Apo Phenotypes Research Articles

Relationship of apolipoprotein E phenotypes to serum lipid and lipoprotein levels in Japanese schoolchildren.

The distribution of apolipoprotein (apo) E phenotypes was investigated in Japanese schoolchildren and an attempt was made to determine whether the apoE phenotypes influence the serum levels of lipid and lipoprotein. The subjects were 289 children from a junior high school (145 M, 144 F, aged 12-13 y). The E3/3 phenotype was demonstrated in 69.9%, E4/3 in 18.7%, E2/3 in 9.0%, E4/2 in 1.4%, E4/4 in 0.7% and E5/4 in 0.3%. E2/2 was not detected. Serum levels of total cholesterol and low-density lipoprotein-cholesterol were progressively lower in the phenotypes E4/3, E3/3 and E2/3. The serum level of high-density lipoprotein-cholesterol showed a progressive increase in the same order. In conclusion, Japanese schoolchildren with E4/3 already had an atherogenic serum lipid profile.

Prevalence of Dementia ofAlzheimer Type and Apolipoprotein EPhenotypes in Aged Patients withDown’s Syndrome

We investigated the exact prevalence of dementia of Alzheimer type (DAT) and apolipoprotein E (ApoE) phenotypes in 106 Japanese Down’s syndrome (DS) patients. Among these patients 16 were diagnosed as having DAT. The prevalence of DAT was 0% in the 30- to 39-year-old group, 16% in the 40- to 49-year-old group, and 38% in those over 50 years old. The frequency of the σ4 allele in DS patients with DAT was 18.8%, which was considerably higher than that of nondemented DS patients (4.5%) and Japanese nondemented controls (6.7%). Especially, the frequency of the σ4 allele in DS patients who developed DAT under 50 years was significantly higher (28.6%). DS patients certainly develop DAT at earlier ages but the prevalence of DAT in each group of patients was lower than previously recognized. It is very likely that the ApoE σ4 is a risk factor for DAT even in DS patients with a genetic predisposition to Alzheimer’s disease.

Apolipoprotein(a) phenotypes as genetic markers of coronary atherosclerosis severity

We investigated Lp(a) levels and apo(a) polymorphism in relation to the severity of coronary artery disease, expressed both by the number of coronary arteries stenosed and three different coronary scoring systems. In a sample of 267 patients with coronary artery disease, a Mono-, Bi- or Multi-vessel coronary stenosis was documented by angiography. Twenty-five apo(a) isoforms were detected by a high resolution phenotyping method. Lp(a) levels did not show any differences among subgroups of patients. Both the percentage of apo(a) isoforms of low molecular weight (<655 kDa) ( P=0.00015) and the percentage of subjects with at least one apo(a) isoform of low molecular weight ( P=0.00027) were significantly correlated with increasing number of coronary vessels stenosed. In multivariate analysis, only apo(a) isoforms of low molecular weight were predictors of coronary atherosclerosis severity, when we used as the dependent variable both the `1-2-multi-vessels' categorization ( P=0.000067) and the Gensini ( P=0.008767), or Green Lane ( P=0.000001) or Dahlen ( P=0.000102) coronary scoring system. Our data show that apo(a) isoforms of low molecular weight are associated with a greater severity of coronary atherosclerosis. If these data are confirmed by prospective studies, apo(a) phenotypes might be used as genetic markers of a greater severity of coronary atherosclerotic lesions.

Lack of interaction of apolipoprotein E phenotype with the lipoprotein response to lovastatin or gemfibrozil in patients with primary hypercholesterolemia

The magnitude of serum lipid changes in response to hypolipidemic drugs varies considerably between individuals. These differences may be due to interactions between genetic and environmental factors that effect drug bioavailability or the capacity of the lipid-regulating enzyme and receptor targets to be affected. The apolipoprotein E (apoE) gene locus has been examined in this regard, but reports are conflicting on the effect of its variability on the response to hypolipidemic drugs. We investigated the effect of apoE polymorphism on the serum lipid response to the hepatic hydroxymethyl glutaryl coenzyme A (HMG CoA) reductase inhibitor lovastatin and the fibric acid derivative gemfibrozil. Lipoprotein changes were assessed after 12 weeks of therapy in 106 patients with primary hypercholesterolemia and combined hyperlipidemia treated with lovastastin and in 63 given gemfibrozil therapy. No significant effect of the apoE phenotypes E3/2, E3/3, or E4/3 on the heterogeneity of lipid responses to either drug was found.

