Introduction

The regulation of the immune system relies on the complex interaction within the cytokine network, where the products of cytokine genes and their respective receptor genes must be finely tuned to maintain an optimal physiological environment. The interleukin-1 (IL-1) system epitomises the highly regulated mechanism required for this to be achieved. Several related gene products interact with each other to adjust the response according to cellular requirements. IL-1 signalling via the interleukin-1 receptor type I (IL-1RI) is an important pathway in the inflammatory process, innate immunity and the immune response.¹ Its effects are various, including the stimulation of fibroblast and keratinocyte proliferation, cartilage breakdown, angiogenesis and increased production of acute phase response proteins by the liver.² These physiological effects are mediated at the molecular level by the activation of JNK and p38 MAP kinases, and consequent upregulation of genes via the transcription factors NF-B, AP-1 and C/EBP³

The products of the IL-1 gene cluster are thought to play a role in a number of diseases, one of which is osteoarthritis (OA). This disease affects an estimated 5 million people in the UK⁴ and over 16 million in the US,⁵ and is characterised by focal loss of cartilage matrix in the affected joints. IL-1 is thought to contribute to this loss by stimulating chondrocytes to produce cartilage matrix degrading enzymes. In support of this contention, a strong correlation exists between specific cleavage of the aggrecan core protein at the Glu373–Ala374 bond and the release of aggrecan catabolites in response to IL-1:⁶ fragments bearing this epitope are found in synovial fluids of OA patients.⁷ In addition, cells from OA synovium produce more IL-1 than normal synovium,^{8, 9} and similarly, chondrocytes for OA cartilage secrete both IL-1 and IL-1 unlike normal chondrocytes.^{10, 11} It is also known that OA cartilage explants are more susceptible to the effects of IL-1 than similar explants from nonarthritic cartilage¹² and that this susceptibility is related to the expression of IL-1 receptor types 1 and 2 on chondrocytes.¹³

A further line of evidence comes from genetic studies. First-degree relatives of patients with generalised radiologic OA are twice as likely to develop the disease as people in the general population,¹⁴ and monozygotic twins are significantly more concordant for hand and knee osteoarthritis than dizygotic twins.¹⁵ Thus, a genetic predisposition to OA exists. Interestingly, some genome-wide scans for predisposing genetic loci to OA reveal potential linkage with the IL-1 gene cluster on chromosome 2.^{16, 17} Taken together, these results implicate both IL-1 and its receptors in the progression of OA in at least some patients, and suggest that a search to determine if genes in the IL-1 cluster are associated with OA is justified.

IL-1 acts through its signalling receptor, IL-1 receptor type 1 (IL-1RI), and the degree of expression of this receptor influences the response of cells to IL-1.¹⁸ The cloning and characterisation of the two IL-1 receptors (type 1 and 2)¹⁹ has revealed the IL-1 receptor type 2 to function as an IL-1 binding protein that acts as a competitive decoy, lacking the cytoplasmic residues necessary for intracellular signalling.²⁰ Three structurally related ligands can bind to the IL-1 receptors: IL-1 and the more abundant IL-1 functioning as agonists, and IL-1Ra as an antagonist that results in no intracellular signalling.

IL-1, IL-1, IL-1Ra and the two IL-1 receptors, type 1 (IL-1RI) and type 2 (IL-1RII), are encoded by a cluster of genes (IL1A, IL1B, IL1RN, IL1R1 and IL1R2, respectively) situated on chromosome 2 (Figure 1a). Recently, six new IL-1 family genes have been mapped between IL1B and IL1RN: IL1F7, IL1F9, IL1F6, IL1F8, IL1F5 and IL1F10.^{21, 22} These genes share the same common ancestor as IL-1 and IL-18, with regions of amino acid and genetic homology, but their function is currently uncharacterised. The IL1A, IL1B and IL1RN genes have been extensively examined for polymorphic nucleotide residues: the literature on these polymorphisms, their effect on gene expression, and their influence on human diseases has been reviewed.^{23, 24, 25}

Figure 1.

