Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Mar 5.
Published in final edited form as: J Geriatr Psychiatry Neurol. 2011 Apr 8;24(2):63–66. doi: 10.1177/0891988711402325

ABCB1 Genotype and CSF β-Amyloid in Alzheimer Disease

R Kohen 1, J B Shofer 2, O Korvatska 1, E C Petrie 1,2, L Y Wang 1,2, G D Schellenberg 1,3, E R Peskind 1,2, C W Wilkinson 1,3
PMCID: PMC3293495  NIHMSID: NIHMS357587  PMID: 21478475

Abstract

The ABCB1 gene, coding for the efflux transporter P-glycoprotein (PGP), is a candidate gene for Alzheimer disease (AD). P-glycoprotein is heavily expressed at the blood–brain barrier, where it mediates the efflux of β-amyloid (Aβ) from the brain. In this study, we investigated a possible association between 2 common ABCB1 polymorphisms, G2677T/A (Ala893Ser/Thr) and C3435T, AD, and cerebrospinal fluid (CSF) levels of Aβ. No strong evidence for association was found.

Keywords: Alzheimer disease, dementia, P-glycoprotein, MDR1, ABCB1, association study

Introduction

P-glycoprotein (PGP) is the product of the ABCB1 gene, also known as MDR1. P-glycoprotein is an ATP-dependent efflux transporter located in tissues with excretory function and at blood–tissue barriers. It is a 1280 amino acid long plasma membrane glycoprotein consisting of 2 homologous parts of approximately equal length. Each half contains 6 hydrophobic transmembrane regions and an ATP binding site.1 P-glycoprotein shields the body from a broad spectrum of potentially toxic substances by limiting their entry through the intestine and promoting their elimination into bile and urine.2 In the brain, PGP is heavily expressed at the blood–brain barrier and the blood–cerebrospinal fluid barrier, particularly at the luminar surface of brain endothelial cells.3,4 P-glycoprotein-knockout mice suffer from enhanced neurotoxicity of centrally acting drugs, illustrating the importance of PGP in actively preventing entry of potentially harmful substances into the brain.5,6

Several lines of evidence implicate PGP as a candidate gene for Alzheimer disease (AD). P-glycoprotein functions as an efflux pump for β-amyloid (Aβ).7,8 A possible role of PGP dysfunction in the etiology of AD is illustrated by studies in mice showing that amyloid precursor protein-transgenic (APP-transgenic) mice have increased Aβ levels in the brain following treatment with a PGP inhibitor. In addition, PGP-knockout mice have reduced clearance of Aβ40 and Aβ42 from the brain following intracerebral microinjection compared to wild type animals. Moreover, hybrid APP-transgenic/PGP-null mice show higher levels of brain Aβ deposition than simple APP-transgenic mice with intact PGP.9 Studies in elderly human participants without dementia have demonstrated cerebral levels of Aβ40 and Aβ42 to be inversely related to PGP expression.10,11

The ABCB1 gene, located on chromosome 7q21, is 209-kb long and contains 29 exons.12 Two polymorphisms of ABCB1, G2677T/A (Ala893Ser/Thr) and C3435T, have been extensively studied and linked to PGP expression.13 A previous small study investigating the distribution of these polymorphisms in patients with dementia and healthy controls found no association.14 The goal of our study was to investigate a possible association of G2677T/A and C3435T with AD in a larger and well-characterized sample of cases and controls collected as part of the Alzheimer's Disease Research Center (ADRC) at the Puget Sound Veterans Affairs Health Care System (VAPSHCS) in Seattle.15 In addition, this is the first study to investigate a possible association of ABCB1 genotype with cerebrospinal fluid (CSF) levels of Aβ42.

Materials and Methods

Participants

All procedures were approved by the institutional review boards of the participating institutions. This study used DNA collected through the Clinical Core of the University of Washington (UW) ADRC. Our sample included DNA from 286 individuals with AD and 240 cognitively intact controls. All participants had normal physical and neurological examinations and laboratory tests (complete blood count, chemistry panel, thyroid stimulating hormone, vitamin B12, and urinalysis). All were free of past or present major psychiatric or neurologic disorders (other than AD) and unstable medical conditions. All participants with AD met clinical diagnostic criteria for probable AD of the National Institute of Neurological and Communicative Disorders and Stroke16 and Diagnostic and Statistical Manual of Mental Disorders (Third Edition Revised) criteria for dementia of the Alzheimer Type. Cognitively normal participants had a Mini-Mental State Examination (MMSE)17 score of 26 to 30, Clinical Dementia Rating Scale score of 0, and no evidence or history of cognitive decline. Patients with mild cognitive impairment (MCI) were excluded. For a subset of participants (n = 267), CSF levels of Aβ42 from samples previously collected and analyzed as described18,19 were used in our analysis. The demographics of our sample are described in Table 1.

