Biomacromolecules 2008, 9, 1117–1130
Polyaspartylhydrazide Copolymer-Based Supramolecular Vesicular Aggregates as Delivery Devices for Anticancer Drugs D. Paolino,† D. Cosco,† M. Licciardi,‡ G. Giammona,‡ M. Fresta,† and G. Cavallaro*,‡ Department of Pharmacobiological Sciences, Faculty of Pharmacy, University “Magna Græcia” of Catanzaro, Campus Universitario, Building of Biosciences, Viale Europa, I-88100 Germaneto (CZ), Italy, and Dipartimento di Chimica e Tecnologie Farmaceutiche, Università degli Studi di Palermo, Via Archirafi 32, 90123, Palermo, Italy Received August 29, 2007; Revised Manuscript Received January 7, 2008
In this paper we report on three different hydrophilic copolymers based on R,β-polyaspartylhydrazide (PAHy) bearing butyric groups in the side chain (C4) (PAHy-C4) or a combination of butyric groups and positive charged residues ((carboxypropyl)trimethylammonium chloride, CPTACl) (PAHy-C4-CPTA) that were synthesized and used for the preparation of new supramolecular vesicular aggregates (SVAs) containing gemcitabine as an antitumor drug. Gemcitabine-loaded SVAs containing synthesized PAHy derivatives were characterized from the physicochemical and technological point of view and the in vitro toxicity and anticancer activity on two different human cancer cell lines, i.e., CaCo-2 (human colon carcinoma) and ARO (human anaplastic thyroid carcinoma) cells, were also evaluated. Moreover, considering that carrier-cell interaction is an important factor to achieve an improvement of anticancer drug activity, confocal laser scanning microscopy and flow cytometric experiments were carried out on the two different cancer cell lines.
1. Introduction Supramolecular systems are considered interesting selforganizing nanodevices whose constituents present well-defined positioning and orientation. In the past few years scientific interest in these organized structures has rapidly increased and included different fields, e.g., electronics, bioengineering, and pharmaceutics.1 This interest stems from the unique features of supramolecular systems that provide the possibility of achieving complex structures with highly specific properties, through the coordinated combination of simple systems.2 Polymeric materials often represent the main constituents of these complex aggregates as polymers have a great versatility, due to the possibility of using different monomers in the polymeric backbone design and to the possibility of introducing several functional groups in their side chains (grafting).3 The combination of suitable polymeric materials and liposomal colloidal systems represents one of the various possibilities to achieve supramolecular systems of pharmaceutical and biomedical interest. In fact, liposomes have clear advantages over other delivery systems because they are nontoxic, biodegradable, and biocompatible devices able to deliver both hydrophilic and lipophilic drugs.4–6 Moreover surface and structural properties of liposomes can be efficiently modified by suitable copolymers.7–10 These copolymers properly used in the liposome formation process led to new colloidal systems with modified surface and bilayer properties due to the presence of copolymers, strongly integrated in the new resulting supramolecular vesicular aggregates (SVAs).7–10 Specific surface properties can be given to the colloidal SVA by proper polymers, such as positive charge density, potentially able to improve cell interaction with the surface of various tumor cells, that are reported to have elevated * Corresponding author. Phone: +39 091 6236 131. Fax: +39 091 6236 150. E-mail: [email protected]
† University “Magna Græcia” of Catanzaro. ‡ Università degli Studi di Palermo.
levels of negatively charged phospholids11–13 and “stealth properties”, useful to prolong blood circulation time;7–10 liposomal bilayer properties can be modified by suitable polymers confering to SVAs specific cell fusion properties and improved drug permeability as a consequence, for example, of a destabilizing effect due to the presence of polymers integrated in the bilayer structure.14,15 In this paper innovative SVAs made up by the combination of lipidic vesicles, formed of dipalmitoylphosphatidylcholine and cholesterol, as conventional lipids, and novel copolymers based on R,β-polyaspartylhydrazide (PAHy), are proposed with the aim of improving the interaction between SVAs and the surface of cancer cells and to increase SVA internalization within cancer cells with a consequent improvement of the anticancer efficacy of the encapsulated drug. PAHy is a freely water soluble, biocompatible, nontoxic and nonantigenic polymer, which has been used for the synthesis of macromolecular prodrugs16 and for the preparation of polymeric hydrogel for controlled drug delivery.17,18 In this paper, three new polyaspartylhydrazide copolymers containing hydrophobic butyric moieties (PAHy-C4) or both butyric and (carboxypropyl)trimethylammonium hydrochloride (CPTA) moieties, as positively charged groups, the last one linked to the polymeric backbone at two different amounts, were prepared and molecularly characterized. PAHy copolymers so obtained have been used to prepare SVAs with the rational to obtain SVAs bearing hydrophobized PAHy copolymers anchored to the bilayer by hydrophobic C4 portions and exposing on the surface of the bilayer just the hydrophilic backbone and, in the other case, together with hydrophilic chain also positive charged groups. With this the presence of hydrophobized PAHy copolymers anchored to the liposome phospholipidc bilayer could positively affect cell interaction and drug penentration,14,15 whereas cationic PAHy copolymers should be expected to improve the interaction between SVAs and the surface of various
10.1021/bm700964a CCC: $40.75 2008 American Chemical Society Published on Web 02/29/2008
Biomacromolecules, Vol. 9, No. 4, 2008
cancer cells characterized by a stronger negative charge in comparison with normal cells.11–13 Resulting SVAs were investigated as potential carriers for gemcitabine, an antitumoral drug which is widely used in clinical practice for the treatment of a number of cancers.19–21 Gemcitabine-loaded SVAs so obtained were characterized for both physicochemical and technological features and for in vitro toxicity and antiproliferative activity on two different human cancer cell lines, i.e., CaCo-2 (human colon carcinoma) and ARO (human anaplastic thyroid carcinoma) cells. Moreover, considering that the liposome-cell interaction is an important factor to achieve an improvement of drug anticancer activity, the intracellular trafficking was investigated on the two different cancer cell lines by confocal laser scanning microscopy and flow cytometric analysis.
