| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,1
* The Faculty of Biotechnology and Food Engineering, The Technion-Israel Institute of Technology, Haifa, Israel;
Department of Neurosurgery, Brigham and Women’s Hospital and Children’s Hospital, Harvard Medical School, Boston, Massachusetts, USA; and
Department of Neurosurgery, Seoul National University Hospital, Seoul, Korea
2Correspondence: M.M., Department of Biotechnology and Food Engineering, Technion-Israel Institute of Technology, Haifa, Israel. E-mail: machlufm{at}tx.technion.ac.il; R.S.C., Department of Neurosurgery, Brigham and Women’s Hospital, 221 Longwood Ave., Boston, MA 02115, USA. E-mail: rcarroll{at}rics.bwh.harvard.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: angiogenesis • brain tumor • PLGA • local delivery
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, which affect different pathways of the angiogenesis process, have been shown to inhibit gliomas. However, these studies and recent preclinical trials show that high inhibitor concentrations at the tumor site are needed to achieve and maintain therapeutic efficacy. Daily high-dose administration is thus required for an extended period of time. Moreover, the route of administration (systemic vs. local) and pattern of injection (continuous vs. bolus) are still in dispute when comparing preclinical and clinical outcome studies. For brain tumor treatment these obstacles are further emphasized, and several important issues need to be addressed, including the local delivery of an agent that permeates the tumor bed at a sufficiently elevated therapeutic concentration for an extended period of time to maintain tumor inhibition. Local delivery of drugs for brain tumor treatment has many advantages, including bypassing the blood brain barrier, localizing the drug at the tumor bed, and reaching tumor cells which have migrated into the normal brain parenchyma (2)
. Such an approach avoids systemic side effects and increases patient compliance.
We have recently demonstrated that the growth of subcutaneous tumors can be reduced by endogenous inhibitors of angiogenesis administered locally using a controlled delivery system in the shape of microspheres. The microspheres were composed of PLGA, which is approved for use in humans by the U.S. Food and Drug Administration (3)
. PLGA polymeric microspheres have been used as a sustained delivery system of many proteins, drugs, and others factors, such as cytokines, hormones, enzymes, and vaccines (4
5
6
7
8)
. Due to their size and shape, polymeric microspheres can be implanted easily by stereotactic injection at the tumor site without the need for a craniotomy and damage to the surrounding brain tissue.
The majority of polymeric studies have characterized the release kinetics and degradation rate of the microspheres by in vitro studies only. It is crucial to understand the difference between these parameters in vitro and in vivo to enable the prediction of the drug release and distribution intracranially. Therefore, the efficacy of this system and the success of this approach needs to be demonstrated in an in vivo orthotopic tumor model. PF-4 is a 70 amino acids protein that inhibits angiogenesis (9
, 10)
. PF-4/CTF is a 23 amino acid C-terminal fragment of PF-4 conserving the antiangiogenic properties of the PF-4 (11
, 12)
, which effectively inhibits glioma tumor growth in mouse models (12
, 13)
. In our studies, PF-4/CTF was labeled with rhodamine, which enables the tracking of the release kinetics of peptide from the microspheres in vitro and in vivo and its distribution in vivo without the need for indirect studies with an agent that can differ in both biological and physical properties. The current study demonstrates for the first time the therapeutic efficacy of tailored PLGA microspheres loaded with the angiogenic inhibitor PF-4/CTF for the inhibition of intracranial human glioma growth.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
using the solid-phase method applying the 9-fluorenylmethyloxycarbonyl Fmoc active ester chemistry on a fully automated peptide synthesizer (model 433A, Applied Biosystems, Foster City, CA, USA), according to the manufacturer’s instructions. Four-methylbenzhydrylamine (MBHA) resin (Novabiochem, Darmstadt, Germany) was used to obtain amidated peptide. Peptide cleavage from the resin was performed in a mixture of 2% triisopropylsilane, 5% phenol, 5% water, and 88% trifluoroacetic acid (TFA). TFA was evaporated using nitrogen gas. The cleaved peptide was precipitated in ice-cold diethyl ether, placed in a fume hood to dry at 60°C for 2 h, dissolved in water for separation from the resin, and lyophilized. The crude peptide was purified (up to >95% purity) by reverse-phase high-performance liquid chromatography (HPLC; Alliance-ZQ, Waters, Milford, MA, USA). The purified peptide was subjected to mass spectrometry analysis in order to confirm its purity. Peptides were stored as lyophilized powder at –20°C.
