Rheumatology Advance Access originally published online on November 22, 2007
Rheumatology 2008 47(2):132-137; doi:10.1093/rheumatology/kem269
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
REVIEWS |
Breakthroughs in genetic studies of ankylosing spondylitis
1Immunogenetics, Diamantina Institute of Cancer, Immunology and Metabolic Medicine, Princess Alexandra Hospital, Woolloongabba, Qld, Australia, University of Oxford Institute of Musculoskeletal Sciences, Botnar Research Centre, Nuffield Orthopaedic Centre, Oxford, UK.
Correspondence to: M. A. Brown, Professor of Immunogenetics, Diamantina Institute of Cancer, Immunology and Metabolic Medicine, Princess Alexandra Hospital, Ipswich Road, Woolloongabba, Qld, 4102, Australia. E-mail: matt.brown{at}uq.edu.au
 |
Abstract |
|---|
Top
Abstract
IntroductionAnkylosing spondylitis (AS), the prototypic seronegative arthropathy, is known to be highly heritable, with >90% of the risk of developing the disease determined genetically. As with most common heritable diseases, progress in identifying the genes involved using family-based or candidate gene approaches has been slow. The recent development of the genome-wide association study approach has revolutionized genetic studies of such diseases. Early studies in ankylosing spondylitis have produced two major breakthroughs in the identification of genes contributing roughly one third of the population attributable risk of the disease, and pointing directly to a potential therapy. These exciting findings highlight the potential of future more comprehensive genetic studies of determinants of disease risk and clinical manifestations, and are the biggest advance in our understanding of the causation of the disease since the discovery of the association with HLA-B27.
KEY WORDS: Ankylosing spondylitis, Genetics, Association
|
Introduction |
|---|
Ankylosing spondylitis (AS), the prototypic seronegative arthropathy, is known to be highly heritable, with >90% of the risk of developing the disease determined genetically. As with most common heritable diseases, progress in identifying the genes involved using family-based or candidate gene approaches has been slow. The recent development of the genome-wide association study approach has revolutionized genetic studies of such diseases. Early studies in AS have produced two major breakthroughs in the identification of genes contributing roughly one-third of the population attributable risk of the disease, and pointing directly to a potential therapy. These exciting findings highlight the potential of future more comprehensive genetic studies of determinants of disease risk and clinical manifestations, and are the biggest advance in our understanding of the causation of the disease since the discovery of the association with HLA-B27.
AS is an enigma in rheumatology. How is it that the aetiopathogenesis of a disease whose main gene was identified more than 30 yrs ago has not been solved? Why is this common condition so resistant to standard, non-biological, rheumatological treatments? Why is it that despite high-quality epidemiological surveys indicating that the condition is common, and in public health terms expensive, that rheumatology clinics and wards are not full of patients with the disease? The resistance of the disease to yield answers as to its aetiopathogenesis such as the mechanism explaining its association with HLA-B27, and its resistance to standard therapeutic options, has discouraged researchers and the pharmaceutical industry from investing the requisite energy and finances into pursuing the research required to develop effective treatments. Patients themselves, many of whom experienced substantial disillusionment with traditional medicine even in establishing a correct diagnosis for their symptoms, gave up seeking conventional treatment, which had little to offer them. This led to a vicious circle—AS was under-researched because it was not perceived to be a major public health problem, and patients often withdrew from traditional medical attention because all that was offered was physiotherapy and anti-inflammatories.
Paradoxically, the recognition that B27 is not the main answer to the genetic causation of the disease but requires an unhealthy dose of modifier genes to cause the condition has led to research breakthroughs in our understanding of AS aetiopathogenesis, and likely in the near future, to novel treatments. Such breakthroughs, enabled largely by the enthusiastic support of AS patient organizations for genetics research, are bringing about the biggest change in our understanding of the disease since the 1970s.
|
Genetic epidemiology of ankylosing spondylitis |
|---|
For more than 40 yrs, AS has been known to run in families. Familiality occurs because of shared susceptibility factors amongst family members, which may be environmental (random or common) or genetic. There is substantial evidence that AS is triggered by exposure to a common environmental pathogen. The condition has a (nearly) worldwide distribution, indicating that whatever environmental exposure triggers the disease in a genetically susceptible person occurs from the Arctic to Asia. Epidemics of AS have not been reported, in contrast to reactive arthritis, another B27-associated disease, which does occur in common source outbreaks. The best environmental model to fit these facts is that the environmental factor involved in AS is ubiquitous. This conclusion is supported by studies of B27-transgenic rats which indicate that the disease develops in rats exposed to normal enteric commensal bacteria, but not in rats maintained in germ-free conditions [1]. Twin studies have provided strong evidence to support this conclusion, with two studies now formally estimating heritability as >90% [2, 3]. Heritability of clinical manifestations of disease is also significant, with age of symptom onset, disease activity measured by the widely employed BASDAI and BASFI questionnaires, and radiographic severity being 40, 51, 76 and 62% heritable, respectively [4–6]. These studies involved families where all cases were B27-positive, and thus the heritability reflects non-B27-linked genetic effects.
Such high heritability is usually associated with monogenic diseases, and the identification of B27 led many investigators to believe that AS was essentially monogenic. Evidence from family and twin studies suggest that at least a moderate number of other genes were involved. Monogenic diseases tend to be quite uncommon, because even minor selection pressure/evolutionary disadvantage leads to adverse genetic effects being eliminated from the general gene pool quite rapidly. It is possible that B27 has been maintained at fairly high frequency because it protects against some disease—for example, there is some evidence that it may partially protect against HIV infection [7]. However, the high prevalence of AS argues against it being a monogenic disease.
