Stabilization of vesicular and supported membranes by glycolipid oxime polymers

Article abstract of DOI:10.1039/c0cc05137c

We report herein new synthetic glycolipid dimers and polymers that provide unprecedented stability to both supported (SLBs) and vesicular lipid bilayers against dehydration and serum exposure. These novel physical properties will enable pharmaceutical delivery and development of SLB bioanalytical devices. The Royal Society of Chemistry.

Full text of DOI:10.1039/c0cc05137c

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 2853–2855
www.rsc.org/chemcomm
COMMUNICATION
Stabilization of vesicular and supported membranes by glycolipid oxime
polymersw
Mingming Ma, Soumitra Chatterjee, Meng Zhang and Dennis Bong*
Received 23rd November 2010, Accepted 5th January 2011
DOI: 10.1039/c0cc05137c
We report herein new synthetic glycolipid dimers and polymers
that provide unprecedented stability to both supported (SLBs)
and vesicular lipid bilayers against dehydration and serum
exposure. These novel physical properties will enable pharma-
ceutical delivery and development of SLB bioanalytical devices.
The most widely used method to create long-circulating vesicle
carriers1 is membrane incorporation of lipid-anchored
polyethylene glycol (PEG),2 which sterically blocks serum
protein binding, thus preventing lipid extraction, opsonization
and immunoclearance. Similar protective effects were observed
with GM1,3 though difficulties in accessing GM1 rendered its
application in delivery impractical. To date, all long-circulating
vesicle applications use PEG–lipids, with few improvements to
Fig. 1 Synthetic compounds used in this study: diketo-trehalose
the original system.1 We report herein new, synthetically
(inset 1), monoaminooxyether lipids 2 and 3, tris-aminooxyetherlipids
accessible glycolipid polymers that stabilize membranes to
4 and 5. Oxime dimers 6 and 7 were derived from coupling of 1 with
exposure to serum, air/freeze-drying and rehydration conditions.
cholesterol 2 and phospholipid 3, respectively, while oxime polymers
We anticipate that glycolipid systems of this type will be useful
were similarly derived from coupling of 1 with cholesterol 4 and
for the stabilization of synthetic membranes for delivery and
SLB bioanalytical devices.4
phospholipid 5, respectively.
(Fig. 1). Boc-cleavage with trifluoroacetic acid yielded the
aminooxyether lipid TFA salts, which were directly reacted
with diketo-trehalose 1 in either water/methanol/chloroform
mixtures or acidic aqueous buffer to cleanly produce oxime
dimers from the monoamino-oxyether lipids, and branched
sugar–lipid oxime polymers from the tris-aminooxyether
lipids.
We have used the a,a-disaccharide of glucose, trehalose, as a
lipid crosslinker: tris-aminooxyether headgroup lipids were
condensed with a synthetic diketo-trehalose derivative to yield
oxime-linked glycolipid polymers (Fig. 1). a,a-Trehalose,
which has previously been utilized in gene delivery,5 is an
ideal disaccharide linker from considerations of chemical
convenience and commercial accessibility. The non-reducing
1,1-glycoside skeleton is unreactive with aminooxyether lipids
and permits symmetric regioselective installation of keto
While in situ reaction of pre-formed vesicles containing
aminooxyether lipids resulted in large vesicle aggregates,
hydration of mixed lipid films containing defined mole fractions
of oxime-linked glycolipids yielded well-behaved suspensions.
Both oxime dimers and polymers were found to be stable at
37 1C at neutral pH (PBS) for several days, as monitored by
mass spectroscopy.6b Polymer size, tunable by reaction time
and concentration, was determined by diffusion ordered NMR
(DOSY) using PMMA molecular weight standards for
calibration (Fig. 2A).7 Larger (20 kD) cholesterol polymers
were less soluble than the 7 kD cholesterol oligomer (Bhexamer)
