Synthesis and evaluation of water-soluble poly(vinyl alcohol)-paclitaxel conjugate as a macromolecular prodrug

Article abstract of DOI:10.1248/bpb.31.963

Paclitaxel (PTX) is an antitumor agent for the treatment of various human cancers. Cremophor EL and ethanol are used to formulate PTX in commercial injection solutions, because of its poor solubility in water. However, these agents cause severe allergic r

Full text of DOI:10.1248/bpb.31.963

May 2008
Biol. Pharm. Bull. 31(5) 963—969 (2008)
963
Synthesis and Evaluation of Water-Soluble Poly(vinyl alcohol)–paclitaxel
Conjugate as a Macromolecular Prodrug
Atsufumi KAKINOKI, Yoshiharu KANEO,* Tetsuro TANAKA, and Yoshitsugu HOSOKAWA
Laboratory of Biopharmaceutics, Faculty of Pharmacy and Pharmaceutical Sciences, Fukuyama University; Fukuyama,
Hiroshima 729–0292, Japan. Received November 18, 2007; accepted January 31, 2008
Paclitaxel (PTX) is an antitumor agent for the treatment of various human cancers. Cremophor EL® and
ethanol are used to formulate PTX in commercial injection solutions, because of its poor solubility in water.
However, these agents cause severe allergic reaction upon intravenous administration. The aim of this study is to
synthesize water-soluble macromolecular prodrugs of PTX for enhancing the therapeutic efficacy. Poly(vinyl
alcohol) (PVA, 80 kDa), water-soluble synthetic polymer, was used as a drug carrier which is safe and stable in
the body. The 2ꢀ-hydroxyl group of PTX was reacted with succinic anhydride and then carboxylic group of the
succinyl spacer was coupled to PVA via ethylene diamine spacer, resulting the water-soluble prodrug of poly(vinyl
alcohol)–paclitaxel conjugate (PVA–SPTX). The solubility of PTX was greatly enhanced by the conjugation to
PVA. The release of PTX from the conjugate was accelerated at the neutral to basic conditions in in vitro release
experiment. [125I]-labeled PVA–SPTX was retained in the blood circulation for several days and was gradually
distributed into the tumorous tissue after intravenous injection to the tumor-bearing mice. PVA–SPTX inhibited
the growth of sarcoma 180 cells subcutaneously inoculated in mice. It was suggested that the water-solubility of
PTX was markedly enhanced by the conjugation to PVA, and PVA–SPTX effectively delivered PTX to the tumor-
ous tissue due to the enhanced permeability and retention (EPR) effect.
Key words poly(vinyl alcohol); paclitaxel; macromolecular prodrug; cytotoxicity; antitumor activity
Paclitaxel (PTX) is an anti-microtubule agent isolated from the body, changes their toxicity and immunogenicity,
from the trunk bark of the Pacific Yew tree, Taxus brevifo- and limits their uptake by cells via endocytosis, thus provid-
lia.1) It has been widely used as an anti-neoplastic agent for a ing the opportunity to direct the drug to the particular cell
variety of human cancers including breast, ovarian, non- type where its activity is needed.19) In addition, these macro-
small cell lung, head and neck cancers, leukemia, and molecular conjugates can accumulate in solid tumors due to
melanoma.2—6)
the enhanced microvasculature of tumor tissue.20,21) This phe-
PTX is a highly hydrophobic drug and is hardly soluble in nomenon has been termed enhanced permeability and reten-
water (water solubility ꢀ0.3 mg/ml).7) Because of its poor tion in relation to tumor targeting (EPR-phenomenon).22—24)
solubility in water and many other acceptable pharmaceutical
Poly(vinyl alcohol) (PVA) is a polymer which is synthe-
solvents, specific emulsionizers, such as Cremophor EL®, are sized by polymerizing not a vinyl alcohol monomer but a
used to formulate PTX in commercial injection solutions. vinyl acetate monomer. This monomer is polymerized in to
However, serious hypersensitivity reactions have been re- poly(vinyl acetate) and then hydrolyzed to produce PVA.
ported in some individuals since the content of Cremophor PVA’s biocompatibility makes it an excellent material for use
EL® used in the PTX formulation is significantly higher than in medical applications such as soft contact lenses. Recently,
in any other marketed drug.8,9) Side effects of PTX formula- PVA has been used for long-term implants, including a bioar-
tion include nausea and vomiting, diarrhea, mucositis, tificial pancreas, artificial cartilage, nonadhesive film, and
myelosuppression, cardiotoxicity and neurotoxicity.10,11)
esophagus or scleral buckling material.25) Furthermore, PVA
In addition, Cremophor EL® is known to leach phthalate has numerous functional groups which are capable of cova-
plasticizers from polyvinylchloride bags and intravenous tub- lently coupling drug molecules.
