NOD2 triggers an interleukin-32-dependent human dendritic cell program in leprosy

Abstract

It is unclear whether the ability of the innate immune system to recognize distinct ligands from a single microbial pathogen via multiple pattern recognition receptors (PRRs) triggers common pathways or differentially triggers specific host responses. In the human mycobacterial infection leprosy, we found that activation of monocytes via nucleotide-binding oligomerization domain-containing protein 2 (NOD2) by its ligand muramyl dipeptide, as compared to activation via heterodimeric Toll-like receptor 2 and Toll-like receptor 1 (TLR2/1) by triacylated lipopeptide, preferentially induced differentiation into dendritic cells (DCs), which was dependent on a previously unknown interleukin-32 (IL-32)-dependent mechanism. Notably, IL-32 was sufficient to induce monocytes to rapidly differentiate into DCs, which were more efficient than granulocyte-macrophage colony-stimulating factor (GM-CSF)-derived DCs in presenting antigen to major histocompatibility complex (MHC) class I-restricted CD8(+) T cells. Expression of NOD2 and IL-32 and the frequency of CD1b(+) DCs at the site of leprosy infection correlated with the clinical presentation; they were greater in patients with limited as compared to progressive disease. The addition of recombinant IL-32 restored NOD2-induced DC differentiation in patients with the progressive form of leprosy. In conclusion, the NOD2 ligand-induced, IL-32-dependent DC differentiation pathway contributes a key and specific mechanism for host defense against microbial infection in humans.

Figure 1

NOD2L and TLR2/1L induce functionally divergent DC-specific pathways. Human monocytes, activated with either NOD2L (1 µg ml⁻¹) or TLR2/1L (1 µg ml⁻¹), were analyzed for their gene expression profiles using Affymetrix microarrays. (a) Ingenuity pathway analysis to associate functional pathways with the corresponding gene sets by enrichment ratios and P values; P values were corrected for multiple hypothesis testing using the Benjamini-Hochberg method. RAN, ras-related nuclear protein; TC, T cell; GR, glucocorticoid receptor; PPAR, peroxisome proliferator–associated receptor; NF-κB, nuclear receptor-κB. (b) Enrichment analysis of the induced gene sets with the four DC-specific pathways, shown as the percentage of genes that are induced within each pathway and the corresponding P value. Association of NOD2L-induced gene sets with corresponding pathway is highlighted in red. (c) Differential gene expression profile of the microbial ligand activated monocytes, illustrated as the fold change ratio (FC) of NOD2L / medium on the x axis and TLR2/1L / medium on the y axis.

Figure 2

NOD2L is a potent inducer of functional CD1b⁺ DCs. Purified human monocytes activated with either NOD2L (1 µg ml⁻¹) or TLR2/1L (1 µg ml⁻¹) for 48 h analyzed by flow cytometry for the expression of CD1b (DCs) and CD209 (macrophages). (a) Representative flow cytometric analyses with double labeling for CD1b and CD209, shown for each condition. Indicated in each quadrant is the percentage of positive cells. (b) CD1b and CD209 expression, shown as mean percentage positive ± s.e.m., n = 8. (c) Surface expression of markers involved in antigen presentation by flow cytometry, gated on CD1b⁺ DCs induced by either NOD2L (1 µg ml⁻¹) or TLR2/1L (1 µg ml⁻¹); data are indicated as mean MFI ± s.e.m., n = 6. (d–f) T cell response using NOD2L- or TLR2/1L-induced purified CD1b⁺ DCs in the context of tetanus toxoid and autologous CD8⁺ T cells (d) as well as various concentrations of M. leprae GroES protein (e) or GroES peptide and an MHC class II–restricted CD4⁺ T cell clone BCD4.9 (f). Proliferation shown as ³H-thymidine incorporation and IFN-γ secretion. Data are representative of triplicate wells of three independent experiments ± s.e.m. Statistical significance was calculated by two-tailed Student’s t test. AU, arbitrary units.

