Van Eyk, Dunn - Proteomic and Genomic Analysis of Cardiovascular Disease - 2003 (522919), страница 75
Текст из файла (страница 75)
anion exchange chromatography). Individualfractions are then subjected to phosphorylationby the addition of protein kinase and 32P-ATP(3). Phosphoproteins are separated by 2DE (4)and identified by mass spectrometry (5).phasizing the need to validate the relevance of a phosphorylation event by furtherbiochemical study.Our prototypical application of the methodology has been in the delineation ofdistinct substrate pools for the individual Ca2+/CaM-dependent protein kinase(CaMK) family members (E.
E. Corcoran, J. Joseph, J. A. MacDonald, T. A. J. Haystead, & A. R. Means, data submitted for publication). Missing from many studiesof CaMK signaling is the differentiation between CaMKI and CaMKIV isozymes;the substrate preferences of CaMKI and CaMKIV are similar and intersect withthe so-called multi-functional CaMK, CaMKII [59]. Unfortunately, the kinase inhibitors currently available inhibit CaMKI, II, and IV similarly and so they cannotbe used alone to define a role for a specific CaMK isotype. Our preliminary studies have identified several novel substrates of CaMKI and IV.
Our analysis, bycomparing the overall similarities and differences between kinases, can answer15.4 Proteome Mining of the Smooth Muscle Cellmore general questions about the biological function of these enzymes. For example, if CaMKI and CaMKIV were essentially the same in their substrate preferences, differences in function could be wholly attributed to differences in localization and regulation by other pathways.One powerful application of functional proteomics is the ability to juxtaposewith genetics to delineate signaling events in vivo.
Currently, genetic manipulationis most easily accomplished in yeast. We have combined proteomics and yeast genetics to identify physiological targets of protein phosphatases. Protein phosphatase-1 (Glc7p) and its binding protein Reg1p are essential for the regulation ofglucose repression pathways in Saccharomyces cerevisiae [60]. In order to identifythe physiological substrates for the Reg1p/Glc7p complex, the effects of deletionof the REG1 gene on the yeast phosphoproteome were examined. Mapping of thephosphoproteome by 2DE revealed two distinct proteins that were greatly increased in phosphate content in REG1D mutants. Microsequencing identified oneof these proteins as hexokinase II (Hxk2p).
A comprehensive biochemical studyindicated that Reg1p targeted PP-1C to dephosphorylate Hxk2p in vivo and thatthe peptide motif (K/R)(X)(I/V)F was necessary for its PP-1C targeting function.The types of proteomic approaches that were utilized in this study to identifyHxk2p as a physiological target of the Reg1p/Glc7p complex herald a new age inwhich the functional consequences of genetic manipulations can be assessed directly and fully on the proteome as a whole. Logical genetic and biochemical experiments can be carried out to probe the function and address the physiologicalrelevance in the context of the intact cell. Clearly, a combined proteomics and genetics approach greatly enhances the ability to ascertain the molecular mechanisms underlying complex biological phenomena.
We believe that a similar strategy could be adopted for SMC systems with transgenic or knock-out mouse work,particularly where there is no obvious phenotype.15.4Proteome Mining of the Smooth Muscle CellGenomic powerhouses have invested billions to identify relevant drug targets byconventional high-throughput drug screening protocols. These screening effortshave exploited the power of combinatorial chemistry (combichem) by generatinglibraries containing millions of molecules.
Regrettably, the massive size of these libraries delayed the identification of new drugs due to the time required to screenthrough them. With reductional iterations the absolute size of combichem libraries has been minimized to a few hundred thousand diverse compounds. Drugscreening initially used these libraries to generate lead molecules. The leads thenserve as scaffolds for iterative substitution using rational combinatorial approaches to improve selectivity and bioavailability. The conventional high-throughput (HT) screen against a designated protein target is still the preferred path tonovel drug discovery.
A designated target, usually a cell or a protein, is used tocomplete a screen against a combichem library for a selective biological effect26927015 Investigations in Smooth Muscle Cell Physiology15.4 Proteome Mining of the Smooth Muscle Cell(i.e., modulation of enzyme activity, inhibition of cell growth, reduction in parasite/viral infection etc.). Many compounds are identified as preliminary candidates, but these often fail in cell and animal studies because they are non-selective or non-reproducible. Ultimately the failure of traditional drug screening methodology rests on the fact that there is no clue as to the mechanism of action [61].Invariably, when a molecule finds its way to clinical trials, deleterious side-effectsare revealed which eliminate the molecule as a drug candidate.