ing.12) Therefore, alternative dosage forms for the PTX
In the previous study, we first synthesized a PVA–doxoru-
administration need to be developed to reduce the undesir- bicin conjugate and demonstrated that PVA provides a poten-
able side effects induced by using Cremophor EL®. tial targetable drug delivery system.26) In this paper we report
In recent years the use of water-soluble polymers as drug on the synthesis of the poly(vinyl alcohol)–paclitaxel conju-
delivery systems has been received increasing attention. Li gate, its release experiment, cytotoxicity, tissue distribution,
et al.13) and Greenwald et al.14) reported the conjugation of and antitumor activity.
polyethylene glycol (PEG) to the 2ꢁ-position of PTX through
a spacer succinyl group. They demonstrated that PEG may MATERIALS AND METHODS
be used as an effective solubilizing carrier for PTX. Poly-L-
glutamic acid was also used to make a water-soluble PTX
Animals Male ddY mice at 5 weeks of age were pur-
conjugate.15—17) It was demonstrated that the poly-(L-glu- chased from Shimizu laboratory supplies (Shizuoka, Japan)
tamic acid)–paclitaxel conjugate was more effective than and housed under a standard condition of temperature and
standard PTX. On the other hand, Sugahara et al. constructed light. Mice were given free access to commercial food and
the PTX delivery system using amino acid linkers in the con- tap water. All animal experiments were conducted in accor-
jugation of PTX with carboxymethyldextran.18)
dance with the institutional guideline for the care and use of
The consequence of attachment of low molecular weight laboratory animals for research, which conforms to the
drugs to macromolecular carriers alters their rate of excretion guideline of Science Council of Japan.
∗ To whom correspondence should be addressed. e-mail: kaneo@fupharm.fukuyama-u.ac.jp
© 2008 Pharmaceutical Society of Japan

964
Vol. 31, No. 5
Materials Paclitaxel (PTX) was obtained from Tokyo system (pH 4.0, 5.0, 6.0, 7.0, 8.0 and 9.0, mꢂ0.15) contain-
Chemical Industry Co., Ltd., Japan. PVA (80 K, MWꢂ80520) ing 1.5 M N,N-diethylnicotinamide (DENA) at 37 °C. The
was kindly supplied by Japan Vam & Poval Co., Ltd., Osaka, experiment was initiated by the addition of the stock solution
Japan. All other chemicals and reagents were of the highest to a preheated buffer solution to give a concentration of 0.25
grade commercially available. Purebright MB37-50T, a solu- mg/ml of PVA–SPTX conjugate, respectively. At a fixed time
bilizer of an amphiphilic polymer, was purchased from NOF intervals, the amount of PTX released was determined by
Corporation, Tokyo.
using a HPLC method described later.
Preparation of Poly(vinyl alcohol)–Paclitaxel Conju-
In Vitro Cytotoxicity Mouse leukemia L1210 was
gate Ethylenediamine residues were introduced to the kindly provided by Institute of Development, Aging and
hydroxyl groups of PVA molecules by the 1,1ꢁ-carbonyldiim- Cancer, Tohoku University (Miyagi, Japan). Cells were typi-
idazole (CDI) activation method.26—29) Four milliliters of cally kept in continuous logarithmic growth at 37 °C in a
dimethyl sulfoxide (DMSO) containing CDI (74 mg) was humidified atmosphere in 5% CO2–95% air in RPMI 1640
added to 400 mg of PVA dissolved in 60 ml of DMSO, fol- medium (Nacalai, Kyoto, Japan), supplemented with 10%
lowed by stirring for 1 h at room temperature. After several heat-inactivated fetal bovine serum (FBS) (Gibco, Tokyo,
precipitations in butanol to remove unreacted reagents, the Japan) and 50 U/ml penicillin and 50 mg/ml streptomycin
fraction of CDI-activated PVA was dried in vacuo. Then, eth- (Dainippon Sumitomo Pharma Co., Ltd., Tokyo, Japan). The
ylenediamine (2 g) was added to the CDI-activated PVA number of viable cells was determined by the trypan blue
(200 mg) dissolved in 50 ml of DMSO and stirred for 48 h at exclusion method by using a Burker-Turk hematocytometer 3
50 °C. After several precipitations in butanol to remove unre- d after incubation.
