The CYPZDGB Allele Is Associated with a Milder Synaptic Pathology in

Alzheimer's Disease Xiaohua Chen, MD," Yu Xia, MS," Michael Alford, BS," &chard DeTeresa, BS," Lawrence Hansen, MD,"

Melville R. Klauber, PhD,? Robert Katzman, MD," Leon Thal, MD," Eliezer Masliah, MD," and Tsunao Saitoh, PhD"

Both genetic and environmental factors affect the progression of Alzheimer's disease (AD). The presence of cortical Lewy bodies in AD patients is associated with an altered presentation of AD pathology suggestive of an interaction between the pathogenesis of Lewy bodies and AD lesions. Since the CYP2D6B mutant allele is often present in patients with Lewy body diseases (Parkinson's disease and Lewy body variant of AD), we extended these prior observations by studying the neuropathology associated with the presence of the CYPZDGB mutant allele in a pure AD population without Lewy bodies. AD patients who possessed the CYP2D6B mutant allele, in comparison with those without the CYP2D6B allele, were found to have a smaller decline in two synaptic markers, choline acetyltransferase and synapto- physin, in the frontal cortex relative to normal control values. On the other hand, senile plaques and neurofibrillary tangles were not significantly affected by the presence of the CYP2D6B mutant allele in the frontal cortex of AD patients. Association of the CYP2D6B mutant allele with Lewy body formation in both Parkinson's disease and the Lewy body variant of AD and with the milder synaptic pathology in pure AD without Lewy bodies suggest that depending on the contribution of other genetic and environmental factors, this mutant allele may be involved with different aspects of neurodegeneration.

Chen X, Xia Y, Alford M, DeTeresa R, Hansen L, Klauber MR, Katzman R, Thal L, Masliah E, Saitoh T. The CYP2D6B allele is associated with a milder synaptic pathology

in Alzheimer's disease. Ann Neurol 1995;38:653-658

Alzheimer's disease (AD) is a multifactorial disease as- sociated with both genetic and epigenetic factors. Mu- tations in the gene encoding amyloid protein precursor (APP) in several patients with familial AD (FAD) of early onset firmly established the genetic cause of this disease 11-31. Later, a mutation on chromosome 14q24.3 was found to be responsible for the majority of early-onset AD 14-10]. However, individuals with identical FAD mutations do not develop clinical symp- toms at the same age. Therefore, it is likely that other genes and epigenetic factors influence the disease prog- ress. The presence of the apolipoprotein E (APOE) €4 allele in more than 50% of AD patients as compared to 15% in the general population amply justifies the idea that a certain genetic makeup makes an individual more prone to AD development 111-161. Except for the genes mentioned above, no other genes that influ- ence the onset or progression of AD are known.

Recently, the CYP2D6B mutation, which previ- ously was determined to be a risk factor for Parkinson's disease C17-201, was identified to be a genetic risk

factor for the Lewy body variant (LBV) of AD 1211. In the LBV the expression of neurofibrillary pathology is relatively mild 1221. If the presence of biochemical or cell biological traits associated with CYP2D6B is involved in the milder expression of AD pathology in the LBV, it is predicted that in pure AD without Lewy bodies the presence of the CYP2D6B allele may also be associated with a reduced expression of AD pathol- ogy. In the current study, we compared biochemical and morphological alterations typically found in AD between AD individuals with at least one CYP2D6B allele and those without this allele. We report that indi- viduals with pure AD with the CYP2D6B allele have milder synaptic pathology than do those without this allele.

Materials and Methods Subjects A total of 121 autopsy subjects from the Alzheimer Disease Research Center at the University of California, San Diego, were utilized for the present analysis. The description of sub-

Address correspondence to Dr Saitoh, Department of Neurosci- ences (0624), School of Medicine, University of California at San Diego, La Jolla, CA 92093-0624.

From the Departments of *Neurosciences and iFamily and Preven- tive Medicine, University of California at San Diego, La Jolla, CA.

