- Email: [email protected]
Preparation and characterization of solid lipid nanospheres containing paclitaxel
European Journal of Pharmaceutical Sciences 10 (2000) 305–309 www.elsevier.nl / locate / ejps
Preparation and characterization of solid lipid nanospheres containing paclitaxel Roberta Cavalli, Otto Caputo, Maria Rosa Gasco* a
Dipartimento di Scienza e Tecnologia del Farmaco, Universita` degli Studi di Torino, Via P. Giuria 9, I-10125 Turin, Italy Received 17 August 1999; received in revised form 12 January 2000; accepted 31 January 2000
Abstract The study describes the development of stealth and non-stealth solid lipid nanospheres (SLNs) as colloidal carriers for paclitaxel, a drug with very low solubility. SLNs are constituted mainly of bioacceptable and biodegradable lipids, such as tripalmitin and phosphatidylcholine, and can incorporate amounts of paclitaxel up to 2.8%. Stealth and non-stealth loaded SLNs are in the nanometer size range and can be sterilized and freeze-dried. Thermal analysis (differential scanning calorimetry) showed that paclitaxel is not able to crystallize in the SLNs. Release of paclitaxel from SLNs is very low. Non-stealth and stealth SLNs are stable over time without precipitation of paclitaxel and can be proposed for its parenteral administration. 2000 Elsevier Science B.V. All rights reserved. Keywords: Solid lipid nanospheres; Paclitaxel; DSC; Colloidal carriers
1. Introduction Paclitaxel (Taxol) is a diterpenoid natural product with therapeutic activity against some kinds of human cancer, such as ovary, breast, lung, head and neck. One of the major problems entailed in using paclitaxel arises from its very low solubility in water due to its extremely hydrophobic character. The commercial formulation of paclitaxel consists of a micellar solution of the drug in Chremophor EL (polyoxyethylated castor oil) containing 50% absolute ethanol. Before administration as a slow infusion, the micellar solution is diluted with saline solution to give 0.6–1.2 mg / ml paclitaxel. Since Cremophor EL has been observed to cause severe hypersensitivity reactions in animals and man, various approaches have been pursued to improve the solubility of paclitaxel, including the synthesis of prodrugs, analogs and congeners. Alternative formulations, such as micelles (Alkan-Onyuksel et al., 1994; Miwa et al., 1998), liposomes (Sharma and Straubinger, 1994), emulsions (Tarr et al., 1987) and nanosuspensions (Bohm et al., 1998) have also been investigated. In the present study we investigated the possibility of incorporating paclitaxel in solid lipid nanospheres (SLNs),
colloidal therapeutic systems proposed for several administration routes (Gasco, 1997), obtained from warm oil-inwater microemulsions. As a vehicle for delivery of paclitaxel, SLNs would have the advantage of being constituted of biocompatible components such as lipids. Moreover, they should avoid drug precipitation, which is one of the problems present when the commercial dosage form is diluted before administration by infusion; indeed, the infusion devices have a filter to eliminate any precipitated paclitaxel. To incorporate paclitaxel we developed a SLN formulation using a triglyceride as lipid matrix, namely tripalmitin, and phosphatidylcholine as surfactant which is acceptable for intravenous administration. To prolong the half-life of SLNs in blood circulation after i.v. administration, stealth SLNs incorporating paclitaxel were also prepared. The steric stabilization of SLNs was achieved using dipalmitoylphosphatidylethanolamine conjugated with polyethylene glycol 2000 (PEG 2000).
2. Experimental
2.1. Materials *Corresponding author. Tel.: 139-11-6707-667; fax: 139-11-6707687. E-mail address: m [email protected] (M.R. Gasco) ]
Tripalmitin and butanol were from Fluka (Buchs, Switzerland); Epikuron 200 (soya phosphatidylcholine 95%)
0928-0987 / 00 / $ – see front matter 2000 Elsevier Science B.V. All rights reserved. PII: S0928-0987( 00 )00081-6
306
R. Cavalli et al. / European Journal of Pharmaceutical Sciences 10 (2000) 305 – 309
was a kind gift from Lucas Meyer (Hamburg, Germany); taurocholate sodium salt was a kind gift from PCA (Basaluzzo, Italy); cholesterylhemisuccinate, trehalose and butanol were from Sigma (Milan, Italy). Dipalmitoylphosphatidylethanolamine–polyethyleneglycol 2000 (DPPE– PEG 2000) was from Avanti Polar Lipids (Alabaster, Al, USA). Paclitaxel was a kind gift of Indena (Milan, Italy). The other chemicals were of analytical reagent grade.
2.2. Preparation of non-stealth solid lipid nanospheres Non-stealth SLNs were prepared from a warm oil-inwater (o / w) microemulsion containing tripalmitin (2.48 mmol, 7% w / w) as internal phase, Epikuron 200 (2.46 mmol, 7% w / w) as surfactant, cholesteryl hemisuccinate (2.14 mmol, 4.5% w / w), butanol (3.50 mmol, 9.2% w / w) and taurocholate (1.14 mmol, 2.4% w / w) as cosurfactants, and distilled filtered water (111.11 mmol, 69.9% w / w) as continuous phase. Paclitaxel was added to the melted tripalmitin; successively, Epikuron 200, cosurfactants and filtered water were added obtaining a clear microemulsion at about 708C. A series of microemulsions was prepared adding different amounts of paclitaxel up to 0.04 mmol. Non-stealth SLNs were obtained by dispersing the warm o / w microemulsion in cold water (2–38C) under mechanical stirring at a ratio of 1:10 (microemulsion–water, v / v); 1 ml of o / w microemulsion weight 0.957 g. The SLN–water dispersion was washed three times with distilled water by diafiltration (TCF2-Amicon, Danvers, USA) using a Diaflo YM 100 membrane (cut-off 100 000 Daltons).
