Cationic lipid–DNA complexes in gene delivery: from biophysics to biological applications

Advanced Drug Delivery Reviews 47 (2001) 277–294 www.elsevier.com / locate / drugdeliv

Cationic lipid–DNA complexes in gene delivery: from biophysics to biological applications Maria C. Pedroso de Lima a

b,c ,

*, Sergio ˜ a,c , Pedro Pires c , Henrique Faneca c , ´ Simoes ¨ ¨ ¸d Nejat Duzgunes

Laboratory of Pharmaceutical Technology, Faculty of Pharmacy, University of Coimbra, 3000 Coimbra codex, Portugal b Department of Biochemistry, University of Coimbra, 3000 Coimbra codex, Portugal c Center for Neurosciences and Cell Biology, University of Coimbra, 3000 Coimbra codex, Portugal d Department of Microbiology, School of Dentistry, University of the Pacific, San Francisco, CA 94115, USA

Abstract Great expectations from the application of gene therapy approaches to human disease have been impaired by the unsatisfactory clinical progress observed. Among others, the use of an efficient carrier for nucleic acid-based medicines is considered to be a determinant factor for the successful application of this promising therapeutic strategy. The drawbacks associated with the use of viral vectors, namely those related with safety problems, have prompted investigators to develop alternative methods for gene delivery, cationic lipid-based systems being the most representative. This review focuses on the various parameters that are considered to be crucial to optimize the use of cationic lipid–DNA complexes for gene therapy purposes. Particular emphasis is devoted to the analysis of the different stages involved in the transfection process, from the biophysical aspects underlying the formation of the complexes to the different biological barriers that need to be surpassed for gene expression to occur.  2001 Elsevier Science B.V. All rights reserved. Keywords: Gene therapy; Non-viral vectors; Lipoplexes; Biological barriers

Contents 1. Introduction ............................................................................................................................................................................ 2. From cationic liposomes to lipoplexes ...................................................................................................................................... 2.1. Liposome composition ..................................................................................................................................................... 2.1.1. Cationic lipids........................................................................................................................................................ 2.1.2. Co-lipids ............................................................................................................................................................... 2.2. Structure and size of cationic liposomes............................................................................................................................. 2.3. Mode of lipoplex formation: structure and morphology of the complexes ............................................................................. 2.4. Parameters affecting the physico-chemical properties of lipoplexes......................................................................................

278 278 278 278 279 279 281 282

Abbreviations: DC-Chol, 3 b[N-(N9,N9-dimethylaminoethane)-carbamoyl]cholesterol; DMRIE, N-(2-hydroxyethyl)-N,N-dimethyl-2,3bis(tetradecyloxy)-1-propanaminium bromide; DOGS, dioctadecyl amino glycyl spermine; DOPE, dioleoylphosphatidylethanolamine; DOSPA, 2,3 dioleyloxy-N-[2(spermine carboxaminino)ethyl]-N,N-dimethyl-1-propanaminium trifluoroacetate; DOTAP, 1,2-dioleoyl-3-trimethylammonium propane; DOTMA, 2,3-bis(oleyl)oxipropyl-trimethylammonium chloride *Corresponding author. Tel.: 1351-239-820-190; fax: 1351-239-853-607. E-mail address: [email protected] (M.C. Pedroso de Lima). 0169-409X / 01 / $ – see front matter  2001 Elsevier Science B.V. All rights reserved. PII: S0169-409X( 01 )00110-7

278

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294

3. Biological activity of the lipoplexes .......................................................................................................................................... 3.1. Lipoplex–cell interaction .................................................................................................................................................. 3.1.1. Cell uptake of lipoplexes ........................................................................................................................................ 3.1.2. Cytoplasmic delivery of DNA ................................................................................................................................. 3.1.3. Nuclear entry of DNA ............................................................................................................................................ 4. Biological stability of lipoplexes: implications for in vivo use .................................................................................................... 5. Concluding remarks ................................................................................................................................................................ Acknowledgements ...................................................................................................................................................................... References ..................................................................................................................................................................................

1. Introduction Research in somatic gene therapy has been focused on the development of suitable carriers that, while exhibiting adequate features for in vivo use, would also mediate efficient intracellular delivery of genetic material. Although the majority of clinical trials have been based on the use of viral vectors, cationic liposomes are emerging as promising nonviral carriers for genetic medicines due to their safety and versatility. The potential advantages of these systems over viral vectors in gene therapy prompted investigators to optimize this strategy by developing novel formulations in which new cationic lipids, different helper lipids, new plasmid constructs, association of proteins or fusogenic peptides and different modes of complex preparation were tested both in vitro and in vivo [1]. However, despite this extensive work in the last decade, which resulted in remarkable progress culminating in the use of cationic liposome–DNA complexes (lipoplexes) in clinical trials, their transfection efficiency is still unsatisfactory, especially when compared with viral vectors. The basic knowledge of the structure–activity relationships of lipoplexes and of the mechanisms involved in the process of intracellular gene delivery is still scarce. It is believed that such knowledge is crucial to further improve the biological performance of these systems and therefore gaining insights into these mechanistic aspects should constitute one of the main goals in this field.

283 284 284 286 286 287 291 291 291

plasmid–DNA encapsulation and consequently the low levels of transfection encouraged investigators to develop other liposome-based strategies [2]. Soon after Behr [3] demonstrated that cationic liposomes could complex and condense DNA, Felgner and collaborators proposed the use of cationic liposomes as efficient carriers for the intracellular delivery of DNA [4]. These authors prepared liposomes composed of the cationic lipid 2,3-bis(oleoyl)oxipropyltrimethylammonium chloride (DOTMA) and dioleoylphosphatidylethanolamine (DOPE), which became commercially available as a transfection reagent designated Lipofectin  . The ability of this system to mediate transfection was attributed to recognition of certain properties, namely: (1) a spontaneous electrostatic interaction between the positively charged liposomes and the negatively charged DNA, which results in an efficient condensation of the nucleic acids; (2) the fact that the resulting cationic liposome / DNA complexes could exhibit a net positive charge that promotes their association with the negatively charged cell surface and (3) the fusogenic properties exhibited by the cationic liposome formulation that can induce fusion and / or destabilization of the plasma membrane, thus facilitating the intracellular release of complexed DNA. These assumptions paved the way to pursue studies aiming at improving the biological activity of lipoplexes, which allowed the identification of several critical parameters that affect their efficacy.

2.1. Liposome composition 2. From cationic liposomes to lipoplexes Although several attempts had been made to use conventional liposomes (neutral or negatively charged) for gene delivery, the limited efficiency of

2.1.1. Cationic lipids Over the last few years an enormous amount of work has been devoted to the development of novel formulations of cationic liposomes, namely through the synthesis of different cationic lipids with low

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294

toxicity and exhibiting different abilities to mediate gene transfer [5–11]. Besides DOTMA and DOTAP (two of the most popular cationic lipids), which are both two-chained amphiphiles, whose acyl chains are linked to the propyl ammonium group (through ether and ester bonds, respectively), numerous new lipids have become commercially available for transfection purposes. As can be observed in Fig. 1, where the chemical structures of the most representative cationic lipids are presented, these lipids differ in their hydrophobic moiety, in the linker group and in the positive valency of the polar headgroup. Although a direct correlation between the nature of the cationic lipids and their ability to mediate transfection and to develop cytotoxicity has been established, the nature of this dependence has not yet been completely clarified. DOSPA and DOGS, which are multivalent cationic lipids, form micellar rather than vesicular structures [12] and exhibit a higher efficacy in condensing DNA than monovalent lipids (e.g. DOTMA, DOTAP, DC-Chol, DMRIE). This property, however, does not necessarily lead to a higher transfection efficiency, since the intracellular dissociation of DNA from the complexes is expected to be more difficult [13]. In general, the transfection activity of cationic lipids decreases with increasing alkyl chain length and saturation. Shorter alkyl chain length favours higher rates of intermembrane transfer of lipid monomers and lipid membrane mixing [5,6]. However, in the case of DC-Chol, one of the most frequently used cationic lipids, the hydrophobic domain consists of a sterol backbone. A direct correlation between the nature of the linker group of the cationic lipids and their potential cytotoxicity was also demonstrated. Lipids with stable ether linkages (e.g. DOTMA, DMRIE) are more toxic than those containing labile ester linkages (e.g. DOTAP) [5,14,15].

