Proteomics: a new diagnostic frontier.
Although the term "proteomics" originated in 1994, the roots of this field are closely entwined with the history of clinical chemistry. The first laboratory test for a protein cancer marker, the Bence Jones protein in urine, was described in 1847. Over the succeeding years, clinical laboratories have developed tests for quantitative and qualitative analysis of hundreds of proteins in biological fluids by use of activity assays, electrophoresis, chromatography, immunoassays, and serologic procedures. There is also a long history of clinical applications of protein electrophoresis, including 50 reports describing applications of 2-dimensional gel electrophoresis in an entire issue of Clinical Chemistry in 1982. For the most part, however, traditional laboratory tests have analyzed one protein at a time. Recent progress in proteomics offers the potential to analyze hundreds of components at a time and to identify proteins and their structural modifications in minuscule specimens. Technologic advances have greatly expanded possibilities in the diagnostic analysis of proteins.
Complexity of the Proteome and Peptidome
The complexity of protein and peptide mixtures in biological fluids such as serum or plasma currently far exceeds the ability to separate different components (1-3). The highest resolution techniques for analysis of proteins can separate a maximum of a few thousand components. There are an estimated 25 000 genes in the human genome, and each gene can yield multiple protein products because of variable splicing and posttranslational processing. Attempts to analyze the maximum number of protein components, therefore, increasingly rely on multidimensional chromatography or other approaches using multiple fractionation steps (1-8).
Proteomic approaches traditionally have been divided into top-down and bottom-up approaches, with some of the characteristics of these approaches listed in Table 1. Both approaches rely heavily on recent advances in mass spectrometry (MS)  (9). The top-down approaches try to separate as many intact proteins as possible by techniques such as 2-dimensional gel electrophoresis, matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) MS, and surface-enhanced laser desorption/ionization (SELDI)-TOF MS. Increased numbers of components are being detected through increasingly complex fractionation of specimens before the final analysis and by increasing the number of dimensions (fractionation steps) (8). Bottom-up approaches start by breaking proteins into small peptides, such as by trypsin digestion. The resulting peptides then are fractionated by chromatographic techniques and analyzed by MS.
Top-down and bottom-up approaches each have advantages and disadvantages. Top-down approaches provide information about the variation, relative quantities, and structures of the proteins that is lacking in the analysis of individual peptides in the bottom-up approach. As an example, analysis of plasma proteins by 2-dimensional gel electrophoresis will yield multiple spots for most proteins, representing variable glycosylation, proteolytic degradation, or other posttranslational modifications, and spot intensities and positions provide information about abundance, molecular size, and isoelectric point. Bottom-up approaches exploit the higher chromatographic and mass spectrometric resolution achieved for small peptides than for large proteins, and the sequences of small peptides can be determined in a very high throughput fashion during analysis of peptides by this approach. Usually, this has enabled the sequence identification of a larger number of proteins than by the top-down approach, but the bottom-up approach does not identify what molecular form of the parent protein served as the source of the peptide. Sequence analysis by the top-down approaches tends to involve more laborious analysis of one protein at a time, although characteristics of the protein such as molecular size and isoelectric point may be known. Some of the top-down approaches, such as profiling by SELDI-TOF or MALDI-TOF MS do not provide sequence identification of peaks in an analysis.
The Plasma Proteome Project found that different proteins were detected by different approaches used by 18 different laboratories (3). A total of 9504 proteins were identified by sequencing of at least 1 peptide. Laboratories using 2-dimensional gels as the sole separation procedure after immunodepletion of abundant proteins identified <200 proteins per specimen, and laboratories applying pure bottom-up approaches found up to ~1000 proteins per specimen. Only one laboratory identified more than 1500 proteins, using a combination of immunodepletion and protein fractionation followed by capillary HPLC-ion trap MS of tryptic peptides. This serves as an illustration of the potential for hybrid approaches that combine the top-down fractionation of intact proteins with subsequent bottom-up analysis of protein-containing fractions. Most of the 9504 protein identifications were based on sequencing of 1 peptide; only 3020 proteins were identified by 2 or more peptides. This set of 3020 was considered to have confirmed identification and serves as the Core Protein Dataset for the HUPO Plasma Proteome Project (3).