PF-4/CTF labeling with rhodamine
PF-4/CTF was labeled via conjugation of rhodamine to the peptide’s N-terminal amino group by exposing the resin-bound peptide (20 mg, 
6 µmol) to 2-fold molar excess of Lissamine rhodamine B sulfonyl chloride (Molecular Probes, Eugene, OR, USA) in dry dimethylformamide (DMF) containing 4-fold molar excess of diisopropylethylamine. After 24 h of incubation, the resin-bound peptide was washed thoroughly with DMF and diethyl ether/dichloromethane (1:1) and dried at 40°C for 4 h. The peptide was then cleaved from the resin, precipitated, extracted, and purified as described above.
Preparation of protein-loaded microspheres
PF-4/CTF-loaded microspheres were prepared as described previously (3)
. Briefly, PLGA (200 mg, RG502 capped and RG502H uncapped, Boehringer Ingelheim, Ingelheim, Germany) with 50:50 lactic-to-glycolic acid ratio was dissolved in 0.5 ml dichloromethane with or without fluorescent 6-coumarin (Sigma-Aldrich, St. Louis, MO, USA). A predetermined solution of PF-4/CTF (1% w/w) was added to the dissolved polymer, and the solution was homogenized using ultraturax (type DI-18, IKA, Staufen, Germany) for 1 min leading to the formation of the first emulsion (W/O). Polyvinyl alcohol (PVA 0.1% w/v) of 85–89 kDa (Sigma-Aldrich) saturated with dichloromethane was rapidly added to the first emulsion, and the solution was homogenized again for the second time. The resulting multiple W/O/W were mixed for 5 min, and a volume of 50 ml of 0.1% w/v aqueous PVA containing 5% (v/v) 2-propanol solution was added to the W/O/W double emulsion. After 30 min of extensive stirring, the microspheres were centrifuged and washed three times. After the final wash, the microspheres were lyophilized (Modulo Edwards, Crawley, UK), resulting in a fine powder of dry PLGA microspheres that contained the PF-4/CTF protein. Empty microspheres were prepared in the same way without PF-4/CTF.
Morphological studies with SEM and size distribution
Microspheres were prepared and incubated in PBS for 1, 10, and 30 days in a shaking incubator at 37°C. At predetermined time points, the microspheres were lyophilized, mounted on an aluminum stub, and sputter-coated with a thin gold layer. Microsphere surface morphology and morphological changes during the degradation process in vitro were determined using SEM (JSM 5400, Jeol, Akishima, Japan). Size distribution of PF-4/CTF-loaded microspheres was analyzed using a Coulter LS 230 particle size analyzer (Beckman Coulter, Fullerton, CA, USA). The results were reported as a volume size distribution by a mathematical computerized analysis for ideal spheres.
Protein loading efficiency
Fractions of 10 mg microspheres loaded with the peptide were digested overnight with 0.1 N NaOH containing 5% sodium dodecyl sulfate (SDS). NaOH increases PLGA hydrolysis rate, and SDS ensures the complete solubility of the protein. The MicroBCA protein assay kit (Pierce, Rockford, IL, USA) was used to determine total protein loading in the microspheres. The loading efficiency was obtained by calculating the percent of total PF-4/CTF loaded in the microspheres divided by the initial protein amount added during the preparation of the microspheres.
