www.elsevier.com/locate/ybbrc

Biochemical and Biophysical Research Communications 351 (2006) 588–594

BBRC

b-adaptin: Key molecule for microglial scavenger receptor functionunder oxidative stress q

Natalia Manzano-Leon, Blanca Delgado-Coello, Margarita Guaderrama-Dıaz,Jaime Mas-Oliva *

Instituto de Fisiologıa Celular, Universidad Nacional Autonoma de Mexico, Mexico, DF, Mexico

Received 3 October 2006Available online 24 October 2006

Abstract

Scavenger receptors internalize chemically modified low density lipoprotein particles (ac-LDL) and other ligands through the processof receptor-mediated endocytosis. During this investigation using amyloid-b as a natural ligand for the SR, we studied under a ligand-induced oxidative stress condition, changes in protein expression of several adaptor proteins important in the organization of the endo-cytic machinery in microglia and macrophages. Differential expression experiments of b-adaptin, a-adaptin, SR-AI, and SR-BI in RAW(macrophages) and EOC (microglia) cells were performed according to dosage and exposure time to amyloid-b. Our results show thataccording to dosage, amyloid-b produces an oxidative stress state that importantly affects the availability of b-adaptin. Under theseconditions, RT-PCR assays show that b-adaptin mRNA is normally synthesized, reason why protein translation or protein structureof b-adaptin might be altered. These observations might have impact in the understanding of the mechanisms microglia employ toprocess amyloid-b in the brain.� 2006 Elsevier Inc. All rights reserved.

Keywords: Microglia; Amyloid-b peptide; Scavenger receptor; Oxidative stress; b-adaptin

Scavenger receptors (SR) divided into six classes consistof a broad family of membrane receptors that internalizeschemically modified low density lipoproteins (LDL) andseveral other ligands [1]. These receptors are localized inmacrophages and similar cells such as microglia, as wellas in hepatocytes, ovary, testicle, suprarenal gland, andastrocytes [2].

SR-AI is able also to bind oxidized LDL (ox-LDL),methylated bovine serum albumin (M-BSA), polyinosinicacid, polyguanosinic acid, polysaccharides such as dextransulfate, fucoidin, lipoteichoic acid (LTA), lipopolysaccha-ride (LPS), and advanced glycosylation end-products(AGE) [3]. Another important ligand for this receptor isthe amyloid-b peptide (Ab) [4]. Ab in high concentrations

0006-291X/$ - see front matter � 2006 Elsevier Inc. All rights reserved.

doi:10.1016/j.bbrc.2006.10.077

q 2006 Prize for Basic Medical Investigation ‘‘Dr. Jorge Rosenkranz’’.Roche-Syntex, Mexico.

* Corresponding author. Fax: +52 55 5622 5611.E-mail address: jmas@ifc.unam.mx (J. Mas-Oliva).

acquires an insoluble b-folded structure that leads to fibrilformation, process directly related to extracellular peptideaccumulation, formation of neurite plaques, developmentof oxidative stress, and cell death [5].

SR-AI endocytosis is carried out through the formationof coated pits on the cell surface [6]. The VXFD amino acidsequence found in the cytoplasmatic tail of SR-AI is funda-mental for the internalization of this receptor, in a similarfashion with specific sequences identified in other receptorsimportant for endocytosis [7].

The class B type I scavenger receptor (SR-BI) was iden-tified due to its binding to ac-LDL, ox-LDL, and nativeLDL. It also binds M-BSA, anionic phospholipids, apop-totic cells, Ab, very low density lipoproteins (VLDL),and high density lipoproteins (HDL) [8]. SR-BI has beenlocalized in diverse cell types, including glia and microglialcells [4,9].

Although SR-BI plays a fundamental role in reversecholesterol transport due to its binding capacity for

N. Manzano-Leon et al. / Biochemical and Biophysical Research Communications 351 (2006) 588–594 589

HDL, it does not internalize lipoprotein particles throughthe classical process of receptor-mediated endocytosis,but rather selectively transfers lipids from the cell mem-brane to lipoproteins [10]. Notwithstanding, when ligandsare of a proteic character, SR-BI is considered an endocyticreceptor since it promotes uptake and degradation ofAGE-BSA, M-BSA, ox-LDL, and ac-LDL [11].

