вход по аккаунту


Membrane-Mimetic Nanocarriers Formed by a Dipalmitoylated Cell-Penetrating Peptide.

код для вставкиСкачать
Drug Delivery
DOI: 10.1002/anie.200500519
Membrane-Mimetic Nanocarriers Formed by a
Dipalmitoylated Cell-Penetrating Peptide**
Sandro Keller,* Ines Sauer, Holger Strauss, Klaus Gast,
Margitta Dathe, and Michael Bienert
Molecular and supramolecular drug-delivery systems have
attracted great attention as pharmaceutical formulations for
the administration of otherwise only poorly soluble, rapidly
degradable, or even toxic molecules.[1] Within the multitude of
colloidal vehicles developed to date, micelles belong to the
most useful tools for the delivery of hydrophobic compounds.[2] However, some serious disadvantages of micellar
carriers remain, among which the problem that some
normally membrane-located biomolecular devices like
vector peptides lose their native conformation and are
severely compromised in their activity when removed from
the lipid bilayer.[3, 4] Here we introduce a dipalmitoylated cellpenetrating peptide that is superior to previously described
biofunctional colloids in its ability to incorporate large
amounts of lipid in mixed micelles. Hence this peptide
constitutes a potent lead for the intracellular delivery of
hydrophobic drugs using bilayer-mimetic nanocarriers.
Ideally, such vehicles would meet a number of criteria.
First, a customizable peptide sequence designed to trigger a
specific response upon interaction with biological targets has
to be exposed on the micellar surface.[5] Second, the nanocarriers should contain an excess of lipid and thus offer a
bilayer-like environment in order to preserve the functionality of the surface-bound peptide. Third, as such a compound
will unavoidably be a strong detergent, its aggregational state
must be easily controllable, and the kinetics of membrane
solubilization should be slow.
In our search for a peptidic vector promoting internalization into a broad range of different cells, we chose a 20residue tandem dimer [A2, (LRKLR KRLLR)2] comprising
binding sites for the low-density lipoprotein receptor and for
cell-surface heparan sulfate proteoglycans, two ubiquitous
mammalian membrane constituents able to mediate uptake of
their ligands.[6] A2 has recently been demonstrated to induce
internalization of liposomes into primary rat brain capillary
endothelial cells when covalently linked to the vesicle surface,[7] and the presence of lipids is known to be crucial for
receptor binding by its parent apolipoprotein E.[3] As fatty
acid modified peptides have turned out to be extraordinarily
powerful detergents for suspending membrane proteins in
mixed micelles providing a bilayer-mimetic environment,[8, 9]
we opted for dipalmitoylation (P2) of two neighboring amino
acids in a 3-residue sequence linked to the N terminus of A2
as a means of hydrophobization. The lipopeptide P2A2 was
prepared with high yields and in quantities of several hundred
milligrams in a simple and effective procedure. A standard
solid-phase protocol based on Na-9-fluorenylmethoxycarbonyl(Fmoc)-protected amino acids was used to synthesize
A2 extended N-terminally with the sequence Fmoc-WK(Dde)G containing an Ne-1-(4,4-dimethyl-2,6-dioxocyclohex1-ylidene)ethyl(Dde)-protected lysine residue. After hydrazine cleavage of the Dde and the amino-terminal Fmoc
moieties, the two amino groups in the lysine side chain and at
the N terminus were acylated simultaneously with palmitic
The resulting chimera P2A2 (Figure 1) with a molar mass
of 3.53 kg mol 1 has an amphipathic structure in which the two
acyl chains form a very hydrophobic tail attached to the
charged peptide, but it is nevertheless well-soluble (> 5 mm)
[*] S. Keller, Dr. I. Sauer, H. Strauss, Dr. M. Dathe, Prof. Dr. M. Bienert
Research Institute of Molecular Pharmacology FMP
Robert-R6ssle-Strasse 10, 13125 Berlin (Germany)
Fax: (+ 49) 30-94793-159
Dr. K. Gast
Physical Biochemistry
Institute of Biochemistry and Biology
University of Potsdam
Karl-Liebknecht-Strasse 24–25, Haus 25
14467 Potsdam-Golm (Germany)
[**] This work was supported by the European Commission (grant no.
