This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ping, P. Search for Related Content PubMed PubMed Citation Articles by Ping, P. Pubmed/NCBI databases Substance via MeSH Related Collections Other Research Cell signalling/signal transduction

(Circulation Research. 2003;93:595.)
© 2003 American Heart Association, Inc.

Review

Identification of Novel Signaling Complexes by Functional Proteomics

Peipei Ping

From the Department of Physiology Medicine and the Division of Cardiology, David Geffen School of Medicine at UCLA, Los Angeles, Calif.

Correspondence to Peipei Ping, PhD, Professor of Physiology and Medicine, Cardiovascular Laboratory, UCLA School of Medicine, MRL Building, Suite 1609, 675 Charles Young Dr, Los Angeles, CA 90095. E-mail peipeiping{at}hotmail.com or peipeiping@earthlink.net

Jennifer E. Van Eyk Guest Editor This Review is part of a thematic series on Proteomics, which includes the following articles:

Cardiovascular Proteomics: Evolution and Potential

Applied Proteomics: Mitochondrial Proteins and Effect on Function

Organelle Proteomics: Implications for Subcellular Fractionation in Proteomics

Identification of Novel Signaling Complexes by Functional Proteomics

Proteomic Approaches to Analyze the Dynamic Relationships Between Nucleocytoplasmic Protein Glycosylation and Phosphorylation

Proteomics in the Cardiomyopathies and Heart Failure: A Step Beyond Genomics

Glycosylation of Apolipoprotein E

	Abstract

Top Abstract Introduction Embracing the Era of... Using Functional Proteomics to... Lessons Learned Through the... References

 
The rapid development of proteomic technologies, combined with the completion of the Human Genome Map, has enabled the compiling of an unprecedented inventory of cellular proteins. Functional proteomics is an emerging field that aims to utilize the enormous amount of information provided by these proteomic technologies to understand the functions of cellular proteins. The utility of functional proteomics has been recently exploited to elucidate cellular mechanisms in numerous fields, of particular salience in the area of signal transduction. This review presents a functional proteomic approach for the study of cardiac cell signaling. It illustrates the strategies by which the subproteome of a targeted signaling system is characterized in an unbiased fashion, the manner in which the biochemical functions of this subproteome are assessed using established molecular and protein chemistry methods, and the challenges associated with these studies.

Key Words: functional subproteomes • multiprotein complexes • signal transduction • ischemic injury • cardiac cell death

	Introduction

Top Abstract Introduction Embracing the Era of... Using Functional Proteomics to... Lessons Learned Through the... References

 
Apreeminent goal in cardiovascular research today is to understand the cellular mechanisms underlying the development of a pathophysiological phenotype. Unraveling the intricate nature of endogenous molecular occurrences is recognized as crucial for a comprehensive understanding of disease pathogenesis. Accordingly, characterization of cellular proteins and the functions they possess will greatly facilitate the design and engineering of therapeutic agents. In this regard, the rapid development of proteomic and genomic technologies has divulged an enormous amount of novel information, leading to the assembly of large protein inventories.^1–4 Consequently, the next challenge is to understand the mechanisms by which these cellular proteins function to impart a phenotype. Functional proteomics has emerged as a powerful tool that uses proteomic technologies to investigate the interaction, integration, and function(s) of proteins.^5,6 This strategy couples proteomic information with biochemical and physiological analyses to advance our understanding of the functional role of proteins in normal and diseased organs.^7–14 In particular, functional proteomics identifies specific proteomic changes, characterizes the biochemical consequences of these changes, and discerns how these subproteomic alterations govern the genesis of a physiological/pathological phenotype.

	Embracing the Era of Functional Proteomics

Top Abstract Introduction Embracing the Era of... Using Functional Proteomics to... Lessons Learned Through the... References

 
Functional Proteomic Strategy
An essential advantage afforded by functional proteomics is its potentially vast application in hypothesis-driven research, in addition to its ostensible discovery-driven utility. This feature makes functional proteomics attractive to a majority of academic investigators. The survival of a multicellular organism is highly dependent on its ability to adapt to changes in its environment. Consequently, an organism’s phenotype in any given circumstance can be viewed as the result of a dynamic equilibrium. It could be hypothesized that cellular homeostasis is governed by an innate infrastructure that adapts in response to impinging stimuli. If proteins are integral components of the aforementioned infrastructure and are actively engaged in the modulation of such an equilibrium, then functional proteomics characterizes the intimate link between this infrastructure and the phenotype.

The conserved conceptual element between all hypotheses to be tested with functional proteomics is the dynamic modulation of a cellular subproteome responsible for the genesis of a phenotype.^7–15 Accordingly, experiments are aimed to characterize the changes that occur in the designated population of cellular proteome and to decipher how these changes result in an altered phenotype. Thus, in addition to the identification of proteins in this subproteome in a nonbiased fashion, functional proteomic studies endeavor to relate this body of protein information not only to what occurs to alter the phenotype, but moreover, how it occurs. With this focus, biochemical analyses and phenotype assessment are conducted to understand the cellular networks that support physiological/pathological phenomena. Recently, our laboratory has used a functional proteomic approach to characterize the PKC-dependent signaling mechanisms underlying cardioprotection.^7,8 We feel that with minor modifications, this strategy can readily be adapted for implementation in a variety of cardiovascular investigations. Figure 1 illustrates a step-by-step course of action of this approach.

View larger version (29K): [in this window] [in a new window]   Figure 1. Schematic representation of the functional proteomic strategy for hypothesis-driven research. If one starts with a hypothesis, the analysis of protein complexes (eg, gel filtration) provides candidate proteins involved in the process of interest. These candidate proteins are then validated by classical biochemical and physiological techniques, which characterize the role (if any) of this protein in the signaling system responsible for the phenotype.

