Prognostic factors in pemphigus vulgaris and pemphigus foliaceus.

BACKGROUND: Pemphigus typically has a chronic course, although there is great variability in disease duration (DD) and time taken to disease remission (DR) between individuals with the disease. The reasons for this are unclear. OBJECTIVES: To explore the prognostic influence of epidemiological, clinical, immunological and genetic factors on disease course and remission in pemphigus vulgaris (PV) and pemphigus foliaceus (PF). METHODS: This was a retrospective study of patients with PV and PF, recruited from a single UK centre. Direct and indirect immunofluorescence and enzyme-linked immunosorbent assay studies for antidesmoglein (Dsg) antibodies were used to assess immunological factors. Polymerase chain reaction with sequence specific primers (PCR-SSP) was used to assess the Class II human leukocyte antigen status of patients. Prognostic endpoints investigated were time to initial first DR and total DD. RESULTS: Ninety-five patients were recruited (79 PV and 16 PF). Patients of Indo-Asian origin were significantly associated with longer DD than White-British patients (P = 0.029). In addition, younger age at onset was associated with a worse prognosis in terms of DD: the mean age at presentation of patients with DD of < 5 years was 49 years (SEM = 3.4) compared with 40 years (SEM = 1.9) in those with DD > 5 years (P = 0.039). A higher initial intercellular antibody titre on normal human skin substrate was associated with a greater time to initial DR (P = 0.007) and high anti-Dsg 3 levels at baseline were associated with a longer total DD (P = 0.03). CONCLUSIONS: Ethnic group, age at presentation, initial intercellular antibody titre and initial Dsg 3 antibody levels all had a significant impact on prognosis of pemphigus.

Evidence of an association between desmoglein 3 haplotypes and pemphigus vulgaris.

BACKGROUND: Pemphigus vulgaris (PV, OMIM 169610) is a severe blistering disorder of the skin and mucous membranes, caused by the production of autoantibodies directed against the epithelial adhesive protein desmoglein 3. Although an association between PV and HLA class II alleles has been established, the genetic factors predisposing to the disease remain poorly understood, the rarity of PV hampering the recruitment of substantial patient cohorts. OBJECTIVES: To investigate DSG3 as a candidate PV susceptibility gene. METHODS: We examined five DSG3 single nucleotide polymorphisms (rs8085532, rs3911655, rs3848485, rs3794925 and rs1466379) in two case-control datasets respectively originating from the U.K. (62 PV patients, 154 controls) and northern India (28 patients, 98 controls). RESULTS: In the U.K. sample, we observed a significant association between PV and the DSG3*TCCTC haplotype (Fisher's exact test P = 0.002). A related haplotype (DSG3*TCCCC) was associated with PV in the Indian dataset (P = 0.002). We also found that all British and Indian patients bearing DSG3 risk haplotypes carried at least one copy of a PV-associated HLA allele. CONCLUSIONS: These results suggest that genetic variation of DSG3 may be an additive risk factor predisposing to PV and warrant further investigations of this gene.

Bullous pemphigoid antigen II (BP180) and its soluble extracellular domains are major autoantigens in mucous membrane pemphigoid: the pathogenic relevance to HLA class II alleles and disease severity.

BACKGROUND: Mucous membrane pemphigoid (MMP), a chronic autoimmune subepithelial blistering disease, is associated with circulating IgG and/or IgA autoantibodies against several basement membrane zone antigens. The heterogeneity of clinical presentation and diversity of target autoantigens have contributed to difficulties in characterizing this condition immunologically. OBJECTIVES: To analyse serum autoantibody profile and HLA class II alleles in MMP patients and to correlate this with the clinical presentation of disease. METHODS: Well-defined subgroups consisting of 124 patients with MMP were examined for IgG and IgA reactivity with immunoblotting using human epidermal, dermal and placental amnion proteins. The results were further analysed on the basis of detailed clinical (sites of involvement and disease severity) and immunopathological criteria (immunofluorescence study and HLA class II alleles). RESULTS: Immunoblot assay revealed that the majority of MMP patients had IgG (93 of 124, 75%) and/or IgA autoantibodies (63 of 124, 51%) to BP180 (including its soluble ectodomains, 120-kDa LAD-1 and 97-kDa LABD97 antigens). Other antigens targeted predominantly by IgG autoantibodies included: BP230 in 34 (27%), beta4 integrin in 26 (21%), and laminin 5 in three (2%). All the BP230+ sera and 23 (88%) beta4 integrin+ sera also reacted with at least one of the BP180 antigens. Over 85% of patients with reactivity to beta4 integrin had ocular involvement. In most cases of MMP, more severe clinical features were associated with antibody reactivity to multiple basement membrane zone antigens, as well as reactivity to multiple BP180 component antigens. Dual BP180/LAD-1 reactivity with IgG and IgA was associated with a more severe phenotype. In addition, the subset-dependent autoantibody reactivity correlated well with specific HLA class II alleles, DQB1*0301, DRB1*04 and DRB1*11. CONCLUSIONS: Our results confirmed that BP180 is a major autoantigen targeted by the sera of patients with MMP. The disease-prevalent HLA class II alleles and humoral autoimmune response against the particular subsets of antigenic epitope(s) within BP180 ectodomain may contribute to the clinicopathological significance and disease severity of MMP.

