UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)
The role of abca1 in atherosclerosis: lessons from in vitro and in vivo models
Singaraja, R.R.
Publication date
2003
Link to publication
Citation for published version (APA):
Singaraja, R. R. (2003). The role of abca1 in atherosclerosis: lessons from in vitro and in vivo
models.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)
and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open
content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please
let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material
inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter
to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You
will be contacted as soon as possible.
Chapter r
Thee variation of phenotypes in Tangier Disease and
FHAA is directly influenced by the nature of DNA
changess in ABCA1
Roshnii R. Singaraja
1, Henk Visscher
1, Erick R. James
1, Veronique Rigot
2,
Yannickk Hamon
2, Susanne M. Gee
1, Scott Newman
3, Giovanna Chimini
2,
andd Michael R. Hayden
11_
rhee Centre for Molecular Medicine and Therapeutics, University of British Columbia and Thee Children's and Women's Hospital, Vancouver, B.C., V5Z 4H4, Canada "Centree d'lmmunologie de Marseille Luminy, INSERM/CNRS et Universite de La Mediterranee,, Case 906, Pare Scientifique de Luminy, 13288, Cedex 09, Marseille, France
3
Xenonn Genetics Inc. 1650 Gilmore Way, Burnaby, B.C., Canada
Abstract t
Tangierr Disease (TD) and familial hypoaiphalipoproteinemia (FHA) are caused by several different mutationss in ABCA1, a gene that transports lipids across the plasma membranes of cells. These lipidss form HDL particles that are taken to the liver for excretion as bile in a process called reversee cholesterol transport (RCT). ABCA1 plays a critical role in the first steps of the RCT pathway.. Interestingly, patients with mutations in ABCA1 have differing clinical lipid phenotypes thatt vary in severity. We hypothesized that the mutations in ABCA1 had different consequences onn ABCA1 trafficking, localization to the cell surface and the ability to induce ApoA-l binding, andd that discrete clinical lipid phenotypes would be expected depending on the impact of the variouss mutations on ABCA1 function. To this end we created in vitro several naturally occurring missensee mutations in the ABCA1 gene. These mutations cause either TD in homozygous individualss or FHA in heterozygotes. We have determined the functional consequences of eachh variant in ABCA1 and correlated the genotypes with the phenotypes observed in patients.
Introduction n
Proteinss are classified as ABC transporters based on the presence of ATP binding domains, also knownn as nucleotide binding folds (NBFs). Genetic variations in the ABC genes are the cause off seveial J I J U I J L I J ui idd m u t a t i o n s IN ABC A I iead to i a n g i e r disease and f a m i l i a l hypoalphalipoproteinemiaa (1-3), both characterized by low to absent levels of HDL-C. The defectt in TD and FHA patients is in their inability to efflux intracellular cholesterol and phospholipids. .
Thee precise mechanism by which ABCA1 effluxes cholesterol and phosphocholine to HDL has untill recently been obscure. Three putative models for efflux by ABCA1 have been suggested (4). Inn the first, both cholesterol and phospholipids are together translocated to the outer plasma membranee leaflet by ABCA1. In the second model, ABCA1 causes phospholipids to be transported too the plasma membrane where it associates with ApoA-l, thus forming pre-B HDL particles. These particless then elicit cholesterol efflux from specific free cholesterol rich plasma membrane microdomains.. The third model is a hybrid of the first two, in which cholesterol and phospholipid effluxess are normally coupled, but can be separated if the free cholesterol rich microdomains are farr apart. Several recent studies have shown that indeed, the translocation of phospholipids and cholesteroll by ABCA1 are separable functions. When media collected from ABCA1 expressing cellss that have been treated with cyclodextrm to remove cellular free cholesterol is added to cells thatt do not contain ABCA1, free cholesterol efflux is still elicited (5), indicating that cholesterol effluxx to ApoA-l is a secondary step, not requiring functional ABCA1.
ABCA11 overexpressing BAC transgenic mice crossed to ApoE-/- mice showed a significant decreasee in atherosclerosis with a minimal increase in HDL-C levels (6, 7). However, the HDL isolatedd from the ABCAI+ApoE-/- mice contained higher amounts of phospholipids, and was aa significantly better acceptor of effluxed free cholesterol, suggesting that the presence of increasedd ABCA1 resulted in an increase in ApoA-l-phospholipid particles, and further supporting thee fact that ApoA-l-phospholipid complexes formed following ABCA1 mediated efflux are efficientt cholesterol acceptors.
Inn the absence of ABCA1 function, lipidation of lipid poor HDL is markedly reduced, causing thee hypercatabolism of HDL particles, and a subsequent reduction in plasma HDL-C levels. This reductionn in lipid efflux results in the accumulation of cholesterol esters observed in peripheral cellss of patients.
Thee aim of the present study was two-fold. First, by generating the precise mutations observed inn TD and FHA patients, we performed genotype/phenotype correlations that might shed furtherr light on ABCAI's function and on the molecular and cellular basis for the phenotypic variabilityy observed in patients with these diseases. Second, although ABCA1 is known to be essentiall for the efflux of lipids across membranes, the exact mechanism by which this occurs stilll remains to be elucidated By determining the biochemical defects caused by ABCA1 variants
inn vitro, and the degree to which each pathogenic variant impaired ABCA1 function, we hopedd to gain further understanding into the structure function relationships of ABCA1. Thuss far, at least fifty mutations in the ABCA1 gene have been identified (1-3, 8-22). These includee 23 missense mutations, 6 nonsense mutations and 21 insertions or deletions. Forty-eightt of the reported mutations occur in exonic regions, whereas 2 occur in introns. These includee a mutation in intron 2 that leads to abnormally spliced proteins and 2 large deletions coveringg introns 12 through 14, and 16 through 3 1 , respectively. The mutations in ABCA1 are nott randomly distributed. Indeed, of all the mutations, 5 are in the first NBD (aa 926-1065), 12 aree in the first extracellular loop {aa 44-640), and 5 are in the c-terminus (aa 2078-2261). A numberr of these mutations, especially those in the ATP binding cassettes occur in amino acids thatt are highly conserved in evolution over 600-1200 million years (23, 24), thus confirming theirr functional relevance. Interestingly, a third of all the mutations are found in the t w o large extracellularr loops, and a quarter of all mutations are found in the NBD domain regions. By contrast,, only one mutation is found in the transmembrane domains that comprise 11.7% of thee total ABCA1 sequence.
Patientss were classified phenotypically by virtue of the extent of lowering of their HDL values comparedd to controls. We have generated in vitro 15 missense mutations described in patients thatt lead to both TD and FHA. First, mutations that represented each patient phenotypic group basedd on absolute HDL-C levels and HDL-C levels relative to the LRC (Lipid Research Clinic) controll populations were chosen. The mutations were also selected on the basis of their locationn on the ABCA1 gene with mutations in any potential functional domain being chosen. Ass an initial step, we performed Western i m m u n o b l o t t i n g and determined intracellular localizationn of the mutants in order to assess the synthesis, stability, and transport of the ABCA11 variants. Since ABCA1 transports lipids to ApoA-l, the apolipoprotein component of HDL,, and since some studies have shown that ApoA-l can be directly cross-linked to cell surfacess with ABCA1 (25), we assessed the ability of the variants to induce the binding of ApoA-l.. In order to also more precisely determine the impact of variants on lipid efflux, we determinedd the ability of each ABCA1 variant to induce efflux of both cholesterol and phosphocholine.. These experiments reveal a spectrum of biochemical defects and provide a mechanismm for the observed variability in patient phenotypes in both TD and FHA and also providee further insights into the function of specific sequences of the ABCA1 gene.
