Acute Ethanol Treatment Modulates Toll-like Receptor-4

Association with Lipid Rafts

Angela Dolganiuc, Genadyi Bakis, Karen Kodys, Pranoti Mandrekar, and Gyongyi Szabo

Background: Alcohol, a substance that is most frequently abused, suppresses innate immune re-sponses to microbial pathogens. The host senses pathogens via Toll-like receptors (TLRs). Recentstudies indicate that alcohol affects TLR signaling.

Methods: Here, we hypothesized that acute alcohol treatment may interfere with early steps ofmembrane-associated TLR2 and TLR4 signaling at the level of lipid rafts. Human monocytes andChinese hamster ovary (CHO) cells, transfected with human TLR2, TLR4, or CD14, were stimulatedwith peptidoglycan (PGN, TLR2 ligand) or lipopolysaccharide (LPS, TLR4 ligand) with or withoutalcohol (50 mM) and analyzed for cytokine production (enzyme-linked immunosorbent assay),nuclear factor-kB (NF-kB) activation (electrophoretic mobility shift assay), membrane fluidity (flu-orescent pyrene eximer formation), and partition of cellular membrane into cholesterol-rich, deter-gent-resistant domains (DRMs; Western blot).

Results: We determined that both TLR2 and TLR4 were located outside the rafts; flotillin, aDRMmarker, was resident in the rafts, while CD14 was equally distributed in and outside the rafts ina steady-state condition. PGN forced TLR2 to migrate into DRMs. Engagement of TLR4 and CD14with LPS induced their migration into the rafts. Alcohol prevented TLR4 partitioning; however, itdid not affect TLR2 migration into the rafts. Furthermore, alcohol downregulated TLR4-induced,but not TLR2-induced, NF-kB activation and cytokine production in monocytes. We found that al-cohol increased membrane fluidity and depleted cellular cholesterol in CHO cells without affectingcell viability.

Conclusions: These data demonstrate for the first time that alcohol disturbs TLR4 and CD14 as-sociation with lipid rafts. We propose that alcohol-induced effects on lipid rafts may contribute tomodulation of TLR4-CD14–triggered early cellular responses.

Key Words: TLR, Rafts, Ethanol, Fluidity, Cholesterol, NF-kB.

ALCOHOL IS A substance that is most frequentlyabused (Gunzerath et al., 2004; Russo et al., 2004).

We, and others, previously reported that alcohol affectsimmune responses (Chiappelli et al., 1995; Choudhryet al., 2004; Cook et al., 2004; Friedman et al., 2003; Goraland Kovacs, 2005; Mandrekar et al., 2004; Nelson andKolls, 2002; Pruet et al., 2004a, 2004b; Szabo et al., 2004).Acute alcohol intake impairs immune functions by inter-fering with inflammatory responses; however, its most det-rimental influence is seen when accompanied by an acutebacterial or viral infection and/or trauma (Boe et al., 2003;Faunce et al., 1998; Haorah et al., 2004; Nelson et al.,1989; Zhang et al., 2002). Acute alcohol abolishes the im-

mune systems’ capacity to clear Streptococcus pneumoniae,aggravates sepsis, and increases the risk of complicationsin trauma and burn patients (Boe et al., 2003; Faunceet al., 1998; Friedman et al., 2003; Lois et al., 1999). Ex-posure to acute alcohol inhibits lipopolysaccharide (LPS)-induced nuclear factor-kB (NF-kB) binding to DNA andinterferes with production of proinflammatory cytokinestumor necrosis factor-a (TNFa) and interleukin-6 in hu-man and murine macrophages, impairs the function ofdendritic cells, and leads to T-cell anergy (Goral and Ko-vacs, 2005; Mandrekar et al., 2002, 2004; Pruett et al.,2004a; Szabo et al., 2004).The host senses pathogens via phylogenetically pre-

served, germ-line–encoded receptor molecules called Toll-like receptors (TLRs) (Takeda et al., 2003). Ten humanTLRs have been described to date, all with similar back-bone structure but with distinct ligand specificity. TLRsrecognize structural components that are conservedamong classes of microorganisms (Kopp and Medzhitov,2003). Engagement of TLRs by their ligands culminates inactivation of NF-kB and production of proinflammatorycytokines; thus, TLRs play a crucial role in the initiation ofthe inflammatory process (Dolganiuc et al., 2004; Goraland Kovacs, 2005; Kirschning and Schumann, 2002; Kopp

From the Department of Medicine, University of Massachusetts Med-ical School, Worcester, Massachusetts.

Received for publication April 27, 2005; accepted September 21, 2005.Supported by PHS grant AA11576 and by the resources of the UMass

Medical School CFAR and DERC core facilities.

Reprint requests: Gyongyi Szabo, MD, PhD, Professor of Medicine,University of Massachusetts Medical School, 364 Plantation Street,LRB 215, Worcester, MA 01605-2324; Fax: 508-856-4770; E-mail:gyongyi.szabo@umassmed.edu

Copyright r 2006 by the Research Society on Alcoholism.

DOI: 10.1111/j.1530-0277.2006.00003.x

Alcohol Clin Exp Res, Vol 30, No 1, 2006: pp 76–8576

ALCOHOLISM: CLINICAL AND EXPERIMENTAL RESEARCH Vol. 30, No. 1January 2006

and Medzhitov, 2003; Pruett et al., 2004; Takeda et al.,2003). Recent publications show that alcohol impairs cel-lular activation via TLR2, TLR3, TLR4, and TLR9(Goral and Kovacs, 2005; Mandrekar et al., 2002; Pruettet al., 2004a, 2004b).Extensive work over the past decades has demonstrated

that the plasma membrane is not a random formation oflipids; rather, it is organized in protein-bearing, glycopho-spholipid, and cholesterol-rich microdomains called lipidrafts (Mayor and Rao, 2004; Simons and Vaz, 2004; Singerand Nicolson, 1972). Depending on cell type, lipid raftsrepresent �10% to 50% of cellular membrane surface andare defined functionally based on their mobility (Hao etal., 2001). Lipid rafts are also called detergent-resistantmembranes (DRMs) because of their in vitro property toresist solubilization in nonionic detergents at low temper-ature (Brown and Rose, 1992). Based on this property,lipid rafts float and concentrate in low-density sucrose gra-dient upon centrifugation. This procedure yields a consist-ent DRM product that is enriched in cholesterol and raftmarker proteins such as flotillin and glycosylphosohatidy-linositol (GPI)-linked proteins, along with other proteinsthat are associated with raft at the time of solubilization(Brown and Rose, 1992; Legler et al., 2003; Li et al., 2003).Among the known raft-associated proteins to date are re-ceptors, such as T-cell and B-cell receptors, costimulatorymolecules CD28 and CD80, major histocompatibilitycomplex class II molecules, human leukocyte antigen(HLA)-DR, tetraspanin CD81, CD95 death-inducingsignaling complex, and TNFR1, but also signaling pro-teins such as signal transducer and activator of transcrip-tion proteins, mitogen-activated protein kinase, proteinkinase C, and PI3 kinase (Goebel et al., 2002; Harder,2004; Li et al., 2003; Meyer zum Bueschenfelde et al., 2004;Pierce, 2002; Schmitz and Orso, 2002). Some of the raft-associated proteins are raft residents, while others are re-cruited upon activation with receptor-specific ligands. Ow-ing to their capacity to concentrate multiple proteins inone tight space, lipid rafts facilitate ‘‘receptor cluster’’ for-mation and play a major role in the regulation of cellsignaling (Goebel et al., 2002; Harder, 2004; Pierce, 2002;Pike, 2003; Schmitz and Orso, 2002). Recent reports havesuggested that TLR2 and TLR4 associate with lipid rafts(Pfeiffer et al., 2001; Schmitz and Orso, 2002; Triantafilouet al., 2002). Furthermore, stimulation with TLR2 orTLR4 ligands leads to the formation of receptor clustersand activation of intracellular signaling machinery thatculminates in cell activation and cytokine production (Do-lganiuc et al., 2004; Goral and Kovacs, 2005; Mandrekaret al., 2002; Pfeiffer et al., 2001; Pruett et al., 2004b; Take-da et al., 2003; Triantafilou et al., 2002).We, and others, have previously shown that alcohol al-

