Article Text

Download PDFPDF

Crohn's disease-associated ATG16L1 polymorphism modulates pro-inflammatory cytokine responses selectively upon activation of NOD2
  1. Theo S Plantinga1,2,
  2. Tania O Crisan1,2,
  3. Marije Oosting1,2,
  4. Frank L van de Veerdonk1,2,
  5. Dirk J de Jong3,
  6. Dana J Philpott4,
  7. Jos W M van der Meer1,2,
  8. Stephen E Girardin5,
  9. Leo A B Joosten1,2,
  10. Mihai G Netea1,2
  1. 1Department of Medicine, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands
  2. 2Nijmegen Institute for Infection, Inflammation and Immunity (N4i), Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands
  3. 3Department of Gastroenterology and Hepatology, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands
  4. 4Department of Immunology, University of Toronto, Toronto, Canada
  5. 5Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Canada
  1. Correspondence to Dr Mihai G Netea, Department of Medicine, Radboud University, Nijmegen Medical Centre, Internal postal code 463, P.O. Box 9101, Geert Grooteplein 8, 6500 HB Nijmegen, The Netherlands; m.netea{at}aig.umcn.nl

Abstract

Objective Autophagy has recently been shown to modulate the production of pro-inflammatory cytokine production and to contribute to antigen processing and presentation through the major histocompatibility complex. Genetic variation in the autophagy gene ATG16L1 has been recently implicated in Crohn's disease pathogenesis. The mechanisms underlying this association are not yet known, although experimental models suggest an inhibitory effect of autophagy on interleukin 1β (IL-1β) responses. Here, the effect of ATG16L1 genetic variation on cytokine responses has been assessed in humans.

Design and setting Peripheral blood mononuclear cells from healthy individuals and patients with Crohn's disease with different ATG16L1 genotypes were stimulated with ligands for Toll-like receptor 2 (TLR2), TLR4 and nucleotide-binding oligomerisation domain 2 (NOD2), with or without the autophagy inhibitor 3-methyladenine. Induction of cytokine production and related factors were measured at the mRNA and protein level. Furthermore, protein levels of ATG16L1 were assessed by western blot.

Results The present study demonstrates that cells isolated from individuals bearing the ATG16L1 Thr300Ala risk variant, which is shown to affect ATG16L1 protein expression upon NOD2 stimulation, display increased production of the pro-inflammatory cytokines IL-1β and IL-6, specifically after stimulation with NOD2 ligands. In contrast, no differences were found when cells were stimulated with TLR2 or TLR4 agonists. These findings were confirmed in two independent cohorts of volunteers and in a group of patients with Crohn's disease. The increased production could be ascribed to increased mRNA expression, while processing of pro-IL-1β by caspase-1 activation was not affected. The effect of the ATG16L1 polymorphism was abrogated when autophagy was blocked.

Conclusions The present study is the first to link the ATG16L1 polymorphism with an excessive production of IL-1β and IL-6 in humans, which may explain the effects of this polymorphism on the inflammatory process in Crohn's disease.

  • Autophagy
  • ATG16L1
  • pro-inflammatory cytokines
  • NOD2
  • genetic polymorphisms
  • immune response
  • inflammatory bowel disease
  • interleukins
View Full Text

Statistics from Altmetric.com

Significance of this study

What is already known about this subject?

  • Genetic variation in ATG16L1 (non-synonymous polymorphism T300A) has been implicated in susceptibility to Crohn's disease.

  • The T300A variant in ATG16L1 has been demonstrated to affect the autophagy process induced by the pattern-recognition receptor NOD2.

  • NOD2 has been shown to induce autophagy by interacting physically with ATG16L1.

  • A potential link has been described between ATG16L1 and pro-inflammatory cytokine responses in mice.

What are the new findings?

  • Inhibition of the autophagy process leads to increased pro-inflammatory cytokine responses in human primary immune cells when stimulated with a ligand for nucleotide-binding, oligomerisation domain 2 (NOD2).

