Many cancer cells have an unusual ability to grow in hypoxia, but the origins of this metabolic phenotype remain unclear. We compared the metabolic phenotypes of three common prostate cancer cell models (LNCaP, DU145, PC3), assessing energy metabolism, metabolic gene expression, and the response to various culture contexts (in vitro and xenografts). LNCaP cells had a more oxidative phenotype than PC3 and DU145 cells based upon respiration, lactate production, [ATP], metabolic gene expression, and sensitivity of these parameters to hypoxia. PC3 and DU145 cells possessed similar Complex II and mtDNA levels, but lower Complex III and IV activities, and were unresponsive to dinitrophenol or dichloroacetate, suggesting that their glycolytic phenotype is due to mitochondrial dysfunction rather than regulation. High passage under normoxia converted LNCaP from oxidative to glycolytic cells (based on respiration and lactate production), and altered metabolic gene expression. Though LNCaP-derived cells differed from the parental line in mitochondrial enzyme activities, none differed in mitochondrial content (assessed as cardiolipin levels). When LNCaP-derived cells were grown as xenografts in immunodeficient mice, there were elements of a hypoxic response (e.g., elevated VEGF mRNA) but line-specific changes in expression of select glycolytic, mitochondrial and fatty acid metabolic genes. Low oxygen in vitro did not influence the mRNA levels of SREBP axis, nor did it significantly alter triglyceride production in any of the cell lines suggesting that the pathway of de novo fatty acid synthesis is not directly upregulated by hypoxic conditions. Collectively, these studies demonstrate important differences in the metabolism of these prostate cancer models. Such metabolic differences would have important ramifications for therapeutic strategies involving metabolic targets.

BACKGROUND: In 2016, an estimated 22,280 new cases of ovarian cancer will be diagnosed with more than 14,000 women succumbing to the malignancy. These figures have not changed in the last decade. During this time new ovarian cancer models have been refined that involve either the fallopian tube or ovarian surface epithelial cells as the origin for tumorigenesis. With refined proteomics methodology, identification of new secreted biomarkers may be revealed that are specific to a primary fallopian tube or ovarian surface epithelium origin and may ultimately aid in the diagnosis, prognosis or therapy of ovarian cancer patients. EXPERIMENTAL PROCEDURES: The secretomes of eight cell lines, represented by SV40 immortalized fallopian tube (FT-194, FT-190), SV40 immortalized ovarian surface epithelium (1816-575), low-grade serous adenocarcinoma (HEY), high-grade serous adenocarcinoma (TOV-1946), clear cell adenocarcinoma (TOV21G), cisplatin-sensitive high-grade serous adenocarcinoma (A2780-S) and cisplatin-resistant high-grade serous adenocarcinoma (A2780-CP), were submitted for isobaric tagging for relative and absolute quantification (iTRAQ®) differential expression analysis. Briefly, each of the eight secretomes were trypsin-digested and labeled with one of eight iTRAQ® 8-plex reagents and subsequently submitted for either two-dimensional capillary liquid chromatography direct data dependent peptide tandem mass spectrometry on an Orbitrap Velos system or one-dimensional mass spectrometry analysis. RESULTS: 456 proteins were identified with the one-dimensional analysis with 2 distinct peptides. 2069 distinct differentially expressed proteins were identified in the two dimensional analysis. Initial relative differential expression rate comparisons of control cell lines, FT-194, FT-190 and 1816-575 versus ovarian cancer subtype and therapy-associated cell lines, HEY, TOV1946, TOV21G, A2780S and A2780CP, are promising. Four proteins were increased greater than three-fold across all ovarian cancer cell lines compared to normal fallopian tube and ovarian surface epithelial control cell lines: stromelysin-1, pigment epithelium-derived factor, matrix metalloproteinase-9 and receptor expression-enhancing protein 6. Large scale inferential statistical methodology will be used to differentiate cell lines. Citation Format: Yaling Zhou PhD, Daniel Brooks, PhD, Sumana Dey PhD, LeeAnn Higgins, PhD, Todd Markowski, BSc, Chad Hamilton MD, Robin Howard, MA, Sorana Raiciulescu MSc, Amy Skubitz PhD, John D. Andersen DO. DIFFERENTIAL PROTEIN EXPRESSION ANALYSIS OF SV40–IMMORTALIZED OVARIAN SURFACE, SV40–IMMORTALIZED FALLOPIAN TUBE AND OVARIAN CANCER SUBTYPE CELL LINE SECRETOMES BY ITRAQ® [abstract]. In: Proceedings of the 11th Biennial Ovarian Cancer Research Symposium; Sep 12-13, 2016; Seattle, WA. Philadelphia (PA): AACR; Clin Cancer Res 2017;23(11 Suppl):Abstract nr DPOC-003.