- Email: [email protected]
Malignant ascites determine the transmesothelial invasion of ovarian cancer cells
Accepted Manuscript Title: Malignant ascites determine the transmesothelial invasion of ovarian cancer cells Authors: Justyna Mikuła-Pietrasik, Paweł Uruski, Sebastian Szubert, Dariusz Szpurek, Stefan Sajdak, Andrzej Tykarski, Krzysztof Ksi˛az˙ ek PII: DOI: Reference:
S1357-2725(17)30214-5 http://dx.doi.org/10.1016/j.biocel.2017.09.002 BC 5209
To appear in:
The International Journal of Biochemistry & Cell Biology
Received date: Revised date: Accepted date:
8-5-2017 3-9-2017 5-9-2017
Please cite this article as: Mikuła-Pietrasik, Justyna., Uruski, Paweł., Szubert, Sebastian., Szpurek, Dariusz., Sajdak, Stefan., Tykarski, Andrzej., & Ksi˛az˙ ek, Krzysztof., Malignant ascites determine the transmesothelial invasion of ovarian cancer cells.International Journal of Biochemistry and Cell Biology http://dx.doi.org/10.1016/j.biocel.2017.09.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Malignant ascites determine the transmesothelial invasion of ovarian cancer cells
Justyna Mikuła-Pietrasik1, Paweł Uruski1, Sebastian Szubert2, Dariusz Szpurek2, Stefan Sajdak2, Andrzej Tykarski1, Krzysztof Książek1 *
1
Department of Hypertensiology, Angiology and Internal Medicine, Poznań University of
Medical Sciences, Długa 1/2 Str., 61-848 Poznań, Poland. E-mail: [email protected] (J.MP); [email protected] (PU); [email protected] (AT); [email protected] (KK) 2
Division of Gynecological Surgery, Poznań University of Medical Sciences, Polna 33 Str,
60-535 Poznań, Poland. E-mail: [email protected] (SSz); [email protected] (DS); [email protected] (SSa)
*Correspondence: Prof. Krzysztof Książek, Department of Hypertensiology, Angiology and Internal Medicine, Poznań University of Medical Sciences, Długa 1/2 Str., 61-848 Poznań, Poland, Phone +48 61 854-92-99, Fax: +48 61 854-90-86, E-mail: [email protected] 1
Graphical Abstract
Highlights
Malignant ascites promote invasion of ovarian cancer cells. Ascites from type II tumors are more proinvasive than those from type I tumors. Ascites deteriorate mesothelial junctions in oxidative stress-dependent mechanism. Ascites affect junctional proteins via p38 MAPK- and NF-B-dependent signaling.
ABSTRACT The exact role of malignant ascites in the development of intraperitoneal ovarian cancer metastases remains unclear. In this report we sought to establish if ascites can determine the efficiency of transmesothelial invasion of ovarian cancer cells, and, if so, whether the fluid generated by highly aggressive serous and undifferentiated tumors will promote the invasion more effectively than ascites from less aggressive clear cell and endometrioid cancers. The study showed that the invasion of ovarian cancer cells (SKOV-3 and primary cancer cells) across monolayered peritoneal mesothelial cells was elevated upon mesothelial cell exposure to fluid produced by serous and undifferentiated cancers, as compared with cells subjected to ascites from clear cell and endometrioid tumors. This effect coincided with decreased mesothelial expression of junctional proteins: connexin 43, E-cadherin, occludin, and desmoglein. Moreover, it was accompanied by transforming growth factor 1-dependent 2
overproduction of reactive oxygen species by these cells. The activity of ascites from serous and undifferentiated tumors was mediated by p38 mitogen-activated protein kinase and nuclear factor B. When the mesothelial cells were protected against oxidative stress, both deterioration of junctional proteins and intensification of cancer cell invasion in response to ascites from serous and undifferentiated tumors were effectively prevented. In conclusion, our findings indicate that the high aggressiveness of some histotypes of ovarian cancer may be related to the ability of malignant ascites generated by these cells to oxidative stressdependent impairment of mesothelial cell integrity and the resulting increase in their transmesothelial invasion.
Key words: intercellular junctions; invasion; malignant ascites; ovarian cancer; peritoneal mesothelium
1. Introduction Epithelial ovarian cancer (EOC) is a heterogenous disease with respect to its origin, genetics, and aggressiveness, and hence it is categorized into type I and type II tumors. Type I is represented by low-grade serous, endometrioid, mucinous, malignant Brenner and clear cell cancers which originate from the ovaries, display high genetic stability, and develop slowly. Type II includes high-grade serous, carcinosarcoma, and undifferentiated cancer, it comes from outside the ovaries and is characterized by a high progression rate and mortality (Shih and Kurman, 2004). As per the genetic background, type I cancers bear mutations in the KRAS, BRAF, PTEN, and PI3K genes, while type II tumors have mutated p53 and mutated or dysfunctional BRCA1/2 genes (Romero and Bast, Jr., 2012). At the same time, although clinical and genetic differences between type I and type II tumors are well defined, cellular
3
pathomechanisms responsible for the high aggressiveness of the latter type remain to be explored. In this study we verified the hypothesis that the diverse aggressiveness of ovarian cancer histotypes may be associated with the activity of malignant ascites, which is a pathological fluid that accumulates in the peritoneum in a significant group of patients (Cvetkovic, 2003). In particular, we examined whether the transmesothelial invasion of cancer cells, which is one of the most critical elements of the formation of intraperitoneal metastasis (Steinkamp et al., 2013), proceeding in the presence of fluid from serous and undifferentiated ovarian cancers (type II tumors) would be more intense than ascites from clear cell and endometrioid cancers (type I tumors). Mechanistically, we addressed the ascites’ effect on the expression of various junctional proteins that provide the integrity of the peritoneal mesothelium and thus restrict the efficiency at which cancer cells move towards the tissue stroma (Defamie et al., 2014). Oxidative stress and related signaling pathways were analyzed to delineate factors regulating the ascites-dependent deterioration of intercellular junctions and those contributing to an increased transmesothelial invasion of ovarian cancer cells.
2. Materials and methods 2.1. Materials Unless otherwise stated, all chemicals and plastics were from Sigma (St. Louis, MO, USA). MG132 (the inhibitor of NF-B), API-1 (the inhibitor of AKT), and SP600125 (the inhibitor of JNK) were purchased from Tocris Bioscience (Ellisville, MO, USA), whereas SB202190 (the inhibitor of p38 MAPK) was from Cell Signaling Technology (Danvers, MA, USA). Exogenous, recombinant human transforming growth factor β1 (TGF-β1) was purchased from R&D Systems (Abingdon, UK).
4
2.2. Cell cultures Human peritoneal mesothelial cells (HPMCs) were isolated from pieces of omentum obtained from 8 patients (28-32 years old) undergoing elective abdominal surgery (institutional consent number 187/14), as described in detail elsewhere (Ksiazek, 2013). The reasons for the surgery included aortic aneurysm (4), hernia (3), and bowel obstruction (1). The cultures were maintained at 37 °C in a humidified atmosphere of 95% air and 5% CO2. Cells were identified as pure mesothelial by their typical cobblestone appearance at confluency and uniform positive staining for cytokeratins and HBME-1 antigen. Primary cultures of HPMCs obtained during the 1st passage (corresponding to ~5% of their replicative lifespan) without any contamination with stromal cells were used in the experiments. Ovarian cancer cells, SKOV-3, were purchased from the ECCC (Porton Down, UK) and propagated in RPMI 1640 medium with L-glutamine (2 mmol/L), penicillin (100 U/ml), streptomycin (100 g/ml), and 10% FBS. Primary epithelial ovarian cancer cells (EOCs) were isolated from a tumor excised during cytoreductive surgery from a patient with serous ovarian carcinoma (stage III according to FIGO). Briefly, the tumor was divided with a scalpel into ten pieces of equal weight and then placed in a solution of 0.05% trypsin and 0.02% EDTA for 20 min at 37 °C with gentle shaking. After resuspension in RPMI1640 containing 20% FBS, the cells were probed with an antibody directed against the epithelial-related antigen (MOC-31; Abcam, Cambridge, UK) to confirm their cancerous nature. Finally, ovarian cancer cells were cultured in RPMI 1640 supplemented with L-glutamine (2 mM) and 20% FBS.
2.3. Malignant and benign ascites Malignant ascites were obtained at the time of cytoreductive surgery from patients with high-grade serous, undifferentiated, clear cell and endometrioid ovarian carcinoma at stage
5
IIIC and IV (n=8 per group). The histopathology, grade, and stage of the tumors were assigned in accordance with the criteria of the International Federation of Gynecology and Obstetrics. Patients received no chemotherapy prior to surgery. Control, benign fluids were obtained from 8 age-matched patients undergoing abdominal surgery due to the presence of non-cancerous lesions - cystadenoma mucinosum multiloculare. Upon collection in sterile conditions, the fluids were centrifuged at 2500 rpm for 5 min and then cell-free supernatants were stored in aliquots at -20 C until required. The study was approved by an institutional ethics committee (consent number 543/14).
2.4. Cancer cell invasion assay Analysis of cancer cell invasion was performed using Cultrex 96 Well BME Cell Invasion Assay (Trevigen Inc. Gaithersburg, MD, USA) as per manufacturer’s instructions. In brief, HPMCs (1x105 cells per well) was seeded on the basement membrane extract (BME) to form a monolayer. Afterwards, the cells were subjected to 10% malignant and benign ascites (50 µl per well) for 72 h. After the incubation, ovarian cancer cells (1x104 cells per well) were placed into the upper chamber of the system, that is on top of HPMCs. The cancer cells invaded through the HPMCs lying on the BME towards a chemotactic gradient generated by 1% FBS. The intensity of fluorescence emitted by the cancer cells was recorded using a SynergyTM 2 spectrofluorometer (BioTek Instruments, Winooski, VT, USA) at 435 nm excitation and 535 nm emission wavelengths, respectively. In some experiments, cancer cell invasion was examined in the presence of HPMCs subjected to malignant and benign ascites (10%, for 72 h) upon their pre-incubation (for 4 h) with the ROS spin-trap scavenger, N-tertbutyl-alpha-phenylnitrone (PBN, Sigma; 800 µM).
