Role of malignant ascites on human mesothelial cells and their gene expression profiles

Background: Malignant ascites is often present at diagnostic in women with advanced ovarian cancer (OC) and its presence is associated with a worse outcome. Human peritoneal mesothelial cells (HPMCs) are key components of malignant ascites. Although the interplay between HPMCs and OC cells is believed to be critical for tumor progression, it has not been well characterized. The purpose of this study was to assess the effect of ascites on HPMCs and clarify the role of HPMCs in OC progression.

Human OC ascites and benign peritoneal fluids were assessed for their ability to stimulate HPMC proliferation. Conditioned medium from ascites- and benign fluid-stimulated HPMCs were compared for their ability to attenuate apoptosis induced by TNF-related apoptosis-inducing ligand (TRAIL). We conducted a comparative analysis of global expression changes in ascites-stimula ted HPMCs using Agilent oligonucleotide microarrays.

As compared to benign peritoneal fluids, malignant ascites stimulated the proliferation of HPMCs. TRAIL-induced apoptosis was attenuated in OC cells exposed to conditioned medium from ascites-stimulated HPMCs as compared to OC cells exposed to conditioned medium from benign fluid-stimulated HPMCs. A total of 649 genes were differentially expressed in ascites-s timulated HPMCs. Based on a ratio of more than 1.5-fold and a P < 0.05, 484 genes were up-regulated and 165 gene s were down-regulated in ascites-exposed HPMCs. Stimulation of HPMCs with OC ascites resulted in diffe rential expression of genes mainly associated with the regulation of cell growth and prolife ration, cell death, cell cycle and cell assembly and organization, compared to benign peritoneal fluids. Top networks up-regulated by OC ascites included Akt and NF- κB survival pathways whereas vascular endothelial growth fact or (VEGF) pathway was down-regulated.

The results of this study not only provide evide nce supporting the importance of the interplay between cancer cells and HPMCs but also define the role that the tumor environment plays in these interactions.

Epithelial ovarian cancer (EOC) is the leading cause of death among gynecological cancers. High-grade serous ovarian carcinomas (HGSOC) are by far the most com- mon (85-90%) subtype and the majority of patients with HGSOC presents with ascites and advanced disease with peritoneal dissemination [1,2]. After initial treatment, the majority of these patients will relapse and eventually die. The mean survival of patients that have advanced disease at presentation is 39 months. This high mortality is mainly attributed to widespread metastasis throughout the peritoneal cavity and the emergence of drug resistance during the course of treatment [3]. OC mortality has not significantly decreased during the last 30 years for reasons including poor understanding of the tumor biology and the interactions with the surrounding environment. Primary tumor growth induces host responses that are believed to support and promote tumor progression. OC mainly spreads by direct extension, through seeding or exfoliation of tumor cells from ovarian/fallopian tubes into ascites, in which tumor cells survive and proliferate, and later implant in the peritoneal cavity. Indeed, the presence of ascites correlates with intraperitoneal tumor spread and a worse prognosis. In this context, ascites that accumulates during OC progression represent a * Correspondence: [email protected] 1Département de Microbiologie et Infectiologie, Faculté de Médecine, Université de Sherbrooke, 3001, 12ième Avenue Nord, Sherbrooke, Québec, J1H 5N4, Canada Full list of author information is available at the end of the article © 2014 Matte et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. Matte et al. BMC Cancer 2014, 14:288 http://www.biomedcentral.com/1471-2407/14/288 particular tumor environment and a survival niche for tumor cells [3,4]. Ascites are complex and heteroge- neous fluids that contain a variety of cytokines, chemo- kines and growth factors as well as other soluble factors such as lysophosphatidic acid (LPA) [5,6]. OC tumorigen- esis is a complex process and a growing body of evidence suggests that although genetic events in the tumor cells themselves are crucial, host and stromal factors in ascites are also important. For example, OC ascites attenuate drug-induced apoptosis in tumor cells and thus provide a protective environment for tumor cells [4]. Soluble factors in ascites activate survival pathways in tumor cells such as Akt and ERK1\2 signaling, through engagement of cell surface receptors such as αvβ5 integrins which attenuate tumor necrosis factor-related apoptosis-inducing ligand (TRAIL)-induced apoptosis [7-9]. A number of studies have also demonstrated that ascites enhance tumor cell proliferation and migration [10,11]. The presence of LPA in ascites has been shown to promote tumor cell prolifera- tion and migration [12]. These data strongly suggest that malignant ascites plays a significant role in facilitating OC progression and metastasis. Human peritoneal mesothelial cells (HPMCs) form the peritoneal lining and serve as a protective anatomical barrier. They are among the most abundant cell type in ascites from patients with OC [13]. Although it is be- coming evident that paracrine factors secreted in the resulting tumor environment subsequently modify the behaviour of tumor cells, a dynamic interaction between HPMCs found in ascites and the surrounding environ- ment could alter their behaviour, which in turn, further affect malignant evolution and contribute to establish a milieu favouring tumor progression. A number of evi- dence suggests that morphological and functional changes of HPMCs occur in the presence of cancer cells due to the secretion of paracrine factors. For example, HPMCs increase in size, become more permeable, and undergo an epithelial to mesenchymal transition (EMT) in the pres- ence of TGF- β [14-17]. However, precisely how HPMCs are influenced by ascites is poorly understood. The aim of this study was to determine the effect of malignant ascites on HPMC behaviour and the paracrine effects of ascites-stimulated HPMCs. We also investi- gated molecular changes that occur in ascites-stimulated HPMCs. We present evidence that ascites impact on HPMCs by altering their behaviour and gene expression profiles.