  • The risk allele of the ATG16L1 T300A variant also increases these responses when cells are specifically triggered with a NOD2 ligand.

  • The T300A polymorphism affects the expression of the ATG16L1 protein, which consequently could influence the balance between induction of autophagy versus cytokine responses by NOD2.

  • In conclusion, genetic variation in ATG16L1 is associated with higher production of pro-inflammatory cytokines that could drive the chronic inflammation observed in Crohn's disease.

How might it impact on clinical practice in the foreseeable future?

  • Patients with Crohn's disease bearing the risk allele of the ATG16L1 T300A polymorphism, of which cells produce more pro-inflammatory cytokines upon NOD2 stimulation, would benefit most from anti-inflammatory therapy by specifically antagonising these cytokines, especially interleukin 1β.

Introduction

Eukaryotic cells are equipped with machinery for collecting and degrading cellular constituents, designated as cellular self-digestion, or autophagy.1 In general, activation of autophagy is triggered by conditions that are threatening for cell survival, including nutritional starvation and hypoxia, and is an important strategy to prevent cell death. Upon activation, cellular components such as cytoplasmic organelles and long-lived proteins are captured into double-membraned vesicles called autophagosomes that fuse with lysosomes for the degradation of its contents.2 Recruitment of autophagy has also been implicated in host defence to intracellular bacteria that can escape from innate recognition by membrane receptors upon engulfment, such as Mycobacterium tuberculosis,3 4 Salmonella typhimurium5 and adherent–invasive Escherichia coli,6 and which is designated as xenophagy.7 In this respect, after the fusion of lysosomes with autophagosomes that have encapsulated bacteria, the bacteria are degraded and the antigens loaded on major histocompatibility complex class II to activate T cell receptor-mediated adaptive immune responses.8 9 One of the main proteins involved in the formation of autophagosomes is autophagy-related 16-like 1 (ATG16L1), which is highly homologous to Atg16 in Saccharomyces cerevisiae. ATG16L1 is a component of a large protein complex, together with ATG5 and ATG12, that is essential for autophagosome formation.

Autophagy, and ATG16L1 in particular, has recently been implicated in the pathogenesis of Crohn's disease. A non-synonymous polymorphism in ATG16L1, Thr300Ala (c.898A→G, rs2241880), was demonstrated to be associated with higher susceptibility to Crohn's disease.10 11 Functional studies have revealed that this ATG16L1 polymorphism affects the phenotype of Paneth cells12 and impairs autophagosome formation specifically after activation of nucleotide-binding oligomerisation domain 2 (NOD2).13 14 However, no abnormalities were observed when autophagy was activated through Toll-like receptor 2 (TLR2) or TLR4 signalling. Recently, it has been demonstrated by Saitoh et al that macrophages derived from ATG16L1 knockout mice, which completely lack autophagy, exhibit elevated interleukin 1β (IL-1β) production after stimulation with lipopolysaccharide (LPS), suggesting an inhibitory role of autophagy on the production of this cytokine.15 While the functional consequences of the ATG16L1 Thr300Ala polymorphism have been demonstrated with regard to the capacity to induce autophagy, no studies have been performed to elucidate its effect on the production of pro-inflammatory cytokines such as IL-1β, which drive the inflammatory process in the inflamed gut mucosa of patients with Crohn's disease.

In the present study we demonstrate that the genetic variant of human ATG16L1, which confers a higher risk for Crohn's disease, is associated with elevated production of pro-inflammatory cytokines after engagement of NOD2. We thereby provide an explanation for the excessive inflammatory response that is observed in Crohn's disease and is caused by microorganisms that reside in the gut. We have assessed production of the pro-inflammatory cytokines IL-1β, tumour necrosis factor (TNF) and IL-6 due to their crucial role for the inflammation in Crohn's disease.

