Snps in the tnf-α gene promoter associated with behçet's disease in moroccan patients

Concise report

SNPs in the TNF-a gene promoter associated with Behc¸et’s disease in Moroccan patients

Asmaa Radouane

, Mounia Oudghiri

, Abdelfettah Chakib

, Siham Bennani

, Isabelle Touitou

and Mouna Barat-Houari

Abstract

Objective. Behc¸et’s disease (BD) is a multisystemic inflammatory disease, mainly characterized by recurrent oral and genital ulcers (GUs), skin lesions and uveitis. Several genetic factors such as the TNF- gene have been evaluated as contributors to the pathogenesis of BD. We aimed to evaluate the association between six TNF- SNPs and susceptibility to BD, or the major clinical manifestations, in Moroccan patients. The six SNPs studied were: c. 1211C>T (rs1799964), c. 1043C>A (rs1800630), c. 1037C>T (rs1799724), c. 556G>A (rs1800750), c. 488G>A (rs1800629) and c. 418G>A (rs361525), known as 1031T>C, 863C>A, 857C>T, 376G>A, 308G>A and 238G>A, respectively.

Methods.SNPs were genotyped by direct sequencing in 120 unrelated Moroccan BD and 112 ethnically matched healthy controls. Allele and genotype distributions were compared between groups using chi-square or Fisher’s exact tests.

Results.The frequency of the 1211C allele was higher in (i) BD patients than in controls [P= 0.02, odds ratio (OR) = 1.68, 95% CI 1.10, 2.56] and in (ii) patients with GUs than in those without (P= 0.002, OR = 3.84, 95% CI 1.55, 9.49). The 418A frequency was lower in patients with uveitis (P= 0.0003, OR = 0.19, 95% CI 0.07, 0.5).

Conclusion. We report the first association between BD and TNF-a SNPs in Moroccan patients. We mainly observed that 1211C constitutes a susceptibility allele for both BD and GU, as previously re- ported for other populations. The 418A allele could be considered as a good prognostic factor for anterior uveitis, in Moroccan BD patients.

Key words: Behc¸et’s disease, genital ulcers, anterior uveitis,TNF-SNP association, Moroccan patients.

Introduction

Behc¸et’s disease (BD) is a multisystemic inflammatory disorder characterized by recurrent oral and genital ulcers (GUs), skin lesions and uveitis [1]. Vasculitis worsens the prognosis, introducing the potential occurrence of

life-threatening complications such as thrombophlebitis, arterial aneurysms and occlusion. BD is mainly prevalent in the Far East (Japan, China and Korea), along the Silk Route, and in the Mediterranean Basin countries, includ- ing Morocco [2]. BD is a multifactorial disease as various environmental and genetic factors may contribute to its development. Among various susceptibility loci, the HLA-B51 locus is most strongly associated with BD in several ethnic groups from Eastern Asia, Middle East/

North Africa, Southern Europe, Northern/Eastern Europe, Turkey [3] and Japan [4]. The TNF- gene, encoding a pro-inflammatory cytokine secreted by monocytes, maps at chromosome 6p21.33, close to the HLA-B locus.

TNF-ahas been shown to be implicated in the patho- genesis of several inflammatory and autoimmune dis- eases, such as sepsis, IBD, RA and BD [5]. Elevated serum levels of TNF-a and its soluble receptor were observed in BD patients [6] as well as increased secretion

1Laboratoire d’immunologie, Institut Pasteur,²Laboratoire de Physiologie et de Ge´ne´tique Mole´culaire, De´partement de Sciences Biologiques de la Faculte´ Hassan II, Universite´ Ain Chock,³Hoˆpital Ibn Rochd, Casablanca, Service des Maladies Infectieuses, Maroc,⁴Unite´

me´dicale des Maladies Auto-inflammatoires, De´partement de Ge´ne´tique, CHRU,⁵Ge´ne´tique des Maladies Auto-inflammatoires et des Oste´o-arthropathies chroniques, INSERM U844 and⁶Universite´

Montpellier 1, UM1, Montpellier, France.

Correspondence to: Mouna Barat-Houari, Unite´ me´dicale des mala- dies Auto-inflammatoires, CHRU Montpellier, Hoˆpital A. de Villeneuve 371, avenue du Doyen Gaston Giraud 34295, Montpellier Cedex 5, France. E-mail: mouna.barat@inserm.fr

Submitted 13 January 2012; revised version accepted 17 April 2012.

BASIC SCIENCE

of TNF-afrom stimulated peripheral blood mononuclear cells [7]. Anti-TNF-atherapies have proven effective, not- ably in patients with uveitis [8].

