Patients who had not been previously treated for keloids prior to surgical excision were included in the study. Using clinical criteria, i.e. a scar with a history of earlier local trauma and growth that has gone beyond the scar’s border, the surgeon was able to identify keloids. Pathologists then confirmed the histology by examining tissue slices using the clinical criteria for a normal scar—light-coloured and flat [24–26 (link)]. The skin tissues of the normal control group were taken from circumcision patients without underlying diseases. Six keloid specimens, six normal scar specimens and six foreskin specimens from urology patients were collected (Table S1, see online supplementary material). Because margins for surgical removal of keloids are relatively narrow, the amount of tissue that can be used to extract normal fibroblasts (NFs) is very small. To ensure that sufficient NFs could be extracted, we used foreskin tissues as the source of NFs. To minimize the impact of cell heterogeneity due to differences in keloid biopsy sites and cell generations, intralesional keloid samples were used to extract KFs, and healthy cells at passages three to five were used in the experiments. The tissues were cut into small pieces ~1 mm3 in size, placed in 20 ml of 0.2% type I collagenase and digested at 37°C for 4–6 h. The digested mixture was filtered through a 100-mesh cell strainer, centrifuged and rinsed twice with serum-containing culture solution. Isolated cells were inoculated into a 25 cm2 disposable culture flask, cultured in Dulbecco's modified eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin and 0.1 g/ml streptomycin at a cell density of 106/ml and placed in a 37°C, 5% CO2 incubator. This study was approved by the Medical Ethics Committee of the Medical Ethics Committee of the First Affiliated Hospital, Sun Yat-sen University (Guangzhou, China). Written informed consent was obtained from all participants.