279

different helper lipid like DOPC [16–18]. This fact has been attributed to the ability of DOPE to facilitate the formation of liposomes in conjuction with cationic lipids and to its tendency to undergo a transition from a bilayer to an hexagonal configuration under acidic pH, which may facilitate fusion with or destabilization of target membranes, in particular endosomal membranes [5,19,20]. In addition, the motional properties of DOPE, in contrast to DOPC, were correlated with the transfection potential of DOPE-containing complexes [17]. More recently, it was suggested that DOPE can also play a role in facilitating the disassembling of the lipidbased DNA formulations after their internalization and escape of DNA from endocytotic vesicles [21,22]. This was based on the assumption that the amine group of PE can interact with DNA phosphate groups, thus leading to weakening of the binding reaction between cationic lipids and DNA [21]. Although the benefits of using DOPE have been demonstrated empirically, recent work has shown that the choice of the helper lipid can dictate the structure and the activity of cationic liposome–DNA complexes as will be described below. Cholesterol has also been employed as a co-lipid to prepare cationic liposomes, resulting in the formation of more stable but less efficient complexes than those containing DOPE. In contrast, cholesterol-containing complexes have shown higher biological activity compared to complexes with DOPE when these complexes were utilized in vivo [7,23–25]. On the other hand, in an attempt to improve colloidal stability of lipoplexes both in vitro and in vivo, incorporation of poly(ethylene glycol) phospholipid-conjugates (PEG-PE) into the liposomal membrane has been explored [24,26]. The advantages and drawbacks associated with this approach will be discussed in more detail below.

2.2. Structure and size of cationic liposomes 2.1.2. Co-lipids The importance of associating a co-lipid to improve the ability of cationic liposomes to transfect cells has been demonstrated. In vitro studies clearly show that liposomes composed of an equimolar mixture of DOPE and cationic lipids (e.g. DOTMA, DOTAP) can mediate higher levels of transfection than those containing only the cationic lipid or a

The mode by which the structure and size of cationic liposomes affect the physico-chemical properties and biological activity of the lipoplexes is an important issue that needs to be clarified. In the great majority of transfection studies lipoplexes are prepared from large unilamellar cationic liposomes (LUVs), exhibiting sizes close to 100 nm, and from

280

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294

Fig. 1. Structures of some cationic lipids commonly used in gene therapy.

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294

small unilamellar cationic liposomes (SUVs) with sizes ranging between 20 and 100 nm. It has been demonstrated that for a given liposome composition, multilamellar liposomes (MLVs), which usually exhibit an average size ranging from 300 to 700 nm, when complexed with DNA mediate higher transfection activity than complexes prepared from SUVs [5,23]. However, a novel method to optimize the properties of lipid–DNA complexes was recently proposed, consisting of both extrusion of the liposomes (to obtain large unilamellar liposomes, LUVs) and controlled mixing of lipid and DNA. This procedure has been shown to result in lipoplexes exhibiting small sizes with a narrow distribution that present a high colloidal stability and adequate features for their in vivo use [27]. Recently, it was reported that lipoplexes resulting from SUVs or MLVs do not differ significantly in their size (ranging from 300 to over 2000 nm, depending on the composition of the medium used in their preparation) or in the extent of their cell association and uptake. Similarly, the transfection activity mediated by the resulting complexes was observed to be dependent only on the final size of the complexes and not on the type of liposomes used [28]. Although recent models proposed for the structure of the lipoplexes, which describe them as multilamellar aggregates, were based on studies using SUVs, recent findings suggest that in the process of lipoplex formation, DNA induces the generation of multilamellar liposomes from unilamellar vesicles [29]. Thus, the effect of using MLVs instead of SUVs on the final properties of the complexes remains to be clarified. The apparent discrepancy of results reinforces the need of further research focused on the understanding and control of pharmaceutical variables involved in the preparation of the cationic lipid-based gene delivery systems.

2.3. Mode of lipoplex formation: structure and morphology of the complexes It is well recognized that the mode of formation of the complexes strongly determines the final physicochemical features of the lipoplexes and, consequently, modulates their biological activity. In the last few years a significant effort has been devoted to gain insights into the parameters that affect the formation

281

and the resulting structure and morphology of the lipoplexes. Both experimental and theoretical studies have been performed and different models have been proposed for the cationic lipid–DNA complexes. Initially, based on light scattering results that indicated a slight size increase of the lipoplexes compared to the cationic liposomes, the calculation of the number of positive charges per liposome, as well as the assumption of complete charge neutralization for lipoplex formation, it was proposed that the lipoplexes resulted from the binding of four intact cationic liposomes to the plasmid–DNA strand mediated by electrostatic interactions [30]. Gershon and collaborators [31], by using electron microscopy, proposed a different model for the formation of the complexes according to which, at low lipid / DNA ( 1 / 2 ) charge ratios, cationic liposomes are adsorbed at the surface of DNA molecules forming aggregates that progressively surround large segments of DNA. The continuous addition of liposomes to a critical concentration and density results in DNA-induced membrane fusion and liposomemediated DNA collapse and condensation. According to these authors, these processes result in coating of DNA with a cationic lipid bilayer along the entire length of the plasmid. Similar observations were also made by Sternberg and collaborators [32] who, using freeze fracture electron microscopy, described the morphology of the lipoplexes as aggregates of cationic liposomes surrounding DNA molecules (designated ‘bead on string’ arrangements) that coexist with tubular structures composed of DNA molecules coated by lipid bilayers. This gave rise to the so-called ‘spaghetti-meatballs’ model [32]. More recently, an important breakthrough towards understanding the interactions between DNA and cationic liposomes as well as the structure of the resulting lipid / DNA complexes was provided by a combined in situ optical microscopy and X-ray diffraction approach [33]. The mixture of cationic liposomes with DNA results in a topological transition from the liposomal structure to a liquid-crystalline, condensed and globular structure, consisting of DNA monolayers, characterized by a uniform interhelical spacing, which are sandwiched between cationic lipid bilayers. This structure was designated the L Ca structure [33]. Using a simple theoretical model that allows the identification of the forces governing

282

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294

DNA adsorption on cationic lamellae, Dan [34] was able to predict the interhelical spacing of complexed DNA molecules and thus providing support to the ¨ experimental observations made by Radler and collaborators [33]. Replacement of DOPC by DOPE in DOTAPcontaining liposomes was shown to lead to a transition from the multilamellar structure (L Ca ) to an inverted hexagonal phase (H CII ) which consists of DNA coated by cationic lipid monolayers arranged on a two-dimensional hexagonal lattice [35]. The authors correlated the self-assembled structure of cationic liposome–DNA complexes with their colloidal stability and ability to mediate transfection, showing that while the presence of DOPC leads to the formation of stable complexes, cationic liposomes containing DOPE are more effective in mediating transfection. The proposal of the inverted hexagonal structure is in close agreement with previous studies performed by May and Ben-Shaul [36] and Dan [37] that described the so-called ‘honeycomb’ structures consisting of a bundle of DNA plasmid, each covered by a monolayer of cationic lipids arranged on a two-dimensional hexagonal lattice. Regarding the formation of the complexes, the benefits of using DOPE were attributed to the fact that DOPE introduces better matching in charge density between lipid and DNA surfaces [38], easier counterion release from the lipid surface by the DNA [20] and lower hydration of the lipid surface [39]. More recently, the role of DOPE in the mode of lipoplex formation was further investigated by performing circular dichroism measurements. These studies showed that the inclusion of DOPE in cationic liposomes induces, in addition to secondary conformational changes observed for pure DOTAP liposomes, a tertiary DNA transition that results from the embedding of DNA within the lipid columnar inverted-hexagonal assembly, which in turn provides the DNA with a given spatial organization and a fixed directionality. The authors proposed that this DNA packaging mode is interrelated and co-exists with short unbound segments of DNA which appear along the hexagonal assembly and converge into a highly condensed bundle characterized by a lefthanded chirality [39]. Overall, these findings contribute to rationalize the choice of the ‘helper’ lipid in the cationic liposomes.

Another interesting contribution to the understanding of the mode of lipoplex formation was reported by Templeton et al. [40]. Based on experimental evidence provided by cryo-electron microscopy and light scattering analysis, these authors described a novel morphology for cationic liposome–DNA complexes consisting of condensed DNA in the interior of invaginated liposomes between two lipid bilayers. As described above, these authors also found that the final structure of the lipoplexes is dependent on the liposome composition. Despite the various morphologies for the lipoplexes presented in the literature and summarized above, it is not yet possible to accurately define which are the conditions or factors that determine each of them. In fact, it is reasonable not to exclude the possibility that the different structures co-exist in the same preparation and that the observed differences can be attributed to the lack of control of the different variables involved in the complex preparation, as well as to the fact that different techniques have been utilized to study lipid bilayers and thus to assess lipoplex morphology. Evidence in support of this hypothesis has recently been reported by Fang and Yang [41] who have directly imaged DNA on lipid membranes by atomic force microscopy (AFM), showing distinct ordered domains.

