Available online at www.sciencedirect.

com

Journal of Controlled Release 123 (2007) 184 – 194

www.elsevier.com/locate/jconrel

Review
Lipoplex morphologies and their influences on transfection
efficiency in gene delivery
Baichao Ma a,b,c , Shubiao Zhang b,⁎, Huiming Jiang b , Budiao Zhao b , Hongtao Lv a,b,c
a
Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, Liaoning, China
b
SEAC-ME Key Laboratory of Biotechnology and Bioresources Utilization, Dalian Nationalities University, Dalian 116600, Liaoning, China
c
Graduate University of Chinese Academy of Sciences, Beijing 100049, China
Received 12 October 2006; accepted 9 August 2007
Available online 24 August 2007

Abstract

Cationic lipid-mediated gene transfer is widely used for their advantages over viral gene transfer because it is non-immunogenic, easy to
produce and not oncogenic. The main drawback of the application of cationic lipids is their low transfection efficiency. Many reports about
transfection efficiency of cationic lipids have been published in recent years. In this review, the current status and prospects for transfection
efficiency of different morphologies of lipoplexes are discussed. High transfection activity will be acquired for HCII structure when membrane
fusion is dominant, but when serum is present LCα lipoplexes show great superiority for their inhibition dissociation by serum during lipoplexes
transporting. Increasing DOPE often gains high activity for the HCII structure promoted by DOPE. High lipofection will be gained from large
lipoplexes when endocytosis is dominant, because large particles facilitate membrane contact and fusion. We suggest morphologies of lipoplex
should be characterized at two levels, lipoplex size and self-assemble structures of lipoplexes, and understanding these would be very important
for scientists to prepare novel cationic lipids and design novel formulations with high transfection efficiency.
© 2007 Elsevier B.V. All rights reserved.

Keywords: Lipoplex morphology; Gene delivery; Transfection efficiency; Cationic lipoplexes

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
2. Formation of cationic lipoplexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
2.1. Driving force and DNA condensation during lipoplex formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
2.2. Lipoplex structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
3. Factors affecting phase behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
3.1. Characteristics of cationic lipid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
3.2. Ionic strength and temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
3.3. Helper lipid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
3.4. DNA concentration (or lipid to DNA ratio) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
3.5. Zeta potential. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
4. Relationship between transfection efficiency and lipoplex morphologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
4.1. Lipoplexe structure-transfection efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
4.2. Lipoplex size-transfection efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
5. Cytotoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190

⁎ Corresponding author. Tel.: +86 411 876 56148; fax: +86 411 876 44496.
E-mail address: [email protected] (S. Zhang).

0168-3659/$ - see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jconrel.2007.08.022
 B. Ma et al. / Journal of Controlled Release 123 (2007) 184–194 185

6. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191

1. Introduction including the driving force of lipoplex formation, DNA con-

densation, and various structures formed due to these factors.
As a promising strategy for the treatment of many inherited The parameters governing the preferred spatial geometry
and acquired diseases, gene therapy which is defined as the focusing on LCα and HCII and transfection efficiency of dif-
genetic modification of cells for therapeutic benefit has attracted ferent structures will also be discussed. Lipoplex size is another
many researchers during the past several decades [1]. The aim parameter of morphology influencing on transfection efficiency,
for gene therapy is to deliver healthy exogenous genic drugs which should be paid much attention, too. We hope the know-
such as plasmid DNA and single-strand oligonucleotides to ledge of transfection activity of lipoplexes with different mor-
replace a missing gene that otherwise have a normal makeup phologies occurring in lipoplexes formation would provide
so as to cure genetic diseases, for instance, cystic fibrosis, valuable insights into the designing of novel cationic lipids and
malignant melanoma, and gaucher's disease [2,3]. Based on formulations with high transfection efficiency, thus yielding
these factors, reliable and efficient vectors delivering exogenous important information for the development of novel vectors for
genes into target cells are urgently awaited. Non-viral vectors in vivo applications.
are widely used for their advantages over viral vectors in recent
years, because they are safe and cheap, easy to manufacture in 2. Formation of cationic lipoplexes
lager scale, and they can also deliver large pieces of DNA [4–8].
During the past years, designing new non-viral system for 2.1. Driving force and DNA condensation during lipoplex
DNA transfer has become an interesting field attracting more formation
and more researchers [9]. Among non-viral vectors, since
cationic lipid/DNA complex (lipoplex) was first introduced by Though the transfection efficiency of more and more novel
Felgner et al. [10] who used N-[1-(2,3-dioleyloxy)propyl]-N,N, cationic lipoplexes has been enhanced compared with tradi-
N-trime-thylammonium chloride (DOTMA) to carry out a tional non-viral delivery systems, the mechanisms of lipoplex
DNA-transfection protocol, cationic lipids and polycations used formations are still being investigated all the time. Felgner et al.
for DNA transfer have been widely investigated [3,11–16]. [33] concluded that the formation of lipoplexes was a result of
Some helper lipids (such as dioleyl phosphatidylethanolamine the electrostatic interaction between cationic charge from lipids
(DOPE), cholesterol or dioleoyl phosphatidyl choline (DOPC)), and anionic charges from DNA. Kreiss et al. [34] reported that
usually neutrally charged, are often employed with cationic lipoplex morphologies were determined by the competitive
lipids in order to gain high transfection efficiency [17]. The interactions between electrostatic forces and elasticity forces
primary reason for current failure and frustration is the in- driven by the lipid hydrophobic moiety. A single plasmid is
adequacy of efficiency of cationic lipids. Tremendous efforts surrounded by sufficient cationic lipids to completely neutralize
are currently paid worldwide to elucidate this problem, in- the negative charges of DNA and provide a complex with a net
cluding the continuous synthesis of new cationic lipids and new positive charge that can associate with the negatively charged
approach of complexes formulations [18–25]. For a given cell surface of cells, which may be correlated with effective trans-
type, the transfection efficiency is dependent on the morpho- fection [10]. To understand the interaction between lipoplexes
logical appearance (or on the packing morphology), besides and cells, it is important to understand the thermodynamics
structural parameters of the lipoplexes [26]. Cationic lipoplexes underlying lipoplex formation.
are often heterogeneous with respect to shape, size and com- In fact, lipoplex formation is a highly dynamic event, in-
position. Many types of morphologies have been reported, volving the relatively uncontrolled interaction between lipo-
including beads on a string structure [11], spaghettis or meat- somes and plasmid DNA [17,24]. Pozharski et al. [35] proposed
balls structure [27], multilamellar structure LCα and inverted that lipoplex formation was endothermic with less than 1 kcal
hexagonal phase structure HCII [28]. In addition, a map–pin absorbed per mole of lipid or DNA charge, and DNA–DNA
structure [29] and a sliding columnar phase [30] have also been repulsion dominated the formation enthalpy of cationic lipid/
identified. The low transfection efficiencies of current cationic DNA complex, simultaneously the lipoplex formation was
lipids may be largely due to the result of incomplete under- driven by an increase of entropy associating with the release of
standing about transfection-related morphology of lipoplex. tightly bound counterions from DNA and the lipid bilayers
Niculescu-Duvaz et al. [18] expatiated in great details the [36,37].
relationship between the structure of the cationic lipid and the Two processes occur during the formation of cationic
transfection activity. Influence of plasmid DNA topology [31] lipoplexes, namely, DNA-induced membrane fusion and lipo-
and liposomal formulation [32] on the transfection properties of somes-induced cooperative DNA collapse. The latter is a key
lipoplexes was discussed elsewhere. In this review we pay our event, and leads to clusters bound to the DNA molecules struc-
attention to the effects of lipoplex morphologies on transfection ture (spherical shape of vesicles can be discerned yet) and short
activity. We will describe the process of lipoplex formation, rod-like structure in which condensed DNA are completely
186 B. Ma et al. / Journal of Controlled Release 123 (2007) 184–194

encapsulated within a smooth lipid bilayers [38]. A concentric

“ring-like” pattern, arising from cationic liposome-induced
DNA compactation, was recognized by Scarzello et al. [39].
DNA molecules undergo a dramatic condensation to a compact,
usually highly ordered toroidal structure, such as so-called ψ-
DNA, a separate phase that is organized very much like liquid
cholesteric crystals, with left-handed chirality [40], because of
the condensation and modification of the cationic lipids [41].
During the complex formation, both lipid and DNA undergo a
complete topological transformation into compact quasisphe-
rical complex particles with ∼0.2 mm diameter, and they are
easy to form string-like colloidal aggregates, inside of which the
complexes have an ordered multilamellar structure (LCα), when
the complexes are neutrally charged [16,42].

