| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
* Department of Pharmacology and Cancer Biology, Duke University, Durham, North Carolina 27710-3686
Serenex Inc., 323 Foster Street, Durham, North Carolina 27710
 |
ABSTRACT |
|---|
 TOP ABSTRACT I. IntroductionII. The Need for Protein Studies in...III. A Proteomics Approach to Signal...IV. Protein IdentificationV. Phosphoproteome AnalysisVI. Proteomics ApplicationsACKNOWLEDGEMENTSREFERENCES |
|---|
|
I. Introduction |
|---|
|
TOPABSTRACTI. IntroductionII. The Need for Protein Studies in...III. A Proteomics Approach to Signal...IV. Protein IdentificationV. Phosphoproteome AnalysisVI. Proteomics ApplicationsACKNOWLEDGEMENTSREFERENCES |
|---|
|
II. The Need for Protein Studies in Signal Transduction |
|---|
|
TOPABSTRACTI. IntroductionII. The Need for Protein Studies in...III. A Proteomics Approach to Signal...IV. Protein IdentificationV. Phosphoproteome AnalysisVI. Proteomics ApplicationsACKNOWLEDGEMENTSREFERENCES |
|---|
Another reason protein analysis is important is because proteins can undergo post-translational modifications and these modifications cannot be predicted from genomic data. Proteins can be modified by the attachment of groups such as phosphates, sulfates, carbohydrates, and lipids. These modifications can change the function of proteins in cells. This is especially important in signal transduction, where signals frequently are transmitted by protein modifications. One type of protein modification that is particularly abundant is protein phosphorylation. It is estimated that 30% of all cellular proteins contain covalently bound phosphate (Hubbard and Cohen, 1993). Thus, to obtain a full understanding of signal transduction pathways, it is essential to study protein modifications at the molecular level.
Finally, protein studies are necessary in signal transduction because signaling pathways are composed of proteins. Many proteins exist in protein complexes whose function may be elucidated only through a study of the intact complex. Currently, the composition and regulation of many protein complexes in signaling remain unknown. In a recent study of protein interactions in Saccharomyces cerevisiae, hundreds of multiprotein complexes were detected, of which more than 90% contained at least one previously uncharacterized protein (Gavin et al., 2002; Ho et al., 2002). To fully understand signaling pathways, the individual proteins in a protein complex will have to be identified and characterized during signaling events. These types of experiments will require a proteomics approach.
|
III. A Proteomics Approach to Signal Transduction |
|---|
|
TOPABSTRACTI. IntroductionII. The Need for Protein Studies in...III. A Proteomics Approach to Signal...IV. Protein IdentificationV. Phosphoproteome AnalysisVI. Proteomics ApplicationsACKNOWLEDGEMENTSREFERENCES |
|---|
Because all these questions must be answered to understand signal transduction, proteomics represents only the first step in this endeavor. It is no longer sufficient to produce long lists of proteins that undergo changes in a given system. It is also necessary to validate the changes by other methods. Validation of a change in a specific protein may involve cloning and expression of the protein in a heterologous system, protein biochemistry to confirm modifications, and cell biology to determine biological significance. Validation of a single protein or its modification often requires 6–12 months of work. Therefore, a proteomics experiment should be designed carefully so that only proteins relevant to the biological problem are studied. Otherwise, significant time can be spent investigating irrelevant proteins.
One of the major hurdles in proteomics occurs at the beginning of the experiment. That is, how can the complex mixture of proteins produced by homogenization of whole cells, tissues, or organisms be analyzed simultaneously? One of the difficulties inherent in a proteomics approach is that a cell extract contains thousands of distinct proteins. These proteins have to be resolved from each other so that changes in individual proteins can be observed. Another difficulty is the detection of low-copy proteins in a cell extract. Finally, because a large number of proteins are involved in proteomic analysis, strategies must be devised to focus the investigation on certain types of proteins.
