Hybrid Assemblies of Fluorescent Nanocrystals and Membrane Proteins in Liposomes

Hybrid Assemblies of Fluorescent Nanocrystals and MembraneProteins in LiposomesVincenzo De Leo,†,‡ Lucia Catucci,†,‡ Andrea Falqui,§ Roberto Marotta,§ Marinella Striccoli,‡

Angela Agostiano,†,‡ Roberto Comparelli,*,‡ and Francesco Milano*,‡

†Department of Chemistry, Universita degli Studi di Bari, Via Orabona 4, 70126 Bari, Italy‡CNR-IPCF Istituto per i Processi Chimici e Fisici, Sez. Bari, Via Orabona 4, 70126 Bari, Italy§Nanochemistry, Istituto Italiano di Tecnologia (I.I.T.), Via Morego 30, 16163 Genova, Italy

*S Supporting Information

ABSTRACT: Because of the growing potential of nanoparticles inbiological and medical applications, tuning and directing theirproperties toward a high compatibility with the aqueous biologicalmilieu is of remarkable relevance. Moreover, the capability tocombine nanocrystals (NCs) with biomolecules, such as proteins,offers great opportunities to design hybrid systems for bothnanobiotechnology and biomedical technology. Here we report onthe application of the micelle-to-vesicle transition (MVT) method forincorporation of hydrophobic, red-emitting CdSe@ZnS NCs into thebilayer of liposomes. This method enabled the construction of anovel hybrid proteo−NC−liposome containing, as model membraneprotein, the photosynthetic reaction center (RC) of Rhodobacter sphaeroides. Electron microscopy confirmed the insertion ofNCs within the lipid bilayer without significantly altering the structure of the unilamellar vesicles. The resulting aqueous NC−liposome suspensions showed low turbidity and kept unaltered the wavelengths of absorbance and emission peaks of the nativeNCs. A relative NC fluorescence quantum yield up to 8% was preserved after their incorporation in liposomes. Interestingly, inproteo−NC−liposomes, RC is not denatured by Cd-based NCs, retaining its structural and functional integrity as shown byabsorption spectra and flash-induced charge recombination kinetics. The outlined strategy can be extended in principle to anysuitably sized hydrophobic NC with similar surface chemistry and to any integral protein complex. Furthermore, the proposedapproach could be used in nanomedicine for the realization of theranostic systems and provides new, interesting perspectives forunderstanding the interactions between integral membrane proteins and nanoparticles, i.e., in nanotoxicology studies.

1. INTRODUCTION

Fluorescent semiconductor nanocrystals (NCs) are veryattractive materials for a wide number of applications indifferent fields of science and technology. The strong and size-dependent tunable photoluminescence (PL) and the broadabsorption spectrum, associated to a very high photostabilitywith respect to conventional organic dyes,1,2 make themappealing for nanomedicine and biological applications.3,4 Therelated literature is being progressively enriched withapplications involving in vivo and in vitro targeting, imaging,bioanalytical assays, drug delivery, disease treatment, and muchmore.5−8 On the other hand, the use of colloidal NCs forbiorelated applications has been limited by difficulties inobtaining biocompatible nano-objects without affecting theirpeculiar features under aqueous biological conditions.9 In fact,the key goal to fully exploit NC properties is getting highcontrol over their size, shape, and surface chemistry.10 Althoughseveral techniques have been proposed to directly synthesizeNCs in water,11 they are not suitable for a wide range of NCs.Conversely, one of the most promising synthetic approachesrelies on the thermal decomposition of suitable precursors in

the presence of terminating agents able to coordinate the NCsurface to control the growth rate and prevent aggregation.Such an approach results in very stable NCs with high controlover size and shape.12 The as-synthesized NCs are surroundedby a layer of terminating agents inferring stability in timeagainst aggregation and making the inorganic NCs soluble inorganic solvents. Nevertheless, the as-synthesized NCs are notwater-dispersible and therefore show poor affinity to abiological environment.5 Various strategies have been devel-oped to prevent such a drawback, based on the modification ofthe NC surface chemistry, preventing their aggregation, andretaining their peculiar optical properties: exchange of ligandson the surface, coating with silica shells, polymers, naturalhydrophilic molecules, and incorporation in micelles.13−17 Analternative and very promising approach exploits the NCinsertion into phospholipid-based liposomes.18

Received: October 28, 2013Revised: January 23, 2014Published: January 26, 2014

Article

pubs.acs.org/Langmuir

© 2014 American Chemical Society 1599 dx.doi.org/10.1021/la404160b | Langmuir 2014, 30, 1599−1608

Liposomes are small vesicles that spontaneously form whenlipid molecules are dispersed in an aqueous medium. Lipidmolecules organize themselves to form a double-layer structurewhich is, in principle, identical to the natural cell membranesand that traps an aqueous volume in their core. Indeedliposomes mimic the lipid scaffolding of biological membranesand have well-characterized physicochemical properties andphase behavior.19 The population of formed vesicles can betuned from tens of nanometers to tens of micrometers indiameter.20 Their appeal is derived from composition, whichmakes them biocompatible and biodegradable, and from well-established clinical use.21 Therefore, they arouse great interestfrom researchers as ideal carriers for biomedical imaging, drugdelivery, targeted therapy, and biosensing.22

Several authors have described the incorporation ofmetallic,23 semiconductor,24 and amorphous25 nanostructuredmaterials in liposomes, using various experimental strategies.Nanomaterials were either directly synthesized within lip-osomes26,27 or, more commonly, inserted after synthesis in aliposomal system.24,28 The fate of nanoparticles depends ontheir surface characteristics: hydrophilic ones can be easilytrapped inside the aqueous compartment of the vesicles,29,30

while hydrophobic ones can be embedded within the lipidbilayer.31,32 The latter possibility is obviously more interesting:to use liposomes as carriers for transferring hydrophobicnanomaterials in an aqueous environment, changing theiraffinity phase, and, at the same time, providing a biocompatibleshield against the biological environment.33 For this purpose,different approaches have been used such as incorporation intopreformed liposomes of nanoparticles dispersed with detergent,bulk hydration of a dried film of lipids and nanoparticles,emulsion processes, and reverse phase evaporation meth-ods.18,19,24,30,31,34−37

In this work, red-emitting hydrophobic CdSe@ZnS NCswere incorporated into liposomes of various phospholipidcomposition by the micelle-to-vesicle transition (MVT)method.38 This method offers a user-friendly strategy for theincorporation of hydrophobic NC into the bilayer of vesicleshaving narrow size distribution and, in addition, enables thesimultaneous insertion of integral membrane proteins, resultingin the production of hybrid vesicles.The construction of liposomes simultaneously containing

NCs and integral membrane proteins is described here for thefirst time by the successful coincorporation of CdSe@ZnS NCsand the photosynthetic reaction center (RC) from thebacterium Rhodobacter sphaeroides. The RC molecular weightis about 100 kDa, and the shape is approximately elliptical withaxes of 3 and 7 nm. It represents a good model for studying theinteraction between proteins and NCs in the phospholipidbilayer, because protocols for its purification and reconstitutionin the membrane are well tested39 and the presence ofbacteriochlorophylls as cofactors in its structure ensure strongand well-assigned signals through optical spectroscopictechniques.40,41

The effectiveness of the incorporation procedure wasinvestigated as a function of NC surface chemistry (namelythe chemical nature of the capping layer), NC concentration,type of detergent (nonionic, cationic, and anionic), and bilayercomposition (zwitterionic, anionic, and poly(ethylene glycol)(PEG)-modified lipids were tested). The surface chemistry andthe optical properties of the NCs were investigated by Fouriertransform infrared (FT-IR), absorbance, and photolumines-cence spectroscopy. The size, the morphology, and the stability

of CdSe@ZnS-loaded liposomes were checked by dynamiclight scattering (DLS) measurements and cryo-EM inves-tigation. The structural and functional survival of RC wasverified by recording the visible-near infrared (vis-NIR) spectraand the charge recombination kinetics.This work creates new perspectives in both nanomedicine

and nanotoxicity: hybrid liposomes can be used in nano-medicine for the realization of theranostic systems, i.e., withcapability of transporting in cells both protein complexes withtherapeutic activity and nanostructured agents useful for cellimaging and other diagnostic applications. Furthermore, theycan be used in nanotoxicology to broaden the knowledge onthe interaction between nanomaterials and integral membraneproteins, which today is limited for the most part to solubleproteins.

