Download pdf - Association of NLRP3 and CARD8 genetic polymorphisms with juvenile idiopathic arthritis in a Taiwanese population

Association of NLRP3 and CARD8 genetic polymorphisms with juvenileidiopathic arthritis in a Taiwanese population

C-A Yang1, S-T Huang1, B-L Chiang1,2

Departments of 1Paediatrics and 2Medical Research, National Taiwan University Hospital, Taipei, Taiwan

Objectives: An elevated interleukin (IL)-1β response in peripheral blood mononuclear cells (PBMCs) has been observedin systemic juvenile idiopathic arthritis (sJIA), suggesting a role for inflammasomes in the pathogenesis of JIA. We aimedto determine whether genetic polymorphisms of the NLRP3 inflammasome components confer risk for oligoarticular andpolyarticular JIA in a Taiwanese population.Method: A total of 118 JIA patients and 103 healthy controls were genotyped for rs4353135 OR2B11/NLRP3 andrs2043211 CARD8 polymorphisms. Clinical laboratory data and serum IL-1β of JIA patients were evaluated by medicalchart review and enzyme-linked immunosorbent assay (ELISA), respectively. The production of IL-17 in lymphocytes ofdifferent genotype carriers was measured using flow cytometry.Results: The variant rs4353135 G allele carrier conferred increased risk for oligoarticular and polyarticular JIA. TheG allele was also found to be associated with higher levels of clinical inflammatory markers. Moreover, G variant carriersenhanced the lymphocyte IL-17 response. The G/G genotype further increased the need for treatment with the tumournecrosis factor (TNF) inhibitor etanercept.Conclusions: Our data indicate that the rs4353135 OR2B11/NLRP3 polymorphism might be functional in, and couldcontribute to, the pathophysiology of oligoarticular and polyarticular JIA in a Taiwanese population.

Juvenile idiopathic arthritis (JIA) is the most commondebilitating chronic disease of childhood. It is an auto-immune disease with heterogeneous subtypes and com-plex immunopathology. Specific human leucocyteantigen (HLA) alleles have been reported to be associatedwith susceptibility to JIA (1, 2). Non-HLA genetic asso-ciations include interleukin-2 receptor subunit alpha(IL2RA), protein tyrosine phosphatase, non-receptortype 22 (PTPN22), tumour necrosis factor alpha(TNF-α), and macrophage migration inhibitory factor(MIF) (3–7). Elevated plasma levels of TNF and MIFhave been reported in patients with JIA, irrespective ofsubtype (8). Furthermore, enhanced IL-1β response hasbeen observed in systemic JIA (sJIA) (9), and the IL-1antagonist had shown efficacy in treating sJIA (10, 11).The involvement of IL-1 suggests that inflammasomesmay play a role in the pathogenesis of JIA.Genetic polymorphisms in IL-6, IL-18, and the IL-1

family have been demonstrated to be associated with sus-ceptibility to different subtypes of JIA (12–14). However,whether genetic variations of inflammasomes contribute torisk for JIA remains unknown. The NOD-like receptor

family, pyrin domain containing 3 (NLRP3) and an adaptorprotein containing caspase recruitment domain (CARD) areessential components of the NLRP3 inflammasome, whichregulates caspase 1-mediated IL-1β activation (15, 16). It hasbeen demonstrated that the rs4353135 polymorphism,which is located between the NLRP3 and OR2B11 genes,affects the expression of NLRP3 and susceptibility toanother autoinflammatory disease, Crohn’s disease (CD)(17). Furthermore, combined genetic polymorphisms ofNLRP3 and CARD8were reported to influence the suscept-ibility and severity of adult rheumatoid arthritis (RA) (18). Itis also likely that functional genetic variations ofNLRP3 andCARD8 play a role in JIA pathogenesis. In our study, wecompared the genotype frequencies of rs4353135 (NLRP3/OR2B11) and rs2043211 (CARD8) variants between JIApatients and healthy controls. The impact of these two singlenucleotide polymorphisms (SNPs) on levels of inflamma-tory markers and treatment responses was also evaluated.

