Effects of the flowering process on the metabolite profiles and bioactivities of Shaanxi Fu brick tea

TL;DR

This study shows how a fermentation process called 'flowering' changes the chemical makeup and health benefits of a type of tea.

Contribution

The study reveals specific metabolites and their roles in antioxidant and lipid-lowering effects after the flowering process.

Findings

Flowering reduces tea polyphenols by 13.09% and amino acids by 38.24%.

Fuzhuanin A, teadenol A, and betaine are linked to lipid-lowering activity.

Echinulin and 6-gingerol are associated with antioxidant capacity.

Abstract

This study investigated the impact of “flowering” process (Aspergillus cristatus-mediated fermentation) on the quality of Shaanxi Fu brick tea. Metabolic profiles of twenty samples, collected before (BF) and after (AF) fermentation were analysed. Moreover, in vitro assessments of their antioxidant and hypolipidaemic activities were also evaluated. Results indicated a significant decrease (p < 0.05) in tea polyphenols (13.09%) and free amino acids (38.24%), while caffeine and flavonoids showed no significant changes after flowering process. Integrated gas chromatography–mass spectrometry (GC–MS) and liquid chromatography-mass spectrometry (LC-MS) found that 15 volatiles (e.g. diallyl sulfide) and 47 non-volatile compounds (e.g. fuzhuanin A) were significantly increased. Correlation analysis revealed that antioxidant capacity was closely associated with echinulin and 6-gingerol, whilst hypolipidaemic effect was correlated with fuzhuanin A, teadenol A, and betaine (p < 0.05). These findings underscore the pivotal role of flowering process in shaping the functional metabolite profile and bioactivity of Fu brick tea.

Unlabelled Image

•Flowering-induced chemical and bioactive changes in Fu-brick tea were investigated.•Flowering reduces tea polyphenols by 13.09% and amino acids by 38.24%.•Volatiles and non-volatiles were analysed by widely target metabolomics.•The antioxidant activity is closely related to echinulin and 6-gingerol.•Fuzhuan A, Teadenol A, and betaine contribute to lipid-lowering activity.

Flowering-induced chemical and bioactive changes in Fu-brick tea were investigated.

Flowering reduces tea polyphenols by 13.09% and amino acids by 38.24%.

Volatiles and non-volatiles were analysed by widely target metabolomics.

The antioxidant activity is closely related to echinulin and 6-gingerol.

Fuzhuan A, Teadenol A, and betaine contribute to lipid-lowering activity.

Figures

Comparative analysis of differences in main components between BF and AF. (a) Moisture content; (b) content of water extract; (c) content of caffeine; (d) content of tea polyphenols; (e) content of total amino acids; (f) Content of total flavonoids; (g) content of catechins; (h) content of free amino acids; (i) content of theanine.Fig. 1Table 1Results of physical and chemical components of Fu brick tea samples.Table 1SampleMoisture content (%)Water extract (%)Caffeine (%)Tea polyphenols (%)Total amino acids (%)Total flavonoids (%)BF17.12 ± 0.0645.02 ± 0.362.59 ± 0.0311.70 ± 0.251.48 ± 0.040.77

Volatile metabolites analysis of BF and AF. (a) Proportion of volatiles classification; (b) volatile compounds with a high peak signal; (c) content heatmap of odor components between BF and AF; (d) volcano map analysis of BF and AF; (e) fold change analysis of the differential volatiles of BF and AF.Fig. 2

Non-volatile metabolites analysis of BF and AF. (a) Proportion of non-volatiles classification; (b) non-volatile compounds with a high peak signal; (c) content heatmap of differential metabolites between BF and AF; (d) volcano map analysis of BF and AF.Fig. 3

Analysis of antioxidant activity and bile salts binding activity of Fu brick tea. (a) DPPH radical scavenging capacity. (b) ABTS radical scavenging capacity. (c) Ferric reducing antioxidant power (FRAP). (d) Binding with sodium cholate. (e) Binding with sodium taurocholate. (f) Binding with sodium glycocholate. (g) Heatmap of correlation analysis. Vitamin C (Vc) and Cholestyramine (Chol.) were set as positive control, respectively.Fig. 4

Full text

Introduction

1

Fu brick tea, a distinctive variety of dark tea products, is characterised by a sophisticated process which encompasses a series of methodical steps: steaming, stacking, weighing, pressing, flowering, and drying (Zhu et al., 2020). Each of these stages critically shapes its final quality and sensory property, ultimately yielding a product with a unique chemical composition and associated health benefit (Lin et al., 2021). The primary production areas for Fu brick tea are the Anhua region in Hunan Province and the Jingyang District in Shaanxi Province, China. It is within these regions that the post-fermentation qualities are particularly pronounced and greatly influenced by the activity of native microorganisms, which thrive in controlled environmental settings (Li, Wang, et al., 2024; Liu et al., 2022).

