Genetics and HLA antigens

Genetics and HLA antigens

2 Genetics and HLA antigens CLIFFORD J. E A S T M O N D Psoriasis is a common disease, known to be familial with a multifactorial pattern of inherit...

928KB Sizes 0 Downloads 69 Views

2 Genetics and HLA antigens CLIFFORD

J. E A S T M O N D

Psoriasis is a common disease, known to be familial with a multifactorial pattern of inheritance. Psoriatic arthritis (PsA) is also familial and is well established as a member of the seronegative spondarthritides. Disorders in this group demonstrate familial associations with each other and with radiological sacroiliitis, in both probands and first-degree relatives. Both psoriasis and PsA have been shown to have human leukocyte antigen (HLA) associations. These associations go some way to explaining the familial pattern of these diseases, but are almost certainly not the only predisposing factors. Additional environmental factors are also likely to have a role. GENETICS OF PSORIASIS

The genetic basis of psoriasis is evident from population surveys, twin studies and analysis of individual pedigrees. Familial aggregation amongst blood relatives has been demonstrated in population studies (Hoede, 1957; Lomholt, 1963). It is found three times more commonly in the offspring of psoriatics than in those of non-psoriatics (Hoede, 1931; Steinberg et al, 1951). A very high incidence of psoriasis has been found in children when both parents have the condition (Lomholt, 1963). There is no evidence that inheritance is sex-linked or sex-limited. The age of onset of psoriasis is not fixed and does not fall into a narrow range. Therefore, with prolonged study of the same population over a period of time, familial aggregation becomes even more evident (Hoede, 1957). The pattern of inheritance does not follow single-gene mendelian inheritance patterns. Concordance in inheritance between monozygotic twins is higher (73%) than that in dizygotic twins (20%) (Farber et al, 1974), providing clear evidence for an important role for genetic predisposition. The lack of complete concordance in identical twins suggests an additional role for environmental factors. In the same study the age of onset, distribution, severity and course of the psoriasis tended to be similar in monozygotic but not in dizygotic twins. This suggests that genetic factors are additionally important in determining these aspects of disease expression. Several large individual pedigrees have been published which demonstrate the variable patterns of inheritance (Spindler, 1934; Grayson and Shair, 1959; Ward and Stephens, 1961; Abele et al, 1963). Bailli~re's Clinical Rheumatology--

Vol. 8, No. 2, May1994 ISBN0-7020-1820-1

263 Copyright9 1994,byBailli6reTindall Allrightsofreproductionin anyformreserved



Hence, there is adequate evidence for multifactorial genetic predisposition in the aetiology and pathogenesis of psoriasis, together with environmental influences. GENETICS OF PSORIATIC ARTHRITIS

In addition to individual pedigrees, family surveys have indicated a heritable factor in PsA. The most extensive of these demonstrated a frequency of PsA in 8.3 % of first-degree relatives of patients with PsA; no spouse controls had the condition (Moll and Wright, 1973). Analysis of these data and individual pedigrees does not support a single-gene mendelian type of inheritance. There is no evidence to support sex-linked or sex-limited inheritance. Formal twin studies have not been undertaken and it is therefore not possible to give any estimate of the comparative importance of genetic and environmental factors. In the Leeds study (Moll and Wright, 1973) an interesting triplet was observed consisting of two identical 'twins' and a non-identical third triplet. The identical brothers both had psoriasis. One, however, had a history of peripheral synovitis compatible with PsA, although at the time of study there was no clinical or radiological evidence of arthritis or sacroiliitis. His identical brother had definite ankylosing spondylitis but no peripheral arthritis. The non-identical brother had no psoriasis, arthritis or spinal disease. These observations suggest that genetic predisposition may have less influence on the type of arthritis than dermal manifestations. This raises the possibility that the arthritic manifestations, as opposed to predisposition to them, may be more influenced by environmental factors, while still supporting an important role for genetic predisposition in both psoriasis and its associated arthropathy. H L A ASSOCIATIONS WITH PSORIASIS

Early studies demonstrated an association between HLA-B17 and HLAB13 and psoriasis (White et al, 1972; Svejgaard et al, 1974; Karvonen, 1975; Krulig et al, 1975; Woodrow et al, 1975; Williams et al, 1976; McMichael et al, 1978). The strength of association with these two antigens varied somewhat in different populations. The association with HLA-B17 was particularly strong in female patients with an early age of onset and a positive family history (White et al, 1972; Woodrow et al, 1975). HLA-B13 did not appear to be associated with a positive family history (White et al, 1972). Subsequent studies have demonstrated an association of psoriasis with the haplotypes B57(B17),Cw6,DR7 and B13,Cw6,DR7 (Espinoza et al, 1978; Murray et al, 1980; Armstrong et al, 1983; Beaulieu et al, 1983; McKendry et al, 1984; Woodrow and Ilchsyn, 1985). In an analysis of these associations Woodrow (1985) demonstrated specific association with B13 (relative risk 3.9), B17 (relative risk 6.8), Cw6 (relative risk 9.1) and DR7 (relative risk 2.7). In a ~LXMcomputer package analysis of data from one study (Woodrow



