OUP user menu

Journal of Crohn's and Colitis: 10 (9)

Editor-in-Chief

Laurence J. Egan, Ireland

Associate Editors

Shomron Ben-Horin, IsraelSilvio Danese, ItalyPeter Lakatos, HungaryMiles Parkes, UKJesús Rivera-Nieves, USABritta Siegmund, GermanyGijs van den Brink, NLSéverine Vermeire, Belgium

6.585
5.586

Published on behalf of

NOD2 activity modulates the phenotype of LPS-stimulated dendritic cells to promote the development of T-helper type 2-like lymphocytes — Possible implications for NOD2-associated Crohn's disease

Matt Butler, Rakesh Chaudhary, David A. van Heel, Raymond J. Playford, Subrata Ghosh
DOI: http://dx.doi.org/10.1016/j.crohns.2007.08.006 106-115 First published online: 1 December 2007

Abstract

Sensing of commensal microorganisms via Toll-like receptors (TLR) in the gut is essential for maintaining intestinal homeostasis in healthy individuals. Conversely, Crohn's disease is characterised by an inappropriate T helper-type 1 (Th1)-mediated immune response towards these same microorganisms. NOD2 is expressed by dendritic cells (DC) and mediates responses to bacterial muramyl-dipeptides (MDP). Mutations in NOD2 (CARD15) have recently been associated with susceptibility to Crohn's disease although the underlying mechanisms have yet to be established. We investigated the functional outcome of NOD2 and TLR4-mediated activation in monocyte-derived DC from wild-type NOD2 healthy controls and NOD2 frame-shift mutation-carrying Crohn's disease patients. In wild-type DC, MDP acted synergistically with LPS to amplify inflammatory cytokine production, enhance co-stimulatory molecule expression, and produce DC that promoted the proliferation of naïve, allogeneic, CD4+ T lymphocytes with a Th2-like cytokine profile. By contrast, DC carrying homozygous NOD2 mutations were unable to react to MDP, responded to LPS only, and promoted the development of Th1 cells. These results suggest activation of the NOD2 pathway in DC modulates their response to TLR agonists and regulates their ability to induce polarised Th1 responses. As a consequence, Crohn's disease patients with defective NOD2 may be predisposed to the generation of strongly polarised Th1 responses against common commensal microorganisms.

1 Introduction

Muramyl dipeptide (MDP) is a conserved moiety of peptidoglycan present in a range of microorganisms and is the minimal essential structure recognised by the intracellular pattern recognition receptor, NOD2 (CARD15).1 NOD2 is expressed by epithelial cells, monocytes and dendritic cells (DC) and activation by MDP leads to NF-κB induction via the serine–threonine kinase, Rip2, and IKKγ (NEMO) pathways.24 Cellular responses to MDP in DC include elevated cytokine secretion, raised expression of membrane-bound major histocompability complexes (MHC) and enhanced co-stimulatory molecule expression which are essential for the generation and differentiation of effector T lymphocytes.57 Interest in MDP has recently been kindled by the strong association of Crohn's disease with three single nucleotide polymorphisms, two mis-sense variations (R702W, G908R), and one frame-shift mutation (1007fs) in the NOD2 gene located within the IBD1 locus on chromosome 16.8,9 Currently, the majority of human data support a loss-of-function effect for these mutations.1,1014

We, and others, recently reported that the TLR and NOD2 signalling pathways appear to interact or ‘synergise’, resulting in amplified cytokine responses upon co-administration of TLR agonists and MDP.6,10,11 This co-operative response is reduced in PBMC from R702W homozygotes, as well as R702W/1007fs or G908R/1007fs double heterozygotes, but is completely abrogated in cells from patients homozygous for the 1007fs (3020insC) mutation.11 NOD2/TLR cross-talk has also been implicated by work from Watanabe et al. who found mice carrying loss-of-function mutations in NOD2 displayed enhanced TLR2-mediated NF-κB activation and Th1 responses.15 Splenocytes from these mice produced elevated levels of IL-12 in response to peptidoglycan suggesting that intact NOD2 activity might have an inhibitory effect on TLR-induced induction of IL-12 production as well as on ‘downstream’ Th1 cytokines. Together these data point to a regulatory role for NOD2 via its ability to modulate the signals received via TLRs.

