Kinetics, physicochemical properties, and antioxidant activities of Angelica keiskei processed under four drying conditions

LWT - Food Science and Technology 98 (2018) 349–357

Contents lists available at ScienceDirect

LWT - Food Science and Technology journal homepage: www.elsevier.com/locate/lwt

Kinetics, physicochemical properties, and antioxidant activities of Angelica keiskei processed under four drying conditions

T

Chengcheng Zhang, Daqun Liu∗∗, Haiyan Gao∗ Food Science Institute, Zhejiang Academy of Agricultural Sciences, Key Laboratory of Post-Harvest Handling of Fruits, Ministry of Agriculture, Key Laboratory of Fruits and Vegetables Postharvest and Processing Technology Research of Zhejiang Province, Key Laboratory of Postharvest Preservation and Processing of Fruits and Vegetables, China National Light Industry, Hangzhou, 310021, China

A R T I C LE I N FO

A B S T R A C T

Keywords: Angelica keiskei Antioxidant capacity Polyphenol components Drying methods

The herb Japanese angelica (Angelica keiskei) has been consumed for centuries as a healthy vegetable. The drying kinetics, microstructure, color, and polyphenol components, as well as antioxidant capacity of stems and leaves of A. keiskei were investigated and compared, after undergoing natural drying (ND), convective drying (CD), freeze-drying (FD), and vacuum oven drying (VOD). The freeze-dried products showed the lowest color differences compared with their fresh counterpart, and also provided the highest retention of the main polyphenol components (i.e., quercetin, luteolin, and chlorogenic acid) and antioxidant activity (DPPH and FRAP), followed by ND. The CD and VOD processes at 60 °C in which heat was applied caused the additional loss of color, total phenolic content (TPC), total flavonoid content (TFC), and antioxidant activity. In addition, the leaves exhibited stronger antioxidant activity and contained higher phenolic content than stems; for example, the quercetin content in fresh stems and leaves was 89.82 vs 638.20 mg/kg dry weight, respectively. Our results implied that FD could be a superior drying technique for A. keiskei, however, high cost of FD procedures may limit its application. Natural drying thus could be an alternative method when finance becomes the main concern.

1. Introduction Angelica keiskei Koidz. (Umbelliferae), commonly known under the Japanese name of Ashitaba or Japanese angelica, is a cold-hardy perennial plant that was originally grown on the Izu Islands and Miura Peninsula of Japan (Nakamura et al., 2012). The aerial parts have been consumed for centuries as a healthy vegetable. Additionally, it is a medicinally important herb used as a diuretic, laxative, and stimulant as well as a tonic for restoring vitality (Kim et al., 2014). There are also several beneficial effects of A. keiskei that have been reported in animal models and clinical trials, including anti-hypertension (Shimizu et al., 1999), anti-tumourigenesis (Kang, Park, Kim, & Kim, 2004), antithrombosis (Son, Park, Yu, Lee, & Park, 2014), vasodilation (Matsuura, Kimura, Nakata, Baba, & Okuda, 2001), and hepatoprotection (Noh, Ahn, Yun, Cho, & Paek, 2015). Angelica keiskei is an important source of health-promoting constituents, such as coumarins, flavanones, chalcones, and phenolic acids (Aldini et al., 2011; Kim et al., 2014). In particular, phenolic acids and flavanones are very important constituents in A. keiskei and have been proven to possess strong free radical scavenging and antioxidant capacities (Kim et al., 2014; Li et al.,

∗

2009). Angelica keiskei is widely cultivated in Asia due to a significant consumer demand for functional foods. Having an initial moisture content of 85–95 g/100 g, fresh A. keiskei is easily perishable because it is highly susceptible to mechanical damage and microbial spoilage under environmental conditions. The drying process reduces the moisture content to a safe level and is the most commonly used method to inhibit microbial growth and delay deteriorative biochemical reactions. Therefore, drying preserves A. keiskei and prolongs its shelf-life. The dry powder is the fundamental ingredient used in A. keiskei products such as snacks, flour, and cosmetics (Bao, 2014; Heo, Kang, Jung, & Shin, 2013; Wu, Yang, & Wang, 2012). Another advantage is easier transport due to reduced weight and volume, which can potentially expand markets for A. keiskei-based products. However, the drying process is complex and significantly influences the bioactive potential of the final ingredient. Previous studies have shown that the drying process could potentiate or reduce the amounts of compounds with potential biological activity (Yábar, Pedreschi, Chirinos, & Campos, 2011). Conventionally, natural drying (ND) and thermal treatments are relatively more common methods used to preserve and extend the shelf-

Corresponding author. Food Science Institute, Zhejiang Academy of Agricultural Sciences, Hangzhou 310021, China. Corresponding author. Food Science Institute, Zhejiang Academy of Agricultural Sciences, Hangzhou 310021, China. E-mail addresses: [email protected] (D. Liu), [email protected] (H. Gao).

∗∗

https://doi.org/10.1016/j.lwt.2018.08.054 Received 5 April 2018; Received in revised form 18 July 2018; Accepted 27 August 2018 Available online 31 August 2018 0023-6438/ © 2018 Elsevier Ltd. All rights reserved.