2.3. Preparation of stealth solid lipid nanospheres Stealth SLNs were prepared by adding DPPE–PEG 2000 to the melted tripalmitin; two different amounts of stealth agent were used (0.2 and 1% calculated on the whole microemulsion). The other components of the microemulsions were the same as those of the microemulsion formulation reported above. Two clear microemulsions were easily obtained with the two percentages of DPPE–PEG 2000. The preparation of the stealth SLNs was the same as that of non-stealth SLNs.
freeze-drying. For this purpose the warm microemulsions were dispersed in a 2% solution of trehalose.
2.5. Characterization of solid lipid nanospheres 2.5.1. Photon correlation spectroscopy The average diameter and polydispersity index of nonstealth and stealth SLNs were determined by photon correlation spectroscopy (PCS) using a 90 PLUS instrument (Brookhaven Instrument, USA) at a fixed angle of 908C and at a temperature of 258C. The SLN–water dispersions were diluted 1:20 with filtered water before analysis. Each value reported is the average of five measurements. The polydispersity index measures the size distribution of the nanosphere population (Koppel, 1972). The average diameter and polydispersity index of SLNs were determined in other than aqueous media, namely saline solution and isotonic glycerol solution (2.6%). 2.5.2. Determination of zeta potential The electrophoretic mobility and the zeta potential were measured using a 90 Plus instrument (Brookhaven Instrument). For the determination of the electrophoretic mobility SLN samples were diluted with 0.1 mM KCl and placed in the electrophoretic cell where an electric field of 15.24 V/ cm was established. Each sample was analyzed in triplicate. The zeta potential values were calculated using the Smolochowski equation. 2.5.3. Transmission electron microscopy Transmission electron microscopy (TEM) analysis was performed using a Philips CM10 instrument. Before analysis, the SLN dispersions were diluted 1:10, stained with a 2% solution of osmium tetraoxide and sprayed on copper grids.
2.4. Sterilization and freeze-drying of solid lipid nanospheres
2.5.4. Thermal analysis of freeze-dried solid lipid nanospheres Differential scanning calorimetry (DSC) analysis was performed using a DSC / 7 differential scanning calorimeter equipped with a TAC 7 / DX instrument controller TAC 7 / DX. The instrument was calibrated with indium for melting point and heat of fusion. A heating rate of 208C / min was employed in the 25–300 8C temperature range. Standard aluminum sample pans (Perkin Elmer) were used; an empty pan was used as reference standard. Analyses were performed under nitrogen purge; triple runs were carried out on each sample.
To obtain sterile products, SLNs dispersions were autoclaved at 1218C and 2 bar for 15 min as prescribed by the European Pharmacopeia. To determine the amount of the incorporated drug, the SLN dispersions were freezedried using a Modulyo freeze-dryer (Edwards, Crawley, UK) before and after sterilization. A cryoprotector, trehalose, was also employed to study its influence during
2.5.5. Percentage of drug incorporated into solid lipid nanospheres For the quantitative determination of paclitaxel, a reverse phase HPLC method was used (Perkin Elmer Binary LC pump 250 liquid chromatograph, RP 18 Waters column 25034.6 mm, 5 mm). The mobile phase was methanol– water (60:40, v / v). The analysis was performed at a
R. Cavalli et al. / European Journal of Pharmaceutical Sciences 10 (2000) 305 – 309
flow-rate of 1.5 ml / min with the UV detector at 227 nm (Sharma et al., 1994b). The amount of paclitaxel incorporated into SLNs was determined on a weighed amount of freeze-dried SLNs dissolved in methanol.
2.5.6. In vitro release kinetics of paclitaxel from solid lipid nanospheres The in vitro release kinetics was determined using a multicompartmental rotating cell system with donor and receptor compartments of 1.5 ml volume each. A hydrophilic membrane Servapor dialysis tubing (Serva, Germany), cut-off 12 000, was used. The experiments were performed using SLN dispersions at pH 7.4; an equal volume of phosphate buffer, pH 7.4, was placed in the receptor compartment. At fixed times, the receptor solution was pipetted out and replaced with phosphate buffer, pH 7.4. To determine drug concentration, the above HPLC method was applied to analyse the receptor solution.
2.6. Determination of taurocholate sodium salt in solid lipid nanospheres The amount of taurocholate present in the SLNs after the washings was determined by HPLC using a Perkin Elmer binary LC Pump 250 and RP 18 Waters column (25034.6 mm, 5 mm). The eluent was acetonitrile–methanol–0.03 M phosphate buffer, pH 4.3 (15:30:55, v / v). The analysis was performed at a flow-rate of 0.6 ml / min with the UV detector operating at 210 nm. The freeze-dried SLNs were dissolved in methanol and analysed.
