Altered host:pathogen interactions conferred by the Blau syndrome mutation of NOD2
- First Online:
- Cite this article as:
- Kim, T., Payne, U., Zhang, X. et al. Rheumatol Int (2007) 27: 257. doi:10.1007/s00296-006-0250-0
- 61 Views
Blau syndrome (BS) is a rare familial granulomatous disease manifested by uveitis, arthritis and skin rash. BS has recently been found to be associated with a distinctive mutation in NOD2, which encodes an intracellular toll-like receptor. We have compared host cell interaction with bacterial challenge in U937 cells expressing wild type human NOD2 (NOD2wt), mutant NOD2 (NOD2Blau), or a vector control (VC). The cells were incubated with Salmonella typhimurium. Intracellular uptake was assessed by harvesting the cells at different time points following invasion and quantitating the CFU, recovered after gentamicin treatment to kill extracellular organisms. Expression of TNF-α, TLR2 and TLR4 was determined by semi-quantitative RT-PCR under resting conditions and after stimulation by bacteria. Invasion of target cells with S. typhimurium was diminished in the presence of NOD2Blau. Expression of TNF-α mRNA was enhanced following bacterial invasion in all cell lines but NOD2Blau was associated with a more rapid decline in TNF-α expression. Kinetics of intracellular clearance of bacteria indicated a relative defect in NOD2Blau compared to controls. This clearance defect may be related to the lack of sustained TNF-α seen in the early stages. These events were not related to differential TLR2 or TLR4 expression since there were no significant differences seen between the cell lines after bacterial stimulation. Our findings indicate that the NOD2 mutation associated with this syndrome alters host:microbial interaction, and this may have relevance to triggering factors in the ocular and joint inflammation seen in BS.
NOD2, also known as CARD15, is a member of the CED4/APAF1 family of apoptosis regulators. It has been mapped to chromosome 16q12 and is reportedly expressed predominantly in cells of the monocytic lineage. This gene encodes a 1,040 amino acid protein composed of two amino terminal caspase recruitment domains (CARDs), linked to a nucleotide binding domain (NBD) and multiple leucine rich repeats (LRRs) in the carboxyl terminal region [1, 2]. In vitro studies have demonstrated that NOD2 binds directly to a muramyl dipeptide (MDP) that is common to all bacteria [3, 4], and this binding leads to activation of nuclear factor-κB through activation of intermediary factors such as RICK and inhibitor of NF-κB kinase complex. Consequently, these findings suggest the NOD2 can act as intracellular receptors for bacterial molecules in mammalian cells [1, 5].
Recently, mutations in NOD2 have been found in Crohn’s disease (CD) and Blau syndrome (BS). The majority of the CD associated variants of NOD2 localize in the LRRs. The three most common mutations result in defective NF-κB activation and the clearance of bacteria is defective in mutated NOD2 transfectants [6–8]. BS is a rare familial granulomatous disease inherited as an autosomal dominant disorder. Affected individuals typically exhibit granulomatous inflammation manifested as uveitis, arthritis, and skin lesions. It is been found that three NOD2 mutations (R334Q, R334W and L469F) located in the region encoding the NBD are associated with BS [9–10].
Because NOD2 is a cytosolic receptor for bacterial components, we sought to investigate the effect of the Blau mutation on biological aspects of bacterial interaction with cells. Since the biological role of NOD2 is interaction with microbial elements, we addressed the question whether the mutations associated with BS might alter host:microbial interactions.
Materials and methods
Transfection of U937 cells
RT-PCR was used to generate cDNA for the NOD2 gene from human peripheral blood. The construct was sequenced and exactly matched with that, previously reported . Using a commercial kit (Stratagene), site-directed mutagenesis was performed to create NOD2Blau (a G1001A nucleotide transition to change an arginine to a glutamine codon—R334Q). The NOD2wt and NOD2Blau were then subcloned into the retroviral vector pGC containing a yellow fluorescent protein (YFP) reporter . Recombinant retrovirus containing each cDNA was produced by transient transfection of the Phoenix-A retroviral packaging line that expresses an amphotropic envelope for targeting human cells. Recombinant retrovirus was then used to infect U937 cells. After 48 h in culture, cells were cloned, at 1-cell per well into a 96-well microtiter plate by FACS, based on expression of YFP. Clones were expanded and analyzed for YFP expression at day 14. Three high expressing clones were selected for further analysis, one expressing the NOD2wt, one the NOD2Blau and one transfected with pGC alone (VC).
