Comparison of free and bound phenolic compositions and antioxidant activities of leaves from different mulberry varieties.

Mulberry leaves are used in traditional Chinese medicine and contain numerous active substances that are known to be beneficial for human health. The aim of this study was to investigate the phenolic compositions and antioxidant activities of the leaves from 23 mulberry cultivars. Qualitative LC-ESI-QTOF analysis revealed the presence of 11 phenolic compounds in the free phenolic extracts and 10 phenolic compounds in the bound fractions. Chlorogenic acid and caffeic acid were the major components in the free and bound fractions, respectively. The results revealed that the changguosang cultivar from Taiwan contained the greatest content of phenolic compounds as well as the highest antioxidant activity among the 23 cultivars examined, as determined using three separate antioxidant assays. The isoquercitrin, chlorogenic acid, and rutin contents of the free phenolic extracts displayed significant correlations with the antioxidant activities, while syringic acid and rutin were the main contributors to the antioxidant activities of the bound phenolic fractions. The obtained results demonstrate that mulberry leaves contain a variety of beneficial phenolic substances and may be suitable for further development as a herbal medicine.

Introduction

Phytonutrients play important roles in improving human health and may protect against heart disease, cancer, the effects of aging, and membrane damage. In particular, phytonutrients possessing antioxidant properties can inhibit the propagation of free-radical reactions implicated in the development of aging-related diseases. Consequently, numerous studies have been conducted to elucidate the characteristics and roles of antioxidant compounds from plants. Antioxidants can delay or suppress the oxidation of molecules by inhibiting the activation or propagation of oxidative chain reactions or scavenging the free radicals generated during oxidative processes [37]. The antioxidant activity of plant tissues is primarily attributable to phenolic, etc. and a high correlation was observed between the antioxidant activity and the total phenolic content of the extract. [1]. Hence, phenolic compounds have attracted considerable attention as potential protective factors against cancer and heart disease owing to their antioxidant activities [8].

Mulberry (Morus alba L.) is a moraceous plant that is extensively cultivated throughout Asia to feed silkworms during the commercial production of silk. It is distributed throughout temperate to subtropical and tropical regions and can be grown under a wide range of climatic, topographic, and soil conditions [11]. Mulberry leaves are considered a nutritious, palatable, and safe food or food additive containing carbohydrates, proteins, calcium, iron, β-carotene, and vitamin B1 [6] and are also a rich source of phenolic compounds such as caffeic acid, rutin, quercetin, isoquercitrin, and astragalin [9, 23]. Mulberry leaves are commonly used as antidiabetic, hypolipidemic, antihypertensive, anti-atherosclerotic, and anticonvulsant agents [4]. These pharmacological effects of mulberry leaves are closely related to their phenolic composition [14]. Plant phenolic compounds have been reported to play a preventive role against various diseases owing to their remarkable antioxidant, antimicrobial, and other activities [7, 29]. For example, Wu et al. [38] reported that mulberry leaf phenolic extract decreases hepatic lipid accumulation via activation of the AMP-activated protein kinase signaling pathway.

There is strong evidence for the variation of phenolic content and antioxidant activity among different species of plants belonging to the same genus [30, 35]. Thus, the identification and quantification of phenolic compounds can provide crucial information regarding the antioxidant function, food quality, and potential health benefits of a specific plant. G1, G6, G8, and GSW are from North China; B-2–8, R2, 7403, and DS are from South China; Z1, Z2, and Z4 are from Southwest China; CGS is from Taiwan; J4-1 and J5 are from Japan; T6, T7, BR60, S54, and QM are from Thailand; Y2 and YXM2 are from India; and Y1 and YD are from Vietnam. Therefore, the aim of this study was to determine and compare the phenolic compositions and antioxidant activities of these 23 samples of mulberry leaves.

Materials and methods

Mulberry leaf samples and preparation

All of the mulberry varieties examined in this study were cultivated in an experimental field in Guangzhou, which was managed by the South China branch of the National Mulberry Germplasm Resource Garden. Mulberry leaves are generally picked in the spring and late autumn frost period, and it is better to pick them in the morning and evening. Leaves from each mulberry variety were harvested, washed with distilled water, and air-dried at 55 °C in a thermostatic hot air drying oven for 7–9 h. The dried leaves were ground into powders using a high-speed pulverizer and stored in airtight containers at − 20 °C prior to analysis [35]. The different mulberry varieties examined are listed in Table 1.

Table 1

Names, species, and abbreviations of the mulberry varieties examined in this study

Cultivar	Species	Abbreviation
Gu 1	M. alba L	G1
Gu 6	M. alba L	G6
Gu 8	M. alba L	G8
Gusangwang	M. mongolica var. diabolica Koidz	GSW
Bei-2-8	M. atropurpurea Roxb	B-2-8
R2	M. atropurpurea Roxb	R2
7403	M. atropurpurea Roxb	7403
Dashi	M. atropurpurea Roxb	DS
Zangjiangxin 1	M. mongolica C.K.Schneid	Z1
Zangjiangxin 2	M. mongolica C.K.Schneid	Z2
Zangjiangxin 4	M. mongolica C.K.Schneid	Z4
Changguosang	M. rotundiloba Koidz	CGS
JP4-1	M. multicaulis Koidz	J4-1
JP5	M. multicaulis Koidz	J5
TL6	M. rotundiloba Koidz	T6
TL7	M. rotundiloba Koidz	T7
BR60	M. rotundiloba Koidz	BR60
S54	M. rotundiloba Koidz	S54
Qingmai	M. rotundiloba Koidz	QM
Yin 2	M. serrata Rox	Y2
Yinximeng 2	M. serrata Rox	YXM2
Yue 1	M. rotundiloba Koidz	Y1
Yueda	M. rotundiloba Koidz	YD

Chemicals and reagents

Acetone, n-hexane, ethyl acetate, concentrated sulfuric acid, sodium hydroxide, phosphoric acid, acetonitrile, ascorbic acid, methanol, formic acid, sodium acetate, acetic acid, hydrochloric acid, iron(III) chloride hexahydrate solution, ammonium acetate, were purchased from Guangzhou Chemical Reagent Factory (Guangzhou, China). phenolic compound standards [rutin (Rut), quercetin (Que), quercitrin (Quer), isoquercitrin (Iso), chlorogenic acid (ChA), syringic acid (SyA), ferulic acid (FeA), caffeic acid (CaA), resveratrol (Res), epicatechin (Epi), astragalin (Ast), scopoletin (Sco), galangal (Gal), catechuic acid (CatA), vanillic acid (VaA), benzoic acid (BeA), gallic acid (GaA), and protocatechuic acid (PrA)] were purchased from the National Institutes for Food and Drug Control (Beijing, China). 2,4,6-tris(2-pyridyl)-S-triazine (TPTZ), 1,1-diphenyl-2-picrylhydrazyl (DPPH), 2,2ʹ-Azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS), and 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (trolox) were purchased from Sigma-Aldrich (St. Louis, MO, USA).