Organisation of the IL1R1–IL1A–IL1B–IL1RN gene cluster showing polymorphic nucleotide sequences. (a) Map of genes within the region, showing base positions on chromosome 2q according to the UCSC Human Genome Browser July 2003 Assembly (http://genome.cse.ucsc.edu/). Direction of transcription is shown beneath each gene by a horizontal arrow. (b) Exon–intron organisation of the IL1R1, IL1A, IL1B and IL1RN genes. Exons are shown as vertical lines or boxes. Exon : intron size ratios are shown to scale. For clarity, genes are not shown to scale: genomic sizes of genes are: IL1R1, 72.3 kb (14 exons: 1A, 1B, 1C, 2–11); IL1A, 10.55 kb (seven exons); IL1B, 7.03 kb (seven exons); IL1RN, 5.25 kb (four exons). Polymorphic nucleotide sequences examined in the IL1A, IL1B and IL1RN genes are shown. (c) Exon–intron organisation of the alternative IL1R1 promoters. Polymorphic nucleotide sequences are shown and those examined are denoted in bold typeface. TSS, transcription start site for the 1A promoter.

Full figure and legend (53K)

The IL1R1 gene (Figure 1) contains 14 exons and has three distinct promoters, generating three alternative transcripts with differing 5'-untranslated regions, derived from exons 1A, 1B or 1C.²⁶ Since the levels of transcription of certain cytokine genes are influenced by polymorphic nucleotides in the promoter sequences,²³ and IL-1RI is expressed on a wide variety of cells subjected to different evolutionary pressures (including those from microorganisms), we predicted that the promoter region of IL1RI should be highly polymorphic. Indeed, we, and others, have previously reported nucleotide sequence polymorphism in the promoter region spanning introns 1A–1C.^{27, 28, 29, 30, 31} In the present report, we describe the identification and characterisation of a further eight novel polymorphisms in the IL1R1 promoter. Recently, a number of reports have indicated that polymorphic IL-1 haplotypes or extended haplotypes, as opposed to single locus polymorphisms, may provide better genetic markers of transcriptional activity and disease association,³² in particular with the complex nature of the IL-1 agonist/antagonist relationship and the two IL-1 receptors.³³ To begin to address this concept with respect to osteoarthritis, this report examines the extent of linkage disequilibrium and the occurrence of haplotypes and extended haplotypes within the IL1R1–IL1A–IL1B–IL1RN region.

Results

Novel polymorphisms in the IL1R1 exon 1A promoter, exon 1A and intron 1A

Sequencing of amplicons showing PCR-SSCP variation revealed eight novel polymorphisms (rows 1–8, Table 1; Figure 2). We were also able to confirm and further characterise a previously described polymorphism in exon 1A at position +103 C>T (dbSNP, rs1019296). The two contiguous exon polymorphisms at positions +102 G>A and +103 C>T formed only three haplotypes from a possible four, with the genotypic combinations GT, GC and AT at the respective positions. These were considered as a single polymorphism with three alleles for the purposes of haplotype analysis, where GT=allele 1, GC=allele 2 and AT=allele 3 (Table 1). Further details of the polymorphisms identified in the study, and used in the subsequent haplotype analyses, are given in Table 1.

Figure 2.

Nucleotide sequence alignment of alleles within the complex microsatellite region of IL1R1 IVS1A. Alleles were designated numerically according to amplified fragment length (AFL) in base pairs: for details of the PCR primers and conditions used see Table 1 and text. Polymorphisms with respect to the reference allele 2 (GenBank AC007271) are: allele 1=102346097-8 del GT, 102346108 del G and allele 3=102346081 insGT, 102346099 insGT, 102346107-8 del GG. Chromosome 2 locations (top line) correspond to the UCSC July 2003 assembly (http://genome.cse.ucsc.edu/). Allele frequencies are shown in Table 1.

Full figure and legend (27K)

Table 1 - Specifications and assay conditions for polymorphisms in the IL1R1, IL1A, IL1B and IL1RN genes.

Full table

Linkage disequilibrium in the IL1R1–IL1A–IL1B–IL1RN region

The patient and control cohorts were genotyped for nine commonly occurring polymorphisms in the IL1R1 gene (Figure 1c) and a series of eight polymorphisms previously described in IL1A, IL1B and IL1RN genes (Table 1). Pairwise linkage disequilibrium (LD) was determined for all 17 loci in 195 control DNAs (Figure 3). This revealed a high level of LD between the polymorphic loci in IL1R1. Similarly, there was considerable intragenic and intergenic LD between polymorphisms in the IL1A–IL1B–IL1RN complex, particularly between the IL1A IVS4 TTA repeat, the IL1RN IVS2 86 bp VNTR and the IL1RN IVS3 +59 loci. Although the LD between the IL1A–IL1B–IL1RN complex and the IL1R1 promoter polymorphisms is much reduced, there are loci which show a significant degree of LD, for example three IL1R1 loci (Exon 1A +102/103 haplotype, IVS1A +463–464 insT and IVS1A +36995) with IL1B and IL1RN; and two IL1R1 loci (IVS1A +783 and IVS1A +35884) with IL1A IVS4 (TTA)n.