The fermentation stage of Jingyang Fu brick tea is especially noteworthy, as it is during this phase that Aspergillus cristatus emerges as the dominant microorganism. This mould is integral to the biochemical transformations occurring during fermentation, driving various enzymatic activities essential for the formation of metabolites (Chen et al., 2022). Notably, A. cristatus is responsible for producing the characteristic “golden flower”, which not only enhances the visual appeal of the tea but also profoundly contributes to its aroma and taste profile (Li, Dai, et al., 2024; Wang et al., 2024). The microbial activities during the flowering process are crucial, facilitating extensive biochemical conversions that underpin the distinctive sensory profile and potential bioactivity of Fu brick tea (Shao et al., 2024).

Given the complexity and uniqueness of the fermentation process, a comparative analysis of the chemical composition and biological activity of Jingyang Fu brick tea before and after the flowering process is of considerable research significance (Han et al., 2024; Wu et al., 2023). Elucidating these changes can contribute greatly to the chemical and biological mechanisms responsible for the tea's unique qualities and health-promoting effects (Xiao et al., 2024). However, systematic investigations into the quality and chemical profiles of Jingyang Fu brick tea at distinct fermentation stages remain scarcely addressed within the existing literature.

To bridge this gap, the present study was designed to conduct a comprehensive analysis of a representative batch of Jingyang Fu brick tea. We aimed to elucidate variations in key chemical constituents including moisture content, water extract, tea polyphenols, caffeine, total free amino acids, flavonoids, catechin profiles, and amino acids compositions before and after the flowering process. Moreover, metabolomic profiling via GC–MS and LC–MS was employed to assess the profiles of volatile and non-volatile metabolites, as well as the shifts in differential metabolites that occur during the flowering process. Finally, this study integrated multiple assays to evaluate in vitro antioxidant activity and conjugated bile acid binding capacity, thereby comprehensively linking processing-induced chemical transformations to functional outcomes.

Materials and methods

2

Samples and reagents

2.1

A total of twenty tea samples were obtained for this study. Following a paired design, ten samples were collected prior to the flowering process and ten afterwards, with each pair sourced from one of ten representative manufacturers in Shaanxi Province, China. Catechin (C ≥ 98%), (−)-epicatechin (EC ≥ 99%), catechin gallate (CG ≥ 98%), gallocatechin gallate (GCG ≥ 98%), (−)-epicatechin gallate (ECG ≥ 97%), (−)-epigallocatechin (EGC ≥ 95%), (−)-epigallocatechin gallate (EGCG ≥97%), and gallic acid (GA ≥ 98%) were purchased from Sigma-Aldrich Co. (St. Louis, MO, USA). Aspartic acid (Asp ≥98%), asparagine (Asn ≥ 98%), alanine(Ala ≥98%), arginine (Arg ≥ 98%), glutamic acid (Glu ≥ 98%), glutamine (Gln ≥ 98%), glycine (Gly ≥ 98%), histidine (His ≥98%), isoleucine(Ile ≥ 98%), leucine (Leu ≥ 98%), lysine(Lys ≥ 98%), phenylalanine (Phe ≥ 98%), proline (Pro ≥98%), serine(Ser ≥ 98%), tryptophan (Try ≥98%), threonine(Thr ≥ 98%), tyrosine(Tyr ≥ 98%), and valine (Val ≥98%) were purchased from ChemFaces (Wuhan, China). Caffeine (CAF ≥ 98%) was obtained from Enzo Life Sciences Inc. (Farmingdale, NY, USA). Folin-Ciocalteu, 2,2-dihydroxyindane-1,3-dione, 2,2-diphenyl-1-picrylhydrazyl (DPPH), 2,4,6-tripyridyl-s-triazine (TPTZ), and 2,2′-azino-bis (3-ethylbenzothiazoline-6-sulphonic acid) diammonium salt (ABTS), potassium persulfate, pepsin from porcine stomach, trypsin from porcine pancreas, sodium cholate, taurocholic acid sodium salt hydrate, sodium glycocholate hydrate, and rutin were purchased from Aladdin Biochemical Technology Co., Ltd. (Shanghai, China).

Instruments

2.2

MTH-100 thermostatic oscillator (Miulab, China), UV-3600 spectrophotometer (Shimadzu, Japan), ultra-performance liquid chromatography (UPLC, Waters, H-CLASS/Qda, USA), S—433D automatic amino acid analyser (Sykam, Germany), ultra high-performance liquid chromatography (1290 Infinity LC, Agilent, USA) coupled to a quadrupole time of flight (AB Sciex TripleTOF 6600), and 8890 gas chromatography with a 7000D mass spectrometer (Agilen, USA).

Methods of chemical composition determination

2.3

Detection of the main physical and chemical quality components

2.3.1

The moisture content, water extract, tea polyphenol, total amino acids, caffein, catechins were determined according to methods detailed in ISO 14502-1:2005. The total flavonoid content was assessed utilizing the aluminum chloride colorimetric method, as described by Mejri et al. (2020).

Volatile analysis by GC–MS

2.3.2

Volatile compounds were extracted by fully automated headspace solid-phase microextraction (HS-SPME) and analysed by gas chromatography–mass spectrometry (GC–MS) according to a previously described method (Wang, Hua, et al., 2021). Specifically, 1 g of each sample was weighed into a 20 mL headspace vial (sourced from Agilent, Palo Alto, CA, USA) containing a saturated NaCl solution to inhibit enzymatic activity. The vials were sealed with a crimp-top caps fitted with TFE‑silicone headspace septa, also provided by Agilent. Prior to SPME, the sample was equilibrated at 60 °C for 5 min. Volatile compounds were then extracted by exposing a 120 μm DVB/CWR/PDMS fiber (Agilent) to the sample headspace for 15 min at 100 °C.

