GRIM-19 Interacts with Nucleotide Oligomerization Domain 2 and Serves as Downstream Effector of Anti-bacterial Function in Intestinal Epithelial Cells*

Nucleotide oligomerization domain 2 (NOD2) functions as a mammalian cytosolic pathogen recognition molecule, and variants have been associated with risk for Crohn disease. We recently demonstrated that NOD2 functions as an anti-bacterial factor limiting survival of intracellular invasive bacteria. To gain further insight into the mechanism of NOD2 activation and signal transduction, we performed yeast two-hybrid screening. We demonstrate that GRIM-19, a protein with homology to the NADPH dehydrogenase complex, interacts with endogenous NOD2 in HT29 cells. GRIM-19 is required for NF-κB activation following NOD2-mediated recognition of bacterial muramyl dipeptide. GRIM-19 also controls pathogen invasion of intestinal epithelial cells. GRIM-19 expression is decreased in inflamed mucosa of patients with inflammatory bowel diseases. GRIM-19 may be a key component in NOD2-mediated innate mucosal responses and serve to regulate intestinal epithelial cell responses to microbes.

Crohn disease (CD) 1 is an inflammatory bowel disease (IBD) characterized by chronic inflammation of the intestine. The pathogenesis of CD is complex and appears to consist of three interacting elements: genetic susceptibility factors, priming by the enteric microflora, and immune-mediated tissue injury (1)(2)(3). Innate immunity relies on the expression of a restricted variety of receptors, including Toll-like receptors (a family of trans-membrane protein) and NODs (a family of intracellular bacterial sensors), able to recognize highly conserved microbial motifs (4). CARD15/NOD2 gene, located on chromosome 16q12, has been identified as the first susceptibility gene for CD (5,6). The role of the two N-terminal CARD domains, a nucleotide-binding domain, and multiple C-terminal leucine-rich repeat regions of NOD2 remains to be established (7).
NOD2 recognizes muramyl dipeptide (MDP-LD) and subsequently activates NF-B through an incompletely understood pathway involving RIP2/RICK and members of the Toll-like receptor-sensing cascade (8 -13). Recent studies leave unclear whether NOD2 is a regulator of TLR2 responses that may be disregulated when NOD2 3020insC mutant associated with CD is present (14 -16). A direct interaction between NOD2 and transforming growth factor-␤-activated kinase 1 (TAK1) was recently shown, and TAK1 regulates NOD2-mediated NF-B activation (17). In addition, NF-B activation induced by Streptococcus pneumoniae depends on NOD2 (18). A recent study of NOD2-deficient mice shows loss of protective immunity in response to bacterial muramyl dipeptide (15). In addition, these mice are susceptible to listeria infection via the oral route. We have previously shown that the 3020insC mutant of CARD15/ NOD2 has impaired function as a defensive factor against intracellular bacteria in intestinal epithelial cells (IEC) (19). We demonstrate here that GRIM-19 is an interactor necessary for NF-B and anti-bacterial effects of NOD2.
Yeast Two-hybrid Screening-Yeast two-hybrid screening was performed by MATHCMAKER GAL4 Two-Hybrid System 3 according to the manufacturer's protocol (BD Biosciences Clontech, Palo Alto, CA). Briefly, pGBKT7-NOD2 was generated by PCR methods from pCMV-FLAG-NOD2 vector (19). pGBKT7-NOD2 was transfected into the AH109 yeast strain. Expression of Myc-tagged NOD2 protein in yeast extract was confirmed by Western blot analysis using anti-Myc monoclonal antibody (Covance, Richmond, CA) and affinity-purified anti-NOD2 anti-sera (19). Screening was performed using a bone marrow pretransformed library (BD Biosciences Clontech) according to the manufacturer's protocol. Co-transformants were selected in S.D. medium lacking histidine, leucine, and tryptophan. Yeast ␣-galactosidase activity, expressed from the MEL1 gene in response to GAL4 activation, was determined in plates containing X-␣-gal (BD Biosciences Clontech).
