Food Chemistry 120 (2010) 184–192

Contents lists available at ScienceDirect

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

Isolation of carotenoids, flavonoids and polysaccharides from Lycium barbarum

L. and evaluation of antioxidant activity
C.C. Wang, S.C. Chang, B. Stephen Inbaraj, B.H. Chen *
Department of Food Science, Fu Jen University, Taipei 242, Taiwan

a r t i c l e i n f o a b s t r a c t

Article history: A preparative column chromatography method was developed to isolate carotenoids, flavonoids and
Received 23 February 2009 polysaccharides, from Lycium barbarum L., possessing vital biological activity, and their antioxidant activ-
Received in revised form 20 August 2009 ity was evaluated. Carotenoids were isolated by a column containing magnesium oxide and diatoma-
Accepted 6 October 2009
ceous earth (1.5:1, w/w), and b-carotene was eluted with hexane, b-cryptoxanthin and neoxanthin
with ethyl acetate and zeaxanthin with ethyl acetate–ethanol (80:20, v/v). Flavonoids and phenolic acids
were separated using a Cosmosil 140 C18-OPN column, with phenolic acids being eluted with deionized
Keywords:
water and neutral flavonoids with methanol. Polysaccharides were fractionated using a DEAE-Sepharose
Lycium barbarum L.
Polysaccharide
CL-6B column; neutral polysaccharides were eluted with water and acidic polysaccharides with different
Flavonoid concentrations of NaCl. For antioxidant activity, the flavonoid fraction was the most effective in scaveng-
Carotenoid ing DPPH and ABTS+ free radicals, chelating metal ions and reducing power, while the zeaxanthin frac-
Preparative column chromatography tion and polysaccharides showed the most pronounced effect in scavenging hydroxy free radicals and
Antioxidant activity superoxide anions, respectively.
Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction Flavonoids, a class of polyphenol compounds, are also widely

distributed in plants, especially fruits and vegetables (Erlund,
Lycium barbarum L., a traditional Chinese herb possessing vital 2004; Kanaze, Gabrieli, Kokkalou, Georgarakis, & Niopas, 2003).
biological activities, such as prevention of cancer and age-related More than 6000 flavonoids have been characterised in nature,
macular degeneration (AMD), is widely used in Asian countries. but their variety and amount vary because of differences in grow-
Many functional components in L. barbarum, including flavonoids, ing environments, maturity and growth conditions (Erlund, 2004).
carotenoids and polysaccharides, have been reported to be closely The physiological activities of flavonoids, such as anti-cancer, anti-
associated with the health-enhancing effect (Fraser & Bramley, inflammation, and anti-atherosclerosis, have been well docu-
2004; Marchand, 2002; Sheng et al., 2007; Yang, Zhao, Yang, & mented (Arai et al., 2000; Havsteen, 2002). Thus, it would be ben-
Ruan, 2008). Several physiological studies have focused on poly- eficial to isolate flavonoids from L. barbarum for use as a nutritional
saccharides; however, both carotenoids and flavonoids have been supplement.
less investigated, especially for their antioxidant activity. Polysaccharides, composed of 100 or more monosaccharides,
Carotenoids, a group of lipid-soluble compounds responsible for are often present in Chinese herbs in large amounts. Polysaccha-
yellow and red colours of many plants and food products, have rides can be water-soluble or water-insoluble, with the former
been demonstrated to be effective in preventing chronic diseases types being glucurono b-glucan and b-glucan, and latter xylo-b-
such as cardiovascular disease and skin cancer (Fraser & Bramley, glucan, xylomann-b-glucan, hetero-b-glucan and manno-b-glucan
2004; Kohlmeier & Hastings, 1995). Inbaraj et al. (2008) reported (Huie & Di, 2004; Sun, Tang, Gu, & Li, 2005). It has been well estab-
the composition of carotenoids in L. barbarum, with zeaxanthin lished that polysaccharides possess antitumor activity (Sheng
constituting the largest portion (0.1%). This is important for pre- et al., 2007), and they can enhance immunity through production
vention of AMD as both lutein and zeaxanthin are the main retina of interleukin and antibody (Yang et al., 2008). However, the anti-
pigments (Cooper, Eldridge, & Peters, 1999; Snodderly, 1995). oxidant activity of polysaccharides in L. barbarum still remains
Therefore, it would be a great advantage to the health food indus- unknown.
try if zeaxanthin in L. barbarum could be isolated for possible pro- In view of the impact of carotenoids, flavonoids and polysaccha-
duction of functional foods in the future. rides on human health, this study was aimed to develop a prepara-
tive chromatography method for isolation of these functional
* Corresponding author. Tel.: +886 2 29053626; fax: +886 2 29021215. components from L. barbarum and to study their antioxidant
E-mail address: 002622@mail.fju.edu.tw (B.H. Chen). activity.

0308-8146/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodchem.2009.10.005
 C.C. Wang et al. / Food Chemistry 120 (2010) 184–192 185

