|Home | About | Journals | Submit | Contact Us | Français|
Gram-negative bacteria contain ligands for Toll-like receptor (TLR) 4 and nucleotide oligomerization domain (NOD) 1 receptors. Lipopolysaccharide (LPS) activates TLR4, while peptidoglycan products activate NOD1. Activation of NOD1 by the specific agonist FK565 results in a profound vascular dysfunction and experimental shock in vivo.
Here, we have analysed a number of pharmacological inhibitors to characterize the role of key signalling pathways in the induction of NOS2 following TLR4 or NOD1 activation.
Vascular smooth muscle (VSM) cells expressed NOD1 mRNA and protein, and, after challenge with Escherichia coli or FK565, NOS2 protein and activity were induced. Macrophages had negligible levels of NOD1 and were unaffected by FK565, but responded to E. coli and LPS by releasing increased NO and expression of NOS2 protein. Classic pharmacological inhibitors for NF-κB (SC-514) and mitogen-activated protein kinase (SB203580, PD98059) signalling pathways inhibited responses in both cell types regardless of agonist. While TLR4-mediated responses in macrophages were specifically inhibited by the pan-caspase inhibitor z-VAD-fmk and the PKC inhibitor Gö6976, NOD1-mediated responses in VSM cells were inhibited by the Rip2 inhibitor PP2.
Our findings suggest a selective role for NOD1 in VSM cells, and highlight NOD1 as a potential novel therapeutic target for the treatment of vascular inflammation.
It is now clear that pattern recognition receptors (PRRs) are intrinsically involved in cardiovascular diseases including atherosclerosis (Michelsen et al., 2004). Moreover, Toll-like receptor (TLR) 4, the receptor for Gram-negative lipopolysaccharide (LPS), or TLR2, the receptor for Gram-positive lipoteichoic acid, is expressed on vascular cells and in vessels where they sense pathogens directly (without the requirement of macrophages), leading to induction of inflammatory genes including NOS 2 (Cartwright et al., 2007a,b;) and COX-2 (Jimenez et al., 2005). Our recent work has shown that activation of the PRR nucleotide oligomerization domain (NOD) 1 on vascular smooth muscle (VSM) cells leads to profound induction of NOS2 in vitro and vascular collapse in vivo (Cartwright et al., 2007a). NOD1 and NOD2 are receptors for products of the synthesis and degradation of peptidoglycan (PGN) from Gram-negative bacteria (Chamaillard et al., 2003; Masumoto et al., 2006; Park et al., 2007). Thus, TLR4 and NOD1 are both receptors for different pathogen-associated molecular patterns (PAMPs) from Gram-negative bacteria. Importantly, others have shown that, in leucocyte populations, activation of NOD1 receptors alone can result in weak activation, and that co-stimulation with TLR agonists can result in increased responses (Chamaillard et al., 2003; Fritz et al., 2005; van Heel et al., 2005; Tada et al., 2005; Masumoto et al., 2006). However, work from our group and others have shown that activation of NOD receptors alone results in induction of inflammatory genes in some types of cell, including VSM cells (Cartwright et al., 2007b; Park et al., 2007). However, the mechanism by which activation of NOD1 in VSM cells results in inflammation was not previously studied. Thus, in the current study, we have compared responses in VSM and macrophages to specific ligands for TLR4 and NOD1, and measured NOS2 activity as a relevant readout of cell activation. Our findings show, for the first time, that NOD1 receptors are selectively expressed in VSM cells. We also go on to establish similarities and differences in how NOD1 versus TLR4 activation (in VSM cells vs. macrophages, respectively) signal resulting in the induction of NOS2. These observations provide important new information about how vascular cells sense pathogens and danger signals, which may well have direct relevance to our understanding of cardiovascular disease in more complex settings.
