Prednisolone-α-cyclodextrin-star PEG polypseudorotaxanes with controlled drug delivery properties

Article abstract of DOI:10.1039/c0ob00039f

The reaction of α-amino-ω-methoxypoly(ethylene glycol) [M = 5000] or star α-amino-poly(ethylene glycol) [M = 20000] with hemiesters of prednisolone dicarboxylic acids (succinic, glutaric, adipic, phthalic acid) has been used to prepare the corresponding conjugates. The rate of esterase catalyzed hydrolysis of the conjugates is controlled by the molecular mass of poly(ethylene glycol) and the length of the linker between prednisolone and poly(ethylene glycol) (τ_1/2 ～ 5-0.5 h). The enzymatic hydrolysis proceeds most rapidly at conjugates with linkers derived from adipic and phthalic acids. The synthesized conjugates form polypseudorotaxanes with α-cyclodextrin which were characterized by 2D NOESY NMR spectra, powder X-ray diffraction patterns and in one case also by STM microscopy. In the case of the polypseudorotaxane having the linker derived from adipic acid, the enzymatic release proceeds ca. five times slower in comparison with the rate of prednisolone release from the corresponding conjugate. The rate of prednisolone release from the carrier can be controlled by three factors: character of the linker between the polymeric carrier and prednisolone, the molecular mass of poly(ethylene glycol) and complex formation with α-cyclodextrin. The synthesized polypseudorotaxanes represent new promising transport systems intended for targeted release of prednisolone in transplanted liver.

Full text of DOI:10.1039/c0ob00039f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Prednisolone-a-cyclodextrin-star PEG polypseudorotaxanes with controlled
drug delivery properties†‡
Elisˇka B´ılkova´,a Milosˇ Sedla´k,*a Bohuslav Dvorˇa´k,b Karel Ventura,b Petr Knotekc and Ludv´ık Benesˇc
Received 23rd April 2010, Accepted 17th August 2010
DOI: 10.1039/c0ob00039f
The reaction of a-amino-w-methoxypoly(ethylene glycol) [M = 5000] or star a-amino-poly(ethylene
glycol) [M = 20 000] with hemiesters of prednisolone dicarboxylic acids (succinic, glutaric, adipic,
phthalic acid) has been used to prepare the corresponding conjugates. The rate of esterase catalyzed
hydrolysis of the conjugates is controlled by the molecular mass of poly(ethylene glycol) and the length
of the linker between prednisolone and poly(ethylene glycol) (t1/2 ~ 5–0.5 h). The enzymatic hydrolysis
proceeds most rapidly at conjugates with linkers derived from adipic and phthalic acids. The synthesized
conjugates form polypseudorotaxanes with a-cyclodextrin which were characterized by 2D NOESY
NMR spectra, powder X-ray diffraction patterns and in one case also by STM microscopy. In the case
of the polypseudorotaxane having the linker derived from adipic acid, the enzymatic release proceeds
ca. five times slower in comparison with the rate of prednisolone release from the corresponding
conjugate. The rate of prednisolone release from the carrier can be controlled by three factors: character
of the linker between the polymeric carrier and prednisolone, the molecular mass of poly(ethylene
glycol) and complex formation with a-cyclodextrin. The synthesized polypseudorotaxanes represent
new promising transport systems intended for targeted release of prednisolone in transplanted liver.
glucosidases (E.C.3.2.1.21) which are specifically present in the
Introduction
infected tissue.14 Local release of PS in transplanted liver can be
Prednisolone (PS) belongs among the group of glucocorticoids
achieved by the action of the liver carboxyesterases (E.C.3.1.1.1),
which exhibit a high anti-inflammatory potential. For this feature
as previously used in the case of dextran–methylprednisolone
conjugates.3,4 The targeted administration to liver is accom-
plished by the fact that the rank order of hydrolytic activity
of carboxyesterases toward ester groups is liver/kidney > small
intestine > lung ꢀ blood.8
they are extensively used for the prevention of rejection in organ
transplantation. Nevertheless, the application of steroids and
other immunosuppressants in these cases is often complicated
due to their adverse side effects, such as diabetes, hypertension,
Cushing’s syndrome and osteoporosis.1 In spite of new pharma-
ceutical formulas of prednisolone and methylprednisolone such
as liposomal forms,2 dextran conjugates3,4 and also nanoparticles
of prednisolone5 have been developed recently, still minimizing of
these side effects is a challenging issue. The results of therapeutic
studies show that it is eligible to release the immunosuppressive
agent directly in the transplanted organ to furnish the tissue with
the active drug.6 Controlled release of the therapeutically active
substance can be achieved by the general principle based on the
presence of specific enzymes.7–8 In our earlier works,9–12 we have
prepared and in vitro tested targeted conjugates of amphotericin
B and nystatin13 with substituted poly(ethylene glycols). The last
types of conjugates12,13 are designed to be sensitive towards b-
The aim of this work was to prepare and characterize such
delivery systems of prednisolone that would highly selectively
release PS in transplanted liver and simultaneously allow variable
control of the rate of PS release at a molecular level. As carriers
we have chosen a-amino-w-methoxypoly(ethylene glycol) [M =
5000] and star a-aminopoly(ethylene glycol) [pentaerythritol a-
aminopoly(ethylene glycol)ether, M = 20 000], because their bio-
compatibility makes them ideal to employ in such pharmaceutical
applications.15 In the structure of the proposed conjugates was
the connection between PS and poly(ethylene glycols) (PEGs)
attained by different linkers from a series of dicarboxylic acids
(succinic, glutaric, adipic, phthalic acid). Prednisolone was linked
to one carboxylic group of the individual acid via an ester function
that can be readily cleaved by esterase (linker), whereas the other
carboxylic group of the acid was suitable for linking to amino-PEG
via amide function (Fig. 1).
