Colloids and Surfaces B: Biointerfaces 115 (2014) 22–28

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

A characterization study of resveratrol/sulfobutyl ether-␤-cyclodextrin inclusion complex and in vitro anticancer activity Valentina Venuti a , Carmela Cannavà b , Maria Chiara Cristiano c , Massimo Fresta c , Domenico Majolino a , Donatella Paolino c , Rosanna Stancanelli b , Silvana Tommasini b , Cinzia Anna Ventura b,∗ a

Dipartimento di Fisica e di Scienze della Terra, Università degli Studi di Messina, Viale Ferdinando Stagno D’Alcontres 31, I-98166 Messina, Italy Dipartimento di Scienze del Farmaco e Prodotti per la Salute, Università degli Studi di Messina, V.le Annunziata, I-98168 Messina, Italy c Dipartimento di Scienze della Salute, Università Magna Graecia di Catanzaro, Viale Europa, I-88100 Catanzaro, Italy b

a r t i c l e

i n f o

Article history: Received 29 July 2013 Received in revised form 18 October 2013 Accepted 13 November 2013 Available online 21 November 2013 Keywords: Resveratrol Sulfobutylether-␤-cyclodextrin FTIR-ATR Job’s plot Anticancer activity

a b s t r a c t A resveratrol/sulfobutylether-␤-cyclodextrin inclusion complex was prepared using the freeze-drying method and characterized in solution through UV–vis spectroscopy, solubility phase studies and Job’s plot methods. At the solid state it was characterized using the FTIR-ATR technique. Sulfobutylether-␤cyclodextrin has a high affinity for the drug, and forms an inclusion complex with a 1:1 molar ratio both in solution and as a solid sample. It also has a high stability constant (Kc , 10,114 M−1 ). Complexation strongly increases the water solubility of resveratrol (from 0.03 mg/ml to 1.1 mg/ml, at 25 ◦ C) and positively influences its in vitro anticancer activity which was observed on a human breast cancer cell line (MCF-7). In solid phase, FTIR-ATR revealed itself as being a useful technique in elucidating the complexation mechanism, which it did by emphasizing the functional groups involved in the activation of non-covalent “host–guest” interactions. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Resveratrol (trans-3,5,4 -trihydroxystilbene – RSV) is a triphenolicphytoalexin found in a variety of plant species [1]. The phenolic nature of RSV explains its antioxidant activity. It has been shown to provide health-promoting benefits, such as lowering the incidence of coronary heart disease, and it possesses qualities that prevent cancer. It also manifests estrogenic activity with varying degrees of estrogen receptor agonist due to its structure which is similar to the synthetic estrogen diethylstilbestrol [2]. In the area of chemoprevention, a great number of studies have been performed, demonstrating that RSV acts through numerous mechanisms, including the regulation of cell cycle progression [3] and apoptosis [4], the inhibition of tumor invasion and angiogenesis [5], the prevention of inflammation [6], the activation of adenosine monophosphate-activated protein kinase (AMPK) [7], the scavenging of reactive oxygen species [8] and the modulation of NFkB [9]. This activation of multiple anti-cancer pathways (pleiotropism) is an attractive feature of RSV, since it may help to overcome drug

∗ Corresponding author. Tel.: +39 090 6766508; fax: +39 090 6766402. E-mail address: [email protected] (C.A. Ventura). 0927-7765/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.colsurfb.2013.11.025

resistance [10]. It has been suggested that RSV is capable of mediating a great number of other biological responses relevant to human health such as protection against viral infections [11] and caloric restriction mimicry [12], as well as having beneficial effects on longevity [13]. RSV is a solid off-white powder soluble in ethanol and in dimethyl sulfoxide but it is practically insoluble in water (∼0.03 mg/ml at 25 ◦ C) according to the European Pharmacopeia definition, and its log P is 3.1 [14]. The drug exists as two structural isomers: cis-(Z) and trans-(E) (Fig. 1). The trans-isomer is biologically more active than the cis-isomer [15], probably due to its non-planar conformation. When protected from light, trans-RSV is stable for at least 42 h and for at least 28 days in pH 1 and 7 buffers, respectively; whereas the cis form is only stable at neutral pH when completely shielded from light [16]. The great hydrophobicity of RSV constitutes a serious problem for its oral bioavailability and for the realization of liquid formulations, it requires the use of solvents that might not be suitable for parenteral administration. So despite its promising beneficial effects, the clinical use of RSV is very limited. To date, it is present on the market only as a nutraceutical product. The realization of fast-dissolving-solid formulations or free solvent–liquid formulations of RSV could open new perspectives for

V. Venuti et al. / Colloids and Surfaces B: Biointerfaces 115 (2014) 22–28

23

(DMSO) were purchased from Sigma–Aldrich (Milan, Italy). MCF-7 cells were provided by IZS of Modena and Reggio-Emilia. 2.2. Preparation of the inclusion complex

Fig. 1. Chemical structures of cis- and trans-resveratrol.

the pharmaceutical use of this active compound, that could show in vivo anticancer activity comparable to other natural products. To this end, cyclodextrins (Cyds) can be used. Compounds that are poorly soluble in water and hydrophobic moieties of amphiphilic molecules can interact non-covalently with Cyd cavities to form the so-called inclusion complexes, which are generally highly water-soluble [17–21]. The solubility of these complexes depends principally on the type of Cyd used. Hydroxypropyl-␤-cyclodextrin (HP-␤-Cyd) and sulfobutylether-␤-cyclodextrin (SBE-␤-Cyd) are non-toxic and biocompatible Cyd derivatives which exhibit solubility and complexing abilities greater than those of the parent Cyd [22]. In particular, SBE-␤-Cyd interacts very well with neutral drugs to facilitate solubility and chemical stability, and because of its polyanionic nature, it interacts particularly well with cationic drugs. Moreover, its four-carbon butyl chain coupled with the repulsion of the end group’s negative charge allows for an extension of the SBE-␤-Cyd cavity, thus binding a drug more strongly, as compared to other modified Cyds [22]. Researchers have recently been studying the inclusion complex of RSV with natural or modified Cyds [23], and they report an increase of water solubility and of the antioxidant activity of the drug, particularly when it is complexed with HP-␤-Cyd [24,25]. Complexation of RSV with SBE-␤-Cyd has been less studied and to our knowledge no detailed characterization studies of this complex are present in literature. In this work we prepared the inclusion complex of RSV with SBE␤-Cyd and the influence of this on the water solubility of the drug was evaluated. The complex was prepared by the freeze-drying method and characterized in the solid state by Fourier transform infrared spectroscopy in attenuated total reflectance (FTIR-ATR) geometry. In solution, solubility phase studies and Job’s plot were performed in order to investigate the stoichiometry of the inclusion complex. In vitro biological assays on a human breast cancer cell line (MCF-7) were performed to evaluate the influence of SBE-␤-Cyd on the anticancer activity of RSV. 2. Materials and methods

SBE-␤-Cyd (200 mg) was solubilized in 10 ml of water at room temperature and added to an amount of RSV exceeding its intrinsic solubility. The flask was repaired from the light in order to prevent the conversion of the RSV from the trans- to the cis-position, sonicated in a Bandelin RK 514 water bath (Berlin, Germany) for 30 min, then stirred for 3 days. Then the suspension was filtered through Sartorius Minisart® -SRP 15 PTFE 0.22 ␮m filters (Germany) and freeze-dried (VirTis Gardiner, USA BenchTop K Series Freeze Dryers). 2.3. Phase-solubility analysis Phase-solubility studies were performed using the method described by Higuchi and Connors [26]. An amount of RSV exceeding its solubility was added to unbuffered aqueous solutions of SBE-␤-Cyd (0.0–8.0 × 10−3 M) in 5 ml capped tubes, then sonicated in a Bandelin RK 514 water bath (Berlin, Germany) for 15 min. The flasks were sealed to avoid changes due to evaporation and magnetically stirred for 3 days in a Telesystem stirring bath thermostat 15.40 with a Telemodul 40 C control unit at 25 ± 0.1 ◦ C. After equilibrium was reached, the suspensions were filtered through Sartorius Minisart® -SRP 15 PTFE 0.22 ␮m filters (Germany) and analyzed through UV–vis spectroscopy, in the 200–400 nm spectral range (FullTech Instruments double beam spectrophotometer, mod. PG T80, Italy). This was done to evaluate the amount of RSV dissolved. All measurements were repeated at least three times. The data obtained were used to determine the binding constant of the RSV–SBE-␤-Cyd inclusion complex, according to Higuchi and Connors equations [26]. No degradation of RSV was observed under experimental conditions. 2.4. Job’s plot method Equimolar (1 × 10−5 M) methanol/water solutions (55/45, v/v) of RSV and SBE-␤-Cyd were mixed to a fixed volume by varying the molar ratio from 0.1 to 0.9, keeping the molar concentration of the species constant. After stirring for 1 h, the absorbance of each solution was measured by UV–vis spectroscopy at 305 nm and abs was determined as the difference between abs without and with Cyd. Then, abs × [RSV] was plotted vs. R (R = [RSV]/[RSV] + [SBE␤-Cyd]) [27].

2.1. Materials

2.5. Water solubility and dissolution rate determination

Resveratrol (3,5,4 -trihydroxystilbene, C14 H12 O3 , MW 228.24) (RSV) is a product of Sigma–Aldrich Chemie® (Italy); sulfobutylether-␤-cyclodextrin (CAPTISOL® , average degree of sulfobutyl substitution: seven; average MW 2162) (SBE-␤-Cyd) was kindly supplied by CyDex Pharmaceutical (Lenexa, Kansas City, USA). They were all employed without any further purification. The water used throughout the study was double-distilled and de-ionised, then filtered through 0.22 ␮m Millipore® GSWP filters (Bedford, USA). Dulbecco’s modified eagle’s medium (DMEM) is a Life Technologies product (Milan, Italy). Fetal bovine serum (FBS), glutamine, penicillin, streptomycin and trypsin/EDTA (1×) solution were purchased from Invitrogen Corporation (Giuliano Milanese, Milan, Italy). 3-[4,5-Dimethylthiazol-2-yl]3,5-diphenyltetrazolium bromide (MTT) dye test (TLC purity 97.5%), phosphate buffer (PBS) solution and dimethyl sulfoxide

The degree of water solubility of the free RSV and the RSV/SBE␤-Cyd inclusion complex was determined by suspending excess amounts of each sample in 3 ml of water and stirring at 25 ± 0.1 ◦ C for 2 days. The suspensions were then filtered through Sartorius Minisart® -SRP 15 PTFE 0.22 ␮m filters (Germany) and analyzed by UV–vis spectroscopy at 305 nm. The determination of the dissolution rates of the same samples was carried out according to the USP 32nd paddle method. Three hundred and thirty mg of free RSV or a corresponding amount in complex were suspended in 900 ml of water and stirred at 100 rpm at 37 ± 0.5 ◦ C. At fixed intervals the concentration of RSV in solution was assayed by UV–vis spectroscopy at 305 nm. The medium was reconstituted with fresh water and the data were corrected for the operated dilution. The experiments were carried out in triplicate and data were presented as mean ± standard deviation.