Further, studies of the recurrence risk of the condition in relatives of cases indicate that it does not segregate as a monogenic condition. The recurrence risk of any disease depends on the genetic model [8]. Monogenic diseases reduce in frequency by half with each increase in distance of relationship to the proband. Polygenic diseases reduce in frequency by roughly the square root with each increase in distance of relationship to the proband. AS probably lies somewhere in between these models; B27 is almost essential for the inheritance of AS, but its penetrance is greatly modified by other genes. Family studies in AS suggest that a moderate number of genes with significant effect are involved in AS, the best-fitting models suggesting five genes, but with a wide range of possible-fitting models ranging from three to nine genes being involved [9]. It is likely that in addition to B27 there will be a small number of genes with moderate effect, and a large number of genes with small effect. Certainly, large multi-case families, typical of diseases with monogenic aetiology, are extremely rare with AS. In contrast, there are many such families reported with chondrocalcinosis and osteoarthritis, for which monogenic variants have been identified.
The strength of the association of B27 with AS led many researchers to believe that it was a monogenic disease. The successful identification of other genes contributing to the disease has now provided the ultimate proof that this was a misconception.
|
Methods for identifying disease-causing genes |
|---|
To a great extent, genetic research has been technologically driven. In the 1980s, the discovery of PCR and the development of restriction fragment length polymorphism approaches to linkage mapping in families enabled a genetics revolution which led to the identification of most genes involved in monogenic diseases. The success in monogenic diseases led to the suggestion that the same approach would work for common, complex, heritable diseases [10]. This was the beginning of ‘hypothesis-free’ genetics research, and encountered a lot of resistance from traditional hypothesis-based researchers. Why should the limited research dollar be spent on ‘fishing-trips’ when the investigators had no proof that the approach would work? As it turns out, these views were—at least in the short-term—validated, because family-based approaches have made little progress in common heritable diseases, including AS. There are at least two main reasons for this.
First, the technological leap required when this approach was first tried was too great. With no human genome map in place, even a firm linkage required an extensive and expensive mapping and sequencing project to identify even the major genes in the linked region. Genotyping error rates were too high; error rates >1% were common, whereas rates >0.1% are now considered insufficiently stringent. Genotyping was inefficient with costs per genotype exceeding £1, whereas nowadays they are less than one percent of that. We had little understanding of the complexity of genetic variation within populations, or methods to identify or control for it.
Second, the genetic effects involved were mostly too small to be detected by family-based linkage approaches. Studies of a few hundred families were at the time considered large, but actually had inadequate power to find genes contributing <10% of the risk to common diseases. Very few success stories in polygenic diseases exist, but there have been some notable exceptions. These include CARD15/NOD2, the major susceptibility gene for Crohn's disease [11]; sequestosome mutations causing Paget's disease [12]; LRP5, a major gene for osteoporosis [13–15]; and a handful of others. These examples all highlight the potential of genetics research, providing breakthroughs in the diseases concerned that decades of hypothesis-driven research had failed to produce. Most genes, though, were below the sensitivity of the linkage approaches that were being pursued, and a new approach was required.
Thankfully, a number of advances—technological, bioinformatic, and computational—have revolutionized the field. Microarray genotyping technologies have enabled high-throughput genotyping at <1 p/genotype with <0.1% error rates. Throughput increased massively. In 1993, two colleagues of mine, working all hours, completed two chromosomes of a family-based linkage study in a common genetic disease. The introduction of fluorescent-tagged microsatellite technology meant that I completed 23 chromosomes in a similar number of samples in the next 12 months. In 2006, my research group took delivery of a microarray SNP genotyper, and in their first week of training completed five times as many genotypes as I had in a year. Duplicate genotyping experiments gave the same genotypes across nearly 316 000 SNPs in two completely independent paired experiments, indicating unprecedented genotyping accuracy. This has removed the need to focus on families (which required a lower marker density) and enabled ultra-high density studies across the entire genome of apparently unrelated cases and controls. The human genome map and International HapMap project provided public resources that enabled intelligent design of mapping projects. This greatly accelerated the leap from low- to high-resolution genetic studies, and led to improved understanding of the complexity of population structure of genetic variation, arising from population migrations and from selection pressures such as infection and malnutrition. Statistical advances enabled this structure to be identified and controlled for. Initially, this consisted of within-family designs such as the parent–case trio transmission-disequilibrium association tests, but these have now been extended to unrelated case–control data.
These frame-shift changes have brought us to the point where we can realistically survey >80% of the genetic variation in human populations to determine its involvement in common diseases, and to a surge in successful identification of human disease genes. Such studies involve genotyping >300 000 SNPs selected across the genome according to the extent of variation observed in populations, in thousands of unrelated cases and controls. These ‘genome-wide association studies’ thus involve billions of genotypes, creating unprecedented analytical challenges, and also unprecedented success. For example, the recently completed Wellcome Trust Case–Control Consortium (WTCCC) study of seven common diseases, including rheumatoid arthritis, has identified 25 genes with significant effects definitely associated with disease, and a further 58 which have high probability of being involved, many times the number of genes that had been hitherto successfully identified [16]. There had been a concern that if genes involved in disease susceptibility harboured multiple variants associated with disease, then genome-wide association studies would not be able to identify them. The WTCCC study shows that for many important disease-susceptibility genes, this is not the case, and the approach works. The hypothesis-driven researchers now have a solid, priceless, foundation of evidence as to the aetiopathogenesis of these diseases upon which to base their research, henceforth. Because of the public funding of these studies, the data are available to all bona fide researchers free of charge (https://www.wtccc.org.uk/). For many years, no serious researcher of the pathogenesis of rheumatic conditions would have ignored the known associations of rheumatoid arthritis with HLA-DRB1, of AS with HLA-B27, or of osteoporosis with LRP5 and the Wnt signalling pathway. Because of these genetic advances, this has now expanded in rheumatoid arthritis to include PTPN22 and several other genes. Over the next few years, similar advances are likely, resources dependent, to occur in most common rheumatic diseases, making this a golden era for human genetics research.