ters TLR signaling pathways (Goral and Kovacs, 2005;Mandrekar et al., 2002, 2004; Pruett et al., 2004a). Here,we investigated the hypothesis that acute alcohol couldaffect the association of TLRs expressed on the cellular

membrane surface with lipid rafts and, thus, influence theoutcome of the cellular activation by pathogens. Weshowed that alcohol impairs TLR4 and CD14, but notTLR2, recruitment to lipid rafts. This correlated with lowNF-kB binding and impaired TNFa production upon en-gagement of TLR4, but not TLR2, in the presence of al-cohol. Our results suggest that alcohol may influence hostdefense by modulating lipid rafts.

MATERIALS AND METHODS

Blood Donors

The study was reviewed and approved by the Institutional Com-mittee for Protection of Human Subjects in Research. The donorswere healthy individuals, aged 18–60 years, females (n5 5) and males(n5 4) who had no previous chronic alcohol abuse history, con-sumed o6 and o9 drinks/w, respectively, and had no alcohol for atleast 48 hours, before blood donation. Blood was obtained by veni-puncture and anticoagulated by heparin. Peripheral blood mononu-clear cells (PBMCs) were separated by centrifugation in Ficollgradient and monocytes were separated from PBMCs by adherenceto plastic, as previously described (Mandrekar et al., 2002).

Cell Lines

Chinese hamster ovary (CHO) cells stably transfected with humanCD14, TLR2, or TLR4 (CHO/CD14, CHO/TLR2, or CHO/TLR4,respectively) were kindly provided by Dr. Golenbock (University ofMassachusetts Medical School). The cells were grown in F12 me-dium supplemented with 10% FBS and selective antibiotics and sub-cultured 3 times a week.

DRM Isolation by Flotation Method

CHO cells were seeded in six-well plates, grown to 70% to 80%confluency, and then stimulated for 1 hour as indicated in the figurelegends. After stimulation, the cells were washed with ice-cold phos-phate-buffered saline (PBS) and lysed in TNE buffer [25 mM Tris-HCl (pH 7.4), 150 mMNaCl, 5 mM ethylenediaminetetraacetic acid]supplemented with N-ethylmaleimide (2.5 mg/ml); protease inhibi-tors including phenylmethanesulfonyl fluoride, aprotinin, antipain,and leupeptin (all Sigma-Aldrich, St. Louis, MO); and 1% Triton X-100 on ice for 30 minutes No alcohol was present during raft extrac-tion to prevent the potential interaction of alcohol with detergent.The extracts were homogenized by passing through a 20-gauge nee-dle (Precision Glide, Becton Dickinson&Co, Franklin Lakes, NJ)and centrifuged for 15 seconds at 1,000 g at 14 1C. The postnuclearsupernatant was kept on ice for an additional 30 minutes, mixed withice-cold 60% sucrose (w/v) to produce a final 50% sucrose solution(final volume 4 ml), placed on top of 1 ml of 70% sucrose (w/v) in apolyallomer tube, and overlayered with 1 ml of 30% sucrose (w/v), 1ml of 15% sucrose (w/v), and 2 ml of 5% sucrose (w/v) (all ice-cold,dissolved in TNE buffer). The samples were centrifuged in an SW49Ti rotor at 100,000�g for 18 hours at 14 1C. Fractions of 500 mlfrom all density layers (except 50% sucrose layer that was collectedin four fractions of 1 ml each) (total of 14 fractions) were collectedfrom centrifuged density gradient and analyzed fresh for the pres-ence of proteins by Western blot.

Cholesterol Depletion

Cells were washed with serum-free warm medium and incubatedwith 5 mM methyl-b-cyclodextrin (MbCD; Sigma) in OptiMEM for1 hour at 37 1C. After incubation, cells were washed with ice-coldPBS, and cellular extracts were prepared as described above. Cho-

77ALCOHOL INHIBITS TLR4 ASSOCIATION WITH LIPID RAFTS

lesterol content in culture medium was determined using the AmplexRed Cholesterol Assay kit (Molecular Probes Inc, Eugene, OR), asrecommended by the manufacturer.

Protein Analysis

Fifty microliters of each density gradient fraction was mixed with2� gel sample buffer [125 mM Tris-HCl (pH 6.8), 2% sodium do-decyl sulfate, 10% glycerol, 1 M b-mercaptoethanol], and the pro-teins were separated in 10% sodium dodecyl sulfate–polyacrylamidegel electrophoresis. After electrophoresis, the gels were equilibratedin transfer buffer [25 mM Tris-HCl (pH 8.2), 192 mM glycine, 10%methanol] and transferred to BioTrace (Pall, Persacola, FL) nitro-cellulose membranes. The membrane was blocked in PBS with 0.5%Tween 20 and 10% nonfat dried milk at 14 1C, washed, and then in-cubated overnight with anti-TLR2, -TLR4, -CD14 (all Santa Cruz,Santa Cruz, CA), or –flotillin (Abcam, Cambridge, MA) antibodiesfollowed by secondary horseradish peroxidase–conjugated antibod-ies (Santa Cruz). The proteins were detected using the enhancedchemiluminescence detection reagent system (LumiGlo, Cell Signa-ling, Beverly,MA). Densitometric analysis was performed for proteinquantification using the GDS-800 system and Labwork 4.0 program(UVP BioImaging Systems, Upland, CA).