2. Materials and Methods 2.1. Chemicals. 1-Ethyl-3-(dimethylamino)propyl)carbodiimide (EDCI), 3-(carboxypropyl)trimethylammonium chloride (CPTACl), and D2O were purchased from Aldrich (Milan, Italy); butyric acid was purchased from Fluka (Milan, Italy). R,β-Poly(aspartylhydrazide) (PAHy) was prepared and purified as elsewhere reported.23 Weight average molecular weight of PAHy was 20.3 kDa. 1,2-Dipalmitoylsn-glycero-3-phosphocholine monohydrate (DPPC) was purchased from Genzyme (Suffolk, U.K.). The following products were purchased from Sigma Chemicals Co. (St. Louis, MO): cholesterol (Chol), N-(fluorescein5-tiocarbamoyl)-1,2-dihexadecanoyl-sn-glycero-3-phosphoethanolamine triethylammonium salt (fluorescein-DHPE), amphotericin B solution (250 µg/mL), 3-[4,5-dimethylthiazol-2-yl]-3,5-diphenyltetrazolium bromide salt (used for MTT-tests), dimethylsulphoxide. RPMI1640 culture medium, Minimum essential medium (MEM) with glutamine, Dulbecco’s modified essential medium (D-MEM) with glutamine, fetal bovine serum (FBS), trypsin-EDTA (1×) solution and penicillin-streptomycin (100 UI/mL) solution were obtained from GIBCO (Invitrogen Corp., Giuliano Milanese (Mi), Italy). Doubledistilled pyrogen-free water was from Sifra SpA (Verona, Italy). Sterile saline was a product of Frekenius Kabi Potenza Srl (Verona, Italy). Gemcitabine (2,2′-difluorodeoxycytidine) hydrochloride (HPLC purity >99%) was a gift of Eli-Lilly Italia SpA (Sesto Fiorentino, Firenze, Italy). All other materials and solvents used in this investigation were of analytical grade (Carlo Erba, Milan, Italy). Human anaplastic thyroid tumor cell line (ARO cells) was a gift from G. Juilliard, UCLA. Human colon adenocarcinoma cell line (CaCo-2 cells) were provided by Istituto Zooprofilattico of Modena and Reggio Emilia. 2.2. Synthesis of r,β-Polyaspartyl[(hydrazide)-co-(butyrylhydrazide)] Copolymer (PAHy-C4). The synthesis of PAHy-C4 was carried out by dissolving 0.25 g of PAHy (1.9 mmol of repeating units) in 5 mL of distilled water and adding 50 µL (0.48 mmol) of butyric acid at a molar ratio between the moles of butyric acid and the moles of repeating units of PAHy (X value) of 0.25. The pH value of the reaction mixture was adjusted to 4.75, and then 0.140 g (0.73 mmol) of EDCI was added under stirring to obtain a molar ratio between EDCI and butyric acid of 1.5. The pH value of the reaction mixture was maintained at 4.75 for 1 h by adding 0.1 N HCl. The solution was then diluted with distilled water (∼5 mL), purified by an exhaustive dialysis using Visking Dialysis Tubing 18/32 in. with a molecular weight cutoff of 12000–14000 and freeze-dried. The pure product was obtained with a yield of 90% (w/w) with respect to the starting PAHy. 2.3. Synthesis of r,β-Polyaspartyl[(hydrazide)-co-(butyrylhydrazide)-co-(4-trimethylamonium chloride butyrylhydrazide)] Copolymer (PAHy-C4-CPTA). To synthesize PAHy-C4-CPTA, a suitable amount of CPTACl (having an X value of 0.7 or 1.4 with respect to free hydrazidic groups of the repeating units of PAHy) was added to an aqueous solution of PAHy-C4 copolymer (0.180 g of copolymer in 8 mL of distilled water). The pH value of the reaction mixture was adjusted to 4.75, and then a suitable amount of EDCI was
Paolino et al. added under stirring to obtain a EDCI/CPTAC1 molar ratio of 1.5. The pH value of the reaction mixture was maintained at 4.75 for 1 h by adding 0.1 N HCl. The same purification procedure as for PAHyC4 was then carried out. Also in this case, the yield of the reaction was 90% (w/w) of pure PAHy-C4-CPTA with respect to the starting PAHy. 2.4. Chemical Characterization of Copolymers. The synthesized copolymers were characterized by both IR and 1H NMR spectroscopy. IR spectra were obtained using a Perkin-Elmer 1720 IR Fourier transform spectrophotometer. Polymer samples were homogeneously mixed with potassium bromide and compressed to obtain disk-shaped tablets. Solid specimens were submitted to FT-IR analysis under a CO2free nitrogen atmosphere at room temperature. The 1H NMR spectra were recorded in D2O using a Bruker AC250 spectrometer operating at 250.13 MHz. Molecular weights and polydispersity indexes of copolymers were determined by a gel permeation chromatography method (GPC1) made up of a Waters (Milford, MA) 600 pump, two Ultraydrogel columns from Waters (1000 and 250 Armstrong), and a Waters 2410 refractive index detector. The mobile phase was a pH 7.8 phosphate buffer solution with 0.1 N NaNO3 and was pumped at a flow rate of 0.8 mL/ min. The GPC1 analysis was carried out at a temperature of 25 ( 0.1 °C. 2.5. Preparation of the SVA Carriers. Three different SVA carriers were prepared by the integration into vesicular system made up of DPPC/Chol (6:3 molar ratio) of one of three different PAHy copolymers; this was carried out using the thin layer evaporation (TLE) technique. Namely, suitable amounts (20 mg) of lipids were solubilized in a round-bottomed test tube with 2 mL of chloroform/methanol (3:1 v/v) and the solvent was then removed by means of a Büchi 461 rotavapor under a slow nitrogen flux, to obtain the formation of a thin film on the inner wall of the tube. This lipid film was hydrated with 100 µL of an aqueous solution of 10 mM gemcitabine and 900 µL of a PAHy copolymer solution. Different PAHy copolymers were added in a molar ratio of 0.1. SVAs were obtained by submitting the lipid/ aqueous dispersion to three alternate cycles (3 min each) of warming to 55 °C (thermostated water bath) and vortexing at 700 rpm. This procedure led to the formation of multilamellar vesicles coated with PAHy-C4 or PAHy-C4-CPTA. The vesicle suspension was keep at 55 °C for 3 h to stabilize the bilayer structures. Multilamellar vesicles were submitted to extrusion through polycarbonate membrane filters (Costar, Corning Inc., Corning, NY) by using a stainless steel extrusion device (Lipex Biomembranes, Vancouver, BC) with a 10 mL waterjacketed thermobarrel connected to a GR 150 thermostat (Grant Instruments Ltd., Cambridge, U.K.). Multilamellar vesicles were first forced through two stacked 400 nm pore size membrane filters and then through two stacked 200 nm pore size membrane filters (10 passages each). The working pressures were 430 and 900 kPa for 400 and 200 nm pore size membranes, respectively, producing 200 nm sized SVAs. Fluorescent-labeled SVAs were prepared by codissolving fluoresceinDHPE (0.1% molar) with the lipid material. SVAs were then prepared as reported above. 