acted reagents, the fraction of PVA–ethylenediamine was
PVA–SPTX and PTX in Purebright MB37-50T were dis-
dried in vacuo. The free amino groups of ethylenediamine solved in RPMI1640 medium. PTX was dissolved in DMSO,
spacers introduced into the PVA molecule were measured by then diluted by the tissue culture medium. The solutions of
the 2,4,6-trinitrobenzenesulfonic acid (TNBS) method.30—32)
each sample were added to test wells to a final volume of
2ꢁ-Succinyl-paclitaxel (SPTX) was synthesized by the 200 ml/well in 96-well plates. Cells were seeded into the
modified method of Dosio et al.33) Briefly, PTX (20 mg, plates at a density of 0.5ꢃ104 cells/200 ml/well. Plates were
0.023 mmol) was added to succinic anhydride (14 mg, 0.14 incubated at 37 °C for 3 d. Subsequently, cellular prolifera-
mmol) in the presence of 4-dimethylamino-pyridine (0.88 tion was examined by an additional incubation for 3 h with
mg, 0.0072 mmol). Then 0.4 ml of dry pyridine was added Alamar Blue (Kanto Reagents, Tokyo, Japan).34) Cellular
and the solution was stirred for 3 h at room temperature. The proliferation induces a chemical reduction of the Alamar
SPTX was purified by chromatography on a 30ꢃ2.5 cm SiO2 Blue which results in a change in redox color from blue to
column eluted with chloroform–methanol mixture (97 : 3 to red. The proliferation of cultures with Alamar Blue was
90 : 10) as determined by TLC (Rf 0.27 in chloroform– determined by measuring absorbance at 570 nm and 600 nm
1
methanol 90 : 10). Its structure was confirmed by H-NMR in an immuno plate reader (NJ-2300, Inter Med Japan,
(500 MHz, CDCl3) d: S 1.13 (s, 17H), 1.22 (s, 16H), 1.67 (s, Tokyo, Japan). Growth inhibition (%) was calculated as fol-
19H), 1.90 (s, 18H), 2.17 (m, 14H), 2.20 (s, 4-OAc), 2.43 (s, lows.
10-OAc), 2.57 (m, CH2CH2), 3.80 (d, 3H), 4.19 and 4.30 (d,
20H), 4.42 (m, 7H), 4.96 (d, 5H), 5.52 (d, 2ꢁH), 5.68 (d, 2H),
growth inhibition (%)ꢂ(1ꢄ(TꢄB)/(CꢄB))ꢃ100
5.98 (dd, 3ꢁH), 6.23 (t, 13H), 6.29 (s, 10H), 7.10 (d, NH), where T, B, and C represent the scaled difference in ab-
7.33 (m, 3ꢁ-Ph), 7.40 (m, 3ꢁ-NBz), 7.50 (m, 2-OBz), 7.75 (d, sorbance at 570 nm and 600 nm of the test compound, the
3ꢁ-NBz), 8.13 (d, 2-OBz).
The SPTX (21.0 mg, 0.022 mmol), dissolved in DMSO–
blank and the control, respectively.
Preparation of [125I]-Labeled PVA–SPTX ([125I]-PVA–S
dimethylformamide (70 : 30) solution containing N-hydroxy- PTX) PVA–SPTX (200 mg) was labeled with 0.25 mCi of
3-sulfo-succinimide (sulfo-NHS, Fluka Chimica, Milan,
[
125I] iodine by using Bolton and Hunter reagent for protein
Italy) (9.7 mg, 0.044 mmol) and 1-(3-dimethylaminopropyl)- iodination (GE Healthcare Bio-Sciences, Tokyo, Japan). Un-
3-ethylcarbodiimide hydrochloride (EDC) (8.6 mg, 0.044 reacted [125I] was removed by chromatography on a PD-10
mmol). After 14 h at room temperature the reaction was column (Amersham Pharmacia Biotech).
complete as determined by TLC (Rfꢂ0.18 in chloroform–
Tissue Distribution in Tumor-Bearing Mice S180 cells
methanol, 90 : 10). Then, PVA–ethylenediamine (200 mg) (1ꢃ106 cells/mouse) were inoculated into the subcutaneous
dissolved in 20 ml DMSO was added to the activated- tissue of the axillary region of ddY mice. Fourteen days after
SPTX and stirred for 48 h at room temperature. After several the inoculation, mice were injected with [125I]-PVA–SPTX
precipitations in butanol to remove unreacted reagents, the (6 mg/kg) in 0.2 ml of saline through the tail vein. Aliquots
fraction of PVA–SPTX conjugate was dried in vacuo. Char- of the sample solution were stored for the calibration of the
acterization of the conjugate was carried out using a three di- disintegration of radioiodine. At appropriate times after the
mensional high-performance size exclusion chromatography administration, blood was collected from the vena cava under
described later.