Experimentall procedures
Generationn of ABCA1 variant constructs
pcDNA3-ABCA11 was generated by RT-PCR amplification of human ABCA1 from liver RNA, and cloningg into the pcDNA3 vector (tnvitrogen). The sequence of ABCA1 was verified and the
stopp codon of ABCA1 was replaced with the EGFP (Clontech) gene as a c-terminal fusion. This fusionn did not cause mislocalization of ABCA1 nor alter its lipid efflux capabilities (data not shown).. All the variants in the ABCA1 gene were generated by PCR based site-directed mutagonesiss as previously described (20) in the pcDNA ABCA1-EGFP vectoi, and all clones weree fully sequenced to ensure that no variations were generated except for those that were desired.. The entire ABCA1 cDNA containing the variants were then cloned into the pcDNA5-FRTT vector (Invitrogen) in order to generate stable cell lines expressing the variants.
Constructionn of stable cell lines expressing ABCA1 variants
2933 Flip-In cells were cultured in DMEM, 10% fetal bovine serum, 20mM L-glutamine and 50U/mll penicillin/streptomycin. Polyclonal ABCA1 stable cell lines were generated by co-transfectmgg pcDNA5/FRT-ABCA1 and pOG44 (Invitrogen) using Fugene-6 (Roche Molecular Biochemicals)) or Lipofectamine (Invitrogen) in growth medium according to manufacturers recommendations.. Seventy-two hours after transfection, hygromycm B (Invitrogen) was added too a final concentration of either 40 (.ig/ml or 50 (.ig/ml, and the media changed every 3-4 days untill hygromycin-resistant colonies were clearly evident. The colonies then were pooled and evaluatedd for ABCA1 expression by Western blot analysis. A control hygromycin-resistant cell linee was generated by co-transfecting pOG44 with the empty pcDNA5/FRT vector.
Proteinn isolation and Western blotting
Cellss were lysed in cold RIPA buffer containing a protease inhibitor cocktail (Roche), and spun att 14000rpm for 5 minutes at 4eC. Supernatants were quantitated for protein using the Lowry assay,, and 40 - 60pg of protein was separated on 7.5% acrylamide gels, and transferred onto polyvinylidenee difluoride (PVDF) membranes (Millipore), ABCA1 was immunodetected using ann anti-ABCA1 monoclonal antibody generated to its c-terminus. Anti-glyceraldehyde phosphate dehydrogenasee (GAPDH) antibody (Chemicon) was used to ensure that equivalent amounts of proteinn were loaded.
Celll surface biotinylation assays
10cmm plates containing confluent cell lines expressing the various ABCA1 mutants were washed thricee with cold PBS containing 1mM MgC! and 0.1 mM CaCL EZ-!ink sulfo NHS LC biotin (2 mg/ml)) was added to the cells and incubated for an hour on ice. Cells were then washed with coldd PBS, and the reaction quenched with 10mM glycine. Cells were washed again and lysed inn RIPA buffer. Immunoprecipitation was performed on the ABCA1 protein using the AC10 monoclonall anti-ABCA1 antibody, and proteins were separated on a 7.5% acrylamide gel. Proteinss were transferred onto Immobilon-P (Millipore) membranes and probed with streptavidin thatt was conjugated to HRP (Horseradish peroxidase) (Biorad).
Determinationn of the glycosylation status of ABCA1
Frozenn cell pellets of stable cell lines expressing various ABCA1 mutations were resuspended inn lysis buffer (1x PBS, 1 % TrintonX-100, 2% Glycerol) and incubated on ice for 10 minutes. Thee lysates were then centrifuged at 14000rpm for 1 minute A Lowry protein assay was performedd on the resulting supernatant, and 20ug of protein was digested with either EndoH orr PNGase (New England Biolabs) according to manufacturers specifications- The digested lysatess were separated on NuPAGE 3-8% Tris-Acetate gels (Invitrogen) and transferred onto
mmobilion-PP membranes (Millipore) for Western analysis. The blots were probed with the monoclonall anti-ABCA1 antibody AC10 (27).
ApoA-ll binding and FACS analysis
Recombinantt ApoA-l (Calbiochem) was coupled to the Cy5 fluorochrome with the " Cy55 monofunctional dye 5-pack (Amersham Pharmacia Biotech) as previously described (28, 29),, Binding was performed in the presence of 20 (.ig/ml of Cy5-ApoA-l in binding buffer (10mMM HEPES pH 7.4, 1.8 mM CaCl., I m M MgCI., 5mM KCI, 150 mM NaCI) for 1 h at 4'C on cellss detached by mild trypsinization (0.05% Trypsin-EDTA) (Life Technologies). At the end of thee incubation, cells were rapidly washed in cold binding buffer. Flow cytometric recordings weree performed on a FACScalibur (Becton Dickinson) and analysed by CellQuest software (CellQuestt 3.3, BD Biosciences). The cells were manually gated for ABCA1 expression as reflected byy EGFP Relative Fluorescence Intensity (RFI). These settings were kept consistent for each cell linee during each experiment. Binding data were calculated as the geometric mean of Cy5-ApoA-ll RFI on the selected cell populations. The induced binding was calculated as the point too point difference between the EGFP positive and negative cells divided by protein expression ass measured by EGFP RFI of the EGFP positive cells and expressed as percent of the binding inducedd by the wild type ABCA1 construct.
Phospholipidd efflux essay
Cellss were plated onto 24 well dishes in D M E M / 1 0 % FBS/50U/ml penicillin-streptomycin / 20mMM L-glutamine. 24 hours later, lOpCi/ml of choline chloride (Perkin-Elmer) was added to the cellss in media. After 16 hours, cells were equilibrated in DMEM/0.2% defatted BSA for one hour, followedd by efflux for four hours to 20ug/ml ApoA-l (Calbiochem). Supernatants were removed andd spun at 8000rpm for 4 minutes. 500ul of 0.2% SDS was added to each well for cell lysis. 400ull of the centrifuged supernatant and 500ul of the cell lysate were added to individual glass tubes.. 1.5ml of a 1:2 mix of chloroform;methanol was added to each glass tube and vortexed for 11 minute each. The tubes were incubated for one hour, followed by the addition of 250ul each off chloroform and water. Tubes were vortexed again, spun at 3000rpm for 15 minutes, and the lowerr phase extracted, placed in eppendorf tubes, and dried down. 200ul of methanol and 400u!! of scintillant were added to each tube, and the radioactivity quantified.
Cholesteroll efflux essay
Cellss were plated onto 24 well dishes in the same media as above. 24 hours later, 1uG/ml of tritiatedd cholesterol (Perkin Elmer) was added to each well in a volume of 500pl of media. 16 hourss later cplk were pqi lilihr^tpd by +hr ^-Idition " f SOOpi ^f H'MEM/0 2% defatted BSA f-_r onee hour, followed by efflux to 20pg/ml of ApoA-l (Calbiochem) for four hours. Supernatant wass removed and centrifuged at 8000rpm for 4 minutes. 200ul of 0.1N NaOH was added to thee cells for lysis. 200ul each of the cleared supernatant and cell lysate were added to individual scintillationn vials containing 400pl of scintillant, and radioactivity quantified.
Confocall microscopy
ABCA1ABCA1 stable cells lines were grown on 0 . 1 % gelatinized coverslips. Cells on coverslips were thenn washed three times with ice cold PBS and fixed with 4 % paraformaldehyde in PBS for 20 minutes.. Cover slips were again washed thrice with PBS and mounted on glass slides with 1-22 drops of Mowiol solution (heated to 55'"C). Images were captured using a Leica TS100 confocall microscope.
Statistics s
Alll statistical analyses were performed using the two-tailed t test using the program GraphPad Prismm (GraphPad Prism version 3 for Windows, GraphPad Software, San Diego, CA, USA).
Results s
TDD and FHA patients display phenotypic variation in levels of plasma HDL-C
Bothh TD and FHA patients show a wide variety of phenotypic defects, the most common of whichh are peripheral neuropathy, hepatosplenomegaly, premature coronary artery disease and variabilityy in plasma HDL-C levels. In order to perform genotype-phenotype correlations, we obtainedd plasma HDL-C levels from every TD and FHA patient harboring missense mutations thatt has been described in the literature. Correlation of genotypes with other phenotypic variationss proved impossible since there was no consistency in the reporting of other phenotypes inn either TD or FHA to provide sufficient sample sizes and representation.
