Soft Matter View Article Online
Published on 17 June 2014. Downloaded by University of Michigan Library on 30/10/2014 12:19:23.
PAPER
Cite this: Soft Matter, 2014, 10, 6451
View Journal | View Issue
Formation, stability, and pH sensitivity of freefloating, giant unilamellar vesicles using palmitic acid–cholesterol mixtures† Nicolas Cottenye,a Gustavo Carbajal,a Zhong-Kai Cui,a Philippe Dauphin Ducharme,b Janine Mauzerollb and Michel Lafleur*a Despite the fact that palmitic acid (PA) and cholesterol (Chol) do not form fluid bilayers once hydrated individually, giant unilamellar vesicles (GUVs) were formed from a mixture of palmitic acid and cholesterol, 30/70 mol/mol. These free-floating GUVs were stable over weeks, did not aggregate and were shown to be highly stable in alkaline pH compared to conventional phospholipid-based GUVs.
Received 23rd April 2014 Accepted 16th June 2014
Acidic pH-triggered payload release from the GUVs was associated with the protonation state of palmitic acid that dictated the mixing lipid properties, thus affecting the stability of the fluid lamellar phase. The
DOI: 10.1039/c4sm00883a
successful formation of PA–Chol GUVs reveals the possibility to create monoalkylated amphiphile-based
www.rsc.org/softmatter
GUVs with distinct pH stability/sensitivity.
1. Introduction Giant unilamellar vesicles (GUVs) are liposomes with micronsize diameters. These self-assembled entities, whose size is on the order of magnitude of biological cells, are valuable biomimetic models used to study the physicochemical properties of membranes, transmembrane trafficking and protein activity.1 Moreover, GUVs, because of their distinct characteristics including a large entrapped volume, have been used in several technological elds including as reactors for chemical and/or enzymatic reactions2 and mRNA synthesis,3 as systems for vesicle self-reproduction,4 and as in vivo drug carriers.5 Their size allows their observation with an optical microscope and, as a consequence, GUVs are very appropriate systems for examining directly vesicle shape deformations6 and for studying micromechanical properties of bilayers.7 GUVs are predominantly formed from phospholipid mixtures.8 Because of the limited chemical stability of
a
Department of Chemistry, Center for Self-Assembled Chemical Structures, Universit´e de Montr´eal, PO Box 6128, Station Downtown, Montr´eal, Qu´ebec, H3C 3J7, Canada. E-mail: michel.la[email protected]; Fax: +1-5143437586; Tel: +15143403205 b
Laboratory for Electrochemical Reactive Imaging and Detection for Biological Systems, Department of Chemistry, McGill University, 801 Sherbrooke Street West, Montr´eal, Qu´ebec, H3A 0B8, Canada † Electronic supplementary information (ESI) available: Fig. S-1: control indicating that the external acidic pH does not affect the uorescence intensity of the entrapped calcein: uorescence microscopy images of POPC/POPG/Chol GUVs at pH 8.4 and 3.4. Fig. S-2: images, recorded during the titration of GUVs from pH 8.4 to 3.4, representing only the uorescence of apolar Nile red. These images are paired with those obtained from the green channel presented in Fig. 5 of the manuscript. See DOI: 10.1039/c4sm00883a
This journal is © The Royal Society of Chemistry 2014
phospholipids under certain conditions (such as under extreme pH or in the presence of phospholipases), phospholipid-based GUVs display limited robustness and signicant permeability that are not always suitable for some applications. Few alternative materials have been examined for GUV formation in order to overcome phospholipid limitations. For example, few studies report GUV formation using block copolymers.9–11 The formation of GUVs with synthetic detergent (monoalkylated amphiphiles) remains largely unexplored. We know of only four other studies reporting synthetic detergent-based GUVs. GUVs were obtained from a mixture of a fatty acid and its salt (namely oleic acid–sodium oleate mixture)12 as well as from fatty acids mixed in equimolar portion with their corresponding fatty alcohol.13 In these two studies, the preparation method based on slow hydration led a wide diversity of self-assembled structures and the produced GUVs represented only a fraction of these (an estimated proportion of 40–50% being GUVs in ref. 12, and was not dened in the other study). In addition, the fatty acid/fatty soap GUVs appeared to be extremely sensitive to pH and they only existed over a limited pH range. GUVs from fatty acid and its salt were also formed in the presence of squalane at the vicinity of a glass surface,14 yet again with limited stability. Finally, GUVs formed by an ammonium amphiphile containing a Schiff base were reported as transient structures observed strictly during the transition from small vesicles to oil droplets caused by the chemical decomposition of the amphiphile.15 Therefore, an effective approach to create free-oating and stable GUVs from the non-phospholipid single-chain amphiphile is still to be ascertained. The formation of stable GUVs with monoalkylated amphiphiles is particularly challenging because, rst, of their critical aggregation concentrations (e.g. 4 mM for PA16), which are considerably higher than those of
Soft Matter, 2014, 10, 6451–6456 | 6451
View Article Online
Published on 17 June 2014. Downloaded by University of Michigan Library on 30/10/2014 12:19:23.
Soft Matter
conventional phospholipids (e.g. 0.46 nM for dipalmitoylphosphatidylcholine17) and, second, they typically self-aggregate to form micelles, not uid lamellar phases. Herein, we report the formation of stable, free-oating GUVs self-assembled from a mixture of cholesterol (Chol), and fatty acid salt, namely deprotonated palmitic acid (PA). It has been shown that cholesterol combined with some single-chain amphiphiles leads to the formation of uid bilayers in the liquid ordered (lo) phase, despite the fact that these components do not form uid lamellar phases once hydrated individually (for a review, see ref. 18). It is proposed that the singlechain amphiphile acts as a gap ller between cholesterol molecules and ensures bilayer uidity while maintaining tight lipid packing. Large unilamellar vesicles (LUVs) were successfully obtained from such uid bilayers.19–22 The distinct chemical composition of these systems is reected in the distinct properties of the resulting liposomes, including a lower permeability compared to their phospholipid-based analogues, a robust protection to the encapsulated materials, and an enhanced chemical stability.18–20,22–24 GUVs formed with PA– Chol mixtures are expected to display similar properties, associated with the high cholesterol content of the assemblies. Therefore the successful preparation of GUVs from a mixture that includes only a single-chain amphiphile and cholesterol creates a new class of GUVs with improved encapsulation performance. In addition, craing functionalities in monoalkylated amphiphile/sterol self-assembled systems is relatively straightforward. For example, bilayers can be made pH responsive by the use of PA. An hydrated mixture of PA and cholesterol, in a molar ratio of 30/70, forms a lo lamellar phase at pH greater than 7.4 as a result of the fatty acid deprotonation.23 The protonation of PA leads to a phase separation between the constituents, and induces the formation of solid particles of PA and cholesterol. This phase transition causes the rupture of the uid bilayer accompanied by the release of the material entrapped inside LUVs.20 This property can be exploited to cra a pH-triggered release from GUVs, a feature that has been used to achieve controlled drug release in specic acidic locations such as cancer tumours and endosomal compartments.25 We selected PA–Chol mixtures as a proof of concept for making GUVs based on single-chain amphiphile–Chol mixtures because the resulting self-assemblies are expected to additionally be pHresponsive. There are few GUV preparation methods reported in the literature including direct lipid lm hydration, alternate current, and double emulsion methods.8 The PA–Chol GUVs were prepared using the double emulsion droplet method,26–29 which was selected for its simplicity and its high exibility in terms of usable materials. Briey, a water-in-oil (chloroform)in-water double emulsion was formed, with monolayers of amphiphiles at the interfaces, stabilizing the emulsion. Upon complete evaporation of the organic phase, the apolar parts of the amphiphile monolayers entered in contact to form bilayers and to give rise to GUVs.30 It was proposed that the size of the internal water droplets dened the size of the resulting GUVs.
