Role for Erbin in Bacterial Activation of Nod2
http://www.100md.com
感染与免疫杂志 2006年第6期
Immunite Innee et Signalisation, Institut Pasteur, 28 rue du Docteur Roux, 75724 Paris, France
GSF—Institut für Molekulare Immunologie, Marchioninistrasse 25, 81377 München, Germany
ABSTRACT
Intracellular peptidoglycan (PG) recognition in human cells is mediated by the NACHT-LRR proteins Nod1 and Nod2. Elicitation of these proteins by PG motifs released from invasive bacteria triggers signaling events, resulting in the activation of the NF-B pathway. In order to decipher the molecular components involved in Nod2 signal transduction, we set out to identify new interaction partners of Nod2 by using a yeast two-hybrid screen. Besides the known interaction partner RIP2, the screen identified the leucine-rich repeat (LRR)- and PDZ domain-containing family member Erbin as a binding partner of Nod2. Erbin showed a specific interaction with Nod2 in coimmunoprecipitation experiments with human HEK 293T cells. Immunofluorescence microscopy with a newly generated anti-Nod2 monoclonal antibody showed that Erbin and Nod2 partially colocalize in human cells. Subsequent analysis of the Erbin/Nod2 interaction revealed that the LRR of Erbin and the caspase activating and recruiting domains of Nod2 were necessary for this interaction. No significant interaction was observed with a Walker B box mutant of Nod2 or a Crohn's disease-associated frameshift mutant of Nod2, indicating that complex formation is dependent on the activity of the molecule. In addition, a change in the dynamics of the Erbin/Nod2 complex was observed during Shigella flexneri infection. Furthermore, ectopic expression of increasing amounts of Erbin or short hairpin RNA-mediated knockdown of Erbin showed a negative influence of Erbin on Nod2/muramyl-dipeptide-mediated NF-B activation. These results implicate Erbin as a potential negative regulator of Nod2 and show that bacterial infection has an impact on Nod2/Erbin complex formation within cells.
INTRODUCTION
Human cells detect bacteria by pattern recognition molecules that confer responsiveness towards bacterium-derived elicitors (25). Prominent bacterial elicitors are bacterial cell wall components such as lipopeptides and peptidoglycan (PG) (25). Recently, it was shown that these molecules trigger immune responses by interactions with proteins from the family of Toll-like receptors at the cell membrane and NACHT-LRR proteins (NLRs) in the cytosol (2, 16, 23, 28). Nod2, a member of the NLR family, was shown to respond to intracellularly localized muramyl-dipeptide (MDP), a subunit of bacterial PG (15). Interestingly, mutations in the Nod2 gene are linked to the onset of Crohn's disease, an inflammatory disorder of the gut (22, 33, 38).
Nod2 is composed of three types of adjacent domains, including two amino-terminal caspase activating and recruiting domains (CARDs), an internal NACHT domain, and a carboxyl-terminal leucine-rich repeat (LRR) domain. Sensing of MDP is mediated through the LRR domain, leading to downstream signaling by a homophilic interaction of the CARD of RIP2 with the CARDs of Nod2 (34, 40). Subsequently, RIP2 triggers NF-B activation through the formation of a complex with IKK (also called NEMO) and ubiquitination thereof (1, 26). Besides RIP2, direct interactions have also been shown for transforming growth factor beta-activated kinase 1 (TAK1) and GRIM-19 with Nod2, and both interactions are able to modulate NOD2-mediated NF-B activation (5, 8).
Nod2 localizes at the cell membrane in an LRR domain-dependent manner when expressed ectopically in human cells (4, 30). Moreover, this membrane association of Nod2 appears to be important for its MDP-sensing function, since MDP-mediated NF-B activation and interleukin-8 release are dependent on cell membrane localization (4).
Besides sensing bacterial products, Nod2 likely plays a role in controlling bacterial infection. For example, Nod2 can detect Streptococcus pneumoniae (35), and the overexpression of Nod2 in epithelial cells correlates negatively with the survival of Salmonella spp. in epithelial cells (19). Accordingly, Nod2–/– mice display increased susceptibility to Listeria monocytogenes infection via the intragastric route (27). Furthermore, bacterial infection positively regulates Nod2 mRNA expression (35).
Only a few interaction partners of Nod2 have been identified so far, and the mechanisms of signal transduction and regulation of Nod2 are only beginning to emerge. In order to gain a better understanding of the Nod2 signaling pathway in human cells, we applied a screen for interaction partners of Nod2. Here we describe the identification of a new binding partner of Nod2, the human Erbin protein, a member of the leucine-rich repeat- and PDZ domain-containing (LAP) family, and assess its function during triggering of Nod2 in infection of epithelial cells with the invasive pathogen Shigella flexneri.
MATERIALS AND METHODS
Identification of proteins that interact with NOD2 by Y2H screening. Yeast two-hybrid (Y2H) screening and data analysis were performed by Hybrigenics, S.A., Paris, France.
Bait cloning. NOD2 was PCR amplified and cloned into a Y2H vector optimized by Hybrigenics. The bait construct was checked by sequencing the entire insert and was subsequently transformed into the Saccharomyces cerevisiae strain L40 GAL4 (14).
Y2H screening. A human activated leukocyte and mastocyte random-primed cDNA library, transformed into the Y187 yeast strain and containing 10 million independent fragments, was used for mating. A high mating efficiency was obtained by using a specific mating method (29a, 29b, 29c). The screen was first performed on a small scale to adapt the selective pressure to the intrinsic property of the bait. Neither toxicity nor autoactivation of the bait was observed. The full-scale screen was then performed under conditions ensuring a minimum of 50 million interactions tested in order to cover five times the primary complexity of the yeast-transformed cDNA library (37). Seventy-one million interactions were actually tested with Nod2. After selection on medium lacking leucine, tryptophan, and histidine, 39 positive clones were picked and analyzed, and the corresponding prey fragments were amplified by PCR and sequenced at their 5' and 3' junctions. Sequences were then filtered and divided into contigs as described previously (12) and compared to the GenBank database using BLASTN (3). A predicted biological score (PBS) was used to assess the reliability of each interaction, as described previously (12). Briefly, the PBS relies on two different levels of analysis, as follows. Firstly, a local score takes into account the redundancy and independency of prey fragments as well as the distributions of reading frames and stop codons in overlapping fragments. Secondly, a global score takes into account the interactions found in all the screens performed at Hybrigenics using the same library. The PBSs have been shown to positively correlate with the biological significance of interactions (37, 41).
Construction of Nod2- and Erbin-encoding plasmids. Plasmids encoding amino-terminally FLAG-tagged Nod2, deletion constructs, and the indicated domains as well as the expression plasmids for the Nod2 mutant proteins were cloned by PCR. Nod2 mRNA derived from a Nod2 cDNA expression plasmid (kindly provided by G. Nunez) was used as the template, and the PCR products were cloned using BamHI and XhoI restriction sites into the pCMV-Tag2B vector (Stratagene). The FLAG-Nod1 expression plasmid was constructed following the same strategy, using a Nod1 cDNA expression vector as the template. Myc-tagged expression plasmids for Erbin and deletion mutants thereof were a kind gift from L. Mei and P. Dai (20, 21). All constructs used were subjected to full-length sequencing (MilleGen, France).
Generation of monoclonal antibodies against Nod2. An internal peptide of Nod2 (129HPARDLQSHRPAIVRRL145) was synthesized and coupled to keyhole limpet hemocyanin or ovalbumin (PSL, Heidelberg, Germany). Rats were immunized with 50 μg peptide-keyhole limpet hemocyanin using CPG 2006 and incomplete Freund adjuvant as an adjuvant. Supernatants were tested in a differential enzyme-linked immunosorbent assay and analyzed by Western blotting using extracts from HEK 293T cells and transiently transfected HEK 293T cells expressing FLAG-Nod2. Anti-Nod2 7E11 and 4A11 of the rat immunoglobulin G2a (IgG2a) subclass were used in this study.
Cell culture, indirect immunofluorescence microscopy, and bacterial infection. HEK 293T, HT29, and HeLa cells were obtained from the ATCC. Cells were cultivated at 37°C with 5% CO2 in Dulbecco's modified Eagle's medium (Gibco-BRL) supplemented with 10% heat-inactivated fetal calf serum (Gibco-BRL) and penicillin-streptomycin (100 IU/ml and 100 mg/ml, respectively; Gibco-BRL). For indirect immunofluorescence microscopy, cells were seeded on HCl-treated coverslips, fixed in 3% paraformaldehyde in phosphate-buffered saline, and permeabilized with 0.5% Triton X-100 for 5 min. Cells were incubated in 3% bovine serum albumin in phosphate-buffered saline. Staining was done by subsequent incubation of primary and secondary antibodies in 3% bovine serum albumin. Primary antibodies were rat anti-Nod2 7E11 (1:100; this study), mouse anti-FLAG M2 (1:100) (F3165; Sigma-Aldrich), rabbit anti-Myc (1:500) (A-14; Santa Cruz Biotechnology), rabbit anti-Erbin serum (1:500) (20), and rabbit anti-S. flexneri 5a lipopeptide (1:500; kindly provided by J. Mounier and A. Phalipon, Institut Pasteur). Primary antibodies were detected with Texas Red-conjugated goat anti-mouse IgG (1:500; Sigma-Aldrich), Alexa 488-conjugated goat anti-rabbit IgG (1:500; Invitrogen Molecular Probes), and fluorescein isothiocyanate-conjugated goat anti-rat IgG (1:500; Jackson Research) secondary antibodies. DNA was stained with Hoechst 33342 (0.1 μg/ml; Invitrogen Molecular Probes).
Bacterial infection of HeLa and HEK 293T cells was performed using the S. flexneri strain M90T afaE as described previously (36). M90T afaE is a wild-type (WT) invasive strain of S. flexneri serotype 5a harboring the plasmid pIL22, which encodes the afimbrial adhesin from uropathogenic Escherichia coli (9). Briefly, bacteria were added to the cells, which were transferred to serum- and antibiotic-free medium and incubated for 10 min at room temperature prior to transfer to 37°C (time zero).