2. Materials and methods (Hesperia, CA, USA). The Cosmosil 5Diol-300-II column
(300  7.5 mm I.D., particle size 5 lm) was from Nacalai Tesque.
2.1. Materials The spectrophotometer (Beckman DU 640) was from Beckman
(Fullerton, CA, USA). The rotary evaporator (Eyela N-1) was from
A total of 12 kg of L. barbarum fruits was purchased from a local the Eyela Co. (Tokyo, Japan). The high-speed centrifuge (Sorvall
drug store in Taipei, Taiwan, and was stored at 20 °C for use. RC5C) was from DuPont (Wilmington, Delaware, USA). The sonica-
Adsorbents for column chromatography, including magnesium tor (model DC 400H) was from Taipei, Taiwan. The homogenizer
oxide, diatomaceous earth and Cosmosil 140C18-OPN (particle size (model 890-68) was from the Oster Co. (Wisconsin, USA). The
140 lm), were obtained from Sigma (St. Louis, MO, USA), J.T. Baker shaker (V–U) was from Hsiang-Tai Co. (Taipei, Taiwan).
(Phillipsburg, NJ, USA) and Nacalai (Kyoto, Japan), respectively. The
ion-exchange resin, DEAE-Sepharose CL-6B, was from GE Amer- 2.3. Extraction and preparation of carotenoids
sham Bioscience (Uppsala, Sweden).
Flavonoid and phenolic acids standards, including caffeic acid, A 10 g powdered fruit sample of L. barbarum was mixed with
gallic acid, chlorogenic acid, p-coumaric acid, rutin and kaempferol, 50 ml of 10% anhydrous sodium sulphate solution, after which
with purities of 99%, 98%, 95%, 98%, 95% and 90%, respectively, the mixture was shaken for 3 min, followed by adding 100 ml of
were from Sigma. Internal standard, taxifolin, with a purity of hexane–ethanol–acetone–toluene (10:6:7:7, v/v/v/v) and shaking
85% was also from Sigma. Carotenoid standards, such as all- for 1 h. Then the solution was centrifuged at 4000 rpm for 1 min
trans-zeaxanthin and all-trans-b-cryptoxanthin, were from Extra- and the supernatant was collected. The residue was repeatedly ex-
synthese Co. (Genay, France), and all-trans-b-carotene was from tracted with 50 ml of hexane until colourless. The supernatants
Sigma. Internal standard, b-apo-80 -carotenal, was from Fluka were combined, evaporated to dryness under vacuum, and dis-
Chemical Co. (Buchs, Switzerland). The glucose standard was from solved in 100 ml of hexane–ethanol–acetone–toluene (10:6:7:7,
Sigma. v/v/v/v). A 5 ml aliquot of 40% methanolic KOH solution was added
The HPLC-grade solvents, including ethyl acetate, methanol, iso- and saponification was carried out under nitrogen in the dark for
propyl alcohol, hexane, acetonitrile, methylene chloride, acetone, 6 h. Next, the extract was evaporated to dryness, and dissolved in
acetic acid and toluene, were from Lab-Scan Co. (Gliwice, Poland). 20 ml of methylene chloride. For preparative chromatography,
Formic acid was from Riedel-de-Haën Co. (Seelze, Germany). 5 ml of concentrated carotenoid extract was poured into a column
Deionized water was made using a Milli-Q water purification sys- (300  16 mm I.D.) containing a mixture of 7.5 g magnesium oxide
tem from Millipore Co. (Bedford, MA, USA). and 5 g diatomaceous earth (1.5:1, w/w) to form a height of about
Chemicals, including iron(II) chloride, iron(III) chloride and 10 cm, followed by adding anhydrous sodium sulphate to form a
potassium ferricyanide, were from Showa Chemical Co. (Tokyo, Ja- layer of approximately 1 cm above the adsorbent. b-Carotene
pan). Ferrozin, trichloroacetic acid (TCA), a-tocopherol, trolox (6- was eluted with 30 ml of 100% hexane, neoxanthin and b-crypto-
hydroxy-2,5,7,8-tetramethylchromane-2-carboxylic acid), gallic xanthin with 50 ml of 100% ethyl acetate, and zeaxanthin with
acid, DPPH(2,2-diphenyl-1,picrylhydrazyl), ABTS (2,20 -azino- 50 ml of ethyl acetate/ethanol (80:20, v/v). The carotenoid compo-
bis(3-ethylbenzothiazoline-6-sulphonic acid), TBA (2-thiobarbitu- sition in each fraction was determined using an HPLC method
ric acid), hydrogen peroxide, NBT (NitroBlue Tetrazolium), NADH based on Inbaraj et al. (2008). A binary mobile phase of metha-
(b-nicotinamide adenine dinucleotide), 2-deoxy-D-ribose, PMS nol/acetonitrile/water (81:14:5, v/v/v) (A) and methylene chloride
(phenazine methosulphate), and proteinase (from Bacillus licheni- (B) was used with the following gradient elution: 84% A and 16% B
formis Subtilisin A) were from Sigma. Disodium hydrogen phos- initially, decreased to 83% A in 22 min, 45% A in 40 min, 25% A in
phate, sodium dihydrogen phosphate, manganese oxide, ascorbic 55 min and returned to the initial ratio in 60 min. A Waters YMC
acid, sodium nitrite, aluminium chloride, EDTA (ethylenediamine- C30 column was used for separation with detection at 450 nm, col-
tetraacetic acid), BHT (butylated hydroxytoluene), iron(II) sulphate umn temperature at 25 °C and flow rate at 1 ml/min. The various
heptahydrate and butylated hydroxyanisole were from Nacalai carotenoids were identified and quantified using an HPLC-MS tech-
Tesque Co. (Kyoto, Japan). The Folin–Ciocalteu reagent was from nique, as described in our previous study (Inbaraj et al., 2008).
Merck (Darmstadt, Germany). Sodium hydroxide was from Rie-
del-de Haën Co. (Seezle, Germany). Anhydrous sodium sulphate 2.4. Quantitation of carotenoids
was from Nacalai Tesque Co. Polyacryl cotton was from Applied
Separation Co. (Allentown, PA, USA). BSA (bovine serum albumin) For quantitation, an internal standard, b-apo-80 -carotenal, was
was from Sigma. Sulphuric acid, boric acid, methyl red and bromo- used for calibration of each standard by taking ten concentrations
phenol blue were from Panreac Co. (Barcelona, Spain). of all-trans-zeaxanthin (0.5, 1, 5, 10, 20, 30, 50, 100, 150 and
170 lg/ml) and seven concentrations each of all-trans-b-crypto-
xanthin and all-trans-b-carotene and mixing with b-apo-80 -carote-
2.2. Instrumentation nal for a final concentration of 15 lg/ml. Standard curves were
prepared by plotting concentration ratio of carotenoid standard
The HPLC instrument was composed of a column temperature to internal standard against its area ratio and the regression
controller (G1316A), a degasser (G1379A), a quaternary pump equations were y = 0.6689x  0.0484, y = 0.7869x  0.0175 and
(G1311A), and a photodiode-array detector (G1315B), all of which y = 0.9425x  0.0293 for all-trans-zeaxanthin, all-trans-b-crypto-
were from Agilent Technologies (Palo Alto, CA, USA). An evapora- xanthin and all-trans-b-carotene, respectively, with the R2 values
tive light scattering detector (ELSD) 800 was from Alltech (Deer- being 0.9982, 0.9985 and 0.9949. Because of the unavailability of
field, IL, USA). The quadrupole LC/MS (model 6130), with multi- commercial standards of cis-carotenoids, the quantitation of the
mode ion source (ESI and APCI), was from Agilent Technologies. cis isomers was based on the standard curves of their correspond-
The preparative column (C26140) was from GE Amersham Biosci- ing all-trans forms, while neoxanthin was quantified by multiply-
ences Co. (Uppsala, Sweden). ing concentration of b-apo-80 -carotenal by peak area ratio of
The YMC C30 reversed-phase column (250  4.6 mm I.D., parti- neoxanthin to b-apo-80 -carotenal. For determination of both detec-
cle size 5 lm) and C30 guard column (6 x 4.6 mm I.D.) were from tion limit (DL) and quantitation limit (QL), three concentrations
the Waters Co. (Milford, MA, USA). The Vydac 201TP54 C18 column (0.025, 0.05 and 0.1 lg/ml) of each standard were prepared and
(250  4.6 mm I.D., particle size 5 lm) was from the Vydac Co. analysed by HPLC. Based on S/N P 3 and S/N P 10, the DL and
186 C.C. Wang et al. / Food Chemistry 120 (2010) 184–192

