Food Research International 51 (2013) 748–755

Contents lists available at SciVerse ScienceDirect

Food Research International

j ourna l homepage: www.e lsev ie r .com/ locate / foodres

Effect of cyclodextrins on the thermodynamic and kinetic propertiesof cyanidin-3-O-glucoside

Ana Fernandes, André Sousa, Joana Azevedo, Nuno Mateus, Victor de Freitas ⁎Centro de Investigação em Química (CIQ), Departamento de Química e Bioquímica, Faculdade de Ciências, Universidade do Porto, Rua do Campo Alegre, 687, 4169-007 Porto, Portugal

⁎ Corresponding author. Tel.: +351 220402558; fax:E-mail address: vfreitas@fc.up.pt (V. de Freitas).

0963-9969/$ – see front matter © 2013 Elsevier Ltd. Allhttp://dx.doi.org/10.1016/j.foodres.2013.01.037

a b s t r a c t

a r t i c l e i n f o

Article history:Received 10 October 2012Accepted 16 January 2013

Keywords:AnthocyaninCyanidin-3-O-glucosideCyclodextrinHydration equilibrium constantInclusion complexKinetic constants

In this study, the effect of α and β-cyclodextrin on cyanidin-3-O-glucoside color was investigated by UV–visible absorption techniques. The equilibrium and kinetic constants of the network of chemical reactionstaking place in cyanidin-3-O-glucoside were also studied in water at 25 °C by UV–visible absorption tech-niques. The results showed that the addition of β-cyclodextrin resulted in the fading of anthocyanin solution,and this fading effect was greater at higher pH. This anti-copigmentation effect is caused by the selective in-clusion and stabilization of the anthocyanin colorless forms into the β-cyclodextrin cavity. Oppositely, nochanges were observed in the cyanidin-3-O-glucoside absorption spectra with the addition ofα-cyclodextrin.Direct pH jump, from thermally equilibrated solutions at pH=1.0 (flavylium cation, AH+), shows threekinetic processes: formation of the base A, hydration reaction to form the hemiketal B and the chalconecis–trans isomerization (Cc–Ct). The results obtained clearly indicated that the equilibrium and kinetic con-stants of the network of chemical reactions taking place in cyanidin-3-O-glucoside were affected by the pres-ence of β-cyclodextrin. Molecular inclusion in the β-cyclodextrin cavity resulted in the increase of theisomerization observed rate constant (kobs) at pH 5.3 and in the increase of the hydration equilibrium con-stant Kh which is in agreement with the fading of the anthocyanin solution. For the macrocycleα-cyclodextrin, no significant changes were observed on the equilibrium and kinetic constants, which sug-gests that the inclusion of cyanidin-3-glucoside in the α-cyclodextrin's cavity is not favored.

© 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Color is one of the most important attributes in food products, asit affects the acceptability of products by consumers (Burin, Rossa,Ferreira-Lima, Hillmann, & Boirdignon-Luiz, 2011). Nowadays, as aconsequence of perceived consumer preferences as well as legisla-tive action, there is a worldwide trend towards the development offood colorants from natural sources, and the replacement of synthet-ic dyes by these natural colorants has been increasing (Idham,Muhamad, & Sarmidi, 2012). In this context, anthocyanins are attrac-tive as food colorants since these natural pigments are responsible forthe colors of flowers, fruits and vegetables, provide high colorantpower and present a low toxicity and water solubility, which permittheir incorporation in many food systems (Brouillard, Chassaing,Isorez, Kueny-Stotz, & Figueiredo, 2010; Ersus & Yurdagel, 2007).Moreover, several studies have suggested that these compounds arebeneficial to human health and have been responsible for several pos-itive therapeutic effects (Bridle & Timberlake, 1997; Clifford, 2000).

+351 220402659.

rights reserved.

However, the use of these colorants in food products may facesome problems due to their instability (Francis & Markakis, 1989).The color and stability of anthocyanins are highly dependent on pH,due to changes in the concentration of the four species present in acid-ic and neutral aqueous solutions: flavylium cation (AH+), quinoidalbase (A), hemiketal (B) and chalcone (C) (Fig. 1) (Brouillard, 1982;Brouillard & Delaporte, 1977; Pina, Melo, Laia, Parola, & Lima, 2012).Conversion of one species to another is typically accompanied by dra-matic changes in color and stability (Burin et al., 2011; Xiong, Melton,Easteal, & Siew, 2006). To increase the stability of anthocyanins, mo-lecular inclusion and encapsulation with several carbohydrates havebeen used (Burin et al., 2011; Cai & Corke, 2000; Ersus & Yurdagel,2007; Idham et al., 2012; Selim, Khalil, Abdel-Bary, & Abdel-Azeim,2008).

