• Aucun résultat trouvé

Active kinase proteome screening reveals novel signal complexity in cardiomyopathy

N/A
N/A
Protected

Academic year: 2021

Partager "Active kinase proteome screening reveals novel signal complexity in cardiomyopathy"

Copied!
11
0
0

Texte intégral

(1)

Publisher’s version / Version de l'éditeur:

Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n’arrivez pas à les repérer, communiquez avec nous à PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca.

Questions? Contact the NRC Publications Archive team at

PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca. If you wish to email the authors directly, please see the first page of the publication for their contact information.

https://publications-cnrc.canada.ca/fra/droits

L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.

Molecular & Cellular Proteomics, 4, 5, pp. 673-682, 2005

READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THIS WEBSITE.

https://nrc-publications.canada.ca/eng/copyright

NRC Publications Archive Record / Notice des Archives des publications du CNRC :

https://nrc-publications.canada.ca/eng/view/object/?id=f1d21755-f192-4553-a401-9f8b7ba3ad65

https://publications-cnrc.canada.ca/fra/voir/objet/?id=f1d21755-f192-4553-a401-9f8b7ba3ad65

Archives des publications du CNRC

This publication could be one of several versions: author’s original, accepted manuscript or the publisher’s version. / La version de cette publication peut être l’une des suivantes : la version prépublication de l’auteur, la version acceptée du manuscrit ou la version de l’éditeur.

For the publisher’s version, please access the DOI link below./ Pour consulter la version de l’éditeur, utilisez le lien DOI ci-dessous.

https://doi.org/10.1074/mcp.M400200-MCP200

Access and use of this website and the material on it are subject to the Terms and Conditions set forth at

Active kinase proteome screening reveals novel signal complexity in

cardiomyopathy

Fernando, Pasan; Deng, Wen; Pekalska, Beata Pekalska; DeRepentigny,

Yves; Kothary, Rashmi; Kelly, John F.; Megeney, Lynn A.

(2)

Active Kinase Proteome Screening Reveals

Novel Signal Complexity in Cardiomyopathy*

S

Pasan Fernando‡, Wen Deng§, Beata Pekalska‡, Yves DeRepentigny‡,

Rashmi Kothary‡¶, John F. Kelly§, and Lynn A. Megeney‡¶

Recent advances in the characterization of the phospho-proteome have been limited to measuring phosphoryla-tion statuses, which imply but do not measure protein kinase activity directly. As such, the ability to screen, compare, and define multiple protein enzymatic activities across divergent samples remains a daunting challenge in proteomics. Here, we describe a gel-based kinase assay coupled to MS identification as an approach to map global kinase activity and assign pathway architecture to spec-ified biologic contexts. We demonstrate the utility of this method as a platform for the comparison of proteomes based on differences in both kinase activities and for use in the de novo substrate identification for individual ki-nases. This approach allowed us to map the signal per-turbations in the post-natal heart that were associated with activation of a myopathic cascade as mediated by the mitogen-activated protein kinase MKK6 and estab-lished the novel observation that MKK6 promotes the development of cardiomyopathy through multiple sub-strate interactions. Molecular & Cellular Proteomics 4: 673– 682, 2005.

Serial modification of proteins is a common mechanism for eukaryotic cellular adaptation. Within this context, protein phosphorylation is believed to be the most common form of modification. Many excellent methods describing the purifi-cation and identifipurifi-cation of phospho-proteins are now avail-able and have significantly contributed to our understanding of cellular regulation through signal transduction (1– 4). Nev-ertheless, phosphorylation status alone does not unambigu-ously characterize the kinetic steps of signaling cascades, namely the activity of protein kinases. Indeed, there are an estimated 519 kinase-related gene products in the human genome (representing ⬃3% of the coded proteome) with a unique substrate range for each kinase (5). As such, success-fully mapping these steps in any biologic model remains a daunting challenge for proteomics.

In addition to the limitation imposed by current screening methodologies, the accurate reconstruction of signaling net-works may also be hindered by the prevailing hypothesis that describes signaling cascades as insulated and linear events. For example, the annotation of mitogen-activated protein ki-nase (MAPK)1pathways describes a three-tier modular

struc-ture involving sequential activation from module to module (6). The p38 MAPK pathway is a typical representation of this kinase family. This pathway contains a phosphorylation se-quence initiated by a MAPK kinase kinase (MKKK), which activates a MAPK kinase (MKK), which then in turn activates a MAPK (p38), with the MAPK then serving as the effector enzyme to stimulate or repress the activity of corresponding protein substrates by targeted phosphorylation (7).

Considerable experimental evidence supports the basic premise of the linear MKK/p38 MAPK pathway, yet ascribing all phenotypic outcomes exclusively to the activity of the effector kinase is most likely an oversimplification. In this regard, the relationship between activation of p38 signal cas-cades and the development of cardiac pathology (hypertro-phy and dilatation) provides an interesting venue in which to model signaling events and their biologic response. To date, numerous studies have implicated p38 activation as a requi-site step in the development of post-natal cardiac dilatation and or hypertrophy (8). In a portion of these studies, a clear relationship was established for p38 and its responsive tran-scription factor substrates, the substrates being directly vali-dated as p38 targets and as hypertrophic agonists (9). Con-versely, other models were predicated on the use of MKK6 as a selective activation source for p38 kinases (10, 11). As predicted, these later models displayed phenotypic alter-ations of the heart, yet the changes were solely attributed to MKK6 activation of p38.