Method

Subjects

A total of 118 JIA patients diagnosed and treated at theNational Taiwan University Hospital between 2004 and2011 were enrolled in this study. The diagnosis and classi-fication of JIA was made according to the InternationalLeague of Associations for Rheumatology (ILAR) revised

Bor-Luen Chiang, Department of Paediatrics, National TaiwanUniversity Hospital, 7 Chung-Shan South Road, Taipei 100, Taiwan.E-mail: [email protected]

Accepted 12 August 2013

146 Scand J Rheumatol 2014;43:146–152

© 2014 Informa Healthcare on license from Scandinavian Rheumatology Research Foundation

DOI: 10.3109/03009742.2013.834962www.scandjrheumatol.dk

Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Im

peri

al C

olle

ge L

ondo

n on

06/

14/1

4Fo

r pe

rson

al u

se o

nly.

criteria for JIA (19). We have included 22 patients withsJIA, 70 with oligoarticular JIA (both persistent andextended oligoarticular), and 26 with polyarticular JIA[both rheumatoid factor (RF) positive and RF negativepolyarticular]. The categories of psoriatic arthritis andenthesitis-related arthritis were not recruited. HLA-B27 iswell known to be highly associated with enthesitis-relatedJIA, and therefore all patients with positive HLA-B27 wereexcluded. Patient profiles are presented in Table 1. Werecruited 103 age- and sex-matched healthy controls fromthe National Taiwan University Hospital.DNA was extracted from whole blood of these subjects.

Serum was collected from 24 of the 96 oligoarticular/poly-articular JIA children at the time of disease remission (atclinical follow-up). Medical records were examined retro-spectively to obtain the initial clinical data with regard to C-reactive protein (CRP), erythrocyte sedimentation rate(ESR), age of onset, and the requirement of TNF-blocker(Enbrel) treatment. The study was approved by theInstitutional Ethical Committees. Informed consent wasobtained from parents or directly from individuals if theywere aged > 15 years, according to the Declaration ofHelsinki (at the General Assembly in October 2008).

Genotyping

Genotyping was performed using genomic DNA isolatedfrom whole blood according to standard procedures, andfluorescence-labelled hybridization fluorescence resonanceenergy transfer (FRET) probes followed by melting curveanalysis as described previously (20, 21). Primers used forNLRP3/OR2B11 (rs4353135) forward and reverse wereGTTTCTTTTCAGAGCCTAAACTGG and TTGCTGAGATATTAAGGCAACATCA, respectively. Primers usedfor CARD8 (rs2043211) forward and reverse were AGCTACCCTGTGTTTCTGAGACC and ATTCATTCTCCCCTGAGTTCG, respectively. Probes for melting curveanalysis were: NLRP3/OR2B11 sensor: GCCGCATACATTTACCCCTC-FL, NLRP3/OR2B11 anchor: Red640-TCTTTCTTGCTTCC TTCATTCTCTCATTTCT, CARD8sensor: AGCACGG ATCA ATAATGGCTC-FL, CARD8anchor: Red640-C C TCTG TCTCATCATCTTCTTGGAAAAAATGT.The PCR reaction mixture included 2 μL DNA (2–10

ng/μL), 0.5 μM of forward and reverse primers, and 0.2μM of each fluorescence probe. PCR was performedon LightCycler1.5 (Roche Diagnostics, Mannheim,Germany). The PCR parameters for the rs4353135 and

rs2043211 polymorphisms were as follows: denaturationat 95˚C for 10 min, then 45 cycles of denaturation (95˚Cfor 15 s), annealing (62˚C for 10 s), and extension (72˚Cfor 10 s). Next, melting curve analysis was performed: 1cycle at 47˚C for 30 s, followed by an increase in tem-perature to 95˚C at a slope of 0.1˚C/s. Melting curveswere converted to melting peaks by plotting the secondnegative derivative of fluorescence with respect to tem-perature. A single peak at higher temperature represents ahomozygously mutated genotype (20, 21). A templatenegative control was included in each run. Duplicateverification was performed in 30 samples.

The genotype frequencies of the rs4353135 andrs2043211 polymorphisms in the controls showed nodeviation from Hardy–Weinberg equilibrium.

Enzyme-linked immunosorbent assay (ELISA)

Wemeasured serum levels of IL-1β, the effector cytokine ofthe inflammasome pathway, in 24 genotyped oligoarticular/polyarticular JIA patients using a commercial human IL-1βELISA kit (R&D Systems, Minneapolis, MN, USA).

Peripheral blood mononuclear cell (PBMC) stimulationassay

PBMCs (1 � 106) of different SNP genotypes were incu-batedwith completeRPMImediumonly or stimulatedwithphorbol 12-myristate 13-acetate (PMA) 100 ng/mL andionomycin 1 μg/mL (both purchased from Sigma, StLouis, MO, USA) for 24 h. Brefeldin A 7.5 μg/mL(Sigma) was added after 1 h of stimulation. Fluorescent-conjugatedmonoclonal antibodies against CD4 (BD, cloneRPA-T4), CD45RA (BioLegend, San Diego, CA, USA;clone HI30), and IL-17 (BD Biosciences, San Jose, CA,USA, clone N49-653) were used to stain intracellular pro-duction of IL-17 in CD4þCD45RA– cells. Cells weregated from the lymphocyte gate on a BD FACS Cantoflow cytometer, and were analysed using FACS Diva soft-ware version 6.0.