Electrophoretic Mobility Shift Assay

To determine NF-kB activation, nuclear extracts from CHOcells and monocytes were prepared as previously described (Haorahet al., 2004). Consensus double-stranded NF-kB oligonucleotides(50-AGTTGAGGGGACTTTCCCAGGC-3 0) from Oligos Etc.(Wilsonville, OR) were labeled with [g32P]ATP (DuPont-NEN) us-ing T4 polynucleotide kinase. Radioactive-labeled oligonucleotideswere purified on a polyacrylamide copolymer column (Bio-Rad,Hercules, CA). Five micrograms of nuclear protein from each exper-imental sample was incubated with radioactive-labeled oligonucleo-tides and NF-kB/DNA complexes were separated in 7.5%polyacrylamide gel. The specificity of the NF-kB band was deter-mined using 5� excess of cold oligonucleotide (competition). Driedgels were subjected to autoradiography. Densitometric analysis ofthe NF-kB band was performed using the GDS-800 system and Lab-work 4.0 program (UVP BioImaging Systems).

Cytokine Production

Two million monocytes per microliter were cultured overnight inRPMI 1640 (Invitrogen Life Technologies, Gaithersburg, MD) sup-plemented with 10% FBS (Defined, from HyClone, Logan, UT) andthen stimulated with lipoprotein-free, phenol-extracted purified LPS(pLPS, List Biologicals, Campbell, CA) or Staphyolococcus aureuspeptidoglycan (PGN, fromFluka,Milwaukee,WI) with or without 50mM ethanol for 12 hours. Cell-free culture supernatants were harvest-ed and analyzed for TNFa production by enzyme-linked immunosor-bent assay (BD Bioscience, San Jose, CA), as recommended by themanufacturer.

Membrane Fluidity

Membrane fluidity was determined using the Membrane FluidityKit (MarkerGene Technologies Inc, Eugene, OR), according to themanufacturer’s recommendations. The assay was based on the abil-ity of cellular membranes to form more and less rigid domains andthe property of lipophylic pyrene probes to undergo fluorescenteximer formation upon spatial interaction. By measuring the ratioof monomer to eximer fluorescence, a quantitative monitoring ofthe membrane fluidity was attained. Briefly, cells were loaded withfluorescent lipid reagent (pyrene decanoic acid, 2 mM) or nonfluores-cent lipid reagent (pyrene, 2 mM), washed extensively, and then stim-ulated with ethanol or MbCD for 1 hour at 37 1C in a 5% CO2

atmosphere. After incubation, the cells were analyzed using a spec-

trophotometer with excitation at 360 nm. Recording the fluorescenceemission at 405 and 450 nm, respectively, monitored the formationof monomers and eximers. The membrane fluidity was determined asthe ratio of mean fluorescence intensity of the monomer/eximer (405/450 nm) in every experimental group. The percent change of mem-brane fluidity of the experimental groups compared with the control(media) group was calculated.

Cell Viability

Cell viability was determined using the ApoTarget kit (Biosource,Camarillo, CA) based on annexin V–fluorescein isothiocyanate andpropidium iodide (PI) staining, as recommended by the manufactur-er. Cell fluorescence was analyzed by flow cytometry.

Statistical Significance

Statistical significance of changes between appropriate groups wascalculated using the two-tailed t-test and Wilcoxon signed-rank non-parametric data analysis. Data are represented as mean � SE.

RESULTS

Acute Ethanol Treatment Inhibits NF-jB Activation andCytokine Production Triggered by a TLR4, but Not aTLR2, Agonist

We have previously reported that ethanol interferes withTLR4 signaling. Specifically, LPS-stimulated humanmon-ocytes treated in vitro with acute alcohol showed lowerNF-kB binding to DNA and reduced TNFa productionthan control cells (Mandrekar et al., 2002). Here, we in-vestigated the influence of alcohol on the signaling path-way of TLR2 and TLR4. Human monocytes express bothTLR2 and TLR4 on their membrane (Mandrekar et al.,2002, 2004). We used PGN to stimulate TLR2 and phenol-purified, lipoprotein-free LPS (LPS) as a ligand for TLR4.Both PGN and LPS induced NF-kB activation (Fig. 1Aand 1B) and TNFa production (Fig. 1C) in human mon-ocytes. In agreement with our previous results (Mandrekaret al., 2002), alcohol inhibited both NF-kB activation andTNFa production when added with LPS. However, to oursurprise, alcohol did not influence PGN-induced cell acti-vation (Fig. 1). Chinese hamster ovary cells stably trans-fected with human TLR2 or with human CD14 (CHO/TLR2 or CHO/CD14, respectively) represent a suitablesurrogate for monocytes in the investigation of TLR func-tions (Mandrekar et al., 2002; Schroder et al., 2003;Triantafilou et al., 2002). Indeed, stimulation of CHO/CD14 cells with LPS induced NF-kB activation (Fig. 2A).CD14 is linked to the cellular membrane via its GPI do-main and has no intracellular domain; thus, CD14 alonecannot activate cells in the presence of LPS. However,CHO cells express endogenous TLR4 that facilitates LPSsignal transduction (Mandrekar et al., 2002). Stimulationwith PGN exclusively induced NF-kB activation in CHO/TLR2 cells. The addition of LPS also stimulated CHO/TLR2 cells (Fig. 2A) owing to the endogenous expressionof TLR4 in CHO cells. These results suggested that CHOcells expressing TLRs and CD14 closely resemble TLRfunctions seen in monocytes, as previously reported by us,

78 DOLGANIUC ET AL.

and others (Mandrekar et al., 2002; Triantafilou et al.,2002). Similar to monocytes (Fig. 1), alcohol inhibitedTLR4-induced cell activation and did not affect TLR2-induced signaling in CHO cells (Fig. 2A). Both TLR2 andTLR4 share a common intracellular signaling pathway(Kopp andMedzhitov, 2003; Takeda et al., 2003). Becausealcohol preferentially inhibited TLR4-mediated, but notTLR2-mediated, signals; this raised the possibility thatthere may be differences in TLR2 and TLR4 recruitmentto the signaling complex.

Acute Alcohol Treatment Prevents Ligand-Induced TLR4,but Not TLR2, Recruitment to Lipid Rafts

Association with lipid rafts represents early events thatoccur immediately after receptor engagement with a spe-

cific ligand (Harder, 2004; Legler et al., 2003; Meyer zumBueschenfelde et al., 2004; Pierce, 2002). We investigatedwhether TLR2, TLR4, and CD14 associated with the lipidrafts. Chinese hamster ovary cells were extracted with non-ionic detergent in cold, and DRMs, representative of lipidrafts, were separated by centrifugation in sucrose gradient.The presence of flotillin, a known constitutive marker ofDRMs (Brown and Rose, 1992; Li et al., 2003; Neumann-Giesen et al., 2004; Pike, 2003), was detected in the low-density sucrose gradient fractions (Fig. 2B). As expected,treatment of cells with MbCD led to raft disruption(Brown and Rose, 1992; Hao et al., 2001); thus, flotillinwas displaced and was now detectable in the high-densitysucrose gradient fractions (Fig. 2B). Stimulation with al-cohol alone minimally affected the association of flotillinwith lipid rafts. The presence of alcohol along with LPS orPGN (data not shown) did not affect flotillin’s associationwith lipid rafts.In the steady-state condition (medium control), TLR2

receptors were identified in high-density sucrose gradientfractions (Fig. 3A), suggesting that in the absence of aspecific ligand, TLR2 does not associate with lipid

