Comparisons in phytochemical components and in vitro digestion properties of corresponding peels, flesh and seeds separated from two blueberry cultivars

Abstract

Highbush blueberries (HB) and rabbiteye blueberries (RB) were separated into peels, flesh, and seeds to assess the compositions of nutriment, anthocyanins, soluble sugars and fatty acids, and the in vitro digesting abilities. Total phenolics contents (TPC) of 51–56 mg GAE/g DW were found in blueberry peels. Compared with HB peels, RB peels showed much higher TPC, but only contained 35 phenolics and lacked peonidin-3-O-rutinoside. Glucose, fructose, and sucrose were all present in HB and RB, but RB flesh had a higher acid-sugar ratio. Unsaturated fatty acid concentrations in HB and RB seeds were comparable (26.65 and 26.43 mg/g, respectively). However, HB seeds have 35 fatty acids, but RB seeds lacked cis-4,7,10,13,16,19-docosahexaenoic acid and cis-10-pentadecenoic acid. The in vitro digestion test showed that the whole fruit/peels/flesh of RB had a higher recovery and bioavailability index of phenolics and anthocyanins. Therefore, the reuse of blueberry pomace needs to be emphasized.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10068-023-01326-w.

Introduction

Blueberries, over 200 species, are cultivated widely in America, Canada, China, etc. (Riihinen et al., 2008; Li et al., 2017). Of which, highbush blueberries (HB, Vaccinium corymbosum L.) and rabbiteye blueberries (RB, Vaccinium ashei Reade) are known as the commercially important blueberry types due to their large fruits and early ripening (Lee et al., 2013; Takeda et al., 2017). Generally, different blueberry cultivars differ in terms of fruit sizes, colors, total phenolics (TP), total anthocyanins (TA), and so on. For example, 9.00–19.00% of soluble solids and 0.13–0.85 g/100 g FW (flesh weight) of TA in blueberries were reported (Wang et al., 2012; Lin et al., 2020). Additionally, according to the database of Phenol-Explorer (http://phenol-explorer.eu/), the average contents of TP in HB and RB are 2.23 mg GAE/g FW and 5.50 mg GAE/g FW, respectively. Nonetheless, little work has been done on the phytochemicals in HB and RB in a systematic manner.

In 2021, about 1.8 million tons of blueberries have been produced, and the global production of blueberries is expected to reach about 2.4 million tons by 2025 (http://nyncj.huizhou.gov.cn/zwzc/xwzx/gjyw/content/post_4798201.html). More than half of the blueberries are processed into products such as beverages, juice, jam, and so on (Liu et al., 2021). Industrial processing of blueberries is usually done with the whole fruit, and generates around 20–30% pomace of the weight of the fruits (Liu et al., 2021; Yang et al., 2022). The pomace, blueberry byproducts, are mainly blueberry peels and seeds which are rich in bioactive components and have been expected to be used as additives, to extract bioactive substance such as anthocyanin, or to produce fermented products in food industries (Liu et al., 2021). However, most of the blueberry byproducts are disposed and rarely used in practice. Identification of the bioactive compounds from the peels and seeds of blueberries could benefit producers and processors economically, provide an opportunity for the development of food products that promote health, and reduce ecological issues due to blueberry pomace disposal. However, the bioactive compounds of the peels and seeds, as well as the flesh of HB and RB, have not been well investigated.

The in vitro digestion model has been increasingly used to develop a mechanistic understanding of the effects of food compositions on human health because it is easy to undertake, inexpensive, and justifiable on ethical grounds (Brodkorb et al., 2019; Jiao et al., 2017). Among the models, the three-stage digestion model consisting of oral, gastric, and intestinal phases and using gastric lipase has been widely accepted to assess the endpoints resulting from the digestion of foods, namely bioaccessibility, which represents the maximum release of a substance from its matrix in the gastrointestinal tract and further becomes available for intestinal absorption (Brodkorb et al., 2019). A more realistic simulation process has been proposed to provide enough information on the beneficial effects or risks in the consumed diet, in which a dialysis membrane is used during or after the simulated intestinal digestion stage, and the bioavailability can be assessed based on the investigation of the substances released from the food matrix and crossed the membrane (Herbello-Hermelo et al., 2018). However, the method of in vitro digestion and dialysis has not been used to evaluate the bioavailability of different individual part of blueberries.

Even though the blueberry anthocyanins have received a lot of attention (Herrera-Balandrano et al., 2021), the phytochemicals and bioavailabilities of peels, flesh, and seeds separated from blueberry fruits were not systematically investigated and compared. The objectives of this study were to: (1) compare the proximate compositions of the peels, flesh, and seeds of two practically important blueberry cultivars, HB and RB, (2) identify and quantify the anthocyanin compositions of the peels, the soluble sugars of the flesh, and the fatty acid compositions of the seeds of HB and RB, and (3) evaluate the gastrointestinal digestion properties of the blueberry whole fruits, peels, and flesh through an in vitro digestion and dialysis approach. This work would help to acquire the nutritional components and achieve the accurate processing of different parts of blueberries, and to acknowledge the added value of blueberry byproducts.