2.6. Physicochemical Characterization of SVAs. SVA mean size, size distribution and Z-potential were determined by the light scattering technique. Photon correlation spectroscopy was used for size analysis. The instrument was a Zetamaster (Malvern Instruments Ltd., Sparing Lane South, Worchester Shine, England) equipped with a 4.5 mW laser diode operating at 670 nm as a light source. Scattered photons were detected at 90°. A third-order cumulant fitting correlation function was performed by a Malvern PCS submicrometer particle analyzer to obtain the mean size and the polydispersity index of SVA colloidal suspensions. The real and imaginary refractive indexes were set at 1.59 and 0.0, respectively. The following parameters were used for experiments: medium refractive index, 1.330; medium viscosity, 1.0 mPa · s; dielectric constant, 80.4. Samples were placed in quartz cuvettes and suitably diluted with filtered water (200 nm pore size filters) to avoid
Supramolecular Vesicular Aggregates
Biomacromolecules, Vol. 9, No. 4, 2008
multiscattering phenomena. The zeta-potential values of SVAs were obtained by laser doppler anemometry, measuring the mean electrophoretic mobility. A Smoluchowsky constant F (Ka) of 1.5 was used to achieve zeta potential values from electrophoretic mobility. The laser doppler anemometry setup was equipped with a He-Ne (633 nm) laser with a nominal power of 5 mW. The light scattering apparatus was calibrated for both size and Z-potential measurements with suitable polystyrene latex before each set of experiments. 2.7. Loading Capacity of SVAs. SVAs were purified from the untrapped gemcitabine by gel permeation chromatography (GPC2) using ¨ KTA prime PLUS (Amersham Biosciences, Uppsala, Sweden) an A equipped with a UV detector at fixed wavelengths (254 and 280 nm). The gel permeation chromatographic separation was carried out with a XK16/20 column (Amersham Biosciences) packed with Sephadex G-25 and using a NaCl saline solution as eluent. The equilibration volume of the column was 80 mL, and the flow rate was 0.5 mL/min. The amount of SVA-associated gemcitabine was determined by HPLC following the destruction of SVAs with methanol.24 Instead of gel permeation chromatography for the purification of SVAs, a centrifugation method was carried out when in vitro experiments had to be carried out, thus avoiding any sample dilution and allowing a more precise dosage of gemcitabine-loaded SVAs. In particular, the untrapped drug was removed with a Beckman Avanti 30 centrifuge equipped with a F1202 fixed angle rotor (Beckman Coulter Inc., Fullerton, CA) at 58000g for 1 h at 4 °C. The amount of untrapped gemcitabine in the supernatant was determined spectrophotometrically at a λmax 268.8 nm by a Perkin-Elmer Lambda 20 UV–vis spectrophotometer (Uberlingen, Germany) with Perkin-Elmer UV WinLab ver. 2.8 acquisition software. The equation of the UV calibration curve of gemcitabine is y ) 0.6954 × 10-3 + 0.3963x, where y is the absorbance and x the drug concentration, with an r2 value of 0.9997. No interference was observed from other components present in the formulation. The aliquot of gemcitabine associated with liposomes is calculated as the difference between the drug amount added during liposome preparation and the amount of gemcitabine spectrophotometrically determined in the supernatant. In both cases, the amount of drug encapsulated is expressed as encapsulation yield according to the equation
EY ) [(Dt - Du) ⁄ Dt] × 100
where Dt is the total amount of the drug used for SVA preparation and Du is the amount of untrapped drug. 2.8. Gemcitabine Release from SVAs. The release of gemcitabine from SVAs was evaluated by using Franz type diffusion flow cells. Cells had a diffusion surface area of 0.75 cm2 and a flow-through receptor compartment with a volume of 4.5 mL. Cell receptor compartments were thermostated to 37 °C and constantly stirred at 700 rpm by small magnetic stirring bars with a Variomag multipoint stirrer (Daytona Beach, FL) to ensure receptor solution homogeneity. Cells were connected to a Minipuls 3 peristaltic pump (Gilson Italia Srl, Cinisello B., Milan) and the receptor solution was collected by means of a fraction collector FC 204 (Gilson). Receptor sink conditions were achieved by pumping the degassed receptor solution (sterile saline solution) at a flow rate of 2 mL/h. Regenerated cellulose membranes (cutoff 10000) (Spectra/PorMembranes, Spectrum Laboratories, Inc. CA) were mounted horizontally after removal of sodium azide solution by hydration for 30 min with pyrogen-free double-distilled water. Various gemcitabine-loaded SVAs (200 µL) were placed in the donor compartment. Release experiments were carried out for 24 h. The amount of gemcitabine released from SVAs was determined spectrophotometrically at 268.8 nm as reported above. Five different experiments were carried out. The amount of released gemcitabine is expressed as percentage with respect to the encapsulated drug amount. 2.9. Cell Cultures. ARO cells were cultured at 37 °C in 5% (v/v) CO2 by using RPMI-1640 medium supplemented with penicillin (100 UI/mL), streptomycin (100 µg/mL), amphotericin B (1% v/v), and FBS (10% v/v). CaCo-2 cells were cultured in D-MEM medium supplemented with the same amounts of antibiotics, antimycotics, and FBS.