Following sample extraction, volatile compounds were thermally desorbed from the SPME fiber coating in the GC injector port (Agilent 8890) at 250 °C for 5 min in splitless mode. Volatile compounds were separated, identified, and quantified using an Agilent 8890 GC coupled with a 7000D mass spectrometer. The system was equipped with a DB-5MS high-performance capillary column (30 m × 0.25 mm × 0.25 μm, 5% phenyl-polymethylsiloxane) with helium as the carrier gas at a constant flow rate of 1.2 mL/min. The injector and transfer line temperatures were set at 250 °C and 280 °C, respectively. The oven temperature programme was as follows: initial 40 °C and holding for 3.5 min; increase to 100 °C at 10 °C/min; then to 180 °C at 7 °C/min; and finally, to 280 °C at 25 °C/min, followed by a 5 min hold at 280 °C.

Mass spectrometric detection was performed in electron ionisation (EI) mode at 70 eV. The quadrupole mass detector, ion source, and transfer line temperatures were precisely set to 150 °C, 230 °C, and 280 °C, respectively. Data were acquired in selected ion monitoring (SIM) mode for the accurate identification and quantification of the analytes. Compound identification was primarily based on mass spectral matching against the NIST 2020, Wiley and the internal MWGC database established by Wuhan Metware Biotechnology Co., Ltd. (Wuhan, China).

Non-volatile analysis by LC-MS

2.3.3

The non-volatile analysis was carried out by metabolomics technology based on ultra-high-performance liquid chromatography coupled with quadrupole time-of-flight mass spectrometry (UHPLC-Q-TOF MS) according to the previous reference (Jing et al., 2017). Freeze-dried tea leaves were pulverized into a homogeneous powder under liquid nitrogen conditions using a ball mill. The powdered samples were stored at −80 °C. For metabolite extraction, 80 mg of tea powder was weighed, and the internal standard for process monitoring (DL-4-chlorophenylalanine, 2 μM) was added immediately before solvent addition. Subsequently, 1000 μL of a methanol/acetonitrile/water solution (2:2:1, v/v/v) was added to the homogenized powder. The mixture was ultrasound for 30 min, and centrifuged for 15 min at 14,000g and 4 °C, separating the supernatant from the insoluble residue. The supernatant was subsequently dried in a vacuum centrifuge to concentrate the metabolites.

Prior to LC-MS analysis, the dried metabolite extract was re-dissolved in 100 μL of an acetonitrile/water mixture (1:1, v/v). All analyses were performed on an Agilent 1290 Infinity UHPLC system interfaced with an AB SCIEX TripleTOF 6600 mass spectrometer (Shanghai Applied Protein Technology Co., Ltd.). Chromatographic separation was carried out on a C18 column maintained at 40 °C with a flow rate of 0.4 mL/min and an injection volume of 2 μL. The mobile phase consisted of two components: mobile phase A, containing 25 mM ammonium acetate and 0.5% formic acid in water, and mobile phase B, which was pure methanol. A gradient programme was implemented as follows: 5% B (0–0.5 min), increased linearly to 100% B over 9.5 min, held at 100% B for 2 min, returned to 5% B in 0.1 min, and re-equilibrated at 5% B for 3.4 min.

The electrospray ionisation (ESI) source was operated with the following parameters: ion source gas 1 (Gas1) 60 psi, ion source gas 2 (Gas2) 65 psi, curtain gas (CUR) at 30 psi, source temperature maintained at 600 °C, and IonSpray Voltage Floating (ISVF) adjusted to ±5500 V. In MS^1^ mode, the acquisition range was m/z 60–1000 with an accumulation time of 0.20 s/spectrum. For data-dependent MS^2^ acquisition, the range was m/z 25–1000 with an accumulation time of 0.05 s/spectrum. Information-dependent acquisition (IDA) in high sensitivity mode was utilized for product ion scanning, with collision energy fixed at 35 eV ± 15 eV, declustering potential set to 60 V (+) and − 60 V (−), isotopes excluded within a 4 Da window, and 10 candidate ions monitored per cycle. Finally, the compound identification was performed by matching MS^1^ and MS^2^ spectra against public databases (HMDB, MassBank, GNPS), and internal laboratory databases.

Methods of bioactivities assay

2.4

Evaluation of antioxidant activity in vitro

2.4.1

In vitro antioxidant activity vitamin C (Vc) was set as a positive control in these assays. Three complementary antioxidant assays—ferric reducing antioxidant power (FRAP), 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging, and 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) radical scavenging—were employed to evaluate Fu brick tea samples. The specific experimental methods are detailed in the related reference (Bannour et al., 2017; Tan et al., 2024).

Evaluation of conjugated cholate activity in vitro

2.4.2

Samples before and after “flowering” were simulated for gastrointestinal digestion in vitro, and their binding rates to cholates such as sodium cholate, sodium taurocholate, and sodium glycocholate were tested, as detailed in the relevant reference (Ma et al., 2022). Specifically, both the tea samples and the positive control (cholestyramine) were evaluated at the same concentration (1 mg/mL) in this assay.

