Cholesteryl-(L-lactic acid)n building blocks for self-assembling biomaterials

Article abstract of DOI:10.1021/ma010907x

The synthesis and properties of a novel set of building blocks for the preparation of self-assembling biomaterials are described. These molecules consist of a relatively short oligo(L-lactic acid)_n segment (n = 10-40) that is substituted with a

Full text of DOI:10.1021/ma010907x

746
Macromolecules 2002, 35, 746-759
Cholesteryl-(L-Lactic Acid)nj Building Blocks for Self-Assembling
Biomaterials
Ha r m -An ton Klok , J u lia J . Hw a n g, Su br a m a n i N. Iyer , a n d Sa m u el I. Stu p p *
Department of Materials Science and Engineering, Department of Chemistry, Medical School,
Northwestern University, Evanston, Illinois 60208
Received May 25, 2001
ABSTRACT: The synthesis and properties of a novel set of building blocks for the preparation of self-
assembling biomaterials are described. These molecules consist of a relatively short oligo(L-lactic acid)nj
segment (nj ) 10-40) that is substituted with a cholesterol moiety as an end group and in some cases
has a second biofunctional substituent at its other terminus. The cholesterol moiety not only serves to
induce liquid crystalline properties and drive the self-assembly of these oligomers, but also is expected
to have an effect on the interaction of cells with these materials. The cholesteryl-(L-lactic acid)nj (CLAnj)
oligomers form thermotropic liquid crystals and self-assemble into lamellar structures consisting of
interdigitated bilayers. In addition, the CLAnj oligomers can be homogeneously blended with poly(L-lactic
acid), thereby offering the possibility to improve the cell interaction properties of this common surgical
biomaterial. Furthermore, the secondary alcohol terminus of the CLAnj oligomers allows the opportunity
to introduce additional bioactive substituents such as cholesterol, indomethacin (an antiinflammatory
drug), and pyrene and rhodamine B (which can act as fluorescent labels for bioimaging purposes) and
various R-amino acids. These biofunctional CLAnj oligomers can also be homogeneously mixed with the
unsubstituted CLAnj oligomers and thus could enable a further noncovalent functionalization of self-
assembling biomaterials.
In tr od u ction
that cholesterol is important in the stability of lipid
rafts, important sites for membrane receptors.4 Thus,
cholesterol delivery to cells could be important in signal
transduction, cell adhesion with its biological conse-
quences, and cell migration. In addition, the cholesterol
moiety was selected because of its mesogenic character
known for many derivatives.5 It was anticipated that
the ability of cholesterol to form liquid crystalline phases
could provide a driving force for the self-assembly of the
molecules discussed here. A layered morphology gener-
ated by self-assembly would be an additional interesting
feature of biodegradable biomaterials since cells would
be exposed to many copies of the same chemistry.
Poly(L-lactic acid) (PLA) is a biocompatible and bio-
degradable material that is widely used in surgery and
tissue engineering research.6 We study here its chemical
modification to generate self-assembling and function-
alized materials. The functions considered here include,
an antiinflammatory drug, R-amino acids, and fluores-
cent dyes for bioimaging. First, we discuss the synthesis
and characterization of the cholesteryl-(L-lactic acid)nj
(CLAnj) oligomers. Then, we evaluate blends of the CLAnj
oligomers with poly(L-lactic acid). Finally, the synthesis
of end functionalized CLAnj oligomers is discussed and
also their blends with nonfunctionalized oligomers.
During the past decade, there has been an increasing
interest in bioactive as opposed to bioinert biomaterials
for human repair. The initial focus of this field was on
nontoxic, biocompatible materials that would be able to
replace or repair natural tissues. Current interest has
shifted to bioactive materials that can induce a specific
biological response. There is also great interest in the
preparation of suitable bioactive materials for biosensor
design. A number of reviews on these topics can be
found in the literature.1
Self-assembly is a powerful strategy for the prepara-
tion of novel materials with molecularly defined proper-
ties.2 From a biomaterials point of view such noncova-
lent strategies could allow precise control of the spatial
distribution of bioactive ligands, which could in turn
impact cell division, cell differentiation as well as the
synthesis of extracellular matrix. In this paper, we
describe the synthesis and properties of a set of bio-
functional building blocks that are designed to self-
assemble into layered structures. These molecules
consist of a cholesterol moiety, a key component of
eukaryotic cell membranes, attached to biodegradable
(L-lactic acid)nj oligomers. The secondary hydroxyl end
group of these cholesteryl-(L-lactic acid)nj (CLAnj) oligo-
mers can be used for further functionalization. The
cholesterol moiety was chosen because of its high
thermodynamic affinity for the cell membrane and its
ability to change the membrane’s permeability and
fluidity.3 These characteristics make cholesterol an
interesting component of bioactive biomaterials since it
could have universal effects regardless of cell type and
receptor map on the membrane. There is recent evidence
Resu lts a n d Discu ssion
Syn th esis. The molecules studied here are schemati-
cally described in Figure 1. The synthesis involves the
preparation of a CLAnj oligomer with a desired degree
of polymerization, followed by end functionalization of
the hydroxyl group.
Ch olester yl)(L-lactic acid)nj Oligom er s. The prepa-
ration of the CLAnj oligomers is outlined in Scheme 1.
Depending on the desired length of the oligo(L-lactic
acid) segment, the polymerizations were performed
either in toluene solution or in bulk. Solution polymer-
* To whom correspondence should be addressed at: 2225 N.
Campus Dr., Northwestern University, Evanston, IL 60208.
E-mail: s-stupp@nwu.edu.
10.1021/ma010907x CCC: $22.00 © 2002 American Chemical Society
Published on Web 01/04/2002

Macromolecules, Vol. 35, No. 3, 2002
Self-Assembling Biomaterials 747
F igu r e 1. Schematic representation of the cholesteryl-(L-lactic acid)nj oligomers that have been prepared and studied.
Sch em e 1
izations were initiated by the aluminumalkoxide gener-
ated in situ upon addition of 1 equiv of Al(Et)3 to 3 equiv
of cholesterol. As expected for a “living” polymerization,
the molecular weights could be controlled by the L-
lactide/cholesterol ratio, and oligomers were obtained
with polydispersities (Mw/Mn) of approximately 1.1.7
This method is particularly suited for the synthesis of
CLAnj oligomers with nj g 20, since products can be
easily isolated by precipitation of the reaction mixture
in excess methanol. Typically, these longer CLAnj oligo-
mers were isolated in 85-90% yield. However, yields
dropped significantly for lower molecular weight oligo-
mers with nj < 20, due to their increased solubility in
methanol. This is in agreement with observations
reported by Kricheldorf et al.,8 who prepared a series
of poly(L-lactic acid)s functionalized with a variety of
hormones, vitamins, and drugs using a similar proce-
dure.
Lower molecular weight CLAnj oligomers (nj < 20) were
synthesized in bulk at 150 °C using tin(II) bis(2-
ethylhexanoate) (Sn(Oct)2) as a catalyst.9 In the absence
of an alcohol as co-initiator, this procedure does not
allow accurate control of molecular weight. However,
in the presence of cholesterol and a catalytic amount of
Sn(Oct)2, narrow distribution oligomers can be prepared
with average molecular weights corresponding to the
L-lactide/alcohol ratio. Performing the polymerization in
bulk makes the workup easier and greatly improves the
yields. Although the Sn(Oct)2-catalyzed bulk-polymer-
ization of L-lactide is not strictly a “living” process, the
molecular weights can be adjusted by varying the
L-lactide/cholesterol ratio, and after workup, oligomers
with fairly narrow polydispersities are obtained (∼1.2).
1
The CLAnj oligomers were characterized by H NMR
spectroscopy and MALDI-TOF mass spectrometry. As
1
a representative example, the H NMR spectrum and
MALDI-TOF mass spectrum of a CLAnj oligomer that
was prepared via the solution method with an initial
L-lactide/cholesterol ratio of 10 are shown in Figure 2.
From the 1H NMR spectrum, the number-average
degree of polymerization of the L-lactid acid block can
be determined by comparing the integrals of the signals
corresponding to the L-lactid C-H protons (labeled b
and c in Figure 2a) with that of the olefinic proton of
the cholesterol moiety (labaelled a in Figure 2a). In this
way, a number-average degree of polymerization of 21
can be calculated. The MALDI-TOF mass spectrum
confirms the chemical composition of the proposed
structure. Since the masses of the cholesterol moiety
(386 Da) and of a L-lactic acid unit (72 Da) are known,
the molecular weight of a CLAnj oligomer at any given
value of n can be calculated. For example, an oligomer
with a number-average degree of polymerization of 20
has a molar mass of 1826 Da. The difference of 23 Da

748 Klok et al.
Macromolecules, Vol. 35, No. 3, 2002
1
F igu r e 2. (a) H NMR spectrum of a CLAnj oligomer with a targeted degree of polymerization (nj) of 20. Only peaks that are
relevant to the discussion of the spectrum in the text are labeled. The complete assignment of the 1H NMR spectrum can be
found in the experimental part. Peaks due to residual solvent in the sample are labeled with “×”. (b) MALDI-TOF mass spectrum
of a CLAnj oligomer with a targeted degree of polymerization (nj) of 20. A number of signals are labeled with the corresponding
masses.
between this number and the observed mass (1849 Da)
is due to Na+, which acts as a cationizing agent in the
MALDI process. Second, using the masses correspond-
ing to the individual peaks and their respective intensi-
ties, the number-average (Mn) and weight-average
molecular weights (Mw) of the CLAnj oligomers can be
calculated from the MALDI-TOF spectra.10 From the
mass spectrum shown in Figure 2b, values of Mh n and
Mh w of 2092 and 2147 Da can be determined, respec-
tively. This Mh n value corresponds to a number-average
degree of polymerization of the L-lactic acid units of 24.
This number is slightly higher, but close to the number-

