Endothelial activation contributes to lung inflammatory disorders by inducing leucocyte recruitment to pulmonary parenchyma. Consequently, vascular-targeted therapies constitute promising strategies for the treatment of inflammatory pathologies. In the present study, we evaluated the effect of 8,9-dehydrohispanolone-15,16-lactol diterpene (DT) on lung endothelium during inflammation. Lung endothelial cells pre-treated with DT and activated with lipopolysaccharide (LPS) or tumour necrosis factor-α (TNF-α) exhibited reduced expression of the pro-inflammatory cytokines Cxcl10, Ccl5 and Cxcl1, whereas the anti-inflammatory molecules IL1r2 and IL-10 were induced. Consistent with this result, DT pre-treatment inhibited nuclear factor κB (NF-κB) nuclear translocation, by interfering with IκBα phosphorylation, and consequently NF-κB transcriptional activity in endothelium activated by LPS or TNF-α. Furthermore, DT, probably through p38 signalling, induced transcriptional activation of genes containing activator protein 1 (AP-1)-binding elements. Inhibition of p38 prevented IL1r2 mRNA expression in endothelium incubated with DT alone or in combination with LPS or TNF-α. Accordingly, conditioned medium (CM) from these cells failed to stimulate leucocytes as measured by a reduction in adhesive ability of the leucocyte cell line J774 to fibronectin (FN). Additionally, DT reduced the expression of the endothelial adhesion molecules E-selectin, vascular cell adhesion molecule 1 (VCAM-1) and intercellular adhesion molecule 1 (ICAM-1) after activation. Similarly, expression of VCAM-1 and ICAM-1 molecules on the lung endothelial layer of C57/BL6 mice pre-treated with DT and challenged with LPS were unchanged. Finally, inhibition of vascular adhesion molecule expression by DT decreased the interaction of J774 cells with lung endothelial cells in an inflammatory environment. Our findings establish DT as a novel endothelial inhibitor for the treatment of inflammatory-related diseases triggered by Gram-negative bacteria or by the associated cytokine TNF-α.
Inflammation is a central feature of many lung pathologies including acute lung injury (ALI), acute respiratory distress syndrome (ARDS) and chronic obstructive pulmonary disease (COPD) [1–3]. Although several factors may influence the development of these diseases, inappropriate immune cell sequestration into lungs as a consequence of sepsis appears to be a key event . Vascular endothelium plays a critical role in regulating leucocyte recruitment in the inflammatory environment . Upon induction by a variety of stimuli, including the pro-inflammatory cytokine tumour necrosis factor-α (TNF-α) and the bacterial endotoxin lipopolysaccharide (LPS), resting endothelium becomes activated and expresses additional pro-inflammatory factors and cell-specific adhesion molecules that enable extravasation of circulating leucocytes to sites of inflammation. Circulating lymphocytes are initially activated by endothelially released chemotactic factors; subsequently, leucocyte rolling occurs on the endothelial surface through the binding of selectins (L-selectin on leucocytes and E- and P-selectins on endothelium) to carbohydrate moieties on the contacting cell, such as PSGL-1. Firm adhesion is maintained by vascular cell adhesion molecule 1 (VCAM-1) and intercellular adhesion molecule 1 (ICAM-1) on the endothelium, which bind previously activated β1- and β2-integrins on the leucocyte respectively. A consequence of this sequential process is that blockade of any of the steps should interrupt extravasation, thus regulating the inflammatory-related disorder. The development of strategies which target endothelial activation to prevent leucocyte–endothelium interaction may therefore constitute novel therapeutic options to treat inflammatory disorders .
Traditional anti-inflammatory therapies, such as corticosteroids and non-steroidal anti-inflammatory drugs, have not provided net clinical benefits for inflammatory lung diseases such as ARDS and COPD [6,7]. The discovery of new bioactive natural products is considered a very promising starting point for the development of therapeutic agents. One of the most important families of natural products known for their medicinal value is the terpenoids. Labdanes, belonging to the bicyclic diterpenoids group, show a broad spectrum of biological properties including antibacterial, anti-inflammatory and anti-tumour activities . The labdane hispanolone, isolated from Ballota hispanica, and the structurally related prehispanolone exhibit anti-inflammatory activity and very low cytotoxicity [9–12]. From this family of compounds, the chemically synthesized diterpenoid derived from hispanolone, 8,9-dehydrohispanolone-15,16-lactol diterpene (DT), has been shown to exert anti-inflammatory effects on macrophages through a mechanism which involves inhibition of nuclear factor κB (NF-κB) .