QL were calculated to be 0.05 and 0.15, 0.025 and 0.075, and 0.025 kaempferol, respectively. For determination of DL and QL, four con-
and 0.075 lg/ml, respectively, for all-trans-zeaxanthin, all-trans-b- centrations of 0.05, 0.1, 0.5 and 1.0 lg/ml for each phenolic acid and
cryptoxanthin and all-trans-b-carotene. flavonoid standard were prepared, and 20 ll were injected into the
HPLC system three times. Based on the criteria described for carote-
2.5. Extraction and preparation of flavonoids and phenolic acids noids, the DL and QL for chlorogenic acid, caffeic acid, p-coumaric
acid, rutin and kaempferol were determined to be 0.1 and 0.3, 0.1
A method, based on Lu (2006), was modified to extract flavo- and 0.3, 0.05 and 0.15, 0.5 and 1.5, and 1.0 and 3.0 lg/ml,
noids and phenolic acids (acidic flavonoids) from L. barbarum. A respectively.
10 g powder sample of L. barbarum fruits was mixed with 100 ml
of 80% ethanol, after which the mixture was shaken in a 90 °C 2.7. Determination of total phenolic compounds in the phenolic acid
water bath for 2 h, followed by centrifuging at 6000 rpm for fraction
25 min, and collecting and filtering supernatant to remove impuri-
ties. For preparative chromatography of flavonoids and phenolic A method, based on Kao and Chen (2006), was used and the
acids, a 5 ml of extract (pH 7) was poured into a glass column amount of total phenolic compounds was expressed as gallic acid
(300  25 mm I.D.) containing 25 g of adsorbent Cosmosil equivalents. Six concentrations of 50, 100, 200, 250, 300 and
140C18-OPN, which was preactivated with 100 ml of methanol 350 lg/ml of gallic acid standard were prepared and 50 ll of each
and 50 ml of deionized water. Then 100 ml of deionized water were collected, followed by adding 200 ll of Folin–Ciocalteau re-
were added to elute phenolic acids and 40 ml of methanol to elute agent, mixing thoroughly, standing at room temperature for
flavonoids. Next, both fractions were separately collected, evapo- 5 min, adding 1000 ll of 15% sodium carbonate solution, reacting
rated to dryness under vacuum, dissolved in 10 ml of acetoni- at room temperature for 60 min and measuring the absorbance
trile–water (1:1, v/v), filtered through a 0.2 lm membrane filter at 750 nm. The gallic acid standard curve was obtained by plotting
and 20 ll were injected for HPLC analysis of phenolic acids and concentration against absorbance and the regression equation was
flavonoids in each fraction. A ternary solvent system of 0.1% formic used to determine the amount of total phenolic compound for a
acid solution (A), tetrahydrofuran (B) and acetonitrile (C), with the 50 ll sample from acidic flavonoid (phenolic acid) fractions.
following gradient elution, was used: 98% A and 2% B initially,
maintained for 3 min, decreased to 93% A, 2% B and 5% C in 2.8. Determination of total flavonoids in the flavonoid fraction
11 min, 76% A, 2% B and 22% C in 20 min, 68% A, 2% B and 30% C
in 25 min, 63% A, 2% B and 35% C in 30 min, and returned to 98% A method, based on Kao and Chen (2006), was used for determi-
A and 2% B in 35 min. The various flavonoids and phenolic acids nation and the total amount of flavonoids was expressed as cate-
were identified and quantified, based on an HPLC-DAD-ESI-MS chin equivalents. One mg of catechin standard was dissolved in
technique developed by Lu (2006). 1 ml of ethanol/water (3:7, v/v) and 5 concentrations of 1, 5, 10,
25, and 50 lg/ml of catechin standard were prepared. A 500 ll
solution of each was collected and mixed with 75 ll of 5% sodium
2.6. Quantitation of phenolic acids and flavonoids
nitrite solution, followed by mixing thoroughly, standing at room
temperature for 5 min, adding 150 ll of 10% aluminium chloride
Both phenolic acids and flavonoids were quantified using an
solution, standing for a further 5 min, adding 500 ll of 1 N sodium
internal standard, taxifolin, which was dissolved in acetonitrile–
hydroxide solution and measuring the absorbance at 510 nm. The
water at a concentration of 200 lg/ml. Seven concentrations of 1,
catechin standard curve was prepared by plotting concentration
5, 7, 10, 15, 17 and 20 lg/ml, for chlorogenic acid, caffeic acid, p-
against absorbance and the regression equation was used to deter-
coumaric acid, quercetin and kaempferol, were separately pre-
mine the total content of flavonoids for a 500 ll sample from the
pared in acetonitrile–water (1:1, v/v). Similarly, seven concentra-
neutral flavonoid fraction.
tions of 5, 7, 10, 15, 17, 20 and 25 lg/ml were prepared for rutin
in acetonitrile–water (1:1, v/v). All the seven concentrations of
2.9. Extraction and preparation of polysaccharides
each phenolic acid and flavonoid standard were then mixed with
internal standard taxifolin for a final concentration at 20 lg/ml
A 10 g fruit sample of L. barbarum was mixed with 100 ml of
and 20 ll were injected into the HPLC system. By plotting concen-
deionized water and homogenised for 1 min, followed by heating
tration ratio (phenolic acid or flavonoid standard vs internal stan-
in boiling water (100 °C) for 30 min and centrifuging at 6000 rpm
dard) against its area ratio, the calibration curve of each standard
for 25 min. After filtering through a filter paper to remove impuri-
was prepared and the regression equations determined were
ties, the crude extract was concentrated under vacuum at 40 °C
y = 2.734x  0.035, y = 1.350x  0.051, y = 0.716x  0.052,
and diluted to 50 ml with deionized water. The crude extract of
y = 12.00x  0.092 and y = 2.810x + 0.027 for chlorogenic acid, caf-
polysaccharide (CE) was used for the subsequent antioxidant activ-
feic acid, p-coumaric acid, rutin and kaempferol, respectively, with
ity study. Then 250 ml of 95% ethanolic solution was added for pre-
R2 values of all being higher than 0.99. The amounts of phenolic
cipitation overnight, and the supernatant was removed after
acids or flavonoids in L. barbarum fruits were quantified using
centrifugation. The precipitate was vacuum-dried at 40 °C to ob-
the following formula:
tain crude polysaccharide (CP) and ground into powder, which
phenolic acid or flavonoid ðlg=gÞ was used for the antioxidant activity study. 0.2 g of dried polysac-
charide was collected and dissolved in 40 ml of 50 mM phosphoric
½ðA=RRFÞ=Ai  Ci  volume of extract  recovery
¼ acid-buffered solution (pH 8), after which the mixture was heated
Ws
in a 60 °C water bath for 5 min and 1 ml of 2.5 U/ml of proteinase
where relative response factor (RRF) = (A/Ai)  (C/Ci), A = peak area (Type III from Bacillus L.) was added to react at 60 °C (pH 8) for 4 h.
of phenolic acid or flavonoid, Ai = peak area of internal standard, This condition was selected as a high degree of protein hydrolysis
C = concentration of phenolic acid or flavonoid (lg/ml), Ci = concen- could be attained. Next, 40 ml of 5% trichloroacetic acid were
tration of internal standard (lg/ml) and Ws = weight of sample (g). added to terminate the reaction, followed by cooling for 30 min
Because of the unavailability of commercial standards, caffeoylqui- and centrifuging at 10,000g for 15 min to collect the supernatant.
nic acid, quercetin-diglycoside and kaempferol-3-O-rutinoside were Five millilitres of filtrate were poured into a glass column
quantified using the standard curves of chlorogenic acid, rutin and (30  2.4 cm) containing 5 g of DEAE-Sepharose CL-6B, which
 C.C. Wang et al. / Food Chemistry 120 (2010) 184–192 187