Wee compared the reported HDL-C levels of patients harbouring 15 different missense mutations (Figuree 1) to HDL-C levels of age and sex matched LRC control subjects. In those with FHA that aree heterozygous for mutations in the ABCA1 gene, three distinct phenotypic groups emerged, onee in which HDL-C levels were about 50% of those of age and sex matched controls (R587W, Q597R,, AL693, N935S, A1046D, C1477R, S1506L, N1800H, R2081W, P2150L), one in which HDL-CC levels were higher than 7 0 % of the levels of age and sex matched controls (A255T, W590S,, T929I), and one in which HDL-C levels were significantly below the expected 50% of thee levels for age and sex matched controls (30.4% of age and sex matched LRC controls)
B B A B C A 1 1 2000 kDa-». o o + J J c c o o O O 4-1 1 £ £ h--lO O ir> > (N N < < 4 H » » ? ? 1— — 00 0 I D D DC C CO O £ £ rr r r--CT) r--CT) i n n O O ro o 0 1 1 t o o I I < < CO O m m n n en n Z Z
O B B
Q Q <o o t t r> > T — — < < H H o> > o o *~ *~ Ï» » CO O CM M - II 5 _j ö CD D O O CO O o o OO e/) o o m m CM M 0. . c c o o o o X X o o o o co oGAPDH H
Figuree 1. Missense mutations in ABCA1 and generation of stable cell lines. (A ) A schematic diagram of the
ABCA11 protein showing the location of the mutations that were studied. Missense mutations described in the ABCA11 gene in patients affected with Tangier disease and Familial hypoalphalipoproteinemia were generated in vitro.. These specific missense mutations were chosen because of the variation in lipid phenotypes observed in the patients.. Stable cell lines were generated in human embryonic kidney 293-Flp-in cells with GFP tagged mutant ABCA1.. (B) The expression of ABCA1 protein in various stable ABCA1 cell lines. Since the Flp-in system generates onlyy single integrants at a defined Flp recombinase insertion site, all cell lines should have one copy of the ABCA1 genee integrated into the genome at a defined site. Any variability in protein expression therefore should arise fromm increased protein turnover and instability caused by the mutations.
( M 1 0 9 1 T )) (Table 1). Of t h o s e w i t h TD, t w o g r o u p s w e r e o b s e r v e d , o n e w i t h t h o s e s h o w i n g
a l m o s tt n o HDL-C levels ( R 5 8 7 W , N 9 3 5 S , N 1 8 0 0 H ) , a n d a n o t h e r w i t h p a t i e n t s displaying > 1 0 %
HDL-CC levels w h e n c o m p a r e d t o age a n d sex m a t c h e d c o n t r o l s ( A 2 5 5 T )
H e t e r o z y g o u ss p a t i e n t s w i t h - 5 0 % o f n o r m a l H D L - C l e v e l s c o n t a i n m u t a n t A B C A 1 a l l e l e ss w i t h null a c t i v i t y
W ee h y p o t h e s i z e d t h a t m u t a t i o n s in A B C A 1 t h a t g e n e r a t e alleles t h a t retain n o residual activity
w o u l dd c o n f e r HDL-C levels t h a t are a b o u t 5 0 % of t h o s e of c o n t r o l s w h e n p a t i e n t s h e t e r o z y g o u s
f o rr A B C A 1 m u t a t i o n s are s t u d i e d . This 5 0 % of n o r m a l HDL-C level w o u l d be g e n e r a t e d f r o m
t h ee w i l d - t y p e allele. In order t o t e s t if i n d e e d t h e m u t a t i o n s in t h e s e p a t i e n t s resulted in null
alleles,, a n d t o p e r f o r m g e n o t y p e - p h e n o t y p e c o r r e l a t i o n s , w e d e t e r m i n e d t h e b i o c h e m i c a l d e f e c t s
Tablee 1. Comparison of patient HDL-C levels to those of age and sex matched controls from the Lipid Research
Clinicc (LRC) database.
'/u-zoo ut normal iiDL-L levels (allelee with residual function) Mutation n
A255T T W590S S T929I I
ABCA11 HETEROZYGOTES
4b-/Ü7oo of normal HDL-C levels -30% of noimal HDL-C levels (allelee with loss of function) (allele with dominant negative function)
%% of LRC controls s 75.94% % 822 62% 75.57% % Mutation n R587W W P2150L/R587W W Q597R R AL693 3 N935S S A1046D D C1477R R R2081W/D1289N N %% of LRC controls s 65.20% % 47.48% % 60.38% % 57.05% % 69.79% % 59.90% % 60.99% % 58.64% % Mutation n M1091T T %% of LRC controls s 30.40% % ABCA11 HOMOZYGOTES >10%% of normal HDL-C levels
(alleless with residual activity)
-10%% of normal HDL-C levels (alleless with no activity) Mutation n A255T T %% of LRC controls s 13.33% % Mutation n R587W W N935S S N1800H H %% of LRC controls s 6.25% % 2.62% % 3.38% %
underlyingg the phenotypes observed in these heterozygous patients. ABCA1 is known to act at thee piasma membrane, inducing binding of ApoA-l to specific domains on cell surfaces, and permittingg the efflux of cholesterol and phosphocholine across plasma membranes to ApoA-l richh acceptor particles. We hypothesized that in conditions where mutant ABCA1 showed no activity,, ABCA1 would not be localized as normal to the plasma membrane and therefore wouldd induce relatively low or absent levels of ApoA-l binding. Alternatively, if defects in the mutantss did result in the localization of ABCA1 to the plasma membrane, reduced induction of ApoA-ll binding could still be observed caused by a mutation that specifically disrupts the bindingg sites or alters conformations necessary for interaction with ApoA-l.
Ass an initial step, we determined the intracellular localization of each of the mutants using GFP taggedd ABCA1 transfected into Hela cells. When localization of the ABCA1 proteins was assessed,, six mutants, R587W, Q597R, AL693, N935S, N1800H and R2081W showed no GFP signall at the plasma membrane, and instead displayed only intracellular accumulation of GFP (Figuree 2A), manifesting as trafficking defective proteins. By contrast C1477R, D1289N and P2150LL alleles showed a subcellular distribution that was similar to wild-type ABCA1 whereas
A1046DD and S1506L showed partially reduced localization at the plasma membrane (Figure 2A). Inn order to further characterize the localization of some of the mutants, we performed assays t oo determine the glycosylation status of the mutants (Figure 2B). Wild-type ABCA1 is glycosylated,, and ABC1 that is retained in the endoplasmic reticulum (ER) is sensitive to EndoH digestion,, and ABCA1 that makes it to the medial and trans-Golgi is EndoH resistant. Endoglycosidasee H cleaves high mannose and some hybrid oligosaccharides from N-linked glycoproteinss (30), and the site of complex oligosaccharide addition is the ER. In the medial andd trans Golgi, these complex oligosaccharides are further modified to high mannose, complex oligosaccharidess that are EndoH insensitive. Alternately, ABCA1 at both the ER and the Golgi iss sensitive to PNGase digestion, since PNGase digests both simple and complex mannose groupss (30, 31). We observed that part of EndoH digested wild-type ABCA1 shows a shift in molecularr weight, indicating some EndoH sensitivity, and reflecting the fact that wild-type ABCA11 is partially localized at the ER and also is found in the medial and trans Golgi and the plasmaa membrane, conferring an EndoH resistant band. R587W, Q597R, \L693 and N935H all showw EndoH sensitivity, indicating that they do not exit the ER, and confirming the intracellular dataa showing that they do not reach the plasma membrane. C1477R is distributed at the ER andd makes it to the medial and trans Golgi since t w o bands are observed for this construct confirmingg the GFP localization data showing that this mutant is transported to the plasma membrane.. S1506L shows a digestion pattern essentially similar to wild-type ABCA1, reflecting thee fact that some GFP-S1506L is localized at the plasma membrane. R2081W, which by GFP localizationn did not show staining at the plasma membrane, by EndoH digestion showed a portionn that was EndoH resistant, and a portion that was EndoH sensitive. This may imply that
Figuree 2. Biochemical characterization of ABCA1 mutations that generate null alleles. (A) Stable cell lines
containingg different mutants were assessed for GFP localization using confocal microscopy. Wild-type (Wt) ABCA11 showed localization at the plasma membrane and was also intracellular^ distributed. Cell lines R587W, Q597R,, AL693, N935S, N1800H and R2081W showed no plasma membrane localization, and instead were only localizedd to intracellular sites A1046D and S1 506L showed reduced GFP localization at the plasma membtane, andd C1477R, D1289L and P2150L showed localization at the plasma membrane and at intracellular sites (B) Determinationn of the glycosylation status of ABCA1 variants Wild-type and mutant ABCA1 protein was immunoprecipitatedd and digested with either EndoH or PNGase in order to determine their localization and glycosylationn status. EndoH sensitivity results in a downward shift of the ABCA1 protein band and indicates that thee protein is present in the ER and therefore only contains high mannose residues. Resistance to EndoH digestionn indicates that the protein has transited through the medial and trans-Golgi, the sites of complex mannosee sugar additions Sensitivity to PNGase indicates that the protein is glycosylated PNGase digests both simplee and complex sugar groups. Wt ABCA1 shows EndoH and PNGase sensitivity, indicating that it is glycosylated, andd that it is found at the ER and also has reached the medial and trans Golgi. R587W, Q597R, AL693 and N935S showw only EndoH sensitivity, which indicates that they are localized in the ER C1477R, S1 506L and R2081W showw both EndoH sensitive and resistant bands indicating that they are found in the ER and also are at the media1 andd trans Golgi. (C) Cell surface biotmylation assays were performed on selected mutants to confirm localization dataa As observed with the GFP localization and EndoH assays, R587W and AL693 showed no ABCA1 at the plasmaa membrane However, C1477R showed normal ABCA1 localization at the membrane when compared to wild-typee ABCAl.