Given this information, we hypothesized that DT might act as a regulator of endothelial cell (EC) participation during inflammation. The well-recognized role of NF-κB in signalling pathways underlying vascular inflammation prompted us to examine the effect of DT during the initial events of immune cell activation. To ascertain novel regulatory functions exerted by DT, we extended our study to a panel of luciferase reporters designed to monitor transcription factor-binding activities. Our results reveal a new function for DT as an inhibitor of endothelial activation during inflammation.
Cells and antibodies
The mouse lung endothelial cell line MLEC-04 has been previously described in detail . MLEC-04 cells are an immortalized EC line that exhibits stable vascular phenotype and reproducible activation ability [15–17]. Mouse monocyte/macrophage J774 cells were purchased from the European Collection of Cell Cultures (ECACC) and cultured following the manufacturer's recommendations. The antibodies used in the study were against platelet endothelial cell adhesion molecule-1 (PECAM-1), ICAM-1, E-selectin and VCAM-1 (BD Pharmingen), p-p38, p-IκBα and IκBα (Cell Signaling Technology), β-actin and polypyrimidine tract-binding protein-associated splicing factor (PSF) (Sigma Chemical Company), Cxcl10 (PeproTech), p65 (Santa Cruz Biotechnology) and VCAM-1 (R&D Systems). All antibodies were applied to a final concentration of 10 μg/ml.
Gene expression analysis
Overnight serum-starved MLEC-04 cells were pre-incubated for 30 min at 37°C with 25 μM DT or DMSO as control, followed by 6 h of stimulation with LPS (100 ng/ml, Escherichia coli 0111:B4; InvivoGen) or TNF-α (10 ng/ml; PeproTech). Total RNA was extracted with TRIzol reagent (Invitrogen) and gene expression was evaluated with the mouse RT2 Profiler PCR Inflammatory Cytokines and Receptors Array APMM-011 (SuperArray Bioscience) on an ABI7900 sequence detector (Applied Biosystems).
MLEC-04 cells were treated as before using a range of DT concentrations as specified. Where indicated, p38 was inhibited by initial incubation of the cells with 10 μM SB202190 (Sigma Chemical Company) for 30 min at 37°C. mRNA expression was analysed by real-time PCR (RT-PCR) using SyBr Green chemistry on an ABI7900 sequence detector (Applied Biosystems). Each sample was run in duplicate, and all genes were analysed in parallel against a housekeeping gene 36B4 (acidic ribosomal phosphoprotein P0), which was used to normalize expression levels of target genes. Fold induction was determined from mean replicate values. Primer sequences are available on request from S.H. or A.L.
Transcription factor activity was evaluated using the gene reporter assay kit Cignal Finder Cancer 10 Pathway Reporter Array (CCA-101L; SABiosciences) which included luciferase reporter vectors from TCF/LEF, RBP-Jk, p53, SMAD2/3/4, E2F/DP1, NF-κB, Myc/Max, HIF1α, Elk-1/SRF and activator protein 1 (AP-1); and the luciferases reporter vectors AP-1 Luc (LR0002) and NF-κB Luc (LR0051, Affymetrix). In both cases, MLEC-04 cells were transfected using Lipofectamine 2000 (Invitrogen) following the manufacturer's recommendations. At 24 h after transfection, cells were serum-starved overnight. The next day, cells were pre-incubated for 30 min at 37°C with 25 μM DT or DMSO as control, followed by 6 h of stimulation with LPS (100 ng/ml) or TNF-α (10 ng/ml). Luciferase activity was determined using the Dual Luciferase Reporter Assay System (Promega) on the Infinite M200-Tecan luminometer (Austria) and expressed as relative luciferase units (RLU) per mg of protein lysate. All transfections were performed in triplicates and repeated at least three times.