Applications of Functional Proteomics
Although functional proteomics has been used to elucidate cellular mechanisms in numerous fields, a particular area shown fruition is signal transduction.^7–15 Soskic et al¹⁴ examined the subproteome containing proteins for the platelet-derived growth factor (PDGF) receptor signaling pathways. They tested the hypothesis that altered phosphorylation events, which occur in an array of proteins, are indicative of molecules that participate in PDGF signaling. Their results showed that PDGF stimulation significantly impacted the serine and tyrosine phosphorylation of proteins within the PDGF receptor signaling system. A similar approach was taken to study N-methyl-D-aspartate (NMDA) receptor signaling in the brain.¹¹ Husi et al¹¹ was the first to report that proteins mediate the NMDA receptor signaling are physically associated with the receptor. The NMDA multiprotein protein complexes may bear great biological significance during the regulation of neuronal synaptic activity. To understand the role of oxygen tension in placental development, Hoang and colleagues¹⁰ determined the effects of hypoxia on the cytotrophoblast proteome. In this study, a subproteome was identified that consisted of proteins that were altered by oxidative stress. This subproteome was found to contain proteins involved in glycolysis, stress response, as well as signal transduction.¹⁰ Using an ensemble of techniques similar to those described above, mitogen-activated protein kinase (MAPK) cascades were also characterized by other investigators.¹² In yeast, Cdc5p, Cef1p, and hCDC5 were previously implicated in the process of pre-mRNA splicing and are components of multiprotein complexes. Proteins stably associated with the multiprotein Cdc5p complexes in both fission and budding yeasts were characterized using tandem affinity purification.¹⁵ This work marks the first comprehensive analysis of CDC5-associated proteins in yeast and provides evidence that the members of this complex must cooperate in certain aspects of pre-mRNA processing.

A unique feature of these functional proteomic studies is the step taken beyond the identification of proteins within a subproteome, that is, to characterize the functions of the identified proteins (Figure 1). For example, in the report on PDGF receptor signaling,¹⁴ the proteins identified by mass spectrometry were further studied to understand their roles in the interaction between multiple signaling pathways, and to examine kinetic changes in the posttranslational modifications after PDGF stimulation. In the investigation of cytotrophoblast proteome,¹⁰ the subcellular location of identified proteins was further determined using confocal microscopy to characterize the redistribution of these proteins in the wake of hypoxia. In this regard, functional proteomic analyses serve to reveal the functions of proteins and to couple these functions with the cellular mechanisms responsible for cell and organ phenotypes.

	Using Functional Proteomics to Map Multiprotein Signaling Complexes

Top Abstract Introduction Embracing the Era of... Using Functional Proteomics to... Lessons Learned Through the... References

 
The Signaling Module Hypothesis
Despite the unprecedented advances in the discovery of novel signaling elements in recent years,^16,17 little is known regarding the specific manner in which these molecules interact with each other and by what mechanisms they transduce signals. Essentially, the blueprint for the cellular signaling architecture remains to be unmasked.

Subcellular signaling has been classically defined as a host of linear pathways, in which molecule X is located either upstream or downstream of molecule Y, and the summation of many Xs and Ys in turn constitutes a system that transduces a signal. This linear paradigm of signal transduction, although simple, has provided satisfactory guidance for the obscured cell signaling mechanisms over several decades. However, the validity of this model is being frequently challenged by many recent observations that suggest signal transduction is a decidedly more dynamic process. Specifically, it has lately been reported that signaling systems appear to accomplish a multitude of distinct tasks with a limited set of proteins,^16,17 ie, the cellular phenotypes A and B may be dramatically different, but the molecular participants in the generation of these phenotypes may be similar or even identical. Signaling by the mitogen-activated protein kinase (MAPK) superfamily in cardiac cells represents an intriguing exemplar of this phenomenon.^8,18–21 Cardiac cell hypertrophy has been shown to involve differential activation of ERK and JNK.^19,20 In addition, both ERK and JNK kinases have been demonstrated to participate in the signaling responses of an entirely different phenotype: cardioprotection against ischemic injury.^8,21 The resultant questions are as follows: via what mechanisms do cardiac cells coordinate ubiquitously expressed signaling proteins to achieve task-specific signal transduction? How does the cell recruit identical or similar molecular entities to engender contrasting phenotypes?

Recently, investigations using a functional proteomic approach have demonstrated that signaling molecules aggregate to form multiprotein complexes,^{7,8,11,15,22–26} a feature that brings molecules into close vicinity. Furthermore, studies from a number of laboratories suggest that the formation of complexes may facilitate signal transduction.^{8,11,22,27–29} On the basis of these lines of evidence, we submitted the signaling module hypothesis,⁶ in which we propose that cardiac cells assemble multiprotein complexes as a means for signal transduction and that a basic functional unit of these complexes are modules. Investigations of PKC modules support this hypothesis.^21,28–30 These findings suggest that the formation of stimulus- and subcellular location–specific modules within a multiprotein complex represents an essential signaling mechanism whereby a cell performs a variety of distinct tasks.^21,28–31

The premise of multiprotein complexes as a means for signal transduction is built on the concept of associating molecules playing critical regulatory roles in cell signaling. Studies using classical molecular and protein chemistry tools in the past decade provided ample evidence in support of this concept. Salient examples in cardiac cell signaling include the ß-adrenergic receptor associating molecule, ß-arrestin,³² the protein kinase A scaffolding protein, AKAP,³³ and the protein kinase c binding partner, RACKs.³⁴ Although the existence of additional associating molecules was postulated, the traditional approach limited an effective search. A key advantage offered by a functional proteomic strategy is that it empowers an unbiased characterization of associating molecules at the subproteome scale, thereby facilitating the elucidation of protein interactions in a systematic fashion rather than the examination of a single molecule in isolation.

Under the scenario of complex formation as a means for signal transduction, ubiquitous signaling molecules may be uniquely organized into distinct multiprotein complexes at different subcellular compartments. Therefore, a molecule of interest may be involved in multiple autonomous signaling systems (such as MAPKs in the development of hypertrophy versus MAPKs in protection against ischemia), an occurrence enabled by the fact that the architectures of these various systems (ie, the makeup of the individual complexes), may be vastly different. Association of a distinct subset of proteins accents the hallmark of individual complexes. It is plausible that by regulating the protein components of various multiprotein complexes, a cell would be able to conduct signal transduction to manifest a wide spectrum of phenotypes.

Functional Proteomic Strategy to Characterize a Signaling Complex
Profiling a Signaling Complex
Several methodologies have been utilized to isolate protein complexes.^7,8,25,26 Often, the first step is to obtain a concentrated subcellular fraction of interest.^35–38 From this fraction, the individual multiprotein protein complex can be purified via a series of nondenaturing techniques.^15,37–40 The goal of these studies is to maintain (1) the associating members of a protein complex and (2) their molecular interactions with each other, because these features exist endogenously. The characterization of PKC multiprotein complexes in cardiac mitochondria is given as example of this approach (Figure 2). The approach will garner information about the molecular composition of the complex, moreover, about the native size and molecular weight of signaling complexes.

