**1. Introduction**

The drug dasatinib (DAS), whose chemical name is N-(2-chloro-6-methylphenyl)-2-[[6-[4-(2- hydroxyethyl)piperazin-1-yl]-2-methylpyrimidin-4-yl]amino]-1,3-thiazole-5-carboxamide (IUPAC)

(Figure 1) is a double inhibitor of kinase proteins, including proto-oncogene tyrosine-protein Src (Src-TK) family kinases [1].

**Figure 1.** Chemical structure of dasatinib (DAS).

DAS is the first-choice oral drug in the treatment of chronic myeloid leukemia (CML) for those patients who are resistant or intolerant to imatinib. In fact, CML is a myeloprolifelative disorder that is caused by the BCR-ABL oncogene and DAS is a potent inhibitor of imatinib-resistant BCR-ABL mutants [2,3].

Until now, DAS was used exclusively for the treatment of adult patients, but new scientific findings have shown its potential in the treatment of CML in paediatric age, where its pharmacokinetic parameters, in particular, absorption and elimination time, were comparable with those in adult, with the same safety and efficacy profiles [4,5]. However, in these clinical trials, the drug was administered to children in the form of tablets or crushed tablets dispersed in fruit juice. In fact, DAS, formulated as monohydrate and marketed under the name of Sprycel®by Bristol Meyer Squibb, is presented in the form of coated tablets with a dosage ranging from 20 to 140 mg of the active ingredient. No liquid formulation is available on the market, and this may be a problem for paediatric patients who may not be able to swallow the tablets.

Moreover, recently, a study was published showing that DAS may be applied in the treatment of Duchenne muscular dystrophy (DMD), a genetic muscle-wasting disorder, whose symptoms occur around the age of four years in boys and ge<sup>t</sup> worse quickly. DMD is characterized by a progressive muscle degeneration and weakness and it is caused by the absence of the subsarcolemmal protein *dystrophin*. Dystrophin preserves sarcolemmal integrity by linking the cytoskeleton to the extracellular matrix via the interaction with the dystrophin-glycoprotein complex (DGC) and allowing for proper force transmission from contractile apparatus to extracellular matrix [6]. Thus, the primary structural defect causes an aberrant transmission of mechanical stimulus across the myofibers, leading to progressive muscle weakness and degeneration [7]. Similar defects occur in animals, such as the widely used C57Bl/10ScSn-Dmd*mdx*/J (*mdx*) mouse model [7,8]. Recent studies in *mdx* mouse model have highlighted that, in dystrophin deficient muscles, Src-TK is both overactivated and overexpressed, due to the excessive ROS production, and contribute also to NOX activation, in an auto-reinforcing loop [9], then playing a key role in DMD pathogenesis. In addition, Src-TK is involved in phosphorylation and degradation of β-dystroglycan (β-DG), a member of DGC, contributing to the loss of this complex in dystrophic myofibers. Thus, either the pharmacological inhibition of Src-TK seems a feasible strategy to ameliorate the pathology [10,11]. Src-TK inhibitors are already clinically available as antitumor drugs, and DAS belongs to this class of drugs.

Based on what has just been outlined, it is evident that it would be useful to develop a new formulation of DAS, which is different from the one currently in use, possibly liquid, so that it could be readily used in paediatric patients either by oral or parenteral route [12].

Therefore, the purpose of the following work was to prepare an aqueous formulation of this drug, evaluating the possibility of using an inclusion complex with cyclodextrins (CDs), as it is a molecule that is characterized by a low water solubility [13]. Cyclodextrins, cyclic oligosaccharides consisting of glucose units joined by α 1,4-glycosidic bond have been widely used to improve the solubility and stability in water of different molecules due to their ability to form host-guest inclusion complexes [14–20]. Thus, in this work, we present an inclusion complex of DAS with the hydroxy-β-cyclodextrin (HP-β-CD), a semisynthetic cyclodextrin that is approved by FDA also for the parenteral administration. The DAS/HP-β-CD inclusion complex was first studied in solution by building the phase solubility diagram according to Higuchi-Connors [21] and a two-dimensional-NMR (2D-NMR) Heteronuclear Multiple Bond Correlation HMBC evaluation was carried out in order to investigate the portion of the molecule actually contained in the HP-β-CD cavity. Subsequently, this complex was prepared in solid state by lyophilization and characterized by Fourier Transform Infrared (FT-IR), Differential Scanning Calorimetry (DSC), evaluation of the incorporation degree, and study of dissolution profiles at different pH values. Finally, in view of potential use of DAS for DMD, we first assessed its cytotoxic action on C2C12 cells, a muscle satellite cell line; secondly, we conducted an in vivo study in wild type C57Bl/6J (WT) mice by administering the inclusion complex in drinking water for one week to test both palatability and the exposure levels of the complex.