Association between BD and SNPs in theTNF-pro- moter has been assessed and confirmed in a meta-analysis of several ethnic groups [9]. Nonetheless, since such a study has never been performed in a Moroccan population, we checked for a possible association between BD and six SNPs of the TNF- gene promoter (NM_000594.2), renamed, according to Human Genome Variation Society (HGVS) nomenclature: c. 1211C>T (rs1799964), c. 1043C>A (rs1800630), c. 1037C>T (rs1799724), c. 556G>A (rs1800750), c. 488G>A (rs1800629) and c. 418G>A (rs361525).

Materials and methods

Patients

One hundred and twenty unrelated Moroccan patients from Casablanca were referred by the Ibn Rochd’s Hospital and Pasteur Institute of Casablanca. The patients satisfy the clinical criteria or the revised criteria of the International Study Group for Behc¸et’s Disease [10], which were mainly oral ulcers (100%), GUs (80%), skin lesions (70%) and uveitis (60%). This study group included 88 males and 32 females, gender ratio: 2.75, age range 14–53 years, mean age (S.D.) at diagnostic 31.0 (7) years.

A total of 112 unrelated, healthy Moroccan (without inflammatory or systemic disease) age- and sex-matched individuals from the same geographic area, were recruited as controls by an expert clinician. All subjects were informed of the study objectives and methods and gave their consent according to the Moroccan hospital ethical committee and the Moroccan Ministry of Health. The study received ethical approval from the Moroccan hospital ethical committee and the Moroccan Ministry of Health.

Genotyping

Genomic DNA was isolated from peripheral EDTA anti-coagulated whole blood using QIAamp DNA Blood Kits (Qiagen, Valencia, CA, USA). We genotyped six SNPs located in the TNF- gene promoter (NM_000594.2), named according to the HGVS nomenclature:

c. 1211C>T (rs1799964), c. 1043C>A (rs1800630), c. 1037C>T (rs1799724), c. 556G>A (rs1800750), c. 488G>A (rs1800629) and c. 418G>A (rs361525) and previously known as 1031T>C, 863C>A, 857C>T,

376G>A, 308G>A and 238G>A, respectively.

PCR were performed with two pairs of primers (sequences available upon request), designed in the TNF- gene promoter (NM_000594.2), using the Promega PCR Master Mix, according to the manufac- turer’s protocol (Promega, Madison, WI, USA). SNPs gen- otyping was performed by direct sequencing using the BigDye Terminator v3.1 (BDT v3.1) Cycle Sequencing Kit, followed by electrophoresis of the amplicons on an ABI 3100XL Genetic Analyzer, according to the manufac- turer’s recommendations (Applied Biosystems, Foster City, CA, USA).

Statistical analysis

Comparisons of allele and genotype frequencies between groups (patientsvscontrol subjects, and within the patient’s group according to different clinical features) were per- formed using the chi-square or Fisher’s exact tests. Odds ratio (OR) and 95% CI were calculated where applicable.

Haplotypes were inferred from individual genotypes using the THESIAS 3.1 program [11] and only haplotypes with a minimal frequency of 5% in either patients or controls were analysed. All statistical analysis was performed using GraphPad Prism 5.04 software for Windows (GraphPad Software, Inc., La Jolla, CA, USA). Bonferroni corrections were applied for multiple associations.

Results

The genotypes of 120 Moroccan BD patients and 112 healthy controls were obtained for six SNPs in the TNF- promoter. The genotype distributions agree with those predicted under Hardy–Weinberg equilibrium in both patients and controls, except for rs361525 (c. 418G>A, previously known as 238G>A) in the con- trol group (data not shown). Genotype and allele distribu- tions are reported in Table 1.

The genotype distribution of TNF- c. 1211T>C (previously known as 1031 T>C) was significantly differ- ent between BD patients and healthy controls (P =0.0056, Bonferroni-adjustedP= 0.033). The 1211C seems to be more frequent in BD Moroccan patients than in healthy controls (31 vs 21%, P =0.02, Bonferroni-adjusted P= 0.11, OR = 1.68, 95% CI 1.10, 2.56). The 1211CC homozygous genotype was over-represented in BD (7.5 vs0%, Fisher’s exact testP =0.0035, Bonferroni-adjusted P =0.021, data not shown).

There was a slightly lower frequency of the 488A allele (previously known as 308A) in patients compared with controls (16vs23%, OR = 0.64, 95% CI 0.40, 1.02,P =0.06, Bonferroni-adjustedP= 0.31). Allele or genotype distribution of the other SNPs c. 1043C>A, c. 1037C>T, c. 556G>A and c. 418G>A (previously known as 863C>A, 857C>T, 376G>A and 238G>A, respectively) were not significantly different between patient and control groups.