and Ilchsyn, 1985), Green et al (1988) established that the primary association is with HLA-Cw6, and that the other HLA associations with psoriasis are due to linkage disequilibrium between these antigens and HLA-Cw6. This analysis, however, does not establish whether the gene for HLA-Cw6 predisposes for psoriasis or is itself in linkage disequilibrium with the disease susceptibility gene. HLA frequencies differ in various populations. Such racial differences may also affect the HLA frequencies in disease. Most studies on psoriasis have been carried out in Caucasian populations, and the above analyses relate to these. One Japanese study found similar HLA associations to those in Caucasians (Ozawa et al, 1988). A recent study (Nakagawa et al, 1991) found an association with HLA-Cw6 with a relative risk of 8.22, but failed to find any association with HLA-B 13, HLA-B 17 or HLA-DR7. A particularly strong association was found with HLA-A1 (relative risk 9.62). Two Chinese studies also found an association with HLA-A1 (Kao et al, 1987; Yi-Ping et al, 1987). Yi-Ping et al (1987) found a significant increase in HLA-B 17 (relative risk 2.5) and HLA-B 13 (relative risk 3.96), whereas Kao et al (1987) failed to find a statistically significant association, although the relative risk for HLA-B13 was 2.0. Kao et al (1987) did not find any association with HLA-DR7 but did find a strong association with HLA-DR1 (relative risk 26.5). Yi-Ping et al (1987) did not test for HLA-DR7 but did find an increased frequency of HLA-DR1 (relative risk 5.2). Neither study appear to have tested for HLA-Cw6. It is difficult from a small number of studies, in which not all antigens of interest have been reported, to determine satisfactorily whether the associations differ in Chinese and Japanese populations from those in Caucasians, but the results to date suggest that there may be some differences that could warrant further exploration. Although there is an association between psoriasis and HLA-Cw6, there is also a high frequency of Cw-blank. Studies of nucleotide sequences in relation to the HLA-C locus have been undertaken by looking at restriction fragment length polymorphisms by polymerase chain reaction. An early study failed to find any specific sequences associated with psoriasis (Ozawa et al, 1988), but Asahina et al (1991) subsequently found a high frequency of C208A in psoriasis. This sequence was found not only in patients typing HLA-Cw6 but also in those with HLA-Cw7 and HLA-Cw-blank. This probe identifies a substitution for threonine by alanine at position 73 on the aa domain helix of HLA-C molecules. A further study by Sakkas et al (1991) found an increased frequency of Cw~,0601, which was identified by the concordant positivity with three probes identifying individually alanine73, lysine8~and tryptophan97. However, positivity for Cw,~0601 was found only in those patients typing Cw6 and not in Cw6-negative patients. Different probes were used in these two studies and it will be important in future studies to determine whether individual probes are identifying identical nucleotide sequences. Asahina et al (1991) undertook a computer search and found that a 0"2 protein of reovirus contained the same amino-acid residues as the area of interest in the HLA-C gene product (amino-acid residues 70-76). At present



it is not known whether this is of significance in relation to the aetiopathogenesis of psoriasis. I-ILA ASSOCIATIONS AND PSORIATIC ARTHRITIS

The H L A associations in PsA are potentially more complex than those in psoriasis. Psoriatic arthritis has heterogeneity of clinical expression: 1.

2. 3. 4.

Peripheral arthritis. (a) Different patterns of peripheral arthritis exist that may be present in different proportions in different series depending on the methods of ascertainment. (b) Distal interphalangeal (DIP) joint disease is defined in different series as exclusively or predominantly present. Spinal disease typical of ankylosing spondylitis. Spinal disease without sacroiliitis. Combined peripheral arthritis and spinal disease.

Failure to acknowledge such heterogeneity may have been significant in the initial controversy regarding the association of PsA with HLA-B27. In addition, some studies were comparatively small. All studies found an association between HLA-B27 and radiological sacroiliitis, whether or not this was associated with clinical or radiological spinal disease (Brewerton et al, 1974; Karvonen et al, 1974, 1975; Metzger et al, 1975; Sany et al al, 1975; Feldman et al, 1976; Eastmond and Woodrow, 1977a; Roux et al, 1977; Espinoza et al, 1978). Some of these studies, however, did not find an association with peripheral arthritis alone (Metzger et al, 1975; Sany et al, 1975; Feldman et al, 1976; Roux et al, 1977; Espinoza et al, 1978). Some studies failed to differentiate sufficiently between patients with peripheral arthritis and those with spinal disease to allow separate analysis (Zachariae et al, 1974; Marcusson et al, 1975; van Romunde et al, 1984). Combined analysis by Woodrow (1985) has, however, established an association with HLA-B27 and PsA, including peripheral PsA alone, but with the strongest association in those patients with radiological sacroiliitis. Woodrow (1985) has also analysed multiple studies and demonstrated the additional association of PsA with HLA-B 13, B 17, Cw6 and DR7. The association with these H L A antigens is, however, not as strong as that in psoriasis alone. Part of the explanation is the increased frequency of HLA-B27 in arthritic patients, but this is unlikely to be the entire explanation. It is suggested that other unknown non-HLA-linked genetic factors are relatively more important in susceptibility to psoriasis in arthritic patients than in those without arthritis. Some studies have attempted to analyse HLA-B27 associations with subdivisions of PsA. Unfortunately, the subdivisions used have not been sufficiently uniform to allow comparisons between studies or statistical combination of data. Those studies that have identified radiological sacroiliitis as a distinct manifestation of spinal disease have all further confirmed the association of this with HLA-B27 (Eastmond and Woodrow, 1977a;