APOLIPOPROTEIN E PHENOTYPE DOES NOT PREDICT RISK OF RENAL IMPAIRMENT AFTER CABG SURGERY

Introduction. Acute renal failure (ARF) following cardiac surgery is associated with a mortality rate ranging from 24% to 70%. While studies have identified preoperative risk factors for ARF [1], a genetic basis for the development of renal dysfunction after coronary artery bypass (CABG) surgery has not been previously investigated. Certain apolipoprotein E (ApoE) phenotypes have been associated with accelerated atherosclerosis (ApoE4) [2] and end-stage renal disease (ApoE2) [3]. We examined the hypothesis that patients with these high risk ApoE phenotypes are predisposed to renal dysfunction after cardiac surgery. Methods. With IRB approval, perioperative data for 159 elective CABG surgical patients was gathered from the Duke Heart Center and Anesthesia Databases including: age, sex, weight, preoperative ejection fraction, New York Heart Association Functional Class, history of diabetes and hypertension, duration of cardiopulmonary bypass, and number of coronary artery grafts. Postoperative change in creatinine (DCR) was calculated from preoperative and peak postoperative serum creatinine values. Peripheral blood samples were obtained from each patient and stored at 4[degree sign]C before processing. ApoE genotypes were determined using standard methods. ApoE phenotype was designated as E2(E2/2 and 2/3), E3(E3/3), or E4(E4/4, 4/3 and 4/2) before analysis. Associations were assessed by standard t test, and Kruskal-Wallis test (ApoE phenotype vs. DCR), p<0.05 was considered significant. Results. Patient ApoE phenotypes were distributed as follows: ApoE2 - 25 (E2/2 - 4, E2/3 - 21), ApoE3 - 89 and ApoE4 - 45 (E4/4 - 2, E4/3 - 38, E4/2 - 5). No significant differences were noted between ApoE phenotype groups for the recorded variables. The association between ApoE phenotypes and DCR is presented in Table 1.Table 1Discussion. We did not find ApoE phenotypes to be related to renal dysfunction after CABG surgery. ApoE phenotype distribution in this study was similar to that in the general population. We conclude that ApoE risk factors which influence the occurrence of atherosclerosis and renal disease are not important in the development of renal dysfunction following CABG surgery.

In vitro factors affecting the concentration of gamma-LpE (γ-LpE) in human plasma

Gamma-LpE (γ-LpE), a sphingomyelin-rich lipoprotein that contains apolipoprotein (apo) E as its only protein component, has been proposed to play a role in cellular cholesterol efflux by acting, like pre-β1-LpA-I, as an initial acceptor of cell-derived cholesterol. In order to further characterize the presence of γ-LpE in human plasma, we have separated γ-LpE by two-dimensional non-denaturing polyacrylamide-gradient gel electrophoresis and detected its presence by immunoblotting with 125I-labeled polyclonal anti-apoE antibody. Five species of γ-LpE were routinely detected in human plasma, ranging in mean particle diameter from 9.5 to 16.5 nm. The largest proportion of γ-migrating apoE was associated with γ2-LpE having a diameter of 13.0 nm. Neither the amount of γ-LpE apoE (representing less than 1–2% of total plasma apoE) nor the number of γ-LpE subfractions was different in serum vs. plasma, or was affected by the presence of agents able to inhibit protein dimerization. γ-LpE subfractions were present in the plasma of patients having different apoE phenotypes (i.e., apoE 2/2, 3/3, or 4/4). Incubation of plasma at 37°C (90 min) caused a significant decrease in plasma γ-LpE (>80%) that was not dependent on LCAT or CETP activity. Storage (at –70°C) of hypertriglyceridemic but not normolipidemic plasma resulted in an increase in γ-LpE. Freezing of postprandial plasma samples, containing increased amounts of triglyceride-rich lipoproteins (TRL) enriched in apoE, also caused an increase in γ-LpE. Incubation of VLDL (d < 1.006 g/ml) with lipase resulted in the production of γ-migrating apoE. These results demonstrate that: 1) different γ-LpE subfractions exist in human plasma; 2) the amount of apoE associated with γ-LpE subfractions is dependent on in vitro conditions of plasma storage; and 3) TRL can act as a source of γ-LpE apoE in vitro.—Krimbou, L., M. Tremblay, H. Jacques, J. Davignon, and J. S. Cohn. In vitro factors affecting the concentration of gamma-LpE (γ-LpE) in human plasma. J. Lipid Res. 1998. 39: 861–872.

