NOVEMBER/DECEMBER 2005 Volume 16, Number 6 © Copyright 2005 by the American Chemical Society
COMMUNICATIONS Lipid Phase Control of DNA Delivery Rumiana Koynova,*,† Li Wang, Yury Tarahovsky,‡ and Robert C. MacDonald Biochemistry, Molecular & Cell Biology, Northwestern University, Evanston, Illinois 60208. Received July 27, 2005; Revised Manuscript Received October 5, 2005
Cationic lipids form nanoscale complexes (lipoplexes) with polyanionic DNA and can be utilized to deliver DNA to cells for transfection. Here we report the correlation between delivery efficiency of these DNA carriers and the mesomorphic phases they form when interacting with anionic membrane lipids. Specifically, formulations that are particularly effective DNA carriers form phases of highest negative interfacial curvature when mixed with anionic lipids, whereas less effective formulations form phases of lower curvature. Structural evolution of the carrier lipid/DNA complexes upon interaction with cellular lipids is hence suggested as a controlling factor in lipid-mediated DNA delivery. A strategy for optimizing lipofection is deduced. The behavior of a highly effective lipoplex formulation, DOTAP/DOPE, is found to conform to this “efficiency formula”.
Efficient delivery of genetic material into biological systems is needed for tasks of utmost importance in laboratory and clinic, such as gene transfection and gene silencing. Synthetic cationic lipids can be used as DNA carriers and are now considered the most promising nonviral gene carriers (1). They form complexes (lipoplexes) with polyanionic DNA. Understanding the mechanism of lipid-mediated DNA delivery (lipofection) is of paramount significance for their effective application, as well as for rational design and synthesis of novel cationic lipid compounds that are promising for superior gene delivery. Here we advance the concept that the controlling factor in lipofection is the structural evolution of lipoplexes upon interaction with cellular (anionic) lipids. * Corresponding author. E-mail:
[email protected]. † Associate member of the Institute of Biophysics, Bulgarian Academy of Sciences. ‡ Current address: Institute of Theoretical and Experimental Biophysics, 142290, Pushchino, Russia.
The unbinding of DNA from a cationic lipid carrier when the lipoplex gets inside the cell has been identified as one of the key steps in lipofection. According to the current understanding, the unbinding is a result of charge neutralization by cellular anionic lipids; indeed, experiments have revealed that addition of negatively charged liposomes to lipoplexes results in dissociation of DNA from the lipid (2-6). A set of significant recent findings suggest that the structure of cationic lipid carriers changes dramatically upon interaction with cellular lipids, and furthermore that such changes may critically affect the delivery efficiency. Hydrated lipids are known for their ability to form an impressive variety of polymorphic and mesomorphic phases, lamellar and nonlamellar, brought about by the optimization of the hydrophobic effect in conjunction with various intra- and intermolecular interactions and geometric packing constraints (7, 8). These phases are generally differentiated by the curvature of their lipidwater interface. The lipid phase state is a key physical
10.1021/bc050226x CCC: $30.25 © 2005 American Chemical Society Published on Web 10/20/2005
1336 Bioconjugate Chem., Vol. 16, No. 6, 2005
Koynova et al.
Figure 1. A. Molecular structures of the cationic phospholipids dioleoyl- and dilauroyl-O-ethylphosphatidylcholine. B. Illustration of lipid monolayers with positive (left) and negative (right) curvature. C. Correlation between the transfection efficiency (red bars) of ethyldilauroylphosphatidylcholine (EDLPC)/ethyldioleoylphosphatidylcholine (EDOPC) lipoplexes, as reported in ref 19 (quantified by expression of β-galactosidase in human umbilical artery endothelial cells) and the phase structure of their mixtures with the anionic lipid cardiolipin (CL). Cartoons of the phases are arranged vertically to give a qualitative notion about the increase of the interfacial curvature. Top row: diffraction patterns of mixtures of EDLPC/EDOPC lipoplex dispersions of different compositions with CL dispersions. The charge ratio of the cationic/anionic lipids in the samples is 1:1. Similar patterns were recorded also for mixtures of EDLPC/EDOPC with the anionic phosphatidylglycerol.