Armstrong et al, 1983; Beaulieu et al, 1983; Gladman et al, 1986). Studies that have analysed data on psoriatic spinal disease in the absence of radiological sacroiliitis have found no association with HLA-B27 (Eastmond and Woodrow, 1977a; McHugh et al, 1987; Scarpa et al, 1988; Suarez-Almazor and Russell, 1990). In studies where all spinal disease has been combined, irrespective of the presence of radiological sacroiliitis, either no association has been found or any association has been extremely weak (Metzger et al, 1975; Eastmond and Woodrow, 1977a). Several patterns of peripheral arthritis have been identified. Although DIP joint involvement was initially recognized as a distinct clinical grouping, the Leeds group in a recent study have cast doubt on the existence of exclusive DIP joint disease (Helliwell et al, 1991). Some studies have identified a group of patients with DIP joint involvement, however, and analysed H L A frequencies. Gladman et al (1986) identified 17 patients with exclusive DIP joint involvement and found HLA-B27 to be present in 41% compared with only 8% of controls. This high frequency compares with a low frequency of 9% in their oligoarticular group (22 patients) and 15% in their polyarticular group (65 patients). Eastmond and Woodrow (1977a) also analysed their data for DIP joint involvement, although they identified this group as those with predominant rather than exclusive DIP joint involvement. They found that HLA-B27 was present in 30% of 26 patients compared with a frequency of 8 % in controls. In both these studies, patients with radiological sacroiliitis were excluded from the analysis. Lopez-Larrea et al (1990) subdivided their 104 patients into those with oligoarticular or polyarticular patterns of peripheral arthritis and found Cw6 to be associated with oligoarthritis. Only five patients had DIP joint disease, and these were not analysed separately. Patients with spinal disease had an increased frequency of HLA-B27 but not of HLA-Cw6. In a subsequent enlarged series of 180 patients (Torre Alonso et al, 1991), these associations were confirmed with the additional observation that HLA-B27 was present in only 22% of patients with unilateral radiological sacroiliitis compared with 85% of those with bilateral changes. In those studies that have separately analysed data for patients with spinal disease characterized by radiological sacroiliitis with or without syndesmophyte formation and no peripheral arthritis, the frequency of HLA-B27 has been found to be similar to that in ankylosing spondylitis in the absence of psoriasis (Brewerton et al, 1974; Eastmond and Woodrow, 1977a; Gladman et al, 1986; Suarez-Almazor and Russell, 1990). Whether there are any associations between other HLA-B antigens and PsA is open to doubt. It has been suggested that there are associations with the subdivisions (splits) of HLA-B16 (B38 and B39). Unfortunately, not all series have given complete data, resulting in difficulties with analysis. One study found an increased frequency of both HLA-B38 and HLA-B39 (Arnett and Bias, 1980). HLA-B38 was found to be associated with the combination of spinal and peripheral arthritis but not with peripheral arthritis alone. No analysis of spinal disease alone was undertaken, nor were patients with spinal disease subdivided into those with and without sacroiliitis. Only 4 of 52 arthritis patients studied were HLA-B39 positive



compared with no controls. These four patients were equally divided into the subgroups of peripheral arthritis either with or without spinal disease. Others have found an increased frequency only of HLA-B38 (Roux et al, 1977; Espinoza et al, 1978; Murray et al, 1980). There are, however, differences in the pattern of arthritis reported to be associated with this antigen. Roux et al (1977) found HLA-B38 to be associated with sacroiliitis with or without spondylitis or peripheral arthritis but not with peripheral arthritis alone, whereas Espinoza et al (1978) found the association of HLA-B38 to be with peripheral arthritis but not with spinal disease. Murray et al (1980) did not subdivide their arthritic patients into clinical subgroups. In addition, 15 of their 52 arthritic patients were Jewish, and in these the association of PsA with HLA-B38 was strongest. The ethnic composition of their control population, however, is unknown, leaving the significance of this observation undetermined. One study (Beaulieu et al, 1983) found an increased frequency only of HLA-B39. No analysis of clinical subdivisions of the arthritis was undertaken. It has been concluded that there is considerable heterogeneity between these studies, with insufficient evidence to support or completely refute an association with the HLA-B16 splits (Woodrow, 1985). Crivellato and Zacchi have more recently re-examined the association of HLA-B16 and its splits. In their first study (1986), in which they did not study HLA-B16 splits, they found an overall association of B16 with psoriatic arthritis in 40 patients, but analysis of subgroups indicated that this association was extremely strong in the ten patients with radiological sacroiliitis and spondylitis (relative risk 67.2). There was, however, no increased frequency in patients with radiological sacroiliitis alone. There was an increase (not statistically significant) of HLA-B16 in 24 patients with psoriatic arthritis without spinal disease (relative risk 9.6). Their subsequent larger study (Crivellato and Zacchi, 1987), which looked at the splits HLAB38 and HLA-B39, revealed no significant increase of HLA-B38 in any of these three subgroups. There was, however, a significant increase of HLAB39 in patients with spinal involvement, evidenced by radiological sacroiliitis, spondylitis or both, irrespective of the presence of peripheral arthritis (29% of 17 patients with spinal disease). In the total group of 50 patients with PsA, there was an overall increased frequency of B39 (10%) and all five B39 subjects had spinal disease. McHugh et al (1987) found a weak, but not statistically significant, association of HLA-B38 (but not of HLA-B39) with spinal disease and not with peripheral arthritis. Gladman et al (1986) found an increased frequency of both HLA-B38 and HLA-B39 in peripheral psoriatic polyarthritis but in no other patterns of peripheral arthritis or spinal disease alone. The issue of any association with HLA-B16 splits therefore remains unresolved. It is possible that there may be different associations depending on disease expression and also in relation to the ethnic origin of the population studied. Some series have analysed patterns of peripheral arthritis in relation to H L A - D R typing and found associations between peripheral arthritis and HLA-DR4 (Gerber et al, 1982; Gladman et al, 1986) and others with



