Cell culture

Human primary epidermal keratinocytes were derived from foreskin tissue as described previously [8] and cultured in a keratinocyte growth medium (calcium concentration 0.1 mM) (KGM, KBM™ Gold™ Basal Medium with KGM™ Gold™ SingleQuots™ supplements, Lonza Bioscience, Durham, NC, USA). The cells were cultured in a humidified incubator in 5% CO2 at 37 °C. From second to fourth passages of keratinocytes were used for all experiments. Human tissue for primary cultures was obtained with the written informed consent of donors, under the approved protocol by the Institutional Review Board of Seoul National University Hospital. For siRNA transfection, cells were seeded on 35 mm culture dishes or 6 well plates and grown until > 90% confluence. To validate siRNA working, cells were harvested after 48 h. Except for siRNA validation experiments, after 24 h of siRNA transfection, the culture medium was switched to a fresh high calcium (2mM)-containing KGM, and the cells were incubated for indicated times. If the incubation time is longer than 2 days, the culture medium was replaced with a fresh high calcium-containing KGM at 48 h after the last medium switching. For the experiment of investigation of EGFR phosphorylation, after 24 h of siRNA transfection, cells were incubated in fresh KBM for 24 h, and then treated with 0 or 20 ng/ml of EGF in fresh KBM. Cells were harvested at 5 min after treatment with EGF. For an epidermal growth factor receptor (EGFR) inhibitor, gefitinib (Santa Cruz Biotechnologies, Santa Cruz, CA) treatment experiments, cells were grown until > 90% confluence, and then incubated with indicated doses of gefitinib (0, 20, or 100 nM) and high calcium-containing KGM for 4 days. The culture medium was additionally replaced with indicated doses of gefitinib and high calcium-containing fresh KGM at 48 h after incubation.

siRNA transfection

Non-targeted negative control siRNA was used as a control siRNA (5’ – CCU ACG CCA CCA AUU UGC U – 3’(dTdT)). Two pools of FUT1 siRNA were used to knockdown in this study. One pool (designated as #1) was purchased from Bioneer (Bioneer Corp., Daejeon, Korea), and composed of three different sequences of siRNAs (5’-CCG UGC UCA UUG CUA ACC A-3’(dTdT), (5’-CGA AAA GCG GAC UGU GGA U-3’(dTdT), and (5’-UGG UAC AGC UUC UGG AGC A-3’(dTdT)). The other (designated as #2) was SMARTPool ON-TARGETplus Human FUT1 siRNA (2523) from Horizon Discovery Ltd. (Dharmacon™, Cambridge, United Kingdom), which is provided as a mixture of four different sequences of siRNAs. siRNA was transfected to keratinocytes at a confluence of > 90% using G-fectin (Genolution, Seoul, South Korea) according to the manufacturer’s manuals. Briefly, 400 nM of each siRNA pool was transfected using 4 µl of G-fectin transfection reagent per 35 mm dish. After 24 h, the culture medium was freshly changed. To confirm the specific downregulation of FUT1 mRNA expression, control or FUT1 siRNA-transfected keratinocytes were harvested 2 days after the transfection, and the expressions of FUT1 mRNA were measured by real-time quantitative PCR, normalized to 36B4 mRNA expression.