The mechanism followed by SR-BI internalization is notknown with certainty, but it is believed to occur throughcaveolae, since approximately 60% of SR-BI within a cellis found in cholesterol-rich lipid domains associated tothese structures. Although the remaining 40% of SR-BI islocalized in caveolin-poor zones [12], there is also evidencethat endocytosis through this receptor might be carried outthrough clathrin-coated pits [13]. On the other hand,CD36, a class B scavenger receptor, has been reported tobe expressed on microglia in Alzheimer disease brainsand mediates production of reactive oxygen species(ROS) [14].

Nearly 30 proteins involved in clathrin-mediated endo-cytosis have been described; many of these are accessorymolecules that aid in the formation of the coated pit.Among these, in addition to clathrin, the clathrin assemblylymphoid myeloid leukemia protein AP180/CALM, Eps15,dynamin, and amphiphysin stand out [15].

Clathrin monomers are arranged within a structuredenominated triskelion, composed of three heavy and threelight chains that bind in polyhedral structures, fundamen-tal in the coated pit formation. The N-terminal end ofthe clathrin heavy chain interacts with proteins involvedin endocytosis [16]. a-adaptin, which makes up togetherwith the b-, l-, and r-adaptin part of the AP2 complex isimportant for the binding of this heterotetramer to theplasma membrane, as well as for its interaction withCALM and amphiphysin [17]. On the other hand, b-adap-tin possesses two clathrin-binding sites [18] and intervenesin the selection of the vesicle cargo molecule [19]. AP180/CALM presents binding sites for clathrin, AP2 [20], andfor PtdIns(4,5)P2-containing membranes [21]. Interestingly,the epidermal growth factor (EGF) receptor pathway sub-strate clone 15 (Eps 15) is a protein that interacts withAP180/CALM, epsin, a-adaptin, as well as several otheraccessory proteins involved in endocytosis [22].

From our laboratory, previous work employing tert-butylhydroperoxide (TBH) has established that SR-AI-transfected CHO cells under oxidative stress present theability to bind ac-LDL, while lipoprotein internalizationpractically disappears [23]. Similar results have been foundwith C6 (glia) and PC12 (pheochromocytoma) cells, whichpossess SR-BI in a natural form (Aguilar-Gaytan et al.,unpublished data). In both cases, oxidative stress did notmodify either mRNA or SR protein expression levels.

Therefore, the main objective of this study has been toevaluate possible modifications in the expression of severalof these endocytic proteins during an episode of naturaloxidative stress provoked by Ab. Together with the possi-ble consequences this situation might have upon SR func-

tion, for the first time we have been able to study theinteractions between the cytoplasmic region of the SRand endocytic proteins in response to binding of Ab tothe SR.

The experiments were conducted with mouse macro-phages (RAW) and mouse microglia (EOC) cells, two celllines that possess SR-AI and SR-BI naturally. Macro-phages were used as control cells and microglia as residentmacrophages of the brain that present a well-establishedphysiological relationship with Ab deposits.

Materials and methods

Materials. Salts and buffers were purchased from Sigma (St. Louis,MO). High-purity synthetic peptides were obtained from PeptidoGenicResearch & Co., Inc. (Livermore, CA). Reagents for cell culture (GIBCO-Invitrogen, Carlsbad, CA). Rat anti-mouse SR-AI/II antibody (Serotec,Raleigh, NC). Rabbit anti-mouse SR-BI antibody (Calbiochem-Merck,Darmstadt, Germany). Anti-endocytic proteins, goat polyclonal anti-Aband secondary antibodies (Santa Cruz Biotechnology, Inc., Santa Cruz,CA). The ECL Plus kit as well as Protein G Sepharose 4 Fast Flow(Amersham Bioscience, Uppsala, Sweden). 6-Carboxy-H2DCFDA [6-carboxy-20,7 0-dichlorodihydrofluorescein diacetate, di (acetoxymethylester)] (Molecular Probes, Eugene, OR).