When cells reached a confluence of 80%, they were treated with trypsin (2 mL) and transferred into 50 mL sterile test tubes to be centrifuged with a Megafuge 1.0 (Heraeus Sepatech) at 1000 rpm for 10 min at room temperature. The pellet was resuspended in 8 mL of culture medium and ∼5 × 105 cells were seeded in plastic culture dishes for in vitro experiments. 2.10. Determination of SVA Toxicity by MTT Test. To evaluate the possible toxicity of SVAs due to the presence of PAHy copolymers, ARO and CaCo-2 cells were submitted to the 3-[4,5-dimethylthiazol2-yl]-3,5-diphenyltetrazolium bromide dye (MTT) test, which determines the amount of colored formazan crystals generated during the biological test by living cells. Cell vitality was measured as a function of the different amounts of SVAs or the PAHy copolymers present in the culture medium. ARO cells (5 × 103 cells/100 µL) and CaCo-2 cells (7 × 103 cells/100 µL) were seeded in 96-well tissue culture plates at 37 °C. After 24 h of incubation, the medium was removed and replaced with 100 µL of the same medium containing different concentrations of SVAs or PAHy copolymers and incubated for 24, 48, or 72 h. At the end of each incubation time 10 µL of MTT (5 mg/mL dissolved in PBS) was added to each well and incubated for a further 3 h. The medium was then removed and a dimethyl sulfoxide/ ethanol (1:1 v/v) mixture (200 µL) was added to dissolve formazan crystals resulting from the cellular conversion of tetrazolium salts. The 96-well culture plates were gently shaken at 230 rpm (KS 130 Control, IKA WERKE GMBH & Co, Staufen, Germany) for 20 min. The sample absorbance was measured with an ELISA microplate reader (Labsystems mod. Multiskan MS, Midland, ON, Canada) at wavelengths of 540 and 690 nm. The amount of formazan is directly proportional to the number of living cells. The percentage of cell viability was calculated using the following equation:
%cellviability ) AbsT ⁄ AbsU × 100
where AbsT is the absorbance of treated cells and AbsU is the absorbance of untreated cells. Cell viability values are the mean of six different experiments ( standard deviation. 2.11. In Vitro Antitumoral Activity of Gemcitabine-Loaded SVAs. To evaluate the cytotoxic effects of free and SVA-associated gemcitabine on the two human cancer cell lines, two different and independent assays were carried out, i.e., the evaluation of both cell vitality by the MTT test and cell mortality by the trypan blue dye exclusion assay. Cell mortality induced by free or SVA-associated gemcitabine on ARO and CaCo-2 cells was assayed in 12-well plastic culture dishes when the cells reached ∼70% confluence. ARO and CaCo-2 cells were seeded at densities of 1.5 × 104 and 2.5 × 104 cells/mL for each well, respectively. After 24 h the cell culture medium was removed and cells were washed with PBS buffer solution (1 mL) and replaced with fresh culture cell medium containing free or SVA-associated gemcitabine at different concentrations and hence submitted to different incubation times. Following the anticancer drug treatment, cells were detached using trypsin/EDTA (1×) solution (2 mL) and then centrifuged (1200 rpm) at 22 °C for 10 min (Megafuge 1.0, Heraeus Sepatech). Supernatants were eliminated and cellular pellets were resuspended in a trypan blue buffer (200 µL). The number of dead cells was determined by a hematocytometer chamber using an optical microscope (Labophot2, Nikon, Japan). Dead cells were colored blue. The antiproliferative activity was expressed as the percentage of cellular mortality and calculated according to the equation
(Nt - Nd) ⁄ Nt × 100 (Nc - Ndc) ⁄ Nc
where Nt is the total number of treated cells, Nd is the number of dead treated cells colored in blue, Nc is the total number of control cells, and Ndc is the number of dead control cells colored blue. ARO and CaCo-2 cell growth following treatment with free or SVAassociated gemcitabine was evaluated by the MTT-test as reported above. Briefly, after different exposure times, MTT (10 µL) was added
Biomacromolecules, Vol. 9, No. 4, 2008
to each well and incubated for 3 h. The amounts of colored formazan crystals were then determined to evaluate cell vitality. For both assays, suitable scalar dilutions of free or SVA-associated gemcitabine were carried out to evaluate the dose-dependent effect of the anticancer agent, while different incubation times, i.e., 24, 48, and 72 h, were investigated to evaluate the anticancer agent exposure time dependent effect. In all experiments, untreated ARO and CaCo-2 cells were the control and cells treated with empty SVAs were the blank. Data are the mean of six different experiments ( standard deviation. 2.12. Confocal Laser Scanning Microscopy. ARO and CaCo-2 cells at 80% confluence were treated with 2 mL of trypsine/EDTA, collected in 50 mL sterile test tubes and centrifuged at 1000 rpm for 10 min. Cells were seeded in six-well plastic culture dishes (3 × 105 cells), in which a sterile cover glass was previously placed. After 24 h of incubation, cells were treated with fluorescein-DHPE labeled SVAs and then incubated for different times. At the end of each incubation period, culture medium was removed and cells were washed three times with pH 7.4 PBS and fixed on glass using an ethanol solution (70% v/v). Fixed cells were washed twice with PBS and stored at 4 °C up to the confocal microscopy analysis. A Leica TCS SP2 MP confocal laser scanning microscopy was used for analyses at λexc ) 496 nm and λem ) 519 nm. Before analysis, samples were placed on a glass treated with glycerol (70% v/v) and fixed with transparent glue to avoid air inclusion between the sample and the glass. 2.13. Intracellular Delivery. Cells were seeded in six-well plastic plates (2.5 × 104 cells/mL) and treated as reported elsewhere.26 When the cells reached a subconfluence growth, they were washed twice with PBS, and fresh medium was added. Cells were treated with 50 µL of the fluorescein-DHPE labeled SVAs and incubated at 37 °C for different times. Untreated cells were used as control. After the incubation, the medium was removed and cells were washed twice with PBS. They were then trypsinized (500 µL), washed with 1.5 mL of PBS, and collected in 15 mL sterile test tubes. Cells were fixed with a 70% (v/v) ethanol solution in PBS (1 mL) to a final concentration of 0.5 × 106 cell/mL. Flow cytometric analysis was carried out using a four-color FACS scanner (Becton-Dickinson Immunocytometry Systems) and LysysII software. For each analysis, 50000–200000 gated events were collected. Fluorescence was measured on a logarithmic scale with 1024 channel resolution. Mean fluorescence intensity values were determined as linear values from Lysys II software.26 2.14. Apoptosis. The presence of apoptosis was determined by analyzing the presence of DNA fragmentation. Cells were treated with the free drug, SVAs, and Gemcitabine loaded SVAs (Gem-SVAs) (1 µM final concentration) for 24, 48, and 72 h. The cells were then harvested with 2 mL of trypsin/EDTA (1×) solution, washed with 4 mL of PBS, and centrifuged at 1700 rpm for 3 min. The pellet was dissolved by adding 700 µL of lysis buffer solution, 800 µL of pH 8 Tris-HCl (10 mM), 200 µL of pH 8 EDTA (1 mM), and 100 µL of sodium dodecyl sulfate. Five microliters of DNase-free RNase (5 µg/ mL) was added to the product of the lysis process, and the solution was incubated at 37 °C for 40 min under continuous stirring. Then, 1 M NaCl (750 µL) was added to the cellular lysate, the mixture was incubated at 4 °C for 1 h and centrifuged at 1200 rpm for 25 min at 4 °C, and the supernatant was withdrawn, placed in a plastic test tube for cells, and mixed with 500 µL of phenol and 250 µL of chloroform. This mixture was vigorously mixed for 3 min and centrifuged at 1200 rpm for 30 min at 4 °C. The supernatant was withdrawn and 650 µL of chloroform was added to form a mixture between the cellular extract solution and the organic solvent. The mixture was vigorously mixed and centrifuged again. The supernatant was separated from the pellet, and the DNA content was precipitated with 1 mL of cooled isopropyl alcohol and stored overnight at -20 °C. The DNA sample was then centrifuged at 1200 rpm for 15 min at 4 °C and the cellular pellet was washed with 1 mL of cooled 70% (v/v) ethanol, centrifuged, and dissolved in 50 µL of Tris EDTA buffer. An aliquot of this solution (5 µL) containing the DNA at a concentration of 50 µg/mL was loaded on agarose gel 1.75% (w/v) in TBE buffer and submitted to electro-
Paolino et al. phoresis at 60 mV for 3 h at room temperature. The electrophoretic gel was stained with a solution of ethidium bromide (10 mg/mL) (5 µL in 5 mL of TBE buffer) for 40 min. Untreated cells were used as negative control (the absence of apoptosis), while HeLa cells treated with staurosporine (1 µM), an apoptosis-inducing agent, were used as a positive control (presence of extensive apoptosis characterized by extensive DNA fragmentation)27 to evaluate the suitability and sensitivity of this assay. 2.15. Statistical Analysis. Statistical analysis of the various experimental results was performed by using one-way ANOVA. A posteriori Bonferroni t test was carried out to check the ANOVA test. A p value SVA-2 > SVA-3. The presence of positive charges probably slowed down the permeation of an ionizable drug, such as gemcitabine, through bilayer structures. It is of interest that after 24 h no significant difference in the amount of gemcitabine released was observed between the various investigated SVAs, which showed a drug-release value of 65–70%. 3.3. In Vitro Tolerability of SVAs. The tolerability of a compoundisacriticalfactorforitspotentialclinicalapplication.36–38 Therefore, the potential toxicity both of free PAHy copolymers and of the three SVAs was investigated to characterize and to define the limit of therapeutic use and the potential application as in vivo anticancer drug carriers.