Statistical analysis

Fisher’s exact test was used to compare the percentage ofvariant allele carriers between JIA and healthy controls, andto compare allele frequencies between JIA with and withouthigh inflammation markers. The percentage of homozygous

Table 1. Clinical characteristics of JIA patients (n ¼ 118).

Onset type

Oligoarticular Polyarticular Systemic Total

Female 31 14 13 58Male 39 12 9 60Age of onset (years), median (range) 11 (2–16) 8 (2–16) 7.5 (2–16) 10 (2–16)

Inflammasome SNPs and JIA 147

www.scandjrheumatol.dk

Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Im

peri

al C

olle

ge L

ondo

n on

06/

14/1

4Fo

r pe

rson

al u

se o

nly.

variant allele carriers in JIA treated with or without etaner-cept was also analysed by Fisher’s exact test. The serumlevels of IL-1β in JIA patients of different genotypes werecompared by using the Kruskal–Wallis test. Delta meanfluorescent intensity (MFI) of IL-17 production in responseto PMA/ionomycin was compared between variant andnon-variant allele carriers by the Mann–Whitney U-test.

Results

The rs4353135 variant G allele carrier is associated withincreased risk for oligoarticular/polyarticular JIA

According to previous reports (7) and our review (22), theoligoarticular subtype shares molecular similarities withpolyarticular JIA. Therefore, we grouped these two sub-types together in this study. The percentage of rs4353135

variant G allele carrier was significantly higher in theoligoarticular/polyarticular group than in the healthy con-trols [odds ratio (OR) 2.127; 95% confidence interval (CI)1.00–4.54, p¼ 0.036; Table 2]. However, there was only atrend of a higher percentage of rs2043211 variant allelecarrier in oligoarticular/polyarticular JIA patients (OR1.779, 95% CI 0.96–3.30, p ¼ 0.046; Table 2). The sJIAgroup showed no difference in percentages of both SNPcarriers compared with the controls (Table 2).

The rs4353135 variant G allele is associated withhigher clinical inflammatory parameters in oligoarticular/polyarticular JIA

We also investigated whether the NLRP3 and CARD8SNPs contributed to raised levels of CRP and ESR,

Table 2. Differences in genotype and allele frequencies of rs4353135 NLRP3 and rs2043211 CARD8 polymorphisms between healthycontrols and patients with juvenile idiopathic arthritis (JIA).

Genotype (%) Allele frequency

rs4353135 n T/T T/G G/G G (%) OR (95% CI) p-value*

Healthy controls 103 24 (23.3) 54 (52.4) 25 (24.3) 50.5Systemic JIA 22 2 (9.1) 15 (68.2) 5 (22.7) 56.8 2.582 (0.56–11.99) 0.174Oligo-/polyarticular JIA 96 12 (12.5) 55 (57.3) 29 (30.2) 58.9 2.127 (1.00–4.54) 0.036

Genotype (%) Allele frequency

rs2043211 n T/T T/A A/A G (%) OR (95% CI) p-value*

Healthy controls 103 37 (35.9) 44 (42.7) 22 (21.4) 42.7Systemic JIA 22 9 (40.9) 10 (45.5) 3 (13.6) 36.4 0.810 (0.32–2.07) 0.417Oligo-/polyarticular JIA 96 23 (24.0) 59 (61.5) 14 (14.5) 45.3 1.779 (0.96–3.30) 0.046

OR, Odds ratio; CI, confidence interval.* Calculated by Fisher’s exact test comparing genotype frequencies of wild type vs. heterozygousþ homozygous variant allele carriers(rs4353135: T/T vs. T/GþG/G; rs2043211: T/T vs. T/AþA/A).

Table 3. Relationship between rs4353135 NLRP3 and rs2043211 CARD8 polymorphisms and CRP levels in patients with juvenileidiopathic arthritis (JIA).