Figure 3

NOD2L induces an IL-32–dependent DC program. (a,b) Induction of GM-CSF mRNA and protein (a) and GM-CSF receptor (GM-CSFRA, GM-CSFRB) (b) by NOD2L (1 µg ml⁻¹) and TLR2/1L (1 µg ml⁻¹) in human monocytes, mean ± s.e.m., n = 5. (c) Gene expression profile data analysis for genes coding for secreted molecules that were enhanced by NOD2L but not TLR2/1L (NOD2L > TLR2/1L). SPP1, secreted phosphoprotein 1; IL-1RN, IL-1 receptor antagonist; TNFSF8, tumor necrosis factor (ligand) superfamily, member 8; CKLF, chemokine-like factor 1; MIF, macrophage migration inhibitory factor; AIMP1, aminoacyl tRNA synthetase complex-interacting multifunctional protein 1; C19ORF10, chromosome 19 open reading frame 10. (d,e) Induction of IL-18 and IL-18R mRNA and IL-32 mRNA and protein by NOD2L (1 µg ml⁻¹) versus TLR2/1L (1 µg ml⁻¹), data represent mean ± s.e.m., n = 5. (f,g) Induction of IL-32 mRNA by TLR and NLR ligands (f) and by live M. leprae (mLEP) at various multiplicities of infection (MOI; g). (h) Effect of siIL32 knockdown on NOD2L induction of IL-32 mRNA and induction of CD1b⁺ DCs. AU, arbitrary units. Data are represented as mean ± s.e.m., n = 4. Statistical significance was calculated by two-tailed Student’s t test.

Figure 4

IL-32–induced DCs are potent antigen-presenting cells for MHC class I–restricted antigens. (a) Ability of recombinant IL-32 to induce CD1b⁺ DCs. (b) Surface expression of markers involved in antigen presentation, gated on CD1b⁺ DCs induced by either recombinant IL-32 (50 ng ml⁻¹) or recombinant GM-CSF (1 U ml⁻¹); data are indicated as mean ± s.e.m., n = 6. (c–f) Ability of IL-32– (50 ng ml⁻¹) or GM-CSF– (1 U ml⁻¹) derived CD1b⁺ DCs to stimulate a T cell response in the context of tetanus toxoid (c), influenza peptide M1 with autologous CD8⁺ T cells (d), M. leprae GroES protein (e) or GroES peptide with the MHC class II–restricted T cell clone BCD4.9. (f) Proliferation shown as ³H-thymidine incorporation and IFN-γ secretion. Data are shown as mean of triplicate wells of at least two independent experiments, ± s.e.m. Statistical significance was calculated by two-tailed Student’s t test.

Figure 5

IL-32 activates a DC program in leprosy. (a) Comparison of gene expression profiles of the microbial ligand–activated monocytes with their differential expression in leprosy skin lesions using an integrative bioinformatics approach. The relative gene expression of NOD2L- versus TLR2/1L-activated monocytes ((log2) ratio) is shown on the x axis, and the relative expression in T-lep versus L-lep lesions ((log2) ratio) is shown on the y axis (hypergeometric distribution P = 0.004, genes ranked on minimal fold change in the two data sets). (b) Expression of CD1B, IL-32, CD205 and CD11b mRNAs in leprosy lesions according to gene expression profile data. (c,d) Immunolabeling of IL-32 (c) and NOD2 (d) in T-lep and L-lep lesions; two representative labeled sections are shown out of at least six, scale bars, 30 µm. (e) Quantification of IL-32– and NOD2–positive cells in T-lep and L-lep lesions, data are indicated as mean ± s.e.m., n = 4. (f) Coexpression by confocal laser microscopy of IL-32 with CD68 (macrophages), CD1b (DCs) and NOD2 in T-lep lesions. In b and e, statistical significance was calculated by two-tailed Student’s t test.

Figure 6

Monocytes from patients with L-lep show reduced induction of CD1b⁺ DCs in response to NOD2L compared to healthy controls. (a) Response of monocytes from patients with L-lep and healthy donors to NOD2L (1 µg ml⁻¹);. (b) Spontaneous IL-10 release in monocytes from patients with L-lep (left), effect of addition of recombinant IL-10 (rIL-10) to normal monocytes on NOD2L-induced CD1b induction (middle) and effect of blocking IL-10 on NOD2L-induced CD1b induction in monocytes from patients with L-lep (right), n = 6. (c) IL-32, IL-18 and GM-CSF mRNA induction by NOD2L (1 µg ml⁻¹) in monocytes from patients with L-lep. (d) Effect of IL-32 (50 ng ml⁻¹) on CD1b induction by NOD2L (1 µg ml⁻¹) in patients with L-lep. Data are shown as mean ± s.e.m., n = 6. Statistical significance was calculated by two-tailed Students t test.