Biomacromolecules, Vol. 9, No. 4, 2008
Figure 3. Release profiles of gemcitabine from various SVAs at 37 °C. Data are expressed as the percentage of the amount of released gemcitabine with respect to the amount of drug entrapped within SVAs. Each value is the mean of five different experiments ( standard deviation.
As previously reported,29 PAHy itself or its copolymers obtained by the conjugation with different molecules such as glycidyltrimethylammonium showed no significant cytotoxicity up to a concentration of 1 mg/mL. However, the conjugation of PAHy with butyric and CPTA moieties can alter the biopharmaceutical properties of the native polymer. In this investigation the cytotoxicity of free PAHy copolymers and relative SVAs was assayed using ARO and CaCo-2 cells as a function of the incubation time, i.e., 24, 48, and 72 h. As shown in Figure 4, the three PAHy copolymers elicited a marked dose-dependent and exposure time dependent cytotoxic effect on ARO cells as assayed by the MTT test. A different cytotoxic effect was observed between the two positively charged PAHy-C4-CPTA and PAHy-C4, which elicited a greater cytotoxic effect at all the investigated incubation times. This completely different biological behavior of the PAHy copolymers prepared and investigated in this paper with respect to the native PAHy29 may be due to their ability to fully integrate with lipid bilayer structures, which on the one hand can represent an advantage for the formation of stable SVAs but on the other is a serious disadvantage in terms of cellular tolerability thus allowing also the interaction with cellular bilayer structures and hence the perturbation of the cellular membrane functions and homeostasis. All the SVAs showed a greater in vitro tolerability with respect to the free polymeric counterparts (Figure 4). These findings are probably due to the fact that in the case of SVAs PAHy copolymers are firmly anchored with the supramolecular carrier and hence no direct effect can be exerted on cells. Similarly to free PAHy copolymers, SVAs are characterized by a concentration- and time-dependent cytotoxic effect with respect to both ARO and CaCo-2 cells (data not reported). The cytotoxic profiles of SVAs are similar for both cell lines (Figure 5). In particular, the three SVAs seem to have a certain safety at concentrations e68 µg/mL considering that a vitality of g85% was observed for all the incubation times. A certain cytotoxicity of SVAs was observed at the highest investigated
Paolino et al.
concentrations and for longer exposure times; i.e., most relevant cytotoxic effects were evident only after 72 h of incubation. Interestingly, significant differences can be observed between the various SVAs at concentrations >68 µg/mL. SVA-1 showed a poor cell toxicity with respect to the control at all concentrations following 24 and 48 h incubation, but only after 72 h was it possible to observe a cytotoxic effect at 0.7 and 7.0 mg/mL. The examination of the cytotoxicity profiles of the positively charged SVA-2 and SVA-3 demonstrated that these two SVAs were more cytotoxic than SVA-1, particularly at concentrations g0.7 mg/mL and after 72 h incubation. Despite of the differences in cytotoxic behaviors of the three SVAs, it is possible to conclude that in any case and at all incubation times no cytotoxic effect was observed up to a PAHy derivative concentration of 68 µg/mL. This amount of PAHy derivatives is present in gemcitabine-loaded SVAs, when a drug concentration of 0.1 µM has to be achieved. This drug concentration represents the highest concentration that can be used in clinical practice for anticancer chemotherapy.39 Therefore, these SVAs might be suitable drug carriers for a potential in vivo application. 3.4. In Vitro Anticancer Activity of GemcitabineLoaded SVAs. It is widely recognized40,41 that the therapeutic use of certain colloidal drug delivery systems in anticancer chemotherapy can provide several biopharmaceutical and hence therapeutic advantages over unentrapped nicked drugs. In the specific case of gemcitabine, recent investigations22,31 have proved that the liposomal carrier can significantly improve the in vitro anticancer activity of this drug against some cancer cell lines. In this attempt and considering our previous encouraging results on gemcitabine liposomal delivery, to evaluate the potential advantages of using SVAs as delivery systems for anticancer drugs, the in vitro cytotoxic effect of three gemcitabine-loaded SVAs was assayed in comparison with drug-loaded uncoated vesicles and the free drug. The cytotoxicity experiments were carried out both on ARO and on CaCo-2 cells in such a way as to evaluate the dose-dependent in vitro antitumoral effect (from 0.001 to 100 µM) and the exposure time dependent effect (from 24 to 72 h). We chose ARO cells as the most widely used in vitro model of human anaplastic thyroid carcinoma, which represents a very aggressive cancer form characterized by a poor frequency of occurrence and a very high mortality (almost 100% of patients die within six months from diagnosis) for which the identification of novel therapeutic strategies is requested, while CaCo-2 cells were chosen as a cancer model with a high frequency of occurrence (especially in developed countries) and a lower aggressivity and mortality than anaplastic thyroid cancer. The results of the cytotoxicity experiments are expressed as the percentage of both cell viability and mortality (data not reported) obtained with two independent biological assays compared with untreated cells (control). Both assays of in vitro cytotoxicity were always in very good agreement throughout the various experiments. As shown in Figure 5, the cytotoxic effect of gemcitabine-loaded SVAs (Gem-SVAs) was different as a function of the PAHy copolymer used for the preparation of the SVAs. In particular, after 24 h of incubation, the most noticeable cytotoxic effect on ARO cells was observed for GemSVA-1 at a drug concentration of 0.01 µM (∼70% cell viability) with a further increase at 0.1 µM (45% cell viability). In the case of positively charged Gem-SVAs (Figure 5), a different behavior was observed as a function of the extent of CPTA substitution. Gem-SVA-3 (DD%CPTA ) 45) showed a cytotoxic effect not significantly different from that observed
Supramolecular Vesicular Aggregates
Biomacromolecules, Vol. 9, No. 4, 2008
Figure 4. Dose-dependent cytotoxic effect of various PAHy copolymers and SVAs against ARO cells at different exposure times. Cytotoxic effect (determined by MTT test) is expressed as the percentage of cell viability with respect to the control (untreated cells, 100% viability), according to eq 2. Error bars, if not shown, are within symbols. Results are presented as the mean of six different experiments ( standard deviation. Statistical analysis by one-way ANOVA and a posteriori Bonferroni t test: /, p < 0.05; //, p < 0.001. Similar results were obtained with CaCo-2cells (data not reported).
for the free gemcitabine in the concentration interval from 0.001 to 1 µM following 24 h of incubation. Gem-SVA-3 became
significantly more effective than the free drug at concentrations g10 µM, while Gem-SVA-2 was significantly more effective
Biomacromolecules, Vol. 9, No. 4, 2008
Paolino et al.
