Serum samples.
Serum samples were obtained from twelve patients with SSc fulfilling 2013 ACR/EULAR criteria [5]. All patients had anti-nuclear antibodies (ANA) at indirect immunofluorescence on HEp-2 cells, at a titer greater than 1:160, with staining patterns consistent with the antigenic specificity. Four patients carried ATA, four ACA, two ARA and two anti-Th/To. The remaining autoantibody profile was negative. In all cases, antibody reactivities against scleroderma antigens were confirmed using two different techniques: line blot (“EUROLINE-SSc profile”, Euroimmun, Lubeck, Germany, which includes the following antigens: Ro52, PDGF receptor, Ku, Pm/Scl75, PM-Scl100, Th/To, NOR90, Fibrillarin, RP155, RNA Polymerase III RP11, CENP B, CENP A, and DNA topoisomerase I) and chemiluminescent immunoassays (“QUANTA Flash CTD Screen Plus”, INOVA Diagnostics, San Diego, CA, USA, which detects antibodies against dsDNA, Sm/RNP, Ro52, Ro60, SSB, DNA topoisomerase I, centromere, Mi-2, Ku, Th/To, RNA Polymerase III, Pm/Scl, PCNA, Jo-1, and ribosomal P). The demographic and clinical features of enrolled SSc patients are detailed in Table 1. Two SLE patients were recruited; one patient carried anti-Sm, anti-U1 ribonucleoprotein (RNP) and anti-double stranded DNA antibodies, the other harbored anti-Sm [16]. Serum was also obtained from two subjects with primary anti-phospholipid syndrome (PAPS) and positive lupus anticoagulant test, anti-cardiolipin and anti-b2 glycoprotein I IgG antibodies [17]. Four normal healthy subjects (NHS), matched on age and gender to patients, with no autoimmune disease and negative autoantibody profile, were enrolled. Serum samples were stored at -20°C.
Endothelial cell culture.
Human umbilical vein endothelial cells (HUVECs) were isolated from normal term umbilical cord vein by type II collagenase perfusion (Worthington, Lakewood, NJ, USA). HUVEC cell cultures were maintained in complete E199 medium (Flow Labs) supplemented with 20% heat inactivated foetal bovine serum (FBS; PAA Laboratories-GE Healthcare), 1% L-glutamine (MP Biomedicals Inc.), 100 U/ml penicillin-streptomycin (MP Biomedicals) and 250 ng/ml Amphotericin B (MP Biomedicals). Confluent cells were passaged with a 0.25% trypsin/EDTA (Gibco-Life Technologies) [18]. In all the experiments, a pool of cells from at least three donors were used at the first passage.
Healthy skin fibroblast cell culture.
Dermal fibroblasts were isolated from skin biopsies from two NHS. Under local anesthesia with 1% xylocaine, 5 mm punch skin biopsies were performed in the distal forearm. Samples were minced into small pieces, and digested by collagenase type I (ThermoFisher Scientific Inc, Waltham, MA, USA) for 2 hours at 37°C with 5% CO2. After a centrifugation at 300g for 10 minutes, pellets were resuspended in 1 ml D-MEM (Gibco-Life Technologies, Groningen, NL) supplemented with 20% Fetal Bovine Serum (FBS, PAA-GE Healthcare, Buckinghamshire, UK), 2 mM glutamine (Sigma-Aldrich), penicillin (100 U/ml)-streptomycin (100 mg/ml) (Sigma-Aldrich) and transferred into a T25 plate (Corning Incorporated, NY, USA). Cultures were maintained at 37°C in 5% CO2-humidified incubator until confluence. Non-adherent cells and dermal tissue were removed by washing, established fibroblasts were passaged after trypsin/EDTA (ThermoFisher Scientific) release up to the eight passage. Cells were maintained in D-MEM with 10% FBS, 2 mM glutamine, penicillin (100 U/ml)-streptomycin (100 mg/ml) (ThermoFisher Scientific) or incubated overnight in D-MEM with 1% FBS before functional studies. The purity of fibroblast culture was 98% as detected by flow cytometry using a mouse anti-human CD90 and a mouse anti-human CD45 antibodies–PE conjugated (BD Biosciences, San Jose, CA, USA).