Dynamic Concentration Range of Proteins in Plasma
The second major challenge of proteomics, beyond the issue of specimen complexity, is the problem of dynamic range. The concentrations of proteins in plasma vary from millimolar for albumin to around femtomolar for low-abundance components (1-3). This represents a [10.sup.12]-fold concentration range, far exceeding the dynamic range of most analytical methods, which typically have dynamic ranges on the order of 100- to 10 000-fold. Immunodepletion of abundant protein components or other fractionation techniques offer approaches for trying to deal with the dynamic range problem as well as with the complexity issue. Removal of the most abundant components by immunodepletion or fractionation before analysis by techniques such as 2-dimensional electrophoresis has markedly expanded the number of lower abundance components that can be detected (1-8).
Many potential diagnostic applications of proteomics depend on quantification of components in addition to their identification. Many new approaches relying on specific labeling or use of internal standards are being developed for either comparative or absolute quantitative analysis of specific components. Use of stable-isotope-labeled peptides or proteins as internal standards in tandem MS appears to offer an approach for quantitative analysis of proteins that uses technologies and approaches similar to those used currently in clinical laboratories for therapeutic drug and metabolite analysis (9,10).
Analysis of large numbers of proteins in a single analysis by proteomic technologies has greatly accelerated the ability to identify new biomarkers for disease processes ranging from cancer to diabetes, stroke, and kidney diseases (4,5,11-14). Proteomic analysis of postmortem cerebrospinal fluid identified nucleoside diphosphate kinase A and PARK7 as proteins released by brain injury and as potential new biomarkers for stroke (13). New biomarkers also may help identify physiologic disorders, such as the example provided of potential biomarkers of anoxia in sleep apnea (15). Analyses of serum or plasma concentrations of proteins such as myoglobin, creatine kinase MB isoenzyme, and cardiac troponins have played an important role in the diagnosis of myocardial infarction. New tools for protein analysis provide approaches for analyzing different molecular forms of traditional markers as well as identifying additional proteins that may serve as biomarkers for heart disease (16,17)
Some potential biomarkers have been determined to be fragments of abundant protein components. A diverse array of protein degradation fragments appear to accumulate in plasma, bound to carrier proteins such as albumin (18-20). Analysis of the total complement of plasma peptides, what has been termed the plasma fragmentome or peptidome, shows promise as a source of new biomarkers. Accumulation of selected peptides may depend on pathways of proteinase activation that are affected by pathophysiologic processes. Some strategies for analysis of the peptidome rely on initial selection of a specific carrier protein such as albumin and analysis of the peptides bound to that carrier. This contrasts with many proteomic analyses, which have used strategies for selective depletion of albumin to detect lower abundance components. It is now apparent that depletion strategies may remove more than the specific intended component.
Proteomic and peptidomic approaches clearly appear to have a valuable role in marker discovery, but it is not yet clear whether these approaches will be used as the platforms for analysis in the clinical laboratory. Standardizing these methods and establishing consistent performance between different laboratories have presented a challenge (21). Implementation of these methods still has not achieved the usual standards of practice for clinical laboratory methods (22). Standardization and thorough understanding of preanalytical factors, including patient preparation and specimen collection and processing techniques, clearly are important factors in discovery efforts and potential clinical applications of tests (3,22). Many specimen collection variables have been examined, but no consensus has been reached about the optimal specimen type. The Specimens Committee of the Plasma Proteome Project concluded that platelet-depleted plasma is preferable because of greater stability, but this is partly counterbalanced by the greater availability of serum specimens in specimen banks for discovery efforts (3, 23). Specimen storage in liquid nitrogen or at -80[degrees]C is recommended (3).
From Research to Clinical Application
Basic research is identifying new biomarkers, and there are initiatives to support multicenter validation of new biomarkers. However, the process of translation to clinical laboratory use has been complex, and, to date, few new biomarkers have progressed to clinical laboratory use (1). This delay may reflect the many steps in the translation of a newly identified biomarker into a useful clinical laboratory test. Analogous to drug development, discovery may be one of the quickest and easiest steps in the test development process. Validation of markers is a challenging process requiring multisite clinical studies (12, 21).
