NOD2 and ATG16L1 polymorphisms affect monocyte responses in Crohn’s disease

Abstract

AIM: To assess whether polymorphisms in NOD2 and ATG16L1 affect cytokine responses and mycobacterium avium subspecies paratuberculosis (MAP) survival in monocytes from Crohn’s disease (CD) patients.

METHODS: Monocytes were isolated from peripheral blood of CD patients of known genotype for common single nucleotide polymorphisms of NOD2 and ATG16L1. Monocytes were challenged with MAP and bacterial persistence assessed at subsequent time-points. Cytokine responses were assayed using a Milliplex multi-analyte profiling assay for 13 cytokines.

RESULTS: Monocytes heterozygous for a NOD2 polymorphism (R702W, P268S, or 1007fs) were more permissive for growth of MAP (P = 0.045) than those without. There was no effect of NOD2 genotype on subsequent cytokine expression. The T300A polymorphism of ATG16L1 did not affect growth of MAP in our model (P = 0.175), but did increase expression of cytokines interleukin (IL)-10 (P = 0.047) and IL-6 (P = 0.019).

CONCLUSION: CD-associated polymorphisms affected the elimination of MAP from ex vivo monocytes (NOD2), or expression of certain cytokines (ATG16L1), implying independent but contributory roles in the pathogenesis of CD.

Key Words: Inflammatory bowel disease; Mycobacterium avium subspecies paratuberculosis; Cytokine; CARD15; Autophagy

INTRODUCTION

Crohn’s disease (CD) has been proposed as being the product of chronic inflammation caused by a dysfunctional interaction between the intestinal immune system and commensal gut microbiota[1]. The inflammation seen in CD is characterized by pronounced Th1 and Th17/23 responses involving the cytokines interleukin (IL)-1, IL-6, tumor necrosis factor (TNF)-α, IFN-γ[1-3], IL-23 and IL-17[4-6].

Although the commensal intestinal microflora appear to play an important role in the etiology of CD, certain bacterial species have also been implicated as putative causal agents of CD. These include Pseudomonas maltophilia[7], Chlamydia trachomatis[8], Bacteroides fragilis[9] Yersinia species[10], adherent invasive Escherichia coli (E. coli)[11] and Mycobacterium avium subspecies paratuberculosis (MAP)[12]. The role of MAP in the etiology of CD is unclear. However, MAP causes chronic intestinal inflammation in ruminants (Johne’s disease) with similar pathophysiology to CD[13] and it has been reported as a causative agent of regional enteritis with similarities to CD in a number of monogastrics, including two primate species[14,15]. MAP has also been cultured from both resected tissue[16] and peripheral blood of CD patients[17], and has been visualized inside macrophages from CD patients[18]. Meta-analyses of epidemiological studies confirm an association of MAP with CD[19,20], although it remains unknown whether MAP is pathogenic in humans, or whether this association reflects a defective host immune system permissive for the survival of MAP.

Twin and family studies have demonstrated a significant genetic component to the development and progression of CD[21-23]. Linkage analysis, candidate gene approaches and, most recently, genome-wide association studies (GWAS) have identified over 30 risk genes for CD[24-26], many of which are involved with bacterial recognition (e.g. NOD2) or processing and elimination of bacteria through the autophagy pathway (e.g. IRGM[27] and ATG16L1[25]).

In this study, we developed an ex vivo monocyte model to assess the impact of the CD-associated single nucleotide polymorphisms (SNPs) in NOD2 (rs2066842, P268S; rs2066844, R702W; and 1007fs, rs2066847) and ATG16L1 (rs2241880, T300A) on cytokine responses to the putative pathogen MAP. MAP can survive and replicate within phagocytic cells, and consequently we also evaluated the impact of these polymorphisms on the intracellular persistence of MAP. The use of ex vivo monocytes allows functional evaluation of SNPs associated with CD, and may provide a more realistic insight into the impact that genotype has on CD compared to studies which involve abrogation of protein expression or whole gene deletion.

MATERIALS AND METHODS

Patient recruitment

Patients for the current study were selected from a New Zealand population-based Caucasian inflammatory bowel disease (IBD) cohort recruited to investigate genetic and environmental factors that contribute to IBD etiology[28-32]. Inclusion criteria for the current study were a confirmed diagnosis of CD and negative MAP status as ascertained by IS900 PCR in peripheral blood[28].

