1017

systems for the identification of Mycobacterium tuberculosis and Mycobacterium avium-intracellulare. Diag Microbiol Infect Dis 1987; 8: 69-77. 7. McFadden JJ, Butcher PD, Thompson J, Hermon Taylor J. Use of DNA probes to detect mycobacterial DNA in tuberculosis tissue. Biochem Soc Trans 1987; 15: 548-49. 8.Hance AJ, Grandchamp B, Levy-Frebault V, et al. Detection and identification of mycobacteria by amplification of mycobacterial DNA. Mol Microbiol 1989; 3: 843-49. 9.Burnet FM. The clonal selection theory of acquired immunity. Cambridge: Cambridge University Press, 1959: 160-63. 10. Mitchell DN, Rees RJW. The nature and physical characteristics of transmissible agents from human sarcoid and Crohn’s disease tissues.

J, Marsac J, Saltiel JC, eds. Proceedings of 9th International Conference on Sarcoidosis, Paris, 1983. Oxford: In: Chretien

Pergamon Press, 1983: 132-41. 11. Graham DY, Markesich DC, Kalter DC, Yoshimura HH. Isolation of cell wall defective acid fast bacteria from skin lesions in patients with sarcoidosis. In: Grassi C, Rizzato G, Pozzi E, eds. Sarcoidosis and other granulomatous disorders, Amsterdam: Elsevier, 1988: 161-64. 12. Hanngren A, Oldham G, Eklund A, Hoffner S, Stjernberg N, Westerdahl G. Tuberculostearic add in lymph nodes from patients with sarcoidosis. Sarcoidosis 1987; 4: 101-04. 13. Parkes SA, Baker SB de C, Bourdillon RE, Murray CRH, Rakshit M. Epidemiology of sarcoidosis in the Isle of Man: a case controlled study. Thorax 1987; 42: 420-26.

SHORT REPORTS Mutant debrisoquine hydroxylation genes in Parkinson’s disease

the CYP2D6 homozygous wild-type allele (wt/wt) and fourteen metabolically deficient or partly deficient phenotypes that arise from combinations of the CYP2D6 A, B, D, and E mutant alleles.8

patients with Parkinson’s disease of mean age 72-3 (SD 11-4) were recruited from two local outpatient clinics with the approval of the local ethics committee. After full informed consent was obtained, 10 ml blood was collected into edetic acid (EDTA) tubes, DNA was prepared, and PCR (for CYP2D6 A and B) and RFLP (for CYP2D6 D and E) assays were done to detect CYP2D6 polymorphism.8 For population controls, genotype distributions in 72 non-age-matched, non-sex-matched healthy volunteers, of the same ethnic origin, were used. Neither age nor gender has been found to influence the distribution of CYP2D6 phenotypes or genotypes in healthy populations. CYP2D6 heterozygotes have in-vivo debrisoquine metabolism intermediate between the two homozygous genotypes.8 Partly impaired debrisoquine metabolism has been observed in Parkinson’s disease,2.4,5 and it was thought important to detect heterozygotes as well as poor metabolisers. 53

years

The

frequency of fifteen genotypes of CYP2D6 (debrisoquine 4-hydroxylase) in 53 patients with Parkinson’s disease was determined by the polymerase chain reaction (PCR) and restriction fragment length polymorphism (RFLP) analyses and compared with the findings in 72 healthy controls. The commonest mutant allele, CYP2D6B, was twice as frequent among patients as in controls, with an approximate relative risk ratio of 2·70 (95% confidence interval 1·14-6·41; p=0·0063) for subjects homozygous or heterozygous for this allele.

Although

exposure to chemical neurotoxins such

1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine

as

(MPTP)

precipitate parkinsonism, idiopathic Parkinson’s disease thought to be caused by a combination of unknown environmental, genetic, and degenerative factors.! Clinical toxicity to environmental agents is more likely to arise in patients with genetically impaired drug metabolism, and Barbeau and colleagues studied polymorphic debrisoquine 4-hydroxylation in a series of parkinsonian patients and controls.2 The tendency towards poorer metabolism of debrisoquine among their patients has since been reevaluated,3,4 with the conclusion that the observed effect was probably an artifact of concomitant drug therapy. Nevertheless, some data indicate that early-onset can

