Unit: Innate immunity and cell signaling
Director: Philpott, Dana
Our group studies innate immunity mediated by a family of cytosolic proteins, call NBS/LRR (nucleotide-binding site/leucine-rich repeat) proteins, which are structurally similar to plant disease resistance proteins or R proteins. The two prototypes of this family that have been best characterised are Nod1 and Nod2. Nod1 and Nod2 recognise distinct motifs of peptidoglycan. We have shown that Nod1 plays an important role in the sensing of Gram-negative bacterial infections of epithelial cells. Activation of defense responses in cells infected with Shigella flexneri or Helicobacter pylori appear to be mediated by this intracellular pattern recognition protein.
Nods and PGRPs in innate immune defense
Nod1 and Nod2 are recently described proteins involved in innate immune defense. These intracellular surveillance proteins both detect bacterial PGN, although requiring distinct motifs from PGN to achieve sensing. Detection through Nod1 and Nod2 initiates pro-inflammatory signaling via NF-B activation, which is necessary for clearance of infecting pathogens from the host. We recently demonstrated that the naturally occurring PGN degradation product sensed by Nod1 is GlcNAc-MurNAc-L-Ala-D-Glu-meso-DAP (GM-triDAP). However, the minimal PGN structure is comprised of the dipeptide D-Glu-meso-DAP, in which meso-DAP amino acid is in the terminal position (Figure 1). The presence of meso-DAP is characteristic of most Gram-negative bacteria plus some Gram-positive bacteria, such as Bacillus and Listeria spp. The specificity of Nod1 to detect this subset of bacteria might represent a selective advantage for the host in certain cases when Gram-negative bacteria represent the main threat, such as in the epithelial cells lining the intestinal mucosa.
In contrast, Nod2 has been implicated as a general sensor for both Gram-positive and Gram-negative PGNs since biochemical and functional analysis have identified muramyl dipeptide MurNAc-L-Ala-D-isoGln (MDP), the minimal motif in all PGNs, as the essential structure recognized by Nod2 (Figure 1). Mutations in Nod2 have been associated with autoinflammatory disease in humans, incluidng Crohn's disease, an inflammatory disorder of the intestine and Blau syndrome, a disease that affects the joints and eye tissues. Interestingly, the most common mutation in Nod2 associated with Crohn's disease, which is a frame-shift mutation resulting in the loss of the terminal LRR, results in protein product that no longer detects MDP. Although the implications of these findings are still not fully understood, it appears that lack of bacterial sensing through a loss of interaction between mutant Nod2 and MDP contributes to the pathology of this disease. A loss of surveillance activity by Nod2 may result in the inability of local responses in the intestinal mucosa to control bacterial infection thereby initiating systemic responses and leading to aberrant inflammation.
Our laboratory focuses on examining the role of peptidoglycan recognition molecules, including Nods and PGRPs, in presentation of PGN to the innate immune system. Our interests are also to understand more about the molecular mechanisms of Nod activation by studying the structure and function of Nod1 and Nod2 in terms of ligand binding and interaction with additional binding partners leading to the activation of downstream signaling pathways. Using murine infection models, we also investigate the role of Nod1 and Nod2 in host defense against bacterial challenge. To this end, mice deficient in Nod1 and Nod2 are available in our laboratory. Finally, we are also interested in the contribution of Nod proteins to TLR signaling and how this affects the adaptive immune response. Outlined below are the present research directions of our laboratory and perspectives of where we believe this highly-evolving field is heading.
1. Nod1: ligand detection to signal transduction
a) Determination of Nod1/Nod2 interaction with their ligands: structure / function analysis of the leucine-rich repeat domain of Nod1. (Muguette Jéhanno, technician; in collaboration with Stephen Girardin, Patrick England (Plate-Forme de Biophysique des Macromolécules et de leurs Interactions), Frederic Pecorari and Pedro Alzari (Plate-Forme 6 - Cristallogénèse et Diffraction des Rayons X ).
Our next aim is to examine if Nod1 directly interacts with its ligand or whether this interaction is indirect through possible co-receptors. So far, for TLRs, there are very few data on the direct interaction of these receptors with the respective ligands. To test for interaction of Nod1/Nod2 and their ligands, we will examine in vitro binding of Nod1 to its ligand and with molecular modelling, determine the potential sites of interaction. Affinity studies will be carried out by BIACORE analysis with the wild-type Nod1 as well as mutants that we have generated by site-directed mutagenesis within the putative binding pocket. So far, we have mutated 40 amino acids within this site and 4 of these mutations severely affect the sensing ability of Nod1. We are currently investigating these mutants in more detail in order to determine the origin of their functional defect.