LWT - Food Science and Technology 98 (2018) 349–357

C. Zhang et al.

Table 1 Parameters of different drying models applied to dried A. keiskei. Model

Modified Page

Model equation

MR = A·exp(-k·tB)

Drying method

CD VOD ND

Henderson Pabis

MR = A·exp (-k·t)

CD VOD ND

Logarithmic

MR = A·exp(-k·t)+B

CD VOD ND

Wang&Singh

MR = A + B·t + C·t2

CD VOD ND

Detected part

stems leaves stems leaves stems leaves stems leaves stems leaves stems leaves stems leaves stems leaves stems leaves stems leaves stems leaves stems leaves

Parameters

Statistics

A

k

B

0.9557 0.9975 0.9241 1.0047 0.9756 0.9854 1.0202 0.9404 1.1136 1.0774 1.0597 0.9900 1.0715 0.9218 4.4988 1.1364 1.1081 0.9827 0.9487 0.7632 1.0231 1.0145 0.9985 0.7974

0.1324 0.6873 0.0115 0.1605 0.0126 0.1143 0.2222 0.4656 0.1659 0.2990 0.0548 0.1212 0.1819 0.5561 0.0213 0.2463 0.0471 0.1249

1.2581 0.6789 2.2496 1.3907 1.4487 1.0247

−0.0770 0.0414 −3.4780 −0.0813 −0.0633 0.0102 −0.1404 −0.1693 −0.0968 −0.2045 −0.0353 −0.0409

C

RMSE

R2

0.0052 0.0091 0.0010 0.0103 3.0216E-4 4.7667E-4

0.0160 0.0036 0.0156 0.0036 0.0061 0.0099 0.0273 0.0261 0.1399 0.0276 0.0330 0.0099 0.0016 0.0166 0.0215 0.0171 0.0243 0.0093 0.0154 0.1191 0.0212 0.0118 0.0179 0.1371

0.9882 0.9956 0.9886 0.9969 0.9962 0.9910 0.9812 0.9702 0.9064 0.9787 0.9807 0.9918 0.9879 0.9793 0.9843 0.9855 0.9845 0.9915 0.9886 0.8518 0.9845 0.9900 0.9886 0.8754

Note:ND: natural drying; CD: convective drying; VOD: vacuum oven drying; FD: Freeze-drying.

life of various vegetables due to their low-cost and simplicity. However, ND has some disadvantages, mainly that it is time-consuming with inconsistent quality standards (Soysal & Oztekin, 2001). In this sense, convective drying (CD) is always an alternative for drying vegetables in the food industry. It should be noted that CD usually causes degradation of thermo sensitive bioactive components, flavors, and colors (An et al., 2016). According to Jiang et al. (2017), convective drying of okra may have negative effects on bioactive compounds, such as flavone and phenolic compounds, compared with their fresh counterparts, thus reducing their antioxidant activity. Freeze-drying (FD) can effectively retain the original properties, including bioactivity of phytochemical compounds, flavor, and shape; however, this method is a time-intensive and costly method (Karam, Petit, Zimmer, Djantou, & Scher, 2016). The effects of drying methods on polyphenolics and antioxidant activities of A. keiskei have not yet been systematically investigated. Therefore, the study of drying technologies in order to determine the most optimal conditions to maximize presentation of color, as well as both active constituents and antioxidant activity, is crucial to produce dried A. keiskei powder for use as a food coloring, as an ingredient for snacks, or a functional ingredient to be included in other products. Additionally, previous research found that the content of polyphenolics in A. keiskei significantly varied across the different parts of plant (Kim et al., 2014; Luo, Ya-Li, Zhu, & Shen, 2017). Accordingly, the aim of this work was to explore the changes in the drying process and their effect on the drying kinetics, structural properties, color characteristics, polyphenolic composition, and antioxidant capacity of A. keiskei between stems and leaves.

width).

2. Materials and methods

2.3. Modeling of drying kinetics

2.1. Raw materials

The drying curves were drawn based on mass losses of A. keiskei samples. The moisture ratio (MR) was calculated using Equation (1):

Fresh A. keiskei was sourced from a farm at Rizhao, Shandong, PR China, in July 2017. The moisture content of the fresh stems and leaves was 11.51 ± 0.43 and 5.78 ± 0.15 kg kg-1 db (dry basis), respectively. Before drying, each plant was separated into leaves and stems before cutting into small pieces (20 ± 0.5 mm, length; 20 ± 0.5 mm,

MR = (Mt – Me)/(M0– Me),

2.2. Drying procedure The initial mass of approximately 400 g of fresh leaves or stems was evenly distributed and subjected to four different drying methods: (i) ND – fresh leaves and stems were dried by placing them in a shaded location between 18 and 25 °C. (ii) CD –A. keiskei samples were subjected to 60 °C in an electric thermostatic drying oven (DHG-9070A; Shanghai Jinghong Experiment Instrument Co., Shanghai, China). (iii) Vacuum oven drying (VOD) –the leaves and stems of A. keiskei were placed in a vacuum oven (DZF-6050; Shanghai Jinghong Experiment Instrument Co., Shanghai, China) at 60 °C. The vacuum was set to 25 Pa and maintained by controlling the vacuum pump and air inlet. (iv) FD –samples were frozen at −80 °C for 12 h, and then were quickly placed into a freeze dryer (SCIENTZ-18N; Ningbo Science Biotechnology Co., Ltd., Ningbo, China) at a vacuum pressure of 47 Pa. During FD, the temperature in the cold trap was −50 °C, while the temperature of the drying chamber reached −25 °C. All of the samples were dried until the moisture content was below 0.1 kg kg-1 db to meet microbial safety requirements. The moisture content was determined by placing samples in a hot air oven at 105 °C until a constant weight was attained (China GB5009.3–2016).

(1)

Where Mt is the moisture content of the sample (kg water kg dry basis1) at time t, M0 is the initial moisture content, and Me is the equilibrium moisture content of the sample. In Equation (1), the value of the 350

LWT - Food Science and Technology 98 (2018) 349–357

C. Zhang et al.

supernatant was removed under nitrogen gas and reconstituted in HPLC mobile phase A for chromatographic analysis. Phenolic acids and flavonoids were analyzed on a ZORBAX SB-C18 column (4.6 × 250 mm, 5 μm) as part of a Waters 2695 high-performance liquid chromatograph equipped with a2489 UV/Visible detector (Waters Corporation; Milford, USA) as previously described (Seong, Hwang, & Chung, 2016). The concentration of each compound was tentatively quantified (mg/kg DW) by the individual linear regression equation obtained from the standard calibration curve.

equilibrium moisture content (Me) is usually infinitesimal and can be omitted (Dadali, Apar, & Özbek, 2007). To obtain the phenomenological parameters of the drying process, four drying models were fitted to predict the drying behaviour (Table 1). 2.4. Color Leaves and stems of A. keiskei before and after each drying method were chosen for color analysis using a CR-400 lab mini colorimeter (Konica Minolta Co., Ltd., Tokyo, Japan) calibrated with a white standard tile. The L*(lightness/darkness), a*(redness/greenness), and b*(yellowness/blueness) values of dried leaves and stems were obtained. The total color difference (ΔE) was estimated using Equation (2):

Δ E=

(L∗ = L0∗)2 + (a∗ − a0∗)2 + (b∗ − b)2 , L∗

a∗

b∗

Where , and are the color values of dried A. keiskei and and b0∗ are the color values of fresh A. keiskei.