In this study we have investigated the possibility that NOD2 activity modulates the outcome of TLR signalling to alter DC phenotype and function and, as a result, adaptive immune responses. DC control the development of adaptive immunity by influencing the Th1/Th2 cytokine profile of responding T lymphocytes. By using DC from NOD2 wild-type donors and Crohn's patients homozygous for the 1007fs NOD2 mutation, we have been able to directly compare the effects of NOD2 signalling on DC-mediated T cell polarisation. This is especially relevant in the pathogenesis of Crohn's disease which is characterised by a distinct polarisation of the adaptive immune response towards Th1 with over-expression of IL-12 and IFNγ.16

2 Materials and methods

2.1 Donor blood

Healthy blood donors were screened for the most common NOD2 mutations — R702W, G908R, 1007fs (3020insC) and those homozygous for wild-type NOD2 alleles were selected as healthy controls. Blood was also obtained by informed consent from Crohn's patients homozygous for the NOD2 1007fs mutation. At the time of the study, all patients had non-active Crohn's disease and were without medication.

2.2 Cell purification and differentiation of monocyte-derived DC

Peripheral blood mononuclear cells (PBMC) were separated from blood samples by centrifugation on a ficoll-hypaque (Nycomed, UK) density gradient according to the manufacturer's instructions. CD14+ monocytes were isolated by positive selection using MACS beads (Miltenyi Biotech, Germany) as per the manufacturer's instructions. CD14+ monocytes were suspended in X-VIVO 15 serum-free medium (Biowhittaker, USA) and were transferred to 6 well tissue culture plates at a concentration of 3–5 × 105 cells per ml. To induce differentiation, cells were given 50 ng/ml IL-4 and 50 ng/ml GM-CSF (First Link Ltd, UK) on day 0 and day 4. Cells were used for experiments at day 6.

2.3 MDP and LPS stimulation

Immature DC were plated at 2.5 × 105 cells per well of a 24 well tissue culture plate in 500 μL of X-VIVO 15 serum-free culture medium (Biowhittaker). Cells were stimulated for 24 h with a range of concentrations (0.1–100 ng/ml) of gel-purified LPS from Escherichia coli 055:B5 (previously shown to lack NOD2 agonist activity).1 Stimulation was carried out in the presence or absence of 100 ng/ml muramyl dipeptide (pharmaceutical grade MDP-Lys18/romurtide; kind gift from Daiichi Pharmaceutical Company) after DC showed maximal cytokine responses (as assessed by TNFα, and IL-10 production) to MDP alone at 100 ng/ml after 24 hour incubation with a range of MDP concentrations (0.1–1000 μg/ml; not shown). In some experiments 25 ng/ml IFNγ (First Link, UK) was added at the same time as MDP (not shown).

2.4 Cytokine measurement

Supernatants were analysed for cytokine content by capture ELISA. TNFα, IL-10, IL-12p70, IFNγ, and IL-13 were detected using antibody pairs (Phamingen, BD Biosciences, UK). Briefly, immunoabsorbant plates (Nunc, UK) were coated with 1 μg/ml capture antibody and incubated for 2 h at 37 °C. Plates were washed in PBS/0.1%Tween 20 before incubating with samples and standards (Peprotech, UK) overnight. Plates were next incubated with the appropriate biotinylated detecting antibody for 2 h at 37 °C. Finally, results were visualised by incubation for 1 h with 0.1 μg/ml avidin–horseradish peroxidase (Biosource Int., USA) followed by addition of substrate solution, TMB (Zymed, USA). Colour development was stopped by the addition of 0.5 M H2SO4 and plates were read for optical density at 450 nm (ELISA plate reader; Titertek Mulitskan). Samples were analysed in triplicate and expressed as mean concentrations +/− standard deviation. Th1 and Th2 cytokines were detected using a flow cytometry-based Multiplex assay system (Bender Medsystems, Vienna, Austria) as per the manufacturer's instructions.