24

V. Venuti et al. / Colloids and Surfaces B: Biointerfaces 115 (2014) 22–28

2.6. FTIR-ATR spectroscopy FTIR-ATR studies were carried out, at room temperature, on a DA8 Fourier transform infrared (FTIR) spectrometer from BOMEM, using a thermoelectrically cooled deuterated triglycene sulphate (DTGS) detector, in combination with a KBr beamsplitter and a Globar source. The powders were contained in a Golden Gate diamond ATR system, based on the attenuated total reflectance (ATR) technique [28]. An ATR setup exhibits various advantages with respect to an ordinary absorption setup. It is nondestructive, it requires only micrograms of sample, and it is at the origin of spectra displaying a very good signal-to-noise ratio, easily avoiding saturation of bands. In addition, a chemical analysis can be performed directly on ATR spectra, avoiding the necessity of using elaborated calculations of optical constants [29]. The spectra were recorded in the 3800–600 cm−1 spectral range. Each spectrum was collected in a dry atmosphere, in order to avoid dirty contributions, with a resolution of 4 cm−1 , and is an average of 100 repetitive scans, hence guaranteeing a good signal-to-noise ratio and high reproducibility. All the IR spectra were normalized for taking into account the effective number of absorbers. No mathematical correction (e.g., smoothing) was done, and spectroscopic manipulations such as baseline adjustment and normalization were performed using the Spectracalc software package GRAMS (Galactic Industries, Salem, NH, USA). For the O H stretching region, second derivative computations were used for evaluating the wavenumbers of the maxima of the different sub-bands. Multiple curve fitting into Voigt profiles was then applied to the experimental profiles based on these wavenumber values, using the routine provided in the PeakFit 4.0 software package. The statistical parameters defined in the software manual were used as a guide to obtain the bestfit and allowed to vary upon iteration until converging solution was reached. The bestfit is characterized by r2 ≈ 0.9999 for all the investigated systems. 2.7. Cell cultures MCF-7 cells were seeded into plastic culture dishes (100 mm × 20 mm) using D-MEM cell culture medium integrated with glutamate, penicillin (100 IU/ml), streptomycin (100 ␮g/ml) and FBS (10%, v/v) and incubated at 37 ◦ C, 5% CO2 (Guaire® TS AutoflowCodue Water-Jacketed-Incubator). At 80% of confluence, MCF-7 cells were harvested using trypsin (1 ml), washed with 2 ml of PBS buffer, collected into centrifuge plastic tubes and further diluted with cell culture medium to obtain a final volume of 8 ml. The samples were further centrifuged at 1200 rpm for 5 min at room temperature using a Megafuge 1.0 (HeraeusSepatech, Osterode/Harz, Germany). The supernatant was withdrawn, and the pellet was re-suspended in medium to obtain a final concentration of approximately 7.6 × 103 cells/cm2 . 2.8. Evaluation of cytotoxic activity MTT assay (3-(4,5-dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide, a yellow tetrazole) was performed to evaluate the cytotoxic activity of RSV, SBE-␤-Cyd and RSV–SBE␤-Cyd. MCF-7 cells were seeded into 96-well cell culture plates (5 × 103 /0.66 cm2 ) and incubated at 37 ◦ C. After 24 h of incubation, the cell culture medium was poured off and replaced with 0.1 ml/well (0.15 mM) of different samples; untreated cells were used as control during experiments. The cytotoxic activity of compounds was evaluated at different incubation times (24, 48, and 72 h). MTT solution (5 mg/ml in PBF buffer) was added to each well and further incubated for 3 h. The cell culture medium was then removed and the obtained formazan crystals, precipitated on the bottom of well, were dissolved using 100 ␮l of a DMSO/ethanol

(1:1, v/v) solution. The plates were further shaken for 20 min at 230 rpm (IKA® KS 130 Control, IKA® WERKE GMBH & Co., Staufen, Germany). The dissolved formazan crystals were quantified using a microplate spectrophotometer (BIORAD, MarkTM Microplate Spectrophotomer) at a wavelength of 570 nm by subtracting background absorbance measured at 690 nm. The percentage of cell viability was calculated according to the following equation: cell viability (%) =