Associations of major histocompatibility complex class I chain-related molecule polymorphisms with Behcet's disease in Caucasian patients.

HLA-B*51 is known to be associated with Behcet's disease (BD) in many ethnic groups. The pathogenic gene, however, may lie close to the HLA-B locus and therefore be in linkage disequilibrium with HLA-B*51. On the basis of the proximity of MIC genes to HLA-B, their expression pattern and their affinity for the activating NKG2D receptor on natural killer (NK) cells and gammadelta T cells, these molecules have been postulated as susceptibility factors in BD. DNA from 56 western European Caucasians with BD and 90 Caucasian controls were analysed by polymerase chain reaction using allele-specific primers for MICA and MICB alleles. An increased allele frequency of MICA*009 was found in the BD patient group (25.0%) when compared with the controls (7.2%). This was associated with a corresponding decrease in MICA*008 in the BD patients (36.6%) compared with the controls (46.7%), which was not significant. MICA*009 was strongly associated with the presence of HLA-B*51 in patients and controls. No significant difference in frequency of MICB alleles was found between patients and controls. Both HLA-B*51 and MICA*009 are strongly associated with BD in a pure Caucasian BD patient group, and the two alleles are in linkage disequilibrium. No MICB allele was found to associate significantly with the disease, an unexpected finding considering the close proximity of the MICA and MICB loci. Our results suggest that while MICB does not influence the development of BD, polymorphisms in MICA may be pathogenic, perhaps through the interaction with NK and gammadelta T cells.

Are cytokine gene polymorphisms associated with outcome in patients with idiopathic intermediate uveitis in the United Kingdom?

BACKGROUND/AIM: Competing levels of cytokines, either locally within the eye or systemically, may influence the eventual outcome of ocular inflammation. Polymorphism in the promoter part of the genes controlling cytokine production may result in either higher or lower production of the relevant cytokine to a given stimulus. The authors hypothesised that such polymorphisms may relate to visual outcome in patients with idiopathic intermediate uveitis. METHODS: DNA was obtained from 125 patients with idiopathic intermediate uveitis and analysed for the interleukin 10 IL-10-1082G/Alpha and IL-10-819C/T, and interferon gamma IFNgamma 874T/A gene polymorphisms. Associations with disease were calculated by both allelic frequency and haplotype analysis, and associations between ocular disease outcomes and the presence of polymorphisms were identified. A bad outcome was defined as loss of vision <6/12 Snellen in both eyes at 5 years from presentation when the eyes were quiet. RESULTS: An initial screen showed that the 874T allele of the IFNgamma gene was more prevalent in patients than controls (chi2= 7.9; p = 0.004 OR 1.7; 95% CI 1.2 to 2.6 (Pc = 0.02), whereas the IL-10-1082/-819 AT haplotype of the interleukin 10 (IL-10) gene was not. Analysis of disease outcome showed an association between IL-10-1082 AA homozygosity and bad outcome (chi2= 13; p = 0.0003). Moreover, the two cytokine polymorphisms taken together showed that up to 75% of patients with a poor visual outcome had the combined IFNgamma 874TA or TT genotype together with the IL-10-1082AA genotype (chi2= 13.2 p = 0.0008 OR 6.4; 95% CI 1.85 to 23.6 Pc = 0.1). CONCLUSION: These results show that disease outcome in intermediate uveitis may be partly determined by a complex interplay between cytokine genes and these results may have implications for future treatment with biological agents that target these cytokines.

Corneodesmosin (CDSN) gene association with psoriasis vulgaris in Caucasian but not in Japanese populations.

