Parallel Session 12: High-throughput Proteomics

Technical advances are still needed at a number of points in strategies for comparative or functional proteomics. In order to simplify mixtures of proteins from human cancer cells, we will present an improved method to isolate membrane proteins and discuss the separation of cytosolic proteins by heparin ion exchange/affinity chromatography. We will review the proteolytic labeling method introduced by this laboratory to provide global labeling of proteolytic peptide products from protein mixtures [Yao et al. (2001) Anal. Chem. 73, 2836 and Reynolds et al. (2002) J. Proteome Res. 1, 27], and demonstrate this approach in a comparative study of protein changes in acquired drug resistance [Hathout et al. (2002) J. Proteome Res. 1, electronically posted].

With a diversity of potential molecular markers it is of clear interest to develop a proteomic approach to the analysis of cancer tissue samples with the ultimate goal of analysis of protein expression patterns in individual patients. One approach to proteomic studies is termed shotgun sequencing in which the components of a proteomic sample are digested with a suitable protease and the resulting peptides are then resolved by chromatography and detected by mass spectrometry (LC/MS/MS). A popular alternative approach is to use 2D gels to achieve much higher resolution of such complex mixtures, but the lower recovery from gels can preclude the analysis of very small samples. This presentation will demonstrate that the LC/MS/MS approach can be used to identify a substantial number of potential tumor markers in a small number of cells from the breast cancer cell line, SKBR-3, isolated by laser capture microdissection. The total number of cells used in this study (e.g., 10,000) is compatible with relatively non-invasive tumor biopsy performed in individual patients and thus opens the possibility of profiling cancer markers in patients during disease progression or treatment. This study also has the potential of correlating mRNA expression profiling with discovery of the corresponding protein products.

The current environment for the discovery and development of medicines of value offers both unique opportunities and unique challenges. Recent advances in genomic and proteomic technologies have provided an array of approaches for the discovery and development of new medicines. Extracting the maximum value of these different technologies requires that they be utilized in an integrated and synergistic manner. To that end, the Genomic and Proteomic Sciences Division was created within Genetics Research at GSK, comprised of seven departments providing platform genomic and proteomic technologies in support of target identification and validation, drug development, and biomarker discovery.

This presentation will provide an overview of our approach to proteom- ics, which emphasizes high throughput, high information content, and information integration. Approaches to the isolation of protein complexes and to protein separation/quantitation will be discussed. Emphasis will be placed on the role of liquid chromatography and mass spectrometry in proteomics. Applications to be discussed include the analysis of 1,898 unique proteins from a nuclear fraction of a human cell line from a single LC injection, and the improvements in proteome coverage obtained by the integrated application of MALDI/MS/MS and ESI/MS/MS as detectors for LC.

The need for using antibodies in discovery of novel proteins, their functions and possible association with diseases, and for high throughput screening of protein levels is immense and so there is a large market potential for the generation of proteomic antibodies. For the production of antibodies against novel proteins based on current bioinformation of human genome, the cost for preparation of native proteins, fusion proteins or DNA vaccines from the human genome as immunogens would be very high and requires a lot of time and effort. Further, production of antibodies against those highly conserved proteins and peptide sequences by mammalian species is normally difficult because these antigens are not recognized as “foreign” by animals. Here we report an alternative method for high-throughput production of proteomic antibodies by using multiple antigenic peptides (MAPs) as immunogens and laying hens as hosts. Antigenic peptides were selected from the known proteins or polypeptides deduced from the genes or gene fragment DNA and MAPs were synthesized by an automatic peptide synthesizer. Laying hens were immunized with MAPs and immune eggs were collected. IgY antibodies were purified from eggs and were used to characterize and quantify proteins. Results from producing antibodies against four different serum proteins and one novel cancer-related protein are reported here. The data suggest that this approach could be an ideal procedure for high-throughput production of antibodies against thousands of novel human proteins.

Two novel dual-vector bacterial expression systems, designed for high- throughput generation of antibodies specific for cDNA-encoded proteins, are presented. The vectors allow for two approaches to a concept that involves parallel expression of cDNA-encoded proteins, in two vector systems as fusion with different tags, both enabling single-step affinity purification. Both approaches utilize fusion tags that include a portion with documented immunopotentiating effect to stimulate antibody production, and generated fusion proteins are used to elicit antibodies. The second fusion protein, expressed using one of the two novel vectors, is used in an immobilized form as an affinity ligand to enrich, from the generated antisera, antibodies with selective reactivity to the cDNA-encoded part. One of the two vectors is taking advantage of a novel Staphylococcus aureus protein A (SPA)-binding affinity tag, Z_SPA-1, enabling straightforward affinity purification and blotting procedures of expressed gene products on readily available reagents and chromatography media. Gene products fused to Z_SPA-1 were recovered on Protein A-Sepharose under both native and semi-denaturing conditions (0.5 M guanidine hydrochloride). In addition this vector has a hexahistidyl (His₆) tag that in fact is present in all vectors in this system, allowing recovery of fusion proteins under highly denaturing conditions. To evaluate the different vector systems, five clones from a mouse testis cDNA library were expressed and purified and it was found; (i) that Protein A-based purification was more stringent than IMAC, (ii) generated antibodies showed high specificity fore three of the clones in blotting procedures on different cell types and tissue homogenates. We thus conclude that the presented dual-vector method offers a highly stringent strategy for generation of monospecific polyclonal antibodies.