A comparative metabolomics study of anthocyanins and taste components in Chinese bayberry (Morella rubra) with different flesh colors

Plant materials and samples

Two Chinese bayberry varieties, BDK and DK, were studied. BDK has pink flesh, while DK has purple-red-colored flesh (Fig. 1A). The fruit was collected from nine-year-old trees at the Fruit Research Institute, Fujian Academy of Agricultural Science, Fuzhou, China (latitude 26°07′32.96″N; longitude 119°20′15.25″E). The Chinese bayberry trees were cultivated in open fields using standard horticultural procedures, including disease and pest control.

Ten mature fruits (87 d after flowering, with the color index of red grapes values (CIRG) indices of 4.95–5.05 and 2.50–2.60 for the ripe DK and BDK fruits, respectively) were picked from each of three trees, enucleated, and the flesh amalgamated into one biological sample. Three biological samples (a total of 30 fruits) were used for each of the Chinese bayberry varieties. All samples were quickly frozen in liquid nitrogen and stored at −80 °C.

Total anthocyanin analysis

The total anthocyanin concentration was measured by a spectrophotometric pH differential technique, as previously described (Romero et al., 2008), 1.00 g of bayberry pulp was fully grind it into powder in liquid nitrogen, added 5 mL of pre-cooled extraction solution (methanol containing 0.05% hydrochloric acid), extracted at 4 °C for 12 h, centrifuged at 10,000×g and 4 °C 20 min. The supernatant was collected and transferred to a clean test tube, 5 mL of extract was added to the precipitate, and leaching at 4 °C for 6 h, the supernatant was collected, and the process was repeated once. The resulting supernatants were pooled as the final measured volume. Three measurements were made for each replicate and the data were expressed as the means ± standard deviations (SD) of the three replicates.

Anthocyanin and proanthocyanin extraction and multiple reaction monitoring (MRM)

Experimental methods and data collection as previously described (Yang et al., 2022; Yuan et al., 2018; Zhu et al., 2020). The samples were freeze-dried in a vacuum freeze-dryer. The dried samples were then pulverized in a mixer mill with a zirconia bead for 1.5 min at 30 Hz. The anthocyanins and proanthocynadins were extracted from 50 mg of powdered samples with 0.5 mL methanol/water/hydrochloric acid (500:500:1, V/V/V). The material was vortexed for 5 min and ultrasonicated for a further 5 min, after which it was centrifuged at 12,000 g for 3 min at 4 °C. This procedure was repeated once. The supernatant was retained, filtered, and subjected to UPLC-MS/MS analysis. Main parameters: solvent system, water (0.1% formic acid): methanol (0.1% formic acid); the gradient program was set to: 95:5 V/V at 0 min, 50:50 V/V at 6 min, 5:95 V/V at 12 min, hold 2 min, 95:5 V/V at 14 min; hold 2 min; temperature, flow rate and injection volume were set at 40 °C, 0.35 mL/min and 2 μL, respectively. The effluent was scanned with LIT and triple quadrupole (QQQ) alternatelyon a triple quadrupole-linear ion trap mass spectrometer (QTRAP), operating in positive ion modes. Main parameters: Ion spray voltage (IS) 5,500 V (positive ion), curtain gas (CUR) was set to 35 psi, the source temperature was set at 550 °C. A specific set of MRM transitions were monitored over each period, determined by the eluted metabolites.

In the detection of anthocyanins, except for pelargonidin-3-O-(6-O-malonyl-beta-D-glucoside) and peonidin-3-O-sophoroside (belonged to semi-quantitative substances), other substances had standard substances and belonged to targeted quantitative substances. The information of anthocyanin standard samples were listed in File S1.