Amyloid-b fibril formation. Ab25–35 peptide (Gly-Ser-Asn-Lys-Gly-Ala-Ile-Ile-Gly-Leu-Met), considered the most toxic fragment from the original40–42 amino acid peptide (Asp-Ala-Glu-Phe-Arg-His-Asp-Ser-Gly-Tyr-Glu-Val-His-His-Gln-Lys-Leu-Val-Phe-Phe-Ala-Glu-Asp-Val-Gly-Ser-Asn-Lys-Gly-Ala-Ile-Ile-Gly-Leu-Met-Val-Gly-Gly-Val-Val-Ile-Ala), wasused to induce oxidative stress. Ab25–35 was solubilized in deionized water at10 lg/ll and stored at �20 �C prior to use. To induce fibril formation,Ab25–35 peptide was resuspended in serum-free culture media andincubated for 3 days at 37 �C. Fibril formation was confirmed by electronmicroscopy (data not shown). In control assays the reverse Ab35–25 peptidewas used (Met-Leu-Gly-Ile-Ile-Ala-Gly-Lys-Asn-Ser-Gly).

Cell culture. RAW cells were grown in Minimum Essential Media(MEM), and EOC cells in Dulbecco’s modified Eagle’s medium (DMEM),containing penicillin (50 U/ml)/streptomycin (50 lg/ml) and supplement-ed with 10% fetal bovine serum at 37 �C in a 5% CO2/95% air atmosphere.EOC cell medium was supplemented with 20% LADMAC conditionedmedia (produced from the LADMAC cell line, rich in growth factorcolony stimulating factor 1).

Oxidative stress assays. 6-Carboxy-H2DCFA was resuspended inDMSO at 10 mM concentration and stored at �20 �C. RAW and EOCcells were exposed at Ab25–35 (0–200 lg/ml) and 1 or 20 lM 6-carboxy-H2DCFA for RAW and EOC cells, respectively, during 24 h in serum-freeculture medium. Cell fluorescence intensity was measured on a FAC-Scalibur instrument from Becton–Dickinson (NJ, USA).

Immunoblots. Cells were treated during 24 h or indicated times withdifferent doses of Ab25–35 in serum-free media. After this procedure, cellswere washed twice with PBS and lysed for 45 min at 4 �C in lysis buffer(150 mM NaCl, 10 mM Tris, pH 7.4, 1% Triton X-100, 0.5% NP40, 1 mMEDTA, 1 mM EGTA, 0.2 mM sodium orthovanadate, 10 mM benzami-dine, 10 lg/ml leupeptin, 10 lg/ml aprotinin, and 250 lM PMSF). Lysedcells were centrifuged at 1500g for 10 min at 4 �C and the supernatantrecovered. Protein concentration was determined using the Micro-BCAprotein assay (Pierce, Rockford, IL) and samples (10 lg/lane) from thetotal protein fraction were analyzed by SDS–PAGE on 10% gels furthertransferred to nitrocellulose paper (Bio-Rad, Hercules, CA). Membraneswere blocked overnight at room temperature with a solution containing5% fat-free milk and 1% Tween 20 in Tris-buffered saline (TBS), pH 7.6.The following antibodies each diluted 500-fold were used: goat polyclonalanti-heavy chain of clathrin, b2 and a2-adaptins and a rabbit polyclonalanti-Eps15. A rat monoclonal anti-mouse scavenger receptor AI/II anti-body and a rabbit polyclonal anti-mouse scavenger receptor BI antibody

590 N. Manzano-Leon et al. / Biochemical and Biophysical Research Communications 351 (2006) 588–594

were also used diluted 100-fold. Secondary antibodies consisted of adonkey anti-goat IgG, a goat anti-rabbit IgG both conjugated to horse-radish peroxidase, and a goat anti-rat horseradish peroxidase mouse-ad-sorbed. Secondary antibodies were used employing 5000-fold dilutionsand were incubated for 1 h at 37 �C in blocking buffer. The membraneswere washed with TBS-Tween 1%, and HRP activity was detected usingthe ECL Plus kit according to the supplier’s instructions. Band imageswere scanned and analyzed using ImageJ software (National Institute ofHealth, Bethesda, Maryland, USA).