PSORS1 on chromosome 6p21.3, which contains the MHC, is a major susceptibility locus for psoriasis vulgaris. This region is characterized by strong linkage disequilibrium and contains the corneodesmosin (CSDN) gene, an attractive candidate for psoriasis susceptibility based on its putative biological function in keratinocyte adhesion, and HLA-Cw6, an established marker for psoriasis susceptibility. We compared two genetically independent populations in order to define the major psoriasis susceptibility gene, a British Caucasian population comprising parent-offspring trios analysed by the transmission disequilibrium test (TDT) and a Japanese case-control population. All individuals were investigated for CDSN polymorphism (+619, +1236, +1240 and +1243) and HLA-C association. Our data confirms strong association with HLA-Cw6 and CDSN allele 5 (+619T, +1240G, +1243C) in the Caucasian cohort (TDT, P = 5.4 x 10(-6)) and in addition defines this region further by identifying a high-risk CDSN haplotype (allele 5 and +1236T, P = 8.5 x 10(-8)). In contrast no association was observed in the Japanese cohort for any HLA-C or CDSN alleles. This data supports a role for the CDSN gene in Caucasian populations with psoriasis. However the lack of association with HLA-Cw6 and CDSN alleles in Japanese psoriasis patients may be because Japanese patients exhibit a form of psoriasis similar to late onset or Type II psoriasis vulgaris in contrast to early onset or Type I disease characterizing our Caucasian population.

Association of HLA-DM polymorphism with the production of antiphospholipid antibodies.

OBJECTIVE: To investigate whether variation in the HLA-DM gene is important in producing a group of pathogenic autoantibodies-antiphospholipid antibodies (aPL)-on the basis that HLA class II restricted antigen presentation is involved in the production of aPL. METHODS: HLA-DMA and DMB polymorphisms were genotyped by polymerase chain reaction combined with restriction enzyme digestion in 51 white patients with primary antiphospholipid syndrome (APS), 82 with systemic lupus erythematosus (SLE) (42 with APS and 40 without APS), and 109 healthy white controls. The association with the aPL profile was examined. RESULTS: The distribution of DMA alleles in APS patients and in patients with APS associated with SLE was significantly different from that in controls by 4x2 chi(2) test with 3 degrees of freedom (p = 0.035 and 0.011, respectively), but it was not different between SLE patients without APS and controls. The allelic distribution of DMA was also different between patients with IgG class anticardiolipin antibody or those with lupus anticoagulant (LA) and controls (p = 0.012 and 0.007, respectively) and between patients with and without LA among SLE patients (p = 0.035). All these differences included the increase in DMA*0102 in the former groups. CONCLUSIONS: The results suggest that HLA-DMA*0102 or its linked gene(s) form one of the genetic risks for the production of aPL.

SNP haplotypes and allele frequencies show evidence for disruptive and balancing selection in the human leukocyte receptor complex.

The human leukocyte receptor complex (LRC) of Chromosome 19q13.4 encodes polymorphic and highly homologous genes that are expressed by cells of the immune system and regulate their function. There is an enormous diversity at the LRC, most particularly the variable number of killer cell immunoglobulin-like receptor (KIR) genes. KIR have been associated with several disease processes due to their interaction with polymorphic human leukocyte antigen class I molecules. We have assessed haplotype compositions, linkage disequilibrium patterns and allele frequencies in two Caucasoid population samples (n=54, n=100), using a composite of single-nucleotide polymorphism (SNP) markers and high-resolution, allele-specific molecular genotyping. Particular KIR loci segregated with SNP and other markers, forming two blocks that were separated by a region with a greater history of recombination. The KIR haplotype composition and allele frequency distributions were consistent with KIR having been subject to balancing selection (Watterson's F: P=0.001). In contrast, there was a high inter-population heterogeneity measure for the LRC-encoded leukocyte immunoglobulin-like receptor A3 (LILRA3), indicating pathogen-driven disruptive selection (Wright's FST=0.32). An assessment of seven populations representative of African, Asian and Caucasoid ethnic groups (total n=593) provided little evidence for long-range LRC haplotypes. The different natural selection pressures acting on each locus may have contributed to a lack of linkage disequilibrium between them.

SNP haplotypes and allele frequencies show evidence for disruptive and balancing selection in the human leukocyte receptor complex.

