Folate-targeted nanoparticles based on albumin and albumin/alginate mixtures as controlled release systems of tamoxifen: synthesis and in vitro characterization.

PURPOSE: Preparation and in vitro characterization of tamoxifen (TMX)-loaded folate-targeted nanoparticles based on disulfide bond reduced bovine serum albumin (BSA-SH) and BSA-SH/alginate-cysteine (BSA-SH/ALG-CYS) mixtures as drug delivery systems. METHODS: Folate-nanoparticles were characterized in terms of folate content, morphology, size, zeta potential, TMX load and drug release kinetics. Additionally, cell viability and cellular uptake of nanoparticles were determined using different cancer cell lines. RESULTS: Folic acid (FOL) was successfully attached to nanoparticles (ranging between 79 and170 µmol folate/g NP). Nanoparticles with 76-417 nm mean size were obtained and loaded with TMX (4.2-7.7 µg/mg NP). Zeta potential and drug extraction revealed major superficial placement of the drug, especially in the case of BSA/ALG-FOL systems. Drug release studies in the presence of surfactant showed a gradual release of the drug between 4-7 h. In general, low cytotoxicity of unloaded systems was found. Internalization of the systems was achieved and mediated by folate receptor, especially in the case of BSA NP-FOL. The administration of 10 µM TMX by TMX-FOL NP showed their efficacy as controlled TMX release systems. CONCLUSIONS: Promising anticancer action of these new TMX-loaded folate-targeted systems was demonstrated, allowing a new administration route to be studied in further in vivo studies in order to improve current TMX therapy.

Targeting tamoxifen to breast cancer xenograft tumours: preclinical efficacy of folate-attached nanoparticles based on alginate-cysteine/disulphide-bond-reduced albumin.

PURPOSE: In vivo evaluation of tamoxifen (TMX)-loaded folate-targeted nanoparticles prepared from a mixture of disulphide bond reduced bovine serum albumin (BSA-SH) and alginate-cysteine (ALG-CYS) as targeted delivery systems of TMX to tumour tissues. METHODS: TMX in solution, TMX included into folate-nanoparticles and their non-targeted analogues were intravenously administered to nude mice carrying xenograft MCF-7 tumours. The antitumor activity of these systems was characterized in terms of tumour growth rate, histological and immunohistochemical analysis of tumour tissues and TMX biodistribution. RESULTS: TMX-folate-attached nanoparticles caused tumour remission whereas free TMX or TMX-non-targeted nanoparticles could only stop the tumour development. The histological evaluation of tumour tissues showed that those treated with folate-conjugated systems presented the most quiescent and disorganized structures. Additionally, the lowest concentrations of TMX accumulated in non-targeted organs were also found after administration of the drug using this formulation. CONCLUSIONS: This study demonstrated that TMX-loaded folate-targeted systems were capable of reaching tumour sites, so enhancing the in vivo anticancer action of TMX, and allowing a new administration route to be applied and some of the current TMX therapy problems to be overcome.

Enhanced preclinical efficacy of tamoxifen developed as alginate-cysteine/disulfide bond reduced albumin nanoparticles.

Tamoxifen (TMX) is the most common clinical choice for the treatment of advanced or metastatic estrogen-dependent breast cancer. However, research on new challenging therapies is necessary due to its undesirable side effects and the limitation of the treatment only to the oral route. In this study, the antitumor activity of TMX-loaded nanoparticles based on different mixtures of alginate-cysteine and disulfide bond reduced bovine serum albumin was tested in vivo in MCF-7 nude mice xenograft model. These systems showed an enhancement of the TMX antitumor activity, since lower tumor evolutions and lower tumor growth rates were observed in mice treated with them. Moreover, histological and immunohistochemical studies revealed that treatments with TMX-loaded nanoparticles showed the most regressive and less proliferative tumor tissues. TMX biodistribution studies determined that TMX-loaded nanoparticles caused more accumulation of the drug into the tumor site with undetectable levels of TMX in plasma, reducing the possibility of delivering TMX to other not-targeted organs and, consequently, developing possible side effects. Thus, these TMX nanoparticulate systems are expected to provide a novel approach to the treatment of breast cancer in the future.

Use of poly(N-isopropylacrylamide) nanohydrogels for the controlled release of pimaricin in active packaging.

