Genetic Variations in ABCA7 Can Increase Secreted Levels of Amyloid-β 40 and Amyloid-β 42 Peptides and ABCA7 Transcription in Cell Culture Models

Abstract

Alzheimer’s disease (AD) is characterized by extracellular deposits of amyloid-β (Aβ) in the brain. ABCA7 is highly expressed in the brain and a susceptibility gene for late-onset AD (LOAD). The minor alleles at two ABCA7 single-nucleotide polymorphisms (SNPs), rs3764650 (T>G; intron13) and rs3752246 at a predicted myristoylation site (C>G; exon33; p.Gly1527Ala), are significantly associated with LOAD risk; however, the mechanism of this association is unknown. Functional consequences of both SNPs were examined in HEK293 and CHO cells stably expressing AβPP_Swe. Luciferase reporter assays in HEK293 cells suggested that intron13 carrying rs3764650 major T-allele (int13-T) possessed promoter-enhancing capabilities. Co-transfection experiments with hABCA7 and int13-T resulted in significantly increased ABCA7 protein level relative to that with int13-G. Expression of hABCA7 carrying rs3752246 risk allele led to increases in secreted Aβ₄₀ and Aβ₄₂ and β-secretase activity in CHO- and HEK-AβPP_Swe cells. Hydroxymyristic acid treatment of cells expressing hABCA7 carrying the rs3752246 major G allele resulted in increased β-secretase activity and levels of Aβ, suggesting that lack of myristoylation contributes to the observed cell-phenotypes. Molecular weight determination, by gel-electrophoresis and mass spectrometry, of hABCA7 peptides spanning position 1527 showed loss of post-translational modification in the risk-allele peptide. These results suggest that decreased expression, or impaired function, of ABCA7 may contribute to AD pathology.

Keywords

ABC transporters ABCA7 protein Alzheimer’s disease amyloid beta-peptides BACE1 protein genome-wide association study myristoylation SNPs

INTRODUCTION

Alzheimer’s disease (AD) is a progressive neurodegenerative disease characterized by cognitive decline and memory dysfunction which correlate well with synaptic loss and neurodegeneration [1]. Pathologically, AD is characterized by extracellular deposits of amyloid-β (Aβ) peptides as neuritic plaques, intraneuronal aggregates of hyperphosphorylated tau and aberrant lipid metabolism in the brain [2]. Aβ₄₀ and Aβ₄₂, the most abundant Aβ species, are produced through sequential cleavage of amyloid-β protein precursor (AβPP) by β-secretase 1 (BACE1) and γ-secretase complex [3]. However, AβPP is mainly processed by the non-amyloidogenic pathway where α-secretase cleaves AβPP within the Aβ domain preventing Aβ generation and releasing soluble AβPPα (sAβPPα) [4]. Late-onset AD (LOAD), which manifests in people over the age of 65 years, accounts for more than 90% of AD cases [5] where the genetic component has been estimated to account for 60–80% [6]. Recent large-scale genome-wide association studies (GWASs) have identified >20 susceptibility genes that implicate three main pathways in LOAD: immune system, lipid metabolism, and synaptic function [7 –13]. Among well-replicated LOAD susceptibility genes is the ATP-binding cassette transporter, sub-family A, member 7 (ABCA7) [9 , 15].

ABCA7 encodes a 220kDa transmembrane protein that is highly expressed in the brain (particularly in hippocampal CA1 neurons and in microglia) and the myelo-lymphatic system [16, 17]. When exogenously expressed, ABCA7 can efflux lipids to apolipoprotein acceptors (eg. apoAI or apoE) and generate high-density lipoprotein particles [18 –22]. In contrast, ABCA7 knock-out and knock-down studies show a largely unchanged cellular lipid efflux capacity but instead suggest that endogenously expressed ABCA7 is important for phagocytosis [17 , 24]. Overexpression of wild-type murine Abca7 has also been shown to regulate AβPP processing and inhibit Aβ secretion [25]. Indeed, ABCA7 was significantly associated with neuritic plaque burden in human brain [26, 27]. In addition, the sterol regulatory element binding protein 2 (SREBP2) has been shown to stimulate ABCA7 expression, which is activated by sterol depletion [23]. Furthermore, murine Abca7 also appears to play gender-biased roles in behavior where male Abca7-KO mice exhibited impaired novel object recognition memory and female knock-out mice exhibited impaired spatial reference memory [28].

The ABCA7 gene maps to chromosome 19p13.3 and consists of 47 exons (http://useast.ensembl.org/index.html). A single GWAS-significant ABCA7 SNP in the intron13 (rs3764650; T>G) was reported in a large European dataset with LOAD risk attributed to minor G-allele [9]. Weak linkage disequilibrium (LD) structure of the ABCA7 locus in the European cohort implicates only a narrow intronic region in risk of LOAD [9]. Meanwhile in North American GWAS, the strongest evidence for association with ABCA7 was obtained for a SNP within exon 33 (rs3752246; C>G) that results in a missense substitution (NP_061985.2: p.Gly1527Ala) within the second extracellular loop of the ABCA7 transporter [12]. LOAD risk was associated with the minor G-allele. Weak LD between SNPs rs3764650 and rs3752246 indicates that the association signals observed in European and North American GWASs are independent of each other (Fig. 1). In contrast, in African American population, the strongest association with LOAD was observed for ABCA7 SNP rs115550680, which is in LD with both rs3764650 and rs3752246. It is tempting to speculate that two variations on the same ABCA7 allele could be potentially functional or act synergistically. Indeed, in African Americans the effect size of ABCA7 (OR = 1.8) was stronger than in other populations and comparable with APOE risk (OR = 2.3) [14]. Importantly, intronic SNP rs3764650 reported in a European cohort [9] and another GWAS-significant intronic SNP rs4147929 identified by a large meta-analysis [29] are located within/nearby CpG-islands, suggesting the contribution of ABCA7 transcriptional regulatory regions to the risk of LOAD (Fig. 1).

In this study we aim to illuminate the functional relevance of rs3764650 and rs3752246 in the context of AD phenotypes using cell culture models. We show data to suggest that the major allele of the SNP rs3764650 can increase the expression of ABCA7; and the minor (risk) allele of rs3752246 potentially eliminates a post-translational modification site on ABCA7 protein and increases the level of secreted Aβ₄₀ and Aβ₄₂ by, at least in part, altering BACE1 activity.

MATERIALS AND METHODS

Materials

The pGL3-promoter vector, Wizard SV Gel and PCR Clean-Up System, and Luciferase Assay System with Reporter Lysis Buffer were purchased from Promega. iScript cDNA Synthesis Kit, Experion RNA StdSens Analysis Kit, SsoFast EvaGreen Supermix, and SureBeadstrademark Protein G kit were obtained from Bio-Rad. All custom primers were purchased through Integrated DNA Technologies, and DNA-free Kit was purchased from Applied Biosystems. The pCR4-TOPO TA Cloning Kit with OneShot TOP10 chemically competent E. coli cells, T4 DNA Ligase, Lipofectamine 2000 Transfection Reagent, the Aβ₄₀ and Aβ₄₂ ELISA kits, TRIzol Reagent, CyQUANT Cell Proliferation Assay Kit and One Shot MAX Efficiency DH5α-T1 Competent cells were purchased from Life Technologies. The QuikChange II XL Site-Directed Mutagenesis Kit was purchased from Agilent Technologies, and the anti-human ABCA7 antibody was purchased from Aviva Systems Biology. The monoclonal anti-β-actin-peroxidase antibody, mevastatin, 2-hydroxymyristic acid (HMA), and Celite S were purchased from Sigma-Aldrich. To detect cellular AβPP and secreted sAβPPα the monoclonal antibody clone 6E10 was purchased from Covance, and the BACE1 61-3E7 antibody was purchased from Santa Cruz Biotechnology, Inc. The pcDNA3.1/hygro(+) empty vector and pcDNA3.1/hygro(+)-hABCA7 plasmid were kindly provided by Dr. Kazumitsu Ueda from Kyoto University, Japan [21]. β-secretase substrate IV, fluorogenic and β-secretase inhibitor III were obtained from CalBiochem. The c-myc monoclonal antibody and pCMV-myc vector were purchased from Clontech Laboratories Inc.

Cell lines and transfections

Chinese hamster ovary (CHO) cells stably expressing human AβPP isoform 695 containing the Swedish mutation (CHO-AβPP_Swe) were kindly provided by Dr. Andrew Hill (The University of Melbourne,Australia). The following double mutation in AβPP found in a Swedish family with familial AD produced 6–8 times more Aβ than normal AβPP, Lys to Asn at residue 595 and Met to Leu at position 596 [30]. The CHO-AβPP_Swe cells were maintained in RPMI 1640 medium supplemented with 10% FBS,1x L-glutamine and 7.5μg/mL puromycin. HEK293 cells were obtained from ATCC and maintained in EMEM containing 10% FBS. HEK293 cells stably expressing AβPP_Swe (HEK293- AβPP_Swe) were obtained from Dr. D. Selkoe (Harvard Medical School, Boston, USA). Unless otherwise noted, transfections were performed with 4μg of DNA at a cell density of 90% in antibiotic-free medium using Lipofectamine Reagent following the manufacturer’s instructions, and cells were harvested for the respective experiment 24 h later.