The human leukocyte receptor complex (LRC) of Chromosome 19q13.4 encodes polymorphic and highly homologous genes that are expressed by cells of the immune system and regulate their function. There is an enormous diversity at the LRC, most particularly the variable number of killer cell immunoglobulin-like receptor (KIR) genes. KIR have been associated with several disease processes due to their interaction with polymorphic human leukocyte antigen class I molecules. We have assessed haplotype compositions, linkage disequilibrium patterns and allele frequencies in two Caucasoid population samples (n=54, n=100), using a composite of single-nucleotide polymorphism (SNP) markers and high-resolution, allele-specific molecular genotyping. Particular KIR loci segregated with SNP and other markers, forming two blocks that were separated by a region with a greater history of recombination. The KIR haplotype composition and allele frequency distributions were consistent with KIR having been subject to balancing selection (Watterson's F: P=0.001). In contrast, there was a high inter - population heterogeneity measure for the LRC-encoded leukocyte immunoglobulin-like receptor A3 (LILRA3), indicating pathogen-driven disruptive selection (Wright's FST=0.32). An assessment of seven populations representative of African, Asian and Caucasoid ethnic groups (total n=593) provided little evidence for long-range LRC haplotypes. The different natural selection pressures acting on each locus may have contributed to a lack of linkage disequilibrium between them.

Chemokine gene polymorphisms associate with gender in patients with uveitis.

Uveitis is an inflammatory condition of ocular tissue characterized by leukocyte infiltration, tissue damage, and decreased visual acuity. Chemokines have been implicated in the pathogenesis of uveitis. Polymorphisms in the genes encoding chemokines have been described as affecting chemokine production or function. We analyzed the frequency of single-nucleotide polymorphisms (SNPs) in genes encoding CCL2 (-2518 and -2076) and CCL5 (-403 and -28) in patients with Behçet's disease (BD), a systemic form of uveitis, and patients with retinal vasculitis (RV), an organ-specific form of disease. We report that there was no association between any SNP and disease. However, when segregated on the basis of gender the CCR5 -403 AA genotype was only found in male patients with BD. Similarly, CCL2 genotypes 1/2 were predominant in males, while genotype 4 was significantly associated with disease in female patients with BD. Differences in disease symptoms and severity between males and females have been described in BD and gender-specific genetic differences in chemokine gene function may be involved.

Analysis of Fc gamma receptor II (CD32) polymorphism in populations of African and South Asian ancestry reveals east-west geographic gradients of allele frequencies.

Analysis of FcgammaRIIA alleles in Pakistanis and in Trinidadians of South Asian, African and mixed ancestry revealed no significant differences between Trinidadian South Asians and Pakistanis. H131 homozygotes were more common among Trinidadian South Asians than among Africans and those of mixed ancestry. Comparison with other populations revealed east-west geographic gradients of allele frequencies.

Behçet's disease: from Hippocrates to the third millennium.

Behçet's disease (BD) is characterised by recurrent episodes of orogenital aphthae, systemic vasculitis, and systemic and retinal venous thrombosis. An association between HLA-B51 and BD was first identified over 20 years ago, but recently identified gene associations implicate regions both within and without the MHC in the immunological events underlying the lesions in BD. These include allelic variants within the tumour necrosis factor gene region and within the MHC class I chain related gene region, the factor V Leiden mutation, which is associated with retinal vascular occlusion, and alleles of the intercellular adhesion molecule gene. No single causative gene for BD has emerged; the evidence indicates that the underlying immune events in BD are triggered by a microbial antigen and subsequently driven by genetic influences which control leucocyte behaviour and the coagulation pathways. Knowledge of these risk factors may permit a more accurate prognosis for a given patient, and identify new pathways for more targeted intervention than is currently available.

Analysis of candidate genes on chromosome 19 in coeliac disease: an association study of the KIR and LILR gene clusters.

Coeliac disease is strongly heritable, with more than half of the genetic susceptibility estimated to come from genes outside the HLA region. Several candidate regions have been suggested from genome-wide linkage studies including chromosome 19q13.4 where linkage has been replicated between populations. The natural killer (NK) cell immunoglobulin-like receptors (KIRs) and leukocyte immunoglobulin-like receptor (LILR, also known as ILT and LIR) gene clusters lie within this region in the leukocyte receptor cluster (LRC). KIR molecules are involved in cytotoxic lymphocyte function and expressed by intraepithelial T and NK cells in the duodenum. We studied 132 unrelated UK Caucasian coeliac patients and their parents together with a control group of 171 UK Caucasians. PCR-SSP for KIR2DL1, KIR2DL2, KIR2DL3, KIR2DL5, LILRA3 (ILT6), LILRA3 deletion and an LILRA3 exon 3 single nucleotide polymorphism (SNP) allowed classification of KIR genotypes into five categories and determination of homozygosity or heterozygosity for the common A and B type KIR haplotypes (as defined in the text) and for the LILRA3 deletion. Case-control analysis found no association of the five KIR genotype categories, the A or B KIR haplotypes, the LILRA3 gene deletion or the LILRA3 exon 3 SNP with coeliac disease. A transmission disequilibrium test also found no association of the A and B KIR haplotypes or the LILRA3 gene deletion with coeliac disease.

