The senescent mesothelial matrix accentuates colonization by ovarian cancer cells

Ovarian cancer is amongst the most morbid of gynecological malignancies due to its diagnosis at an advanced stage, a transcoelomic mode of metastasis, and rapid transition to chemotherapeutic resistance. Like all other malignancies, the progression of ovarian cancer may be interpreted as an emergent outcome of the conflict between metastasizing cancer cells and the natural defense mounted by microenvironmental barriers to such migration. Here, we asked whether senescence in coelom-lining mesothelia, brought about by drug exposure, affects their interaction with disseminated ovarian cancer cells. We observed that cancer cells adhered faster on senescent human and murine mesothelial monolayers than on non-senescent controls. Time-lapse epifluorescence microscopy showed that mesothelial cells were cleared by a host of cancer cells that surrounded the former, even under sub-confluent conditions. A multiscale computational model predicted that such colocalized mesothelial clearance under sub-confluence requires greater adhesion between cancer cells and senescent mesothelia. Consistent with the prediction, we observed that senescent mesothelia expressed an extracellular matrix with higher levels of fibronectin, laminins and hyaluronan than non-senescent controls. On senescent matrix, cancer cells adhered more efficiently, spread better, and moved faster and persistently, aiding the spread of cancer. Inhibition assays using RGD cyclopeptides suggested the adhesion was predominantly contributed by fibronectin and laminin. These findings led us to propose that the senescence-associated matrisomal phenotype of peritoneal barriers enhances the colonization of invading ovarian cancer cells contributing to the metastatic burden associated with the disease. Supplementary Information The online version contains supplementary material available at 10.1007/s00018-023-05017-x.

The confirmation of senescent cells acquired from ex-vivo murine peritoneal treated with 100 nM of doxorubicin.(A) The representative snapshots show stained DNA (using DAPI) and F-actin (using phalloidin) of the untreated and treated samples for morphological examination of the non-senescent and senescent cells extracted from the murine peritonea.The magnification is 10x, the scale bar is 100 μm, and the experiment was performed in duplicates.The confirmation of senescent MeT-5A cells upon treatment with doxorubicin.(A) Staining DNA (using DAPI) and F-actin (using phalloidin) of the treated and untreated samples for morphological examination of the non-senescent and senescent cells, (B) SA-β-gal staining of the treated and untreated samples for the detection of senescent cells with the dichloro-dibromo indigo precipitate (greenish-blue dye) accumulation, (C) qPCR fold change analysis of p21 expression on treated and untreated samples.The magnification is 10x; the scale bar is 100 μm, and p-values were computed using a two-tailed, parametric unpaired t-test with Welch's correction.A snapshot of the initial layout of the different cells on the lattice -The picture depicts the initial layout of senescent, non-senescent, and cancer cells in a two-dimensional setup and the incorporated plugins.The relative sizes and the proliferation rates of the three cell types were calibrated from the experiments.

S11
Screenshot showing the analysis performed on images collected from mesothelial clearance assay to calculate the area occupied by MeT-5A and SKOV-3/OVCAR-3 cells in the same field.The highest value of the threshold with a neat background was chosen for generating the mask.The area of cells in the field was measured by summing up the individual particle area and is calculated under the 'Analyze particles' options with size limit = 0.01-Infinity.