Relationship of apolipoprotein E phenotypes to serum lipid and lipoprotein levels in Japanese schoolchildren

The distribution of apolipoprotein (apo) E phenotypes was investigated in Japanese schoolchildren and an attempt was made to determine whether the apoE phenotypes influence the serum levels of lipid and lipoprotein. The subjects were 289 children from ajunior high school (145 M, 144 F, aged 12–13 y). TheE3/3 phenotype was demonstrated in 69.9%, E4/3 in 18.7%, E2/3 in 9.0%, E4/2 in 1.4%, E4/4 in 0.7% and E5/4 in 0.3%. E2/2 was not detected. Serum levels of total cholesterol and low‐density lipoprotein‐cholesterol were progressively lower in the phenotypes E4/3, E3/3 and E2/3. The serum level of high‐density lipoprotein‐cholesterol showed a progressive increase in the same order. In conclusion, Japanese schoolchildren with E4/3 already had an atherogenic serum lipid profile.

The Apolipoprotein E Phenotype Has a Strong Influence on Tracking of Serum Cholesterol and Lipoprotein Levels in Children: A Follow-Up Study from Birth to the Age of 11 Years

The extent to which an individual maintains his position relative to the rest of the population is called tracking. The objective of this study was to examine the effect of the apolipoprotein E (apoE) phenotype on the tracking of serum cholesterol and lipoproteins from birth to the age of 11 y. In a longitudinal follow-up study of healthy children, concentrations of total serum cholesterol and triglyceride were determined at birth (n = 193), and at the ages of 2 (n = 192), 4 (n = 192), 6 (n = 190), 9 (n = 188), and 12 mo (n = 196), and 5 (n = 162) and 11 y (n = 153). Concentrations of total HDL, HDL2, and HDL3, VLDL, and LDL cholesterol were determined at 2, 6, 9, and 12 mo (n = 36), and 5 (n = 162) and 11 y (n = 153). The apoE phenotype was determined in 151 children. The children had the following apoE phenotypes: 4 had type 4/4 and 40 type 3/4 (group apoE4), 94 had type 3/3 (group apoE3), and 11 had type 2/3 and 2 type 2/4 (group apoE2). The correlation coefficients for total cholesterol levels during childhood compared with the level at 11 y of age were: 0.03 at birth, 0.26 (p < 0.001) at 2 mo, 0.24 (p < 0.001) at 4 mo, 0.24 (p < 0.001) at 6 mo, 0.28 (p < 0.001) at 9 mo, 0.41 (p < 0.001) at 12 mo, and 0.60 (p < 0.001) at 5 y. When the children were divided into three groups according to their apoE phenotypes, these three groups had the following correlation coefficients at 4 mo, 12 mo, or 5 y of age compared with the level at the age of 11 y; group apoE2: r = 0.65 (p < 0.01), r = 0.59 (p < 0.01), and r = 0.72 (p < 0.01); group apoE3: r = 0.27 (p < 0.01), 0.43 (p < 0.001), and r = 0.64 (p < 0.001); and group apoE4: r = 0.14 (p = NS), r = 0.33 (p < 0.05), and 0.42 (p < 0.01). The apoE phenotype also strongly influenced the tracking of the LDL cholesterol levels; the correlation coefficients between 5 and 11 y of age were for group apoE2 r = 0.84 (p < 0.001), for group apoE3 r = 0.70 (p < 0.001), and for group apoE4 r = 0.37 (p < 0.05). Our results indicate that the apoE phenotype strongly influences the tracking of lipids. The children having apoE 2/3, 2/4, and 3/3 phenotypes maintained their relative cholesterol and lipoprotein levels better than the others throughout the first 11 y of age. Because the apoE phenotype strongly affects the tracking of serum cholesterol, the usefulness of cholesterol screening in predicting future cholesterol values should be analyzed, keeping the apoE phenotype in mind.

Chimpanzee lipoprotein(A): Relationship between apolipoprotein(A) isoform size and the density profile of lipoprotein(A) in animals with different heterozygous apo(A) phenotypes.