Materials and methods

Genotyping for ATG16L1 Thr300Ala polymorphism

DNA was isolated from whole blood by using the isolation kit Puregene (Gentra Systems, Minneapolis, Minnesota, USA), according to the manufacturer's protocol. Genotyping for the presence of the ATG16L1 Thr300Ala polymorphism was performed by applying the TaqMan single nucleotide polymorphism assay C_9095577_20 on the 7300 ABI Real-Time PCR system (Applied Biosystems, Foster City, California, USA). Two independent cohorts of healthy volunteers have been assessed (N=46 and N=90). We also selected five patients homozygous for the 300Thr allele, five patients homozygous for the 300Ala allele, and five heterozygous patients from a cohort of 74 patients with Crohn's disease, and from these 15 patients we isolated cells for the cytokine stimulation assays. All volunteers gave written informed consent.

Stimulation assays for peripheral blood mononuclear cells

Venous blood was drawn from the cubital vein of healthy volunteers or patients with Crohn's disease into 10 ml EDTA tubes (Monoject, Covidien, Mansfield, Massachusetts, USA). In a separate experiment, cells isolated from four individuals homozygous for the NOD2 3020insC mutation were stimulated in order to assess the role of NOD2 for cytokine production. At time of blood donation, patients with Crohn's disease were in a quiescent phase, defined as a prolonged period of at least 3 months of mild disease without relapses or exacerbations in the absence of immunomodulatory therapy. The mononuclear cell fraction was obtained by density centrifugation of blood diluted 1:1 in pyrogen-free saline over Ficoll-Paque (Pharmacia Biotech, Pittsburgh, Pennsylvania, USA). Cells were washed twice in saline and suspended in culture medium (RPMI; Invitrogen, Carlsbad, California, USA) supplemented with gentamicin 10 μg/ml, l-glutamine 10 mM and pyruvate 10 mM. Cells were counted in a Coulter counter (Coulter Electronics, Brea, California, USA) and the number was adjusted to 5×106 cells/ml. A total of 5×105 mononuclear cells in a 100 μl volume was added to round-bottom 96-well plates (Greiner, Monroe, North Carolina, USA) and incubated with either 100 μl of culture medium (negative control), or the various stimuli: E coli lipopolysaccharide (LPS, 10 ng/ml; Sigma, St Lousi, Missouri, USA), Pam3Cys (10 μg/ml, EMC Microcollections, Tübingen, Germany) or muramyl dipeptide (MDP, 10 μg/ml; Sigma), IL-1 receptor antagonist (10 μg/ml; Amgen, Breda, The Netherlands). Autophagy was inhibited by 3-methyl adenine (3-MA, 10 mM; Sigma) or was activated by Earle's Balanced Salt Solution (EBSS) starvation medium (Invitrogen). Cytokine measurements of IL-1β, IL-6 and TNFα were performed in the supernatants, whereas pro-IL-1β was measured intracellularly. All cytokines were measured after 24 h incubation, using commercial ELISA kits (R&D Systems, Minneapolis, Minnesota, USA (TNFα, IL-1β, pro-IL-1β) or Sanquin, Amsterdam, The Netherlands (IL-6)).

Real-time PCR

Peripheral blood mononuclear cells (PBMCs) stimulated for 4 or 24 h at 37°C were treated with TRIzol Reagent (Invitrogen) and total RNA purification was performed according to manufacturer's instructions. Isolated RNA was subsequently transcribed into complementary DNA using iScript cDNA Synthesis Kit (Bio-Rad, Hercules, California, USA) followed by quantitative PCR using the SYBR Green method (Applied Biosystems). The following primers were used: for IL-1β, forward 5′-GCCCTAAACAGATGAAGTGCTC-3′ and reverse, 5′-GAACCAGCATCTTCCTCAG-3′; for ATG16L1 forward, 5′-ACGTACCAAACAGGCACGAG-3′ and reverse, 5′-CAGGTCAGAGATAGTCTGCAAAC-3′. Data were corrected for expression of the housekeeping gene β2 microglobulin, for which the primers forward 5′-ATGAGTATGCCTGCCGTGTG-3′ and reverse 5′-CCAAATGCGGCATCTTCAAAC-3′ were used.