Expression analysis

HEK293 cells were treated with mevastatin in serum-free DMEM containing 10μg/mL fatty-acidfree BSA. RNA was extracted using TRIzol reagent,DNA was removed from the RNA preparation with DNA-free, and RNA integrity was evalua-ted with Experion following the manufacturer’s instructions. cDNA was generated using iScript following the manufacturer’s instructions, and qPCRwas performed on SREBF2 and β-actin usingSsoFast EvaGreen Supermix with the followingprimers: SREBF2, forward 5’- AGGCAGGCTTTGAAGACGAA-3’ and reverse 5’-AGCTTCCCTGTGATGTGCAG-3’ and actin, forward 5’-TGTCCACCTTCCAGCAGATGT-3’ and reverse 5’- AGTCCGCCTAGAAGCATTTGC-3’ and the reaction: 98°C 2 min, 98°C 2 s, 55°C 5 s, go to step 2 39x using the CFX96 Real-time System. SREBF2 expression was determined relative to vehicle and was normalized to β-actin.

Generation of vectors containing int13-T and int13-G

First, intron13 containing rs3764650 major-T allele (int13-T) of the ABCA7 gene was cloned into the pCR4-TOPO vector. Primers were designed that flanked intron13 (forward 5’-GCCCTATCCGTGCTATGTGGACG-3’ and reverse 5’-CAGCGACCGGCTCAGCACAC-3’), and used to amplify this 395 bp sequence from human non-demented genomic DNA by PCR (94°C 2 min, 94°C 1 min, 68°C 1 min, 72°C 1 min, go to step 2 32 times, 72°C 7 min). The PCR product was purified from agarose gel using Promega Wizard SV Gel and PCR Clean up Kit and cloned into the pCR4-TOPO vector following the manufacturer’s instructions. Briefly, the PCR product was mixed and incubated with the vectorin the provided salt solution for 20 min, an aliquot of the mixture was transformed into OneShot TOP10 cells using heat shock at 42°C, and the cells were spread onto LB plates containing 50μg/mL kanamycin. The sequence and orientation of select pCR4-TOPO-int13-T clones were verified. Site-directed mutagenesis created pCR4-TOPO-intron13 containing the rs3764650 minor G-allele (int13-G, see below). These are the int13-T- and int13-G-containing pCR4-TOPO plasmids usedfor the co-transfection experiment. Next, the int13-T and int13-G inserts were cloned into the multiple cloning site of the pGL3-promoter luciferase reporter vector. Int13-T and int13-G were amplified from the TOPO vectors using forward and reverse primers containing KpnI and XhoI restriction enzyme sites, respectively (forward 5’-GGTTTAAAGGTACCCGCCCTT-3’ and reverse 5’-GCGGCCGCCTCGAGGCCCTT-3’) by PCR (94°C 2 min, 94°C 1 min, 50°C 1 min, 72°C 1.5 min, go to step 2 30 times, 72°C 7 min) which were cloned again into pCR4-TOPO vector. KpnI- and XhoI-digested int13-T and int13-G inserts were individually ligated to KpnI- and XhoI-digested and dephosphorylated pGL3-promoter vector using T4 DNA Ligase. These are the pGL3-promoter-int13-T and –int13-G vectors used for luciferase assay and co-transfection experiments.

Luciferase reporter assay

HEK293 cells were transfected with pGL3-promoter (empty vector), pGL3-promoter-int13-T, or pGL3-promoter-int13-G. Twenty-four hours post-transfection, the cells were treated with 20μM mevastatin (in DMSO) for 6 h (to stimulate ABCA7 expression machinery), washed in clear HBSS and harvested in 1 X lysis buffer provided in the Promega Luciferase assay kit. The cell lysates were placed on ice, vortexed for 10–15 s, centrifuged at 12000 xg for 2 min at 4°C and the supernatants were transferred to new tubes which were stored at –80°C until the assay. The luciferase activity from the cell lysates of three independent treatment experiments (n = 3) was determined following the manufacturer’s instructions. Briefly, the samples were diluted 1 : 100 with 1 X lysis buffer supplemented with 1 mg/mL BSA and an aliquot was placed into an opaque 96-well plate. Using the injecting port of the plate reader, the luminescence was read immediately after the addition of the Luciferase Assay Reagent. The luminescence counts were multiplied by the appropriate dilution factor and normalized to cell protein levels which were determined using the detergent compatible Bio-Rad BCA protein assay.

Co-transfection experiments with int13-T or int13-G

HEK293 cells were co-transfected with equal molar ratios of pcDNA3.1/hygro(+)-hABCA7 and pCR4-TOPO-int13-T, -int13-G or empty pCR4-TOPO vector using Lipofectamine reagent. Twenty-four hours post-transfection, the cells were harvested. The proteins were resolved by SDS-PAGE and ABCA7 and β-actin were detected by western blot analysis. In another set, HEK293 cells were co-transfected with pcDNA3.1/hygro(+)-hABCA7 and pGL3-promoter-int13-T, or pGL3-promoter-int13-G, and ABCA7 and β-actin levels were analyzed. Further still, the CMV promoter was removed from the hABCA7-encoding mammalian expression vector pcDNA3.1/hygro(+)-hABCA7 using HindIII and NruI, and then blunt-ended and self-ligated to create hABCA7-noCMV. This vector was co-transfected in HEK293 cells with pGL3-promoter-int13-T or pGL3-promoter-int13-G, and ABCA7 and β-actin levels were analyzed.

Creating the variant clones

The thymine at position 115 of intron13 (within pCRTOPO-int13-T) and the cytosine at position 164 of exon33 (within pcDNA3.1/hygro(+)-hABCA7) were respectively mutated to a guanine using the QuikChange Site-Directed Mutagenesis Kit as per the manufacturer’s instructions. Briefly, a PCR reaction was performed (95°C 1 min, 95°C 50 s, 60°C 50 s, 68°C 12 min, go to step 2 18 times, 68°C 7 min) using the aforementioned dsDNA template and custom primers containing the mutation of interest (for intronic rs3764650 SNP: forward 5’-CTGCGAACTTTGCACCGTTACACCACTCCACGT-3’ and reverse 5’-ACGTGGAGTGGTGTAACGGTGCAAAGTTCGCAG-3’, and for exonic rs3752246 SNP: forward 5’-CAGCTGTCTGAGGGTGCACTGATGGCC-3’ and reverse 5’- GGCCATCAGTGCACCCTCAGACAGCTG-3’). Next, restriction enzyme DpnI was added to digest the parental strand and the reactions were transformed into XL10-Gold ultracompetent cells. The DNA was extracted from the resulting clones and sequenced to verify the presence of the mutation.

Western blotting analysis

Five μg of each protein sample was resolved on SDS-PAGE gels, transferred to PVDF membrane, and the membrane was blocked with 5% skim milk in TBST for at least 1 h. Primary antibody incubations were performed overnight in 1% skim milk in TBST, and after washes with TBST (3 times for 15 min), secondary antibody incubations (if required) were performed in 5% skim milk in TBST for 1 h. Chemiluminescence was captured on film and densitometry on scanned images was performed using Alpha View version 3.2.3.0.

Aβ determination

The medium was harvested, mixed with protease inhibitors and frozen at –80°C until the analysis. The medium samples from four separate experiments (n = 4) were then thawed, appropriately diluted and the levels of Aβ₄₀ and Aβ₄₂ were determined by ELISA. The values were multiplied by the respective dilution factors, normalized to cell protein level, and presented relative to vector-transfected cells.

β-secretase activity

Cells were washed twice with clear HBSS and scraped into 100μL cold Reaction Buffer (20 mM sodium citrate, pH 4.8 and 0.06% (v/v) TritonX-100), mixed well and stored at –80°C. The cell lysates from three separate experiments (n = 3) were incubated on ice for 10 min and centrifuged at 10,000 xg for 1 min. The supernatants were transferred to a new tube and their protein content was determined using the detergent compatible Bio-Rad BCA protein assay. The enzymatic reaction for secretase activity was carried out in black microtiter plates. Approximately 1μg of total protein was measured. Twenty μM of fluorogenic β-secretase substrate, which was reconstituted in DMSO, was used for each reaction. The plate was protected from light, tapped gently to mix and incubated at 37°C for 30 min. The plate was read on a fluorescent microplate reader using light filters that allow for EDANS excitation at 335–355 nm and collected emitted light at 495–510 nm. FiveμM of β-secretase inhibitor III reduced the activity produced by 1μg of recombinant human BACE1.