6452 | Soft Matter, 2014, 10, 6451–6456
Paper
2.
Results and discussion
In order to establish the formation of GUVs, epiuorescence imaging was carried out. Two uorescent probes were used: Nile red, a hydrophobic probe that was inserted in the hydrophobic core of lipid bilayers, and calcein, a hydrophilic probe that was trapped in the GUV internal pool. This approach allowed the observation of both the walls of the GUVs and their entrapped content. The GUV stability at alkaline pH was investigated by assessing its permeability to a uorophore (sulforhodamine B (SRB)) and compared to conventional phospholipid-based GUVs made of 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC)–Chol (60/40 mixture) as control. The pH-sensitivity of the GUVs was investigated by light scattering variations and by uorescence microscopy. The uorescence micrographs presented in Fig. 1 demonstrate the successful self-assembly of GUVs from a PA–Chol 30/ 70 mixture using the lipid-stabilized double-emulsion method. The GUVs had a spherical morphology with diameters mainly ranging from 1 to 5 mm. Occasionally, GUVs had diameters larger than 10 mm. The diameter distribution is presented in Fig. 2. Overall, the average diameters of PA–Chol 30/70 GUVs were 3.3 2.1 mm as established from uorescence imaging (n ¼ 154), and 3.4 1.9 mm as measured from dynamic light scattering. Multilamellar vesicles and more complex particles were also detected but they represented at most 10% of the observed self-assemblies. PA– Chol GUVs were highly stable and could be stored at room temperature for a month without noticeable change in morphology or size. The enhanced stability is consistent with the ndings obtained with PA–Chol LUVs, which also exhibited a stable behaviour over extended periods of time.24 We have assessed the permeability of PA–Chol GUVs by measuring the passive release of SRB (Fig. 3). PA–Chol GUVs remained essentially impermeable to SRB at pH 12 as less than 10% of entrapped uorophore had leaked out aer 4.5 days. This very limited membrane permeability is consistent with that previously reported for PA–Chol LUVs.20,31 In addition, because
Fluorescence microscopy images of PA–Chol GUVs at pH 8.4. The red channel corresponds to membranes labelled with Nile red whereas the green channel represents calcein entrapped in the inner aqueous pool of the GUVs. The image on the left is obtained by the superposition of the two images (right, top and bottom). The scale bars correspond to 10 mm.
Fig. 1
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 17 June 2014. Downloaded by University of Michigan Library on 30/10/2014 12:19:23.
Paper
Soft Matter
Fig. 2 Size distribution of PA–Chol GUVs at pH 8.4, sorted by a 0.5mm steps, as obtained from the measurements from fluorescence microscopy pictures (n ¼ 154). The average diameter was 3.3 2.1 mm. Fig. 4 Destabilization of PA–Chol GUVs induced by a decrease of pH, as observed by light scattering measurements. The dashed line is only a guide to the eyes. A strong decrease in scattered light is observed below pH 6, corresponding to GUV disruption, and sedimentation of the resulting solid particles at acidic pH.
PA–Chol GUV permeability to SRB at pH 12 is compared with that obtained for POPC–Chol GUVs. PA–Chol GUVs display under these alkaline conditions a considerable enhanced chemical stability compared to POPC–Chol ones.
Fig. 3
of their chemical composition, PA–Chol GUVs exhibited an enhanced stability compared to conventional phospholipid formulation. For example, POPC–Chol GUVs, which also exist in the lo phase, were leakier under these alkaline conditions, leading to the release of 50% of trapped SRB over 4.5 days. This behaviour was associated with the alkaline hydrolysis of the ester groups linking the acyl chains to the glycerol backbone of POPC, a chemical reaction that disrupted the bilayer integrity. This difference illustrates very well the good chemical stability of PA–Chol GUVs. The sensitivity of PA–Chol GUVs to acidic pH was rst assessed using light scattering (Fig. 4) by titrating GUVs prepared at pH 8.4 with HCl, down to an external pH of 3.4. Between pH 8.4 and 6.5, a considerable scattering was observed due to the presence of numerous GUVs in the suspension. Between pH 6.5 and 5.0, a large decrease of the scattered light intensity was observed. The pH range associated with this
This journal is © The Royal Society of Chemistry 2014
transformation is consistent with the apparent pKa of PA inserted in uid bilayers, that was estimated to be between 6.4 and 8.2.20, 32–33 In these studies, PA–Chol uid bilayers were shown to become unstable when a large fraction of the fatty acid was protonated, as the electrostatic intermolecular interactions did not favour lipid mixing. Under these conditions, there was the formation of solid particles of cholesterol and of PA.20 On the basis of these results, we interpreted the reduced light scattering intensity observed at low pH by the destabilization of the GUVs that underwent a transition towards solid cholesterol and solid palmitic acid. These solid particles likely sedimented in the measurement cell, reducing, as a consequence, the light scattering power. To corroborate that the pH-triggered disruption of the GUVs observed from the light scattering experiments effectively leads to the formation of solid lipid particles and a triggered release of the GUV payload, epiuorescence studies of PA–Chol GUVs were performed (Fig. 5). Nile-red labeled GUVs were initially formed in a sucrose- and a calcein-containing buffer, and then sedimented at the glass-slide interface in a glucose-containing buffer. The pH of both the internal and external buffers was 8.4. The pH was subsequently modied by sequential additions of diluted HCl. Despite the fact that the HCl aliquots were added as gently as possible, the pH variations created convective ows that were sufficient to displace the GUVs, preventing a proper tracking of specic vesicles. Nevertheless, pH-triggered calcein release could be clearly observed during titration experiments. Down to pH 6.4, no signicant morphological variations, swelling, or uorophore leakage could be detected, indicating unaltered GUV permeability. Below pH 6.4, the GUV density, as observed by their internal content labeled with calcein, decreased. It should be pointed out that calcein uorescence is known to be considerably decreased at pH below 5;34 however the addition of HCl essentially modied the external pH of the
Soft Matter, 2014, 10, 6451–6456 | 6453
View Article Online
Soft Matter
Paper
is believed that the Nile red-labeled fragments observed at low pH corresponded to PA and/or cholesterol solid particles, which would have essentially a hydrophobic core. At pH 3.4, GUVs were no longer observed. These results clearly indicated that the GUVs were pH-sensitive and became unstable due to pH trigger.
3.
Experimental
Published on 17 June 2014. Downloaded by University of Michigan Library on 30/10/2014 12:19:23.