2.2. Lipoplex structures

Two fundamentally different types of models have been

employed in order to interprete the cationic lipoplex structures, an
“external” model, in which DNA is adsorbed onto the surface of
cationic liposomes, and an “internal” model, in which the DNA is
surrounded or “coated” by a lipid envelope [43]. In one model of
lipopelx structures, as exemplified by Felgner et al. [11], the DNA
Fig. 2. Schematic of (a) the lamellar structure (LCα) of cationic lipid/DNA
is bound electrostatically to the outside of the cationic lipid
lipoplexes, where the DNA rods are sandwiched between the lipid bilayers;
vesicles, i.e., it is adsorbed onto the vesicle, gaining a beads on a (b) the inverted hexagonal structure (HCII), where DNA rods are coated with
string structure. Cry-TEM applied to DOTAP or DDAB and DNA lipid monolayer arranged on a hexagonal lattice; and (c) the intercalated
complexes containing DOPE revealed entrapment of DNA into hexagonal structure (HI), where DNA rods are covered by three honeycombs of
aggregated multilamellar structures at low lipid-DNA ratios. An lipid micelles that are arranged on a hexagonal lattice [49,50].
excess of cationic surfactant to DNA in terms of charge, leads to
entrapment of the DNA molecules between the lamellas in gonal structures containing inverted hexagonal HCII (Fig. 2 b)
clusters of aggregated multilamellar structures, independent on and hexagonal HI (Fig. 2 c). Most complexes assume lamellar
the choice of surfactant [44]. By a three-step mechanism, several phase (LCα) structures with DNA sandwiched between the
single bean-like structures are aggregated in a parallel or cationic lipids [19,51,52]. A report proposed that a transient
perpendicular orientation in the end, and their diameters match spaghetti-like structure existed between LCα and HCII, and it
the length of single plasmid DNA (approximately 200 nm), possibly served as a precursor to the phase HCII [53]. Cryo-
implying that the condensing effect is fairly small [45]. Vesicles electron images of LCα and HCII structures can be seen in Fig. 3.
made from pure pyridinium-based lipids analog SAINT-2 (Fig. 1) Formation of lipoplexes, even if it may seem very simple from
associate rapidly with plasmid DNA, leading to the formation of its concept, could determine the morphology and transfection
ellipsoid-shaped particles. When vesicles consisting of a mixture potential of lipoplexes. In a mechanism of two-step lipoplex
of SAINT-2 and DOPE (1:1 molar ratio) were incubated with formation processes, the first step is attributed to the electrostatic
plasmids, round-shaped lipoplexes bearing smoother appearance binding of DNA to the liposome surface, and the second one
were formed, whereas free DNA was not visible [46]. involves fusion and rearrangement of the liposomes [54]. Though
A recent study revealed that lamellar structure was present lipoplex morphologies are influenced by the second step, it is not
during the condensation and transport of the DNA, whereas a the dominant factor governing lipoplex morphologies during
more aggressive inverted hexagonal structure was formed upon lipoplex formation.
contact with the cell membrane [47]. The appearance of the
lipoplexes is often a highly ordered tubular structure when they 3. Factors affecting phase behavior
are endocytosed by cells and assume perinuclear localization in
these endosomes [48]. The observed ordered cationic lipoplexes Now it is known that lipoplex structure and stability depend
mainly have multilamellar structure LCα (Fig. 2 a) and hexa- on many parameters, including the molecular characteristics of
the cationic and helper co-lipids [55] and ratios of liposomes to
DNA [38]. Some thermodynamic properties of lipoplex
structures limiting to the two primary structures (LCα and
HCII) and relating to their phase diagram have been described
previously [53]. Now we will discuss the parameters governing
the preferred spatial geometry focusing on LCα and HCII in the
Fig. 1. Molecular structure of SAINT-2. following sections.
 B. Ma et al. / Journal of Controlled Release 123 (2007) 184–194 187

less packing parameter, P = V / a · l, in which “V” denotes the

hydrophobic chain volume , “a” the optimal cross-sectional
headgroup area and “l” the length of the hydrophobic tails ( e.g.
1/2 b P ≤ 1 → lamellar organization; P N 1 → inverted structures)
[57,58]. Lipid containing four alkyl chains exhibits high ability
to induce HCII phases for their large acyl chain cross-section
[24]. Limiting to pyridinium amphiphiles, the lipoplex tends to
assume a well defined HCII phase structure with the increase of
lipid alkyl chain length. It is established that the introduction of
a double bond leads to increased preference for the HCII
organization, which may attribute to an increase in the cross-
sectional area in the hydrocarbon chain region [59,60], but
variations in the position of the unsaturation do not lead to any
morphological differences [55].

3.2. Ionic strength and temperature

The LCα phase keeps stable in a wide range of ionic strength

and temperatures. The mobility of the lipid chains which is
believed related to the transition from LCα phase to HCII phase
increases when temperature is elevated [55]. At high tempera-
ture, it results in decrease of “a” as a consequence of the decrease
in the hydration shell and increase of hydrophobic chain volume
“V” due to the increased mobility of the alkyl chains [39]. Upon
an increase in ionic strength, mixtures aggregate into a relatively
stable, dispersed cubic phase and finally undergo a quantitative
Fig. 3. Cryo-electron microscopy images of lipoplexes. (a) condensed lamellar transition to the HCII phase upon DNA complexation. With
phase; (b) inverted hexagonal (HCII) structure; (c) Fourier transform pattern
derived from (b), (b) and (c) shown that HCII structure was present in different
increasing of ionic strength (such as presence of NaCl), the main
orientations, and it highlights the columnar characteristics and the hexagonal consequence is the screening of the electrostatic repulsion
order, respectively. Reprinted with permission from J. Am. Chem. Soc. 2003, between the headgroups of the lipids, then it leads to a decrease
125, 1551–1558. Copyright 2003 American Chemicial Society. of cross-sectional headgroup area “a” and, consequently, an
increase of packing parameter “p”, consistent with the transition
3.1. Characteristics of cationic lipid to the HCII phase [39,55]. Recently, using a Nile Red-based
assay, Wasungu et al. [61] showed that lipoplexes made from
The relationship between the phase behavior of cationic sugar-based gemini surfactants GS1 and GS2 (Fig. 4) underwent
lipids and the final morphologies after DNA complexation is a lamellar phase to non-inverted micellar phase (HI) transition
especially important for the prediction of transfection efficien- upon decreasing the pH from neutral to mildly acidic, differing
cies on the basis of the molecular structure of the carriers [39]. A from that of SAINT-2/DOPE, a classical HCII phase forming
curvature theory has been reported in which the “shape” of the system. The two lipoplexes of GS1 and GS2 are capable of
molecule that determines the natural curvature of the mem- remaining a good colloidal stability in salt and in serum at
brane, Co = 1/Ro, will also determine the actual curvature, physiological pH owing to their lamellar organization consistent
C = 1/R in the lipid systems. Herein “R” denotes the actual with a prolonged stability in vivo [62].
radius and “Ro” the natural radius of curvature of the lipid
monolayer. The actual curvature describes the structures of 3.3. Helper lipid
lipoplex, for example, when C = 0 lamellar LCα structure is
favored, Co b 0 an inverted hexagonal H C II structure is The presence of the helper lipid DOPE could increase the
preferred, and Co N 0 hexagonal HI structure is dominant [56]. average packing parameter and lead to mixed bilayers [55]. Most
In the HI phase, tubular lipid micelles are arranged on a
hexagonal lattice and the DNA rods, depending on the extent of
hydration, are arranged on a honeycomb lattice in the interstices
of the lipid micelle arrangement. Space of surplus in the HI
structure is filled with the headgroups, water, and counterions
[50].
Using sunfish amphiphiles Scarzello et al. [39,55] conclud-
ed, differing only in their hydrophobic domain, the dependence
of the aggregate morphology on the size of the hydrophobic
region of an amphiphile could be expressed by the dimension- Fig. 4. Molecular structures of GS1 and GS2.
188 B. Ma et al. / Journal of Controlled Release 123 (2007) 184–194