Macromolecules, Vol. 35, No. 3, 2002
Self-Assembling Biomaterials 749
average degree of polymerization calculated from 1H
NMR (21). Generally, the results from 1H NMR spec-
troscopy and MALDI mass spectrometry were in good
agreement and differed at most by three to four units
in the calculated degree of polymerization. Finally, the
MALDI-TOF mass spectrum indicate that transesteri-
fication reactions occur during the polymerization of
L-lactide. The difference in mass between two neighbor-
ing peaks is 72 Da, which cannot be explained by a
simple ring-opening polymerization of L-lactide without
side reactions. In this case, since the monomer L-lactide
is composed of two L-lactic acid molecules, only polyester
chains containing an even number of repeat units would
be obtained and two neighboring peaks in the MALDI-
TOF mass spectrum would be spaced at a distance of
144 Da. The observed difference in mass of 72 Da can
only be explained by transesterification reactions, which
cause chain scrambling and result in the formation of
CLAnj oligomers that can also contain an odd number
of L-lactic acid repeat units. The occurrence of such
transesterification reactions during the polymerization
of L-lactide has been reported before.11
Th er m a l Beh a vior a n d Self-Or ga n iza tion . The
thermal behavior of the CLAnj oligomers was investi-
gated by differential scanning calorimetry (DSC) and
optical microscopy experiments. As a representative
example a DSC trace as well as an optical micrograph
of CLA10 (1a ) are shown in Figure 3, parts a and b,
respectively.
The DSC trace and optical micrograph shown in
Figure 3, parts a and b, indicate that 1a self-assembles
into a liquid crystalline phase. The DSC trace reveals
a glass transition temperature (Tg) at 32 °C and two
phase transitions at 57 and 88 °C. Between crossed
polarizers, the material displays a characteristic smec-
tic-like birefringent texture (Figure 2b) until it becomes
optically isotropic upon heating to approximately 88 °C.
Electron diffraction experiments indicate that under
ambient conditions the CLAnj oligomers can form smectic
liquid crystalline phases.12 Thus, the material exists in
a liquid crystalline glassy state below 32 °C, and it is
liquid crystalline up to 88 °C when isotropization occurs.
The phase transition observed by DSC at 57 °C must
involve a phase change between two different types of
liquid crystals. Increasing the length of the oligo(L-lactic
acid) segment results in an increase of both the glass-
transition temperature and the isotropization temper-
ature (Ti) (see Figure 3c). Tg and Ti, however, are still
below the Tg and the melting point of high molecular
weight poly(L-lactic acid) (PLA), which are 57 °C and
184 °C, respectively.13 In addition, increasing the length
of the (L-lactic acid) segment results in a narrowing of
the high-temperature liquid crystalline regime. The
observation of liquid crystalline behavior in 1a is of
particular interest since monodisperse (L-lactic acid)nj
oligomers with nj < 11 are known to be completely
amorphous.14 Thus, the results obtained for 1a clearly
demonstrate the ability of the cholesterol segment to
induce self-assembly in these oligomers.
F igu r e 3. (a) Differential scanning calorimetry scan (2nd
heating) of 1a . As is schematically indicated, the glass transi-
tion temperaure (Tg) was taken as the temperature at which
half of the change in heat capacity (∆Cp) has occurred. (b)
Optical micrograph between cross-polarizers (taken at room
temperature) of 1a . (c) Phase diagram comparing the thermal
behavior of the CLAnj oligomers (1a -1c) with that of poly(L-
lactic acid).
F igu r e 4. Small-angle X-ray scattering scans of 1a at 25 and
65 °C.
To obtain more information about the nature of the
liquid crystal to liquid crystal transition, small-angle
X-ray scattering (SAXS) experiments were performed
at different temperatures (SAXS scans of 1a at 25 and
65 °C are shown in Figure 4). In both cases the presence
of (001) and (002) reflections indicates a lamellar
organization of these molecules, and also confirms the
formation of a second smectic liquid crystalline phase
at elevated temperature. Increasing the temperature
results in a increase of the d spacing from 58 to 76 Å,
most probably due to a more extended conformation of
the lactic acid segment. Taking into account an esti-
mated extended length (L) of 43 Å for this oligomer (see
also Table 1),15 these data suggest that the molecules
organize into interdigitated bilayers.

750 Klok et al.
Macromolecules, Vol. 35, No. 3, 2002
Ta ble 1. d Sp a cin gs a n d Estim a ted Molecu la r Len gth s of
CLAnj Oligom er s
av no. of
d spacing
calcd
compound
repeats (nj)a
(Å)b
length, L (Å)c
1a
1b
1c
10
24
37
58
88
94
43
84
121
a
Average number of L-lactic acid repeat units calculated from
b
the MALDI-TOF mass spectrum. Layer spacing measured by
small-angle X-ray scattering. c Estimated extended length of aver-
age-sized molecules. These numbers were calculated using a length
of 14 Å for the cholesterol moiety (obtained from molecular
dynamics simulations (SYBYL)) and a length of 2.9 Å per L-lactic
acid repeat.15
F igu r e 6. Differential scanning calorimetry heating traces
of several blends of 1c and PLA.
F igu r e 7. Small-angle X-ray scattering scans recorded at 25
°C of several blends of 1c and PLA.
F igu r e 5. Small-angle X-ray scattering scans recorded at 25
°C of the different CLAnj oligomers: 1a (nj ) 10); 1b (nj ) 24);
1c (nj ) 37).
from 48.5 °C for pure 1c to 59.5 °C for the pure PLA.
The SAXS data presented in Figure 7 show that the
organization of 1c is not completely disrupted by the
addition of up to ∼50 wt % PLA. As was discussed
before, SAXS experiments with pure 1c indicate that
the molecules organize into interdigitated bilayers.
Upon the addition of 20 wt % PLA, the (001) reflection
shifts to a smaller angle, corresponding to a d spacing
of 100 Å. With 49 wt % PLA in the blend, the (001)
reflection is visible as a broad shoulder and indicates a
d spacing of ∼110 Å. The linear increase in d spacing
with increasing amounts of PLA in the blend suggests
that this material is incorporated in the L-lactic acid
domain of the bilayers, resulting in “swelling” of the
lamellae. Similar to the pure PLA, no SAXS signal could
be observed for blends containing ∼80 wt % or more
PLA. The wide-angle X-ray scattering scans (WAXS)
shown in Figure 8 indicate a high level of order on a
molecular scale for all the investigated 1c/PLA blends.
Throughout the range of investigated compositions,
clear WAXS peaks are observed, which do not signifi-
cantly broaden with increasing amounts of PLA. In
particular, the WAXS data show that even blends
containing more than 80 wt % PLA (for which no SAXS
signals could be observed) are not completely amor-
phous. This is understandable given the semicrystalline
nature of PLA.13
In Figure 5, the SAXS scans of 1a recorded at 25 °C
are compared with those of two longer oligomers 1b and
1c which contain on average 24 and 37 repeat units of
L-lactic acid, respectively. The corresponding d spacings
together with the estimated extended lengths of an
average-sized oligomer are listed in Table 1. Indepen-
dent of the length of the L-lactic acid segment, both an
(001) and an (002) reflection are observed, indicating a
lamellar organization of the CLAnj oligomers. The cor-
responding layer spacings gradually increase from 58
Å for 1a to 94 Å for 1c. Except for 1b, the observed layer
spacings significantly deviate from the estimated ex-
tended lengths, suggesting that the molecules organize
into interdigitated bilayers.
Blen d s of Oligom er s w ith P oly(L-la ctic a cid ).
Binary blends were prepared by solution mixing of PLA
with a certain amount of an CLAnj oligomer followed by
evaporation of the solvent, and were subsequently
investigated by means of DSC and small-angle as well
as wide-angle X-ray scattering experiments. DSC heat-
ing traces and SAXS diffraction scans are shown in
Figures 6 and 7, respectively, for several blends of PLA
and 1c.
The DSC traces in Figure 6 clearly show a single Tg
for all the investigated 1c/PLA mixtures, indicating
miscibility of both components of the blend.16 The Tg’s
of the blends are not significantly broadened in com-
parison with the pure components, and gradually shift
The miscibility of CLAnj oligomers with high molecular
weight PLA offers the possibility to modify the proper-
ties of biomaterials based on this polymer. Mixing a
small amount of a CLAnj oligomer with PLA prior to the
final processing step, or coating a thin layer of these