2.7. Determination of cholesteryl hemisuccinate in SLNs The amount of cholesteryl hemisuccinate present in the SLNs after the washings was determined by HPLC using a Perkin Elmer binary LC Pump 250 and RP 18 Waters column (25034.6 mm, 5 mm). The eluent was acetonitrile–isopropanol (60:40, v / v); the analysis was performed at a flow-rate of 0.6 ml / min with the UV detector operating at 210 nm. The freeze-dried SLNs were analysed directly after dissolution in eluent.
307
3. Results
3.1. Characterization of stealth and non-stealth solid lipid nanospheres 3.1.1. Non-stealth SLNs The average diameters, polydispersity indices and zeta potentials of SLNs prepared adding 0.011 mmol or 0.022 ml of paclitaxel to the microemulsion are reported in Table 1. The results differ according to the amount of paclitaxel incorporate: increasing the amount incorporated increase the sizes of SLNs. SLNs prepared with higher amounts of paclitaxel had average diameters above 300 nm. The average diameter of SLNs (0.011 mmol) in saline solution increased up to 420 nm and the polydispersity index to 0.26; in the isotonic glycerol solution the average diameter of the SLNs did not increase. TEM analysis showed the spherical shape of the SLNs and confirmed their sizes. 3.1.2. Stealth SLNs The average diameters, the polydispersity indices and the zeta potentials of stealth SLNs did not show differences from non-stealth SLNs. The same results were obtained using the two different amounts of stealth agent into the microemulsion (0.2 or 1%). In the isotonic glycerol solution the average diameter of stealth SLNs did not increase, while in the presence of saline solution the average diameters increased as with non-stealth SLNs. The zeta potential values of stealth SLNs decreased to 223 and 220 mV according to the amount of stealth agent. 3.2. Freeze-dried solid lipid nanospheres Freeze-dried SLNs readily redispersed in water under mechanical stirring, even 1 year after their preparation. The size of non-stealth and stealth SLNs remained in the nanometer size range (average diameter: 410 nm and polydispersity index: 0.4). The presence of cryoprotector (trehalose) favoured the redispersions of SLNs (average diameter: 340 nm and polydispersity index: 0.3).
3.3. Percentage of paclitaxel incorporated The amounts of paclitaxel incorporated adding 0.011 mmol and 0.022 mmol to the microemulsion are reported in Table 2. No paclitaxel was detected in the washing waters.
2.8. Solubility of paclitaxel over time SLN dispersions, both stealth and non-stealth, were ultracentrifuged to separate SLNs 1 week and 1 month after their preparation. Precipitation of SLNs was complete after ultracentrifugation at 40 000 g for 37 min using an L60 centrifuge (Beckman). The supernatants and SLNs were then analyzed by HPLC to determine the paclitaxel concentration.
Table 1 Average diameter, polydispersity index and zeta potential of paclitaxelloaded non-stealth SLNs Amount of paclitaxel (mmol)
Average diameter (nm)
Polydispersity index
Zeta potential (mV)
0.011 0.022
200 250
0.19 0.22
228 231
R. Cavalli et al. / European Journal of Pharmaceutical Sciences 10 (2000) 305 – 309
308
Table 2 Percentages of paclitaxel incorporated into SLNs Paclitaxel amount in the microemulsion (mmol)
Paclitaxel in the SLNs (%)
0.011 0.022
1.3 2.8
3.4. Thermal analysis of freeze-dried solid lipid nanospheres All SLNs prepared were analysed by DSC to investigate the crystal habit of paclitaxel. Fig. 1 shows the DSC curves of freeze-dried SLNs containing 2.8% of paclitaxel and paclitaxel alone. The SLNs did not show the melting peak of paclitaxel at 2308C, suggesting that paclitaxel is present in SLNs in the amorphous form or molecularly dispersed. DSC analysis showed that paclitaxel is still present in a non-crystalline form in SLNs 1 year after their preparation.
3.5. Release kinetics of paclitaxel from solid lipid nanospheres
Fig. 2. Percentage of paclitaxel released vs. time from paclitaxel-loaded SLNs (2.8%).
solution of paclitaxel; for paclitaxel was reported a solubility range from 0.5 to 35 mmol / l (Sharma et al., 1995).
3.7. Amount of cosurfactants into SLNs Fig. 2 reports the percentages of paclitaxel released from SLNs; after 120 min the percentage of paclitaxel release was about 0.1%. Sink conditions were not required as the release was very low with a linear trend.
No taurocholate was detected into freeze-dried SLNs after three washings. The amount of cholesteryl hemisuccinate present in the freeze-dried SLNs was 8%.
3.6. Solubility of paclitaxel over time 4. Discussion The same amount of paclitaxel (18 ng / ml) was found in the supernatant of all types of SLNs after ultracentrifugation. This solubility was lower than that a saturated
In recent years, paclitaxel has been used to treat ovarian carcinoma and breast carcinoma alone or together with
Fig. 1. DSC thermograms of paclitaxel (A) [– – –] and SLNs containing 2.8% paclitaxel (B) [——].