Received May 2, 1995, and in revised form Jun 23 Accepted for publication Jun 26, 1995

~

Copyright 0 1995 by the American Neurological Association 653

Table I. Description of AD Subjects

Age at Death Duration MMSE APOE €4” AD Subjects N (yr) ( yr) Score (5%) With CYP2D6B 42 77.8 2 1.2 8.2 % 0.6 9.5 2 1.3 73.8

Used for ChAT 26 77.5 ? 1.4 7.6 ? 0.8 11.5 % 1.6 69.2

Without CYP2D6B 79 78.4 2 0.9 8.9 2 0.5 8.5 % 1.3 60.8 Used for ChAT 57 77.5 ? 1.2 9.0 * 0.6 7.4 ? 1.6 63.2

Used for synaptophysin 18 78.2 ? 2.3 6.9 ? 0.9 10.6 ? 2.0 83.3

Used for synaptophysin 34 77.2 t 1.6 9.2 * 0.6 9.5 2 2.0 55.9

aPercentage of individuals with one or two APOE €4 alleles. There were no statistically significant differences in the age at death, duration, MMSE score, or APOE €4 represenration, using an unpaired two-tailed t test or Fisher’s exact test, between any pairs with and those without CYPZDGB mutant alleles. There is a nonsignificant trend for the APOE €4 allele to be found more frequently among pure AD individuals with the CYP2DGB mutant allele.

MMSE = Mini-Mental Stare Examination; APOE = apolipoprotein E; ChAT = choline aceryltransferase

jects with and those without the CYP2D6B mutant allele is found in Table 1. These individuals were extensively studied neurologically and neuropathobgically , and morphometric scores, onset of dementia, and APOE genotype were avail- able for correlation with the CYP2D6 genotype. All subjects were diagnosed as having probable AD by National Institute of Neurological and Communicative Disorders and Stroke/ Alzheimer’s Disease and Related Disorders Association (NINCDSIADRDA) criteria [23] and this was confirmed at autopsy by the presence of adequate numbers of plaques [24]. These subjects did not have Lewy bodies. Frozen sam- ples from the midfrontal cortex were available from a subset of subjects and were homogenized for biochemical determi- nation.

CYP2DG Genotyping We genotyped CYP2D6 in all 1;!1 brains. The procedure of CYP2D6 genotyping has been published [21). Briefly, the 334-bp CYP2D6 genomic DNA was amplified by polymer- ase chain reactions (PCRs) and di,gested by the B J ~ N I restric- tion enzyme. The digested products were run on a 1.55% agarose gel and stained with 0.5 mgiml of ethidium bromide. The mutant CYP2D6B allele having a G-to-A transition gave a 334-bp band whereas the wild-type allele was digested to yield 105- and 229-bp bands.

Neuropatbological Analysis Neuritic plaques and neurofibrillary tangles in the midfrontal region were counted on paraffin sections 10 pm thick stained with thioflavine-S as previously described [25]. These counts were done manually with a fluorescence microscope using a 540-nm wavelength filter {26]. Plaques were enumerated in three fields along the side of the gyms at a total magnifica- tion of X 125 (field size, 1.60 mm2). Tangles were enumer- ated similarly at a magnification of x 500 (field size, 0.10 mm2).

Immunocbemical Determination of Synaptopbysin Dot blot immunoquantification of synaptophysin was per- formed in a subset of frozen frontal cortex samples as pre- viously described [27]. Briefly, homogenates (particulate fraction) from each sample were applied in triplicate on nitro- cellulose with a MiniFold Dot blot apparatus. Blots were incubated with mouse monoclonal antibody against syn-

aptophysin (SY38, Boehringer Mannheim, Indianapolis, I N ) [28), followed by rabbit polyclonal anti-mouse sec- ondary antibody and ’*’ I-protein A. Radiolabeled blots were analyzed with a phosphorimager (Molecular Dynam- ics, Sunnyvale, CA) for subsequent quantification with the ImageQuant software relative to a standard curve. We pre- viously showed that there is a strong correlation between synaptophysin quantified by dot blot and estimation of the number of synaptophysin-immunolabeled terminals by laser scanning confocal microscopy {27]. Furthermore, experimen- tal studies have shown that immunoquantification of synapto- physin is a fast and reliable method to study synaptic popula- tions in animal models of denervatiodreinnervation and in neurodegenerative disorders 129, 301.