Clonal populations of VSM cells with a contractile (‘spindle’) phenotype derived from the mixed polyclonal rat aortic VSM cell line WKY12-22 were kindly provided by Professor Timothy Warner (William Harvey Research Institute, London, UK). J774.2, a murine macrophage cell line, was obtained from the European collection of cell cultures. Murine peritoneal macrophages were collected from C57BL/6 mice or Wistar–Kyoto rats by peritoneal lavage (Terenzi et al., 1995). Cells were cultured as described previously (Jimenez et al., 2005) using Dulbecco's modified Eagle's medium (DMEM) containing 10% FBS and antibiotic/mycotic mix. Cells were treated for 24 (VSM and J774.2 cells) or 48 (rat peritoneal macrophages) hours with either FK565 (10 nM, NOD1 agonist), LPS (1 µg·mL−1, TLR4 agonist) or whole heat-killed Escherichia coli (3 × 107 CFU·mL−1). In some experiments, cells were pretreated for 60–90 min with pharmacological inhibitors of different signalling cascades before the addition of bacterial ligands.
NO production by cells was measured by the accumulation of its oxidation product, nitrite, using the Griess reaction, as we have described previously (Jimenez et al., 2005; Paul-Clark et al., 2006). Nitrite levels obtained in the presence of different inhibitors are expressed as a percentage of the response to the agonist alone.
Cell respiration, a marker of viability, was assessed after all treatments by measuring the reduction of MTT (Sigma Chemical Company, Poole, Dorset, UK) to formazan. Unless shown in Supporting Information Tables S1 and S2, none of treatments affected cell viability.
Western blot analysis using specific NOS2 (1:500; Santa Cruz Biotechnology, Santa Cruz, CA, USA), NOD1 (1:1000, Cell Signaling Technology, New England Biolabs Ltd., Hitchin, Hertfordshire, UK) and COX-2 (1:1000; Cayman Chemical, Ann Arbor, MI, USA) antibodies was performed as previously described (Jimenez et al., 2005; Paul-Clark et al., 2006; Cartwright et al., 2007a). The expression of NOD1 was also analysed on murine macrophages collected from C57BL/6 mice by peritoneal lavage (Bosca et al., 2005).
Reverse transcription–PCR (RT–PCR) analysis was performed as previously described (Cartwright et al., 2007a). Total RNA was extracted using RNeasy Minikit (Qiagen, Crawley, UK), and converted into complementary DNA by RT following the protocol of the manufacturer (Omniscript Reverse Transcription kit, Qiagen). Primers for the PCR were selected from the regions of the NOD1 mRNA sequence with the highest degree of homology in rat (NM_001109236.1, GenBank) and mouse (NM_172729.2, GenBank). Both sequences were compared with GenBank using the Blast software (National Center for Biotechnology Information website). The following primers, synthesized on request by Invitrogen (Paisley, UK), were used for the amplification of NOD1; sense: 5′- GTC CTC AAC GAG CAT GGC GAG ACT -3′ and antisense: 5′-TGC AGC TCA TCC AGG CCG TCA A-3′ resulting in a product of 302 bp; and for β-actin, sense: 5′-GAT CAT GTT TGA GAC CTT CAA CAC C-3′ and antisense: 5′-GTT TCA TGG ATG CC ACA GGA TTC-3′, which yielded a product of 467 bp. In preliminary experiments, we tested a number of cycles ranging from 24 to 35, with cDNA diluted over a range of three magnitudes, to establish the linearity of the reaction. PCR of NOD1 was performed using a MyCycler Thermal cycler (Bio-Rad, Hercules, CA, USA) by using Taq PCR Master Mix Kit (Qiagen) for 30 cycles (94°C for 45 s, 64.7°C for 45 s and 72°C for 45 s) followed by a final extension step at 72°C for 10 min. The same protocol was used for β-actin except for the annealing temperature, which was 60.4°C. The products of the PCR were separated in 2% agarose gels stained with ethidium bromide, and bands were quantified using image analysis software (Quantity One, Bio-Rad). Results were normalized to the bands of β-actin from the same samples.
