9598
Langmuir 2008, 24, 9598-9606
Brewster Angle Microscopy and PMIRRAS Study of DNA Interactions with BGTC, a Cationic Lipid Used for Gene Transfer Sabine Castano,*,† Brigitte Delord,‡ Annie Fe´vrier,‡ Jean-Marie Lehn,§ Pierre Lehn,|,⊥ and Bernard Desbat† UMR 5248 CBMN, CNRS-UniVersite´ Bordeaux 1-ENITAB, IECB, 2 rue Robert Escarpit, 33607 Pessac, France, UPR 8641 CRPP, rue du Dr A. Schweitzer, 33600 Pessac, France, Laboratoire de Chimie des Interactions Mole´culaires, Colle`ge de France, Paris, France, INSERM U613, Brest F-29200, France, and UniVersite´ de Bretagne Occidentale UBO, Brest F-29200, France ReceiVed NoVember 8, 2007. ReVised Manuscript ReceiVed May 26, 2008 The lipid bis(guanidinium)-tris(2-aminoethyl)amine-cholesterol (BGTC) is a cationic cholesterol derivative bearing guanidinium polar headgroups which displays high transfection efficiency in Vitro and in ViVo when used alone or formulated as liposomes with the neutral colipid 1,2-di-[cis-9-octadecenoyl]-sn-glycero-3-phosphoethanolamine (DOPE). Since transfection may be related to the structural and physicochemical properties of the self-assembled supramolecular lipid-DNA complexes, we used the Langmuir monolayer technique coupled with Brewster angle microscopy (BAM) and polarization modulation infrared reflection absorption spectroscopy (PMIRRAS) to investigate DNA-BGTC and DNA-BGTC/DOPE interactions at the air/water interface. We herein show that BGTC forms stable monolayers at the air/water interface. When DNA is injected into the subphase, it adsorbs to BGTC at 20 mN/m. Whathever the (+/-) charge ratio of the complexes used, defined as the ratio of positive charges of BGTC in the monolayer versus negative charges of DNA injected in the subphase, the DNA interacts with the cationic lipid and forms either an incomplete (no constituent in excess) or a complete (DNA in excess) monolayer of oriented double strands parallel to the lipid monolayer plan. We also show that, under a homogeneous BGTC/DOPE (3/2) monolayer at 20 mN/m, DNA adsorbs homogeneously to form an organized but incomplete layer whatever the charge ratio used (DNA in default or in excess). Compression beyond the collapse of these mixed DNA-BGTC/DOPE systems leads to the formation of dense DNA monolayers under an asymmetric lipid bilayer with a bottom layer of BGTC in contact with DNA and a top layer mainly constituted of DOPE. These results allow a better understanding of the mechanisms underlying the formation of the supramolecular BGTC-DNA complexes efficient for gene transfection.
Introduction Gene therapy is believed by many to become the promising therapy of the 21st century because it aims to suppress the cause of a genetic disease rather than its symptoms by delivering a functional version of the disease-causing gene into the cell nucleus.1-5 Its clinical success is strongly dependent on the development of efficient and safe gene delivery systems termed “vectors”. Over the past decade, various gene delivery systems have been investigated for human gene therapy approaches. They are mainly of two types: viral and nonviral vectors. Even if the transfection efficiency of viral vectors is in general superior to that of the nonviral ones, many problems (such as immunogenicity, insertional mutagenesis, toxicity) as well as practical issues of bulk production and quality control are associated with the use of recombinant viruses.6-11 Therefore, various nonviral gene * To whom correspondence should be addressed. Telephone: +33 540003046. Fax: +33 540003073. E-mail:
[email protected]. † UMR 5248 CBMN, CNRS-Universite´ Bordeaux 1-ENITAB. ‡ UPR 8641 CRPP. § Colle`ge de France. | INSERM. ⊥ Universite´ de Bretagne Occidentale UBO.
(1) Mulligan, R. C. Science 1993, 260, 926. (2) Verma, I. M.; Somina, M. Nature 1997, 389, 239. (3) Anderson, W. F. Nature 1998, 392, 25. (4) Yla-Herttuala, S.; Martin, J. F. Lancet 2000, 355, 213. (5) Kumar, V. V.; Singh, R. S.; Chaudhuri, A. Curr. Med. Chem. 2003, 10, 1297. (6) Hollon, T. Nat. Med. 2000, 6, 6. (7) Schroder, A. R.; Shinn, P.; Chen, H.; Berry, C.; Ecker, J. R.; Bushman, F. Cell 2002, 110, 521. (8) Li, Z.; Dullmann, J.; Schiedlmeier, B.; Schmidt, M.; von Kalle, C.; Meyer, J.; Forster, M.; Stocking, C.; Wahlers, A.; Frank, O.; Ostertag, W.; Kuhlcke, K.; Eckert, H. G.; Fehse, B.; Baum, C. Science 2002, 296, 497.
delivery systems have also been developed.12-18 Among these synthetic vectors, numerous cationic lipids have been synthetized and some of them displayed significant transfection efficiency not only in in Vitro and in ViVo experiments but also in the clinical setting for treatment of diseases such as cancer and cystic fibrosis.19-24 However, despite some positive results, it is at present clear that their transfection efficiency has to be increased in order for them to become real therapeutic agents. (9) Woods, N. B.; Muessig, A.; Schmidt, M.; Flygare, J.; Olsson, K.; Salmon, P.; Trono, D.; von Kalle, C.; Karlsson, S. Blood 2003, 101, 1284. (10) Hacein-Bey-Abina, S.; Von Kalle, C.; Schmidt, M.; McCormack, M. P.; Wulffraat, N.; Leboulch, P.; Lim, A.; Osborne, C. S.; Pawliuk, R.; Morillon, E.; Sorensen, R.; Forster, A.; Fraser, P.; Cohen, J. I.; de Saint Basile, G.; Alexander, I.; Wintergerst, U.; Frebourg, T.; Aurias, A.; Stoppa-Lyonnet, D.; Romana, S.; Radford-Weiss, I.; Gross, F.; Valensi, F.; Delabesse, E.; Macintyre, E.; Sigaux, F.; Soulier, J.; Leiva, L. E.; Wissler, M.; Prinz, C.; Rabbitts, T. H.; Le Deist, F.; Fischer, A.; Cavazzana-Calvo, M. Science 2003, 302, 415. (11) Woods, N. B.; Bottero, V.; Schmidt, M.; von Kalle, C.; Verma, I. M. Nature 2006, 440, 1123. (12) Behr, J. P. Acc. Chem. Res. 1993, 26, 274. (13) Gao, X.; Huang, L. Gene Ther. 1995, 2, 710. (14) Ledley, F. D. Hum. Gene Ther. 1995, 6, 1129. (15) Lehn, P.; Fabrega, S.; Oudrhiri, N.; Navarro, J. AdV. Drug DeliVery ReV. 1998, 30, 5. (16) Miller, A. D. Angew. Chem., Int. Ed. 1998, 37, 1768. (17) Wagner, E. In NonViral Vectors for Gene Therapy; Huang, L., Hung, M. C., Wagner, E., Eds.; Academic Press: San Diego, CA, 1999; p 207. (18) Martin, B.; Aissaoui, A.; Sainlos, M.; Oudrhiri, N.; Hauchecorne, M.; Vigneron, J. P.; Lehn, J. M.; Lehn, P. Gene Ther. Mol. Biol. 2003, 7, 273. (19) Roth, J. A.; Cristiano, R. J. J. Natl. Cancer Inst. 1997, 89, 21. (20) Hersh, E. M.; Stopeck, A. T. In Self-assembling Complexes for Gene DeliVery; Kabanov, A. V., Felgner, P. L., Seymour, L. W., Eds.; John Wiley and Sons: Chichester, U.K., 1998; p 421. (21) Alton, E. W. F. W.; Stern, M.; Farley, R.; Jaffe, A.; Chadwick, S. L.; Phillips, J.; Davies, J.; Smith, S. N.; Browning, J.; Davies, M. G.; Hodson, M. E.; Durham, S. R.; Li, D.; Jeffery, P. K.; Scallan, M.; Balfour, R.; Eastman, S. J.; Cheng, S. H.; Smith, A. E.; Meeker, D.; Geddes, D. M. Lancet 1999, 353, 947. (22) Boucher, R. C. J. Clin. InVest. 1999, 103, 441.
