Antioxidant Capacity, Anticancer Ability and Flavonoids Composition of 35 Citrus (Citrus reticulata Blanco) Varieties.

Citrus (Citrus reticulate Blanco) is one of the most commonly consumed and widely distributed fruit in the world, which is possessing extensive bioactivities. Present study aimed to fully understand the flavonoids compositions, antioxidant capacities and in vitro anticancer abilities of different citrus resources. Citrus fruits of 35 varieties belonging to 5 types (pummelos, oranges, tangerines, mandarins and hybrids) were collected. Combining li quid chromatography combined with electrospray ionization mass spectrometry (LC-ESI-MS/MS) and ultra-performance liquid chromatography combined with diode array detector (UPLC-DAD), a total of 39 flavonoid compounds were identified, including 4 flavones, 9 flavanones and 26 polymethoxylated flavonoids (PMFs). Each citrus fruit was examined and compared by 4 parts, flavedo, albedo, segment membrane and juice sacs. The juice sacs had the lowest total phenolics, following by the segment membrane. Four antioxidant traits including 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging activity, ferric reducing antioxidant power (FRAP), oxygen radical absorbance capacity (ORAC) and cupric reducing antioxidant capacity (CUPRAC) were applied for the antioxidant capacities evaluation. Three gastric cancer cell lines, SGC-7901, BGC-823 and AGS were applied for the cytotoxicity evaluation. According to the results of correlation analysis, phenolics compounds might be the main contributor to the antioxidant activity of citrus extracts, while PMFs existing only in the flavedo might be closely related to the gastric cancer cell line cytotoxicity of citrus extracts. The results of present study might provide a theoretical guidance for the utilization of citrus resources.

1. Introduction

Citrus (Citrus reticulate Blanco) is a tropical or subtropical fruit widely distributed around the world. As one of the most consumed fruits it also has great economic importance. Besides its value as a delicious fruit, its nutritional values are also important. Previous studies have reported a variety of bioactivities of citrus fruit, like antioxidant [1], anticancer [2,3], anti-inflammation [4], anti-fat [5] and anti-diabetes properties [6,7]. Many of the bioactivities are attributed to the phenolics and flavonoids that are abundant in citrus fruit [5,8,9].

Citrus can be classified in several types, including mandarins, tangerines, oranges, pummelos, hybrids, lemons, limes, etc. [10]. Zhejiang Province, an important citrus production region in China, is rich in citrus germplasm resources. Several distinctive local characteristic citrus varieties had been found in this region, such as Ougan (OG), Yuhuanyou (YHY), Mantouhong (MTH), Huyou (HY) and Ponkan (PG). Thus, the citrus varieties in this region might provide some very distinctive resources for the bioactivity study of citrus fruits.

The present study aimed to carry out a comprehensive investigation on the flavonoid composition and distribution, antioxidant capacity and gastric tumor cytotoxicity of citrus fruits from Zhejiang Province. A total of 35 varieties belonging to five types of citrus fruits were collected for the study. Ultra-performance liquid chromatography (UPLC) and lipid chromatography combined with electrospray ionization mass spectrometry (LC-ESI-MS/MS) were used for identification and quantification of flavonoid compounds. 2,2-Diphenyl-1-picrylhydrazyl (DPPH) radical scavenging activity, ferric reducing antioxidant power (FRAP), oxygen radical absorbance capacity (ORAC) and cupric reducing antioxidant capacity (CUPRAC), four commonly used antioxidant tests, were applied for the antioxidant capacity evaluation. Gastric tumor cell lines were applied for the cytotoxicity evaluation of citrus flavonoid-rich extracts.

2. Results

2.1. Fruit Basic Quality Index of Different Citrus Varieties

As shown in Table 1, five types (pummelos, oranges, tangerines, mandarins and hybrids) of 35 varieties of citrus fruits were collected and their fruit basic qualities were tested. Significant differences were found in fruit color, weight, edible rate and total soluble solids (TSS) among the 35 tested citrus varieties. MTH showed the highest Citrus Color Index (CCI) value of 16.75, while Qingougan (QOG) showed the lowest CCI at −6.97. Pummelos showed higher weight than the other types of citrus fruits, among which YHY and Mabuwendan (MBWD) had the highest weights at 1754.10 g and 1676.14 g, respectively. Kouzhijin22 (KZJ22) had the lowest weight of 67.97 g. Hongmeiren (HMR) had the highest edible rate of 85.31%, while Zaoxiangyou (ZXY) only had an edible rate of 56.12%. TSS varied from 8.42 to 16.08 °Brix. KZJ22 had the highest TSS value of 16.08 °Brix, followed by Aiyuan31 (AY31), Rinan1 (RN1), Gongchuan (GOC), Mixiagan (MXG), Aiyuan27 (AY27) and Yuanxiaochun (YXC), while QOG and Wuheougan (WHOG) showed the lowest TSS. Correlation analysis showed that CCI and TSS had significant relationships with each other (r = 0.690, p < 0.01), indicating that citrus varieties with better color look may have better taste.

Table 1

Appearance and taste qualities of Citrus fruits of 35 cultivars. TSS = total soluble solids.

Cultivar (Abbreviation)	Fruit Type	Color (CCI)	Weight (g) *	Edible Rate (%)	TSS (°Brix) *
Aiyuan27 (AY27)	Hybrids	5.09 ± 0.09	397.97 ± 27.47 ef	84.18	12.85 ± 0.05 abcdef
Aiyuan30 (AY30)	Hybrids	14.10 ± 0.26	114.11 ± 11.15 lmno	80.06	12.51 ± 0.29 bcdefg
Aiyuan31 (AY31)	Hybrids	13.05 ± 0.32	385.06 ± 21.19 efg	84.54	14.79 ± 0.04 ab
Buzhihuo (BZH)	Hybrids	4.69 ± 0.38	216.62 ± 12.05 ij	75.95	10.91 ± 0.50 cdefghij
CaRaCaRa (CR)	Oranges	5.63 ± 1.85	159.36 ± 2.56 ijklm	75.12	11.87 ± 0.60 bcdefghi
Chunxiang (CX)	Hybrids	−4.95 ± 0.44	200.58 ± 9.23 ijk	67.34	9.17 ± 0.28 ghij
Dafen (DF)	Mandarins	3.29 ± 0.33	120.31 ± 12.15 klmno	79.82	10.49 ± 0.28 cdefghij
Gaocheng (GAC)	Hybrids	7.35 ± 0.09	343.87 ± 20.94 gh	74.51	12.44 ± 0.25 bcdefgh
Gongneiyiyugan (GN)	Hybrids	2.76 ± 0.66	241.55 ± 21.74 ij	74.14	10.21 ± 0.56 cdefghij
Gongchuan (GOC)	Mandarins	8.27 ± 0.33	141.94 ± 5.75 jklmno	82.51	13.23 ± 0.45 abcd
Hongmeiren (HMR)	Hybrids	7.59 ± 0.49	137.70 ± 4.03 jklmno	85.31	11.10 ± 0.08 cdefghij
Huyou (HY)	Pummelos	1.43 ± 0.80	571.61 ± 46.24 d	67.97	9.66 ± 0.08 efghij
Kouzhijin22 (KZJ22)	Hybrids	14.68 ± 0.22	67.97 ± 3.80 o	67.59	16.08 ± 0.71 a
Mabuwendan (MBWD)	Pummelos	1.22 ± 0.05	1676.14 ± 81.74 a	59.74	8.93 ± 0.08 cdefghij
Mantouhong (MTH)	Tangerines	16.75 ± 0.63	84.27 ± 3.26 no	79.89	12.53 ± 0.38 bcdefg
Mixiagan (MXG)	Hybrids	1.97 ± 0.08	303.59 ± 18.62 gh	72.13	12.96 ± 0.25 abcde
Nangan20 (NG20)	Mandarins	9.22 ± 0.12	100.61 ± 3.60 mno	74.07	11.76 ± 0.15 bcdeghi
Newhall (NH)	Oranges	8.72 ± 0.38	168.28 ± 5.15 ijklm	71.66	11.96 ± 0.70 bcdefghi
Ougan (OG)	Mandarins	−0.56 ± 1.09	175.42 ± 3.55 ijklm	64.12	9.46 ± 0.22 fghij
Ponkan (PG)	Tangerines	4.89 ± 0.94	140.61 ± 2.34 jklmno	74.68	9.08 ± 0.12 hij
Putaoyou (PTY)	Pummelos	0.24 ± 1.02	442.38 ± 21.67 e	77.49	9.56 ± 0.23 fghij
Qingji (QJ)	Hybrids	6.92 ± 0.20	170.73 ± 8.25 ijklm	73.18	10.55 ± 0.41 cdefghij
Qingougan (QOG)	Mandarins	−6.97 ± 0.44	146.74 ± 3.78 ijklmno	69.33	8.42 ± 0.05 j
Rinan1 (RN1)	Mandarins	8.99 ± 0.08	118.75 ± 5.82 klmno	76.04	13.48 ± 0.14 abc
Shangyexinxi (SYXX)	Mandarins	9.09 ± 0.20	228.56 ± 5.10 hi	80.21	10.34 ± 0.36 cdefghij
Shiwen (SW)	Mandarins	3.12 ± 0.30	74.16 ± 1.31 o	80.46	11.54 ± 0.09 bcdefghi
Sijiyou (SJY)	Pummelos	1.70 ± 0.19	1486.37 ± 21.38 b	65.72	9.68 ± 0.30 efghij
Tiancao (TC)	Hybrids	7.32 ± 0.90	138.75 ± 2.68 jklmno	85.24	9.89 ± 0.42 defghij
Tianchun (TCH)	Hybrids	2.86 ± 0.21	188.85 ± 6.50 ijkl	69.11	11.22 ± 0.18 cdefghij
Weizhang (WZ)	Mandarins	8.39 ± 0.54	128.33 ± 4.86 klmno	73.29	10.05 ± 0.22 defghij
Wuheougan (WHOG)	Mandarins	−1.09 ± 0.98	113.50 ±4.12 lmno	64.25	8.74 ± 0.18 ij
Youliang (YL)	Mandarins	10.45 ± 0.17	138.13 ± 12.12 jklmno	75.32	12.22 ± 0.50 bcdefgh
Yuanxiaochun (YXC)	Hybrids	1.34 ± 0.07	135.94 ± 6.26 jklmno	79.54	12.76 ± 0.42 abcdef
Yuhuanyou (YHY)	Pummelos	0.68 ± 0.08	1754.10 ± 76.73 a	73.10	10.92 ± 0.20 cdefghij
Zaoxiangyou (ZXY)	Pummelos	1.80 ± 0.18	1585.77 ± 76.21 b	56.12	11.38 ± 0.09 cdefghij

* Results were the mean ± SD (n = 15) on a fresh weight (FW) (g) and TSS (°Brix) of citrus fruit. Values within each column followed by different superscript letters were significantly different at p < 0.05 according to Tukey’s tests.