Immunoprecipitation and Immunoblotting Experiments-Cells were grown on 6-well plates. Medium was removed, and 300 l of 1% Triton lysis buffer (1.25% sodium dodecyl sulfate, 2.5% glycerol, 62.5 mmol/ liter Tris/HCl, pH 6.8, 5% 2-mercaptoethanol) supplemented with protease inhibitor mixture (Complete Mini; Roche Applied Science) was added. Supernatant of cell lysate was obtained by centrifugation at 12,000 ϫ g for 10 min. Protein concentration was determined using the DC protein assay kit (Bio-Rad). Proteins were blotted onto polyvinylidene difluoride membranes and stained for FLAG-NOD2 using anti-FLAG monoclonal antibody (Sigma), for Xpress-GRIM-19 using anti-Xpress monoclonal antibody (Invitrogen), and for NOD1-HA using anti-HA monoclonal antibody (Roche Applied Science). Two milligrams of cell lysate were immunoprecipitated with 2 g of anti-FLAG, anti-Xpress, or anti-HA monoclonal antibodies and 100 l of Hiptrap protein A/G-Sepharose beads. After overnight incubation at 4°C, immunoprecipitated proteins were separated on 4 -12% or on 4 -20% Tris-glycine gel (Invitrogen). Proteins were blotted onto polyvinylidene difluoride membranes and stained using specific antibodies. For endogenous binding, 500 g of total protein from HT29 cells were subjected to immunoprecipitation/blotting as described above using rabbit anti-NOD2 antiserum HM2559 (19) and affinity-purified rabbit antiserum against human GRIM-19 produced by Affinity Bioreagents (Project A203001, immunizing peptide IMKDVPDWKVGESVF).
Confocal Microscopy-Caco-2 cells were grown on sterile permanox coverslips for 24 h and then transfected with pCDNA4/HisMAX-GRIM-19 and pCMVFLAG-NOD2. After 48 h, cells were washed twice with ice-cold PBS, fixed for 20 min with cold methanol at Ϫ20°C, and washed three times with ice-cold PBS. Cells were saturated for 30 min with PBS containing 5% donkey serum. Cells were then incubated for 2 h with primary antibody (mouse monoclonal anti-Xpress antibody and/or rabbit polyclonal anti-FLAG antibody (Sigma). Immunostaining was performed, respectively, with Texas-Red-conjugated anti-mouse IgG or with fluorescein isothiocyanate-conjugated anti-rabbit IgG (Vector Laboratories) secondary antibodies. Salmonella typhimurium was detected with rabbit antibody to Salmonella fluorescein conjugate (Biodesign International, Saco, ME). The coverslips were mounted in Vectashield (Vector Laboratories) and examined with a confocal laser scanning microscope.
Bacterial Invasion Assays-Invasion assays were performed with Salmonella enterica serovar typhimurium or Escherichia coli TOP10 (Invitrogen). Monolayers were seeded in 24-well tissue culture plate with 10 5 cells/well and incubated for 20 h. Monolayers were then infected in 1 ml of cell culture medium without antibiotic and with heat-inactivated fetal calf serum at a multiplicity of infection (MOI) of 10 bacteria/epithelial cell. After a 2-h incubation period at 37°C, monolayers were washed two times with PBS. Fresh cell culture medium containing 100 g/ml gentamicin (Sigma) was added for 1 h to kill extracellular bacteria. The epithelial cells were then lysed with 1% Triton X-100 in deionized water. Samples were diluted and plated onto Luria-Bertani agar plates to determine the number of colony-forming units corresponding to the number of intracellular bacteria.