2.4. Parameters affecting the physico-chemical properties of lipoplexes In addition to the morphology of the lipoplexes, other physico-chemical characteristics including size, charge density and colloidal stability are relevant properties of lipoplexes that determine their successful use both in vitro and in vivo. Therefore, understanding the parameters that modulate such properties is of crucial importance. Like in other systems, the physico-chemical properties of the lipoplexes depend on several thermodynamic as well as kinetic factors that affect complex formation, which therefore should be considered in their preparation. In this context, although several questions still remain to be clarified, it is well known that the concentrations of cationic lipid and DNA, the ionic strength and temperature of the suspending medium, the order of addition, the mixing rate of the components, and the extent of complex formation represent critical parameters [2,27,42,43].

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294

The relative proportion of cationic lipid and DNA determines the properties of the lipoplexes, namely size, surface charge (zeta potential), efficacy of complexation, colloidal stability and biological activity. Recent studies have shown that highly positively charged complexes, in which DNA is completely sequestered and condensed, exhibit an homogeneous size distribution (mean diameter between 100 and 450 nm). A similar size distribution is also observed when complexes are prepared with an excess of DNA over cationic lipids (i.e. negatively charged complexes), although in this case the presence of free DNA is generally observed [33,44–46]. On the other hand, complexes prepared from a lipid / DNA charge ratio of approximately 1 / 1 exhibit a neutral zeta potential, suggesting that all the cationic lipid molecules are neutralized by DNA [18,43,46,47]. Such neutral complexes are characterized by a heterogenous size distribution (mean diameter from 350 to 1200 nm) and usually present a much lower colloidal stability than those exhibiting an excess of net positive or negative charge. This can be attributed to a lack of electrostatic repulsive forces among the complexes that would prevent their aggregation [5,18,27,46,48]. The influence of lipid / DNA stoichiometry on the physico-chemical properties of the complexes becomes even more difficult to evaluate considering that, for a fixed lipid / DNA charge ratio, the increase in concentration of lipid and DNA results in a significant change of their size and colloidal stability, which can be attributed to enhanced precipitation at higher concentrations due to smaller interparticle separation [44]. Since thermodynamic parameters have been recognized to be involved in the formation of lipoplexes, the DLVO theory has been applied to understand how different parameters affect the final properties of lipoplexes [49]. In this regard, lipoplexes are usually prepared in low ionic strength solutions in order to decrease precipitation. Curiously, the precipitation effect caused by high ionic strength is particularly more pronounced during complex formation than in the stability of pre-formed lipoplexes, despite the fact that under the same conditions they still obey the DLVO theory. At low ionic strength, the attractive electrostatic forces required for lipoplex formation are enhanced, thus leading to a faster and more intense interaction between DNA and cationic liposomes, which seems to prevent the aggregation and

283

sedimentation of the complexes. Temperature has been described not to be a critical parameter affecting lipid–DNA complex formation as well as their final properties, as long as DNA denaturation is avoided. Regarding the effect of kinetic parameters, it is well established that rapid mixing of components results in small lipoplexes, while a very slow mixing often results in precipitation [50]. By analogy to crystal growth, this can be explained by the slow growth of a few nucleation embryos and critical ratios (neutral intermediates) on a local scale that favour aggregation. In an attempt to prevent the formation of neutral lipoplexes that tend to aggregate, empirical rules have been followed regarding the order of addition of the lipid and DNA. If positively charged complexes are desired, DNA should be added to a cationic liposome suspension. The inverse will be true to obtain negatively charged complexes.

3. Biological activity of the lipoplexes Much effort has been devoted to the development of cationic liposomes as gene transfer vectors and elegant strategies have been explored to enhance their biological activity. Although these systems have been used for gene therapy in animal models and tested even in clinical trials, they are still far from constituting viable alternatives to the use of viral vectors. Ideally, it would be desirable to produce complexes of small size with a narrow distribution, that ensure the protection of DNA, that present a neutral or negatively charged surface (to prevent non-specific interactions with blood components), that interact readily with the cells, preferably in a specific manner, and that mediate high levels of transgene expression without causing cytotoxicity. However, so far a combination of such properties confined to cationic liposome-based carriers has proven to be very difficult to achieve and the reported progress on the development of lipoplexes represents a compromise between their adequate features and biological activity. In fact, positively charged complexes have been described as being able to completely condense DNA and to mediate the highest levels of transfection, both in vitro and in vivo [18,46,51]. Favorable

284

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294

interactions with and binding to the cell surface as well as an efficient protection of the foreign DNA against nucleases can partially explain these observations [4,13,44]. Surprisingly, it has been recognized that within this type of complexes, those exhibiting large sizes (.200 nm) are more effective in mediating transfection than small complexes (50–100 nm) [40,44,52]. Whether these findings are due to more efficient lipoplex–cell interactions (presumably favored by a more extensive deposition of the large complexes at the cell surface), to the ability of certain size classes of the complexes to trigger cellular internalization events (like phagocytosis), or to the fact that more copies of the plasmids may be carried in the larger complexes, are still open questions. It was demonstrated recently that, for highly positively charged complexes, free liposomes coexist with the lipid–DNA complexes and play an important role in mediating transfection in vivo, namely upon i.v. administration in mice [51]. This enhancing effect was attributed to an increase of the retention time and efficient protection of DNA, and presumably to the ability of the free liposomes in promoting intracellular gene delivery by the lipid– DNA complexes [25,51]. In addition, it was shown that lipoplexes prepared at high cationic lipid / DNA (1 / 2) charge ratios were also able to overcome the inhibitory effect of serum on lipofection [40,53]. This resistance to the inhibitory effect of serum on transfection can also be achieved by prolonging the time of complex formation [54]. Curiously, this timedependent ‘maturation’ was only observed for monovalent cationic lipids. Moreover, the process of maturation, which was accelerated by high charge ratios, high concentration and high temperature, resulted in the formation of homogeneous particles with a mean diameter of 170–400 nm. In an attempt to improve the biological activity of the lipoplexes through modulation of physico-chemical properties, namely size and DNA protection, condensing DNA with polycations prior to its mixture with cationic liposomes was shown to be advantageous [55–57]. The addition of poly( Llysine), in contrast with small polycations, was shown to reduce the particle size of the resulting complexes, to render DNA resistant to nuclease activity and to enhance transfection of different cell lines [58]. More recently, the association of the polycationic peptide, protamine sulphate, with DNA

followed by addition of monovalent cationic liposomal formulations (DC-Chol and lipofectin) was found to increase transfection activities to levels comparable to those seen for the multivalent liposome formulations (lipofectamine) [57]. This effect was attributed to the ability of protamine to condense DNA efficiently, although the presence of certain amino acids resembling those of nuclear localization signals (NLS) in its sequence may also play a role in enhancing transfection, as will be discussed below. The resulting cationic lipid–protamine–DNA (LPD) formulation led not only to better protection of plasmid DNA against enzymatic digestion, but also to higher gene expression in mice upon i.v. injection, as compared with DOTAP–DNA complexes [55]. On the other hand, when protamine was added to DOTAP liposomes followed by addition of DNA not only were small condensed LPD particles (135 nm) obtained, but also a wide range of protamine / DNA ratios could be prepared without the problem of aggregation. This new protocol was shown to reduce the optimal dose of cationic lipid, to allow the use of large concentrations of each of the lipoplex components and to result in higher levels of in vivo gene expression compared to the particles prepared by the protocol mentioned above [56].

3.1. Lipoplex–cell interaction Despite the extensive use of cationic liposomes for gene delivery in vitro and in vivo, the mechanisms by which they deliver DNA into cells are not well understood. Gene transfer mediated by lipoplexes is strongly dependent on their physico-chemical features and on the cellular internalization mechanisms. Several obstacles, including the cytoplasmic, the endosomal and the nuclear membranes are recognized to restrict the successful application of cationic liposomes to mediate transfection. Therefore, in an attempt to achieve effective intracellular delivery and expression of the transgene into the desired cells, several strategies have been explored to surpass the referred barriers, some of which will be discussed below.