Western blot

For western blotting, 5×106 cells were lysed in 100 ml of lysis buffer (50 mM Tris (pH 7.4), 150 mM NaCl, 2 mM EDTA, 2 mM EGTA, 10% glycerol, 1% Triton X-100, 40 mM α-glycerophosphate, 50 mM sodium fluoride, 200 mM sodium vanadate, 10 mg/ml leupeptin, 10 mg/ml aprotinin, 1 mM pepstatin A, and 1 mM phenylmethylsulfonyl fluoride). The homogenate was frozen, then thawed and centrifuged at 4°C for 10 min at 15 000×g, and the supernatant was taken for western blot analysis. Equal amounts of protein were subjected to SDS-PAGE using 10%, 12% and 15% polyacrylamide gels at a constant voltage of 100 V. After SDS-PAGE, proteins were transferred to a nitrocellulose membrane (0.2 mm). The membrane was blocked with 5% (wt/vol.) milk powder in TBS/Tween 20 for 1 h at room temperature, followed by incubation overnight at 4°C with a caspase-1 p10 antibody (SC-515; Santa Cruz Biotechnology, Santa Cruz, California, USA) in 5% BSA/TBS/Tween 20 or with an ATG16L1 antibody (NB110-60928, Novus Biologicals, Colorado, USA) in 5% milk powder in TBS/Tween 20. After overnight incubation, the blots were washed three times with TBS/Tween 20 and then incubated with HRP-conjugated swine anti-rabbit antibody at a dilution of 1:5000 in 5% (wt/vol.) milk powder in TBS/Tween 20 for 1 h at room temperature. After being washed three times with TBS/Tween 20, the blots where developed with ECL (GE Healthcare, Pittsburgh, Pennsylvania, USA) according to the manufacturer's instructions. The intensity of the bands on the western blots was assessed by Image Lab statistical software (Bio-Rad).

Statistical analysis

Differences in cytokine production capacity between groups were analysed using the Mann–Whitney U test. Data on mRNA expression levels and immunoblot intensity values were statistically tested with the Student t test. Differences were considered statistically significant at p<0.05.

Results

Autophagy blockade and activation modulates MDP-induced IL-1β

To assess the role of autophagy in the induction of IL-1β by NOD2 signalling, PBMCs obtained from healthy volunteers were stimulated with the NOD2 ligand MDP alone or in combination with the autophagy inhibitor 3-MA for 24 h (figure 1A). The production of IL-1β was higher when cells were stimulated with both MDP and 3-MA compared to MDP alone. In contrast, when autophagy was induced in cells by culturing in starvation medium (EBSS), the MDP-induced IL-1β response was decreased compared to MDP stimulation of cells in nutrient-rich medium (figure 1B).