10.1021/la703491r CCC: $40.75 2008 American Chemical Society Published on Web 07/30/2008
DNA-BGTC Interactions Scheme 1. Chemical Structure of BGTC
That led some of us (P.L. and J.-M.L.) to develop novel cationic derivatives of cholesterol characterized by guanidinium polar headgroups.25 The rationale (at the molecular level) underlying the approach was that the positively charged guanidinium group (i) naturally occurs in arginine residues which play a key role in DNA-binding proteins and (ii) remains charged over a wide range of pH due to its high pKa value;26 thus, the guanidinium group appeared well-suited for interaction with the negatively charged phosphate residues of polynucleotides. It was indeed shown that the guanidinium group and phosphate can establish electrostatic interactions with the formation of a pair of parallel hydrogen bonds.27 Furthermore, it was also demonstrated that the guanidinium group can develop hydrogen bonding with a nucleic-base-like guanine.28,29 On the other hand, cholesterol has been shown to facilitate cellular uptake of oligonucleotides and polar drugs.30-34 Therefore, it was reasoned that the combination of such two subunits might generate efficient potential vectors for gene transfection. Among the synthetized guanidinium cholesterol derivatives, bis(guanidinium)- tris(2-aminoethyl)amine-cholesterol (BGTC; Scheme 1) was found to show efficient transfection of a wide variety of mammalian cell lines in Vitro when used alone or formulated as cationic liposomes with 1,2-di-[cis-9-octadecenoyl]-sn-glycero-3-phosphoethanolamine (DOPE).25 With a view to treatment of cystic fibrosis and other lung diseases, BGTC lipoplexes were found to be efficient for gene transfer into airway epithelium cells in Vitro, into the mouse airway epithelium in ViVo via direct intratracheal administration or intranasal instillation,35,36 and also into fetal sheep airways in utero.37 In addition, BGTC/DOPE-DNA formulations showed promising results for (23) Griesenbach, U.; Geddes, D. M.; Alton, E. W. F. W. In NonViral Vectors for Gene Therapy; Huang, L., Hung, M. C., Wagner, E., Eds.; Academic Press: San Diego, CA, 1999; 337. (24) Davies, J. C.; Geddes, D. M.; Alton, E. W. J. Gene Med. 2001, 3, 409. (25) Vigneron, J. P.; Oudrhiri, N.; Fauquet, M.; Vergely, L.; Bradley, J. C.; Basseville, M.; Lehn, P.; Lehn, J. M. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 9682. (26) Wirth, T. H.; Davidson, N. J. Am. Chem. Soc. 1964, 86, 4325. (27) Cotton, F. A.; Day, V. W.; Hazen, E. E., Jr.; Larsen, S. J. Am. Chem. Soc. 1973, 95, 4834. (28) Rould, M. A.; Perona, J. J.; Steitz, T. A. Nature (London) 1991, 352, 213. (29) Puglisi, J. D.; Tan, R.; Calman, B. J.; Frankel, A. D.; Williamson, J. R. Science 1992, 257, 76. (30) Letsinger, R. L.; Zhang, G.; Sun, D. K.; Ikeuchi, T.; Sarin, P. S. Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 6553. (31) Boutorin, A. S.; Gus’kova, L. V.; Ivanova, E. M.; Kobetz, N. D.; Zarytova, V. F.; Ryte, A. S.; Yurchenko, L. V.; Vlassov, V. V. FEBS Lett. 1989, 254, 129. (32) Bodor, N.; Prokai, L.; Wu, W. M.; Farag, H.; Jonalagadda, S.; Kawamura, M.; Simpkins, J. Science 1992, 257, 1698. (33) Koster, F.; Schroder, A.; Finas, D.; Hauser, C.; Diedrich, K.; Felberbaum, R. Int. J. Mol. Med. 2006, 18, 1201. (34) Cheng, K.; Ye, Z.; Guntaka, R. V.; Mahato, R. I. J. Pharmacol. Exp. Ther. 2006, 317, 797. (35) Oudrhiri, N.; Vigneron, J. P.; Peuchmaur, M.; Leclerc, T.; Lehn, J. M.; Lehn, P. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 1651. (36) Pitard, B.; Oudrhiri, N.; Lambert, O.; Vivien, E.; Masson, C.; Wetzer, B.; Hauchecorne, M.; Scherman, D.; Rigaud, J. L.; Vigneron, J. P.; Lehn, J. M.; Lehn, P. J. Gene Med. 2001, 3, 478. (37) Luton, D.; Oudrhiri, N.; de Lagausie, P.; Aissaoui, A.; Hauchecorne, M.; Julia, S.; Oury, J. F.; Aigrain, Y.; Peuchmaur, M.; Vigneron, J. P.; Lehn, J. M.; Lehn, P. J. Gene Med. 2004, 6, 328.
Langmuir, Vol. 24, No. 17, 2008 9599
aerosol gene delivery.38 Finally, efficient BGTC-mediated gene transfection was also observed in a murine model of pancreatic peritoneal carcinomatosis.39,40 Despite all these promising results, it is noteworthy that only very few investigations36,41 were performed to identify and characterize the basic features of the DNA-BGTC/lipid interactions even if it clearly appears that the transfection efficiency of this cationic lipid system is directly related to the structural and physicochemical properties of the self-assembled supramolecular systems. Therefore, in this work, we used the monolayer Langmuir technique, which already demonstrated its usefulness to analyze DNA-cationic lipid interactions,42-47 coupled with Brewster angle microscopy (BAM) and polarization modulation infrared reflection absorption spectroscopy (PMIRRAS) to investigate DNA-BGTC and DNA-BGTC/DOPE interactions at the air/ water interface.