3.1.1. Cell uptake of lipoplexes Following binding to the cells, lipid–DNA complexes may be internalized via two different pathways: (1) through fusion with the plasma membrane

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294

promoted by the cooperative action of both cationic lipids and co-lipids such as DOPE, as suggested by initial studies in this field [4]; (2) through endocytosis involving clathrin and non-clathrin coated vesicles [46,59–61]. These two pathways are not mutually exclusive and may occur to varying extents depending on the type, confluence, and age of the cell. Moreover, it has been demonstrated that the physico-chemical properties of the lipoplexes (namely size and charge) may influence their mode of cellular uptake, leading in some cases to promotion of phagocytosis [22]. Most of the reported studies aiming at clarifying these mechanisms indicate that the endocytotic pathway plays a major role in the internalization of lipoplexes [46,59,62], resulting in efficient transfection. Nevertheless, we and others have provided evidence that membrane fusion is an important event in lipopolex formation as well as in their interaction with cells, being involved in the various stages of this process. Indeed it was demonstrated that fusion (as assessed by lipid mixing) between cationic liposomes occurs upon addition of DNA (Fig. 2). Additionally, it is generally accepted that the complexes have the ability to fuse with

Fig. 2. Fusion was evaluated by measuring the extent of Rh-PE fluorescence dequenching upon addition of DNA to a mixture of fluorescently labelled (4 mM) and unlabeled (16 mM) DOTAP liposomes (3-min incubation).

285

cellular membranes, including the plasma and endosomal membranes. However, the contribution of such a process was shown not to be relevant to intracellular gene delivery [46,63]. A possible explanation for the lack of correlation between transfection and extent of fusion is that the topology of the cationic liposome–DNA complexes does not allow the entry of the DNA into the cell through fusion with the plasma membrane but rather leads to the release of DNA to the extracellular medium. In addition to the above hypotheses on the mechanism of DNA entry, van der Woude and collaborators [64] have suggested the formation of pores in the plasma membrane, based on the hemolytic activity of lipoplexes. In this case, whether the DNA permeates through the pore or whether the breakdown of the permeability barrier of the cell membrane is only a by-product of the fusion reaction must be resolved. Based on the evidence that endocytosis represents the major pathway of lipoplex internalization, attempts have been made to enhance cell internalization by specifically targeting cationic lipid-based systems to cells, through the association of protein or peptide ligands or antibodies directed toward receptors that mediate endocytosis, such as lectins and asialoglycoprotein, asialofetuin, integrin, folate, Her2 / Neu and LDL receptors [65]. In this regard, we and others have demonstrated that association of transferrin to the lipoplexes enhanced transfection in a large variety of cells, including dividing and nondividing cells [18,66]. In general, transfection was most effective with the use of optimized lipid / DNA (1 / 2) charge ratios at which the complexes presented a net negative charge. Moreover, our studies also indicated that triggering internalization of the lipoplexes through a non-specific endocytotic process (namely phagocytosis) can also be achieved by associating certain proteins (e.g. albumin) to the lipoplexes [67]. It should be noted that promotion of the extent of binding and internalization of the lipoplexes does not necessarily translate into a similar enhancement of transgene expression. Several studies from various laboratories, including ours, have indicated that no correlation can be established between the extent of binding / cell association of the lipoplexes or the amount of DNA associated with the cell and the observed levels of transfection [65,68,69]. These observations reinforce the importance of other ele-

286

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294

ments that also play a role in determining the ability of carried DNA to be expressed.

3.1.2. Cytoplasmic delivery of DNA Following internalization, the next step is the release of the complexes from the endocytotic compartments into the cytoplasm. The release into the cytoplasm at this stage is crucial in order to avoid DNA degradation at the lysosomal level. How the lipoplex induces the disruption of the endosome in order to gain access into the cytoplasm is a question that still needs to be resolved. Recently, it has been proposed that the destabilization of the endosomal membrane by the internalized complexes induces flip-flop of anionic lipids from the cytoplasmic leaflet to the lumenal leaflet. The formation of a chargeneutral ion pair is thought to result in the displacement of the DNA from its complex with the cationic lipid leading to the release of DNA into the cytoplasm [60]. It should be noted that the presence of DOPE in the liposome formulations also plays an important role in mediating destabilization of the endosomal membrane since the acidification of the endosomal lumen activates the fusogenic properties of this lipid. More recently, it was demonstrated that following internalization of the complexes via endocytosis, DOPE-containing cationic liposomes promote fusion with the endosomal membrane under acidic conditions, thus allowing release of DNA into the cytoplasm [70]. Moreover, as stated above, DOPE may also be involved in helping the DNA dissociation from the lipoplexes due to the ability of its amine group to compete with cationic lipid for DNA phosphate groups, upon lipoplex internalization [21]. It is possible that pore formation at the endosomal membrane is also involved in the escape of the complexes or of free DNA into the cytoplasm. The association of pH-sensitive fusogenic peptides to polymer-based transfection reagents has been explored to enhance DNA delivery into the cytoplasm [71–73]. This strategy is expected to result in destabilization of the endosomal membrane upon acidification of its lumen, in a manner similar to that used by certain types of enveloped viruses to infect their target cells. Recently, we demonstrated that association of either the GALA fusogenic peptide [74] or the fusion peptide derived from the influenza virus hemagglutinin, to lipoplexes resulted in a

significant enhancement of transfection, this effect being particularly relevant for professional phagocytic cells (human macrophages) [66]. It remains to be clarified whether the benefits of such a strategy are not counteracted by potential immune responses elicited upon in vivo application. Although addition of lysosomotropic agents, such as chloroquine, or of compounds that promote osmotic swelling of the endosomes (e.g. sucrose or lipopolyamines) [65] was also shown to enhance transfection, it does not seem feasible to apply this approach for in vivo purposes.

3.1.3. Nuclear entry of DNA Once in the cytoplasm, DNA has to reach the nucleus and surpass the nuclear membrane for transcription to occur. As with the other steps involved in intracellular gene delivery by lipoplexes, the knowledge of DNA trafficking to the nucleus is still scarce. Assuming that DNA is lipid-free, a rapid movement to the nucleus appears to be required in order to avoid its degradation by nucleases, as indicated by the finding that free DNA microinjected into the cytoplasm is degraded within a short time [75]. In the absence of cell division, whether DNA penetrates the nuclear membrane through pores by a passive diffusion process or through active transport mechanisms involving, for example, its non-specific association with receptors for nuclear localization signal (NLS) peptides remains to be clarified. The former mode of entry is unlikely to occur, since pores act as a size exclusion sieve avoiding the free exchange of macromolecules larger than 70 kDa, which is significantly lower than the molecular weight of DNA. Another question that should be answered relates to the degree of DNA condensation / compaction when it reaches the nucleus. In this regard, it seems that partial coating of DNA with lipid would be advantageous at this stage, not only to reduce the size of the plasmid but also to ensure its protection against cytoplasmic nucleases. Moreover, it can be speculated that traces of cationic lipid still associated with DNA may play a role in the destabilization of the nuclear membrane. However, so far no evidence has been reported to support this hypothesis. In this context, it should be noted that microinjection of free plasmids into the nucleus of oocytes results in gene expression, whereas microin-

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294

jection of lipoplexes do not, suggesting that total lipid coating of DNA inhibits transcription [59]. Therefore, it would be desirable to have complete DNA uncoating at the interface of the nuclear membrane, just prior to DNA injection into the nucleus. In view of the very limited number of plasmid copies that are translocated into the nucleus when transfection is mediated by cationic liposomes, different strategies have been attempted to promote the nuclear entry of DNA. Most of these strategies have been inspired by the observation that certain proteins (e.g. histones, transcription factors, viral proteins, etc.) bearing Nuclear Localization Signal (NLS) peptides have the ability to be translocated into the nucleus by a receptor-mediated process [76]. In this regard, peptide sequences composed of a short stretch of 5–10 basic amino acids based on the SV40 core nuclear localization signal, are those that have been associated most frequently with non-viral vectors aiming at enhancing transfection. Different versions of applying this strategy have been reported, including the simple association of NLS peptides with lipoplexes [77] or its conjugation with polycations [78]. Recently, Zanta et al. [79] described an elegant procedure consisting of covalent coupling of a single NLS sequence to capped DNA constructs containing a hairpin oligonucleotide enriched with amino groups. These authors showed that with such a procedure the enhancement of transfection of a large variety of cell types, including non-dividing cells, was so significant (10–1000 fold) that only minute amounts of DNA were required to observe transfection. This represents an important progress in the use of lipid-based vectors, taking into account the putative cytotoxic effects of cationic lipids at higher concentrations. More recently, a less cumbersome but equally effective technique to couple NLS sequences to plasmids and to target them to the nucleus was reported [80]. This involves the combination of a peptide nucleic acid (PNA) with the SV40 NLS creating a bifunctional PNA–NLS peptide which has the ability to hybridize with certain target sites in the plasmid and to transport DNA into the nucleus. Both of the above referred investigators point out that possible saturation of NLS receptors at the nuclear pore complex (karyopherin-a / b proteins), the optimization of the number of NLS per plasmid, and