3.1. Materials Palmitic acid (PA) (99%), cholesterol (Chol) (>99%), Nile red (>98%), tris(hydroxymethyl)aminomethane (TRIS) (99%), 2-[Nmorpholino]ethanesulfonic acid (MES) (>99%), diethyl ether (>99%), glucose (ACS reagent) and sucrose (ACS reagent) were supplied by Sigma Chemical Co. (St. Louis, MO, USA). KCl (ACS reagent) and chloroform (>99%) were obtained from Biopharm and Anachemia (Burlignton, ON, CA), respectively. Calcein (high purity) and SRB were obtained from Invitrogen Corporation (Burlignton, ON, CA). POPC (>99%) was purchased from Avanti Polar Lipids, Inc. (Birmingham, AL, USA). Milli-Q water with a resistivity of 18.2 MU cm was used for the buffer preparation. 3.2. GUV formation
Fig. 5 Effect of pH on the GUV stability monitored by fluorescence microscopy. Larger images at extreme pH values: 8.4 (top) and 3.4 (bottom) display the overlaid calcein and Nile red fluorescence signals. The smaller images (middle), recorded during the titration of GUVs from pH 8.4 to 3.4, present only the fluorescence of calcein from the inner pool of the GUVs. The scale bars represent 10 mm. Vesicle burst was observed from pH 6.4. At pH 3.4, GUVs were no longer observed.
suspension and should have a very limited impact on GUV internal pH. In fact, we showed, as a control, that the uorescence of calcein trapped in phospholipid-based and pH-insensitive GUVs was not signicantly affected over the pH range from 3.4 to 8.4 (Fig. S1 of the ESI†). The disappearance of calcein labeled GUVs upon pH decrease was concomitant with an increased number of particles completely labeled with Nile red, without inner pool (see Fig. 5 and S2 of the ESI†). It has been established that the PA–Chol system undergoes a pH triggered phase transition from a uid lamellar phase to solid PA and solid cholesterol when the pH of a suspension is decreased.23 It
6454 | Soft Matter, 2014, 10, 6451–6456
PA–Chol (30/70) mixtures were obtained by solubilizing weighed amounts of each component in a benzene–methanol mixture (75/25 (v/v)). POPC–Chol (60/40) mixtures were obtained by solubilizing weighed amounts of each component in a benzene–methanol mixture (90/10 (v/v)). The solutions were then freeze-dried for at least 16 h. Subsequently, the resulting powder was dissolved in chloroform (10 mg mL1). For the labeling of the GUV membrane, Nile red was added to this solution (50 mg mL1). In order to make a double emulsion, two suspensions were prepared. In the rst vial, 320 mL of the lipid solution, 1 mL of chloroform, and 1 mL of a sucrose-containing buffer (TRIS 5 mM; MES 5 mM; sucrose 215 mM, pH 8.4) were mixed. For the labeling of the GUV internal aqueous compartment, calcein (0.1 mM) was added to this buffer. The second vial contained 320 mL of the lipid solution, 0.5 mL of diethyl ether and 2.5 mL of the sucrose-containing buffer. The rst and second vials were sonicated for 45 and 15 s respectively. Their contents were mixed and sonicated for an additional 10 s period in order to form the water-in-oil-in-water double emulsion. The sample was transferred to a round bottom ask and heated under an argon atmosphere between 65 and 68 C for 45 min to allow the evaporation of the organic solvents. In order to isolate the GUVs, the suspension was centrifuged at 1500g for 30 min, the supernatant was removed, and the GUVs were resuspended in 5 mL of the sucrose-containing buffer. This procedure was repeated twice. 3.3. Fluorescence microscopy A 100 mL aliquot of a glucose-containing buffer (TRIS 5 mM; MES 5 mM; glucose 50 mM; KCl 100 mM, pH 8.3) was deposited on a glass slide and 20 mL of the GUV suspension were added. The glucose- and sucrose-containing buffers were iso-osmolar (250 mOsm L1) to prevent GUV collapse, shrinkage or bursting
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 17 June 2014. Downloaded by University of Michigan Library on 30/10/2014 12:19:23.
Paper
as a result of an osmotic stress. A 5 min period was allowed for the GUVs to sediment to the glass slide surface. In order to study the pH sensitivity of the GUVs, the pH of a sample prepared at pH 8.4 as described above was decreased by the successive addition of 2 mL aliquots of HCl, 0.05 M, added every 3 min. The pH was measured aer each HCl addition using a combination pH microelectrode (Microelectrodes Inc., Bedford, NH, USA). Aer 8 aliquots, the pH of the suspension reached 3.6 0.6. Fluorescence images of Nile red- and calcein-labeled GUVs were obtained using lexc/lem: 561/600–650 nm and 488/ 508–540 nm, respectively. Images were digitally processed using the FIJI/ImageJ image processing soware. Images of static GUVs were acquired using an upright uorescence microscope (Axioskop 2 Plus, Carl Zeiss, Inc.) equipped with a Qimaging cooled charge-coupled-device camera, a 100 Plan-Neo Zeiss objective, and the Image-Pro Plus acquisition soware whereas images assessing pH sensitivity of the GUVs were acquired using an inverted uorescence microscope (Olympus IX81) equipped with a 100 objective, and the Image-Pro Plus acquisition soware. 3.4. Light scattering measurement The light (633 nm) scattering was measured at an angle of 173 , using a Zetasizer NanoZS equipped with an MPT-2 Autotitrator (Malvern Instrument). Typically, the pH of a GUV suspension in the sucrose-containing buffer was varied from 8.4 to 3.0 with 0.5-unit steps. Autotitrations were carried out with HCl, 0.1 and 0.05 M. The titrations were performed in triplicates. 3.5. Stability assessment at alkaline pH PA–Chol (30/70) and POPC–Chol (60/40) GUVs were prepared as mentioned above except that the lipids were hydrated with a buffer containing TRIS (5 mM), SRB (80 mM), pH 8.4. The SRB concentration led to self-quenching of its uorescence.35 Free SRB was separated from SRB-loaded GUVs using Amicon Ultra10 centrifugal lter units (Milipore, Billerica, MA) spun at 1500g. The GUVs were harvested on the lter and resuspended in a buffer containing TRIS (5 mM), and NaCl (450 mM), pH 8.4. The osmolarity of the GUV internal and external buffers was carefully matched to ensure iso-osmolarity over the experiment course. The washing of the SRB-loaded GUVs was repeated until the SRB self-quenching factor was at least 0.9. The nal GUV concentration was approximately 1 mM of PA or POPC. To measure the stability of these GUVs at pH 12, the percentage of entrapped SRB was calculated, as previously reported by Phoeung et al.22 In this experiment an initial reference consisting of a sample kept at pH 8.4 diluted 10 times in TRIS (5 mM), NaCl (450 mM) buffer, pH 8.4 then 5 times in TRIS (50 mM), NaCl (370 mM) buffer, pH 8.4, was measured before (Ii) and aer (Ii + T) the addition of Triton X-100 (1 (v/v)% in the pH-8.4 buffer). Then 1 mL of the GUV initial suspension was diluted in 9 mL of an iso-osmolar buffer, pH 12, containing NaCl (350 mM); this was set as time ¼ 0. At different times, a 200 mL aliquot of GUVs in alkaline buffer was diluted in an isoosmolar buffer at pH 8.4 (TRIS 50 mM, NaCl 370 mM, pH 8.4) in order to avoid the inuence of pH on SRB uorescence. The This journal is © The Royal Society of Chemistry 2014
Soft Matter
nal lipid concentration was 20 mM of PA or POPC. Immediately aer the nal dilution, the uorescence intensity (If) of the samples was measured (lex/lem ¼ 565/585 nm). Subsequently, 20 mL of Triton X-100 (1 (v/v)% in the pH 8.4 buffer) was added to completely release the encapsulated SRB and the resulting uorescence intensity was measured (If + T) to normalize the subsequent measurements. The percentage of encapsulation for each time point of the series was calculated using eqn (1): ! IfþT If IiþT 100: % encapsulated SRB ¼ (1) ðIiþT Ii ÞIfþT
The reported average values and standard deviations (Fig. 2) were obtained from duplicate measurements of 2 parallel release experiments per batch on 2 independent GUV batches.
