Letter pubs.acs.org/Langmuir
Shear-Induced Membrane Fusion in Viscous Solutions Maxim Kogan, Bobo Feng, Bengt Nordén, Sandra Rocha,* and Tamás Beke-Somfai* Department of Chemical and Biological Engineering, Physical Chemistry, Chalmers University of Technology, SE-412 96 Gothenburg, Sweden ABSTRACT: Large unilamellar lipid vesicles do not normally fuse under fluid shear stress. They might deform and open pores to relax the tension to which they are exposed, but membrane fusion occurring solely due to shear stress has not yet been reported. We present evidence that shear forces in a viscous solution can induce lipid bilayer fusion. The fusion of 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) liposomes is observed in Couette flow with shear rates above 3000 s−1 provided that the medium is viscous enough. Liposome samples, prepared at different viscosities using a 0−50 wt % range of sucrose concentration, were studied by dynamic light scattering, lipid fusion assays using Förster resonance energy transfer (FRET), and linear dichroism (LD) spectroscopy. Liposomes in solutions with 40 wt % (or more) sucrose showed lipid fusion under shear forces. These results support the hypothesis that under suitable conditions lipid membranes may fuse in response to mechanical-force-induced stress.
■
INTRODUCTION Membrane fusion occurs during cellular processes such as intracellular traffic. Membranes in vivo do not fuse spontaneously but under a selective process controlled by specific proteins. Despite an extensive study of protein-induced membrane fusion, the mechanistic details are not yet understood. Phospholipid vesicle fusion can be induced in vitro by molecules such as poly(ethylene glycol), dextran, and multivalent ions.1−4 Seemingly, the role of these compounds is to reduce the distance between the two bilayers by overcoming the hydration forces between them. Direct dehydration of stack bilayers results in contact points between apposed monolayers, which merge into a membrane intermediate state known as stalk.5 Forcing bilayers into close proximity to one another using DNA bases or membrane-anchored DNA strands also triggers vesicle hemifusion or fusion.6−8 Those studies suggest that direct mechanical force could initiate fusion; however, how to achieve fusion in a solution-phase system is far from trivial. Although there has been some indication that shear flow might assist lipid fusion in the presence of surfactants9 and is known to induce the deformation of lipid bilayer vesicles,9−12 shear force alone has not been seen to promote fusion. Laminar shear flow led to the formation of nanotubes of surface-immobilized giant vesicles containing pegylated lipids.12 We report that shear force can induce the lipid bilayer fusion of vesicles without the need to add any catalyzing agents. We prepared the liposomes in buffer containing sucrose and exposed them to shear forces in a Couette flow cell. The increased viscosity and shear rate of 3100 s−1 resulted in lipid bilayer fusion. An important aspect of fusion is the membrane curvature. If the ratio of the cross sections between the lipid headgroup and its hydrocarbon chain is smaller than 1, the lipid monolayer has a tendency to bend toward the side of the headgroup (negative curvature). It is known that lipids that supply critical negative spontaneous curvature to bilayers tend to promote fusion.13,14 We used a lipid with effectively neutral © 2014 American Chemical Society
curvature, which does not support vesicle fusion, to address the efficiency of the applied mechanical force on lipid mixing.15 Our system consisted of large unilamellar liposomes of DOPC (1,2-dioleoyl-sn-glycero-3-phosphocholine) prepared in 50 wt % sucrose buffer. The liposomes were subjected to shear flow and eventually fused. This process can be characterized by dynamic light scattering (DLS) and fluorescence measurements based on fluorescence resonance energy transfer (FRET). Additionally, we observed that membrane fusion improves the linear dichroism (LD) signal of lipid-embedded molecules such as retinoic acid, used as an internal membrane orientation probe.