ether anesthesia and the tumor tissue, liver, spleen, kidney,
PTX Content of the Conjugate The conjugate was dis- lung, and heart were excised and weighed. The radioactivity
solved in distilled water, and the absorbance at 227 nm was was determined with a gamma counter (Aloka 301).
measured. The PTX content of the conjugates was estimated
using the calibration curve of PTX standard.
In Vivo Antitumor Activity S180 cells (1ꢃ106 cells/
mouse) were inoculated into the subcutaneous tissue of the
In Vitro Release Experiment The release of PTX from axillary region of ddY mice. PVA–SPTX (25 mg/kg in PTX
the conjugate was determined in a 0.05 M phosphate buffer equivalents) or PTX (25 mg/kg) was given as intermittent

May 2008
965
intravenous injections on days 7, 10, 14, 17 and 21 after was the preparation of PVA–ethylenediamine, the second the
tumor inoculation. Saline was used as a negative control. preparation of SPTX and the third the binding of SPTX and
Tumor sizes were measured with calipers, and the tumor vol- PVA–ethylenediamine.
umes (V) were calcualted as: VꢂLꢃW2/2, where L and W
are length (mm) and width (mm), respectively.
Hydroxyl group of PVA was activated by CDI. Then, a 10-
fold weight excess of ethylenediamine was reacted with CDI-
Analytical Methods The amount of PTX was deter- activated PVA. An excess of the ethylenediamine was used in
mined by HPLC. Chromatography was carried out using a order to prevent crosslinking and cyclisation of the PVA
Shimadzu liquid chromatographic system (LC-6A, Kyoto, chains. The number of free amino groups of the PVA–ethyl-
Japan) with a variable-wavelength UV detector (SPD-6A, enediamine was estimated by the TNBS method.30—32) One
Shimadzu) operated at 227 nm. A 4.6ꢃ150 mm, 5-mm parti- mole of PVA showed the color intensity of 25 mol of free
cle size, C18 reversed-phase column (TSK gel ODS-80TM, amino group.
Tosho, Japan) was used at ambient temperature. The mobile
SPTX was synthesized by the modified method of Dosio
phase was a mixture of acetonitrile and 2 mM phosphoric acid et al.33) The crude SPTX was purified by chromatography on
1
(55 : 45, v/v). The injection volume was 20 ml, and the flow a SiO2 column. It was confirmed by the H-NMR analysis
rate was 1.0 ml/min. For every experimental sample the con- that the hydroxyl group of 2ꢁ position of PTX was reacted
tent of PTX was calculated by measuring the relevant peak with succinic anhydride. The yield of SPTX was 94.0%.
area and calibrating against the corresponding peak area de-
rived from the n-hexyl p-hydroxybenzoate internal standard.
Free carboxylic group of SPTX molecule was activated
with twice molar of EDC and sufo-NHS, and then reacted
Three dimensional high-performance size-exclusion chro- with PVA–ethylenediamine. The yield of PVA–SPTX was
matography was carried out using a Shimadzu liquid chro- 91.4%. The PTX content of the conjugates estimated by the
matographic system (LC-9A, Kyoto, Japan) equipped with a UV method was 6.2 w/w%, corresponding to 5.8 mol of
variable-wavelength UV detector (MCPD-3600, Otsuka, PTX/mol of PVA.
Osaka, Japan). A 7.8ꢃ300 mm, TSKge1 G4000PWXL col-
Size-exclusion chromatography of the conjugate was per-
umn (Tosoh) was used at 40 °C. The mobile phase was 20% formed by HPLC on an TSK gel G4000PWXL column. Elu-
acetonitrile in 50 mM LiCl. The injection volume was 50 ml, tion peaks were detected by a photodiode array detector. The
and the flow rate was 1.0 ml/min.
retention time of the peak top of PVA–SPTX was 8.4 min in
the cross section at 227 nm (Fig. 2). PVA–SPTX eluted at the
same retention time to the original PVA detected by a differ-
ential refractometer.
