Food Chemistry 370 (2022) 131001

Contents lists available at ScienceDirect

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

In vitro digestion of the whole blackberry fruit: bioaccessibility, bioactive

variation of active ingredients and impacts on human gut microbiota
Zuman Dou a, Chun Chen a, b, c, d, *, Qiang Huang a, c, Xiong Fu a, c, d, *
a
SCUT-Zhuhai Institute of Modern Industrial Innovation, School of Food Science and Engineering, South China University of Technology, 381 Wushan Road, Guangzhou
510640, China
b
Guangzhou Inst Modern Ind Technol, Nansha 511458, China
c
Guangdong Province Key Laboratory for Green Processing of Natural Products and Product Safety, Guangzhou 510640, China
d
Overseas Expertise Introduction Center for Discipline Innovation of Food Nutrition and Human Health (111 Center), Guangzhou, China

A R T I C L E I N F O A B S T R A C T

Keywords: In vitro digestion and fermentation of blackberry fruit was investigated, and results showed that the phenolics
Blackberry were mainly released in gastric phase while carbohydrates in small intestinal phase. The bioaccessibility for
Bioaccessibility phenolics and carbohydrates were 42.80% and 69.30%, indicating most of phenolics still remain in colon and
Antioxidant
available for intestinal flora. The total phenolics released during the digestion account for the improvement of
Hypoglycemic
antioxidant and hypoglycemic activities. Especially, cyanidin-3-O-glucoside with higher released amount and
Gut microbiota
bioaccessibility index (63.21%), exhibited the strongest α-glucosidase inhibitory activity. After fermentation, the
non-digestible fractions of blackberry affected the ecosystem of the intestinal tract by decreasing the colonic pH
(△pH = 1.10), enhancing the production of SCFAs and modulating gut microbiota composition (the ratio of
Firmicute/Bacteroidetes decreased from13.18 to 0.87). The results provided insights into the digestive properties
and health benefits of blackberry fruit after consumption.

1. Introduction matrix, digestive stabilities and their metabolism in human body (Dou,
Chen, & Fu, 2019a; Zheng et al., 2018). The stability of these phyto­
Type 2 diabetes mellitus (T2DM) as one of the most serious chronic chemicals are easily affected by the environmental factors during the
diseases is characterized by insulin resistance and impaired insulin human digestion, such as pH, digestive enzymes and temperature
secretion (Zhang et al., 2019). Current treatments for diabetes include (Zheng et al., 2018). In addition, not all phytochemicals released during
drug delivery of insulin and its analogues, such as thiazolidinediones, the digestion can be absorbed, some of them will reach the colon and
sulfonylureas, and biguanide. However, these drugs with high price and fermented by gut microbiota to produce some specific active metabolites
side effects are not suitable for people (Wang et al., 2019). Increasing (Mosele et al., 2015). For instance, Stalmach et al reported that coffee
evidences indicated that the consumption of fruits and vegetables rich in chlorogenic acids and phenolics of Concord grape juice exhibited a low
phytochemicals is beneficial for diabetes to control the blood glucose absorption (<10%) (Angelique Stalmach, Edwards, Wightman, &
and restore of pancreas (Chen & Fu, 2019; Chen et al., 2019; Chen et al., Crozier, 2012; Stalmach et al., 2014). However, there was little infor­
2017; Zhang et al., 2019). As a kind of berry, blackberry belonging to the mation about the bioaccessibility and gut microbiota modulations of
genus of Rubus is widely distributed in South and North America, Europe blackberry after consumption.
(Gowd et al., 2018; Martins et al., 2014). Numerous studies have indi­ Therefore, the aim of this study was to determine the bioaccessibility
cated that blackberries are rich in phytochemicals and exhibit various of bioactive components of blackberry fruit using a static in vitro
biological activities including protection against oxidant stress, anti­ digestion model. The antioxidant and hypoglycemic activity changes of
cancer and hypoglycemic (Gowd et al., 2018; Prior et al., 2008; Wada & these phytochemicals during the digestion were also evaluated. In
Ou, 2002; Dou, Chen, & Fu, 2019b). addition, intestinal microbial modulations of blackberry were further
The in vitro properties of these functional components are seriously investigated. The present study could provide insight into the health
related to their bioaccessibility after consumption which are greatly benefits of blackberry after consumption.
affected by several factors such as the extent they are released from food

* Corresponding authors.

https://doi.org/10.1016/j.foodchem.2021.131001
Received 21 March 2021; Received in revised form 24 August 2021; Accepted 29 August 2021
Available online 1 September 2021
0308-8146/© 2021 Elsevier Ltd. All rights reserved.
Z. Dou et al. Food Chemistry 370 (2022) 131001