Cyclodextrins (CDs) have been used to stabilize anthocyanins(Chandra, Nair, & Iezzoni, 1993; Lewis, Walker, & Lancaster, 1995;Mourtzinos et al., 2008). These compounds are cyclic oligosaccharidescomposed of β-D-glucopyranosyl units linked byα-(1→4) bonds. Themost common areα-,β- andγ-cyclodextrinwhich have six, seven andeight glucose units, respectively. The glucose monomers are orientat-ed in a cyclic manner, forming a typical conical or truncated conestructure with a hydrophobic hollow cavity of specific diameter andvolume and a relatively hydrophilic outer surface (Pegg & Shahidi,

Fig. 1. Structural transformations of flavylium cation in strongly acidic to alkaline aqueous media (Brouillard, 1982; Brouillard & Delaporte, 1977).

749A. Fernandes et al. / Food Research International 51 (2013) 748–755

2007). The hollow molecular shape gives cyclodextrins their uniqueability to form reversible inclusion complexes with a wide variety oforganic compounds (often phenolic substances) yielding supramole-cules (Hendy & Breslin, 2011). Formation of the host–guest complexcan be easily monitored using techniques such as nuclear magneticresonance (NMR) (Cai et al., 1990; Ishizu, Kintsu, & Yamamoto, 1999;Ishizuka et al., 2002), UV–visible spectroscopy (UV–vis) (Divakar,1993), fluorescence spectroscopy (Divakar & Maheswaran, 1997), IRspectroscopy, electrochemical approaches and solubility measure-ments (Hendy & Breslin, 2011).

Concerning anthocyanins, they include in their structure hydro-phobic aromatic moieties and hydrophilic polar groups like hydroxylgroups. This amphiphilic character makes anthocyanins a very goodcandidate for molecular inclusion with cyclodextrins (Dangles,Stoeckel, Wigand, & Brouillard, 1992). Several studies have reportedthe formation of inclusion complexes of anthocyaninswith cyclodextrinmacrocycles (Dangles & Brouillard, 1992; Dangles, Stoeckel, et al., 1992;Dangles, Wigand, & Brouillard, 1992; Lewis et al., 1995; Mourtzinos etal., 2008; Nagatomo, 1985; Tamura, Takada, Yamagami, & Shiomi,1998; Yamada, Komiya, & Akaki, 1980).

However, interaction between cyclodextrins and some anthocya-nins was found to promote the anthocyanin discoloration and lead toan anti-copigmentation effect (Dangles & Brouillard, 1992; Dangles,Stoeckel, et al., 1992; Lewis et al., 1995; Yamada et al., 1980).

The main goal of this study was to promote the use of anthocya-nins as food colorants and to overcome the limitations of the applica-tion of these natural pigments in food systems, enhancing its stability.In this study, the effect of cyclodextrins (α and β) on the color ofcyanidin-3-O-glucoside was evaluated. The effect of these two cyclo-dextrins on the equilibrium and kinetic constants of the network ofchemical reactions taking place in cyanidin-3-O-glucoside was alsoevaluated by UV–visible absorption techniques.

2. Materials and methods

2.1. Materials

β-Cyclodextrin (β-CD) and α-cyclodextrin (α-CD) were pur-chased from Sigma-Aldrich® (Madrid, Spain). All aqueous solutionswere prepared with distilled water. Cyanidin-3-O-glucoside (cy3glc)was extracted and purified in the laboratory from blackberries(Rubus fruticosus) by semipreparative HPLC with C18 reversed phasecolumn, as described elsewhere (Azevedo et al., 2010).

A universal buffer of Theorell Stenhagen was made by dissolving2.25 mL of phosphoric acid (85% w/w), 7.00 g of monohydrated citricacid, 3.54 g of boric acid and 343 mL of 1 M NaOH solution in water(until 1 L completion) (Küster & Thiel, 1982).

2.2. pH measurements

All pH measurements were made in a pH meter WTW pH 320fitted with a Crison electrode. The calibration was made with stan-dard buffers pH 7.00 and pH 4.00 purchased from Crison.

2.3. Effect of cyclodextrins on the color of cyanidin-3-O-glucoside

A stock solution of cy3glc (1.50×10−4 M) was prepared using asolution of 0.1 M HCl. The low pH value of the stock solution ensuredthat most of the cy3glc was in the flavylium form and thus preventedpigment degradation. The solution composed of 1.0 mL of cy3glc stocksolution, 1.0 mL of NaOH solution (0.1 M) and 0.5 mL of universalbuffer solution at pH 1.0 and 2.0 was allowed to equilibrate for 2 h.To these anthocyanin solutions β-CD was added at various concentra-tions (6.00×10−4; 1.80×10−3 and 7.50×10−3 M) and the solutionswere left to equilibrate for 2 h. The absorbance (260–700 nm) was

Fig. 2. A, UV–visible spectra of pure cyanidin-3-O-glucoside and in the presence of β-CDat different concentrations at pH 2.0 (25 °C); Cy3glc: 1.50×10−4 M (0); β-CD: 6.00×10−4 M (1); 1.80×10−3 M (2); 7.50×10−3 M (3); B, UV–visible spectra of purecyanidin-3-O-glucoside and in the presence of α-CD at different concentrations atpH 2.0 (25 °C); Cy3glc: 1.50×10−4 M (0); α-CD: 6.00×10−4 M (1); 1.80×10−3 M (2);7.50×10−3 M (3).