2.9. TFC and TPC The TFC and TPC were analyzed using the method recently described by Jiang et al. (2017) with the same analytical setup. TFC is expressed as mg rutin/100 g dry weight (DW), and TPC is expressed as mg gallic acid/100 g DW.

(2)

L0∗,

a0∗

2.10. Antioxidant capacity measurement The procedures for the 2,2-diphenyl-1-picryl-hydrazyl-hydrate (DPPH) and ferric reducing antioxidant potential (FRAP) antioxidant assays were carried out in accordance with the methods described by Chuyen, Roach, Golding, Parks, and Nguyen (2017). Briefly, 0.15 mL of each extract was allowed to react with 2.85 mL of working solution for each antioxidant assay in a test tube for 30 min in the dark. The antioxidant activity of A. keiskei is expressed as mg trolox equivalent/100 g DW.

2.5. Microstructure Analysis of structural characteristics was carried out using a scanning electron microscope (SU-8010, Hitachi Limited, Japan) at an accelerating voltage of 3 kV. Prior to analysis, the dried samples were coated with a very thin layer of gold under high vacuum to impart conductivity. The magnification was adjusted to 300 × . 2.6. Sample extraction

2.11. Statistical analysis

For sample analysis, 1 g dried powder or 5 g fresh homogenized sample was extracted with 70% ethanol (10 mL × 3) using sonication for 30 min at 40 °C. The extracts were then filtered and dried using a rotary evaporator at 40 °C. After concentration, 70% ethanol solvent was added to adjust the final volume to 10 mL for total flavonoid content (TFC), total phenolic content (TPC), and antioxidant activity determination. For chromatographic analysis, the extracts were filtered through a 0.22-μm syringe filter, and 1 mL was transferred into a bottle prior to injection into the chromatographic systems.

All of the experiments were performed at least in triplicate. The results are expressed as the mean ± standard deviation (SD) for each group. Statistical analyses were performed using SAS 9.1 software. All of the data were analyzed using the ANOVA procedure with significance analysis at the p < 0.05 level. Correlations among data obtained were performed using the standard Pearson's correlation. 3. Results and discussion 3.1. Drying kinetics

2.7. Identification of polyphenol compounds by UPLC-ESI-MS The drying curves obtained from A. keiskei dried by CD, VOD, and ND are presented in Fig. 1. The moisture content obtained during the drying experiments was converted to the moisture ratio (MR) and then fitted to four common drying models for plant material (Table 1). According to the statistical results of the root mean square (RMSE) and correlation coefficient (R2), the exponential model (modified Page, Henderson and Pabis, and Logarithmic) exhibited better performance for A. keiskei samples with higher values of R2 (mean values of 0.9928, 0.9682, and 0.9855, respectively) and lower values of RSME (0.0091, 0.0439, and 0.0151, respectively). Among them, the modified Page model was found as the best fit, and it also has the simplest form and has been widely used with other fruits and vegetables (Szychowski et al., 2018). The information concerning the influence of different drying methods on the drying time, residual moisture content, and water activity of A. keiskei is displayed in Table 2. The drying time for reducing the moisture content of A. keiskei to less than 0.1 kg kg-1 db was found to vary widely among the four drying methods. As expected, drying A. keiskei in the shade (ND) took the longest time (64–76 h), followed by FD (40–48 h); however, the drying time needed to remove 90% of the water from A. keiskei was 13–16 h and 12–13 h for CD and VOD at 60 °C, respectively. At the beginning of a drying process, the moisture content in A. keiskei leaves and stems rapidly decreased with increased drying time under various drying conditions as high initial moisture content (Fig. 1), but the drying rate decreased later when the moisture content

Identification of the polyphenol compounds in A. keiskei was performed using a Waters ACQUITY UPLC coupled with a Waters SYNAPT G2 quadrupole time-of-flight mass spectrometer system (Milford, MA, USA). Chromatographic separation was carried out using an ACQUITY UPLC HSS C18 column (2.1 mm × 100 mm × 1.8 μm; Waters, Milford, MA, USA) at a flow rate of 0.2 mL/min; 0.1% formic acid in deionized water and 0.1% formic acid in acetonitrile (high-performance liquid chromatography (HPLC) grade) were chosen as solvents A and B, respectively. A gradient was generated under the following conditions: 0–3 min, 5% B; 3–10 min, 5–30% B; 10–20 min, 30–90% B; 20–30 min, 90–99% B; and 30–32 min, 99–5% B. The ESI source parameters were as follows: positive and negative ion modes were run separately with a cone voltage of 40 V and capillary voltage 2.8 kV, and the source temperature was 105 °C. Nitrogen was used as the desolvation gas at a flow rate of 800 L/h and desolvation temperature at 300 °C. The data were acquired and analyzed with Waters Masslynx software with scans from m/z 100 to 1000. 2.8. Determination of quercetin, luteolin, and chlorogenic acid by HPLC Prior to analysis, flavonoids in the extract were obtained after sugar cleavage by an acid hydrolysis treatment, as described by Zhang, Kou, Fugal, and McLaughlin (2004). After acid boiling, the reactive solution was mixed with 10 mL of ethyl acetate and vortexed, and then, the 351

LWT - Food Science and Technology 98 (2018) 349–357

C. Zhang et al.

Fig. 1. Drying curves of A. keiskei samplesdried by (A) convective drying, (B) vacuum oven drying, and (C)natural ”represents stems, and“ ” represents leaves. drying; “