4. Conclusions Synthetic detergent-based GUVs were produced from a palmitic acid salt–cholesterol mixture. Despite the fact that PA, analogous to other synthetic detergents, has a propensity to form micelles and generally induces destabilization of lamellar structures, it is believed that the high cholesterol content in the lamellar phase led to a tight lipid packing that ensured the bilayer cohesion.23 These GUVs were stable at room temperature and displayed a pH-triggered release of their content, analogous to that previously observed for PA–Chol LUVs.31 Different monoalkylated amphiphile/sterol systems have been reported to form uid bilayers and LUVs could be derived from these mixtures.18,21,24,36,37 Several of their properties, including their overall charge, their response to stimuli such as pH and light, and their permeability, can be modulated in a rational and predictable manner by the selection of the constituents. The successful formation of the PA–Chol GUVs presented here as a proof of concept indicates the possibility of developing a novel and versatile platform of self-assembled free-oating GUVs prepared from non-phospholipid molecules that could present distinct advantages, including good chemical stability, low permeability, tunable surface charge, tunable sensing properties, and preparation from inexpensive and biocompatible components.
Acknowledgements This work was supported by the Universit´ e de Montr´ eal, the Natural Sciences and Engineering Research Council of Canada and by the Fonds Qu´ eb´ ecois de la Recherche sur la Nature et les Technologies through its Strategic Cluster program.
Notes and references 1 K. Kitatokarczyk, J. Grumelard, T. Haefele and W. Meier, Polymer, 2005, 46, 3540–3563. 2 K. Kato, P. Walde, H. Mitsui and N. Higashi, Biotechnol. Bioeng., 2003, 84, 415–423.
Soft Matter, 2014, 10, 6451–6456 | 6455
View Article Online
Published on 17 June 2014. Downloaded by University of Michigan Library on 30/10/2014 12:19:23.
Soft Matter
3 A. Fischer, A. Franco and T. Oberholzer, ChemBioChem, 2002, 3, 409–417. 4 K. Kurihara, M. Tamura, K. Shohda, T. Toyota, K. Suzuki and T. Sugawara, Nat. Chem., 2011, 3, 775–781. 5 B. A. Timofeev, I. M. Bolotin, L. P. Stepanova, A. A. Bogdanov, K. Georgiu, S. N. Malyshev, V. V. Petrovsky, A. L. Klibanov and V. P. Torchilin, J. Microencapsulation, 1994, 11, 627–632. 6 Y. Inaoka and M. Yamazaki, Langmuir, 2007, 23, 720–728. 7 D. Needham and R. S. Nunn, Biophys. J., 1990, 58, 997–1009. 8 P. Walde, K. Cosentino, H. Engel and P. Stano, ChemBioChem, 2010, 11, 848–865. 9 M. Sauer, T. Haefele, A. Graff, C. Nardin and W. Meier, Chem. Commun., 2001, 2452–2453. 10 S. Sugiura, T. Kuroiwa, T. Kagota, M. Nakajima, S. Sato, S. Mukataka, P. Walde and S. Ichikawa, Langmuir, 2008, 24, 4581–4588. 11 X. Zhu and M. Liu, Langmuir, 2011, 27, 12844–12850. 12 R. Wick, P. Walde and P. L. Luisi, J. Am. Chem. Soc., 1995, 117, 1435–1436. 13 W. R. Hargreaves and D. W. Deamer, Biochemistry, 1978, 17, 3759–3768. 14 K. Morigaki and P. Walde, Langmuir, 2002, 18, 10509–10511. 15 T. Toyota, K. Takakura and T. Sugawara, Chem. Lett., 2004, 33, 1442–1443. 16 S. Y. Ho and J. Storch, Am. J. Physiol., 2001, 281, C1106– C1117. 17 R. Smith and C. Tanford, J. Mol. Biol., 1972, 67, 75–83. 18 Z.-K. Cui and M. Laeur, Colloids Surf., B, 2014, 114, 177– 185. 19 G. Carbajal, Z.-K. Cui and M. Laeur, Sci. China: Chem., 2012, 56, 40–47. 20 G. Bastiat, P. Oliger, G. Karlsson, K. Edwards and M. Laeur, Langmuir, 2007, 23, 7695–7699.
6456 | Soft Matter, 2014, 10, 6451–6456
Paper
21 Z.-K. Cui, G. Bastiat and M. Laeur, Langmuir, 2010, 26, 12733–12739. 22 T. Phoeung, L. M. Huber and M. Laeur, Langmuir, 2009, 25, 5778–5784. 23 C. Par´ e and M. Laeur, Langmuir, 2001, 17, 5587–5594. 24 Z.-K. Cui, G. Bastiat, C. Jin, A. Keyvanloo and M. Laeur, Biochim. Biophys. Acta, Biomembr., 2010, 1798, 1144–1152. 25 D. C. Drummond, M. Zignani and J.-C. Leroux, Prog. Lipid Res., 2000, 39, 409–460. 26 H. C. Shum, D. Lee, I. Yoon, T. Kodger and D. A. Weitz, Langmuir, 2008, 24, 7651–7653. 27 T. Foster, K. D. Dorfman and H. T. Davis, J. Colloid Interface Sci., 2010, 351, 140–150. 28 T. Foster, K. D. Dorfman and H. T. Davis, Langmuir, 2010, 26, 9666–9672. 29 D. Correia-Ledo, A. A. Arnold and J. Mauzeroll, J. Am. Chem. Soc., 2010, 132, 15120–15123. 30 F. Szoka and D. Papahadjopoulos, Proc. Natl. Acad. Sci. U. S. A., 1978, 75, 4194–4198. 31 G. Bastiat and M. Laeur, J. Phys. Chem. B, 2007, 111, 10929– 10937. 32 J. Ouimet, S. Cro, C. Par´ e, J. Katsaras and M. Laeur, Langmuir, 2003, 19, 1089–1097. 33 T. Phoeung, P. Aubron, G. Rydzek and M. Laeur, Langmuir, 2010, 26, 12769–12776. 34 T. Allen, in Liposome Technology, ed. G. Gregoriadis, CRC Press, Boca Raton, FL, 3rd edn, 1984, vol. 3, pp. 177–182. 35 R. El Jastimi and M. Laeur, Biospectroscopy, 1999, 5, 133– 140. 36 Z.-K. Cui, A. Bouisse, N. Cottenye and M. Laeur, Langmuir, 2012, 28, 13668–13674. 37 N. Cottenye, Z.-K. Cui, K. J. Wilkinson, J. Barbeau and M. Laeur, Biofouling, 2013, 29, 817–827.