In vitro release of PF-4/CTF from PLGA microspheres
For the in vitro release studies, 20 mg microspheres (capped and uncapped PLGA), loaded with 200 µg PF-4/CTF, were incubated in PBS at pH 7.3 and maintained in a shaking incubator at 37°C. Every 2–3 days over a 30-day period, microspheres were recovered from the release media by centrifugation, and media were collected and replaced with fresh media. PF-4/CTF concentration in the media was quantitatively determined using a MicroBCA protein assay. The percent of released protein of the total loaded protein was calculated at each time point.
In vitro microsphere degradation studies
PF-4/CTF-rhodamine was loaded into PLGA 50:50 capped microspheres. The polymer was labeled with fluorescent marker 6-coumarin. Microspheres from the same preparation procedure were divided into 20 mg aliquots. The microspheres were incubated under shaking conditions in PBS, pH 7.3, and 37°C. After 1, 7, 14, and 30 days of incubation, the microspheres were mounted and visualized with confocal microscopy. The micrographs (n=6) were analyzed using image-analysis software (Lucia G, Nikon, Badhoevedorp, Netherlands). The results were depicted as peptide-to-polymer ratio (red-to-green areas), which gives an indication of PF-4/CTF release and drug-to-polymer ratio.
Cell culture
The human glioblastoma cell line U87-MG was obtained from the American Type Culture Collection (Manassas, VA, USA). U87-MG was cultured as described previously (15)
. Human umbilical vein endothelial cells (HUVECs) were isolated from umbilical veins as described previously (16)
.
Cell proliferation assays
Proliferation assays were performed on HUVECs using a tritiated thymidine assay as described previously (17)
. HUVECs (20,000 cells/well) were seeded on 24-well plates coated with gelatin, incubated overnight with medium 199 (Biological Industries, Bet Haemek, Israel) supplemented with 10% fetal calf serum, glutamine, 1% (v/v) penicillin/streptomycin, and 2 ng/ml of bFGF. After 24 h, media were replaced with media containing 5% serum and 250 µl of the released medium taken from the different incubation periods of the microspheres containing PF-4/CTF-rhodamine. Medium from empty microspheres was used as controls. After 48 h, [3H]-thymidine (1 µCi/ml) was added, and cells were incubated for an additional 24 h in 37°C, 10% CO2. Cells were washed with PBS and lysed with 0.2 N NaOH for 20 min. Samples were suspended in liquid scintillation fluid and analyzed with β-counter. The results are expressed in counts per minute (cpm).
In vivo subcutaneous model: microsphere degradation, peptide distribution, and glioma growth inhibition studies
To evaluate the in vivo degradation of microspheres, peptide distribution, and efficacy, male Swiss nude mice (4–6 wk old, Harlan Lab, Jerusalem, Israel) were inoculated subcutaneously with 5 x 106 U87-MG cells. After 10 days, when tumors reached 200 mm3 in size, animals were randomly divided into three treatment groups. The first group (n=40) received a single subcutaneous injection of microspheres labeled with coumarin (20 mg/mice) loaded with PF-4/CTF-rhodamine (100 µg) adjacent to the tumor. The second group (n=15) received a single subcutaneous injection of empty microspheres as a control. The third group (n=15) did not receive any treatment. The animals in the first group (PLGA PF-4/CTF) were further subdivided into four groups. At 1, 7, 14, and 30 days after microsphere injection, 10 animals at each time point were sacrificed and the tissue surrounding the injected microspheres was harvested. All samples were embedded in OCT (Sakura Finetek USA, Torrance, CA, USA), frozen in dry ice/butane, and stored at –80°C. Frozen sections (8 µm) were cut using a cryostat and stained with DAPI (Sigma). The microspheres, tumors, and surrounded tissue were visualized using fluorescence and confocal microscopy. The micrographs (n=6) were then analyzed with image-analysis software (Lucia G, Nikon). For the animals in the 30-day period, subcutaneous tumor growth was also measured transcutaneously with a caliber at the termination of the study, and the tumor volume was calculated (18)
.