Here, we report the application of a novel proteomic screening assay to interrogate the complexity of MAPK-in-duced cardiac adaptation. This assay combined the utility of solid-phase kinase assays with the resolving power of nano-flow-LC-MS/MS. We have termed this assay kinase sweep

From the ‡Molecular Medicine Program, Ottawa Health Research Institute, Ottawa Hospital, Ottawa, Ontario K1H 8L6, Canada; §Insti-tute for Biologic Sciences, National Research Council, Ottawa, On-tario K1A 0R63, Canada; and ¶Department of Cellular and Molecular Medicine, University of Ottawa, Ottawa K1H 8M5, Ontario, Canada

Received, December 12, 2004, and in revised form, February 17, 2005

Published, MCP Papers in Press, February 18, 2005, DOI 10.1074/mcp.M400200-MCP200

1The abbreviations used are: MAPK, mitogen-activated protein

kinase; MKKK, MAPK kinase kinase; MKK, MAPK kinase; KSE, kinase sweep; KSS, kinase substrate; MBP, myelin basic protein; 2D, two-dimensional; PAK, p21 activated kinase; AMPK, 5⬘-AMP activated kinase; RAK, Rho-associated kinase; IP, immunoprecipitation; CRMP, collapsin response mediator protein; PKN, protein kinase N.

Research

© 2005 by The American Society for Biochemistry and Molecular Biology, Inc.

Molecular & Cellular Proteomics 4.5

673

This paper is available on line at http://www.mcponline.org

(3)

(KSE)/MS. Using this assay with gel-embedded substrates, we identified a broad range of kinase activities in lysates derived from the hearts of transgenic mice expressing acti-vated MKK6. Second, we used a modification in the assay which we termed kinase substrate (KSS)/MS, to identify a range of probable substrates for MKK6, suggesting that this kinase may influence hypertrophic adaptation through a com-plex signaling architecture. Together, these results provide a proof of concept that KS/MS (KSE/MS and KSS/MS) is appli-cable for large scale identification and analyses of both pro-tein kinase activities and their effective substrate range.

EXPERIMENTAL PROCEDURES

Transgenic Mice—The full-length murine MKK6 gene carrying

mu-tations at S207E and T211E (MKK6EE) was subcloned into a pBlue-script backbone containing a murine␣-myosin heavy chain promoter (a gift from Jeffrey Robbins, Children’s Hospital, Cincinnati, OH) and a 600-bp fragment of the human growth hormone poly(A) tail. The transgenic mice were derived as described previously (12).

Tissue Lysis—Hearts from MKK6EE and wild-type mice were

ex-cised and snap-frozen in liquid nitrogen. Total protein was extracted in modified RIPA buffer (50 mMTris-HCl, pH 7.4, 150 mMNaCl, 1 mM

EDTA, 1% Nonidet P-40, 1% glycerol) containing 1.0␮g/ml aprotinin, 0.5␮g/ml leupeptin, 0.7 ␮g/ml pepstatin, and 50 ␮g/ml PMSF and supplemented with 20 mMNaF and 5 mMsodium orthovanadate. Tissues were Dounce homogenized on ice and rotated at 4° C for 30 min. Samples were centrifuged at 10,000⫻ g, and soluble fractions were stored at – 80° C. Total protein was measured using the bicin-choninic acid protein assay (Pierce, Rockford, IL).

IEF and Gel-based Kinase Assays—Samples were prepared for IEF

using standard protocols as previously described (13). Three hundred to 400 ␮g of total protein was separated on immobilized pH 4–7 gradient strips (Bio-Rad, Hercules, CA).

KSE/MS—For identification of protein kinases, 0.1 mg/ml myelin

basic protein (MBP) was mixed with a 10 –12% polyacrylamide solu-tion (30%T:2.6%C). In some instances, 0.1 mg/ml histone H1 or both MBP and histone H1 (0.05 mg/ml of each) were used. Gel-based kinase assays were performed as described previously for one-di-mensional SDS-PAGE (14) with several modifications. All steps were performed at room temperature unless otherwise indicated. Following electrophoresis, gels were washed in three changes of a propanol-Tris buffer (50 mMTris-HCl, pH 8.0, 20% 2-propanol). Gels were then soaked in three changes of a DTT-Tris buffer (50 mMTris-HCL, pH 8.0, 5.0 mMDTT). A denaturing solution (50 mMTris, pH 8.0, 20 mM

DTT, 6.0Mguanidine-HCl) facilitated protein unfolding within the gel. Gel-bound proteins were refolded in a protein folding buffer (50 mM

Tris-HCl, pH 8.0, 5.0 mMDTT, 0.04% Tween-40, 150 mMNaCl, and 5.0 mMMgCl2). The solution was changed four times and then once

prior to an overnight incubation at 4 °C. Subsequent changes of protein folding buffer were carried out the following day (8 –10 times). Prior to the kinase reaction, gels were soaked in two changes of pre-kinase reaction buffer (20 mMMOPS, pH 7.4, 25 mM

␤-glycero-phosphate, 5 mMEGTA, 1 mMDTT, 1 mMsodium orthovanadate, and 50 mMMgCl2) for 30 min each. Gels were then incubated in a kinase

buffer (20 mMMOPS, pH 7.4, 25 mM ␤-glycerophosphate, 5 mM

EGTA, 1 mMDTT, 1 mMsodium orthovanadate including 5␮Ci/␮l [␥-32P]ATP) and 500

Mcold ATP for 1.5–2 h. Incubation was per-formed with occasional agitation. In order to minimize nonspecific background and precipitate incorporated radiolabel, gels were washed with several changes of a TCA wash solution (5% TCA, 1% sodium pyrophosphate). Finally, the gels were air-dried (Bio-Rad), and kinases were detected by autoradiography.