was previously equilibrated with 150 ml of deionized water and water and 1 ml of 0.1% ferrous chloride. The solution was mixed
the neutral fraction of polysaccharide (LBPN) was eluted with thoroughly for 10 min and the absorbance was measured at
200 ml of deionized water and the acidic fractions of polysaccha- 700 nm. The higher the absorbance, the better was the reducing
rides LBPa1, LBPa2 and LBPa3 were eluted with 200, 100 and power.
100 ml of 0.1, 0.26 and 0.6 M of sodium hydroxide, respectively.
All of the LBPN, LBPa1, LBPa2 and LBPa3 were used for antioxidant 2.14. Chelating of ferrous ion
activity study.
The method of Kao and Chen (2006) was used. One ml solutions
2.10. Quantitation of polysaccharides of carotenoids, polysaccharides, flavonoids, a-tocopheol, ascorbic
acid, EDTA or BHA were mixed with 3.7 ml of methanol and
The amounts of polysaccharide in CE, CP, LBPN, LBPa1, LBPa2 0.1 ml of ferrous chloride solution (2 mM), separately, followed
and LBPa3 were determined using the phenol–sulphuric acid by mixing thoroughly for 30 min, and 0.2 ml of ferrozine solution
method (Dubois, Gilles, Hamilton, Rebers, & Smith, 1956). Briefly, (5 mM) was added. The mixture was reacted at room temperature
0.2 ml of polysaccharide solution was mixed with 0.2 ml of 5% phe- for 10 min, and the absorbance was measured at 562 nm. A lower
nol solution, followed by adding 1 ml of concentrated sulphuric absorbance indicated a better ferrous ion-chelating ability. The
acid and shaking the mixture for 30 min. The absorbance was mea- chelating effect was calculated by using the formula described by
sured at 490 nm and used to quantify polysaccharide, based on the Kao and Chen (2006).
standard curve of glucose, which was prepared by plotting six con-
centrations (10–100 lg/ml) against their absorbances. The poly- 2.15. Scavenging ability of hydroxyl free radical
saccharides in each fraction were further diluted to adjust
concentration within the linear range of the standard curve. A method based on Ghiselli, Nardini, Baldi, and Scaccini (1998)
and Ren et al. (2008) was used. A 200 ll solution of flavonoids,
2.11. Scavenging of DPPH free radical carotenoids, polysaccharides, a-tocopherol, ascorbic acid, EDTA
or BHA was mixed with 500 ll each of phosphate-buffered solution
The method of Kao and Chen (2006) was used. One ml of flavo- (pH 7.4), EDTA-ferrous ion solution (10 mM) and a-deoxyribose
noids, polysaccharides, carotenoids, Vit E, ascorbic acid, EDTA or solution (10 mM), after which each solution was mixed thoroughly
BHA solutions was mixed with a 0.2 ml of 1 mM DPPH solution and 100 ll of hydrogen peroxide (10 mM) were added and the
(in ethanol) separately, after which each solution was mixed thor- whole shaken for a few seconds. The mixture was settled at 37 °C
oughly and then stood in the dark for 30 min, and the absorbance for 15 min, and 500 ll each of 2.8% trichloroacetic acid and 1%
was measured at 517 nm. The scavenging effect (%) was deter- TBA (thiobarbituric acid) reagent were added. After heating at
mined using the formula described by Kao and Chen (2006). 100 °C for 10 min, the absorbance was measured at 532 nm. A low-
er absorbance indicated a better hydroxyl radical-scavenging abil-
2.12. Trolox equivalent antioxidant capacity (TEAC) ity. The scavenging effect was determined using the formula
described by Ren et al. (2008).
The method of Kao and Chen (2006) was used. A 20 ml solution
of 1 mM ABTS+ was filtered through a filter paper containing 2 g of
2.16. Scavenging ability of superoxide anion
manganese dioxide, after which the filtrate was passed through a
0.2 lm PVDC syringe filter to obtain a blue–green solution. Then
The method of Li, Zhou, and Han (2006) was used. A 0.2 ml solu-
the solution was diluted with 5 mM phosphate-buffered saline
tion of flavonoids, carotenoids, polysaccharides, a-tocopherol,
(pH 7.7) and the absorbance was 1.00 ± 0.02 at 734 nm. For prepa-
ascorbic acid, EDTA or BHA was mixed with 0.4 ml of 150 lM nitro
ration of the trolox standard curve, one ml of ABTS+ solution was
blue tetrazolium (NBT) in phosphate-buffered solution (pH 7.4)
mixed with 0.1 ml of trolox standard solutions (50, 100, 150, 200,
and 0.4 ml of 470 lM b-nicotinamide adenine dinucleotide (b-
300 and 400 lM) prepared in 5 mM of phosphate-buffered saline
NADH), followed by adding 0.05 ml of 60 lM phenazine methosul-
(pH 7.7), after which each solution was mixed thoroughly for
phate (PMS), shaking for few seconds, standing at room tempera-
30 s and then stood for a further 30 s, and the absorbance was
ture for 5 min and measuring the absorbance at 560 nm. A lower
measured at 734 nm. The trolox standard curve was obtained by
absorbance indicated a better superoxide anion-scavenging ability.
plotting concentration against absorbance, and the regression
The scavenging effect was calculated by using the formula de-
equation was automatically calculated. Next, 0.1 ml of flavonoids,
scribed by Li et al. (2006).
carotenoids, polysaccharides, a-tocopherol, ascorbic acid, EDTA
or BHA was mixed with 1 ml of ABTS+ solution separately, and
each solution was mixed thoroughly for 30 s and stood for a further 2.17. Statistical analysis
30 s, after which the absorbance was measured at 734 nm. Then
the relative trolox concentration was obtained, based on the All the treatments were performed in duplicate and the data
regression equation, and the higher the trolox concentration, the were analysed by variance (ANOVA) and Duncan’s multiple range
better was the antioxidant activity. test for statistical significance (a = 0.05) by using a SAS software
system (2004).
2.13. Reducing power
3. Results and discussion
The method of Kao and Chen (2006) was used. One ml solutions
of carotenoids, polysaccharides, flavonoids, a-tocopherol, ascorbic 3.1. Composition of functional components in fractions of L. barbarum
acid, EDTA or BHA were mixed with 0.5 ml of 0.2 M phosphate-buf-
fered solution (pH 6.6) and 0.5 ml of 1% potassium ferricyanide, The HPLC chromatograms of different carotenoid fractions, as
separately, after which each mixture was heated in a 50 °C water well as phenolic acid and flavonoid fractions isolated from L. bar-
bath for 20 min and cooled immediately, followed by adding barum by preparative column chromatography, are shown in
0.5 ml of 10% trichloroacetic acid. After mixing thoroughly, one Figs. 1 and 2, respectively. Table 1 shows contents of functional
ml of supernatant was collected and treated with 1 ml of deionized components in different fractions of L. barbarum extract. In the
188 C.C. Wang et al. / Food Chemistry 120 (2010) 184–192