2.5 FACS analysis

DC were washed 3 times in cold FACS buffer (PBS, 3% FCS, 0.02% NaN3) and then stained with the appropriate, FITC-conjugated, antibodies – MHC class II (DakoCytomation, UK), CD80 (Pharmingen, BD), CD86 (Pharmingen, BD), and CD40 (Diaclone, UK) – at a final concentration of 1 μg/ml for 30 min on ice and in the dark. Isotype control antibodies were mouse IgG1 (MOPC21, Sigma-Aldrich) and mouse IgG2a (Sigma-Aldrich). Cells were washed 3 times in FACS buffer and analysed immediately by flow cytometry (FACScalibur, Becton Dickinson).

2.6 Mixed lymphocyte reactions and T cell proliferation

DC were harvested and plated out in 96-well round bottomed plates at 104 cells per well in 100 μl of X-VIVO 15 serum-free medium. Allogeneic CD4+CD45RA+T cells (isolated from PBMCs of wild-type NOD2 healthy volunteers by negative selection) were used as responders and added at 1 × 105 cells/well in 100 μl medium. Plates were incubated for 4 days. 18 h prior to harvesting, 1 μCi of tritiated methyl thymidine (Amersham Int., UK) was added to each well. Cells were harvested on to glass fibre filter mats (Wallac, UK) and thymidine incorporation was measured by liquid scintillation spectroscopy in a beta counter (1205 Betaplate counter, Wallac).

3 Results

To test our hypothesis, DC from healthy wild-type NOD2 donors and Crohn's patients (homozygous for NOD2 1007fs) were analysed for their ability to respond to either highly purified LPS (previously shown to lack NOD2 activity), MDP-Nopia (endotoxin-free), or a combination of MDP and LPS. Both wild-type and patient DC displayed dose-dependent TNFα, IL-10, and IL-12p70 responses to LPS (Fig. 1a). Compared to cells from healthy donors, patient DC consistently elicited higher cytokine responses to LPS alone (Fig. 1b; p < 0.05 for TNFα by Mann Whitney U test) suggesting these cells possessed elevated baseline levels of activation. This effect was not related to NOD2 status and has been previously reported.1721 In the presence of MDP, wild-type NOD2 DC consistently produced significantly elevated TNFα and IL-10 levels in response to LPS (Fig. 1b; p < 0.05 Wilcoxon matched ranks test). MDP-mediated enhancement was detectable at all doses of LPS but was maximal at, or below, the peak response seen to LPS alone. By contrast, MDP failed to induce significant increases in the IL-12p70 response to LPS unless given in the presence of 25 ng/ml IFNγ (not shown). Similar results were observed using cells from five other NOD2 wild-type donors (Fig. 1b). DC homozygous for the NOD2 1007fs mutation all failed to respond to MDP— even in the presence of IFNγ. (Fig. 1a and b). These results led us to hypothesise that NOD2 activity might have an effect on the polarisation of T cell cytokine responses by influencing the balance of IL-10 and IL-12p70 production — cytokines known to affect the development of Th1, Th2, and Th3/Treg cells.22

Figure 1

Inflammatory cytokine analysis: (a) Representative data using DC from a NOD2 wild-type or NOD2 1007fs donor. DC were stimulated for 24 h with a range of LPS doses (0–100 ng/ml) +/− 100 ng/ml MDP. The presence of MDP enhanced TNFα and IL-10 production at all LPS concentrations in NOD2 wild-type cells only. Results are mean values +/− SD of triplicate samples. (b) Accumulated data (n = 6 NOD2 wild-type, and n = 6 NOD2 1007fs donors) showing DC responses to 10 ng/ml LPS +/− 100 ng/ml MDP. p < 0.05 by Wilcoxon matched pairs test.