2. Materials and methods taken any antibiotic or prebiotic supplements for a year prior to fecal
sample donation. For 60 h before providing stool samples, the volun­
2.1. Chemicals and reagents teers followed a low polyphenolic diet, which included avoiding fruits
and vegetables, tea, coffee, wine and wholemeal foods. Volunteers
Phenolic standards, monosaccharide standards, organic acid stan­ provided a stool sample in a fasted state on the morning of the study and
dards, 1,1-diphenyl-2-picrylhydrazyl (DPPH), 2,2′ -azino-bis (3-ethyl­ collected in an anaerobic tube. Meanwhile, samples were processed
benzothiazoline-6-sulfonic acid) (ABTS), 6-hydroxy-2,5,7,8- within 1 h of passage. The nondigestible fractions of blackberry fruit
etramethylchroman-2-carboxylic acid (Trolox), α-amylase, pepsin, were suspended with 30 mL of anaerobic medium and 30 mL of 10%
pancreatin, α-glucosidase, and p-Nitrophenyl-α-D-glucopyranoside were fecal slurry in a 100-mL fermentation vessel at 37 ◦ C for 24 h. The fer­
all purchased from Sigma-Aldrich (St. Louis, MO, USA). Chromato­ mented culture without NFBF and with inulin were used as blank and
graphic grade of acetonitrile and formic acid were purchased from Anpel positive control under the same conditions.
Scientific Instrument Co., Ltd. (Shanghai, China). All other reagents used
to prepare the simulated saliva fluid, simulated gastric fluid, and 2.5. Extraction and identification of phenolics and tannins
simulated intestinal fluid and colonic fluid were analytically grade and
provided by Guangzhou Chemical Reagent Co. Ltd (Guangzhou, China). The extraction of phenolic compounds was performed according to
the reported method (Agudelo et al., 2018; Li et al., 2017). The total
2.2. Plant materials content of phenolic compounds in the ethanol extract and digested
fractions of blackberry fruit were determined by the Folin-Ciocalteu
The freeze-dried blackberry fruit of the cultivar of “Boysen” were method as previously reported (Singleton & Rossi, 1965; Zheng et al.,
provided by Harbin jiacheng xinrong food co. LTD (Heilongjiang, 2018). The total tannins content was measured according the published
China). The freeze-dried blackberry fruits were crushed and passed study (Ivanovic et al., 2014) with slight modifications. Briefly, the
through a 100-mesh sieve to obtain the superfine blackberry powder. blackberry powder (5 g) was extracted deionized water and diluted to
The obtained blackberry powder was packed in an airtight bag and 100 mL and boiled for 30 min. After cooling down, the extraction was
stored at − 80 ◦ C to keep desiccation for further use. centrifuged (7513 g, 4 min) and mixed with sodium tungstate-sodium
molybdate mixture and sodium carbonate solution (75 g/L). The
2.3. In vitro gastrointestinal digestion mixture was incubated for 2 h at room temperature and the absorbance
was recorded at 760 nm. Pyrogallol was used as a standard and the re­
In vitro digestion of blackberry was conducted according to the sults were expressed as gram pyrogallol equivalents per 100 g of dry
method published earlier with few modifications (Brodkorb et al., 2019; fruit (g PE/100 g fruit, DW).
Dou, Chen, & Fu, 2019a; Minekus et al., 2014), which consisted of The major phenolic acid profiles in the ethanol extract and digested
mouth, stomach and small intestine phases. For oral digestion, 2.0 g of fraction of blackberry fruit were analyzed by an Agilent 1260 chro­
the blackberry fruit powder in triplicate was suspended in 18.0 mL of matograph (Agilent Technologies, Santa Clara, CA, USA) equipped with
simulated salivary electrolyte with pH adjusted to 7.0 after the addition a diode array detector (HPLC-DAD) as the method we described earlier
of 2.0 mL of α-amylase solution (75 U/mL solution in 1.5 mM CaCl2). (Gowd et al., 2018; Li et al., 2017). Analyses were carried out on a
The mixture was mixed thoroughly and incubated for 5 min at 37 ◦ C and ZARBAX SB-C18 column (Agilent, 4.6 × 250 mm, 5 μm), and the mobile
pH was kept at 7.0. In gastric phase, the oral remained part was mixed phase consisted of 0.1% (v/v) formic acid in water (solvent A) and
with simulated gastric electrolyte with pH adjusted to 2.5 with HCl (6 acetonitrile (solvent B) using a gradient elution at 0.8 mL/min with the
M). Then the pepsin and CaCl2 was added to afford a final concentration following gradient program, started with 5% B, 0–10 min; 20% B at 15
of 2000 U/mL and 1.5 mM, and incubated for 2 h at 37 ◦ C, the pH of the min; 38% B at 25 min; 40% B at 30 min; 100% B at 31–35 min; 5% B at
mixture was kept at 2.5 all the time. In the intestinal digestion, the pH of 36–50 min. The temperature of the column was maintained at 30 ◦ C, the
the mixture after gastric digestion was adjusted to 7.0 with NaOH (6 M) injection was 5 μL, and the chromatograms were recorded at 254, 280,
immediately. Then, the simulated intestinal electrolyte was added at a 325 and 520 nm. The phenolic compounds were identified by the
ratio of 1:1. The bile salt, CaCl2 and pancreatin were also added (10 mM, retention time comparison with authentic standards injected in the same
0.6 mM and 100 U/mL, respectively). After incubation for 2 h at 37 ◦ C, conditions and with refences to scientific literatures published previ­
the supernatant was transferred to dialysis tube with molecular cut-off ously (Gowd et al., 2018; Schulz et al., 2019). The quantitative analysis
12 KDa to determine the bioaccessibility of active ingredients. Dialysis of the components was achieved through calibration curves of standard
process was against NaCl (10 mM) overnight at 37 ◦ C, and the solution compounds, and all the determinations were conducted in triplicate. The
remained in the dialysis tube was taken as non-absorbable fractions (IN), standard compounds used were include gallic acid, 2,4,6-trihydroxyben­
indicating the residual substance in gastrointestinal tract and reach to zoic acid, 3,4-dihydroxybenzoic acid, catechol, cyanidin-3-O-glucoside,
colon (colon-available). The solution diffused into the dialysis tube was caffeic acid, rutin, p-coumaric acid, quercetin-3-O-glucoside, ferulic acid
considered as gastrointestinal-available fractions (OUT). All the super­ and coumarin.
natant was stored at − 80 ◦ C until analysis, and the residues after
simulated intestinal digestion were freeze-dried and regarded as non­ 2.6. Soluble carbohydrate and organic acid analysis
digestible fractions of blackberry fruit (NFBF). After each phase, 5 mL of
mixture was taken out and centrifuged (4500 g, 10 min) for analysis. Preparation of water extracts from blackberry fruit was performed as
previous (Caicedo-Lopez et al., 2019; Dou, Chen, & Fu, 2019b). In short,
2.4. In vitro colonic fermentation 2.0 g of the blackberry fruit powder was suspended with 100 mL of
deionized water and stirred for 1 h in an oscillating bath (90 ◦ C). The
The nondigestible fractions of blackberry fruit (NFBF) were further mixture was centrifuged (4500 g, 10 min) and the residues were re-
fermented using fresh fecal samples as previously described (Chen et al., extracted twice under the same conditions. The supernatants com­
2018; Stalmach, Edwards, Wightman, & Crozier, 2013; Dou, Chen, & Fu, bined and concentrated to 100 mL for further analysis. The total content
2019a) with some modifications. The study protocol was approved by of soluble carbohydrate was measured by the method of phenol–sulfuric
the South China University and Technology Ethics Committee and the acid with glucose as standard (Chen, You, Abbasi, Fu, & Liu, 2015;