Western blot analysis

For Western blot analysis, at indicated times, cells were lysed with a lysis buffer (2% SDS, 2-mercaptoethanol (715 mM), 5% glycerol, 50 mM Tris pH 6.8, and 0.01% bromophenol blue) containing protease inhibitor cocktail (cOmplete™, Mini Protease Inhibitor Tablets, Roche, Mannheim, Germany) and phosphatase inhibitor cocktail 2 (Sigma-Aldrich, St. Louis, MO), and then boiled at 95ºC for 7 min. Same amounts of protein samples were separated by SDS-PAGE gel (6, 8, or 12%) and transferred onto nitrocellulose membranes. After blocking with 5% skim milk in Tris-buffered saline containing 0.1% Tween 20, membranes were incubated with antibodies specific for desmoglein 1 (B-11), desmoglein 3 (5H10), H type 2 antigen (BRIC231), α-tubulin (B-5-1-2), EGFR (A-10), filaggrin (AKH-1) (Santa Cruz Biotechnologies), p-EGFR(Y1068) (polyclonal, Cell Signaling Technology, Danvers, MA, USA), desmoplakin 1,2 (DP2.15, abcam, Cambridge, UK), desmocollin 1 (772906, R&D systems, Minneapolis, MN, USA), loricrin (Poly19051, Covance, Princeton, NJ, USA), plakophilin 1 (10B2) (Thermo Fisher Scientific, Waltham, MA), or FUT1 (rabbit polyclonal) (Cusabio Technology LLC, Houston, TX, USA). Membranes were washed using Tris-buffered saline with 0.1% Tween-20, incubated with secondary antibodies, and signals were detected using an enhanced chemiluminescent solution (Supersignal West Pico Plus or Femto, Thermo Scientific) with Amersham Imager 680 (GE Healthcare, Chicago, IL, USA). The densitometry of bands was analyzed using the Image J software (National Institutes of Health, Bethesda, MD, USA) and normalized with the quantified values of α-tubulin or β-actin as a loading control.

Dispase-based dissociation assay

The cell-cell adhesive strength was assessed by dispase-based dissociation assay. After 48 h of siRNA transfection, when cells got a confluent cell monolayer, the culture medium was switched to a high calcium medium for an additional 8 h. The cells were then incubated in phosphate-buffered saline (PBS) containing dispase II (Dispase®, Roche, Mannheim, Germany) for 15 min to make the cells floating monolayer sheets. Released cell sheets were carefully transferred to 15 ml conical tubes with 5 ml PBS. The tubes were then subjected to mechanical stress by rotating 10–20 times in a tube rotator (Labquake™, Thermo Scientific, Waltham, MA). To give consistent and reproducible mechanical stress, a rotator was applied instead of the manual pipetting used in the literature [9]. PBS containing fragmented cell sheets was transferred to 35-mm culture dishes and then scanned by a digital scanner. The total number of fragments was counted by Image J software.

Transmission electron microscopy (TEM)

Ultrastructures of desmosomes were analyzed using TEM images. After siRNA transfection, dispase II was treated as above mentioned to make floating monolayer sheets. Dispase was used to harvest the cells while maintaining the intercellular adhesive structure. Dispase primarily disrupts cell-extracellular matrix adhesion while leaving intercellular adhesion of epithelial cells intact [10]. The cell sheets were transferred to 1.5-ml microcentrifuge tubes with 1 ml PBS. Harvested sheets were centrifuged at 500 g for 3 min. After washing with PBS twice, the cell pellets were fixed overnight in a mixture of cold 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M phosphate buffer (pH 7.2). The rest of the procedures were performed according to the common protocols for conventional electron microscopic exams (processed in the laboratory of the electron microscope, Seoul National University Hospital). More than 50 images were obtained per group, and the number of desmosomes used in size analysis was 193 and 207 in the control and FUT1 knockdown groups, respectively.

Immunoprecipitation

Primary keratinocytes were grown with KGM in the 100-mm dish until > 90% confluence, and the medium was switched to a high-calcium medium for an additional 48 h. Then cells were lysed in lysis buffer (1x RIPA) containing protease inhibitor cocktail (cOmplete™, Roche) and phosphatase inhibitor cocktail 2 (Sigma-Aldrich) on ice for 1 h with intermittent vortexing, and the supernatant was obtained by centrifugation at 13,000 rpm for 20 min was used for immunoprecipitation. For immunoprecipitation, 500 ug of cell lysates were incubated with 2 ug of mouse IgG or mouse monoclonal primary antibodies against H2 (BRIC231), and EGFR (A-10) and 30 ul of Protein A/G plus-Agarose (Santa Cruz Biotechnologies) for overnight at 4℃ on the rotator. Then, the samples were centrifuged at 3,000 rpm for 3 min, and the pellets were washed three times with 1x PBS at 3,000 rpm for 3 min. After the final wash, the pellets were added with 2x SDS sample buffer, and heated at 95℃ for 5 min. After cooling on ice, the samples were centrifuged at 13,000 rpm for 3 min, and the supernatants were subjected to the SDS-PAGE for Western blot analysis. All the procedures for Western blot analysis were the same as mentioned above, except using EasyBlot anti-mouse IgG (Genetex, Irvine, CA, USA) as an HRP-conjugated secondary antibody.