R. Cavalli et al. / European Journal of Pharmaceutical Sciences 10 (2000) 305 – 309
some other antineoplastic agents, such as cis-platinum or carboplatinum, obtaining positive results. Paclitaxel is commonly administered as a micellar solution to increase its solubility. High amount of Cremophor EL are employed to avoid precipitation of the drug when the concentrated solution is diluted. As an alternative carrier, we studied the incorporation of paclitaxel into SLNs. Paclitaxel-loaded SLNs were prepared from a warm o / w microemulsion not containing synthetic surfactant, but only natural molecules such as phosphatidylcholine. The other components of the SLNs formulation were biocompatible and biodegradable. Moreover, the amount of cosurfactants present in the SLNs dispersions can be decreased very much by diafiltration washings, obtaining a purified final product. Stealth SLNs incorporating paclitaxel are easily obtained with an average diameter suitable for i.v. administration. The presence of hydrophilic PEG chains on the surface of the SLNs decreased the zeta potential of the SLNs. We have previously shown that the presence of the stealth agent could increase the blood circulation time of SLNs (Fundaro` et al., submitted for publication). Stealth and non-stealth SLNs were in a colloidal size range and showed a spherical shape by TEM analysis. SLN dispersions can be sterilized by autoclaving, obtaining a sterile product without changing the amount of incorporated drug; sterile SLN dispersions are stable for more than 18 months stored at 48C. The SLN dispersions can be freeze dried (alone or in the presence of a cryoprotector) to facilitate storage. The cryoprotector favours the redispersion of the freeze-dried SLNs. The paclitaxel loading capacity of this SLN formulation is suitable for paclitaxel administration, and even at high drug incorporation the average diameter of SLNs remains in the colloidal size range. DSC analysis of paclitaxel-loaded SLNs showed that the drug melting peak at 2308C is not present, whereas with paclitaxel as such, the melting peak of the drug occurs before its decomposition. This thermal behaviour may be ascribed to the presence of paclitaxel in an amorphous form or molecularly dispersed. This effect on the crystalline habit of paclitaxel may be related to the preparative method of the SLNs. The rapid quenching of the microemulsion does not allow the drug to crystallize. From the experimental data the release kinetics of paclitaxel from SLNs was seen to be of first pseudo zero order and the amount of paclitaxel released over time was very low, avoiding its precipitation. SLNs could therefore be considered as a slow releasing carrier; this behaviour has already noted with other drugs, such as doxorubicin, which showed a slower release time (Cavalli et al., 1993) and consequently a longer circulation time in comparison to the doxorubicin solution (Fundaro` et al., submitted for
309
publication), after i.v. administration to rats in both nonstealth and in stealth SLNs. The problem of the separation of paclitaxel from the commercial used micellar solution would presumably be avoided as the release kinetics of the drug from SLNs is slow. A used daily dose of paclitaxel is 250 mg / m 2 which means about 10 mg / m 2 / h. Thus assuming an incorporation of paclitaxel of 2%, 10 mg of paclitaxel are contained in 500 mg of SLNs, which should be easily administered in 20–40 ml. The sizes of SLNs in the isotonic glycerol solution were maintained. In conclusion, non-stealth and stealth SLNs constituted of bioacceptable lipids were prepared and can be proposed as paclitaxel carriers. They showed average diameters in the colloidal size range, an incorporation of paclitaxel suitable for i.v. administration and a slow drug release. Works are in progress to study the cytotoxicity of paclitaxel-loaded SLNs in different cell cultures.
Acknowledgements The work has been supported by University of Turin (ex 60%).
References Alkan-Onyuksel, H., Ramakkrishnan, S., Chai, H.B., Pezzuto, J.A., 1994. A mixed micellar formulation suitable for the parenteral administration of taxol. Pharm. Res. 11, 206–212. ¨ ¨ Bohm, B.H.L., Behnke, D., Muller, R.H., 1998. Paclitaxel nanosuspensions: production, lyophilization and stability. In: Proceedings 2nd World Meeting APGI /APV, Paris, 25–28 May, 1998. Cavalli, R., Caputo, O., Gasco, M.R., 1993. Solid lipospheres of doxorubicin and idarubicin. Int. J. Pharm. 89, R9–R12. Fundaro` A., Cavalli R., Bargoni A., Vighetto D., Zara G.P., Gasco M.R. Non-stealth and stealth solid lipid nanospheres carrying doxorubicin: pharmacokinetics and tissue distribution after ev administration to rats. Pharmacol. Res., submitted for publication. Gasco, M.R., 1997. Solid lipid nanoparticles from microemulsions. Pharm. Technol. Eur. 9, 52–58. Koppel, D.E., 1972. Analysis of macromolecular polydispersity in intensity correlation spectroscopy: the method of cumulants. J. Chem. Phys. 57, 4814–4816. Miwa, A., Ishibe, A., Nakano, M., Yamashira, T., Itai, S., Jinno, S., Kawahara, H., 1998. Development of novel chitosan derivatives as micellar carriers of taxol. Pharm. Res. 15, 1844–1850. Sharma, A., Straubinger, R.M., 1994. Novel taxol formulations: preparation and characterization of taxol-containing liposomes. Pharm. Res. 11, 889–896. Sharma, A., Conway, W.D., Struabinger, R.M., 1994. Reversed-phase high performance liquid chromatographic determination of taxol in mouse plasma. J. Chromatogr. B 655, 315–319. Sharma, U.S., Balasubramanian, S.V., Straubinger, R.M., 1995. Pharmaceutical and physical properties of paclitaxel (taxol) complexes with cyclodextrins. J. Pharm. Sci. 84, 1223–1230. Tarr, B.D., Sanbandam, T.G., Yalkowsky, S.H., 1987. A new parenteral emulsion for the administration of taxol. Pharm. Res. 4, 162–165.