2.5. Analysis of intercellular junctions
6
In order to examine expression of junctional proteins, HPMCs were seeded into semitransparent 96-well plates (1x105 cells per well), and then they were allowed to form a monolayer for next 24 h. Upon reaching the confluency, the plates were carefully washed to remove all non-adherent cells and debris, and then the cells were subjected to 10% malignant and benign ascites (50 µl per well) for 72 h. After the incubation, the cells were fixed in paraformaldehyde, washed and treated with antibodies against connexin 43 (cat # ab11370, Abcam, 1:100, overnight), E-cadherin (cat # ab15148, Abcam, 1:100, overnight), occludin (cat # NBP1-87402, Novus Biologicals, Littleton, CO, USA, 1:100, overnight), and desmoglein (cat # ab12077, Abcam, 1:10, overnight). Then the cells were extensively washed with phosphate-buffered saline (PBS) and incubated with DyLight 488 antibody (cat # ab96899, Abcam; 1:500, 1h) to quantify connexin 43, E-cadherin and occludin, and with Alexa Fluor 488 antibody (cat # ab150113, Abcam, 1:500, 1h) to quantify desmoglein. Finally, the cells were washed three times with PBS and fluorescence emitted was recorded using a SynergyTM 2 spectrofluorometer (BioTek Instruments, Winooski, VT, USA). Representative pictures of immunoreactions were taken using an Axio Vert.A1 microscope (Carl-Zeiss, Jena, Germany). In some experiments, the expression of junctional proteins was measured in HPMCs exposed to both malignant and benign ascites (10%, for 72 h) upon their pre-incubation (12 h) with MG132 (10 µM) and SB202190 (10 µM) for 12 h. In other experiments the proteins were analyzed upon HPMC pre-treatment (for 4 h) with PBN (800 µM). Additionally, the expression of the junctional proteins was quantified in cell homogenates using specific, colorimetric ELISA-based kits purchased from Abbexa Ltd (Cambridge, UK), as per manufacturer’s instructions. In order to collect appropriate amounts of cellular protein, the experiments were performed with HPMCs seeded in 24-well plates (5x105 cells per well) and allowed to form the monolayer for 24 h. Upon the exposure to 10% malignant and benign
7
ascites (250 µl per well) for 72 h or the exposure to MG132 and SB202190 for 12 h, the cells were homogenized by sonication. The homogenates were centrifuged at 5000 × g for 5 min and the supernatants collected were stored at -80 C until assayed.
2.6. Production of reactive oxygen species (ROS) In order to examine the production of ROS, HPMCs were seeded into semi-transparent 96well plates (1x105 cells per well), and then they were allowed to form a monolayer for next 24 h. Upon reaching the confluency, the plates were carefully washed to remove all non-adherent cells and debris, and then the cells were subjected to 10% malignant and benign ascites (50 µl per well) for 72 h. The ROS release by HPMCs was assessed using 2’,7’dichlorodihydrofluorescein diacetate (H2DCFDA), essentially as described in (MikulaPietrasik et al., 2012). In some experiments, ROS production was monitored in cells exposed to exogenous, recombinant TGF-β1 used at concentrations of 100-400 pg/ml for 72 h. In another group of experiments, ROS were measured in HPMCs subjected to malignant and benign ascites (10%, for 72 h) upon their pre-incubation with MG132 (10 µM), API-1 (50 µM), SP600125 (10 µM), and SB202190 (10 µM) for 12 h.
2.7. Statistics Statistical analysis was conducted with GraphPad Prism 5.00 software (GraphPad Software, San Diego, USA). The means were compared using repeated measures analysis of variance (ANOVA) with the Newman-Keuls test as a post-hoc test. When appropriate, the Wilcoxon matched pairs test was used. The results were expressed as means SD. Differences with a P value < 0.05 were considered to be statistically significant.
3. Results and Discussion 8
There is a growing body of evidence that malignant ascites may play some role in the development of ovarian tumors (Cvetkovic, 2003). It has been found that ascites may promote the development of tumor metastases by suppressing peritoneal inflammation (SimpsonAbelson et al., 2013) and that the ascites’ biochemical composition may, at least theoretically, facilitate angiogenesis and proliferation of cancer cells (Matte et al., 2012). Interestingly, a recent report revealed that the proinflammatory characteristics of the fluid generated by undifferentiated tumors, known to be the most lethal ovarian cancer histotype (Silva et al., 1991), may underlie the high proliferative and migratory potential of the cancer cells (MikulaPietrasik et al., 2016c). In general, however, the mechanistic aspects of the intraperitoneal spread of ovarian tumors associated with the presence and activity of malignant ascites, including the background of differences in the aggressiveness of type I and type II tumors, remain elusive. In this report we showed that malignant ascites determines the efficiency at which both established (SKOV-3) and primary EOC cells invade across the peritoneal mesothelial cells (HPMCs). Taking into account that a single layer of HPMCs forms a physical barrier on the way of the cancer cells towards the peritoneal stroma (Iwanicki et al., 2011), our observation sheds new light on the pathomechanism controlling the development of intraperitoneal ovarian tumors. Significantly, as is shown in Fig. 1, the effectiveness of cancer cell movement was particularly high when the cells invaded through the mesothelium subjected to ascites from serous and undifferentiated tumors (vs. clear cell and endometrioid cancer, and vs. benign fluid from non-cancerous patients). Among the two cancer histotypes displaying high aggressiveness, ascites from patients with undifferentiated tumors were more pro-invasive. This effect resembles that when malignant ascites generated by undifferentiated tumors promoted migration of A2780, OVCAR-3, and SKOV-3 ovarian cancer cells more efficiently
9
than ascites from serous carcinoma (Mikula-Pietrasik et al., 2016c). In this context it should be emphasized that the high ability of cancer cells to invade across monolayered HPMCs as revealed in the current study may to some extent be overlapped by increased cancer cell migration which proceeds in an HPMCs-independent manner. In fact, ascites generated by type II tumors, in particular by the undifferentiated histotype, are rich in several chemotactic agents, e.g. CCL2, CXCL1, CXCL5, CXCL8, and CXCL12 (Mikula-Pietrasik et al., 2016c). In our opinion, both increased invasion and migration should be considered in order to visualize the whole process of transmesothelial cancer cell movement. It is worth noting that the findings available in the literature regarding the role of ascitic fluid in cancer cell invasion, in particular the findings provided by Puiffe et al., are inconclusive (Puiffe et al., 2007). The authors of the study cited above show that malignant ascites differ in their capacity to modulate cancer cell invasion, i.e. some of them exert the inhibitory effect while some stimulate the process. We believe that this inconsistent picture may result from at least three independent variables: i) referring the results to 5% serum instead of benign ascites, ii) the use of mixed, heterologous fluids instead of an analysis of fluids derived from a given histotype, and iii) the measurement of cell invasion across the Matrigel instead of an analysis of cell movement through the mesothelial cells lying on the Matrigel or its equivalent. Because our experimental approach includes all of the aspects as postulated above, we are convinced that the results presented here conclusively indicate that the malignant ascites possesses clearly proinvasive capabilities. The integrity of the HPMCs’ monolayer largely depends on the expression and functionality of various intercellular junctions and is considered a limiting factor for cancer cell invasion (Kenny et al., 2007). Having this in mind, we examined the effect of malignant ascites on four junctional proteins which are representative of the peritoneal mesothelium, i.e. connexin 43 (gap junction), E-cadherin (adherens junction), occludin (tight junction), and
10
desmoglein (desmosome) (Simionescu and Simionescu, 1977). As shown in Figs. 2 and 3, and in Tab. 1, HPMCs exposed to malignant ascites from type I tumors (clear cell and endometrioid) displayed unaltered expression of E-cadherin, occludin, and desmoglein, and slightly decreased expression of connexin 43. In contrast, HPMCs subjected to malignant ascites from type II tumors (serous and undifferentiated) were characterized by significantly decreased expression of all four tested intercellular junctions. As per connexin 43, its decline was much more pronounced as compared with the effects exerted by ascites from clear cell and endometrioid cancers. Significantly, in the case of E-cadherin, occludin, and desmoglein but not connexin 43 (where the situation was opposite), the effects generated by ascites from undifferentiated tumors were again considerably stronger than those resulting from the activity of their serous counterparts. These results suggest a scenario in which the increased potential of ovarian cancer cells to invade through HPMCs treated with malignant ascites from aggressive tumors may be associated with increased “permeability” of the mesothelium due to decreased expression of junctional proteins. This possibility is in accordance with results of Giverso et al., who combined experiments in vitro with 3D multiscale mathematical analysis and proved that an efficient invasion of ovarian cancer cells requires the loosening of junctional proteins in the peritoneal mesothelium (Giverso et al., 2010). Having established that malignant ascites-related deterioration of the junctional proteins leading to a weakening of the structural and functional connection between adjacent HPMCs may be the reason for the intensified transmesothelial invasion of ovarian cancer cells, we focused on a mechanistic background of this phenomenon. Because decreased expression of intercellular junctions, e.g. connexin 43, has been found in senescent cells (Statuto et al., 2002), and the intensification of HPMCs’ senescence upon their exposure to malignant ascites was described recently (Mikula-Pietrasik et al., 2016a), we paid special attention to the role of
