ARTHRITIS & RHEUMATISM

Vol. 50, No. 6, June 2004, pp 2032–2033

© 2004, American College of Rheumatology
CONCISE COMMUNICATION

DOI 10.1002/art.20215 The results of the present study of 2 independent

Lack of association between the I␬BL promoter cohorts could not confirm the recent finding that I␬BL was the
second MHC-linked susceptibility locus for RA (7). Validation
polymorphism and rheumatoid arthritis
of these findings in other populations is needed to establish the
Various strategies have been pursued to elucidate the role of I␬BL polymorphism in susceptibility to RA. In accor-
genetic background of rheumatoid arthritis (RA) (for review, dance with our data, a recent study conducted in a British
see ref. 1). Genome-wide linkage analyses using microsatellite population, reported in abstract form, revealed no evidence of
markers have consistently demonstrated that the HLA region association of the I␬BL ⫺62 marker with RA (9). Further-
has the largest genetic contribution to RA, which is consistent more, we previously reported a lack of association of the
with the well-documented association of RA with HLA class II I␬BL ⫹ 738T/C mutation with RA in a Spanish population
genes (for review, see refs. 1 and 2). Although models have (10). Okamato et al (7) have speculated that I␬BL ⫺62 T/A
been proposed to explain the contribution of HLA to RA, polymorphism may have biologic significance. Nevertheless,
none of them can entirely explain the estimated 37% Allcock et al did not observe any effect of SNP ⫺62 on the
contribution of the major histocompatibility complex transcription level of I␬BL messenger RNA, studied using
(MHC) region to the genetic component of RA. In fact, it transformed B cell lines (11). Taken together, these data
has been suggested that other loci within the MHC region suggest that I␬BL gene polymorphism may not play a relevant
may contribute to the genetic risk of RA (1,2). Among other role in RA.
considerations, different shared epitope–encoding DRB1 Lack of replication of a previous association is a
alleles vary in the strength of their association with RA (2). common event in the search for genetic determinants of
The search for additional predisposing loci within the MHC complex human traits. In order to identify the most likely
is complicated by the strong disequilibrium among loci in causes of the lack of replication in these kinds of studies,
the MHC region and by the heterogeneity of RA in terms of recent meta-analyses have been conducted (12,13). Typically,
disease severity. the results of the first study show a stronger effect than is found
We and others, analyzing microsatellite markers, have in subsequent studies, due either to genuine population hetero-
demonstrated that the telomeric portion of the HLA region geneity or to various types of bias. Among the most common
contains a genetic factor that predisposes to RA independently causes of irreproducibility are population stratification, publi-
of the HLA class II genes (3–6). The I␬BL (inhibitor of cation bias, and time-lag bias. In addition, in some cases there
NF-␬B–like) gene, located in this RA-relevant segment, is a may be statistical uncertainty in the first study. Of note,
potential regulator of the NF-␬B pathway and may therefore Okamato et al (7) did not address the multiple comparison
be important in the regulation of proinflammatory cytokine consideration, leading to a possible false-positive association
production. Interestingly, a recent study has implicated a between I␬BL genotypes and RA occurring by random chance.
polymorphism in the promoter region of I␬BL in susceptibility Our lack of significant results is unlikely to have resulted from
to RA (7). The above considerations prompted us to investi- low power, because the number of RA patients and control
gate the I␬BL gene promoter polymorphism in RA, using 2 subjects was higher than in the original study (7). Based on the
independent clinical cohorts. previous study, one would expect that statistical power of
We analyzed a case–control panel consisting of 339 RA 99.0% would be achieved with 305 and 270 subjects in the
patients and 382 control subjects. The first group comprised control and patient groups, respectively (P ⫽ 0.05 and TT
203 RA patients and 204 controls referred from the Hospital genotype frequency in the control group 43.7%).
Virgen de las Nieves (Granada), and the second group con- The effects of genetic and population (and clinical)
sisted of 141 RA patients and 180 controls from the Hospital heterogeneity must be considered when attempting to detect
Virgen del Rocio (Seville). Cases and controls were of Spanish susceptibility genes for RA in different populations. Our
Caucasian ethnicity and were matched for age and sex. The findings, together with those of Coote et al in a study of
patients were diagnosed according to the American College of another Caucasian population (9), raise the question of
Rheumatology revised criteria for RA (8). whether I␬BL may be an RA susceptibility gene only in the
The allele distribution of I␬BL ⫺62 gene variation in Japanese population. Such differences might result from the
the case–control cohorts from Granada and Seville showed no differing evolutionary histories of the populations or from
significant differences in allele frequencies; hence, we com- interaction of I␬BL polymorphism with various environmental
bined the 2 cohorts to form an RA case–control group which exposures. The risk of the disease may also be modified by
was used in the further analysis. We did not observe statisti- other polymorphisms located in and around the functional
cally significant differences among I␬BL ⫺62 genotypic or variants or by haplotypic effects created by alleles at I␬BL loci
allelic frequencies between cases and controls, for either that are in linkage disequilibrium with other important RA
cohort individually or in the 2 cohorts together, as shown in genetic markers, which differ between Caucasian and Japanese
Table 1. Also, the control genotype frequencies of single- populations. Another plausible explanation for the apparently
nucleotide polymorphism (SNP) ⫺62 were in Hardy-Weinberg contradictory findings across studies is that the underlying
equilibrium. In addition, no differences in I␬BL ⫺62 allele biologic mechanism of the I␬BL gene is complex and its
distribution were found when RA patients and controls were functional relevance and physiologic roles in inflammation and
stratified by the presence or absence of the shared epitope immunity are yet to be understood. The question remains as to
(data not shown). whether other polymorphisms within the genes of the NF-␬B