AbsT × 100 AbsC

where AbsT is the absorbance of treated cells and AbsC is the absorbance of control (untreated) cells. The concentration of the formazan crystals is directly correlated to cell viability. Data are the average of three different experiments (6 replicates for each point) ± standard deviation. 2.9. Statistical analysis One-way ANOVA testing was carried out to evaluate statistical significance. A Bonferroni t-test analysis was used to validate the ANOVA test. A value of p < 0.05 was considered as the minimal level of significance in the various experiments. 3. Results and discussion 3.1. Characterization of the complex in solution To exclude any degradation of the RSV under experimental conditions, acqueous solutions of the drug (32.8 ␮M) were stored at 25 ± 0.1 ◦ C, in the dark, for 5 days, in the absence or in the presence of increasing amounts of SBE-␤-Cyd (RSV:SBE-␤-Cyd molar ratio, 1:1, 1:10, 1:50 and 1:100). Portions of each sample were periodically withdrawn and observed through UV–vis spectroscopy. Trans-RSV shows two UV bands, one centered at around 215 nm and the other one at about 300 nm. This latter band presents a bandwidth of 20 nm and two small maxima centered at 305 and 316 nm [30]. After conversion from trans- to cis-RSV the band at 300 nm disappears and a band is present at 285 nm [16]. No variation of the UV–vis spectra was observed for the RSV solution prepared in the absence of SBE-␤-Cyd, showing there was no conversion from trans- to cis-RSV under experimental conditions (spectra not showed). In the presence of increasing amounts of SBE-␤-Cyd no conversion from trans- to cis-RSV was observed, but a variation in the intensity of the two bands of RSV was present for the entire duration of the experiment (the spectra obtained after 5 days are shown in Fig. 2). Particularly, a different influence was exerted on the two UV bands by the Cyd: (i) a hypochromic effect for one band at 300 nm up to 1:50 molar ratio (the spectra overlap at this wavelength) and an hyperchromic effect on the same band in the presence of the highest SBE-␤-Cyd concentration (1:100 molar ratio); and (ii) a slight hyperchromic effect for the other band at 215 nm up to 1:10 molar ratio; the same band was not well defined and almost disappeared at higher SBE-␤-Cyd molar ratios (1:50 and 1:100). A marginal red shift of about 3 nm was registered for both bands at all considered molar ratios. The variations that occur in the UV–vis spectrum are due to a perturbation of the microenvironment of a drug as a consequence of its complexation with the macrocycle [31,32], generally accompanied by the establishment of dipole–dipole, electrostatic, van der Walls and/or hydrogen bond-type interactions. The trend observed in our study demonstrated the existence in solution of an interaction between RSV and SBE-␤-Cyd which probably involves different binding processes as a function of the Cyd concentration. At lower RSV:Cyd molar ratios, an inclusion complex formed between the RSV and the SBE-␤-Cyd cavity (the aforementioned hypochromic effect on

V. Venuti et al. / Colloids and Surfaces B: Biointerfaces 115 (2014) 22–28

1,6

25

0,006

1,4 1,2

RSV(in solution) (M)

0,004

1,0

Abs

0,8 0,6

0,002

0,4

A

0,000

0,2 0,000

0,002

0,0 200

250

300 Wavelength (nm)

350

0,004

0,006

0,008

SBE- -Cyd (M)

400

120 Fig. 2. UV–vis spectra of trans-RSV free (32.8 ␮M, solid line) and in presence of SBE␤-Cyd at different drug-Cyd molar ratios (1:1 – ·· – ··; 1:10 - · - · - · -; 1:50 ·········; 1:100 - - - - -), after 5 days in the dark at 25 ± 0.1 ◦ C.

% Dissolution

60

B

30

0 0

20

40

60

80

100 120 140 160 180 200

Time (min) Fig. 3. Phase solubility diagram of RSV in the presence of increasing concentrations of SBE-␤-Cyd (0.0–8.0 mM) in water at 25.0 ± 0.1 ◦ C (A) and dissolution profile of free RSV (open squares) and RSV/SBE-␤-Cyd 1:1 inclusion complex (open circles) (B) in water at 37.0 ± 0.1 ◦ C. Each value is the average of three different experiments ± standard deviation.

3.2. Characterization of the complex in the solid state The FTIR-ATR technique was chosen in order to detect complex formation in solid phase and to point out the implication of the different functional groups of guest and host molecules in the inclusion process. This is done by analyzing the significant changes in the vibrational features of the complex with respect to those 0,50 0,45 0,40

-5

Abs[RSV] (10 )

the band at 300 nm); the complex formed in 1:1 stoichiometry, as subsequently demonstrated by means of a solubility phase diagram and Job’s plot method. At this point, the excess of Cyd in the solution (1:10 and 1:50 molar ratios) does not further influence the intensity of the band that appears overlapped in practically all solutions. By increasing the Cyd molar ratio, other binding processes could occur in the solution, for example a non-inclusion complexation between RSV and SBE-␤-Cyd aggregates, present in solution at high concentrations [33], or else an interaction between the sulfate group linking at the edge of the SBE-␤-Cyd cavity and the phenolic groups of the RSV. As far as the band at 215 nm is concerned, any variation observed in the spectra should cover an alteration of the dimer/monomer equilibrium of the RSV and (as reported by López-Nicolás and García-Carmona [30] for the interaction of RSV with HP-␤-Cyd), the inclusion of the RSV into the SBE-␤-Cyd cavity probably alters this equilibrium toward the presence of a large amount of monomer in the solution, producing a suppression of this band at the highest Cyd concentration. The most common method used to evaluate the ability of Cyd to complex a molecule is the phase-solubility study. The solubility diagram shown in Fig. 3A evidences the fact that the degree of solubility of RSV was increased in a concentration-dependent way by the complexation of this drug with SBE-␤-Cyd, showing favorable host/guest interaction. In particular, an AL type diagram was obtained and, since the slope of the diagram is less than 1, the stoichiometry of the complex was assumed to be 1:1 over the investigated concentration range. The stability constants Kc were calculated from the straight-line portion of the phase-solubility diagram, according to the equation Kc = ˛/S0 (1 − ˛), where ˛ is the slope of the linear plot reporting the amount of complexed RSV as a function of Cyd and S0 is the solubility of RSV in water. The Kc value turned out to be 10,114 M−1 . The stoichiometry of the complex was confirmed by Job’s plot, considering the variation of the UV–vis spectra of RSV in the region of the 300 nm band, in the presence of SBE-␤-Cyd. A maximum value at R = 0.5 and a highly symmetrical shape evidenced the presence of a complex with 1:1 stoichiometry within the range of the investigated concentrations (Fig. 4). These results are in agreement with the phase solubility studies. Complexation strongly increased the water solubility of RSV (∼0.03 mg/ml and ∼1.1 mg/ml, at 25 ◦ C for free RSV and RSV–SBE␤-Cyd inclusion complex, respectively) producing a very rapid dissolution of the complexed drug (Fig. 3B).