### **2. Results and Discussion**

### *2.1. Evaluation of the Inclusion Complex in Solution*

First of all, the solubility of DAS was determined at 25 ◦C both in ultra-pure water and in buffered aqueous solutions at pH 1.2 (HCl 0.05 M, for oral administration) and at pH 7.4 (phosphate buffer 0.05 M, for parenteral administration). The results are shown in Table 1. DAS is a strong base with a pKa value of 10.28 [11], so it is more soluble in acid environments where the protonation of the NH groups occurs.

The lowest solubility value was recorded in ultrapure water with a pH value of about 6.0, so the phase solubility diagram relating to the complexation of DAS with HP-β-CD has been studied at 25 ◦C in water. According to the Higuchi and Connors classification [20], it shows an AP-type profile, as reported in Figure 2a, and this result clearly show that DAS solubility in water is linearly influenced by the presence of HP-β-CD until a percentage of cyclodextrin equal to about 6%, with the formation of an inclusion complex with 1:1 host:guest stoichiometry, while in the presence of major cyclodextrin percentages the formation of complexes with different stoichiometry occurs. From the analysis of the first linear portion of the Higuchi-Connors diagram (Figure 2b), it is possible to obtain the complexation constant for the complex with 1: 1 host:guest stoichiometry and it was found to be equal to 922.13 <sup>M</sup>−1, with an increase of about 21 times of the DAS solubility in the presence of 6% *w*/*v* of cyclodextrin (0.014 mg/mL, 2.9 × 10−<sup>5</sup> M) as compared to the solubility value of the drug in the absence of the complexant, which results to be 6.49 × 10−<sup>4</sup> mg/mL (1.33 × 10−<sup>6</sup> M).


**Table 1.** DAS solubility at 25 ◦C in presence of different environments.

In order to determine the exact stoichiometric ratio between DAS and HP-β-CD in the formation of the inclusion complex, the Job's plot (Figure 2C) was constructed, as described in the experimental section. In detail, this study was conducted via 1H-NMR observing the variation of chemical shifts of methyl hydrogens (CH3) on the pyrimidine ring of DAS.

**Figure 2.** Phase solubility and Job's plot diagrams of DAS and hydroxypropyl-β-cyclodextrin (HP-β-CD) in water at 25 ◦C. (**a**) Phase solubility diagram in the HP-β-CD concentration range 0–10%; (**b**) Phase solubility diagram in the HP-β-CD concentration range 0–6%; (**c**) Job's plot diagram.

As shown in the graph, a highly symmetrical trend with a maximum value being recorded at *r* = 0.5 is observed, and this finding highlights the formation of a 1:1 inclusion complex. This result is quite in agreemen<sup>t</sup> with the phase solubility diagram because for the construction of Job's plot very low concentration of HP-β-CD are used and at low concentration of cyclodextrin the formation of an inclusion complex with a 1:1 host:guest stoichiometry occurs. This behavior has already been widely described in the literature [15,22]. It is in fact known that the balance of complexation between drug and cyclodextrin is strongly influenced by the concentrations in solution of the two components and that in the presence of high concentration of cyclodextrin different solubilization phenomena take place that modify the stoichiometry of the inclusion complex, leading to higher-order complexes.

Furthermore, the construction of the Job's diagram has been carried out on the basis of the displacement, in terms of chemical shift, of the methyl group protons on the DAS pyrimidine ring. It would therefore seem that this methyl group is directly involved in the formation of the inclusion complex with the cavity of the cyclodextrin and in order to obtain more information about the interactions of the drug with the cyclodextrin in solution, a 1H- and 2D-NMR (HMBC) study was conducted, keeping the DAS concentration constant and varying the molar ratio DAS: HP-β-CD.

Figure 3 shows the 2D 1H- 13C-NMR spectrum of DAS in DMSO-d6, which was used to make the correct assignment of DAS protons while in Figure 4a–c are reported the 1H-NMR spectra of methyl resonances of CH3 on pyrimidine and benzene rings at different DAS: HP-β-CD molar ratios.

**Figure 3.** 2D 1H- 13C-NMR spectrum of DAS in DMSO-d6. The cross-peaks displayed by HMBC were used to identify the structure of the drug, including the correlation of the δ of hydrogens and carbons separated from each other with two and three chemical bond.