Heat-killed E. coli suspensions were prepared as previously described (Cartwright et al., 2007a) from a clinical blood culture isolate of E. coli 0111.B4. The NOD1 agonist FK565 (heptanoyl-gamma-d-glutamyl-l-meso-diaminopimelyl-d-alanine), a generous gift from Professor David Van Heel (QMUL, London, UK), was obtained from Fujisawa Pharmaceutical (Osaka, Japan), the NOD2 agonist MDP (muramyl N-acetylmuramyl-l-alanyl-d-isoglutamine, also known as muramyl dipeptide) was purchased from Invivogen (San Diego, CA, USA), the cell-permeable JNK peptide inhibitor d-JNKi ((d)-hJIP[175-157]-DPro-DPro-(d)-HIV-TAT[57-48]) was purchased from Enzo Life Sciences (Exeter, UK) and the TLR4 agonist LPS (from E. coli 055:B5) was purchased from Sigma. All other drugs were obtained from Calbiochem (San Diego, CA, USA). Drugs were dissolved initially in dimethyl sulphoxide (DMSO) (except for LPS and FK565, which were dissolved in PBS) to prepare a 0.1 or 0.01 mol·L−1 stock solution; further dilutions were made in DMEM. The maximum concentration of DMSO achieved in any experiment, 0.3%, did not show significant effects on cell viability.
Results were normalized to ‘control’ in most figures, expressed as mean ± SEM, and were analysed by Student's one sample t-test. Values of P < 0.05 were considered to be significant.
We have recently shown that FK565 activates the VSM component of intact blood vessels via a NOD1-specific pathway resulting in NOS2 induction and vascular dysfunction (Cartwright et al., 2007b). The effect of FK565 was found to be independent of TLR4 or TLR adapter proteins (Cartwright et al., 2007b). Here, we have confirmed that FK565 induces NOS2 activity (Figure 1A and Supporting Information Figure S1) and protein expression (Figure 1C) in cultured VSM cells. Furthermore, NOS2 induction was accompanied by an increase in the activity and protein expression of COX2 (Supporting Information Figure S1), another enzyme involved in the production of key vasoactive inflammatory mediators (such as prostacyclin). NOS2 activity and protein were also induced in VSM cells stimulated with whole Gram-negative E. coli and, to a lesser extent, LPS (Figure 1A,C). Similarly, J774.2 murine macrophages showed increased levels of NOS2 activity (Figure 1B) and protein (Figure 1D) when stimulated with E. coli or LPS. However, in direct contrast to results seen in VSM cells, FK565 had no effect on NOS2 activity in macrophages (Figure 1B,D). VSM cells and J774.2 macrophages were unaffected by the NOD2 agonist MDP (nitrite µM; VSM cells: basal, 0.0 ± 0.0; plus 10 µM MDP, 0.8 ± 0.4; macrophages: basal, 1.0 ± 0.9, plus 10 µM MDP, 0.0 ± 0.0; n= 9; 24 h). In order to ensure these observations were not due to any species differences, we performed separate studies using primary cultures of rat macrophages. Similar to experiments using murine J774 macrophages, rat macrophages released increased levels of NO when stimulated with LPS, but not FK565 (nitrite µM; basal, 0 ± 0; plus 1 µg·mL−1 LPS, 28 ± 10; plus 10 nM FK565, 0 ± 0; n= 6; 48 h). In line with these observations, VSM cells expressed clear detectable levels of NOD1 protein, while both J774.2 and freshly elicited mouse macrophages expressed low or undetectable levels of NOD1 protein (Figure 1E,F). These findings were confirmed by RT–PCR (Figure 1G,H). Levels of NOD1 gene expression (NOD1/β-actin) were not altered by stimulation of cells with FK565 for 24 h (VSM cells: control 0.100 ± 0.030, plus FK565 0.147 ± 0.033; J774.2 macrophages: 0.006 ± 0.002, plus FK565 0.003 ± 0.006). In previous studies, enhancement of TLR4-induced secretion of pro-inflammatory cytokines (including IL-1β, IL-6 or IL-8) by NOD1 stimulation has been shown in macrophages and dendritic cells (Chamaillard et al., 2003; Fritz et al., 2005; Masumoto et al., 2006). In our study, we found no interaction between FK565 and LPS for the induction of NOS2 in macrophages (co-stimulation with 10 nM FK565 plus 1 µg·mL−1 LPS induced 105 ± 5.2% of nitrite produced by 1 µg·mL−1 LPS alone). In VSM cells, there was a small, but statistically significant, increase in the ability of FK565 to induce NOS2 activity when cells were co-treated with threshold concentrations of LPS (co-stimulation with 0.1 µg·mL−1 LPS plus 10 nM FK565 induced 132 ± 5.6% of nitrite produced by FK565 alone, P < 0.05).