A. ANALYSIS OF PROTEIN MIXTURES BY TWO-DIMENSIONAL GEL ELECTROPHORESIS
An example of how two-dimensional gel electrophoresis (2-DGE) can be used in proteomics is shown in Figure 1A. In this approach, a cell extract is prepared from two different samples and resolved by 2-DGE for comparison of protein expression or changes in protein modification. This method can provide valuable information and has been used to investigate specific signaling pathways (Lewis et al., 2000). However, this approach also suffers from many limitations, including poor protein representation and an inability to identify low-copy proteins (Gygi et al., 1999b; Graves and Haystead, 2002). In addition, it is difficult to screen large numbers of samples for changes because 2-DGE is very labor intensive and time consuming. Therefore, alternative strategies are needed.
|
The advantage of these methods is that, theoretically, more proteins can be identified, especially low-copy proteins. In addition, because protein gels are avoided, time is saved from loading and extracting proteins from gels, a process that is not easily automated. However, protein gels perform an important function. They allow for the visual selection of specific proteins from a complex mixture. In the absence of such visual selection, every protein (or peptide) in a cell extract has to be analyzed to find those that undergo changes in response to the biological condition. Frequently, it is important to activate specific signaling pathways and determine early events to understand how signals are transmitted. In these experiments, activation of a specific signaling pathway will not result in changes in the majority of cellular proteins. Therefore, if all proteins in a cell are analyzed, most of the analysis time will be spent on irrelevant proteins. In addition, the analysis of large numbers of proteins requires specialized equipment and computer systems to handle massive amounts of data.
C. TARGETED ANALYSIS OF SUBPROTEOMES
In our laboratory, we utilize an approach that enriches for low-copy proteins and allows for their visual selection. In this method, affinity chromatography is used to isolate subproteomes or specific types of proteins related to the biological question. These proteins are then visualized using one-dimensional gel electrophoresis (1-DGE) or 2-DGE (Figure 1C). The isolation of subproteomes of the cell serves two important functions. First, it reduces the complexity of the analysis because fewer proteins are analyzed. This saves data analysis time and focuses attention on only those proteins relevant to the biological question. Second, it enriches for low-abundance proteins that otherwise would go undetected in cell extracts. Because of the reduction in protein complexity achieved with this approach, the protein samples can be analyzed for differences using conventional 1-DGE. As a result, many more proteins are represented in the gels and large numbers of samples can be screened quickly.
An example of subproteome enrichment is demonstrated by the use of the specific affinity resin, 
-phosphate-linked adenosine triphosphate (ATP)-Sepharose, originally developed in our laboratory for the purification of protein kinases (Haystead et al., 1993). Unlike previous nucleotide resins, we linked ATP in an orientation favorable for binding protein kinases based upon the crystal structure of cyclic AMP (cAMP)-dependent protein kinase (Knighton et al., 1991). From this structure, it was shown that the -phosphate group of ATP extended to the solvent. Therefore, this part of ATP was used for linkage to the matrix. We have shown that, in addition to binding protein kinases, -ATP-Sepharose has the ability to bind a large variety of purine-utilizing enzymes, including dehydrogenases, heat shock proteins, nonprotein kinases, DNA ligases, and miscellaneous purine-utilizing enzymes (Graves et al., 2001). This subproteome can be termed the "purine-binding proteome."
Because many components of signal transduction pathways (e.g., protein kinases) are present in the purine-binding proteome, -ATP-Sepharose can be used as a tool to dissect protein-signaling pathways. In this approach, the entire purine-binding proteome is captured from control cells or cells that have undergone some form of treatment. Proteins then are eluted from the matrix with free ATP and either resolved by 2-DGE or further fractionated by ion-exchange chromatography (Figure 2). This approach can provide information about changes in protein expression, in the activation state of protein kinases (as reflected in their ability to bind -ATP-Sepharose), or in the post-translational modification of proteins.
|
|
IV. Protein Identification |
|---|
|
TOPABSTRACTI. IntroductionII. The Need for Protein Studies in...III. A Proteomics Approach to Signal...IV. Protein IdentificationV. Phosphoproteome AnalysisVI. Proteomics ApplicationsACKNOWLEDGEMENTSREFERENCES |
|---|
A. MASS SPECTROMETRY
MS is now the dominant technology for protein identification (Pandey and Mann, 2000). Using MS, most proteins can be identified at the femtomolar level. However, the success of protein identification can depend upon the sample preparation and on the type of mass spectrometer used. In functional proteomics, considerable effort frequently is expended to obtain a relatively small number of proteins for analysis. In this case, the most important feature of a mass spectrometer is high sensitivity and ability to identify a protein with the highest level of confidence. To this end, we have found that electrospray ionization and tandem mass spectometry (MS/MS) are the most useful. For MS analysis, samples must be converted to gas-phase ions. Two of the most common methods to accomplish this are matrix-assisted laser desorption ionization (MALDI) and electrospray ionization (ESI).
