Assemblies of Linear-Dendritic Copolymers
A R T I C L E S
carboxylic acid groups remained along the copolymer back-
bone.13 Since then, we have determined that it is possible to
quantitatively convert pendant amine groups along a copolymer
backbone to carbamates by reaction with 4-nitrophenyl carbon-
ates. Methods for the high-yielding functionalization of hydroxyl
groups on a polymeric backbone have also been previously
developed in our group.15 With these issues in mind, the target
copolymers for this work are designed to contain blocks with
pendant amines or hydroxyl groups for functionalization.
Although linear block copolymers of PEO with polylysine
and polyserine have the desired pendant amine and hydroxyl
groups, respectively, a drawback to this approach is that the
polymerization of amino acid N-carboxyanhydrides initiated
from PEO-NH216 in our hands was not well controlled, leading
to relatively high polydispersities and homopolymer impurities
that were difficult to remove. Linear-dendrimer hybrid copoly-
mers are attractive alternatives because they can be prepared
by a stepwise sequence leading to well-defined structures with
a controlled number of peripheral functional groups on the
dendrimer for subsequent functionalization. The self-assembly
of various linear-dendrimer hybrids in water has been investi-
gated by several groups, who have reported the formation of
aggregates ranging from unimolecular micelles to multimolecu-
lar spherical micelles and even vesicles.17 In pioneering work,
Chapman and co-workers have prepared PEO-dendritic poly-
lysine copolymers having amine or BOC-protected amine groups
on the dendrimer periphery, with the BOC-functionalized
copolymers showing micellar properties in aqueous solution.18
We therefore hypothesized that it should be possible to attach
acid-sensitive hydrophobic acetal groups to the peripheral amine
groups in an efficient coupling reaction to afford acid-sensitive
micelles. PEO-dendritic polyester block copolymers have
previously been synthesized in our group, but their micellization
properties have not yet been investigated.15a These copolymers
have peripheral hydroxyl groups on the dendrimers that can
serve as functional handles for attachment of the acetals. Using
both of these well-defined systems, it should be possible to tune
the micelle properties such as size, critical micelle concentration
(CMC), and drug release rates by varying the length of the PEO
chain, the dendrimer generation, and the linkage between the
acetals and dendrimer backbone.
Figure 1. Schematic for drug release from a pH-sensitive micelle.
degrade under acidic conditions.12 However, the number of
systems that are responsive within the physiologically accessible
pH range of approximately 4.5-7.4 is quite limited. Here we
describe the development of a new and general approach to
stimuli-responsive micelles using PEO-dendrimer hybrids as
the copolymer backbone. This approach involves the attachment
of hydrophobic groups to the surface of the core-forming
dendrimer block by highly sensitive acetal linkages.13 The
system is designed such that upon hydrolysis of the acetals, the
hydrophobic dendrimer periphery becomes hydrophilic, thus
removing the driving force for self-assembly and destabilizing
the micelle. This destabilization should enable release of the
contents of the micelle from its encapsulating compartment as
illustrated in Figure 1.
The stepwise synthesis and plurivalency of linear-dendrimer
block copolymers make them ideal copolymer scaffolds for the
development of these pH-responsive systems. The length of the
linear poly(ethylene oxide) block and the dendrimer generation
can be easily tuned to afford micelles with a range of sizes and
stabilities. In addition, subtle changes in the copolymer structure
affect the microenvironment at the micelle core and can
significantly affect the rate of acetal hydrolysis, enabling the
control of the rate of micelle degradation. Because these micelle
systems are designed to be very stable at neutral or physiological
pH (7.4), but degrade under mildly acidic conditions, they are
potentially applicable as pH-responsive drug delivery systems
capable of selectively releasing drug molecules in tumor tissue
or within tumor cells.
Results and Discussion
Design. Acetals formed from the 1,3-diols and 2,4,6-
trimethoxybenzaldehyde were chosen as the pH-sensitive link-
ages such that polar diol moieties are revealed on the core-
forming block upon hydrolysis. While cyclic acetals are known
to hydrolyze relatively slowly in comparison to their noncyclic
analogues,14 the electron-donating methoxy groups in the ortho
and para positions accelerate the rate of hydrolysis, providing
a half-life of 1 h at pH 5.0 and 37 °C as previously measured
for low molecular weight model compound.
The amine of a diol such as serinol or the third hydroxyl of
1,1,1-tris(hydroxymethyl)ethane can provide a functional handle
for attachment of the acetals to the polymer backbone. Although
in initial studies the amine of serinol could be conjugated directly
to a copolymer backbone having pendant carboxylic acids, the
yields for this conjugation were not quantitative and titratable
Synthesis. To prepare the target acetal with a single remaining
reactive group available for subsequent coupling to the amine
functionalized copolymers, 2,4,6-trimethoxybenzaldehyde (1)
was reacted with 1,1,1-tris(hydroxymethyl)ethane in dry THF
in the presence of molecular sieves and catalytic p-toluene-
sulfonic acid to give the acetal 2. The remaining hydroxyl group
was then activated by reaction with 4-nitrophenyl chloroformate,
providing the carbonate 3 as shown in Scheme 1. The third-
generation PEO-dendritic polylysine copolymers 4a (PEO-5K-
(15) (a) Ihre, H.; Padilla De Jesu´s, O. L.; Fre´chet, J. M. J. J. Am. Chem. Soc.
2001, 123, 5908. (b) Grayson, S. M.; Fre´chet, J. M. J. Macromolecules
2001, 6542-6544.
(16) (a) Harada, A.; Kataoka, K. Macromolecules 1995, 28, 5294-5299. (b)
Thunemann, A. F.; Beyermann, J.; Kukula, H. Macromolecules 2000, 33,
5906-5911.
(17) (a) Gitsov, I.; Wooley, K. L.; Fre´chet, J. M. J. Angew. Chem., Int. Ed.
Engl. 1992, 31, 1200-1202. (b) Gitsov, I.; Fre´chet, J. M. J. Macromolecules
1993, 26, 6536-6546. (c) van Hest, J. C. M.; Delnoye, D. A. P.; Baars,
M. W. P. L.; van Genderen, M. H. P.; Meijer, E. W. Science 1995, 268,
1592-1595. (d) Iyer, J.; Fleming, K.; Hammond, P. T. Macromolecules
1998, 31, 8757-8765. (e) Chang, Y.; Kwon, Y. C.; Lee, S. C.; Kim, C.
Macromolecules 2000, 33, 4496-4500.
(11) (a) Yoo, H. S.; Lee, E. A.; Park, T. G. J. Controlled Release 2002, 82,
17-27. (b) Bae, Y.; Fukushima, S.; Harada, A.; Kataoka, K. Angew. Chem.,
Int. Ed. 2003, 42, 4640-4643.
(12) (a) Heller, J.; Barr, J.; Ng, S. Y.; Abdellauoi, K. S.; Gurny, R. AdV. Drug
DeliVery ReV. 2002, 54, 1015-1039. (b) Ng, S. Y.; Heller, J. U.S. Patent
2,003,152,630, 2003.
(13) Gillies, E. R.; Fre´chet, J. M. J. Chem. Commun. 2003, 1640-1641.
(14) Fife, T.; Jao, L. J. Org. Chem. 1965, 30, 1492-1495.
(18) Chapman, T. M.; Hillyer, G. L.; Mahan, E. J.; Shaffer, K. A. J. Am. Chem.
Soc. 1994, 116, 11195-11196.
9
J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004 11937