*2.9. Immunocytochemistry (ICC)*

iSGECs and SGCLs were plated onto Coating Matrix (2D) (Gibco, Thermo Fisher Scientific, Waltham, MA, USA,) covered 8-well Nunc™ Lab-Tek™ II Chamber Slides™ (Thermo Fisher Scientific, Waltham, MA, USA). Antibodies and concentrations are listed in Supplementary Table S2. Cells were grown for 48–72 h, then fixed with ice-cold methanol (−20 ◦C, 10 min) and incubated with the following antibodies directed against them: KRT8, KRT18, KRT19, AQP5, ZO-1, α-SMA, Vimentin, and E-cadherin proteins. Cells were also fixed with 4% paraformaldehyde (room temperature, 15 min) to stain Ki-67 and α-amylase. Fixed cells were permeabilized with 0.25% Triton X-100 in TBS (room temperature, 10 min).

Cells were fixed then washed (3x) in TBS and incubated (room temperature, 5 min) in 3% H202 (*v*/*v*) TBS to block endogenous peroxidases. Next, cells were incubated in blocking buffer consisting of 1% BSA (*w*/*v*), 2% skim-milk (*w*/*v*), and 0.1% Tween-20 in TBS (room temperature, 1 h). Primary antibodies were diluted in TBS with 0.1% Tween-20 and 1% BSA. The fixed cells were incubated with the primary antibody (4 ◦C, 12 h) (Supplementary Table S2). Cells were washed with TBS (3 times, 5 min each) and incubated with anti-mouse secondary antibodies at different concentrations (Supplementary Table S2) and diluted in TBS with 0.1% Tween-20 (room temperature, 1–2 h). After incubation, cells were washed with TBS (3X).

Monolayer culture-fixed cells were incubated in diaminobenzidine (DAB) (Pierce, Thermo Fisher Scientific, Waltham, MA, USA) 1x solution (2–7 min). DAB was washed away using TBS to stop the reaction and the cells were counterstained with Gills II hematoxylin (Richard-Allan Scientific, Kalamazoo, MI, USA) (diluted 1:5 *v*/*v* DPBS) for 30 s to highlight nuclei. Slides without primary antibody served as negative controls.

For immunocytochemistry (ICC)–immunofluorescence (IF) visualization of 3D spheroid cultures grown on matrigel, cells were fixed using 4% paraformaldehyde (room temperature, 25 min) and permeabilized with 0.25% triton X-100 in TBS (room temperature, 10 min). The same blocking and incubation procedure previously outlined was followed except 4 ,6-diamidino-2-phenylindole (DAPI) (Abcam, Cambridge, UK) was used to highlight DNA. Slides without primary antibody served as negative controls.

Slides were observed using an Olympus BX51 fluorescence microscope (Shinjuku City, Tokyo, Japan) and photographed with an Olympus DP70 (Shinjuku City, Tokyo, Japan) mounted camera. Cells in culture plates were viewed by phase contrast on an Olympus IX-71 inverted fluorescence microscope and photographed using an Olympus DP70 mounted camera. Proliferation scores were determined based on Ki-67 protein detection, a nuclear protein expressed during several cell cycle phases (G1/S/G2/M), which is frequently employed as an indicator of actively proliferating cells [22]. Proliferation scores (%) were generated by selection of three random images for nuclei counting using Fiji-ImageJ software [23]. The total number of cell nuclei was divided by the total number of Ki-67 (+) nuclei and averaged among three images.