Fig. 1. HPLC chromatograms of three different carotenoid fractions isolated from L. barbarum fruits. (A) b-Carotene fraction: peak 1, 13- or 130 -cis-b-carotene; peak 2, all-
trans-b-carotene; peak 3, 9- or 90 -cis-b-carotene. (B) Neoxanthin and b-cryptoxanthin fraction: peak 4, neoxanthin; peak 5, all-trans-b-cryptoxanthin; peak 6, 9- or 90 -cis-b-
cryptoxanthin. (C) Zeaxanthin fraction: peak 7, 9- or 90 -cis-zeaxanthin; peak 8, 13- or 130 -cis-zeaxanthin; peak 9, 15- or 150 -cis-zeaxanthin; peak 10, all-trans-zeaxanthin;
peak 11, 9- or 90 -cis-zeaxanthin.

carotenoid fraction, all-trans-zeaxanthin and its cis isomers con- metric method described in the preceding section to avoid quan-
stituted the largest portion (1403 lg/g), followed by neoxanthin titation error. This was also used as a basis for calculation of
and cryptoxanthin fraction (72.1 lg/g) and b-carotene fraction concentration for subsequent antioxidant study. The total amount
(35.9 lg/g). In total, 10 carotenoids were identified in L. barbarum of flavonoid in the flavonoid fraction, based on catechin equiva-
extract, including all-trans forms of zeaxanthin, b-cryptoxanthin lents, was shown to be 116 lg/g whereas the total phenolic com-
and b-carotene, as well as their cis isomers and neoxanthin. In pounds, based on gallic acid equivalents, was 340 lg/g, which was
the flavonoid fraction, 3 neutral flavonoids and 4 phenolic acids substantially greater than that by HPLC analysis. Nevertheless,
were present, with the former containing quercetin-diglycoside this difference may be also due to overestimation of phenolic con-
(66.0 lg/g), rutin (42.0 lg/g) and kaempferol-3-O-rutinoside tent by the Folin–Ciocalteu method, as the reagent added can also
(11.3 lg/g), and the latter containing chlorogenic acid (12.4 lg/ react with non-phenolic reducing compounds, such as organic
g), caffeoylquinic acid (0.34 lg/g), caffeic acid (3.73 lg/g) and p- acids, sugars and amino acids. It is also possible that different
coumaric acid (6.06 lg/g). Obviously the neutral flavonoid was phenolic compounds may respond differently to Folin–Ciocalteu
present in a much larger amount than the phenolic acids, on reagent. For instance, while catechin, caffeic acid, rutin and gallic
the basis of HPLC analysis. For the polysaccharide fraction, a level acid show similar absorption behaviours with the reagent, several
of 25.6 lg/g was shown for the neutral polysaccharide, whereas flavonoids may show a low absorption, resulting in underestima-
26.9 lg/g was found for the acidic polysaccharide, including tion of total phenolic content. Likewise, both carotenoid extract
LPBa1 (9.26 lg/g), LPBa2 (9.26 lg/g) and LPBa3 (8.41 lg/g). As and zeaxanthin fraction were selected for antioxidant activity
several peaks in the flavonoid fraction remained unidentified on study as the former contained 10 carotenoids and the latter con-
the HPLC chromatogram, the total amounts of neutral flavonoid stituted the largest portion, with the total amount in Table 1
and phenolic acids were also determined using the spectrophoto- being used for concentration calculation, so also were the neutral
 C.C. Wang et al. / Food Chemistry 120 (2010) 184–192 189

Fig. 2. HPLC chromatograms of phenolic acid (A) and flavonoid (B) fractions isolated from L. barbarum fruits. (A) phenolic acid fraction: peak 1, chlorogenic acid; peak 2,
caffeoylquinic acid; peak 3, caffeic acid; peak 4, p-coumaric acid. (B) flavonoid fraction: peak 5, quercetin-diglycoside; peak 6, rutin; peak 7 kaempferol-3-O-rutinoside.

Table 1 acid, BHA and EDTA. With the exception of EDTA, most treatments
Contents of functional components in different fractions of L. barbarum extract. were effective in scavenging DPPH free radicals, with Vit C showing
Fraction Components Content (mg/g)a the most pronounced effect (96.8%), followed by acidic flavonoids
Carotenoid
(Fla-A, 96.6%), Vit E (94.1%), BHA (93.7%), rutin (87.7%), LBPa3
b-Carotene fraction 13- or 13’-cis-b-carotene 8.97 ± 0.00 (84.9%), neutral flavonoids (82.5%), carotenoid extract (Caro-ex-
All-trans-b-carotene 16.6 ± 0.35 tract, 70.6%), LBPa2 (69.9%), crude polysaccharide extract (CE,
9- or 9’-cis-b-carotene 10.3 ± 0.24 66.6%), crude polysaccharide (CP, 58.2%), LBPa1 (52.5%), LBPN
Neoxanthin and Neoxanthin 13.2 ± 0.04 (38.1%), zeaxanthin standard (Zea-std, 25.2%) and zeaxanthin frac-
cryptoxanthin fraction All-trans-b-cryptoxanthin 53.3 ± 0.73 tion (19.5%). This outcome clearly revealed a better antioxidative ef-
9- or 9’-cis-b-cryptoxanthin 5.61 ± 0.03
fect of acidic polysaccharides compared to neutral polysaccharides,
Zeaxanthin fraction 9- or 9’-cis-zeaxanthin 39.3 ± 5.61 which should be due to the ability of galacturonic acid present in the
13- or 13’-cis-zeaxanthin 4.85 ± 0.80
former to chelate metal ion and in turn scavenge DPPH radical. A
15- or 15’–cis-zeaxanthin 32.8 ± 2.65
All-trans-zeaxanthin 1326 ± 14.5 lower DPPH free radical-scavenging activity occurred for both the
zeaxanthin fraction and the zeaxanthin standard than for the carot-
Flavonoid enoid extract, which may be accounted for by the presence of some
Neutral fraction Quercetin-diglycoside 66.0 ± 0.53 other lipophilic antioxidant components in the carotenoid extract.
Rutin 42.0 ± 0.39
Kaempferol-3-O-rutinoside 11.3 ± 0.34
In addition, the presence of the b-ionone ring in zeaxanthin may de-
crease the resonance effect of p electrons because of steric hin-
Acidic fraction Chlorogenic acid 12.4 ± 0.67
Caffeoylquinic acid 0.34 ± 0.30
drance to lower the free radical-scavenging activity (Jiménez-
Caffeic acid 3.73 ± 0.26 Escrig, Jiménez-Jiménez, Sánchez-Moreno, & Saura-Calixto, 2000).
p-Coumaric acid 6.06 ± 0.16 Moreover, the absorbance measured at 517 nm may interfere with
absorption of the chromophore in the zeaxanthin fraction and zea-
Polysaccharide
xanthin standard. Like the acidic polysaccharide, the acidic flavo-
Content (mg/g)
Crude fraction CPb 57.2 ± 0.29 noids also showed a larger DPPH free radical-scavenging activity
Neutral fraction LPBNc 25.6 ± 1.32 than did the rutin standard and the neutral flavonoid.
Acidic fraction LPBa1d 9.26 ± 0.06
LPBa2d 9.26 ± 0.57 3.3. TEAC value
LPBa3d 8.41 ± 1.49
Crude extract CEe 580 ± 20.0
The TEAC values of carotenoids, flavonoids, polysaccharides, a-
a
Means of duplicate analyses ± standard deviation. tocopherol, ascorbic acid, BHA and EDTA are shown in Fig. 3B. The
b
Crude polysaccharide.
c TEAC value was used to assess the scavenging ability of ABTS+ free
Neutral polysaccharide.
d
Acidic polysaccharides with different molecular weights. radicals with the water-soluble vitamin E (trolox) as reference
e
Crude extract of polysaccharide. standard for comparison. Of the various treatments, BHA, Vit C,
neutral flavonoids, rutin and acidic flavonoids were the most effi-
polysaccharide (LPBN), acidic polysaccharide (LPBa1, LPBa2 and cient in scavenging ABTS+, which were equivalent to levels of
LPBa3), crude polysaccharide (CP) and crude extract of polysac- trolox at 367, 366, 365, 364 and 331 lM, respectively. This phe-
charide (CE). nomenon is similar to a report by Soobrattee, Neergheen, Luxi-
mon-Ramma, Aruoma, and Bahorun (2005), who observed the
3.2. DPPH free radical-scavenging activity ability of flavonoids and phenolic acids, in scavenging ABTS+, to
follow the order: procyanidin > flavanol > flavonol > hydroxycin-
Fig. 3A shows the DPPH free radical-scavenging activities of namic acid > simple phenolic acids. Obviously, the high antioxidant
carotenoids, flavonoids, polysaccharides, a-tocopherol, ascorbic activity of flavonoids can be attributed to hydroxy groups in the
190 C.C. Wang et al. / Food Chemistry 120 (2010) 184–192

