Decreased Peritoneal Ovarian Cancer Growth in Mice Lacking Expression of Lipid Phosphate Phosphohydrolase 1

March Decreased Peritoneal Ovarian Cancer Growth in Mice Lacking Expression of Lipid Phosphate Phosphohydrolase 1 John Nakayama 0 1 2 3 Timothy A. Raines 0 1 2 3 Kevin R. Lynch 0 1 2 3 Jill K. Slack-Davis 0 1 2 3 0 1 Department of Obstetrics and Gynecology, The Cancer Center, University of Virginia, Charlottesville, Virginia, United States of America, 2 Department of Microbiology, Immunology and Cancer Biology, The Cancer Center, University of Virginia, Charlottesville, Virginia, United States of America, 3 Department of Pharmacology, The Cancer Center, University of Virginia, Charlottesville, Virginia, United States of America, 4 The Cancer Center, University of Virginia , Charlottesville, Virginia , United States of America 1 Funding: This work was supported by the National Institutes of Health/National Cancer Institute grant CA142783 supported JKS-D and KRL, who was also supported by National Institutes of Health General Medical Sciences grant GM067958-09. TAR was supported by a diversity supplement to CA142783. JN was supported by the University of Virginia Gynecologic Oncology Fellowship Program. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript 2 Academic Editor: Tanya V. Kalin, Cincinnati Children's Hospital Medical Center , UNITED STATES 3 Current address: Case Western Reserve University, School of Medicine , Cleveland, Ohio , United States of America - Lysophosphatidic acid (LPA) is a bioactive lipid that enhances ovarian cancer cell proliferation, migration and invasion in vitro and stimulates peritoneal metastasis in vivo. LPA is generated through the action of autotaxin or phospholipases, and degradation begins with lipid phosphate phosphohydrolase (LPP)-dependent removal of the phosphate. While the effects of LPA on ovarian cancer progression are clear, the effects of LPA metabolism within the tumor microenvironment on peritoneal metastasis have not been reported. We examined the contribution of lipid phosphatase activity to ovarian cancer peritoneal metastasis using mice deficient in LPP1 expression. Homozygous deletion of LPP1 (LPP1 KO) results in elevated levels and decreased turnover of LPA in vivo. Within 2 weeks of intraperitoneal injection of syngeneic mouse ovarian cancer cells, we observed enhanced tumor seeding in the LPP1 KO mice compared to wild type. However, tumor growth plateaued in the LPP1 KO mice by 3 weeks while tumors continued to grow in wild type mice. The decreased tumor burden was accompanied by increased apoptosis and no change in proliferation or angiogenesis. Tumor growth was restored and apoptosis reversed with exogenous administration of LPA. Together, these observations demonstrate that the elevated levels of LPA per se in LPP1 KO mice do not inhibit tumor growth. Rather, the data support the notion that either elevated LPA concentration or altered LPA metabolism affects other growthpromoting contributions of the tumor microenvironment. Lysophsphatidic acid (LPA) is a bioactive lipid that regulates several cellular functions critical for tumorigenesis and metastasis including proliferation, survival, cytoskeletal reorganization, migration, invasion and cytokine production [15]. The importance of LPA to ovarian cancer Competing Interests: The authors have declared that no competing interests exist. progression was established when it was identified as a growth factor in malignant ascites [6]. LPA stimulates cellular activities via at least three (LPA1, LPA2, and LPA3) and perhaps as many as 6 (LPA46) G-protein coupled receptors. LPA1 is expressed on normal ovarian surface epithelium; the expression of LPA2 and LPA3 is induced in the cancer cells [2]. Upon binding its receptor, LPA stimulates ovarian cancer cell proliferation through activation of G12 [4]. All three receptors regulate ovarian cancer cell migration and invasion directly by activating pro-migratory Rac and Rho-dependent signaling pathways [1,7]. In addition, LPA promotes ovarian cancer growth and metastasis indirectly by stimulating the production of proteases (MMP and urokinase plasminogen activator) [8,9] and cytokines (IL-6 and IL-8), which play a role in ovarian cancer invasion and metastasis [10]. LPA binding to LPA2 or LPA3 increases production of IL-6, IL-8, and VEGF. Indeed, knockdown of LPA2 or LPA3 decreases IL-6 production, and their over-expression leads to increased serum levels of IL-6 and VEGF, increased tumor burden and shortened survival times in a mouse model of ovarian cancer peritoneal metastasis; modulation of LPA1 had no significant effect in this study [11]. LPA is produced by a variety of cells within the tumor microenvironment including platelets, mesothelial cells, adipocytes, endothelial cells and ovarian cancer cells [12,13], and in the absence of cancer, concentrations are tightly maintained below 1 M. Levels of LPA are significantly elevated in plasma and ascites (up to 50 M) of women with ovarian cancer [14], and increased plasma LPA has been suggested as a biomarker for ovarian cancer [15]. Members of the phospholipase A1 (PLA1) and PLA2 families remove a fatty acid chain from phosphatidic acid to form LPA. Autotaxin (ATX), an extracellular lysophospholipase D also generates LPA following the removal of choline from lysophosphatidylcholine. PLA2 and autotaxin are elevated in ovarian cancer patients [16,17], and a positive feedback loop exists between vascular endothelial growth factor (VEGF) and ATX production by ovarian cancer cells [18]. LPA catabolism is initiated by lipid phosphate phosphohydrolases (LPPs), types 1, 2 and 3, which remove the phosphate to generate monoacylglycerol (MAG). MAG is further cleaved by monoacylglycerol lipase to release the fatty acid chain from glycerol. LPP1 and LPP3 expression is reduced in human ovarian cancers relative to normal ovarian tissue [19], while forced over-expression of either LPP1 or LPP3 decreases tumorigenesis of ovarian cancer cells in mouse models presumably from decreased levels of LPA [19,20]. In addition to regulating ovarian cancer cell proliferation, survival, migration and invasion in vitro, the ability of LPA to promote ovarian cancer invasion and growth has been demonstrated in mouse models of peritoneal metastasis; daily injection or implantation of pumps producing high concentrations of LPA increased tumor burden in immune compromised and syngeneic mouse models [21,22]. However, the effects of LPA metabolism within the tumor microenvironment on peritoneal metastasis have not been reported. We sought to determine whether impaired LPA phosphatase (specifically LPP1) activity affected ovarian cancer peritoneal metastasis. Rather than target LPP1 activity in the tumor cells, we examined the effect of LPP1 loss in the tumor microenvironment using mice lacking LPP1 expression following the insertion of an exon-trap (LPP1 KO) [23] and syngeneic mouse ovarian cancer cells [24]. Lipid phosphatase ac (...truncated)