studies have shown that lipoplexes containing the non-bilayer- DNA ratios, and the structure of lipoplexes with a positive zeta
phase-preferring lipid DOPE or cholesterol would promote HCII potential is different from lipoplexes with a negative zeta potential
organization [17]. A transition from the LCα phase to the HCII [72]. Lipoplexes with positive zeta potential might correspond to
phase could be expected by increasing weight fraction of DOPE, the aggregated multilamellar structure (LCα) [66]. Negative zeta
via controlling the spontaneous radius of curvature “Ro” of the potential leads to free plasmids or protruding DNA-strings. It is
lipid layers, favored by the elastic free energy [56]. On the other also possible that the “external” structure is formed at low lipid:
hand, the ability of DOPE to control the structure of the cationic DNA charge ratios, in which the majority of the DNA is bound to
lipoplexes also depends on the molecular shape of the cationic the exterior of the complex and not encapsulated within a lipid
lipid [63], which is determined by the ratio of head group area coat [44,72,74]. However, zeta potential is just one fact affecting
over hydrophobic area [64]. Conversely, it would be expected transfection activity.
that lipids such as dioleoylphosphatidylcholine (DOPC) would There are still many other parameters affecting lipoplex
hinder the ability of cationic lipids to induce non-bilayer morphologies, including processing parameters [75,76], formu-
structure [60]. By interaction with anionic vesicles, to simulate lation procedure [77], and the ambient conditions [43] during
lipoplexes–endosomal membrane interaction, SAINT-2/DOPE lipoplexes formation and DNA incubation time [27]. Although
lipoplex showed a perfect hexagonal phase, whereas SAINT-2/ influences of these parameters aren't neglected, we speculate
DPPE lipoplexes formed a mixed lamellar–hexagonal phase molecule shape of cationic lipid, helper lipids (especially the
when containing lamellar-phase-forming dipalmitoylphosphati- nonbilayer-phase-preferring lipid DOPE and lamellar-phase-
dylethanolamine (DPPE) [17]. preferring DPPE) and lipid to DNA ratio (or charge ratio of
positive and negative) are three key factors governing lipoplex
3.4. DNA concentration (or lipid to DNA ratio) structures during their formation. The most general principle
that arises from above is that if one can correlate the efficiency
Globular condensates are formed generally at high lipid/DNA of transfection with the morphology of the lipoplex, one will be
ratio in terms of charge [38,44,65]. With the addition of DNA, able to make lipofection predictions based on known parameters
larger chain-like structures emerge subsequently and they are governing lipoplex morphologies.
linked by an invisible thread revealed by Brownian motion of
these globules [66]. Large unilamellar vesicles (LUV) and 4. Relationship between transfection efficiency and lipoplex
sedimented multilamellar vesicles (sMLV), opposed to small morphologies
unilamellar vesicles (SUV), formed lipoplexes that existed as a
single phase over a relatively broad range of mixing (+/−) ratios, Lipofection of the lipoplexes may occur though several stages:
which was revealed by sedimentation in sucrose density gradients (a) protection of DNA (prevention dissociation by serum);
[67]. Precompaction of DNA with the single-tailed cationic (b) transporting of lipoplexes to target tissues; (c) cellular uptake;
surfactant CTAB, followed by complexation with DMTAP: (d) fusion of an internalized cationic lipoplex with the endo-
DOPE (1:1) liposomes, led to ternary complexes with a multi- some membrane; (e) escape of DNA from the endosome;
lamellar organization proved by electron microscopy data. The (f) transporting of DNA to the nuecleus; (g) nuclear entry and
study about complexes formed by dsDNA with CTAB, in the (h) release of DNA from lipoplexes [18,78,79]. Transfection for
presence of the cosurfactant hexanol ( a membrane soluble some of the above stages can be enhanced by some structures
cosurfactant) [28], using small angle X-ray diffraction, revealed or sizes of lipoplexes, and other stages may be promoted by
an intercalated hexagonal (HI) → lamellar (LCα ) → inverted lipoplexes with other structures or sizes.
hexagonal (HCII ) transformation on increasing hexanol content
at low DNA concentrations. A HCII → LCα transformation was 4.1. Lipoplexe structure-transfection efficiency
observed as increasing of DNA concentration at higher hexanol
content. Thus, an LCα → HCII → LCα reentrant phase transitions In vivo, highly stabilized lipoplexes, in particular, those that
was obtained as DNA concentration increasing at higher hexanol have some projections, map–pin structures lead to high
content [68]. Detailed parameters of two hexagonal structures HI transfection efficiency [29,78]. Worm-like structures or tubes of
and HCII were described further using quantitative analysis of the lipoplexes containing DOTAP/MOG, DOTAP or DOTAP/PC,
diffraction data elsewhere [69]. A report revealed that cationic and DOTAP/DOPE were observed in freeze-fracture electron
lipoplex structures changed with various lipid to DNA charge micrographs. The tubes are extremely short and appear bead-like
ratio but were not affected at above isoelectric point, where the in lipoplexes containing DOTAP/MOG, slightly longer in those
charges on the DNA exactly matched those on the cationic lipids containing DOTAP or DOTAP/PC, and extensively elongated in
[37,70]. DOTAP/DOPE lipoplexes. The Extensively elongated tubular
structures are not required for their inability in terms of trans-
3.5. Zeta potential fection activity [74]. Small Angle X-Ray Scattering and Fluore-
scence Correlation Spectroscopy measurements also confirmed
Zeta potential is an indirect measurement of the vesicle surface the existence of worm-like or tube structures [80]. The spaghetti-
charge, and it can be used to evaluate the extent of interaction of like structures, occurring at DNA: lipid concentrations which are
the liposomal surface cationic charges with the anionic charges of typically used during transfection (2 μg of DNA to 20 nmol DC-
DNA [71–73]. In general, zeta potential is a function of lipid to Chol liposomes) and their diameter comes closest to the diameter
 B. Ma et al. / Journal of Controlled Release 123 (2007) 184–194 189

endosomal membrane into the cytosol for transport to the nucleus.