This journal is © The Royal Society of Chemistry 2014
Published on 17 June 2014. Downloaded by University of Michigan Library on 30/10/2014 12:19:23.
PAPER
Cite this: Soft Matter, 2014, 10, 6451
View Journal | View Issue
Formation, stability, and pH sensitivity of freefloating, giant unilamellar vesicles using palmitic acid–cholesterol mixtures† Nicolas Cottenye,a Gustavo Carbajal,a Zhong-Kai Cui,a Philippe Dauphin Ducharme,b Janine Mauzerollb and Michel Lafleur*a Despite the fact that palmitic acid (PA) and cholesterol (Chol) do not form fluid bilayers once hydrated individually, giant unilamellar vesicles (GUVs) were formed from a mixture of palmitic acid and cholesterol, 30/70 mol/mol. These free-floating GUVs were stable over weeks, did not aggregate and were shown to be highly stable in alkaline pH compared to conventional phospholipid-based GUVs.
Received 23rd April 2014 Accepted 16th June 2014
Acidic pH-triggered payload release from the GUVs was associated with the protonation state of palmitic acid that dictated the mixing lipid properties, thus affecting the stability of the fluid lamellar phase. The
DOI: 10.1039/c4sm00883a
successful formation of PA–Chol GUVs reveals the possibility to create monoalkylated amphiphile-based
www.rsc.org/softmatter
GUVs with distinct pH stability/sensitivity.
1. Introduction Giant unilamellar vesicles (GUVs) are liposomes with micronsize diameters. These self-assembled entities, whose size is on the order of magnitude of biological cells, are valuable biomimetic models used to study the physicochemical properties of membranes, transmembrane trafficking and protein activity.1 Moreover, GUVs, because of their distinct characteristics including a large entrapped volume, have been used in several technological elds including as reactors for chemical and/or enzymatic reactions2 and mRNA synthesis,3 as systems for vesicle self-reproduction,4 and as in vivo drug carriers.5 Their size allows their observation with an optical microscope and, as a consequence, GUVs are very appropriate systems for examining directly vesicle shape deformations6 and for studying micromechanical properties of bilayers.7 GUVs are predominantly formed from phospholipid mixtures.8 Because of the limited chemical stability of
a
Department of Chemistry, Center for Self-Assembled Chemical Structures, Universit´e de Montr´eal, PO Box 6128, Station Downtown, Montr´eal, Qu´ebec, H3C 3J7, Canada. E-mail: michel.la[email protected]; Fax: +1-5143437586; Tel: +15143403205 b
Laboratory for Electrochemical Reactive Imaging and Detection for Biological Systems, Department of Chemistry, McGill University, 801 Sherbrooke Street West, Montr´eal, Qu´ebec, H3A 0B8, Canada † Electronic supplementary information (ESI) available: Fig. S-1: control indicating that the external acidic pH does not affect the uorescence intensity of the entrapped calcein: uorescence microscopy images of POPC/POPG/Chol GUVs at pH 8.4 and 3.4. Fig. S-2: images, recorded during the titration of GUVs from pH 8.4 to 3.4, representing only the uorescence of apolar Nile red. These images are paired with those obtained from the green channel presented in Fig. 5 of the manuscript. See DOI: 10.1039/c4sm00883a
This journal is © The Royal Society of Chemistry 2014
phospholipids under certain conditions (such as under extreme pH or in the presence of phospholipases), phospholipid-based GUVs display limited robustness and signicant permeability that are not always suitable for some applications. Few alternative materials have been examined for GUV formation in order to overcome phospholipid limitations. For example, few studies report GUV formation using block copolymers.9–11 The formation of GUVs with synthetic detergent (monoalkylated amphiphiles) remains largely unexplored. We know of only four other studies reporting synthetic detergent-based GUVs. GUVs were obtained from a mixture of a fatty acid and its salt (namely oleic acid–sodium oleate mixture)12 as well as from fatty acids mixed in equimolar portion with their corresponding fatty alcohol.13 In these two studies, the preparation method based on slow hydration led a wide diversity of self-assembled structures and the produced GUVs represented only a fraction of these (an estimated proportion of 40–50% being GUVs in ref. 12, and was not dened in the other study). In addition, the fatty acid/fatty soap GUVs appeared to be extremely sensitive to pH and they only existed over a limited pH range. GUVs from fatty acid and its salt were also formed in the presence of squalane at the vicinity of a glass surface,14 yet again with limited stability. Finally, GUVs formed by an ammonium amphiphile containing a Schiff base were reported as transient structures observed strictly during the transition from small vesicles to oil droplets caused by the chemical decomposition of the amphiphile.15 Therefore, an effective approach to create free-oating and stable GUVs from the non-phospholipid single-chain amphiphile is still to be ascertained. The formation of stable GUVs with monoalkylated amphiphiles is particularly challenging because, rst, of their critical aggregation concentrations (e.g. 4 mM for PA16), which are considerably higher than those of
Soft Matter, 2014, 10, 6451–6456 | 6451
View Article Online
Published on 17 June 2014. Downloaded by University of Michigan Library on 30/10/2014 12:19:23.
Soft Matter
conventional phospholipids (e.g. 0.46 nM for dipalmitoylphosphatidylcholine17) and, second, they typically self-aggregate to form micelles, not uid lamellar phases. Herein, we report the formation of stable, free-oating GUVs self-assembled from a mixture of cholesterol (Chol), and fatty acid salt, namely deprotonated palmitic acid (PA). It has been shown that cholesterol combined with some single-chain amphiphiles leads to the formation of uid bilayers in the liquid ordered (lo) phase, despite the fact that these components do not form uid lamellar phases once hydrated individually (for a review, see ref. 18). It is proposed that the singlechain amphiphile acts as a gap ller between cholesterol molecules and ensures bilayer uidity while maintaining tight lipid packing. Large unilamellar vesicles (LUVs) were successfully obtained from such uid bilayers.19–22 The distinct chemical composition of these systems is reected in the distinct properties of the resulting liposomes, including a lower permeability compared to their phospholipid-based analogues, a robust protection to the encapsulated materials, and an enhanced chemical stability.18–20,22–24 GUVs formed with PA– Chol mixtures are expected to display similar properties, associated with the high cholesterol content of the assemblies. Therefore the successful preparation of GUVs from a mixture that includes only a single-chain amphiphile and cholesterol creates a new class of GUVs with improved encapsulation performance. In addition, craing functionalities in monoalkylated amphiphile/sterol self-assembled systems is relatively straightforward. For example, bilayers can be made pH responsive by the use of PA. An hydrated mixture of PA and cholesterol, in a molar ratio of 30/70, forms a lo lamellar phase at pH greater than 7.4 as a result of the fatty acid deprotonation.23 The protonation of PA leads to a phase separation between the constituents, and induces the formation of solid particles of PA and cholesterol. This phase transition causes the rupture of the uid bilayer accompanied by the release of the material entrapped inside LUVs.20 This property can be exploited to cra a pH-triggered release from GUVs, a feature that has been used to achieve controlled drug release in specic acidic locations such as cancer tumours and endosomal compartments.25 We selected PA–Chol mixtures as a proof of concept for making GUVs based on single-chain amphiphile–Chol mixtures because the resulting self-assemblies are expected to additionally be pHresponsive. There are few GUV preparation methods reported in the literature including direct lipid lm hydration, alternate current, and double emulsion methods.8 The PA–Chol GUVs were prepared using the double emulsion droplet method,26–29 which was selected for its simplicity and its high exibility in terms of usable materials. Briey, a water-in-oil (chloroform)in-water double emulsion was formed, with monolayers of amphiphiles at the interfaces, stabilizing the emulsion. Upon complete evaporation of the organic phase, the apolar parts of the amphiphile monolayers entered in contact to form bilayers and to give rise to GUVs.30 It was proposed that the size of the internal water droplets dened the size of the resulting GUVs.
