Copyright © 2011 American Scientific Publishers
All rights reserved
Printed in the United States of America
Journal of
Nanoscience and Nanotechnology
Vol. 11, 1–5, 2011
Budding Dynamics of the Lipid Membrane
Sun Ju Bae1 , Sunghwan Jung2 , and Soong Ho Um1 ∗
1
School of Materials Science and Engineering, Gwangju Institute of Science and Technology,
Gwangju, 500-712, Republic of Korea
2
Department of Engineering Science and Mechanics, Virginia Polytechnic Institute and State University,
Blacksburg, VA 24061, USA
Liposomes are widely used in industrial engineering systems including cosmetics and pharmaceutical and biotechnological applications. A fundamental understanding of the dynamics of lipid bud
formation is required to efficiently produce various high-technological liposomes with well-controlled
sizes and shapes. In this study, a novel patterning method of controlling the uniformity of the sizes
and shapes of liposome buds nucleated from a flat lipid film casted on a flat substrate is reported.
The dried-out lipid components had been swelling during aqueous hydration, and they pinched away
in approximately one hour. The sizes and spacing of the liposome buds were monitored during
hydration. It was observed that the average size of the buds slowly increased, but their spacing did
not significantly change. Moreover, the bud size distribution was a very narrow Gaussian, which
implies the formation of buds with uniform sizes. The analytical calculation of the equilibrium state
was theoretically developed and compared with the experiments. It is envisioned that this study will
provide insights on a sustained-release drug vehicle.
Keywords: Liposome, Spin-Coating, Budding.
As mimics of the dynamic cellular membranes in a living organism, lipid molecules are primarily self-assembled
into spherical bilayers owing to the hydrophobic effect.1
Even with a low critical micelle concentration (e.g.,
0.46 nM in the case of 16:0 phosphocholine), such
hydrophobic interaction is sufficient to form a stable
bilayer structure, which has been studied in various fields,
such as mathematics, physics, chemistry, biology, pharmaceutics, and medicine.2 3
Owing to the intrinsic biocompatibility and simple
manipulability of a vesicle, a liposome have been used
as common drug carriers in the pharmaceutical industry.4
Several liposome species have been utilized as novel stimuli drug delivery vehicles because they respond to various stimuli, such as the composition, surface charges,5 and
sizes of liposomes,6 and the pH7 and temperature8 variation in the environment.9 In particular, the uniform sizes
of liposomes is one of the dominant factors contributing
to systemic administration through an intravascular drug
delivery pathway.10 11 It is well known that the liposomal
size plays a significant role in in-vivo biodistribution.12 13
Liposomal size control is currently generally achieved
∗
Author to whom correspondence should be addressed.
J. Nanosci. Nanotechnol. 2011, Vol. 11, No. xx
using either of two popular techniques: sonication14 or
an extrusion.15 Sonication changes the sizes of liposomes
through the strong agitation of sound waves.16 Hence,
nucleated liposomes are smaller than those prepared simply by shaking. Two decades ago, Papahadjopoulos et al.
developed an extrusion method where liposomes are permeated through a polycarbonate membrane with a regular
pore size.17 Even though this method is very reliable in
controlling the liposomal size, it is generally considered
inconvenient and time-consuming in the entire process and
few studies have been undertaken to allow it to be effectively controlled as designed.
Estes and Mayer et al. have presented a spin-casting
method that is capable of producing a homogeneous lipid
film on a two-dimensional substrate.18 They optimized
the lipid concentration and the spinning speed to be able
to control the thickness and homogeneity of such lipid
film, after which they obtained the optimal thickness of
the homogeneous lipid film to generate giant liposomes
via electroformation. The nucleated giant vesicles made
through this method are two to five times larger than
those made through the conventional droplet-derived-film
technique. Another method involves the use of chemical additives. Kanta Tsumoto and Hideki Matsuo et al.
reported that the sugar moieties involved in lipid films
may enhance the repulsion between the lipid lamellae,
1533-4880/2011/11/001/005
doi:10.1166/jnn.2011.4388
1
RESEARCH ARTICLE
1. INTRODUCTION
Budding Dynamics of the Lipid Membrane
RESEARCH ARTICLE
and they confirmed through the study that this may help
obtain more and larger giant liposomes with a narrow size
distribution.19 Alternatively, there have been some efforts
to systematically observe and understand the lipid budding
process for size control. The budding process of lipid vesicles has been observed with varying lipid compositions
to come up with fluid- and gel-like domains.20 The bud
growth results from the fluid-like domains of the lipid vesicles above the phase transition temperature of the fluid-like
domain, and a morphological change was observed over
time.21 22 The modeling study of the budding process has
also been investigated.23
In this study, the spatial distribution and time evolution
of lipid vesicles budding from a flat film were investigated.