HLA-DR7 (Armstrong et al, 1983). There has been particular interest in the possible association of HLA-DR4 with the pattern of peripheral PsA resembling seronegative rheumatoid arthritis in its joint distribution and symmetry. Gerber et al (1982) found HLA-DR4 in 86% of 12 patients with this pattern compared with 36% of normal controls, but no association in patients with other patterns of peripheral arthritis. Gladman et al (1986) found that 61% of 12 patients were HLA-DR4 positive compared with 30% of controls. However, two patients in the latter series had immunoglobulin M (IgM) rheumatoid factor and both were HLA-DR4 positive. Salvarani et al (1989) failed to find any association with this or any other pattern of PsA and HLA-DR4, although they acknowledge a low control population frequency of HLA-DR4 and a relatively weak association of HLA-DR4 with seropositive rheumatoid arthritis in their region of Italy compared with other series of similar patients. In addition, no association was found between HLA-DR7 and psoriatic arthritis of any clinical subtype. Sakkas et al (1990) confirmed the association of psoriasis with HLA-DR7. They also found an association of HLA-DR7 with PsA, but less strong than that between this antigen and psoriasis. In addition, using Southern blot analysis of genomic D N A they found a stronger association of the subtype HLA-DR7a with PsA than with psoriasis. Variation in the frequency of this subtype may account for some of the differences between reported series of PsA. HLA ASSOCIATION AND PALMOPLANTAR PUSTULOSIS (SAPHO)

An association between HLA-B27 and palmoplantar pustulosis has been noted. The association is especially present in patients with anterior chest wall osteitis and sacroiliitis with or without spondylitis (Szanto and Linse, 1991; Khan and Chaot, 1992). No increased frequency of those H L A antigens associated with psoriasis, however, has been found. This suggests that palmopustulosis is an immunogenetically distinct disorder from psoriasis but may have some links with PsA and other seronegative spondarthritides. HLA ASSOCIATION WITH THE DERMAL AND RHEUMATIC

MANIFESTATIONS OF HIV INFECTION Reiter's syndrome associated with human immunodeficiency virus (HIV) infection and acquired immune deficiency disease (AIDS) is strongly associated with HLA-B27 (Brancato et al, 1989). However, although there is a high frequency of psoriasiform skin lesions in HIV-infected individuals, this manifestation is not associated with HLA-Cw6 or other H L A antigens normally associated with psoriasis. This suggests a different aetiopathogenesis for these skin lesions than in psoriasis vulgaris, although the arthropathy associated with these skin lesions and other manifestations of Reiter's syndrome is likely to occur as a result of similar mechanisms as in



the absence of HIV infection. It also suggests that CD4 helper T cells are not important for the development of this arthropathy in HIV infection and equally may have no role in PsA, whereas CD8 suppressor cytotoxic T cells may prove to have a significant role in the development of these arthropathies. DISCUSSION There is clear evidence of an important role for genetic factors in both psoriasis and PsA. The pattern of inheritance, however, is not that of a single mendelian gene. The association of both these diseases with histocompatibility antigens provides evidence for at least part of the genetic susceptibility, although in both psoriasis and PsA it is probable that genes outwith the major histocompatibility complex (MHC) are also operative. In addition, there is evidence from family and other studies of a role for environmental factors. It remains to be determined how such factors may inter-relate with genes both within and outwith the MHC. Psoriasis

Present evidence suggests that the primary association in psoriasis is with HLA-Cw6, although there is also a high frequency of HLA-Cw-blank. Further study of individual nucleotide sequences of the HLA-C locus may lead to recognition of a single sequence specifically associated with psoriasis. It seems likely that this nucleotide sequence will be in the c~1domain helix of the HLA-C locus. Psoriatic arthritis