We next analysed DC surface expression of major histocompatability complex class II (MHC II) and the co-stimulatory molecules CD80, CD86 and CD40, since co-stimulatory molecules are also important factors governing T cell polarisation.22,23 In both NOD2 wild-type and NOD2 1007fs DC, LPS alone induced increases in MHC class II, CD80, CD86, and CD40 (Fig. 2). This was enhanced by the presence of MDP in NOD2 wild-type cells (particularly for CD80 and CD40). However, while NOD2 1007fs cells appeared to respond normally to LPS stimulation, they were unable to respond to MDP. These data suggested that NOD2 activity could alter the co-stimulatory profile of responding DC — an effect lost in cells carrying the NOD2 1007fs mutation.

Figure 2

Phenotypic analysis: DC from NOD2 wild-type and NOD2 1007fs donors were treated with 10 ng/ml LPS +/− 100 ng/ml MDP for 24 h. Cells were subsequently analysed for MHC class II, CD80, CD86 and CD40 expression by flow cytometry. In both sets of cells, LPS induced the elevated expression of all markers. However, only NOD2 wild-type DC showed increased expression following the addition of MDP. Results are representative data and show mean fluorescence intensities for each phenotypic marker.

Finally, we analysed the functional outcome of these phenotypic changes by using stimulated DC (24 hour treatment followed by washing) as stimulators in an allogeneic MLR with CD4+CD45RA+ cells as responders (Fig. 3). In accordance with the flow cytometric data, NOD2 wild-type and NOD2 1007fs DC possessed enhanced stimulatory capacities following treatment with 10 ng/ml LPS only. In addition, the stimulatory capacity of NOD2 wild-type DC was further enhanced following treatment with LPS and MDP. By contrast, the stimulatory capacity of NOD2 1007fs DC was unaffected by stimulation with MDP. Taken together these data suggested that, while they are unable to respond to MDP, NOD2 1007fs DC could respond normally to TLR4 ligands in terms of their ability to make cytokines, undergo phenotypic maturation, and stimulate T cells.

Figure 3

Stimulatory capacity in MLR: Proliferation of 105 allogeneic CD4+CD45RA+ T cells to 104 DC following treatment with 10 ng/ml LPS, 100 ng/ml MDP, or LPS and MDP in combination in a 5 day MLR. Both sets of DC induced elevated levels of proliferation following treatment with LPS. Only NOD2 wild-type DC induced elevated levels of proliferation in the presence of MDP. Results are representative data and show the mean of 5 data points per condition +/− SD. p < 0.05 by Wilcoxon matched pairs test.

While NOD2-mediated DC maturation has been shown to result in enhanced T cell stimulatory capacity, it is currently unclear how this affects T-helper cell cytokine profiles. We thus analysed the type of cytokines released from stimulated CD4+CD45RA+ T cells following interaction with DC pre-treated for 24 h with 10 ng/ml LPS +/− 100 ng/ml MDP. A 1:10 ratio of DC was used to stimulate allogeneic CD4+CD45RA+ cells for 4 days after which the supernatants were by analysed using a multiplex cytokine bead assay (Fig. 4a and b). Following treatment with LPS alone, both NOD2 wild-type and NOD2 1007fs DC induced T cells to produce elevated levels of the Th1 cytokines IL-1β, IL-8, and TNFα. However, stimulation of wild-type DC with LPS in the presence of MDP resulted in significant reductions in IFNγ, IL-1β and IL-8 (Fig. 4a), and increased levels of the Th2 cytokines IL-4, IL-5, IL-6, IL-10 and IL-13 (Fig. 4b). Interestingly, these cells also produced elevated levels of the Th1-associated TNFα and TNFβ (Fig. 4a). Conversely, prior stimulation with MDP had no effect on the Th1/Th2 cytokine balance induced by NOD2 1007fs DC while LPS-induced Th1 induction was unaffected. These cytokine responses were confirmed as T cell-derived by the detection of similar cytokine profiles from cultures using irradiated DC as stimulators (not shown).