Extraction of free phenolic compounds

It was necessary to remove lipids from the mulberry leaves prior to extraction of the free phenolic compounds [12]. Considering that the lipid content of mulberry leaves is not high, the degreasing procedure in this study was adjusted. Free phenolic compounds were extracted according to a modified version of the methods reported by Li et al. [20] and Kim et al. [16]. Briefly, samples of the dried leaves (10 g) were extracted with n-hexane (250 mL) for 10 min under continuous stirring (10,000 rpm) in an ice bath, followed by centrifugation (5000 rpm, 5 min) and removal of the supernatant. The residue was added to chilled acetone/water (8:2, v/v, 250 mL) followed by homogenization (5000 rpm, 5 min) and centrifugation (5000 rpm, 5 min). The residue was extracted again under the same conditions. The two supernatants were combined and evaporated to dryness at 50 °C on a rotary evaporator (RE-52AA, Shanghai Yarong Biochemical instrument Factory, Shanghai, China). The dried samples were dissolved in methanol/water (8:2, v/v), filtered through 0.22 μm membrane filters, and stored at − 80 °C prior to analysis. Each sample was extracted in duplicate. The residue was used to measure the bound phenolic compounds, as described in the following subsection.

Extraction of bound phenolic compounds

The bound phenolic compounds were extracted by alkaline hydrolysis according to previously described procedures [5, 18, 21] with some modifications. Briefly, the mulberry leaf residues obtained after extraction of the free phenolic compounds were hydrolyzed in 2 M NaOH (100 mL) at room temperature for 1.5 h with continuous stirring under nitrogen atmosphere. The resulting mixtures were acidified to pH 2 with 6 M HCl and then extracted six times with ethyl acetate (120 mL each time). The ethyl acetate fractions were combined and evaporated to dryness at 45 °C on a rotary evaporator (RE-52AA). The dried extracts were redissolved in 80% methanol, filtered through 0.22 μm membrane filters, and stored at − 80 °C prior to analysis.

LC-ESI-QTOF analysis

Samples were analyzed by LC-ESI-QTOF according to a modified version of the method described by Tomas et al. [36]. Each sample was vortexed for 30 s, filtered through a 0.22 μm organic membrane, and transferred into an injection vial. The temperature of the chromatographic column was set to 35 °C, and the injection volume was 1 μL. The mobile phase consisted of 0.1% formic acid in water (solvent A) and 0.1% formic acid in acetonitrile (solvent B) in positive ion mode or 2 mM ammonium acetate in water (solvent A) and acetonitrile (solvent B) in negative ion mode. The flow rate was set to 400 μL/min with a gradient from 5 to 95% solvent B over 22 min, as shown in Table 2. The Agilent 6545A QTOF mass spectrometer is controlled by the control software (LC/MS Data Acquisition, Version B.08.00) based on the Auto MS/MS mode for primary and secondary mass spectrometry data acquisition. The MS and MS2 acquisition rates were 5 and 10 spectra per second, respectively. The secondary collision energy is 0v and 10v respectively.Eight ions were selected in the first-level spectrum for the second-level scan. The m/z range of primary and secondary mass scanning are both 50–1100. Spectra were collected in both positive and negative modes, and the acquired data were saved in centroid format. The ESI ion source parameters were as follows: ion source gas temperature, 320 °C; nitrogen flow rate, 8 L/min; sheath gas flow rate, 12 L/min; sheath gas temperature, 350 °C; capillary voltage, 4000 V (positive ion mode) or 3500 V (negative ion mode).

Table 2

LC-ESI-QTOF mobile phase gradient

Time (min)	Flow rate (μL/min)	A (%)	B (%)
0	400	95	5
1.5	400	95	5
2.5	400	90	10
14	400	60	40
22	400	5	95
25	400	5	95
26	400	95	5
30	400	95	5

Determination of phenolic content using HPLC

The phenolic compounds present in the 23 mulberry leaf samples were determined by HPLC analysis (Agilent 1200, Agilent Technologies Inc., Karlsruhe, Germany) according to a modified version of the methods reported by Eom et al. [10] and Subhashinee et al. [33]. Chromatographic separation was performed using an Capcell Pak ADME column (250 × 4.6 mm, 5 μm). The mobile phase consisted of phosphoric acid/water (0.2:100, v/v, solvent A) and acetonitrile (solvent B). The solvent gradient was as follows: 0–20 min 10% B, 20–30 min 16% B, 30–40 min 16% B, 40–60 min 20% B, 60–70 min 30% B, 70–71 min 35% B, 71–75 min 80% B, 75–76 min 80% B, 76–90 min 10% B. The solvent flow rate was 1.0 mL/min, the column temperature was set to 25 °C, and chromatograms were recorded at 280 nm and 350 nm.

Dionex Acclaim 120 C-18 analytical column.

CAPCELL PAK ADM.

Determination of FRAP activity

The ferric reducing antioxidant power (FRAP) assay was performed according to a modified version of a previously reported method [39, 40]. A working solution was prepared by mixing 10 mL of 300 mM acetate buffer (0.1870 g of sodium acetate and 1.6 mL of acetic acid), 1 mL of TPTZ solution (10 mM TPTZ in 40 mM HCl), and 1 mL of 20 mM iron(III) chloride hexahydrate solution. This mixture was pre-warmed to 37 °C and should always be prepared fresh. Samples of the mulberry leaf extracts (100 μL) were incubated with 3.0 mL of the FRAP reagent for 4 min at 25 °C. The absorbance at 593 nm was then measured using a spectrophotometer (UV-1700, Shimadzu Instruments Manufacturing, CO., LTD, Suzhou, China). The FRAP values were expressed as μmol Fe2+ equivalents per gram of dry weight (μmol Fe2+/g DW).

Determination of DPPH radical scavenging activity

The DPPH radical scavenging activities of the mulberry leaf samples were determined as described in previous reports [2, 3, 22]. Briefly, the mulberry leaf extracts and DPPH solutions were diluted to appropriate concentrations, and a solution of ascorbic acid in methanol was used to prepare a standard curve (R2 = 0.993). Next, 1 mL of the diluted sample was mixed with 5 mL of 2 mM DPPH solution in methanol. The mixture was stirred vigorously and incubated for 50 min in the dark at room temperature. The absorbance at 520 nm was then measured using a UV2300II spectrophotometer and the results were expressed in μmol of ascorbic acid equivalent antioxidant capacity (AEAC) per gram of dry weight (μmol AEAC/g DW).

Determination of ABTS radical scavenging activity

The ABTS radical scavenging activities of the mulberry leaf samples were determined as described in previous reports [27, 31, 34]. Briefly, an ABTS stock solution was prepared by mixing 7 mM ABTS and 2.45 mM potassium persulfate (1:1, v/v) followed by incubation for 16 h in the dark at room temperature; the resulting ABTS radical solution was used within 24 h. The ABTS stock solution was diluted with methanol until its absorbance at 734 nm reached 0.70 ± 0.02. Next, 100 μL of the diluted sample was mixed with 3.8 mL of the ABTS working solution and the absorbance at 734 nm was measured after incubation at room temperature for exactly 6 min. Trolox was used as a reference to generate a standard curve (R2 = 0.995), and the results were expressed as μmol of trolox equivalent antioxidant capacity (TEAC) per gram of dry weight (μmol TEAC/g DW).