Reverse Transcription-Polymerase Chain Reaction-Total RNA of IEC lines were extracted using Trizol (Invitrogen) following the manufacturer's protocols. For reverse transcription, 2 g of total RNA were transcribed by the SuperScript First-strand Synthesis system (Invitrogen). Real-time RT-PCR was performed in an ABI Prism 7000 sequence detector using the SYBR Green JumpStart TM detection system. Briefly, 50 ng of the reversed transcribed cDNA were used for each PCR reaction with 200 nM of forward and reverse primers. Primers for PCR were forward 5-accggaagtgtgggatactg-3, reverse 5-gctcacggttccacttcatt-3 (GRIM-19, 194 bp); forward 5-tcatctctgccccctctgct-3, reverse 5-cgacgcctgcttcaccacct-3 (glyceraldehyde-3-phosphate dehydrogenase (GAPDH), 440 bp). The PCR program used was 50°C for 2 min, then 95°C for 10 min followed by 40 cycles of 95°C 15 s, 60°C 15 s, and 72°C 15 s. The threshold cycle (C T ) values were obtained for the reactions reflecting quantity of the template in the sample. GRIM-19 Delta C T (⌬C T ) was calculated by subtracting the GAPDH C T value from the GRIM-19 C T value and thus represented the relative quantity of the target molecule after normalizing with the internal standard GAPDH. The GRIM-19 ⌬C T values of Caco-2 cells infected with Salmonella or transfected with pCDNA4/HisMAX-GRIM19 were expressed as percentage of GRIM-19 ⌬C T values of control Caco-2 cells.
PCR products were sequenced using an ABI 3700 PRISM (PerkinElmer Life Sciences) automated sequencer. Sequences were analyzed using NCBI BLAST software.
Cytotoxicity Assays-To evaluate the effect of overexpression of GRIM-19 and NOD2 in Caco-2 cells, we measured the release of adenylate kinase from damage cells using the ToxiLight non-destructive cytotoxicity assay according to the manufacturer's instructions (Cambrex Bio Science, Rockland, ME). As a positive control, we infected Caco-2 cells with S. typhimurium at an MOI of 50 during 2 h.
NF-B Activation Assays-HEK293 cells were transfected overnight with 1 ng of NOD2, 10 ng of grim-19 siRNA plus 1 ng of pIV luciferase reporter plasmid and Renilla plasmid. At the same time, 1 g of MDP-LD (Sigma) was added. After 24 h, luciferase activity was measured using the dual-luciferase reporter assay system (Promega, Madison, WI) according to the manufacturer's instructions and normalized relative to Renilla activity.
Statistical Analysis-For analysis of the significance of differences in invasion, NF-B, or mRNA levels, Student's t test was used for comparison of two groups of data. All experiments were repeated at least three times. A p value equal to or less than 0.05 was considered to be statistically significant.

GRIM-19 Is a New Binding
Partner of NOD2-A yeast twohybrid screen was performed to identify cellular proteins that interact with NOD2. The N-terminal deletion construct, NOD2 without the CARD15 domain, was used as bait (Fig. 1A). We screened a human bone marrow cDNA library expressing proteins fused to the AD transcriptional activation domain. One positive clone was identified to encode the human GRIM-19 protein, a novel cell death-related gene (21). Co-expression of NOD2 and GRIM-19 in yeast survival assays in S.D./ϪAde/ ϪHis/ϪLeu/ϪTrp/X-gal-selective medium confirmed a strong interaction between these two proteins (data not shown).