Materials and Methods Materials. The cationic lipid BGTC was synthetized by the J.-M. Lehn group as previously described.25 1,2-Di-[cis-9-octadecenoyl]sn-glycero-3-phosphoethanolamine (DOPE) and calf thymus DNA were purchased from Sigma, France. Chloroform (CHCl3) was purchased from Sigma-Aldrich, Steinheim, Germany. Long DNA strands were fractionnized by sonication to obtain double strands of ≈200 base pairs. The ultrapure water subphase was obtained from a MilliQ (Millipore, Molsheim, France) system with a nominal resistivity of 18.2 MΩ · cm. Film Formation and Surface Pressure Measurements. Monolayer experiments were performed on a computer-controlled Langmuir film balance (Nima Technology, Coventry, England). The rectangular trough (V ) 110 cm3, S ) 145 cm2) and the barrier were made of Teflon. The surface pressure (Π) was measured by the Wilhelmy method using a filter paper plate. The trough was filled with a salted solution (NaCl 150 mM) in ultrapure water. The experiments were carried out at 22 ( 2 °C. Pure BGTC films were obtained by deposition of a few microliters of CHCl3 stock solutions at the air/water interface. Mixed BGTC/ DOPE films were obtained by deposition at the surface of a mixture in CHCl3 of both lipids at the defined ratio. After complete evaporation of the solvent (≈15 min), the lipid film was then slowly compressed up to the defined lateral pressure. For DNA interaction experiments, a defined volume of a DNA dispersion in pure water was injected under the compressed lipid monolayer to reach a defined R(+/-) ratio of positive charges of BGTC in the monolayer versus negative charges of DNA injected in the subphase. Kinetics of lateral pressure evolution, PMIRRAS spectra, and BAM pictures were then recorded. PMIRRAS Spectroscopy. PMIRRAS spectra were recorded on a Nicolet Nexus 870 spectrometer equipped with a photovoltaic HgCdTe detector cooled at 77 K. Generally, 600 scans were coadded at a resolution of 8 cm-1 for BGTC, BGTC/DOPE monolayers alone or in interaction with DNA. In short, PMIRRAS combines Fourier (38) Densmore, C. L.; Giddings, T. H.; Waldrep, J. C.; Kinsey, B. M.; Knight, V. J. Gene Med. 1999, 1, 251. (39) Hajri, A.; Wack, S.; Lehn, P.; Vigneron, J. P.; Lehn, J. M. Cancer Gene Ther. 2000, 7, 1393. (40) Hajri, A.; Wack, S.; Lehn, P.; Vigneron, J. P.; Lehn, J. M.; Marescaux, J.; Aprahamian, M. Cancer Gene Ther. 2004, 11, 16. (41) Pitard, B.; Oudrhiri, N.; Vigneron, J.-P.; Hauchecorne, M.; Aguerre, O.; Toury, R.; Airiau, M.; Ramasawmy, R.; Scherman, D.; Crouzet, J.; Lehn, J.-M.; Lehn, P. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 2621. (42) Symietz, C.; Schneider, M.; Brezesinski, G.; Mohwald, H. Macromolecules 2004, 37, 3865. (43) McLoughlin, D.; Dias, R.; Lindman, B.; Cardenas, M.; Nylander, T.; Dawson, K.; Miguel, M.; Langevin, D. Langmuir 2005, 21, 1900. (44) Ca´rdenas, M.; Nylander, T.; Jo¨nsson, B.; Lindman, B. J. Colloid Interface Sci. 2005, 286, 166. (45) Gromelski, S.; Brezesinski, G. Langmuir 2006, 22, 6293. (46) Erokhina, S.; Berzina, T.; Cristofolini, L.; Konovalov, O.; Erokhin, V.; Fontana, M. P. Langmuir 2007, 23, 4414. (47) Cristofolini, L.; Berzina, T.; Erokhina, S.; Konovalov, O.; Erokhin, V. Biomacromolecules 2007, 8, 2270.
9600 Langmuir, Vol. 24, No. 17, 2008
Castano et al.
transform infrared (FT-IR) reflection spectroscopy with fast polarization modulation of the incident beam between the parallel (p) and perpendicular (s) polarization. Two-channel processing of the detected signal makes it possible to obtain the differential reflectance spectrum:
∆R ⁄ R ) (Rp - Rs) ⁄ (Rp) + (Rs)J2
(1)
To remove the contribution of liquid water absorption and the dependence on Bessel functions J2, the monolayer spectra are divided by that of the subphase. With an incidence angle of 75°, transition moments preferentially oriented in the plane of the interface give intense and upward oriented bands, while perpendicular ones give weaker and downward oriented bands.48 Brewster Angle Microscopy (BAM) and Ellipsometry Measurements. The morphology of pure BGTC or mixed BGTC/DOPE layers before and after DNA interaction at the air/water interface were observed using a Brewster angle microscope (NFT BAM2plus, Go¨ttingen, Germany) mounted on the Langmuir trough. The microscope was equipped with a frequency doubled Nd:Yag laser (532 nm, 50 mW), a polarizer, an analyzer, and a CCD camera. The exposure time (ET), depending on the image luminosity, was adjusted to avoid saturation of the camera. The spatial resolution of the Brewster angle microscope was about 2 µm, and the image size was 600 × 450 µm with ×10 magnification lens used. The BAM images are coded in gray level. To determine the thickness of the layer at the surface, we used the calibration procedure of the BAM software that determines the linear function between the reflectance and the gray level. This function is established by comparison between the experimental curve of the gray level as a function of the incidence angle and the Fresnel curve (curve of the reflectance as a function of the incidence angle) that can be fitted by a parabola around the Brewster angle minimum. From the reflectance value, the BAM thickness model allows evaluation of the thickness of the layer at the surface with the knowledge of the experimental Brewster angle and the optical index of the film. This model is based on the proportionality relation between the reflectance and the square of the interfacial film thickness when the optical index of the film is assumed constant.49 Moreover, with Brewster angle microscopy, information on the fluidity of the film can be obtained by observing the geometry of the domains at the water surface. For ellipsometric measurements the same setup as that for BAM was used as an imaging ellipsometer at an incidence angle close to the Brewster one (52°). It operates on the principle of classical null ellipsometry.50 The angles of the polarizer, compensator, and analyzer that obtained the null condition allow one to get the (∆, Ψ) angles that are related to the optical properties of the sample. In ultrathin film conditions, ∆ is proportionnal to the film thickness. Comparison of the measured data with computerized optical modeling included in the BAM software leads to a deduction of film thickness when an estimation of the refractive index can be obtained.
Results BGTC Monolayer at the Air/Water Interface. BGTC was spread at the air/water interface and the Π-A isotherm was registered (Figure 1). The pressure regularly increases up to pressure collapse at 52 mN/m. The layer formed at the air/water interface is stable: the pressure remains constant at high lateral pressure when the barrier stops and the isotherm is reversible (not shown). Whatever the lateral pressure, BAM micrographs show an homogeneous layer where reflectance increases with compression from Rf ) 7.70 × 10-7 in the gas phase at 0.5 mN/m to 5.05 × 10-6 before collapse (Figure 1). Estimations of layer thickness taking a refractive index of 1.47 give ≈10 ( 1 Å in the gas phase (image a, Figure 1), ≈27 ( 1 Å at 20 mN/m in the LC phase (48) Blaudez, D.; Buffeteau, T.; Cornut, J. C.; Desbat, B.; Escafre, N.; Pezolet, M.; Turlet, J. M. Thin Solid Films 1994, 242, 146. (49) de Mul, M. N. G.; Mann, J. A. Langmuir 1998, 14, 2455. (50) Azzam, R. M. A.; Bashara, N. M. Ellipsometry and Polarized Light; North-Holland Physics Publishing: New York, 1977.