120 A a b a b
d c

Scavenging effect (%)

96.6 96.8
100 e 94.1 93.7
84.9 87.7
g f f 82.5
80 h 69.9 70.6
66.6
58.2 i
60 j
52.5

38.1
40 l k
25.2
19.5
20 m
0.0
0
CE CP LBPN LBPa1 LBPa2 LBPa3 Caro- Zea Zea-std Fla-A Fla-N Rutin Vit E Vit C BHA EDTA
extract

a a a a
400 B b 365.2 364.2 365.6 366.8

350 330.5
Conc. of Trolox (µM)

300 c c
246.4 243.7
250
200
d
150 e 110.3
95.2
100 f f g f f
h 37.7 42.9 37.9 37.8 i
50 14.2
26.1
0.0
0
CE CP LBPN LBPa1 LBPa2 LBPa3 Caro- Zea Zea-std Fla-A Fla-N Rutin Vit E Vit C BHA EDTA
extract

Fig. 3. DPPH free radical-scavenging activity (A) and TEAC values (B) of carotenoids, flavonoids, polysaccharides, a-tocopherol, ascorbic acid, BHA and EDTA. LBPN: neutral
polysaccharide of Lycium fruit; LBPa1: acidic polysaccharide of Lycium fruit (eluted with 0.1 M NaCl); LBPa2: acidic polysaccharide of Lycium fruit (eluted with 0.26 M NaCl);
LBPa3: acidic polysaccharide of Lycium fruit (eluted with 0.6 M NaCl); CE: crude extract of polysaccharide; CP: crude polysaccharide; Caro-extract: carotenoids extract; Zea:
zeaxanthin fraction; Zea-std; zeaxanthin standard; Fla-A: flavonoids (acidic group); Fla-N: flavonoids (neutral group); Rutin: rutin standard. Data with different letters are
significantly different at P < 0.05.

A- and B-rings, and the larger the number of hydroxy groups, the As expected, EDTA possessed the maximum chelating activity of
higher is the capacity to scavenge free radicals (Kao & Chen, ferrous ion (99.9%), followed by LBPa2 (95.3%), acidic flavonoid
2006). Both carotenoid extract and Vit E showed similar effects (94.4%) and neutral flavonoid (86.4%). As described previously,
in TEAC values with 246 and 244 lM, respectively, whereas lower the acidic groups in LBPa2 should contribute to the metal ion-che-
TEAC values of 110 and 37.8 lM were found for the zeaxanthin lating ability, so also should the hydroxy groups at C-30 and C-40 of
fraction and zeaxanthin standard. As mentioned above, the pres- the B-ring, as well as at C-5 of the A-ring and keto group at the C-
ence of some other antioxidant components in the carotenoid ex- ring of flavonoids (Kao & Chen, 2006). In contrast, a poor ferrous
tract should be responsible for this effect. However, all the ion-chelating effect was observed for the carotenoid extract
polysaccharides showed low TEAC values, ranging from 14.2 lM (1.8%), zeaxanthin fraction (9.1%) and zeaxanthin standard (0%),
for CE to 95.2 lM for LBPa3, which should be due to the absence which should be caused by absence of the phenolic-type structure.
of phenolic-type structures in polysaccharides. Surprisingly, ascorbic acid (Vit C), a reducing agent containing a
carboxyl group, and rutin, a flavonoid containing 4 hydroxy groups,
3.4. Reducing power both exhibited a low metal ion-chelating activity, 3.4% for the for-
mer and 4.1% for the latter. It was assumed that ascorbic acid may
Fig. 4A shows the reducing powers of carotenoids, flavonoids, undergo degradation amid the presence of iron(II) chloride in the
polysaccharides, a-tocopherol, ascorbic acid, BHA and EDTA. A peak system used for determination of chelated ferrous ion (Kao & Chen,
reducing power was observed for acidic flavonoids, with ABS at 2006). Comparatively, both acidic flavonoid and neutral flavonoid
2.79, followed by neutral flavonoid (1.80), Vit C (1.00), BHA showed a much greater metal ion-chelating ability than did rutin,
(0.84), rutin (0.54) and Vit E (0.31). Yet, a weak reducing power as some other functional components, e.g. polyphenol compounds
was found for the other treatments, ranging from 0.05 in ABS for in the former, may play a significant role. Similar to reducing
CE to 0.22 for LBPN. Moran, Klucas, Grayer, Abain, and Becana power, a poor ferrous ion-chelating effect occurred for LBPN
(1997) studied the prooxidant and antioxidant properties of iron (5.0%) and CE (0%), whereas moderate chelating activities were
complex with phenolic compounds from soybean nodules and re- found for CP (38.8%), LBPa3 (42.9%) and LBPa1 (47.0%).
ported the hydroxy groups at C-30 and C-40 of the B-ring to be more
active in reducing iron concentration, while a hydroxy group at C-3 3.6. Scavenging activity of superoxide anion
and keto group at C-4 was less active, with the keto group at C-4 and
the hydroxy group at C-5 being inactive. However, no activity was The superoxide anion-scavenging activities of carotenoids,
observed for the keto group at C-4 and hydroxy group at C-3. In flavonoids, polysaccharides, a-tocopherol, ascorbic acid, BHA and
our experiment, the neutral flavonoid was shown to contain rutin, EDTA are shown in Fig. 5A, characterised by a poor superoxide an-
quercetin-diglycoside and kaempferol-3-O-rutinoside, all of which ion-scavenging activity for most treatments, with the exception of
belong to the flavonols and the presence of a catechol group in rutin and LBPa3, which amounted to 28.3% and 14.1%, respectively.
the B-ring should contribute to the reducing power. It has been well documented that, for flavonoids, the keto group at
C-4 and hydroxy group at C-3 or C-5 of the C-ring were effective in
3.5. Chelating of ferrous ion scavenging superoxide anion, and the greater the number of hydro-
xy groups, the better is the scavenging activity (Das & Pereira, 1990;
Fig. 4B shows the ferrous ion-chelating ability of carotenoids, van Ackerm et al., 1995). In addition, the hydroxy groups at C-30 and
flavonoids, polysaccharides, a-tocopherol. Vit C, BHA and EDTA. C-40 of the B-ring in rutin should also be responsible for this effect
 C.C. Wang et al. / Food Chemistry 120 (2010) 184–192 191

