Introduction

Biomarkers are indicators of specific biological states. Cancer biomarkers can aid in diagnosis and/or patient management by refining staging and predicting or monitoring response to treatment (1,2). Proteomics is particularly promising for the analysis of biological fluids and biomarker identification. The current interest in proteomics is due in part to its power to interrogate the end player encoded for in the genome, namely the proteins. Here, we discuss innovative proteomic approaches that have been successful for biomarker discovery.

Challenges of Plasma-Based Biomarker Discovery

Plasma is among the most accessible biological materials available. However, the plasma proteome is particularly challenging because of its complexity and vast dynamic range. Plasma contains several thousand proteins with concentrations ranging from as high as 20-50 mg/mL for serum albumin to femtomolar concentrations for some known biomarkers (3). Plasma consists of vast assemblies of proteins and complexes that reflect the physiologic or pathologic state of cells, tissues, and organs, with many of the plasma proteins found in multiple forms (4). Some forms result from alternative splicing at the RNA level, and others result from cleavages and posttranslational modifications (5). This plethora of protein forms, combined with the vast dynamic range of concentrations, presents the greatest challenge to plasma biomarker discovery. While mass spectrometry (MS) has evolved sufficiently to detect and identify femtomoles of peptides, the dynamic range of detection is still a limiting factor for comprehensive serum profiling (6,7).

In order to overcome the challenge presented by the wide range of concentrations of plasma proteins, three basic strategies have been developed: (i) removal of high abundance proteins, such as albumin and immunoglobulins that interfere with the detection of less abundant proteins (8); (ii) fractionation of samples, by chromatographic or other means of separation, to reduce complexity prior to MS (4,7,9); and (iii) isolation of targeted groups of proteins or peptides of interest (10,11,12). Using a combination of depletion of abundant proteins and extensive fractionation, together with isotopic labeling (13), our group was able to identify several cancer biomarkers in plasma other than nonspecific inflammatory and acute phase protein markers (Figure 1).

Figure 1. Proteomic platform for identification of cancer biomarker in plasma. (Click to enlarge)

Extensive fractionation of intact proteins permits a more detailed investigation of the particular protein forms identified and their posttranslational modifications that may be associated with disease; a feature that is not readily achievable with protein digestion-based fractionation approaches. Figure 2 illustrates this feature, with the identification of a C-terminal fragment of insulin-like growth factor binding protein 2 (IGFBP2) separate from the intact protein. Several proteolytic enzymes have been identified that cleave IGFBPs. It has been suggested that proteolytic cleavage of IGFBPs is associated with regulation of the proliferative effects of IGFs on their target cells (14).

Figure 2. Identification of protein fragments of insulin-like growth factor binding protein 2 (IGFBP2). (Click to enlarge)

Cell-Based Proteomics

Tumor cells and cancer cell lines are also valuable sources of potential cancer biomarkers. A multitude of approaches for cell protein analysis have yielded quantitative information. Here we address strategies that fall under the heading of functional proteomics.

Stable isotope labeling of amino acids in cell culture (SILAC) has become a popular approach to study the proteomes of various cell types and microorganisms and how they change in response to various conditions (15). In a nutshell, cells are grown in culture in the presence of one amino acid, such as lysine, for which all the 12C were substituted by 13C. Incorporation of this isotopically labeled amino acid occurs in the process of cell growth, protein synthesis, and turnover. SILAC allows light and heavy proteomes to be distinguished by MS.

This approach has been used to study cancer cell spread (16) and cancer cell secretion (17), as well as to identify therapeutic targets (18).

More focused approaches within the category of functional proteomics aim to facilitate the analysis of complex proteomes by selecting specific protein functions of interest. This strategy, referred to as activity-based protein profiling, uses active site-directed probes to interrogate, for example, the functional state of enzyme families (19,20,21). The delineation of enzyme activities selectively associated with tumor cells or tissues has the potential to yield a rich source of biomarkers and targets for cancer diagnosis and therapy. Quantitative profiling of the activity of the SH family of enzymes demonstrated that the membrane and secreted proteomes are especially enriched in enzyme activities that serve as markers of cellular phenotype (22).