Healthy skin fibroblasts were stimulated with supernatants from HUVECs incubated with scleroderma and control ICs for 24 hours. The mRNA levels of collagen (col)Iα1 and matrix metalloproteinases (mmp)-1, the secretion of transforming growth factor (TGF) -b1 and the protein expression levels of a smooth muscle actin (a-SMA) and IL-6 were evaluated. In the latter experiments, fibroblasts were also stimulated with tumour necrosis factor (TNF) -a (10 ng/mL, R&D Systems) and with supernatants from HUVECs incubated with TNF-a for 24 hours as positive controls.
Immune complexes.
ICs were precipitated from NHS’ and patients’ sera. Briefly, serum samples were mixed with ice-cold 5% polyethylene-glycol (PEG) 6000 (Sigma-Aldrich; Saint Louis, MO, USA)-0.1 M EDTA (Bioscience, Inc, La Jolla, CA, USA) and incubated overnight at 4°C. Samples were diluted three times with 2.5% PEG 6000 in RPMI (Euroclone S.p.A., Pero, Italy), layered on top of 2.5% PEG 6000 supplemented with 5% human serum albumin (Sigma-Aldrich) and centrifuged at 3900 rpm at 4°C for 20 minutes. Pellets were dissolved in D-PBS to initial serum volume and immediately used at 1:2 dilution [19].
Every sample was used in triplicates, each experiment was repeated twice using SSc-ICs isolated from all patients for each autoantibody specificity and control ICs.
The potential endotoxin contamination of IC preparations was ruled out by limulus amoebocyte lysate (LAL) gel-clot test (Pyrosate Kit, Associates of Cape Cod Incorporated, East Falmouth, MA, USA; sensitivity 0.25 EU/ml).
ICAM-1 expression.
Inter-cellular adhesion molecule (ICAM) -1 surface levels were evaluated by home-made cell ELISA, as in previous studies [20]. Confluent EC monolayers were rested in D-MEM with 1% FBS overnight in 96-well plate.
After 24-hour incubation with 100 ml/well of SSc-ICs, PAPS-ICs, SLE-ICs, NHS-ICs, IL-1b (50 U/ml, PeproTech, Rocky Hill, NJ, USA), LPS (1 mg/ml, R&D Systems, Minneapolis, MN, USA) or medium alone, cells were washed twice with HBSS (Sigma-Aldrich), and incubated for 60 minutes at room temperature with 100 ml/well of murine monoclonal IgG specific for human ICAM-1 (CD54, R&D Systems). The antibody was used at a final dilution of 1:500 in HBSS-FBS 2.5%. After two additional washes, cells were incubated for 90 minutes at room temperature with 100 ml of phosphatase-conjugated goat anti-mouse IgG (Cappel, Cochranville, PA, USA). The secondary antibody was used at a dilution of 1:1.000 in HBSS-FBS 10%. After two washes with HBSS, 100 ml of the enzymatic substrate (p-nitrophenylphosphate in 0.05 M Mg-carbonate buffer pH 9.8, Sigma-Aldrich) were added. The optical density (OD) values were evaluated at 405 nm after 30 minutes of incubation by a semiautomatic reader (Titertek Multiskan MCC/340, Titertek Instruments Inc, Pforzheim, Germany).
IL-6, IL-8, and TGF-b1 protein secretion.
IL-6, IL-8 and TGF-b1 release was evaluated in culture supernatants after 24-hour incubation with SSc-ICs, PAPS-ICs, SLE-ICs, NHS-ICs or agonists [IL-1b and LPS] by commercial ELISAs (R&D Systems).
Fcg receptor expression.