287

the target sites are crucial issues to be considered for a successful application of these strategies. Very recently, the nuclear import machinery was further exploited in an attempt not only to increase the amount of DNA that can reach the nucleus but also to take advantage of cell-specific nuclear translocation mechanisms [81]. Similarly to what has been observed for transcription factors, this involves the formation of a protein–DNA complex that targets the plasmid to the nucleus in a cell-specific manner. Use of nonclassical NLS containing the M9 sequence of heterogeneous nuclear ribonucleoprotein (hnRNP) A1 was reported as being a promising strategy to enhance transfection of non-dividing cells mediated by cationic liposomes [82]. In this study, the NLS peptides were crosslinked with cationic peptides (derived from a scrambled sequences of the SV40 T-antigen consensus NLS) in order to promote DNA binding of the M9 sequence, presumably in a nonspecific manner.

4. Biological stability of lipoplexes: implications for in vivo use Although cationic liposomes have proven to be promising systems in transfecting cells in tissue culture, it has been recognized that their in vitro efficiency does not correlate with their ability to deliver DNA after in vivo administration [7,8,11,24,83,84]. Reasons for these observations have been attributed to: (1) differences in the biology, functionality and complexity between cell cultures and animal models; and (2) changes in the colloidal properties of the cationic lipid–DNA complexes upon their interaction with cells and biological fluids. Some of the parameters involved in lipoplex-mediated intracellular gene delivery in vivo, such as the physical chemistry of the cell surface, the availability of surface receptors and the cell propensity for endocytosis are significantly more complex and depend on factors that are not present in a simple cell culture system. Moreover, factors affecting the bioavailability of the administered complexes, which are not modelled by the cell culture system, strongly determine the in vivo performance of such carriers. This includes avid interaction with serum components, resulting in colloidal instability, including both

288

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294

aggregation and dissociation of the complexes, and rapid elimination from blood circulation [85,86]. Association of serum proteins to lipoplexes was also reported to affect their mode of interaction with cells. Evidence on the effect of serum components on the physico-chemical properties of cationic lipid / oligonucleotide complexes and on their interactions with cells was recently presented showing that serum components can decrease oligonucleotide delivery into cells and promote dissociation of the complex [86]. Based on the finding that bovine serum albumin (BSA) can protect the complex against dissociation from certain polyanions present in serum, the authors suggest that coating of lipoplexes with neutral or negatively charged proteins may constitute a promising strategy to modulate their colloidal stability and transfection efficacy in the presence of serum. Our studies showing a significant enhancement of transfection and resistance to the presence of serum (as compared to plain lipoplexes) using negatively charged ternary complexes which were prepared through the association of human serum albumin to optimized lipid / DNA complexes provide support for this hypothesis [67]. Transferrin-lipoplexes were also found to mediate extensive transfection in the presence of serum [87]. We have observed that incubation of cationic liposomes with serum leads to an increase in their size which is mainly due to aggregation promoted by serum components [46]. The negatively charged serum proteins may form bridges between cationic liposomes, similarly to what occurs in the case of DNA addition to liposomes. However, in the latter case this increase in size was shown to result from DNA-induced fusion of liposomes, which was not observed for the effect of serum [46]. It should be noted that under the experimental conditions we used for lipoplex preparation, no aggregation was observed upon incubation with serum (20%) when complexes were prepared at high (1 / 2) charge ratios (4:1) (Fig. 3). Apparently, in this case the binding of serum proteins was not sufficient to overcome the electrostatic repulsion between the complexes. In this context, it is worthwhile to refer to the work of Ross and Hui [28] who found that the size growth of liposome–DNA complexes promoted by the presence of polyanions in the medium could be

arrested upon addition of serum, this being dependent on the charge ratio and type of liposomes used. Since it has been found that larger lipoplexes are those exhibiting the highest extents of cellular association and uptake, consequently leading to high levels of transfection, the authors suggest that the inhibition of transfection efficiency usually reported in the presence of serum can be attributed to impairment of lipoplex size growth as well as to interruption of the cationic lipid–DNA interaction, or prevention of lipoplex association with the cell membrane. Different approaches have been taken to modulate the properties of lipoplexes in vivo aiming to overcome some of the above referred limitations. As mentioned before, inclusion of cholesterol in the bilayer of cationic liposomes resulted in very active complexes upon in vivo administration [7,23,24,26,51,88]. Besides being resistant to the inhibitory effect of serum (up to 80%), the significant transfection activity achieved under these conditions was attributed to an improved cell binding and uptake of the complexes promoted by the presence of cholesterol [83]. Moreover, inclusion of cholesterol in the liposome composition enables the use of increased concentrations of lipid and DNA without affecting lipoplex stability. This, in turn, allows increased doses of DNA to be delivered and expressed. On the other hand, the choice of DOPE as the helper lipid for cationic liposomes was described to result in a decrease of the levels of transfection in vivo [88]. These findings suggest that the function of the helper lipid in liposomes is different in vivo from in vitro, thus supporting the prediction that the in vivo behavior cannot be established from in vitro data. The use of cationic liposomes modified with the phospholipid derivative of the polymer poly(ethylene glycol) (PEG-PE) may constitute a promising approach to the development of an efficient pharmaceutical carrier for systemic in vivo gene delivery. This strategy is expected to avoid the tendency of lipoplexes to form large aggregates, as well as to reduce their blood clearance upon i.v. administration, thus conferring prolonged circulation. In addition, shifting of the lipoplex biodistribution pattern to target tissues such as tumors, rather than the lung and heart, is also expected to take place.

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294

289

Fig. 3. The effect of serum on the mean diameter of DOTAP liposomes and of their complexes with DNA (A) and on liposome–cell fusion (B). Liposomes were prepared by extrusion of multilamellar vesicles (suspended in 10 mM Hepes, 150 mM NaCl, pH 7.4) through two stacked 100-nm polycarbonate filters. Liposome–DNA complexes were prepared at different lipid / DNA (1 / 2) charge ratios immediately before analysis. DNA-induced liposome fusion was evaluated in the absence or presence of different amounts of serum.

Incorporation of small amounts of PEG-PE in cationic liposome–DNA complexes was shown to stabilize the resulting particles for prolonged storage while maintaining their biological activity [24], and to result in complexes that are highly active in vivo [26]. Meyer et al. [89] have reported that incorporation of PEG-PE into cationic liposomes prevents

aggregation and increases the particle stability of cationic liposome–ODN complexes, without changing the ultrastructural characteristics of the original liposomes. These authors have also found that complexes of ODN with PEG-modified cationic liposomes are stable in human plasma and enhance cellular uptake of ODN in serum-supplemented cell

290

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294

culture medium. The advantage of such modification of the liposome surface was further demonstrated by the finding that conjugation of specific antibodies to the distal termini of PEG chains triggers efficient intracellular delivery of ODN to target cancer cells [89]. In contrast, in vitro studies performed in our laboratory clearly show that ternary complexes, prepared through the association of sterically stabilized cationic liposomes (DOTAP/ DOPE / DSPEPEG 2000 (1:1:0.02)) with transferrin and DNA at optimized lipid / DNA charge ratios, although being significantly more effective in mediating transfection than plain lipoplexes, were less active than nonsterically stabilized transferrin-lipoplexes in cell ˜ V. Slepushkin, R. Gaspar, M.C. culture (S. Simoes, ¨ ¨ ¸, unpublished Pedroso de Lima and N. Duzgunes data) (Fig. 4). These findings suggest that the successful use of such an approach can been restricted by certain limitations, namely: (1) lower reactivity of PEGylated-cationic liposomes with DNA, which may prevent the formation of active complexes; (2) unfavourable interactions with cells resulting in lower extent of cellular internalization of the complexes either by endocytosis or membrane fusion / poration and (3) possible impairment of the processes involved in the escape of DNA (either

complexed or in the free form) from the endosomes into the cytoplasm, including dissociation of the lipoplex and nuclear entry of DNA. Potential solutions to circumvent these limitations include the reduction of the density of PEG at the liposomal surface (while maintaining their Stealth properties), insertion of the polymer after the formation of cationic liposome–DNA complexes [90] and use of PEG–lipid conjugates with cleavable bonds that would allow the shedding of PEG molecules under acidic conditions such as those found in the endosomal lumen. In this context, it should be mentioned that a promising strategy was reported recently involving the efficient entrapment of plasmid DNA in cationic liposomes containing a PEG–ceramide construct by employing a detergent dialysis procedure [91]. The resulting PEG–lipid constructs were shown to regulate the fusion and aggregation of the cationic lipid–DNA complexes, and to confer long circulation times as well as a protective effect against degradation by DNase I and serum nucleases. Furthermore, this PEG–lipid construct has the ability to dissociate from the complexes when in the endosome, thus facilitating delivery of DNA into the cytoplasm. Besides the effects of the incorporation of poly(ethylene glycol) referred to above, the hydrophilic shield imposed by the polymer at the liposom-