In vivo intracranial model: microspheres degradation, peptide distribution, and glioma growth inhibition studies
To address the release profiles of PF-4/CTF from microspheres and therapeutic efficacy in an in vivo intracranial model, microspheres were injected into the established glioma. Briefly, Swiss nude male mice (n=20, 6–8 wk old; Charles River, Wilmington, MA, USA) were anesthetized and stereotactically inoculated with U87-MG cells (120,000 cells in 3 µl of PBS) via a 30-gauge Hamilton syringe (Hamilton, Reno, NV, USA) into the left forebrain (2.5 mm lateral and 1 mm anterior to bregma, at a 2.5 mm depth from the skull surface). Five days after tumor inoculation, the animals were randomized into two groups (n=10) and treated with intratumoral implantation of PLGA microspheres containing 15 µg PF-4/CTF or empty PLGA microspheres as a control via a 22-gauge Hamilton syringe at the established tumor site using the same burr hole and stereotactic coordinates. Implantation of microspheres was made into 2 sites: directly into the tumor center (7.5 µl) and 0.5 mm below the lower poles of the tumor (7.5 µl) along the trajectory of tumor inoculation.
MRI experiments
Two weeks after intracranial injection of microspheres, MRI experiments were performed on a Bruker 4.7 T system, operating on a Paravision (version 3.0.1) software platform (Bruker, Billerica, MA, USA). Tumor volumes were estimated using Gd-enhanced T1-weighted spin-echo images, from which 3D renderings of the tumors were generated with in-house 3D software (19
, 20)
.
After MRI, animals were perfused with 4% paraformaldehyde under deep anesthesia, and brains were processed as described previously (21)
. Tissue was stained with DAPI or hematoxylin and eosin as per standard protocol. Intracranial distribution of microspheres was assessed using fluorescent microscopy. Tumor volumes were estimated using the formula for ellipsoid and expressed as means ± SE as described previously (21)
.
Immunohistochemistry
Immunohistochemistry was carried out using the Vectastain Elite ABC kit (Vector Laboratories, Burlingame, CA, USA) as described previously. Primary antibodies included anti-CD31 (1:100; BD Biosciences PharMingen, San Jose, CA, USA) for blood vessel density, anti-Ki67 nuclear antigen (1:100; DAKO, Carpinteria, CA, USA) for proliferating cells and anti-cleaved caspase 3 (1:100; Cell Signaling Technology, Beverly, MA, USA) for the detection of apoptosis. The blood vessel density and apoptotic indices were defined as the percentage of positively stained cells of 100 nuclei from 5 randomly chosen high-power fields.
Statistics
In the subcutaneous model, significant differences in tumor growth among the groups were analyzed by analysis of variance (ANOVA). In the intracranial model, significant differences in tumor volume, microvessel density, proliferation, and apoptosis index were determined using the Mann-Whitney U test. Values of P < 0.05 were considered significant. All of the values were calculated as means ± SE or were expressed as percentage of control ± SEM.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A). Two types of microspheres were examined: uncapped PLGA50:50H (Fig. 1Aa-c
) and capped PLGA50:50 (Fig. 1Ad-f
). On the day of preparation (day 0) the microspheres composed of capped and uncapped polymers exhibited a spherical shape with smooth and uniform surface area (Fig. 1Aa, d
). After 10 days of incubation, the microspheres started to loose their smooth surface morphology (Fig. 1Ab, e
). The microspheres composed of uncapped polymers were broken and showed a rough surface morphology when compared to the capped microspheres (Fig. 1Ab
). In contrast, the capped microspheres were still relatively intact and smooth, and their shape remained relatively unchanged (Fig. 1Ae
). Thirty days postincubation, the microspheres composed of the uncapped polymer completely lost their shape and exhibited a porous bulk surface (Fig. 1Ac
). However, some of the microspheres composed of the capped polymer maintained their intact shape (Fig. 1Af
).