RT-PCR. RAW and EOC cells were treated with Ab25–35 (0–200 lg/ml) for 24 h in serum-free medium. RNA was isolated with Trizol(Invitrogen Co.), according to the manufacturer’s instructions. The levelof transcripts from encoding RNAs for scavenger receptor AI and BI, andendocytic proteins were determined by RT-PCR starting with 2 lg of totalRNA using the GeneAmp RNA Kit (Applied Biosystems). cDNAs werederived using 2.5 lM random hexanucleotides and oligo(dT)16, and 2.5 UMuLV of reverse transcriptase at 42 �C for 45 min followed by denatu-ralization at 99 �C for 5 min. cDNAs were amplified by PCR using specificprimers (0.2 lM) and 2.5 U of AmpliTaq DNA polymerase during 30cycles; denaturalization at 99 �C for 2 min, annealing at indicated tem-perature for each primer pair (60–63 �C) for 30 s, and extension at 72 �Cfor 1 min. SR-AI sense primer was 5 0-CTGGAC AAACTG GTCCACCT-30 and antisense primer 5 0-TCCCCT TCTCTC CCTTTT GT-3 0. SR-BI sense primer 5 0-CAGGCT GTGGGA ACTCTA GC-3 0 and antisenseprimer 5 0-GAAAAA GCGCCA GATACA GC-3 0. a-adaptin sense primer5 0-TTTGTG GGCATT GGACTG TA-3 0 and antisense primer 5 0-AGACCC GTGCAT AACTGA CC-30. b-adaptin sense primer 5 0-CGCTCCTTCACC TACTCC TG-3 0 and antisense primer 5 0-GTGAGT AAACGTCCCCGA AA-3 0. Clathrin sense primer 5 0-GTGAGC ACAGGAATCTGC AA-3 0 and antisense primer 5 0-CAATCA ACACCT GCACTGCT-30. b-actin sense primer 5 0-GGGTCA GAAGGA TTCCTA TG-3 0

and antisense 5 0-GGTCTC AAACAT GATCTG GG-3 0.Immunoprecipitations. Cells were treated with Ab25–35 (20 lg/ml) for 0,

5, 10, 15, and 30 min, and lysed as previously described. Five hundredmicrograms of cellular protein lysate was incubated with the goat poly-clonal antibody anti-b2-adaptin (1:50) for 1 h at 4 �C, and immune com-plexes were precipitated with protein G–Sepharose. Immunoprecipitatedproteins were washed twice with lysis buffer and resuspended in 40 ll of 5·SDS loading buffer. Twenty microliter samples were run on a 10% dena-turing SDS–polyacrylamide gel for immunodetection of associated pro-teins in the complex.

Results and discussion

Oxidative stress

In order to determine, whether cells in contact withAb25–35 present evidence of an oxidative stress state, we uti-lized the 6-carboxy-H2DCFA reagent, a molecule that eas-ily passes through the cell membrane and reacts withreactive oxygen species (ROS). Fig. 1A shows fluorescenceintensity of RAW cells exposed to increasing doses ofAb25–35. We observe that RAW control cells without serumpresent a basal fluorescence similar to the one shown bycontrol cells with serum, indicating a low stress level dueto its absence. Once exposed to Ab25–35, RAW cell fluores-cence increases gradually. EOC cells treated with increasingdoses of Ab25–35 (0–200 lg/ml) in the presence of 20 lM 6-carboxy-H2DCFA and cultured in the absence of serum for24 h showed the same tendency in fluorescence increase(Fig. 1B). Since in the past microglia have shown to be rel-atively resistant to Ab-induced cell death, which must haveprobably takes place through the development of different

mechanisms against the production of ROS [24], in ourhands, using incubation times up to 24 h and maintainingcell viability above 50%, success was achieved in establish-ing an oxidative stress state in this cell type. Moreover, theamount of Ab employed had to be maintained relativelyhigh in order to obtain clearcut responses. This set ofresults supports the fact that RAW and EOC cells developan Ab25–35 dependent oxidative stress state, consistent withprevious reports showing that this peptide produces highlevels of ROS not only in rat cortical neurons [25] and hip-pocampus neurons [26], but also in microglia [27].

Cell viability of RAW and EOC cells was evaluated bythe trypan blue method when in contact with different dos-es of Ab25–35 during 24 h. Viability percentages decreasedup to 50% in the 200 lg/ml dose in both cell lines. Viabilityshown by control cells grown with or without serum wascarried out in order to avoid possible interference withother receptors, since assays were always serum-free. Con-trol experiments performed with the reverse Ab35–25 pep-tide showed that cell viability was not affected in either ofthe experiments carried out with or without the presenceof serum (data not shown).