0.06 to 0.10 kg kg-1 db, and 0.368 to 0.462, respectively (Table 2). The observed ranges of moisture content and water activity are considered microbiologically safe (Chuyen et al., 2017), and could be regarded as suitable for long-term storage of the dried product.

decreased (Fig. 1). Additionally, our results showed that the drying time for stems was longer than that for leaves, agreeing with the lower drying constant k of the modified Page model (0.1324 vs 0.6873, 0.0115 vs 0.1605, and 0.0126 vs 0.1143; Table 2). The final moisture content and water activity of the dried products was in the range from 352

LWT - Food Science and Technology 98 (2018) 349–357

C. Zhang et al.

Table 2 The effect of different drying methods on the drying time, residual moisture content and water activity of A. keiskei. Drying method

Leaves Drying time

Fresh Natural drying Convective drying Vacuum oven drying Freeze-drying

– 64 h 13 h 12 h 40 h

Stems −1

Moisture content (kg kg 5.62 0.10 0.09 0.06 0.07

± ± ± ± ±

a

0.003 0.008b 0.009bc 0.009d 0.010cd

db)

Water activity

Drying time

– 0.460 0.452 0.375 0.402

– 76 h 16 h 13 h 48 h

± ± ± ±

0.07a 0.05a 0.09c 0.03b

Moisture content (kg kg−1 db) 11.05 ± 0.003 0.10 ± 0.007b 0.09 ± 0.005c 0.09 ± 0.004c 0.06 ± 0.004d

a

Water activity – 0.462 0.425 0.422 0.368

± ± ± ±

0.06a 0.04b 0.04b 0.02c

Note: Results are expressed as mean ± standard deviation of three replicates.Moisture content is shown as kg water kg dry solid−1.For each column, means followed by different letters are significantly different (P < 0.05).

Fig. 2. Scanning electron micrographs of A. keiskei samples dried by the four drying processes. FD: Freeze-drying; ND: natural drying; CD: convective drying; VOD: vacuum oven drying.

353

LWT - Food Science and Technology 98 (2018) 349–357

C. Zhang et al.

Table 3 The effect of different drying methods on color parameters of A. keiskeisamples. Drying method

Surface color L*

Leaves

Fresh ND CD VOD FD Fresh ND CD VOD FD

Stems

a*

42.21 38.52 37.18 36.84 45.32 64.19 57.19 57.23 58.26 68.12

± ± ± ± ± ± ± ± ± ±

c

0.15 0.13c 0.44d 0.25d 0.61a 5.59ab 0.19c 0.18c 0.52bc 0.60a

ΔE

b*

−12.60 ± 0.19 −10.97 ± 0.06b −10.78 ± 0.17b −6.67 ± 0.05a −13.59 ± 0.16d −29.95 ± 3.81c −11.84 ± 0.06b −10.83 ± 0.03b −4.85 ± 0.08a −14.32 ± 0.05b

b

14.93 14.21 14.03 15.05 16.80 37.33 19.66 19.98 20.50 20.74

± ± ± ± ± ± ± ± ± ±

b

0.29 0.06c 0.23c 0.15b 0.23a 3.97a 0.05b 0.05b 0.2b 0.10b

– 4.11 ± 0.17c 5.43 ± 0.51b 8.01 ± 0.21a 3.78 ± 0.64c – 26.26 ± 1.70b 26.75 ± 1.65ab 30.80 ± 1.48a 23.15 ± 1.43c

Note: Results are expressed as mean ± standard deviation of three replicates.For each column, means followed by different letters are significantly different (P < 0.05).ND: natural drying; CD: convective drying; VOD: vacuum oven drying; FD: Freeze-drying. Table 4 Phenolic compounds identified in A. keiskei stems and leaves by LC-MS. Peak No.

Rt (min)

MS (m/z)

Phenolic acids and derivatives 1 7.880 377 [M+Na]+ Flavonols 2 9.89 595 [M+H]+ 3 10.110 457 [M+Na]+ 4 10.169 487[M+Na]+ 5 10.272 449[M+H]+ Coumarin 6 17.02 367[M+Na]+ 7 17.42 367[M+Na]+ 8 18.91 351[M+Na]+ Chalcone 9 20.61 339[M+H]+ 10 20.78 393[M+H]+

Measured Mass

MS/MS fragments (m/z)

Tentative identification

References

Detected part

354.1

355, 163

Chlorogenic acid

(Aldini et al., 2011; Li et al., 2009)

Stems, Leaves

594.2 434.4 464.1 448.1

449, 287 435, 303 465, 303 287

Luteolin 7-O-rutinoside Quercetin-3-O-arabinopyranoside Quercetin 3-O-galactoside Luteolin-C-glucoside

(Aldini et al., 2011; Li et al., 2009) (Kim et al., 2005) (Aldini et al., 2011; Li et al., 2009) (Aldini et al., 2011; Li et al., 2009)

Stems Stems, Leaves Stems, Leaves Stems, Leaves

344.4 344.4 328.4

327 327 329, 229

Laserpitin Isolaserpitin Selinidin

(Kim et al., 2014) (Kim et al., 2014) (Kim et al., 2014)

Stems, Leaves Stems, Leaves Stems

338.2 392.2

283 269

4-Hydroxyderricin Xanthoangelol

(Aldini et al., 2011; Li et al., 2009) (Aldini et al., 2011; Li et al., 2009)

Stems Stems

chromavalue) in such circumstances (Mphahlele, Fawole, Makunga, & Opara, 2016). In addition, FD products were found to have the lowest ΔE, followed by the ND samples. ND samples with lower L*and b*values and higher a* value than FD samples showed that the longer drying time during ND may have caused the A. keiskei samples to darken. In contrast, the VOD and CD samples generated a large decrease in the L* value compared to their fresh counterpart (p > 0.05) during the thermal processing at 60 °C. Higher redness and ΔE value were also observed in thermal process (VOD and CD) samples as compared to FD and ND (Table 3). This could indicate that discoloration and browning during thermal drying may be the result of various chemical reactions including chlorophyll destruction or a non-enzymatic Maillard reaction, which occurs between proteins or amino acids and saccharides during heating (Carabasa-Giribet & Ibarz-Ribas, 2000), leading to the formation of brown compounds. As a result, the VOD and CD samples showed higher color differences (higher ΔE values) compared to fresh A. keiskei.

3.2. Effect of the drying process on microstructure, and color parameters The effects of different drying methods on the internal structural properties of A. keiskei samples are shown in Fig. 2. The tissue structure of the FD sample was a porous honeycomb-type, and the hole walls were very thin. The porous honeycomb-type tissue structure was formed by the removal of water that occurred by sublimating directly from the frozen substances to the gas phase with the simultaneous effect of the vacuum without damage from external forces (Chen et al., 2017). VOD samples under a vacuum could also easily produce porous structures (Jiang et al., 2017); however, a thicker hole wall was observed in the VOD samples than that of FD when they were drying at 60 °C under fast vaporization. As shown in Fig. 2, A. keiskei dried by the ND and CD methods showed more dense structures and less porosity, which was in agreement with the observation of severe tissue shrinkage and collapse. The effect of different drying methods on the color values of dried A. keiskei is shown in Table 3. L*, a*, and b* values of dried samples (except for the FD samples) were found to be significantly lower (p < 0.05) than the fresh ones. In addition, significant differences in L* and a* values among different drying methods were also observed, and the b* value was the less affected color variable between drying treatments (Table 3). Freeze-drying was the best option to obtain dried samples with the highest luminosity (L* = 45.32 for leaves, L* = 68.12 for stems on average) and lowest redness (a* = -13.59 for leaves, a* = 14.32 for stems on average). From Table 3, it is clear that the color parameters of the FD samples are similar to those of fresh samples. Those observations can be explained by the fact that the enzymatic browning reaction is relatively weak when exposed to a vacuum and low temperature environment (Jiang et al., 2017), which caused less loss of color degradation of the dried samples. Furthermore, it has a slight capacity for bleaching (lower

3.3. Identification of the chemical composition by UPLC-ESI-MS The phenolic compounds within A. keiskei stems and leaves were analyzed by UPLC-ESI-MS, and the mass spectral data of the identified compounds are listed in Table 4. The compounds were subsequently tentatively identified based on the matching of retention time and exact relative molecular mass, and also by comparing the generated fragments with reference substances or those mentioned in previously reported studies (Aldini et al., 2011; Kim et al., 2014, 2005; Li et al., 2009; Ogawa, Nakashima, & Baba, 2003). MS spectra were studied by ESI in both the positive and negative ionisation modes. It was found that the sensitivity for flavonols and coumarins was higher in the positive ionisation mode as compared to the negative mode. Therefore, phenolic compounds in A. keiskei are preferentially detected using the 354

LWT - Food Science and Technology 98 (2018) 349–357

C. Zhang et al.

positive mode in the current study. Table 4 shows that 10 phenolic compounds were certainly or putatively identified in the stems of A. keiskei, including 4 flavonols, 1 phenolic acid, 3 coumarins, and 2 chalcones. Two types of flavonol derivatives with a fragment at m/z 303 and 287 were found in A. keiskei with characteristics of quercetin and luteolin derivatives, respectively (Table 4). Quercetin and luteolin derivatives are the main flavonols found in A. keiskei, which was in accordance with previous findings (Li et al., 2009). Nevertheless, chlorogenic acid was found to be the most abundant phenolic acid compound in stems. Additionally, three compounds belonging to coumarins, including laserpitin, isolaserpitin, and selinidin, were also identified. Laserpitin and isolaserpitin (with Rt = 17.02 min and 17.42 min, respectively) exhibited a [M+Na]+m/z at 367 and a MS/MS fragment at m/z 327, while selinidin with a [M +Na]+m/z at 351 was identified according to previous studies (Kim et al., 2014). The peaks eluting with an RT of 20.61 min and 20.78 min exhibited an isotopic mass at 339 and 393; thus, they were identified as 4-hydroxyderricin (MW 338.2) and xanthoangelol (MW 392.2), which was in agreement with published research (Ogawa et al., 2003). The two mentioned chalcones identified in A. keiskei are considered to be the major bioactive compounds of the plant because of their physiological functions (Nakamura et al., 2012). The peak intensity chromatograms in positive ionisation mode for the A. keiskei sample showed significant variation for phenolic compounds according to different parts of the plant. Compared to stem samples, a lower abundance of phenolic compounds was detected in the leaves: 3 flavonols, 1 phenolic acid, and 2 coumarins (Table 4). 3.4. Effect of drying methods on the quantities of quercetin, luteolin, and chlorogenic acid Fig. 3. The quercetin, luteolin, and chlorogenic acid content in the (A)leaves and (B)stems of A. keiskei samples dried by the four drying processes. The data are presented as the mean ± SE (n = 3). For each compound (quercetin, luteolin and chlorogenic acid), bars followed by the different letter (a–d) means statistically different (P < 0.05). FD: Freeze-drying; ND: natural drying; CD: ”represents quercetin; “ ” convective drying; VOD: vacuum oven drying. “ represents luteolin; “ ” represents chlorogenic acid.

Flavonoids, such as quercetin and luteolin, found in A. keiskei in relatively high proportion possess strong antioxidant activity (Kim et al., 2005; Li et al., 2009). Chlorogenic acid, the major phenolic acid compound identified by UPLC-ESI-MS analysis from the aerial parts of A. keiskei in the present study, is also reported to exhibit antioxidant capacity in vitro (Rice-Evans, Miller, & Paganga, 1996). As shown in Fig. 3, the quercetin levels in fresh stems and leaves were 89.82 and 638.20 mg/kg DW, respectively; the luteolin levels were found to be 9.18 and 262.53 mg/kg DW, respectively; and chlorogenic acid levels were 26.09 and 80.94 mg/kg DW, respectively. Luo et al. (2017) reported that the levels of phenolic compounds in leaves were significantly greater than those in the stems and root of A. keiskei. Therefore, A. keiskei leaf is a promising potential source of flavonoid compounds. It has high levels of quercetin and luteolin that were found to be greater than those inedible plants such as broccoli (60 mg quercetin/kg DW and 74.4 mg luteolin/kg DW) and mint (48.5 mg quercetin/kg DW) (Miean & Mohamed, 2001). In our study, we found that the quercetin, luteolin, and chlorogenic acid content in dried samples, except for chlorogenic acid in the freezedried sample, was less than that in their fresh counterparts. Quercetin levels of ND, CD, VOD, and FD in A. keiskei leaves were 590.99, 406.95, 384.39, and 553.17 mg/kg DW, respectively; luteolin levels were 200.97, 153.81, 70.82, and 186.64 mg/kg DW, respectively; and chlorogenic acid levels were 76.47, 41.17, 31.20, and 86.79 mg/kg DW, respectively (Fig. 3A). Additionally, quercetin, luteolin, and chlorogenic acid levels of ND, CD, VOD, and FD stems showed a similar overall trend to that of leaves(Fig. 3B). The VOD samples under a low oxygen but heating environment exhibited the lowest concentration of quercetin, luteolin, and chlorogenic acid. This may have resulted from the high-humidity air during VOD, and thus, it may lead to the activation of oxidative enzymes such as polyphenoloxidase and peroxidase (Gumusay, Borazan, Ercal, & Demirkol, 2015; Menon, Hii, Law, Shariff, & Djaeni, 2017). In general, FD and ND products showed higher retention of quercetin, luteolin, and chlorogenic acid compared to CD and VOD. The losses seen in the amounts of quercetin, luteolin, and

chlorogenic acid after using the thermal drying method may have been the result of accelerated degradation due to heat. 3.5. Effect of the drying process on TPC and TFC of A. keiskei extract As shown in Table 5, the TPC and TFC in this plant significantly varied among its leaves and stems. The TFC of leaf extracts (37.47 mg RT/g DW) was almost 4.5-fold higher than that of stems (8.42 mg RT/g DW), and the TPC of leaf extracts (17.65 mg GAE/g DW) was 2.05-fold higher than that of stems (8.60 mg GAE/g DW). The TPC content in A. keiskei was different from that given in a previous report (8.6–9.7 mg GAE/g DW) (Li et al., 2009), where an entire plant, including leaves, stems, and branches, was used. When compared with other health-relevant plants, such as okra (18.54 mg RT/g DW and 12.73 mg GAE/g DW) (Jiang et al., 2017) and ginger (11.97, 13.49), the TFC and TPC of A. keiskei leaf was considerably higher (An et al., 2016). Additionally, we observed comparatively lower values of TFC and TPC in dried samples as compared to their fresh counterparts, except for those samples that underwent the drying process of FD whose TPC was significantly increased compared with fresh ones. According to Asami, Hong, Barrett, and Mitchell (2003), FD could lead to a greater rupturing of the plant cell structure as ice crystals develop within the plant matrix, which would promote the extraction of active ingredients in dried samples, and as a result, increased TPC was detected in FD samples. The highest TFC was observed in plants that underwent the freezedried method, followed by ND and CD; VOD showed the lowest TFC value (Table 5). Similarly, the highest TPC content was also obtained by 355

LWT - Food Science and Technology 98 (2018) 349–357

Note: Results are expressed as Mean ± SE (n = 3). For each line, means followed by different letters are significantly different (P < 0.05).FD: Freeze-drying; ND: natural drying; CD: convective drying; VOD: vacuum oven drying; S: stems; L: leaves.

40.33 15.18 25.92 31.19 15.09 ± 1.57c 9.28 ± 0.82b 11.18 ± 0.62d 14.60 ± 0.52c 35.32 14.08 21.71 27.20 36.96 15.35 23.67 30.51 1.33ab 0.36ab 0.64a 0.76a ± ± ± ± 37.47 17.28 27.66 30.18 TFC (mg RT/g d.w.) TPC(mg GAE/g d.w.) DPPH (mg TE/g d.w.) FRAP(mg TE/g d.w.)

8.42 8.60 6.28 3.99

± ± ± ±

0.20d 0.28cd 0.62e 0.33de

8.04 6.84 4.46 3.19

± ± ± ±

0.78d 0.68de 0.76efg 0.21d

6.90 5.91 3.00 2.56

± ± ± ±

0.77de 0.47e 0.18gf 0.45f

5.37 6.53 1.37 1.85

± ± ± ±

0.45e 0.45e 0.47g 0.28de

8.34 7.00 5.15 3.66

± ± ± ±

0.53d 0.70de 0.70ef 0.14de

Fresh-L FD-S VOD-S CD-S ND-S Fresh-S

Table 5 Changes of total flavonoid (TFC), total phenolic content (TPC) and antioxidant activity of A. keiskeisamples dried by ND, CD, VOD and FD.

ND-L

± ± ± ±

0.89b 0.66b 2.84bc 0.95a

CD-L

± ± ± ±

2.41b 0.82b 1.06c 1.68b

VOD-L

FD-L

± ± ± ±

0.88a 1.36ab 1.12ab 1.41a

C. Zhang et al.

the freeze-drying method, followed by ND, CD, and VOD. The vacuum oven-dried samples had the lowest concentration of TPC and TFC; this result was similar to that obtained by Correia, Grace, Esposito, and Lila (2017), who reported the lowest obtained TPC in wild blueberry after it was VOD. Similarly, thermal drying resulted in significant depletion of polyphenols (p < 0.05). The CD and VOD process resulted in TPC degradation of 15.05% and 42.05%, respectively, in leaves, whereas the FD and ND process caused the TPC to degrade as low as 6.02% and 8.75%, respectively. According to several published studies, this behaviour may be due to the fact that thermal treatment has significant effects on the depletion of polyphenols, promotion of oxidation, or binding of phenolics to proteins that cannot be extracted by presently available methods (Asami et al., 2003; Martin-Cabrejas et al., 2009). 3.6. Effect of the drying process on the antioxidant capacity of A. keiskei extract Table 5 shows the results of the in vitro antioxidant potentials of A. keiskei extract under different drying processes. In accordance with TPC and TFC, the DPPH and FRAP values of the extract of A. keiskei leaves showed a 4.4-fold and 3.6-foldincreaseas compared to the extract of stems, respectively. In the DPPH assay, the highest residual DPPH radical scavenging ability was found in FD products (82.00% for stems vs 93.71% for leaves), followed by ND, CD, and VOD. Many previous studies have reported that an increase in the DPPH free radical scavenging activity in the plant extract is in agreement with a greater abundance of phenolic compounds (TFC and TPC) (Barroso et al., 2018; Zitnanova et al., 2006), which was consistent with a high correlation between TPC (R2 = 0.991), TFC (R2 = 0.988), and DPPH free radical scavenging activity in the present study. In the FRAP assays, the FD and ND samples caused small losses in FRAP values; however, there was not a sizeable difference (p > 0.05) with fresh A. keiskei (Table 4). On the contrary, significantly decreased values (p < 0.05) of antioxidant activity for CD and VOD samples were noted; this was due to the thermal degradation of phytochemicals during the high temperature-based drying method (Reblova, 2012), while FRAP values also exhibited high correlation with the TFC (R2 = 0.992) and TPC (R2 = 0.975). In general, the antioxidant capacity of A. keiskei was positively correlated with the phenolic compound content. Moreover, the samples that underwent the CD and VOD process in which heat was applied showed lower residual antioxidant capacity than those that underwent FD and ND. 4. Conclusion Fresh stems and leaves of A. keiskei were separately dried using ND, CD, FD, and VOD. The drying kinetics showed that the modified Page model exhibited the best fit for A. keiskei. In addition, 10 phenolic compounds were identified in the stems of A. keiskei and 6 in leaves. However, the leaves presented a relatively higher content of antioxidant components, i.e., quercetin, luteolin, and chlorogenic acid, as well as antioxidant activity than that of stems. Samples that underwent the FD and ND processes showed higher retention of color, quercetin, luteolin, and chlorogenic acid content, as well as antioxidant properties (DPPH, FRAP, TPC, and TFC), compared with the CD and VOD processes at 60 °C in which heat was applied. FD product exhibited the best quality characteristic, however, high cost of FD procedures may limit its application. ND thus could be an alternative method when finance become the main concern. Acknowledgements This work was financially supported by the Key Research and Development Project of Zhejiang Province (grant nos. 2017C02021 and 2018C02048). 356

LWT - Food Science and Technology 98 (2018) 349–357

C. Zhang et al.

References

Characterisation, extraction efficiency, stability and antioxidant activity of phytonutrients in Angelica keiskei. Food Chemistry, 115, 227–232. Luo, Y. L., Ya-Li, L. I., Zhu, C. L., & Shen, L. Y. (2017). Growth characteristics and quality components of angelica keiskei in shanghai. Journal of Shanghai Jiaotong University, 35, 45–49. Martin-Cabrejas, M. A., Aguilera, Y., Pedrosa, M. M., Cuadrado, C., Hernandez, T., Diaz, S., et al. (2009). The impact of dehydration process on antinutrients and protein digestibility of some legume flours. Food Chemistry, 114, 1063–1068. Matsuura, M., Kimura, Y., Nakata, K., Baba, K., & Okuda, H. (2001). Artery relaxation by chalcones isolated from the roots of Angelica keiskei. Planta Medica, 67, 230–235. Menon, A. S., Hii, C. L., Law, C. L., Shariff, S., & Djaeni, M. (2017). Effects of drying on the production of polyphenol-rich cocoa beans. Drying Technology, 35, 1799–1806. Miean, K. H., & Mohamed, S. (2001). Flavonoid (myricetin, quercetin, kaempferol, luteolin, and apigenin) content of edible tropical plants. Journal of Agricultural and Food Chemistry, 49, 3106–3112. Mphahlele, R. R., Fawole, O. A., Makunga, N. P., & Opara, U. L. (2016). Effect of drying on the bioactive compounds, antioxidant, antibacterial and antityrosinase activities of pomegranate peel. BMC Complementary and Alternative Medicine, 16(143). Nakamura, T., Tokushima, T., Kawabata, K., Yamamoto, N., Miyamoto, M., & Ashida, H. (2012). Absorption and metabolism of 4-hydroxyderricin and xanthoangelol after oral administration of Angelica keiskei (Ashitaba) extract in mice. Archives of Biochemistry and Biophysics, 521, 71–76. Noh, H., Ahn, E., Yun, J., Cho, B., & Paek, Y. (2015). Angelica keiskei Koidzumi extracts improve some markers of liver function in habitual alcohol drinkers: A randomized double-blind clinical trial. Journal OfMedicinal Food, 18, 166–172. Ogawa, H., Nakashima, S., & Baba, K. (2003). Effects of dietary Angelica keiskei on lipid metabolism in stroke-prone spontaneously hypertensive rats. Clinical and Experimental Pharmacology and Physiology, 30, 284–288. Reblova, Z. (2012). Effect of temperature on the antioxidant activity of phenolic acids. Czech Journal of Food Sciences, 30, 171–177. Rice-Evans, C. A., Miller, N. J., & Paganga, G. (1996). Structure-antioxidant activityrelationships of flavonoids and phenolic acids. Free Radical Biology and Medicine, 20, 933–956. Seong, G., Hwang, I., & Chung, S. (2016). Antioxidant capacities and polyphenolics of Chinese cabbage (Brassica rapa L. ssp Pekinensis) leaves. Food Chemistry, 199, 612–618. Shimizu, E., Hayashi, A., Takahashi, R., Aoyagi, Y., Murakami, T., & Kimoto, K. (1999). Effects of angiotensin I-converting enzyme inhibitor from Ashitaba (Angelicakeiskei) on blood pressure of spontaneously hypertensive rats. Journal Of Nutritional Science and Vitaminology (Tokyo), 45, 375–383. Son, D. J., Park, Y. O., Yu, C., Lee, S. E., & Park, Y. H. (2014). Bioassay-guided isolation and identification of anti-platelet-active compounds from the root of Ashitaba (Angelica keiskei Koidz.). Natural Product Research, 28, 2312–2316. Soysal, Y., & Oztekin, S. (2001). Technical and economic performance of a tray dryer for medicinal and aromatic plants. Journal of Agricultural Engineering Research, 79, 73–79. Szychowski, P. J., Lech, K., Sendranadal, E., Hernández, F., Figiel, A., Wojdyło, A., et al. (2018). Kinetics, biocompounds, antioxidant activity, and sensory attributes of quinces as affected by drying method. Food Chemistry, 255, 157–164. Wu, J., Yang, C., & Wang, Z. (2012). Angelica keiskei health-care chewable tablet and preparation method thereof. CN 102326776 A. Yábar, E., Pedreschi, R., Chirinos, R., & Campos, D. (2011). Glucosinolate content and myrosinase activity evolution in three maca (lepidium meyenii walp.) ecotypes during preharvest, harvest and postharvest drying. Food Chemistry, 127, 1576–1583. Zhang, Z., Kou, X. L., Fugal, K., & McLaughlin, J. (2004). Comparison of HPLC methods for determination of anthocyanins and anthocyanidins in bilberry extracts. Journal of Agricultural and Food Chemistry, 52, 688–691. Zitnanova, I., Ranostajova, S., Sobotova, H., Demelova, D., Pechan, I., & Durakova, Z. (2006). Antioxidative activity of selected fruits and vegetables. Biologia, 61, 279–284.

