Printer Friendly

Genotypes and phenotypes for apolipoprotein E and alzheimer disease in the Honolulu-Asia Aging Study.

Apolipoprotein E (ApoE) [6] plays a role in the transport and redistribution of lipids, including cholesterol. Of the three common isoforms of this glycoprotein (E2, E3, and E4), E3 is by far the most common. Isoform E4 is associated with higher plasma concentrations of cholesterol, cardiovascular disease, and Alzheimer disease (AD) (1-4). The corresponding genetic alleles, which are found on chromosome 19 ([epsilon]2, [epsilon]3, [epsilon] 4), account for 99% of the genetic variance (5), and involve differences in codons 112 and 158.

The ApoE [epsilon] 4 allele has been shown to be associated with AD in African-American, European, Japanese, and Asian-ancestry populations (6-15). Increased AD risk is associated with either one or two copies of the E4 allele and has led some to advocate [epsilon] 4 genotyping for diagnostic support in early dementia or as an adjunct to the differential diagnosis of dementia (9,16,17). Other reports have suggested that ApoE2 is protective for AD (18,19). This report addresses the reliability and comparability of ApoE phenotyping and genotyping, and the implications of using these two methods for epidemiological studies, population screening, and patient care. There have been previous reports of inconsistencies between ApoE genotypes and phenotypes, possibly related to methodological or other problems (20-26). One possibility is that posttranslational protein glycation might influence accurate characterization of the phenotype. Potential inconsistencies between ApoE genotyping and phenotyping are important considerations in assessing the utility of genetic testing for ApoE in epidemiological studies and in patient care. Although several statements about the usefulness of ApoE testing have been made, these include no mention of the possibility of laboratory errors (27-32). This report aims to answer the following questions: (a) Are results of phenotyping and genotyping of ApoE the same? (b) Is the association of ApoE with AD different for phenotype and genotype? (c) What are the consequences of any difference in ApoE typing for specific applications (research, screening, patient care)?

Data and Methods


The Honolulu-Asia Aging Study (HAAS) cohort consists of Japanese-American men born from 1900 through 1919 and living on Oahu, Hawaii when the study began. Of men followed since 1965, 3734 received evaluations for cognitive function and dementia during the 1991-1993 examination cycle (33-35). Subjects were fully informed regarding participation in the study and provided informed consent. Interviews and testing were carried out by trained interviewers, either in the research center or at the subjects' homes or nursing homes, in the subjects' preferred language [Japanese (12%) or English].


Dementia was assessed using Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R) diagnostic criteria (36) by a panel consisting of the study neurologist and at least two other physicians with expertise in geriatric medicine. Criteria of the National Institute of Neurological and Communicative Disorders-Alzheimer's Disease and Related Dementias Association were similarly applied for the diagnosis of probable or possible AD. A total of 105 cases included 61 individuals classified as having probable AD, 6 classified as having possible AD with no other cause apparent, and 38 classified as having possible AD judged the primary cause of the dementia, but with another contributing cause also recognized. Odd ratios (ORs) were calculated by comparing AD patients with the rest of the HAAS sample.


ApoE phenotyping and genotyping were done on plasma and buffy coat preparations separated from anticoagulated blood within 2 h of collection and then frozen and held at -70[degrees]C until thawed for this use. Phenotyping was done at the Northwest Lipid Research Laboratory, Seattle, under the direction of one of the authors (S.M.), using a modification of the method described by Kataoka et al. (37). Briefly, 10 [micro]L of plasma sample was incubated with dithiothreitol (0.7 g/L) and Tween 20 (2.5 mL/L) for 15 min in preparation for monodimensional isoelectric focusing. Flatbed gels of 5% polyacrylamide containing ampholytes (pH range 4.0-8.0) and 3 mol/L urea were prepared, placed in an LKB Electrophor electrofocusing unit, and prefocused by applying constant power of 20 W for 15 min, with 1 mol/L NaOH and 1 mol/L phosphoric acid as cathode and anode buffers, respectively. Plasma samples adsorbed onto filter strips were applied on the gel ~15 mm from the cathode. Known samples of the common phenotypes were included with unknowns in each gel. A constant power of 20 W for 30 min was applied to allow samples to enter the gel. Sample wicks were removed, and sample proteins were focused by further power application for 90 min. Protein fractions in the gel were eluted and transferred to a nitrocellulose filter by overnight passive capillary adsorption. The nitrocellulose filter containing the protein bands was incubated for 60 min in Tris-buffered saline (0.25 mol/L NaCl, 0.03 mol/L Tris-HCl, pH 8.0) containing 20 g/L nonfat milk. The filter was exposed for 60 min to a monospecific goat anti-human ApoE antibody (Inkstar), washed in Tris-buffered saline, and reacted with a second antibody, goat anti-rabbit conjugated with horseradish peroxidase (Kirkegaard & Perry Laboratories). After several washings, the banding patterns on the filter were visualized using an ECL chemiluminescence system (Amersham Pharmacia Biotech), and a permanent record of the results was made by exposing an autoradiographic film (Kodak).