The expression of FcgRI (CD64), FcgRII (CD32) and FcgRIII (CD16) was measured on HUVECs by flow cytometry after subtraction of background signals. Briefly, cells were detached by trypsin/EDTA and washed; 200.000 cells per tube were incubated for 20 minutes at 4°C with FITC mouse monoclonal anti-human CD64 (ThermoFisher), FITC mouse monoclonal anti-human CD32 (Beckman Coulter, Brea, CA, USA), FITC mouse monoclonal anti-human CD16 (BD Pharmingen, San Diego, CA, USA) or mouse anti-human isotype control (BD Biosciences, San Jose, CA). Samples were washed again and suspended in 250 μL cold DPBS/1% FCS. 10.000 events were acquired at medium flow rate. A single fluorochrome dot plot strategy was used to identify FcgR positive endothelial cells. In the experiments, control procedures to establish proper calibration and linearity were performed. Analyses were performed using BD FACSLyric cytometer and BD FACSuite software (BD Biosciences).
tlr2, tlr3, tlr4, tlr7, tlr8 and tlr9,interferon-a,interferon-b, endothelin-1, matrix metalloproteinase-1 and collagenIa1 mRNA expression levels.
tlr2, tlr3, tlr4, tlr7, tlr8, tlr9,interferon (ifn)-a, ifn-b and endothelin (et)-1 mRNA expression levels were evaluated in HUVECs stimulated for 24 hours with SSc-ICs, PAPS-ICs, SLE-ICs, NHS-ICs or agonists (LPS, Poly I:C and ODN CpG [5 mM, InvivoGen, San Diego, CA, USA]). mmp-1 and colIa1 were evaluated in healthy skin fibroblasts stimulated for 24 hours with supernatants from HUVECs treated with SSc-ICs, NHS-ICs or recombinant human TGF-b1 (10 ng/ml, PreproTech, Rocky Hill, JN, USA). Total RNA from cells was purified using Trizol Reagent (ThermoFisher Scientific). Amplification Grade DNase I (ThermoFisher Scientific) was used to eliminate residual genomic DNA. A reverse transcription reaction was performed using SuperScriptTM First-Strand Synthesis System for RT-PCR (ThermoFisher Scientific). Universal PCR Master Mix No AmpErase UNG (ThermoFisher Scientific) was used for Quantitative RT-PCR, by ABIPRISM 7900 HT Sequence Detection System (ThermoFisher Scientific). Quantification of mRNA expression was performed with TaqMan® Gene Expression Assay (ThermoFisher Scientific) for each target gene. The following TaqMan® Gene Expression assays were used: Hs01872448_s1 (tlr2), Hs01551078_m1 (tlr3), Hs00152939_m1 (tlr4), Hs01933259_s1 (tlr7), Hs00152972_m1 (tlr8), Hs00370913_s1 (tlr9), Hs00855471_g1 (ifn-a), Hs01077958_s1 (ifn-b), Hs00174961_m1 (et-1), Hs00164004_m1 (colIa1), Hs00899658_m1 (mmp-1) and Hs99999905_m1 (gapdh). Expression levels of target genes (tlr2, tlr3, tlr4, tlr7 tlr8 and tlr9, ifn-a, ifn-b, et-1, mmp-1 and coIa1) were determined by the comparative Ct method normalizing the target to the endogenous gene (gapdh). Relative values of target to reference were expressed as Fold change (RQ). The optimal time point to evaluate the mRNA levels of colIa1 was set at 24 hours based on a kinetics curve of the mRNA response to stimulation with TGF-b.
IL-6 and a-SMA protein expression and nuclear factor k B, p38 mitogen activated kinase, SAPK-JNK and Akt activation rate.