Aldini, G., Regazzoni, L., Pedretti, A., Carini, M., Cho, S. M., Park, K. M., et al. (2011). An integrated high resolution mass spectrometric and informatics approach for the rapid identification of phenolics in plant extract. Journal of Chromatography A, 1218, 2856–2864. An, K., Zhao, D., Wang, Z., Wu, J., Xu, Y., & Xiao, G. (2016). Comparison of different drying methods on Chinese ginger (Zingiberofficinale Roscoe): Changes in volatiles, chemical profile, antioxidant properties, and microstructure. Food Chemistry, 197, 1292–1300. Asami, D. K., Hong, Y. J., Barrett, D. M., & Mitchell, A. E. (2003). Comparison of the total phenolic and ascorbic acid content of freeze-dried and air-dried marionberry, strawberry, and corn grown using conventional, organic, and sustainable agricultural practices. Journal of Agricultural and Food Chemistry, 51, 1237–1241. Bao, G. (2014). Angelica keiskei-containing nutritional and health protecting flour and its preparation method. CN 104187320 A. Barroso, M. R., Martins, N., Barros, L., Antonio, A. L., Angelo Rodrigues, M, Sousa, M. J., et al. (2018). Assessment of the nitrogen fertilization effect on bioactive Compounds of frozen fresh and dried samples of Stevia rebaudiana Bertoni. Food Chemistry, 243, 208–213. Carabasa-Giribet, M., & Ibarz-Ribas, A. (2000). Kinetics of colour development in aqueous glucose systems at high temperatures. Journal of the Science of Food and Agriculture, 44(3), 181–189. Chen, Q., Li, Z., Bi, J., Zhou, L., Yi, J., & Wu, X. (2017). Effect of hybrid drying methods on physicochemical, nutritional and antioxidant properties of dried black mulberry. Lebensmittel-Wissenschaft und -Technologie- Food Science and Technology, 80, 178–184. Chuyen, H. V., Roach, P. D., Golding, J. B., Parks, S. E., & Nguyen, M. H. (2017). Effects of four different drying methods on the carotenoid composition and antioxidant capacity of dried Gac peel. Journal of the Science of Food and Agriculture, 97, 1656–1662. Correia, R., Grace, M. H., Esposito, D., & Lila, M. A. (2017). Wild blueberry polyphenolprotein food ingredients produced by three drying methods: Comparative physicochemical properties, phytochemical content, and stability during storage. Food Chemistry, 235, 76–85. Dadali, G., Apar, D. K., & Özbek, B. (2007). Microwave drying kinetics of okra. Drying Technology, 25, 917–924. Gumusay, O. A., Borazan, A. A., Ercal, N., & Demirkol, O. (2015). Drying effects on the antioxidant properties of tomatoes and ginger. Food Chemistry, 173, 156–162. Heo, J. M., Kang, H. M., Jung, H. K., & Shin, A. R. (2013). Cosmetic composition containing extract of Dioscorea bulbifera L., Angelica keiskei, Hibiscus esculentus and Colocasia antiquorumSchott et Endl. var. esculenta Engl. for improving skin moisture. KR1020130168410. Jiang, N., Liu, C., Li, D., Zhang, Z., Liu, C., Wang, D., et al. (2017). Evaluation of freeze drying combined with microwave vacuum drying for functional okra snacks: Antioxidant properties, sensory quality, and energy consumption. LWT - Food Science and Technology, 82, 216–226. Kang, M. H., Park, Y. K., Kim, H. Y., & Kim, T. S. (2004). Green vegetable drink consumption protects peripheral lymphocytes DNA damage in Korean smokers. BioFactors, 22, 245–247. Karam, M. C., Petit, J., Zimmer, D., Djantou, E. B., & Scher, J. (2016). Effects of drying and grinding in production of fruit and vegetable powders: A review. Journal of Food Engineering, 188, 32–49. Kim, S. J., Cho, J. Y., Wee, J. H., Jang, M. Y., Kim, C., Rim, Y. S., et al. (2005). Isolation and characterization of antioxidative compounds from the aerial parts of Angelica keiskei. Food Science and Biotechnology, 14, 58–63. Kim, D. W., Curtis-Long, M. J., Yuk, H. J., Wang, Y., Song, Y. H., Jeong, S. H., et al. (2014). Quantitative analysis of phenolic metabolites from different parts of Angelica keiskei by HPLC-ESI MS/MS and their xanthine oxidase inhibition. Food Chemistry, 153, 20–27. Li, L., Aldini, G., Carini, M., Chen, C. Y. O., Chun, H., Cho, S., et al. (2009).

357