Caspase activating and recruitment domain 8 (CARD8) potently inhibits NF-κB signaling, which plays a key role in inflammation, and may contribute to avoid a pathologic activation of NF-κB; however, the transcriptional mechanisms regulating CARD8 expression and the relevance of this protein in inflammatory diseases are poorly understood. We found a NF-κB-binding element within the human CARD8 promoter that was required for transcriptional activity in response to TNF-α and the p65 subunit of NF-κB. Moreover, TNF-α and overexpression of p65 induced the formation of NF-κB-CARD8 promoter complexes. Thus, CARD8 may control NF-κB activation through a regulatory loop. To study the relevance of CARD8 in chronic inflammatory disorders, we functionally characterized a deleterious polymorphism (p.C10X) and studied its association with rheumatoid arthritis (RA). Transfection of cell lines with the allelic variants of CARD8 revealed that full-length (CARD8-L) but not truncated (CARD8-S) protein inhibits NF-κB transcriptional activity, and abrogates the binding of NF-κB to its consensus site. Furthermore, in contrast to the full-length protein, CARD8-S did not modify the expression of NF-κB target genes (cIAP, A1), in response to TNF-α. We analyzed the p.C10X polymorphism in 200 patients with RA, and found that homozygous carriers of the CARD8-S allele have higher disease activity score (p = 0.014), more extra-articular manifestations (p = 0.03), and a lower probability of clinical remission (p = 0.03) than the CARD8-L allele carriers. Overall, our findings provide molecular insight into the expression of CARD8 by NF-κB, and suggest that a deleterious polymorphism of CARD8 may help predict the severity of RA.

Activation of the NF-κB pathway is involved in the pathogenesis of chronic inflammatory diseases, including rheumatoid arthritis (RA),3 and inflammatory bowel disease. In addition, altered NF-κB regulation may participate in other diseases such as atherosclerosis and Alzheimer’s disease, in which the inflammatory response is at least partially implicated (1). NF-κB regulates immune and inflammatory responses, and is also involved in protecting cells from undergoing apoptosis in response to a variety of apoptotic stimuli. There is increasing evidence that NF-κB is a major, if not the main, transcription factor controlling inflammation. This transcription factor is regulated by two kinases, IκB kinase 1 (IKK1) and 2 (IKK2). The latter is particularly important as it phosphorylates IκB, which is subsequently ubiquitinated and degraded, thus leading to the activation of NF-κB. In addition, some members of the nucleotide-binding domain and leucine-rich repeat containing (NLR) family (formerly known as CATERPILLER family), including NOD1, NOD2, NLRP3/NALP3, and NLRP4/NALP4, regulate the activity of NF-κB (2). This family is comprised of proteins grouped in subfamilies based on the domain architecture and provide positive and negative signals for the control of immune and inflammatory responses (3). An interesting example is found in NLRP3 which has been shown to activate NF-κB in the presence of apoptosis-associated speck-like protein containing the caspase activating and recruitment domain (CARD) (4), and is also able to exert an inhibitory effect on TNF and TNF-receptor-associated factor 6-induced NF-κB activation (5). Another protein, that shares structural and functional features with some members of the NLR family (e.g., NLRP1), CARD8, has also been shown to possess NF-κB suppressor activity in vitro (6, 7, 8). Different protein-protein interactions provide a molecular platform triggering activation of NF-κB. For instance, NOD1 has been shown to engage the NF-κB pathway by a recruitment of the receptor-interacting protein kinase 2, subsequently activating the IKK complex via induced proximity signaling (9). Furthermore, NLRP4 associates with IKK1 and suppresses cytokine-mediated activation of this kinase, which plays a critical role in controlling degradation of IκB, thus releasing NF-κB (10). Coimmunoprecipitation experiments also revealed that CARD8 interacts with the regulatory subunit of the IκB kinase, NEMO, providing a molecular basis for CARD8 inhibitory activity (6). Thus, the emerging view is of a complex balance between inducers and inhibitors of the NF-κB-signaling pathway that in the proper context serve to amplify or suppress inflammatory processes. However, the contribution of each single gene to the response to inflammatory signals and its association with inflammatory disorders still awaits further investigation.

Genetic variants of NLRP3 have been associated with familial cold autoinflammatory syndrome and Muckle-Wells syndrome (11). Consistently, macrophages from Muckle-Wells patients spontaneously secrete active IL-1β (12). It has also been shown that genetic variations of NOD2 increase the susceptivity to Crohn’s disease and Blau syndrome, two chronic inflammatory disorders (13, 14, 15). Therefore, mutations or polymorphisms in genes that can act as regulators of NF-κB activation are associated with persistent autoinflammatory and fever diseases.

A number of studies suggest that NF-κB is essential for the expression of both inflammatory cytokines and tissue-destructive enzymes in RA (16). Consistently, activation of NF-κB has been shown in synovial tissue from RA patients, and this appears to be related to the clinical diagnosis (17). In this study, we describe that NF-κB transactivates the CARD8 gene, which may lead to a regulatory loop that controls NF-κB activation in response to inflammatory stimuli. We also showed that a truncating polymorphism abrogates the capacity of CARD8 to inhibit the NF-κB transcriptional activity, and that this polymorphism modifies the severity of RA.

Two hundred patients with RA, diagnosed according to the 1987 American College of Rheumatology (ACR) criteria (18) were assessed. The patients were followed in two different rheumatology units and have different characteristics (Table I). Patients studied at Hospital Universitario Marques de Valdecilla (Santander, Spain) included 91 unselected patients with long-lasting RA. DNA samples and clinical information, including demographic data, disease characteristics, and treatment, were available from previous studies. Patients followed at Hospital Universitario La Paz (Madrid, Spain) included 109 unselected patients attended in the Early Arthritis Clinic of the Rheumatology Unit. All these patients were prospectively studied every 6 mo and demographic, clinical, radiological, and therapy data were recorded in a database. For genotype distribution, a control group of 200 sex- and age-matched individuals who had no known history of serious disease, including autoimmune or chronic infectious disorders, was also included in this study. The study was approved by the Hospital Universitario Marques de Valdecilla Research Ethics Committee and all subjects gave informed consent before participation in this study.