subjects gave written informed consent. Fresh fecal samples were DuBois et al., 1956; Dou, Chen, & Fu, 2019b). The results were expressed
collected from ten healthy volunteers (five male and five female, 20–30 as milligram glucose equivalents per gram of dry fruit (mg GE/g fruit,
years of age) who declared no gastrointestinal diseases and had not DW), all the experiments were performed three times.

2
Z. Dou et al. Food Chemistry 370 (2022) 131001

The identification of the free sugars and oligosaccharide in water the water used as blank control. The results were expressed as milligram
extracts and digested fractions of blackberry fruit was performed using acarbose equivalents per gram of dry fruit (mg AE/g fruit, DW).
an Agilent 1260 chromatograph (Agilent Technologies, Santa Clara, CA, The major phenolic compounds released from blackberry fruit dur­
USA) equipped with a differential refractive index detector (HPLC-RID) ing the in vitro during the digestion were selected to molecular docking
and a ZARBAX Carbohydrate column (Agilent, 4.6 × 150 mm, 5 μm) with α-glucosidase. The X-ray crystallographic structures of the
(Muzquiz et al., 1999). The temperature of the column was controlled at α-glucosidase (PDB: 3A4A) was retrieved from the RCSB Protein Data
30 ◦ C, and the 75% (v/v) acetonitrile solution was used as mobile phase Bank (http://www.rcsb.org/pdb). The structure of 2,4,6-trihydroxyben­
with a flow rate of 1 mL/min. All the free sugars and oligosaccharide zoic acid, cyanidin-3-O-glucoside, gallic acid, caffeic acid and α-gluco­
were identified and quantified by comparing the retention times of pure sidase were downloaded from PubChem Compound (NCBI, htt
standards (glucose, fructose, mannose, maltose, sucrose, raffinose and ps://pubchem.ncbi.nlm.nih.gov) database and converted to 3D struc­
stachyose). tures by Chembio3D ultra (version 14.0) software. The docking pro­
The organic acid profiles were measured according to the method cedure was conducted with AutoDock 4.2 software, and the number of
described previously with few changes (Zeppa et al., 2001). The sepa­ docking runs was set to 100. Furthermore, the obtained docking results
ration was conducted using an Agilent 1260 chromatograph (Agilent were visualized and analyzed by PyMOL (version 1.8) and Discovery
Technologies, Santa Clara, CA, USA) equipped with a diode array de­ Studio 2016 software.
tector (HPLC-DAD) at a wavelength of 215 nm. The temperature of
column was kept at 30 ◦ C. The mobile phase comprised of KH2PO4 (0.01 2.11. SCFAs and gut microbiota analysis
M, pH 2.5, solvent A) and acetonitrile (solvent B) with a flow rate of 0.8
mL/min. Tartaric acid, ascorbic acid, citric acid and malic acid were The pH value of each fermented products was determined by pH
used as standards. meter. The content of SCFAs in fermented culture was determined using
GC–MS according to the method previously reported (Chen et al., 2018).
2.7. Calculation of recovery index and bioaccessibility index Total microbial DNA was extracted from fermented products using Fast
DNA SPIN extraction kits (MP Biomedicals, Santa Ana, CA, USA) fol­
The recovery index and bioaccessibility index were calculated to lowed the manufacturer’s manual. The quantity and quality of extracted
evaluate the effect of the matrix composition on the digestion of the DNAs were measured using a NanoDrop ND-1000 spectrophotometer
phenolics and carbohydrates from blackberry (Martínez-Las Heras et al., (Thermo Fisher Scientific, Waltham, MA, USA) and agarose gel elec­
2017). trophoresis, respectively. DNA samples were sequenced for 16S rRNA
genes using Illumina MiSeq (Illumina, San Diego), targeting the V3-V4
B
Recoveryindex(%) = × 100 region with barcoded 338F (5′ -ACTCCTACGGGAGGCAGCA-3′ ) and
A
806R (5′ -GGACTACHVGGGTWTCTAAT-3′ ) universal primers and pro­
C cessed as previously described (Dou, Chen, & Fu, 2019; Khan et al.,
Bioaccessibilityindex = × 100 2018). Sequence data analyses were mainly performed using QIIME and
A
R packages (v3.2.0). OTU-level alpha diversity indices, such as Chao1
where A is the total phenolics or carbohydrates extracted with ethanol richness estimator, ACE metric (Abundance-based Coverage Estimator),
(90%, v/v) or water (before digestion); B is the total phenolics or car­ Shannon diversity index, and Simpson index, were calculated using the
bohydrates released during the simulated oral, gastric and small intes­ OTU table in QIIME. Beta diversity analysis was performed to investi­
tinal digestion; C is the total phenolics or carbohydrates diffuse into the gate the structural variation of microbial communities across samples
dialysis tube (OUT, gastrointestinal-available) after intestinal dialysis using nonmetric multidimensional scaling (NMDS), principal compo­
step. nent analysis (PCA) was also conducted based on the genus-level
compositional profiles. Microbial functions were predicted by PICRUSt
2.8. Morphological analysis (SEM) (Phylogenetic investigation of communities by reconstruction of unob­
served states), based on high-quality sequences (Langille et al., 2013).
The morphological properties of spray fried powder and digested Using 16S rRNA information, PICRUSt software (package at http://pi
samples of blackberry were observed using a SEM system (Model JEM- crust.github.com/) recaptures key findings from the Human Micro­
2100F, Bruker, Germany) at a 15 KV acceleration voltage (Wang biome Project and accurately predicts the abundance of gene families in
et al., 2019). host-associated and environmental communities, with quantifiable
uncertainty.
2.9. Antioxidant activity analysis
2.12. Statistical analysis
The DPPH and ABTS radical scavenging capacity assay were deter­
mined as the modified method described earlier (Zhang et al., 2019). The experimental data were expressed as mean ± standard deviation
The results were expressed as molar trolox equivalents per gram of dry (SD) of triplicates. Statistical analysis was performed using IBM SPSS
fruit (mol TE/g fruit, DW). version 24.0 (IBM Corp., Armonk, NY, USA). Significant changes were
analyzed by one-way analysis of variance (ANOVA) followed by Dun­
2.10. α-Glucosidase inhibitory activity and mechanism analysis can’s multiple-range test and p < 0.05 was regarded as statistically
significant difference.
The α-glucosidase inhibitory activity was measured according to the
reported method (Chen, You, Abbasi, Fu, Liu, & Li, 2016). α-Glucosidase 3. Results and discussions
from Saccharomyces cerevisiae with a 99.9% purity was purchased from
Sigma-Aldrich (St. Louis, MO, USA). Sample solution was premixed with 3.1. Release dynamics of phenolic compounds and carbohydrates
α-glucosidase (0.35 U/mL) at an equal volume ration and incubated for
10 min in a water bath (37 ◦ C). Then p-nitrophenyl-α-D-glucopyranoside The total content of phenolic compounds and carbohydrates in
(dissolved in 0.1 M of phosphate-buffered saline, pH 6.9) was added and blackberry fruit powder are 2290.62 ± 34.48 mg GAE/100 g, DW and
incubated for another 20 min, followed by the addition of Na2CO3 (1 M) 285.00 ± 10.89 mg GE/g, DW. As shown in Fig. 1A, after oral digestion,
solution to terminate the reaction. The absorbance of the mixture was the total phenolics released were1316.76 ± 3.96 mg GAE/100 g, DW
recorded at 405 nm, and the acarbose was used as positive control, while and increased to 1906.83 ± 4.45 mg GAE/100 g, DW after gastric