Cell culture and clinical samples The three malignant ascites used in this study (OVC346, OVC508, OVC509) were obtained at the time of initial cytoreductive surgery from three ovarian cancer patients at the Centre hospitalier universitaire de Sherbrooke. Peritoneal fluids were obtained from three patients oper- ated for conditions other than cancer. This study has been performed in accordance with the Declaration of Helsinki and was approved by the «Comité d ’éthique de la recherche en santé chez l ’humain du centre hospitalier universitaire de Sherbrooke». Fluids were centrifuged at 1000 rpm for 15 min and the cell-free fractions were stored at -20°C until assayed. All fluids were supplied by the Banque de tissus et de données of the Réseau de Recherche en Cancer of the Fonds de la Recherche du Québec en Santé affiliated to the Canadian Tumor Repository Network (CTRNet). Histopathological diagnosis, grade, and stage of ovarian tumor samples were assigned according to the crit eria of the International Fed- eration of Gynecology and Obstetrics. The three malignant ascites were from patients with HGSOC (stage III/IV) and were chosen because they are representative HGSOC asci- tes with regards to their properties and cytokine profiles [5,7,8]. The ovarian cancer cell lines CaOV3 and SKOV3 were obtained from American Type Culture Collection, (Manassas, VA) and maintained in a humidified 5% CO 2 in- cubator at 37°C. Cells were passaged twice weekly. CaOV3 and SKOV3 cells were cultured in DMEM/F12 (Wisent) supplemented with 10% FBS, 2 mM glutamine and antibi- otics. HPMCs were isolated from peritoneal lavages of two women operated for conditio ns other than cancer. After centrifugation, the cell pe llet is placed on T25 culture plates. The medium is changed the next day and, in our ex- perience, adhered cells typically represent HPMCs. The na- ture of HPMCs was confirmed by immunostaining with antibodies against calreticulin (Life Technology) and epithelial marker MOC31 ( Oncogen Research Product, S a nD i e g o ,C A ) .H P M C sw e r eg r o w ni nD M E M / F 1 2 supplemented with 0.4 μg/ml of hydrocortisone and 10 ng/ml EGF (Sigma, Oakville, Canada), 10% FBS and antibiotics. The media was changed every 3 days while the cells were maintained at 37°C in a humidified 5% CO 2 incubator. HPMCs were used between passage 5-8. Immunofluorescence Cells were grown on glass slides, fixed in cold methanol and blocked in PBS/2% BSA at room temperature for 1 h. Anti-calreticulin and anti-MOC31 primary antibodies were diluted in PBS/BSA and slides were incubated at room temperature for 1 h. Slides were washed twice in cold PBS, incubated 1 h at room temperature either with FITC or Texas-Red conjugated antibodies and visualized with a Olympus IX70 fluorescence microscope (Olympus, Hamburg, Germany). In vitro proliferation assay HPMCs were seeded in medium either with 10% FBS, with 10% benign fluids or with 10% malignant ascites in six-well plates and incubated at 37°C. Cells were monitored for up Matte et al. BMC Cancer 2014, 14:288 Page 2 of 13 http://www.biomedcentral.com/1471-2407/14/288 to 48 h and representative wells were photographed. In some experience, hydroxyurea (30 mM) (Sigma) was added to inhibit cell proliferation. Two independent experiments were performed for each assay and representative photo- graphs were taken. Cell growth was also quantitatively determined using XTT assay as previously described [7]. RNA preparation and quantitative PCR validation HPMCs were incubated in medium with either 10% benign fluids or 10% malignant ascites for 4 h. Cells were washed with PBS and total RNA was extracted from HPMCs using TRIzol reagent (Life Technologies) according to the manufacturer ’s protocol and subjected to reverse transcription (RT) with oligodT from Promega (Madison, WI) and MMULV reverse transcriptase en- zyme. The quality and concentration of RNA was deter- mined by capillary electrophoresis using a Agilent 2100 Bioanalyzer (Agilent Technologies, Mississauga, Ontario, Canada). The integrity of the cDNA was assessed with the Taqman gene expression assays (Life Technologies), done on 18S housekeeping gene. Each sample was normal- ized to the housekeeping gene levels. For quantitative PCR validation, total RNA was extracted and cDNA was ob- tained as described above, The FAST T aqman gene expres- sion assay was used with 50 ng of cDNA. Conditions were as follow: initial cycle 50°C, 2 min, 95°C, 10 min. 40 cycles at 95°C, 15 s and 60°C, 1 min on a StepOnePlus TM Real- Time PCR system (Life Technologies). Data were analyzed using the StepOne TM software and comparative ΔΔ Ct measure was used to express the results as fold changes. Gene expression profiling and data analysis Microarray hybridization was performed using the Whole Human Genome Oligonucleot ide Microarray (Agilent), containing ~ 44,000 genes, at the Cancer Research Centre, Hôpital Hôtel-Dieu de Québec. Upon hybridization and washing, the arrays were scanned using a dual-laser DNA microarray scanner (Agilent). The data were extracted from images by the Feature Extraction software 6.1 (Agilent). The GeneSpring software (Agilent) was used to generate lists of selected genes for statis tical analysis. An intensity- dependent normalization (Lowess normalization) was ap- plied to correct for artifacts caused by non-linear rates of dye incorporation as well as inconsistencies of the relative fluorescence intensity between dyes. Consecutive lists of differentially expressed genes were generated considering a 1.5-fold expression as the gene selection criteria. The genes in the gene lists were classifi ed according to their function using the Gene Ontology (GO SLIMS) classification sys- tem. Network analysis of the microarray data was com- pleted using the Ingenuity Pathway Analysis software (http://www.Ingenuity.com). The microarray data have been deposited to the GEO dat abase (http://www.ncbi. nlm.nih.gov/geo/) with accession number GSE55065. Conditioned media and apoptosis assay To generate HPMC-conditioned media, HPMCs were seeded at 80% density in six-well plates and cultured in media containing either 10% FBS, 10% benign fluids or 10% malignant ascites overnight. Cells were washed twice and fresh medium without FBS or growth factors was added. HPMCs were cultured for 8 to 24 h. Medium conditioned by ascites-stimulated and benign fluids- stimulated HPMCs were applied at a ratio of 50% v/v to CaOV3 cells cultured at 70% density in 12-well plates. CaOV3 cell apoptosis in the presence of TRAIL (25 ng/ml) (PeproTech Inc, Rocky Hill, NJ) was measured using the Cell Death Detection ELISA kit (Roche, Laval, Québec, Canada) according to the manufacturer ’s instruction. CaOV3 cells were pre-treated for 1 h with HPMC-conditioned medium before the addition of TRAIL overnight. Three independent sets of experiments were performed for each type of condi- tioned medium. Determination of growth factor levels in ascites LPA levels in benign peritoneal fluids and malignant asci- tes were determined by ELISA using the Echelon Biosci- ences kit (Salt Lake City). TGF- β1 levels were determined using the RayBio® Human Cytokine Antibody Array G series 1000 from RayBiotech Inc. (Norcross, GA). With this method, TGF-β1 levels are expressed as relative fluor- escent units (FU) and can be used to compare levels in dif- ferent ascites. The signal intensities were quantified using the ScanArray Express dual-color confocal laser scanner (Perkin Elmer). Data were collected in Cy3 channel and stored as paired TiFF images. Spots were identified and local background substracted using the TIGR_Spotfinder 3.1.1 software. The internal negative controls were used to determine the cut-off intensity for a positive signal. Inten- sities up to 750 FU were considered negative.

