Fig. 5: Inhibition of HSP90 attenuates TIMP1^EV mediated ECM remodelling.

A Immunoblot of CD63, HSP90AA and TIMP1 following immunoprecipitation of TIMP1 from CRC-EVs (HCT116, HT29, and SW620) or Rabbit IgG as negative control. Input of lysates of HCT116, HT29, and SW620 EV protein were used as internal control. ACTB and CD9 were used as loading controls. Whole cell lysate (WCL) of SW620 EV treated pFs was used as an additional control. B Immunoblot analysis of TIMP1 expression in pFs treated with EVs and CM from HCT116, HT29 and SW620 in the presence of HSP90AA AB (20 ng/ml) and corresponding controls after 48 h. ACTB was used as loading control. C Densitometry quantification of TIMP1 expression normalised to ACTB shown in (B) using imageJ. D Immunoblot analysis of TIMP1 expression in pFs treated with EVs from HCT116, HT29 and SW620 in the presence or absence of 17AAG (0.3 µM) after 48 h. E Densitometry quantification of TIMP1 expression normalized to ACTB shown in (D) using imageJ. F Representative bright field images of collagen matrigel lattices with embedded pFs treated with CRC-EVs (HCT116, HT29 and SW620) in the presence or absence of 17AAG (0.3 µM) after 48 h. G Representative bright field images of collagen matrigel lattices with embedded pFs treated with serum EVs from HD, CRC liver MET, and CRC treated pFs in the presence or absence of 17AAG (0.3 µM). H Graphical overview of the TIMP1^EV mediated ECM remodelling. TIMP1^EV binds to HSP90AA and CD63 receptor on the cell surface of fibroblasts to regulate TIMP1 in the recipient cells leading to ECM remodelling (left). Inhibition of HSP90 or TIMP1^EV levels affects TIMP1 mediated ECM remodelling (right). All experiments were performed at least 3 times. Error bars depict mean ± SEM. P values were calculated by unpaired t test. *p < 0.05.