Liver X Receptor Regulates Rheumatoid Arthritis Fibroblast-like Synoviocyte Invasiveness, Matrix Metalloproteinase 2 Activation, Interleukin-6 and CXCL10
- 6 Downloads
Fibroblast-like synoviocyte (FLS) invasiveness correlates with articular damage in rheumatoid arthritis (RA), yet little is known about its regulation. In this study we aimed to determine the role of the nuclear receptor liver X receptor (LXR) in FLS invasion. FLS were isolated from synovial tissues obtained from RA patients and from DA rats with pristane-induced arthritis. Invasion was tested on Matrigel-coated chambers in the presence of the LXR agonist T0901317, or control vehicle. FLS were cultured in the presence or absence of T0901317, and supernatants were used to quantify matrix metalloproteinase 1 (MMP-1), MMP-2, MMP-3, interleukin-6 (IL-6), tumor necrosis factor-α and C-X-C motif chemokine ligand 10 (CXCL10). Nuclear factor-κB (NF-κB) (p65) and Akt activation, actin cytoskeleton, cell morphology and lamellipodia formation were also determined. The LXR agonist T0901317 significantly reduced DA FLS invasion by 99% (P ≤ 0.001), and RA FLS invasion by 96% (P ≤ 0.001), compared with control. T0901317-induced suppression of invasion was associated with reduced production of activated MMP-2, IL-6 and CXCL10 by RA FLS, and with reduction of actin filament reorganization and reduced polarized formation of lamellipodia. T0901317 also prevented both IL-1β-induced and IL-6-induced FLS invasion. NF-κB (p65) and Akt activation were not significantly affected by T0901317. This is the first description of a role for LXR in the regulation of FLS invasion and in processes and pathways implicated both in invasion as well as in inflammatory responses. These findings provide a new rationale for considering LXR agonists as therapeutic agents aimed at reducing both inflammation and FLS-mediated invasion and destruction in RA.
Rheumatoid arthritis (RA) is a chronic autoimmune disease that affects 0.5–1% of the population and is associated with increased risk for joint deformities, disability, and reduced longevity (1,2). The RA synovial tissue is characterized by synovial hyperplasia, also called “pannus,” which is infiltrated with inflammatory cells. The RA synovial pannus produces proinflammatory cytokines, chemokines and proteases, and invades and destroys cartilage and bone (3). The fibroblast-like synoviocyte (FLS) has a central role in the formation of the RA synovial pannus and in joint destruction (3,4).
The in vitro invasive properties of FLS from patients with RA and from rats with pristane-induced arthritis (PIA) through collagen-rich Matrigel have been shown to correlate with radiographic erosive changes and with histological joint damage, respectively (5,6). Erosive changes and joint damage correlate with worse disease outcome, including increased risk for disability and for the development of deformities (7, 8, 9). Therefore, understanding of the processes and genes regulating FLS invasion has the potential to generate new targets for therapies aimed at reducing articular damage as well as improving disease outcome.
We have recently determined that synovial tissues from arthritic DA rats, a strain that develops severe and erosive PIA, have significantly reduced expression of nuclear receptors (NRs) compared with synovial tissues from rats with mild and nonerosive disease (10). Additionally, DA rats have reduced expression of NR target genes, suggesting not only reduced NR expression but also reduced NR activity. NRs are a family of ligand-activated transcription factors that regulate several cellular processes, including lipid metabolism, calcium metabolism and, importantly for arthritis, inflammatory responses (11). Members of the NR family include the vitamin D receptor (VDR), retinoid X receptor γ (RXRγ) and the liver X receptor α (LXRα), among others (12). Therefore, we hypothesized that the increased expression and activity of one or more NRs in synovial tissues might have a protective effect in maintaining inflammation-free and noninvasive synovial tissue and FLS (13).