U937 cells were lysed at 5 × 106 cells/200 μl lysis buffer (0.8 M Tris–HCl pH 8.0, 1% NP-40, 0.33 M NaCl and 5 mM EDTA). The whole cell extracts (13 μl, equivalent to 3.25 × 105 cells) were subjected to 4–12% NuPage gel (Invitrogen) with 2 μl sample reducing agent (Invitrogen) and 5 μl 4× sample buffer. The samples were boiled at 95°C for 8 min before loading and transferred to polyvinylidene difluoride membranes. Post-transfer, nonspecific binding was blocked with 2% BSA in PBS for 1 h. Membranes were then incubated for 1 h with 2D9, a monoclonal antibody specific for NOD2 (a gift of Gabriel Nunez, Ann Arbor, USA), at a final concentration of 30 μg/ml on rocking platform . After washing with PBS-1% Tween, goat anti-mouse-HRP (Santa Cruz) was added for 1 h at 1:1,000 dilution. Blots were developed using Western Lightning chemiluminescence Reagent Plus (Perken Elmer).
Cell preparation, bacterial invasion and clearance
Each U937 cell line was cultured in tissue culture flasks under lipopolysaccharide (LPS)-free conditions in RPMI 1640 medium containing 10% heat-inactivated fetal calf serum (FCS) in humidified 5% CO2 at 37°C. The cells were seeded into 24-well plates or tubes at a concentration of 1 × 106 cells/ml for 24 h . The cells were washed once with filtrated phosphate-buffered saline (PBS) and then overlaid with RPMI 1640 medium containing 10% FCS. The Salmonellatyphimurium was originally isolated from a patient with Salmonella-triggered ReA and were grown in Luria-Bertani broth at 37°C to the logarithmic phase of growth. U937 cells were cocultured with Salmonella at a 1:5-to-10 cell/bacterium ratio for 90 min at 37°C. After three washes with PBS, the cells were overlaid with fresh incubation medium containing 50 μg of gentamicin/ml (Gibco) and incubated for 60 min at 37°C. This was defined as the time point 0 (150 min after invasion) for quantitating invasion. Cells were then harvested at different time points. The U937 cells are only marginally adherent and great care was taken in performing the washes. The plate was spun after each wash to minimize cell loss. At each time point, cells were enumerated using a hemocytometer and serial dilutions, and cell viability was assessed by trypan blue exclusion. Colony forming units (CFU) were determined after cell lysis (with 0.2% Triton-X-100) to calculate the total intracellular bacteria load . Previous studies using U937 have variably used PMA-activated cells . In pilot studies we found that there was no difference in the invasion profile between cells pretreated with PMA and non-stimulated cells.
In experiments addressing cytokine expression, PMA-stimulated U937 cells were incubated with either Salmonella LPS (500 ng/ml, Sigma) or the live pathogen. Total RNA from cells at different time points was prepared using TRIZOL reagent (Invitrogen). To assess mRNA expression, a semiquantitative RT-PCR method was used. The generated cDNA was amplified by using primers for TLR2 (5′-GGCCAGCAAATTACCTGTGTG-3′, 5′-CCAGGTAGGTCTTGGTGTTCA-3′), TLR4 (5′-GCTTCTTGCTGGCTGCATAA-3′, 5′-GAAATGGAGGCACCCCTTC-3′), β-actin (5′-GACAGCAGTCGGTTGGAGC-3′, 5′-CAGGTAAGCCCTGGCTGC-3′), TNF-α (5′-ACAAGCCTGTAGCCCATGTT-D′, 5′-AAAGTAGACCTGCCCAGACT-3′). After 26 (TNF-α), 32(TLR4, β-actin) or 35 (TLR2) amplification cycles, the expected PCR products were size fractionated onto a 1.7% agarose gel and stained with ethidium bromide (EtBr). Expected size for the PCR products for TNF-α, TLR4, TLR2 and β-actin were 427, 150, 617 and 540 bp, respectively. Quantitative analyses of PCR products were performed using Fluor-STM (Bio-Rad). The gel stained with EtBr was placed and scanned by Fluor-STM. We generated a negative image of the EtBr-containing bands.
The invasion data and relative ratio of TNF-α and β-actin was analyzed by ANOVA and post hoc test of SPSS 11 version. Statistical significance was defined as a P < 0.05.