attribute because it mediates virtually all properties of lipid dispersions. Correlations between the mesomorphic phase state of the lipoplexes (lamellar vs inverted hexagonal) and their DNA delivery activity have been sought, but so far there is no clear consensus as to whether the hexagonal phase per se is beneficial for transfection (9-14). It was recently reported that mixtures of cationic lipids with anionic lipids of the type found in cell membranes are unusually prone to form nonlamellar phases, even when the pure components form only the lamellar phase: even small amounts of anionic lipid added to certain cationic lipids generate virtually the entire panoply of possible lipid arrays (15-17). Since escape of DNA from the lipoplexes must involve neutralization of cationic lipid by cellular anionic lipids, most likely by fusion of cell membranes with lipoplexes, a wide variety of nonlamellar arrays can potentially appear in treated cells. Could the efficiency of transfection depend on which of those phases is preferred? This now seems very likely: in a previous study on this topic we demonstrated that DNA release measured after treating lipoplexes with anionic membrane lipids varies widely for the different cationic and anionic lipid combinations and, remarkably, showed an unambiguous correlation between the DNA unbinding and the phases formed by the mixtures of the anionic lipids with the cationic lipid of the lipoplexes (18).
Specifically, the higher the negative interfacial curvature (polar/nonpolar interface curved toward the water, see Figure 1B) of the phase assumed by the cationic/anionic lipid mixture, the faster and more complete the release of DNA. We thus hypothesized that the structural evolution of lipoplexes upon interaction with cellular (anionic) lipids is a controlling factor in lipid-mediated DNA delivery. To test this suggestion, we compared the lipofection efficiency of certain lipoplexes with their phase behavior in mixtures with several anionic membrane lipids. In our study we used lipoplex formulations comprising mixtures of the dilauroyl and dioleoyl derivatives of cationic ethylphosphatidylcholine (2), EDLPC/EDOPC (Figure 1A). It was discovered recently that delivery efficiency of such lipoplex formulations could be strongly modulated by merely varying the ratio of the two components (19). For example, the EDLPC/EDOPC mixture delivered DNA into cells more than 30-fold more efficiently at certain intermediate compositions than either compound separately (19). Given this astonishing synergy, we applied synchrotron X-ray diffraction to study the structural correlates of the lipoplex formulations and their mixtures with anionic lipids. The widely explored lipid transfection formulation, DOTAP/DOPE 1:1 (both from Avanti Polar Lipids) was also examined. EDLPC/EDOPC cationic lipid dispersions were prepared by mixing lipids as chloroform solutions, with
Communications
Bioconjugate Chem., Vol. 16, No. 6, 2005 1337
Figure 2. Kinetics of phase changes in the EDLPC/EDOPC lipoplexes of the highest activity (6:4 wt ratio (19)), upon addition of the anionic lipid DOPS, in the time interval 10 min to 30 min after the addition: bilayer cubic Pn3m phase forms initially in parallel to the lamellar phase; later, micellar cubic Fd3m could be distinguished. The charge ratio of the cationic/anionic lipids in the sample is 1:1.
subsequent chloroform evaporation under vacuum and overnight hydration in PBS (17). Lipoplexes were prepared by adding aqueous solution of herring sperm DNA (Invitrogen, Carlsbad, CA) to the cationic lipid dispersion at isoelectric ratio, vortexing, and overnight equilibration (17, 20). Anionic lipid dispersions were prepared from dioleoylphosphatidylserine (DOPS), dioleoylphosphatidylglycerol (DOPG), or tetraoleoylcardiolipin (CL) (Avanti Polar Lipids), as described in ref 18, and added to the lipoplexes. These mixed dispersions were then equilibrated for 1-2 days, filled into glass capillaries, and flame-sealed. For kinetic measurements, lipoplexes and anionic lipid dispersions were mixed and immediately mounted on the X-ray beamline for measurement. Smallangle X-ray diffraction (SAXD) measurements were performed at Argonne National Laboratory, Advanced Photon Source, BioCAT (beamline 18-ID) and DND-CAT (beamline 5-IDD), using 12 keV X-rays. A Linkam thermal stage (Linkam Sci Instruments, Surrey, England) provided temperature control. Linear heating scans were performed at rates of 0.8-2 °C/min. Exposure times were typically ∼0.5-1 s. Both the lipid dispersions and the lipid/DNA complexes of the EDLPC/EDOPC mixture arrange into lamellar arrays at all compositions; the same is valid for the anionic lipid dispersions (not illustrated). When anionic lipids interact with the lipoplexes however, precisely those lipoplex compositions that exhibit particularly high transfection efficiency, EDLPC/EDOPC 60:40 w/w, formed the inverted micellar cubic phase, Fd3m, while either components and