Genotype (%)

rs4353135 n T/T T/G G/G G allele (%) OR (95% CI) p-value*

Systemic JIACRP � 4 mg/dL 17 1 (5.9) 11 (64.7) 5 (29.4) 61.7 2.432 (0.57–10.25) 0.195CRP < 4 mg/dL 5 1 (20) 4 (80) 0 (0) 40

Oligo-/polyarticular JIACRP � 4 mg/dL 10 1 (10) 1 (10) 8 (80) 85 4.486 (1.27–15.88) 0.009CRP < 4 mg/dL 86 11 (12.8) 54 (62.8) 21 (24.4) 55.8

Genotype (%)

rs2043211 n T/T T/A A/A A allele (%) OR (95% CI) p-value*

Systemic JIACRP � 4 mg/dL 16 6 (37.5) 7 (43.7) 3 (18.8) 40.6 2.053 (0.46–9.07) 0.276CRP < 4 mg/dL 6 3 (50) 3 (50) 0 (0) 25

Oligo-/polyarticular JIACRP � 4 mg/dL 13 1 (10) 9 (10) 3 (80) 85 1.780 (0.77–4.11) 0.125CRP < 4 mg/dL 83 22 (12.8) 50 (62.8) 11 (24.4) 55.8

OR, Odds ratio; CI, confidence interval; CRP, C-reactive protein.* Calculated by Fisher’s exact test comparing variant allele (G allele for rs4353135; A allele for rs2043211) frequencies between patientswith higher CRP ( � 4mg/dL) and lower CRP levels.

148 C-A Yang et al


Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Im

peri

al C

olle

ge L

ondo

n on

06/

14/1

4Fo

r pe

rson

al u

se o

nly.

which are the most common clinical tests in evaluatinginflammation status of autoimmune diseases. Thers4353135 G allele frequency was found to be higher inoligoarticular/polyarticular JIA with elevated inflamma-tory markers at initial disease presentation (CRP � 4 mg/dL vs. CRP < 4 mg/dL, OR 4.486, 95% CI 1.27–15.88,p¼ 0.009, Table 3; ESR� 20 mm/h vs. ESR < 20 mm/h,OR 1.879, 95% CI 1.01–3.51, p ¼ 0.034, Table 4). Bycontrast, the rs2043211 variant A allele frequency wassimilar between oligoarticular/polyarticular JIA patientswith and without higher CRP/ESR. As for the sJIA group,no significant difference in NLRP3 and CARD8 SNPfrequencies was detected in patients with low and high

inflammatory markers. Of note, all sJIA patients had anincreased ESR level at the first clinical visit (Table 4).

The rs4353135 variant G allele is associatedwith enhancedIL-17 response in lymphocytes

As IL-1β is the effector cytokine of inflammasomes, weevaluated the level of IL-1β in serum of 24 out of 96oligoarticular/polyarticular JIA patients at the time of clin-ical follow-up (at disease remission; duration of disease andtreatment was between 1 and 2 years). However, the med-ian IL-1β level was zero in each rs4353135 genotypecarrier (Kruskal–Wallis test p > 0.05, Figure 1A).

Table 4. Relationship between rs4353135 NLRP3 and rs2043211 CARD8 polymorphisms and ESR levels in patients with juvenileidiopathic arthritis (JIA)

Genotype (%)


Systemic JIAESR � 20 mm/h 22 2 (9.1) 15 (68.2) 5 (22.7) 56.8 NA NAESR < 20 mm/h 0 0 (0) 0 (0) 0 (0) 0

Oligo-/polyarticular JIAESR � 20 mm/h 58 5 (8.6) 32 (55.2) 21 (36.2) 63.8 1.879 (1.01–3.51) 0.034ESR < 20 mm/h 31 7 (22.6) 18 (58.1) 6 (19.3) 48.4

Genotype (%)


Systemic JIAESR � 20 mm/h 22 9 (9.1) 10 (68.2) 3 (22.7) 56.8 NA NAESR < 20 mm/h 0 0 (0) 0 (0) 0 (0) 0

Oligo-/polyarticular JIAESR � 20 mm/h 57 11 (19.3) 37 (64.9) 9 (15.8) 48.2 1.199 (0.65–2.22) 0.338ESR < 20 mm/h 32 9 (28.1) 18 (56.3) 5 (15.6) 43.8

OR, Odds ratio; CI, confidence interval; ESR, erythrocyte sedimentation rate; NA, not available.* Calculated by Fisher’s exact test comparing variant allele (G allele for rs4353135; A allele for rs2043211) frequencies between patientswith higher ESR ( � 20 mm/h) and lower ESR levels.

A B **

5000

4000

3000

2000

1000

0

n.s.