6452 | Soft Matter, 2014, 10, 6451–6456
Paper
2.
Results and discussion
In order to establish the formation of GUVs, epiuorescence imaging was carried out. Two uorescent probes were used: Nile red, a hydrophobic probe that was inserted in the hydrophobic core of lipid bilayers, and calcein, a hydrophilic probe that was trapped in the GUV internal pool. This approach allowed the observation of both the walls of the GUVs and their entrapped content. The GUV stability at alkaline pH was investigated by assessing its permeability to a uorophore (sulforhodamine B (SRB)) and compared to conventional phospholipid-based GUVs made of 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC)–Chol (60/40 mixture) as control. The pH-sensitivity of the GUVs was investigated by light scattering variations and by uorescence microscopy. The uorescence micrographs presented in Fig. 1 demonstrate the successful self-assembly of GUVs from a PA–Chol 30/ 70 mixture using the lipid-stabilized double-emulsion method. The GUVs had a spherical morphology with diameters mainly ranging from 1 to 5 mm. Occasionally, GUVs had diameters larger than 10 mm. The diameter distribution is presented in Fig. 2. Overall, the average diameters of PA–Chol 30/70 GUVs were 3.3 2.1 mm as established from uorescence imaging (n ¼ 154), and 3.4 1.9 mm as measured from dynamic light scattering. Multilamellar vesicles and more complex particles were also detected but they represented at most 10% of the observed self-assemblies. PA– Chol GUVs were highly stable and could be stored at room temperature for a month without noticeable change in morphology or size. The enhanced stability is consistent with the ndings obtained with PA–Chol LUVs, which also exhibited a stable behaviour over extended periods of time.24 We have assessed the permeability of PA–Chol GUVs by measuring the passive release of SRB (Fig. 3). PA–Chol GUVs remained essentially impermeable to SRB at pH 12 as less than 10% of entrapped uorophore had leaked out aer 4.5 days. This very limited membrane permeability is consistent with that previously reported for PA–Chol LUVs.20,31 In addition, because
Fluorescence microscopy images of PA–Chol GUVs at pH 8.4. The red channel corresponds to membranes labelled with Nile red whereas the green channel represents calcein entrapped in the inner aqueous pool of the GUVs. The image on the left is obtained by the superposition of the two images (right, top and bottom). The scale bars correspond to 10 mm.
Fig. 1
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 17 June 2014. Downloaded by University of Michigan Library on 30/10/2014 12:19:23.
Paper
Soft Matter
Fig. 2 Size distribution of PA–Chol GUVs at pH 8.4, sorted by a 0.5mm steps, as obtained from the measurements from fluorescence microscopy pictures (n ¼ 154). The average diameter was 3.3 2.1 mm. Fig. 4 Destabilization of PA–Chol GUVs induced by a decrease of pH, as observed by light scattering measurements. The dashed line is only a guide to the eyes. A strong decrease in scattered light is observed below pH 6, corresponding to GUV disruption, and sedimentation of the resulting solid particles at acidic pH.
PA–Chol GUV permeability to SRB at pH 12 is compared with that obtained for POPC–Chol GUVs. PA–Chol GUVs display under these alkaline conditions a considerable enhanced chemical stability compared to POPC–Chol ones.
Fig. 3
of their chemical composition, PA–Chol GUVs exhibited an enhanced stability compared to conventional phospholipid formulation. For example, POPC–Chol GUVs, which also exist in the lo phase, were leakier under these alkaline conditions, leading to the release of 50% of trapped SRB over 4.5 days. This behaviour was associated with the alkaline hydrolysis of the ester groups linking the acyl chains to the glycerol backbone of POPC, a chemical reaction that disrupted the bilayer integrity. This difference illustrates very well the good chemical stability of PA–Chol GUVs. The sensitivity of PA–Chol GUVs to acidic pH was rst assessed using light scattering (Fig. 4) by titrating GUVs prepared at pH 8.4 with HCl, down to an external pH of 3.4. Between pH 8.4 and 6.5, a considerable scattering was observed due to the presence of numerous GUVs in the suspension. Between pH 6.5 and 5.0, a large decrease of the scattered light intensity was observed. The pH range associated with this
This journal is © The Royal Society of Chemistry 2014
transformation is consistent with the apparent pKa of PA inserted in uid bilayers, that was estimated to be between 6.4 and 8.2.20, 32–33 In these studies, PA–Chol uid bilayers were shown to become unstable when a large fraction of the fatty acid was protonated, as the electrostatic intermolecular interactions did not favour lipid mixing. Under these conditions, there was the formation of solid particles of cholesterol and of PA.20 On the basis of these results, we interpreted the reduced light scattering intensity observed at low pH by the destabilization of the GUVs that underwent a transition towards solid cholesterol and solid palmitic acid. These solid particles likely sedimented in the measurement cell, reducing, as a consequence, the light scattering power. To corroborate that the pH-triggered disruption of the GUVs observed from the light scattering experiments effectively leads to the formation of solid lipid particles and a triggered release of the GUV payload, epiuorescence studies of PA–Chol GUVs were performed (Fig. 5). Nile-red labeled GUVs were initially formed in a sucrose- and a calcein-containing buffer, and then sedimented at the glass-slide interface in a glucose-containing buffer. The pH of both the internal and external buffers was 8.4. The pH was subsequently modied by sequential additions of diluted HCl. Despite the fact that the HCl aliquots were added as gently as possible, the pH variations created convective ows that were sufficient to displace the GUVs, preventing a proper tracking of specic vesicles. Nevertheless, pH-triggered calcein release could be clearly observed during titration experiments. Down to pH 6.4, no signicant morphological variations, swelling, or uorophore leakage could be detected, indicating unaltered GUV permeability. Below pH 6.4, the GUV density, as observed by their internal content labeled with calcein, decreased. It should be pointed out that calcein uorescence is known to be considerably decreased at pH below 5;34 however the addition of HCl essentially modied the external pH of the
Soft Matter, 2014, 10, 6451–6456 | 6453
View Article Online
Soft Matter
Paper
is believed that the Nile red-labeled fragments observed at low pH corresponded to PA and/or cholesterol solid particles, which would have essentially a hydrophobic core. At pH 3.4, GUVs were no longer observed. These results clearly indicated that the GUVs were pH-sensitive and became unstable due to pH trigger.
3.
Experimental
Published on 17 June 2014. Downloaded by University of Michigan Library on 30/10/2014 12:19:23.
