International Journal of Pharmaceutics 468 (2014) 258–263
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
b-Cyclodextrin-dextran polymers for the solubilization of poorly soluble drugs Massimiliano di Cagno a, *, Thorbjørn Terndrup Nielsen b , Kim Lambertsen Larsen b , Judith Kuntsche a , Annette Bauer-Brandl a a b
Department of Physics, Chemistry and Pharmacy, University of Southern Denmark, Campusvej 55, Odense M 5230, Denmark Department of Biotechnology, Chemistry and Environmental Engineering, Aalborg University, Sohngaardsholmsvej 57, Aalborg 9000, Denmark
A R T I C L E I N F O
A B S T R A C T
Article history: Received 12 March 2014 Received in revised form 9 April 2014 Accepted 12 April 2014 Available online 16 April 2014
The aim of this study was to assess the potential of novel b-cyclodextrin (bCD)-dextran polymers for drug delivery. The size distribution of bCD-dextrans (for eventual parenteral administration), the influence of the dextran backbones on the stability of the bCD/drug complex, the solubilization efficiency of poorly soluble drugs and drug release properties were investigated. Size analysis of different bCDdextrans was measured by size exclusion chromatography (SEC) and asymmetrical flow field-flow fractionation (AF4). Stability of drug/bCD-dextrans was assessed by isothermal titration calorimetry (ITC) and molar enthalpies of complexation and equilibrium constants compared to some commercially available bCD derivatives. For evaluation of the solubilization efficiency, phase-solubility diagrams were made employing hydrocortisone (HC) as a model of poorly soluble drugs, whereas reverse dialysis was used to detect potential drug supersaturation (increased molecularly dissolved drug concentration) as well as controlled release effects. Results indicate that all investigated bCD-polymers are of appropriate sizes for parenteral administration. Thermodynamic results demonstrate that the presence of the dextran backbone structure does not affect the stability of the bCD/drug complex, compared to native bCD and commercially available derivatives. Solubility studies evidence higher solubilizing abilities of these new polymers in comparison to commercially available bCDs, but no supersaturation states were induced. Moreover, drug release studies evidenced that diffusion of HC was influenced by the solubilization induced by the bCD-derivatives. ã 2014 Elsevier B.V. All rights reserved.
Keywords: Cyclodextrins Calorimetry (ITC) bCD-dextran polymers Solubilization Poorly soluble drugs
1. Introduction Poor solubility of active pharmaceutical ingredients (API) is one of the biggest challenges in drug development: a large number of drugs on the market (approximately 40%), and even more those under development (approx. 70%), belong to class II, or respectively class IV of the Biopharmaceutics Classification System (BCS) and are therefore associated with solubility issues and as a consequence, poor oral bioavailability (Williams et al., 2013; Fahr and Liu, 2007; Amidon et al., 1995). In the last few years, a number of techniques have been suggested to increase bioavailability of such APIs (Williams et al., 2013; Fahr and Liu, 2007; Zhang et al., 2008). Amongst these, the use of cyclodextrins (CDs) has emerged as an
* Corresponding author. Tel.: +45 65502504; fax: +45 66158780. E-mail addresses: [email protected] (M. di Cagno), [email protected] (T. Terndrup Nielsen), [email protected] (K. Lambertsen Larsen), [email protected] (J. Kuntsche), [email protected] (A. Bauer-Brandl). http://dx.doi.org/10.1016/j.ijpharm.2014.04.029 0378-5173/ ã 2014 Elsevier B.V. All rights reserved.
efficient approach for solubilization. Cyclodextrins are cyclic oligosaccharides consisting of (a-1,4)-linked a-D-glucopyranose units. CD molecules are shaped like truncated cones, with a hydrophobic cavity inside and a hydrophilic surface outside. For many years CDs have been intensively investigated for pharmaceutical use due to their ability to spontaneously complex poorly soluble drug molecules, inducing solubilization (enhancement in apparent solubility) (Loftsson et al., 2002, 2004; Connors, 1997). For pharmaceutical purposes, b-cyclodextrins (bCDs, 7 glucose units in the ring) are mostly studied. Although bCDs are hydrophilic (as well as their complexes), their aqueous solubility is comparably low. The reason for this fact is associated with the high crystal lattice energy (Loftsson and Brewster, 2010) as well as intramolecular hydrogen bond formation that compromise the interaction with water molecules (solvation). To overcome this limitation, derivatives of the bCDs with higher aqueous solubility have been developed by introducing substituents at specific positions of the glucose rings, breaking the intramolecular net of hydrogen bonds and thereby enhancing aqueous solubility. The
M. di Cagno et al. / International Journal of Pharmaceutics 468 (2014) 258–263
most commonly used derivatives are methyl-b-cyclodextrin (MbCD), hydroxypropyl-b-cyclodextrin (HPbCD), and sulfobutylether-b-cyclodextrins (SBEbCD) (Brewster et al., 2008). Some of these bCD derivatives are of high solubility (e.g., methylderivatives), but were found to have strong systemic toxicity. Among the different derivatives, hydroxypropyl and sulfobutylether bCDs were found to have the lowest hemolytic effect and are therefore regarded suitable for parenteral administration (as well as some a- and g-cyclodextrin derivatives with 6 and 8 glucose units, respectively) (Loftsson and Brewster, 2010). Moreover, it has been shown that pharmacokinetic parameters of a number of drugs are not significantly modified after parenteral co-administration with these specific types of bCD derivatives (Kurkov et al., 2012). The general mechanism of the complexing reaction comprises exothermic interaction between the drug molecules and the CD cavity (negative enthalpy value), and the replacement of water from the cavity by the drug molecule (negative entropy). Furthermore, the presence of small quantities of hydrophilic polymers (such as PVP) or specific salts (e.g., sodium acetate or sodium salicylate) together with bCDs has previously been shown to stabilize drug/CD complexes through increased drug/CD interaction (enthalpy) or by formation of water-soluble drug/CD aggregates (Loftsson et al., 1994; Loftsson and Brewster, 2012). It has also been described that hydrophilic polymers can induce supersaturation states (increased molecularly dissolved drug concentration) (Di Cagno and Luppi, 2013). In order to overcome low aqueous solubility of native bCD and to take advantage of the stabilizing effect that hydrophilic polymers may have on the complex drug/CD, a new series of polymers based on hydrophilic backbones (i.e., dextrans) with conjugated bCD units have been synthetized and studied (Nielsen et al., 2010, 2009). The advantages of these CD-modified polymers are their much higher aqueous solubility compared to native bCD and the absence of substituents that may lead to steric hindrance at the CD opening. It should also be mentioned that FDA acknowledges dextran as GRAS when in use as food additives. Dextran is in pharmaceutical use as a plasma expander (Roberts and Bratton, 1998) and it is biodegraded by enzymatic activity. Considering all these facts, bCD-dextran polymers appear very promising as solubilizing agents for poorly
259
soluble drugs and drug delivery vehicles. In the present work, we studied the suitability of some new bCD-dextran polymers for drug delivery purposes. We investigated size distribution of different bCD-dextrans to evaluate their potential for parenteral administration, the influence of the dextran backbones on the stability of the bCD/drug complexes, solubilization efficiencies of a selected poorly soluble drug and drug release properties. 2. Materials and methods 2.1. Materials Three bCD modified dextrans (Fig. 1) were synthetized as described earlier (Nielsen et al., 2010). bCD, HPbCD with a number of substituent per glucose unit (molar degree of substitution (MS)) of 0.87 and MbCD (MS of 0.57) were a generous gift from Roquette Freres (Lestrem, France). Dextrans of different molecular weights (10, 20 and 25 kDa) were purchased from Pharmacosmos A/S (Holbaek, Denmark). Sodium di-hydrogen phosphate (NaH2PO4), di-sodium hydrogen phosphate (Na2HPO4) and sodium chloride (NaCl) (Sigma–Aldrich Chemie GmbH, Steinheim, Germany) were used for the preparation of the isotonic (275–285 mOsm, SemiMicro Osmometer K-7400, Herbert Knauer GmbH, Berlin, Germany) and isohydric (pH 7.4, pH meter Lab 827, Metrohm AG, Herisau, Switzerland) buffer saline (73 mM). Ibuprofen (IBP, Caesar & Lorenz GmbH, Hilden, Germany) and hydrocortisone (HC, Sigma–Aldrich) were employed as model drugs (Fig. 1). 2.2. Molecular mass determination 2.2.1. Size exclusion chromatography (SEC) Size exclusion chromatography was performed employing a TSK-gel column type SW 4000-3000, and polymer sizes were detected by a Wyatt miniDAWN 3 angle light scattering detector (MALS) (Wyatt Technology Inc., Santa Barbara, USA) and RIdetector (Optilab rEX, Wyatt) using a refractive index increment (dn/dc) of 0.15 mL/g. A 0.1 mM lithium nitrate (LiNO3) solution preserved with 0.05% sodium azide was used as eluent at a flow rate of 1 mL/min.