Table I.

Main characteristics of the RA patients included in the studya

Cohort 1 (n = 91)Cohort 2 (n = 109)
Females (%) 68.1 72.5 
Age (years) 54.5 ± 13b 52 ± 16b 
Interval to diagnosisc (months) 33.6 ± 6b 4 ± 2.3b 
Duration of follow-up (years) 13.6 + 0.7b 
RF positive (%) 59.3 74.1 
SE positive (%) 58.2 61.7 
Erosive disease (%) 90.1 95.9 
Patients (%) treated with:   
 DMARDS 98.9 88.3 
 Combined therapy 52.7 34 
 Biologic therapy 23.1 22.3 
Cohort 1 (n = 91)Cohort 2 (n = 109)
Females (%) 68.1 72.5 
Age (years) 54.5 ± 13b 52 ± 16b 
Interval to diagnosisc (months) 33.6 ± 6b 4 ± 2.3b 
Duration of follow-up (years) 13.6 + 0.7b 
RF positive (%) 59.3 74.1 
SE positive (%) 58.2 61.7 
Erosive disease (%) 90.1 95.9 
Patients (%) treated with:   
 DMARDS 98.9 88.3 
 Combined therapy 52.7 34 
 Biologic therapy 23.1 22.3 
a

RF, rheumatoid factor; SE, shared epitope (HLA-DRB1); DMARDs, disease-modifying anti-rheumatic drugs.

b

Mean ± SD.

c

The lag time between onset of symptoms and diagnosis of RA.

Human cell lines Namalwa, 1411, DU145, 2102, SHSY5Y, Elijah, MCF7, HL-60, MCF7, and HEK293T were maintained in RPMI 1640 medium (Biochrom) supplemented with 10% FCS (Flow Laboratories). SUM159 cells were incubated in DMEM/HAM’S F-12 (Biochrom) supplemented with 5% FCS, 5 μg/ml insulin, and 1 μg/ml hydrocortisone (both obtained from Sigma-Aldrich). When indicated, peripheral blood was incubated with 10 μg/ml LPS from Salmonella typhimurium (Sigma-Aldrich) for 4 h and then analyzed for the production of IL-8 by using an enzyme immunoassay (BD Biosciences).

We analyzed a single nucleotide polymorphism (SNP identification: rs2043211:T > A) located at the third nucleotide of codon 10 of CARD8 (RefSeq NM_014959.1). This polymorphism consists of a T to A transversion (c.30T > A) that generates a premature stop codon (p.C10X). DNA was extracted from whole blood by using the QIAamp DNA blood kit (Qiagen) and amplified with primers for human CARD8 5′-GCCTATGCTATCATCAGGCACC-3′ and 5′-TTCATTCTCCCCTGAGTTCGAT-3′. The amplified fragment (317 bp) was digested with AlwI (New England Biolabs), which recognizes the A allele generating two fragments of 165 and 152 bp. The digestion products were run on a 2% agarose gel. Approximately 30% of all DNA samples have also been sequenced in both directions with the same primers used for amplification, and we verified the authenticity of the polymorphism in all cases.

Cells were lysed and nuclear fractions were resuspended in 20 mM HEPES (pH 7.9), 420 mM NaCl, 1 mM EDTA, 1 mM EGTA, and 20% glycerol. Nuclear extracts (10 μg of total protein) were incubated with a [32P]dsDNA probe from the promoter region of the CARD8 gene (5′-GGGATTCTCC-3′) or a probe corresponding to the consensus NF-κB site (5′-GGGAATTTCC-3′). Samples were run on a 5% nondenaturing polyacrylamide gel in 200 mM Tris-borate, 2 mM EDTA. Gels were dried and visualized by autoradiography. Supershifts were performed using rabbit polyclonal Abs specific for p50, p65 and GATA1 (Santa Cruz Biotechnology).

Total RNA was prepared using TRIzol reagent (Invitrogen Life Technologies). To assess mRNA expression, a quantitative real-time PCR was performed as previously described (19). The generated cDNA was amplified by using primers for human cIAP-1, GAPDH (19), A1 (20), and CARD8 (5′-TGGTATCTGTGGTCAGCCAC-3′, and 5′-GAAGCTGGGGCTTTCCAG-3′). The ratio of the abundance of cIAP-1, A1, and CARD8 transcripts to that of GAPDH transcripts was calculated as described (19).

A genomic PCR fragment of 1170 bp from the promoter region of CARD8 (CARD8pt), starting 86 bases downstream from the transcription start site, were cloned into KpnI and XhoI sites of the pGL2-basic luciferase reporter vector (Promega). HEK293T cells were cotransfected with 1 μg of pGL2-CARD8pt and 0.2 μg of pRSV-β-galactosidase by lipofection using Superfect (Qiagen). Twenty-four hours posttransfection, cells were incubated with 10 ng/ml TNF-α (Sigma-Aldrich) for 6 h, and then cell extracts were prepared and analyzed for the relative luciferase activity by a dual-light reporter gene assay system (Applied Biosystems). When indicated, cells were cotransfected with 1 μg of pGL2-CARD8pt and 0.2 μg of pRc/CMV-p65 (provided by I. Udalova, University of Oxford, Oxford, U.K.) as above and 24 h later, cell extracts were analyzed for luciferase activity. Site-directed mutagenesis of the pGL2-CARD8pt vector was conducted by using the QuickChange Site-Directed Mutagenesis kit (Stratagene) with the following primers: 5′-ATGGTAGCATAACCT-3′ and 5′-AGGTTATGCTACCAT-3′. The CARD8pt DNA insert was sequenced to verify the mutation.

In another set of experiments, HEK293T cells were cotransfected with 1–2 μg of CARD8 cDNA variants cloned into the pCR3.1 vector, 0.25 μg of the reporter plasmid pBVIx-Luc, containing six NF-κB recognition sites within the promoter sequence linked to the luciferase gene (21) and 0.2 μg of pRSV-β-galactosidase by lipofection with Superfect. Twenty-four hours posttransfection, cells were incubated with 5 ng/ml TNF-α for 16 h and then cell extracts were prepared and analyzed for the relative luciferase activity. Results were normalized for transfection efficiency with values obtained with pRSV-β-gal.