Macromolecules, Vol. 35, No. 3, 2002
Self-Assembling Biomaterials 751
are of interest for further functionalization reactions and
are obtained upon deprotection of 2, 3, 4, 7, and 15.
CLAnj oligomers substituted with a fluorescent dye (e.g.,
10 and 13) could be useful for fluorescent microscopy
studies, whereas the attachment of an antiinflammatory
drug such as indomethacin (5, 9) results in an oligomeric
prodrug.17 In the latter case, degradation of the L-lactic
acid spacer could result in a gradual release of indo-
methacin.
Most of the CLAnj derivatives depicted in Scheme 2
were prepared by direct esterification of the secondary
alcohol terminus of the oligomer with the corresponding
carboxylic acid. Most aromatic acids could be attached
under mild conditions using the diisopropylcarbodi-
imide/(dimethylamino)pyridinium 4-toluenesulfonate
(DIPC/DPTS) method.18 CLAnj oligomers 2 and 3 were
obtained by esterification of 4-tert-butoxycarbonylamino-
benzoic acid and 4-tert-butoxybenzoic acid, respectively,
which were prepared as reported previously.19 The end
functionalized CLAnj oligomers were characterized by
means of 1H NMR spectroscopy and MALDI-TOF mass
spectrometry. As a representative example, the 1H NMR
and mass spectra of 2 are shown in Figure 9. The good
agreement between the integrals of the doublets in the
aromatic part of the region (indicated with d and e in
Figure 9a) and the complex peak at ∼ 5.35 ppm caused
by the olefinic proton of the cholesterol moiety and the
C-H proton of the terminal L-lactic acid residue in the
1H NMR spectrum (labeled a and c) indicates a quan-
titative end functionalization of the CLAnj oligomer.
From the integrals of the signals at ∼5.15 ppm, which
F igu r e 8. Wide-angle X-ray scattering scans recorded at 25
°C of several blends of 1c and PLA.
oligomers on the surface of a preformed PLA-based
implant could modify cell response.12
Biofu n ction a l Ch olester yl-(L-La ctic a cid )nj Oli-
gom er s. Syn th esis. The secondary alcohol terminus
of the CLAnj oligomers can be used for further function-
alization of the molecules. Several of the CLAnj deriva-
tives that have been prepared are shown in Schemes 2,
6, and 7. The purpose of these functionalizations is
2-fold. First of all, the introduction of new end groups
could facilitate further substitution reactions (e.g., by
replacement of the secondary alcohol end group by a
primary amine). Second, the introduction of new end
groups could be useful to control biological response or
allow imaging of the oligomers in biological experiments.
CLAnj oligomers with aliphatic and aromatic amino end
groups, or phenol or carboxy groups at their chain end
Sch em e 2

752 Klok et al.
Macromolecules, Vol. 35, No. 3, 2002
1
F igu r e 9. (a) H NMR spectrum of the end functionalized CLAnj derivative 2. Only peaks that are relevant to the discussion of
1
the spectrum in the text are labeled. The complete assignment of the H NMR spectrum can be found in the experimental part.
Peaks due to residual solvent in the sample are labeled with “×”. (b) MALDI-TOF mass spectrum of the end functionalized CLAnj
derivative 2. A number of signals are labeled with the corresponding masses.
is due to the C-H protons of the L-lactic acid repeats
(marked b in Figure 9a) and the peak at 5.35 ppm a
number-average degree of polymerization of 27 can be
calculated. Unambiguous proof for the chemical integ-
rity of 2 was obtained from the MALDI-TOF mass
spectrum, which is shown in Figure 9b. In this mass
spectrum a series of peaks is labeled with the observed
masses. The mass calculated for a Na+-labeled oligomer
2 with a L-lactic acid segment composed of 21 repeat
units (2118 + 23 Da) perfectly matches the measured
mass in the mass spectrum, which underlines the
success of the end modification. The sodium ions were
added during sample preparation and act as cationizing
agents to facilitate the MALDI process.
Whereas the synthesis of 5 and 6 could be ac-
complished using the DIPC/DPTS method, this proce-

Macromolecules, Vol. 35, No. 3, 2002
Self-Assembling Biomaterials 753
Sch em e 3
Sch em e 4
dure was not found to be effective for the preparation
of the Rhodamine B functionalized oligomer 10. In this
case, the coupling of the fluorescent dye could be
performed in good yields using using 1-(3-dimethylami-
nopropyl)-3-ethylcarbodiimide-HCl (EDC)20 as the cou-
pling agent in the presence of 4-(dimethylamino)-
pyridine (DMAP) as the acylation catalyst.21,22 The
indomethacin and cholesterol-substituted aromatic acids
16 and 17, which were used for the preparation of CLAnj
oligomers 5 and 6, were synthesized as depicted in
Schemes 3 and 4. Indomethacin was first esterified with
benzyl-4-hydroxybenzoate using the DIPC/DPTS meth-
odology to give intermediate 18, which was subse-
quently hydrogenated to afford the final compound (16)
(Scheme 3). Cholesterol derivative 17 was prepared in
a single step by acylation of 4-aminobenzoic acid with
cholesterylchloroformate in the presence of triethy-
lamine (Scheme 4).
The esterification of aliphatic carboxylic acids with
the CLAnj oligomers was generally accomplished using
the EDC/DMAP methodology described above.20,21 The
carboxylic acid terminated oligomer 8 was obtained by
alcoholysis of succinic anhydride in the presence of
DMAP. Under EDC/DMAP conditions, indomethacin
could be coupled directly to the CLAnj oligomers, without
the need for a 4-hydroxybenzoic acid linker (compound
9). The protected L-aspartic acid derivative (21), which
was required for the preparation of oligomers 7 and 15,
was synthesized as shown in Scheme 5. The synthesis
first involved the preparation of N-(tert-butoxycarbonyl)-
OR,Oâ-di(9-fluorenylmethyl)-L-asparate (19), to protect
the carboxylic acid functionalities,23 followed by removal
of the amine protecting group under the action of
trifluoroacetic acid (CF3COOH)24 and aminolysis of
succinic anhydride to afford the desired carboxylic acid
(21).
It was anticipated that replacement of the secondary
alcohol terminus by a phenolic alcohol, a primary or
aromatic amine or a carboxylic acid would facilitate the
attachment of functional groups to the CLAnj oligomers.
As an example, the derivatization of the CLAnj-4-(tert-
butoxy)benzoate (3) and the CLAnj-N-(tert-butoxycar-
bonyl)glycinate oligomers (4) is shown in Schemes 6 and
7, respectively. The 4-hydroxybenzoic acid substituted
CLAnj oligomer 11, which was obtained after treatment
of 3 with a solution of CF3COOH in CH2Cl2, can be
easily reacted with various aromatic carboxylic acids
under mild DIPC/DPTS conditions.18 In this way,
1-pyrenecarboxylic acid could be successfully attached
to CLAnj oligomers (compound 13; Scheme 6).
Deprotection of the tert-butoxycarbonyl protected
oligomer 4 in a 2/1 (v/v) mixture of CH2Cl2 and CF3-
COOH afforded the corresponding ammonium trifluo-
roacetate (14), which was acylated with N-succinyl-
OR,Oâ-di(9-fluorenylmethyl)-L-aspartate (21) yielding
the L-aspartic acid substituted oligomer 15 (Scheme 7).
Blen d s w ith Ch olester yl)(L-La ctic a cid )nj. We
investigated the miscibility of CLAnj oligomers with the
functionalized derivatives, which could be useful in
tailoring biomaterials. The miscibility of CLA37-4-
(indomethacin)benzoate (5) and CLA37-4-(choles-
terylcarbamate)benzoate (6) with 1c was investigated
with mixtures containing 0-100 wt % of 5 or 6 and 1c
prepared in solution, followed by evaporation of solvent.
The homogeneity of the resulting blends was analyzed
by DSC, SAXS and WAXS experiments. Representative
DSC scans and X-ray diffraction scans for a series of
mixtures of 1c and 6 are shown in Figures 10-12,
respectively.
The DSC scans shown in Figure 10 reveal a single Tg
for all blend compositions, suggesting miscibility.16 With
increasing amounts of 6, the Tg gradually shifts from
49 °C for 1c to 64 °C for pure 6.
The SAXS scan of pure 6 shows both a (001) and an
(002) reflection, suggesting lamellar organization of the
oligomers. From these diffraction peaks, a d spacing of
∼115 Å can be calculated, which is significantly smaller
Among the structures presented in Scheme 2, the
4-(tert-butoxycarbonyl)aminobenzoic acid (2), the 4-(tert-
butoxy)benzoic acid (3), the N-(tert-butoxycarbonyl)-
glycine (4), and the succinic acid substituted CLAnj
oligomers (8) were used for further functionalization.