During the early phases of the study, the procedure was validated by running 200 samples in double-blind fashion. Confirmed samples were included as quality controls in subsequent analyses. Data entry and final types were checked regularly by a second person. Phenotypes for a panel of ~20 samples identified as nonconcordant with genotype were independently confirmed at a second laboratory. Although it may be possible to identify the ApoE5 and ApoE7 phenotype patterns on gels such as those used for this study, samples of these isoforms were not ordinarily included as known positives in gels, and neither the antibody reagents nor the methods have been evaluated to determine whether these isoforms would have been identified.

ApoE genotyping was done at Duke University under the direction of one of the authors (A.S.). Genomic DNA was extracted from peripheral blood leukocytes (buffy coat samples heavily contaminated with erythrocytes) using Puregene kits (Gentra Systems) according to the manufacturer's protocol for blood. ApoE gene amplification and typing were performed as described by Saunders et al. (7) with the exception that reactions were nonradioactive and restriction digest fragments were visualized using a fluorimager after SYBR Green staining. Efforts to minimize human error included assigning sample-specific barcodes to all buffy coat samples, aliquots, and extracted DNA and using these to track the sample through DNA extraction, PCR set up, and reading. Risk of pipetting and transferring errors were minimized by rigorous standardization of procedures for aliquoting DNA samples and loading gels, and by spacing of samples and controls (water, ApoE calibrators, molecular weight markers). Gels were read and genotypes transcribed by two persons independently, with correspondence checked after data entry.


Because ApoE2, -3, and -4 alleles and the isoforms of a pair can be considered independent of each other, our analysis is based both on allele and isoform frequencies (given two observations per person), and on gene pairs and persons (where each person is one observation and can be homo- or heterozygous). ORs were calculated by logistic regression, controlling for age and education. Confidence intervals are 95% confidence intervals, and significance testing is at the 5% level, unless otherwise stated.

Multivariate logistic regression analyses to evaluate associations of E2 and E4 ApoE alleles or isoforms with AD (Table 4) were carried out as a series of four separate models. The dependent variable was the dichotomous (present or not present) diagnosis of AD, based on National Institute of Neurological and Communicative Disorders-Alzheimer's Disease and Related Dementias Association diagnostic criteria for probable or possible AD. This included the 105 cases described above and excluded other cases of possible AD in which the most important cause of dementia was not thought to be AD. All regression models included age (in single years) and education (as single years of schooling completed) as covariates. To assess possible interactions of ApoE type with age or education, all possible two-variable products of the E2 or E4 allele or isoform with age or education were registered in models in the presence of the primary variables.


ApoE genotype and phenotype results were available for 3564 (95%) of the 3734 men who participated at the 1991-1993 examination. Data are presented based on individuals (Table 1) and on isoforms or alleles (two per person, n = 7128; Table 2). As expected, the most common ApoE type was 3-3. Distinguishing the probability of a phenotype-genotype nonconcordance according to isoforms, alleles, or individuals is important both for assessing the clinical utility of the two methods and for understanding the biological meaning of such nonconcordance. Of participants with genotype 3-3, 90% were also phenotyped as 3-3. Of participants with phenotype 3-3, 94% were also genotyped as 3-3. The discrepancies were largely attributable to an excess of the E2 isoform (Table 2A). When calculated as a conditional probability, 43% of ApoE phenotyped as 2 were genotyped as 3 (Table 2B). The probability of an ApoE2 phenotype among persons genotyped as 3 was 4% (Table 2C).