But HCII structures formed from DDAB/DNA/DOPE lipoplexes,
have high transfection activity in vitro, and low in vivo either due
to serum instability or a high clearance rate [29].
Interestingly, novel ternary detergent/DNA/lipid lipoplex was
Fig. 5. Molecular structure of MVL5. found to be more efficient in the presence of serum, for their
highly dense and bent lamellar structure adopted by these
of the nuclear pores, may be the active cationic lipoplex. The lipoplexes which prevented their dissociation by serum proteins
reason for their high activity is likely attributed to that spaghetti, and allowed subsequent efficient internalization in the target cells
similar to microvilli, is extremely curved structure with very small [82]. Dioctadecylamidoglycyl-spermine(DOGS)-DNA lipoplex
radii (especially at the ending tip) and therefore is able to adhere also could maintain transfection efficiency in the presence of
and fuse to flat cells easily. Meanwhile, residual positive charges serum when the complexes adopted a lamellar arrangement [83].
on their surfaces enhance the interaction and fusion with cell and For LCα, DOPC/DOTAP-DNA lipoplexes shows a strong
probably nuclear membranes, thereby promoting the transfer of dependence on the molar fraction of neutral lipid DOPC
DNA into the cytoplasm and eventually into the nucleus of cells (ΦDOPC) and therefore membrane charge density σΜ. The
[27]. transfection efõficiency starts low for 0.5 b ΦDOPC b 0.7 and
The structure-efficiency relationships studied widely recent increases dramatically to a similar value, at ΦDOPC = 0.2, with
years mainly focus on LCα structure and HCII structure, and the HCII lipoplex achieved by the DOPE/DOTAP-DNA [37]. For LCα
conclusions are not consistent either. It is proposed that the structure of multivalent cationic lipid MVL5/DOPC/DNA
absence of a propensity for transition to the HCII phase may result lipoplex (Fig. 5), the transfection efficiency increases exponen-
in a lower transfection potential compared to the lipoplexes that tially with a linear increase of σΜ [51]. But others found that the
exhibit a higher-order inverted hexagonal structures [28,55,64]. curve of transfection efficiency versus σΜ assumed a bell-shape
The difference of transfection efficiency between LCα phase and with increasing σΜ using classes of MVL cationic lipids recently
HCII phase comes from the reason that HCII lipoplexes fuse and [84]. In LCα phase at high concentrations of cationic lipid, the
release DNA when in contact with anionic vesicles, which are enhanced transfection efficiency due to the formation of pores
cell-free models of cellular membranes, in particular, anionic opening in the membranes induced by the large electrostatic
endosomal vesicles, but LCα lipoplexes remain stable when in pressure through which the DNA molecules may escape the
contact with vesicle membranes, revealed by optical microscopy complex into cytoplasm was supported by a Meso-scale computer
[28]. On the other hand, HCII phase facilitates the interaction with modeling of cationic lipid lipoplexes [85]. In conclusion, most
the cell membrane and/or enables it to escape from the endosome reports approximately consider that membrane charge density σΜ
[47]. The ability of cationic lipids to promote inverted hexagonal is a universal parameter governing the transfection efficiency of
phase structures can at least in part lead to facilitate its fusion with LCα lipoplexs [19,51,52,84,86].
the endosomal membrane and disruption of the endosomal A recent study for lipoplex of dendritic lipid MVLBG2/
membrane following uptake of nucleic acid–cationic lipid DOPC/DNA, with a composition around 25 mol% MVLBG2
lipoplexes into cells, thus facilitating cytoplasmic release of the (Fig. 6), showed that lipoplex in the HI phase efficiently
plasmid or oligonucleotide into the cytosol [17,60,76,81]. So transfected mouse and human cells in culture. The high
Zuhorn et al. [17]concluded that inverted hexagonal phase efficiency may be ascribed to the existence of a continuous
formation in lipoplexes was not only a prerequisite for nucleic DNA substructure which likely facilitates release of the DNA
acid release from the complex, but also highly critical for cargo, as in principle all DNA is accessible once a part of it is
accomplishing efficient translocation of nucleic acids across the exposed to the inside of the cell [50].

Fig. 6. Molecular structure of MVLBG2.

190 B. Ma et al. / Journal of Controlled Release 123 (2007) 184–194

The behavior of lipoplexes containing univalent lipids

and multivalent lipids may be different, and the transfection
efficiency of the same structure can not maintain consistent in
Fig. 7. Molecular structure of RPR120535.
vivo and in vitro [28,29]. It is widely accepted that the main
entry route of lipoplex to mammalian cells is endocytosis, so
any structure facilitating membrane fusion, DNA release from sedimentation and better cellular trafficking [100]. However,
lipoplexes and escape from endosome could acquire high trans- highly effective lipoplex for gene transfer was obtained in NIH
fection efficiency, and HCII is a representative in these aspects. 3 T3 cells from RPR120535/DNA lipoplex (Fig. 7) character-
High transfection efficiency from LCα mainly benefits from ized by their small diameter (50 nm) [16]. A recent report also
their stability which can restrain dissociation of serum during supported that particles with smaller size would gain high
lipoplexes transporting. Therefore, it will be much important to transfection efficiency [101]. The requirement for efficient
find out more how formulations affect lipoplex structures so that transfection may be different in vivo and in vitro. Small particle
we can better predict the character of the lipoplex supramolec- size (40–80 nm) is required for high efficiency in vivo delivery
ular assemblies with high transfection efficiencies. owing to traversal of the capillary network (e.g. in the lungs),
while 200–400 nm is the optimal size for lipoplexes in vitro
4.2. Lipoplex size-transfection efficiency [78,99,102].
Although conflicting reports exist upon the optimal size of
Another parameter of morphologies affecting transfection lipoplexes for lipofection, there is no doubt that high lipofection
efficiency is lipoplex size, for the important role of lipoplex size in would be gained from large lipoplexes when endocytosis is
determining the nature of the entry pathway by endocytosis [76]. dominant, because large particles facilitate membrane contact
The cationic lipoplex stability depends on the charge ratio of lipid and fusion. When the type of cell is not actively endocytosing
to DNA [16,26,87]. Lipoplex size and the heterogeneity of the cell, either small particles may have high transfection efficiency,
structures both appear to increase along with the increasing of or lipoplex size doesn't correlate with lipofection efficiency.
lipid to DNA ratio [43]. Condensing ability of cationic lipids Besides lipoplex size, different conditions (such as cationic
(especially monovalent lipids) is another role affecting complex lipids, cell types and in vivo or vitro) may result in different
size [88]. However, size effect of lipoplexs on the transfection transfection efficiency.
activity has not been unified so far. Some consider that there is no
apparent correlation of the size of lipoplexes with transfection 5. Cytotoxicity
efficiency [89–91], others suggest that lipoplex size influence
[77,92] transfection efficiency, even lipoplex size is a major factor Nguyen et al. [103] clearly indicated that there was a synergism
[93,94] in terms of lipofection efficiency. between cationic lipid and pDNA in causing cytotoxicity. They
Lipoplex size is very important for gene transfer to actively found that cationic lipid (Lipofectamine 2000) alone induced only
endocytosing cells [93]. The influences on the transfection a slight cellular toxicity, irrespective of the absence or presence of
efficiency of lipoplexes by cationic lipid:DNA ratio, types of serum, and free pDNA did not show any cytotoxicity. However,
liposomes, incubation time in polyanion containing media, and lipoplexes induced a significant cytotoxicity toward HeLa cells,
time of serum addition, are channeled mostly through their B16BL6 cells and RGC-6 cells compared to cationic lipid alone,
influences on lipoplex size. For example, lipofection inhibition and the cytotoxicity increased as the cationic lipid content in the
by serum is largely due to the serum inhibition of lipoplex size lipoplex increased. Another study proposed that cationic
growth, and may be overcome by using large, stable lipoplexes liposomes formulated from DOPE and cationic lipids (such as
[95]. Lipoplexes of less than 250 nm in size measured by DOTAP, DMTAP, DPTAP, DSTAP), whether or not they were
dynamic light scattering show efficient transfection only in the complexed with DNA, were highly toxic in vitro toward
absence of serum. Conversely, lipoplexes of over 700 nm mean macrophages, but not toward non-phagocytic T cells. The
diameter induce efficient transfection in the presence or absence incorporation of DNA marginally reduced cationic liposome
of serum [87]. It was reported that the particle size may be one of toxicity toward macrophages [104]. Other factors affecting
the factors that were contributed to serum resistance of EDL cytotoxicity of lipoplexs include zeta potential, incubation time,
(ethanol-dried lipid-DNA) lipoplexes, and the large cationic cell type and cell density [105,106]. Toxicity of the lipoplex may
lipoplexes may delay the dissociation of DNAwith lipid, thereby depend upon the nature of the aggregates formed. For example,
enhancing DNA transfection efficiency [96]. Konopka et al. [97] the same lipoplex has been shown to exhibit a significantly
proposed that serum decreased the size of lipoplexes but was reduced toxicity when present in a vesicular as opposed to a
essentially not inhibitory to transfection activity. micellar solution [107]. Literatures about direct relationship
Large lipoplexes size is more efficient to transfer gene between toxicity and morphologies of lipoplex are very few at
because large particles taken up by cells lead to the formation of present, and it is worthy to study for scientists in the future.
large intracellular vesicles which are more easily disrupted, thus
releasing DNA into the cytoplasm [98]. In addition, the advan- 6. Discussion
tageous effect of large particles upon lipofection has been
attributed to maximum contact with cells [99], increased phago- Different stages in the pathway of transfection may be
cytic activity accompanied by endosomal escape [74] and faster promoted by different types of morphological behavior of the
 B. Ma et al. / Journal of Controlled Release 123 (2007) 184–194 191