B B
01 1 re re>, ,
_ J J **#* * -* -* W T T o o TO TO t/> >>< <
_ l l + + T T C C T3 3 C C LU U + + T T 0 0 c c LU U i i T T 0 0 T3 3 c c LU U--i --i X X 0 0 T3 3 C C LU U + + O O IA A TO TO ü ü a. a.
H H
+ + 0) ) VI I O O 2 2 CL L i i <D D V) ) re re (5 5 2 2 a. .1^1 1
01 1 to o CD D 2 2 0 . . 01 1 TO TO>. .
—I Ito. to.
+ + T T 0 0 T3 3 C C UJ J i i 0 0 C C LU U^ N , ,
R587W W o o TO TO^ ^
_ i i + + T T O O c c UJ J i i T T O O a a c c LU U + + O O w w O O 2 2 o_ _ + + a) ) i/> > O O 2 2 a a i i o o CD D:>: :>:
a. . t t a a w w O O 2 2 CL L 0) ) re re VI I^ ^
_ j ja: :
+ +
T T 0 0 T3 3 C C LU U M tt ' Q597R R a; ; re re>> >
_ l l + + T T 0 0 T3 3 C C LU U 1 1 T T 0 0 - D D C C LU U :: : t 1 1 T T 0 0 T3 3 C C LU U + + CJ J V) ) O O 2 2 a. . i + + <D D w wo o
2 2 CL L I I Ü Ü V) )o o
2 2 a. .i i
QJ J(/) )
o o
2 2 a a < j j TO TO>, ,
_! !
++ I XX X OO 0 DD - a CC C LUU LUfa fa
mmmm f ' I i i AL693 3 01 1 TO TO^ ^
_ J J ++ 1 XX X 00 o T33 T3 cc c LUU LU ++ I a)) o V)) V) roro a; OO O 22 2 a.. a.IJ J
+ + 0) ) VI I re re (3 (3 2 2 C L Lfr..* fr..*
I-- :*4 S) S) VI I TO TO CD D 2 2 CL L N935S S C1477R R S1506L Lmmmm *»»*
R2081W W 3 0 0--< --<
OO 2 0 0-< -<
c c 'S S 3 3 %% 1 0 0 -0.2 2* *
0.03 3 r ^ - ii 0.006II In
Ctr l l w t t R587 W W AL69 3 3 C1477 R Rthee R2081W mutant enters the Golgi, but does not transit to the plasma membrane and is sequesteredd in a compartment between the late Golgi and the plasma membrane {Figure 2B). Wee next performed cell surface biotmylation assays on some of the constructs in order to confirmm our localization findings. When the ratio of total cellular ABCA1 to cell surface ABCA1 wass measured, in agreement with our previous data, we found that both R587W and AL693 showedd a significant reduction in cell surface ABCA1 expression. However, C1477R, which wass found to be at the plasma membrane both by GFP staining, and by EndoH assays, was found att the plasma membrane in similar quantities as the wild-type ABCA1 protein (Figure 2C). Wee next assessed the extent to which each of these mutants elicited cell surface ApoA-l binding.. Varying amounts of cell surface ApoA-l binding was observed with these constructs. Interestingly,, C1477R, which localized to the plasma membrane elicited almost no ApoA-l bindingg , n=3, p= 0.0008 ) when compared to controls (Figure 3A), Constructs A1046DD and S1506L both showed partial ability to induce cell surface ApoA-l binding
,, n=3, p= 0.02 and , n=3, p= 0.004 respectively) (Figure 3A), indicating somee localization of the mutant protein at the plasma membrane. Mutants R587W, Q597R, AL693,, N935S, N1800H, and R2081W all showed significantly reduced cell surface ApoA-l bindingg (R587W, , n=3, p= 0.006; Q597R, 17.4 14.0%, n=3, p= 0.009; AL693,
,, n=3, p= 0.008; N935S, , n=3, p= 0.01; N1800H, 1 5.5%, n=3, p=0.01;; R2081W, , n=3, p= 0.02) {Figure 3A) which correlated with the defects in traffickingg of these mutant ABCAIs at the plasma membrane.
Alll mutants with the exception of D1289N and P2150L showed a significantly reduced choline andd cholesterol efflux capability (Figure 3B and C). The mutant P21 50L showed normal subcellular localization,, normal ApoA-l b i n d i n g , normal cholesterol efflux, and a slightly reduced phosphocholinee efflux, suggesting that perhaps this variant is a cSNP rather than a mutation. Additionall evidence for this is that the variant P2150L has only been described in patients that alsoo have the R587W variant, which clearly shows biochemical defects {Hayden, unpublished data).. In these patients the defects in ABCA1 function are more appropriately ascribed to the R587WW variant, D1289N has been described in patients homozygous for the R2081W mutation (14).. The R2081W mutation results in defects in intracellular trafficking, ApoA-l binding and lipidd efflux, and thus is likely the functional defect in these patients. As a whole, in all variants tested,, the cholesterol and phosphocholine efflux abilities were closely correlated.
<50%% of normal HDL-C levels in ABCA1 heterozygotes imply a dominant-negative functionn for the mutant allele
Off the ABCA1 heterozygote patients described thus far, only one missense mutation, M1091T, resultss in HDL-C levels that are significantly below the 50" percentile of age and sex matched controll levels. ABCA1 harboring this mutation is localized to intracellular regions and not
Figuree 3. Functional characterization of mutations that result in loss of function. (A) ABCA1 has been shown to
inducee ApoA-l binding to cell surfaces, which then facilitates the efflux of lipids across plasma memhrgnes. The abilityy of the various mutants to induce cell surface ApoA-l binding was determined by FACS analysis. Cy5 conjugatedd ApoA-l binding to the cell surface was guantitated and the values normalized to those induced by wild-typee ABCA1 All the mutants tested showed significantly lowered ApoA-l binding with the exception of D1289LL and P2150F which showed normal ApoA-l binding. (B) Since ABCA1 promotes the efflux of phosphocholme,, the ability of each of the mutants to efflux choline was determined. Tntiated choline chloride wass added to the stable cell lines, and efflux to ApoAT wasquantitated. Values were normalized to those for wild-typee ABCA1. All mutants except D1289L showed reduced choline efflux. (C) Since ABCA1 also transports cholesteroll across plasma membranes, the ability of the various mutants to induce cholesterol efflux was determinedd by loading the stable cell lines with tritiated cholesterol and quantitating the ApoA-l mediated efflux Valuess were normalized to those of wild-type ABCA1 All mutants tested with the exception of D1289L and P211 50L showed significant defects in their ability to efflux cholesterol.