There is now conclusive evidence of an association between HLA-B27 and peripheral PsA, irrespective of the presence of radiological sacroiliitis. The association, however, is highest in the presence of the combination of psoriasis, sacroiliitis and spondylitis but without peripheral arthritis. Indeed the strength of association is identical to that in ankylosing spondylitis in the absence of psoriasis. It may therefore be that, whichever of the mechanisms currently under consideration to explain the association of HLA-B27 and ankylosing spondylitis is finally determined to be relevant, it is likely also to be relevant in this particular form of psoriatic arthritis. Analysis of the various structural subtypes of HLA-B27 in ankylosing spondylitis has failed to yield any results to suggest that a particular subtype is associated with ankylosing spondylitis (McLean, 1992). No similar studies have been undertaken in PsA or psoriatic spondylitis. There does not appear to be any association between HLA-B27 and psoriatic spondylitis in the absence of radiological sacroiliitis. Although there is clearly an association between HLA-B27 and peripheral PsA, the weaker association of this pattern of PsA in contrast to radiological sacroiliitis suggests a potential role for: (1) other genes within the MHC; (2) other genes outwith the MHC, and (3) environmental factors.



At the present time it is not known whether there is any particular pattern of peripheral PsA with a stronger or weaker association with HLA-B27, although some results do suggest that DIP joint disease may be more strongly associated with HLA-B27 than with the other forms of peripheral arthritis. It is important to note that this particular pattern of peripheral arthritis also has a stronger association with sacroiliitis. This clinical association, however, cannot be the explanation of the observations of HLA-B27 and DIP joint disease since studies specifically excluding those patients with sacroiliitis have also found a tendency towards a higher frequency of HLA-B27. It may be, however, that some of these patients will develop sacroiliitis in the future; clinical disease manifestations are not static. Although PsA is associated with HLA-Cw6, the association is weaker than that between HLA-Cw6 and psoriasis alone. It is not possible with present knowledge to determine whether or not HLA-Cw6 itself, a specific nucleotide sequence of the HLA-Cw locus or a gene in strong linkage disequilibrium with the HLA-C locus predisposes to psoriasis. Further, it is not possible to know whether the same gene predisposing to psoriasis located in the MHC will itself have any predisposition to the development of arthritis. The combination of peripheral arthritis and radiological sacroiliitis appears to hold an intermediate position in relation to the frequency of HLA-B27 between that of peripheral PsA alone and psoriatic spondylitis with radiological sacroiliitis alone (Eastmond and Woodrow, 1977a; SuarezAlmazor and Russell, 1990). Patients with HLA-B27-negative ankylosing spondylitis do not have an increased frequency of HLA-Cw6 or those HLA antigens in linked disequilibrium with HLA-Cw6. They do, however, have evidence of an increased family history of psoriasis and chronic inflammatory bowel disease (Woodrow and Eastmond, 1978). This suggests that there are genes outwith the MHC that predispose to HLA-B27-negative ankylosing spondylitis, psoriasis and chronic inflammatory bowel disease. The nature and location of these genes is as yet unknown. There is some evidence for an association between PsA and HLA-B16 and its splits, HLA-B38 and HLA-B39. The available studies for analysis, however, do not allow this to be determined with complete certainty and, in particular, there is conflicting evidence in relation to the individual pattern of clinical disease that may be associated with these splits. They may, however, represent additional genes to HLA-B27 within the MHC having a role in predisposing to PsA. Similarly, there is a hint of a possible association between HLA-DR4 and the symmetrical form of peripheral PsA. In view of the known association between HLA-DR4 and rheumatoid arthritis, it is of some importance to determine whether this pattern is truly associated with HLA-DR4, although existing studies are insufficientto provide conclusive evidence. If there is such an association, this could further suggest that individual HLA genes may have an influence on the variable disease pattern of PsA. On the other hand, it may be that this pattern of peripheral arthropathy is simply a coincidence of two common diseases, rheumatoid arthritis and psoriasis. Alternatively, in