2032
CONCISE COMMUNICATION 2033

Table 1. I␬BL ⫺62 genotypes and allele frequencies in the rheumatoid arthritis (RA) case–control
cohorts*
Granada Seville Total

RA patients Controls RA patients Controls RA patients Controls

(n ⫽ 203) (n ⫽ 204) (n ⫽ 141) (n ⫽ 180) (n ⫽ 344) (n ⫽ 384)
Genotype
AA 16 (7.9) 17 (8.3) 10 (7.1) 16 (8.9) 26 (7.6) 33 (8.6)
AT 112 (55.2) 99 (48.5) 64 (45.4) 84 (46.7) 176 (51.2) 183 (47.7)
TT 75 (36.9) 88 (43.1) 67 (47.5) 80 (44.4) 142 (41.3) 168 (43.8)
Allele
A 144 (35.5) 133 (32.6) 84 (29.8) 116 (32.2) 228 (33.1) 249 (32.4)
T 262 (64.5) 275 (67.4) 198 (70.2) 244 (67.8) 460 (66.9) 519 (67.6)

* I␬BL ⫺62 typing was carried out directly on genomic DNA obtained by standard methods, as previously
described (10). Allelic and genotypic frequencies were compared by Pearson’s chi-square test with
contingency tables. Odds ratios with 95% confidence intervals were also calculated. Statistical power
calculations were made using the Statcalc program (Epi Info 2002; Centers for Disease Control and
Prevention, Atlanta, GA). Values are the number (%).

cascade inhibitors may act in concert to enhance RA predis- 5. Pascual M, Mataran L, Jones G, Shing D, van der Slik AR,
position, as seems to occur in multiple sclerosis (14). Giphart MJ, et al. HLA haplotypes and susceptibility to rheuma-
Supported by grant SAF2003-0346 from the Comision toid arthritis: more than class II genes. Scand J Rheumatol
2002;31:275–8.
Interministerial de Ciencia y Technologı́a, Spain. 6. Jawaheer D, Li W, Graham RR, Chen W, Damle A, Xiao X, et al.
Lola Collado, PhD Dissecting the genetic complexity of the association between
Blanca Rueda, PhD human leukocyte antigens and rheumatoid arthritis. Am J Hum
Genet 2002;71:585–94.
CSIC 7. Okamoto K, Makino S, Yoshikawa Y, Takaki A, Nagatsuka Y, Ota
Granada, Spain M, et al. Identification of I␬BL as the second major histocompat-
Rafael Cáliz, MD ibility complex-linked susceptibility locus for rheumatoid arthritis.
Hospital Virgen de las Nieves Am J Hum Genet 2003;72:303–12.
Granada, Spain 8. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF,
Belen Torres, PhD Cooper NS, et al. The American Rheumatism Association 1987
Alicia Garcı́a, MD revised criteria for the classification of rheumatoid arthritis.
Antonio Nuñez-Roldan, MD Arthritis Rheum 1988;31:315–24.
Maria Francisca González-Escribano, PhD 9. Coote AL, Kilding R, Iles MM, Wilson AG. The IKBL-62
polymorphism is not associated with rheumatoid arthritis [ab-
Hospital Virgen de Rocio
stract]. Arthritis Rheum 2003;48 Suppl 9:S412.
Seville, Spain 10. Martinez A, Pascual M, Pascual-Salcedo D, Balsa A, Martin J, de
Javier Martin, MD, PhD la Concha EG. Genetic polymorphisms in Spanish rheumatoid
CSIC arthritis patients: an association and linkage study. Genes Immun
Granada, Spain 2003;4:117–21.
11. Allcock RJ, Baluchova K, Cheong KY, Price P. Haplotypic single
1. Barton A, Ollier W. Genetic approaches to the investigation of nucleotide polymorphisms in the central MHC gene IKBL, a
rheumatoid arthritis. Curr Opin Rheumatol 2002;14:260–9. potential regulator of NF-␬B function. Immunogenetics 2001;52:
2. Gregersen PK. Teasing apart the complex genetics of human 289–93.
autoimmunity: lessons from rheumatoid arthritis. Clin Immunol 12. Ioannidis JP, Ntzani EE, Trikalinos TA, Contopoulos-Ioannidis
2003;107:1–9. DG. Replication validity of genetic association studies. Nat Genet
3. Zanelli E, Jones G, Pascual M, Eerligh P, van der Slik AR, 2001;29:306–9.
Zwinderman AH, et al. The telomeric part of the HLA region 13. Lohmueller KE, Pearce CL, Pike M, Lander ES, Hirschhorn JN.
predisposes to rheumatoid arthritis independently of the class II Meta-analysis of genetic association studies supports a contribu-
loci. Hum Immunol 2001;62:75–84. tion of common variants to susceptibility to common disease. Nat
4. Ota M, Katsuyama Y, Kimura A, Tsuchiya K, Kondo M, Naruse T, Genet 2003;33:177–82.
et al. A second susceptibility gene for developing rheumatoid 14. Miterski B, Bohringer S, Klein W, Sindern E, Haupts M, Schim-
arthritis in the human MHC is localized within a 70-kb interval rigk S, et al. Inhibitors in the NF␬B cascade comprise prime
telomeric of the TNF genes in the HLA class III region. Genomics candidate genes predisposing to multiple sclerosis, especially in
2001;71:263–70. selected combinations. Genes Immun 2002;3:211–9.