lipoplexes, and different structures lead to a distinctly different Reference

mechanism of cell entry [53]. The high transfection efficiency of
lipoplexes containing DOPE may be ascribed to potential [1] I. Niculescu-Duvaz, R. Spooner, R. Marais, C.J. Springer, Gene-directed
enzyme prodrug therapy, Bioconjug. Chem. 9 (1998) 4–22.
transformation to an invert hexagonal structures HCII promoted
[2] K.K. Hunt, S.A. Vorburger, Hurdles and hopes for cancer treatment,
by DOPE. A transition from the LCα structure to the HCII structure Science 297 (2002) 415–416.
and the subsequent increasing of transfection efficiency could be [3] M.J. Bennett, A.M. Aberle, R.P. Balasubramaniam, J.G. Malone, R.W.
expected by increasing weight fraction of DOPE. This, however Malone, M.H. Nantz, Cationic lipid-mediated gene delivery to murine
does not mean that all lipoplexes containing DOPE do have high lung: correlation of lipid hydration with in vivo transfection activity,
J. Med. Chem. 40 (1997) 4069–4078.
transfection efficiency, because we are not sure that HCII have
[4] W.F. Anderson, Human gene therapy, Nature 392 (1998) 25–26.
high transfection efficiency in every gene delivery. High [5] D. Ferber, Gene therapy: safer and virus-free, Science 294 (2001)
lipofection will be acquired for HCII structure when membrane 1638–1642.
fusion is dominant, but when serum is present LCα lipoplexes [6] K. Mukherjee, J. Sen, A. Chaudhuri, Common co-lipids, in synergy,
show great superiority for their inhibition dissociation by serum impart high gene transfer properties to transfection-incompetent cationic
lipids, FEBS Lett. 579 (2005) 1291–1300.
during lipoplexes transporting. High lipofection would be gained
[7] I.M. Verma, M.D. Weitzman, Gene therapy: twenty-first century
from large lipoplexes when endocytosis is dominant, because medicine, Annu. Rev. Biochem. 74 (2005) 711–738.
large particles facilitate membrane contact and fusion. When the [8] J. Zabner, Cationic lipids used in gene transfer, Adv. Drug. Deliv. Rev. 27
type of cell is not actively endocytosing cell, either small particles (1997) 17–28.
may have high transfection efficiency, or lipoplex size doesn't [9] M.A. Ilies, W.A. Seitz, A.T. Balaban, Cationic lipids in gene delivery:
principles, vector design and therapeutical applications, Curr. Pharm.
correlate with lipofection.
Des. 8 (2002) 2441–2473.
It is difficult to say which structure or lipoplex size is better for [10] P.L. Felgner, T.R. Gadek, M. Holm, R. Roman, H.W. Chan, M. Wenz, J.P.
gene delivery, because transfection efficiency of the cationic Northrop, G.M. Ringgold, M. Danielsen, Lipofection: a highly efficient,
lipoplexes may depend on not only morphologies of the lipid-mediated DNA-transfection procedure, Proc. Natl. Acad. Sci. U. S. A.
complexes, but also other factors, such as cell types. Different 84 (1987) 7413–7417.
[11] P.L. Felgner, G.M. Ringold, Cationic liposome-mediated transfection,
types of cells may possess different ability to the uptake and
Nature. 337 (1989) 387–388.
expression of DNA [108]. Therefore, a coherent and comprehen- [12] X. Gao, L. Huang, A novel cationic liposome reagent for efficient
sive conclusion of the morphology-efficiency relationship of transfection of mammalian cells, Biochem. Biophys. Res. Commun. 179
cationic lipoplexes is often difficult, due to the fact that it is (1991) 280–285.
difficult for scientists to perform a comparison of different classes [13] B. Sternberg, K. Hong, W. Zheng, D. Papahadjopoulos, Ultrastructural
characterization of cationic liposome-DNA complexes showing enhanced
of cationic lipids and different cell types within one study.
stability in serum and high transfection activity in vivo, Biochim.
The morphologies of lipoplex should be physicochemically Biophys. Acta. 1375 (1998) 23–35.
characterized at two different levels, one level, which relates to [14] W.T. Godbey, K.K. Wu, G.J. Hirasaki, A.G. Mikos, Improved packing of
size and size instability, and the other level, which relates to the poly (ethylenimine)/DNA complexes increases transfection efficiency,
self-assembled structures including all parameters influencing Gene. Ther. 6 (1999) 1380–1388.
[15] C. Dai, B. Wang, H. Zhao, B. Li, Factors affecting protein release
lipoplex structures and optimized structure with high lipofec-
from microcapsule prepared by liposome in alginate, Colloids. Surf., B
tion. Clearly, more work will be required before we have a Biointerfaces. 42 (2005) 253–258.
perfect understanding of the lipoplex morphology at macro [16] B. Pitard, O. Aguerre, M. Airiau, A.M. Lachagès, T. Boukhnikachvili, G.
level and an understanding of all the critical parameters that Byk, C. Dubertret, C. Herviou, D. Scherman, J.F. Mayaux, J. Crouzet,
control and affect the lipoplex structures at micro levels. In the Virus-sized self-assembling lamellar complexes between plasmid DNA
and cationic micelles promote gene transfer, Proc. Natl. Acad. Sci. U. S. A.
future, detailed data generated from these studies in the process
94 (1997) 14412–14417.
of gene delivery will help scientists to design new generations [17] I.S. Zuhorn, U. Bakowsky, E. Polushkin, W.H. Visser, M.C. Stuart, J.B.
of vectors with low toxicity and relatively high transfection Engberts, D. Hoekstra, Nonbilayer phase of lipoplex — membrane
efficiency. Influences of lipoplex morphologies on transfection mixture determines endosomal escape of genetic cargo and transfection
efficiency are greatly depended on internalisation mechanisms efficiency, Mol. Ther. 11 (2005) 801–810.
[18] D. Niculescu-Duvaz, J. Heyes, C.J. Springer, Structure-activity relation-
and intracellular trafficking. Some studies have expatiated
ship in cationic lipid mediated gene transfection, Curr. Med. Chem. 10
lipoplex internalisation mechanisms and intracellular trafficking (2003) 1233–1261.
[109,110], but they are currently not well documented. How- [19] K. Ewert, A. Ahmad, H.M. Evans, C.R. Safinya, Cationic lipid-DNA
ever, because the multiple barriers to entry of the lipoplexes into complexes for non-viral gene therapy: relating supramolecular structures
the target cell are becoming clear at the molecular level, the role to cellular pathways, Expert. Opin. Biol. Ther. 5 (2005) 33–53.
[20] J.H. Felgner, R. Kumar, C.N. Sridhar, C.J. Wheeler, Y.J. Tsai, R. Border,
of morphologies of lipoplex is becoming more and more im-
P. Ramsey, M. Martin, P.L. Felgner, Enhanced gene delivery and
portant for scientists in vector design, formulation and molec- mechanism studies with a novel series of cationic lipid formulations,
ular assembly of lipoplexes. J. Biol. Chem. 269 (1994) 2550–2561.
[21] J. Wang, X. Guo, Y. Xu, L. Barron, F.C. Szoka Jr., Synthesis and
Acknowledgement characterization of long chain alkyl acyl carnitine esters. Potentially
biodegradable cationic lipids for use in gene delivery, J. Med. Chem. 4
(1998) 2207–2215.
The authors of this paper gratefully thank the financial [22] Z. Hyvonen, A. Plotniece, I. Reine, B. Chekavichus, G. Duburs, A. Urtti,
support from the Educational Department of Liaoning Province Novel cationic amphiphilic1,4-dihydropyridine derivatives for DNA
(20040084) and Dalian Nationalities University (20036214). delivery, Biochim. Biophys. Acta. 1509 (2000) 451–466.
192 B. Ma et al. / Journal of Controlled Release 123 (2007) 184–194