Immunoprecipitation and immunoblotting. For immunoprecipitation, HEK 293T cells were transiently transfected, using FuGene6 (Roche) according to the manufacturer's conditions, with the indicated plasmids (1 μg plasmid per 6-cm dish) and incubated for 48 h. Cells were lysed in NP-40 buffer (150 mM NaCl, 1% NP-40, 50 mM Tris-HCl, pH 7.5) containing phosphatase inhibitors (20 mM -glycerophosphate, 5 mM NaF, 100 μM Na3VO4, and Complete protease inhibitor cocktail [Roche]). Lysates were cleared for 20 min at 14,000 x g at 4°C. Immunoprecipitation was subsequently carried out for 4 h at 4°C by adding anti-FLAG beads (M2 gel; Sigma-Aldrich) or 9E10 anti-Myc antibody (sc-40; Santa Cruz Biotechnology) and protein G-Sepharose matrix (Amersham Pharmacia) to the cell extracts. The beads were precipitated by centrifugation steps and washed five times in NP-40 buffer before sodium dodecyl sulfate loading buffer was added. Typically, about 10 to 20 times more precipitate than input was loaded into the gel. Proteins were separated by Laemmli sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred by semidry Western transfer to a nitrocellulose membrane (Bio-Rad). Proteins were detected by incubation of the membrane subsequently with primary and secondary antibodies and by a final incubation with SuperSignal West Femto maximum sensitivity substrate (Pierce). Primary antibodies were mouse anti-FLAG M2 (1:1000) (F3165; Sigma-Aldrich), rabbit anti-Myc (1:1,000) (A-14; Santa Cruz Biotechnology), rabbit anti-Erbin serum (1:1,000) (20), rat anti-Nod2 4A11 and 7E11 (1:100; this study), mouse anti--tubulin (1:4,000) (T-7816; Sigma-Aldrich), and rabbit anti-IB (1:1,000) (C-21; Santa Cruz Biotechnology). Secondary antibodies were horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG (1:4,000) (170-6616; Bio-Rad), HRP-conjugated goat anti-rabbit IgG (1:4,000) (170-6515; Bio-Rad), and HRP-conjugated goat anti-rat IgG (1:4,000; Jackson Research Laboratory).
Luciferase reporter assays. Activation of NF-B was measured using a luciferase reporter assay as described previously (36). Briefly, HEK 293T cells were seeded in 24-well plates and transfected using Fugene6 (Roche), 0.25 ng Nod2 or Nod2 K305R, 75 ng Igk-luciferase NF-B reporter (31), and 25 ng -galactosidase expression plasmid per well. Cells were directly stimulated with 200 nM MDP or 0.01 μg/ml tumor necrosis factor (TNF), as indicated. After 16 h of incubation, the cells were lysed and luciferase activity was measured. The mean and standard error of the mean (SEM) were calculated from triplets, and luciferase activity was normalized as a ratio to -galactosidase activity. For short hairpin RNA (shRNA) experiments, cells were transfected with the indicated amount of an Erbin-specific shRNA plasmid (pBLOCK-it; Invitrogen), a kind gift of P. Dai and L. Mei (10), by using FuGene 72 h prior to transfection with the reporter and stimulation. Western analysis of the Erbin protein amounts after shRNA treatment was conducted with the lysates used for the luciferase activity test. For infection, S. flexneri M90T afaE was added to the transfected cells, and the medium was replaced with Dulbecco's modified Eagle's medium containing 50 μg/ml gentamicin 15 min after incubation at 37°C.
RESULTS
Identification of Erbin. In order to identify new interaction partners of the human Nod2 protein, a yeast two-hybrid screen was applied. We found that full-length Nod2 was functional in a yeast two-hybrid system by using a directed interaction test with the known binding partner RIP2 (data not shown). Since Nod2 is expressed in leukocytes (13, 18, 34), a yeast two-hybrid screen using full-length Nod2 as bait was therefore performed with a human leukocyte/mastocyte library. Interestingly, one of the interacting clones obtained encoded the CARD of RIP2, which is known to interact with Nod2 (34), thereby validating our approach. The clones with the highest probability scores for interaction among the candidates were two independent clones spanning the LRR domain of the human LAP family member Erbin (6, 7) (Fig. 1). Moreover, the expression profile reported for Erbin was closely related to the expression pattern of Nod2, with both proteins being highly expressed in CD14+ monocytes and CD33+ myeloid cells in both humans and mice (39).
The physical interaction of Nod2 and Erbin was subsequently confirmed in human cells by coimmunoprecipitation from transiently transfected HEK 293T cells. Endogenous Erbin protein was found to coprecipitate with FLAG-Nod2 (Fig. 2A). Moreover, ectopically expressed Myc-Erbin, but not the matrix alone, coprecipitated with FLAG-Nod2 (Fig. 2B). The interaction between Nod2 and Erbin was shown to be specific, as no binding of Erbin was observed with the Nod2-related protein Nod1 (Fig. 2A), nor did Nod2 bind the LAP protein Scribble, a close relative of Erbin (7; data not shown). Both a P-loop mutant of Nod2 (K305R) and a Crohn's disease-associated frameshift mutant of Nod2 (L1007fs) are known to be compromised in sensing of MDP (15, 22, 33, 34, 40). Whereas Erbin bound to WT Nod2, no significant interaction was observed with Nod2 K305R or Nod2 L1007fs (Fig. 2C).
Taken together, these findings show that Erbin forms a complex with Nod2 in human cells and that the affinity for Nod2 seems to depend on the MDP-sensing functionality of the Nod2 protein.
Mapping of interaction domains. To establish the protein domains responsible for the interaction between the two proteins, coimmunoprecipitation experiments with Erbin and Nod2 deletion constructs were conducted with lysates of transiently transfected HEK 293T cells. FLAG-Nod2 not only bound to the full-length Myc-Erbin protein but also showed binding at nearly the same affinity to a deletion protein of Erbin lacking the PDZ domain (amino acids [aa] 1 to 1279) (Fig. 2D, lanes 1 and 3). On the other hand, no interaction was observed with an amino-terminal deletion protein of Erbin containing the PDZ domain alone (aa 965 to 1371) (Fig. 2D, lane 6). However, a mutant of Erbin lacking the LRR domain (aa 391 to 1371) still showed very weak binding, and also the LRR domain (aa 1 to 391) and the central domain of Erbin (aa 391 to 965) itself bound weakly to Nod2 (Fig. 2D, lanes 2, 4 and 5).
Using the same approach with Nod2, the expected binding of full-length Erbin to Nod2 was observed, whereas no binding was observed to Nod1 or the matrix alone (Fig. 2E). Furthermore, a mutant lacking the LRR domain of Nod2 (aa 1 to 600) bound Myc-Erbin, although to a lesser extent (Fig. 2E). Importantly, no interaction was observed with a mutant lacking the two CARDs of Nod2 (aa 250 to 1040) or with the NACHT domain (aa 250 to 600) or CARD (aa 1 to 250) alone (Fig. 2E).
In conclusion, these results support the hypothesis that the domain responsible for mediating the interaction with Nod2 lies within the carboxyl terminus of the LRR domain of Erbin. On the side of Nod2, the interaction is mediated by the CARDs of Nod2; however, these domains seemed not to be sufficient to mediate the interaction.
Localization in human cells. To explore the site of complex formation between Nod2 and Erbin within the cell, the subcellular localization of the two proteins in human cells was examined. HeLa cells were transfected transiently with FLAG-tagged human Nod2 and Myc-Erbin, and the proteins were visualized using the indicated antibodies. Consistent with recent reports describing the localization of Erbin in human cells, the Erbin antibody gave a strong signal at the cell cortex and a weak signal in the cytosol (6, 29, 30) (Fig. 3A). As reported recently, we also observed Nod2 at the cell membrane and in the cytoplasm (4, 30) (Fig. 3A). Depending on the expression level of ectopically expressed Nod2, the protein was found in the cytosol and was more or less prominent at the cell membrane and in membrane ruffles, where the merged signals of Nod2 and Erbin revealed partial colocalization (Fig. 3A). Notably, about half of the transfected cells exhibited nuclear localization of the ectopically expressed Nod2 protein (Fig. 3A and B).
Erbin was initially reported to play a role in targeting the ErbB2 receptor to the membrane (6), and Erbin associates with basolateral membranes though an LRR-domain-dependent mechanism (29). Recently, a domain within the LRR of Nod2 was also reported to be essential for the membrane association of Nod2 (4). The deletion of the LRR domain in Nod2 interrupted the interaction with Erbin (Fig. 2C and E), making it tempting to speculate that Erbin might target Nod2 to the membrane. To investigate a possible involvement of Erbin in localizing Nod2 at the membrane, endogenous Erbin was silenced by shRNA. HeLa cells transiently cotransfected with FLAG-Nod2 and the Erbin shRNA plasmid showed a clear reduction in the Erbin-specific immunofluorescence signal in transfected cells (Fig. 3B, arrows). However, the membrane localization pattern of ectopically expressed Nod2 in these cells was indistinguishable from that in normal HeLa cells (Fig. 3B, arrows). Thus, Erbin seems not to be essential for the targeting of Nod2 to the membrane.
To determine the localization of endogenous Nod2, Nod2-specific monoclonal antibodies were generated. Two clones, 4A11 and 7E11, specifically recognized the ectopically expressed FLAG-Nod2 protein expressed in HEK 293T cells but gave no signal in lysates from untransfected HEK 293T cells (Fig. 3C; data not shown). Whereas 4A11 gave the strongest signal in Western blot analysis, 7E11 gave the best signal by immunofluorescence and recognized endogenous Nod2 protein in lysates from HT29 cells, which were shown to express higher levels of endogenous Nod2 (4) (Fig. 3C). Using these antibodies and a polyclonal serum specific for Erbin (20), colocalization of Nod2 and Erbin at the cell membrane in undifferentiated cells from the human colon cancer cell line HT29 was observed (Fig. 3D). Notably, the Nod2 antibody also gave signals in the nuclei of HT29, SW480, HeLa, and HEK cells (Fig. 3D; data not shown).
Taken together, these data provide evidence that a partial pool of Nod2 is localized in an Erbin-independent manner at the cell surface, where it partially colocalizes with Erbin.
Influence of Erbin on Nod2-mediated NF-B activation. Nod2 triggers the NF-B signaling pathway via activation of RIP2 (26, 34). This activation can be studied specifically in HEK cells transiently transfected with Nod2 and a NF-B reporter construct by using the Nod2-specific elicitor MDP (15, 17, 40). Since HEK 293T cells do express detectable amounts of Erbin (Fig. 2A and 4B), Nod2-mediated NF-B activation was assessed in these cells, and the influence of Erbin on this signaling pathway was tested by shRNA-mediated Erbin knockdown. As shown in Fig. 4B, the Erbin shRNA significantly reduced the amount of total Erbin protein detected by Western analysis. Quantification of NF-B reporter activity in these cells showed that a knockdown of Erbin correlated with higher reporter activity when cells were treated with MDP; however, no correlation between the Erbin protein level and basal Nod2-driven NF-B reporter activity was observed (Fig. 4A). In contrast, ectopic expression of Erbin resulted in reduced Nod2/MDP-mediated NF-B reporter activity (Fig. 4C). This influence on Nod2/MDP-mediated NF-B activation was specific, as TNF-mediated NF-B reporter activity was unaltered when increasing amounts of Erbin were expressed (Fig. 4C).
In conclusion, these results support an involvement of Erbin as a negative regulator of the MDP-mediated activation of NF-B by Nod2.