11
oxidative stress, which is the key determinant of several senescence-related abnormalities in HPMCs (Ksiazek, 2013). Our study revealed that HPMCs subjected to malignant ascites generate a higher amount of ROS than cells exposed to benign fluid. Moreover, the magnitude of oxidative stress in cells treated with ascites from type II tumors was higher as compared with their counterparts exposed to fluid generated by type I tumors. Here also the greatest ROS production was observed in HPMCs stimulated with ascites obtained from patients with undifferentiated ovarian cancer (Fig. 4A). The prooxidative impact of malignant ascites on HPMCs is an entirely new finding; however, its seems to adhere to earlier results obtained by Yang et al., who found that the ascites modulates the antioxidative potential of ovarian cancer cells, thus supporting their ability to adhere and invade (Yang et al., 2005). Intervention studies using a spin-trap ROS scavenger, PBN, showed that either increased cancer cell invasion across HPMCs (Tab. 2) or decreased expression of junctional proteins (Tab. 3) elicited by malignant ascites (particularly those generated by undifferentiated and serous tumors) can be effectively prevented when the HPMCs are protected against oxidative stress. These results, pointing to negative regulation of mesothelial intercellular junctions by ROS, are in line with the observations of another group which revealed the deterioration of connexin 43 in cells exposed to various prooxidative stimuli (Berthoud and Beyer, 2009). They are also in agreement with a report showing a disruption of epithelial tight junctions upon the activity of calcium-mediated oxidative stress (Gangwar et al., 2017). The search for a soluble mediator of increased oxidative stress in HPMCs treated with malignant ascites directed our attention towards transforming growth β1 (TGF-β1), a cytokine whose capacity to elevate ROS release has already been well documented for various cell types, including HPMCs (Mikula-Pietrasik et al., 2013). Interestingly, our previous study showed that the concentration of TGF-β1 in the malignant ascites from undifferentiated
12
ovarian tumors was higher when compared with fluids generated by clear cell and endometrioid cancers, and comparable with that characterizing the ascites from patients with serous carcinoma (Mikula-Pietrasik et al., 2016c). The role of TGF-β1 as an ascites-derived, prooxidative agent was confirmed experimentally when we used an exogenous recombinant form of this cytokine and applied it on monolayered HPMCs at concentrations corresponding to its level in the four tested types of malignant ascites. The production of ROS by HPMCs treated with TGF-β1 appeared to increase in a dose-dependent manner (Fig. 4B) under such a regimen. In order to identify the signaling pathways underlying the oxidative stress-related decrease in the expression of junctional proteins in HPMCs exposed to malignant ascites from serous and undifferentiated tumors, the cells were pre-incubated with chemical inhibitors of four pathways known to be associated with the activity of either TGF-β1 or ROS, i.e. AKT, JNK, NF-B, and p38 MAPK (Krstic et al., 2015). Afterwards, the magnitude of oxidative stress and the expression of the intercellular junctions in HPMCs were re-investigated. The results shown in Fig. 5 indicate that the production of ROS in HPMCs subjected to malignant ascites can be inhibited upon neutralization of NF-B and p38 MAPK but not AKT and JNK. Inhibition of the same signaling molecules resulted in restoration of the expression of all of the tested junctional proteins to values characterizing cells treated with benign ascites (Fig. 6 and Tab. 4). The contribution of NF-B and p38 MAPK in an ascites-dependent deterioration of junctional proteins in HPMCs may suggest that the functional alterations in those cells that promote cancer cell invasion may contain a component associated with their senescence. This statement is based on the fact that both NF-B and p38 MAPK have already been recognized to mediate some senescence-related changes in HPMCs (e.g. the development of the secretory phenotype (Mikula-Pietrasik et al., 2015)). In addition, the neutralization of p38 MAPK has been found to rejuvenate the HPMCs and to postpone their cancer-promoting capabilities 13
(Mikula-Pietrasik et al., 2016b). Taking into account the fact that malignant ascites from serous ovarian cancer have already been found to stimulate senescence in HPMCs (MikulaPietrasik et al., 2016a), the current study extends this effect to ascites produced by undifferentiated tumors whose activity seems to be even more pronounced.
4. Conclusions In conclusion, the results we provide in this paper indicate that the severe aggressiveness of serous and undifferentiated ovarian cancers may be related, at least to some extent, to the activity of the malignant ascites that is generated by these cells. In particular, it may be associated with their ability to destroy the integrity of HPMCs, thus leading to improved transmesothelial invasion of the cancer cells. A significant role in all of these processes is played by oxidative stress. Taking this into account, there is an urgent need to examine whether intraperitoneal administration of antioxidants, e.g. catalase (Nishikawa et al., 2009), could prevent or at least restrict the malignant ascites-dependent progression of highly aggressive ovarian tumors.
Acknowledgments The study was supported by a grant from the National Science Centre, Poland (registration number 2014/15/B/NZ3/00421).
Conflict of interest The authors declare no conflict of interest.
References
14
Berthoud, V.M., Beyer, E.C., 2009. Oxidative stress, lens gap junctions, and cataracts. Antioxid. Redox. Signal. 11, 339-353. Cvetkovic, D., 2003. Early events in ovarian oncogenesis. Reprod. Biol. Endocrinol. 1, 68. Defamie, N., Chepied, A., Mesnil, M., 2014. Connexins, gap junctions and tissue invasion. FEBS Lett. 588, 1331-1338. Gangwar, R., Meena, A., Shukla, P., Nagaraja, A., Dorniak, P., Pallikuth, S., Waters, C. Sood, A., Rao, R., 2017. Calcium-Mediated Oxidative Stress: a Common Mechanism in Tight Junction Disruption by Different Types of Cellular Stress. Biochem. J. 474, 731-749. Giverso, C., Scianna, M., Preziosi, L., Lo Buono, N., Funaro, A., 2010. Individual Cell-Based Model for In-Vitro Mesothelial Invasion of Ovarian Cancer. Math. Model. Nat. Phenom 5, 1-21. Iwanicki, M.P., Davidowitz, R.A., Ng, M.R., Besser, A., Muranen, T., Merritt, M., Danuser, G., Ince, T.A., Brugge, J.S., 2011. Ovarian cancer spheroids use myosin-generated force to clear the mesothelium. Cancer Discov. 1, 144-157. Kenny, H.A., Krausz, T., Yamada, S.D., Lengyel, E., 2007. Use of a novel 3D culture model to elucidate the role of mesothelial cells, fibroblasts and extra-cellular matrices on adhesion and invasion of ovarian cancer cells to the omentum. Int. J. Cancer 121, 14631472. Krstic, J., Trivanovic, D., Mojsilovic, S., Santibanez, J.F., 2015. Transforming Growth Factor-Beta and Oxidative Stress Interplay: Implications in Tumorigenesis and Cancer Progression. Oxid. Med Cell Longev. 2015, 654594. Ksiazek, K. 2013. Mesothelial cell: A multifaceted model of aging. Ageing Res Rev. 12, 595604. Matte, I., Lane, D., Laplante, C., Rancourt, C., Piche, A., 2012. Profiling of cytokines in human epithelial ovarian cancer ascites. Am. J Cancer Res 2, 566-580.
15
Mikula-Pietrasik, J., Kuczmarska, A., Kucinska, M., Murias, M., Wierzchowski, M., Winckiewicz, M., Staniszewski, R., Breborowicz, A., Ksiazek, K., 2012. Resveratrol and its synthetic derivatives exert opposite effects on mesothelial cell-dependent angiogenesis via modulating secretion of VEGF and IL-8/CXCL8. Angiogenesis. 15, 361-376. Mikula-Pietrasik, J., Sosinska, P., Janus, J., Rubis, B., Brewinska-Olchowik, M., Piwocka, K., Ksiazek, K., 2013. Bystander senescence in human peritoneal mesothelium and fibroblasts is related to thrombospondin-1-dependent activation of transforming growth factor-beta1. Int. J. Biochem. Cell Biol. 45, 2087-2096. Mikula-Pietrasik, J., Sosinska, P., Maksin, K., Kucinska, M., Piotrowska, H., Murias, M. Wozniak, A., Szpurek, D., Ksiazek, K., 2015. Colorectal cancer-promoting activity of the senescent peritoneal mesothelium. Oncotarget. 6, 29178-29195. Mikula-Pietrasik, J., Uruski, P., Matuszkiewicz, K., Szubert, S., Moszynski, R., Szpurek, D., Sajdak, S., Tykarski, A., Ksiazek, K., 2016a. Ovarian cancer-derived ascitic fluids induce a senescence-dependent pro-cancerogenic phenotype in normal peritoneal mesothelial cells. Cell Oncol (Dordr. ) 39, 473-481. Mikula-Pietrasik, J., Uruski, P., Sosinska, P., Maksin, K., Piotrowska-Kempisty, H., Kucinska, M., Murias, M., Szubert, S., Wozniak, A., Szpurek, D., Sajdak, S., Piwocka, K., Tykarski, A., Ksiazek, K., 2016b. Senescent peritoneal mesothelium creates a niche for ovarian cancer metastases. Cell Death. Dis. 7, e2565. Mikula-Pietrasik, J., Uruski, P., Szubert, S., Moszynski, R., Szpurek, D., Sajdak, S., Tykarski, A., Ksiazek, K., 2016c. Biochemical composition of malignant ascites determines high aggressiveness of undifferentiated ovarian tumors. Med Oncol 33, 94. Nishikawa, M., Hashida, M., Takakura, Y., 2009. Catalase delivery for inhibiting ROSmediated tissue injury and tumor metastasis. Adv Drug Deliv. Rev. 61, 319-326.