In a previous study [C. Doucet et al., J. Lipid Res 35:263-270, 1994], we have shown that plasma lipoprotein (a) [Lp(a)] levels were significantly elevated in a population of unrelated chimpanzees as compared to those in normolipidemic human subjects. Nonetheless, the inverse correlation between Lp(a) levels and apolipoprotein (a) [apo(a)] isoforms typical of man was maintained in the chimpanzee. In the present study, we describe the density profiles of apo B- and apo A1-containing lipoproteins and of Lp(a) in chimpanzee plasmas heterozygous for apo(a) isoforms after fractionation by single spin ultracentrifugation in an isopycnic gradient. The distribution of apo(a) isoforms in the density gradient was also examined by SDS-agarose gel electrophoresis and immunoblotting using chemiluminescence detection. In all double-band phenotypes examined, the smallest isoform was present along the entire length of the density gradient. The density distribution of the second isoform varied according to the size difference between the respective isoforms. Two isoforms close in size (difference in apparent molecular mass = 60 kDa) were present together in every gradient subfraction. On the contrary, when the two isoforms displayed distinct molecular mass (maximal difference in apparent molecular mass = 340 kDa), then the largest was principally present in the densest fractions of the gradient (d > 1.1 mg/ml). These observations suggest that Lp(a) particles with small apo(a) isoforms are more susceptible to interact with other lipoproteins than are Lp(a) particles with large isoforms.

Fractional removal rate of fat emulsion (K 2) remains to be low in apoE3/3 phenotype subjects with serum triglyceride level above [formula omitted

The relation between fractional catabolic rate (K 2) of an intravenously injected fat emulsion, Intralipid ®, and the level of serum triglyceride (TG) was evaluated to cast light on TG-rich lipoprotein metabolism in 182 subjects who were homozygotes for the most common form of apoE 3 3 . Both normolipidemic individuals and primary hyperlipidemic patients were included. To assess the influence of variation in the apoE phenotype on fat emulsion metabolism, 25 subjects with the apoE 4 3 phenotype and 21 with apoE 3 2 phenotype were also evaluated. In the apoE 3 3 subjects, K 2 decreased with increasing TG level up to 180mg dl , but above 180mg dl , K 2 remained at a constant level. This TG value was therefore concluded to be a cut off beyond which the TG-dependent decrease in K 2 disappeared. No aparent correlation between K 2 and serum TG was observed in subjects with apoE 4 3 or 3 2 phenotypes. In subjects with TG above 180mg dl , presumably both accelerated synthesis and limited removal are involved in the development of their hypertriglyceridemia. Since K 2 demonstrated little change with increase of TG over 180mg dl in apoE 3 3 subjects, we concluded that the capacity to catabolize fat emulsion reaches a kinetic saturation. The TG value of 180mg dl may be a physiological significance. The relation between K 2 and TG was specific in apoE 3 3 , as it was not observed in subjects having apoE 4 3 or 3 2 phenotypes.

Genetic risk factors and ischaemic cerebrovascular disease: role of common variation of the genes encoding apolipoproteins and angiotensin-converting enzyme

DNA polymorphisms in genes encoding apolipoproteins (apo) A-I, C-III, B and E and angiotensin-converting enzyme (ACE) have been proposed to be associated with the risk of coronary artery disease (CAD). We studied whether the same genetic markers would also be associated with the occurrence and extent of atherosclerosis in cervical arteries. DNA samples from 234 survivors of stroke or a transient ischaemic attack aged 60 years or less were examined. The presence of atherosclerosis was assessed using aortic arch angiograms. The SstI polymorphism of apoA-I/C-III gene locus, the Xbal polymorphism of apoB gene, common apoE phenotypes and the insertion/deletion polymorphism of the ACE gene were analysed. The allele frequencies of the apoA-I/C-III, apoB, apoE or ACE gene did not differ between the groups with (n = 148) or without (n = 85) cervical atherosclerosis. However, when patients with at least one apoE4 allele and one X2 allele of apoB were combined and compared with those without either of them (E2E3 or E3E3 and X1X1), a significant association with the presence of cervical atherosclerosis was found (P = 0.03). The patients having the E2E3 phenotype had a significantly elevated serum triglyceride level compared with those with the E3E3 phenotype (P = 0.03). Serum high-density lipoprotein (HDL) cholesterol was lower in the patients with the E2E3 phenotype than in those with the E3E3 and E3E4 (P = 0.01 and P = 0.06, respectively). The apoB or ACE genotypes were not significantly associated with serum lipid or lipoprotein levels. There was no association between the ACE gene polymorphism and the occurrence of hypertension. In conclusion, the interaction of common apoB and apoE alleles may increase the risk of atherosclerosis in cervical arteries.