Table 1.

Comparison of Participants With AD and Controls by Demographic Variablesa

AD, n (%) Controls, n(%)
Gender Male 148 (52) 121 (50)
Female 138 (48) 119 (50)
Race White 272 (95) 205 (87)
African American 5 (2) 16 (7)
Asian or Pacific Islander 5 (2) 9 (4)
American Indian/Alaskan native 2 (1) 3 (1)
Other/unknown 2 (1) 4 (2)
Age 68.2 ± 9.4 52.5 ± 19.5
Open in a new tab

Abbreviations: AD, Alzheimer's disease.

a

For categorical variables (gender and race) the numbers (n) as well as the percentages (%) of participants with AD (n = 286) and controls (n = 237) that met the demographic criteria listed are given. For age at entry into the study, mean ± SD is given. Differences in gender distribution between subjects with AD and controls were not significant, but differences in race distribution (Fisher exact test P = .004) and age (t test P = .000) were highly significant.

ABCB1 Genotyping

We genotyped the G2677T/A (Ala893Ser/Thr) polymorphism in exon 22 (dbSNP: rs2032582) and the C3435T polymorphism in exon 27 (dbSNP: rs1045642). Genomic DNA (100 ng) was amplified in the presence of gene-specific primers and allele-specific fluorescent probes, using inventoried TaqMan SNP genotyping assays (Applied Biosystems, California) and following the manufacturers' instructions.

Statistical Analysis

Calculations for deviation from Hardy–Weinberg equilibrium and for linkage disequilibrium between G2677T/A and C3435T were performed using χ2 tests. Haplotype analysis was done using Multiallelic Interallelic Disequilibrium Analysis Software (MIDAS).20 Bivariate associations between diagnosis (AD vs control) and G2677T/A or C3435T genotype (C/C, C/T, or T/T) were assessed using Fisher exact tests. Due to low frequencies for several G2677T/A genotypes, statistical inference for G2677T/A focused on the comparison between T allele carriers vs those with no T-allele. To account for the potential confounding of age, ethnicity, and APOE4 status on the relationship between diagnosis and G2677T/A T allele, logistic regression was used with G2677T/A T allele as the dependent variable and diagnosis, age, ethnicity (white vs. non-white), and APOE4 status (presence of E4 allele vs absence) as independent covariates. Similarly, to account for confounding in the association between C3435T genotype and AD diagnosis, multinomial regression was carried out with C3435T genotype as the dependent variable (C/C as the reference group) and diagnosis, age, ethnicity, and APOE4 status as independent covariates. Results are presented as odds ratios.

Linear regression was used to assess the association between Aβ42 (the dependent variable) and G2677T/A T allele or C3435T genotype with covariates age, ethnicity, AD diagnosis, and APOE4 status. Normality of Aβ42 was assessed using the Kolmogorov-Smirnov test. Analyses were carried out using Stata 9 (StataCorp) and R 2.9.1.21

Results

Demographics and ABCB1 Genotype

The demographics of AD participants and controls are shown in Table 1. The control group was significantly younger and had higher minority participation. Genotype distributions for the G2677T/A and C3435T polymorphisms were checked for deviation from Hardy–Weinberg equilibrium and no deviation was observed (AD: G2677T/A χ2 = 6.69, P = .08; C3435T χ2 = 0.15, P = .69. Controls: G2677T/A χ2 = 0.39, P = .94; C3435T χ2 = 0.59, P = .44). G2677T/A and C3435T were in linkage disequilibrium (LD) (D′ = −0.90, r2 = −0.59). Since vgenotype distributions differed significantly by race (G2677T/A P = .000, C3435T P = .005, data not shown), we controlled for age and race in the following analyses.