aInstitute of Organic Chemistry and Technology, Faculty of Chemical
Technology, University of Pardubice, Studentska´ 573, 532 10 Pardubice,
Czech Republic. E-mail: milos.sedlak@upce.cz; Fax: +420 466 037 068;
Tel: +420 466 037 506
We have explored several approaches to control the PS release
rate in enzyme kinetics. One of the possibilities is to modify
the linker employing the above-mentioned dicarboxylic acids
having different lengths and characters of chain. In addition,
formation of inclusion compounds of medical drugs with cy-
clodextrins was previously proposed as an efficient tool for
modulation of drug delivery features.16 For example, modification
with cyclodextrins was adopted17 for slowing down the release
of prednisolone as well as methylprednisolone. However, a more
bDepartment of Analytical Chemistry, Faculty of Chemical Technology,
University of Pardubice, Studentska´ 573, 532 10 Pardubice, Czech Republic
cJoint Laboratory of Solid State Chemistry of the University of Pardubice
and the Institute of Macromolecular Chemistry AS CR, v.v.i. University of
Pardubice, Studentska´ 95, Pardubice 532 10, Czech Republic
† Electronic supplementary information (ESI) available: 1H and 13C NMR
and IR spectra. See DOI: 10.1039/c0ob00039f
‡ Dedicated to Professor Jitka Moravcova´ on the occasion of her 60th
birthday.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 5423–5430 | 5423
©

View Article Online
Fig. 1 Structures of conjugates 1a–e.
recent insight into this area is represented by conjugates based
Results and discussion
on polypseudorotaxanes18–28 (i.e. complexes of substituted PEGs
with cyclodextrins). Polypseudorotaxanes obtained through the
complex formation of substituted PEGs and a-cyclodextrin (a-
CD) have proven distinct retardation of the overall excretion
rate of the conjugate from organism.18–22 Therefore, in continuing
our efforts to develop potential pharmaceutical formulations
of prednisolone with prolonged effects, we have prepared and
characterized polypseudorotaxanes 3b and 3d (Fig. 2.) derived
from a-CD and from conjugates 1b and 1d.
Synthesis and characterization of conjugates 1a–e
The new hemiesters of dicarboxylic acids and prednisolone 2b–
d were prepared in analogy to the earlier described method29
for prednisolone 21-hemisuccinate (2a), i.e. by the reaction of
correspondinganhydrides (succinic, glutaric, adipic, phthalicacid)
with prednisolone in the presence of pyridine (57–67%). All the
prepared hemiesters 2a–d were characterized by melting points,
microanalysis, 1H, 13C NMR and FT-IR spectroscopy.
After activation with dicyclohexylcarbodiimide (DCC), the
hemiesters of prednisolone 2a–d were subsequently connected
to amino PEGs (mPEG, sPEG) via amide function (Scheme 1).
The conjugates 1a–e prepared in this way were recrystallized
1
from propan-2-ol and characterized by H NMR spectroscopy
and GPC. The integral intensities of two signals of –CH in
prednisolone (~7.3; ~6.0 ppm) and the signals of –CH2– in PEG
1
(3.2–3.8 ppm) in the H NMR spectra showed that in conjugate
1a the molar ratio is PS : mPEG = 1 : 1. In conjugates 1b–e the
molar ratio was PS : sPEG = 4 : 1. The 1H NMR spectra were also
used for determination of the Mn value and the GPC analysis for
determination of the Mw value, thus giving the Mw/Mn ratio for
the individual conjugates 1a–e.
Scheme 1 Synthesis of PS-PEG conjugates 1a–e.
Synthesis and characterization of polypseudorotaxanes 3b and 3d
The formation and decomposition of polypseudorotaxanes
PEG/a-CD represent an equilibrium process;23–28 therefore the
complex formation of conjugates 1b and 1d was performed with
excess of a-CD. Polypseudorotaxanes 3b and 3d precipitated
Fig.
2
Schematic structure of polypseudorotaxanes derived from
a-cyclodextrin-star poly(ethylene glycol)amid-linker-ester-prednisolone; n
~ 26 for linker succinate (3b); n ~ 12 for linker adipate (3d).
5424 | Org. Biomol. Chem., 2010, 8, 5423–5430
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
within 14 days standing of a mixture of saturated aqueous solution
of a-CD and saturated aqueous solution of the corresponding
conjugate at room temperature; they were isolated by centrifu-
gation, purified by repeated decantation with distilled water,
again isolated by centrifugation, and dried in vacuum at room
temperature. Polypseudorotaxanes 3b (37%) and 3d (62%) were
characterized by using powder X-ray diffraction, 1H and 2D-
NOESY NMR spectroscopy. In the case of polypseudorotaxane
3d the characterization was supplemented by Scanning Tunneling
Microscopy (STM).
Fig. 3 shows powder X-ray diffraction patterns of the a-
CD alone (Fig. 3(a)) as compared with the physical mixture of
a-CD and conjugate 1b (Fig. 3(b)), and polypseudorotaxanes
3b (Fig. 3(c)) and 3d (Fig. 3(d)). The diffraction pattern of
polypseudorotaxanes 3b,d is obviously different from those of the
physical mixture and a-CD alone. When compared to previously
published results,26–28 it indicates that the PEG chain forms a
polypseudorotaxane arrangement in which the long PEG chain
is embedded in the stacked host channel. The diffraction peaks at
2H = 12.9, 19.8 and 22.6◦ resembled arrangement of a-CD in the
order head-to-head/tail-to-tail,30 as shown in Fig. 3(c) (d).
The stoichiometry of both polypseudorotaxanes 3b and 3d was
determined by integration of peak surface areas in 1H NMR
spectrum of anomeric proton of a-CD (4.84 ppm) and the ethylene
oxide protons of PEG chain (~ 3.55 ppm) in dimethyl sulfoxide
solution (Fig. 4). The 2D NOESY NMR spectra of polypseudoro-
taxanes 3b and 3d (Fig. 5) clearly show the crosspeaks belonging
to non-bonding interaction through space, namely between the
hydrogen atoms of the poly(ethylene glycol) unit [CH2CH2O] and
the hydrogen atom 3 of the a-CD units, which also confirms
the polypseudorotaxane arrangement in accordance with the
literature.26 The integral intensities show that the stoichiometries
Fig. 5 2D NOESY NMR spectrum of polypseudorotaxane 3d. Arrows
point to crosspeaks belonging to non-bonding interaction through space
between the hydrogen atoms in -CH2CH2O- of PEG and hydrogen atom 3
of a-cyclodextrin.
