Lipid-based nanotubes as functional architectures with embedded fluorescence and recognition capabilities

Article abstract of DOI:10.1021/ja0446449

A limited combinatorial strategy was used to synthesize a small library of soft lipid-based materials ranging from structurally unordered fibers to highly uniform nanotubes. The latter nanotubes are comprised of a bilayer structure with interdigitated alk

Full text of DOI:10.1021/ja0446449

Published on Web 11/02/2004
Lipid-Based Nanotubes as Functional Architectures with Embedded
Fluorescence and Recognition Capabilities
George John,*,†,‡ Megan Mason,† Pulickel M. Ajayan,§ and Jonathan S. Dordick*,†
Department of Chemical and Biological Engineering and the Rensselaer Nanotechnology Center, and
Department of Materials Science and Engineering, Rensselaer Polytechnic Institute, 110 Eighth Street,
Troy, New York 12180-3590
Received September 3, 2004; E-mail: dordick@rpi.edu; john@sci.ccny.cuny.edu
Soft nanotubular structures represent a potentially powerful
architecture generated through self-assembly of amphiphilic
molecules.1-5 Several classes of amphiphiles are known to pro-
vide these materials, including lipid-modified peptides2, bola-
amphiphiles3, and sugar-lipid conjugates.4 An example of the latter
is the sugar-derivatized cardanols, which consist of a carbohydrate
headgroup and an aliphatic alkyl chain connected through a phenyl
moiety.5a Under optimal solution conditions, these alkylphenyl-
glucopyranosides form fibrous aggregates and nanotubes upon
dispersion in water, particularly when the alkyl chain is unsaturated
leading to a bent structure that induces supramolecular chirality.6
Nevertheless, despite the ability of a wide range of compounds to
assemble into coiled fibers,5,6 no clear design rules have been
formulated. Such information is critical to advance applications in
medicine,7 chemical and biological sensing,8 and sub-micro-total
analysis system (sub-µ-TAS) designs.9 For this reason, we embarked
on a limited combinatorial strategy to synthesize lipid-based
nanotubes. Specifically, by combining simple monosaccharides,
fatty acids, and diamino-aromatic linkers, we generated a library
of products ranging from fibers that lack structural regularity to
highly uniform nanotubes. These results represent a systematic
strategy to design functional lipid nanotubes with precise structural
and functional features.
Three monosaccharides (â-D-glucose, â-D-galactose, and â-D-
mannose), three C-18 fatty acids (stearic, oleic, and linoleic), and
six diamino-aromatic linkers (2,3-; 2,5-; and 2,6-diaminopyridine
(DAP) and 1,2-; 1,3-; and 1,4-diaminobenzene) were reacted (see
Supporting Information) according to Figure 1a to give the
monomer components used in this study. Of particular interest were
compounds 1 and 2, which formed fibrous assemblies upon
Figure 1. (a) Synthetic scheme for sugar-containing DAP alkylamides.
dispersing in water following vortexing at 100 °C for 30 min, slow
cooling to room temperature, and incubation for 12 h.
(b) TEM images of self-assembled morphologies of 1 in water, helical coiled
fibers, and nanotubes. The dimensions of the nanotubes were uniform and
reproducible using 0.16 mM of 1 at pH 7.
Visible and fluorescence microscopy showed bundles of self-
assembled structures, with the latter showing fluorescence properties
identical to those of free DAP (λex ) 263 nm; λem ) 535 nm).
TEM images after 4 h showed that the precipitates from 1 formed
helical ribbon morphologies, which upon aging for an additional
12 h yielded nanotubes with an outer diameter of 60-80 nm and
an inner diameter of ca. 20 nm (Figure 1b). DSC analysis of the
nanotubes from 1 showed a gel-to-liquid crystalline-phase transition
temperature (Tm) of 70 °C, which was far higher than the Tm of 42
°C for the unsaturated cardanyl glucoside system.5a,b Despite the
relatively high Tm, the nanostructures from 1 were not highly
crystalline. The lack of the double bond in the alkyl chain to give
2 had a significant effect on the morphology of the resulting
nanostructure. Instead of nanotubes, the precipitates from 2 formed
fibrous structures with a thickness of ca. 80-150 nm. These fibers
had a relatively high Tm of 90 °C, indicating more significant
crystalline packing compared to the unsaturated systems from 1.