Preparation and characterization of solid lipid nanospheres containing paclitaxel Roberta Cavalli, Otto Caputo, Maria Rosa Gasco* a
Dipartimento di Scienza e Tecnologia del Farmaco, Universita` degli Studi di Torino, Via P. Giuria 9, I-10125 Turin, Italy Received 17 August 1999; received in revised form 12 January 2000; accepted 31 January 2000
Abstract The study describes the development of stealth and non-stealth solid lipid nanospheres (SLNs) as colloidal carriers for paclitaxel, a drug with very low solubility. SLNs are constituted mainly of bioacceptable and biodegradable lipids, such as tripalmitin and phosphatidylcholine, and can incorporate amounts of paclitaxel up to 2.8%. Stealth and non-stealth loaded SLNs are in the nanometer size range and can be sterilized and freeze-dried. Thermal analysis (differential scanning calorimetry) showed that paclitaxel is not able to crystallize in the SLNs. Release of paclitaxel from SLNs is very low. Non-stealth and stealth SLNs are stable over time without precipitation of paclitaxel and can be proposed for its parenteral administration. 2000 Elsevier Science B.V. All rights reserved. Keywords: Solid lipid nanospheres; Paclitaxel; DSC; Colloidal carriers
1. Introduction Paclitaxel (Taxol) is a diterpenoid natural product with therapeutic activity against some kinds of human cancer, such as ovary, breast, lung, head and neck. One of the major problems entailed in using paclitaxel arises from its very low solubility in water due to its extremely hydrophobic character. The commercial formulation of paclitaxel consists of a micellar solution of the drug in Chremophor EL (polyoxyethylated castor oil) containing 50% absolute ethanol. Before administration as a slow infusion, the micellar solution is diluted with saline solution to give 0.6–1.2 mg / ml paclitaxel. Since Cremophor EL has been observed to cause severe hypersensitivity reactions in animals and man, various approaches have been pursued to improve the solubility of paclitaxel, including the synthesis of prodrugs, analogs and congeners. Alternative formulations, such as micelles (Alkan-Onyuksel et al., 1994; Miwa et al., 1998), liposomes (Sharma and Straubinger, 1994), emulsions (Tarr et al., 1987) and nanosuspensions (Bohm et al., 1998) have also been investigated. In the present study we investigated the possibility of incorporating paclitaxel in solid lipid nanospheres (SLNs),
colloidal therapeutic systems proposed for several administration routes (Gasco, 1997), obtained from warm oil-inwater microemulsions. As a vehicle for delivery of paclitaxel, SLNs would have the advantage of being constituted of biocompatible components such as lipids. Moreover, they should avoid drug precipitation, which is one of the problems present when the commercial dosage form is diluted before administration by infusion; indeed, the infusion devices have a filter to eliminate any precipitated paclitaxel. To incorporate paclitaxel we developed a SLN formulation using a triglyceride as lipid matrix, namely tripalmitin, and phosphatidylcholine as surfactant which is acceptable for intravenous administration. To prolong the half-life of SLNs in blood circulation after i.v. administration, stealth SLNs incorporating paclitaxel were also prepared. The steric stabilization of SLNs was achieved using dipalmitoylphosphatidylethanolamine conjugated with polyethylene glycol 2000 (PEG 2000).
2. Experimental
2.1. Materials *Corresponding author. Tel.: 139-11-6707-667; fax: 139-11-6707687. E-mail address: m [email protected] (M.R. Gasco) ]
Tripalmitin and butanol were from Fluka (Buchs, Switzerland); Epikuron 200 (soya phosphatidylcholine 95%)
0928-0987 / 00 / $ – see front matter 2000 Elsevier Science B.V. All rights reserved. PII: S0928-0987( 00 )00081-6
306
R. Cavalli et al. / European Journal of Pharmaceutical Sciences 10 (2000) 305 – 309
was a kind gift from Lucas Meyer (Hamburg, Germany); taurocholate sodium salt was a kind gift from PCA (Basaluzzo, Italy); cholesterylhemisuccinate, trehalose and butanol were from Sigma (Milan, Italy). Dipalmitoylphosphatidylethanolamine–polyethyleneglycol 2000 (DPPE– PEG 2000) was from Avanti Polar Lipids (Alabaster, Al, USA). Paclitaxel was a kind gift of Indena (Milan, Italy). The other chemicals were of analytical reagent grade.
2.2. Preparation of non-stealth solid lipid nanospheres Non-stealth SLNs were prepared from a warm oil-inwater (o / w) microemulsion containing tripalmitin (2.48 mmol, 7% w / w) as internal phase, Epikuron 200 (2.46 mmol, 7% w / w) as surfactant, cholesteryl hemisuccinate (2.14 mmol, 4.5% w / w), butanol (3.50 mmol, 9.2% w / w) and taurocholate (1.14 mmol, 2.4% w / w) as cosurfactants, and distilled filtered water (111.11 mmol, 69.9% w / w) as continuous phase. Paclitaxel was added to the melted tripalmitin; successively, Epikuron 200, cosurfactants and filtered water were added obtaining a clear microemulsion at about 708C. A series of microemulsions was prepared adding different amounts of paclitaxel up to 0.04 mmol. Non-stealth SLNs were obtained by dispersing the warm o / w microemulsion in cold water (2–38C) under mechanical stirring at a ratio of 1:10 (microemulsion–water, v / v); 1 ml of o / w microemulsion weight 0.957 g. The SLN–water dispersion was washed three times with distilled water by diafiltration (TCF2-Amicon, Danvers, USA) using a Diaflo YM 100 membrane (cut-off 100 000 Daltons).