Cboline Acetyltransferase Choline acetyltransferase (ChAT) activity was assayed in a subset of brain samples as described previously 1311. Ho- mogenates were incubated in 20-ml vials with a mixture containing 0.1 mCi acetyl-{ 1-’*C1 coenzyme A, 300 mM so- dium chloride (NaCI), 50 mM sodium phosphate (pH 7.4), 8 mM choline bromide, 20 mM ethylenediaminetetra- acetic acid (EDTA; p H 7.4), and 0.1 mM physostigmine at 37°C for 60 minutes in a volume of 0.2 ml. The reaction was terminated by the addition of 5 ml of 10 mM sodium phosphate buffer (pH 7.4) followed by 2 ml of acetonitrile containing 20 mg of sodium tetraphenylborate (Kalignost) and 10 ml of toluene-based scintillation fluid containing 0.5% 2, 5-diphenyloxazole (PPO) and 0.03% 1, 4-bis(2-(5- phenyloxazolyl)) benzene (POPOP). The vials were gently shaken before counting, to facilitate the extraction of acetyl- choline into the toluene phase.

Statistical Analyses Due to sample size limitations, for the purpose of statistical analysis, AD subjects were divided into two groups: (1) those with the CYP2D6B mutant allele (includes homozygote and heterozygote) and (2) those without the CYP2D6B mutant allele. Comparisons of the CYP2D6 genotype among AD subjects were done with the unpaired, two-tailed Student’s t test. Values were expressed as the mean 2 standard error of the mean (SEM). Statistical significance was determined at the 5% level.

654 Annals of Neurology Vol 38 N o 4 October 1995

F i g 1. Choline acetjiltransferase (ChAT) actitfity in the mid- frontal region of cerebral cortex in pure A D subjects with (B) a,2d those u'ithol,t (w, the C Y P ~ D G B mutant al/elt. ~ ~ ~ ~ ~ , ~ k iiidicates that ChAT actiiity in ~ubje6-t~ u i t h the CYP2DGB allele was significantly higher than the artit,ity in those uith- out this allele (p = 0.037 by tu:o-tailed Student's t test). The tlC,O gvoups co7npared are 7aatrhed f o r age, dl,ratio,l of disease. st- imerity of dementia, and apolipoprotein E €4 status.

F i g 2. Synaptophyszn roncetitration in the midfrontal region of cerebral cortex in pure AD subjects with (B, and those without (W) the CYP2DbB mutant allele. Synaptophy~in ler,els in sub- jects u h h the CYP2DGB allele were signiJcantIy higher than thosefound in .subjects without this allele (p = 0.049 by the two-tailed Student's t test). The f u ' o groups are matched as i d - cated in the Figure I legend.

Results The Presence of the CYP2D6B Mutant Allele Is Associated with a Smaller Choline Acetyltransferase Decline in the Frontal Cortex A cholinergic deficit is consistently found in AD neo- cortex in conjunction with cholinergic neuronal loss in the nucleus basalis of Meynert (321. This is demon- strated by a decline in ChAT activity in homoge- nates of cortex tissue from A D patients (Fig 1). How- ever, compared to controls, AD subjects without the CYP2D6B mutant allele showed an average decline of 50% in ChAT activity in the frontal cortex whereas those with the B mutant allele showed only a 20%) decline. The difference between AD subjects with and those without the CYP2D6B mutant allele was signifi- cant.

The Presence of the CYP206B Mutant Allele Is Associated with a Smaller Decrease in a Generic Presynaptic Marker. Synaptophysin Frontal ChAT activity reflects the integrity of choliner- gic nerve terminals, which are a minor fraction of total nerve terminals. To evaluate the intactness of total nerve terminals, the concentration of synaptophysin, a presynaptic vesicle membrane protein, was quantified in homogenates of frontal cortex. Compared to con- trols, AD subjects without the CYP2D6B mutant al-

lele showed an average decline of 45%) in synaptophy- sin concentration in the frontal cortex whereas those with the B mutation showed only a 30% decline (Fig 2). The difference between AD subjects with and those without the CYP2D6B mutant allele was significant.

