Researchers have used a live cell imaging technique to demonstrate how clusters of ovarian cancer cells effectively snow-plough their way through the mesothelial cell layer covering peritoneal organs in order to invade new tissues. The studies suggest that when ovarian cancer spheroids settle on a mesothelial monolayer they forcibly push the mesothelial cells out from under them to facilitate cancer cell attachment to the connective tissue beneath.
The research was conducted by scientists at Harvard Medical School and Brigham and Women’s Hospital. The team suggests that acquisition of contractile properties by ovarian tumor cells could represent a step in the process by which ovarian cancers become more malignant. Results from the in vitro time-lapse microscopy studies are described in Cell Discovery in a paper titled "Ovarian Cancer Spheroids Use Myosin-Generated Force to Clear the Mesothelium."
To date exactly how tumor cells effect mesothelial clearance has remained unknown. To understand this process further, Joan Brugge, M.D., and colleagues used a live, image-based in vitro model to monitor in real time the interactions between tumor spheroids and mesothelial cells expressing green-fluorescent protein (GFP).
Initial data confirmed that as the spheroid spreads on the mesothelial monolayer, mesothelial cells were displaced from the area directly underneath the spreading spheroid, a phenomenon they termed mesothelial clearance. This occurred whether the cells were grown on a glass surface or on a more physiological relevant substrate.
Closer observations further suggested that the cancer spheroids adhere to the mesothelial layer and induce localized de-adhesion of the mesothelial cells, allowing them to be pushed away from under the spheroid. Moreover, the mesothelial cells that were in direct contact with a tumor spheroid migrated further away from the cancer cells than those that weren’t directly in contact with the spheroid.
To investigate the mechanism responsible for pushing mesothelial cells away, Dr. Brugge and colleagues treated cancer cells with an shRNA and siRNA that blocked myosin IIA and myosin IIB heavy chains, respectively. Treated cancer cells were still able to spread over surfaces coated with fibronectin and collagen I, but they were unable to sustain mesothelial clearance.
Moreover, attenuating myosin II in the cancer cells had no effect on the adhesion of spheroids to a mesothelial cell monolayer. These results suggested that myosin II expression is dispensible for spheroid attachment but is needed for mesothelial clearance.
The ability of cells to exert force on the outside environment depends on linkage of the actin and myosin network to integrins through recruitment of talin I to adhesion sites. Further studies in ovarian cancer cells showed that blocking talin 1 but not talin 2 in the cancer spheroids significantly reduced mestoehlial clearance, even though, it didn’t affect initial adherence to the monolayer.
“These data indicate that talin I is required for tumor cell intercalation into the mesothelium and suggest that the linkage of integrins to the actomyosin network in ovarian spheroids contributes to mesothelial clearance,” the authors state.
Expression of the α5 integrin fibronectin receptor has previously been shown to correlate with the development of myosin-driven contractility and increased invasion of ovarian cancer cells. With this observation in mind the researchers then investigated whether α5β1 integrin-mediated activation of contractility contributes to ovarian spheroid-induced mesothelial clearance.
They first blocked the function of α5 integrin in ovarian cancer cells that express high levels of the protein. The result was significantly decreased spheroid-induced mesothelial clearance but not initial adherence of the spheroids to the mesothelial monolayer. Interestingly, blocking other adhesion receptors expressed by the spheroids, including CD44 and integrins α2 and αv, had no significant effects on spheroid-induced mesothelial clearance.
In parallel experiments the team overexpressed α5 integrin in ovarian cancer cells that would normally express only low levels of α5 integrin and were naturally unable to clear a mesothelial monolayer. Ectopic expression of α5 integrin by these cells increased the activation of myosin, promoted cell spreading, increased stress fibers and other cortical actin contractile structures, and increased mesothelial clearance.
Results thus far suggested that engagement of the fibronectin receptor α5β1 integrin is an important step in spheroid-induced mesothelial clearance. Indeed, while fibronectin fibrils and organized collagen fibers are both found on the dorsal surfaces of mesothelial cells, blocking the α2β1 integrin collagen receptor did not affect mesothelial clearance.
Focusing therefore on fibronectin, the researchers showed that cancer spheroids induced detachment of fibronectin fibrils from the mesothelial cells. When α5β1 was inhibited using a function-blocking antibody, fibronectin did not dissociate from the mesothelial cells, “suggesting that dissociation of fibronectin from the top of the mesothelial monolayer was dependent on functional α5β1 integrin expressed by the cancer spheroids.”
The collective data strongly suggests that α5β1 integrin, myosin, and talin are all required in ovarian cancer cells for mesothelial clearance and that α5β1 integrin-dependent binding of the ovarian cancer spheroids to fibronectin organized by the mesothelium is important for clearance processes, the authors state. “These findings would suggest that the traction force exerted on the substratum by the spreading spheroids contributes to mesothelial clearance.”
To investigate this notion further they used traction force microscopy to determine if modulation of α5, talin I, or myosin affected force generation in ovarian cancer cells. The combined results of these studies were consistent with a model in which talin I and myosin act downstream of α5 integrin to generate force as ovarian cancer cells interact with fibronectin matrix.
“These studies provide new insights into the mechanism whereby ovarian tumor spheroids induce mesothelial cell clearance,” the authors conclude. “Clearance-competent tumor spheroids were found to adhere to the dorsal surface of the mesothelial cells and initiate spreading. Protrusions from the spreading cells penetrated underneath the mesothelial cells, causing localized breakdown of the mesothelial cell matrix adhesions, and provoked migration of the cells.”
Lead author Marcin Iwanicki, Ph.D., a postdoc in Dr. Brugge’s laboratory, adds “The cancer cells act like bullies. Instead of relying on a sophisticated biochemical process to achieve their goal, they simply push mesothelial cells apart.”
In patients with advanced disease, ovarian tumor clusters predominantly implant into the mesothelial lining of peritoneal cavity-associated organs, and invasive tumors can cross the mesothelial layer and gain access to stroma beneath, the researchers note.
“These observations suggest that the mesothelium presents a functional barrier to the spread and progression of ovarian tumors. Hence, one would expect progression toward invasive disease to be associated with alterations that enable tumor cells to adhere to the mesothelium and break the mesothelial barrier by provoking mesothelial clearance. Our results may show that acquisition of contractile phenotype in ovarian tumor cells represents a step toward malignant progression.”