QLK3-CT-2002-01989 to SK) and by the Deutsche Forschungsgemeinschaft (grant no. DA 324/4-1/2 to IS and SFB 498 to HS). We
thank Prof. Hartmut Oschkinat and Prof. Peter Pohl for helpful
comments on the manuscript and Dr. Annette Meister, Dr. Michael
SchDmann, Dr. Michael Beyermann, and Dr. Heiko Heerklotz for
fruitful discussions. We are indebted to Heidemarie Lerch for massspectrometric measurements, to Bernhard Schmikale for peptide
synthesis, and to Heike Nikolenko, Carolyn Vargas, Seong-Ji Han,
Jens H. Laettig, and Stephan Pritz for excellent technical assistance.
Supporting information for this article is available on the WWW
under or from the author.
Figure 1. Structure of P2A2. The two palmitoyl chains are linked
through amide groups to the N-terminal tryptophan and to the side
chain of the neighboring lysine (one-letter amino acid code, formal
valence z = 12 at pH 7.4). The hydrophobic methylene and methyl
groups of the palmitoyl residues are highlighted in yellow in the spacefilling model.[11]
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 5252 –5255
in aqueous solution (10 mm 2-amino-2-(hydroxymethyl)propane-1,3-diol (Tris), 154 mm NaCl, 0.1 mm ethylenediaminetetraacetate (EDTA), pH 7.4).[12] Unlike double-chained
phospholipids possessing relatively small headgroups, the
bulky peptide of P2A2 does not appear to allow selfassociation into bilayers but rather implies detergent-like
Analytical ultracentrifugation revealed indeed that P2A2
assembles into well-defined aggregates with a molar mass of
M = 68 kg mol 1 corresponding to 19 monomers (Figure 2 a).
detergent/lipid mole ratio, Rsat:
e , at which membrane solubilization began (Figure 3 a). A further increase in cP2A2 led to a
drastic drop in I until solubilization was completed at Rsol:
e .
This was in excellent agreement with a second type of
Figure 3. Detergent activity of P2A2 on 100-nm POPC vesicles as monitored by right-angle light scattering. a) Scattering intensity, I, obtained
from 100 mm POPC extruded along with increasing P2A2 concentrations, cP2A2. b) I values for various lipid concentrations, cPOPC, prepared
in the presence (circles) or absence (squares) of 10 mm P2A2.
Figure 2. Mass and size characterization of P2A2 micelles. a) Analytical ultracentrifugation at cP2A2 = 96 mm and 20 krpm gave the local
refractive index difference, Dn, (open circles) as a function of the
squared distance from the rotor axis, d2, at sedimentation equilibrium.
The best single-species fit (solid line) was obtained for a molar mass
of M = (67.7 0.3) kg mol 1. b) Dynamic light scattering at
cP2A2 = 244 mm yielded the distribution function, f(rS), of the Stokes
radius, rS. The stronger peaks at small radii observed at four different
concentrations were used to extrapolate r expt
S = (5.25 0.30) nm.
Static light scattering (LS) confirmed this result;[10] however,
dynamic LS yielded a Stokes radius of rexpt
= 5.3 nm (FigS
ure 2 b), which is distinctly larger than the value of rcalcd
3.4 nm calculated for a spherical particle of this mass.[10] This
discrepancy and the small aggregation number most likely
arise from a nonspherical micellar geometry as invoked for
other lipopeptides[8] and are also reminiscent of bile salts.[13]
However, as a practical advantage over bile salts, the
critical micellar concentration (CMC) of P2A2 is expected to
be considerably lower owing to the pronounced hydrophobicity conferred by the two palmitoyl chains, which should
thus preserve the colloidal state also upon extreme dilution.
In fact, the subnanomolar CMC of the lipopeptide is below
the ranges experimentally accessible by fluorescence spectroscopy (nanomolar) and calorimetry (micromolar),[12, 14]
suggesting that the concentration of monomeric P2A2 is
always negligible.
The efficacy of P2A2 in forming lipid-rich micelles was
monitored on the basis of the light scattering intensity, I, of its
mixtures with the zwitterionic phospholipid 1-palmitoyl-2oleoyl-sn-glycero-3-phosphocholine (POPC). To this end, dry
lipid films were suspended in buffer containing various
concentrations of P2A2, and standard procedures for preparing unilamellar vesicles were then applied.[10] For a series of
samples with identical lipid concentration, the I values rose
only slightly with the P2A2 concentration up to a saturating
Angew. Chem. Int. Ed. 2005, 44, 5252 –5255
scattering experiment employing a constant lipopeptide
concentration but different amounts of lipid (Figure 3 b).
Plotting the characteristic P2A2 concentrations at the
break points in the curves obtained with LS and calorimetry[10, 15] versus the lipid concentration yields a simple phase
diagram (Figure 4).[16, 17] The two straight lines with slopes of
Figure 4. Phase diagram of dilute aqueous P2A2/POPC mixtures. Data
are taken from solubilization (filled symbols) and reconstitution (open
symbols) experiments conducted with light scattering (squares) and
titration calorimetry (circles), [10] respectively. The linear regression
lines separate the micellar (m) and the vesicular (v) areas from the
transition range (m + v).[17]
e = 0.012 and Re = 0.13 delimit three areas ascribed to the
micellar, the transitional, and the vesicular range, respectively.[18] Thus, to our knowledge P2A2 is by far the most
efficient detergent in that it can solubilize about seven lipid
molecules per lipopeptide, that is, 50 times more than other
peptides with two fatty acyl chains.[8] The partition coefficient
of P2A2 from vesicles (v) to micelles (m) amounts to Pm=v
P2A2 =
10 and reflects the preference of the lipopeptide for the
micellar state. In contrast, POPC reveals only a slight
penchant for membranes with Pm=v
POPC = 0.9, indicating that
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
the lipid can be accommodated in the micelle almost as well as
in the liposome and underpinning the suitability of P2A2 for
micellar vehicles offering a bilayer-mimetic environment.
Most importantly, the clearcut boundaries in the phase
diagram (Figure 4) permit direct control of the aggregational
state of P2A2/POPC mixtures by adjusting the concentrations
of the components. Hence, in addition to micellar also
liposomal structures are available as supramolecular carriers,
and vesicles doped with P2A2 reveal superior properties in
terms of high lipid affinity, stable bilayer anchorage, and low
membrane permeabilization when compared with the properties of other hydrophobized A2 derivatives.[19] Preliminary
studies with confocal laser scanning microscopy revealed the
efficient internalization of micelles composed of a fluorescently labeled P2A2 derivative (P2fA2) into mouse brain
capillary endothelial cells (Figure 5). Moreover, the slow
kinetics of solubilization[10] even suggests application of
nonequilibrium formulations in which P2A2-oversaturated
vesicles gradually rearrange into mixed micelles. This is
particularly intriguing in light of the increasing importance of
delayed and sustained release of drugs from delivery systems.[20]
Figure 5. Internalization of a carboxyfluorescein-labeled P2A2 derivative (P2fA2) into immortalized b.End3 mouse brain capillary endothelial cells as visualized by confocal laser scanning microscopy. After
exposure to 1 mm P2fA2 for 30 min at 37 8C, this representative cell
displayed a punctuate intracellular fluorescence pattern pointing to a
predominantly endocytotic mode of uptake. Membrane integrity and
cell viability were confirmed by staining with trypan blue.[12]
We have shown that a novel lipopeptide composed of a
target-specific peptidic vector and two adjacent fatty acyl
chains is a highly promising candidate for solubilizing hydrophobic drugs in lipid-rich, membrane-mimetic nanocarriers.