Serendipity manyplay a role in determining a drug”s ultimate success by identifying the real biological target via some unexpected but desirable side effect [62]. The list of drugs inwhich serendipity played a role in the final clinical outcome includes the majorityof our pharmaceutical repertoire (a highly popularized example being ViagraTMand penile erectile dysfunction).The more information we have concerning the biology of different SMC diseases the more we are able to reduce the role of serendipity in the discovery ofnew pharmacotheraputics.
Toward this end, the various genome projects will playa significant role by identifying a host of new genes in SMC. However, the benefitof a strictly genomic-based approach to the identification of new targets in SMCdisease is not so obvious. Recent advances in DNA sequencing, genotyping, andDNA microarray chip technologies enable the identification of specific nucleotidepoint mutations (SNPs) in genes that can, in principle, be the molecular basis ofdisease. One scenario is to sequence the entire genome of each disease sub-population and compare it with the genomes of individuals without the disease. Amonumental task that will most surely identify many genomic differences between the individuals, any one of which could be the molecular basis for the disease, but with equal probability could be harmless polymorphisms.
In addition,the vastness of the human genome may obscure the defective gene. Genomicstrategies make sense as long as one gene produces one messenger RNA that inturn codes for one protein, as the conventional dogma has dictated. But genesclearly do not tell the whole story. The etiology of diseases involve ill-defined environmental factors that can trigger inappropriate activation of rogue genes. Thesefactors cannot be measured by examining the genome alone.3Fig. 15.6 A. The effects of REG1 deletion onthe yeast phosphoproteome.
Yeast were labeled to steady state with [32P] orthophosphate and whole cell extracts characterizedby 2DE and autoradiography. Spots of interestwere identified by microsequencing. Molecular weight (kDa) and isoelectric points (pI)are theoretical values calculated by the Expasycompute pI/MW tool (http://expasy.hcuge.ch/ch2d/pi_tool.html) and are derived from theprimary amino acid sequence of the identifiedprotein. Spots C and D are exposure artifactsare were not reproduced in additional experiments. B.
Identification of REG1 as a PP-1binding protein by biochemical study. Assayof phosphatase activity toward Hxk2p in yeastextracts. Cell extracts were prepared fromwild-type and REG1D yeast and fractionatedby anion-exchange chromatography. Columnfractions were assayed for Hxk2p phosphatase activity. Column fractions were also Western blotted with rabbit anti-PP-1C antibody.(Redrawn from [60].)27127215 Investigations in Smooth Muscle Cell PhysiologyThe foundation underlying the recent growth of the proteomic field was themass of genomics data delivered by the Human Genome Project and the majoradvances in bioinformatics in the late 1990s.
With the newfound ability to identifyalmost any protein isolated from a tissue sample, researchers are able to devise logical genetic and biochemical experiments to probe the function and address thephysiological relevance in the context of the intact cell. Genomics and proteomicsare now being used as synergistic partners rather than competitors in drug discovery. A mixture of genomic and proteomic technologies are increasingly being usedto identify relevant drug targets for conventional HT screens, but they do nothingto provide key information needed to further advance the HT screen.
The conventional HT screen (proteomic or genomic) lacks information on possible serendipitous interactions with other protein targets. Also, the true value of the combichemlibrary is not fully exploited. Theoretically, a structurally diverse combichem library contains within it every molecular shape that could fit selectively into abinding site on every protein target in the cell. By screening with a single proteintarget rather than with the entire cellular complement of proteins (i.e., the proteome), important biological information is lost for future analysis. Technologiesneed to be developed that maximize the information obtained from HT combichem library screening when coupled with genomic and proteomic protocols.Proteome mining technologies may provide a solution to this challenge [62].The principles of proteome mining are based on the assumption that all drug-likemolecules selectively compete with a natural cellular ligand for a binding site on aprotein target.
In a proteome mine, natural ligands are immobilize on beads athigh density and in an orientation that sterically favors interaction with their protein targets. The immobilized ligand is exposed to whole animal or tissue extract,and bound proteins are evaluated for specificity by protein sequencing by massspectrometry. In the prototypical example in our laboratory, the proteome mine ischarged with ATP ligand.
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