Data processing

2.5

The volatile metabolomics data were integrated and corrected for chromatographic peaks using MassHunter software. The non-volatile metabolomics data were used for peak alignment, retention time correction, and peak area extraction using MSDAL software. The data were preprocessed in Excel 2013, and IBM SPSS Statistics26 was used for pairwise samples t-test, and the Duncan test was used for multiple comparisons in ANOVA. The levels of major constituents (moisture, water extract, tea polyphenols, total free amino acids; caffeine, and total flavonoids) were analysed using paired t-tests. For the metabolomics dataset, the false discovery rate (FDR) correction was applied to adjust p-values for multiple hypothesis testing. Orthogonal partial least squares discriminant analysis (OPLS-DA) was performed by SIMCA-P software.

Results and discussion

3

Impact of the flowering process on the key quality components in Fu brick tea

3.1

Tea quality and bioactivity are generally influenced by the major constituents, including moisture content, water-soluble extracts, tea polyphenols, amino acids, caffeine, and flavonoids (Sun et al., 2023). To elucidate the impact of the flowering process on these quality indicators, the levels of the major constituents were determined before and after flowering process. Fig. 1 depicts the chemical compositions of the major constituents in each sample. By comparing the impact of the flowering process on the variations in the quality-related constituents, as illustrated in Fig. 1a, it is evident that 70% of the samples show a statistically significant increase in moisture content following the flowering process (p < 0.05), and the average moisture content of the AF samples increased by 10.68% compared to BF samples. This phenomenon can be attributed to the general necessity of maintaining a specific humidity level to support the growth and reproduction of microorganisms during the flowering process (Rui et al., 2019). Maintaining optimal moisture content during the flowering process plays a pivotal role in facilitating the transformation of both volatile and non-volatile components, which underpin the development of the unique aged aroma, fungal-floral notes, and mellow taste characteristic of Fu brick tea (Li, Zhou, et al., 2023). As shown in Fig. 1b, 70% of these samples demonstrate a statistically significant increase (p < 0.05) in water extract content after flowering, and the average water extract of the AFs increased by 2.31% compared to BFs. This finding aligns with prior research, in which extracellular enzymes secreted by microorganisms break down large molecules in tea into smaller, water-soluble compounds. Consequently, this process elevates the water extract content, thereby enhancing the intensity of tea flavour and the richness of the tea liquor (Guo et al., 2024). In contrast, 50% of the samples exhibit a statistically significant decrease (p < 0.05) in polyphenol content after flowering (Fig. 1c), with a mean reduction of 13.09%. A statistically significant decrease (p < 0.05) in the total free amino acids content was also observed in all samples after flowering (Fig. 1d), with a mean reduction of 38.24%. Regarding caffeine content (Fig. 1e), 60% of the samples showed no significant (p > 0.05) difference in caffeine content before and after flowering. As for the total flavonoid content (Fig. 1f), the majority of these samples (90%) exhibited no statistically significant (p > 0.05) difference in total flavonoid content. Collectively, these results demonstrate that the flowering process exerts a differential influence on the major quality-related constituents of Fu brick tea: the levels of moisture and water-soluble extracts increase significantly, the levels of polyphenols and free amino acids decline, while the levels of caffeine and flavonoids remain essentially unaltered. (See Table 1.)Fig. 1. Comparative analysis of differences in main components between BF and AF. (a) Moisture content; (b) content of water extract; (c) content of caffeine; (d) content of tea polyphenols; (e) content of total amino acids; (f) Content of total flavonoids; (g) content of catechins; (h) content of free amino acids; (i) content of theanine.Fig. 1. Table 1Results of physical and chemical components of Fu brick tea samples.Table 1. SampleMoisture content (%)Water extract (%)Caffeine (%)Tea polyphenols (%)Total amino acids (%)Total flavonoids (%)BF17.12 ± 0.0645.02 ± 0.362.59 ± 0.0311.70 ± 0.251.48 ± 0.040.77 ± 0.10BF27.15 ± 0.0850.04 ± 0.023.39 ± 0.1016.60 ± 0.762.85 ± 0.020.81 ± 0.11BF37.98 ± 0.1042.47 ± 0.202.55 ± 0.2011.16 ± 1.192.52 ± 0.080.85 ± 0.14BF47.60 ± 0.0052.04 ± 0.443.65 ± 0.0520.95 ± 0.463.15 ± 0.050.54 ± 0.11BF59.55 ± 0.1238.72 ± 0.491.90 ± 0.0610.92 ± 0.550.98 ± 0.040.95 ± 0.14BF68.33 ± 0.0240.47 ± 0.012.15 ± 0.1310.07 ± 0.050.93 ± 0.010.47 ± 0.06BF77.98 ± 0.2245.90 ± 0.053.18 ± 0.068.67 ± 0.432.60 ± 0.160.61 ± 0.03BF89.83 ± 0.1450.05 ± 0.603.10 ± 0.0615.41 ± 0.303.59 ± 0.040.53 ± 0.04BF97.08 ± 0.1845.96 ± 0.492.79 ± 0.0712.99 ± 0.761.51 ± 0.030.71 ± 0.14BF108.50 ± 0.0444.73 ± 0.432.33 ± 0.0714.29 ± 0.661.87 ± 0.010.67 ± 0.03AF110.38 ± 0.0146.44 ± 0.672.46 ± 0.0510.37 ± 0.170.98 ± 0.020.81 ± 0.08AF27.75 ± 0.0450.27 ± 0.143.26 ± 0.1214.90 ± 0.921.80 ± 0.040.87 ± 0.18AF37.70 ± 0.0844.42 ± 0.422.92 ± 0.1210.75 ± 0.651.64 ± 0.070.87 ± 0.04AF48.10 ± 0.0555.09 ± 1.953.58 ± 0.0319.56 ± 0.772.13 ± 0.070.57 ± 0.03AF510.95 ± 0.2940.70 ± 0.412.08 ± 0.027.99 ± 0.130.43 ± 0.010.71 ± 0.06AF68.30 ± 0.1243.64 ± 0.082.56 ± 0.088.13 ± 0.420.76 ± 0.050.37 ± 0.08AF710.35 ± 0.0441.00 ± 0.073.40 ± 0.064.15 ± 0.190.93 ± 0.030.45 ± 0.02AF810.33 ± 0.3152.62 ± 0.283.30 ± 0.3515.58 ± 1.472.82 ± 0.141.12 ± 0.18AF98.83 ± 0.1045.78 ± 0.032.64 ± 0.0810.61 ± 0.600.75 ± 0.100.66 ± 0.02AF107.08 ± 0.0045.95 ± 0.222.67 ± 0.0913.33 ± 0.721.02 ± 0.080.83 ± 0.18