|
Linkage studies and ankylosing spondylitis |
|---|
Three groups have published linkage studies in AS [17–20], and the genotypes from these studies have been combined in formal meta-analysis to include data from 488 multiple affected families involving 589 affected sibling pairs [21]. Unsurprisingly, the strongest linkage observed was with the MHC. In addition, regions on chromosome 10 and 16q achieved ‘suggestive’ evidence of linkage (likely to occur once by chance per genome screen), and regions on chromosomes 1, 3, 5, 6, 9, 17 and 19q showed at least nominal linkage in two or more scans. Two regions where association had previously been reported (chromosome 2, the IL-1 gene cluster [22–25]; chromosome 22, CYP2D6 [26, 27]) showed nominal linkage. The linkages observed were quite broad, and there is a risk that any particular linkage might be a false positive, given the moderate evidence for linkage. When these scans were performed, it had been planned to follow them up by systematic screens of linked regions by high-density association mapping, but the development of genome-wide association studies has made this approach obsolete. Nonetheless, these linkage scans remain helpful in prioritization of genes for further study, either for candidate gene association studies or following genome-wide association studies.
|
High-density association studies and AS |
|---|
No genome-wide association study has yet been completed in AS. The WTCCC AS study, involving genotyping of 14 436 non-synonymous (coding) and 897 MHC SNPs, in 1000 cases and 1500 controls, is by far the largest study to date in AS genetics, but nonetheless has only investigated a small proportion of the overall genetic diversity in relation to the disease [28]. The findings of the MHC component of this study are discussed subsequently.
The AS component of this study has identified two major genes involved in the disease which represent major breakthroughs in our understanding of the aetiology of AS, and which will stimulate new research efforts in the disease. These two genes are ARTS1 and IL23R, for both of which strong, independently replicated association has been identified (Table 1). These two genes, respectively, are responsible for 26 and 9% of the population- attributable risk of AS; in contrast, the population-attributable risk of PTPN22 in rheumatoid arthritis and type 1 diabetes mellitus is 8% [16].
|
IL23R encodes a critical cytokine receptor in the TH17 subset of T cells. TH17 cells were originally identified as a distinct subset of T-cells expressing high levels of the pro-inflammatory cytokine IL-17 in response to stimulation [29]. This subset of effector T-cells was identified by immunologists researching paradoxical effects of interferon-
blockade in the mouse models extrinsic autoimmune encephalomyelitis [30] and collagen-induced arthritis [31], models most closely resembling multiple sclerosis and rheumatoid arthritis, respectively. In the past 12 months, genetic variation in IL23R has been demonstrated to substantially affect susceptibility to Crohn's disease [32, 33], psoriasis [34] and AS [28]. Whether they also influence multiple sclerosis is unknown, but the WTCCC did not identify association of IL23R with rheumatoid arthritis, so it appears not to be important in that disease. This is, thus, not only a major gene for AS, but also at least partially explains the clinical association of AS with Crohn's disease and psoriasis. The association in AS was stronger in patients lacking clinical IBD, indicating that the genetic effects independently influence AS susceptibility. This is, thus, the first seronegative pleiotropic gene (affecting more than one disease), as the previously identified major genes for these conditions, such as CARD15 for Crohn's disease, HLA-B27 for AS and HLA-Cw6 for psoriasis, are disease-specific. Although further research is required to determine the primary associated variant, and whether the variants involved are gain- or loss-of-function, the therapeutic potential of this finding is immediately apparent. Our own data (unpublished) indicate that IL23R genetic variants also influence disease age of onset and severity. The strong association with susceptibility to disease suggests that IL23/IL23R targeted therapies may even be capable of disease prevention, a particularly exciting potential.
Perhaps more exciting was the finding of association of the gene ARTS1 and disease (Table 1). The gene has two known functions, either of which may explain its association with AS. Within the endoplasmic reticulum, ARTS1 is involved in trimming peptides to the optimal length for major histocompatibility complex (MHC) Class I presentation [35, 36]. AS is primarily an HLA-Class I-mediated autoimmune disease, with >90% of cases carrying the HLA-B27 allele. How B27 causes AS is unknown, but if the mechanism of association of ARTS1 with the disease is through effects on peptide presentation, this would have significant implications for research into the mechanisms of association of the B27 with AS, substantially favouring mechanisms that involve abnormalities of peptide presentation rather than those involving abnormal forms of B27. The second known function of ARTS1 is that it cleaves cell surface receptors for the pro-inflammatory cytokines IL-1 (IL-1R2) [37], IL-6 (IL-6R
) [38] and tumour necrosis factor (TNF) (TNFR1) [39], thereby down-regulating their signalling. Loss-of-function ARTS1 variants could thus have pro-inflammatory effects through this mechanism. Further research will be required as to the mechanism, but one clue in this research is that no evidence was seen for association with ARTS1 in a large study of ulcerative colitis (n = 1011) and Crohn's disease (n = 755) cases compared with healthy controls (n = 633). Neither of these conditions is HLA Class I mediated, whereas up-regulation of pro-inflammatory cytokines such as IL-1, IL-6 and TNF is a feature of these conditions. This suggests that it is the peptide trimming function of ARTS1 which explains the mechanism of the association with AS, though, of course, much more direct evidence will be required to confirm this theory.
