MCP
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/mcp.M600383-MCP200 on December 26, 2006.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
M600383-MCP200v1
6/3/503    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Glossary
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Stoevesandt, O.
Right arrow Articles by Brock, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stoevesandt, O.
Right arrow Articles by Brock, R.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Molecular & Cellular Proteomics 6:503-513, 2007.
© 2007 by The American Society for Biochemistry and Molecular Biology, Inc.

Research

A Network Analysis of Changes in Molecular Interactions in Cellular Signaling*,S

Oda Stoevesandt{ddagger},§, Karsten Köhler{ddagger},, Susann Wolf||, Thomas André**, Wilfred Hummel and Roland Brock{ddagger}{ddagger}

From the Interfaculty Institute for Cell Biology, University of Tübingen, Auf der Morgenstelle 15, 72076 Tübingen, Germany


   ABSTRACT
TOP
 ABSTRACT
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Multiprotein complexes play an essential role in the propagation and integration of cellular signals. However, systems level analyses of signaling-dependent changes in the pattern of molecular interactions are still missing. Signaling in T-lymphocytes is one prominent example in which multiprotein complexes orchestrate signal transduction. We implemented peptide microarrays comprising a set of interaction motifs of signaling proteins for network-based analyses of signaling-dependent changes in molecular interactions. Lysates of resting or stimulated cells were incubated on these arrays, and the binding of signaling proteins was detected by immunofluorescence. Signaling-dependent complex formation led to changes of signals on the microarrays in two ways. 1) Masking of a binding site of a signaling protein for a peptide on the array resulted in a signal decrease. 2) Interaction of a protein with a second protein, which in turn binds to a peptide on the array, resulted in a signal increase for the first protein. Dissipation of complexes led to the reverse changes. Competition with peptides corresponding to interaction motifs provided detailed information on the architecture of complexes; lack of individual signaling proteins revealed the functional interdependence of interactions in the network. We show that complex formation through phosphorylation of the scaffolding protein LAT (linker for activation of T-cells) acted as a signal amplifier. PLC{gamma}1 deficiency increased the resting state levels of LAT-dependent complexes and augmented the recruitment of the phosphatase SHPTP2 into complexes. For the analysis of signaling networks, the parallel detection of changes in interactions enabled the identification of functional interdependencies with minimum a priori knowledge.

In recent years the molecular characterization of cellular processes has moved to a systems level. For example, the cellular response of T-lymphocytes to various stimuli has been systematically characterized at the level of transcription (1), translation (2), and protein phosphorylation (3). Given the major technological advances in the analysis of proteins and protein complexes, it is somewhat surprising that comparable systems level approaches for analyzing signaling-dependent changes in protein interaction networks are still missing. Complexes of signaling proteins are key determinants for the integration of signals from different cell surface receptors.

Instead of a direct analysis of interaction networks in signaling-active cells, interaction networks have mostly been inferred by mapping of binary interactions as exemplified by two-hybrid screens (4). Moreover system-wide mappings of molecular complexes and detection of signaling-dependent formation of complexes have been performed through tag-based pulldowns (5, 6). However, this approach requires the introduction of tagged proteins into many separate cell populations and therefore does not provide comprehensive information on the overall pattern of molecular interactions in unmodified cells. The highly parallel detection of signaling-dependent changes in the pattern of molecular interactions in an unmodified cellular system is therefore still an unmet challenge. Here we demonstrate the use of peptide microarrays for this purpose.

We have shown previously that a peptide immobilized on a surface may be used to probe for changes in the availability of a protein domain involved in a signaling-dependent protein-protein interaction (7). This approach was now generalized for the parallel detection of changes in the pattern of interactions in T-cell receptor-dependent signal transduction. In addition, we exploit that proteins directly binding to a capture peptide bring along their interaction partners, making these microarrays a highly parallel variant of coprecipitation techniques.

Interaction domains recognizing short linear peptide motifs play a major role in mediating protein-protein interactions in signaling pathways (8). T-cell receptor-dependent signal transduction is representative for signaling pathways in which interactions mediated via linear peptide motifs play a major role. Src homology (SH)1 2 domains interacting with phosphotyrosine (Tyr(P)) peptide motifs and SH3 domains interacting with polyproline (poly(Pro)) peptide motifs dominate in this network. Other prominent examples are the signaling pathways downstream of the ErbB receptor family, of the platelet-derived growth factor receptor, and of the insulin receptor (9, 10). The InterPro database (www.ebi.ac.uk/interpro) catalogues 322 human SH2 domains interacting with Tyr(P) peptide motifs and 661 human SH3 domains interacting with poly(Pro) peptide motifs.

Upon T-cell activation, a tyrosine kinase cascade involving LCK and ZAP70 leads to the phosphorylation of several tyrosine residues of LAT (11). The phosphorylated LAT scaffold then acts as a germinal center for the formation of a multiprotein complex (12–14). The so-called LAT signalosome plays a key role in the distribution of the signal to various downstream pathways (15, 16). For these early steps in T-cell signaling our peptide arrays provided comprehensive information on the molecular organization and functional interdependence of molecular interactions.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Cell Culture, Stimulation, and Preparation of Cell Lysates—
The leukemic human T-helper cell line Jurkat (17), its LCK-deficient derivative JCaM1.6 (18), and the PLC{gamma}1-deficient derivative J.gamma1 (19) (ATCC, Manassas, VA) were cultivated in RPMI 1640 medium (PAN-Biotech, Aidenbach, Germany) with 10% fetal calf serum (PAN-Biotech) at 37 °C in 5% CO2. For TCR stimulation, cells at a density of 1 x 107 cells/ml were resuspended in ice-cold HEPES-buffered saline (HBS; 10 mM HEPES, 135 mM NaCl, 5 mM KCl, 1 mM MgCl2, 1.8 mM CaCl2, pH 7.4) supplemented with 0.1% BSA and 5 mM glucose containing 10 µg/ml anti-CD3 and/or anti-CD28 antibody (clones OKT3/9.3, G. Jung, Department of Immunology, University of Tübingen) or the isotype control (mouse IgG2a, Sigma) and were incubated on ice for 10 min. Then the cross-linking secondary antibody (goat anti-mouse Ig, Calbiochem) was added to a final concentration of 20 µg/ml, and samples were immediately transferred to 37 °C for the time indicated. For stimulation via broad range phosphatase inhibition, cells were treated in the same medium for 20 min at 37 °C with 0.5 mM pervanadate (20). Using cells expressing a green fluorescent protein fusion protein of the kinase ZAP70 we had established previously that 20 min of incubation with pervanadate yielded a pronounced translocation of the protein to the plasma membrane (7). 5 min of incubation with pervanadate was too short to observe this effect. Pervanadate was freshly prepared by reaction of 5 mM Na3VO4 with 5 mM H2O2 in HBS for 15 min at room temperature.