Senile Plaques and Neurofbrilla y Tangles, Two Hallmark Lesions of A D , Are Not Significantb Affected by the Presence of the CYP2D6B Mutant Allele AD is characterized by the presence of senile plaques and neurofibrillary tangles [ 331. The comparison of plaque numbers in the frontal cortex between individu- als with and those without the CYP2D6B mutant allele did not reveal significant differences (Fig 3). In the frontal cortex of A D patients, the number of senile plaques is so high that quantification may not ade- quately reflect the severity of the pathology due to a ceiling effect. Therefore, quantification of plaques was performed in the hippocampus where the number of plaques is one fourth to one third of that found in the frontal cortex. Again no effect of the CYP2D6B mu- tant allele was found (1 1.4 ? 0.7 for wild type, N = 78, vs 11.8 * 1.1 for the B mutation, N = 41). Simi- larly, comparison of tangle numbers between individu- als with and those without the CYP2D6B mutant allele did not reveal any significant differences either (Fig 4).

Chen et al: CYPZDGB Effect on AD Pathology 655

W (N=79)

T I

B (N=42)

CYP2D6 Allele

5/

f I Q T

W (N=79)

B (N=42)

CYP2D6 Allele

F i g 3. Effect of the presence of the CYP2D6B mutant allele on the plaque number in AD. Plaque numbers in the midfrontal re- gion of the cerebral cortex in pure A D subjects with (B) and those without tW) the CYPZDGB mutant allele were compared. There was no significant group dz#erence (two-sided Student’s t test, t = 1.511, p = 0.133).

F i g 4. Effect of the presence of the CYP2D6B mutant allele on the tangle number in AD. Tangle numbers in the midfrontal re- gion of the cerebral cortex in pure A D subjects with (B) and those without (W) the CYP2D6B mutant allele were compared. There was no significant group difference (two-sided Student’s t test, t = 0.911, p = 0.364).

Discussion The recent discovery that the presence of the APOE €4 allele is the major genetic risk factor for AD prompted the analysis of many psychometric and neu- ropathological variables in relation to APOE alleles [12, 14, 34, 351. However, the ability of other genetic influences to modify the neur opathological severity of AD has not been investigated The current study dem- onstrated that the presence of the CYP2D6B mutant allele is associated with less severe synaptic pathology in AD. Interestingly, the milder synaptic neuropathol- ogy is not accompanied by the presence of significantly fewer senile plaques or neurofibrillary tangles in frontal cortex. Although we do not know the molecular or cellular mechanisms of amyloid and senile plaque for- mation, the lack of significant effect of the CYP2D6B allele on this process suggests that the primary target of the alteration associated with the CYP2D6B mutant allele is in the synaptic compartment of neurons. Con- sistent with this interpretation, we did not observe the effect of the CYP2D6 genotype on neuronal counts in the frontal cortex (data not shown). However, a possi- bility remains that cholinergic neuron counts in the nucleus basalis may be affected by the CYP2D6 geno- type in view of the genetic effect on ChAT activity in the frontal cortex that receives the cholinergic projec- tion from the nucleus basalis. Future studies need to clarify this point.

What might be a potential mechanism for the CYP2D6B allele association with its interfering effect on synaptic pathology in AD? There are two possibili-

ties. It is possible that the CYP2D6B allele itself is involved or, alternatively, it is also possible that a muta- tion close to the CYP2D6B allele might interfere with synaptic deterioration. If the former is correct, a possi- ble role for environmental compounds may be put forward for the synaptic pathology in AD since the CYP2D6 gene that encodes debrisoquine 4- hydroxylase is involved in detoxifying environmental toxins r36). The B mutation eliminates the splicing signal after exon 3, and thus the active enzyme. Re- duced levels of debrisoquine 4-hydroxylase in the brain tissue of CYP2D6B individuals {37) may fail to inacti- vate a potential neurotoxin that is yet to be identified in the brain tissue of AD victims.

Although the current study demonstrated the associ- ation of the CYP2D6B mutation with blunted synaptic pathology in AD, there is still a possibility that certain mutations or variations of a gene closely linked to the CYP2D6B mutation are responsible for this reduced neuropathology. The CYP2D6 gene is linked to sev- eral genes for neuronal growth-associated proteins, in- cluding platelet-derived growth factor b (PDGFb) and neurofilament-H (NF-H) on chromosome 22 at q13.1 r3S). Involvement of many growth-associated proteins E391, and PDGFb in particular C401, in the pathogene- sis of AD warrants consideration of the presence of a mutation in these genes that is linked to the CYP2D6B allele.