In different cell types, induction of NOS2 by a variety of agents is sensitive to inhibition by corticosteroids. In this study, we found that NOS2 activity and protein expression induced by either FK565 in VSM cells or by LPS in macrophages were inhibited by dexamethasone (Figure 2A–D; Supporting Information Table S3). The I-κB kinase (IKK)-2-inhibitor SC-514 (4-amino-[2′,3′-bithiophene]-5-carboxamide) (Kishore et al., 2003) inhibited NOS2 activity and protein expression in VSM cells stimulated with FK565 (Figure 2E,G; Supporting Information Table S3) or macrophages (Figure 2F,H; Supporting Information Table S3). Similarly, SC-514 inhibited NOS2 activity induced by whole E. coli in VSM cells and macrophages (Supporting Information Table S4). The p38 and extracellular signal-regulated (ERK) mitogen-activated protein kinase (MAPK) inhibitors SB 203580 (4-[5-(4-fluorophenyl)-2-[4-(methylsulphonyl)phenyl]-1H-imidazol-4-yl]pyridine) and PD 98059 (2-(2-amino-3-methoxyphenyl)-4H-1-benzopyran-4-one) (Davies et al., 2000), respectively, inhibited NOS2 activity and protein expression in VSM cells stimulated with FK565 or macrophages stimulated with LPS (Figure 3; Supporting Information Table S3). NOS2 activity induced in macrophages by LPS was inhibited by the c-Jun N-terminal kinase (JNK) inhibitor SP 600125 (anthra[1-9-cd]pyrazol-6(2H)-one) (Bennett et al., 2001) (Figure 3; Supporting Information Table S4), but not by the cell-permeable JNK peptide inhibitor d-JNKi ((d)-hJIP[175-157]-DPro-DPro-(d)-HIV-TAT[57-48]) (Borsello and Bonny, 2004). By contrast, SP 600125 showed a weak trend to inhibit NOS2 activity induced by FK565 in VSM cells, while d-JNKi induced a statistically significant inhibition of activity in these cells (Figure 3; Supporting Information Tables S3 and S4). These results are in contrast to those found in cells stimulated with whole E. coli (Supporting Information Table S4), where SP 600125 inhibited NOS2 activity in both VSM cells and macrophages. Gö6983 (3-[1-[3-(dimethylamino) propyl]-5-methoxy-1H-indol-3-yl]-4-(1H-indol-3-yl)-1H-pyrrole-2,5-dione), which at the concentration tested is a selective inhibitor of conventional (α, βI, βII, γ), some novel (δ) and atypical (ι/λ, ζ) isoforms of PKC (Way et al., 2000), had no effect on NOS2 activity induced in VSM cells by FK565 or macrophages induced by LPS (Figure 4). Gö6976 (5,6,7,13-tetrahydro-13-methyl-5-oxo-12H-indolo[2,3-a]pyrrolo[3,4-c]carbazole-12-propanenitrile), which, in addition to conventional isoforms, inhibits PKCµ (Way et al., 2000), also had no effect on NOS2 activity induced by FK565 in VSM cells. By contrast, Gö6976 inhibited NOS2 activity induced by LPS in macrophages (Figure 4). Similar results were found when cells were stimulated with whole E. coli. The Src family kinase inhibitor PP2 (3-(4-chlorophenyl) 1-(1,1-dimethylethyl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine), at concentrations reported to inhibit RIP2 [EC50= 100 nM for Src and EC50= 10 nM for RIP2 (Windheim et al., 2007)] inhibited NOS2 activity induced in VSM cells stimulated with either FK565 or whole E. coli, but had little or no effect on activity induced in macrophages stimulated with LPS or whole E. coli (Figure 5, Supporting Information Table S4).