Association of NOD2 and GRIM-19 -To confirm the interaction of NOD2 and GRIM-19 in mammalian cells, we transfected COS7 or HEK293 cells with FLAG-tagged NOD2 and Xpresstagged GRIM-19. As shown in Fig. 1B, GRIM-19 was detected in anti-FLAG immunoprecipitates from NOD2 co-transfectants, but not from cells co-transfected with the control plasmid (Fig. 1B). A reciprocal immunoprecipitation/blotting experiment using anti-Xpress monoclonal antibody also showed NOD2 co-precipitating with GRIM-19 (Fig. 1B). To study the possible physiological significance of NOD2/GRIM-19 interaction, we investigated endogenous interaction between NOD2 and GRIM-19. For this, we used the HM2559 rabbit antiserum against NOD2 previously generated (19) and a newly produced rabbit antiserum against GRIM-19. Anti-GRIM-19 antibody specifically recognized Xpress-tagged GRIM-19 overexpressed in COS7 or HT29 cells (data not shown). The rabbit antiserum against NOD2 HM2559 showed a good endogenous NOD2 expression in HT29 cells, whereas COS7 and HEK293 cells expressed low amounts of endogenous NOD2 (data not shown). We expressed GRIM-19 in HT29 cells and showed after immunoprecipitation using anti-GRIM-19 antiserum an association with endogenous NOD2 (Fig. 1C). Similarly, immunoprecipitation of HT29 lysate overexpressing NOD2 using anti-GRIM-19 antiserum showed increased NOD2 binding compared with lysate from non-transfected HT29 cells (data not shown). The interaction between GRIM-19 and CARD4/NOD1 was also examined. In contrast to NOD2, no association was observed between NOD1 and GRIM-19 by immunoprecipitation (Fig.  1D). These data suggest a specific functional link between GRIM-19 and NOD2.
GRIM-19 and NOD2 Colocalize in Caco-2 and COS7 Cells-To determine in which cellular compartment NOD2 and GRIM-19 interact, we examined their subcellular localization by immunofluorescence confocal microscopy. COS7 and Caco-2 cells were transfected with Xpress-GRIM-19 and GFP-NOD2 expression plasmids (Fig. 2). The NOD2 protein was observed throughout the cytoplasm and also near the plasma membrane. In COS7 and Caco-2 cells co-expressing GFP-NOD2 and Xpress-GRIM-19, GRIM-19 partially colocalized with NOD2 in intracellular vesicles but not near the membrane (Fig. 2).

GRIM-19 Is Expressed in IBD Tissues and Intestinal
Epithelial Cell Lines-Expression of GRIM-19 was also determined in colonic biopsies from Crohn disease and ulcerative colitis patients taken from both involved and non-involved areas and compared with expression in mucosal biopsies obtained from controls without IBD. In the non-involved mucosa from IBD patients, grim-19 mRNA expression was comparable with that in control patients. In contrast, grim-19 mRNA expression was significantly decreased in involved areas from mucosa of both ulcerative colitis and CD patients (Fig. 3A). Expression of grim-19 mRNA was also assessed by RT-PCR in several human intestinal epithelial cell lines, the THP-1 macrophage cell line, and Jurkat cells. GRIM-19 was expressed in THP-1, Jurkat cell lines, and in all the IEC lines used in this study (Fig. 3B).
The effect of bacterial invasion on GRIM-19 expression was evaluated. Caco-2 cells were infected with invasive S. typhimurium and non-pathogenic and non-invasive E. coli for 2 h at an MOI 100. Invasive ability of S. typhimurium was verified using gentamicin protection assay (data not shown). S. typhimurium infection up-regulated grim-19 mRNA expression in Caco-2 cells (2.26-fold), whereas infection with a non-pathogenic and non-invasive E. coli had no effect on grim-19 mRNA expression (Fig. 4).