Materials and methods

Materials and chemicals

Fresh HB (duke) from the same lot were obtained in bulk from a local supermarket. Fresh RB (Britewell) were provided by Anhui Huiwang Food Co., Ltd. (Hefei, China). Gallic acid (GA), cyanidin-3-O-glucoside (CG), 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox), 2,2-diphenyl-1-picrylhydrazyl (DPPH), 2,2′-azino-bis (3-ethylbenzothiazoline-6-sulphonic acid) (ABTS), and 12 standards of soluble sugars (Table 3) were obtained from Solarbio Science & Technology Co., Ltd. (Beijing, China). The 37-component FAMEs (fatty acid methyl esters) standard mixture, ranging from C4:0 to C22:6, was bought from Sigma-Aldrich (St. Louis, MO, USA). Alpha-amylase (40–60 U/mg), pepsin (3000 U/g), pancreatin (4000 U/g), and bile salts were supplied by Yuanye Biochemical Reagent Co., Ltd. (Shanghai, China). All of the other chemicals used were of analytical grade and purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China).

Table 3.

Soluble sugars in FHB and FRB

No.	Soluble sugars	Retention time (min)	Molecular formula	FHB (mg/g DW)	FRB (mg/g DW)
1	Glucose	14.6837	C6H12O6	305.1047	328.0004
2	Fructose	17.1753	C6H12O6	373.4841	407.9016
3	Sucrose	18.142	C12H22O11	46.0466a	39.5501b
4	Trehalose	3.4503	C12H22O11	ND	ND
5	Fucose	5.4003	C6H12O5	ND	ND
6	Rhamnose	9.892	C6H14O6	ND	ND
7	Arabinose	11.0003	C5H10O5	ND	ND
8	Galactose	13.8503	C6H12O6	ND	ND
9	Lactose	21.7587	C12H22O11	ND	ND
10	Raffinose	25.8837	C18H32O16	ND	ND
11	Stachyose	27.5087	C24H44O22	ND	ND
12	Maltose	29.7253	C12H22O11	ND	ND
Total soluble sugars				718.6354	775.4521

FHB and FRB peels of highbush blueberry and rabbiteye blueberry, respectively; DW dry weight, ND not detected. Means with different superscript in the same row are significantly different (p < 0.05)

Separation of peels, flesh and seeds from blueberry fruits

The peels, flesh, and seeds of fresh HB and RB, were separated after quick-freezing, peeling, drying, and sieving. After quickly frozen at – 38 ℃ for 6 h, the blueberry fruits were removed, put into the tap water for 10–12 s to soften the peels while the flesh remained frozen, and then immediately peeled by hand. The peels and the seed-containing flesh were separately collected and freeze-dried in an FD-1B-50 freeze-dryer (Boyikang Experimental Instrument Co., Ltd., Beijing, China) at − 38 ℃ for 24 h. After that, the dried peels of HB and RB, denoted PHB and PRB, respectively, were smashed in a grinder, while the dried seed-containing flesh was broken slightly and screened using a sieve with a 0.297 mm opening to separate the seeds (SHB and SRB, respectively) from the flesh (FHB and FRB). The separated peels, flesh, and seeds were separately stored at − 40 ℃ for further use.

Color measurement of peels, flesh and seeds

The surface color of the dried powder of peel, flesh, and seed from HB and RB was measured with a CR-400 spectrophotometer (Konica Minolta, Osaka, Japan) (Zielinska & Michalska, 2016) and the L*, a*, b*, and ΔE* values were recorded.

Proximate compositions analysis of peels, flesh and seeds

The contents of moisture, proteins, fat, ash, and crude fiber of the peels, flesh, and seeds from HB and RB were analyzed according to the method of Palmeira et al. (2019). The pectin content was determined through a colorimetric method reported by Deng et al. (2019) and the results were expressed as g of galacturonic acid equivalent (GAE) per 100 g. The total sugar content and total acidity were measured through the phenol sulfuric acid method and the reference titration method, respectively. The contents of total phenols (TP) and total anthocyanins (TA) were assessed with the Folin-Ciocalteu method and the pH differential method, respectively, using GA and CG, respectively, for calibration (Souza et al., 2014).

