Changes in bio‐accessibility, polyphenol profile and antioxidants of quinoa and djulis sprouts during in vitro simulated gastrointestinal digestion

Abstract This study aimed to evaluate the bio‐accessibility of the phenolics and flavonoid, the polyphenolic profile and the antioxidant activity of sprouts obtained from four different quinoa genotypes and one djulis cultivar during in vitro gastrointestinal digestion. Compared to their content in sprouts, the bioavailable phenolics after the oral phase, the gastric phase, the intestinal phase, and in the dialyzable fraction were in the ranges of 45.7%–63.5%, 87.6%–116.7%, 89.6%–124.5%, and 7.4%–10.9%, respectively. The trend in flavonoid bio‐accessibility was similar to the polyphenols. The dialyzable flavonoid recoveries varied between 4.2% and 12.4%. Correspondingly, the free radical scavenging activity of the dialyzable phase decreased significantly from 84.7% to 96.5%. The main phenolic acids were vanillic acid, caffeic acid, and syringic acid during digestion. The results suggest that gastrointestinal digestion greatly affected the absorption of polyphenols and flavonoid of quinoa and djulis sprouts, as well as their antioxidant capacity.


| INTRODUC TI ON
Quinoa (Chenopodium quinoa Willd.), an annual dicotyledonous plant in the Amaranthaceae family, originates from the Andean region of South America. It has attracted much attention in recent times for its high nutritional value and resistance to abiotic stresses (Ramesh, Devi, Gopinath, & Praveen, 2019). As a multipurpose crop, quinoa has been cultivated as a leafy vegetable and subsidiary grain for both human consumption and animal feed in different parts of the world (Villa, Russo, Kerbab, Landi, & Rastrelli, 2014). Djulis (Chenopodium formosanum Koidz) is also a pseudocereal crop of the Amaranthaceae family with a close botanical similarity to quinoa and is an indigenous plant to Taiwan that has been consumed for many years. Its beneficial nutritional and health attributes have been highlighted recently (Huang, Chu, Sridhar, & Tsai, 2019;Narkprasom, Wang, Hsiao, & Tsai, 2012). The current global interest in healthier lifestyles has increased market demands for leafy vegetables that are rich in bioactive compounds. Previous studies have shown that quinoa sprouts are a good source of total phenolic compounds with high antioxidant activity that would be suitable for human consumption (Paśko et al., 2009;Paśko, Sajewicz, Gorinstein, & Zachwieja, 2008). The leaves of quinoa possess a broad spectrum of biological activities, such as antioxidant, anti-inflammatory, and cytostatic effects (Chen et al., 2017;Gawlik-Dziki et al., 2013). Therefore, quinoa and djulis could be cultivated as nutraceutically valuable green leafy vegetables due to the high nutraceutical potential of their leaves and sprouts.
Polyphenols are being widely studied for their potential positive effects in foods. Phenolic compounds are the major low molecular weight bioactive components usually found in plants, and have shown various positive bioactivities, such as antioxidant, anti-inflammatory and anticarcinogenic properties (Bellion et al., 2010;Chai et al., 2020;Gawlik-Dziki, 2012;Kim, Hwang, Kim, & Choi, 2020;Shen et al., 2019). Although the level of phenolic compounds may be obtained from composition tables, the total phenolic compounds and other natural antioxidants present in food are not necessarily direct predictors of their real effects on human health (Gunathilake, Ranaweera, & Rupasinghe, 2018;Pellegrini et al., 2017). In order to be absorbed and effective within the human body, phenolic compounds must be released from the plant tissue matrix and modified in the gastrointestinal tract and then absorbed in the gut up to a certain level (Parada & Aguilera, 2007). Bio-accessibility is defined as the proportions of phenolic compounds that are available for absorption during gastrointestinal digestion (Palafox-Carlos, Ayala-Zavala, & González-Aguilar, 2011). Bioactive compounds will exert health benefits if they remain available for absorption after all the phases of the gastrointestinal digestion process are finished. Moreover, there are a number of factors associated with the bio-accessibility of phenolic acids, including proteins, carbohydrates, and lipids, which surround polyphenols inside the gastrointestinal tract (Gong, Chi, Zhang, Wang, & Sun, 2019;Jakobek, 2015;Zhu, 2015). Among the methods used to determine the bio-accessibility of bioactive compounds, simulated gastrointestinal digestion is considered a valid and rapid alternative that does not possess the ethical restrictions of in vivo methods (Pellegrini et al., 2017).
However, there have been no systematic studies on the bio-accessibility of phenolic compounds from quinoa and djulis sprouts during digestion. Understanding the factors that affect the bio-accessibility and bioavailability of bioactive substance is important for evaluating their biological significance and efficacy as functional food ingredients. Therefore, this study aimed to investigate the effect of simulated in vitro gastrointestinal digestion on the recovery and bio-accessibility indexes of polyphenols and flavonoids, the polyphenolic profile and the antioxidant activity of sprouts from quinoa and djulis.