2.3. Preparation of stealth solid lipid nanospheres Stealth SLNs were prepared by adding DPPE–PEG 2000 to the melted tripalmitin; two different amounts of stealth agent were used (0.2 and 1% calculated on the whole microemulsion). The other components of the microemulsions were the same as those of the microemulsion formulation reported above. Two clear microemulsions were easily obtained with the two percentages of DPPE–PEG 2000. The preparation of the stealth SLNs was the same as that of non-stealth SLNs.
freeze-drying. For this purpose the warm microemulsions were dispersed in a 2% solution of trehalose.
2.5. Characterization of solid lipid nanospheres 2.5.1. Photon correlation spectroscopy The average diameter and polydispersity index of nonstealth and stealth SLNs were determined by photon correlation spectroscopy (PCS) using a 90 PLUS instrument (Brookhaven Instrument, USA) at a fixed angle of 908C and at a temperature of 258C. The SLN–water dispersions were diluted 1:20 with filtered water before analysis. Each value reported is the average of five measurements. The polydispersity index measures the size distribution of the nanosphere population (Koppel, 1972). The average diameter and polydispersity index of SLNs were determined in other than aqueous media, namely saline solution and isotonic glycerol solution (2.6%). 2.5.2. Determination of zeta potential The electrophoretic mobility and the zeta potential were measured using a 90 Plus instrument (Brookhaven Instrument). For the determination of the electrophoretic mobility SLN samples were diluted with 0.1 mM KCl and placed in the electrophoretic cell where an electric field of 15.24 V/ cm was established. Each sample was analyzed in triplicate. The zeta potential values were calculated using the Smolochowski equation. 2.5.3. Transmission electron microscopy Transmission electron microscopy (TEM) analysis was performed using a Philips CM10 instrument. Before analysis, the SLN dispersions were diluted 1:10, stained with a 2% solution of osmium tetraoxide and sprayed on copper grids.
2.4. Sterilization and freeze-drying of solid lipid nanospheres
2.5.4. Thermal analysis of freeze-dried solid lipid nanospheres Differential scanning calorimetry (DSC) analysis was performed using a DSC / 7 differential scanning calorimeter equipped with a TAC 7 / DX instrument controller TAC 7 / DX. The instrument was calibrated with indium for melting point and heat of fusion. A heating rate of 208C / min was employed in the 25–300 8C temperature range. Standard aluminum sample pans (Perkin Elmer) were used; an empty pan was used as reference standard. Analyses were performed under nitrogen purge; triple runs were carried out on each sample.
To obtain sterile products, SLNs dispersions were autoclaved at 1218C and 2 bar for 15 min as prescribed by the European Pharmacopeia. To determine the amount of the incorporated drug, the SLN dispersions were freezedried using a Modulyo freeze-dryer (Edwards, Crawley, UK) before and after sterilization. A cryoprotector, trehalose, was also employed to study its influence during
2.5.5. Percentage of drug incorporated into solid lipid nanospheres For the quantitative determination of paclitaxel, a reverse phase HPLC method was used (Perkin Elmer Binary LC pump 250 liquid chromatograph, RP 18 Waters column 25034.6 mm, 5 mm). The mobile phase was methanol– water (60:40, v / v). The analysis was performed at a
R. Cavalli et al. / European Journal of Pharmaceutical Sciences 10 (2000) 305 – 309
flow-rate of 1.5 ml / min with the UV detector at 227 nm (Sharma et al., 1994b). The amount of paclitaxel incorporated into SLNs was determined on a weighed amount of freeze-dried SLNs dissolved in methanol.
2.5.6. In vitro release kinetics of paclitaxel from solid lipid nanospheres The in vitro release kinetics was determined using a multicompartmental rotating cell system with donor and receptor compartments of 1.5 ml volume each. A hydrophilic membrane Servapor dialysis tubing (Serva, Germany), cut-off 12 000, was used. The experiments were performed using SLN dispersions at pH 7.4; an equal volume of phosphate buffer, pH 7.4, was placed in the receptor compartment. At fixed times, the receptor solution was pipetted out and replaced with phosphate buffer, pH 7.4. To determine drug concentration, the above HPLC method was applied to analyse the receptor solution.
2.6. Determination of taurocholate sodium salt in solid lipid nanospheres The amount of taurocholate present in the SLNs after the washings was determined by HPLC using a Perkin Elmer binary LC Pump 250 and RP 18 Waters column (25034.6 mm, 5 mm). The eluent was acetonitrile–methanol–0.03 M phosphate buffer, pH 4.3 (15:30:55, v / v). The analysis was performed at a flow-rate of 0.6 ml / min with the UV detector operating at 210 nm. The freeze-dried SLNs were dissolved in methanol and analysed.