Frequencies of AD for each ApoE type are shown in Table 3. Of 17 men with the 4-4 phenotype and 16 men with the 4-4 genotype, 3 were found to have AD, compared with a computed expected number of 0.5 AD cases. Table 4 shows the association with AD of ApoE genotype and phenotype using logistic regression adjusted for age and education. The OR for ApoE4 heterozygotes and AD was similar for genotype and phenotype and was approximately a twofold excess risk (P <0.01) over non-ApoE4 individuals. There was similarity in AD risk association for genotype and phenotype comparisons of ApoE4 homozygotes with an OR of ~14 (P <0.001). For ApoE2, a moderate protective effect was evident with an OR of ~0.5 (P >0.09), similar for phenotype and genotype. There were insufficient 2-2 cases to determine whether a dose-response relationship existed with homozygous ApoE2 (expected numbers [approximately equal to] 0.5; found, none). In all models, the OR estimating the increasing occurrence of AD with a single year of age was 1.24, whereas the OR of AD for a single year of schooling completed was 0.96.

The increased frequency of AD among ApoE4 homozygotes (OR = 14.7) was higher than expected from an independent heterozygote ApoE4 effect [[(OR = 2.0).sup.2] = 4.0]; the square of the upper confidence limit of the OR of [epsilon] 4 heterozygotes ([3.2.sup.2] = 10.2) is less than the point estimate of the [epsilon] homozygotes (14.7), and conversely, the root of the lower confidence limit of the OR of [epsilon] 4 homozygotes ([square root of] 4.0 = 2.0) is equal to the point estimate of the [epsilon] 4 heterozygotes (2.0). This suggests that homozygosity is associated with a higher risk of AD than two independent alleles (at a 5% significance level). This "recessive" character of the ApoE4 gene risk was equally apparent when isoform typing was used.

The association of ApoE (both 2 and 4, both homo- and heterozygous) with AD appeared to be almost independent of age. The OR without age adjustment was 12.2 (3.8-39.0) for E4 homozygotes and 12.0 (3.7-38.8) for [epsilon] 4 homozygotes. (For the ORs with adjustment, see Table 4.) The independence of the effects of age and ApoE4 was further supported by the introduction of an interaction term, which was nonsignificant (P <0.38 for E4 homozygotes and P <0.36 for [epsilon] 4 homozygotes). This finding points to at least partially different mechanisms underlying the influences of age and ApoE on the development of AD, and suggests that the increased occurrence of AD with ApoE4 positivity is generally constant across the age range in this population (71-93 years).

Because of this age independence, the strength of the association of AD with ApoE4 can be expressed in a "corresponding" age difference (Table 4). Estimated from logistic regression, ApoE4 heterozygosity confers on an individual a probability of having AD approximating that of an otherwise similar 3-3 individual who is 3 years older. Similarly, an ApoE4 homozygous person has a probability of AD similar to that of a 3-3 person who is 12 years older.

Despite the considerable and significant OR (-14) for 4-homozygosity and AD, the effect of age in predicting AD is quite dominant. It is useful, therefore, to compare the number of additional cases of AD one might predict using ApoE4 status in addition to age and education, compared with age and education only. The classification table of the logistic regression, presented in the form of a so-called ROC curve (38) in Fig. 1, shows the trade-off of sensitivity against specificity. The two ROC curves are almost identical. In our study population, given a "fixed" sensitivity of 66.7%, the specificity in predicting AD increased from 83.0% to 83.3% when ApoE4 status was included in the logistic model in addition to age and education. Thus, in population screening, ApoE4 status contributes little to predicting dementia, if age and education are known.



There is growing interest in identifying and using genetic polymorphisms for disease risk, including risk for AD, in the general population. This interest has been particularly strong for ApoE4 in relation to AD. Although there has been some endorsement of the use of genotyping for dementia patients, population screening of asymptomatic individuals has not been recommended (27). It is, however, important to distinguish among use for epidemiological research, patient care, and population screening.


An important aspect of epidemiological research is the discovery and modeling of risk associations. An important conclusion of our study is that the phenotypes and genotypes of ApoE2 and ApoE4 are associated with similar magnitude with AD: ApoE2 appears to be moderately protective and ApoE4 is a risk factor. The robustness of these relationships supports the use of phenotyping, especially when DNA specimens are not available, as is the case in many large longitudinal population-based studies. Confirmatory observations might also support the use of specimens collected many years previously in retrospective cohort studies. Such investigations might add to existing knowledge of incidence of AD and whether differential mortality related to the presence of ApoE4 might distort subsequent relationships (39, 40). However, for studies of ApoE2, considerable differences in persons identified as "at risk" by genotyping or phenotyping must be kept in mind. It has yet to be established whether the high frequency of ApoE2 phenotype-genotype nonconcordance we observed is specific to the Japanese-American subjects in the HAAS or occurs in other populations as well.