Fig. 3 Powder X-ray diffraction patterns: (a) a-CD alone; (b) physical
mixture of a-CD with conjugate 1b; (c) polypseudorotaxane 3b; (d)
polypseudorotaxane 3d.
Fig. 4 1H NMR spectrum of polypseudorotaxane 3d. Insets represent parts of the spectrum at increased intensity with the assigned signals of prednisolone.
This journal is The Royal Society of Chemistry 2010 Org. Biomol. Chem., 2010, 8, 5423–5430 | 5425
©

View Article Online
Fig. 6 STM images of hydrated 3d (a, b, c) and dehydrated 3d (d, e), histogram of a-CD distances (f).
of polypseudorotaxanes 3b and 3d prepared repeatedly under
the same conditions are substantially different. In the case of
polypseudorotaxane3b each arm of the PEG chain is threaded into
32 a-CD units on average, and polypseudorotaxane 3b contains in
total 128 a-CD units, i.e. there are 3.5 [CH2CH2O] units per one
a-CD unit. However, in the case of polypseudorotaxane 3d the
stoichiometry found is half the previous, i.e. there are ca. 6 units
[CH2CH2O] units per one a-CD unit. These findings show that the
stoichiometry ratio [CH2CH2O]/a-CD is significantly influenced
by the character of the linker between PS and PEG. For instance,
the ratio determined earlier31 for non-functionalized sPEG [M =
7400] was 2.2 units [CH2CH2O] per one unit of a-CD.
The necklace like structure of a-CD units threaded onto PEG
chains was directly observed by using STM for hydrated (Fig. 6(a,
b)) and dehydrated (d, e) samples of 3d. The a-CD units were
detected as bright dots and PEG chains as lines connecting them.
The distance between two a-CD units was analyzed (Fig. 6(f);
the L value according to Koji Miyake32). The minimum L value
found was 6.3 nm and the typical value was 30 nm. No shuttle
manipulation according to Ref. 32 was observed during STM
experiment due to the higher interaction to HOPG in comparison
to MoS2 substrate. Self-assembly effects show lamellar structure
of 3d (Fig. 6(c)).
The lateral dimension of a-CD units in the STM image of
polypseudorotaxane 3d was changed dramatically according to
degree of hydration of the sample. The diameter of hydrated
particles was found between 6–8 nm and the observed diameter
of PEG reached a similar value (Fig. 7(a)). This was caused by
residues of the solvating layer covering a-CD units as well as the
PEG chain. After more exhaustive drying the particle diameter
detected decreased to 2.5–3.5 nm, which is in better agreement with
the particle size of a-CD (1.7 nm) and with previously published
Fig. 7 Line-scans of hydrated 3d (a) and dehydrated 3d (b) a-CD particles.
results.32–35 The bigger size resulted as an artefact due to tunneling
current acquisition from the side of the a-CD particle as well as
residues of solvating layer from inadequate drying or progressive
hydration adsorbed during the experiment.
Stability of conjugates 1a–e, 3b,d in phosphate buffer solutions and
hydrochloric acid solutions
Conjugates 1a–e, 3b,d did not exhibit any measurable increase in
PS concentration at least for one week at the temperature of 37 ◦C
in the phosphate buffers (pH: 7.4, 5.8, 1 ¥ 10-1 mol L-1) and in
acetate buffer (pH 4.0, 1 ¥ 10-1 mol L-1). In the hydrochloric
acid solution (1 ¥ 10-2 mol L-1; I = 1 mol L-1; KCl) an
observable increase in PS concentration took place only in more
than 72 h.
5426 | Org. Biomol. Chem., 2010, 8, 5423–5430
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 1 Rate constants (k) and half-lives (t1/2) of enzymatic hydrolysis
of conjugates 1a–e and polypseudorotaxanes 3b,d (pig liver esterase, E.C.
3.1.1.1, pH 7.4, 37 ◦C)
Conjugate or
polypseudorotaxane
k/min-1
t1/2/min
1a
1b
1c
1d
1e
3b
3d
3.8 ¥ 10-3 3 ¥ 10-4
2.7 ¥ 10-3 1 ¥ 10-5
1.3 ¥ 10-2 2 ¥ 10-3
1.8 ¥ 10-2 6 ¥ 10-4
1.4 ¥ 10-2 3 ¥ 10-3
2.5 ¥ 10-3 1 ¥ 10-4
3.8 ¥ 10-3 3 ¥ 10-4
184 15
259 10
52
37
48
7
2
8
275 11
184 16
Enzymatic hydrolysis of conjugates 1a–e and polypseudorotaxanes
3b,d by carboxyesterase (E.C.3.1.1.1)
Fig. 8 Concentration-time dependence (mmol L-1 versus hours) of
prednisolone release in a solution of pig liver esterase (E.C.3.1.1.1; 0.1 mg;
17 units g-1, phosphate buffer pH 7.4; 1 ¥ 10-1 mol L-1; 1 mL; 37 ◦C)
for conjugate 1d (᭺) and polypseudorotaxane 3d(᭹). The points were
determined experimentally using HPLC at 245 nm; the curves correspond
to pseudo-first-order rate equation.
For the purpose of verification of release of PS from conjugates
1a–e and polypseudorotaxanes 3b,d the kinetics of enzymatic
hydrolysis was studied by means of HPLC measuring the free
PS concentration increase (Table 1).