RESULTS
Preparation of PVA–SPTX Conjugate PTX was bound
Kinetics of Regeneration of PTX from the Conjugate
to PVA according to the synthetic scheme shown in Fig. 1. The degradation of PTX was investigated in citrate and phos-
The overall synthetic pathway involved three steps. The first phate buffer solutions containing 1.5 M DENA over the pH
Fig. 1. Synthetic Pathway of PVA–SPTX Conjugate
EDC: 1-(3-dimethylaminopropyl)3-ethylcarbodiimide hydrochloride, sulfo-NHS: N-hydroxy-3-sulfo-succinimide.

966
Vol. 31, No. 5
Fig. 2. Three Dimensional Chromatogram of PVA–SPTX
The retention time of the peak top of PVA–SPTX was 8.4 min in the cross section at 227 nm. High-performance size-exclusion chromatography was carried out using a HPLC
system equipped with a photodiode array detector. A 7.8ꢃ300 mm, TSKge1 G4000PWXL column was used at 40 °C. The mobile phase was 20% acetonitrile in 50 mM LiCl and
the flow rate was 1.0 ml/min.
Fig. 3. Stability of PTX in 0.05 M Phosphate Buffer Solutions (mꢂ0.15)
Containing 1.5 M DENA of pH 7.0 (ꢀ), pH 8.0 (ꢁ) and pH 9.0 (ꢂ) at 37 °C
Fig. 4. Regeneration of PTX from PVA–SPTX in 0.05 M Citrate Buffer
Solutions (mꢂ0.15) of pH 4.0 (ꢃ), 5.0 (ꢄ) or 6.0 (ꢅ) and in 0.05 M Phos-
phate Buffer Solutions (mꢂ0.15) of pH 7.0 (ꢀ), 8.0 (ꢁ) or 9.0 (ꢂ) Contain-
range 4—9 at 37 °C. At pH 7—9 the degradation of PTX fol- ing 1.5 M DENA at 37 °C
lowed pseudo first-order kinetics as shown in Fig. 3, whereas
PTX was quite stable at acidic conditions (pH 4—6). Time The concentration of PVA–SPTX to produce a 50% inhibi-
courses for PTX regenerated after incubation of PVA–SPTX tion of normal cell growth (IC50) was 13.8 ng/ml (equivalent
in the buffer solutions are shown in Fig. 4. The apparent first- to PTX), whereas the IC50 of PTX in DMSO was 3.0 ng/ml.
order rate constants for regeneration of PTX from PVA–
Tissue Distribution in Tumor-Bearing Mice Mice
SPTX (k1) and for decomposition of PTX (k2) were estimated were injected with [125I]-PVA–SPTX (6 mg/kg). At appropri-
by the curve fitting using a nonlinear least squares program ate times until 7 d, the radioactivities in a variety of organs
(MULTI)35) and are depicted in Fig. 5. The profiles of k1 and were determined with a gamma counter. Figure 7 shows the
k2 showed pH-dependency in the investigated pH range.
time profile of the tissue distribution of [125I]-PVA–SPTX
In Vitro Cytotoxicity In vitro biological efficacy of the after intravenous injection to mice. It was found that [125I]-
soluble PVA–SPTX was tested using murine leukemia cell PVA–SPTX was retained in the blood circulation for several
line L1210. The cytotoxicities of PVA–SPTX and PTX are days and was gradually distributed into the tumorous tissue.
shown in Fig. 6. As is evident, the cytotoxic effect of these
In Vivo Antitumor Activity Figure 8 shows the growth
compounds on L1210 cells was concentration dependent. inhibition curves against S180 subcutaneously inoculated in
PTX in DMSO showed the highest activity, whereas both mice. To determine the effect of these drugs on S180 tumor
PVA–SPTX and PTX in Purebright MB37-50T exhibited growth, PVA–SPTX (25 mg/kg in PTX equivalents) or PTX
similar activity to that of the positive control of doxorubicin. (25 mg/kg) was given as intermittent intravenous injections

May 2008
967
Fig. 5. pH Profile of Regeneration Rate Constant of PTX from
PVA–SPTX (ꢂ) and Decomposition Rate Constant of PTX (ꢀ) in 0.05 M
Citrate Buffer Solutions (mꢂ0.15) or Phosphate Buffer Solutions (mꢂ0.15)
Containing 1.5 M DENA at 37 °C
Fig. 8. Antitumor Effect of PVA–SPTX and PTX on Implanted S180
Tumor
Mice were inoculated with S180 cells (1ꢃ106 cells/mouse) subcutaneously.