View larger version (33K): [in this window] [in a new window]   Figure 2. Example application of a functional proteomic strategy to examine PKC multiprotein signaling complexes. Cardiac mitochondria are isolated and their functional viability confirmed. Next, mitochondrial lysates are incubated with recombinant GST-tagged PKC protein, and GST-based affinity pull-down assays are performed to purify PKC protein complexes. These purified protein complexes are separated by blue native polyacrylamide gel electrophoresis (PAGE), a nondenaturing separation technique developed for the mitochondria that does not disrupt protein-protein interactions within the complexes. Parallel purification experiments are conducted using the GST epitope alone (ie, GST-null proteins incubated with mitochondria) and solely the GST-PKC protein (without mitochondrial proteins) to serve as negative controls for the blue native PAGE step. Thus, after blue native PAGE, specific protein complexes isolated via GST-PKC pull-down are excised. These multiprotein complexes are in-gel digested and identified by liquid chromatography–coupled tandem mass spectrometry analysis (ionization by electrospray). Lastly, the specific suborganelle distribution of the identified proteins across various compartments of the mitochondria are determined by fractionation of the mitochondria into outer membrane, intermembrane space, inner membrane, and matrix components, followed by immunoprecipitation for PKC and immunoblotting with antibodies corresponding to identified proteins.

Multiprotein complexes may be isolated and purified by a variety of methods, among which, antibody-based immunoprecipitation and recombinant protein-based affinity pull-down are most effective. Importantly, these two techniques provide complimentary information about protein complexes when used in parallel. Antibody-based immunoprecipitation targets preexisting multiprotein complexes using an antibody against one of the components. This method isolates the complexes without disrupting them, but is limited in its ability to identify specific complexes in which the antibody-directed epitope may be masked by other interacting proteins. Additionally, the antibody-binding event may displace proteins that interact with the epitope region of the target protein, and hence these interacting proteins would escape the analysis as well. In contrast, recombinant protein-based affinity pull-down assays introduce exogenous full-length proteins, ie, the "bait" protein, into the system. In this assay, the bait protein competes with endogenous protein to form a new multiprotein complex. This method is advantageous because epitope tagging of the recombinant protein facilitates the pull-down and reduces the likelihood that the epitope for isolation is masked. When used together, these two tools are a powerful approach to isolate and purify multiprotein complexes. In addition, liquid chromatography and other nondenaturing tools (eg, blue native electrophoresis) can be used in combination with antibody-based immunoprecipitation and recombinant tag protein based affinity pull-downs to enhance the sensitivity of the purification procedure (Figure 2). At this stage, the multiprotein complexes can also be visualized by electron microscopy after isolation to gain knowledge on the physical properties of the complexes.^41,42,44

Subsequently, the proteins in the complexes can be further separated by standard denaturing electrophoresis and identified by either matrix-assisted laser deionization (MALDI) mass spectrometry or tandem mass spectrometry.^3,4,45 Based on the experience we have obtained in characterizing PKC multiprotein complexes in the mitochondria, successful identification of molecular components within these complexes requires a high-sensitivity approach. Often, low-flow rate electrospray ("nanospray") with reverse-phase LC and tandem MS on gel excised and trypsinized proteins are effective.^4,7,43 This approach has high sensitivity and mass accuracy, it generates primary amino acid sequence data, and increases the likelihood of definitive identification of proteins in the subsequent database search. LC/MS/MS has been implemented to determine components of the splicesome multiprotein complex⁴³ and the NMDA receptor complex,¹¹ and to examine multiprotein complexes in yeast when coupled with affinity pull-down and PAGE separation.^25,26 After mass spectrometry analysis, the protein sequence data are searched against the protein sequence database (such as the NCBI non-redundant database), using one of numerous publicly available search engines.^4,7,45

Characterization of Signaling Modules
There are four important features of a signaling module that delineate the strategies by which it is characterized. First, a module consists of a set of proteins that are physically associated with each other. The identities of proteins in a module can be determined via mass spectrometry or immunoblotting; whereas their physical interactions are assessed via either in vitro affinity pull-down assays and/or coimmunoprecipitation. Second, the assembly of a module usually leads to chemical modification (ie, posttranslational modification) of one or more of the module components, which can be determined via mass spectrometry and/or 2-dimensional electrophoresis. Third, a signaling module is a basic functional unit within the complex, and thus its ability to transduce signals from one molecule to the next is analyzed using biochemical assays (eg, kinase activity assays). Finally, a signaling module consists of those members of a complex that necessarily associate in order to carry out a signaling event, and hence the functional significance of modules must be verified by the phenotype that they engender. The necessary role of a given module in the signaling system underlying a phenotype is often assessed by pharmacological studies and the use of transgenic animals.^27,28 For example, the establishment of a hypertrophic signaling module will require studies that involve the disruption of this module’s assembly in vivo and the examination of its impact on cardiac hypertrophy.

Identification of Binding Partners
A number of techniques have been utilized to identify binding partners of a protein.^{27,28,46–49} Often the protein of interest is used as a bait to screen for candidate binding partners via affinity pull-down assays. An alternative method that has been widely exploited is the yeast two-hybrid screen,⁴⁸ whereby a reporter gene is activated as the result of interactions between the bait and "prey" proteins. The yeast two-hybrid assay was recently used in a high-throughput fashion to delineate the protein-protein interaction map of Helicobacter pylori.⁴⁹ Once two proteins have been characterized as binding partners, the molecular interactions between various domains of these molecules of interest can subsequently be deciphered using mutagenesis studies.^27,30 It is noteworthy that most of the technologies discussed herein will identify a large pool of proteins that are able to interact with a protein of interest. However, these findings are not necessarily indicative of what occurs in a living cell. Therefore, it is essential that the putative binding partners identified in vitro are further validated with in vivo experiments (Figure 1).