Figure 1

Autophagy, ATG16L1 genotype and cytokines. (A) IL-1β production by PBMCs after stimulation for 24 h at 37°C with the NOD2 ligand MDP, in the absence or presence of 3-MA, an inhibitor of autophagy (N=4). Values are mean±SD, *p<0.05. (B) IL-1β mRNA expression after stimulation of PBMCs unstimulated or stimulated with MDP, cultured either in nutrient rich medium (RPMI) or in EBSS starvation medium (N=4). Values are means±SD, *p<0.05. (C) Cytokine production capacity of TNFα, IL-1β and IL-6 by PBMCs obtained from healthy volunteers after stimulation for 24 h with MDP, Pam3Cys or LPS, stratified for ATG16L1 Thr300Ala genotype (A allele equals 300Thr allele, G allele equals 300Ala allele). For MDP stimulation PBMCs from two independent cohorts of healthy volunteers are shown, for LPS and Pam3Cys stimulation only the largest cohort. Values are means±SEM, *p<0.05, **p<0.005. (D) Cytokine production capacity of IL-1β by PBMCs obtained from patients with Crohn's disease with functional NOD2 after stimulation for 24 h with MDP, Pam3Cys or LPS, stratified for ATG16L1 Thr300Ala genotype (A allele equals 300Thr allele, G allele equals 300Ala allele). Values are means±SEM, *p<0.05. (E) IL-1β dependent IL-6 production. PBMCs were stimulated with MDP with or without IL-1 receptor antagonist for 24 h (N=4). EBSS, Earle's Balanced Salt Solution; IL, interleukin; LPS, lipopolysaccharide; 3-MA, 3-methyl adenine; MDP, muramyl dipeptide; NOD2, nucleotide-binding, oligomerisation domain 2; PBMCs, peripheral blood mononuclear cells; TNF, tumour necrosis factor.

ATG16L1 Thr300Ala genotype and MDP-induced cytokine responses

The effect observed by blocking autophagy on MDP-induced IL-1β responses led to the hypothesis that the ATG16L1 Thr300Ala genotype may influence these responses in a similar manner. Thus, PBMCs obtained from healthy volunteers with different ATG16L1 Thr300Ala genotypes were stimulated with LPS, Pam3Cys or MDP for 24 h (figure 1C). Significantly higher production of IL-1β and IL-6, but not TNFα, was observed in the individuals homozygous for the 300Ala allele compared to individuals homozygous or heterozygous for the 300Thr allele, after stimulation of the cells with MDP. However, no differences were observed after stimulation with either Pam3Cys or LPS. These effects were consistent and reproducible in two independent cohorts of volunteers (figure 1C) and in a group of Crohn's disease patients (figure 1D). Of note, none of these individuals were bearing Crohn's disease-associated NOD2 variants (including R702W, G908R and 1007fsinsC). To study the role of IL-1β in the observed increased production of IL-6, PBMCs were stimulated with MDP with or without IL-1Ra for 24 h. IL-1Ra significantly decreased the MDP-induced production of IL-6 (figure 1E).

ATG16L1 Thr300Ala genotype and MDP stimulation in relation to 3-MA-induced IL-1β

In order to examine whether the effect of the ATG16L1 polymorphism on IL-1β responses is mediated through autophagy, PBMCs collected from individuals with different ATG16L1 genotypes were stimulated for 24 h with either MDP, 3-MA or the combination of these two stimuli. Indeed, when autophagy was completely blocked, the IL-1β response was similar between the genotypes. Notably, as also shown in figure 1A, 3-MA also induces secretion of IL-1β (figure 2A). To confirm that NOD2 mutations lead to a completely defective MDP induced IL-1β production, cells from patients with Crohn's disease and from healthy volunteers were stimulated with MDP for 24 h and IL-1β was measured in the supernatants. MDP-induced production of IL-1β was completely dependent on recognition of MDP by NOD2 (figure 2B).

Figure 2

(A) Production capacity of IL-1β by PBMCs after stimulation for 24 h with MDP, 3-MA or the combination, stratified for ATG16L1 Thr300Ala genotype (A allele equals 300Thr, G allele equals 300Ala). Data are mean±SEM, *p<0.05. (B) Production capacity of IL-1β by PBMCs from healthy volunteers and NOD2 deficient patients with Crohn's disease stimulated with MDP for 24 h (N=4). IL, interleukin; 3-MA, 3-methyl adenine; MDP, muramyl dipeptide; NOD2, nucleotide-binding, oligomerisation domain 2; PBMCs, peripheral blood mononuclear cells.