Characterization of mesothelial cultures from the peritoneal lining We established HPMC cultures of peritoneal fluids from two women with benign conditions. The morphology of two primary HPMC samples (Meso-7 and Meso-9) cul- tured in presence of 10% FBS is shown in Figure 1A. These cells show spindle fibroblastic-like pattern consist- ent with a mesenchymal phenotype. The primary HPMC cultures of Meso-7 were further characterized using MOC31 epithelial marker and calretinin mesothelial marker [18]. As shown in Figure 1B, mesothelial cultures stained positive for calretinin and negative for MOC31 as expected, confirming the absence of epithelial ce lls in HPMCs. In contrast, the SKOV3 OC cell line stained positive for MOC31 and nega- tive for calretinin. Furthermore, as previously reported [14], HPMCs cultured in serum-free medium exhibited a polygonal, even cobblestone-like morphology (Figure 1C). Matte et al. BMC Cancer 2014, 14:288 Page 3 of 13 http://www.biomedcentral.com/1471-2407/14/288 In contrast, HPMCs cultured in 10% malignant ascites exhibited a more fibroblastic-like pattern. Because TGF- β1 has been previously associated with morphologic changes in HMPCs [14], we examined the levels of TGF- β1 from benign fluids and malignant asci- tes. Interestingly, the levels of TGF- β1 were significantly higher ( P < 0.001) in malignant ascites compared to benign fluids (Figure 1D). TGF- β1 levels were below the threshold for positivity (750 FU) in the two benign peri- toneal fluids tested. Malignant ascites stimulate the growth of HPMCs Malignant ascites constitute a dynamic reservoir of soluble factors, which individually and in a combined fashion may affect cell behavior. To assess the putative effect of malig- nant ascites on the growth of HPMC cultures, we se- lected two representative ascites (OVC346 and OVC508) obtained from women with newly diagnosed HGSOC. These malignant ascites have been previously described [5,7,8]. This study included only HGSOC ascites because they are the most clinically relevant as the majority of patients presenting with ov arian cancer (80-90%) have HGSOC. HPMCs were incubated with OVC346 and OVC508 cell-free ascites fra ctions and two peritoneal fluids from women with benign gynecological condi- tions. Compared to the peritoneal benign fluids, a growth-enhancing effect was observed with the two malignant ascites as shown by an increased in overall cell number after 12 h (Figure 2A). Both OVC346 and OVC508 malignant ascites had growth-enhancing activity compared to benign fluids. The growth-enhancing effect of malignant ascites was completely inhibited by the addition hydroxyurea, a cell cycle inhibitor. When com- pared to benign fluid OV401, a growth-enhancing activity SKOV3 Meso-7 MOC31Calretinin Meso-7 Meso-7Meso-9 A B CD Meso-7 Serum free FBS 10% Ascites 10% * * Relative TGF-β1 expression Malignant ascitesBenign fluids 0 200 400 600 800 1000 1200 1400 1600 1800 2000 OV370 OV410 OVC346 OVC508 Figure 1 Characterization of HPMCs. (A) Phase contrast pictures of HPMCs (Meso-7 and Meso-9) cultured in 10% FBS (x 100 magnification). Bars 200 μm. (B) Immunofluorescence detection of MOC31 and calretinin in human ovarian cancer cells SKOV3 and HPMCs. The cells were fixed with cold methanol and stained with FITC-conjugated anti-MOC31 and Texas Red-conjugated anti-calretinin (x 1000 magnification). HPMCs stained positive for calretinin and negative for MOC31 confirming that they were mesothelial cells. Bars 30 μm. (C) HPMCs were cultured either with absence of FBS, 10% FBS or 10% malignant ascites (OVC508) and representation phase contrast images were taken (x 200 magnification). Bars 100 μm. (D) The relative expression of TGF-β1 was determined as described in Material & Methods for peritoneal benign fluids (OV370 and OV401) and malignant ascites (OVC346 and OVC508). The solid line indicates the cut -off intensity (750 FU) for a p ositive signal. *indicate P < 0.001, T-student test. Matte et al. BMC Cancer 2014, 14:288 Page 4 of 13 http://www.biomedcentral.com/1471-2407/14/288 on HPMCs was observed for up to 48 h with malignant ascites (Figure 2B). To ensure that the effect of ascites was not limited to a single HPMC culture, we also tested the effect of ascites on Meso-9 mesothelial culture. Malignant ascites (OVC509) also enhanced the growth of Meso-9, although these cells grew at a much slower rate than the OVC508 (10%) OV370 (10%) 0 h 12 h FBS (10%)OV401 (10%) Benign fluids OVC346 (10%) Malignant ascites 12 h + HU OV401 (10%)OVC508 (10%) 0 h 48 hA B 0 h 48 h OV370 (10%)OVC509 (10%) Meso-7 cells Meso-7 cells Meso-9 cells C LPA (uM) E Malignant ascitesBenign fluids 0 1 2 3 4 5 6 7 8 9 10 OV370 OV410 OVC346 OVC508 * 0.00 0.10 0.20 0.30 0.40 0.50 0.60 02 4 4 8 7 2 9 6 Control FBS OV401 OVC346 O.D. (nm) Time (h) D Figure 2 Effect of ascites on HPMC proliferation. (A) HPMCs were cultured either with 10% FBS, 10% benign peritoneal fluids or 10% malignant ascites in the presence or absence of hydroxyurea (HU) for 12 h and phase contrast images were taken (x 200 magnification). Bars 100 μm. (B-C) Phase contrast pictures of Meso-7 and Meso-9 cells cultured either with 10% benign fluid (OV370 or OV401) or 10% malignant ascites (OVC508 or OVC509).Bars 100 μM. (D) HPMC (meso-7 cells) were seeded and cell growt hf o ru pt o9 6hw a sd e t e r m i n e db yX T Ta s s a y .(E) Determination of LPA levels in benign fluids or malignant ascites. There was no significant difference between levels of LPA in OV401, OVC508 and OVC509 (P > 0.05). Levels of LPA in OV370 were however significantly higher. *indicateP < 0.01, T-student test. Matte et al. BMC Cancer 2014, 14:288 Page 5 of 13 http://www.biomedcentral.com/1471-2407/14/288 Meso-7 cells suggesting that the effect of malignant ascites on growth is reproducible in different HPMC culture (Figure 2C). The cell growth of HPMCs in the pres- ence of benign fluid (OV401) and malignant ascites OVC346 was also monitored by XTT assay and dem- onstrated that OVC346 stimulated cell growth whereas OV401 did not (Figure 2D). These data suggest that ascites contain soluble factors that stimulate the prolif- eration of the two patient-derived HPMC cultures. LPA is a growth factor-like phospholipid present in the serum and ascites of patients with OC and promotes tumor cell proliferation [6]. LPA has been reported to be present at higher concentration in malignant ascites when compared to benign fluids [6]. However, we found that LPA levels were not consistently higher in malignant ascites OVC346 and OVC508 when compared to benign fluids (Figure 2E). A more extensive analysis of LPA levels in benign fluids (n = 17) versus serous OC (n = 20) also failed to show higher levels of LPA in serous OC (median 1.9 μM±1 . 