IL-1

β (p

g/m

L)

IL-1

7 (d

elta

MFI

)

15

10

5

0

T/T T/G G/G T/T T/G G/G

Figure 1. Influence of the rs4353135 polymorphism on IL-1β and IL-17 production. (A) Serum IL-1βwas detected in rs4353135 T/T (n¼ 5), T/G (n¼12), and G/G (n¼ 7) genotype carriers of oligoarticular/polyarticular JIA patients at disease remission using an ELISA. Lines represent medians. n.s., notsignificant by the Kruskal–Wallis test. (B) IL-17 production measured in CD4þCD45RA– lymphocytes derived from control subjects of differentrs4353135 genotypes using flow cytometry. PBMCswere stimulated with PMA/ionomycin for 24 h; the response to medium alone was subtracted. Linesrepresent medians. ** p < 0.01 calculated by the Mann–Whitney U-test comparing IL-17 production of T/T vs. T/GþG/G.



Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Im

peri

al C

olle

ge L

ondo

n on

06/

14/1

4Fo

r pe

rson

al u

se o

nly.

Because of the recently identified role of IL-17 in JIA(reviewed by Miossec (23)), we further investigatedwhether PMA/ionomycin-stimulated IL-17 productionwould be altered by these two SNPs in PBMCs of thecontrols. Remarkably, the rs4353135 G allele carriersshowed significantly higher IL-17 response in theirCD4þCD45RA–lymphocytes (median delta MFI ofIL-17 in T/T vs. T/GþG/G ¼ 2308 vs. 1077, Mann–Whitney U-test p < 0.01, Figure 1B).

The homozygous rs4353135 variant allele carrier isassociated with the need for etanercept treatmentin oligoarticular/polyarticular JIA

Etanercept has been used as a second-line agent for JIA (24,25). According to the National Health Insurance guidelinesin Taiwan, the use of etanercept is limited to systemic,polyarticular, and extended oligoarticular JIA patients whofailed to respond to 3 months of treatment with an adequatedose of methotrexate (MTX)/prednisolone. We comparedthe genotype percentages between etanercept and non-etanercept users in JIA. The percentage of homozygousrs4353135 G allele genotype was found to be higher inoligoarticular/polyarticular JIA patients on etanercept (OR2.672, 95% CI 1.04–6.88, p¼ 0.036, Table 5). The G allelehomozygosity, however, did not affect the etanerceptrequirement in sJIA patients. Furthermore, the homozygousrs2043211 A allele carrier was not associated with the needfor etanercept treatment in both sJIA and oligoarticular/polyarticular JIA groups (Table 5).

Discussion

Our study reveals for the first time that carriers of thers4353135 G allele in a Taiwanese population have

increased risk for oligoarticular/polyarticular JIA. Inoligoarticular/polyarticular JIA, the G allele is furtherassociated with higher CRP and ESR levels at initialdisease presentation, and the G allele homozygosity cor-relates with the need for etanercept treatment. By contrast,rs4353135 and rs2043211 SNPs were not associated withsusceptibility, inflammatory status, and etanercept usagein sJIA patients.

It has been shown that the monocyte NLRP3 expres-sion level increases in the rs4353135 G/T genotype, andis highest in the G/G carrier (17). Consistently, we foundthat lymphocytes of the G allele carrier produced moreIL-17 than the wild type in response to PMA/ionomycinstimulation. Of note, in oligoarticular/polyarticular JIA,the G variant is associated with raised inflammatorymarkers, but only the homozygous G/G genotypeincreases the requirement of Enbrel. It could be thatthe treatment response of oligoarticular/polyarticularJIA correlates not only with proinflammatory cytokinesbut also with the degree of NLRP3 expression. AlthoughIL-1β is the effector cytokine of NLRP3 inflamma-somes, a previous study (17) did not detect a differencein monocyte IL-1β secretion among rs4353135 geno-types. In our study, the barely detected IL-1β in JIAserum might be related to the time of blood sampling.The low IL-1β level at disease remission could masksubtle differences. Alternatively, the differential influ-ence of rs4353135 on IL-17 and IL-1β production mightindicate distinct underlying mechanisms.

In contrast to our findings, the ‘hypo-inflammatory’rs4353135 T variant has been reported to be associatedwith risk for CD in European descendants (17). However,this association could not be replicated in a large UK cohortanalysing 1298 cases (26). In fact, other genetic polymor-phisms in NLRP1 and NLRP3 have been reported to

Table 5. Associations of rs4353135 NLRP3 and rs2043211 CARD8 polymorphisms with the need for etanercept treatment in juvenileidiopathic arthritis (JIA).