Owing to their small size, these supramolecular vehicles share
many advantages with other micelles,[2] while the bioactivity
of the peptide moiety is maintained thanks to the bilayer-like
environment offered by the palmitoyl tails. These unique
properties and the sound physicochemical characterization
make the colloidal state of P2A2/lipid systems readily
manageable and provide versatile formulations for in vitro
experiments, which have indeed demonstrated rapid cellular
internalization. The straightforward synthesis and good
solubility should render P2A2 amenable to a more detailed
structure elucidation and might also prove useful for other
fields, such as membrane protein reconstitution.[8, 9] Finally,
the peptide sequence need not be restricted to the cell-
penetrating A2 but may be adaptable to a broad spectrum of
ligands destined to diverse biological targets. We are conducting further studies to shed light on the mechanism of
uptake and to extend the approach presented in this work to
similar chimerical lipopeptides.
Received: February 10, 2005
Revised: April 12, 2005
Published online: July 20, 2005
Keywords: drug delivery · lipopeptides · membranes · micelles ·
[1] D. E. Discher, A. Eisenberg, Science 2002, 297, 967 – 973; H.
Dou, M. Jiang, H. Peng, D. Chen, Y. Hong, Angew. Chem. 2003,
115, 1554 – 1557; Angew. Chem. Int. Ed. 2003, 42, 1516 – 1519; R.
Duncan, Nat. Rev. Drug Discovery 2003, 2, 347 – 360; J. Kreuter,
Colloidal Drug Delivery Systems, Marcel Dekker, New York,
1994; R. Langer, Science 2001, 293, 58 – 59; D. D. Lasic, D.
Papahadjopoulos, Science 1995, 267, 1275 – 1276; M. A. Moses,
H. Brem, R. Langer, Cancer Cell 2003, 4, 337 – 341; P. C. Rensen,
R. L. de Vrueh, J. Kuiper, M. K. Bijsterbosch, E. A. Biessen, T. J.
van Berkel, Adv. Drug Delivery Rev. 2001, 47, 251 – 276.
[2] Y. Bae, S. Fukushima, A. Harada, K. Kataoka, Angew. Chem.
2003, 115, 4788 – 4791; Angew. Chem. Int. Ed. 2003, 42, 4640 –
4643; R. Haag, Angew. Chem. 2004, 116, 280 – 284; Angew.
Chem. Int. Ed. 2004, 43, 278 – 282; Y. Kakizawa, K. Kataoka,
Adv. Drug Delivery Rev. 2002, 54, 203 – 222; K. Kataoka, T.
Matsumoto, M. Yokoyama, T. Okano, Y. Sakurai, S. Fukushima,
K. Okamoto, G. S. Kwon, J. Controlled Release 2000, 64, 143 –
153; K. Kataoka, A. Harada, Y. Nagasaki, Adv. Drug Delivery
Rev. 2001, 47, 113 – 131; M. KrGmer, J. F. StumbH, H. TIrk, S.
Krause, A. Komp, L. Delineau, S. Prokhorova, H. Kautz, R.
Haag, Angew. Chem. 2002, 114, 4426 – 4431; Angew. Chem. Int.
Ed. 2002, 41, 4252 – 4256; N. Nasongkla, X. Shuai, H. Ai, B. D.
Weinberg, J. Pink, D. A. Boothman, J. Gao, Angew. Chem. 2004,
116, 6483 – 6487; Angew. Chem. Int. Ed. 2004, 43, 6323 – 6327; R.
Savic, L. Luo, A. Eisenberg, D. Maysinger, Science 2003, 300,
615 – 618; V. P. Torchilin, Cell. Mol. Life Sci. 2004, 61, 2549 –
2559; J. V. Weaver, Y. Tang, S. Liu, P. D. Iddon, R. Grigg, N. C.
Billingham, S. P. Armes, R. Hunter, S. P. Rannard, Angew.
Chem. 2004, 116, 1413 – 1416; Angew. Chem. Int. Ed. 2004, 43,
1389 – 1392.
[3] T. L. Innerarity, R. E. Pitas, R. W. Mahley, J. Biol. Chem. 1979,
254, 4186 – 4190; K. H. Weisgraber, Adv. Protein Chem. 1994, 45,
249 – 302.
[4] L. Nobs, F. Buchegger, R. Gurny, E. Allemann, J. Pharm. Sci.
2004, 93, 1980 – 1992; M. Zaiou, K. S. Arnold, Y. M. Newhouse,
T. L. Innerarity, K. H. Weisgraber, M. L. Segall, M. C. Phillips, S.
Lund-Katz, J. Lipid Res. 2000, 41, 1087 – 1095.