Results and discussion

Qualitative analysis of phenolic compounds in mulberry leaves

To determine the presence of various biologically active phenolic compounds in the mulberry leaf samples, the obtained mass spectra of the samples were compared with spectra of standard compounds with respect to retention time (RT), molecular ion peak, and structural fragments observed in the secondary mass spectra.

In LC–MS, a sample is first subjected to liquid chromatography to separate the sample components, which are subsequently ionized and separated according to their mass-to-charge ratios to reveal information regarding the molecular weight, structure, and component amount.

Mulberry leaves are rich in a variety of phenolic compounds, the composition and content of which can vary for mulberry leaves from different cultivars and regions. LC-ESI-QTOF analysis revealed the presence of 11 highly matched phenolic substances in the free phenolic extracts and 10 phenolic substances in the bound phenolic extracts, as summarized in Tables 3 and 4, respectively. The free phenolic compounds detected in the mulberry leaf samples were BeA, PrA, GaA, CaA, Sco, Epi, Que, ChA, Iso, Rut, and Ast, while the bound phenolic compounds were SyA, GaA, BeA, PrA, CaA, Iso, Rut, FeA, Ast, and Que. Among these compounds, Sco, Epi, and ChA were only detected in the free phenolic extracts, which indicates that these substances may be either absent from the bound phenolic extracts. In contrast, SyA and FeA were only detected in the bound phenolic extracts and not in the free phenolic extracts. Zou et al. [40] reported certain differences in the phenolic contents of different varieties of mulberry leaves, which is useful for helping farmers to select which varieties to cultivate for higher quality mulberry leaves. Qadir et al. [26] applied gas chromatography–mass spectrometry to identify and quantify the main phenolic compounds present in M. alba leaf extracts, including Que, GaA, and SyA. In the bound phenolic extracts, ChA was not detected, indicating that ChA may only exist in the free phenolic extracts. Most of the phenolic substances present in high contents in the free and bound phenolic extracts could be qualitatively analyzed by LC–MS.

Table 3

Qualitative analysis of free phenolic compounds in mulberry leaves

Compound	Molecular formula	Expected m/z	Observed m/z	Mass error (ppm)	RT (min)	Fragment ions (relative intensity, %)
BeA	C7H6O2	121.0295	121.0294	− 0.83	7.3186	121.02985 (100%), 77.04045 (14.2%), 82.49994 (1.9%)
PrA	C7H6O4	153.0193	153.0194	0.65	7.9276	153.01923 (100%), 153.04488 (3.3%), 109.02855 (2.6%)
GaA	C7H6O5	169.0142	169.0145	1.77	3.4645	169.01408 (100%), 125.02387 (3.3%), 61.7975 (2.5%)
CaA	C9H8O4	179.0350	179.0351	0.56	7.6609	179.03479 (100%), 135.04578 (5.3%)
Sco	C10H8O4	191.0350	191.0350	0	9.9880	191.03455 (100%), 176.01178 (9.3%), 147.02908 (3.5%)
Epi	C15H14O6	289.0718	289.0678	− 13.84	2.2628	243.0616 (100%), 289.06708 (41.6%), 244.06404 (12.5%)
Que	C15H10O7	301.0354	301.0356	0.66	13.8163	301.03537 (100%), 178.99699 (6.5%), 151.00305 (4.4%), 63.02433 (2.1%)
ChA	C16H18O9	353.0878	353.0865	− 3.8	6.7420	191.05605 (100%), 192.05885 (8.7%)
Iso	C21H20O12	463.0882	463.0903	4.53	8.6035	463.08832 (100%), 301.03555 (3%)
Rut	C27H30O16	609.1461	609.1478	2.79	8.9101	609.14642 (100%), 609.18945 (9%), 300.02475 (2.2%)
Ast	C21H20O11	447.0933	447.0942	− 2.01	10.6400	447.0932 (100%), 285.0388 (18.98%), 449.0994 (8.2%)

All spectra were recorded in negative ion mode with a collision energy of 10 V

Table 4

Qualitative analysis of bound phenolic compounds in mulberry leaves

Compound	Molecular formula	Expected m/z	Observed m/z	Mass error (ppm)	RT (min)	Fragment ions (relative intensity, %)
SyA	C9H10O5	197.0455	197.0457	1.01	3.3027	135.04504 (100%), 153.05786 (31.6%), 162.83853 (26.5%), 61.98833 (25.3%)
GaA	C7H6O5	169.0142	169.0144	1.18	3.4643	169.01404 (100%), 168.88683 (15.4%), 125.02482 (8.8%), 122.89381 (2.7%)
BeA	C7H6O2	121.0295	121.0305	8.26	7.3001	121.02959 (100%), 77.04009 (10.4%), 77.05341 (0.8%)
PrA	C7H6O4	153.0193	153.0193	0	7.7071	153.01918 (100%), 109.02876 (6.1%)
CaA	C9H8O4	179.0350	179.0352	1.12	7.7366	179.03502 (100%), 135.04532 (12.9%)
Iso	C21H20O12	463.0882	463.0877	− 1.08	8.5514	463.08813 (100%), 301.03839 (3.6%), 193.01373 (1.9%)
Rut	C27H30O16	609.1461	609.1469	1.31	9.2957	609.14612 (100%), 300.02649 (3.7%)
FeA	C10H10O4	193.0506	193.0508	1.04	10.0027	193.05028 (100%), 134.03656 (4.7%)
Ast	C21H20O11	447.0933	447.0937	0.89	10.9455	447.0921 (100%), 285.03876 (3.7%)
Que	C15H10O7	301.0354	301.0353	− 0.33	13.7901	301.03534 (100%), 300.99191 (5.2%)

All spectra were recorded in negative ion mode with a collision energy of 10 V

Total phenolic contents

The total phenolic contents in the 80% acetone extracts of the 23 samples were measured by HPLC as presented in Table 5 and Fig. 1, revealing clear differences between the various cultivars. The total phenolic contents (including both free and bound compounds) increased in the following order: G1 (2.61 mg/g DW) < S54 (4.0 mg/g DW) < G8 (4.22 mg/g DW) < BR60 (4.55 mg/g DW) < 7403 (4.62 mg/g DW) < Z1 (4.97 mg/g DW) < Z2 (5.98 mg/g DW) < J5 (7.08 mg/g DW) < G6 (7.47 mg/g DW) < Z4 (7.50 mg/g DW) < T7 (8.07 mg/g DW) < YXM2 (8.24 mg/g DW) < R2 (9.66 mg/g DW) < T6 (13.49 mg/g DW) < B-2–8 (16.27 mg/g DW) < QM (16.75 mg/g DW) < GSW (17.3 mg/g DW) < J4-1 (18.46 mg/g DW) < Y2 (23.83 mg/g DW) < YD (25.72 mg/g DW) < DS (27.99 mg/g DW) < Y1 (39.12 mg/g DW) < CGS (51.81 mg/g DW).