Autophosphorylated proteins were excluded from analyses by run-ning replicate gels that did not contain embedded substrate. The radiolabeled spots appearing on these gels were compared with spots on gels that contained embedded substrate. Similar protein spots from both gels did not undergo further analysis.

KSS/MS—A 10 to 12% polyacrylamide gel (30%T:2.6%C) without

copolymerized substrate was used for the substrate-sweep/MS anal-yses. In this assay, substrates for the kinase of interest are endoge-nous to the sample. Following the second-dimension electrophoresis, gels were treated as described for the KSE/MS. After incubation in a pre-kinase buffer as above, the gel was soaked for 1.5–2.0 h in a kinase buffer containing a commercial recombinant activated kinase (MKK6; Upstate Biotechnology, Lake Placid, NY). The gel was then washed and treated as above.

In-gel Tryptic Digests and MS—Autoradiograms from gel-based

kinase assays were matched with their corresponding dried gels. Picked spots were excised and re-swollen in 50 mMNH4HCO3. Gel

pieces were reduced, alkylated, and processed for in-gel tryptic di-gestion using standard protocols (15). All nano-HPLC-ESI MS/MS experiments were conducted using a Q-TOF UltimaTMhybrid Q-TOF

mass spectrometer coupled with a CapLC™ capillary LC system (Waters, Millford, MA). Samples were first loaded onto a 0.5-cm⫻ 300-␮m C18peptide trap (LC-Packings, San Francisco, CA). Trapped

peptides were eluted off to a 15-cm⫻ 75-␮m C18capillary column for

separation using a gradient of 5–50% ACN over 45 min and 50 – 85% ACN over 5 min. The flow rate to the ESI sprayer was 250 nl/min. The mass spectrometer was set to operate in automated MS/MS acqui-sition mode during the analysis. In order to increase the probability of detecting isolated kinases with the KSE/MS method, MBP parent ions were excluded from analyses using an in silico-generated MBP ion exclusion list. The peptide MS/MS spectra were searched against the NCBInr database using the MASCOT searching algorithm.

Immunoprecipitation/in Vitro Kinase and Immunoblot

Assays—Im-mune-complex kinase analyses were performed using tissue protein lysates from wild-type and MKK6EE mice as described previously (13). Western transfer and immunoblot analyses were performed as described previously (13). Antibodies against MKK6 (Calbiochem, La Jolla, CA), p38␣, p38␥ (Upstate Biotechnology), AMPK␤1, AMPK␣, and PAK1 (Cell Signaling Technologies, Beverly, MA), and pan-phos-phoserine (Zymed Laboratories Inc., San Francisco, CA; Chemicon, Temecula, CA; Abcam, Cambridge, MA; and Cell Signaling Technol-ogies) were used as indicated by the manufacturer.

RESULTS

To begin to address the depth of the MKK6/p38 signaling cascade, we chose to utilize the post-natal heart as a mod-eling environment. To this end, we generated transgenic mice overexpressing an activated MKK6 transgene (MKK6EE). Post-natal cardiac expression was achieved by using the ␣-myosin heavy chain promoter (16, 17) (Fig. 1D). Although MKK6EE mice appeared normal at birth, these animals had attenuated lifespans (⬍2 months). Post-mortem analysis of MKK6EE mice revealed evidence of a significant cardiac my-opathy, with enlarged ventricular walls and a pronounced increase in ventricular chamber size in comparison to wild-type littermates (Fig. 1, A and B). In addition, the fiber diam-eter from MKK6EE hearts were larger than those from wild-type littermates (Fig. 1C). Interestingly, we observed that overexpression of activated MKK6EE resulted in the prefer-ential induction of p38␣ kinase activity with little or no alter-ation in p38␥ despite similar levels of protein expression (Fig.

(4)

1, E–G). Indeed, variable MKK6 activation has been demon-strated to alter the substrate preference of MKK6 between the targets of p38␣ and p38␥, an interaction that originates from the more efficient binding of MKK6 to p38␣ in comparison to p38␥ (18).

These results indicated that the signaling pathways in-volved in MKK6 activation were more complex than initially anticipated. As such, it was evident that the standard

predic-tive models for interrogating MAPK signaling were limited, and the use of broad-range screening tools were required to identify the relevant kinases and phosphoproteins. As a first step in screening the pathways, we performed standard two-dimensional (2D) gel separation of cardiac protein lysates derived from wild-type and MKK6EE transgenic mice followed by Western blot analysis of these gel profiles using a number of pan-phosphoserine antibodies, i.e. antibodies raised to a

FIG. 1. Characterization of the MKK6EE phenotype. A, whole mount view of wild-type (left) and MKK6EE (right) hearts at 2 months of age.

The MKK6EE heart displays enlarged atria and ventricular hypertrophy. B, cross section of MKK6EE heart. Distention of the left ventricle is accompanied by thickening of the ventricular walls. Section was photographed at⫻0.63 magnification. C, longitudinal sections of wild-type heart (left) and MKK6EE heart (right). Individual fibers from MKK6EE hearts appear larger than fibers from wild-type hearts. Sections were photographed at⫻60 magnification. D, MKK6 protein levels are higher in MKK6EE hearts. Western blot analysis of wild-type (wt) and MKK6EE (tg) heart lysates using an anti-MKK6 antibody. The blot is representative of the MKK6EE transgenic strain. E, MKK6 activity is higher in MKK6 EE transgenic hearts. MKK6 kinase activities from wt and tg tissue lysates were compared by IP/kinase analyses. MKK6 was immunopre-cipitated with an anti-MKK6 antibody and incubated in a kinase reaction buffer containing [32P]ATP and MBP. Kinase activity was assessed

by autoradiography. Control reactions with (⫹) or without (–) recombinant activated MKK6 or from the skeletal myobast cells (C2) are shown.