MLEC-04 cells were treated as above, but were stimulated with LPS or TNF-α for the times indicated. Extraction of nuclear proteins was performed as described in .
Endothelial activation and protein expression were monitored by standard immunoblot analysis of selected targets. Band intensity was quantified using ImageJ software (http://rsbweb.nih.gov/ij/) and normalized to β-actin levels.
Cell-surface antigen expression was analysed by flow cytometry following standard procedures . ECs pre-treated with DT or DMSO were incubated with LPS or TNF-α as described above. Data acquisition was performed in a FACSCalibur flow cytometer (Becton Dickinson).
Nuclear translocation of NF-κB p65 was evaluated in stimulated ECs by confocal microscopy (Leica TCS SP5, Leica Microsystems) . ECs were co-stained with the DNA marker Hoechst 58 (Promega).
Control or DT-pre-treated (1 h before at 20 mg/kg) C57/BL6 mice (four animals per group) were challenged by intraperitoneal injection of LPS (10 mg/kg). DT administration was not toxic in this experimental condition, in agreement with previously reported in vitro and in vivo models performed with derivatives from this family of compounds [13,20]. After 6 h, mice were killed by CO2 inhalation and lungs were aseptically dissected, fixed with 4% paraformaldehyde and paraffin-processed. Tissue sections from paraffin-embedded lungs were immunostained for endothelial markers following standard protocols. Sections were evaluated by light microscopy using an Axio Imager A1 Zeiss microscope. Immunohistochemistry stained area was quantified using ImageJ software (http://rsbweb.nih.gov/ij/). The study was approved by the Comité de Bioética y Bienestar Animal from Instituto de Salud Carlos III and Comunidad de Madrid (PROEX 042/15) and conformed to the guidelines of Directive 2010/63/EU of the European Parliament on the protection of animals used for scientific purpose.
Leucocyte activation assessment by released endothelial factors: cell adhesion assay
Endothelial factors released into the conditioned medium (CM) of ECs stimulated overnight with LPS (100 ng/ml) or TNF-α (10 ng/ml) alone or in combination with DT (25 μM) were evaluated for leucocyte activation in an adhesion assay with J774 cells to fibronectin (FN). Collected CM (100 μl) was added to FN (0.5 μg/ml; Sigma Chemical Company)-coated 96-well plates (Falcon, Becton Dickinson) in triplicate, and 200000 J774 cells were added to each well. After incubation for 1 h at 37°C, the number of adhered cells following three washes with warm PBS was calculated by measuring the absorbance of wells at 540 nm after fixation and staining with 0.5% Crystal Violet in 20% methanol. As a complementary strategy, cells were evaluated by light microscopy.
Leucocyte–endothelium co-adhesion assay
Control or DT-treated endothelial monolayers in 96-well plates were incubated with LPS or TNF-α for 6 h. EC cultures were then washed three times and CFSE (5 mM/20 min; Molecular Probes)-labelled J774 cells were added to each well (200 000 cells per well). After 1 h, wells were washed and fluorescently labelled cells were homogenized with detergent and measured using a spectrofluorometer plate reader (495 nm/517 nm; Infinite M200, Tecan). Alternatively, wells were fixed with 3% paraformaldehyde and evaluated by fluorescence microscopy. Each condition was performed in triplicate.
All data are means ± S.E.M. for a representative experiment performed in triplicate from three to four independent experiments. Statistical significance was estimated by one-way ANOVA with Tukey's post-test for multiple comparisons. A value of P<0.05 was considered significant. Analysis was performed using Prism version 5.01 for Windows, (GraphPad Software).