Figure 3.

Linkage disequilibrium analysis of polymorphisms in the IL1R1, IL1A, IL1B and IL1RN genes. P-values of the Lewontin's |D'| coefficient for each two-way comparison are as shown in the key; ns, not significant.

Full figure and legend (108K)

Haplotypes in the IL1R1 promoter and IL1A–IL1B–IL1RN complex and association with OA

Genotypes at the 17 polymorphic loci (Table 1) represented a total of 95 IL1R1–IL1A–IL1B–IL1RN haplotypes in the control subjects. In view of the overall weak LD between the IL1R1 promoter and the IL1A–IL1B–IL1RN complex, separate haplotype analyses were conducted. The nine IL1R1 promoter polymorphisms revealed a total of 33 haplotypes in controls and patients from Bristol and London with knee OA. A total of 299 of the 304 patients were included in the IL1R1 haplotype study: the five remaining individuals were excluded due to incomplete genotyping data. Table 2 shows details of the five haplotypes with a frequency of 0.05 or more in either controls or patients (haplotype nomenclature is explained in Table 2 footnote). There was no significant difference between any of the IL1R1 promoter haplotype frequencies in controls or patients.

Table 2 - IL1R1 haplotype analysis of controls and OA patients.

Full table

A parallel analysis of the eight polymorphisms in the IL1A–IL1B–IL1RN complex revealed a total of 37 haplotypes in controls and knee OA patients. Table 3 shows details of the nine haplotypes with a frequency of 0.05 or higher, in either controls or patients (haplotype nomenclature is explained in Table 3 footnote).

Table 3 - IL1A-IL1B-IL1RN haplotype analysis of controls and OA patients.

Full table

The Bristol knee OA cohort was examined first. Here, we found one haplotype (2C-CTG-1TT) was significantly more frequent in patients with knee OA than in controls (OR=3.93, CI=1.68–9.25, P=0.00043, P_c=0.0043). When sexes were analysed separately, the association remained significant (males, P=0.0069; females, P=0.0037). Further scrutiny of these data indicated that there may be a protective haplotype, masked by the inclusion of the IL1A genotypes. We thus conducted a further analysis of the six polymorphisms within the IL1B–IL1RN region. This revealed a total of 20 haplotypes in controls and patients. Table 4 shows details of the seven haplotypes with a frequency of 0.05 or higher, in either controls or patients (haplotype nomenclature is explained in Table 4 footnote). One haplotype (CCA-1TT) was significantly less frequent in patients than in controls (OR=0.37, CI=0.18–0.77, P=0.0036, P_c=0.029), representing a putative protective factor against knee OA. When sexes were analysed as subgroups, the trends were the same: the significance of the association reduced in males (P=0.051), although the association with females remained significant (P=0.014).

Table 4 - IL1B-IL1RN haplotype analysis of controls and OA patients.

Full table

To verify these associations in a different population, we duplicated the analyses in a knee OA patient cohort from London. The putative risk haplotype in the IL1A–IL1B–IL1RN region (2C-CTG-1TT) was also significantly increased in this OA cohort (OR=2.32, CI=1.05–5.08, P=0.02; Table 3). When sexes were analysed separately, the association remained significant (males, P=0.00036; females, P=0.038). A duplicate analysis of the IL1B–IL1RN region in the London OA cohort showed that the frequency of the putative protective haplotype was also significantly reduced (OR=0.19, CI=0.09–0.40, P=0.0000008; Table 4). As observed with the Bristol cohort, the statistical significance was lost in males, but was still highly significant in females (P=0.0000008), although the numbers (n=38) of males in the London cohort was low.