**Figure 4.** 1H-NMR spectra of methyl resonances of CH3 on pyrimidine (δ ~2.5) and benzene (δ ~2.3) rings in the presence of cyclodextrin at different DAS:HP-β-CD molar ratios. (**a**) DAS:HP-β-CD molar ratios 1:0; (**b**) DAS:HP-β-CD molar ratios 1:1; and (**c**) DAS:HP-β-CD molar ratios 1:10.

The relative positions of the peaks were in agreemen<sup>t</sup> with the assignment. 1H- and 2D-NMR (HMBC) investigations confirmed the structure of the molecule and elucidated the interactions with the cyclodextrin in solution. Our results give a direct evidence of the formation of an inclusion complex between the drug and the cyclodextrin. In Table 2, we reported the variation of chemical shifts of methyl hydrogens (CH3) on the pyrimidine ring of DAS in the presence of different concentrations of cyclodextrin (i.e., molar ratio drug:cyclodextrin 1:1, 1:2, and 1:3). As one can see from Table 2, the chemical shifts of this CH3 are affected during complexation, showing changes in the ppm values. In particular, as shown in Figure 4a–c increasing the concentration of cyclodextrin in solution, we observed that the chemical shift of the CH3 hydrogens on the pyrimidine ring shifted downfield (higher ppm). These findings sugges<sup>t</sup> that the methyl hydrogens on the pyrimidine ring were directly involved in the complexation with cyclodextrin. In detail, the hydrogen nuclei of the drug included in the cyclodextrin cavity established hydrophobic interactions with cyclodextrin hydrogens, resulting in a their deshielding. No significant variation of the chemical shifts of the methyl hdrogens on the aromatic ring benzene was observed. This would sugges<sup>t</sup> that this portion of the molecule is not interested in the complex formation with cyclodextrin.

**Table 2.** Shifts of CH3 hydrogens in the presence of cyclodextrin at different DAS: HP-β-CD molar ratios.


### *2.2. Characterization of the Inclusion Complex in the Solid State*

In order to exploit the complexation with HP-β-CD in the preparation of a powder formulation of the drug that instantly dissolves when placed in water to be administered orally or parenterally, the solid-state complex was prepared by lyophilization. The freeze-dried complex was characterized by the assessment of the degree of incorporation, expressed as g of DAS per 100 g of product and it was found to be 4.23 ± 0.42 g of DAS per 100 g of lyophilized powder. This solid inclusion complex has been characterized by FT-IR, DSC, and dissolution profile. Figure 5 shows the IR spectra of DAS, HP-β-CD, and HP-β-CD solid inclusion complex.

The IR spectrum of DAS shows an absorption band at 1609 cm<sup>−</sup><sup>1</sup> due to stretching of the carbonyl group in the amidic bound, and two absorption bands at 2945 and 2930 cm<sup>−</sup><sup>1</sup> due to C-H stretching of methylenic and alchilic groups. In addition, the bands at 1583, 1498, and 1417 cm<sup>−</sup><sup>1</sup> corresponding to the C-C strain of the aromatic ring, and the bands at 3461 and 3225 cm<sup>−</sup>1, corresponding to the stretching of the N-H and O-H are highlighted, respectively.

The spectra of CDs inclusion complex appear to be very similar to those of cyclodextrin, since the cyclodextrins exhibit a high number of polar groups (OH, CO) that give rise to very broad absorption bands, which in some regions often overlap with those of DAS, also because a large excess of cyclodextrin is present in the complexes. This finding is confirmed by the DSC study reported in Figure 6.

**Figure 5.** Fourier Transform Infrared (FT-IR) spectra (**a**) DAS, (**b**) HP-β-CD, and (**c**) DAS/HP-β-CD complex.

**Figure 6.** DSC thermograms: (**a**) DAS, (**b**) HP-β-CD, and (**c**) DAS/HP-β-CD complex.

In the DAS thermogram (Figure 6a), it is evident the crystalline nature and the high degree of purity of this compound that shows an endothermic spike at 285 ◦C, according with data reported in literature [11].

The HP-β-CD thermogram (Figure 6b) highlights the amorphous nature of the same, which does not exhibit an endothermic melting peak but only a sloping peak between 80 and 100 ◦C due to the loss of the water present in the sample. The thermogram of the DAS/HP-β-CD complex (Figure 6c) has a trend that is comparable to that of cyclodextrin alone and this indicates the drug's inclusion within the cavity of the complexing agen<sup>t</sup> with its amorphization.

### *2.3. Dissolution Studies*

Moreover, dissolution studies have been performed at 37 ◦C in two different media: phosphate buffer 0.05M pH = 7.4 and HCl 0.05M pH = 1.2. Figure 7 shows obtained dissolution profiles.