As the CYP2D6B mutant allele is a genetic risk factor for LBV C213, it is of interest to ask whether pure AD with the CYP2D6B allele is an incipient form

656 Annals of Neurology Vol 38 No 4 October 1995

Table 2. Effect of CYP2D6B and Lewy Body Variant (LBV) on A D Pathology

Variable

Pathology AD with CYP2D6B LBV

Synaptic pathology Choline acetyltransferase Less severe More severe Synaptophysin Less severe More severe

Plaques Comparable Less severe Tangles Comparable Less severe

Classic lesions

of the LBV. In other words, will such patients develop Lewy bodies should they live longer? The answer ap- pears to be negative. The comparison of neuropathol- ogy in pure AD subjects with the CYP2D6B and LBV does not show clear similarities (Table 2). Synaptic pa- thology is less severe in pure AD subjects with the CYP2D6B mutant allele than in those without this mu- tant allele. However, synaptic pathology is actually more severe in the LBV than in pure AD. On the other hand, classic lesions of AD, plaques and tangles, are less prominent in the LBV whereas the severity of this pathology is not significantly affected by the pres- ence of CYP2D6B in pure AD.

There are two possible explanations for these find- ings. The first possibility is that the two separate muta- tions or variations linked to CYP2D6B are, respec- tively, responsible for the LBV and less severe synaptic pathology in pure AD subjects with CYP2D6B alleles. Previous work 121) and the current study revealed the linkage of two apparently independent phenom- ena (LBV and less severe synaptic pathology) to CYP2D6B. However, from this type of analysis, it can- not be concluded that the CYP2D6B mutation is re- sponsible for the phenotype. It is possible that two entirely different genes closely linked to CYP2D6B are involved in our current and previous findings, al- though it is not probable that two such independent genes are found fortuitously.

The second possibility is that either the CYP2D6B mutant or a closely linked gene is associated with both the LBV and the reduction of synaptic pathology in pure AD. It is likely that there is yet a third group of genes (modifier) that affects the expression of a CYP2D6B or a linked gene. Manifestation of less se- vere synaptic pathology in pure AD individuals with the CYP2D6B allele may be dictated by an as yet un- known modifier gene. LBV-associated pathology may manifest with CYP2D6B in the presence of the other type of modifier gene. In this model the combination of several sets of genes dictates the neuropathology of AD. There is a precedent for different phenotypes caused by a single mutation. A mutation on the Alzhei-

mer APP at position 692 manifests as AD in several individuals and as amyloid angiopathy in other individ- uals {4 11, probably due to the different genetic back- ground of these individuals. Involvement of multiple genes and their interaction in the expression of AD neuropatholog have been discussed previously 142).

In spite of the less severe reduction of synaptophysin and ChAT activity in the frontal cortex of AD subjects with the CYP2D6B allele, there was no significant dif- ference in the duration of the disease, age at death, neuropsychological test scores, or senile plaque and neurofibrillary tangle numbers in these subjects as compared to those without this allele. The lack of the effect of less severe reduction of synaptic pathology in AD patients with the B mutant allele on the clinical aspects of the disease might be partly due to the choice of our AD subjects. Because our subjects were se- verely demented (Mini-Mental State Examination score -9), any subtle changes may be masked by the floor effect. On the other hand, the lack of an effect of less severe reduction of synaptic pathology on the other neuropathological aspects of the disease might indicate a relatively independent nature of these le- sions. Thus, the current genetic and neuropathological analysis revealed that the presence of the CYP2D6B mutant allele is a genetic factor that reduces the sever- ity of synaptic pathology in AD. The molecular and cellular mechanisms for this phenomenon and its sig- nificance in the progression of AD neuropathology need to be elucidated in the future.

This research was funded by grants from the National Institutes of Health (AGO5 13 1 and AG08205).

We thank Robert Davignon for editorial assistance

References 1.

2.

3.

4.

5.

6.

7.

8.

Goate A, Chartier-Harlin MC, Mullan M, et al. Segregation of a missense mutation in the amyloid precursor protein gene with familial Alzheimer’s disease. Nature 199 1;349:704-706 Naruse S, Igarashi S, Kobayashi H, et al. Missense mutation Val+Ile in exon 17 of amyloid precursor protein gene in Japa- nese familial Alzheimer’s disease. Lancet 1991;337:978-979 Mullan M, Crawford F, Axelman K, et al. A pathogenic muta- tion for probable Alzheimer’s disease in the APP gene at the N-terminus of p-amyloid. Nature Genet 1992;1:345-347 Schellenberg GD, Bird TD, Wijsman EM, et al. Genetic linkage evidence for a familial Alzheimer’s disease locus on chromo- some 14. Science 1992;258:668-671 Schellenberg GD, Payami H, Wijsman EM, et al. Chromosome 14 and late-onset familial Alzheimer disease (FAD). Am J Hum Genet 1973;5 3:6 19-628 St. George-Hyslop P, Haines JL, Rogaev E, et al. Genetic evi- dence for a novel familial Alzheimer’s disease locus on chromo- some 14. Nature Genet 1992;2:330-334 Van Broeckhoven C, Backhovens H, Cruts M, et al. Mapping of a gene predisposing to early-onset Alzheimer’s disease to chromosome 14q24.3. Nature Genet 1992;2:335-339 Mullan M, Houlden H, Windelspecht M, et al. A locus for

Chen e t al: C Y P 2 D 6 B Effect o n AD Pathology 657

familial early-onset Alzheimer’s disease on the long arm of chro- mosome 14, proximal to the cul-antichymotrypsin gene. Nature Genet 1992;2:340-342

9. Haltia M, Viitanen M, Sulkava R, et al. Chromosome 14- encoded Alzheimer’s disease: genetic and clinicopathological de- scription. Ann Neurol 1994;36: 362-367

10. Nechiporuk A, Fain P, Kort E, e t al. Linkage of familial Alzhei- mer disease to chromosome 14 in two large early-onset pedi- grees: effects of marker allele frequencies on lod scores. Am J Med Genet 1993;48:63-66

11. Strittmarter WJ, Weisgraber KH., Huang DY, e t al. Binding of human apolipoprotein E to synthetic amyloid P peptide: iso- form-specific effects and implications for late-onset familial Alz- heimer disease. Proc Natl Acad Sci USA 1993;90:8098-8102

12. Schmechel DE, Saunders AM, Strittmatter WJ, et d. Increased amylaici P-peptide deposition in cerebral cortex as a conse- quence of apolipoprotein E genotype in late-onset Alzheimer disease. Proc Natl Acad Sci USA 1993;90:9649-9653

13. Roses AD. Apolipoprotein E affects the rate of Alzheimer dis- ease expression: P-amyloid burden is a secondary consequence dependent on APOE genotype a.nd duration of disease. J Neu- ropathol Exp Neurol 1994;53:429-437

14. Rebeck GW, Reiter JS, Strickland DK, Hyman BT. Apolipo- protein E in sporadic Alzheimer’s disease: allelic variation and receptor interactions. Neuron 1993;11:575-580

15. Okuizumi K, Onodera 0, Tanaka H , et al. ApoE-€4 and early- onset Alzheimer’s. Nature Genet 1994;7:10-11

16. Nathan BP, Bellosta S, Sanan DlA, et al. Differential effects of apolipoproteins E3 and E4 on neuronal growth in vitro. Science

17. Armstrong M, Daly AK, Cholerton S, et al. Mutant debriso- quine hydroxylation genes in Parkinson’s disease. Lancet 1992; 3391017-1018

18. Smith CAD, Gough AC, Leigh PN, et al. Debrisoquine hydrox- ylase gene polymorphism and susceptibility to Parkinson’s dis- ease. Lancet 1992;339:1375-1377

19. Kondo 1, Kanazawa I. Debrisoquine hydroxylase and Parkin- son’s disease. Adv Neurol 1993:60:338-342

20. Kurrh MC, Kurth JH. Variant cytochrome P,,, CYP2D6 allelic frequencies in Parkinson’s disease. Am J Med Genet 1993;48: 166-168

21. Saitoh T, Xia Y, Chen X, et al. The CYP2D6B mutant allele is overrepresented in the Lewy body variant of Alzheimer’s dis- ease. Ann Neurol 1995;37:110--112