The combined experimental and theoretical results on the
formation of lipid buds are presented. In the experiments,
lipids were deposited on a flat surface and were monitored during hydration at a fixed temperature. A theoretical model was developed by considering various energies
related to the elastic deformation of lipids. In Section 2,
the experimental methods that were used in this study to
characterize the dynamics of the lipid budding process are
explained. The details of the developed theoretical model
are presented in Section 3. In Section 4, the results of
the use of the theoretical model are compared with the
experimental results. Discussion is featured, and the future
research directions are presented, in Section 5.
2. EXPERIMENTAL DETAILS
An experimental method of studying the budding process
from the lipid film is presented herein. The strategy for
preparing the lipid film for the budding-off test is outlined.
A square glass substrate (12 × 12 mm) is first cleaned with
a piranha solution (H2 SO4 :H2 O2 =7:3). A dry lipid component, DOPC (1,2-dioleoyl-sn-glycero-3-phosphocholine,
Avanti Polar Lipids), is completely reconstituted into a
mixing co-solvent of chloroform and acetonitrile at a 95:5
ratio. (It is reported that the most homogeneous lipid film
formed via spin-coating is successfully achieved under the
mixing ratio of 95% chloroform and 5% acetonitrile.18 The stock concentration is in 2.5 mg/ml. The 50 l lipid
precursor solution is slowly poured at the center of the
glass substrate until the glass substrate is entirely covered
with the solution. The glass substrate is then moved to the
chamber of a spin-caster and is immediately rotated at the
speed of 600 rpm for 60 sec after which the substrate is
coated with a homogenous lipid film (DOPC film) that has
been dried under high vacuum for over 18 h to remove all
traces of mixing organic co-solvents therefrom.
To observe the budding effects in uniform sizes, the
dry DOPC film is hydrated by adding 100 l deionized
(DI) water prewarmed at room temperature. Images of the
in-situ budding process are taken, using a phase contrast
microscope (TE2000-U, Nikon), every min until 60 min
2
Bae et al.
after hydration. The images of lipid budding in a dry film
and 20, 30, and 40 min after hydration captured and shown
in Figure 2(A) are further analyzed using the Image J software (NIH) to determine the sizes (diameters) of the individual liposomes and the distances between them. In the
microscope images, the liposomes show bright white dots
against the background. The image intensity of the liposomes was measured using the plot profile of the Image J
software. The boundaries of the liposome and of the background were estimated, and the number of pixels of the
liposomes were converted to micrometer using the scale
bars of the images.
3. THEORETICAL MODEL
The formation of micron-sized vesicles is a complicated
non-equilibrium process involving elastic deformation and
energetic dissipation of the lipid membrane and fluid bulk.
When a solvent keeps in touch with a thin layer of membrane attached to a flat substrate, this membrane gets
swollen and becomes unstable in space and time. Many
small buds are then nucleated from the flat surface and
are eventually detached from the remaining membrane. If
this process occurs fast, dissipations will become predominant on both the membrane and the bulk. If the process is
slow, a system satisfies the quasi-equilibrium state at any
instance of time. The experiment that was conducted in
this work was performed on time scales of an hour. The
budding process was so slow that the equilibrium state is
an acceptable assumption for the theoretical model developed herein.
The total energy due to the area increase during the formation of a vesicle mainly consists of viscous dissipations,
elastic deformation and stretching. For the slow budding
process, the dissipation is neglected because the dissipated
energy in both the membrane and the bulk fluid is inversely
proportional to the time scale. Then, the total energy is
composed of three parts: the bending, edge, and stretching
energies.24
1
2
2 2
E = S
+ 2L + S − L2 −
2
R Rc
where R is the radius of curvature, Rc is the intrinsic radius
of curvature, S is the surface area of a bud, and L is the
neck radius as shown in Figure 1. There are several physical parameters;25 26 the typical value of the bending modulus is about 4 × 10−19 Nm, that of the line tension is
about 5 × 10−12 N, and that of the stretching modulus is
about 1 × 10−7 N/m.
As shown in Figure 1, cap area S is expressed as
S = 2R2 1−cos = 4R2 sin2 /2 ≡ 4R2 Bd , and the
difference between the cap area and the projected area
2 2
is S − L2 = S = SBd =
4R Bd , where the radius of
S
the projected area is L =
1 − Bd = 2R Bd 1 − Bd ,
J. Nanosci. Nanotechnol. 11, 1–5, 2011
Bae et al.
Budding Dynamics of the Lipid Membrane
4. RESULTS
Fig. 1.