1. MALDI-MS
In MALDI, a laser is used to promote the ionization of analyte molecules embedded in a crystalline matrix (Karas and Hillenkamp, 1988). The single biggest advantage to MALDI analysis is speed. In MALDI, the entire process of sample preparation and analysis in the mass spectrometer can be automated. This is not yet the case for ESI. As a result, MALDI is the preferred method for high-throughput proteomics operations. However, for many functional proteomics applications, MALDI may not be the best choice. This is because the type of data obtained may not be specific enough for protein identification. In MALDI, the masses of individual peptides derived from a protein are measured in what is known as a peptide mass "fingerprint." In general, peptide mass fingerprint analysis is not as reliable for protein identification as peptide amino acid sequence obtained by MS/MS. Because of this deficiency, a new breed of MALDI mass spectrometers have emerged that can, in addition to peptide mass fingerprinting, also obtain peptide amino acid sequence. Examples of these machines include MALDI-Q-TOF (Loboda et al., 2000; Shevchenko et al., 2000) and MALDI-TOF-TOF (Medzihradszky et al., 2000). It remains to be seen which machine will be most successful at protein identification.
2. ESI-MS
In ESI, a sample is directly introduced into the mass spectrometer as a fine mist of charged droplets (Fenn et al., 1989; Hunt et al., 1992). In an adaptation of electrospray known as nanospray, samples are introduced through microcapillary tubes at very low flow rates (Wilm and Mann, 1996; Wilm et al., 1996). ESI has a number of advantages over MALDI in functional proteomics applications and, as a result, is the method of choice for sample analysis in our laboratory. First, nano-electrospray MS can be used to obtain peptide amino acid sequence in a procedure known as MS/MS (Wilm et al., 1996). This is the most specific method for protein identification and greater confidence can be achieved in protein assignment. Second, from our experience, nanospray provides a higher sensitivity than MALDI, allowing for the identification of low-copy proteins. Finally, and perhaps most importantly, MS/MS is less affected by protein mixtures than peptide mass fingerprinting. This is critically important because, frequently, protein "bands" isolated from polyacrylamide gels contain more than one protein.
B. OUR APPROACH TO PROTEIN IDENTIFICATION
The general approach for protein identification in our laboratory is outlined in Figure 3. In this method, a protein band is excised from a gel, digested to peptides by trypsin, and the peptides purified by the method of Wilm and colleagues (Wilm et al., 1996). The resultant peptides are ionized by nano-electrospray and mass analyzed with a quadrupole-time-of-flight (TOF) mass spectrometer (Q-STAR, PE-SCIEX) to obtain a spectrum of the peptides. From this spectrum, a parent ion is selected for fragmentation and amino acid sequencing. We are capable of identifying faint, silver-stained bands that equal approximately 5–10 ng of protein. However, because MS is not quantitative, frequently several different proteins can be identified from a single band isolated from a polyacrylamide gel. Thus, additional experiments often are required to confirm that the protein identified in the mass spectrometer correlates with a given protein observed on a gel.
|
1. Peptide Mass Fingerprint Database Searching
In this approach, peptide masses obtained from MALDI-MS are compared against theoretical spectra obtained from primary-sequence databases (Eng et al., 1994). Although this type of search method is rapid, it is not error tolerant and requires a high degree of agreement for success. Factors that alter peptide masses from those in the database (e.g., post-translational modifications) can invalidate this search method. Since many proteins in signaling pathways are modified, this search method may not result in correct protein identification. Another factor that can contribute to inaccurate protein identification is mass degeneracy. For example, a peptide with the sequence VSLNP will have the same mass as the peptide LSNVP. In addition, the masses of certain amino acids, when combined, can equal the mass of other amino acids. For example, the mass of glycine plus glutamate or aspartate plus alanine equals the mass of tryptophane. Because of these limitations, peptide mass fingerprinting is not well suited for protein identification from organisms with large genomes (e.g., human) or towards genomes that are not completely annotated (Qin et al., 1997).