LXRα can be stimulated by endogenous ligands such as oxysterols and forms heterodimers with RXRγ. The LXRα-RXRγ dimer binds to LXR-responsive elements located in the promoter of gene targets to modulate gene expression. LXRα is of particular interest for four additional reasons. First, LXR agonists have been shown to ameliorate several rodent models of inflammation and autoimmunity, such as lung inflammation (14), acute dermatitis (15), experimental allergic encephalomyelitis (EAE) (16) and murine lupus (17). Second, LXR agonists reduce macrophage expression of proinflammatory cytokines (15). Third, LXR agonists successfully treated collagen-induced arthritis (CIA) in mice, and reduced disease severity and articular damage via unknown mechanisms (18,19). And fourth, DA synovial tissues expressed increased levels of CYP7b1, which is an enzyme that inactivates oxysterols, the natural ligands and activators of LXR synthesized from cholesterol (13), suggesting reduced levels of activating ligands for LXRα in arthritis-susceptible DA synovial tissues.
In the present study we investigated the potential role of LXR in the regulation of the invasive properties of FLS from patients with RA and DA rats.
Materials and Methods
Rats and the Induction of PIA
Arthritis-susceptible DA rats (DA/Hsd) (8–12 wks old) were housed in a specific pathogen-free environment. For the induction of PIA, DA rats received 150 µL of pristane by intradermal injection at the base of the tail (20). All rats developed PIA and were euthanized on d 21 after induction, and synovial tissues were collected from the ankle joints for the isolation of FLS. All animal work was approved by the Feinstein Institute’s institutional animal care and use committee.
Synovial tissues were obtained from RA patients who met the 1987 American College of Rheumatology criteria (21) and were undergoing an elective orthopedic surgery. All patients signed an informed consent form under a protocol approved by the Feinstein Institute’s institutional review board.
Isolation and Culture of FLS
FLS were isolated by use of enzymatic digestion from synovial tissues from DA rats and from RA patients, as previously described (5,22). Briefly, tissues were minced and incubated with a solution containing 0.15 mg/mL DNase, 0.15 mg/mL hyaluronidase (type I-S) and 1 mg/mL collagenase (type IA) (Sigma-Aldrich, St. Louis, MO, USA) in Dulbecco’s modified Eagle’s medium (DMEM) (Invitrogen, Carlsbad, CA, USA) for 1 h at 37°C. Cells were washed and resuspended in complete medium containing DMEM with 10% fetal bovine serum (FBS), glutamine 30 ng/mL, amphotericin B 250 µg/mL and gentamycin 20 ng/mL (Invitrogen) for culture. All experiments were done with cells cultured for at least four passages (>95% FLS purity).
The T0901317 LXR Agonist
T0901317 (Cayman Chemicals, Ann Arbor, MI, USA) is a synthetic agonist that specifically binds and activates both LXRα and β at 1 µmol/L and 10 µmol/L concentrations (15,23). These concentrations were used in the present study. T0901317 was reconstituted in dimethyl sulfoxide (DMSO), and DMSO was used as vehicle control in all experiments.
MTT Survival Assay
FLS were plated in triplicate at 1 × 104 cells per well in 96-well flat-bottomed plates and allowed to adhere for 24 h in complete medium (described above). FLS were treated with T0901317 at 10 µmol/L or DMSO in serum-free medium for 24 h. Cell survival was determined by using the colorimetric 3-(4),5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay (Millipore, Billerica, MA, USA) according to manufacturer’s instructions.
FLS Invasion Assay
We tested the effect of T0901317 on FLS invasion using an in vitro trans-well system through collagen-rich Matrigel (BD, Franklin Lakes, NJ, USA) as previously described (5,24). Briefly, 70–80% confluent FLS were harvested by trypsin-EDTA digestion. Then 2 × 104 cells were resuspended in 500 µL of serum-free DMEM and plated in the upper compartment of the Matrigel-coated inserts. T0901317 (final concentration of 1 µmol/L or 10 µmol/L) or the same amount of the control solvent DMSO was added to the upper chambers. The lower compartment was filled with complete media. After 24 h incubation at 37°C the supernatant in the upper chamber was collected, and the upper surface of the insert was scraped with cotton swabs to remove noninvading cells and the Matrigel layer. The opposite side of the insert was stained with Crystal Violet (Sigma, Saint Louis, MO, USA) and the total number of cells that invaded through Matrigel was counted at 100× magnification. Experimental treatments were done in duplicate.
Quantitative Real-Time PCR
FLS RNA was extracted with the RNeasy Mini Kit (Qiagen, Valencia, CA, USA). Samples were digested with DNase (Qiagen) and eluted with 30 µL of RNase-free water. RNAs were quantified and assessed for purity with a NanoDrop spectrophotometer (Rockland, DE, USA). We used 200 ng of total RNA from each sample for cDNA synthesis with the Superscript III kit (Invitrogen).