16
Puiffe, M.L., Le, P.C., Filali-Mouhim, A., Zietarska, M., Ouellet, V., Tonin, P.N., Chevrette, M., Provencher, D.M., Mes-Masson, A.M., 2007. Characterization of ovarian cancer ascites on cell invasion, proliferation, spheroid formation, and gene expression in an in vitro model of epithelial ovarian cancer. Neoplasia. 9, 820-829. Romero, I., Bast Jr, R.C., 2012. Minireview: human ovarian cancer: biology, current management, and paths to personalizing therapy. Endocrinology 153, 1593-1602. Shih, I., Kurman, R.J., 2004. Ovarian tumorigenesis: a proposed model based on morphological and molecular genetic analysis. Am. J Pathol. 164, 1511-1518. Silva, E.G., Tornos, C., Bailey, M.A., Morris, M., 1991. Undifferentiated carcinoma of the ovary. Arch Pathol. Lab Med 115, 377-381. Simionescu, M., Simionescu, N., 1977. Organization of cell junctions in the peritoneal mesothelium. J Cell Biol. 74, 98-110. Simpson-Abelson, M.R., Loyall, J.L., Lehman, H.K., Barnas, J.L., Minderman, H., O'Loughlin, K.L., Wallace, P.K., George, T.C., Peng, P., Kelleher Jr, R.J., Odunsi, K., Bankert, R.B., 2013. Human ovarian tumor ascites fluids rapidly and reversibly inhibit T cell receptor-induced NF-kappaB and NFAT signaling in tumor-associated T cells. Cancer Immun. 13, 14. Statuto, M., Bianchi, C., Perego, R., Del, M.U., 2002. Drop of connexin 43 in replicative senescence of human fibroblasts HEL-299 as a possible biomarker of senescence. Exp Gerontol. 37, 1113-1120. Steinkamp, M.P., Winner, K.K., Davies, S., Muller, C., Zhang, Y., Hoffman, R.M., Shirinifard, A., Moses, M., Jiang, Y., Wilson, B.S., 2013. Ovarian tumor attachment, invasion, and vascularization reflect unique microenvironments in the peritoneum: insights from xenograft and mathematical models. Front Oncol. 3, 97.
17
Yang, W., Toffa, S.E., Lohn, J.W., Seifalian, A.M., Winslet, M.C., 2005. Malignant ascites increases the antioxidant ability of human ovarian (SKOV-3) and gastric adenocarcinoma (KATO-III) cells. Gynecol. Oncol 96, 430-438.
Tables
Table 1 The quantification of junctional proteins in HPMCs exposed to benign and malignant ascites. Concentration (ng/ml) Connexin 43 E-cadherin Occludin Desmoglein
Benign ascites and malignant ascites generated by ovarian cancer cells Benign Clear cell Endometrioid Serous Undifferentiated 2.4±0.3 6.3±0.2 1.6±0.1 9.6±0.1
1.9±0.1* 6.5±0.1 1.7±0.2 10.3±0.5
1.8±0.2* 6.2±0.3 1.4±0.2 10.6±0.4
0.8±0.2*# 6.2±0.3 0.8±0.3*# 6.3±0.2*#
1.3±0.3*# 2.7±0.1*# 0.5±0.2*# 2.1±0.3*#
* P<0.05 vs. HPMCs subjected to benign ascites, # P<0.05 vs. CC, P<0.05 vs. Endo. The experiments were performed with HPMCs established from 8 different donors. Ascites (malignant and benign) were obtained from 8 patients per group.
18
Table 2 The effect of HPMC protection against oxidative stress on the invasion of ovarian cancer cells across HPMCs subjected to malignant ascites.
Malignant ascites generated by ovarian cancer cells Clear cell Endometrioid Serous Undifferentiated -PBN +PBN -PBN +PBN -PBN +PBN -PBN +PBN SKOV-3 Primary EOC
127±8* 145±3*
104±2# 111±6#
125±4* 136±9*
103±1# 109±4#
158±9* 192±6*
104±2# 122±6#
200±11* 217±6*
111±3# 104±9#
The results are expressed as the percentage of invasion of cancer cells through HPMCs exposed to benign fluid from non-cancerous patients, treated as 100%. * P<0.05 vs. HPMCs exposed to benign fluid; # P<0.05 vs. HPMCs not exposed to PBN. The experiments were performed with HPMCs established from 8 different donors. Cancer cells were used in duplicates. Ascites (malignant and benign) were obtained from 8 patients per group.
19
Table 3 The effect of HPMC protection against oxidative stress on the expression of junctional proteins in cells treated with malignant ascites. Malignant ascites generated by ovarian cancer cells Clear cell Endometrioid Serous Undifferentiated -PBN +PBN -PBN +PBN -PBN +PBN -PBN +PBN Connexin 43 E-cadherin Occludin Desmoglein
82±7* 102±2 104±3 102±7
99±1# 103±2 102±1 107±2
85±7* 101±4 93±1* 105±2
101±3# 101±2 99±2# 108±8
42±4* 101±5 85±7* 78±8*
92±6# 99±3 102±1# 106±6#
72±7* 68±4* 61±4* 44±9*
101±3# 101±2# 106±2# 102±6#
The results are expressed as the percentage of the expression of junctional proteins in HPMCs subjected to benign fluid from non-cancerous patients, treated as 100%. * P<0.05 vs. HPMCs exposed to benign fluid; # P<0.05 vs. HPMCs not exposed to PBN. The experiments were performed with HPMCs established from 8 different donors. Ascites (malignant and benign) were obtained from 8 patients per group.
20
Table 4
The quantification of junctional proteins in HPMCs exposed to malignant ascites from serous and undifferentiated ovarian tumors upon the inhibition of NF-B and p38 MAPK
Concentration (ng/ml) Connexin 43 E-cadherin Occludin Desmoglein Concentration (ng/ml) Connexin 43 E-cadherin Occludin Desmoglein
Benign ascites and malignant ascites from serous ovarian cancer Serous + Serous + Benign Serous MG132 SB202190 2.4±0.2 6.5±0.1 1.5±0.2 9.1±0.4
0.9±0.2* 6.3±0.6 1.0±0.1* 7.2±0.2*
2.5±0.1# 6.1±0.4 1.7±0.2# 9.6±0.2#
2.3±0.3# 6.5±0.2 1.7±0.1# 9.9±0.3#
Benign ascites and malignant ascites from undifferentiated ovarian cancer Undiff + Undiff + Benign Undiff MG132 SB202190 2.3±0.2 6.6±0.4 1.9±0.3 9.4±0.3
1.1±0.2* 4.1±0.2* 1.3±0.2* 3.4±0.2*
2.2±0.4# 7.4±0.2# 2.1±0.3# 10.1±0.7#
2.6±0.2# 7.1±0.1# 2.0±0.1# 11.1±0.6#
* P<0.05 vs. HPMCs subjected to benign ascites, # P<0.05 vs. HPMCs subjected to malignant ascites without their pre-incubation with signaling pathway inhibitors. The experiments were 21
performed with HPMCs established from 8 different donors and with ascites (malignant and benign) from 8 patients per group.
Figure legends
Fig. 1. Effect of malignant ascites from patients with clear cell (CC), endometrioid (Endo), serous (Ser), and undifferentiated (Undiff) ovarian tumors on transmesothelial invasion of SKOV-3 cells (A) and primary epithelial ovarian cancer cells (B). The results are presented as scatter dot plots in order to show the variability of ascites obtained from different patients. * P<0.05 vs. HPMCs subjected to benign ascites, # P<0.05 vs. CC, P<0.05 vs. Endo. The experiments were performed with HPMCs established from 8 different donors. The cancer cells were used in duplicates. Ascites (malignant and benign) were obtained from 8 patients per group. BA – benign ascites from non-cancerous patients; RFU – relative fluorescence units.
Fig. 2. Effect of malignant ascites from patients with clear cell (CC), endometrioid (Endo), serous (Ser), and undifferentiated (Undiff) ovarian tumors on the expression of connexin 43 (A), E-cadherin (B), occludin (C), and desmoglein (D) in HPMCs. * P<0.05 vs. HPMCs subjected to benign ascites, # P<0.05 vs. CC, P<0.05 vs. Endo. The experiments were performed with HPMCs established from 8 different donors. Ascites (malignant and 22
benign) were obtained from 8 patients per group. BA – benign ascites from non-cancerous patients; RFU – relative fluorescence units.
Fig. 3. Representative images of immunofluorescence against junctional proteins in HPMCs exposed to malignant ascites from patients with clear cell (CC), endometrioid (Endo), serous (Ser), and undifferentiated (Undiff) ovarian tumors and to benign ascites (BA) from non-cancerous patients. The stainings were performed with HPMCs established from 8 different donors and with ascites (malignant and benign) from 8 patients per group. Magnification x 200. Fig. 4. Effect of malignant ascites (A) from patients with clear cell (CC), endometrioid (Endo), serous (Ser), and undifferentiated (Undiff) ovarian tumors and the exogenous, recombinant form of human TGF-β1 (B) on the production of ROS in HPMCs. * P<0.05 vs. HPMCs subjected to benign ascites, # P<0.05 vs. CC, P<0.05 vs. Endo, ** P<0.05 vs. cells not treated with rhTGF-β1. The experiments were performed with HPMCs established from 8 different donors and with ascites (malignant and benign) from 8 patients per group. BA – benign ascites from non-cancerous patients; RFU – relative fluorescence units.
Fig. 5. Analysis of signaling pathways underlying the activity of malignant ascites from type II ovarian tumors. Effect of the inhibition of AKT, JNK, NF-B, and p38 MAPK on the production of ROS by HPMCs exposed to malignant ascites from serous (A) and undifferentiated cancer (B). The experiments were performed with HPMCs established from 8 different donors and with ascites (malignant and benign) from 8 patients per group. * P<0.05 vs. HPMCs subjected to benign ascites, # P<0.05 vs. HPMCs subjected to malignant ascites without their pre-incubation with signaling pathway inhibitors. MA – malignant ascites; BA – benign ascites; RFU – relative fluorescence units.
23
Fig. 6. Effect of NF-B and p38 MAPK inhibition on the expression of connexin 43, Ecadherin, occludin, and desmoglein in HPMCs exposed to malignant ascites from serous (A-D) and undifferentiated (E-H) ovarian tumors. The experiments were performed with HPMCs established from 8 different donors and with ascites (malignant and benign) from 8 patients per group. * P<0.05 vs. HPMCs subjected to benign ascites, # P<0.05 vs. HPMCs subjected to malignant ascites without their pre-incubation with signaling pathway inhibitors. MA – malignant ascites; BA – benign ascites; RFU – relative fluorescence units.