addition to such a coincidence of two common diseases, it may be that some genes for psoriasis inhibit the formation of IgM rheumatoid factor or add a further predisposition to the development of arthritis in individuals already carrying genetic susceptibility to the development of rheumatoid arthritis through HLA-DR4. Twin studies in psoriasis suggest a role for environmental factors. Such studies have not been undertaken in PsA. Twin studies in ankylosing spondylitis suggest a role for environmental factors in both disease susceptibility and expression (Eastmond and Woodrow, 1977b). Those twins with PsA who have been observed suggest that there may be a role for environmental factors in the individual articular manifestations. Exacerbations of PsA have been associated with infections known to trigger reactive arthritis (Eastmond, 1983; Trull et al, 1986). Many patients with PsA have a course of relapse and remission. The disease may appear entirely quiescent for considerable periods. It would therefore be surprising if environmental factors did not have a role in predisposing to PsA, although as yet completely convincing evidence of their importance is lacking and there remains considerable speculation with regard to the possible mechanisms of the interrelationship between environmental factors and H L A antigens in ankylosing spondylitis. The pattern of disease associated with HIV infection is of interest in relation to possible interactions between HLA antigens and environmental factors. The failure to find an association between the psoriasiform lesions found in HIV infection and genes within the MHC suggests either that these psoriasiform manifestations are different from psoriasis or that this particular manifestation is less dependent than psoriasis vulgaris on genes within the MHC. The finding that the arthropathy similar to reactive arthritis and PsA in HIV infection is associated with HLA-B27 suggests that these articular manifestations at least are likely to have the same genetic basis within the MHC as similar articular diseases not associated with HIV infection. This also suggests that CD4 helper T cells are unlikely to have a significant role in the aetiopathogenesis of these diseases either in HIV-infected persons or in the absence of such infection, and that CD8 suppressor cytotoxic T cells may prove to have a more significant role in the development of these arthropathies. It may be necessary to await further clarification of the nature of t h e association between HLA-Cw6 and psoriasis and between HLA-B27 and ankylosing spondylitis before the mechanisms and role of the MHC associations with PsA can be studied, in view of the considerably more complex nature of these associations. There may, however, be justification in the study of comparatively large and well-documented series of patients with PsA in as genetically homogeneous a population as possible to determine the precise associations between the HLA antigens of interest and different patterns of articular and dermal disease since the data currently available provide the tantalizing prospect that the individual clinical manifestations of disease may be determined by a complex interplay between individual genes within and outwith the MHC and variable environmental factors. It could also be of interest in considering such a



complex interplay to try to collect a sufficiently large number of twins for comparison of the frequency of disease in monozygotic and dizygotic twins, as well as for study of the differences in disease expression. A multinational collaborative effort would be required to find sufficient twins to provide adequate data. We can only look forward to the possibility of a fascinating unravelling of such a complex interplay of genetic and environmental factors as more is learnt of the mechanisms by which genes within the M H C predispose to these diseases. SUMMARY

There is convincing evidence of a genetic basis for both psoriasis and psoriatic arthritis (PsA). Part of this genetic predisposition is due to genes within the major histocompatibility complex (MHC). In psoriasis, the primary association is with HLA-Cw6. Further work on specific nucleotide frequencies, especially those in the oq domain helix of the HLA-C molecule, will be of interest in determining whether a specific nucleotide frequency is present in all patients. The situation in PsA is considerably more complex. It is now established that there is an association between HLA-B27 and PsA, both in its peripheral arthropathy and in spinal disease in which radiological sacroiliitis is present. Spinal disease without radiological sacroiliitis is probably not associated with HLA-B27. There is some suggestion that HLA-B16 or its splits, HLA-B38 and HLA-B39, may also be associated with PsA, but there is considerable heterogeneity between the series, which prevents a firm conclusion being made. It is possible, but again not conclusive, that there is an association between H L A - D R 4 and the symmetrical seronegative pattern of peripheral PsA. It is also likely that genes outwith the M H C predispose to psoriasis and PsA. It is further likely that a role will be found for environmental factors in both psoriasis and PsA. There is a tantalizing possibility of a complex interplay between a variety of environmental factors and genetic factors, both within and outwith the MHC, determining not only susceptibility but also the individual clinical pattern of disease. Further clarification of these possibilities is likely to depend primarily on understanding the role of genes within the MHC in predisposing to comparatively more homogeneous diseases, such as psoriasis and ankylosing spondylitis, before the mechanisms operating in PsA can be analysed and better understood. REFERENCES

Abele DC, Dobson RL & GrahamJB (1963) Heredityand psoriasis. A studyof a large family. Archives of Dermatology 88: 38-47. Armstrong RD, Panayi GS & Welsh KI (1983) Histocompatibilityantigens in psoriasis, psoriatic arthropathy and ankylosingspondylitis.Annals of the Rheumatic Diseases 42: 142-146.