Fig. 1. Chemical structures of bCD-dextran derivatives (A), ibuprofen (B) and hydrocortisone (C).
260
M. di Cagno et al. / International Journal of Pharmaceutics 468 (2014) 258–263
2.2.2. Asymmetrical flow field-flow fractionation (AF4) Asymmetrical flow field-flow fractionation (Eclipse 3+ AF4, Wyatt Europe, Dernbach, Germany) was employed for comparison to determine the molecular masses for the different bCD-dextran polymers. The trapezoidal separation channel (length 265 mm, width 350 mm) equipped with a membrane of regenerated cellulose (molecular weight cut off 5 kDa, Wyatt Europe) was coupled with a UV–vis detector (Agilent Technologies GmbH, Boeblingen, Germany), a refractometry detector (Optilab rEX) and a multi-angle light scattering detector (DAWN HELEOS II) (both from Wyatt Technology Inc., Santa Barbara, USA). A 50 mM sodium chloride solution preserved with 0.02% sodium azide and filtered through a 0.01 mm membrane was used as an eluent. The samples were injected over 2 min (focus flow 2 mL/min) and eluted either applying constant cross flow (3 mL/min over 20 min initially followed by a cross flow decreasing from 3 to 0 mL/min within 5 min) or by only applying a cross flow gradient (cross flow decreasing from 3 to 0.1 mL/min over 40 min). In both cases, an elution step without cross flow finally was applied to ensure complete elution. Molar masses were calculated with Astra software (version 6.0.1, Wyatt) based on the light scattering and RI detector signals (baseline RI signal were subtracted) and using refractive index increment (dn/dc) of 0.14 mL/g. The average molecular weights are given together with the polydispersity values (Mw/Mn), whereas a polydispersity value of one corresponds to a complete homogeneous (monodisperse) sample. 2.3. Degree of substitution determination (DS) Determination of the degree of substitutions of the bCDdextran polymers was made by NMR spectroscopy using a Bruker DRX400 spectrometer (Bruker BioSpin GmbH, Rheinstetten, Germany). Samples were measured at 20 mg/mL in D2O at 42 C and the DS was calculated by comparison of the integral of the characteristic triazol proton and the anomeric protons of bCD and dextran as previously described (Nielsen et al., 2009). 2.4. Evaluation of complexation efficiency Calorimetric analyses were performed at 25 C using a 2277 Thermal Activity Monitor, TAM (Thermometric, Järfälla, Sweden) following a method previously described (Di Cagno et al., 2011). Small volumes (2.5 mL) of titrand solution (0.6 mM IBP isotonic and isohydric solution in phosphate buffer saline) were placed into 5 mL steel vessels (sample and reference). A 250 mL syringe (Hamilton, Bonaduz, Switzerland) filled with the titration solution containing the bCD-dextrans (Table 1) was inserted in the 612 Lund Syringe Pump (Thermometric). Fifteen drops of 12 mL volume each were titrated at 45 min intervals to complete the complexation reaction of the IBP solution. Each titration experiment was performed in triplicate. The values of standard enthalpy (DH ), mass equilibrium constants (K) and the amount of bCD units expressed as a concentration were calculated using the 2277-131 Digitam software by fitting the raw experimental data with titration curves, assuming a 1:1 stoichiometry of complexation between bCD units and IBP (as suggested earlier for polyethylene glycol bCD derivatives (Nielsen et al., 2009)). Table 1 Characteristics of the different bCD-dextrans titrant solutions used in the isothermal titration calorimetry experiments.
bCD-dextran (code)
pH
D10GPbCD D20GPbCD D25GPbCD
7.37 282 7.38 283 7.36 277
Osmolality (mOsm/ kg)
Polymer concentration (mg/ mL) 17.5 14.0 22.5
2.5. Phase-solubility studies HC was added in excess to aqueous solutions of different concentrations (3–4 mM) of the respective bCD-dextran polymer and commercial bCD derivatives for comparison in isotonic PBS (pH 7.4). After shaking for 72 h in a thermostatic water bath at 25 C (Julabo SW 23, Julabo Labortechnik GmbH, Seelbach, Germany), the solutions were filtered through 0.45 mm pore size Minisart RC4 filters (Sartorious Stedim Biotech GmbH, Göttingen, Germany). Concentrations of HC were detected in the supernatant by UV–vis spectroscopy (maximum wavelength 248 nm) by a Genesis 10 UV/ VIS spectrophotometer (Thermo Electron Corporation, Cambridge, UK). 2.6. Dialysis studies The dialysis method previously described (Di Cagno and Luppi, 2013) was employed to characterize possible supersaturation of HC induced by the bCD-dextrans, as well as to evaluate drug release from these polymers. Frantz (permeability) cells (SES GmbHAnalysesysteme, Bechenheim, Germany) with surface areas of 1 cm2, small chamber volume of 2 mL and big chamber volume of 8 mL were employed. Reverse dialysis setup was chosen: the big chamber was used as a donor compartment and filled with drug suspension and drug/bCD-dextrans saturated solutions (prepared as described in Section 2.5, the concentration of HC is reported in Table 3). Pre-washed dialysis membranes (Sigma–Aldrich) with a molecular weight cut-off (MWCO) of 10 kDa were placed between the donor and acceptor chambers. The MWCO was chosen to prevent bCD-dextrans to permeate, whereas HC could cross the membrane without impediment. At time zero (start of the experiment) the acceptor compartment was filled with 1.8 mL of fresh buffer. For supersaturation studies (no in-sink conditions needed), sampling of acceptor solution (0.2 mL) was performed after 50 h (reaching equilibrium), whereas, for the drug release studies, samples of 0.5 mL at 30 min intervals were withdrawn over five hours (replacement of the withdrawn volume with an equal amount of fresh buffer). Drug concentration was detected by UV–vis spectroscopy and each experiment was carried out in three replicates. 2.7. Statistic analysis Student’s t-test was performed to evaluate the significance of the differences identified between the data set obtained for each bCDdextran studied. P 0.05 was considered significantly different. 3. Results 3.1. Molecular weight determination and degree of substitution The actual molecular weights of the polymers were determined by both size exclusion chromatography and AF4. For all bCDdextrans, molecular weights detected by AF4 were in reasonable agreement with the results obtained by the SEC technique (Table 2). Polydispersity values were in the range between 1.247 (D10GPbCD solutions) and 1.512 (for D25GPbCD solutions). The degrees of substitution of all bCD-dextran polymers were between 55% (D10GPbCD) and 63% (D25GPbCD) (Table 2). No significant aggregation was macroscopically observed over time. 3.2. Complexation efficiency studies The standard enthalpy and Gibbs free energy of complexation between bCD-dextran polymers and IBP are reported in Fig. 2 in comparison to results obtained previously for some bCD derivatives (Di Cagno et al., 2011). Standard enthalpies of complexation were not
M. di Cagno et al. / International Journal of Pharmaceutics 468 (2014) 258–263
261
Table 2 Molecular weights (Mw), polydispersity and degree of substitutions (DS) identified for the different bCD-dextran derivatives. Relative standard deviations (SDrel) are provided.