DT regulates LPS- and TNF-α-triggered endothelial activation by inhibiting gene expression of cytokines and their receptors
To evaluate the effect of DT on endothelial activation during inflammation, we analysed endothelial gene expression of cytokines and receptors essential for immune cell activation. MLEC-04 cells pre-treated with DT or DMSO (Figure 1A) were stimulated with LPS or TNF-α for 6 h and gene expression was surveyed using a mouse inflammatory cytokines and receptors cDNA microarray. From 89 inflammation-related genes evaluated in the microarray, 44 genes were significantly down-regulated by DT under the experimental conditions. DT impaired the expression of 35 and 27 genes induced in response to LPS and TNF-α respectively (Figure 1B). Thus, chemokines such as Ccl5, Ccl7, Ccl8, Cxcl1 and Cxcl10, which play an active role in recruiting leucocytes [21,22], were markedly inhibited by DT (Figure 1C). Comparison of the genes inhibited by DT in ECs stimulated with LPS or TNF-α revealed substantial overlap in the signalling pathways regulated by this compound. In particular, of 23 genes similarly induced in both LPS- and TNF-α-stimulated ECs, 18 (78%) were down-regulated by pre-treatment with DT (Figures 1B and 1C). In contrast, DT alone induced the transcription of ten genes (Figure 1C). Selected genes from the microarray were validated by RT-PCR. DT inhibited mRNA expression of Cxcl10 in response to LPS or TNF-α in ECs; additionally, DT specifically decreased the expression of Ccl5 in ECs stimulated with TNF-α and LPS-stimulated expression of Cxcl1 (Figure 1D). In accordance with its anti-inflammatory effects, DT alone up-regulated the expression of Ilr2 and, in combination with LPS or TNF-α, this effect was additive (Figure 1D). DT pre-treatment also up-regulated LPS-induced expression of IL-10 in MLEC-04 cells (Figure 1D). Expression levels of Cxcl10 were also evaluated by Western blotting. As anticipated, LPS and TNF-α stimulation increased the protein expression of Cxcl10 in MLEC-04 cells and pre-treatment with DT blocked this up-regulation (Figure 1E).
Inflammatory-related genes regulated by DT in response to LPS or TNF-α in lung ECs
Transcriptional regulation of cell adhesion molecules by DT in activated endothelium
We next evaluated the effect of DT on the expression of adhesion molecules essential for the interaction of circulating lymphocytes to the endothelial layer during the inflammatory response. Expression of PECAM-1 was used as an invariable control under these experimental conditions . Stimulation of ECs with LPS or TNF-α triggered an appropriate inflammatory reaction as demonstrated by overexpression of the endothelial adhesion molecules E-selectin, VCAM-1 and ICAM-1 (Figure 2A; left and right panels respectively). Pre-treatment of MLEC-04 cells with 25 μM DT markedly impaired this induction (Figure 2A). In the absence of inflammatory stimulation, DT-incubated ECs showed no detectable gene transcription and were similar to resting endothelium (results not shown). To assess the dose–response relationship between DT and inflammation-induced adhesion molecule expression, MLEC-04 cells were pre-treated with increasing amounts of DT before LPS or TNF-α stimulation. Whereas PECAM-1 gene expression was unaffected by DT, a partial inhibitory effect of DT on gene expression of the remaining adhesion molecules was detected at doses as low as 1 μM. Since 25 μM was the most effective dose without apparent toxicity to ECs (results not shown), we elected to use this concentration of DT for subsequent in vitro experiments.
Transcriptional regulation of cell adhesion molecules by DT in activated endothelium
DT down-regulates endothelial adhesion proteins in an inflammatory environment
To validate the adhesion protein gene expression results, we analysed protein expression by Western blotting and flow cytometry. Compared with virtually undetectable levels in resting cells, MLEC-04 cells activated with LPS or TNF-α overexpressed VCAM-1 and ICAM-1 proteins as demonstrated by Western blotting, and pre-treatment with DT significantly inhibited VCAM-1 and ICAM-1 expression in response to the inflammatory stimuli (Figures 3A and 3B respectively). To examine the cell-surface expression of the adhesion molecules, we carried out flow cytometry. As expected, ECs stimulated with LPS or TNF-α showed increased cell-surface overexpression of E-selectin, VCAM-1 and ICAM-1 as compared with resting cells (Figure 3C). Confirming the gene expression analysis, DT modulated the EC response to LPS or TNF-α by reducing the surface expression of E-selectin, VCAM-1 and ICAM-1, whereas protein levels were unaffected in resting endothelium (Figure 3C). PECAM-1 was used as an invariable control under these experimental conditions and was unaffected by treatments.