Figure 1. Π-A isotherm and BAM images of BGTC at the air/water interface. Images: (a) Π ) 0.7 mN/m, Rf ) 7.70 × 10-7; (b) Π ) 10 mN/m, Rf ) 3.80 × 10-6; (c) Π ) 20 mN/m, Rf ) 4.35 × 10-6; and (d) Π ) 40 mN/m, Rf ) 5.05 × 10-6 (Rf ) surface reflectance). Subphase: H2O, NaCl 150 mM.
Figure 2. IR spectra of BGTC: (- - -) evaporated film on ATR diamond crystal and (s) PMIRRAS spectrum at 20 mN/m at the air/water interface. Subphase: H2O, NaCl 150 mM.
(image c, Figure 1), and ≈29 ( 1 Å at 40 mN/m in the solid phase (image d, Figure 1). The PMIRRAS spectrum of the BGTC monolayer was registered in parallel with that of an evaporated film of BGTC by attentuated total reflection (ATR) (Figure 2). The assignment of the major bands of BGTC is summarized in Table 1.51-53 If we keep in mind that the ATR spectrum is characteristic of an isotropic organization of BGTC while the PMIRRAS spectrum is sensitive to the orientation of the BGTC molecule in the monolayer at the air/water interface, it is noticeable that strong (51) Angell, L.; Sheppard, N.; Yamaguchi, A.; Shimanouchi, T.; Miyazawa, T.; Mizushima, S. Trans. Faraday Soc. 1957, 53, 589. (52) Furer, V. L. J. Appl. Spectrosc. 1990, 53, 860. (53) Socrates, G. Infrared Characteristic Group Frequencies; John Wiley and Sons: New York, 1980.
DNA-BGTC Interactions
Langmuir, Vol. 24, No. 17, 2008 9601
Table 1. Wavenumbers and Assignment of Main BGTC IR Bandsa wavenumbers (cm-1)
type of vibration (molecular unit)
≈1690 ≈1670/1640 ≈1620 1540
ν(CdO) ester (carbamate linker) νas(CN) (guanidinium) amide I (carbamate linker) δ(NH2) (guanidinium) + amide II (carbamate linker) δ(CH2) δ(OH) (cholesterol) amide III (carbamate linker) r(NH2)in plane (guanidinium) ν(CsO) (cholesterol)
1470 1380 1260 1150 1050 a
ν, stretching; δ, bending; r, rocking.
Table 2. Reflectance and (∆,Ψ) Angle Values of Layers Obtained by BAM and Ellipsometry
layer composition
0.7 10.0 20.0 40.0
0.77 3.80 4.35 5.05
+DNA, R(+/-) ) 1
24.0
10.10
+DNA, R(+/-) ) 1/10
30.0
16.10
BGTC/DOPE (3/2)
20.0
1.64
40.0
1.98
20.0
3.58
b. coll.a
34.5
20.0
4.44
b. coll.a
61.2
+DNA, R(+/-) ) 1/30
a
variations occur in the 1600-1700 cm-1 region showing peculiar orientations of the guanidinium polar head at the interface compared to the ATR spectrum, but it is difficult to go further in the interpretation due to the complex attribution in this region due to various couplings. DNA Interaction with a BGTC Monolayer (20mN/m) at the Air/Water Interface. When DNA is injected into the subphase of a preformed BGTC monolayer at 20 mN/m, kinetics of lateral pressure variation show an increase of Π (for a defined surface) due to DNA interaction with BGTC (Figure 3). ∆Π increases in parallel with the injected DNA quantity. Indeed, ∆Π ) 4 mN/m for a quantity of DNA corresponding to a stoichiometric ratio of positive charge of BGTC versus negative charge of DNA (R(+/-) ) 1) and ∆Π ) 10 mN/m for an excess of DNA (R(+/-) ) 1/10. Furthermore, the final Π value is reached faster when DNA is in excess: 15 min at R(+/-) ) 1/10 compared to 30 min at R(+/-) ) 1. BAM pictures registered in the final states after DNA interaction display an increase of the reflectance compared to the case of BGTC alone (Rf ) 1.01 × 10-5 for R(+/-) ) 1 and Rf ) 1.61 × 10-5 for R(+/-) ) 1/10 compared to 4.35 × 10-6 for BGTC; Table 2), and the layer remains homogeneous (no domain appears at the resolution of the Brewester angle microscope). The thickness of the BGTC monolayer at 20 mN/m was estimated from reflectance measurements at 27 ( 1 Å. If we assume that this thickness remains unchanged during DNA adsorption, using a two layer optical model, it is possible to estimate the thickness of the DNA layer. Taking a refractive index for hydrated DNA
BAM reflectance (×10-6)b
BGTC
+DNA, R(+/-) ) 5
Figure 3. Kinetics of DNA interaction on a BGTC monolayer (Π ) 20 mN/m) at the air/water interface: (- - -) R(+/-) ) 1/1 and (s) R(+/-) ) 1/10. R(+/-) ) ratio of positive charges of BGTC in the monolayer versus negative charges of DNA injected in the subphase. Subphase: H2O, NaCl 150 mM.
Π (mN/m)
b. coll.: beyond collapse.
b
ellipsometric angles (∆,Ψ)
187.1 ( 1.0°, 1.90 ( 0.2° 191.5 ( 1.0°, 1.75 ( 0.2° 207.5 ( 1.0°, 2.50 ( 0.2° 194.5 ( 1.0°, 1.63 ( 0.2° 213.4 ( 1.0°, 2.31 ( 0.2°
Error estimation: 5%.
between 1.47 and 1.50,54 we obtained 14.5 ( 1.5 Å for the DNA layer thickness at R(+/-) ) 1 and 25.0 ( 2.0 Å when DNA is in large charge excess at R(+/-) ) 1/10. To go further in the organization of DNA under the BGTC monolayer, IR spectroscopy was performed. First, the bulk spectrum of DNA in solution was registered by ATR (Figure 4B). On this spectrum, the main bands of DNA can be assigned: ν(CdO) of the bases in the 1600-1700 cm-1 region, νas(PO2-) at 1220 cm-1, ν(PO2-) at 1080 cm-1, ν(CO) deoxyribose around 1050 cm-1, and ribose-phosphate main chain vibration around 975 cm-1. As expected, the νas(PO2-) frequency is characteristic of the fully hydrated B-form of DNA.55-59 One can also notice that, in the isotropic conformation of DNA in buffer, the intensity ratio of the antissymetric PO2- at 1220 cm-1 versus symmetric PO2- at 1080 cm-1, IPO2-as/IPO2-s < 1. This ratio IPO2-as/IPO2-s will further be used as a marker of the DNA orientation. Indeed, PMIRRAS is sensitive to the orientation of transition moments and transition moments of νas(PO2-) and νs(PO2-) pointed in perpendicular directions in the DNA double helix (scheme 2), and then the intensity ratio of these two bands will be very sensitive to this group orientation which allows one to estimate the orientation of DNA. PMIRRAS spectra obtained for the DNA adsorbed under the BGTC monolayer at the stoichiometric charge ratio (R(+/-) ) 1) and with an excess of DNA (R(+/-) ) 1/10) are presented Figure 4A. They clearly show the presence of DNA characterized by the main bands in the 1250-950 cm-1 region free from BGTC contribution. The 1600-1700 cm-1 region also clearly demonstrates the presence of DNA bases, but it is more complex to attribute due to the BGTC contributions. (54) Samoc, A.; Miniewicz, A.; Samoc, M.; Grote, J. G. J. Appl. Polym. Sci. 2007, 105, 236. (55) Shimanouchi, T.; Tsuboi, M.; Kyogoku, Y. In The Structure and Properties of Biomolecules and Biological Systems-AdVances in Chemical Physics; Duchesne, J., Ed.; Interscience: London, 1964; p 435. (56) Tsuboi, M. In Applied Spectroscopy ReViews; Brame, E. G. J., Ed.; Dekker: New York, 1969; 45. (57) Taillandier, E.; Liquier, J. Methods Enzymol. 1992, 211, 307. (58) Liquier, J.; Taillandier, E. In Infrared Spectroscopy of Biomolecules; Mantsch, H. H., Chapman, D., Eds.; Wiley-Liss, Inc.: New York, 1996; p 131. (59) Banyay, M.; Sarkar, M.; Graslund, A. Biophys. Chem. 2003, 104, 477.
