Senescent peritoneal mesothelium creates a niche for ovarian cancer metastases

Abstract

Although both incidence and aggressiveness of ovarian malignancy rise with age, the exact reason for this tendency, in particular the contribution of senescent cells, remains elusive. In this project we found that the patient's age determines the frequency of intraperitoneal metastases of ovarian cancer. Moreover, we documented that senescent human peritoneal mesothelial cells (HPMCs) stimulate proliferation, migration and invasion of ovarian cancer cells in vitro, and that this effect is related to both the activity of soluble agents released to the environment by these cells and direct cell-cell contact. The panel of mediators of the pro-cancerous activity of senescent HPMCs appeared to be cancer cell line-specific. The growth of tumors in a mouse peritoneal cavity was intensified when the cancer cells were co-injected together with senescent HPMCs. This effect was reversible when the senescence of HPMCs was slowed down by the neutralization of p38 MAPK. The analysis of lesions excised from the peritoneum of patients with ovarian cancer showed the abundance of senescent HPMCs in close proximity to the cancerous tissue. Collectively, our findings indicate that senescent HPMCs which accumulate in the peritoneum in vivo may create a metastatic niche facilitating intraperitoneal expansion of ovarian malignancy.

Figure 1

Effect of senescent HPMCs on the progression of ovarian cancer cells in vitro ovarian cancer cells were subjected to CM from young and senescent HPMCs, then their proliferation (a), distribution in the cell cycle (b), and migration (c) were measured. In addition, the cancer cells were seeded on top of young and senescent HPMCs in order to examine their proliferation (d-e) and invasion (f). The hatched areas in the histograms shown in panel (b) indicate cells in the S phase of the cell cycle. Panel (e) shows representative pictures of fluorescence emitted by GFP-transfected cancer cells growing in direct contact with the HPMCs (× 100; bar, 100 μm). The asterisks indicate significant differences as compared with young HPMCs. The experiments were performed in duplicates using HPMCs obtained from 14 different donors. The results are expressed as mean±S.D.

Figure 2

Examination of the intraperitoneal development of ovarian tumors upon i.p. injection of ovarian cancer cells together with young or senescent HPMCs. Representative images showing bioluminescence intensity recorded 5 and 12 (A2780) or 20 (OVCAR-3, SKOV-3) days after cell implantation (a). The dynamics of xenograft development, estimated according to the difference between the highest bioluminescence intensity recorded throughout the experiment and the initial value, were recorded 5 days after cell injection (b). The asterisks indicate a significant difference as compared with xenografts established in the presence of young HPMCs. Experiments were performed on seven animals per group with HPMCs established from six different donors. The results are expressed as mean±S.D.

Figure 3

Identification of soluble mediators of the pro-cancerous activity of senescent HPMCs ovarian cancer cell proliferation (a-c) and migration (g-i) were investigated in the presence of a CM generated by senescent HPMCs upon its pre-incubation with specific neutralizing antibodies directed against proteins that were initially identified as plausible mediators of increased proliferation and migration of the cancer cells (see Table 3). Because the reduction of the cancer cell progression that was observed in these experiments was only partial, subsequent studies on cancer cell proliferation (d-f) and migration (j-l) were conducted in which proteins whose neutralization reduced the stimulatory effect of senescent HPMCs were blocked simultaneously in one sample of the CM. The asterisks indicate significant differences as compared with cells exposed to CM from senescent HPMCs. The hashes indicate significant differences as compared with cells exposed to CM from young HPMCs. The experiments were performed in triplicates using HPMCs obtained from six different donors. The results are expressed as mean±S.D.

Figure 4

Role of p38 MAPK as a mediator of HPMC senescence replication (a), the level of γ-H2A.X foci (b) and SA-β-Gal (c-d), and the secretion of proteins (e) were compared in HPMCs undergoing senescence in standard culture conditions (control cells) and in cells exposed to the p38 MAPK inhibitor – SB202190. The frame located in panel (a) indicates a time point at which HPMCs propagated in standard conditions reached senescence. The results shown in panels (b, c, d, and e) refer to this moment. Panel (d) shows representative results of staining against SA-β-Gal (positive cells are green; × 100; bar, 100 μm). The asterisks indicate significant differences as compared with senescent HPMCs. The experiments were performed by using HPMCs obtained from 12 different donors. The results are expressed as mean±S.D.

Figure 5

Effect of HPMC rejuvenation by p38 MAPK inhibition on the intraperitoneal development of ovarian tumors in mice. Representative pictures showing bioluminescence intensity of A2780 cells co-injected i.p. with senescent HPMCs or with middle-aged HPMCs treated with SB202190 (a). The dynamics of xenograft development, estimated according to the difference between the highest bioluminescence intensity recorded throughout the experiment and the initial value, were recorded 2 days after cell implantation (b). Comparison of masses of tumors excised from a mouse peritoneum at the end of the experiment (c). The asterisks indicate a significant difference as compared with xenografts established in the presence of senescent HPMCs. Experiments were performed on five animals per group with HPMCs established from five different donors. The results are expressed as mean±S.D.

Figure 6

Presence of senescent HPMCs in proximity to cancerous tissue in tumors excised from the peritoneal cavity of patients with ovarian cancer HPMCs were identified according to a positive (brown) reaction against D2-40 antigen. Senescent cells were identified according to the expression (green) of SA-β-Gal and a positive reaction (brown) against γ-H2A.X. The pictures shown in this figure are representative examples obtained from 4 out of 11 patients examined. Arrows indicate exemplary positive reactions against SA-β-Gal. CA – cancerous tissue. Magnification × 400; bar, 50 μm