UNLABELLED: We propose here a delivery drug-polymer system using poly(N-isopropylacrylamide) (PNIPA) nanohydrogels that enables pimaricin to be protected from hostile environments and allows the controlled release of the antifungal through environmental stimuli. We synthesized 2 nanohydrogels, 1 with 100% N-isopropylacrylamide (PNIPA(5)) and 1 with 80% N-isopropylacrylamide copolymerized and 20% acrylic acid (PNIPA-20AA(5)). Both were then, loaded with a pimaricin aqueous solution. The pimaricin release profiles of these 2 nanohydrogels were considerably different: PNIPA(5) released 10% and PNIPA-20AA(5) released 30% with respect to the free pimaricin release. Moreover, the diffusion experiments showed that pimaricin was released from the PNIPA-20AA(5) nanohydrogel for up to 3 times longer than free pimaricin. Therefore, incorporating acrylic acid as comonomer into the PNIPA nanohydrogel resulted in a slower but more continuous release of pimaricin. The highest pimaricin levels were reached when the most hydrophilic nanohydrogel was used. The bioassay results showed that the pimaricin-nanohydrogel system was highly effective in inhibiting the growth of the indicator strain in conditions of thermal abuse. The spoilage in acidified samples stored under fluorescent lighting was reduced by 80.94% ± 33.02% in samples treated with a pimaricin-loaded nanohydrogel, but only by 19.91% ± 6.68% in samples treated with free pimaricin. Therefore, 2 conclusions emerge from this study. One is that the nanohydrogel delivery system could impede the degradation of pimaricin. The other is that the inhibitory effect of the antifungal on yeast growth is more pronounced when it is added included into the nanohydrogel to the food, especially in an acidic environment. PRACTICAL APPLICATION: This article presents relevant results on the use of nanohydrogels in food packaging. Nanohydrogels could provide protection so that the pimaricin remains active for a longer time. They also allow the controlled release of pimaricin, which thus regulates the unnecessary presence of the antifungal in the food.

Tamoxifen-loaded thiolated alginate-albumin nanoparticles as antitumoral drug delivery systems.

Nanoparticles based on disulfide bond reduced bovine serum albumin and thiolated alginate (alginate-cysteine conjugate) have been prepared by coacervation method and have been loaded with tamoxifen (TMX). The TMX load into the nanoparticles was optimized (4-6 µg/mg NP) by freeze-drying the systems before the loading procedure. Maximum TMX release (45-52%) took place between 2 and 25 h. Cytotoxicity of unloaded nanoparticles in MCF-7 and HeLa cells was not observed, although a small decrease in viability took place at very high concentration. Cell uptake of nanoparticles occurred in both cell types and the presence of polysaccharide in the nanoparticle composition allowed a better interaction with cells. The administration of 10 µM TMX by TMX-nanoparticles was effective in both cellular lines, and the effect of the drug-loaded systems on MCF-7 cell cycle showed the efficacy of the TMX-loaded nanoparticles.

Paclitaxel-loaded polyester nanoparticles prepared by spray-drying technology: in vitro bioactivity evaluation.

Paclitaxel (PTX), an antimicrotubular agent used in the treatment of ovarian and breast cancer, was encapsulated in nanoparticles (NPs) of poly(lactide-co-glycolide) (PLGA) and poly(ε-caprolactone) (PCL) polymers using the spray-drying technique. Morphology, size distribution, drug encapsulation efficiency, thermal degradation and drug release were characterized. MCF7 cells were employed to evaluate the efficacy of the systems on cell cycle and cytotoxicity. The particle size was in the range 0.8-1 µm. The incorporation efficiency of PTX was more than 80% in all NPs obtained. In vitro drug release took place during 35 days, and drug release rates were in the order PCL > PLGA 50:50 > PLGA 75:25. Unloaded NPs showed to be cytocompatible at MCF7 cells. PTX-loaded NPs demonstrated the release of the drug block cells in the G2/M phase. All PTX-loaded formulations showed their efficacy in killing MCF7 cells, mainly PTX-loaded PLGA 50:50 and PLGA 75:25 that cause a decrease in cell viability lower than 20%.

Structural and functional implications of the hexokinase-nickel interaction.