Stimulation was stopped by placing the samples on ice and washing the cells in ice-cold HBS. For lysis, cells at a density of 2 x 107 cells/ml were resuspended in ice-cold lysis buffer (1% Triton X-100, 50 mM n-octyl ß-D-glucopyranoside (Fluka), 20 mM Tris, 1 mM EDTA, 150 mM NaCl, 1 mM Na3VO4, pH 7.5, and protease inhibitor mixture (Roche Applied Science)) and incubated on ice for 60 min. ß-D-Glucopyranoside ensured the efficient extraction of phosphorylated LAT residing in membrane rafts. For the assessment of peptide effects, peptides were added from DMF stock solutions to the lysis buffer (to a DMF concentration of 1% in the lysate) prior to lysis in the following concentrations unless indicated otherwise: LATpY191, 20 µM; LATpY132, 0.5 µM; SLP228, 20 µM; and SLP179, 20 µM. Crude lysates were clarified by centrifugation at 20,000 x g for 15 min, and total protein concentration was determined by a standard protein assay (Bio-Rad) and adjusted to the same concentration for each sample. These lysates were used on the peptide arrays without further workup.

Peptides—
Peptide sequences were derived from binding motifs described in the literature (Supplemental Table 1). Peptides were 15–20 amino acids long. The binding motifs themselves are only about 4–7 amino acids long and were centrally positioned in the peptides. Peptides containing Cys in positions close to the binding motif itself were avoided. Instead of including spacer building blocks between the N-terminal cysteine residue and the domain-binding motif such as for example 6-aminohexanoic acid, we added some extra amino acid residues from the protein sequence. Peptides were acquired from EMC microcollections (Tübingen, Germany), analyzed by MALDI-TOF and HPLC, and purified as described previously (7). All peptides carried an additional N-terminal Cys residue for immobilization on epoxy-activated surfaces. Peptide designations begin with a superscript "pep" and are composed from the name of the protein from which they were derived and either the position of the Tyr(P) residue in the whole protein (for Tyr(P) peptides) or the position of the first residue in the peptide (for poly(Pro) and other peptides).

Generation of Microarrays—
Peptides were freshly diluted from DMF stock solutions to a spotting concentration of 100 µM in 0.1 M phosphate buffer (pH 8.0). 16 identical arrays of up to 9 x 9 spots per array were spotted on epoxy-functionalized glass slides (Nexterion Slide E, Schott, Jena, Germany) at a relative humidity of 70% using a GeSiM NP2.0 nanopipettor (GeSiM, Dresden, Germany). A fluorescein-labeled control peptide was spotted to assist the orientation. Per peptide, duplicate spots of 1.2 nl each were spotted with a center-to-center spacing of 500 µm. Slides were incubated with of O,O'-bis(2-aminopropyl)polyethylene glycol 800 (Fluka) at 70 °C for 16 h to quench the remaining reactive epoxy groups. After cooling, slides were rinsed with deionized water, dried in a stream of filtered processed air, and stored at 4 °C.

Incubation of Arrays and Immunofluorescence—
Before incubation, slides carrying arrays were washed with lysis buffer and double distilled water and dried in a stream of filtered processed air. A 16-well clip-on frame (ProPlate Multiarray System, Grace Biolabs, Molecular Probes, Eugene, OR) was assembled with the slide to create a separate incubation chamber for each array. Arrays were blocked with 1% Top-Block (Fluka) in PBS for 1 h, washed with TPBSB (PBS, 0.05% Tween, 0.05% BSA), and incubated with 50 µl of cell lysate/chamber at 4 °C for 1 h. Then the arrays were washed with TPBSB, incubated with 2 µg/ml primary antibody in TPBSB at room temperature for 15 min, washed again with TPBSB, and incubated with 1 µg/ml secondary antibody in TPBSB at room temperature for 10 min. Subsequently arrays were washed with TPBSB, rinsed with double distilled H2O, and dried in an air stream. Per array, two primary antibodies were used, one from mouse and one from rabbit. Secondary antibodies were goat anti-mouse Alexa546 and goat anti-rabbit Alexa633 (Molecular Probes/Invitrogen). For each primary antibody used, a negative control was carried out by immunostaining of an array without prior incubation with lysate. The specificity of detection antibodies was validated by Western blotting. Slides were scanned at 10-µm resolution on a ScanArray array scanner (PerkinElmer Life Sciences) using lasers of 543 and 633 nm for excitation.

Image Analysis and Data Processing—
Semiautomatic image analysis was performed using Array Pro (Media Cybernetics, Silver Spring, MD). For each experiment, the image of the array was overlaid with a grid corresponding to the array layout. The net signal intensity of each spot (Ispot net) was determined by subtracting the background intensity of a ring around the spot (Ispot net = Ispot – Ilocal background). Further data processing was carried out in Excel. Negative Ispot net values were considered no signals and set to 0. From Ispot net of each peptide spot and incubation condition, the respective signal of the antibody control was subtracted (Ispot cont = Ispot net – I controlspot net). If a negative Ispot cont value was the result, it was set to 1 to enable further batch calculations. Ispot cont values of the stimulated conditions were divided by the corrected signal intensity of the unstimulated condition from respective peptide spots if at least one of the values was above a certain threshold. Thresholds were 100 for pervanadate (PV) stimulation experiments and 50 for antibody stimulation. Thresholds were established empirically. Results of several experiments were combined by calculation of the medians of the ratios, rendering the analysis more robust toward outliers.


   RESULTS
TOP
 ABSTRACT
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Experimental Design—
16 identical peptide arrays per glass slide were individually incubated with cell lysates. Each array was immunostained for two bound proteins using primary antibodies from mouse and rabbit and species-specific distinctly labeled secondary antibodies (Fig. 1A). The peptides were selected as a set of known binding motifs for interaction domains of proteins involved in T-cell signal transduction with a focus on interactions up- and downstream of the scaffolding protein LAT (Supplemental Table 1). To limit redissociation of weakly bound proteins, antibody incubation and washing were as brief as possible. Judged by the intensity of immunostaining, proteins showed a strong preference for peptides representing binding motifs for the respective domain subtype, notably peptides derived from known binding partners (Fig. 1B).