[23] H. Gao, K.M. Hui, Synthesis of a novel series of cationic lipids [43] S.J. Eastman, C. Siegel, J. Tousignant, A.E. Smith, S.H. Cheng, R.K.
that can act as efficient gene delivery vehicles through systematic Scheule, Biophysical characterization of cationic lipid: DNA complexes,
heterocyclic substitution of cholesterol derivatives, Gene. Ther. 8 (2001) Biochim. Biophys. Acta 1325 (1997) 41–62.
855–863. [44] J. Gustafsson, G. Arvidson, G. Karlsson, M. Almgren, Complexes
[24] J. Gaucheron, T. Wong, K.F. Wong, N. Maurer, P.R. Cullis, Synthesis and between cationic liposomes and DNA visualized by cryo-TEM, Biochim.
properties of novel tetraalkyl cationic lipids, Bioconjug. Chem. 13 (2002) Biophys. Acta 1235 (1995) 305–312.
671–675. [45] V. Oberle, U. Bakowsky, I.S. Zuhorn, D. Hoekstra, Lipoplex formation
[25] J. Pelisek, L. Gaedtke, J. DeRouchey, G.F. Walker, S. Nikol, E. Wagner, under equilibrium conditions reveal a three-step mechanism, Biophys. J.
Optimized lipopolyplex formulations for gene transfer to human colon 79 (2000) 1447–1454.
carcinoma cells underin vitro conditions, J. Gene. Med. 8 (2006) 186–197. [46] I.S. Zuhorn, V. Oberle, W.H. Visser, J.B. Engberts, U. Bakowsky, E.
[26] F. Boffi, A. Bonincontro, F. Bordi, E. Bultrini, C. Cametti, A. Congiu- Polushkin, D. Hoekstra, Phase behavior of cationic amphiphiles and their
Castellano, F. De Luca, G. Risuleo, Two-step mechanism in cationic mixtures with helper lipid Influences lipoplex shape, DNA translocation,
lipoplex formation as observed by dynamic light scattering, dielectric and transfection efficiency, Biophys. J. 83 (2002) 2096–2180.
relaxation and circular dichroism methods, Phys. Chem. Chem. Phys. 4 [47] R. Hulst, I. Muizebelt, P. Oosting, C. Pol, A. Wagenaar, J. Šmisterová, E.
(2002) 2708–2713. Bulten, C. Driessen, D. Hoekstra, J.B. Engberts, Sunfish amphiphiles:
[27] B. Sternberg, F.L. Sorgib, L. Huang, New structures in complex conceptually new carriers for DNA delivery, Eur. J. Org. Chem. 4 (2004)
formation between DNA and cationic liposomes visualized by freeze- 835–849.
fracture electron microscopy, FEBS. Lett. 356 (1994) 361–366. [48] J. Zabner, A.J. Fasbender, T. Moninger, K.A. Poellinger, M.J. Welsh,
[28] I. Koltover, T. Salditt, J.O. Radler, C.R. Safinya, An inverted hexagonal Cellular and molecular barriers to gene transfer by a cationic lipid, J. Biol.
phase of cationic liposome-DNA complexes related to DNA release and Chem. 270 (1995) 18997–19007.
delivery, Science. 281 (1998) 78–81. [49] R. Ghirlando, E.J. Wachtel, T. Arad, A. Minsky, DNA packaging induced
[29] B. Sternberg, K. Hong, W. Zheng, D. Papahadjopoulos, Ultrastructural by micellar aggregates: a novel in vitro DNA condensation system,
characterization of cationic liposome-DNA complexes showing enhanced Biochemistry 31 (1992) 7110–7119.
stability in serum and high transfection activity in vivo, Biochim. [50] K.K. Ewert, H.M. Evans, A. Zidovska, N.F. Bouxsein, A. Ahmad, C.R.
Biophys. Acta 1375 (1998) 23–35. Safinya, A columnar phase of dendritic lipid-based cationic liposome-
[30] C.S. O'Hern, T.C. Lubensky, Sliding columnar phase of DNA-lipid DNA complexes for gene delivery: hexagonally ordered cylindrical
complexes, Phys. Rev. Lett. 80 (1998) 4345–4348. micelles embedded in a DNA honeycomb lattice, J. Am. Chem. Soc. 128
[31] K. Remaut, N.N. Sanders, F. Fayazpour, J. Demeester, S.C. De Smedt, (2006) 3998–4006.
Influence of plasmid DNA topology on the transfection properties of [51] K. Ewert, A. Ahmad, H.M. Evans, H.W. Schmidt, C.R. Safinya, Efficient
dotap/dope lipoplexes, J. Control. Release 115 (2006) 335–343. synthesis and cell-transfection properties of a new multivalent cationic
[32] M.A. Ilies, W.A. Seitz, B.H. Johnson, E.L. Ezell, A.L. Miller, E.B. lipid for nonviral gene delivery, J. Med. Chem. 45 (2002) 5023–5029.
Thompson, A.T. Balaban, Lipophilic pyrylium salts in the synthesis of [52] K. Ewert, H.M. Evans, A. Ahmad, N.L. Slack, A.J. Lin, A. Martin-
efficient pyridinium-based cationic lipids, gemini surfactants, and Herranz, C.R. Safinya, Lipoplex structures and their distinct cellular
lipophilic oligomers for gene delivery, J. Med. Chem. 49 (2006) pathways, Adv. Genet. 53 (2005) 119–155.
3872–3887. [53] S. May, A. Ben-Shaul, Modeling of cationic lipid-DNA complexes, Curr.
[33] P.L. Felgner, Y. Barenholz, J.P. Behr, S.H. Cheng, P. Cullis, L. Huang, Med. Chem. 11 (2004) 151–167.
J.A. Jessee, L. Seymour, F. Szoka, A.R. Thierry, E. Wagner, G. Wu, [54] V. Pector, J. Backmann, D. Maes, M. Vandenbranden, J.M. Ruysschaert,
Nomenclature for synthetic gene delivery systems, Hum. Gene. Ther. Biophysical and structural properties of DNA diC(14)-amidine com-
8 (1997) 511–512. plexes. Influence of the DNA/lipid ratio, J. Biol. Chem. 275 (2000)
[34] P. Kreiss, B. Cameron, R. Rangara, P. Mailhe, O. Aguerre-Charriol, M. 29533–29538.
Airiau, D. Scherman, J. Crouzet, B. Pitard, Plasmid DNA size dose not [55] M. Scarzello, J. Smisterova, A. Wagenaar, M.C. Stuart, D. Hoekstra, J.B.
affect the physicochemical properties of lipoplexes but modulates gene Engberts, R. Hulst, Sunfish cationic amphiphiles: toward an adaptative
transfer efficiency, Nucleic. Acids. Res. 27 (1999) 1792–3798. lipoplex morphology, J. Am. Chem. Soc. 127 (2005) 10420–10429.
[35] E. Pozharski, R.C. MacDonald, Thermodynamics of cationic lipid-DNA [56] C.R. Safinya, Structures of lipid-DNA complexes: supramolecular
complex formation as studied by isothermal titration calorimetry, assembly and gene delivery, Curr. Opin. Struct. Biol. 11 (2001) 440–448.
Biophys. J. 83 (2002) 556–565. [57] H. Edlund, A. Sadaghiani, A. Khan, Phase behavior and phase structure
[36] E. Pozharski, R.C. MacDonald, Lipoplex thermodynamics: determination for catanionic surfactant mixtures: dodecyltrimethylammonium chloride-
of DNA-cationic lipoid interaction energies, Biophys. J. 85 (2003) sodium nonanoate-water system, Langmuir 13 (1997) 4953–4963.
3969–3978. [58] R. Nagarajan, Molecular packing parameter and surfactant self-assembly:
[37] K. Ewert, N.L. Slack, A. Ahmad, H.M. Evans, A.J. Lin, C.E. Samuel, C.R. the neglected role of the surfactant tail, Langmuir 18 (2002) 31–38.
Safinya, Cationic lipid-DNA complexes for gene therapy: understanding [59] R. Koynova, L. Wang, R.C. MacDonald, An intracellular lamellar-
the relationship between complex structure and gene delivery pathways nonlamellar phase transition rationalizes the superior performance of
at the molecular level, Curr. Med. Chem. 11 (2004) 133–149. some cationic lipid transfection agents, Proc. Natl. Acad. Sci. U. S. A.
[38] H. Gershon, R. Ghirlando, S.B. Guttman, A. Minsky, Mode of formation 103 (2006) 14373–14378.
and structural features of DNA-cationic liposome complexes used for [60] I.M. Hafez, N. Maurer, P.R. Cullis, On the mechanism whereby cationic
transfection, Biochemistry. 32 (1993) 7143–7151. lipids promote intracellular delivery of polynucleic acids, Gene. Ther.
[39] M. Scarzello, V. Chupin, A. Wagenaar, M.C. Stuart, J.B. Engberts, R. 8 (2001) 1188–1196.
Hulst, Polymorphism of pyridinium amphiphiles for gene delivery: [61] L. Wasungu, M.C. Stuart, M. Scarzello, J.B. Engberts, D. Hoekstra,
influence ofIonic strength, helper lipid content, and plasmid DNA Lipoplexes formed from sugar-based gemini surfactants undergo a
complexation, Biophys. J. 88 (2005) 2104–2113. lamellar-to-micellar phase transition at acidic pH. Evidence for a non-
[40] D. Simberg, D. Danino, Y. Talmon, A. Minsky, M.E. Ferrari, C.J. Wheeler, inverted membrane-destabilizing hexagonal phase of lipoplexes, Bio-
Y. Barenholz, Phase behavior, DNA ordering, and size instability of chim. Biophys. Acta 1758 (2006) 1677–1684.
cationic lipoplexes, J. Biol. Chem. 276 (2001) 47453–47459. [62] L. Wasungu, M. Scarzello, G. van Dam, G. Molema, A. Wagenaar, J.B.
[41] V.A. Bloomfield, DNA condensation by multivalent cations, Biopoly- Engberts, D. Hoekstra, Transfection mediated by pH-sensitive sugar-
mers 44 (1997) 269–282. based gemini surfactants; potential for in vivo gene therapy applications,
[42] I. Koltover, T. Salditt, C.R. Safinya, Phase diagram, stability, and J. Mol. Med. 84 (2006) 774–784.
overcharging of lamellar cationic lipid-DNA self-assembled complexes, [63] J. Šmisterová, A. Wagenaar, M.C. Stuart, E. Polushkin, G. ten Brink, R.
Biophys. J. 77 (1999) 915–924. Hulst, J.B. Engberts, D. Hoekstra, Molecular shape of the cationic lipid
 B. Ma et al. / Journal of Controlled Release 123 (2007) 184–194 193