Quantitative real-time PCR (qPCR) was done as previously described (5). Briefly, we used the Universal ProbeLibrary (Roche, Indianapolis, IN, USA). Probes were used at a final concentration of 250 nmol/L. We designed primers targeting rat and human genes with the Universal ProbeLibrary Assay Design Center (Roche), and used a 400-nmol/L concentration with Absolute Blue QPCR Master Mix (Thermo Scientific, Surrey, UK). Samples were run in duplicate on a Roche 480 qPCR thermocycler, and the means were used for analysis. Data were analyzed with LC480 software version 1.5 (Roche). Relative expression of all the genes was adjusted for GAPDH (glyceraldehyde-3-phosphate dehydrogenase) in each sample (ΔCt), and the ΔCt was used for t test analysis. A P value ≤0.01 was considered significant. Fold-differences were calculated with the 2−ΔΔCt method (25).
FLS were cultured for 24 h in the upper chambers of the Matrigel-coated inserts as part of the invasion experiments. Supernatant from each upper chamber was collected and concentrated with Microcon YM30 columns (Millipore). Equal amounts of concentrated supernatant protein was mixed with Tris-glycine-sodium dodecyl sulfate sample buffer (Invitrogen), then loaded into a precasted zymogram gel (Invitrogen) and run for 90 min at 125 V, as previously described (5). Gels were then treated with renaturing buffer (Invitrogen), followed by overnight incubation in developing buffer (Invitrogen) at 37°C. Gels were stained with SimplyBlue Safe Stain (Invitrogen) for 1 h at room temperature and washed. Gelatin zymography was used to assess MMP-2, and casein zymography was used for MMP-3 (5,26).
Cytometric Bead-Based Cytokine Detection
Supernatants from the upper chambers of RA FLS invasion experiments were concentrated with Microcon columns (Millipore) and analyzed for total protein content using a bicinchoninic acid (BCA) protein assay kit (Thermo Scientific). Concentrations of IL-6, IL-8, IL-10, TNFα, monocyte chemotactic protein-1 (MCP-1) and CXCL10 were determined with a bead-based detection system (Cytometric Bead Array, CBA, BD Bioscience, San Jose, CA, USA) according to the manufacturer’s protocol.
MMP-1. Equal volumes of the upper-chamber supernatants (as described in the zymography section) were loaded on a NuPAGE 10% Bis-Tris gel (Invitrogen) in the presence of 2-(N-morpholino) ethanesulfonic acid (MES) buffer (Invitrogen) and run under reducing conditions. Proteins were transferred overnight to a polyvinylidene difluoride (PVDF) membrane (Immobilion, Millipore). The membrane was then blocked with 5% blottinggrade nonfat dry milk (Bio-Rad, Hercules, CA, USA) and incubated with mouse monoclonal anti-MMP-1 antibody (Calbiochem, EMD, La Jolla, CA, USA). Horseradish peroxidase-conjugated anti-mouse IgG antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA) was used as the secondary antibody.
Akt and phospho-Akt. For detection of Akt, FLS were plated at 70–80% confluence in 6-well plates and allowed to adhere for 24 h. Cells were then serum starved for 24 h and pretreated with T0901317 10 µmol/L or DMSO (control) for 2 h, followed by 5-min culture with IL-1β 10 ng/mL. Total cell lysates were collected with radioimmunoprecipitation assay buffer (RIPA, Thermo Scientific) containing Halt Protease & Phosphatase inhibitor cocktail (Thermo Scientific). Protein content was determined with a BCA protein assay kit (Thermo Scientific) and the same amount of total proteins was loaded on a NuPAGE 10% Bis-Tris gel (Invitrogen) in the presence of MES buffer (Invitrogen) and in reducing conditions. Proteins were transferred to PVDF membranes, blocked with 5% nonfat milk, then probed with antibodies against the phosphorylated form of Akt (Cell Signaling, Danvers, MA, USA). Horseradish peroxidase-conjugated anti-mouse IgG (Santa Cruz Biotechnologies, Santa Cruz, CA, USA) was used as secondary antibody for 1 h at room temperature in Tris-buffered saline with 0.1% Tween-20. Protein bands were detected by Amersham ECL plus (GE Healthcare, Buckinghamshire, UK) and visualized using X-OMAT Kodak film. Following the analyses of phospho-Akt, membranes were stripped and reprobed with a rabbit antibody against total Akt (Cell Signaling) and an anti-rabbit IgG (Amersham, GE Healthcare) used as secondary antibodies. Densitometry was done with Adobe Photoshop, version 7.0 (San Jose, CA, USA).