Association of ABCB1 Genotype With AD

Genotype distributions did not differ significantly by disease status (Table 2). In the dominant model of 2677T (presence of T allele vs other), we observed slightly increased odds for AD among T-allele carriers (OR 1.49, 95% CI 1.03 to 2.16, P = .034), however, the effect was no longer significant if we adjusted for age (n = 408, OR 1.23, 95% CI 0.81 to 1.87, P = .3). Our results remained essentially unchanged if we adjusted for ethnicity (OR = 1.18, 95% CI 0.77 to 1.80, P = .4) or ethnicity plus APOE4 status (OR = 1.23, 95% CI 0.78 to 1.92, P = .4) in addition to age. Our study had 65% power (α = 0.05) to detect a 10% increase in G2677T/A T-allele frequency among participants with AD, 94% power to detect a 15% increase, and over 99% power to detect a 20% increase using 2-sided tests. No association between C3435T and AD was observed. These results remained the same when age, ethnicity, and APOE genotype were controlled for (data not shown). For participants with AD (n = 285), mean age of disease onset was 67 ± 9.4 years. There were no significant differences in age of AD onset between participants of different races, or between genotype groups for G2677T/A or C3435T (data not shown).

Table 2.

Distribution of Genotype Frequencies for G2677T/A and C3435T by AD Diagnosisa

AD
 Controls
Polymorphism Genotype n (%)
G2677T/A G/G 73 (26) 78 (33)
G/T 142 (50) 110 (46)
G/A 5 (2) 9 (4)
T/A 5 (2) 5 (2)
T/T 60 (21) 38 (16)
A/A 1 (0.4) 0
C3435T C/C 63 (22) 66 (28)
C/T 138 (49) 114 (48)
T/T 83 (29) 60 (25)
Open in a new tab

Abbreviations: AD, Alzheimer's disease.

a

Shown are the allele frequencies (%) for subjects with AD (n = 286) and controls (n = 240): 2 AD participants had missing C3435T genotypes. Genotype distributions did not differ significantly between participants with AD and controls (G2677T/A P = .17, C3435T P = .30, Fisher exact test).

Association of ABCB1 Genotype With CSF Levels of Aβ42

In a subset of 267 participants, CSF levels of Aβ42 were significantly lower in participants with AD (n = 54, 109 ± 31 pg/mL CSF) than in controls (n = 213, 163 ± 33 pg/mL CSF), P < .0001 (linear regression). Age of entry into the study and APOE genotype were both significantly correlated with CSF levels of Aβ42 (linear regression, age: P = .000, adjusted R2 = 0.21; APOE4: P = .000, adjusted R2 = .17). Neither the C3435T nor the G2677T/A polymorphism were associated with CSF levels of Aβ42 (Table 3). These results did not change if we applied a dominant model of 2677T or if participants with AD and controls were analyzed separately (not shown). For both polymorphisms, linear regression controlling for age, ethnicity, AD, and APOE4 status showed no significant associations between ABCB1 genotype and CSF levels of Aβ42 (data not shown). The Kolmogorov-Smirnov test showed a slight deviation from normality in the distribution of Aβ42 (P = .043). Using a power of Aβ42 derived from the Box-Cox set of transformation did not change the results above.

Table 3.

Cerebrospinal Fluid (CSF) Levels of Amyloid-β42 by G2677T/A and C3435T Genotype Groupa

Total
Polymorphism Genotype n Aβ42
G2677T/A G/G 83 155 ± 41
G/T 129 152 ± 38
G/A 6 175 ± 35
T/A 4 147 ± 47
T/T 45 145 ± 40
P value 0.44
C3435T C/C 66 152 ± 38
C/T 130 155 ± 41
T/T 71 148 ± 38
P value 0.49
Open in a new tab
a

Shown are the number of participants (n) and mean ± SD CSF levels of Aβ42 (pg/mL CSF) per genotype group for all subjects (n = 267). The P values (ANOVA) for comparison of Aβ42 levels by genotype are shown below. The G2677T/A A/A genotype was not represented in the subset of our sample for which Aβ42 CSF levels had been determined.

Abbreviation: AD, Alzheimer's disease.