F, p38␥ activity is lower in MKK6EE hearts. p38␥ kinase activities were assessed as in D using an anti-p38␥ antibody. Reactions without MBP

(–) and with recombinant activated p38␥ (⫹) are shown as controls. G, p38␣ activity is higher in MKK6EE hearts. p38␣ kinase activities were assessed as in D using an anti-p38␣ antibody. A control reaction using recombinant activated p38␣ (⫹) was also included. For E–F, the position equal loading of each kinase reaction was assessed with Coomassie blue (shown below each panel).

Activity-based Proteomic Screening of Cardiomyopathy

(5)

phosphoserine epitope in general. This screening approach demonstrated observable differences between the conditions (see supplemental Fig. 1), yet the reproducibility of this ap-proach was subject to considerable variability. The variability originated in part from the pan-phosphoserine antibodies, i.e. antigen detection appeared to be dependent on the class of phosphoserine proteins examined and did not provide global or uniform coverage of serine phosphoproteins (data not shown). In addition, the information acquired from this screen like other proteomic tools did not provide a direct measure of protein activities.

Based on these limitations, we sought to develop a repre-sentative proteomic screen in which global kinase activity could be coupled to kinase identification. We had previously shown that innate kinase activities of muscle protein lysates could be retained following IEF and SDS-PAGE separation (13). Therefore, we postulated that additional alterations to this preparative protein separation could be made to maxi-mize kinase activities and enhance identification through the use of LC-MS/MS (Fig. 2). Protein lysates were subject to IEF and SDS-PAGE with a gel matrix consisting of a ubiquitous kinase substrate cross-linked to the matrix, in this case MBP. The protocol was enhanced with the use of a chaotropic buffer designed to encourage protein re-folding and mainte-nance of enzymatic activity. These gels were then subject to a modified kinase assay and then autoradiography to identify matrix-bound kinase activities. Absolute identification of the speculative kinase could then be made using LC-MS/MS (Fig. 2). We have termed this assay KSE/MS. As a proof of principle experiment, we subjected active MKK6 and p38 fusion pro-teins to the KSE/MS procedure with 2D-PAGE gels containing MBP. We observed proteins in these gels with mass and isoelectric points within ranges consistent for recombinant MKK6 and p38, which retained robust kinase activity (Fig. 3, A and B). These protein spots were then cut from the gel, trypsin-digested, and positively identified by LC-MS/MS as MKK6 and p38 (Fig. 3, A and B). Importantly, the MS analysis was conducted using an MBP exclusion list to circumvent any signal squelching that may have arisen from the excess gel-bound MBP.

Following the successful reconstitution of these recombi-nant kinases in the KSE/MS, we then tested the utility of this method for monitoring multiple kinase activities from tissue protein lysates. Total protein lysates were extracted from wild-type and MKK6EE hearts, and comparative kinase activ-ity maps were generated (Fig. 4A). LC-MS/MS identification of these proteins demonstrated that the wild-type heart ap-peared to maintain a kinase activity profile consistent with a baseline cellular function, displaying measurable kinase ac-tivity for intermediate metabolic enzymes (such as protein kinase C and pyruvate kinase) and cytoskeletal organizers p21 activated kinase (PAK) and a kinase activity associated with the cellular chaperone 14-3-3) (Fig. 4, A and B). The kinase activity profile that was generated from MKK6EE

pro-tein lysates was clearly divergent from the profile observed using wild-type lysates (Fig. 4A). For example, in addition to elevation of MKK6 and p38␣ activity, MKK6EE lysates also demonstrated robust enzymatic activity for three kinases of particular interest, 5⬘-AMP activated kinase (AMPK), Rho-associated kinase (RAK), and the serine/threonine kinase pro-tein kinase N (PKN) (Fig. 4). To validate the observations obtained from the 2D-KS/MS screen, we made use of stand-ard immunoprecipitation (IP) kinase assays. Individual kinases were immunoprecipitated from wild-type and MKK6EE car-diac protein lysates and incubated with MBP in a comparative

in vitro kinase assay. The results of the IP kinase assay were

entirely consistent with the results obtained in the initial KSE/MS assay for each protein tested (data only shown for PAK and AMPK), which confirmed the utility of KSE/MS in identifying multiple kinase activities from one sample and constructing comparative kinase activity maps (Fig. 4C).

The KSE/MS screen yielded an enzymatic activity map that was consistent with the physiologic demands for cardiac hypertrophy. Nevertheless, many of these responding kinases do not contain the requisite interaction domains that are typ-ical hallmarks of MAPK substrates (19). It is probable then that a portion of these kinases were responding to the hypertrophy environment and not MKK6 directly. Therefore to distinguish primary MKK6/p38␣ signaling alterations from secondary changes, we adapted our kinase assay to identify probable kinase substrates for MKK6 (KSS/MS). Briefly, this KS/MS variation utilized IEF-2D-separated protein lysates as the ki-nase substrate, i.e. second-dimension gel without MBP (Fig. 2b). These gels were then incubated with a recombinant ac-tivated kinase in an appropriate incubation buffer (Fig. 2), followed by MS identification of the [␥-32P]ATP-labeled

pro-tein spots. Subsequently, we utilized the KSS/MS variation with recombinant MKK6 and p38␣ to identify corresponding substrates in cardiac protein lysates.