754 Klok et al.
Macromolecules, Vol. 35, No. 3, 2002
Sch em e 5
Sch em e 6
Sch em e 7
Å, possibly an intermolecular distance) for the blends
does not significantly broaden or shift to different
scattering angles in comparison with the pure com-
pounds.
than the estimated extended length of an average-sized
molecule in this material (∼141 Å).15 To fulfill these
geometric requirements, 6 is probably organized in an
interdigitated bilayer structure similar to 1c. The SAXS
scans of blends of 1c and 6 also show two orders of
reflection, demonstrating that they also possess the
lamellar order observed in their components with a
single periodicity, and also further suggesting molecular
miscibility. The corresponding d spacings of the blends
gradually increase from 94 Å for 1c to 115 Å for the
pure 6. WAXS scans of the blends also indicated a high
degree of order. With the exception of the blend that
contained 83 wt % 6, the wide angle diffraction peak (5
A series of blends of 5 and 1c were prepared and
analyzed as described above. The results of DSC, SAXS,
and WAXS experiments performed with these blends
are presented in Figures 13-15, respectively. Again, a
single Tg is observed for all the investigated blends,
suggesting that both components are miscible. With
increasing amounts of 5, the Tg of the blends gradually
shifts from 49 °C for the pure 1c to 61 °C for 5. No signal
could be observed in SAXS scans of pure 5, suggesting
very poor ordering in this material. Interestingly,
however, the addition of a small amount of 1c (∼11 wt

Macromolecules, Vol. 35, No. 3, 2002
Self-Assembling Biomaterials 755
F igu r e 13. Differential scanning calorimetry scans of a series
of mixtures of 1c and 5.
F igu r e 10. Differential scanning calorimetry scans for a
series of mixtures of 1c and 6.
F igu r e 14. Small-angle X-ray scattering scans recorded at
25 °C of a series of mixtures of 1c and 5.
F igu r e 11. Small-angle X-ray scattering scans recorded at
25 °C for a series of mixtures of 1c and 6.
F igu r e 15. Wide-angle X-ray scattering scans recorded at 25
°C of a series of mixtures of 1c and 5.
F igu r e 12. Wide-angle X-ray scattering scans recorded at 25
°C for a series of mixtures of 1c and 6.
induce self-assembly, but also make these systems
interesting biomaterials since cholesterol is ubiquitous
and critically needed in all mammalian cells. In the solid
state, the CLAnj oligomers and their blends with func-
tionalized derivatives self-assemble into interdigitated
bilayers of potential interest as temporary resorbable
scaffolds in tissue engineering applications. Further-
more, since the CLAnj oligomers can be homogeneously
mixed with PLA, these materials could be used to
modify the surfaces of polymeric scaffolds that are
already approved for human use. The CLAnj oligomers
can also be easily functionalized with bioactive drugs
or fluorescent labels to follow their fate in cells. Such
materials could very easily be used as blends with
functionalized derivatives given their molecular com-
patibility.
%) is sufficient to induce order in the blend (see Figure
14). The estimated average extended length of 5 is
approximately 138 Å, which is larger than the d spacing
obtained by SAXS, again indicating that the molecules
interdigitate to some extent. The WAXS scans shown
in Figure 15 confirm the amorphous nature of 5, but
interestingly local order within the layers is observed
even at very low concentrations of 1c. A possible
explanation here is the fact that steric disruption of
packing occurs in the indomethacin derivative. How-
ever, addition of small amounts of unsubstituted oligo-
mers allows the blend system to order locally.
Con clu sion s
The covalent attachment of a cholesterol moiety to low
molecular weight L-lactic acid oligomers with number-
average degrees of polymerization (nj) varying from 10
to 40 results in the formation of thermotropic liquid
crystalline materials. The cholesterol moieties not only
Exp er im en ta l Section
Gen er a l Da ta . Unless stated otherwise, all reagents and
solvents were of commercial grade and were used as received.
All reactions were performed under a nitrogen atmosphere.

756 Klok et al.
Macromolecules, Vol. 35, No. 3, 2002
Methylene chloride (CH2Cl2) and toluene were freshly distilled
from P2O5 and sodium/benzophenone, respectively, prior to use.
N,N′-Dimethylformamide (DMF) was stored over molecular
sieves (4 Å). Triethylamine (TEA) was stored over KOH.
Cholesterol and L-lactide were recrystallized from ethanol and
ethyl acetate, respectively, and vacuum-dried at room tem-
perature to constant weight. 4-(Dimethylamino)pyridinium
4-toluenesulfonate (DPTS) was prepared according to a lit-
erature procedure.18 Reactions were monitored by thin-layer
chromatography, using 0.250 mm precoated silicagel 60 F254
glass-plates (Merck). Column chromatography was performed
with Merck silica gel 60 (0.040-0.063 mm, 230-400 mesh,
60 Å).
(q, -(O(CdO)CH(CH3))n-, nH), 5.10 (q, cholesterol-O(CdO)-
CH(CH3)-, 1H), 4.65 (m, cholesterol C-3 H, 1H), 4.35 (q,
-O(CdO)CH(CH3)OH, 1H), 1.55 (d, -(O(CdO)CH(CH3))n-,
3nH), 0.65 (s, cholesterol C-18 H, 3H). nj (1H NMR) ) 11. MS
(MALDI-TOF): Mw ) 1938, Mn ) 1603, Mw/Mn ) 1.21. nj
(MALDI-TOF MS) ) 17.
Ch olester yl)(L-la ctic a cid )nj 4-(ter t-bu toxyca r bon yl)-
a m in oben zoa te (2). As a typical example, the substitution
of a cholesteryl-(L-lactic acid)24 oligomer will be described:
DIPC (0.25 mL, 1.60 mmol) was added to a solution of 1 (1.0
g, ∼0.48 mmol), 4-(N-tert-butoxycarbonyl)aminobenzoic acid
(0.28 g, 1.18 mmol), and DPTS (0.41 g, 1.39 mmol) in CH2Cl2
(25 mL). The reaction mixture was stirred at room-tempera-
ture overnight. After that, the reaction mixture was evapo-
rated to dryness. The residue was redissolved in CH2Cl2 (20
1H and 13C NMR spectra were recorded on Varian U400 or
U500 spectrometers at room temperature. Chemical shifts are
expressed in parts per million (δ) using residual protons in
the indicated solvents as internal standard. Mass spectrometry
was performed at the Mass-Spectrometry Laboratory of the
School of Chemical Sciences at the University of Illinois at
UrbanasChampaign. Low-resolution fast atom bombardment
mass-spectra (LR-FAB) were recorded on a Micromass ZAB-
SE mass-spectrometer. High-resolution fast atom bombard-
ment mass-spectrometry (HR-FAB) was performed on a