Arnett FC & Bias WB (1980) HLA-Bw38 and Bw39 in psoriatic arthritis: relationship and implications for peripheral and axial involvement. Arthritis and Rheumatism 23:649-650 (abstract). Asahina A, Akazaki S, Nakagawa H et al (1991) Specific nucleotide sequence of HLA-C is strongly associated with psoriasis vulgaris. Journal of Investigative Dermatology 97: 254258 (Erratum Journal of Investigative Dermatology 97: 762). Beaulieu AD, Roy R, Mathon G e t al (1983) Psoriatic arthritis: risk factors for patients with psoriasis--a study based on histocompatibility antigen frequencies. Journal of Rheumatology 10: 633-636. Brancato L, Itescu S, Skovron ML, Solomon G & Winchester R (1989) Aspects of the spectrum, prevalence and disease susceptibility determinants of Reiter's syndrome and related disorders associated with HIV infection. Rheumatology International 9: 13%141. Brewerton DA, Nicholls A, Caffrey M, Waiters D & James DCO (1974) HL-A27 and arthropathies associated with ulcerative colitis and psoriasis. Lancet i: 956-958. Crivellato E & Zacchi T (1986) Association between HLA-B16 and psoriatic spondylitis. Acta Dermatologica and Venereologica (Stockholm) 66: 262-264. Crivellato E & Zacchi T (1987) HLA-B39 and the axial type of psoriatic arthritis. Acta Dermatologica and Venereologica (Stockholm) 67: 249-250. Eastmond CJ (1983) Gram-negative bacteria and B27 disease. BritishJournal of Rheumatology 22 (supplement 2): 67-71. Eastmond CJ & Woodrow JC (1977a) The HLA system and the arthropathies associated with psoriasis. Annals of the Rheumatic Diseases 36: 112-120. Eastmond CJ & Woodrow JC (1977b) Discordance for ankylosing spondylitis in monozygotic twins. Annals of the Rheumatic Disease 36: 360-364. Espinoza LR, Vasey FB, Oh JH, Wilkinson R & Osterland CK (1978) Association between HLA-Bw38 and peripheral psoriatic arthritis. Arthritis and Rheumatism 21: 72-75. Farber EM, Nall ML & Watson W (1974) Natural history of psoriasis in 61 twin pairs. Archives of Dermatology 109: 207-211. Feldman JL, Amor B, Cazalis P, Dryll A, Hors JO & Haquart B (1976) Antig~nes HLA chez les malades atteints de rhumatisme psoriasique. La Nouvelle Presse M~dicale 5: 477-480. Gerber LH, Murray CL, Perlman SG et al (1982) Human lymphocyte antigens characterising psoriatic arthritis and its subtypes. Journal of Rheumatology 9: 703-707. Gladman DD, Anhorn KA, Schachter RK & Mervart H (1986) HLA antigens in psoriatic arthritis. Journal of Rheumatology 13: 586-592. Grayson LD & Shair HM (1959) Psoriatic family tree. Archives of Dermatology 79: 661-664. Green J, Montasser M, Low HC & Woodrow JC (1988) Investigation of the association of a number of HLA antigens with psoriasis and psoriatic arthritis. Statistics in Medicine 7: 443-450. Helliwell P, Marchesoni A, Peters M, Barker M & Wright V (1991) A re-evaluation of the osteoarticular manifestations of psoriasis. British Journal of Rheumatology 30: 339-345. Hoede K (1931) Umwelt und erblichkeit bei der entstehung der Schuppenflechte. Wfirzburg, Habilitationsschrift. Hoede K (1957) Zur frage der erblichkeit der psoriasis. Hautarzt 8: 433-438. Kao J-S, Yu H-S, Shey H-M & C h e n G-S (1987) HLA-A, B, C, and DR in Chinese with psoriasis: relation to age of onset and clinical expression of the disease. Journal of the Formosan Medical Association 86: 259-263. Karvonen J (1975) HL-A antigens in psoriasis with special reference to the clinical type of onset, exacerbations after respiratory infection and occurrence of arthritis. Annals of Clinical Research 7: 301-311. Karvonen J, Lassus A, Sievers U & Tiilikainen A (1974) HL-A antigens in psoriatic arthritis. Annals of Clinical Research 6: 304-307. Karvonen J, Tiilikainen A & Lassus A (1975) HL-A antigens in patients with persistent palmoplantar pustulosis and pustular psoriasis. Annals of Clinical Research 7: 112-115. Khan MF & Chaot AM (1992) SAPHO syndrome. Rheumatic Diseases Clinics of North America 18: 225-246. Krulig L, Farber EM, Grumet FC & Payne RO (1975) Histocompatibility (HL-A) antigens in psoriasis. Archives of Dermatology 111: 857-860. Lomholt G (1963) Psoriasis, Prevalence, Spontaneous Course and Genetics. Copenhagen: GEC Gad.