3
Z. Dou et al. Food Chemistry 370 (2022) 131001

Fig. 1. Changes of the total content of phenolic compounds (A) and soluble carbohydrates (B) of blackberry fruit released during the in vitro gastrointestinal
digestion at different time point. (A) phenolic compounds; (B) carbohydrates.

digestion, while significantly decreased in intestinal phase. Similar re­ g), DW, the results were in accordance with the release trends of phe­
sults were observed in mulberry and Rubus idaeus L. fruit digestion, nolics that tannins were mainly released in gastric digestion.
which may due to the phenolics especially anthocyanins are sensitive to As for soluble carbohydrates (Fig. 1B), the total soluble carbohy­
alkaline conditions (Agudelo et al., 2018; Qin et al., 2018). The total drates released during the oral and gastric phases showed no significant
tannins content in undigested blackberry fruit was determined as 0.15 g difference (p > 0.05), while obviously increased during small intestinal
PE/100 g, DW, which was higher than that of blackberry pulp and juice digestion with 1.26-fold higher than that of oral phase. This may be due
reported previously (Ivanovic et al., 2014). P. Moraes et al reported that to the rich digestive enzymes that could hydrolyze fiber in small intes­
the content of total condensed tannins varied significantly among tine. The release kinetics of total phenolics and soluble carbohydrates
different cultivars of blackberry, which may account for the higher were further fitted by gaussian function and the R2 were 0.9831 and
tannins content in the present study (Moraes et al., 2020). After oral 0.9899, which provided useful information to monitor the phytochem­
digestion, the tannins content was 0.06 g PE/100 g, DW and increased to icals released from blackberry after consumption (Fig. S1). Recovery
0.12 g PE/100 g, DW after gastric digestion. However, the content of index for soluble carbohydrates was higher than that of extracts, while
tannins decreased significantly after intestinal digestion (0.08 g PE/100 the phenolics was lower. The bioaccessibility index for phenolics

Table 1
The major free phenolic acids and flavonoids identified in undigested samples and digested fractions of blackberry during the in vitro digestion (oral, gastric and
intestinal phase).
a
Sample/ phenolic acids (mg/100 g, DW) flavonoids (mg/100 g, DW) a
Stage
Gallic 2,4,6- 3,4- Caffeic p- Ferulic Catechol Cyanidin- Rutin Quercetin- Coumarin
acid Trihydroxybenzoic Dihydroxybenzoic acid coumaric acid 3-O- 3-O-
acid acid acid glucoside glucoside

undigested 19.7 ± 2433 ± 23.8bc 12.1 ± 2.14a 24.6 ± 12.1 ± 9.37 ± 35.1 ± 651 ± 57.5 ± 33.1 ± 10.0 ±
2.31a 3.41a 1.88a 1.34a 3.03a 12.3c 2.12a 2.31a 1.59a
mouth 5.53 ± 1913 ± 16.6d 1.93 ± 0.35de 2.91 ± nd 2.18 ± 9.06 ± nd 13.8 ± 21.7 ± 8.50 ±
0.65e 0.35f 0.41e 1.02e 0.99d 1.46c 1.21b
stomach
15 min 14.0 ± 2498 ± 25.0ab 2.67 ± 0.65d 21.9 ± 5.27 ± 4.26 ± 11.7 ± 720 ± 20.8 ± 20.6 ± 8.43 ±
0.98c 2.48a 0.97c 0.56d 0.98 cd 15.7b 1.68c 1.33c 0.99b
30 min 17.2 ± 2527 ± 30.2a 4.15 ± 0.77c 22.2 ± 6.09 ± 5.78 ± 18.3 ± 769 ± 24.7 ± 30.0 ± 8.30 ±
0.76b 1.92a 0.56bc 0.45c 1.34b 10.4a 1.43b 2.02a 0.79b
60 min 17.0 ± 2499 ± 18.5ab 5.16 ± 0.74bc 22.7 ± 5.73 ± 6.48 ± 12.1 ± 725 ± 23.4 ± 25.5 ± 8.41 ±
0.66b 0.98a 0.64c 0.32bc 0.87c 12.4b 1.51bc 1.98b 0.67b
120 min 13.9 ± 2476 ± 20.3b 6.82 ± 0.98b 21.9 ± 5.25 ± 6.90 ± 12.2 ± 720 ± 21.6 ± 20.3 ± 8.44 ±
1.02c 1.87a 0.45c 0.46b 0.91c 14.6b 1.06c 1.20c 0.81b
Small
Intestine
15 min 13.1 ± 2428 ± 19.9bc 10.5 ± 1.02a 11.8 ± 7.16 ± 6.34 ± 9.54 ± 375 ± 21.8 ± 6.72 ± 8.49 ±
0.43c 0.64b 0.61b 0.55bc 0.88de 2.31e 1.28c 0.78d 0.76b
30 min 12.8 ± 2411 ± 21.3c 10.7 ± 0.99a 11.6 ± 7.28 ± 6.46 ± 9.47 ± 390 ± 21.7 ± 6.58 ± 8.45 ±
1.21c 0.35b 0.77b 0.61bc 0.76de 2.89d 1.09c 0.54d 0.76b
60 min 12.9 ± 2406 ± 23.6c 10.9 ± 1.21a 11.7 ± 7.28 ± 6.50 ± 9.99 ± 376 ± 20.3 ± 6.77 ± 8.48 ±
0.88c 0.46b 0.56b 0.57bc 1.00d 3.13e 1.33c 0.73d 0.84b
120 min 13.0 ± 2421 ± 28.7c 9.18 ± 1.00a 10.9 ± 7.14 ± 5.90 ± 9.41 ± 366 ± 21.3 ± 6.99 ± 8.48 ±
1.30c 0.61b 0.70b 0.55bc 0.78de 2.99f 1.22c 0.81d 0.75b
IN 7.21 ± 1456 ± 11.4e 5.04 ± 0.23bc 6.42 ± 6.15 ± 2.51 ± 7.35 ± 141 ± 10.9 ± 4.31 ± 5.05 ±
0.31d 0.37c 0.41bc 0.38e 0.31f 2.56 h 0.87e 0.15e 0.43c
OUT 5.02 ± 980 ± 19.9f 5.22 ± 0.49bc 4.97 ± 2.84 ± 4.40 ± 2.73 ± 243 ± 12.7 ± 2.99 ± 3.37 ±
0.21e 0.24d 0.14d 0.43d 0.58 g 1.33 g 1.21d 0.40f 0.52d
BI (%) 41.0 40.2 50.9 43.6 31.6 63.7 27.1 63.2 53.7 41.0 40.0