For specificity of characterization, the standard for patient care has become the genotype, and this procedure is recommended, especially because many laboratories are offering this determination. In addition, in the future it is likely that there will be other polymorphisms best studied with DNA, and these will completed in conjunction with ApoE determinations. Although the association of ApoE4 with a risk for AD is similar for genotype and phenotype, the remaining individual nonconcordance is not to be neglected, and--as is usual in clinical situations--the clinician must consider the gained information in the context of all available information bearing on the diagnosis.


Population screening for AD would involve persons without any known predisposition for this condition. The expected prevalence of AD would be mainly dependent on the age composition of the population and would be rather low. In our population sample of men 71-93 years of age, screening for AD by means of ApoE would not have been fruitful, except for identification of the limited number of ApoE4-4 persons. In the general population, use of ApoE status in addition to the freely available information on age and education only marginally improves prediction, as demonstrated by the ROC curves. Furthermore, screening is at present--without a fairly effective therapy for AD--not justified.


Although major discrepancies between ApoE phenotypes and genotypes have been reported previously, the extent of nonconcordance has varied dramatically. The initial descriptions of phenotype-genotype nonconcordance were focused on their associations with diabetes and hyperglycemia, leading to speculation that a glucose-driven posttranslational modification of the protein might lead to alteration in the band pattern on isoelectric focusing and ultimately to errors in typing (23-26). For the most part, these observations have not been confirmed, and the glycation of the molecule appears an unlikely cause for substantial nonconcordance. A second possibility is clerical or laboratory error, occurring by chance (22). A third possibility, as yet not demonstrated to be an important cause of phenotype-genotype nonconcordance, is that there are rare genetic polymorphisms associated with ApoE2 that affect protein expression and/or alter the primary structure of the gene product. Some reports of "rare" polymorphisms have appeared, notably in the Japanese population (ApoE-[epsilon]7, ApoE-E1, ApoE-E5, ApoE-E7) (41-43), but unusual mutations have also been reported that cause an ApoE-[member] 4/E4 discrepancy with usual assessment methods (44). Finally, "normal" variations in test reproducibility may explain part of the nonconcordance; these may become visible especially in studies with large numbers of subjects. In fact, there is a point to be made for routinely genotyping DNA in duplicate, as soon as "cheaper" techniques allow the extra effort.


Our finding of the independence of risk for AD associated with age and ApoE (in a group of men over age 70) may reflect the presence of AD and not its progress. Independence was also found by others (19, 45). However, some dependencies have been found in clinical case-control studies with a wider age range (11, 46). It has also been reported that the influence of ApoE4 may diminish rather dramatically after age 80 (47). ApoE4 does not seem to be related to the progress of AD (48-50), which is in contrast to the association of ApoE with the progress of cognitive impairment in similar age groups (51-53).

The choice of genotyping or phenotyping should also be informed by the other correlates of ApoE type, especially as related to lipid metabolism. Before identification of the relationship between ApoE4 and AD, most work with ApoE type related to cardiovascular disease and lipid concentrations. Because this research goes back several years and because of limitations in technology, ApoE serum proteins were measured either directly or in the VLDL subfraction (22). As genotyping became available, there were some attempts to compare results. In one such comparison, serum triglyceride concentrations were higher in the phenotyped ApoE2 subgroup compared with the genotyped designation (22). The reasons for this discrepancy were unclear. The possibility that unrecognized genotypes or different intermediary processes might affect physiologic and metabolic measurements has been suggested (22). Such phenomena could vary with ethnicity, comorbidity, or other factors.

The observation of Lahoz et al. (22) that many ApoE phenogenotype-genotype discrepancies could be attributed to errors in labeling or handling must not be minimized. When ApoE typing is important for the care or diagnosis of an individual, only very low laboratory error is tolerable. When typing is done as part of epidemiologic research, the identification of true phenotype-genotype differences may well lead to a better understanding of several illnesses, including AD, atherosclerosis, and diabetes. The importance of reliable typing for research or clinical purposes is obvious. Although duplicate testing has rarely been done for genetic assays, this or some other method for detecting test inconsistencies could ultimately be an important quality-control strategy for laboratories conducting such tests.

In conclusion, the HAAS provided a valuable opportunity to evaluate various aspects of the use of ApoE phenotyping and genotyping in a population-based epidemiological study. In this specific population, either genotyping or phenotyping was adequate for determining associations of ApoE4 with AD. Genotype and phenotype showed similar associations, although mild discrepancies for ApoE4 and substantial discrepancies for ApoE2 produced some differences in which individuals were identified. The effect of age and ApoE status on AD presence appeared to rather independent. Our findings indicate that the use of ApoE status for screening of the general population would provide only minimal improvement in prediction of cases over the use of education and age only.

