Genetics of B27-associated diseases--1.

The genetic analysis of those conditions that show familial aggregation, in the absence of recognisable patterns of Mendelian segregation, has proved to be very difficult. A generalised model would allow for the possibility of more than one disease susceptibility gene and for the influence of environmental factors. The pedigree illustrated in Fig. 1 includes patients with Reiter's syndrome (RS), ankylosing spondylitis (AS) and uveitis, ulcerative colitis, and psoriasis. It can be seen that the associations with the HLA-B locus genes enable us to achieve an understanding of some genetic structure in pedigrees of this kind. With regard to the arthropathies being considered at this meeting, clearly the HLA-linked 'ankylosing spondylitis gene' (which may or may not be B27 itself) and genes that play a role in susceptibility to psoriasis and inflammatory bowel disease have to be considered. If we look first at AS itself I think we must look at two phenomena. Firstly, there are the genetic and environmental factors that determine w-hether an individual gets or does not get the disease. Secondly, there is the question of the severity. Experience in recent years increasingly suggests that we are dealing here with a very graded phenotype from mild asymptomatic radiologically detectable sacroiliitis to the full picture of severe spinal disease. This evidence comes from the study of members of the families of patients with spondylitis and from studies of populations, particularly those made up of B27-positive individuals. It seems to me that clinically there is a much greater difference between a B27-positive individual with asymptomatic radiological sacroiliitis and one with very severe sacroiliitis than between the former and a normal B27-negative individual. I think it is going to be just as important to ask why some B27-positive individuals get very severe disease while in others the only evidence of disease is a slight radiological change in the sacroiliac joints. The source of variation in relation to both these questions must be either genetic or environmental or both. If we accept for the moment that the 'spondylitis gene' is B27 itself then clearly B27 has an essential role in determining whether any degree of sacroiliitis or spondylitis occurs. We can produce a general model (Fig. 2) in which the much increased relative risk associated with B27-something of the order of 100 to 150-is evident but which allows for further variation in liability within B27-negative and B27-positive individuals. This underlying variability could be environmental or genetic. Evidence of important environmental factors comes from a study of identical twins in which it can be