DT down-regulates endothelial adhesion proteins in an inflammatory context
DT blocks LPS- or TNF-α-mediated NF-κB activation in ECs
Because LPS and TNF-α signal through NF-κB  and DT has been described to block this transcription factor activity in macrophages , we evaluated the effects of DT on NF-κB signalling in stimulated ECs as a possible mechanism of endothelial inhibition. To do this, first we analysed the NF-κB inhibitor IκBα because of its key regulatory role in transcription factor activity. As expected, MLEC-04 cells incubated with LPS or TNF-α induced IκBα phosphorylation and subsequent degradation. The presence of DT blocked IκBα phosphorylation thus maintaining the protein level in the cell cytoplasm (Figure 4A). Next, we examined NF-κB signalling by analysing the nuclear localization of the p65 subunit as evidence of transcription factor activity. Western blot analysis of nuclear extracts showed p65 internalization occurring approximately 60 or 30 min after LPS or TNF-α stimulation of ECs respectively. This effect was abrogated by pre-treatment of ECs with DT (Figures 4B and 4C). Furthermore, compared with control conditions, LPS or TNF-α stimulation led to abundant p65 nuclear staining in MLEC-04 cells as shown by confocal microscopy of p65 and Hoechst 58-counterstained nuclei. DT pre-treatment impeded nuclear localization of p65 (Figure 4D).
DT blocks NF-κB and contributes to AP-1 transcriptional activities in ECs exposed to inflammation
DT mediates AP-1 transcriptional activity
To extend our knowledge on the mechanisms regulated by DT on ECs, and provide new insights into the potential pathways involved in the up-regulation of anti-inflammatory genes (e.g. IL1r2), we analysed transcriptional factors known to be important in the signalling pathways elicited by LPS or TNF-α, using a luciferase reporter assay (CCA-101L kit from SABiosciences). Out of ten different pathways analysed, only NF-κB and AP-1 transcriptional activities were regulated after stimulation with LPS or TNF-α (results not shown). Data from the luciferase reporter assay were confirmed using the luciferase reporter vectors NF-κB Luc and AP-1 Luc as described in the Experimental section (Figures 4E and 4F). According to the previous results showing inhibitory effects of DT on NF-κB signalling (Figures 4A–4D), NF-κB-dependent luciferase activity was significantly reduced on MLEC-04 cells treated with DT previous to the addition of LPS or TNF-α (Figure 4E). Interestingly, DT induced the activation of the transcription factor AP-1 at a similar level to LPS or TNF-α (Figure 4F). This result suggests that IL1r2 up-regulation by DT might be due to AP-1 activation, since AP-1-binding sites have been described in its promoter . Transcriptional activation of AP-1 after stimulation with pro-inflammatory cytokines has been described to be mediated by p38 signalling cascade [25–27], therefore we evaluated the involvement of p38 on DT-mediated signalling by Western blot assays. As expected, stimulation of ECs with LPS or TNF-α activated p38, whereas DT alone induced phosphorylation of p38 and co-operated with LPS or TNF-α in p38 activation (Figure 4G). To confirm the role of p38 in DT function, we performed experiments in the presence of the p38 inhibitor SB202190, using the expression of IL1r2 mRNA as a representative target of anti-inflammatory genes induced by DT (Figure 1D). As shown previously, IL1r2 mRNA was increased after MLEC-04 incubation with DT alone or in combination with LPS or TNF-α. Nevertheless, pharmacological inhibition of p38 significantly reduced IL1r2 expression in response to DT alone or in combination with LPS or TNF-α (Figure 4H).
VCAM-1 and ICAM-1 expression in vivo is regulated by DT in an LPS-induced mouse model of inflammation
To connect our in vitro observation with what happens in vivo, we evaluated endothelial adhesion marker expression in the lung of an LPS-induced endotoxemic mouse sensitized with DT. Consistent with our initial data (Figures 3A and 3C) and from other studies , VCAM-1 protein is virtually undetectable by immunohistochemistry of lung paraffin sections from control animals [Figures 5A(a) and 5A(d)]. In contrast, LPS-challenged animals strongly overexpressed VCAM-1 on the endothelial layer, indicating endothelial inflammatory activation [Figures 5A(b) and 5A(e)]. Prior sensitization with DT before LPS injection abrogated VCAM-1 expression in lung tissue, approaching levels close to untreated samples [Figures 5A(c) and 5A(f)].