(8), while POPE-derived 20 kD polymer (B15mer) (9) and
both sterol and POPE dimers (6 and 7, respectively) were
soluble in methanol/chloroform mixtures. Organic solvent
solubility permitted the preparation of mixed lipid films, which
could be hydrated into vesicles. Notably, hydration and
extrusion of a mixed lipid film containing ePC and either
0
functionality at the C6 and C6 positions without the use of
0
protecting groups. Transformation of the C6 and C6 primary
alcohols into iodides6 was followed by a high-yielding 3-step
sequence of thiolacetate addition, hydrolysis and conjugate
addition to methylvinyl ketone (ESIw). Mono and tris-
aminooxyether lipids were prepared by acylation of cholesterol
or 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-ethanolamine
(POPE) with Boc-protected aminooxyether headgroups
Department of Chemistry, The Ohio State University,
100 W. 18th Avenue, Columbus, OH 43210, USA.
E-mail: bong@chem.osu.edu
w Electronic supplementary information (ESI) available: Synthetic
and experimental procedures, compound characterization, DOSY
spectra, DLS and zeta potential data. See DOI: 10.1039/c0cc05137c
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 2853–2855 2853

View Article Online
percent sterol (cholesterol/8) was used to allow release to be
detected on the time scale of a few days. Vesicle stabilization
by the glycolipid derivatives was strongly affected by degree of
oligomerization and solubility. With both cholesterol and
phospholipid anchors, the trehalose lipid dimers offered
significant stabilization in FBS, but the polymeric trehalose–
lipids always yielded superior protection per mole lipid
anchor. Though there is clearly an advantage to structures
larger than a dimer, larger oligomer or polymers are limited by
their ability to incorporate into the lipid membrane. Thus, it
was difficult to achieve mole percent incorporation levels
higher than 10% for 9 and 30% for 8.
A unique property of 30% 8 in DPPC or hydrogenated ePC
LUVs is that they may be lyophilized and rehydrated with
minimal contents release (average of three trials was B10%
contents loss) when calcein dye was encapsulated; further, no
significant changes in vesicular size or structure were observed
by DLS and cryo-TEM (Fig. 3). However, total contents loss
and extensive fusion was observed when cholesterol or dimer 6
was used in place of 8, or if hydration was attempted at
elevated temperatures (B70 1C). Sugar solutions (10 : 1
sugar : lipid) are known to protect LUVs from contents loss
and fusion upon lyophilization and rehydration, and we find
that covalent presentation of polymeric trehalose on multiple
lipid anchors confers significant membrane anhydrobiotic and
cryo-protection at sugar : lipid ratios of 0.1–0.3. Rehydration
of freeze-dried gel phase LUVs without size change requires
steric inhibition of vesicle fusion, which appears to be
provided by the membrane-anchored trehalose–cholesterol
oligomers.11 Cholesterol–polymer containing vesicles deposited
poorly on glass, resulting in low yields of SLB formation,
while LUVs containing PE-derived 7 and 9 readily deposited,
possibly by virtue of higher membrane fluidity. Though dimer
7 preserved SLBs from delamination (uncoating from the glass
Fig. 2 (A) Diffusion coefficients for PMMA standards and oxime
polymers derived from cholesterol (8) and phospholipid (9). (B) DLS
and zeta potential of ePC LUVs with 8 or 9. Calcein release on
exposure to 30% FBS at 37 1C from (C) ePC LUVs with the indicated
mole percent POPG ( ), 5% PEG-DSPE ( ), dimer 7 (- - -) and 20 kD
PE–polymer 9 (—); (D) ePC LUVs with 30% cholesterol ( ), 30%
cholesterol, 5% PEG-DSPE ( ), cholesterol dimer 6 (- - -) and
cholesterol polymer 8 (—). Total cholesterol (free and glycolipid)
was constant at 30%. Measurement made over 20 hours and
100 hours in (C) and (D), respectively.
glycolipid polymer required elevated temperatures (50 1C)
while ePC films are easily handled at room temperature.
Neither polymer could be suspended in water in the absence
of a lipid host environment. While this is unsurprising for the
sterol polymer, we expected the phospholipid glycopolymer 9
to be more soluble. These altered properties may derive from
the rigidity of the trehalose linker which may inhibit proper
hydrophobic burial and suspension in water. Notably, the
phospholipid dimer 7 could easily be hydrated to form a
suspension, underscoring the differences in dimer and polymer
of both physical properties and function.