2.2. Total Phenolics and Antioxidant Capacities

Each citrus fruit was divided into four parts from outside to inside: flavedo, albedo, segment membrane and juice sacs. Ultrasonic assistant extraction was used to improve the extraction efficiency according to previous studies [11,12]. As shown in Table 2, Table 3, Table 4 and Table 5, the total phenolics contents varied between the varieties and tissues of citrus fruits.

Table 2

Total phenolics and antioxidant properties of citrus fruit flavedo of 35 cultivars.

Cultivars	Total Phenolics	DPPH	FRAP	ORAC	CUPRAC	APC Index *	Rank #
AY27	15.05 ± 0.56 no	11.36 ± 0.71 jklm	9.63 ± 0.44 hijk	153.73 ± 8.78 kl	29.35 ± 1.09 pqr	46.57	26
AY30	27.55 ± 0.29 a	20.73 ± 0.12 a	21.33 ± 1.31 a	319.01 ± 44.63 bc	54.80 ± 2.34 a	92.01	1
AY31	13.73 ± 0.36 op	9.99 ± 0.93 mno	7.12 ± 0.24 lmnop	188.39 ± 15.19 hijkl	32.85 ± 3.32 nop	45.42	28
BZH	19.68 ± 0.58 defg	16.91 ± 0.24 bc	12.36 ± 1.65 efg	221.30 ± 13.70 defghijkl	42.76 ± 0.59 defg	66.18	7
CR	18.90 ± 0.45 ghi	15.34 ± 0.80 bcdef	11.94 ± 0.48 efgh	224.91 ± 26.69 defghijk	42.69 ± 0.59 defg	63.96	10
CX	11.67 ± 0.40 q	5.48 ± 0.56 r	1.94 ± 0.16 s	162.38 ± 18.82 jkl	27.62 ± 0.34 qrs	30.14	35
DF	18.07 ± 0.41 ghijk	10.29 ± 0.83 lmn	7.93 ± 0.30 klmno	206.96 ± 18.53 efghijkl	37.07 ± 0.65 hijklmn	49.65	24
GAC	18.09 ± 0.38 ghijk	17.48 ± 1.04 b	14.80 ± 0.59 cd	185.27 ± 27.39 hijkl	35.14 ± 0.37 klmno	64.33	8
GN	16.44 ± 0.57 klmn	10.72 ± 0.23 lmn	4.85 ± 1.16 pqr	208.48 ± 30.42 efghijkl	36.08 ± 2.81 jklmn	46.19	27
GOC	19.58 ± 0.50 efg	13.46 ± 0.51 efghij	11.12 ± 0.30 efghi	163.88 ± 20.46 jkl	42.45 ± 1.68 defg	57.37	15
HMR	24.20 ± 0.69 b	20.40 ± 0.24 a	17.23 ± 0.29 b	270.31 ± 13.88 cdefg	51.74 ± 0.27 ab	82.81	3
HY	15.20 ± 0.60 mno	14.47 ± 0.60 defgh	8.67 ± 0.28 jklm	196.63 ± 16.56 ghijkl	33.55 ± 0.14 nop	53.40	21
KZJ22	21.11 ± 1.05 cde	14.92 ± 0.91 cdefg	12.51 ± 1.15 def	357.54 ± 0.53 b	43.78 ± 1.07 def	71.69	4
MBWD	13.13 ± 0.18 pq	9.17 ± 0.94 nopq	4.92 ± 0.45 pqr	178.20 ± 22.21 ijkl	25.62 ± 0.50 rs	38.01	32
MTH	21.65 ± 0.65 c	14.77 ± 0.68 cdefg	12.82 ± 0.48 de	231.29 ± 39.15 defghij	41.82 ± 1.70 efgh	64.25	9
MXG	14.86 ± 0.40 no	13.04 ± 0.25 ghijk	9.98 ± 0.96 ghijk	75.78 ± 35.22 m	24.50 ± 0.86 s	42.64	30
NG20	19.30 ± 0.11 fgh	13.39 ± 0.64 efghij	9.43 ± 0.55 ijkl	288.99 ± 3.61 bcd	37.61 ± 0.26 hijklmn	59.76	13
NH	20.60 ± 0.35 cdef	16.50 ± 0.27 bcd	12.82 ± 0.73 de	213.25 ± 26.59 efghijkl	45.70 ± 1.07 cde	67.14	5
OG	19.04 ± 0.60 fghi	11.44 ± 0.83 jklm	9.59 ± 0.49 hijk	249.79 ± 21.33 cdefghi	41.04 ± 0.60 efghi	57.08	16
PG	18.36 ± 0.22 ghij	14.39 ± 0.37 defgh	11.31 ± 1.08 efghi	250.70 ± 42.77 cdefghi	39.87 ± 1.39 fghijk	62.16	11
PTY	17.20 ± 0.76 jkl	15.54 ± 0.58 bcde	7.71 ± 0.45 klmno	201.56 ± 18.04 fghijkl	34.19 ± 0.78 mno	54.12	19
QJ	21.45 ± 0.54 c	16.36 ± 0.06 bcd	10.36 ± 0.73 fghij	256.04 ± 37.59 cdefgh	46.81 ± 2.56 cd	66.88	6
QOG	18.29 ± 0.36 ghij	10.62 ± 0.34 lmn	8.42 ± 0.59 jklmn	277.70 ± 15.99 cde	40.36 ± 0.96 fghij	55.89	17
RN1	16.33 ± 0.48 lmn	12.41 ± 0.95 hijkl	8.52 ± 0.71 jklm	214.83 ± 20.09 defghijkl	30.71 ± 1.99 opq	50.41	23
SJY	12.18 ± 0.32 pq	7.80 ± 0.55 pq	3.87 ± 0.07 rs	149.67 ± 4.81 lm	24.34 ± 0.32 s	33.03	34
SW	21.32 ± 0.60 cd	13.24 ± 0.92 fghij	10.39 ± 0.76 fghij	242.22 ± 28.50 defghi	40.02 ± 1.70 fghij	59.31	14
SYXX	16.85 ± 0.52 jklm	11.74 ± 0.81 ijklm	8.35 ± 0.83 jklmn	209.35 ± 24.08 efghijkl	36.65 ± 1.81 ijklmn	51.83	22
TC	24.80 ± 0.55 b	19.94 ± 0.30 a	16.88 ± 1.60 bc	468.96 ± 34.84 a	49.16 ± 2.51 bc	91.26	2
TCH	16.82 ± 0.31 jklm	10.04 ± 0.50 1mno	5.57 ± 0.24 opqr	227.70 ± 11.03 defghijk	38.84 ± 0.88 ghijklm	48.49	25
WHOG	17.64 ± 0.37 hijkl	11.02 ± 0.23 klmn	7.72 ± 0.62 klmno	256.28 ± 28.81 cdefgh	39.12 ± 2.01 fghijkl	53.85	20
WZ	17.66 ± 0.14 hijkl	11.77 ± 0.53 ijklm	9.81 ± 0.53 hijk	210.70 ± 16.96 efghijkl	40.07 ± 1.88 fghij	55.20	18
YHY	13.57 ± 0.17 op	9.76 ± 0.63 mnop	6.05 ± 0.53 nopqr	158.18 ± 8.30 jkl	26.28 ± 1.21 qrs	39.28	31
YL	17.57 ± 0.99 ijkl	13.75 ± 1.56 efghij	9.22 ± 0.89 ijkl	275.85 ± 27.24 cdef	41.02 ± 0.88 efghi	60.81	12
YXC	15.49 ± 0.73 mn	7.89 ± 0.80 opq	6.64 ± 0.04 mnopq	188.26 ± 8.62 hijkl	34.69 ± 0.80 lmno	43.16	29
ZXY	11.52 ± 0.03 q	7.40 ± 0.13 qr	4.70 ± 0.08 qr	158.83 ± 17.20 jkl	24.44 ± 2.01 s	34.05	33

Results were the mean ± SD (n = 3) on a dried weight (g) of citrus basis. Total phenolics were calculated as mg gallic acid equivalents (GAE)/g DW. Antioxidant capacities (DPPH, FRAP, ORAC and CUPRAC) were calculated as mg trolox equivalent antioxidant capacities (TEAC)/g DW. Values within each column followed by different superscript letters were significantly different (p < 0.05) according to Tukey’s tests. * Antioxidant index score = [(sample score/best score) × 100], averaged for all four tests for each cultivar for the antioxidant potency composite (APC) index. # Ranked according to the APC index.

Table 3

Total phenolics and antioxidant properties of citrus fruit albedo of 35 cultivars.