Activation of several NOD-like receptors (NLRs) from the NALP subfamily has also been linked to the formation of an inflammasome, leading to the activation of caspase, which processes pro-IL-1β into mature IL-1β to initiate an inflammatory response. Neither the pan-caspase inhibitor (Garcia-Calvo et al., 1998) Z-VAD-fmk (benzyloxycarbonyl-Val-Ala-Asp(OMe)-fluoromethylketone; Figure 6A) nor binding antibodies for IL-1β or IL-18 (Figure 6C) affected NOS2 activity induced in VSM cells stimulated with FK565. By contrast, Z-VAD-fmk inhibited NOS2 activity and protein expression in J774.2 macrophages stimulated with whole E. coli or LPS (Figure 6B,D; Supporting Information Tables S3 and S4). However, this was associated with a direct effect of the drug on cell viability (Supporting Information Tables S1 and S2). The selective inhibition of caspase 1 and caspase 8 isoforms by Z-WEHD-fmk (benzyloxycarbonyl-Trp-Glu(OMe)-His-Asp(OMe)-fluoromethylketone) and Z-IETD-fmk (benzyloxycarbonyl-Ile-Glu(OMe)-Thr-Asp(OMe)-fluoromethylketone) (Maelfait et al., 2008), respectively, mimicked the effects of Z-VAD-fmk on NOS2 induction while preserving cell viability (Figure 6F).
It is now well accepted that stromal cells of the vasculature, including VSM cells, sense pathogens and danger signals directly, and that this is important for the development of septic shock and atherosclerosis. In the current study, we identified a clear difference in how VSM and macrophages sense NOD1 agonists. NOD1 receptors are PRR for PAMPs of Gram-negative bacteria. VSM cells express NOD1 receptors and respond to agonists by expressing increased NOS2. By contrast, macrophages do not express NOD1 receptors and are unresponsive to the selective agonist FK565. Our findings also suggest that the signalling pathways for NOD1 and TLR4 are separate, but converge at the level of NF-κB and MAPK activation for NOS2 induction.