|
In vitro release of PF-4/CTF from capped and uncapped PLGA microspheres
The release profiles of PF-4/CTF-rhodamine from microspheres composed of PLGA 50:50 capped (esterified end groups) and uncapped (free carboxyl acid) PLGA 50:50H were determined by calculating the cumulative percentage of peptide released (Fig. 1B
). When using uncapped PLGA (50:50H) the release profile of PF-4/CTF- rhodamine was characterized by a continuous release over 17 days. In contrast, the capped PLGA (50:50) microspheres continued to release PF-4/CTF-rhodamine for up to 30 days. Both release profiles were characterized by a small initial burst of PF-4/CTF-rhodamine during the first day (10%). However, the release profile of PF-4/CTF-rhodamine from the capped polymer microspheres showed a significantly slower release (smaller slope) than the PF-4/CTF-rhodamine released from of the uncapped polymer microspheres.
Microsphere size and loading efficiency
Microspheres composed of capped PLGA had a narrow range of diameter distribution of 40 µm. The amounts of PF-4/CTF (labeled and unlabeled with rhodamine) were similar in all microspheres preparations, with high encapsulation efficiency of 44.8–69.8% (Fig. 1C
).
In vitro degradation kinetics of PLGA microspheres and release profile of PF-4/CTF from microspheres
PLGA microspheres were labeled with 6-coumarin and loaded with PF-4/CTF labeled with rhodamine to follow microsphere degradation and PF-4/CTF release in a physiological environment, both in vitro and in vivo. Figure 2
A shows a confocal micrograph of a single microsphere after 1, 7, 14, and 30 days of incubation. The confocal microscopy allows z-section analysis of the peptide distribution inside the microsphere. At all the time points, the labeled PF-4/CTF was located inside the microsphere matrices, even when microspheres undergo morphology changes during the longer incubation periods. One day after incubation, the PF-4/CTF signal was the highest and the distribution was relatively uniform. The PF-4/CTF-rhodamine was concentrated in discrete areas. Seven days after incubation, the microspheres were still spherical and uniform in shape, and their PF-4/CTF content was easily detected. The microspheres started to lose their shape and were found in clusters after only 14 days (Fig. 2A
). After 30 days of incubation, the majority of the microspheres had lost their spherical shape, and the polymer was detected as a bundle. Micrographs were analyzed using special software for image analysis, and the percent of the peptide (red) was normalized per microsphere (the polymeric fraction in green, Fig. 2B
). As can be seen, the amount of PF-4/CTF per microsphere decreased with increased incubation time. The PF-4/CTF-rhodamine signal was significantly lower after 7, 14, and 30 days of incubation than on day 1, but a considerable amount of PF-4/CTF-rhodamine (35%) was still detected after 30 days (Fig. 2B
).
|
Cell proliferation assay
Cell proliferation assays were performed in order to determine whether the PF-4/CTF retained its biological activity after being labeled with rhodamine (data not shown) and after its release from the microspheres. HUVEC proliferation was significantly inhibited by PF-4/CTF-rhodamine released from the PLGA microspheres on days 1, 3, and 7 of release by 14%, 33%, and 28%, respectively (Fig. 2C
). Samples taken from empty microspheres had no effect on HUVEC proliferation.
In vivo subcutaneous model: microsphere degradation, peptide distribution, and glioma growth inhibition studies
For the in vivo degradation studies PLGA 50:50 microspheres (the same formulation as used in the in vitro degradation experiments) loaded with PF-4/CTF rhodamine were implanted subcutaneously into male Swiss nude mice (20 mg per mouse) bearing subcutaneous glioma tumors (Fig. 3
A). The microspheres remained at the site of injection during the entire 30-day experimental period.
|
Image analysis of these micrographs showed a decrease in peptide/polymer content in the tissues, although at a slower rate than with the in vitro conditions (Figs. 3B
and 2B)
. Only after 30 days of release was the labeled PF-4/CTF signal significantly lower than the signal detected at 1 day postinjection (Fig. 3B
). The PF-4/CTF-rhodamine was still easily detectable in the microspheres and the surrounding tissue. The amount of peptide in the tissue was 30% of the content found in microspheres 1 day post-subcutaneous injection. Most important, PF-4/CTF-rhodamine released during these time points significantly inhibited the growth of U87-MG glioma subcutaneous xenografts. The tumor volume of labeled PF-4-/CTF-treated groups was inhibited by 80% when compared to the control groups (P<0.05, Fig. 3C
).