Based on the above analysis results, the total polyphenols and amino acids contents in Fu brick tea exhibited some significant changes after flowering process. Therefore, we quantified individual catechin and amino acids profiles (Table S1). As shown in Fig. 1g, gallic acid (GA) was the only phenolic compound that exhibited increased levels in AF samples, whereas all eight quantified catechins exhibited statistically significant decreases (p < 0.05) compared to BF samples. Notably, epigallocatechin gallate (EGCG) emerged as the most abundant catechin in Fu brick tea. And the most significant decrease in EGCG content was observed during the flowering process, with the mean values declining from 2.54% to 1.34%. A parallel trend was observed for free amino acids. Fig. 1h shows significantly lower concentrations of all 19 analysed amino acids in AF samples compared to BF samples. Theanine is the most abundant amino acids in the samples (Fig. 1i), and its average content decreased from 0.49% to 0.26% during the flowering process. Catechins and amino acids constitute the primary flavour components of tea, contributing to its bitter and umami taste, respectively. Generally, under the characteristic elevated temperature and humidity conditions of the flowering phase, these components undergo accelerated depletion. For instance, catechins and related phenolic compounds are subject to oxidative polymerisation, giving rise sequentially to theaflavins, thearubigins, and ultimately theabrownins. These pigment polymers are key contributors to the dark reddish colour and mellow taste of the tea infusion (Li et al., 2021).

Flowering-induced metabolic changes in Fu brick tea revealed by untargeted metabolomics

3.2

Volatile metabolomics analysis

3.2.1

A total of 608 volatile compounds were identified in all samples using SPME-GC–MS as listed in Table S2. These compounds were classified into 14 chemical categories, including 48 alcohols, 45 aldehydes, 56 ketones, 10 carboxylic acids, 78 esters, 9 phenols, 32 aromatic compounds, 3 nitrogen-containing compounds, 141 terpenes, 7 sulfur-containing compounds, 68 hydrocarbons, 97 heterocyclic compounds, 10 amines, and 4 others. As depicted in Fig. 2a, terpenes were the most abundant class, accounting for 23% of all detected volatiles, followed by heterocyclic compounds (16%) and esters (13%). The average relative content of compounds such as limonene, β-ocimene, β-phellandrene, p-cymene, and 3-ethyl-2-methyl-1,3-hexadiene, 3-acetyl-1H-pyrroline, 2-acetylpyridine, m-tolualdehyde, (E)-4-thujanol, γ-terpinene, methyl salicylate and phenylethyl alcohol are relatively abundant in all samples (Fig. 2b). Among these compounds, several abundant terpenoids contribute distinctive flavours to Fu brick tea. Characteristic terpenoid constituents of Fu brick tea, particularly limonene (citrus, lemon), and β-ocimene (floral, fruity), were confirmed to derive from geranyl pyrophosphate (GPP) through the methylerythritol phosphate (MEP) pathway, aligning with canonical terpenoid biosynthesis mechanisms (Jin et al., 2020). Likewise, β-phellandrene (pine, refreshing) is generated from GPP through catalysis by specific monoterpene synthases. Methyl salicylate (holly, mint), although an oxygenated monoterpenoid derivative, retains a terpenoid backbone derived from GPP and undergoes additional cyclisation and oxidative modification steps. These terpenoids are generally regarded as contributors to the varietal aroma profile of tea. In the present study, a modest increase in the content of these compounds was observed following the flowering process.Fig. 2. Volatile metabolites analysis of BF and AF. (a) Proportion of volatiles classification; (b) volatile compounds with a high peak signal; (c) content heatmap of odor components between BF and AF; (d) volcano map analysis of BF and AF; (e) fold change analysis of the differential volatiles of BF and AF.Fig. 2