Table 1 presents rate constants and half-lives of enzymatic
hydrolysis for individualconjugates1a–e and for polypseudorotax-
anes 3b,d. The values given in the table show the influence of three
parameters upon the rate of PS release from the carrier. The effect
of the first parameter, viz. molecular mass of the PEGs, results from
the comparison of hydrolysis rates of conjugates 1a and 1b which
contain the same linker (succinic acid) but differ in molecular
mass and character of polymeric carrier. The conjugate with lower
molecular mass 1a (MPEG = 5000) undergoes the enzymatic release
of PS ca. 1.5 times faster than conjugate 1b (MPEG = 20 000),
which is in agreement with earlier findings.10,36 The second tool
for influencing the enzymatic hydrolysis rate is seen from the
comparison of PS release rates of conjugates 1b–d which differ
only in the length of chain in the used homologous series of linkers
derived from dicarboxylic acids and the conjugate 1e derived
from phthalic acid. The enzymatic hydrolysis rate increases with
increasing length of the hydrophobic chain linked to poly(ethylene
glycol) chain. Behavior of these systems exhibits a similar order as
the compared enzymatic hydrolysis of butyryl, propionyl, acetyl
esters and esters of aromatic acids.37
Furthermore, we investigated the PS release rates from
polypseudorotaxanes 3b,d which were compared with the PS
release rates of the corresponding conjugates 1b,d. The overall
PS release rate from polypseudorotaxanes 3b,d involves two
kinetic processes: 3b,d → 1b,d → PS, i.e. dethreading of a-
CD units (3b,d → 1b,d) and enzymatic hydrolysis of the ester
bond. The overall rate of PS release from the polypseudorotaxane
is controlled by the slowest process, i.e. the rate-limiting step. In
the case of conjugate 3b the decomposition half-lives of 3b and 1b
are comparable (Table 1), hence the rate-limiting step is enzymatic
hydrolysis 1b → PS, which means that dethreading of a-CD units
must be fast enough that it is kinetically insignificant. From Fig. 8
follows that the enzymatic hydrolysis 1d → PS obeys the pseudo-
first-order kinetics (analogous kinetic dependencies were also
obtained in the case of conjugates 1a–e and polypseudorotaxane
3b). In the case of hydrolysis 3d → 1d → PS the kinetic dependence
is significantly affected by the rate of the first process 3d → 1d, and
the experimental dependence begins to differ from pseudo-first-
order kinetics (Fig. 8). The PS release rate of polypseudorotaxane
3d was more than five times slower than that of conjugate 1d
(Table 1). The described results show that the PS release rates from
the carrier can be controlled in a modular way by three factors: the
linker between the polymeric carrier and PS, the molecular mass
of PEGs, and complex formation with a-CD.
Conclusions
The polypseudorotaxanes derived from a-CD and PS-PEG con-
jugates were prepared with the aim to accomplish high selectivity
of prednisolone release in transplanted liver. The selectivity of
targeted release is ensured by the presence of ester linkage which
is sensitive to liver esterase (E.C. 3.1.1.1). The rates of release
of PS from the carrier can be controlled in a modular way by
three factors: linker between polymeric carrier and PS, molecular
mass of the PEGs, and complex formation with a-CD. The
enzyme catalyzed PS release can be slowed down by shortening
the aliphatic chain of the linker and increasing the molecular mass
of the polymeric carrier. The character of the linker at the end
of the PEG chain significantly affects also the stoichiometry of
complex formation of PEG with a-CD, which is even reflected
in the course of dethreading of a-CD units. All the conjugates
and polypseudorotaxanes are relatively stable in acid medium of
hydrochloric acid (1 ¥ 10-2 mol L-1), which is a precondition of
the possibility of peroral administration that is considerate of the
patient.
Experimental
General
Unless otherwise stated pentaerythritol poly(ethylene glycol) ether
(M = 20 000) was obtained from JenKem Technology USA.
Prednisolone and other chemicals and solvents were obtained from
Fluka or Aldrich and used without further purification.
The 1H NMR spectra were calibrated with respect to
the middle signal in the multiplet of solvent (d = 2.55;
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 5423–5430 | 5427
©

View Article Online
DMSO-d6) or (d = 7.24; CDCl3). The 1H NMR spectra of polymer
conjugates were measured38,39 with a relaxation delay of 6 s and
an acquisition time of 4 s. The 13C NMR spectra were measured
in a standard way using broad-band proton decoupling and/or
pulse sequence APT. The 13C NMR spectra were calibrated with
respect to the middle signal in the multiplet of solvent (d = 39.6;
DMSO-d6) or (d = 77.0; CDCl3). 2D NOESY measurements were
performed using the standard Bruker pulse sequence library with
the experimental conditions as follows: spectral width of 4807 Hz,
128 transients per increment for the 128 increments, with 400 ms
mixing time duration and 2048 data points in the F2 domain. The
spectra were processed with a sine-bell squared in both dimensions.
The IR spectra were measured in solid state on an FT-IR-Perkin
Elmer Spectrum BX instrument with horizontal ATR modul with
ZnSe crystal in the range from 650 to 4000 cm-1. The HPLC
analysis was performed using a Shimadzu (Kyoto, Japan) HPLC
system consisting of two Model LC-10ADvp pumps, a Model
SPD-M10Avp UV/vis spectrophotometric detector, a DGU-14A
degasser, a CTO-10ASvp column oven and a SCL-10Avp system
controller at 25 ◦C. Separation was performed using a guard
column Security Guard, 4 mm ¥ 3 mm, C18 (Phenomenex,
Torrance, CA, USA). Elution was carried out with a gradient
of methanol: water at a flow rate of 1 mL min-1. The mobile
phase was filtered through a 0.45 mm Hydrophilic Polypropylene
Membrane Filter (Pall Corporation, Ann Arbor, Michigan, USA).
The injection volume was set at 20 mL. The detection wavelength
was 245 nm. The purity of conjugates 1a–e was estimated by
in vacuum (dehydrated samples). The STM experiment was
performed under ambient conditions by using Solver ProM, Nt-
MDT (Russia) with a Pt/Ir tip with calibration based on atomic
scale resolution. The measurement was carried out in constant
current mode with a sample bias voltage of +400 mV, a tunneling
current of 2.10 nA with resolution 256 ¥ 256 pxs2.