1f o r benign fluids versus 3.0 μM±1 . 9f o rs e r o u sO C ;P >0 . 0 5 ) . Malignant ascites-stimulated HPMCs secrete soluble factors that attenuate TRAIL-induced apoptosis Soluble factors produced by cancer-associated fibroblasts and bone marrow stromal cells have been shown to con- fer resistance to TRAIL-induced apoptosis in tumor cells [19-21]. We reasoned that malignant ascites-stimulated HPMCs might also secrete soluble factors that could attenuate TRAIL-induced apoptosis. HPMCs were incu- bated with benign fluids or malignant ascites overnight. The cells were then washed twice and conditioned media (CM) were collected 12 h later. Ovarian cancer CaOV3 cells were treated with TRAIL in presence of CM from HPMCs exposed to either benign fluids or ma- lignant ascites and apoptosis was measured (Figure 3A). As shown in Figure 3B, TRAIL-induced apoptosis was decreased in CaOV3 cells exposed to CM from malig- nant ascites-exposed HPMCs as compared to CM from benign fluid-exposed HPMCs. These results suggest that ascites-stimulated HPMCs secrete soluble factors that attenuate TRAIL-induced apoptosis. To examine the ef- fect of ascites exposure on the secretion of soluble factors overtime, HPMCs were stimulated with malignant ascites or benign fluids overnight. Cells were then washed twice and CM were collected after 8, 12 and 24 h. Whereas CM from benign fluid-stimulated HPMCs collected at differ- ent time did not affect TRAIL-induced apoptosis (OV370, OV401), CM from ascites-stimulated HPMCs significantly reduced apoptosis in CaOV3 cells (Figure 3C). The max- imum protection was observed at 12 h. Gene expression changes induced by malignant ascites The expression profiles from HPMC cultures exposed to peritoneal fluids and OC ascites were compared using Malignant ascitesBenign fluids Apoptosis Fold increase relative to control (no TRAIL) P = 0.02B Apoptosis Fold increase relative to control (no TRAIL) Time (h) C * * * HPMCs primed by incubation with ascites o/n HPMCs cultured in FBS 10% OC cells incubated in HPMC-conditioned medium Medium conditioned by primed-HPMCs 8-24 hA HPMCs washed TRAIL (25 ng/ml) CaOV3 cells 24 h 0 1 2 3 4 5 6 7 OV370 OV401 OVC346 OVC508 0 1 2 3 4 5 6 7 8 9 10 0 8 12 24 OVC346 OVC508 OV370 OV401 Figure 3 TRAIL-induced apoptosis in ascites-stimulated HPMCs. (A) Diagram of HPMC-priming assays. Ascites-stimulated or benign fluid-stimulated HPMCs were culture overnight (shown in yellow), washed TWICE and cultured in serum/hormone-free medium for 8 to 24 h to generate HPMC-conditioned medium (shown in pink) that were collected at either 12 h(B) or different time points(C). HPMC-conditioned medium was then added to CaOV3 tumor cells in the presence of TRAIL (25 ng/ml). TRAIL-induced apoptosis was measured in CaOV3 cells incubated with the indicated HPMC-conditioned medium overnight and expressed as fold increased relative to cells th at were exposed to HPMC-conditio ned medium but not to TRAIL. Data are expressed as means of triplicates from three independent experiments ± SD. * indicate P < 0.01. Matte et al. BMC Cancer 2014, 14:288 Page 6 of 13 http://www.biomedcentral.com/1471-2407/14/288 Table 1 Up-regulated genes in HPMCs exposed to malignant ascites versus to benign fluids (Fold change > 2, P < 0.05) Unigene Symbol Description Functions Fold change Hs.425633 CKMT1B Creatine kinase mitochondrial 1B Mitochondria 31.80 Hs.282409 CYP2C19 Cytochrome P450, family 2, subfamily C Electron transport 29.10 Hs.560 APOBEC1 Apolipoprotein B mRNA editing enzyme, catalytic polypeptide 1 mRNA processing 17.19 Hs.516922 NKX2-2 NK2 transcription factor related locus 2 Transcription 16.20 Hs.553484 ANGPT2 Angiopoietin-2A Angiogenesis 16.16 Hs.145932 MTL5 Metallothionein-like 5 Anti-apoptotic 11.40 Hs.445098 DEPDC1 DEP domain containing 1 Intracellular signaling 7.06 Hs.149924 LILRB1 Leucocyte immunoglobulin-like receptor, subfamily B Immune response 6.31 Hs.471494 EIF2C4 mRNA for KIAA1567 Protein synthesis 6.06 Hs.272215 KLF15 Kruppel-like factor 15 Transcription 5.74 Hs.285829 DLEU7 Deleted in lymphocytic leukemia 7 Intracellular signaling 5.23 Hs.6702 KCNB2 Potassium voltage-gated chanel Cation transport 5.08 Hs.351403 DZIP1L DAZ interacting protein 1-like Ion binding 4.60 Hs.479853 EPHA5 EPH receptor A5 Intracellular signaling 4.54 Hs.122583 UTG2A3 UDP glucuronosyltransferase 2 family Metabolism 4.23 Hs.446021 SLC37A3 Solute carrier family 37 Transport 4.02 Hs.270833 AREG Amphiregulin Cell proliferation 3.52 Hs.467793 FLJ40869 Hypothetical protein FLJ40869 DNA repair 3.31 Hs.441975 BIRC4BP XIAP-associated factor 1 Ion binding 3.26 Hs.531941 MYB myeloblastosis viral oncogene homolog Transcription 2.83 Hs.585869 ZNF224 Zinc finger protein 224 Transcription 2.57 Hs.188518 MT1K Metallothionein 1 K Ion binding 2.54 Hs.190043 MOSPD2 Motile sperm domain containing 2 Membrane protein 2.52 Hs.270543 GNB4 Guanine nucleotide binding protein, beta polypeptide 4 Intracellular signaling 2.44 Hs.190622 DDX58 DEAD box polypeptide 58 DNA binding 2.37 Hs.658169 SFRP4 Secreted frizzled-related protein 4 Intracellular signaling 2.37 Hs.13852 DNAJB4 Hsp40 homolog, subfamily B, member 4 Protein folding 2.36 Hs.189920 PB1 Polybromo 1 Transcription 2.35 Hs.513044 CSPG4 Chondroitin sulfate proteoglycan 4 Cell motility 2.32 Hs.520506 FBXO5 F-box protein 5 Protein degradation 2.28 Hs.83634 HCFC1 Host cell factor C1 Cell proliferation 2.25 Hs.14794 ZFP28 Zinc finger protein 28 homolog Transcription 2.19 Hs.10319 UGT2B7 UDP glucuronosyltransferase 2 family Metabolism 2.17 Hs.503093 ZFP36L2 Zinc finger protein 36, C3H type-like 2 Cell proliferation 2.14 Hs.110915 IL22RA1 Interleukin 22 receptor, alpha 1 Immune response 2.13 Hs.442344 IRS2 Insulin receptor