Influence of oxidized lipids on palmitoyl-oleoyl-phosphatidylcholine organization, contribution of Langmuir monolayers and Langmuir–Blodgett films

(1)

HAL Id: hal-01391772

https://hal.archives-ouvertes.fr/hal-01391772

Submitted on 3 Nov 2016

HAL is a multi-disciplinary open access

archive for the deposit and dissemination of

sci-entific research documents, whether they are

pub-lished or not. The documents may come from

teaching and research institutions in France or

abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est

destinée au dépôt et à la diffusion de documents

scientifiques de niveau recherche, publiés ou non,

émanant des établissements d’enseignement et de

recherche français ou étrangers, des laboratoires

publics ou privés.

Distributed under a Creative Commons Attribution - ShareAlike| 4.0 International

License

Influence of oxidized lipids on

palmitoyl-oleoyl-phosphatidylcholine organization,

contribution of Langmuir monolayers and

Langmuir–Blodgett films

Christine Grauby-Heywang, Fabien Moroté, Marion Mathelié-Guinlet,

Ibtissem Gammoudi, Ndeye Rokhaya Faye, Touria Cohen-Bouhacina

To cite this version:

Christine Grauby-Heywang, Fabien Moroté, Marion Mathelié-Guinlet, Ibtissem Gammoudi, Ndeye

Rokhaya Faye, et al.. Influence of oxidized lipids on palmitoyl-oleoyl-phosphatidylcholine organization,

contribution of Langmuir monolayers and Langmuir–Blodgett films. Chemistry and Physics of Lipids,

Elsevier, 2016, 200, pp.74 - 82. �10.1016/j.chemphyslip.2016.07.001�. �hal-01391772�

(2)

In

ﬂuence

of

oxidized

lipids

on

palmitoyl-oleoyl-phosphatidylcholine

organization,

contribution

of

Langmuir

monolayers

and

Langmuir

–

Blodgett

ﬁlms

Christine

Grauby-Heywang

a,

*

,

Fabien

Moroté

a

,

Marion

Mathelié-Guinlet

a

,

Ibtissem

Gammoudi

b

,

Ndeye

Rokhaya

Faye

a,1

,

Touria

Cohen-Bouhacina

a

LaboratoireOndesetMatièred’Aquitaine(LOMA),UMRCNRS5798,UniversitédeBordeaux,351coursdelalibération,33405TalenceCedex,France

b

CelluledetransfertNanoPhyNov,UniversitédeBordeaux,351coursdelalibération,TalenceCedex33405,France

Keywords: Oxidizedlipid Langmuirmonolayer Langmuir–Blodgettﬁlm Surfacepressuremeasurements Atomicforcemicroscopy

ABSTRACT

In this work, we studied theinteraction of two oxidizedlipids, PoxnoPC and PazePC, withPOPC phospholipid.Mean molecularareasobtained from (p A)isotherms ofmixed PoxnoPC POPCand PazePC POPCmonolayersrevealeddifferentbehaviorsofthesetwooxidizedlipids:thepresenceof PoxnoPCinthemonolayersinducestheirexpansion,meanmolecularareasbeinghigherthanthose expectedinthecaseofidealmixtures.PazePC POPCbehaveonthewholeideally.Thisdifferencecanbe explainedbyadifferentconformationofoxidizedlipids.MoreoverthecarboxylicfunctionofPazePCis protonatedunderourexperimentalconditions,asshownby(p A)isothermsofPazePCatdifferentpH values.Bothoxidizedlipidsinducealsoanincreaseofthemonolayerelasticity,PoxnoPCbeingslightly moreefﬁcientthanPazePC.Thesemonolayersweretransferredfromtheair waterinterfaceontomica supportsforastudybyAFM.AFMimagesareonthewholehomogenous,suggestingthepresenceofonly onelipidphaseinbothcases.However,inthecaseofPazePC POPCmonolayers,AFMimagesshowalso thepresenceofareasthickerof7nmto10nmthanthesurroundinglipidphase,probablyduetothelocal formationofmultilayersystemsinducedbycompression.

1. Introduction

Oxidationofmembranelipidscanoccurviadifferentprocesses, suchasoxidativestressresponsiblefortheformationofreactive oxygen species (ROS) damaging biomolecules (Deigner and Hermetter,2008),photodynamictherapies(Girotti,2001;Castano

etal.,2004)orenzymaticreactions(Greigetal.,2012).Damages duetooxidationcanbedramatic,sinceitleadstotheformationof deeplymodiﬁedlipids,suchasphospholipidscontaininghydroxyl orhydroperoxygroupsontheunsaturatedchain,phospholipids containingatruncatedchainwithaterminalcarbonylorhydroxyl group,or lysophospholipids (O’Donnel, 2011).The main conse quenceofthesemodiﬁcationsisthatoxidizedlipidsaremorepolar thantheintactones.Chemically,polyunsaturatedphospholipids aremoresensitivetooxidation,andtheirpolarheadgroupisvery oftenphosphatidylcholine(PC),PCbeingthemajorphospholipid inmammalianmembranes(Catala,2012).

Effectsoftheseoxidizedlipidsaremultiple.Forinstance,they canberecognizedbyspeciﬁcreceptorswhichactivatedifferent signalingpathways leading tovariouscellularresponses (Fruh wirthetal.,2007;Greigetal.,2012).Inthiscontext,ithasbeen shownthatlipidoxidationplaysaroleininﬂammatoryprocesses observedinvariousdiseases,suchasatherosclerosis,cardiovascu lardiseasesorcancers(Greigetal.,2012;Fruhwirthetal.,2007; KoppakaandAxelsen,2000).Oxidizedlipidscouldalsostimulate theformationofamyloidplaques,responsiblefortheAlzheimer disease (Koppaka and Axelsen, 2000; Volinsky and Kinnunen,

Abbreviations:p,surfacepressure;A,meanmoleculararea;POPC, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine;OPPC, 1-oleoyl-2-palmitoyl-sn-glycero-3-phosphocholine; PoxnoPC, 1-palmitoyl-2-(90 -oxo-nonanoyl)-sn-glycero-3-phos-phocholine; PazePC, 1-palmitoyl-2-azelaoyl-sn-glycero-3-phosphocholine; PC, phosphatidylcholine;PS,phosphatidylserine;ROS,reactiveoxygenspecies;AFM, atomicforcemicroscopy;LB,Langmuir–Blodgett.

* Correspondingauthor.

E-mailaddresses:christine.grauby-heywang@u-bordeaux.fr

(C.Grauby-Heywang),fabien.morote@u-bordeaux.fr(F.Moroté),

marion.mathelie-guinlet@u-bordeaux.fr(M.Mathelié-Guinlet),

ibtissem.gammoudi@u-bordeaux.fr(I.Gammoudi),rokhaya.faye@ihu-liryc.fr

(N.R.Faye),touria.cohen-bouhacina@u-bordeaux.fr(T.Cohen-Bouhacina).