The p38␣-based KSS/MS resulted in direct phosphoryla-tion of and identificaphosphoryla-tion for multiple known substrates (data not shown), an established characteristic for this effector ki-nase during cardiac hypertrophy (17). Not surprisingly, MKK6-based KSS/MS experiments identified p38␣, p38␦, and p38␥ (low level) kinases as targeted substrates (Fig. 5, A and B). However, these experiments also identified other probable targets for MKK6 in cardiac protein lysates, including cy-toskeletal regulatory factors and translational control proteins (Fig. 5, A and B). Interestingly, we also identified eIF-4E and tubulin, proteins known to be phosphorylated by kinases (Fig. 5, C and D). Phosphorylation status for these MKK6 targets were independently confirmed using a combination of anti-phospho-specific antibodies for IP/Western and Western blotting analysis in wild-type and MKK6EE heart lysates.

DISCUSSION

The phosphorylation of cellular proteins represents an im-portant mechanism for eukaryotic cellular adaptation. The

(6)

biochemical pathways that mediate these adaptive responses have become a central focus in signaling biology. Several broad-based screening methods, now available for the iden-tification of phospho-proteins, have established their utility in as far as revealing the outcome of cellular regulation. How-ever, the current approaches that recount signaling events

often rely on methods that depict only linear and isolated outcomes. Clearly, illustrating the dynamic nature of signal transduction cascades remains a difficult and onerous chal-lenge. In this report, we describe an activity-dependent pro-teomic screening assay used to interrogate the complexity of MAPK signaling. Using this assay, we demonstrate that

FIG. 2. The KS/MS assay. A schematic diagram depicting the essential steps of the KSE/MS assay. Samples are IEF on immobilized pH gradient of a chosen pI range (a). The focused sample is loaded onto a denaturing polyacrylamide matrix containing an embedded kinase substrate of choice (e.g. MBP, histone-H1) (b). Following electrophoresis, the gel is washed in a Tris-propanol buffer to remove SDS (c). Proteins within the gel are denatured in denaturing buffer (d) and then renatured in a folding buffer (e). The gel is incubated in a kinase buffer containing [32P]ATP (f) and washed to remove unincorporated radiolabel (g). Autoradiographic exposure of the dried gel reveals a number of

spots of interest (h), where each spot represents a putative kinase that has phosphorylated the in-gel substrate. The spots of interest are excised from the gel, subjected to tryptic digest, and further purified for mass spectrometric analyses using nano-LC-ESI-MS (i).

Activity-based Proteomic Screening of Cardiomyopathy

(7)

MAPK signals in cardiac disease can behave in an unconven-tional manner.

Broad spectrum analysis of protein kinase activity using gel-based methods has been reported previously (14, 20). In these instances, proteins were subjected to one-dimensional separation based on molecular mass through a substrate-embedded polyacrylamide matrix. Upon exposing the gel to denaturation/renaturation and kinase buffers, single bands representing protein kinases were visualized. However, the resolution of this assay provides only qualitative information because several protein kinases may share the same molec-ular mass. Using an improved gel-based approach, we incor-porated a high-resolution 2D polyacrylamide separation cou-pled with mass spectrometric interrogation to unambiguously identify protein kinases, termed KSE/MS. Importantly, these modifications allowed us to draw more enumerative conclu-sions between test samples.

An activated MKK6 murine model of cardiomyopathy re-vealed several interesting and unanticipated results following the application of KSE/MS. Aside from the expected

activa-tion of downstream signaling components, i.e. p38 MAPK, we identified kinases that affect cellular metabolism, cytoskeletal dynamics, and cell survival. In general terms, it is known that AMPK acts as an independent rate-limiting step in oxidative metabolic pathways. More interestingly in the present con-text, activation of AMPK has been implicated in metabolic regulatory events during the development of cardiac hyper-trophy (21). Similarly, elevated RAK activity has been shown to be a crucial component of hypertension-induced left ven-tricular hypertrophy (22, 23), and PKN has been shown to mediate Rho GTPase activation of transcriptional activity leading to hypertrophy in cardiomyocytes (24). In addition to the cardiac-specific effects, both RAK and PKN are known to be modulators of the cytoskeleton.

An important consideration of the KSE/MS method is the choice of substrate that is co-polymerized within the second-dimension gel. While MBP is a general substrate for a wide variety of serine/threonine protein kinases (25, 26), a number of kinases have alternate substrate preferences. In this in-stance, the KSE/MS method may be best utilized with a

FIG. 3. Validation of the KS/MS assay. A, recombinant active Mal-E-MKK6 was subjected to KS/MS analysis. A representative autora-diogram is shown. The spot of interest (circled in red) was tryptic digested and prepared for MS analysis. The corresponding MS spectra for the double-charged parent ion is shown on the right. B, recombinant active GST-p38␥ was subjected to KS/MS analyses as in A. The resulting spectra (parent ion, double charge) from an in-gel tryptic digest of the circled spot (indicated in red) is shown on the right. For both A and B, the peptide MS/MS spectra were searched against public databases using the MASCOT (Matrix Science, London, UK) database searching algorithm.

(8)

multisubstrate gel matrix. For example, when we co-polymer-ized both MBP and histone H1 in combination, a larger num-ber of kinase activities were detected. By using the two gen-eral substrates together, we were able to cover a more extensive kinase substrate range (see Supplemental Fig. 2).

A novel feature of the KS/MS approach is the ability to identify substrates of a given protein kinase, termed KSS/MS. By separating IEF samples in a substrate-free second-dimen-sion gel and incorporating a recombinant active protein ki-nase in the kiki-nase incubation buffer, we were able to identify several substrates of MKK6. Several of these substrates are known to be targeted by kinases (eIF-4E, tubulin, etc.), yet this is the first demonstration to imply that these same proteins act as bona fide MKK6 substrates (Fig. 5, C and D).

