Van Eyk, Dunn - Proteomic and Genomic Analysis of Cardiovascular Disease - 2003 (522919), страница 102
Текст из файла (страница 102)
In subsequent studies, improvements to the MudPIT system have demonstrated a dynamic range of 10,000:1 between the most and least abundant proteins in a mixture, i.e. cell lysate, which translates to a level of detection at low-femtomole levelsor 100 copies/cell [32].
The MudPIT technology potentially provides a robust alternative in proteome analysis though it is limited by its qualitative properties, lowsample throughput and constraints in data management.Recently, an experimental strategy was described for systematically sequencingand quantifying proteins in complex mixtures based on differential guanidinationof C-terminal lysine residues on tryptic peptides followed by capillary LC/ESI-MS[33]. This approach, termed mass-coded abundance tagging or MCAT, is simple,economic, sensitive in the pmol to fmol range, and can be applied to large-scalestudies of relative protein abundance present in distinct cell states under the effects of mutation, disease, or pharmacological treatment. Furthermore, the MCATapproach can be performed in tandem with multidimensional separation techniques, such as MudPIT, to achieve an even more detailed and robust proteomicanalysis.Another emerging labeled tag technology for the rapid comparative analysis ofproteomes is capillary isoelectric focusing (CIEF) coupled with Fourier transformion cyclotron resonance mass spectrometry (FTICR) being developed by RichardSmith’s group at Pacific Northwest National Labs [34, 35].
This approach claims toprovide a means of characterizing large numbers of proteins/peptides (i.e., ~1,000proteins from > 100,000 peptides detected) with exceptional resolution, highermass measurement accuracy (i.e., *1 ppm), and greater sensitivity than possiblewith conventional MS-based strategies described above [36, 37]. The ESI-FTICRMS approach has also been applied recently to the identification of phosphopeptides [38].36736821 Proteomics in Pharmaceutical R & DFicarro et al., working in conjunction with MDS Proteomics, described an interesting approach to “phosphoprofiling” in yeast [39].
Proteins derived from wholecell lysates were digested and resulting peptides converted to methyl esters, enriched for phosphopeptides by immobilized metal-affinity chromatography orIMAC, and analyzed by nanoflow HPLC/ESI-MS. This method has the apparentadvantage of eliminating nonspecific binding, thereby allowing for the identification of hundreds of phosphorylation sites in a single analysis with a sensitivity of5 fmol. Indeed, the method achieves a level of detection sufficient to identify rareprotein phosphorylation, such as tyrosine phosphorylation and of proteins withlow codon bias or abundance.Researchers at MDS Proteomics, as well as at Cellzome, have independently developed technologies with the ability to characterize hundreds of distinct multiprotein complexes as described in two new large-scale studies [40, 41]. These approaches involved the tagging of individual proteins in Saccharomyces cerevisiaethat in turn were used to pull down associated proteins for identification by MS.Ho et al.
detected over 3,600 associated proteins covering 25% of the yeast proteome with high-throughput mass spectrometric protein complex identification orHMS-PCI, an approach that only used 10% of the predicted yeast proteins as bait[41]. The HMS-PCI data set revealed a high degree of cellular connectivity and anaverage 3-fold higher success rate in detection of known complexes comparedwith large-scale two-hybrid studies. Gavin et al. identified over 1,400 distinct proteins within 232 multi-protein complexes using tandem affinity purification orTAP [40]. In doing so, they predicted the function of many previously unknownproteins, principally through homology, and more importantly, reconstructed aproteome network of functional units.
While these two new approaches are clearlypowerful, they do suffer from significant number of false-positive interactions (ashigh as 30% according to Gavin et al. [40]), and a failure to identify many knownassociations. Nonetheless, the HMS-PCI and TAP approaches, in conjunction withothers hold great promise and should make it feasible to characterize all cellularproteins interactions, though this will surely be labor intensive.A promising technology for rapid protein pattern analysis developed by Ciphergen Biosystems is surface-enhanced laser desorption ionization TOF (SELDI-TOF)mass spectrometry [42, 43].