bCD-dextran
Dextran
Dextran
bCD-dextran
bCD-dextran
(Code)
Mw (kDa)
Polydispersity (Mw/Mn)
nominal from manufacturer
Mw Mw (kDa) (kDa) AF4 (n = 2–4, SDrel 4%)
Mw (kDa) SEC (n = 3, SDrel 5%)
DS (% w/w) NMR (SDrel 5%)
10 20 25
16 20 31
1.247 1.503 1.512
25 54 83
55 57 63
D10GPbCD D20GPbCD D25GPbCD
bCD-dextran
31 67 83
the same for all modified polymers. Whereas D10GPbCD, D20GPbCD expressed enthalpies of complexation (12.8 0.9 and 14.7 0.7, respectively) which were in the same range as that of native bCD (14.6 0.7 kJ/mol), D25GPbCD expressed higher (P < 0.05) energy of complexation (19.3 0.8 kJ/mol) in comparison to the other polymers and standard bCD derivatives.
amount of HC molecularly dissolved in the aqueous phase. It can thus be concluded that the observed increase in apparent solubility is exclusively due to complexation of HC with the bCD units bonded to the dextran backbones.
3.3. Solubility and true supersaturation studies
Permeability studies with dialysis membranes were performed in order to characterize if these polymers were affecting the diffusion of drug molecules in aqueous environment, in comparison to the drug in the crystalline form. In the case of crystalline drug suspension, the flux of HC was different in comparison to bCD-dextrans solutions (Fig. 4). Specifically, in the case of the crystalline drug suspension, the flux was significantly (P 0.05) higher in comparison with all bCD-dextran derivatives (Table 3). Apparent permeability coefficients (Papp) were also calculated using as initial donor concentration the fraction of molecularly dissolved drug and the results are reported in Table 3.
All solubility studies were performed in isotonic PBS pH 7.4 and phase-solubility diagrams for HC in the presence of bCD-dextrans and classical bCDs are reported in Fig. 3. For all bCD-dextrans, starting from the thermodynamic solubility of HC at 25 C (0.90 0.03 mM), the apparent solubility of HC increases linearly with increasing concentration of the polymers (reported as bCD unit concentration), while bCD (and analogues of higher solubility) were not capable of inducing solubilization for this drug. All bCDdextrans lead to a higher apparent solubility for HC. Specifically, the apparent solubilities in the case of D10GPbCD, D20GPbCD and D25GPbCD were 3.11 0.31 mM, 4.05 0.14 mM and 3.21 0.09 mM, respectively, whereas for native bCD, MbCD and HPbCD were much lower (1.38 0.31 mM, 1.54 0.05 mM and 1.49 0.15 mM, respectively). In contrast to what has been observed for other polymers (e.g., PVP (Loftsson et al., 1994)), no enhancement of apparent solubilities of the drug by any of the native unmodified dextrans has been observed. Characterization of the fraction of molecularly dissolved HC indicates no increase of molecularly dissolved drug induced by the bCD dextrans (no supersaturation observed, Table 3). On the contrary, it appears that the presence of bCD-dextrans results in a certain reduction of the
3.4. bCD-dextrans drug release study
4. Discussion The aim of this work was to explore the potential of bCDdextran polymers in drug delivery. All bCD-dextrans sizes were studied in aqueous environment to confirm homogeneity of solutions and to detect eventual aggregates that may hamper their use for aqueous parenteral formulations. AF4 is a very powerful and sensitive technique for analyzing molecular weights of polymers and would detect aggregation in solutions. All bCDdextrans studied were of appropriate sizes for potential parenteral administration (Wong et al., 2008) and the molecular weighs
Fig. 2. Thermodynamic parameters of complexation of the model drug ibuprofen with bCD-dextrans (D10GPbCD, D20GPbCD and D25GPbCD) and some bCDs (HPbCD, MbCD and native bCD). Results are reported as mean standard deviation (n = 3).
262
M. di Cagno et al. / International Journal of Pharmaceutics 468 (2014) 258–263
Fig. 3. Phase-solubility diagram of hydrocortisone (HC) in the presence of different types of bCDs and bCD-dextrans. Results are reported as mean standard deviation (n = 3).
detected via the AF4 technique were in good agreement with the results obtained by SEC analysis. The discrepancies between the results obtained with the two techniques are not larger than expected and are probably due to the different separation principles of the two techniques. Moreover, in the SEC experiments, a three angles light scattering detector was employed whereas in the AF4 analysis, an 18 angles light scattering detector was employed. The average polydispersity value was 1.43, indicating a good homogeneity of the samples. IBP was chosen as model drug for thermodynamic complexation studies for comparability with previous results (Di Cagno et al., 2011). All data have been analysed assuming a 1:1 stoichiometry of reaction between IBP molecules and the bCD units (Xing et al., 2009). Results indicate that enthalpies of complexation between bCD units and IBP are comparable with those measured earlier for classical bCD derivatives (hydroxypropyl- and methyl-moieties, Fig. 2). Just in the case of D25GPbCD the complexation reaction was slightly more exothermic. However, it should be noted that for all bCD-dextrans, Gibbs free energies were comparable (no significant differences, P 0.05) to native bCD and derivatives (HPbCD and MbCD), indicating classical entropy–enthalpy compensation. From these studies it can be concluded that stabilities of the bCD-drug complex is not negatively influenced by the presence of the dextran backbone. Concerning the phase-solubility studies, solubilization efficiencies of all bCD-dextrans tested were much higher compared to those of bCD and commercially available derivatives (Fig. 3). The reason for this phenomenon is probably due to the fact that bCD/HC complexes form aggregates that are poorly soluble when standard bCDs are used, and the apparent solubility of the drug is therefore not improved (Loftsson and Table 3 Apparent solubility, concentration of molecularly dissolved drug, flux and apparent permeability coefficient (Papp) of HC in the presence of different type of bCDdextrans. Results are reported as the mean standard deviation (n = 3). Apparent solubility (mM)
Fraction molecularly Flux dissolved (mM) (105 mmol/ s cm2)
HC powder Not applicable 0.90 0.03 D10GPbCD 4.00 0.12 0.89 0.04 D25GPbCD 4.53 0.04 0.71 0.07
3.31 0.07 2.40 0.23 2.55 0.15
Papp calculated from the fraction of molecularly dissolved HC.
Papp (105 cm/s)
Fig. 4. Permeability profiles of HC through dialysis membranes of bCD-dextran formulations and HC crystalline suspension. Results are reported as mean standard deviation (n = 3).