The interaction between nickel and yeast hexokinase was studied. The binding of nickel showed a positive cooperativity, and saturation was not reached. The nickel binding induced modifications in the secondary structure of the protein; thus, a lost of alpha helix and beta turns, as well as an increase of the random structure and beta sheet was observed. The monomer/dimmer equilibrium of the protein was modified in the presence of nickel, and the monomer state was mainly obtained at the highest nickel concentrations studied. These changes on the protein structure caused a decrease in the enzyme activity. According to kinetic studies, nickel caused a non-competitive inhibition when glucose was the variable substrate and a linear competitive inhibition when ATP was the variable substrate.

5-Fluorouracil-loaded microspheres prepared by spray-drying poly(D,L-lactide) and poly(lactide-co-glycolide) polymers: characterization and drug release.

5-Fluorouracil (5-FU), a hydrosoluble anti-neoplastic drug, was encapsulated in microspheres of poly(D,L-lactide) (PLA) and poly(lactide-co-glycolide) (PLGA) polymers using the spray-drying technique, in order to obtain small size microspheres with a significant drug entrapment efficiency. Drug-loaded microspheres included between 47 +/- 11 and 67 +/- 12 microg 5-FU mg(-1) microspheres and the percentage of entrapment efficiency was between 52 +/- 12 and 74 +/- 13. Microspheres were of small size (average diameter: 0.9 +/- 0.4-1.4 +/- 0.8 microm microspheres without drug; 1.1 +/- 0.5-1.7 +/- 0.9 microm 5-FU-loaded microspheres) and their surface was smooth and slightly porous, some hollows or deformations were observed in microspheres prepared from polymers with larger Tg. A fractionation process of the raw polymer during the formation of microspheres was observed as an increase of the average molecular weight and also of Tg of the polymer of the microspheres. The presence of 5-FU did not modify the Tg values of the microspheres. Significant interactions between the drug and each one of the polymers did not take place and total release of the included drug was observed in all cases. The time needed for the total drug release (28-129 h) was in the order PLA > PLGA 75/25 > PLGA 50/50. A burst effect (17-20%) was observed during the first hour and then a period of constant release rate (3.52 +/- 0.82-1.46 +/- 0.26 microg 5-FU h(-1) per milligram of microspheres) up to 8 or 13 h, depending on the polymer, was obtained.

Cytarabine release from comatrices of albumin microspheres in a poly(lactide-co-glycolide) film: in vitro and in vivo studies.

Cytarabine (ara-C) was included in albumin microspheres and these microspheres were immersed in a poly(lactide-co-glycolide) (PLGA) film to constitute a comatrix system to develop a prolonged form of release. Cytarabine-loaded albumin microspheres were synthesized by emulsion, and 25 or 50 mg of drug were included in the disperse phase. Thus, microspheres with 46+/-4 microg drug/mg microspheres and 50+/-5 microg drug/mg microspheres were obtained, which means a percentage of incorporation efficiency of 42+/-4% and 25+/-2%, respectively. These cytarabine-loaded microspheres were used to prepare PLGA-comatrices. Kinetic release studies indicated that total cytarabine release only takes place in the presence of protease, probably due to the fact that glutaraldehyde establishes covalent links with the amine side group of the drug and cross-links it with the protein matrix. A slower kinetic release of the drug was obtained from PLGA-comatrices, although only 80% of the included cytarabine was released on day 7. The comatrices were subcutaneously implanted in the back of rats and in both cases the ara-C administered dose was 36 mg of ara-C per kg of body weight. The drug was detected in plasma 10 days. The mean residence time (MRT) of the drug administered by these comatrices was 87-91 times larger when compared to the value obtained when the drug was administered in solution by intraperitoneal injection. The histological studies show that a degradative process of the comatrices takes place. The comatrices do not damage surrounding tissue; a normal regeneration of the implanted zone was observed.

Sustained release of bupivacaine from devices based on chitosan.

Chitosan beads loaded with bupivacaine (16+/-3 microg of drug per milligram of beads) were prepared by cross-linking with glutaraldehyde. In vitro drug release at pH and temperature conditions similar to those of the biological systems were studied. Maximum release of bupivacaine was obtained between 100 and 120 h, depending on the presence of lysozyme in the release medium, since the enzyme facilitates the release process. A constant release rate of the drug, between 11 and 15 microg/h, was observed for 30 h. In order to prolong bupivacaine release, the drug-loaded chitosan beads were coated with a poly(DL-lactide-co-glycolide) film. The resulting device allows the drug to be released in a sustained form; a constant release rate between 28.5 and 29.5 microg/h was obtained for 3 days, and the maximum release of bupivacaine took place at day 9. The in vitro results indicate a possible application of these bupivacaine loaded chitosan systems as drug release devices with an analgesic action. Thus, they could be used in the treatment of dental pain in the buccal cavity, where drug release would be made easier by lysozyme of the saliva.