Other genes for which association with AS has been confirmed include the IL-1 gene complex and CYP2D6 [26, 27]. Several reports of association with IL-1 gene complex members have been made [22–25, 40, 41]. Like the MHC, this is a very complex locus to study, being subject to strong evolutionary pressure and with marked preservation of genetic sequence in haplotypes, making it very difficult to identify true disease-associated variants. The most consistent associations have been with IL1A variants, which are also associated with psoriatic arthritis [42]. The relative inefficacy of IL-1 inhibition with anakinra has unfortunately discouraged further research of IL-1 as a potential therapeutic target in AS, despite these genetic findings.
Unconfirmed reports of association with AS have been published with several other candidate genes including ANKH [43, 44], TLR4 [45–48], CARD15 [49–57] and KIR [58]. Of these, TLR4 and CARD15 were investigated in the WTCCC study, and neither showed any association with disease (P > 0.1), and thus are very unlikely to be truly associated with the condition. Insufficient markers were studied in the other genes, which await further studies to determine their true significance.
|
MHC genetics of ankylosing spondylitis |
|---|
There are now over 40 different allelic variants of HLA-B27 reported, all of which are ancestrally related to B*2705. This variant is found in 95% of British Caucasians [59]. Two subtypes, HLA-B*2706 and B*2709, have been demonstrated to have reduced penetrance for AS [60, 61]. No cases of AS have yet been documented that carry B*2709, whereas the B*2706 subtype is incompletely protective, and a small number of cases have been reported carrying this allele. B*2709 is common in Sardinia but is very rare elsewhere. B*2706 is typically found in people of Asian descent, and is particularly common in Malay ethnic groups compared with Chinese [61]. The marked effects that minor allelic variations in B27 have on penetrance have provided useful clues in research into the mechanism underlying the association of B27 with AS. The polymorphic residues involved in determining the strength of association of the various subtypes must be important in the mechanism of B27's association. An ultimate test of hypotheses regarding the mechanism of association is whether the hypothesis can explain the much reduced or absent association of B*2706 and B*2709. It is difficult to envisage how some hypotheses, such as homodimer formation (which depends on the invariant Cys67 residue), or mechanisms related to inefficient folding and endoplasmic reticulum stress, can explain the observed variation in strength of association of B27 subtypes.
The MHC, situated on chromosome 6 (6p21.3), extends over 3.6 Mb and contains about 220 genes, many of which have immunoregulatory functions. There is compelling evidence that the MHC contains several other non-B27 determinants of disease susceptibility, including the HLA-B allele HLA-B60 and non-HLA-B genes. The association of HLA-B60 with AS is much weaker than the association with B27, with an odds ratio (OR) of 3.6 [59]. It is uncertain as to whether HLA-B60 is also disease-causing itself, or a marker of an MHC haplotype bearing other disease-causing genes. The association of HLA-B60 with disease is well established in B27-positive cases [59, 62], and there is data suggesting a role in B27-negative AS [59, 63]. An interesting report exists suggesting an association of HLA*B1403 and AS in West Africans with AS. If confirmed, this finding may advance research into how B27 causes AS [64], as B*1403 has a similar peptide binding motif to HLA-B27 [65].
To identify other MHC genes involved in AS, investigators have studied association of other MHC Class II and Class III genes with disease. We have reported association of HLA-DRB1*01 with AS in a B27-matched case–control study [66] and in twins [2]. Carriage of HLA-DRB1*08 in B27-positive AS cases is associated with younger age of symptom onset and the occurrence of uveitis [67, 68], and weakly with susceptibility to AS [64], suggesting that non-B27 MHC genes may also influence severity and clinical manifestations of AS. Several small association studies have implicated other MHC genes in AS, although the studies have been too small and targeted to determine whether these are primary associations or due to linkage disequilibrium with other loci (reviewed in [69]).
The WTCCC AS study genotyped 897 MHC SNPs in 1000 AS cases and 1500 healthy controls. These studies showed extremely strong and broad association of the MHC with AS, with association with P-values <10–50 present from 30.9 to 32.5 Mb from the p-telomere of chromosome 6 (Fig. 1). As the controls in this analysis are not matched for HLA-B27 with the cases, this association probably reflects both linkage disequilibrium with HLA-B27, and the presence of non-B27 MHC-associated genes.
|
To test for the presence of non-B27 MHC associations of AS, we recently completed a study of B27-matched MHC haplotypes in cases and controls [70]. Comparing B27-matched case and control haplotypes, strong association was observed with DRB1 irrespective of whether the haplotype carried HLA-B27 (B27-positive strand P = 4 x 10–4, B27-negative strand P = 5 x 10–8). Specific MHC haplotypes were demonstrated to have highly significant association with AS, controlling fully for the carriage of B27 (Table 2). The effect size of these associations is substantial. The population attributable fractions from specific MHC haplotypes range from 16.5% for the B27–/DR1+ haplotype, down to 3.5% for the B27+/DRB1*07 [71]; assuming an additive model, the attributable risk from these haplotypes is 34%. This strongly suggests that further studies of the MHC for AS-susceptibility genes other than B27 are likely to be quite fruitful.