Protein samples are first prepared, directly appliedand separated on solid supports, or ProteinChipsTM, engineered with hydrophobic, normal-phase, metal-affinity, and cationic or anionic bait surfaces. Mass signature “peaks” derived from the MS are displayed as a standard chromatograph butroutine identification of the protein peaks is not yet achievable. Nonetheless, withthis technique, peptide and protein profiles are easily obtained within minutesand hundreds of proteins can be profiled simultaneously from a small number ofcells or complex mixtures (tissue of body fluids). It also has the advantage of having greater sensitivity for protein < 25 kDa, but on average resolves approximatelyonly half the number of proteins compared to 2-DE.
The SELDI technology maybe an important tool for the molecular fingerprinting of disease samples, especially for providing insights into protein expression changes as potential diagnostic and prognostic markers.21.3 Target IdentificationLastly, the use of CALI (chromophore-assisted laser inactivation) in protein target validation should be mentioned here. This approach involves the conversionof a nonfunction-blocking antibody into a neutralizing molecule through photochemical modifications [44, 45].
A temporal and locally restricted protein “knockout” is achieved in cell-based assays that results in systematic inhibition of function. This is followed by subsequent identification of the target proteins by MSand other techniques. Xerion Pharmaceuticals has adapted the CALI approach onan industrial scale with a technology termed XCALIbur that combines CALI withautomation (i.e., for 96- and 384-well platforms), combinatorial binder generationand cell-based assays.
Most recently, this approach has been modified to applyfluoroscein-labeled probes for inactivation referred to as fluorophore-assisted lightinactivation (FALI) [46]. The use of CALI in a disease-based setting precludesprior knowledge of a target, takes into account post-translational modifications,and can be applied to both cell-free as well as cell-based assays, demonstrating itsversatility for validating disease relevant targets.Clearly, it is anticipated that the aforementioned quantitative analytical technologies will become fully integrated and adopted for the high-throughput, automated,global analysis of protein expression profiling.
Additional applications and refinements to bioanalytical separations for protein profiling are certainly expected inthe near future and will no doubt assist in revolutionizing the proteomics field.21.3Target IdentificationA proteomics approach offers great advantages in identifying potential moleculartargets. Strategies for target-driven drug discovery and subsequent rational drug design require identifying key cellular proteins that are causally related to disease processes and the validation of such molecular mechanisms as targets for therapeuticintervention.
As noted previously, the functions of gene products are frequently unknown and many proteins undergo post-translational modifications that greatly influence their biological properties. The value of proteomics is in its ability to providea global way of understanding molecular mechanisms involved in a defined systemas well as their interactions. Well-conceived biological questions and carefully designed experimental paradigms that have measurable physiological and molecularchanges will allow for the elucidation of the interrelationships between expressedproteins.
The development of a functional protein network database begins to establish the means to identify therapeutic targets of consequence.21.3.1Cardiovascular DiseasesThe complex nature of cardiovascular (CV) diseases offers great potential for discoveries of novel therapeutic targets by applying proteome-wide analysis to humanand animal models of CV pathophysiology. However, it is just such complexity36937021 Proteomics in Pharmaceutical R & Dthat underscores the many challenges, such as cellular heterogeneity, awaitingproteomic studies in CV biology. The assessment of various models of CV diseaseand dysfunction have revealed changes in cardiac protein expression patterns, andidentified a significant number that were previously unknown. For comprehensiveand recent reviews the reader is directed to Dunn [47], Arrell et al.
[48], Van Eyk[49], and Macri and Rapundalo [50]. A few notable CV-related proteomic studiesare highlighted in the sections below.One area of CV disease research that has received significant attention throughproteomics is cardiomyopathy or progressive heart failure. A number of groups including those led by Michael Dunn (London, UK), Peter Jungblutt, and JoachimKlose (the latter two in Berlin) have been pioneers in characterizing cardiac proteinprofiles from human, rat, canine and bovine models of cardiomyopathy and establishing public databases of identified cardiac proteomes.A number of other studies have utilized 2-DE protein profiling of myocardialtissue and cells in an effort to elucidate the cellular and molecular mechanismsassociated with cardiac hypertrophy and failure. Arnott et al. at Genentech observed that phenylephrine-induced hypertrophy in neonatal rat cardiomyocyteswas associated with substantial protein expression changes including molecularchaperones, ubiquitin-related and cytoskeletal proteins, and enzymes associatedwith cellular metabolism [51].
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