Micromass 70-SE-4F instrument. MALDI-TOF mass-spectra
were acquired on a PerSeptive Biosystems Voyager-DE STR
spectrometer. For the MALDI-TOF experiments, 3-indoly-
lacetic acid (IAA) was used as the matrix and sodium ions were
added to facilitate cationization. Elemental analysis was
performed in the Microanalysis Laboratory of the School of
Chemical Sciences of the University of Illinois at Urbanas
Champaign.
Small-angle X-ray scattering (SAXS) experiments were
carried out using a Bruker instrument with an Anton Paar
high-resolution small angle camera, a Hi-Star area detector,
a M18XHF22-SRA rotating anode generator, a camera dis-
tance of 630 nm, and Bruker software. Powder diffraction rings
were integrated over 360° to yield the diffraction pattern, and
the system was calibrated using a silver behenate standard.
The samples used for variable temperature studies were placed
in sealed 0.7 mm diameter capillary tubes. Differential scan-
ning calorimetry (DSC) experiments were performed using a
TA 2920 Modulated DSC instrument with a ramp speed of 10
°C/min.
Syn th esis. Ch olester yl-(L-La ctic a cid )nj (1). (i) Solu -
tion P olym er iza tion . As a typical example, the synthesis of
an oligomer with a targeted length of 20 lactic acid residues
will be described: a 1.9 M toluene solution of Al(Et)3 (3 mL,
5.7 mmol Al(Et)3) was added to a mixture of L-lactide (24.65
g, 171 mmol) and cholesterol (6.61 g, 17.1 mmol) in toluene
(120 mL). This mixture was stirred for 15 min at room
temperature and was then placed in a preheated oil bath at
80 °C. After 5 h, the polymerization was quenched with MeOH
(5 mL), and subsequently the warm reaction mixture was
precipitated in MeOH (1500 mL). Solids were filtered and
vacuum-dried at room temperature. Finally, the crude product
was passed over a short column (SiO2, CH2Cl2/MeOH 100/5
(v/v)). Yield: 16.8 g (54%). 1H NMR (CDCl3):25 δ ) 5.35 (d,
cholesterol C-6 H, 1H), 5.15 (q, -(O(CdO)CH(CH3))n-, nH),
5.10 (q, cholesterol-O(CdO)CH(CH3)-, 1H), 4.65 (m, choles-
terol C-3 H, 1H), 4.35 (q, -O(CdO)CH(CH3)OH, 1H), 1.55 (d,
-(O(CdO)CH(CH3))n-, 3nH), 0.65 (s, cholesterol C-18 H, 3H).
nj (1H NMR) ) 21. MS (MALDI-TOF): Mw ) 2147, Mn ) 2092,
Mw/Mn ) 1.03. nj (MALDI-TOF MS) ) 24.
mL) and precipitated in MeOH (400 mL). Solids were filtered
1
and vacuum-dried. Yield: 1.01 g (91%). H NMR (CDCl3):25
δ
) 7.75 (d, ArH, 2H), 6.65 (d, ArH, 2H), 6.70 (s, NH, 1H), 5.35
(m, cholesterol C-6 H and -O(CdO)CH(CH3)O(CdO)Ar-, 2H),
5.15 (q, -(O(CdO)CH(CH3))n-, nH), 5.10 (q, cholesterol-O(Cd
O)CH(CH3)-, 1H), 4.65 (m, cholesterol C-3 H, 1H), 1.55 (d,
-(O(CdO)CH(CH3))n-, 3nH), 1.40 (s, (CH3)3CO(CdO)-, 9H),
0.65 (s, cholesterol C-18 H, 3H). nj (1H NMR) ) 27. MS
(MALDI-TOF): Mw ) 2384, Mn ) 2345, Mw/Mn ) 1.02. nj
(MALDI-TOF MS) ) 24.
Ch olester yl)(L-La ctic a cid )nj N-(ter t-bu toxyca r bon yl)-
glycin a te (4). As a typical example the end-functionalization
of a cholesteryl-(L-lactic acid)25 oligomer will be described:
EDC (0.30 g, 1.56 mmol) was added to an ice-cooled solution
of 1 (1.00 g, ∼0.46 mmol), N-(tert-butoxycarbonyl)glycine (0.26
g, 1.48 mmol) and DMAP (0.10 g, 0.82 mmol) in CH2Cl2 (20
mL). The reaction mixture was stirred at 0 °C for 4 h. Then,
the ice bath was removed, and stirring was continued for
another 4 h at room temperature. The reaction mixture was
evaporated to dryness, the residue redissolved in CH2Cl2 (5
mL) and precipitated in MeOH (100 mL). Solids were filtered
and vacuum-dried at room temperature. Yield: 0.90 g (84%).
1H NMR (CDCl3):25 δ ) 5.35 (d, cholesterol C-6 H, 1H), 5.15
(q, -(O(CdO)CH(CH3))n-, nH), 5.10 (q, cholesterol-O(CdO)-
CH(CH3)-, 1H), 5.00 (b, -CH(CH3)O(CdO)CH2-, 1H), 4.65
(m, cholesterol C-3 H, 1H), 4.00 (m, -O(CdO)CH2NH-, 2H),
1.55 (d, -(O(CdO)CH(CH3))n-, 3nH), 1.45 (s, (CH3)3CO(Cd
O)-, 9H), 0.65 (s, cholesterol C-18 H, 3H). nj (1H NMR) ) 27.
MS (MALDI-TOF): Mw ) 2687, Mn ) 2380, Mw/Mn ) 1.13. nj
(MALDI-TOF MS) ) 25.
Ben zyl 4-(In d om eth a cin )ben zoa te (18). DIPC (1.0 mL,
6.40 mmol) was added to a solution of indomethacin (1.0 g,
2.80 mmol), DPTS (1.24 g, 4.21 mmol) and benzyl-4-hydroxy-
benzoate (0.77 g, 3.37 mmol) in DMF (30 mL). The solution
was stirred overnight at room temperature. Then, the reaction
mixture was poured into water (250 mL). The aqueous mixture
was extracted with CH2Cl2 (3×), and the combined CH2Cl2
extracts were washed with water (1×) and brine (1×). The
organic phase was separated, dried over MgSO4, filtered, and
evaporated to dryness. The oily residue was triturated with
MeOH. Solids were filtered and vacuum-dried. Yield: 1.24 g
(79%). 1H NMR (CDCl3): δ ) 8.10 (d, ArH(CdO)O-, 2H), 7.68
(d, ArH(CdO)n-, 2H), 7.48 (d, ArHCl, 2H), 7.40 (m, ArHbenzyl,
5H), 7.15 (d, ArHO(CdO)-, 2H), 7.04 (d, ArH, 1H), 6.88 (d,
ArH, 1H), 6.70 (dd, ArH, 1H), 5.35 (s, ArCH2O-, 2H), 3.95 (s,
-CH2(CdO)O-, 2H), 3.85 (s, ArOCH3, 3H), 2.45 (s, -CH3, 3H).
MS (LR-FAB) m/z ) 568.2 (MH+). MS (HR-FAB) m/z )
567.144800 (calculated 567.144866) for C33H26NO6Cl.
(ii) Bu lk P olym er iza tion . As a typical example the
synthesis of an oligomer with a targeted length of 10 lactic
acid residues will be described: A mixture of L-lactide (10.19
g, 71 mmol) and cholesterol (5.44 g, 14 mmol) was placed in a
preheated oil bath at 150 °C and stirred until everything was
molten. Then, a solution of Sn(oct)2 in toluene (1 mL, 0.37 gr
Sn(oct)2/mL) was added, and the reaction mixture was stirred
at 150 °C. After 5 h, the reaction mixture was allowed to cool
to room temperature, and the residual solid was triturated
with MeOH (100 mL). Solids were collected by filtration and
finally vacuum-dried at room temperature. Yield: 7.82 g (50%).
1H NMR (CDCl3):25 δ ) 5.35 (d, cholesterol C-6 H, 1H), 5.15
4-(In d om eth a cin )ben zoic a cid (16). Pd-C (0.47 g, 50 wt
% H2O) was added to a solution of 18 (1.21 g, 2.13 mmol) in a
mixture of EtAc (80 mL) and EtOH (40 mL), and the reaction
mixture was stirred overnight at room temperature under a
hydrogen atmosphere. Then, the reaction mixture was filtered
over a short plug of Celite and evaporated to dryness, affording
1
the free acid in quantitative yield. H NMR (DMSO-d6): δ )
7.98 (d, ArH(CdO)O-, 2H), 7.65 (m, -N(CdO)ArHCl, 4H),
7.25 (d, ArHO(CdO)-, 2H), 7.15 (d, ArH, 1H), 6.95 (d, ArH,
1H), 6.72 (dd, ArH, 1H), 4.10 (s, -CH2(CdO)O-, 2H), 3.75 (s,
ArOCH3, 3H), 2.25 (s, -CH3, 3H). MS (LR-FAB) m/z ) 478.2
(MH+).