Anthocyanin compositions analysis of peels

The dried PHB or PRB powder (0.50 g) was dissolved in 1 mL of the liquor consisting of methanol, water, and formic acid (70:30:1, v/v/v), and analyzed using a UPLC-Orbitrap-MS system (Thermo Fisher Scientific (Shanghai) Instruments Co., Ltd., Shanghai, China) connected with a Waters Accuracy UPLC HSS T3 column (2.1 mm × 100 mm; Thermo Fisher Scientific (Shanghai) Instruments Co., Ltd., Shanghai, China). Mobile phase A was the ultrapure water containing 0.1% fomic acid, whereas mobile phase B was 100% acetonitrile. The elution gradient was as follows: 0–2 min, 5% B; 2–15 min, 5–30% B; 15–20 min, 30–95% B; 20–26 min, 95–5% B. The injection volume was 20 μL and the running temperature was 25 ℃. The detection was conducted at 520 nm at a flow rate of 0.30 mL/min. HRMS data were recorded on a QE xactive hybrid Q-Orbitrap mass spectrometer (Thermo Fisher Scientific (Shanghai) Instruments Co., Ltd., Shanghai, China) equipped with a heated ESI source utilizing the FSMS2 MS acquisition methods. The ESI source parameters were set as follows: spray voltage, − 2.8 kV/3.0 kV; sheath gas pressure, 40 arb; aux gas pressure, 10 arb; sweep gas pressure, 0 arb; capillary temperature, 320 ℃; and aux gas heater temperature, 350 ℃. ESI was performed with the scan range between 100 and 1200 m/z. The anthocyanins were identified on the basis of the retention time, m/z, and exact fragment masses compared with the existing database (Barnes et al., 2009).

Soluble sugars evaluation of flesh

The contents and compositions of the soluble sugars in the FHB or FRB powder were evaluated according to the method of Lee et al. (2013). Briefly, 0.10 g of FHB or FRB powder was extracted thrice with a 0.80 mL 80% ethanol (v/v) solution at 50 ℃ for 2 h. The slurry after each extraction was diluted with 0.80 mL H₂O and centrifuged at 10,000 rpm for 5 min. Each supernatant was collected, pooled in a centrifuge tube, and analyzed using a Thermo ICS5000 ion chromatographic system (ICS5000; Thermo Fisher Scientific, San Jose, CA, USA) with a prewashed CarboPac PA-10 anion-exchange column (250 × 4.0 mm; Thermo Fisher Scientific, San Jose, CA, USA) and a pulsed amperometric detector.

Fatty acid composition measurement of seeds

The fatty acid compositions of SHB and SRB powder were measured through GC-MS analysis according to the method reported by Tang et al. (2021) with some modifications. SHB or SRB powder (0.10 mg) was put into 4 mL chloroform, mixed with a vortex for 30 s, centrifuged at 3500 rpm for 15 min at room temperature, let stand for 10 min, and then the lower liquid layer was removed and extracted twice with 2 mL dichloromethane at 40 ℃ for 1 h. The slurry after each extraction was centrifuged at 5000 rpm for 15 min. The lower layer of liquid was collected, pooled in a centrifuge tube, blow-dried with nitrogen, esterified to form fatty acid methyl esters and then injected into a gas chromatography system (Agilent 6890; Agilent Technologies Inc., Santa Clara, CA, USA) with a capillary column CP-Sil 88 (100 m × 0.25 mm; Agilent Technologies Inc., Santa Clara, CA, USA) for quantifying fatty acid methyl esters. The sample intake quantity was 1 μL, the shunt ratio was 10:1, and the gas load was high purity nitrogen with a flow rate of 1.0 mL/min. The initial temperature of the column temperature chamber was 100 ℃ for 5 min, and the temperature was heated to 240 ℃ at 4 ℃/min for 15 min. The quadrupole mass spectrometry detection system (Agilent 5975; Agilent Technologies, San Jose, CA, USA) was used. The inlet temperature was 260 ℃, and the four-stage rod temperature was 150 ℃. The full SCAN mode and 30–550 m/z of SCAN range were set. The fatty acids were identified by comparing the retention times of 37-component FAMEs standards with the retention times of samples.

Bioavailability of blueberry whole fruits, peels and flesh

The bioavailability of the whole fruits, peels, and flesh of HB and RB was performed based on the in vitro simulated digestion and dialysis test, and the oral, gastric, and intestinal stock solutions were prepared before the simulated digestion test according to Herbello-Hermelo et al. (2018). Briefly, each sample (0.50 g) was dissolved in 6 mL of ultrapure water and digested sequentially at 37 ℃ using a water bath and performing pH-test adjustment as set: oral phase (addition of oral stock solution and salivary amylase, incubating at pH 7 for 2 min); gastric phase (addition of gastric stock solutions, pepsin and gastric lipase, mixing at pH 3 for 2 h); intestinal phase (addition of an intestinal stock solution mixture of 10 mM bile salt, mixing at pH 7 for 2 h). For the simulated intestinal phase, dialysis bags (pH = 7) were added to the mixture at the beginning of digestion (Cao et al., 2021). The supernatant was collected and heated to 100 ℃ for 5 min, then stored at − 20 ℃ to evaluate the contents of TP and TA, and the DPPH and ABTS scavenging activities (Souza et al., 2014; Zhou et al., 2020). The in vitro simulated digestion indices of release and bioavailability index of anthocyanins and phenolics were determined according to the method of Cao et al. (2021) and expressed as the following formula: RI (recovery index, %) = (TP/TA in digesta / TP/TA in undigested samples) × 100, BI (bioavailability index, %) = (TP/TA in the dialysis bag remaining / TP/TA in undigested samples) × 100.

Statistical analysis

All the tests were performed in triplicate. The results were expressed as mean values and standard error. Statistical analysis and multiple comparison between the means were analyzed by one way analysis of variance and LSD test at 5% level of significance using SPSS 17.0 software.