2.7. Determination of cholesteryl hemisuccinate in SLNs The amount of cholesteryl hemisuccinate present in the SLNs after the washings was determined by HPLC using a Perkin Elmer binary LC Pump 250 and RP 18 Waters column (25034.6 mm, 5 mm). The eluent was acetonitrile–isopropanol (60:40, v / v); the analysis was performed at a flow-rate of 0.6 ml / min with the UV detector operating at 210 nm. The freeze-dried SLNs were analysed directly after dissolution in eluent.
307
3. Results
3.1. Characterization of stealth and non-stealth solid lipid nanospheres 3.1.1. Non-stealth SLNs The average diameters, polydispersity indices and zeta potentials of SLNs prepared adding 0.011 mmol or 0.022 ml of paclitaxel to the microemulsion are reported in Table 1. The results differ according to the amount of paclitaxel incorporate: increasing the amount incorporated increase the sizes of SLNs. SLNs prepared with higher amounts of paclitaxel had average diameters above 300 nm. The average diameter of SLNs (0.011 mmol) in saline solution increased up to 420 nm and the polydispersity index to 0.26; in the isotonic glycerol solution the average diameter of the SLNs did not increase. TEM analysis showed the spherical shape of the SLNs and confirmed their sizes. 3.1.2. Stealth SLNs The average diameters, the polydispersity indices and the zeta potentials of stealth SLNs did not show differences from non-stealth SLNs. The same results were obtained using the two different amounts of stealth agent into the microemulsion (0.2 or 1%). In the isotonic glycerol solution the average diameter of stealth SLNs did not increase, while in the presence of saline solution the average diameters increased as with non-stealth SLNs. The zeta potential values of stealth SLNs decreased to 223 and 220 mV according to the amount of stealth agent. 3.2. Freeze-dried solid lipid nanospheres Freeze-dried SLNs readily redispersed in water under mechanical stirring, even 1 year after their preparation. The size of non-stealth and stealth SLNs remained in the nanometer size range (average diameter: 410 nm and polydispersity index: 0.4). The presence of cryoprotector (trehalose) favoured the redispersions of SLNs (average diameter: 340 nm and polydispersity index: 0.3).
3.3. Percentage of paclitaxel incorporated The amounts of paclitaxel incorporated adding 0.011 mmol and 0.022 mmol to the microemulsion are reported in Table 2. No paclitaxel was detected in the washing waters.
2.8. Solubility of paclitaxel over time SLN dispersions, both stealth and non-stealth, were ultracentrifuged to separate SLNs 1 week and 1 month after their preparation. Precipitation of SLNs was complete after ultracentrifugation at 40 000 g for 37 min using an L60 centrifuge (Beckman). The supernatants and SLNs were then analyzed by HPLC to determine the paclitaxel concentration.
Table 1 Average diameter, polydispersity index and zeta potential of paclitaxelloaded non-stealth SLNs Amount of paclitaxel (mmol)
Average diameter (nm)
Polydispersity index
Zeta potential (mV)
0.011 0.022
200 250
0.19 0.22
228 231
R. Cavalli et al. / European Journal of Pharmaceutical Sciences 10 (2000) 305 – 309
308
Table 2 Percentages of paclitaxel incorporated into SLNs Paclitaxel amount in the microemulsion (mmol)
Paclitaxel in the SLNs (%)
0.011 0.022
1.3 2.8
3.4. Thermal analysis of freeze-dried solid lipid nanospheres All SLNs prepared were analysed by DSC to investigate the crystal habit of paclitaxel. Fig. 1 shows the DSC curves of freeze-dried SLNs containing 2.8% of paclitaxel and paclitaxel alone. The SLNs did not show the melting peak of paclitaxel at 2308C, suggesting that paclitaxel is present in SLNs in the amorphous form or molecularly dispersed. DSC analysis showed that paclitaxel is still present in a non-crystalline form in SLNs 1 year after their preparation.
3.5. Release kinetics of paclitaxel from solid lipid nanospheres
Fig. 2. Percentage of paclitaxel released vs. time from paclitaxel-loaded SLNs (2.8%).
solution of paclitaxel; for paclitaxel was reported a solubility range from 0.5 to 35 mmol / l (Sharma et al., 1995).
3.7. Amount of cosurfactants into SLNs Fig. 2 reports the percentages of paclitaxel released from SLNs; after 120 min the percentage of paclitaxel release was about 0.1%. Sink conditions were not required as the release was very low with a linear trend.
No taurocholate was detected into freeze-dried SLNs after three washings. The amount of cholesteryl hemisuccinate present in the freeze-dried SLNs was 8%.
3.6. Solubility of paclitaxel over time 4. Discussion The same amount of paclitaxel (18 ng / ml) was found in the supernatant of all types of SLNs after ultracentrifugation. This solubility was lower than that a saturated
In recent years, paclitaxel has been used to treat ovarian carcinoma and breast carcinoma alone or together with
Fig. 1. DSC thermograms of paclitaxel (A) [– – –] and SLNs containing 2.8% paclitaxel (B) [——].
