Human amniotic epithelial cells (hAECs) isolation and characterization
hAECs were isolated as previously described (21)Briefly, the amniotic membrane was mechanically peeled from the underlying chorion, washed in Hanks' Balanced Salt Solution (HBSS, ThermoFisher Scientific, Reinach, Switzerland), cut into small pieces and trypsinised (0.25% Trypsin/EDTA, ThermoFisher Scientific) to release cells. Dispersed hAECs were collected by centrifugation, seeded at a density 2x105 cells/cm2 and cultured in hAEC culture medium consisting in DMEM/F-12 medium (ThermoFisher Scientific,) supplemented 2 mmol/l L-Glutamin, 100 U/ml Penicillin and 0.1 mg/ml Streptomycin (1% (v/v) of a L-Glutamin-Penicillin-Streptomycin stock solution from Sigma-Aldrich), 1 mmol/l sodium pyruvate (Sigma-Aldrich), 1% (v/v) MEM NEAA 100X (ThermoFisher Scientific), 0.1% fungin (InvivoGen, San Diego, CA), 10% fetal bovine serum (FBS; Merk Millipore, Zug, Switzerland), 0.05 mmol/l 2-mercaptoethanol (ThermoFisher Scientific), 10 ng/ml human recombinant epidermal growth factor (EGF, Sigma-Aldrich).
Confluent cells were trypsinised (0.05% Trypsin/EDTA, ThermoFisher Scientific) and characterized by flow cytometry (FC; figure 1.A). Cells were washed in FC buffer (PBS-0.1 % BSA supplemented with 0.01% sodium azide) and incubated for 30 minutes at 4°C with the following antibodies: FITC-conjugated anti-human CD105 (clone 266), BV421-conjugated anti-human CD326 (clone EBA-1), PerCP-Cy5.5 conjugated anti-SSEA4 (clone MC813-70) (1:50 dilution; all from BD Biosciences, Allschwil, Switzerland), PE-Cy7 conjugated anti-human CD90 (clone 5E10) (1:100 dilution, BD Biosciences), PE-conjugated anti-human HLA-E (clone 3D12) and APC-conjugated anti-human HLA-G (clone 87G) (1:20 dilution, Biolegend, London, UK) antibodies. Controls were stained with isotype-matched irrelevant antibody to evaluate non-specific binding to target cells. Cells were analyzed on a Gallios cytometer (Beckman Coulter, Indianapolis, Indiana, US) using Kaluza Analysis software from Beckman Coulter (Version 1.5.20365.16139). The percentage of positive cells for the different markers was assessed on a gate set on hAECs by using forward- and side-scatter analysis during the acquisition of data, followed by doublets and Draq7-positive (dead) cells exclusion. Representative histograms were plotted using FlowJo software (version 10.6.1, BD Biosciences). hAECs cultured on collagen-coated coverslips were fixed in 4% paraformaldehyde (PFA), rinsed and permeabilized. For histological characterization, after two washes, unspecific binding sites were blocked and samples were stained with the following primary antibodies: monoclonal anti-SSEA-4 (1:75 dilution, clone MC813, Abcam, Cambridge, UK), polyclonal anti-Oct4 (1:200 dilution, Abcam, Cambridge, UK) and monoclonal anti-human HLA-G (1:50 dilution, clone 4H84, BD Biosciences). The secondary antibodies were Alexa 555 anti-mouse or anti-rabbit antibodies (1:300 dilution, ThermoFisher Scientific). Stained cells were mounted with aqueous mounting medium containing DAPI (Fluoroshield Mounting Medium with DAPI, Abcam). Images were captured using a Zeiss Axioscan.Z1 slide scanner (Zeiss, Feldbach, Germany).
Rat pancreatic islets isolation
Animal experiments were performed under protocols reviewed and approved by the Geneva Veterinary authorities and the University of Geneva Institutional Animal Care and Use Committee.
Ten-week old male Lewis rats (LEW/OrlRj; Janvier Labs, Le Genest St-Isle, France) were used for pancreatic islet isolation. Rat pancreatic islets were isolated as previously described (22, 23), purified by Ficoll density centrifugation and cultured in islet culture medium consisting of DMEM (ThermoFisher Scientific) supplemented with 10% (v/v) FBS (ThermoFisher Scientific), 1 mmol/l sodium pyruvate, 11mmol/l glucose (Bichsel, Interlaken, Switzerland), 0.05 mmol/l 2-mercaptoethanol, 2 mmol/l L-Glutamine, 100 U/ml Penicillin and 0.1 mg/ml Streptomycin for one day prior to be used in co-culture experiments.