Table 5

Total, free, and bound phenolic contents of the 23 mulberry leaf samples

Cultivar	Total phenolic content(μg/g DW)	Free phenolic content(μg/g DW)	Bound phenolic content(μg/g DW)
G1	2607.7 ± 161.35 k	2264.56 ± 161.35 k (86.84%)	343.14 ± 46.25ghi (13.16%)
G6	7472.27 ± 572.98hij	7147.13 ± 572.98ijk (95.65%)	325.14 ± 43.63ghi (4.35%)
G8	4224.72 ± 418.4ijk	4023.23 ± 418.4ijk (95.23%)	201.49 ± 33.66j (4.77%)
GSW	17,302.22 ± 1667.62ef	16,191.26 ± 1667.62ef (93.58%)	1110.96 ± 105.93a (6.42%)
DS	27,990.69 ± 2711.73c	27,156.6 ± 2711.73c (97.02%)	834.08 ± 80.98bc (2.98%)
7403	4621.51 ± 367.57ijk	4347.98 ± 367.57ijk (94.08%)	273.53 ± 40.51hij (5.92%)
R2	9660.23 ± 831.38gh	9322.52 ± 831.38gh (96.5%)	337.71 ± 40.26ghi (3.5%)
B-2–8	16,272.83 ± 1808.31ef	15,757.86 ± 1808.31ef (96.84%)	514.97 ± 66.57f (3.16%)
Z1	4965.79 ± 548.09ijk	4526.59 ± 548.09ijk (91.16%)	439.19 ± 56.41 fg (8.84%)
Z2	5984.99 ± 571.56hijk	5110.54 ± 571.56ijk (85.39%)	874.44 ± 95.62b (14.61%)
Z4	7498.51 ± 682.07hij	6746.59 ± 682.07ijk (89.97%)	751.92 ± 87.75 cd (10.03%)
CGS	51,811.86 ± 5233.05a	51,279.3 ± 5233.05a (98.97%)	532.56 ± 46.46ef (1.03%)
T6	13,490.12 ± 1792.09 fg	12,954.33 ± 1792.09 fg (96.03%)	535.78 ± 65.37ef (3.97%)
T7	8068.84 ± 973.34hi	7828.8 ± 973.34hi (97.03%)	240.04 ± 27.12ij (2.97%)
BR60	4545.99 ± 456.22ijk	4300.87 ± 456.22ijk (94.61%)	245.12 ± 27.41ij (5.39%)
S54	4000 ± 397.67jk	3693.5 ± 397.67jk (92.34%)	306.5 ± 34.73hij (7.66%)
QM	16,748.36 ± 2401.89ef	16,240.99 ± 2401.89ef (96.97%)	507.38 ± 53.43f (3.03%)
J4-1	18,463.88 ± 2285.15e	17,712.43 ± 2285.15e (95.93%)	751.45 ± 80.13 cd (4.07%)
J5	7082.78 ± 636.06hij	6796.07 ± 636.06ijk (95.95%)	286.71 ± 27.65hij (4.05%)
Y2	23,833.07 ± 2884.06d	23,127.05 ± 2884.06d (97.04%)	706.03 ± 74.29d (2.96%)
YXM2	8235.79 ± 611.32hi	7849.57 ± 611.32hi (95.31%)	386.21 ± 36.6gh (4.69%)
YD	25,720.37 ± 3059.38 cd	25,347.54 ± 3059.38 cd (98.55%)	372.83 ± 33.87gh (1.45%)
Y1	39,120.86 ± 3514.82b	38,478.25 ± 3514.82b (98.36%)	642.61 ± 60.36de (1.64%)

Different letters indicate significant differences in total, free, and bound phenolic contents in 23 mulberry leaf samples (P < 0.05)

Values with no letters in common in each column are significantly different (P < 0.05), n = 3

Values in parentheses indicate the percentage contributions of the free and bound fractions to the total phenolic content for each sample

Fig. 1

Free and bound phenolic contents of the 23 mulberry leaf samples

Table 5 also lists the free and bound phenolic contents and their corresponding percentages for the 23 samples. The contribution of the free phenolic fraction to the total phenolic content ranged from 85.39% (Z2) to 98.97% (CGS). The free phenolic content in the 23 varieties of mulberry leaves varied from 2.26 mg/g DW (G1) to 51.28 mg/g DW (CGS) and followed almost the same order as the total phenolic content: G1 (2.26 mg/g DW) < S54 (3.69 mg/g DW) < G8 (4.02 mg/g DW) < BR60 (4.3 mg/g DW) < 7403 (4.35 mg/g DW) < Z1 (4.53 mg/g DW) < Z2 (5.11 mg/g DW) < Z4 (6.75 mg/g DW) < J5 (6.80 mg/g DW) < G6 (7.15 mg/g DW) < T7 (7.83 mg/g DW) < YXM2 (7.85 mg/g DW) < R2 (9.32 mg/g DW) < T6 (12.95 mg/g DW) < B-2–8 (15.76 mg/g DW) < GSW (16.19 mg/g DW) < QM (16.24 mg/g DW) < J4-1 (17.71 mg/g DW) < Y2 (23.13 mg/g DW) < YD (25.35 mg/g DW) < DS (27.16 mg/g DW) < Y1 (38.48 mg/g DW) < CGS (51.28 mg/g DW).

In contrast, the bound phenolic content in the samples followed a quite different order: G8 (0.20 mg/g DW) < T7 (0.24 mg/g DW) < BR60 (0.25 mg/g DW) < 7403 (0.27 mg/g DW) < J5 (0.29 mg/g DW) < S54 (0.31 mg/g DW) < G6 (0.33 mg/g DW) < R2 and G1 (0.34 mg/g DW) < YD (0.37 mg/g DW) < YXM2 (0.39 mg/g DW) < Z1 (0.44 mg/g DW) < QM (0.51 mg/g DW) < B-2–8 (0.51 mg/g DW) < CGS (0.53 mg/g DW) < T6 (0.54 mg/g DW) < Y1 (0.64 mg/g DW) < Y2 (0.71 mg/g DW) < J4-1 (0.75 mg/g DW) < Z4 (0.75 mg/g DW) < DS (0.83 mg/g DW) < Z2 (0.87 mg/g DW) < GSW (1.11 mg/g DW).

Overall, the results revealed that CGS from Taiwan possessed the highest total and free phenolic contents, whereas G1 from North China displayed the lowest values among the 23 cultivars studied. In contrast, the bound phenolic contents of the mulberry cultivars followed a different trend, with the highest and lowest values observed for GSW and G8, respectively. The results further demonstrate that the phenolic compounds present in mulberry leaves predominantly exist in the free state. Furthermore, the total phenolic contents of mulberry leaf cultivars from the same geographical area varied considerably; for instance, GSW, DS, Z4, QM, J4-1, Y2, and Y1 displayed the highest free phenolic contents of the specimens from North China, South China, Southwest China, Thailand, Japan, India, and Vietnam, respectively.