To elucidate the compositional variations in volatile compounds between BF and AF, orthogonal partial least-squares discriminant analysis (OPLS-DA) was applied. A total of 283 differential metabolites (variable importance in projection, VIP > 1) were identified, effectively discriminating the two groups, as visualised in the heatmap (Fig. 2c). Overall, the majority of volatile compounds significantly decreased in AF samples. As illustrated in Fig. 2d, a rigorous screening process was employed using fold change criteria (FC > 2 or FC < 0.5) combined with a statistical significance threshold of p < 0.05 to pinpoint volatile compounds exhibiting significant increases or decreases in content between BF and AF samples. Consequently, 15 compounds were identified with notable increases and 9 compounds with significant decreases in content, as presented in Fig. 2e. The aroma compounds with significantly decreased content mainly include 1,6-dihydrocarveol, p-tolyl isobutyrate, and umbellulone. Based on reported odor descriptors, 1,6-dihydrocarveol contributes minty and grassy notes, p-tolyl isobutyrate imparts floral and fruity characteristics, and umbellulone provides minty and green notes (Tavares et al., 2020; Ye et al., 2024). After the flowering process, a significant decrease was observed in the content of volatile compounds associated with green, minty, floral, and fruity fragrance notes. This reduction in odor-active compounds thereby contributes to the shift in the overall aroma profile of Fu brick tea after flowering. Conversely, aroma compounds with increased content included diallyl sulfide, (Z)-3-hexenol, methyl salicylate, and 1-hepten-3-one. Diallyl sulfide, characterised by its distinctive garlic and onion odor, is a sulfide compound that may originate from Strecker aldehyde degradation or microbial-driven protein hydrolysis (Choi et al., 2019; Wang et al., 2020). As such odor notes are typically considered off-flavours in most commercial teas and can adversely affect sensory quality at elevated concentrations, the observed increase in diallyl sulfide content warrants further investigation into its biosynthetic origin in Fu brick tea. It has been reported that the concentrations of (Z)-3-hexenol (green), methyl salicylate (holly, mint), and 1-hepten-3-one (metallic) are increased in Fu brick tea, and these compounds are associated with the distinctive “fungal flower” aroma of the tea (Li et al., 2020; Ma et al., 2021). Beyond individual compound accumulation, perceptual interactions among aroma constituents also shape the final olfactory profile. According to the literature (Zheng et al., 2025), various interactions exist among aroma compounds in Fu brick tea, including synergistic effects, as observed between methyl salicylate and citral, and masking effects, as demonstrated between methyl salicylate and decanal. These perceptual interactions and the substantial differential accumulation of aroma-active volatiles during the flowering transition therefore mediate the characteristic aroma formation and quality evolution in Fu brick tea.

In addition, processing parameters during the flowering stage are critical for determining tea quality. Moisture content, in particular, serves as a critical determinant of microbial fermentation dynamics, which directly shape the volatile compound profile. Excessive moisture may promote anaerobic conditions conducive to off-flavour formation, whereas insufficient moisture may constrain microbial and enzymatic activity, limiting aroma development. A comparative study contrasting dry-pile (moisture content <30%) fermentation and wet-pile (moisture content ≈ 60%) fermentation revealed distinct impacts on volatile composition; the fungal genera Debaryomyces, Thermomyces, and Setophoma were identified as important drivers of the aroma differences between the two fermentation regimes, with 3,7-dimethyl-1,5,7-octatrien-3-ol and d-limonene pinpointed as discriminatory volatile markers (Chen et al., 2025). Overall, these findings demonstrate that the flowering process profoundly modulates the volatile metabolome of Fu brick tea. Systematic investigation into the optimisation of process parameters is warranted to enable targeted quality enhancement.

Non-volatile metabolomics analysis

3.2.2

In this study, UPLC-Q-TOF MS analysis detected a total of 1263 and 876 compounds in positive and negative ion modes, respectively (Table S3), demonstrating superior detection capability compared to conventional chromatographic methods. The identified compounds were categorized into 13 distinct chemical classes (Fig. 3a), with lipids and lipid-like molecules representing the most abundant category (23%), followed by phenylpropanoids and polyketides (19%). Notably, hydroxy fatty acids were found to be more abundant in AF samples after flowering process. This finding consisted with previous report that the increase of hydroxy fatty acids can be attributed to their production through microbial-associated secondary metabolism (Zhu et al., 2022). Mass spectrometric analysis revealed distinct profiles of abundant secondary metabolites in Fu brick tea (Fig. 3b). In positive ion mode, the predominant signals corresponded to caffeine, theobromine, neohesperidin, delphinidin-3-O-rutinoside, quercetogetin, and hibiscetin heptamethyl ether. In negative ion mode, higher abundance of catechin gallate, rutin, delphinidin 3-O-glucoside, and myricetin-3-O-galactoside were observed.Fig. 3. Non-volatile metabolites analysis of BF and AF. (a) Proportion of non-volatiles classification; (b) non-volatile compounds with a high peak signal; (c) content heatmap of differential metabolites between BF and AF; (d) volcano map analysis of BF and AF.Fig. 3