Functional Role of GRIM-19 in Caco-2 Cells-To determine the effect of GRIM-19 on cell death, we assessed a non-destructive bioluminescence cytotoxicity assay on Caco-2 cells. Cells were transfected with Xpress-tagged GRIM-19 or FLAG-tagged NOD2 or infected with S. typhimurium. Overexpression of GRIM-19 or NOD2 did not induce cell death under the conditions of this study, whereas infection by S. typhimurium induced cell death (Fig. 5A). To determine whether GRIM-19 protein results in an alteration of the functional outcome of bacterial survival, untransfected and transiently transfected Caco-2 cells were infected with S. typhimurium. After a 2-h incubation period with the monolayer, the percentage of intracellular bacteria in comparison with the inoculum was significantly decreased (72.0 Ϯ 5.4%) in Caco-2 cells expressing GRIM-19 in comparison with non-transfected cells or cells transfected with vector alone (Fig. 5B). Consistent with these findings, S. typhimurium invasion increased (162.0 Ϯ 43.8%) in Caco-2 cells harboring a plasmid encoding the grim-19 siRNA-1 (Fig. 5B), which significantly decreased grim-19 mRNA levels (data not shown), whereas the control grim-19 siRNA (sequence 2), which has no effect on grim-19 mRNA, did not modify S. typhimurium invasive ability. To confirm these results, immunostaining was performed with Caco-2 cells transfected with Xpress-tagged GRIM-19 and infected with S. typhimurium. As shown in Fig. 5C, few bacteria were found in Caco-2 cells expressing GRIM-19, whereas numerous bacteria were observed in adjacent cells that did not express GRIM-19 (Fig. 5C). Consistent with these results, the mean number of intracellular bacteria found in Caco-2 cells expressing GRIM-19 after immunostaining examination by confocal microscopy is significantly lower (7.5 bacteria/cell) (p Ͻ 0.001) than that of untransfected cells (14.8 bacteria/cells) (Fig. 5D). Thus, GRIM-19 protected host cells by reducing the intracellular survival of S. typhimurium.
Retinoic Acid and IFN-␤ Exert Anti-bacterial Activity by Inducing GRIM-19 Expression-To investigate whether endoge- The precipitates were fractionated through 4 -12% or 4 -20% Trisglycine SDS-PAGE and blotted with anti-FLAG or anti-Xpress monoclonal antibodies. Total cell lysates (TCL) were subjected to Western blot analysis with anti-Xpress antibody or anti-FLAG antibody to detect the expression of GRIM-19 or NOD2 in transfected COS7 cells. C, HT29 cells were transfected with Xpress-tagged GRIM-19. The cell lysates were immunoprecipitated with anti-GRIM-19 antibody. Immunoprecipitates were subjected to Western blot analysis using rabbit antiserum against human GRIM-19 or rabbit anti-serum HM2559 against human NOD2. D, COS7 cells were transfected with CARD4/NOD1-HAtagged and/or Xpress-tagged GRIM-19. After immunoprecipitation with anti-HA or anti-Xpress monoclonal antibodies, immunoprecipitates were subjected to Western blot analysis using anti-HA or anti-Xpress monoclonal antibodies.
nous GRIM-19 directly exerts anti-microbial activity, we induced endogenous GRIM-19 expression by stimulating Caco-2 cells with a combination of retinoic acid (RA) and interferon-␤ (INF-␤). After stimulation with RA and IFN-␤, real-time RT-PCR showed that grim-19 mRNA expression was significantly increased (Fig. 6), whereas Caco-2 cells stimulated with either RA or IFN-␤ alone did not induce GRIM-19 expression. Subsequently the ability of S. typhimurium to invade Caco-2 cells that were unstimulated or stimulated with RA or INF-␤, or both, was evaluated. After a 2-h incubation period with the epithelial monolayer, the percentage of intracellular bacteria was significantly decreased in Caco-2 cells stimulated with both RA and INF-␤ (Fig. 6), whereas the invasive ability of S. typhimurium remained the same in Caco-2 cells stimulated only with either RA or INF-␤. Increased grim-19 expression therefore correlated with a decrease in the number of recovered viable S. typhimurium in bacterial invasion assays. Finally, grim-19 siRNA-1 restored invasive ability of S. typhimurium in stimulated Caco-2 cells in comparison with non-stimulated Caco-2 cells (Fig. 6), indicating that retinoic acid and INF-␤ exert anti-bacterial activity by inducing endogenous GRIM-19 expression.