substrate 2 Cell proliferation 2.11 Hs.479670 TEC Tec protein tyrosine kinase Intracellular signaling 2.10 Hs.167700 SMAD5 SMAD, family member 5 Transcription 2.10 Hs.541894 ANKRD36 Ankirin repeat domain 36 Catalytic activity 2.09 Hs.655199 MT1A Metallothionein 1A Ion binding 2.09 Hs.407190 FKBP5 FK506 binding protein 5 Protein folding 2.06 Hs.744289 ZNF549 Zinc finger protein 549 Ion binding 2.06 Hs.431550 MAP4K4 Mitogen-activated protein kinase kinase kinase kinase 4 Intracellular signaling 2.04 Hs.433702 EIF5 Eukaryotic translation initiation factor 5 Translation 2.04 Hs.167584 SLC2A2 Solute carrier family 2 Transport 2.03 Hs.405144 SFRS3 Splicing factor, arginine/serine-rich 3 RNA binding 2.02 Matte et al. BMC Cancer 2014, 14:288 Page 7 of 13 http://www.biomedcentral.com/1471-2407/14/288 the Whole Human Genome Oligonucleotide Microarray (Agilent), containing ~ 44,000 genes. Microarrays were performed on HPMCs exposed to 3 malignant ascites from women with advanced (stage III/IV) serous OC and two benign peritoneal fluids. First, we generated lists of significantly up-reg ulated and down-regulated genes that were differentially expressed between OC ascites (OVC346, OVC508 and OVC509) and control OV370 peritoneal fluid. Then, the set of genes that were commonly expressed between control peritoneal fluids (OV370 and OV401) were subtracted from the first list of genes to generate a dataset of differentially Table 2 Down-regulated genes in HPMCs exposed to malignant ascites versus to benign fluids (Fold change > 2, P < 0.05) Unigene Symbol Description Functions Fold change Hs.602085 PHLDA1 Pleckstrin homology-like domain, member 1 Nuclear protein -3.40 Hs.624 IL-8 Interleukin 8 Angiogenesis -3.38 Hs.98367 SOX17 SRY-box 17 Transcription -3.33 Hs.345139 GEM GTP binding protein overexpressed in skeletal muscle Intracellular signaling -2.58 Hs.73853 BMP2 Bone morphogenetic protein 2 Cell proliferation -2.58 Hs.251526 CCL7 Chemokine ligand 7 Inflammatory response -2.54 Hs.591159 PTHLH Parathyroide hormone-like hormone Ion binding -2.44 Hs.643357 ADAMTS1 A disintegrin and metallopeptidase with thrombospondin type 1 Cell proliferation -2.34 Hs.517310 RIPK4 Receptor-interacting serine-threonine kinase 4 Intracellular signaling -2.32 Hs.616962 GDF15 Growth differentiation factor 15 Intracellular signaling -2.31 Hs.279522 NR4A3 Nuclear receptor subfamily 4, group A, member 3 Transcription -2.30 Hs.25829 RASD1 RAS, dexamethasone-induced 1 Intracellular signaling -2.27 Hs.516826 TRIB3 Tribbles homolog 3 Transcription -2.22 Hs.406714 KRTAP2-4 Keratin associated protein 2-4 Cell structure -2.20 Hs.505146 CLDN14 Claudin 14 Cell-cell adhesion -2.20 Hs.799 HBEGF Heparin-binding EFG-like growth factor Growth factor -2.18 Hs.525572 BDKRB1 Bradykinin receptor B1 Inflammatory response -2.18 Hs.191215 PSCD1 Pleckstrin homology, Sec7 and coiled-coil domains Transport -2.07 Hs.76095 IER3 Immediate early response 3 Apoptosis -2.04 Hs.18676 SPRY2 Sprouty homolog 2 Cell-cell signaling -2.03 AB Up-regulated upon malignant ascites treatment Down-regulated upon malignant ascites treatment Figure 4 Functional analysis for the dataset of differentially expressed genes (≥ 1.5-fold) in ascites-stimulated HPMC cells.Top functions that meet a P value cutoff of 0.05 are displayed. The orange line represents the cutoff value for significance. (A) Genes that were up-regulated and (B) genes down-regulated. Matte et al. BMC Cancer 2014, 14:288 Page 8 of 13 http://www.biomedcentral.com/1471-2407/14/288 expressed genes between malignant ascites and benign peritoneal fluids. A subset of 649 genes was thus selected by filtering on confidence at P value = 0.05, followed by filtering on expression levels ( ≥ 1.5 fold). We found 484 genes to be commonly up-regulated and 185 genes to be down-regulated in HPMCs exposed to malignant ascites. Top molecules that were up-regulated are shown in T able 1 and those down-regulated in T able 2. Pathway and network analysis based on the 649 genes list were generated through the use of Ingenuity Pathways Analysis (IPA). IPA showed that the top two pathways up-regulated in this gene list were functionally associated with the regulation of cell cycle and apoptosis (Figure 4A) which is consistent with data from Figures 2 and 3. Genes implicated in cell death and cell growth and proliferation (mostly negative regulators) were among the top pathways down-regulated (Figure 4B) (T able 3). Networks linked to cancer, inflammatory response, cell movement, cell assem- bly and organization, cell-to-cell signaling, DNA replica- tion, and repair and recombination were both induced or suppressed. The analysis recognized several important nodes linked with numerous partners, including nuclear factor-κB( N F -κB), Akt, heat shock protein 90 (Hsp90), hepatocyte nuclear factor 4 α (HNF-4α), KRAS, SMAD1, RNA helicase p68 (DDX5B, p68), c-KIT ligand (KITLG), vascular endothelial growth factor (VEGF), interleukin-8 (IL-8), follicle stimulating hormone (FSH), colony stimu- lating factor 2 (CSF2), cyclin-dependent kinase inhibitor 1A (CDKN1A, p21, Clip1), bone morphogenetic protein 2 (BMP2) (Figure 5A and B). While some of the up-regulated gene nodes and related pathways were associated with posi- tive feedbacks on the cell cycle (NF- κB, KRAS, KITLG, MAP4K4, HCFC1), some down-regulated genes were nega- tive regulators of the cell cycle (CDKN1A/2B, DUSP6/10). Validation of microarray findings with quantitative RT-PCR To validate the results of the microarray analysis, we used quantitative real-time PCR to quantify the expres- sion of selected genes including PTHLH, INHBA, PHLDA1, IRS2 and KTR-18 in ascites-stimulated HPMCs compared to benign fluid (OVC370, O VC401)-stimulated HPMCs. qRT-PCR analysis confirmed our microarray findings for PTHLH, INHBA and PHLDA1 genes which were down- regulated, and for IRS2 and KTR-18 which were up- regulated (Figure 6A). qRT-PCR analysis was also performed with a third peritoneal fluid OV1081 along with OV370 to validate the differential expression of IL-8 and BMP2 in malignant ascites (Figure 6B). The expression of IL-8 and BMP2 were down-regulated in HPMCs stimulated with malignant ascites as compared to both OV1081 and OV370 benign fluids.