Generation and purification of ABCA7 peptides

Peptides of ABCA7 were created that span position 1527 of ABCA7/Gly1527 or ABCA7/Ala1527 (each 96 amino acids long) called 96-Gly and 96-Ala, respectively. The forward primer 5’- GAATTCCCCTGCCTGGCGGGGCC-3’ carrying an EcoRI site and the reverse primer 5’-GGTACCTCAGATGGAGACGAGGACGTCCAC-3’ carrying a KpnI site followed by a stop codon were used to amplify pcDNA3.1/hygro(+)-hABCA7 major allele containing (Gly1527) vector and pcDNA3.1/hygro(+)-hABCA7 minor-allele containing (Ala1527) vector, respectively, by PCR (94°C 30 s, 60°C 30 s, 72°C 30 s, for 30 cycles). The PCR fragments were cloned into pCR4-TOPO vector; EcoRI- and KpnI-digested fragments from the pCR4-TOPO vectors were cloned into pCMV-myc vector which was also cleaved by EcoRI and KpnI. The insertion, orientation and sequence accuracy of the fragments were confirmedby sequencing analysis. The pCMV-myc constructs encoding the ABCA7/Gly1527 peptide (96-Gly) or the ABCA7/Ala1527 peptide (96-Ala) (where the peptides were fused to the myc-tag in the pCMV vector) were transfected and expressed in HEK293 cells. The cells were harvested at 24 h post-transfection, and the cell lysates were prepared in a buffer containing 50 mM Tris pH 8.0, 0.137 M NaCl, 1% NP-40 (v/v), 2 mM glucose, 5% glycerol (v/v), and 1x protease inhibitor cocktail. The SureBeadstrademark Protein G kit was used to purify the myc-tagged peptides, following manufacturer’s instructions. Briefly, 10μg c-myc antibody was mixed with protein G beads in 1x PBS for 20 min at room temperature (RT). The beads were magnetized and washed, and cell lysates containing 900μg protein were added to each tubeand rotated for 1 hour at RT. The beads were magnetized, supernatants discarded, and washed. The recombinant peptides were eluted with 20 mM glycine pH 2.0.

Mass spectrometry

Aliquots of 1μl purified ABCA7 peptides (0.3μg/μL) were mixed with an equal volume of α-cyano-hydroxycinnamic acid in 80% (v/v) acetonitrile, 0.1% (v/v) trifluoroacetic acid. Spectra were acquired on a 4800 MALDI-TOF-TOF instrument (Applied Biosystems) and analyzed using the Data Explorer software.

Inhibition of myristoylation

The cells were treated with vehicle (5 mg/mL BSA-containing media) or HMA (1 mM HMA, 5 mg/mL BSA in media) for 24 h. Media was harvested and treated as outlined under ‘Aβ determination’ and the cells were treated as outlined in ‘β-secretase activity’ above. One mM HMA has been routinely used in various cell types to inhibit N-myristoyltransferase [31 –33].

Statistical analysis

All results were reported as mean±SD. Statistical significance was analyzed by unpaired two-tailed t-test for comparison of two means, or by one-way ANOVA with Bonferroni post-hoc test for comparison of three or more means. Statistical significance was defined as p < 0.05.

RESULTS

The rs3764650 major T-allele may increase transcription

Intron13 of the human ABCA7 gene is 395 bp in length, spanning position 1,046,406 to 1,046,801 on chromosome 19. SNP rs3764650 is found at position 1,046,520, where the major allele is T and the minor allele is G. Luciferase reporter vectors were used to determine whether the intron13 sequence carrying the major T-allele of rs3764650 (named int13-T) enhances or represses transcription and whether intron13 carrying the minor G-allele of rs3764650 (named int13-G) disrupts this function. Int13-T or int13-G were cloned into the multiple cloning site upstream of the SV40 promoter within the pGL3-promoter luciferase reporter vector and the sequences were verified (Fig. 2A). HEK293 cells were transfected with these constructs and the resulting luminescence was measured after treatment with mevastatin or vehicle. Mevastatin was used in order to activate cellular ABCA7 machinery since it had been shown that cholesterol depletion stimulates SREBP2 expression which activates ABCA7 expression machinery [23, 34]. HEK293 cells transfected with int13-T exhibited increased luminescence after mevastatin treatment relative to vector control (Fig. 2B, one-way ANOVA, Bonferroni post-hoc test ^**p < 0.01). There was no significant change in luciferase activity from mevastatin-treated int13-G-transfected cells relative to vector control cells (Fig. 2B). These data provided the first indication that intron13 of ABCA7 carrying the major allele of rs3764650 may act as a promoter-enhancer capable of increasing ABCA7 expression. In the absence of mevastatin, the levels of luciferase activity from int13-G or int13-T transfected cells did not differ from the control (data not shown). To show that mevastatin treatment stimulates the expression machinery of ABCA7, qPCR analysis of SREBF2 gene expression from vehicle and mevastatin treated cells was analyzed. The relative gene expression of SREBF2 was increased in mevastatin-treated cells relative to vehicle control (Fig. 2C, t-test ^*p < 0.05). ABCA7 mRNA levels were undetected in HEK293 cells, as has been previously reported [18 , 21]. To further support the notion that int13-T possesses promoter enhancer capabilities, a hABCA7-encoding mammalian expression vector pcDNA3.1/hygro(+)-hABCA7 was co-transfected with equal molar ratios of either int13-T or int13-G (cloned within aforementioned pGL3-promoter vectors) in HEK293 cells and the levels of ABCA7 were measured. Western blot analysis shows that co-transfection with int13-T significantly increased the levels of ABCA7 protein as compared to co-transfection with int13-G or empty vector (Fig. 2D, one-way ANOVA, Bonferonni post-hoc test ^*p < 0.05, ^**p < 0.01). These data suggest that int13-T can enhance protein expression relative to co-transfection with int13-G or empty-vector. To provide additional support for the promoter enhancer capabilities of intron13 carrying the rs3764650-T allele on ABCA7 expression, the hABCA7-encoding mammalian expression vector was co-transfected with either int13-T or int13-G (cloned within the pCR4-TOPO vector, see Reporter gene assay in Materials and Methods section for generation) into HEK293 cells. Equal molar ratios of the hABCA7 plasmid, and the int13-T- or int13-G-encoding vectors were used for transfection. Western blot analysis shows that int13-T significantly increased the levels of ABCA7 protein as compared to the co-transfection with int13-G or empty vector (Fig. 2E, one-way ANOVA, Bonferonni post-hoc test ^*p < 0.05, ^**p < 0.01). These data provide further support that the intron13 sequence containing the major T allele can enhance expression. Next, to determine whether int13-T requires a promoter for functionality to initiate transcription, the CMV promoter upstream of the hABCA7 gene was removed from the hABCA7-encoding mammalian expression vector and this new plasmid (hABCA7-noCMV) was co-transfected into HEK293 cells with either int13-Tor int13-G (within the pCR4-TOPO vectors) and the levels of ABCA7 were measured. No ABCA7 expression was detected by western blot analysis from HEK293 cells co-transfected with the CMV-deleted hABCA7 plasmid and int13-T (Fig. 2F). This suggests that a promoter is necessary upstream of the ABCA7 gene for int13-T functionality and that int13-T likely functions as a promoter enhancer. To summarize, int13-T enhances the functioning of a downstream SV40 promoter within the pGL3-vector in the luciferase assay. In the subsequent co-transfection experiments, int13-T (in either pGL3-promoter or pCR4-TOPO vectors) enhances the expression of ABCA7 from the pcDNA3.1/hygro(+) vector. In the last set of co-transfection experiments, ABCA7 expression was not detected when int13-T was cotransfected with an hABCA7-encoding vector lacking the CMV promoter. These findings suggest that intron13 of ABCA7 carrying the rs3764650 major T allele can enhance expression.

The rs3752246 minor G-allele (ABCA7/Ala1527) increased secreted levels of Aβ₄₀and Aβ₄₂

The minor G-allele at rs3752246 (in exon33) creates a missense substitution in the ABCA7 protein replacing the glycine at position 1527 for an alanine residue (ABCA7/Ala1527). To determine the implications of ABCA7/Ala1527, the C>G point mutation was created by site-specific mutagenesis within the hABCA7-encoding mammalian expression vector pcDNA3.1/hygro(+)-hABCA7 (Fig. 3A), and various cellular phenotypes were analyzed after transfection of the ABCA7/Gly1527- and ABCA7/Ala1527-expressing vectors. Western blot analysis showed that both ABCA7/Gly1527 and ABCA7/Ala1527 were robustly expressed in HEK293 cells (Fig. 3B). These data show that the minor allele of rs3752246 does not alter the expression of the ABCA7 protein. To determine whether ABCA7/Ala1527 expression affected the levels of secreted Aβ, full-length ABCA7/Gly1527 or ABCA7/Ala127 were expressed in CHO cells stably expressing AβPP_Swe (CHO-AβPP_Swe) and the levels of Aβ₄₂ and Aβ₄₀ were measured in the media by commercial ELISA kits. This cell line was used because cells expressing AβPP_Swe generate a high level of Aβ relative to cells expressing normal AβPP [30], and because it has previously been reported that Abca7 expressed in CHO-AβPP_Swe cells result in decreased levels of secreted Aβ [25]. ABCA7/Gly1527-expressing CHO-AβPP_Swe cells exhibited significantly reduced levels of secreted Aβ₄₂ relative to vector-transfected cells (Fig. 3C, left panel, one-way ANOVA, Bonferroni post-hoc test ^***p < 0.001), which confirmed what was reported by Chan et al that overexpression of wild-type Abca7 reduced the levels of Aβ in CHO-AβPPs_we cells, as detected by western blot [25]. Another study showed that loss of Abca7 increased the levels of insoluble Aβ in the J20 AD mouse model [35]. The levels of secreted Aβ₄₀ from ABCA7/Gly1527-expressing CHO-AβPP_Swe cells were not significantly decreased relative to vector-transfected cells (Fig. 3C, right panel). Interestingly, ABCA7/Ala1527-expressing CHO-AβPP_Swe cells exhibited significantly increased levels of secreted Aβ₄₂ and Aβ₄₀ relative to ABCA7/Gly1527-expressing cells (Fig. 3C, one-way ANOVA, Bonferroni post-hoc test ^**p < 0.01 ^***p < 0.001). The levels of secreted Aβ₄₂ from CHO-AβPP_Swe cells expressing ABCA7/Ala1527 were also increased relative to vector control (Fig. 3C, left panel, one-way ANOVA, Bonferonni post-hoc test ^***p < 0.001). The inset shows the expression of hABCA7 from these ABCA7/Gly1527- and ABCA7/Ala1527-expressing cells. These data show that the major allele of rs3752246 decreases the levels of secreted Aβ₄₂ (as previously reported); while the minor allele of rs3752246 seems to increase the levels of secreted Aβ₄₂ and Aβ₄₀ in CHO-AβPP_Swe cells.