750 A. Fernandes et al. / Food Research International 51 (2013) 748–755

measured in an UV–vis spectrophotometer (Thermo Scientific Evolu-tion Array UV–vis spectrophotometer) fitted with a thermostated1.0 cm path length cuvette device at 25 °C. Similar studies wereperformed with α-CD.

2.4. pH jump

All thermodynamic and kinetic constants of the cy3glc solution andof the inclusion complexes cy3glc–β-CD and cy3glc–α-CD were deter-mined by a spectrophotometric method, as described elsewhere (Naveet al., 2010; Pina, 1998). A stock solution of cy3glc (1.50×10−4 M)was prepared as described above. For the molecular inclusion complex,a β-CD and an α-CD solution (7.50×10−3 M) were prepared in 0.1 MHCl. Cy3glc was dissolved in this solution at an approximate molarratio of 1:50 (cy3glc:α or β-CD, the molar ratio which induced thelargest flavylium cation absorbance decrease) and left to equilibrate.pH jumps were carried out by adding the necessary amount of NaOH(0.1 M) to neutralize the stock solution of flavylium cation (cy3glc)and the necessary amount of a universal buffer solution (at differentpH values). The ionic strengthwasmaintained at 0.1 M so that the effectof ion-pair formation could beminimized. The concentration of the buff-er (Theorell Stenhagen universal buffer)wasmaintained low in order toavoid the buffer effects. Throughout the whole experiment the temper-ature was kept at 25 °C.

3. Results and discussion

The ability of a macrocycle molecule to form an inclusion complexwith a guest molecule depends mainly on two factors: the first con-cerns the steric requirements and depends on the relative size ofthe cyclodextrin and the size of the molecule to include; the secondcritical requirement corresponds to the thermodynamic interactionsbetween different components of the system (cyclodextrin, moleculeto include and solvent) (Bourvellec, 2003). Several forces such as vanderWaals forces (hydrophobic interaction, dipole–dipole interaction)and hydrogen bonding interaction are involved in the binding ofguest molecules to the cyclodextrin cavity. These forces are capableof forming a stable complex, but not permanently as the guest mole-cule can still be released from the complex to become available for theintended effect of the guest molecule (Hedges, Shieh, & Sikorski, 1995).Among different naturally occurring anthocyanins, cyanidin-3-O-glucoside was selected for this study because of its structural sim-plicity and reduced steric hindrance, a property needed for a goodinteraction with cyclodextrin (Dangles, Wigand, et al., 1992). Indeedfor sterically hindered pigments such as anthocyanidin diglycosidesand anthocyanidin monoglycosides which are heavily substitutedon their aromatic rings, the inclusion on the β-CD cavity is not fa-vored (Dangles, Stoeckel, et al., 1992).

3.1. Effect of cyclodextrins on the color of cyanidin-3-O-glucoside

Addition ofβ-CD to an acidic solution of cy3glc induces a strong de-crease in the anthocyanin visible absorption band (hypochromism)(Fig. 2) and this decrease was greater with higher concentrations ofβ-CD added. This effect was lower at pH 1.0, in which cy3glc is practi-cally 100% in its flavylium equilibrium form (8% absorbance decreaseat pH 1.0 instead of 38% absorbance decrease at pH 2.0, at molarratio 1:50). This fading effect, also known as anti-copigmentation phe-nomenon has previously been reported in several works (Dangles,Stoeckel, et al., 1992; Dangles, Wigand, et al., 1992; Lewis et al.,1995; Yamada et al., 1980). Since no change in the shape of the ab-sorption band corresponding to the flavylium form was observed, itcan be assumed that the flavylium cation is not the species interactingwith β-CD (Dangles, Wigand, et al., 1992). Indeed, anthocyaninflavylium cation maximum absorption is very sensitive to solvent ef-fects and the formation of an inclusion complex of the flavylium cation

with cyclodextrin would induce a large modification in its solva-tion shell, shifting the visible maximum of absorption of flavylium(Dangles, Wigand, et al., 1992). The large loss of color following theaddition of β-CD to the cy3glc solution points to the preferentialinclusion of the colorless forms (B, Cc, Ct) into the β-CD cavity, aphenomenon leading to the shifting of the pigment hydration equilib-rium towards the formation of more colorless forms (Brouillard et al.,2010; Yamada et al., 1980). A small increase of absorbance at λmax inthe ultraviolet region was observed with the addition of β-CD, whichagrees with the formation of the anthocyanin colorless forms. Theseforms are more abundant at pH 2.0 comparing to pH 1.0 whichcould explain the higher anti-copigmentation effect at the prior pH.