2. Peptide Mass Tag Database Searching
Peptide mass tag database searching can be conducted using peptide sequence obtained from MS/MS. In this approach, a partial amino acid sequence, known as the sequence tag, is combined with the mass of the peptide to search relevant databases (Eng et al., 1994; Mann and Wilm, 1994). The advantage to this method is that it is fast and, if enough amino acid sequence is obtained, can be a very specific method for protein identification. However, this search method also depends upon a complete and well-annotated sequence database.
3. FASTS Database Searching
The FASTS program uses amino acid sequence obtained from MS/MS to search databases from organisms whose genomes have not been sequenced completely or whose databases are not fully annotated (Mackey et al., 2002). In this approach, de novo peptide sequence is obtained from MS, entered into the FASTS program, and all peptide sequences are searched against a database simultaneously. Because the program utilizes amino acid sequence, it is error tolerant (Mackey et al., 2002). This is important because sequencing errors, polymorphisms, and conservative amino acid substitutions often can prevent protein identification with other search methods. A schematic of how the FASTS program functions is shown in Figure 4. The FASTS program is available to the public at http://fasta.bioch.virginia.edu/.
|
|
V. Phosphoproteome Analysis |
|---|
|
TOPABSTRACTI. IntroductionII. The Need for Protein Studies in...III. A Proteomics Approach to Signal...IV. Protein IdentificationV. Phosphoproteome AnalysisVI. Proteomics ApplicationsACKNOWLEDGEMENTSREFERENCES |
|---|
A. PHOSPHOPROTEOME ANALYSIS BY PROTEIN RADIOLABELING
Protein labeling with either inorganic phosphate (32Pi) or -(32P)-ATP is still one of the most practical ways to study protein phosphorylation. In proteomics, a common approach is to label cells in vivo with 32Pi, then resolve the 32P-labeled proteins by 2-DGE. The major drawback to this approach is that although many phosphoproteins can be visualized by autoradiography, they are not present in sufficient amounts to be identified. To overcome this problem, we utilize methods to enrich for specific subproteomes prior to analysis by electrophoresis (Figure 5).
|
2. Protein Kinase Profiling in Vitro
Protein kinase profiling is a way to identify protein kinases and their substrates under different biological conditions. In this strategy, a cell extract is fractionated by ATP-Sepharose or ion-exchange chromatography and the individual fractions are incubated with kinase buffer containing -32P-ATP to allow labeling of phosphorylated proteins (Figure 5B). Individual fractions are resolved by 1-DGE and proteins that undergo changes in phosphorylation are selected by autoradiography. If specific proteins undergo changes in phosphorylation, they are excised from the gel and identified by MS. If individual proteins cannot be identified in the fraction, then the entire fraction can be reapplied to another ion-exchange column under different conditions. Alternatively, the fraction can be applied to -ATP-Sepharose to isolate protein kinases. The advantage to this approach is that a large number of samples can be screened easily and quickly for changes in protein phosphorylation and both the substrates and the protein kinases can be identified. One caveat to this approach is that because the phosphorylation is conducted in vitro, the results must be confirmed by in vivo studies.
B. METHODS FOR PHOSPHOPROTEOME ISOLATION
A major obstacle in the study of phosphorylated proteins is that phosphoproteins comprise only a small fraction of the total protein in a cellular lysate. As a result, many phosphoproteins cannot be identified in a cell extract. To address this problem, methods have emerged to enrich for phosphorylated proteins from a complex mixture of proteins. These techniques may facilitate the isolation and study of the entire phosphoproteome under different biological conditions.