■
EXPERIMENTAL SECTION
Sample Preparation. DOPC liposomes were prepared by the lipid film hydration method. The lipid (Avanti) was dissolved at a concentration of 1.3 mM in chloroform, which was then evaporated, and the resultant dried film was hydrated with buffer (10 mM TES, 100 mM NaCl, 0.1 mM EDTA, pH 7.4) containing the desired sucrose concentration (range 0−50 wt %). The lipid film was hydrated with buffer already containing sucrose to ensure that the viscosities of the inner and outer liquid phases of the liposome are the same. The suspension was then freeze−thawed and extruded through polycarbonate filters with a pore diameter of 100 nm using a Lipex extruder (Northern Lipids). Retinoic acid in ethanol was added to the liposome samples at a probe-to-lipid molar ratio of 1:300 (final volume of ethanol was less than 2% (v/v)), and the samples were equilibrated for 2 h prior to the measurements. Dynamic Light Scattering (DLS). DLS measurements of liposomes in 50 wt % sucrose were performed using a Zetasizer Nano ZS instrument (Malvern Instruments Ltd.). The time-dependent autocorrelation function for each measurement was acquired every 10 s, with 12 acquisitions. The parameters were the following: laser Received: December 19, 2013 Revised: March 17, 2014 Published: April 23, 2014 4875
dx.doi.org/10.1021/la404857r | Langmuir 2014, 30, 4875−4878
Langmuir
Letter
wavelength of 633 nm, scattering angle of 173°, temperature of 25 °C, medium viscosity of 12.500 cP, and medium refractive index of 1.420. Fluorescence Resonance Energy Transfer. Vesicle fusion was monitored via lipid and content mixing and assayed by fluorescence dequenching of the lipidic fluorescence resonance energy transfer (FRET) pair of N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl) phosphatidylethanolamine (NBD-PE, Avanti)/N-(lissamine rhodamine B sulfonyl) phosphoethanolamine (Rh-PE, Avanti) and dissociation and dequenching of the soluble 8-aminonaphthalene-1,3,6 trisulfonic acid (ANTS, Life Technology)/p-xylene-bis-pyridinium bromide (DPX, Life Technology) complex. Labeled liposomes were prepared with 0.8 mol % each of NBD-PE and Rh-PE in the initial chloroform solution of DOPC. Unlabeled liposomes and calibration liposomes (mock fused liposomes) containing 0.08 mol % of each fluorophore were also prepared. Labeled and unlabeled liposomes, prepared in 50 wt % sucrose buffer were mixed at a molar ratio of 1:9. The percentage of lipid mixing is given by M(t) = 100 × [I(s) − I(0)]/[I(c) − I(0)], where I(s) is the fluorescence intensity of the samples after rotation, I(0) is the fluorescence of the mixture of labeled and unlabeled liposomes before rotation, and I(c) is the fluorescence of mock fused liposomes. Liposomes containing ANTS (25 mM), DPX (90 mM), or a ANTS/DPX mixture (12.5 mM ANTS/45 mM DPX) were prepared. The osmolality of ANTS and DPX solutions was adjusted with sucrose using a Fiske 210 micro-sample osmometer in order to match the osmolality of the buffer used to remove the free probes (50 wt % sucrose in 10 mM TES containing 100 mM NaCl, 0.1 mM EDTA, pH 7.4), which was carried out by dialysis using a Float-A-Lyzer G2MWCO 100 kDa (Spectrum Laboratories, Inc.). The ANTSloaded liposomes were mixed with DPX-loaded liposomes in a 1:1 volume ratio, and changes in fluorescence were monitored (intensity of ANTS-liposomes at 25 μM lipid is adjusted to an arbitrary unit of 100% and the fluorescence of 50 μM ANTS/DPX-liposomes is taken as 0%). The fluorescence of (ANTS/DPX)-loaded liposomes was measured and normalized against that of the same samples in the presence of Triton. The fluorescence intensity spectra were obtained with a Varian Cary Eclipse fluorescence spectrophotometer. The excitation wavelengths were set at 460 and 354 nm for NBD-PE/Rh-PE and ANTS/DPX, respectively. Flow Linear Dichroism (LD) Spectroscopy. LD spectra were recorded using a Chirascan CD spectrometer equipped with an LD detector and a custom-made outer cylinder rotation Couette flow cell with a sample volume of 2 mL and a path length of 1 mm. The spectra were collected between 200 and 450 nm in 1 nm increments at a scan speed of 100 nm/min. At least three data accumulations were undertaken to generate an average for each measurement. Additionally, baselines at zero shear gradients were measured and subtracted from all spectra.
Figure 1. Effect of shear flow at 6200 s−1 on the size of DOPC liposomes prepared in 50 wt % sucrose: (A) dynamic light scattering autocorrelation function and (B) intensity size distribution before and after shearing for 2 h.
(NBD), confirming the fusion of the lipid bilayers (Figure 2). The lipid mixing percentages were 46, 84, and 100% after 20,
Figure 2. FRET efficiency, in 50 wt % sucrose buffer, of (NBD-PE/ Rh-PE)-labeled liposomes mixed with unlabeled liposomes before (black line) and after rotating flow at 6200 s−1 (A) and 3100 s−1 (B) for 20 min (dark cyan), 1 h (purple), and 2 h (gray dotted line). The spectrum of mock fused liposomes, corresponding to 100% lipid mixing, is shown in light gray (overlapped by the spectra at 1 and 2 h in A and with the spectrum at 2 h in B).