Nod2/Erbin interaction during bacterial infection. Nod2 is proposed to be the major sensor for intracellularly located, bacterially derived MDP (15, 28). Accordingly, we found that NF-B activation triggered by invasion of HEK 293T cells with the gram-negative bacterium S. flexneri was enhanced when Nod2, but not the Nod2 K305R mutant, was ectopically expressed (Fig. 5A). Since Erbin was shown to modulate NF-B activation after elicitation of Nod2, the subcellular localization of Erbin and Nod2 was analyzed during biological activation of Nod2, using the S. flexneri infection model. For this purpose, HeLa cells transiently transfected with FLAG-Nod2 were infected with the invasive S. flexneri strain M90T expressing the E. coli AfaE adhesin to mediate a higher infection efficiency. During bacterial invasion, in all infected cells, the Nod2 signal was predominant at the entry foci of the bacteria (Fig. 5B and C). At the same time, the Erbin signal was also enhanced at the site of bacterial entry, where it showed colocalization with Nod2 (Fig. 5C). S. flexneri induces membrane ruffling at the site of entry, leading to a higher surface area at the entry site (32), which underscores the observation that both Nod2 and Erbin are localized at the cell membrane.
The coincidence of the two proteins being present at the bacterial entry focus makes it tempting to speculate that this is of biological relevance in sensing bacteria through Nod2. To examine this possibility in greater detail, we asked whether colocalization at the site of bacterial entry could reflect changes in the stoichiometry of the Nod2/Erbin complex. To this end, HEK 293T cells were transiently transfected with FLAG-Nod2 and subsequently infected with S. flexneri M90T afaE. At the indicated times after infection, cells were lysed, and equal amounts were subjected to immunoprecipitation using anti-FLAG antibody. Coprecipitated endogenous Erbin protein was detected by Western analysis. Monitoring the amount of IB in the total lysates showed a degradation of IB 60 min after infection, as expected for successful bacterial invasion using this infection model (36). Importantly, a change in the amounts of coprecipitated endogenous and ectopically expressed Erbin was observed, whereas virtually equal amounts of Nod2 were precipitated at each time point of the experiment, and the Erbin signal in the total lysate did not change over time (Fig. 5D and data not shown). The amount of coprecipitated Erbin protein was lower than that in uninfected cells at early times of infection (10 to 20 min) and increased over time to reach a peak at around 30 to 40 min postinfection (Fig. 5D). To rule out the possibility that this might reflect small changes in Erbin expression after infection that are not detectable by Western analysis, the same experiment was conducted with coexpression of ectopic FLAG-Nod2 and Myc-Erbin. Again, a clear change over time in the amount of coprecipitated Myc-Erbin was observed, supporting the finding of the previous experiment (Fig. 5D).
These results show conclusively that S. flexneri triggers Nod2-mediated NF-B activation, and they support a role of Erbin in bacterially triggered Nod2 activation. Indeed, Nod2 and Erbin are present at the entry foci during S. flexneri infection of human cells, and the stoichiometry of the Nod2/Erbin complex is altered during infection.
DISCUSSION
Using a yeast two-hybrid screen with Nod2 as bait, this work identified the LAP family member Erbin as an interaction partner of human Nod2. This interaction was verified in coimmunoprecipitation experiments with human cells, which showed that the interaction domains lie within the LRR domain of Erbin and the CARDs of Nod2. In epithelial cells, the two proteins were further shown to partially colocalize at the cell membrane and at the entry foci during infection with invasive S. flexneri.
To date, only a few components of the Nod2 signaling pathway are known, with the best-studied example being the serine/threonine kinase RIP2. Studies using dominant-negative mutants and knockout mice showed that this protein is essential for Nod2-mediated NF-B signaling (26, 34). The exact mechanism of signal transduction through RIP2 is still not fully understood, but recent evidence shows that proximity activation of RIP2 and subsequent ubiquitination of IKK (also called NEMO) are crucial for downstream NF-B activation (1, 24, 34). Molecules that enhance Nod2 signaling have also been reported and include TAK1 (8) and GRIM-19 (5). Interestingly, GRIM-19 is also implicated in bacterial infection, and a role for this protein in intracellular Salmonella survival was proposed (5). On the other hand, the only negative regulation process for Nod2 was proposed by heteromerization with the NLR member Ipaf (also called CLAN) (11). The identification of a new negative regulator of Nod2 is therefore a further step towards an understanding of the signaling events mediated by Nod proteins.
Recently, Erbin was independently identified in a biochemical screen (tandem affinity purification) for new Nod2 binding proteins and reported to negatively regulate Nod2-mediated NF-B activation (30). The data reported here support this previous finding with regard to the identification of a physical interaction between Erbin and Nod2 and a function of Erbin in influencing Nod2-mediated NF-B activation. However, some minor discrepancies exist with regard to the mapping of the interaction domain. Whereas, for example, both data from the work of McDonald et al. and this study show that the PDZ domain of Erbin is not involved in mediating the interaction with Nod2, McDonald et al. reported that a carboxyl-terminal domain of Erbin (aa 965 to 1371) can bind to Nod2 (30). In the present study, however, no affinity for Nod2 was observed for this domain. This result is further strengthened by the results from the yeast two-hybrid screen, which did not yield clones for the carboxyl-terminal part of Erbin. The discrepancy in these results might be partially explainable by the use of different constructs to conduct the coimmunoprecipitation experiments. Further independent examination might help to clarify this issue.
Furthermore, our findings showed that the Nod2/Erbin interaction is dependent on the CARDs of Nod2, a finding that was also supported by the work of McDonald et al. (30). However, a dramatic decrease in binding affinity was also seen when the LRR domain of Nod2 was partially (Nod2 L1007fs) or completely deleted. Accumulated data from studies using Nod2 deletion mutants have led us to propose that this decrease was due to a loss in proper functionality of the deletion proteins. Most likely, this might be explained by improper folding or aggregate formation of these mutant Nod2 proteins.
A Nod2 membrane association was recently reported (4) and was confirmed in this study for endogenous and ectopically expressed Nod2 protein, using monoclonal antibodies directed against Nod2. Intriguingly, a nuclear signal for Nod2 was also observed. The reasons for the discrepancy in this observation with regard to recent reports on the localization of Nod2 (4, 30) are currently unclear but might reflect differences in the methodologies used or the affinities of the antibodies. The L1007 frameshift mutant of Nod2 was reported to exhibit a more cytoplasmic localization than the WT Nod2 protein when ectopically expressed in human cells (4, 30). Interestingly, this was also the case for the K305R P-loop mutant of Nod2 and a deletion mutant of Nod2 lagging the LRR domain (data not shown). Since the LRR domain of Nod2 alone was not sufficient to mediate membrane localization (data not shown), this suggests that membrane targeting of Nod2 may not be mediated by a domain in the LRR of Nod2, as proposed recently (4), but rather may depend on the functionality of the protein. This indicates that the localization of Nod2 at the cell cortex might be a secondary event dependent on the signaling integrity of the Nod2 protein and, moreover, might be influenced by activation of Nod2 by its bacterial elicitor.
Importantly, Erbin and Nod2 colocalized at the entry foci of S. flexneri. Since both proteins are recruited to the cell membrane, the higher signal intensity at this site might only correlate with the higher surface area at the entry site provoked by bacterially induced membrane ruffling (32). However, this finding illustrates that the molecules are in close proximity at the site of first peptidoglycan release from the bacteria in the cell. Nod2 was shown to be an antibacterial factor that limits the survival of intracellular bacteria when ectopically expressed (19). Therefore, even if recruitment to the entry foci might not be an active principle, it offers a biological explanation for Nod2 localization at the membrane for immediate sensing of bacterial products that might be transferred to the cytosol upon bacterial invasion.
An analysis of Nod2-mediated NF-B activation showed that ectopic expression of Erbin had a negative effect on Nod2-mediated NF-B signaling, whereas knockdown of Erbin exhibited a positive effect. These findings support the view that Erbin is a negative regulator of Nod2-mediated NF-B signaling. Furthermore, infection of cells with invasive S. flexneri triggered Nod2 signaling and led to a change in the stoichiometry of the Erbin/Nod2 complex, showing a maximal affinity of Nod2 for Erbin at 30 to 40 min postinfection. Erbin was further shown to bind only WT Nod2, but not MDP-sensing inactive mutants, such as a P-loop mutant (Nod2 K305R) and a Crohn's disease-associated frameshift mutant lacking part of the LRR domain of Nod2 (Nod2 L1007fs) (15, 34, 40). Taken together, these results suggest that activated Nod2 has a higher affinity for Erbin than does the inactive protein. Notably, Nod2 and RIP2 might also form a transient complex after elicitation of Nod2 where Nod2 forms active oligomers, presumably via homophilic NACHT/NACHT domain interactions (34). Since overexpression of Erbin had a negative impact on Nod2-mediated NF-B activation, it is tempting to speculate that this negative regulatory mechanism is exerted by inhibiting the binding of RIP2. Thus, Erbin might play an important role in dampening Nod2 signaling events in the cell. However, no significant influence of overexpression of RIP2 was observed on binding of Erbin to Nod2 (data not shown).
More importantly, this study expands previous observations regarding the involvement of Nod2 and Erbin in bacterial infection. A clear change in the dynamics of the Nod2/Erbin complex was observed after bacterial infection, even if the precise biological function of the observation remains to be established. Together with the in vivo colocalization data, which revealed the two proteins at the first contact site of bacterial invasion, and the inhibitory effect of Nod2 on the replication of invasive bacteria (19), we assume that Erbin plays a pivotal role in orchestrating Nod2 signaling during bacterial infection. Further studies will show if other proteins are involved in the regulation of Nod2 and if there might be a functional interplay between Nod2-interacting proteins, such as GRIM-19 or TAK1, and the Nod2/Erbin complex.
ACKNOWLEDGMENTS
We thank S. Girardin and J. H. Fritz for advice and critical discussions as well as all the members of the IIS for helpful comments. We are grateful to M. Jehanno for help in cloning the Nod2FS construct and M. Parmier for technical assistance. We thank L. Mei and P. Dai for sharing their expertise on Erbin and the kind gift of Erbin reagents and G. Nunez for providing the Nod2 expression plasmid.
T. A. Kufer acknowledges support by a FEBS long-term fellowship and funding from the Institut Pasteur for the yeast two-hybrid screen.
REFERENCES
1. Abbott, D. W., A. Wilkins, J. M. Asara, and L. C. Cantley. 2004. The Crohn's disease protein, NOD2, requires RIP2 in order to induce ubiquitinylation of a novel site on NEMO. Curr. Biol. 14:2217-2227.
2. Akira, S., and K. Takeda. 2004. Toll-like receptor signalling. Nat. Rev. Immunol. 4:499-511.
3. Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang, W. Miller, and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25:3389-3402.
4. Barnich, N., J. E. Aguirre, H. C. Reinecker, R. Xavier, and D. K. Podolsky. 2005. Membrane recruitment of NOD2 in intestinal epithelial cells is essential for nuclear factor-B activation in muramyl dipeptide recognition. J. Cell Biol. 170:21-26.