Moreover, we observed that the frequency of the 1211C allele was higher in patients with GUs than in patients without (35.4 vs 12.5%, P =0.002, Bonferroni- adjusted P= 0.012, OR = 3.84, 95% CI 1.55, 9.49, data not shown). Although the frequency of the 1211C allele slightly increased in BD patients with arthritis, the difference did not reach statistical significance (P =0.09, Bonferroni-adjusted P= 0.43, OR = 1.67, 95% CI 0.96, 2.92, data not shown).

The 418A allele was significantly under-represented in patients with uveitis compared with those without uveitis (4.16vs18.8%,P =0.0003,Pcorr= 0.0018, OR = 0.19; 95%

CI 0.07, 0.5, data not shown) and in the subgroup with an- terior uveitis (4.7vs18.8%,p =0.0009, Bonferroni-adjusted P= 0.005, OR = 0.21; 95% CI 0.081, 0.56, data not shown).

The six SNPs were in linkage disequilibrium within our Moroccan population (data not shown). We observed that

seven haplotypes represented 99% of the 64 possible haplotypes and that their overall distribution was signifi- cantly different between cases and controls (Table 2, P =0.0005). Two haplotypes, CCCGGG and CCCGGA, were significantly over-represented among patients (8.75 vs 3.12%, P =0.011, OR = 2.97, 95% CI 1.24, 7.14 and 8.34 vs 1.78%, P =0.0014, OR = 4.98, 95% CI 1.67, 14.8, respectively) and, in the patients sub-group with GU (19.8vs4.91%,P<0.0001, OR = 4.78, 95% CI 2.37, 9.65, data not shown). The total frequency of these two haplotypes was significantly higher in patients compared with healthy controls (17.1vs4.91%, OR = 3.99, 95% CI 1.99, 7.98,P<0.0001, Table 2).

Discussion

The pathogenesis of BD is unknown; however, it has been suggested that genetic factors may be implicated in its development, such as the HLA-B51 locus and theTNF- gene [3, 4, 9]. TheTNF-promoter allele 1211C consti- tutes a major susceptibility factor for BD, in either HLA-B51- positive or negative UK white Caucasian patients [12]. We analyse the association between BD in the Moroccan population and six SNPs in the TNF- gene: rs1799964 (c. 1211C>T), rs1800630 (c. 1043C>A), rs1799724 (c. 1037C>T), rs1800750 (c. 556G>A), rs1800629 (c. 488G>A) and rs361525 (c. 418G>A).

Allele frequences of theTNF-SNPs in our Moroccan population are quite similar to those reported in the Tunisian sample [9, 13]. The 1211C frequency was higher in BD patients than in controls, as previously

reported [7, 9, 12, 14] And this allele tended to be over-represented in the sub-group of patients suffering from arthritis or GU as observed in a Korean sample [14]. It was shown that 1211C contributes to an increased transcriptional activity, resulting in TNF-aover- production and chronic inflammation in BD [15]. Although c. 488G>A was not associated with BD in Moroccan pa- tients, as reported in other populations [9], we observed that the 488A allele tended to be more frequent in the control group.

HLA-B27 was recently associated with anterior uveitis in another sample of Moroccan patients [16]. Interestingly, we report for the first time a strong association between the 418A allele and the absence of uveitis. Since 89% of Moroccan patients with uveitis displayed an anterior form, the 418A could play a protective role against anterior uveitis in this population. Such an association was not observed in the Korean sample [14].

TheTNF-SNPs were in linkage disequilibrium in both the control and patient groups of our Moroccan sample (data not shown). The overall distribution of the seven major haplotypes was significantly different between pa- tients and controls (p =0.0005). The most frequent haplo- type, TCCGGG, composed of the six common alleles, was not associated with BD in our Moroccan sample (Table 2), unlike in Caucasians and Koreans [12, 14]. The CCCGGG and CCCGGA haplotypes were significantly associated with a higher susceptibility to BD and GU, as previously observed in UK white Caucasoid patients [12]. The CCCG GA haplotype was associated with a decreased suscep- tibility to anterior uveitis as was the 418A allele.