GRIM-19 Is Required for NOD2-mediated NF-B Activation following MDP-LD Stimulation-HEK293 cells transfected with 1 ng of NOD2 were stimulated with 1 g of MDP-LD and transfected with siRNA specific for grim-19. Transfection with grim-19 siRNA-1 significantly decreased grim-19 mRNA level and inhibited the MDP-LD-driven response to NOD2. NF-B activation in HEK293 transfected with NOD2 and grim-19 siRNA-1 after MDP-LD stimulation was only 50% of that observed in HEK293 transfected only with NOD2 or with pSUPER control vector (Fig. 7A). Control grim-19 siRNA, which did not knock down grim-19 mRNA, had no significant effect on NF-B activation via NOD2 after MDP-LD stimulation.
The anti-bacterial activity of NOD2 was also dependent on the presence of GRIM-19. The invasive ability of S. typhimurium decreased in HEK293 cells overexpressing NOD2 compared with non-transfected HEK293 cells (Fig. 7B). This effect was reversed in the presence of grim-19 siRNA-1 (Fig. 7B), suggesting that anti-bacterial activity conferred by NOD2 correlates with NF-B activation. DISCUSSION We have previously shown that NOD2, but not the NOD2 mutant 3020insC associated with CD, protects intestinal epithelial cells against Salmonella infection (19). In the present study, we performed yeast two-hybrid screening and identified GRIM-19 as an interacting protein with NOD2 in mammalian cells. GRIM-19, gene associated with retinoid-IFN-induced mortality 19, is located on chromosome 19 and induces cell death in a number of tumor cell lines. GRIM-19 protein expression is induced by the combination of IFN-␤ and all-transretinoic acid RA (21,22). The subcellular location of GRIM-19 action remains to be established. Originally GRIM-19 was observed in the nucleus (21) and more recently in both nucleus and cytoplasm (23). Its nuclear, but also cytoplasmic, distribution and punctate staining patterns observed in cells prompted speculation that GRIM-19 might interact with various protein or protein complexes to regulate cellular responses (21,23). GRIM-19 is also a subunit of the mitochondrial NADPH: ubiquinone oxidoreductase (respiratory complex I) (24) and colocalized with mitochondria in MCF-7 and COS1 cells (25). Recently, GRIM-19 was detected in the native form in mitochondrial complex I. Homologous deletion of GRIM-19 in mice causes embryonic lethality at embryonic day 9.5 (26). In the present study, we showed a cytoplasmic colocalization of GRIM-19 and NOD2. Furthermore, interaction between GRIM-19 and NOD2 is NOD2-specific; no binding was observed with NOD1, another NOD protein family member.
Present and previous studies indicate that GRIM-19 binds additional proteins that play a crucial role in IBD, including NOD2, Stat3, and GW112. GRIM-19 binds Stat3 in various cell types but did not bind other Stat proteins, such as Stat1 or Stat5a (25), and the interaction between GRIM-19 and Stat3 suppresses Stat3 activity. Stat3 has a critical role in the development and regulation of innate immunity, and deletion of Stat3 during hematopoiesis causes Crohn disease-like pathogenesis and lethality in mice (27). GRIM-19 has also been reported to bind GW112, a protein expressed in various human normal and malignant tissues with higher expression in organs/tumors of the digestive system. GW112 plays an antiapoptotic role that promotes tumor growth, and GW112 could be involved in the regulation of cellular apoptosis under inflammatory conditions in the digestive system (28).
We provide evidence here that Salmonella infection increases grim-19 mRNA in infected Caco-2 cells, whereas expression remains unchanged in Caco-2 cells infected with noninvasive E. coli. Epithelial cells of the human intestinal mucosa are the initial site of host invasion by bacterial enteric pathogens. Human colonic epithelial cells were shown to undergo apoptosis following infection with different invasive bacteria, such as enteroinvasive E. coli or Salmonella. Apoptosis in response to bacterial infection may function to delete infected and damaged epithelial cells and restore epithelial cell growth regulation and epithelial integrity (29). It has previously been shown that after invasion of intestinal macrophages, virulence proteins secreted by Salmonella specifically induce apoptotic cell death by activating the cysteine protease caspase-1 (30).