Freeprobe

me

dia

PG

N

PG

N+

EtO

H

EtO

H

LPS

LPS

+E

tOH

com

petit

ion

NFκB

A

B

0

20

40

60

80

Den

sity

uni

ts

med

ia

PG

N

PG

N+

EtO

H

EtO

H

LPS

LPS

+E

tOH

med

ia

PG

N

PG

N+

EtO

H

EtO

H

LPS

LPS

+E

tOH

com

petit

ion

p<0.016

012345

TN

Fα

(ng

/ml)

p<0.028

C

Fig. 1. Alcohol inhibits Toll-like receptor (TLR)4- but not TLR2-mediatedcell activation. (A) Human monocytes were stimulated with lipopolysaccharide(LPS) (0.1 mg/ml) or peptidoglycan (PGN) (1 mg/ml) with or without 50 mMethanol (EtOH) for 1 hour, washed, and lysed into cytoplasmic and nuclearfractions as indicated in the Materials and Methods section. Five micrograms ofnuclear protein was incubated with radioactive-labeled nuclear factor-kB (NF-kB) oligonucleotide, protein/DNA complex was separated from free probe inpolyacrylamide gel, and dried gel was exposed to autoradiography. Thespecificity of the NF-kB band was confirmed using 5� excess of cold oligonuc-leotide (competition). One representative gel (A) and densitometric analysis ofsix gels (B, mean � SE) are shown. (C) Human monocytes (2�106/ml) werestimulated overnight and cell-free culture supernatants were analyzed for tumornecrosis factor-a (TNFa) production by enzyme-linked immunosorbent assay.Mean � SE value from 6 experiments using cells from 9 individuals is shown.

CHO/CD14/TLR2

med

ia

LPS

LPS

+E

tOH

PG

N

med

ia

PG

N

PG

N+

EtO

H

LPS

CHO/CD14

0

80

204060

Den

sity

uni

ts

NFκB

A

Media

EtOH

MβCD

FlotillinB

Non-raft Raft

Fig. 2. (A) Human Toll-like receptors (TLRs) expressed in Chinese ham-ster ovary (CHO) cells are functional. CHO/CD14 and CHO/TLR2 cells werestimulated with lipopolysaccharide (LPS) (0.1 mg/ml) or peptidoglycan (PGN)(1 mg/ml) with or without 50 mM ethanol for 1 hour, washed, and lysed intocytoplasmic and nuclear fractions as indicated in the Materials and Methodssection. Nuclear factor kB (NF-kB) activation was assessed as in Fig. 1. Onerepresentative gel (top) and densitometric analysis of 5 with similar results(bottom) are shown. (B) Flotillin is indicative of detergent-resistant mem-branes (rafts). CHO/CD14 cells were stimulated as indicated (50 mM ethanol,EtOH; 5 mM methyl-b-cyclodextrine, MbCD) for 1 hour, lysed, and subjectedto centrifugation in sucrose density gradient. Proteins from sucrose fractionswere separated in a 7.5% polyacrylamide gel and transferred to a nitrocellu-lose membrane. Immunostaining was performed with anti-flotillin primary Abs,followed by incubation with secondary Ab conjugated to horseradish per-oxidase. Immunoreactive bands were visualized using enhanced chemilumi-nescence reagents. Results from a representative Western blot of three blotsare shown.

79ALCOHOL INHIBITS TLR4 ASSOCIATION WITH LIPID RAFTS

rafts. Upon stimulation with PGN, TLR2 partitionedinto lipid-enriched domains, as most of the protein migrat-ed from high-density sucrose gradient fractions into low-density sucrose gradient fractions representing lipid rafts(Fig. 3B). The presence of 50 mM ethanol during thestimulation with PGN did not affect TLR2 associationwith lipid rafts (Fig. 3C). Disruption of lipid rafts withMbCD prevented PGN-induced TLR2 migration into lip-id rafts (Fig. 3D).Similar to TLR2, TLR4 was not associated with lipid

rafts in the steady-state condition (medium control, Fig.4A). Stimulation with LPS for 1 hour induced TLR4translocation from high-density sucrose gradient fractionsinto the low-density sucrose gradient fractions, suggestingthat upon engagement with a specific ligand, TLR4 mi-grates into DRMs of the cellular membrane (Fig. 4B).However, the addition of alcohol to LPS-stimulated cellschanged the LPS-induced distribution of TLR4 from high-density into low-density sucrose gradient fractions, thusincreasing residual TLR4 in the nonraft fractions (Fig.4C). These data suggested that alcohol prevented LPS-triggered TLR4 association with lipid rafts. Disruption ofDRMs with MbCD led to the sequestration of TLR4 out-side the rafts, suggested by the presence in high-density su-crose gradient fractions (Fig. 4D).CD14 is a coreceptor for TLR4 that enhances TLR4-

mediated cell activation (Akashi et al., 2003; Kopp andMedzhitov, 2003; Takeda et al., 2003). We found thatCD14 was constitutively associated with lipid rafts (Fig.5A). Upon stimulation with LPS, we identified a pattern ofcellular redistribution of CD14 that favored migration of

CD14 from mid-density sucrose fractions into DRM-rep-resenting low-density sucrose gradient fractions (Fig. 5B).More important, the presence of alcohol during stimula-tion with LPS prevented cellular redistribution of CD14(Fig. 5C). Disruption of DRMs with MbCD resulted inthe displacement of CD14 from low-density sucrose frac-tions into high-density fractions (Fig. 5D). Overall, byplotting the data as a fraction of total density obtained bysummation of the density in all raft or nonraft fractions,we identified that alcohol significantly prevented TLR4migration into lipid rafts (Fig. 6). Based on these data, weconcluded that alcohol inhibited TLR4, but not TLR2, re-cruitment to the lipid raft upon engagement with a specificligand. Furthermore, alcohol prevented LPS-induced re-distribution of CD14 within the cellular membrane, sup-porting the idea that the LPS receptor complex is affectedby alcohol.