Macromolecules, Vol. 35, No. 3, 2002
Self-Assembling Biomaterials 757
Ch olester yl)(L-La ctic a cid )nj 4-(In d om eth a cin )ben -
zoa te (5). As a typical example, the functionalization of a
cholesteryl-(L-lactic acid)34 oligomer will be described: DIPC
(0.11 mL, 0.70 mmol) was added dropwise to a solution of 1
(1.87 g, ∼0.66 mmol) 16 (0.38 g, 0.80 mmol) and DPTS (0.20
g, 0.68 mmol) in CH2Cl2 (10 mL). After overnight stirring at
room temperature, the reaction mixture was evaporated to
dryness. The residue was redissolved in a minimal amount
CH2Cl2 and precipitated in MeOH. Solids were filtered and
4H), 4.20 (m, CHFmCH2O-, 2H), 2.70 (m, -CHCH2(CdO)O-,
2H), 1.35 (s, (CH3)3CO(CdO)-, 9H). 13C NMR (DMSO-d6): δ
) 171.19, 170.04, 155.30, 143.51, 140.72, 127.74, 127.15,
125.22, 120.14, 78.56, 66.48, 66.01, 50.03, 46.26, 46.12, 35.53,
28.09. MS (LR-FAB) m/z ) 590.2 (MH+). MS (HR-FAB) m/z
) 590.254300 (calculated 590.254263) for C37H36NO6.
Or,Oâ-Di(9-flu or en ylm eth yl)-L-a sp a r ta te Tr iflu or oa ce-
tic Acid Sa lt (20). TFA (30 mL) was added to an ice-cooled
solution of 19 (2.93 g, 4.97 mmol) in CH2Cl2 (60 mL). The
reaction mixture was stirred at 0 °C for 1 h, and at room
temperature for another hour. Then, the reaction mixture was
evaporated to dryness, and the residue triturated with Et2O.
Solids were filtered and vacuum-dried at room temperature.
vacuum-dried. Yield: 1.90 g (88%). 1H NMR (DMSO-d6):25
δ
) 8.08 (d, ArH(CdO)O-, 2H), 7.70 (m, -N(CdO)ArHCl, 4H),
7.35 (d, ArHO(CdO)-, 2H), 7.22 (d, ArH, 1H), 6.98 (d, ArH,
1H), 6.78 (dd, ArH, 1H), 5.40 (d, cholesterol C-6 H and -O(Cd
O)CH(CH3)O(CdO)Ar-, 2H), 5.25 (q, -(O(CdO)CH(CH3))n-,
nH), 5.10 (q, cholesterol-O(CdO)CH(CH3)-, 1H), 4.55 (m,
cholesterol C-3 H, 1H), 4.15 (s, -CH2(CdO)O-, 2H), 3.80 (s,
ArOCH3, 3H), 2.25 (s, -CH3, 3H), 1.55 (d, -(O(CdO)CH-
(CH3))n-, 3nH), 0.65 (s, cholesterol C-18 H, 3H). nj (1H NMR)
1
Yield: 2.82 g (94%). H NMR (DMSO-d6): δ ) 8.70 (s, -NH3,
3H), 7.85 (m, ArH, 4H), 7.60 (m, ArH, 4H), 7.30 (m, ArH, 8H),
4.35 (m, -CHNH(CdO)- and FmCH2O- and CHFmCH2O-,
7H), 2.85 (m, -CHCH2(CdO)O-, 2H). 13C NMR (DMSO-d6):
δ ) 169.19, 168.41, 143.29, 140.75, 127.85, 127.22, 125.12,
120.16, 67.36, 66.49, 48.37, 46.06, 46.00, 33.97. MS (LR-FAB)
m/z ) 490.2 (M - CF3COO-). MS (HR-FAB) m/z ) 490.201800
(calculated 490.201834) for C32H28NO4.
) 36. MS (MALDI-TOF): Mw ) 3598, Mn ) 3278, Mw/Mn
1.10. nj (MALDI-TOF MS) ) 34.
)
Ben zoic Acid -4-ch olester ylca r ba m a te (17). A solution
of cholesteryl chloroformate (1.0 g, 2.23 mmol) in CH2Cl2 (10
mL) was added to a mixture of 4-aminobenzoic acid (0.306 g,
2.23 mmol) and triethylamine (0.31 mL, 2.23 mmol) in CH2-
Cl2 (20 mL). Then, TEA (0.31 mL, 2.23 mmol) was added
dropwise to the reaction mixture, which was subsequently
stirred overnight at room temperature. After that, the reaction
mixture was washed with an aqueous KHSO4 solution (1×),
water (1×), and brine (1×). The organic phase was separated,
dried over MgSO4, filtered, and evaporated to dryness. The
residue was triturated with MeOH. Solids were filtered and
vacuum-dried. Yield: 0.92 g (74%). 1H NMR (CDCl3):25 δ )
7.85 (d, ArH(CdO)OH, 2H), 6.65 (d, ArHNH(CdO)-, 2H), 5.40
(d, cholesterol C-6 H, 1H), 4.65 (m, cholesterol C-3, 1H), 0.65
(s, cholesterol C-18, 3H). 13C NMR (CDCl3): δ ) 161.09, 149.19,
138.91, 132.99, 123.34, 114.16, 79.64, 56.62, 56.08, 49.91,
42.28, 39.66, 39.49, 37.64, 36.77, 36.51, 36.15, 35.77, 31.87,
31.80, 28.21, 28.00, 27.40, 24.26, 23.80, 22.81, 22.55, 21.02,
19.38, 19.26, 18.70, 11.84. MS (LR-FAB) m/z ) 550.4 (MH+).
Ch olester yl)(L-La ctic a cid )nj 4-(Ch olester ylca r ba m -
a te)ben zoa te (6). As a typical example, the functionalization
of a cholesteryl-(L-lactic acid)34 oligomer will be described:
DIPC (20 µL, 0.13 mmol) was added dropwise to a solution of
1 (0.215 g, ∼0.076 mmol) 17 (0.051 g, 0.093 mmol) and DPTS
(0.023 g, 0.078 mmol) in CH2Cl2 (25 mL). After overnight
stirring at room temperature, the reaction mixture was
evaporated to dryness. The residue was redissolved in a
minimal amount of CH2Cl2 and precipitated in MeOH. Solids
were filtered and vacuum-dried. Yield: 0.21 g (83%). 1H NMR
(CDCl3):25 δ ) 8.00 (d, ArH(CdO)O-, 2H), 7.45 (d, ArHNH-
(CdO)-, 2H), 5.40 (d, cholesterol′ C-6 H, 1H), 5.35 (d,
cholesterol C-6 H, 1H), 5.30 (q, -O(CdO)CH(CH3)O(CdO)Ar,
1H), 5.15 (q, -(OC(dO)CH(CH3))n-, nH), 5.10 (q, cholesterol-
OC(dO)CH(CH3)-, 1H), 4.65 (m, cholesterol C-3 H and
cholesterol′ C-3 H, 2 × 1H), 1.55 (d, -(OC(dO)CH(CH3))n-,
3nH), 0.65 (ds, cholesterol C-18 and cholesterol′ C-18 H, 2 ×
3H). nj (1H NMR) ) 28. MS (MALDI-TOF): Mw ) 3483, Mn )
3119, Mw/Mn ) 1.12. nj (MALDI-TOF MS) ) 30.
N-su ccin yl-Or,Oâ-di(9-flu or en ylm eth yl)-L-aspar tate (21).
N,N-Diisopropylethylamine (0.60 mL, 3.44 mmol) was added
dropwise to a mixture of 20 (1.00 g, 1.66 mmol) and succinic
anhydride (0.165 g, 1.65 mmol) in CH2Cl2 (60 mL). The
reaction mixture was stirred at room temperature for 15 h.
Then, the reaction mixture was evaporated to dryness, and
the crude product was purified by column chromatography
(SiO2, CH2Cl2/MeOH 100/10 (v/v)). Yield: 0.82 g (84%). 1H
NMR (DMSO-d6): δ ) 8.45 (d, NH, 1H), 7.86 (m, ArH, 4H),
7.64 (m, ArH, 4H), 7.40 (m, ArH, 4H), 7.30 (m, ArH, 4H), 4.65
(m, -CHNH(CdO)-, 1H), 4.35 (m, FmCH2O-, 4H), 4.25 (m,
CHFmCH2O-, 2H), 2.65 (m, -CHCH2(CdO)O-, 2H), 2.35 (m,
-NH(CdO)CH2CH2(CdO)OH, 4H). 13C NMR (DMSO-d6): δ
) 173.79, 171.27, 170.77, 169.98, 143.57, 140.75, 127.78,
127.19, 125.16, 120.14, 66.45, 66.01, 48.53, 46.17, 46.12, 35.51,
29.82, 29.08. MS (LR-FAB) m/z ) 590.3 (MH+). MS (HR-
FAB) m/z ) 590.217900 (calculated 590.217878) for C36H32
-
NO7.
Ch olest er yl)(L-La ct ic a cid )nj Or,Oâ-d i(9-flu or en yl-
m eth yl)-L-a sp a r ta te (7). As a typical example, the function-
alization of a cholesteryl-(L-lactic acid)25 oligomer will be
described: EDC (0.040 g, 0.21 mmol) was added to an
ice-cooled mixture of 1 (0.161 g, ∼0.073 mmol), 21 (0.100 g,
0.170 mmol), and DMAP (0.007 g, 0.057 mmol) in CH2Cl2 (12
mL) and DMF (2 mL). The reaction mixture was stirred at 0
°C for 3.5 h. Then, the ice bath was removed, and the reaction
mixture was stirred at room temperature for another 5 h. After
that, the reaction mixture was diluted with CH2Cl2 and
washed with water (1×) and brine (1×). The organic phase
was separated, dried over MgSO4, filtered, and evaporated to
dryness. The crude product was purified by column chroma-
tography (SiO2, CH2Cl2/MeOH 100/5 (v/v)). Yield: 0.19 g (94%).
1H NMR (CDCl3):25 δ ) 7.75 (m, ArH, 4H), 7.50 (m, ArH, 4H),
7.38 (m, ArH, 4H), 7.25 (m, ArH, 4H), 6.28 (d, NH, 1H), 5.35
(d, cholesterol C-6 H, 1H), 5.15 (q, -(O(CdO)CH(CH3))n-, nH),
5.10 (q, cholesterol-O(CdO)CH(CH3)-, 1H), 4.85 (m, -CH-
(CH3)O(CdO)CH2-, 1H), 4.65 (m, cholesterol C-3 H, 1H), 4.40
(br m, -CHNH(CdO)- and FmCH2O-, 5H), 4.10 (br m,
CHFmCH2O-, 2H), 2.95 (m, -CHCH2(CdO)O-, 2H), 2.50 (m,
-NH(CdO)CH2CH2(CdO)NH-, 4H), 1.55 (d, -(O(CdO)CH-
(CH3))n-, 3nH), 0.65 (s, cholesterol C-18 H, 3H). nj (1H NMR)
N-(ter t-Bu toxycar bon yl)-Or,Oâ-di(9-flu or en ylm eth yl) L-
Asp a r ta te (19). DCC (5.47 g, 26.51 mmol) was added in three
portions to an ice-cooled mixture of N-tert-BOC-L-Asp (2.00 g,
8.58 mmol), 9-fluorenylmethanol (5.00 g, 25.5 mmol) and
DMAP (0.17 g, 1.39 mmol) in CH2Cl2 (100 mL). After the
addition of the DCC was completed, the reaction mixture was
stirred at 0 °C for another hour. Then, the ice bath was
removed, and the reaction mixture was stirred overnight at
room temperature. The precipitated dicyclohexylurea was
removed by filtration, and the filtrate was successively washed
with 5% NaHCO3 (1×), H2O (1×), and brine (1×). The organic
phase was separated, dried over MgSO4, filtered, and evapo-
rated to dryness. The crude product was purified by column
chromatography (SiO2, CH2Cl2/MeOH 100/1 (v/v)). Yield: 4.02
g (79%). 1H NMR (DMSO-d6): δ ) 7.85 (m, ArH, 4H), 7.65
(m, ArH, 4H), 7.45 (d, NH, 1H), 7.40 (m, ArH, 4H), 7.30 (m,
ArH, 4H), 4.55 (m, -CHNH(CdO)-, 1H), 4.35 (m, FmCH2O-,
) 22. MS (MALDI-TOF): Mw ) 2767, Mn ) 2646, Mw/Mn
1.05. nj (MALDI-TOF MS) ) 23.
)
Ch olester yl)(L-La ctic a cid )nj Su ccin a te (8). As a typical
example, the modification of a cholesteryl-(L-lactic acid)25
oligomer will be described: DMAP (0.139 g, 1.14 mmol) was
added to a solution of 1 (0.103 g, ∼0.047 mmol) and succinic
anhydride (0.019 g, 0.19 mmol) in CH2Cl2 (10 mL). After
overnight stirring at room temperature, the reaction mixture
was evaporated to dryness, and the crude product was purified
by column chromatography (SiO2, CH2Cl2/MeOH 100/10 (v/
v)). Yield: 0.080 g (74%). 1H NMR (CDCl3):25 δ ) 5.35 (d,
cholesterol C-6 H, 1H), 5.15 (q, -(O(CdO)CH(CH3))n-, nH),
5.10 (q, cholesterol-O(CdO)CH(CH3)-, 1H), 4.65 (m, choles-