1. Immobilized Metal Affinity Chromatography
Immobilized metal affinity chromatography (IMAC) has been used to purify phosphopeptides (Andersson and Porath, 1986) and, more recently, has been combined with MS to identify phosphoproteins from cell lysates (Posewitz and Tempst, 1999; Cao and Stults, 2000; Stensballe et al., 2001; Ficarro et al., 2002). However, in addition to binding phosphopeptides, IMAC is known to bind nonphosphorylated, acidic peptides through their carboxylate groups. To overcome this problem, Ficarro and colleagues converted the carboxylate groups in peptides to their corresponding methyl esters. This step was reported to reduce the binding of nonphosphorylated peptides to the IMAC matrix (Ficarro et al., 2002).
In the approach of Ficarro and colleagues, a protein extract is prepared from the sample and digested to peptides by the action of a protease. The resultant peptide mixture is treated with methanolic hydrochloric acid (HCl) to convert carboxylate groups to methyl esters and the mixture is applied to an IMAC column. The recovered peptides are separated further by high-performance liquid chromatography (HPLC) and analyzed online by an ion-trap mass spectrometer (Ficarro et al., 2002). However, if one assumes that approximately 30% of all proteins are phosphorylated (Hubbard and Cohen, 1993), then the total number of phosphoproteins in a cell lysate easily could exceed 1000. This approach requires that all phosphoproteins in a cell lysate (i.e., thousands of phosphoproteins) be analyzed to detect those that undergo changes.
2. Chemical Derivitization of Phosphorylated Amino Acids
One approach to isolate phosphorylated proteins or peptides is to take advantage of the unique chemistry of phosphoamino acids in peptides. Previously, Meyer and colleagues developed a method for the identification of phosphoserine in peptides by its conversion to S-ethylcysteine and Edman sequencing (Meyer et al., 1991). Under alkaline conditions, phosphoserine will undergo ß-elimination to dehydroalanine, which can be converted to S-ethylcysteine by Michael addition with ethanethiol. In our laboratory, we optimized conditions for the derivitization of phosphoserine with 1,2-ethanedithiol (EDT) for the purpose of fluorophore labeling of proteins (Fadden and Haystead, 1995). This same strategy can be used to attach biotinylated moieties to purify phosphoproteins or peptides. An example of this approach recently was provided by Oda and colleagues who, using the same chemistry described herein, converted phosphoserine residues to biotinylated residues in proteins. The biotinylated proteins were purified by avidin chromatography, digested to peptides, and identified by MS (Oda et al., 2001). In an alternate method, Zhou and colleagues devised a technique to isolate peptides phosphorylated on serine, threonine, and tyrosine residues (Zhou et al., 2001).
The main disadvantage of these two methods is that the current chemistries require significant amounts of protein or peptide for identification by MS to be successful. As a result, we have not found these methods to be a practical approach for isolation of phosphoproteins. In addition, the selectivity of these methods has not been confirmed. Several proteins were isolated using these methods that are not known phosphoproteins (e.g., glyceraldehyde 3-phosphate dehydrogenase). Therefore, further work needs to be done to confirm these results and to determine the specificity of this approach (Oda et al., 2001; Zhou et al., 2001).
C. IDENTIFICATION OF PROTEIN PHOSPHORYLATION SITES
In addition to identifying phosphoproteins, it is important to identify the phosphorylation sites in proteins, so that information can be obtained about the function of the phosphorylation event. The two predominant techniques for phosphorylation site analysis are Edman degradation and MS.
1. Phosphorylation Site Identification by Edman Degradation
Edman degradation is still one of the most practical methods to determine phosphorylation sites in peptides. This is because the technique is relatively simple, is very sensitive, and can be applied to a large variety of peptides (Aebersold et al., 1991; Boyle et al., 1991; Wettenhall et al., 1991). If enough radioactivity can be incorporated into the phosphoprotein of interest, sites can be determined at the sub-fmol level. In our hands, this can be as little as 1000 cpm (not ideal). In this approach, a 32P-labeled protein is digested with a protease and the resulting phosphopeptides are purified by reverse-phase HPLC or thin-layer chromatography (TLC) (Figure 6). The isolated peptides then are cross-linked via their C-termini to an inert membrane and the radioactive membrane is subjected to several rounds of Edman cycles. The radioactivity is collected after each cleavage step and the released 32P is measured in a scintillation counter. This method positionally places the phosphoamino acid within the sequenced phosphopeptide (Figure 6). Of course, this is meaningful only if the sequence of the phosphopeptide is known. The analysis ceases to become quantitative beyond 30 Edman cycles (even with efficient, modern Edman machines) due to well-understood issues with repetitive yield associated with Edman chemistry.