5. Barnich, N., T. Hisamatsu, J. E. Aguirre, R. Xavier, H. C. Reinecker, and D. K. Podolsky. 2005. GRIM-19 interacts with nucleotide oligomerization domain 2 and serves as downstream effector of anti-bacterial function in intestinal epithelial cells. J. Biol. Chem. 280:19021-19026.
6. Borg, J. P., S. Marchetto, A. Le Bivic, V. Ollendorff, F. Jaulin-Bastard, H. Saito, E. Fournier, J. Adelaide, B. Margolis, and D. Birnbaum. 2000. ERBIN: a basolateral PDZ protein that interacts with the mammalian ERBB2/HER2 receptor. Nat. Cell Biol. 2:407-414.
7. Bryant, P. J., and A. Huwe. 2000. LAP proteins: what's up with epithelia Nat. Cell Biol. 2:E141-E143.
8. Chen, C. M., Y. Gong, M. Zhang, and J. J. Chen. 2004. Reciprocal cross-talk between Nod2 and TAK1 signaling pathways. J. Biol. Chem. 279:25876-25882.
9. Clerc, P., and P. J. Sansonetti. 1987. Entry of Shigella flexneri into HeLa cells: evidence for directed phagocytosis involving actin polymerization and myosin accumulation. Infect. Immun. 55:2681-2688.
10. Dai, P., W. Xiong, and L. Mei. 2006. Erbin inhibits RAF activation by disrupting the Sur-8-Ras-Raf complex. J. Biol. Chem. 281:927-933.
11. Damiano, J. S., V. Oliveira, K. Welsh, and J. C. Reed. 2004. Heterotypic interactions among NACHT domains: implications for regulation of innate immune responses. Biochem. J. 381:213-219.
12. Formstecher, E., S. Aresta, V. Collura, A. Hamburger, A. Meil, A. Trehin, C. Reverdy, V. Betin, S. Maire, C. Brun, B. Jacq, M. Arpin, Y. Bellaiche, S. Bellusci, P. Benaroch, M. Bornens, R. Chanet, P. Chavrier, O. Delattre, V. Doye, R. Fehon, G. Faye, T. Galli, J. A. Girault, B. Goud, J. de Gunzburg, L. Johannes, M. P. Junier, V. Mirouse, A. Mukherjee, D. Papadopoulo, F. Perez, A. Plessis, C. Rosse, S. Saule, D. Stoppa-Lyonnet, A. Vincent, M. White, P. Legrain, J. Wojcik, J. Camonis, and L. Daviet. 2005. Protein interaction mapping: a Drosophila case study. Genome Res. 15:376-384.
13. Fritz, J. H., S. E. Girardin, C. Fitting, C. Werts, D. Mengin-Lecreulx, M. Caroff, J. M. Cavaillon, D. J. Philpott, and M. Adib-Conquy. 2005. Synergistic stimulation of human monocytes and dendritic cells by Toll-like receptor 4 and NOD1- and NOD2-activating agonists. Eur. J. Immunol. 35:2459-2470.
14. Fromont-Racine, M., J. C. Rain, and P. Legrain. 1997. Toward a functional analysis of the yeast genome through exhaustive two-hybrid screens. Nat. Genet. 16:277-282.
15. Girardin, S. E., I. G. Boneca, J. Viala, M. Chamaillard, A. Labigne, G. Thomas, D. J. Philpott, and P. J. Sansonetti. 2003. Nod2 is a general sensor of peptidoglycan through muramyl dipeptide (MDP) detection. J. Biol. Chem. 278:8869-8872.
16. Girardin, S. E., P. J. Sansonetti, and D. J. Philpott. 2002. Intracellular vs extracellular recognition of pathogens—common concepts in mammals and flies. Trends Microbiol. 10:193-199.
17. Girardin, S. E., L. H. Travassos, M. Herve, D. Blanot, I. G. Boneca, D. J. Philpott, P. J. Sansonetti, and D. Mengin-Lecreulx. 2003. Peptidoglycan molecular requirements allowing detection by Nod1 and Nod2. J. Biol. Chem. 278:41702-41708.
18. Gutierrez, O., C. Pipaon, N. Inohara, A. Fontalba, Y. Ogura, F. Prosper, G. Nunez, and J. L. Fernandez-Luna. 2002. Induction of Nod2 in myelomonocytic and intestinal epithelial cells via nuclear factor-kappa B activation. J. Biol. Chem. 277:41701-41705.
19. Hisamatsu, T., M. Suzuki, H. C. Reinecker, W. J. Nadeau, B. A. McCormick, and D. K. Podolsky. 2003. CARD15/NOD2 functions as an antibacterial factor in human intestinal epithelial cells. Gastroenterology 124:993-1000.
20. Huang, Y. Z., Q. Wang, W. C. Xiong, and L. Mei. 2001. Erbin is a protein concentrated at postsynaptic membranes that interacts with PSD-95. J. Biol. Chem. 276:19318-19326.
21. Huang, Y. Z., M. Zang, W. C. Xiong, Z. Luo, and L. Mei. 2003. Erbin suppresses the MAP kinase pathway. J. Biol. Chem. 278:1108-1114.
22. Hugot, J. P., M. Chamaillard, H. Zouali, S. Lesage, J. P. Cezard, J. Belaiche, S. Almer, C. Tysk, C. A. O'Morain, M. Gassull, V. Binder, Y. Finkel, A. Cortot, R. Modigliani, P. Laurent-Puig, C. Gower-Rousseau, J. Macry, J. F. Colombel, M. Sahbatou, and G. Thomas. 2001. Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn's disease. Nature 411:599-603.
23. Inohara, N., M. Chamaillard, C. McDonald, and G. Nunez. 2005. NOD-LRR proteins: role in host-microbial interactions and inflammatory disease. Annu. Rev. Biochem. 74:355-383.
24. Inohara, N., T. Koseki, J. Lin, L. del Peso, P. C. Lucas, F. F. Chen, Y. Ogura, and G. Nunez. 2000. An induced proximity model for NF-kappa B activation in the Nod1/RICK and RIP signaling pathways. J. Biol. Chem. 275:27823-27831.
25. Janeway, C. A., Jr., and R. Medzhitov. 2002. Innate immune recognition. Annu. Rev. Immunol. 20:197-216.
26. Kobayashi, K., N. Inohara, L. D. Hernandez, J. E. Galan, G. Nunez, C. A. Janeway, R. Medzhitov, and R. A. Flavell. 2002. RICK/Rip2/CARDIAK mediates signalling for receptors of the innate and adaptive immune systems. Nature 416:194-199.
27. Kobayashi, K. S., M. Chamaillard, Y. Ogura, O. Henegariu, N. Inohara, G. Nunez, and R. A. Flavell. 2005. Nod2-dependent regulation of innate and adaptive immunity in the intestinal tract. Science 307:731-734.
28. Kufer, T. A., J. H. Fritz, and D. J. Philpott. 2005. NACHT-LRR proteins (NLRs) in bacterial infection and immunity. Trends Microbiol. 13:381-388.
29. Legouis, R., F. Jaulin-Bastard, S. Schott, C. Navarro, J. P. Borg, and M. Labouesse. 2003. Basolateral targeting by leucine-rich repeat domains in epithelial cells. EMBO Rep. 4:1096-1102.
29a. Legrain, P., et al. 1998. U.S. patent 6187535.
29b. Legrain, P., et al. 2000. U.S. patent 6531284.
29c. Legrain, P., et al. 2002. U.S. patent 6913886.
30. McDonald, C., F. F. Chen, V. Ollendorff, Y. Ogura, S. Marchetto, P. Lecine, J. P. Borg, and G. Nunez. 2005. A role for Erbin in the regulation of Nod2-dependent NF-kappaB signaling. J. Biol. Chem. 280:40301-40309.
31. Munoz, E., G. Courtois, P. Veschambre, P. Jalinot, and A. Israel. 1994. Tax induces nuclear translocation of NF-B through dissociation of cytoplasmic complexes containing p105 or p100 but does not induce degradation of IB/MAD3. J. Virol. 68:8035-8044.
32. Nhieu, G. T., and P. J. Sansonetti. 1999. Mechanism of Shigella entry into epithelial cells. Curr. Opin. Microbiol. 2:51-55.
33. Ogura, Y., D. K. Bonen, N. Inohara, D. L. Nicolae, F. F. Chen, R. Ramos, H. Britton, T. Moran, R. Karaliuskas, R. H. Duerr, J. P. Achkar, S. R. Brant, T. M. Bayless, B. S. Kirschner, S. B. Hanauer, G. Nunez, and J. H. Cho. 2001. A frameshift mutation in NOD2 associated with susceptibility to Crohn's disease. Nature 411:603-606.
34. Ogura, Y., N. Inohara, A. Benito, F. F. Chen, S. Yamaoka, and G. Nunez. 2001. Nod2, a Nod1/Apaf-1 family member that is restricted to monocytes and activates NF-kappaB. J. Biol. Chem. 276:4812-4818.
35. Opitz, B., A. Puschel, B. Schmeck, A. C. Hocke, S. Rosseau, S. Hammerschmidt, R. R. Schumann, N. Suttorp, and S. Hippenstiel. 2004. Nucleotide-binding oligomerization domain proteins are innate immune receptors for internalized Streptococcus pneumoniae. J. Biol. Chem. 279:36426-36432.
36. Philpott, D. J., S. Yamaoka, A. Israel, and P. J. Sansonetti. 2000. Invasive Shigella flexneri activates NF-kappa B through a lipopolysaccharide-dependent innate intracellular response and leads to IL-8 expression in epithelial cells. J. Immunol. 165:903-914.
37. Rain, J. C., L. Selig, H. De Reuse, V. Battaglia, C. Reverdy, S. Simon, G. Lenzen, F. Petel, J. Wojcik, V. Schachter, Y. Chemama, A. Labigne, and P. Legrain. 2001. The protein-protein interaction map of Helicobacter pylori. Nature 409:211-215.
38. Schreiber, S., P. Rosenstiel, M. Albrecht, J. Hampe, and M. Krawczak. 2005. Genetics of Crohn disease, an archetypal inflammatory barrier disease. Nat. Rev. Genet. 6:376-388.
39. Su, A. I., M. P. Cooke, K. A. Ching, Y. Hakak, J. R. Walker, T. Wiltshire, A. P. Orth, R. G. Vega, L. M. Sapinoso, A. Moqrich, A. Patapoutian, G. M. Hampton, P. G. Schultz, and J. B. Hogenesch. 2002. Large-scale analysis of the human and mouse transcriptomes. Proc. Natl. Acad. Sci. USA 99:4465-4470.
40. Tanabe, T., M. Chamaillard, Y. Ogura, L. Zhu, S. Qiu, J. Masumoto, P. Ghosh, A. Moran, M. M. Predergast, G. Tromp, C. J. Williams, N. Inohara, and G. Nunez. 2004. Regulatory regions and critical residues of NOD2 involved in muramyl dipeptide recognition. EMBO J. 23:1587-1597.