Free and bound phenolic profiles

Chromatograms of phenolic compounds of the leaves of the mulberry varieties from Z2 are shown in Figs. 2 and 3. Identify the main phenolic compounds by comparing the retention time and other data of the standards. In addition, phenolic compounds are quantified by using a corresponding standard curve with a higher correlation value. Tables 6 and 7 summarize the contents of each phenolic substance in the mulberry leaf samples in the free and bound states, respectively. As shown in Table 6, HPLC analysis revealed the presence of 11 phenolic compounds in the free state, of which ChA, Epi, CaA, Rut, Iso, and Ast together accounted for over 50% of the free phenolic content. This result is consistent with previous reports by Zou et al. [40] and Onogi et al. [25]. The values of ChA ranged from 1.396 to 44.627 mg/g DW and essentially determined the free phenolic content of the mulberry leaves. The leaves of the mulberry varieties from Vietnam (YD and Y1) and Taiwan (CGS) were found to be especially rich in ChA. As mentioned in previous reports [17, 28], ChA can serve as an antioxidant in vitro and may hinder the formation of mutagenic and carcinogenic N-nitroso compounds by inhibiting N-nitrosation reactions. Furthermore, ChA was reported to inhibit the oxidation of low-density lipoprotein in vitro and could protect against cardiovascular disease [19]. Hence, ChA could potentially be extracted from mulberry leaves and refined for use as a medicine to treat human diseases.

Fig. 2

HPLC Chromatograms at 280 nm of the leaves of the mulberry varieties from Z2

Fig. 3

HPLC Chromatograms at 350 nm of the leaves of the mulberry varieties from Z2

Table 6

Free phenolic compounds detected in the 23 mulberry leaf samples by HPLC

Compound(mg/g DW)	ChA	Epi	CaA	Rut	Iso	Ast
G1	1.396 ± 0.128j	0.3 ± 0.039gh	0.126 ± 0.016 cd	0.073 ± 0.01 k	0.178 ± 0.017 k	0.123 ± 0.008 lm
G6	4.268 ± 0.65ghij	0.73 ± 0.096bc	0.124 ± 0.016 cd	0.309 ± 0.043ijk	0.854 ± 0.089f	0.756 ± 0.071ef
G8	2.402 ± 0.242ij	0.388 ± 0.052efgh	0.145 ± 0.03c	0.285 ± 0.055ijk	0.513 ± 0.063fghijk	0.183 ± 0.042ijklm
GSW	12.603 ± 1.26d	0.467 ± 0.061e	0.047 ± 0.006fgh	1.068 ± 0.085de	1.494 ± 0.194de	0.472 ± 0.058 g
R2	21.127 ± 1.93c	0.81 ± 0.088ab	0.235 ± 0.049ab	1.136 ± 0.179d	2.093 ± 0.216b	1.553 ± 0.228a
7403	3.122 ± 0.301hij	0.3 ± 0.025gh	0.037 ± 0.003gh	0.181 ± 0.019ijk	0.29 ± 0.028ijk	0.274 ± 0.022hijkl
DS	7.494 ± 0.614e	0.338 ± 0.036fgh	0.223 ± 0.031b	0.409 ± 0.033hijk	0.447 ± 0.041ghijk	0.279 ± 0.052hijkl
B-2-8	13.766 ± 1.593d	0.352 ± 0.043efgh	0.045 ± 0.006gh	0.252 ± 0.046ijk	0.733 ± 0.061 fg	0.46 ± 0.039 g
Z1	3.49 ± 0.416ghij	0.341 ± 0.035fgh	0.071 ± 0.009efg	0.134 ± 0.025jk	0.341 ± 0.045hijk	0.067 ± 0.009 m
Z2	3.42 ± 0.333ghij	0.376 ± 0.071efgh	0.118 ± 0.013 cd	0.289 ± 0.044ijk	0.681 ± 0.071fgh	0.14 ± 0.026klm
Z4	4.9 ± 0.433fghi	0.395 ± 0.052efg	0.277 ± 0.053a	0.305 ± 0.043ijk	0.631 ± 0.056fghi	0.16 ± 0.032jklm
CGS	44.627 ± 4.149a	0.295 ± 0.056gh	0.042 ± 0.005gh	2.248 ± 0.417c	3.187 ± 0.476a	0.754 ± 0.1ef
T6	8.47 ± 1.062e	0.432 ± 0.073ef	0.091 ± 0.011def	1.263 ± 0.194d	1.881 ± 0.35bc	0.672 ± 0.073f
T7	5.627 ± 0.737efgh	0.756 ± 0.07bc	0.032 ± 0.004gh	0.411 ± 0.045hijk	0.582 ± 0.049fghij	0.354 ± 0.056ghi
QM	2.695 ± 0.326hij	0.313 ± 0.029gh	0.041 ± 0.003gh	0.416 ± 0.032hij	0.412 ± 0.021ghijk	0.315 ± 0.033ghijk
BR60	2.217 ± 0.242ij	0.302 ± 0.032gh	0.059 ± 0.005fgh	0.378 ± 0.042hijk	0.382 ± 0.038ghijk	0.257 ± 0.025hijkl
S54	12.413 ± 1.943d	0.601 ± 0.065d	0.09 ± 0.009def	0.931 ± 0.09def	1.231 ± 0.159e	0.861 ± 0.114de
J4-1	12.969 ± 1.612d	0.923 ± 0.098a	0.25 ± 0.044ab	0.771 ± 0.074efg	1.555 ± 0.279cde	1.044 ± 0.143bc
J5	4.577 ± 0.509fghi	0.276 ± 0.041 h	0.053 ± 0.009fgh	0.666 ± 0.097fgh	0.619 ± 0.084fghi	0.367 ± 0.056gh
Y2	18.341 ± 2.335c	0.666 ± 0.092 cd	0.109 ± 0.014cde	1.165 ± 0.133d	1.739 ± 0.165bcd	0.934 ± 0.121 cd
YXM2	6.429 ± 0.437efg	0.282 ± 0.033gh	0.133 ± 0.024 cd	0.507 ± 0.047ghi	0.237 ± 0.029 k	0.136 ± 0.022 lm
YD	19.797 ± 2.348c	0.153 ± 0.026i	0.017 ± 0.002 h	3.57 ± 0.468b	1.332 ± 0.135e	0.336 ± 0.062ghij
Y1	29.099 ± 2.512b	0.367 ± 0.037efgh	0.035 ± 0.005gh	4.397 ± 0.352a	3.077 ± 0.363a	1.128 ± 0.184b

ND not detected

Different letters indicate significant differences in free phenolic compounds detected in the 23 mulberry leaf samples by HPLC (P < 0.05)

Values with no letters in common in each column are significantly different (P < 0.05), n = 3