Discussion

IL-1 ligand–receptor signalling is a highly sensitive immune process, where even small changes in gene expression may result in widespread changes to the cell's environment. Indeed, fewer than 10 ligand-occupied IL-1 receptors are sufficient to transduce a strong intracellular signal.² IL-1 is a highly proinflammatory cytokine and the levels of signalling are greatly increased in many disorders, including autoimmune diseases. Recent work on IL-1-associated disorders has led researchers to search for predisposing loci within the highly polymorphic IL1R1–IL1A–IL1B–IL1RN gene cluster that may help to pinpoint genetic markers of disease susceptibility, severity or outcome. Impetus for examination of these loci has been further provided by results of genome-wide scans for susceptibility loci.^{16, 17} Much of the research to date has focused on individual polymorphisms or small numbers of polymorphisms in single genes. However, this approach may be oversimplistic due to the complex interaction of the ligands and receptors involved in the transduction and inhibition of IL-1 signalling, and because of linkage disequilibrium within the genes encoding the ligands and receptors. Since the immunopathology of osteoarthritis includes induction of high levels of IL-1 in the synovial fluid, we hypothesised that haplotype or extended haplotype analysis across a region encompassing the IL1R1, IL1A, IL1B and IL1RN genes might indicate haplotypic markers of disease susceptibility or severity which individual genotypic markers may fail to reveal.

From analysis of 17 polymorphic loci, we identified 95 extended IL1R1–IL1A–IL1B–IL1RN haplotypes. These data also revealed that strong LD exists between polymorphic loci in the IL1R1 promoter and in the IL1A–IL1B–IL1RN complex, but that between these two regions, considerably less LD exists. Allelic or haplotypic polymorphism within the IL1R1 promoter might involve alteration in transcription factor binding sites and thus exert a regulatory influence on receptor expression, thereby contributing to the aetiology or immunopathology of knee osteoarthritis. However, we found no association between promoter haplotypes in the IL1R1 gene and knee osteoarthritis. TRANSFAC 6.0 analysis (http://www.gene-regulation.com/pub/databases.html#transfac) of the IL1R1 promoter polymorphism at position -227 upstream of exon 1A, revealed a potential NF-B binding site motif. We tested the potential functional significance of this polymorphism using two reporter gene assays. However, no differences were observed in expression or response to IL-1 induction in -227G and -227A clones, either in a pGL3-SV40 luciferase reporter system, or in a lentiviral-GFP vector.

In contrast to the IL1R1 results, we identified an extended IL1A–IL1B–IL1RN risk haplotype (2C-CTG-1TT), which was significantly more frequent in patients with knee osteoarthritis. In an analysis of IL1B–IL1RN region haplotypes, one of these (CCA-1TT) was associated with a significant protective effect against knee osteoarthritis. In order to reconcile our extended haplotype approach with potential functional differences, we examined individual polymorphisms within the haplotypes which have been previously associated with regulation of expression. The IL1B–IL1RN CCA-1TT protective haplotype differed from the risk haplotype in the genotypes at positions -31 and -511 in the IL1B gene. These two polymorphisms are in significant LD (see Figure 3). The IL1B -31T allele creates a TATA box, which has been shown to bind TATA–binding protein (TBP),⁴¹ whereas the IL1B -31C allele does not. It has been suggested that the strongly linked IL1B -31T and the IL1B -511G alleles may be associated with higher levels of gene expression than the IL1B -31C and IL1B -511A alleles:^{41, 42} we are currently investigating this possibility. This could represent a partial functional explanation for the association of the risk haplotype (2C-CTG-1TT) with knee OA, since this haplotype contains the IL1B -31T and IL1B -511G genotypes, is TATA-box positive, and may be associated with high-level expression of IL-1 (a potent mediator of inflammation in OA joints). In contrast, the protective haplotype (CCA-1TT) contains the IL1B -31C genotype, is TATA-box negative, and is associated with lower levels of IL-1 production.

The importance of examining extended haplotypes of the IL-1 gene cluster is made clear by this study. We identified no association of the individual IL1B polymorphisms with osteoarthritis, but statistically significant associations when haplotypic analysis was conducted. The level of IL-1Ra expression will undoubtedly have a major impact on a cell's responsiveness to IL-1 and IL-1. Both the identified risk and protective extended haplotypes possess the IL1RN '1TT' haplotype, which includes the VNTR allele 1, implicated to express IL-1Ra at lower levels than VNTR allele 2.⁴³ It is believed, however, that IL-1Ra expression is also influenced by IL1B genotype: previous research has identified association between the IL1B -511, IL1B exon 5 +14 and IL1RN VNTR genotypes and their combined influence on the level of IL-1Ra plasma levels.³³ One logical explanation for our findings is that in the risk haplotype population, the IL1A and IL1B alleles may not only lead to high IL-1 production, but may also confer a lower level of IL-1Ra expression: for the protective haplotype, the opposite might be the case. This might explain the observation in some individuals, of both higher levels of IL-1 found in OA synovium, and the increased susceptibility of OA cartilage to IL-1, due to the concurrent reduction of IL-1Ra expression. Our results indicate potential IL1A–IL1B–IL1RN extended haplotypes associated with susceptibility to OA. This could help to explain previous reports of linkage from some groups between markers in this region and OA in genome-wide scans.^{16, 17} However, it should be stressed that the disease is a multigenic disorder and our data identifies only one of the susceptibility loci involved, each of which represent an as yet unknown proportion of the overall genetic influence.