Schematic picture of the budding process.
where Bd is the budding parameter describing the membrane geometry (i.e., Bd is equal to zero for a flat surface and approaches unity for a = closed spherical
bud). Finally, total energy is calculated using the following
equation:
E = 8Bd 1 − R/Rc 2 + 4R Bd 1 − Bd + 4R2 Bd2
When the bud is about to be detached from the substrate,
Bd becomes unity, and the line tension is not important. At
this moment, the energy has a local minimum value at a
given R. This minimum value can be obtained by taking a
derivative on the total energy with respect to R. The energy
minimum is achieved when the bud radius becomes
R=
Rc
R2c + 1
(1)
where = /4. Then, the bud size will have a maximum
value when the intrinsic radius becomes
Bd
2 1 − Bd /Bd + −
1 − Bd /Bd ± Rc =
(3)
Bd
The time scale of the budding process is presumably governed by the molecule diffusion to the membrane matrix. It
is assumed that
area of the membrane increases
√
√ the surface
in time as S/∼ Dt, where D is the diffusion coefficient. This implies that budding parameter Bd linearly
increases in time as Bd ∼t/T , where T = 4R2 /D. By substituting Bd with the normalized time in Eqs. (2) and (3),
the time evolution of bud radiuses can be obtained. (The
further results are shown in the next session.)
J. Nanosci. Nanotechnol. 11, 1–5, 2011
5. DISCUSSION
It was observed that the budding process of the lipid film
via spin-coating uniformly occurs at the same time. The
size and spacing distribution of the buds are narrowly dispersed, following the Gaussian distribution. The time evolution of the bud size can be well predicted by the newly
developed model using the energy balance in the equilibrium state. From the point of view of the experimental
technique, the presented result implies that the spin-casting
technique can be effectively utilized to come up with
uniform-sized liposomes based on the film format. It is
envisioned that the spin-casting technique will become an
alternative way of efficiently and simply controlling the
sizes of liposomes. Further studies are under way to minimize the water drop volume during the hydration step
of this method, and attempts are being made to encapsulate some drug candidates, such as insulin or testerone,
into liposome buds. There is also a plan of developing
a theory for the prediction of the spacing between buds,
3
RESEARCH ARTICLE
= /2 is about 125 × 1011 m−2 . The bud radius
where R
(R) has a maximum value when
= 0, that is, when
Rc
R2c = 1. When put back to Eq. (1), R will have a maxi
mum value of about 1.4 m, which is close to the observed
value (the average diameter of the ∼2 m bud). The uncertainty comes from the estimated values of and .
At this point, the above analysis is generalized to predict
the changes in the bud size over time. In this case, it is not
necessary for budding parameter Bd to be in the middle of
the budding process. By modifying the above analysis, the
energy is minimized at
1 − Bd /Bd
Rc − R2c
(2)
R=
R2c Bd + 1
Figure 2(A) shows the images of the lipid film before
and after hydration, as observed using a brightfield/phase
contrast microscope. The lipid film is observed to have a
non-transparent flat surface before hydration. After hydration, however, the flat film develops a wrinkled surface,
and 10 min after hydration, it starts to form localized buds.
The lipid buds are observed to continuously grow during
the first 60 min. The budding of the lipid film occurs easily
at room temperature because of the lower melting temperature of the DOPC lipid film (Tm = −20 C). The lipid
film then undergoes a morphological change into spherical
buds around 45 min after hydration.
In the image analysis of the lipid buds after 45 min, 215
buds were identified, and their average size and spacing
were determined to be 23 ± 05 and 41 ± 15 m, respectively. The bud sizes increased with time after hydration,
while the spacing between the buds, was almost constant
(see Fig. 2(B)). Moreover, it was found that the sizes and
spacing of the buds are distributed in a narrow Gaussian
curve (see Figs. 2(C and D)). This suggests that the homogeneity of the lipid film deposited on a substrate may
have the significant effect of producing uniform liposome
patterns.
The theoretical model developed herein predicts the
time dependence of the bud size via quasi-equilibrium
analysis. Figure 3 compares the bud size (D) from the theoretical model with that obtained via experimental observation. Both the theoretical and experimental results show
the trend of the bud size gradually growing from zero and
then sharply increasing just before the bud takes off from
the surface.
RESEARCH ARTICLE
Budding Dynamics of the Lipid Membrane
Bae et al.
Fig. 2. (A) Images of the lipid film before and after hydration, obtained using a brightfield/phase contrast microscope. The lipid film had a clear, flat
surface before hydration and began to wrinkle and bud after hydration (scale bar: 10 m). (B) Average diameter and spacing of the buds. The bud size
slowly increased while the spacing between the buds was almost constant over time. (C) Size distribution of the buds depending on the time (min). (D)
Spacing distribution between the buds depending on the time (min). The size and spacing distribution of the buds followed the Gaussian distribution.
which is not yet well understood. This may provide sizecontrollable and systematically patterned liposomes with
the higher loading efficiency of drugs.
