Food Chemistry 138 (2013) 2338–2345

Contents lists available at SciVerse ScienceDirect

Food Chemistry

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

High pressure homogenization increases antioxidant capacity and short-chainfatty acid yield of polysaccharide from seeds of Plantago asiatica L.

Jie-Lun Hu, Shao-Ping Nie ⇑, Ming-Yong Xie ⇑State Key Laboratory of Food Science and Technology, Nanchang University, Nanchang 330047, China

a r t i c l e i n f o

Article history:Received 10 August 2012Received in revised form 19 November 2012Accepted 5 December 2012Available online 26 December 2012

Keywords:Plantago asiatica L.PolysaccharideHigh pressure homogenizationAntioxidant capacityShort-chain fatty acid

0308-8146/$ - see front matter � 2013 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.foodchem.2012.12.016

⇑ Corresponding authors. Tel./fax: +86 791 839688304452 (S.-P. Nie).

E-mail addresses: spnie@ncu.edu.cn (S.-P. Nie), my

a b s t r a c t

Physiological properties of homogenized and non-homogenized polysaccharide from the seeds of Plan-tago asiatica L., including antioxidant capacity and short-chain fatty acid (SCFA) production, were com-pared in this study. High pressure homogenization decreased particle size of the polysaccharide, andchanged the surface topography from large flake-like structure to smaller porous chips. FT-IR showedthat high pressure homogenization did not alter the primary structure of the polysaccharide. However,high pressure homogenization increased antioxidant capacity of the polysaccharide, evaluated by 4 anti-oxidant capacity assays (hydroxyl radical-scavenging, superoxide radical-scavenging, 1,1-diphenyl-2-picryl-hydrazyl radical (DPPH)-scavenging and lipid peroxidation inhibition). Additionally, the produc-tion of total SCFA, propionic acid and n-butyric acid in ceca and colons of mice significantly increasedafter dieting supplementation with homogenized polysaccharide. These results showed that high pres-sure homogenization treatment could be a promising approach for the production of value-added poly-saccharides in the food industry.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Polysaccharides have been widely used in food processing andpreparation as stabilizers, thickeners, texturizers or emulsifiers(Huck-Iriart, Pizones Ruiz-Henestrosa, & Candal, 2012). An emul-sion is a mixture of two immiscible liquids in which one phase isdispersed through the other as small, discrete droplets. Polysaccha-rides as emulsifiers can assist in obtaining a fine dispersion and inmaintaining the homogeneous mixture through homogenization(Lapasin & Pricl, 1995). Many studies have shown that viscosity,rheological and gelling properties of polysaccharides could beinfluenced by the high pressure homogenization process (Pouliot,Britten, & Latreille, 1990; Silvestri & Gabrielson, 1991). In addition,it was also reported that high pressure homogenization could re-duce the particle size of polysaccharides and induce some degrada-tion, but it did not destroy the primary structure of polysaccharides(Chen et al., 2012; Floury, Desrumaux, Axelos, & Legrand, 2002).Concerning the effects of high pressure homogenization on physi-ological properties of polysaccharides, Lyons and Deidra Shannon(2011) reported that the antimicrobial properties of chitosan couldbe improved after being treated by high pressure homogenization.However, no study has been conducted to examine the effects ofhigh pressure homogenization on several other properties of

ll rights reserved.

9009 (M.-Y. Xie), +86 791

xie@ncu.edu.cn (M.-Y. Xie).

polysaccharides, such as their antioxidant capacity or their influ-ence on the production of total SCFA, propionic acid and n-butyricacid in ceca and colon once ingested. Antioxidant capacity of poly-saccharides is considered as one of the important physiologicalproperties, for it could help prolong the shelf life of food products(Gan & Latiff, 2011). Short-chain fatty acid production of polysac-charides is also increasingly attracting attention, because it is ben-eficial for human intestinal health (Huang, Chu, Dai, Yu, & Chau,2012). Researchers showed that antioxidant activity of the polysac-charides may be influenced by the reduction of the particle size(Pérez-Jiménez et al., 2008), and the particle size of the polysaccha-rides (determining the surface available to bacteria) may also influ-ence fermentation and thus affect the SCFA production (Guillon,Auffret, Robertson, Thibault, & Barry, 1998). Thus, our hypothesisis that the antioxidant activity and SCFA production of the polysac-charides may be influenced by high pressure homogenizationtreatment, resulting in the particle size reduction.

Polysaccharides could be extracted from many kinds of herbalor plant materials. Some Plantago plants, such as Plantago afra L.,Plantago psyllium L., Plantago ovata Forsk (isabgul), Plantago indicaL. and Plantago major L., are often used in traditional medicinethroughout the world because the soluble fibres in their seedsare able to improve some intestinal functions (Mahady, Fong, &Farnsworth, 1999; Samuelsen, 2000). Our research group has re-cently isolated a pure and homogeneous polysaccharide from theseeds of P. asiatica L. with a molecular weight of 1894 kDa (Yin,Nie, Zhou, Wan, & Xie, 2010). The polysaccharide was found to

J.-L. Hu et al. / Food Chemistry 138 (2013) 2338–2345 2339

be a highly branched heteroxylan which consisted of a b-1, 4-linked Xylp backbone with side chains attached to O-2 or O-3.The side chains consisted of b-T-linked Xylp, a-T-linked Araf, a-T-linked GlcAp, b-Xylp-(1 ? 3)-a-Araf and a-Araf-(1 ? 3)-b-Xylp(Yin, Lin, Li, Wang, Cui, Nie, & Xie, 2012). In addition, our recentstudies have also shown that this polysaccharide may induce mat-uration of murine dendrite cells, and may have antioxidant activityand promote defaecation (Huang, Xie, Yin, Nie, & Xie, 2009; Tang,Huang, Yin, Zhou, Xie, & Xie, 2007; Wu, Tian, Xie, & Li, 2007).

In the study, the effect of high pressure homogenization on thepolysaccharide from the seeds of P. asiatica L. was evaluated byusing dynamic light scattering, environmental scanning electronmicroscopy and FT-IR spectroscopy. In addition, the effects of highpressure homogenization on some properties of the polysaccha-ride, such as antioxidant activity and SCFA production, were forthe first time examined.