Transdermal application of bupivacaine-loaded poly(acrylamide(A)-co-monomethyl itaconate) hydrogels.

Bupivacaine, an amide local anaesthetic agent of long-acting and intense anaesthesia, was incorporated into poly(acrylamide(A)-co-monomethyl itaconate (MMI)) hydrogels. The swelling behaviour of two gel compositions, without drug, 75A/25MMI and 60A/40MMI, through rabbit ear skin, mounted on a modified Franz diffusion cell, was studied. Both gel compositions reach the equilibrium swelling degree (88.9+/-0.7 wt.% for 75A/25MMI and 92.5+/-0.1 wt.% for 60A/40MMI). The swelling kinetics was in accordance with the second Fick's Law; diffusion coefficients indicate faster swelling for gels with lower amount of monomethyl itaconic acid. The skin flux of bupivacaine solution through rabbit ear skin was 105+/-24 microg/cm(2)/h, the effective permeability coefficient was 26 x 10(-3)+/-9 x 10(-3)cm/h, and 77+/-15% of bupivacaine was permeated. Bupivacaine-loaded gels allow the drug was permeated through the skin. 47+/-4% and 36+/-3% of the drug amount included in 75A/25MMI and 60A/40MMI hydrogels, respectively, was permeated. The skin flux of the drug was between 90+/-5 and 16+/-7 microg/cm(2)/h depending on the amount of bupivacaine included in the gel and the gel composition. Skin flux increases with the drug load of the gels. Furthermore, as more MMI in the gel slower skin flux of the drug due to bupivacaine-gel interactions.

Preparation of bupivacaine-loaded poly(epsilon-caprolactone) microspheres by spray drying: drug release studies and biocompatibility.

Poly(epsilon-caprolactone) microspheres containing bupivacaine were prepared by the spray-drying process. The average size of drug loaded microspheres was less than 3 microm in diameter, and the percentage of entrapment efficiency was 91 +/- 3%. In vitro drug release kinetic in phosphate buffer at 37 degrees C showed a hyperbolic profile, with a burst-effect during the first hour. Subcutaneous injection of bupivacaine-loaded microspheres in the back of rats caused an increase in drug concentration in plasma. Maximum bupivacaine concentration in plasma was 237 +/- 58 ng/ml at 105 h, and drug was detected in plasma for 16 days. The half-life time of the drug was increased by more than 125 times with regard to that of the drug administered in a solution by intraperitoneal injection. After 30 days of injection, a mass formed by microspheres surrounded by a thin fibrous capsule was observed. Small blood vessels and multinucleate foreign body giant cells with macrophagic function around microspheres were detected. After 60 days of injection a subcutaneous mass was also observed, which was formed of more degraded dispersed microspheres in conjunctive tissue, which had a normal structure. Thus, bupivacaine-loaded poly(epsilon-caprolactone) microspheres could be considered as a device to be used in the treatment of severe pain that is not responsive to opioids for example in cancer-related syndromes or in intractable herpetic neuralgia.

Effect of cadmium acetate on the conformation of lysozyme: functional implications.

Structural variations of lysozyme as a consequence of its interaction with CdAc2, as well as the implications on the protein functionality have been studied. Variations in the conformation of the macromolecule are seen, however these changes are not reflected on the secondary structure. The interaction of the salt with the polypeptide chain is weak and thermodynamically unfavourable. Molecular aggregates (dimer forms) are observed at the highest salt concentrations. This interaction causes an inhibitory effect on lysozyme, the activity loss being 50% at the highest salt concentration studied. The inhibition is of mixed type with an uncompetitive component. Thus cadmium does not bind to the active site of the enzyme which is in accordance with the not very large activity loss observed. The substrate inhibition of lysozyme is favoured in the presence of the salt, so interaction with the macromolecule is at low affinity sites.

In vitro controlled release of bupivacaine from albumin microspheres and a co-matrix formed by microspheres in a poly(lactide-co-glycolide) film.

