Verteporfin

Human HSC cell line, LX2, was obtained from American Type Culture Collection (Manassas, USA) and cultured in DMEM with 1% penicillin/streptomycin and 10% fetal bovine serum at 37 °C. LX2 cells were incubated under hypoxia (1% oxygen, 5% carbon dioxide and 94% nitrogen) in the hypoxia chamber (Galaxy 48 R, Eppendorf, Hamburg, Germany) for 3 h, followed by reoxygenation (hypoxia/ reoxygenation) under the normoxic conditions (21% oxygen, 5% carbon dioxide and 74% nitrogen) for 6 h to mimic activation of HSCs under ischemia–reperfusion injury^{23 (link)}.
To evaluate the dynamic change after hypoxia/reoxygenation, LX2 cells were harvested at 0 h, 1 h, 2 h, 4 h and 6 h after normoxic conditions for RT-qPCR and Western blot analyses (Supplementary Fig. 1).
To investigate the roles of YAP in regulation of CTGF expression, YAP siRNA or YAP function inhibitor (Verteporfin, 2 μM, s1786, Selleck Chemicals, Houston, USA) was incubated before hypoxia to observe the roles of YAP in the regulation of CTGF expression LX2 cells.

To evaluate the growth rate, NPCs were seeded at fixed cell densities (1 × 10⁵/well in a 12-well plate) in a standard maintenance culture medium and cell number was evaluated after 3, 6, and 9 days in culture. For the MTT assay (Roche), 10,000 NPCs have been seeded in a 96-well plate, and after 3 days in culture with NPC-specific medium, cell proliferation has been evaluated according to the kit protocol. The measurement of the released absorbance has been detected through the Spark multiplate reader (Tecan). When cells have been treated with 0.5 µM Verteporfin (Selleckchem), NPCs have been treated for 24 h before detecting cell proliferation through MTT assay.
Cell death and apoptosis were analysed through flow cytometry (MACSQuant Analyzer, Mylteni); NPC were detached with Accutase, washed once with PBS, and stained with propidium iodide (PI) and AnnexinV (FITC) for 30 min in PBS 4% FBS before data acquisition. Flow cytometry data were analysed with the FlowJo software.

PS lumen devices were prepared as described above to create tubular void structures embedded in hydrogel. GFP fluorescent MDA-MB-231 cells were resuspended at 5 million cells/mL and loaded into the lumen structure as described above and left to adhere for 24 h. Verteporfin (Selleckchem, S1786, Pittsburgh, PA, USA), was dissolved at 25 mg/mL in DMSO, and later diluted 1:50,000 (to a final concentration of 500 ng/mL) in relevant media. Verteporfin solution was perfused through the lumen and incubated at 37 °C for 1 h. Thereafter, cells in the lumen were exposed to fluorescent light using a Nikon Eclipse Ti microscope and a 485/35 nm filter for 45 s. After the light activation of Verteporfin, cells were left in the incubator overnight. Viability was assessed via propidium iodide staining as described in subsequent sections, since cells were already fluorescently green, calcein was not required to stain live cells. Images were analyzed using ImageJ by plotting the fluorescence profile in three different region of interests (ROIs) in the image within the lumen (at the beginning, center and end of the lumen along the axis). The area under the curve was calculated and normalized to the highest value (among green and red) to demonstrate which of the fluorescence channels was more prevalent in each section (and ROI) of the lumen.

KDM3A (ab91252, Abcam, Cambridge, MA, USA), Ki67 (ab16667, Abcam, Cambridge, MA, USA), E-cadherin (CST, #3195, Cell Signaling Technology, Danvers, MA, USA), Vimentin (CST, #5741, Cell Signaling Technology, Danvers, MA, USA), Hippo pathway antibody sampler kit (CST 8579, Cell Signaling Technology, Danvers, MA, USA), including LATS1, MST1, MST2, YAP1, TAZ, and GAPDH (sc-365062, Santa Cruz, CA, USA) were used in this study. H3K9m2 inhibitor Bix01294, UNC0631, Hippo pathway inhibitor verteporfin were purchased from Selleck (Houston, TX, USA).