2. Materials and methods

2.1. Materials and reagents

The seeds of P. asiatica L. were purchased from Ji’an, JiangxiProvince, China and dried in the sun before use. Highly pure SCFAswere used to prepare the standard solutions for gas chromatogra-phy (GC) determination. Acetic acid (P99.5% purity) and n-valericacid (P99.5% purity) were obtained from Merck (Darmstadt,Germany). Propionic acid (P99.5% purity) was purchased fromJanssen Chimica (Belgium), while i-butyric acid (P99.5% purity),n-butyric acid (P99% purity), i-valeric acid (P99% purity) and4-methylvaleric acid (internal standard) (P99% purity) were pur-chased from Sigma (St. Louis, MO, USA). All other reagents usedwere of analytical grade and purchased from Shanghai Chemicalsand Reagents Co. (Shanghai, China).

2.2. Animals

Kunming mice, weighing 20.0 ± 2.0 g [Grade II, Certificate Num-ber SCXK (gan) 2006-0001], were purchased from Jiangxi College ofTraditional Chinese Medicine, Jiangxi Province, China. In this study,36 mice were used in total. Mice were randomly divided into threetreatment groups and in each treatment group there were 12 mice.All animals used in this study were cared for in accordance withthe Guidelines for the Care and Use of Laboratory Animals, pub-lished by the United States National Institute of Health (NIH, Pub-lication No. 85-23, 1996), and all procedures were approved by theAnimal Care Review Committee (Animal application approvalnumber 0064257), Nanchang University, China.

2.3. Polysaccharide preparation

Polysaccharide from P. asiatica L. seeds was prepared using ourpublished method (Yin et al., 2010). Briefly, the seeds of P. asiaticaL. (50 g) were defatted with 1 l of ethanol at room temperature for24 h under stirring, and then extracted with 500 ml of doubly dis-tilled water at 100 �C for 2 h. The residue was re-extracted. Thecombined aqueous extract, a gel of high viscosity, was centrifuged(4800�g, 10 min), prefiltered through a cotton cloth bag and con-centrated in a rotary evaporator under reduced pressure at 55 �C,to yield P. asiatica L. water extract. The filtrate was mixed with1.5 g/l of papain and heated in water at 60 �C for 2 h. The resultingaqueous solution was extensively dialyzed against doubly distilledwater for 72 h and precipitated by adding 4 volumes of anhydrousethanol at 4 �C for 12 h. After centrifugation the precipitate waswashed with anhydrous ethanol, dissolved in water and lyophi-lized to yield the polysaccharide (soluble fibre content >90.0%).

2.4. High pressure homogenization treatment

The polysaccharide solution (2.5 mg/ml) was dispersed indeionized water and stirred gently at room temperature to achievecomplete solubilization. High pressure homogenization was per-formed using a M-100EH-30 microfluidizer (Microfluidics Co.,Newton, USA) at 160 MPa for 5 passes. The experiment was con-ducted three times with different solutions of polysaccharide in or-der to verify the repeatability of the homogenization process. Partsof the solutions, after high pressure homogenization, were lyophi-lized for the Fourier transform infrared (FT-IR) and scanningmicroscopy analyses.

2.5. Average particle size (PS) and distribution determination

The average particle size and distribution of the polysaccharidein solutions were determined by Nicomp 380/ZLS Zeta potential/Particle sizer (PSS Nicomp, Santa Barbara, USA), based on dynamiclight scattering (DLS). The solutions were diluted to a concentra-tion of 0.5 mg/ml with deionized water, and all measurementswere carried out at 25 �C (Chen et al., 2012).

2.6. Scanning microscopy analysis

Polysaccharide samples were taken after freeze–drying, andsamples were prepared by sticking the polysaccharide onto dou-ble-sided adhesive tape attached to a circular specimen stub. Thesamples were viewed using an environmental scanning electronmicroscope (ESEM) (Quanta 200F, FEI Deutschland GmbH, Kassel,Germany) at 30 kV voltage and 3.0 spot size. Low vacuum modewas used for the ESEM.

2.7. FT-IR spectroscopy

The FT-IR spectra of polysaccharide samples were recorded on aNicolet 5700 spectrometer (Thermo Co., Madison, USA). The driedsamples were ground with KBr powder (spectroscopic grade) andpressed into pellets for spectra measurement in the frequencyrange of 4000–400 cm�1. The data was collected and plotted astransmittance (%) in function of the wave number (cm�1) and ana-lyzed with Ominic 7.2 software.

2.8. Antioxidant activity assays

2.8.1. Hydroxyl radical-scavenging activityThe hydroxyl radical-scavenging ability was measured, using a

modified method of Halliwell, Gutteridge, and Aruoma (1987). Areaction mixture was prepared by adding 0.1 ml of EDTA (1 mM),0.01 ml of FeCl3 (10 mM), 0.1 ml of H2O2 (10 mM), 0.36 mlof deoxyribose (10 mM), 1.0 ml of polysaccharide samples(0.1–10 mg/ml) dissolved in distilled water, 0.33 ml of phosphatebuffer (50 mM, pH 7.4) and 0.1 ml of ascorbic acid (1 mM) in se-quence. The mixture was then incubated at 37 �C for 1 h. 1.0 mlof the incubated mixture was mixed with 1.0 ml of 10% trichloro-acetic acid and 1.0 ml of 0.5% thiobarbituric acid (in 25 mM NaOHcontaining 0.025% butylated hydroxyl anisole) to develop the pinkchromogen measured at 532 nm. The hydroxyl radical-scavengingactivity of the polysaccharide samples was reported as the percent-age of inhibition of deoxyribose degradation and was calculatedaccording to the following equation:

% inhibition ¼ ðA0 � AtÞ=A0 � 100 ð1Þ

where A0 was the absorbance of the control (blank, without poly-saccharide samples) and At was the absorbance in the presence ofthe polysaccharide samples. All of the tests were carried out in

2340 J.-L. Hu et al. / Food Chemistry 138 (2013) 2338–2345

triplicate and IC50 values were expressed as means ± standard devi-ation (SD). Ascorbic acid was used as a positive control.