foundd at the plasma membrane (Figure 4A). Both EndoH sensitive and EndoH resistant bands weree present, implying that some protein resided in the ER and another portion of the protein wass localized to the Golgi (Figure 4B). When the cell surface localized ABCA1 levels were comparedd to those of wild-type ABCA1 by biotinylation, a significant reduction in the percentage off mutant ABCA1 was observed, thus corroborating our localization and EndoH data (Figure 4C).. The lack of localization at the plasma membrane corresponded to the finding that M1091T showedd a significantly reduced ability to induce ApoA-l binding (1 , n=3, p=0.004) (Figuree 4D). Markedly reduced efflux of phosphocholine and cholesterol were observed, with effluxx levels of % (n=6, p<0.0001) for choline efflux and % (n=5, p=0.0005) forr cholesterol efflux (Figure 4E and F) being observed. This finding, along with the ~30% of normall HDL-C levels observed in patients harboring this mutation led us to hypothesize that thee M1091T mutant acted in a dominant-negative manner and prevented the functioning of thee normal ABCA1 allele, either through directly interacting with it, or by having a disturbed interactionn with a protein partner that is required for the proper functioning of the normal ABCA11 protein. That mutant ABCA1 can act in a dominant-negative manner has previously beenn demonstrated for truncation mutations of ABCA1 (27) adding further proof for the fact thatt ABCA1 either acts as part of a dimer, or acts as part of a complex involving other proteins.
M u t a n tt ABCA1 in heterozygoses w i t h >70% HDL-C levels retains partial activity
Wee next addressed biochemical defects in heterozygous patients in whom mutations result in >70%% of normal HDL-C levels. We hypothesized that the mutant allele of ABCA1 in these cases wouldd retain partial activity, thus resulting in the observed HDL-C levels. At the plasma membrane, ABCA11 induces binding of ApoA-l to specific domains on cell surfaces, permitting the efflux of cholesteroll and phosphocholine across plasma membranes to the ApoA-l rich acceptor particles. Wee hypothesized that in these patients with mutant ABCA1 showing partial activity, ABCA1 wouldd be normally localized to the plasma membrane and would induce significant levels of ApoA-ll binding.
Ass the first step, we determined the intracellular localization of each of the mutants using GFP taggedd mutant ABCA1 transfected into Hela cells. All three mutants A255T, W590S and T929I,
Figuree 4. Characterization of the ABCA1 mutant showing dominant-negative effects. (A) Intracellular localization
off the M1091T mutant indicated that it was retained in an intracellular compartment and did not reach the plasmaa membrane. (B) Endo H digestion of the M1091T mutant showed two bands, one sensitive to EndoH digestion,, indicating that a portion of M1091T was localized to the ER, and the other band, was EndoH insensitive,, indicating that a portion of M1091T reached the mediai and cis Golgi, where they are retained {C) Celll surface biotinylation assays of the M1091T mutant confumed its lack of localization at the plasma membrane whenn compared to wiid-type ABCA1. (D) The Ml 091 T mutant showed a significant ieduction in its ability to inducee ApoA-l binding when compared to wild-type ABCA1. A significant reduction in the ability of Ml 091 1 to promotee efflux of both phosphocholine (E) and cholesterol efflux (F) to ApoA-i acceptors was observed in the celll line compared to wild-type ABCA.1
O O
ro o
(A A>, ,
_i i+ +
X X O O • n nc c
LU U X X 0 0 • n nc c
LLI I+ +
O O<n n
ro ro
CD D
<£ £ Q. . 1 1 0) ) tfl lro ro
CD D
<£ £ CL LM1091T T
mmmm . mm
D D
300 0< <
O O 000 200< <
o o jaa 100 o o ra ra 0.02 2 Ctrll WT M1091T 150 0 125 5°£ £
Q .. XI 5 11 N <u<u — oo ra •ÊË Ë 33 O l/>> C 100 0 75 5 50 0 25 5 WTT M1091T 175 5 1 5 00 -X -X oo | vv Z. cc o OO "O x:: o OO N11 1
O. . 51255 -• m n n 755 500 255 -WT T PP < 0.0001 M1091T T <DD p •*-!! N IAIA — mm ra öö E 250 0 2000 -100 0 50 0 PP = 0.0005A A
WTT M 1 0 9 1 Tweree localized to the plasma membrane as well as to intracellular regions in a manner indistinguishablee from wild-type ABCA1 (Figure 5A).
Thiss finding was confirmed in the A255T mutant, which showed cell surface ABCA1 levels identicall to that of wild-type ABCA1 when assessed by biotinylation (Figure 5B).
Sincee these mutants do localize at the plasma membrane, we next looked at their ability to inducee cell surface ApoA-l binding. Binding of cells expressing each of these mutants to fluorescently conjugatedd ApoA-l was quantified using FACS analysis. All three of these ABCA1 mutants expressed inn cells showed normal binding compared to wild-type ABCA1 (A255T, 98.Oil 0.2%, n=4; W590S, 94.9+26.7%,, n= 3; T929I, 83.6+14.5, n=3) (Figure 5C), consistent with our hypothesis that these mutantss would display normal localization and normal ApoA-l binding.
W77 A?S5T W590S T929I W I A?MT WS90S Ï929I
Figuree 5. Functional characterization of ABCA1 variants with partial retention of function. Heterozygous
patientss with the mutations A255T, W590S and T929I show >70% of normal HDL-C levels, and therefore are hypothesizedd to have mutant alleles that partially retain function. (A) All three mutations were localized to the plasmaa membrane and to intracellular locations in a distribution pattern identical to that of wild-type ABCA1. (B) Inn order to confirm the GFP localization, cell surface biotinylation was performed on A255T and showed that normall levels of ABCA1 were localized at the plasma membrane when compared to wild-type ABCA1 (C) All of thee mutants showed normal ability to induce cell surface ApoA-l binding when compared to wild-type ABCA1. Alll three mutants however, showed significantly reduced ability to promote the efflux of phosphocholine (D) andd cholesterol (E) leading to the phenotypes observed in patients.
However,, all these mutant ABCA1 showed defects in both cholesterol (Figure 5D) and phosphocholmee {Figure 5E) efflux (A255T, cholesterol , n=5, p= 0.0001, choline
,, n=8, p= 0.0002; W5905, cholesterol , n = 5, p= 0.0008, choline 4 4 . 7 ^ 2 1 . 1 % ,, n o , p CO1., T929I, d i u i u . t u u l G9.9i11 1 % , n-5, p-0.004, cholintr , n=7,, p<0.0001). Mutant W5905 is defective in annexinV binding, a measure of plasma membrane flippingg (32). Plasma membrane flipping is induced by wild-type ABCA1 and is likely a critical factorr for efflux to occur. A255T and T929I have yet to be assessed for their ability to induce annexinVV binding. Defects in annexinV binding may underlie the defects in efflux observed withh all these mutants, since they were normal for all other functions tested.
Increasedd plasma HDL-C levels (>10,h percentile) in homozygotes confirm the retention
off partial activity by mutant alleles
Dataa generated thus far would predict that mutant alleles that retain partial activity in patients homozygouss for mutations in ABCA1 would confer higher plasma HDL-C levels than of those inn whom both mutant alleles lack complete function. We had previously hypothesized and confirmedd in heterozygotes that the mutant A255T allele retained partial activity. Similarly, patientss with TD who were homozygous for A255T, show 13.3% of age and sex matched controll HDL-C levels.
Nearr complete absence of plasma HDL-C in homozygotes for mutations in ABCA1 impliess the presence of null alleles of ABCA1
Patientss homozygous for R587W mutations showed 6.3% of normal levels of HDL-C, those with N935SS showed 2.62% of normal levels of HDL-C, and those with N1800H homozygous mutations showedd 3.4% of normal levels of plasma HDL-C. These findings confirmed our previous hypothesis thatt these mutations result in a complete lack of function of the ABCA1 protein.