Supported by National Institute on Aging Contract NO1AG-4-2149 and National Heart, Lung, and Blood Institute Contract N01-HC-05102 from the NIH.


(1.) Braeckman L, De Bacquer D, Rosseneu M, De Backer G. Apolipoprotein E polymorphism in middle-aged Belgian men: phenotype distribution and relation to serum lipids and lipoproteins. Atherosclerosis 1996;120:67-73.

(2.) Olichney JM, Sabbagh MN, Hofstetter CR, Galasko D, Grundman M, Katzman R, Thai U. The impact of apolipoprotein E4 on cause of death in Alzheimer's disease. Neurology 1997;49:76-81.

(3.) Roses AD, Saunders AM, Alberts MA, Strittmatter WJ, Schmechel D, Gorder E, Pericak-Vance MA. Apolipoprotein E E4 allele and risk of dementia [Letter]. JAMA 1995;273:374-5.

(4.) Slooter AJC, Tang MX, Duijn CMV, Stern Y, Ott A, Bell K, et al. Apolipoprotein E [epsilon] 4 and the risk of dementia with stroke. JAMA 1997;277:818-21.

(5.) Hansen PS, Gerdes LU, Klausen IC, Gregersen N, Faergeman O. Genotyping compared with protein phenotyping of the common apolipoprotein E polymorphism. Clin Chim Acta 1994;224:131-7.

(6.) Strittmatter WJ, Weisgraber KH, Goedert M, Saunders AM, Huang D, Corder EH, et al. Hypothesis: microtubule instability and paired helical filament formation in the Alzheimer disease brain are related to apolipoprotein E genotype. Exp Neurol 1994;125:163-71.

(7.) Saunders AM, Strittmatter WJ, Schmechel D, George-Hyslop PH, Pericak-Vance MA, Joo SH, et al. Association of apolipoprotein E allele epsilon 4 with late-onset familial and sporadic Alzheimer's disease. Neurology 1993;43:1467-72.

(8.) Poirier J, Davignon J, Bouthillier D, Kogan S, Bertrand P, Gauthier S. Apolipoprotein E polymorphism and Alzheimer's disease. Lancet 1993;342:697-9.

(9.) Myers RH, Schaefer EJ, Wilson PW, D'Agostino R, Ordovas JM, Espino A, et al. Apolipoprotein E epsilon4 association with dementia in a population-based study: The Framingham Study. Neurology 1996;46:673-7.

(10.) Kukull WA, Schellenberg GD, Bowen JD, McCormick WC, Yu CE, Teri L, et al. Apolipoprotein E in Alzheimer's disease risk and case detection: a case-control study. J Clin Epidemiol 1996;49: 1143-8.

(11.) Bickeboller H, Campion D, Brice A, Amouyel P, Hannequin D, Didierjean O, et al. Apolipoprotein E and Alzheimer disease: genotype-specific risks by age and sex. Am J Hum Genet 1997; 60:439-46.

(12.) Nunomura A, Chiba S, Eto M, Saito M, Makino I, Miyagishi T. Apolipoprotein E polymorphism and susceptibility to early- and late-onset sporadic Alzheimer's disease in Hokkaido, the northern part of Japan. Neurosci Lett 1996;206:17-20.

(13.) Hendrie HC, Hall KS, Hui S, Unverzagt FW, Yu CE, Lahiri DK, et al. Apolipoprotein E genotypes and Alzheimer's disease in community study of elderly African Americans. Ann Neurol 1995;37:118-20.

(14.) Noguchi S, Murakami K, Yamada N. Apolipoprotein E genotype and Alzheimer's disease [Letter]. Lancet 1993;342:737.

(15.) Hong CJ, Liu TY, Liu HC, Wang SJ, Fuh JL, Chi CW, et al. Epsilon 4 allele of apolipoprotein E increases risk of Alzheimer's disease in a Chinese population. Neurology 1996;46:1749-51.

(16.) Saunders AM, Hulette O, Welsh-Bohmer KA, Schmechel DE, Crain B, Burke JR, et al. Specificity, sensitivity, and predictive value of apolipoprotein-E genotyping for sporadic Alzheimer's disease. Lancet 1996;348:90-3.

(17.) Nalbantoglu J, Gilfix BM, Bertrand P, Robitaille Y, Gauthier S, Rosenblatt DS, Poirier J. Predictive value of apolipoprotein E genotyping in Alzheimer's disease: results of an autopsy series and an analysis of several combined studies. Ann Neurol 1994; 36:889-95.