Multivariate statistical analysis was conducted to compare the non-volatile metabolite profiles between AFs and BFs. As illustrated in Fig. 3c, the hierarchical clustering heatmap was constructed based on significantly differential compounds (VIP > 1.0 and p < 0.05). Notably, quantitative analysis revealed that 68.3% of these differential metabolites exhibited elevated abundance in AFs compared to BFs. To enhance analytical rigour, a fold-change (FC) threshold combined with statistical significance (p < 0.05) were applied to identify metabolites with substantial quantitative variation during flowering process. In positive ion mode, 34 metabolites showed significant up-regulation (FC > 2.0, p < 0.05), while only 2 compounds were down-regulated. Similarly, negative ion mode analysis identified 13 up-regulated and 2 down-regulated metabolites. Overall, the vast majority (87.2%) of differentially accumulated non-volatile compounds were present at higher levels in post-flowering samples.

Among the significantly up-regulated metabolites, several compounds of particular interest were tentatively identified, including fuzhuanin A, teadenol A, betaine, 6-gingerol, reserpic acid, echinulin, flavoglaucin and several others. Conversely, the compounds with notable decreases in content primarily encompass benzyl-β-D-glucopyranoside, 12-hydroxydodecan-3-enoic acid, trehalose, and several other compounds. Interestingly, a substantial proportion of the metabolites that accumulated significantly after the flowering process are known to be associated with microbial metabolism. These include both direct microbial secondary metabolites (e.g. echinulin and flavoglaucin) and compounds derived from the microbial biotransformation of endogenous tea constituents. For instance, fuzhuanin A and teadenol A have been characterised as oxidation products derived from the B- and C-rings of epigallocatechin gallate during fungal fermentation of Fu brick tea (Nagasawa et al., 2020; Xie et al., 2020). The fungus Eurotium cristatum is recognised for its capacity to produce a diverse array of indole alkaloids and other bioactive secondary metabolites. Representative examples include neoechinulin A, isoechinulin A, tardioxopiperazine A, and echinulin—the latter originally isolated from the strain E. cristatum EN-220 (Du et al., 2012). The increased abundance of echinulin in AF samples, coinciding with the proliferation of E. cristatum, corroborates this biosynthetic capacity. Similarly, a series of hydroxybenzaldehyde derivatives biosynthesized by E. cristatum via the polyketide pathway, including isodihydroauroglucin, dihydroauroglucin, tetrahydroauroglucin, and flavoglaucin, have been identified in Fu brick tea (Song et al., 2022). Our analysis further confirmed a significant increase in the content of flavoglaucin after flowering process driven by E. cristatum. Moreover, the formation of these compounds during the flowering process warrants further investigation using reference standards, including fuzhuanin A, teadenol A, betaine, 6-gingerol, reserpic acid, echinulin, and flavoglaucin.

In this study, these findings demonstrate that the flowering process not only elevates the abundance of the majority of non-volatile metabolites but also introduces characteristic microbial secondary metabolites into the tea matrix. Accordingly, precise regulation of processing parameters during the flowering stage is of considerable importance. Moisture content, in particular, acts as a critical determinant of microbial community structure and metabolic output in fermented tea (Li, Zhou, et al., 2023). This principle is exemplified by a recent study on Tibetan tea, in which 407 differential non-volatile metabolites were identified between dry-pile and wet-pile fermentation regimes, underscoring the profound biochemical consequences of moisture modulation (Chen et al., 2025). By governing water activity, it is possible to influence the proliferation and secondary metabolism of dominant fungal populations, thereby shaping the profile of microbial-derived flavour compounds and bioactive constituents in the finished product.