Methods

Patients and controls

Osteoarthritis patients (n=304) were studied from two UK centres: Bristol (n=141, 65 male and 76 female Caucasians: mean age=62; median age=63); and London (n=163, 38 male and 125 female Caucasians: mean age=71; median age=71). All patients had osteoarthritis as defined by the ACR criteria with at least grade 2 radiological change. Control subjects were 195 (97 male and 98 female) unrelated healthy Caucasian volunteer blood donors from South West England. Informed consent was obtained with appropriate local ethics committee approval. Genomic DNA was isolated from peripheral blood leucocytes using a salting-out procedure.⁴⁴

Identification of novel polymorphisms in the IL1R1 exon 1A region

PCR-SSCP analysis⁴⁵ was used to determine the extent of nucleotide sequence polymorphism around the first untranslated exon of the human IL1R1 exon, using a large series of overlapping PCR primers derived from a published IL1R1 sequence (GenBank AC007271), designed to produce amplicons of approximately 250 bp, overlapping each other by 20–50 bp. DNAs from 195 healthy controls (see above) were screened for polymorphism, within the region 2 kb upstream to 4 kb downstream of the exon 1A transcription start site (Figure 1c). PCR conditions were essentially as described below. Following amplification, PCR-SSCP analysis was performed using polyacrylamide gel electrophoresis (PAGE) as described⁴⁵ in 12% nondenaturing polyacrylamide gels (Protogel, National Diagnostics, Atlanta, GA, USA) containing 1 TBE buffer. Electrophoresis was carried out at 300 V for 140 min on a triple-wide minigel system (CBS Scientific Company Inc., Del Mar, USA). DNA was visualised by staining for 10 min with ethidium bromide (0.5 g/ml in 1 TBE buffer) and gels were imaged using a Kodak EDAS 120 digital camera. Nucleotide sequencing of SSCP variants was performed by automated nucleotide sequencing on an ABI 310 sequencer (Perkin-Elmer, Norwalk, USA).

SNP and VNTR genotyping

For each polymorphic locus to be analysed, PCR amplifications were set up in a total volume of 20 l containing 200 ng genomic DNA, 0.1 U Taq polymerase (Abgene, Epsom, Surrey, UK), 0.05 M of each PCR primer (Table 1), 2 l of 10 reaction buffer (160 mM (NH₄)₂SO₄, 670 mM Tris-HCl (pH 8.8 at 25°C), 15 mM MgCl₂, 0.1% v/v Tween-20 (Ameresco Chemicals, Anachem, Luton, UK)), 0.8 l DNA polymerisation mix (Abgene). Amplification was carried out in a PTC-100 thermal cycler (MJ Research Inc., Waltham, MA, USA) using cycle parameters of: 95°C for 5 min (initial denaturation), then 35 cycles of: 95°C (30 s), annealing temp (see Table 1) (40 s) and 72°C (30 s); with a final extension for 5 min at 72°C.

RFLP analysis was carried out for the IL1R1 SNPs where appropriate (Table 1). The restriction endonucleases, MspI, PstI, HinfI, HincII, AluI, FokI, NlaIII and HhaI (New England Biolabs Inc., MD, USA), were used according to the manufacturer's instructions. Briefly, 4 U of enzyme were used per 10 l amplified DNA, and incubated at 37°C for 6 h. Following digestion, DNA samples were electrophoresed in 15% nondenaturing polyacrylamide gels for 60 min and imaged on a UV transilluminator using a Kodak EDAS 120 digital camera. Analysis of repeat sequences was carried out by PAGE using 12% Accugel (National Diagnostics, Atlanta, GA, USA) for 50 min (IL1RN IVS2 86 bp VNTR, IL1A IVS4 TTA repeat, and 100 min (IL1R1 IVS1A complex microsatellite) (see Table 1). For the IL1R1 –227 A>G SNP, induced heteroduplex generator (IHG) analysis was used, incorporating an [AAAA] insertion adjacent to the polymorphic site.³⁴ The same method was used to identify the polymorphisms IL1A –889 C>T, IL1B Exon 5 +14 C>T, IL1B –31 C>T, IL1B –511 G>A, IL1RN IVS3 +59 T>C and IL1RN Exon 4 +72 T>C (Table 1).