|
|
Future studies |
|---|
In an ongoing study, a consortium of Australian, British and North American investigators (the TASC group) is performing a genome-wide association study in AS. It is expected that this will be completed early in 2008, and will involve 2000 AS cases and >6000 healthy controls. The record of previous genome-wide association studies suggests that this scan is likely to produce a small number of clear-cut hits (low-hanging fruit), and a moderately large number of intermediate strength associations (higher-hanging fruit) which will require further studies to determine their true significance. This scan will also be underpowered to detect genes involved in disease severity, although arms of the consortium are investigating genetic determinants of clinical and radiographic severity. Other areas which will require further research include the role of copy number variation and methylation patterns, and the genetic determinants of the disease in other populations. There is a notable paucity of non-MHC genetic data about both AS and rheumatoid arthritis in Asian populations, despite clear evidence in the case of rheumatoid arthritis at least that the genetic determinants are different in Asian ethnic groups.
Thus, there will be a need for further studies involving even more cases. Studies in different ethnic groups may also be required to identify genes specific to those ethnic groups, and to assist with defining the true associated variants in involved genes. Collaboration will be the key to the success of these studies. The record demonstrates that genetic studies involving a few hundred cases and controls are underpowered, and even positive findings are insufficiently robust to be valuable, leading some commentators to the conclusion that ‘association analysis ... has thus far led to a whole lot of nothing’ [72]. Successful follow-up studies to genome-wide association studies in other common diseases have required the study of thousands of cases to produce convincingly clear-cut findings, generally requiring the collaboration of many research and clinical groups. Thus, whilst the TASC study will provide an unprecedented view of the genetic determinants of AS susceptibility and clinical manifestations, it should be considered the foundation of future research programmes into AS aetiopathogenesis rather than answering all the questions itself.
|
Conclusions |
|---|
The successful identification of ARTS1 and IL23R should give those involved in AS genetics research great encouragement of the potential of this research. The main requirements now will be the support of AS patients, the rheumatology community and funding agencies, to complete the task. We have shown that hypothesis-free genetics research can identify the genes which are the main determinants of who develops AS, and its clinical manifestations. The challenge will then be for the hypothesis-driven researchers to work out why these genes are associated with AS, and what to do about it. You never know, it may end up explaining why HLA-B27 causes the disease.
|
Acknowledgements |
|---|
The studies described in this review have involved many colleagues whose vital contribution I would like to acknowledge. I would also like to thank the thousands of individuals who have participated in the studies described and without whom the progress I describe would not have been made.
Funding: M.A.B. is funded by an NHMRC (Australia) Principal Research Fellowship. Research presented in this review was largely funded by the Arthritis Research Campaign (UK), Wellcome Trust and National Institute of Arthritis, Musculoskeletal and Skin Diseases (USA).
Disclosure statement: The author has declared no conflict of interest.
|
References |
|---|
- Taurog JD, Richardson JA, Croft JT, et al. The germfree state prevents development of gut and joint inflammatory disease in HLA-B27 transgenic rats. J Exp Med (1994) 180:2359–64.
[Abstract/Free Full Text] - Brown MA, Kennedy LG, MacGregor AJ, et al. Susceptibility to AS in twins: the role of genes, HLA, and the environment. Arthritis Rheum (1997) 40:1823–8.[Web of Science][Medline]
- Pedersen O, Svendsen A, Ejstrup L, Skytthe A, Harris J, Junker P. Heritability estimates on ankylosing spondylitis. Clin Exp Rheumatol (2006) 24:463.
- Brown MA, Brophy S, Bradbury L, et al. Identification of major loci controlling clinical manifestations of ankylosing spondylitis. Arthritis Rheum (2003) 48:2234–9.[CrossRef][Web of Science][Medline]
- Hamersma J, Cardon LR, Bradbury L, et al. Is disease severity in ankylosing spondylitis genetically determined? Arthritis Rheum (2001) 44:1396–400.[CrossRef][Web of Science][Medline]
- Brophy S, Hickey S, Menon A, et al. Concordance of disease severity among family members with ankylosing spondylitis? J Rheumatol (2004) 31:1775–8.
[Abstract/Free Full Text] - Rowland-Jones S, Colbert RA, Dong T, et al. Distinct recognition of closely-related HIV-1 and HIV-2 cytotoxic T-cell epitopes presented by HLA-B*2703 and B*2705. Aids (1998) 12:1391–3.[Web of Science][Medline]
- Risch N. Linkage strategies for genetically complex traits. I. Multilocus models. Am J Hum Genet (1990) 46:222–8.[Web of Science][Medline]
- Brown MA, Laval SH, Brophy S, Calin A. Recurrence risk modelling of the genetic susceptibility to ankylosing spondylitis. Ann Rheum Dis (2000) 59:883–6.
[Abstract/Free Full Text] - Lander ES, Botstein D. Strategies for studying heterogeneous genetic traits in humans by using a linkage map of restriction fragment length polymorphisms. Proc Natl Acad Sci USA (1986) 83:7353–7.