IL-1β mRNA expression, pro-IL-1β measurements and caspase-1 activation

To further dissect the mechanism underlying the effect of the ATG16L1 Thr300Ala genotype on NOD2 induced IL-1β responses, we assessed transcriptional and post-transcriptional mechanisms of IL-1β regulation. First, IL-1β mRNA expression was measured after stimulation of PBMCs with MDP for 24 h. An increased mRNA expression of IL-1β was observed in the cells isolated from individuals homozygous for the ATG16L1 300Ala allele, as compared to cells from individuals heterozygous or homozygous for the 300Thr allele (figure 3A). Similarly, an increased concentration of intracellular pro-IL-1β was observed in cells homozygous for the 300Ala allele (figure 3B). In contrast, no difference could be observed between the ATG16L1 genotypes regarding caspase-1 activation, either in the unstimulated condition or after stimulation with MDP. The active p35 subunit of caspase-1 could be detected in both the unstimulated and MDP stimulated condition, and was not influenced by the ATG16L1 genotype (figure 3C).

Figure 3

IL-1β mRNA expression, pro-IL-1β measurements and caspase-1 activation in relation to the ATG16L1 Thr300Ala genotype (A allele equals 300Thr, G allele equals 300Ala). (A) IL-1β mRNA expression in cells stimulated for 24 h with the indicated stimuli and stratified for ATG16L1 Thr300Ala genotype. Data are mean±SEM, *p<0.05. (B) Intracellular pro-IL-1β in cells stimulated for 24 h with MDP, stratified for ATG16L1 Thr300Ala genotype. Data are mean±SEM, *p<0.05. (C) Western blot of the active p35 subunit of caspase-1 and β-actin as loading control after incubation of PBMCs for 4 h, either unstimulated or stimulated with MDP, stratified for the ATG16L1 Thr300Ala genotype. IL, interleukin; MDP, muramyl dipeptide; PBMCs, peripheral blood mononuclear cells.

ATG16L1 mRNA and protein expression

Expression levels of ATG16L1 at the mRNA level were assessed between cells with different ATG16L1 genotypes that were left unstimulated or were stimulated with either Pam3Cys, LPS or MDP for 24 h. No differences were observed that would explain the effect of the ATG16L1 Thr300Ala polymorphism on NOD2 induced IL-1β (figure 4A). To study ATG16L1 protein expression, western blot was performed with cell lysates obtained from PBMCs, either homozygous for the 300Thr allele or homozygous for the 300Ala allele, which were left unstimulated or stimulated with MDP, Pam3Cys or LPS for 3 h (figure 4B). It can be clearly observed that there are large differences in the amount of ATG16L1 between the genotypes comparing the unstimulated and stimulated conditions. In fact, in cells homozygous for the 300Thr allele, ATG16L1 protein expression is enhanced after MDP, Pam3Cys or LPS stimulation, whereas in cells homozygous for the 300Ala allele the amount of ATG16L1 protein remains unchanged. In addition, the absolute amount of protein after stimulation is higher in the 300Thr genotype compared to the 300Ala genotype. These differences in ATG16L1 protein expression are quantified and shown in figure 4C.

Figure 4

ATG16L1 mRNA and protein expression. (A) ATG16L1 mRNA expression after incubation of PBMCs for 24 h, either unstimulated or stimulated with MDP, LPS or Pam3Cys, stratified for ATG16L1 Thr300Ala genotype (A allele equals 300Thr, G allele equals 300Ala). Data are mean±SEM. (B) Western blot of ATG16L1 protein and β-actin as loading control after incubation of PBMCs for 3 h, either unstimulated or stimulated with MDP, P3C (Pam3Cys) or LPS, stratified for the ATG16L1 Thr300Ala genotype. Picture is representative of three independent experiments with different donors. (C) Intensity of ATG16L1 bands on the western blot corrected for the intensity of the corresponding β-actin bands. Data are based on three experiments with different donors. Data are mean±SD, *p<0.05. LPS, lipopolysaccharide; MDP, muramyl dipeptide; PBMCs, peripheral blood mononuclear cells.