1

Presentaddress:CentredeRechercheCardio-ThoraciquedeBordeaux—INSERM U1045,IHU-LIRYC,PTIB—HôpitalXavierArnozan,AvenueduHautLévêque,33604 Pessac,France.

(3)

2013).Atlast,theyareinvolvedinapoptosis,sincetheexternali zationofoxidizedlipidsisnecessaryfortheclearanceofapoptotic cells by macrophages (Fruhwirth et al., 2007; Volinsky and Kinnunen,2013).

Concerning molecular mechanisms, lipids being essential componentsof cells, it is clearthat theiroxidation can induce strongmodiﬁcationinthecellfunctioning,oxidizedlipidsactingas second toxic messengers or inducing structural damages in membranes (Catala, 2012; Volinsky and Kinnunen, 2013). In particular, oxidizedlipids canbe locally concentrated in mem branes, thus inﬂuencing theirbiophysicalproperties (Kinnunen etal.,2012;VolinskyandKinnunen,2013).Inthiscontext,alotof studieshave been performedon lipidsfrom themitochondrial membrane,sincethemainpartofendogenousfreeradicals,acting asstrongoxidants, isproduced in mitochondriaduringaerobic respiration (Gabbita et al.,1998). It has been shown that the accumulationofoxidizedlipidsinmicrosomesinducesanincrease ofthelipidacylchainorder(Eichenbergeretal.,1982).Inthesame idea, the lipid oxidation induced by FeSO4 on mitochondria

decreasesthemembranefluidity(Nepomucenoetal., 1997).Onthe contrary,Gabbita etal. observed an increase of themembrane fluidityafterthe oxidation of synaptosomes and mitochondrial membranes,thishigherfluiditybeingassignedtotheformationof gapsinthemembranes(Gabbitaetal.,1998).Disorderinthelipid packinginthepresence ofoxidizedlipidswas alsoreportedin supportedbilayersmadewithphospholipidsfrommitochondria incubatedinROS generatingconditions,evenifalltheoxidizing conditionsarenotsimilarlyefficient(MegliandSabatini,2003).In particular,inthecaseofmitochondriaexposedtoCCl4,knownto

increasetheoxidativestressinashortterm,thebilayerdisorderis proportionaltotheoxidantamountand totheincubation time (MegliandSabatini,2004).

However,theanalysisoftheseexperimentsisdifficult,sincethe diversityofmoleculesproducedbyoxidationinaninitialcomplex lipidmixtureiswideandthenatureofoxidizedmoleculesisnot always known in details. However, perfectly defined oxidized lipidsarenowavailable,leadingtolessambiguousresults.Thisis the case of 1 palmitoyl 2 (90 _{oxo nonanoyl) sn glycero 3 phos} phocholine and 1 palmitoyl 2 azelaoyl sn glycero 3 phospho choline, PoxnoPC and PazePC,respectively, which are oxidized derivativesof1 palmitoyl 2 oleoyl phosphatidylcholineorPOPC. Atthecellularlevel,ithasbeenshownthatPoxnoPCisinvolvedin apoptosisandnecrosisprocesses(Uhlsonet al.,2002), whereas PazePCbehavesasaweakligandperoxisomereceptor(Davisetal., 2001).The behaviorof theselipids orsimilar onesincludedin membranemodelshasbeenalreadystudiedbydifferentexperi mentalmethods:electronparamagneticresonance(Meglietal., 2005),surfacepressureandsurfacedipolemeasurementscoupled withfluorescencemicroscopyattheair waterinterface(Sabatini etal.,2006), fluorescenceenergytransfer(Mattila etal.,2008), fluorescence correlation spectroscopy z scan (Beranova et al., 2010; Parkkila et al., 2015), fluorescence solvent relaxation (Volinskyetal.,2011),scatteringstoppedflowexperiments(Lis et al., 2011), differential scanning calorimetry and nuclear magnetic resonance (Wallgren et al., 2012; Wallgren et al., 2013) ... Molecular simulations have been also successfully applied, giving precious complementary information, such as moleculeorientationandconformation(KhandeliaandMouritsen,

2009; Beranova et al., 2010; Cwiklik and Jungwirth, 2010;

Khandelia et al.,2014).On thewhole, data showthat oxidized lipids,insufficientamount,caninducebilayer/micelletransitions (Meglietal.,2005),ortheformationofdefectsandtheincreaseof lipidflip flop(Volinskyetal.,2011).Ontheotherhand,oxidized lipidsareforinstanceabletostabilizesphingomyelin/cholesterol domainsinternaryPC/sphingomyelin/cholesterolmixtures(Volin skyetal.,2012;Parkkilaetal.,2015).Moreover,theoxidized,and

thusmorepolar,chainisabletoreverseinamoreorless“extended lipidconformation”(KhandeliaandMouritsen,2009),inorderto getclosertotheinterfacialregionoreventopointoutinwater.This particular orientation, described recently in a “lipid whisker model”(Catala,2012)improvingthe“ﬂuidmosaicmodel”ofSinger and Nicolson (Singer and Nicolson, 1972), induces important changes in the membrane structure and dynamics (mean molecular areas, bilayer thickness, hydration proﬁle, phase separation ...), depending onthe percentage of theoxidized lipid,theexperimentalconditions(pH,cations)andthenatureof surroundinglipids.

In a previous work, we studied the natural oxidation by atmosphericoxygenofLangmuir Blodgett(LB)filmsmadeoftwo monounsaturated lipids, POPC and OPPC (1 oleoyl 2 palmitoyl phosphatidylcholine),usingatomicforcemicroscopy(AFM)(Faye et al., 2013). AFM is a highresolution scanning probe surface analysis technique, based on the measurement of interaction forces between a fine tipand the sample surface, these forces dependingonthetip sampledistance.Byscanningthesurfaceitis possibletoimageitwithlateralandperpendicularresolutionsof 1nmand 0.1nm, respectively.To ourknowledge,this powerful imagingtechniquehasnotbeenappliedinthecaseofmembrane systemsincludingoxidizedlipids,whereasitisperfectlyadapted tothenon damagingstudyoflipidplanarmonolayersandbilayers (Garcia Manyèsetal.,2007;Garcia ManyèsandSanz,2010;Faye etal.,2013).Moreover,AFMpresentsanimportantadvantageas comparedtowidespreadfluorescencemicroscopy,sincethishigh resolution and non intrusive technique does not require the presenceofafluorophore,whichcaninduceitselflipidoxidation (AyuyanandCohen,2006).