Inhibition of VCAM-1 and ICAM-1 by DT on lung vasculature in an LPS-induced inflammation model
In contrast with VCAM-1, ICAM-1 is constitutively expressed on a variety of cells including ECs and alveolar epithelial cells . Accordingly, immunohistochemistry of lung sections from control animals demonstrated strong ICAM-1 staining on the alveolar epithelium and weaker staining on the vascular endothelium [Figures 5B(a) and 5B(d)]. LPS injection triggered overexpression of ICAM-1 in both epithelial and endothelial compartments as demonstrated by the increased cellular labelling [Figures 5B(b) and 5B(e)]. Similar to the results for VCAM-1, DT-sensitized mice had reduced ICAM-1 expression after LPS injection [Figures 5B(c) and 5B(f)], suggesting protection from inflammation.
DT-treated ECs regulate leucocyte behaviour during inflammation
We next assessed the physiological impact of DT-dependent inhibition of the endothelial inflammatory response using the leucocyte cell line J774. We first measured EC-triggered leucocyte activation using EC CM. Activated endothelium releases chemokines and cytokines that prime leucocyte function such as integrin-mediated cell adhesion  (Figure 1). We therefore performed adhesion assays to FN as readout for the activity of its receptor α4β1 that is expressed in J774 cells and is an essential integrin for leucocyte firm adhesion to the endothelial layer during inflammation [14,31]. CM from LPS- or TNF-α-stimulated MLEC-04 cells significantly increased J774 relative adhesion to FN as compared with CM collected from resting or DT pre-treated ECs (Figure 6A). Moreover, CM collected from ECs pre-treated with DT before stimulation with LPS or TNF-α failed to increase J774 adhesive properties. Appropriate controls were carried out to ensure that residual LPS, TNF-α or DT did not interfere with the assay (results not shown). These findings indicate that molecules released from activated ECs are responsible for J774 activation, and inhibition of endothelium with DT contributes to maintain leucocyte resting during inflammation (Figure 6A).
DT-treated ECs regulate leucocyte behaviour during inflammation
Finally, we explored the functional relevance of DT-dependent inhibition of endothelial VCAM-1 and ICAM-1 expression on J774–EC co-adhesion as readout for leucocyte–EC firm adhesion during the inflammatory response. MLEC-04 cells were stimulated with LPS or TNF-α in the presence or the absence of DT and, after several washes, CFSE-labelled leucocytes were added to allow cell interactions. As expected, LPS- or TNF-α-stimulated ECs induced a significant increase in J774 cells adhering to the endothelial monolayer (Figure 6B). In contrast, DT-treated ECs activated with LPS or TNF-α failed to support J774 attachment to MLEC-04 cells (Figure 6B). Thus, DT-dependent regulation of endothelial adhesion molecule expression is an efficient strategy to prevent leucocyte adhesion to the vasculature during inflammation. Collectively, these results identify DT as an endothelial regulator of inflammatory-related disorders that can control leucocyte functions and, consequently, contribute to health recovery.
LPS is a component of the outer membrane of Gram-negative bacteria that promotes activation of macrophages, neutrophils and the adjacent endothelium. Direct activation of endothelium by LPS is the earliest step in leucocyte recruitment into the endotoxaemic lung . In this setting, TNF-α is released by activated endothelium and stimulates the expression of pro-inflammatory cytokines and cell-surface adhesion molecules [32,33], thus contributing to pathology. Accordingly, endothelium-targeting strategies involving the regulation of leucocyte recruitment might constitute bona fide therapies for inflammatory disorders . LPS and TNF-α both signal through activation of NF-κB. Many reports associate NF-κB activation with the establishment of different human pathologies, including chronic inflammatory disease, therefore the manipulation of NF-κB function represents one of the most interesting research disciplines in this field . We have previously described a novel compound, DT, chemically derived from the natural diterpenoid hispanolone, which shows inhibitory activity against NF-κB and consequently blocks the inflammatory programme in innate immune cells . In the present study, we establish a new role for the diterpenoid DT as an efficient anti-inflammatory drug in the endothelial compartment. DT blocks EC activation triggered by LPS or TNF-α, resulting in a reduction in released pro-inflammatory cytokines and chemokines and expression of cell- surface adhesion molecules by the endothelium. At the same time, the expression of anti-inflammatory molecules such as IL-10 or IL1r2 is induced, contributing to the resolution of the inflammation. As a consequence of these activities, leucocyte activation and recruitment to the vascular layer is hindered (see summary model in Figure 7).