| Plant material and sample preparation
Four different quinoa genotypes and one djulis cultivar were analyzed in this study. The varieties chosen were: Red quinoa (RQ), collected from Yunnan province; White quinoa (WQ), collected Qinghai province; Mengli 1 (Gray quinoa, GQ), which is a new variety from Inner Mongolia; Altiplano from Peru, which was collected from Xinjiang province (AQ); and One djulis cultivar collected from Shandong province (DK). The experiment was conducted in a greenhouse in Qingdao (Shandong, China) in November 2018. Aerial parts of fresh samples were harvested after 30 days of growth. The sprouts were dried at 70°C, and then ground into powder before they were stored at −18°C until further use.

| Extraction procedures
Samples (0.2 g each) were extracted with 40 ml of 80% methanol at 70°C for 2 hr in a water shaker with agitation (120 rpm) and centrifuged at 4,000 rpm for 10 min. The supernatants were subsequently filtered through 0.22 μm filter paper and stored at −18°C until further use.

| Gastrointestinal digestion
The in vitro digestion model was performed as previously described (Gunathilake et al., 2018;Pellegrini et al., 2017) with some modification and the steps are shown in Figure 1. Approximately, 10 g of sample was mixed with 200 ml distilled water and placed in a 500 ml beaker. The oral phase was simulated by adding 1.2 ml of 8 mg/ml salivary α-amylase dissolved in CaCl 2 (1 mmol/L, pH 7.0) solution.
The mixture was incubated in a water bath (37°C) with agitation (100 rpm) for 10 min, protected from light. After the oral digestion, 10 ml of suspension was transferred to a test tube for analysis of bioactive compounds and antioxidant activity, and the reactions were stopped by cooling the test tubes in ice. For the gastric phase, the pH of each sample solution was adjusted to 2.0 with a 6 M HCl solution and 5 ml of pepsin solution (40 mg/ml in 0.1 M HCl) was added, followed by incubation (1 hr, 37°C, 100 rpm). After the gastric digestion, the 10 ml suspension was transferred to a test tube and immediately stopped reaction by cooling in ice. For the intestinal phase with dialysis, segments of cellulose membrane dialysis tubing (flat width: 25 mm, MWCO 14,000 Da) were cut into 15.0 cm lengths, washed with 0.9% NaCl solution, and then one end of each segment was sealed with clips. The dialysis bags are a simplified model that represents the epithelial barrier. The prepared dialysis bags were filled with 5 ml of 0.9% NaCl and 5 ml of 0.5 M NaHCO 3 without leaving any air bubbles inside. Then the other ends of the dialysis bags were sealed with clips and immediately immersed into the appropriate gastric digest, followed by incubation (45 min, 37°C, 100 rpm). After this step, the digest transitioned from the gastric phase to the intestinal phase. The pH of the mixture was adjusted to 6.8 with 6 M NaOH, followed by the addition of 10 ml pancreatin and bile mixture (22 mg/ml pancreatin and 70 mg/ml bile extract dissolved in 0.5 M NaHCO 3 ). The mixture was incubated for 2 hr in a water bath (37°C) with agitation (100 rpm). After incubation, the 10 ml suspension was transferred and stopped reaction in ice. The dialysis bags were carefully separated from the beakers and rinsed with water. The solution in each of the dialysis bags was transferred to a test tube and diluted to a final volume of 20 ml with an addition of 0.9% NaCl and stopped reaction in ice. The proteins and polysaccharides from each of the digested phases were precipitated the suspensions by adding acetone to each test tube make an 80% solution.
Then, the resulting suspension was centrifuged to remove sediment and the supernate was stored at −18°C until analysis.

| Analysis of phenolic compounds
Total polyphenol content (TPC) was determined using the Folin-Ciocalteu method described by our laboratory (Qin, Wu, Yao, & Ren, 2013)  B. The flow rate was 1.0 ml/min, and the injection volume was 10 μL.

| Analysis of flavonoid compounds
Total flavonoid content (TFC) was determined using the aluminum chloride colorimetric method as described by Sarker and Oba (2018) and the results were expressed as milligrams of rutin equivalents (RE) per gram of dry weight of the sample.

| Antioxidant activities
The DPPH radical scavenging assay and the ABTS radical scavenging assay were used for the evaluation of antioxidant activities according our previous study .

| Statistical analysis
Results are expressed as means ± SD of triplicate determinations.
All calculations were performed using SPSS (Statistics for Social Science) version 17.0. Statistical significance was established at probability values of <0.05.

| Effect of in vitro gastrointestinal digestion on the TPC
The TPC in different fractions following in vitro gastrointestinal digestion are shown in Figure 2. The TPC in the methanolic extracts of the sprouts were in the range of 8.34-10.07 mg GAE/g DW, with the RQ sprouts having the highest TPC level while GQ had the lowest level. The behavior of the different quinoa sprout samples differed after the oral phase, with the percentage recovery of polyphenols being within the range of 45.7%-63.5%, relative to the TPC of the methanolic extracts. It can be speculated that only a portion of polyphenols was released during this step. After gastric digestion, the bio-accessible TPC increased dramatically, recovering 87.6%, 98.0%, 116.7%, 102.5% and 100.3% of the TPC from RQ, WQ, GQ, AQ, and DK, respectively. A similar trend has also been reported from quinoa seeds and grape (Pellegrini et al., 2017;Tagliazucchi, Verzelloni, Bertolini, & Conte, 2010). This may be due to the breaking of bonds between the bioactive compounds and nutrients by the acid medium, which helps to the release polyphenolic compounds from the food matrix (Alminger et al., 2014). The bio-accessible TPC after intestinal digestion were in the range of 89.6%-124.5%. The TPC bio-accessibility of the sprouts of the four quinoa genotypes significantly (p < .05) increased after the intestinal phase. In contrast, the TPC bio-accessibility of djulis decreased significantly (p < .05). Tagliazucchi et al. (2010) also reported that phenolic compounds continued to be released during intestinal digestion. Polyphenols bound to vegetal matrices, in the form of esters, glycosides, or polymers, cannot be absorbed (da Silva Haas et al., 2019). However, other studies have found that phenolic compounds can be released from these matrices during digestion under the action of pepsin, trypsin and the environmental pH, and are then absorbed in the gut (Manach, Williamson, Morand, Scalbert, & Rémésy, 2005;Parada & Aguilera, 2007), which may explain why the polyphenol recovery index increased during the gastric and intestinal phases.
The recoveries of dialyzable polyphenols from RQ, WQ, GQ, AQ, and DK were 7.4%, 9.4%, 10.9%, 9.9%, and 8.5%, respectively. These results indicated that almost 90% of the polyphenols could not be absorbed by the intestines. Gunathilake et al. (2018) also reported that the dialyzable polyphenols of six edible green leaves had recoveries that varied between 3.06% and 12.30% relative to their fresh

| Effect of in vitro gastrointestinal digestion on the TFC
The TFC obtained after the oral, gastric, and intestinal phases, and the dialysis and methanol extractions are presented in Figure 3.

| Effect of in vitro digestion on the individual phenolic compounds
The results obtained from the phenolic profile characterization following methanol extraction are presented in Table 1 The results of individual phenolic acid profiles of the oral, gastric, and intestinal samples are presented in Table 2. After digestion, F I G U R E 3 Total flavonoid content of sprouts subjected to simulated in vitro oral, gastric, and intestinal digestion and dialysis (potential uptake); (methanol extract of sprouts). The data presented in this figure consist of average quantities ± SD of three independent samples. Different letters in the bars within each sprout represent statistically significant differences (p < .05) TA B L E 1 HPLC analysis of the polyphenolic profile of quinoa sprouts following methanol extraction (µg/g DW)  & García-Conesa, 2007;Gullon, Pintado, Fernández-López, Pérez-Álvarez, & Viuda-Martos, 2015;Tagliazucchi et al., 2010).
The highest total phenolic content of each digestion phase was observed in GQ sprout, whereas the lowest total phenolic content was seen in WQ sprout.
The potential to take up phenolic acid after the simulated in vitro digestion phase was observed to have decreased significantly ( Note: Data are the means ± SD (n = 3), different letters within the same column indicate significant difference between different sprout varieties at the 0.05 level; Total, sum of detected compounds.

| Antioxidant capacity
The antioxidant activity of different fractions from the in vitro gastrointestinal digestion procedure is shown in  (Cheng et al., 2019;Dziki et al., 2015).
There was a positive correlation between free radical scavenging activity and TPC released from the different digestion steps (ABTS: r = .756, p = .001; DPPH: r = .774, p = .001), as well as between free radical scavenging activity and TFC released from the different digestion steps (ABTS: r = .739, p = .002; DPPH: r = .757, p = .001) ( Figure 4). Similarly, previous study found that there was a positive correlation between total polyphenols of plant material and free radical scavenging activity (Gunathilake & Ranaweera, 2016). Based on the literature, the antioxidant activity of a polyphenol is related to its chemical structure. For example, polyphenol aglycones display a higher antioxidant activity than their glycosides (Kamiloglu, Pasli, Ozcelik, & Capanoglu, 2014). The type and quantity of polyphenolic compounds are related to the scavenging of free radicals (Cao, Sofic, & Prior, 1997

| CON CLUS IONS
The present study suggests that polyphenols and flavonoids of quinoa and djulis sprouts are progressively released and made available for absorption in an in vitro model of gastrointestinal digestion.
Compared to the oral phase, the recovery index of polyphenols and flavonoid significantly increased (p < .05) in the gastric and intestinal phase. However, the final potential uptake of polyphenols and flavonoids after simulated in vitro digestion was significantly lower (p < .05) than the amount originally available in the leaves. Some individual phenolic compounds might degrade or transform during digestion. There was a positive correlation between the antioxidant activity and TPC or TFC released from different digestion steps.
Further studies should be undertaken to support the findings of this study and to better understand the factors that affect the bio-accessibility of bioactive compounds in quinoa and djulis sprouts.

ACK N OWLED G M ENTS
We thank associate professor Xiao-Yan Zhang from Institute of

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

E TH I C A L A PPROVA L
Neither animal nor human testing was involved in this study. anthocyanins and antioxidant activity following simulated gastro-intestinal digestion and dialysis of apple varieties: Bioaccessibility F I G U R E 4 (a) correlation between the phenolic release content and ABTS free radical scavenging activity; (b) correlation between the phenolic release content and DPPH free radical scavenging activity; (c) correlation between the flavonoid release content and ABTS free radical scavenging activity; (d) correlation between the flavonoid release content and DPPH free radical scavenging activity