Study of stability of conjugates 1a–e in phosphate buffer solutions
and hydrochloric acid solution
A methanolic solution of studied conjugates 1a–e (20 mL; 30 mg
mL-1) was injected into solutions (1 mL) of buffers with pH
values: 7.4, 5.8 (phosphate), 4.0 (acetate), 1 ¥ 10-1 mol L-1 and
into solution of hydrochloric acid (1 ¥ 10-2 mol L-1; I = 1 mol L-1;
KCl). The increases of PS concentration were monitored by means
of HPLC at 37 ◦C.
Study of hydrolysis of conjugates 1a–e by esterase
The kinetic measurements were carried out at 37 ◦C. A methanolic
solution of the conjugate studied (20 mL; 30 mg mL-1) was injected
into a solution of commercial pig liver esterase (E.C.3.1.1.1;
Sigma; 0.1 mg; 17 units g-1) in phosphate buffer (pH 7.4; 1 ¥
10-1 mol L-1; 1 mL). The solution obtained was kept at constant
temperature and samples were taken at definite time intervals
for determination of PS concentration in the course of time by
using HPLC. The concentration-time dependence was optimized
to obtain the pseudo-first-order rate constants (k = ln2/t1/2) and
half-lives (t1/2) of enzymatic hydrolysis.
R
means of HPLC using LiChroCARTꢀ 125 mm ¥ 4 mm column
R
packed with LiChrospherꢀ 100 RP-18e 5 mm (Merck, Darmstadt,
Germany) and eluted with a mobile phase of acetonitrile with
20 mM chelaton II. Gel permeation chromatography was used for
the estimation of Mw of conjugates 1a–e. The HPLC device was
identical with that used for purity determination but for the follow-
ing parameters: HEMA-BIO columns (hydrophilic modification
HEMAGel, particle size 10 mm, porosity 40/100/300/1000) at
25 ◦C using an RI detector and UV/vis detector. Redistilled water
(pH 7.1) was used as the eluent. The columns were calibrated with
a series of standard PEGs (Merck) and standard sPEGs (JenKem
Technology USA) of various molecular masses. The microanalyses
were performed on an apparatus of FISONS.
General procedure for synthesis of prednisolone hemiesters (2a–d)
A solution of prednisolone (1 g; 2.77 mmol) and the corresponding
anhydride (12.55 mmol) in pyridine (8 mL) was stirred at room
temperature. After 24 h, the reaction mixture was poured onto a
mixture of ice (25 g), water (25 mL) and conc. HCl (10 mL). The
separated crystals were collected by filtration, washed with water,
dried, recrystallized from toluene and dried again.
Prednisolone 21-hemisuccinate (2a). Yield: 0.8 g (63%), mp
◦
◦
1
172–174 C (lit.,29 176 C). H NMR (500.13 MHz, DMSO-d6):
dH: 0.82 (s, 3H); 0.92 (m, 1H); 1.06 (m, 1H); 1.33 (m, 1H); 1.42 (s,
3H); 1.46 (m, 1H); 1.58 (m, 1H); 1.68 (m, 2H); 1.91 (m, 1H); 2.06
(m, 2H); 2.33 (dd, J = 13 Hz, J = 3 Hz, 1H); 2.52 (m, 4H); 2.65 (m,
2H); 4.11 (d, J = 19 Hz, 1H); 4.32 (s, 1H); 4,53 (d, J = 19 Hz, 1H);
4.80 (d, J = 17 Hz, 1H); 5.12 (d, J = 17 Hz, 1H); 5.95 (s, 1H); 6.19
(d, J = 10 Hz, 1H); 7.37 (d, J = 10 Hz, 1H). 13C NMR (DMSO-d6):
dC: 16.6, 20.9, 23.6, 28.5, 28.7, 31.0, 31.4, 33.0, 34.0, 43.9, 46.6,
47.1, 51.1, 55.5, 66.0, 67.7, 68.4, 77,2, 88.6, 121.6, 127.10, 156.9,
170.7, 171.7, 173.3, 185.3, 205.2. FT-IR: nmax/cm-1 3407, 2938,
2855, 1721, 1708, 1654 and 1611. Anal. calcd for C25H32O8 (460)
(%): C, 65.24; H, 7.00. Found: C, 64.89; H, 7.21.
X-Ray diffraction
˚
Powder X-ray diffraction data (Cu-Ka, l = 1.5418 A) were
collected on a D8 Advance diffractometer (Bruker AXS,
Germany) with Bragg-Brentano H-H goniometer (radius
217.5 mm) equipped with a secondary beam curved graphite
monochromator and Na(Tl)I scintillation detector. The generator
was operated at 40 kV and 30 mA. The scan was performed at room
temperature from 2 to 65◦ (2 H) in 0.02◦ step with a counting time
of 10 s per step.
Predni◦solone 21-hemiglutarate (2b). Yield: 0.7 g (67%), mp
173–175 C. 1H NMR (500.13 MHz, DMSO-d6): dH: 0.77 (s, 3H);
0.87 (dd, J = 11 Hz, J = 3,4 Hz, 1H); 0.99 (m, 1H); 1.29 (m,
1H); 1.37 (s, 3H); 1.43 (m, 1H); 1.63 (m, 3H); 1.76 (m, 2H); 1.87
(m, 1H); 2.01 (m, 2H); 2.30 (m, 4H); 2.42 (m, 2H); 2.47 (m, 1H);
4.28 (m, 1H); 4.75 (d, J = 17.6 Hz, 1H); 5.06 (d, J = 17.6 Hz,
1H); 5.91 (s,1H); 6.15 (dd, J = 10.1 Hz, J = 1.8 Hz, 1H); 7.32 (d,
J = 10.1 Hz, 1H). 13C NMR (125.76 MHz, DMSO-d6): dC: 16.5,
Scanning tunneling microscopy (STM)
STM was performed for characterization of particle topol-
ogy/location. An aqueous solution of 3d (0.5 wt.%) was dropped
onto a fresh, highly ordered pyrolytic graphite (HOPG) surface
at room temperature, and then the samples were dried at a
temperature of 60 ◦C for 2 h (hydrated samples) or for 72 h
5428 | Org. Biomol. Chem., 2010, 8, 5423–5430
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
20.0, 20.9, 23.5, 30.9, 31.4, 32.4, 32.5, 33.1, 34.0, 43.8, 47.1, 51.1,
55.4, 67.6, 68.4, 88.7, 121.6, 127.0, 156.9, 170.8, 172.1, 174.1, 185.3,
205.4. FT-IR: nmax/cm-1 3477, 2936, 1744, 1710, 1651, 1591 and
892. Anal. calcd for C26H34O8 (474) (%): C, 65.81; H, 7.22. Found:
C, 65.62; H, 7.56.