The rs3752246 minor G-allele (ABCA7/Ala1527) increased BACE1 activity

To try to understand the mechanism underlying the increase in secreted levels of Aβ₄₂ and Aβ₄₀ by ABCA7/Ala1527-expressing CHO-AβPP_Swe cells, it was ascertained whether BACE1 activity was affected by ABCA7/Gly1527 or ABCA7/Ala1527 expression. CHO-AβPP_Swe cells were transfected with either empty vector pcDNA3.1/hygro(+), full-length ABCA7/Gly1527 or full-length ABCA7/Ala1527 constructs, and 24 h later the cells were harvested and the EDANS emission (495–510 nm) was measured in each sample after incubation with fluorogenic BACE1 substrate IV using excitation at 335–355 nm. ABCA7/Ala1527-expressing CHO-AβPP_Swe cells exhibited a marginal but significant increase in BACE1 activity compared to the empty vector control (Fig. 4A, one-way ANOVA, Bonferroni post-hoc test ^*p < 0.05). To confirm these data, ABCA7/Gly1527 and ABCA7/Ala1527 were also expressed in another cell line that stably expressed AβPP_Swe, HEK293-AβPP_Swe. Indeed, ABCA7/Ala1527 expression resulted in a significant increase in BACE1 activity relative to ABCA7/Gly1527-expressing cells in HEK293 cells stably expressing AβPP_Swe (Fig. 4B, one-way ANOVA, Bonferroni post-hoc test ^**p < 0.01). These data suggest that the minor allele of rs3752246 in ABCA7 is associated with increased BACE1 activity.

To determine whether the increase in BACE1 activity was due to either a decrease in the non-amyloidogenic pathway, an increase in AβPP, or an increase in BACE1 protein levels, the levels of various proteins were determined by western blot analysis. The levels of sAβPPα, AβPP and BACE1 did not change in CHO-AβPP_Swe cells expressing ABCA7/Gly1527 or ABCA7/Ala1527 relative to control cells (Fig. 5). These data indicatethat the increase in BACE1 activity observed in ABCA7/Ala1527-expressing cells, relative to control or ABCA7/Gly1527-expressing cells, is neither associated with a reduction in the non-amyloidogenic pathway (as evidenced by no change in the levels of sAβPPα with ABCA7/Ala1527 expression), nor an increase in the levels of the AβPP or BACE1 proteins.

The minor G-allele of rs3752246 (ABCA7/Ala1527) abolishes a protein modification site

Analysis of the ABCA7 protein sequence by in silico methods revealed several putative and known N-glycosylation, phosphorylation and myristoylation sites (using Prosite: http://prosite.expasy.org/). Myristoylation is an irreversible protein modifi-cation whereby a myristoyl group is covalently attached to a glycine residue of target proteins [31, 36]. A myristoylation site was predicted at posi-tion 1527; the minor allele of rs3752246 (ABCA7/Ala1527) replaces the required glycine with an alanine thereby potentially removing the myristoylation site (Major allele ABCA7 : 1527-GALMAS-1532; minor allele ABCA7 : 1527-AALMAS-1532). Loss of this myristoylation site may have a role in the phenotypes we have described above. To test this hypothesis, HEK293-AβPP_Swe cells expressing either full-length ABCA7/Gly1527 or full-length ABCA7/Ala1527 constructs were treated with 1 mM HMA to inhibit myristoylation after which the levels of BACE1 activity and secreted Aβ were measured. HMA (1 mM) has been used in cells as a potent inhibitor of myristoyl-CoA-protein N-myris-toyltransferase (NMT), the enzyme that cataly-zes protein N-myristoylation [31]. Similar to what was observed in Fig. 4B, ABCA7/Ala1527-express-ing cells exhibited increased BACE1 activity relative to ABCA7/Gly1527-expressing cells under vehicle treatment (Fig. 6A, one-way ANOVA, Bonferroni post-hoc test ^*p < 0.05). ABCA7/Gly1527-expressing cells treated with 1 mM HMA exhibited a significant increase in BACE1 activity relative to vehicle treatment (Fig. 6A, one-way ANOVA, Bonferroni post-hoc test ^*p < 0.05). HMA treatment did not alter the levels of BACE1 activity in ABCA7/Ala1527-expressing cells (Fig. 6A). These data suggest that inhibiting myristoylation increased BACE1 activity in cells expressing ABCA7/Gly1527, that is presumably myristoylated at position 1527, but not in cells expressing ABCA7/Ala1527, which presumably does not harbor a myristoylation site at position 1527. The levels of secreted Aβ were also assessed after HMA treatment in HEK293-AβPP_Swe cells to determine whether a similar pattern was observed. As such, HEK293-AβPPswe cells were transfected with either ABCA7/Gly1527 or ABCA7/Ala1527, and 24 h later the cells were treated with 1 mM HMA and the levels of Aβ were assessed in the media after 24 h. Again, ABCA7/Ala1527-expressing cells exhibited increa-sed levels of Aβ₄₂ relative to ABCA7/Gly1527-expressing cells under vehicle treatment (Fig. 6B, left panel, one-way ANOVA, Bonferroni post-hoc test ^*p < 0.05), as seen in Fig. 3C. HMA treatment increased the levels of Aβ₄₂ and Aβ₄₀ relative to vehicle from cells expressing ABCA7/Gly1527, which presumably is myristoylated at position 1527, but not from cells expressing ABCA7/Ala1527, which presumably does not contain a myristoylation site at that position (Fig. 6B, one-way ANOVA, Bonferroni post-hoc test ^*p < 0.05). Importantly, HMA treatment (1 mM HMA, 5 mg/mL BSA) was not toxic to the HEK293-AβPPswe cells as compared to vehicle treatment (5 mg/mL BSA), as determined by CyQUANT Cell Proliferation Assay Kit. HEK-APPswe cells were transfected with pcDNA3.1/hygro(+), -hABCA7/Gly1527 or –hABCA7/Ala1527 and 24 h later, the cells were treated with HMA/BSA solution for an additional 24 h and harvested to measure cell viability. The HMA treatment (1 mM HMA, 5 mg/mL BSA) was not toxic to the cells since there was no appreciable difference in the cell number between HMA treatment and vehicle treatment (5 mg/mL BSA) (Fig. 6C). These data provide evidence to support the hypothesis that the lack of myristoylation of ABCA7 (either ABCA7/Gly1527 versus ABCA7/Ala1527, or ABCA7/Gly1527+vehicle versus ABCA7/Gly1527+HMA) may contribute to an in-crease in BACE1 activity and an increase in secreted Aβ levels.

Elimination of a post-translational modification site by minor allele rs3752246 (ABCA7/Ala1527)

In an effort to determine whether ABCA7/Gly1527 contains a protein modification that is eliminated in ABCA7/Ala1527, we generated short ABCA7 peptides spanning position 1527 and measured their molecular weight by gel electrophoresis and mass spectrometry. The ABCA7 peptides (each 96-amino acids long), spanning position 1527 of ABCA7, from either the major (96-Gly) or minor rs3752246 allele (96-Ala) were generated in HEK293 cells after transfection with pCMV-myc constructs encoding these sequences. The only difference in the protein sequence between the 96-Gly (MW: 12800.84) and the 96-Ala (MW: 12814.87) peptides was a glycine to an alanine substitution (Fig. 7A). When resolved by 15% acrylamide:bis-acrylamide gel-electrophoresis followed by western blot analyses, the 96-Ala peptide moved faster than the 96-Gly peptide, indicating a smaller molecular weight (Fig. 7B). A protein G kit was used to purify the myc-tagged peptides, and the recombinant peptides were eluted with 20 mM glycine pH 2.0, as described in the methods section. Mass spectrometry analysis of the purified peptides show that the mass of the 96-Ala peptide (12544.1) is 171 daltons smaller than that of the 96-Gly peptide (12715.1) (Fig. 7C), a difference of 57 daltons from the expected size of 228 daltons for a myristoyl group. Importantly, the mass spectrometer was calibrated with BSA; the measured molecular weight of BSA was 66406.4 daltons while the predicted molecular weight is 66463 daltons, a difference of 57 daltons. These results suggest that the minor allele of rs3752246 results in the elimination of a protein modification site which may contribute to the increase in BACE1 activity and the increase in secreted levels of Aβ₄₂ and Aβ₄₀.