41. Wojcik, J., I. G. Boneca, and P. Legrain. 2002. Prediction, assessment and validation of protein interaction maps in bacteria. J. Mol. Biol. 323:763-770.(T. A. Kufer, E. Kremmer, )
GSF—Institut für Molekulare Immunologie, Marchioninistrasse 25, 81377 München, Germany
ABSTRACT
Intracellular peptidoglycan (PG) recognition in human cells is mediated by the NACHT-LRR proteins Nod1 and Nod2. Elicitation of these proteins by PG motifs released from invasive bacteria triggers signaling events, resulting in the activation of the NF-B pathway. In order to decipher the molecular components involved in Nod2 signal transduction, we set out to identify new interaction partners of Nod2 by using a yeast two-hybrid screen. Besides the known interaction partner RIP2, the screen identified the leucine-rich repeat (LRR)- and PDZ domain-containing family member Erbin as a binding partner of Nod2. Erbin showed a specific interaction with Nod2 in coimmunoprecipitation experiments with human HEK 293T cells. Immunofluorescence microscopy with a newly generated anti-Nod2 monoclonal antibody showed that Erbin and Nod2 partially colocalize in human cells. Subsequent analysis of the Erbin/Nod2 interaction revealed that the LRR of Erbin and the caspase activating and recruiting domains of Nod2 were necessary for this interaction. No significant interaction was observed with a Walker B box mutant of Nod2 or a Crohn's disease-associated frameshift mutant of Nod2, indicating that complex formation is dependent on the activity of the molecule. In addition, a change in the dynamics of the Erbin/Nod2 complex was observed during Shigella flexneri infection. Furthermore, ectopic expression of increasing amounts of Erbin or short hairpin RNA-mediated knockdown of Erbin showed a negative influence of Erbin on Nod2/muramyl-dipeptide-mediated NF-B activation. These results implicate Erbin as a potential negative regulator of Nod2 and show that bacterial infection has an impact on Nod2/Erbin complex formation within cells.
INTRODUCTION
Human cells detect bacteria by pattern recognition molecules that confer responsiveness towards bacterium-derived elicitors (25). Prominent bacterial elicitors are bacterial cell wall components such as lipopeptides and peptidoglycan (PG) (25). Recently, it was shown that these molecules trigger immune responses by interactions with proteins from the family of Toll-like receptors at the cell membrane and NACHT-LRR proteins (NLRs) in the cytosol (2, 16, 23, 28). Nod2, a member of the NLR family, was shown to respond to intracellularly localized muramyl-dipeptide (MDP), a subunit of bacterial PG (15). Interestingly, mutations in the Nod2 gene are linked to the onset of Crohn's disease, an inflammatory disorder of the gut (22, 33, 38).
Nod2 is composed of three types of adjacent domains, including two amino-terminal caspase activating and recruiting domains (CARDs), an internal NACHT domain, and a carboxyl-terminal leucine-rich repeat (LRR) domain. Sensing of MDP is mediated through the LRR domain, leading to downstream signaling by a homophilic interaction of the CARD of RIP2 with the CARDs of Nod2 (34, 40). Subsequently, RIP2 triggers NF-B activation through the formation of a complex with IKK (also called NEMO) and ubiquitination thereof (1, 26). Besides RIP2, direct interactions have also been shown for transforming growth factor beta-activated kinase 1 (TAK1) and GRIM-19 with Nod2, and both interactions are able to modulate NOD2-mediated NF-B activation (5, 8).
Nod2 localizes at the cell membrane in an LRR domain-dependent manner when expressed ectopically in human cells (4, 30). Moreover, this membrane association of Nod2 appears to be important for its MDP-sensing function, since MDP-mediated NF-B activation and interleukin-8 release are dependent on cell membrane localization (4).
Besides sensing bacterial products, Nod2 likely plays a role in controlling bacterial infection. For example, Nod2 can detect Streptococcus pneumoniae (35), and the overexpression of Nod2 in epithelial cells correlates negatively with the survival of Salmonella spp. in epithelial cells (19). Accordingly, Nod2–/– mice display increased susceptibility to Listeria monocytogenes infection via the intragastric route (27). Furthermore, bacterial infection positively regulates Nod2 mRNA expression (35).
Only a few interaction partners of Nod2 have been identified so far, and the mechanisms of signal transduction and regulation of Nod2 are only beginning to emerge. In order to gain a better understanding of the Nod2 signaling pathway in human cells, we applied a screen for interaction partners of Nod2. Here we describe the identification of a new binding partner of Nod2, the human Erbin protein, a member of the leucine-rich repeat- and PDZ domain-containing (LAP) family, and assess its function during triggering of Nod2 in infection of epithelial cells with the invasive pathogen Shigella flexneri.
MATERIALS AND METHODS
Identification of proteins that interact with NOD2 by Y2H screening. Yeast two-hybrid (Y2H) screening and data analysis were performed by Hybrigenics, S.A., Paris, France.
Bait cloning. NOD2 was PCR amplified and cloned into a Y2H vector optimized by Hybrigenics. The bait construct was checked by sequencing the entire insert and was subsequently transformed into the Saccharomyces cerevisiae strain L40 GAL4 (14).
Y2H screening. A human activated leukocyte and mastocyte random-primed cDNA library, transformed into the Y187 yeast strain and containing 10 million independent fragments, was used for mating. A high mating efficiency was obtained by using a specific mating method (29a, 29b, 29c). The screen was first performed on a small scale to adapt the selective pressure to the intrinsic property of the bait. Neither toxicity nor autoactivation of the bait was observed. The full-scale screen was then performed under conditions ensuring a minimum of 50 million interactions tested in order to cover five times the primary complexity of the yeast-transformed cDNA library (37). Seventy-one million interactions were actually tested with Nod2. After selection on medium lacking leucine, tryptophan, and histidine, 39 positive clones were picked and analyzed, and the corresponding prey fragments were amplified by PCR and sequenced at their 5' and 3' junctions. Sequences were then filtered and divided into contigs as described previously (12) and compared to the GenBank database using BLASTN (3). A predicted biological score (PBS) was used to assess the reliability of each interaction, as described previously (12). Briefly, the PBS relies on two different levels of analysis, as follows. Firstly, a local score takes into account the redundancy and independency of prey fragments as well as the distributions of reading frames and stop codons in overlapping fragments. Secondly, a global score takes into account the interactions found in all the screens performed at Hybrigenics using the same library. The PBSs have been shown to positively correlate with the biological significance of interactions (37, 41).
Construction of Nod2- and Erbin-encoding plasmids. Plasmids encoding amino-terminally FLAG-tagged Nod2, deletion constructs, and the indicated domains as well as the expression plasmids for the Nod2 mutant proteins were cloned by PCR. Nod2 mRNA derived from a Nod2 cDNA expression plasmid (kindly provided by G. Nunez) was used as the template, and the PCR products were cloned using BamHI and XhoI restriction sites into the pCMV-Tag2B vector (Stratagene). The FLAG-Nod1 expression plasmid was constructed following the same strategy, using a Nod1 cDNA expression vector as the template. Myc-tagged expression plasmids for Erbin and deletion mutants thereof were a kind gift from L. Mei and P. Dai (20, 21). All constructs used were subjected to full-length sequencing (MilleGen, France).
Generation of monoclonal antibodies against Nod2. An internal peptide of Nod2 (129HPARDLQSHRPAIVRRL145) was synthesized and coupled to keyhole limpet hemocyanin or ovalbumin (PSL, Heidelberg, Germany). Rats were immunized with 50 μg peptide-keyhole limpet hemocyanin using CPG 2006 and incomplete Freund adjuvant as an adjuvant. Supernatants were tested in a differential enzyme-linked immunosorbent assay and analyzed by Western blotting using extracts from HEK 293T cells and transiently transfected HEK 293T cells expressing FLAG-Nod2. Anti-Nod2 7E11 and 4A11 of the rat immunoglobulin G2a (IgG2a) subclass were used in this study.
Cell culture, indirect immunofluorescence microscopy, and bacterial infection. HEK 293T, HT29, and HeLa cells were obtained from the ATCC. Cells were cultivated at 37°C with 5% CO2 in Dulbecco's modified Eagle's medium (Gibco-BRL) supplemented with 10% heat-inactivated fetal calf serum (Gibco-BRL) and penicillin-streptomycin (100 IU/ml and 100 mg/ml, respectively; Gibco-BRL). For indirect immunofluorescence microscopy, cells were seeded on HCl-treated coverslips, fixed in 3% paraformaldehyde in phosphate-buffered saline, and permeabilized with 0.5% Triton X-100 for 5 min. Cells were incubated in 3% bovine serum albumin in phosphate-buffered saline. Staining was done by subsequent incubation of primary and secondary antibodies in 3% bovine serum albumin. Primary antibodies were rat anti-Nod2 7E11 (1:100; this study), mouse anti-FLAG M2 (1:100) (F3165; Sigma-Aldrich), rabbit anti-Myc (1:500) (A-14; Santa Cruz Biotechnology), rabbit anti-Erbin serum (1:500) (20), and rabbit anti-S. flexneri 5a lipopeptide (1:500; kindly provided by J. Mounier and A. Phalipon, Institut Pasteur). Primary antibodies were detected with Texas Red-conjugated goat anti-mouse IgG (1:500; Sigma-Aldrich), Alexa 488-conjugated goat anti-rabbit IgG (1:500; Invitrogen Molecular Probes), and fluorescein isothiocyanate-conjugated goat anti-rat IgG (1:500; Jackson Research) secondary antibodies. DNA was stained with Hoechst 33342 (0.1 μg/ml; Invitrogen Molecular Probes).
Bacterial infection of HeLa and HEK 293T cells was performed using the S. flexneri strain M90T afaE as described previously (36). M90T afaE is a wild-type (WT) invasive strain of S. flexneri serotype 5a harboring the plasmid pIL22, which encodes the afimbrial adhesin from uropathogenic Escherichia coli (9). Briefly, bacteria were added to the cells, which were transferred to serum- and antibiotic-free medium and incubated for 10 min at room temperature prior to transfer to 37°C (time zero).