Table 7

Bound phenolic compounds detected in the 23 mulberry leaf samples by HPLC

Compound(mg/g DW)	CaA	Iso	Ast	PrA	FeA
G1	0.125 ± 0.021j	0.086 ± 0.008fgh	0.049 ± 0.006ef	0.02 ± 0.003ef	0.013 ± 0.003abc
G6	0.133 ± 0.022j	0.08 ± 0.008fghi	0.068 ± 0.008d	0.016 ± 0.002efgh	0.005 ± 0.001jklm
G8	0.141 ± 0.024j	ND	0.027 ± 0.003ghij	ND	0.011 ± 0.002cdef
GSW	0.531 ± 0.04a	0.412 ± 0.04a	0.059 ± 0.008de	0.046 ± 0.007a	0.012 ± 0.003abcd
R2	0.317 ± 0.024def	0.25 ± 0.022c	0.185 ± 0.023a	0.016 ± 0.003efgh	0.011 ± 0.002bcde
7403	0.123 ± 0.018j	0.047 ± 0.005ij	0.04 ± 0.006 fg	0.018 ± 0.003defg	0.014 ± 0.003ab
DS	0.201 ± 0.021hj	0.044 ± 0.006ij	0.017 ± 0.003ij	0.013 ± 0.002gh	0.013 ± 0.002abc
B-2–8	0.263 ± 0.031 fg	0.127 ± 0.02e	0.072 ± 0.007d	0.003 ± 0j	0.004 ± 0klm
Z1	0.284 ± 0.035 fg	0.081 ± 0.011fghi	0.017 ± 0.003ij	0.023 ± 0.003 cd	0.007 ± 0.001hijkl
Z2	0.365 ± 0.039 cd	0.357 ± 0.035b	0.032 ± 0.005ghi	0.032 ± 0.004b	0.009 ± 0.001efghi
Z4	0.359 ± 0.038cde	0.255 ± 0.034c	0.036 ± 0.004fgh	0.028 ± 0.003bc	0.01 ± 0.001defg
CGS	0.383 ± 0.034c	0.075 ± 0.006ghi	0.022 ± 0.002hij	0.008 ± 0ij	0.007 ± 0.001ghijk
T6	0.198 ± 0.024hi	0.198 ± 0.028d	0.063 ± 0.005de	0.012 ± 0.002hi	0.009 ± 0.001defghi
T7	0.094 ± 0.01j	0.057 ± 0.007ghij	0.033 ± 0.003ghi	0.014 ± 0.002gh	0.014 ± 0.002abc
QM	0.108 ± 0.012j	0.046 ± 0.005ij	0.018 ± 0.002ij	0.016 ± 0.001efgh	0.007 ± 0ijklm
BR60	0.147 ± 0.017ij	0.062 ± 0.007ghij	0.022 ± 0.003hij	0.018 ± 0.002defg	0.004 ± 0 m
S54	0.307 ± 0.03ef	0.116 ± 0.013ef	0.028 ± 0.004ghij	0.016 ± 0.002fgh	0.008 ± 0fghi
J4-1	0.314 ± 0.026def	0.209 ± 0.025d	0.135 ± 0.019b	0.022 ± 0.003d	0.015 ± 0.002a
J5	0.138 ± 0.01j	0.051 ± 0.007hij	0.02 ± 0.001ij	0.012 ± 0.002hi	0.007 ± 0.001ghij
Y2	0.307 ± 0.031ef	0.22 ± 0.024 cd	0.099 ± 0.011c	0.021 ± 0.002de	0.009 ± 0.001efghi
YXM2	0.273 ± 0.022 fg	0.032 ± 0.004j	0.014 ± 0.002j	0.011 ± 0.002hi	0.01 ± 0.001defgh
YD	0.23 ± 0.02gh	0.06 ± 0.006ghij	0.016 ± 0.002j	0.003 ± 0j	0.008 ± 0.001ghij
Y1	0.468 ± 0.045b	0.09 ± 0.006efg	ND	0.006 ± 0.001j	0.004 ± 0 lm

ND not detected

Different letters indicate significant differences in bound phenolic compounds detected in the 23 mulberry leaf samples by HPLC(P < 0.05)

Values with no letters in common in each column are significantly different (P < 0.05), n = 3

As shown in Table 6, the leaves of the various mulberry cultivars contained different amounts of individual phenolic compounds. However, not all of the 11 phenolic compounds were detected in all samples; for example, BeA was only observed in G6 and G8, while Sco was only found in G1, R2, Z2, QM, BR60, J4-1, J5, Y1. Although none of the 23 cultivars contained all 11 components, J4-1,R2, Z2 and J5 were the richest in free phenolic compounds among the samples tested, with ten components detected. In contrast, YD contained the smallest variety of phenolic compounds, with only six of the components detected.

As shown in Table 7, CaA, Iso, Ast, PrA, FeA, and SyA accounted for the majority of the bound phenolic content for most of the samples, although Iso and PrA were not detected in G8 and SyA was not detected in Z4. Among these six main bound phenolic compounds, CaA was the major component. Previous studies have indicated that CaA, as an α-tocopherol protectant in low-density lipoprotein, is a superior antioxidant compared with FeA, which can also serve as a potent antioxidant to eliminate free radicals and singlet oxygen [15, 24].GSW, 7403, DS, Z1, Z2, T6, QM and J4-1were the richest among the 23 cultivars in terms of bound phenolic compounds.

The data presented in Tables 6 and 7 show that the total phenol contents of Rut (21.472 mg/g DW), Ast (12.700 mg/g DW), ChA (245.249 mg/g DW), and BeA(0.055 mg/g DW) were generally consistent with those determined by Zou et al. [40], who reported a Rut content of 0.1–0.7 mg/g DW, an Ast content of 0.1–0.5 mg/g DW, a ChA content of 0.9–2.1 mg/g DW, and a BeA content of 0–0.2 mg/g DW.

Combined with Tables 6 and 7, only 10 phenolic compounds could be found in the bound phenolic content, which was one phenolic kindless than free phenol. Seven phenolic compounds were detected in both the free and bound fractions, indicating that they occur in mulberry leaves in both forms. Similar to the results for the free phenolic compounds, BeA was detected in the bound fraction for samples G6, G8, and G1. Although the bound phenolic content was relatively low, it cannot be neglected, especially in the case of CaA, because the conjugated forms have been demonstrated to act as more powerful antioxidants in various systems [13, 24].

Antioxidant activity and its correlation with phenolic content

The antioxidant activities of the free phenolic fractions of the 23 mulberry leaf samples were determined using three separate assays (FRAP, ABTS, and DPPH). As shown in Table 8, the FRAP, ABTS, and DPPH values ranged from 35.13 μmol Fe2+/g DW (G1) to 227.8 μmol Fe2+/g DW (CGS), from 19.81 μmol TEAC/g DW (7403) to 120.42 μmol TEAC/g DW (Y2), and from 23.11 μmol AEAC/g DW (G1) to 256.63 μmol AEAC/g DW (CGS), respectively. The free phenolic content and antioxidant activity exhibited a certain degree of positive correlation.