AFMimagesofPOPCandOPPCLBfilmsshowedthatbothfilms aresensitivetooxidation,evenifthis processisratherlowand occurs after a few days.This process,which is notobserved if samplesarekeptundervacuum,isresponsiblefortheappearance of domainsregularlydistributedonthefilmsurfaceand higher than the surrounding intact phase, suggesting that oxidation occurslocally,likelyinareaspresentingalocaldefect.Weassumed thatthesedomainsresultfromthereversaloftheoxidizedchain, as previously described, leading to the raising of the whole molecule.

Unfortunaly,itwasnotpossibleinthisfirststudytodefinethe composition of LB films in terms of oxidized lipids, since this oxidationwasnotchemicallycontrolled.Therefore,inthepresent work,wecontinue toexplorethebehaviorofoxidizedlipidsin membrane models, by introducing PoxnoPC and PazePC, two knownderivativesofPOPC,inPOPCmonolayers.Theselipidswere addedtoPOPCataknownamount(inthe4 20mol%range).Ina firststep,surfacepressuremeasurementsenabledustostudythe lateral packingoftheselipid mixtures,whereas compressibility modulusgaveinformationonthemonolayerelasticity.Asinthe caseofAFM,suchmethodshavebeenrelativelyrarelyappliedto thestudyofoxidizedlipids(Sabatinietal.,2006;Volinskyetal., 2012).Ina secondstep,mixedmonolayersweretransferredon planarsupportsbytheLBtechniquetobecharacterizedbyAFM. OurresultsshowthatPoxnoPCandPazePCbehavedifferentlyin POPC monolayers in terms of mean molecular areas, PoxnoPC inducinganexpansionofmixedmonolayers,contrarytoPazePC. Bothoxidizedlipidsinduceanincreaseofthemonolayerelasticity. AFM images do not show any phase separation in mixed monolayers,butrevealthepresenceofthickerareasinLBfilms containingPazePC,likelyduetothelocalformationofmultilayer systemsinducedbycompression.Atlast, theseimages arevery differentfromthosepreviouslyobtainedinthecaseofPOPCLB films submitted to atmospheric oxygen, which confirms that oxidationoccurslocallyandpropagatesbychainreactioninpure POPCLBfilms.

(4)

2. Materialandmethods 2.1.Material

PoxnoPC (1 palmitoyl 2 (90 oxo nonanoyl) sn glycero 3 phosphocholine)andPazePC(1 palmitoyl 2 azelaoyl sn glycero 3 phosphocholine)werepurchasedfromAvantiPolarLipids.POPC (1 palmitoyl 2 oleoyl sn glycero 3 phosphocholine) was pur chasedfromSigma Aldrich.Fig.1showsthechemicalstructure ofthesethreelipids.Theywereatleast99%pureandusedwithout furtherpuriﬁcation. Chloroformand ethanol (bothHPLC grade) werepurchasedfromSigma Aldrich.Milliporewaterwasusedas subphase(pH5.6,resistivityhigherthan18.2M

V

cm).Inthecase of purePazePCmonolayers, NaOHand HCl (bothfrom Sigma Aldrich) were occasionally added to change the subphase pH. MuscovitemicasubstratesforAFMexperimentswerepurchased fromElectronMicroscopySciences(USA).

2.2.SurfacepressuremeasurementsandLangmuir Blodgett(LB) transfers

Surfacepressureexperimentswerecarriedoutinair,usinga Nima Langmuir through (spreading surface between 40 and 234cm2_),_equipped_with_a_Wilhelmy_balance_(Nima)._Brie_ﬂy,_lipids

weredissolvedina1/1(v/v)mixtureofchloroformandethanolata concentration of 1mM. In the case of mixed monolayers, appropriate volumes of each solution were mixed just before spreading,anewmixedsolutionbeingusedfateachspreading. Afterspreadingandevaporationofsolvents(15min),lipidswere compressed continuouslyat a rate of 5cm2_min 1_. _During _the

compression, surface pressure

p

was measured by using a WhatmanchromatographypaperasWilhelmyplate.Thetemper atureoftheroomwaskeptconstantto201C.

p

Aisothermswereanalyzedfollowingtwocomplementary methods.First,averagesofexperimentalmeanmolecularareasata givensurfacepressure(obtainedfromseveralisothermsrealized independently)werecomparedtotheoreticalones,obtainedatthe samepressurefromtheadditionofmeanmolecularareasofsingle componentstakingintoaccounttheirmolefraction(Gaines,1966). Secondly the reciprocal isothermal compressibility CS 1 (com

pressibilitymodulus)was determinedfrom thecompressibility CS ð 1=ApÞðdA=d

p

ÞwhereApisthemeanmolecularareaata

givensurfacepressure

p

.AhighCS 1valueisindicativeofalow

interfacialelasticity.

Monolayersweretransferredfromtheair waterinterfaceonto freshly cleaved micaby the LB method,by using a dip coated mechanism from Nima. Mica was ﬁrst immersed in water, perpendicularly to the interface,and the lipid monolayer was spread as previously described and compressed until a surface pressure of 30 mN.m 1_. _This _surface _pressure _was _next _kept

constantthankstoacontrolsystemmaintainingthepressureby adjustingthesurfaceoccupiedbythemonolayer.Afterstabiliza tionof thelipid monolayer (afew minutesat most),micawas removedfromwateratarateof5mmmin 1_,_and_kept_protected

fromdustuntilitsimagingbyAFMinthefollowinghours. 2.3.Atomicforcemicroscopy

AFM images were recorded with a BioscopeII AFM setup (Veeco Brucker,SantaCA)equippedwithaGscanner(maximum XYZ scan range of 150

m

m150

m

m12

m

m). Samples were scannedin tappingmodeusingPPP NCLsiliconprobes(NANO SENSORSTM₎ _with _a _spring _constant _of _about ₃₂_N_m 1 _and _a

correspondingmeasuredresonancefrequencyof about165kHz. Allscansweredoneatairandroomtemperature(201C)with scanratesbetween0.3Hzand1Hz(accordingtothescansizeand thescanningmode).Sampleswerescannedinthehoursfollowing themonolayerdepositioninordertoavoidanydamagesdueto oxidation (Faye et al., 2013). Three similar LB ﬁlms were systematically studied to ensure reproducibility, and different areasof each samplewereobserved.AFM datawereprocessed using the Nanoscope (version 7.30, Veeco) and the Gwiddion softwares.

3. Results

Fig.2Ashowsthe

p

AisothermsofPoxnoPC,POPCandtheir mixturescontainingfrom4to20mol%ofPoxnoPC.Isothermsof PoxnoPC and POPC are in agreement with those previously described (Sabatini etal.,2006;Laietal.,1994;Grechishnikova etal.,1999),andsuggestthattheselipidsareinliquidexpanded phase,astheirmixtures.ThecollapsesurfacepressureforPoxnoPC isratherlow,around31mNm 1,butthepresenceofPOPCintothe monolayersincreasestheirstabilityathighpressure.