A crucial step in OC progression is the ability of tumor cells to shed from the primary tumor site and proliferate and survive in ascites. In this context, ascites must pro- vide a milieu that support tumor cell growth. OC ascites are rich, heterogeneous and complex fluids that harbor a wide variety of soluble factors that are part of an auto- crine and paracrine network in tumor cells. In line with these observations, the presence of ascites correlates with peritoneal spread of OC tumors [22] and signifi- cantly decreases the 5-year survival rate for women with advanced OC [23]. Malignant ascites offer OC cells a network of proliferative and survival factors; thus OC cells floating in ascites receive signals that alter gene expression which confer a survival advantage. Indeed, it was recently demonstrated that ascites promote the acti- vation of survival pathways in tumor cells, which contrib- ute to attenuate drug-induced apoptosis [7-9]. Changes in tumor cell behavior are mediated by the activation of vari- ous signaling pathways such as PI3K/Akt and MAPK/ERK pathways in these cells [8,9]. HPMCs present in ascites are theoretically exposed to those same factors and conse- quently receive similar signals. To better understand the role of HPMCs in OC progression and how ascites signals may alter their behavior, we characterized the effects of malignant ascites on HPMC morphology and prolifera- tion, and correlated these effects with molecular alter- ations in gene expression occurring in HPMCs after Table 3 Cell death biological function category showing a significant fold change in genes belonging to HPMCs exposed to malignant ascites compared to benign fluids (Fold change > 2, P < 0.05) Symbol Description Fold change BDKRB1 Bradykinin receptor B1 -2.18 IL-8 Interleukin 8 -3.38 SOX17 SRY-box 17 -3.33 BMP2 Bone morphogenetic protein 2 -2.58 NR4A3 Nuclear receptor subfamily 4, group A, member 3 -2.30 RASD1 RAS, dexamethasone-induced 1 -2.27 PHLDA1 Pleckstrin homology-like domain, member 1 -3.40 TRIB3 Tribbles homolog 3 -2.22 HBEGF Heparin-binding EFG-like growth factor -2.20 SPRY2 Sprouty homolog 2 -2.03 AREG Amphiregulin 3.52 DDX58 DEAD box polypeptide 58 2.37 DEPDC1 DEP domain containing 1 7.06 IRS2 Insulin receptor substrate 2 2.11 SMAD5 SMAD, family member 5 2.10 MT1A Metallothionein 1A 2.10 MAP4K4 Mitogen-activated protein kinase kinase kinase kinase 4 2.04 Matte et al. BMC Cancer 2014, 14:288 Page 9 of 13 http://www.biomedcentral.com/1471-2407/14/288 exposure to malignant OC ascites. We used low passage two patient-derived HPMC cultures that were derived from peritoneal fluids and exposed these cells to either malignant ascites or benign peritoneal fluids. We analyzed functionally related genes that were commonly differen- tially expressed following exposure of HPMCs to all ma- lignant ascites compared to benign peritoneal fluids. The current study demonstrates that OC ascites con- sistently induce a switch of morphology in HPMCs from an epithelial to a fibroblastic pattern, a finding that has been reported by other groups when HPMCs were incu- bated with TGF- β1 [14,15]. In contrast, benign fluids failed to induce such a switch. Interestingly, levels of TGF-β1 were below the threshold of positivity in benign fluids whereas TGF-β1 was detectable in malignant ascites, although levels were low (~ 1400 FU). TGF- β1i sc o n s i d - ered a critical regulator of epithelial-to-mesenchymal tran- sition (EMT). The essential features of EMT include the downregulation of epithelial cell markers ( e.g. E-cadherin, claudins) and the upregulated expression of fibroblastic markers ( e.g. vimentin, N-cadherin). TGF- β1-induced EMT is mediated by Smad-dependent and –independent signaling [24]. Whether the low level of TGF- β1f o u n di n malignant ascites is responsible for the morphologic changes that were observed in HPMCs is unclear. Smad1 and Smad5 genes were up-regulated by malignant ascites which is consistent with the involvement of TGF- β1. Sig- naling pathways involved in EMT such as PI3K/Akt and Ras/MAPK were also up-regulated by malignant ascites (Figure 5A). All these findings are consistent with an im- portant role for TGF- β1. However, growth factors other than TGF- β1, such as hepatocyte growth factor (HGF), fibroblast growth factor (FGF) or epidermal growth factor (EGF), which are found in malignant ascites [5], may also activate these signaling pathways and induce EMT [24]. In the current study, we observed that the three OC ascites tested stimulated the proliferation of HPMCs. In contrast, the two peritoneal fluids did not stimulate proliferation. This suggests that the malignant ascites tested contain growth-promoting activity. In line with this observation, malignant ascites were also found to stimulate the prolif- eration of OC cells in vitro [10]. Malignant ascites contain several growth factors that could potentially stimulate the proliferation of mesothelial cells [5]. Among these factors, LPA is of particular interest. In the present study, we showed that LPA is detectable in both malignant ascites Figure 5 Network analysis of dynamic gene expression in ascites-stimulated HPMCs based on the common up-regulated (A) or down- regulated (B) ( ≥ 1.5-fold) gene expression list obtained following stimulation with all three malignant ascites. The top-scoring networks were merged and displayed graphically as node (gene/gene product) and edges (the biological relationships between the nodes). Nodes are displayed using various shapes that represent the functional class of the gene product (square, cytokine; vertical oval, transmembrane receptor; rectangle, nuclear receptor; diamond, enzyme; rhomboid, transporter; hexagon, translation factor; horizontal oval, transcription factor; circle, other). Edges are displayed with various labels that described the nature of relationship between the nodes: — binding only; → acts on. The length of an edge reflects the evidence supporting that node-to-node relationship, in that edges supported by article from literature are shorter. Dotted edges represent indirect interaction. Matte et al. BMC Cancer 2014, 14:288 Page 10 of 13 http://www.biomedcentral.com/1471-2407/14/288 and in benign fluids (Figure 2E). It has been previously reported that LPA is present at 20-80 μM concentrations in the ascites of OC patients [25-27]. LPA is a factor in ascites from OC patients that promote the proliferation as well as the migration of OC cells [28-30]. Serous OC LPA levels were lower in this study (median 3.0 μM) compared to previous studies. Most importantly however, the levels of LPA were not significantly different in serous OC com- pared to benign fluids. These observations suggest