**Figure 7.** Dissolution profiles at 37 ◦C: (**a**) pH 7.4 and (**b**) pH 1.2 of DAS alone (-) and DAS/HP-β-CD complex (). All values are mean ± SD, *n* = 3.

It is clear that the DAS/HP-β-CD lyophilized complex exhibits a better dissolution profile than the drug alone. In particular, this is especially evident at pH 7.4 where it was not possible to obtain the dissolution profile of DAS alone due to its very low solubility at this pH value, which prevents the quantitative determination of the drug via HPLC in the dissolution medium. The hydrophobic nature of the drug prevented its contact with the dissolution medium, causing it to float on the surface and hindering its dissolution. Instead, in the same dissolution medium the presence of HP-β-CD allows for the achievement of a quantity of dissolved drug equal to about 77% after 420 min. At pH 1.2, DAS appears to be more soluble, as demonstrated by the previously described solubility analysis. In this case, hence, it was possible to obtain the solubility profile of the drug alone at this pH value and it is evident that after approximately 420 min the quantity of drug dissolved is approximately 81%, as compared to the 100% that is reached from the complex with the HP-β-CD.

Therefore, the complexation with the selected cyclodextrin certainly represents an effective strategy for improving the solubility characteristics and the dissolution profile of DAS, also allowing an improvement of these characteristics with respect to those of monohydrate and polymorphic forms that are patented [23,24], and enabling it to be administered parenterally, in addition to oral administration, which is currently the only possible DAS route of administration.

### *2.4. Cytotoxicity Studies*

The cell viability study was performed to compare the cytotoxicity, and then the pharmacological activity, of the DAS/HP-β-CD inclusion complex with that of the free drug both solubilized in DMEM. In view of potential use for DMD, this effect has been assessed on C2C12 myoblasts. For DAS alone, due to its very low water solubility, a DMSO solution was prepared and this solution was subsequently diluted in DMEM so that the final DMSO concentration in each well was less than 0.15% in order to ensure cellular vitality. The two vehicles were also tested, i.e., HP-β-CD and DMSO, both diluted in DMEM and both at the highest concentration tested in the presence of DAS. The results that were obtained in terms of cellular viability are shown in Figure 8.

**Figure 8.** Effect of DAS on cell viability. The figure shows the cytotoxic effect on cell viability of increasing concentration of DAS (0.1–100 μM) alone or complexed with HP-β-CD. The results are expressed as the percentage of the control (ctrl) and presented as the mean ± S.E.M. Each data is from 24–48 replicates (wells) and 6–9 different culture dishes. The statistical significance between groups was evaluated by Student's *t*-test, as follows: significantly different with respect to \* the control value (0.001 < *p* < 0.05); ◦ DAS/HP-β-CD at the same concentration (0.001 < *p* < 0.05).

It is clear that all of the compounds tested show, as expected, a cytotoxicity that is concentration dependent. In particular, the DAS/HP-β-CD complex has a relatively higher effect on cell viability than free DAS, with significantly different statistical results (0.001 < *p* < 0.005 and 0.025 < *p* <0.001). Furthermore, since both vehicles are not cytotoxic, because they guarantee 100% cellular viability, the cytotoxicity recorded in the test is attributable exclusively to the effect of the drug. The obtained result suggests that DAS complexation with HP-β-CD increases the cytotoxicity of the drug, and this effect is probably a consequence of the increased solubility of DAS in water-like phase. Anyway, it is important to underline that the concentration at which DAS exerted cytotoxic actions on C2C12 cells, both free than complexed with HP-β-CD, is higher that the IC50 values known to inhibit cancer cell growth, which are in the nM range. Therefore, this in vitro experiment underlines that DAS is relatively safe on satellite muscle precursors being cytotoxic only at high concentrations. In fact, the concentrations that are used in the cell viability test are above the therapeutic plasma levels of DAS, which range in the low μM values.

### *2.5. Pharmacokinetic Results*

HPLC analyses were carried out to evaluate the DAS traceability in main target tissues (quadriceps and liver) of treated mice. Appreciable drugs' levels were found in quadriceps and livers of treated animals (Figure 9). These results are in line with the finding that DAS is rapidly distributes in tissues [25]. Also, the level reached in skeletal muscle allows for predicting sufficient exposure for the action of DAS to take place, considering that the inhibition of Src-TK occurs in the nanomolar range [26].

**Figure 9.** Pharmacokinetic analysis in quadriceps and livers of DAS/HP-β-CD inclusion complex administered at 15 mg/kg in drinking water for 1 week. All values are mean ± S.E.M. from 7–8 mice for each group. No significant difference was found by Student *t*-test analysis.

### **3. Materials and Methods**