Experimentalmeanmolecularareas(Aexp)ofmixedmonolayers

have been compared to ideal ones deﬁned as

A_idealmixture X1A1þð1 X1ÞA2,where A1 and A2are the mean

molecularareasoflipids1and2atagivensurfacepressure,andX1

themolefractionofthelipid1inthemonolayer(Gaines,1966). Any deviation

D

A Aexp Aidealmixture6 0from this expression indicatesthat the mixtureis not ideal becauseof attractive or repulsive forces between the two molecules, inducing the condensation(

D

A<0)ortheexpansion (

D

A>0)ofthemixed monolayers,respectively.Italsoprovidesevidenceformolecular miscibility(Gaines,1966).

However, this analysismakes senseonly if uncertainties in meanmolecularareasdeducedfrom

p

Aisothermsareweak.In the case of pure POPC monolayers,

p

A isotherms were reproducible with typical variations of mean molecular area around1.5Å2_in_the_surface_pressure_range_of_{5 30}_mN_m 1_._In_the

samesurfacepressurerange,variationsofmeanmolecularareas werearound2.5 3.5Å2_in_the_case_of_PazePC,_and_{1.5 2.0}_Å2_in_the

case of PoxnoPC, uncertainties for correspondingmixedmono layersbeinginthesamerange.

TheanalysisofmeanmolecularareasofmixedPoxnoPC POPC monolayers at different surface pressures shows that they are higherthanthoseexpectedinidealmixtures,

D

Avaluesranging

(5)

from4.0to6.5Å2_according_to_the_surface_pressure_and_the_amount

ofPoxnoPC.Inallcases,theamplitudeoftheexpansiondecreases whenthesurfacepressureincreases.TheexpansionofPoxnoPC POPC monolayers suggests that molecules are at least partly miscible.

At last, Fig. 2B shows the variation of the compressibility modulus CS 1 versus the percentage of PoxnoPC into the

monolayer. Pure PoxnoPC monolayer is characterized by CS 1

valuesroughlybetween30and60mNm 1inthesurfacepressure rangeof5 30mNm 1_._These_values_are_in_agreement_with_those

previously reported by Sabatini and coworkers for this lipid (Sabatinietal.,2006).Compressibilitymoduluslogicallyincreases when surface pressure increases, showing the decrease of the monolayer elasticity. Taking into account the experimental uncertainty on this parameter, it gets to a constant value for surfacepressureabove20mNm 1_._In_the_case_of_POPC,_values_of

compressibility modulus are consistent with those previously

reported(Lietal.,2001;Mattietal.,2001).Atlast,inthecaseof mixedPoxnoPC POPCmonolayers,thevariationofcompressibility modulus with surface pressure is similar, an increase of the PoxnoPCamountintothemonolayerincreasingitselasticity.On thewhole,asimilarvariationwasobservedinthecaseofPoxnoPC DPPCmonolayersstudiedbySabatiniandcoworkers,withhigher CS 1valuesduetothehigherrigidityofDPPCascomparedtoPOPC

one(Sabatinietal.,2006).

Monolayerswerethentransferredonmicaat30mNm 1_to_be

observedbyAFM.Thissurfacepressureisclosefromthesurface pressureof PoxnoPCcollapse, butthis valuecorresponds tothe estimated membrane surface pressure (Janmey and Kinnunen, 2006).Moreover,asufficientlyhighsurfacepressureisrequiredto keeplipidcohesion(moreparticularlyinthecaseoffluidPOPC), and the presence of POPC in mixed monolayers increases the collapsesurfacepressure,asshowninFig.2A.Fig.3Ashowsthe AFM image of a pure PoxnoPC LB film. The film appears homogenous, with an average surface roughness of 0.1nm, in agreementwiththefactthatthislipidisinanexpandedphase.We observedpreviouslythesamefeaturewithPOPCLBfilms(Faye etal.,2013).Fig.3BshowsanimageofaPoxnoPC POPCLBfilm containing 20mol% of PoxnoPC. This uniform image (average surface roughness around 0.2nm) suggests that PoxnoPC is homogenouslydistributedinPOPC(similarimageswereobtained atothermol%ofPoxnoPCwithacomparablesurfaceroughness,or atlowerscale,datanotshown).

InthecaseofPazePC,animportantpointisﬁrsttodetermineif thecarboxylicfunctionoftheoxidizedchainisprotonatedornot under our experimental conditions. Fig. 4A shows the (

p

A) isothermofPazePCspreadonultrapurewater.Thisisothermisin agreementwiththosepreviouslyreported(Sabatinietal.,2006; Volinskyetal.,2012)andshowsthatthislipidisinexpandedphase asPoxnoPC,withmeanmolecularareasshiftedtohighervalues (around 10Å2_)._It _has_been _reported_that _the _pKa_value _of _the

carboxylicfunctionoftheoxidizedchainisunusuallyhighinthe 7 8range(Nagleetal.,2013).Thisunusualrangeissupportedbya studyofthecytochromc PazePCinteractionusingquenchingof pyreneemission(Mattilaetal.,2008):theelectrostaticinteraction decreaseswhenpHdecreasesfrom7.4to5.0,becauseofPazePC protonation at low pH. These results suggest that PazePC carboxylic function is protonated under our experimental con ditions.However,toconﬁrmthishypothesis,(

p

A)isothermsof PazePCwerealsoperformedonwatersubphaseatpH3.0and10.0 (Fig. 4B). Isotherm obtainedat acidic pHis similar tothe one obtainedwithultrapurewater,whereasisothermobtainedatbasic pHis clearlydifferent:in thiscase, compressioninducesalsoa regular increase of surface pressure but witha different slope. Compressibility modulus was determined at various surface pressures for all pH values: in the case of PazePC monolayers spread on pure water, they are in the same range than the compressibility of PoxnoPC monolayers, but slightly higher, increasingfrom30mNm 1_to₇₅_mN_m 1_when_surface_pressure

increasesfrom5mNm 1_to₃₀_mN_m 1_._They_are_on_the_whole_in

thesamerangeifthepHofthesubphaseisacidic,butclearlylower at basic pH (around 25 30mNm 1 _at _a _surface _pressure _of

10mNm 1_for_instance)._Moreover,_at_surface_pressure_higher_than

15mNm 1, theisotherm bends in a potential phase transition. Surface ofthethroughbeingtoosmalltofurthercompressthe monolayer, the same experiment was repeated with a higher numberofspreadmolecules.Evenunderthesenewconditionsit was not possible to compress the monolayer above a surface pressure of 20 22mNm 1_, _suggesting _that _the _monolayer _is

unstable.Finally,wecanconcludefromtheseresultsthatPazePC carboxylic function is mainly protonated when it is spread on ultrapure water, since (

p

A) isotherm is similar to the one

0 5 10 15 20 25 30 35 40 40 60 80 100 120 140 160 180 S u rf ac e p re ss u re (m N .m -1 )

Mean molecular area (Å2) 100 mol% (pure PoxnoPC) 0 mol% (pure POPC) 20 mol% 8 mol%

A

20 30 40 50 60 70 80 90 100 C s -1 (m N .m -1 ) Percentage of PoxnoPC 5 mN.m1 15 mN.m1 25 mN.m1 30 mN.m1 0 20 40 60 80 100

B

Fig.2.(A)p–AisothermsofPoxnoPC,POPCandtheirmixtures(subphaseultrapure water pH 5.6, T=201_C, _compression _speed ₅_cm2

min 1

). Percentages of PoxnoPCareindicated.(B)CompressibilitymodulusCS 1versusthepercentage

of PoxnoPC into mixed monolayers (average values obtained from several independentisotherms).Straightlinesareaddedtoguidetheeye.