Figure 4

Th1/Th2 polarisation: DC treated with 10 ng/ml LPS, 100 ng/ml MDP, or LPS and MDP in combination were used as stimulators in a 4 day MLR using 105 CD4+CD45RA+ T cells as responders. Supernatants were harvested and analysed for (a) Th1-like cytokines and, (b) Th2-like cytokines, by multiplex cytokine array. (c) DC from 6 healthy donors and 6 NOD2 1007fs patients were used as stimulators in a 4 day MLR. Supernatants were analysed by capture ELISA for IFNã or IL-13 production as markers for Th1 and Th2 differentiation respectively. Data are means of triplicate samples for each donor. p < 0.05 by Wilcoxon matched pairs test.

To confirm these results, DC from 6 NOD2 wild-type healthy controls and 6 NOD2 1007fs Crohn's patients NOD2 were analysed for their ability to induce IFNγ (Th1) or IL-13 (Th2) secretion by naïve T cells (Fig. 4c). By this analysis, wild-type DC stimulated with LPS alone promoted increased production of both IFNã and IL-13. However, wild-type DC stimulated with LPS in the presence of MDP promoted a greatly reduced level of IFNγ and significantly elevated levels of IL-13 (p < 0.05 Wilcoxon matched ranks test) suggesting a polarisation towards Th2. Conversely, following stimulation with LPS alone, NOD2 1007fs DC induced high levels of IFNγ with only small increases in IL-13 suggesting a strong polarisation towards Th1. This polarisation was maintained even when DC were stimulated in the presence of MDP. Interestingly, DC from Crohn's patients induced significantly more IFNγ from naïve T cells as compared to healthy control DC (p < 0.005 Mann Whitney U test) — particularly following stimulation with LPS. Conversely, patient DC tended to induce less IL-13 production in responding T cells compared to healthy control DC — although this failed to reach statistical significance). These differences were unrelated to NOD2 genotype and appeared to reflect the general enhanced baseline response of cells isolated from Crohn's patients.

Taken together, these data indicate that intact NOD2 activity could modulate DC phenotype and function to elicit an altered adaptive T cell response and that this activity is likely to be lost in individuals homozygous carrying the NOD2 1007fs mutation.

4 Discussion

Bacterial sensing by immune cells is likely to involve multiple receptors and activation of numerous signalling pathways. Understanding how these signals interact to produce functional cellular responses is clearly an important goal. In this study we investigated the response of DC to simultaneous activation of the TLR4 and NOD2 bacterial sensing pathways. By utilising DC from Crohn's patients carrying homozygous loss-of-function mutations in NOD2 (the 1007 frameshift mutation; 3020insC), we were able to observe the effect of defective innate signalling on the development of the adaptive immune response.

Our initial results showed that both NOD2 wild-type and NOD2 1007fs DC could respond normally to TLR4 agonist (LPS) in terms of cytokine production, MHC expression, co-stimulatory molecule expression, and stimulatory capacity. Interestingly, patient DC routinely displayed elevated baseline levels of cytokine production and often displayed enhanced cytokine responses upon LPS stimulation. Patient DC also tended to express higher levels of CD86 and were more stimulatory in the allogeneic MLR. Many of these specific differences have been reported by other investigators and do not appear to be linked to NOD2 genotype.1721 In addition, a number of groups have shown that cellular responses to TLR pathway activation are unaffected by Crohn's-associated NOD2 mutations.7,15,24