Hence, the unsaturated oleic acid moiety appears to be critical in
nanotube formation.
To gain insight into the molecular orientation and packing profile
within the assembled morphologies from 1 and 2 and to understand
why the two similar monomers gave strikingly different self-
assembled morphologies, we examined the wet and dry forms of
the self-assembled nanofibers via small-angle X-ray scattering. The
molecular length of the amphiphiles was calculated by CPK
modeling on the basis of single-crystalline data of oleic acid,10 and
then X-ray diffraction patterns were obtained. The small-angle
diffraction patterns of the nanotubes from 1 revealed ordered
reflection peaks with a long period of 3.5 nm, which is substantially
† Department of Chemical and Biological Engineering.
‡ Current address: Department of Chemistry, The City College of the City
University of New York, Convent Avenue at 138th Street, New York, NY 10031.
§ Department of Materials Science and Engineering.
9
15012
J. AM. CHEM. SOC. 2004, 126, 15012-15013
10.1021/ja0446449 CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Figure 3. (a) Fluorescence quenching of DAP nanotubes by thymidine
(0.16 mM nanotube based on concentration of 1). The thymidine concentra-
tions ranged from (top to bottom) 0 to 10 mM, emission λmax ) 535 nm.
(b) Schematic representation of the possible interaction between nanotubes
of 1 and thymidine.
less effective in quenching fluorescence. The interaction of the
nanotubes from 1 with thymidine was extended to polyT (Mw
8
kDa; up to 0.16 mM and giving an apparent binding constant of
1.7 × 105 M-1), which quenched fluorescence similar to that of
thymidine itself.
Figure 2. Proposed self-assembled nanostructures from 1 and 2. The
unsaturation in 1 results in a kink and a slightly less layered interdigitation.
The planarity of DAP is shown (as a side view) in the bottom picture.
On the basis of these results, the structural requirements for self-
assembly of amphiphilic monomers into highly organized nanotubes
have begun to be elucidated. These include the combination of
strongly hydrophobic and hydrophilic moieties, hydrogen bonding
interactions of hydrophilic groups that are favored in sugars with
equatorial anomers at the 2 and 4 positions of a pyranose ring, and
potentially a linker with suitable planarity. In addition, substantial
bending of the monomers is required, which arises from the meta
orientation of the linker, along with unsaturation of the alkyl chain.
This information can be used to design single-chain amphiphiles
that form high-axial-ratio nanostructures starting from simple
molecules, which also contain molecular recognition groups that
can be used to monitor the chemical selectivity of supramolecular
aggregates toward guest binding.
smaller than twice the extended molecular length of 1 (d-spacing
of 3.03 nm by the CPK molecular modeling).
These results strongly suggest that the nanotubes from 1 form a
bilayer structure with interdigitated alkyl chains associated through
hydrophobic interactions (Figure 2). Moreover, according to powder
X-ray diffraction analysis (see Supporting Information), the gluco-
pyranoside moieties of the bilayer participate in strong inter-
molecular hydrogen bonding, which results in a highly ordered
chiral packing structure. This combination of hydrophobic and
hydrogen bonding interactions appears to favor the formation of
the nanotubular structure. Conversely, the diffraction pattern of the
nanofibers from 2 (molecular length of 3.2 nm) indicated a shorter
d-spacing of 3.3 nm, which translates into a greater degree of
interdigitation of the bilayer structure (Figure 2). The “kink” in
self-assembled structures from 1 appears to reduce the crystallinity
of the nanostructure, enabling more facile formation of a nanotube.
Further differences between nanostructures from 1 and 2 were
revealed by FT-IR spectroscopy. The C-H stretching peaks were
slightly shifted between 1 and 2 (2854 vs 2850 cm-1, respectively)
and the CdO (amide I) stretching patterns were shifted (1651 and
1654 cm-1, respectively), which are consistent with a more
crystalline structure for the nanofibers derived from 2.5a,c
Acknowledgment. This work was supported by an NSF-
Nanoscale Science and Engineering Center (DMR-0117792).
Supporting Information Available: Synthesis of a set of am-
phiphiles and the self-assembly procedure, optical microscopy images
of the fluorescent nanotubes, and TEM images of nanotubes and
fibers. This material is available free of charge via the Internet at
http://pubs.acs.org.