DISCUSSION

Our findings show for the first time cell-phenotypes associated with the major and minor alleles of two ABCA7 SNPs that contribute to the risk of LOAD. The mechanisms of action seem to be different for the minor alleles of each of the two SNPs tested – potential omission of a promoter enhancer versus modification of protein function. A similar observation was reported for the CLU gene, for which AD risk could be independently associated with either rare coding variations or common variations in the regulatory region [37].

We provided evidence to suggest that intron13 carrying the major T-allele of SNP rs3764650 can enhance expression while the minor allele does not, supporting a loss of function disease mechanism associated with ABCA7 [38]. Genomic DNA may be coiled so that transcription factors (TFs) at the promoter and enhancer, which may be located several Kb away, interact to form a large protein complex. Our data suggest that TFs present in HEK293 cells are involved in the promoter enhancing ability of the ABCA7 intron13 sequence carrying the major T-allele of rs3764650. This implies that the ABCA7 DNA coils at least once to allow intron13 into the proximity of the ABCA7 promoter region where it binds to the TF complex and initiates transcription. In fact, according to available ChIP-seq data, SNP rs3764650 is in close proximity to an insulator regulatory element (chr19 : 995401-995800/hg18), which could block the interaction between an adjacent enhancer (chr19 : 995801-996200/hg18) and the promoter (Fig. 1). Insulator activity is thought to occur primarily through the 3D structure of DNA [39]. Therefore, differences in the 3D DNA structure caused by the risk haplotypes might potentially change the insulator function in transcriptional regulation. It is assumed that the sequence of intron13 may bring co-factor(s) or co-activator(s) to the transcription complex for initiating ABCA7 transcription. The putative promoter sequence of ABCA7 contains TF binding sites with roles in macrophage activation, hematopoiesis and lipid metabolism (AP-1, Ets-2, c-Myb, Sp1, SREBP, C/EBP, AML1a, LYF1, MZF1, GATA and Ikaros2) [23, 40]. Sequence analysis showed that intron13 contains potential binding sites for Ets-1 (GGAA [core motif], [41]), and may contain a Sp1 binding site [(G/T)GGGCGG(G/A)(G/A)(C/T), http://en.wikipedia.org/wiki/Sp1_transcription_factor]. Further investigation is warranted to determine which TF(s) or co-factor(s)/co-activator(s) bind to the potential promoter enhancer created by the major T-allele of rs3764650 in intron13 of ABCA7. Notably, cholesterol-depletion was not required to stimulate int13-T transcription in the co-transfection experiments because int13-T transcription was under the control of the lac promoter. What is clear is that, regardless of how int13-T is transcribed, int13-T enhances the expression. Our data is in agreement with reports stating that the T-allele of rs3764650 is associated with increased ABCA7 expression [42], and that rs3764650 minor-G allele correlated with decreased ABCA7 expression and increased neuritic plaques in humans [27]. However, the association between clinical presentation and ABCA7 levels has generated conflicting results. It was reported that higher ABCA7 levels were associated with more advanced cognitive decline [34 , 43]. Yet, it was suggested that increased ABCA7 expression reduces AD risk [17 , 24], and recently, that altered DNA methylation at the ABCA7 locus was associated with pathological AD diagnosis [44]. The authors argue that the increase in ABCA7 observed in AD may reflect an inadequate compensatory change [42] or potentially accumulation of dysfunctional protein.

There have been cell culture and in vivo reports showing that murine wild-type Abca7 expression prevents Aβ accumulation, and that Abca7 loss increases Aβ levels [25, 35]. In addition, a recent study demonstrated that suppression of ABCA7 in vitro or in vivo results in an elevation of amyloid production [45]. This phenomenon was also observed in our study, where expression of human wild-type ABCA7/Gly1527 resulted in reduced levels of secreted Aβ₄₂ and Aβ₄₀. In contrast, expression of the minor (risk) allele of rs3752246 (ABCA7/Ala1527) resulted in an increase in secreted Aβ₄₂ and Aβ₄₀. It is as yet unclear how ABCA7 alters Aβ levels; whether impaired Aβ production relates to the ABCA7 lipid efflux function, whether ABCA7 interacts and impacts AβPP processing, or whether increased Aβ clearance is due to the role of ABCA7 in phagocytosis [46]. We showed that the observed increase in Aβ levels from ABCA7/Ala1527-expressing cells relative to ABCA7/Gly1527-expressing cells was not due to a decrease in the non-amyloidogenic pathway or an increase in AβPP or BACE1 protein levels. Our data show that the increase in secreted Aβ levels from ABCA7/Ala1527 expressing cells was at least in part due to an increase in BACE1 activity. We speculate that if 60–70% transfection efficiency resulted in a 1.2-fold increase in BACE1 activity in ABCA7/Ala1527-expressing cells relative to control or ABCA7/Gly1527-expressing cells, then presumably 100% efficiency would result in a greater difference. Regardless, a 1.2-fold increase in BACE1 activity has been reported in the brains of AD versus non-demented patients [36, 47], thus there is biological implication to this observed significant difference.

Our results also suggest that the lack of myristoylation of ABCA7 at position 1527 plays a role in increasing BACE1 activity and secreted Aβ levels.It is curious that there are such post-translational modifications within ABCA7 because myristoylation of internal glycines are typically preceded by proteolytic cleavage [36]. We did not observe cleavage products via western blot analysis, nor are there known reports of myristoylation or proteolyticcleavage of ABC transporters even though in silico scans of other ABC transporters (such as ABCA1, a close homolog of ABCA7) also yield several predicted myristoylation sites (prosite.expasy.org/). However, it is worth noting that in a recent profile of co- and post-translationally N-myristoylated proteins, mitochondrial ABCB6 was listed in a human cell line [31]. Given that there is a difference in post-translational modification between ABCA7/Gly1527 and ABCA7/Ala1527, the lack of the myristate at position 1527 could influence protein-protein interactions that favor BACE1 activity. This hypothesis is based on a couple of facts. First, myristoylation modulates protein–protein interactions, and influences the activity of lipid metabolism enzymes [48, 49].Second, the low-density lipoprotein receptor-related protein has been shown to associate with both AβPP [50] and ABCA7 [51], suggesting that ABCA7 is in close proximity to AβPP, and its interacting proteins (e.g., BACE1). Interestingly, position 1527 of ABCA7 protein is very close to a missense mutation site in homolog ABCA1 found in Tangier disease and familial HDL deficiency [52] showing that this locus may be evolutionarily conserved and important for proper functioning of ABCtransporters.

Although Hollingworth et al. dubbed the non-synonymous rs37552246 SNP as being the unlikely functional variant based on in silico studies [9], our study provides evidence to show that both the rs3764650 and rs3752246 SNPs have functional relevance. Interactions between APOEɛ4, the strongest risk factor for LOAD, and both ABCA7 SNPs, rs3764650 and rs3752246, were associated with all three cognitive factor scores tested [53]. The AD-associated polymorphism rs3851179 near PICALM, another LOAD susceptibility gene that encodes a clathirin-coated pit accessory protein, was evaluated and found to modestly increase PICALM expression [54] much like we report for rs3764650 within ABCA7. Other ABC transporters have also shown to affect AβPP processing. ABCA2 favors BACE1-dependent cleavage of AβPP, while ABCG1 inhibits Aβ production [55, 56]. GWAS uncover associations between genetic variations and disorders, but findings cannot be directly applied to the prevention or treatment of disease; interventions can only be recommended once the full pathway of disease development is understood. As such, there is a need to supplement GWAS with studies such as ours to understand the mechanism behind the genetic variation and the disorder.

Footnotes

Acknowledgments

This study was supported by research grants - CIHR#109606 to W. Zhang; CIHR#106886 to E. Rogaeva and W. Zhang; CIHR#TAD 125698 to J. Woodgett and W. Zhang; NSFC#81061120527 toZ. Yang, and was conducted at the National ResearchCouncil Canada. The authors would like to thankDr. K. Ueda (Kyoto, Japan) for providing the pcDNA3.1/hygro(+) empty vector and the pcDNA3.1/hygro(+)-hABCA7 wild-type vector,Dr. A. Hill (Melbourne, Australia) for providing the CHO-AβPP_Swe cells, and Dr. D. Selkoe (Harvard Medical School, Boston, USA) for providing HEK293-AβPP_Swe cells.

Authors’ disclosures available online ().

References

Goedert

(2015) NEURODEGENERATION. Alzhei-mer’s and Parkinson’s diseases: The prion concept in relation to assembled Abeta, tau, and alpha-synuclein. Science 349, 1255555.