2.8.2. Superoxide radical-scavenging activityThis activity was measured using nitro blue tetrazolium (NBT)

reagent as described by Sabu and Kuttan (2002). The method isbased on generation of superoxide radical O�2 by auto-oxidation ofhydroxylamine hydrochloride in the presence of NBT, which getsreduced to nitrite. Nitrite, in the presence of EDTA, gives a colourthat was measured at 560 nm. Test solutions of polysaccharide(0.1–10 mg/ml) were taken in a test tube. To this, a reaction mixtureconsisting of 1 ml of (50 mM) sodium carbonate, 0.4 ml of (24 mM)NBT and 0.2 ml of 0.1 mM EDTA solution was added to the test tubeand an immediate reading was taken at 560 nm. 0.4 ml of (1 mM) ofhydroxylamine hydrochloride was added to initiate the reaction.The reaction mixture was then incubated at 25 �C for 15 min andthe reduction of NBT was measured at 560 nm. Decreased absor-bance of the reaction mixture indicates increased superoxide an-ion-scavenging activity. All the samples were treated in a similarmanner, absorbance was recorded and the percentage of inhibitionwas calculated according to the following equation:

% inhibition ¼ ðA0 � AtÞ=A0 � 100 ð2Þ

where A0 was the absorbance of the control (blank, without poly-saccharide samples) and At was the absorbance in the presence ofthe polysaccharide samples. All the tests were performed in tripli-cate and IC50 values were obtained. Ascorbic acid was used as a po-sitive control.

2.8.3. DPPH radical-scavenging activityDPPH radical-scavenging activity of polysaccharide samples was

measured according to our published method (Chen, Xie, Nie, Li, &Wang, 2008). The 0.2 mM solution of DPPH in 95% ethanol was pre-pared daily before UV measurements were taken. One millilitre ofthe polysaccharide samples of different quantities (0.1–10 mg) inwater was thoroughly mixed with 2 ml of freshly prepared DPPH�and 2 ml of 95% ethanol. The mixture was shaken vigorously and leftto stand for 30 min in the dark. The absorbance of the supernatantobtained after centrifugation was then measured at 517 nm.

The DPPH radical-scavenging ability was calculated using thefollowing equation:

I% ¼ ½1� ðAi � AjÞ=Ac� � 100% ð3Þ

where Ac is the absorbance of DPPH solution without polysaccha-ride sample (2 ml DPPH + 3 ml of 95% ethanol), Ai is the absorbanceof the test polysaccharide sample mixed with DPPH� solution (1 mlof polysaccharide sample + 2 ml of DPPH + 2 ml of 95% ethanol) andAj is the absorbance of the polysaccharide sample without DPPH�solution (1 ml of polysaccharide sample + 4 ml of 95% ethanol). Allthe tests were performed in triplicate and IC50 values were ob-tained. Ascorbic acid was used as a positive control.

2.8.4. Inhibition of lipid peroxidationThis test was conducted using the method of Zhang, Yu, Zhou,

and Xiao (1996) with some modifications. Briefly, an equal volumeof egg yolk was added to 0.1 M phosphate-buffered saline (PBS, pH7.45). The mixture was stirred magnetically for 10 min and then di-luted with 24 volumes of PBS. The yolk homogenate (1 ml), poly-saccharide samples (0.5 ml, 0.1–10 mg/ml), PBS (1 ml) and25 mM FeSO4 (1 ml) were mixed in a tube and shaken at 37 �Cfor 15 min. The reaction was stopped by the addition of trichloro-acetic acid and the mixture was centrifuged. Then 1 ml of 8 g/lthiobarbituric acid solution was added to 3 ml of the supernatant.This solution was heated at 10 �C for 10 min, after which its absor-bance at 532 nm was measured. The ability to inhibit lipid perox-idation was calculated as follows:

inhibitory effectð%Þ ¼ ½ðB0 � BÞ=B0� � 100 ð4Þ

where B0 is the absorbance of the control and B is the absorbance inthe presence of polysaccharide samples. All the tests were per-formed in triplicate and IC50 values were obtained. Ascorbic acidwas used as a positive control.

2.9. Changes in SCFA production of polysaccharide treated by highpressure homogenization

2.9.1. Animal experiment designSix-week-old male Kunming mice (20.0 ± 2.0 g) were individu-

ally housed in stainless steel cages in a room with controlled tem-perature (25 ± 0.5 �C), relative humidity (50 ± 5%) and 12 h/12 hlight/dark cycle. All mice were fed with the same amount of basaldiet (Supplementary Table 1) which was prepared according to thepublished formula (Adachi et al., 2004; Berggren, Björck, Nyman, &Eggum, 1993), and water was provided ad libitum. Mice were ran-domly divided into three groups: (1) non-treated polysaccharidegroup: mice were given oral administration of non-treated poly-saccharide at the dose of 0.1 g/kg body weight (BW); (2) homoge-nization-treated polysaccharide group: mice were given oraladministration of homogenization-treated polysaccharide at thedose of 0.1 g/kg BW; (3) control group: mice were given distilledwater of the same volume as the mean volume of the polysaccha-ride groups. Non-treated polysaccharide and homogenization-trea-ted polysaccharide were dissolved in distilled water beforeadministration. Each group had 12 mice, and each mouse washoused in one cage.

All the mice were given oral administration of polysaccharide orwater (control) by gavage at about 9:00 am everyday for 4 weeks.In addition, we determined the body weight of each mouse every-day before gavage and changed the volume of the polysaccharidein order to make sure the mice were given polysaccharide at thedose of 0.1 g/kg BW by gavage everyday. Throughout the experi-ment, the animals’ general health status and body mass were ob-served twice daily. At the end of the designated experimentalperiod, the mice were sacrificed, and the liver, kidney, spleen,heart, lung, intestine, cecum and colon were excised and immedi-ately weighed. After that, the ceca and colons were aseptically re-moved, immediately, and placed on an ice-cold plate,longitudinally opened and the cecal and colonic contents were col-lected for further use.