controls the structure of cationic lipid/dioleylphosphatidylethanolamine- [85] O. Farago, N. Gronbech-Jensen, P. Pincus, Mesoscale computer modeling
DNA complexes and the efficiency of gene delivery, J. Biol. Chem. 276 of lipid-DNA complexes for gene therapy, Phys. Rev. Lett. 96 (2006)
(2001) 47615–47622. 018102.
[64] J. Šmisterová, A. Wagenaar, M.C. Stuart, E. Polushkin, G. ten Brinke, R. [86] A.J. Lin, N.L. Slack, A. Ahmad, C.X. George, C.E. Samuel, C.R.
Hulst, J.B. Engberts, D. Hoekstra, Molecular shape of the cationic lipid Safinya, Three-dimensional imaging of lipid gene-carriers: membrane
controls the structure of cationic lipid/dioleylphosphatidylethanolamine- charge density controls universal transfection behavior in lamellar
DNA complexes and the efficiency of gene delivery, J. Biol. Chem. 276 cationic liposome-DNA complexes, Biophys. J. 84 (2003) 3307–3316.
(2001) 47615–47622. [87] J. Turek, C. Dubertret, G. Jaslin, K. Antonakis, D. Scherman, B. Pitard,
[65] C.M. Wiethoff, M.L. Gill, G.S. Koe, J.G. Koe, C.R. Middaugh, The Formulations which increase the size of lipoplexes prevent serum-
structural organization of cationic lipid-DNA complexes, J. Biol. Chem. associated inhibition of transfection, J. Gene. Med. 2 (2000) 32–40.
277 (2002) 44980–44987. [88] F. Liu, L. Huang, Development of non-viral vectors for systemic gene
[66] J.O. Rädler, I. Koltover, T. Salditt, C.R. Safinya, Structure of DNA- delivery, J. Control. Release 78 (2002) 259–266.
cationic liposome complexes: dna intercalation in multilamellar mem- [89] T. Stegmann, J.Y. Legendre, Gene transfer mediated by cationic lipids:
branes in distinct interhelical packing regimes, Science 275 (1997) lack of a correlation between lipid mixing and transfection, Biochim.
810–814. Biophys. Acta 1325 (1997) 71–79.
[67] E. Gonçalves, R.J. Debs, T.D. Heath, The effect of liposome size on the final [90] Y.P. Zhang, D.L. Reimer, G. Zhang, P.H. Lee, M.B. Bally, Self-assembling
lipid/DNA ratio of cationic lipoplexes, Biophys. J. 86 (2004) 1554–1563. DNA-lipid particles for gene transfer, Pharm. Res. 14 (1997) 190–196.
[68] R. Krishnaswamy, V.A. Raghunathan, A.K. Sood, Reentrant phase [91] Z. Hassani, G.F. Lemkine, P. Erbacher, K. Palmier, G. Alfama, C.
transitions of DNA-surfactant complexes, Phys. Rev., E 69 (2004) Giovannangeli, J.P. Behr, B.A. Demeneix, Lipid-mediated siRNA
031905. delivery down-regulates exogenous gene expression in the mouse brain
[69] R. Krishnaswamy, G. Pabst, M. Rappolt, V.A. Raghunathan, A.K. Sood, at picomolar levels, J. Gene. Med. 7 (2005) 198–207.
Structure of DNA-CATB-hexanol complexes, Phys. Rev., E 73 (2006) [92] J. Rejman, V. Oberle, I.S. Zuhorn, D. Hoekstra, Size-dependent
031904. internalization of particles via the pathways of clathrin-and caveolae-
[70] K.K. Ewert, H.M. Evans, N.F. Bouxsein, C.R. Safinya, Dendritic cationic mediated endocytosis, Biochem. J. 337 (2004) 159–169.
lipids with highly charged headgroups for efficient gene delivery, [93] P.C. Ross, S.W. Hui, Lipoplex size is a major determinant of in vitro
Bioconjug. Chem. 17 (2006) 877–888. lipofection efficiency, Gene. Ther. 6 (1999) 651–659.
[71] Y. Perrie, G. Gregoriadis, Liposome-entrapped plasmid DNA: character- [94] M.R. Almofti, H. Harashima, Y. Shinohara, A. Almofti, W. Li, H.
isation studies, Biochim. Biophys. Acta 1475 (2000) 125–132. Kiwada, Lipoplex size determines lipofection efficiency with or without
[72] S.J. Eastman, C. Siegel, J. Tousignant, A.E. Smith, S.H. Cheng, R.K. serum, Mol. Membr. Biol. 20 (2003) 35–43.
Scheule, Biophysical characterization of cationic lipid:DNA complexes, [95] M. Carriere, I. Tranchant, P.A. Niore, G. Byk, N. Mignet, V. Escriou, D.
Biochim. Biophys. Acta 1325 (1997) 41–62. Scherman, J. Herscovici, Optimization of cationic lipid mediated
[73] Y. Perrie, P.M. Frederik, G. Gregoriadis, Liposome-mediated DNA gene transfer: structure-function, physico-chemical, and cellular studies,
vaccination: the effect of vesicle, Vaccine 19 (2001) 3301–3310. J. Liposome. Res. 12 (2002) 95–106.
[74] Y. Xu, S.W. Hui, P. Frederik, F.C. Szoka Jr., Physicochemical [96] T. Lian, R.J. Ho, Design and characterization of a novel lipid-DNA
characterization and purification of cationic lipoplexes, Biophys. J. 77 complex that resists serum-induced destabilization, J. Pharm. Sci. 92
(1999) 341–353. (2003) 2373–2385.
[75] Y. Capan, B.H. Woo, S. Gebrekidan, S. Ahmed, P.P. Deluca, Influence of [97] K. Konopka, N. Overlid, A.C. Nagaraj, N. Duzgunes, Serum decreases