(18.) Lippa CF, Smith TW, Saunders AM, Hulette C, Pulaski-Salo D, Roses AD. Apolipoprotein E-epsilon 2 and Alzheimer's disease: genotype influences pathologic phenotype. Neurology 1997;48: 515-9.

(19.) Farrer LA, Cupples LA, Haines JL, Hyman B, Kukull WA, Mayeux R, et al. Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease-a meta analysis. JAMA 1997;278:1349-56.

(20.) Kim JH, Lee EJ, Kwon OH. Apolipoprotein E genotyping and phenotyping in type II diabetes mellitus patients with hypertriglyceridemia. Clin Biochem 1997;30:47-52.

(21.) Brouwer DA, Vam Doormaal JJ, Muskiet FA. Clinical chemistry of common apolipoprotein E isoforms. J Chromatogr B Biomed Appl 1996;678:23-41.

(22.) Lahoz C, Osgood D, Wilson PW, Schaefer EJ, Ordovas JM. Frequency of phenotype-genotype discrepancies at the apolipoprotein E locus in a large population study. Clin Chem 1996;42: 1817-23.

(23.) Mailly F, Moll P, Kottke BA, Kamboh MI, Humphries SE, Ferrell RE. Estimation of the frequency of isoform-genotype discrepancies at the apolipoprotein E locus in heterozygotes for the isoforms. Genet Epidemiol 1992;9:239-48.

(24.) Liinamaa MJ, Kervinen K, Hannuksela ML, Kesaniemi YA, Savolainen MJ. Effect of apolipoprotein E phenotype on plasma lipids and lipoproteins in alcohol abusers. Alcohol Clin Exp Res 1997; 21:606-12.

(25.) Wenham PR, Sedky A, Spooner RJ. Apolipoprotein E phenotyping: a word of caution. Ann Clin Biochem 1991;28:599-605.

(26.) Snowden C, Houlston RS, Arif MH, Laker MF, Humphries SE, Alberti KG. Disparity between apolipoprotein E phenotypes and genotypes (as determined by polymerase chain reaction and oligonucleotide probes) in patients with non-insulin-dependent diabetes mellitus. Clin Chim Acta 1991;196:49-57.

(27.) National Institute on Aging/Alzheimer's Association Working Group. Apolipoprotein E genotyping in Alzheimer's disease. Lancet 1996;347:1091-5.

(28.) Post SG, Whitehouse PJ, Binstock RH, Bird TD, Eckert SK, Farrer LA, et al. The clinical introduction of genetic testing for Alzheimer disease-an ethical perspective. JAMA 1997;277:832-6.

(29.) Mayeux R, Schupf N. Apolipoprotein E and Alzheimer's disease; the implications of progress in molecular medicine. Am J Public Health 1995;85:1280-4.

(30.) Annas GJ. Genetic prophecy and genetic privacy-can we prevent the dream from becoming a nightmare? Am J Public Health 1995;85:1196-7.

(31.) Hyman BT. Apolipoprotein E genotype: utility in clinical practice in Alzheimer's disease. J Am Geriatr Soc 1996;44:1469-71.

(32.) Pericak-Vance MA, Bass MP, Yamaoka LH, Gaskell PC, Scott WK, Terwedow HA, et al. Complete genomic screen in late-onset familial Alzheimer disease. JAMA 1997;278:1237-41.

(33.) Rodriguez BL, Curb JD, Burchfiel CM, Huang B, Sharp DS, Lu GY, et al. Impaired glucose tolerance, diabetes, and cardiovascular disease risk factor profiles in the elderly. The Honolulu Heart Program. Diabetes Care 1996;19:587-90.

(34.) Kagan A, Harris BR, Winkelstein W, Johnson KG, Kato H, Syme SL, et al. Epidemiological studies of coronary heart disease and stroke in Japanese men living in Japan, Hawaii and California: demographic, physical, dietary and biochemical characteristics. J Chronic Dis 1974;27:345-64.

(35.) White L, Petrovitch H, Ross GW, Masaki KH, Abbott RD, Teng EL, et al. Prevalence of dementia in older Japanese-American men in Hawaii: the Honolulu Asia Aging Study. JAMA 1996;276:955-60.

(36.) American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 3rd ed., revised. Washington, DC: American Psychiatric Association, 1987.