Cell extracts (60 μg of protein) were separated on a 8% polyacrylamide gel and transferred to nitrocellulose as previously described (22). Blots were blocked with 3% BSA and incubated with rabbit anti-CARD8 (6), or mouse anti-α-tubulin (Sigma-Aldrich) Abs followed by incubation with goat anti-rabbit or anti-mouse Abs conjugated to alkaline phosphatase. Bound Ab was detected by a chemiluminescence system (Applied-Biosystems).

All statistics were calculated with the SPSS statistical package (version 13.0). Continuous variables are summarized as means ± SD. Categoric variables were compared using the χ2 test or Fisher exact test when appropriate. The Student t test was used to compare continuous variables between two groups. A one-way ANOVA with Tukey’s post-hoc multiple-comparison test was used to analyze differences in mean disease activity score (DAS) among patients with different genotypes of CARD8. DAS at different time intervals was determined by performing a repeated measures ANOVA. All statistical tests were two-sided and the significance level was set at p < 0.05.

CARD8 has been shown to inhibit NF-κB activity (6, 7, 8). Interestingly, some of the target genes of NF-κB (IL-1β, TNF-α) in turn control the activity of this transcription factor (23). In line with this, we previously described that NOD2, a NF-κB inducer, is transcriptionally activated by NF-κB (19). Based on these observations, we first analyzed the expression of CARD8 mRNA in myeloblastic HL-60 cells after treatment with a NF-κB activator. Following incubation with LPS, the mRNA levels of CARD8 increased ∼4-fold by 24 h and 6-fold by 48 h (Fig. 1,A). We then stimulated a nonhemopoietic cell line with TNF-α, another NF-κB activator, and found a similar pattern of CARD8 expression, although in these cells the highest mRNA levels (∼5-fold induction compared with untreated cells) were reached at the 24 h time point (Fig. 1,A). In both cell lines, cotreatment with Bay11-7082, an inhibitor of NF-κB activation, abrogated the induction of CARD8 (data not shown). Based on the up-regulation of CARD8 in response to NF-κB activators, we searched for consensus sites within the CARD8 promoter region and found a putative NF-κB recognition sequence in the reverse orientation 18 bases upstream from the transcription start site (Fig. 1,B). To assess the transcriptional activity of the CARD8 promoter, a 1170-bp fragment containing the NF-κB consensus site of the CARD8 promoter (CARD8pt) was cloned into a luciferase vector and this construct was transiently transfected in HEK293T cells. Stimulation of the cells with TNF-α induced the transcriptional activity ∼3.5-fold when compared with unstimulated cells (Fig. 1,C). To confirm the capacity of NF-κB to transactivate CARD8, we performed the gene reporter assays in the presence of the p65 subunit of NF-κB. As shown in Fig. 1,C, overexpression of p65 as detected by Western blot, increased the luciferase activity by 16-fold compared with cells transfected with the empty expression vector. To assess the relevance of the putative NF-κB site, we mutated four bases within this sequence motif in the CARD8 promoter fragment (Fig. 1,B). In contract to the wild-type promoter, the mutant CARD8pt-luciferase construct was not induced by either TNF-α or p65 (Fig. 1,C). To directly prove the binding of NF-κB to the CARD8 promoter, MCF7 cells were incubated with TNF-α and then nuclear extracts were subjected to EMSA. We detected a protein-DNA complex following activation of NF-κB that was supershifted in the presence of Abs against p65 (Fig. 2,A). Similar result was obtained when HEK293T cells were transfected with a p65-containing expression vector (Fig. 2 B), although this time the complexes formed by p65 homodimers were predominant as revealed by the use of NF-κB subunit-specific Abs.

FIGURE 1.

Transcriptional activation of the CARD8 promoter. A, HL-60 and MCF7 cells were cultured with 10 μg/ml LPS and 10 ng/ml TNF-α, respectively, for the indicated time intervals, and then total RNA was extracted and analyzed for the expression of CARD8 by real-time RT-PCR. B, Consensus site for NF-κB recognition in the CARD8 promoter (CARD8pt). C, HEK293T cells were transfected with a 1170-bp fragment of the CARD8 promoter or a mutant sequence (mutCARD8pt) (mutation shown in B). Promoter activation was assayed by cotransfection with a p65-containing vector or treatment with TNF-α for 6 h. After 24 or 30 h of transfection, CARD8pt-dependent transcription was determined. Units of luciferase activity were normalized based on values of pRSV-β-galactosidase activity to control for transfection efficiency. Data are presented as the mean of triplicate cultures ± SD.

FIGURE 1.

Transcriptional activation of the CARD8 promoter. A, HL-60 and MCF7 cells were cultured with 10 μg/ml LPS and 10 ng/ml TNF-α, respectively, for the indicated time intervals, and then total RNA was extracted and analyzed for the expression of CARD8 by real-time RT-PCR. B, Consensus site for NF-κB recognition in the CARD8 promoter (CARD8pt). C, HEK293T cells were transfected with a 1170-bp fragment of the CARD8 promoter or a mutant sequence (mutCARD8pt) (mutation shown in B). Promoter activation was assayed by cotransfection with a p65-containing vector or treatment with TNF-α for 6 h. After 24 or 30 h of transfection, CARD8pt-dependent transcription was determined. Units of luciferase activity were normalized based on values of pRSV-β-galactosidase activity to control for transfection efficiency. Data are presented as the mean of triplicate cultures ± SD.

Close modal
FIGURE 2.