Results and discussion

Moisture content and color of peels, flesh and seeds from HB and RB

The percentages of the peels, flesh, and seeds in the weight of HB were 27.00%, 61.16% and 11.83%, respectively, while those of RB were 28.05%, 64.19% and 7.56%, respectively (Fig. 1A). The peels and seeds of blueberries represent nearly 40% of the whole fruit weight. Considering the blueberry industry, the biomass of the by-products is so large that the reevaluation could not be ignored.

Fig.1.

Percentage (A) and visual appearance (B) of peels, flesh, and seeds from HB and RB. HB highbush blueberry, RB rabbiteye blueberry, PHB and PRB peels of HB and RB, respectively, FHB and FRB flesh of HB and RB, respectively, SHB and SRB seeds of HB and RB, respectively.

There was no significant difference (p > 0.05) in the moisture content among PHB, PRB, FHB and FRB, which ranged from 83 to 86% and was comparable to the range (80.70–86.80%) of the whole blueberries reported by the USDA database (https://fdc.nal.usda.gov/ndb/), however, the moisture content of SHB (12.80 g/100 g FW) was much lower (p < 0.05) than that of SRB (17.82 g/100 g FW) (Table 1), which may be related with the smaller size of SHB.

Table 1.

Moisture content, color, and proximate compositions of peels, flesh, and seeds from HB and RB

Items	PHB	PRB	FHB	FRB	SHB	SRB
Moisture (g/100 g FW)	84.67 ± 1.08a	83.22 ± 1.26a	86.02 ± 0.98a	85.96 ± 1.29a	12.80 ± 0.17c	17.82 ± 0.28b
Color
L*	40.01 ± 0.21d	41.57 ± 0.64d	58.54 ± 0.49b	69.93 ± 0.38a	52.00 ± 0.58c	53.62 ± 0.46c
a*	3.99 ± 0.09c	3.93 ± 0.05c	12.96 ± 0.23b	19.16 ± 0.04a	13.63 ± 0.12b	13.64 ± 0.04b
b*	-0.06 ± 0.02e	0.03 ± 0.00d	13.95 ± 0.16b	6.35 ± 0.08c	16.04 ± 0.25a	16.58 ± 0.44a
∆E*	54.33 ± 0.21a	52.77 ± 0.67a	38.98 ± 0.42c	30.92 ± 0.12d	45.69 ± 0.53b	44.36 ± 0.22b
Proximate compositions (g/100 g DW)
Protein	0.81 ± 0.02b	0.84 ± 0.03b	0.42 ± 0.02c	0.44 ± 0.02c	7.39 ± 0.69a	8.13 ± 0.22a
Fat	2.45 ± 0.01b	2.59 ± 0.04b	1.24 ± 0.18c	1.00 ± 0.09d	24.91 ± 0.80a	26.90 ± 0.42a
Ash	1.54 ± 0.09b	1.77 ± 0.08a	1.49 ± 0.01b	0.98 ± 0.07d	1.31 ± 0.02c	0.92 ± 0.02d
Crude fiber	4.79 ± 0.02b	4.95 ± 0.42b	3.65 ± 0.03c	3.60 ± 0.03c	21.70 ± 1.06a	22.19 ± 0.40a
Pectin	2.90 ± 0.15b	3.76 ± 0.22a	1.16 ± 0.07d	1.66 ± 0.10c	NP	NP
Total sugar	50.06 ± 3.04a	44.04 ± 1.92b	41.71 ± 3.42c	38.98 ± 0.79d	NP	NP
Total acidity	2.81 ± 0.07b	2.94 ± 0.03b	4.38 ± 0.06a	2.46 ± 0.04c	2.24 ± 0.02c	1.59 ± 0.01d
Acid-sugar ratio (g/g)	17.79 ± 0.77a	14.24 ± 0.58c	9.52 ± 0.43d	15.86 ± 0.59b	NP	NP
TP (mg GAE/g DW)	51.69 ± 0.82b	55.39 ± 1.59a	7.26 ± 0.53d	8.59 ± 0.24c	ND	ND
TA (mg CGE/g DW)	26.74 ± 0.44a	28.05 ± 0.57a	1.01 ± 0.02b	0.89 ± 0.06c	ND	ND

HB highbush blueberry, RB rabbiteye blueberry, PHB and PRB peels of HB and RB, respectively; FHB and FRB the flesh of HB and RB, respectively; SHB and SRB seeds of HB and RB, respectively; FW fresh weight, DW dry weight, TP total phenolics, TA total anthocyanins, GAE gallic acid equivalent, CGE cyanidin-3-O-glucoside equivalent, NP not performed, ND not detected. Data are expressed as mean ± SD. Means with different superscript in the same row are significantly different (p < 0.05)

The surface color is an important criterion of blueberry variety and maturity, and the ripe blueberries have higher a* values and lower L* and b* values (Cesa et al., 2017; Lin et al., 2020). The values of L*, a*, b*, and ∆E* between PHB and PRB, as well as between SHB and SRB, were similar (p > 0.05), except for the b* value of PHB (–0.06) which had a much lower b* value (p < 0.05) than that of PRB (0.03) (Table 1), indicating that PHB is a bluish color. Additionally, FHB had lower L* and a* values but higher b* and ∆E* values than FRB, indicating that FHB is reddish. The colorimeter readings (Table 1) were consistent with the surface color of PHB, PRB, FHB, FRB, SHB, and SRB (Fig. 1B). The difference in the color of the peels and flesh of HB and RB may be related to the difference in their varieties.