Impor-tantly, many of these proteins are known to be involved in cytoskeletal reorganization and have been reported to retain cardiac-specific functions. For example,␣-adducin is a spec-trin-like protein that modulates ankryin interactions with actin (27) and has been linked to hypertension-associated cardiac hypertrophy (28). Semaphorins are secreted glycoproteins that bind and activate a family of transmembrane receptor proteins termed plexins. Semaphorins were originally identi-fied as critical axon guidance factors through deconstruction of the actin cytoskeleton (29). Interestingly, semaphorin 6 has been shown to have profound effects on cardiac morphogen-esis by limiting the expansion of ventricular mass, presumably by inducing cardiomyocyte cell contraction (30, 31). Finally, it is germaine to note that collapsin response mediator

pro-FIG. 4. KS/MS profiling of protein kinases from wild-type and MKK6EE cardiac lysates. A, KSE/MS analyses of wild-type and MKK6EE cardiac lysates revealed distinct differences in kinase activities. Cardiac lysates were prepared and subjected to the KSE/MS assay using MBP as a gel-bound substrate. MKK6 (1), p38 (2), and cAMP-dependent protein kinase (3) are shown for the MKK6EE analysis. B, a partial list of protein kinases identified by KSE/MS in wild-type and transgenic cardiac lysates. MS results were searched against the NCBI databases using MASCOT database search algorithms. C, validation of KSE/MS targets using in vitro kinase assays. Adenosine monophosphate-activated protein kinase (AMPK) (left) and p21 activated kinase (PAK) (right) was immunoprecipitated from wild-type (wt) and MKK6EE transgenic (tg) heart and skeletal muscle protein lysates. Kinase activity using MBP as the target substrate was assessed as described in Fig. 1. A positive control containing recombinant activated AMPK was included (input). D, representative spectra for PAK (top) and serine/threonine protein kinase PKN (bottom) are shown. Spectra were generated by searching MS results against NCBI databases using the MASCOT database search algorithms. The peptide sequence and parent ion mass is shown for each. Additional spectra for other identified peptides are shown in supplemental Fig. 3.

Activity-based Proteomic Screening of Cardiomyopathy

(9)

tein-1 (CRMP-1) was also identified as an MKK6 substrate. CRMP proteins act to mediate semaphorin signals, and CRMP activity is known to be suppressed by phosphoryla-tion, at least in neuronal cell types (32). As such, it is reason-able to suggest that MKK6 may promote the progression of cardiomyocyte hypertrophy by targeted phosphorylation of these proteins, and it is the phosphorylation and consequent inhibition of these proteins that permits the characteristic re-organization of the cytoskeleton during hypertrophy (33).

The prevailing hypothesis for MAPK-induced cardiac hy-pertrophy describes a simple linear relationship in which the

effector kinase acts as the final arbiter of the hypertrophic signal by up-regulating cardiac transcription factor activity (34, 35). This model presumes that the cytoskeletal and sar-comereic reorganization that takes place during cardiomyo-cyte hypertrophy is the consequence of effector kinase func-tion. Our results challenge this model and suggest that MAPK induction of cardiac hypertrophy originates from both effector kinase-dependent and -independent mechanisms, i.e. indi-rect activation of protein/substrate activity.

The application of KS/MS relies on the efficient separation of proteins by isoelectric point and then by molecular mass.

FIG. 5. KS/MS identification of substrates for MKK6. A, cardiac protein lysates from MKK6EE mice were subject to KSS/MS analyses. The

second-dimension PAGE was run with the lysates as the substrate. Activated MKK6 was included in the kinase reaction buffer (see Fig. 2). Numbers correspond to proteins identified by MS analyses (B). B, a partial list of potential MKK6 substrates identified in A. Peptide fragments were searched against the NCBI databases using MASCOT database search algorithms. Asterisk indicates protein spots not shown on the gel (A). C, phosphorylation of tubulin by MKK6. The identified substrate tubulin was immunoprecipitated from MKK6EE transgenic cardiac lysates and subjected to an in vitro kinase assay using recombinant active MKK6 and then Western transferred to a membrane support. Increasing amounts of recombinant active enzyme were used to validate phosphorylation of tubulin by MKK6 (shown by autoradiography). A control lane showing tubulin without active MKK6 is shown. The blot was then immunostained to show comparative levels of immunoprecipitated tubulin.

D, the identified substrate adducin was verified as an MKK6 substrate as in C. An equivalent level of immunoprecipitated adducin was verified

by an anti-adducin immunoblot (lower panel). E, representative spectra for adducing (top) and tubulin (bottom). The parent ion mass and peptide sequence are also shown for each. Additional spectra for other identified peptides are shown in supplemental Fig. 4.

(10)

Although the resolution of an entire proteome using gel-based separation is still a challenge, these methods remain versatile, dependable, and cost efficient. More recently, significant ad-vances in proteomic profiling techniques have been made with gel-free methods of sample separation followed by LC-MS (36 –38). These approaches have the advantage of enhanced resolution through comprehensive sample analyses including the detection of low abundance proteins and pro-teins within extreme isoelectric point ranges. We are currently making efforts to improve the resolution of the KS/MS meth-odologies by incorporating gel-free matrices coupled to LC separation.

To date, proteomic applications in signaling biology have been limited to annotation of the phosphoproteome. These protocols are based on one general strategy; the enrichment of phosphorylated proteins/peptides from a protein lysate followed by mass spectrometric identification of the altered proteins/peptides. Despite the promise of these innovations, assigning true functional outcomes to the resulting phospho-proteome maps has remained elusive. Our observations have demonstrated that the KS/MS protocol circumvents this limi-tation, coupling signal protein function (kinase activity) to protein identification. As such, application of the KS/MS methodologies will provide a platform for accelerating signal-ing discovery in complex biologic models.