Mechanisms That Define Complex Formation and Module Assembly
Although a number of reports have documented the existence of multiprotein complexes in a variety of systems,^{7,8,11,15,22–26} little is known regarding the cellular mechanisms that regulate the assembly of these complexes. Current investigations from our laboratory and others suggest that subcellular location, cell stimulus, and the molecular configuration of proteins may contribute to complex assembly and module formation.^{6,8,21,27,28,38}

Evidence indicating that protein complexes may be compartmentalized at targeted subcellular locations has come from several investigations.^{16,17,35,36,50} In particular, characterization of PKC complexes from total cardiac cell lysates⁸ and analyses of MAPK complexes from the K562 cells¹² both identified a group of organelle-specific structural proteins that associate with these kinases. Specifically, PKC was found to associate with caveolin-3 and prohibitin,⁸ whereas the MAPKs were colocalized with the Golgi reassembly stacking protein and the nuclear transport RNA-binding protein, suggesting that these complexes might be differentially sequestered to various subcellular compartments.¹² Further analysis of the PKC complexes showed that the composition of all PKC complexes was not identical across different cellular compartments. Instead of a singular PKC complex that simultaneously includes all PKC-associated molecules, each subcellular compartment appears to posses a unique PKC complex, the composition of which appears to be distinctly tailored to perform the functions of the specific subcellular organelle (Figure 2).³⁸ For example, a cardiac module containing PKC and Src tyrosine kinase formed only in the particulate regions of the cell. Although PKC and Src were both independently found in the cytosolic fraction, PKC-Src modules were not detected in this subcellular location of the cardiac cell.²⁸ Taken together, these findings are consistent with the signaling module hypothesis⁶ that subcellular location modulates the assembly of complexes and modules.

Stimulus-dependent regulation of a subproteome has recently been demonstrated in multiple cell types.^{10,12,14,51,52} For example, the phosphorylation events in the PDGF receptor signaling subproteome were significantly affected by PDGF stimulation.¹⁴ In the human K562 cell proteome, phorbol esters dynamically impacted the posttranslational modifications of MAPK cascade members.¹² These data illustrate that a stimulus (such as PDGF and phorbol esters) can profoundly influence the assembly of protein complexes, both by modulating the composition of the complex and by governing the biochemical characteristics of molecules associated with the complex. Recent data from our laboratory demonstrated that formation of PKC-Lck modules was significantly increased during two forms of cardioprotection: ischemic preconditioning and PKC-induced protection,²⁷ indicating that the impinging stimulus also appears to heavily regulate module formation.

For most cellular kinases, subcellular translocation and activation is accompanied by molecular conformational changes.^16,47 The manner by which the configuration of a molecule influences the formation of signaling modules was established through mutation-based in vitro analyses of PKC and Src tyrosine kinase.³⁰ Additional data from our laboratory indicates that Src tyrosine kinase favors binding to the open configuration of PKC, whereas its interactions with the wild-type PKC were considerably weaker.³⁰ It has been well documented that the particulate PKC protein exhibits an open conformation, whereas in the cytosol, the kinase assumes a closed configuration. Thus, the finding that PKC-Src modules reside exclusively in the particulate fraction and that the binding of Src and PKC is enhanced when PKC is in an open configuration, underscores the notion that subcellular location and molecular configuration play essential roles in module assembly.

Complex Formation as a Mean for Cardiac Cell Signaling
What are the advantages of conceptualizing cardiac cell signaling as multiprotein complexes? Cell signaling via complex formation affects the tenets of the current signaling paradigm in at least three ways. First, this model focuses on not only the central kinase/molecule orchestrating the signaling event, but importantly, it also takes into account the battery of molecules that function in concert to accomplish signal transduction. In this regard, emphasis is placed on both the associating partners of the central kinase/molecule as well as the specific manner by which these partners interact. Second, spatial distribution of these multiprotein complexes appears to contribute to the dynamics of this model. Information obtained by our group^{21,27–29,38} and others,^35–37,39 although limited, indicates that the subcellular location of the complex is a major determinant of its molecular composition and the task that the complex performs. Third, as many signaling molecules are being explored as potential drug targets, the signaling module hypothesis highlights novel strategies for drug design: agents that target protein-protein interactions, or moreover, that influence the assembly of specific complexes, are more appealing than drugs that aim at a single molecule in isolation. If multiprotein complexes are essential mechanisms of cell signaling underlying phenotype in health and disease, then therapeutic agents designed to effectively promote or disrupt their formation would likely achieve a level of specificity and efficacy not possible with present strategies.

	Lessons Learned Through the Characterization of PKC Multiprotein Complexes

Top Abstract Introduction Embracing the Era of... Using Functional Proteomics to... Lessons Learned Through the... References

 
Despite the enormous promise functional proteomic strategies afford for the identification of novel multiprotein signaling complexes, a number of limitations associated with the technology also present tremendous challenges for its implementation. Based on our experience with the characterization of PKC multiprotein complexes, several issues are worthy of careful consideration.

Identification of Members Residing in the Multiprotein Complex
In contrast to what had been anticipated in the early stages of functional proteomic studies, the rate-limiting step in this approach is not mass spectrometry–based protein identification (although it is an exceptional technically demanding step), rather, it is the functional validation of proteins identified by mass spectrometry. Our experience with characterizing PKC multiprotein complexes in cardioprotection^7,8 illustrates, in part, the challenges inherent with these investigations, and the specific measures that may be applied to overcome these obstacles.

The story begins with the PKC hypothesis in preconditioning introduced by Downey and colleagues.⁵³ In their 1994 report, inhibition of PKC was shown to block ischemic preconditioning, and activation of this kinase produced an infarct-sparing effect. Subsequently, mounting evidence has demonstrated that the isoform of PKC is a critical mediator of preconditioning.^54–59 In addition to the heart, activation of PKC appears to be a conserved signaling step in protection against cell death in a number of organs.^60–62 Furthermore, multiple ancillary studies have characterized an array of molecules that participate in preconditioning in a PKC-dependent fashion (see review^54,55). The data suggest that during the genesis of cardioprotection, PKC does not function in isolation. Rather, PKC appears to form alliances with a battery of other proteins. These findings support a cardioprotective signaling system that is reliant on PKC, but in addition, that includes a repertoire of molecules whose participation is tightly coupled to the activation of PKC. This enigma of PKC signaling provided an ideal opportunity for the implementation of functional proteomics to characterize a cardioprotective subproteome.