In the current study, we found that the synthetic NOD1 agonist FK565 induced profound increases in NO release and increased expression of NOS2 protein in VSM cells. Similarly VSM cells showed increased levels of COX-2 activity and expression when stimulated with FK565. These observations confirm our previous results showing that activation of NOD1 in murine blood vessels induces NOS2 activity and vascular dysfunction (Cartwright et al., 2007a). Our previous studies confirmed that these actions of FK565 are totally specific to NOD1 receptors in vitro and are accompanied by extensive vascular collapse and shock in vivo (Cartwright et al., 2007a). In that model of shock, the decline in blood pressure is directly associated with the induction of NOS2 in blood vessels (Szabo et al., 1993). We also found that the PAMP for TLR4, LPS induced NOS2 activity in our cultures of VSM cells, although the magnitude of response was reduced compared to the response we obtained when cells were stimulated with FK565. This observation is also in line with, and explains, our previous study where the ‘shock’ induced by FK565 in vivo was far greater than that induced by LPS (Cartwright et al., 2007a). It is therefore significant to note that in VSM, NOD1 appears to dominate TLR4 responses when it comes to induction of NOS2. Importantly, the reverse seems to be true for macrophages. As expected, we found that LPS induced a robust induction of NOS2 activity in the murine macrophage cell line, J774, and in primary cultures of rat macrophages. By contrast, we found that the NOD1 agonist FK565 was completely inactive as an inducer of NOS2 activity or protein. In line with these observations, we found that VSM cells expressed detectable levels of NOD1 mRNA and protein, while macrophages did not. NOD2 signalling is often studied in parallel with NOD1. We have previously shown that the NOD2 agonists, MDP, have little effect on NOS2 activity in VSM cells. In the current study, we have confirmed these observations in experiments where VSM cells and macrophages were treated in parallel. Specifically, we found that MDP did not induce NOS2 activity in either cell type. These observations strengthen our assertion that NOD1 is the predominant pathway for NOS2 induction in these cells. This result supports published data showing that NOD2 agonists are poor inducers of cytokine production (Park et al., 2007) and vascular dysfunction (Cartwright et al., 2007b). In other studies, activators of NOD receptors and activators of TLRs were found to have a synergistic effect (Chamaillard et al., 2003; Fritz et al., 2005; van Heel et al., 2005; Tada et al., 2005; Masumoto et al., 2006). However, in our study, where NOS2 was the gene of interest, we found no interaction between LPS and FK565 in macrophages, and only a very modest interaction in VSM cells. Nevertheless, because macrophages represent a phenotypically highly diverse cell population, the role of NOD1 signalling pathways in macrophages from different sources may well vary and requires further study before extensive conclusions can be drawn.
The pathway by which TLR4 signals in the VSM component of blood vessels is well worked out (Yang et al., 2005; Cartwright et al., 2007a; Son et al., 2008). However, this is the first study to investigate NOD1-mediated NOS2 induction in a vascular model. Firstly, we demonstrated that similar to NOS2 induced via the TLR4 pathway, the glucocorticoids dexamethasone and prednisolone inhibited NOS2 induced by FK565 in VSM cells. NF-κB is a major inflammatory factor triggered by both TLRs and NLRs (Shaw et al., 2008). Accordingly, the effects induced by TLR4 and NOD1 agonists were blocked by the IKK-2 inhibitor SC-514 in both types of cell. More recently, it was shown that Gram-negative bacterial LPS is also able to activate MAPK signalling pathways in macrophages (Shaw et al., 2008). Our data confirm that the p38 MAPK and ERK1/2 pathways are needed for NOS2 induction in VSM cells and J774.2 macrophages. The role of JNK is, however, uncertain in this cell system and requires further study. In addition, the apparent lack of specificity of SP 600125 in murine macrophages highlights that caution must be applied when drawing conclusions using these compounds.
In other cell types, the details of how activation of NF-κB and MAPK is linked to TLR4 and NOD1 are not clearly established, but different reports have suggested that PKC and some tyrosine kinases could be involved in these signalling pathways. Several broad inhibitors of PKC isoforms have been demonstrated to be effective in inhibiting secretion of pro-inflammatory cytokines following TLR4 activation in macrophages, neutrophils and neurones (Asehnoune et al., 2005; Song et al., 2007). Gö6983 and Gö6976 were used in this study to assess the role of PKC in the induction of NOS2 by TLR4 and NOD1 agonists. Gö6983, which at the concentration tested shows selectivity for the conventional (α, βI, βII, γ), some novel (δ) and atypical (ι/λ, ζ) PKC isoforms (Way et al., 2000), had no effect on NOS2 activity induced in VSM cells by FK565 or that in macrophages induced by LPS. By contrast, Gö6976, which in addition to the conventional isoforms, also inhibits PKCµ (Way et al., 2000) inhibited NOS2 activity induced by LPS in macrophages, but had no effect on the activity induced in VSM cells by FK565. As the only isoform known to be sensitive to Gö6976 and insensitive to Gö6983 is PKCµ (also known as protein kinase D or PKD), our data suggest that this isoform might be involved in the responses induced by LPS in macrophages. PKD isoforms are present in a wide variety of cells where they control cellular processes such as cell proliferation, apoptosis or immune responses (Van Lint et al., 2002; Song et al., 2007). PKDs have been shown to control gene transcription by activating NF-κB and MAPK pathways, and by inhibiting HDACs (Matthews et al., 2006). PKD can be activated by a variety of stimuli via PKC-dependent and -independent pathways (Johannessen et al., 2007). Previous reports have shown that the involvement of specific PKC isoforms seems to be time dependent. Because we analysed the effects of these inhibitors after 24 h, the possibility that other PKC isoforms are involved in earlier events cannot be ruled.