a
3.00 A 2.79

2.50
b

ABS at 700 nm
2.00 1.80

1.50 c
d
1.00
1.00 e 0.84
gh g ij 0.54 f
k k j hi kl
0.50 l 0.20 0.22
0.31 m
0.14 0.16 0.18
0.05 0.090.09 0.07 0.00
0.00
CE CP LBPN LBPa1 LBPa2 LBPa3 Caro- Zea Zea-std Fla-A Fla-N Rutin Vit E Vit C BHA EDTA
extract

120 B ab b
a
c 99.9
100 95.3 94.4
Chelating effect (%)

86.4

80
d de
60 e 47.0
42.9
38.8
40
gf
ghf gh f gfh gfh
20 h 9.1 h 5.5 gh
5.0 1.8 4.1 3.4 4.0
0.0 0.0
0
CE CP LBPN LBPa1 LBPa2 LBPa3 Caro- Zea Zea-std Fla-A Fla-N Rutin Vit E Vit C BHA EDTA
extract

Fig. 4. Reducing power (A) and metal ion-chelating ability (B) of carotenoids, flavonoids, polysaccharides, a-tocopherol, ascorbic acid, BHA and EDTA. The expansion of
abbreviations and statistical significance are the same as given for Fig. 3.

a
35 A 28.3
30
Scavenging effect (%)

25
20 b
15 14.1 cd
c
10 7.5
5.1 de
5 e e e e e e e e e e e
1.6
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.5 0.0 0.0
0
CE CP LBPN LBPa1 LBPa2 LBPa3 Caro- Zea Zea-std Fla-A Fla-N Rutin Vit E Vit C BHA EDTA
extract

a
100
90
B b
79.4
87.0
b
Scavenging effect (%)

79.3
80 c c c c
68.1
70 66.2 d 65.4 67.2
57.6
60 e
50 gf gf 42.6
ef ef
36.3 36.0
40.0 g 39.1
40 32.7
30
20 h h
10 5.9
2.6
0
CE CP LBPN LBPa1 LBPa2 LBPa3 Caro- Zea Zea-std Fla-A Fla-N Rutin Vit E Vit C BHA EDTA
extract

Fig. 5. Superoxide anion-(A) and hydroxyl radical-(B) scavenging activity of carotenoids, flavonoids, polysaccharides, a-tocopherol, ascorbic acid, BHA and EDTA. The
expansion of abbreviations and statistical significance are the same as given for Fig. 3.

because of the hydrogen donation ability (Zhishen, Mengcheng, & are shown in Fig. 5B. Of all the treatments, carotenoids showed the
Jianming, 1999). Conversely, a low content of rutin in acidic flavo- most distinct effect in scavenging hydroxyl radicals, which equalled
noids and neutral flavonoids may minimise the superoxide anion- 87.0, 79.4 and 79.3% for the zeaxanthin fraction, carotenoid extract
scavenging activity. But for the various treatments of polysaccha- and zeaxanthin standard, respectively. The long chain of conjugated
rides and carotenoids, the weak superoxide anion-scavenging double bonds in carotenoids has been demonstrated to play a vital
activity could be due to a lack of phenolic-type structure. role in scavenging hydroxyl radicals (Trevithick-Sutton, Foote, Col-
lins, & Trevithick, 2006). A similar phenomenon was observed by
3.7. Hydroxyl radical-scavenging activity Trevithick-Sutton et al. (2006), who used an electron spin reso-
nance spectrophotometer to study the hydroxyl radical-scavenging
The hydroxyl radical-scavenging activities of carotenoids, flavo- activity of various carotenoids, with zeaxanthin possessing the
noids, polysaccharides, a-tocopherol, ascorbic acid, BHA and EDTA maximum inhibition effect, followed by b-carotene, lycopene and
192 C.C. Wang et al. / Food Chemistry 120 (2010) 184–192