Binding of NF-κB to the CARD8 promoter is induced by TNF-α and p65. A, MCF7 cells were incubated with (T) or without (C) 10 ng/ml TNF-α for 1 h. Formation of binding complexes was determined by EMSA using a radiolabeled probe from the CARD8 promoter or a NF-κB consensus probe. Nuclear extracts from stimulated cells were preincubated with 100-fold molar excess of unlabeled probes, an irrelevant nonspecific probe (NSP), or a specific Ab against p65. Anti-GATA1 was included as a negative control. B, HEK293T cells were transfected with p65 and 24 h later nuclear extracts were incubated with a probe from the CARD8 promoter containing the NF-κB site and then subjected to EMSA. Treatment with TNF-α was also included for comparison. Extracts were preincubated with the indicated Abs to determine the NF-κB subunit composition of the complexes.

FIGURE 2.

Binding of NF-κB to the CARD8 promoter is induced by TNF-α and p65. A, MCF7 cells were incubated with (T) or without (C) 10 ng/ml TNF-α for 1 h. Formation of binding complexes was determined by EMSA using a radiolabeled probe from the CARD8 promoter or a NF-κB consensus probe. Nuclear extracts from stimulated cells were preincubated with 100-fold molar excess of unlabeled probes, an irrelevant nonspecific probe (NSP), or a specific Ab against p65. Anti-GATA1 was included as a negative control. B, HEK293T cells were transfected with p65 and 24 h later nuclear extracts were incubated with a probe from the CARD8 promoter containing the NF-κB site and then subjected to EMSA. Treatment with TNF-α was also included for comparison. Extracts were preincubated with the indicated Abs to determine the NF-κB subunit composition of the complexes.

Close modal

The impact of other NF-κB regulators on human physiology and disease has been assessed through the identification of sequence variations that modify the activity of the protein. With this in mind, we first searched for SNPs that may introduce damaging changes into the protein. The SNP rs2043211 comprises a T to A transversion (Fig. 3, A and B), and changes a cysteine residue (TGT) to a stop codon (TGA) at amino acid position 10 (p.C10X) which, presumably leads to a severely truncated protein. Analysis of this SNP in 200 DNA samples from normal donors revealed 8% of AA and 44% of AT genotype carriers. The structural relevance of this polymorphism was confirmed by analyzing the expression of CARD8 protein in peripheral blood cells from carriers of the different genotypes. As shown in Fig. 3,C, homozygous carriers of the A allele lack the full-length protein, and those carrying both variant alleles show a reduction in CARD8 protein levels. To further confirm this result, HEK293T cells were transfected with CARD8 cDNA carrying either the T or the A polymorphism and then subjected to Western blot analysis. Fig. 3 D shows that those cells transfected with the T-containing construct (long isoform) produced a protein of the expected size, and those carrying the stop polymorphism (short isoform) did not.

FIGURE 3.

Analysis of the p.C10X polymorphism in CARD8. A, Sequence of the CARD8 genotypes. B, Analysis of the different genotypes by PCR followed by AlwI digestion. Note that the lower band visualized in the agarose gel contains both restriction fragments (152 and 165 bp). C, The expression of CARD8 protein in peripheral blood cells was analyzed by Western blot with a specific polyclonal Ab. The levels of α-tubulin were also determined to assure equal loading. D, HEK293T cells were transfected with the full-length, L, or the truncated, S, isoform of CARD8 and then analyzed by Western blot. V, cells transfected with the empty expression vector as a negative control.

FIGURE 3.

Analysis of the p.C10X polymorphism in CARD8. A, Sequence of the CARD8 genotypes. B, Analysis of the different genotypes by PCR followed by AlwI digestion. Note that the lower band visualized in the agarose gel contains both restriction fragments (152 and 165 bp). C, The expression of CARD8 protein in peripheral blood cells was analyzed by Western blot with a specific polyclonal Ab. The levels of α-tubulin were also determined to assure equal loading. D, HEK293T cells were transfected with the full-length, L, or the truncated, S, isoform of CARD8 and then analyzed by Western blot. V, cells transfected with the empty expression vector as a negative control.

Close modal

The region responsible for the inhibitory effect on NF-κB is contained within the first 320 aa of CARD8 (6). Thus, we studied whether the protein truncated at amino acid position 10 retained, at least in part, the inhibitory capacity. We first analyzed the p.C10X polymorphism in a large number of cell lines and identified four homozygous for the TT genotype and four for the AA genotype (Fig. 4,A). The activity of NF-κB in these cells was assessed by EMSA using a NF-κB consensus probe. As shown in Fig. 4,B, three of the TT homozygous cells have very little or no constitutive activation of NF-κB as compared with most of the AA homozygous cell lines. Specificity of protein-DNA complexes was verified with Abs against the NF-κB subunits p50 and p65 and appeared as supershifted bands. To further study the relevance of the p.C10X polymorphism on the activity of NF-κB, we cotransfected HEK293T cells with the genetic variants of CARD8 and a luciferase reporter gene driven by a promoter sequence containing six NF-κB-responsive elements (Fig. 5,A). Expression of exogenous CARD8 was assessed by Western blot analyses (Fig. 5,B). Because TNF-α is widely regarded as one of the most potent inflammatory stimulants, and leads to NF-κB activation, the reporter gene was induced following incubation of transfected cells with TNF-α. This induction was reduced >3-fold in cells expressing the long isoform (CARD8-L), whereas no change was observed in cells transfected with the short variant (CARD8-S) (Fig. 5,A). Consistent with this result, full-length CARD (CARD-L) but not truncated CARD (CARD-S), was able to inhibit the binding of p50-p65 complexes to a NF-κB consensus sequence following treatment with TNF-α as assessed by mobility shift assay (Fig. 5,C). Furthermore, CARD8-L-transfected cells showed a significant reduction (>5-fold) in the mRNA levels of cIAP1, a NF-κB target gene, in response to TNF-α. Again, expression of CARD8-S did not modify the transcriptional activity of NF-κB (Fig. 6,A). Similar results were obtained when DU145 cells, which do not express the full-length CARD8 protein, were transfected with either CARD8-L or CARD8-S. Although CARD8-L blocked the induction of another NF-κB target gene, A1, the truncated protein had no significant effect (Fig. 6 B). Consistently, the LPS-stimulated production of IL-8, a NF-κB target gene, in whole blood from three homozygous carriers of each allele showed a tendential increase in IL-8 secretion from cells carrying the CARD8-S allele (mean ± SD: 1.21 ± 0.57) compared with that of cells carrying the CARD8-L allele (mean ± SD: 0.65 ± 0.27). Statistical significance would most likely be reached by increasing the sample size to overcome individual variability due to differences in the mononuclear cell populations and other extrinsic and intrinsic factors that modulate the production of proinflammatory cytokines such as IL-8.