2. EXPERIMENTAL SECTIONMaterials. All chemicals were purchased with the highest purity

available and were used without further purification. Cadmium oxide(CdO, powder 99.5%), selenium (Se, powder 99,99%), oleic acid(OLEA, technical grade 90%), trioctylphosphine oxide (TOPO, 99%and technical grade), tributylphosphine (TBP, 99%), trioctylphosphine(TOP, technical grade), tert-butylphosphonic acid (TBPA), diethylzinc(1.0 M solution in heptane), 1-octanethiol (OTT), 1-dodecanethiol(DDT), and hexamethyldisilathiane (HMDT) were purchased fromAldrich. Hexadecylamine (HDA) was purchased from Fluka.

The reagent grade salts for the 50 mM K-phosphate, 100 mM KCl(pH 7.0) buffer solutions, sodium cholate (SC), octyl glucoside (OG),cetyltrimethylammonium bromide (CTAB), phosphatidylglycerol(PG), diphosphatidylglycerol (cardiolipin, CL), and phosphatidylser-ine (brain extract, type III: Folch fraction III from bovine brain, PS)were purchased from Sigma. 1,2-Dipalmitoyl-sn-glycero-3-phosphoe-thanolamine-N-[methoxy(polyethylene glycol)-2000] (16:0 PEG-2000-PE) and 1,2-dipalmitoyl-sn-glycero-3-phosphothioethanol(PLT) were purchased from Avanti Polar Lipids. Phosphatidylcholine(PC) was a kind gift from Lipoid.

Synthesis of CdSe@ZnS Nanocrystals. CdSe@ZnS colloidalcore−shell NCs were prepared by using a slightly modified literaturesynthetic procedure.42 All the synthetic steps were performed understandard airless conditions using a Schlenk line. In a typical synthesis,CdO (1 mmol) was dissolved in a TOPO:HDA:TBPA mixture(15:25:1). The resulting mixture was heated at 290 °C, and 2 mL ofTBP was added to the reaction mixture. CdSe nucleation waspromoted at 300 °C upon sudden injection of Se precursor solution,consisting of 5 mmol of Se dissolved in 20 mmol of TBP. The NCgrowth was carried out at 270 °C for 10 min. After that, thetemperature was lowered to 110 °C and the NCs were annealed for 1h to promote surface reconstruction before growing the ZnS shell.This step was performed using the same reaction mixture without anyintermediate purification step, raising the temperature to 150 °C. Theproper amount of a Zn(C2H5)2:HMDT (1:1) stock solution in TBP(50 mmol) was dropwise injected, and the shell growth was monitoredby UV−vis absorption and PL spectra. The as-prepared core−shellNCs were precipitated by methanol and recovered by centrifugationthree times. Then the obtained reddish powder was dissolved inCHCl3 for optical and structural characterization.

CdSe@ZnS Capping Exchange. The capping exchange proce-dure exploits the strong affinity of thiol groups for the CdSe@ZnS NCsurface. Here several alkyl thiols, basically differing in their alkyl chainlength (i.e., in their steric hindrance), were tested, namely OTT, DDT,EDT, and PLT. In a typical procedure, alkyl thiol was added to aCHCl3 solution of washed CdSe@ZnS NCs (usually in molar ratio2000:1) and vigorously stirred for 24 h at room temperature underambient atmosphere. The excess of capping ligand was removed byrepeatedly washing via the standard precipitation/centrifugationprocedure. Then the precipitate was dissolved in the proper amountof CHCl3 for incorporation in liposomes. The effectiveness of thecapping exchange procedure was assessed by UV−vis absorbance

Langmuir Article

dx.doi.org/10.1021/la404160b | Langmuir 2014, 30, 1599−16081600

measurements, steady-state and time-resolved photoluminescencespectroscopy, and FT-IR-ATR measurements.Encapsulation of the CdSe@ZnS Core−Shell NCs in the

Phospholipid Bilayer of the Liposome. NC−liposomes wereprepared by suitable modification of the micelle-to-vesicle transition(MVT) method, consisting of the detergent removal fromphospholipid-containing mixed micelles to induce liposome forma-tion.38 Several nonionic and anionic phospholipids were tested, such asPC, PG, CL, and PS, either separately or in mixture.43 For thesterically stabilized liposomes, PEG-2000-PE was added to the lipidblend. Lipids and NCs were dissolved in CHCl3 to give the desiredratio and then dried with a gentle nitrogen flux to form ahomogeneous film on the walls of a conical glass tube. Solventremoval was completed under vacuum conditions (24 h) by a no-oilpump operating at 1 mbar. After that, 0.5 mL of 4% SC, or OG, orCTAB in 50 mM K-phosphate/100 mM KCl (pH 7.0) was added tothe dry lipid film and then sonicated (20 shots with a BransonSonicator 250) to form a clear, translucent mixed micelle solution. Thelatter was loaded into a glass column (20 × 1 cm) packed with G-50Sephadex Superfine (Sigma) equilibrated with 50 mM K-phosphate/100 mM KCl (pH 7.0) for detergent removal by size exclusionchromatography (SEC). NC−liposomes eluted after a void volume ofabout 1.5 mL.Encapsulation of Photosynthetic RC in the Phospholipid

Bilayer of the Liposome. RC was isolated from Rhodobactersphaeroides strain R-26.1 as previously described.44 RC− and RC−NC−liposomes were prepared by the MVT method. The mixedmicelle solution, obtained as described in the previous section, wasadded to 70 μM RC stock solution to a desired NC/RC ratio. The RCmixed micelle solution was vigorously shaken for 30 s and kept at 4 °Cfor 20 min before loading it on the column for the SEC.UV−Vis−NIR and Emission Spectroscopy. Absorption meas-

urements were performed by means of a UV/vis/NIR Cary 5000Spectrophotometer (Varian). The photoluminescence (PL) spectrawere recorded by using an Eclypse Spectrofluorimeter (Varian). Therelative PL quantum yields (QY) of NCs in solution and liposomesuspension were estimated by using rhodamine 101 as reference dyeand comparing the integrated PL intensity of the NCs and the dye,both recorded in excited solutions or suspension at the sameabsorbance (<0.1 au, to minimize possible reabsorption effects), asreported elsewhere.42