As pointed out in the introduction, the signalling events associated with NOD1 are not completely understood. It is clear that RIP2 and NOD1 interact, in order for downstream signalling to occur (Kobayashi et al., 2002), but whether this interaction provides a mere physical link between NLR and downstream effectors molecules or whether the kinase activity of RIP2 plays a direct role in transduction mechanisms remains to be determined (Windheim et al., 2007). In the present study, PP2, a potent inhibitor of RIP2 (Windheim et al., 2007), inhibited the production of NO induced by FK565 in VSM cells. Because activation of NF-κB by TLR agonists has also been shown to signal through this kinase, RIP2 was proposed as a point of synergy (Kobayashi et al., 2002). However, in our model, inhibition of RIP2 had little or no effect on macrophages stimulated with either LPS or E. coli. These results suggest that RIP2 kinase activity is required for NOS2 induction following NOD1 activation. SB 203580, which is commonly used as a specific inhibitor of p38 MAPK, has also been found to inhibit RIP2 kinase activity in vitro with even greater potency (Argast et al., 2005; Windheim et al., 2007). Therefore, the inhibition observed in the production of NO by VSM cells following treatment with SB 203580 might be revealing the consequences of inhibiting both p38 MAPK and RIP2.
Activation of TLRs can result in the activation of caspase 1 and caspase 8 (Maelfait et al., 2008; Shaw et al., 2008) by the inflammasome complex, which induces the maturation of the pro-inflammatory cytokines IL-1β and IL-18. Functional co-operation has also been observed between inflammasome NLRs and NOD1 receptors, generating controversy over whether inflammasome NLRs are able to sense products of the degradation of PGN (Shaw et al., 2008). When we analysed the role of the inflammasome in NOD1 responses using FK565 or whole Gram-negative bacteria in VSM cells, we found that neither the pan-caspase inhibitor Z-VAD-fmk nor the binding antibodies for IL-1β or IL-18 affected the production of NO. By contrast, Z-VAD-fmk inhibited TLR4-mediated NOS2 activity and protein expression, and reduced cell viability in macrophages stimulated with LPS or E. coli. The effects of Z-VAD-fmk on NOS2 induction, but not on macrophage viability, were mimicked when caspase 1 and caspase 8 were inhibited selectively using Z-WEHD-fmk and Z-IETD-fmk respectively. It should be noted that, in contrast to some preceding reports, in the current study where NOS2 activity was our readout, no account was taken of any influence of exogenous or endogenous ATP. Finally, our data suggest that activation of caspases and PKD are both necessary and specific events for the induction of NOS2 following TLR4 activation in macrophages. It is noteworthy that caspase-dependent activation of PKD has been demonstrated both in vitro and in vivo (Endo et al., 2000; Van Lint et al., 2002).