Currently, the pipeline for translation of new biomarkers into tests appears to have a bottleneck at the early stages of translation of research markers into clinical tests. Research groups performing discovery and clinical studies rarely have the resources to develop prototype analyzers or test reagent sets, to manufacture them, or to proceed with other steps in commercialization. These steps usually rely on the in vitro diagnostics industry, which has had relatively low investment in the development of new markers. After development of tests, there is a need for evaluation in clinical laboratories, submission for approval by the US Food and Drug Administration (24), establishment of reimbursement rates by the Medicare system and insurers, and education of physicians about test ordering and interpretation. Testing in the clinical environment requires extensive standardization and development of detailed quality assurance processes (22, 25). The process of translating new markers into clinical laboratory tests entails contributions from multiple disciplines, including scientists; engineers; business, legal, and regulatory professionals; clinicians; and clinical laboratorians. Clinical chemistry professionals can provide important contributions to many of the steps in translation of new markers into clinical laboratory tests. The large number of well-validated quantitative tests for individual proteins that are currently provided by clinical laboratories also may have a role in helping to standardize proteomic analyses (26) or to provide information that should be integrated with the proteomic analysis.
Targeted Proteomics and Qualitative Variation and Proteins
A more immediate impact of proteomics on the clinical laboratory than the pipeline of new products from proteomic discovery efforts may be applications of MS to examine qualitative variations of proteins. Some examples that are already in clinical laboratory use include analysis of genetic variants of the protein transthyretin and detection of carbohydrate-deficient transferrin (27, 28). These might be considered applications of targeted proteomics, i.e., analysis of structural changes in specific proteins. Antibody capture serves as a means to enrich target proteins for subsequent analysis by MS (16, 27-29). The increased ability to identify structural variations of specific proteins may lead to development of a greater range of tests that analyze qualitative changes in proteins. Another development that is expected to arise in part from proteomic investigations is increased availability of multiplex immunoassays. These have already appeared in various forms in the clinical laboratory as limited panels, and rapid growth is occurring in the research applications of multiplex assays. Multiplex assays are likely to pose many new challenges for quality assurance and assay validation that are not seen with traditional laboratory testing of one component at a time (25).
Proteomics in the Clinical Laboratory
Although proteomics is expected to have a large impact on the clinical laboratory, it is not yet clear whether this will appear as a series of new markers adapted to current testing technologies or as new technology platforms appearing in the clinical laboratory. MS offers great potential for both quantitative and qualitative analysis of proteins that may open new dimensions for analysis of proteins in the clinical laboratory. Clinical laboratory experience with quantitative analysis of immunosuppressant drugs by MS may serve as a useful model for quantitative analysis of peptides. Considering that the current menu of a few hundred clinical laboratory tests for specific proteins analyzes fewer than 1% of all gene products and only a small sampling of protein modifications, there is great unexplored diagnostic potential in the analysis of proteins.
This minireview is based on presentations at a conference entitled, "Proteomics: A New Diagnostic Frontier", held October 24-25, 2005, in Washington, DC. The conference was organized by the AACC Proteomics Division, with co-sponsorship by the US Human Proteome Organization (US HUPO). Introductory comments from Mitchell Scott, AACC President, and Glen Hortin, Chair of the AACC Proteomics Division, described historical roots of protein analysis in the field of clinical chemistry. Sixteen speakers reviewed progress in clinical proteomics as well as pathways and processes for bringing the new technologies and markers into application in the clinical laboratory. Overviews of proteomics and technologies were provided by N. Leigh Anderson, Denis Hochstrasser, and Eleftherios P. Diamandis in a session moderated by Roland Valdes, Jr. Approaches for discovering biomarkers for cancer detection were described by Samir Hanash, Lance A. Liotta, and Daniel W. Chan. New approaches for biomarker discovery for cardiovascular and kidney disease and sleep apnea were presented by William S. Hancock, Jon B. Klein, and Saeed A. Jortani. US HUPO organized a session in which Jennifer Van Eyk and Catherine Fenselau described proteomic studies of cardiovascular disease and acquired drug resistance, and Gilbert S. Omenn summarized recent findings of the HUPO Plasma Proteome Project. In the final session, Glen Hortin, Steven I. Gutman, Stephen R. Master, and John F. O'Brien examined important issues in the translation of newly discovered markers into clinical laboratory tests. Panel discussions included contributions from Bruce C. Haywood, Mary F. Lopez, and Gail S. Page in addition to the speakers named above.
(1.) Anderson NL, Anderson NG. The human plasma proteome: history, character, and diagnostic prospects. Mol Cell Proteomics 2002; 1:845-67.