24
S1357-2725(17)30214-5 http://dx.doi.org/10.1016/j.biocel.2017.09.002 BC 5209
To appear in:
The International Journal of Biochemistry & Cell Biology
Received date: Revised date: Accepted date:
8-5-2017 3-9-2017 5-9-2017
Please cite this article as: Mikuła-Pietrasik, Justyna., Uruski, Paweł., Szubert, Sebastian., Szpurek, Dariusz., Sajdak, Stefan., Tykarski, Andrzej., & Ksi˛az˙ ek, Krzysztof., Malignant ascites determine the transmesothelial invasion of ovarian cancer cells.International Journal of Biochemistry and Cell Biology http://dx.doi.org/10.1016/j.biocel.2017.09.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Malignant ascites determine the transmesothelial invasion of ovarian cancer cells
Justyna Mikuła-Pietrasik1, Paweł Uruski1, Sebastian Szubert2, Dariusz Szpurek2, Stefan Sajdak2, Andrzej Tykarski1, Krzysztof Książek1 *
1
Department of Hypertensiology, Angiology and Internal Medicine, Poznań University of
Medical Sciences, Długa 1/2 Str., 61-848 Poznań, Poland. E-mail: [email protected] (J.MP); [email protected] (PU); [email protected] (AT); [email protected] (KK) 2
Division of Gynecological Surgery, Poznań University of Medical Sciences, Polna 33 Str,
60-535 Poznań, Poland. E-mail: [email protected] (SSz); [email protected] (DS); [email protected] (SSa)
*Correspondence: Prof. Krzysztof Książek, Department of Hypertensiology, Angiology and Internal Medicine, Poznań University of Medical Sciences, Długa 1/2 Str., 61-848 Poznań, Poland, Phone +48 61 854-92-99, Fax: +48 61 854-90-86, E-mail: [email protected] 1
Graphical Abstract
Highlights
Malignant ascites promote invasion of ovarian cancer cells. Ascites from type II tumors are more proinvasive than those from type I tumors. Ascites deteriorate mesothelial junctions in oxidative stress-dependent mechanism. Ascites affect junctional proteins via p38 MAPK- and NF-B-dependent signaling.
ABSTRACT The exact role of malignant ascites in the development of intraperitoneal ovarian cancer metastases remains unclear. In this report we sought to establish if ascites can determine the efficiency of transmesothelial invasion of ovarian cancer cells, and, if so, whether the fluid generated by highly aggressive serous and undifferentiated tumors will promote the invasion more effectively than ascites from less aggressive clear cell and endometrioid cancers. The study showed that the invasion of ovarian cancer cells (SKOV-3 and primary cancer cells) across monolayered peritoneal mesothelial cells was elevated upon mesothelial cell exposure to fluid produced by serous and undifferentiated cancers, as compared with cells subjected to ascites from clear cell and endometrioid tumors. This effect coincided with decreased mesothelial expression of junctional proteins: connexin 43, E-cadherin, occludin, and desmoglein. Moreover, it was accompanied by transforming growth factor 1-dependent 2
overproduction of reactive oxygen species by these cells. The activity of ascites from serous and undifferentiated tumors was mediated by p38 mitogen-activated protein kinase and nuclear factor B. When the mesothelial cells were protected against oxidative stress, both deterioration of junctional proteins and intensification of cancer cell invasion in response to ascites from serous and undifferentiated tumors were effectively prevented. In conclusion, our findings indicate that the high aggressiveness of some histotypes of ovarian cancer may be related to the ability of malignant ascites generated by these cells to oxidative stressdependent impairment of mesothelial cell integrity and the resulting increase in their transmesothelial invasion.
Key words: intercellular junctions; invasion; malignant ascites; ovarian cancer; peritoneal mesothelium
1. Introduction Epithelial ovarian cancer (EOC) is a heterogenous disease with respect to its origin, genetics, and aggressiveness, and hence it is categorized into type I and type II tumors. Type I is represented by low-grade serous, endometrioid, mucinous, malignant Brenner and clear cell cancers which originate from the ovaries, display high genetic stability, and develop slowly. Type II includes high-grade serous, carcinosarcoma, and undifferentiated cancer, it comes from outside the ovaries and is characterized by a high progression rate and mortality (Shih and Kurman, 2004). As per the genetic background, type I cancers bear mutations in the KRAS, BRAF, PTEN, and PI3K genes, while type II tumors have mutated p53 and mutated or dysfunctional BRCA1/2 genes (Romero and Bast, Jr., 2012). At the same time, although clinical and genetic differences between type I and type II tumors are well defined, cellular
3
pathomechanisms responsible for the high aggressiveness of the latter type remain to be explored. In this study we verified the hypothesis that the diverse aggressiveness of ovarian cancer histotypes may be associated with the activity of malignant ascites, which is a pathological fluid that accumulates in the peritoneum in a significant group of patients (Cvetkovic, 2003). In particular, we examined whether the transmesothelial invasion of cancer cells, which is one of the most critical elements of the formation of intraperitoneal metastasis (Steinkamp et al., 2013), proceeding in the presence of fluid from serous and undifferentiated ovarian cancers (type II tumors) would be more intense than ascites from clear cell and endometrioid cancers (type I tumors). Mechanistically, we addressed the ascites’ effect on the expression of various junctional proteins that provide the integrity of the peritoneal mesothelium and thus restrict the efficiency at which cancer cells move towards the tissue stroma (Defamie et al., 2014). Oxidative stress and related signaling pathways were analyzed to delineate factors regulating the ascites-dependent deterioration of intercellular junctions and those contributing to an increased transmesothelial invasion of ovarian cancer cells.
2. Materials and methods 2.1. Materials Unless otherwise stated, all chemicals and plastics were from Sigma (St. Louis, MO, USA). MG132 (the inhibitor of NF-B), API-1 (the inhibitor of AKT), and SP600125 (the inhibitor of JNK) were purchased from Tocris Bioscience (Ellisville, MO, USA), whereas SB202190 (the inhibitor of p38 MAPK) was from Cell Signaling Technology (Danvers, MA, USA). Exogenous, recombinant human transforming growth factor β1 (TGF-β1) was purchased from R&D Systems (Abingdon, UK).
4
2.2. Cell cultures Human peritoneal mesothelial cells (HPMCs) were isolated from pieces of omentum obtained from 8 patients (28-32 years old) undergoing elective abdominal surgery (institutional consent number 187/14), as described in detail elsewhere (Ksiazek, 2013). The reasons for the surgery included aortic aneurysm (4), hernia (3), and bowel obstruction (1). The cultures were maintained at 37 °C in a humidified atmosphere of 95% air and 5% CO2. Cells were identified as pure mesothelial by their typical cobblestone appearance at confluency and uniform positive staining for cytokeratins and HBME-1 antigen. Primary cultures of HPMCs obtained during the 1st passage (corresponding to ~5% of their replicative lifespan) without any contamination with stromal cells were used in the experiments. Ovarian cancer cells, SKOV-3, were purchased from the ECCC (Porton Down, UK) and propagated in RPMI 1640 medium with L-glutamine (2 mmol/L), penicillin (100 U/ml), streptomycin (100 g/ml), and 10% FBS. Primary epithelial ovarian cancer cells (EOCs) were isolated from a tumor excised during cytoreductive surgery from a patient with serous ovarian carcinoma (stage III according to FIGO). Briefly, the tumor was divided with a scalpel into ten pieces of equal weight and then placed in a solution of 0.05% trypsin and 0.02% EDTA for 20 min at 37 °C with gentle shaking. After resuspension in RPMI1640 containing 20% FBS, the cells were probed with an antibody directed against the epithelial-related antigen (MOC-31; Abcam, Cambridge, UK) to confirm their cancerous nature. Finally, ovarian cancer cells were cultured in RPMI 1640 supplemented with L-glutamine (2 mM) and 20% FBS.
2.3. Malignant and benign ascites Malignant ascites were obtained at the time of cytoreductive surgery from patients with high-grade serous, undifferentiated, clear cell and endometrioid ovarian carcinoma at stage
5
IIIC and IV (n=8 per group). The histopathology, grade, and stage of the tumors were assigned in accordance with the criteria of the International Federation of Gynecology and Obstetrics. Patients received no chemotherapy prior to surgery. Control, benign fluids were obtained from 8 age-matched patients undergoing abdominal surgery due to the presence of non-cancerous lesions - cystadenoma mucinosum multiloculare. Upon collection in sterile conditions, the fluids were centrifuged at 2500 rpm for 5 min and then cell-free supernatants were stored in aliquots at -20 C until required. The study was approved by an institutional ethics committee (consent number 543/14).
2.4. Cancer cell invasion assay Analysis of cancer cell invasion was performed using Cultrex 96 Well BME Cell Invasion Assay (Trevigen Inc. Gaithersburg, MD, USA) as per manufacturer’s instructions. In brief, HPMCs (1x105 cells per well) was seeded on the basement membrane extract (BME) to form a monolayer. Afterwards, the cells were subjected to 10% malignant and benign ascites (50 µl per well) for 72 h. After the incubation, ovarian cancer cells (1x104 cells per well) were placed into the upper chamber of the system, that is on top of HPMCs. The cancer cells invaded through the HPMCs lying on the BME towards a chemotactic gradient generated by 1% FBS. The intensity of fluorescence emitted by the cancer cells was recorded using a SynergyTM 2 spectrofluorometer (BioTek Instruments, Winooski, VT, USA) at 435 nm excitation and 535 nm emission wavelengths, respectively. In some experiments, cancer cell invasion was examined in the presence of HPMCs subjected to malignant and benign ascites (10%, for 72 h) upon their pre-incubation (for 4 h) with the ROS spin-trap scavenger, N-tertbutyl-alpha-phenylnitrone (PBN, Sigma; 800 µM).
