25g needle

The amount of CO in the tissue samples was measured by GC according to the reported method^{18 (link)} with slight modifications. Tissues (100 mg) were homogenized (Power MasherII, nippi) in PBS (0.3 ml) and then disrupted by sonication (time: 10 s × 2, on ice, amplitude: 15; QSONICA). The solution and a 5 mmφ glass ball were placed in a 5 mL 25 G syringe (TERUMO) sealed with a septum cap (natural rubber, for a 7 mm tube). The atmosphere in the syringe (1 mL) was carefully replaced with He gas through inserting a 25 G needle (TERUMO) into the septum cap. Three drops of 30% sulfosalicylic acid (SSA) (Wako) were injected through the septum into the syringe. The mixture was then mixed well using a 5 mmφ glass ball. Methane gas (50 μl) was added to the head-space atmosphere through the septum in the syringe as an internal standard. CO liberated from the tissues into the gas phase (0.5 ml) was analyzed by GC (Shimadzu GC-2014, Shimadzu). The GC conditions were as follows: detector = TCD; column = SHINCARBON: carrier gas = He; column temperature = 40 °C; temperature at vaporizing chamber and detector = 120 °C; flow rate = 50 ml/min.

Tissue samples were placed in a 5 ml 25G syringe (TERUMO) sealed with a septum cap (natural rubber, for a 7 mm tube). The atmosphere in the syringe was replaced with pure CO gas (4 ml) (Sumitomo Seika) through inserting a 25G needle (TERUMO) into the septum cap. Tissue samples were then incubated at 4 °C under a CO atmosphere. After 1 h, the syringe was opened under air atmosphere, samples were taken out from the syringe, and the amount of CO was measured by the hemoCD1 assay.

A modified protocol was used based on previous studies 23. In brief, NIKS cells were first collected by trypsinization and after centrifugation and washing, cells were resuspended to ∼1 × 10⁶ cells/ml in F medium and fixed in 1.5% PFA (J61899; Alfa Aesar, Lancaster, UK) for 10 min at room temperature. Cells were then recovered by centrifugation (3000 rpm) and permeabilized in ice‐cold methanol (1 × 10⁶ cells/500 µl) (Sigma, Haverhill, UK) at 4 °C for 10 min. Cells were subsequently washed in PBS–1% bovine serum albumin (BSA) and passed gently through a 25G needle (Terumo, Leuven, Belgium) for to five times to avoid the formation of cell clumps. Cells were then incubated (1 × 10⁶ cells/100 µl) with a mouse primary antibody to keratin 10 (Krt10) (PA5‐32459; Thermo Scientific, East Grinstead, UK) for 1 h on ice with occasional agitation. The optimal concentration of Krt10 antibody for use in these assays was determined experimentally to be 1 µg/µl. Cells were subsequently washed in PBS–1% BSA and incubated with Alexa Fluor 488 donkey anti‐mouse secondary antibody (A21202; Life Technologies, Paisley, UK) diluted to 1 µg/ml for 30 min at room temperature. After extensive washing (at least three times in PBS), cells were subjected to FACS sorting using either a DxP8 (Cytek, Ely, UK) or a MoFlo MLS cell sorter (DakoCytomation, London, UK).