Analyzing proteomic expression in a clinical screening environment using mass spectrometry

The age of the ‘omics’ has clearly arrived in biochemical research and clinical chemistry laboratories. The impact of the first of these, genomics, is felt in the everyday practice of clinical and laboratory science. With the recent elaboration of the human genome, there are few diagnostic evaluations that fail to include an investigation into whether a particular mutation exists for a specific disease, whether that mutation is correlated with severity or outcome, and how the mutation might affect therapeutic considerations. The difficulties with genomics are the sheer numbers of mutations that exist for many diseases and the inadequacy of laboratory tests to detect every mutation for a particular disorder without becoming prohibitively expensive. Furthermore, an abnormal gene is not necessarily the causative agent unless it is expressed.

Enter proteomics. It has been recognized that the translation of the gene to a final product may produce a myriad of possibilities from severe, disabling disease to mild symptoms or no apparent effect at all. The answer to explain this most likely resides in protein expression. Will an abnormal protein always produce disease? Not necessarily, since, like DNA mutations, amino acid alterations in the structure of proteins may have little effect on function or activity if it is not at the active site or if the alteration is minor (e.g., substitution of a leucine for isoleucine). Nevertheless, there is much to be gained from studying proteins from a clinical perspective, particularly in the area of biomarker research and screening.

Mass spectrometry plays a key role as a tool for the identification and characterization of proteins [1]. Its use expanded significantly in the 1990s with the widespread introduction and use of new ionization techniques such as electrospray and matrix-assisted laser desorption/ionization (MALDI). In fact, the Nobel Prize in Chemistry in 2003 was awarded to two mass spectrometrists for their role in this development. In addition, many new mass spectrometric analyzers specifically designed for protein analysis and identification have been developed, the most common of which is time-of-flight mass spectrometry. During the early part of this decade, 2000–2005, many proteomic facilities have been established in universities around the world, with the primary purpose of supporting research and development in proteomic applications. But what about the clinical applications? Most current clinical applications are based on prior investigations in biomedical or pharmaceutical research. Clinical laboratories, however, require a higher level of performance and validation, thereby slowing down the introduction of clinical proteomic applications. The area of greatest promise in the clinical laboratory or ‘space’, as it is sometimes referred to, is in the area of protein fingerprinting [2]. The analysis of proteins is quite complex and many proteins have yet to be identified. However, a qualitative MS analysis, which shows a particular pattern of peaks, tends to remain consistent in specific disease states or conditions, not dissimilar to the patterns of whorls in a thumbprint. The challenge that remains is the validation of this type of assay and its reproducibility across many laboratories. In truth, mass spectrometry analysis of a protein in a biologic matrix is still somewhat complex, and more novel methods are needed [3].

By all accounts, there are few routine clinical laboratory applications that perform protein analysis by mass spectrometry. The reason for this deficiency resides in the complexity of the patient sample being collected. The mass spectrometer actually performs its job well by identifying the proteins presented to it, best illustrated by a review of the few applications that have been described in the clinical literature, such as hemoglobin analysis. Very little sample preparation is required due to the abundance of this protein in blood, plasma and tissue. With respect to concentration, hemoglobin is present in abundance, whereas many proteins are only present in trace quantities. Mass spectrometry is actually a relatively insensitive technique compared with immunoassays, although its selectivity is quite high. Most technology that utilizes mass spectrometry has shifted to the sample preparation side of the analysis and actually utilizes antibodies and other means to isolate the protein of interest. The analysis by MS of this isolate thereby becomes much more accurate. Quantitatively, however, mass spectrometry still faces challenges in protein analysis, as do other methods. The key to the success of proteomics in the clinical laboratory may ultimately reside in the ability to extract the compounds of interest in a relatively pure state (or less complex matrix) and then identify them in the form of a profile or fingerprint. For quantitative work, new approaches are needed; alternatively, old standbys such as hydrolyzing the protein using enzymes such as trypsin may permit more accurate quantification of the peptides produced.

However, there is an alternative to protein analysis, and that lies in the remaining ‘omics’, metabolomics. Metabolomics is actually a measure of protein expression or function. For enzymes, the expression of the protein activity is performed by quantitatively accessing either substrate or product, or both. Those products or substrates interact with other enzymes to produce alterations in other metabolites. These interactions are extremely useful in the identification of disease states, and may ultimately be even more important, since it is the excess or deficiency of a metabolite that produces toxicity and illness.

Fortunately, metabolites are more easily measured by current mass spectrometry techniques such as liquid chromotography mass spectrometry and are generally more quantitative. It is thus easier to compare mass spectrometry techniques with established clinical laboratory tests, allowing them to be more readily validated. We are beginning to observe this process in areas such as newborn screening and metabolism, and the expansion of this effort should be striking in the next several years [4]. It is the authors’ view that metabolomics will play an important early role in mass spectrometry applications in clinical chemistry [5]. As new techniques of isolating proteins improve, we will then see more application of mass spectrometry in peptide messenger/hormone analysis, followed eventually by more robust protein analysis.