In NOD2 wild-type DC, the combination of LPS and MDP consistently resulted in significant increases in inflammatory cytokine production with particularly large increases in IL-10 production. Conversely, LPS-stimulated DC were poor producers of IL-12p70 — even in the presence of MDP. Since IL-12 and IL-10 are critical cytokines controlling the development of Th1 and Th2 responses respectively, the enhancement of IL-10 – but not IL-12 secretion – by MDP suggested these DC might have divergent effects on their ability to drive T cell polarisation.22,23 A number of groups have reported enhanced levels of IL-12 following stimulation by TLR agonists in the presence of MDP6,7 — although in each case very high concentrations (∼1 μg/ml) of MDP were used. Moreover, these studies used a form of MDP lacking the lysine tail of MDP-Nopia and consequently, different forms of MDP may trigger alternative pathways of cellular uptake and pathway activation. In correspondence with published data, we were able to show MDP-induced augmentation of IL-12p70 production but only when DC were treated with LPS and MDP in the presence of IFNγ (not shown). It is unlikely this type of stimulation reflects physiological conditions and the presence of IFNγ automatically drives T cells towards a Th1 phenotype. We therefore pursued the effects of MDP and LPS in the absence of other stimulatory pathway activation. In contrast to wild-type DC, NOD2 1007fs DC produced IL-10 and IL-12 only in response to LPS and were unable to respond to MDP — even in the presence of IFNγ. Crucially, unlike the NOD2 wild-type DC, these cells also failed to show any bias towards IL-10 secretion on stimulation. These results corresponded with conclusions reached by Netea et al. and Watanabe et al. who found that mononuclear cells from either patients carrying the 1007fs NOD2 mutation, or Card15−/− mice, had defective IL-10 release after stimulation via TLR2 and developed enhanced Th1 responses.15,2527

Analysis of DC phenotype and function showed NOD2 wild-type cells could respond to a combination of LPS and MDP by upregulating their expression of MHC and co-stimulatory molecules (particularly CD80) while NOD2 1007fs DC were only able to upregulate these molecules in response to LPS stimulation alone. The contribution of co-stimulatory molecules to Th1/Th2 development is still a contentious issue but a number of studies have implicated CD80 in the development of Th2 or regulatory T cell responses.2830 Indeed, evidence increasingly supports the view that CD80 is a more effective ligand for the immuno-regulatory receptor CTLA-4 on T cells30 — although more work is clearly required to investigate the expression of other co-stimulatory molecules on DC following NOD2 pathway activation. Consistent with the phenotypic changes, NOD2 wild-type DC treated with LPS and MDP displayed enhanced stimulatory capacity while NOD2 1007fs DC could only show increased stimulation following activation by LPS.

Finally, we analysed the Th1/Th2 polarisation of naïve CD4+ T cells following stimulation by DC treated with MDP, LPS, or MDP and LPS in combination. LPS stimulation alone resulted in NOD2 wild-type DC which promoted the development of IFNγ, IL-8, and IL-1β-secreting Th1-like cells. However, DC stimulated in the presence of LPS and MDP induced T cells which predominantly secreted IL-4, IL-5, IL-6, IL-10, IL-13, (and, interestingly, TNFα) with a concomitant down-regulation of the Th1 cytokines, IFNγ, IL-8, and IL-1β. The same analysis using NOD2 1007fs DC showed the same Th1 bias (elevated IFNγ, IL-8, and IL-1β) in response to LPS, even in the presence of MDP. Thus, while TLR activation of NOD2 wild-type DC generally resulted in Th1 polarisation,25 the additional activation of the NOD2 signalling pathway appeared to modulate this and produce a Th2-like response. With the enhanced levels of TNFα, the T cell cytokine pattern induced by MDP-treated NOD2 wild-type DC actually resembled the unique Th2 phenotype described for pro-allergic (high TNFα, high IL-4, high IL-13 and relatively low IL-10), rather than ‘classical’ Th2 cells (low TNFα, high IL-4, high IL-13 and high IL-10).28,31 It is tempting to speculate that these Th2 ‘effector’ cells might play a role in immunity to particular types of gut pathogen such as gram positive bacteria or even parasitic worms (both of which are rich in peptidoglycans and thus NOD2 agonists). Furthermore, it is worth noting that TNFα is known to have an important role in regulating Th2-cytokine mediated responses at mucosal sites during parasitic worm infestation32 and helminths have shown efficacy in treating Crohn's disease.33