TABLE1 Genotype and allele distribution of theTNF-SNPs of BD patients and control subjects

SNPs

Genotype distribution

P-value MAF OR (95% CI)^b (frequency, %) (Pcorr)^a % P-value (Pcorr)^a

rs1799964 (c.-1211T>C) TT TC CC C

BD 55 (45.8) 56 (46.7) 9 (7.5) 0.0056 31.2 1.68 (1.10, 2.56)

Controls 65 (58.0) 47 (42.0) 0 (0) (0.033) 21.0 0.02(0.11)

rs1800630 (c.-1043C>A) CC CA AA A

BD 89 (74.2) 31 (25.8) 0 (0) 0.88 12.9 1.13 (0.65, 1.97)

Controls 85 (75.9) 27 (24.1) 0 (0) – 12.0 0.67

rs17999724 (c.-1037C>T) CC CT TT T

BD 105 (87.5) 15 (12.5) 0 (0) 0.46 6.25 0.76 (0.37, 1.55)

Controls 94 (83.9) 18 (16.1) 0 (0) – 8.03 0.47

rs1800750 (c.-556G>A) GG GA AA A

BD 116 (96.7) 4 (3.3) 0 (0) 0.15 1.67 0.40 (0.12, 1.3)

Controls 103 (92.0) 9 (8.0) 0 (0) – 4.02 0.16

rs1800629 (c.-488G>A) GG GA AA A

BD 82 (68.3) 38 (31.7) 0 (0) 0.018 15.8 0.64 (0.40, 1.02)

Controls 68 (60.7) 37 (33.0) 7 (6.3) – 22.8 0.06

rs361525 (c.-418G>A) GG GA AA A

BD 102 (85.0) 12 (10.0) 6 (5.0) 0.045 10.0 1.55 (0.79, 3.03)

Controls 97 (86.6) 15 (13.4) 0 (0) – 6.70 0.24

aChi-square or Fisher’s exact testP-values for comparison of genotype and allele distribution between cases and controls;

andPcorr= Bonferroni-adjustedP-value,^bOR of the MAF. SignificantP-values and OR are reported in bold. MAF = minor allele frequency.

The c. 418G>A, c. 488G>A and c. 556G>A were shown to be clustered in a series of conserved repetitive motifs, which could constitute consensus binding sites for a transcriptional factors or modulate DNA structure [17].

The functional significance of the 488A allele was largely demonstrated by an increased TNF- transcriptional activity of a reporter gene expressed in various cell lines [18] and an increased TNF-secretion by LPS-stimulated peripheral blood mononuclear cells from heterozygous individuals [19].

In conclusion, we succeeded in replicating in a Moroccan population the previously reported associations of BD (or related phenotypes, such as GU) with (i) 1211C allele and (ii) the three major haplotypes ofTNF-SNPs.

This study also pinpointed the 418A allele as a good prognostic factor for anterior uveitis. We suggest that the common allele 418G could increase transcriptional activity of theTNF-promoter resulting in increased levels of this cytokine in the sera and aqueous fluids of patients suffering from intraocular inflammation. Our observations add pieces to the puzzle of the genetic susceptibility to BD and support the role played by TNF- in BD physiopathology.

Rheumatology key messages

. The c. 1211C allele constitutes a susceptibility factor for BD in Moroccans.

. The c. 1211C allele is a susceptibility factor for GUs in BD in Moroccan patients.

. The c. 418A allele is a good prognostic factor for uveitis in BD in Moroccan patients.

Acknowledgements

We thank Mrs Imane El Mouraghi for her help in the DNA isolation and Dr Francis Vasseur for his contributions to the choice of statistical analyses. The manuscript was edited for English language by Angloscribe.

Funding: This work was supported by the French Ministry of Health, the Moroccan Ministry of Health and the ‘Institut Pasteur in Morocco’.

Disclosure statement: The authors have declared no conflicts of interest.

References

1 Kapsimali VD, Kanakis MA, Vaiopoulos GA,

Kaklamanis PG. Etiopathogenesis of Behc¸et’s disease with emphasis on the role of immunological aberrations.

Clin Rheumatol 2010;29:1211–6.

2 Davatchi F, Shahram F, Chams-Davatchi Cet al. Behc¸et’s disease: from East to West. Clin Rheumatol 2010;29:

823–33.

3 Remmers EF, Cosan F, Kirino Yet al. Genome-wide as- sociation study identifies variants in the MHC class I, IL10, and IL23R-IL12RB2 regions associated with Behc¸et’s disease. Nat Genet 2010;42:698–702.