3.1. Materials Palmitic acid (PA) (99%), cholesterol (Chol) (>99%), Nile red (>98%), tris(hydroxymethyl)aminomethane (TRIS) (99%), 2-[Nmorpholino]ethanesulfonic acid (MES) (>99%), diethyl ether (>99%), glucose (ACS reagent) and sucrose (ACS reagent) were supplied by Sigma Chemical Co. (St. Louis, MO, USA). KCl (ACS reagent) and chloroform (>99%) were obtained from Biopharm and Anachemia (Burlignton, ON, CA), respectively. Calcein (high purity) and SRB were obtained from Invitrogen Corporation (Burlignton, ON, CA). POPC (>99%) was purchased from Avanti Polar Lipids, Inc. (Birmingham, AL, USA). Milli-Q water with a resistivity of 18.2 MU cm was used for the buffer preparation. 3.2. GUV formation
Fig. 5 Effect of pH on the GUV stability monitored by fluorescence microscopy. Larger images at extreme pH values: 8.4 (top) and 3.4 (bottom) display the overlaid calcein and Nile red fluorescence signals. The smaller images (middle), recorded during the titration of GUVs from pH 8.4 to 3.4, present only the fluorescence of calcein from the inner pool of the GUVs. The scale bars represent 10 mm. Vesicle burst was observed from pH 6.4. At pH 3.4, GUVs were no longer observed.
suspension and should have a very limited impact on GUV internal pH. In fact, we showed, as a control, that the uorescence of calcein trapped in phospholipid-based and pH-insensitive GUVs was not signicantly affected over the pH range from 3.4 to 8.4 (Fig. S1 of the ESI†). The disappearance of calcein labeled GUVs upon pH decrease was concomitant with an increased number of particles completely labeled with Nile red, without inner pool (see Fig. 5 and S2 of the ESI†). It has been established that the PA–Chol system undergoes a pH triggered phase transition from a uid lamellar phase to solid PA and solid cholesterol when the pH of a suspension is decreased.23 It
6454 | Soft Matter, 2014, 10, 6451–6456
PA–Chol (30/70) mixtures were obtained by solubilizing weighed amounts of each component in a benzene–methanol mixture (75/25 (v/v)). POPC–Chol (60/40) mixtures were obtained by solubilizing weighed amounts of each component in a benzene–methanol mixture (90/10 (v/v)). The solutions were then freeze-dried for at least 16 h. Subsequently, the resulting powder was dissolved in chloroform (10 mg mL1). For the labeling of the GUV membrane, Nile red was added to this solution (50 mg mL1). In order to make a double emulsion, two suspensions were prepared. In the rst vial, 320 mL of the lipid solution, 1 mL of chloroform, and 1 mL of a sucrose-containing buffer (TRIS 5 mM; MES 5 mM; sucrose 215 mM, pH 8.4) were mixed. For the labeling of the GUV internal aqueous compartment, calcein (0.1 mM) was added to this buffer. The second vial contained 320 mL of the lipid solution, 0.5 mL of diethyl ether and 2.5 mL of the sucrose-containing buffer. The rst and second vials were sonicated for 45 and 15 s respectively. Their contents were mixed and sonicated for an additional 10 s period in order to form the water-in-oil-in-water double emulsion. The sample was transferred to a round bottom ask and heated under an argon atmosphere between 65 and 68 C for 45 min to allow the evaporation of the organic solvents. In order to isolate the GUVs, the suspension was centrifuged at 1500g for 30 min, the supernatant was removed, and the GUVs were resuspended in 5 mL of the sucrose-containing buffer. This procedure was repeated twice. 3.3. Fluorescence microscopy A 100 mL aliquot of a glucose-containing buffer (TRIS 5 mM; MES 5 mM; glucose 50 mM; KCl 100 mM, pH 8.3) was deposited on a glass slide and 20 mL of the GUV suspension were added. The glucose- and sucrose-containing buffers were iso-osmolar (250 mOsm L1) to prevent GUV collapse, shrinkage or bursting
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 17 June 2014. Downloaded by University of Michigan Library on 30/10/2014 12:19:23.
Paper
as a result of an osmotic stress. A 5 min period was allowed for the GUVs to sediment to the glass slide surface. In order to study the pH sensitivity of the GUVs, the pH of a sample prepared at pH 8.4 as described above was decreased by the successive addition of 2 mL aliquots of HCl, 0.05 M, added every 3 min. The pH was measured aer each HCl addition using a combination pH microelectrode (Microelectrodes Inc., Bedford, NH, USA). Aer 8 aliquots, the pH of the suspension reached 3.6 0.6. Fluorescence images of Nile red- and calcein-labeled GUVs were obtained using lexc/lem: 561/600–650 nm and 488/ 508–540 nm, respectively. Images were digitally processed using the FIJI/ImageJ image processing soware. Images of static GUVs were acquired using an upright uorescence microscope (Axioskop 2 Plus, Carl Zeiss, Inc.) equipped with a Qimaging cooled charge-coupled-device camera, a 100 Plan-Neo Zeiss objective, and the Image-Pro Plus acquisition soware whereas images assessing pH sensitivity of the GUVs were acquired using an inverted uorescence microscope (Olympus IX81) equipped with a 100 objective, and the Image-Pro Plus acquisition soware. 3.4. Light scattering measurement The light (633 nm) scattering was measured at an angle of 173 , using a Zetasizer NanoZS equipped with an MPT-2 Autotitrator (Malvern Instrument). Typically, the pH of a GUV suspension in the sucrose-containing buffer was varied from 8.4 to 3.0 with 0.5-unit steps. Autotitrations were carried out with HCl, 0.1 and 0.05 M. The titrations were performed in triplicates. 3.5. Stability assessment at alkaline pH PA–Chol (30/70) and POPC–Chol (60/40) GUVs were prepared as mentioned above except that the lipids were hydrated with a buffer containing TRIS (5 mM), SRB (80 mM), pH 8.4. The SRB concentration led to self-quenching of its uorescence.35 Free SRB was separated from SRB-loaded GUVs using Amicon Ultra10 centrifugal lter units (Milipore, Billerica, MA) spun at 1500g. The GUVs were harvested on the lter and resuspended in a buffer containing TRIS (5 mM), and NaCl (450 mM), pH 8.4. The osmolarity of the GUV internal and external buffers was carefully matched to ensure iso-osmolarity over the experiment course. The washing of the SRB-loaded GUVs was repeated until the SRB self-quenching factor was at least 0.9. The nal GUV concentration was approximately 1 mM of PA or POPC. To measure the stability of these GUVs at pH 12, the percentage of entrapped SRB was calculated, as previously reported by Phoeung et al.22 In this experiment an initial reference consisting of a sample kept at pH 8.4 diluted 10 times in TRIS (5 mM), NaCl (450 mM) buffer, pH 8.4 then 5 times in TRIS (50 mM), NaCl (370 mM) buffer, pH 8.4, was measured before (Ii) and aer (Ii + T) the addition of Triton X-100 (1 (v/v)% in the pH-8.4 buffer). Then 1 mL of the GUV initial suspension was diluted in 9 mL of an iso-osmolar buffer, pH 12, containing NaCl (350 mM); this was set as time ¼ 0. At different times, a 200 mL aliquot of GUVs in alkaline buffer was diluted in an isoosmolar buffer at pH 8.4 (TRIS 50 mM, NaCl 370 mM, pH 8.4) in order to avoid the inuence of pH on SRB uorescence. The This journal is © The Royal Society of Chemistry 2014
Soft Matter
nal lipid concentration was 20 mM of PA or POPC. Immediately aer the nal dilution, the uorescence intensity (If) of the samples was measured (lex/lem ¼ 565/585 nm). Subsequently, 20 mL of Triton X-100 (1 (v/v)% in the pH 8.4 buffer) was added to completely release the encapsulated SRB and the resulting uorescence intensity was measured (If + T) to normalize the subsequent measurements. The percentage of encapsulation for each time point of the series was calculated using eqn (1): ! IfþT If IiþT 100: % encapsulated SRB ¼ (1) ðIiþT Ii ÞIfþT
The reported average values and standard deviations (Fig. 2) were obtained from duplicate measurements of 2 parallel release experiments per batch on 2 independent GUV batches.