Immunoprecipitation and immunoblotting. For immunoprecipitation, HEK 293T cells were transiently transfected, using FuGene6 (Roche) according to the manufacturer's conditions, with the indicated plasmids (1 μg plasmid per 6-cm dish) and incubated for 48 h. Cells were lysed in NP-40 buffer (150 mM NaCl, 1% NP-40, 50 mM Tris-HCl, pH 7.5) containing phosphatase inhibitors (20 mM -glycerophosphate, 5 mM NaF, 100 μM Na3VO4, and Complete protease inhibitor cocktail [Roche]). Lysates were cleared for 20 min at 14,000 x g at 4°C. Immunoprecipitation was subsequently carried out for 4 h at 4°C by adding anti-FLAG beads (M2 gel; Sigma-Aldrich) or 9E10 anti-Myc antibody (sc-40; Santa Cruz Biotechnology) and protein G-Sepharose matrix (Amersham Pharmacia) to the cell extracts. The beads were precipitated by centrifugation steps and washed five times in NP-40 buffer before sodium dodecyl sulfate loading buffer was added. Typically, about 10 to 20 times more precipitate than input was loaded into the gel. Proteins were separated by Laemmli sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred by semidry Western transfer to a nitrocellulose membrane (Bio-Rad). Proteins were detected by incubation of the membrane subsequently with primary and secondary antibodies and by a final incubation with SuperSignal West Femto maximum sensitivity substrate (Pierce). Primary antibodies were mouse anti-FLAG M2 (1:1000) (F3165; Sigma-Aldrich), rabbit anti-Myc (1:1,000) (A-14; Santa Cruz Biotechnology), rabbit anti-Erbin serum (1:1,000) (20), rat anti-Nod2 4A11 and 7E11 (1:100; this study), mouse anti--tubulin (1:4,000) (T-7816; Sigma-Aldrich), and rabbit anti-IB (1:1,000) (C-21; Santa Cruz Biotechnology). Secondary antibodies were horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG (1:4,000) (170-6616; Bio-Rad), HRP-conjugated goat anti-rabbit IgG (1:4,000) (170-6515; Bio-Rad), and HRP-conjugated goat anti-rat IgG (1:4,000; Jackson Research Laboratory).
Luciferase reporter assays. Activation of NF-B was measured using a luciferase reporter assay as described previously (36). Briefly, HEK 293T cells were seeded in 24-well plates and transfected using Fugene6 (Roche), 0.25 ng Nod2 or Nod2 K305R, 75 ng Igk-luciferase NF-B reporter (31), and 25 ng -galactosidase expression plasmid per well. Cells were directly stimulated with 200 nM MDP or 0.01 μg/ml tumor necrosis factor (TNF), as indicated. After 16 h of incubation, the cells were lysed and luciferase activity was measured. The mean and standard error of the mean (SEM) were calculated from triplets, and luciferase activity was normalized as a ratio to -galactosidase activity. For short hairpin RNA (shRNA) experiments, cells were transfected with the indicated amount of an Erbin-specific shRNA plasmid (pBLOCK-it; Invitrogen), a kind gift of P. Dai and L. Mei (10), by using FuGene 72 h prior to transfection with the reporter and stimulation. Western analysis of the Erbin protein amounts after shRNA treatment was conducted with the lysates used for the luciferase activity test. For infection, S. flexneri M90T afaE was added to the transfected cells, and the medium was replaced with Dulbecco's modified Eagle's medium containing 50 μg/ml gentamicin 15 min after incubation at 37°C.
RESULTS
Identification of Erbin. In order to identify new interaction partners of the human Nod2 protein, a yeast two-hybrid screen was applied. We found that full-length Nod2 was functional in a yeast two-hybrid system by using a directed interaction test with the known binding partner RIP2 (data not shown). Since Nod2 is expressed in leukocytes (13, 18, 34), a yeast two-hybrid screen using full-length Nod2 as bait was therefore performed with a human leukocyte/mastocyte library. Interestingly, one of the interacting clones obtained encoded the CARD of RIP2, which is known to interact with Nod2 (34), thereby validating our approach. The clones with the highest probability scores for interaction among the candidates were two independent clones spanning the LRR domain of the human LAP family member Erbin (6, 7) (Fig. 1). Moreover, the expression profile reported for Erbin was closely related to the expression pattern of Nod2, with both proteins being highly expressed in CD14+ monocytes and CD33+ myeloid cells in both humans and mice (39).
The physical interaction of Nod2 and Erbin was subsequently confirmed in human cells by coimmunoprecipitation from transiently transfected HEK 293T cells. Endogenous Erbin protein was found to coprecipitate with FLAG-Nod2 (Fig. 2A). Moreover, ectopically expressed Myc-Erbin, but not the matrix alone, coprecipitated with FLAG-Nod2 (Fig. 2B). The interaction between Nod2 and Erbin was shown to be specific, as no binding of Erbin was observed with the Nod2-related protein Nod1 (Fig. 2A), nor did Nod2 bind the LAP protein Scribble, a close relative of Erbin (7; data not shown). Both a P-loop mutant of Nod2 (K305R) and a Crohn's disease-associated frameshift mutant of Nod2 (L1007fs) are known to be compromised in sensing of MDP (15, 22, 33, 34, 40). Whereas Erbin bound to WT Nod2, no significant interaction was observed with Nod2 K305R or Nod2 L1007fs (Fig. 2C).
Taken together, these findings show that Erbin forms a complex with Nod2 in human cells and that the affinity for Nod2 seems to depend on the MDP-sensing functionality of the Nod2 protein.
Mapping of interaction domains. To establish the protein domains responsible for the interaction between the two proteins, coimmunoprecipitation experiments with Erbin and Nod2 deletion constructs were conducted with lysates of transiently transfected HEK 293T cells. FLAG-Nod2 not only bound to the full-length Myc-Erbin protein but also showed binding at nearly the same affinity to a deletion protein of Erbin lacking the PDZ domain (amino acids [aa] 1 to 1279) (Fig. 2D, lanes 1 and 3). On the other hand, no interaction was observed with an amino-terminal deletion protein of Erbin containing the PDZ domain alone (aa 965 to 1371) (Fig. 2D, lane 6). However, a mutant of Erbin lacking the LRR domain (aa 391 to 1371) still showed very weak binding, and also the LRR domain (aa 1 to 391) and the central domain of Erbin (aa 391 to 965) itself bound weakly to Nod2 (Fig. 2D, lanes 2, 4 and 5).
Using the same approach with Nod2, the expected binding of full-length Erbin to Nod2 was observed, whereas no binding was observed to Nod1 or the matrix alone (Fig. 2E). Furthermore, a mutant lacking the LRR domain of Nod2 (aa 1 to 600) bound Myc-Erbin, although to a lesser extent (Fig. 2E). Importantly, no interaction was observed with a mutant lacking the two CARDs of Nod2 (aa 250 to 1040) or with the NACHT domain (aa 250 to 600) or CARD (aa 1 to 250) alone (Fig. 2E).
In conclusion, these results support the hypothesis that the domain responsible for mediating the interaction with Nod2 lies within the carboxyl terminus of the LRR domain of Erbin. On the side of Nod2, the interaction is mediated by the CARDs of Nod2; however, these domains seemed not to be sufficient to mediate the interaction.
Localization in human cells. To explore the site of complex formation between Nod2 and Erbin within the cell, the subcellular localization of the two proteins in human cells was examined. HeLa cells were transfected transiently with FLAG-tagged human Nod2 and Myc-Erbin, and the proteins were visualized using the indicated antibodies. Consistent with recent reports describing the localization of Erbin in human cells, the Erbin antibody gave a strong signal at the cell cortex and a weak signal in the cytosol (6, 29, 30) (Fig. 3A). As reported recently, we also observed Nod2 at the cell membrane and in the cytoplasm (4, 30) (Fig. 3A). Depending on the expression level of ectopically expressed Nod2, the protein was found in the cytosol and was more or less prominent at the cell membrane and in membrane ruffles, where the merged signals of Nod2 and Erbin revealed partial colocalization (Fig. 3A). Notably, about half of the transfected cells exhibited nuclear localization of the ectopically expressed Nod2 protein (Fig. 3A and B).
Erbin was initially reported to play a role in targeting the ErbB2 receptor to the membrane (6), and Erbin associates with basolateral membranes though an LRR-domain-dependent mechanism (29). Recently, a domain within the LRR of Nod2 was also reported to be essential for the membrane association of Nod2 (4). The deletion of the LRR domain in Nod2 interrupted the interaction with Erbin (Fig. 2C and E), making it tempting to speculate that Erbin might target Nod2 to the membrane. To investigate a possible involvement of Erbin in localizing Nod2 at the membrane, endogenous Erbin was silenced by shRNA. HeLa cells transiently cotransfected with FLAG-Nod2 and the Erbin shRNA plasmid showed a clear reduction in the Erbin-specific immunofluorescence signal in transfected cells (Fig. 3B, arrows). However, the membrane localization pattern of ectopically expressed Nod2 in these cells was indistinguishable from that in normal HeLa cells (Fig. 3B, arrows). Thus, Erbin seems not to be essential for the targeting of Nod2 to the membrane.
To determine the localization of endogenous Nod2, Nod2-specific monoclonal antibodies were generated. Two clones, 4A11 and 7E11, specifically recognized the ectopically expressed FLAG-Nod2 protein expressed in HEK 293T cells but gave no signal in lysates from untransfected HEK 293T cells (Fig. 3C; data not shown). Whereas 4A11 gave the strongest signal in Western blot analysis, 7E11 gave the best signal by immunofluorescence and recognized endogenous Nod2 protein in lysates from HT29 cells, which were shown to express higher levels of endogenous Nod2 (4) (Fig. 3C). Using these antibodies and a polyclonal serum specific for Erbin (20), colocalization of Nod2 and Erbin at the cell membrane in undifferentiated cells from the human colon cancer cell line HT29 was observed (Fig. 3D). Notably, the Nod2 antibody also gave signals in the nuclei of HT29, SW480, HeLa, and HEK cells (Fig. 3D; data not shown).
Taken together, these data provide evidence that a partial pool of Nod2 is localized in an Erbin-independent manner at the cell surface, where it partially colocalizes with Erbin.
Influence of Erbin on Nod2-mediated NF-B activation. Nod2 triggers the NF-B signaling pathway via activation of RIP2 (26, 34). This activation can be studied specifically in HEK cells transiently transfected with Nod2 and a NF-B reporter construct by using the Nod2-specific elicitor MDP (15, 17, 40). Since HEK 293T cells do express detectable amounts of Erbin (Fig. 2A and 4B), Nod2-mediated NF-B activation was assessed in these cells, and the influence of Erbin on this signaling pathway was tested by shRNA-mediated Erbin knockdown. As shown in Fig. 4B, the Erbin shRNA significantly reduced the amount of total Erbin protein detected by Western analysis. Quantification of NF-B reporter activity in these cells showed that a knockdown of Erbin correlated with higher reporter activity when cells were treated with MDP; however, no correlation between the Erbin protein level and basal Nod2-driven NF-B reporter activity was observed (Fig. 4A). In contrast, ectopic expression of Erbin resulted in reduced Nod2/MDP-mediated NF-B reporter activity (Fig. 4C). This influence on Nod2/MDP-mediated NF-B activation was specific, as TNF-mediated NF-B reporter activity was unaltered when increasing amounts of Erbin were expressed (Fig. 4C).
In conclusion, these results support an involvement of Erbin as a negative regulator of the MDP-mediated activation of NF-B by Nod2.