Genotype (%)


Systemic JIAEtanercept (þ) 17 2 (11.8) 11 (64.7) 4 (23.5) 55.9 1.231 (0.11–14.43) 0.687Etanercept (–) 5 0 (0) 4 (80) 1 (20) 60

Oligo-/polyarticular JIAEtanercept (þ) 26 3 (11.5) 11 (42.3) 12 (46.2) 67.3 2.672 (1.04–6.88) 0.036Etanercept (–) 70 9 (12.8) 44 (62.9) 17 (24.3) 55.7

Genotype (%)


Systemic JIAEtanercept (þ) 17 7 (41.2) 7 (41.2) 3 (17.6) 38.2 2.655 (0.12–60.27) 0.442Etanercept (–) 5 2 (40) 3 (80) 0 (0) 30

Oligo-/polyarticular JIAEtanercept (þ) 26 10 (38.5) 13 (50) 3 (11.5) 36.5 0.700 (0.18–2.74) 0.439Etanercept (–) 70 13 (18.6) 46 (65.7) 11 (15.7) 48.6

OR, Odds ratio; CI, confidence interval.* Calculated by Fisher’s exact test comparing percentages of homozygous and non-homozygous variant allele carriers between thetwo groups (with or without the usage of etanercept).

150 C-A Yang et al


Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Im

peri

al C

olle

ge L

ondo

n on

06/

14/1

4Fo

r pe

rson

al u

se o

nly.

increase NLRP mRNA expression and enhance IL-1β andIL-18 production (27, 28). These SNPs were shown toconfer risk for autoimmune diseases, such as type 1 dia-betes, RA, and systemic lupus erythematosus (SLE) (28–30). Our finding of the association between the ‘hyper-inflammatory’ rs4353135 G variant and risk for oligoarti-cular/polyarticular JIA is in accordance with these reports.In our study, the oligoarticular/polyarticular JIA patients

showed only a trend of a higher percentage of the rs2043211variant allele carriers. TheCARD8 rs2043211 (C10X) poly-morphism, which results in a truncated protein, has beendemonstrated to confer susceptibilities to CD and RA whenoccurring in combination with another gain-of-functionNLRP3 SNP, rs35829419 (18, 31). However, a Spanishstudy including 1530 RA patients failed to verify the asso-ciation with the rs2043211 SNP (32). Furthermore, unlikeEuropeans, the rs35829419 minor allele is almost absent inthose of Chinese descent.We did not detect any rs35829419SNP carrier in 40 patients from our JIA study group in apreliminary experiment (data not shown). Whether theCARD8 variant, together with other SNPs in the inflamma-some component, influences the risk of autoimmune dis-eases in a Taiwanese population requires furtherinvestigation. It has been reported that the rs2043211 variantallele is associated with elevated ESR, disease severity, andthe need for anti-TNF therapy in RA patients (18, 33).However, we did not detect an association of the CARD8SNP with elevated inflammatory markers or etanerceptusage in JIA, suggesting a different aetiology betweenadult and juvenile autoimmune arthritis.Of note, although the percentage of rs4353135 G allele

carrier is also high in sJIA patients, it is not statisticallydifferent when compared with healthy controls. In addi-tion, the sJIA patients all presented with high ESR levels,irrespective of rs4353135 or rs2043211 genotype. Unlikethe results for oligoarticular/polyarticular JIA, rs4353135G allele homozygosity is not associated with anti-TNFtreatment in sJIA. It had been proposed that the patho-genesis of oligoarticular/polyarticular JIA and sJIA aredifferent: whereas oligoarticular/polyarticular JIA is anantigen-driven autoimmune disease involving an imba-lance of Th1/Th17 and regulatory T (Treg) cells (22, 34),sJIA is more of an autoinflammatory disease, producinghigher amounts of proinflammatory cytokines includingIL-1β (8, 9). Increased plasma IL-17 had been observed inactive oligoarticular/polyarticular JIA but not in sJIApatients (8). Furthermore, joint IL-17þT cell numberswere shown to be related to disease severity of oligoarti-cular JIA (34). Our finding that the rs4353135 variantallele enhanced lymphocyte IL-17 response lends supportsto a role for Th17 in oligoarticular/polyarticular JIA. Moreresearch is needed to evaluate the contribution of thers4353135 polymorphism to IL-1β production in activeoligoarticular/polyarticular JIA.In conclusion, our results suggest that the rs4353135

polymorphism in NLRP3/OR2B11 enhanced lymphocyte

IL-17 production, which might contribute to the pathophy-siology of oligoarticular/polyarticular JIA. Because of therelatively small sample size, replication studies on largerindependent cohorts including all JIA subtypes are required.

Acknowledgements

This study was supported by Research Collaboration Funding of theNational Taiwan University Hospital and Cathay General Hospital,project no. 101-CGN04.