Discussion

Variation in the autophagy gene ATG16L1 has emerged as a major genetic susceptibility factor for Crohn's disease. The present study provides functional data that complement this genetic association, by demonstrating that immune cells that bear the risk variant of ATG16L1 display enhanced production of pro-inflammatory cytokines after activation of NOD2.

Autophagy, a basic machinery in eukaryotic cells responsible for bulk degradation of cellular constituents, is also indispensable in host defence against intracellular bacteria. In this respect, autophagosomes are formed that encapsulate bacteria upon their entry into innate immune cells. Next, lysosomes fuse with the assembled autophagosomes, after which its content is degraded and made available for antigen presentation. One of the crucial proteins involved in this process is ATG16L1, of which a genetic variant is linked to Crohn's disease.

Recent studies on the functional consequences of the risk variant of ATG16L1 have focussed on the capacity to form autophagosomes after microbial stimulation. Indeed, in these studies it could be demonstrated that the risk variant of ATG16L1 is associated with an impaired induction of autophagy specifically after NOD2 engagement,13 14 which could favour bacterial persistence and secondary inflammation. Interestingly, studies with cells derived from ATG16L1 knockout mice have demonstrated a pivotal role of this protein for the inhibition of IL-1β production after endotoxin treatment in a TIR-domain-containing adapter-inducing interferon-β (TRIF) and caspase-1-dependent manner.15 This initial finding in mice that autophagy can modulate cytokine responses upon microbial triggering of immune cells, led us to hypothesise that genetic variation in human ATG16L1 could modulate cytokine responses, especially IL-1β.

An initial argument for a role of autophagy in NOD2-induced IL-1β responses was that inhibition of autophagy by 3-MA led to a robust increase in secretion of IL-1β, while activation of autophagy by culturing the cells in starvation medium decreased the IL-1β response after stimulation of PBMCs with a pure NOD2 ligand, MDP. To examine whether the ATG16L1 Thr300Ala polymorphism exerts a similar effect on cytokine responses, PBMCs obtained from a large group of healthy volunteers were stimulated with ligands for TLR2, TLR4 and NOD2. Strongly increased IL-1β and IL-6 responses were observed when cells from individuals bearing the ATG16L1 genotypes that confer an increased susceptibility to Crohn's disease were stimulated with MDP, whereas TNFα responses were not different. Similarly, no differences were observed after stimulation of PBMCs with either LPS or Pam3Cys. As a confirmation, these findings were reproduced in a second set of experiments in PBMCs from another independent cohort of healthy subjects. Furthermore, similar experiments with PBMCs obtained from a group of patients with Crohn's disease also revealed the specific effect of the ATG16L1 T300A polymorphism on NOD2-induced IL-1β. Interestingly, similar experiments with cells stimulated with MDP alone or in combination with the autophagy inhibitor 3-MA, resulted in a loss of the difference between the ATG16L1 genotypes, indicating that the autophagy machinery is indeed involved in the effect on cytokine production. One may hypothesise whether these effects of ATG16L1 polymorphisms are dependent on mutations in NOD2 that are known to increased susceptibility to Crohn's disease. However, no role for NOD2 mutations in the observed differences in IL-1β production could be demonstrated, since these mutations lead to a complete deficiency of NOD2 signalling, as reported previously.16 17

To study the mechanism underlying the effect of the ATG16L1 genotype on MDP-induced IL-1β, several stages of transcriptional and translational regulation of IL-1β were investigated. In line with the increased secretion, higher IL-1β mRNA expression and intracellular pro-IL-1β were observed. However, no difference in caspase-1 activation was apparent between cells with different ATG16L1 genotypes, arguing for an effect on the transcription level of IL-1β. Interestingly, caspase-1 was demonstrated to be constitutively active when cells were left unstimulated. This constitutive active state of caspase-1, expressed in the monocyte fraction of PBMCs, is in line with a previous study that demonstrated that the regulatory step of IL-1β production in human monocytes is represented by transcription, and not inflammasome activation.18