In vivo intracranial model: microsphere degradation, peptide distribution, and glioma growth inhibition
Microspheres (50:50 capped) labeled with 6-coumarin and loaded with PF-4/CTF rhodamine were successfully injected intracranially to nude mice bearing U-87MG glioma tumors (Fig. 4
). The microspheres were clearly detected and well distributed in the tumor site at 14 days postinjection. Fluorescent microscopy demonstrated strong and bright red and green signals, thus indicating that after 14 days the microspheres remained intact and contained a considerable amount of labeled PF-4/CTF.
|
To assess the therapeutic efficacy of PF-4/CTF-loaded microspheres injected intracranially, we monitored tumor volume by MRI and histological analysis. According to the MRI results, PF-4-/CTF-loaded microspheres significantly inhibited tumor growth 14 days postinjection by 65.3% (34.8±14.8 for PF-4-/CTF microspheres and 100±22.5 for empty microspheres; Fig. 5
A). These findings were confirmed by tumor volume measurements performed on tumors harvested from the mice 14 days after microsphere injection (Fig. 5B
). A single injection of PF-4-/CTF-loaded microspheres resulted in a 72% reduction of tumor volume compared to empty microspheres (27.9±12.1 for PF-4-/CTF microspheres and 100±26.0 for empty microspheres; *P<0.05). Immunohistochemical studies demonstrated a significant 55.7% decrease in microvessel count (26.3±1.1 for empty microspheres and 11.7±1.7 for PF-4/CTF microspheres) and a 4-fold increase in tumor cell apoptosis (0.84±0.07 for empty microspheres and 3.3±0.85 for PF-4/CTF microspheres). No significant difference was found in the proliferation index (38.8±1.3 for PF-4/CTF microspheres and 48.8±1.3 for empty microspheres; Fig. 6
; **P<0.01).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. We have previously shown that a single local injection of PLGA microspheres, loaded with very low amounts of the endogenous inhibitors PEX or PF-4/CTF, led to a significant inhibition of subcutaneous tumor growth. However, the efficacy of such a delivery system needs to be also demonstrated intracranially, to overcome barriers such as administration, sufficient protein loading, and therapeutic efficacy.
Prior to in vivo studies, several important properties such as drug-loading efficiency, polymer composition, microsphere size, drug release kinetic profile, and polymer degradation (23
, 24
, 25
) need to be determined in vitro. Later, it is essential to address the effect of the in vivo environment on these parameters. For these studies, labeling of the polymer and the therapeutic drug of interest is required so the degradation rate of the polymer in vivo and the release and distribution of the therapeutic drug in the tissue can be followed. Most important, the studies need to be performed with the therapeutic drug of interest rather than with a model drug, as such parameters as size and charge may affect the actual results.
In the current study, all the experiments were performed with the therapeutic molecule PF-4/CTF, which was conjugated, for the first time, to the fluorescent marker rhodamine. The labeling of the PF-4/CTF did not significantly affect its biological activity (data not shown). The PF4/CTF-rhodamine released from the microspheres retained its biological activity and inhibited HUVEC proliferation. The percent inhibition obtained in the current study is lower than that obtained in our previous study, which is due to a lower amount of PF-4/CTF loaded in microspheres (200 vs. 500 µg; 3
). In addition, for all experiments we labeled the PLGA polymer with 6-coumarin, a common labeling agent for PLGA. Labeling the PF-4/CTF and the PLGA allows the simultaneous and direct evaluation of protein release, distribution, and microsphere degradation in the subcutaneous and intracranial tumor environment. Previous studies have evaluated the release of the drug using indirect measurements of the amount of drug remaining in the microspheres or in the tissue only, with no follow up of polymer degradation (26
, 27)
. In addition, the morphological changes of the microspheres were conducted using SEM or TEM or by following labeled polymer, but not in the presence of the drug.