is

Parkinson’s

disease

is

indeed

associated

with

The table shows the distribution in patients with Parkinson’s disease and controls of the fifteen CYP2D6 genotypes which our assays would detect. 30 of 53 (57%) cases and 45 of 72 (63%) controls had no detectable mutations (wt/wt genotype). The frequency of the CYP2D6B mutant allele was doubled from 15 of 144 (10-4%) control chromosomes to 23 of 106 (21-6%) in cases (X2 with Yates’ correction 6.26; p < 0-025); the estimated odds ratio for heterozygous/homozygous CYP2D6B in Parkinson’s disease is therefore 2-70 (95% CI 1-14-6-41; p=0-0063). Interestingly, the total frequency of the rarer alleles CYP2D6 D and E, which arise from deletion of the CYP2D6 gene,8was reduced from 12 of 144 (8.3%) =

CYP2D6 GENOTYPES IN PATIENTS WITH PARKINSON’S DISEASE AND CONTROLS

the

debrisoquine poor metabolism (PM) phenotypes Similarly, studies in animals (the neurotoxins MPTp6 and tetrahydroisoquinoline [TIQ]7 are both presumed to be metabolised in rats by the equivalent of the human enzyme CYP2D6) have linked the CYP2D6 allele to chemical neurotoxicity and Parkinson’s disease. DNA analysis obviates the confounding effects of concurrent drug therapy on metabolic phenotyping, and unambiguous analysis of CYP2D6 polymorphism and its possible relation to Parkinson’s disease is now possible. The polymerase chain reaction (PCR) and restriction fragment length polymorphism (RFLP) techniques can be used to identify

wt= wild type allele, = none detected, CYP2D6A and B alleles detected by PCR analysis, CYP2D6 D and E alleles characterised by fragments of 13 and 11 kb, respectively, on Southern blots of genomic DNA after Xbal digestion and use of a full-length cDNA probe8

1018

chromosomes in controls to 1 of 106 (0-9%) in cases (r’=119, p < 0-001). The excess of CYP2D6B alleles among patients with Parkinson’s disease was almost entirely accounted for by a doubled frequency of the wt/CYP2D6B genotype compared with controls (37-7% 18-1%). The numbers studied were too small to detect any changes in homozygous PM frequency (2 in both groups), and we did not attempt to identify the rare CYP2D6C allele, also associated with deficient CYP2D6-mediated metabolism.9 How might the hepatic CYP2D6 isozyme influence the onset of Parkinson’s disease? Apart from a possible role in environmental neurotoxin inactivation, recent evidence indicates that the gene product may influence brain function. A partial cDNA with 100% sequence homology to CYP2D6 has recently been cloned from the caudate nucleus in man, to and CYP2D6-like metabolic activity is distributed throughout the canine brain; this protein has an affinity for drugs that interact with the neuronal dopamine transporter. to Individuals homozygous or heterozygous for CYP2D6 mutations may therefore have lower rates of CYP2D6-mediated chemical metabolism in the brain and altered dopamine homoeostasis. In this preliminary study we found that the commonest genotype to confer impaired debrisoquine metabolism, wt/CYP2D6B, is twice as common among patients with Parkinson’s disease than in the normal population. These findings would not be influenced by concomitant drug therapy. However, we did not observe any overall increase in the total number of mutant alleles among parkinsonian patients and it is possible that a specific genetic mechanism, such as linkage disequilibrium, underlies the apparent association between Parkinson’s disease and the CYP2D6B allele, or indeed that the findings of this small pilot study, with controls matched for ethnicity alone, could be explained by chance. Larger controlled studies, perhaps with an emphasis on patients with early onset of Parkinson’s

disease, are now required. We thank Dr R. G. Cooper and Dr D. Bates for assistance in recruitment and Dr N. Caporaso for statistical advice. Supported Wellcome Trust, BAT Ltd, and Bayer UK plc.

patient by the

REFERENCES 1. Calne DB, Langston JW. Aetiology of Parkinson’s disease. Lancet 1983; ii: 1457-59. 2. Barbeau A, Cloutier T, Roy M, Plasse L, Paris S, Poirier J. Ecogenetics of Parkinson’s disease: 4-hydroxylation of debrisoquine. Lancet 1985; ii: 1213-16. 3. Poirier J, Roy M, Campanella G, Cloutier T, Paris S. Debrisoquine metabolism in parkinsonian patients treated with antihistamine drugs. Lancet 1987; ii: 386. 4. Benitez J, Ladero JM, Jimenez FJ, et al. Oxidative polymorphism of debrisoquine in Parkinson’s disease. J Neurol Neurosurg Psychiatry 1990; 53: 289-92. 5. Tanner CM, Chen B, Wang WZ, et al. Environmental factors in the etiology of Parkinson’s disease. Can J Neurol Sci 1987; 14 (suppl): 419-23. 6. Fonne-Pfister R, Bargetzi MJ, Meyer UA. MPTP, the neurotoxin inducing Parkinson’s disease is a potent competitive inhibitor of human