Di Paolo

, Kim

(2011) Linking lipids to Alzheimer’s disease: Cholesterol and beyond. Nat Rev Neurosci 12, 284–296.

Ghani

, Rogaeva

(2014) Autosomal Dominant Alzheimer’s Disease: Underlying Causes, Springer, London.

LaFerla

, Green

, Oddo

(2007) Intracellular amyloid-beta in Alzheimer’s disease. Nat Rev Neurosci 8, 499–509.

Eisenstein

(2011) Genetics: Finding risk factors. Nature 475, S20–S22.

Gatz

, Reynolds

, Fratiglioni

, Johansson

, Mortimer

, Berg

, Fiske

, Pedersen

(2006) Role of genes and environments for explaining Alzheimer disease. Arch Gen Psychiatry 63, 168–174.

Corneveaux

, Myers

, Allen

, Pruzin

, Ramirez

, Engel

, Nalls

, Chen

, Lee

, Chewning

, Villa

, Meechoovet

, Gerber

, Frost

, Benson

, O’Reilly

, Chibnik

, Shulman

, Singleton

, Craig

, Van Keuren-Jensen

, Dunckley

, Bennett

, De Jager

, Heward

, Hardy

, Reiman

, Huentelman

(2010) Association of CR1, CLU and PICALM with Alzheimer’s disease in a cohort of clinically characterized and neuropathologically verified individuals. Hum Mol Genet 19, 3295–3301.

Harold

, Abraham

, Hollingworth

, Sims

, Gerrish

, Hamshere

, Pahwa

, Moskvina

, Dowzell

, Williams

, Jones

, Thomas

, Stretton

, Morgan

, Lovestone

, Powell

, Proitsi

, Lupton

, Brayne

, Rubinsztein

, Gill

, Lawlor

, Lynch

, Morgan

, Brown

, Passmore

, Craig

, McGuinness

, Todd

, Holmes

, Mann

, Smith

, Love

, Kehoe

, Hardy

, Mead

, Fox

, Rossor

, Collinge

, Maier

, Jessen

, Schürmann

, Heun

, van den Bussche

, Heuser

, Kornhuber

, Wiltfang

, Dichgans

, Frölich

, Hampel

, Hüll

, Rujescu

, Goate

, Kauwe

, Cruchaga

, Nowotny

, Morris

, Mayo

, Sleegers

, Bettens

, Engelborghs

, De Deyn

, Van Broeckhoven

, Livingston

, Bass

, Gurling

, McQuillin

, Gwilliam

, Deloukas

, Al-Chalabi

, Shaw

, Tsolaki

, Singleton

, Guerreiro

, Mühleisen

, Nöthen

, Moebus

, Jöckel

, Klopp

, Wichmann

, Carrasquillo

, Pankratz

, Younkin

, Holmans

, O’Donovan

, Owen

, Williams

(2009) Genome-wide association study identifies variants at CLU and PICALM associated with Alzheimer’s disease. Nat Genet 41, 1088–1093.

Hollingworth

, Harold

, Sims

, Gerrish

, Lambert

, Carrasquillo

, Abraham

, Hamshere

, Pahwa

, Moskvina

, Dowzell

, Jones

, Stretton

, Thomas

, Richards

, Ivanov

, Widdowson

, Chapman

, Lovestone

, Powell

, Proitsi

, Lupton

, Brayne

, Rubinsztein

, Gill

, Lawlor

, Lynch

, Brown

, Passmore

, Craig

, McGuinness

, Todd

, Holmes

, Mann

, Smith

, Beaumont

, Warden

, Wilcock

, Love

, Kehoe

, Hooper

, Vardy

, Hardy

, Mead

, Fox

, Rossor

, Collinge

, Maier

, Jessen

, Rüther

, Schürmann

, Heun

, Kölsch

, van den Bussche

, Heuser

, Kornhuber

, Wiltfang

, Dichgans

, Frölich

, Hampel

, Gallacher

, Hüll

, Rujescu

, Giegling

, Goate

, Kauwe

, Cruchaga

, Nowotny

, Morris

, Mayo

, Sleegers

, Bettens

, Engelborghs

, De Deyn

, Van Broeckhoven

, Livingston

, Bass

, Gurling

, McQuillin

, Gwilliam

, Deloukas

, Al-Chalabi

, Shaw

, Tsolaki

, Singleton

, Guerreiro

, Mühleisen

, Nöthen

, Moebus

, Jöckel

, Klopp

, Wichmann

, Pankratz

, Sando

, Aasly

, Barcikowska

, Wszolek

, Dickson

, Graff-Radford

, Petersen

, Alzheimer’s Disease Neuroimaging Initiative, van Duijn

, Breteler

, Ikram

, DeStefano

, Fitzpatrick

, Lopez

, Launer

, Seshadri

, CHARGE consortium, Berr

, Campion

, Epelbaum

, Dartigues

, Tzourio

, Alpérovitch

, Lathrop

, EADI1 consortium, Feulner

, Friedrich

, Riehle

, Krawczak

, Schreiber

, Mayhaus

, Nicolhaus

, Wagenpfeil

, Steinberg

, Stefansson

, Snaedal

, Björnsson

, Jonsson

, Chouraki

, Genier-Boley

, Hiltunen

, Soininen

, Combarros

, Zelenika

, Delepine

, Bullido

, Pasquier

, Mateo

, Frank-Garcia

, Porcellini

, Hanon

, Coto

, Alvarez

, Bosco

, Siciliano

, Mancuso

, Panza

, Solfrizzi

, Nacmias

, Sorbi

, Bossú

, Piccardi

, Arosio

, Annoni

, Seripa

, Pilotto

, Scarpini

, Galimberti

, Brice

, Hannequin

, Licastro

, Jones

, Holmans

, Jonsson

, Riemenschneider

, Morgan

, Younkin

, Owen

, O’Donovan

, Amouyel

, Williams

(2011) Common variants at ABCA7, MS4A6A/MS4A4E, EPHA1, CD33 and CD2AP are associated with Alzheimer’s disease. Nat Genet 43, 429–435.

10.

Lambert

, Heath

, Even

, Campion

, Sleegers

, Hiltunen

, Combarros

, Zelenika

, Bullido

, Tavernier

, Letenneur

, Bettens

, Berr

, Pasquier

, Fievet

, Barberger-Gateau

, Engelborghs

, De Deyn

, Mateo

, Franck

, Helisalmi

, Porcellini

, Hanon

, European Alzheimer’s Disease Initiative I, de Pancorbo

, Lendon

, Dufouil

, Jaillard

, Leveillard

, Alvarez

, Bosco

, Mancuso

, Panza

, Nacmias

, Bossu

, Piccardi

, Annoni

, Seripa

, Galimberti

, Hannequin

, Licastro

, Soininen

, Ritchie

, Blanche

, Dartigues

, Tzourio

, Gut

, Van Broeckhoven

, Alperovitch

, Lathrop

, Amouyel

(2009) Genome-wide association study identifies variants at CLU and CR1 associated with Alzheimer’s disease. Nat Genet 41, 1094–1099.

11.

Morgan

(2011) The three new pathways leading to Alzheimer’s disease. Neuropathol Appl Neurobiol 37, 353–357.

12.

Naj

, Jun

, Beecham

, Wang

, Vardarajan

, Buros

, Gallins

, Buxbaum

, Jarvik

, Crane

, Larson

, Bird

, Boeve

, Graff-Radford

, De Jager

, Evans

, Schneider

, Carrasquillo

, Ertekin-Taner

, Younkin

, Cruchaga

, Kauwe

, Nowotny

, Kramer

, Hardy

, Huentelman

, Myers

, Barmada

, Demirci

, Baldwin

, Green

, Rogaeva

, St George-Hyslop

, Arnold

, Barber

, Beach

, Bigio

, Bowen

, Boxer

, Burke

, Cairns

, Carlson

, Carney

, Carroll

, Chui

, Clark

, Corneveaux

, Cotman

, Cummings

, DeCarli

, DeKosky

, Diaz-Arrastia

, Dick

, Dickson

, Ellis

, Faber

, Fallon

, Farlow

, Ferris

, Frosch

, Galasko

, Ganguli

, Gearing

, Geschwind

, Ghetti

, Gilbert

, Gilman

, Giordani

, Glass

, Growdon

, Hamilton

, Harrell

, Head

, Honig

, Hulette

, Hyman

, Jicha

, Jin

, Johnson

, Karlawish

, Karydas

, Kaye

, Kim

, Koo

, Kowall

, Lah

, Levey

, Lieberman

, Lopez

, Mack

, Marson

, Martiniuk

, Mash

, Masliah

, McCormick

, McCurry

, McDavid

, McKee

, Mesulam

, Miller

, Parisi

, Perl

, Peskind

, Petersen

, Poon

, Quinn

, Rajbhandary

, Raskind

, Reisberg

, Ringman

, Roberson

, Rosenberg

, Sano

, Schneider

, Seeley

, Shelanski

, Slifer

, Smith

, Sonnen

, Spina

, Stern

, Tanzi

, Trojanowski

, Troncoso

, Van Deerlin

, Vinters

, Vonsattel

, Weintraub

, Welsh-Bohmer

, Williamson

, Woltjer

, Cantwell

, Dombroski

, Beekly

, Lunetta

, Martin

, Kamboh

, Saykin

, Reiman

, Bennett

, Morris

, Montine

, Goate

, Blacker

, Tsuang

, Hakonarson

, Kukull

, Foroud

, Haines

, Mayeux

, Pericak-Vance

, Farrer

, Schellenberg

(2011) Common variants at MS4A4/MS4A6E, CD2AP, CD33 and EPHA1 are associated with late-onset Alzheimer’s disease. Nat Genet 43, 436–441.