|
2. Phosphorylation Site Identification by MS
Phosphorylation sites in peptides can also be analyzed by MS/MS. In this approach, the phosphopeptide is sequenced in the mass spectrometer and the site of phosphorylation is determined unambiguously (Figure 6). Phosphopeptides can be identified from a mixture of peptides by a method known as precursor ion scanning (Neubauer and Mann, 1999). Serine and threonine phosphorylated peptides undergo loss of 98 Da (H3PO3-) and 80 Da (HPO3-) from the molecular ion. These ions can be used to identify phosphorylated peptides. Having identified a phosphopeptide, the peptide mixture is sprayed under acidic conditions and the phosphopeptide sequenced by conventional MS/MS. Upon fragmentation of the phosphopeptide, phosphoserine can be identified by the formation of dehydroalanine (69 Da), the ß-elimination product of phosphoserine. Similarly, phosphothreonine can be identified by the formation of its ß-elimination product, dehydroamino-2-butyric acid at 83 Da (Neubauer and Mann, 1999).
|
VI. Proteomics Applications |
|---|
|
TOPABSTRACTI. IntroductionII. The Need for Protein Studies in...III. A Proteomics Approach to Signal...IV. Protein IdentificationV. Phosphoproteome AnalysisVI. Proteomics ApplicationsACKNOWLEDGEMENTSREFERENCES |
|---|
subunit of pyruvate dehydrogenase.
We went on to validate these findings in a comprehensive biochemical study. Consistent with increased phosphorylation of Hxk2p in response to REG1 deletion, fractionation of yeast extracts by anion-exchange chromatography identified a Hxk2p phosphatase activity in wild-type strains that was selectively lost in the reg1
mutant. Having carried out these studies, we attempted to rescue the reg1 phosphoprotein phenotype by overexpressing both wild-type and mutant Reg1p in the deletion strains. Here, both the phosphorylation state of Hxk2p and Hxk2p phosphatase activity were restored to wild-type levels in the reg1 mutant by expression of a LexA-Reg1p fusion protein. In contrast, expression of a LexA-Reg1p protein containing mutations at phenylalanine in a putative PP-1C (the catalytic subunit) binding site motif (K/R)(X)(I/V)XF was unable to rescue Hxk2p dephosphorylation in intact yeast or restore Hxk2p phosphatase activity. These results demonstrate that Reg1p targets PP-1C to dephosphorylate Hxk2p in vivo and that the peptide motif (K/R)(X)(I/V)XF is necessary for its PP-1 targeting function. These studies, therefore, demonstrate how a proteomics approach can be used to first identify enzyme targets in cells, then direct further analysis to verify the findings. Clearly, a combined proteomics and genetics approach greatly enhances one’s ability to directly answer key biological questions. We believe that a similar strategy could be adopted with transgenic or knockout mouse work, particularly in cases where there is no obvious phenotype.
B. SMOOTH MUSCLE CELL PHYSIOLOGY
Proteomics, when combined with cell physiology, can be a valuable tool to understand basic cellular processes. In our laboratory, we apply proteomics techniques to study the regulation of smooth muscle contraction (SMC). SMC is regulated principally by free intracellular calcium levels and the level of myosin light-chain phosphorylation (Sellers and Adelstein, 1985). The steady-state level of myosin light-chain phosphorylation is, in turn, regulated by the relative activities of myosin light-chain kinase and myosin light-chain phosphatase (SMPP-1M). SMPP-1M itself is regulated by phosphorylation and its activity can be inhibited by phosphorylation of its targeting subunit. However, until recently, the protein kinase responsible for this phosphorylation was unknown. Therefore, we utilized a proteomics strategy to identify the relevant endogenous kinase responsible for phosphorylation and inhibition of SMPP-1M.
