Abstract
Background: Acidosis, due to poor perfusion coupled with increased glycolysis, alters the immediate tumor microenvironment; tumor cells that adapt to lower pH have a growth advantage and become more invasive leading to metastasis. The phenotypic changes in these cells also lead to differences in protein expression, which can be measured with quantitative proteomics. Identification of the proteins that are modulated will advance the studies of the mechanisms of tumor survival and progression and suggest candidate biomarkers for patient assessment.
Methods: We assessed differential protein expression between breast cancer cells, DCIS, MCF-7, and MDA-MB-231, grown under low pH, 6.7, and the same cells grown at physiological pH, 7.4. After acclimation to low pH for 3 months until growth rates were similar to cells at physiological pH, cell lines were grown for 2 weeks with stable isotope labeling using amino acids in cell culture (SILAC) heavy medium (pH 6.7) or the corresponding light amino acids (pH 7.4). Equal numbers of cells were mixed prior to lysis; the proteins in the whole cell lysate were separated by SDS-PAGE, reduced, alkylated, digested in the gel, analyzed by LC-MS/MS peptide sequencing, and identified by database searching. The differential protein expression level was analyzed by MaxQuant, developed by Cox and Mann, and GeneGo software (Metacore) was used to analyze the pathways represented by the proteins detected by LC-MS/MS. Furthermore, quantitative assays, based on liquid chromatography coupled to multiple reaction monitoring (LC-MRM) are developed for assessment of protein expression in cell lines and tissue samples. LC-MRM quantifies proteins by measuring them against known concentrations of internal standards in a triple quadruple mass spectrometer. Unique peptides corresponding to each protein were chosen for monitoring to measure the expression levels of each protein.
Results: Extraction of data from LC-MS/MS enabled MaxQuant to calculate the ratio between heavy and light peptide ion signals, which describes the differential protein expression. Analysis of SILAC heavy and light pairs resulted in quantification of between 2,500 and 3,500 proteins in each experiment. Heavy to light ratios (H/L) were calculated and log transformed to calculate the standard deviation. Significant changes were determined to be those that were more than 2 standard deviations from the median. MaxQuant revealed that the expression of seven proteins/genes were enhanced at low pH in all three breast cancer cell lines: Agrin, Clusterin, EBP, ERMP1, G6PE, HBS1L, and LASS2. GeneGo associated these proteins with different biological processes such as immune response and the pentose phosphate pathway.
Conclusions: Proteomics analysis discovered seven over-expressed proteins involved in acidosis in breast cancer. These proteins might be potential biomarkers for characterizing acidosis and improve clinical outcome. Future in vivo experiments that test these biomarkers in xenograft tumors are necessary. ∗This presenting student is supported by an ARRA supplement to the NCI Cancer Center Support Grant awarded to Moffitt (3P30 CA076292-11S6 PI WS Dalton) to create a training program for underrepresented undergraduate students in clinical proteomics.
Citation Information: Cancer Epidemiol Biomarkers Prev 2010;19(10 Suppl):A65.