Cultivars	Total Phenolics	DPPH	FRAP	ORAC	CUPRAC	APC Index *	Rank #
AY27	15.19 ± 0.73 fghijklm	8.22 ± 0.38 defghi	15.01 ± 0.08 d	55.49 ± 7.65 m	18.30 ± 0.83 mnop	38.40	22
AY30	19.84 ± 1.30 bcd	16.53 ± 0.49 a	18.45 ± 0.88 c	288.00 ± 39.86 efg	36.77 ± 0.89 cd	72.09	2
AY31	11.68 ± 1.28 lmn	4.73 ± 0.35 kl	8.24 ± 0.68 ijkl	73.31 ± 2.11 lm	16.26 ± 0.55 opq	26.31	33
BZH	14.45 ± 2.04 ghijklmn	10.62 ± 1.57 cd	11.64 ± 0.92 efg	220.49 ± 24.87 ghijk	31.25 ± 1.49 efg	51.52	11
CR	13.46 ± 0.15 ijklmn	5.88 ± 0.74 ijkl	7.28 ± 0.27 jklmno	232.58 ± 11.48 ghij	17.42 ± 0.21 nop	33.49	26
CX	10.35 ± 0.48 n	4.37 ± 0.69 l	5.68 ± 0.24 lmno	177.82 ± 13.67 hijkl	22.21 ± 1.10 jklm	30.33	32
DF	13.13 ± 0.44 jklmn	5.00 ± 0.36 jkl	7.12 ± 0.03 jklmno	417.66 ± 22.58 cd	22.77 ± 0.44 jkl	41.79	20
GAC	27.15 ± 3.53 a	17.43 ± 0.93 a	25.26 ± 2.77 a	420.79 ± 53.79 cd	39.34 ± 3.01 bc	86.39	1
GN	15.74 ± 1.16 defghijkl	7.00 ± 0.81 fghijk	7.24 ± 0.34 jklmno	429.84 ± 62.99 cd	26.20 ± 1.09 hij	47.04	15
GOC	16.12 ± 1.71 defghijk	6.97 ± 0.57 fghijk	8.06 ± 0.88 jkl	382.81 ± 24.59 de	23.24 ± 0.15 jk	44.48	16
HMR	17.78 ± 1.65 cdefgh	11.65 ± 0.26 bc	14.50 ± 0.44 d	358.36 ± 33.63 def	33.35 ± 0.31 def	62.03	7
HY	17.59 ± 0.15 cdefg	8.60 ± 1.49 defg	6.80 ± 0.51 jklmno	424.62 ± 24.53 cd	25.56 ± 0.60 ij	48.33	14
KZJ22	16.39 ± 0.63 defghij	6.90 ± 0.67 fghijk	11.84 ± 0.46 ef	245.27 ± 4.35 g	23.53 ± 0.75 opq	43.21	17
MBWD	13.82 ± 0.27 hijklmn	7.37 ± 0.67 fghij	8.02 ± 0.19 jkl	206.45 ± 15.84 ghijk	14.13 ± 2.06 pqr	33.60	25
MTH	18.96 ± 0.50 cdef	13.92 ± 0.55 b	17.77 ± 1.23 c	215.75 ± 28.81 ghijk	34.77 ± 1.47 de	63.98	6
MXG	14.97 ± 2.42 fghijklm	5.13 ± 0.06 jkl	10.71 ± 0.64 fg	177.63 ± 2.90 hijkl	12.12 ± 0.39 qr	30.94	31
NG20	15.35 ± 0.27 fghijklm	4.86 ± 0.43 jkl	9.22 ± 0.17 ghij	176.15 ± 30.27 hijkl	20.09 ± 0.61 klmno	33.32	27
NH	13.73 ± 0.57 hijklmn	6.55 ± 0.33 ghijkl	7.67 ± 0.19 jklm	116.37 ± 1.22 klm	22.21 ± 1.39 jklm	33.12	29
OG	20.98 ± 0.60 bc	8.96 ± 1.10 defg	13.28 ± 1.25 de	549.63 ± 73.64 b	46.43 ± 1.98 a	71.10	4
PG	15.49 ± 0.57 efghijklm	17.65 ± 0.26 a	21.56 ± 1.56 b	179.13 ± 2.73 hijkl	34.96 ± 2.78 de	71.77	3
PTY	24.00 ± 1.50 ab	12.28 ± 0.33 bc	11.57 ± 0.34 efgh	508.19 ± 85.71 bc	33.86 ± 1.85 def	65.82	5
QJ	15.49 ± 0.72 efghijklm	10.23 ± 0.82 cde	11.51 ± 0.36 efgh	198.88 ± 12.12 ghijk	28.14 ± 0.85 g	48.37	13
QOG	19.69 ± 1.54 cde	8.49 ± 0.86 defgh	10.65 ± 0.56 fg	404.23 ± 4.79 cd	42.67 ± 1.95 ab	60.45	8
RN1	14.18 ± 1.50 higklmn	8.04 ± 0.78 efghi	11.89 ± 0.52 ef	126.91 ± 13.19 jklm	20.40 ± 0.25 klmno	38.82	21
SJY	14.37 ± 1.23 higklmn	7.31 ± 0.74 fghij	7.53 ± 0.33 jklmn	224.79 ± 19.36 ghij	12.41 ± 1.49 qr	32.78	30
SW	15.54 ± 0.81 efghijklm	7.35 ± 0.99 fghij	9.12 ±0.83 ghij	412.33 ± 29.49 cd	26.11 ± 0.83 hij	48.70	12
SYXX	13.41 ± 1.48 ijklmn	5.22 ± 0.73 jkl	6.51 ± 0.29 klmno	360.47 ± 49.12 de	18.66 ± 1.6l mno	37.18	23
TC	16.12 ± 1.82 defghijk	8.56 ± 0.66 defg	9.04 ± 0.48 hijk	677.90 ± 51.50 a	22.61 ± 0.27 jkl	58.25	9
TCH	11.37 ± 1.58 mn	4.75 ± 0.44 kl	5.42 ± 0.60 mno	253.81 ± 16.73 fg	22.83 ± 1.49 jkl	33.75	24
WHOG	18.67 ± 1.05 cdefg	8.59 ± 0.44 defg	10.66 ± 0.42 fg	282.05 ± 78.31 efgh	41.25 ± 1.21 b	55.33	10
WZ	11.68 ± 1.24 lmn	4.72 ± 0.18 kl	6.33 ± 0.56 lmno	248.27 ± 30.14 g	20.81 ± 0.62 klmn	33.31	28
YHY	16.85 ± 0.57 cdefghij	4.56 ± 2.19 kl	4.82 ± 0.37 o	198.22 ± 10.99 ghijk	11.92 ± 1.59 r	24.96	35
YL	13.91 ± 0.64 higklmn	5.92 ± 0.17 ijkl	7.31 ± 0.84 jklmno	436.49 ± 14.06 cd	21.00 ± 0.18 klmn	43.02	19
YXC	12.06 ± 1.41 klmn	9.11 ± 0.70 def	11.30 ± 1.01 efgh	76.45 ± 8.52 lm	30.19 ± 1.02 fgh	43.16	18
ZXY	14.00 ± 0.39 higklmn	6.09 ± 0.74 hijkl	5.07 ± 0.70 no	168.30 ± 11.89 ijkl	11.34 ± 1.08 r	25.96	34

Results were the mean ± SD (n = 3) on a dried weight (g) of citrus basis. Total phenolics were calculated as mg gallic acid equivalents (GAE)/g DW. Antioxidant capacities (DPPH, FRAP, ORAC and CUPRAC) were calculated as mg trolox equivalent antioxidant capacities (TEAC)/g DW. Values within each column followed by different superscript letters were significantly different (p < 0.05) according to Tukey’s tests. * Antioxidant index score = [(sample score/best score) × 100], averaged for all four tests for each cultivar for the antioxidant potency composite (APC) index. # Ranked according to the APC index.

Table 4

Total phenolics and antioxidant properties of citrus fruit segment membrane of 35 cultivars.

Cultivars	Total Phenolics	DPPH	FRAP	ORAC	CUPRAC	APC Index *	Rank #
AY27	8.84 ± 0.10 cdefgh	5.12 ± 0.53 defghijk	15.92 ± 0.77 cd	49.03 ± 7.40 pq	5.03 ± 0.20 def	63.12	18
AY30	8.59 ± 0.21 efgh	6.84 ± 0.98 ab	19.13 ± 0.70 ab	102.13 ± 6.06 jklmn	6.13 ± 0.09 ab	83.70	4
AY31	7.84 ± 0.38 hi	3.15 ± 0.29 opq	10.60 ± 0.08 klm	69.27 ± 5.44 nopq	3.16 ± 0.23 lmn	44.79	33
BZH	8.90 ± 0.16 cdefgh	5.61 ± 0.15 cdef	15.41 ± 0.28 cde	129.18 ± 10.93 fghijk	4.77 ± 0.26 defghi	72.72	8
CR	8.18 ± 0.67 gh	3.62 ± 0.38 lmnopq	9.40 ± 0.91 mno	98.24 ± 3.84 klmno	4.05 ± 0.33 ghijkl	51.83	30
CX	6.27 ± 0.29 j	3.55 ± 0.03 mnopq	8.30 ± 0.27 o	77.50 ± 5.45 lmnop	2.57 ± 0.17 n	41.92	34
DF	8.71 ± 0.18 defgh	5.25 ± 0.50 defghi	11.69 ± 0.07 ijkl	132.53 ± 8.86 efghijk	3.91 ± 0.30 ijklm	63.83	17
GAC	12.56 ± 0.70 a	5.73 ± 0.12 bcde	19.83 ± 0.90 a	183.38 ± 5.16 ab	6.24 ± 0.34 ab	90.96	1
GN	9.70 ± 0.32 bcdef	3.67 ± 0.17 lmnopq	9.96 ± 0.16 lmno	164.49 ± 10.28 bcdef	3.30 ± 0.23 klmn	57.70	25
GOC	9.44 ± 0.43 bcdefg	5.27 ± 0.17 defgh	12.83 ± 0.49 ghij	139.82 ± 14.40 defghi	3.89 ± 0.25 ijklm	66.13	12
HMR	10.63 ± 0.33 b	7.34 ± 0.39 a	17.32 ± 0.13 ab	133.80 ± 7.90 efghijk	5.45 ± 0.21 bcd	84.24	3
HY	9.88 ± 0.74 bcde	4.80 ± 0.52 efghijkl	13.53 ± 0.89 efghi	171.10 ± 19.05 bcd	6.37 ± 0.18 a	78.89	5
KZJ22	8.55 ± 0.60 efgh	2.96 ± 0.22 pq	10.39 ± 0.18 klmn	98.58 ± 4.77 jklmno	3.01 ± 0.17 mn	46.79	32
MBWD	8.67 ± 0.12 defgh	4.55 ± 0.55 efghijklmn	8.50 ± 0.94 no	147.93 ± 12.91 cdefg	4.60 ± 0.46 defghi	61.97	19
MTH	7.64 ± 0.34 hij	5.73 ± 0.05 bcde	13.26 ± 0.65 ghi	72.25 ± 5.65 mnopq	4.13 ± 0.08 ghijk	61.09	20
MXG	7.77 ± 0.25 hi	4.34 ± 0.15 hijklmno	15.78 ± 0.92 cd	59.10 ± 6.67 pq	4.85 ± 0.29 defgh	60.78	21
NG20	8.84 ± 0.37 cdefgh	3.42 ± 0.47 nopq	9.96 ± 0.24 lmno	139.79 ± 19.17 defghi	3.43 ± 0.03 jklmn	54.40	28
NH	9.55 ± 0.63 bcdefg	4.50 ± 0.54 fghijklmn	14.00 ± 0.35 defgh	111.73 ± 2.57 hijkl	4.64 ± 0.23 defghi	64.56	16
OG	10.63 ± 0.56 b	5.29 ± 0.34 defgh	10.80 ± 0.42 jklm	141.78 ±12.77 defgh	2.73 ± 0.09 n	59.32	23
PG	7.72 ± 0.38 hij	6.31 ± 0.05 abcd	14.59 ± 0.83 defg	71.47 ± 6.54 mnopq	4.23 ± 0.04 fghij	65.04	14
PTY	9.66 ± 0.46 bcdef	3.91 ± 0.17 klmopq	14.86 ± 0.79 defg	133.13 ± 6.13 efghijk	4.60 ± 0.17 defghi	66.04	13
QJ	8.70 ± 0.39 defgh	5.64 ± 0.30 bcdef	13.33 ± 0.13 fghi	104.83 ± 9.21 ijklm	5.12 ± 1.03 def	68.66	10
QOG	9.06 ± 0.71 cdefgh	4.44 ± 0.47 fghijklmn	9.52 ± 0.35 mno	123.99 ± 3.26 ghijk	3.07 ± 0.13 mn	54.01	29
RN1	8.90 ± 0.58 cdefgh	3.95 ± 0.25 jklmnop	11.64 ± 0.71 ijkl	106.21 ± 13.69 ijklm	3.98 ± 0.28 hijkl	56.46	27
SJY	8.49 ± 0.57 efgh	3.52 ± 0.43 mnopq	9.78 ± 0.63 lmno	178.73 ± 28.31 abc	4.88 ± 0.12 defg	64.86	15
SW	9.01 ± 0.13 cdefgh	5.57 ± 0.51 cdefg	14.63 ± 1.00 defg	152.56 ± 10.44 bcdefg	4.42 ± 0.09 efghi	73.02	7
SYXX	9.70 ± 0.18 bcdef	3.50 ± 0.33 mnopq	9.51 ± 0.27 mno	166.46 ± 2.43 bcde	4.00 ± 0.07 ghijkl	59.53	22
TC	9.58 ± 0.69 bcdefg	6.64 ± 0.53 abc	18.86 ± 1.38 ab	170.14 ± 2.32 bcd	6.02 ± 0.08 abc	90.38	2
TCH	6.42 ± 0.17 ij	2.72 ± 0.18 q	9.37 ± 0.27 mno	64.69 ± 9.16 opq	2.87 ± 0.10 n	40.08	35
WHOG	8.44 ± 0.47 efgh	4.12 ± 0.33 hijklmnop	9.50 ± 0.83 mno	101.30 ± 10.04 jklmn	3.06 ± 0.10 mn	50.14	31
WZ	8.36 ± 0.57 fgh	4.65 ± 0.24 efghijklm	10.00 ± 0.71 lmno	131.91 ± 17.31 efghijk	3.36 ± 0.03 jklmn	57.42	26
YHY	10.10 ± 0.34 bcd	4.06 ± 0.31 ijklmnop	10.82 ± 0.61 jklm	208.86 ± 0.33 a	5.13 ± 0.12 cde	72.60	9
YL	10.20 ± 0.24 bc	4.38 ± 0.21 ghijklmn	12.24 ± 0.28 hijk	153.20 ± 25.10 bcdefg	4.63 ± 0.28 defghi	66.86	11
YXC	6.66 ± 0.43 ij	4.82 ± 0.47 efghijkl	15.33 ± 0.52 cdef	37.26 ± 3.92 q	4.66 ± 0.43 defghi	58.49	24
ZXY	8.96 ± 0.61 cdefgh	5.14 ± 0.23 defghij	9.95 ± 0.10 lmno	179.93 ± 8.53 abc	6.30 ± 0.19 ab	76.31	6

Results were the mean ± SD (n = 3) on a dried weight (g) of citrus basis. Total phenolics were calculated as mg gallic acid equivalents (GAE)/g DW. Antioxidant capacities (DPPH, FRAP, ORAC and CUPRAC) were calculated as mg trolox equivalent antioxidant capacities (TEAC)/g DW. Values within each column followed by different superscript letters were significantly different (p < 0.05) according to Tukey’s tests. * Antioxidant index score = [(sample score/best score) × 100], averaged for all four tests for each cultivar for the antioxidant potency composite (APC) index. # Ranked according to the APC index.

Table 5

Total phenolics and antioxidant properties of citrus fruit juice sacs of 35 cultivars.

Cultivars	Total Phenolics	DPPH	FRAP	ORAC	CUPRAC	APC Index *	Rank #
AY27	6.35 ± 0.13 a	4.73 ± 0.15 bcd	13.00 ± 1.09 a	62.32 ± 7.68 bc	5.01 ± 0.27 cde	84.23	1
AY30	6.10 ± 0.20 abc	5.02 ± 0.36 abc	8.54 ± 0.37 defgh	32.26 ± 2.36 ghijk	6.04 ± 0.20 ab	70.68	8
AY31	4.75 ± 0.21 ghijk	2.09 ± 0.27 opq	7.42 ± 0.56 ghijkl	40.59 ± 2.64 efg	2.85 ± 0.16 mno	47.04	31
BZH	5.25 ± 0.13 defghij	4.29 ± 0.18 cdefgh	8.36 ± 0.43 defghij	20.35 ± 0.70 l	6.32 ± 1.08 ab	64.22	12
CR	5.84 ± 0.21 abcdef	4.54 ± 0.12 cdefg	8.27 ± 0.39 defghij	53.06 ± 4.85 cd	4.90 ± 0.20 de	70.85	7
CX	4.18 ± 0.10 k	1.66 ± 0.03 pq	4.28 ± 0.34 o	27.56 ± 1.35 ijkl	2.51 ± 0.04 no	33.60	34
DF	5.89 ± 0.04 abcde	3.74 ± 0.28 ghijk	7.07 ± 0.55 hijklm	38.83 ± 2.42 efgh	3.97 ± 0.10 fghijkl	57.01	20
GAC	6.06 ± 0.39 abcd	4.41 ± 0.18 cdefg	11.96 ± 0.53 ab	54.12 ± 4.41 cd	4.58 ± 0.20 defghi	76.58	5
GN	5.27 ± 0.05 cdefghij	3.58 ± 0.26 hijk	6.60 ± 0.28 jklmn	28.76 ± 1.54 hijkl	4.53 ± 0.51 defghij	54.09	26
GOC	5.42 ± 0.18 bcdefghi	3.80 ± 0.21 fghij	6.74 ± 0.46 ijklmn	29.49 ± 1.13 ghijkl	3.87 ± 0.23 ghijkl	53.19	28
HMR	5.27 ± 0.15 cdefghij	3.23 ± 0.08 ijkl	7.45 ± 0.61 ghijkl	28.78 ± 1.70 hijkl	4.47 ± 0.05 defghij	53.99	27
HY	5.98 ± 0.31 abcd	5.40 ± 0.21 ab	10.04 ± 0.65 cd	27.64 ± 0.39 hijkl	6.96 ± 0.09 a	77.00	4
KZJ22	5.70 ± 0.29 abcdef	1.46 ± 0.15 q	5.63 ± 0.33 mno	75.63 ± 3.81 a	2.30 ± 0.19 o	50.46	30
MBWD	4.80 ± 0.14 ghijk	4.22 ± 0.20 cdefgh	7.61 ± 0.29 ghijkl	20.15 ± 3.74 l	4.53 ± 0.19 defghij	55.98	22
MTH	5.39 ± 0.39 bcdefghi	3.58 ± 0.15 hijk	8.36 ± 0.43 defghij	35.82 ± 3.47 ghi	3.88 ± 0.27 ghijkl	57.47	19
MXG	5.02 ± 0.11 fghij	2.41 ± 0.11 mnop	7.43 ± 0.34 ghijkl	48.81 ± 3.20 de	3.17 ± 0.30 lmno	52.32	29
NG20	5.44 ± 0.03 bcdefghi	4.25 ± 0.18 cdefgh	11.9 ± 1.19 ab	26.84 ± 0.92 ijkl	4.77 ± 0.14 defg	67.43	9
NH	6.39 ± 0.42 a	4.74 ± 0.51 bcd	9.13 ± 1.08 cdefg	57.02 ± 2.10 bcd	5.87 ± 0.13 bc	78.17	3
OG	6.18 ± 0.39 ab	3.25 ± 0.15 ijkl	5.97 ± 0.81 lmno	55.16 ± 7.92 bcd	3.62 ± 0.20 jklm	56.90	21
PG	6.00 ± 0.23 abcd	4.17 ± 0.16 defgh	10.32 ± 0.46 bc	36.58 ± 1.61 ghi	4.36 ± 0.23 defghij	65.79	10
PTY	5.84 ± 0.11 abcedf	4.28 ± 0.27 cdefgh	7.57 ± 0.29 ghijkl	60.32 ± 7.35 bc	5.02 ± 0.29 cde	71.20	6
QJ	5.10 ± 0.26 efghij	4.68 ± 0.22 bcde	7.82 ± 0.14 fghijk	22.59 ± 1.15 kl	5.10 ± 0.30 cd	61.24	14
QOG	5.91 ± 0.35 abcde	2.68 ± 0.36 lmno	6.74 ± 0.12 ijklmn	48.80 ± 2.53 de	4.14 ± 0.15 efghijk	55.66	23
RN1	4.73 ± 0.42 ghijk	3.31 ± 0.15 ijkl	9.04 ± 0.21 cdefg	37.03 ± 4.50 fghi	4.15 ± 0.01 efghijk	58.97	16
SJY	4.62 ± 0.15 ijk	4.57 ± 0.09 cdef	8.44 ± 0.39 defghi	26.43 ± 4.43 ijkl	4.81 ± 0.31 def	62.18	13
SW	5.90 ± 0.42 abcde	4.01 ± 0.20 defghi	7.95 ± 0.05 efghijk	35.24 ± 0.94 ghij	4.63 ± 0.10 defghi	61.06	15
SYXX	5.49 ± 0.10 bcdefgh	3.51 ± 0.26 hijk	6.18 ± 0.24 klmn	53.77 ± 1.61 cd	3.80 ± 0.18 ijkl	58.62	17
TC	6.20 ± 0.17 ab	5.73 ± 0.62 a	9.73 ± 0.58 cde	35.38 ± 5.10 ghi	6.54 ± 0.22 ab	78.90	2
TCH	4.63 ± 0.31 hijk	2.23 ± 0.26 nopq	5.19 ± 0.51 no	24.15 ± 1.38 jkl	3.12 ± 0.20 lmno	38.90	33
WHOG	6.14 ± 0.46 ab	3.04 ± 0.03 jklm	6.29 ± 0.02 klmn	47.89 ± 1.22 def	3.84 ± 0.12 hijkl	54.98	24
WZ	4.67 ± 0.20 ghijkl	2.98 ± 0.21 klmn	5.83 ± 0.15 lmno	23.84 ± 1.42 kl	3.38 ± 0.04 klmn	44.23	32
YHY	4.83 ± 0.13 ghijk	4.61 ± 0.55 bcde	9.48 ± 1.15 cdef	27.52 ± 3.15 ijkl	4.74 ± 0.14 defgh	64.47	11
YL	5.46 ± 0.28 bcdefghi	2.97 ± 0.10 klmn	6.23 ± 0.57 klmn	65.67 ± 2.39 ab	3.29 ± 0.26 klmn	58.46	18
YXC	4.36 ± 0.11 k	2.79 ± 0.10 lmno	7.05 ± 0.43 hijklm	24.17 ± 1.26 jkl	4.54 ± 0.47 defghij	50.03	35
ZXY	4.53 ± 0.15 jk	3.90 ± 0.23 efghi	6.78 ± 0.28 hijklmn	27.78 ± 1.10 hijkl	4.35 ± 0.42 defghij	54.86	25