4. Conclusions Synthetic detergent-based GUVs were produced from a palmitic acid salt–cholesterol mixture. Despite the fact that PA, analogous to other synthetic detergents, has a propensity to form micelles and generally induces destabilization of lamellar structures, it is believed that the high cholesterol content in the lamellar phase led to a tight lipid packing that ensured the bilayer cohesion.23 These GUVs were stable at room temperature and displayed a pH-triggered release of their content, analogous to that previously observed for PA–Chol LUVs.31 Different monoalkylated amphiphile/sterol systems have been reported to form uid bilayers and LUVs could be derived from these mixtures.18,21,24,36,37 Several of their properties, including their overall charge, their response to stimuli such as pH and light, and their permeability, can be modulated in a rational and predictable manner by the selection of the constituents. The successful formation of the PA–Chol GUVs presented here as a proof of concept indicates the possibility of developing a novel and versatile platform of self-assembled free-oating GUVs prepared from non-phospholipid molecules that could present distinct advantages, including good chemical stability, low permeability, tunable surface charge, tunable sensing properties, and preparation from inexpensive and biocompatible components.
Acknowledgements This work was supported by the Universit´ e de Montr´ eal, the Natural Sciences and Engineering Research Council of Canada and by the Fonds Qu´ eb´ ecois de la Recherche sur la Nature et les Technologies through its Strategic Cluster program.
Notes and references 1 K. Kitatokarczyk, J. Grumelard, T. Haefele and W. Meier, Polymer, 2005, 46, 3540–3563. 2 K. Kato, P. Walde, H. Mitsui and N. Higashi, Biotechnol. Bioeng., 2003, 84, 415–423.
Soft Matter, 2014, 10, 6451–6456 | 6455
View Article Online
Published on 17 June 2014. Downloaded by University of Michigan Library on 30/10/2014 12:19:23.
Soft Matter
3 A. Fischer, A. Franco and T. Oberholzer, ChemBioChem, 2002, 3, 409–417. 4 K. Kurihara, M. Tamura, K. Shohda, T. Toyota, K. Suzuki and T. Sugawara, Nat. Chem., 2011, 3, 775–781. 5 B. A. Timofeev, I. M. Bolotin, L. P. Stepanova, A. A. Bogdanov, K. Georgiu, S. N. Malyshev, V. V. Petrovsky, A. L. Klibanov and V. P. Torchilin, J. Microencapsulation, 1994, 11, 627–632. 6 Y. Inaoka and M. Yamazaki, Langmuir, 2007, 23, 720–728. 7 D. Needham and R. S. Nunn, Biophys. J., 1990, 58, 997–1009. 8 P. Walde, K. Cosentino, H. Engel and P. Stano, ChemBioChem, 2010, 11, 848–865. 9 M. Sauer, T. Haefele, A. Graff, C. Nardin and W. Meier, Chem. Commun., 2001, 2452–2453. 10 S. Sugiura, T. Kuroiwa, T. Kagota, M. Nakajima, S. Sato, S. Mukataka, P. Walde and S. Ichikawa, Langmuir, 2008, 24, 4581–4588. 11 X. Zhu and M. Liu, Langmuir, 2011, 27, 12844–12850. 12 R. Wick, P. Walde and P. L. Luisi, J. Am. Chem. Soc., 1995, 117, 1435–1436. 13 W. R. Hargreaves and D. W. Deamer, Biochemistry, 1978, 17, 3759–3768. 14 K. Morigaki and P. Walde, Langmuir, 2002, 18, 10509–10511. 15 T. Toyota, K. Takakura and T. Sugawara, Chem. Lett., 2004, 33, 1442–1443. 16 S. Y. Ho and J. Storch, Am. J. Physiol., 2001, 281, C1106– C1117. 17 R. Smith and C. Tanford, J. Mol. Biol., 1972, 67, 75–83. 18 Z.-K. Cui and M. Laeur, Colloids Surf., B, 2014, 114, 177– 185. 19 G. Carbajal, Z.-K. Cui and M. Laeur, Sci. China: Chem., 2012, 56, 40–47. 20 G. Bastiat, P. Oliger, G. Karlsson, K. Edwards and M. Laeur, Langmuir, 2007, 23, 7695–7699.
6456 | Soft Matter, 2014, 10, 6451–6456
Paper
21 Z.-K. Cui, G. Bastiat and M. Laeur, Langmuir, 2010, 26, 12733–12739. 22 T. Phoeung, L. M. Huber and M. Laeur, Langmuir, 2009, 25, 5778–5784. 23 C. Par´ e and M. Laeur, Langmuir, 2001, 17, 5587–5594. 24 Z.-K. Cui, G. Bastiat, C. Jin, A. Keyvanloo and M. Laeur, Biochim. Biophys. Acta, Biomembr., 2010, 1798, 1144–1152. 25 D. C. Drummond, M. Zignani and J.-C. Leroux, Prog. Lipid Res., 2000, 39, 409–460. 26 H. C. Shum, D. Lee, I. Yoon, T. Kodger and D. A. Weitz, Langmuir, 2008, 24, 7651–7653. 27 T. Foster, K. D. Dorfman and H. T. Davis, J. Colloid Interface Sci., 2010, 351, 140–150. 28 T. Foster, K. D. Dorfman and H. T. Davis, Langmuir, 2010, 26, 9666–9672. 29 D. Correia-Ledo, A. A. Arnold and J. Mauzeroll, J. Am. Chem. Soc., 2010, 132, 15120–15123. 30 F. Szoka and D. Papahadjopoulos, Proc. Natl. Acad. Sci. U. S. A., 1978, 75, 4194–4198. 31 G. Bastiat and M. Laeur, J. Phys. Chem. B, 2007, 111, 10929– 10937. 32 J. Ouimet, S. Cro, C. Par´ e, J. Katsaras and M. Laeur, Langmuir, 2003, 19, 1089–1097. 33 T. Phoeung, P. Aubron, G. Rydzek and M. Laeur, Langmuir, 2010, 26, 12769–12776. 34 T. Allen, in Liposome Technology, ed. G. Gregoriadis, CRC Press, Boca Raton, FL, 3rd edn, 1984, vol. 3, pp. 177–182. 35 R. El Jastimi and M. Laeur, Biospectroscopy, 1999, 5, 133– 140. 36 Z.-K. Cui, A. Bouisse, N. Cottenye and M. Laeur, Langmuir, 2012, 28, 13668–13674. 37 N. Cottenye, Z.-K. Cui, K. J. Wilkinson, J. Barbeau and M. Laeur, Biofouling, 2013, 29, 817–827.
This journal is © The Royal Society of Chemistry 2014