Nod2/Erbin interaction during bacterial infection. Nod2 is proposed to be the major sensor for intracellularly located, bacterially derived MDP (15, 28). Accordingly, we found that NF-B activation triggered by invasion of HEK 293T cells with the gram-negative bacterium S. flexneri was enhanced when Nod2, but not the Nod2 K305R mutant, was ectopically expressed (Fig. 5A). Since Erbin was shown to modulate NF-B activation after elicitation of Nod2, the subcellular localization of Erbin and Nod2 was analyzed during biological activation of Nod2, using the S. flexneri infection model. For this purpose, HeLa cells transiently transfected with FLAG-Nod2 were infected with the invasive S. flexneri strain M90T expressing the E. coli AfaE adhesin to mediate a higher infection efficiency. During bacterial invasion, in all infected cells, the Nod2 signal was predominant at the entry foci of the bacteria (Fig. 5B and C). At the same time, the Erbin signal was also enhanced at the site of bacterial entry, where it showed colocalization with Nod2 (Fig. 5C). S. flexneri induces membrane ruffling at the site of entry, leading to a higher surface area at the entry site (32), which underscores the observation that both Nod2 and Erbin are localized at the cell membrane.
The coincidence of the two proteins being present at the bacterial entry focus makes it tempting to speculate that this is of biological relevance in sensing bacteria through Nod2. To examine this possibility in greater detail, we asked whether colocalization at the site of bacterial entry could reflect changes in the stoichiometry of the Nod2/Erbin complex. To this end, HEK 293T cells were transiently transfected with FLAG-Nod2 and subsequently infected with S. flexneri M90T afaE. At the indicated times after infection, cells were lysed, and equal amounts were subjected to immunoprecipitation using anti-FLAG antibody. Coprecipitated endogenous Erbin protein was detected by Western analysis. Monitoring the amount of IB in the total lysates showed a degradation of IB 60 min after infection, as expected for successful bacterial invasion using this infection model (36). Importantly, a change in the amounts of coprecipitated endogenous and ectopically expressed Erbin was observed, whereas virtually equal amounts of Nod2 were precipitated at each time point of the experiment, and the Erbin signal in the total lysate did not change over time (Fig. 5D and data not shown). The amount of coprecipitated Erbin protein was lower than that in uninfected cells at early times of infection (10 to 20 min) and increased over time to reach a peak at around 30 to 40 min postinfection (Fig. 5D). To rule out the possibility that this might reflect small changes in Erbin expression after infection that are not detectable by Western analysis, the same experiment was conducted with coexpression of ectopic FLAG-Nod2 and Myc-Erbin. Again, a clear change over time in the amount of coprecipitated Myc-Erbin was observed, supporting the finding of the previous experiment (Fig. 5D).
These results show conclusively that S. flexneri triggers Nod2-mediated NF-B activation, and they support a role of Erbin in bacterially triggered Nod2 activation. Indeed, Nod2 and Erbin are present at the entry foci during S. flexneri infection of human cells, and the stoichiometry of the Nod2/Erbin complex is altered during infection.
DISCUSSION
Using a yeast two-hybrid screen with Nod2 as bait, this work identified the LAP family member Erbin as an interaction partner of human Nod2. This interaction was verified in coimmunoprecipitation experiments with human cells, which showed that the interaction domains lie within the LRR domain of Erbin and the CARDs of Nod2. In epithelial cells, the two proteins were further shown to partially colocalize at the cell membrane and at the entry foci during infection with invasive S. flexneri.
To date, only a few components of the Nod2 signaling pathway are known, with the best-studied example being the serine/threonine kinase RIP2. Studies using dominant-negative mutants and knockout mice showed that this protein is essential for Nod2-mediated NF-B signaling (26, 34). The exact mechanism of signal transduction through RIP2 is still not fully understood, but recent evidence shows that proximity activation of RIP2 and subsequent ubiquitination of IKK (also called NEMO) are crucial for downstream NF-B activation (1, 24, 34). Molecules that enhance Nod2 signaling have also been reported and include TAK1 (8) and GRIM-19 (5). Interestingly, GRIM-19 is also implicated in bacterial infection, and a role for this protein in intracellular Salmonella survival was proposed (5). On the other hand, the only negative regulation process for Nod2 was proposed by heteromerization with the NLR member Ipaf (also called CLAN) (11). The identification of a new negative regulator of Nod2 is therefore a further step towards an understanding of the signaling events mediated by Nod proteins.
Recently, Erbin was independently identified in a biochemical screen (tandem affinity purification) for new Nod2 binding proteins and reported to negatively regulate Nod2-mediated NF-B activation (30). The data reported here support this previous finding with regard to the identification of a physical interaction between Erbin and Nod2 and a function of Erbin in influencing Nod2-mediated NF-B activation. However, some minor discrepancies exist with regard to the mapping of the interaction domain. Whereas, for example, both data from the work of McDonald et al. and this study show that the PDZ domain of Erbin is not involved in mediating the interaction with Nod2, McDonald et al. reported that a carboxyl-terminal domain of Erbin (aa 965 to 1371) can bind to Nod2 (30). In the present study, however, no affinity for Nod2 was observed for this domain. This result is further strengthened by the results from the yeast two-hybrid screen, which did not yield clones for the carboxyl-terminal part of Erbin. The discrepancy in these results might be partially explainable by the use of different constructs to conduct the coimmunoprecipitation experiments. Further independent examination might help to clarify this issue.
Furthermore, our findings showed that the Nod2/Erbin interaction is dependent on the CARDs of Nod2, a finding that was also supported by the work of McDonald et al. (30). However, a dramatic decrease in binding affinity was also seen when the LRR domain of Nod2 was partially (Nod2 L1007fs) or completely deleted. Accumulated data from studies using Nod2 deletion mutants have led us to propose that this decrease was due to a loss in proper functionality of the deletion proteins. Most likely, this might be explained by improper folding or aggregate formation of these mutant Nod2 proteins.
A Nod2 membrane association was recently reported (4) and was confirmed in this study for endogenous and ectopically expressed Nod2 protein, using monoclonal antibodies directed against Nod2. Intriguingly, a nuclear signal for Nod2 was also observed. The reasons for the discrepancy in this observation with regard to recent reports on the localization of Nod2 (4, 30) are currently unclear but might reflect differences in the methodologies used or the affinities of the antibodies. The L1007 frameshift mutant of Nod2 was reported to exhibit a more cytoplasmic localization than the WT Nod2 protein when ectopically expressed in human cells (4, 30). Interestingly, this was also the case for the K305R P-loop mutant of Nod2 and a deletion mutant of Nod2 lagging the LRR domain (data not shown). Since the LRR domain of Nod2 alone was not sufficient to mediate membrane localization (data not shown), this suggests that membrane targeting of Nod2 may not be mediated by a domain in the LRR of Nod2, as proposed recently (4), but rather may depend on the functionality of the protein. This indicates that the localization of Nod2 at the cell cortex might be a secondary event dependent on the signaling integrity of the Nod2 protein and, moreover, might be influenced by activation of Nod2 by its bacterial elicitor.
Importantly, Erbin and Nod2 colocalized at the entry foci of S. flexneri. Since both proteins are recruited to the cell membrane, the higher signal intensity at this site might only correlate with the higher surface area at the entry site provoked by bacterially induced membrane ruffling (32). However, this finding illustrates that the molecules are in close proximity at the site of first peptidoglycan release from the bacteria in the cell. Nod2 was shown to be an antibacterial factor that limits the survival of intracellular bacteria when ectopically expressed (19). Therefore, even if recruitment to the entry foci might not be an active principle, it offers a biological explanation for Nod2 localization at the membrane for immediate sensing of bacterial products that might be transferred to the cytosol upon bacterial invasion.
An analysis of Nod2-mediated NF-B activation showed that ectopic expression of Erbin had a negative effect on Nod2-mediated NF-B signaling, whereas knockdown of Erbin exhibited a positive effect. These findings support the view that Erbin is a negative regulator of Nod2-mediated NF-B signaling. Furthermore, infection of cells with invasive S. flexneri triggered Nod2 signaling and led to a change in the stoichiometry of the Erbin/Nod2 complex, showing a maximal affinity of Nod2 for Erbin at 30 to 40 min postinfection. Erbin was further shown to bind only WT Nod2, but not MDP-sensing inactive mutants, such as a P-loop mutant (Nod2 K305R) and a Crohn's disease-associated frameshift mutant lacking part of the LRR domain of Nod2 (Nod2 L1007fs) (15, 34, 40). Taken together, these results suggest that activated Nod2 has a higher affinity for Erbin than does the inactive protein. Notably, Nod2 and RIP2 might also form a transient complex after elicitation of Nod2 where Nod2 forms active oligomers, presumably via homophilic NACHT/NACHT domain interactions (34). Since overexpression of Erbin had a negative impact on Nod2-mediated NF-B activation, it is tempting to speculate that this negative regulatory mechanism is exerted by inhibiting the binding of RIP2. Thus, Erbin might play an important role in dampening Nod2 signaling events in the cell. However, no significant influence of overexpression of RIP2 was observed on binding of Erbin to Nod2 (data not shown).
More importantly, this study expands previous observations regarding the involvement of Nod2 and Erbin in bacterial infection. A clear change in the dynamics of the Nod2/Erbin complex was observed after bacterial infection, even if the precise biological function of the observation remains to be established. Together with the in vivo colocalization data, which revealed the two proteins at the first contact site of bacterial invasion, and the inhibitory effect of Nod2 on the replication of invasive bacteria (19), we assume that Erbin plays a pivotal role in orchestrating Nod2 signaling during bacterial infection. Further studies will show if other proteins are involved in the regulation of Nod2 and if there might be a functional interplay between Nod2-interacting proteins, such as GRIM-19 or TAK1, and the Nod2/Erbin complex.
ACKNOWLEDGMENTS
We thank S. Girardin and J. H. Fritz for advice and critical discussions as well as all the members of the IIS for helpful comments. We are grateful to M. Jehanno for help in cloning the Nod2FS construct and M. Parmier for technical assistance. We thank L. Mei and P. Dai for sharing their expertise on Erbin and the kind gift of Erbin reagents and G. Nunez for providing the Nod2 expression plasmid.
T. A. Kufer acknowledges support by a FEBS long-term fellowship and funding from the Institut Pasteur for the yeast two-hybrid screen.
REFERENCES
1. Abbott, D. W., A. Wilkins, J. M. Asara, and L. C. Cantley. 2004. The Crohn's disease protein, NOD2, requires RIP2 in order to induce ubiquitinylation of a novel site on NEMO. Curr. Biol. 14:2217-2227.
2. Akira, S., and K. Takeda. 2004. Toll-like receptor signalling. Nat. Rev. Immunol. 4:499-511.
3. Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang, W. Miller, and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25:3389-3402.
4. Barnich, N., J. E. Aguirre, H. C. Reinecker, R. Xavier, and D. K. Podolsky. 2005. Membrane recruitment of NOD2 in intestinal epithelial cells is essential for nuclear factor-B activation in muramyl dipeptide recognition. J. Cell Biol. 170:21-26.
5. Barnich, N., T. Hisamatsu, J. E. Aguirre, R. Xavier, H. C. Reinecker, and D. K. Podolsky. 2005. GRIM-19 interacts with nucleotide oligomerization domain 2 and serves as downstream effector of anti-bacterial function in intestinal epithelial cells. J. Biol. Chem. 280:19021-19026.