Figure 1
View larger version (29K):
[in this window]
[in a new window]

  		FIG. 1. Peptide microarrays for the detection of signaling-dependent changes of molecular interactions. A, one slide holding 16 identical peptide microarrays incubated with cell lysates and different antibodies for immunostaining. Each subarray was incubated with lysate of 106 Jurkat cells. Green, secondary immunostaining with anti-mouse Alexa546; red, secondary immunostaining with anti-rabbit Alexa633. For the total of 31 different immobilized peptides, binding of key proteins in T-cell signal transduction could be detected on 23 peptides, comprising 16 Tyr(P) motif peptides, six poly(Pro) motif peptides, and one WASP homology domain 1 domain-binding peptide. Detailed view of the outlined area, set of arrays used for determining background binding of anti-PLC{gamma}1, anti-LAT, anti-GADS, and anti-SLP76 and binding of PLC{gamma}1 and LAT from resting and PV-treated cells. B, background-corrected signal intensities of anti-PLC{gamma}1 and anti-GADS staining on selected peptides after incubation of arrays with lysates of resting cells. Results from five independent experiments distinguished by color and median (black bar) are shown. Proteins showed a strong preference for peptides representing binding motifs for the respective domain subtype. PLC{gamma}1 contains two SH2 and one SH3 domain and GADS contains one SH2 and two SH3 domains but of different motif subtype specificities: pepLATpY132 (SHP/PLC motif) matches the SHP/PLC-type SH2 domain of PLC{gamma}1; pepLATpY191 (GRB2/SFK/PLC motif) and pepLATpY226 (GRB2/SFK motif) match the GRB2-type SH2 domain of GADS; pepSLP179 (type 2 poly(Pro) motif) matches the type 2 SH3 domain of PLC{gamma}1; and pepSLP228 (atypical SH3 domain-binding motif) matches the atypical SH3 domain of GADS. C, fluorescence intensities F derived for resting and PV-treated conditions were corrected for background binding of antibodies. Median values of five experiments are shown if a minimum of four of five experiments produced a signal above threshold (see "Experimental Procedures"). For PLC{gamma}1, the strong, activation-dependent signal increases on pepSLP228 and pepGAB2-509 are indicative for the signaling-dependent formation of protein complexes. Similarly the signals for LAT may only be explained by the binding of complexes as LAT lacks a signaling domain. In B and C peptides are indicated without superscript pep. Proteins are printed in boldface italics. a.u., arbitrary units; stim., stimulated;pY, phosphotyrosine.

 
Phosphorylation-dependent Changes in Interaction Profiles—
The formation of protein complexes upon tyrosine phosphorylation plays a major role in T-cell signaling. To initiate the formation of such complexes, we used the phosphotyrosine phosphatase inhibitor pervanadate (20). Lysates of resting or pervanadate-treated Jurkat T-cell leukemia cells were incubated on separate arrays (Fig. 1, A and C), and signal ratios of corresponding peptides were calculated (Fig. 2A). For interactions of SH2 domain-containing proteins on Tyr(P) peptides representing known binding partners, pervanadate induced a moderate signal decrease as for example for GADS, GRB2, and PLC{gamma}1 on the respective pepLATpY motifs. This decrease was in line with our previously demonstrated concept that phosphorylation-dependent complex formation masks binding sites of proteins and therefore reduces binding to cognate immobilized phosphopeptides (7). For proteins directly binding to cognate poly(Pro) motifs through SH3 domains, we observed only moderate signal changes, often minor increases, as for example for PLC{gamma}1 binding to pepSLP179 or GRB2 binding to pepGAB2-509 (21) (for the nomenclature of peptides refer to Supplemental Table 1).


Figure 2
View larger version (49K):
[in this window]
[in a new window]

  		FIG. 2. Changes in protein signal intensities on the peptide microarrays upon treatment of cells with pervanadate. Jurkat cells and derivative cell lines were used. Peptides immobilized on the array are plotted from left to right, and proteins probed by immunostaining are plotted from top to bottom, i.e. each row represents the background-corrected ratios of two arrays (lysates from stimulated versus lysates from resting cells) probed for one protein. On the right, the numbers of SH2 and SH3 domains of the respective proteins are listed. A, Jurkat cells. The median ratio of five experiments is shown if a minimum of four of five experiments produced a signal above threshold (see "Experimental Procedures"). B, J.gamma1 cells; and C, JCaM1.6 cells. The median ratio of four experiments is shown if a minimum of three of four experiments produced a signal above threshold. atyp., atypical; stim., stimulated; unstim., unstimulated; FYB, FYN-binding protein; WIP, WASP-interacting protein; WH1, WASP homology domain 1; PTB, phosphotyrosine-binding domain; PAK, p21-activated kinase, serine/threonine-protein kinase; cCBL, Casitas B-lineage lymphoma proto-oncogene; SIG, SIGLEC-7, Sialic acid-binding Ig-like lectin 7.

 
In addition, proteins lacking SH3 domains were also detected on the SH3 domain-binding poly(Pro) peptides (Fig. 2A), suggesting recruitment to the array in a complex with a direct binding partner for an immobilized poly(Pro) peptide. Signal increases for these proteins were much stronger than for proteins for which direct binding was possible. For example, LAT displayed a stimulation-dependent median 74-fold increase of signal on pepSLP228 (Figs. 1C and 2A). This binding could be mediated by an SH2/SH3 domain adapter like GADS or GRB2, which were also detected. Signals for which the subtype preference of the respective domain was violated could also originate from the binding of complexes rather than direct interactions. For example, the LAT interactor PLC{gamma}1 was detected on pepSLP228 with a median increase upon stimulation of 4400. Likewise the increase of the GRB2 signal on pepSIGpY437 could be explained by a phosphorylation-dependent interaction between GRB2 and SHPTP2 and binding of SHPTP2 to pepSIGpY437. The absolute signal intensities also provided evidence for the binding of protein complexes to the arrays (Fig. 1C). For lysates of resting cells, immunostaining for PLC{gamma}1 was intense on pepSLP179 and pepPAK6 and negligible on pepSLP228 and pepGAB2-509. For lysates of pervanadate-treated cells, signals on pepSLP228 and pepGAB2-509 were stronger than signals on pepSLP179 and pepPAK6. If a mere change in the availability of the SH3 domain of PLC{gamma}1 was the reason for a signal increase, then the signal should increase uniformly on all peptides. The reversal in the order of signal intensities therefore suggests that a portion of PLC{gamma}1 was present in complexes recruited to the array via the SH3 domain of another protein.

Dissection of Complex Architectures by Peptide Competition and Titration—
Next we sought to devise a general strategy for translating the changes of signals on the microarrays into testable hypotheses on the architecture of signaling complexes. First an interaction network was deduced from the microarray information (Fig. 3). In the network, signals involving GADS, PLC{gamma}1, SLP76, and LAT and their respective peptides, all proteins that are part of the LAT signalosome, are highly interconnected (Fig. 4A) (12–14, 16). Considering direct and indirect interactions, different possibilities exist to explain the signals on pepSLP228 (Fig. 4B) or on pepLATpY191 (Fig. 4C).