Maximum MCS 8100
Table s2 -Table showing all the parameters and their possible values used in the computational model.
microenvironment on the adhesivity of ovarian cancer cells.(A) Representative images show the adhesion of GFP-labelled OVCAR-3 cells on control and senescent MeT-5A monolayers, (B) The bar graph shows the number of OVCAR-3 cells per field for control and senescent MeT-5A monolayers at 10x magnification with a scale bar of 100 μm.The experiments were performed in triplicates.The data are presented as mean +/-SEM and significance is obtained using an unpaired parametric t-test with Welch'of mixed cultures of RFP-labelled MeT-5A monolayers with GFP-labelled SKOV-3 cells.(A) Representative images show the area of senescent MeT-5A monolayer without SKOV-3 cells at 0-hour and 72-hour time points (can be seen in video S3).(B) The bar graph compares the area of senescent MeT-5A monolayers with and without SKOV-3 cells at a 72-hour time point.The fields are at 10x magnification with a scale bar of 100 μm.The experiments were performed in triplicates.The data are presented as mean +/-SEM and significance is obtained using an unpaired parametric t-test with Welch's correction. of mixed cultures of RFP-labelled MeT-5A monolayers with GFP-labelled OVCAR-3 cells.(A) Representative images show the area of control MeT-5A monolayer and OVCAR-3 cells at 0-hour and 60-hour time points.(B) Representative images show the area of senescent MeT-5A monolayer and SKOV-3 cells at 0-hour and 60-hour time points.(C) (i) Quantification of the area occupied by OVCAR-3 cells within control and senescent MeT-5A monolayers for 60 hours, plotted with 12-hour intervals.(ii) The bar graph compares the area of OVCAR-3 cells in the two monolayers at a 60hour time point.(D) (i) Quantification of the area of control and senescent MeT-5A monolayers for 60 hours, plotted with 12hour intervals.(ii) The bar graph compares the area of control and senescent MeT-5A monolayers at a 60-hour time point.(E) Representative images show the area of senescent MeT-5A monolayer without OVCAR-3 cells at 0-hour and 60-hour time points.(F) The bar graph compares the area of senescent MeT-5A monolayers with and without OVCAR-3 cells at a 72-hour time point.Quantification of the area of senescent MeT-5A monolayers with and without OVCAR-3 cells for 60 hours, plotted with 12-hour intervals shown in Di (brown line).The fields are at 10x magnification with a scale bar of 100 μm.The experiments were performed in triplicates.The data are presented as mean +/-SEM and significance is obtained using an unpaired parametric t-test with Welch's correction.Effect of senescence-associated secretory phenotype (SASP) and extracellular matrix secreted by control and senescent MeT-5A monolayers on the proliferation of SKOV-3 cells.(A) (i) Schematics of proliferation assay performed on SKOV-3 cells using conditioned media from control and senescent MeT-5A monolayers.(ii) Plot shows the fold change in fluorescence intensity of SKOV-3 cells for 30 hours of incubation with conditioned media, plotted with 12-hour intervals.(B) (i) Schematics of adhesion and proliferation assay performed on SKOV-3 cells seeded on extracellular matrix secreted by control and senescent MeT-5A monolayers.(ii) Plot shows the fold change in fluorescence intensity of SKOV-3 cells that adhered to the extracellular matrix for 36 hours, plotted with 12-hour intervals.The experiments were performed in duplicates, and the data are presented as mean +-associated secretory phenotype (SASP) and extracellular matrix secreted by control and senescent MeT-5A monolayers on the proliferation of OVCAR-3 cells.(A) The plot shows the fold change in fluorescence intensity of OVCAR-3 cells for 48 hours of incubation with conditioned media, plotted with 24-hour intervals.(B) The plot shows the fold change in fluorescence intensity of OVCAR-3 cells that adhered to the extracellular matrix for 24 hours, plotted with 12-hour intervals.The experiments were performed in duplicates, and the data are presented as mean +/-SEM.matrix secreted by control and senescent MeT-5A monolayers on the adhesivity and spread of OVCAR-3 cells.(A) Representative images of adhered OVCAR-3 cells on extracellular matrix secreted by control and senescent MeT-5A monolayers along with insets showing the spread of OVCAR-3 cells on senescent extracellular matrix.(B) (i) Bar graph shows the number of adhered OVCAR-3 cells per field on the extracellular matrix secreted by control and senescent MeT-5A monolayers.(ii) Bar graph shows the mean OVCAR-3 spread area on the extracellular matrix secreted by control and senescent MeT-5A monolayers.The fields are at 10x magnification with a scale bar of 100 μm and the colours are inverted in (A) to visualize cell edges.The experiments were performed in triplicates.The data are presented as mean +/-SEM and significance is obtained using an unpaired parametric t-test with Welch's correction.

S13A
snapshot of the active clearance code from Twedit++5 -The code shows the implementation of the killing of senescent cells if more than 55% of their perimeter is shared with ovarian cancer cells.

Table 1 -
RT PCR primer sequence information for the gene expression studies in non-senescent and senescent MeT-5A monolayers