Alcohol Increases Cellular Membrane Fluidity andDisplaces Cellular Cholesterol

To investigate the mechanisms by which alcohol influ-enced TLR translocation into the lipid rafts, we exploredthe possibility that alcohol may cause lipid raft disruptionand change membrane fluidity. We labeled the cells witheximer-forming lipid and analyzed membrane fluidity inthe presence of alcohol. Increasing alcohol concentrationsand eximer formation showed a linear relationship, indi-cating that alcohol increased cellular membrane fluidity(Fig. 7A). Maximum fluidity of the membrane (�150% ofcontrol cells) was observed in MbCD-treated cells that

TLR2media

Non-raft Raft

0102030405060

APGN

Non-raft Raft

1030

50

70

90

Den

sity

un

its

B

PGN+ethanol

Non-raft Raft

0

20

40

60

80

CPGN+MβCD

Non-raft Raft

Den

sity

un

its

Den

sity

un

its

Den

sity

un

its

0

40

80

120

D

Fig. 3. Peptidoglycan (PGN) engages Toll-like receptor (TLR)2 into lipid rafts. Chinese hamster ovary (CHO)/TLR2 cells were stimulated with PGN (1 mg/ml)with or without 50 mM ethanol (EtOH) or 5 mM methyl-b-cyclodextrine (MbCD) for 1 hour, lysed, and subjected to centrifugation in sucrose density gradient.Proteins from sucrose fractions were separated in a 7.5% polyacrylamide gel and transferred to a nitrocellulose membrane. TLR2 protein detection was per-formed using anti-TLR2 primary Abs, followed by secondary Ab conjugated to horseradish peroxidase. The bands corresponding to TLR2 were visualized usingenhanced chemiluminescence reagents. One representative Western blot and densitometric analysis of 3 blots are shown.

80 DOLGANIUC ET AL.

were used as positive control. In contrast, neither alcohol-nor MbCD-induced changes were observed in the fluores-cence of cells labeled with nonfluorescent lipids that were

used as negative control of the assay. Because both alcoholand MbCD increased membrane fluidity and MbCD isknown for its property to displace cellular cholesterol, we

CD14

020406080

100120

media

A

Den

sity

un

its

Den

sity

un

its

Den

sity

un

its

Den

sity

un

its

LPS

020406080

100120

B

020406080

100120140

LPS+ethanol

Non-raft Raft

C D

LPS+MβCD

020406080

100120140 Non-raft Raft

Non-raft Raft Non-raft Raft

Fig. 5. Alcohol affects lipopolysaccharide (LPS)-induced CD14 partition into lipid rafts. Chinese hamster ovary (CHO)/CD14 cells were stimulated with LPS(0.1 mg/ml) with or without 50 mM ethanol (EtOH) or 5 mM methyl-b-cyclodextrine (MbCD) for 1 hour, lysed, and subjected to centrifugation in sucrose densitygradient. Proteins from sucrose fractions were separated in a 7.5% polyacrylamide gel and transferred to a nitrocellulose membrane. CD14 detection wasperformed using anti-CD14 primary Abs, followed by horseradish peroxidase–labeled secondary Ab. The bands corresponding to CD14 were visualized usingenhanced chemiluminescence reagents. One representative Western blot and densitometric analysis of 3 blots are shown.

TLR4media

0

40

80

120

A

0

20

40

60

LPSB

LPS+MβCD

0

40

80

120Non-raft Raft

D

0

20

40

60

LPS+ethanol

Non-raft Raft

Non-raft RaftNon-raft Raft

C

Den

sity

un

its

Den

sity

un

its

Den

sity

un

its

Den

sity

un

its

Fig. 4. Alcohol prevents lipopolysaccharide (LPS)-triggered Toll-like receptor (TLR)4 migration into lipid rafts. Chinese hamster ovary (CHO)/TLR4 cellswere stimulated with LPS (0.1 mg/ml) with or without 50 mM ethanol (EtOH) or 5 mM methyl-b-cyclodextrine (MbCD) for 1 hour, lysed, and subjected to cen-trifugation in sucrose density gradient. Proteins from sucrose fractions were separated in a 7.5% polyacrylamide gel and transferred to a nitrocellulose mem-brane. TLR4 detection was performed using anti-TLR4 primary Abs, followed by horseradish peroxidase–labeled secondary Ab. The bands corresponding toTLR4 were visualized using enhanced chemiluminescence reagents. One representative Western blot and densitometric analysis of 3 blots are shown.

81ALCOHOL INHIBITS TLR4 ASSOCIATION WITH LIPID RAFTS

hypothesized that alcohol, a potent solvent, may also alterthe integrity of cellular cholesterol. We successfully deplet-ed cellular cholesterol with MbCD, as shown by increasedcholesterol in culture supernatants (Fig. 7B). Alcohol alsoincreased cholesterol content in culture media, suggestingthat cholesterol depletion from the cells may account forlipid raft disturbance in the presence of alcohol. We analy-zed cell viability under the same experimental conditions(Fig. 7C) and observed that up to 100 mM alcohol is welltolerated by the cells for 1 hour, suggesting that the chang-es in eximer formation (Fig. 7A) and increased cholesterolrelease (Fig. 7B) seen in the presence of 25–100 mM alco-hol were not due to membrane damage by the alcohol.To analyze the role of lipid rafts in TLR signaling, we

stimulated the cells with TLR2 or TLR4 ligands in thepresence of increasing concentrations of cholesterol-de-pleting drugMbCD. As shown in Fig. 8, there was a loss ofLPS and PGN-induced NF-kB activation in CHO cells ex-pressing TLR4 and TLR2, respectively. We titratedMbCD and found similar trends in the degree of lipid raftdisruption and the extent of NF-kB activation with eitherTLR2 or TLR4 ligands. However, TLR4-mediated signa-ling was more readily disrupted by lower concentrations ofMbCD compared with TLR2-mediated signaling. Theseresults suggested that lipid rafts are prerequisites for TLRsignaling and indicated a different degree of association ofTLR2 and TLR4 with the rafts.

DISCUSSION

In the present study, we demonstrated for the first timethat acute alcohol influences TLR4 association with lipid

rafts. We employed purified ligands to activate TLR2 andTLR4 and evaluated the influence of alcohol on cellularactivation steps from the membrane, by investigating re-ceptor migration into lipid rafts, to intracellular signalingcascade represented by activation and nuclear transloca-tion of NF-kB, to the final event of cellular activation,such as proinflammatory cytokine production. We ob-served that all those steps were affected by alcohol; how-ever, alcohol affected TLR2 and TLR4 signalingdifferentially. First, we identified that, in nonactivatedcells, both TLR2 and TLR4 were localized to the nonraftarea of the cellular domain, as we detected them in thehigh-density sucrose gradient fractions. Second, weshowed that upon engagement of the receptor with specif-ic ligands, both TLR2 and TLR4 migrated from high-density sucrose gradient fractions into detergent-resistantlow-density sucrose gradient fractions, suggesting thatboth TLR2 and TLR4 were associated with cholesterol-rich domains of the cellular membrane, representative oflipid rafts. Third, and most important, we demonstratedthat alcohol preferentially targeted the TLR4 receptorcomplex by preventing TLR4 recruitment into the lipidrafts, while association of TLR2 with the rafts was not af-fected by alcohol. There could be several explanations forthe different effects of alcohol on TLR2 and TLR4 migra-tion into lipid rafts.Alcohol may influence TLRs at the level of protein

structure, conformation, or expression/trafficking or couldinterfere with cellular membrane organization. While it ishighly unlikely that alcohol would preferentially changeTLR4, but not TLR2, structure or tridimensional confor-mation, it is possible that alcohol would act in a more in-direct way by disrupting protein–protein associations inselect raft structures. Recent reports showed that upon cellstimulation with specific ligands, TLRs are not the solemolecules recruited to lipid rafts (Pfeiffer et al., 2001). Forreasons unknown to date, TLR4 migrates into the lipidraft with Fcg receptors, adhesion molecules, low-densitylipoprotein receptor, and signal transduction molecules(Pfeiffer et al., 2001). So far it is not known whether TLR4directly interacts with any of these molecules. Recent stud-ies have shown that upon engagement with LPS, TLR4follows a specific migration pathway from the membraneto Golgi and back to the membrane (Latz et al., 2002). Theturnover of TLR2 is less studied and it is not entirely clearif TLR2 follows the same migration pathway as TLR4. Itis also possible that alcohol affects receptor degradation/turnover. In a different scenario, it is also possible that al-cohol could induce alterations of membrane fluidity andthus facilitate TLR4, but not TLR2, docking to a nonraftlocated protein.In regards to cellular membrane organization, here we

show that alcohol changes protein content in the rafts andmodulates cellular membrane fluidity. Previous studieshave shown that ethanol affects membrane fluidity in dif-ferent cell types, including brain cells and hepatocytes

00.20.40.60.8

1

RaftNon-rafts

Media LPS LPS+EtOH

Media PGN PGN+EtOH

Media LPS LPS+EtOH

TLR2

TLR4

CD14

0.20.40.60.8

1

0

0.20.40.60.8

1

0

No

n-r

aft /

raft

rat

io

*

Fig. 6. Alcohol affects the distribution of Toll-like receptor (TLR)4, but notTLR2, into lipid rafts. We plotted the densitometric data from Figs. 3, 4, and 5 astotal density obtained by summation of the densities in all-raft or all-nonraftfractions. The ratio of total raft to nonraft fractions is shown. �po0.05.

82 DOLGANIUC ET AL.

(Colles et al., 1995; Da Silveira et al., 2003; Iqbal and Sze,1994; Sergent et al., 2005). Our experiments indicate thatalcohol interferes with fluorescent lipid eximer formation,suggesting that alcohol may induce changes in partitioningcellular membrane in raft and nonraft structures. We iden-tified that alcohol impaired only TLR4, but not flotillin orTLR2, migration into rafts. These data suggest that alco-hol causes dissolution of a distinct class of raft structures

that included TLR4 and some CD14, but not flotillin orTLR2. We also found some degree of cholesterol depletionby alcohol in the cells. These effects were not because ofalcohol-induced cell death or apoptosis. Assuming that theraft structure may have a relatively constant content ofcholesterol, we speculate that alcohol-induced cholesterolrelease into the culture supernatants suggests heterogene-ity in raft structures that may account for the different as-

A B

C D

NFκB

med

iaLP

S

MβC

Dco

mp

+LPS

020406080

Den

sity

uni

ts 100

*****

med

iaP

GN

MβC

Dco

mp

+PGN

020406080

Den

sity

uni

ts 100

***

CHO/CD14/TLR2CHO/CD14

MβCD MβCD

Fig. 8. Lipid raft disruption leads to impairment in Toll-like receptor (TLR) signaling. Chinese hamster ovary (CHO)/CD14 (A, C) or CHO/CD14/TLR2 cells (B,D) were stimulated for 1 hour, as indicated [lipopolysaccharide (LPS) 0.1 mg/ml, peptidoglycan (PGN) 1 mg/ml, methyl-b-cyclodextrine (MbCD) 5 mM alone orMbCD 20-15-10-5-2.5-1-0.5 mM for dose-dependent curve]. Five micrograms of nuclear protein/stimulation group was analyzed to determine the nuclear factor-kB (NF-kB) binding capacity by electrophoretic mobility shift assay. One representative gel (A, B) and densitometric analysis of 3 gels (C, D) are shown.�po0.05.

0

40

80

120

160(%

cha

nge

of40

5nm

/450

nm r

atio

)

p<0.059

fluorescent lipid

non-fluorescent lipid

control EtOH25mM

EtOH50mM

EtOH100mM

EtOH500mM

MβCD5mM

p<0.025

p<0.047p<0.039 p<0.034

A

0

0.5

1.0

1.5

2.0

2.5

3.0

media EtOH 50mM

EtOH200mM

MβCD5mM

Cho

lest

erol

(nM

) p<0.046 p<0.042

p<0.027B

Media MβCD 5mMEtOH 25mM EtOH 50mM EtOH 100mM

100 101 102 103 104100

101

102

103

104

100 101 102 103 104100

101

102

103

104

100 101 102 103 104100

101

102

103

104

100 101 102 103 104100

101

102

103

104

100 101 102 103 104100

101

102

103

104

100 101 102 103 104100

101

102

103

1042.03 2.26

0.5495.2

2.14 3.69

0.6793.595.4

1.69 2.06

0.8294.1 0.52

1.96 3.41 0.83 3.2

0.7995.2

22.58 25.4

3.1248.9

EtOH 500mMC

Annexin V-FITC

Pro

pidi

um io

dide

Fig. 7. Alcohol increases membrane fluidity and displaces cellular cholesterol. (A) Chinese hamster ovary (CHO) cells were labeled with fluorescent (ex-imer-forming, pyrene-decanoic acid 2 mM) lipid or nonfluorescent lipid (as negative control, pyrene 2 mM), washed, incubated with increasing concentrations ofethanol (25–500 mM, EtOH) or 5 mM methyl-b-cyclodextrine (MbCD) for 1 hour, and immediately analyzed using a reader with exitation at 350 nm. Fluores-cence emission was collected at 405 nm for monomers and 450 nm for eximers. The membrane fluidity was determined as ratio 405/450 in every experimentalgroup. Shown is the mean � SE value of percentage change in membrane fluidity from each experimental group compared with the control (medium) group,determined in 6 separated experiments. (B) CHO cells were incubated with increasing concentrations of ethanol (50 and 200 mM, EtOH) or5 mM methyl-b-cyclodextrine (MbCD) for 1 hour. Cell-free culture supernatants were analyzed for cholesterol content using Amplex Red Cholesterol Assaykit. The mean � SE value of cholesterol determined in 6 separated experiments is shown. (C) CHO cells were treated as above–labeled with annexin V–fluorescein isothiocyanate (FITC) and propidium iodide, and cellular fluorescence was analyzed by flow cytometry. The upper left quadrant shows dead cells,the lower left quadrant shows viable cells, and the right lower and upper quadrants show early and late apoptotic cells, respectively. One representative ex-periment of 2 with similar results is shown.