(2.) Lescuyer P, Hochstrasser DF, Sanchez JC. Comprehensive proteome analysis by chromatographic protein prefractionation. Electrophoresis 2004;25:1125-35.
(3.) Omenn GS, States DJ, Adamski M, Blackwell TW, Menon R, Hermjakob H, et al. Overview of the HUPO Plasma Proteome Project: results from the pilot phase with 35 collaborating laboratories and multiple analytical groups, generating a core dataset of 3020 proteins and a publicly-available database. Proteomics 2005;5:3226-45.
(4.) Yang Z, Hancock WS, Chew TR, Bonilla L. A study of glycoproteins in human serum and plasma reference standards (HUPO) using multilectin affinity chromatography couple with RPLC-MS/MS. Proteomics 2005;5:3353-66.
(5.) Wang H, Clouthier SG, Galchev V, Misek DE, Duffner U, Min CK, et al. Intact-protein-based high-resolution three-dimensional quantitative analysis system for proteome profiling of biological fluids. Mol Cell Proteomics 2005;4:618-25.
(6.) Sheng S, Chen D, Van Eyk JE. Multidimensional liquid chromatography separation of intact proteins by chromatographic focusing and reversed phase of the human serum proteome: optimization and protein database. Mol Cell Proteomics 2006;5:26-34.
(7.) Rahbar AM, Fenselau C. Unbiased examination of changes in plasma membrane proteins in drug resistant cancer cells. J Proteome Res 2005;4:2148-53.
(8.) Anderson NL, Polanski M, Pieper R, Gatlin T, Tirumalai RS, Conrads TP, et al. The human plasma proteome: a nonredundant list developed by combination of four separate sources. Mol Cell Proteomics 2004;3:311-26.
(9.) Diamandis EP. Mass spectrometry as a diagnostic and a cancer biomarker discovery tool: opportunities and potential limitations. Mol Cell Proteomics 2004;3:367-78.
(10.) Anderson L, Hunter CL. Quantitative mass spectrometric multiple reaction monitoring assays for major plasma proteins. Mol Cell Proteomics 2006;5:573-88.
(11.) Calvo KR, Liotta, Petricoin EF 3rd. Clinical proteomics: from biomarker discovery and cell signaling profiles to individualized personal therapy. Biosci Rep 2005;25:107-25.
(12.) Li J, Orlandi R, White CN, Rosenzweig J, Zhao J, Seregni E, et al. Independent validation of candidate breast cancer serum biomarkers identified by mass spectrometry. Clin Chem 2005;51:2229-35.
(13.) Allard L, Burkhard PR, Lescuyer P, Burgess JA, Walter N, Hochstrasser DF, et al. PARK7 and nucleoside diphosphate kinase A as plasma markers for the early diagnosis of stroke. Clin Chem 2005;51:2043-51.
(14.) Merchant ML, Klein JB. Proteomics and diabetic nephropathy. Curr Diab Rep 2005;5:464-9.
(15.) Shah ZA, Jortani SA, Tauman R, Valdes R Jr, Gozal D. Serum proteomic patterns associated with sleep-disordered breathing in children. Pediatr Res 2006;59:466-70.
(16.) Labugger R, Simpson JA, Quick M, Brown HA, Collier CE, Neverova I, et al. Strategy for analysis of cardiac troponins in biological samples with a combination of affinity chromatography and mass spectrometry. Clin Chem 2003;49:873-9.
(17.) McDonough JL, Van Eyk JE. Developing the next generation of cardiac markers: disease-induced modifications of troponin I. Prog Cardiovasc Dis 2004;47:207-16.
(18.) Lowenthal MS, Mehta AI, Frogale K, Bandle RW, Araujo RP, Hood BL, et al. Analysis of albumin-associated peptides and proteins from ovarian cancer patients. Clin Chem 2005;51:1933-45.
(19.) Liotta LA, Petricoin EF 3rd. Serum peptidome for cancer detection: spinning biological trash into diagnostic gold. J Clin Invest 2006; 116:26-30.
(20.) Lopez MF, Mikulskis A, Kuzdzal S, Bennett DA, Kelly J, Golenko E, et al. high-resolution serum proteomic profiling of Alzheimer disease samples reveals disease-specific, carrier-protein-bound mass signatures. Clin Chem 2005;51:1946-54.