Results were the mean ± SD (n = 3) on a dried weight (g) of citrus basis. Total phenolics were calculated as mg gallic acid equivalents (GAE)/g DW. Antioxidant capacities (DPPH, FRAP, ORAC and CUPRAC) were calculated as mg trolox equivalent antioxidant capacities (TEAC)/g DW. Values within each column followed by different superscript letters were significantly different (p < 0.05) according to Tukey’s tests. * Antioxidant index score = [(sample score/best score) × 100], averaged for all four tests for each cultivar for the antioxidant potency composite (APC) index. # Ranked according to the APC index.

Among the four citrus fruit parts, the juice sacs had the lowest total phenolics contents (4.18 mg gallic acid equivalents (GAE)/g dry weight of extracts (DW) in CX to 6.39 mg GAE/g DW in Newhall (NH)), followed by the segment membrane (6.27 mg GAE/g DW in CX to 12.56 mg GAE/g DW in GAC) in all the 35 varieties. The total phenolics contents in the flavedo ranged from 11.52 mg GAE/g DW in ZXY to 27.55 mg GAE/g DW in AY30. While the total phenolics contents in the albedo ranged from 10.35 mg GAE/g DW in CX to 27.15 mg GAE/g DW in GAC. For the flavedo and albedo part, 12 citrus varieties had higher total phenolics contents in flavedo than in albedo, including six pummelos (HY, YHY, MBWD, Zaoxiangyou (ZXY), Putaoyou (PTY), Sijiyou (SJY)), three hybrids (AY27, GOC, MXG) and three tangerines (OG, QOG, WHOG). Zhang et al. [13] had previously tested the total phenolics of 14 wild mandarin genotypes and two cultivars, which ranged from 29.38 to 51.14 mg GAE/g DW. The total phenolics contents of the sum of flavedo and albedo (22.02 to 47.39 mg GAE/g DW) were close to the results of Zhang et al. [13].

The DPPH radical scavenging activity, FRAP, ORAC and CUPRAC assays were used to measure the antioxidant capacities of the four citrus parts. The mechanisms of these four antioxidant tests can be divided into two types: the DPPH, FRAP and CUPRAC tests are mainly electron transfer type antioxidant methods [14,15,16,17] while ORAC is a hydrogen supply ability type antioxidant method [18,19]. The results showed significant differences among different varieties for the four fruit parts with each measurement method.

DPPH antioxidant tests are commonly used in determining the primary antioxidant capacities. The DPPH radical scavenging mechanisms include two types: the electron transfer type when components dissolve in polar solutions and the hydrogen supply ability type in nonpolar solutions [14,20]. In this study, the DPPH radical scavenging abilities were mainly based on the electron transfer ability of the antioxidant components since the citrus extracts were dissolved in water. DPPH values ranged from 5.48 to 20.73 mg Trolox equivalent antioxidant capacities (TEAC)/g DW in flavedo and from 4.37 to 17.43 mg TEAC/g DW in albedo. AY30 in flavedo and PG in albedo ranked the first, respectively, while CX showed the lowest value in both parts. In the segment membrane, DPPH value varied from 2.72 (TCH) to 7.34 mg TEAC/g DW (HMR). In the juice sacs, TC had the highest DPPH value (5.73 mg TEAC/g DW) and KZJ22 had the lowest (1.46 mg TEAC/g DW).

FRAP is another antioxidant test widely used in the determination of plant antioxidant capacity [15]. FRAP values in flavedo varied from 1.94 to 21.33 mg TEAC/g DW, with the highest and lowest value corresponding to AY30 and CX, respectively, which is similar to the DPPH result. GAC showed the highest FRAP value in both albedo (25.26 mg TEAC/g DW) and segment membrane part (19.83 mg TEAC/g DW) while YHY showed the lowest value in albedo (4.82 mg TEAC/g DW). CX ranked the last in flavedo, segment membrane, juice sacs, which may due to its low total phenolics content. FRAP value of AY27 (13.00 mg TEAC/g DW) ranked the first in juice sacs, which is almost three times higher than CX (4.28 mg TEAC/g DW).

The ORAC method, which was widely used in animal and botany materials, tests the hydrogen atom transfer ability of antioxidant components [18]. The ORAC values were the highest among the four methods, which may due to its unique antioxidant type. TC showed the highest ORAC value in flavedo (468.96 mg TEAC/g DW) and albedo (677.90 mg TEAC/g DW), while NG20 showed the lowest value in flavedo (75.78 mg TEAC/g DW) and AY27 showed the lowest in albedo (55.49 mg TEAC/g DW). In the segment membrane, YHY and YXC showed the highest value (208.86 mg TEAC/g DW) and lowest value (37.26 mg TEAC/g DW), respectively. In juice sacs parts, interestingly, KZJ22 showed the highest ORAC value of 76.53 mg TEAC/g DW, although KZJ22 showed the lowest DPPH value in the same part, suggesting that different antioxidant measurements could show great differences in the antioxidant index.

The CUPRAC assay determines the capacities of tested materials to reduce divalent copper ions to cuprous ions [17]. The CUPRAC value ranking was similar to that of the FRAP value, which may be due to the similar antioxidant mechanisms involved. AY30 showed the highest CUPRAC value in flavedo (54.80 mg TEAC/g DW), OG ranked the first in albedo (46.63 mg TEAC/g DW), while ZXY showed the lowest CUPRAC value in both parts (24.44 mg TEAC/g DW in flavedo and 11.34 mg TEAC/g DW albedo). In segment membrane, HY showed the highest value (6.37 mg TEAC/g DW) and CX showed the lowest value (2.57 mg TEAC/g DW). In juice sacs, HY ranked the first with 6.54 mg TEAC/g DW and KZJ22 ranked the last with CUPRAC value of 2.30 mg TEAC/g DW.

To comprehensively compare the antioxidant capacities, an overall antioxidant potency composite (APC) index was calculated according to the method described by Seeram et al. [21]. The overall APC index showed obvious variations, ranging from 33.03 (SJY) to 92.01 (AY30) in flavedo, from 24.96 (YHY) to 86.39 (GAC) in albedo, from 40.08 (TCH) to 90.96 (GAC) in segment membrane, and from 50.03 (YXC) to 84.23 (AY27) in juice sacs

2.3. Tumor Cytotoxicity

In vitro tumor cytotoxicity of the citrus extracts were measured on three gastric cancer cell lines, i.e., SGC-7901, BGC-823 and AGS. Cell viability assays were performed with a Cell Counting Kit-8 (CCK-8) and the corresponding IC50 values were calculated. Among the four parts, only flavedo extracts of each citrus variety showed significant tumor cytotoxicity, with IC50 values as shown in Table 6, while the other three parts showed no significant cytotoxicity effects (data not shown). Among the three cancer cell lines, the AGS cell line seem to be generally more sensitive to citrus extracts treatments than other two cell lines, which can be inferred from the lower IC50 values. Among the citrus types, pummelo fruits extracts showed the high IC50 value (>100 μg/mL) in all three cell lines, suggested that the tumor cytotoxicity of pummelo fruits to the cancer cells tested was weak. QJ (IC50 value = 20.36 μg/mL), CR (IC50 value = 18.71 μg/mL), NH (IC50 value = 15.77 μg/mL) showed the highest antitumor activity to SGC-7901, BGC-823, AGS cell, respectively. The highest IC50 value for each cell is more than 15 times higher than the lowest.

Table 6

The IC50 value of citrus flavedo extracts treatment to three gastric cancer cell lines.