2.5. Analysis of intercellular junctions
6
In order to examine expression of junctional proteins, HPMCs were seeded into semitransparent 96-well plates (1x105 cells per well), and then they were allowed to form a monolayer for next 24 h. Upon reaching the confluency, the plates were carefully washed to remove all non-adherent cells and debris, and then the cells were subjected to 10% malignant and benign ascites (50 µl per well) for 72 h. After the incubation, the cells were fixed in paraformaldehyde, washed and treated with antibodies against connexin 43 (cat # ab11370, Abcam, 1:100, overnight), E-cadherin (cat # ab15148, Abcam, 1:100, overnight), occludin (cat # NBP1-87402, Novus Biologicals, Littleton, CO, USA, 1:100, overnight), and desmoglein (cat # ab12077, Abcam, 1:10, overnight). Then the cells were extensively washed with phosphate-buffered saline (PBS) and incubated with DyLight 488 antibody (cat # ab96899, Abcam; 1:500, 1h) to quantify connexin 43, E-cadherin and occludin, and with Alexa Fluor 488 antibody (cat # ab150113, Abcam, 1:500, 1h) to quantify desmoglein. Finally, the cells were washed three times with PBS and fluorescence emitted was recorded using a SynergyTM 2 spectrofluorometer (BioTek Instruments, Winooski, VT, USA). Representative pictures of immunoreactions were taken using an Axio Vert.A1 microscope (Carl-Zeiss, Jena, Germany). In some experiments, the expression of junctional proteins was measured in HPMCs exposed to both malignant and benign ascites (10%, for 72 h) upon their pre-incubation (12 h) with MG132 (10 µM) and SB202190 (10 µM) for 12 h. In other experiments the proteins were analyzed upon HPMC pre-treatment (for 4 h) with PBN (800 µM). Additionally, the expression of the junctional proteins was quantified in cell homogenates using specific, colorimetric ELISA-based kits purchased from Abbexa Ltd (Cambridge, UK), as per manufacturer’s instructions. In order to collect appropriate amounts of cellular protein, the experiments were performed with HPMCs seeded in 24-well plates (5x105 cells per well) and allowed to form the monolayer for 24 h. Upon the exposure to 10% malignant and benign
7
ascites (250 µl per well) for 72 h or the exposure to MG132 and SB202190 for 12 h, the cells were homogenized by sonication. The homogenates were centrifuged at 5000 × g for 5 min and the supernatants collected were stored at -80 C until assayed.
2.6. Production of reactive oxygen species (ROS) In order to examine the production of ROS, HPMCs were seeded into semi-transparent 96well plates (1x105 cells per well), and then they were allowed to form a monolayer for next 24 h. Upon reaching the confluency, the plates were carefully washed to remove all non-adherent cells and debris, and then the cells were subjected to 10% malignant and benign ascites (50 µl per well) for 72 h. The ROS release by HPMCs was assessed using 2’,7’dichlorodihydrofluorescein diacetate (H2DCFDA), essentially as described in (MikulaPietrasik et al., 2012). In some experiments, ROS production was monitored in cells exposed to exogenous, recombinant TGF-β1 used at concentrations of 100-400 pg/ml for 72 h. In another group of experiments, ROS were measured in HPMCs subjected to malignant and benign ascites (10%, for 72 h) upon their pre-incubation with MG132 (10 µM), API-1 (50 µM), SP600125 (10 µM), and SB202190 (10 µM) for 12 h.
2.7. Statistics Statistical analysis was conducted with GraphPad Prism 5.00 software (GraphPad Software, San Diego, USA). The means were compared using repeated measures analysis of variance (ANOVA) with the Newman-Keuls test as a post-hoc test. When appropriate, the Wilcoxon matched pairs test was used. The results were expressed as means SD. Differences with a P value < 0.05 were considered to be statistically significant.
3. Results and Discussion 8
There is a growing body of evidence that malignant ascites may play some role in the development of ovarian tumors (Cvetkovic, 2003). It has been found that ascites may promote the development of tumor metastases by suppressing peritoneal inflammation (SimpsonAbelson et al., 2013) and that the ascites’ biochemical composition may, at least theoretically, facilitate angiogenesis and proliferation of cancer cells (Matte et al., 2012). Interestingly, a recent report revealed that the proinflammatory characteristics of the fluid generated by undifferentiated tumors, known to be the most lethal ovarian cancer histotype (Silva et al., 1991), may underlie the high proliferative and migratory potential of the cancer cells (MikulaPietrasik et al., 2016c). In general, however, the mechanistic aspects of the intraperitoneal spread of ovarian tumors associated with the presence and activity of malignant ascites, including the background of differences in the aggressiveness of type I and type II tumors, remain elusive. In this report we showed that malignant ascites determines the efficiency at which both established (SKOV-3) and primary EOC cells invade across the peritoneal mesothelial cells (HPMCs). Taking into account that a single layer of HPMCs forms a physical barrier on the way of the cancer cells towards the peritoneal stroma (Iwanicki et al., 2011), our observation sheds new light on the pathomechanism controlling the development of intraperitoneal ovarian tumors. Significantly, as is shown in Fig. 1, the effectiveness of cancer cell movement was particularly high when the cells invaded through the mesothelium subjected to ascites from serous and undifferentiated tumors (vs. clear cell and endometrioid cancer, and vs. benign fluid from non-cancerous patients). Among the two cancer histotypes displaying high aggressiveness, ascites from patients with undifferentiated tumors were more pro-invasive. This effect resembles that when malignant ascites generated by undifferentiated tumors promoted migration of A2780, OVCAR-3, and SKOV-3 ovarian cancer cells more efficiently
9
than ascites from serous carcinoma (Mikula-Pietrasik et al., 2016c). In this context it should be emphasized that the high ability of cancer cells to invade across monolayered HPMCs as revealed in the current study may to some extent be overlapped by increased cancer cell migration which proceeds in an HPMCs-independent manner. In fact, ascites generated by type II tumors, in particular by the undifferentiated histotype, are rich in several chemotactic agents, e.g. CCL2, CXCL1, CXCL5, CXCL8, and CXCL12 (Mikula-Pietrasik et al., 2016c). In our opinion, both increased invasion and migration should be considered in order to visualize the whole process of transmesothelial cancer cell movement. It is worth noting that the findings available in the literature regarding the role of ascitic fluid in cancer cell invasion, in particular the findings provided by Puiffe et al., are inconclusive (Puiffe et al., 2007). The authors of the study cited above show that malignant ascites differ in their capacity to modulate cancer cell invasion, i.e. some of them exert the inhibitory effect while some stimulate the process. We believe that this inconsistent picture may result from at least three independent variables: i) referring the results to 5% serum instead of benign ascites, ii) the use of mixed, heterologous fluids instead of an analysis of fluids derived from a given histotype, and iii) the measurement of cell invasion across the Matrigel instead of an analysis of cell movement through the mesothelial cells lying on the Matrigel or its equivalent. Because our experimental approach includes all of the aspects as postulated above, we are convinced that the results presented here conclusively indicate that the malignant ascites possesses clearly proinvasive capabilities. The integrity of the HPMCs’ monolayer largely depends on the expression and functionality of various intercellular junctions and is considered a limiting factor for cancer cell invasion (Kenny et al., 2007). Having this in mind, we examined the effect of malignant ascites on four junctional proteins which are representative of the peritoneal mesothelium, i.e. connexin 43 (gap junction), E-cadherin (adherens junction), occludin (tight junction), and
10
desmoglein (desmosome) (Simionescu and Simionescu, 1977). As shown in Figs. 2 and 3, and in Tab. 1, HPMCs exposed to malignant ascites from type I tumors (clear cell and endometrioid) displayed unaltered expression of E-cadherin, occludin, and desmoglein, and slightly decreased expression of connexin 43. In contrast, HPMCs subjected to malignant ascites from type II tumors (serous and undifferentiated) were characterized by significantly decreased expression of all four tested intercellular junctions. As per connexin 43, its decline was much more pronounced as compared with the effects exerted by ascites from clear cell and endometrioid cancers. Significantly, in the case of E-cadherin, occludin, and desmoglein but not connexin 43 (where the situation was opposite), the effects generated by ascites from undifferentiated tumors were again considerably stronger than those resulting from the activity of their serous counterparts. These results suggest a scenario in which the increased potential of ovarian cancer cells to invade through HPMCs treated with malignant ascites from aggressive tumors may be associated with increased “permeability” of the mesothelium due to decreased expression of junctional proteins. This possibility is in accordance with results of Giverso et al., who combined experiments in vitro with 3D multiscale mathematical analysis and proved that an efficient invasion of ovarian cancer cells requires the loosening of junctional proteins in the peritoneal mesothelium (Giverso et al., 2010). Having established that malignant ascites-related deterioration of the junctional proteins leading to a weakening of the structural and functional connection between adjacent HPMCs may be the reason for the intensified transmesothelial invasion of ovarian cancer cells, we focused on a mechanistic background of this phenomenon. Because decreased expression of intercellular junctions, e.g. connexin 43, has been found in senescent cells (Statuto et al., 2002), and the intensification of HPMCs’ senescence upon their exposure to malignant ascites was described recently (Mikula-Pietrasik et al., 2016a), we paid special attention to the role of