nd = not detectable; BI = bioaccessibility;

In: colon-available fractions; OUT: gastrointestinal-available fractions;
The undigested sample refers to the blackberry extracts extracted from ethanol (90%, v/v).
a
Different letter in the same column means significant differences among the groups at p < 0.05; The results were expressed as mean ± standard deviation (SD);

4
Z. Dou et al. Food Chemistry 370 (2022) 131001

(42.79%) was also lower than soluble carbohydrates (69.30%). This Table 2
could be explained that when the food matrix is exposed to gastroin­ The major organic acids and free sugars identified in undigested samples and
testinal conditions, a proportion of phenolics may be converted into digested fractions of blackberry during the in vitro digestion (oral, gastric and
different structure forms or interacted with other compounds and hardly intestinal phase).
be detected (Agudelo et al., 2018; Mosele et al., 2015). Additionally, the Sample/ organic acid (mg/g, DW) a
Free sugars (mg/g,
phenolics could interact with food matrix such as binding to lipid and Stage DW) a
protein and affect its absorption (Gullon et al., 2015). Tartaric Ascorbic Citric Malic fructose glucose
acid acid acid acid
3.2. Identification of main phenolic compounds, organic acids and free undigested 26.9 ± 4.22 ± 62.8 ± 4.83 130 ± 153 ±
sugars 1.20a 0.88ab 3.12a ± 4.38e 5.30e
0.45 g
mouth 7.82 ± 1.09 ± 16.4 ± 20.4 147 ± 181 ±
As shown in Table 1 and Fig. S2, a total of eleven phenolic com­
0.77e 0.57c 2.44e ± 6.36d 5.87d
pounds were identified in the digestive process, including gallic acid, 1.21e
2,4,6-trihydroxybenzoic acid, 3,4-dihydroxybenzoic acid, catechol, stomach
cyanidin-3-O-glucoside, caffeic acid, rutin, p-coumaric acid, quercetin- 15 min 11.2 ± 3.37 ± 29.5 ± 22.7 148 ± 184 ±
3-O-glucoside, ferulic acid and coumarin. 2,4,6-trihydroxybenzoic acid 0.43d 0.94b 2.76d ± 7.22d 5.42 cd
0.97e
and cyanidin-3-O-glucoside were the two most widely released com­ 30 min 23.6 ± 5.50 ± 35.8 ± 22.5 153 ± 194 ±
pounds during the all digestion steps. It has been reported that cyanidin- 1.28b 0.67a 2.31c ± 5.78d 6.03bc
3-O-glucoside plays a key role in the antioxidant capacity (Bao et al., 1.33e
2019). However, p-coumaric acid and cyanidin-3-O-glucoside could not 60 min 23.9 ± 5.49 ± 36.2 ± 25.5 159 ± 196 ±
1.09b 0.71a 2.55c 6.75d 6.40b
be detected during the oral digestion, same results were observed in
±
1.19d
previous study (Gowd et al., 2018). All phenolic compounds signifi­ 120 min 24.0 ± 5.43 ± 39.5 ± 28.4 180 ± 203 ±
cantly released in gastric digestion decreased in small intestinal phase 1.34b 0.89a 2.08bc ± 7.84c 5.92b
which may account for the total content phenolics variation, and similar 1.16c
results were observed in the digestion of maqui berry and Concord grape Small
Intestine
juice (Lucas-Gonzalez et al., 2016; Angelique Stalmach, Edwards, 15 min 23.5 ± nd 42.2 ± 31.8 193 ± 213 ±
Wightman, & Crozier, 2012). As for individual phenolics, ferulic acid 1.18b 2.74b ± 6.67bc 5.15ab
(63.68%) and cyanidin-3-O-glucoside (63.21%) exhibited higher bio­ 1.34b
accessibility than other phenolics, especially catechol was just 27.08% 30 min 22.6 ± nd 40.3 ± 37.9 207 ± 217 ±
0.99b 2.71b 5.42a 6.37a
and showed the lowest bioaccessibility index. The results indicated that ±
1.47a
different phenolics exhibited different bioaccessibility, which may be 60 min 22.2 ± nd 41.1 ± 37.6 203 ± 210 ±
due to that some individual phenolic compounds could be degraded or 0.97b 2.56b ± 6.13ab 6.01ab
converted into other chemical compounds. However, most of phenolic 1.49a
compounds could not be absorbed by gastrointestinal tract and reached 120 min 22.1 ± nd 41.6 ± 37.2 197 ± 194 ±
0.86b 2.40b 5.45ab 5.94bc
to colon for fermentation (A. Stalmach, Edwards, Wightman, & Crozier,
±
1.35a
2013), therefore, the colonic catabolism of blackberry polyphenols IN 13.0 ± nd 27.5 ± 11.1 29.9 ± 36.6 ±
needs to be further investigated. 0.65c 1.39d ± 1.23 g 2.38 g
The common organic acid and free sugars released during the 0.94f
OUT 6.02 ± nd 12.4 ± 27.4 154 ± 129 ±
digestion were also determined. As shown in Table 2, citric acid and
0.90e 0.83f ± 5.01d 1.32f
malic acid were the two most released during the whole digestion. The 1.08
malic acid exhibited the highest bioaccessibility (68.90%), while cd
ascorbic acid could not be detected after gastric digestion. The same BI (%) 31.6 – 31.0 68.9 83.8 77.9
phenomenon was found in Andean berry that the ascorbic acid may be nd = not detectable; BI = bioaccessibility;
degraded in alkaline conditions (Agudelo et al., 2018). Fructose and In: colon-available fractions; OUT: gastrointestinal-available fractions;
glucose determined during the digestion were all with high bio­ The undigested sample refers to the blackberry extracts extracted from ethanol
accessibility (near 80%), which may provide energy for the body’s (90%, v/v).
a
metabolism. However, no disaccharides and oligosaccharides could be Different letter in the same column means significant differences among the
detected, which was different from Andean berry digestion that some groups at p < 0.05; The results were expressed as mean ± standard deviation
oligosaccharides could be detected (Agudelo et al., 2018), indicating (SD);
that different berries exhibited different digestive properties.
blackberry powder.
3.3. Morphological characteristics
3.4. Antioxidant activity changes during digestion
SEM is a qualitive method to analyze the surface morphology of
materials, which is helpful to understand the digestive character of The antioxidant activities are linked to the phytochemicals present in
blackberry. As shown in Fig. S3A, the initial blackberry powder showed fruits. However, these phytochemicals may convert into other forms and
a smooth, compact blocky structure with a diameter of 117.41 μm. exert different bioactivities induced by digestive conditions, such as
Obviously, in vitro digestion changed the surface morphology and par­ temperature, pH and enzymes. As shown in Fig. S4, the DPPH radical
ticle size of blackberry. After oral digestion, the powders agglomerated scavenging activity of undigested samples (ethanol extracts), oral,
under chewing in mouth with particle size increased to 151.56 μm, and gastric, colon-available (IN) and gastrointestinal-available (OUT) frac­
the surface exhibited a rough, large loose fragment (Fig. S3B). However, tions were 324.95 ± 4.60, 210.42 ± 2.76, 277.98 ± 10.15, 81.06 ± 1.13
after gastric digestion, the surface showed a small fragment with particle and 79.78 ± 2.99 mol TE/g, DW. After simulated gastric digestion, the
size decreased to 100.61 μm (Fig. S3C). The apparent structure also had antioxidant activity was significantly (p < 0.05) improved compared
a little change after intestinal digestion that small granules appeared and with oral digestion, which was 1.32-fold higher than that of oral phase,
the particle size was just 80.30 μm (Fig. S3D). These results suggested while still lower than that of undigested samples (85.55%). After dialysis
that in vitro digestion had a great impact on morphological properties of for 12 h, the DPPH radical scavenging activity of gastrointestinal-