Protein expression and endocytosis

We evaluated Ab25–35 internalization in RAW and EOCcells through immunofluorescence assays with an anti-Abantibody and a rhodamine-coupled secondary antibody.During these assays, we observe according to the dosageemployed that cells internalize Ab25–35 and, found distrib-uted throughout the cytoplasm. Although, we observedhigh fluorescence levels at the higher doses of peptide, fluo-rescence is lost when using 200 lg/ml. It is possible thatexposure of cells to high concentration of Ab25–35, as fur-ther explored in our study, promotes cells to lose the capac-ity to internalize Ab through the SR, a phenomenonapparently directly related to the disruption of the endo-cytic machinery (data not shown).

To date, there are few data concerning the influence Abexerts upon proteins that participate in the endocyticmachinery [28]. Therefore, this study, for the first time,explores the effect of an oxidative stress state producedby exposure to Ab25–35, upon the expression of proteinsinvolved in the process of endocytosis. This conditionmight be considered of special interest when studying cellsthat normally present the SR and under physiological con-ditions can be found in direct contact with Ab deposits. Tostudy this phenomenon, we analyzed the expression levelsof SR-AI, SR-BI, a-adaptin, b-adaptin, and Eps15 allinvolved in receptor-mediated endocytosis.

First, we used total protein extracts from RAW andEOC cells treated for 24 h with different dosages of controlpeptide Ab35–25 (0–200 lg/ml). These immunotransfersshow that cell treatment with the control peptide doesnot alter the levels of a-adaptin, b-adaptin, Eps15, SR-AIor SR-BI (Fig. 2A and C). Nevertheless, expression levelsof the same proteins studied with RAW and EOC cells

Fig. 1. Cell oxidative stress. Fluorescence intensity analyzed by flow cytometry of (A) RAW and (B) EOC cells exposed for 24 h to Ab25–35 (0–200 lg/ml).RAW and EOC monolayers were grown in the presence of Ab25–35 and exposed to (A) 1 lM or (B) 20 lM 6-carboxy-H2DCFA, washed, scraped, andresuspended in PBS buffer. Fetal bovine serum (FBS). Values in this set of results are expressed as means of four different experiments ± standard error(*p < 0.05).

Fig. 2. Endocytic proteins. Expression levels of endocytic proteins a- and b-adaptin, Eps15, SR-AI, and SR-BI. (A) RAW cells treated with different doses(0–200 lg/ml) of Ab35–25 (control peptide), (B) RAW cells incubated with Ab25–35 (0–200 lg/ml), (C) EOC cells treated with Ab35–25, and (D) EOC cellsexposed at Ab25–35. All experiments were performed in serum-free medium after an incubation period of 24 h. Representative experiment of a series of 3.

N. Manzano-Leon et al. / Biochemical and Biophysical Research Communications 351 (2006) 588–594 591

treated for 24 h with Ab25–35 (0–200 lg/ml) show an impor-tant decline of b-adaptin levels in direct correlation to theAb25–35 concentration employed (Fig. 2B and D). A cleardecrease in the level of b-adaptin at a moderate concentra-tion of Ab25–35 (50 lg/ml) is evident. Protein band analysisfrom these experiments was consistent with this importantdecrease (data not shown).

The importance of b-adaptin in this process lies on thefact that it shows two clathrin-binding sites: a clathrin-box motif in the hinge region and a specific site in theear region [18] that participates in the selection of vesi-

cle-charged molecules [19]. On the other hand, RAWand EOC cells treated for 24 h with Ab25–35 show anincreased level of SR-BI in relationship with the concen-tration of Ab employed (Fig. 2B and D). Moreover,EOC cells also show an increase in a-adaptin accordingto the Ab25–35 dosage concentration used. We believechanges in the level of expression of this accessory pro-tein, as well as those observed for SR-BI, might be con-sidered as compensatory mechanisms in response to theimportant decrease in the level of b-adaptin during anoxidative stress state.

592 N. Manzano-Leon et al. / Biochemical and Biophysical Research Communications 351 (2006) 588–594

Interestingly, when ac-LDL was used as a ligand, thelevels of b-adaptin remained stable (data not shown). Thisresult seems to be of relevance since although both ligandsare recognized by the SR, only one is capable to affect b-adaptin levels. This might be related to the fact thatAb25–35 is capable to initiate and maintain an oxidativestress condition, in contrast to ac-LDL.