[Abstract/Free Full Text] - Hugot JP, Chamaillard M, Zouali H, et al. Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn's disease. Nature (2001) 411:599–603.[CrossRef][Medline]
- Laurin N, Brown JP, Morissette J, Raymond V. Recurrent mutation of the gene encoding sequestosome 1 (SQSTM1/p62) in Paget disease of bone. Am J Hum Genet (2002) 70:1582–8.[CrossRef][Web of Science][Medline]
- Ferrari SL, Deutsch S, Choudhury U, et al. Polymorphisms in the low-density lipoprotein receptor-related protein 5 (LRP5) gene are associated with variation in vertebral bone mass, vertebral bone size, and stature in whites. Am J Hum Genet (2004) 74:866–75.[CrossRef][Web of Science][Medline]
- Koay M, Woon P-Y, Zhang Y, et al. Influence of LRP5 polymorphisms on normal variation in BMD. J Bone Miner Res (2004) 19:1619–27.[CrossRef][Web of Science][Medline]
- Little R, Carulli J, Del Mastro R, et al. A mutation in the LDL receptor-related protein 5 gene results in the autosomal dominant high-bone-mass trait. Am J Hum Genet (2002) 70:11–19.[CrossRef][Web of Science][Medline]
- Consortium WTCC. Genomewide association study of 14,000 cases of seven common diseases and 3000 controls. Nature (2007) 447:661–83.[CrossRef][Web of Science][Medline]
- Zhang G, Luo J, Bruckel J, et al. Genetic studies in familial ankylosing spondylitis susceptibility. Arthritis Rheum (2004) 50:2246–54.[CrossRef][Web of Science][Medline]
- Miceli-Richard C, Zouali H, Said-Nahal R, et al. Significant linkage to spondyloarthropathy on 9q31-34. Hum Mol Genet (2004) 13:1641–8.
[Abstract/Free Full Text] - Laval SH, Timms A, Edwards S, et al. Whole-genome screening in ankylosing spondylitis: evidence of non-MHC genetic-susceptibility loci. Am J Hum Genet (2001) 68:918–26.[CrossRef][Web of Science][Medline]
- Brown MA, Pile KD, Kennedy LG, et al. A genome-wide screen for susceptibility loci in ankylosing spondylitis. Arthritis Rheum (1998) 41:588–95.[CrossRef][Web of Science][Medline]
- Carter KW, Pluzhnikov A, Timms AE, et al. Combined analysis of three whole genome linkage scans for ankylosing spondylitis. Rheumatology (2007) 46:763–71.
[Abstract/Free Full Text] - Timms AE, Crane AM, Sims AM, et al. The interleukin 1 gene cluster contains a major susceptibility locus for ankylosing spondylitis. Am J Hum Genet (2004) 75:587–95.[CrossRef][Web of Science][Medline]
- van der Paardt M, Crusius JB, Garcia-Gonzalez MA, et al. Interleukin-1beta and interleukin-1 receptor antagonist gene polymorphisms in ankylosing spondylitis. Rheumatology (2002) 41:1419–23.
[Abstract/Free Full Text] - McGarry F, Neilly J, Anderson N, Sturrock R, Field M. A polymorphism within the interleukin 1 receptor antagonist (IL-1Ra) gene is associated with ankylosing spondylitis. Rheumatology (2001) 40:1359–64.
[Abstract/Free Full Text] - Chou CT, Timms AE, Wei JC, Tsai WC, Wordsworth BP, Brown MA. Replication of association of IL1 gene complex members with ankylosing spondylitis in Taiwanese Chinese. Ann Rheum Dis (2006) 65:1106–9.
[Abstract/Free Full Text] - Brown MA, Edwards S, Hoyle E, et al. Polymorphisms of the CYP2D6 gene increase susceptibility to ankylosing spondylitis. Hum Mol Genet (2000) 9:1563–6.
[Abstract/Free Full Text] - Beyeler C, Armstrong M, Bird HA, Idle JR, Daly AK. Relationship between genotype for the cytochrome P450 CYP2D6 and susceptibility to ankylosing spondylitis and rheumatoid arthritis. Ann Rheum Dis (1996) 55:66–8.
[Abstract/Free Full Text] - WTCCC TASC. A genome-wide scan of 14,000 non-synonymous coding SNPs in 5,500 individuals: The Wellcome Trust Case Control Consortium. Nat Genet (2007) Oct 22; [Epub ahead of print].
- Park H, Li Z, Yang XO, et al. A distinct lineage of CD4 T cells regulates tissue inflammation by producing interleukin 17. Nat Immunol (2005) 6:1133–41.[CrossRef][Web of Science][Medline]
- Cua DJ, Sherlock J, Chen Y, et al. Interleukin-23 rather than interleukin-12 is the critical cytokine for autoimmune inflammation of the brain. Nature (2003) 421:744–8.[CrossRef][Medline]
- Murphy CA, Langrish CL, Chen Y, et al. Divergent pro- and antiinflammatory roles for IL-23 and IL-12 in joint autoimmune inflammation. J Exp Med (2003) 198:1951–7.
[Abstract/Free Full Text] - Duerr RH, Taylor KD, Brant SR, et al. A genome-wide association study identifies IL23R as an inflammatory bowel disease gene. Science (2006) 314:1461–3.
