|Paul Russo, George Mason University, Manassas , VA|
Mark M. Ross, George Mason University, Manassas , VA
Weidong Zhou, George Mason University, Manassas , VA
Francesco Meani, George Mason University, Manassas , VA
Lance Liotta, George Mason University, Manassas , VA
Emanuel Petricoin, George Mason University, Manassas , VA
Mary Lopez, Thermo BRIMS, Cambridge, MA
Taha Rezai, Thermo BRIMS, Cambridge, MA
Bin Wei, National Cancer Institute, Bethesda, MD
Jennifer Edwards, National Cancer Institute, Bethesda, MD
Shelley Hoover, National Cancer Institute, Bethesda, MD
Paul Goldsmith, National Cancer Institute, Bethesda, MD
Gregory Alvord, National Cancer Institute, Bethesda, MD
Octavio Quinones, National Cancer Institute, Bethesda, MD
Elizabeth Spehalski, National Cancer Institute, Bethesda, MD
Mark Simpson, National Cancer Institute, Bethesda, MD
|The objective of this study is to discover, verify and compare assay methods for a novel serum protein ovarian cancer biomarker. A xenograft model of human serous SKOV3 ovarian cancer cells was established in nude mice to enable study of differential serum protein expression as a function of peritoneal tumor burden. Differential expression of a candidate biomarker protein, S100A6 (SA6), was verified in a larger number and variety of control and cancer samples by immunohistochemical and Western blot analyses. Several methods, including ELISA, reverse-phase protein microarray (RPMA) and multiple reaction monitoring mass spectrometry (MRM MS), were compared for assay of this new cancer biomarker.|
Samples consisted of serum and tissue from control and cancer xenograft mice, as well as human cancer cells and tissue. Biomarker discovery was achieved by analysis of proteins from control and cancerous mouse serum samples using gel fractionation to isolate low molecular weight (LMW) proteins followed by in-gel digestion and LC-MS/MS analysis (Thermo LTQ-Orbitrap). For candidate biomarker verification, custom rabbit polyclonal antibodies (pAbs) were developed against two different human SA6 peptide sequences and used in Western assays and an electrochemiluminescent immunosorbant assay (ECLISA). One of the pAbs was used also in RPMA assays. The LC-MS MRM analyses were performed using a Thermo Quantum triple quadrupole mass spectrometer.
A human-specific SA6 peptide was identified only in cancer samples. Ovarian cancer-specific SA6 expression was verified by immunohistochemical analysis of mouse cancer xenograft tissue. No SA6 expression was detected in normal human ovary and non-neoplastic tissue adjacent to tumors.
Three SA6 assay methods were evaluated for analytical performance, including sensitivity, specificity and reproducibility. First efforts were to develop and optimize a custom-designed ECLISA platform that was demonstrated to have a ~10ng/mL (~1fmol/uL)detection limit for SA6 in serum. Subsequent work was focused on developing and testing a RPMA-based method using the pAb used in the ECLISA method. Third, we are developing and optimizing an LC-MS MRM method using synthetic SA6 tryptic peptides, and a recombinant human SA6 protein. Initial MRM experiments indicate a detection limit of ~100amol/uL for the neat SA6 sample. The MRM method provides a unique capability to simultaneously monitor many analytes, such as peptides from unchanging endogenous serum proteins, spiked isotopically-labeled synthetic SA6 peptides for true quantitation, and potentially confounding analytes such as peptides originating from unique mouse versions of human proteins.
Initial studies indicate that mean SA6 concentration is directly proportional to tumor burden predictions.