FIGURE 4.

Correlation between the CARD8 genotypes and the DNA-binding capacity of NF-κB. A, Different cell lines carrying the TT or the AA genotypes as assessed by PCR followed by AlwI digestion. B, Cells were analyzed for the formation of NF-κB-DNA-binding complexes by EMSA using a radiolabeled probe containing a NF-κB consensus sequence. Nuclear extracts were preincubated with Abs against NF-κB subunits (p50 and p65) and with irrelevant anti-GATA1 Abs to demonstrate the specificity of the binding.

FIGURE 4.

Correlation between the CARD8 genotypes and the DNA-binding capacity of NF-κB. A, Different cell lines carrying the TT or the AA genotypes as assessed by PCR followed by AlwI digestion. B, Cells were analyzed for the formation of NF-κB-DNA-binding complexes by EMSA using a radiolabeled probe containing a NF-κB consensus sequence. Nuclear extracts were preincubated with Abs against NF-κB subunits (p50 and p65) and with irrelevant anti-GATA1 Abs to demonstrate the specificity of the binding.

Close modal
FIGURE 5.

Capacity of the CARD8 variants to modulate the transcriptional activity of NF-κB. A, HEK293T cells were transfected with the indicated amounts of the short, S, and long, L, variants of CARD8 along with a luciferase reporter plasmid containing six NF-κB responsive elements. After 16 h of treatment with the NF-κB inducer TNF-α, the luciferase activity was determined. B, Expression of the CARD8 protein variants in transfected cells was analyzed by Western blotting. The levels of α-tubulin were also determined to assure equal loading. C, Cells were transfected as in A, and maintained in the presence of TNF-α for 1 h. Then nuclear extracts were analyzed for the formation of protein-DNA complexes by EMSA using a NF-κB consensus probe. Nuclear extracts from cells cotransfected with p50 and p65 were run in parallel to confirm the presence of both proteins in the binding complex. V, cells transfected with the empty expression vector as a negative control. Histograms show the means ± SD of three independent experiments.

FIGURE 5.

Capacity of the CARD8 variants to modulate the transcriptional activity of NF-κB. A, HEK293T cells were transfected with the indicated amounts of the short, S, and long, L, variants of CARD8 along with a luciferase reporter plasmid containing six NF-κB responsive elements. After 16 h of treatment with the NF-κB inducer TNF-α, the luciferase activity was determined. B, Expression of the CARD8 protein variants in transfected cells was analyzed by Western blotting. The levels of α-tubulin were also determined to assure equal loading. C, Cells were transfected as in A, and maintained in the presence of TNF-α for 1 h. Then nuclear extracts were analyzed for the formation of protein-DNA complexes by EMSA using a NF-κB consensus probe. Nuclear extracts from cells cotransfected with p50 and p65 were run in parallel to confirm the presence of both proteins in the binding complex. V, cells transfected with the empty expression vector as a negative control. Histograms show the means ± SD of three independent experiments.

Close modal
FIGURE 6.

A and B, Cells were transfected with the CARD8 variants as indicated in Fig. 5 and then treated or not with 10 ng/ml TNF-α for 24 h. The mRNA levels of two NF-κB target genes, cIAP1 and A1, were determined by quantitative RT-PCR. Histograms show the means ± SD of three independent experiments.

FIGURE 6.

A and B, Cells were transfected with the CARD8 variants as indicated in Fig. 5 and then treated or not with 10 ng/ml TNF-α for 24 h. The mRNA levels of two NF-κB target genes, cIAP1 and A1, were determined by quantitative RT-PCR. Histograms show the means ± SD of three independent experiments.

Close modal

Our previous data indicated that CARD8-S has completely lost the capacity to interfere with the activation of NF-κB in response to inflammatory stimuli. NF-κB plays a relevant role in the induction of inflammatory mediators in chronic inflammatory disorders such as RA (16). To assess the contribution of the CARD8-S allele in predicting RA diagnosis or modifying the course of the disease, we studied the p.C10X polymorphism in DNA samples from 200 RA patients. The initial part of the study was conducted in 91 patients with long-lasting RA. The genotype frequencies were not significantly different between patient (9% AA, 45% AT, 46% TT) and control (8% AA, 44% AT, 48% TT) populations. However, the presence of the homozygous CARD8-S allele was associated with extra-articular manifestations comprising s.c. nodules, secondary Sjögren syndrome, episcleritis/scleritis, pericardial or pleural effusion, interstitial lung disease, and systemic vasculitis (p = 0.03), and with a lower probability of clinical remission (p = 0.03) as defined according to the ACR criteria (24) (Fig. 7,A). These results suggested that the p.C10X polymorphism was associated with a more severe disease in a sample of patients with a considerable delay in diagnosis and long-lasting RA. The clinical relevance of this association was further investigated in an independent cohort involving 109 patients with early RA assessed at regular intervals (every 6 mo) according to a well-established protocol that include DAS for extended joint counts, functional capacity as measured by Health Assessment Questionnaire score, laboratory (rheumatoid factor, anti-cyclic citrullinated peptide Abs, acute phase reactants) findings and radiographic data assessed with the Sharp-van der Heijde score (25, 26). Patients were followed for a period of 2 years in contrast to the mean 13.6-year follow-up of patients in cohort 1. Additionally, clinical remission was assessed according to DAS-based criteria (27). In agreement with the previous group of RA patients, the genotype frequencies were not significantly different between patients (9% AA, 47% AT, 52% TT) and controls. However, we found that those patients carrying the CARD8-S allele had significantly higher levels of anti-cyclic citrullinated peptide Abs than CARD8-L carriers (891.7 ± 101.8 vs 549.4 ± 91.7 U/ml, 95% confidence interval 70.5–614, p = 0.014). Moreover, the extended DAS was significantly higher in patients homozygous for the CARD8-S allele than in homozygous carriers of the CARD8-L allele over 2 years of follow-up (Fig. 7 B), as determined by the repeated measures ANOVA (p = 0.014 at the 2-year time point). As it is frequently observed that the disease activity of a patient may fluctuate around the level of “no or minimal” disease activity, we studied the cumulative amount of disease activity over 2 years (area under the curve). Consistent with our previous result, on Tukey’s post-hoc testing, a significant increase of the cumulative DAS was found in AA carriers relative to TT carriers (p = 0.03; mean differences 3.73; 95% confidence interval 0.23–7.24).