90

0,35 0,30 0,25 0,20 0,15 0,10 0,0

0,2

0,4

0,6

0,8

1,0

R Fig. 4. Continuous variation plot (Job’s plot) for the complexation of RSV with SBE␤-Cyd from absorbance measurements.

V. Venuti et al. / Colloids and Surfaces B: Biointerfaces 115 (2014) 22–28

a

IR absorbance (arb. units)

26

4 3 5

2

6

1 3800

3600

3400

3200

3000

2800

-1

Wavenumber (cm )

of the individual components and the physical mixture (a simple blending of the two constituents without complexation) [34]. The FTIR-ATR spectra of RSV, SBE-␤-Cyd, RSV + SBE-␤-Cyd 1:1 physical mixture and RSV/SBE-␤-Cyd 1:1 inclusion complex are represented in Fig. 5. The FTIR-ATR spectrum of RSV showed the presence of the following main peaks: ∼ 3214 cm−1 (large band), assigned to the free O H stretching vibration, a double peak at ∼1600 and ∼1587 cm−1 , corresponding to C C aromatic double bond stretching and C C olefinic stretching. Again, the peaks at ∼1513 and ∼1463 cm−1 reflect benzene skeleton vibrations, and C C stretching vibrations are evidenced by the peak at ∼1380 cm−1 . The peaks at ∼988 and ∼965 cm−1 are ascribed to the bending vibration of C C H, and demonstrate the trans-form of RSV [35]. The spectrum of SBE-␤Cyd is mainly characterized by intense bands at 3700–3000 cm−1 due to the O H stretching vibration, which overlapped with the band associated to the vibration of the CH and CH2 groups that appears in the 3000–2800 cm−1 region. The band at ∼1644 cm−1 reflects the ␦-HOH bending of water molecules attached to Cyd, whereas the peaks at ∼1148 and ∼1016 cm−1 are respectively ascribed to C H and C O stretching vibrations. The FTIR-ATR spectrum of the physical mixture contains all the contributions coming form both RSV and SBE-␤-Cyd without showing any significant differences. This demonstrates that the drug remains intact and that there is no interaction between drug and macrocycle. On the contrary, when compared to that of the physical mixture, the spectrum of the inclusion complex exhibits relevant changes in center-frequencies, intensities and widths of the characteristic absorption peaks, revealing the formation of a new chemical bond between the RSV and SBE-␤-Cyd [36]. Information concerning the “host–guest” interactions driving the complexation mechanism has been gathered principally by observing the high-frequency region of the spectra. Previous studies demonstrated that the curve-fitting technique can be applied to the FTIR-ATR O H stretching vibration band in order to quantitatively account for the modification in the H-bond scheme upon complexation with ␤-Cyds [37]. According to these studies, the O H stretching band is composed of several unresolved components corresponding to primary and secondary OH groups of the host molecule (SBE-␤-Cyd in our case), of the guest (RSV), and of interstitial and intracavity crystallized water molecules. Because of the partial overlapping of the O H and C H stretching bands in our experimental spectra, the first extending from ∼3700 to ∼3000 cm−1 , and the second from ∼3000 to ∼2800 cm−1 , we initially proceeded by fitting the whole range, by means of

b

IR absorbance (arb. units)

Fig. 5. Experimental FTIR-ATR spectra of RSV (a), SBE-␤-Cyd (b), RSV + SBE-␤-Cyd physical mixture (c), and RSV–SBE-␤-Cyd, 1:1 inclusion complex (d).

3

4

5

2

6

1 3800

3600

3400

3200

3000

2800

-1

Wavenumber (cm ) Fig. 6. The fit of the FTIR-ATR bands in the OH stretching region for RSV + SBE-␤-Cyd, 1:1 physical mixture (a), and RSV–SBE-␤-Cyd, 1:1 inclusion complex (b).