Preparation of an adenovirus vector encoding human FUT1 gene

For overexpression of FUT1 in primary human keratinocytes, we generated a recombinant adenovirus vector encoding human wild-type FUT1. FUT1 gene was amplified from human cDNA and subcloned into a pENTR(TM)/D-TOPO(TM) vector that has attL sites for site-specific recombination with a Gateway destination vector (Invitrogen, Carlsbad, CA). The replication-incompetent adenoviruses were created using a Vira-power adenovirus expression system (Invitrogen) in HEK293A cells, as previously described [11]. The amplified virus was purified by double cesium chloride-gradient ultracentrifugation. Viral particles in the cesium chloride gradient were collected and dialyzed using a Slide-A-Lyzer Dialysis Cassette (Thermo Scientific). An adenovirus vector encoding GFP was used as a control vector [11]. For adenovirus-mediated gene transduction, primary human epidermal keratinocytes were infected with an indicated multiplicity of infection (MOI) of the adenovirus in a high-calcium growth medium. If the incubation time is longer than 2 days, the culture medium was additionally changed with high calcium-containing fresh KGM at 48 h after infection.

Real-time quantitative PCR

siRNA-transfected keratinocytes or adenoviral vector-treated keratinocytes were harvested at indicated time points for the analysis of mRNA expression, and total RNA was extracted using RNAiso plus (Takara Bio Inc., Shiga, Japan) according to the manufacturer’s protocols. 1 µg of total RNA was used to convert to cDNA using a First Strand cDNA Synthesis Kit (Thermo Fisher Scientific). Quantitative real-time PCR was performed on a CFX96™ Real-Time PCR Detection System (Bio-Rad Laboratories, Inc., Hercules, CA, USA) using 2X GreenStar™ qPCR Master Mix (Bioneer Corp., Daejeon, Korea) according to the manufacturer’s instructions. The PCR conditions were 95°C for 3 min, followed by 40 cycles at 95°C for 22 sec and 60°C for 1 min. The data were analyzed through the comparative CT method (ΔΔCT method), normalized to 36B4, and presented as relative folds ± standard deviation. Primers used were: FUT1 forward: 5’-AAC AGA TCC GCA GAG AGT TCA, reverse: 5’-GCA TAA CCT GCA GAT AGT CCC, AREG forward: 5’-ACC TAC TCT GGG AAG CGT GA-3’ and reverse: 5’-AGC CAG GTA TTT GTG GTT CG-3’. HBEGF forward: 5′-CCT CCA GTG CTG GAT TTG AT-3′, and reverse: 5′-GCC AGG AAA TTG CCA AAG TA-3′, TGFA forward: 5′-CTT CAA GCC AGG TTT TCG AG-3′, and reverse: 5′- CTT CTG TGA CTG GGC AGG TT -3′, and 36B4 forward: 5’- TCG ACA ATG GCA GCA TCT AC -3’ and reverse: 5’- TGA TGC AAC AGT TGG GTA GC -3’.

Enzyme-linked immunosorbent assay (ELISA)

Conditioned media were collected from keratinocytes at indicated time, and secretions of amphiregulin (AREG), heparin binding-epidemal growth factor (HB-EGF), and tumor growth factor-α (TGF-α) in conditioned media were measured by ELISA using commercial kits for human AREG (#CSB-04486 h), HB-EGF (#CSB-E09716h), and TGF-α (#CSB-E04724h), purchased from Cusabio Technology LLC, according to the manufacturer’s instructions. Data are represented as mean relative expression (%) ± SD relative to levels in the control groups (NC siRNA-transfected or control adenovirus-infected cells).

Statistical analysis

Statistical analyses were performed using the Wilcoxon signed-rank test or one-way ANOVA followed by Bonferroni’s post-hoc test. Data are represented as the mean values ± SD of four or more independent experiments. Statistical significance was set at p < 0.05.