2.9.2. Measurement of cecal and colonic SCFA concentrationA portion (0.1 g) of the sample (cecal or colonic content) was

rapidly put into a round-bottomed stoppered tube in an ice-coldwater bath. One millilitre of deionized water was added to the tubeand mixed intermittently on a vortex-mixer for 2 min. The tubewas allowed to stand in the ice-cold water bath for 20 min andthen centrifuged at 4800g for 20 min at 4 �C. The supernatantwas transferred to another round-bottomed stoppered tube. Thisprocess was repeated once. The supernatant was analyzed by injec-tion into the chromatographic system.

Chromatographic analysis was carried out according to our pub-lished method (Hu, Nie, Min, & Xie, 2012), using an Agilent 6890NGC system equipped with a flame ionization detector (FID) and anN10149 automatic liquid sampler (Agilent, USA). A GC column(HP-INNOWAX, 190901N-213, J & W Scientific, Agilent Technolo-gies Inc., USA) of 30 m � 0.32 mm ID coated with 0.50 lm filmthickness was used. Nitrogen was supplied as the carrier gas at aflow rate of 19.0 ml/min with a split ratio of 1:10. The initial oventemperature was 100 �C and it was kept there for 0.5 min, and thenraised to 180 �C by 4 �C/min. The temperatures of the FID andinjection port were 240 �C. The flow rates of hydrogen and syn-thetic air were 30 and 300 ml/min, respectively. The injected sam-

J.-L. Hu et al. / Food Chemistry 138 (2013) 2338–2345 2341

ple volume for GC analysis was 0.2 ll, and the running time foreach analysis was 20.5 min. The independently replicated determi-nations were performed three times for standard solutions andeach sample. Data handling was carried out with a HP ChemStationPlus software (A.09.xx, Agilent). Mean values and standard devia-tions were calculated from triplicate determinations. Results wereexpressed as means ± standard deviation.

2.10. Statistical analysis

All the experiments were done in triplicate. Results were ex-pressed as means ± standard deviation (SD). Data were evaluatedby 1-way analysis of variance, using SPSS 10.0 software (Version16.0, Chicago, United States). The difference between differentgroups was evaluated by SNK test. The level of significance wasset at p < 0.05.

3. Results and discussion

3.1. Particle size and surface topography

The particle size and distribution of the polysaccharide from P.asiatica L. seeds, before and after being treated by high pressurehomogenization, are shown in Fig. 1. After high pressure homoge-

Fig. 1. Particle size of polysaccharide from P. asiatica L.

nization, the distribution profile of particle size shifted left andnarrowed (a typical Gaussian distribution), indicating that the par-ticle size of the polysaccharide decreased. The size of the polysac-charide decreased from 4758.7 ± 153.8 nm to 226.3 ± 4.7 nm afterhigh pressure homogenization.

In addition, as shown by ESEM, the non-treated polysaccharidewas relatively tidy, with flake-like lamella (Fig. 2A). However, afterhigh pressure homogenization, many pores appeared in the poly-saccharide sample (Fig. 2B). The original flake-like structure wastotally changed into smaller fragments after homogenization.Interestingly, it was found that Fig. 2A and B were similar to themicrostructure of pectin, with and without high pressure homoge-nization, respectively (Chen et al., 2012).

It was reported that the complexes could be broken into micro-fragments by homogenization and thus produce spherical particleswith a very low diameter (Chen et al., 1989). Homogenizationcould result in a change in the degree of aggregation of polymer,as well as irreversible disruption of the polymer (Lagoueyte & Pa-quin, 1998). These results were similar to ours.

3.2. FT-IR spectroscopy

The chemical structures of the non-treated and homogeniza-tion-treated polysaccharide samples were analyzed by FT-IR

seeds. A: non-treated; B: homogenization-treated.

Fig. 2. Environmental scanning electron micrograph of polysaccharide from P.asiatica L. seeds. A: non-treated; B: homogenization-treated.

2342 J.-L. Hu et al. / Food Chemistry 138 (2013) 2338–2345

spectroscopy (Fig. 3). The strong absorption at 1048 cm�1 was dueto stretching vibration of the pyranose ring. In the anomeric region(950–700 cm�1), the spectrum exhibited characteristic absorptionat 810 cm�1 due to the presence of mannose. There are two typesof end carbon–glucoside bonds (a and b) that can be distinguishedby IR. In IR spectra, the a-type C–H bond has an absorption peaknear 844 cm�1, while that of the b-type C–H bond is near891 cm�1 (Chen et al., 2008). A characteristic absorption at899 cm�1, indicating the b configuration of the sugar units, wasalso observed, but there was no absorption near 844 cm�1 corre-sponding to the a configuration. The absorption band at1628 cm�1 indicated the presence of uronic acid. The band at3423 cm�1 was due to hydroxyl stretching vibration of the poly-saccharide. The bands at 2928 cm�1 and 1403 cm�1 were due toC–H stretching vibration. For non-treated and homogenization-treated polysaccharide samples, the spectra were rather similarand there was no significant difference for characteristic absorp-tion bands, as described above between them. These resultsshowed that high pressure homogenization treatment had not al-tered the primary structure of the polysaccharide.

3.3. Antioxidant activity

Extended shelf-life is a key factor for making any food commod-ity more profitable and commercially available for long periods of

time at the best possible quality. The producer will benefit fromthe longer shelf-life to develop the market over greater distances.Natural antioxidant compounds could help prolong the shelf-lifeof food, and might also be beneficial for human health to some ex-tent. In this study, antioxidant activity of the polysaccharide sam-ples was evaluated using four different assays. The IC50 values ofdifferent polysaccharide samples are summarized in Table 1.

Hydroxyl radical is an extremely reactive free radical formed inbiological systems and has been implicated as a highly damagingspecies in free radical pathology, capable of damaging almost everymolecule found in the living cells (Hochstein & Atallah, 1988). Asshown in Table 1, homogenization-treated polysaccharide hadmuch more scavenging power for the hydroxyl radical than hadnon-treated polysaccharide (p < 0.05).