6. Borg, J. P., S. Marchetto, A. Le Bivic, V. Ollendorff, F. Jaulin-Bastard, H. Saito, E. Fournier, J. Adelaide, B. Margolis, and D. Birnbaum. 2000. ERBIN: a basolateral PDZ protein that interacts with the mammalian ERBB2/HER2 receptor. Nat. Cell Biol. 2:407-414.
7. Bryant, P. J., and A. Huwe. 2000. LAP proteins: what's up with epithelia Nat. Cell Biol. 2:E141-E143.
8. Chen, C. M., Y. Gong, M. Zhang, and J. J. Chen. 2004. Reciprocal cross-talk between Nod2 and TAK1 signaling pathways. J. Biol. Chem. 279:25876-25882.
9. Clerc, P., and P. J. Sansonetti. 1987. Entry of Shigella flexneri into HeLa cells: evidence for directed phagocytosis involving actin polymerization and myosin accumulation. Infect. Immun. 55:2681-2688.
10. Dai, P., W. Xiong, and L. Mei. 2006. Erbin inhibits RAF activation by disrupting the Sur-8-Ras-Raf complex. J. Biol. Chem. 281:927-933.
11. Damiano, J. S., V. Oliveira, K. Welsh, and J. C. Reed. 2004. Heterotypic interactions among NACHT domains: implications for regulation of innate immune responses. Biochem. J. 381:213-219.
12. Formstecher, E., S. Aresta, V. Collura, A. Hamburger, A. Meil, A. Trehin, C. Reverdy, V. Betin, S. Maire, C. Brun, B. Jacq, M. Arpin, Y. Bellaiche, S. Bellusci, P. Benaroch, M. Bornens, R. Chanet, P. Chavrier, O. Delattre, V. Doye, R. Fehon, G. Faye, T. Galli, J. A. Girault, B. Goud, J. de Gunzburg, L. Johannes, M. P. Junier, V. Mirouse, A. Mukherjee, D. Papadopoulo, F. Perez, A. Plessis, C. Rosse, S. Saule, D. Stoppa-Lyonnet, A. Vincent, M. White, P. Legrain, J. Wojcik, J. Camonis, and L. Daviet. 2005. Protein interaction mapping: a Drosophila case study. Genome Res. 15:376-384.
13. Fritz, J. H., S. E. Girardin, C. Fitting, C. Werts, D. Mengin-Lecreulx, M. Caroff, J. M. Cavaillon, D. J. Philpott, and M. Adib-Conquy. 2005. Synergistic stimulation of human monocytes and dendritic cells by Toll-like receptor 4 and NOD1- and NOD2-activating agonists. Eur. J. Immunol. 35:2459-2470.
14. Fromont-Racine, M., J. C. Rain, and P. Legrain. 1997. Toward a functional analysis of the yeast genome through exhaustive two-hybrid screens. Nat. Genet. 16:277-282.
15. Girardin, S. E., I. G. Boneca, J. Viala, M. Chamaillard, A. Labigne, G. Thomas, D. J. Philpott, and P. J. Sansonetti. 2003. Nod2 is a general sensor of peptidoglycan through muramyl dipeptide (MDP) detection. J. Biol. Chem. 278:8869-8872.
16. Girardin, S. E., P. J. Sansonetti, and D. J. Philpott. 2002. Intracellular vs extracellular recognition of pathogens—common concepts in mammals and flies. Trends Microbiol. 10:193-199.
17. Girardin, S. E., L. H. Travassos, M. Herve, D. Blanot, I. G. Boneca, D. J. Philpott, P. J. Sansonetti, and D. Mengin-Lecreulx. 2003. Peptidoglycan molecular requirements allowing detection by Nod1 and Nod2. J. Biol. Chem. 278:41702-41708.
18. Gutierrez, O., C. Pipaon, N. Inohara, A. Fontalba, Y. Ogura, F. Prosper, G. Nunez, and J. L. Fernandez-Luna. 2002. Induction of Nod2 in myelomonocytic and intestinal epithelial cells via nuclear factor-kappa B activation. J. Biol. Chem. 277:41701-41705.
19. Hisamatsu, T., M. Suzuki, H. C. Reinecker, W. J. Nadeau, B. A. McCormick, and D. K. Podolsky. 2003. CARD15/NOD2 functions as an antibacterial factor in human intestinal epithelial cells. Gastroenterology 124:993-1000.
20. Huang, Y. Z., Q. Wang, W. C. Xiong, and L. Mei. 2001. Erbin is a protein concentrated at postsynaptic membranes that interacts with PSD-95. J. Biol. Chem. 276:19318-19326.
21. Huang, Y. Z., M. Zang, W. C. Xiong, Z. Luo, and L. Mei. 2003. Erbin suppresses the MAP kinase pathway. J. Biol. Chem. 278:1108-1114.
22. Hugot, J. P., M. Chamaillard, H. Zouali, S. Lesage, J. P. Cezard, J. Belaiche, S. Almer, C. Tysk, C. A. O'Morain, M. Gassull, V. Binder, Y. Finkel, A. Cortot, R. Modigliani, P. Laurent-Puig, C. Gower-Rousseau, J. Macry, J. F. Colombel, M. Sahbatou, and G. Thomas. 2001. Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn's disease. Nature 411:599-603.
23. Inohara, N., M. Chamaillard, C. McDonald, and G. Nunez. 2005. NOD-LRR proteins: role in host-microbial interactions and inflammatory disease. Annu. Rev. Biochem. 74:355-383.
24. Inohara, N., T. Koseki, J. Lin, L. del Peso, P. C. Lucas, F. F. Chen, Y. Ogura, and G. Nunez. 2000. An induced proximity model for NF-kappa B activation in the Nod1/RICK and RIP signaling pathways. J. Biol. Chem. 275:27823-27831.
25. Janeway, C. A., Jr., and R. Medzhitov. 2002. Innate immune recognition. Annu. Rev. Immunol. 20:197-216.
26. Kobayashi, K., N. Inohara, L. D. Hernandez, J. E. Galan, G. Nunez, C. A. Janeway, R. Medzhitov, and R. A. Flavell. 2002. RICK/Rip2/CARDIAK mediates signalling for receptors of the innate and adaptive immune systems. Nature 416:194-199.
27. Kobayashi, K. S., M. Chamaillard, Y. Ogura, O. Henegariu, N. Inohara, G. Nunez, and R. A. Flavell. 2005. Nod2-dependent regulation of innate and adaptive immunity in the intestinal tract. Science 307:731-734.
28. Kufer, T. A., J. H. Fritz, and D. J. Philpott. 2005. NACHT-LRR proteins (NLRs) in bacterial infection and immunity. Trends Microbiol. 13:381-388.
29. Legouis, R., F. Jaulin-Bastard, S. Schott, C. Navarro, J. P. Borg, and M. Labouesse. 2003. Basolateral targeting by leucine-rich repeat domains in epithelial cells. EMBO Rep. 4:1096-1102.
29a. Legrain, P., et al. 1998. U.S. patent 6187535.
29b. Legrain, P., et al. 2000. U.S. patent 6531284.
29c. Legrain, P., et al. 2002. U.S. patent 6913886.
30. McDonald, C., F. F. Chen, V. Ollendorff, Y. Ogura, S. Marchetto, P. Lecine, J. P. Borg, and G. Nunez. 2005. A role for Erbin in the regulation of Nod2-dependent NF-kappaB signaling. J. Biol. Chem. 280:40301-40309.
31. Munoz, E., G. Courtois, P. Veschambre, P. Jalinot, and A. Israel. 1994. Tax induces nuclear translocation of NF-B through dissociation of cytoplasmic complexes containing p105 or p100 but does not induce degradation of IB/MAD3. J. Virol. 68:8035-8044.
32. Nhieu, G. T., and P. J. Sansonetti. 1999. Mechanism of Shigella entry into epithelial cells. Curr. Opin. Microbiol. 2:51-55.
33. Ogura, Y., D. K. Bonen, N. Inohara, D. L. Nicolae, F. F. Chen, R. Ramos, H. Britton, T. Moran, R. Karaliuskas, R. H. Duerr, J. P. Achkar, S. R. Brant, T. M. Bayless, B. S. Kirschner, S. B. Hanauer, G. Nunez, and J. H. Cho. 2001. A frameshift mutation in NOD2 associated with susceptibility to Crohn's disease. Nature 411:603-606.
34. Ogura, Y., N. Inohara, A. Benito, F. F. Chen, S. Yamaoka, and G. Nunez. 2001. Nod2, a Nod1/Apaf-1 family member that is restricted to monocytes and activates NF-kappaB. J. Biol. Chem. 276:4812-4818.
35. Opitz, B., A. Puschel, B. Schmeck, A. C. Hocke, S. Rosseau, S. Hammerschmidt, R. R. Schumann, N. Suttorp, and S. Hippenstiel. 2004. Nucleotide-binding oligomerization domain proteins are innate immune receptors for internalized Streptococcus pneumoniae. J. Biol. Chem. 279:36426-36432.
36. Philpott, D. J., S. Yamaoka, A. Israel, and P. J. Sansonetti. 2000. Invasive Shigella flexneri activates NF-kappa B through a lipopolysaccharide-dependent innate intracellular response and leads to IL-8 expression in epithelial cells. J. Immunol. 165:903-914.
37. Rain, J. C., L. Selig, H. De Reuse, V. Battaglia, C. Reverdy, S. Simon, G. Lenzen, F. Petel, J. Wojcik, V. Schachter, Y. Chemama, A. Labigne, and P. Legrain. 2001. The protein-protein interaction map of Helicobacter pylori. Nature 409:211-215.
38. Schreiber, S., P. Rosenstiel, M. Albrecht, J. Hampe, and M. Krawczak. 2005. Genetics of Crohn disease, an archetypal inflammatory barrier disease. Nat. Rev. Genet. 6:376-388.
39. Su, A. I., M. P. Cooke, K. A. Ching, Y. Hakak, J. R. Walker, T. Wiltshire, A. P. Orth, R. G. Vega, L. M. Sapinoso, A. Moqrich, A. Patapoutian, G. M. Hampton, P. G. Schultz, and J. B. Hogenesch. 2002. Large-scale analysis of the human and mouse transcriptomes. Proc. Natl. Acad. Sci. USA 99:4465-4470.
40. Tanabe, T., M. Chamaillard, Y. Ogura, L. Zhu, S. Qiu, J. Masumoto, P. Ghosh, A. Moran, M. M. Predergast, G. Tromp, C. J. Williams, N. Inohara, and G. Nunez. 2004. Regulatory regions and critical residues of NOD2 involved in muramyl dipeptide recognition. EMBO J. 23:1587-1597.
41. Wojcik, J., I. G. Boneca, and P. Legrain. 2002. Prediction, assessment and validation of protein interaction maps in bacteria. J. Mol. Biol. 323:763-770.(T. A. Kufer, E. Kremmer, )