FIGURE 7.

Association between the outcome of patients with RA and the genotypes of CARD8. A, Frequency of different clinical features studied in carriers of the different genotypes in a retrospective study (n = 91). B, Prospective study (n = 109) showing the activity of the disease (DAS) during a follow-up period of 2 years.

FIGURE 7.

Association between the outcome of patients with RA and the genotypes of CARD8. A, Frequency of different clinical features studied in carriers of the different genotypes in a retrospective study (n = 91). B, Prospective study (n = 109) showing the activity of the disease (DAS) during a follow-up period of 2 years.

Close modal

NF-κB is a major, if not the major, transcription factor controlling immunity and inflammation (28). The best known regulation of the NF-κB pathway is through the association of NF-κB complexes with the IκB inhibitor proteins which leads to cytoplasmic retention of NF-κB. Recently, CARD8 has been shown to potently inhibit NF-κB as it blocks the activation of this transcription factor in response to a number of stimuli, including IL-1 and TNF-α (6, 7, 8). In this study, we have described that the mechanism of CARD8 regulation involves transcriptional activation of the CARD8 promoter through NF-κB. Because CARD8 represses the activity of NF-κB and this process is likely to limit the induction of inflammatory mediators, up-regulation of CARD8 may be part of a negative regulatory loop to control the inflammatory response. A similar regulatory mechanism has been recently described for NOD2 (19), a protein known to activate NF-κB by means of receptor-interacting protein kinase 2. NOD2 is up-regulated in response to LPS and TNF-α through a transcriptionally active NF-κB site in its promoter, and this mechanism has been suggested to contribute to the amplification of the innate immune response and susceptibility to inflammatory disease. We have also described the functional consequences of a polymorphism of CARD8 that generates a premature stop codon (p.C10X). We showed that this polymorphism severely truncated the protein and that cell lines carrying both truncating (CARD8-S) alleles have an increased basal level of NF-κB-DNA-binding complexes. Furthermore, in contrast to the full-length protein, CARD8-S was unable to inhibit the NF-κB transcriptional activity in response to TNF-α. Thus, although the inhibitory capacity of CARD8 has been located within the N-terminal end of the protein (6, 7), the short peptide containing the first 10 aa, that results from the truncating polymorphism, does not retain this functional capacity. Mutations in the NF-κB-signaling pathway that abolish or reduce NF-κB activation have been associated with a number of inherited genetic disorders, mainly immunodeficiency syndromes (29). A heterozygous mutation affecting the gene-encoding IκBα has been described in patients exhibiting a severe immunodeficiency syndrome (30, 31). This alteration resulted in a specific impairment of IκBα degradation, which is needed for NF-κB to translocate to the nucleus where it promotes the expression of a number of genes. Mutations in NEMO, another component of the NF-κB regulatory pathway, abolish NF-κB activation and may give rise to a large number of defects in humans (29, 32). Thus, the existence of immunodeficiency-associated mutations that lead to inactivation of a key transcription factor to immune and inflammatory responses provides a rationale to consider that a genetic variant that reduces the inhibitory action on NF-κB may be clinically relevant in chronic inflammatory disorders where this factor is constitutively active. We studied the p.C10X polymorphism of CARD8 in RA, a chronic disease where NF-κB has been well-recognized as a pivotal regulator of inflammation (33). Although there were no differences in genotype distribution between control and patient groups, those patients carrying the CARD8-S allele in homozygosis had a more severe disease than carriers of the full-length allele. We found that after 2 years of follow-up of a group of patients with early RA, the activity of the disease measured as DAS, was significantly higher in patients homozygous for CARD8-S. It is worth noting that DAS, a combined index including tender and swollen joint counts, determination of acute phase reactants, and the global evaluation by the patient of the overall disease activity, is one of the most widely used activity indices in RA (27). Therefore, our data suggest that despite an early diagnosis and an appropriate early treatment, those patients carrying the truncating allele have a higher amount of disease activity and joint inflammation. In line with this, activation of NF-κB in vascular endothelium and type A synovial lining cells has been described in synovial tissue from patients with RA (34), and consistently, one of the major determinants of the severity of RA is the persistence of synovial inflammation (35). Thus, a polymorphism that renders the CARD8 protein inactive may reduce the control on the NF-κB-signaling pathway giving rise to a more active transcription factor and an amplification of the inflammatory process. A recent work described that the p.C10X polymorphism of CARD8 slightly increased the likelihood of acquiring Crohn’s disease (36), a chronic inflammatory disorder that has been previously shown to be associated with NOD2 polymorphisms (14). However, this work focused on the association with disease susceptibility, which resulted to be modest (odds ratio, 1.35). Animal models showing that inhibition of NF-κB impairs collagen-induced arthritis development (37), and that intra-articular administration of NF-κB blockers prevented the recurrence of streptococcal cell wall-induced arthritis in treated joints (38), provide further evidences that regulation of NF-κB is of major importance in RA. Taking into account the complex control on the activation of NF-κB and the increasing number of proteins involved, analyses of deleterious or activating polymorphisms in candidate genes will be needed to decipher their contribution to chronic inflammatory disorders.

Our observations shed light on the regulation of CARD8 and suggest, based on current data, that CARD8 may be involved in a negative regulatory loop to limit the inflammatory response of NF-κB. Our data also indicate that a deleterious polymorphism of CARD8 (p.C10X) contributes to disease outcome in patients with RA, and thus analysis of p.C10X may be relevant in this inflammatory and destructive disease. The clinical interest of this polymorphism in RA and other chronic inflammatory disorders will need to be confirmed in prospective studies on larger series of patients with different genetic background.