Superoxide anion plays an important role in the formation ofreactive oxygen species, such as hydrogen peroxide, hydroxyl rad-ical, and singlet oxygen, and induces oxidative damage in lipids,protein, and DNA (Pietta, 2000). In this study, the IC50 value forsuperoxide radical-scavenging ability of non-treated polysaccha-ride (2421 ± 2 lg/ml, Table 1) was higher than that for homogeni-zation-treated polysaccharide (2271 ± 3 lg/ml) (p < 0.05).

The model of scavenging the stable DPPH radical is a widelyused method for evaluating the free radical-scavenging ability ofnatural compounds. The effect of antioxidants on DPPH radical-scavenging was thought to be due to their hydrogen-donating abil-ity (Chen et al., 2008). The DPPH�-scavenging activity of the non-treated polysaccharide, expressed in terms of IC50, was503 ± 8 lg/ml (Table 1), with a lower DPPH�-scavenging powerthan that of homogenization-treated polysaccharide (384 ± 5 lg/ml) or ascorbic acid (205 ± 3 lg/ml) (p < 0.05).

Lipid peroxidation (oxidative degradation of polyunsaturatedfatty acid in the cell membranes) generates a number of degrada-tion products, such as malondialdehyde (MDA), which is found tocause cell membrane destruction and cell damage, leading to liverinjury, atherosclerosis, kidney damage, ageing, and susceptibilityto cancer (Rice-Evans & Burdon, 1993). The IC50 value (lg/ml), re-lated to the lipid peroxidation inhibitory capacity assay, showedthat the lipid peroxidation inhibitory capacity of non-treated poly-saccharide (1703 ± 8 lg/ml) was only about 60% of the homogeni-zation-treated polysaccharide (1076 ± 8 lg/ml, Table 1).

In these four antioxidant activity assays, except for the DPPH�-scavenging assay, the non-treated polysaccharide and homogeni-zation-treated polysaccharide both have good solubility and noaggregates formed. Possibly some small aggregates were formedafter addition of 95% ethanol in the DPPH�-scavenging assay. How-ever, these small aggregates could be removed by centrifugationbefore absorbance determination; thus they did not influence theaccuracy of the absorbance. The decrease of absorbance was obvi-ous, in the presence of polysaccharides, compared to the control. Inaddition, the procedures for DPPH�-scavenging assay have beensuccessfully used in our previous studies (Chen et al., 2008; Yinet al., 2010).

The antioxidant capacity results indicated that the antioxidantcapacity of the polysaccharide from P. asiatica L. was improvedafter the high pressure homogenization treatment. It is reportedthat one of the antioxidant mechanisms of polysaccharides maybe due to the hydrogen supplied from some substituent groupsin the polysaccharides (such as ferulic acids), which combine withradicals and form a stable radical to terminate the radical chainreaction (Chen et al., 2008). The other possibility is that the poly-saccharide itself combines with metal ions or free radicals thatare necessary for the radical chain reaction, so the reaction is ter-minated (Yin et al., 2010). In this study, the antioxidant capacityof original polysaccharide from P. asiatica L. may be attributed toboth of them. On the one hand, the polysaccharide may havestrong ability to chelate metal ions or free radicals, as inferred from

Fig. 3. Effect of high pressure homogenization on FT-IR spectra of polysaccharide from P. asiatica L. seeds.

Table 1Half-inhibition (IC50) values of antioxidant activities of non-treated polysaccharide and homogenization-treated polysaccharide measured using hydroxyl radical-scavenging,superoxide radical-scavenging, DPPH radical-scavenging and inhibition of lipid peroxidation assays.

Sample IC50 (lg/ml)a

Hydroxyl radical-scavenging Superoxide radical-scavenging DPPH radical-scavenging Lipid peroxidation inhibition

Non-treated polysaccharide 3084 ± 2ab 2421 ± 2a 503 ± 8a 1703 ± 8aHomogenization-treated polysaccharide 2394 ± 3b 2271 ± 3b 384 ± 5b 1076 ± 8bAscorbic acid c 220 ± 3c 304 ± 5c 205 ± 3c 440 ± 3c

a The IC50 value is expressed as lg/ml and represents the concentration of sample that is required for 50% inhibition of hydroxyl radical, superoxide radical, DPPH radical,and lipid peroxidation. A lower IC50 value indicates a higher antioxidant activity. Each value in the table was obtained by calculating the average of three determina-tions ± standard deviation.

b Mean values in the same column with different letters are significantly different (Tukey test, p < 0.05).c Ascorbic acid was used as a positive control.

Table 2Effect of non-treated and homogenization-treated polysaccharide from seeds ofP. asiatica L. on short-chain fatty acid production in mouse cecum.

Control group NTP groupa HTP groupb

SCFA (lg/g of cecal content)Acetic acid 2525 ± 46ac 3026 ± 51b 3043 ± 41bPropionic acid 1022 ± 24a 1108 ± 28b 1499 ± 17cButyric acid 1386 ± 21a 2101 ± 33b 2604 ± 25ci-Butyric acid 117 ± 11a 105 ± 19a 112 ± 12aValeric acid 73 ± 6a 74 ± 7a 72 ± 9ai-Valeric acid 125 ± 15a 120 ± 11a 123 ± 12aTotal SCFA 5248 ± 132a 6534 ± 144b 7453 ± 152c

a NTP group, non-treated polysaccharide group.b HTP group, homogenization-treated polysaccharide group.c Each value is the mean ± standard deviation (n = 12); means in the same line

not sharing a common letter are significantly different (p < 0.05).

Table 3Effect of non-treated and homogenization-treated polysaccharide from seeds ofP. asiatica L. on short-chain fatty acid production in mouse colon.

Control group NTP groupa HTP groupb

SCFA (lg/g of colonic content)Acetic acid 2332 ± 57ac 3074 ± 61b 3090 ± 49bPropionic acid 949 ± 39a 1064 ± 27b 1468 ± 40cButyric acid 1167 ± 21a 1509 ± 36b 2206 ± 30ci-Butyric acid 171 ± 14a 164 ± 15a 160 ± 20aValeric acid 62 ± 8a 58 ± 6a 60 ± 10ai-Valeric acid 146 ± 11a 144 ± 13a 146 ± 18aTotal SCFA 4827 ± 147a 6013 ± 142b 7130 ± 156c

a NTP group, non-treated polysaccharide group.b HTP group, homogenization-treated polysaccharide group.c Each value is the mean ± standard deviation (n = 12); means in the same line

not sharing a common letter are significantly different (p < 0.05).