LXRα and LXRβ. DA FLS were stimulated and cultured under the same conditions described above for Akt in the presence or absence of DMSO (control), T0901317 10 µmol/L and IL-1β 10 ng/mL for 24 h, 48 h or 72 h, followed by cell lysis and protein extraction. A rabbit polyclonal antibody that recognizes both LXRα and LXRβ (Santa Cruz Biotechnology) was used, and a monoclonal antibody against human Vinculin (Sigma) was used as control.
p65 Nuclear Factor-κB Activity
DNA-binding activity of the nuclear factor-κB (NF-κB) protein p65 was quantified in nuclear extracts using the TransAM™ NF-κB (p65) (Active Motif, Carlsbad, CA, USA) according to the manufacturer’s instructions. Briefly, FLS were pretreated for 20 h with T0901317 10 µmol/L or DMSO and then stimulated with IL-1β 10 ng/mL for 30 min.
Confluent FLS (10–20%) were cultured on cover slips, starved overnight and then treated with either DMSO or T0901317 10 µmol/L in complete media for 20 min. FLS were then fixed with 4% formaldehyde (Ted Pella Inc., Redding, CA, USA) and permeabilized with phosphate-buffered saline (PBS)-Triton X-100 0.1%, and blocked with 5% nonfat milk. Cells were incubated with rabbit antibodies against phosphorylated focal adhesion kinase-1 (FAK) (Cell Signaling) for 1 h at room temperature, washed with PBS with 0.1% Triton X-100, and then incubated with TexRed-conjugated anti-rabbit IgG antibody (Jackson ImmunoResearch, West Grove, PA, USA) for 30 min, and Phalloidin-fluorescein isothiocyanate (FITC) (Sigma-Aldrich) used to stain actin filaments. Stained and washed cells were mounted and observed under fluorescent microscopy (Zeiss Axiovert 200M with Zeiss software Axioversion 4.7).
Actin Cytoskeleton, Lamellipodia and Phospho-FAK Scoring System
To quantify the potential changes induced by T0901317 in FLS morphology, we recently developed a new scoring system (27), which includes a previously established actin scoring system (28), plus the addition of scores for the following major parameters of interest: (a) actin filament characteristics and distribution (0–3); (b) cell morphology (0 or 2); (c) lamellipodia location (0 or 2); and (d) distribution of phosphorylated FAK (0–2). The cell scoring ranged from zero to nine, and 20 cells per treatment group and per cell line were analyzed.
All supplementary materials are available online at www.molmed.org .
LXR is Expressed and Functional in Arthritic DA FLS
LXRα and LXRβ proteins were expressed by FLS, but, unlike mRNA levels, protein levels did not change following IL-1β treatment in the presence or absence of T0901317 during 24-72-h experiments. These results suggest that these proteins have a long half-life, and that changes in their levels of protein did not explain the effect of T0901317 on FLS invasion over a 24-h period (Figure 1C).
IL-6 and IL-17 did not cause the same effects as IL-1β on LXR. IL-6 treatment increased LXRα expression by 3.5-fold, and IL-6 plus T0901317-treated FLS continued to have a 1.85-fold increase in LXRα expression compared with controls (Supplementary Figure S1A). IL-17 treatment did not affect the mRNA levels of LXRα, but the addition of T0901317 was still able to increase the expression of LXRα by 2.8-fold (Supplementary Figure S1B). LXRβ mRNA levels were not affected by IL-6 or IL-17 treatment (Supplementary Figures S1A, B).
The LXR Agonist T0901317 Significantly Reduced the Invasive Properties of FLS from DA Rats and from RA Patients
The MTT assay showed that T0901317 at 1 µmol/L or 10 µmol/L concentrations did not affect FLS numbers or viability during the same period of time used for the invasion experiments (24 h) (Figure 2C). Therefore, the reduced invasion of FLS treated with T0901317 was unlikely due to drug-induced cell toxicity.