that, in the two malignant ascites tested, LPA may not be a critical factor for ascites-mediated proliferation of the two sam- ples of HPMCs. Consistent with the findings that malignant ascites stimulate HPMC proliferation in vitro , we found that cell cycle- and cell growth-related genes were up- and down- regulated by malignant ascites. In total, the expression of 85 genes involved in cell proliferation was altered by malignant ascites. In particular, several cyclin-dependent kinase (Cdk) inhibitors (p21, p15) and dual specificity phosphatases (DUSP6, DUSP10) were down-regulated. Upon stimulation by growth factors, downstream targets such as cyclin D1 are activated by the ERK pathway, which is activated by LPA [30], resulting in progression from G1 to S. Cdks inhibitors such as p21 and p15 can block G1 progression. Dusp6 and Dusp10 acts as nega- tive feedback regulators of ERK signalling [24,31]. Con- versely, genes such as receptor tyrosine kinase KIT, its ligand stem cell factor (SCF) and KRAS, which induce ERK phosphorylation and promote cell proliferation [32], were upregulated by ascites. 0 0.2 0.4 0.6 0.8 1 1.2 OV1081 OVC346 OVC509 0 0.2 0.4 0.6 0.8 1 1.2 OV1081 OVC346 OVC509 IL-8 BMP-2 qRT-PCR fold change qRT-PCR fold change qRT-PCR fold change qRT-PCR fold change qRT-PCR fold change PTHLH qRT-PCR fold change INHBA qRT-PCR fold change PHLDA1 IL-8 BMP-2 A B IRS2 qRT-PCR fold change KTR-18 Malignant ascites Malignant ascites Malignant ascites Malignant ascites Malignant ascites Malignant ascites Malignant ascites Malignant ascitesMalignant ascites 0 0.2 0.4 0.6 0.8 1 1.2 1.4 OVC370 OVC346 OVC509 0 0.2 0.4 0.6 0.8 1 1.2 OVC370 OVC346 OVC509 * * * * 0 0.2 0.4 0.6 0.8 1 1.2 OVC370 OVC346 OVC509 * * 0 0.2 0.4 0.6 0.8 1 1.2 OVC370 OVC346 OVC509 * * 0 0.2 0.4 0.6 0.8 1 1.2 1.4 OVC401 OVC346 * 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 OVC401 OVC346 qRT-PCR fold change * 0 0.2 0.4 0.6 0.8 1 1.2 OVC370 OVC346 OVC509 * * Figure 6 Correlation between gene expression data and quantitative PCR. (A) Levels of mRNA expression in ascites-stimlated HPMCs expressed as fold changes relative to benign fluiids OVC370- or OVC401-stimulated HPMCs. (B) Relative expression of IL-8 and BMP-2 mRNA in ascites-stimulated HPMCs compared to either benign fluid OV1081- or OVC370-stimulated HPMCs. * indicate P < 0.03. Matte et al. BMC Cancer 2014, 14:288 Page 11 of 13 http://www.biomedcentral.com/1471-2407/14/288 Our data indicate that the two OC ascites tested in- duce the secretion of factors by HPMCs (at least in the meso-7 samples) that attenuate TRAIL-induced apop- tosis in tumor cells. This observation implies that ascites activate HPMCs through paracrine interactions and acti- vated HPMCs secrete factors that promote the survival of tumor cells. Indeed, many genes differentially expressed in HPMCs stimulated by malignant ascites are closely related to the regulation of apoptosis. The apoptosis-related genes include a total of 47 genes that were down-regulated and 58 that were up-regulated (Figure 4). Interestingly, stem cell factor (SCF) and its receptor (c-kit) were among the genes that were up-regulated. Myb transcription factor, which serves as a regulator of c-kit expression, was up-regulated by ascites in HPMCs. SCF/c-kit pathway has been implicated in a variety of processes including cell survival [33]. SCF signals via c-kit through PI3K/Akt and Ras/MAPK pathways, two well-establish survival pathways [33]. Ahmed et al. showed that ascites activate Ras/MAPK signaling in OC cells [34]. Our group also demonstrated that OC ascites stimulate MAPK/ERK1/2 pathway leading to the regulation of Mcl-1 antiapoptotic protein in OC cells [9].

In summary, this study provides evidence that activation of HPMCs is mediated by paracrine interactions with soluble factors in malignant ascites. These factors stimulate a phenotypic shift from an epithelial to a fibroblastic morphology in HPMCs. Asci tes-stimulated HPMCs are proliferative and secrete soluble factors that promote tumor cell survival. Although the nature of these factors remains to be determined, they likely promote a survival advantage for tumor cells. Paracrine factors in ascites activate intracel- lular signaling network such as Akt and NF- κBi nH P M C s which mediate, in turn, the up-regulation of HPMC-secreted factors that impact OC progression. One limitation of this study is that data were derived from a small number of samples, thus conclusions should be viewed appropriately. Validation in a larger set of patients will be beneficial. Future studies assessing the nature of paracrine and autocrine stimulating signals will help to better define the interplay between HPMCs and tumor cells that is important for OC progression. Competing interests The authors declare that they have no competing interest. Authors’ contributions IM participated in the design of the study and performed all functional assays. IM was also responsible for obtaining the clinical samples. DL performed the LPA measurements in the clinical samples. DB performed the microarray data analysis. CB performed the characterization of HPMCs by immunofluorescence. CR participated in the design of the study and helped to draft the manuscript. AP conceived the study, participated in its design and drafted the manuscript. All authors read and approved the final manuscript. Acknowledgments This work was supported by funds from the Canadian Institute for Health Research (MOP-244194-CPT-CFDA-48852) and Centre de Recherche Clinique Étienne-Lebel. Tumor banking was supported by the Banque de tissus et données of the Réseau de recherche sur le cancer of the Fonds de recherche du Québec – Santé (FRQS), associated with the Canadian Tumor Repository Network (CTRNet). The tissue bank provided malignant ascites, peritoneal fluids and mesothelial cells derived from primary samples. Author details 1Département de Microbiologie et Infectiologie, Faculté de Médecine, Université de Sherbrooke, 3001, 12ième Avenue Nord, Sherbrooke, Québec, J1H 5N4, Canada. 2Cancer Research Centre, Hôpital L ’Hôtel-Dieu de Québec, Centre Hospitalier Universitaire de Québec, 9 rue McMahon, Québec, G1R 2J6, Canada. Received: 22 July 2013 Accepted: 3 April 2014 Published: 24 April 2014