Linkage disequilibrium, haplotype frequency and disease association analysis

Allele frequencies were estimated by gene counting and potential deviation in Hardy–Weinberg equilibrium was tested using the ² test with 1df. For the control population Arlequin version 2.000⁴⁶ was used to determine pairwise linkage disequilibrium (LD), expressed as Lewontin's |D'| coefficient—statistical significance of the P-value was defined at the 5% level.

Haplotypes were constructed from genotypic data using the PHASE 2.0 software platform.^{47, 48} Haplotypes with an observed frequency of less than 0.05 were analysed as a combined group. To identify potential risk or protective haplotypes, haplotype frequency comparisons between control and OA patients were performed using haplotypic odds ratios (OR) with Cornfield 95% confidence intervals (CI). For all analyses, the reference haplotype was that which occurred most frequently.⁴⁹ Pearson's ² test or Fisher's exact test were used to determine uncorrected P-values and corrections were made for multiple comparisons (p_c) where appropriate.

References

1. Dinarello CA. Interleukin-1. Cytokine Growth Factor Rev 1997; 8: 253–265. | Article | PubMed | ChemPort |
2. Dinarello CA. Biologic basis for interleukin-1 in disease. Blood 1996; 87: 2095–2147. | PubMed | ISI | ChemPort |
3. O'Neill LA, Greene C. Signal transduction pathways activated by the IL-1 receptor family: ancient signaling machinery in mammals, insects, and plants. J Leukocyte Biol 998; 63: 650–657.
4. Felson D. Epidemiology of the rheumatic diseases. Arthritis and Allied Conditions 1993, 12th edn. McCarty and Koopman (Lea & Febiger): Malvern, PA.
5. Simon LS. Osteoarthritis: a review. Clin Cornerstone 1999; 2: 26–37. | Article | PubMed |
6. Arner EC, Hughes CE, Decicco CP, Caterson B, Tortorella MD. Cytokine-induced cartilage proteoglycan degradation is mediated by aggrecanase. Osteoarthritis Cartilage, Osteoarthritis Res Soc 1998; 6: 214–228. | Article |
7. Fosang AJ, Last K, Knauper V, Murphy G, Neame PJ. Degradation of cartilage aggrecan by collagenase-3 (MMP-13). FEBS Lett 1996; 380: 17–20. | Article | PubMed | ChemPort |
8. Westacott CI, Taylor G, Atkins R, Elson C. Interleukin-1 alpha and beta production by cells isolated from membranes around aseptically loose total joint replacements. Ann Rheum Dis 1992; 51: 638–642. | PubMed |
9. Elson CJ, Mortuza FY, Perry MJ, Warnock MG, Webb GR, Westacott CI. Cytokines and focal loss of cartilage in osteoarthritis. Br J Rheumatol 1998; 37: 106–107. | Article | PubMed |
10. Towle CA, Hung HH, Bonassar LJ, Treadwell BV, Mangham DC. Detection of interleukin-1 in the cartilage of patients with osteoarthritis: a possible autocrine/paracrine role in pathogenesis. Osteoarthritis Cartilage, Osteoarthritis Res Soc 1997; 5: 293–300.
11. Moos V, Rudwaleit M, Herzog V, Hohlig K, Sieper J, Muller B. Association of genotypes affecting the expression of interleukin-1beta or interleukin-1 receptor antagonist with osteoarthritis. Arthrit Rheum 2000; 43: 2417–2422. | Article |
12. Ismaiel S, Atkins RM, Pearse MF, Dieppe PA, Elson CJ. Susceptibility of normal and arthritic human articular cartilage to degradative stimuli. Br J Rheumatol 1992; 31: 369–373. | PubMed |
13. Martel-Pelletier J, McCollum R, DiBattista J et al. The interleukin-1 receptor in normal and osteoarthritic human articular chondrocytes. Identification as the type I receptor and analysis of binding kinetics and function. Arthrit Rheum 1992; 35: 530–540.
