121

of the pyridine ring by cytochrome P-450 or xanthine oxidase,followed by oxidation by aldehyde oxidase. Cytochrome P-450activity in Parkinson’s disease when probed with debrisoquineseems normal but probing with phenytoin showed that there maybe impairment in a few patients since reduced para-hydroxylationwas seen.s Our results might be explained by dysfunction ofhydroxylation mediated by an isoenzyme of cytochrome P-450 orperhaps xanthine or aldehyde oxidases, leading to accumulation ofthe Nl-methyl derivative. This may provide a basis for individualsusceptibility to selective neuronal damage from pyridines,isoquinolines, or B-carbolines, whose N-methylated derivatives areknown to be more toxic and more MPP+-like than the parentcompounds.

University Department of Neurology andSchool of Biochemistry,

University of Birmingham,Edgbaston, Birmingham B15 2TH, UK

SANDRA GREENSTEVE BUTTRUMHELEN MOLLOYGLYN STEVENTONSTEPHEN STURMANROSEMARY WARINGHARDEV PALLADRIAN WILLIAMS

1. McCeaner GM, Bender DA. The metabolism of high intakes of tryptophannicotinamide and nicotinic acid in the rat. Br J Nutr 1986; 56: 577-86.

2. Markey SP, Johannessen JN, Chiueh CC, Burns RS, Herkenham MA. Intraneuronalgeneration of a pyridinium metabolite may cause drug-induced parkinsonism.Nature 1984; 311: 464-67.

3. McKee RW, Kang-Lee YA, Panaqua M, Swenseid ME. Determination ofnicotinamide and metabolic products in urine by high performance liquidchromatography. J Chromatogr 1982; 230: 309-17.

4. Steventon GB, Heafield MTE, Sturman SG, Waring RH, Williams AC, Ellingham J.Degenerative neurological disease and debrisoqume 4-hydroxylation capacity.Med Sci Res 1989; 17: 163-64.

5. Ferrari MD, Peeters EAJ, Haan J, et al. Cytochrome p450 and Parkinson’s disease,poor parahydroxylation of phenytoin. J Neurol Sci 1990; 96: 153-57.

CSF S-adenosylmethionine inneuropsychiatric disorders

SIR,-Low cerebrospinal fluid (CSF) S-adenosylmethionine(SAM) has been linked to folate deficiency in HIV-positive childrenwith neurological involvement.1 However, Dr Keating andcolleagues (April 20, p 935) found no evidence of folate or vitaminB12 deficiency in adults. Our investigations of CSF SAM inneuropsychiatric disorders have implications for the interpretationof the HIV studies: CSF SAM was significantly lower in patientswith severe depression than in neurological controls.2 However,some of the lowest values were in smaller groups of patients witheither Alzheimer’s dementia or subacute combined degeneration(SCD). As is predictable, low CSF SAM can be associated withfolate or vitamin B 12 deficiency, whether in SCD3 or in depression,in which folate deficiency occurs in 20-30% of patients. However,not all patients with low CSF SAM have folate or vitamin B12deficiency. One possible explanation for the low CSF SAM inKeating and co-workers’ patients is that they may have been moredepressed than the HIV-negative controls.These findings for HIV-positive patients are likely to give rise to

trials of treatment with SAM. We have shown that administrationof SAM parenterally or orally is associated with a rise in CSF SAMin depressed or demented patients, indicating that this metabolitecrosses the blood-brain barrier The antidepressant effect of SAMhas been confirmed in several studies.5 Our finding of low CSFSAM in some patients with depression provides a rational basis forthe use of the metabolite as an antidepressant in this subgroup.The very low concentrations of CSF SAM in our Alzheimer’s

groUp2 suggest a possible role for disturbed methylation in somedementias. A depletion of SAM in the nervous system will probablyhave widespread metabolic consequences since it is the sole methyldonor in many metabolic reactions involving neurotransmitters(including acetylcholine), monoamines, polyamines, membranephospholipids, proteins, and nucleoproteins.

Furthermore, it has been suggested that the methylation of DNAis an epigenetic mechanism that determines the long-termfunctioning of cells and that the gradual loss of methyl groups

derived from SAM could contribute to ageing. Our open6 andsingle-blind’ pilot studies of intravenous and oral SAM in dementiasuggest that controlled trials of SAM in this clinical situation wouldbe worth while.