Table 8

Antioxidant activities of the free phenolic fractions of the 23 mulberry leaf samples

Cultivar	FRAP(μmol Fe2+/g DW)	ABTS(μmol TEAC/g DW)	DPPH(μmol AEAC/g DW)
G1	35.13 ± 2.47 h	23.53 ± 1.66gh	23.11 ± 1.91i
G6	70.21 ± 5.63ef	32.98 ± 2.64 g	55.21 ± 4.43efg
G8	47.33 ± 4.95fgh	24.97 ± 2.57gh	33.87 ± 3.54ghi
GSW	108.14 ± 11.13d	45.52 ± 4.68f	102.77 ± 10.57d
R2	66.68 ± 5.95ef	30.04 ± 2.68gh	40.81 ± 3.64fghi
7403	41 ± 3.46gh	19.81 ± 1.67 h	26.37 ± 2.23c
DS	166.13 ± 16.59c	81.23 ± 8.11c	158.15 ± 15.79i
B-2–8	79.97 ± 9.18ef	35.74 ± 4.1 fg	41.23 ± 4.73fghi
Z1	40.72 ± 4.93gh	25.08 ± 3.04gh	30.04 ± 3.64hi
Z2	53.17 ± 5.95fgh	28.66 ± 3.21gh	42.01 ± 4.7fghi
Z4	55.91 ± 5.65efgh	32.27 ± 3.26gh	44 ± 4.45fghi
CGS	227.8 ± 23.25a	96.22 ± 9.82b	256.63 ± 26.19a
T6	129.14 ± 17.87d	63.25 ± 8.75d	100.27 ± 13.87d
T7	63.16 ± 7.85efg	28.73 ± 3.57gh	43.08 ± 5.36fghi
QM	41.26 ± 4.44gh	25.09 ± 2.7gh	37.15 ± 4ghi
BR60	105.83 ± 15.65d	48.07 ± 7.11ef	97.37 ± 14.4d
S54	45.76 ± 4.85fgh	24.06 ± 2.55gh	28.9 ± 3.07hi
J4-1	114.68 ± 14.8d	59.61 ± 7.69de	66.83 ± 8.62e
J5	61.98 ± 5.8efg	29.48 ± 2.76gh	42.59 ± 3.99fghi
Y2	129.24 ± 16.12d	120.42 ± 15.02a	61.05 ± 7.61ef
YXM2	51.81 ± 4.04fgh	27.41 ± 2.13gh	48.73 ± 3.8efgh
YD	196.81 ± 23.75b	82.42 ± 9.95c	182.94 ± 22.08b
Y1	221.99 ± 20.28a	94.89 ± 8.67b	247.82 ± 22.64a

As shown in Table 9, the FRAP values of the bound phenolic fractions of the 23 mulberry leaf samples ranged from 12.74 μmol Fe2+/g DW (T7) to 45.93 μmol Fe2+/g DW (Y2), which did not vary a lot probably. The ABTS values varied from 5.54 μmol TEAC/g DW (T7) to 18.27 μmol TEAC/g DW (T6), where T6 and T7 originated from the same country (Thailand). The DPPH values ranged from 5.11 μmol AEAC/g DW (T7) to 24.05 μmol AEAC/g DW (Y1).

Table 9

Antioxidant activities of the bound phenolic fractions of the 23 mulberry leaf samples

Cultivar	FRAP(μmol Fe2+/g DW)	ABTS(μmol TEAC/g DW)	DPPH(μmol AEAC/g DW)
G1	16.5 ± 2.16ij	9.34 ± 1.22fghi	9.85 ± 1.29fghi
G6	14.79 ± 1.98ij	6.52 ± 0.88jk	7.35 ± 0.99ijk
G8	14.52 ± 2.37ij	6.56 ± 1.07jk	7.23 ± 1.18ijk
GSW	32.26 ± 3.09 cd	13.37 ± 1.28 cd	14.31 ± 1.37de
R2	22.71 ± 2.71fgh	9.55 ± 1.14efgh	8.77 ± 1.05hij
7403	13.52 ± 2.04j	6.93 ± 1.05ijk	7.78 ± 1.18ijk
DS	37.2 ± 3.61bc	16.01 ± 1.55ab	20.89 ± 2.03b
B-2-8	39.15 ± 5.06bc	15.31 ± 1.98bc	17.36 ± 2.24c
Z1	17.03 ± 2.19hij	8.94 ± 1.15ghij	9.73 ± 1.25ghi
Z2	23.28 ± 2.55 fg	11.81 ± 1.29def	11.42 ± 1.25efgh
Z4	22.76 ± 2.66fgh	11.15 ± 1.3defg	9.95 ± 1.16fghi
CGS	29.45 ± 2.57de	12.38 ± 1.08d	15 ± 1.31 cd
T6	41.89 ± 5.11ab	18.27 ± 2.24a	22.52 ± 2.75ab
T7	12.74 ± 1.44j	5.54 ± 0.63 k	5.11 ± 0.58 k
QM	16.91 ± 1.92hij	8.94 ± 1.01ghij	10.02 ± 1.14fghi
BR60	26.21 ± 2.76ef	11.92 ± 1.26de	14.32 ± 1.51de
S54	13.65 ± 1.53j	6.92 ± 0.77ijk	5.99 ± 0.67jk
J4-1	28.47 ± 3.04def	13.31 ± 1.42 cd	12.72 ± 1.36def
J5	17.72 ± 1.71ghij	7.75 ± 0.75hijk	7.74 ± 0.75ijk
Y2	45.93 ± 4.83a	14.99 ± 1.58bc	20.48 ± 2.15b
YXM2	19.64 ± 1.86ghi	9.24 ± 0.88ghi	11.93 ± 1.13efg
YD	26.98 ± 2.45def	10.83 ± 0.98defg	14.31 ± 1.3de
Y1	37.86 ± 3.56bc	16.4 ± 1.54ab	24.05 ± 2.26a

Overall, the results of the three assays revealed a positive correlation between the free and bound phenolic contents and the antioxidant activities. Thus, the correlations between the antioxidant activities and the contents of individual phenolic compounds as determined by HPLC were evaluated to further examine the differences between the 23 cultivars.

Correlation between antioxidant activities and contents of individual phenolic compounds

Correlation analysis was conducted to determine whether any linear relationships existed between the antioxidant activities, total phenolic content, free phenolic content, bound phenolic content, and content of each phenolic component, and the results are summarized in Tables 10, 11, and 12. Owing to the diversity of the tested cultivars and differences in climate and other factors between different regions, the correlation of the different measured data was also different.

Table 10

Correlation coefficients (R2) for the linear relationships between the total phenolic content, free phenolic content, bound phenolic content and FRAP, DPPH, and ABTS activities

	FRAPa	ABTSb	DPPHc	FRAPd	ABTSe	DPPHf	PCtotalg	PCfreeh	PCboundi
FRAPa	1	0.8898	0.96	0.6751	0.6437	0.7375	0.795	0.7973	0.1861
ABTSb	–	1	0.7676	0.7876	0.6866	0.7878	0.7052	0.7059	0.2342
DPPHc	–	–	1	0.5208	0.5338	0.6443	0.7722	0.7755	0.1234
FRAPd	–	–	–	1	0.9532	0.9389	0.4926	0.4873	0.4582
ABTSe	–	–	–	–	1	0.9495	0.4193	0.4131	0.4749
DPPHf	–	–	–	–	–	1	0.4629	0.4606	0.2929
PCtotalg	–	–	–	–	–	–	1	0.9998	0.3899
PCfreeh	–	–	–	–	–	–	–	1	0.3732
PCboundi	–	–	–	–	–	–	–	–	1

aFerric reducing antioxidant power of the free phenolic extract

bABTS radical scavenging activity of the free phenolic extract

cDPPH radical scavenging activity of the free phenolic extract

dFerric reducing antioxidant power of the bound phenolic extract

eABTS radical scavenging activity of the bound phenolic extract

fDPPH radical scavenging activity of the bound phenolic extract

gTotal phenolic content

hFree phenolic content

iBound phenolic content

Table 11

Correlation coefficients (R2) for the linear relationships between the antioxidant activities and contents of each free phenolic component