Exposure of hAECs to IFN-γ
hAECs were seeded onto 25cm² tissue culture-treated flasks at a density of 17’500 cells/cm2. Culture media were replaced the next day and cells were exposed to various concentrations of recombinant murine IFN-γ for 48 h. Untreated hAEC served as controls. Supernatants from treated and untreated hAECs were analyzed to detect secreted anti- and pro-inflammatory factors. hAEC phenotype were assessed by FC analysis.
Exposure of hAECs: Islet co-culture to pro-inflammatory cytokines
hAECs were thawed and added (1.0×106 cells) to a 100 mm tissue culture-treated dish with 500 islet equivalents (IEQ) in a total volume of 10 mL hAEC culture medium. Rat islet (RI) and hAECs monocultures were used as controls. After 24 h incubation, a pro-inflammatory cytokine cocktail consisting of 100 U/mL recombinant murine Interferon-gamma (IFN-γ), 800 U/mL recombinant human Tumor Necrosis Factor-alpha (TNF-α) and 50 u/mL recombinant human Interleukine-1 beta (IL-1β) was added to 10 ml of culture medium for an additional 48 h (figure 3.A).
All cytokines were obtained from Peprotech (London, UK).
Analysis of culture supernatant for soluble cytoprotective factors
Culture supernatants from hAECs monocultures were sampled before and after cytokine exposure and stored at -20°C until further assessment. Qualitative detection of secreted anti- and pro-inflammatory factors (IL6, IL10, TNF-α, G-CSF and TGF-β1) in the culture media was performed using the commercial Multi-Analyte ELISArray assay MEH003A from Qiagen (Courtaboeuf, France). Released IL6, IL10 and G-CSF were quantified using Quantikine ELISA for human IL-6, IL-10 and G-CSF (R&D Systems, Abingdon, UK), according to manufacturer instructions.
Cell apoptosis assay
RI (100 IEQ), hAECs (2x105 cells) and RI + hAECs (2x105 hAECs+100 IEQ) cultured in 35 mm² petri dishes with or without cytokines were washed with PBS and cytoplasmic histone-associated DNA fragments were extracted and quantified using the Cell Death Detection ELISA kit from Roche (Sigma-Aldrich) according to manufacturer instructions.
Islet functional assay
RI and RI + hAECs cultured with or without cytokines were assessed in duplicates for glucose stimulated insulin secretion. After a 1 hour pre-incubation in Krebs–Ringer buffered HEPES (pH 7.4) with 0.1% (w/v) BSA (KRB) at 37°C, islets and cells were successively incubated for 1 hour at 37°C in KRB solutions containing glucose at low (2.8 mmol/l) or high (16.7 mmol/l) concentration. Total insulin content was extracted using acid-ethanol solution. Supernatants were collected after each incubation time, insulin concentration was determined by ELISA (Mercodia, Uppsala, Sweden) and normalized to the total insulin content of the corresponding lysates. Islet responsiveness to glucose was defined as the ratio of insulin secretion in high glucose to insulin secretion in low glucose solution, hereafter referred to as the stimulation index (SI).
Real-Time quantitative Polymerase Chain Reaction
RNA was extracted from hAECs monocultures (106 cells in 10 cm² petri dishes) or RI+hAEC co-cultures using the RNeasy minikit (Qiagen, Courtaboeuf, France). cDNAs were synthesized by reverse-transcription using the GoScriptTM Reverse Transcription Kit (Promega, Dübendorf, Switzerland). RT-qPCR was performed using the Takyon No-Rox SYBR Core Kit blue dTTP (Eurogentec, Liège, Belgium), or the Taqman Fast Advance Master Mix (Thermofisher Scientific). Primers used for amplification were purchased from Microsynth (Balgach, Switzerland) and Thermofisher Scientific and targeted the following genes: human IL4, IL6, IL8, IL10, HLA-G, HLA-E, STAT1, STAT3, JAK1, JAK2 NFKB1, and rat Bcl2 and Nfkb1. Gene expression values were normalized to the housekeeping genes human RPLP0, human EIF2B1, rat Rplp1 and rat Actb, and calculated with the comparative cycle threshold Ct method (2-ΔCt method). All primer forward and reverse sequences are detailed in table 1.
Statistical analysis
Continuous and categorical variables are presented as mean ± SEM. Comparisons between groups were performed with unpaired two-tailed Student’s t-test or one-way / two-way ANOVA with Tukey’s or Sidak’s post-hoc test wherever appropriate. All statistical analyses were performed with Prism software 7.02 (GraphPad, La Jolla, CA, USA), and p < 0.05 was considered statistically significant.