Time-Resolved Photoluminescence Spectroscopy. The fluo-rescence decay was investigated by time-correlated single-photoncounting (TCSPC). The measurements were performed for both NCsolutions and liposome suspensions using a FluoroHub (HORIBAJobin-Yvon). The samples were excited at 375 nm using a picosecondlaser diode (NanoLED 375L) emitting 80 ps pulses at a 1 MHzrepetition rate. The PL signals were dispersed by a double gratingmonochromator and detected by a picosecond photon counter (TBXps Photon Detection Module, HORIBA Jobin-Yvon). Time resolutionof experimental setup was estimated in ∼200 ps.Charge Recombination Kinetics of RC. RC activity was checked

by recording at 865 nm the charge recombination kinetics using akinetic spectrometer of local design as previously described.44 Theexcitation was accomplished by a xenon flash lamp (Hamamatsu) witha pulse length of ∼100 μs, well below the time constant of the fastestrecorded process (100 ms). The rapid changes in the opticalabsorbance were collected on a R928 photomultiplier (Hamamatsu)whose output, after filtering and amplification, was collected with adigital Oscilloscope (Tektronix, Inc., TKS3052, Beaverton, OR) andstored on a computer. The fitting on the kinetic traces was performedusing a nonlinear least-squares fitting algorithm implemented onlocally developed software.FT-IR-ATR Spectroscopy. Midinfrared spectra were acquired with

a Varian 670-IR spectrometer equipped with a DTGS (deuteratedtryglycine sulfate) detector. The spectral resolution used for allexperiments was 4 cm−1. For attenuated total reflection (ATR)measurements, the internal reflection element (IRE) used was a one-bounce 2 mm diameter diamond microprism. Cast films have beenprepared directly onto the internal reflection element, by depositing

the solution or suspension of interest (3−5 μL) on the upper face ofthe diamond crystal and allowing the solvent to evaporate.

Dynamic Light Scattering. Dynamic light scattering measure-ments (DLS) were performed using a HORIBA Dynamic LightScattering Particle Size Analyzer LB-550 instrument (Horiba JobinYvonne), equipped with a laser diode source (wavelength 650 nm, 5mW). Measurements were performed at 25 °C with 100 sampling.The experimental time-course changes in the measured light intensityare converted in the frequency distribution, or Power Spectrum (PS),by fast Fourier transformation. The particle size distribution is thenobtained by comparing iteratively, using the Twomey method, theexperimental PS with calculated frequency distribution based on theintensity of the Brownian movement for particles of different size.

Cryoelectron Microscopy. For cryo-EM analysis, 3 μL of eachsample was deposited on Quantifoil grids (Electron MicroscopySciences) previously glow-discharged in a Solarus plasma cleaner(Gatan, Inc.). Grids were automatically vitrified in a Vitrobot Mark IV(FEI) cryosample plunger. TEM images were recorded using low-doseconditions at −170 °C using a Titan Krios electron microscope (FEI)equipped with a FEG (field emission gun) and operating at 300 kV.Images were recorded using a FEI Falcon Direct Electron Detector.

3. RESULTS AND DISCUSSIONWe used the MVT method to encapsulate CdSe@ZnS NCs inthe lipid bilayer of liposomes. The experimental steps of theprocedure can be briefly described as follows. First, thetrioctylphosphine oxide (TOPO) capping on the NC surfacewas replaced by alkyl thiols to improve the hydrophobicinterdigitation between the organic shell of NCs and thehydrophobic tails of lipids. NCs were then dispersed in water(Figure 1A) together with the phospholipids of interest in amicroheterogeneous system such as mixed micelles, by meansof a detergent solution. Finally, NC−liposomes wereassembled, removing detergent molecules by size exclusionchromatography (SEC) (see details below). For the RC−NC−liposome preparation, the appropriate amount of detergent-stabilized RC stock solution was added to micellar solution andthen vigorously shaken for 30 s (Figure 1B).

CdSe@ZnS NC Capping Agents. Differently from whatwas reported in a previous work,32 under our experimentalconditions, the presence of the native TOPO capping layer wasdetrimental for the incorporation of NCs into the lipid bilayerof liposomes, not allowing the formation of mixed micelles,irrespective of the NC size. The average diameter of the usedred NCs measured by TEM (Figure S3 in the SupportingInformation) is 3.3 ± 0.3 nm, considering only the inorganicmaterial. Fully extended TOPO molecules should add about 2nm to the particle diameter.18 Nevertheless, it is reasonable tothink that the organic molecules are arranged around the NCsurface in a not fully elongated way, so the values reportedabove can be considered an upper limit to the effective particlesizes. The theoretical thickness of a pure PC bilayer is around 4nm.45 In our PC:PS:PEG-2000PE liposomes, by TEM wemeasured an average thickness of 5 nm. Therefore, theunsuccessful incorporation of TOPO-coated NCs can beascribed to the steric hindrance of the three alkyl chains of asingle TOPO molecule rather than to a size mismatch. Aproper NC surface functionalization was thus necessary toachieve an effective incorporation of NCs into liposomes. Alkylthiols with different lengths of the carbon chain along with 1,2-dipalmitoyl-sn-glycero-3-phosphothioethanol (PLT), a syn-thetic S−H-terminated phospholipid, were chosen to replacethe pristine TOPO ligand, owing to the high affinity of the thiolmoiety for the CdSe@ZnS surface. Indeed the presence of alinear alkyl chain promoted the hydrophobic interaction with

Langmuir Article

dx.doi.org/10.1021/la404160b | Langmuir 2014, 30, 1599−16081601

lipid tails, despite their greater length with respect to TOPOmolecules. For example, in the case of fully extended DDT andeven for the PLT molecules, 3.6 and 4.8 nm should be addedrespectively to the particle diameter, confirming that thedecisive factor for the NC incorporation is not only the particlesize but also the coating affinity for the lipid bilayer.The effectiveness of the capping exchange procedure was

verified by means of FT-IR spectroscopy (Figure S1 inSupporting Information) besides the NC ability to incorporatein the liposome bilayer. The spectral characteristics of NCswere similar before and after treatment with thiols, andnegligible shifts of the absorbance and fluorescence peaks wereobserved (Figure 2), suggesting that no NC aggregation occurupon capping exchange, as confirmed by TEM analysis (FigureS3, panel B, in Supporting Information). The NC quantumyield (QY) in chloroform, calculated with respect to rhodamine101, after capping exchange showed that the emissionproperties were mainly retained, irrespective of the testedmolecules and, interestingly, even improved in the case of 1-octanethiol (OTT) and DDT (Figure 2).To gain a better insight on how the capping exchange can

modify the surface properties of the CdSe@ZnS NCs underinvestigation, the recombination dynamics was investigated bytime-correlated single-photon counting (TCSPC). The highsensitivity to surface states and to the chemical environmentsafforded by luminescence makes this spectroscopic technique

an extremely useful tool to investigate modifications ofrelaxation dynamics of NCs in the presence of differentsurrounding media. The PL decay of NCs appeared onlyweakly influenced by the capping exchange, as a slightly slowerkinetics was observed for DDT-capped NCs with respect to thenative ones (Figure S2 in Supporting Information), in line withthe increase in PL QY. This effect can be probably ascribed to abetter surface passivation of CdSe@ZnS NCs in the presence ofthiol groups, as recently observed by Aguilera-Sigalat et al.46

CdSe@ZnS NC−Liposomes: Preparation and Charac-terization. NC−liposomes were prepared by a suitablemodification of the MVT method starting from mixed micelles,consisting of a colloidal, microheterogeneous suspension ofmicelle-forming detergent molecules and lipid molecules. Theyusually appear as disclike, sheetlike, or cylindrical aggregates,38

and their structure consists of lipid bilayer fragments in whichthe detergent molecules are mainly distributed at thehydrophobic edges. In this work, mixed micelles were formedby sonication of a lipid−NC thin film added to a high criticalmicellar concentration (CMC) detergent solution. In this way,NCs remain trapped within the lipid bilayer region of the mixedmicelles. Detergent is also present in the bulk phase asnonaggregated monomeric molecules that are in rapidequilibrium with mixed micelles.38 By means of the SECtechnique, the detergent was easily removed from the solutionbecause, during the elution, detergent monomers are slowerwith respect to the bigger aggregates. To maintain theequilibrium, detergent molecules move progressively frommixed micelles into the bulk solution, promoting the mergingof bilayer fragments until closed vesicles are formed47 (Figure1A). To verify the complete removal of detergent, FT-IRspectra were recorded during the steps of the experiment.