Voigt profiles. Then we subtracted the contributions coming from symmetric and antisymmetric methyl stretches from the total fits. Finally, a more detailed curve fitting was exclusively applied to the O H stretching region. The second derivative computation of the experimental profiles gave us a first indication of the number of the band components and of the wave number of their maxima [28,38]. Five sub-bands were distinguished, as clearly reported in Fig. 6a and b in the case of RSV + SBE-␤-Cyd 1:1 physical mixture and RSV–SBE-␤-Cyd 1:1 inclusion complex, respectively. The corresponding center-frequencies and percentage intensities are reported in Table 1, together with an interpretation according to recent pertinent literature [39]. All of the center frequencies of the sub-bands shifted to higher wave numbers during the passage from physical mixture to inclusion complex. Since the center-frequencies are indicative of the strength of the environment to which the OH oscillators belong, this spectral change indicates a rearrangement, upon complexation, of the hydrogen bond network toward a less intense association, i.e., in favor of schemes of reduced co-operativity. By looking at the percentage intensities which account for the population of the OH groups involved in a specific environment, we can hypothesize a complexation mechanism similar to that revealed for a variety of similar systems [36,37]. According to this mechanism, the insertion of the guest inside the host will cause, on one hand, the partial escape of intracavity H2 O molecules (reflected in the decrease of I1 ), that will redistribute outside of the torus, and, on the other hand, a general altering of the H-bond network, in which interstitial aggregates are preferred (as taken into account by all the other changes of intensities). In particular, the observed increase of I4 , passing from physical mixture to inclusion complex

V. Venuti et al. / Colloids and Surfaces B: Biointerfaces 115 (2014) 22–28

27

Table 1 Main fitting parameters, i.e., wavenumber (␻i, cm−1 , i = 1–5) and percentage intensity (Ii/IOH, %, i = 1–5) of the FTIR-ATR sub-bands in the O H stretching region. O H stretching intra-cavity H2 O

O H stretching primary OH groups

O H stretching interstitial H2 O

O H stretching O H groups of RSV

O H stretching secondary OH groups

O H stretching interstitial H2 O

ω1 (cm−1 )

I1 (%)

ω2 (cm−1 )

I2 (%)

ω3 (cm−1 )

I3 (%)

ω4 (cm−1 )

I4 (%)

ω5 (cm−1 )

I5 (%)

ω6 (cm−1 )

I6 (%)

3588.5 3607.2

4.9 3.8

3428.5 3539.4

15.8 11.9

3311.6 3430.9

22.4 26.5

3215.7 3301.2

30.6 32.0

3119.9 3210.3

18.5 15.6

3012.6 3105.5

7.8 10.2

120,0 100,0

Cell Viability (%)

and accounting for the population of the O H groups of RSV, testifies to the breaking of the intramolecular hydrogen bond of RSV upon complexation, and the involvement of these groups in less intense environments, as indicated by the high wave number shift of the center-frequency of the sub-band. Evident changes upon complexation were also detected in the C H stretching vibration (see Fig. 5). The two peaks at ∼2940 cm−1 and ∼2889 cm−1 in the SBE-␤-Cyd spectrum, still distinguishable in the physical mixture, tend to completely overlap in the inclusion complex, suggesting the involvement of these groups in the establishment of host–guest interactions. Proceeding to the lower frequencies, the double peak seen at ∼1600 cm−1 and ∼1587 cm−1 of RSV, ascribed to C C aromatic double bond stretching and C C olefinic stretching, is still clearly detectable, almost unaltered, in the RSV + SBE-␤-Cyd 1:1 physical mixture, whereas it appears enlarged and strongly reduced in intensity in the spectrum of the RSV–SBE-␤-Cyd 1:1 inclusion complex. Both of these changes reveal a hindering of the corresponding vibrations as a consequence of the inclusion of these groups in the SBE-␤-Cyd cavity. Continuing, the benzene skeleton vibrations (peaks at ∼1513 and ∼1463 cm−1 in the RSV spectrum) and C C stretching vibrations (peak at ∼1380 cm−1 in the RSV spectrum), still evident in the case of the physical mixture, are strongly bumped and tend to disappear in the case of the inclusion complex. Again, a close fitting of these groups inside the host cavity can explain these spectral variations. The peak at ∼1380 cm−1 in the RSV spectrum, accounting for the C C stretching vibrations, undergoes a low-frequency shift, a broadening and a reduction in intensity passing from a physical mixture to an inclusion complex. The low-frequency shift can be explained in terms of a more intense electrostatic environment in which the C C bond finds itself as a consequence of inclusion, which contributes to the weakening of the corresponding dipole moment. On the other hand, the inclusion of this functional group will hinder the corresponding stretching vibration, justifying its broadening and diminishing in intensity. Finally, we also observed an intense increase of the band at ∼1644 cm−1 of SBE-␤-Cyd during the passage from physical mixture to inclusion complex, ascribed to the ␦-HOH bending of water molecules attached to the Cyd. Since it is an established fact that this band is associated with water molecules that are not involved in a symmetric tetrahedral network [40], this occurrence could testify to the breaking, upon complexation, of extended tetrahedral H-bonded arrangements, and the tendency of crystallized water molecules to organize in networks which are not fully bonded, in agreement with what has already been revealed by the analysis of the O H stretching region.