[Abstract/Free Full Text] - Tremelling M, Cummings F, Fisher SA, et al. IL23R variation determines susceptibility but not disease phenotype in inflammatory bowel disease. Gastroenterology (2007) 132:1657–64.[CrossRef][Web of Science][Medline]
- Cargill M, Schrodi S, Chang M, et al. A large-scale genetic association study confirms IL12B and leads to the identification of IL23R as psoriasis-risk genes. Am J Hum Genet (2007) 80:273–90.[CrossRef][Web of Science][Medline]
- Hammer GE, Gonzalez F, Champsaur M, Cado D, Shastri N. The aminopeptidase ERAAP shapes the peptide repertoire displayed by major histocompatibility complex class I molecules. Nat Immunol (2006) 7:103–12.[CrossRef][Web of Science][Medline]
- Kanaseki T, Blanchard N, Hammer GE, Gonzalez F, Shastri N. ERAAP synergizes with MHC class I molecules to make the final cut in the antigenic peptide precursors in the endoplasmic reticulum. Immunity (2006) 25:795–806.[CrossRef][Web of Science][Medline]
- Cui X, Rouhani FN, Hawari F, Levine SJ. Shedding of the type II IL-1 decoy receptor requires a multifunctional aminopeptidase, aminopeptidase regulator of TNF receptor type 1 shedding. J Immunol (2003) 171:6814–9.
[Abstract/Free Full Text] - Cui X, Rouhani FN, Hawari F, Levine SJ. An aminopeptidase, ARTS-1, is required for interleukin-6 receptor shedding. J Biol Chem (2003) 278:28677–85.
[Abstract/Free Full Text] - Cui X, Hawari F, Alsaaty S, et al. Identification of ARTS-1 as a novel TNFR1-binding protein that promotes TNFR1 ectodomain shedding. J Clin Invest (2002) 110:515–26.[CrossRef][Web of Science][Medline]
- Vazquez-Del MM, Garcia-Gonzalez A, Munoz-Valle JF, et al. Interleukin 1beta (IL-1beta), IL-10, tumor necrosis factor-alpha, and cellular proliferation index in peripheral blood mononuclear cells in patients with ankylosing spondylitis. J Rheumatol (2002) 29:522–6.
[Abstract/Free Full Text] - Maksymowych WP, Rahman P, Reeve JP, Gladman DD, Peddle L, Inman RD. Association of the IL1 gene cluster with susceptibility to ankylosing spondylitis: an analysis of three Canadian populations. Arthritis Rheum (2006) 54:974–85.[CrossRef][Web of Science][Medline]
- Rahman P, Sun S, Peddle L, et al. Association between the interleukin-1 family gene cluster and psoriatic arthritis. Arthritis Rheum (2006) 54:2321–5.[CrossRef][Web of Science][Medline]
- Timms AE, Zhang Y, Bradbury L, Wordsworth BP, Brown MA. Investigation of the role of ANKH in ankylosing spondylitis. Arthritis Rheum (2003) 48:2898–902.[CrossRef][Web of Science][Medline]
- Tsui FW, Tsui HW, Cheng EY, et al. Novel genetic markers in the 5'-flanking region of ANKH are associated with ankylosing spondylitis. Arthritis Rheum (2003) 48:791–7.[CrossRef][Web of Science][Medline]
- Adam R, Sturrock RD, Gracie JA. TLR4 mutations (Asp299Gly and Thr399Ile) are not associated with ankylosing spondylitis. Ann Rheum Dis (2006) 65:1099–101.
[Abstract/Free Full Text] - Gergely P Jr, Blazsek A, Weiszhar Z, Pazar B, Poor G. Lack of genetic association of the Toll-like receptor 4 (TLR4) Asp299Gly and Thr399Ile polymorphisms with spondylarthropathies in a Hungarian population. Rheumatology (2006) 45:1194–6.
[Abstract/Free Full Text] - Snelgrove T, Lim S, Greenwood C, et al. Association of toll-like receptor 4 variants and ankylosing spondylitis: a case-control study. J Rheumatol (2007) 34:368–70.
[Abstract/Free Full Text] - van der Paardt M, Crusius JB, de Koning MH, et al. No evidence for involvement of the Toll-like receptor 4 (TLR4) A896G and CD14-C260T polymorphisms in susceptibility to ankylosing spondylitis. Ann Rheum Dis (2005) 64:235–8.
[Abstract/Free Full Text] - Crane AM, Bradbury L, van Heel DA, et al. Role of NOD2 variants in spondylarthritis. Arthritis Rheum (2002) 46:1629–33.[CrossRef][Web of Science][Medline]
- Miceli-Richard C, Zouali H, Lesage S, et al. CARD15/NOD2 analyses in spondylarthropathy. Arthritis Rheum (2002) 46:1405–6.[CrossRef][Web of Science][Medline]
- van der Paardt M, Crusius JB, de Koning MH, et al. CARD15 gene mutations are not associated with ankylosing spondylitis. Genes Immun (2003) 4:77–8.[CrossRef][Web of Science][Medline]
- van Heel DA, McGovern DP, Cardon LR, et al. Fine mapping of the IBD1 locus did not identify Crohn disease-associated NOD2 variants: implications for complex disease genetics. Am J Med Genet (2002) 111:253–9.[CrossRef][Web of Science][Medline]
- van Heel DA, Dechairo BM, Dawson G, et al. The IBD6 Crohn's disease locus demonstrates complex interactions with CARD15 and IBD5 disease-associated variants. Hum Mol Genet (2003) 12:2569–75.
[Abstract/Free Full Text] - D’Amato M. The Crohn's associated NOD2 3020InsC frameshift mutation does not confer susceptibility to ankylosing spondylitis. J Rheumatol (2002) 29:2470–1.