(37.) Kataoka S, Paidi M, Howard BV. Simplified isoelectric focusing/ immunoblotting determination of apoprotein E phenotype. Clin Chem 1994;40:11-3.

(38.) Kardaun JWPF, Kardaun OJWF. Comparative diagnostic performance of three radiological procedures for the detection of lumbar disk herniation. Methods Inf Med 1990;29:12-22.

(39.) Zerba KE, Ferrell RE, Sing CF. Genotype-environment interaction: apolipoprotein E (ApoE) gene effects and age as an index of time and spatial context in the human. Genetics 1996;143:463-78.

(40.) Tilvis RS, Strandberg TE, Juva K. Apolipoprotein E phenotypes, dementia and mortality in a prospective population sample. J Am Geriatr Soc 1998;46:712-5.

(41.) Yanagi K, Yamashita, S, Hiraoka H, Ishigami M, Kihara S, Hirano K, et al. Increased serum remnant lipoproteins in patients with apolipoprotein E7 (apo E Suita). Atherosclerosis 1997;131:49-58.

(42.) Ando M, Sasaki J, Hua H, Matsunaga A, Uchida K, Jou K, et al. A novel 18-amino acid deletion in apolipoprotein E associated with lipoprotein glomerulopathy. Kidney Int 1999;56:1317-23.

(43.) Imura T, Kimura H, Murata S, Yamaguchi T, Kanou A, Nakashima M, at al. [Detection of apolipoproteins E5 and E7 by a widely used commercial ApoE IFE kit-evidence provided by genotyping]. Rinsho Byori 1998;46:289-93.

(44.) Kang AK, Jenkins DJ, Wolever TM, Huff MW, Maguire GF, Connelly PW, Hegele RA. Apolipoprotein E R112; R251G: a carboxyterminal variant found in patients with hyperlipidemia and coronary heart disease. Mutat Res 1997;382:57-65.

(45.) Slooter AJ, Cruts M, Kalmijn S, Hofman A, Breteler MM, Van Broeckhoven C, Van Duijn CM. Risk estimates of dementia by apolipoprotein E genotypes from a population based incidence study: the Rotterdam Study. Arch Neurol 1998;55:964-8.

(46.) Frisoni GB, Manfredi M, Geroldi C, Binetti G, Zanetti O, Bianchetti A, Trabucchi M. The prevalence of apoE-epsilon4 in Alzheimer's disease is age dependent. J Neurol Neurosurg Psychiatry 1998; 65:103-6.

(47.) Breitner JC, Wyse BW, Anthony JC, Welsh-Bohmer KA, Steffens DC, Norton MC, et al. APOE-epsilon4 count predicts age when prevalence of AD increases, then declines: the Cache County Study. Neurology 1999;53:321-31.

(48.) Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC Jr, Rimmler JB, et al. Apolipoprotein E, survival in Alzheimer's disease patients, and the competing risks of death and Alzheimer's disease. Neurology 1995;45:1323-8.

(49.) Dal Forno G, Rasmusson DX, Brandt J, Carson KA, Brookmeyer R, Troncosp J, Kawas CH. Apolipoprotein E genotype and rate of decline in probable Alzheimer's disease. Arch Neurol 1996;53: 345-50.

(50.) Growdon JH, Locascio JJ, Corkin S, Gomez-Isla T, Hyman BT. Apolipoprotein E genotype does not influence rates of cognitive decline in Alzheimer's disease. Neurology 1996;47:444-8.

(51.) Henderson AS, Easteal S, Jorm AF, Mackinnon AJ, Korten AE, Christensen H, et al. Apolipoprotein E allele epsilon 4, dementia, and cognitive decline in a population sample Lancet 1995;346: 1387-90.

(52.) Bondi MW, Salmon DP, Monsch AU, Galasko D, Butters N, Klauber MR, et al. Episodic memory changes are associated with the APOE-epsilon 4 allele in nondemented older adults. Neurology 1995;45:2203-6.

(53.) Helkala EL, Koivisto K, Hanninen T, Vanhanen M, Kervinen K, Kuusisto J, et al. The association of apolipoprotein E polymorphism with memory: a population based study. Neurosci Lett 1995;191:141-4.


[1] Archa B.V., 3039 HK Rotterdam, The Netherlands.

[2] Kuakini Medical Center and the Pacific Health Research Institute, Honolulu, Hawaii 96813.

[3] University of Washington School of Medicine, Northwest Lipid Laboratories, Seattle, WA 98103.

[4] Joseph and Kathleen Bryan Alzheimer's Disease Research Center, Duke University School of Medicine, Durham, NC 27710.