(6)

obtainedatpH3.0.TheinstabilityofthemonolayeratpH10.0, likelyduetoelectrostaticrepulsion,conﬁrmsthishypothesis.

Mixedmonolayers(POPC+4to20mol%ofPazePC)arealsoin an expanded phase, but contrary to the case of PoxnoPC, the presence of POPC doesnot changesigniﬁcantlythestability of monolayers at high surface pressure, collapse occurring in the same range of surface pressure values. The analysis of mean molecularareasshowsamodestexpansionofthemonolayersin thepresenceofPazePC(datanotshown),but

D

Avaluesareinthe rangeofexperimentalerrors,leadingtotheﬁnalconclusionthat thesemixtureswould ratherbehaveasidealones.Asindicated previously,compressibilitymodulusofpurePazePCmonolayersis slightlyhigherthanthoseofPoxnoPCmonolayers.Thisshowsthat PazePCmonolayersareslightlylesselastic.InthecaseofPazePC POPC mixed monolayers, an increase of PazePC percentage increasesthemonolayerelasticitybutinaslightlyweakermanner thanPoxnoPC(datanotshown).

PazePC forms homogenous films with an average surface roughness of 0.1nm (data not shown), in agreement with its expandedphase.InthecaseofPazePC POPCLBfilms,heightand phaseAFMimages(Fig.5AandB)areonthewholealsouniform (averagesurfaceroughnessaround0.1nm),butit ispossibleto observesmallareascharacterized bya higher thicknessusually from7nmto10nmabovethehomogenoussurroundingphaseas shownbyheightprofilesextractedfromheightimage(Fig.5C).The contrastobserved inphase image (Fig. 5B) suggestsa different molecular organization between the homogenous surrounding phaseandprotrusions,thephasedifference betweenbothareas beinginarangeof50 55,commonlymetinthecaseoflipids.At last,typicaldimensionsofthesestructuresareinthe70 100nm range,thebiggestoneshavingdimensionaround500nm.Their densityintoLBfilmsincreaseswiththepercentageofPazePCinthe monolayer, and they are not observed if the monolayer is transferredat20mNm 1_instead_of₃₀_mN_m 1_(data_not_shown).

4. Discussion

SurfacepressuremeasurementsshowthatPoxnoPCandPazePC arebothinaliquidexpandedphase,PazePC

p

Aisothermbeing shifted to higher mean molecular areas as compared to the PoxnoPCone.AFMimagesofcorrespondingLBﬁlmsconﬁrmsthe presenceofonlyonephaseforbothlipidsatthelateralresolution oftheAFMsetup(around1nm).

Ithasbeenproposedthankstomolecularsimulationsthatpolar oxidizedchainsareabletoorientatemoreorlessparalleltothe

Fig.3. (A)HeightAFMimages(3030mm2₎_of_a_pure_PoxnoPC_LB_film._(B)_Height_AFM_image₍₃₀₃₀_m_m2₎_of_a_mixed_PoxnoPC₍₂₀_mol%)–POPC_LB_film._Both_films_are

transferredat30mNm 1

.

A

B

Fig.4.(A)p–AisothermsofPazePC,POPCandtheirmixtures(subphaseultrapure waterpH5.6,T=201_C,_compression_speed₅_cm2

min 1

).PercentagesofPazePC areindicated.(B)p–AisothermsofPazePConultrapurewateratpH3.0and10.0 (T=201C,compressionspeed5cm2

min 1

(7)

air waterinterfaceoreventoplungeintowaterwithanextended conformation.TheoxidizedchainofPazePCbeingmorepolarthan the chain of PoxnoPC, the extended conformation is favored (Khandelia andMouritsen,2009).Thisresultisin contradiction withhigher mean molecular areasobserved on ourisotherms, sinceanextendedconformationwouldratherfavoragatheringof moleculesintothemonolayerduringcompression.Moreover,the reversal of the oxidized chain of PazePC is favored by the deprotonation of the COOH group at the end of this chain, leadingtoahigheraf_ﬁnityofthispartofthemoleculeforwater. Underourexperimentalconditions,thecarboxylicfunctionofthe oxidizedchainisprotonatedasshownbyexperimentsperformed atdifferentpH valuesand in agreementwithitsexpectedpKa valuearound7 8(Nagleetal.,2013).Thisleadstotheconclusion thatthehighermeanmolecularareasobservedon(

p

A)isotherm of PazePC are probably not due to electrostatic repulsion, carboxylicfunctionontheoxidizedchainbeingneutral.Atlast, takingintoaccountthefactthatthisfunctionintheprotonated formisstableinlipidbilayersasshownbysimulationsonGluand Aspaminoacids(MacCallumetal.,2008),thecarboxylicfunction islikelyembeddedinlipidandtheoxidizedchainnotreversed.In thecaseofPoxnoPC,apartialreversal,oratleastatiltofthechain ascomparedtomonolayernormal,isnotexcluded.

p

A isotherms of lipid mixtures suggest that they are in a liquid expanded phase under our experimental conditions.

Analysis of mean molecularareas and compressibility modulus ofmixedmonolayersshowsthatthepresenceofPoxnoPCinduces their expansion and increases the monolayer elasticity. These resultscanbepartlycomparedtothoseobtainedbySabatiniand coworkers(Sabatinietal.,2006),whostudiedmonolayersofDPPC containing PoxnoPC. They observed that this lipid induces an expansion of the monolayers until a surface pressure around 20mNm 1_,_coupled_with_the_gradual_{disappearance}_of_the_phase

transition of DPPC and an increase of the ﬁlm elasticity. AFM images of PoxnoPC POPC LB _ﬁlmsarehomogenous, suggesting that PoxnoPC is regularly distributed in the monolayers, in agreementwiththeexpansion(Gaines,1966).