Oppositely, addition of α-CD to a cy3glc solution at pH 2.0 causedonly a small increase at λmax in the UV–visible spectrum. These re-sults appear to indicate that the inclusion of cy3glc on the α-CD'scavity at pH 2.0 is not favored (Fig. 2). α-CD differs from β-CD onlyin one glucose unit, having six instead of seven glucose units. Thismacrocycle has an inside diameter of 5.7 Å and an outside diameterof 13.7 Å while β-CD presents an inside diameter of 7.8 Å and an out-side diameter of 15.3 Å (Pegg & Shahidi, 2007). The dimensions of thecyclodextrin's cavity allow some selectivity for the complexation ofguest molecules and the smaller cavity of α-CD probably could not af-ford so many interactions between the colorless forms of cy3glc andthe cavity walls, thus resulting in weaker binding. On the otherhand, by increasing the size of the cyclodextrin, its hydrophobicityis enhanced leading to a significant increase in its affinity with poly-phenols (Tønnesen, Másson, & Loftsson, 2002).

Fig. 3. A, Absorption spectra of equilibrated solutions of cy3glc (6.0×10−5 M) at differ-ent pH values. Inset: Determination of pKa′ through the fitting of the absorption datataken at 511 nm by means of Eq. (6); B, absorption spectra of equilibrated solutionsof the inclusion complex cy3glc:β-CD (6.0×10−5 M:3.0×10−3 M) at different pHvalues. Inset: Determination of pKa′ through the fitting of the absorption data takenat 511 nm by means of Eq. (6).

751A. Fernandes et al. / Food Research International 51 (2013) 748–755

3.2. pH jumps

One useful way to follow the kinetic process occurring in the net-work of flavylium compounds is the use of relaxation techniques, inwhich the equilibrium of the system is shifted bymeans of an externalinfluence, applied very rapidly, forcing the metastable system thusformed to shift to the new state of equilibrium. The most commonperturbations used are instantaneous variations of temperature, pres-sure, concentration or pH, the latter being more advantageous since itdoes not produce secondary physical effects (Brouillard, Delaporte, &Dubois, 1978). The application of these relaxation methods allowsthe determination of both equilibrium constants and the reactionrate constants (Brouillard & Dubois, 1977; Maçanita et al., 2002).Among the structural transformations that an anthocyanin undergoesin an aqueous solution, the proton transfer equilibrium and thecycle-chain tautomerism are quite instantaneous. By contrast, the hy-dration reaction and the chalcone isomerization are relatively slowprocesses occurring on very distinct time scales. For instance, hydra-tion is a process whose equilibrium state is achieved within about10 s at pH 3–4 and about a fewminutes at a pH value closer to neutral-ity. The chalcone isomerization is much slower and pH independentand its equilibrium state is fully established within about 2 h. The ki-netics of both latter phenomena and the way they are influenced bythe presence of β-CD can thus be easily studied using a standardUV–visible spectrophotometer (Dangles, Stoeckel, et al., 1992).

In this work, the chemical reactions were studied by performingdirect pH jumps from acidic to basic, in thermally equilibrated solu-tions. As previously reported, in acidic media the following equationsaccount for the thermodynamic and kinetic processes (Brouillard &Delaporte, 1977; Pina, 1998):

AHþ ⇔Ka A þHþ Ka proton transfer ð1Þ

AHþ ⇔Kh BþHþ Kh hydration ð2Þ

B⇔Kt Cc Kt tautomerization ð3Þ

Cc ⇔K i

Ct K i isomerization: ð4Þ

These can be simplified in one single acid–base equilibrium withconstant K′a (Eqs. (5) and (6)) if the A, B, Cc and Ct species are repre-sented together as one generic conjugate base CB (given by the sumof the species A, B, Cc and Ct), which is in equilibrium with the acidicflavylium cation AH+:

K ′a ¼ Ka þ Kh þ KhKt þ KhKtK i ð5Þ

K ′a ¼

CB½ � Hþ� �

AHþ ; CB½ � ¼ A½ � þ B½ � þ Cc½ � þ Ct½ � : ð6Þ

3.2.1. Thermal equilibriumThe constant K′a can be obtained by spectrophotometric measure-

ment (Pina, 1998; Pina, Benedito, Melo, Parola, & Bernardo, 1996). Theabsorption spectra of the thermally equilibrated solutions of cy3glcand of the inclusion complex cy3glc–β-CD are displayed in Fig. 3.This figure shows the pH dependence of the absorption spectra ofthe equilibrated solutions together with the representation of thefitting to Eq. (6). The main feature that emerges from these spectrais the progressive decrease, with increasing pH, of the absorptionband centered at 511 nm, which can be attributed to the flavyliumcation (Pina et al., 1996). The data represented in this figure andthe fitting of Eq. (6) allow us to calculate pK′a=2.8 for cy3glc andpK′a=2.1 for the inclusion complex cy3glc–β-CD. The difference

observed on the pK′a calculated for cy3glc and for cy3glc–β-CD prob-ably indicates that in the presence of β-CD the colorless forms aremore favorable.

Three main processes with very different timescales can generallybe observed for anthocyanins: (i) proton transfer is the fastest andoccurs in the micro-second timescale; (ii) hydration (includingtautomerization) takes place on a timescale from seconds to severalminutes (if pH is not much higher than the pKa); (iii) isomerizationoccurs on a timescale of seconds or days, depending on the cis–transisomerization barrier.

3.2.2. Proton transferThe first process that occurs upon a pH jump from pHb1 to higher

values is the transfer of a proton. This reaction is very fast and forthis reason, conventional spectrophotometry including stopped-flowanalysis is useless (Formosinho & Arnaut, 1993; Pina, 1998). The acid-ity constant Ka, can be calculated through an alternative method asfollows: pH jump from pHb1 to a value where pH≫pKa+2. At thispH the flavylium cation is totally converted into a quinoidal baseand the hydration process is generally slow enough to be followedby a conventional spectrophotometer. Under these conditions, the

Fig. 4. A1, Spectral variations versus time of cy3glc (6.0×10−5 M) upon a direct pH jump from pH 1.0 to pH 5.3; A2, variation of the absorbance at 530 nm versus time, pKa=3.7determined by means of Eq. (7); B1, spectral variations versus time of the inclusion complex cy3glc:β-CD (6.0×10−5 M:3.0×10−3 M) upon a direct pH jump from pH 1.0 to pH 5.3;B2, variation of the absorbance at 530 nm versus time, pKa=3.2 determined by means of Eq. (7).

Fig. 5. A1, Variation of the absorbance at 530 nm versus time of cy3glc (6.0×10−5 M) upon a direct pH jump from pH 1.0 to pH 5.3; A2, variation of the absorbance at 350 nm versustime of cy3glc (6.0×10−5 M) at the same conditions, showing two kinetic processes; B1, variation of the absorbance at 530 nm versus time of the inclusion complex cy3glc–β-CD(6.0×10−5 M:3.0×10−3 M) upon a direct pH jump from pH 1.0 to pH 5.3; B2, variation of the absorbance at 350 nm versus time of the inclusion complex cy3glc–β-CD(6.0×10−5 M:3.0×10−3 M) at the same conditions, showing two kinetic processes.

752 A. Fernandes et al. / Food Research International 51 (2013) 748–755

Fig. 6. A, Fitting of the faster observed rate constants as a function of pH for the cy3glc so-lution (6.0×10−5 M), pKh=2.9, determined by means of Eq. (8); B, fitting of the fasterobserved rate constants as a function of pH for the inclusion complex cy3glc–β-CD(6.0×10−5 M:3.0×10−3 M), pKh=2.5, determined by means of Eq. (8).

753A. Fernandes et al. / Food Research International 51 (2013) 748–755

absorbance taken immediately after the pH jump, A0, corresponds tothe maximum concentration of this species. At the end of the kineticprocess (final equilibrium) the absorbance of the quinoidal base indi-cates its final concentration A1. The ratio A1/A0 is equal to the molarfraction distribution of the quinoidal base at the equilibrium (whenβ′=1, there is practically no flavylium cation) (Eq. (7)). For cy3glc,Ka/K′a=0.12, giving pKa=3.7. For the molecular complex cy3glc–β-CD,Ka/K′a=0.07, giving pKa=3.1 (Fig. 4).

A½ �C0

¼ Ka

K ′aβ′; β′ ¼ A½ � þ B½ � þ Cc½ � þ Ct½ �

C0ð7Þ

3.2.3. Hydration and isomerization reactionsDirect pH jumps followed in a longer scale of time (using a com-

mon spectrophotometer) put in evidence the existence of a muchslower process that was assigned to the Cc–Ct isomerization. ThreepH jumps to pH 4.5, 5.3 and 6.1 were performed and followed for sev-eral hours to study the isomerization process. An example of one ofthose experiments is displayed in Fig. 5 at pH 5.3, in which the de-crease of absorbance at 530 nm together with the increase of absor-bance at 350 nm could be an indication of the isomerization process,which led to the formation of colorless forms (Ct). Taking the ab-sorbance values of the consecutive spectra at λmax=530 nm and350 nm and plotting them against time, one can obtain two graphsfrom Fig. 5(A and B). The fitting of the biexponential curve at350 nm allowed the determination of the observed rate constants(kobs) of the hydration (fastest process — absorbance decrease) andthe isomerization process (slowest process — absorbance increase)(Pina, 1998). It is possible to observe that at pH 5.3, the isomerizationoccurs with a mean lifetime of 0.96 h for cy3glc and of 0.55 h for theinclusion complex cy3glc–β-CD (1/kobs). Inclusion of this anthocyaninon the β-CD cavity clearly induced acceleration on the isomerizationobserved rate constant (approximately 1.7 times faster).

According to previous works (Houbiers, Lima, Maçanita, & Santos,1998; Nave et al., 2010), for anthocyanins (oenin) the value of Kt isvery small and considering only the fastest process (hydration reac-tion), it is possible to determine the pseudo equilibrium constantK^

a. Bearing this in mind, Eq. (5) could be simplified to give Eq. (8),which allows the calculation of Kh. Nevertheless, this only constitutesan approximation in the measurement of Kh.

K a ¼ Ka þ Kh ð8Þ

As shown in Fig. 6, the fitting of the faster observed rate constant(kobs) for cy3glc and for the inclusion complex cy3glc–β-CD, as a func-tion of pH, allowed the determination of Kh. Hydration of cy3glc isspeeded up by the presence of β-CD and hydration appears about1.5 times as fast inside the β-CD cavity. From the data presented inFig. 6, for cy3glc, Kh=1.8×10−3 M−1 and for the inclusion complexcy3glc–β-CD, Kh=2.7×10−3 M−1. The equilibrium and the kineticconstants determined by UV–vis absorption, for cy3glc and for theinclusion complex cy3glc–β-CD are presented in Table 1. The isomer-ization observed rate constant and hydration constant (Kh) obtainedfor the molecular inclusion complex cy3glc–β-CD are in agreementto previous works. Dangles, Stoeckel, et al. (1992) and Dangles,Wigand, et al. (1992) reported the inclusion-induced acceleration forthe hydration and isomerization reaction of a synthetic, non-naturalanthocyanin, the 3,4′-dimethoxy-7-hydroxyflavylium. In this work,the evaluation of hydration reaction was made through one pH jumpto pH 3.5 and the observed rate constantwas about twice as fast insidethe β-CD cavity. Similarly, this macrocycle catalyzed the chalcone cis–trans isomerization, and the isomerization observed rate constant hasbeen estimated to be about ten times faster inside the β-CD cavity(Dangles, Stoeckel, et al., 1992).

Similar studies were performed with β-CD's smaller homologue,α-CD. The interaction of this macrocycle with cy3glc did not alter sig-nificantly the equilibrium and kinetic rate constants of this anthocyaninwhich clearly indicates that the inclusion of cy3glc on theα-CD's cavityis not so favored and thus the impact of this macrocycle on theequilibrium and kinetic constants of cy3glc is not significant (Table 1).

4. Conclusion

The need for safe natural food colors has prompted researchers allover the world to look for new sources of food colors. In order to in-crease the knowledge about the impact of cyclodextrins on the equilib-rium and kinetic constants of the network of chemical reactions takingplace in cyanidin-3-O-glucoside a spectroscopic study was performed.

Among several cyclodextrins, β-CD deserves special attention be-cause it is the most readily available cyclodextrin and is preferredcompared to its homologues especially because of its wide use in thefood industry and also for its strong interaction with anthocyanins.The spectral properties of cy3glc were changed by the molecularinclusion with β-CD. The anti-copigmentation phenomenon observedshows dependence on the pH and on the β-CD concentration. Thecomparison of the values of the equilibrium and kinetic equilibriumconstants between cy3glc and the inclusion complex cy3glc–β-CDshowed that the inclusion causes significant modifications in the

Table 1Thermodynamic and kinetic constants determined by UV–vis absorption for the transformations of cy3glc (6.0×10−5 M) and of inclusion complexes cy3glc–β-CD(6.0×10−5 M:3.0×10−3 M) and cy3glc–α-CD (6.0×10−5 M:3.0×10−3 M). Thermodynamic and kinetic constants described for malvidin-3-O-glucoside in the literature(Mv3glc) (Brouillard et al., 1978; Houbiers et al., 1998; Maçanita et al., 2002; Nave et al., 2010).

Determined values Described values

Cy3glc Cy3glc–β-CD Cy3glc–α-CD Mv3glc

K′a (M−1) 1.7×10−3

(2.8±0.1)8.4×10−3

(2.1±0.1)2.3×10−3

(2.6±0.1)5.35×10−3 (2.3)2.74×10−3 (2.54)

Ka (M−1) 2.0×10−4 (3.7) 5.9×10−4 (3.2) 2.9×10−4 (3.5) 1.99×10−4 (3.7)1.6×10−4 (3.8)

Kh (M−1) 1.8×10−3

(2.9±0.1)2.7×10−3

(2.5±0.1)2.0×10−3

(2.7±0.1)1.6×10−3 (2.8)4.3×10−3 (2.4)

kh (s−1) 0.087 0.050 0.081 0.08k−h (M−1 s−1) 47.5 18.0 39.6 19

754 A. Fernandes et al. / Food Research International 51 (2013) 748–755

kinetic and thermodynamic properties. Inclusion of cy3glc in the β-CDcavity induced acceleration on the isomerization observed rate con-stant (kobs) and on the hydration constant (Kh), which is in agreementwith the anti-copigmentation phenomenon observed. The resultsobtained contribute for the elucidation of the impact of cyclodextrinson the color and on the equilibrium and kinetic constants of cyanidin-3-O-glucoside, increasing the knowledge regarding the stabilization ofanthocyanins. Despite the color diminishmolecular inclusion with cy-clodextrins may lead to the anthocyanin stabilization, which couldpromote its use as food colorant.

Acknowledgments

The authors thank Professor Fernando Pina for the insights duringthis research. This research was supported by POPH/FSE and FCT(Fundação para a Ciência e Tecnologia) fromPortugal by two PhD schol-arships (SFRH/BD/65350/2009 and SFRH/BD/68736/2010). This workwas also supported by one project grant (PTDC/QUI-QUI/117996/2010).

References

Azevedo, J., Fernandes, I., Faria, A., Oliveira, J., Fernandes, A., de Freitas, V., et al. (2010).Antioxidant properties of anthocyanidins, anthocyanidin-3-glucosides and respec-tive portisins. Food Chemistry, 119(2), 518–523.

Bourvellec, C. (2003). Association entre les procyanidols et les polymères pariétaux depommes: Quantification et conséquences. : Universite de Rennes I.

Bridle, P., & Timberlake, C. F. (1997). Anthocyanins as natural food colours—Selectedaspects. Food Chemistry, 58(1–2), 103–109.

Brouillard, R. (1982). In P. Markakis (Ed.), Anthocyanins as food colors. New York: Aca-demic Press.

Brouillard, R., Chassaing, S., Isorez, G., Kueny-Stotz, M., & Figueiredo, P. (2010). The vis-ible flavonoids or anthocyanins: From research to applications. In C. Santos-Buelga,M. Escribano-Bailon, & U. Lattanzio (Eds.), Recent advances in polyphenols research,2. (pp. 1–23): John Wiley & Sons Ltd.

Brouillard, R., & Delaporte, B. (1977). Chemistry of anthocyanin pigments. 2. Kinetic andthermodynamic study of proton transfer, hydration, and tautomeric reactions ofmalvidin 3-glucoside. Journal of the American Chemical Society, 99(26), 8461–8468.

Brouillard, R., Delaporte, B., & Dubois, J. E. (1978). Chemistry of anthocyanin pigments. 3.Relaxation amplitudes in pH-jump experiments. Journal of the American ChemicalSociety, 100(19), 6202–6205.

Brouillard, R., & Dubois, J. E. (1977). Mechanism of the structural transformations of antho-cyanins in acidic media. Journal of the American Chemical Society, 99(5), 1359–1364.

Burin, V. M., Rossa, P. N., Ferreira-Lima, N. E., Hillmann, M. C. R., & Boirdignon-Luiz, M. T.(2011). Anthocyanins: Optimisation of extraction from Cabernet Sauvignon grapes,microcapsulation and stability in soft drink. International Journal of Food Science andTechnology, 46(1), 186–193.

Cai, Y. Z., & Corke, H. (2000). Production and properties of spray-dried Amaranthusbetacyanin pigments. Journal of Food Science, 65(7), 1248–1252.

Cai, Y., Gaffney, S. H., Lilley, T. H., Magnolato, D., Martin, R., Spencer, C. M., et al. (1990).Polyphenol interactions. Part 4. Model studies with caffeine and cyclodextrins.Journal of the Chemical Society, Perkin Transactions, 2(12), 2197–2209.

Chandra, A., Nair, M. G., & Iezzoni, A. F. (1993). Isolation and stabilization of anthocyaninsfrom tart cherries (Prunus cerasus L.). Journal of Agricultural and Food Chemistry, 41(7),1062–1065.

Clifford, M. N. (2000). Anthocyanins — Nature, occurrence and dietary burden. Journalof the Science of Food and Agriculture, 80(7), 1063–1072.

Dangles, O., & Brouillard, R. (1992). Inclusion of natural pigments into cyclodextrins.Natural pigments in competitive spectrophotometric studies of host–guest systems.

In A. R. Hedges (Ed.), Minutes of the Sixth International Symposium on Cyclodextrins(pp. 117–122). Paris: Editions de Santé.

Dangles, O., Stoeckel, C., Wigand, M. C., & Brouillard, R. (1992). Two very distinct typesof anthocyanin complexation: Copigmentation and inclusion. Tetrahedron Letters,33(36), 5227–5230.

Dangles, O., Wigand, M. C., & Brouillard, R. (1992). Anthocyanin anti-copigment effect.Phytochemistry, 31(11), 3811–3812.

Divakar, S. (1993). A structural study of the naringin-β-cyclodextrin complex. Journalof Inclusion Phenomena and Macrocyclic Chemistry, 15(4), 305–316.

Divakar, S., & Maheswaran, M. M. (1997). Structural studies on inclusion compounds ofβ-cyclodextrin with some substituted phenols. Journal of Inclusion Phenomena andMacrocyclic Chemistry, 27(2), 113–126.

Ersus, S., & Yurdagel, U. (2007). Microencapsulation of anthocyanin pigments of blackcarrot (Daucuscarota L.) by spray drier. Journal of Food Engineering, 80(3), 805–812.

Formosinho, S. J., & Arnaut, L. G. (1993). Excited-state proton transfer reactions II. In-tramolecular reactions. Journal of Photochemistry and Photobiology A: Chemistry,75(1), 21–48.

Francis, F. J., & Markakis, P. C. (1989). Food colorants: Anthocyanins. Critical Reviews inFood Science and Nutrition, 28(4), 273–314.

Hedges, A. R., Shieh, W. J., & Sikorski, C. T. (1995). Use of cyclodextrins for encapsula-tion in the use and treatment of food products. Encapsulation and controlled releaseof food ingredients, 590. (pp. 60–71): American Chemical Society.

Hendy, G. M., & Breslin, C. B. (2011). A spectrophotometric and NMR study on the for-mation of an inclusion complex between dopamine and a sulfonated cyclodextrinhost. Journal of Electroanalytical Chemistry, 661(1), 179–185.

Houbiers, C., Lima, J. C., Maçanita, A. L., & Santos, H. (1998). Color stabilization ofmalvidin 3-glucoside: Self-aggregation of the flavylium cation and copigmentationwith the Z-chalcone form. The Journal of Physical Chemistry. B, 102(18), 3578–3585.

Idham, Z., Muhamad, I. I., & Sarmidi, M. R. (2012). Degradation kinetics and color sta-bility of spray-dried encapsulated anthocyanins from Hibiscus sabdariffa L. Journalof Food Process Engineering, 35(4), 522–542.

Ishizu, T., Kintsu, K., & Yamamoto, H. (1999). NMR study of the solution structures of theinclusion complexes of β-cyclodextrin with (+)-catechin and (−)-epicatechin. TheJournal of Physical Chemistry. B, 103(42), 8992–8997.

Ishizuka, Y., Fujiwara, M., Kanazawa, K., Nemoto, T., Fujita, K., & Nakanishi, H. (2002).Three-dimensional structure of the inclusion complex between phloridzin andβ-cyclodextrin. Carbohydrate Research, 337(19), 1737–1743.

Küster, W. F., & Thiel, A. (1982). Tabelle per le Analisi Chimiche e Chimico-Fisiche (12 ed.).Milano: Hoepli.

Lewis, C. E., Walker, J. R. L., & Lancaster, J. E. (1995). Effect of polysaccharides on the col-our of anthocyanins. Food Chemistry, 54(3), 315–319.

Maçanita, A. L., Moreira, P. F., Lima, J. C., Quina, F. H., Yihwa, C., & Vautier-Giongo, C.(2002). Proton transfer in anthocyanins and related flavylium salts. Determinationof ground-state rate constants with nanosecond laser flash photolysis. The Journalof Physical Chemistry. A, 106(7), 1248–1255.

Mourtzinos, I., Makris, D. P., Yannakopoulou, K., Kalogeropoulos, N., Michali, I., &Karathanos, V. T. (2008). Thermal stability of anthocyanin extract of Hibiscussabdariffa L. in the presence of β-cyclodextrin. Journal of Agricultural and FoodChemistry, 56(21), 10303–10310.

Nagatomo, S. (1985). Cyclodextrins: Expanding the development of their functions andapplications. Chemical Economy and Engineering Review, 17, 28–34.

Nave, F., Petrov, V., Pina, F., Teixeira, N.,Mateus, N., & de Freitas, V. (2010). Thermodynamicand kinetic properties of a redwine pigment: Catechin-(4,8)-malvidin-3-O-glucoside.The Journal of Physical Chemistry. B, 114(42), 13487–13496.

Pegg, R. B., & Shahidi, F. (2007). Encapsulation, stabilization, and controlled release offood ingredients and bioactives. In L. Taylor & Frances Group (Eds.), Handbook offood preservation (2 ed.).

Pina, F. (1998). Thermodynamics and kinetics of flavylium salts. Malvidin revisited.Journal of the Chemical Society, Faraday Transactions, 94(15), 2109–2116.

Pina, F., Benedito, L., Melo, M. J., Parola, A. J., & Bernardo, A. (1996). Photochemistry of3,4′-dimethoxy-7-hydroxyflavylium chloride. Photochromism and excited-stateproton transfer. Journal of the Chemical Society, Faraday Transactions, 92(10),1693–1699.

Pina, F., Melo, M. J., Laia, C. A. T., Parola, A. J., & Lima, J. C. (2012). Chemistry and appli-cations of flavylium compounds: A handful of colours. Chemical Society Reviews,41(2), 869–908.

755A. Fernandes et al. / Food Research International 51 (2013) 748–755

Selim, K. A., Khalil, K. E., Abdel-Bary, M. S., & Abdel-Azeim, N. A. (2008). Extraction, en-capsulation and utilization of red pigments from roselle (Hibiscus sabdariffa L.) asnatural food colourants. Alexandria Journal of Food Science and Technology, 7–20.

Tamura, H., Takada, M., Yamagami, A., & Shiomi, K. (1998). The color stability andantioxidative activity of an anthocyanin and γ-cyclodextrin complex. Functionalfoods for disease prevention, 701. (pp. 157–171): American Chemical Society.

Tønnesen, H. H., Másson, M., & Loftsson, T. (2002). Studies of curcumin and curcuminoids.XXVII. Cyclodextrin complexation: Solubility, chemical and photochemical stability.International Journal of Pharmaceutics, 244(1–2), 127–135.

Xiong, S., Melton, L. D., Easteal, A. J., & Siew, D. (2006). Stability and antioxidant activityof black currant anthocyanins in solution and encapsulated in glucan gel. Journal ofAgricultural and Food Chemistry, 54(17), 6201–6208.

Yamada, T., Komiya, T., & Akaki, M. (1980). Formation of an inclusion complex ofanthocyanin with cyclodextrin. Agricultural and Biological Chemistry, 44(6),1411–1413.