80,0 60,0 40,0 20,0 0,0 Control

RSV

RSV-SBE-β-Cyd

Fig. 7. Cytotoxic effects of free RSV and RSV–SBE-␤-CyD inclusion complex against an MCF-7 human breast cancer cell line at different exposition times:

24 h,

48 h and 72 h. Results are presented as the mean of three different experiments (6 replicates for each points) ± standard deviation.

as a suspension, while the inclusion complex and free Cyd were solutions. As expected, the free SBE-␤-Cyd elicited no significant advantage in terms of in vitro anticancer activity (data not shown). At all incubation times, it was possible to observe a well-defined reduction of MCF-7 viability elicited by free and complexed RSV (Fig. 7). However, it was impossible to detect any appreciable difference in antitumoral activity between the RSV–SBE-␤-Cyd inclusion complex and the free drug at 24 h. The best results in terms of reduction of cell viability were obtained after 48 h and 72 h for the RSV–SBE␤-Cyd inclusion complex, which induced a significant degree of cytotoxicity with respect to the free drug. The improved anticancer activity observed for the complex with respect to RSV alone was probably due to the balance of two factors: (i) the solubilizing effect of SBE-␤-Cyd on RSV and (ii) the Kc value of the inclusion complex. In fact, only solubilized RSV (and not associated to Cyd) can penetrate the viable cells and exert its anticancer activity. The effect of RSV alone was limited by its poor solubility. On the other hand, the complexed RSV was totally solubilized, and showed greater anticancer activity with respect to RSV alone. However, this activity was not as much as we expected on the basis of the drug water solubility improvement (0.03 mg/ml and 1.1 mg/ml for free and complexed RSV, respectively). Probably, due to the high Kc value (10,114 M−1 ) of the inclusion complex, large amounts of the drug were associated to the Cyd and were not able to penetrate within the cells, resulting in a reduced effect.

3.3. In vitro biological assay

4. Conclusions

Due to the important cancer chemopreventive activity shown by RSV [41], we assayed the influence of complexation within SBE␤-Cydon its in vitro anticancer activity in terms of cytotoxicity by using the cell viability MTT test. The cytotoxic effect was evaluated as a function of the incubation time (24, 48 or 72 h) in order to define the time-exposition effect. The test was performed on RSV–SBE-␤Cyd inclusion complex and compared with free RSV and SBE-␤-Cyd alone. Due to its poor water solubility, the free RSV was assayed

Phase solubility analysis, UV–vis spectroscopy and FTIR-ATR spectroscopy were used to study the inclusion complex formed by RSV and SBE-␤-Cyd both in solution and in solid state. In a liquid phase, the formation of an inclusion complex with 1:1 stoichiometry and a high stability constant was observed. Complexation produced consistent improvement in the solubility of RSV in water and consequently a high dissolution rate. Subsequently, FTIR-ATR spectroscopy was employed to evidence the

28

V. Venuti et al. / Colloids and Surfaces B: Biointerfaces 115 (2014) 22–28

formation of RSV–SBE-␤-CyD 1:1 inclusion complexes in a solid phase, allowing the investigation of the host–guest interactions and considerable differences were revealed between the spectra of the physical mixtures and the complex. We demonstrated that the formation and/or modification of polar bonds play the main role in inclusion phenomena. In particular, the complexation mechanisms were successfully monitored through the decomposition of the O H stretching FTIR-ATR band. Preliminary in vitro biological results showed that SBE-␤-Cyd improves the anticancer activity of RSV on MCF-7 cell lines, and its action is the result of a balance between the increased water solubility of the drug and the high stability constant. This study demonstrated that SBE-␤-Cyd has a good potential of being a suitable delivery system for the oral and/or parenteral administration of RSV. In fact, the high degree of water solubility and rapid dissolution obtained for the drug could improve its oral bioavailability, at the moment very limited. Moreover, due to the biocompatibility of the Cyd used, a liquid formulation with anticancer properties could be developed for parenteral administration. Other in vitro and in vivo studies are in progress to confirm the effectiveness of a parenterally administered RSV–SBE-␤-Cyd inclusion complex. Conflict of interest The authors report no conflicts of interest. Acknowledgment This study was financially supported by MIUR (Ministerodell’Istruzione, dell’Università e dellaRicerca), PRIN2009 Grant No. 2009Z8YTYC. The authors are grateful to Lynn Whitted for her revision of the language. References [1] [2] [3] [4] [5]

L. Frémont, Life Sci. 66 (2000) 663. S. Fulda, Drug Discov. Today 15 (2010) 757. S.M. Meeran, S.K. Katiyar, Front. Biosci. 13 (2008) 2191. S.M. Hadi, S.F. Asad, S. Singh, A. Ahmad, IUBMB 50 (2000) 167. M. Belleri, D. Ribatti, S. Nicoli, F. Cotelli, L. Forti, V. Vannini, L.A. Stivala, M. Presta, Mol. Pharmacol. 67 (2005) 1451. [6] Y.J. Surh, K.S. Chun, H.H. Cha, S.S. Han, Y.S. Keum, K.K. Park, S.S. Lee, Mutat. Res. 480/481 (2001) 243. [7] B. Dasgupta, J. Milbrandt, Proc. Natl. Acad. Sci. U.S.A. 104 (2007) 7217. [8] S.S. Leonard, C. Xia, B.H. Jiang, B. Stinefelt, H. Klandorf, G.K. Harris, X. Shi, Biochem. Biophys. Res. Commun. 309 (2003) 1017.