Compound	ChA	Epi	CaA	Rut	Iso	Ast	PrA	Que	GaA	Sco	FRAP	ABTS	DPPH
ChA	1	0.06109	− 0.15278	0.74647	0.90622	0.62171	− 0.45845	− 0.04528	− 0.54529	0.58749	0.78381	0.69221	0.76781
Epi	–	1	0.43503	− 0.14782	0.23562	0.65714	− 0.41698	0.46368	− 0.09566	− 0.31809	− 0.13601	0.00211	− 0.27078
CaA	–	–	1	− 0.31581	− 0.08295	0.18897	− 0.12246	0.06754	0.05357	− 0.55668	− 0.17605	− 0.07144	− 0.23755
Rut	–	–	–	1	0.77122	0.46117	− 0.25405	− 0.10427	− 0.51705	0.86516	0.81368	0.66942	0.81723
Iso	–	–	–	–	1	0.7687	− 0.34297	0.02399	− 0.57575	0.60393	0.74951	0.67925	0.71518
Ast	–	–	–	–	–	1	− 0.31615	0.15592	− 0.42047	0.23725	0.37703	0.40483	0.27307
PrA	–	–	–	–	–	–	1	− 0.51773	− 0.08846	0.40536	− 0.22612	− 0.39174	0.03339
Que	–	–	–	–	–	–	–	1	0.2282	− 0.41755	− 0.18283	− 0.12509	− 0.22011
GaA	–	–	–	–	–	–	–	–	1	− 0.41773	− 0.52943	− 0.52334	− 0.4482
Sco	–	–	–	–	–	–	–	–	–	1	0.76458	0.7181	0.83016
FRAP	–	–	–	–	–	–	–	–	–	–	1	0.88975	0.96
ABTS	–	–	–	–	–	–	–	–	–	–	–	1	0.76762
DPPH	–	–	–	–	–	–	–	–	–	–	–	–	1

Table 12

Correlation coefficients (R2) for the linear relationships between the antioxidant activities and contents of each bound phenolic component

Compound	CaA	Iso	Ast	PrA	FeA	SyA	Que	Rut	GaA	BeA	FRAP	ABTS	DPPH
CaA	1	0.65768	0.25927	0.37507	− 0.09959	0.62789	0.55234	0.54925	0.22599	− 0.98429	0.46961	0.49066	0.40303
Iso	–	1	0.47181	0.74687	0.22374	0.18262	0.73828	0.43122	0.06553	1	0.32322	0.3395	0.12059
Ast	–	–	1	0.12189	0.25964	0.50799	0.46127	0.17347	− 0.20314	0.34391	0.2905	0.22378	0.09693
PrA	–	–	–	1	0.37708	− 0.33017	0.70274	− 0.01405	0.29171	1	− 0.09967	− 0.046	− 0.21309
FeA	–	–	–	–	1	− 0.3119	0.32327	− 0.07737	− 0.24806	0.47515	− 0.20614	− 0.17771	− 0.27261
SyA	–	–	–	–	–	1	0.09162	0.69018	− 0.17523	− 0.77855	0.6659	0.57881	0.64429
Que	–	–	–	–	–	–	1	0.38495	− 0.04223	–	0.16827	0.21116	0.05313
Rut	–	–	–	–	–	–	–	1	− 0.23894	–	0.72964	0.6797	0.67301
GaA	–	–	–	–	–	–	–	–	1	–	− 0.18081	− 0.11587	− 0.16261
BeA	–	–	–	–	–	–	–	–	–	1	0.98417	0.94582	0.963
FRAP	–	–	–	–	–	–	–	–	–	–	1	0.95325	0.93894
ABTS	–	–	–	–	–	–	–	–	–	–	–	1	0.94952
DPPH	–	–	–	–	–	–	–	–	–	–	–	–	1

As shown in Table 10, between the phenolic contents and the antioxidant activities of the various extracts, the strongest correlation was observed between the free phenolic content and the FRAP values of the free phenolic extracts (R2 = 0.7978), followed by the relationship between the total phenolic content and the FRAP values of the free phenolic extracts (R2 = 0.795). The weakest correlation occurred between the bound phenolic content and the DPPH values of the free phenolic extracts (R2 = 0.1234). Previous studies have indicated that phenolic compounds greatly contribute to the antioxidant activity of mulberry leaves, especially the free phenolic content, which has the greatest influence. A strong correlation was also observed between the bound phenolic content about Rut and the ABTS values of the bound phenolic extracts(R2 = 0.6797), indicating that certain phenolic compounds in the bound phenolic fraction may influence the removal of ABTS free radicals, such as CaA, Iso, and Que (Table 12). However, the correlation between total phenolic content and ABTS, free phenolic content and ABTS were the weakest, indicating that there may have been other substances present in the mulberry leaves that also affected the removal of ABTS free radicals, such as PrA (Table 11).

As shown in Tables 11 and 12, the contents of four phenolic compounds, namely, Iso, Rut, Sco, and ChA, exhibited positive correlations with the antioxidant activities of the free phenolic extracts as determined using all three assays, whereas the other seven phenolic compounds displayed no obvious correlation with the antioxidant activities. The strength of these positive correlations followed the order Rut > ChA > Sco > Iso for the FRAP activity, Sco > Rut > ChA > Iso for the DPPH activity, and Sco > ChA > Iso > Rut for the ABTS activity. However, as Sco was not detected in most of the samples, the free phenolic substances predominantly responsible for the antioxidant activities of the mulberry leaf samples appear to be Iso, ChA, and Rut. The correlation coefficient between the FRAP and DPPH assays was similar in Table 11, which indicates that these two assays were more suitable than the ABTS assay for measuring the antioxidant activity of free phenolic compounds. This may explain why the FRAP and DPPH assays have been more frequently applied to determine the phenolic antioxidant activity in previous studies [32, 39]. With respect to the bound phenolic compounds, SyA and Rut (Rut > SyA) displayed a significant positive correlation with the antioxidant activities of the bound phenolic extracts as determined using all three assays, whereas no obvious correlation was observed for the other compounds.

Overall, the correlation between the phenolic contents and antioxidant activities fluctuated greatly, reflecting that phenolic compounds are the main antioxidants in mulberry leaves.

Conclusions

The phenolic compositions and antioxidant activities of leaves from 23 mulberry varieties cultivated in several countries and regions were characterized using LC-ESI-QTOF and three separate antioxidant assays (FRAP, ABTS, and DPPH). The results revealed significant differences in the phytochemical contents and antioxidant activities between the samples. The phenolic compounds in mulberry leaves predominantly existed in the free form. CGS from Taiwan displayed the highest phenolic content as well as superior antioxidant activity compared with the other 22 cultivars, as determined by the FRAP, ABTS, and DPPH assays. Furthermore, the obtained results demonstrated that ChA and CaA were the main phenolic compounds in the free and bound fractions, respectively. Iso, ChA, and Rut accounted for the majority of the antioxidant activity in the free phenolic fractions, while SyA and Rut were positively correlated with the antioxidant activity of the bound phenolic fractions. Taken together, these results indicate that the antioxidant activity of mulberry leaves is related to the content and types of phenolic compounds present.