Acknowledgment—We thank Nathalie Ahn for helpful discussions.

* P. F. was supported by a fellowship from the Heart and Stroke Foundation of Canada (HSFC). L. A. M. is a scholar of the Canadian Institutes of Health Research (CIHR). This work was supported by grants from CIHR (to J. F. K., R. K. and L. A. M.), HSFC (to L. A. M.), and Genome Canada (to L. A. M.). 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 accord-ance with 18 U.S.C. Section 1734 solely to indicate this fact.

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

储 To whom correspondence should be addressed: Ottawa Health Research Institute, Ottawa Hospital, General Campus, 501 Smyth Rd., Ottawa, Ontario K1H 8L6, Canada. Tel.: 613-737-8618; Fax: 613-737-8803; E-mail: lmegeney@ohri.ca.

REFERENCES

1. Knight, Z. A., Schilling, B., Row, R. H., Kenski, D. M., Gibson, B. W., and Shokat, K. M. (2003) Phosphospecific proteolysis for mapping sites of protein phosphorylation. Nat. Biotechnol. 21, 1047–1054

2. Affolter, M., Watts, J. D., Krebs, D. L., and Aebersold, R. (1994) Evaluation of two-dimensional phosphopeptide maps by electrospray ionization mass spectrometry of recovered peptides. Anal. Biochem. 223, 74 – 81 3. Zhou, H., Watts, J. D., and Aebersold, R. (2001) A systematic approach to the analysis of protein phosphorylation. Nat. Biotechnol. 19, 375–378 4. Ficarro, S. B., McCleland, M. L., Stukenberg, P. T., Burke, D. J., Ross,

M. M., Shabanowitz, J., Hunt, D. F., and White, F. M. (2002) Phospho-proteome analysis by mass spectrometry and its application to Saccha-romyces cerevisiae. Nat. Biotechnol. 20, 301–305

5. Manning, G., Whyte, D. B., Martinez, R., Hunter, T., and Sudarsanam, S. (2002) The protein kinase complement of the human genome. Science 298, 1912–1934

6. Garrington, T. P., and Johnson, G. L. (1999) Organization and regulation of mitogen-activated protein kinase signaling pathways. Curr. Opin. Cell

Biol. 11, 211–218

7. Johnson, G. L., and Lapadat, R. (2002) Mitogen-activated protein kinase pathways mediated by ERK, JNK, and p38 protein kinases. Science 298, 1911–1912

8. Han, J., and Molkentin, J. D. (2000) Regulation of MEF2 by p38 MAPK and its implication in cardiomyocyte biology. Trends Cardiovasc. Med. 10, 19 –22

9. Sugden, P. H. (2001) Signalling pathways in cardiac myocyte hypertrophy. Ann. Med. 33, 611– 622

10. Braz, J. C., Bueno, O. F., Liang, Q., Wilkins, B. J., Dai, Y.-S., Parsons, S., Braunwart, J., Glascock, B. J., Klevitsky, R., Kimball, T. F., Hewett, T. E., and Molkentin, J. D. (2003) Targeted inhibition of p38 MAPK promotes hypertrophic cardiomyopathy through upregulation of calcineurin-NFAT signaling. J. Clin. Invest. 111, 1475–1486

11. Liao, P., Georgakopoulos, D., Kovacs, A., Zheng, M., Lerner, D., Pu, H., Saffitz, J., Chien, K., Xiao, R.-P., Kass, D. A., and Wang, Y. (2001) The in vivo role of p38 MAP kinases in cardiac remodeling and restrictive cardiomyopathy. Proc. Natl. Acad. Sci. U. S. A. 98, 12283–12288 12. Guy, L. G., Kothary, R., DeRepentigny, Y., Delvoye, N., Ellis, J., and Wall, L.

(1996) The␤-globin locus control region enhances transcription of but does not confer position-independent expression onto the lacZ gene in transgenic mice. EMBO J. 15, 3713–3721

13. Fernando, P., Kelly, J. F., Balazsi, K., Slack, R. S., and Megeney, L. A. (2002) Caspase 3 activity is required for skeletal muscle differentiation. Proc. Natl. Acad. Sci. U. S. A. 99, 11025–11030

14. Kameshita, I., and Fujisawa, H. (1989) A sensitive method for detection of calmodulin-dependent protein kinase II activity in sodium dodecyl sul-fate-polyacrylamide gel. Anal. Biochem. 183, 139 –143

15. Westermeier, R., and Naven, T. (2002) Proteomics in Practice: A Laboratory Manual of Proteome Analysis, Wiley-VCH Verlag-GmbH, Weinheim, Germany

16. Jones, W. K., Sanchez, A., and Robbins, J. (1994) Murine pulmonary myocardium: developmental analysis of cardiac gene expression. Dev. Dyn. 200, 117–128

17. Kolodziejczyk, S. M., Wang, L., Balazsi, K., DeRepentigny, Y., Kothary, R., and Megeney, L. A. (1999) MEF2 is upregulated during cardiac hyper-trophy and is required for normal post-natal growth of the myocardium. Curr. Biol. 9, 1203–1206

18. Alonso, G., Ambrosino, C., Jones, M., and Nebreda, A. R. (2000) Differential activation of p38 mitogen-activated protein kinase isoforms depending on signal strength. J. Biol. Chem. 275, 40641– 40648

19. Barsyte-Lovejoy, D., Galanis, A., and Sharrocks, A. D. (2002) Specificity determinants in MAPK signaling to transcription factors. J. Biol. Chem. 277, 9896 –9903