Proximate compositions of peels, flesh and seeds from HB and RB

The main constituents in both the peels and the flesh of HB and RB were total sugar, showing 38.98–58.06 g/100 g DW, however, in the seeds, were fats, showing both approximately 25 g/100 g DW (dry weight) (Table 1). PHB showed higher total sugar content (50.06 g/100 g DW) and acid-sugar ratio (17.79) but lower contents of ash (1.54 g/100 g DW), pectin (2.90 g/100 g DW) and TP (51.69 mg GAE/g DW) than PRB (p < 0.05). The difference in the TP and TA contents betweeen PHB and PRB may lead to the difference in their color (Fig. 1). FRB had similar contents (p > 0.05) of protein and crude fiber, higher contents (p < 0.05) of pectin (1.66 g/100 g DW) and TP (8.59 mg GAE/g DW), and lower contents (p < 0.05) of other proximate compounds, compared to FHB. Except for the ash content and total acidity, no differences in the contents of the other proximate compounds were found between SHB and SRB. In general, as compared with HB, RB had larger fruits in sizes, tasted more delicious with slight sweetness and sour probably due to the suitable acid-sugar ratio (Lin et al., 2020), and had higher contents of pectin and TP in the peels and flesh (Table 1). Thus, RB would be more favored by consumers and more suitable for industrial applications. Blueberry peels would be more suitable as a natural toner, flavor additive, and raw material for functional food and medicine. Blueberry flesh would be better used for processed drinks, fruit wines, and jam. Blueberry seeds could be sold as extract capsules and seed oils (Liu et al., 2021; Zhou et al., 2020, 2022).

Anthocyanin compositions of PHB and PRB

The anthocyanins present in blueberries are various and related to the varieties, genetics, ripeness, processing and growth environments, and about 90% of them are concentrated in blueberry peels (Herrera-Balandrano et al., 2021). In the present study, 36 and 35 anthocyanins, in the form of aglycone or arabinoside, glucoside, galactoside, rutinoside, diglucoside, and sophoroside, were tentatively identified from PHB and PRB, respectively (Table 2 and Supplementary Fig. 1). And Mv-3-O-gal/glu (25.13% in PHB and 24.63% in PRB), Mv-3-O-ara (16.08% in PHB and 15.31% in PRB), and Pn-3-O-gal/glu (15.94% in PHB and 15.25% in PRB) were the top three in both PHB and PRB. Malvidin and malvidin glycosides were the most dominant anthocyanins (> 40%) in PHB and PRB, which was in agreement with the results reported by Nicoué et al. (2007) and Chai et al. (2021). As compared with PHB, PRB contained higher percentages (p < 0.05) of Dp, Dp-3-O-rha, and Pn-3-O-(6″-acetyl)-gal/glu, lower percentages (p < 0.05) of Mv-3-O-(6″-acetyl)-gal/glu, Pg-3-O-gal/glu, and Pn-3-O-ara, and similar percentages (p > 0.05) of the other anthocyanins; meanwhile, Pn-3-O-rut was found in PHB while not in PRB. That a variety of anthocyanins in blueberry peels were identified may be ascribed to the different research object, varieties and identification methods Chai et al. (2021). Anthocyanins have been applied for the prevention of obesity, cardiovascular disease, and cancer therapy (Lin et al., 2017). Therefore, the blueberry peels are a good source of bioactive molecules in food and medicine due to the abundant anthocyanins.

Table 2.

