Journal of the American Chemical Society
Communication
to document pH lability. At predetermined time intervals,
dialysis media were completely replaced, and the concentration
of estradiol was determined by HPLC. HPLC and LC-MS
analyses confirmed that estradiol was released from DCE
microparticles in its native form (at pH 7.4, Figure S5). No
burst release was observed, and 50% of estradiol was released
from DCE microparticles at pH 7.4 in ∼14 weeks (Figure 3A).
particles; ∼14 μm, Table S2) with a yield of 67.2 wt % and
drug loading of 6.0 wt %. Scanning electron microscopy (SEM)
showed spherical DCE and DCE/PLGA microparticles (Figure
S7). We documented an acidic microenvironment in DCE/
PLGA microparticles (Figure 3B) by producing maps of the
ratio of the fluorescence intensity at 450 nm to that at 520 nm
(I450 nm /I520 nm) of LysoSensor Yellow/Blue dextran,14,23
encapsulated within the particles; a higher I450/I520 indicates a
higher pH. (These particles were larger [∼30 μm] than those
made of polymer alone [10−14 μm, Table S2], presumably
because a double emulsion method was used to load the dye.)
The pH in PLGA microparticles was lower than in DCE/PLGA
microparticles, which was lower than in DCE microparticles
(Figure 3C). This low-pH microenvironment was the likely
cause of the much more rapid release of estradiol from DCE/
PLGA microparticles in the first 3 weeks (Figures 3A and S8),
compared to DCE microparticles.
DCE microparticle biocompatibility was evaluated after
injection at the rat sciatic nerve, where nerve, muscle, and
connective tissues are in proximity. On dissection 4 and 21 days
after injection, microparticles were visible at the injection site
(Figure S9). Tissues were sectioned and stained with
hematoxylin and eosin. By light microscopy, in all tissues,
particle-shaped lucencies (from particles dissolved in staining)
and inflammatory cells were seen near muscle and nerve.
Macrophages, lymphocytes, and some neutrophils were
observed on day 4, while macrophages, lymphocytes, and
some foreign-body giant cells were observed on day 21.
Inflammation outside of the particle mass and myotoxicity were
scored (see Methods in SI). Scores were low at both time
points, with minimal myotoxicity and mild inflammation that
S3). Deeper layers within the muscle had normal morphology
without inflammation. Nerve tissue appeared intact. These
results were similar to results obtained with polyester
microparticles at the same location.24
In summary, we have developed a facile method to synthesize
di-isopropenyl ether monomers and high molecular weight
polyketals by Lewis acid-catalyzed addition polymerization of
di-isopropenyl ether and diol monomers. Estradiol was used as
a model drug in the synthesis of a hydrophobic estradiol
polyketal conjugate. Because estradiol itself is a building block
of the polymer, drug loading was high, and estradiol release was
slow. In vitro release of native estradiol from estradiol-polyketal
conjugate microparticles occurred over more than 4 months.
Drug release kinetics was altered by adding PLGA. Tissue
reaction in vivo was benign. These materials may be useful for
very prolonged sustained delivery as well as for pH-responsive
biomaterials.25 The approach described here could be
applicable to other alcohol- or thiol-containing drugs which
do not have interfering reactive functional groups, such as
unprotected amines and carboxylates. Polyketals with func-
tional pendant groups could also be synthesized from
functional diols, such as protected amine or maleimide group
containing diols, to allow drug conjugation to the polymer.
Figure 3. DCE microparticles: drug release and acidic microenviron-
ment. (A) Cumulative release of estradiol (%) from microparticles at
37 °C in PBS (pH = 7.4, 0.1 wt % Tween 80). §: Release experiments
were interrupted prior to completion; see text for further discussion.
Data are means SD; n = 4. (B) Mapping microenvironment pH in
microparticles containing lysosensor yellow/blue dextran. A higher
I450 nm/I520 nm reflects a higher pH. (C) Quantitation of the average
I450/I520 in each microparticle type (data are means SD; n = 8); *
indicates p < 0.05.
(Note that the release experiment was not taken to completion
with the DCE and DCE/PLGA groups, but interrupted at 20
weeks. At that time a large amount of both types of
microparticles were still observed.) In comparison, at pH 7.4
PLGA microparticles exhibited burst release, released 50% of
estradiol in <1 week, and all the estradiol was released in 4
weeks, before degradation of PLGA microparticles was
complete. No oligomers of estradiol conjugates were detected
by HPLC in the release media from DCE microparticles. In
contrast, conjugates of drugs to polyester polymers (such as
PLGA, PLA) release the native forms of drugs over a relatively
short period (typically less than a month) and subsequently
release soluble carboxyl group-containing oligomer-drug
conjugates.21,22 At pH 5, DCE microparticles released 50% of
their estradiol in 3 days, indicating that their degradation was
acid sensitive (Figure 3A). For DCE microparticles, release at
pH 5 was stopped after a week when the point was clearly made
that degradation at pH 5 was much more rapid. Other
degradation byproducts of DCE microparticles were acetone
1
and CDM (confirmed by H NMR studies of release media at
pH 5.0; see Figure S6, which also shows release from DE
microparticles). This pH sensitivity may be useful in
applications such as intracellular and anticancer drug delivery.
Given the acid-catalyzed degradation of polyketals,3 we
hypothesized that drug release could be modulated by co-
incorporating PLGA, the degradation of which is known to
create a local acidic microenvironment.14 Microparticles with
50 wt % DCE and PLGA (50:50, ester terminated, Mw 24,000−
38,000) were prepared (abbreviated DCE/PLGA micro-
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
Experimental details and data (PDF)
C
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX