caffeic acids on colour intensification and anthocyanin stability

Food Chemistry 228 (2017) 526–532

Contents lists available at ScienceDirect

Food Chemistry journal homepage: www.elsevier.com/locate/foodchem

The effects of gallic/ferulic/caffeic acids on colour intensification and anthocyanin stability Bing-Jun Qian a,1, Jian-Hua Liu b,1, Shu-Juan Zhao a, Jian-Xiong Cai a, Pu Jing a,⇑ a Research Center for Food Safety and Nutrition, Key Lab of Urban Agriculture (South), Bor S. Luh Food Safety Research Center, School of Agriculture & Biology, Shanghai Jiao Tong University, Shanghai 200240, China b College of Resources and Environment Engineering, Yibin University, Yibin, Sichuan 644000, China

a r t i c l e

i n f o

Article history: Received 26 August 2016 Received in revised form 25 December 2016 Accepted 25 January 2017 Available online 27 January 2017 Chemical compounds studied in this article: Peonidin 3-sophoroside-5-glucoside (PubChem CID: 44256845) Cyanidin 3-sophoroside-5-glucoside (PubChem CID: 44256732) Cyanidin 3-(600 -ferulylsophoroside)-5-gluco side (PubChem CID: 44256782) Cyanidin 3-(600 -caffeylsophoroside)-5-gluco side (PubChem CID: 44256781) Cyanidin 3-(600 -caffeyl-6000 -ferulylsophoro side)-5-glucoside (PubChem CID: 44256788)

a b s t r a c t The mechanism by which copigments stabilize colour, by protecting anthocyanin chromophores from nucleophilic attack, seems well accepted. This study was to determine effects of gallic/ferulic/caffeic acids on colour intensification and anthocyanin stability. Molecular dynamics simulations were applied to explore molecular interactions. Phenolic acids intensified the colour by 19%  27%. Colour fading during heating followed first-order reactions with half-lives of 3.66, 9.64, 3.50, and 3.39 h, whereas anthocyanin degradation, determined by the pH differential method (or HPLC-PDA), followed second-order reactions with half-lives of 3.29 (3.40), 3.43 (3.39), 2.29 (0.39), and 2.72 (0.32) h alone or with gallic/ferulic/caffeic acids, respectively, suggesting that anthocyanin degradation was faster than the colour fading. The strongest protection of gallic acids might be attributed to the shortest distance (4.37 Å) of its aromatic ring to the anthocyanin (AC) panel. Hyperchromic effects induced by phenolic acids were pronounced and they obscured the accelerated anthocyanin degradation due to self-association interruption. Ó 2017 Elsevier Ltd. All rights reserved.

Keywords: Purple sweet potato Peonidin-3-sophoroside-5-glucoside Copigmentation Phenolic acid p-p Stacking

1. Introduction The main challenge in the use of anthocyanins as natural food colorants is their relatively low stability. Anthocyanins owe their colour to the high resonance of a fully conjugated 10-electron A-C ring system, with some contribution by the B ring as well. If the resonance structure is disrupted, the colour is lost (Brouillard, 1982). The stability of anthocyanins is determined mainly by their chemical structure and also depends on a combination of environmental factors, including temperature, light, presence of other phenolic compounds, metal ions, ascorbic acid, ⇑ Corresponding author. 1

E-mail address: [email protected] (P. Jing). These authors made equivalent contributions to the manuscript.

http://dx.doi.org/10.1016/j.foodchem.2017.01.120 0308-8146/Ó 2017 Elsevier Ltd. All rights reserved.

and oxygen (Delgado-Vargas & Paredes-López, 2003; Shahidi & Naczk, 2004). Copigmentation represents an important factor in anthocyanin chromophore stabilization (Boulton, 2001; Mazza & Brouillard, 1990; Trouillas et al., 2016). The anthocyanin copigmentation, in model solutions and in wines, has been evaluated and interpreted by the visible kmax or differential colorimetry, both in model solutions and in wines (Chung, Rojanasasithara, Mutilangi, & McClements, 2017; García Marino, Escudero-Gilete, Heredia, Escribano-Bailón, & Rivas-Gonzalo, 2013; Gordillo et al., 2015; Navruz, Türkyılmaz, & Özkan, 2016). Anthocyanins are more stable in an aquatic system via intermolecular copigmentation with other compounds (Mazza & Brouillard, 1990). The compounds or copigments that can interact with anthocyanins include simple phenols, such as catechin (Dangles & Brouillard, 1992; Mazza & Brouillard,

B.-J. Qian et al. / Food Chemistry 228 (2017) 526–532

1987), flavanol (McDougall, Gordon, Brennan, & Stewart, 2005), chlorogenic acid, caffeic acid, or rutin (Davies & Mazza, 1993) and polyphenols, such as tannins (Remy, Fulcrand, Labarbe, Cheynier, & Moutounet, 2000; Salas, Fulcrand, Meudec, & Cheynier, 2003; Thorngate & Singleton, 1994). The instant p-p overlap, dipole-dipole interactions, and hydrogen bonding are main molecular interactions of anthocyanin complexation with copigments (Dangles & Brouillard, 1992). The stacking of the copigment molecule on the planar polarizable nuclei of the coloured anthocyanin forms (flavylium ion or quinonoid forms) prevents the nucleophilic attack of water at position 2 of the pyrylium nucleus, leading to colourless hemiketal and chalcone forms (Davies & Mazza, 1993; Mazza & Brouillard, 1990). The anthocyanin self-association also protects their molecular stability (Boulton, 2001). However, it is not clear whether the anthocyanin association with phenolic acids would be better than their selfassociation to protect pigments from the water nucleophilic attack. The aims of this study were to investigate effects of gallic/ferulic/caffeic acids on anthocyanins from purple sweet potatoes and their colour, and to explore the mechanism underling the interaction between anthocyanins and phenolic acids. The dynamic changes in colour and anthocyanins were monitored during heating and the intermolecular interactions of peonidin 3-O-(2-O-bd-glucopyranosyl-b-d-glucopyranoide)-5-O-b-d-glucopyranoside with gallic/ferulic/caffeic acids were explored, using molecular dynamics simulation.

527

series of test tubes was filled with 10 ml of each solution, closed with screw caps and covered with aluminium foil. The tubes were then immersed in a water bath at 95 °C. Twelve tubes (three replicates for each solution) were prepared and sampled at regular time intervals (0, 0.5, 1, 2, 5, 10, and 15 h) and rapidly cooled to room temperature for analysis. The absorption spectra were recorded, using an L5S UV–visible spectrophotometer (Shanghai Analytical Instrument, China), by scanning the visible range from 400 to 800 nm. A possible hypochromic effect was detected as an increase in the absorbance value at kmax and a bathochromic shift as a shift in the wavelength (nm) of kmax. Samples were stored in a refrigerator and analyzed for monomeric anthocyanins within one day. 2.4. Total monomeric anthocyanins The total monomeric anthocyanins in the PSP extracts was determined by the pH differential method (Giusti & Wrolstad, 2001). An L5S UV–visible spectrophotometer (Shanghai Analytical Instrument, China) was used to read the absorbance at 520 and 700 nm. Total monomeric anthocyanins were expressed as cyanidin-3-glucoside, using a molecular weight of 449.2 and a molar absorptivity of 26,900 l  cm1 mg1. Cuvettes with a 1 cm path length were used. Measurements were performed in triplicate. 2.5. UPLC-ESI-HRMS analysis of PSP anthocyanins

2. Materials and methods 2.1. Materials and reagents Purple sweet potatoes (PSP; Ipomoea batatas) were purchased at a local market in Shanghai, China. Ferulic, gallic, and caffeic acids were purchased from Shanghai Pureone Biotechnology (Shanghai, China). All other chemicals were purchased from Sigma-Aldrich (Shanghai, China). 2.2. Extraction and purification of anthocyanins The extraction and purification of PSP anthocyanins were performed as described by Jing et al. (2014). Fresh purple sweet potatoes (5 kg) were ground into powder, which was immersed in 10 L of methanol containing 0.01% HCl and extracted for 2 h. The extracts were applied to a 600 cm  50 cm Amberlite XAD-7HP column (Huideyi, Beijing, China) and washed with 0.01% HCl in water to remove water-soluble compounds. The anthocyanin fraction was eluted with 0.01% HCl in ethanol. The anthocyanin fraction was applied to a 100 cm  2.5 cm open column packed with Sephadex LH-20 and separated by 50% aqueous ethanol containing 0.01% HCl. The anthocyanin fractions were further purified in an Agilent preparative HPLC system equipped with a semipreparative ZORBAX Eclipse XDB-C18 column. The anthocyanin peak fractions were collected and evaporated to dryness under a nitrogen-blow evaporator. The purity of the anthocyanins obtained from purple sweet potato was >30% (w/w). 2.3. Thermal stability of copigmentation The pigment stability of PSP anthocyanins dissolved in pH 3.2 buffer (0.06 M sodium acetate and 0.02 M phosphoric acid) was evaluated, using an accelerated stability test at 95 °C. The four treated solutions containing 22.5 mg monomeric anthocyanins/l of PSP anthocyanin extract alone or with additional gallic acid (851 mg), ferulic acid (971 mg), or caffeic acid (901 mg) were prepared to satisfy the 1:100 M ratio of cyanidin-3-glucoside and phenolic acid. A

The anthocyanins were separated by an ACQUITY UPLC system (Waters, MA, USA) equipped with an Acquity BEH C18 column (1.7 lm, 100 mm  2.1 mm i.d., Waters, MA, USA). The analysis was conducted at a flow rate of 0.4 ml/min. Mobile phase A was 0.1% formic acid in water (v/v), and mobile phase B was acetonitrile. The following gradient programme was used for the mobile phases: 0–15 min, 5%–20% B; 15–20 min, 20–40% B; 20–22 min, 40–85%. Spectral information was collected over the wavelength range of 200–800 nm. The LC-HRMS system consisted of a Waters Micromass Q-TOF Premier mass spectrometer equipped with an electrospray interface (Waters Corporation, Milford, MA) at the Instrumental Analysis Center of Shanghai Jiao Tong University. The positive ionization mode was used for the detection of anthocyanins. The applied electrospray/ion optics parameters were set as follows: capillary voltage, 3.0 kV; sampling cone, 35 V; collision energy, 4 eV; source temperature, 115 °C; desolvation temperature, 250 °C; desolvation gas flow, 300 L/h. The scan time was 0.5 s in an m/z range of 50– 1500 au. 2.6. Quantitative analysis of individual PSP anthocyanins The identification of monomeric anthocyanins in PSPs during thermal treatment was performed, using a LC-2030C HPLC system (Shimadzu, Japan) equipped with a binary solvent delivery system, an online vacuum degasser, a diode array detection (PDA) system, an automatic sampler, a thermostatically controlled column compartment and a Shimadzu LabSolutions workstation. Separation was achieved by reverse phase elution on an InertSustain C18 column (5 lm, 250 mm  4.6 mm i.d.). The analysis was conducted at a flow rate of 1 ml/min. The injection volume was 20 ll. The mobile phase consisted of 1% formic acid in water (v/v; eluent A) and of acetonitrile (eluent B). The gradient programme was as follows: 0–5 min, 10% B; 5–20 min, 10–15% B; 20–30 min, 15–20% B; 30–40 min, 20–25%B; 40–45 min, 25–40% B; 45–50 min, 40–60% B. Spectral information was collected over the wavelength range of 200–800 nm, and the detection wavelength was set at 530 nm, whereas the total absorbance of anthocyanin peaks was recorded.

528

B.-J. Qian et al. / Food Chemistry 228 (2017) 526–532

Samples of each thermal treatment were analyzed in triplicate. The total absorbance retention of anthocyanin on thermal treatment was analyzed using linear regression analysis. 2.7. Molecular dynamics simulation The initial geometries of peonidin 3-O-(2-O-b-d-glucopyrano syl-b-d-glucopyranoide)-5-O-b-d-glucopyranoside and the phenolic acids (caffeic, ferulic, and gallic acids) were built into the SYBYL X-1.3 software on a Windows operating system. Energy minimization of each structure was conducted by the Powell method, using the Tripos force field, where a convergence criterion of 0.005 kal/ (mol Å) was used as the termination of the Powell conjugate gradient algorithm, and the maximum iterations were set to 1000 steps. The partial atomic charges were calculated, using the GasteigerHücke method. Other parameters were the defaults according to our previous publication (Jing et al., 2012). Molecular docking was performed by the Surflex-dock model, using the lowestenergy conformations. The peonidin-3-sophoroside-5-glucoside molecular model was set as the receptor and the phenolic acid molecular models as the ligands. The water molecules and bound ligand were removed from the receptor prior to docking analysis. All parameters were default. CScore calculation was off. After successful docking, all the putative docking models were analyzed and selected, based on the MolDock score and hydrogen bonding interactions, and the best docked complexes were further analyzed by molecular dynamics simulations. Molecular dynamics (MD) simulations in periodic boundary conditions were performed, using the Tripos force field. The algorithm for solvation was set as silverware. One counterion (Cl-) was applied to neutralize the positive charge of each studied system. Each complex was minimized in two stages. In the first stage, the complexes were fixed, and only the water and position of the counterion were minimized. In the second stage, the full system was minimized. Afterwards, the whole system was equilibrated by 100 ps MD, followed by a 10 ns MD simulation, using the NPT ensemble at 368 K with 50 fs of integration time and a cut-off of 10 Å for long-range interactions. A 31 Å  31 Å  31 Å box around the water was applied to simulate the periodic boundary conditions. The final modelled system contained 1393 peonidin-3-sophoroside-5-glucoside, phenolic acid, and water molecules. The stabilized structure served to visualize the capability to form hydrogen bonds (H bonding ability), as well as to determine the sites and number of direct and intermediate hydrogen bonds stabilizing it (Kunsagi-Mate, Szabo, Nikfardjam, & Kollar, 2006). 2.8. Statistical analysis The accelerated stability test was carried out according to a completely randomized design (CRD) with three replicates for each treatment. Statistics were analyzed, using SPSS for Windows (version rel. 10.05, 1999, SPSS Inc., Chicago, IL). Values of p < 0.05 were considered significant. Reaction rates for colour fading/ anthocyanin degradation were obtained from linear regression analysis (95% confidence interval). 3. Results 3.1. Structure of anthocyanins in purple sweet potatoes A qualitative analysis of anthocyanin composition was conducted, using UPLC-ESI-MSn. The identification was performed by comparing the ion fragments obtained from MSn and literature reports (Kim et al., 2012; Lee, Park, Choi, & Jung, 2013; Li et al., 2013; Truong et al., 2010) . In total, 18 anthocyanins were

tentatively identified in PSPs (Table 1). The basic structures were identified as peonidin-3-sophoroside-5-glucoside (787 [M]+, 463 [M + glucoside]+, and 301 [peonidin]+) or cyanidin-3-sophoro side-5-glucoside (773 [M]+, 449 [M + glucoside]+, and 287 [cyanidin]+. The fragmentation patterns of individual peaks are clear and characteristic, as shown in Fig. 1: all glycosylated substitutes at position C3 and/or C5 of the flavylium ring were cleaved, which is consistent with previous reports on the fragmentation patterns for triglycosidated anthocyanins (Lee et al., 2013; Terahara et al., 2007; Tian, Konczak, & Schwartz, 2005;). With respect to the acyl group, peaks 1–8 and peaks 9–18 were characterized as mono-acylated and di-acylated with caffeic, ferulic and/ or p-hydroxybenzoic acids, respectively, as described in Table 1, whereas peaks 12–14 were primary anthocyanins. 3.2. Hyperchromic effect of PSP anthocyanin extracts combined with gallic, ferulic, and caffeic acids Copigmentation experiments of PSP anthocyanins with gallic, ferulic, or caffeic acids were conducted at a 1:100 M ratio, based on our preliminary data. A hyperchromic shift was observed, as shown in Fig. 2, suggesting that the colour of the PSP anthocyanins was enhanced significantly by the tested phenolic acids. Ferulic acid increased the colour intensity of PSP anthocyanins by 26.5%, as shown in Fig. 2, followed by caffeic acid (20.1%) and gallic acid (19.1%). Gallic/ferulic/caffeic acids intensified the colour of PSP anthocyanins by 19  27% at pH = 3.2. In other studies, ferulic acid was found to improve the colour intensities of strawberry, raspberry, cranberry, and lingonberry juices rich in different anthocyanins to varying degrees (Rein & Heinonen, 2004), especially strawberry juices. No bathochromic or hypsochromic shift was observable for the copigmentation of PSP anthocyanins with ferulic, caffeic or gallic acids at pH = 3.5, which is consistent with reports that a bathochromic shift was not in evidence for the copigmentation of malvin by 7-O-sulfoquercetin at pH = 3.5 but was obvious at pH = 1 (Alluis & Dangles, 2001). 3.3. Thermal stability of colour and monomeric anthocyanins of purple sweet potato extracts An accelerated thermal-stability assay was established to evaluate the copigmentation effect of gallic, ferulic, and caffeic acids. Colour intensity and monomeric anthocyanins were monitored during heat treatment. Linear regression analysis was applied to determine the rate of colour fading or anthocyanin degradation in each copigmentation. The equations and R-square values are presented in Table 2. All equations had an R-square value of 0.919 or higher, showing that most of the variability was explained by the model. Fig. 2A shows that the colour of the PSP anthocyanin solution at pH 3.2 decreased sharply at 95 °C, and only 63.7% and 31.3% remained after 1 h and 10 h at 95 °C, respectively. Approximately 66.0%, 71.8% and 66.5% of the colour of PSP anthocyanins copigmented with gallic, ferulic or caffeic acids remained after 1 h at 95 °C, respectively, showing greater stability than that of anthocyanins alone, as illustrated in Fig. 2B–D. However, after 10 h of heat treatment at 95 °C, the colour intensities of the copigmentation solutions with gallic, ferulic and caffeic acids were reduced to 23.8%, 52.7% and 24.7%, respectively, suggesting that gallic acid acted to protect the anthocyanin solution from colour fading, but ferulic and caffeic acids appeared to promote the colour degradation under longer heat treatment. The colour faded, following first-order kinetics during the heat treatment, as described in Table 2. The half-lives of the colour intensities were 3.66, 9.64,

529

B.-J. Qian et al. / Food Chemistry 228 (2017) 526–532 Table 1 Qualitative analyses of anthocyanins in purple sweet potatoes based on literature sources.a Peaks

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 a b

Retention time (min)

Anthocyanins

HPLC

UPLC

17.47 18.56 22.29 23.17 24.55 28.50 29.90 30.96 31.27 32.41 33.29 34.24 34.79 36.59 37.73 38.24 39.00 40.45

4.81 5.27 6.36 6.76 7.24 8.50 8.76 9.54 9.54 9.54 10.72 11.10 11.27 12.36 12.63 13.09 13.36 14.38

m/z (amu) +

Cy 3-p-hydroxybenzoyl soph-5-glu Cy-3-(600 -caffeoyl soph)-5-glub Pn-3-p-hydroxybenzoyl soph-5-glc Pn-3-caffeoyl soph-5-glu Cy-3-(600 -feruloyl soph)-5-glub Cy-3-caffeoyl soph-5-glu isomer Pn-3-(600 -feruloyl soph)-5-glub Pn-3-caffeoyl soph-5-glu isomer Cy-3-(600 ,6000 -dicaffeoyl soph)-5-glub Cy-3-caffeoyl-p-hydroxybenzoyl soph-5-glu Cy-3-(600 -caffeoyl-6000 -feruloyl soph)-5-glub Pn-3-(600 ,6000 -dicaffeoyl soph)-5-glub Pn-3-caffeoyl p-hydroxybenzoyl-soph-5-glu Pn-3-(600 caffeoyl-6000 -feruloyl soph)-5-glub Cy-3-(600 ,6000 -diferuloyl soph)-5-glub Pn-3-caffeoyl feruloyl soph-5-glu isomer Pn-3-feruloyl p-hydroxybenzoyl-soph-5-glu Pn-3-diferuloyl soph-5-glu

t1/2 (hour) (alone or with acids)

[M]

Fragments

893.238 935.247 907.256 949.266 949.271 935.255 963.280 949.269 1097.290 1055.278 1111.309 1111.306 1069.292 1125.321 1125.316 1125.319 1083.311 1139.338

731.184, 773.189, 745.207, 787.201, 787.211, 773.185, 801.223, 787.212, 935.228, 893.219, 949.250, 949.245, 907.237, 963.262, 963.257, 963.263, 921.250, 977.277,

449.131, 449.109, 463.126, 463.121, 449.109, 449.103, 463.119, 463.123, 449.108, 449.108, 449.107, 463.125, 463.124, 463.118, 449.106, 463.124, 463.124, 463.124,

287.058 287.052 301.071 301.072 287.058 287.046 301.070 301.069 287.054 287.054 287.054 301.060 301.060 301.058 287.054 301.069 301.072 301.072

Alone

Gallic

Ferulic

Caffeic

2.91 1.83 3.75 3.70 2.57 2.89 2.26 2.50 3.06 4.48 3.41 3.23 1.17 7.47 2.89 6.01 8.18 4.02

3.10 2.58 3.70 3.84 2.54 3.08 3.38 2.80 3.15 5.69 4.20 4.22 4.66 8.92 2.94 5.18 8.91 5.12

0.54 0.54 0.69 0.46 0.51 0.38 0.56 0.35 0.35 1.43 0.23 0.52 0.31 0.84 0.34 0.38 0.65 0.56

0.75 0.68 0.61 0.37 0.30 0.43 0.74 0.04 0.14 1.48 0.40 0.26 0.20 0.46 0.16 0.34 0.64 0.90

Cy, cyanidin; Pn, peonidin; soph, sophoroside; glu, glucoside. structural determination based on the literature (Kim et al., 2012; Truong et al., 2010).

and those copigmented with gallic, ferulic, or caffeic acids were 3.29 (3.40), 3.43 (3.39), 2.29 (0.39), and 2.72 (0.32) h, respectively. Generally, monomeric anthocyanin degradation was faster than the colour fading according to their halflives. 3.4. Molecular dynamic simulation of copigmentation of peonidin 3-O-(2-O-b-d-glucopyranosyl-b-d-glucopyranoide)-5-O-b-d-glucopyranoside and phenolic acids

R2 , R3

Fig. 1. Chemical structures of anthocyanins in purple sweet potatoes. Arrows indicate the MS1 fragmentation scheme (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.).

3.50, and 3.39 h for the PSP anthocyanin extracts alone and copigmentation with gallic, ferulic, or caffeic acids, respectively. The remaining anthocyanins during heat treatment at 95 °C were monitored via the pH differential method and HPLC-PDA. The HPLC profiles during heat treatment are shown in Data in brief. The PSP anthocyanins showed similar profiles alone and copigmented with gallic acid over heating time, whereas PSP anthocyanins that copigmented with ferulic or caffeic acids almost disappeared after 2 h. The monomeric anthocyanin content decreased during heating treatment, and the change followed second-order kinetics under both evaluation methods. The halflives of monomeric anthocyanins determined via the pH differential method (or HPLC-PDA method) for PSP anthocyanins alone

Molecular dynamics simulations were performed to understand the differences in copigmentation behaviour of PSP anthocyanins and phenolic acids (ferulic, caffeic, or gallic acid). The peonidin 3-O-(2-O-b-d-glucopyranosyl-b-d-glucopyranoide)-5-O-b-d-glucopyranoside was applied as compound 1 in the simulation study since peonidin derivatives were predominant anthocyanins in tested PSPs. The closest geometries to the average structures after molecular dynamics simulations at 368 K are shown in Data in brief. Table 2 shows the average minimal distance from the aromatic ring of ferulic/caffeic/gallic acids to the AC panel or chromophore of peonidin 3-O-(2-O-b-d-glucopyranosyl-b-d-glucopyranoide)-5-O-b-d-glucopyranoside at 368 K. The minimal distance from the aromatic ring of phenolic acids to the AC panel of compound 1 followed the order gallic acid (4.37 Å) < ferulic acid (4.59 Å) < caffeic acid (5.02 Å). The closest distance was from gallic acid to the AC panel of peonidin 3-O-(2-O-b-d-glucopyranosyl-b-dglucopyranoide)-5-O-b-d-glucopyranoside. Ferulic and caffeic acids were 4.59 Å and 5.02 Å from the AC panel of compound 1, respectively. 4. Discussion The results showed that monomeric anthocyanin degradation was faster than the colour fading. Anthocyanin self-stacking was assumed to have higher thermal resistance than copigmentation with phenolic acids via p–p overlap. Additional phenolic acids might interrupt anthocyanin self-stacking or self-association, leading to obvious anthocyanin degradation. Structural transformations of anthocyanins occur in acidic aqueous media at pH 1–6. Four forms coexist in equilibrium: the quinoidal base A, the flavylium cation AH+, the carbinol pseudobase B, and the chalcone C (Raymond Brouillard & Delaporte, 1977). The self-association of anthocyanins occurs when the

530

B.-J. Qian et al. / Food Chemistry 228 (2017) 526–532

Fig. 2. Changes in the absorbance spectra of PSP anthocyanins alone or with phenolic acids at 95 °C. Letters a, b, c, d, e and f indicate the measurements at intervals of 0, 0.5, 1, 2, 5 and 10 h, respectively.

Table 2 Thermal degradation kinetics of colour intensity and total monomeric anthocyanins of purple sweet potatoes at 95 °C. R2

t1/2 (hour)

Reaction order

Colour intensity via visible spectrum PSPA Ln (A/A0) = 0.0987  t  0.3312 PSPA + Gallic Ln (A/A0) = 0.0434  t  0.2748 PSPA + Ferulic Ln (A/A0) = 0.1148  t  0.2914 PSPA + Caffeic Ln (A/A0) = 0.1207  t  0.2845

0.990 0.995 0.997 0.997

3.66 9.64 3.50 3.39

First First First First

Monomeric anthocyanins via pH differential methodb PSPA C0/C = 0.288  t + 1.054 PSPA + Gallic C0/C = 0.255  t + 1.129 PSPA + Ferulic C0/C = 0.418  t + 1.045 PSPA + Caffeic C0/C = 0.319  t + 1.134

0.919 0.925 0.964 0.965

3.29 3.43 2.29 2.72

Second Second Second Second

Monomeric anthocyanins via HPLC-DADc PSPA A0/A = 0.228  t + 1.223 PSPA + Gallic A0/A = 0.302  t + 0.975 PSPA + Ferulic A0/A = 2.457  t + 1.037 PSPA + Caffeic A0/A = 2.386  t + 1.231

0.955 0.920 0.997 0.995

3.40 3.39 0.39 0.32

Second Second Second Second

Equations a

a Colour intensity was determined by the absorbance at 535 nm in a UV–vis spectrometer. A0 is the absorbance at time 0 min and A is the absorbance after t minutes of heating at 95 °C. b Total amount of monomeric anthocyanins was determined via the pH differential method. C0 is the total monomeric anthocyanins (mg/l) at time 0 min and C is the total monomeric anthocyanins (mg/l) after t minutes of heating at 95 °C. c Total amount of monomeric anthocyanins was determined via the HPLC-DAD method. A0 is the total peak area of anthocyanin peaks at time 0 min and A is the peak area of anthocyanins after t minutes of heating at 95 °C.

concentration increases to some point. Additional phenolic acids might interrupt anthocyanin self-stacking or self-association and cause obvious anthocyanin degradation. Scheme 1 illustrates the aqueous anthocyanin solution before and after adding gallic acid. The self-association between the flavylium cation (AH+), quinoidal base (A) and carbinol pseudobase (B) forms and the copigmentation of peonidin-3-sophoroside-5-glucoside and gallic acid is shown in Scheme 1. The hyperchromism and anthocyanin stability caused by the addition of phenolic acid are considered as follows: 4.1. Hyperchromism Hyperchromic shifts (19  26%) were observed when PSP anthocyanins were mixed with ferulic/caffeic/gallic acids in a

1:100 M ratio. Two approaches have been applied to explain the mechanisms. First, the p–p stacking of planar aromatic molecules facilitates the formation of excimer (excited dimer) states, in which excitation is shared between two closely spaced molecules, and the hyperchromic effect originates from the higher delocalization of the excitonic states in the denatured DNA with respect to the double-stranded conformation (D’Abramo, Castellazzi, Orozco, & Amadei, 2013). Additional phenols in the anthocyanin solution might increase the p-electron delocalization, consequently producing hyperchromic effects. Ferulic and caffeic acids, unlike gallic acid, are cinnamic acids, where the unsaturated carboxylic acid could provide additional p-electron delocalization and promote the hyperchromic effect. Additionally, the methoyl group on the ferulic aromatic ring contributes electrons to the phenol ring,

531

B.-J. Qian et al. / Food Chemistry 228 (2017) 526–532

Scheme 1. Systems of self-stacking anthocyanins. (A) aqueous anthocyanin system; (B) aqueous anthocyanin system after adding gallic acids; (C) prototypical interaction of anthocyanins with gallic acids.

which could explain the enhancement of colour intensity by phenolic acids in the order ferulic > caffeic > gallic acids in this study. An alternative explanation could be the disturbance of the relative orientation of neighbouring chromophores in the presence of phenolic acids, which could increase the absorptivity of the anthocyanins. Similarly, the hypochromism or hyperchromism of nucleic acids, a typical example of orientation effects, showed absorbance decreases in the trend free nucleotides > single-stranded DNA > double-stranded DNA (Crespo-Hernandez, Cohen, & Kohler, 2005). 4.2. Stabilizing anthocyanins? Similar structural features make the intermolecular interactions between the flavylium cation and quinonoidal base forms of the anthocyanin or with copigments much easier (Asen, Stewart, & Norris, 1972). Clearly, phenolic acids rather than the same amount of additional anthocyanins interacted with anthocyanins less efficiently, based on the similarity of structural features. Consequently, the molecular stacking of the aromatic units of phenolic acids with the chromophore (AC panel) of anthocyanins, instead of perfect anthocyanin self-stacking, reduced their ability to prevent hydration of the anthocyanidin nucleus (Kondo, Ueda, Isobe, & Goto, 1998). This mechanism might be why anthocyanin degradation proceeded faster with phenolic acids, except for gallic acid, than alone in this study. The results of the computer simulation show that the distance of the aromatic surface of gallic acid to the AC panel of peonidin-3-sophoroside-5-glucoside was 4.37 Å, as shown in Table 3, which is much shorter than the other two phenolic acids, suggesting that gallic acid might protect the anthocyanidin nucleus from hydration better than did the ferulic and caffeic acids. 5. Conclusion Both experimental and theoretical studies have allowed a molecular understanding of the anthocyanin copigmentation phenomenon over recent decades. In this study, gallic/ferulic/caffeic acids showed a hypochromic effect. However, these copigments did not protect anthocyanins from thermal degradation. Additionally, the anthocyanins degraded faster than the colour fading. These results allowed us to reconsider the well-accepted mechanism in which molecular interactions between anthocyanins

Table 3 Average minimal distances between approximately planar surfaces of peonidin 3-O(2-O-b-D-glucopyranosyl-b-D-glucopyranoide)-5-O-b-D-glucopyranoside (compound 1) and phenolic acids.

Ferulic acid

Caffeic acid

Gallic acid

Phenolic acids

Average minimal distance to compound 1 (Å) AC panel

B-ring

Gallic acid Ferulic acid Caffeic acid

4.37 4.59 5.02

4.92 4.87 4.89

and copigment molecules protect the chromophore from nucleophilic attack, thereby stabilizing the colour. We propose that the anthocyanin self-stacking might offer better thermal resistance than the copigmentation with phenolic acids via p–p overlap, based on the similarity of structural features; additional phenolic acids might interrupt anthocyanin self-stacking or selfassociation and accelerate the pigment degradation.

Acknowledgements This study was funded by the National Nature Science Foundation of China (Grant No. 31371756) and Science and technology plan projects in Sichuan province (No. 2015JY0112).

References Alluis, B., & Dangles, O. (2001). Quercetin (=2-(3,4-dihydroxyphenyl)-3,5,7trihydroxy-4h-1-benzopyran-4-one) glycosides and sulfates: Chemical synthesis, complexation, and antioxidant properties. Helvetica Chimica Acta, 84(5), 1133–1156. Asen, S., Stewart, R. N., & Norris, K. H. (1972). Co-pigmentation of anthocyanins in plant tissues and its effect on color. Phytochemistry, 11(3), 1139–1144.

532

B.-J. Qian et al. / Food Chemistry 228 (2017) 526–532

Boulton, R. (2001). The copigmentation of anthocyanins and its role in the color of red wine: A critical review. American Journal of Enology and Viticulture, 52(2), 67–87. Brouillard, R., & Delaporte, B. (1977). Chemistry of anthocyanin pigments. 2. Kinetic and thermodynamic study of proton transfer, hydration, and tautomeric reactions of malvidin 3-glucoside. Journal of the American Chemical Society, 99 (26), 8461–8468. Brouillard, R. (1982). Chemical structure of anthocyanins. In P. Markakis (Ed.), Anthocyanins as food colours (pp. 1–40). New York: Academic Press. Chung, C., Rojanasasithara, T., Mutilangi, W., & McClements, D. J. (2017). Stability improvement of natural food colors: Impact of amino acid and peptide addition on anthocyanin stability in model beverages. Food Chemistry, 218, 277–284. Crespo-Hernandez, C. E., Cohen, B., & Kohler, B. (2005). Base stacking controls excited-state dynamics in AT DNA. Nature, 436(7054), 1141–1144. D’Abramo, M., Castellazzi, C. L., Orozco, M., & Amadei, A. (2013). On the Nature of DNA Hyperchromic Effect. J. Phys. Chem. B, 117(29), 8697–8704. Dangles, O., & Brouillard, R. (1992). Polyphenol interactions. The copigmentation case: Thermodynamic data from temperature variation and relaxation kinetics. Medium effect. Canadian Journal of Chemistry, 70(8), 2174–2189. Davies, A. J., & Mazza, G. (1993). Copigmentation of simple and acylated anthocyanins with colorless phenolic compounds. Journal of Agriculture and Food Chemistry, 41(5), 716–720. Delgado-Vargas, F., & Paredes-López, O. (2003). Anthocyanins and betalains. In Natural Colorants for Food and Nutraceutical Uses (pp. 167–219). Boca Raton: CRC Press. García Marino, M., Escudero-Gilete, M. L., Heredia, F. J., Escribano-Bailón, M. T., & Rivas-Gonzalo, J. C. (2013). Color-copigmentation study by tristimulus colorimetry (CIELAB) in red wines obtained from Tempranillo and Graciano varieties. Food Research International, 51, 123. Giusti, M. M., & Wrolstad, R. E. (2001). Characterization and measurement of anthocyanins by UV-visible spectroscopy. In R. E. Wrolstad, T. E. Acree, E. A. Decker, M. H. Penner, D. S. Reid, S. J. Schwartz, C. F. Shoemaker, D. Smith, & P. Sporns (Eds.), Current Protocals in Food Analytical Chemistry, 1 ed., Vol. 1 (pp. F1.2.1-F1.2.13). NY: John Wiley & Sons Inc. Gordillo, B., Rodríguez-Pulido, F. J., González-Miret, M. L., Quijada-Morín, N., RivasGonzalo, J. C., García-Estévez, I., et al. (2015). Application of differential colorimetry to evaluate anthocyanin–flavonol–flavanol ternary copigmentation interactions in model solutions. Journal of Agricultural and Food Chemistry, 63(35), 7645–7653. Jing, P., Zhao, S. J., Jian, W. J., Qian, B. J., Dong, Y., & Pang, J. (2012). Quantitative studies on structure-DPPH center dot scavenging activity relationships of food phenolic acids. Molecules, 17(11), 12910–12924. Jing, P., Zhao, S. J., Ruan, S. Y., Sui, Z. Q., Chen, L. H., Jiang, L. L., et al. (2014). Quantitative studies on structure-ORAC relationships of anthocyanins from eggplant and radish using 3D-QSAR. Food Chemistry, 145, 365–371. Kim, H. W., Kim, J. B., Cho, S. M., Chung, M. N., Lee, Y. M., Chu, S. M., et al. (2012). Anthocyanin changes in the Korean purple-fleshed sweet potato, Shinzami, as affected by steaming and baking. Food Chemistry, 130(4), 966–972. Kondo, T., Ueda, M., Isobe, M., & Goto, T. (1998). A new molecular mechanism of blue color development with protocyanin, a supramolecular pigment from cornflower Centaurea cyanus. Tetrahedron Letters, 39(45), 8307–8310.

Kunsagi-Mate, S., Szabo, K., Nikfardjam, M. P., & Kollar, L. (2006). Determination of the thermodynamic parameters of the complex formation between malvidin-3O-glucoside and polyphenols. Copigmentation effect in red wines. Journal of Biochemical and Biophysical Methods, 69(1–2), 113–119. Lee, M. J., Park, J. S., Choi, D. S., & Jung, M. Y. (2013). Characterization and Quantitation of Anthocyanins in Purple-Fleshed Sweet Potatoes Cultivated in Korea by HPLC-DAD and HPLC-ESI-QTOF-MS/MS. Journal of Agriculture and Food Chemistry, 61(12), 3148–3158. Li, J., Li, X. D., Zhang, Y., Zheng, Z. D., Qu, Z. Y., Liu, M., et al. (2013). Identification and thermal stability of purple-fleshed sweet potato anthocyanins in aqueous solutions with various pH values and fruit juices. Food Chemistry, 136(3–4), 1429–1434. Mazza, G., & Brouillard, R. (1987). Recent developments in the stabilization of anthocyanins in food products. Food Chemistry, 25(3), 207–225. Mazza, G., & Brouillard, R. (1990). The mechanism of copigmentation of anthocyanins in aqueous solutions. Phytochemistry, 29(4), 1097–1102. McDougall, G. J., Gordon, S., Brennan, R., & Stewart, D. (2005). Anthocyanin-flavanol condensation products from black currant (Ribes nigrum L.). Journal of Agriculture and Food Chemistry, 53(20), 7878–7885. Navruz, A., Türkyılmaz, M., & Özkan, M. (2016). Colour stabilities of sour cherry juice concentrates enhanced with gallic acid and various plant extracts during storage. Food Chemistry, 197(Part A), 150–160. Rein, M. J., & Heinonen, M. (2004). Stability and enhancement of berry juice color. Journal of Agriculture and Food Chemistry, 52(10), 3106–3114. Remy, S., Fulcrand, H., Labarbe, B., Cheynier, V., & Moutounet, M. (2000). First confirmation in red wine of products resulting from direct anthocyanin-tannin reactions. Journal of the Science of Food and Agriculture, 80(6), 745–751. Salas, E., Fulcrand, H., Meudec, E., & Cheynier, V. (2003). Reactions of Anthocyanins and Tannins in Model Solutions. Journal of Agriculture and Food Chemistry, 51 (27), 7951–7961. Shahidi, F., & Naczk, M. (2004). Phenolics in food and nutraceuticals. Washington, D. C.: CRC Press. Terahara, N., Oki, T., Matsui, T., Fukui, K., Sugita, K., Matsumoto, K., et al. (2007). Simultaneous determination of major anthocyanins in purple sweet potato. Journal of Japan Society of Nutrition and Food Sciences, 54(1), 33–38. Thorngate, J. H., & Singleton, V. L. (1994). Reactions of monomeric and polymeric flavan-3-ols with monomeric pigment in model wine solutions. American Journal of Enology and Viticulture, 45(3), 349–352. Tian, Q. G., Konczak, I., & Schwartz, S. J. (2005). Probing anthocyanin profiles in purple sweet potato cell line (Ipomoea batatas L. cv. Ayamurasaki) by highperformance liquid chromatography and Electrospray ionization tandem mass spectrometry. Journal of Agriculture and Food Chemistry, 53(16), 6503–6509. Trouillas, P., Sancho-García, J. C., De Freitas, V., Gierschner, J., Otyepka, M., & Dangles, O. (2016). Stabilizing and modulating color by copigmentation: Insights from theory and experiment. Chemical Reviews. Truong, V. D., Deighton, N., Thompson, R. T., McFeeters, R. F., Dean, L. O., Pecota, K. V., et al. (2010). Characterization of anthocyanins and anthocyanidins in purplefleshed sweetpotatoes by HPLC-DAD/ESI-MS/MS. Journal of Agriculture and Food Chemistry, 58(1), 404–410.