mixtures of other compositions all formed phases of lower curvature (Figure 1C), as revealed by our small-angle X-ray diffraction data. The micellar cubic phase exhibits even higher interfacial curvature than the more familiar inverted hexagonal phase (8). In the case of the anionic lipid cardiolipin (CL), in a sample of 1:1 cationic/anionic lipid charge ratio, this phase is identified by 11 diffraction maxima with reciprocal spacings fitting the ratio: x8: x11: x12: x16: x24: x32: x35: x36: x48: x51: x56, characteristic for the cubic aspect #15 (21). The anionic lipid dioleoylphosphatidylglycerol similarly formed micellar cubic phase when mixed with the most efficient EDLPC/EDOPC 6:4 formulation (not illustrated). Another anionic membrane lipid, phosphatidylserine, was found to form the micellar cubic phase transiently when mixed extempore with the
EDLPC/EDOPC 6:4 formulation, as revealed by our kinetic synchrotron X-ray experiments (Figure 2). The pure cationic lipid components (EDLPC, EDOPC) and their mixtures of other, less effective compositions, formed phases of lower curvature: inverted hexagonal, bilayer cubic, and/or coexisting lamellar (Figure 1C). Thus, transfection outcome appears to corellate with the propensity of the cationic/anionic lipid mixture to evolve into highly curved mesomorphic structure when they interact. According to our earlier study, DNA unbinding after treating lipoplexes with anionic lipids also unambiguously correlated with the disposition of the cationic/anionic lipid mixtures to form assemblies of highly curved topologies (18). The propensity of lipid bilayers to form nonlamellar phases has been described by the concept of bilayer “frustration” brought about by the imbalance of forces in the bilayer and which imposes a nonzero intrinsic curvature of the two opposing monolayers (8, 22, 23). In terms of this concept, those cationic lipid formulations, which tend to easily transform into curved nonlamellar arrays when mixed with anionic lipid, presumably retain higher degree of “frustration”. The stored curvature elastic energy in a “frustrated” bilayer seems to be comparable to the cationic lipid-DNA binding energy: the same magnitude of ∼1-2kBT has been estimated for both the binding free energy of lipoplexes (24-26) and the curvature elastic energy stored in a flat DOPE-dominated membrane (27, 28). The release of stored curvature elastic energy upon lamellar-nonlamellar phase transition and the balance between these energy terms could therefore play a significant role in lipoplex-membrane interactions and DNA unbinding energetics. This result corroborates our hypothesis that structural evolution of lipoplexes upon interaction with cellular lipids is the controlling factor in lipid-mediated DNA delivery. Here we recall that, for effective delivery, DNA must be retained until the lipoplex is inside the cell; thus, initial lamellar lipoplex structures are likely responsible for the successful import of DNA into cells. Juxtaposing our results, a strategy for optimizing lipofection can be deduced: lamellar lipoplex formulations with compositions close to the lamellar-nonlamellar phase boundary, which could easily undergo phase transition upon mixing with cellular anionic lipids (Figure 3), are predicted to be especially efficient because they would be equally successful in transporting DNA into
1338 Bioconjugate Chem., Vol. 16, No. 6, 2005
Koynova et al.
Figure 3. Hypothetical scheme for efficiency of DNA delivery: Lamellar lipoplexes residing close to the lamellar-nonlamellar phase transition form nonlamellar structures upon contacting the cellular lipids and thus exhibit optimum delivery efficiency (alternatively, lipoplexes retaining their lamellar structure after contacting the endosomal membrane may not efficiently release DNA; on the other hand, nonlamellar lipoplexes may release DNA prematurely).
of minor amounts of anionic lipids and/or elevation of the temperature to 37 °C. ACKNOWLEDGMENT
Supported by NIH Grants GM52329 and GM57305. Synchrotron X-ray measurements were performed at BioCAT and DND-CAT at APS, Argonne National Laboratory. BioCAT is NIH-supported, Grant RR08630. DNDCAT is supported by E.I. DuPont de Nemours & Co., The Dow Chemical Company, NSF Grant DMR-9304725, and the State of Illinois through the Department of Commerce and the Board of Higher Education Grant IBHE HECA NWU 96. Use of APS was supported by the U.S. DOE, Basic Energy Sciences, Office of Energy Research, Contract No. W-31-102-Eng-38. Figure 4. Small-angle X-ray diffraction patterns of DOTAP/ DOPE (1:1 mol/mol) dispersion recorded during heating at 1 °C/ min (exposure time 1 s): at low temperatures, the sample forms lamellar phase (with minor traces of cubic). Upon heating, hexagonal phase starts to form at ∼39-40 °C, which coexists with the lamellar phase. Inset: Upon addition of 5% of the anionic membrane lipid DOPS, cubic phase dominates in the dispersion at room and physiological temperatures.