Cultivars	SGC-7901 (μg/mL)	BGC-823 (μg/mL)	AGS (μg/mL)
AY27	28.74 ± 0.61 abc	22.61 ± 1.04 abc	20.49 ± 0.97 abc
AY30	27.50 ± 1.34 abc	28.51 ± 0.81 abcd	24.32 ± 1.01 abcd
AY31	62.45 ± 2.49 efg	68.62 ± 3.07 gh	53.41 ± 0.78 ef
BZH	40.65 ± 1.86 bcd	38.90 ± 0.93 cde	34.64 ± 1.11 bcd
CR	28.50 ± 1.80 abc	18.71 ± 0.82 a	24.25 ± 1.32 abcd
CX	80.58 ± 1.12 gh	83.40 ± 5.94 hijk	74.89 ± 2.43 gh
DF	64.34 ± 2.31 efg	42.39 ± 0.90 de	36.52 ± 0.65 cde
GAC	60.86 ± 1.78 ef	70.37 ± 3.65 gh	61.94 ± 2.18 fg
GN	33.81 ± 2.01 abc	35.80 ± 1.10 abcde	34.51 ± 0.99 bcd
GOC	55.61 ± 3.35 def	49.43 ± 2.05 ef	32.39 ± 1.39 abcd
HMR	100.97 ± 4.35 i	99.05 ± 1.97 k	59.95 ± 1.34 fg
HY	164.86 ± 6.31 k	150.54 ± 7.45 l	139.8 ± 8.13 j
KZJ	39.07 ± 1.72 abcd	37.80 ± 1.94 cde	30.78 ± 1.55 abcd
MBWD	203.51 ± 8.28 l	153.78 ± 4.51 lm	114.64 ± 4.46 i
MTH	37.79 ± 1.71 abcd	37.27 ± 1.36 cde	29.34 ± 1.40 abcd
MXG	89.16 ± 3.69 hi	77.75 ± 4.93 ghij	83.65 ± 2.86 h
NG20	120.43 ± 6.27 j	89.07 ± 3.92 ijk	74.59 ± 2.65 gh
NH	21.65 ± 0.63 ab	19.88 ± 0.52 ab	15.77 ± 0.66 a
OG	34.60 ± 0.42 abc	37.01 ± 1.37 bcde	32.33 ± 0.64 abcd
PG	37.76 ± 1.14 abcd	33.21 ± 1.54 abcde	34.02 ± 1.89 bcd
PTY	207.37 ± 11.72 l	192.10 ± 3.72 n	165.46 ± 4.93 k
QJ	20.36 ± 0.81 a	24.44 ± 1.01 abc	17.49 ± 0.65 ab
QOG	34.39 ± 3.05 abc	30.71 ± 0.63 abcd	23.37 ± 0.57 abcd
RN1	205.73 ± 5.71 l	168.73 ± 4.87 m	134.84 ± 4.77 j
SJY	368.40 ± 20.35 o	355.32 ± 17.47 p	311.41 ± 21.54 m
SW	92.66 ± 2.19 hi	97.75 ± 8.99 k	79.95 ± 4.45 h
SYXX	81.40 ± 1.75 gh	74.28 ± 3.77 ghi	72.35 ± 2.11 gh
TC	35.88 ± 1.39 abc	33.75 ± 2.10 abcde	24.24 ± 1.65 abcd
TCH	69.14 ± 3.79 fg	61.86 ± 2.55 fg	53.82 ± 0.63 ef
WHOG	26.42 ± 1.72 ab	30.14 ± 1.59 abcd	26.24 ± 0.93 abcd
WZ	137.53 ± 6.09 j	95.04 ± 1.86 jk	76.47 ± 1.98 gh
YHY	313.09 ± 12.31 n	253.47 ± 14.60 o	255.56 ± 18.56 l
YL	81.39 ± 1.85 gh	73.77 ± 5.29 ghi	66.24 ± 0.78 fgh
YXC	46.44 ± 2.30 cde	34.07 ± 0.76 abcde	39.32 ± 2.39 de
ZXY	291.47 ± 13.15 m	244.66 ± 9.86 o	173.76 ± 5.97 k

Results were the mean ± SD (n = 3) on Half inhibition rate (IC50). Values within each column followed by different superscript letters were significantly different (p < 0.05) according to Tukey’s tests.

2.4. Identification and Quantification of Individual Flavonoid Compound

Further identification and quantification of individual flavonoid in citrus fruits were performed by LC-ESI-MS/MS and UPLC-DAD. A total of 39 flavonoid compounds, including four flavones, nine flavanones and 26 PMFs, were identified. The identification was based on comparison of the retention times and the maximum absorption wavelength of standards, as well as the fragment ion information reported in previous studies (Table 7 and Table S6, Figures S1–S5) [22,23,24].

Table 7

Determination of flavonoids from citrus fruits by LC-ESI-MS/MS (+ESI mode) and UPLC-DAD.

Peak No.	RT (min)	λmax (nm)	[M + H]+ (m/z)	Formula	Fragment Ions (m/z)	Tentative Compounds
Flavone C-glycosides
1	2.30	240.4, 330.6	595.1665	C27H30O15	577, 559, 541, 529, 523, 505	Vicenin-2
3	3.53	255.0, 348.0	433.1142	C21H20O10	415, 397, 379, 367, 337, 313	Apigenin-8-C-glucoside
5	4.46	272.3	463.1234	C22H22O11	445, 427, 397, 380, 367, 343, 313	Diosmetin-6-C-glucoside
Flavanone O-glycosides
2	3.18	284.2	597.1823	C27H32O15	435, 399, 355, 289, 263, 195	Eriocitrin
4	3.61	284.2	597.1806	C27H32O15	433, 399, 355, 289, 263, 195	Neoeriocitrin
6	5.02	283.0, 329.4	581.186	C27H32O14	419, 383, 339, 273, 195, 153	Narirutin
8	5.94	283.0, 329.4	581.1865	C27H32O14	419, 383, 339, 315, 273, 195	Naringin
9	6.68	284.2, 331.8	611.1987	C28H34O15	449, 413, 369, 303, 263, 195, 153	Hesperidin
10	7.92	284.2	611.1974	C28H34O15	413, 369, 303, 263, 195	Neohesperidin
11	10.38	283.0	595.2019	C28H34O14	377, 353, 329, 287, 263, 195	Didymin
12	10.62	328.2	595.2017	C28H34O14	463, 379, 287, 263, 153	Poncirin
15	11.68	337.7	725.2283	C33H30O31	419, 404, 390, 361	Melitidin
Flavone O-glycoside
7	5.65	267.0, 334.0	579.1704	C27H30O14	433, 315, 271, 195, 153, 127	Rhoifolin
Polymethoxyflavonoids
13	11.02	215.6, 324.7	329.1019	C18H16O6	314, 299, 271, 228	Monohydroxytrimethoxyflavone (1)
14	11.29	328.2	359.1114	C19H18O7	344, 329, 314, 286, 257	Gardenin B
16	11.97	322.3	329.1016	C18H16O6	314, 299, 268, 136	Monohydroxytrimethoxyflavone (2)
17	12.40	343.4	331.0817	C17H14O7	316, 301, 273, 245, 217, 168	Trihydroxydimethoxyflavone
18	12.57	348.0	389.1227	C20H20O8	374, 359, 341, 298	Monohydroxy-pentamethoxyflavone (1)
19	12.84	323.5	373.1277	C20H20O7	358, 343, 315, 287, 181, 153	Isosinensetin
20	13.26	281.8	359.1129	C19H18O7	344, 329, 314, 301, 163, 147	Monohydroxytetramethoxyflavone
21	13.53	335.4	389.1227	C20H20O8	374, 359, 341, 298	Monohydroxypentamethoxyflavone (2)
22	13.55	351.6	403.1389	C21H22O8	388, 373, 359, 327, 183, 163	Hexamethoxyflavone (1)
23	14.25	240.4, 330.6	373.1289	C20H20O7	357, 343, 329, 312, 297, 153	Sinensetin
24	14.60	268.0, 334.0	343.118	C19H18O6	328, 313, 285, 257, 181, 153	Tetramethyl-O-isoscutellarein
25	14.80	343.4	345.0974	C18H16O7	330, 318	Dihydroxytrimethoxyflavone
26	14.96	210.0, 336.6	403.1392	C21H22O8	388, 373, 359, 327, 183, 163	Hexa-O-methylgossypetin
27	15.17	337.7	373.1282	C20H20O7	253, 211, 196, 181, 168, 150	5,7,3′,4′,5′-Pentamethoxyflavone
28	15.78	249.8, 334.2	403.1388	C21H22O8	388, 373, 358, 355, 327, 211, 183	Nobiletin
29	15.97	321.1	343.1175	C19H18O6	328, 313, 285, 257, 181, 153	Tetramethyl-O-scutellarein
30	16.32	350.4	375.1074	C19H18O8	345, 327. 197	5,4′-Dihydroxyl-3,7,8,3′-tetramethoxyflavonol
31	16.49	254.6, 341.2	433.1486	C22H24O9	417, 403, 388, 373, 303, 217	3,5,6,7,8,3′,4′-Heptamethoxyflavone
32	17.42	271.1, 323.5	373.1281	C20H20O7	358, 343, 328, 297, 211, 183	Tangeretin
33	17.82	282.0, 341.0	359.1128	C19H18O7	344, 329, 311, 283	6-O-Desmethyltangeritin/7-O-desmethyltangeritin
34	18.49	349.2	403.1388	C21H22O8	388, 373, 359, 327, 183, 163	Hexamethoxyflavone (2)
35	18.56	283.0, 341.3	389.1229	C20H20O8	374, 359, 341, 331, 197	5-Hydroxy-6,7,8,3′,4′-pentamethoxyflavone
36	18.99	331.8	329.1018	C18H16O6	299, 285, 268, 153	Monohydroxytrimethoxyflavone
37	19.36	273.5, 357.6	389.1232	C20H20O8	374, 359, 341, 298	Monohydroxypentamethoxyflavone (3)
38	19.65	346.9	419.1326	C21H22O9	404, 389, 372, 218	Natsudaidai
39	20.88	290.0, 329.4	359.1116	C19H18O7	344, 326, 298, 282, 255, 162	5-Hydroxy-7,8,3′,4′-tetramethoxyflavone

The four flavones included three flavone C-glucosides (vicenin-2, apigenin-8-C-glusoide and diosmetin-6-C-glucoside) and one flavone O-glycoside (rhoifolin). Diosmetin-6-C-glucoside was only found in the Buzhihuo (BZH), NG20, SJY and MXG, with the highest concentration of 9.05 mg/g DW in the albedo part of BZH. Rhoifolin was only found in pomelo type fruit (Table S1 and S3–S5).