14. Kellgren J, Lawrence J, Bier F. Genetic factors in generalized osteo-arthrosis. Ann Rheum Dis 1963: 237–255. | PubMed | ChemPort |
15. Spector TD, Cicuttini F, Baker J, Loughlin J, Hart D. Genetic influences on osteoarthritis in women: a twin study. Br Med J 1996; 312: 940–943.
16. Leppavuori J, Kujala U, Kinnunen J et al. Genome scan for predisposing loci for distal interphalangeal joint osteoarthritis: evidence for a locus on 2q. Am J Hum Genet 1999; 65: 1060–1067. | Article | PubMed |
17. Loughlin J, Mustafa Z, Smith A et al. Linkage analysis of chromosome 2q in osteoarthritis. Rheumatology 2000; 39: 377–381. | Article | PubMed |
18. Sims JE, Gayle MA, Slack JL et al. Interleukin 1 signaling occurs exclusively via the type I receptor. Proc Natl Acad Sci USA 1993; 90: 6155–6159. | Article | PubMed | ChemPort |
19. Sims JE, Dower SK. Interleukin-1 receptors. Eur Cytokine Network 1994; 5: 539–546.
20. Colotta F, Re F, Muzio M et al. Interleukin-1 type II receptor: a decoy target for IL-1 that is regulated by IL-4. Science 1993; 261: 472–475. | Article | PubMed | ISI | ChemPort |
21. Taylor SL, Renshaw BR, Garka KE, Smith DE, Sims JE. Genomic organization of the interleukin-1 locus. Genomics 2002; 79: 726–733. | Article | PubMed |
22. Nicklin MJ, Barton JL, Nguyen M, FitzGerald MG, Duff GW, Kornman K. A sequence-based map of the nine genes of the human interleukin-1 cluster. Genomics 2002; 79: 718–725. | Article | PubMed | ChemPort |
23. Bidwell J, Keen L, Gallagher G et al. Cytokine gene polymorphism in human disease: on-line databases. Genes Immun 1999; 1: 3–19. | Article | PubMed | ISI | ChemPort |
24. Bidwell J, Keen L, Gallagher G et al. Cytokine gene polymorphism in human disease: on-line databases, supplement 1. Genes Immun 2001; 2: 61–70. | Article | PubMed | ISI | ChemPort |
25. Haukim N, Bidwell JL, Smith AJP et al. Cytokine gene polymorphism in human disease: on-line databases, supplement 2. Genes Immun 2002; 3: 313–330. | Article | PubMed | ISI | ChemPort |
26. Ye K, Vannier E, Clark BD, Sims JE, Dinarello CA. Three distinct promoters direct transcription of different 5' untranslated regions of the human interleukin 1 type I receptor: a possible mechanism for control of translation. Cytokine 1996; 8: 421–429. | Article | PubMed |
27. Bergholdt R, Karlsen AE, Johannesen J et al. Characterization of polymorphisms of an interleukin 1 receptor type 1 gene (IL1RI) promotor region (P2) and their relation to insulin-dependent diabetes mellitus (IDDM). The Danish Study Group of Diabetes in Childhood. Cytokine 1995; 7: 727–733. | Article | PubMed |
28. Bergholdt R, Larsen ZM, Andersen NA et al. Characterization of new polymorphisms in the 5' UTR of the human interleukin-1 receptor type 1 (IL1R1) gene: linkage to type 1 diabetes and correlation to IL-1RI plasma level. Genes Immun 2000; 1: 495–500. | Article | PubMed |
29. Sitara D, Olomolaiye O, Wood N et al. Identification of novel single nucleotide polymorphisms in intron 1B and exon 1C of the human interleukin-1 receptor type I (IL-1RI) gene. Genes Immun 1999; 1: 161–163. | Article | PubMed |
30. Sitara D, Wood N, Keen L, Morse H, Bidwell J. Characterization of a rare Sty I polymorphism in exon 1C of the human interleukin-1 receptor type I (IL-1RI) gene. Eur J Immunogenet 2000; 27: 103–104. | Article | PubMed |
31. Sitara D, Wood N, Morse H, Keen L, Bidwell J. A Bsr Bl polymorphism in exon 1C of the human interleukin-1 receptor type I (IL-1RI) gene. Genes Immun 2000; 1: 234–235. | Article | PubMed |