5
Z. Dou et al. Food Chemistry 370 (2022) 131001

available fraction was almost equal to that of colon-available fractions. of gastrointestinal-available fractions (221.23 ± 19.21 mg AE/g, DW)
In line with DPPH assay, the blackberry after gastric digestion also was far below than that of colon-available fractions (974.98 ± 9.66 mg
exhibited the highest ABTS radical scavenging activity, while that of AE/g, DW). A linear relationship was found between the α-glucosidase
gastrointestinal-available fractions was significantly lower than colon- inhibitory activity and total phenolics (r = 0,936, p = 0.02), while no
available fractions (p < 0.05). The antioxidant activity of digested significant correlation with carbohydrates (r = 0.412, p = 0.49). The
fractions was lower than undigested fractions was also observed in results are in accordance with the reported raspberries that the pheno­
digestion of Madeiran elderberry (Sambucus lanceolata) (Pinto et al., lics play a vital role in inhibiting α-glucosidase (Qin et al., 2018).
2017) and aronia juice (Du & Myracle, 2018), however, the digested To further investigate the inhibitory mechanism of phenolics on
Merlot (Vitis vinifera) grape exhibited a better antioxidant activity than α-glucosidase, the molecular docking (MD) analysis was applied. In this
undigested fractions (Corrêa et al., 2017), manifesting that different raw study, four major phenolics released from blackberry during the diges­
materials showed a different digestive properties. In addition, these re­ tion were selected to bind with α-glucosidase using AutoDock 4.2 soft­
sults showed that the antioxidant components of blackberry were ware. In Fig. S5, the results showed that four phenolics all inserted into
mainly released in gastric digestion. The results were similar with in the hydrophobic cavity of α-glucosidase (Fig. S5A-D). The binding en­
vitro digestion of hawthorns (Zheng et al., 2018) and maqui berry ergy for 2,4,6-trihydroxybenzoic acid, cyanidin-3-O-glucoside, gallic
(Aristotelia chilensis (Molina)Stuntz) (Lucas-Gonzalez et al., 2016), acid and caffeic acid were − 2.92, − 7.77, − 3.67 and − 4.58 Kcal/mol,
whereas the antioxidant components of whole grains, such as beans and respectively (Table S1). Among them, cyanidin-3-O-glucoside, with the
whole wheat, were mainly released in small intestine or utilized by gut lowest binding energy, suggested the highest affinity to bind the
microbiota (Papillo et al., 2014). α-glucosidase and four conventional hydrogen bonds were found be­
Correlations between the release of total phenolics and carbohy­ tween cyanidin-3-O-glucoside and residues Glu 411, Pro 312, Asp 242,
drates were analyzed using Pearson’s correlation coefficient analysis. Ser 240, and two carbon hydrogen interactions (Asp 242, Pro 312) as
The correlation coefficient results were as following: phenolics vs DPPH well as one Pi-Anion (Asp 307) and Pi-Alkyl (Arg 315) bond. Previous
radical scavenging activity (r = 0.996**, p = 0.00), phenolics vs ABTS study has indicated that the residue Glu 411 is one of the most important
radical scavenging activity (r = 0.927*, p = 0.02), carbohydrates vs active sites of amino acid on the surface of α-glucosidase (Hua et al.,
DPPH radical scavenging activity (r = 0.716, p = 0.17), carbohydrates 2018). In the present study, cyanidin-3-O-glucoside exhibited the
vs ABTS radical scavenging activity (r = 0.388, p = 0.52). It was found strongest inhibitory effect against α-glucosidase, followed by caffeic
that there were significant positive correlations between the total acid, which may be due to that they all bind with residue Glu 411 and
released phenolics and the antioxidant activity of blackberry, while no had a higher binding energy than the two others. The number of binding
significant correlation between carbohydrates and antioxidant activity. sites is another factor affects the inhibitory capacity against α-glucosi­
The results suggested that the released phenolics during the digestion of dase. Gallic acid, with the maximum number of binding sites, also
blackberry played an important role in their antioxidant activities. exhibited a good inhibitory against α-glucosidase and mainly interacted
with conventional hydrogen bonds (Gln 279, Glu 277, Arg 213, Asp 215,
3.5. α-Glucosidase inhibitory activity and its mechanism Asp 69) and Pi-Anion bonds (Asp 352, Arg 442). These results suggested
that the phenolics could tightly bind to the active cavity of α-glucosidase
α-Glucosidase located in the intestinal cells could degrade the and hindered the entrance of substrate to α-glucosidase leading to the
oligosaccharide into monosaccharide leading to the increase of blood decrease of enzyme activity.
glucose level (Chen, You, Abbasi, Fu, Liu, & Li, 2016). Therefore, it is an
effective method to control postprandial blood glucose level by inhib­ 3.6. pH changes during the fermentation
iting the activity of α-glucosidase. As shown in Fig. 2, the gastric
digestive fractions showed the strongest α-glucosidase inhibitory activ­ The pH shift of the fermented products could reflect the fermentation
ity (6326.24 ± 184.43 mg AE/g, DW), which was 34.30-fold higher than process. As shown in Fig. S6, after fermentation for 24 h, the pH value of
that of oral digestions (1459.34 ± 39.52 mg AE/g, DW), while lower blackberry group decreased from 5.87 (0 h) to 4.77 (24 h, ΔpH = 1.10),
than that of undigested samples (10104.34 ± 68.60 mg AE/g, DW), while the blank group decreased from 6.08 (0 h) to 5.04 (24 h, ΔpH =
similar results were observed in digestion of Rubus idaeus L. fruit that 1.04). The results demonstrated that the addition of blackberry had a
gastric digestion fractions exhibited best α-glucosidase inhibitory ac­ greater impact on the pH value of gut environment compared with blank
tivity (Qin et al., 2018). Moreover, the α-glucosidase inhibitory activity group. The pH value of blackberry group was significantly (p < 0.05)
lower than that of blank group at all fermented time points, but higher
than that of inulin group. The similar results were observed in fermen­
tation of polysaccharides from bee collected pollen of Chinese wolfberry
(Agudelo et al., 2018). The proper decrease of pH value in intestinal
tract can promote the reproduction of some beneficial bacteria and
inhibit harmful microorganisms (Dou, Chen, & Fu, 2019; Zhou et al.,
2018).