Albumin microspheres cross-linked with glutaraldehyde and loaded with bupivacaine, a local anaesthetic, were synthesized (138 +/- 59 microm diameter). A matrix formed by bupivacaine-loaded microspheres in a poly(lactide-co-glycolide) film was prepared in order to improve the controlled release of the drug. In vitro release of the drug was determined in phosphate buffer at 37 degrees C in the absence and in the presence of protease type VIII to mimic a biological system. The effect of temperature and protease on bupivacaine as a function of time was examined; both of them cause a degradative effect on the drug. A rapid release (60 +/- 8% of the drug) takes place at 1 h, and maximum release is found at 50 +/- 6 h from microspheres with swelling. In the presence of protease, maximum release of bupivacaine from microspheres is found at 28 +/- 2 h; the microspheres disappear at 8 days. Inclusion of bupivacaine-loaded microspheres in a poly(lactide-co-glycolide) film causes a slower release of the drug, up to 18 days, with swelling. In the presence of protease, the polymer protects bupivacaine-loaded microspheres from degradation, which takes place at 20 days.

Analysis of aluminum-yeast hexokinase interaction: modifications on protein structure and functionality.

The aluminum and yeast hexokinase interaction was studied. Structural changes were correlated with variations in protein functionality. Results show two different behaviors: At low metal concentrations preferential adsorption of metal (and water exclusion) induces aggregate formation. No significant changes in the protein structure occur, but there is a continuous loss of activity (from the first concentration). At large salt concentrations a monomerization process and a conformational change in the secondary structure as well as in the three-dimensional structure take place. This change reduces the percentage of alpha-helix conformation, gives thermal stability to the protein, and allows the exposure of some tryptophan residue and hydrophobic regions. The protein inhibition increases. Conformational change and monomerization may allow access of the metal to the substrate site, mainly the ATP site. The inhibition in any case is of mixed type with a competitive component.

Chitosan microspheres in PLG films as devices for cytarabine release.

Cytarabine was included in chitosan microspheres and several of these microspheres were embedded in a poly(lactide-co-glycolide) (PLG) film to constitute a comatrix system, to develop a prolonged release form. Chitosan microspheres, in the range of 92+/-65 microm, having good spherical geometry and a smooth surface incorporating cytarabine, were prepared. The cytarabine amount included in chitosan microspheres was 43.7 microg of ara-C per milligram microsphere. The incorporation efficiency of the cytarabine in microspheres was 70.6%. Total cytarabine release from microspheres in vitro was detected at 48 h. Inclusion of cytarabine-loaded microspheres in poly(lactide-co-glycolide) film initiated a slower release of the drug and, in this way, the maximum of cytarabine released (80%) took place in vitro at 94.5 h. Comatrices, with 8.7 mg of cytarabine, signifying a dose of 34.5 microg/kg, were subcutaneously implanted in the back of rats. Maximum plasma cytarabine concentration was 18.5+/-1.5 microg/ml, 48 h after the device implantation and the drug was detected in plasma for 13 days. The histological studies show a slow degradative process. After 6 months of implantation, most of the microspheres of the matrix seemed to be intact, the comatrix appeared surrounded by conjunctive tissue and small blood vessels and nerve packets were detected in the periphery of the implant.

In-vivo drug delivery of 5-fluorouracil using poly(2-hydroxyethyl methacrylate-co-acrylamide) hydrogels.

Poly(2-hydroxyethyl methacrylate-co-acrylamide) hydrogels crosslinked with ethylen glycol dimethacrylate were used as devices for the in-vivo drug release of 5-fluorouracil (5-FU). Drug-loaded hydrogels were subcutaneously implanted in the back of Wistar rats. All hydrogel discs reached an equilibrium swelling degree, which was slightly larger than that determined in-vitro. After 30 days of implantation, the hydrogel discs were transparent, and without fracture or apparent degradation. In addition, a fibrous capsule was not detected around the hydrogels that had greater hydration degrees. Release of 5-FU from these hydrogels allows the drug to remain in the plasma from 1 to 5 days, in spite of its short plasma half-life (15 min). This was an improvement of up to 98-times compared with the intraperitoneal drug administration. Administration of 5-FU by implantation of 2-hydroxyethylmethacrylate-co-acrylamide copolymeric hydrogels seems to be a good candidate for 5-FU therapy, since the drug released results in a therapeutically suitable plasma concentration of 5-FU for an extended period of time, despite the short half-life of the drug.