13.

Seshadri

, Fitzpatrick

, Ikram

, DeStefano

, Gudnason

, Boada

, Bis

, Smith

, Carassquillo

, Lambert

, Harold

, Schrijvers

, Ramirez-Lorca

, Debette

, Longstreth

Jr , Janssens

, Pankratz

, Dartigues

, Hollingworth

, Aspelund

, Hernandez

, Beiser

, Kuller

, Koudstaal

, Dickson

, Tzourio

, Abraham

, Antunez

, Du

, Rotter

, Aulchenko

, Harris

, Petersen

, Berr

, Owen

, Lopez-Arrieta

, Varadarajan

, Becker

, Rivadeneira

, Nalls

, Graff-Radford

, Campion

, Auerbach

, Rice

, Hofman

, Jonsson

, Schmidt

, Lathrop

, Mosley

, Au

, Psaty

, Uitterlinden

, Farrer

, Lumley

, Ruiz

, Williams

, Amouyel

, Younkin

, Wolf

, Launer

, Lopez

, van Duijn

, Breteler

, CHARGE Consortium, GERAD1 Consortium, EADI1 Consortium (2010) Genome-wide analysis of genetic loci associated with Alzheimer disease. JAMA 303, 1832–1840.

14.

Reitz

, Jun

, Naj

, Rajbhandary

, Vardarajan

, Wang

, Valladares

, Lin

, Larson

, Graff-Radford

, Evans

, De Jager

, Crane

, Buxbaum

, Murrell

, Raj

, Ertekin-Taner

, Logue

, Baldwin

, Green

, Barnes

, Cantwell

, Fallin

, Go

, Griffith

, Obisesan

, Manly

, Lunetta

, Kamboh

, Lopez

, Bennett

, Hendrie

, Hall

, Goate

, Byrd

, Kukull

, Foroud

, Haines

, Farrer

, Pericak-Vance

, Schellenberg

, Mayeux

, Alzheimer Disease Genetics Consortium (2013) Variants in the ATP-binding cassette transporter (ABCA7), apolipoprotein E 4,and the risk of late-onset Alzheimer disease in African Americans. JAMA 309, 1483–1492.

15.

Shen

, Jia

(2016) An overview of genome-wide association studies in Alzheimer’s disease. Neurosci Bull 32, 183–190.

16.

Kaminski

, Orso

, Diederich

, Klucken

, Drobnik

, Schmitz

(2000) Identification of a novel human sterol-sensitive ATP-binding cassette transporter (ABCA7). Biochem Biophys Res Commun 273, 532–538.

17.

Kim

, Fitzgerald

, Kang

, Okuhira

, Bell

, Manning

, Koehn

, Lu

, Moore

, Freeman

(2005) Abca7 null mice retain normal macrophage phosphatidylcholine and cholesterol efflux activity despite alterations in adipose mass and serum cholesterol levels. J Biol Chem 280, 3989–3995.

18.

Abe-Dohmae

, Ikeda

, Matsuo

, Hayashi

, Okuhira

, Ueda

, Yokoyama

(2004) Human ABCA7 supports apolipoprotein-mediated release of cellular cholesterol and phospholipid to generate high density lipoprotein. J Biol Chem 279, 604–611.

19.

Wang

, Lan

, Gerbod-Giannone

, Linsel-Nitschke

, Jehle

, Chen

, Martinez

, Tall

(2003) ATP-binding cassette transporter A7 (ABCA7) binds apolipoprotein A-I and mediates cellular phospholipid but not cholesterol efflux. J Biol Chem 278, 42906–42912.

20.

Hayashi

, Abe-Dohmae

, Okazaki

, Ueda

, Yokoyama

(2005) Heterogeneity of high density lipoprotein generated by ABCA1 and ABCA7. J Lipid Res 46, 1703–1711.

21.

Ikeda

, Abe-Dohmae

, Munehira

, Aoki

, Kawamoto

, Furuya

, Shitara

, Amachi

, Kioka

, Matsuo

, Yokoyama

, Ueda

(2003) Posttranscriptional regulation of human ABCA7 and its function for the apoA-I-dependent lipid release. Biochem Biophys Res Commun 311, 313–318.

22.

Abe-Dohmae

, Ueda

, Yokoyama

(2006) ABCA7, a molecule with unknown function. FEBS Lett 580, 1178–1182.

23.

Iwamoto

, Abe-Dohmae

, Sato

, Yokoyama

(2006) ABCA7 expression is regulated by cellular cholesterol through the SREBP2 pathway and associated with phagocytosis. J Lipid Res 47, 1915–1927.

24.

Linsel-Nitschke

, Jehle

, Shan

, Cao

, Bacic

, Lan

, Wang

, Tall

(2005) Potential role of ABCA7 in cellular lipid efflux to apoA-I. J Lipid Res 46, 86–92.

25.

Chan

, Kim

, Kwok

, Hill

, Cappai

, Rye

, Garner

(2008) ATP-binding cassette transporter A7 regulates processing of amyloid precursor protein in vitro. J Neurochem 106, 793–804.

26.

Hughes

, Lopez

, Evans

, Kamboh

, Williamson

, Klunk

, Mathis

, Price

, Cohen

, Snitz

, Dekosky

, Kuller

(2014) Markers of cholesterol transport are associated with amyloid deposition in the brain. Neurobiol Aging 35, 802–807.

27.

Shulman

, Chen

, Keenan

, Chibnik

, Fleisher

, Thiyyagura

, Roontiva

, McCabe

, Patsopoulos

, Corneveaux

, Yu

, Huentelman

, Evans

, Schneider

, Reiman

, De Jager

, Bennett

(2013) Genetic susceptibility for Alzheimer disease neuritic plaque pathology. JAMA Neurol 70, 1150–1157.

28.

Logge

, Cheng

, Chesworth

, Bhatia

, Garner

, Kim

, Karl

(2012) Role of Abca7 in mouse behaviours relevant to neurodegenerative diseases. PLoS One 7, e45959.

29.

Lambert

, Ibrahim-Verbaas

, Harold

, Naj

, Sims

, Bellenguez

, DeStafano

, Bis

, Beecham

, Grenier-Boley

, Russo

, Thorton-Wells

, Jones

, Smith

, Chouraki

, Thomas

, Ikram

, Zelenika

, Vardarajan

, Kamatani

, Lin

, Gerrish

, Schmidt

, Kunkle

, Dunstan

, Ruiz

, Bihoreau

, Choi

, Reitz

, Pasquier

, Cruchaga

, Craig

, Amin

, Berr

, Lopez

, De Jager

, Deramecourt

, Johnston

, Evans

, Lovestone

, Letenneur

, Morón

, Rubinsztein

, Eiriksdottir

, Sleegers

, Goate

, Fiévet

, Huentelman

, Gill

, Brown

, Kamboh

, Keller

, Barberger-Gateau

, McGuiness

, Larson

, Green

, Myers

, Dufouil

, Todd

, Wallon

, Love

, Rogaeva

, Gallacher

, St George-Hyslop

, Clarimon

, Lleo

, Bayer

, Tsuang

, Yu

, Tsolaki

, Bossú

, Spalletta

, Proitsi

, Collinge

, Sorbi

, Sanchez-Garcia

, Fox

, Hardy

, Deniz Naranjo

, Bosco

, Clarke

, Brayne

, Galimberti

, Mancuso

, Matthews

, European Alzheimer’s Disease Initiative (EADI), Genetic and Environmental Risk in Alzheimer’s Disease, Alzheimer’s Disease Genetic Consortium, Cohorts for Heart and Aging Research in Genomic Epidemiology, Moebus

, Mecocci

, Del Zompo

, Maier

, Hampel

, Pilotto

, Bullido

, Panza

, Caffarra

, Nacmias

, Gilbert

, Mayhaus

, Lannefelt

, Hakonarson

, Pichler

, Carrasquillo

, Ingelsson

, Beekly

, Alvarez

, Zou

, Valladares

, Younkin

, Coto

, Hamilton-Nelson

, Gu

, Razquin

, Pastor

, Mateo

, Owen

, Faber

, Jonsson

, Combarros

, O’Donovan

, Cantwell

, Soininen

, Blacker

, Mead

, Mosley

Jr , Bennett

, Harris

, Fratiglioni

, Holmes

, de Bruijn

, Passmore

, Montine

, Bettens

, Rotter

, Brice

, Morgan

, Foroud

, Kukull

, Hannequin

, Powell

, Nalls

, Ritchie

, Lunetta

, Kauwe

, Boerwinkle

, Riemenschneider

, Boada

, Hiltuenen

, Martin

, Schmidt

, Rujescu

, Wang

, Dartigues

, Mayeux

, Tzourio

, Hofman

, Nöthen

, Graff

, Psaty

, Jones

, Haines

, Holmans

, Lathrop

, Pericak-Vance

, Launer

, Farrer

, van Duijn

, Van Broeckhoven

, Moskvina

, Seshadri

, Williams

, Schellenberg

, Amouyel

(2013) Meta-analysis of 74,046 individuals identifies 11 new susceptibility loci for Alzheimer’s disease. Nat Genet 45, 1452–1458.

