Gabi U. Dachs

Impact of tumor necrosis factor‐alpha and lysophosphatidic acid on the behavior of ovarian cancer cells in a three‐dimensional collagen hydrogel

Abstract Background Advanced ovarian cancer involves tumor cells proliferating within ascitic fluid, which is rich in chemokines, cytokines, growth factors, and phospholipids essential for tumor progression. Tumor necrosis factor‐alpha (TNF‐α) and lysophosphatidic acid (LPA) are critical components that modulate ovarian tumor cell behavior in vivo, though most in vitro studies have relied on cell monolayers that do not accurately represent the tumor microenvironment. We thus investigated TNF‐α and LPA effects on ovarian tumor cells cultured in collagen scaffolds, which are a model more reflective of in vivo conditions. Methods Ovarian cancer cell lines SKOV‐3, OVCAR‐8, OVCAR‐5, and OVCAR‐4 were cultured in collagen I hydrogels at various concentrations of TNF‐α and LPA for 2–8 days. Outcomes included cell morphology, collagen topography, secretion of vascular endothelial growth factor (VEGF), interleukin‐8 (IL‐8), interleukin‐6 (IL‐6), and cell survival endpoints, with comparisons to collagen topography observed in ovarian tumor tissue. Results Collagen I fiber topography in 3D hydrogels remodeled over time, resembling the ovarian cancer tissue. LPA significantly reduced hydrogel size in SKOV‐3 cells. While LPA modestly altered VEGF secretion, both LPA and TNF‐α increased IL‐8 secretion over time, with TNF‐α also elevating IL‐6 levels. TNF‐α reduced cell survival in OVCAR‐8 and OVCAR‐5, while LPA promoted cell growth in SKOV‐3, OVCAR‐8, and OVCAR‐5. Conclusions TNF‐α and LPA have diverse, cell line‐specific effects on ovarian cancer cells in collagen hydrogels, underscoring cell line heterogeneity and the utility of 3D models.

Targeting Phosphoinositide 3-Kinase to Reduce the Progression of Ovarian Cancer Cells in a 3D Collagen Model

Ovarian cancer remains a major cause of mortality in women aged 74 years and under. Dysregulation of the PI3K/AKT/mTOR and NFκB signaling pathways has been associated with poor outcomes and treatment resistance. This study evaluated three potential anticancer agents targeting these pathways: buparlisib (a pan-PI3K/mTORC1 inhibitor), SN32976 (a PI3K p110α inhibitor), and pterostilbene (a resveratrol analogue that downregulates PI3K/AKT and NFκB signaling). Their efficacy was tested in 3D collagen models of ovarian cancer, using SKOV3 and OVCAR8 cell lines, activated by tumor necrosis factor-alpha (TNFα) and lysophosphatidic acid (LPA). Using concentrations derived from 2D assays, viability, collagen gel sizes, secretion of interleukin 6/8 (IL-6/8) and signal pathway proteins were analyzed. All compounds were less effective in 3D models than in 2D cultures, with high cell viability maintained. TNFα and LPA did not significantly alter drug sensitivity, and collagen gel contraction was largely unaffected. While the compounds did not consistently change signaling protein levels, they generally reduced secretion of pro-inflammatory cytokines IL-6 and IL-8. Growth in 3D collagen gels conferred drug resistance on OVCAR8 but not SKOV3 models. Overall, these findings provide preclinical support for further investigation of SN32976 and pterostilbene in ovarian cancer models.