9602 Langmuir, Vol. 24, No. 17, 2008
Figure 4. IR spectra of DNA/BGTC interactions. (A) PMIRRAS spectra of DNA in interaction with a BGTC monolayer at 20 mN/m: (- - -) R(+/-) ) 1/1and (s) R(+/-) ) 1/10. (B) ATR spectrum of DNA in solution. R(+/-) ) ratio of positive charges of BGTC in the monolayer versus negative charges of DNA injected in the subphase. Subphase: H2O, NaCl 150 mM. Scheme 2. Projections of the PO2- Antisymmetric and Symmetric Transition Moments in the Double Helix of DNA
From the great differences in this region, whether the charge ratio R(+/-), it is possible to conclude that the DNA base/ BGTC interactions are different when DNA is in excess or in stoichiometric charge ratio. Whatever the charge ratio, one can notice that the intensity ratio IPO2-as/IPO2-s > 1, which means an anisotropic orientation of the DNA under the BGTC monolayer compared to the DNA in buffer. DNA Interaction with a BGTC/DOPE (3/2) Monolayer (20mN/m) at the Air/Water Interface. Experiments were also performed to study the DNA interaction with the BGTC/DOPE (3/2) mixture used for transfection assays. Two different charge
Castano et al.
ratios were investigated: an excess of negative charges of DNA, R(+/-) ) 1/30, and an excess of positive charge of BGTC, R(+/-) ) 5. Whatever the charge ratio, when DNA is injected into the subphase of a preformed BGTC/DOPE monolayer at 20 mN/m, kinetics of lateral pressure variation show a weak but reproducible decrease of Π, ∆Π ) -2 mN/m (data not shown), which can only be explained by a condensation of molecules at the lipid interface induced by DNA interaction with the monolayer. A BAM picture registered for the BGTC/DOPE monolayer alone at 20 mN/m displays an homogeneous luminosity (not shown) with a mean reflectance Rf ) 1.64 × 10-6 (Table 2), which traduces a good miscibility of both lipids. Taking a refractive index of 1.47, the layer thickness is estimated as 19.3 ( 1.0 Å. The (∆,Ψ) angles obtained by ellipsometry (Table 2) give 18.5 ( 1.0 Å for the same refractive index, which is in good agreement with the BAM estimation. At the final states after DNA interaction, there is an increase of the reflectance compared to the BGTC/DOPE (3/2) alone (Rf ) 3.58 × 10-6 for R(+/-) ) 5 and Rf ) 4.44 × 10-6 for R(+/-) ) 1/30; Table 2). In both cases, the layer remains homogeneous: no domain appears at the spatial resolution of the Brewester angle microscope. Thickness estimations assuming that the thickness of the BGTC/DOPE monolayer remains unchanged after DNA interaction and taking a refractive index for DNA between 1.47 and 1.50, we obtained 7.0 ( 1.5 Å for the DNA layer average thickness when the charge ratio R(+/-) ) 5, and 10.5 ( 1.5 Å when the charge ratio R(+/-) ) 1/30. The (∆,Ψ) angles obtained by ellipsometry in the presence of DNA (Table 2) display an increase of the ∆ angle, which corresponds to the increase of the layer thickness. Thickness estimations using the previous refractive index for DNA give 7.0 ( 1.0 Å for the DNA layer average thickness when the charge ratio R(+/-) ) 5 and 12.0 ( 1.0 Å for the charge ratio R(+/-) ) 1/30, which are very comparable to the BAM estimations. A PMIRRAS spectrum of the BGTC/DOPE (3/2) monolayer at 20 mN/m is presented in Figure 5A in the 1800-900 cm-1 region. The main bands are the ν(CdO) of DOPE ester groups around 1737 cm-1, the δ(CH2) of the acyl chains, and the vas,s(PO2-) around 1220 and 1080 cm-1, respectively.52,60 The BGTC contributions in the 1600-1700 cm-1 region cannot be seen due to the strong negative band around 1660 cm-1 due to the strong dispersion of the refractive index of water in this spectral range.61,62 When DNA was injected under the lipid monolayer, whether the charge ratio R(+/-) ) 5 or 1/30, the PMIRRAS spectra (Figure 5) show a significant increase of the νas,s(PO2-) around 1220 and 1080 cm-1 which is more important in the case of an excess of DNA. Since we can assume that DNA preferentially interacts with BGTC (no interaction observed with a pure DOPE monolayer; data not shown) and phosphate groups of DOPE will not be perturbed by DNA interaction, we can conclude that the increase of the vas,s(PO2-) is due to accumulation of DNA phosphate groups under the monolayer and then to the interaction of DNA strands with the monolayer. One can also notice that, whatever the charge ratio, the intensity ratio of DNA bands IPO2-as/IPO2-s > 1: this means an anisotropic orientation of the DNA under the lipid monolayer. One can also see that when DNA is in excess (R(+/-) ) 1/30, Figure 5B), modifications occur in the 1600-1700 cm-1 spectral region due to the contribution of the DNA bases which superimpose to the lipid (60) Mantsch, H. H.; McElhaney, R. N. Chem. Phys. Lipids 1991, 57, 213. (61) Dluhy, R. A. J. Phys. Chem. 1986, 90, 1373–1379. (62) Blaudez, D.; Turlet, J. M.; Dufourcq, J.; Bard, D.; Buffeteau, T.; Desbat, B. J. Chem. Soc., Faraday Trans. 1996, 92, 525.
DNA-BGTC Interactions
Langmuir, Vol. 24, No. 17, 2008 9603
Figure 6. BAM images of the compression beyond the collapse of (A,B) mixed DNA-BGTC/DOPE (3/2) monolayers at the air/water interface, Π ) 46 mN/m, A ) 25 A2/mol, (A) R(+/-) ) 5, Rf ) 34.5 × 10-6 and (B) R(+/-) ) 1/30, Rf ) 61.2 × 10-6, and (C) pure BGTC/DOPE monolayer at collapse, Π ) 40 mN/m, A ) 51 A2/mol, Rf ) 1.98 × 10-6. R(+/-) ) ratio of positive charges of BGTC in the monolayer versus negative charges of DNA injected in the subphase. Rf ) surface reflectance. Subphase: H2O, NaCl 150 mM.