Anthocyanin compositions of PHB and PRB

No.	Anthocyanins	Retention time (min)		[M-H]−(m/z)	Fragment masses	Percentage (%)
		PHB	PRB			PHB	PRB
1	Cy	16.83	16.83	287.05	449.11, 287.06	0.003	0.001
2	Cy-3,5-O-dig	1.99	1.99	611.16	463.12, 625.18, 301.07	0.005	0.005
3	Cy-3-O-gal/glu	5.60	5.57	449.11	287.06	8.519	8.101
4	Cy-3-O-ara	6.58	6.57	419.10	287.06	2.521	2.232
5	Cy-3-O-sop	11.74	11.75	611.16	317.07	0.107	0.099
6	Cy-3-O-rut	13.04	13.05	595.17	287.06	0.019	0.015
Total percentage						11.174	10.453
7	Dp	13.21	13.22	303.05	317.07	0.429b	0.752a
8	Dp-3-O-(6''-p-coumaryl)-glu	16.42	16.42	611.14	303.05	0.083	0.070
9	Dp-3-O-(6-O-acetyl)-glu	14.50	14.51	507.11	303.05	0.399	0.373
10	Dp-3-O-ara	13.21	13.22	435.09	303.05	0.429	0.397
11	Dp-3-O-gal/glu	12.21	12.23	465.10	303.05	1.963	1.869
12	Dp-3-O-rha	5.60	5.57	449.11	317.07	4.929b	8.064a
13	Dp-3-O-rut	1.99	1.99	611.16	317.07	0.004	0.005
Total percentage						8.236b	11.530a
14	Mv-3-O-(6''-acetyl)-gal/glu	9.82	10.65	535.14	331.08, 493.13	1.419b	1.883a
15	Mv-3-O-(6''-malonyl)-glu	9.51	9.51	579.13	331.08, 493.13	0.135	0.123
16	Mv-3-O-(6-O-p-coumaryl)-O-glu	12.36	12.37	639.17	331.08	0.124a	0.083b
17	Mv-3-O-ara	8.16	8.15	463.12	331.08	16.081	15.306
18	Mv-3-O-gal/glu	7.44	7.42	493.13	331.08	25.131	24.625
19	Mv-3,5-O-dig	13.09	13.09	655.19	287.06	0.005	0.004
Total percentage						42.895	42.024
20	Pg	9.06	9.07	271.06	301.07	0.002	0.001
21	Pg-3,5-O-dig	13.04	13.05	595.17	303.05	0.019	0.015
22	Pg-3-O-gal/glu	7.85	7.84	433.11	493.13, 331.08	1.510a	1.274b
23	Pn	15.73	15.76	301.07	303.05	0.002	0.001
24	Pn-3-O-rut	7.51	ND	ND	465.10, 303.05	0.011	ND
25	Pn-3-O-(6''-acetyl)-gal/glu	10.44	10.49	505.13	331.08	0.166b	0.265a
26	Pn-3-O-ara	7.85	7.84	433.11	301.07, 463.12	1.510a	1.276b
27	Pn-3-O-gal/glu	8.16	8.15	463.12	433.11, 271.06	15.943	15.248
28	Pn-3-O-sop-5-O-glu	8.89	8.89	787.23	271.06	0.003	0.002
Total percentage						19.166	18.082
29	Pt	16.84	16.84	317.06	303.05	0.021	0.018
30	Pt-3-O-rut-5-O-glu	8.89	8.89	787.23	317.07, 479.12, 625.18	0.003	0.002
31	Pt-3-O-(6''-acetyl) gal/glu	9.25	9.26	521.13	301.07	0.802	0.887
32	Pt-3-O-(6''-malonyl)-glu	14.86	14.86	565.12	317.07, 479.12	0.075	0.063
33	Pt-3-O-gal/glu	6.27	6.30	479.12	317.07, 479.12	8.731	8.582
34	Pt-3-O-rut	12.99	13.00	625.18	317.07, 479.12	0.236	0.212
35	Pt-3-O-p-coumaroyl-O-glu	16.65	16.65	625.15	317.07, 479.12	0.014	0.010
36	Pt-3-O-ara	5.60	5.57	449.11	301.07	8.647	8.137
Total percentage						18.523	17.911

PHB and PRB peels of highbush blueberry and rabbiteye blueberry, respectively; m/z mass-to-charge ratio, Cy cyanidin, dig diglucoside, gal galactoside, glu glucoside, ara arabinoside, sop sophoroside, rut rutinoside, Pt petunidin, Dp delphinidin, rha rhamnoside, Mv malvidin, Pn peonidin, Pg pelargonidin, ND not detected. Means with different superscript in the same row are significantly different (p < 0.05)

Soluble sugars of flesh from HB and RB

Soluble sugars not only give fruits a sweet and sour taste, but are directly involved in the synthesis of anthocyanins, such as Mv-3-O-glu, and of function as osmoprotectants, preventing intracellular ice formation and protecting membranes and macromolecules from freeze-induced dehydration (Korn et al., 2008). Only 3 soluble sugars, including glucose (305.10 mg/g DW in FHB and 328.00 mg/g DW in FRB), fructose (373.50 mg/g DW in FHB and 407.90 mg/g DW in FRB), and sucrose (46.00 mg/g DW in FHB and 39.60 mg/g DW in FRB), were detected (Table 3 and Supplementary Fig. 2). The fructose content of FRB was slightly higher than that of FHB, which may contribute to the sweet taste of FRB. Additionally, the significantly lower content (p < 0.05) of sucrose in FRB would help RB have a little longer shelf-life (Korn et al., 2008; Lee et al., 2013), compared to HB. Kader et al. (1993), as well as Kalt and McDonald (1996), also found that glucose and fructose were the main soluble sugars in blueberry fruits.