Aspreviouslymentioned,simulationsshowedthattheoxidized chainofPoxnoPCisabletoliemoreorlessparalleltotheinterface, expandingthemonolayerofafewÅ2₍_Khandelia_and_Mouritsen,

2009). In the case of PoxnoPC DPPC monolayers studied by Sabatiniandcoworkers,theexpansionisinthesamerange,except at a surface pressure of 10mNm 1 _where _expansion _is _higher

(around10Å2_)._Expansion_observed_in_the_case_of_{PoxnoPC POPC}

monolayers(between4.0and6.5Å2₎_are_in_agreement_with_these

previous results,even ifnoclearcorrelationhasbeenobserved betweentheamplitudeoftheexpansion andthepercentage of PoxnoPCcontrarytothecaseofDPPC(itcouldbeduetodifferent parameterssuchasdifferentorientationsoftheoxidizedchainin the same sample combined with the high ﬂuidity of POPC).

Fig.5.(A)HeightAFMimage(1010mm2

)ofamixedPazePC(20mol%)-POPCLBﬁlmtransferredat30mNm 1

.(B)Correspondingphaseimage.(C)Heightproﬁlesobtained fromsectionsindicatedintheimageofPazePC–POPCLBﬁlm.

(8)

Moreover,it suggests that theoxidized chain is more tiltedas comparedtothemonolayer normal inthepresenceof POPCas schematicallyshowninFig.6A,sinceasimilarorientationwould notcauseanexpansion.Italsoexplainsthehighercompressibility ofmixedmonolayers.

ResultsaredifferentinthecaseofPazePC POPCmonolayers, evenifPazePCincreasesalsothemonolayerelasticity(inalower extendthanPoxnoPC).Asmentionedintheresults,weassumethat thesemixturesbehaveideallyaccordingtothedeﬁnitionofGaines (Gaines, 1966). In the case of ideal mixtures, either lipids are miscible without interaction, or they form separated phases. Assuming that oxidized chains of PazePC molecules remain embedded in the plane of the monolayer without reversal (carboxylicgroupsbeingprotonated),bothhypothesesarepossible ifwesupposethattheoverallmolecularorganizationremainsthe samebetweenpureandmixedmonolayers.

Undertheseconditions,AFMisapertinenttooltounderstand thebehaviorofPazePC inPazePC POPCLBﬁlms.Corresponding imagesareonthewholehomogenous,atleastatsurfacepressure lowerthan30mNm 1_,_suggesting_a_mainly_homogenous_distribu

tionofboth lipids.However,imagesalsorevealthepresenceof areasthickerthanthesurroundingphase,thethicknessrangingon thewholefrom7to10nm.Theseareasareprobablyduetothe local formation of multilayer systems beneath the monolayer induced by compression (the monolayer transfer inducing the elevationofthesemultilayersystemsabovethesurfaceoftheLB ﬁlm, as sketched in Fig. 6B). Such structures appear probably duringthetimewherethesurfacepressureiskeptconstant(just beforeand during thetransfer),since nosignis visiblein

p

A isothermsandinCS 1measurements.However,someirregulari

tiesintheoveralllineardecreasewithtimeofthesurfaceoccupied by the monolayer are observed during the transfer (data not shown).Suchirregularitiescouldbethesignofadjustmentsinthe monolayer.Moreoverthesimultaneouspresenceofbothlipidsis necessarytoobservethesemultilayersystems,sincetheyarenot observedinthecaseofsinglecomponentLBﬁlms.Theseareasare eithernotobservediftheLBﬁlmistransferredatalowersurface

pressure,and theirdensity intotheLBﬁlmincreaseswhen the percentageofPazePCincreases.

Taking into account the average length of a lipid molecule around2 3nm(Endersetal.,2004;Kucerkaetal.,2005;Roiter etal.,2009;Orsietal.,2010),thethicknessrangeoftheseareas suggeststhat up toﬁve monomolecularlayerscould besuper imposedlocallyunderthemonolayerattheair waterinterface. Moreover,thephasesignalinAFMbeingsensitive,inthecaseof softmaterials,totheviscoelasticpropertiesandadhesionforces, phaseimagescanthusbelinkedtodissipativeprocesses.Inour case, contrastin thephase image (Fig. 5B) suggestsa different molecularorganizationbetweenthetwokindsofareas,onemade oflipidorganizedinahomogenousmonolayer(darkones)andone madeofmoreorganized/rigidmolecules(brightones).

Theseobservationsareonceagainpartlyinagreementwiththe studyofSabatiniandcoworkers(Sabatinietal.,2006),whonoted an apparent condensation of DPPC monolayers containing PoxnoPCandPazePCatsurfacepressuresabove45mNm 1,due infacttothedissolvingofoxidizedlipidsandmicelleformation. Theyalsosuggestedthat themonolayerareaswhere dissolving occurscouldbesmallenoughthatthesurroundingmatrixwould retainoxidizedlipids.Ourresultsconﬁrmthishypothesis,atleast inthecaseofPazePC,sincethemultilayersystemsobservedby AFMdonotdetachfromthemonolayer.Moreoverheightsections performed on AFM images enable also to estimate the mean diameter of suchareas around a few hundreds of nm. Micelle formation has also been proposed in ternary lipid mixtures containinghighpercentagesofoxidizedlipids(Meglietal.,2005). At last, the fact that we do not observe such protrusions in PoxnoPC POPCmonolayerscouldbeduetothefactthatrequired experimental conditions to observe them would be slightly different,forinstanceintermsofsurfacepressure.

Finally, both oxidized lipids are miscible with POPC. This miscibility conﬁrms results of a previous study by electron paramagnetic resonance of lipid vesicles containing oxidized lipidssuggestingthatﬂuidmembraneswouldbeprotectedagainst phaseseparation(Meglietal.,2005).Thisprotectionislikelydue

Fig.6.(A)SchemesummarizingtheorganizationofPoxnoPCandPazePCinpureandmixedmonolayers,deducedfromsurfacepressuremeasurementsandAFM;(B) SchemeofapossibleorganizationoflipidsinPazePC–POPCLBﬁlms,withtheformationofmultilayersystemsinducedbycompression,raisingabovetheleveloftheLBﬁlm duringthetransferofthemonolayerfromtheair–waterinterface.

(9)

tothefactthatPOPCisafluidlipid,ableofadjustmentattherootof only one phase. On the contrary, the study by fluorescence microscopyof DPPC monolayerscontaining PoxnoPCor PazePC shows a phase separation, with domains madeof DPPC in the liquidcondensed phase remainingintothe monolayers,even if theirsizeandshapearemodified(Sabatinietal.,2006).Inthesame idea,theadditionofPazePCtoPOPC POPE cardiolipinmixtures inducesabroadeningoftheheatprofileobtainedbydifferential scanningcalorimetry,suggestingthepresenceofPazePC richand PazePC poordomains(Wallgrenetal.,2012).Aphaseseparation wasalsoobservedinthecaseofDMPCvesiclescontainingeither PazePCorPoxnoPC,theeffectofoxidizedlipidsbeingvisibleevent ataverylowpercentageof2mol%(Wallgrenetal.,2013).