References
Aminopyridine derivatives are known to function as artificial
receptors that can bind various ligands through complementary
multipoint H-bonding.11 Hence, we reasoned that the DAP residue
could serve as a functional recognition element, and in the process,
its intrinsic fluorescence would be affected by selective interaction
with external ligands. To test this hypothesis, we added water-
soluble compounds that can undergo H-bonding to the nanotube
from 1. Addition of up to 10 mM thymidine caused the nearly
immediate quenching of fluorescence (Figure 3a), with an apparent
binding constant of ∼2.5 × 103 M-1. In addition, thymidine
analogues such as uracil and the anticancer compounds 5-fluoro-
uracil and its prodrug derivative Tegafur also quenched fluores-
cence. The fluorescence quenching was selective for nucleosides;
â-D-glucose and urea, while capable of undergoing extensive
H-bonding, did not quench the fluorescence of the nanotubes from
1, even at concentrations as high as 16 mM (100 fold higher than
the nanotube concentration, based on the concentration of 1). These
results suggest that the interaction between the nanotubes and
thymidine may occur through a three-point hydrogen bonded
network11 (Figure 3b). Consistent with this hypothesis, adenosine,
which cannot bind in this three-point manner, was at least 2-fold
(1) (a) Nakashima, N.; Asakuma, S.; Kunitake, T. J. Am. Chem. Soc. 1985,
107, 509. (b) Yager, P.; Schoen, P. E. Mol. Cryst. Liq. Cryst. 1984, 106,
371. (c) Thomas, B. N.; Safinya, C. R.; Plano, R. J.; Clark, N. A. Science
1995, 267, 1635.
(2) (a) Schnur, J. M. Science 1993, 262, 1669. (b) Hartgerink, J. D.; Beniash,
E.; Stupp, S. I. Science 2001, 294, 1684. (c) Kunitake, T. Angew. Chem.,
Int. Ed. Engl. 1992, 31, 709.
(3) (a) Fuhrhop, J.-H.; Helfrich, W. Chem. ReV. 1993, 93, 1565. (b) Shimizu
T.; Kogiso, M.; Masuda, M. Nature 1996, 383, 487. (c) Oda, R.; Huc, I.;
Schmutz, M.; Candau, S. J.; MacKintosh, F. C. Nature 1999, 399, 566-
569. (d) Matsui, H.; Gologan, B. J. Phys. Chem. B 2000, 104, 3383.
(4) (a) O’Brien, D. F. J. Am. Chem. Soc. 1994, 116, 10057. (b) Engelkamp,
H.; Middlebeck, S.; Nolte, R. J. M. Science 1999, 284, 785.
(5) (a) John, G.; Masuda, M.; Okada, Y.; Yase, K.; Shimizu, T. AdV. Mater.
2001, 13, 715. (b) John, G.; Jung, J. H.; Minamikawa, H.; Yoshida, K.;
Shimizu, T. Chem. Eur. J. 2002, 8, 5494. (c) Jung, J. H.; John, G.; Yoshida,
K.; Shimizu, T. J. Am. Chem. Soc. 2002, 124, 10674.
(6) (a) Selinger, J. V.; Spector, M. S.; Schnur, J. M. J. Phys. Chem. B, 2001,
105, 7157. (b) Berthier, D.; Buffeteau, T.; Leger, J.-M.; Oda, R.; Huc, I.
J. Am. Chem. Soc. 2002, 124, 13486.
(7) Guo, X.; Szoka, F. C., Jr. Acc. Chem. Res. 2003, 36, 335.
(8) (a) Cullum, B. M.; Vo-Dinh, T. TIBTECH 2000, 18, 388. (b) Rothenberg,
B. E.; Hayes, B. K.; Toomre, D.; Manzi, A. E.; Varki, A. Proc. Natl.
Acad. Sci. U.S.A. 1993, 90, 11939.
(9) Vilkner, T.; Janasek, D.; Manz, A. Anal. Chem. 2004, 76, 3373.
(10) Ernst, J.; Sheldrick, W. S.; Fuhrhop, J.-H. Z. Naturforsch. 1979, 34b, 706.
(11) Shenhar, R.; Rotello, V. M. Acc. Chem. Res. 2003, 36, 549.
JA0446449
9
J. AM. CHEM. SOC. VOL. 126, NO. 46, 2004 15013