Figure 3
View larger version (62K):
[in this window]
[in a new window]

  		FIG. 3. Interaction network in wild-type Jurkat cells constructed from changes of signals for the individual spots shown in Fig. 2A. Nodes of the network represent the peptides on the arrays (white boxes) and/or the proteins probed for by immunofluorescence (gray boxes). The line width represents the signal intensity of the protein on the respective peptide for the unstimulated condition with the tip of the arrow going from the detected protein to the peptide. The line color represents the signal ratio of PV-treated to resting condition as introduced in Fig. 2A. In contrast to networks inferred from a systematic mapping of molecular interactions using e.g. yeast two-hybrid screens (4) or microarrays with immobilized recombinant protein domains (24, 25), this network carries information on changes in the pattern of molecular interactions in the signaling-active cells themselves, expressing endogenous proteins only. PAK, p21-activated kinase, serine/threonine-protein kinase; WIP, WASP-interacting protein; FYB, FYN-binding protein; pY, phosphotyrosine.

 

Figure 4
View larger version (37K):
[in this window]
[in a new window]

  		FIG. 4. Stepwise dissection of signaling complexes through peptide competition. A, subnetwork of interactions derived from the binding patterns observed for resting and PV-stimulated Jurkat cells (Fig. 2A). Arrows represent either direct or indirect interactions and point toward the binding partner represented by a peptide on the array. Arrowheads capped with a perpendicular line represent indirect interactions for which the binding protein lacks an interaction domain for the respective peptide motif. Arrow colors represent the change of signal upon stimulation (cf. Fig. 2); arrow width represents the signal intensity of the unstimulated condition. B, I–III depict possible architectures for the binding of GADS, LAT, and PLC{gamma}1 to the arrayed pepLATpY191 based on the binding domains of the proteins. C, changes of binding of GADS, LAT, and PLC{gamma}1 to pepLATpY191 upon competition with pepLATpY191 (20 µM), pepLATpY132 (0.5 µM) (22), pepSLP228 (20 µM), and pepSLP179 (20 µM). The peptide-induced dissociation of proteins is consistent with option III. D, I and II depict possible architectures for binding of GADS, SLP76, and PLC{gamma}1 to the arrayed pepSLP228. E, changes of binding of GADS, SLP76, and PLC{gamma}1 to pepSLP228 upon competition with pepLATpY191, pepLATpY132, pepSLP228, and pepSLP179 (concentrations as in C) for lysates of resting and PV-activated cells. C and E, means and mean deviations from the mean of two independent experiments. pY, phosphotyrosine.

 
We had shown previously that in cell lysates the PLC{gamma}1-LAT interaction was competed by incubation with pepLATpY132 and less effectively with pepLATpY191 (22). Incubation of lysates with competitor peptides prior to incubation on the array should therefore also disrupt complexes. The resulting changes in the pattern of signals should then enable the determination of the binding mode. For example, if three proteins were bound as a linear complex, then incubation with peptides corresponding to individual interaction motifs should selectively remove one, two, or all three proteins from the array. On pepSLP228, the signals of GADS, LAT, and PLC{gamma}1 were all sensitive to pepSLP228 (Fig. 4B). Upon addition of pepLATpY191, the signals of LAT and PLC{gamma}1 decreased, whereas the signal of GADS was not affected. In contrast to the LAT signal, the PLC{gamma}1 signal was also sensitive to pepLATpY132. Taken together, these data are fully consistent with the GADS-mediated binding of a complex consisting of GADS, LAT, and PLC{gamma}1 (Fig. 4C). The dependence of these interactions on the activation-induced phosphorylation of LAT also explains the strong activation-dependent increase of the LAT and PLC{gamma}1 signals on the array.

Peptide competition of GADS, SLP76, and PLC{gamma}1 on the immobilized pepLATpY191 provided further information on the signaling-dependent reorganization of these proteins. All proteins contain SH2 domains and might therefore bind the peptide independently. Alternatively GADS and PLC{gamma}1 might engage in activation-independent interactions mediated via specific poly(Pro) motifs on SLP76 (Fig. 4D). GADS, SLP76, and PLC{gamma}1 were all sensitive to pepLATpY191. However, only SLP76 and PLC{gamma}1 were sensitive to pepSLP228. Peptides pepSLP179 and pepLATpY132 most strongly affected PLC{gamma}1; however, signals for the other proteins were also reduced (Fig. 4E). The latter observations suggest that further interactions, e.g. of the PLC{gamma}1 SH2 domain with the peptide microarray, also stabilize the complex. To determine to which degree the SLP76 poly(Pro)-mediated interactions were present constitutively, lysates of resting cells were also incubated with competitor peptides. SLP76 was fully dissociated by incubation with pepSLP228, consistent with the presence of a preformed complex between GADS and SLP76 mediated by the constitutive interaction of GADS with the atypical pepSLP228 poly(Pro) motif (12). For PLC{gamma}1, the changes were less pronounced. Incubation with pepSLP228 led to some reduction of signal; however, pepSLP179 was without effect. In accordance with the changes observed on the array after activation for the pepSLP228 spot, these results indicate that more PLC{gamma}1 entered a GADS-containing complex (Fig. 4E).

Systems-oriented Dissection of Molecular Interactions—
Having performed a targeted disruption of molecular complexes, we next aimed to determine the significance of individual interaction motifs on a systems level. For seven proteins, binding to the array was titrated with either pepLATpY132 or pepLATpY191 (Fig. 5 and Supplemental Fig. S1). Overall pepLATpY191 had a more pronounced effect on the pattern of interactions than did pepLATpY132 that may be explained by the broader specificity of pepLATpY191 for different types of SH2 domains (11, 14, 23) (Fig. 5A). Binding of PLC{gamma}1 to most poly(Pro) peptides was also inhibited, indicating a mediation of interactions by other proteins to which PLC{gamma}1 binds via its SH2 domains. This experiment also demonstrates that the interaction of LAT with proteins that contain polyproline motif-binding SH3 domains is mediated via proteins binding to pepLATpY191. For LAT and PLC{gamma}1 on the pepSLP228 spot, signal losses upon addition of pepLATpY191 occurred with the same concentration dependence (Fig. 5B), whereas for pepLATpY132 only PLC{gamma}1 dissociated, supporting the binding order inferred from the simple peptide competition experiment (Fig. 4B).