Discussion

Our study showed no significant associations of the ABCB1 G2677T/A and C3435T polymorphisms of ABCB1 with AD or with CSF levels of Aβ42. This negative result is in keeping with previous conflicting reports about associations of these genetic variants with PGP expression or function.13,22,23 Our findings also confirm the results of a prior small association study in 113 patients with various types of dementia and 41 controls which reported no association of ABCB1 genotype with dementia.14 Our results also agree with previous studies that showed CSF Aβ42 to be lower in patients with AD than in healthy control participants.24

The strength of this study is the simultaneous investigation of ABCB1 genotype and CSF levels of Aβ42 in a large cohort of AD participants and controls. We are not reporting here on the influence of APOE genotype on CSF levels of Aβ42, as this has previously been described in a subset of our sample.18,19 A limitation of this study is the lack of detailed clinical information about our AD participants, as it prevents us from making observations about possible effects of ABCB1 genotype on disease severity and progression. Moreover, we investigated only 2 common variants of the ABCB1 gene whose effects on PGP expression remain controversial. A possible effect of rare deleterious variants on AD risk is worthy of further investigation. The ABCB1 gene is subject to complex regulation.25–27 A disturbance of PGP expression due to disruption of regulatory mechanisms remains a plausible risk factor for AD.

Acknowledgments

Funding The authors disclosed receipt of the following financial support for the research and/or authorship of this article: NIH/NIA grant P50AG005136 and by the Northwest Network VISN 20 Mental Illness Research, Education and Clinical Center (MIRECC), the VA Puget Sound Geriatric Research, Education and Clinical Center (GRECC), and the Office of Research and Development Medical Research Service, Department of Veterans Affairs.

Footnotes

Declaration of Conflicting Interests The authors declared no conflicts of interest with respect to the authorship and/or publication of this article.