Atlast,AFMimagesofmixedmonolayersareclearlydifferent from those obtained in the case of POPC LB ﬁlms oxidating “naturally” in thepresence ofoxygen, showingthe appearance withtimeofdomainsregularlydistributed(Fayeetal.,2013).This differenceconﬁrmsourprevioushypothesis,based onthelocal oxidationofPOPCmoleculesandtheoxidationpropagation.

5. Conclusion

Thisworkshowsthatcouplingsurfacepressuremeasurements andAFMimagingoftransferredLBﬁlmsisapowerfulapproachto betterunderstandthebehaviorof oxidizedlipidsinmembrane models.Thiscombinationenabledustoconﬁrmthemiscibilityof PoxnoPCandPazePCinPOPCmonolayers,eveniftheseoxidized lipidsdonothaveprobablythesameconformation.Atlast,AFM showed,atleastinthecaseof PazePC,theformationofthicker areassuggestingthelocalformationofmultilayersystemsbeneath themonolayer.

Acknowledgements

We thank the Région Aquitaine and CNRS (France) for supportingthis workthrough thePh.D. grant of N.R. Faye and theequipmentoftheNanoSpectroImagerie(NSI LOMA)platform used in this work. We also thank the Direction Générale de l’Armement(DGA,MinistèredelaDéfense,France)andtheRégion Aquitaine(France)forthePh.D.grantofM.Mathelié Guinlet.

References

Ayuyan,A.G.,Cohen,F.S.,2006.Lipidperoxidespromotelargerafts:effectsof

excitationofprobesinﬂuorescencemicroscopyandelectrochemicalreactions

duringvesicleformation.Biophys.J.91,2172–2183.

Beranova,L.,Cwiklik,L.,Jurkiewicz,P.,Hof,M.,Jungwirth,P.,2010.Oxidation

changesphysicalpropertiesofphospholipidbilayers:ﬂuorescence

spectroscopyandmolecularsimulations.Langmuir26,6140–6144.

Castano,A.P.,Demidova,T.N.,Hamblin,M.R.,2004.Mechanismsinphotodynamic

therapy:partone-photosensitizers,photochemistryandcellularlocalization.

PhotodiagnosisPhotodyn.Ther.1,279–293.

Catala,A.,2012.Lipidperoxidationmodiﬁesthepictureofmembranesfromthe

ﬂuidmosaicmodeltothelipidwhiskermodel.Biochimie94,101–109.

Cwiklik,L.,Jungwirth,P.,2010.Massiveoxidationofphospholipidmembraneleads

toporecreationandbilayerdisintegration.Chem.Phys.Lett.486,99–103.

Davis,S.S.,Pontsler,A.V.,Marathe,G.K.,Harrison,K.A.,Murphy,R.C.,Hinshaw,J.C.,

Prestwich,G.D.,StHilaire,A.,Prescott,S.M.,Zimmerman,G.A.,McIntyre,T.M.,

2001.Oxidizedalkylphospholipidsarespeciﬁc,highafﬁnityperoxisome

proliferator-activatedreceptorgligandsandagonists.J.Biol.Chem.276,16015–

16023.

Deigner,H.-P.,Hermetter,A.,2008.Oxidizedphospholipids:emerginglipid

mediatorsinpathophysiology.Curr.Opin.Lipidol.19,289–294.

Eichenberger,K.,Böhni,P.,Winterhalter,K.H.,Kawato,S.,Richter,C.,1982.

Microsomallipidperoxidationcausesanincreaseintheorderofthemembrane

lipiddomain.FEBSLett.142,59–62.

Enders,O.,Ngezahayo,A.,Wiechmann,M.,Leisten,F.,Kolb,H.-A.,2004.Structural

calorimetryofmaintransitionofsupportedDMPCbilayersby

temperature-controlledAFM.Biophys.J.87,2522–2531.

Faye,N.R.,Moroté,F.,Grauby-Heywang,C.,Cohen-Bouhacina,T.,2013.Oxidationof

Langmuir–Blodgettﬁlmsofmonounsaturatedlipidsstudiedbyatomicforce

microscopy.Int.J.Nanotechnol.390–403(5/6/7).

Fruhwirth,G.O.,Loidl,A.,Hermetter,A.,2007.Oxidizedphospholipids:from

molecularpropertiestodisease.Biochim.Biophys.Acta1772,718–736.

Gabbita,S.P.,Subramaniam,R.,Allouch,F.,Carney,J.M.,Butterﬁeld,D.A.,1998.

Effectsofmitochondrialrespiratorystimulationonmembranelipidsand

proteins:anelectronparamagneticresonanceinvestigation.Biochim.Biophys.

Acta1372,163–173.

Gaines,G.L.,1966.Mixedmonolayers.In:Prigogine,I.(Ed.),FromInsoluble

MonolayersatLiquid–GasInterfaces.IntersciencesPublishers,pp.81–300.

Garcia-Manyès,S.,Sanz,F.,2010.Nanomechanicsoflipidbilayersbyforce

spectroscopywithAFM:aperspective.Biochim.Biophys.Acta1798,741–749.

Garcia-Manyès,S.,Domenèch,O.,Sanz,F.,Montero,M.T.,Hernandez-Borrel,J.,2007.

AtomicforcemicroscopyandforcespectroscopystudyofLangmuir–Blodgett

ﬁlmsformedbyheteroacidphospholipidsofbiologicalinterest.Biochim.

Biophys.Acta1768,1190–1198.

Girotti,A.W.,2001.Photosensitizedoxidationofmembranelipids:reaction

pathways,cytotoxiceffects,andcytoprotectivemechanisms.J.Photochem.

Photobiol.B63,103–113.

Grechishnikova,I.V.,Bergström,F.,Johansson,L.,Brown,R.E.,Molotkovsky,J.G.,

1999.Newﬂuorescentcholesterolanalogsasmembraneprobes.Biochim.

Greig,F.H.,Kennedy,S.,Spickett,C.M.,2012.Physiologicaleffectsofoxidized

phospholipidsandtheircellularsignalingmechanismsininﬂammation.Free

Radic.Biol.Med.52,266–280.

Janmey,P.A.,Kinnunen,P.K.J.,2006.Biophysicalpropertiesoflipidsanddynamic

membranes.TrendsCellBiol.16,538–546.

Khandelia,H.,Mouritsen,O.G.,2009.Lipidgymnastics:evidenceofcompleteacyl

chainreversalinoxidizedphospholipidsfrommolecularsimulations.Biophys.J.

96,2734–2743.