Figure 5
View larger version (35K):
[in this window]
[in a new window]

  		FIG. 5. Titration of lysates of PV-treated cells with pepLATpY132 and pepLATpY191. A, overview of protein signals affected by peptide titration. x indicates signals that decreased by at least 50% during the titration. Peptides on which protein binding was not affected by peptide titration are omitted. Titration with pepLATpY191 has a more severe effect on molecular interactions. B, relative signals between lysates of PV-treated cells incubated with increasing peptide concentrations and resting cells on pepSLP228. The pepLATpY191-induced dissociation of PLC{gamma}1 and LAT is consistent with the results shown in Fig. 4C. C, influence of peptide titration on the signal intensities of PLC{gamma}1 and SHPTP2 on different peptide spots. Data from one representative experiment are shown. The different dissociation patterns induced by titration with peptides suggest that both proteins do not directly compete for the same binding sites. Data from one representative experiment are shown. PAK, p21-activated kinase, serine/threonine-protein kinase.

 
Effect of Single Protein Deficiency on the Interaction Network—
The titration experiments had revealed the significance of individual interaction motifs for the functional organization of the interaction network. Next we examined the impact of the loss of a multidomain protein involved in multiple complexes by using the PLC{gamma}1-deficient Jurkat derivative J.gamma1 (19). For proteins supposed to bind directly to immobilized peptides (e.g. GADS and GRB2 on pepLATpY191 and pepSLP228), only marginal effects of PLC{gamma}1 deficiency were observed (Fig. 2B and Supplemental Fig. S2). For resting PLC{gamma}1-deficient cells, binding of LAT to poly(Pro) peptides was increased compared with resting Jurkat cells. In contrast, for activated PLC{gamma}1-deficient cells, the phosphorylation-dependent recruitment of LAT to poly(Pro) peptides was inhibited compared with activated Jurkat cells. Also SLP76 on pepLATpY132 became undetectable for both resting and activated PLC{gamma}1-deficient cells. In contrast, signals of SHPTP2 on poly(Pro) peptides were increased for PLC{gamma}1-deficient cells compared with Jurkat cells for resting and for activated cells (Fig. 6A and Supplemental Fig. S3). As SHPTP2 itself does not contain an SH3 domain, this interaction has to be indirect.


Figure 6
View larger version (46K):
[in this window]
[in a new window]

  		FIG. 6. Impact of single protein deficiencies on the signaling network in Jurkat cells. Shown are changes in protein signal intensity on the peptide microarrays between signaling-defective cells and "wild-type" Jurkat cells expressed as background-corrected ratio "deficient/WT." Lysates of resting (–) and pervanadate-treated (+) cells were used. The median ratio of four experiments is shown if a minimum of three of four experiments produced a signal above threshold. A, PLC{gamma}1-deficient J.gamma1 cells. For proteins supposed to bind directly to immobilized peptides, only marginal changes were observed. Prominent effects were observed for LAT and SHPTP2. PLC{gamma}1 deficiency led to an increase of LAT on polyproline peptides in lysates from resting state cells, indicative of an increase in resting state fluorescence, and favored the entry of SHPTP2 into signaling complexes. B, LCK-defective JCaM1.6 cells. The predominant signal decreases on Tyr(P) peptides were largely missing in agreement with the role of LCK in starting a phosphorylation cascade. atyp., atypical; FYB, FYN-binding protein; WIP, WASP-interacting protein; WH1, WASP homology domain 1; PTB, phosphotyrosine-binding domain; PAK, p21-activated kinase, serine/threonine-protein kinase.

 
As a model for a generally impaired TCR downstream signaling, the LCK-defective Jurkat derivative JCaM1.6 (18) was selected. For the LCK-defective cells, the predominant stimulation-dependent signal decreases that were observed on Tyr(P) peptides for Jurkat cells were largely missing in agreement with the role of LCK in starting a phosphorylation cascade (Fig. 2C and Supplemental Fig. S4). For the LCK-defective cells, PLC{gamma}1 became undetectable on most poly(Pro) peptides likely due to a lack of phosphorylated LAT and SLP76 as mediators of binding (Fig. 2C). By contrast, pervanadate-dependent recruitment of SHPTP2 to poly(Pro) peptides even increased for LCK-defective cells compared with Jurkat cells (Fig. 2C).

Western blotting showed that, except for the deficient proteins, the levels of most proteins probed for on the arrays, notably of LAT and SHPTP2, were the same in Jurkat, J.gamma1, and JCaM1.6 cells (Supplemental Fig. S5). Only for PI3K and VAV was less protein detected in JCaM1.6 cells.

Stimulus and Time Dependence of Protein Interactions in T-cell Activation—
We next used the peptide arrays to determine the dependence of complex formation on different specific stimuli. Jurkat cells were activated by stimulatory anti-CD3, costimulatory anti-CD28, or both antibodies for variable times, and the microarrays were probed for PLC{gamma}1, LAT, PI3K, SLP76, GADS, and LCK. Distinct signal profiles were obtained for PLC{gamma}1, LAT, SLP76, and PI3K (Fig. 7). For PLC{gamma}1 and LAT, signal increases on poly(Pro) peptides were strongest after 2 min of CD3 stimulation, suggesting fast Tyr(P)-dependent complex formation and corecruitment to the array. Only on pepSLP228 did CD28 stimulation alone induce slight increases in PLC{gamma}1 recruitment. With CD3/CD28 costimulation signal increases for PLC{gamma}1 were augmented compared with CD3 stimulation and detectable for a longer period of time. For SLP76, signal decreases on LATpY peptides were detected upon CD3 stimulation. This decrease was most pronounced after 10 min of stimulation, indicating that recruitment of SLP76 to the LAT complex occurred with a slower kinetics than the phosphorylation of LAT and recruitment of PLC{gamma}1. For PI3K, no systematic signal changes were observed upon stimulation with CD3 or CD28 alone, whereas costimulation led to a signal decrease. Signal changes upon antibody stimulation had the same direction as signal changes upon pervanadate stimulation (Fig. 7). Nevertheless the effects of pervanadate stimulation carried out alongside the antibody stimulations were weaker than in the experiments presented above (Fig. 2); this may be explained by a preincubation step on ice.


Figure 7
View larger version (46K):
[in this window]
[in a new window]

  		FIG. 7. Stimulus and time dependence of changes in molecular interactions. Shown are signal changes on peptide microarrays following treatment of Jurkat cells with stimulatory antibodies against the TCR/CD3 complex ({alpha}CD3) and the CD28 coreceptor ({alpha}CD28) for different time points. Given are the means and mean deviations of the mean of three experiments of the signal ratios of stimulated versus resting cells. The respective proteins are indicated on top of each column of the graphs, the peptide spot is indicated in each graph, and the incubation conditions are indicated at the lower right. A negative control was generated by mock stimulation with an isotype antibody (IgG); a positive control was generated by incubation with PV. In the presence of {alpha}CD28, signals of murine detection antibodies could not be quantified because of high overall backgrounds, accounting for the data points missing for LAT and SLP76. For PLC{gamma}1 and LAT, signal increases on poly(Pro) peptides were strongest after 2 min of CD3 stimulation, suggesting fast Tyr(P)-dependent complex formation and corecruitment to the array. For SLP76, CD3-induced signal decreases on LATpY peptides were most pronounced after 10 min of stimulation, indicating that recruitment of SLP76 to the LAT complex occurred with a slower kinetics than the phosphorylation of LAT and recruitment of PLC{gamma}1. PAK, p21-activated kinase, serine/threonine-protein kinase.