cells (since lamellar phase optimally protects DNA (18)) and in releasing it subsequently. Indeed, our X-ray diffraction experiments have now provided explicit support for this “efficiency formula”: we found that one of the very effective and widely explored lipid transfection formulations, DOTAP/DOPE 1:1 (mol/ mol), exhibits precisely this behavior: Samples prepared at low (below physiological) temperature form the lamellar phase, whereas upon heating, lamellar f inverted hexagonal phase transition begins just above physiological temperature, at ∼39-40 °C (Figure 4). Addition of a minor proportion (e5%) of the anionic lipid phosphatidylserine was enough to convert the lipoplex structure into a predominantly cubic bicontinuous Pn3m phase at physiological temperature (Figure 4, inset). Thus, after uptake into the cell, these lipoplexes would very easily be converted into nonlamellar arrays that allow the DNA release, which is the likely reason for their effectiveness. Therefore, rational design of cationic lipid formulations residing close to the lamellar-nonlamellar phase transition may bring about superior delivery efficiency, especially if that phase transition is facilitated by acquisition
LITERATURE CITED (1) Felgner, P. L., and Ringold, G. M. (1989) Cationic liposomemediated transfection. Nature 337, 387-388. (2) MacDonald, R. C., Ashley, G. W., Shida, M. M., Rakhmanova, V. A., Tarahovsky, Y. S., Pantazatos, D. P., Kennedy, M. T., Pozharski, E. V., Baker, K. A., Jones, R. D., Rosenzweig, H. S., Choi, K. L., Qiu, R. Z., and McIntosh, T. J. (1999) Physical and biological properties of cationic triesters of phosphatidylcholine. Biophys. J. 77, 2612-2629. (3) Xu, Y. H., and Szoka, F. C. (1996) Mechanism of DNA release from cationic liposome/DNA complexes used in cell transfection. Biochemistry 35, 5616-5623. (4) Zelphati, O., and Szoka, F. C. (1996) Mechanism of oligonucleotide release from cationic liposomes. Proc. Natl. Acad. Sci. U.S.A. 93, 11493-11498. (5) Tarahovsky, Y., Koynova, R., and MacDonald, R. C. (2004) Efficiency of anionic lipids to release DNA from lipoplexes correlates with the interfacial curvature of the mesomorphic phases of cationic/anionic lipid mixtures. Biophys. J. 86, 37A. (6) Ashley, G. W., Shida, M. M., Qiu, R., Lahiri, M. K., Levisay, P. C., Jones, R. D., Baker, K. A., and MacDonald, R. C. (1996) Phosphatidylcholinium compounds: A new class of cationic phospholipids with transfection activin and unusual physical properties (Abstract). Biophys. J. 70, A88. (7) Luzzati, V., and Tardieu, A. (1974) Lipid Phases - Structure and Structural Transitions. Annu. Rev. Phys. Chem. 25, 7994. (8) Seddon, J. M., and Templer, R. H. (1995) In Handbook of Biological Physics. (Lipowsky, R., and Sackmann, E., Eds.) pp 97-160, Elsevier Science, Amsterdam.