T0901317 Reduced the Levels of Activated MMP-2 in RA FLS
Levels of total and activated forms of MMP-1 and MMP-3 measured in the same FLS supernatants by Western blot and zymography, respectively, were not significantly affected by T0901317 (data not shown).
T0901317 Did Not Affect NF-κB (p65) Activation
DA FLS were stimulated with IL-1β (10 ng/mL) to induce increased NF-κB activation. However, pretreatment with T0901317 did not consistently reduce the activation of p65 (data not shown).
T0901317 Did Not Affect Akt Phosphorylation
FLS were treated with FBS 10% to induce Akt activation and phosphorylation. Pretreatment with T0901317 did not significantly reduce the phosphorylation of Akt (n = 3; data not shown).
Stimulation of LXR Inhibits FLS Morphological Changes Required for Polarized Migration and Invasion
The LXR Agonist T0901317 Reduces the Expression of IL-1 β, IL-6, CXCL10 and MMP-3
Next, supernatants from RA FLS cultured without serum on Matrigel (upper compartment of the invasion chambers) were used to quantify IL-6, IL-8, IL-10, MCP-1, TNFα and CXCL10 protein levels. These culture supernatants were chosen because they provide a direct correlation with the invasion studies. Similar to results from the invasion studies, treatment with the LXR agonist T0901317 significantly reduced levels of IL-6 by 74% (P = 0.0000002, Figures 5D, E) and levels of CXCL10 by 60% (P = 0.004, Figures 5F, G), respectively. MCP-1, IL-10 and TNFα were nearly undetectable. Levels of IL-8 were not significantly changed by T0901317 (data not shown). Therefore, our results suggest that the LXR agonist affects mRNA expression and protein synthesis of key mediators of both inflammatory responses in arthritis, as well as mediators of FLS invasion and articular damage.
LXR Stimulation with T0901317 Prevents IL-1β and IL-6-Induced FLS Invasion
Treatment with T0901317 completely prevented IL-1β and IL-6-induced increases in FLS invasion, suggesting that LXR interferes with signaling or effector pathways regulated by the receptors for these cytokines (Figure 6). T0901317-treated cells were less invasive than controls, suggesting that, in addition to interfering with IL-1β and IL-6 activity, T0901317 also interferes with other invasion-mediating processes.
Although the LXR agonist reduced IL-1β and IL-6 expression, our results demonstrated that its effect on invasion was not dependent on the suppression of cytokine expression as it retained its invasion-suppressing effect despite cytokine treatment/supplementation.
Synovial tissue hyperplasia and tissue invasion and destruction of cartilage and bone are characteristic findings of RA (3), and FLS have a central role on these processes (4). Cartilage, bone and ultimately joint destruction correlate with disease severity and with increased risk for deformities and disability, yet little is known about the genes regulating these processes. Previous studies have shown that the in vitro invasive properties of FLS through collagen-rich Matrigel correlate with histologic damage in rats with PIA (5), and with radiographic erosions and damage in patients with RA (6). Therefore, this model has direct clinical correlation and relevance, and we consider it an important and useful strategy to understand the processes and genes regulating FLS invasion, and to identify new potential targets for therapies aimed at preserving joint architecture and reducing articular damage.
We have previously determined that the expression of the antiinflammatory NR LXRα is decreased in synovial tissues from arthritic rats (10). In the present study we show that IL-1β can reduce the expression of LXRα in FLS, providing a new explanation for the reduced expression and activity of LXRα in vivo.
Treatment with the LXRα agonist T0901317 significantly reduced the invasive properties of FLS from arthritic rats and from patients with RA by more than 90%. Our results demonstrate that the LXR-mediated suppression of invasion involves: (a) reducing the production of the active form of MMP-2, a protease implicated in FLS invasion (31) and radiographic damage in RA (32); (b) reducing levels of MMP-3, another protease implicated in RA joint damage (33) and FLS invasion (24); and (c) interfering with actin cytoskeleton reorganization, with FLS ability to form lamellipodia in a polarized manner and with colocalization of phospho-FAK with lamellipodia, all required for cell mobility and invasion (22,31,34,35).