 

   DISCUSSION
TOP
 ABSTRACT
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
Here we demonstrate the highly parallel detection of changes in the pattern of protein-protein interactions and acquisition of information on the architecture of signaling complexes using peptide microarrays. One slide containing 16 arrays with 23 capture peptides each, probed with two antibodies per array, yields 736 data points per slide.

Other recent examples for microarray-based approaches in the highly parallel analysis of molecular interactions have inferred possible signal transduction networks from interactions obtained with recombinant proteins (24, 25). In contrast, the interaction network derived by our approach provides functional information on activation-dependent changes in molecular interactions. An expression of tagged proteins is not required. We show that time dependence, stimulus dependence, the effect of signaling deficiencies, and the disruption of molecular interactions can be assessed.

Especially on poly(Pro) peptides, we detected numerous proteins without a cognate binding domain. Through titration with peptides corresponding to binding motifs of domains of the detected proteins, we were able to dismantle complexes on the arrays in a stepwise and dose-dependent manner. These experiments allowed a hypothesis-driven verification of models on the architecture of complexes giving rise to the observed combination of signals. Moreover as detailed for Jurkat derivatives deficient in single proteins, our experiments provided comprehensive insights into the functional interdependence of molecular interactions.

Stronger activation-dependent signal increases were observed for those proteins that could not bind directly to the peptide. In general these peptides bound via phosphorylation-dependent interactions to the poly(Pro) peptide-binding proteins (Supplemental Fig. S6). For example, for the activation-induced GADS-LAT-PLC{gamma}1 complex bound to pepSLP228 the magnitude of the activation-dependent signal changes increased with distance from the peptide microarray (Figs. 2A and 4, B and C, and Supplemental Fig. S6). This observation is readily explained by the formation of the binding site only after stimulation. If each of the interactions contributing to a linear complex on the immobilized peptide (peptide-A-B-C) occurred with a different probability after stimulation (peptide-A changing x-fold, A-B changing y-fold, and B-C changing z-fold), then the chance to find C on the immobilized peptide would change x x y x z-fold. This result is consistent with the concept that a scaffolding protein, which links two signaling proteins, may act as signal amplifier (26).

On phosphotyrosine peptides, secondary interactions were mostly mediated by poly(Pro) motifs, such as detailed for the GADS-SLP76-PLC{gamma}1 complex on pepLATpY191. Correspondingly within one potential complex, relative changes were comparable.

In resting and activated PLC{gamma}1-deficient and activated LCK-defective cells, the binding of SHPTP2 on poly(Pro) peptides was increased compared with Jurkat cells. SHPTP2 on pepSLP228 was sensitive to titration with pepLATpY191 (Fig. 5C) but was not detected on immobilized pepLATpY191 (Figs. 2 and 6). This suggests that the titration with pepLATpY191 rather targeted an SH2 domain binding to phosphorylated SHPTP2 than the SH2 domain of SHPTP2 itself. GRB2 is a likely candidate for mediating this interaction (27). The different dissociation patterns induced by titration with pepLATpY132 and pepLATpY191 (Fig. 5A) also suggest that PLC{gamma}1 and SHPTP2 do not directly compete for the same binding sites. Interestingly the levels of LAT on the poly(Pro) spots in resting PLC{gamma}1-deficient cells were increased. These results indicate that in resting cells the presence of PLC{gamma}1 has an effect on the resting state phosphorylation of key signaling proteins in the network and thereby the pattern of preformed complexes. In the PLC{gamma}1-deficient cells, the propagation of the signal should therefore not only be affected by the mere loss of this protein but also by a change in the predisposition of a cell to respond to a stimulus. For PLC{gamma}1-deficient cells, signals for SHPTP2 on poly(Pro) peptides were stronger than for Jurkat cells (Fig. 6A), further supporting a role of PLC{gamma}1 in antagonizing the recruitment of SHPTP2 into complexes. Additional evidence for this role of PLC{gamma}1 is provided by the analysis of LCK-defective Jurkat cells. Recruitment of PLC{gamma}1 into complexes was strongly reduced, whereas recruitment of SHPTP2 was increased. Moreover this observation indicates that PLC{gamma}1 but not SHPTP2 is downstream of LCK.

For the interplay of the stimulation of TCR/CD3 and CD28 by stimulatory antibodies, our peptide microarrays allowed the analysis of time courses of protein complex formation in a parallel manner. We demonstrate that the recruitment of SLP76 to the LAT signalosome followed a different time course than the recruitment of PLC{gamma}1. Recently an analysis of phosphorylation kinetics in early T-cell signaling showed that phosphorylation of LAT residues Tyr-132 and Tyr-191 reaches maximum levels within 2 min (28). Our data show that the arrayed pepLATpY191 was maximally competed by its cellular counterpart after 10 min of stimulation, suggesting that LAT phosphorylation is not the rate-limiting step for formation of the LAT signalosome. Only for PI3K was costimulation with anti-CD3 and anti-CD28 required for the detection of changes. For PLC{gamma}1, costimulation led to stronger changes. These results demonstrate the potential of the peptide microarrays for integrated network analysis in systems with multiple stimuli.

In conclusion, peptide microarrays provide a tool for the highly parallel analysis of protein-protein interactions in cellular signal transduction. The benefit of using peptide microarrays goes far beyond a mere saving of time due to the parallel processing of samples and a minimization of sample consumption as demonstrated for the titration of interactions with LAT-derived Tyr(P) motifs. Only through the parallel detection of signaling-dependent changes of the binding of proteins to the array was information obtained on the phosphorylation-dependent amplification of signal propagation. Moreover the antagonizing effect of PLC{gamma}1 on the recruitment of SHPTP2 into complexes was detected with minimum a priori knowledge. Classical biochemical co-immunoprecipitation approaches are strictly hypothesis-driven, i.e. pairs of interactions are probed for intentionally.

This approach will further benefit from the use of arrayed peptide libraries derived from the literature and by motif prediction (23, 29) and from antibody libraries that together may simultaneously probe for all potential peptide motif-mediated molecular interactions in a network. Furthermore protein domains immobilized on the array and as competitors for titration will extend the approach to domain-domain interactions. Quantitative analyses of signal changes upon stimulation and competition experiments provide a wealth of information on the functional organization of protein complexes and thus further our understanding of interactions in signal transduction in a systems biological view.