We are grateful to Seamus Martin for providing the anti-CARD8 Ab.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by Grants PI050169 and ISCIII-RETIC RD06/0020 from the Spanish Fondo de Investigacion Sanitaria, and API/06/02 from Fundacion Marques de Valdecilla-Instituto de Formacion e Investigacion Marques de Valdecilla.

3

Abbreviations used in this paper: RA, rheumatoid arthritis; IKK, IκB kinase; NLR, nucleotide-binding domain and leucine-rich repeat containing; DAS, disease activity score; SNP, single nucleotide polymorphism; CARD, caspase activating and recruitment domain.

1
Yamamoto, Y., R. B. Gaynor.
2001
. Therapeutic potential of inhibition of the NF-κB pathway in the treatment of inflammation and cancer.
J. Clin. Invest.
107
:
135
-142.
2
Ogura, Y., N. Inohara, A. Benito, F. F. Chen, S. Yamaoka, G. Nunez.
2001
. Nod2, a Nod1/Apaf-1 family member that is restricted to monocytes and activates NF-κB.
J. Biol. Chem.
276
:
4812
-4818.
3
Ting, J. P., D. L. Kastner, H. M. Hoffman.
2006
. CATERPILLERs, pyrin and hereditary immunological disorders.
Nat. Rev. Immunol.
6
:
183
-195.
4
Manji, G. A., L. Wang, B. J. Geddes, M. Brown, S. Merriam, A. Al-Garawi, S. Mak, J. M. Lora, M. Briskin, M. Jurman, et al
2002
. PYPAF1, a PYRIN-containing Apaf1-like protein that assembles with ASC and regulates activation of NF-κB.
J. Biol. Chem.
277
:
11570
-11575.
5
O’Connor, W., Jr, J. A. Harton, X. Zhu, M. W. Linhoff, J. P. Ting.
2003
. Cutting edge: CIAS1/cryopyrin/PYPAF1/NALP3/CATERPILLER 1.1 is an inducible inflammatory mediator with NF-κB suppressive properties.
J. Immunol.
171
:
6329
-6333.
6
Bouchier-Hayes, L., H. Conroy, H. Egan, C. Adrain, E. M. Creagh, M. MacFarlane, S. J. Martin.
2001
. CARDINAL, a novel caspase recruitment domain protein, is an inhibitor of multiple NF-κB activation pathways.
J. Biol. Chem.
276
:
44069
-44077.
7
Razmara, M., S. M. Srinivasula, L. Wang, J. L. Poyet, B. J. Geddes, P. S. DiStefano, J. Bertin, E. S. Alnemri.
2002
. CARD-8 protein, a new CARD family member that regulates caspase-1 activation and apoptosis.
J. Biol. Chem.
277
:
13952
-13958.
8
Stilo, R., A. Leonardi, L. Formisano, B. Di Jeso, P. Vito, D. Liguoro.
2002
. TUCAN/CARDINAL and DRAL participate in a common pathway for modulation of NF-κB activation.
FEBS Lett.
521
:
165
-169.
9
Inohara, N., T. Koseki, J. Lin, L. del Peso, P. C. Lucas, F. F. Chen, Y. Ogura, G. Nunez.
2000
. An induced proximity model for NF-κB activation in the Nod1/RICK and RIP signaling pathways.
J. Biol. Chem.
275
:
27823
-27831.
10
Fiorentino, L., C. Stehlik, V. Oliveira, M. E. Ariza, A. Godzik, J. C. Reed.
2002
. A novel PAAD-containing protein that modulates NF-κB induction by cytokines tumor necrosis factor-α and interleukin-1β.
J. Biol. Chem.
277
:
35333
-35340.
11
Hoffman, H. M., J. L. Mueller, D. H. Broide, A. A. Wanderer, R. D. Kolodner.
2001
. Mutation of a new gene encoding a putative pyrin-like protein causes familial cold autoinflammatory syndrome and Muckle-Wells syndrome.
Nat. Genet.
29
:
301
-305.
12
Agostini, L., F. Martinon, K. Burns, M. F. McDermott, P. N. Hawkins, J. Tschopp.
2004
. NALP3 forms an IL-1β-processing inflammasome with increased activity in Muckle-Wells autoinflammatory disorder.
Immunity
20
:
319
-325.
13
Hugot, J., M. Chamaillard, H. Zouali, S. Lesage, J. Cezard, J. Belaiche, S. Almer, C. Tysk, C. O’Morain, M. Gassull, et al
2001
. Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn’s disease.
Nature
411
:
599
-603.
14
Ogura, Y., D. Bonen, N. Inohara, D. Nicolae, F. Chen, R. Ramos, H. Britton, T. Moran, R. Karaliuskas, R. Duerr, et al
2001
. A frameshift mutation in NOD2 associated with susceptibility to Crohn’s disease.
Nature
411
:
603
-606.
15
Miceli-Richard, C., S. Lesage, M. Rybojad, A. Prieur, S. Manouvrier-Hanu, R. Hafner, M. Chamaillard, H. Zouali, G. Thomas, J. Hugot.
2001
. CARD15 mutations in Blau syndrome.
Nat. Genet.
29
:
19
-20.
16
Andreakos, E., S. Sacre, B. M. Foxwell, M. Feldmann.
2005
. The Toll-like receptor-nuclear factor κB pathway in rheumatoid arthritis.
Front. Biosci.
10
:
2478
-2488.
17
Marok, R., P. G. Winyard, A. Coumbe, M. L. Kus, K. Gaffney, S. Blades, P. I. Mapp, C. J. Morris, D. R. Blake, C. Kaltschmidt, P. A. Baeuerle.
1996
. Activation of the transcription factor nuclear factor-κB in human inflamed synovial tissue.
Arthritis Rheum.
39
:
583
-591.
18
Arnett, F. C., S. M. Edworthy, D. A. Bloch, D. J. McShane, J. F. Fries, N. S. Cooper, L. A. Healey, S. R. Kaplan, M. H. Liang, H. S. Luthra, et al
1988
. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis.
Arthritis Rheum.
31
:
315
-324.
19
Gutierrez, O., C. Pipaon, N. Inohara, A. Fontalba, Y. Ogura, F. Prosper, G. Nunez, J. L. Fernandez-Luna.
2002
. Induction of Nod2 in myelomonocytic and intestinal epithelial cells via nuclear factor-κB activation.
J. Biol. Chem.
277
:
41701
-41705.
20
Inohara, N., Y. Ogura, A. Fontalba, O. Gutierrez, F. Pons, J. Crespo, K. Fukase, S. Inamura, S. Kusumoto, M. Hashimoto, et al
2003
. Host recognition of bacterial muramyl dipeptide mediated through NOD2: implications for Crohn’s disease.
J. Biol. Chem.
278
:
5509
-5512.
21
Inohara, N., Y. Ogura, F. F. Chen, A. Muto, G. Nunez.
2001
. Human Nod1 confers responsiveness to bacterial lipopolysaccharides.
J. Biol. Chem.
276
:
2551
-2554.
22
Benito, A., M. Silva, D. Grillot, G. Nunez, J. Fernandez-Luna.
1996
. Apoptosis induced by erythroid differentiation of human leukemia cell lines is inhibited by Bcl-xL.
Blood
87
:
3837
-3843.
23
May, M., S. Ghosh.
1998
. Signal transduction through NF-κB.
Immunol. Today
19
:
80
-88.
24
Pinals, R. S., A. T. Masi, R. A. Larsen.
1981
. Preliminary criteria for clinical remission in rheumatoid arthritis.
Arthritis Rheum.
24
:
1308
-1315.
25
Gonzalez-Alvaro, I., L. Carmona, A. Balsa, R. Sanmarti, M. A. Belmonte, X. Tena.
2003
. Patterns of disease modifying antirheumatic drug use in a Spanish cohort of patients with rheumatoid arthritis.
J. Rheumatol.
30
:
697
-704.
26
Welsing, P. M., A. M. van Gestel, H. L. Swinkels, L. A. Kiemeney, P. L. van Riel.
2001
. The relationship between disease activity, joint destruction, and functional capacity over the course of rheumatoid arthritis.
Arthritis Rheum.
44
:
2009
-2017.
27
Fransen, J., P. L. van Riel.
2005
. The Disease Activity Score and the EULAR response criteria.
Clin. Exp. Rheumatol.
23
:
S93
-S99.
28
Ghosh, S., M. J. May, E. B. Kopp.
1998
. NF-κB and Rel proteins: evolutionarily conserved mediators of immune responses.
Annu. Rev. Immunol.
16
:
225
-260.
29
Courtois, G., T. D. Gilmore.
2006
. Mutations in the NF-κB signaling pathway: implications for human disease.
Oncogene
25
:
6831
-6843.
30
Courtois, G., A. Smahi, J. Reichenbach, R. Doffinger, C. Cancrini, M. Bonnet, A. Puel, C. Chable-Bessia, S. Yamaoka, J. Feinberg, et al
2003
. A hypermorphic IκBα mutation is associated with autosomal dominant anhidrotic ectodermal dysplasia and T cell immunodeficiency.
J. Clin. Invest.
112
:
1108
-1115.
31
Janssen, R., A. van Wengen, M. A. Hoeve, M. ten Dam, M. van der Burg, J. van Dongen, E. van de Vosse, M. van Tol, R. Bredius, T. H. Ottenhoff, et al
2004
. The same IκBα mutation in two related individuals leads to completely different clinical syndromes.
J. Exp. Med.
200
:
559
-568.
32
Smahi, A., G. Courtois, P. Vabres, S. Yamaoka, S. Heuertz, A. Munnich, A. Israel, N. S. Heiss, S. M. Klauck, P. Kioschis, et al
2000
. Genomic rearrangement in NEMO impairs NF-κB activation and is a cause of incontinentia pigmenti. The International Incontinentia Pigmenti (IP) Consortium.
Nature
405
:
466
-472.
33
Makarov, S. S..
2001
. NF-κB in rheumatoid arthritis: a pivotal regulator of inflammation, hyperplasia, and tissue destruction.
Arthritis Res.
3
:
200
-206.
34
Marok, R., P. G. Winyard, A. Coumbe, M. L. Kus, K. Gaffney, S. Blades, P. I. Mapp, C. J. Morris, D. R. Blake, C. Kaltschmidt, P. A. Baeuerle.
1996
. Activation of the transcription factor nuclear factor-κB in human inflamed synovial tissue.
Arthritis Rheum.
39
:
583
-591.
35
van der Heijde, D. M., P. L. van Riel, M. A. van Leeuwen, M. A. van ’t Hof, M. H. van Rijswijk, L. B. van de Putte.
1992
. Prognostic factors for radiographic damage and physical disability in early rheumatoid arthritis: a prospective follow-up study of 147 patients.
Br. J. Rheumatol.
31
:
519
-525.
36
McGovern, D. P., H. Butler, T. Ahmad, M. Paolucci, D. A. van Heel, K. Negoro, P. Hysi, J. Ragoussis, S. P. Travis, L. R. Cardon, D. P. Jewell.
2006
. TUCAN (CARD8) genetic variants and inflammatory bowel disease.
Gastroenterology
131
:
1190
-1196.
37
Seetharaman, R., A. L. Mora, G. Nabozny, M. Boothby, J. Chen.
1999
. Essential role of T cell NF-κB activation in collagen-induced arthritis.
J. Immunol.
163
:
1577
-1583.
38
Miagkov, A. V., D. V. Kovalenko, C. E. Brown, J. R. Didsbury, J. P. Cogswell, S. A. Stimpson, A. S. Baldwin, S. S. Makarov.
1998
. NF-κB activation provides the potential link between inflammation and hyperplasia in the arthritic joint.
Proc. Natl. Acad. Sci. USA
95
:
13859
-13864.