R. Cavalli et al. / European Journal of Pharmaceutical Sciences 10 (2000) 305 – 309
some other antineoplastic agents, such as cis-platinum or carboplatinum, obtaining positive results. Paclitaxel is commonly administered as a micellar solution to increase its solubility. High amount of Cremophor EL are employed to avoid precipitation of the drug when the concentrated solution is diluted. As an alternative carrier, we studied the incorporation of paclitaxel into SLNs. Paclitaxel-loaded SLNs were prepared from a warm o / w microemulsion not containing synthetic surfactant, but only natural molecules such as phosphatidylcholine. The other components of the SLNs formulation were biocompatible and biodegradable. Moreover, the amount of cosurfactants present in the SLNs dispersions can be decreased very much by diafiltration washings, obtaining a purified final product. Stealth SLNs incorporating paclitaxel are easily obtained with an average diameter suitable for i.v. administration. The presence of hydrophilic PEG chains on the surface of the SLNs decreased the zeta potential of the SLNs. We have previously shown that the presence of the stealth agent could increase the blood circulation time of SLNs (Fundaro` et al., submitted for publication). Stealth and non-stealth SLNs were in a colloidal size range and showed a spherical shape by TEM analysis. SLN dispersions can be sterilized by autoclaving, obtaining a sterile product without changing the amount of incorporated drug; sterile SLN dispersions are stable for more than 18 months stored at 48C. The SLN dispersions can be freeze dried (alone or in the presence of a cryoprotector) to facilitate storage. The cryoprotector favours the redispersion of the freeze-dried SLNs. The paclitaxel loading capacity of this SLN formulation is suitable for paclitaxel administration, and even at high drug incorporation the average diameter of SLNs remains in the colloidal size range. DSC analysis of paclitaxel-loaded SLNs showed that the drug melting peak at 2308C is not present, whereas with paclitaxel as such, the melting peak of the drug occurs before its decomposition. This thermal behaviour may be ascribed to the presence of paclitaxel in an amorphous form or molecularly dispersed. This effect on the crystalline habit of paclitaxel may be related to the preparative method of the SLNs. The rapid quenching of the microemulsion does not allow the drug to crystallize. From the experimental data the release kinetics of paclitaxel from SLNs was seen to be of first pseudo zero order and the amount of paclitaxel released over time was very low, avoiding its precipitation. SLNs could therefore be considered as a slow releasing carrier; this behaviour has already noted with other drugs, such as doxorubicin, which showed a slower release time (Cavalli et al., 1993) and consequently a longer circulation time in comparison to the doxorubicin solution (Fundaro` et al., submitted for
309
publication), after i.v. administration to rats in both nonstealth and in stealth SLNs. The problem of the separation of paclitaxel from the commercial used micellar solution would presumably be avoided as the release kinetics of the drug from SLNs is slow. A used daily dose of paclitaxel is 250 mg / m 2 which means about 10 mg / m 2 / h. Thus assuming an incorporation of paclitaxel of 2%, 10 mg of paclitaxel are contained in 500 mg of SLNs, which should be easily administered in 20–40 ml. The sizes of SLNs in the isotonic glycerol solution were maintained. In conclusion, non-stealth and stealth SLNs constituted of bioacceptable lipids were prepared and can be proposed as paclitaxel carriers. They showed average diameters in the colloidal size range, an incorporation of paclitaxel suitable for i.v. administration and a slow drug release. Works are in progress to study the cytotoxicity of paclitaxel-loaded SLNs in different cell cultures.
Acknowledgements The work has been supported by University of Turin (ex 60%).
References Alkan-Onyuksel, H., Ramakkrishnan, S., Chai, H.B., Pezzuto, J.A., 1994. A mixed micellar formulation suitable for the parenteral administration of taxol. Pharm. Res. 11, 206–212. ¨ ¨ Bohm, B.H.L., Behnke, D., Muller, R.H., 1998. Paclitaxel nanosuspensions: production, lyophilization and stability. In: Proceedings 2nd World Meeting APGI /APV, Paris, 25–28 May, 1998. Cavalli, R., Caputo, O., Gasco, M.R., 1993. Solid lipospheres of doxorubicin and idarubicin. Int. J. Pharm. 89, R9–R12. Fundaro` A., Cavalli R., Bargoni A., Vighetto D., Zara G.P., Gasco M.R. Non-stealth and stealth solid lipid nanospheres carrying doxorubicin: pharmacokinetics and tissue distribution after ev administration to rats. Pharmacol. Res., submitted for publication. Gasco, M.R., 1997. Solid lipid nanoparticles from microemulsions. Pharm. Technol. Eur. 9, 52–58. Koppel, D.E., 1972. Analysis of macromolecular polydispersity in intensity correlation spectroscopy: the method of cumulants. J. Chem. Phys. 57, 4814–4816. Miwa, A., Ishibe, A., Nakano, M., Yamashira, T., Itai, S., Jinno, S., Kawahara, H., 1998. Development of novel chitosan derivatives as micellar carriers of taxol. Pharm. Res. 15, 1844–1850. Sharma, A., Straubinger, R.M., 1994. Novel taxol formulations: preparation and characterization of taxol-containing liposomes. Pharm. Res. 11, 889–896. Sharma, A., Conway, W.D., Struabinger, R.M., 1994. Reversed-phase high performance liquid chromatographic determination of taxol in mouse plasma. J. Chromatogr. B 655, 315–319. Sharma, U.S., Balasubramanian, S.V., Straubinger, R.M., 1995. Pharmaceutical and physical properties of paclitaxel (taxol) complexes with cyclodextrins. J. Pharm. Sci. 84, 1223–1230. Tarr, B.D., Sanbandam, T.G., Yalkowsky, S.H., 1987. A new parenteral emulsion for the administration of taxol. Pharm. Res. 4, 162–165.