Acknowledgments: This study was supported by the
Dasan project through a grant provided by the Gwangju
Institute of Science and Technology in 2010, by the NCRC
program of Korean Science and Engineering Foundation,
which is being funded by the South Korean government
(Grant No. R15-2008-006-02002-0), and by a National
Research Foundation of Korea (NRF) grant funded by
the South Korean government (MEST) (No. 20100007782;
Midcareer Researcher Program).
References and Notes
Fig. 3. Equilibrium size of a bud versus a normalized time. The
blue line is from theoretical model and the red circles are from the
experiments.
4
1. S. Svetina, Chem. Phys. Chem. 10, 2769 (2009).
2. D. Marsh, Handbook of Lipid Bilayers, CRC press, New
York (1990).
3. D. D. Lasic, Handbook of Biological Physics, edited by R. Lipowsky
and E. Sackmann, Elsevier, Amsterdam (1995), Vol. 1, p. 494.
J. Nanosci. Nanotechnol. 11, 1–5, 2011
Bae et al.
4. Z. Drulis-Kawa and A. Dorotkiewicz-Jach, Int J. Pharm. 387, 187
(2010).
5. D. G. Fatouros, P. V. Ioannou, and S. G. Antimisiaris, J. Nanosci.
Nanotechnol. 6, 9 (2006).
6. E. Goncalves, R. J. Debs, and T. D. Heath, Biophys. J. 86, 1554
(2004).
7. D. C. Drummond, M. Zignani, and J. Leroux, Prog. Lipid Res.
39, 409 (2000).
8. K. Kono, Adv. Drug Deliv. Rev. 53, 307 (2001).
9. V. P. Torchilin, Nat. Rev. Drug Discovery 4, 145 (2005).
10. Brigger, C. Dubernet, and P. Couvreur, Advanced Drug Delivery
Reviews 54, 631 (2002).
11. M. Ferrari, Nat. Nanotechnol. 3, 131 (2008).
12. S. Unezaki, K. Maruyama, J. Hosoda, I. Nagae, Y. Koyanagi,
M. Nakata, O. Ishida, M. Iwatsuru, and S. Tsuchiya, Int. J. Pharm.
144, 11 (1996).
13. Ishida, K. Maruyama, K. Sasaki, and M. Iwatsuru, Int. J. Pharm.
190, 49 (1999).
14. S. J. Kost and Y. Barenholz, Chem. Phys. Lipids 162, 1 (2009).
15. G. Zhai, J. Wu, G. Xiang, W. Mao, B. Yu, H. Li, L. Piao, L. J. Lee,
and R. J. Lee, J. Nanosci. Nanotechnol. 9, 2155 (2009).
Budding Dynamics of the Lipid Membrane
16. D. Papahadjopoulos and J. C. Watkins, Biochim. Biophys. Acta.
135, 639 (1967).
17. F. Olson, C. A. Hunt, F. C. Szoka, W. J. Vail, and
D. Papahadjopoulos, Biochim. Biophys. Acta. 557, 9 (1979).
18. D. J. Estes and M. Mayer, Colloids Surf., B-Biointerfaces 42, 115
(2005).
19. K. Tsumoto, H. Matsuo, M. Tomita, and T. Yoshimura, Colloids and
Surfaces, B-Biointerfaces 68, 98 (2009).
20. C. Leirer, B. Wunderlich, V. M. Myles, and M. F. Schneider, Biophys. Chem. 143, 106 (2009).
21. L. Li, X. Liang, M. Lin, F. Qiu, and Y. Yang, J. Am. Chem. Soc.
127, 17996 (2005).
22. T. Franke, C. T. Leirer, A. Wixforth, N. Dan, and M. F. Schneider,
Chem. Phys. Chem. 10, 2852 (2009).
23. E. Hernandez-Zapata, L. Martinez-Balbuena, and I. SantamariaHolek, J. Biol. Phys. 35, 297 (2009).
24. R. Lipowsky, J. Phys. II France 2, 1825 (1992).
25. S. Semrau, T. Idema, L. Holtzer, T. Schmidt, and C. Storm, Phys.
Rev. Lett. 100, 088101 (2008).
26. W. Rawicz, K. C. Olbrich, T. McIntosh, D. Needham, and E. Evans,
Biophys. J. 79, 328 (2000).
Received: 9 July 2010. Accepted: 15 January 2011.
RESEARCH ARTICLE
J. Nanosci. Nanotechnol. 11, 1–5, 2011
5