Because -ATP-Sepharose is capable of binding protein kinases (Haystead et al., 1993), we applied the entire SMC complement of proteins to -ATP-Sepharose. Using this affinity purification, we successfully obtained amino acid sequence for the previously unidentified MYPT1-kinase (MacDonald et al., 2001a). Further experiments, including addition of recombinant enzyme to permeabilized SMC, confirmed a role for the MYPT1 kinase as a regulator of SMC contractile state (MacDonald et al., 2001a,b; Borman et al., 2002).
We currently are using a functional proteomics approach to identify novel protein kinase substrates that are phosphorylated in response to agonist-induced smooth muscle relaxation. In our approach, permeabilized smooth muscle strips, contracted by treatment with calcium, are treated with 8-bromo-cGMP to induce relaxation. Cell extracts are prepared and proteins are resolved by 2-DGE. Proteins that undergo changes in phosphorylation in response to 8Br-cAMP are sequenced and identified by MS. Using this method, several known phosphoproteins — including HSP20, HSP27, and telokin (MacDonald et al., 2000; Walker et al., 2001; Borman et al., 2002) — were identified. In addition to these known proteins, we have identified novel phosphoproteins. Their role in smooth muscle relaxation is being investigated (J.A. MacDonald and T.A. Haystead, unpublished results).
C. PROTEOME MINING
Proteome mining is an iterative process in which a targeted proteome is screened against compound libraries to identify highly selective drug-protein interactions (Graves and Haystead, 2002). The advantage of this approach is the ability to identify a drug-like molecule from a compound library that is specific for a protein or group of proteins and simultaneously identify the protein targets of the drug. The process begins with the isolation of a specific subproteome from a cell, tissue, or animal source by application of saturating amounts of tissue extract to an affinity matrix (Figure 7). The affinity matrix is composed of a natural ligand immobilized in an orientation that favors interaction with its respective protein targets. For example, if the natural ligand is ATP, only those proteins that bind ATP or ATP-resembling molecules will be captured on the matrix. This can be validated by sequencing and identifying proteins that are specifically absorbed to the matrix. The captured proteome then is screened against large chemical libraries and, if a compound from the library is able to compete with the immobilized ligand for protein binding, the bound protein is displaced from the matrix and recovered. Since the drug has the potential to interact with all of the proteins bound to the matrix, information about drug specificity can be obtained by identifying the eluted proteins.
|
-ATP-Sepharose and screened the bound proteomes against several of the quinoline antimalarial drugs (e.g., chloroquine, primaquine). Interestingly, several quinoline antimalarial drugs eluted two proteins from the human red blood cell purine-binding proteome identified as aldehyde dehydrogenase and quinone reductase 2 (Graves et al., 2001). We are investigating the role of these proteins in the pathology of malaria. Thus, proteome mining can allow the identification of drug targets and further investigation of those targets can reveal the molecular action of the drugs in vivo.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTI. IntroductionII. The Need for Protein Studies in...III. A Proteomics Approach to Signal...IV. Protein IdentificationV. Phosphoproteome AnalysisVI. Proteomics ApplicationsACKNOWLEDGEMENTSREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTI. IntroductionII. The Need for Protein Studies in...III. A Proteomics Approach to Signal...IV. Protein IdentificationV. Phosphoproteome AnalysisVI. Proteomics ApplicationsACKNOWLEDGEMENTSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  H. M. Wu, M. Jin, and C. B. Marsh Toward functional proteomics of alveolar macrophages Am J Physiol Lung Cell Mol Physiol, April 1, 2005; 288(4): L585 - L595. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. J. Lee, Y. Yan, G. Marriott, and R. M Corn Quantitative functional analysis of protein complexes on surfaces J. Physiol., February 15, 2005; 563(1): 61 - 71. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. A. Wetmore and B. A. Merrick Invited Review: Toxicoproteomics: Proteomics Applied to Toxicology and Pathology Toxicol Pathol, October 1, 2004; 32(6): 619 - 642. [Abstract] [PDF]
|
|
|
|
 |
|
 K. S. Nair, A. Jaleel, Y. W. Asmann, K. R. Short, and S. Raghavakaimal Proteomic research: potential opportunities for clinical and physiological investigators Am J Physiol Endocrinol Metab, June 1, 2004; 286(6): E863 - E874. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||