Figure 5. In vitro cytotoxic activity of various Gem-SVAs against ARO and CaCo-2 cells with respect to the free gemcitabine at different exposure times determined by MTT test (cell viability). Cell viability is expressed as a percentage and calculated with respect to untreated cells (control) according to eq 2. The mixture of gemcitabine with unloaded SVAs showed no significantly different cell viability value from those observed for the free drug (data not reported). Error bars, if not shown, are within symbols. Results are presented as the mean of six different experiments ( standard deviation. Statistical analysis by one-way ANOVA and a posteriori Bonferroni t test: /, p < 0.05; //, p < 0.001.
than the free drug just at a concentration of 0.001 µM with a maximum at 100 µM eliciting a cellular mortality of 30%. Gem-
SVA-3 (DD%CPTA ) 45) showed a cytotoxic activity similar to that of free gemcitabine after 24 h of incubation. The
Supramolecular Vesicular Aggregates
cytotoxic profile of Gem-SVA-3 changed noticeable with respect to the free drug following 48 and 72 h of exposure, in these situations a significant cytotoxic activity on ARO cells was already observed at 0.1 µM after 48 h incubation and an intense activity was achieved at 1 µM (∼82% cell mortality) after 72 h of exposure. The cytotoxic activity exerted by Gem-SVA-1 and GemSVA-2 after 48 and 72 h of exposure showed similar profiles to those observed following 24 h of exposure but was characterized by a greater extent of the cytotoxic effect (Figure 5). These findings are evidence of a time-exposure effect, i.e., the longer the ARO cell exposure to Gem-SVAs, the greater the cytotoxic effect. Gem-SVAs, following 48 and 72 h of exposure, provided a greater overall cytotoxic effect than the free drug, which was not at all or poorly effective up to a concentration of 1 µM. It is noteworthy that both Gem-SVA-1 and Gem-SVA-2 showed an intense in vitro anticancer activity just at 1 µM after 48 h of exposure, thus showing a cellular mortality of ∼75% with respect to the untreated cells. Taking into consideration the tolerability of the various SVAs (Figure 4) and the fact that no significant difference of the in vitro cytotoxic activity was observed between the free drug and a mixture of gemcitabine with unloaded SVAs, it can be claimed that the improvement of the antitumoral activity of gemcitabine is mediated by SVA delivery and no contribution comes from the carrier per se being safe up to a PAHy copolymer concentration of 0.1 mg/mL at all investigated exposure times. A positive contribution to the cytotoxic effect of the drug from SVAs can be hypothesized, especially for Gem-SVA-3, only for concentrations g10 µM and at an exposure time of 72 h. All these findings on various Gem-SVAs appeared particularly encouraging compared with the in vitro antitumoral effect on ARO cells observed for free gemcitabine and drug-loaded uncoated vesicles. In fact, the drug solution was not effective at any investigated concentration following 24 h of incubation but exterted its cytotoxic effect only after 72 h of incubation at the highest investigated concentrations, i.e., 10 and 100 µM, which are not compatible with an in vivo administration.39 In vitro anticancer activity experiments on CaCo-2 cells (Figure 5) showed a slightly different situation with respect to that observed for ARO cells. In fact, Gem-SVA-2 and GemSVA-3 showed cytotoxic activity profiles similar to that observed for the free drug after 24 h of exposure, while both Gem-SVA-2 and Gem-SVA-3 determined a slight but significant (p < 0.05) improvement of the antitumoral activity with respect to the free drug after exposure times of 48 and 72 h for concentrations g0.01 µM. In all cases Gem-SVA-2 and GemSVA-3 showed a lower antitumoral efficacy than that elicited by Gem-SVA-1, which exerted a significant (p < 0.001) anticancer activity already at a concentration of 0.1 µM after 24 h of exposure (∼40% cell viability). The in vitro antitumoral activity data demonstrated the presence of both a dose-dependent and an exposure-time effect for various Gem-SVAs. Differences were also observed between the various Gem-SVAs, and in particular the presence and the density of PAHy-C4-CPTA seemed to be the parameter governing the different interactions between carriers and cells, hence determining the different in vitro antitumoral activity profiles. To investigate whether the different rates of cytotoxic activity (especially in the case of ARO cells) observed for the various Gem-SVAs could be attributed to a different rate and extent of interaction with cells, CLSM and FACS experiments were carried out.