lutein. It may be inferred that the non-conjugated double bond Fraser, P. D., & Bramley, P. M. (2004). The biosynthesis and nutritional uses of
carotenoids. Progress in Lipid Research, 43, 228–265.
within the cyclic ring is more susceptible to bonding with free rad-
Ghiselli, A., Nardini, M., Baldi, A., & Scaccini, C. (1998). Antioxidant activity of
icals. However, for acidic flavonoids and neutral flavonoids, only different phenolic fractions separated from an Italian red wine. Journal of
moderate inhibition effects of 32.7% and 39.1% were shown, respec- Agricultural and Food Chemistry, 46, 361–367.
tively, which were much lower than the rutin standard (66.2%). As Havsteen, B. H. (2002). The biochemistry and medical significance of the flavonoids.
Pharmacology and Therapeutics, 96, 67–202.
explained above, the hydroxyl groups at C-30 and C-40 of the B-ring Huie, C. W., & Di, X. (2004). Chromatographic and electrophoretic methods for
in rutin should contribute to this effect. For the control treatments, Lingzhi pharmacologically active components. Journal of Chromatography B, 812,
significant scavenging effects of 57.6%, 68.1%, 65.4% and 67.2% were 241–257.
Inbaraj, B. S., Lu, H., Hung, C. F., Wu, W. B., Lin, C. L., & Chen, B. H. (2008).
found for Vit E, Vit C, BHA and EDTA, respectively, which were sub- Determination of carotenoids and their esters in fruits of Lycium barbarum
stantially higher than those of flavonoids and polysaccharides. Both Linnaeus. by HPLC–DAD–APCI–MS. Journal of Pharmaceutical and Biomedical
CP and CE exhibited poor scavenging activities of 2.6% and 5.9%, Analysis, 47, 812–818.
Jiménez-Escrig, A., Jiménez-Jiménez, I., Sánchez-Moreno, C., & Saura-Calixto, F.
respectively, whereas moderate inhibitions occurred for LBPN (2000). Evaluation of free radical scavenging of dietary carotenoids by the stable
(36.3%), LBPa1 (36.0%), LBPa2 (42.6%) and LBPa3 (40.0%). This out- radical 2,2-diphenyl-1-picryhydrazyl. Journal of the Science of Food and
come revealed that polysaccharides, purified through a column, Agriculture, 80, 1686–1690.
Kanaze, F. I., Gabrieli, C., Kokkalou, E., Georgarakis, M., & Niopas, I. (2003).
possessed a stronger free radical-scavenging activity than those Simultaneous reversed-phase high-performance liquid chromatographic
without purification. Furthermore, the galacturonic acid content method for the determination of diosmin, hesperidin and naringin in different
in acidic polysaccharide should be an imperative factor in enhanc- citrus fruit juices and pharmaceutical formulations. Journal of Pharmaceutical
and Biomedical Analysis, 33, 243–249.
ing antioxidative activity. This phenomenon was observed by Asker,
Kao, T. H., & Chen, B. H. (2006). Functional components in soybean cake and their
Mahmoud, and Ibrahim (2007), who isolated acidic polysaccharide effects on antioxidant activity. Journal of Agricultural and Food Chemistry, 54,
from Bacillus polymyxa NRC-A and reported the antioxidant activity 7544–7555.
of polysaccharides to be closely associated with the amount of Kohlmeier, L., & Hastings, S. B. (1995). Epidemiologic evidence of a role of
carotenoids in cardiovascular disease prevention. American Journal of Clinical
galacturonic acid. That is, the higher the content of galacturonic Nutrition, 62, 1370S–1376S.
acid, the better was the antioxidant activity. Li, X. L., Zhou, A., & Han, Y. (2006). Anti-oxidation and anti-microorganism activities
In conclusion, carotenoids in L. barbarum L. were extracted with of purification polysaccharide from Lygodium japonicum in vitro. Carbohydrate
Polymers, 66, 1–9.
hexane–ethanol–acetone–toluene (10:7:6:7, v/v/v/v), followed by Lu, H. (2006). Determination of carotenoids, phenolic acids and flavoniods in Lycium
80% ethanol for flavonoids and phenolic acids, and 100 °C boiling barbarum Linnaeus. Master’s thesis, Fu Jen Catholic University, Taipei, Taiwan.
water for polysaccharides. For preparative chromatography, Marchand, L. L. (2002). Cancer preventive effects of flavonoids. Biomedicine and
Pharmacotherapy, 56, 296–301.
carotenoids were separated by a column containing magnesium Moran, J. F., Klucas, R. V., Grayer, R. J., Abain, J., & Becana, M. (1997). Complexes of
oxide and diatomaceous earth at 1:5:1 (w/w), while flavonoids iron with phenolic compounds from soybean nodules and other legume tissues:
and phenolic acids were separated with a Cosmosil 140C18-OPN Prooxidant and antioxidant properties. Free Radical Biology & Medicine, 22,
861–870.
column and polysaccharides with a Sepharose CL-6B column. The Ren, J., Zhao, M., Shi, J., Wang, J., Jiang, Y., Cui, C., et al. (2008). Purification and
flavonoid fraction showed the most pronounced effect in scaveng- identification of antioxidant peptides from grass carp muscle hydrolysates by
ing DPPH and ABTS+ free radicals, chelating metal ions and reduc- consecutive chromatography and electrospray ionization-mass spectrometry.
Food Chemistry, 108, 727–736.
ing power, whereas the zeaxanthin and polysaccharide fractions
SAS. (2004). SAS procedures and SAS/graph user’s guide, version 6; software release 8.2
were the most effective in scavenging hydroxy free radicals and (TS2M0) of the SAS system for Microsoft windows. SAS Institute, Inc.: Cary, NC.
superoxide radical anions, respectively. The results of this study Sheng, J. C., Yu, F., Xin, Z. H., Zhao, L. Y., Zhu, X. J., & Hu, Q. H. (2007). Preparation,
may provide a basis for possible production of functional foods in identification and their antitumor activities in vitro of polysaccharides from
Chlorella pyrenoidosa. Food Chemistry, 105, 533–539.
the future with L. barbarum as a raw material. Snodderly, D. M. (1995). Evidence for protection against age-related macular
degeneration by carotenoids and antioxidant vitamins. American Journal of
References Clinical Nutrition, 62, 1448S–1461S.
Soobrattee, M. A., Neergheen, V. S., Luximon-Ramma, A., Aruoma, O. I., & Bahorun, T.
(2005). Phenolics as potential antioxidant therapeutic agents: mechanism and
Arai, Y., Watanabe, S., Kimira, M., Shimoi, K., Mochizuki, R., & Kinae, N. (2000).
actions. Mutation Research, 579, 200–213.
Dietary antioxidant flavonols, flavones and isoflavones by Japanese women and
Sun, Y., Tang, J., Gu, X., & Li, D. (2005). Water-soluble polysaccharides from Angelica
the inverse correlation between quercetin intake and plasma LDL cholesterol
sinensis (Oliv.) Diels: preparation, characterization and bioactivity. International
concentration. Journal of Nutrition, 130, 2243–2250.
Journal of Biological Macromolecules, 36, 283–289.
Asker, M. M. S., Mahmoud, M. G., & Ibrahim, G. S. (2007). Structural characterization
Trevithick-Sutton, C. C., Foote, C. S., Collins, M., & Trevithick, J. R. (2006). The retinal
and biological activity of acidic polysaccharide fractions isolated from Bacillus
carotenoids zeaxanthin and lutein scavenge superoxide and hydroxyl
polymyxa NRC-A. Journal of Applied Science Research, 3, 1170–1177.
radicals: A chemiluminescence and ESR study. Molecular Vision, 12, 1127–
Cooper, D. A., Eldridge, A. L., & Peters, J. C. (1999). Dietary carotenoids and certain
1135.
cancers, heart disease, and age-related macular degeneration: A review of
van Ackerm, S. A., van den Berg, D. J., Tromp, M. N., Griffioen, D. H., van Bennekom,
recent research. Nutrition Reviews, 57, 201–214.
W. P., van der Vijgh, W. J., & Bast, A. (1995). Structural aspects of antioxidant
Das, N. P., & Pereira, T. A. (1990). Effect of flavonoids on thermal autoxidation of
activity of flavonoids. Free Radical Biology & Medicine, 20, 331–342.
Plam oil: Structure–activity relationship. Journal of the American Oil Chemists’
Yang, X. B., Zhao, Y., Yang, Y., & Ruan, Y. (2008). Isolation and characterization of
Society, 67, 255–258.
immunostimulatory polysaccharide from an herb tea, Gynostemma
Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., & Smith, F. (1956). Colorimetric
pentaphyllum Makino. Journal of Agricultural and Food Chemistry, 56, 6905–
method for determination of sugars and related substances. Analytical
6909.
Chemistry, 28, 350–356.
Zhishen, J., Mengcheng, T., & Jianming, W. (1999). The determination of flavonoids
Erlund, I. (2004). Review of the flavonoids quercetin, hespertin, and naringenin:
contents in mulberry and their scavenging effects on superoxide radicals. Food
Dietary sources, bioactivities, bioavailability, and epidemiology. Nutrition
Chemistry, 64, 555–559.
Research, 24, 851–874.