Bioconjugate Chem., Vol. 16, No. 6, 2005 1339
Communications (9) Koltover, I., Salditt, T., Radler, J. O., and Safinya, C. R. (1998) An inverted hexagonal phase of cationic liposome-DNA complexes related to DNA release and delivery. Science 281, 78-81. (10) Smisterova, J., Wagenaar, A., Stuart, M. C. A., Polushkin, E., ten Brinke, G., Hulst, R., Engberts, J. B. F. N., and Hoekstra, D. (2001) Molecular shape of the cationic lipid controls the structure of cationic lipid/dioleylphosphatidylethanolamine-DNA complexes and the efficiency of gene delivery. J. Biol. Chem. 276, 47615-47622. (11) Rakhmanova, V. A., McIntosh, T. J., and MacDonald, R. C. (2000) Effects of dioleoylphosphatidylethanolamine on the activity and structure of O-alkyl phosphatidylcholine-DNA transfection complexes. Cell. Mol. Biol. Lett. 5, 51-65. (12) Simberg, D., Danino, D., Talmon, Y., Minsky, A., Ferrari, M. E., Wheeler, C. J., and Barenholz, Y. (2001) Phase behavior, DNA ordering, and size instability of cationic lipoplexes - Relevance to optimal transfection activity. J. Biol. Chem. 276, 47453-47459. (13) Congiu, A., Pozzi, D., Esposito, C., Castellano, C., and Mossa, G. (2004) Correlation between structure and transfection efficiency: a study of DC-Chol-DOPE/DNA complexes. Colloids Surfaces B: Biointerfaces 36, 43-48. (14) Ross, P. C., Hensen, M. L., Supabphol, R., and Hui, S. W. (1998) Multilamellar cationic liposomes are efficient vectors for in vitro gene transfer in serum. J. Liposome Res. 8, 499520. (15) Tarahovsky, Y. S., Arsenault, A. L., MacDonald, R. C., McIntosh, T. J., and Epand, R. M. (2000) Electrostatic control of phospholipid polymorphism. Biophys. J. 79, 3193-3200. (16) Lewis, R. N. A. H., and McElhaney, R. N. (2000) Surface charge markedly attenuates the nonlamellar phase-forming propensities of lipid bilayer membranes: Calorimetric and P-31-nuclear magnetic resonance studies of mixtures of cationic, anionic, and zwitterionic lipids. Biophys. J. 79, 1455-1464. (17) Koynova, R., and MacDonald, R. C. (2003) Mixtures of cationic lipid O-ethylphosphatidylcholine with membrane lipids and DNA: Phase diagrams. Biophys. J. 85, 2449-2465.
(18) Tarahovsky, Y. S., Koynova, R., and MacDonald, R. C. (2004) DNA release from lipoplexes by anionic lipids: Correlation with lipid mesomorphism, interfacial curvature, and membrane fusion. Biophys. J. 87, 1054-1064. (19) Wang, L., and MacDonald, R. C. (2004) New strategy for transfection: mixtures of medium-chain and long-chain cationic lipids synergistically enhance transfection. Gene Ther. 11, 1358-1362. (20) Koynova, R., and MacDonald, R. C. (2004) Columnar DNA superlattices in lamellar o-ethylphosphatidylcholine lipoplexes: Mechanism of the gel-liquid crystalline lipid phase transition. Nano Lett. 4, 1475-1479. (21) Kasper, J. S., and Lonsdale, K. (1985) International Tables for X-ray Crystallography, Riedel Publishing Company, Dordrecht, The Netherlands. (22) Gruner, S. M. (1985) Intrinsic curvature hypothesis for biomembrane lipid-composition - A role for nonbilayer lipids. Proc. Natl. Acad. Sci. U.S.A. 82, 3665-3669. (23) Anderson, D. M., Gruner, S. M., and Leibler, S. (1988) Geometrical aspects of the frustration in the cubic phases of lyotropic liquid-crystals. Proc. Natl. Acad. Sci. U.S.A. 85, 5364-5368. (24) Bruinsma, R. (1998) Electrostatics of DNA cationic lipid complexes: isoelectric instability. Eur. Phys. J. B 4, 75-88. (25) Pozharski, E., and MacDonald, R. C. (2002) Thermodynamics of cationic lipid-DNA complex formation as studied by isothermal titration calorimetry. Biophys. J. 83, 556-565. (26) Pozharski, E., and MacDonald, R. C. (2003) Lipoplex thermodynamics: Determination of DNA-cationic lipoid interaction energies. Biophys. J. 85, 3969-3978. (27) Attard, G. S., Templer, R. H., Smith, W. S., Hunt, A. N., and Jackowski, S. (2000) Modulation of CTP: phosphocholine cytidylyltransferase by membrane curvature elastic stress. Proc. Natl. Acad. Sci. U.S.A. 97, 9032-9036. (28) Botelho, A. V., Gibson, N. J., Thurmond, R. L., Wang, Y., and Brown, M. F. (2002) Conformational energetics of rhodopsin modulated by nonlamellar-forming lipids. Biochemistry 41, 6354-6368.
BC050226X