Discussion n
Inn this study we sought to delineate how different mutations in ABCA1 influence its function inn order to provide insights into the phenotypic variability in patients with TD and FHA. In the past,, ascertainment of these diseases was based purely on phenotype, with those that completely lackedd plasma HDL-C and showing a combination of hepatosplenomegaly, peripheral neuropathy andd accumulation of lipids in the reticuloendothelial system being diagnosed as having TD, andd those with a significant reduction in HDL-C levels alone being diagnosed as having FHA. Sincee the discovery of ABCA1 as the gene in which mutations result in TD and FHA (1-3), it is noww possible to classify those that have mutations in both alleles as having TD, and those that aree heterozygous for mutations in ABCA1 as being affected with FHA. This has led to a range off phenotypes being observed in both FHA and TD patients, with varying symptoms and
severity.. Between patients who are heterozygous for mutations, there is considerable variability inn the lipid phenotype. Similarly, it is now recognized that some TD patients have a much more severee phenotype than others.
Thee first patient with TD was described in 1961 by Frederickson and colleagues (33). Since then,, close to 50 mutations have been described in the ABCA1 gene (1-3, 8-22). Since both TD andd FHA are relatively rare diseases, the numbers of individuals affected remain small. In addition,, the actual descriptions of the patient phenotypes vary widely, resulting in very few reportss that consistently describe the status of patients for the various clinical parameters. As such,, we chose plasma HDL-C levels as the phenotype for comparison of the various mutations sincee it is the only consistently reported phenotype, and this allowed us to gain sufficient, albeitt few individuals with each mutation for phenotypic analysis. Since in several reports HDL-CC levels for unaffected family members were not provided, we compared every patient with thee specific missense mutations described in the literature to the LRC database in order to determinee age and sex matched control HDL-C values.
AA clear spectrum of phenotypes was evident when patient HDL-C levels were compared to age andd sex matched controls. In those that were heterozygous for mutations in the ABCA1 gene, thee wild-type allele would contribute 5 0 % of normal plasma HDL-C levels. We classified the effectss of heterozygous ABCA1 mutations into three phenotypic groups, those with >50% HDL-C,, those with approximately 50% HDL-C, and those with <50% HDL-C when compared to controls.. We hypothesized that patients that had HDL-C >50!' percentile represented a group inn w h o m ABCA1 mutant alleles retained partial activity. By contrast, ABCA1 mutant alleles that hadd no residual function would result in a phenotype of HDL accounted for completely by the wild-typee allele. Patients with ABCA1 mutant alleles and HDL significantly less than the 50'[ percentilee would be predicted to have suppression of the wild-type allele (a dominant-negative mutation).. Those homozygous for a particular mutation in ABCA1 were classified into t w o groups,, those in w h o m the ABCA1 alleles showed partial retention of activity, and those in w h o mm the ABCA1 alleles showed a complete loss of activity.
Wee then performed detailed biochemical analysis on the specific mutations in each group in orderr to determine the defects underlying these hypotheses. We first determined the subcellular localizationn of the mutants. Wild-type ABCA1 is found at the plasma membrane and also intracellular^^ (34, 35), and the plasma membrane localization of ABCA1 is critical for its functionn in efflux. Therefore, mutants that do not localize to the plasma membrane will show functionall defects in lipid efflux. We next determined the glycosylation status of mutations in eachh group, since ABCA1 that is retained in the ER would show EndoH sensitivity and protein thatt transits to the medial and trans Golgi will show EndoH insensitivity, providing further informationn on the trafficking defects of mutant ABCA1. We also performed cell surface biotinylationn assays in order to quantify the expression of mutant ABCA1 protein at the plasma
membranee and to confirm the localization and EndoH data. We next performed ApoA-l binding assays.. Wild-type ABCA1 has been shown to induce the binding of ApoA-l at the cell surface, andd this is essential for lipid efflux. Therefore, mutants inducing no cell surface ApoA-l binding willl be efflux defective Finally, we quantified both cholesterol and phosphochohne efflux fromm the mutant cell lines.
Ass expected, almost all mutations resulted in defects in both phosphocholine and cholesterol efflux,, although different biochemical defects were identified as the underlying cause of the loweredd efflux levels. Of the three missense mutations tested that showed residual function, A255T,, W590S and T929I, all localized to the plasma membrane, and all three induced cell surfacee ApoA-l binding to similar levels as wild-type ABCA1. Our findings for the mutant W590SS therefore agree with previous findings that W590S does localize normally, and shows normall ApoA-l binding (32, 36, 37). This data implies that although mutant ABCA1 localizes to thee plasma membrane, and induces ApoA-l binding, these events alone are insufficient for normall lipid efflux to occur. That additional events are needed is evident from a previous study thatt showed that the mutant W590S induces significantly less membrane phosphatidyl serine flippingg when compared to wild-type ABCA1 (32), and implies that in addition to ApoA-l binding,, other membrane structural requirements exist for ABCA1 mediated efflux to occur. Mutantss A255T and T929I have not been assessed for their ability to induce membrane flipping andd phosphatidyl serine exposure.
Off the mutants that show complete loss of function, most showed intracellular localization, andd unlike wild-type ABCA1, showed no plasma membrane localization and thus displaying reducedd ApoA-l binding and no lipid efflux. C1477R however, did reach the plasma membrane, andd showed similar localization as the wild-type ABCA1 protein. However, this mutant induced almostt no cell surface ApoA-l binding (32, 36). This resulted in a significant reduction in lipid effluxx capacity. The defect in the C1477R mutant suggests that although plasma membrane localizationn is critical for lipid efflux to occur, clearly other mechanisms result in failure of bindingg to ApoA-l. C1477R is localized within the second large extracellular loop in ABCA1, andd studies in ABCR have revealed that its t w o large extracellular loops interact with each otherr through disulfide bonds (38). Since ABCR and ABCA1 are closely related, it is a possibility thatt this disrupted cysteine in the second large extracellular loop prevents appropriate interaction betweenn the two large extracellular loops of ABCA1, which could be essential for appropriate celll surface ApoA-l binding. Both A1046D and S1506L showed partial localization at the plasmaa membrane, and revealed significantly reduced ApoA-l binding. The data for 51506L agreess with data previously described (36), The S1506L mutation also occurs in the second largee extracellular portion of ABCA1, and could therefore disrupt the normal folding of the protein,, and may potentially destroy a domain of interaction for the extracellular loops necessary forr ApoA-l binding. The mutation A1046D however, occurs in the intracellular domain of
ABCA11 and is localized between the first Walker A and B motifs. That this intracellular mutation wouldd cause a similar biochemical defect as those in the second large extracellular loop suggests thatt some intracellular sites may also determine the three dimensional folding efficiency of ABCA11 vital for its ability to induce ApoA-l binding perhaps by affecting the formation of a channell through which lipid is transported for ApoA-l binding.
Bothh P2150L and D1289N traffic normally, and show normal ApoA-l binding. They both show normall cholesterol efflux and P2150L shows a mild reduction in choline efflux. Thus, despite theirr description as mutations, our data is more consistent with these t w o variants being cSNPs.. P2150L has only been found in compound heterozygous patients (Hayden, unpublished data)) who contain the R587W mutation and another unidentified mutation. Since biochemical characterizationn of R587W clearly results in defects in function, the latter mutation likely contributess to the observed phenotypes. In addition, patients homozygous for the R587W mutationn have been described (12, 18), presumably without the P2150L variant, and showing aa TD phenotype, making it more likely that P2150L is a cSNP. D1289N has been described as aa variant in patients that are homozygous for R2081W (14). Biochemical characterization of R2081WW does result in defects in subcellular localization and lipid efflux, suggesting that D1289NN is another cSNP.