References1. Fernandez-Vina M, Fink CW, Stastny P. HLA associations in

juvenile arthritis. Clin Exp Rheumatol 1994;12:205–14.2. Thomson W, Barrett JH, Donn R, Pepper L, Kennedy LJ, Ollier

WE, et al. Juvenile idiopathic arthritis classified by the ILARcriteria: HLA associations in UK patients. Rheumatology (Oxford)2002;41:1183–9.

3. Hinks A, Ke X, Barton A, Eyre S, Bowes J, Worthington J, et al.Association of the IL2RA/CD25 gene with juvenile idiopathicarthritis. Arthritis Rheum 2009;60:251–7.

4. Viken MK, Amundsen SS, Kvien TK, Boberg KM, Gilboe IM,Lilleby V, et al. Association analysis of the 1858C>T polymorphismin the PTPN22 gene in juvenile idiopathic arthritis and otherautoimmune diseases. Genes Immun 2005;6:271–3.

5. Schmeling H, Wagner U, Peterson A, Horneff G. Tumor necrosisfactor alpha promoter polymorphisms in patients with juvenileidiopathic arthritis. Clin Exp Rheumatol 2006;24:103–8.

6. Donn R, Alourfi Z, Zeggini E, Lamb R, Jury F, Lunt M, et al. Afunctional promoter haplotype of macrophage migration inhibitoryfactor is linked and associated with juvenile idiopathic arthritis.Arthritis Rheum 2004;50:1604–10.

7. Macaubas C, Nguyen K, Milojevic D, Park JL, Mellins ED.Oligoarticular and polyarticular JIA: epidemiology and pathogenesis.Nat Rev Rheumatol 2009;5:616–26.

8. de Jager W, Hoppenreijs EP, Wulffraat NM, Wedderburn LR, KuisW, Prakken BJ. Blood and synovial fluid cytokine signatures inpatients with juvenile idiopathic arthritis: a cross-sectional study.Ann Rheum Dis 2007;66:589–98.

9. Pascual V, Allantaz F, Arce E, Punaro M, Banchereau J. Role ofinterleukin-1 (IL-1) in the pathogenesis of systemic onset juvenileidiopathic arthritis and clinical response to IL-1 blockade. J ExpMed 2005;201:1479–86.

10. Quartier P, Allantaz F, Cimaz R, Pillet P, Messiaen C, Bardin C,et al. A multicentre, randomised, double-blind, placebo-controlledtrial with the interleukin-1 receptor antagonist anakinra in patientswith systemic-onset juvenile idiopathic arthritis (ANAJIS trial).Ann Rheum Dis 2011;70:747–54.

11. Ruperto N, Quartier P, Wulffraat N, Woo P, Ravelli A, Mouy R,et al. A phase II, multicenter, open-label study evaluating dosingand preliminary safety and efficacy of canakinumab in systemicjuvenile idiopathic arthritis with active systemic features. ArthritisRheum 2012;64:557–67.

12. McDowell TL, Symons JA, Ploski R, Forre O, Duff GW. A geneticassociation between juvenile rheumatoid arthritis and a novelinterleukin-1 alpha polymorphism. Arthritis Rheum 1995;38:221–8.

13. Fishman D, Faulds G, Jeffery R, Mohamed-Ali V, Yudkin JS,Humphries S, et al. The effect of novel polymorphisms in theinterleukin-6 (IL-6) gene on IL-6 transcription and plasma IL-6levels, and an association with systemic-onset juvenile chronicarthritis. J Clin Invest 1998;102:1369–76.

14. Sugiura T, Maeno N, Kawaguchi Y, Takei S, Imanaka H, KawanoY, et al. A promoter haplotype of the interleukin-18 gene isassociated with juvenile idiopathic arthritis in the Japanesepopulation. Arthritis Res Ther 2006;8:R60.



Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Im

peri

al C

olle

ge L

ondo

n on

06/

14/1

4Fo

r pe

rson

al u

se o

nly.

15. Agostini L, Martinon F, Burns K, McDermott MF, Hawkins PN,Tschopp J. NALP3 forms an IL-1beta-processing inflammasomewith increased activity inMuckle-Wells autoinflammatory disorder.Immunity 2004;20:319–25.

16. Martinon F, Tschopp J. Inflammatory caspases: linking anintracellular innate immune system to autoinflammatory diseases.Cell 2004;117:561–74.

17. Villani AC, Lemire M, Fortin G, Louis E, Silverberg MS, ColletteC, et al. Common variants in the NLRP3 region contribute toCrohn’s disease susceptibility. Nat Genet 2009;41:71–6.

18. KastbomA, Verma D, Eriksson P, Skogh T,Wingren G, SoderkvistP. Genetic variation in proteins of the cryopyrin inflammasomeinfluences susceptibility and severity of rheumatoid arthritis (theSwedish TIRA project). Rheumatology (Oxford) 2008;47:415–17.