60, and 120 min respectively, at a shear rate of 3100 s−1. At 6200 s−1, the lipid mixing was 76% after 20 min and reached 100% within 1 h. The dye/quencher pair ANTS/DPX liposome fusion assay, which probes the intermixing of aqueous contents, was also performed. The initial fluorescence emission from a 1:1 mixture of ANTS-loaded liposomes and DPX-loaded liposomes did not change after rotating flow. The incorporation of the dye/ quencher pair in the same liposome batch shows that shear forces induce the leakage of content from the liposomes (Figure 3), which is most likely the reason that it is not possible to establish internal mixing of the vesicle contents after applying the shear force. The changes induced on the liposomes by shear flow were also monitored by LD spectroscopy. LD measures the difference in the absorption between parallel and perpendicularly polarized light, and thus a prerequisite is the macroscopic orientation of the sample, which is achieved by subjecting liposomes to shear flow in a Couette cell.17 Molecules associated with the liposomes will also be aligned relative to the flow. Recently, it has become clear that flow-LD is an attractive technique for studying membrane systems, providing information on the structure, function, orientation angle, and insertion kinetics of peptides and proteins.18,19 The degree of flow alignment is characterized by the macroscopic orientation
■
RESULTS AND DISCUSSION The DLS data confirm that the initial size of DOPC liposomes in 50 wt % sucrose buffer (120 nm) changes and becomes polydisperse (mean diameter 360 nm) after shear flow for 120 min at 6200 s−1 (Figure 1). The liposomes neither recovered their initial size nor precipitated or flocculated, even after 24 h of having stopped the shearing. Lipid mixing was assessed by FRET between labeled lipids, NBD-PE and Rh-PE. The presence of the two probes in the liposomes at a concentration near 1 mol % results in an efficient energy transfer between NBD-PE and Rh-PE, corresponding to a minimal fluorescence intensity of the energy donor NBD-PE and high fluorescence of Rh-PE.16 When membrane fusion occurs, the average distance separating the fluorophores increases, leading to a decrease in the efficiency of the energy transfer and consequently the fluorescence intensity of NBDPE increases. Shear flow, at a rate of either 3100 or 6200 s−1, led to a dramatic increase in the fluorescence from the donor 4876
dx.doi.org/10.1021/la404857r | Langmuir 2014, 30, 4875−4878
Langmuir
Letter
The final orientation factor was S = 0.25 (LDr = 0.38), which corresponds to a nearly 4-fold increase of the initial value (Figure 4). We also investigated the influence of the sucrose concentration (0−50 wt %) on liposomes under a shear rate of 3100 s−1 to address the effect of the medium viscosity on the LD signal (Figure 4). No changes in the LD could be observed for samples with 0 and 10 wt % sucrose, and samples containing 30 wt % sucrose showed only a slight increase in the LD signal in absolute value. However, for 40 and 50 wt % sucrose samples a rapid increase in the absolute value of the LD signal is observed. This difference might be attributed to the sample viscosity, which is 2.8, 5.2, and 12.5 cP for 30, 40, and 50 wt % sucrose buffers, respectively.23 The higher viscosity also increases the applied shear stress, which leads to bilayer fusion. In the shear flow, the vesicles likely adapt to the applied stress by developing a tension in the membrane and by deforming into transiently ellipsoidal shapes.11 The deformation changes the curvature of the bilayer, and it is expected that the more viscous the solution, the more the viscous drag will predominate and the more deformed the liposomes will be. The highly curved areas of two vesicles, extremities of the ellipsoids, contact with each other, and, due to the high membrane tension, the lipids in apposing bilayers may form a fusion pore.24 Membrane curvature stress induced by deformation is known to control vesicle fusion on solid substrates.25 When the deformation reaches a critical value, vesicles rupture and fuse with each other to release the membrane curvature stress.25 Thus, a way of relaxing the tension during flow may in this case also be via pore formation. Due to the high viscosity of the medium, the diffusion of the lipids is restrained and the liposome pores will not close fast enough, leading to the formation of an open structure that eventually fuses with another structure. It is also possible that a stack of bilayers are being formed and that they are able to contact at different points and fuse. Both scenarios would result in a better alignment of the lipid phase along the axis of the applied shear force. This can be readily followed by the increase in the LD intensity, which shows that the lipid phase has an increasing orientation factor with time. In summary, we demonstrate how mechanical forces can lead to lipid bilayer mixing in viscous media, which so far has been an unexplored way of achieving membrane fusion. The high tension induced by shear forces on the membrane creates high curvature areas on the liposomes, which probably relaxes this tension by forming pores that may also fuse.24,26
Figure 3. Fluorescence spectra of (ANTS/DPX)-loaded liposomes in 50 wt % sucrose buffer before (black line) and after shearing at 3100 s−1 for 1 h (gray line). The spectra were normalized to the spectrum of liposomes in the presence of Triton (dark cyan).