RT-PCR

Our study for the first time reports the effect of Ab25–35

exposure upon the SR, a-adaptin, b-adaptin, and clathrintranscript levels (Fig. 3). Interestingly, an increase inmRNA expression of a-adaptin and SR-BI is observed inaccordance with the Ab25–35 dosage employed in bothRAW and EOC cells. We believe this phenomenon againmight be considered as a cell response in an attempt to ren-der endocytosis in an effective way as a compensatorymechanism to the decrease in the availability of b-adaptin.

Since immunoblot assays indicate an important decreasein the signal shown by b-adaptin in both cell types (Fig. 2Band D) and mRNA expression assays do not show anychanges when cells are exposed to an oxidative stress state,

Fig. 3. Protein transcript levels. Ethidium bromide-stained gels of RT-PCR asscells treated for 24 h with Ab25–35 (0–200 lg/ml). Representative experiment o

Fig. 4. Interaction of endocytic proteins. Immunoprecipitation assays using an(20 lg/ml) for 0, 5, 10, 15, and 30 min in serum-free medium. Representative

our results indicate the possibility that damage is exertedupon b-adaptin at the translational or post-translationallevel.

Immunoprecipitation assays

Although several interactions between diverse proteinsof the endocytosis machinery are known in direct relation-ship to a series of receptors, there are no specific data ofthis phenomenon with respect to the SR. To conduct moredetailed experiments of the interactions that occur betweenproteins during the process of clathrin-mediated endocyto-sis of Ab25–35 through the SR, we carried out immunopre-cipitation assays.

These assays were performed with RAW and EOC cellsduring 0, 5, 10, 15, and 30 min using a single concentrationof Ab25–35 (20 lg/ml), and immunoprecipitating with anti-b-adaptin antibody. We found that b-adaptin immunopre-cipitates with a-adaptin in both cell lines even in theabsence of Ab (Fig. 4A and B). From this result, we con-clude that b-adaptin must probably interact directly witha-adaptin, in support of results that show these two pro-teins form part of the AP-2 complex [17,29]. Although

ays of a- and b-adaptin, clathrin, SR-AI, and SR-BI from RAW and EOCf a series of 3.

ti-b-adaptin antibody of (A) RAW and (B) EOC cells exposed to Ab25–35

immunoblots of a series of three independent experiments.

N. Manzano-Leon et al. / Biochemical and Biophysical Research Communications 351 (2006) 588–594 593

b-adaptin interacts with Eps15 also, independently of thepresence of the ligand in EOC cells, it seems that in the caseof RAW cells, Eps15 does not coimmunoprecipitate withb-adaptin.

On the other hand, SR-AI immunoprecipitates with b-adaptin in the absence of ligand in RAW cells (Fig. 4A).In contrast, the b-adaptin antibody is not able to immuno-precipitate SR-AI in EOC cells. Therefore, in the first case,this interaction can be considered a preformed conditionthat does not change in response to ligand binding. More-over, SR-BI immunoprecipitates with b-adaptin in bothcell lines only when Ab25–35 is used as a ligand (Fig. 4Aand B). In this case, the interaction between b-adaptinand SR-BI seems to be promoted by Ab (Fig. 4A and B).

In sum, we report b-adaptin as an important molecule ofthe protein machinery involved in clathrin-mediated endo-cytosis of the SR. The availability of b-adaptin decreasesnotably in response to Ab25–35 binding that in turn medi-ates a state of oxidative stress. Since binding of ligandssuch ac-LDL at concentrations like the ones used in thisstudy does not unchain oxidative stress by itself, in this casewe observe b-adaptin levels to remain constant. It is inter-esting to mention that although the degree of similaritybetween the scavenger A and scavenger B receptors isnon-existent, both receptors decrease their endocyticcapacity under the action of high concentrations of Ab.Since ligand binding is not affected, this result togetherwith previous findings from our laboratory [23], suggeststhat an impaired internalization of Ab is a consequenceof the lack of organization of the endocytic machinerydue to b-adaptin alteration, secondary to the establishmentof an oxidative stress state. Since RT-PCR assays showthat b-adaptin mRNA is normally synthesized in both celllines studied under oxidative stress, our results put forwardthe possibility that this condition might be affecting proteintranslation or directly altering the protein structure of b-adaptin. These possibilities are currently being exploredin our laboratory.