758 Klok et al.
Macromolecules, Vol. 35, No. 3, 2002
terol C-3 H, 1H), 2.70 (m, -O(CdO)CH2CH2(CdO)OH, 4H),
1.55 (d, -(O(CdO)CH(CH3))n-, 3nH), 0.65 (s, cholesterol C-18
H, 3H). nj (1H NMR) ) 21. MS (MALDI-TOF): Mw ) 2030,
Mn ) 1784, Mw/Mn ) 1.14. nj (MALDI-TOF MS) ) 18.
Ch olester yl)(L-La ctic a cid )nj)In d om eth a cin (9). As a
typical example, the modification of a cholesteryl-(L-lactic
acid)25 oligomer will be described: EDC (0.096 g, 0.50 mmol)
was added to an ice-cooled solution of 1 (0.507 g, ∼0.23 mmol),
indomethacin (0.162 g, 0.45 mmol), and DMAP (0.038 g, 0.31
mmol) in CH2Cl2 (60 mL). The reaction mixture was stirred
overnight in the ice bath, which gradually melted during the
course of the reaction. After that, the reaction mixture was
evaporated to dryness and the residue redissolved in CH2Cl2
(5 mL) and precipitated in MeOH (100 mL). Solids were
filtered and vacuum-dried at room temperature. Yield: 0.49
modification of an oligomer with an average of 25 repeat units
of L-lactic acid will be described: TFA (10 mL) was added to
an ice-cooled solution of 4 (0.86 g, ∼0.36 mmol) in CH2Cl2 (20
mL). The reaction mixture was stirred at 0 °C for 1 h and at
room temperature for another hour. Then, the reaction mixture
was evaporated to dryness, and the residue was triturated with
n-hexane. Solids were filtered and vacuum-dried at room
temperature. Yield: 0.80 g (93%). 1H NMR (CDCl3):25 δ ) 5.35
(d, cholesterol C-6 H, 1H), 5.15 (q, -(O(CdO)CH(CH3))n- and
-CH(CH3)O(CdO)CH2-, (n + 1)H), 5.10 (q, cholesterol-O(Cd
O)CH(CH3)-, 1H), 4.65 (m, cholesterol C-3 H, 1H), 4.00 (br,
-O(CdO)CH2NH-, 2H), 1.55 (d, -(O(CdO)CH(CH3))n-, 3nH),
0.65 (s, cholesterol C-18 H, 3H). nj (1H NMR) ) 26. MS
(MALDI-TOF): Mw ) 2313, Mn ) 2173, Mw/Mn ) 1.06. nj
(MALDI-TOF MS) ) 23.
1
g (84%). H NMR (CDCl3):25 δ ) 7.65 (d, -N(CdO)ArH, 2H),
Ch olester yl)(L-la ctic a cid )nj)Glycin e-Or,Oâ-d i(9-flu o-
r en ylm eth yl)-L-a sp a r ta te (15). As a typical example the
endfunctionalization of a cholesteryl-(L-lactic acid)23-Gly‚
TFA oligomer will be described: EDC (0.025 g, 0.130 mmol)
was added to an ice-cooled solution of 14 (0.109 g, ∼0.046
mmol), 21 (0.061 g, 0.103 mmol), and DMAP (0.009 g, 0.074
mmol) in CH2Cl2 (20 mL). The reaction mixture was stirred
at 0 °C for 3 h. Then, the reaction mixture was diluted with
CH2Cl2 and washed with water (1×) and brine (1×). The
organic phase was separated, dried over MgSO4, and evapo-
rated to dryness. The crude product was purified by column
chromatography (SiO2, CH2Cl2/MeOH 100/5 (v/v)). Yield: 0.09
7.45 (d, ArHCl, 2H), 6.96 (d, Ar′H, 1H), 6.88 (d, Ar′H, 1H),
6.65 (dd, Ar′H, 1H), 5.35 (d, cholesterol C-6 H, 1H), 5.15 (q,
-(O(CdO)CH(CH3))n- and O(CdO)CH(CH3)O(CdO)CH2-, (n
+ 1)H), 5.10 (q, cholesterol-O(CdO)CH(CH3)-, 1H), 4.65 (m,
cholesterol C-3 H, 1H), 3.85 (s, ArOCH3, 3H), 3.75 (s, -CH2-
(CdO)-, 2H), 2.35 (s, -CH3, 3H), 1.55 (d, -(O(CdO)CH-
(CH3))n-, 3nH), 0.65 (s, cholesterol C-18 H, 3H). nj (1H NMR)
) 25. MS (MALDI-TOF): Mw ) 2551, Mn ) 2429, Mw/Mn
1.05. nj (MALDI-TOF MS) ) 23.
)
Ch olester yl)(L-la ctic a cid )nj)Rh od a m in e B (10). As a
typical example, the modification of a cholesteryl-(L-lactic
acid)25 oligomer will be described: EDC (0.096 g, 0.50 mmol)
was added to an ice-cooled solution of 1 (0.52 g, ∼0.24 mmol),
Rhodamine B (0.21 g, 0.44 mmol), and DMAP (0.035 g, 0.29
mmol) in CH2Cl2 (60 mL). The reaction mixture was stirred
overnight in the ice bath, which gradually melted during the
course of the reaction. After that, the reaction mixture was
evaporated to dryness, redissolved in CH2Cl2 (5 mL), and
precipitated in MeOH (100 mL). Solids were filtered, vacuum-
dried, and purified by column chromatography (SiO2, CH2Cl2/
MeOH 100/10 (v/v)). Yield (i): 0.050 g. The MeOH phase was
evaporated and subsequently purified by column chromatog-
raphy (SiO2, CH2Cl2/MeOH 100/10 (v/v)) to provide a second
batch of the product. Yield (ii): 0.060 g. Total yield (i + ii):
1
g (69%). H NMR (CDCl3):25 δ ) 7.75 (m, ArH, 4H), 7.50 (m,
ArH, 4H), 7.35 (m, ArH, 4H), 7.25 (m, ArH, 4H), 6.48 (d, NH,
1H), 6.38 (t, NH, 1H), 5.35 (d, cholesterol C-6 H, 1H), 5.15 (q,
-(O(CdO)CH(CH3))n-, nH), 5.10 (q, cholesterol-O(CdO)-
CH(CH3)-, 1H), 4.85 (m, -CH(CH3)O(CdO)CH2-, 1H), 4.65
(m, cholesterol C-3 H, 1H), 4.40 (br m, -CHNH(CdO)- and
FmCH2O-, 5H), 4.10 (br m, CHFmCH2O- and -O(CdO)CH2-
NH-, 4H), 2.95 (m, -CHCH2(CdO)O-, 2H), 2.50 (m, -NH-
(CdO)CH2CH2(CdO)NH-, 4H), 1.55 (d, -(O(CdO)CH(CH3))n-,
3nH), 0.65 (s, cholesterol C-18 H, 3H). nj (1H NMR) ) 23. MS
(MALDI-TOF): Mw ) 2806, Mn ) 2644, Mw/Mn ) 1.06. nj
(MALDI-TOF MS) ) 23.
1
0.11 g (17%). H NMR (CDCl3):25 δ ) 8.35 (d, ArH, 1H), 7.80
Ack n ow led gm en t. This research was partially sup-
ported by a grant from Boehringer Mannheim (India-
napolis, IN). H.-A.K. gratefully acknowledges The Neth-
erlands Organization for Scientific Research for a
postdoctoral fellowship.
(m, ArH, 1H), 7.70 (m, ArH, 1H), 7.30 (d, ArH, 1H), 7.05 (m,
ArH, 2H), 6.90 (d, ArH, 1H), 6.80 (m, ArH, 3H), 5.35 (d,
cholesterol C-6 H, 1H), 5.15 (q, -(O(CdO)CH(CH3))n- and
O(CdO)CH(CH3)O(CdO)Ar-, (n + 1)H), 5.10 (q, cholesterol-
O(CdO)CH(CH3)-, 1H), 4.65 (m, cholesterol C-3 H, 1H), 3.60
(q, -NCH2CH3, 8H), 1.55 (d, -(O(CdO)CH(CH3))n-, 3nH),
1.30 (t, -NCH2CH3, 12H), 0.65 (s, cholesterol C-18 H, 3H). nj
Refer en ces a n d Notes
(i) (1H NMR) ) 24. MS (i) (MALDI-TOF): Mw ) 2740, Mn
)
(1) For some recent references, see, e.g.: (a) Knoll, W.; Mat-
suzawa, M.; Offenha¨usser, A.; Ru¨he, J . Isr. J . Chem. 1996,
36, 357-369. (b) Ra¨dler, J .; Sackmann, E. Curr. Opin. Solid
State Mater. Sci. 1997, 2, 330-336. (c) Cook, A. D.; Hrkach,
J . S.; Gao, N. N.; J ohnson, I. M.; Pajvani, U. B.; Cannizzaro,
S. M.; Langer, R. J . Biomed. Mater. Res. 1997, 35, 513-523.
(d) Chen, C. S.; Mrksich, M.; Huang, S.; Whitesides, G. M.;
Ingber, D. E. Biotechnol. Prog. 1998, 14, 356-363. (e)
Hubbell, J . A. Curr. Opin. Solid State Mater. Sci. 1998, 3,
246-254. (f) Hench, L. L. Mater. Sci. Forum 1999, 293, 37-
63.
(2) See, e.g.: (a) Lowe, C. R. Curr. Opin. Struct. Biol. 2000, 10,
428-434. (b) Stupp, S. I.; Pralle, M. U.; Tew, G. N.; Li, L.
M.; Sayar, M.; Zubarev, E. R. MRS Bull. 2000, 25, 42-48.
(c) Emrick, T.; Fre´chet, J . M. J . Curr. Opin. Colloid Interface
Sci. 1999, 4, 15-23. (d) Sijbesma, R. P.; Meijer, E. W. Curr.
Opin. Colloid Interface Sci. 1999, 4, 24-32. (e) Ciferri, A.
Supramolecular Polymers; Marcel Dekker: New York 2000.
(3) See, e.g.: (a) Yeagle, P. L. Biochim. Biophys. Acta 1985, 822,
267-287. (b) Yeagle, P. L. Biochimie 1991, 73, 1303-1310.
(c) Alberts, B.; Bray, D.; Lewis, J .; Raff, M.; Roberts, K.;
Watson, J . D. Molecular Biology of the Cell, 2nd ed.; Garland
Publishing Inc.: New York, 1989.
2652, Mw/Mn ) 1.03. nj (i) (MALDI-TOF MS) ) 25. nj (ii) (1H
NMR) ) 22. MS (ii) (MALDI-TOF): Mw ) 2185, Mn ) 2113,
Mw/Mn ) 1.03. nj (ii) (MALDI-TOF MS) ) 17.
Ch olester yl)(L-La ctic a cid )nj 4-(1-P yr en eca r boxyla te)-
ben zoa te (13). As a typical example, the end-functionalization
of a cholesteryl-(L-lactic acid)25-4-(hydroxy)benzoate oligo-
mer will be described: DIPC (50 µL, 0.32 mmol) was added
dropwise to a mixture of 11 (0.103 g, ∼0.045 mmol), 1-pyren-
ecarboxylic acid (0.025 g, 0.10 mmol), and DPTS (0.047 g, 0.16
mmol) in CH2Cl2 (15 mL). After overnight stirring at room
temperature, the reaction mixture was evaporated to dryness.
The residue was redissolved in CH2Cl2 (5 mL) and precipitated
in MeOH (100 mL). Solids were filtered and vacuum-dried.
The crude product was purified by column chromatography
(SiO2, CH2Cl2/MeOH 100/2.5 (v/v)). Yield: 0.075 g (66%). 1H
NMR (CDCl3):25 δ ) 9.38 (d, pyreneArH, 1H), 8.90 (d, pyreneArH,
1H), 8.28 (m, ArH(CdO)O- and pyreneArH, 7H), 8.12 (m, pyrene
-
ArH, 2H), 7.45 (d, ArHO(CdO)-, 2H), 5.40 (q, -O(CdO)-
CH(CH3)O(CdO)Ar-, 1H), 5.35 (d, cholesterol C-6 H, 1H), 5.15
(q, -(O(CdO)CH(CH3))n-, nH), 5.10 (q, cholesterol-O(CdO)-
CH(CH3)-, 1H), 4.65 (m, cholesterol C-3 H, 1H), 1.55 (d,
-(O(CdO)CH(CH3))n-, 3nH), 0.65 (s, cholesterol C-18 H, 3H).
nj (1H NMR) ) 26. MS (MALDI-TOF): Mw ) 2611, Mn ) 2519,
Mw/Mn ) 1.04. nj (MALDI-TOF MS) ) 25.
(4) Heino, S.; Lusa, S.; Somerharju, P.; Ehnholm, C.; Olkkonen,
V. M.; Ikonen, E. Proc. Natl. Acad. Sci. U.S.A. 2000, 97,
8375-8380.
(5) The liquid crystalline behavior of certain cholesteryl-esters
was discovered more than one century ago. See, e.g.: Rein-
itzer, F. Monatsch. Chem. 1888, 9, 421-441. For more recent
Ch olester yl)(L-La ctic a cid )nj)Glycin a te Tr iflu or oa ce-
tic Acid Sa lt (14). As a typical example the end group