parameter S. The macroscopic ordering refers to the degree of deformation of the lipid vesicles and their alignment in the direction of the flow, which is governed by the S factor. The value of S is determined from the reduced LD values (LDr), a dimensionless concentration-independent variable defined as LDr = LD/Aiso, where Aiso is the absorption of the isotropic sample. For every specific transition dipole moment of a membrane probe oriented in the lipid bilayer, the following relationship holds20 LD 3 LDr = = S(1 − 3 cos2 α) A iso 4 (1) where α is the angle between the membrane normal and the transition moment of the probe. Retinoic acid was incorporated into DOPC liposomes, and LD measurements were carried out under the same conditions at which fusion is observed. Sucrose also has the advantage of reducing the light scattering of the liposomes by matching their refractive index.21 On the basis of previous reports, the angle α of retinoic acid is nearly equal to 0° for the transition moment at 350 nm.18,22 LD of retinoic acid in DOPC liposomes was measured over time at a shear force of 6200 s−1. At t = 0 min, the LD signal shows the typical magnitude range of the flow-oriented liposomes with an orientation factor of ∼0.07 (LDr = 0.1) (Figure 4). The LD signal increases in absolute value with time, almost linearly until it reaches a plateau at about 80 min after continuous shearing.
■
OUTLOOK The described setup enables, among others, studies on how vesicles behave in viscous media in the presence of fusion proteins and under different types of shear forces, which are expected to give novel insights into this vital biological process. It should also be noted that LD spectroscopy seems particularly useful in continuous follow up for this process. LD will be further tested to study molecular interactions formed between membrane-bound proteins by mixing liposomes containing each of the different molecules.
Figure 4. (A) LD of retinoic acid incorporated into DOPC liposomes in 50 wt % sucrose under a shear force of 6200 s−1 at initial times (0) and after 20, 40, 60, and 80 min. (B) Variation over time of the LDr values at 350 nm for retinoic acid incorporated into DOPC liposomes in the presence of 0, 10, 30, 40, and 50 wt % sucrose under an applied shear force of 3100 s−1 compared to the sample with 50 wt % sheared at 6200 s−1. Note that LDr values are positive due to their normalization to the first measured point (considered to be the 0 min point) for better comparison.
■
AUTHOR INFORMATION
Corresponding Authors
*E-mail:
[email protected]. Fax: +46 31 7723858. Tel: +46 31 772 2815. *E-mail:
[email protected]. Fax: +46 31 7723858. Tel: +46 31 7723807. 4877
dx.doi.org/10.1021/la404857r | Langmuir 2014, 30, 4875−4878
Langmuir
Letter
Notes
(19) Svensson, F. R.; Lincoln, P.; Norden, B.; Esbjorner, E. K. Tryptophan orientations in membrane-bound gramicidin and melittina comparative linear dichroism study on transmembrane and surfacebound peptides. Biochim. Biophys. Acta, Biomembr. 2011, 1808 (1), 219−228. (20) Nordén, B.; Rodger, A.; Dafforn, T. Linear Dichroism and Circular Dichroism. A Textbook on Polarized-Light Spectroscopy; The Royal Society of Chemistry: London, 2010. (21) Ardhammar, M.; Lincoln, P.; Norden, B. Invisible liposomes: Refractive index matching with sucrose enables flow dichroism assessment of peptide orientation in lipid vesicle membrane. Proc. Natl. Acad. Sci. U.S.A. 2002, 99 (24), 15313−15317. (22) Svensson, F. R.; Lincoln, P.; Norden, B.; Esbjorner, E. K. Retinoid chromophores as probes of membrane lipid order. J. Phys. Chem. B 2007, 111 (36), 10839−10848. (23) Asadi, M. Beet-Sugar Handbook; Wiley-Interscience: Hoboken, NJ, 2007. (24) Gao, L. H.; Lipowsky, R.; Shillcock, J. Tension-induced vesicle fusion: pathways and pore dynamics. Soft Matter 2008, 4 (6), 1208− 1214. (25) Zhu, T.; Jiang, Z. Y.; Nurlybaeva, E. M. R.; Sheng, J.; Ma, Y. Q. Effect of Osmotic Stress on Membrane Fusion on Solid Substrate. Langmuir 2013, 29 (21), 6377−6385. (26) Shillcock, J. C.; Lipowsky, R. Tension-induced fusion of bilayer membranes and vesicles. Nat. Mater. 2005, 4 (3), 225−228.