[5] National Institute on Aging, Bethesda, MD 20892.

[6] Nonstandard abbreviations: ApoE, apolipoprotein E; AD, Alzheimer disease; HAAS, Honolulu-Asia Aging Study; and OR, odds ratio.

* Address correspondence to this author at: Pacific Health Research Institute, Suite 306, 846 South Hotel St., Honolulu, HI 96813. Fax 808-524-4315; e-mail

Received in revised form July 7, 2000; accepted July 10, 2000.
Table 1. Phenotype vs genotype of 2-, 3-, 4-allele/isoform
of ApoE, per person.


Phenotype 2-2 2-3 2-4 3-3 3-4 4-4 Total

2-2 5 7 4 2 0 0 18
2-3 2 261 2 227 5 0 497
2-4 1 2 28 0 1 0 32
3-3 0 26 3 2332 107 0 2468
3-4 0 1 0 32 497 2 532
4-4 0 0 0 0 3 14 17
Total 8 297 37 2593 613 16 3564

Table 2. Phenotype vs genotype of ApoE, per gen (twice
the number of persons).

A. Cross table


Phenotype 2 3 4 Total

2 317 244 4 565
3 32 5814 119 5965
4 1 38 559 598
Total 350 6096 682 7128

B. Conditional probability of genotype, given phenotype (to
be read rowwise)


Phenotype 2 3 4 Total

2 0.56 0.43 0.007 1.0
3 0.005 0.97 0.02 1.0
4 0.002 0.06 0.94 1.0

C. Conditional probability of phenotype, given genotype (to
be read columnwise)


Phenotype 2 3 4

2 0.91 0.04 0.006
3 0.09 0.95 0.17
4 0.003 0.006 0.82
Total 1.0 1.0 1.0

Table 3. AD vs genotype and phenotype of ApoE.

 Phenotype Genotype
Isoform No AD AD %AD No AD AD %AD

2-2 18 0 0 8 0 0
2-3 490 7 1 291 6 2
2-4 30 2 6 36 1 3
3-3 2397 71 3 2526 67 3
3-4 511 21 4 586 27 4
4-4 13 4 24 12 4 25
Total 3459 105 3 3459 105 3

Table 4. Association of ApoE alleles/isoforms with AD.

 Confidence interval

 OR (a) CI difference (b)

 2 homozygous -- (c)
 2 heterozygous 0.5 0.3-1.1 -2.88
 4 heterozygous 1.9 1.1-3.0 2.90
 4 homozygous 14.2 3.9-51.6 12.24

 2 homozygous -- (c)
 2 heterozygous 0.6 0.3-1.4 -2.16
 4 heterozygous 2.0 1.3-3.2 3.23
 4 homozygous 14.7 4.0-53.6 12.36

(a) OR calculated by logistic regression, adjusting for age and

(b) Years difference: the risk for AD corresponds to the risk of
a ApoE 3-3 person who is x years older.

(c) No occurrences of 2-2 allele/isoform in AD group; OR
considered "low" and confidence interval "wide".
COPYRIGHT 2000 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2000 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Molecular Diagnostics and Genetics
Author:Kardaun, Jan W.P.F.; White, Lon; Resnick, Helaine E.; Petrovitch, Helen; Marcovina, Santica M.; Saun
Publication:Clinical Chemistry
Date:Oct 1, 2000
Previous Article:Rapid single-tube screening of the C282Y hemochromatosis mutation by real-time multiplex allele-specific PCR without fluorescent probes.
Next Article:Evaluation of the performance of a p53 sequencing microarray chip using 140 previously sequenced bladder tumor samples.

Related Articles
Four Alzheimer's susceptibility genes discovered.
An unusual case of severe hypertriglyceridemia and splenomegaly.
Increased serum lipoprotein(a) concentrations and low molecular weight phenotypes of apolipoprotein(a) are associated with symptomatic peripheral...
Apolipoprotein E and transferrin genotyping by ligation detection reaction and universal array.
Apolipoprotein E genotyping by capillary electrophoretic analysis of restriction fragments.
Genotype-phenotype correlations (II): assessing the influence of sequence variants on the clinical phenotype in multifactorial disorders.
Glutathione S-transferase PI *C allelic variant increases susceptibility for late-onset Alzheimer disease: association study and relationship with...
Apolipoprotein E in cerebrospinal fluid: relation to phenotype and plasma apolipoprotein E concentrations.
Detection of prevalent genetic alterations predisposing to hemochromatosis and other common human diseases.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters