Intake of Passiflora Edulis Leaf Extract Improves Antioxidant and Anti-Inflammatory Status in Rats with 2,4,6-Trinitrobenzenesul

Journal of Functional Foods 17 (2015) 575–586

Available online at www.sciencedirect.com ScienceDirect

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

Intake of Passiﬂora edulis leaf extract improves antioxidant and anti-inﬂammatory status in rats with 2,4,6-trinitrobenzenesulphonic acid induced colitis

Cinthia Baú Betim Cazarin a, Juliana Kelly da Silva a, Talita Cristina Colomeu b, Ângela Giovana Batista a, Laura Maria Molina Meletti c, Jonas Augusto Rizzato Paschoal a, Stanislau Bogusz Junior d, Patrícia Aparecida de Campos Braga a, Felix Guillermo Reyes Reyes a, Fábio Augusto e, Luciana Rodrigues de Meirelles b, Ricardo de Lima Zollner b,*, Mário Roberto Maróstica Júnior a,** a School of Food Engineering, University of Campinas, Rua Monteiro Lobato, 80, Cidade Universitária Zeferino Vaz, Campinas, São Paulo, Brazil b Faculty of Medical Science, University of Campinas, Rua Tessália Vieira de Camargo, 126, Cidade Universitária Zeferino Vaz, Campinas, São Paulo, Brazil c Center of Research and Development on Plant Genetic Resources, Agronomic Institute of Campinas, Campinas, SP, Brazil d Institute of Chemistry of São Carlos (IQSC), University of São Paulo (USP), São Carlos, São Paulo, Brazil e Institute of Chemistry, Instituto Nacional de Ciência e Tecnologia em Bioanalítica (INCTBio), University of Campinas, Campinas, São Paulo, Brazil

ARTICLE INFO ABSTRACT

Article history: The antioxidant and anti-inflammatory effects of the aqueous extract of Passiflora edulis leaves, Received 18 December 2014 which is a source of vitexin, isovitexin and isoorientin, was investigated in a 2,4,6- Received in revised form 27 May trinitrobenzenesulphonic acid induced colitis model. The oral intake of P. edulis extract 2015 (1100 gmL−1) improved significantly (P < 0.05) the endogenous antioxidant status and de- Accepted 28 May 2015 creased lipid peroxidation in serum, liver, and colon. The consumption of P. edulis extract Available online 24 June 2015 decreased the level of pro-inflammatory in colon tissue, especially by reducing IL-1β five- fold and TNF-α two-fold compared to the control colitis group. The extract of P. edulis leaves Keywords: possessed therapeutic activity in healthy rats and in TNBS-induced colitis, especially by con- Passiflora edulis leaves trolling the oxidative stress and inflammation observed in this disease. Colitis © 2015 Elsevier Ltd. All rights reserved. Oxidative stress Antioxidant potential Anti-inflammatory activity

* Corresponding author. Faculty of Medical Science, University of Campinas, Rua Tessália Vieira de Camargo, 126, Cidade Universitária Zeferino Vaz, Campinas, São Paulo, Brazil. Tel.: +55 19 32893709; fax: +55 19 32893709. E-mail address: [email protected] (R.L. Zollner). ** Corresponding author. School of Food Engineering, University of Campinas, Rua Monteiro Lobato, 80, Cidade Universitária Zeferino Vaz, Campinas, São Paulo, Brazil. Tel.: +55 19 3521 4059; fax: +55 19 3521 4059. E-mail address: [email protected] (M.R. Maróstica Júnior). http://dx.doi.org/10.1016/j.jff.2015.05.034 1756-4646/© 2015 Elsevier Ltd. All rights reserved. 576 Journal of Functional Foods 17 (2015) 575–586

thus modulating the production of short-chain fatty acids 1. Introduction (SCFA). The improvement in the mucosal protection against inflammatory damage has been observed in many studies in Passiflora is a native tropical and subtropical species culti- which the formation of SCFA increased after the consump- vated in several parts of the world, especially in the American tion of microbiota substrates (Duda-Chodak, 2012; Parkar et al., continent, and encompasses approximately 450 species, usually 2013). Therefore, phenolic compounds may play a role in im- called passion fruit (Dhawan, Dhawan, & Sharma, 2004). Their proving intestinal health (Duda-Chodak, 2012; Parkar et al., cultivation is primarily focused on the production of pulp juice 2013). and frozen pulp from the fruit. However, in folk medicine, leaves Inflammatory bowel disease (IBD) is a chronic immune- of Passiflora are used due to their anxiolytic (Li et al., 2011) and mediated intestinal disorder that encompasses Crohn’s disease anti-inflammatory activities (Montanher, Zucolotto, Schenkel, (CD) and ulcerative colitis (UC). CD is a transmural inflamma- & Frode, 2007), which may be mainly associated with the pres- tion that can affect the entire gastrointestinal tract, while UC ence of phenolic compounds (lucenin-2, vicenin-2, isoorientin, is a non-transmural inflammation restricted to the colon, and isovitexin, luteolin). In a previous work, we showed that Passiflora there are many factors related to colitis pathogenesis, includ- alata extract contained seven-fold more of the total amounts ing genetic, environmental, microbiological and immunological of flavonoids vitexin, isovitexin, and isoorientin than Passiflora factors (Podolsky, 2002). edulis extract (Colomeu et al., 2014; da Silva et al., 2013). Some Increases in reactive oxygen species (ROS) in colorectal evidence about the anti-inflammatory activity of P. alata was biopsy specimens of subjects and animal models of IBD are observed by Colomeu et al. (2014) in a type 1 diabetes experi- also observed (Keshavarzian et al., 1992; Nieto et al., 2000). The mental model. Intake of aqueous extract of P. alata promoted cause of the oxidative stress is an imbalance between protec- a decrease in cell infiltration and insulitis in NOD mice com- tor systems and production of free radicals. Our previous data pared to the control group. However, the amounts of isovitexin indicate that the extract of P. edulis leaves possessed antioxi- in P. edulis extract (0.50 ± 0.04 mg g−1) were almost three-fold dant activity in vitro and in vitro against oxidative damage, and more than that in P. alata extract (0.19 ± 0.01 mg g−1)(Colomeu is considered as an important source of natural antioxidants et al., 2014; da Silva et al., 2013). Protection against oxidative (da Silva et al., 2013). In addition, we demonstrated that it could stress was observed in cells treated with isovitexin extract from reduce oxidative stress in vivo by improving the antioxidant rice Oryza sativa, mainly through decreased lipid peroxidation power, reducing TBARS (thiobarbituric acid reactive sub- and inhibition of xanthine oxidation activity (Lin, Chen, Lee, stances) level, and by increasing the number of useful colonic & Lin, 2002). Amongst the flavonoids identified in the aqueous microbiota in healthy rats (da Silva et al., 2013). In the face of extract of P. edulis, isoorientin showed a higher amount this, the aim of the present study was to evaluate the effect (1.05 ± 0.03 mg g−1) compared with others flavonoids (da Silva of intake of aqueous P. edulis extract on inflammation and oxi- et al., 2013). dative stress using an experimental model of rat colitis induced The administration of hydroalcoholic extract of P. edulis in by 2,4,6-trinitrobenzenesulphonic acid (TNBS). rat colon anastomosis showed an anti-inflammatory effect through increased infiltration of mononuclear cells, fibro- blast proliferation, and granulation tissue, promoting the colon 2. Materials and methods healing (Bezerra et al., 2006). In another study, aqueous extracts of leaves of P. alata (100 mg kg−1, i.p.) and P. edulis 2.1. P. edulis leaves (250 mg kg−1, i.p.) showed a significant anti-inflammatory ac- Leaves of P. edulis were collected in June 2011 from the Agro- tivity on carrageenan-induced pleurisy in mice, especially by nomic Institute of Campinas (Campinas, SP, Brazil). A voucher inhibition of influx of leukocytes and a reduction in the serum specimen was deposited in the Herbarium at the same insti- C-reactive protein levels (Vargas et al., 2007). tute. The leaves were dried in an oven with circulating air In addition, different aqueous extract sources of polyphe- (Marconi, Piracicaba, SP, Brazil) at 50 °C for 48 h, ground to a nols have shown anti-carcinogenic activity (de Mejía, Song, fine homogeneous powder, and stored in amber glass bottles Heck, & Ramírez-Mares, 2010; Li et al., 2013; Zhao et al., 2014). at 8 °C. The anti-carcinogenic activity from P. edulis was evaluated in the fruit and the peel showed 89% inhibition of HT-29 (human 2.2. Aqueous P. edulis leaf extract – Tea colon adenocarcinoma cells) cancer cell line activity, and the pulp and seed showed the highest inhibitory activity (97%) (Li The extract was prepared by infusing powdered leaves in boiling et al., 2013). water for 25 min (leaves/water, 1:25, w/v). The extract was fil- These aforementioned effects may be correlated with the tered, and the volume adjusted to 25 mL. The antioxidant polyphenols found in the extracts, due to their antioxidant prop- potential and the composition of phenolic compounds were erties and ability to neutralize or quench oxidants (Bezerra et al., analysed in the extract. The concentration (1100 gmL−1)ofthe 2006; Vargas et al., 2007). Thus, some hypotheses about the poly- aqueous extract (named Tea) given to the animals was deter- phenol metabolism have been studied, especially because, mined in accordance to a previous study (da Silva et al., 2013). before their absorption, most dietary polyphenols are trans- formed in the colon by microbiota, which may influence their 2.3. LC–MS/MS analysis absorption and biological activities (Parkar, Trower, & Stevenson, 2013). Unabsorbed phenolic compounds that could remain in The LC–MS/MS equipment used in this study for the identifi- the intestinal lumen could be metabolized by the microbiota, cation of the flavonoids in P. edulis aqueous extract consists Journal of Functional Foods 17 (2015) 575–586 577

of an AcquityTM UPLC system (Waters, Milford, MA, USA) coupled for 2 weeks. Colitis was induced by intracolonic administra- to a mass spectrometer comprising hybrid quadrupole (Q) and tion of 2,4,6-trinitrobenzenesulphonic acid (TNBS) after 1 week time-of-flight (Tof) mass analysers XevoTM G2 Q-Tof (Waters), of the experiment. Briefly, the animals were anaesthetized with with an electrospray source interface (ESI) as ionization source halothane and 10 mg of TNBS dissolved in 0.25 mL of 50% in negative mode, and the instrument control and data pro- ethanol (v/v) was administered by Teflon cannula inserted 8 cm cessing were performed by MassLynx software (Waters) version into the anus (Morris et al., 1989). The same procedure was per- 4.1. formed with the Saline groups by intracolonic administration An MSE method (where E represents collision energy) was of 0.25 mL of 0.9% saline solution instead of TNBS. established applying low-energy collision (collision energy = 0) Access to food and drink was ad libitum and was moni- and high-energy collision (ranging from 10 to 20 V), in order tored three times a week. Tea was prepared and changed every to monitor the precursor ion as well for fragmentation infor- 48 hours to avoid oxidation. After 2 weeks, the animals were mation (Fig. 1). The data were acquired in centroid resolution anaesthetized with ketamine and xylazine and killed by ex- mode, using a mass range from 100 to 600 Da. The Elemental sanguination by cardiac puncture. Composition tool from the MassLynx software was applied to predict the molecular formula for the analytes. After that, the 2.5. Blood sampling and analysis founded presumed molecular formula was searched in the Chemspider database in order to determine the potential struc- Blood samples were obtained by cardiac puncture under an- tures, and by the MassFragment tool of the MassLynx software, aesthesia, collected in tubes sprayed internally with clot the fragmentation pattern was achieved. The chromato- accelerator (SiO2), and centrifuged at 2600 g for 20 min. Serum graphic separation was carried out on a Eclipse Plus C18 column was stored at −80 °C until analyses. Serum albumin and total RRHD (1.8 m, 2.1 mm × 100 mm column) (Agilent, Santa Clara, proteins were measured using commercial kits (Laborlab, CA, USA) using as mobile phase 0.1% formic acid (A) and ace- Guarulhos, SP, Brazil) and cytokines (IL-1β, IL-6, and TNF-α)were tonitrile + 0.1% formic acid (B). The gradient program was used measured using commercial Elisa kits (Peprotech, Ribeirão Preto, as follows: 0 min 95% A and 5% B; 10.00 min 1% A and 99% B; SP, Brazil). 13.01–15.00 min 95% A and 5% B. The elution conditions were The antioxidant activity was determined in serum. The optimized with a flow rate of 0.5 mL min−1, the injection volume samples were treated with ethanol/ultrapure water and was 2 L and the oven temperature was 35 °C. The sample was 0.75 mol L−1 metaphosphoric acid (Leite et al., 2011). These ex- diluted in ACN + 0.1% formic acid (1:1, v/v) and filtered through tracts were used in hydrophilic oxygen radical absorbance 0.22 m polytetrafluoroethylene membrane before being in- capacity (ORAC) (Prior et al., 2003) and ferric reducing antioxi- jected into the chromatograph. dant power (FRAP) (Benzie & Strain, 1996) assays. For the MS operating conditions the following the param- The ORAC assay was performed in the dark. Samples (20 L, eters were set: capillary voltage 2.5 kV, sampling cone voltage prepared as described above), 120 L of fluorescein in phos- 40 V, extraction cone voltage 4 V, source temperature 150 °C, phate buffer (pH 7.4), and 60 Lof2,2′-azobis(2- desolvation temperature 600 °C, cone gas flow 50 L h−1 and methylpropionamidine) dihydrochloride (AAPH) were added in desolvation gas flow 1000 L h−1. black microplates. The microplate reader was adjusted as pre- A solution of leucine–enkephalin at a concentration of viously described (fluorescent filters, excitation wavelength, 2ngg−1 was used as reference solution (Lockspray)innegative 485 nm; emission wavelength, 520 nm) (Prior et al., 2003). ORAC mode in order to obtain the accurate mass of the target mol- values are expressed in mol trolox equivalent (TE) per mL ecules. The [M-H]− ion at 554.2615 was used as the lock mass serum by using a Trolox standard curve (2.5–50.0 M TE). and the infusion flow rate of the solution was 5 L/min. For The ferric reducing ability of the serum was determined by calibration of the equipment a 0.5 mM sodium formate solu- the FRAP method (Benzie & Strain, 1996) with adaptations. In tion was infused at a flow rate of 5 L min−1. An exploratory the dark, FRAP reagent was made with 300 mmol L−1 acetate method was developed in order to confirm previous data and buffer (pH 3.6), 10 mmol TPTZ in 40 mmol L−1 HCl solution, and −1 assess the P. edulis aqueous extract composition. 20 mmol L FeCl3. Sample or standard solutions, ultrapure water, and FRAP reagent were mixed and incubated in a waterbath 2.4. In vivo experimental design for 30 min at 37 °C. After cooling to room temperature, samples and standards were read at 595 nm. The trolox standard curve This study was carried out in accordance with the ‘Guide for was made (10–800 mol TE). The results are expressed in mol the Care and Use of Laboratory Animals’, as promoted by the trolox equivalent (TE) mL−1. Brazilian College of Animal Experimentation (COBEA), and was approved by the Animal Research and Ethic Committee of the 2.6. Tissue sampling and analysis University of Campinas (Brazil) (protocol 2887-1/2012). Male Wistar rats (77-days-old, weight ~288.4 ± 21.09 g) were Liver and colon tissue were quickly removed, cleaned with saline housed under standard conditions of temperature (22 °C ± 2 °C), solution (0.9% NaCl), and weighed. The colon’s length was mea- humidity (60–70%), and a light–dark cycle (12/12 h). The rats sured and the macroscopically visible damage was evaluated were randomized into four groups of six animals each, named on a 0–10 scale (Bell, Gall, & Wallace, 1995). For histological Control Saline, Control Colitis, Tea Saline, and Tea Colitis. The analyses, freshly dissected colon samples were cryoconserved animals were fed with semi-purified experimental diet AIN- in a freezing medium (Jung; Leica Instruments, Nussloch, 93M (Reeves, Nielsen, & Fahey, 1993). Water was given to the Germany). Serial cross-sections (7 m) were prepared using a control group and P. edulis aqueous extract to the Tea groups cryostat (CM 1850 Cryostat, Leica) and fixed on silanized slides. 578 Journal of Functional Foods 17 (2015) 575–586

Fig. 1 – MSE spectra of Orientin/Isoorientin (A1 and A2) and Vitexin/Isovitexin (B1 and B2) identiﬁed in the P. edulis leaf extracts. A1 and A2: low and high energy spectra generated by combining the data acquired at retention time 2.27 min; B1 and B2: low and high energy spectra generated by combining the data acquired at retention time 2.48 min.

After staining the samples with haematoxylin–eosin, the se- samples were frozen in liquid nitrogen, and kept at −80 °C. Colon verity of each colitis case was analysed (Eclipsi 80i microscope, tissue homogenates were prepared in 50 mmol phosphate Nikon, Melville, NY, USA) by a pathologist, according to the cri- buffer (pH 7.4) using a Polytron homogenizer (MA102/Mini; teria described by Chinen et al. (2011). Afterwards, the tissue Marconi, Piracicaba, SP, Brazil) and the supernatant was kept Journal of Functional Foods 17 (2015) 575–586 579

at −80 °C until analyses of the cytokines (IL-1β, IL-6, and TNF- enized for faecal pH determination using a pH meter (model α) by commercial ELISA kits (Peprotech). TEC-5; Tecnal, Piracicaba, SP, Brazil) (da Silva et al., 2013).In Lipid peroxidation (Ohkawa, Ohishi, & Yagi, 1979) was done addition, lactobacilli, bifidobacteria, enterobacteria, and total aerobic by the TBA (thiobarbituric acid) assay and the resultant TBA bacteria counts were determined in the faeces. Caecal con- reactive substances were determined. Liver and colon tissue tents were homogenized in peptone water (100 mg mL−1) and samples were macerated in liquid nitrogen and 10 mg mL−1 was then ten-fold serial dilutions were made in the same medium. sonicated in acetate buffer (pH 3.5) on ice. The samples were Aliquots of 0.1 mL of the appropriate dilution were spread onto mixed with 8.1% sodium dodecyl sulphate (SDS) plus working the MRS agar media for Lactobacillus counts, and onto supple- reagent (TBA, 20% acetic acid, and 5% sodium hydroxide). After mented MRS agar (0.5 mg L−1 dicloxacillin, 1 g L−1 LiCl, and heating at 95 °C for 60 min, the samples were maintained in 0.5gL−1 L-cysteine hydrochloride) for Bifidobacterium. Culture an ice-bath for 10 min, and centrifuged at 10,000 g for 10 min. plates were incubated under anaerobic conditions at 37 °C for The supernatant was read at 532 nm, using a clear 96-well 24–48 h. Similarly, 1 mL of the diluted sample was spread onto microplate. The results are expressed as nmol MDA equiva- specific Petrifilm count plates (3M®, São Paulo, Brazil) for En- lents mg−1 tissue (or nmol MDA equivalents mL−1 serum). terobacteriaceae and total aerobic bacteria. Plates were incubated Reduced glutathione (GSH) levels were determined in the at 37 °C for 24–48 h. After the incubation, the specific colo- phosphate buffer homogenates using Ellman’s reagent (5,5′- nies grown on the selective culture media were counted, and dithiobis-(2-nitrobenzoic acid)) (Ellman, 1959). GSH solution (2.5– the number of viable microorganisms per gram of faeces 500 nmol GSH mL−1) was used as the standard, and absorbance (CFU g−1) was calculated. The mean and standard error were was read at 412 nm. Reduced thiol contents are expressed in calculated from the log-10 values of the CFU g−1. nmol GSH mg−1 protein. The protein concentration of tissue The content of SCFA was analysed by gas chromatogra- homogenates was determined by the Bradford method phy (G. Zhao, Nyman, & Jonsson, 2006) using Agilent 6890N (Bradford, 1976). equipment coupled with a flame ionization detector (FID) and Glutathione peroxidase (GPx) activity was determined based autosampler N10149 (Agilent, Santa Clara, CA, USA). A capil- on the oxidation of 10 mmol GSH by GPx coupled to the oxi- lary column NukolTM (Supelco, Bellefonte, PA, USA) with dation of 4 mmol NADPH by 1 U enzymatic activity of 30 m × 0.25 mm i.d. × 0.25 m was used. Chromatographic con- glutathione reductase (GR) in the presence of 0.25 mmol H2O2. ditions were injector and detector at 250 °C, split mode (1:10), The rate of NADPH oxidation was monitored by the decrease injection volume 1 L; helium carrier gas, with flow rate of in absorbance at 365 nm (Flohe & Gunzler, 1984). The results 1mLmin−1; initial temperature 100 °C (0.5 min), increasing 8 °C are expressed as nmol NADPH consumed min−1 mg−1 protein. min−1 until 180 °C (1 min), and then increasing at a rate of 20 °C Glutathione reductase activity was measured by following min−1 until 200 °C for 5 min. the decrease in absorbance at 340 nm induced by the reduction of 1 mmol oxidized glutathione in the presence of 0.1 mmol 2.8. Statistical analyses NADPH in phosphate buffer (Carlberg & Mannervik, 1985). The −1 −1 results are expressed as nmol NADPH consumed min mg Parametric data are expressed as means ± standard error of the protein. mean (SEM). The statistical analyses were carried out using Superoxide dismutase (SOD) activity was determined in 1 L GraphPad Prism 5.0 (GraphPad Software Inc., La Jolla, CA, USA) of appropriately diluted phosphate buffer. The homogenates software. Analyses were based on the Student’s t-test and were added to a 96-well microplate and 150 L of the working ANOVA, and the limit of significance was set at P < 0.05. solution (0.1 mmol hypoxanthine, 0.07 U xanthine oxidase, and 0.6 mmol tetranitrotetrazolium blue (NTB) in phosphate buffer in 1:1:1 proportions) was added, and the kinetic reaction was monitored at 560 nm for 10 min (Winterbourn, Hawkins, Brian, 3. Results and discussion & Carrell, 1975). The area under the curve (AUC) was calculated, and the SOD activity is expressed as U mg−1 protein. The characterization of Vitexin/Isovitexin and Orientin/ Myeloperoxidase (MPO) activity was determined in colon Isoorientin was performed by UPLC-ESI-Q-TOF/MS, applying the tissue homogenates to evaluate neutrophil accumulation MSE technique which has been proved to be a powerful and (Krawisz, Sharon, & Stenson, 1984). The samples were centri- reliable analytical approach for molecule identification (Bateman fuged at 3000 g for 15 min at 4 °C, and aliquots of the et al., 2007). Through the MSE technique, both precursor and supernatant were then mixed with a reaction buffer of fragment mass spectra can be simultaneously acquired by al- −1 50 mmol L phosphate buffer (pH 6.8), containing 0.005% H2O2 ternating between high and low collision energy during a single and 1.25 mg mL−1 o-dianisidine dihydrochloride, and the re- chromatographic run. According to the obtained results, the action was measured at 460 nm. The results were expressed low collision energy provided the mass spectra for deprotonated as MPO units mg−1 of protein, where 1 unit of MPO activity was molecules in their intact forms, whose molecular formulas sug- defined as that degrading 1 mol hydrogen peroxide min−1 at gested by the elemental composition tool are related to vitexin/ 25 °C. isovitexin and orientin/isoorientin. The high energy spectra obtained in MSE experiments provided the fragmentation in- 2.7. Caecal content analysis formation for the molecules. On the basis of the accurate mass and fragmentation profile obtained through the Mass Frag- The faeces samples were collected in the cecum and diluted ment tool combined with Chemspider database, it was possible with deionized water (1 mg mL−1), and then were homog- to assign the peaks with retention time 2.27 min and 2.48 min 580 Journal of Functional Foods 17 (2015) 575–586

for the molecules orientin/isoorientin and vitexin/isovitexin, inhibiting around to 28–43% of the inflammation (Kupeli, Aslan, respectively. However, it is noteworthy that this analytical ap- Gurbuz, & Yesilada, 2004). proach still has some limitations in differentiating isomers, The concentration of the P. edulis aqueous extract used in which does not allow us to differentiate them by accurate this study (1100 g leaves mL−1) was chosen based on the 50% masses or fragmentation profile. scavenge of DPPH radical (da Silva et al., 2013). Aqueous P. edulis The interest in natural antioxidants has increased consid- extract was tested in acute and subacute assays, and no sig- erably due to their action in the prevention of certain chronic nificant alterations were found in the organ weight, or diseases, and these compounds can be supplied by a bal- haematological and biochemical parameters, showing that the anced diet rich in fruits, vegetables, and leaves, or by oral administration up to 2000 mg kg−1 of this extract, and the supplementation (Willcox, Ash, & Catignani, 2004). Further- intake of different concentrations (100, 200, 300, 400 mg kg−1), more, teas are rich in flavonoids and are the second-most is safe (Devaki, Beulah, Akila, & Gopalakrishnan, 2012). popular beverage consumed around the world (El-Beshbishy, The consumption of aqueous P. edulis extract significantly 2005). Although the Brazilian legislation establishes that the decreased the weight gain in the Tea Saline group (1.14-fold) P. edulis L. tea should be prepared from the pulp, we observed compared to the Control Saline group (Table 2). In addition, in- that the leaves also have bioactive compounds in their com- creases in food (1.34-fold) and fluid (1.54-fold) intake were position with potential to improve health. The aqueous P. edulis observed between the Tea Colitis group and the Control Colitis extract was shown to be a source of three flavone C-glycosides: group (Table 2), without change in the body weight gain. This isoorientin, isovitexin, and vitexin (Table 1 and Fig. 1). Anti- reduction in the weight gain may be related to the flavones inflammatory activity of vitexin was observed in mice paw after present in the tea, which corroborates with other studies that treatment with 10 mg kg−1 (30 min before the i.p. injection of showed a significant reduction in the weight after consump- carrageenan). Vitexin was able to decrease the expression of tion of flavonoid-enriched extract from Nelumbo nucifera leaf pro-inflammatory cytokines (TNF-α, IL-1β, IL-6, and IL-33) re- (Devaki et al., 2012) and green tea polyphenols (Lu, Zhu, Shen, sponsible by hyperalgesia induced by carrageenan (Borghi et al., & Gao, 2012). Decrease in body weight gain could be related 2013). Pro-inflammatory cytokines, like IL-1β, TNF-α and IL-6, to the ability of vitexin and isovitexin to inhibit the are closely related to the oxidative stress, mainly by NADPH α-glucosidase activity, contributing to a decrease in glucose oxidase activation and superoxide anion production. The in- uptake (Choo, Sulong, Man, & Wong, 2012). In the same way, creased production of this reactive oxygen species promotes Yang, He, and Lu (2014) studied the interaction of four C-glycosyl the activation of the NF-kB pathway that leads to an increase flavonoids (orientin, isoorientin, vitexin and isovitexin) ex- in pro-inflammatory cytokine production, feeding the inflam- tracted from Phyllostachys edulis leaves, with starch and matory cycle (Morgan & Liu, 2011). Lin et al. (2005) observed α-amylase. The authors showed that these flavones can inter- the anti-inflammatory activity of isovitexin in vitro by inter- act with the starch decreasing its digestibility and the activity fering IKK phosphorylation, decreasing the NF-kB translocation of the α-amylase (flavones C6-glycosides flavones C8-glycosides). from the cytoplasm to the nucleus. In addition, isoorientin (15 The consumption of aqueous P. edulis extract improved the and 30 mg kg−1), another C-glycosyl flavonoid, showed anti- content of serum proteins, albumin, and antioxidant status ac- inflammatory activity in carrageenan-induced inflammation cording to ORAC assays in healthy animals compared to the

Table 1 – C-glycosyl flavonoids identified in the Passiflora edulis leaves aqueous extract. Analite Molecular Nominal accurate Experimental Error Error Fragments RT (min) formula mass accurate mass (ppm) (mDa) exact mass [M-H]− [M-H]− MSE m/z

− − Isoorientin/ C21H20O11 447.0927 447.0925 −0.4 −0.2 [M-H -90] : −C3H5O3 (m/z 357.0612) 2.27 − − orientin [M-H -90-30] : −C4H7O4 (m/z 327.0504) − − [M-H -90-30-28] : −C5H7O5 (m/z 299.0553) − − Vitexin/ C21H20O10 431.0978 431.0971 −1.3 −0.7 [M-H -90] : −C3H6O3 (m/z 341.0656) 2.49 − − isovitexin [M-H -90-30] : −C4H8O4 (m/z 311.0549) − − [M-H -90-30-28] : −C5H8O5 (m/z 283.0612)

Table 2 – Results of weight gain (g), food (g day−1) and hydro daily intake (mL day−1) of male Wistar rats with and without TNBS-induced colitis. Group Subgroup Weight gain Feed intake Hydro intake

TEA Colitis 286.0 ± 19.46c 18.5 ± 5.17a 54.2 ± 13.48a Saline 343.1 ± 25.80b 22.0 ± 4.62a 43.4 ± 8.34a,b Control Colitis 312.2 ± 34.31b,c 13.8 ± 2.85b 35.1 ± 1.45b,c Saline 391.7 ± 10.16a 24.9 ± 2.49a 25.0 ± 6.07c TEA = rats drank P. edulis aqueous extract; control = rats drank water. Results expressed in mean ± SEM. Values (columns) followed by different superscript represent signiﬁcant difference (P < 0.05). Journal of Functional Foods 17 (2015) 575–586 581

Control Saline group (Table 3). In addition, a decreased level Increases in serum cytokine (IL-1β, IL-6, and TNF-α) levels of serum TBARS was observed in the Tea Colitis group (1.38- were reported in subjects with IBD (Street et al., 2004), but this fold) compared to the Control Colitis group. Similar results were was not observed amongst the groups in this study (Table 3). observed after the consumption of two cups of green tea This result could be related to the time that the flavones remain (250 mg of total catechins) by healthy volunteers during 42 days; in the intestine. Zeng et al. (2013) observed that more than 50% increases in total antioxidant capacity in plasma and de- of the flavone C-glucosides remain until 12 h in the gastroin- creases in the peroxide level were shown in the Tea group testinal tract and that this increase in passage time through compared to the control (Erba et al., 2005). Moreover, the con- the colon could favour their antioxidant activity. sumption of green tea by weight-trained men increased the Aerobic organisms use a series of antioxidant defences, en- antioxidant activity after exercise, suggesting that the con- dogenous and exogenous, in an attempt to protect themselves sumption of the tea may offer protection to athletes against against oxidative damage (Roessner, Kuester, Malfertheiner, & oxidative damage caused by the exercise (Panza et al., 2008). Schneider-Stock, 2008). There are several mechanisms in- In a previous study, we demonstrated that the consumption volved in the immune and inflammatory responses that lead of aqueous P. edulis extract did not change the antioxidant ac- to tissue injury in IBD, and ROS may play an important role. tivity or TBARS level in the serum of Wistar rats; however, Continuous production of ROS by the immune system (neu- decreases of 20% in the liver TBARS were detected in the group trophils, macrophages, and eosinophils) may deplete the local that drank the same tea compared to the control (da Silva et al., antioxidant enzymes leading to oxidative stress and damage 2013). to tissues (Roessner et al., 2008).

Table3–Protein content, antioxidant status, lipid peroxidation and cytokine level in serum of male Wistar rats with and without TNBS-induced colitis. Serum analyses Tea colitis Tea saline Control colitis Control saline

Total protein* 3.9 ± 0.45a 4.1 ± 0.50a,b 3.4 ± 0.17b,c 3.0 ± 0.04c Albumin* 3.4 ± 0.17a 3.6 ± 0.18a 3.4 ± 0.17a,b 3.0 ± 0.04b FRAP** 461.2 ± 23.15b 579.3 ± 17.13a 504.1 ± 35.17a,b 598.9 ± 32.98a ORAC** 3772.4 ± 262.2a,b 3340.6 ± 296.5b 4290.2 ± 47.58a 2649.5 ± 137.16c TBARS*** 1.3 ± 0.17b 1.3 ± 0.02b 1.8 ± 0.09a 1.5 ± 0.12b IL-1β# 0.6 ± 0.07a 0.6 ± 0.09a 0.6 ± 0.07a 0.6 ± 0.06a IL-6# 0.5 ± 0.34a,b 0.5 ± 0.21a,b 0.3 ± 0.02b 0.4 ± 0.02a TNF-α# 1.0 ± 0.25a 0.5 ± 0.19a 1.0 ± 0.34a 0.9 ± 0.17a *mgdL−1;** mol TE L−1; *** nmol MDA mL−1; # ng mL−1. FRAP – ferric reducing power assay; ORAC – oxygen radical absorbance capacity. Results expressed in mean ± SEM. Values (rows) followed by different superscript represent signiﬁcant difference (P < 0.05).

Table 4 – Hepatic antioxidant activity of male Wistar rats with and without TNBS-induced colitis. Liver analyses Tea colitis Tea saline Control colitis Control saline

GSH* 46.5 ± 3.66a,b 57.0 ± 5.36a 35.2 ± 1.38c 40.0 ± 3.86b,c GR** 18.3 ± 2.79a 20.6 ± 2.02a 10.3 ± 0.79b 9.4 ± 1.16b GPx** 6.8 ± 1.52b 2.9 ± 0.55c 12.0 ± 0.82a 9.5 ± 1.26a,b SOD*** 6.5 ± 0.83b 12.6 ± 1.66a 6.9 ± 0.61b 10.5 ± 0.77a TBARS# 0.6 ± 0.06b 1.1 ± 0.12a 1.1 ± 0.07a 0.9 ± 0.04a * nmol GSH mg−1 protein; ** nmol NADPH consumed min−1 mg−1 protein; *** SOD Units mg−1 protein; # nmol MDA mL−1. Results expressed in mean ± SEM. Values (rows) followed by different superscript represent signiﬁcantly difference (P < 0.05).

Table 5 – Microbiota (Biﬁdobacterium, Lactobacillus, total aerobic bacteria and Enterobacteriaceae), pH and short-chain fatty acid in caecal content of male Wistar rats with and without TNBS-induced colitis. Microbiota analyses Tea colitis Tea saline Control colitis Control saline

Lactobacillus* 10.8 ± 0.17a,b 10.9 ± 0.06a 10.5 ± 0.29a,b 10.4 ± 0.11b Biﬁdobacterium* 10.2 ± 0.30a 10.2 ± 0.30a 10.3 ± 0.23a 9.9 ± 0.20a Total aerobic* 10.1 ± 0.17a 9.9 ± 0.13a 9.8 ± 0.16a 9.6 ± 0.12a Enterobacteriaceae* 9.7 ± 0.24a 10.0 ± 0.02a 9.3 ± 0.22a 9.6 ± 0.23a pH 8.8 ± 0.07a 8.3 ± 0.10c 8.8 ± 0.03a 8.5 ± 0.14b,c Acetic acid** 101.2 ± 33.84a,b 94.5 ± 26.54a,b 51.1 ± 7.59b 83.1 ± 8.15a Butyric acid** 3.6 ± 0.95b 3.6 ± 0.97b 12.0 ± 1.52a 13.4 ± 1.92a Propionic acid** 4.0 ± 0.49b,c 3.4 ± 0.58c 9.4 ± 1.64b 11.1 ± 1.28a,b * CUF g−1 wet faeces; ** nmol mg−1 wet faeces. Results expressed in mean ± SEM. Values (rows) followed by different superscript represent sig- niﬁcant difference (P < 0.05). 582 Journal of Functional Foods 17 (2015) 575–586

The endogenous antioxidant defences and some markers et al., 2007). A significant decrease in the liver TBARS level in of oxidative stress were analysed in the colon and liver. The the Tea Colitis groups (1.81-fold) was observed in comparison macroscopic evaluation of the damage on colon tissue did not to the Control Colitis group (Table 4). Similar hepatic effects show any difference between the tea colitis (7.8 ± 0.25) and in MDA and GSH levels were observed in animals fed olive leaf control colitis (6.7 ± 0.61) groups, similar to the microscopic extract (Coban et al., 2014). The animals from groups that drank results and the colonic weight/length ratio (Fig. 2). The con- Tea showed decreased activities of hydrogen peroxide enzyme sumption of the tea did not change the GSH, GPx, or GR values GPx in the liver and lower levels of lipid peroxides in the liver in colon tissue, but the animals that drank Tea showed de- and colon tissues. In this context, the decreased GPx in the liver creased SOD (1.70-fold Colitis and 1.81-fold Saline) compared could be related to the decrease in ROS formation, as well as to the control groups, respectively (Fig. 3). However, GSH and in the colon where decreases in the SOD activity (the major GR activities were 1.3- and 1.8-fold enhanced, respectively, in source of hydrogen peroxide) were detected. Similar results were the liver of the Tea Colitis group compared to the Control Colitis observed in our previous study, i.e., the healthy rats that drank group (Table 4). Glutathione, which has reducing power, is main- aqueous P. edulis extract showed decreases in GPx and SOD in tained in the reduced form by GR activity, which acts together the liver compared to the control group that did not drink the with NADPH (Nieto et al., 2000). Therefore, the increment or tea (da Silva et al., 2013). maintenance of total glutathione and GR could be an indica- The Tea groups showed decreases in the cytokine expres- tion of an improvement in antioxidant status after aqueous sion in the colon tissue compared to the Control groups (Fig. 4). P. edulis extract intake. An increase in the pro-inflammatory interleukins expres- MPO is primarily found in lysosomes of neutrophils and pro- sion, such as IL-1β, IL-6, and TNF-α, is observed in intestinal duces high amounts of ROS (Krawisz et al., 1984). The Tea and mucosa and these interleukins are one of several inflamma- Control groups showed the same MPO activity in the colon tory mediators that induce lesions in the intestinal mucosa tissue; the Tea Colitis group showed 3.06-fold decreased colonic (Neurath, 2014). The consumption of the tea promoted de- TBARS level in comparison to the control colitis group. Fur- creased expression of IL-1β (5.07-fold) and IL-6 (2.16 fold) in thermore, a decrease of 4.64-fold in MPO activity was observed the colon tissue of the Tea Colitis group compared to the Control in the Tea Saline group compared to the Control Saline group Colitis group. In addition, the cytokine TNF-α, which might in the colon tissue (Fig. 3). The anti-inflammatory effects of prolong the inflammation by activation of nuclear factor kappa P. edulis were previously observed by Vargas et al. (2007). Using B, was 1.99-fold lower in the colon tissue of the Tea Colitis group a mouse model of pleurisy the authors observed that the dose than in the Control Colitis group. of 250 mg kg−1, i.p., of P. edulis aqueous leaf extract was effec- Large amounts of dietary polyphenols are not absorbed in tive to reduce 49 + 11% of the myeloperoxidase activity (Vargas the gastrointestinal tract and these unabsorbed compounds

Fig. 2 – Colonic weight/length ratio, and macroscopic and microscopic scores (A) and, macroscopic changes in the architecture of the colon tissue in the TNBS model of colitis (B). Colitic mucosa is typically inﬂamed and is characterized by bowel wall thickening, hyperemia, and areas of necrosis. Tea = animals that drank aqueous P. edulis extract; Control = drank water. Data expressed as mean ± SEM. Different letters in the columns indicate statistical differences (P < 0.05) between groups. Journal of Functional Foods 17 (2015) 575–586 583

Fig. 3 – Antioxidant status, TBARS myeloperoxidase activity in colon tissue. Tea = animals that drank aqueous P. edulis extract; Control = drank water. Data expressed as mean ± SEM. Different letters in the columns indicate statistical differences (P < 0.05) between groups.

Fig.4–Pro-inﬂammatory cytokines in colon tissue. Tea = animals that drank aqueous P. edulis extract; Control = drank water. Data expressed as mean ± SEM. Different letters in the columns indicate statistical differences (P < 0.05) between groups. 584 Journal of Functional Foods 17 (2015) 575–586

that remain in the lumen can be fermented and can modu- REFERENCES late the microbiota through their metabolites (Duda-Chodak, 2012; Parkar et al., 2013). This modulation is based on the inhibition of the growth and adhesion of bacterial pathogens in Bateman, K. P., Castro-Perez, J., Wrona, M., Shockcor, J. P., Yu, K., the enterocytes, and on improving the proliferation and ad- Oballa, R., & Nicoll-Griffith, D. A. (2007). MSE with mass defect hesion of probiotic bacteria (Duda-Chodak, 2012). filtering for in vitro and in vivo metabolite identification. The consumption of green tea by volunteers who did not Rapid Communications in Mass Spectrometry, 21(9), 1485–1496. doi:10.1002/rcm.2996. usually consume this tea, for 10 days, was able to improve the Bell, C. J., Gall, D. G., & Wallace, J. L. (1995). Disruption of colonic proportion of Bifidobacterium species in the microbiota (Jin, electrolyte transport in experimental colitis. American Journal Touyama, Hisada, & Benno, 2012). The intake of the aqueous of Physiology, 268(4 Pt. 1), G622–G630. P. edulis extract improved by 1.05-fold the count of Lactobacil- Benzie, I. F., & Strain, J. J. (1996). The ferric reducing ability of lus in the caecal content of the Tea Saline group when compared plasma (FRAP) as a measure of “antioxidant power”: The FRAP to the Control Saline group, and changes in the Bifidobacterium, assay. Analytical Biochemistry, 239(1), 70–76. doi:10.1006/ abio.1996.0292. total aerobic bacteria, and Enterobacteriaceae counts were ob- Bezerra, J. A., Campos, A. C., Vasconcelos, P. R., Nicareta, J. R., served between the groups (Table 5). The influence of some Ribeiro, E. R., Sebastiao, A. P., Urdiales, A. I., Moreira, M., & polyphenols on microbiota composition was evaluated by Parkar Borges, A. M. (2006). Extract of Passiflora edulis in the healing et al. (2013). The authors incubated the polyphenols with faecal of colonic anastomosis in rats: A tensiometric and fresh samples and after 24 and 48 h, they observed increase morphologic study. Acta Cirúgica Brasileira, 21(Suppl. 3), growth of Bacteroidetes and Firmicutes, as well as an in- 16–25. crease in short-chain fatty acid content (Parkar et al., 2013). Borghi, S. M., Carvalho, T. T., Staurengo-Ferrari, L., Hohmann, M. S. N., Pinge-Filho, P., Casagrande, R., & Verri, W. A. (2013). There is evidence of the correlation between microbiota and Vitexin inhibits inflammatory pain in mice by targeting the development of inflammatory bowel disease, especially TRPV1, oxidative stress, and cytokines. Journal of Natural because the microbe Lactobacillus appears to be protective, while Products, 76(6), 1141–1149. doi:10.1021/np400222v. Bacteriodes is detrimental (Perez-Munoz et al., 2014). Our results Bradford, M. M. (1976). A rapid and sensitive method for the showed that the consumption of the tea was not able to quantitation of microgram quantities of protein utilizing the improve the level of SCFA in the faeces. This result is in ac- principle of protein-dye binding. Analytical Biochemistry, 72, cordance with our previous data that showed decreases in acetic 248–254. Carlberg, I., & Mannervik, B. (1985). Glutathione reductase. and butyric acids in the animals that received the tea (da Silva Methods in Enzymology, 113, 484–490. et al., 2013). Chinen, T., Komai, K., Muto, G., Morita, R., Inoue, N., Yoshida, H., New strategies for IBD therapies have been investigated in Sekiya, T., Yoshida, R., Nakamura, K., Takayanagi, R., & order to improve the health status of the subjects. Tea is com- Yoshimura, A. (2011). Prostaglandin E2 and SOCS1 have monly consumed worldwide and could be a source of natural a role in intestinal immune tolerance. [10.1038/ncomms1181]. antioxidants to improve the body antioxidant status. The tea Nature Communications, 2, 190. doi:10.1038/ncomms1181. Choo, C. Y., Sulong, N. Y., Man, F., & Wong, T. W. (2012). Vitexin prepared from P. edulis leaves provides significant amounts of and isovitexin from the Leaves of Ficus deltoidea with vitexin, isovitexin and isoorientin. Their intake has been shown in-vivo α-glucosidase inhibition. Journal of Ethnopharmacology, to be effective in the improvement of the hepatic antioxidant 142(3), 776–781. http://dx.doi.org/10.1016/j.jep.2012 status; the tea decreased serum, colonic and hepatic lipid .05.062. peroxidation and decreased the pro-inflammatory cytokine level Coban, J., Oztezcan, S., Dogru-Abbasoglu, S., Bingul, I., Yesil- in the colon tissue. In conclusion, the consumption of this Mizrak, K., & Uysal, M. (2014). Olive leaf extract decreases potent source of antioxidants by IBD patients could mini- age-induced oxidative stress in major organs of aged rats. Geriatrics & Gerontology International., doi:10.1111/ggi.12192. mize inflammatory events by preventing lipid peroxidation and Colomeu, T. C., Figueiredo, D., Cazarin, C. B. B., Schumacher, N. S. by improving the in vivo antioxidant status. G., Maróstica, M. R., Jr., Meletti, L. M. M., & Zollner, R. L. (2014). Antioxidant and anti-diabetic potential of Passiflora alata Curtis aqueous leaves extract in type 1 diabetes mellitus (NOD-mice). International Immunopharmacology, 18(1), 106–115. Conflict of interest doi:10.1016/j.intimp.2013.11.005. da Silva, J. K., Cazarin, C. B. B., Colomeu, T. C., Batista, A. G., Meletti, L. M. M., Paschoal, J. A. R., Bogusz, S., Furlan, M. F., The authors declare that they have no conflicts of interest per- Reyes, F. G. R., Augusto, F., Marostica, M. R., & Zollner, R. D. taining to this study. (2013). Antioxidant activity of aqueous extract of passion fruit (Passiflora edulis) leaves: In vitro and in vivo study. Food Research International, 53(2), 882–890. doi:10.1016/ j.foodres.2012.12.043. Acknowledgments de Mejía, E. G., Song, Y. S., Heck, C. I., & Ramírez-Mares, M. (2010). Yerba mate tea (Ilex paraguariensis): Phenolics, antioxidant The authors wish to thank Fundação de Amparo à Pesquisa capacity and in vitro inhibition of colon cancer cell do Estado de São Paulo (FAPESP grant n° 2010/16752-3), proliferation. Journal of Functional Foods, 2(1), 23–34. http://dx.doi.org/10.1016/j.jff.2009.12.003. Coordenação de Aperfeiçoamento de Pessoal de Nível Supe- Devaki, K., Beulah, U., Akila, G., & Gopalakrishnan, V. K. (2012). rior, and Conselho Nacional de Desenvolvimento Científico e Effect of aqueous extract of Passiflora edulis on biochemical Tecnológico for their financial support. MRMJ acknowledges and hematological parameters of Wistar albino rats. Toxicology CNPq for financial support (CNPq 300533/2013-6). International, 19(1), 63–67. doi:10.4103/0971-6580.94508. Journal of Functional Foods 17 (2015) 575–586 585