Figure 5. PMIRRAS spectra of DNA interaction with a BGTC/DOPE (3/2) monolayer at 20mN/m: (A) (- - -) BGTC/DOPE (3/2), 20 mN/m; (s) DNA, R(+/-) ) 5. (B) (- - -) BGTC/DOPE (3/2), 20 mN/m; (s) +DNA, R(+/-) ) 1/30. Subphase: H2O, NaCl 150 mM.
spectrum and a strong intensification of the 1025 cm-1 band corresponding to the furanose ring vibration. Compression beyond the Collapse of DNA-BGTC/DOPE (3/2) Monolayers. Previous studies demonstrated that compression beyond the collapse of phospholipid monolayers can lead to the formation of stable multilayers at the air/water interface which are relevant new models for studying the properties of biological membranes and for understanding the nature of interactions between membranes and peptides, proteins, or nucleotides.63 Therefore, the behavior of the mixed adsorbed DNA-BGTC/DOPE (3/2) monolayer was investigated beyond the collapse. After adsorption of DNA under the BGTC/DOPE (3/2) monolayer at 20 mN/m, the resulting layer was compressed to reach the collapse (around Π ≈ 46 mN/m and molecular area of 50 Å2/mol). The compression was carried on to divide the molecular area by a factor of 2 (final molecular area around 25 Å2/mol). BAM pictures and PMIRRAS spectra were then registered. BAM pictures display very high luminosity at the end of the compression, whatever the charge ratio, compared to the same monolayers before compression at 20 mN/m. When DNA is in default, R(+/-))5 (Figure 6A), we observe a relatively homogeneous layer of higher luminosity compared to that before collapse (Rf ) 34.5 × 10-6), and a reflectance multiplied by 10 compared to that of the layer before collapse. Since thickness is proportional to the square of the reflectance, this means that roughly the thicknesses are multiplied by 3. When DNA is in excess, R(+/-) ) 1/30 (Figure 6B), the layer remains quite homogeneous after collapse, but one can distinguish granularities on the picture. The average luminosity is multiplied by a factor (63) Saccani, J.; Castano, S.; Beaurain, F.; Laguerre, M.; Desbat, B. Langmuir 2004, 20, 9190.
of 17 compared to the same before collapse (Rf ) 61.2 × 10-6), which means that the thickness is roughly multiplied by a factor 4. This increased thickness is also confirmed by the increase of the (∆,Ψ) angles determined by ellipsometry (Table 2). To go further in the identification of these thick layers, PMIRRAS spectra of the layers compressed beyond the collapse were recorded. In the 1800-900 cm-1 region, they display an increase of all the bands present on the spectrum whatever the BGTC/DNA charge ratio compared to the spectrum of a pure compact BGTC/DOPE monolayer at 40 mN/M (Figure 7A). Since all the bands can contain contributions of both lipid and DNA, we investigate in parallel the 3000-2800 cm-1 spectral region of the antisymmetric (νas(CH2)) and symmetric (νsCH2) stretching bands located at 2920 and 2850 cm-1, respectively, mainly characteristic of the lipids chains and free of DNA contribution (Figure 7B). Whatever the charge ratio, their intensity is multiplied by a factor of ≈2 compared to the pure lipid monolayer. This value proves the formation of a lipid bilayer at the surface of the trough. The important enlargment of the bands and their shift toward higher frequencies, all the more important than DNA quantity increases, correspond to the perturbation of the organization of the lipid acyl chain with an increase in gauche conformation. In parallel, it was shown that no bilayer is formed during the compression beyond the collapse of a pure BGTC/ DOPE (3/2) monolayer which collapses with the formation of aggregates (data not shown) and that it is the presence of DNA adsorbed under the layer which induces the bilayer formation. In the 1800-900 cm-1 region, the intensity increase noticed for all bands shows both the DNA interaction and the bilayer formation. When DNA is in default, R(+/-) ) 5, all the main band intensities are multiplied by a factor almost equal to 2, but since the bands intensities are sensitive both to the surface concentration of the molecules and to their orientation, it is quite difficult to decorrelate both phenomena which can occur simultaneously with antagonist effects on band intensities. In parallel, a decrease of the wavelengths of νas,s(PO2-) is observed compared to the case of pure BGTC/DOPE lipid monolayer at 40 mN/m (-1.3 and -4.6 cm-1, respectively). When DNA is in excess, R(+/-) ) 1/30, all the bands are multiplied by a factor greater than 2, which corresponds to both the lipid bilayer formation and DNA adsorption. In parallel, a red shift of the νas,s(PO2-) wavelengths (-8.4 and -9.8 cm-1, respectively) compared to the case of the BGTC/DOPE monolayer at 40 mN/m is observed, showing both the DNA phosphate contribution due to its interaction with the bilayer and perhaps also a reorientation of the DOPE phosphate group due to the bilayer formation.
9604 Langmuir, Vol. 24, No. 17, 2008
Figure 7. PMIRRAS spectra of a DNA-BGTC/DOPE (3/2) layer after compression beyond the collapse: (A) 1850-900 cm-1 region and (B) CH region, BGTC/DOPE (3/2), 40 mN/m, before collapse, A ) 50 A2/molecule, (- - -) +DNA, R(+/-) ) 5; beyond the collapse, A ) 25 A2/molecule; (s) +DNA, R(+/-) ) 1/30; beyond the collapse, A ) 25 A2/molecule.
Discussion DNA Interaction with Pure BGTC or Mixed BGTC/DOPE (Molar Ratio 3/2) Monolayers at the Air/Water Interface. Since the cationic lipid BGTC appeared to be very efficient for gene transfection into a variety of mammalian cell lines, we first investigated its physicochemical behavior at the air/water interface. We were quite surprised to obtain stable films up to 52 mN/m for concentrations below µM, since BGTC was described to form true micellar solutions with a critical micelle concentration (CMC) ≈ 90 µM.25 BGTC films behave much more like Langmuir films than Gibbs ones, and BGTC is probably stabilized by hydrophobic interactions between the cholesterol units of the molecule. The thickness of 29 ( 1 Å and the 38 ( 2 Å2/mol molecular area obtained for the densely packed film are coherent with a model of a monolayer formed with extended structures of the molecule, the cholesterol units being packed together in the top of the monolayer, and the guanidinium polar heads diving into the water subphase. DNA injected into the subphase at a stoichiometric charge ratio or in excess strongly interacts with the BGTC monolayer compressed at 20 mN/m as demonstrated by (i) the lateral pressure variations, (ii) the reflectance increases observed by BAM microscopy, and (iii) the PMIRRAS spectra which clearly show the presence of DNA, a presence characterized by the main bands of the phosphate groups in the 1250-950 cm-1 region free from BGTC contribution and the influence of DNA bases in the 1600-1700 cm-1 region. The lateral pressure increases of +4
Castano et al.