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS
■
REFERENCES
This work was supported by King Abdullah University of Science and Technology (KAUST) (grant KUK-11-008-23), the European Research Council (EC-2008 AdG 227700SUMO), and the Swedish Research Council - Linnaeus grant SUPRA 349-2007-8680.
(1) Macdonald, R. I. Membrane-Fusion Due to Dehydration by Polyethylene-Glycol, Dextran, or Sucrose. Biochemistry 1985, 24 (15), 4058−4066. (2) Hui, S. W.; Kuhl, T. L.; Guo, Y. Q.; Israelachvili, J. Use of poly(ethylene glycol) to control cell aggregation and fusion. Colloids Surf., B 1999, 14 (1−4), 213−222. (3) Arnold, K.; Ohki, S.; Krumbiegel, M. Interaction of Dextran Sulfate with Phospholipid Surfaces and Liposome Aggregation and Fusion. Chem. Phys. Lipids 1990, 55 (3), 301−307. (4) Wilschut, J.; Hoekstra, D. Membrane-Fusion - from Liposomes to Biological-Membranes. Trends Biochem. Sci. 1984, 9 (11), 479−483. (5) Yang, L.; Huang, H. W. Observation of a membrane fusion intermediate structure. Science 2002, 297 (5588), 1877−1879. (6) Heuvingh, J.; Pincet, F.; Cribier, S. Hemifusion and fusion of giant vesicles induced by reduction of inter-membrane distance. Eur. Phys. J. E 2004, 14 (3), 269−276. (7) Stengel, G.; Zahn, R.; Hook, F. DNA-induced programmable fusion of phospholipid vesicles. J. Am. Chem. Soc. 2007, 129 (31), 9584−+. (8) Hadorn, M.; Hotz, P. E. DNA-Mediated Self-Assembly of Artificial Vesicles. PLoS One 2010, 5, (3), doi: 10.1371/journal.pone.0009886. (9) Bernard, A. L.; Guedeau-Boudeville, M. A.; Marchi-Artzner, V.; Gulik-Krzywicki, T.; di Meglio, J. M.; Jullien, L. Shear-induced permeation and fusion of lipid vesicles. J. Colloid Interface Sci. 2005, 287 (1), 298−306. (10) de Haas, K. H.; Blom, C.; van den Ende, D.; Duits, M. H. G.; Mellema, J. Deformation of giant lipid bilayer vesicles in shear flow. Phys. Rev. E 1997, 56 (6), 7132−7137. (11) Marmottant, P.; Biben, T.; Hilgenfeldt, S. Deformation and rupture of lipid vesicles in the strong shear flow generated by ultrasound-driven microbubbles. Proc. R. Soc. London, Sect. A 2008, 464 (2095), 1781−1800. (12) Sekine, Y.; Abe, K.; Shimizu, A.; Sasaki, Y.; Sawada, S.; Akiyoshi, K. Shear flow-induced nanotubulation of surface-immobilized liposomes. RSC Adv. 2012, 2 (7), 2682−2684. (13) Chernomordik, L.; Kozlov, M. M.; Zimmerberg, J. Lipids in Biological Membrane-Fusion. J. Membr. Biol. 1995, 146 (1), 1−14. (14) Churchward, M. A.; Rogasevskaia, T.; Brandman, D. M.; Khosravani, H.; Nava, P.; Atkinson, J. K.; Coorssen, J. R. Specific lipids supply critical negative spontaneous curvature - An essential component of native Ca2+-triggered membrane fusion. Biophys. J. 2008, 94 (10), 3976−3986. (15) Szule, J. A.; Fuller, N. L.; Rand, R. P. The effects of acyl chain length and saturation of diacylglycerols and phosphatidylcholines on membrane monolayer curvature. Biophys. J. 2002, 83 (2), 977−984. (16) Düzgüneş, N.; Faneca, H.; Lima, M. P. Methods to Monitor Liposome Fusion, Permeability, and Interaction with Cells. In Liposomes; Weissig, V., Ed. Humana Press: New York, 2010; Vol. 606, pp 209−232. (17) Wada, A. Chain Regularity and Flow Dichroism of Deoxyribonucleic Acids in Solution. Biopolymers 1964, 2 (4), 361− 380. (18) Caesar, C. E. B.; Esbjorner, E. K.; Lincoln, P.; Norden, B. Assigning Membrane Binding Geometry of Cytochrome c by Polarized Light Spectroscopy. Biophys. J. 2009, 96 (8), 3399−3411. 4878
dx.doi.org/10.1021/la404857r | Langmuir 2014, 30, 4875−4878