Brewster, 2012). When standard bCDs are employed, solubility diagrams show a small increase in apparent solubility of HC, followed by a “plateau” and subsequent reduction in drug concentration (BS/BL phase-solubility diagram (Al-Sou’od, 2008; Preiss et al., 1994) due to precipitation of the complex. On the contrary, the presence of the dextran backbone allows the bCD/HC complex to be solubilized to a much higher extent. The polymers can be ranked according to their equilibrium constants for the complexation with HC (calculated from the slopes of phase solubility studies using Eq. (1)) as D20GPbCD (8.1 kM1) > D25GPbCD (3.9 kM1) > D10GPbCD (2.4 kM1) > bCDs (0.3 kM1), indicating a higher stability of bCD/HC complex in all cases where the dextran backbone is present. K eq: ¼
slope S0 ð1 slopeÞ
(1)
It should also be noted that the phase-solubility profile of D25GPbCD differs from the other dextran derivatives of lower molecular weighs (more related to “plateau” AN type than to a fully linear AL type), inducing the thought that a longer dextran backbone chain might reduce solubilizing properties of these polymers as earlier observed (Fülöp et al., 2013). Supersaturation studies (Table 3) clearly indicate that these polymers do not increase the concentration of molecularly dissolved HC and the observed solubilization is just related to complexation. Drug release results (Fig. 4 and Table 3) evidence that when HC is associated with the bCD-dextrans, the flux through the dialysis membrane of the drug is significantly (P 0.05) lower in comparison to that of the drug suspension. Apparent permeability coefficients (Papp) of HC were calculated taking only the fraction of molecularly dissolved HC into account in the donor solution as suggested earlier (Di Cagno and Luppi, 2013) (Table 3). The calculated Papp values are very similar for the powder suspension and the bCD-dextrans, suggesting that the flux of HC through the membranes is mainly driven by the concentration of molecularly dissolved HC, and that the high amount of HC loaded on the bCD-dextrans acts as a drug reservoir. 5. Conclusion
3.67 0.07 2.60 0.26 3.73 0.22
The potential pharmaceutical usefulness of a series of newly synthetized bCD-dextran polymers has been tested and was compared to native bCDs available on the market. All investigated
M. di Cagno et al. / International Journal of Pharmaceutics 468 (2014) 258–263
polymers were of appropriate sizes to allow parenteral administration, and no significant aggregation was macroscopically observed over time. Thermodynamic results evidence that the presence of the dextran backbone did not affect the stability of the bCD/drug complex independently of its molecular weight. All the investigated bCD polymers showed superior solubilizing properties for the poorly soluble drug HC in comparison to conventional bCD derivatives. This higher solubilization efficiency is solely due to complexation within bCD units and not due to supersaturation. Moreover, all bCD-dextrans provided a reservoir of poorly soluble HC in the aqueous solutions (complexed drug) altering also the diffusion of HC through the dialysis membranes. In conclusion, the described bCD-dextran polymers provide interesting options for both local and systemic drug delivery. Acknowledgments Our thanks go to the bachelor students Suad M. Awad, Philipp A. Elvang, Mette M. Tang, and De Xing Zheng for their contributions to the experimental work and to Roquette Freres for the donation of the native bCDs. References Al-Sou’od, K.A., 2008. Investigation of the hydrocortisone-b-cyclodextrin complex by phase solubility methods: some theoretical and practical consideration. Journal of Solution Chemistry 37, 119–133. Amidon, G.L., Lennernas, H., Shah, V.P., Crison, J.R., 1995. A theoretical basis for a biopharmaceutic drug classification: the correlation of in vitro drug product dissolution and in vivo bioavailability. Pharmaceutical Research 12, 413– 420. Brewster, M.E., Vandecruys, R., Peeters, J., Neeskens, P., Verreck, G., Loftsson, T., 2008. Comparative interaction of 2-hydroxypropyl-b-cyclodextrin and sulfobutylether-b-cyclodextrin with itraconazole: phase solubility behavior and solubilization of supersaturated drug solution. European Journal of Pharmaceutical Sciences 3, 103–494. Connors, A.K., 1997. The stability of cyclodextrin complexes in solution. Chemical Reviews 97, 1257–1325. Di Cagno, M., Luppi, B., 2013. Drug “supersaturation’’ states induced by polymeric micelles and liposomes: a mechanistic investigation into permeability enhancements. European Journal of Pharmaceutical Sciences 48, 775–780.
263
Di Cagno, M., Stein, P.C., Skalko-Basnet, N., Brandl, M., Bauer-Brandl, A., 2011. Solubilization of ibuprofen with b-cyclodextrin derivatives: energetic and structural studies. Journal of Pharmaceutical and Biomedical Analysis 55, 446–451. Fahr, A., Liu, X., 2007. Drug delivery strategies for poorly water-soluble drugs. Expert Opinion on Drug Delivery 4, 403–416. Fülöp, Z., Nielsen, T.T., Larsen, K.L., Loftsson, T., 2013. Dextran-based cyclodextrin polymers: their solubilizing effect and self-association. Carbohydrate Polymers 97, 635–642. Kurkov, S.V., Madden, D.E., Carr, D., Loftsson, T., 2012. The effect of parenterally administered cyclodextrins on the pharmacokinetics of coadministered drugs. Journal of Pharmaceutical Sciences 101, 4402–4408. Loftsson, T., Brewster, M.E., 2010. Pharmaceutical applications of cyclodextrins: basic science and product development. Journal of Pharmacy and Pharmacology 62, 1607–1621. Loftsson, T., Brewster, M.E., 2012. Cyclodextrins as functional excipients: methods to enhance complexation efficiency. Journal of Pharmaceutical Sciences 101, 3019– 3032. Loftsson, T., Brewster, M.E., Másson, M., 2004. Role of cyclodextrins in improving oral drug delivery. American Journal of Drug Delivery 2, 1–15. Loftsson, T., Frioriksdottir, H., Sigurdardottir, A.M., Ueda, H., 1994. The effect of water-soluble polymers on drug-cyclodextrin complexation. International Journal of Pharmaceutics 110, 169–177. Loftsson, T., Magnúsdóttir, A., Másson, A., Sigurjónsdóttir, J.F., 2002. Self-association and cyclodextrin solubilization of drugs. Journal of Pharmaceutical Sciences 91, 2307–2316. Nielsen, T.T., Wintgens, V., Larsen, K.L., Amiel, C., 2009. Synthesis and characterization of poly (ethylene glycol) based b-cyclodextrin polymers. Journal of Inclusion Phenomena and Macrocyclic Chemistry 65, 341–348. Nielsen, T.T., Wintgens, V., Amiel, C., Wimmer, R., Larsen, K.L., 2010. Facile synthesis of b-cyclodextrin-dextran polymers by “click” chemistry. Biomacromolecules 11, 1710–1715. Preiss, A., Mehnert, W., Frömming, K., 1994. Complexation of hydrocortisone with b-cyclodextrin and hydroxypropyl-b-cyclodextrin. Archiv der Pharmazie 327, 729–734. Roberts, J.R., Bratton, S.L., 1998. Colloid volume expanders. Problems, pitfalls and possibilities. Drugs 55, 621–630. Williams, H.D., Trevaskis, N.L., Charman, S.A., Shanker, S.M., Charman, W.N., Pouton, C.W., Porter, C.J.H., 2013. Strategies to address low drug solubility in discovery and development. Pharmacological Reviews 65, 315–499. Wong, J., Brugger, A., Khare, A., Chaubal, M., Papadopoulos, P., Rabinow, B., Kipp, J., Ning, J., 2008. Suspensions for intravenous(IV) injection: a review of development, preclinical and clinical aspects. Advanced Drug Delivery Reviews 60, 939–954. Xing, S., Zhang, Q., Zhang, C., Zhao, Q., Ai, H., Sun, D., 2009. Isothermal titration calorimetry and theoretical studies on host–guest interaction of ibuprofen with a-, b- and g-cyclodextrin. Journal of Solution Chemistry 38, 531–543. Zhang, L., Gu, F.X., Chan, J.M., Wang, A.Z., Langer, R.S., Farokhzad, O.C., 2008. Nanoparticles in medicine: therapeutic applications and developments. Clinical Pharmacology and Therapeutics 83, 761–769.