The LXR agonist T0901317 also reduced the expression of cytokines central to RA pathogenesis and joint damage such as IL-1β (36, 37, 38) and IL-6 (39,40). IL-1β and IL-6 increase the production of several MMPs (41), and IL-1β directly increases FLS and synovial tissue invasion and destruction of cartilage (37, 38). Furthermore, the synovial levels of IL-1β are predictive of joint damage progression in RA (42). IL-6 had not been previously studied in the regulation of FLS invasion, but it was a reasonable possibility given its proinvasive and prometastatic effects in cancers (43). This is the first study to show that IL-6 increases the invasive properties of RA FLS. Therefore, our results suggest that IL-1β and IL-6 may have mutually suppressive and antagonistic effects against LXRα, and that LXR agonists are capable of preventing the increased FLS invasiveness induced by these cytokines.
CXCL10 levels are increased in RA (44,45), and antibodies targeting this chemokine ameliorated disease activity in patients (46) and dramatically reduced cartilage and bone damage in rodent arthritis (47). We have recently determined that in addition to its chemotactic activity for CXCR3-positive mast cells and T cells, CXCL10 increases invasion in an autocrine and paracrine manner in FLS from arthritic rats and patients with RA (27). Therefore, the reduction in levels of CXCL10 induced by the LXR agonist T0901317 provides yet another mechanism to reduce FLS invasion.
LXR has a known direct effect on gene expression via its binding to LXR-responsive elements that might explain part of our observations. Yet, there is some suggestion from the literature that LXR has nongenomic effects, potentially interfering with signaling pathways such as NF-κB in brain tissues (48) and Akt in prostate cancer cells (49). We studied the effect of T0901317 on FLS NF-κB (p65) and Akt activation, but, similarly to other investigators (14), we did not observe consistent suppression, suggesting that some of the LXR nongenomic effects are cell or tissue specific.
LXR agonists have been successfully used by two different groups to reduce susceptibility, disease severity and articular damage in collagen-induced arthritis in mice (18,19), although a third group unexpectedly found opposite results (50). In addition to suppressive effects on FLS invasion and cytokine/chemokine and MMP production described here, others have reported that LXR agonists suppress the differentiation of Th17 T cells (16), which are central to the pathogenesis of RA and other autoimmune diseases. LXR agonists also significantly improved another Th17-mediated disease, EAE, which is a model of multiple sclerosis (16). Our results combined with other data reported in the literature suggest that the use of LXR agonists may provide a beneficial effect that involves interference with processes mediating both FLS invasion and articular damage, as well as regulation of proinflammatory mediators and the development of pathogenic Th17 cells.
In the present study we determined that LXR is functional and responsive in DA FLS and in RA FLS, and treatment with a synthetic agonist significantly reduces FLS invasion. Our observations provide a new mechanistic explanation for the disease-ameliorating and joint-protecting effects of LXR agonists, including the inhibition of IL-1β- and IL-6-induced effects on FLS, and provide a new rationale for considering clinical trials with this group of drugs in patients with RA.
The authors declare that they have no competing interests as defined by Molecular Medicine, or other interests that might be perceived to influence the results and discussion reported in this paper.
The authors thank M Keogh and C Mason of the Feinstein Institute’s Tissue Donation Program for their assistance in obtaining RA synovial tissues. Funding was provided by the National Institutes of Health grants R01-AR46213, R01-AR052439 (NIAMS) and R01-AI54348 (NIAID) to P Gulko.
- 36.Jiang Y, et al. (2000) A multicenter, double-blind, dose-ranging, randomized, placebo-controlled study of recombinant human interleukin-1 receptor antagonist in patients with rheumatoid arthritis: radiologic progression and correlation of Genant and Larsen scores. Arthritis Rheum. 43:1001–9.CrossRefGoogle Scholar
- 46.Yellin M, et al. (2009) A phase II, randomized, double-blind, placebo-controlled study evaluating the efficacy and safety of MDX-1100, a fully human anti-CXCL10 monoclonal antibody, in combination with methotrexate in patients with rheumatoid arthritis. Arthritis Rheum. 60(Suppl 10):414.Google Scholar
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, and provide a link to the Creative Commons license. You do not have permission under this license to share adapted material derived from this article or parts of it.
The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
To view a copy of this license, visit (http://creativecommons.org/licenses/by-nc-nd/4.0/)