83ALCOHOL INHIBITS TLR4 ASSOCIATION WITH LIPID RAFTS

sociation of TLR2 or TLR4 with the membranes. Whilethe differential effect of alcohol on TLR2 and TLR4 can-not not be fully explained by alcohol’s ability to increasemembrane fluidity, we could not exclude the possibilitythat TLR2 and TLR4 migrate into different types of raftsupon cell stimulation. All lipid-rich domains of the cellularmembrane are generically called rafts; however, they differin lipid composition, protein composition, size, and evenfunction (Chateray-Rivauday et al., 2004; Melkonian etal., 1999; Pike, 2003). To date, neither a unified classifica-tion of lipid rafts nor specific protein markers of differenttypes of lipid rafts have been discovered. We identified thatthe disruption of TLR2 and TLR4 association with lipidrafts is instrumental in TLR signaling. The gradual dis-ruption of rafts withMbCD leads to the progressive loss ofsignal transduction, shown by a decrease in NF-kB bind-ing capacity. We observed that alcohol significantly pre-vented TLR4 migration into lipid rafts (�72%) (Fig. 6),which translated into loss of NF-kB binding activity(�66%) (Fig. 2).Finally, we cannot exclude the possibility that alcohol

can modify the dimensional structure of LPS but not thatof PGN. However, previously published data by us, andothers, showed that alcohol-pretreated cells respond toLPS in a manner similar to cells concomitantly stimulatedwith LPS1alcohol, suggesting that alcohol acts on the celldirectly rather than on LPS (Mandrekar et al., 1997; Rivi-er, 1997).Our discovery that alcohol interferes with TLR4 associ-

ation with lipid rafts brings significant new information onthe molecular mechanisms by which alcohol influencesearly events of cell activation. Plasma membrane domainsrange from the ones involved in receptor cluster formationto those that form the immunological synapse (Buracket al., 2002). Considering that cell activation is modulatedby the formation and distribution of receptor clusters, theunderstanding of alcohol’s influence on this process willhelp explain multiple abnormalities of immune responsesseen in individuals after alcohol intake.

ACKNOWLEDGMENTS

We thank Dr. Golenbock (University of MassachusettsMedical School) for CHO cell lines and appreciate blooddonations from our donors.

REFERENCES

Akashi S, Saitoh S, Wakabayashi Y, Kikuchi T, Takamura N, Nagai Y,

Kusumoto Y, Fukase K, Kusumoto S, Adachi Y, Kosugi A, Miyake

K (2003) Lipopolysaccharide interaction with cell surface Toll-like re-

ceptor 4-MD-2: higher affinity than that with MD-2 or CD14. J ExpMed 198:1035–1042.

Boe DM, Nelson S, Zhang P, Quinton L, Bagby GJ (2003) Alcohol-

induced suppression of lung chemokine production and the host def-ense response to Streptococcus pneumoniae. Alcohol Clin Exp Res 27:

1838–1845.

Brown DA, Rose JK (1992) Sorting of GPI-anchored proteins to glycol-

ipid-enriched membrane subdomains during transport to the apicalcell surface. Cell 68:533–544.

BurackWR, Lee KH, Holdorf AD, Dustin ML, Shaw AS (2002) Cutting

edge: quantitative imaging of raft accumulation in the immunologicalsynapse. J Immunol 169:2837–2841.

Chatenay-Rivauday C, Cakar ZP, Jeno P, Kuzmenko ES, Fiedler K

(2004) Caveolae: biochemical analysis. Mol Biol Rep 31:67–84.

Chiappelli F, Kung M, Lee P, Pham L, Manfrini E, Villanueva P (1995)Alcohol modulation of human normal T-cell activation, maturation,

and migration. Alcohol Clin Exp Res 19:539–544.

ChoudhryMA, Ren X, Romero A, Kovacs EJ, Gamelli RL, SayeedMM

(2004) Combined alcohol and burn injury differentially regulate p38and ERK activation in mesenteric lymph node T cell. J Surg Res

121:62–68.

Colles S, Wood WG, Myers-Payne SC, Igbavboa U, Avdulov NA, Jose-ph J, Schroeder F (1995) Structure and polarity of mouse brain synap-

tic plasma membrane: effects of ethanol in vitro and in vivo.

Biochemistry 34:5945–5959.

Cook RT, Zhu X, Coleman RA, Ballas ZK, Waldschmidt TJ, Ray NB,LaBrecque DR, Cook BL (2004) T-cell activation after chronic

ethanol ingestion in mice. Alcohol 33:175–181.

Da Silveira MG, Golovina EA, Hoekstra FA, Rombouts FM, Abee T

(2003) Membrane fluidity adjustments in ethanol-stressed Oenococcusoeni cells. Appl Environ Microbiol 69:5826–5832.

Dolganiuc A, Oak S, Kodys K, Golenbock DT, Finberg RW, Kurt-

Jones E, Szabo G (2004) Hepatitis C core and nonstructural 3 proteins

trigger toll-like receptor 2-mediated pathways and inflammatory acti-vation. Gastroenterology 127:1513–1524.

Faunce DE, Gregory MS, Kovacs EJ (1998) Acute ethanol exposure pri-

or to thermal injury results in decreased T-cell responses mediated inpart by increased production of IL-6. Shock 10:135–140.

Friedman H, Newton C, Klein TW (2003) Microbial infections,

immunomodulation, and drugs of abuse. Clin Microbiol Rev 16:

209–219.Goebel J, Forrest K, Flynn D, Rao R, Roszman TL (2002) Lipid rafts,

major histocompatibility complex molecules, and immune regulation.

Hum Immunol 63:813–820.

Goral J, Kovacs EJ. (2005) In vivo ethanol exposure down-regulatesTLR2-, TLR4-, and TLR9-mediated macrophage inflammatory

response by limiting p38 and ERK1/2 activation. J Immunol 174:

456–463.Gunzerath L, Faden V, Zakhari S, Warren K (2004) National Institute

on Alcohol Abuse and Alcoholism report on moderate drinking. Al-

cohol Clin Exp Res 28:829–847.

Hao M, Mukherjee S, Maxfield FR (2001) Cholesterol depletion induceslarge scale domain segregation in living cell membranes. Proc Natl

Acad Sci USA 98:13072–13077.

Haorah J, Heilman D, Diekmann C, Osna N, Donohue TM Jr., Ghor-

pade A, Persidsky Y (2004) Alcohol and HIV decrease proteasome andimmunoproteasome function in macrophages: implications for im-

paired immune function during disease. Cell Immunol 229:139–148.

Harder T (2004) Lipid raft domains and protein networks in T-cell re-

ceptor signal transduction. Curr Opin Immunol 16:353–359.Iqbal Z, Sze PY (1994) Correlation between [125I] calmodulin binding

and lipid fluidity in synaptic plasma membranes: effects of ethanol and

other short-chain alcohols. Brain Res Mol Brain Res 27:333–336.Kirschning CJ, Schumann RR (2002) TLR2: Cellular sensor for micro-

bial and endogenous molecular patterns. Curr Top Microbiol Immu-

nol 270:121–144.

Kopp E, Medzhitov R (2003) Recognition of microbial infection by Toll-like receptors. Curr Opin Immunol 15:396–401.