Bupivacaine-loaded comatrix formed by albumin microspheres included in a poly(lactide-co-glycolide) film: in vivo biocompatibility and drug release studies.

Bupivacaine-loaded comatrix, formed by bupivacaine-loaded microspheres included in a poly(lactide-co-glycolide) film, was assayed for the controlled release of the drug 'in vivo'. The comatrix, with 66.37 microg of bupivacaine, signifying a dose of 265.5 microg/kg, was subcutaneously implanted in the back of rats. Maximum plasma bupivacaine concentration was 147.6 +/- 5.0 ng/ml 95 h after the device implantation, and the drug was detected in plasma for 17 days. The half-life time of bupivacaine improves by more than 50 times with regard to that of the drug administered in a solution by intraperitoneal injection. After 15 days of implantation the comatrix was included in a thin fibrous capsule and degradation of the microspheres was observed. The histological studies show good biocompatibility of this comatrix. After 50 days the comatrix was degraded and its remains were almost indistinguishable from the surrounding tissue. Small number of microspheres was observed and they were surrounded by conjunctive tissue. Nerve packets and small blood vessels were also observed in the periphery of the implant.

Poly(acrylamide-co-monoethyl itaconate) hydrogels as devices for cytarabine release in rats.

This study has tested the application of three different copolymeric poly(acrylamide-co-monoethyl itaconate; A/MEI) hydrogels, 90A/10MEI, 75A/25MEI and 60A/40MEI, on the release of cytarabine (ara-C). The drug was incorporated in gels by placing it in the polymerization feed mixture and discs loaded with 5-50 mg ara-C were obtained. The amount of swelling at equilibrium in saline solution (NaCl, 0.9% w/w) was between 78 and 82% w/w, depending on the composition of the copolymer. The diffusion studies followed Fick's second law. The diffusion coefficients for swelling of the gels were between 9.30 x 10(-11) m2 s(-1) and 37.42 x 10(-11) m2 s(-1); those for release of ara-C were between 3.42 x 10(-11) m2 s(-1) and 10.25 x 10(-11) m2 s(-1). The activation energies for swelling were in the range 16.60 +/- 2.59-21.85 +/- 1.78 kJ mol(-1); those for ara-C release were 28.13 +/- 3.1-29.7 +/- 4.6 kJ mo(-1). To determine the applicability of these copolymers, 75A/25MEI gel was subcutaneously implanted in rats and the plasma concentration of the drug was determined by high-performance liquid chromatography. The concentration of ara-C in plasma (range 17.67 +/- 5.68-10.76 +/- 2.15 microg mL(-1)) was maintained during the first stages (2-8 h) and no drug was detected after 32 h. This route of administration was compared with intraperitoneal injection of the drug. In conclusion, ara-C can be incorporated in hydrogels and released in a pharmacologically active form. The concentration of ara-C in plasma is maintained for long enough to improve therapeutic results.

Slow releasing of ara-C from poly(2-hydroxyethyl methacrylate) and poly(2-hydroxyethyl methacrylate-co-N-vinyl-2-pyrrolidone) hydrogels implanted subcutaneously in the back of rats.

The release of cytarabine (ara-C) from poly(2-hydroxyethyl methacrylate) and poly(2-hydroxyethyl methacrylate-co-N-vinyl-2pyrrolidone) hydrogels cross-linked with different amounts of ethyleneglycol dimethacrylate (EGDMA) 'in vivo' has been studied. Two ara-C loaded hydrogel discs, each with 25 mg of the drug, were subcutaneously implanted in the back of male Wistar rats. Total ara-C dose was 230 mg kg(-1). Ara-C and ara-U plasmatic concentration were determined by HPLC. Periods of constant drug concentration are observed from all gels. Ara-C concentrations in the steady-state are between 19.0 +/- 2.0 and 2.2 +/- 0.8 micromol l(-1). The release time of ara-C was between 3 days from pH EMA 0.5% and 16 days from H80/VP20/E15 gels. These results are very different of that obtained when ara-C is administered by intraperitoneal injection, in this case peaks of maximum concentration (between 24 +/- 1 and 3.9 +/- 0.4 microg ml(-1)) 30 min after the injection are originated, and no drug is detected 4 h after the injection.