There are two successful approaches to identifying mutations that contribute to multifactorial disorders. One is to recognise polymorphisms that are closely associated with the disease or with one aspect of it and to search for closely linked mutations in populations or families. Those so far found in association with common diseases generally have a frequency greater than 1% in the general population and therefore are more appropriately called polymorphisms. The second method is to conduct breeding experiments using mouse models.
Investigations of Type I diabetes mellitus and of Parkinson's disease exemplify the first type of approach. In the former condition there is a long known association of disease with certain HLA haplotypes. More information about the precise influential alleles was found by studying patients of different races (``Caucasians'' from the USA and ``blacks'' from Britain) whose HLA haplotypes showed different patterns (Todd et al 1987, 1989). In both groups, specific DQI alleles contributed to susceptibility or resistance to Type I diabetes mellitus. Regarding Parkinson's disease, two groups of workers have recently found that mutations in the CYP206 gene are unduly common in patients (Armstrong et al 1992, Smith et al 1992). What is this gene and why did the authors study it? The gene codes for one of the P450 cytochromes which is responsible for the oxidation of debrisoquine and for about twenty other drugs (Kagimoto et al 199). Five to ten percent of Europeans are poor oxidisers of these drugs and a previous example of the harmful effects of poor oxidation was when neurotoxicity to the drug perhexiline was found almost exclusively in those members of the population who were poor oxidisers (Shah et al 1982). In 1985 Barbeau et al observed that poor oxidisers of debrisoquine were unduly common among patients who developed Parkinson's disease early in life. The possible relevance of this has emerged since. In the early 1980's it was observed that the meperidine derivative, MPTP, caused irreversible Parkinson's disease in drug addicts using a synthetic heroin (Langston et al 1983, Tertrud & Langston 1989). Experiments in rats have since shown that MPTP is metabolised by the same cytochrome that is responsible for the oxidation of debrisoquine (Fonne-Prister et al 1987). If MPTP analogues are formed from endogenous methylation (Testa et al 1985) it is possible that these have a chronic toxic effect on the basal ganglia, particularly in those patients who are poor oxidisers. Therefore the finding of an excess of mutations of the CYP206 gene in patients with Parkinson's disease is particularly interesting, and leads the way to identifying other components of an environmental/genetic interaction. Parkinsonian patients also have poor sulphur metabolism (Steventon et al 1989) and there are different patterns of impaired detoxification systems in Alzheimer's disease and motor neurone disease, so similar mechanisms may be operating in all these neurodegenerative disorders.
The second approach, that of performing breeding experiments using mouse models has been particularly successful with understanding the genetic background for Type I insulin-dependent diabetes mellitus (Prochazka et al 1987, Todd et al 1991, Cornall et al 1991). The NOD (non-obese diabetic) mouse is a model for Type I diabetes as it spontaneously develops insulin dependent diabetes, together with an inflammation of the pancreatic islet cells and the presence of autoantibodies to these cells. Diabetes in the NOD mouse depends upon H2 linked genes which are compatible to HLA genes in humans, but this is not the whole story. For, those congenic progeny of NOD/resistant strains of mice who receive the HLA region from the NOD mouse but the remainder of their genotype from the resistant mice, do not develop diabetes. Series of breeding experiments between these congenic NOD mice and strains of resistant mice, together with the analysis of a large number of progeny using variable locus-specific micro satellites have shown that loci on mouse chromosomes 1, 3, 9 and 11 influence the development of diabetes. Now that a large and widespread number of DNA markers are available for the mouse genome, and now that it is realised that there is considerable homology between mouse and human genes, these sorts of breeding experiments could help in identifying susceptibility genes for other diseases, provided that a suitable mouse model is available. There is however no mouse model for Parkinson's disease, which can only be reproduced in primates and then only after the administration of MPTP.