IL-6 and a-SMA protein expression was evaluated in fibroblasts stimulated for 24 hours with TNF-a and supernatants from HUVECs treated with SSc-ICs, NHS-ICs and TNF-a. The activation rate of nuclear factor k B (NFkB), p38 mitogen activated kinase (p38MAPK), SAPK-JUN N terminal kinase (JNK) and RAC-a serine/threonine protein kinase (Akt) was assessed in HUVECs incubated for 24 hours with SSc-ICs, PAPS-ICs, SLE-ICs, NHS-ICs and IL-1b. Total proteins were isolated using RIPA Lysis Buffer added with Protease and Phosphatase inhibitor cocktail (Sigma-Aldrich). Protein concentration was evaluated using BCA Protein Assay Kit (ThermoFisher Scientific). Proteins were fractionated by NuPAGE BIS-TRIS by 4-12% SDS-polyacrylamide pre-cast gel electrophoresis and transferred to nitrocellulose using iBlot Transfer Stacks Nitrocellulose (ThermoFisher Scientific). Membranes were blocked for 2 hours at room temperature in PBS/0.05% Tween 20 (PT) (Bio-Rad Laboratories, Hercules, CA, USA) containing 5% non-fat milk powder (Mellin, Milan, Italy), and incubated with anti-human IL-6 (Cell Signaling Technology, Danvers, MA, USA), anti-a smooth muscle actin (a-SMA, Sigma-Aldrich), anti-human α-tubulin (Sigma-Aldrich), anti-human nuclear factor k B (NFkB), anti-human phosphorylated NFkB (pNFkB), anti-human p38 mitogen activated kinase (p38MAPK), anti-human phosphorylated p38MAPK (pp38MAPK), anti-human SAPK-JUN N terminal kinase (JNK) or anti-human phosphorylated SAPK-JNK (anti-pSAPK-JNK) antibodies, anti-human RAC-a serine/threonine protein kinase (Akt) or anti-human phosphorylated Akt (anti-pAkt, Cell Signaling Technology, Danvers, MA, USA). After washes, membranes were incubated in PT/5% non-fat milk powder plus HRP-conjugated secondary antibodies (MP Biomedicals, Santa Ana, CA, USA) and developed using ECL Plus Detection System (ThermoFisher Scientific). Signals were detected using radiographic films (Kodak, Rochester, NY, USA). Image J software (LI-COR Biosciences, Lincoln, NE, US) was used to analyze and quantify gels. Proteins expression levels of IL-6 and a-SMA were normalized to the housekeeping gene, α-tubulin, and expressed as relative protein levels.
An activation kinetics to evaluate the phosphorylation of intracellular mediators in response to different ICs has been performed at different time-points: 0 – 6 – 24 – 36 hours. At 6 hours, the activation rate of all intracellular study mediators was low. In all cases, the highest activation levels were observed at 24 and 36 hours. However, we observed a decrease in cell viability at 36 hours. Therefore, cells were incubated for 24 hours with the different stimuli.
Bafilomycin pre-treatment.
HUVECs were incubated for 2 hours at 37°C with bafilomycin A1 (Millipore, Temecula, CA, USA, 100 nmol) and then treated with SSc-ICs, PAPS-ICs, NHS-ICs and Poly I:C. RT-PCR for il-6 and et-1 was then performed.
IL-6 and TGF-b inhibitors.
Fibroblasts were pre-incubated for 30 minutes at 37° C with inhibitors of IL-6 (rat anti-human monoclonal antibody, ThermoFisher Scientific Inc; 2.5 mg/mL) and TGF-b (mouse anti-human monoclonal antibody, R&D Systems; 5 mg/mL). Pre-treated fibroblasts were then stimulated with supernatants from HUVECs incubated with ATA-ICs and with recombinant TGF-b1 as positive control. The mRNA levels of et-1 were evaluated.
Statistical analysis.
Descriptive statistics was used to calculate mean and standard deviation (SD). Since our data were derived from in vitro experiments conducted under high controlled conditions and originated from a high number of cells, ANOVA test was used to compare different experimental conditions, post-hoc comparisons were assessed by Dunnett’s test. With regards to not homogeneity of variance assumption, Welch’s correction was applied when required. Paired or unpaired t-tests were performed to compare mean values between two groups. All analyses were performed with GraphPad Prism 5.01. P-values <0.05 were considered significant.
The approval of the Institutional Review Board of Istituto G. Pini, Milan, Italy was obtained; all subjects provided written informed consent.