Fatty acid compositions of SHB and SRB

Both SHB and SRB had rich unsaturated fatty acids (USFA), showing 26.59 mg/g DW and 26.37 mg/g DW, respectively. And the major presented were linoleic acid (38.14% in SHB and 32.76% in SRB), followed by α-linolenic acid (27.26% in SHB and 26.49% in SRB) and oleic acid (17.98% in SHB and 26.19% in SRB) (Table 4 and Supplementary Fig. 3), suggesting that USFA content in SHB and SRB reached up to more than 83% and was higher than that of most other plant seeds (Zhou et al., 2020). SHB contained 35 fatty acids, while SRB had only 33 fatty acids and lacked of cis-10-pentadecenoic acid (0.03 mg/g DW) and cis-4,7,10,13,16,19-docosahexaenoic acid (0.19 mg/g DW), which were not found in SRB; in addition, as compared with SHB, SRB showed much higher contents of monounsaturated fatty acids (MUSFA) including oleic acid, significantly lower contents of saturated fatty acids (SFA) including palmitic acid and arachidic acid. It follows then that the development and utilization of SHB and SRB to extract oil would have great prospects in nutraceuticals and functional foods due to the plentiful USFAs. Blueberry seeds from HB and RB can be used as valuable sources of healthy edible oil and medicinal oil, providing a new way for the development of enterprises.

Table 4.

Fatty acid compositions of SHB and SRB

No.	Fatty acids	Retention time (min)		Molecular formula	Contents (mg/g DW)
		SHB	SRB		SHB	SRB
1	Caproic acid (C6:0)	10.186	10.236	C7H14O2	0.0000	0.0000
2	Caprylic acid (C8:0)	11.480	11.476	C9H18O2	0.0011	0.0007
3	Capric acid (C10:0)	13.381	13.377	C11H22O2	0.0252	0.0201
4	Undecanoic acid (C11:0)	14.646	14.659	C12H24O2	0.0056	0.0052
5	Lauric acid (C12:0)	16.102	16.103	C13H26O2	0.0340	0.0311
6	Tridecanoic acid (C13:0)	17.696	17.696	C14H28O2	0.0197	0.0181
7	Myristic acid (C14:0)	19.372	19.372	C15H30O2	0.0530	0.0468
8	Myristoleic acid (C14:1)	20.863	20.859	C15H28O2	0.0305	0.0292
9	Pentadecanoic acid (C15:0)	21.071	21.076	C16H32O2	0.0102	0.0075
10	Cis-10-pentadecenoic acid (C15:1)	22.554	22.568	C16H30O2	0.0302	ND
11	Palmitic acid (C16:0)	22.776	22.776	C17H34O2	2.4785a	2.0414b
12	Palmitoleic acid (C16:1)	24.012	24.012	C17H32O2	0.0657	0.0579
13	Heptadecanoic acid (C17:0)	24.419	24.420	C18H36O2	0.0424	0.0376
14	Cis-10-heptadecanoic acid (C17:1)	25.562	25.563	C18H34O2	0.0528	0.0496
15	Stearic acid (C18:0)	26.019	26.019	C19H38O2	0.8397	0.8789
16	Elaidic acid (C18:1n9t)	26.783	26.787	C19H36O2	0.0495	0.0482
17	Oleic acid (C18:1n9c)	27.083	27.091	C19H36O2	5.4817b	7.8051a
18	Linolelaidic acid (C18:2n6t)	27.884	27.892	C19H34O2	0.0522	0.0499
19	Linoleic acid (C18:2n6c)	28.603	28.600	C19H34O2	11.6341a	9.7664b
20	Arachidic acid (C20:0)	29.029	29.029	C21H42O2	0.1022	0.0754
21	γ-Linolenic acid (C18:3n6)	29.729	29.726	C19H32O2	0.0614	0.0607
22	Cis-11-eicosenoic acid (C20:1)	30.018	30.014	C21H40O2	0.0583	0.0671
23	α-Linolenic acid (C18:3n3)	30.275	30.275	C19H32O2	8.3151	7.8955
24	Henicosanoic acid (C21:0)	30.424	30.411	C22H44O2	0.0348	0.0280
25	Cis-11,14-eicosadienoic acid (C20:2)	31.457	31.453	C21H38O2	0.0634	0.0588
26	Behenic acid (C22:0)	31.851	31.852	C23H46O2	0.0862	0.0784
27	Cis-8,11,14-eicosatrienoic acid (C20:3n6)	32.567	32.440	C21H36O2	0.0236	0.0287
28	Erucic acid (C22:1n9)	32.861	32.861	C23H44O2	0.2008	0.1841
29	Cis-11,14,17-eicosatrienoic acid (C20:3n3)	33.169	33.165	C21H36O2	0.0514	0.0489
30	Tricosanoic acid (C23:0)	33.282	33.282	C24H48O2	0.0662	0.0630
31	Arachaidonic acid (C20:4n6)	33.458	33.431	C21H34O2	0.0172	0.0159
32	Cis-13,16-docosadienoic acid (C22:2)	34.407	34.407	C23H42O2	0.0776	0.0760
33	Lignoceric acid (C24:0)	34.786	34.783	C25H50O2	0.1151	0.1096
34	Cis-5,8,11,14,17-eicosapentaenoic acid (C20:5n3)	35.394	35.494	C21H32O2	0.0540	0.0518
35	Nervonic acid (C24:1)	36.004	35.989	C25H48O2	0.0749	0.0719
36	Cis-4,7,10,13,16,19-docosahexaenoic acid (C22:6n3)	40.248	40.257	C23H34O2	0.1941	ND
Total saturated fatty acids					3.9139a	3.4418b
Total unsaturated fatty acids					26.5885a	26.3657a
Total monounsaturated fatty acids					6.0471b	8.3148a
Total polyunsaturated fatty acids					20.5414a	18.0509b