   ACKNOWLEDGMENTS
 
Gundram Jung, Ludger Grosse-Hovest, Alexander Ganser, Reiner Lammers (Tübingen, Germany), Christian Freund (Berlin, Germany), and Anne-Marie Lellouch and Hans Rogl (Marseille, France) kindly provided antibodies. Georg Otto (Max Planck Institute for Developmental Biology Tübingen) and Michael Hartmann and Thomas Joos (Natural and Medical Sciences Institute at the University of Tübingen) kindly provided access to microarray scanners.


  FOOTNOTES
 
Received, October 5, 2006, and in revised form, December 21, 2006.

Published, MCP Papers in Press, December 26, 2006, DOI 10.1074/mcp.M600383-MCP200

1 The abbreviations used are: SH, Src homology; GAB2, GRB2-associated binding protein 2; GADS, GRB2-related adapter protein downstream of SHC; GRB2, growth factor receptor-bound protein; LAT, linker for activation of T-cells; LCK, lymphocyte cell-specific protein-tyrosine kinase; PI3K, phosphatidylinositol 3-kinase; PLC, phospholipase C; PV, pervanadate; SHPTP2, SH domain-containing protein tyrosine phosphatase 2; SLP76, SH2 domain-containing leukocyte phosphoprotein of 76 kDa; TCR, T-cell receptor; ZAP70, {zeta}-associated protein-tyrosine kinase of 70 kDa; HBS, HEPES-buffered saline; DMF, N,N-dimethylformamide; WASP, Wiskott-Aldrich syndrome proteins; SFK, Src family of protein kinases. Back

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

S The on-line version of this article (available at http://www.mcponline.org) contains supplemental material. Back

{ddagger} Both authors contributed equally to this work. Back

§ Present address: Technology Research Group, The Babraham Inst., Babraham, Cambridge CB2 4AT, UK. Back

A scholar of the Graduiertenkolleg 794. Present address: Division of Cell and Molecular Biology, Imperial College, South Kensington, London SW7 2AZ, UK. Back

|| Present address: Inst. for Biochemistry, University of Leipzig, Johannisallee 30, 04103 Leipzig, Germany. Back

** Present address: EMC microcollections GmbH, Sindelfinger Strasse 3, 72070 Tübingen, Germany. Back

{ddagger}{ddagger} Supported by the Volkswagen Foundation ("Nachwuchsgruppen an Universitäten," I/77 472) and the University of Tübingen (Strukturfond). To whom correspondence should be addressed: Dept. of Biochemistry, Nijmegen Centre for Molecular Life Sciences, Radboud University Nijmegen Medical Centre, P. O. Box 9101, 6500 HB Nijmegen, The Netherlands. Tel.: 31(0)24-36-16413; Fax: 31(0)24-36-66213; E-mail: r.brock{at}ncmls.ru.nl


   REFERENCES
TOP
 ABSTRACT
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Diehn, M., Alizadeh, A. A., Rando, O. J., Liu, C. L., Stankunas, K., Botstein, D., Crabtree, G. R., and Brown, P. O. (2002) Genomic expression programs and the integration of the CD28 costimulatory signal in T cell activation. Proc. Natl. Acad. Sci. U. S. A. 99, 11796 –11801[Abstract/Free Full Text]

  2. Grolleau, A., Bowman, J., Pradet-Balade, B., Puravs, E., Hanash, S., Garcia-Sanz, J. A., and Beretta, L. (2002) Global and specific translational control by rapamycin in T cells uncovered by microarrays and proteomics. J. Biol. Chem. 277, 22175 –22184[Abstract/Free Full Text]

  3. Kim, J. E., and White, F. M. (2006) Quantitative analysis of phosphotyrosine signaling networks triggered by CD3 and CD28 costimulation in Jurkat cells. J. Immunol. 176, 2833 –2843[Abstract/Free Full Text]

  4. Rual, J. F., Venkatesan, K., Hao, T., Hirozane-Kishikawa, T., Dricot, A., Li, N., Berriz, G. F., Gibbons, F. D., Dreze, M., Ayivi-Guedehoussou, N., Klitgord, N., Simon, C., Boxem, M., Milstein, S., Rosenberg, J., Goldberg, D. S., Zhang, L. V., Wong, S. L., Franklin, G., Li, S., Albala, J. S., Lim, J., Fraughton, C., Llamosas, E., Cevik, S., Bex, C., Lamesch, P., Sikorski, R. S., Vandenhaute, J., Zoghbi, H. Y., Smolyar, A., Bosak, S., Sequerra, R., Doucette-Stamm, L., Cusick, M. E., Hill, D. E., Roth, F. P., and Vidal, M. (2005) Towards a proteome-scale map of the human protein-protein interaction network. Nature 437, 1173 –1178[CrossRef][Medline]

  5. Gavin, A. C., Bosche, M., Krause, R., Grandi, P., Marzioch, M., Bauer, A., Schultz, J., Rick, J. M., Michon, A. M., Cruciat, C. M., Remor, M., Hofert, C., Schelder, M., Brajenovic, M., Ruffner, H., Merino, A., Klein, K., Hudak, M., Dickson, D., Rudi, T., Gnau, V., Bauch, A., Bastuck, S., Huhse, B., Leutwein, C., Heurtier, M. A., Copley, R. R., Edelmann, A., Querfurth, E., Rybin, V., Drewes, G., Raida, M., Bouwmeester, T., Bork, P., Seraphin, B., Kuster, B., Neubauer, G., and Superti-Furga, G. (2002) Functional organization of the yeast proteome by systematic analysis of protein complexes. Nature 415, 141 –147[CrossRef][Medline]

  6. Bouwmeester, T., Bauch, A., Ruffner, H., Angrand, P. O., Bergamini, G., Croughton, K., Cruciat, C., Eberhard, D., Gagneur, J., Ghidelli, S., Hopf, C., Huhse, B., Mangano, R., Michon, A. M., Schirle, M., Schlegl, J., Schwab, M., Stein, M. A., Bauer, A., Casari, G., Drewes, G., Gavin, A. C., Jackson, D. B., Joberty, G., Neubauer, G., Rick, J., Kuster, B., and Superti-Furga, G. (2004) A physical and functional map of the human TNF-{alpha}/NF-{kappa}B signal transduction pathway. Nat. Cell Biol. 6, 97 –105[CrossRef][Medline]

  7. Stoevesandt, O., Elbs, M., Köhler, K., Lellouch, A. C., Fischer, R., André, T., and Brock, R. (2005) Peptide microarrays for the detection of molecular interactions in cellular signal transduction. Proteomics 5, 2010 –2017[CrossRef][Medline]

  8. Pawson, T., and Nash, P. (2003) Assembly of cell regulatory systems through protein interaction domains. Science 300, 445 –452[Abstract/Free Full Text]