formulation parameters on the characteristics of poly(D,L-lactide-co- the size of metafectene-and genejammer-DNA complexes but does not
glycolide) microspheres containing poly(L-lysine) complexed plasmid affect significantly their transfection activity in SCCVII murine
DNA, J. Control. Release 60 (1999) 279–286. squamous cell carcinoma cells, Cell. Mol. Biol. Lett. 11 (2006) 171–190.
[76] L. Wasungu, D. Hoekstra, Cationic lipids, lipoplexes and intracellular [98] V. Escriou, C. Ciolina, F. Lacroix, G. Byk, D. Scherman, P. Wils, Cationic
delivery of genes, J. Control. Release 116 (2006) 255–264. lipid-mediated gene transfer: effect of serum on cellular uptake and
[77] V.A. Rakhmanova, E.V. Pozharski, R.C. MacDonald, Mechanisms of intracellular fate of lipopolyamine/DNA complexes, Biochim. Biophys.
lipoplex formation: dependence of the biological properties of transfec- Acta 1368 (1998) 276–288.
tion complexes on formulation procedures, J. Membr. Biol. 200 (2004) [99] M.T. Kennedy, E.V. Pozharski, V.A. Rakhmanova, R.C. MacDonald,
35–45. Factors governing the assembly of cationic phospholipid-DNA com-
[78] R.I. Zhdanov, O.V. Podobed, V.V. Vlassov, Cationic lipid-DNA plexes, Biophys. J. 78 (2000) 1620–1633.
complexes — lipoplexes for gene transfer and therapy, Bioelectrochemistry [100] L.K. Lee, E.K. Siapati, R.G. Jenkins, R.J. McAnulty, S.L. Hart, P.A.
58 (2002) 53–64. Shamlou, Biophysical characterization of an integrin-targeted non-viral
[79] S. Chesnoy, L. Huang, Structure and function of lipid-DNA complexes vector, Med. Sci. Monit. 9 (2003) 54–61.
for gene delivery, Annu. Rev. Biophys. Biomol. Struct. 29 (2000) 27–47. [101] C. Kneuer, C. Ehrhardt, H. Bakowsky, M.N. Kumar, V. Oberle, C.M.
[80] A. Hohner, J. Bayer, J.O. Radler, Wormlike lipid/DNA micelles in a non- Lehr, D. Hoekstra, U. Bakowsky, The influence of physicochemical
polar solvent, Eur. Phys. J., E Soft Matter 21 (2006) 41–48. parameters on the efficacy of non-viral DNA transfection complexes:
[81] P.C. Bell, M. Bergsma, I.P. Dolbnya, W. Bras, M.C. Stuart, A.E. Rowan, a comparative study, J. Nanosci Nanotechnol. 6 (2006) 2776–2782.
M.C. Feiters, J.B. Engberts, Transfection mediated by gemini surfactants: [102] R.I. Zhdanov, O.V. Podobed, V.V. Vlassov, Cationic lipid-DNA
engineered escape from the endosomal compartment, J. Am. Chem. Soc. complexes-lipoplexes-for gene transfer and therapy, Bioelectrochemistry
125 (2003) 1551–1558. 58 (2002) 53–64.
[82] D. Llères, J.M. Weibel, D. Heissler, G. Zuber, G. Duportail, Y. Mély, [103] L.T. Nguyen, K. Atobe, J.M. Barichello, T. Ishida, H. Kiwada, Complex
Dependence of the cellular internalization and transfection efficiency on formation with plasmid dna increases the cytotoxicity of cationic
the structure and physicochemical properties of cationic detergent/DNA/ liposomes, Biol. Pharm. Bull. 30 (2007) 751–757.
liposomes, J. Gene. Med. 6 (2004) 415–428. [104] M.C. Filion, N.C. Phillips, Toxicity and immunomodulatory activity of
[83] T. Boukhnikachvili, O. Aguerre-Chariol, M. Airiau, S. Lesieur, M. liposomal vectors formulated with cationic lipids toward immune effector
Ollivon, J. Vacus, Structure of in-serum transfecting DNA-cationic lipid cells, Biochim. Biophys. Acta 1329 (1997) 345–356.
complexes, FEBS. Lett. 409 (1997) 188–194. [105] P. Pinnaduwage, L. Schmitt, L. Huang, Use of a quaternary ammonium
[84] A. Ahmad, H.M. Evans, K. Ewert, C.X. George, C.E. Samuel, C.R. detergent in liposome mediated DNA transfection of mouse L-cells,
Safinya1, New multivalent cationic lipids reveal bell curve for Biochim. Biophys. Acta 985 (1989) 33–37.
transfection efficiency versus membrane charge density: lipid-DNA [106] C.R. Dass, Lipoplex-mediated delivery of nucleic acids: factors affecting
complexes for gene delivery, J. Gene. Med. 7 (2005) 739–748. in vivo transfection, J. Mol. Med. 82 (2004) 579–591.
194 B. Ma et al. / Journal of Controlled Release 123 (2007) 184–194

[107] M.J. Lawrence, Surfactant systems: their use in drug delivery, Chem. Soc. [109] M.B. Bally, P. Harvie, F.M. Wong, S. Kong, E.K. Wasan, D.L. Reimer,
Rev. 23 (1994) 417–423. Biological barriers to cellular delivery of lipid-based DNA carriers, Adv.
[108] S.W. Kim, T. Ogawa, Y. Tabata, I. Nishimura, Efficacy and cytotoxicity of Drug Deliv. Rev. 38 (1999) 291–315.
cationic-agent-mediated nonviral gene transfer into osteoblasts, J. Biomed. [110] R. Zhou, R.C. Geiger, D.A. Dean, Intracellular trafficking of nucleic
Mater. Res., A. 71 (2004) 308–315. acids, Expert Opin. Drug Deliv. 1 (2004) 127–140.