Natural killer cell immunoglobulin-like receptor (KIR) locus profiles in African and South Asian populations.

Natural killer (NK) and some T cells express killer cell immunoglobulin-like receptors (KIRs), which interact with HLA class I expressed by target cells and consequently regulate cytolytic activity. The number of KIR loci can vary and so a range of genetic profiles is observed. We have determined the KIR genetic profiles from one African (n = 62) and two South Asian (n = 108, n = 78) populations. Several of the KIRs are present at significantly different frequencies between the two major ethnic groups (eg KIR2DS4 gene frequency 0.82 African, 0.47 S Asian. Pc < 1 x 10(-6)) and this is due to uneven distribution of two KIR haplotype families 'A' and 'B'. All three populations described here displayed a greater degree of diversity of KIR genetic profiles than other populations investigated, which indicates further complexity of underlying haplotypes; in this respect we describe two individuals who appear homozygous for a large deletion including the previously ubiquitous 2DL4. We have also reanalysed three populations that we studied previously, for the presence of a KIR which is now known to be an indicator of the 'B' haplotype. South Asians had the highest overall frequencies of all KIR loci characteristic of 'B' haplotypes (Pc < 0.0001 to < 0.004). Furthermore, gene frequency independent deviances in the linkage disequilibrium were apparent between populations.

Quantitative trait loci on chromosomes 1, 2, 3, 4, 8, 9, 11, 12, and 18 control variation in levels of T and B lymphocyte subpopulations.

Lymphocyte subpopulation levels are used for prognosis and monitoring of a variety of human diseases, especially those with an infectious etiology. As a primary step to defining the major gene variation underlying these phenotypes, we conducted the first whole-genome screen for quantitative variation in lymphocyte count, CD4 T cell, CD8 T cell, B cell, and natural killer cell numbers, as well as CD4:CD8 ratio. The screen was performed in 15 of the CEPH families that form the main human genome genetic project mapping resource. Quantitative-trait loci (QTLs) that account for significant proportions of the phenotypic variance of lymphocyte subpopulations were detected on chromosomes 1, 2, 3, 4, 8, 9, 11, 12, and 18. The most significant QTL found was for CD4 levels on chromosome 8 (empirical P=.00005). Two regions of chromosome 4 showed significant linkage to CD4:CD8 ratio (empirical P=.00007 and P=.003). A QTL for the highly correlated measures of CD4 and CD19 levels colocalized at 18q21 (both P=.003). Similarly, a shared region of chromosome 1 was linked to CD8 and CD19 levels (P=.0001 and P=.002, respectively). Several of the identified chromosome regions are likely to harbor polymorphic candidate genes responsible for these important human phenotypes. Their discovery has important implications for understanding the generation of the immune repertoire and understanding immune-system homeostasis. More generally, these data show the power of an integrated human gene-mapping approach for heritable molecular phenotypes, using large pedigrees that have been extensively genotyped.

Human Ro60 (SSA2) genomic organization and sequence alterations, examined in cutaneous lupus erythematosus.