SHB and SRB seeds of highbush blueberry and rabbiteye blueberry, respectively; DW dry weight, ND not detected. Means with different superscript in the same row are significantly different (p < 0.05)

In vitro digestion properties of blueberry whole fruits, peels and flesh

In vitro digestion and dialysis methods, which are rapid, safe, and no ethical restrictions, were used to simulate the gastrointestinal condition to test the bioavailability of the whole fruits, peels, and flesh of HB and RB. After simulated digestion, the RI of TP and TA was 40.42–60.12% and 10.46–19.93%, respectively,. After the gastrointestinal digestion, the maximum RI (56.05–60.13%) of TP for the peels was found, followed by the whole fruits with a RI of 45.91–49.39%, and the flesh was the lowest (40.42–42.75%) (Fig. 2A). However, the RI of TA of the peels, the whole fruit, and the flesh were only 18.35–19.94%, 11.97–12.99%, and 0.46–10.77%, respectively. Interestingly, the RI values of the whole fruits, peels, and flesh of RB were significantly higher than those of HB. Similar results were found in the wild blueberry (Correa-Betanzo et al., 2014) and other berries (Kaeswurm et al., 2022; Marhuenda et al., 2016; Pinto et al., 2017). The difference in the RI of the whole fruits, peels, and flesh may be due to the differences in the contents and compositions of the bioactive substances, such as pectin and anthocyanins (Table 1).

Fig. 2.

In vitro digestion properties of the whole fruit, peels, and flesh of HB and RB. A: RI and BI; B: DPPH and ABTS scavenging activities; C: Pearson correlation coefficients among TP and TA contents with antioxidant capacities. RI recovery index, BI bioavailability index, TP total phenolics, TA total anthocyanins, HB highbush blueberry, RB rabbiteye blueberry, U undigested samples, S salivary fluid after 2 min of digestion, G1 and G2 gastric fluid after 1 h and 2 h of digestion, respectively; I1 and I2 intestinal fluid after 1 h and 2 h of digestion, respectively; GAE gallic acid equivalent, CGE cyanidin-3-O-glucoside equivalent. PHB and PRB peels of HB and RB, respectively; FHB and FRB flesh of HB and RB, respectively. Different small letters (a-l) represent there are significant difference among the recovery index, and different capital letters (A-L) represent there are significant difference among the bioaccessibility index (p < 0.05)

The BI of the phenols or anthocyanins indicates that they are absorbed by the small intestine (Cao et al., 2021). The change trends of the BI were similar to that of the RI, but the BI of TP and TA both retained low after simulated digestion, which were < 16% (Fig. 2A). McDougall et al. (2005) observed a bioavailability of 5.00% of TA and 10.30% of TP in raspberry, which is in agreement with this study. A lower BI was also reported by Liang et al. (2012), who reported 0.30% for anthocyanin and 12.00% for phenols. Nevertheless, a BI of 27.00–29.00% for TP and 15.00–21.00% for TA in frozen sweet cherries has also been found (Fazzari et al., 2008). The differences in the BI may be due to the differences in the chemical structures of the bioactive substances. In general, the RI and BI of TP and TA in the whole fruits, peels, and flesh of RB were significantly higher than those of HB.

No significant differences in the changes of DPPH and ABTS free radical scavenging capacities were found in individual parts of HB and RB during the oral and gastric digestion phases (p > 0.05), and the antioxidant activity decreased in the order of peels > whole fruits > flesh (Fig. 2B). The DPPH free radical scavenging ability was 58.26–64.50 mg/g DW for the peels (availability of 47.27–48.13%), 28.19–29.49 mg/g DW for the whole fruits (availability of 42.45–43.92%), and 11.01–12.18 mg/g DW for the flesh (availability of 29.07–30.77%); meanwhile, the ABTS free radical scavenging capacity of HB, RB, PHB, PRB, FHB and FRB declined approximately 80.46%, 78.45%, 66.80%, 62.17%, 85.88% and 85.66%, respectively. The differences in the antioxidant activities among the whole fruits, peels, and flesh may be related to the differences in their compositions and contents of the individual polyphenols and anthocyanins (Souza et al., 2014), which could be seen from the results of the Pearson correlation analysis (Fig. 2C). After in vitro digestion, the TP and TA showed a substantial positive association with antioxidant capabilities, which was consistent with the literature (Cao et al., 2021; Marhuenda et al., 2016).

Supplementary Information

Below is the link to the electronic supplementary material.

Declarations

Conflict of interest

The authors declare no conflict of interest.