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
b-Cyclodextrin-dextran polymers for the solubilization of poorly soluble drugs Massimiliano di Cagno a, *, Thorbjørn Terndrup Nielsen b , Kim Lambertsen Larsen b , Judith Kuntsche a , Annette Bauer-Brandl a a b
Department of Physics, Chemistry and Pharmacy, University of Southern Denmark, Campusvej 55, Odense M 5230, Denmark Department of Biotechnology, Chemistry and Environmental Engineering, Aalborg University, Sohngaardsholmsvej 57, Aalborg 9000, Denmark
A R T I C L E I N F O
A B S T R A C T
Article history: Received 12 March 2014 Received in revised form 9 April 2014 Accepted 12 April 2014 Available online 16 April 2014
The aim of this study was to assess the potential of novel b-cyclodextrin (bCD)-dextran polymers for drug delivery. The size distribution of bCD-dextrans (for eventual parenteral administration), the influence of the dextran backbones on the stability of the bCD/drug complex, the solubilization efficiency of poorly soluble drugs and drug release properties were investigated. Size analysis of different bCDdextrans was measured by size exclusion chromatography (SEC) and asymmetrical flow field-flow fractionation (AF4). Stability of drug/bCD-dextrans was assessed by isothermal titration calorimetry (ITC) and molar enthalpies of complexation and equilibrium constants compared to some commercially available bCD derivatives. For evaluation of the solubilization efficiency, phase-solubility diagrams were made employing hydrocortisone (HC) as a model of poorly soluble drugs, whereas reverse dialysis was used to detect potential drug supersaturation (increased molecularly dissolved drug concentration) as well as controlled release effects. Results indicate that all investigated bCD-polymers are of appropriate sizes for parenteral administration. Thermodynamic results demonstrate that the presence of the dextran backbone structure does not affect the stability of the bCD/drug complex, compared to native bCD and commercially available derivatives. Solubility studies evidence higher solubilizing abilities of these new polymers in comparison to commercially available bCDs, but no supersaturation states were induced. Moreover, drug release studies evidenced that diffusion of HC was influenced by the solubilization induced by the bCD-derivatives. ã 2014 Elsevier B.V. All rights reserved.
Keywords: Cyclodextrins Calorimetry (ITC) bCD-dextran polymers Solubilization Poorly soluble drugs
1. Introduction Poor solubility of active pharmaceutical ingredients (API) is one of the biggest challenges in drug development: a large number of drugs on the market (approximately 40%), and even more those under development (approx. 70%), belong to class II, or respectively class IV of the Biopharmaceutics Classification System (BCS) and are therefore associated with solubility issues and as a consequence, poor oral bioavailability (Williams et al., 2013; Fahr and Liu, 2007; Amidon et al., 1995). In the last few years, a number of techniques have been suggested to increase bioavailability of such APIs (Williams et al., 2013; Fahr and Liu, 2007; Zhang et al., 2008). Amongst these, the use of cyclodextrins (CDs) has emerged as an
* Corresponding author. Tel.: +45 65502504; fax: +45 66158780. E-mail addresses: [email protected] (M. di Cagno), [email protected] (T. Terndrup Nielsen), [email protected] (K. Lambertsen Larsen), [email protected] (J. Kuntsche), [email protected] (A. Bauer-Brandl). http://dx.doi.org/10.1016/j.ijpharm.2014.04.029 0378-5173/ ã 2014 Elsevier B.V. All rights reserved.
efficient approach for solubilization. Cyclodextrins are cyclic oligosaccharides consisting of (a-1,4)-linked a-D-glucopyranose units. CD molecules are shaped like truncated cones, with a hydrophobic cavity inside and a hydrophilic surface outside. For many years CDs have been intensively investigated for pharmaceutical use due to their ability to spontaneously complex poorly soluble drug molecules, inducing solubilization (enhancement in apparent solubility) (Loftsson et al., 2002, 2004; Connors, 1997). For pharmaceutical purposes, b-cyclodextrins (bCDs, 7 glucose units in the ring) are mostly studied. Although bCDs are hydrophilic (as well as their complexes), their aqueous solubility is comparably low. The reason for this fact is associated with the high crystal lattice energy (Loftsson and Brewster, 2010) as well as intramolecular hydrogen bond formation that compromise the interaction with water molecules (solvation). To overcome this limitation, derivatives of the bCDs with higher aqueous solubility have been developed by introducing substituents at specific positions of the glucose rings, breaking the intramolecular net of hydrogen bonds and thereby enhancing aqueous solubility. The
M. di Cagno et al. / International Journal of Pharmaceutics 468 (2014) 258–263
most commonly used derivatives are methyl-b-cyclodextrin (MbCD), hydroxypropyl-b-cyclodextrin (HPbCD), and sulfobutylether-b-cyclodextrins (SBEbCD) (Brewster et al., 2008). Some of these bCD derivatives are of high solubility (e.g., methylderivatives), but were found to have strong systemic toxicity. Among the different derivatives, hydroxypropyl and sulfobutylether bCDs were found to have the lowest hemolytic effect and are therefore regarded suitable for parenteral administration (as well as some a- and g-cyclodextrin derivatives with 6 and 8 glucose units, respectively) (Loftsson and Brewster, 2010). Moreover, it has been shown that pharmacokinetic parameters of a number of drugs are not significantly modified after parenteral co-administration with these specific types of bCD derivatives (Kurkov et al., 2012). The general mechanism of the complexing reaction comprises exothermic interaction between the drug molecules and the CD cavity (negative enthalpy value), and the replacement of water from the cavity by the drug molecule (negative entropy). Furthermore, the presence of small quantities of hydrophilic polymers (such as PVP) or specific salts (e.g., sodium acetate or sodium salicylate) together with bCDs has previously been shown to stabilize drug/CD complexes through increased drug/CD interaction (enthalpy) or by formation of water-soluble drug/CD aggregates (Loftsson et al., 1994; Loftsson and Brewster, 2012). It has also been described that hydrophilic polymers can induce supersaturation states (increased molecularly dissolved drug concentration) (Di Cagno and Luppi, 2013). In order to overcome low aqueous solubility of native bCD and to take advantage of the stabilizing effect that hydrophilic polymers may have on the complex drug/CD, a new series of polymers based on hydrophilic backbones (i.e., dextrans) with conjugated bCD units have been synthetized and studied (Nielsen et al., 2010, 2009). The advantages of these CD-modified polymers are their much higher aqueous solubility compared to native bCD and the absence of substituents that may lead to steric hindrance at the CD opening. It should also be mentioned that FDA acknowledges dextran as GRAS when in use as food additives. Dextran is in pharmaceutical use as a plasma expander (Roberts and Bratton, 1998) and it is biodegraded by enzymatic activity. Considering all these facts, bCD-dextran polymers appear very promising as solubilizing agents for poorly
259
soluble drugs and drug delivery vehicles. In the present work, we studied the suitability of some new bCD-dextran polymers for drug delivery purposes. We investigated size distribution of different bCD-dextrans to evaluate their potential for parenteral administration, the influence of the dextran backbones on the stability of the bCD/drug complexes, solubilization efficiencies of a selected poorly soluble drug and drug release properties. 2. Materials and methods 2.1. Materials Three bCD modified dextrans (Fig. 1) were synthetized as described earlier (Nielsen et al., 2010). bCD, HPbCD with a number of substituent per glucose unit (molar degree of substitution (MS)) of 0.87 and MbCD (MS of 0.57) were a generous gift from Roquette Freres (Lestrem, France). Dextrans of different molecular weights (10, 20 and 25 kDa) were purchased from Pharmacosmos A/S (Holbaek, Denmark). Sodium di-hydrogen phosphate (NaH2PO4), di-sodium hydrogen phosphate (Na2HPO4) and sodium chloride (NaCl) (Sigma–Aldrich Chemie GmbH, Steinheim, Germany) were used for the preparation of the isotonic (275–285 mOsm, SemiMicro Osmometer K-7400, Herbert Knauer GmbH, Berlin, Germany) and isohydric (pH 7.4, pH meter Lab 827, Metrohm AG, Herisau, Switzerland) buffer saline (73 mM). Ibuprofen (IBP, Caesar & Lorenz GmbH, Hilden, Germany) and hydrocortisone (HC, Sigma–Aldrich) were employed as model drugs (Fig. 1). 2.2. Molecular mass determination 2.2.1. Size exclusion chromatography (SEC) Size exclusion chromatography was performed employing a TSK-gel column type SW 4000-3000, and polymer sizes were detected by a Wyatt miniDAWN 3 angle light scattering detector (MALS) (Wyatt Technology Inc., Santa Barbara, USA) and RIdetector (Optilab rEX, Wyatt) using a refractive index increment (dn/dc) of 0.15 mL/g. A 0.1 mM lithium nitrate (LiNO3) solution preserved with 0.05% sodium azide was used as eluent at a flow rate of 1 mL/min.