  9. Yaffe, M. B. (2002) Phosphotyrosine-binding domains in signal transduction. Nat. Rev. Mol. Cell. Biol. 3, 177 –186[CrossRef][Medline]

  10. Yarden, Y., and Sliwkowski, M. X. (2001) Untangling the ErbB signalling network. Nat. Rev. Mol. Cell. Biol. 2, 127 –137[CrossRef][Medline]

  11. Paz, P. E., Wang, S., Clarke, H., Lu, X., Stokoe, D., and Abo, A. (2001) Mapping the Zap-70 phosphorylation sites on LAT (linker for activation of T cells) required for recruitment and activation of signalling proteins in T cells. Biochem. J. 356, 461 –471[CrossRef][Medline]

  12. Liu, S. K., Fang, N., Koretzky, G. A., and McGlade, C. J. (1999) The hematopoietic-specific adaptor protein gads functions in T-cell signaling via interactions with the SLP-76 and LAT adaptors. Curr. Biol. 9, 67 –75[CrossRef][Medline]

  13. Yablonski, D., Kadlecek, T., and Weiss, A. (2001) Identification of a phospholipase C-{gamma}1 (PLC-{gamma}1) SH3 domain-binding site in SLP-76 required for T-cell receptor-mediated activation of PLC-{gamma}1 and NFAT. Mol. Cell. Biol. 21, 4208 –4218[Abstract/Free Full Text]

  14. Zhang, W., Trible, R. P., Zhu, M., Liu, S. K., McGlade, C. J., and Samelson, L. E. (2000) Association of Grb2, Gads, and phospholipase C-{gamma}1 with phosphorylated LAT tyrosine residues. Effect of LAT tyrosine mutations on T cell antigen receptor-mediated signaling. J. Biol. Chem. 275, 23355 –23361[Abstract/Free Full Text]

  15. Anderson, S. M., Burton, E. A., and Koch, B. L. (1997) Phosphorylation of Cbl following stimulation with interleukin-3 and its association with Grb2, Fyn, and phosphatidylinositol 3-kinase. J. Biol. Chem. 272, 739 –745[Abstract/Free Full Text]

  16. Lin, J., and Weiss, A. (2001) Identification of the minimal tyrosine residues required for linker for activation of T cell function. J. Biol. Chem. 276, 29588 –29595[Abstract/Free Full Text]

  17. Gillis, S., and Watson, J. (1980) Biochemical and biological characterization of lymphocyte regulatory molecules. V. Identification of an interleukin 2-producing human leukemia T cell line. J. Exp. Med. 152, 1709 –1719[Abstract/Free Full Text]

  18. Straus, D. B., and Weiss, A. (1992) Genetic evidence for the involvement of the lck tyrosine kinase in signal transduction through the T cell antigen receptor. Cell 70, 585 –593[CrossRef][Medline]

  19. Irvin, B. J., Williams, B. L., Nilson, A. E., Maynor, H. O., and Abraham, R. T. (2000) Pleiotropic contributions of phospholipase C-{gamma}1 (PLC-{gamma}1) to T-cell antigen receptor-mediated signaling: reconstitution studies of a PLC-{gamma}1-deficient Jurkat T-cell line. Mol. Cell. Biol. 20, 9149 –9161[Abstract/Free Full Text]

  20. Secrist, J. P., Burns, L. A., Karnitz, L., Koretzky, G. A., and Abraham, R. T. (1993) Stimulatory effects of the protein tyrosine phosphatase inhibitor, pervanadate, on T-cell activation events. J. Biol. Chem. 268, 5886 –5893[Abstract/Free Full Text]

  21. Lock, L. S., Royal, I., Naujokas, M. A., and Park, M. (2000) Identification of an atypical Grb2 carboxyl-terminal SH3 domain binding site in Gab docking proteins reveals Grb2-dependent and -independent recruitment of Gab1 to receptor tyrosine kinases. J. Biol. Chem. 275, 31536 –31545[Abstract/Free Full Text]

  22. Stoevesandt, O., Köhler, K., Fischer, R., Johnston, I. C., and Brock, R. (2005) One-step analysis of protein complexes in microliters of cell lysate. Nat. Methods 2, 833 –835[CrossRef][Medline]

  23. Puntervoll, P., Linding, R., Gemund, C., Chabanis-Davidson, S., Mattingsdal, M., Cameron, S., Martin, D. M. A., Ausiello, G., Brannetti, B., Costantini, A., Ferre, F., Maselli, V., Via, A., Cesareni, G., Diella, F., Superti-Furga, G., Wyrwicz, L., Ramu, C., McGuigan, C., Gudavalli, R., Letunic, I., Bork, P., Rychlewski, L., Kuster, B., Helmer-Citterich, M., Hunter, W. N., Aasland, R., and Gibson, T. J. (2003) ELM server: a new resource for investigating short functional sites in modular eukaryotic proteins. Nucleic Acids Res. 31, 3625 –3630[Abstract/Free Full Text]

  24. Jones, R. B., Gordus, A., Krall, J. A., and MacBeath, G. (2006) A quantitative protein interaction network for the ErbB receptors using protein microarrays. Nature 439, 168 –174[CrossRef][Medline]

  25. Stiffler, M. A., Grantcharova, V. P., Sevecka, M., and MacBeath, G. (2006) Uncovering quantitative protein interaction networks for mouse PDZ domains using protein microarrays. J. Am. Chem. Soc. 128, 5913 –5922[CrossRef][Medline]

  26. Levchenko, A., Bruck, J., and Sternberg, P. W. (2000) Scaffold proteins may biphasically affect the levels of mitogen-activated protein kinase signaling and reduce its threshold properties. Proc. Natl. Acad. Sci. U. S. A. 97, 5818 –5823[Abstract/Free Full Text]

  27. Vogel, W., and Ullrich, A. (1996) Multiple in vivo phosphorylated tyrosine phosphatase SHP-2 engages binding to Grb2 via tyrosine 584. Cell Growth Differ. 7, 1589 –1597[Abstract]

  28. Houtman, J. C., Houghtling, R. A., Barda-Saad, M., Toda, Y., and Samelson, L. E. (2005) Early phosphorylation kinetics of proteins involved in proximal TCR-mediated signaling pathways. J. Immunol. 175, 2449 –2458[Abstract/Free Full Text]

  29. Neduva, V., Linding, R., Su-Angrand, I., Stark, A., De Masi, F., Gibson, T. J., Lewis, J., Serrano, L., and Russell, R. B. (2005) Systematic discovery of new recognition peptides mediating protein interaction networks. PLoS Biol. 3, e405[CrossRef][Medline]


Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
M600383-MCP200v1
6/3/503    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Glossary
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google