Latz E, Visintin A, Lien E, Fitzgerald KA, Monks BG, Kurt-Jones EA,

Golenbock DT, Espevik T (2002) Lipopolysaccharide rapidly traffics

to and from the Golgi apparatus with the toll-like receptor 4-MD-2-CD14 complex in a process that is distinct from the initiation of signal

transduction. J Biol Chem 277:47834–47843.

84 DOLGANIUC ET AL.

Legler DF, Micheau O, Doucey MA, Tschopp J, Bron C (2003) Recruit-

ment of TNF receptor 1 to lipid rafts is essential for TNFalpha-medi-ated NF-kappaB activation. Immunity 18:655–664.

Li N, Mak A, Richards DP, Naber C, Keller BO, Li L, Shaw AR (2003)

Monocyte lipid rafts contain proteins implicated in vesicular traffick-ing and phagosome formation. Proteomics 3:536–548.

Lois M, Brown LA, Moss IM, Roman J, Guidot DM (1999) Ethanol

ingestion increases activation of matrix metalloproteinases in rat

lungs during acute endotoxemia. Am J Respir Crit Care Med 160:1354–1360.

Mandrekar P, Catalano D, Szabo G (1997) Alcohol-induced regulation

of nuclear regulatory factor-kB in human monocytes. Alcohol Clin

Exp Res 21:988–994.Mandrekar P, Dolganiuc A, BelleroseG,KodysK, Romics L, Nizamani R,

Szabo G (2002) Acute alcohol inhibits the induction of nuclear regu-

latory factor kappa B activation through CD14/toll-like receptor 4,interleukin-1, and tumor necrosis factor receptors: a common mecha-

nism independent of inhibitory kappa B alpha degradation? Alcohol

Clin Exp Res 26:1609–1614.

Mandrekar P, Catalano D, Dolganiuc A, Kodys K, Szabo G (2004) In-hibition of myeloid dendritic cell accessory cell function and induction

of T cell anergy by alcohol correlates with decreased IL-12 production.

J Immunol 173:3398–3407.

Mayor S, RaoM (2004) Rafts: Scale-dependent, active lipid organizationat the cell surface. Traffic 5:231–240.

Melkonian KA, Ostermeyer AG, Chen JZ, Roth MG, Brown DA (1999)

Role of lipid modifications in targeting proteins to detergent-resistant

membrane rafts. Many raft proteins are acylated, while few are preny-lated. J Biol Chem 274:3910–3917.

Meyer zum Bueschenfelde CO, Unternaehrer J, Mellman I, Bottomly K

(2004) Regulated recruitment of MHC class II and costimulatory mol-ecules to lipid rafts in dendritic cells. J Immunol 173:6119–6124.

Nelson S, Bagby GJ, Bainton BG, Summer WR (1989) The effects of

acute and chronic alcoholism on tumor necrosis factor and the inflam-

matory response. J Infect Dis 160:422–429.Nelson S, Kolls JK (2002) Alcohol, host defence and society. Nat Rev

Immunol 2:205–209.

Neumann-Giesen C, Falkenbach B, Beicht P, Claasen S, Luers G,

Stuermer CA, Herzog V, Tikkanen R (2004) Membrane and raft asso-ciation of reggie-1/flotillin-2: role of myristoylation, palmitoylation

and oligomerization and induction of filopodia by overexpression.

Biochem J 378:509–518.Pfeiffer A, Bottcher A, Orso E, Kapinsky M, Nagy P, Bodnar A,

Spreitzer I, Liebisch G, Drobnik W, Gempel K, Horn M, Holmer S,

Hartung T, Multhoff G, Schutz G, Schindler H, Ulmer AJ, Heine H,

Stelter F, Schutt C, Rothe G, Szollosi J, Damjanovich S, Schmitz G

(2001) Lipopolysaccharide and ceramide docking to CD14 provokes

ligand-specific receptor clustering in rafts. Eur J Immunol 31:3153–3164.

Pierce SK (2002) Lipid rafts and B-cell activation. Nat Rev Immunol

2:96–105.Pike LJ (2003) Lipid rafts: bringing order to chaos. J Lipid Res 44:

655–667.

Pruett SB, Zheng Q, Fan R, Matthews K, Schwab C (2004a) Ethanol

suppresses cytokine responses induced through Toll-like receptors aswell as innate resistance to Escherichia coli in a mouse model for binge

drinking. Alcohol 33:147–155.

Pruett SB, Zheng Q, Fan R, Matthews K, Schwab C (2004b) Acute ex-

posure to ethanol affects Toll-like receptor signaling and subsequentresponses: an overview of recent studies. Alcohol 33:235–239.

Rivier C (1997) Effect of pretreatment with alcohol on subsequent endo-

crine and immune responses in the adult male rat. Alcohol Clin ExpRes 21:1690–1694.

Russo D, Purohit V, Foudin L, Salin M (2004) Workshop on Alcohol

Use and Health Disparities 2002: a call to arms. Alcohol 32:37–43.

Schmitz G, Orso E (2002) CD14 signalling in lipid rafts: new ligands andco-receptors. Curr Opin Lipidol 13:513–521.

Schroder NW, Morath S, Alexander C, Hamann L, Hartung T, Zah-

ringer U, Gobel UB, Weber JR, Schumann RR (2003) Lipoteichoic

acid (LTA) of Streptococcus pneumoniae and Staphylococcus aureusactivates immune cells via Toll-like receptor (TLR)-2, lipopolysaccha-

ride-binding protein (LBP), and CD14, whereas TLR-4 and MD-2 are

not involved. J Biol Chem 278:15587–15594.

Sergent O, Pereira M, Belhomme C, Chevanne M, Huc L, Lagadic-Gossmann D (2005) Role for membrane fluidity in ethanol-induced

oxidative stress of primary rat hepatocytes. J Pharmacol Exp Ther Jan

5; [Epub ahead of print].Simons K, Vaz WL (2004) Model systems, lipid rafts, and cell mem-

branes. Annu Rev Biophys Biomol Struct 33:269–295.

Singer SJ, Nicolson GL (1972) The fluid mosaic model of the structure of

cell membranes. Science 175:720–731.Szabo G, Dolganiuc A, Mandrekar P, White B (2004) Inhibition of an-

tigen-presenting cell functions by alcohol: implications for hepatitis C

virus infection. Alcohol 33:241–249.

Takeda K, Kaisho T, Akira S (2003) Toll-like receptors. Annu RevImmunol 21:335–376.

Triantafilou M, Miyake K, Golenbock DT, Triantafilou K (2002) Medi-

ators of innate immune recognition of bacteria concentrate in lipidrafts and facilitate lipopolysaccharide-induced cell activation. J Cell

Sci 115:2603–2611.

Zhang P, Bagby GJ, Happel KI, Summer WR, Nelson S (2002) Pulmo-

nary host defenses and alcohol. Front Biosci 7:d1314–d1330.

85ALCOHOL INHIBITS TLR4 ASSOCIATION WITH LIPID RAFTS