(21.) Semmes OJ, Feng Z, Adam BL, Banez LL, Bigbee WL, Campos D, et al. Evaluation of serum protein profiling by surface-enhanced laser desorption/ionization time-of-flight mass spectrometry for the detection of prostate cancer: I. Assessment of platform reproducibility. Clin Chem 2005;51:102-12.
(22.) Hortin GL. Can mass spectrometric protein profiling meet desired standards of clinical laboratory practice? [Editorial]. Clin Chem 2005;51:3-5.
(23.) Rai AJ, Gelfand CA, Haywood BC, Warunek DJ, Yi J, Schuchard MD, et al. HUPO Plasma Proteome Project specimen collection and handling: towards the standardization of parameters for plasma proteome samples. Proteomics 2005;5:3262-77.
(24.) Hackett JL, Gutman SI. Introduction to the Food and Drug Administration (FDA) regulatory process. J Proteome Res 2005;4: 1110-3.
(25.) Master SR. Diagnostic proteomics: back to basics? [Editorial]. Clin Chem 2005;51:1333-4.
(26.) Rossi L, Martin BM, Hortin GL, White RL, Foster M, Moharram R, et al. Inflammatory protein profile during systemic high dose interleukin-2 administration. Proteomics 2006;6:709-20.
(27.) Bergen HR, Lacey JM, O'Brien JF, Naylor S. Online single-step analysis of blood proteins: the transferrin story. Anal Biochem 2001;296:122-9.
(28.) Bergen HR 3rd, Zeldenrust SR, Butz ML, Snow DS, Dyck PJ, Klein CJ, et al. Identification of transthyretin variants by sequential proteomic and genomic analysis. Clin Chem 2004;50:1544-52.
(29.) Zhu Y, Valdes R Jr, Jortani SA. Application of bioaffinity mass spectrometry for analysis of ligands. Ther Drug Monit 2005;27: 694-9.
GLEN L. HORTIN,  * SAEED A. JORTANI  JAMES C. RITCHIE, JR.,  ROLAND VALDES, JR.,  and DANIEL W. CHAN 
 Department of Laboratory Medicine, Intramural research program of the NIH Clinical Center, National Institutes of Health, Bethesda, MD.
 Department of Pathology and Laboratory Medicine, University of Louisville, Louisville, KY
 Department of Pathology, Emory University, Atlanta, GA.
 Department of Pathology, Johns Hopkins University, Baltimore, MD.
 Nonstandard abbreviations: MS, mass spectrometry; MALDI-TOF, matrix-assisted laser desorption/ionization time-of-flight; and SELDI, surface-enhanced laser desorption/ionization.
* Address correspondence to this author at: Department of Laboratory Medicine, NIH, Bldg 10, Room 2C-407, Bethesda, MD 20892-1508. Fax 301-4021885; e-mail email@example.com.
Received January 27, 2006; accepted April 7, 2006.
Previously published online at DOI: 10.1373/clinchem.2006.067280
Table 1. Top-down vs bottom-up proteomics. (a) Top-down proteomics Bottom-up proteomics Approach Separation of intact Digestion of all proteins proteins; analysis of to peptides; separation isolated proteins and analysis of peptides Examples Two-dimensional Tryptic digestion electrophoresis or followed by liquid multidimensional chromatography--ion chromatography followed trap MS by tryptic digestion of isolated proteins and analysis of tryptic peptides by MS Profiling of proteins by MALDI-TOF or SELDI-TOF MS Advantages Identification of High efficiency of variations of intact chromatographic proteins separations of peptides Quantitative information High-throughput sequence about forms of proteins identification Information about protein Larger number of proteins characteristics, e.g., identified size and pI Disadvantages Lower throughput sequence Lack of information on identification variation of intact Fewer proteins identified forms of proteins Incomplete structural information on forms of a protein serving as the source of peptides No sequence identification by some profiling methods (a) Either approach may be augmented by prior depletion of the most abundant plasma components or other preliminary fractionation steps.
|Printer friendly Cite/link Email Feedback|
|Author:||Hortin, Glen L.; Jortani, Saeed A.; Ritchie, James C., Jr.; Valdes, Roland, Jr.; Chan, Daniel W.|
|Date:||Jul 1, 2006|
|Previous Article:||The battle against infectious diseases in developing countries: the inseparable twins of diagnosis and therapy.|
|Next Article:||The MALDI-TOF mass spectrometric view of the plasma proteome and peptidome.|