[Free Full Text] - Ferreiros-Vidal I, Amarelo J, Barros F, Carracedo A, Gomez-Reino JJ, Gonzalez A. Lack of association of ankylosing spondylitis with the most common NOD2 susceptibility alleles to Crohn's disease. J Rheumatol (2003) 30:102–4.[Web of Science][Medline]
- Peeters H, Vander Cruyssen B, Laukens D, et al. Radiological sacroiliitis, a hallmark of spondylitis, is linked with CARD15 gene polymorphisms in patients with Crohn's disease. Ann Rheum Dis (2004) 63:1131–4.
[Abstract/Free Full Text] - Kim TH, Rahman P, Jun JB, et al. Analysis of CARD15 polymorphisms in Korean patients with ankylosing spondylitis reveals absence of common variants seen in Western populations. J Rheumatol (2004) 31:1959–61.
[Abstract/Free Full Text] - Lopez-Larrea C, Blanco-Gelaz MA, Torre-Alonso JC, et al. Contribution of KIR3DL1/3DS1 to ankylosing spondylitis in human leukocyte antigen-B27 Caucasian populations. Arthritis Res Ther (2006) 8:R101.[CrossRef][Medline]
- Brown MA, Pile KD, Kennedy LG, et al. HLA class I associations of ankylosing spondylitis in the white population in the United Kingdom. Ann Rheum Dis (1996) 55:268–70.
[Abstract/Free Full Text] - D’Amato M, Fiorillo MT, Carcassi C, et al. Relevance of residue 116 of HLA-B27 in determining susceptibility to ankylosing spondylitis. Eur J Immunol (1995) 25:3199–201.[Web of Science][Medline]
- Nasution AR, Mardjuadi A, Kunmartini S, et al. HLA-B27 subtypes positively and negatively associated with spondyloarthropathy. J Rheumatol (1997) 24:1111–4.[Web of Science][Medline]
- Robinson WP, van der Linden SM, Khan MA, et al. HLA-Bw60 increases susceptibility to ankylosing spondylitis in HLA-B27+ patients. Arthritis Rheum (1989) 32:1135–41.[Web of Science][Medline]
- Wei JC, Tsai WC, Lin HS, Tsai CY, Chou CT. HLA-B60 and B61 are strongly associated with ankylosing spondylitis in HLA-B27-negative Taiwan Chinese patients. Rheumatology (2004) 43:839–42.
[Abstract/Free Full Text] - Lopez-Larrea C, Mijiyawa M, Gonzalez S, Fernandez-Morera JL, Blanco-Gelaz MA, Martinez-Borra J, Lopez-Vazquez A. Association of ankylosing spondylitis with HLA-B*1403 in a West African population. Arthritis Rheum (2002) 46:2968–71.[CrossRef][Web of Science][Medline]
- Merino E, Montserrat V, Paradela A, Lopez de Castro JA. Two HLA-B14 subtypes (B*1402 and B*1403) differentially associated with ankylosing spondylitis differ substantially in peptide specificity but have limited peptide and T-cell epitope sharing with HLA-B27. J Biol Chem (2005) 280:35868–80.
[Abstract/Free Full Text] - Brown MA, Kennedy LG, Darke C, et al. The effect of HLA-DR genes on susceptibility to and severity of ankylosing spondylitis. Arthritis Rheum (1998) 41:460–5.[CrossRef][Web of Science][Medline]
- Ploski R, Maksymowych W, Forre O. HLA-DR8 and susceptibility to acute anterior uveitis in ankylosing spondylitis: comment on the article by Monowarul Islam et al. Arthritis Rheum (1996) 39:351–2.[Web of Science][Medline]
- Monowarul Islam SM, Numaga J, Fujino Y, et al. HLA-DR8 and acute anterior uveitis in ankylosing spondylitis. Arthritis Rheum (1995) 38:547–50.[Medline]
- Sims AM, Wordsworth BP, Brown MA. Genetic susceptibility to ankylosing spondylitis. Curr Mol Med (2004) 4:13–20.[CrossRef][Web of Science][Medline]
- Sims AM, Barnardo M, Herzberg I, et al. Non-B27 MHC associations of ankylosing spondylitis. Genes Immun (2007) 8:115–23.[CrossRef][Web of Science][Medline]
- Miettinen OS. Proportion of disease caused or prevented by a given exposure, trait or intervention. Am J Epidemiol (1974) 99:325–32.
[Abstract/Free Full Text] - Terwilliger JD, Weiss KM. Linkage disequilibrium mapping of complex disease: fantasy or reality? Curr Opin Biotechnol (1998) 9:578–94.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Karaderi, D. Harvey, C. Farrar, L. H. Appleton, M. A. Stone, R. D. Sturrock, M. A. Brown, P. Wordsworth, and J. J. Pointon Association between the interleukin 23 receptor and ankylosing spondylitis is confirmed by a new UK case-control study and meta-analysis of published series Rheumatology, April 1, 2009; 48(4): 386 - 389. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. W. Quinn, N. A. Sims, H. Saleh, D. Mirosa, K. Thompson, S. Bouralexis, E. C. Walker, T. J. Martin, and M. T. Gillespie IL-23 Inhibits Osteoclastogenesis Indirectly through Lymphocytes and Is Required for the Maintenance of Bone Mass in Mice J. Immunol., October 15, 2008; 181(8): 5720 - 5729. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Pradeep, A. Keat, and K. Gaffney Predicting outcome in ankylosing spondylitis Rheumatology, July 1, 2008; 47(7): 942 - 945. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||