Using a functional proteomic approach, we isolated and purified the PKC multiprotein complexes (an example of such is shown in Figure 2). Although performing the appropriate negative controls for the immunoprecipitation (eg, IgG-based immunoprecipitation as negative controls^7,8) or the affinity pull-downs (eg, GST-null protein as negative controls^7,8; Figure 2) significantly reduced the number of false-positives, the most effective measure was the implementation of criteria, on the basis of which we could discern the hundreds of gel plugs collected from either 1-dimensional or 2-dimensional gel electrophoreses and subsequently analyzed using both MALDI mass spectrometry and LC tandem mass spectrometry. We reasoned that a protein must satisfy the following prerequisites to be considered as a true candidate: (1) identification of the protein was confirmed in multiprotein complexes purified by two independent approaches (eg, immunoprecipitation and GST-affinity pull-down); (2) the protein must have been identified by mass spectrometry on the basis of at least two peptides; and (3) its association with PKC verified by subsequent immunoprecipitation and immunoblotting studies. Finally, studies demonstrate that the association with PKC complexes result in the modification of the activity and/or biological function of a molecule are taken as solid evidence to support a functional role of such molecule in the PKC complexes. With this set of criteria, a total of 93 proteins (from several hundred analyzed spots) have been identified as candidate members of various PKC multiprotein complexes in the heart.^7,8

Functional Characterization of Modules
Based on data in the literature, these 93 proteins can be divided into at least 6 functional groups and are known to localize to various subcellular locations. Our studies show that these components were dynamically altered during cardioprotection. Thus, the task of target validation continues at the level of establishing a causative relationship between a particular signaling module and the genesis of a specific phenotype. To this end, we have thus far characterized the PKC-Src tyrosine kinase signaling modules,^28,30,63 the PKC-Lck signaling modules,²⁷ and the PKC-MAPK²¹ and PKC-VDAC²⁹ signaling modules at the mitochondria. Interestingly, the data suggest that although each modules contains PKC, the downstream signaling events that these modules regulate appear to be distinct. For example, our studies demonstrate that an important downstream signaling task of the PKC-Lck module is the mobilization of the transcription factor NF-B,²⁷ a molecule that had been previously shown to play a protective role by several other investigators (see review⁵⁴) although its mechanism of activation in cardioprotection was not understood. In another example, a signaling target of the mitochondrial PKC-ERK modules was found to be the proapoptotic protein Bad, phosphorylation and inhibition of which was observed in the protected myocardium.²¹

The characterization of PKC signaling complexes and modules not only aided the verification of protective molecules suggested by previous studies, but more importantly, it has led to the identification of novel signaling modules that had not been discovered by classical approaches. Two examples in this regard are the classification of PKC-Bmx and the PKC-Oracle/Cypher signaling modules^64,65 within the PKC multiprotein complexes. Both proteins were previously unrecognized as potentially participating in cardioprotective signaling. Bmx and Oracle/Cypher are identified by MALDI MS (Bmx)⁸ and LC/MS/MS (Oracle/Cypher)⁷ as members of the PKC multiprotein complexes. In view of these lines of evidence, it is conceivable that multiple distinct modules within the PKC complex may exist. These modules function to facilitate the genesis of the cardioprotective phenotype by the accomplishment of as yet unknown signaling tasks in targeted subcellular locations.

Challenges Ahead
Through distinct applications in a variety of studies, functional proteomics has proven to be a powerful tool for the identification of novel multiprotein complexes, signaling modules, and binding partners. Nevertheless, the current technology has several limitations, including the following: (1) it identifies stably associated multiprotein complexes and may not faithfully recapitulate all protein-protein interactions present in vivo; thus, a protein that transiently associates with a multiprotein complex may escape the identification; (2) the current approach is not robustly quantitative and does not provide information regarding the stoichiometry of any components within the multiprotein complex; (3) the sensitivity is low and may not identify proteins that are less abundant; and (4) only those proteins whose interaction are not affected by the given affinity purification procedure can be identified. The last issue may have significant impact, because the purification procedure may affect posttranslational modifications and protein-protein interactions.⁴⁵ Ostensibly, efforts to address these issues are much needed and are necessary steps to make this approach more effective.

Concluding Remarks
As is the case with the genome, cellular or whole animal proteomes will not in themselves represent usable ends. If the tremendous amount of proteomic data are instead taken in the light of a specific hypothesis and used to understand the signaling architecture that is responsible for physiological and pathophysiological organ condition, then the true power of proteomics can achieve fruition. In this regard, functional proteomics affords a nonbiased investigation of a subproteome of interest, along with subsequent examination of this subproteome on the basis of a specific hypothesis to interrogate specific protein events that manifest cell or organ phenotypes. Importantly, functional proteomics allows not only the identification of novel proteins, signaling modules, and signaling complexes, but moreover, it aids our understanding of the causative relationship between the modification of these proteins and development of pathological organ conditions. It is with this end in hand that we can hope to see the conceptual elegance of molecular medicine become a reality.

	Acknowledgments

 
This work was supported in part by NIH R01 grants HL63901 and HL65431 and by the Laubisch Endowment. The author is deeply indebted to Dr Thomas M. Vondriska, who has provided essential discussions and critical comments on this review.

	Footnotes

 
Original received August 1, 2001; resubmission received July 8, 2003; revised resubmission received August 18, 2003; accepted August 19, 2003.