Fig. 4. Effect of incorporating DSPE-PEG 2000 (2 mol% of phospholipid content) into cationic liposomes on their ability to mediate transfection. Cationic liposomes composed of DOTAP:DOPE with or without DSPE-PEG 2000 were complexed, in the absence or presence of transferrin, with 1 mg of pCMVluc at the indicated lipid / DNA charge ratios. Complexes were incubated with COS-7 cells for 4 h at 378C and transfection activity evaluated following a further 48-h incubation.

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294

al surface is also expected to lead to a significant reduction of the interaction of positively charged lipoplexes with serum proteins [91–93].

5. Concluding remarks Although over the last years it has been demonstrated that the synthesis of new cationic lipids and the design of new plasmid constructs with more efficient promotors / enhancers are valid strategies to improve transfection activity mediated by lipoplexes, the achieved progress is still far from being satisfactory. It seems, therefore, that development of new approaches to improve the features of lipoplexes aiming at facilitating their use in vivo (e.g. colloidal stability) is crucial to generate viable alternatives to viral vectors. As stated above, to achieve such a goal, attempts have been made to confer viral attributes to lipoplexes, namely through the association of certain proteins or peptides. Whether these improvements result in a system that, while exhibiting satisfactory ability to mediate in vivo transfection, would lead to such a complexity that could endanger its versatility and large scale production or could limit extended / repeated in vivo use due to immunogenicity, are important questions that remain to be addressed. Taking together the variables affecting the formation and structure of lipoplexes, their biological stability, and thus their biodistribution and pharmacokinetics, and those affecting their mode of interaction with cells, it will be rather laborious and difficult to design a non-viral vector capable of fulfilling the conflicting requirements imposed by each of the different stages involved in the gene delivery process. It seems therefore that a compromise between different properties has to be established to be able to achieve a carrier system that overall exhibits high transfection efficacy in vivo.

Acknowledgements This work was supported by Grant BIO4-CT972191 from the European Union and a NATO Collaborative Linkage Grant CLG.976106.

291

References ˜ ¨ ¨ ¸, [1] S. Simoes, P. Pires, M.C. Pedroso de Lima, N. Duzgunes Cationic liposomes as gene transfer vectors: Barriers to successful application in gene therapy, Curr. Opin. Mol. Ther. 1 (1999) 147–157. [2] D.D. Lasic, N.S. Templeton, Liposomes in gene therapy, Adv. Drug Deliv. Rev. 20 (1996) 221–266. [3] J.P. Behr, DNA strongly binds to micelles and vesicles containing lipopolyamines or lipointercalants, Tetrahedron Lett. 27 (1986) 5861–5864. [4] P.L. Felgner, T.R. Gadek, M. Holm, R. Roman, H.W. Chan, M. Wenz, J.P. Northrop, G.M. Ringold, M. Danielsen, Lipofection: a highly efficient, lipid-mediated DNA-transfection procedure, Proc. Natl. Acad. Sci. USA 84 (1987) 7413– 7417. [5] J.H. Felgner, R. Kumar, C.N. Sridhar, C.J. Wheeler, Y.J. Tsai, R. Border, P. Ramsey, M. Martin, P.L. Felgner, Enhanced gene delivery and mechanism studies with a novel series of cationic lipid formulations, J. Biol. Chem. 269 (1994) 2550–2561. [6] E.R. Lee, J. Marshall, C.S. Siegel, C. Jiang, N.S. Yew, M.R. Nichols, J.B. Nietupski, R.J. Ziegler, M.B. Lane, K.X. Wang, N.C. Wan, R.K. Scheule, D.J. Harris, A.E. Smith, S.H. Cheng, Detailed analysis of structures and formulations of cationic lipids for efficient gene transfer to the lung, Hum. Gene Ther. 7 (1996) 1701–1717. [7] J. Wang, X. Guo, Y. Xu, L. Barron, F.C. Szoka Jr, Synthesis and characterization of long chain alkyl acyl carnitine esters. Potentially biodegradable cationic lipids for use in gene delivery, J. Med. Chem. 41 (1998) 2207–2215. [8] I. Solodin, C.S. Brown, M.S. Bruno, C.Y. Chow, E.H. Jang, R.J. Debs, T.D. Heath, A novel series of amphiphilic imidazolinium compounds for in vitro and in vivo gene delivery, Biochemistry 34 (1995) 13537–13544. [9] C.J. Wheeler, P.L. Felgner, Y. Tsai, J. Marshall, L. Shukhu, S.G. Doh, J. Hartikka, J.B. Nietupski, M. Manthrope, M. Nichols, M. Plewe, X. Liang, J. Norman, A. Smith, S.H. Cheng, A novel cationic lipid reagent greatly enhances plasmid DNA delivery and expression in mouse lung, Proc. Natl. Acad. Sci. USA 93 (1996) 11454–11459. [10] T. Paukku, S. Lauraeus, I. Huhtaniemi, P.K. Kinnunen, Novel cationic liposomes for DNA-transfection with high efficiency and low toxicity, Chem. Phys. Lipids 87 (1997) 23–29. [11] C.M. Gorman, M. Aikawa, B. Fox, E. Fox, C. Lapuz, B. Michaud, H. Nguyen, E. Roche, T. Sawa, J.P. WienerKronish, Efficient in vivo delivery of DNA to pulmonary cells using the novel lipid EDMPC, Gene Ther. 4 (1997) 983–992. [12] J.P. Behr, B. Demeneix, J.P. Leoffler, J. Perez-Mutul, Efficient gene transfer into mammalian primary endocrine cells with lipolyamine-coated-DNA, Proc. Natl. Acad. Sci. USA 86 (1989) 6982–6986. [13] M.E. Ferrari, C.M. Nguyen, O. Zelphati, Y. Tsai, P.L. Felgner, Analytical methods for the characterization of

292

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294 cationic lipid–nucleic acid complexes, Hum. Gene Ther. 9 (1998) 341–351. P.L. Felgner, Y.J. Tsai, L. Sukhu, C.J. Wheeler, M. Manthorpe, J. Marshall, S.H. Cheng, Improved cationic lipid formulations for in vivo gene therapy, Ann. NY Acad. Sci. 772 (1995) 126–139. R.K. Scheule, J.A. George, R.G. Bagley, J. Marshall, J.M. Kaplan, G.Y. Akita, K.X. Wang, E.R. Lee, D.J. Harris, C. Jiang, N.S. Yew, A.E. Smith, S.H. Cheng, Basis of pulmonary toxicity associated with cationic lipid-mediated gene transfer to the mammalian lung, Hum. Gene Ther. 8 (1997) 689–707. S.W. Hui, M. Langner, Y.L. Zhao, E. Hurley, K. Chan, The role of helper lipids in cationic liposome-mediated gene transfer, Biophys. J. 71 (1996) 590–599. K.W.C. Mok, P.R. Cullis, Structural and fusogenic properties of cationic liposomes in the presence of plasmid DNA, Biophys. J. 73 (1997) 2534–2545. ˜ V. Slepushkin, R. Gaspar, M.C. Pedroso de Lima, S. Simoes, ¨ ¨ ¸, Gene delivery by negatively charged ternary N. Duzgunes complexes of DNA, cationic liposomes and transferrin or fusigenic peptides, Gene Ther. 5 (1998) 955–964. ¨ I. Koltover, T. Salditt, J.O. Radler, C.R. safinya, An inverted hexagonal phase of cationic liposome–DNA complexes related to DNA release and delivery, Science 281 (1998) 78–81. N.J. Zuidam, Y. Barenholz, Electrostatic and structural properties of complexes involving plasmid DNA and cationic lipids commonly used for gene delivery, Biochim. Biophys. Acta 1368 (1998) 115–128. P. Harvie, F.M.P. Wong, M.B. Bally, Characterization of lipid DNA interactions. I. Destabilization of bound lipids and DNA dissociation, Biophys. J. 75 (1998) 1040–1051. ˜ V. Slepushkin, R. Gaspar, M.C. Pedroso de Lima, S. Simoes, ¨ ¨ ¸, Mechanisms of gene transfer mediated by N. Duzgunes lipoplexes associated with targeting ligands and pH-sensitive peptides, Gene Ther. 6 (1999) 1798–1807. Y. Liu, L.C. Mounkes, H.D. Liggitt, C.S. Brown, I. Solodin, T.D. Heath, R.J. Debs, Factors influencing the efficiency of cationic liposome-mediated intravenous gene delivery, Nat. Biotech. 15 (1997) 167–173. K. Hong, W. Zheng, A. Baker, D. Papahadjopoulos, Stabilization of cationic liposome–plasmid DNA complexes by polyamines and poly(ethylene glycol)–phospholipid conjugates for efficient in vivo gene delivery, FEBS Lett. 400 (1997) 233–237. Y.K. Song, D. Liu, Free liposomes enhance the transfection activity of DNA / lipid complexes in vivo by intravenous administration, Biochim. Biophys. Acta 1372 (1998) 141– 150. B. Sternberg, K. Hong, W. Zheng, D. Papahadjopoulos, Ultrastructural characterization of cationic liposome–DNA complexes showing enhanced stability in serum and high transfection activity in vivo, Biochim. Biophys. Acta 1375 (1998) 23–35. O. Zelphati, C. Nguyen, M. Ferrari, J. Felgner, Y. Tsai, P.L. Felgner, Stable and monodisperse lipoplex formulations for gene delivery, Gene Ther. 5 (1998) 1272–1282.