References

  • 1.Chen CJ, Chin JE, Ueda K, et al. Internal duplication and homology with bacterial transport proteins in the mdr1 (P-glycoprotein) gene from multidrug-resistant human cells. Cell. 1986;47(3):381–389. doi: 10.1016/0092-8674(86)90595-7. [DOI] [PubMed] [Google Scholar]
  • 2.Fromm MF. Importance of P-glycoprotein at blood-tissue barriers. Trends Pharmacol Sci. 2004;25(8):423–429. doi: 10.1016/j.tips.2004.06.002. [DOI] [PubMed] [Google Scholar]
  • 3.Demeule M, Regina A, Jodoin J, et al. Drug transport to the brain: key roles for the efflux pump P-glycoprotein in the blood-brain barrier. Vascul Pharmacol. 2002;38(6):339–348. doi: 10.1016/s1537-1891(02)00201-x. [DOI] [PubMed] [Google Scholar]
  • 4.Lee G, Bendayan R. Functional expression and localization of P-glycoprotein in the central nervous system: relevance to the pathogenesis and treatment of neurological disorders. Pharm Res. 2004;21(8):1313–1330. doi: 10.1023/b:pham.0000036905.82914.8e. [DOI] [PubMed] [Google Scholar]
  • 5.Schinkel AH, Wagenaar E, Mol CA, van Deemter L. P-glycoprotein in the blood-brain barrier of mice influences the brain penetration and pharmacological activity of many drugs. J Clin Invest. 1996;97(11):2517–2524. doi: 10.1172/JCI118699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Loscher W, Potschka H. Drug resistance in brain diseases and the role of drug efflux transporters. Nat Rev Neurosci. 2005;6(8):591–602. doi: 10.1038/nrn1728. [DOI] [PubMed] [Google Scholar]
  • 7.Lam FC, Liu R, Lu P, et al. beta-Amyloid efflux mediated by p-glycoprotein. J Neurochem. 2001;76(4):1121–1128. doi: 10.1046/j.1471-4159.2001.00113.x. [DOI] [PubMed] [Google Scholar]
  • 8.Kuhnke D, Jedlitschky G, Grube M, et al. MDR1-P-Glycoprotein (ABCB1) Mediates transport of Alzheimer's amyloid-beta peptides—implications for the mechanisms of Abeta clearance at the blood-brain barrier. Brain Pathol. 2007;17(4):347–353. doi: 10.1111/j.1750-3639.2007.00075.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Cirrito JR, Deane R, Fagan AM, et al. P-glycoprotein deficiency at the blood-brain barrier increases amyloid-beta deposition in an Alzheimer disease mouse model. J Clin Invest. 2005;115(11):3285–3290. doi: 10.1172/JCI25247. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Vogelgesang S, Cascorbi I, Schroeder E, et al. Deposition of Alzheimer's beta-amyloid is inversely correlated with P-glycoprotein expression in the brains of elderly non-demented humans. Pharmacogenetics. 2002;12(7):535–541. doi: 10.1097/00008571-200210000-00005. [DOI] [PubMed] [Google Scholar]
  • 11.Vogelgesang S, Warzok RW, Cascorbi I, et al. The role of P-glycoprotein in cerebral amyloid angiopathy; implications for the early pathogenesis of Alzheimer's disease. Curr Alzheimer Res. 2004;1(2):121–125. doi: 10.2174/1567205043332225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Bodor M, Kelly EJ, Ho RJ. Characterization of the human MDR1 gene. AAPS J. 2005;7(1):E1–E5. doi: 10.1208/aapsj070101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Leschziner GD, Andrew T, Pirmohamed M, Johnson MR. ABCB1 genotype and PGP expression, function and therapeutic drug response: a critical review and recommendations for future research. Pharmacogenomics J. 2007;7(3):154–179. doi: 10.1038/sj.tpj.6500413. [DOI] [PubMed] [Google Scholar]
  • 14.Frankfort SV, Doodeman VD, Bakker R, et al. ABCB1 genotypes and haplotypes in patients with dementia and age-matched non-demented control patients. Mol Neurodegener. 2006;1:13. doi: 10.1186/1750-1326-1-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Tsuang D, Kukull W, Sheppard L, et al. Impact of sample selection on APOE epsilon 4 allele frequency: a comparison of two Alzheimer's disease samples. J Am Geriatr Soc. 1996;44(6):704–707. doi: 10.1111/j.1532-5415.1996.tb01836.x. [DOI] [PubMed] [Google Scholar]
  • 16.McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer's disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer's disease. Neurology. 1984;34(7):939–944. doi: 10.1212/wnl.34.7.939. [DOI] [PubMed] [Google Scholar]
  • 17.Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12(3):189–198. doi: 10.1016/0022-3956(75)90026-6. [DOI] [PubMed] [Google Scholar]
  • 18.Peskind ER, Wilkinson CW, Petrie EC, Schellenberg GD, Raskind MA. Increased CSF cortisol in AD is a function of APOE genotype. Neurology. 2001;56(8):1094–1098. doi: 10.1212/wnl.56.8.1094. [DOI] [PubMed] [Google Scholar]
  • 19.Peskind ER, Li G, Shofer J, et al. Age and apolipoprotein E*4 allele effects on cerebrospinal fluid beta-amyloid 42 in adults with normal cognition. Arch Neurol. 2006;63(7):936–939. doi: 10.1001/archneur.63.7.936. [DOI] [PubMed] [Google Scholar]
  • 20.Gaunt TR, Rodriguez S, Zapata C, Day IN. MIDAS: software for analysis and visualisation of interallelic disequilibrium between multiallelic markers. BMC Bioinformatics. 2006;7:227. doi: 10.1186/1471-2105-7-227. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.R Development Core Team . R Foundation for Statistical Computing. Vienna, Austria: 2008. [Accessed December 5, 2010]. R: A language and environment for statistical computing. http://www.R-project.org. [Google Scholar]
  • 22.Woodahl EL, Ho RJ. The role of MDR1 genetic polymorphisms in interindividual variability in P-glycoprotein expression and function. Curr Drug Metab. 2004;5(1):11–19. doi: 10.2174/1389200043489108. [DOI] [PubMed] [Google Scholar]
  • 23.Marzolini C, Paus E, Buclin T, Kim RB. Polymorphisms in human MDR1 (P-glycoprotein): recent advances and clinical relevance. Clin Pharmacol Ther. 2004;75(1):13–33. doi: 10.1016/j.clpt.2003.09.012. [DOI] [PubMed] [Google Scholar]
  • 24.Shaw LM, Vanderstichele H, Knapik-Czajka M, et al. Cerebrospinal fluid biomarker signature in Alzheimer's disease neuroimaging initiative subjects. Ann Neurol. 2009;65(4):403–413. doi: 10.1002/ana.21610. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Sukhai M, Piquette-Miller M. Regulation of the multidrug resistance genes by stress signals. J Pharm Pharm Sci. 2000;3(2):268–280. [PubMed] [Google Scholar]
  • 26.Su F, Ouyang N, Zhu P, et al. Psychological stress induces chemoresistance in breast cancer by upregulating mdr1. Biochem Biophys Res Commun. 2005;329(3):888–897. doi: 10.1016/j.bbrc.2005.02.056. [DOI] [PubMed] [Google Scholar]
  • 27.Spudich A, Kilic E, Xing H, et al. Inhibition of multidrug resistance transporter-1 facilitates neuroprotective therapies after focal cerebral ischemia. Nat Neurosci. 2006;9(4):487–488. doi: 10.1038/nn1676. [DOI] [PubMed] [Google Scholar]

RESOURCES