Szoka and co-workers have shown that membranes with
multiply-anchored cholesterol lipids exhibit enhanced stability.8
Consistent with these findings, cholesterol glycopolymer 8 in
LUVs provided the most effective stabilization in 30% fetal
bovine serum at 37 1C as judged by a contents release assay9
(Fig. 2). The cholesterol and phospholipid anchored polymers
8 and 9 both yielded significant protection in FBS relative to
controls, as did the respective trehalose–lipid dimers 6 and 7.
The superior stabilization exhibited by the cholesterol-based
glycopolymer likely derives from the greater inherent stability
of cholesterol-containing membranes.10 The phospholipid
series was compared to control LUVs wherein negatively
charged phospholipid (POPG) replaced 20 kD polymer 9,
which has one negative charge per lipid anchor; indeed, dimer
7 and polymer 9–LUVs had zeta potentials similar to LUVs
containing the same mole fraction of POPG (Fig. 2B).
Likewise, cholesterol dimer 6 and polymer 8–LUVs had sizes
and low magnitude zeta potentials similar to LUVs composed
of only zwitterionic PC and neutral cholesterol lipids. A fluid
phase membrane composed of 70% egg PC (ePC) and 30 mole
Fig. 3 LUVs containing 30% 8 analyzed using (A) dynamic light
scattering (1731) and cryo-TEM (B, C). Cryo-TEM images of (left)
freshly hydrated and (right) lyophilized and resuspended: (B) DPPC
LUVs and (C) hEPC (scale bar = 200 nm).
c
2854 Chem. Commun., 2011, 47, 2853–2855
This journal is The Royal Society of Chemistry 2011

View Article Online
we are optimistic that glycolipid polymer systems may be
further engineered to completely block contents loss while
maintaining vesicular shape and size during freeze-drying.
Furthermore, though some glycosylated lipids are highly
immunogenic,21 mycobacteria-derived trehalose dimycolate
lipid (TDM) is itself nonimmunogenic when membrane
anchored, suggesting that synthetic trehalose lipids may be
similarly inert.22 We anticipate that the glycolipid polymer
stabilized membranes reported herein will be of use as
long-circulating drug carriers. Overall, these data indicate that
glycolipid polymers stabilize vesicular and supported lipid
membranes to anhydrobiotic and cryogenic conditions,
which has not been previously demonstrated with lipid
polymerization23 or any other strategy. Thus, membrane-
protection with glycolipids and related biomaterials holds
promise as an enabling technology for delivery and membrane-
based devices.24
Fig. 4 Supported lipid bilayers containing (A) 7 and (B) 9, mole
percent as indicated. Top rows (A) and (B): freshly formed SLBs;
bottom rows: after drying and rehydration, 10ꢀ magnification, scale
bar = 100 mm. FRAP in SLBs with 10% 9, before dehydration
(C) and after dehydration/rehydration (D), 40ꢀ magnification, scale
bar = 40 mm.
surface) during dehydration–rehydration, considerable surface
scarring occurred, whereas PE–polymer 9 preserved both
surface uniformity and membrane fluidity, as judged by
fluorescence recovery after photobleaching (FRAP) at just
10 mole percent loading of 9 (Fig. 4), similar to prior reports
of protection of lipid monolayers by synthetic trehalose
lipids.12 As with vesicular membranes, polymeric glycolipid 9
offered superior anhydrobiotic SLB protection relative to
dimer 7, again suggesting that presentation of covalently
clustered sugars as afforded by the polymeric structure of 9
is key. Stabilization of membranes to physical insults over a
wide range of hydration is hugely enabling for the development
of bilayer-based devices.13 The current method provides a
strategy that may be suitable for stabilization of free-standing
membranes. Trehalose exhibits exceptional function among
disaccharides with regard to protection against anhydrobiotic
and cryogenic of SLBs,14 protein15 and vesicles,11 leading to its
use in pharmaceutical formulations;16,17 synthetic and native
trehalose lipids are known similarly to protect SLBs.12,18
Trehalose also possesses a number of unique physical properties:
it has the largest hydrated radius of any disaccharide, the
highest glass transition temperature, reversible, hydration-
dependent polymorphism, and kosmotropic function in water.