J.-L. Hu et al. / Food Chemistry 138 (2013) 2338–2345 2343

our previous research (Yin, Lin, Li, Wang, Cui, Nie, & Xie, 2012). Onthe other hand, there are some ferulic acids present in the mole-cule of the polysaccharide (Yin et al., 2012), which have beenproved to be hydrogen-donors and have obvious antioxidant abil-ity (Graf, 1992). After high pressure homogenization, the particlesize of the polysaccharide was significantly reduced (Fig. 1), result-ing in an increase of total particle surface area for combining with

metal ions or free radicals. Hence, the antioxidant capacity of thepolysaccharide was improved after high pressure homogenization.In addition, it was reported that the viscosity of the polysaccharidemay influence its chelation ability towards metal ions or free rad-icals, and polysaccharides with lower viscosity exhibited higherantioxidant ability due to their greater chelation capacity (Kamil,Jeon, & Shahidi, 2002). In this study, it was found that the viscosity

2344 J.-L. Hu et al. / Food Chemistry 138 (2013) 2338–2345

of the polysaccharide decreased after high pressure homogeniza-tion treatment, which was consistent with another report (Chenet al., 2012). Thus, the higher antioxidant ability of the homogeni-zation-treated polysaccharide may also be attributed to its lowerviscosity as compared to the non-treated polysaccharide. Further-more, the process of homogenization may lead the polysaccharidemolecule being dispersed more homogeneously, which resulted inthe ferulic acid present in the molecule also being dispersed morehomogeneously. This could mean that the ferulic acid provideshydrogen more effectively and performs antioxidant activity bet-ter. For the above reasons, the antioxidant capacity of the polysac-charide could be improved after high pressure homogenizationtreatment.

3.4. SCFA production of polysaccharide

3.4.1. General health status and organ weight of miceThroughout the experimental period, no noticeable behavioral

or activity changes were observed in the mice, and no treatment-related illness or death occurred. The growth of mice appeared nor-mal throughout the experimental period, and no mouse experi-enced diarrhea or constipation. After 4 weeks, the average bodyweight gain was 18.6 g per mouse, and the average dietary intakewas 213.5 g per mouse. There was no observable difference in theanimals’ body mass and hair lustre between the polysaccharide-treated groups and control group.

Supplementary Table 2 summarizes the effects of the polysac-charide administration on the weights of various mouse organs.Organ weights were not different amongst the groups, except forthe cecal weight and colon weight. The cecum and colon of thepolysaccharide-treated groups were all significantly heavier thanthose of the control group (p < 0.05). This phenomenon has gener-ally been observed when mice were fed with soluble dietary fibreand other carbohydrates (Arjmandi, Craig, Nathani, & Reeves,1992; Moundras, Behr, Demigné, Mazur, & Rémésy, 1994). Therewas no significant difference in mouse cecum and colon weight be-tween the homogenization-treated polysaccharide group and thenon-treated polysaccharide group (p > 0.05).

3.4.2. Cecal and colonic SCFA concentration of miceIn mice, the cecum and colon are sites of vigorous microbial

activity where polysaccharides undergo fermentation, yieldingSCFA. SCFA was reported to provide more than 70% of the oxygenconsumed by human cecal and colonic tissue. Acetate is oxidizedby brain, heart and peripheral tissues; propionic acid affects the li-ver and cholesterol metabolism; butyric acid serves as an energysource for colonic epithelium, regulates epithelial and immune cellgrowth and apoptosis, and provides protection against colonic can-cer and colitis (Pryde, Duncan, Hold, Stewart, & Flint, 2002).

The SCFA concentrations in the mouse cecum and colon in dif-ferent groups are presented in Tables 2 and 3. The polysaccharide-treated groups had significantly higher concentrations of totalSCFA, acetic acid, propionic acid and n-butyric acid in mouse ce-cum and colon than had the control group (p < 0.05). It was re-ported that the production of propionic acid could be promotedby the fermentation of glucose, xylose and arabinose (Mortensen,Holtug, & Rasmussen, 1988). The polysaccharide from P. asiatica Lwas high in xylose and arabinose content with a relatively loweramount of glucose (Yin et al., 2010), so the increase of propionicacid might result from the fermentation of xylose, arabinose andglucose of the polysaccharide. It was also found that xylose tendedto have a greater impact on the production of butyric acid (Salva-dor et al., 1993). The polysaccharide from P. asiatica L was rich inxylose (Yin et al., 2010), so the increase of butyric acid might bedue to the fermentation of xylose in the polysaccharide. In addi-tion, there were no significant differences in the concentrations

of i-butyric acid, n-valeric acid, and i-valeric acid in mouse cecumand colon amongst non-treated polysaccharide group, homogeni-zation-treated polysaccharide group and control group, throughoutthe experimental period (p > 0.05).

It could also be seen from Tables 2 and 3 that the total SCFA,propionic acid and n-butyric acid concentrations in mouse cecumand colon of homogenization-treated polysaccharide group wereboth significantly higher than that of the non-treated polysaccha-ride group (p < 0.05). Researchers showed that the porosity of thepolysaccharide and the particle size of the polysaccharide (deter-mining the surface available to bacteria) may influence the fermen-tation and thus affect the SCFA production (Guillon et al., 1998). Todegrade polysaccharides, microbial glycosidases must have accessto their substrates. One factor, important for the control of accessi-bility, is particle size. Decreasing particle size increases the exter-nal surface area and so increases the area exposed to bacteria(Gama, Teixeira, & Mota, 1994). For example, the small-particlewheat bran produced greater SCFA concentrations during in vitrofermentation than did large-particle bran (Stewart & Slavin,2009); the concentrations of total SCFA, acetic acid, propionic acidand n-butyric acid in the cecal content generally increased in ratsas the particle size of corn bran decreased (Ebihara & Nakamoto,2001); vegetable fibre of smaller particle size results in longer tran-sit times in human intestine and is more extensively degraded(Lampe, Slavin, Melcher, & Potter, 1992). In addition, the pores inthe polysaccharides could also improve the surface availabilityfor bacteria and the surface availability for enzymes, and thusthe polysaccharides are more extensively degraded (Guillonet al., 1998). In the present study, the particle size of the polysac-charide from P. asiatica L. decreased significantly after high pres-sure homogenization (Fig. 1). Meanwhile, the surface topographyof the polysaccharide was changed from large flake-like structureto smaller porous chips, which increased the porosity of the poly-saccharide (Fig. 2). Therefore, the accessible surface area of thepolysaccharide for the cecal and colonic microbiota to contactwas increased, and the glucose, xylose and arabinose in the poly-saccharide could be more easily utilized by the microbiota afterhigh pressure homogenization.