Efforts continue to try to identify environmental triggers for these multifactorial diseases, so that they may be controlled in individuals who are at increased risk, namely twins, sibs and offspring of patients. The environmental causes of Type I diabetes mellitus have been elusive, but the longstanding suggestion of viruses (because twins and sibs who are going to become diabetic do so soon after the index patient), and of Coxsackie B virus in particular, has received a recent boost from the observation that glutamate decarboxlyase is highly homologous with the P2-C protein of Coxsackie virus B4 (Clare-Salzler et al 1992). Antibodies to glutamate decarboxylase are an early and consistent finding in Type I diabetes mellitus and are presumed to play an important part in initiating this autoimmune disease.
Sarah Bundey
References
Armstrong M, Daly A K, Cholerton S, Bateman D N, Idle J R (1992) Mutant debrisoquine hydroxylation genes in Parkinson's disease. Lancet 339: 1017-1018
Barbeau A, Cloutier T, Roy M, Plasse L, Paris S, Poitier J 1985 Ecogenetics of Parkinson's disease; 4-hydroxylation of debrisoquine. Lancet 2: 1213-1216
Clare-Salzler M J, Tobin A J, Kaufman D L 1992 Glutamate decarboxylase: an autoantigen in IDDM. Diabetes Care 15: 132-135
Cornall R J, Prins J-B, Todd J A et al 1991 Type I diabetes in mice is linked to the interleukin-1 receptor and Lsh/Ity/Bcg genes on chromosome 1. Nature 353: 262-265
Fonne-Pfister R, Bargetz M J, Meyer U A 1987 MPTP, the neurotoxin inducing Parkinson's disease, is a potent competitive inhibitor of human and rat cytochrome P450 isozymes catalysing debrisoquine 4-hydroxylation. Biochemical and Biophysiocal Research Communications 148: 1144-1150
Kagimoto M, Heim M, Kagimoto K, Zeugin T, Meyer U A 1990 Multiple mutations of the human cytochrome P450IID6 gene (CYP2D6) in poor metabolisers of debrisoquine. Journal of Biological Chemistry 265: 17209-17214
Langston J W, Ballard P, Tertrud J, Irwin I 1983 Chronic Parkinsonism in humans due to a product of meperidine-analog synthesis, Science 219: 979-980
Prochazka M, Leiter E H, Serreze D V, Coleman D L 1987 Three recessive loci required for insulin-dependent diabetes in non-obese diabetic mice. Science 237: 186-289
Shah R R, Oates N S, Idle J R, Smith R L, Lockhart J D P 1982 Impaired oxidation of debrisoquinein patients with perhexiline neuropathy. British Medical Journal 284: 295-299
Smith C A D, Gough A C, Leigh P N et al 1992 Debrisoquine hydroxylase gene polymorphism and susceptibility to Parkinson's disease. Lancet 339: 1375-1377
Steventon G B, Heafield M T E, Waring R H, Williams A C 1989 Xenobiotic metabolism in Parkinson's disease. Neurology 39: 883-887
Testa B, Naylor R, Costall B, Jenner P, Marsden C D 1985 Does an endogenous methylpyridinium analogue cause Parkinson's disease? Journal of Pharmacy and Pharmacology 37: 679-680
Tetrud J W, Langston J W 1989 MPTP-induced Parkinsonism as a model for Parkinson's disease. Acta Neurologica Scandinavica Suppl 126: 35-40
Todd J A, Bell J I, McDevitt H O 1987 HLA-DQbeta gene contributes to susceptibility and resistance to insulin-dependent diabetes mellitus. Nature 329: 599-604
Todd J A, Mijovic C, Fletcher J, Jenkins D, Bradwell A R, Barnett A H 1989 Identification of susceptibility loci for insulin-dependent diabetes mellitus by trans-racial gene mapping. Nature 338: 587-589
Todd J A, Aitman T J, Cornall R J et al 1991 Genetic analysis of autoimmune type I diabetes mellitus in mice. Nature 351: 542-547