PVA–SPTX and free PTX were given as intermittent intravenous injections on days 7,
10, 14, 17 and 21 after the tumor inoculation. ꢀ, PVA–SPTX (25 mg/kg in PTX equiv-
alents); ꢁ, PTX (25 mg/kg); ꢂ, control (5% glucose). Each point represents the
meanꢅS.E. of six mice. Student’s t-test was performed between control and PVA–
SPTX, ∗ pꢀ0.05.
Chart 1
on days 7, 10, 14, 17 and 21 after tumor inoculation. PVA–
SPTX caused significant tumor growth inhibition compared
to the control. The control group of mice showed a progres-
sive increase in tumor growth with the mean tumor volume
increasing to 4278ꢅ1426 mm3 on day 30 after tumor inocu-
lation. The mice treated with PVA–SPTX showed significant
tumor growth inhibition (mean tumor volume 329ꢅ187
mm3) compared to the control group, which was almost simi-
lar to or less than the PTX-treated group (mean tumor vol-
ume 1304ꢅ864 mm3). In this experiment, although both
PTX and PVA–SPTX did not cause the side effect of body
weight loss at 25 mg/kg in PTX equivalents, the macromole-
cular prodrug was expected to show the tolerability at the
higher dose.
Fig. 6. Cytotoxicity of PVA–SPTX, PTX, and Doxorubicin against L1210
Cells
L1210 cells were cultivated in the medium containing each compound for 3 d, and
the cells viability was determined by the alamar blue method. ꢀ, PVA–SPTX; ꢁ, PTX
(dissolved in Purebright MB37-50T solution); ꢂ, PTX (dissolved in DMSO); ꢃ,
doxorubicin. Values are given as meansꢅS.D. of three experiments.
DISCUSSION
The kinetics of regeneration of PTX from the conjugate
was examined in citrate and phosphate buffer solutions con-
taining 1.5 M DENA over the pH range 4—9 at 37 °C. Since
the degradation of PTX followed pseudo first-order kinetics,
the overall reactions were described by Chart 1, where k1, k2
and k3 are apparent first-order rate constants for the depicted
reactions.
Then the concentration of PTX ([PTX]) has a time
dependence given by the following equation:
Fig. 7. Time Profile of Tissue Distribution of 125I-PVA–SPTX after Intra-
venous Injection (6 mg/kg) to Mice
[PTX]ꢂk1[PTX]*/(k2ꢄ(k1ꢆk3))ꢃ(exp(ꢄ(k1ꢆk3)t)ꢄexp(ꢄk2t))
(1)
ꢀ, plasma; ꢁ, liver; ꢆ, kidney; ꢂ, tumor; ꢃ, spleen; ꢄ, lung; ꢅ, heart. Each point
represents the meanꢅS.D. of three mice.
where [PTX]* is the initial concentration of PTX covalently

968
Vol. 31, No. 5
bound to PVA–SPTX. On the other hand, the stability of PTX as forming a polymeric lipid nanosphere.38) As shown
PTX is expressed by the following equation:
in Fig. 6, PVA–SPTX inhibited the proliferation of L1210
cells to the same extent as PTX dissolved in the Purebright
MB37-50T. The cytotoxicity of PTX in the Purebright
[PTX]ꢂ[PTX]0 exp(ꢄk2t)
(2)
where [PTX]0 is the initial concentration of PTX. At first the MB37-50T was similar to that reported by Wada et al.39)
parameters of k2 were calculated from the observed time
In the previous work, we examined the pharmacokinetics
courses shown in Fig. 3 using Eq. 2. Equation 1 was then and biodisposition of PVA in experimental animals. [125I] la-
fitted to the observed time courses shown in Fig. 4 using a beled PVA was retained in the blood circulation for several
nonlinear least squares program (MULTI).35)
days after intravenous injection to mice. Although the tissue
The model suggested in Chart 1 suffered the divergence in distribution of PVA was small, a significant accumulation in
most of the computations; even if it was converged, the nega- the liver, kidney and spleen was observed.27) In this study,
tive value of k3 with an extraordinarily large standard devia-
[
125I]-PVA–SPTX was also retained in the blood circulation
tion was obtained. When it was assumed that the degrada- very well and was gradually accumulated in the tumorous tis-
tionof PTX bound to PVA is negligible, however, the excel- sue as shown in Fig. 7. These findings indicated that PVA–
lent convergence could be obtained in the curve fitting using SPTX was accumulated efficiently in the tumorous tissue by
the algorithm of Gauss–Newton method where the following EPR effect.22—24)
equation was adopted instead of Eq. 1.