Figure 1. Schematic representation of the MVT method applied to theformation of NC−liposomes (A) and RC−NC−liposomes (B). Lipidsand NCs were first combined in a dried film. Mixed micelles were thenobtained by sonication in a detergent solution. For RC−NCliposomes, purified detergent-solubilized RCs were added after thesonication step to avoid protein denaturation. Detergent removal wasachieved by SEC to induce reorganization of lipids in liposomes. Thissketch is merely qualitative. The dimensions of the differentcomponents are not to scale.

Figure 2. Absorption (blue line) and photoluminescence (red line)spectra (λex = 400 nm) of CdSe@ZnS core−shell in chloroform withdifferent capping agents: TOPO (trioctylphosphine oxide), PLT (1,2-dipalmitoyl-sn-glycero-3-phosphothioethanol), OTT (1-octanethiol),DDT (1-dodecanethiol). QY is relative to rhodamine 101.

Langmuir Article

dx.doi.org/10.1021/la404160b | Langmuir 2014, 30, 1599−16081602

Figure 3 shows the actual formation of pure phosphatidylcho-line (PC) liposome containing DDT-capped NCs. The PC is

identified by a characteristic peak at 1736 cm−1 attributable toester carbonyl stretching, while the presence of detergentsodium cholate (SC) is easily revealed by the 1555 cm−1 banddue to the asymmetric carboxylate stretching.39 The infraredspectrum relevant to liposomes evidently is lacking thecarboxylate stretching band and has the same overall featuresof the PC spectrum.Cryo-EM images of NC−liposomes (PC:phosphatidylserine

(PS):phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000 (PEG-2000PE) 2:1:0.1; DDT-capped CdSe@ZnS; lipid−NC ratio of about 4000:1) show a population mainly composedof small unilamellar vesicles with inorganic NCs, identifiable asdark spots in the thickness of the lipid bilayer (Figure 4A). TheMVT method was unambiguously able to generate hybrid NC−vesicles, with the same architecture of vesicles made in the

absence of NCs (Figure 4B), with narrow size distribution andwithout large aggregates of NCs or lipids.The lipid bilayer thus appears to accommodate hydrophobic

particles with a diameter close to its thickness by a mechanismsimilar to that reported for the integral membrane proteins, aspreviously hypothesized.19 It is remarkable that, differentlyfrom previous reports,32 this result was obtained with large red-emitting CdSe@ZnS that are considered a suitable tool in tissueimaging due to lower autofluorescence background comparedto the smaller green NCs. Although a great number of vesiclescan be seen showing some well-dispersed nanoparticles in thelipid bilayer, empty vesicles and entirely loaded liposomes werealso observed (Figure 4C). This phenomenon was previouslynoticed in hybrid Au nanoparticle (NP)−liposomes and wasprobably related to the thermodynamics of hydrophobic NPinsertion into the lipid bilayer.18 The side by side association ofnanoparticles into the bilayer may indeed reduce strainedregions at the NC−lipid interface and void space around thenanoparticles, minimizing the energy penalty to deform thebilayer due to NC insertion.18,48 The sporadic presence ofJanus-type NC−vesicle hybrids (Figure 4D), similar to thosefound by Rasch et al., further supports this hypothesis. Theresolution of the reported cryo-EM imaging is lower than thatwhich can be obtained in a standard HRTEM experiment. Thisis due to the concurrence of some experimental conditions.First, the material containing NC−liposomes is constituted byvitrified water, the thickness and scattering power of such anembedding layer not being low. Second, the electronabsorption as well as the intrinsic size of the nanocrystalscontained in the liposomes is quite small. As a consequence, thesimultaneous occurrence of these conditions did not allow us toachieve a resolution better than that reported here. As expected,the average diameter of the vesicles, as estimated by EM, wassignificantly smaller than that obtained by DLS measurement,as this latter technique is strongly influenced by the presence ofeven a small fraction of large particles (Table S2 in SupportingInformation). Indeed, the intensity of the scattered light isproportional to the sixth power of the particle radius. Liposomediameter was affected by the NC content and, when the lipid−NC ratio was decreased below 3000:1, isolated vesicles denselyloaded with NCs were observed (Figure S4 in SupportingInformation). The number of aggregates and debris under theseconditions was high and increased progressively with a decreasein the ratio of lipid to NC. These systems showed largeinstability (a behavior similar to that found with other methodsof preparation),18 and also the optical properties of the systemundergo a sharp deterioration due to the phenomena ofscattering, hampering the characterization with conventionalspectroscopic techniques.Liposomes and NC−liposomes were also prepared by

varying several parameters, such as the chemical nature of theNC capping agent and detergent as well as the composition ofthe lipid blend. Table S2 (Supporting Information) shows that,irrespective of the length of thiol molecules and of thecomposition of the lipid blend, NCs were successfullyincorporated in liposomes. The method used allowed us toprepare vesicles of sizes ranging from about 50 to 400 nm,namely from small unilamellar vesicles (SUVs) to largeunilamellar vesicles (LUVs), depending on the lipid composi-tion and on the amount of NCs hosted in the bilayer. The sizedistribution of the obtained vesicles was narrow enough so thatany subsequent extrusion step was not necessary. Differentlyfrom other reports,18 we were able to obtain NC−liposomes

Figure 3. FT-IR spectra of DDT-capped CdSe@ZnS NCs (A), purePC (B), pure SC (C), mixed micelles of PC, SC, and NCs (D), andNC−liposome (E). Note the similarity between B and E spectra.

Figure 4. Cryo-EM images of NC−liposomes (PC:PS:PEG-2000PE)2:1:0.1, CdSe@ZnS DDT-capped, lipid−NC ratio of about 4000:1).(A) Hybrid NC−vesicles (arrowheads); (B) empty vesicles (arrow-heads); (C) entirely NC-loaded vesicles (arrowheads), (D) sporadicJanus-type NC−vesicles (arrowheads).

Langmuir Article

dx.doi.org/10.1021/la404160b | Langmuir 2014, 30, 1599−16081603

using a different type of detergent, such as the nonionic octylglucoside (OG), the cationic cetyltrimethylammonium bromide(CTAB), and the anionic SC (Table S2, SupportingInformation). The use of SC is particularly advantageousbecause of its biocompatibility, low cost, stability at roomtemperature, good solubility, and high CMC of about 14 mM.UV−vis absorbance and fluorescence spectra of NCs

embedded in liposomes (Figure 5) closely resemble those of

NCs in chloroform (see also Figure 2) although a weak red-shift of a few nanometers in the spectra occurs, indicating anelectronic rearrangement of the charge carriers due to themodified chemical surrounding. The absorption spectrum ofthe liposome suspension is distorted by the well-known Tyndalleffect, namely the light scattering by particles in a fine colloidalsuspension: the longer wavelength light is transmitted morewhile the shorter wavelength light is reflected more viascattering with a wavelength dependence proportional to thefourth power of the frequency.This effect is evident when the cross-section of the

suspended particles falls in the range 40−900 nm, i.e.,somewhat below or near the wavelength of visible light.Nevertheless our NC−liposome suspensions, with a meandiameter below 100 nm, showed limited turbidity (see FigureS5 of Supporting Information).Figure 6A shows a comparison between normalized steady-

state fluorescence spectra of NCs in their native chloroformsolution, and fluorescence spectra when the luminescentparticles were incorporated in the liposome bilayer. From the