Biomacromolecules, Vol. 9, No. 4, 2008
3.5. SVA-Cancer Cell Interaction. Two independent methods, i.e., CLSM and FACS, were used to evaluate the cellular trafficking and the interaction rate of SVAs with the investigated cancer cells. CLSM micrographs (Figure 6) of monolayers of ARO and CaCo-2 cells following Gem-SVA uptake experiments supported the results of in vitro cytotoxicity by showing string fluorescence in the cell cytoplasm as well as in the nucleus. Furthermore, it was possible to observe the effects of both the exposure time and the characteristics of the SVA surface on the interaction with the two cancer cell lines. All micrographs show a fluorescent layer coinciding with the cell outline, and the stronger this green fluorescence, the longer the exposure time. Moreover, after 24 h of exposure to Gem-SVA-1, a variation of the ARO cell morphology can be clearly observed probably due to the cytotoxic effect of the encapsulated gemcitabine, in agreement with data on in vitro antitumoral efficacy. In the case of charged systems (Figure 6), only a slight green fluorescence and poor or no cell morphology alteration (due to the anticancer effect of gemcitabine) are detectable after 24 h of exposure. The green fluorescence increases after 48 and 72 h of exposure especially in cells treated with Gem-SVA-2 (data not reported). Also in this case CLSM data support in vitro findings on the antitumoral activity of 1 µM positively charged Gem-SVAs, namely, the rate and the extent of appearance of the anticancer drug activity are similar to those observed for the induction of green fluorescence following the interaction of fluorescence-labeled Gem-SVAs with ARO cells. Also in the case of CaCo-2 cells (Figure 6), different rates and extents of Gem-SVAs-cell interaction were observed as a function of the carrier type and were always in very good agreement with in vitro experiments on antitumoral activity. Namely, Gem-SVA-1 was able to noticeably interact with CaCo-2 cell membranes just after 24 h of exposure, showing an intense fluorescence in the cytoplasm. While in the case of positively charged systems the interaction with the cells is less intense and more gradual and an overall distribution of the green fluorescence can be observed both in cellular cytoplasm and in membranes, thus confirming the slight improvement of the anticancer drug activity for exposure time g48 h and/or drug concentrations g10 µM. CLSM experiments clearly showed a different rate of interaction between the various SVAs with both ARO and CaCo-2 cells according to the following decreasing order: Gem-SVA-1 > Gem-SVA-2 g Gem-SVA-3. In both cell lines no fluorescence (Figure 6) could be detected from untreated cells (control), thus showing that there is no cell autofluorescence phenomenon which may interfere and lead to a misinterpretation of the data. Moreover, CLSM threedimensional analysis of cells incubated with the three different fluorescence-labeled Gem-SVAs demonstrated that fluorescent signals are inside the cells, thus providing evidence of SVA uptake and internalization by the ARO and CaCo-2 cells (data not shown). Time-dependent cellular interaction of Gem-SVAs was also evaluated by FACS experiments, which can easily take into account the whole cell population. As shown in Figure 7, fluorescein-labeled Gem-SVA-1 confirmed CLSM data, thus showing the highest rate of interaction with respect to positively charged systems. The extent of carrier-cell interaction showed a difference between the various Gem-SVAs following 48 h of incubation, i.e., Gem-SVA-1 > Gem-SVA-2> Gem-SVA-3. Therefore, the type of SVA seems to be a determinant factor for the cellular delivery of the fluorescent probe.
Biomacromolecules, Vol. 9, No. 4, 2008
Paolino et al.
Figure 6. Confocal laser scanning micrographs of ARO and CaCo-2 cells treated with fluorescein-labeled Gem-SVA-1 and Gem-SVA-2 at different incubation times: 3 h (A); 6 h (B); 9 h (C); 24 h (D). Reflectance confocal laser scanning micrographs of untreated ARO (E) and CaCo-2 (F) cells.
Experimental data of CLSM and FACS showed, in good agreement with each other, the faster entry into tumor cells of neutral SVA-1 in comparison with charged SVA-2 and SVA3. Moreover these data are in agreement with in vitro cytotoxicity experiments. In effect the rate and efficiency of cell uptake of considered SVAs should be explained not only taking into account the possible electrostatic interaction between positively
charged (SVA-2 ans SVA-3) and negatively charged tumor cell surface but also considering the role played just by the hydrophobized PAHy (PAHy-C4 without positive charges) on the cell interaction. In effect the ability of neutral amphiphilic copolymers to modify fluidity and stability of the liposomal bilayer and then the ability to interact with the cell membrane is well-known in the literature.14,15 We can speculate that the
Supramolecular Vesicular Aggregates
Figure 7. Time-course of fluorescent probe penetration in cultured ARO cells as measured by FACS. Each point is the average of five different experiments ( standard deviation. Similar time-course profiles were observed for cultured CaCo-2 cells (data not reported).
presence of neutral hydrophilic copolymer PAHy-C4, integrated in the SVAs-1 bilayer, can increase vesicular fluidity, improving in vitro drug release, besides cell interaction, whereas in the case of PAHy-C4-CPTA (present in SVAs-2 and SVAs-3), probably the positive charges cause conformational changes in the copolymers, modifying (negatively) its ability to change fluidity of bilayer (of DPPC/chol vesicles) without offering a compensating effect in terms of interaction through positive charges. 3.6. Apoptosis Induction. Data on in vitro cytotoxicity showed a certain anticancer efficacy of Gem-SVAs by eliciting an increased cancer cell mortality with respect to the free drug. A particular aspect that can be useful to define is the evaluation of the type of cell death, i.e., necrosis (cell death due to an acute tissue injury) and apoptosis (an ordered and programmed cell death process). The apoptosis induction in cancer cells is a further advantage in terms of antitumoral efficacy, particularly for very aggressive forms of tumors.42 Therefore, the last aspect investigated in this paper was the occurrence of apoptosis following cell treatment with free drug or gem-SVAs. The analysis of the DNA fragmentation (data not reported) showed no apoptosis in the case of both untreated cells and cells treated with unloaded SVAs. Cell treatment with free gemcitabine determined a DNA fragmentation only after 72 h of exposure. The induction of an apoptotic pathway was much more rapid in the case of gem-SVA-1, which elicited a DNA fragmentation after just 24 h of incubation. After 48 h of incubation also gem-SVA-2 and Gem-SVA-3 induced apoptosis. The extent of DNA fragmentation was greater for Gem-SVAs than the free drug, particularly in the case of ARO cells. Also in this case the findings on extent and kinetics of apoptosis induction are in agreement with in vitro cytotoxicity and SVA-cancer cell interaction experiments. Most probably the rate and extent of apoptosis induction are related to the different intracellular drug uptakes elicited by the different Gem-SVAs with respect to the free drug.
Biomacromolecules, Vol. 9, No. 4, 2008
therapeutic field having the possibility of fulfilling a variety of requirements of a number of pathological situations. In particular, data reported herein demonstrate that PAHy copolymerbased SVAs may be strategic carriers in anticancer chemotherapy due to two important features, (i) they are quite safe at the concentrations used in clinical practice and (ii) they are able to improve the in vitro antitumoral activity of gemcitabine, thus representing a possible therapeutic approach for cancers such as human anaplastic thyroid carcinoma, which at the moment has no valid cure. A wide application range is envisaged for the supramolecular carriers investigated in this paper considering the multiplicity of possible variations that can be achieved by the chemical modification/conjugation of the PAHy polymeric backbone. The first evidence of these opportunities was provided by the different rate and extent of interactions as a function of both the cell and the SVA type. Namely, a faster and greater interaction with cells was observed for SVA-1 with respect to both SVA-2 and SVA-3. Therefore, our findings enforced the evidence of the importance of carrier surface characteristics in the recognition and interaction with biological substrates. The extremely encouraging results suggest possible in vivo applications for SVAs based on PAHy copolymers. Acknowledgment. This paper was financially supported by a grant from the Italian Ministry of University and Research (PRIN 2006) and a grant from the Italian Ministry of HealthsRegione Calabria (Dipartimento Tutela della Salute Politiche Sanitarie e Sociali). The authors are very grateful to Dr. Christian Celia (Department of Pharmacobiological Sciences, University of Catanzaro) for his excellent and valuable support in the in vitro apoptosis experiments and Mr. Nicola Rotiroti for his skillful technical assistance. The authors are very grateful to Dr. Antony Bridgewood for his revision of the language of this paper.