3.7. The production of SCFAs during the fermentation

As an important metabolite of intestinal flora, short-chain fatty acids

(SCFAs) play an important role in maintaining intestinal homeostasis
and body health. As shown in Table 3, compared to the blank group, the
fermentation of non-digestible fractions of blackberry significantly (p <
0.05) improved the production of SCFAs at each time point. After
fermentation for 24 h, the total content of SCFAs in blackberry group
Fig. 2. Inhibitory effects on α-glucosidase activity of undigested sample and was improved from 319.50 ± 5.21 to 2314.70 ± 17.75 μg/g, which was
digested fractions of blackberry during the in vitro digestion (oral, gastric and 1.13-fold higher than that of blank group (2047.83 ± 15.34 μg/g), while
intestine phase). In: colon-available fractions; OUT: gastrointestinal-available lower than that of inulin group (2544.47 ± 18.86 μg/g). The signifi­
fractions. The undigested sample refers to the blackberry extracts extracted cantly higher production of SCFAs in blackberry group was mainly due
from ethanol (90%, v/v). to the utilization of dietary fiber by gut microbiota as previously

6
Z. Dou et al. Food Chemistry 370 (2022) 131001

Table 3
The production of short-chain fatty acids (SCFAs) after in vitro fermentation of blackberry, water (Blank) and inulin.
Treatment Time (h) SCFA concentrations (μg/g) a

Acetic Propionic Isobutyric Butyric Isovaleric Valeric Total

Fecal samples 0 110 ± 3.24 g 106 ± 0.98f 4.85 ± 0.12e 86.2 ± 1.14 h 5.16 ± 0.14e 1.16 ± 0.02e 313 ± 5.51 g
Fecal samples + NFBF 0 112 ± 2.24 g 107 ± 1.44f 4.69 ± 0.08e 89.3 ± 1.68 g 4.94 ± 0.31e 1.02 ± 0.04f 320 ± 5.21 g
Fecal samples + Inulin 0 111 ± 2.68 g 106 ± 1.01f 4.71 ± 0.11e 87.2 ± 1.28gh 5.02 ± 0.21e 1.08 ± 0.06ef 315 ± 5.43 g
Blank 12 412 ± 4.31f 592 ± 2.97e 4.94 ± 0.08e 377 ± 2.90f 6.75 ± 0.14d 3.29 ± 0.15d 1395 ± 12.3f
24 556 ± 4.24d 845 ± 4.66c 5.30 ± 0.18d 629 ± 3.58c 7.61 ± 0.19c 4.98 ± 0.19b 2048 ± 15.3d
Blackberry 12 533 ± 3.98e 763 ± 3.92d 5.51 ± 0.11 cd 469 ± 3.12e 6.61 ± 0.18d 4.37 ± 0.11c 1781 ± 15.2e
24 645 ± 4.59c 981 ± 4.79b 6.34 ± 0.17b 669 ± 4.03b 8.35 ± 0.11b 5.14 ± 0.19b 2315 ± 17.8b
Inulin 12 661 ± 5.06b 840 ± 4.64c 5.65 ± 0.14c 599 ± 3.64d 6.86 ± 0.12d 5.22 ± 0.20b 2118 ± 16.6c
24 754 ± 5.51a 1033 ± 5.31a 7.11 ± 0.17a 734 ± 4.41a 9.31 ± 0.15a 6.04 ± 0.21a 2544 ± 18.9a

The results were expressed as mean ± standard deviation (SD).

a
Different letter in the same column means significant differences among the groups at p < 0.05;

Fig. 3. The fermentation behavior of each treatment group. (A) The numbers of OTUs analyzed by Venn diagram; (B) Observed species; (C) The principal component
analysis (PCA) of the overall gut microflora composition; (D) The Non-Metric Multi-Dimensional scaling (NMDS) analysis of the overall gut microflora composition;
(E) Diversity distribution of intestinal microorganisms at phylum level; (F) The ratio of Firmicutes to Bacteroidetes (F/B).

7
Z. Dou et al. Food Chemistry 370 (2022) 131001

reported in Moringa oleifera Lam. Leaves (Dou, Chen, & Fu, 2019a). Unobserved States (PICRUSt) according to 16S data. As shown in Fig. S7,
Obviously, acetic, propionic and butyric acid were the predominate fermentation of blackberry could significantly enhance the glycan
fermentation products and mainly contributed to the increase of total biosynthesis and metabolism, while lower the rate of metabolism of
SCFAs. After 24 h fermentation, the concentration of acetic acid was amino acid, energy and lipid (Fig. S7A), which were closely associate
increased from 112.31 ± 2.24 (0 h) to 645.26 ± 4.59 μg/g (24 h), which with chronic diseases, such as diabetes and obesity, and may paly vital
was 1.61-fold higher than that of blank group (555.94 ± 4.24 μg/g), roles in metabolic improvement. Additionally, blackberry supplement
showing significant difference with blank group. In addition, the content could obviously regulate the cellular processes, organismal systems and
of propionic and butyric acids were increased by 1.16-fold and 1.06-fold human diseases induced by alterations of gut microbiota (Fig. S7B-D).
in comparison with blank group. These results may also account for the For example, blackberry treatment could inhibit the cell motility,
pH change during fermentation. Obviously, there was no significant (p improve the transport and catabolism, intensify the digestive, endo­
> 0.05) change for isobutyric, isovaleric and valeric acids. Reported crine, immune and nervous system, and protect host against cancers,
studies has demonstrated that acetic acid can be used as an energy immune system diseases, infectious and neurodegenerative diseases.
source for intestinal microbiota and peripheral tissues (Zhou et al., These results suggested blackberry can be used as a functional fruit for
2018), propionic acid could lower serum cholesterol level and prevent host healthy. However, the exact signaling mechanism of blackberry
the obesity caused by diet (Zhou et al., 2018), and butyric acid could fruit in host peripheral tissues is still unclear, and more research in vivo
alleviate the diabetes and insulin resistance (Lin et al., 2012). about its regulatory mechanism is needed in the future.

3.8. Intestinal flora analysis 4. Conclusions

As shown in Fig. 3A, the addition of blackberry significantly (p < In this study, the bioaccessibility and bioactivity changes of major
0.05) improved the OTU numbers compared with the fermentation of active compounds present in blackberry during the in vitro digestion
inulin (positive) and water (blank) group, which was consistent with were monitored. Results showed that the phenolics mainly released in
observed species (Fig. 3B) that 1.59-fold, 1.10-fold higher than that of gastric phase while carbohydrates in small intestinal digestion. The
blank and inulin group, respectively. The alpha diversity of each treat­ bioaccessibility of phenolics and carbohydrates were relatively low, and
ment is shown in Table S2. The Chao 1 and ACE index could reflect the most of them remain in colon and available for intestinal microflora. In
species richness while Simpson and Shannon indices are commonly used addition, the gastrointestinal digestion obviously altered the
to estimate the species diversity in sample. In this study, after fermen­ morphology properties and particle size of blackberry fruit powder. The
tation for 24 h, the alpha diversity in blackberry group was obviously (p alterations of total phenolics may account for the changes of antioxidant
< 0.05) enhanced compared with blank and inulin group, suggesting and hypoglycemic activities during the whole digestion, especially
that fermentation of blackberry had a greater effect on gut microbiota cyanidin-3-O-glucoside with higher amount and bioaccessibility index.
than inulin. The fermentation of polyphenols from seabuckthorn berries Furthermore, the fermentation of non-digestible fractions of blackberry
juice also significantly enhanced alpha diversity of gut microbiota, affected the ecosystem of the intestinal tract by lowering the colonic pH,
which was consistent with the present study (Attri & Goel, 2018). enhancing the production of SCFAs and modulating of gut microbiota
The principal component analysis (PCA) and Non-Metric Multi- composition, especially reducing the ratio of Firmicutes to Bacter­
Dimensional scaling (NMDS) scores plot were applied to reveal a distinct oidetes. This study systematically investigated the digestion and
clustering of intestinal flora compositions from different groups (Fig. 3C- fermentation of blackberry. Generally, blackberry exhibited a promising
D). The results clearly showed that each group was obviously different candidate for application in functional foods.
from others, suggesting a statistically separation among the intestinal
flora of the different groups. The results of PCA analysis were shown in CRediT authorship contribution statement
Fig. 3C, and the first two axes could account for 79.74% variation in the
different groups. The different groups were separated from each other Zuman Dou: Conceptualization. Chun Chen: Writing - original
on PC 1 while close on PC 2, suggesting that the effect of different draft. Huang Qiang: Investigation. Xiong Fu: Supervision.
treatments on intestinal community structure account for 46.30%. The
blackberry and inulin group were isolated from each other in PCA and Declaration of Competing Interest
NMDS analysis, demonstrating that blackberry and inulin had different
impacts on gut microbiota. The authors declare that they have no known competing financial
Gut microbial composition of different groups at phylum level was interests or personal relationships that could have appeared to influence
further analyzed and the results were demonstrated in Fig. 3E-F and the work reported in this paper.
Table S3. Notably, all groups mainly comprised of Firmicutes, Bacter­
oidetes, Proteobacteria and Actinobacteria. Especially, the content of Acknowledgements
Firmicutes and Bacteroidetes was consistent with the fact that>90% of
bacteria in the colon belong to the phyla of Bacteroidetes and Firmicute Guangdong Basic and Applied Basic Research Foundation
(Dou, Chen, & Fu, 2019a; Zhou et al., 2018). After fermentation with (2019A1515011996), Financial and moral assistances supported by the
blackberry, the Firmicute level was decreased from 81.46% to 32.25%, National Natural Science Foundation of China (31972022), China
while the Bacteroidetes level increased from 6.18% to 37.14%. A Postdoctoral Science Foundation (2018 M643092), National Natural
significantly (p < 0.05) lower ratio of Firmicute to Bacteroidetes (F/B) Science Foundation of China (31972011), the Guangzhou Science and
was found with the addition of blackberry compared to water (blank) or Technology Program (201907010035), the Natural Science Foundation
inulin (positive) group (Fig. 3F). Increasing studies have shown that F/B of Guangdong Province (2019A1515011670), National Key R&D Pro­
is closely associated with obesity that high ratio of F/B was found in gram (2017YFD0400703), Guangzhou Science Technology and Inno­
obese individuals, and low F/B may lead to the decrease of energy vation Commission (201803050001), Science & Technology Planning
harvest, which is conducive to reducing the risk of human obesity Project of Nansha, Guangzhou (2016GJ001), and 111 Project (B17018)
(Turnbaugh et al., 2009; Turnbaugh et al., 2006). Therefore, these re­ to conduct the project are gratefully acknowledged.
sults indicate that blackberry is expected to become a functional anti-
obese fruit by adjusting the composition of bacterial flora. Appendix A. Supplementary data
Functional analysis of intestinal flora was also investigated with
Phylogenetic Investigation of Communities by Reconstruction of Supplementary data to this article can be found online at https://doi.