The nine flavanone O-glycosides included eriocitrin, neoeriocitrin, narirutin, naringin, hesperidin, neohesperidin, didymin, poncirin and melitidin, in which narirutin, naringin, hesperidin, neohesperidin were the most common flavonoid components in the 35 citrus varieties (Table S1 and 3–S5). Naringin was very abundant in pummelo type fruit. The content of narirutin was generally low in the pomelo type fruit. Melitidin was only found in the flavedo part of MXG, while vicenin-2 was only existed in KZJ22. The content of hesperidin and neohesperidin showed a seesawing like relationship, that is, the varieties with abundant hesperidin tend to have low level of neohesperidin, which may due to the differentiation in synthetic metabolism.

The PMFs were only found in flavedo, including one trihydroxydimethoxyflavone, four trimethoxyflavones, seven tetramethoxyflavones, eight pentamethoxyflavones, five hexa-methoxyflavones and one heptamethoxyflavone (Table S2). Among the 26 PMFs identified, there were 12 monohydroxy PMFs, one dihydroxy PMFs and one trihydroxy PMF. The exact location of the methoxy groups of two monohydroxypentamethoxyflavones, two hexamethoxyflavones and three monohydroxypentamethoxyflavones need further identification and we added a number after their names according to the order they appeared in the UPLC chromatogram. The PMF contents differed among varieties. Trihydroxydimethoxyflavone only existed in HMR (3.78 mg/g DW) and 5,7,3′,4′,5′-pentamethoxyflavone only existed in AY27 (1.22 mg/g DW). Isosinensetin sinensetin, tetramethyl-O-isoscutellarein, nobiletin, tetramethyl-O-scutellarein, 3,5,6,7,8,3′,4′-heptamethoxy-flavone, tangeretin, 5-hydroxy-6,7,8,3′,4′-pentamethoxyflavone existed in most citrus varieties. Nobiletin was the maximum PMFs in almost all of the varieties except for TCH, WZ, DF, SYXX, SW and MXG.

2.5. Correlations between Total Phenolics and Bioactivites

Correlation analyses were performed to investigate the relationship between the phenolics content, antioxidant ability and cytotoxicity on gastric cancer cell (Table 8 and Table 9).

Table 8

Pearson’s correlation coefficients among total phenolics, antioxidant values and in vitro anticancer abilities in citrus flavedo.

Bioactive Capacities	Total Phenolic	DPPH	FRAP	ORAC	CUPRAC	APC Index	1/IC50 SGC-7901	1/IC50 BGC-823	1/IC50 AGS
Total Phenolic	1
DPPH	0.853 **	1
FRAP	0.888 **	0.923 **	1
ORAC	0.735 **	0.560 **	0.563 **	1
CUPRAC	0.936 **	0.766 **	0.787 **	0.741 **	1
APC Index	0.958 **	0.919 **	0.930 **	0.781 **	0.916 **	1
1/IC50 SGC-7901	0.513 **	0.375 *	0.421 **	0.366 **	0.590 **	0.482 **	1
1/IC50 BGC-823	0.456 **	0.316	0.395 **	0.309	0.536 **	0.433 **	0.947 **	1
1/IC50 AGS	0.548 **	0.392 **	0.456 **	0.396 **	0.617 **	0.514 **	0.965 **	0.949 **	1

1/IC50 means the reciprocal value of IC50; One and two asterisks represent statistical significance at p < 0.05 and p < 0.01, respectively.

Table 9

Pearson’s correlation coefficients among total phenolics, antioxidant values in albedo, segment membrane and juice sacs.

Bioactive Capacities	Total Phenolic	DPPH	FRAP	ORAC	CUPRAC	APC Index
Albedo
Total Phenolic	1
DPPH	0.679 **	1
FRAP	0.645 **	0.901 **	1
ORAC	0.500 **	0.156	0.015	1
CUPRAC	0.656 **	0.677 **	0.626 **	0.404 **	1
APC Index	0.804 **	0.875 **	0.806 **	0.534 **	0.876 **	1
Segment membrane
Total Phenolic	1
DPPH	0.322	1
FRAP	0.346 *	0.741 **	1
ORAC	0.725 **	0.086	−0.059	1
CUPRAC	0.422 *	0.589 **	0.681 **	0.358	1
APC Index	0.649 **	0.779 **	0.754 **	0.538 **	0.884 **	1
Juice sacs
Total Phenolic	1
DPPH	0.452 **	1
FRAP	0.421 *	0.696 **	1
ORAC	0.576 **	−0.183	0.033	1
CUPRAC	0.409 *	0.896 **	0.606 **	−0.255	1
APC Index	0.705 **	0.853 **	0.827 **	0.282	0.786 **	1

One and two asterisks represent statistical significance at p < 0.05 and p < 0.01, respectively.

For the 4 antioxidant traits, DPPH, FRAP, CUPRAC showed significant correlation with each other (p < 0.01), indicating that these 3 traits determined the same type of antioxidant capacities. However, ORAC method only showed correlation with other 3 traits in flavedo. In other 3 parts of citrus, ORAC traits showed very weak relation with other 3 methods, suggesting that ORAC value reflected a different type of antioxidant ability. Total phenolics contents in all the 4 parts showed significant correlations with all 4 the antioxidant traits and the APC overall index (Table 8 and Table 9), indicating that phenolics compounds were the principle contributor to antioxidant capacities of citrus. High correlation of total phenolics contents and antioxidant capacities also showed in the peach fruit [25], Chinese bayberry [26], vegetables and grains [27], indicating that this is a common phenomenon in nature.

The cytotoxicity of extracts showed high correlations among 3 cell traits. However, the correlations between antioxidant capacities and anticancer abilities were relatively low, indicating that the in vitro cytotoxicity of citrus extracts may not be caused by its antioxidant ability. In plenty previous anticancer studies, most of the flavanones were reported to be functioned through enhancing the body’s own function which was based on antioxidant capacities of flavanone [28,29,30]. However, the results of present study indicated a different functional mechanism of citrus flavonoids rich extracts.

For the cytotoxicity of individual compounds, the contents of 11 PMFs showed significant correlations with cytotoxicity of extracts on all three gastric cancer cell lines, in which nobiletin showed the highest correlation coefficient (r = 0.587, p < 0.01; r = 0.530, p < 0.01; r = 0.534, p < 0.01) (Figure 1, Tables S7–S10). None of the flavanones or flavones showed significant positive relationships with cytotoxicity and rhoifolin showed a significant negative relationship (r = −0.378, p < 0.05; r = −0.366, p < 0.05; r = −0.361, p < 0.05). These results suggested that PMFs might be the main contributors to the cytotoxicity of extracts. Similar results were observed in previous studies. PMFs have shown anti-proliferation abilities to many cancers in vitro [31,32] and in vivo [33,34]. They were suggested to be functioned through induction of cell apoptosis [33,35], cell cycle blocking [36,37] and autophagy, etc. [38,39].

Figure 1

Correlation analysis of bioactive traits and individual flavonoid compounds in flavedo (A); albedo (B); segment membrane (C) and juice sacs (D).

For the antioxidant activity of individual compounds, the didymin showed the positive relationship with APC index and all four antioxidant tests in flavedo. Hesperidin in flavedo (r = 0.558, p < 0.01), neohesperidin in albedo (r = 0.718, p < 0.01), poncirin (r = 0.618, p < 0.01) in albedo showed strong relationship with CUPRAC. Naringin showed significant relationship with ORAC value in segment membrane (r = 0.592, p < 0.01). These results showed that didymin, hesperidin, neohesperidin, poncirin, naringin might played dominant roles in antioxidant ability of citrus extracts.

3. Experimental Section

3.1. Materials

Citrus fruits at commercial maturity were harvested from orchards of Zhejiang Province in December 2015 (Table 1), and transported to the laboratory of Zhejiang University, Hangzhou, within 6 h of harvest. Uniform fruit free from blemishes and mechanical injury was selected for the present study. The fruits were separated into four parts, i.e., flavedo, albedo, segment membrane and juice sacs, and immediately frozen in liquid nitrogen. After freeze-drying (FM 25EL-85, VirTis, Gardiner, NY, USA), all samples were ground into a fine powder and stored at −80 °C for further experiments. The SGC-7901, BGC-823, AGS gastric cancer cell lines were obtained from the Department of Surgery, Second Affiliated Hospital, School of Medicine, Zhejiang University.

3.2. Chemicals and Reagents

All standards and reagents were of HPLC grade. Narirutin, neoeriocitrin, hesperidin, naringenin, poncirin, naringin, hesperidin, neohesperidin, nobiletin, tangeretin, 2,2-diphenyl-1-picrylhydrazyl (DPPH), 2,4,6-tris(2-pyridyl)-s-triazine (TPTZ), Trolox, Folin-Ciocalteu reagent, 2,2′-azobis(2-methylpropionamidine) dihydrochloride (AAPH), rutin, gallic acid, fluorescein sodium, copper chloride, neocuproine, ammonium acetate, methanol, acetonitrile were purchased from Sigma-Aldrich (St. Louis, MO, USA). Eriocitrin was purchased from Aladdin Industrial Inc. (Shanghai, China). Isosinensetin, and 5-demethylnobiletin were purchased from Biobiopha Co., Ltd. (Kunming, China). Didymin was purchased from J & K Scientific (Shanghai, China). Cell Counting Kit-8 was purchased from Dojindo Molecular Technologies, Inc. (Shanghai, China). Samples for HPLC were filtered through a 0.22 μm membrane before injection. Double-distilled water (ddH2O) was used in all experiment. RPMI 1640 medium, fetal bovine serum (FBS), N-2-hydroxyethylpiperazine-N-2-ethane sulfonic acid (HEPES), trypsin-EDTA were purchased from Gibco (Waltham, MA, USA). Penicillin-streptomycin solution was purchased from Hangzhou Keyi Biotechnology Co., Ltd. (Hangzhou, China). All other solvents and reagents were of analytical grade purchased from Sinopharm Chemical Reagents Co., Ltd. (Shanghai, China).