Expression of NOD2
Uptake and clearance of bacteria
Whereas NODBlau appeared to minimize the enhanced invasion conferred by the NODwt, the profiles on intracellular bacteria clearance followed a different pattern. When comparing clearance of the intracellular bacteria over time, we noted that clearance at day 6 and day 9, was comparable between NODwt and VC. However NODBlau conferred in impairment in intracellular clearance, with significantly higher intracellular pathogen load in these cells at day 6 and day 9, in comparison with both NODwt and VC transfectants (Fig. 2).
Expression of TNF-α mRNA
Expression of TLR2 and TLR4 mRNA
The BS is a rare autosomal dominant disorder characterized by early-onset granulomatous arthritis, uveitis and skin rash in which the susceptibility locus has been mapped chromosome 16p12-q21 [9–10]. The finding that BS is associated with a mutation in NOD2 has raised the question whether innate immunity might be altered by virtue of this mutation. Innate immunity represents the first line of defence against bacteria invasion. Toll-like receptors (TLR) are critical components of this defense system. NOD2 is a cytosolic counterpart of TLR and interacts with MDP present in all bacteria, activating NF-κB [3–5, 15]. Recent genetic studies have established an important role for NOD2 mutations in CD. One copy of the mutated alleles can induce a two- to fourfold increased risk of developing CD, whereas homozygosity or compound heterozygosity for NOD2 mutations increases the risk 20–40-fold [6–7, 16].
The impact of the CD-related NOD2 mutation on bacterial invasion in vitro has been examined by measuring bacteria recovery from infected cells . In that study, wild type NOD2 appeared to confer diminished invasion, compared to nontransfected cells and cells expressing the CD-related NOD2 mutation. These results suggested that wild type NOD2 protects host cells by decreasing invasion of S. typhimurium, and that this defensive role of NOD2 is functionally defective in the presence of CD-mutant NOD2. In our study with the BS-related NOD2 mutation, the initial uptake of S. typhimurium was diminished in cells expressing NOD2Blau compared to cells expressing NOD2wt or VC alone. However, the intracellular clearance of S.typhimurium after 6 days was diminished in NOD2Blau cells compared with controls. These data indicate that NOD2Blau influences both the uptake and clearance of arthritogenic pathogens. Since NOD2 is an intracellular protein, the mechanism by which NOD2 affects the invasion by bacteria is not immediately apparent, although experimental conditions for our early invasion time points do incorporate initial intracellular processing of the bacteria. In any event, the diminished intracellular bacterial load associated with the NOD2Blau is associated with a diminished TNF-α expression pattern, since NODBlau cells demonstrate both decreased invasion and more rapid decline in TNF-α expression compared to NODwt. Although initial invasion was diminished, intracellular bacterial clearance profiling indicated that NOD2Blau conferred a relative defect in clearance, which might in turn have related to the relative lack of sustained TNF-α. This might result in persistence of intracellular pathogens in vivo, but there has been no systematic analysis of BS patients in this regard. TNF-α itself has been shown to play a critical role in host clearance of arthritogenic bacteria, and studies from our lab have demonstrated more accelerated reactive arthritis in mice, deficient in TNF-α receptors . This has clinical precedent in the altered host clearance of Chlamydia, in which defective host clearance might set the stage for chronic antigenic stimulation in the joint, with chronic reactive arthritis occurring as a result . It would be valuable to examine this question, using cells from patients with BS, to address whether the in vitro results of our study are translated to the clinical syndrome itself.
The relative decrease in TNF-α expression in NOD2Blau following challenge with S. typhimurium was also seen when Salmonella LPS was used to stimulate the cells (data not shown). Since U937 cells express TLR4 on their cell surface, the induced TNF-α can be triggered by both the interaction of LPS with surface TLR4, as well as NOD1 binding diaminoopimelic containing peptidoglycan and NOD2 binding MDP—which may be contaminating the preparation of LPS [3, 4, 19]. Because of the importance of TLR in the first line of defense against bacteria, we addressed whether the NOD2Blau mutation might influence expression of TLR2 and TLR4 after Salmonella invasion. There were no differences observed in any of the cell lines by quantitative analyses. It is known that TLR4 and TLR2 expression can be influenced by signaling from the cell surface via TLR themselves. Whether NOD1 or NOD2 can play a role in this pathway is not known, but our results suggest that any such process would not be influenced by the BS mutation in NOD2.
The role of bacterial infection is unresolved in BS, as indeed it is in CD. NOD2 mutation associated with BS alters host:microbial interactions in a significant way and warrants further study.