Identification of Apo(a) Phenotypes in a Korean Population Using a Standardized Nomenclature System Based on the Number of Kringle IV Repeats

We determined serum lipoprotein(a) [Lp(a)] concentrations and apolipoprotein(a) [apo(a)] phenotypes in 193 healthy Koreans. We analysed the apo(a) phenotypes by a simplified sodium dodecyl sulphate-polyacrylamide gel electrophoresis method and classified apo(a) isoforms objectively using an apo(a) phenotype standard with a known number of kringle IV repeats. The frequency distribution of Lp(a) levels showed marked positive skew with a mean of 0.143 g/L and a median of 0.052 g/L. Of the 193 subjects tested, no bands were detected in three, and single- and double-band phenotypes were observed in 103 and 87, respectively. Among the Koreans, the most frequent phenotype was S5(39.4%), followed by S4S5(17.1%), S5S5(14.0%), S4(11.4%), S3S5(5.2%), and the remaining phenotypes (13.0%). The calculated apo(a) allele frequencies were LpF, 0.003; LpS1, 0.013; LpS2, 0.010; LpS3, 0.048; LpS4, 0.198; LpS5, 0.563 and Lp0, 0.165. We found that the serum Lp(a) concentration in Koreans was similar to that of Caucasians, but the apo(a) allele size distribution was shifted toward higher molecular weights.

Prevalence of double pre-beta lipoproteinemia in hyperlipidemic patients is influenced by gender, menopausal status, and ApoE phenotype.

Double pre-beta lipoproteinemia (DPBL) is a plasma lipoprotein phenotype characterized by the presence of two agarose gel electrophoretic populations of very low density lipoproteins (VLDLs, d < 1.006 g/mL), i.e., normal pre-beta-migrating VLDL and slow pre-beta VLDL. Slow pre-beta VLDL represents remnant lipoproteins derived from the hydrolysis of triglyceride (TG)-rich lipoproteins (TRLs), and thus DPBL is a characteristic of plasma remnant lipoprotein accumulation. To determine the prevalence of DPBL in our lipid clinic population, patients (n = 2501) were selected who (1) had an unambiguous VLDL electrophoretic phenotype and could be classified as having either DPBL (DPBL+), beta-migrating VLDL (beta-VLDL +), or an absence of both (DPBL/beta-VLDL-/-) and (2) had hypercholesterolemia (HC: plasma cholesterol > or = 6.2 mmol/L, n = 1017), hypertriglyceridemia (HTG: plasma TG > or = 2.3 mmol/L but < 15 mmol/L, n = 554) or combined hyperlipidemia (HC + HTG, n = 930). Patients with TG < 2.3 mmol/L and cholesterol < 5.2 mmol/L acted as control subjects (n = 343). Using a commercially available agarose gel electrophoresis system, we identified 220 hyperlipidemic patients (8.8%) with DPBL (versus < 1% of control). The prevalence of DPBL was higher in (1) male than in female patients (10.7% versus 6.7%), (2) postmenopausal than in premenopausal females (7.3% versus 4.1%), and (3) patients with HC + HTG than in those with HTG or HC alone (15.8% versus 8.3% versus 2.7%, respectively). Patients with an epsilon 2 allele had a higher prevalence of DPBL; i.e., 26.9% of apoE 3/2 and 26.2% of apoE 4/2 patients had DPBL compared with 6.5%, 6.8%, and 7.4% of apoE 3/3, 4/3, and 4/4 patients, respectively. DPBL patients consistently had increased levels of VLDL-C and (LDL + HDL)-TG and decreased levels of LDL-C, and their plasma lipid profiles were intermediate between those of beta-VLDL+ and DPBL/beta-VLDL -/- patients. These results demonstrate that male sex, postmenopausal status in women, and the presence of an apoE 3/2 or apoE 4/2 phenotype are associated with an increased incidence of DPBL in hyperlipidemic patients.