1. Partridge EE, Barnes MN: Epithelial ovarian cancer: prevention, diagnosis, and treatment. CA Cancer J Clin 1999, 49:297–320. 2. Kipps E, Tan DS, Kaye SB: Meeting the challenge of ascites in ovarian cancer: new avenues for therapy and research. Nat Rev Cancer 2013, 13:273–282. 3. Bast RC Jr, Hennessy B, Mills GB: The biology of ovarian cancer: new opportunities for translation. Nat Rev Cancer 2009, 9:415–428. 4. Lane D, Matte I, Rancourt C, Piché A: The prosurvival activity of ascites against TRAIL is associated with a shorter disease-free interval in patients with ovarian cancer. J Ovarian Res 2010, 3:1. 5. Matte I, Lane D, Laplante C, Rancourt C, Piché A: Profiling of cytokines in human epithelial ovarian cancer ascites. Am J Cancer Res 2012, 2:566–580. 6. Mills GB, Moolenaar WH: The emerging role of lysophosphatidic acid in cancer. Nat Rev Cancer 2003, 3:582–591. 7. Lane D, Robert V, Grondin R, Rancourt C, Piché A: Malignant ascites protect against TRAIL-induced apo ptosis by activating the PI3K/Akt pathway in human ovarian carcinoma cells.Int J Cancer 2007, 121:1227–1237. 8. Lane D, Goncharenko-Khaider N, Rancourt C, Piché A: Ovarian cancer ascites protects from TRAIL-induced cell death through αvβ5 integrin-mediated focal adhesion kinase and Akt activation. Oncogene 2010, 29:3519–3531. 9. Goncharenko-Khaider N, Matte I, Lane D, Rancourt C, Piché A: Ovarian cancer ascites increase Mcl-1 expression in tumor cells through ERK1/2-Elk-1 signaling to attenuate TRAIL-induced apoptosis. Mol Cancer 2012, 11:84. 10. Puiffe ML, Le Page C, Filali-Mouhim A, Zietarska M, Ouellet V, Tonin PN, Chevrette M, Provencher DM, Mes-Masson AM: Characterization of ovarian cancer ascites on cell invasion, proliferation, spheroid formation, and gene expression in anin vitro model of epithelial ovarian cancer. Neoplasia 2007, 9:820–829. 11. Mills GB, May C, Hill M, Campbell S, Shaw P, Marks A: Ascitic fluid from human ovarian cancer patients contains growth factors necessary for intraperitoneal growth of human ovarian adenocarcinoma cells. J Clin Invest 1990, 86:851–855. 12. Yamada T, Sato K, Komachi M, Malchinkhuu E, Tobo M, Kimura T, Kuwabara A, Yanagita Y, Ikeya T, Tanahashi Y, Ogawa T, Ohwada S, Morishita Y, Ohta H, I mD S ,T a m o t oK ,T o m u r aH ,O k a j i m aF :Lysophosphatidic acid (LPA) in malignant ascites stimulates motility of human pancreatic cancer cells through LPA1. JB i o lC h e m2004, 279:6595–6605. 13. Davidson B: Malignant effusions: from diagnosis to biology. Diagn Cytopathol 2004, 31:246–254. 14. Na D, Lv ZD, Liu FN, Xu Y, Jiang CG, Sun Z, Miao ZF, Li F, Xu HM: Transforming growth factor β1 produced in autocrine/paracrine manner affects the morphology and function of mesothelial cells and promotes peritoneal carcinomatosis. Int J Mol Med 2010, 26:325–332. 15. Ikubo A, Morisaki T, Katano M, Kitsuki H, Anan K, Uchiyama A, Tanaka M, Torisu M: Ap o s s i b l er o l eo fT G F -β in the formation of malignant effusions. Clin Immunol Immunopathol 1995, 77:27–32. 16. Zhang H, Liu X, Liu Y, Yi B, Yu X: Epithelial-mesenchymal transition of rat peritoneal mesothelial cells via RhoA/Rock pathway. In Vitro Cell Dev Biol Anim 2011, 47:165–172. 17. Kajiyama H, Shibata K, Ino K, Nawa A, Mizutani S, Kikkawa F: Possible involvement of SDF-1α/CXCR4-DPPIV axis in TGF-β1-induced enhancement Matte et al. BMC Cancer 2014, 14:288 Page 12 of 13 http://www.biomedcentral.com/1471-2407/14/288 of migratory potential in human peritoneal mesothelial cells. Cell Tissue Res 2007, 330:221–229. 18. Oates J, Edwards C: HBME-1, MOC-31, WT1 and calretinin: an assessment of recently described markers for mesothelioma and adenocarcinoma. Histopathology 2000, 36:341–347. 19. Shipman CM, Croucher PI: Osteoprotegerin is a soluble decoy receptor for tumor necrosis factor-related apoptosis-inducing ligand/Apo2 ligand and can function as a paracrine survival factor for human myeloma cells. Cancer Res 2003, 63:912–916. 20. Neville-Webbe HL, Cross NA, Eaton CL, Nyambo R, Evan CA, Coleman RE, Holen I: Osteoprotegerin (OPG) produced by bone marrow stromal cells protects breast cancer cells from TRAIL-induced apoptosis. Breast Cancer Res Treat 2004, 86:269–279. 21. Meads MB, Hazlehurst LA, Dalton WS: The bone marrow microenvironment as a tumor sanctuary and contributor to drug resistance. Clin Cancer Res 2008, 14:2519–2526. 22. Parsons SL, Lang MW, Steele RJ: Malignant ascites: a 2-year review from a teaching hospital. Eur J Surg Oncol 1996, 22:237–239. 23. Ayantunde AA, Parsons SL: Pattern and prognostic factors in patients with malignant ascites : a retrospective study. Ann Oncol 2007, 18:945–949. 24. Xu J, Lamouille S, Derynck R: TGF-β-induced epithelial to mesenchymal transition. Cell Res 2009, 19:156–172. 25. Kazi JU, Vaapil M, Agarwal S, Bracco E, Pahlman S, Rönnstrand L: The tyrosine kinase CSK associates with FLT3 and c-Kit receptors and regulates downstream signaling. Cell Signal 2013, 25:1852–1860. 26. Shen Z, Wu M, Elson P, Kennedy AW, Belinson J, Casey G, Xu Y: Fatty acid composition of lysophosphatidic acid and lysophosphatidylinositol in plasma from patients with ovarian cancer and other gynecological diseases. Gynecol Oncol 2001, 83:25–30. 27. Xu Y, Shen Z, Wiper DW, Wu M, Morton RE, Elson P, Kennedy AW, Belinson J, Markman M, Casey G: Lysophosphatidic acid as a potential biomarker for ovarian and other gynecologic cancers. JAMA 1998, 280:719–723. 28. Westermann AM, Havik E, Postma FR, Beijnen JH, Dalesio O, Moolenaar WH, Rodenhuis S: Malignant effusions contain lysophosphatidic acid (LPA)-like activity. Ann Oncol 1998, 9:437–442. 29. Mills GB, May C, McGill M, Roifman CM, Mellors A: A putative new growth factor in ascitic fluid from ovarian cancer patients: identification, characterization, and mechanism of action. Cancer Res 1988, 48:1066–1071. 30. Xu Y, Fang XJ, Casey G, Mills GB: Lysophospholipids activate ovarian and breast cancer cells. Biochem J 1995, 309:933–940. 31. Zeliadt NA, Mauro LJ, Wattenberg EV: Reciprocal regulation of extracellular signal regulated kinase 1/2 and mitogen activated protein kinase phosphatase-3. Toxicol Appl Pharmacol 2008, 232:408–417. 32. Nomura M, Shiiba K, Katagiri C, Kasugai I, Masuda K, Sato I, Sato M, Kakugawa Y, Nomura E, Hayashi K, Nakamura Y, Nagata T, Otsuka T, Katakura R, Yamashita Y, Sato M, Tanuma N, Shima H: Novel function of MKP-5/DUPS10, a phosphatase of stress-activated kinases, on ERK-dependent gene expression, and upregulation of its gene expression in colon carcinomas. Oncol Rep 2012, 28:931–936. 33. Bian D, Su S, Mahanivong C, Cheng RK, Han Q, Pan ZK, Sun P, Huang S: Lysophosphatidic acid stimulates ovarian cancer cell migration via a Ras-MEK kinase 1 pathway. Cancer Res 2004, 64:4209–4217. 34. Lennartsson J, Rönnstrand L: Stem cell factor receptor/c-Kit: from basic science to clinical implications. Physiol Rev 2012, 92:1619–1649. doi:10.1186/1471-2407-14-288 Cite this article as: Matte et al. : Role of malignant ascites on human mesothelial cells and their gene expression profiles. BMC Cancer 2014 14:288. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Matte et al. BMC Cancer 2014, 14:288 Page 13 of 13 http://www.biomedcentral.com/1471-2407/14/288