11
oxidative stress, which is the key determinant of several senescence-related abnormalities in HPMCs (Ksiazek, 2013). Our study revealed that HPMCs subjected to malignant ascites generate a higher amount of ROS than cells exposed to benign fluid. Moreover, the magnitude of oxidative stress in cells treated with ascites from type II tumors was higher as compared with their counterparts exposed to fluid generated by type I tumors. Here also the greatest ROS production was observed in HPMCs stimulated with ascites obtained from patients with undifferentiated ovarian cancer (Fig. 4A). The prooxidative impact of malignant ascites on HPMCs is an entirely new finding; however, its seems to adhere to earlier results obtained by Yang et al., who found that the ascites modulates the antioxidative potential of ovarian cancer cells, thus supporting their ability to adhere and invade (Yang et al., 2005). Intervention studies using a spin-trap ROS scavenger, PBN, showed that either increased cancer cell invasion across HPMCs (Tab. 2) or decreased expression of junctional proteins (Tab. 3) elicited by malignant ascites (particularly those generated by undifferentiated and serous tumors) can be effectively prevented when the HPMCs are protected against oxidative stress. These results, pointing to negative regulation of mesothelial intercellular junctions by ROS, are in line with the observations of another group which revealed the deterioration of connexin 43 in cells exposed to various prooxidative stimuli (Berthoud and Beyer, 2009). They are also in agreement with a report showing a disruption of epithelial tight junctions upon the activity of calcium-mediated oxidative stress (Gangwar et al., 2017). The search for a soluble mediator of increased oxidative stress in HPMCs treated with malignant ascites directed our attention towards transforming growth β1 (TGF-β1), a cytokine whose capacity to elevate ROS release has already been well documented for various cell types, including HPMCs (Mikula-Pietrasik et al., 2013). Interestingly, our previous study showed that the concentration of TGF-β1 in the malignant ascites from undifferentiated
12
ovarian tumors was higher when compared with fluids generated by clear cell and endometrioid cancers, and comparable with that characterizing the ascites from patients with serous carcinoma (Mikula-Pietrasik et al., 2016c). The role of TGF-β1 as an ascites-derived, prooxidative agent was confirmed experimentally when we used an exogenous recombinant form of this cytokine and applied it on monolayered HPMCs at concentrations corresponding to its level in the four tested types of malignant ascites. The production of ROS by HPMCs treated with TGF-β1 appeared to increase in a dose-dependent manner (Fig. 4B) under such a regimen. In order to identify the signaling pathways underlying the oxidative stress-related decrease in the expression of junctional proteins in HPMCs exposed to malignant ascites from serous and undifferentiated tumors, the cells were pre-incubated with chemical inhibitors of four pathways known to be associated with the activity of either TGF-β1 or ROS, i.e. AKT, JNK, NF-B, and p38 MAPK (Krstic et al., 2015). Afterwards, the magnitude of oxidative stress and the expression of the intercellular junctions in HPMCs were re-investigated. The results shown in Fig. 5 indicate that the production of ROS in HPMCs subjected to malignant ascites can be inhibited upon neutralization of NF-B and p38 MAPK but not AKT and JNK. Inhibition of the same signaling molecules resulted in restoration of the expression of all of the tested junctional proteins to values characterizing cells treated with benign ascites (Fig. 6 and Tab. 4). The contribution of NF-B and p38 MAPK in an ascites-dependent deterioration of junctional proteins in HPMCs may suggest that the functional alterations in those cells that promote cancer cell invasion may contain a component associated with their senescence. This statement is based on the fact that both NF-B and p38 MAPK have already been recognized to mediate some senescence-related changes in HPMCs (e.g. the development of the secretory phenotype (Mikula-Pietrasik et al., 2015)). In addition, the neutralization of p38 MAPK has been found to rejuvenate the HPMCs and to postpone their cancer-promoting capabilities 13
(Mikula-Pietrasik et al., 2016b). Taking into account the fact that malignant ascites from serous ovarian cancer have already been found to stimulate senescence in HPMCs (MikulaPietrasik et al., 2016a), the current study extends this effect to ascites produced by undifferentiated tumors whose activity seems to be even more pronounced.
4. Conclusions In conclusion, the results we provide in this paper indicate that the severe aggressiveness of serous and undifferentiated ovarian cancers may be related, at least to some extent, to the activity of the malignant ascites that is generated by these cells. In particular, it may be associated with their ability to destroy the integrity of HPMCs, thus leading to improved transmesothelial invasion of the cancer cells. A significant role in all of these processes is played by oxidative stress. Taking this into account, there is an urgent need to examine whether intraperitoneal administration of antioxidants, e.g. catalase (Nishikawa et al., 2009), could prevent or at least restrict the malignant ascites-dependent progression of highly aggressive ovarian tumors.
Acknowledgments The study was supported by a grant from the National Science Centre, Poland (registration number 2014/15/B/NZ3/00421).
Conflict of interest The authors declare no conflict of interest.
References
14
Berthoud, V.M., Beyer, E.C., 2009. Oxidative stress, lens gap junctions, and cataracts. Antioxid. Redox. Signal. 11, 339-353. Cvetkovic, D., 2003. Early events in ovarian oncogenesis. Reprod. Biol. Endocrinol. 1, 68. Defamie, N., Chepied, A., Mesnil, M., 2014. Connexins, gap junctions and tissue invasion. FEBS Lett. 588, 1331-1338. Gangwar, R., Meena, A., Shukla, P., Nagaraja, A., Dorniak, P., Pallikuth, S., Waters, C. Sood, A., Rao, R., 2017. Calcium-Mediated Oxidative Stress: a Common Mechanism in Tight Junction Disruption by Different Types of Cellular Stress. Biochem. J. 474, 731-749. Giverso, C., Scianna, M., Preziosi, L., Lo Buono, N., Funaro, A., 2010. Individual Cell-Based Model for In-Vitro Mesothelial Invasion of Ovarian Cancer. Math. Model. Nat. Phenom 5, 1-21. Iwanicki, M.P., Davidowitz, R.A., Ng, M.R., Besser, A., Muranen, T., Merritt, M., Danuser, G., Ince, T.A., Brugge, J.S., 2011. Ovarian cancer spheroids use myosin-generated force to clear the mesothelium. Cancer Discov. 1, 144-157. Kenny, H.A., Krausz, T., Yamada, S.D., Lengyel, E., 2007. Use of a novel 3D culture model to elucidate the role of mesothelial cells, fibroblasts and extra-cellular matrices on adhesion and invasion of ovarian cancer cells to the omentum. Int. J. Cancer 121, 14631472. Krstic, J., Trivanovic, D., Mojsilovic, S., Santibanez, J.F., 2015. Transforming Growth Factor-Beta and Oxidative Stress Interplay: Implications in Tumorigenesis and Cancer Progression. Oxid. Med Cell Longev. 2015, 654594. Ksiazek, K. 2013. Mesothelial cell: A multifaceted model of aging. Ageing Res Rev. 12, 595604. Matte, I., Lane, D., Laplante, C., Rancourt, C., Piche, A., 2012. Profiling of cytokines in human epithelial ovarian cancer ascites. Am. J Cancer Res 2, 566-580.
15
Mikula-Pietrasik, J., Kuczmarska, A., Kucinska, M., Murias, M., Wierzchowski, M., Winckiewicz, M., Staniszewski, R., Breborowicz, A., Ksiazek, K., 2012. Resveratrol and its synthetic derivatives exert opposite effects on mesothelial cell-dependent angiogenesis via modulating secretion of VEGF and IL-8/CXCL8. Angiogenesis. 15, 361-376. Mikula-Pietrasik, J., Sosinska, P., Janus, J., Rubis, B., Brewinska-Olchowik, M., Piwocka, K., Ksiazek, K., 2013. Bystander senescence in human peritoneal mesothelium and fibroblasts is related to thrombospondin-1-dependent activation of transforming growth factor-beta1. Int. J. Biochem. Cell Biol. 45, 2087-2096. Mikula-Pietrasik, J., Sosinska, P., Maksin, K., Kucinska, M., Piotrowska, H., Murias, M. Wozniak, A., Szpurek, D., Ksiazek, K., 2015. Colorectal cancer-promoting activity of the senescent peritoneal mesothelium. Oncotarget. 6, 29178-29195. Mikula-Pietrasik, J., Uruski, P., Matuszkiewicz, K., Szubert, S., Moszynski, R., Szpurek, D., Sajdak, S., Tykarski, A., Ksiazek, K., 2016a. Ovarian cancer-derived ascitic fluids induce a senescence-dependent pro-cancerogenic phenotype in normal peritoneal mesothelial cells. Cell Oncol (Dordr. ) 39, 473-481. Mikula-Pietrasik, J., Uruski, P., Sosinska, P., Maksin, K., Piotrowska-Kempisty, H., Kucinska, M., Murias, M., Szubert, S., Wozniak, A., Szpurek, D., Sajdak, S., Piwocka, K., Tykarski, A., Ksiazek, K., 2016b. Senescent peritoneal mesothelium creates a niche for ovarian cancer metastases. Cell Death. Dis. 7, e2565. Mikula-Pietrasik, J., Uruski, P., Szubert, S., Moszynski, R., Szpurek, D., Sajdak, S., Tykarski, A., Ksiazek, K., 2016c. Biochemical composition of malignant ascites determines high aggressiveness of undifferentiated ovarian tumors. Med Oncol 33, 94. Nishikawa, M., Hashida, M., Takakura, Y., 2009. Catalase delivery for inhibiting ROSmediated tissue injury and tumor metastasis. Adv Drug Deliv. Rev. 61, 319-326.
