Development of multiplexed techniques using two-dimensional HPLC, protein microarrays and mass spectrometry for investigations in protein posttranslational modifications and disease progression pathways

Over the last few years cancer has replaced heart disease as the major killer in the US where the incidence of cancer is rising steadily. In order to understand the underlying biology of cancer and the role posttranslational modifications play in its development, several studies were performed to characterize human cancer cell lines and tissues. High throughput and multiplexed methods using reversed phase protein microarrays employing fractionated proteins from whole cell and tissue lysates were used to study the autoantibody response in cancer cells, the role of phosphorylation cascades in cancer development and mining disease response pathways in cancer. In these experiments, the cell and tissue lysates were fractionated using a 2D-HPLC based method where proteins were first separated according to their pI and then using non-porous reversed phase HPLC. Monolithic capillary HPLC methods were developed to increase the reliability of mass spectrometry based protein identifications employing peptide mass fingerprinting (PMF) and for high-throughput proteomics. In the first study, protein microarrays were used to study the global changes in phosphorylation following the inhibition of FGFR2 gene/protein in SUM-52PE human breast cancer cell lines. A small molecule universal phospho-sensor dye was used to detect phosphorylations on the microarrays. Peptide mass fingerprinting was used to identify the phospho-proteins and LC-MS/MS was used to validate the protein IDs. MAPK interacting protein, STAT3 and Lamin A/C for instance were detected to be over-expressed while the expression levels of Eps15 and RalBP1 were observed to be supressed following inhibition of FGFR2. Further, human pancreatic cancer Panc-1 cell-lines and metastatic and localized prostate cancer tissue proteins were fractionated respectively and used as baits for obtaining auto-antibody response data with an aim of mining biomarkers for improved classification of disease including predicting disease progression stage. This study also used molecular concept modeling to obtain information pathways that are perturbed in the progression of cancer. The last two studies use monolithic-HPLC for increasing sequence coverage for peptide mass fingerprinting and ESI-MS/MS. Sequence coverage exceeding 85% could be obtained in 2D liquid fractionated human breast cancer CA1a.c11 and CA1d.c11 cell line proteins. An automated platform was developed for spotting of MALDI plates and used for on-plate digestion and proteins separated using monolithic capillary HPLC for PMF. These methods can often detect low abundance proteins and posttranslational modifications.