NF-κB activity is a decisive step in the inflammatory response. In resting cells, NF-κB is sequestered in the cytoplasm by association with the inhibitory proteins IκB. Upon cell activation by inflammatory stimuli, IKK complex phosphorylates IκB and promotes its degradation, thus allowing NF-κB (constituted by the heterodimer p65/p50) to translocate to the nucleus to activate a gene transcription programme [23,36]. DT interferes with NF-κB activation in ECs during inflammation as observed by the inhibition of IκBα phosphorylation and the absence of the NF-κB subunit p65 in the cell nucleus (Figures 4A–4D). The presence of two Michael acceptor groups in DT molecule with probability to target specific residues on IKK complex has been postulated to be involved in the NF-κB regulation [13,37]. In addition, transfection of ECs with luciferase reporter vectors containing specific transcription factor-binding elements revealed that DT negatively regulates transcriptional activity mediated by NF-κB. These findings further establish DT as a novel inhibitor of NF-κB function, with the ability to act on several cell types such as ECs, as shown in the present study, and on macrophages . Further studies are needed in other biological situations to ascertain the universality of DT inhibition of NF-κB-dependent transcriptional regulation .
In addition to NF-κB, LPS and TNF-α also signal through AP-1 to activate the inflammatory transcriptional programme. Luciferase reporter assays not only showed activation of AP-1 after LPS and TNF-α stimulation, but also demonstrated that AP-1 was activated after DT treatment. The induction of AP-1 by pro-inflammatory cytokines is mediated, among other members of the mitogen-activated protein kinases (MAPK) family, by p38 [25–27]. In the present study, we report that DT regulates p38 signalling cascade by itself and in collaboration with LPS or TNF-α, suggesting that the likely molecular mechanism by which DT up-regulates the expression of the anti-inflammatory cytokine IL1r2, may involve the activation of AP-1 via p38 modulation, since its promoter presents AP-1-binding sites . Indeed, pharmacological inhibition of p38 signalling significantly reduced IL1r2 expression. This is the first demonstration that DT can induce IL1r2 expression via activation of p38 and AP-1, providing a dual role for this compound on the inflammatory response of ECs. This complex regulation is not unusual as other terpenoids have been described to block NF-κB signalling allowing p38 cascade .
Cytokines, such as TNF-α and chemokines, such as Ccl5, Cxcl1 or Cxcl10, are released by the endothelium in an inflammatory environment and mediate leucocyte activation and recruitment across the vascular layer to the injured tissue [21,22]. It has been proposed that these molecules play a key role in the pathophysiology of ALI, ARDS and COPD. In fact, anti-cytokine therapy has led to improved recovery from inflammatory-related diseases . In our lung endothelial model, DT inhibited endothelial inflammatory activation by impairing the expression of a wide panel of cytokines and chemokines, including those mentioned above. In addition, DT induced the endothelial expression of the inflammatory repressor cytokine IL-10 and the decoy receptor for IL-1, IL1r2 [40,41], indicating a double regulatory system for leucocyte performance. In line with our data, cytokine inhibition has been described as a mechanism underlying the anti-inflammatory effects of the agonist of the liver X receptor, GW3965, in COPD . In this sense, GW3965 down-regulated Cxcl10 and Ccl5 expression while inducing IL-10 expression, similar to our findings for DT in ECs.