19. Petty RE, Southwood TR, Manners P, Baum J, Glass DN,Goldenberg J, et al. International League of Associations forRheumatology classification of juvenile idiopathic arthritis:second revision, Edmonton, 2001. J Rheumatol 2004;31:390–2.

20. Hamann L, Hamprecht A, Gomma A, Schumann RR. Rapid andinexpensive real-time PCR for genotyping functional polymorphismswithin the Toll-like receptor -2, -4, and -9 genes. J Immunol Methods2004;285:281–91.

21. Johnson CM, Lyle EA, Omueti KO, Stepensky VA, Yegin O,Alpsoy E, et al. Cutting edge: a common polymorphism impairscell surface trafficking and functional responses of TLR1 butprotects against leprosy. J Immunol 2007;178:7520–4.

22. Lin YT, Wang CT, Gershwin ME, Chiang BL. The pathogenesis ofoligoarticular/polyarticular vs systemic juvenile idiopathic arthritis.Autoimmun Rev 2011;10:482–9.

23. Miossec P. Interleukin-17 and Th17 cells: from adult to juvenilearthritis - now it is serious! Arthritis Rheum 2011;63:2168-71.

24. van Luijn JC, Danz M, Bijlsma JW, Gribnau FW, Leufkens HG.Post-approval trials of new medicines: widening use or deepeningknowledge? Analysis of 10 years of etanercept. Scand J Rheumatol2011;40:183–91.

25. Iannone F, Fanizzi R, Scioscia C, Anelli MG, Lapadula G. Bodymass does not affect the remission of psoriatic arthritis patients onanti-TNF-alpha therapy. Scand J Rheumatol 2013;42:41–4.

26. Lewis GJ, Massey DC, Zhang H, Bredin F, Tremelling M, Lee JC,et al. Genetic association between NLRP3 variants and Crohn’sdisease does not replicate in a large UK panel. Inflamm Bowel Dis2011;17:1387–91.

27. Verma D, Sarndahl E, Andersson H, Eriksson P, Fredrikson M,Jonsson JI, et al. The Q705K polymorphism in NLRP3 is a gain-of-function alteration leading to excessive interleukin-1beta and IL-18production. PLoS One 2012;7:e34977.

28. Sui J, Li H, Fang Y, Liu Y, Li M, Zhong B, et al. NLRP1 genepolymorphism influences gene transcription and is a risk factorfor rheumatoid arthritis in Han Chinese. Arthritis Rheum 2012;64:647–54.

29. Pontillo A, Brandao L, Guimaraes R, Segat L, Araujo J, Crovella S.Two SNPs in NLRP3 gene are involved in the predisposition totype-1 diabetes and celiac disease in a pediatric population fromnortheast Brazil. Autoimmunity 2010;43:583–9.

30. Pontillo A, Girardelli M, Kamada AJ, Pancotto JA, Donadi EA,Crovella S, et al. Polymorphisms in inflammasome genes areinvolved in the predisposition to systemic lupus erythematosus.Autoimmunity 2012;45:271–8.

31. Schoultz I, Verma D, Halfvarsson J, Torkvist L, Fredrikson M,Sjoqvist U, et al. Combined polymorphisms in genes encoding theinflammasome components NALP3 and CARD8 confersusceptibility to Crohn’s disease in Swedish men. Am JGastroenterol 2009;104:1180–8.

32. Garcia-Bermudez M, Lopez-Mejias R, Gonzalez-Juanatey C,Corrales A, Castaneda S, Ortiz AM, et al. CARD8 rs2043211 (p.C10X) polymorphism is not associated with disease susceptibilityor cardiovascular events in Spanish rheumatoid arthritis patients.DNA Cell Biol 2013;32:28–33.

33. Kastbom A, Johansson M, Verma D, Soderkvist P, Rantapaa-Dahlqvist S. CARD8 p.C10X polymorphism is associated withinflammatory activity in early rheumatoid arthritis. Ann RheumDis 2010;69:723–6.

34. Nistala K, Moncrieffe H, Newton KR, Varsani H, Hunter P,Wedderburn LR. Interleukin-17-producing T cells are enriched in thejoints of children with arthritis, but have a reciprocal relationship toregulatory T cell numbers. Arthritis Rheum 2008;58:875–87.

152 C-A Yang et al


Scan

d J

Rhe

umat

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Im

peri

al C

olle

ge L

ondo

n on

06/

14/1

4Fo

r pe

rson

al u

se o

nly.