Fig. 1. Chemical structures of bCD-dextran derivatives (A), ibuprofen (B) and hydrocortisone (C).
260
M. di Cagno et al. / International Journal of Pharmaceutics 468 (2014) 258–263
2.2.2. Asymmetrical flow field-flow fractionation (AF4) Asymmetrical flow field-flow fractionation (Eclipse 3+ AF4, Wyatt Europe, Dernbach, Germany) was employed for comparison to determine the molecular masses for the different bCD-dextran polymers. The trapezoidal separation channel (length 265 mm, width 350 mm) equipped with a membrane of regenerated cellulose (molecular weight cut off 5 kDa, Wyatt Europe) was coupled with a UV–vis detector (Agilent Technologies GmbH, Boeblingen, Germany), a refractometry detector (Optilab rEX) and a multi-angle light scattering detector (DAWN HELEOS II) (both from Wyatt Technology Inc., Santa Barbara, USA). A 50 mM sodium chloride solution preserved with 0.02% sodium azide and filtered through a 0.01 mm membrane was used as an eluent. The samples were injected over 2 min (focus flow 2 mL/min) and eluted either applying constant cross flow (3 mL/min over 20 min initially followed by a cross flow decreasing from 3 to 0 mL/min within 5 min) or by only applying a cross flow gradient (cross flow decreasing from 3 to 0.1 mL/min over 40 min). In both cases, an elution step without cross flow finally was applied to ensure complete elution. Molar masses were calculated with Astra software (version 6.0.1, Wyatt) based on the light scattering and RI detector signals (baseline RI signal were subtracted) and using refractive index increment (dn/dc) of 0.14 mL/g. The average molecular weights are given together with the polydispersity values (Mw/Mn), whereas a polydispersity value of one corresponds to a complete homogeneous (monodisperse) sample. 2.3. Degree of substitution determination (DS) Determination of the degree of substitutions of the bCDdextran polymers was made by NMR spectroscopy using a Bruker DRX400 spectrometer (Bruker BioSpin GmbH, Rheinstetten, Germany). Samples were measured at 20 mg/mL in D2O at 42 C and the DS was calculated by comparison of the integral of the characteristic triazol proton and the anomeric protons of bCD and dextran as previously described (Nielsen et al., 2009). 2.4. Evaluation of complexation efficiency Calorimetric analyses were performed at 25 C using a 2277 Thermal Activity Monitor, TAM (Thermometric, Järfälla, Sweden) following a method previously described (Di Cagno et al., 2011). Small volumes (2.5 mL) of titrand solution (0.6 mM IBP isotonic and isohydric solution in phosphate buffer saline) were placed into 5 mL steel vessels (sample and reference). A 250 mL syringe (Hamilton, Bonaduz, Switzerland) filled with the titration solution containing the bCD-dextrans (Table 1) was inserted in the 612 Lund Syringe Pump (Thermometric). Fifteen drops of 12 mL volume each were titrated at 45 min intervals to complete the complexation reaction of the IBP solution. Each titration experiment was performed in triplicate. The values of standard enthalpy (DH ), mass equilibrium constants (K) and the amount of bCD units expressed as a concentration were calculated using the 2277-131 Digitam software by fitting the raw experimental data with titration curves, assuming a 1:1 stoichiometry of complexation between bCD units and IBP (as suggested earlier for polyethylene glycol bCD derivatives (Nielsen et al., 2009)). Table 1 Characteristics of the different bCD-dextrans titrant solutions used in the isothermal titration calorimetry experiments.
bCD-dextran (code)
pH
D10GPbCD D20GPbCD D25GPbCD
7.37 282 7.38 283 7.36 277
Osmolality (mOsm/ kg)
Polymer concentration (mg/ mL) 17.5 14.0 22.5
2.5. Phase-solubility studies HC was added in excess to aqueous solutions of different concentrations (3–4 mM) of the respective bCD-dextran polymer and commercial bCD derivatives for comparison in isotonic PBS (pH 7.4). After shaking for 72 h in a thermostatic water bath at 25 C (Julabo SW 23, Julabo Labortechnik GmbH, Seelbach, Germany), the solutions were filtered through 0.45 mm pore size Minisart RC4 filters (Sartorious Stedim Biotech GmbH, Göttingen, Germany). Concentrations of HC were detected in the supernatant by UV–vis spectroscopy (maximum wavelength 248 nm) by a Genesis 10 UV/ VIS spectrophotometer (Thermo Electron Corporation, Cambridge, UK). 2.6. Dialysis studies The dialysis method previously described (Di Cagno and Luppi, 2013) was employed to characterize possible supersaturation of HC induced by the bCD-dextrans, as well as to evaluate drug release from these polymers. Frantz (permeability) cells (SES GmbHAnalysesysteme, Bechenheim, Germany) with surface areas of 1 cm2, small chamber volume of 2 mL and big chamber volume of 8 mL were employed. Reverse dialysis setup was chosen: the big chamber was used as a donor compartment and filled with drug suspension and drug/bCD-dextrans saturated solutions (prepared as described in Section 2.5, the concentration of HC is reported in Table 3). Pre-washed dialysis membranes (Sigma–Aldrich) with a molecular weight cut-off (MWCO) of 10 kDa were placed between the donor and acceptor chambers. The MWCO was chosen to prevent bCD-dextrans to permeate, whereas HC could cross the membrane without impediment. At time zero (start of the experiment) the acceptor compartment was filled with 1.8 mL of fresh buffer. For supersaturation studies (no in-sink conditions needed), sampling of acceptor solution (0.2 mL) was performed after 50 h (reaching equilibrium), whereas, for the drug release studies, samples of 0.5 mL at 30 min intervals were withdrawn over five hours (replacement of the withdrawn volume with an equal amount of fresh buffer). Drug concentration was detected by UV–vis spectroscopy and each experiment was carried out in three replicates. 2.7. Statistic analysis Student’s t-test was performed to evaluate the significance of the differences identified between the data set obtained for each bCDdextran studied. P 0.05 was considered significantly different. 3. Results 3.1. Molecular weight determination and degree of substitution The actual molecular weights of the polymers were determined by both size exclusion chromatography and AF4. For all bCDdextrans, molecular weights detected by AF4 were in reasonable agreement with the results obtained by the SEC technique (Table 2). Polydispersity values were in the range between 1.247 (D10GPbCD solutions) and 1.512 (for D25GPbCD solutions). The degrees of substitution of all bCD-dextran polymers were between 55% (D10GPbCD) and 63% (D25GPbCD) (Table 2). No significant aggregation was macroscopically observed over time. 3.2. Complexation efficiency studies The standard enthalpy and Gibbs free energy of complexation between bCD-dextran polymers and IBP are reported in Fig. 2 in comparison to results obtained previously for some bCD derivatives (Di Cagno et al., 2011). Standard enthalpies of complexation were not
M. di Cagno et al. / International Journal of Pharmaceutics 468 (2014) 258–263
261
Table 2 Molecular weights (Mw), polydispersity and degree of substitutions (DS) identified for the different bCD-dextran derivatives. Relative standard deviations (SDrel) are provided.