20. Wooten, M. W. (2002) In-gel kinase assay as a method to identify kinase substrates. Sci. STKE 2002, PL15

21. Tian, R., Musi, N., D’Agostino, J., Hirshman, M. F., and Goodyear, L. J. (2001) Increased adenosine monophosphate-activated protein kinase activity in rat hearts with pressure-overload hypertrophy. Circulation 104, 1664 –1669

22. Satoh, S., Ueda, Y., Koyanagi, M., Kadokami, T., Sugano, M., Yoshikawa, Y., and Makino, N. (2003) Chronic inhibition of Rho kinase blunts the process of left ventricular hypertrophy leading to cardiac contractile dysfunction in hypertension-induced heart failure. J. Mol. Cell Cardiol. 35, 59 –70

23. Higashi, M., Shimokawa, H., Hattori, T., Hiroki, J., Mukai, Y., Morikawa, K., Ichiki, T., Takahashi, S., and Takeshita, A. (2003) Long-term inhibition of Rho-kinase suppresses angiotensin II-induced cardiovascular hypertro-phy in rats in vivo: Effect on endothelial NAD(P)H oxidase system. Circ. Res. 93, 767–775

24. Morissette, M. R., Sah, V. P., Glembotski, C. C., and Brown, J. H. (2000) The Rho effector, PKN, regulates ANF gene transcription in cardiomyo-cytes through a serum response element. Am. J. Physiol. Heart Circ. Physiol. 278, H1769 –H1774

25. Martenson, R., Law, M., and Deibler, G. (1983) Identification of multiple in vivo phosphorylation sites in rabbit myelin basic protein. J. Biol. Chem. 258, 930 –937

26. Cicirelli, M., Pelech, S., and Krebs, E. (1988) Activation of multiple protein kinases during the burst in protein phosphorylation that precedes the first meiotic cell division in Xenopus oocytes. J. Biol. Chem. 263, 2009 –2019

Activity-based Proteomic Screening of Cardiomyopathy

(11)

27. Bennett, V., and Baines, A. J. (2001) Spectrin and ankyrin-based pathways: Metazoan inventions for integrating cells into tissues. Physiol. Rev. 81, 1353–1392

28. Morrison, A. C., Doris, P. A., Folsom, A. R., Nieto, F. J., Boerwinkle, E., and Hegele, R. A. (2001) G-Protein␤3 subunit and ␣-adducin polymorphisms and risk of subclinical and clinical stroke editorial comment: candidate genes for stroke: If elected, will they serve? Stroke 32, 822– 829 29. Tamagnone, L., and Comoglio, P. M. (2004) To move or not to move?

Semaphorin signalling in cell migration. EMBO Rep. 5, 356 –361 30. Toyofuku, T., Zhang, H., Kumanogoh, A., Takegahara, N., Suto, F., Kamei,

J., Aoki, K., Yabuki, M., Hori, M., Fujisawa, H., and Kikutani, H. (2004) Dual roles of Sema6D in cardiac morphogenesis through region-specific association of its receptor, Plexin-A1, with off-track and vascular endo-thelial growth factor receptor type 2. Genes Dev. 18, 435– 447 31. Toyofuku, T., Zhang, H., Kumanogoh, A., Takegahara, N., Yabuki, M.,

Harada, K., Hori, M., and Kikutani, H. (2004) Guidance of myocardial patterning in cardiac development by Sema6D reverse signalling. Nat. Cell Biol.

32. Deo, R. C., Schmidt, E. F., Elhabazi, A., Togashi, H., Burley, S. K., and

Strittmatter, S. M. (2004) Structural bases for CRMP function in plexin-dependent semaphorin3A signaling. EMBO J. 23, 9 –22

33. de Tombe, P. P., and Solaro, R. J. (2000) Integration of cardiac myofilament activity and regulation with pathways signaling hypertrophy and failure. Ann. Biomed. Eng. 28, 991–1001

34. Bueno, O. F., and Molkentin, J. D. (2002) Involvement of extracellular signal-regulated kinases 1/2 in cardiac hypertrophy and cell death. Circ. Res. 91, 776 –781

35. Molkentin, J. D. (2004) Calcineurin-NFAT signaling regulates the cardiac hypertrophic response in coordination with the MAPKs. Cardiovasc. Res. 63, 467– 475

36. Aebersold, R., and Mann, M. (2003) Mass spectrometry-based proteomics. Nature 422, 198 –207

37. Liu, H., Lin, D., and Yates, J. R., 3rd (2002) Multidimensional separations for protein/peptide analysis in the post-genomic era. BioTechniques 32, 898, 900, 902 passim

38. Wu, C. C., and Yates, J. R., 3rd (2003) The application of mass spectrom-etry to membrane proteomics. Nat. Biotechnol. 21, 262–267

Références

Documents relatifs

[r]

Keywords: sparse representations, dictionary learning, inverse problems, Indian Buffet Process, Bayesian non parametric.. I Thanks to the BNPSI ANR project no ANR-13-BS-03-0006-01

L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.

The three different stacking configurations yield distinct low-frequency Raman response, because the interlayer phonon modes are sensitive to the stacking

Mass spectrometry sequencing of long digital polymers facilitated by programmed inter-byte fragmentation.. Abdelaziz Al Ouahabi, Jean-Arthur Amalian, Laurence Charles,

Having developed this luminescent CDK10/CycM kinase activity assay in the 384-well plate format, we set out to determine whether it is amenable to high-throughput screening

A few residual molecules, active on the plant cell cytoskeleton and plantlets, but showing no detectable effect on human cells, were finally tested for their effect on in vitro