shape of the spectra, a quite inhomogeneous broadening in thelow energy side occurs for NCs confined in liposomes. Inaddition, as shown in Figure 6B, the transfer of the NCs fromchloroform solution to lipid bilayer caused a decrease inlifetimes of the excited state of NCs due to the lower QY(below 10% in liposomes vs about 40% in chloroform).This behavior, together with the shape of PL band, can be

ascribed to energy transfer processes from higher energy smallNCs to lower energy large NCs belonging to the samenanocrystal population, when packed together in the liposome,as also confirmed by cryo-EM images. However, thefluorescence intensity of the aqueous suspension of NC−liposomes was lower compared to that of NC solution inchloroform having the same concentration. This effect wasfrequently noticed during various strategies adopted to transferNCs in water.24,49,50 The mechanism beyond this phenomenonis not yet fully understood51 because many parameters cangenerally affect the fluorescence intensity of NCs: (i)aggregation phenomena, (ii) self-quenching, (iii) changes inthe surface of nanoparticles due to interaction with water, (iv)interaction with quenchers,24 if present in the system.Nonetheless, we observed that the QY measured in liposomesof identical lipid composition decreased with an increase in theconcentration of NCs (Table S3), suggesting that suchfluorescence quenching could be ascribed to self-quenchingdue to the high concentration of NCs in the lipid bilayer. Theconcentration of nanocrystals within the lipid bilayer [NC]bilayercan be calculated given the bulk concentration of phospholipid[PL] and NCs [NC]bulk, the radius of the liposomes (rH), thebilayer thickness (ρ), and the polar head area of thephospholipids (σ):39

ρ σ=

− −r

r r N[NC] [NC]

6( ( ) ) [PL]bilayer bulk

H2

H3

H3

A (1)

Under the realized conditions, by using ρ= 4 nm, σ = 0.72nm2, and assuming that all NCs added are incorporated in thelipid bilayer, the concentration of NCs in liposomes can reachmaximum values on the order of 10−5 M (Table S3, SupportingInformation) starting from bulk concentrations of 10−7 M.Extracting NCs from lipids and transferring them back to theorganic phase, a restitutio of the native QY was observed (seeSupporting Information for details). Therefore, it can beassumed that the lower value of QY of NCs in liposomes ismainly ascribed to their peculiar arrangement in a highlyorganized system and that self-quenching phenomena comefrom NC crowding within the bilayer. Anyway, the decrease ofbrightness observed after the insertion of CdSe@ZnS in

Figure 5. Absorbance (blue line) and fluorescence spectra (red line)(λex = 400 nm) of CdSe@ZnS NC−liposome (PC:PS 2:1, overalllipid−NC molar ratio 15 000:1).

Figure 6. Normalized steady-state fluorescence spectra (λex = 400 nm) of DDT-capped NCs in CHCl3, and NC−liposomes (PC:PS 2:1) in water.(B) Time-resolved PL decay curves (λex = 375 nm, λem = 600 nm) from DDT-capped NCs in CHCl3, and NC−liposomes (PC:PS 2:1) in water.

Langmuir Article

dx.doi.org/10.1021/la404160b | Langmuir 2014, 30, 1599−16081604

liposomes did not invalidate the possibility of its use forbiological labeling, because a fluorescence QY between 3% and8% was measured for our vesicles whereas a QY ranging from1% to 5% has been reported to be sufficient for cell-labelingstudies.52

NC−Liposome Stability. The stability of liposomes overtime is determined by a chemical factor, depending onoxidation and hydrolysis of phospholipids, and by a physicalfactor, depending on the tendency of liposomes to aggregateand to settle. The chemical stability can be increased byavoiding exposure to light, high temperature, and oxygen.Aggregation may be limited by appropriately altering theformulation of the liposomes. The introduction of negativecharges on the liposome surface by adding anionic lipids suchas phosphatidylglycerol (PG), diphosphatidylglycerol (CL), orPS allowed us to obtain preparations that were stable for severaldays through electrostatic repulsion. Finally, the introduction ofPEG lipids to form a polymeric cloud on the liposome surfaceresulted in a reduction of the attractive forces through sterichindrance from the poly(ethylene glycol) chains. Liposomesmade with PC:PS:PEG-2000-PE in a 2:1:0.05 proportion,containing NCs at an overall lipid−NC molar ratio of 3000:1,remained stable for several weeks (see Supporting Informationfor more details). The introduction of PEG lipids to theliposome is also advantageous for bioapplications because theyare nonimmunogenic and nonantigenic, favor systemiccirculation time, and lower the toxicity of the encapsulatednanoparticles.5,50 Furthermore, PEG lipids can be function-alized with specific reactive groups for targeting purposes.50

Interaction between RC Membrane Proteins andCdSe@ZnS in RC−NC−Liposomes. Among the advantagesof the MVT method for incorporating NCs in liposomes (themethod is simple, efficient, and fast53), there is the possibility ofthe simultaneous functional reconstitution of integral mem-brane proteins. Indeed among all the techniques available forliposome preparation, those using detergents are most efficientto reconstitute proteoliposomes.53 This opens the possibility tostudy the possible interactions between NCs and hydrophobicmembrane proteins in close proximity when coincorporated inthe same vesicle. To investigate this potentiality, we selected asmodel protein the well-characterized photosynthetic RCisolated from Rhodobacter sphaeroides R26.54 RC is amembrane-spanning protein composed of three subunits (L,M, and H) and nine cofactors in the protein scaffold: twochemically identical ubiquinone-10 (UQ10) molecules, one ironion, two bacteriopheophytins (BF), and four bacteriochlor-ophylls (Bchl), two of which form a functional dimer (D). Thepresence of BF and Bchl in its structure ensures strong andwell-assigned signals through optical spectroscopic techni-ques.40

The integrity of the protein in hybrid RC−NC−liposomeswas verified at structural and functional level. Figure 7 shows acomparison among the absorption spectra of NC−liposomes,of RC−liposomes, and RC−NC−liposomes.The peak at 595 nm of the hybrid vesicle clearly results from

the convolution of the peak at 600 nm of the RC and the peakat 590 nm of the CdSe@ZnS NCs, confirming their successfulcoincorporation in the vesicles. Indeed in this study we carriedout an SEC for the detergent depletion in the MVT method,assuring the removal of the proteins eventually not embeddedin the liposomes. DLS data also confirm the coincorporation ofRC and NCs in the same vesicles (see Figure S9 in SupportingInformation), showing for the hybrid RC−NC system a

narrow, unimodal size distribution, with an average size greaterthan those of the RC−liposomes and NC−liposomes. Effectiveincorporation of RC in the bilayer is further confirmed by achange in the rate constant of the charge-separated state decay,which is about 1 s−1 in the starting micellar solution and about0.5 s−1 in our PC liposomes (see below and SupportingInformation for more details).The 865 nm signal, arising from excitonic coupling between

two Bchl that are spatially arranged to form dimer D, is alsovery sensitive to small changes in the mutual positions of thetwo molecules; we observed that, in hybrid vesicles, the ratiobetween this signal and peak at 800 nm, referred to as themonomers of Bchl, remained constant (taking into account thecontribution of scattering) with respect to RC−liposomes.Moreover, no significant shifts in the peaks were observed, andtherefore it can be concluded that the chemical environmentaround the cofactors was not modified, indicating that theprotein scaffold of RC remains structurally intact even in thepresence of NCs.Further insight on the protein integrity can be gained by a

photochemical assay: being the pivotal centrum of thephotosynthetic apparatus, RCs undergo a charge separationreaction upon light excitation between the primary electrondonor D and the quinone electron acceptors, followed by acharge recombination (CR) reaction. This reaction can beeasily spectrophotometrically followed by the light-induceddisappearance and subsequent dark recovery of the 865 nmband, whose kinetics is extremely sensitive to the chemicalenvironment surrounding the RC.55