8
Z. Dou et al. Food Chemistry 370 (2022) 131001

org/10.1016/j.foodchem.2021.131001. Mechanisms. PLoS ONE, 7(4), e35240. https://doi.org/10.1371/journal.

pone.0035240.
Lucas-Gonzalez, R., Navarro-Coves, S., Pérez-Álvarez, J. A., Fernández-López, J.,
References Muñoz, L. A., & Viuda-Martos, M. (2016). Assessment of polyphenolic profile
stability and changes in the antioxidant potential of maqui berry (Aristotelia
Agudelo, C. D., Luzardo-Ocampo, I., Campos-Vega, R., Loarca-Piña, G., & Maldonado- chilensis (Molina) Stuntz) during in vitro gastrointestinal digestion. Industrial Crops
Celis, M. E. (2018). Bioaccessibility during in vitro digestion and antiproliferative and Products, 94, 774–782.
effect of bioactive compounds from Andean berry (Vaccinium meridionale Swartz) Martínez-Las Heras, R., Pinazo, A., Heredia, A., & Andrés, A. (2017). Evaluation studies
juice. Journal of Agriculture and Food Chemistry, 66(28), 7358–7366. of persimmon plant (Diospyros kaki) for physiological benefits and bioaccessibility
Attri, S., & Goel, G. (2018). Influence of polyphenol rich seabuckthorn berries juice on of antioxidants by in vitro simulated gastrointestinal digestion. Food Chemistry, 214,
release of polyphenols and colonic microbiota on exposure to simulated human 478–485.
digestion model. Food Research International, 111, 314–323. Martins, A., Barros, L., Carvalho, A. M., Santos-Buelga, C., Fernandes, I. P., Barreiro, F., &
Bao, T., Li, Y., Xie, J., Jia, Z., & Chen, W. (2019). Systematic evaluation of polyphenols Ferreira, I. C. F. R. (2014). Phenolic extracts of Rubus ulmifolius Schott flowers:
composition and antioxidant activity of mulberry cultivars subjected to Characterization, microencapsulation and incorporation into yogurts as
gastrointestinal digestion and gut microbiota fermentation. Journal of Functional nutraceutical sources. Food & Function, 5(6), 1091–1100.
Foods, 58, 338–349. Minekus, M., Alminger, M., Alvito, P., Ballance, S., Bohn, T., Bourlieu, C., … Brodkorb, A.
Brodkorb, A., Egger, L., Alminger, M., Alvito, P., Assunção, R., Ballance, S., … Recio, I. (2014). A standardised static in vitro digestion method suitable for food - an
(2019). INFOGEST static in vitro simulation of gastrointestinal food digestion. international consensus. Food & Function, 5(6), 1113–1124.
Nature Protocols, 14(4), 991–1014. Moraes, D. P., Lozano-Sánchez, J., Machado, M. L., Vizzotto, M., Lazzaretti, M., Leyva-
Caicedo-Lopez, L. H., Luzardo-Ocampo, I., Cuellar-Nuñez, M. L., Campos-Vega, R., Jimenez, F. J. J., … Barcia, M. T. (2020). Characterization of a new blackberry
Mendoza, S., & Loarca-Piña, G. (2019). Effect of the in vitro gastrointestinal cultivar BRS Xingu: Chemical composition, phenolic compounds, and antioxidant
digestion on free-phenolic compounds and mono/oligosaccharides from Moringa capacity in vitro and in vivo. Food Chemistry, 322, 126783. https://doi.org/10.1016/
oleifera leaves: Bioaccessibility, intestinal permeability and antioxidant capacity. j.foodchem.2020.126783.
Food Research International, 120, 631–642. Mosele, J. I., Macià, A., Romero, M.-P., Motilva, M.-J., & Rubió, L. (2015). Application of
Chen, C., & Fu, X. (2019). Spheroidization on Fructus Mori polysaccharides to enhance in vitro gastrointestinal digestion and colonic fermentation models to pomegranate
bioavailability and bioactivity by anti-solvent precipitation method. Food Chemistry, products (juice, pulp and peel extract) to study the stability and catabolism of
300, 125245. https://doi.org/10.1016/j.foodchem.2019.125245. phenolic compounds. Journal of Functional Foods, 14, 529–540.
Chen, C., Wang, P.-P., Huang, Q., You, L.-J., Liu, R. H., Zhao, M.-M., … Luo, Z.-G. (2019). Muzquiz, M., Burbano, C., Pedrosa, M. M., Folkman, W., & Gulewicz, K. (1999). Lupins as
A comparison study on polysaccharides extracted from Fructus Mori using different a potential source of raffinose family oligosaccharides: Preparative method for their
methods: Structural characterization and glucose entrapment. Food & Function, 10 isolation and purification. Industrial Crops and Products, 9(3), 183–188.
(6), 3684–3695. Papillo, V. A., Vitaglione, P., Graziani, G., Gokmen, V., & Fogliano, V. (2014). Release of
Chen, C., You, L.-J., Abbasi, A. M., Fu, X., & Liu, R. H. (2015). Optimization for antioxidant capacity from five plant foods during a multistep enzymatic digestion
ultrasound extraction of polysaccharides from mulberry fruits with antioxidant and protocol. Journal of Agriculture and Food Chemistry, 62(18), 4119–4126.
hyperglycemic activity in vitro. Carbohydrate Polymers, 130, 122–132. Pinto, J., Spinola, V., Llorent-Martinez, E. J., Fernandez-de Cordova, M. L., Molina-
Chen, C., You, L.-J., Abbasi, A. M., Fu, X., Liu, R. H., & Li, C. (2016). Characterization of Garcia, L., & Castilho, P. C. (2017). Polyphenolic profile and antioxidant activities of
polysaccharide fractions in mulberry fruit and assessment of their antioxidant and Madeiran elderberry (Sambucus lanceolata) as affected by simulated in vitro
hypoglycemic activities in vitro. Food & Function, 7(1), 530–539. digestion. Food Research International, 100(Pt 3), 404–410.
Chen, C., Zhang, B., Huang, Q., Fu, X., & Liu, R. H. (2017). Microwave-assisted extraction Prior, R. L., Wu, X., Gu, L., Hager, T. J., Hager, A., & Howard, L. R. (2008). Whole berries
of polysaccharides from Moringa oleifera Lam. leaves: Characterization and versus berry anthocyanins: Interactions with dietary fat levels in the C57BL/6J
hypoglycemic activity. Industrial Crops and Products, 100, 1–11. mouse model of obesity. Journal of Agricultural and Food Chemistry, 56(3), 647–653.
Chen, G., Xie, M., Wan, P., Chen, D., Ye, H., Chen, L., … Liu, Z. (2018). Digestion under Qin, Y., Wang, L.u., Liu, Y., Zhang, Q., Li, Y., & Wu, Z. (2018). Release of phenolics
saliva, simulated gastric and small intestinal conditions and fermentation in vitro by compounds from Rubus idaeus L. dried fruits and seeds during simulated in vitro
human intestinal microbiota of polysaccharides from Fuzhuan brick tea. Food digestion and their bio-activities. Journal of Functional Foods, 46, 57–65.
Chemistry, 244, 331–339. Schulz, M., Seraglio, S. K. T., Della Betta, F., Nehring, P., Valese, A. C., Daguer, H., …
Corrêa, R. C. G., Haminiuk, C. W. I., Barros, L., Dias, M. I., Calhelha, R. C., Kato, C. G., … Fett, R. (2019). Blackberry (Rubus ulmifolius Schott): Chemical composition,
Ferreira, I. C. F. R. (2017). Stability and biological activity of Merlot (Vitis vinifera) phenolic compounds and antioxidant capacity in two edible stages. Food Research,
grape pomace phytochemicals after simulated in vitro gastrointestinal digestion and 122, 627–634.
colonic fermentation. Journal of Functional Foods, 36, 410–417. Singleton, V. L., & Rossi, J. A. (1965). Colorimetry of Total Phenolics with
Dou, Z., Chen, C., & Fu, X. (2019a). Bioaccessibility, antioxidant activity and modulation Phosphomolybdic-Phosphotungstic Acid Reagents. American Journal of Enology and
effect on gut microbiota of bioactive compounds from Moringa oleifera Lam. leaves Viticulture, 16(3), 144.
during digestion and fermentation in vitro. Food Funct. Stalmach, A., Edwards, C. A., Wightman, J. D., & Crozier, A. (2012). Gastrointestinal
Dou, Z., Chen, C., & Fu, X. (2019b). The effect of ultrasound irradiation on the stability and bioavailability of (poly)phenolic compounds following ingestion of
physicochemical properties and α-glucosidase inhibitory effect of blackberry fruit Concord grape juice by humans. Molecular Nutrition & Food Research, 56(3),
polysaccharide. Food Hydrocolloids, 96, 568–576. 497–509.
Du, X., & Myracle, A. D. (2018). Fermentation alters the bioaccessible phenolic Stalmach, A., Edwards, C. A., Wightman, J. D., & Crozier, A. (2013). Colonic catabolism
compounds and increases the alpha-glucosidase inhibitory effects of aronia juice in a of dietary phenolic and polyphenolic compounds from Concord grape juice. Food &
dairy matrix following in-vitro digestion. Food & Function., 9(5), 2998–3007. Function, 4(1), 52–62.
DuBois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., & Smith, F. (1956). Colorimetric Stalmach, A., Williamson, G., & Crozier, A. (2014). Impact of dose on the bioavailability
Method for Determination of Sugars and Related Substances. Analytical Chemistry, 28 of coffee chlorogenic acids in humans. Food & Function, 5(8), 1727–1737.
(3), 350–356. Turnbaugh, P. J., Hamady, M., Yatsunenko, T., Cantarel, B. L., Duncan, A., Ley, R. E., …
Gowd, V., Bao, T., Wang, L., Huang, Y., Chen, S., Zheng, X., … Chen, W. (2018). Gordon, J. I. (2009). A core gut microbiome in obese and lean twins. Nature, 457
Antioxidant and antidiabetic activity of blackberry after gastrointestinal digestion (7228), 480–484.
and human gut microbiota fermentation. Food Chemistry, 269, 618–627. Turnbaugh, P. J., Ley, R. E., Mahowald, M. A., Magrini, V., Mardis, E. R., & Gordon, J. I.
Gullon, B., Pintado, M. E., Barber, X., Fernández-López, J., Pérez-Álvarez, J. A., & Viuda- (2006). An obesity-associated gut microbiome with increased capacity for energy
Martos, M. (2015). Bioaccessibility, changes in the antioxidant potential and colonic harvest. Nature, 444(7122), 1027–1031.
fermentation of date pits and apple bagasse flours obtained from co-products during Wada, L., & Ou, B. (2002). Antioxidant activity and phenolic content of oregon
simulated in vitro gastrointestinal digestion. Food Research International, 78, caneberries. Journal of Agricultural and Food Chemistry, 50(12), 3495–3500.
169–176. Wang, L., Li, C., Huang, Q., & Fu, X. (2019). Biofunctionalization of selenium
Hua, F., Zhou, P., Wu, H.-Y., Chu, G.-X., Xie, Z.-W., & Bao, G.-H. (2018). Inhibition of nanoparticle with a polysaccharide from Rosa roxburghii fruit and its protective
flavonoid glycosides from Lu’an GuaPian tea on α-glucosidase and α-amylase: effect against H2O2-induced apoptosis in INS-1 cells. Food & Function.
Molecular docking and interaction mechanism. Food & Function. Wang, P.-P., Huang, Q., Chen, C., You, L.-J., Liu, R. H., Luo, Z.-G., … Fu, X. (2019). The
Ivanovic, J., Tadic, V., Dimitrijevic, S., Stamenic, M., Petrovic, S., & Zizovic, I. (2014). chemical structure and biological activities of a novel polysaccharide obtained from
Antioxidant properties of the anthocyanin-containing ultrasonic extract from Fructus Mori and its zinc derivative. Journal of Functional Foods, 54, 64–73.
blackberry cultivar “Čačanska Bestrna”. Industrial Crops and Products, 53, 274–281. Zeppa, G., Conterno, L., & Gerbi, V. (2001). Determination of organic acids, sugars,
Khan, I., Huang, G., Li, X., Leong, W., Xia, W., & Hsiao, W. L. W. (2018). Mushroom diacetyl, and acetoin in cheese by high-performance liquid chromatography. Journal
polysaccharides from Ganoderma lucidum and Poria cocos reveal prebiotic of Agricultural and Food Chemistry, 49(6), 2722–2726.
functions. Journal of Functional Foods, 41, 191–201. Zhang, C., Wu, W., Xin, X., Li, X., & Liu, D. (2019). Extract of ice plant
Langille, M. G. I., Zaneveld, J., Caporaso, J. G., McDonald, D., Knights, D., Reyes, J. A., … (Mesembryanthemum crystallinum) ameliorates hyperglycemia and modulates the
Huttenhower, C. (2013). Predictive functional profiling of microbial communities gut microbiota composition in type 2 diabetic Goto-Kakizaki rats. Food & Function.,
using 16S rRNA marker gene sequences. Nature Biotechnology, 31(9), 814–821. 10(6), 3252–3261.
Li, F., Zhang, B., Chen, G.u., & Fu, X. (2017). The novel contributors of anti-diabetic Zhang, J.-Q., Li, C., Huang, Q., You, L.-J., Chen, C., Fu, X., & Liu, R. H. (2019).
potential in mulberry polyphenols revealed by UHPLC-HR-ESI-TOF-MS/MS. Food Comparative study on the physicochemical properties and bioactivities of
Research International, 100, 873–884. polysaccharide fractions extracted from Fructus Mori at different temperatures. Food
Lin, H. V., Frassetto, A., Kowalik Jr, E. J., Nawrocki, A. R., Lu, M. M., Kosinski, J. R., … & Function, 10(1), 410–421.
Brennan, L. (2012). Butyrate and Propionate Protect against Diet-Induced Obesity
and Regulate Gut Hormones via Free Fatty Acid Receptor 3-Independent

9
Z. Dou et al. Food Chemistry 370 (2022) 131001

Zhang, J., Chen, C., & Fu, X. (2019). Fructus mori L. polysaccharide-iron chelates formed Zhou, W., Yan, Y., Mi, J., Zhang, H., Lu, L.u., Luo, Q., … Cao, Y. (2018). Simulated
by self-embedding with iron(iii) as the core exhibit good antioxidant activity. Food & Digestion and Fermentation in Vitro by Human Gut Microbiota of Polysaccharides
Function, 10(6), 3150–3160. from Bee Collected Pollen of Chinese Wolfberry. Journal of Agriculture and Food
Zheng, G., Deng, J., Wen, L., You, L., Zhao, Z., & Zhou, L. (2018). Release of phenolic Chemistry, 66(4), 898–907.
compounds and antioxidant capacity of Chinese hawthorn “ Crataegus pinnatifida ”
during in vitro digestion. Journal of Functional Foods, 40, 76–85.

10