22. Hansen L, Salmon D, Galasko D, et al. The Lewy body variant of Alzheimer’s disease: a clinical and pathological entity. Neurol-

23. McKhann G, Drachman D, Folstein M, et al. Clinical diagnosis of Alzheimer’s disease: report of‘ the NINCDS-ADRDA work groups under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 1984; 34939-944

24. Khachaturian 2s. Diagnosis of Alzheimer’s disease. Arch Neu- rol 1985;42: 1097-1 105

25. Terry RD, Peck A, DeTeresa R, et al. Some morphometric aspects of the brain in senile dementia of the Alzheimer type. Ann Neurol 1981 ;lo: 184- 192

1994;264:850-852

ogy 1990;40:1-8

26. Hansen LA, Masliah E, Quijada FS, Rexin D. Entorhinal neuro- fibrillary tangles in Alzheimer disease with Lewy bodies. Neu- rosci Lett 1991;129:269-272

27. Alford MF, Masliah E, Hansen LA, Terry RD. A simple dot- immunobinding assay for quantification immunoreactivity in hu- man brain. J Histochem Cytochem 1994;42:283-287

28. Wiedenmann B, Franke WW. Identification and localization of synaptophysin, an integral membrane glycoprotein of M, 38,000 characteristic of presynaptic vesicles. Cell 1985;41:1017- 1028

29. Walaas S1, Jahn R, Greengard P. Quantitation of nerve terminal populations: synaptic vesicle associated proteins on markers of synaptic density in the rat neostriatum. Synapse 1988;2:5 16- 520

30. Masliah E, Hansen L, Albright T , et al. Immunoelectron micro- scopic study of synaptic pathology in Alzheimer’s disease. Actd Neuropathol (Berl) 1991;81:428-433

31. Fonnum FJ. A rapid radiochemical method for the determina- tion of choline acetyltransferase. J Neurochem 1975;24:407- 407

32. Whirehouse PJ, Clark AW, Price DL. Alzheimer’s disease (AD): loss of cholinergic neurons in the nucleus basalis. J Neuropathol Exp Neurol 198 1 $0: 32 3-329

33. Price DL, Sisodia SS. Cellular and molecular biology of Alzhei- mer’s disease and animal models. Annu Rev Med 1994;45:435- 446

34. Harrington CR, Louwagie J, Rossau R, et al. Influence of apoli- poprotein E genotype on senile dementia of the Alzheimer and Lewy body types. Significance for etiological theories of Alzhei- mer’s disease. Am J Pathol 1994;145:1472-1484

35. Hansen LA, Galasko D, Samuel W, et al. Apolipoprotein-E €4 is associated with increased neurofibrillary pathology in the Lewy body variant of Alzheimer disease. Neurosci Lett 1994; 182:63-65

36. Evans DAP, Mahgoub A, Sloan TP, et al. A family and popula- tion study of the genetic polymorphism of debrisoquine oxida- tion in a white Bricish population. J Med Genet 1980;17:102- 105

37. Tyndale RF. Sunahara R, Inaba T, et al. Neuronal cytochrome P4501ID1 (debrisoquineisparreine-type): potent inhibition of activity by ( -)-cocaine and nucleotide sequence identity to hu- man hepatic P450 gene CYP2D6. Mol Pharmacol 1991;40:

38. Adams MD, Dubnick M, Kerlavage AR, et al. Sequence identi- fication of 2,375 human brain genes. Nature 1992;355:632-634

39. Masliah E, Mallory M, G e N , Saitoh T. Protein kinases and growth associated proteins in plaque formation in Alzheimer’s disease. Rev Neurosci 1992;3:99-107.

40. Masliah E, Mallory M, Alford M, et al. PDGF is associated with neuronal and glial alterations of Alzheimer’s disease. Neurobiol Aging 1995;16:549-556

41. Hendriks L, van Duijn CM, Cras P, e t al. Presenile dementia and cerebral haemorrhage linked to a mutation at codon 692 of the p-amyloid precursor protein gene. Nature Genet 1992;l: 2 18-22 1

42. Masliah E, Saitoh T. Lewy bodies, Alzheimer pathology and APP mutation. Neurosci Lett 1994;180:292-293

63-68

658 Annals of Neurology Vol 38 No 4 October 1995