The inset in Figure 7 shows the CR kinetics relative to ahybrid RC−NC−liposome of pure PC compared with the CRkinetics of the RC-only loaded liposome of the samecomposition. Both the amplitude and rate constant of the CRfor RCs in liposomes with or without NCs are very similar,indicating that RC in the presence of NCs are completelyphotoactive and that the energetic levels of the cofactors remainunperturbed (more detailed information about CR kinetics ofRC are provided in paragraph 2 of Supporting Information).Therefore, we can infer that the protein within the hybrid RC−NC−liposomes retains its structural and functional integrity, asalso recently found in micellar systems.56

Figure 7. Absorbance spectra of RC−liposomes ([RC] = 2 μM), NC−liposomes ([NC] = 0.8 μM), and hybrid RC−NC−liposomes ([RC] =2 μM, [NC] = 0.8 μM). In all samples, PC:PS = 2:1. Inset: chargerecombination kinetics of RC in liposomes of pure PC in the presenceand absence of CdSe@ZnS NCs. In both cases, an excess ofubiquinone was added to saturate the QB site of the protein.

Langmuir Article

dx.doi.org/10.1021/la404160b | Langmuir 2014, 30, 1599−16081605

On the other hand, NCs also mainly retained their opticalproperties in the presence of RC protein. Indeed relaxationdynamics of NCs in liposomes, with and without RC protein,did not show substantial differences under the realizedconditions (Figure 8 A).However, a slight fluorescence quenching of NCs in the

presence of RCs, that rises with an increase in RCconcentration, was measured. Such a quenching can be ascribedto a mere inner filter effect due to the overlap between theemission peak of the NCs and the RC absorption spectrumaround 600 nm (Figure 8 B). The average distance between theNC and the RC in the bilayer was found to be 4 times the valueof their Forster distance (R0), as calculated and discussed inSupporting Information; therefore, fluorescence resonanceenergy transfer (FRET) was not observed.The reported data clearly demonstrate that vesicle scaffolds

can support multiple components with different functionalities,without reciprocal destructive effects, as needed for manyappealing applications, such as combined diagnosis and therapyin medicine.57

4. CONCLUSIONWe have shown that the MVT method can be advantageouslyused for the preparation of hybrid NC−liposome systems withnarrow size distribution and good stability. Optical propertiesand in particular a good level of emitted fluorescence weremaintained for NCs loaded in the liposome vesicles. Theseresults raise new possibilities for fluorescent imaging ofbiological targets in vitro and in vivo. Furthermore, thefunctionalization techniques for the liposomes and thepossibility of using them as carriers for hydrophobic andhydrophilic drugs offer new opportunities for the simultaneousdelivery of therapeutic and diagnostic agents in tissues(theranostic system).The simultaneous incorporation in the bilayer of an integral

membrane protein was demonstrated for the first time for theRhodobacter sphaeroides RC chosen as model protein. Experi-ments have shown that the developed proteo−NC−liposomehybrid assemblies are stable and can preserve the structural andfunctional integrity of both biologic and inorganic componentswithout reciprocal destructive effects. This work will certainlyhelp to bridge the gap between the knowledge of NC−solubleprotein interaction and NC−integral membrane proteininteraction. Considering that hydrophobic membrane proteinsrepresent 50% by weight of the outer membrane of the cell andthat they play an important role in various diseases, this result iscertainly interesting for biomedical applications.

The proposed method is indeed general and can be, inprinciple, extended to any nanoparticle (metallic, semi-conductor, magnetic) with similar size and surface chemistry,phospholipid, and transmembrane protein.

■ ASSOCIATED CONTENT*S Supporting InformationDetails on NC capping exchange procedure, with FT-IRspectra, time-resolved PL decay curves, and TEM images;details on charge recombination kinetics of RC; cryo-EM imageof NC−liposomes at high NC/lipid ratio; digital picture of vialscontaining several NC−liposome suspensions under visible andUV illumination; confocal fluorescence microscopy image ofNC−liposome suspension; hydrodynamic diameter of NC−liposomes of various formulations and with variable startingdetergent in mixed micelles and overall lipid−NC ratio; detailson extraction of NCs from liposomes to hexane with calculationof NC local concentration in the bilayer and measurements ofNC QY in the liposome and after extraction in hexane; time-resolved PL decay curves from native NCs in hexane, NC−liposomes in water, and NCs extracted from liposomes andrecovered in hexane; evaluation of NC−liposome stability;FRET calculations; size distribution obtained by DLS for emptyvesicles, RC−liposomes, NC−liposomes, and NC−RC−lip-osomes. This material is available free of charge via the Internetat http://pubs.acs.org.

■ AUTHOR INFORMATIONCorresponding Authors*E-mail: [email protected].*E-mail: [email protected] authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis work was supported by the financial contribution of PRIN2010-2011 “Organizzazione Funzionale a Livello Nanoscopicodi (Bio)Molecole e Ibridi per Applicazioni nel Campo dellaSensoristica, della Medicina e delle Biotecnologie”. The authorsthank Lipoid GmbH for providing the phospholipid PC.

■ REFERENCES(1) Alivisatos, A. P. Semiconductor Clusters, Nanocrystals, andQuantum Dots. Science 1996, 271 (5251), 933−937.(2) Murray, C. B.; Norris, D. J.; Bawendi, M. G. Synthesis andCharacterization of Nearly Monodisperse CdE (E = Sulfur, Selenium,

Figure 8. (A) Fluorescence lifetime measurements relative to CdSe@ZnS NCs in NC−liposome and RC−NC−liposome hybrid assemblies (λex =375 nm, λem = 600 nm). (B) Absorption spectra of RC−NC hybrid vesicles for increasing amounts of RCs. In the inset are reported thecorresponding PL spectra (λex = 400 nm).