[5] Throughout the text, the term “micelle” is used to denote
aggregates of colloidal size, without making any assumptions
regarding their shape or composition.
[6] C. A. Dyer, L. K. Curtiss, J. Biol. Chem. 1991, 266, 22 803 –
22 806; C. A. Dyer, R. S. Smith, L. K. Curtiss, J. Biol. Chem.
1991, 266, 15 009 – 15 015; C. A. Dyer, D. P. Cistola, G. C. Parry,
L. K. Curtiss, J. Lipid Res. 1995, 36, 80 – 88; M. P. Mims, A. T.
Darnule, R. W. Tovar, H. J. Pownall, D. A. Sparrow, J. T.
Sparrow, D. P. Via, L. C. Smith, J. Biol. Chem. 1994, 269,
20 539 – 20 547.
[7] I. Sauer, I. R. Dunay, K. Weisgraber, M. Bienert, M. Dathe,
Biochemistry 2005, 44, 2021 – 2029.
[8] C. L. McGregor, L. Chen, N. C. Pomroy, P. Hwang, S. Go, A.
Chakrabartty, G. G. PrivH, Nat. Biotechnol. 2003, 21, 171 – 176.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 5252 –5255
[9] C. R. Sanders, A. K. Hoffmann, D. N. Grayn, M. H. Keyes, C. D.
Ellis, ChemBioChem 2004, 5, 423 – 426.
[10] See the Supporting Information for details.
[11] For simplicity, the peptide moiety is depicted as an ideal a helix
even though circular dichroism spectroscopy has shown that the
helicity depends on the aggregational state and never exceeds
50 %.
[12] Data not shown.
[13] M. Almgren, Biochim. Biophys. Acta 2000, 1508, 146 – 163; P.
Garidel, A. Hildebrand, R. Neubert, A. Blume, Langmuir 2000,
16, 5267 – 5275; M. Janich, J. Lange, H. Graener, R. Neubert, J.
Phys. Chem. B 1998, 102, 5957 – 5962; A. Jover, F. Meijide,
E. R. NfflPez, J. V. Tato, M. Mosquera, Langmuir 1997, 13,
[14] S. Paula, W. SIs, J. Tuchtenhagen, A. Blume, J. Phys. Chem. 1995,
99, 11 742 – 11 751.
[15] H. Heerklotz, J. Seelig, Biochim. Biophys. Acta 2000, 1508, 69 –
85; H. Heerklotz, G. Lantzsch, H. Binder, G. Klose, A. Blume,
Chem. Phys. Lett. 1995, 235, 517 – 520.
[16] H. Heerklotz, G. Lantzsch, H. Binder, G. Klose, A. Blume, J.
Phys. Chem. 1996, 100, 6764 – 6774.
[17] Vesicle–micelle coexistence is the simplest but not the only
model providing a consistent interpretation, and other scenarios
might be envisaged, for example, a transition from vesicles to
disk-shaped bilayers or bicelles at Rsat:
followed by a gradual
shrinking caused by a further increase in cP2A2. Rather than a
thermodynamically meaningful parameter, Rsol:
would then be
an arbitrarily defined threshold at which the bilayer disks
become hardly distinguishable in size from pure P2A2 micelles
as judged from the I values.
[18] D. Lichtenberg, Biochim. Biophys. Acta 1985, 821, 470 – 478.
[19] I. Sauer, H. Nikolenko, S. Keller, M. Bienert, M. Dathe, K. Abu
Ajaj, submitted .
[20] S. Mantripragada, Prog. Lipid Res. 2002, 41, 392 – 406; M. P.
Ramprasad, G. M. Anantharamaiah, D. W. Garber, N. V. Katre,
J. Controlled Release 2002, 79, 207 – 218.
Angew. Chem. Int. Ed. 2005, 44, 5252 –5255
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Без категории
Размер файла
137 Кб
penetration, nanocarriers, former, membranes, peptide, cells, mimetic, dipalmitoylated
Пожаловаться на содержимое документа