3.3. Fruit Quality Analysis (Color, Fruit Weight, Edible Proportion and Total Soluble Sugar)

Fruit color measurement was carried out using the Citrus Color Index (CCI) according to a previous report [40]. The raw data were adopted as L*, a* and b* with a MiniScan XE Plus Colorimeter (HunterLab, Reston, VA, USA) and the CCI was calculated as CCI = [1000 × a*/(L* × b*)]. Four evenly distributed equatorial sites were measured for each fruit and a mean value was obtained from the measurement of 15 fruits per variety. The edible rate was calculated as the weight percentage of pulp to the whole fruit. TSS of 15 fruits per variety were measured with a portable digital refractometer (Atago PR-101α, Tokyo, Japan) at 25 °C and the data were expressed as °Brix.

3.4. Extraction and Determination of Total Phenolics

One gram of citrus fruit ground powder was ultrasonically extracted in 20 mL of 95% ethanol at 25 °C in a material-to-solvent ratio of 1:20 (w/v) for three times. The extract was centrifuged at 10,000 ×g for 5 min and the supernatants were evaporated under reduced pressure at 35 °C to remove the ethanol. The phenolics were enriched by solid-phase extraction using a Sep-pak C18 cartridge (12 cc, 2 g sorbent, Waters Corp., Milford, MA, USA). The citrus phenolic-rich extracts were used for further analysis.

Total phenolics contents were measured with a Folin-Ciocalteu method according to previous report [41] with slight modification. In brief, 4 mL of ddH2O and 0.5 mL appropriately-diluted citrus extracts was placed into a test tube, added with 0.5 mL Folin-Ciocalteu (0.5 mol/L) and incubated for 3 min. Then 1 mL of saturated sodium carbonate was added into the mixture following by incubating the reaction for 2 h in 30 °C water bath. The absorbance of the reaction product were measured at 760 nm using a microplate reader (Synergy H1, Biotek, Winooski, VT, USA). Gallic acid was used as the standard and the results were expressed as mg gallic acid equivalent (GAE)/g DW.

3.5. Antioxidant Capacity Assays

The DPPH radical scavenging activity was measured according to our previous publication [42] with some modification. In brief, 2 μL diluted citrus extracts was mixed with 198 μL DPPH solution (60 μM), the mixture was allowed to react for 2 h at room temperature, away from light. Then the absorbance at 515 nm was measured using a microplate reader. Trolox was used as the standard and the results were expressed as mg Trolox equivalent antioxidant capacities (TEAC)/g DW.

The FRAP assay was carried out according to Zhang et al. [43] with modifications. Briefly, the FRAP working solution was prepared by mixing 100 mL acetate buffer (300 mmol/L, pH 3.6), 10 mL TPTZ solution (10 mmol/L in 40 mmol/L HCl) and 10 mL FeCl3 (20 mmol/L). 10 μL appropriately diluted citrus extracts and 90 μL of FRAP working solution was mixed in a 96-well plate and incubated for 5 min. Then the absorbance of 593 nm was recorded with a microplate reader. Trolox was used as the standard and the results were expressed as mg Trolox equivalent antioxidant capacities (TEAC)/g DW.

ORAC antioxidant activity was measured according to previous report [44] with some minor modifications. 25 μL appropriately diluted citrus extraction was placed into a black-walled 96-well plate, and mixed with 150 μL sodium fluorescein (40 nmol/L). The mixture was incubated for 10 min at 37 °C. Then 25 μL AAPH (150 mmol/L) was added and the fluorescence detection was performed immediately with a microplate reader (set with excitation wavelength of 485 nm, emission wavelength of 535 nm, time interval of 2 min for the 2 h detection). Phosphate buffer solution (PBS) was used as a blank control and the final fluorescence measurements were expressed relative to the initial reading (fn). Results were calculated based on the differences in areas under the sodium fluorescein decay curve (AUC) between the blank and samples. Trolox was used as standard and the results were expressed as mg Trolox equivalent antioxidant capacities (TEAC)/g DW. The AUC was calculated as: AUC = (f

The CUPARC assay was carried out according to previous report [16] with some modifications. Reaction system consisted of 20 μL appropriately diluted citrus extraction, 50 μL copper chloride (10 mmol/L), 50 μL neocuproine (7 mmol/L), 50 μL ammonium acetate (pH 7) and 50 μL ddH2O was added into a 96-well plate in order. After 30 min of incubation away from light in room temperature, absorbance of 450 nm was measured in a microplate reader. Trolox was used as standard and the results were expressed as mg Trolox equivalent antioxidant capacities (TEAC)/g DW.

An overall APC index was applied to comprehensively evaluated the antioxidant traits of the extracts. For each antioxidant trait, antioxidant index score = [(sample score/best score) × 100]. The APC index was calculated as the average of the antioxidant index scores of the referred 4 methods.

3.6. Cell Culture and Cell Viability Assay

The human gastric cancer cell lines SGC-7901, BGC-823, AGS were cultured in RPMI 1640 medium containing 10% fetal bovine serum (FBS), 100 U/mL penicillin, 100 μg/mL streptomycin, and 20 mmol/L HEPEs, at 37 °C in an incubator (Thermo Scientific 3111, Thermo Scientific, Waltham, MA, USA) containing 5% CO2. Cells were passaged every 48 h using trypsin (0.25%)/EDTA (0.02%) solution. Exponentially growing cells were used for experimentation.

Cell viability assay was performed with cell counting kit-8 (CCK-8) analysis according to the methods described in previous study [45]. Briefly, cells (4000 cells per well for SGC-7901, 8000 cells per well for BGC-823 and 9000 cells per well for AGS) were seeded into 96-well plates. After 24 h incubation, the medium were moved and cells were treated with or without citrus extractions in a total volume of 200 μL each well. After 48 h incubation, the supernatant was removed and washed with PBS for 2 times. The cell viability was measured using CCK-8 kit according to the instruction. Taxol was used as positive control. The inhibition ratio was calculated: Inhibition ratio = (A

The IC50 value was calculated by probit analysis method using SPSS 19.0 software (IBM, Armonk, NY, USA).

3.7. UPLC-DAD and LC-ESI-MS/MS Analysis of Phonilic Compounds

Individual flavonoid compounds were identified and quantified combining LC-ESI-MS/MS and UPLC-DAD. Flavones and flavanones were detected at 280 nm and polymethoxylated flavonoids were detected at 330 nm. 13 flavanoids, i.e., eriocitrin, neoericitrin, narirutin, naringin, hesperidin, neohesperidin, didmin, poncirin, isosinensetin, sinensitin, nobiletin, tangeretin, 5-hydroxy-6,7,8,3′,4′-pentamethoxyflavone, were quantified with their own standard curves according to the retention time and the chromatographic peak area in the UPLC analysis. Other 26 flavonoids were quantified as equivalents of hesperidin at 280 nm. All tests were run in triplicate and data were expressed as mg/g DW.

The flavonoid compounds were determined with a UPLC system (2695 pump, 2996 diode array detector, Waters Corp.) coupled with an BEH C18 analytical column (ACQUITY UPLC, 2.1 × 150 mm, Waters Corp.). The column was operated at a temperature of 25 °C. The injection of sample was 2 μL and the flow rate was 0.3 mL/min. The compounds were detected between 200 and 500 nm. The mobile phase of UPLC consisted waters (Eluent A) and acetonitrile (Eluent B). the gradient program was as follows: 0–5 min, 20% of B; 5–8 min, 20–34% of B; 8–20 min, 34–60% of B; 20–22 min, 60–100% of B; 22–23 min, 100% of B; 23–24 min, 100–20% of B; 24–25 min, 20% of B.

Mass spectrometric analysis were performed according to our previous publication [25]. Briefly, an Agilent 6460 triple quadrupole mass spectrometer equipped with an ESI source (Agilent Technologies, Santa Clara, CA, USA) was used for mass analysis and the analysis were operated in positive ionization mode. The nebulizer pressure was set to 45 psi and drying gas flow rate was 5 L/min. The flow rate and the temperature of the sheath gas was 11 L/min and 350 °C, respectively. Chromatographic separations were done on an BEH C18 analytical column (ACQUITY UPLC, 2.1 × 150 mm) using an Agilent 1290 Infinity UPLC system (Agilent Technologies). The eluent was split and with a rate of 0.3 mL/min going into the mass detector. The data acquisition and processing were performed at an Agilent Mass Hunter Workstation.

3.8. Statistical Analysis

All data were obtained from at least three replications and expressed as the means ± standard deviation. The statistical analyses were carried out using SPSS 19.0 software (IBM, Armonk, NY, USA). Significant differences among the sample were analyzed using one-way ANOVA, followed by Tukey’s test at p < 0.05. Pearson correlation coefficients were calculated at p < 0.05.

4. Conclusions

A total of 39 flavonoids, including four flavones, nine flavanones and 26 PMFs were identified and quantified from 35 varieties of five types of citrus fruit growing in Zhejiang Province of China. Among them, all 39 compounds could be found in the flavedo, three flavones and nine flavanones were found in the albedo, segment membrane and juice sacs, while PMFs were existing only in the flavedo. The flavonoids composition and bioactivity varied depending on the types and tissues of citrus fruit. According to the results of correlation analysis, phenolics were deduced to be the chief contributor for the antioxidant capacity of citrus fruit, in which the individual flavanone compounds including didymin, hesperidin, neohesperidin, poncirin, naringin were the principal contributing components. Phenolics extracts from flavedo showed significant cytotoxicity effects on gastric tumor cell lines, and PMFs were deduced to be the dominant contributors, with the nobiletin as the principle contributing component. The correlation between antioxidant capacities and the cytotoxicity effects was not significant. These results may offer important information for breeding and further utilization of citrus resources.