Dhawan, K., Dhawan, S., & Sharma, A. (2004). Passiflora: A review Morris, G. P., Beck, P. L., Herridge, M. S., Depew, W. T., Szewczuk, update. Journal of Ethnopharmacology, 94(1), 1–23. doi:10.1016/ M. R., & Wallace, J. L. (1989). Hapten-induced model of chronic j.jep.2004.02.023. inflammation and ulceration in the rat colon. Gastroenterology, Duda-Chodak, A. (2012). The inhibitory effect of polyphenols on 96(3), 795–803. human gut microbiota. Journal of Physiology and Pharmacology, Neurath, M. F. (2014). Cytokines in inflammatory bowel disease. 63(5), 497–503. [Review]. Nature Reviews. Immunology, 14(5), 329–342. El-Beshbishy, H. A. (2005). Hepatoprotective effect of green tea doi:10.1038/nri3661. (Camellia sinensis) extract against tamoxifen-induced liver Nieto, N., Torres, M. I., Fernandez, M. I., Giron, M. D., Rios, A., injury in rats. Journal of Biochemistry and Molecular Biology, 38(5), Suarez, M. D., & Gil, A. (2000). Experimental ulcerative colitis 563–570. impairs antioxidant defense system in rat intestine. Digestive Ellman, G. L. (1959). Tissue sulfhydryl groups. Archives of Diseases and Sciences, 45(9), 1820–1827. Biochemistry and Biophysics, 82(1), 70–77. Ohkawa, H., Ohishi, N., & Yagi, K. (1979). Assay for lipid peroxides Erba, D., Riso, P., Bordoni, A., Foti, P., Biagi, P. L., & Testolin, G. in animal tissues by thiobarbituric acid reaction. Analytical (2005). Effectiveness of moderate green tea consumption on Biochemistry, 95(2), 351–358. doi:10.1016/0003-2697(79) antioxidative status and plasma lipid profile in humans. The 90738-3. Journal of Nutritional Biochemistry, 16(3), 144–149. doi:10.1016/ Panza, V. S., Wazlawik, E., Ricardo Schutz, G., Comin, L., Hecht, K. j.jnutbio.2004.11.006. C., & da Silva, E. L. (2008). Consumption of green tea favorably Flohe, L., & Gunzler, W. A. (1984). Assays of glutathione affects oxidative stress markers in weight-trained men. peroxidase. Methods in Enzymology, 105, 114–121. Nutrition (Burbank, Los Angeles County, Calif.), 24(5), 433–442. Jin, J. S., Touyama, M., Hisada, T., & Benno, Y. (2012). Effects of doi:10.1016/j.nut.2008.01.009. green tea consumption on human fecal microbiota with Parkar, S. G., Trower, T. M., & Stevenson, D. E. (2013). Fecal special reference to Bifidobacterium species. Microbiology and microbial metabolism of polyphenols and its effects on Immunology, 56(11), 729–739. doi:10.1111/j.1348- human gut microbiota. Anaerobe, 23(0), 12–19. doi:10.1016/ 0421.2012.00502.x. j.anaerobe.2013.07.009. Keshavarzian, A., Sedghi, S., Kanofsky, J., List, T., Robinson, C., Perez-Munoz, M. E., Bergstrom, K., Peng, V., Schmaltz, R., Jimenez- Ibrahim, C., & Winship, D. (1992). Excessive production of Cardona, R., Marsteller, N., McGee, S., Clavel, T., Ley, R., Fu, J., reactive oxygen metabolites by inflamed colon: Analysis by Xia, L., & Peterson, D. A. (2014). Discordance between changes chemiluminescence probe. Gastroenterology, 103(1), 177–185. in the gut microbiota and pathogenicity in a mouse model of Krawisz, J. E., Sharon, P., & Stenson, W. F. (1984). Quantitative spontaneous colitis. Gut Microbes, 5(3), 286–295. doi:10.4161/ assay for acute intestinal inflammation based on gmic.28622; 28622 [pii]. myeloperoxidase activity. Assessment of inflammation in rat Podolsky, D. K. (2002). Inflammatory bowel disease. The New and hamster models. Gastroenterology, 87(6), 1344–1350. England Journal of Medicine, 347(6), 417–429. doi:10.1056/ Kupeli, E., Aslan, M., Gurbuz, I., & Yesilada, E. (2004). Evaluation of NEJMra020831. in vivo biological activity profile of isoorientin. Verlag der Prior, R. L., Hoang, H., Gu, L., Wu, X., Bacchiocca, M., Howard, L., Zeitschrift für Naturforschung, 59(11–12), 787–790. Hampsch-Woodill, M., Huang, D., Ou, B., & Jacob, R. (2003). Leite, A. V., Malta, L. G., Riccio, M. F., Eberlin, M. N., Pastore, G. M., Assays for hydrophilic and lipophilic antioxidant capacity & Marostica Junior, M. R. (2011). Antioxidant potential of rat (oxygen radical absorbance capacity (ORAC(FL))) of plasma plasma by administration of freeze-dried jaboticaba peel and other biological and food samples. Journal of Agricultural (Myrciaria jaboticaba Vell Berg). Journal of Agricultural and Food and Food Chemistry, 51(11), 3273–3279. doi:10.1021/jf0262256. Chemistry, 59(6), 2277–2283. doi:10.1021/jf103181x. Reeves, P. G., Nielsen, F. H., & Fahey, G. C., Jr. (1993). AIN-93 Li, F., Li, S., Li, H.-B., Deng, G.-F., Ling, W.-H., Wu, S., Xu, X.-R., & purified diets for laboratory rodents: Final report of the Chen, F. (2013). Antiproliferative activity of peels, pulps and American Institute of Nutrition ad hoc writing committee on seeds of 61 fruits. Journal of Functional Foods, 5(3), 1298–1309. the reformulation of the AIN-76A rodent diet. Journal of http://dx.doi.org/10.1016/j.jff.2013.04.016. Nutrition, 123(11), 1939–1951. Li, H., Zhou, P., Yang, Q., Shen, Y., Deng, J., Li, L., & Zhao, D. (2011). Roessner, A., Kuester, D., Malfertheiner, P., & Schneider-Stock, R. Comparative studies on anxiolytic activities and flavonoid (2008). Oxidative stress in ulcerative colitis-associated compositions of Passiflora edulis ‘edulis’ and Passiflora edulis carcinogenesis. Pathology, Research and Practice, 204(7), 511–524. ‘flavicarpa. Journal of Ethnopharmacology, 133(3), 1085–1090. doi:10.1016/j.prp.2008.04.011. doi:10.1016/j.jep.2010.11.039. Street, M. E., de’Angelis, G., Camacho-Hubner, C., Giovannelli, G., Lin, C. M., Chen, C. T., Lee, H. H., & Lin, J. K. (2002). Prevention of Ziveri, M. A., Bacchini, P. L., Bernasconi, S., Sansebastiano, G., cellular ROS damage by isovitexin and related flavonoids. & Savage, M. O. (2004). Relationships between serum IGF-1, Planta Medica, 68(4), 365–367. doi:10.1055/S-2002-26753. IGFBP-2, interleukin-1beta and interleukin-6 in inflammatory Lin, C. M., Huang, S. T., Liang, Y. C., Lin, M. S., Shih, C. M., Chang, bowel disease. Hormone Research, 61(4), 159–164. doi:10.1159/ Y. C., Chen, T. Y., & Chen, C. T. (2005). Isovitexin suppresses 000075699. lipopolysaccharide-mediated inducible nitric oxide synthase Vargas, A. J., Geremias, D. S., Provensi, G., Fornari, P. E., Reginatto, through inhibition of NF-kappa B in mouse macrophages. F. H., Gosmann, G., Schenkel, E. P., & Fröde, T. S. (2007). Planta Medica, 71(8), 748–753. doi:10.1055/s-2005-871287. Passiflora alata and Passiflora edulis spray-dried aqueous Lu, C., Zhu, W., Shen, C. L., & Gao, W. (2012). Green tea extracts inhibit inflammation in mouse model of pleurisy. polyphenols reduce body weight in rats by modulating Fitoterapia, 78(2), 112–119. obesity-related genes. PLoS ONE, 7(6), e38332. doi:10.1371/ http://dx.doi.org/10.1016/j.fitote.2006.09.030. journal.pone.0038332. Willcox, J. K., Ash, S. L., & Catignani, G. L. (2004). Antioxidants Montanher, A. B., Zucolotto, S. M., Schenkel, E. P., & Frode, T. S. and prevention of chronic disease. Critical Reviews in Food (2007). Evidence of anti-inflammatory effects of Passiflora Science and Nutrition, 44(4), 275–295. doi:10.1080/ edulis in an inflammation model. Journal of Ethnopharmacology, 10408690490468489. 109(2), 281–288. doi:10.1016/j.jep.2006.07.031. Winterbourn, C. C., Hawkins, R. E., Brian, M., & Carrell, R. W. Morgan, M. J., & Liu, Z. G. (2011). Crosstalk of reactive oxygen (1975). The estimation of red cell superoxide dismutase species and NF-κB signaling. Cell Research, 21(1), 103–115. activity. Journal of Laboratory and Clinical Medicine, 85(2), 337– doi:10.1038/cr.2010.178. 341. 586 Journal of Functional Foods 17 (2015) 575–586

Yang, J. P., He, H., & Lu, Y. H. (2014). Four flavonoid compounds Zhao, G., Nyman, M., & Jonsson, J. A. (2006). Rapid determination from Phyllostachys edulis leaf extract retard the digestion of of short-chain fatty acids in colonic contents and faeces of starch and its working mechanisms. Journal of Agricultural and humans and rats by acidified water-extraction and direct- Food Chemistry, 62(31), 7760–7770. doi:10.1021/jf501931m. injection gas chromatography. Biomedical Chromatography, Zeng, P., Zhang, Y., Pan, C., Jia, Q., Guo, F., Li, Y., Zhu, W., & Chen, 20(8), 674–682. doi:10.1002/bmc.580. K. (2013). Advances in studying of the pharmacological Zhao, X., Wang, Q., Li, G., Chen, F., Qian, Y., & Wang, R. (2014). In activities and structure–activity relationships of natural vitro antioxidant, anti-mutagenic, anti-cancer and anti- C-glycosylflavonoids. Acta Pharmaceutica Sinica B, 3(3), 154–162. angiogenic effects of Chinese Bowl tea. Journal of Functional http://dx.doi.org/10.1016/j.apsb.2013.04.004. Foods, 7(0), 590–598. http://dx.doi.org/10.1016/j.jff.2013.12.026.