The therapeutic release from the PLGA microspheres occurs by homogenous degradation of the polymeric matrix. Many factors, including polymer composition, molecular weight of the drug, and polymer chain end groups, affect the release kinetics of a drug and the loading efficiency and degradation rate of the microspheres (28
, 29)
. When using polymers with free carboxyl acid end groups (uncapped PLGA 50:50H), the release kinetics of the protein are accelerated when compared to particles composed of polymers with esterified end groups (capped polymers, PLGA 50:50). From our in vitro studies, it is evident that the polymer composition affects the protein release kinetics and degradation rate of the microspheres. Both polymers (capped and uncapped) have a similar molecular weight (20,000 kDa), but they differ in their hydrophobicity, which in turn affects water penetration and surface erosion, especially in the first few days of release (30)
. Capped PLGA has esterified end groups and is less hydrophilic; thus, water penetration occurs at a slower rate than with uncapped polymer. SEM studies revealed that uncapped PLGA 50:50 microspheres loaded with PF-4/CTF undergo progressive changes in surface morphology over time when compared to the capped polymer at 30 days postincubation. However, the polymer composition (capped or uncapped) did not affect microsphere size and loading efficiencies.
Typical release kinetics from 50:50 (capped) PLGA microspheres are characterized by an initial burst release on incubation in vitro. Despite the fact that PF-4/CTF is a small protein (23 amino acids), in both polymer types used for microsphere preparation, a small initial burst of the PF-4/CTF was detected on the first day of release. However, the release of PF-4/CTF from uncapped (50:50H) microspheres was faster than from capped microspheres, with an almost complete release of PF-4/CTF after 2 wk of incubation. The PF4/CTF released from capped polymer showed typical triphase release kinetics (31)
, in which a fast initial release phase (burst) is followed by a second slow release (days 5–14) phase that lasts days and a third rapid release phase.
From the in vitro confocal studies, it appears that the PF-4/CTF-rhodamine is distributed throughout the microspheres with almost no absorption on the microsphere surface, which may explain the slower release profile during the first days of incubation. The moderate decrease in peptide-to-polymer ratio between days 14 and 30 indicates that some polymer erosion occurred together with peptide release.
Considering these results and our former results with unlabeled PF4/CTF release from capped PLGA (3)
, it is evident that the capped polymer microspheres have a longer release profile and slower degradation rate than the uncapped polymer. Therefore, we have chosen to use these microspheres for the in vivo kinetics and efficacy studies. The second step of our study was to inject microspheres, labeled with 6-coumarin and loaded with PF-4/CTF-rhodamine adjacent to glioma tumor inoculated subcutaneously, and follow their degradation rate, distribution of protein, and efficacy. The labeling of the polymer and protein allows the quantification of the drug-to-polymer ratio using confocal imaging. This ratio indicates the PF-4/CTF release kinetics from the microspheres over time. As with the in vivo studies, the polymer and PF-4/CTF fluorescent signals were well detected, and some of the microspheres were still intact after 30 days of implantation. The microspheres were localized near the tumor site and did not migrate to the surrounding tissue. However, compared to the in vitro release studies, the in vivo release of PF-4/CTF was slower, a trend that was also demonstrated by others (32)
. The significant release of PF-4/CTF in vivo occurred after only 30 days, as indicated by the PF-4/CTF-to-polymer ratio (a 70% release), compared to the in vitro condition, in which a significant decrease was observed after only a week (50% in vitro, compared to 30% in vivo). The slower release rate, which was related to the degradation of the microspheres in vivo, may be explained by the low availability of water in the tissue, compared with the in vitro conditions, in which the microspheres were incubated in PBS at 37°C and with shaking (100 rpm). In addition, different proteins, which are present in the tissue environment, may absorb to the surface of the microspheres and slow the release of the PF-4/CTF (33)
. Our results are in contrast to previous reported studies, which showed that the degradation in vivo was faster than that in vitro (23)
. This difference was explained by the effects of biological compounds present in the in vivo setting, such as lipids, that act as plasticizers and lead to increased uptake of water into the polymeric matrix (23
, 34
, 35)
. In addition, in these studies the microsphere degradation was followed by measuring polymer molecular weight as a function of time using gel-permeation chromatography, which is an indirect method to study degradation (23)
or TEM (36)
. Our study is the first to use a direct method to measure the in vivo degradation by using the therapeutic agent as the imaging marker at the same time.