3.39–3.78 (m, 460H); 4.45 (m, 1H); 4.89 (d, J = 17 Hz, 1H); 5.03
(d, J = 17 Hz, 1H); 6.02 (s, 1H); 6.27 (d, J = 9,9 Hz, 1H); 6.73 (d,
J = 7.2 Hz, 1H); 7.33 (d, J = 9.5 Hz, 1H); 8.13 (m, 1H). Mw/Mn =
1.08.
Star poly(ethylene glycol)amid-glutarate-ester-prednisolone (1c).
Yield: 450 mg (82%).1H NMR (500.13 MHz, CDCl3): dH: 0.91 (s,
3H); 1.03 (m, 1H); 1.17 (m, 2H); 1.42 (s, 3H); 1.70 (m, 5H); 2.01
(m, 3H); 2.31 (m, 1H); 2.40 (m, 4H); 2.65 (m, 3H); 3.20–3.90 (m,
473H); 4.42 (m, 1H); 4.88 (m, 1H); 4.98 (m, 1H); 5.99 (s, 1H); 6.24
(d, J = 11 Hz, 1H); 7.30 (m, 1H); 7.80 (m, 1H). Mw/Mn = 1.06.
Prednisolone 21-hemiadipate (2c). Yield: 0.8 g (59%), mp 177–
179 ◦C. 1H NMR (500.13 MHz, DMSO-d6): dH: 0.75 (s, 3H); 0.85
(dd, J = 11 Hz, J = 3.3 Hz, 1H); 0.98 (m, 1H); 1.26 (m, 1H); 1.35
(s, 3H); 1.41 (m, 1H); 1.53 (m, 4H); 1.62 (m, 3H); 1.84 (m, 1H);
2.00 (m, 2H); 2.21 (m, 3H); 2.26 (m, 1H); 2.36 (m, 2H); 2.45 (m,
1H); 4.27 (m, 1H); 4.74 (d, J = 17.5 Hz, 1H); 5.02 (d, J = 17.6 Hz,
1H); 5.90 (s,1H); 6,14 (dd, J = 10.1 Hz, J = 1.8 Hz, 1H); 7.2 (d, J =
10.1 Hz, 1H). 13C NMR (125.76 MHz, DMSO-d6): dC: 16.7, 21.1,
23.7, 24.1, 24.2, 31.1, 31.6, 32.4, 33.2, 33.3, 33.6, 33.2, 44.1, 47.3,
51.3, 55.6, 67.8, 68.6, 88.9, 121.7, 127.2, 157.4, 171.3, 172.6, 174.7,
185.8, 205.7. FT-IR: nmax/cm-1 3480, 2936, 1713, 1651 and 1593.
Anal. calcd for C27H36O8 (489) (%): C, 66.38; H, 7.43. Found: C,
66.24; H, 7.52.
Star poly(ethylene glycol)amid-adipate-ester-prednisolone (1d).
Yield: 450 mg (82%).1H NMR (500.13 MHz, CDCl3): dH: 0.90 (s,
3H); 1.05 (m, 1H); 1.15 (m, 2H); 1.42 (s, 3H); 1.60–1.75 (m, 7H);
2.05 (m, 4H); 2.29 (m, 3H); 2.41 (m, 2H); 2.53 (m, 1H); 2.65 (m,
1H); 3.35–3.85 (m, 481H); 4.24 (m, 1H); 4.88 (m, 1H); 4.94 (m,
1H); 5.96 (s, 1H); 6.22 (d, J = 9.8 Hz, 1H); 7.28 (m, 1H); 7.85 (m,
1H). Mw/Mn = 1.05.
Predni◦solone 21-hemiphthalate (2d). Yield: 0.8 g (57%), mp
182–184 C. 1H NMR (500,13 MHz, DMSO-d6): dH: 0.79 (s, 3H);
0.87 (dd, J = 11.2 Hz, J = 3.1 Hz, 1H); 0.98 (m, 1H); 1.25 (m, 1H);
1.36 (s, 3H); 1.47 (m, 1H); 1.65 (m, 2H); 1.90 (m, 1H); 2.00 (m,
2H); 2.27 (m, 1H); 2.47 (m, 2H); 4.28 (m, 1H); 4.93 (d, J = 17.3 Hz,
1H); 5.26 (d, J = 17.3 Hz, 1H); 5.89 (s,1H); 6.13 (d, J = 10.1 Hz,
1H); 7.32 (d, J = 10.1 Hz, 1H); 7.63 (m, 2H); 7.73 (m, 2H). 13C
NMR (125.76 MHz, DMSO-d6): dC: 16.6, 20.8, 23.5, 30.9, 31.3,
31.4, 33.1, 33.9, 43.9, 47.1, 51.1, 55.4, 68.2, 68.4, 88.7, 121.5, 127.9,
127.2, 128.3, 128.6, 128.8, 131.6, 132.3, 156.8, 167.9, 170.6, 185.2,
205.0. FT-IR: nmax/cm-1 3368, 2939, 1770, 1740, 1651, 1303 and
1014. Anal. calcd for C29H32O8 (509) (%): C, 68.49; H, 6.34. Found:
C, 68.38; H, 6.47.
Star poly(ethylene glycol)amid-phthalate-ester-prednisolone (1e).