Langmuir Article

dx.doi.org/10.1021/la404160b | Langmuir 2014, 30, 1599−16081606

Tellurium) Semiconductor Nanocrystallites. J. Am. Chem. Soc. 1993,115 (19), 8706−8715.(3) Klostranec, J. M.; Chan, W. C. W. Quantum Dots in Biologicaland Biomedical Research: Recent Progress and Present Challenges.Adv. Mater. 2006, 18 (15), 1953−1964.(4) Pelaz, B.; Jaber, S.; de Aberasturi, D. J.; Wulf, V.; Aida, T.; de laFuente, J. M.; Feldmann, J.; Gaub, H. E.; Josephson, L.; Kagan, C. R.;Kotov, N. A.; Liz-Marzan, L. M.; Mattoussi, H.; Mulvaney, P.; Murray,C. B.; Rogach, A. L.; Weiss, P. S.; Willner, I.; Parak, W. J. The State ofNanoparticle-Based Nanoscience and Biotechnology: Progress, Prom-ises, and Challenges. ACS Nano 2012, 6 (10), 8468−8483.(5) Biju, V.; Itoh, T.; Ishikawa, M. Delivering Quantum Dots toCells: Bioconjugated Quantum Dots for Targeted and NonspecificExtracellular and Intracellular Imaging. Chem. Soc. Rev. 2010, 39 (8),3031−3056.(6) de la Fuente, J. M.; Berry, C. C.; Riehle, M. O.; Curtis, A. S. G.Nanoparticle Targeting at Cells. Langmuir 2006, 22 (7), 3286−3293.(7) Biju, V.; Mundayoor, S.; Omkumar, R. V.; Anas, A.; Ishikawa, M.Bioconjugated Quantum Dots for Cancer Research: Present Status,Prospects and Remaining Issues. Biotechnol. Adv. 2009, 28 (2), 199−213.(8) Tian, Q.; Hu, J.; Zhu, Y.; Zou, R.; Chen, Z.; Yang, S.; Li, R.; Su,Q.; Han, Y.; Liu, X. Sub-10 nm Fe3O4@Cu2−xS Core−ShellNanoparticles for Dual-Modal Imaging and Photothermal Therapy.J. Am. Chem. Soc. 2013, 135 (23), 8571−8577.(9) Comparelli, R.; Curri, M. L.; Cozzoli, P. D.; Striccoli, M.Nanomaterials for Biosensors. In Nanotechnologies for the Life Sciences;Kumar, C. S. S. R., Ed.; Wiley-VCH: Weinheim, 2006; Vol. 8, pp 123−174.(10) Curri, M. L.; Comparelli, R.; Striccoli, M.; Agostiano, A.Emerging Methods for Fabricating Functional Structures by Patterningand Assembling Engineered Nanocrystals. Phys. Chem. Chem. Phys.2010, 12, 11197−11207.(11) Zhao, Q.; Rong, X. L.; Ma, H. B.; Tao, G. H. Aqueous Synthesisof CdSe and CdSe/CdS Quantum Dots with ControllableIntroduction of Se and S sources. J. Mater. Sci. 2013, 48 (5), 2135−2141.(12) Yin, Y.; Alivisatos, A. P. Colloidal Nanocrystal Synthesis and theOrganic-Inorganic Interface. Nature 2005, 437 (7059), 664−670.(13) Depalo, N.; Carrieri, P.; Comparelli, R.; Striccoli, M.; Agostiano,A.; Bertinetti, L.; Innocenti, C.; Sangregorio, C.; Curri, M. L.Biofunctionalization of Anisotropic Nanocrystalline Semiconductor−Magnetic Heterostructures. Langmuir 2011, 27 (11), 6962−6970.(14) Yu, W. W.; Chang, E.; Drezek, R.; Colvin, V. L. Water-SolubleQuantum Dots for Biomedical Applications. Biochem. Biophys. Res.Commun. 2006, 348, 781−786.(15) Pellegrino, T.; Manna, L.; Kudera, S.; Koktysh, D.; Rogach, A.L.; Natile, G.; Parak, W. J. Hydrophobic Nanocrystals Coated with anAmphiphilic Polymer Shell: A General Route to Water SolubleNanocrystals. Nano Lett. 2004, 4 (4), 703−707.(16) Dubertret, B.; Skourides, P.; Norris, D. J.; Noireaux, V.;Brivanlou, A. H.; Libchaber, A. In Vivo Imaging of Quantum DotsEncapsulated in Phospholipid Micelles. Science 2002, 298, 1759−1762.(17) Di Marco, M.; Shamsuddin, S.; Razak, K. A.; Aziz, A. A.; Devaux,C.; Borghi, E.; Levy, L.; Sadun, C. Overview of the Main MethodsUsed to Combine Proteins With Nanosystems: Absorption,Bioconjugation, and Encapsulation. Int. J. Nanomed. 2010, 5, 37−49.(18) Rasch, M. R.; Rossinyol, E.; Hueso, J. L.; Goodfellow, B. W.;Arbiol, J.; Korgel, B. A. Hydrophobic Gold Nanoparticle Self-Assemblywith Phosphatidylcholine Lipid: Membrane-Loaded and JanusVesicles. Nano Lett. 2010, 10 (9), 3733−3739.(19) Bothun, G. D. Hydrophobic Silver Nanoparticles Trapped inLipid Bilayers: Size Distribution, Bilayer Phase Behavior, and OpticalProperties. J. Nanobiotechnol. 2008, 6 (13).(20) Roger, R. C. Introduction. In Liposomes: A Pratical Approach,New., R. R. C., Ed.; IRL Press at Oxford University Press: New York,1992; Chapter 1, p 301.

(21) Immordino, M. L.; Dosio, F.; Cattel, L. Stealth Liposomes:Review of the Basic Science, Rationale, and Clinical Applications,Existing and Potential. Int. J. Nanomed. 2006, 1 (3), 297−315.(22) Malam, Y.; Loizidou, M.; Seifalian, A. M. Liposomes andNanoparticles: Nanosized Vehicles for Drug Delivery in Cancer.Trends Pharmacol. Sci. 2009, 30 (11), 592−599.(23) Park, S.-H.; Oh, S.-G.; Suh, K.-D.; Han, S.-H.; Chung, D. J.;Mun, J.-Y.; Han, S.-S.; Kim, J.-W. Control Over Micro-Fluidity ofLiposomal Membranes by Hybridizing Metal Nanoparticles. ColloidsSurf., B 2009, 70 (1), 108−113.(24) Feng, L.; Kong, X.; Chao, K.; Sun, Y.; Zeng, Q.; Zhang, Y.Efficient Phase Transfer of Hydrophobic CdSe Quantum Dots: FromNonpolar Organic Solvent to Biocompatible Water Buffer. Mater.Chem. Phys. 2005, 93 (2−3), 310−313.(25) Liu, J. W.; Jiang, X. M.; Ashley, C.; Brinker, C. J. ElectrostaticallyMediated Liposome Fusion and Lipid Exchange with a Nanoparticle-Supported Bilayer for Control of Surface Charge, Drug Containment,and Delivery. J. Am. Chem. Soc. 2009, 131 (22), 7567−7569.(26) Sangregorio, C.; Wiemann, J. K.; O’Connor, C. J.; Rosenzweig,Z. A New Method for the Synthesis of Magnetoliposomes. J. Appl.Phys. 1999, 85 (8), 5699−5701.(27) Korgel, B. A.; Monbouquette, H. G. Controlled Synthesis ofMixed Core and Layered (Zn,Cd)S and (Hg,Cd)S Nanocrystalswithin Phosphatidylcholine Vesicles. Langmuir 2000, 16 (8), 3588−3594.(28) Binder, W. H.; Sachsenhofer, R.; Farnik, D.; Blaas, D. Guidingthe Location of Nanoparticles into Vesicular Structures: AMorphological Study. Phys. Chem. Chem. Phys. 2007, 9 (48), 6435−6441.(29) Al-Jamal, W. T.; Al-Jamal, K. T.; Bomans, P. H.; Frederik, P. M.;Kostarelos, K. Functionalized-Quantum-Dot-Liposome Hybrids AsMultimodal Nanoparticles for Cancer. Small 2008, 4 (9), 1406−1415.(30) Chen, C.-S.; Yao, J.; Durst, R. Liposome encapsulation offluorescent nanoparticles: Quantum dots and silica nanoparticles. J.Nanopart. Res. 2006, 8 (6), 1033−1038.(31) Park, S. H.; Oh, S. G.; Mun, J. Y.; Han, S. S. Loading of GoldNanoparticles Inside the DPPC Bilayers of Liposome and TheirEffects on Membrane Fluidities. Coll. Surf., B 2006, 48 (2), 112−118.(32) Al-Jamal, W. T.; Al-Jamal, K. T.; Tian, B.; Lacerda, L.; Bomans,P. H.; Frederik, P. M.; Kostarelos, K. Lipid−Quantum Dot BilayerVesicles Enhance Tumor Cell Uptake and Retention in Vitro and inVivo. ACS Nano 2008, 2 (3), 408−418.(33) Gopalakrishnan, G.; Danelon, C.; Izewska, P.; Prummer, M.;Bolinger, P.-Y.; Geissbuhler, I.; Demurtas, D.; Dubochet, J.; Vogel, H.Multifunctional Lipid/Quantum Dot Hybrid Nanocontainers forControlled Targeting of Live Cells. Angew. Chem., Int. Ed. 2006, 45(33), 5478−5483.(34) Liu, M.; Gan, L.; Chen, L.; Xu, Z.; Zhu, D.; Hao, Z.; Chen, L.Supramolecular Core−Shell Nanosilica@Liposome Nanocapsules forDrug Delivery. Langmuir 2012, 28 (29), 10725−10732.(35) Seth, A.; Bealle, G.; Santanach-Carreras, E.; Abou-Hassan, A.;Menager, C. Design of Vesicles Using Capillary Microfluidic Devices:From Magnetic to Multifunctional Vesicles. Adv. Mater. 2012, 24 (26),3544−3548.(36) Shan, G.-Y.; Li, D.; Feng, L.-Y.; Kong, X.-G.; Liu, Y.-C.; Bai, Y.-B.; Li, T.-J.; Sun, J.-Z. Encapsulation of CdSe/ZnSe Quantum Dots byLiposome Complexes. Chin. J. Chem. 2005, 23 (12), 1688−1692.(37) Kim, S. J.; Wi, H. S.; Kim, K.; Lee, K.; Kim, S. M.; Yang, H. S.;Pak, H. K. Encapsulation of CdSe Nanoparticles Inside LiposomeSuspended in Aqueous Solution. J. Korean Phys. Soc. 2006, 49, S684−S687.(38) Walde, P. Preparation of Vesicles (Liposomes). In Encyclopediaof Nanoscience and Nanotechnology; Nalwa, H. S., Ed.; AmericanScientific Publishers: Stevenson Ranch, CA, 2003; Vol. X, pp 1−37.(39) Milano, F.; Italiano, F.; Agostiano, A.; Trotta, M. Character-isation of RC-Proteoliposomes at Different RC/Lipid Ratios. Photo-synth. Res. 2009, 100, 107−112.(40) Milano, F.; Tangorra, R. R.; Hassan Omar, O.; Ragni, R.;Operamolla, A.; Agostiano, A.; Farinola, G. M.; Trotta, M. Enhancing