30.

Citron

, Oltersdorf

, Haass

, McConlogue

, Hung

, Seubert

, Vigo-Pelfrey

, Lieberburg

, Selkoe

(1992) Mutation of the beta-amyloid precursor protein in familial Alzheimer’s disease increases beta-protein production. Nature 360, 672–674.

31.

Nadler

, Harrison

, Ashendel

, Cassady

, Geahlen

(1993) Treatment of T cells with 2-hydroxymyristic acid inhibits the myristoylation and alters the stability of p56lck. Biochemistry 32, 9250–9255.

32.

Kwiatkowska

, Frey

, Sobota

(2003) Phosphorylation of FcgammaRIIA is required for the receptor-induced actin rearrangement and capping: The role of membrane rafts. J Cell Sci 116, 537–550.

33.

Martin

, Vilas

, Prescher

, Rajaiah

, Falck

, Bertozzi

, Berthiaume

(2008) Rapid detection, discovery, and identification of post-translationally myristoylated proteins during apoptosis using a bio-orthogonal azidomyristate analog. FASEB J 22, 797–806.

34.

Tanaka

, Abe-Dohmae

, Iwamoto

, Fitzgerald

, Yokoyama

(2011) HMG-CoA reductase inhibitors enhance phagocytosis by upregulating ATP-binding cassette transporter A7. Atherosclerosis 217, 407–414.

35.

Kim

, Li

, Ruberu

, Chan

, Elliott

, Low

, Cheng

, Karl

, Garner

(2013) Deletion of Abca7 increases cerebral amyloid-beta accumulation in the J20 mouse model of Alzheimer’s disease. J Neurosci 33, 4387–4394.

36.

Cheng

, He

, Lee

, Yao

, Li

, Shen

(2014) High activities of BACE1 in brains with mild cognitive impairment. Am J Pathol 184, 141–147.

37.

Bettens

, Brouwers

, Engelborghs

, Lambert

, Rogaeva

, Vandenberghe

, Le Bastard

, Pasquier

, Vermeulen

, Van Dongen

, Mattheijssens

, Peeters

, Mayeux

, St George-Hyslop

, Amouyel

, De Deyn

, Sleegers

, Van Broeckhoven

(2012) Both common variations and rare non-synonymous substitutions and small insertion/deletions in CLU are associated with increased Alzheimer risk. Mol Neurodegener 7, 3.

38.

Ordonez-Gutierrez

, Anton

, Wandosell

(2015) Peripheral amyloid levels present gender differences associated with aging in AbetaPP/PS1 mice. J Alzheimers Dis 44, 1063–1068.

39.

Phillips

, Corces

(2009) CTCF: Master weaver of the genome. Cell 137, 1194–1211.

40.

Kaminski

, Piehler

, Schmitz

(2000) Genomic organization of the human cholesterol-responsive ABC transporter ABCA7: Tandem linkage with the minor histocompatibility antigen HA-1 gene. Biochem Biophys Res Commun 278, 782–789.

41.

von der Ahe

, Nischan

, Kunz

, Otte

, Knies

, Oderwald

, Wasylyk

(1993) Ets transcription factor binding site is required for positive and TNF alpha-induced negative promoter regulation. Nucleic Acids Res 21, 5636–5643.

42.

Vasquez

, Fardo

, Estus

(2013) ABCA7 expression is associated with Alzheimer’s disease polymorphism and disease status. Neurosci Lett 556, 58–62.

43.

Karch

, Jeng

, Nowotny

, Cady

, Cruchaga

, Goate

(2012) Expression of novel Alzheimer’s disease risk genes in control and Alzheimer’s disease brains. PLoS One 7, e50976.

44.

, Chibnik

, Srivastava

, Pochet

, Yang

, Xu

, Kozubek

, Obholzer

, Leurgans

, Schneider

, Meissner

, De Jager

, Bennett

(2015) Association of brain DNA methylation in SORL1, ABCA7, HLA-DRB5, SLC24A4, and BIN1 with pathological diagnosis of Alzheimer disease. JAMA Neurol 72, 15–24.

45.

Satoh

, Abe-Dohmae

, Yokoyama

, St George-Hyslop

, Fraser

(2015) ATP-binding cassette transporter A7 (ABCA7) loss of function alters Alzheimer amyloid processing. J Biol Chem 290, 24152–24165.

46.

Bacskai

, Kajdasz

, McLellan

, Games

, Seubert

, Schenk

, Hyman

(2002) Non-Fc-mediated mechanisms are involved in clearance of amyloid-beta in vivo by immunotherapy. J Neurosci 22, 7873–7878.

47.

Xiong

, Callaghan

, Jones

, Walker

, Lue

, Beach

, Sue

, Woulfe

, Xu

, Stanimirovic

, Zhang

(2008) Cholesterol retention in Alzheimer’s brain is responsible for high beta- and gamma-secretase activities and Abeta production. Neurobiol Dis 29, 422–437.

48.

Martin

, Beauchamp

, Berthiaume

(2011) Post-translational myristoylation: Fat matters in cellular life and death. Biochimie 93, 18–31.

49.

Rioux

, Pedrono

, Legrand

(2011) Regulation of mammalian desaturases by myristic acid: N-terminal myristoylation and other modulations. Biochim Biophys Acta 1811, 1–8.

50.

Rebeck

, Moir

, Mui

, Strickland

, Tanzi

, Hyman

(2001) Association of membrane-bound amyloid precursor protein APP with the apolipoprotein E receptor LRP. Brain Res Mol Brain Res 87, 238–245.

51.

Jehle

, Gardai

, Li

, Linsel-Nitschke

, Morimoto

, Janssen

, Vandivier

, Wang

, Greenberg

, Dale

, Qin

, Henson

, Tall

(2006) ATP-binding cassette transporter A7 enhances phagocytosis of apoptotic cells and associated ERK signaling in macrophages. J Cell Biol 174, 547–556.

52.

Tanaka

, Ikeda

, Abe-Dohmae

, Arakawa

, Sadanami

, Kidera

, Nakagawa

, Nagase

, Aoki

, Kioka

, Amachi

, Yokoyama

, Ueda

(2001) Human ABCA1contains a large amino-terminal extracellular domain homologous to an epitope of Sjogren’s Syndrome. Biochem Biophys Res Commun 283, 1019–1025.

53.

Engelman

, Koscik

, Jonaitis

, Okonkwo

, Hermann

, La Rue

, Sager

(2013) Interaction between two cholesterol metabolism genes influences memory: Findings from the Wisconsin Registry for Alzheimer’s Prevention. J Alzheimers Dis 36, 749–757.

54.

Parikh

, Fardo

, Estus

(2014) Genetics of PICALM expression and Alzheimer’s disease. PLoS One 9, e91242.

55.

Kim

, Rahmanto

, Kamili

, Rye

, Guillemin

, Gelissen

, Jessup

, Hill

, Garner

(2007) Role of ABCG1 and ABCA1 in regulation of neuronal cholesterol efflux to apolipoprotein E discs and suppression of amyloid-beta peptide generation. J Biol Chem 282, 2851–2861.

56.

Pahnke

, Frohlich

, Krohn

, Schumacher

, Paarmann

(2013) Impaired mitochondrial energy production and ABC transporter function-A crucial interconnection in dementing proteopathies of the brain. Mech Ageing Dev 134, 506–515.

Genetic Variations in ABCA7 Can Increase Secreted Levels of Amyloid-β 40 and Amyloid-β 42 Peptides and ABCA7 Transcription in Cell Culture Models

Abstract

Keywords

INTRODUCTION

MATERIALS AND METHODS

Materials

Cell lines and transfections

Expression analysis

Generation of vectors containing int13-T and int13-G

Luciferase reporter assay

Co-transfection experiments with int13-T or int13-G

Creating the variant clones

Western blotting analysis

Aβ determination

β-secretase activity

Generation and purification of ABCA7 peptides

Mass spectrometry

Inhibition of myristoylation

Statistical analysis

RESULTS

The rs3764650 major T-allele may increase transcription

The rs3752246 minor G-allele (ABCA7/Ala1527) increased secreted levels of Aβ40and Aβ42

The rs3752246 minor G-allele (ABCA7/Ala1527) increased BACE1 activity

The minor G-allele of rs3752246 (ABCA7/Ala1527) abolishes a protein modification site

Elimination of a post-translational modification site by minor allele rs3752246 (ABCA7/Ala1527)

DISCUSSION

Footnotes

Acknowledgments

References

The rs3752246 minor G-allele (ABCA7/Ala1527) increased secreted levels of Aβ₄₀and Aβ₄₂