TABLE2Estimateddistributionandfrequenciesofsevenhaplotypesamong64possiblehaplotypesinferredfromcontrolandpatientgroupsusingtheThesias3.1program HaplotypereconstructionHaplotypedistribution(frequency,%)Fisher’sexacttesta c.-1211T>Cc.-1043C>Ac.-1037C>Tc.-556G>Ac.-488G>Ac.-418G>AControls(n=224)BD(n=240)P-valueOR(95%CI) TCCGGG106(47.3)111(46.2)0.850.96(0.59,1.55) CACGGG27(12.0)31(12.9)0.891.08(0.52,2.23) CCCGGG7(3.12)21(8.75)0.0112.97(1.24,7.14) CCCGGA4(1.79)20(8.34)0.00144.98(1.67,14.8) TCCGAG51(22.8)38(15.8)0.060.64(0.40,1.02) TCTGGG18(8.0)15(6.25)0.280.67(0.331.36) CCCAGA9(4.0)2(0.84)0.030.20(0.04,0.94) Otherhaplotypes2(0.89)2(0.84)1.000.93(0.13,6.68) Ofall64possiblehaplotypes,only7werereportedtohavereachedaminimalfrequencyof5%ineitherthecontrolorBehc¸et’sdiseasegroup.n:numberofchromosomes.a Fisher’s exacttestP-valuesforcomparisonofonehaplotypevsallotherpossiblehaplotypes,significantP-valuesandORarereportedinbold.

4 Mizuki N, Meguro A, Ota Met al. Genome-wide associ- ation studies identify IL23R-IL12RB2 and IL10 as Behc¸et’s disease susceptibility loci. Nat Genet 2010;42:703–6.

5 Kapoor M, Martel-Pelletier J, Lajeunesse D, Pelletier JP, Fahmi H. Role of proinflammatory cytokines in the pathophysiology of osteoarthritis. Nat Rev Rheumatol 2011;7:33–42.

6 Duzgun N, Ayaslioglu E, Tutkak H, Aydintug OT. Cytokine inhibitors: soluble tumor necrosis factor receptor 1 and interleukin-1 receptor antagonist in Behc¸et’s disease.

Rheumatol Int 2005;25:1–5.

7 Akman A, Sallakci N, Coskun Met al. TNF-alpha gene 1031 T/C polymorphism in Turkish patients with Behc¸et’s disease. Br J Dermatol 2006;155:350–6.

8 Benitah NR, Sobrin L, Papaliodis GN. The use of biologic agents in the treatment of ocular manifestations of Behc¸et’s disease. Semin Ophthalmol 2011;26:295–303.

9 Touma Z, Farra C, Hamdan Aet al. TNF polymorphisms in patients with Behcet disease: a meta-analysis. Arch Med Res 2010;41:142–6.

10 Criteria for diagnosis of Behc¸et’s disease. International Study Group for Behc¸et’s Disease. Lancet 1990;335:1078–80.

11 Tregouet DA, Escolano S, Tiret L, Mallet A, Golmard JL. A new algorithm for haplotype-based association analysis: the Stochastic-EM algorithm. Ann Hum Genet 2004;68:165–77.

12 Ahmad T, Wallace GR, James Tet al. Mapping the HLA association in Behc¸et’s disease: a role for tumor necrosis factor polymorphisms? Arthritis Rheum 2003;48:807–13.

13 Kamoun M, Chelbi H, Houman MH, Lacheb J,

Hamzaoui K. Tumor necrosis factor gene polymorphisms in Tunisian patients with Behc¸et’s disease. Hum Immunol 2007;68:201–5.

14 Park K, Kim N, Nam J, Bang D, Lee ES. Association of TNFA promoter region haplotype in Behc¸et’s disease.

J Korean Med Sci 2006;21:596–601.

15 Higuchi T, Seki N, Kamizono Set al. Polymorphism of the 5’-flanking region of the human tumor necrosis factor (TNF)-alpha gene in Japanese. Tissue Antigens 1998;

51:605–12.

16 Radouane A, Oudghiri M, Chakib A et al. HLA-B*27 allele associated to Behc¸et’s disease and to anterior uveitis in Moroccan patients. Ann Biol Clin 2011;69:

419–24.

17 Allen RD. Polymorphism of the human TNF-alpha promoter–random variation or functional diversity?

Mol Immunol 1999;36:1017–27.

18 Karimi M, Goldie LC, Cruickshank MN, Moses EK, Abraham LJ. A critical assessment of the factors affecting reporter gene assays for promoter SNP function: a reassessment of -308 TNF polymorphism function using a novel integrated reporter system. Eur J Hum Genet 2009;

17:1454–62.

19 Louis E, Franchimont D, Piron Aet al. Tumour necrosis factor (TNF) gene polymorphism influences TNF-alpha production in lipopolysaccharide (LPS)-stimulated whole blood cell culture in healthy humans. Clin Exp Immunol 1998;113:401–6.