Cytokines released from activated endothelium mediate the first step of leucocyte recruitment during inflammation . These soluble factors activate leucocytes and, among other acquired abilities, they induce integrin-mediated cell adhesion . α4β1 is a cell-surface integrin found on leucocytes and mediates cell adhesion to FN and VCAM-1, the latter being a fundamental ligand for leucocyte firm adhesion to the endothelial layer . In this regard, several studies point to α4β1 as a therapeutic target for treatment of several inflammatory disorders such as asthma and COPD . CM from DT-treated activated ECs failed to support adhesion of J774 leucocytes to FN, suggesting a role for endothelial cytokines on leucocyte activation, at least for the heterodimer α4β1. J774 cells express this integrin family member and J774 cells have been validated as a leucocyte cell model for studies on inflammatory regulation ( and references therein). In this scenario, the lack of leucocyte α4β1 functionality implies an inhibitory step for adhesion to the vascular ligand VCAM-1, thus preventing leucocyte attachment to the endothelial surface and hampering progression of inflammation.
Expression of vascular adhesion molecules is essential to mediate leucocyte firm adhesion to the endothelial layer to maintain the inflammatory response . Anti-adhesion therapies targeting endothelial VCAM-1 or ICAM-1 have been effective for the treatment of inflammatory disorders. These strategies are based on the regulation of the protein expression or by addition of inhibitory compounds [46–48]. In our model, DT inhibited the expression of the adhesion molecules VCAM-1 and ICAM-1 during inflammation. Functionally, DT-inactivated ECs failed to support firm interaction of the J774 cells to the vascular layer and, apparently, impaired leucocyte extravasation.
Consistent with our previous findings , the present study establishes DT as an effective anti-inflammatory agent by regulating endothelial functionality and innate immunity during inflammation. Currently, a wide range of anti-inflammatory strategies have been developed as effective therapies for the treatment of different chronic disorders such as atherosclerosis, Crohn's disease and inflammatory lung diseases [49,50]. Physicians and researchers must pay special attention to the immunocompromised state of the patient resulting from prolonged drug administration or severe treatments. This artificial immunodeficient situation may lead to vulnerability to common infections and opportunistic organisms such as bacteria, viruses and fungi, and could be life-threatening [51–53]. Given this, it could be interesting to analyse the functionality of the defence system in individuals treated systemically with DT. In the case of compromising the integrity of the immune response of the patient, it would be important to evaluate the possibility of specifically targeting the vascular compartment with DT to ameliorate adverse consequences of widespread administration . This endothelium-specific therapy might interfere with immune cell recruitment without affecting their activity, and could open the possibility to combine this treatment with other verified anti-inflammatory therapies .
Lidia Jiménez-García, Sandra Herranz, Paqui Través, Raquel López-Fontal and María Angeles Higueras performed experimental work and reviewed the paper before submission. Beatriz de las Heras provided essential reagents and revised the paper. Sonsoles Hortelano contributed to the design of the work, interpretation of data and revised the paper. Alfonso Luque performed and supervised experimentation, designed the work, analysed data and wrote the paper.
We thank Raquel Pérez Tavarez and Manuel Priego from the Histology Unit and Silvia Hernández and Fernando Gonzalez-Camacho from the Confocal Facility for their technical assistance.
CONFLICT OF INTEREST
Beatriz de las Heras and Sonsoles Hortelano are inventors on a Spanish patent application on labdane diterpenoids as anti-tumoral agents. The other authors declared no conflict of interest.
This work was supported by the Ministerio de Economía y Competitividad (MINECO) and the Instituto de Salud Carlos III (ISCIII) [grants numbers TPY-M-1068/13 and IERPY 1149/16 (to A.L.); and MPY 1410/09 (to S.H.)]; by the MINECO through the Fondo de Investigación en Salud (FIS) [grant number PI11.0036 (to S.H.); and FI12/00340 (to L.J-G.)]; and by the MINECO-FIS through the Miguel Servet Program [grant number CP12/03087 (to A.L.)].
acute lung injury
activator protein 1
acute respiratory distress syndrome
chronic obstructive pulmonary disease
intercellular adhesion molecule 1
mitogen-activated protein kinases
nuclear factor κB
p38 mitogen-activated protein kinase
platelet endothelial cell adhesion molecule-1
polypyrimidine tract-binding protein-associated splicing factor
relative luciferase units
tumour necrosis factor-α
vascular cell adhesion molecule 1