Yield: 500 mg (90%).1H NMR (500.13 MHz, CDCl3): dH: 0.89 (s,
3H); 0.96 (m, 1H); 1.16 (m, 2H); 1.40 (s, 3H); 1.60–1.80 (m, 4H);
2.03 (m, 3H); 2.26 (m, 1H); 2.56 (m, 2H); 3.20–3.80 (m, 505H);
4.40 (m, 1H); 5.05 (d, J = 17 Hz, 1H); 5.26 (d, J = 17 Hz, 1H); 5.97
(s, 1H); 6.22 (d, J = 10.7 Hz, 1H); 6.70 (d, J = 7 Hz, 1H); 7.34 (d,
J = 9.5 Hz, 1H); 7.49 (m, 1H); 7.72 (m, 1H); 7.77 (m, 1H); 8.09
(m, 1H). Mw/Mn = 1.12.
Preparation of polypseudorotaxanes 3b and 3d
A mixture of aqueous solutions of a-cyclodextrin (889 mg;
0.91 mmol) (8 mL) and conjugate 1a or 1d (100 mg; 4.6 mmol)
(3.5 mL) was stirred in a closed vial at room temperature. The
stirring was finished after 5 h. A turbidity began to appear after
24 h, and the conjugate separated after 14 days standing was
centrifuged, washed with small amount of water, centrifuged again
and dried in vacuum at room temperature.
General procedure for synthesis of conjugates 1a–e
A solution of 4-dimethylaminopyridine (31 mg; 0.25 mmol) with
the corresponding hemiester 2a–2d (0.5 mmol) in CH2Cl2 (5 mL)
was added to a solution of a-amino-w-methoxypoly(ethylene
glycol) (mPEG-NH2) (500 mg; 0.1 mmol; Mw = 5 000) or star
a-amino poly(ethylene glycol) (sPEG-NH2) (500 mg; 0.025 mmol;
Mw = 20 000) and dicyclohexylcarbodiimide (DCC) (103 mg;
0.5 mmol) in CH2Cl2 (10 mL), and the reaction mixture was stirred
at room temperature. After 2 days, the mixture was filtered and
poured into diethyl ether (250 mL). The crystals formed were
collected by filtration, and the crude product was recrystallized
twice from propan-2-ol.
Polypseudorotaxane of a-cyclodextrin-star poly(ethylene glycol)-
amid-succinate-ester-prednisolone (3b). Yield: 250 mg (37%). 1H
NMR (500.13 MHz, DMSO-d6): dH: 0.83 (s, 3H); 1.28 (m, 3H);
1.43 (s, 3H); 1.6–1.7 (m, 3H); 1.95 (s, 2H); 2.13 (s, 3H); 2.38 (m,
2H); 2.71 (m, 4H); 3.32 (m, 192H); 3.43 (m, the signal is partially
overlapped by the water signal); 3.55 (m, 366H); 3.60–3.64 (m,
192H); 3.67–3.72 (m, 384H); 3.82 (t, J = 9.4 Hz, 192H); 4.54 (t,
J = 5.7 Hz, 192H); 4.84 (d, J = 3.2 Hz, 192H); 5.49 (d, J = 2.7 Hz,
192H); 5.57 (d, J = 7.1 Hz, 192H); 5.96 (s, 1H); 6.20 (d, J = 10 Hz,
1H); 7.36 (d, J = 10 Hz, 1H); 8.05 (m, 1H).
Methoxypoly(ethylene glycol)amid-succinate-ester-prednisolone
(1a). Yield: 450 mg (82%). 1H NMR (500.13 MHz, CDCl3): dH:
0.88 (s, 3H); 0.99 (m, 1H); 1.19 (m, 2H); 1.41 (s, 3H); 1.49 (m,
1H); 1.68 (m, 3H); 2.03 (m, 3H); 2.29 (m, 1H); 2.63 (m, 4H); 2.84
(m, 2H); 3.33 (s, 3H); 3.41–3.74 (m, 446H); 4.41 (m, 1H); 4.88 (m,
1H); 5.00 (m, 1H); 5.97 (s, 1H); 6.22 (d, J = 10 Hz, 1H); 6.72 (d,
J = 7 Hz, 1H); 7.30 (m, 1H); 8.10 (t, J = 6.8 Hz, 1H). Mw/Mn =
1.18.
Polypseudorotaxane of a-cyclodextrin-star poly(ethylene glycol)-
1
amid-adipate-ester-prednisolone (3d). Yield: 300 mg (62%). H
NMR (500.13 MHz, DMSO-d6): dH: 0.82 (s, 3H); 1.1–1.2 (m,
3H); 1.29 (s, 3H); 1.43 (m, 1H); 1.5–1.6 (m, 3H); 1.84 (s, 1H);
1.95 (s, 2H); 2.13 (s, 4H); 2.23 (t, J = 7.4 Hz, 2H); 3.32 (d, J =
8.4 Hz, 108H); 3.41 (m, the signal is partially overlapped by the
water signal); 3.55 (m, 366H); 3.60–3.72 (m, 216H); 3.81 (t, J =
9 Hz, 108H); 4.54 (brs, 108H); 4.84 (d, J = 2.8 Hz, 108H); 5.49
(brs, 108H); 5.57 (brs, 108H) 5.96 (s, 1H); 6.20 (d, J = 10 Hz, 1H);
7.37 (d, J = 10 Hz, 1H); 7.90 (m, 1H).
Star poly(ethylene glycol)amid-succinate-ester-prednisolone (1b).
Yield: 500 mg (91%). 1H NMR (500.13 MHz, CDCl3): dH: 0.92 (s,
3H); 1.05 (m, 1H); 1.25 (m, 2H); 1.46 (s, 3H); 1.55 (m, 1H); 1.72
(m, 3H); 2.10 (m, 3H); 2.33 (m, 1H); 2.68 (m, 4H); 2.87 (m, 2H);
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 5423–5430 | 5429
©

View Article Online
16 T. Loftsson and D. Ducheˆne, Int. J. Pharm., 2007, 329, 1.
17 K. L. Larsen, F. L. Aachmann, R. Wimmer, V. J. Stella and U. M.
Kjølner, J. Pharm. Sci., 2005, 94, 507.
18 T. Ooya and N. Yui, J. Biomater. Sci., Polym. Ed., 1997, 8, 437.
19 N. Yui, T. Ooya and T. Kumeno, Bioconjugate Chem., 1998, 9, 118.
20 T. Ooya and N. Yui, J. Controlled Release, 1999, 58, 251.
21 T. Higashi, F. Hirayama, H. Arima and K. Uekama, Bioorg. Med.
Chem. Lett., 2007, 17, 1871.