BACKGROUND: The Ro 60 kDa protein (Ro60 or SSA2) is the major component of the Ro ribonucleoprotein (Ro RNP) complex, to which an immune response is a specific feature of several autoimmune diseases. The genomic organization and any sequence variation within the DNA encoding Ro60 are unknown. OBJECTIVES: To characterize the Ro60 gene structure and to assess whether any sequence alterations might be associated with serum anti-Ro antibody in subacute cutaneous lupus erythematosus (SCLE), thus potentially providing new insight into disease pathogenesis. METHODS: The cDNA sequence for Ro60 was obtained from the NCBI database and used for a BLAST search for a clone containing the entire genomic sequence. The intron-exon borders were confirmed by designing intronic primer pairs to flank each exon, which were then used to amplify genomic DNA for automated sequencing from 36 caucasian patients with SCLE (anti-Ro positive) and 49 with discoid LE (DLE, anti-Ro negative), in addition to 36 healthy caucasian controls. RESULTS: Heteroduplex analysis of polymerase chain reaction (PCR) products from patients and controls spanning all Ro60 exons (1-8) revealed a common bandshift in the PCR products spanning exon 7. Sequencing of the corresponding PCR products demonstrated an A > G substitution at nucleotide position 1318-7, within the consensus acceptor splice site of exon 7 (GenBank XM001901). The allele frequencies were major allele A (0.71) and minor allele G (0.29) in 72 control chromosomes, with no significant differences found between SCLE patients, DLE patients and controls. CONCLUSIONS: The genomic organization of the DNA encoding the Ro60 protein is described, including a common polymorphism within the consensus acceptor splice site of exon 7. Our delineation of a strategy for the genomic amplification of Ro60 forms a basis for further examination of the pathological functions of the Ro RNP in autoimmune disease.

The clinical importance of variation within the HLA-Bw4 complex.

BACKGROUND: Antibody screening of a patient with a failed renal transplant showed positive reactions with most, but not all HLA-Bw4-associated B-locus antigens. However, the patient's serological HLA class I type suggested the presence of HLA-Bw4. METHODS: Standard molecular techniques were used to re-type the patient and donor. ELISA antibody screening helped determine the patient's antibody specificity. RESULTS: The patient's type was HLA-B*1402,4703;Bw6 and the donor HLA-B*4703,51011;Bw4,6. Analysis of ELISA results identified three amino acids (positions 77,80,81) as the most likely epitope recognised by the patient's serum. These corresponded to HLA-B*51011 amino acid mismatches, explaining the lymphocytotoxic reactivity pattern. This epitope is located on a subgroup of the HLA-Bw4 antigen suggesting anti-Bw4 was not a sufficient description of this antibody. CONCLUSIONS: This report identifies an antibody to a sub-group of the Bw4 public specificity and also confirms the need for sequence-level analysis in the tissue-typing laboratory to determine future unacceptable mismatches.

Genetic determinism in the relationship between human CD4+ and CD8+ T lymphocyte populations?

The adaptive immune system in mammals acts in a coordinated manner to eliminate environmentally derived pathogens. Humans, mice and rats show within species variation in the levels and ratios of their peripheral CD4+ and CD8+ T cells and to a significant degree this variation is under the control of polymorphic genes. Whether genes act separately to specify CD4+ and CD8+ subpopulation levels or whether CD8+ variation is controlled through gene and environmental action on CD4+ cells or vice versa, is not known. We use a quantitative modelling approach in identical and non-identical female human twins to delineate the lines of control which act upon and between CD4+ and CD8+ subsets. The major findings of the study are: (1) genetic variation controls CD8+ T cell levels through two major routes-the first is via an effect on CD4+ T cells which accounts for the observed co-variation between CD4+ and CD8+ T cells, the second is through direct action on CD8+ T cell levels. (2) No evidence of a gene effect from CD8+ T cells on CD4+ cells is observed. Our findings have implications for the evolution of the complex defence system of which CD4+ and CD8+ T cells are a crucial part and encourage further work towards locating common pleiotropic quantitative trait loci responsible for variation in numbers of T cells.

Tumour necrosis factor alpha promoter polymorphism at position -308 is not associated with actinic prurigo.

Actinic prurigo (AP) has been found to be strongly associated with HLA DR4 and in particular with the DR4 subtype DRB1*0407. However, AP may occur in the absence of HLA-DR4. Furthermore, it has been shown that HLA-DR4 and DRB1*0407, even in association with polymorphic light eruption (PLE), are insufficient for the expression of the AP phenotype. It seems likely, therefore, that other genes in the HLA DR or adjacent regions may contribute to AP susceptibility. One possible predisposing factor in AP may be tumour necrosis factor (TNF)alpha as suggested by the good response of AP to the TNFalpha inhibitor thalidomide, and by the involvement of this cytokine in many immune responses. The aim of this study was to explore the relationship between AP and TNFalpha by examining the frequency of TNF2 in patients with AP, PLE and in normal controls. TNF1 and TNF2 are biallelic polymorphisms at position -308 of the TNFalpha gene promoter and are known to affect transcription of TNFalpha. TNF2 is the rarer of the two alleles and is associated with high functional levels of TNFalpha. This study confirms the positive linkage disequilibrium that has been described between HLA DR3 and TNF2, but fails to show an association between TNF2 and AP.