GRIM-19 has been shown to interact with multiple proteins such as mitochondrial NADH:ubiquinone oxidoreductase and to have several biological activities, including cell growth, transcription, and cell death (21). Here, we show that under the conditions of this study, GRIM-19 did not induce cell death. Expression of GRIM-19 in Caco-2 cells decreased the invasive ability of Salmonella, revealing its protective role in IEC. Given that the overall transfection efficiency in the Caco-2 cells was ϳ30 -35%, the 28% reduction in colony-forming units suggests that GRIM-19 is highly effective in controlling intracellular survival in those cells expressing the transfected protein. Down-regulation of GRIM-19 expression using siRNA increased the invasive ability of S. typhimurium. Consistent with these findings, the invasive ability of S. typhimurium also decreased in Caco-2 cells expressing increased endogenous GRIM-19 induced by the combination of IFN␤/ RA. In addition, siRNA against GRIM-19 restored invasive activity of S. typhimurium in Caco-2 cells stimulated with IFN␤/RA, confirming that endogenous GRIM-19 protects cells against bacteria. We also show that expression of GRIM-19 is decreased in IBD tissues from both ulcerative colitis and CD patients in affected areas, compared with uninvolved tissue. This decrease could be due to the loss of epithelium in the involved area or to the down-regulation of GRIM-19 expression in inflamed areas in IBD patients. Given the fact that we demonstrate a protective role for GRIM-19, decreased GRIM-19 expression in involved colonic areas in IBD patients could induce a higher susceptibility of commensal and/or pathogenic bacteria to invade and/or survive in involved areas of the epithelium.
The question remains how does GRIM-19 function in the innate immune mucosal response? It appears from our data Invasive ability of S. typhimurium in Caco-2 cells stimulated with a combination of retinoic acid (RA) and interferon-␤ (IFN-␤) to induce endogenous GRIM-19 expression was determined as described in Fig. 5. grim-19 mRNA levels were determined by RT-PCR as described in Fig. 3. *, p Ͻ 0.05. that GRIM-19 acts as a key component of the innate immune mucosal response by modulating NF-B activation via NOD2. Decreased expression of GRIM-19 using specific siRNA in HEK293 overexpressing NOD2 decreased NF-B activation in response to MDP-LD. Importantly, NF-B activation correlates with a decrease in the number of viable intracellular S. typhimurium, indicating that NOD2 anti-bacterial activity is dependent on NF-B activation.
In addition to its effects on NOD2-mediated NF-B activation, it is possible that GRIM-19 has other functions that contribute to its effects on epithelial response to invasive bacteria. GRIM-19 is a subunit of the mitochondrial NADPH:ubiquinone oxidoreductase (24 -26). The NADPH enzyme complex catalyzes the transfer of electrons from NADPH to molecular oxygen, generating reactive oxygen species (ROS), in particular superoxide anion. Reactive oxygen species operate on a variety of physiological processes, including host defense against pathogens (31). The best known O 2 . -producing enzyme is the phagocyte-associated respiratory enzyme NADPH oxidase burst that plays a crucial role in the process of killing microorganisms (31). Helicobacter pylori lipopolysaccharide stimulated Toll-like-receptor 4 signaling and activated the NADPH oxidase 1 (32,33). In addition, a yeast two-hybrid screen led to the observation of a direct interaction of TLR4 with NADPH oxidase 4 that mediates lipopolysaccharide-induced reactive oxygen species generation and NF-B activation (34). It is possible that GRIM-19 increases the production of reactive oxygen species in intestinal epithelial cells after bacterial invasion, protecting the intestinal mucosa against pathogens.
GRIM-19 could support mucosal defense by mediating NOD2 function in the recognition of bacterial pathogens and their elimination in intestinal epithelial cells.