Cryo-electron tomography of nanoparticle transmigration into liposome Olivier Le Bihan a , Pierre Bonnafous a , Laszlo Marak b , Thomas Bickel c , Sylvain Trépout a,1 , Stéphane Mornet d , Felix De Haas e , Hugues Talbot b , Jean-Christophe Taveau a , Olivier Lambert a, * a Structure of Membrane Complexes and Cell Process, CBMN UMR-CNRS 5248, Université Bordeaux, ENITAB, IECB, Avenue des Facultés, F-33405 Talence, France b ESIEE, BP99, 2 Bd Blaise Pascal, F-93162 Noisy-le-Grand Cedex, France c CPMOH, Université Bordeaux, 316 cours de la Libération, F-33405 Talence, France d ICMCB, CNRS, Université Bordeaux, 87 Avenue du Dr. A. Schweitzer, F-33608 Pessac, France e FEI Electron Optics B.V. Achtseweg Noord 5, 5651 GG Eindhoven, The Netherlands article info Article history: Received 10 April 2009 Accepted 2 July 2009 Available online 23 July 2009 Keywords: Liposome Cryo-electron microscopy Cryo-electron tomography Nanoparticle Nanoparticle transport abstract Nanoparticle transport across cell membrane plays a crucial role in the development of drug delivery sys- tems as well as in the toxicity response induced by nanoparticles. As hydrophilic nanoparticles interact with lipid membranes and are able to induce membrane perturbations, hypothetic mechanisms based on membrane curvature or hole formation have been proposed for activating their transmigration. We report on the transport of hydrophilic silica nanoparticles into large unilamellar neutral DOPC liposomes via an internalization process. The strong adhesive interactions of lipid membrane onto the silica nano- particle triggered liposome deformation until the formation of a curved neck. Then the rupture of this membrane neck led to the complete engulfment of the nanoparticle. Using cryo-electron tomography we determined 3D architectures of intermediate steps of this process unveiling internalized silica nano- particles surrounded by a supported lipid bilayer. This engulfing process was achieved for a large range of particle size (from 30 to 200 nm in diameter). These original data provide interesting highlights for nano- particle transmigration and could be applied to biotechnology development. Ó 2009 Elsevier Inc. All rights reserved. 1. Introduction The growing interest for nanoparticles is motivated by potential developments in nanotechnology, biotechnology or in medicine, but is also due to the health risk associated with their use (Nel et al., 2006). Indeed, toxicological effects have been already identi- fied and further investigations are required to examine fundamen- tal mechanisms underlying these undesirable biological responses. Among predictive pathways leading to nanoparticle toxicity, trans- migration of nanoparticles into cell is one of main concerns. Transport of nanoparticles into mammalian cells was proposed to be mediated by a non endocytotic pathway as evidenced by the entry of ultrafine particles within red blood cell and cyt-D – blocked macrophages that both lack of endocytotic capabilities (Geiser et al., 2005; Rothen-Rutishauser et al., 2006). However, the passive transmembrane transport of gold particles studied with liposomes mimicking biological membrane (Banerji and Hayes, 2007) had revealed that gold particles did not diffuse through the lipid membrane suggesting that interactions with lipid membrane are likely required. Adhesion of colloidal particles onto giant vesicles based on elec- trostatic interactions induced the coverage of the particle by lipid membrane. This coverage was not limited to the vesicle-bead con- tact but rather extended to the entire bead while remaining at- tached to the vesicle periphery (Fery et al., 2003). Theoretical models have been proposed to predict adhesion and wrapping mechanism for colloid–vesicle complexes. The penetration is pos- sible for large vesicle radii compared to the particle radii whereas for small vesicles (R = 300 nm) it will not be complete for any col- loid radius (Deserno and Gelbart, 2002). Simulations of adhesive interaction of lipid bilayers with a spherical particle have shown the inducement of an engulfing process (Smith et al., 2007; Reyn- war et al., 2007). The particle is wrapped by the membrane, but may remain tethered to it. To have a complete engulfing process, the presence of phase-separated membrane domains is needed to promote fission, allowing the particle detachment from the larger membrane (Smith et al., 2007). Other simulations based on modeling thermodynamics of interactions between charged nanoparticles and lipid membrane, confirmed the deposition of a lipid bilayer around the particle triggering the formation of a hole 1047-8477/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.jsb.2009.07.006 Abbreviations: TEM, transmission electron microscopy; ET, electron tomogra- phy; DOPC, 1,2-dioleoyl-sn-glycero-3-phosphocholine; SLB, supported lipid bilayer; LUV, large unilamellar vesicle; SNP, silica nanoparticle; AFM, atomic force microscopy. * Corresponding author. Fax: +33 5 40002200. E-mail address: o.lambert@cbmn.u-bordeaux.fr (O. Lambert). 1 Present address: European Molecular Biology Laboratory, Meyerhofstrasse 1, D–69117 Heidelberg, Germany. Journal of Structural Biology 168 (2009) 419–425 Contents lists available at ScienceDirect Journal of Structural Biology journal homepage: www.elsevier.com/locate/yjsbi