4. Conclusions

The effect of high pressure homogenization on the physiologicalproperties of polysaccharide, such as antioxidant capacity andSCFA production, was for the first time investigated. It was foundthat the particle size of polysaccharide from the seeds of Plantagoasiatica L. decreased and the surface topography of the polysaccha-ride changed from large flake-like structure to smaller porous chipsafter high pressure homogenization. The antioxidant capacity andSCFA production of the polysaccharide were improved by thehomogenization treatment. Our results showed that high pressurehomogenization treatment could be a promising approach for theproduction of value-added polysaccharides in the food industry.

5. Conflict of interest

The authors declare no competing financial interest.

Acknowledgements

This study is financially supported by Key Program of NationalNatural Science Foundation of China (No. 31130041), National KeyTechnology R & D Program of China (2012BAD33B06), NationalNatural Science Foundation of China (20802032 and 21062012),Objective-Oriented Project of State Key Laboratory of Food Scienceand Technology (SKLF-MB-201001), Training Project of Young

J.-L. Hu et al. / Food Chemistry 138 (2013) 2338–2345 2345

Scientists of Jiangxi Province (Stars of Jing gang) and the Open Pro-ject Program of State Key Laboratory of Food Science and Technol-ogy of Nanchang University (No. SKLF-KF-201202), are gratefullyacknowledged.

Appendix A. Supplementary data

Supplementary data associated with this article can be found,in the online version, at http://dx.doi.org/10.1016/j.foodchem.2012.12.016.

References

Adachi, T., Ono, Y., Koh, K. B., Takashima, K., Tainaka, H., Matsuno, Y., et al. (2004).Long-term alteration of gene expression without morphological change in testisafter neonatal exposure to genistein in mice. Toxicogenomic analysis usingcDNA microarray. Food and Chemical Toxicology, 42(3), 445–452.

Arjmandi, B. H., Craig, J., Nathani, S., & Reeves, R. D. (1992). Soluble dietary fiber andcholesterol influence in vivo hepatic and intestinal cholesterol biosynthesis inrats. The Journal of Nutrition, 122(7), 1559–1565.

Berggren, A. M., Björck, I. M. E., Nyman, E., & Eggum, B. O. (1993). Short-chain fattyacid content and pH in caecum of rats given various sources of carbohydrates.Journal of the Science of Food and Agriculture, 63(4), 397–406.

Chen, W. S., Henry, G. A., Gaud, S. M., Miller, M. S., Kaiser, J. M., Balmadeca, E. A.,Morgan, R. G., Baer, C. C., Borwankar, R. P., & Hellgeth, L. C. (1989).Microfragmented ionic polysaccharide/protein complex dispersions. EP Patent0340035.

Chen, J., Liang, R. H., Liu, W., Liu, C. M., Li, T., Tu, Z. C., et al. (2012). Degradation ofhigh-methoxyl pectin by dynamic high pressure microfluidization and itsmechanism. Food Hydrocolloids, 28(1), 121–129.

Chen, Y., Xie, M. Y., Nie, S. P., Li, C., & Wang, Y. X. (2008). Purification, compositionanalysis and antioxidant activity of a polysaccharide from the fruiting bodies ofGanoderma atrum. Food Chemistry, 107(1), 231–241.

Ebihara, K., & Nakamoto, Y. (2001). Effect of the particle size of corn bran on theplasma cholesterol concentration, fecal output and cecal fermentation in rats.Nutrition Research, 21(12), 1509–1518.

Floury, J., Desrumaux, A., Axelos, M. A. V., & Legrand, J. (2002). Degradation ofmethylcellulose during ultra-high pressure homogenisation. Food Hydrocolloids,16(1), 47–53.

Gama, F., Teixeira, J., & Mota, M. (1994). Cellulose morphology and enzymaticreactivity: A modified solute exclusion technique. Biotechnology andBioengineering, 43(5), 381–387.

Gan, C. Y., & Latiff, A. A. (2011). Extraction of antioxidant pectic-polysaccharide frommangosteen (Garcinia mangostana) rind: Optimization using response surfacemethodology. Carbohydrate Polymers, 83(2), 600–607.

Graf, E. (1992). Antioxidant potential of ferulic acid. Free Radical Biology andMedicine, 13(4), 435–448.

Guillon, F., Auffret, A., Robertson, J., Thibault, J. F., & Barry, J. L. (1998). Relationshipsbetween physical characteristics of sugar-beet fibre and its fermentability byhuman faecal flora. Carbohydrate Polymers, 37(2), 185–197.

Halliwell, B., Gutteridge, J., & Aruoma, O. I. (1987). The deoxyribose method: Asimple ‘‘test-tube’’ assay for determination of rate constants for reactions ofhydroxyl radicals. Analytical Biochemistry, 165(1), 215–219.

Hochstein, P., & Atallah, A. S. (1988). The nature of oxidants and antioxidantsystems in the inhibition of mutation and cancer. Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis, 202(2), 363–375.

Hu, J.-L., Nie, S.-P., Min, F.-F., & Xie, M.-Y. (2012). Polysaccharide from seeds ofPlantago asiatica L. increases short-chain fatty acid production and fecalmoisture along with lowering pH in mouse colon. Journal of Agricultural andFood Chemistry. http://dx.doi.org/10.1021/jf302169u.