Lopez-Larrea C, Torre Alonso JC, Roriguez Perez A & Coto E (1990) HLA antigens in psoriatic arthritis subtypes of a Spanish population. Annals of the Rheumatic Diseases 49: 318-319. McHugh NJ, Laurent MR, Treadwell BL, Tweed JM & Dagger J (1987) Psoriatic arthritis: clinical subgroups and histocompatibility antigens. Annals of the Rheumatic Diseases 46: 184--188. McKendry RJR, Senger DPS, DesGroseilliers JP & Dunne JV (1984) Frequency of HLA antigens in patients with psoriasis or psoriatic arthritis. Canadian Medical Association Journal 130: 411-415. McLean L (1992) HLA-B27 subtypes: implications for the spondyloarthropathies. Annals of the Rheumatic Diseases 51: 929-931. McMichael A J, Morhenn V, Payne R, Sasazuki T & Farber EM (1978) HLA C and D antigens associated with psoriasis. British Journal of Dermatology 98: 287-292. Marcusson J, M611or E & Thyresson N (1975) HL-A antigens (17,27,UPS) in psoriasis with special reference to patients with arthritic lesions. Acta Dermato-Venereologica (Stockholm) 55: 297-300. Metzger AL, Morris RI, Bluestone R & Terasaki PI (1975) HL-A W27 in psoriatic arthropathy. Arthritis and Rheumatism 18: 111-115. Moll JMH & Wright V (1973) Familial occurrence of psoriatic arthritis. Annals of the Rheumatic Diseases 32: 181-201. Murray C, Mann DL, Gerber N e t al (1980) Histocompatibility alloantigens in psoriasis and psoriatic arthritis. Journal of Clinical Investigation 66: 670-675. Nakagawa H, Akazaki S, Asahina A et al (1991) Study of HLA class I, class II and complement genes (C2, C4A, C4B and BF) in Japanese psoriatics and analysis of a newly-found haplotype by pulsed field gel electrophoresis. Archives of Dermatological Research 283: 281-284. Ozawa A, Ohkido M, Inoko H, Ando A & Tsuji K (1988) Specific restriction fragment length polymorphism on the HLA-C region and susceptibility to psoriasis vulgaris. Journal of Investigative Dermatology 90: 402-405. Roux H, Mercier P, Maestracci D et al (1977) Psoriatic arthritis and HLA antigens. Journal of Rheumatology 4 (supplement 3): 64-65. Sakkas LI, Loqueman N, Bird H, Vaughan RW, Welsh KI & Panayi GS (1990) HLA class II and T cell receptor gene polymorphisms in psoriatic arthritis. Journal of Rheumatology 17: 1487-1490. Sakkas LI, Vaughan RW, Panayi GS & Welsh KI (1991) Application of polymerase chain reaction to HLA-C locus typing. European Journal oflmmunogenetics 18: 185-187. Salvarani C,~Macchioni PL, Zizzi F et al (1989) Clinical subgroups and HLA antigens in Italian patients with psoriatic arthritis. Clinicaland Experimental Rheumatology 7: 391-396. Sany J, Seibnalet J, Guilhou J-J & Serre H (1975) HL-A et rhumatisme psoriasique. Revue du Rhumatisme 42: 451-460. Scarpa R, Oriente P, Pucino A et al (1988) The clinical spectrum of psoriatic spondylitis. British Journal of Rheumatology 27: 133-137. Spindler A (1934) Zur vererbung der psorisis vulgaris. Archives of Dermatology and Syphilis 169: 417-420. Steinberg AG, Becker SW, Fitzpatrick TB & Kierland RR (1951) A genetic and statistical study of psoriasis. American Journal of Human Genetics 3: 267-281. Suarez-Almazor ME & Russell AS (1990) Sacroiliitis in psoriasis: relationship to peripheral arthritis and HLA B27. Journal of Rheumatology 17: 804-808. Svejgaard A, Staub-Nielsen L, Svejgaard E, Kissmeyer-Nielsen F, HjortshOj A & Zachariae H (1974) HLA in psoriasis vulgaris and in pustular psoriasis--population and family studies. British Journal of Dermatology 91: 145-153. Szanto E & Linse U (1991) Arthropathy associated with palmoplantar pustulosis. Clinical Rheumatology 10: 130-135. Torre Alonso JC, Rodriguez Perez A, Arribas Castrillo JM & Ballina Garc/a J (1991) Psoriatic arthritis (PA): a clinical, immunological and radiological study of 180 patients. British Journal of Rheumatology 30: 245-250. Trull AK, Eastmond C J, Panayi GS & Reid TM (1986) Salmonella reactive arthritis: serum and secretory antibodies in eight patients identified after a large outbreak. British Journal of Rheumatology 25: 13-19.



van Romunde LKJ, Cats A, Hermans J, Valkenburg HA & de Vries E (1984) Psoriasis and arthritis. III: A cross-sectional comparative study of patients with 'psoriatic arthritis' and seronegative and seropositive polyarthritis: radiological and HLA aspects. Rheumatology International 4: 67-73. Ward JH & Stephens FE (1961) Inheritance of psoriasis in a Utah kindred. Archives of Dermatology and Syphilis 84: 589-592. White SH, Newcomer VD, Mickey MR & Terasaki PI (1972) Disturbance of HL-A antigen frequency in psoriasis. New England Journal of Medicine 287: 740-743. Williams RC, McKenzie AW, Roger JH & Joysey VC (1976) HLA antigens in patients with guttate psoriasis. British Journal of Dermatology 95: 163-167. Woodrow JC (1985) Genetic aspects of the spondyloarthropathies. Clinics in the Rheumatic Diseases 11" 1-24. Woodrow JC & Eastmond CJ (1978) HLA-B27 and the genetics of ankylosing spondylitis. Annals of the Rheumatic Diseases 37" 504-509. Woodrow JC & Ilchsyn A (1985) HLA antigens in psoriasis and psoriatic arthritis. Journal of Medical Genetics 22: 492-495. Woodrow JC, Dave VK, Usher N & Anderson J (1975) The HL-A system and psoriasis. British Journal of Dermatology 92: 427-436. Yi-Ping S, Jia-bin A, Shi-tai Let al (1987) HLA antigen distribution and haplotype segregation in Chinese patients with psoriasis. Chinese Medical Journal 100: 236-241. Zachariae H, HjortshOj A, Kissmeyer-Nielsen F, Svejgaard A, Svejgaard E & Zachariae E (1974) HL-A antigens and psoriatic arthritis. Acta Dermato-Venereologica (Stockholm) 54: 443-447.