mN/m for a stoichiometric charge ratio (R(+/-) ) 1) and of +10 mN/m for an excess of DNA (R(+/-) ) 1/10) correspond to a molecular area decrease of 2 and 4.2 Å2/mol per BGTC molecule, respectively. This suggests that DNA occupies only 4.1% of the layer surface for R(+/-) ) 1 and 9.1% for R(+/-) ) 1/10, a finding showing that the DNA does not deeply insert into the BGTC layer but preferentially adsorbs to the positively charged polar heads of BGTC. The measured reflectances allow one to estimate the DNA layer thickness assuming that BGTC remains unchanged. Here, taking a refractive index for DNA between 1.47 and 1.50, we obtained 14.5 ( 1.5 Å for the DNA layer thickness at R(+/-) ) 1 and 25.0 ( 2.0 Å when DNA is in large charge excess at R(+/-) ) 1/10. The dimensions of the calibrated 200 base pair DNA strands used in the present study can be estimated through the Watson-Crick structure to 24 Å for the diameter of the double helix and 680 Å for the length of helix, taking 34 Å for 10 base pairs assuming a B-form since the DNA is fully hydrated,64 which is confirmed by the 1225 cm-1 wavelength of νas(PO2-). When DNA is in excess, the average layer thickness measured corresponds to the diameter of the DNA double helix, which means that the double-strand form is preserved,45,47 and let us suppose that there is a monolayer of DNA under the BGTC one. On the contrary, at the stoichiometric charge ratio R(+/-) ) 1, the DNA layer under the BGTC one is incomplete, since the average thickness is inferior to the smallest dimension of the DNA but the DNA molecules are homogeneously dispersed under the BGTC monolayer, leading thereby to a homogeneous increase of reflectance. Such a hypothesis of an incomplete layer is favored over that of the denaturation of the double-stranded DNA helix which could also lead to a thinner thickness as it was observed for DNA interaction with aminecontaining amphiphiles46,65 where DNA splitting is observed. The intensity ratio IPO2-as/IPO2-s > 1 measured by PMIRRAS, whatever the charge ratio, indicates an anisotropic orientation of the DNA under the BGTC monolayer compared to the DNA in buffer. Using the PMIRRAS selection rules at the air/water interface, such an intensity ratio means that the antisymmetric PO2- vibration moment is favored compared to the symmetric one. Considering the geometry of the double helix of DNA and the projections of the symmetric and antisymmetric transition moments of the PO2- group (Scheme 2), this suggests that the DNA double helix is mainly parallel to the interface plane, a result which is in agreement with the BAM data. Next, in order to better mimic real transfection conditions, we investigated the BGTC/DOPE (3/2) lipid mixture, as such a formulation was demonstrated to allow efficient gene transfection.25,35-38 Indeed, DOPE is often used as neutral helper lipid to prepare stable liposomes when using non-bilayer forming cationic lipids such as cationic cholesterol derivatives like BGTC.13 Furthermore, it was demonstrated that DOPE may also provide membrane fusion activity, a property which may allow enhanced escape of the lipoplexes from the endosomes, as it is a non-bilayer-forming lipid at acidic pH capable of destabilizing the endosomal membrane, thereby leading to increased transfection efficacy.66-72 We also choose two charge ratios which correspond to (i) an excess of BGTC positive charges compared to DNA negative charges (R(+/-) ) 5) and (ii) an excess of DNA negative charges compared to BGTC positive charges (R(+/-) ) 1/30). Previous work studying the supramolecular assemblies formed by BGTC/DOPE demonstrated indeed a threezone model of colloidal stability of the DNA/BGTC/DOPE lipoplexes,41 where only two zones lead to colloidally stable (64) Watson, J. D.; Crick, F. H. C. Nature 1953, 4356, 737. (65) Sukhorukov, G. B.; Feigin, L. A.; Montre, M. M.; Sukhorukov, B. I. Thin Solid Films 1995, 259, 79.
DNA-BGTC Interactions
complexes: the zone of highly positive transfection-efficient complexes (corresponding to R(+/-) > 3 in our expression of charge ratio) and the zone of negative complexes (R(+/-) < 1). Therefore, we investigated two DNA ratios belonging to these two domains of colloidal stability of the lipoplexes. Whatever the charge ratio, BAM results show a homogeneous increase of luminosity, corresponding to a homogeneous DNA adsorption. Thus, even if DNA preferentially interacts with BGTC, this interaction does not induce a phase separation of both lipids. Thickness estimations by BAM and ellipsometry lead to a DNA layer of 7.0 ( 1.5 Å when DNA is in default (R(+/-) ) 5) and of 11.0 ( 2.0 Å when DNA is in excess (R(+/-) ) 1/30), which means that in both cases the DNA layer under the BGTC is incomplete since its thickness is inferior to the smallest dimension of the DNA double helix. These thickness differences are in good agreement with the PMIRRAS spectra which display a more significant increase of the νas,s(PO2-) due to DNA adsorption and important modifications in the 1600-1700 cm-1 spectral region due to the contribution of the DNA bases in the case of R(+/-) ) 1/30. Whatever the charge ratio, the intensity ratio of DNA band contributions IPO2-as/IPO2-s > 1 shows an anisotropic orientation of the DNA under the lipid monolayer; this favors the antisymmetric PO2- vibration moment over the symmetric one and means that the double helix of DNA is mainly parallel to the interface plane in both cases, but with a different orientation of the furanose ring in the case of DNA excess as shown by the important intensification of the band around 1025 cm-1. Organized Collapse of the BGTC/DOPE (3/2)-DNA Monolayers beyond Compression. Since previous studies have demonstrated that compression beyond the collapse of some phospholipid monolayers can lead to the formation of stable multilayers at the air/water interface,62 we next investigated the behavior beyond the collapse of the mixed DNA/BGTC/DOPE layer obtained at 20 mN/m. BAM pictures obtained after reduction of a factor 2 of the molecular area of collapse display an important and homogeneous increase of reflectance, all the more important than DNA quantity increases. These BAM pictures with homogeneous and high luminosity are characteristic of an organized collapse from a bi- to a tridimensional system; indeed, collapses inducing formation of aggregates at the interface or folds in the subphase would show either bright spots of saturated luminosity in the first case or flickers due to reflection of the diving objects in the second one.62 To explain such a behavior, the hypothesis of the formation of an asymmetric lipid bilayer on the top of the DNA layer can be formulated (Scheme 3). Indeed, it may be suggested that the BGTC molecules remain in the layer facing the aqueous phase due to their strong interactions with the DNA adsorbed underneath, while DOPE (which does not directly interact with DNA in the subphase) flip-flops beyond compression to form a top layer. A tail-to-tail contact organization seems to be more stable than an arrangement where the polar headgroups of the top layer would be in interaction with the alkyl chains of the bottom layer. This tail-to-tail contact organization supposes that a hydration layer exists at the polar head level to minimize repulsion between the polar heads of the top layer and the air. Such a lipid distribution beyond compression is fully compatible with the respective molecular areas of both lipids. Indeed, the molecular area of a BGTC molecule in a compressed monolayer may have a value of 38 ( 1A2/mol using the compression isotherm (cf. results). Concerning DOPE, the molecular area was estimated around 60 A2/mol using the same method (data not shown and ref 73). Since the lipid molar ratio used here is 3 BGTC for 2 DOPE, (66) Ellens, H.; Bentz, J.; Szoka, F. C. Biochemistry 1986, 25, 4141.