[9] M. Athar, J.H. Back, X. Tang, K.H. Kim, L. Kopelovich, D.R. Bickers, A.L. Kim, Toxicol. Appl. Pharmacol. 224 (2007) 274. [10] X. Hu, Y. Xuan, Cancer Lett. 259 (2008) 127. [11] K.P.L. Bhat, J.W. Kosmeder, J.M. Pezzuto, Antioxid. Redox Signal. 3 (2001) 1041. [12] K.T. Howitz, K.J. Bitterman, H.Y. Cohen, D.W. Lamming, S. Lavu, J.G. Wood, R.E. Zipkin, P. Chung, A. Kisielewski, L.L. Zhang, B. Scherer, D.A. Sinclair, Nature 425 (2003) 191. [13] S. Wei, S.K. Kulp, C.S. Chen, J. Biol. Chem. 285 (2010) 9780. ´ M. Sentjurc, Eur. J. Pharm. Biopharm. [14] J. Kristl, K. Teskac, C. Caddeo, Z. Abramovic, 73 (2009) 253. [15] C. Rius, M. Abu-Taha, C. Hermenegildo, L. Piqueras, J.M. Cerda-Nicolas, A.C. ˜ J. Cortijo, E.J. Morcillo, F. Orallo, M.J. Sanz, J. Immunol. 185 Issekutz, L. Estan, (2010) 3718. [16] C. Bonda, J. Zhang, A. Pavlovic, Cosmet. Toiletries 126 (2011) 652. [17] E.M.M. Del Valle, Process Biochem. 39 (2004) 1033. [18] C.A. Ventura, S. Tommasini, A. Falcone, I. Giannone, D. Paolino, V. Sdrafkakis, M.R. Mondello, G. Puglisi, Int. J. Pharm. 314 (2006) 37. [19] R. Stancanelli, A. Mazzaglia, S. Tommasini, M.L. Calabrò, V. Villari, M. Guardo, P. Ficarra, R. Ficarra, J. Pharm. Biomed. Anal. 44 (2007) 980. [20] C. dos Santos, M. Pilar Buera, M. Florencia Mazzobre, Food Chem. 132 (2012) 2030. [21] H. Liang, Q. Yuan, F. Vriesekoop, F. Lv, Food Chem. 135 (2012) 1020. [22] V.J. Stella, Q. He, Toxicol. Pathol. 36 (2008) 30. ˜ [23] C. Lucas-Abellán, I.l. Fortea, J.M. López-Nicolás, E. Núnez-Delicado, Food Chem. 104 (2007) 39. [24] C. Lucas-Abellán, M.T. Mercader-Ros, M.P. Zafrilla, M.L. Fortea, J.A. Gabaldón, E. ˜ Núnez-Delicado, J. Agric. Food Chem. 56 (2008) 2254. [25] Z. Lu, B. Cheng, Y. Hu, Y. Zhang, G. Zou, Food Chem. 113 (2009) 17. [26] T. Higuchi, K.A. Connors, Adv. Anal. Chem. Instrum. 4 (1965) 117. [27] S. Gibaud, S.B. Zirar, P. Mutzenhardt, I. Fries, A. Astier, Int. J. Pharm. 306 (2005) 107. [28] V. Crupi, F. Longo, D. Majolino, V. Venuti, J. Phys.: Condens. Mater. 18 (2006) 3563. [29] B.D. Hahn, R.H.H. Neubert, S. Wartewig, A. Christ, C. Hentzsch, J. Pharm. Sci. 89 (2000) 1106. [30] J.M. López-Nicolás, F. García-Carmona, Food Chem. 118 (2010) 648. [31] C.A. Ventura, I. Giannone, T. Musumeci, R. Pignatello, L. Ragni, C. Landolfi, C. Milanese, D. Paolino, G. Puglisi, Eur. J. Med. Chem. 41 (2006) 233. [32] C. Tablet, L. Dumitrache, L. Minea, M. Hillebrand, Rev. Roum. Chim. 57 (2012) 665. [33] T. Loftsson, M. Másson, M.E. Brewster, J. Pharm. Sci. 93 (2004) 1091. [34] C. Aram˘a, C. Nicolescu, A. Nedelcu, C. Monciu, J. Incl. Phenom. Macrocycl. Chem. 70 (2011) 421. [35] V. Bertacche, N. Lorenzi, D. Nava, E. Pini, C. Sinico, J. Incl. Phenom. Macrocycl. Chem. 55 (2006) 279. [36] C. Cannavà, V. Crupi, M. Guardo, D. Majolino, R. Stancanelli, S. Tommasini, C.A. Ventura, V. Venuti, J. Incl. Phenom. Macrocycl. Chem. 75 (2013) 255. [37] V. Crupi, G. Guella, D. Majolino, I. Mancini, A. Paciaroni, B. Rossi, V. Venuti, P. Verrocchio, G. Viliani, Philos. Mag. 91 (2011) 1776 (and reference therein). [38] V. Crupi, R. Ficarra, M. Guardo, D. Majolino, R. Stancanelli, V. Venuti, J. Pharm. Biomed. Anal. 44 (2007) 110. [39] C. Cannavà, V. Crupi, P. Ficarra, M. Guardo, D. Majolino, R. Stancanelli, V. Venuti, Vib. Spectrosc. 48 (2008) 172. [40] J.B. Brubach, A. Mermet, A. Filabozzi, A. Gerschel, D. Lairez, M.P. Krafft, J. Phys. Chem. B105 (2001) 430. [41] J.K. Kundu, Y.J. Surh, Cancer Lett. 269 (2008) 243.