Onee mutant, M1091T, when present in heterozygous patients, results in significantly lower thann 50% of normal HDL-C levels. This mutant was the most functionally deficient biochemically, andd resulted in almost complete lack of cholesterol and choline efflux. It does not localize to thee plasma membrane, and also showed almost no ApoA-l binding. It has been previously suggestedd that ABCA1 acts as a dimer or as part of a complex, since truncation mutations of ABCA11 also result in lower than expected HDL-C levels for heterozygous mutations, suggesting aa dominant-negative role for these mutants. Further evidence for the fact that ABCA1 either dimerizess or acts as part of a complex is that when fibroblasts from patients with truncation mutationss were treated with oxysterols that normally significantly upregulate ABCA1 RNA and protein,, cell lines with truncated ABCA1 showed increased mRNA, identical to controls, but no increasee in ABCA1 protein levels were observed, but rather, a decrease in the amount protein contributedd by the wild-type allele was observed (27). M1091T, although a missense mutation, likelyy acts in a dominant-negative manner, and interferes with the function of the normal allele off ABCA1 by an as of yet unknown mechanism.
Overall,, the biochemical data presented in this study provide an explanation for the varying HDL-CC levels in patients with mutations in ABCA1 and provide novel insights into genotype/ phenotypee correlations between various ABCA1 mutants and plasma HDL-C levels, offering somee insight into the biochemical defects underlying these observed phenotypes
References: :
1.. Brooks-Witson, A., Marcil, M., Clee, S.M., Zhang, L., Roomp, K, van Dam, M , Yu, L, Brewer, C , Collins, J.A., Molhuizen,, H.O.F., Loubser, 0., Ouellette, B.F.F., Fichter, K., Ashbourne Excoffon, K.J.D., Sensen, C.W., Scherer,, S., Mott, 5., Denis, M., Martmdale, D , Frohlich, J,, Morgan, K., Koop, B., Pimstone, S.N., Kastelein, JJ J P . Genest. J Jr and HavH^n M R 1999 r .-lutati-.n-- in ARM i,, T ^ u j ^ r H ^ H S " ^ f ^ n ^ i i~:^^ ^:; -y lipoproteinn deficiency. Nat. Genet. 22: 336-345.
2.. Rust, S., Rosier, M., Funke, H , Amoura, Z., Piette, J.-C, Deleuze, J.-F., Brewer, H.B. Jr., Duverger, N., Denefle, P.,, and Assmann, G. 1999 Tangier Disease is caused by mutations in the gene encoding ATP-bmding cassette transporterr 1. Nat.Genet. 22: 352-355
33 Bodzioch, M., Orsó, E., Klucken, J., Langmann, T., Bottcher, A., Diederich, W., Drobnik, W., Bailage, S., Buchler,, C , Porsch-Ozcurumez, M., Kaminski, W E . , Hahmann, H.W., Oette, K., Rothe, G., Aslanidis, C., Lackner,, K.J., and Schmitz, G. 1999. The gene encoding ATP-bmding cassette transporter 1 is mutated in Tangierr Disease. Nat.Genet. 22: 347-351.
4.. Tall, A.R., Costet P, and Wang, N. 2002 Regulation and mechanisms of macrophage cholesterol efflux. J. Clin. InvestInvest 110(7)-899-904
5.. Wang, N., Silver, D.L., Thiele, C, and Tall, A.R 2 0 0 1 . ATP-binding cassette transporter A1 (ABCA1) functions ass a cholesterol efflux regulatory protein. J. Biol. Chem. 276(26):23742-7.
6.. Singaraja, R.R., Bocher, V„ James, E.R., Clee, S.M., Zhang, L.H , Leavitt, B.R., Tan, B., Brooks-Wilson, A., Kwok, A.,, Bissada, N., Yang, Y.Z., Liu, G., Tafuri, S.R., Fievet, C , Wellington, C.L., Staels, B., and Hayden, M.R. 2 0 0 1 . Humann ABCA1 BAC transgenic mice show increased high density lipoprotein cholesterol and ApoAI-dependent effluxx stimulated by an internal promoter containing liver X receptor response elements in intron 1. J. Biol. Chem.Chem. 276(36):33969-79.
77 Singaraja, R R., Fievet, C, Castro, G , James, E.R., Hennuyer, N., Clee, S.M., Bissada, N., Choy, J.C., Fruchart, J.C.,, McManus, B.M., Staels, B., and Hayden, M.R. 2002. Increased ABCA1 activity protects against atherosclerosis.. J.Clin Invest. 110: 35-42.
8.. Hong, S.H., Riley, W., Rhyne, J., Friel, G.,and Miller, M. 2002. Lack of association between increased carotid ntima-mediaa thickening and decreased HDL-cholestero! in a family with a novel ABCA1 variant, G2265T. Clin.Clin. Chem 48:2066-2070.
9.. Wang, J., Burnett, J.R., Near, S., Young, K., Zinman, B., Hanley, A.J , Connelly, P.W., Harris, SB., and Hegele, R.A.. 2000. Common and rare ABCA1 variants affecting plasma HDL cholesterol. ArtenosclerThromb. Vase.Biol. 20:: 1983-1989
100 Nishida, Y., Hirano, K., Tsukamoto, K., Nagano, M., Ikegami, C , Roomp, K., Ishihara, M., Sakane, N., Zhang, Z.,, Tsujn, K.K., Matsuyama, A , Ohama, T., Matsuura, F , Ishigami, M., Sakai, N., Hiraoka, H., Hattori, H , Wellington,, C , Yoshida, Y , Misugi, S., Hayden, M.R., Egashira, T., Yamashita, S , and Matsuzawa, Y 2002 Expressionn and functional analyses of novel mutations of ATP-binding cassette transporter-1 in Japanese patientss with high-density lipoprotein deficiency. Biochem. Biophys. Res. Commun. 290: 713-721.
111 Brousseau, M.E., Shaefer, E.J., Dupuis, J., Eustace, B., Van Eerdewegh, P., Goldkamp, A.L., Thurston, L.M., FitzGerald,, M.G., Yasek-McKenna, D , O'Neill, G., Eberhart, G.P., Weiffenbach, B., Ordovas, J.M., Freeman, W W ,, Brown, R.H.Jr, and Gu, J.Z. 2000 Novel mutations in the gene encoding ATP-binding cassette 1 in four Tangierr disease kindreds. / Lipid Res. 41: 433^441
122 Lawn, R M , Wade, D P., Garvin, M R, Wang, X , Schwartz, K., Porter, J.G., Seilhamer. J.J., Vaughan, A.M., andd Oram, J F, 1999. The Tangier disease gene product ABC1 controls the cellular apolipoprotein-mediated lipidd removal pathway J Clin. Invest 104: R25-R31
13.. Lapicka-Bodzioch, K , Bodzioch, M., Krull, M., Kielar, D , Probst, M., Kiec, B., Andnkovics, H., Bottcher, A., Hubacek,, J., Aslanidis, C, Suttorp, N., and Schmitz, G. 2001 Homogeneous assay based on 52 primer sets to scann for mutations of the ABCA1 gene and its application in genetic analysis of a new patient with familial high-densityy lipoprotein deficiency syndrome. Biochim. Biophys Acta 1 537 4 2 4 8
14.. Huang, W., Monyama, K., Koga, T , Hua, H., Ageta, M., Kawabata, S., Mawatari, K , Imamura, T , Eto, T , Kawamura,, M., Teramoto, T., and Sasaki, J. 2 0 0 1 . Novel mutations in ABCA1 gene in Japanese patients with Tangierr disease and familial high density lipoprotein deficiency with coronary heart disease Biochim. Biophys, Acta.Acta. 1 5 3 7 : 7 1 7 8
11 5. Ishii, j . , Nagano, M , Kujiraoka, T , Ishihara, M., Egashira, T , Takada, D , Tsuji, M , Hattori, H., and Emi, M. 2002. Clinicall variant of Tangier disease in Japan: mutation of the ABCA1 gene in hypoalphalipoproteinemia with corneall lipidosis J. Hum. Genet. 47: 366-369.