Departments of Neurology and Psychiatry,King’s College Hospital,London SE5 9RS, UK

T. BOTTIGLIERIE. H. REYNOLDSB. K. TOONE

Department of Psychiatry,Northwick Park Hospital,Harrow, Middlesex M. W. P. CARNEY

1. Surtees R, Hyland K, Smith I. Central-nervous-system methyl-group metabolism inchildren with neurological complications of HIV infection. Lancet 1990; 335:619-21.

2. Bottiglieri T, Godfrey P, Flynn T, et al. Cerebrospinal fluid S-adenosylmethionine indepression and dementia: effects of treatment with parenteral and oral S-

adenosylmethionine. J Neurol Neurosurg Psychiatry 1990; 53: 1096-98.3. Bottiglieri T, Godfrey P, Carney MWP, Toone BK, Reynolds EH. CFS-S-

adenosylmethionine in neurological and psychiatric patients: effects of treatmentwith parenteral and oral S-adenosylmethionine. In: Linnell JC, Bhatt R, eds.Biomedicine and physiology of vitamin B12. London: Children’s Medical Charity,1990. 167-74.

4 Godfrey PSA, Toone BK, Carney MWP, et al. Enhancement of recovery from

psychiatric illness by methylfolate. Lancet 1990; 336: 392-95.5. Kagan BL, Sultzer DL, Rosenlicht N, Gemer RH. Oral S-adenosylmethionine in

depression: a randomized double-blind, placebo-controlled trial. Am J Psychiatry1990; 147: 591-95.

6. Reynolds EH, Carney MWP, Toone BK, et al. Transmethylation and

neuropsychiatry. Cell Biol Rev 1987; (suppl 1): 93-102.7. Reynolds EH, Godfrey PSA, Bottiglieri T, et al. S-adenosylmethionine and

Alzheimer’s disease. Neurology 1989; 39 (suppl 1): 397.

Gastric-aspirate angiotensin-convertingenzyme in gastrosarcoidosis

SIR,-We have detected angiotensin-converting enzyme (ACE)in gastric aspirates of a patient with gastrosarcoidosis.A symptom-free 55-year-old Japanese woman was admitted for

evaluation of abnormal upper gastrointestinal findings. No crackleswere heard on chest auscultation and no lymphadenopathy or skinlesions were present. Gastroscopy revealed multiple yellowish ulcerscars and nodules in the body and fomix of the stomach.

Transendoscopic gastric biopsy showed multiple non-caseatingepithelioid granulomas containing multinucleated giant cells in themucosa and submucosa without evidence of malignant disease.Staining for mycobacteria and fungi was negative, as was culture ofthe biopsy specimen. Transbronchial biopsies showed no

granulomas. Consequently, gastrosarcoidosis was diagnosed.Serum ACE was 22-2 IU/1 (normal 7-0-25-0), but ACEconcentration in gastric aspirate was double this value (48-1 IU/1).Serum ACE has proved useful in the diagnosis and management

of sarcoidosis,l and Chan Seem et al2 detected ACE in the

cerebrospinal fluid of a patient with neurosarcoidosis. The presenceof ACE in gastric aspirate has not, to our knowledge, beenpreviously reported in gastrosarcoidosis. Although ACE in gastricaspirate can be accounted for by leakage of systemic ACE, thiscannot be the whole explanation because the ACE concentrationwas strikingly higher than that in serum and because, in otherdiseases with presumed leakage of serum (gastritis, gastric ulcer),ACE was hardly detected in gastric aspirates (05 [0.4] IU/1,n = 10). We believe that ACE in gastric aspirate derives from localproduction of sarcoid tissue and so more directly reflects diseaseactivity than does serum ACE.When gastrosarcoidosis is suspected, ACE in gastric aspirate

may be more helpful as both an index of disease activity and an aid indiagnosis than serum ACE.

First Department of Internal Medicine,Gunma University School of Medicine,3-39 Showa-Machi, Maebashi, Gunma 371, Japan

SHINICHI OKAMURAKEIGO IESAKISEIJI MURAMATSU

1. Studdy P, Bird R, James DG, Sherlock S. Serum angiotensin-converting enzyme(SACE) in sarcoidosis and other granulomatous disorders. Lancet 1978; ii:1331-34.

2. Chan Seem CP, Norfolk G, Spokes EG. CSF angiotensin-converting enzyme inneurosarcoidosis. Lancet 1985; i: 456-57.

3. Silverstein E, Friedland J, Lyons HA, Gourin A. Elevation of angiotensin-convertingenzyme in granulomatous lymph nodes and serum in sarcoidosis: clinical andpossible pathogenic significance. Ann NY Acad Sci 1976; 278: 498-513.