In conclusion, in this study the conjugate was generated
from the combination of PVA and PTX for the first time. It
was suggested that the water-solubility of PTX was markedly
[PTX]ꢂk1[PTX]*/(k2ꢄk1)ꢃ(exp(ꢄk1t)ꢄexp(ꢄk2t))
(3)
The apparent first-order rate constants for regeneration of enhanced by the conjugation to PVA, and the conjugate,
PTX from PVA–SPTX (k1) and for decomposition of PTX PVA–SPTX, effectively delivered PTX to the tumorous tis-
(k2) are depicted in Fig. 5. The profiles of k1 and k2 showed sue due to the EPR effect. These studies demonstrated that
pH-dependency in the investigated pH range. PTX was very PVA may be used as an effective solubilizing carrier for
stable at acidic conditions, but the decomposition was in- PTX.
creased with the elevation of the pH. It was reported that
PTX was susceptible to mild basic hydrolysis which resulted
Acknowledgements This work was supported in part by
in the formation of baccatin III as the major product.36) The Grants in Aid (No. 13672406) for Scientific Research (C)
regeneration rate of PTX was also enhanced with the rise in from the Ministry of Education, Culture, Sports, Science and
pH and reached a maximum at pH 9. At pH 7, the regenera- Technology, Japan (to YK). The authors are also grateful to
tion of PTX (k1ꢂ0.187 hꢄ1) was 54-times faster than the Mr. Hiroshi Noguchi of Japan Vam & Poval Co., Ltd., Osaka,
degradation of PTX (k2ꢂ0.00345 hꢄ1), showing half-lives of Japan for the gift of the PVA samples. The authors particu-
3.7 and 201 h, respectively. These findings indicate that cova- larly acknowledge T. Sunagari and D. Ishii for their technical
lent binding to PVA stabilizes PTX (k3ꢂ0) and the macro- assistance.
molecular prodrug, PVA–SPTX, releases PTX gradually at
physiological conditions.
REFERENCES
In addition, the amount of PTX covalently bound to
PVA–SPTX was estimated by the value of [PTX]*, which
was similar to that measured by the UV method. These find-
ings indicated that approximately 100% of PTX was released
from the conjugate.
1) Wani M. C., Taylor H. L., Wall M. E., Coggon P., McPhail A. T., J.
Am. Chem. Soc., 93, 2325—2327 (1971).
2) Arbuck S. G., Christian M. C., Fisherman J. S., Cazenave L. A.,
Sarosy G., Suffness M., Adams J., Canetta R., Cole K. E., Friedman
M. A., J. Natl. Cancer Inst. Monogr., 15, 11—24 (1993).
3) Cahan M. A., Walter K. A., Colvin O. M., Brem H., Cancer
Chemother. Pharmacol., 33, 441—444 (1994).
4) Foa R., Norton L., Seidman A. D., Int. J. Clin. Lab. Res., 24, 6—14
(1994).
5) Long H. J., Mayo Clin. Proc., 69, 341—345 (1994).
6) Wiernik P. H., Einzig A. I., J. Natl. Cancer Inst. Monogr., 15, 185—
187 (1993).
7) Lee J., Lee S. C., Acharya G., Chang C. J., Park K., Pharm. Res., 20,
1022—1030 (2003).
8) Fjallskog M. L., Frii L., Bergh J., Lancet, 342, 8873—8875 (1993).
9) Panchagnula R., Int. J. Pharm., 172, 1—15 (1998).
10) Einzig A. I., Wiernik P. H., Sasloff J., Runowicz C. D., Goldberg G. L.,
J. Clin. Oncol., 10, 1748—1753 (1992).
11) Rogers B. B., Oncol. Nurs. Forum, 20, 1483—1489 (1993).
12) Goldspiel B. R., Ann. Pharmacother., 28, S23—S26 (1994).
13) Li C., Yu D., Inoue T., Yang D. J., Milas L., Hunter N. R., Kim E. E.,
Wallace S., Anticancer Drugs, 7, 642—648 (1996).