16.. Altilia, S., Pisciotta, L, Garuti, R., Tarugi, P., Cantafora, A , Calabresi, L., Tagliabue, J., Maccari, S., Bernini, F., Zanotti,, I , Vergani, C, Bertolini, S, and Calandra, 5 2003. Abnormal splicing of ABCA1 pre-mRNA in Tangier diseasee due to a IVS2 +5G>C mutation in ABCA1 gene J Lipid Res. 44: 254-264.
11 7. Guo, Z , Inazu, A., Yu, W., Suzumura, T., Okamoto, M., Nohara, A , Higashikata, T., Sano, R., Wakasugi, K., Hayakawa,, T , Yoshida, K., Suehiro, T., Schmitz, G , and Mabuchi, H 2002. Double deletions and missense mutationss in the first nucleotide-binding fold of the ATP-binding cassette transporter A1 ( A B C A I ) g e n e in Japanesee patients with Tangier disease. J Hum. Genet. 47: 325-329.
11 8. Bertolini, S., Pisciotta, L, Sen, M., Cusano, R., Cantafora, A., Calabresi, L, Franceschini, G., Ravazzolo, R., and Calandra,, S, 2001. A point mutation in ABC1 gene in a patient with severe premature coronary heart disease andd mild clinical phenotype of Tangier disease. Atheroscler. 154: 599-605.
19.. Hong.S.H., Rhyne, J., Zeiler, K.,and Miller, M. 2002. Novel ABCA1 compound variant associated with HDL cholesteroll deficiency. Biochim. Biophys. Acta. 1587: 60-64.
200 Hong, S.H., Rhyne, J , Zeiler, K., and Miller, M. 2002. ABCA1 (Alabama): a novel variant associated with HDL deficiencyy and premature coronary artery disease. Atheroscler. 164: 245-250.
211 Remaley, A.T., Rust, S., Rosier, M , Knapper, C , Naudin, L., Broccardo, C, Peterson, K.M., Koch, C , Arnould, I.,, Prades, C , Duverger, N., Funke, H , Assman, G., Dinger, M., Dean, M., Chimini, G., Santamarina-Fojo, S., Fredrickson,, D.S., Denefle, P , and Brewer, H.BJr 1999 Human ATP-binding cassette transporter 1 (ABC1): genomicc organization and identification of the genetic defect in the original Tangier disease kindred. Proc NatlNatl Acad. Sci. U.S.A. 96: 12685-12690
22.. Asztalos, B.F., Brousseau, M.E , McNamara, J.R., Horvath, K.V., Roheim, P 5 , and Schaefer, E.J. 2 0 0 1 . Subpopulationss of high density lipoproteins in homozygous and heterozygous Tangier disease. Atheroscler. 156:217-225. .
233 Blaxter, M 1998. Caenorhabditiseleganss is a nematode. Science. 282(5396)2041-6.
244 Feng, D.F , Cho, G., and Doolittle, R.F. 1997. Determining divergence times with a protein clock: update and reevaluation.. Proc. Natl. Acad. So. USA. 94(24): 13028-33.
255 Wang, N., Silver, D.L , Costet, P., and Tall, A.R. 2000 Specific binding of ApoA-l, enhanced cholesterol efflux, andd altered plasma membrane morphology in cells expressing ABC 1. J. Biol. Chem. 275(42):33053-8.
26.. Wellington, C.L., Singaraja, R., Ellerby, L, Savill, J., Roy, S., Leavitt, B , Cattaneo, E., Hackam, A., Sharp, A , Thornberry,, N., Nicholson, D.W., Bredesen, D.E., and Hayden MR. 2000. Inhibiting caspase cleavage of huntingtinn reduces toxicity and aggregate formation in neuronal and nonneuronal cells. J. Biol. Chem. 275(26)49831-8 8
21.. Wellington, C I . , Yang, Y.Z., Zhou, S., Clee, S.M., Tan, B„ Hirano, K., Zwarts, K., Kwok, A., Gelfer, A., Marcil, M.,, Newman, S., Roomp, K., Singaraja, R., Collins, J., Zhang, L.H., Groen, A.K., Hovingh, K., Brownlie, A , Tafuri,, S., Genest, J. Jr., Kastelein, J.J., and Hayden, M R . 2002 Truncation mutations in ABCA1 suppress normall upregulation of full- length ABCA1 by 9-cis-retinoic acid and 22-R-hydroxycholesterol. J. Lipid Res. 43: 1939-1949. .
28.. Chambenoit, O., Hamon, Y., Marguet, D., Rigneault, H , Rosseneu, M., and Chimini, G. 2001. Specific docking of apolipoproteinA-ll at the cell surface requires a functional ABCA1 transporter. 7. Biol. Chem. 276(13):9955-60
29.. Attie, A.D., Hamon, Y„ Brooks-Wilson, A.R., Gray-Keller, MP., MacDonald, M L , Rigot, V., Tebon, A., Zhang, L.H.,, Mulligan, J.D., Singaraja, R.R., Bitgood, J J , Cook, M E , Kastelein, J.J., Chimini, G., and Hayden MR. 2002. Identificationn and functional analysis of a naturally occurring E89K mutation in the ABCA1 gene of the W H A M chicken.. J. Lipid. Res. 43(10): 1610-7.
300 Maley, F, Trimble, RB., Tarentino, A.L., and Plummer, T.H.Jr. 1989. Characterization of glycoproteins and theirr associated oligosaccharides through the use of endoglycosidases. Anal. Biochem. 180(2): 195-204.
31.. Tarentino, A.L., Quinones, G., Trumble, A , Changchien, L.M., Duceman, B., Maley, F., and Plummer, T.H Jr. 1990.. Molecular cloning and amino acid sequence of peptide-N4-(N-acetyl-beta-D-glucosaminyl)asparagine amidasefromfiavobacteriummeningosepttcumJ.. Biol. Chem. 265(12):6961 -6.
32.. Rigot, V., Hamon, Y., Chambenoit, 0., Alibert, M., Duverger, N., and Chimini, G. 2002. Distinct sites on ABCA11 control distinct steps required for cellular release of phospholipids. J. Lipid Res. 43: 2077-2086.
33.. Fredenckson, DS., Altrocchi, P.H., Avioli, L.V., Goodman, D.W.S. and Goodman, H.C. 1961. Tangier Disease. Combinedd clinical staff conference at the National Institutes of Health. Annals Int Med. 55:1016-1031
34.. Neufeld, E.B., Remaley, A.T., Demosky, S.J., Stonik, J.A., Cooney, A.M., Comly, M., Dwyer, N.K., Zhang, M., Blanchette-Mackie,, J., Santamarina-Fojo. S., and Brewer, H.B. Jr 2001 Cellular localization and trafficking of thee human ABCA1 transporter. J. Biol. Chem. 276(29)27584-90.
35.. Santamarina-Fojo, S., Remaley, A T . , Neufeld, E.B., and Brewer, H.B. Jr. 2 0 0 1 . Regulation and intracellular traffickingg of the ABCA1 transporter. Lipid Res. 42(9): 133945.
36.. Fitzgerald, M l . , Morns, A.L., Rhee, J.S., Andersson, LP., Mendez, A.J., and Freeman, M.W. 2002. Naturally occurringg mutations in the largest extracellular loops of ABCA1 can disrupt its direct interaction with apolipoproteinA-l.J.. Biol. Chem. 277:33178-33187.
37.. Tanaka, A.U , Abe-Dohmae, S., Oiinisni, T., Auki, R., fviorinaga, G., ökuhira, K.I., ikeda, Y., Kano, h, Matsuo, M., Kioka,, N., Amachi, T., Murata, M., Yokoyama, S., and Ueda, K. 2003. Effects of mutations of ABCA1 in the first extracellularr domain on subcellular trafficking and ATP binding/hydrolysis. J. Biol. Chem. 278(10)881 5-9
38.. Bungert, S., Molday, L.L.and Molday, R.S.2001 Membrane topology of the ATP binding cassette transporter ABCRR and its relationship to ABC 1 and related ABCA transporters: identification of N-linked glycosylation sites. J.J. Biol. Chem. 276: 23539-23546.