	References

Top Abstract Introduction Embracing the Era of... Using Functional Proteomics to... Lessons Learned Through the... References

 

1. Costanzo MC, Hogan JD, Cusick ME, Davis BP, Fancher AM, Hodges PE, Kondu P, Lengieza C, Lew-Smith JE, Lingner C, Roberg-Perez KJ, Tillberg M, Brooks JE, Garrels JI. The yeast proteome database (YPD) and Caenorhabditis elegans proteome database (WormPD). Nucleic Acids Res. 2000; 28: 73–76.[Abstract/Free Full Text]
2. Dunn MJ. Studying heart disease using the proteomic approach. Drug Discov Today. 2000; 5: 76–84.[CrossRef][Medline] [Order article via Infotrieve]
3. Gygi SP, Aebersold R. Mass spectrometry and proteomics. Curr Opin Chem Biol. 2000; 4: 489–494.[CrossRef][Medline] [Order article via Infotrieve]
4. Pandey A, Mann M. Proteomics to study genes and genomes. Nature. 2000; 405: 837–846.[CrossRef][Medline] [Order article via Infotrieve]
5. Arrell DK, Neverova I, Van Eyk JE. Cardiovascular proteomics. Circ Res. 2001; 88: 763–773.[Abstract/Free Full Text]
6. Vondriska TM, Klein JB, Ping P. Use of functional proteomics to investigate PKC-mediated cardioprotection: the signaling module hypothesis. Am J Physiol. 2001; 280: H1434–H1441.
7. Edmondson RD, Vondriska TM, Biederman KJ, Zhang J, Jones RC, Zheng Y, Allen DL, Xiu JX, Cardwell EM, Pisano MR, Ping P. PKC signaling complexes include metabolism- and transcription/translation-related proteins: complimentary separation techniques with liquid chromatography and tandem mass spectrometry. Mol Cell Proteomics. 2002; 1: 421–433.[Abstract/Free Full Text]
8. Ping P, Zhang J, Pierce WM Jr, Bolli R. Functional proteomic analysis of PKC signaling complexes in the normal heart and during cardioprotection. Circ Res. 2001; 88: 59–62.[Abstract/Free Full Text]
9. Godovac-Zimmermann J, Soskic V, Poznanovic S, Brianza F. Functional proteomics of signal transduction by membrane receptors. Electrophoresis. 1999; 20: 952–961.[CrossRef][Medline] [Order article via Infotrieve]
10. Hoang VM, Foulk R, Clauser K, Burlingame A, Gibson BW, Fisher SJ. Functional proteomics: examining the effects of hypoxia on the cytotrophoblast protein repertoire. Biochemistry. 2001; 40: 4077–4086.[CrossRef][Medline] [Order article via Infotrieve]
11. Husi H, Ward MA, Choudhary JS, Blackstock WP, Grant SG. Proteomic analysis of NMDA receptor-adhesion protein signaling complexes. Nat Neurosci. 2000; 3: 661–669.[CrossRef][Medline] [Order article via Infotrieve]
12. Lewis TS, Hunt JB, Aveline LD, Jonscher KR, Louie DF, Yeh JM, Nahreini TS, Resing KA, Ahn NG. Identification of novel MAP kinase pathway signaling targets by functional proteomics and mass spectrometry. Mol Cell. 2000; 6: 1343–1354.[CrossRef][Medline] [Order article via Infotrieve]
13. McDonough JL, Labugger R, Pickett W, Tse MY, MacKenzie S, Pang SC, Atar D, Ropchan G, Van Eyk JE. Cardiac troponin I is modified in the myocardium of bypass patients. Circulation. 2001; 103: 58–64.[Abstract/Free Full Text]
14. Soskic V, Gorlach M, Poznanovic S, Boehmer FD, Godovac-Zimmermann J. Functional proteomics analysis of signal transduction pathways of the platelet-derived growth factor ß receptor. Biochemistry. 1999; 38: 1757–1764.[CrossRef][Medline] [Order article via Infotrieve]
15. Ohi MD, Link AJ, Ren L, Jennings JL, McDonald WH, Could KL. Proteomics analysis reveals stable multiprotein complexes in both fission and budding yeast containing Myb-related Cdc5p/Cef1p, novel pre-mRNA splicing factors, and snRNAs. Mol Cell Biol. 2002; 22: 2011–2024.[Abstract/Free Full Text]
16. Hunter T. Signaling-2000 and beyond. Cell. 2000; 100: 113–127.[CrossRef][Medline] [Order article via Infotrieve]
17. Jordan JD, Landau EM, Iyengar R. Signaling networks. Cell. 2000; 103: 193–2000.[CrossRef][Medline] [Order article via Infotrieve]
18. Heidkamp MC, Bayer AL, Martin JL, Samarel AM. Differential activation of mitogen-activated protein kinase cascades and apoptosis by protein kinase C and in neonatal rat ventricular myocytes. Circ Res. 2001; 89: 882–890.[Abstract/Free Full Text]
19. Bueno OF, De Windt LJ, Tymitz KM, Witt SA, Kimball TR, Klevitsky R, Hewett TE, Jones SP, Lefer DJ, Peng CF, Kitsis RN, Molkentin JD. The MEK1-ERK1/2 signaling pathway promotes compensated cardiac hypertrophy in transgenic mice. EMBO J. 2000; 19: 6341–6350.[CrossRef][Medline] [Order article via Infotrieve]
20. Clerk A, Pham FH, Fuller SJ, Sahai E, Aktories K, Marais R, Marshall C, Sugden PH. Regulation of mitogen-activated protein kinases in cardiac myocytes through the small G protein Rac 1. Mol Cell Biol. 2001; 21: 1173–1184.[Abstract/Free Full Text]
21. Baines PC, Zhang J, Wang GW, Zheng YT, Xiu JX, Cardwell EM, Bolli R, Ping P. Mitochondrial PKC and MAPK form signaling modules in the murine heart. Circ Res. 2002; 90: 390–397.[Abstract/Free Full Text]
22. Fraser ID, Cong M, Kim J, Rollins EN, Daaka Y, Lefkowitz RJ, Scott JD. Assembly of an A kinase-anchoring protein-ß₂-adrenergic receptor complex facilitates receptor phosphorylation and signaling. Curr Biol. 2000; 10: 409–412.[CrossRef][Medline] [Order article via Infotrieve]
23. MacCoss MJ, McDonald WH, Saraf A, Sadygov R, Clark JM, Tasto JJ, Gould KL, Wolters D, Washburn M, Weiss A, Clark JI, Yates JR III. Shotgun identification of protein modifications from protein complexes and lens tissue. Proc Natl Acad Sci U S A. 2002; 99: 7900–7905.[Abstract/Free Full Text]
24. Nedelkov D, Nelson RW. Delineation of in vivo assembled multiprotein complexes via biomolecular interaction analysis mass spectrometry. Proteomics. 2001; 1: 1441–1446.[CrossRef][Medline] [Order article via Infotrieve]
25. Ho Y, Gruhler A, Heilbut A, Bader GD, Moore L, Adams SL, Millar A, Taylor P, Bennett K, Boutilier K, et al. Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry. Nature. 2002; 415: 180–183.[CrossRef][Medline] [Order article via Infotrieve]
26. Gavin AC, Bosche M, Krause R, Grandi P, Marioch M, Bauer A, Schultz J, Rick JM, Michon AM, Cruciat CM, et al. Functional organization of the yeast proteome by systematic analysis of protein complexes. Nature. 2002; 415: 141–147.[CrossRef][Medline] [Order article via Infotrieve]
27. Ping P, Song C, Zhang J, Guo Y, Cao X, Li RC, Wu W, Vondriska TM, Pass JM, Tang XL, Pierce WM Jr, Bolli R. Formation of PKC-Lck signaling modules confers cardioprotection. J Clin Invest. 2002; 109: 499–507.[CrossRef][Medline] [Order article via Infotrieve]
28. Vondriska TM, Zhang J, Song C, Tang XL, Cao X, Baines CP, Pass JM, Wang S, Bolli R, Ping P. PKC-Src modules direct signal transduction in NO-induced cardioprotection. Circ Res. 2001; 88: 1306–1313.[Abstract/Free Full Text]
29. Baines CP, Song CX, Zheng YT, Wang GW, Zhang J, Wang OL, Guo Y, Bolli R, Cardwell EM, Ping P. PKC interacts with and inhibits the permeability transition pore in cardiac mitochondria. Circ Res. 2003; 92: 873–880.[Abstract/Free Full Text]
30. Song C, Vondriska TM, Wang GW, Klein JB, Cao X, Zhang J, Kang Y, D’Souza S, Ping P. Conformation dictates module formation: example of PKC and Src tyrosine kinase. Am J Physiol. 2002; 282: H1166–H1171.
31. Vondriska TM, Ping P. Multiprotein signaling complexes and regulation of cardiac phenotype. J Mol Cell Cardiol. 2003; 35: 1027–1033.[CrossRef][Medline] [Order article via Infotrieve]