In addition to immune cells, structural cells of the cardiovascular system including VSM cells sense pathogens and danger signals directly, clearly contributing to the development of septic shock and atherosclerosis. In sepsis or vascular inflammation, the endothelium tends to become activated, resulting in recruitment of immune cells to the site of injury. However, if the inflammation persists, the endothelium can become dysfunctional with reduced endothelium-dependent dilator responses (Clapp et al., 2004). Under these conditions, the underlying VSM becomes activated by PAMPs or cytokines, and expresses inducible enzymes including NOS2 (Cartwright et al., 2007b; Mitchell et al., 2007), leading to hypo-responsiveness to constrictor agents. This impaired vascular contractility is independent of the endothelium (Julou-Schaeffer et al., 1990; Szabo et al., 1993) and associated specifically with the induction of NOSII in the VSM component of blood vessels (Bishop-Bailey et al., 1997)
In vivo, the host is exposed to multiple ligands derived from Gram-negative bacteria, including both NOD1 and TLR4 agonists. Here, we have identified some similarities and differences in how VSM cells and macrophages sense Gram-negative PAMPs. Our findings show a novel and selective role for NOD1 receptors in VSM cells. Our data also show that the signalling pathways resulting in NOS2 induction are separate for NOD1 and TLR4, but converge at the level of NF-κB and MAPK. An effective host response to infection and injury relies on the ability to regulate the spatial and temporal patterns of the signalling pathways activated during the inflammatory response. The data presented here suggest that NOD1 is a central PRR pathway in VSM cells, and highlight NOD1 as a potential novel therapeutic target for the treatment of vascular inflammation.
This study was funded in part by research awards from the British Heart Foundation (FS/09/017/26810) and the Wellcome Trust (083429/Z/07/Z). LM was a recipient of a Marie Curie Intra-European Fellowship within the 6th European Community Framework Programme.
The authors declare no conflicts of interest.
Additional Supporting Information may be found in the online version of this article:
Figure S1 Effects of the NOD1 agonist FK565 on COX and NOS activity in cultured VSM cells. At 24 h, FK565 (0.1–100 nM) induced a concentration-dependent increase in both COX (A, measured by the accumulation of 6-ketoPGF1α) and NOS2 (B, measured by the accumulation of nitrite) activity in VSM cells. Data are expressed as mean ± SEM. n = 6. (Inset) Representative Western blots of COX2 (≈72 kDa) and β-actin (≈47 kDa) from VSM cells.
Table S1 Effects of drug treatments on VSM cell viability assayed by MTT assay. Results shown are selected concentrations (μM) from experiments where drugs were found to affect viability in either VSM cells or J774 macrophages. Where concentrations of drugs are not shown on S1 (or S2), no effect on cell viability was seen. Data are expressed as mean absorbance values ± SEM, and were analysed using one-way ANOVA followed by Dunnett's post-test; * denotes P < 0.05 for each inhibitor compared to respective control. n = 6–12.
Table S2 Effects of drug treatments on J774 cell viability assayed by MTT assay. Results shown are selected concentrations (μM) from experiments where drugs were found to affect viability in either VSM cells or J774 macrophages. Where concentrations of drugs are not shown on S1 (or S2), no effect on cell viability was seen. Data are expressed as mean absorbance values ± SEM, and analysed using one-way ANOVA followed by Dunnett's post-test; * denotes P < 0.05 for each inhibitor compared to respective control. n = 6–18.
Table S3 Densitometry analysis for NOS2 protein in cells treated with various inhibitors. Results were normalized by β-actin and expressed as a percentage of the response to FK565 (10 nM) or LPS (1 μg·mL–1). Data were analysed using one sample t-test; * denotes P < 0.05 for each inhibitor compared to respective control. n = 3–5. Inhibitors were used at the concentration shown in parentheses (μM).
Table S4 Effect of signalling pathway inhibitors on the ability of Escherichia coli (3 × 107 CFU·mL–1), FK565 (10 nM) and LPS (1 μg·mL–1) to induce nitrite production in VSM cells and macrophages. Results were expressed as a percentage of the corresponding response to E. coli, FK565 or LPS alone. Results were analysed using one sample t-test, and * denotes P < 0.05 for treatments versus respective control (100%). n = 6–15. Inhibitors were used at the concentration shown in parentheses (μM).
Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.