Langmuir, Vol. 24, No. 17, 2008 9605 Scheme 3. Scheme of the DNA Interactions with BGTC/DOPE (3/2) Layer, R(+/-) ) 5a
a (A) At 20 mN/m, formation of an incomplete DNA layer (≈7 Å thick) under the BGTC/DOPE monolayer (≈19 Å thick). (B) Beyond the collapse, formation of an asymmetric lipid bilayer (≈50 Å thick); DOPE on the top, and BGTC in contact with an organized DNA layer (≈27 Å thick).
the above data suggest that 2 DOPE molecules occupy an area similar to that of 3 BGTC molecules, a result in agreement with the hypothesis of a bottom layer of BGTC and a top layer of DOPE. The hypothesis of such an asymmetric bilayer formed beyond compression is reinforced by the observation that a pure BGTC monolayer with an adsorbed DNA layer does not lead to an organized collapse (data not shown). This may be explained by the fact that most (if not all) BGTC molecules are involved in interactions with DNA and stabilized by the DNA layer, so that they cannot flip-flop beyond compression. The previous hypothesis allowed us to estimate the thickness of the lipid layer, whereas the reflectances obtained by BAM and the measured ellipsometric angles could be used to evaluate the thickness of the adsorbed DNA layer. The thickness of a compressed pure BGTC layer was estimated around 29 ( 1 Å (cf. results), while that of a compressed DOPE monolayer was estimated around 21 ( 1 Å using the same method (data not shown), with this DOPE thickness being coherent with a fluid monolayer of mainly dioleyl (C18:1) lipid acyl chains. Thus, the thickness of the lipid bilayer can be estimated around 50 Å. Taking a two-layer model and a refractive index of 1.47 for the lipid bilayer and of 1.47-1.50 for the DNA, one can assume a DNA layer thickness around 27.3 ( 3.0 Å by BAM (23.0 ( 3.0 Å using the ellipsometric angles) when DNA is in default R(+/-) ) 5 and around 51.2 ( 6.0 Å by BAM (48.0 ( 4.0 Å using the ellipsometric angles) when DNA is in excess R(+/-) ) 1/30. It is important to notice the good correlation obtained for the thickness estimations using both techniques (BAM and ellipsometry). In the case of a default of DNA (R(+/-) ) 5), since the DNA layer thickness corresponds to the dimension of the double helix diameter, these results are therefore in agreement with the formation of a DNA monolayer under the lipid bilayer with the DNA double strands mainly parallel to the bilayer plane, an arrangement coherent with the incomplete but parallel oriented DNA layer observed before compression at 20 mN/m (scheme 3). On the other hand, when DNA is in excess, the average DNA (67) Felgner, J. H.; Kumar, R.; Sridhar, C. N.; Wheeler, C. J.; Tsai, Y. J.; Border, R.; Ramsey, P.; Martin, M.; Felgner, P. L. J. Biol. Chem. 1994, 269, 2550. (68) Zhou, X.; Huang, L. Biochim. Biophys. Acta 1994, 1189, 195. (69) Farhood, H.; Serbina, N.; Huang, L. Biochim. Biophys. Acta 1995, 4, 289–295. (70) Schwartz, B.; Benoist, C.; Abdallah, B.; Scherman, D.; Behr, J. P.; Demeneix, B. A. Hum. Gen. Ther. 1995, 6, 1515. (71) Wrobel, I.; Collins, D. Biochim. Biophys. Acta 1995, 1235, 296. (72) Aissaoui, A.; Oudrhiri, N.; Petit, L.; Hauchecorne, M.; Kan, E.; Sainlos, M.; Julia, S.; Navarro, J.; Vigneron, J. P.; Lehn, J. M.; Lehn, P. Curr. Drug Targets 2002, 3, 1. (73) Rathman, J. F.; Sun, P. Faraday Discuss. 2005, 129, 193.
9606 Langmuir, Vol. 24, No. 17, 2008
layer appears to be thicker than the double helix diameter, a finding suggesting a more disoriented organization of the DNA strands with loops adsorbed under the bilayer and in agreement with the granularities observed (which is related to small regions of higher thickness due to DNA accumulation). The results obtained for the R(+/-) ) 5 charge ratio, which corresponds to an excess of positive charges of BGTC versus DNA negative charges, may be compared to previous data reported by Pitard and co-workers41 for positive lipoplexes of the same composition. Indeed, it was found that BGTC/DOPE (3/2) unilamellar liposomes of 150 nm in diameter were converted by addition of DNA in default (R(+/-) > 3 in our charge ratio expression) into concentric multilamellar structures of 250 nm in diameter characterized by DNA monolayers intercalated between the lipid bilayers. Furthermore, it was also demonstrated by small-angle X-ray scattering (SAXS) that the lamellar organization displayed a regular spacing of 70 Å, a value close to the 75 Å obtained for the total thickness (lipid bilayer + DNA layer) obtained herein by compression over the collapse, with the difference being possibly explained by the higher confinement induced by the multilamellar structure. Here, it should be stressed that quite similar results obtained with two different experimental systems may allow us to formulate some hypothesis on the lipid repartition in the lipoplexes bilayers. Taking into account our present data showing (i) that, under a homogeneous BGTC/ DOPE (3/2) layer, the DNA adsorbs homogeneously to form an organized but incomplete layer and (ii) that mechanical constraint (compression) leads to the formation of an organized and complete DNA monolayer under an asymmetric lipid bilayer, one can indeed wonder whether the DNA interaction with BGTC/DOPE large unilamellar vesicles (LUV) may lead to mechanical constraints resulting in an increased curvature tension of the liposomes and thereby promoting a complete lipid reorganization of the bilayer leaflets, multibilayer liposomes characterized by DNA monolayers sandwiched between asymmetric lipid bilayers being finally formed. Furthermore, Pitard and co-workers41 demonstrated that the BGTC/DOPE-DNA multilamellar li-
Castano et al.
poplexes (with DNA intercalated between the lipid bilayers) were not sensitive to ethidium bromide (absence of fluorescence due to ethidium bromide intercalation between the DNA base pairs with no ethidium bromide staining of the DNA observed by agarose gel electrophoresis). Such results suggest that no DNA is accessible to ethidium bromide, a fact only possible if no significant amount of DNA is adsorbed onto the external leaflet of the most peripheral lipid bilayer of the lipoplexes. Thus, in agreement with our present data obtained by using the Langmuir technique, this last lipid bilayer may display an asymmetric lipid composition with essentially DOPE in the external leaflet (no interaction with DNA) and BGTC mainly in the internal leaflet. Of note, such a lipid repartition in the most external bilayer should affect the surface charge (zeta potential) of the lipoplexes and therefore their efficiency and toxicity. Finally, since Pitard and co-workers did not study the supramolecular organization of the lipoplexes obtained by addition of DNA in excess (as they display a much lower transfection efficiency), it is thus not possible to compare the results obtained in the present work with a charge ratio R(+/-) ) 1/30. One can thus only note that the main difference observed herein between the charge ratios R(+/-) ) 5 and R(+/-) ) 1/30 is in fact the formation of a thicker and more disorganized DNA layer under the asymmetric lipid bilayer when DNA is in excess. In conclusion, our study demonstrates that important results allowing a better characterization of the BGTC-DNA interactions and organization can be obtained by the Langmuir technique coupled with BAM microscopy and PMIRRAS spectroscopy. It also suggests that more systematic work on cationic lipid-based gene delivery systems exhibiting a high transfection activity may allow one to unveil critical features playing an important role in the transfection efficiency. This could in turn lead to a better understanding of the mechanisms underlying cationic lipidmediated transfection as well as to the design of improved lipidbased gene delivery systems. LA703491R