If the interaction is indeed direct, we can then envision crystallographical analysis of Nod1 with its ligand, through collaborations with the crystallography platform directed by Dr. Pedro Alzari.
b) Nod1 interaction with its ligand or other putative binding partners: Cell biological approaches. (Thomas Kufer, post-doctoral fellow).
In parallel with the in vitro work described above, we are also examining the interaction of Nod1 with its PGN ligand or other down-stream signaling partners in the context of infection with Gram-negative bacteria. Co-immunoprecipitation experiments using an antibody to Nod1, which we produced in our laboratory, are planned from infected cells. We can then determine if during the course of infection, Nod1 interacts with its ligand (antibodies to the PGN motif are currently being prepared). If the Nod1/ligand interaction is indirect, using this strategy, we can also define other binding partners required for the interaction. This can be accomplished by silver stain of co-immunoprecipitates followed by mass spectroscopy. In parallel, yeast two hybrid screens have been commenced with full length Nod1 as well as with distinct domains of Nod1 including the LRR, NBS and CARD domains.
c) Role of peptidoglycan recognition proteins - PGRPs. (Rafika Athman, post-doctoral fellow).
PGRPs in Drosophila have been shown to play a critical role in host defense. Mammals possess 4 PGRPs, however, their role in the innate immune response has been less clear. For instance, mice lacking PGRP-L or PGRP-S have only minor defects in combating certain microbial infections. We have examined the potential contribution of PGRPs to bacterial sensing by Nod proteins. PGRPs expression appears to be upregulated by bacterial infection as well as treatment with the individual Nod ligands. Further studies will focus on the mechanims of this regulation of expression as well as the contribution of PGRPs to host defense mediated by Nod proteins.
2. Analysis of the in vivo role of Nod1.
(Jérôme Viala (PhD student), Lionel LeBourhis (DEA student), Richard Ferrero (Unité de Pathogénie Bactérienne des Muqueuses), in collaboration with Catherine Chaput, Ivo Boneca, Stephen Girardin, Philippe Sansonetti (Unité de Pathogénie Microbienne Moléculaire), John Bertin and Peter DeStefano, Millennium Pharmaceuticals, Cambridge MA).
The most important step for determining the role of intracellular bacterial sensing by Nod1 is to validate our in vitro findings in vivo. John Bertin at Millennium Pharmaceuticals provided us with Nod1-deficient mice. So far, no spontaneous phenotype is observed in these mice.
a) Nod1-deficient mice: primary cell culture.
Mouse embryonic fibroblas possess extracellular TLR recognition systems, thereby excluding the possibility for us to use these cells in these assays. Therefore, we developed and refined a technique for isolating intestinal and gastric epithelial cells from mice. Consistent with what we observed in cultured epithelial cells, these primary cells lack extracellular TLR sensing as extracellular bacteria or their products do not activate NF-κ B. Indeed, this lack of extracellular sensing is likely a general feature of epithelial cells since it has been shown that these cells often lack TLR2/TLR4 sensing systems. This may reflect the environment that these cells exist in the body as they are in constant contact with the luminal flora. In contrast, intracellular presentation of Gram-negative bacterial supernatants (through microinjection) or infection of these cells with S. flexneri led to the activation of NF-κB. Strikingly, this was only observed in cells isolated from wild-type mice since cells from mice deficient in Nod1 showed no activation of NF-κB by bacterial products, although these cells could be proficiently activated by TNFα. These studies demonstrate the crucial role of Nod1 in the intracellular sensing of Gram-negative bacteria. Furthermore, these cells are incapable of sensing Gram-positive bacteria, thereby showing that Nod1 is the sole sentinel molecule in epithelial cells capable of bacterial sensing.
b) In vivo infection models
Our work described above showed that Nod1 in epithelial cells senses bacterial products internalized into the cell during bacterial invasion. We have now shown that a non-invasive bacteria, Helicobacter pylori, also stimulates Nod1. Stimulation of this intracellular receptor is dependent on the delivery of peptidoglycan into the host through the type IV secretion apparatus of H. pylori, which is likened to a molecular syringe through which the bacterium can directly translocate bacterial products into epithelial cells. Furthermore, the importance of Nod1 in this process is exemplified by the fact that mice deficient in Nod1 are much more susceptible to infection with H. pylori than their wild-type counterparts; more than 100 times more bacteria were found in the stomachs isolated from the knock-out mice compared to controls (Figure 2).
c) Understanding the role of Nod1 in bacterial clearance. (Lionel LeBourhis, DEA student).