cytochrome P450 isozymes (P450 buf1, P450db1) catalyzing debrisoquine 4-hydroxylation. Biochem Biophys Res Commun 1987; and

rat

148: 1144-50. 7. Ohtas S, Tachikawa O, Makino Y, et al. Metabolism and brain accumulation of tetrahydroisoquinoline (TIQ) a possible parkinsonism inducing substance, in an animal model of a poor debrisoquine metabolizer. Life Sci 1990; 46: 599-605. 8. Daly AK, Armstrong M, Monkman SC, et al. Genetic and metabolic criteria for the assignment of debrisoquine 4-hydroxylation (cytochrome P4502D6) phenotypes. Pharmacogenetics 1991; 1: 33-41. 9. Tyndale R, Aoyama T, Broly F, et al. Identification of a new variant CYP2D6 allele lacking the codon encoding Lys-281: possible association with the poor metabolizer phenotype. Pharmacogenetics 1991; 1: 26-32.

10. Tyndalc RF, Sunahara R, Inaba T, P45011D1

et

al. Neuronal cytochrome

(debrisoquine/sparteine-type): potent inhibition of activity

by (-)-cocaine and nucleotide sequence identity to human hepatic P450 gene CYP2D6. Mol Pharmacol

1991; 40: 63-68.

ADDRESSES. Pharmacogenetics Research Unit (M. Armstrong, BSc, A. K. Daly, PhD, S. Cholerton, PhD, Prof J. R Idle, PhD), and Wolfson Unit of Clinical Pharmacology (D N Bateman, FRCP), Department of Pharmacological Sciences, Medical School, Framlington Place, Newcastle upon Tyne NE2 4HH, UK. Correspondence to Prof J R Idle.

Sensitivity of serological assays to identify blood donors with hepatitis C viraemia

Blood donors at high risk of hepatitis C virus (HCV) infection were tested for viraemia by the polymerase chain reaction (PCR). PCR results were accepted as positive only if reactive in 3 of 4 tests and if confirmed in an independent laboratory. The sera were also tested by 6 different assays to determine the ability of current serological assays to detect viraemic blood donors. Of 19 PCR-positive sera, only 13 (68%) were detected by the most sensitive of the serological assays. If these results are confirmed, automated PCR assays may be required for blooddonor screening to prevent transmission of HCV.

Anti-HCV-positive blood readily transmits HCV infection.1 But how sensitive are existing serological assays for the detection of anti-HCV antibodies and thereby the identification of infectious blood donors? Zanetti et aP describe 3 blood donors with viraemia detected by the polymerase chain reaction (PCR) who were seronegative by standard anti-HCV enzyme-linked immunosorbent assay (ELISA, Ortho). Aach et aP found that first-generation (anti-ClOO-3) ELISA assays detected 81%, and secondgeneration ELISA assays 93%, of donors implicated in the transmission of non-A, non-B hepatitis, but cases had to meet strict clinical criteria for the diagnosis of hepatitis. We have observed transfused patients with little or no increase in circulating transaminase concentrations who acquired chronic HCV viraemia detected by PCR, with or without accompanying anti-HCV seroconversion (unpublished observations). To investigate the sensitivity of anti-HCV serological assays we tested samples from blood donors at high risk of HCV infection by PCR, and then retested these samples by available immunoassays.

Disposable pipettes were used to sample plasma from 108 donors who had alanine aminotransferase concentrations above 100 IU/1. Samples were tested on 4 different occasions in our laboratory; nested PCR assays were done with primer pairs from the 5’ non-coding region as previously described4 and only samples positive in at least 3 of 4 tests were considered positive and then sent under code to the Liver Diseases Section, National Institutes of Health, where they were retested. Weakly positive samples were tested after cold precipitation with 10% polyethylene glycol. Blood from 25 of these 108 high-risk donors was found to be positive by PCR, 19 of which were independently confirmed to be positive at NIH. To provide an estimate of HCV-RNA titre, cDNA from positive samples was serially diluted in 0 01 mol/1 "tris", and