Langmuir Article

dx.doi.org/10.1021/la404160b | Langmuir 2014, 30, 1599−16081607

the Light Harvesting Capability of a Photosynthetic Reaction Centerby a Tailored Molecular Fluorophore. Angew. Chem., Int. Ed. 2012, 51,11019−11023.(41) De Leo, V.; Catucci, L.; Ventrella, A.; Milano, F.; Agostiano, A.;Corcelli, A. Cardiolipin Increases in Chromatophores Isolated fromRhodobacter sphaeroides After Osmotic Stress: Structural and Func-tional Roles. J. Lipid Res. 2009, 50, 256−264.(42) Mekis, I.; Talapin, D. V.; Kornowski, A.; Haase, M.; Weller, H.One-Pot Synthesis of Highly Luminescent CdSe/CdS Core−ShellNanocrystals via Organometallic and “Greener” Chemical Approaches.J. Phys. Chem. B 2003, 107 (30), 7454−7462.(43) Catucci, L.; De Leo, V.; Milano, F.; Giotta, L.; Vitale, R.;Agostiano, A.; Corcelli, A. Oxidoreductase Activity of Chromatophoresand Purified Cytochrome bc1 Complex from Rhodobacter sphaeroides:A Possible Role of Cardiolipin. J. Bioenerg. Biomembr. 2012, 44 (4),487−493.(44) Agostiano, A.; Milano, F.; Trotta, M. Investigation on theDetergent Role in the Function of Secondary Quinone in BacterialReaction Centers. Eur. J. Biochem. 1999, 262 (2), 358−364.(45) Lewis, B. A.; Engelman, D. M. Lipid Bilayer Thickness VariesLinearly with Acyl Chain Length in Fluid PhosphatidylcholineVesicles. J. Mol. Biol. 1983, 166 (2), 211−217.(46) Aguilera-Sigalat, J.; Rocton, S.; Sanchez-Royo, J. F.; Galian, R.E.; Perez-Prieto, J. Highly Fluorescent and Photostable Organic- andWater-Soluble CdSe/ZnS Core-Shell Quantum Dots Capped withThiols. RSC Adv. 2012, 2 (4), 1632−1638.(47) Jiskoot, W.; Teerlink, T.; Beuvery, E. C.; Crommelin, D. A.Preparation of Liposomes Via Detergent Removal From MixedMicelles by Dilution. Pharm. Weekbl. 1986, 8 (5), 259−265.(48) Wi, H. S.; Lee, K.; Pak, H. K. Interfacial Energy Considerationin the Organization of a Quantum Dot−Lipid Mixed System. J. Phys.:Condens. Matter 2008, 20, 494211.(49) Pinaud, F.; King, D.; Moore, H.-P.; Weiss, S. Bioactivation andCell Targeting of Semiconductor CdSe/ZnS Nanocrystals withPhytochelatin-Related Peptides. J. Am. Chem. Soc. 2004, 126 (19),6115−6123.(50) Erogbogbo, F.; Yong, K.-T.; Roy, I.; Xu, G.; Prasad, P. N.;Swihart, M. T. Biocompatible Luminescent Silicon Quantum Dots forImaging of Cancer Cells. ACS Nano 2008, 2 (5), 873−878.(51) Geissbuehler, I.; Hovius, R.; Martinez, K. L.; Adrian, M.;Thampi, K. R.; Vogel, H. Lipid-Coated Nanocrystals as Multi-functionalized Luminescent Scaffolds for Supramolecular BiologicalAssemblies. Angew. Chem., Int. Ed. 2005, 44, 1388−1392.(52) Zimmer, J. P.; Kim, S. W.; Ohnishi, S.; Tanaka, E.; Frangioni, J.V.; Bawendi, M. G. Size Series of Small Indium Arsenide−ZincSelenide Core−Shell Nanocrystals and Their Application to in VivoImaging. J. Am. Chem. Soc. 2006, 128, 2526−2527.(53) Ollivon, M.; Lesieur, S.; Grabielle-Madelmont, C.; Paternostre,M. t. Vesicle Reconstitution from Lipid−Detergent Mixed Micelles.Biochim. Biophys. Acta, Biomembr. 2000, 1508 (1−2), 34−50.(54) Allen, J. P.; Williams, J. C. Photosynthetic Reaction Centers.FEBS Lett. 1998, 438 (1−2), 5−9.(55) Nagy, L.; Milano, F.; Dorogi, M.; Agostiano, A.; Laczko, G.;Szebenyi, K.; Varo, G.; Trotta, M.; Maroti, P. Protein/LipidInteraction in the Bacterial Photosynthetic Reaction Center:Phosphatidylcholine and Phosphatidylglycerol Modify the Free EnergyLevels of the Quinones. Biochemistry 2004, 43 (40), 12913−23.(56) Nabiev, I.; Rakovich, A.; Sukhanova, A.; Lukashev, E.;Zagidullin, V.; Pachenko, V.; Rakovich, Y. P.; Donegan, J. F.; Rubin,A. B.; Govorov, A. O. Fluorescent Quantum Dots as ArtificialAntennas for Enhanced Light Harvesting and Energy Transfer toPhotosynthetic Reaction Centers. Angew. Chem., Int. Ed. 2010, 49(40), 7217−7221.(57) Rasch, M. R.; Yu, Y.; Bosoy, C.; Goodfellow, B. W.; Korgel, B.A. Chloroform-Enhanced Incorporation of Hydrophobic Gold Nano-crystals into Dioleoylphosphatidylcholine (DOPC) Vesicle Mem-branes. Langmuir 2012, 28 (36), 12971−12981.

Langmuir Article

dx.doi.org/10.1021/la404160b | Langmuir 2014, 30, 1599−16081608