16
Puiffe, M.L., Le, P.C., Filali-Mouhim, A., Zietarska, M., Ouellet, V., Tonin, P.N., Chevrette, M., Provencher, D.M., Mes-Masson, A.M., 2007. Characterization of ovarian cancer ascites on cell invasion, proliferation, spheroid formation, and gene expression in an in vitro model of epithelial ovarian cancer. Neoplasia. 9, 820-829. Romero, I., Bast Jr, R.C., 2012. Minireview: human ovarian cancer: biology, current management, and paths to personalizing therapy. Endocrinology 153, 1593-1602. Shih, I., Kurman, R.J., 2004. Ovarian tumorigenesis: a proposed model based on morphological and molecular genetic analysis. Am. J Pathol. 164, 1511-1518. Silva, E.G., Tornos, C., Bailey, M.A., Morris, M., 1991. Undifferentiated carcinoma of the ovary. Arch Pathol. Lab Med 115, 377-381. Simionescu, M., Simionescu, N., 1977. Organization of cell junctions in the peritoneal mesothelium. J Cell Biol. 74, 98-110. Simpson-Abelson, M.R., Loyall, J.L., Lehman, H.K., Barnas, J.L., Minderman, H., O'Loughlin, K.L., Wallace, P.K., George, T.C., Peng, P., Kelleher Jr, R.J., Odunsi, K., Bankert, R.B., 2013. Human ovarian tumor ascites fluids rapidly and reversibly inhibit T cell receptor-induced NF-kappaB and NFAT signaling in tumor-associated T cells. Cancer Immun. 13, 14. Statuto, M., Bianchi, C., Perego, R., Del, M.U., 2002. Drop of connexin 43 in replicative senescence of human fibroblasts HEL-299 as a possible biomarker of senescence. Exp Gerontol. 37, 1113-1120. Steinkamp, M.P., Winner, K.K., Davies, S., Muller, C., Zhang, Y., Hoffman, R.M., Shirinifard, A., Moses, M., Jiang, Y., Wilson, B.S., 2013. Ovarian tumor attachment, invasion, and vascularization reflect unique microenvironments in the peritoneum: insights from xenograft and mathematical models. Front Oncol. 3, 97.
17
Yang, W., Toffa, S.E., Lohn, J.W., Seifalian, A.M., Winslet, M.C., 2005. Malignant ascites increases the antioxidant ability of human ovarian (SKOV-3) and gastric adenocarcinoma (KATO-III) cells. Gynecol. Oncol 96, 430-438.
Tables
Table 1 The quantification of junctional proteins in HPMCs exposed to benign and malignant ascites. Concentration (ng/ml) Connexin 43 E-cadherin Occludin Desmoglein
Benign ascites and malignant ascites generated by ovarian cancer cells Benign Clear cell Endometrioid Serous Undifferentiated 2.4±0.3 6.3±0.2 1.6±0.1 9.6±0.1
1.9±0.1* 6.5±0.1 1.7±0.2 10.3±0.5
1.8±0.2* 6.2±0.3 1.4±0.2 10.6±0.4
0.8±0.2*# 6.2±0.3 0.8±0.3*# 6.3±0.2*#
1.3±0.3*# 2.7±0.1*# 0.5±0.2*# 2.1±0.3*#
* P<0.05 vs. HPMCs subjected to benign ascites, # P<0.05 vs. CC, P<0.05 vs. Endo. The experiments were performed with HPMCs established from 8 different donors. Ascites (malignant and benign) were obtained from 8 patients per group.
18
Table 2 The effect of HPMC protection against oxidative stress on the invasion of ovarian cancer cells across HPMCs subjected to malignant ascites.
Malignant ascites generated by ovarian cancer cells Clear cell Endometrioid Serous Undifferentiated -PBN +PBN -PBN +PBN -PBN +PBN -PBN +PBN SKOV-3 Primary EOC
127±8* 145±3*
104±2# 111±6#
125±4* 136±9*
103±1# 109±4#
158±9* 192±6*
104±2# 122±6#
200±11* 217±6*
111±3# 104±9#
The results are expressed as the percentage of invasion of cancer cells through HPMCs exposed to benign fluid from non-cancerous patients, treated as 100%. * P<0.05 vs. HPMCs exposed to benign fluid; # P<0.05 vs. HPMCs not exposed to PBN. The experiments were performed with HPMCs established from 8 different donors. Cancer cells were used in duplicates. Ascites (malignant and benign) were obtained from 8 patients per group.
19
Table 3 The effect of HPMC protection against oxidative stress on the expression of junctional proteins in cells treated with malignant ascites. Malignant ascites generated by ovarian cancer cells Clear cell Endometrioid Serous Undifferentiated -PBN +PBN -PBN +PBN -PBN +PBN -PBN +PBN Connexin 43 E-cadherin Occludin Desmoglein
82±7* 102±2 104±3 102±7
99±1# 103±2 102±1 107±2
85±7* 101±4 93±1* 105±2
101±3# 101±2 99±2# 108±8
42±4* 101±5 85±7* 78±8*
92±6# 99±3 102±1# 106±6#
72±7* 68±4* 61±4* 44±9*
101±3# 101±2# 106±2# 102±6#
The results are expressed as the percentage of the expression of junctional proteins in HPMCs subjected to benign fluid from non-cancerous patients, treated as 100%. * P<0.05 vs. HPMCs exposed to benign fluid; # P<0.05 vs. HPMCs not exposed to PBN. The experiments were performed with HPMCs established from 8 different donors. Ascites (malignant and benign) were obtained from 8 patients per group.
20
Table 4
The quantification of junctional proteins in HPMCs exposed to malignant ascites from serous and undifferentiated ovarian tumors upon the inhibition of NF-B and p38 MAPK
Concentration (ng/ml) Connexin 43 E-cadherin Occludin Desmoglein Concentration (ng/ml) Connexin 43 E-cadherin Occludin Desmoglein
Benign ascites and malignant ascites from serous ovarian cancer Serous + Serous + Benign Serous MG132 SB202190 2.4±0.2 6.5±0.1 1.5±0.2 9.1±0.4
0.9±0.2* 6.3±0.6 1.0±0.1* 7.2±0.2*
2.5±0.1# 6.1±0.4 1.7±0.2# 9.6±0.2#
2.3±0.3# 6.5±0.2 1.7±0.1# 9.9±0.3#
Benign ascites and malignant ascites from undifferentiated ovarian cancer Undiff + Undiff + Benign Undiff MG132 SB202190 2.3±0.2 6.6±0.4 1.9±0.3 9.4±0.3
1.1±0.2* 4.1±0.2* 1.3±0.2* 3.4±0.2*
2.2±0.4# 7.4±0.2# 2.1±0.3# 10.1±0.7#
2.6±0.2# 7.1±0.1# 2.0±0.1# 11.1±0.6#
* P<0.05 vs. HPMCs subjected to benign ascites, # P<0.05 vs. HPMCs subjected to malignant ascites without their pre-incubation with signaling pathway inhibitors. The experiments were 21
performed with HPMCs established from 8 different donors and with ascites (malignant and benign) from 8 patients per group.
Figure legends
Fig. 1. Effect of malignant ascites from patients with clear cell (CC), endometrioid (Endo), serous (Ser), and undifferentiated (Undiff) ovarian tumors on transmesothelial invasion of SKOV-3 cells (A) and primary epithelial ovarian cancer cells (B). The results are presented as scatter dot plots in order to show the variability of ascites obtained from different patients. * P<0.05 vs. HPMCs subjected to benign ascites, # P<0.05 vs. CC, P<0.05 vs. Endo. The experiments were performed with HPMCs established from 8 different donors. The cancer cells were used in duplicates. Ascites (malignant and benign) were obtained from 8 patients per group. BA – benign ascites from non-cancerous patients; RFU – relative fluorescence units.
Fig. 2. Effect of malignant ascites from patients with clear cell (CC), endometrioid (Endo), serous (Ser), and undifferentiated (Undiff) ovarian tumors on the expression of connexin 43 (A), E-cadherin (B), occludin (C), and desmoglein (D) in HPMCs. * P<0.05 vs. HPMCs subjected to benign ascites, # P<0.05 vs. CC, P<0.05 vs. Endo. The experiments were performed with HPMCs established from 8 different donors. Ascites (malignant and 22
benign) were obtained from 8 patients per group. BA – benign ascites from non-cancerous patients; RFU – relative fluorescence units.
Fig. 3. Representative images of immunofluorescence against junctional proteins in HPMCs exposed to malignant ascites from patients with clear cell (CC), endometrioid (Endo), serous (Ser), and undifferentiated (Undiff) ovarian tumors and to benign ascites (BA) from non-cancerous patients. The stainings were performed with HPMCs established from 8 different donors and with ascites (malignant and benign) from 8 patients per group. Magnification x 200. Fig. 4. Effect of malignant ascites (A) from patients with clear cell (CC), endometrioid (Endo), serous (Ser), and undifferentiated (Undiff) ovarian tumors and the exogenous, recombinant form of human TGF-β1 (B) on the production of ROS in HPMCs. * P<0.05 vs. HPMCs subjected to benign ascites, # P<0.05 vs. CC, P<0.05 vs. Endo, ** P<0.05 vs. cells not treated with rhTGF-β1. The experiments were performed with HPMCs established from 8 different donors and with ascites (malignant and benign) from 8 patients per group. BA – benign ascites from non-cancerous patients; RFU – relative fluorescence units.
Fig. 5. Analysis of signaling pathways underlying the activity of malignant ascites from type II ovarian tumors. Effect of the inhibition of AKT, JNK, NF-B, and p38 MAPK on the production of ROS by HPMCs exposed to malignant ascites from serous (A) and undifferentiated cancer (B). The experiments were performed with HPMCs established from 8 different donors and with ascites (malignant and benign) from 8 patients per group. * P<0.05 vs. HPMCs subjected to benign ascites, # P<0.05 vs. HPMCs subjected to malignant ascites without their pre-incubation with signaling pathway inhibitors. MA – malignant ascites; BA – benign ascites; RFU – relative fluorescence units.
23
Fig. 6. Effect of NF-B and p38 MAPK inhibition on the expression of connexin 43, Ecadherin, occludin, and desmoglein in HPMCs exposed to malignant ascites from serous (A-D) and undifferentiated (E-H) ovarian tumors. The experiments were performed with HPMCs established from 8 different donors and with ascites (malignant and benign) from 8 patients per group. * P<0.05 vs. HPMCs subjected to benign ascites, # P<0.05 vs. HPMCs subjected to malignant ascites without their pre-incubation with signaling pathway inhibitors. MA – malignant ascites; BA – benign ascites; RFU – relative fluorescence units.
24