Whether the NOD2-mediated polarisation of the adaptive immune response represents a mechanism of innate tolerance induction or a means of eliminating specific pathogens remains unclear at this stage. However, recent studies indicate TLR expression is relatively low in the healthy gut34 and, since NOD2 is expressed by epithelial cells under normal conditions, the balance of bacterial sensing in a healthy gut could conceivably occur via NOD-like receptors (NLRs) such as NOD1 and 2, with only minimal activation of TLR pathways. In accordance with our data, this might result in DC which promote Th2-like responses, induce IgA production, and regulate intestinal responses to harmless enteric antigens. In an inflamed gut, TLR expression and activity appear to increase.34 In this setting, mucosal DC could promote the development of Th1 cells as a result of predominant TLR-mediated (rather than NOD-dominated) bacterial antigen detection.

We are currently investigating the means by which MDP mediates its immuno-modulatory effects in DC — whether by altering the IL-10/IL-12p70 balance or by altered co-stimulatory molecule expression of activating DC. However, identification of NOD2 as a Th2-promoting bacterial sensor is further evidence that the immune system ‘tailors’ its responses via consolidation of signals received through multiple pattern recognition receptors. Since microorganisms consist of a wide range of potential TLR agonists, it will be important to determine whether the NOD2 modulatory effect affects each TLR pathway and persists in the face of more than one TLR signal. Recent reports suggest many TLRs can act synergistically with each other during responses to multiple TLR agonists, leading to increased production of IL-12 and IL-23 by DC and induction of enhanced and sustained Th1 polarisation.25 In addition, high concentrations of TLR agonist may override the modulatory effect of NOD2 stimulation since preliminary observations suggest the effects of MDP on T cell polarisation are lost when high concentrations of LPS (> 100 ng/ml) are used (M. Butler — unpublished observations).

While it is clear NOD2 acts synergistically with TLR signalling to enhance NF-κB activation and pro-inflammatory cytokine secretion,11 this runs counter to the fact that NOD2 mutations apparently lead to the development of Crohn's disease and strongly polarised Th1 immune responses.1 Our present study offers a plausible explanation by showing that NOD2-mediated MDP synergy with TLRs produces a quantitatively different adaptive immune response resulting in DC with Th2-polarising properties. In Crohn's patients carrying loss-of-function mutations in NOD2, this immuno-modulatory effect of NOD2 signalling is lost, potentially leading to the persistent induction of Th1 cells in the presence of TLR agonists such as LPS. It thus seems likely that defects in NOD2 bacterial sensing deliver a ‘double hit’ — weakening the innate immune response by limiting inflammatory cytokine and α-defensin release,35,36 whilst simultaneously allowing the development of strongly polarised Th1-dominated adaptive immune responses towards these same enteric organisms. The result could be an over-aggressive adaptive immune response to common enteric microorganisms and the foundation of inflammatory bowel disease.

Acknowledgements

This work was partially funded by the Eli and Edythe L. Broad Foundation and the National Association of Crohn's and Colitis, UK.

Abbreviations
CARD15
caspase recruitment domain 15
DC
dendritic cell
LPS
lipopolysaccharide
MDP
muramyl dipeptide
NOD2
nucleotide oligomerization domain-2
PBMC
peripheral blood mononuclear cell
PRR
pattern recognition receptor
Th1/2
T helper-type 1/2
TLR
Toll-like receptor.

References

View Abstract