bCD-dextran
Dextran
Dextran
bCD-dextran
bCD-dextran
(Code)
Mw (kDa)
Polydispersity (Mw/Mn)
nominal from manufacturer
Mw Mw (kDa) (kDa) AF4 (n = 2–4, SDrel 4%)
Mw (kDa) SEC (n = 3, SDrel 5%)
DS (% w/w) NMR (SDrel 5%)
10 20 25
16 20 31
1.247 1.503 1.512
25 54 83
55 57 63
D10GPbCD D20GPbCD D25GPbCD
bCD-dextran
31 67 83
the same for all modified polymers. Whereas D10GPbCD, D20GPbCD expressed enthalpies of complexation (12.8 0.9 and 14.7 0.7, respectively) which were in the same range as that of native bCD (14.6 0.7 kJ/mol), D25GPbCD expressed higher (P < 0.05) energy of complexation (19.3 0.8 kJ/mol) in comparison to the other polymers and standard bCD derivatives.
amount of HC molecularly dissolved in the aqueous phase. It can thus be concluded that the observed increase in apparent solubility is exclusively due to complexation of HC with the bCD units bonded to the dextran backbones.
3.3. Solubility and true supersaturation studies
Permeability studies with dialysis membranes were performed in order to characterize if these polymers were affecting the diffusion of drug molecules in aqueous environment, in comparison to the drug in the crystalline form. In the case of crystalline drug suspension, the flux of HC was different in comparison to bCD-dextrans solutions (Fig. 4). Specifically, in the case of the crystalline drug suspension, the flux was significantly (P 0.05) higher in comparison with all bCD-dextran derivatives (Table 3). Apparent permeability coefficients (Papp) were also calculated using as initial donor concentration the fraction of molecularly dissolved drug and the results are reported in Table 3.
All solubility studies were performed in isotonic PBS pH 7.4 and phase-solubility diagrams for HC in the presence of bCD-dextrans and classical bCDs are reported in Fig. 3. For all bCD-dextrans, starting from the thermodynamic solubility of HC at 25 C (0.90 0.03 mM), the apparent solubility of HC increases linearly with increasing concentration of the polymers (reported as bCD unit concentration), while bCD (and analogues of higher solubility) were not capable of inducing solubilization for this drug. All bCDdextrans lead to a higher apparent solubility for HC. Specifically, the apparent solubilities in the case of D10GPbCD, D20GPbCD and D25GPbCD were 3.11 0.31 mM, 4.05 0.14 mM and 3.21 0.09 mM, respectively, whereas for native bCD, MbCD and HPbCD were much lower (1.38 0.31 mM, 1.54 0.05 mM and 1.49 0.15 mM, respectively). In contrast to what has been observed for other polymers (e.g., PVP (Loftsson et al., 1994)), no enhancement of apparent solubilities of the drug by any of the native unmodified dextrans has been observed. Characterization of the fraction of molecularly dissolved HC indicates no increase of molecularly dissolved drug induced by the bCD dextrans (no supersaturation observed, Table 3). On the contrary, it appears that the presence of bCD-dextrans results in a certain reduction of the
3.4. bCD-dextrans drug release study
4. Discussion The aim of this work was to explore the potential of bCDdextran polymers in drug delivery. All bCD-dextrans sizes were studied in aqueous environment to confirm homogeneity of solutions and to detect eventual aggregates that may hamper their use for aqueous parenteral formulations. AF4 is a very powerful and sensitive technique for analyzing molecular weights of polymers and would detect aggregation in solutions. All bCDdextrans studied were of appropriate sizes for potential parenteral administration (Wong et al., 2008) and the molecular weighs
Fig. 2. Thermodynamic parameters of complexation of the model drug ibuprofen with bCD-dextrans (D10GPbCD, D20GPbCD and D25GPbCD) and some bCDs (HPbCD, MbCD and native bCD). Results are reported as mean standard deviation (n = 3).
262
M. di Cagno et al. / International Journal of Pharmaceutics 468 (2014) 258–263
Fig. 3. Phase-solubility diagram of hydrocortisone (HC) in the presence of different types of bCDs and bCD-dextrans. Results are reported as mean standard deviation (n = 3).
detected via the AF4 technique were in good agreement with the results obtained by SEC analysis. The discrepancies between the results obtained with the two techniques are not larger than expected and are probably due to the different separation principles of the two techniques. Moreover, in the SEC experiments, a three angles light scattering detector was employed whereas in the AF4 analysis, an 18 angles light scattering detector was employed. The average polydispersity value was 1.43, indicating a good homogeneity of the samples. IBP was chosen as model drug for thermodynamic complexation studies for comparability with previous results (Di Cagno et al., 2011). All data have been analysed assuming a 1:1 stoichiometry of reaction between IBP molecules and the bCD units (Xing et al., 2009). Results indicate that enthalpies of complexation between bCD units and IBP are comparable with those measured earlier for classical bCD derivatives (hydroxypropyl- and methyl-moieties, Fig. 2). Just in the case of D25GPbCD the complexation reaction was slightly more exothermic. However, it should be noted that for all bCD-dextrans, Gibbs free energies were comparable (no significant differences, P 0.05) to native bCD and derivatives (HPbCD and MbCD), indicating classical entropy–enthalpy compensation. From these studies it can be concluded that stabilities of the bCD-drug complex is not negatively influenced by the presence of the dextran backbone. Concerning the phase-solubility studies, solubilization efficiencies of all bCD-dextrans tested were much higher compared to those of bCD and commercially available derivatives (Fig. 3). The reason for this phenomenon is probably due to the fact that bCD/HC complexes form aggregates that are poorly soluble when standard bCDs are used, and the apparent solubility of the drug is therefore not improved (Loftsson and Table 3 Apparent solubility, concentration of molecularly dissolved drug, flux and apparent permeability coefficient (Papp) of HC in the presence of different type of bCDdextrans. Results are reported as the mean standard deviation (n = 3). Apparent solubility (mM)
Fraction molecularly Flux dissolved (mM) (105 mmol/ s cm2)
HC powder Not applicable 0.90 0.03 D10GPbCD 4.00 0.12 0.89 0.04 D25GPbCD 4.53 0.04 0.71 0.07
3.31 0.07 2.40 0.23 2.55 0.15
Papp calculated from the fraction of molecularly dissolved HC.
Papp (105 cm/s)
Fig. 4. Permeability profiles of HC through dialysis membranes of bCD-dextran formulations and HC crystalline suspension. Results are reported as mean standard deviation (n = 3).