[28] P.C. Ross, S.W. Hui, Lipoplex size is major determinant of in vitro lipofection efficiency, Gene Ther. 6 (1999) 651–659. [29] B.J. Battersby, R. Grimm, S. Huebner, G. Cevc, Evidence for three-dimensional interlayer correlations in cationic lipid– DNA complexes as observed by cryo-electron microscopy, Biochim. Biophys. Acta 1372 (1998) 379–383. [30] P.L. Felgner, G.M. Ringold, Cationic liposomes mediate transfection, Nature 337 (1989) 387–388. [31] H. Gershon, R. Ghirlando, S.B. Guttman, A. Minsky, Mode of formation and structural features of DNA–cationic liposome complexes used for transfection, Biochemistry 32 (1993) 7143–7151. [32] B. Sternberg, F.L. Sorgi, L. Huang, New structures in complex formation between DNA and cationic liposomes visualized by freeze–fracture electron microscopy, FEBS Lett. 356 (1994) 361–366. [33] J.O. Radler, I. Koltover, T. Salditt, C.R. Safinya, Structure of DNA–cationic liposome complexes: DNA intercalation in multilamellar membranes in distinct interhelical packing regimes, Science 275 (1997) 810–814. [34] N. Dan, Multilamellar structures of DNA complexes with cationic liposomes, Biophys. J. 73 (1997) 1842–1846. [35] I. Koltover, T. Salditt, J.O. Radler, C.R. Safinya, An inverted hexagonal phase of cationic liposome–DNA complexes related to DNA release and delivery, Science 281 (1998) 78–81. [36] S. May, A. Ben-Shaul, DNA–lipid complexes: stability of honeycomb-like and spaghetti-like structures, Biophys. J. 73 (1997) 2427–2440. [37] N. Dan, The structure of the DNA complexes with cationic liposomes-cylindrical or flat bilayers?, Biochim. Biophys. Acta 1369 (1998) 34–38. [38] N.J. Zuidam, Y. Barenholz, Electrostatic parameters of cationic liposomes commonly used for gene delivery as determined by 4-heptadecyl-7-hydroxycoumarin, Biochim. Biophys. Acta 1329 (1997) 211–222. [39] N.J. Zuidam, Y. Barenholz, A. Minsky, Chiral DNA packaging in DNA–cationic liposome assemblies, FEBS Lett. 457 (1999) 419–422. [40] N.S. Templeton, D.D. Lasic, P.M. Frederik, H.H. Strey, D.D. Roberts, G.N. Pavlakis, Improved DNA:liposome complexes for increased systemic delivery and gene expression, Nat. Biotech. 15 (1997) 647–652. [41] Y. Fang, J. Yang, Two-dimensional condensation of DNA molecules on cationic lipid membranes, J. Phys. Chem. B 101 (1997) 441–449. [42] D.D. Lasic, Recent developments in medical applications of liposomes: sterically stabilized liposomes in cancer therapy and gene delivery in vivo, J. Controlled Release 48 (1997) 203–222. [43] G.S. Koe, H.L. Way, G.M. Quetingco, J.G. Smith, T. Wedeking, R.W. Niven, The effect of mixing on the formation of DNA / liposome complexes, Pharm. Res. 14 (1997) S57. [44] E. Tomlinson, A.P. Rolland, Controllable gene therapy: pharmaceutics of non-viral gene delivery systems, J. Controlled Release 39 (1996) 357–372.

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294 [45] S.J. Eastman, C. Siegel, J. Tousignant, A.E. Smith, S.H. Cheng, R.K. Scheule, Biophysical characterization of cationic lipid:DNA complexes, Biochim. Biophys. Acta 1325 (1997) 41–62. ˜ ¨ ¨ ¸, M.C. [46] P. Pires, S. Simoes, S. Nir, R. Gaspar, N. Duzgunes Pedroso de Lima, Interaction of cationic liposomes and their DNA complexes with monocytic leukemia cells, Biochim. Biophys. Acta 1418 (1999) 71–84. [47] F.D. Ledley, Pharmaceutical approach to somatic gene therapy, Pharm. Res. 13 (1996) 1595–1614. [48] I.R. Mahato, A. Rolland, E. Tomlinson, Cationic lipid-based gene delivery systems: pharmaceutical perspectives, Pharm. Res. 14 (1997) 853–859. [49] D.D. Lasic, Gene delivery, in: D.D. Lasic (Ed.), Liposomes in Gene Delivery, CRC Press, Boca Raton, FL, 1997, pp. 53–66. [50] S. Hirota, C. Tros de Ilarduya, L.G. Barron, F.C. Szoka Jr, Simple mixing device to reproducibly prepare cationic lipid– DNA complexes (lipoplexes), Biotechniques 27 (1999) 286– 290. [51] J.G. Smith, T. Wedeking, J.H. Vernachio, H. Way, R.W. Niven, Characterization and in vivo testing of a heterogeneous cationic lipid–DNA formulation, Pharm. Res. 15 (1998) 1356–1363. [52] D. Deshpande, P. Blezinger, R. Pillai, J. Duguid, B. Freimark, A. Rolland, Target specific optimization of cationic lipid-based systems for pulmonary gene therapy, Pharm. Res. 15 (1998) 1340–1347. [53] J.P. Yang, L. Huang, Overcoming the inhibitory effect of serum on lipofection by increasing the charge ratio of cationic liposome and DNA, Gene Ther. 4 (1997) 950–960. [54] J.P. Yang, L. Huang, Time-dependent maturation of cationic liposome–DNA complex for serum resistance, Gene Ther. 5 (1998) 380–387. [55] S. Li, L. Huang, In vivo gene transfer via intravenous administration of cationic lipid–protamine–DNA (LPD) complexes, Gene Ther. 4 (1997) 891–900. [56] S. Li, M.A. Rizzo, S. Bhattacharya, L. Huang, Characterization of cationic lipid–protamine–DNA (LPD) complexes for intravenous gene delivery, Gene Ther. 5 (1998) 930–937. [57] F.L. Sorgi, S. Bhattacharya, L. Huang, Protamine sulfate enhances lipid-mediated gene transfer, Gene Ther. 4 (1997) 961–968. [58] X. Gao, L. Huang, Potentiation of cationic liposome-mediated gene delivery by polycations, Biochemistry 35 (1996) 1027–1036. [59] J. Zabner, A.J. Fasbender, T. Moninger, K.A. Poellinger, M.J. Welsh, Cellular and molecular barriers to gene transfer by a cationic lipid, J. Biol. Chem. 270 (1995) 18997–19007. [60] Y. Xu, F.C. Szoka Jr, Mechanism of DNA release from cationic liposome / DNA complexes used in cell transfection, Biochemistry 35 (1996) 5616–5622. [61] D.S. Friend, D. Papahadjopoulos, R.J. Debs, Endocytosis and intracellular processing accompanying transfection mediated by cationic liposomes, Biochim. Biophys. Acta 1278 (1996) 41–50. [62] I. Wrobel, D. Collins, Fusion of cationic liposomes with