22 T. Higashi, F. Hirayama, S. Yamashita, S. Misumi, H. Arima and K.
Uekama, Int. J. Pharm., 2009, 374, 26.
23 A. Harada and M. Kamachi, Macromolecules, 1990, 23, 2821.
24 A. Harada and M. Kamachi, Nature, 1992, 356, 325.
25 A. Harada and M. Kamachi, Nature, 1994, 370, 126.
26 G. Wenz, B.-H. Han and A. Mu¨ler, Chem. Rev., 2006, 106, 782.
27 S. Loethen, J.-M. Kim and D. H. Thompson, Polym. Rev., 2007, 47,
383.
Acknowledgements
The authors acknowledge the financial support from the MSM
002 162 7501 and MSM 002 162 7502.
References
1 C. H. Robert, Jr, and M. Ferid, in Goodman and Gilman’s The Phar-
macological Basis of Therapeutics, ed. P. B. Molinoff, R. W. Ruddon,
The McGraw-Hill Companies, Inc. Press, New York, 1996, pp. 1459–
1489.
2 M. Teshima, S. Kawakami, S. Fumoto, K. Nishida, J. Nakamura, M.
Nakashima, H. Nakagawa, N. Ichikawa and H. Sasaki, Biol. Pharm.
Bull., 2006, 29, 1436.
3 R. Mehvar, R. O. Dann and D. A. Hoganson, J. Controlled Release,
2000, 68, 53.
4 S. Penugonda, A. Kumar, H. K. Agarwal, K. Parang and R. Mehvar,
J. Pharm. Sci., 2008, 97, 2649.
5 K. Moribe, M. Fukino, Y. Tozuka, K. Higashi and K. Yamamoto,
Int. J. Pharm., 2009, 380, 201.
6 S. A. Gruber, Transplantation, 1992, 54, 1.
7 F. Hiyarama and K. Uekama, in Prodrugs: Challenges and Rewards,
ed. V. J. Stella, R. T. Borchardt, M. J. Hageman, R. Oliyai, H. Maag,
J. W. Tilley, Part 1 p. 683, Springer, 2007.
8 F. Hiyarama and K. Uekama, in Prodrugs: Challenges and Rewards,
ed. V. J. Stella, R. T. Borchardt, M. J. Hageman, R. Oliyai, H. Maag,
J. W. Tilley, Part 2 p. 351, Springer, 2007.
28 A. Harada, A. Hashidzume, H. Yamaguchi and Y. Takashima, Chem.
Rev., 2009, 109, 5974.
29 C. Wang, M. Zhao, X. Qiu and S. Peng, Bioorg. Med. Chem., 2004, 12,
4403.
30 I. N. Topicheva, A. E. Tonelli, I. G. Panova, E. V. Matuchina, F. A.
Kalashnikov, V. I. Gerasimov, C. C. Rusa, M. Rusa and M. A. Hunt,
Langmuir, 2004, 20, 9036.
31 S. Jiao, S. H. Goh and S. Valiyaveettil, Macromolecules, 2002, 35, 1980.
32 K. Miyake, S. Yasuda, A. Harada, J. Sumaoka, M. Komiyama and H.
Shigekawa, J. Am. Chem. Soc., 2003, 125, 5080.
33 H. Shigekawa, K. Miyake, J. Sumaoka, A. Harada and M. Komiyama,
J. Am. Chem. Soc., 2000, 122, 5411.
34 U. E. Majewska, K. Chmurski, K. Biesiada, R. Andrzej, A. R. Olszyna
and A. R. Bilewicz, Electroanalysis, 2006, 18, 1463.
35 Y. Liu, P. Liang, Y. Chen, Y.-M. Zhang, J.-Y. Zheng and H. Yue,
Macromolecules, 2005, 38, 9095.
36 L. Hu, in Drug Delivery Principles and Applications, ed. B. Wang,
T. Siahaan, R. A. Soltero, John Wiley & Sons, New Jersey, 2005, p.139.
37 R. N. V. S. Mamidi, R. Mullangri, J. Kota, R. Bhamidipati, A. A. Khan,
K. Katneni, S. Datla, S. K. Singh, K. Y. Rao, C. S. Rao, N. R. Srinivas
and R. Rajagopalan, R., Biopharm. Drug Dispos., 2002, 23, 273.
38 G. Tocco, M. Begala, G. Delogu, G. Meli, C. Picciau and G. Podda,
Synlett, 2005, 1296.
ˇ
9 M. Sedla´k, V. Buchta, L. Kubicova´, P. Simu˚nek, M. Holcˇapek and P.
Kasˇparova´, Bioorg. Med. Chem. Lett., 2001, 11, 2833.
10 M. Sedla´k, M. Pravda, L. Kubicova´, P. Mikulcˇ´ıkova´ and K. Ventura,
Bioorg. Med. Chem. Lett., 2007, 17, 2554.
11 M. Sedla´k, M. Pravda, F. Staud, L. Kubicova´, K. Ty´cˇova´ and K.
Ventura, Bioorg. Med. Chem., 2007, 15, 4069.
ˇ
12 M. Sedla´k, P. Drabina, E. B´ılkova´, P. Simu˚nek and V. Buchta, Bioorg.
Med. Chem. Lett., 2008, 18, 2952.
13 E. B´ılkova´, A. Imramovsky´, V. Buchta and M. Sedla´k, Int. J. Pharm.,
2010, 386, 1.
39 M. Sedla´k, P. Drabina, M. Svobodova´ and J. Hanusek, Synlett, 2008,
1230.
14 M. Sedla´k, Mini-Rev. Med. Chem., 2009, 9, 1306.
15 M. Sedla´k, Collect. Czech. Chem. Commun., 2005, 70, 269.
5430 | Org. Biomol. Chem., 2010, 8, 5423–5430
This journal is
The Royal Society of Chemistry 2010
©