Brewster, 2012). When standard bCDs are employed, solubility diagrams show a small increase in apparent solubility of HC, followed by a “plateau” and subsequent reduction in drug concentration (BS/BL phase-solubility diagram (Al-Sou’od, 2008; Preiss et al., 1994) due to precipitation of the complex. On the contrary, the presence of the dextran backbone allows the bCD/HC complex to be solubilized to a much higher extent. The polymers can be ranked according to their equilibrium constants for the complexation with HC (calculated from the slopes of phase solubility studies using Eq. (1)) as D20GPbCD (8.1 kM1) > D25GPbCD (3.9 kM1) > D10GPbCD (2.4 kM1) > bCDs (0.3 kM1), indicating a higher stability of bCD/HC complex in all cases where the dextran backbone is present. K eq: ¼
slope S0 ð1 slopeÞ
(1)
It should also be noted that the phase-solubility profile of D25GPbCD differs from the other dextran derivatives of lower molecular weighs (more related to “plateau” AN type than to a fully linear AL type), inducing the thought that a longer dextran backbone chain might reduce solubilizing properties of these polymers as earlier observed (Fülöp et al., 2013). Supersaturation studies (Table 3) clearly indicate that these polymers do not increase the concentration of molecularly dissolved HC and the observed solubilization is just related to complexation. Drug release results (Fig. 4 and Table 3) evidence that when HC is associated with the bCD-dextrans, the flux through the dialysis membrane of the drug is significantly (P 0.05) lower in comparison to that of the drug suspension. Apparent permeability coefficients (Papp) of HC were calculated taking only the fraction of molecularly dissolved HC into account in the donor solution as suggested earlier (Di Cagno and Luppi, 2013) (Table 3). The calculated Papp values are very similar for the powder suspension and the bCD-dextrans, suggesting that the flux of HC through the membranes is mainly driven by the concentration of molecularly dissolved HC, and that the high amount of HC loaded on the bCD-dextrans acts as a drug reservoir. 5. Conclusion
3.67 0.07 2.60 0.26 3.73 0.22
The potential pharmaceutical usefulness of a series of newly synthetized bCD-dextran polymers has been tested and was compared to native bCDs available on the market. All investigated
M. di Cagno et al. / International Journal of Pharmaceutics 468 (2014) 258–263
polymers were of appropriate sizes to allow parenteral administration, and no significant aggregation was macroscopically observed over time. Thermodynamic results evidence that the presence of the dextran backbone did not affect the stability of the bCD/drug complex independently of its molecular weight. All the investigated bCD polymers showed superior solubilizing properties for the poorly soluble drug HC in comparison to conventional bCD derivatives. This higher solubilization efficiency is solely due to complexation within bCD units and not due to supersaturation. Moreover, all bCD-dextrans provided a reservoir of poorly soluble HC in the aqueous solutions (complexed drug) altering also the diffusion of HC through the dialysis membranes. In conclusion, the described bCD-dextran polymers provide interesting options for both local and systemic drug delivery. Acknowledgments Our thanks go to the bachelor students Suad M. Awad, Philipp A. Elvang, Mette M. Tang, and De Xing Zheng for their contributions to the experimental work and to Roquette Freres for the donation of the native bCDs. References Al-Sou’od, K.A., 2008. Investigation of the hydrocortisone-b-cyclodextrin complex by phase solubility methods: some theoretical and practical consideration. Journal of Solution Chemistry 37, 119–133. Amidon, G.L., Lennernas, H., Shah, V.P., Crison, J.R., 1995. A theoretical basis for a biopharmaceutic drug classification: the correlation of in vitro drug product dissolution and in vivo bioavailability. Pharmaceutical Research 12, 413– 420. Brewster, M.E., Vandecruys, R., Peeters, J., Neeskens, P., Verreck, G., Loftsson, T., 2008. Comparative interaction of 2-hydroxypropyl-b-cyclodextrin and sulfobutylether-b-cyclodextrin with itraconazole: phase solubility behavior and solubilization of supersaturated drug solution. European Journal of Pharmaceutical Sciences 3, 103–494. Connors, A.K., 1997. The stability of cyclodextrin complexes in solution. Chemical Reviews 97, 1257–1325. Di Cagno, M., Luppi, B., 2013. Drug “supersaturation’’ states induced by polymeric micelles and liposomes: a mechanistic investigation into permeability enhancements. European Journal of Pharmaceutical Sciences 48, 775–780.
263
Di Cagno, M., Stein, P.C., Skalko-Basnet, N., Brandl, M., Bauer-Brandl, A., 2011. Solubilization of ibuprofen with b-cyclodextrin derivatives: energetic and structural studies. Journal of Pharmaceutical and Biomedical Analysis 55, 446–451. Fahr, A., Liu, X., 2007. Drug delivery strategies for poorly water-soluble drugs. Expert Opinion on Drug Delivery 4, 403–416. Fülöp, Z., Nielsen, T.T., Larsen, K.L., Loftsson, T., 2013. Dextran-based cyclodextrin polymers: their solubilizing effect and self-association. Carbohydrate Polymers 97, 635–642. Kurkov, S.V., Madden, D.E., Carr, D., Loftsson, T., 2012. The effect of parenterally administered cyclodextrins on the pharmacokinetics of coadministered drugs. Journal of Pharmaceutical Sciences 101, 4402–4408. Loftsson, T., Brewster, M.E., 2010. Pharmaceutical applications of cyclodextrins: basic science and product development. Journal of Pharmacy and Pharmacology 62, 1607–1621. Loftsson, T., Brewster, M.E., 2012. Cyclodextrins as functional excipients: methods to enhance complexation efficiency. Journal of Pharmaceutical Sciences 101, 3019– 3032. Loftsson, T., Brewster, M.E., Másson, M., 2004. Role of cyclodextrins in improving oral drug delivery. American Journal of Drug Delivery 2, 1–15. Loftsson, T., Frioriksdottir, H., Sigurdardottir, A.M., Ueda, H., 1994. The effect of water-soluble polymers on drug-cyclodextrin complexation. International Journal of Pharmaceutics 110, 169–177. Loftsson, T., Magnúsdóttir, A., Másson, A., Sigurjónsdóttir, J.F., 2002. Self-association and cyclodextrin solubilization of drugs. Journal of Pharmaceutical Sciences 91, 2307–2316. Nielsen, T.T., Wintgens, V., Larsen, K.L., Amiel, C., 2009. Synthesis and characterization of poly (ethylene glycol) based b-cyclodextrin polymers. Journal of Inclusion Phenomena and Macrocyclic Chemistry 65, 341–348. Nielsen, T.T., Wintgens, V., Amiel, C., Wimmer, R., Larsen, K.L., 2010. Facile synthesis of b-cyclodextrin-dextran polymers by “click” chemistry. Biomacromolecules 11, 1710–1715. Preiss, A., Mehnert, W., Frömming, K., 1994. Complexation of hydrocortisone with b-cyclodextrin and hydroxypropyl-b-cyclodextrin. Archiv der Pharmazie 327, 729–734. Roberts, J.R., Bratton, S.L., 1998. Colloid volume expanders. Problems, pitfalls and possibilities. Drugs 55, 621–630. Williams, H.D., Trevaskis, N.L., Charman, S.A., Shanker, S.M., Charman, W.N., Pouton, C.W., Porter, C.J.H., 2013. Strategies to address low drug solubility in discovery and development. Pharmacological Reviews 65, 315–499. Wong, J., Brugger, A., Khare, A., Chaubal, M., Papadopoulos, P., Rabinow, B., Kipp, J., Ning, J., 2008. Suspensions for intravenous(IV) injection: a review of development, preclinical and clinical aspects. Advanced Drug Delivery Reviews 60, 939–954. Xing, S., Zhang, Q., Zhang, C., Zhao, Q., Ai, H., Sun, D., 2009. Isothermal titration calorimetry and theoretical studies on host–guest interaction of ibuprofen with a-, b- and g-cyclodextrin. Journal of Solution Chemistry 38, 531–543. Zhang, L., Gu, F.X., Chan, J.M., Wang, A.Z., Langer, R.S., Farokhzad, O.C., 2008. Nanoparticles in medicine: therapeutic applications and developments. Clinical Pharmacology and Therapeutics 83, 761–769.