Primary metabolite extraction and MRM

The primary metabolites were extracted, and MRMs performed as previously described (Liang et al., 2022; Yang et al., 2022; Zou et al., 2020). After freeze-drying and pulverizing, 100 mg of the powdered material was added to 1.2 mL of 70% methanol solution, vortexed for 30 s every 30 min for a total of six times, and stored at 4 °C overnight. The material was then centrifuged at 12,000 g for 10 min and filtered. The extracts were then subjected to analysis on the UPLC-ESI-MS/MS system. The mobile phase used was composed of solvent A and solvent B, solvent A containing 0.1% formic acid in pure water, and solvent B containing 0.1% formic acid in acetonitrile. Sample analysis was performed using a gradient program with the column oven, flow rate and injection volume set to 40 °C, 0.35 mL/min and 4 μL, respectively. The effluent was scanned with LIT and triple quadrupole (QQQ) alternatelyon a triple quadrupole-linear ion trap mass spectrometer (QTRAP), operating in both positive and negative ion modes. Main parameters: Ion spray voltage (IS) 5,500 V (positive ion)/−4,500 V (negative ion), curtain gas (CUR) was set to 25 psi, ion source gas I (GSI) and gas II (GSII) were set to 50 and 60 psi, respectively, the source temperature was set at 550 °C. A specific set of MRM transitions were monitored over each period, determined by the eluted metabolites.

Statistical analysis

All data were expressed as means ± standard deviation (SD) of three replicates and analyzed using SPSS version 22 (IBM Corp., Armonk, NY, USA). The metabolomic data were analyzed by principal component analysis (PCA) and heatmaps and processed using the web-based tool MetaboAnalyst (https://cloud.metware.cn/). Significantly upregulated and downregulated metabolites between the groups were determined by VIP >= 1 and absolute Log₂FC (fold change) >= 1.

Anthocyanins and proanthocyanins in the two Chinese bayberry varieties

Chinese bayberry fruits are usually white, pink, red, purple-red, and purple-black, dependent on the contents and types of anthocyanins (Jia et al., 2015; Zhang et al., 2009). To investigate the differences in the anthocyanin components of the different Chinese bayberry pulp colors, the purple-red variety DK and its pink-fleshed variant BDK were selected for the study. Visual observation of the two varieties showed that the DK fruit was significantly darker than BDK fruit (Fig. 1A). The total anthocyanin concentrations in the pulp were evaluated using a spectrophotometric pH differential method. This showed that the anthocyanin concentrations were higher in the DK variety (652 mg/kg of FW) in comparison with the BDK fruit (81.6 mg/kg of FW), resulting in the darker red color of DK fruit compared with the light pink coloration of the BDK fruit (Fig. 1B, File S2). This indicated that the anthocyanin content was specifically related to the variety and thus the genetic composition of the Chinese bayberry.

To fully explore the anthocyanin composition of the two Chinese bayberry varieties, the DK and BDK pulp extracts were analyzed by MRM. Eighteen anthocyanins, including cyanidin-3,5-O-diglucoside, cyanidin-3-O-(6-O-malonyl-beta-d-glucoside), cyanidin-3-O-arabinoside, cyanidin-3-O-glucoside, cyanidin-3-O-sambubioside, cyanidin-3-O-sophoroside, cyanidin-3-O-xyloside, delphinidin-3-O-glucoside, delphinidin-3-O-rhamnoside, malvidin-3-O-arabinoside, malvidin-3-o-galactoside, malvidin-3-O-glucoside, pelargonidin-3-O-(6-O-malonyl-beta-d-glucoside), pelargonidin-3-O-glucoside, peonidin-3,5-O-diglucoside, peonidin-3-O-glucoside, peonidin-3-O-sophoroside, and petunidin-3-O-glucoside were isolated and characterized; peonidin-3,5-O-diglucoside was not detected in BDK. Of these 18 anthocyanins, 13 have not been identified in the Chinese bayberry fruit previously. The major anthocyanin identified was cyanidin-3-O-glucoside (Table 1), in agreement with previous observations (Feng et al., 2012). In addition, although the same kinds of anthocyanins were identified in both bayberry varieties, the levels of the compounds differed markedly, with the concentrations of the 18 anthocyanins in DK (with dark-colored pulp) higher than those in BDK (with lighter-colored pulp) (Table 1), suggesting the possible inhibition of genes related to anthocyanin biosynthesis or regulation in BDK, resulting in lower anthocyanin levels in the fruit. These observations were confirmed by the quantitative determination of the anthocyanin concentrations between the two varieties (Fig. 1B).

Three proanthocyanins were identified and analyzed in the pulp of the two Chinese bayberry varieties by MRM. These included procyanidin A2, procyanidin B2, and procyanidin C1. DK was rich in both procyanidin A2 and procyanidin C1, with the concentration of procyanidin A2 in DK being 5.11 times that of BDK, while BDK had a comparable procyanidin B2 content (Table 2).