References and Notes (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18)
4. Conclusions The possibility to prepare stable SVAs with tailor-made properties represents a great applicative opportunity in the
(19) (20) (21)
Tsukruk, V. V. Prog. Polym. Sci. 1997, 22, 247–311. Harada, A.; Kataoka, K. Prog. Polym. Sci. 2006, 31, 949–982. Torchilin, V. J. Controlled Release 2001, 73, 137–172. Lian, T.; Ho, R. J. Y. J. Pharm. Sci. 2001, 90, 667–680. Banerjee, R. J. Biomater. Appl. 2001, 16, 3–21. Khuller, G. K.; Kapur, M.; Sharma, S. Curr. Pharm. Des. 2004, 10, 3263–3274. Klibanov, A. L.; Maruyama, K.; Torchilin, V. P.; Huang, L. FEBS Lett. 1990, 268, 235–237. Allen, T. M.; Hansen, C.; Martin, F.; Redemann, C.; Yau-Young, A. Biochim. Biophys. Acta 1991, 1066, 29–36. Simoes, S; Moreira, J. N.; Fonseca, C.; Duzgunes, N; Pedroso de Lima, M. C. AdV. Drug DeliVery ReV. 2004, 56, 947–965. Xiaomei, M.; Zhensheng, Z. Int. J. Pharm. 2006, 318, 55–61. Schroder-Borm, H.; Bakalova, R.; Andra, J. FEBS Lett. 2005, 579, 6128–6134. Utsugi, T.; Schroit, A. J.; Connor, J.; Bucana, C. D.; Fidler, I. J. Cancer Res. 1991, 51, 3062–3066. Ran, S.; Downes, A.; Thorpe, P. E. Cancer Res. 2002, 62, 6132– 6140. Castelli, F.; Messina, C.; Martinetti, E.; Licciardi, M.; Craparo, E. F.; Pitarresi, G. Thermochim. Acta 2004, 423, 18–28. Zhang, L.; Peng, T.; Cheng, S.-X.; Zhuo, R.-X. J. Phys. Chem. 2004, 108, 7783–7770. Giammona, G.; Carlisi, B.; Cavallaro, G.; Pitarresi, G.; Spampinato, S. J. Controlled Release 1994, 29, 63–72. Giammona, G.; Pitarresi, G.; Tomarchio, V.; De Guidi, G.; Giuffrida, S. J. Controlled Release 1998, 29, 249–257. Pitarresi, G.; Cavallaro, G.; Carlisi, B.; Giammona, G.; Bulone, D.; San Biagio, P. L. Macromol. Chem. Phys. 2000, 201, 2542–2549. Hendricksen, K.; Witjes, J. A. Curr. Opin. Urol. 2006, 16, 361–366. Smith, I. E. Semin. Oncol. 2006, 33, S19–23. Toschi, L.; Finocchiaro, G.; Bartolini, S.; Gioia, V.; Cappuzzo, F. Future Oncol 2005, 1, 7–17.
Biomacromolecules, Vol. 9, No. 4, 2008
(22) Celano, M.; Calcagno, M. G.; Bulotta, S.; Paolino, D.; Arturi, F.; Rotiroti, D.; Filetti, S.; Fresta, M.; Russo, D. BMC Cancer 2004, 4, 63. (23) Giammona, G.; Carlisi, B.; Palazzo, A.; Palazzo, S. Boll. Chim. Farm. 1989, 128, 62–64. (24) Pavelic, Z.; Skalko-Basnet, N.; Filipovic-Grcic, J.; Martinac, A.; Jalsenjak, I. J. Controlled Release 2005, 106, 34–43. (25) Benita, S.; Poly, P. A.; Puisieux, F.; Delattre, J. J. Pharm. Sci. 1984, 73, 1751–1754. (26) Touitou, E.; Godin, B.; Dayana, N.; Weissa, C.; Piliponskyb, A.; LeviSchaffer, F. Biomaterials 2001, 22, 3053–3059. (27) Tewari, M.; Quan, L. T.; O’Rourke, K.; Desnoyers, S.; Zeng, Z.; Beidler, D. R.; Poirier, G. G.; Salvesen, G. S.; Dixit, V. M. Cell 1995, 81, 801–809. (28) Giammona, G.; Tomarchio, V.; Pitarresi, G.; Cavallaro, G. Polymer 1997, 38, 3315–3323. (29) Pedone, E.; Cavallaro, G.; Richardson, S. C. W.; Duncan, R.; Giammona, G. J. Controlled Release 2001, 77, 139–153. (30) Cavallaro, G.; Palumbo, F. S.; Licciardi, M.; Giammona, G. Drug DeliV. 2005, 12, 377–384. (31) Calvagno, M. G.; Celia, C.; Paolino, D.; Cosco, D.; Iannone, M.; Castelli, F.; Doldo, P.; Fresta, M. Curr. Drug DeliVery 2007, 4, 89– 101.
Paolino et al. (32) De Campos, A. M.; Sanchez, A.; Gref, R.; Calvo, P.; Alonso, M. J. Eur. J. Pharm. Sci. 2003, 20, 73–81. (33) Ulrich, A. S. Biosci. Rep. 2002, 22, 129–150. (34) Nicholas, A. R.; Scott, M. J.; Kennedy, N. I.; Jones, M. N. Biochim. Biophys. Acta 2000, 1463, 167–178. (35) Mu, X.; Zhong, Z. Int. J. Pharm. 2006, 318, 55–61. (36) Paolino, D.; Muzzalupo, R.; Ricciardi, A.; Celia, C.; Picci, N.; Fresta, M. Biomed. Microdevices 2007;in press. (37) Paolino, D.; Lucania, G.; Mardente, D.; Alhaique, F.; Fresta, M. J. Controlled Release 2005, 106, 99–110. (38) Zhong, Z.; Feijen, J.; Lok, M. C.; Hennink, W. E.; Christensen, L. V.; Yockman, J. W.; Kim, Y. H.; Kim, S. W. Biomacromolecules 2005, 6, 3440–3448. (39) Touroutoglou, N.; Gravel, D.; Raber, M. N.; Plunkett, W.; Abruzzese, J. L. Ann. Oncol. 1998, 9, 1003–1008. (40) Maurer, N.; Fenske, D. B.; Cullis, P. R. Expert Opin. Biol. Ther. 2001, 1, 923–947. (41) Hofheinz, R. D.; Gnad-Vogt, S. U.; Beyer, U.; Hochhaus, A. Anticancer Drugs 2005, 16, 691–707. (42) Baumann, P.; Mandl-Weber, S.; Emmerich, B.; Straka, C.; Schmidmaier, R. Anticancer Drugs 2007, 18, 405–410.