The next aim of this project is to try to understand how Nod1 plays a role in bacterial clearance from the host. Our hypothesis is that Nod1 triggers the production of anti-microbial factors in the epithelial cell that aid the in destruction of bacteria. Consistent with this, we have shown that Nod1 appears to be implicated in the production of b-defensins from H. pylori infected epithelial cells. Dominant-negative Nod1 constructs were able to block H. pylori induction of a human b-defensin-reporter transgene. Moreover, stimulation of Nod1 with its peptidoglycan ligand was able to drive a defensin-promotor driven luciferase construct in treated epithelial cells. Future studies will attempt to examine the role of Nod1 in H. pylori-triggered defensin production in our mouse infection model.
3. Signalling to the adaptive immune response: role of Nod1 and Nod2. (Jörg Fritz, post-doctoral fellow, in collaboration with Stephen Girardin, Philippe Sansonetti, PMM, Minou Adib-Conquy and Jean-Marc Cavaillon, UCI, Jean-Pierre Hugot and Marco Giovaninni, CEPH, for providing Nod2-deficient mice, and Mathieu Allez, Hôptial St. Louis for human samples).
Rather than encountering individual PAMPs during an infection, the host cell is confronted with the entire microbe, which may express many different PAMPs at the same time. Therefore, it is important to study not only the function of individual PRRs following interaction with their ligands but also the combinatorial effects of multiple PRR engagement. Recent findings have suggested that Crohn's disease patients have altered responses to TLR ligands. Indeed, it appears that lack of Nod2 compromises a key "brake" in the inflammatory response against bacteria through TLR2 stimulation. Therefore, we aim to investigate the cross-talk between TLRs and Nods (and Nod1 and Nod2) and the impact of these interactions on the development of the immune response against bacterial infection.
In immune cells such as macrophages and dendritic cells, a wide repertoire of TLR/Nod proteins is expressed to detect the presence of bacteria or bacterial products. While both TLR and Nod sensing systems are known to trigger a pro-inflammatory program independently, the existence of synergistic responses induced by TLRs and Nods remains largely uncharacterized.
Preliminary results have been aimed to identify the optimal conditions in which Nod ligands (M-TriDAP and MDP, for Nod1 and Nod2, respectively) enhance pro-inflammatory signals induced by TLR ligands (LPS, lipoproteins, CpG DNA, etc..). Synergy in TLR and Nod responses will be first examined using cytokine (TNF, IL-6, IL-1b, IL-10, IL-12, IL-1a) secretion as a read-out. Additionally, in the case of dendritic cells, the cell surface expression of co-stimulatory molecules (such as HLA-DR, CD86, CD80, CD40, CD83) will be examined by FACS, following exposure to TLR ligands with/without Nod ligands. Monocytes derived from human blood and macrophages from mouse peritoneal cavity will be used. In addition, dendritic cells from either human donors or mice will be derived for in vitro studies.
In the human model, the cross-talk between TLR and Nod pathways will be studied using immune cells isolated from Crohn's disease patients either wild-type or carrying mutations in the CARD15/Nod2 gene. Monocytes and dendritic cells from these patients will be isolated and studied as described above. In the mouse model, we will take advantage of the large collection of knock-out mice presently available: Nod1-/-, Nod2-/-, TLR2-/-, TLR4-/-, MyD88-/-, TLR2/4-/-. Isolated murine macrophages or dendritic cells will be stimulated in vitro with various TLR or Nod ligands or whole Gram-negative or Gram-positive bacteria. The results from these studies will allow an interpretation of the relative contribution of TLRs and Nod proteins to cytokine production, and cell differentiation and maturation in response to various ligand as well as whole bacteria.
Figure 1: Naturally occurring (red or green) and minimal (blue or orange) peptidoglycan motifs sensed by Nod1 and Nod2. For Nod1, the minimal naturally occurring peptidoglycan (PGN) motif that is detected is the sugar backbone of N-acetylglucosamine (GlcNAc) and N-acetylmuramic acid (MurNAc) linked to a stem peptide consisting of L-Ala-g-D-Glu-meso-DAP (GM-triDAP). The minimal PGN structure detected by Nod1 is the dipeptide D-Glu-meso-DAP, in which meso-DAP amino acid is in the terminal position. Nod2 detects GlcNAc-MurNAc linked to L-Ala-g-D-Glu as the naturally occurring PGN motif whereas the minimal motif consists of MurNAc-L-Ala-g-D-Glu (MDP).
Figure 2: Bacterial loads in the stomachs of Nod1 wild-type or Nod1-deficient mice at seven and thirty days post-infection. Nod1-deficient mice cannot limit bacterial growth as well as wild-type mice.
Keywords: Nod1, Nod2, innate immunity, Crohn's disease, bacterial infection