Primary metabolites in the two Chinese bayberry varieties

To better understand flavor and quality differences between the two varieties, we performed widely targeted UPLC-MS/MS-based metabolite profiling of the differently flavored DK and BDK. A total of 229 metabolites were identified, many of which were likely to have contributed to taste (14 saccharides and alcohols, 41 organic acids, and 47 amino acids and derivatives), together with other primary metabolites (Files S3 and S4).

PCA of the 229 metabolites showed a clear separation between the DK and BDK samples (Fig. 2A), with the first PC accounting for over 75% of the total variability. We used a log10 transformation of the metabolite peaks, followed by hierarchical cluster analysis, to remove the influence of quantity on pattern recognition. As shown by the heatmap (Fig. 2B), the biological replicates from each variety were clustered, indicating the reliability of the metabolic profiles. Thus, both PCA and cluster analysis demonstrated distinct metabolic profiles for the two varieties.

Differentially accumulated primary metabolites in the fruit of the two varieties

To identify differential primary metabolites between DK and BDK, metabolites with fold changes ≥ 2.0 (upregulated) or ≤ 0.5 (downregulated) were selected. The metabolites were analyzed using a variable importance in projection (VIP) value (VIP ≥ 1.0), resulting in the identification of 85 differential metabolites between the varieties (Fig. 3A). In all, 37% of the total 229 metabolites were found to be differentially expressed, indicating significant differences in the metabolite contents of the two varieties. Of these, 62 metabolites were downregulated in BDK compared with DK, while 23 were upregulated (Fig. 3A). These 85 metabolites belonged to seven classes, with most being phenolic acids, lipids, amino acids and amino acid derivatives, and organic acids (Fig. 3B).

Associations of saccharides, organic acids, and other components with fruit taste and quality

The Chinese bayberry fruit pulp contained primarily fructose, glucose, and sucrose. In all, 17 saccharides were detected in the two varieties. To the best of our knowledge, this is the first identification of these saccharides in the pulp of the Chinese bayberry. It was found that the overall sweet taste of the fruit was dependent not only on the concentrations of the various saccharides but also on their relative proportions. The DK variety had relatively higher levels of fructose and glucose and a lower level of sucrose compared to BDK (File S3), while the concentrations of D-glucoronic acid, gluconic acid, D-sorbitol, and mannitol in DK were over twice those seen in BDK, with D-fructose 6-phosphate and D-glucose 6-phosphate levels of less than one-half those of BDK (Table 3).

A total of 41 organic acids were detected in the two varieties, including major organic acids such as citric acid, malic acid, succinic acid, and quinic acid; of these, 37 have not been previously reported (File S3). In terms of differential metabolites, eight organic acids, namely anchoic acid, 2-isopropylmalate acid, muconic acid, D-galacturonic acid, 2-hydroxyhexadecanoic acid, 4-acetamidobutyric acid, (S)-(−)-2-hydroxyisocaproic acid, and 2-hydroxy-4-methylpentanoic acid, were found to be more abundant in DK. Fumaric acid, 3-dehydro-L-threonic acid, diethyl phosphate, 2-hydroxycinnamic acid, and 2,4,6-trihydroxybenzoic acid were found to be more abundant in BDK (Table 3).

Forty-seven amino acids and their derivatives were identified in the two varieties. Of these, 16 amino acids were differentially accumulated using thresholds of VIP ≥1.0 and fold change ≥ 2.0 (upregulated) or ≤ 0.5 (downregulated). Thirteen of these, 3-hydroxy-3-methylpentane-1,5-dioic acid, L-(+)-arginine, N-acetyl-L-glutamic acid, N-acetyl-L-arginine, L-glycyl-L-phenylalanine, L-prolyl-L-leucine, L-valyl-L-leucine, DL-alanyl-DL-phenylalanine, L-leucyl-L-leucine, L-isoleucyl-L-aspartate, L-alanyl-L-phenylalanine, L-leucyl-L-phenylalanine, and L-phenylalanyl-L-phenylalanine showed greater accumulation in DK compared with BDK (Table 3).

Phenolic acids, lipids, nucleotides and their derivatives, and vitamins were also identified. Of these, seven phenolic acids, 18 lipids, four nucleotides and derivatives, and two vitamins were present in greater concentrations in DK than BDK, as determined by thresholds of VIP ≥1.0 and fold change ≥ 2.0 (File S5).