2.P.99 A novel kringle-4 number-based recombinant apo(a) standard for human Lp(a) phenotyping

Apolipoprotein(a) phenotyping is a critically im- portant method to explore the role of kringle-4 repeat num- ber as a modulator of lipoprotein(a)-associated cardiovas- cular risk. The availability of a kringle-4 number-based reference standard is therefore necessary for a reliable and generally accepted classification of apo(a) phenotypes. We propose here a battery of recombinant apo(a) isoforms that may be used as the reference standard in various gel sys- tems. Five plasmids encoding for r-apo(a) containing a known number (n 5 9, 13, 17, 25, 33) of plasminogen-like kringle-4 copies were constructed, and transfected into the human embryonic kidney cell line 293. The electrophoretic mobility of the recombinant apo(a) isoforms expressed by these cells in a hollow-fiber bioreactor was determined after reduction by SDS-gel (agarose, acrylamide or a mixture of both) electrophoresis and immunoblotting using an anti- body specific for human apo(a). The equation of the linear relationship between log r-apo(a) kringle number and rela- tive migration was used to determine the isoform size of apo(a) in normal human plasma. A very good correlation ( r 5 0.97) was found with the genotype (pulsed-field gel eletrophoresis of kpn I-digested restriction fragments of ge- nomic DNA) and among electrophoretic methods. The pro- posed recombinant standard offers the possibility to iden- tify apo(a) isoforms within a large range of molecular sizes, 9 to 33 kringle-4 copies, using simple electrophoretic tech- niques and a nomenclature based on its molecular struc- ture, i.e., the number of kringle-4 repeats.— Angles-Cano, E., S. Loyau, G. Cardoso-Saldana, R. Couderc, and P. Gillery. A novel kringle-4 number-based recombinant apo(a) stan- dard for human apo(a) phenotyping. J. Lipid Res. 1999. 40: 354-359.

2.P.133 Lp(a) levels and apo(a) phenotypes in angiographically established coronary heart disease and in controls

2.P.133 Lp(a) levels and apo(a) phenotypes in angiographically established coronary heart disease and in controls

2.P.105 Apo (a) phenotypes and Lp(a) levels in acute coronary syndromes

2.P.105 Apo (a) phenotypes and Lp(a) levels in acute coronary syndromes

2.P.97 Relation of lipoprotein Lp(a) concentration and apolipoprotein apo(a) phenotypes to the presence and severity of carotid atherosclerosis

2.P.97 Relation of lipoprotein Lp(a) concentration and apolipoprotein apo(a) phenotypes to the presence and severity of carotid atherosclerosis

Relation between lipoprotein(a) concentrations in patients with acute-phase response and risk analysis for coronary heart disease

This study investigated whether lipoprotein(a) [Lp(a)] is an acute-phase reactant that can cause important bias in risk factor analysis for coronary heart disease among patients with an acute-phase response (APR patients). To determine whether serum Lp(a) concentrations increase among APR patients, we compared the Lp(a) concentrations and apolipoprotein(a) [apo(a)] phenotypes of 100 controls with those of a random sampling of 100 APR patients. Serum Lp(a) concentration was measured by ELISA; Lp(a) phenotyping was performed by electrophoresis on sodium dodecyl sulfate-polyacrylamide gel. Lp(a) was significantly (P <0.0001) higher among APR patients (mean +/- SD 0.300 +/- 0.284 g/L) than among controls (0.118 +/- 0.193 g/L) even though the distribution of apo(a) phenotypes did not differ significantly. The 100 APR patients were grouped into 4 categories: 48 patients with infections, 25 postoperative patients, 17 patients with tumors, and 10 patients with other diseases, all of whom showed substantially higher Lp(a) values than did the controls. For the S5, S4S5, S5S5, and S4 phenotypes, the mean concentrations of serum Lp(a) were substantially higher among the APR patients.

1.P.326 Effect of apoE phenotype on postprandial triglycerides levels in young male with and without a familial history of myocardial infarction: The EARS II study

1.P.326 Effect of apoE phenotype on postprandial triglycerides levels in young male with and without a familial history of myocardial infarction: The EARS II study

1.P.334 The association of apoE phenotype and Lp(a) with cognitive function in Japanese centenarians

1.P.334 The association of apoE phenotype and Lp(a) with cognitive function in Japanese centenarians