The genotype combinations and patient phenotype information are detailed in Table 1. Briefly, the potential impact of ATG16L1 and NOD2 polymorphisms were assessed separately. For analysis of ATG16L1, monocytes were collected from CD patients who had a wild-type NOD2 genotype and were homozygous for either the major (G) allele (n = 6) or minor (A) allele (n = 6) of ATG16L1 1138G > A (rs2241880). Conversely, for experiments evaluating the effect of NOD2 genotype, monocytes were collected from patients who were ATG16L1 1138G homozygotes and were heterozygous for one of the three NOD2 SNPs previously associated with CD; 2104C > T (R702W, rs2066844), 2722G > C (G908R, rs2066845), or 3020insC (1007fs, rs2066847), and were heterozygous for the background variant 802C > T (P268S, rs2066842). A total of 12 patients were recruited, 6 carried the polymorphisms described, and 6 were homozygous at these NOD2 loci (Table 1). None of the patients included in this study had the SNPs rs13361189 and rs4958847. These SNPs are in complete linkage disequilibrium with a 20 kb insertion/deletion polymorphism which has been shown to alter expression of the autophagy gene IRGM[33].

Table 1 IRGM, ATG16L1, and NOD2 genotypes of Crohn’s disease patients from whom monocytes were collected for ex vivo experiments.

	IRGM		ATG16L1	NOD2				Clinical characteristics
Patient	rs133611891	rs49588471	T300A 1138G > Ars2241880	R702W 2104C > Trs2066844	P268S 802C > Trs2066842	G908R 2722G > Crs2066845	1007fs 3020insCrs2066847	Gender	Time since diagnosis (yr)	Disease location	Harvey bradshaw index2
1	TT	GG	AA	CC	CC	GG	00	M	53	Ileal	0
2	TT	GG	AA	CC	CC	GG	00	M	5	Ileo-colonic	0
3	TT	GG	AA	CC	CC	GG	00	F	5	Colonic	0
4	TT	GG	GG	CC	CC	GG	00	M	25	Ileo-colonic	2
5	TT	GG	AA	CC	CC	GG	00	F	6	Ileo-colonic	1
6	TT	GG	AA	CC	CC	GG	00	F	9	Ileal	3
7	TT	GG	AA	CC	CC	GG	00	M	1	Ileal	0
8	TT	GG	GG	CC	CC	GG	00	M	6	Colonic	2
9	TT	GG	GG	CC	CC	GG	00	F	10	Colonic	5
10	TT	GG	GG	CC	CC	GG	00	M	24	Ileo-colonic	1
11	TT	GG	GG	CC	CC	GG	00	F	11	Colonic	0
12	TT	GG	GG	CC	CC	GG	00	F	7	Ileo-colonic	4
13	TT	GG	GG	CT	CT	GG	00	F	5	Colonic	4
14	TT	GG	GG	CT	CT	GG	00	F	9	Ileo-colonic	0
15	TT	GG	GG	CT	CT	GG	00	M	17	Ileal	0
16	TT	GG	GG	CT	CT	GG	00	F	18	Colonic	4
17	TT	GG	GG	CC	CT	GG	0C	M	7	Ileal	1
18	TT	GG	GG	CT	CT	GG	00	F	16	Ileo-colonic	4

¹SNPs located upstream (5’) to IRGM;

²Non-invasive clinical index used to assess disease activity in patients with Crohn's disease. A score of ≥ 7 indicates active disease.

Preparation of monocytes from peripheral blood

Blood (40 mL) was drawn into heparin tubes (Sigma-Aldrich), divided into 20 mL aliquots, and 15 mL of Phosphate Buffered Saline (PBS) was added to each. Ficoll-Paque™ PREMIUM (10 mL) (GE Healthcare Bio-Sciences Uppsala, Sweden) was layered under each aliquot and the samples centrifuged (1000 ×g, 20 min). Mononuclear cells collected from the interface were added to 30 mL of PBS, centrifuged (350 ×g, 10 min) and resuspended in 15 mL of PBS. In order to standardize the number of monocytes used in experiments, mononuclear cells were enumerated using a hemocytometer and a 200 μL aliquot was analyzed on a Beckman Coulter FC500 MPL flow cytometer to determine the percentage of monocytes based on forward and side scatter characteristics. After centrifugation (350 ×g, 10 min), cells were resuspended in RPMI medium 1640 supplemented with 10% fetal bovine serum (FBS) (Invitrogen, Carlsbad, CA, USA) to a concentration of 4 × 10⁵ monocytes/mL. Then 500 μL of this preparation were plated per well in a 24-well tissue culture plate (Nunc, Roskilde, Denmark). Monocytes were left to adhere for 1 h in a humidified incubator at 37°C with 5% CO₂ and non-adherent cells were removed by washing three times with 1 mL of room temperature (RT) PBS. Adherent monocytes were incubated overnight in 500 μL of RPMI 1640/10% FBS.

Culture of MAP

MAP strain Dominic (ATCC 43545) was inoculated into 10 mL volumes of Difco™ Middlebrook 7H9 broth (BD Biosciences, Sparks, MD, USA) supplemented with 10% v/v BBL™ Middlebrook OADC Enrichment (BD Biosciences), 0.05% v/v Tween 80 (Sigma-Aldrich, St Louis, MO, USA) and 2 mg/mL Mycobactin J (Allied Monitor, Fayette, MO, USA) and grown at 37°C. A standard growth curve of MAP was obtained by measuring the optical density at 600 nm (OD₆₀₀) of an aliquot of bacterial suspension during the log phase, which had been passed repeatedly through a 25-gauge needle, and enumerating the bacteria by plating on the same medium supplemented with 1.5% agar (Invitrogen, Carlsbad, CA, USA).

Monocyte challenge experiments

Bacteria were grown to mid log phase (approximately 1 × 10⁸ cells/mL, with reference to previously obtained growth curve data) and harvested by centrifugation (13 000 ×g, 5 min). Bacteria were resuspended in 500 μL of PBS and passed ten times through a 25-gauge needle to break up clumps of cells. One 500 μL aliquot of MAP was heat-inactivated at 90°C for 5 min and then both aliquots were diluted to 4 × 10⁶ cells/mL in RPMI 1640/10% FBS. The optimal temperature and incubation time for heat-inactivation had been previously confirmed by plating heat-treated MAP cells onto agar. Growth medium was removed from the monocytes and 500 μL of the MAP suspension was added to each well. Challenge experiments were performed in triplicate for each time-point/genotype combination. After incubation for 4 h at 37°C, 5% CO_2, the supernatant was removed and stored at -80°C as a zero time-point sample for cytokine assays. Monolayers were washed three times with 1 mL of PBS to remove extracellular bacteria. A zero time-point lysate to assess bacterial uptake and persistence was removed by incubating 500 μL of a 0.1% sodium deoxycholate (Sigma-Aldrich, St Louis, MO, USA) solution with the monolayer for 5 min at RT. Lysates were centrifuged (13 000 ×g, 5 min) and resuspended in 200 μL of PBS. Serial dilutions of the lysates were made with PBS and 50 μL were spread on to a Middlebrook agar plate. The agar plates were left to incubate for four weeks at 37°C before counting MAP colonies. Subsequent samples at time-points of 24, 48, 72 and 96 h were similarly processed.

Confirmation of genotypes of study participants

ATG16L1 genotypes were confirmed at recruitment by direct DNA sequencing of PCR products. Briefly, genomic DNA was extracted from fresh peripheral blood using GenElute™ (Sigma-Aldrich, St Louis, MO, USA) spin columns according to manufacturer’s protocols. A 480 bp fragment containing the ATG16L1 1138G > A SNP was amplified for DNA sequencing from patient genomic DNA using the following primers: 5'-CCACAGGTTAGTGTGCAGGA-3' (forward primer) and 5'-CACAGCTGACAGAGCCAAAA-3' (reverse primer). PCR was carried out in a 20 μL volume containing 1 μL of genomic DNA, 0.3 μmol/L of each primer, 200 μmol/L dNTPs, 0.75 mmol/L MgCl₂, 1 × TAQ-Ti reaction buffer (Thermo Fisher Scientific, Pittsburgh, PA, USA) and 0.25 U of TAQ-Ti DNA polymerase (Thermo Fisher Scientific). After an initial denaturation step of 94°C for 2 min, 35 cycles were performed at 94°C for 30 s, 60°C for 30 s and 72°C for 30 s. Five microlitres of each PCR product was checked on 1% agarose. Another 5 μL aliquot was purified with Exo-SAP-IT (USB Corporation, Ohio, USA) and sequenced using BigDye chemistry (Applied Biosystems, California, USA) on an ABI 3730 Genetic Analyzer (Foster City, California, USA). The NOD2 and IRGM genotypes of study participants were established as previously described using allele-specific PCR and pre-designed TaqMan SNP genotyping assays, respectively[34,35].

Multiplex cytokine analysis

Cytokine analysis was performed using a 13-plex MILLIPLEX™ MAP human cytokine kit according to manufacturer’s recommendations (Millipore) for the following: IFNγ, IL-10, IL-12p40, IL-12p70, IL-17, IL-1b, IL-2, IL-4, IL-5, IL-6, IL-8, TNFα, TNFβ.

Statistical analysis

Data from the five time-points were analyzed by repeated measures ANOVA, with fixed effects for time (as categorical variable) and random effects for subject i.e. genotype. For cytokine analyses, where readings were below the threshold for detection (3.2 pg/mL) they were included as 3.2 pg/mL. Results were considered significant at P≤ 0.05.

Ethical considerations

Informed written consent was obtained from all participants in this study and ethical approval for this work was granted by the Upper South B Regional Ethics Committee of New Zealand.

RESULTS

Bacterial persistence ATG16L1

Numbers of MAP increased from a mean log₁₀ colony forming units (cfu) of 4.24 at 0 h to 5.08 at 96 h for the AA genotype, and from log10 cfu of 4.47 to 5.29 for the GG genotype (Figure 1). There was no evidence (T-testing) for the effect of the ATG16L1 T300A polymorphism on MAP numbers at any individual time-point, or overall (pooled time-points ANOVA, P = 0.175) (Figure 1).

Figure 1 Effect of ATG16L1 T300A variant on mycobacterium avium subspecies paratuberculosis numbers. Time course of mycobacterium avium subspecies paratuberculosis (MAP) growth in the context of the ATG16L1 1138 G > A variant. Monocytes were derived from subjects homozygous for either ATG16L1 allele and carrying NOD2 wild-type alleles (n = 6 for each group). MAP growth is expressed as colony forming units (cfu).

NOD2

MAP also grew in the monocytes with different NOD2 genotypes, from a mean log₁₀cfu of 4.30 at 0 h to 5.12 for the cells homozygous for the major alleles of the NOD2 variants, and from log₁₀cfu of 4.58 to 5.42 for cells heterozygous for any NOD2 variant (Figure 2). There was no significant effect (T-testing) of NOD2 genotype on MAP numbers at any individual time-point. However, analysis of all time-points indicated that monocytes heterozygous for a NOD2 polymorphism were more permissive for growth of MAP (ANOVA, P = 0.045) (Figure 2).

Figure 2 Effect of NOD2 variants on mycobacterium avium subspecies paratuberculosis numbers. Time course of mycobacterium avium subspecies paratuberculosis (MAP) growth in the context of NOD2 genetic variation. Monocytes were derived from subjects homozygous for the ATG16L1 1138G allele and with/without NOD2 genetic variants (n = 6 for each group). MAP growth is expressed as colony forming units (cfu).

Multiplex cytokine panels

Thirteen cytokines were evaluated from ex vivo monocyte supernatants using a MILLIPLEX™ MAP human cytokine kit. Each time-point/genotype combination was assayed in triplicate. All assays passed quality controls and r² values for the standard curves were ≥ 0.99.

Of the thirteen cytokines analyzed in the multiplex format, only four, IL-10, IL-6, IL-8 and TNFα, had measurable responses above the detection threshold (> 3.2 pg/mL) on the multiplex ELISA platform used. Where samples had values of < 3.2 pg/mL, they were considered to be 3.2 pg/mL for statistical purposes.

Effect of ATG16L1 genotype

Cytokine expression results are shown in Figure 3. The AA genotype of ATG16L1 was associated with greater expression of cytokines IL-10 and IL-6 in response to challenge with MAP (P = 0.047 and P = 0.019, respectively). No significant difference was seen between ATG16L1 genotypes AA and GG for expression of either IL-8 or TNFα (P = 0.758 and P = 0.289, respectively) (Figure 3).

Figure 3 Effect of the ATG16L1 T300A polymorphism on expression of cytokines following challenge with mycobacterium avium subspecies paratuberculosis. Cytokine concentrations are in pg/mL. Results are the means (and standard error of the mean) of triplicate samples from patients in Table 1. P-values are derived from repeated measures ANOVA testing for the effect of genotype on cytokine expression. Fixed effects for time (categorical variable) and ATG16L1 genotype, random effect for subject. IL: Interleukin; TNF: Tumor necrosis factor.

Effect of NOD2 genotype

Cytokine expression changes for IL-10, IL-6, IL-8 and TNFα by cells of different NOD2 genotypes are shown in Figure 4. In general, the presence of a NOD2 polymorphism resulted in a trend of lower expression of these four cytokines throughout the time-course, compared to monocytes without NOD2 variants.

Figure 4 Effect of NOD2 polymorphisms on expression of cytokines following challenge with mycobacterium avium subspecies paratuberculosis. Cytokine concentrations are in pg/mL. Results are the means (and standard error of the mean) of triplicate samples from patients in Table 1. P-values are derived from repeated measures ANOVA testing for the effect of genotype on cytokine expression. Fixed effects for time (categorical variable) and NOD2 genotype, random effect for subject. IL: Interleukin; TNF: Tumor necrosis factor.

However, this effect was not significant, with P-values of 0.56, 0.32, 0.41 and 0.97 for cytokines IL-10, IL-6, IL-8 and TNFα, respectively (Figure 4).

DISCUSSION

Numerous genetic association studies have indicated a strong role for NOD2 and ATG16L1 in the etiology of CD. However, functional studies have yet to fully elucidate whether, and to what extent, polymorphic variation in these genes affects bacterial clearance and inflammation in CD. Models of gene/environment interaction have mostly used epithelial or monocyte cell lines in vitro with bacterial pathogens such as Salmonella typhimurium that are not generally associated with CD. Also, the use of gene silencing or ablation techniques in these models may assist our understanding of the function of these genes, but may not be representative of the effects of SNPs which have been associated with CD. In this study we developed an ex vivo cellular model using monocytes from CD patients and a putative CD pathogen, MAP, in order to assess the effect of SNPs in NOD2 and ATG16L1 on bacterial survival and subsequent inflammatory response.

NOD2 is located in the cytoplasm, and plays an important role in cellular responses to bacterial infection[36] through recognition of muramyl dipeptide (MDP), a subunit of bacterial peptidoglycan (PGN)[37]. Furthermore, the R702W, G908R and 1007fs NOD2 variants, which confer susceptibility to CD, have been shown to impair responses to the bacterial antigen lipopolysaccharide[38].

We found the effect of NOD2 polymorphisms appeared to be primarily on bacterial persistence/growth, with heterozygosity at R702W or 1007fs making monocytes from CD patients significantly more permissive to growth of MAP. Indeed, a recent study has observed that these variants are associated with an impairment of monocyte phagocytosis and the development of bacteremia in intensive care unit patients[39]. Despite the increased bacterial load, no differences were seen in subsequent cytokine responses for the different host NOD2 genotypes. Whilst there has been no prior published research with respect to MAP in this type of model of CD, the functional effects of NOD2 SNPs have been examined for adherent invasive E. coli (AIEC). No differences were found in the persistence or growth of AIEC in monocytes from patients who were heterozygous or homozygous for the minor allele of NOD2 variants R702W, G908R and 1007fs compared to NOD2 major allele homozygotes[40]. It has also been shown that in peripheral blood mononuclear cells from CD patients, homozygosity for R702W does not affect the cytokine response to Gram-negative (Helicobacter pylori) bacterial peptidoglycan[41], although homozygosity for the 1007 fs mutation did. In our study we recruited no 1007fs homozygotes and only a single 1007 fs heterozygote so were unable to make any direct comparisons with this previous study. Comparative studies for a broader range of bacterial species in ex vivo cells from patients who are homozygous for these NOD2 SNPs will help clarify the role of this genotype in disease pathogenesis.

Both MAP and AIEC are capable of intracellular survival and growth in monocytes and macrophages, and it appears that the presence of NOD2 polymorphisms may influence their respective intracellular survival and growth in different ways[42].

Autophagy has been identified as a mechanism for clearing intracellular pathogens, and two autophagy genes, ATG16L1 and IRGM, have been associated with CD[25,27]. In our study, we investigated the effect of the ATG16L1 T300A polymorphism, and controlled for genotypic variation in IRGM (Table 1). ATG16L1 T300A (rs2241880; 1138G > A) is a common non-synonymous SNP where the G major allele confers greater disease risk and results in a threonine-to-alanine substitution at amino acid position 300 of the ATG16L1 protein (T300A). This SNP appears to account for all of the disease risk conferred by this locus[43], and functionally, this polymorphism has been proposed to contribute to defective macrophage killing of bacteria[44]. This assertion is supported by the results of two in vitro studies. Kuballa et al[45] found that the T300A variant impaired handling and autophagy of Salmonella within human epithelial cells, and Lapaquette et al[46] showed that siRNA knockdown of ATG16L1 led to loss of autophagy of intracellular AIEC bacteria by HeLa cells. Transfection of affected HeLa cells with wild-type ATG16L1 restored autophagic function, whereas transfection with the T300A polymorphic form did not. In contrast, ATG16L1 T300A had no effect on the survival of either S. typhimurium or group A Streptococcus in mouse embryonic fibroblasts[47]. Although direct comparison of different model systems is complex and potentially misleading, our results with MAP in ex vivo monocytes generate the question as to whether the T300A polymorphism of ATG16L1 affects autophagic clearance of certain intracellular bacteria as profoundly as indicated by knockdown or silencing models of gene function.

In our study, the T300A polymorphism was associated with significant changes in production of the cytokines IL-6 and IL-10 in response to bacterial challenge with MAP. These two cytokines are components of the Th1 (pro-inflammatory) and Th2 (modulatory) pathways of inflammation, respectively, and it is likely that CD results from an imbalance between these two pathways. It is tempting to speculate that the relative levels of the two cytokines that were induced are indicative of an imbalance, but extrapolating from a very specific model to describe a complex disease state would be misleading and clearly further comparative work is required in this area in disease-relevant models.

Our study is the first to investigate the effect of NOD2 and ATG16L1 genotype on the response of ex vivo human monocytes to the putative CD pathogen MAP. Although our results are preliminary and need to be replicated in a larger sample, they provide novel insights into the effect of disease-associated SNPs in innate immunity genes on detection, handling, and elimination of bacteria, and ultimately CD pathogenesis. Our observations indicate that NOD2 SNPs R702W, P268S, or 1007 fs impair the elimination of MAP yet do not impact on cytokine production. They may, therefore, increase susceptibility to prolonged intracellular bacterial infection. Conversely, the ATG16L1 T300A polymorphism significantly alters the expression of certain Th1 and Th2 cytokines after MAP challenge, but does not seem to affect the autophagic clearance of this putative CD pathogen.

COMMENTS

Background

Polymorphisms of the genes NOD2 and ATG16L1 have been associated with susceptibility to Crohn’s disease (CD). These genes are important for an effective innate immune response against potential bacterial pathogens [such as Mycobacterium avium subspecies paratuberculosis (MAP)] which may trigger or exacerbate inflammation. Monocytes from CD patients of known genotype were used to determine whether polymorphisms in NOD2 and ATG16L1 alter cytokine responses and bacterial survival following challenge with MAP.

Research frontiers

Previous research has investigated the role of polymorphisms in NOD2 and ATG16L1 in various model systems. In general, these have used gene silencing strategies that may not realistically reflect the biological consequences of single nucleotide changes in these genes, or model bacterial pathogens that have little relevance to CD. None have reported the functional consequences of the naturally occurring single nucleotide polymorphisms using patient-derived cells and bacteria such as MAP that have been implicated in CD etiology.

Innovations and breakthroughs

The authors findings indicate that monocytes heterozygous for a NOD2 polymorphism were more permissive for the intracellular growth of MAP than those without. However, these polymorphisms did not affect subsequent cytokine expression. The T300A polymorphism of ATG16L1 did not affect growth of MAP in our monocyte model but did result in increased expression of certain cytokines - interleukin (IL)-10 and IL-6.

Applications

By understanding how naturally occurring disease-related polymorphisms of NOD2 and ATG16L1 influence bacterial survival and also the production of inflammatory mediators, the authors may gain insight into the contribution of these genetic changes to the function of the host innate immune system. Development of this model system that utilizes patient cells with known single nucleotide changes in key CD-susceptibility genes will provide another research tool to assist better understanding of disease pathogenesis related to bacterial handling.

Terminology

NOD2 (CARD15) - nucleotide oligomerization domain 2 - is a cytosolic pattern recognition receptor that recognizes muramyl dipeptide, a component of bacterial peptidoglycan. Polymorphisms in NOD2 have been associated with altered susceptibility to CD in many genetic studies. ATG16L1 - autophagy-related 16-like 1 - is a key component of the autophagic apparatus that is involved with uptake and digestion of intracellular bacteria. The T300A polymorphism of ATG16L1 has also been associated with altered susceptibility to CD. MAP is an intracellular bacterium that has been cited in several studies as a putative causal agent of CD.

Peer review

This is a very well-designed and well-written study, with interesting and important scientific merit. Not just simple polymorphism descriptions, but their effect on human monocyte cytokine production and intracellular pathogen survival were examined with a very functional methodology. Their ex vivo model is much closer to the real pathogenesis of CD than any earlier one. Using the author’s concept, more descriptive polymorphism analysis of CD and other diseases may be placed into functional analysis.