32. Luttrel LM, Roudabush FL, Choy EW, Miller WE, Field ME, Pierce KL, Lefkowitz RJ. Activation and targeting of extracellular signal-regulated kinases by ß-arrestin scaffolds. Proc Natl Acad Sci U S A. 2001; 98: 2449–2454.[Abstract/Free Full Text]
33. Fink MA, Zakhary DR, Mackey JA, Desnoyer RW, Appereson-Hansen C, Damron DS, Bond M. AKAP-mediated targeting of protein kinase A regulates contractility in cardiac myocytes. Circ Res. 2001; 88: 291–297.[Abstract/Free Full Text]
34. Mochly-Rosen D. Localization of protein kinases by anchoring proteins: a theme in signal transduction. Science. 1995; 268: 247–251.[Abstract/Free Full Text]
35. Huber LA, Pfaller K, Vietor I. Organelle proteomics: implications for subcellular fractionation in proteomics. Circ Res. 2003; 92: 962–968.[Abstract/Free Full Text]
36. Bell AW, Ward MA, Blackstock WP, Freeman HN, Choudhary JS, Lewis AP, Chotai D, Fazel A, Gushue JN, Paiement J, Paley S, Chevet E, Lafreniere-Roula M, Solari R, Thomas DY, Rowley A, Bergeron J. Proteomics characterization of abundant Golgi membrane proteins. J Biol Chem. 2001; 276: 5152–5165.[Abstract/Free Full Text]
37. Taylor SW, Fahy E, Zhang B, Glenn GM, Warnock DE, Wiley S, Murphy AN, Gaucher SP, Capaldi RA, Gibson BW, Ghosh SS. Characterization of the human heart mitochondrial proteome. Nat Biotechnol. 2003; 21: 281–286.[CrossRef][Medline] [Order article via Infotrieve]
38. Zhang J, Bolli R, Zheng Y, Xiu JX, Wang G, Pierce WM, Ping P. Subcellular location is a major determinant of PKC complex assembly: a view gained via functional proteomic analyses in cardioprotection. Circulation. 2001; 104 (suppl II): II-209. Abstract.
39. Nijtmans LG, Henderson NS, Holt IJ. Blue native electrophoresis to study mitochondrial and other protein complexes. Methods. 2002; 26: 327–334.[CrossRef][Medline] [Order article via Infotrieve]
40. Brookes PS, Pinner A, Ramachandran A, Coward L, Barnes S, Kim H, Darley-Usmar VM. High throughput two-dimensional blue-native electrophoresis: a tool for functional proteomics of mitochondria and signaling complexes. Proteomics. 2002; 2: 969–977.[CrossRef][Medline] [Order article via Infotrieve]
41. Huang CC, Hall DH, Hedgecock EM, Kao G, Karantza V, Vogel BE, Hutter H, Chisholm AD, Yurchenoco PD, Wadsworth WG. Laminin subunits and their role in C. elegans development. Development. 2003; 130: 3343–3358.[Abstract/Free Full Text]
42. Perez-Terzic C, Behfar A, Mery A, van Deursen JM, Terzic A, Puceat M. Structural adaptation of the nuclear pore complex in stem cell-derived cardiomyocytes. Circ Res. 2003; 92: 444–452.[Abstract/Free Full Text]
43. Neubauer G, King A, Rappsilber J, Calvio C, Watson M, Ajuh P, Sleeman J, Lamond A, Mann M. Mass spectrometry and EST-database searching allows characterization of the multi-protein spliceosome complex. Nat Genet. 1998; 20: 46–50.[CrossRef][Medline] [Order article via Infotrieve]
44. Wedaman KP, Reinke A, Anderson S, Yates J III, McCaffery JM, Powers T. Tor kinases are in distinct membrane-associated protein complexes in Saccharomyces cerevisiae. Mol Biol Cell. 2003; 14: 1204–1220.[Abstract/Free Full Text]
45. Aebersold R, Mann M. Mass spectrometry-based proteomics. Nature. 2003; 422: 198–207.[CrossRef][Medline] [Order article via Infotrieve]
46. Hazbun TR, Fields S. Networking proteins in yeast. Proc Natl Acad Sci U S A. 2001; 98: 4277–4278.[Free Full Text]
47. Jaken S, Parker PJ. Protein kinase C binding partners. Bioessays. 2000; 22: 245–254.[CrossRef][Medline] [Order article via Infotrieve]
48. Legrain P, Jestin JL, Schachter V. From the analysis of protein complexes to proteome-wide linkage maps. Curr Opin Biotechnol. 2000; 11: 402–407.[CrossRef][Medline] [Order article via Infotrieve]
49. Rain JC, Selig L, De Reuse H, Battaglia V, Reverdy C, Simon S, Lenzen G, Petel F, Wojcik J, Schachter V, Chemama Y, Labigne A, Legrain P. The protein-protein interaction map of Helicobacter pylori. Nature. 2001; 409: 211–215.[CrossRef][Medline] [Order article via Infotrieve]
50. Yanagida M, Shimamoto A, Nishikawa K, Furuichi Y, Isobe T, Takahashi N. Isolation and proteomic characterization of the major proteins of the nucleolin-binding ribonucleoprotein complexes. Proteomics. 2001; 1: 1390–1404.[CrossRef][Medline] [Order article via Infotrieve]
51. Kanamoto T, Hellman U, Heldin CH, Souchelnytskyi S. Functional proteomics of transforming growth factor-ß1-stimulated Mv1Lu epithelial cells: Rad51 as a target of TGF-ß1–dependent regulation of DNA repair. EMBO J. 2002; 21: 1219–1230.[CrossRef][Medline] [Order article via Infotrieve]
52. Lin TK, Hughes G, Muratovska A, Blaikie FH, Brookes PS, Darley-Usmar V, Smith RAJ, Murphy MP. Specific modification of mitochondrial protein thiols in response to oxidative stress. J Biol Chem. 2002; 277: 17048–17056.[Abstract/Free Full Text]
53. Ytrehus K, Liu GS, Downey JM. Preconditioning protects ischemic rabbit heart by protein kinase C activation. Am J Physiol. 1994; 266: H1145–H1152.[Medline] [Order article via Infotrieve]
54. Bolli R. The late phase of preconditioning. Circ Res. 2000; 87: 972–983.[Abstract/Free Full Text]
55. Cohen MV, Baines CP, Downey JM. Ischemic preconditioning. Annu Rev Physiol. 2000; 62: 79–109.[CrossRef][Medline] [Order article via Infotrieve]
56. Dorn GW II, Souroujon MC, Liron T, Chen CH, Gray MO, Zhou HZ, Csukai M, Wu G, Lorenz JN, Mochly-Rosen D. Sustained in vivo cardiac protection by a rationally designed peptide that causes protein kinase C translocation. Proc Natl Acad Sci U S A. 1999; 96: 12798–12803.[Abstract/Free Full Text]
57. Gray MO, Karliner JS, Mochly-Rosen D. A selective protein kinase C antagonist inhibits protection of cardiac myocytes from hypoxia-induced cell death. J Biol Chem. 1997; 272: 30945–30951.[Abstract/Free Full Text]
58. Sato T, O’Rourke B, Marban E. Modulation of mitochondrial ATP-dependent K⁺ channels by protein kinase C. Circ Res. 1998; 83: 110–114.[Abstract/Free Full Text]
59. Pyle WG, Chen Y, Hofmann PA. Cardioprotection through a PKC-dependent decrease in myofilament ATPase. Am J Physiol. 2003; 285: H1220–H1228.
60. Aley KO, Messing RO, Mochly-Rosen D, Levine JD. Chronic hypersensitivity for inflammatory nociceptor sensitization mediated by the isozyme of protein kinase C. J Neurosci. 2000; 20: 4680–4685.[Abstract/Free Full Text]
61. Chow JY, Uribe JM, Barrett KE. A role for protein kinase C in the inhibitory effect of epidermal growth factor on calcium-stimulated chloride secretion in human colonic epithelial cells. J Biol Chem. 2000; 275: 21169–21176.[Abstract/Free Full Text]
62. Knauf JA, Elisei R, Mochly-Rosen D, Liron T, Chen XN, Gonsky R, Korenberg JR, Fagin JA. Involvement of PKC in thyroid cell death. J Biol Chem. 1999; 274: 23414–23425.[Abstract/Free Full Text]
63. Tang XL, Kodani E, Takano H, Hill M, Shinmura K, Vondriska TM, Ping P, Bolli R. Protein tyrosine kinase signaling is necessary for NO donor-induced late preconditioning against myocardial stunning. Am J Physiol. 2003; 284: H1441–H1448.
64. Vondriska TM, Tang XL, Zhang J, Bolli R, Ping P. Bmx, a member of the Tec family of non-receptor tyrosine kinases, is a novel participant in the signaling system of NO-induced cardioprotection. Circulation. 2001; 104 (suppl II): II-61. Abstract.
65. Vondriska TM, Chen J, Wang GW, Lu M, Qiao X, Zong NC, Zhang J, Pantaleon D, Roos K, Ping P. The scaffolding protein Oracle is a necessary mediator of nitric oxide-induced cardioprotection. Circulation. In press. Abstract.