Khandelia,H.,Loubet,B.,Olzynska,A.,Jurkiewics,P.,Hof,M.,2014.Pairingof

cholesterolwithoxidizedphospholipidspeciesinlipidbilayers.SoftMatter10,

639–947.

Kinnunen,P.K.J.,Kaarniranta,K.,Mahalka,A.K.,2012.Protein-oxidizedphospholipid

interactionsincellularsignalingforcelldeath:frombiophysicstoclinical

correlations.Biochim.Biophys.Acta1818,2446–2455.

Koppaka,V.,Axelsen,P.H.,2000.Acceleratedaccumulationofamyloidbproteins

andoxidativelydamagedlipidmembranes.Biochemistry39,10011–10016.

Kucerka,N.,Liu,Y.,Chu,N.,Petrache,H.I.,Tristram-Nagle,S.,Nagle,J.F.,2005.

StructureoffullyhydratedﬂuidphaseDMPCandDLPClipidbilayersusingX-ray

scatteringfromorientedmultilamellararraysandfromunilamellarvesicles.

Biophys.J.88,2626–2637.

Lai,C.C.,Yang,S.H.,Finlayson-Pitts,B.J.,1994.Interactionsofmonolayersof

unsaturatedphosphocholineswithozoneattheair–waterinterface.Langmuir

10,4637–4644.

Li,X.-M.,Momsen,M.M.,Smaby,J.M.,Brockman,H.L.,Brown,R.E.,2001.Cholesterol

decreasestheinterfacialelasticityanddetergentsolubilityofsphingomyelins.

Biochemistry40,5954–5963.

Lis,M.,Wizert,A.,Przybylo,M.,Langner,M.,Swiatek,J.,Jungwirth,P.,Cwiklik,L.,

2011.Theeffectsoflipidoxidationonthewaterpermeabilityofphospholipids

bilayers.Phys.Chem.Chem.Phys.13,17555–17563.

MacCallum,J.L.,DrewBennet,W.F.,Tieleman,D.P.,2008.Distributionofaminoacids

inalipidbilayerfromcomputersimulations.Biophys.J.94,3393–3404.

Matti,V.,Säily,J.,Ryhänen,S.J.,Holopainen,J.M.,Borocci,S.,Mancini,G.,Kinnunen,

P.K.J.,2001.Characterizationofmixedmonolayersofphosphatidylcholineanda

dicationicGeminisurfactantSS-1withaLangmuirbalance:effectsofDNA.

Biophys.J.81,2135–2143.

Mattila,J.-P.,Sabatini,K.,Kinnunen,P.K.J.,2008.Interactionofcytochromcwith

1-palmitoyl-2-azelaoyl-sn-glycero-3-phosphocholine:evidenceforacylchain

reversal.Langmuir24,4157–4160.

Megli,F.M.,Sabatini,K.,2003.RespirationstateIV-generatedROSdestroythe

mitochondrialbilayerpackingorderinvitro.AnEPRstudy.FEBSLett.550,185–

189.

Megli,F.M.,Sabatini,K.,2004.Mitochondrialphospholipidbilayerstructureis

ruinedafterliveroxidativeinjuryinvivo.FEBSLett.573,68–72.

Megli,F.M.,Rosso,L.,Sabatini,K.,2005.Oxidizedphospholipidsinducephase

separationinlipidvesicles.FEBSLett.579,4577–4584.

Nagle,J.F.,etal.,2013.Generaldiscussion.FaradayDiscuss.161,563–589.

Nepomuceno,M.F.,Alonso,A.,Pereira-da-Silva,L.,Tabak,M.,1997.Inhibitoryeffect

ofdipyridamoleanditsderivativesonlipidperoxidationinmitochondria.Free

Radic.Biol.Med.23,1046–1054.

O’Donnel,V.B.,2011.Massspectrometryanalysisofoxidizedphosphatidylcholine

andphosphatidylethanolamine.Biochim.Biophys.Acta1811,818–826.

Orsi,M.,Michel,J.,Essex,J.W.,2010.Coarse-grainmodellingofDMPCandDOPC

lipidbilayers.J.Phys.Condens.Matter22,155106–155120.

Parkkila,P.,Steﬂ,M.,Olzynska,A.,Hof,M.,Kinnunen,P.K.J.,2015.Phospholipid

lateraldiffusioninphosphatidylcholine-sphingomyelin-cholesterol

monolayers;effectsofoxidativelytruncatedphosphatidylcholines.Biochim.

Roiter,Y.,Ornatska,M.,Rammohan,A.R.,Balakrishnan,J.,Heine,D.R.,Minko,S.,

2009.Interactionoflipidmembranewithnanostructuredsurfaces.Langmuir

25,6287–6299.

Sabatini,K.,Mattila,J.-P.,Megli,F.M.,Kinnunen,P.K.J.,2006.Characterizationoftwo

oxidativelymodiﬁedphospholipidsinmixedmonolayerswithDPPC.Biophys.J.

90,4488–4499.

(10)

Singer,S.J.,Nicolson,G.L.,1972.Theﬂuidmosaicmodelofthestructureofcell

membranes.Science175,720–731.

Uhlson,C.,Harrison,K.,Allen,C.B.,Ahmad,S.,White,C.W.,Murphy,R.C.,2002.

Oxidizedphospholipidsderivedfromozone-treatedlungsurfactantextract

reducemacrophageandepithelialcellviability.Chem.Res.Toxicol. 15,896–906.

Volinsky,R.,Kinnunen,P.K.J.,2013.Oxidizedphosphatidylcholinesin

membrane-levelcellularsignaling:frombiophysicstophysiologyandmolecularpathology.

FEBSJ.280,2806–2816.

Volinsky,R.,Cwiklik,L.,Jurkiewicz,P.,Hof,M.,Jungwirth,P.,Kinnunen,P.K.J.,2011.

Oxidizedphosphatidylcholinesfacilitatephospholipidﬂip-ﬂopinliposomes.

Biophys.J.101,1376–1384.

Volinsky,R.,Paananen,R.,Kinnunen,P.K.J.,2012.Oxidizedphosphatidylcholines

promotephaseseparationofcholesterol-sphingomyelindomains.Biophys.J.

103,247–254.

Wallgren,M.,Lidman,M.,Pham,Q.D.,Cyprych,K.,Gröbner,G.,2012.Theoxidized

phospholipidPazePCmodulatesinteractionsbetweenBaxandmitochondrial

membranes.Biochim.Biophys.Acta1818,2718–2724.

Wallgren,M.,Beranova,L.,Pham,Q.D.,Linh,K.,Lidman,M.,Procek,J.,Cyrpych,K.,

Kinnunen,P.K.J.,Hof,M.,Gröbner,G.,2013.Impactofoxidizedphospholipidson

thestructuralanddynamicorganizationofphospholipidmembranes:a