Macromolecules, Vol. 35, No. 3, 2002
Self-Assembling Biomaterials 759
overviews, see, e.g.: (a) Weiss, R. G. Tetrahedron 1988, 44,
3413-3475. (b) Goodby, J . W. Liq. Cryst. 1998, 24, 25-38.
(6) See, e.g.: (a) Kulkarni, R. K.; Pani, K. C.; Neuman, C.;
Leonard, F. Arch. Surg. 1966, 93, 839-843. (b) Cao, Y.;
Vacanti, J . P.; Paige, K. T.; Upton, J .; Vacanti, C. A. Plast.
Reconstr. Surg. 1997, 100, 297-302. (c) Maquet, V.; J e´roˆme,
R. Mater. Sci. Forum 1997, 250, 15-42.
estimated to ∼14 Å using molecular dynamics simulations
(SYBYL). The length of a single L-lactic acid repeat (∼2.9 Å)
was obtained from X-ray diffraction data published in the
literature: de Santis, P.; Kovacs, A. J . Biopolymers 1968, 6,
299-306
(16) Olabisi, O.; Robeson, L. M.; Shaw, M. T. Polymer-Polymer
Miscibility; Academic Press: New York, 1979.
(7) For the polymerization of L-lactide using functional alu-
minium alkoxides, see, e.g.: (a) Dubois, Ph.; J e´roˆme, R.;
Teyssie´, Ph. Makromol. Chem., Macromol. Symp. 1991, 42/
43, 103-116. (b) Barakat, I.; Dubois, Ph.; J e´roˆme, R.; Teyssie´,
Ph. J . Polym. Sci.: Part A: Polym. Chem. 1993, 31, 505-
514.
(17) (a) Ringsdorf, H. J . Polym. Sci., Part C, Polym. Symp. 1975,
51, 135-153. (b) Akashi, M.; Takemoto, K. Adv. Polym. Sci.
1990, 97, 107-146.
(18) Moore, J . S.; Stupp, S. I. Macromolecules 1990, 23, 65-70.
(19) (a) Klok, H.-A.; Hwang, J . J .; Iyer, S.; Tew, G. N.; Li, L. S.;
Stupp, S. I. Polym. Prepr. (Div. Polym. Chem., Am. Chem.
Soc.) 1998, 39 (2), 166-167. (b) Klok, H.-A.; Hwang, J . J .;
Hartegerink, J . D.; Stupp, S. I. To be submitted for publica-
tion.
(20) Sheehan, J . C.; Cruickshank, P. A.; Boshart, G. L. J . Org.
Chem. 1961, 26, 2525-2528.
(8) Kricheldorf, H. R.; Kreiser-Saunders, I. Polymer 1994, 35,
4175-4180.
(9) For other examples of the Sn(oct)2 catalyzed bulk-polymer-
ization of L-lactide, see, among others: (a) Kricheldorf, H.
R.; Kreiser-Saunders, I.; Boettcher, C. Polymer 1995, 36,
1253-1259. (b) Van Dijk-Wolthuis, W. N. E.; Tsang, S. K.
Y.; Kettenes-van den Bosch, J . J .; Hennink, W. E. Polymer
1997, 38, 6235-6242.
(21) Scriven, E. F. C. Chem. Soc. Rev. 1983, 12, 129-161.
(22) We found that the coupling of aromatic acids to aromatic
amines or phenols generally proceeded smoothly using the
DIPC/DPTS method. Reactions between aliphatic acids and
aliphatic alcohols, in contrast, generally were most successful
under EDC/DMAP conditions. However, we did not carry out
systematic efforts to optimize the coupling conditions.
(23) For other examples of the use of the 9-fluorenylmethylester
as a carboxyl protecting group, see amongst others: (a)
Bednarek, M. A.; Bodansky, M. Int. J . Pept. Protein Res. 1983,
21, 196-201. (b) Kessler, H.; Siegmaier, R. Tetrahedron Lett.
1983, 281-282. (c) Bolin, D. R.; Wang, C.-T.; Felix, A. M. Org.
Prep. Proc. Int. 1989, 21, 67-74.
(10) Mn and Mw were calculated from the MALDI-TOF mass
spectra using the following well-known equations: Mn
)
(ΣiniMi)/(Σini); Mw ) (ΣiniMi2)/(ΣiniMi). In these equations, Mi
and ni represent, respectively, the mass and intensity of the
individual peaks in the mass spectrum.
(11) See, e.g.: (a) Bero, N.; Kasperczyk, J .; J edlinski, Z. J .
Makromol. Chem. 1990, 191, 2287-2296. (b) Montaudo, G.;
Montaudo, M. S.; Puglisi, C.; Samperi, F.; Spassky, N.;
Leborgne, A.; Wisniewski, M. Macromolecules 1996, 29,
6461-6465.
(12) Hwang, J . J .; Iyer, S. N.; Li, L.-S.; Stupp, S. I. Submitted for
publication. Electron diffraction patterns indicate a smectic
Eh phase occurs at room temperature, this phase was
previously discovered in our group (see: Li, L.-S.; Hong, X.
J .; Stupp, S. I. Liq. Cryst. 1996, 21, 469-483).
(13) J amshidi, K.; Hyon, S.-H.; Ikada, Y. Polymer 1988, 29, 2229-
2234.
(14) De J ong, S. J .; van Dijk-Wolthuis, W. N. E.; Kettenes-van
den Bosch, J . J .; Schuyl, P. J . W.; Hennink, W. E. Macro-
molecules 1998, 31, 6397-6402.
(15) For the calculation of the approximate extended length of the
molecules (L), the length of the cholesterol-moiety was
(24) Tarbell, D. S.; Yamamoto, Y.; Pope, B. M. Proc. Natl. Acad.
Sci. U.S.A. 1972, 69, 730-732.
(25) Only the shifts of the protons at C-3, C-6 and C-18 positions
of the cholesterol-moiety will be reported. All other reso-
nances are in agreement with the reference 1H NMR spec-
trum: Handbook of Proton-NMR Spectra and Data (Asahi
Research Center Co., Ltd, Sasaki, S.); Academic Press: San
Diego, CA, 1985; Vol 5, spectrum no. 3958.
MA010907X