32. Cox A, Camp NJ, Nicklin MJ, di Giovine FS, Duff GW. An analysis of linkage disequilibrium in the interleukin-1 gene cluster, using a novel grouping method for multiallelic markers. Am J Hum Genet 1998; 62: 1180–1188. | Article | PubMed | ISI | ChemPort |
33. Hurme M, Santtila S. IL-1 receptor antagonist (IL-1Ra) plasma levels are co-ordinately regulated by both IL-1Ra and IL-1beta genes. Eur J Immunol 1998; 28: 2598–2602. | Article | PubMed | ISI | ChemPort |
34. Wood N, Bidwell J. Genetic screening and testing by induced heteroduplex formation. Electrophoresis 1996; 17: 247–254. | PubMed |
35. Zuliani G, Hobbs HH. A high frequency of length polymorphisms in repeated sequences adjacent to Alu sequences. Am J Hum Genet 1990; 46: 963–969. | PubMed | ISI | ChemPort |
36. Kornman KS, di Giovine FS. Genetic variations in cytokine expression: a risk factor for severity of adult periodontitis. Ann Periodontol 1998; 3: 327–338. | PubMed | ChemPort |
37. Pociot F, Molvig J, Wogensen L, Worsaae H, Nerup J. A TaqI polymorphism in the human interleukin-1 beta (IL-1 beta) gene correlates with IL-1 beta secretion in vitro. Eur J Clin Invest 1992; 22: 396–402. | PubMed | ISI | ChemPort |
38. Guasch JF, Bertina RM, Reitsma PH. Five novel intragenic dimorphisms in the human interleukin-1 genes combine to high informativity. Cytokine 1996; 8: 598–602. | Article | PubMed |
39. di Giovine FS, Takhsh E, Blakemore AI, Duff GW. Single base polymorphism at -511 in the human interleukin-1 beta gene (IL1 beta). Hum Mol Genet 1992; 1: 450. | PubMed | ChemPort |
40. Tarlow JK, Blakemore AI, Lennard A et al. Polymorphism in human IL-1 receptor antagonist gene intron 2 is caused by variable numbers of an 86-bp tandem repeat. Hum Genet 1993; 91: 403–404. | Article | PubMed | ISI | ChemPort |
41. El Omar EM, Carrington M, Chow WH et al. Interleukin-1 polymorphisms associated with increased risk of gastric cancer. Nature 2000; 404: 398–402. | Article | PubMed | ISI | ChemPort |
42. Santtila S, Savinainen K, Hurme M. Presence of the IL-1RA allele 2 (IL1RN^*2) is associated with enhanced IL-1beta production in vitro. Scand J Immunol 1998; 47: 195–198. | Article | PubMed | ISI | ChemPort |
43. Danis VA, Millington M, Hyland VJ, Grennan D. Cytokine production by normal human monocytes: inter-subject variation and relationship to an IL-1 receptor antagonist (IL-1Ra) gene polymorphism. Clin Exp Immunol 1995; 99: 303–310. | PubMed | ISI | ChemPort |
44. Miller SA, Dykes DD, Polesky HF. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 1988; 16: 1215. | Article | PubMed | ISI | ChemPort |
45. Orita M, Suzuki Y, Sekiya T, Hayashi K. Rapid and sensitive detection of point mutations and DNA polymorphisms using the polymerase chain reaction. Genomics 1989; 5: 874–879. | Article | PubMed | ISI | ChemPort |
46. Schneider S, Roessli D, Excoffier L. Arlequin: A Software for Population Genetics Data Analysis. Ver 2.000. Genetics and Biometry Lab, Department of Anthropology, University of Geneva, 2000.
47. Stephens M, Smith NJ, Donnelly P. A new statistical method for haplotype reconstruction from population data. Am J Hum Genet 2001; 68: 978–989. | Article | PubMed | ISI | ChemPort |
48. Stephens M, Donnelly P. A comparison of Bayesian methods for haplotype reconstruction from population genotype data. Am J Hum Genet 2003; 73: 1162–1169. | Article | PubMed | ISI | ChemPort |
49. 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, (on-line early).

Acknowledgements

We thank Dr Nigel Wood for oligonucleotide synthesis, Doris Culpan for nucleotide sequencing and Gabriela Surdulescu for sample preparation from the London OA cohort.