The increased production of IL-6 concomitantly with IL-1β is most probably a consequence of increased IL-1β production, since it is demonstrated that signalling through the IL-1 receptor is a major inducer of IL-6,19 20 as we have confirmed in the present study. Interestingly, transcription of TNFα was not affected by the ATG16L1 genotype, which argues for different pathways of gene transcription induction for IL-1β and TNFα. Indeed, recent studies suggest that TNFα transcription is mainly induced by p38 MAPK and NF-κB, whereas IL-1β transcription also involves extracellular signal-related kinase/nuclear factor kappa B (Erk/NF-κB) pathways.21 22

Measurements of ATG16L1 mRNA and protein demonstrated that ATG16L1 expression is not influenced by the ATG16L1 genotype on the transcriptional level. However, an evident effect of the polymorphism on ATG16L1 protein expression could be observed. Stimulation with either MDP, Pam3Cys or LPS led to either an increase in or an unchanged amount of ATG16L1 protein compared to unstimulated cells, depending on the ATG16L1 genotype. Also, a higher total amount of ATG16L1 protein was observed after stimulation in cells bearing the 300Thr genotype compared to cells with the 300Ala genotype. Together with the recent finding that NOD2 specifically interacts with ATG16L1 in a direct fashion,13 these observations could provide an explanation for the specific effect of the polymorphism on NOD2 induced cytokine responses.

The exact mechanism underlying enhanced production of IL-1β upon MDP stimulation by cells bearing the ATG16L1 300Ala allele remains to be examined in further detail. Previously it has been shown that the polymorphism affects ATG16L1 protein expression.23 Hence, one possible explanation could be that the extent of NF-κB activation by NOD2 is influenced by this lower availability of the ATG16L1 300Ala variant. Importantly, our findings also point to an effect of the polymorphism on protein expression. Consequently, the balance between the two effects exerted by NOD2, that is, binding to ATG16L1 to induce autophagy on the one hand and activating NF-κB on the other hand, could be skewed in favour of Erk/NF-κB activation and IL-1β production in the case of autophagy defects in individuals bearing the ATG16L1 polymorphism. This would implicate this polymorphism as a modulator of the balance between NOD2 induced autophagy versus cytokine production.

Some important differences in the effect on regulation of IL-1β responses are apparent between mouse ATG16L1 knockout cells and the human genetic variants of ATG16L1. Studies on ATG16L1 knockout mouse cells revealed an inhibitory role of the protein on caspase-1 activation, whereas the human genetic variants of ATG16L1 exert an effect on IL-1β mRNA expression. These discrepancies emphasise that important differences exist between mice and humans as well as between complete knockout and non-synonymous polymorphism, leading to amino acid substitution.

Innate immune responses mediated by NOD2, including cytokine production and induction of autophagy, have appeared to play a central role in pathogenesis of Crohn's disease. At first, this was demonstrated by genetic association studies that revealed NOD2 mutations as a risk factor for developing Crohn's disease.24 25 Second, the role of innate host responses for the pathogenesis of Crohn's disease is further emphasised by the association of genetic variants of ATG16L1 and IRGM with Crohn's disease, and the influence of these variants specifically on NOD2-induced immune responses. This has already been shown concerning the induction of autophagy by NOD2 signalling, which is affected by the ATG16L1 genotype.13 14 In addition to these findings, the present study demonstrates the association of ATG16L1 genotype with increased NOD2 induced IL-1β responses, providing a direct link of the Crohn's disease risk-associated ATG16L1 300Ala allele with excessive inflammation that is pathognomonic of Crohn's disease.

References

View Abstract

Footnotes

  • Funding This study was performed within the framework of the Dutch Top Institute Pharma # D1-101. MGN was supported by a Vici grant of the Netherlands Organization for Scientific Research (NWO).

  • Competing interests None.

  • Ethics approval This study was conducted with the approval of the Ethical Committee of the Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.