Multidimensional analyses of the proteome using nonporous reversed-phase HPLC coupled with MALDI-TOF and ESI mass spectrometry.

摘要

This thesis work consists of four major parts: solubilization of large proteins, profiling of large proteins from whole cell lysates, the use of nonporous (NP) RP-HPLC columns for protein separation, and the development of multidimensional nonporous LC-MS methods. In order to gain a better understanding of the biological system, new techniques must be developed to complement 1-D and 2-D gel electrophoresis methods. The use of a nonionic detergent (either Triton X-100 or n-octyl glucopyranoside) in the lysis buffer, along with guanidine-hydrochloride, has significantly improved the solubilization of large proteins (50 kDa). Using these procedures, proteins over 100 kDa have been profiled from E. coli extracts with MALDI-TOF MS. NP-RP-HPLC has been used to separate proteins up to 90 kDa from human breast cancer and Siberian Permafrost bacterial whole cell lysates with sufficient resolution for subsequent MS or enzymatic digest analysis. The C18 coated 1.5 μm diameter spherical NP packing materials provide a uniform surface that prevents proteins from sticking inside the pores. This means improved protein recovery (up to 80%) and separation time (15–30 min). Oncoproteins implicated in human breast cancer, including C-src, P53, and Hsp 27, have been identified from various MCF10 cell lines. In addition, Csp C, a cold shock protein from Permafrost 7-3 strain, has also been isolated. The 1-D image generated from the on-line NP-RP-HPLC-ESI-MS method, therefore, provides a direct means of comparing the highly expressed proteins in different samples or at different stages of tumorigenic progression. The development of chromatofocusing NP-RP-HPLC-ESI-MS method, as an alternative, allows for rapid separation of proteins based upon pI and hydrophobicity in the first and second chromatographic dimensions. Molecular weight (MW) is determined on-line in the third dimension using an ESI-MS. The resulting 2-D protein image is analogous to a 2-D gel image, pI vs. MW along with abundance information. This technique can be fully automated as well. Consequently, it is a practical alternative to the more time consuming and labor-intensive 2-D gel methods in the study of proteomics.