We have also tested the biological efficacy of the rhodamine-labeled PF-4/CTF in vivo by following tumor inhibition over time. Subcutaneous tumor volume was significantly inhibited (80%) 30 days post-microsphere administration, in agreement with our previous work (3)
. This effect was obtained with a single subcutaneous injection of microspheres loaded with a very small amount of PF-4/CTF (100 µg/mouse) compared to systemic administration, which required milligram quantities of PF-4/CTF to achieve the same level of inhibition (13)
.
The most significant and clinically relevant aspect of our study was to effectively administer labeled PF-4/CTF microspheres (50:50 capped) intracranially and to follow their efficacy for the first time. The intracranial environment is unique because of specific extracellular matrix proteins, which modulate many crucial cell processes, as in survival, proliferation, differentiation, and migration. From the intracranial studies, it is evident that 14 days post-microsphere injection, the microspheres were still clearly detected, with some retention of morphological spherical shape. The microspheres were localized only near the tumor site where they were injected. Most important, the PF-4/CTF-rhodamine exhibited a strong red fluorescent signal. These results clearly demonstrate that the PF-4/CTF is still in the microspheres and surrounding tissue even 14 days postinjection. Taking into account the fact that PF-4/CTF has a rapid clearance in the body (minutes to a few hours; ref. 38
), the long-term efficacy over the 2-wk time period, using a single injection, is not a trivial achievement.
The presence of the PF-4/CTF for such a long period led to the significant inhibition of tumors inoculated intracranially. MRI was successfully implemented for monitoring the dynamic changes of tumor growth and inhibition in response to the slow-release treatment of the polymeric microspheres. This can give a very significant advantage when evaluating the effect of a continuous release of angiogenesis inhibitors, such as PF-4/CTF, due to the fact that the same tumor can be monitored over different stages of the therapeutic release. The inhibition of tumor growth was supported further by the H&E staining of the harvested brains. The amount of PF-4/CTF in the microspheres (15 µg/mouse) is significantly less than the amounts used by osmotic minipumps implanted intracranially (37)
. For minipump studies, 1, 0.5, and 0.25 mg/kg/day of human PF-4/CTF was given for 29 days, leading to 80–90% reduction in tumor volume. These results also demonstrated that the local treatment was far more effective than the systemic administration using similar amounts of PF-4/CTF (0.25–0.5 mg). The effectiveness of the microsphere system compared to minipump administration may be explained by the shape and size of the microspheres, which enable their distribution in the tumor bed, thus reaching larger tumor areas at the same time. In addition, microspheres can serve as a protective system for therapeutics, as indicated by the long retention of their biological activity (3)
. Therefore, such a system targets and concentrates the drug at the site of the tumor rather than in its surrounding tissue. Similar results have also been observed with gene therapy studies, which localized the DNA that encode for PF-4 at the site of the tumor (39
, 40)
.
In summary, our data clearly demonstrate the efficacy of PLGA microspheres as a continuous delivery system for angiogenic inhibitors for the treatment of malignant gliomas. The labeling of the therapeutic agent itself and the polymer enables the in vivo tracking of the fate of microspheres and correlates it to the in vitro conditions. The PLGA microspheres maintained their morphological integrity and continued to release biologically active PF-4/CTF for 30 days. This system may overcome many problems associated with the present clinical trials using antiangiogenic therapy, including the high doses needed and poor efficacy. Moreover, the PLGA microsphere system can be tailored to the therapeutic need, including increasing the length of delivery time and repeated administration.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication April 30, 2007. Accepted for publication August 16, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