Our finding identifying b-adaptin as a key molecule inthe internalization of Ab advances the understanding ofthe possible mechanisms involved in the accumulation ofAb in the extracellular space of neurons from patients suf-fering of Alzheimer disease. The possibility this phenome-non might apply to other diseases such as atherosclerosis,a process started to be considered as an amyloid relateddisease, remains also to be studied.

Acknowledgments

This project was supported by CONACyT (Grant38551-N) and DGAPA-UNAM (Grant IN230303) to J.Mas-Oliva. We thank Paul Gaytan and Santiago Becerrafor synthesis of high quality oligonucleotides, Ma. ElenaGutierrez for word processing and Maggie Brunner foreditorial revision.

References

[1] M. Krieger, The other side of scavenger receptors: pattern recognitionfor host defense, Curr. Opin. Lipidol. 8 (1997) 275–280.

[2] V. Terpstra, E.S. van Amersfoort, A.G. van Velzen, J. Kuiper, T.J.C.van Berkel, Hepatic and extrahepatic scavenger receptors: function inrelation to disease, Arterioscler. Thromb. Vasc. Biol. 20 (2000)1860–1872.

[3] M.P.J. de Winther, K.W. van Dijk, L.M. Havekes, M.H. Hofker,Macrophage scavenger receptor class A: a multifunctional receptor inatherosclerosis, Arterioscler. Thromb. Vasc. Biol. 20 (2000) 290–297.

[4] D.M. Paresce, R.N. Ghosh, F.R. Maxfield, Microglial cells internal-ize aggregates of the Alzheimer’s disease amyloid beta-protein via ascavenger receptor, Neuron 17 (1996) 553–565.

[5] C.J. Pike, D. Burdick, A.J. Walencewicz, C.G. Glabe, C.W. Cotman,Neurodegeneration induced by b-amyloid peptides in vitro: the role ofpeptide assembly state, J. Neurosci. 13 (1993) 1676–1687.

[6] S. Esbach, M.F. Stins, A. Brouwer, P.J. Roholl, T.J. van Berkel, D.L.Knook, Morphological characterization of scavenger receptor-med-iated processing of modified lipoproteins by rat liver endothelial cells,Exp. Cell Res. 210 (1994) 62–70.

[7] K. Morimoto, Y. Wada, J. Hinagata, T. Imanishi, T.U. Kodama, T.Doi, VXFD in the cytoplasmic domain of macrophage scavengerreceptors mediates their efficient internalization and cell-surfaceexpression, Biol. Pharm. Bull. 22 (1999) 1022–1026.

[8] S.L. Acton, P.E. Scherer, H.F. Lodish, M. Krieger, Expressioncloning of SR-BI, a CD36-related class B scavenger receptor, J. Biol.Chem. 269 (1994) 21003–21009.

[9] J. Hussemann, J.D. Loike, T. Kodama, S.C. Silverstein, Scavengerreceptor class type (SR-BI) mediates adhesion of neonatal murinemicroglia to fibrillar b-amyloid, J. Neuroimmunol. 114 (2001)142–150.

[10] M. Krieger, Charting the fate of the ‘‘good cholesterol’’: identificationand characterization of the high-density lipoprotein receptor SRBI,Annu. Rev. Biochem. 68 (1999) 523–558.

[11] D. Rhainds, L. Falstrault, C. Tremblay, L. Brissette, Uptake and fateof class B scavenger ligands in HepG2 cells, Eur. J. Biochem. 261(1999) 227–235.

[12] G.A. Graf, P.M. Connell, D.R. van der Westhuyzen, E.J. Smart, Theclass B, type I scavenger receptor promotes the selective uptake ofhigh density lipoprotein cholesterol ethers into caveolae, J. Biol.Chem. 274 (1999) 12043–12048.

[13] G.H. Hansen, L.L. Niels-Christiansen, L. Immerdal, E.M. Danielsen,Scavenger receptor class B type I (SR-BI) in pig enterocytes:trafficking from the brush border to lipid droplets during fatabsortion, Gut 52 (2003) 1424–1431.