In the in vitro release experiment, we observed that PTX
released from the conjugate was precipitated as a white pow-
der in buffer solutions. Consequently 1.5 M DENA was added
to the release medium providing a chemical sink condition
for the release of PTX from PVA–SPTX. Baek et al. reported
that the PTX released from poly(lactic-co-glycol acid) (PLGA)
matrices was much faster in DENA solution than in serum or
phosphate buffer solution. They suggested that the presence
of DENA in the release medium increased the hydrolysis rate
of PLGA polymers.37) However, in this study the regenera-
tion rate of PTX from PVA–SPTX was not accelerated in
1.5 M DENA, being similar to those obtained in the ordinary
buffer solutions, because PVA was stable in the 1.5 M DENA
solution (data are not shown).
The in vitro cytotoxicity of PVA–SPTX was evaluated by
using L1210 cells. PTX previously dissolved in a small
amount of DMSO was a little more effective than the other
compounds. Furthermore, one of the amphiphilic 2-
methacryloxyethyl phosphorylcholine (MPC) polymers,
Purebright MB37-50T, was also used as the solubilizer of
PTX. It was reported that Purebright MB37-50T dissolved
14) Greenwald R. B., Gilbert C. W., Pendri A., Conover C. D., Xia J., Mar-
tinez A., J. Med. Chem., 39, 424—431 (1996).
15) Li C., Price J. E., Milas L., Hunter N. R., Ke S., Yu D. F., Charnsan-
gavej C., Wallace S., Clin. Cancer Res., 5, 891—897 (1999).
16) Li C., Yu D. F., Newman R. A., Cabral F., Stephens L. C., Hunter N.,
Milas L., Wallace S., Cancer Res., 58, 2404—2409 (1998).
17) Singer J. W., Baker B., De Vries P., Kumar A., Shaffer S., Vawter E.,
Bolton M., Garzone P., Adv. Exp. Med. Biol., 519, 81—99 (2003).

May 2008
969
18) Sugahara S., Kajiki M., Kuriyama H., Kobayashi T. R., Biol. Pharm.
Bull., 25, 632—641 (2002).
19) Ringsdorf H., J. Polym. Sci. Polym. Symp., 51, 135—153 (1975).
20) Seymour L. W., Crit. Rev. Ther. Drug Carrier Syst., 9, 135—187
(1992).
28) Hearn M. T., Methods Enzymol., 135, 102—117 (1987).
29) Beauchamp C. O., Gonias S. L., Menapace D. P., Pizzo S. V., Anal.
Biochem., 131, 25—33 (1983).
30) Habeeb A. F., Anal. Biochem., 14, 328—336 (1966).
31) Fields R., Biochem. J., 124, 581—590 (1971).
21) Yuan F., Dellian M., Fukumura D., Leunig M., Berk D. A., Torchilin V. 32) Fields R., Methods Enzymol., 25, 464—468 (1972).
P., Jain R. K., Cancer Res., 55, 3752—3756 (1995).
22) Maeda H., Adv. Enzyme Regul., 41, 189—207 (2001).
33) Dosio F., Brusa P., Crosasso P., Arpicco S., Cattel L., J. Controlled Re-
lease, 47, 293—304 (1997).
23) Maeda H., Wu J., Sawa T., Matsumura Y., Hori K., J. Controlled Re- 34) de Fries R., Mitsuhashi M., J. Clin. Lab. Anal., 9, 89—95 (1995).
lease, 65, 271—284 (2000).
35) Yamaoka K., Tanigawara Y., Nakagawa T., Uno T., J. Pharmacobio-
24) Maeda H., Matsumura Y., Crit. Rev. Ther. Drug Carrier Syst., 6, 193—
210 (1989).
Dyn., 4, 879—885 (1981).
36) Ringel I., Horwitz S. B., J. Pharmacol. Exp. Ther., 242, 692—698
25) Mallapragada S. K., McCarthy-Schroeder S., “Handbook of Pharma-
(1987).
ceutical Controlled Release Technology,” Marcel Dekker, New York, 37) Baek N., Lee J., Park K., J. Biomater. Sci. Polym. Ed., 15, 527—542
2000, pp. 31—46.
(2004).
26) Kakinoki A., Kaneo Y., Ikeda Y., Tanaka T., Fujita K., Biol. Pharm.
Bull., 31, 103—110 (2008).
38) Konno T., Watanabe J., Ishihara K., J. Biomed. Mater. Res. A, 65,
209—214 (2003).
27) Kaneo Y., Hashihama S., Kakinoki A., Tanaka T., Nakano T., Ikeda Y., 39) Wada M., Jinno H., Ueda M., Ikeda T., Kitajima M., Konno T., Wata-
Drug Metab. Pharmacokinet., 20, 435—442 (2005). nabe J., Ishihara K., Anticancer Res., 27, 1431—1435 (2007).