[63]

[64]

[65]

[66]

[67]

[68]

[69]

[70]

[71]

[72]

[73]

[74]

[75]

[76]

293

mammalian cells occurs after endocytosis, Biochim. Biophys. Acta 1235 (1995) 296–304. T. Stegmann, J.Y. Legendre, Gene transfer mediated by cationic lipids: lack of a correlation between lipid mixing and transfection, Biochim. Biophys. Acta 1325 (1997) 71– 79. I. Van der Woude, H.W. Visser, M.B.A. Ter Beest, A. Wagenaar, M.H.J. Ruiters, J.B.F.N. Engberts, D. Hoekstra, Parameters influencing the introduction of plasmid DNA into cells by the use of synthetic amphiphiles as a carrier system, Biochim. Biophys. Acta 1240 (1995) 34–40. M.B. Bally, P. Harvie, F.M.P. Wong, S. Kong, E.K. Wasan, D.L. Reimer, Biological barriers to cellular delivery of lipidbased DNA carriers, Adv. Drug Deliv. Rev. 38 (1999) 291–315. ˜ S. Simoes, V. Slepushkin, E. Pretzer, R. Gaspar, M.C. ¨ ¨ ¸, Transfection of human Pedroso de Lima, N. Duzgunes macrophages by lipoplexes via the combined use of targeting ligands and fusogenic peptides, J. Leukoc. Biol. 65 (1999) 270–279. ˜ V. Slepushkin, P. Pires, R. Gaspar, M.C. Pedroso S. Simoes, ¨ ¨ ¸, Enhanced gene delivery by lipoplexde Lima, N. Duzgunes es associated with human serum albumin, Biochim. Biophys. Acta 1463 (2000) 459–469. P.W. Cheng, Receptor ligand-facilitated gene transfer: Enhancement of liposome-mediated gene transfer and expression by transferrin, Hum. Gene Ther. 7 (1996) 275–282. ˜ S. Simoes, ˜ P. Pires, S. Nir, M.C. Pedroso de Lima, T. Girao, Kinetic analysis of the initial steps involved in lipoplex–cell interactions: effect of various factors that influence transfection activity, Biochim. Biophys. Acta 1510 (2001) 136–151. A. Noguchi, T. Furuno, C. Kawaura, M. Nakanishi, Membrane fusion plays an important role in gene transfection mediated by cationic liposomes, FEBS Lett. 433 (1998) 169–173. E. Wagner, C. Plank, K. Zatloukal, M. Cotten, M.L. Birnstiel, Influenza virus hemagglutinin HA-2 N-terminal fusogenic peptides augment gene transfer by transferrin– polylysine–DNA complexes: Toward a synthetic virus-like gene-transfer vehicle, Proc. Natl. Acad. Sci. USA 89 (1992) 7934–7938. A. Kichler, K. Mechtler, J.P. Behr, E. Wagner, Influence of membrane-active peptides on lipospermine / DNA complex mediated gene transfer, Bioconjug. Chem. 8 (1997) 213– 221. C. Plank, B. Oberhauser, K. Mechtler, C. Koch, E. Wagner, The influence of endosome-disruptive peptides on gene transfer using synthetic virus-like gene transfer systems, J. Biol. Chem. 269 (1994) 12918–12924. N.K. Subbarao, R.A. Parente, F.C. Szoka, L. Nadasdi, K. Pongracz, pH-dependent bilayer destabilization by an amphipathic peptide, Biochemistry 26 (1987) 2964–2972. R.L. Page, S.P. Butler, A. Subramanian, F.C. Gwazdauskas, J.L. Johnson, W.H. Velander, Transgenesis in mice by cytoplasmic injection of polylysine / DNA mixtures, Transgenic Res. 4 (1995) 353–360. ´ ´ Signalling and transport P. Csermely, T. Schnaider, I. Szanto,

294

[77]

[78]

[79]

[80]

[81]

[82]

[83]

[84]

[85]

M.C. Pedroso de Lima et al. / Advanced Drug Delivery Reviews 47 (2001) 277 – 294 through the nuclear membrane, Biochim. Biophys. Acta 1241 (1995) 425–452. A.I. Aronsohn, J.A. Hughes, Nuclear localization signal peptides enhance cationic liposome-mediated gene therapy, J. Drug Target. 5 (1997) 163–169. C.K. Chan, D.A. Jans, Enhancement of polylysine-mediate transferrinfection by nuclear localization sequences: polylysine does not function as a nuclear localization sequence, Hum. Gene Ther. 10 (1999) 1695–1702. M.A. Zanta, P.B. Valladier, J.P. Behr, Gene delivery: A single nuclear localization signal peptide is sufficient to carry DNA to the cell nucleus, Proc. Natl. Acad. Sci. USA 96 (1999) 91–96. ´ A.J. Mohamed, C.I.E. Smith, A peptide nucleic L.J. Branden, acid-nuclear localization signal fusion that mediates nuclear transport of DNA, Nat. Biotech. 17 (1999) 784–787. J. Vacik, B.S. Dean, W.E. Zimmer, D.A. Dean, Cell-specific nuclear import of plasmid DNA, Gene Ther. 6 (1999) 1006– 1014. A. Subramanian, P. Ranganathan, S.L. Diamond, Nuclear targeting peptide scaffolds for lipofection of nondividing mammalian cells, Nat. Biotech. 17 (1999) 873–877. K. Crook, B.J. Stevenson, M. Dubochet, D.J. Porteous, Inclusion of cholesterol in DOTAP transfection complexes increases the delivery of DNA to cells in vitro in the presence of serum, Gene Ther. 5 (1998) 137–143. N.K. Egilmez, Y. Iwanuma, R.B. Bankert, Evaluation and optimization of different cationic liposome formulations for in vivo gene transfer, Biochem. Biophys. Res. Commun. 221 (1996) 169–173. C. Plank, K. Mechtler, F.C. Szoka Jr, E. Wagner, Activation of the complement system by synthetic DNA complexes: a potential barrier for intravenous gene delivery, Hum. Gene Ther. 7 (1996) 1437–1446.

[86] O. Zelphati, L.S. Uyechi, L.G. Barron, F.C. Szoka Jr, Effect of serum components on the physico-chemical properties of cationic lipid / oligonucleotide complexes and on their interactions with cells, Biochim. Biophys. Acta 1390 (1998) 119–133. ¨ ¨ ¸, Efficient gene transfer by [87] C. Tros de Ilarduya, N. Duzgunes transferrin lipoplexes in the presence of serum, Biochim. Biophys. Acta 1463 (2000) 333–342. [88] Y.K. Song, F. Liu, S. Chu, D. Liu, Characterization of cationic liposome-mediated gene transfer in vivo by intravenous administration, Hum. Gene Ther. 8 (1997) 1585– 1594. [89] O. Meyer, D. Kirpotin, K. Hong, B. Sternberg, J.W. Park, M.C. Woodle, D. Papahadjopoulos, Cationic liposomes coated with polyethylene glycol as carriers for oligonucleotides, J. Biol. Chem. 273 (1998) 15621–15627. [90] P.S. Uster, T.M. Allen, B.E. Daniel, C.J. Mendez, M.S. Newman, G.Z. Zhu, Insertion of poly(ethylene glycol) derivatized phospholipid into pre-formed liposomes results in prolonged circulation time, FEBS Lett. 386 (1996) 243–246. [91] N. Maurer, B. Mui, K.W.C. Mok, Q.F. Akhong, P.R. Cullis, Lipid-based systems for the intracellular delivery of genetic drugs, Mol. Membr. Biol. 16 (1999) 129–140. [92] Y.P. Zhang, L. Sekirov, E.G. Saravolac, J.J. Wheeler, P. Tardi, K. Clow, E. Leng, R. Sun, P.R. Cullis, P. Scherrer, Stabilized plasmid–lipid particles for regional gene therapy: formulation and transfection properties, Gene Ther. 6 (1999) 1438–1447. [93] J.J. Wheeler, L. Palmer, M. Ossanlou, I. MacLachlan, R.W. Graham, Y.P. Zhang, M.J. Hope, P. Scherrer, P.R. Cullis, Stabilized plasmid–lipid particles: construction and characterization, Gene Ther. 6 (1999) 271–281.