Huang, D. F., Xie, M. Y., Yin, J. Y., Nie, S. P., & Xie, M. Y. (2009). Immunomodulatoryactivity of the seeds of Plantago asiatica L. Journal of Ethnopharmacology, 124(3),493–498.

Huang, Y.-L., Chu, H.-F., Dai, F.-J., Yu, T.-Y., & Chau, C.-F. (2012). Intestinal healthbenefits of the water-soluble carbohydrate concentrate of wild grape (Vitisthunbergii) in hamsters. Journal of Agricultural and Food Chemistry, 60(19),4854–4858.

Huck-Iriart, C., Pizones Ruiz-Henestrosa, V. M., Candal, R. J., & Herrera, M. L. (2012).Effect of aqueous phase composition on stability of sodium caseinate/sunfloweroil emulsions. Food and Bioprocess Technology. http://dx.doi.org/10.1007/s11947-012-0901-y.

Kamil, J. Y. V. A., Jeon, Y. J., & Shahidi, F. (2002). Antioxidative activity of chitosans ofdifferent viscosity in cooked comminuted flesh of herring (Clupea harengus).Food Chemistry, 79(1), 69–77.

Lagoueyte, N., & Paquin, P. (1998). Effects of microfluidization on the functionalproperties of xanthan gum. Food Hydrocolloids, 12(3), 365–371.

Lampe, J. W., Slavin, J. L., Melcher, E. A., & Potter, J. D. (1992). Effects of cereal andvegetable fiber feeding on potential risk factors for colon cancer. CancerEpidemiology Biomarkers & Prevention, 1(3), 207–211.

Lapasin, R., & Pricl, S. (1995). Rheology of industrial polysaccharides: theory andapplications (1st ed.). London: Blackie Academic & Professional.

Lyons, Deidra Shannon. (2011). Depolymerization of chitosan by high-pressurehomogenization and the effect on antimicrobial properties. Master’s Thesis.University of Tennessee. <http://trace.tennessee.edu/utkgradthes/1001>.

Mahady, G. B., Fong, H. H. S., & Farnsworth, N. R. (1999). WHO monographs onselected medicinal plants (1st ed.). Geneva: World Health Organisation.

Mortensen, P. B., Holtug, K., & Rasmussen, H. S. (1988). Short-chain fatty acidproduction from mono-and disaccharides in a fecal incubation system:Implications for colonic fermentation of dietary fiber in humans. The Journalof Nutrition, 118(3), 321.

Moundras, C., Behr, S. R., Demigné, C., Mazur, A., & Rémésy, C. (1994). Fermentablepolysaccharides that enhance fecal bile acid excretion lower plasma cholesteroland apolipoprotein E-rich HDL in rats. The Journal of Nutrition, 124(11), 2179.

Pérez-Jiménez, J., Arranz, S., Tabernero, M., Díaz-Rubio, M. E., Serrano, J., Goñi, I.,et al. (2008). Updated methodology to determine antioxidant capacity in plantfoods, oils and beverages: Extraction, measurement and expression of results.Food Research International, 41(3), 274–285.

Pietta, P. G. (2000). Flavonoids as antioxidants. Journal of Natural Products, 63(7),1035–1042.

Pouliot, Y., Britten, M., & Latreille, B. (1990). Effect of high-pressure homogenizationon a sterilized infant formula: Microstructure and age gelation. Food Structure,9(1), 1–8.

Pryde, S. E., Duncan, S. H., Hold, G. L., Stewart, C. S., & Flint, H. J. (2002). Themicrobiology of butyrate formation in the human colon. FEMS MicrobiologyLetters, 217(2), 133–139.

Rice-Evans, C., & Burdon, R. (1993). Free radicals–lipid interaction and theirpathological consequences. Progress in Lipid Research, 32, 71–110.

Sabu, M., & Kuttan, R. (2002). Anti-diabetic activity of medicinal plants and itsrelationship with their antioxidant property. Journal of Ethnopharmacology,81(2), 155–160.

Salvador, V., Cherbut, C., Barry, J. L., Bertrand, D., Bonnet, C., & Delort-Laval, J. (1993).Sugar composition of dietary fibre and short-chain fatty acid production duringin vitro fermentation by human bacteria. British Journal of Nutrition, 70(1),189–197.

Samuelsen, A. B. (2000). The traditional uses, chemical constituents and biologicalactivities of Plantago major L. A review. Journal of Ethnopharmacology, 71(1–2),1–21.

Silvestri, S., & Gabrielson, G. (1991). Degradation of tragacanth by high shear andturbulent forces during microfluidization. International Journal of Pharmaceutics,73(2), 163–169.

Stewart, M. L., & Slavin, J. L. (2009). Particle size and fraction of wheat bran influenceshort-chain fatty acid production in vitro. British Journal of Nutrition, 102(10),1404–1407.

Tang, Y., Huang, D., Yin, J., Zhou, C., Xie, X., & Xie, M. (2007). Effects of Semenplantaginis polysaccharides on phenotypic and endocytosis of murine dendriticcells. Food Science, 28(10), 517–520.

Wu, G., Tian, Y., Xie, M., & Li, C. (2007). Effect of Semen plantaginis polysaccharideson defecating function in constipated mice. Food Science, 28(10), 514–516.

Yin, J.-Y., Nie, S.-P., Zhou, C., Wan, Y., & Xie, M.-Y. (2010). Chemical characteristicsand antioxidant activities of polysaccharide purified from the seeds of Plantagoasiatica L. Journal of the Science of Food and Agriculture, 90(2), 210–217.

Yin, J., Lin, H., Li, J., Wang, Y., Cui, S. W., Nie, S., et al. (2012). Structuralcharacterization of a highly branched polysaccharide from the seeds ofPlantago asiatica L. Carbohydrate Polymers, 87(4), 2416–2424.

Zhang, E. X., Yu, L. J., Zhou, Y. L., & Xiao, X. (1996). Studies on the peroxidation ofpolyunsaturated fatty acid from lipoprotein induced by iron and the evaluationof the anti-oxidative activity of some natural products. Sheng Wu Hua Xue YuSheng Wu Wu Li Xue Bao, 28(2), 218–222.