RESUMEN
We made a comprehensive study of protein expression of androgen-independent (AI) prostate cancer by means of agarose two-dimensional gel electrophoresis (agarose 2-DE) followed by liquid chromatography-tandem mass spectrometric analysis. The agarose 2-DE method being good at separating high molecular mass (HMM) proteins and alkaline ones, so we were able to successfully reveal differences between the proteomes of androgen-dependent (AD) tumors and those of AI tumors. During the creation of agarose 2-DE protein maps, we successfully identified 295 proteins (91.0%) out of 324 spots excised in total. Excluding redundant and mouse serum proteins, we considered the remaining 225 proteins to be related to the cancer. We divided the 225 cancer-related proteins into HMM and low molecular mass (LMM) groups by their molecular mass being above or below 80 kDa. Functional classification of the proteins in these two groups showed clear differences between the two: more than half (54.8%) of the HMM proteins, but less than one-third (29.1%) of the LMM ones were classified among transcription- or translation-related proteins. Eighteen proteins were regulated when the tumor progressed from an AD to an AI state. Five of these proteins, including antioxidant protein 2, superoxide dismutase 1, thioredoxin peroxidase, GTP-binding protein beta chain homolog, and the ha1225 gene product, had a function to protect cells against oxidant stress-induced apoptosis. We suggest that prevention against oxidative stress is one of the key points for prostate cancer to obtain androgen independency. We also found proteins that had not been previously reported in prostate cancer by use of genomic or conventional 2-DE based proteomic approaches. The proteomic approach, using agarose 2-DE focused on HMM proteins, has the capability to find novel biomarkers of prostate cancer.