These physical properties are thought to allow trehalose to
protect biomolecular assemblies (such as membranes) in a
glassy solid sugar coating that is stable to varying hydration
levels.15 We are continuing to investigate the significance of
sugar structure to membrane stability, though we note that the
present trehalose-based system is conveniently accessible
and exhibits unprecedented stability over a wide range of
hydration. Our glycolipid strategy complements existing
stabilization approaches1,19 and allows access to neutral or
charged glycolipid polymer membranes that have unique
anhydrobiotic and cryogenic stability. PEG-protected vesicles
are the only marketed immunoevasive liposomal drug carriers
but cannot be freeze-dried without external protectant;1,20
Notes and references
1 T. Kaasgaard and T. L. Andresen, Expert Opin. Drug Delivery,
2010, 7, 225–243; M. L. Immordino, F. Dosio and L. Cattel, Int. J.
Nanomed., 2006, 1, 297–315.
2 T. M. Allen, C. Hansen, F. Martin, C. Redemann and A. Yau-
Young, Biochim. Biophys. Acta, Biomembr., 1991, 1066, 29–36.
3 T. M. Allen and A. Chonn, FEBS Lett., 1987, 223, 42–46.
4 E. Reimhult and K. Kumar, Trends Biotechnol., 2008, 26, 82–89.
5 S. Srinivasachari, Y. Liu, G. Zhang, L. Prevette and
T. M. Reineke, J. Am. Chem. Soc., 2006, 128, 8176–8184.
6 (a) J. M. G. Fernandez, C. O. Mellet, J. L. J. Blanco, J. F. Mota,
A. Gadelle, A. Coste-Sarguet and J. Defaye, Carbohydr. Res.,
1995, 268, 57–71; (b) J. Kalia and R. T. Raines, Angew. Chem.,
Int. Ed., 2008, 47, 7523–7526.
7 A. Chen, D. Wu and C. S. Johnson, J. Am. Chem. Soc., 1995, 117,
7965–7970.
8 Z. Huang and F. C. Szoka, Jr., J. Am. Chem. Soc., 2008, 130,
15702–15712.
9 R. R. C. New, Liposomes: A Practical Approach, Oxford, NY,
1990.
10 R. B. Gennis, Biomembranes, 1990.
11 J. H. Crowe, L. M. Crowe, A. E. Oliver, N. Tsvetkova, W. Wolkers
and F. Tablin, Cryobiology, 2001, 43, 89–105.
12 C. W. Harland, Z. Botyanszki, D. Rabuka, C. R. Bertozzi and
R. Parthasarathy, Langmuir, 2009, 25, 5193–5198.
13 Y. Deng, Y. Wang, B. Holtz, J. Li, N. Traaseth, G. Veglia,
B. J. Stottrup, R. Elde, D. Pei, A. Guo and X. Y. Zhu, J. Am.
Chem. Soc., 2008, 130, 6267–6271.
14 F. Albertorio, V. A. Chapa, X. Chen, A. J. Diaz and P. S. Cremer,
J. Am. Chem. Soc., 2007, 129, 10567–10574.
15 N. K. Jain and I. Roy, Protein Sci., 2009, 18, 24–36.
16 United States Pat., 6221385, 1997.
17 J. F. Willart and M. Descamps, Mol. Pharmacol., 2008, 5, 905–920.
18 C. W. Harland, D. Rabuka, C. R. Bertozzi and R. Parthasarathy,
Biophys. J., 2008, 94, 4718–4724.
19 S.-M. Lee, H. Chen, C. M. Dettmer, T. V. O’Halloran and
S. T. Nguyen, J. Am. Chem. Soc., 2007, 129, 15096–15097.
20 G. L. Plosker, Drugs, 2008, 68, 2535–2551.
21 G. De Libero and L. Mori, Prog. Lipid Res., 2010, 49, 120–127.
22 R. L. Hunter, M. R. Olsen, C. Jagannath and J. K. Actor, Ann.
Clin. Lab Sci., 2006, 36, 371–386.
23 A. Mueller, B. Bondurant and D. F. O’Brien, Macromolecules,
2000, 33, 4799–4804.
24 F. Albertorio, A. J. Diaz, T. Yang, V. A. Chapa, S. Kataoka,
E. T. Castellana and P. S. Cremer, Langmuir, 2005, 21, 7476–7482.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 2853–2855 2855