This article has been cited by other articles:

				D. A. Liem, H. M. Honda, J. Zhang, D. Woo, and P. Ping Past and present course of cardioprotection against ischemia- reperfusion injury J Appl Physiol, December 1, 2007; 103(6): 2129 - 2136. [Abstract] [Full Text] [PDF]

				T. A. Drake and P. Ping Thematic review series: Systems Biology Approaches to Metabolic and Cardiovascular Disorders. Proteomics approaches to the systems biology of cardiovascular diseases J. Lipid Res., January 1, 2007; 48(1): 1 - 8. [Abstract] [Full Text] [PDF]

				S. Arab, A. O. Gramolini, P. Ping, T. Kislinger, B. Stanley, J. van Eyk, M. Ouzounian, D. H. MacLennan, A. Emili, and P. P. Liu Cardiovascular Proteomics: Tools to Develop Novel Biomarkers and Potential Applications J. Am. Coll. Cardiol., November 7, 2006; 48(9): 1733 - 1741. [Abstract] [Full Text] [PDF]

				A. D. T. Costa, R. Jakob, C. L. Costa, K. Andrukhiv, I. C. West, and K. D. Garlid The Mechanism by Which the Mitochondrial ATP-sensitive K+ Channel Opening and H2O2 Inhibit the Mitochondrial Permeability Transition J. Biol. Chem., July 28, 2006; 281(30): 20801 - 20808. [Abstract] [Full Text] [PDF]