[14] I.S. Coraci, J. Husemann, J.W. Berman, C. Hulette, J.H. Dufor, G.K.Campanella, A.D. Luster, S.C. Silverstein, J.B. El-Khoury, CD36, aclass B scavenger receptor, is expressed on microglia in Alzheimer’sdisease brains and can mediate production of reactive oxygen speciesin response to beta-amyloid fibrils, Am. J. Pathol. 160 (2002) 101–112.

[15] S.A. Mousavi, L. Malerod, R. Kjeken, Clathrin-dependent endocy-tosis, Biochem. J. 377 (2004) 1–16.

[16] E. ter Haar, S.C. Harrison, T. Kirchhausen, Peptide-in-grooveinteractions link target proteins to the beta-propeller of clathrin,Proc. Natl. Acad. Sci. USA 97 (2000) 1096–1100.

[17] B.M. Pearse, C.J. Smith, D.J. Owen, Clathrin coat construction inendocytosis, Curr. Opin. Struct. Biol. 10 (2000) 220–228.

[18] D.J. Owen, Y. Vallis, B.M. Pearse, H.T. McMahon, P.R. Evans, Thestructure and function of the beta 2-adaptin appendage domain,EMBO J. 19 (2000) 4216–4227.

[19] I. Rapoport, Y.C. Chen, P. Cupers, S.E. Shoelson, T. Kirchhausen,Dileucine-based sorting signals bind to the beta chain of AP-1 at a sitedistinct and regulated differently from the tyrosine-based motif-binding site, EMBO J. 19 (1998) 2148–2155.

[20] F. Tebar, S.K. Bohlander, A. Sorkin, Clathrin assembly lymphoidmyeloid leukemia (CALM) protein: localization in endocytic-coated

594 N. Manzano-Leon et al. / Biochemical and Biophysical Research Communications 351 (2006) 588–594

pits, interactions with clathrin, and the impact of over expression onclathrin-mediated traffic, Mol. Biol. Cell 10 (1999) 2687–2702.

[21] M.G. Ford, B.M. Pearse, M.K. Higgins, Y. Vallis, D.J. Owen, A.Gibson, C.R. Hopkins, P.R. Evans, H.T. McMahon, Simultaneousbinding of PtdIns(4,5)P2 and clathrin by AP180 in the nucleation ofclathrin lattices on membranes, Science 291 (2001) 1051–1055.

[22] A.E. Salcini, H. Chen, G. Iannolo, P. De Camilli, P.P. Di Fiore,Epidermal growth factor pathway substrate 15, Eps15, Int. J.Biochem. Cell Biol. 31 (1999) 805–809.

[23] R. Aguilar-Gaytan, J. Mas-Oliva, Oxidative stress impairs endocyto-sis of the scavenger receptor class A, Biochem. Biophys. Res.Commun. 305 (2003) 510–517.

[24] J. Hirrlinger, J.M. Gutterer, L. Kussmaul, B. Hamprecht, R.Dringen, Microglial cells in culture express a prominent glutathionesystem for the defense against reactive oxygen species, Dev. Neurosci.22 (2000) 384–392.

[25] X.Q. Xiao, H.Y. Zhang, X.C. Tang, Huperzine A attenuates amyloidbeta-peptide fragment 25–35-induced apoptosis in rat cortical neu-rons via inhibiting reactive oxygen species formation and caspase-3activation, J. Neurosci. Res. 67 (2002) 30–36.

[26] H. Zeng, Q. Chen, B. Zhao, Genistein ameliorates beta-amyloidpeptide (25–35)-induced hippocampal neuronal apoptosis, Free Rad-ic. Biol. Med. 36 (2004) 180–188.

[27] M.H. Jang, S.B. Jung, M.H. Lee, C.J. Kim, Y.T. Oh, I. Kang, J. Kim,E.H. Kim, Melatonin attenuates amyloid beta25–35-induced apop-tosis in mouse microglial BV2 cells, Neurosci. Lett. 380 (2005) 26–31.

[28] P.J. Yao, M. Zhu, E.I. Pyun, A.I. Brooks, S. Therianos, V.E. Meyers,P.D. Coleman, Defects in expression of genes related to synapticvesicle trafficking in frontal cortex of Alzheimer’s disease, Neurobiol.Dis. 12 (2003) 97–105.

[29] J.E. Heuser, J. Keen, Deep-etch visualization or proteins involved inclathrin assembly, J. Cell Biol. 107 (1988) 877–886.