Flowering process-induced changes in bioactive functions of Fu brick tea

3.3

Analysis of antioxidant activity

3.3.1

Antioxidant capacity constitutes a key functional attribute of tea. As shown in Table 2, the positive control, vitamin C (Vc), exhibited IC_50_ values of 0.053 ± 0.001 mg/mL (DPPH), 0.385 ± 0.002 mg/mL (ABTS), and 8.153 ± 0.009 mg/mL (FRAP), indicating a substantially higher antioxidant capacity than that of all tested tea samples. For samples, the results demonstrate significant variations in antioxidant activities among different Fu brick tea samples. Moreover, distinct changes in antioxidant capacity were observed within the same sample when comparing pre-flowering and post-flowering stages. Antioxidant results obtained by the three methods are shown in Fig. 4a-c. Following the flowering process, certain samples (F6, F8, F10) exhibited a significant decrease (p < 0.05) in antioxidant activity, whereas the others (F2, F7) displayed a significant increase (p < 0.05). This non-uniform changes in antioxidant capacity observed during flowering process may reflect intrinsic variations in flowering intensity and associated metabolic shifts among samples.Table 2. Antioxidant activity of Fu brick tea.Table 2. SamplesDPPH-IC_50_(mg/mL)ABTS-IC_50_(mg/mL)FRAP value(mmol/L)BFAFBFAFBFAFF10.566 ± 0.0020.597 ± 0.0034.069 ± 0.0093.316 ± 0.0021.032 ± 0.0141.083 ± 0.009F20.435 ± 0.0050.355 ± 0.0002.397 ± 0.0192.037 ± 0.0101.840 ± 0.0321.963 ± 0.018F30.258 ± 0.0060.360 ± 0.0092.073 ± 0.0561.909 ± 0.0192.233 ± 0.0412.982 ± 0.006F40.562 ± 0.0360.643 ± 0.0002.839 ± 0.2193.044 ± 0.0591.266 ± 0.0171.208 ± 0.003F50.238 ± 0.0140.228 ± 0.0041.598 ± 0.0081.559 ± 0.0432.385 ± 0.0292.945 ± 0.041F60.578 ± 0.0341.051 ± 0.0013.115 ± 0.0814.524 ± 0.0231.277 ± 0.0140.745 ± 0.002F70.917 ± 0.0610.537 ± 0.0155.936 ± 0.1263.603 ± 0.1150.722 ± 0.0041.088 ± 0.000F81.275 ± 0.0712.524 ± 0.0766.925 ± 0.01014.606 ± 0.0490.51 ± 0.0060.257 ± 0.001F90.369 ± 0.0090.335 ± 0.0022.126 ± 0.0062.353 ± 0.0061.797 ± 0.0071.769 ± 0.015F100.45 ± 0.0050.53 ± 0.0123.217 ± 0.0093.854 ± 0.0551.279 ± 0.0021.062 ± 0.011Vc0.053 ± 0.0010.385 ± 0.0028.153 ± 0.009Fig. 4Analysis of antioxidant activity and bile salts binding activity of Fu brick tea. (a) DPPH radical scavenging capacity. (b) ABTS radical scavenging capacity. (c) Ferric reducing antioxidant power (FRAP). (d) Binding with sodium cholate. (e) Binding with sodium taurocholate. (f) Binding with sodium glycocholate. (g) Heatmap of correlation analysis. Vitamin C (Vc) and Cholestyramine (Chol.) were set as positive control, respectively.Fig. 4

Previous studies have established that the antioxidant activity of Fu brick tea is closely associated with components such as catechins, gallic acid, polysaccharides and tea pigments (Chen et al., 2022; Tan et al., 2024; Yao et al., 2017). During the flowering process, although the content of most catechins declined, the content of gallic acid, polysaccharides and tea pigments increased. Moreover, microbial secondary metabolites generated during fermentation contribute substantially to the bioactivity profile of Fu brick tea. For instance, flavoglaucin—a polyketide derivative biosynthesised by E. cristatum—has been reported to possess both antioxidant and hypolipidaemic activities (Song et al., 2022). To further explore the mechanistic basis of the observed functional alterations, correlation analysis was performed between bioactivity indices and non-volatile components exhibiting significant changes during the flowering process (Fig. 4g). It can be seen that the antioxidant activity showed positive associations with echinulin, flavoglaucin, and 6-gingerol. Statistically significant positive correlations were identified specifically for echinulin (r = 0.45–0.48, p < 0.05) and 6-gingerol (r = 0.72–0.74, p < 0.05). This further confirms the impact of microbial metabolites on antioxidant activity. Notably, the abundance of these antioxidant-related metabolites varied considerably among Fu brick tea samples, reflecting differences in processing parameters across manufacturers. Such variability likely underpins the inconsistent direction and magnitude of antioxidant activity changes observed between samples. Collectively, these findings indicate that the impact of flowering process on the antioxidant activity of Fu brick tea is primarily governed by the dynamic compositional shifts of these key bioactive constituents.

Analysis of binding capacities with bile salts

3.3.2

The lipid-lowering potential of tea can be evaluated in vitro by measuring its bile acid salt-binding capacity, as a positive correlation has been established between this binding affinity and the hypolipidaemic effects of tea-derived bioactive compounds (Ma et al., 2022). As presented in Table 3, the positive control, cholestyramine, exhibited substantially higher binding rates for sodium cholate (73.23% ± 2.36%), sodium taurocholate (75.02% ± 1.85%), and sodium glycocholate (80.17% ± 1.26%) relative to all Fu brick tea samples, confirming its potent in vitro lipid-lowering activity. Notable variations in the bile salt binding activity were observed among different tea samples. The binding activity of Fu brick tea samples with sodium cholate (Fig. 4d), sodium taurocholate (Fig. 4e), and sodium glycocholate (Fig. 4f) presenting significant differences between BF and AF samples. Overall, the majority of AF samples exhibited enhanced binding affinity toward sodium cholate and sodium taurocholate, with particularly significant increases (p < 0.05) observed for samples F5, F7, F9, and F10. These results suggest that the flowering process confers improved in vitro hypolipidaemic potential upon Fu brick tea.Table 3. Results of conjugated bile salts before and after the flowering of Fu brick tea.Table 3. SamplesBinding rate with sodium cholate %Binding rate with sodium taurocholate %Binding rate with sodium glycocholate %BFAFBFAFBFAFF125.00 ± 2.1137.38 ± 3.0533.69 ± 0.8739.47 ± 1.1934.81 ± 0.5244.54 ± 0.23F234.20 ± 0.7039.25 ± 1.1433.33 ± 4.0544.74 ± 1.1942.41 ± 0.5250.29 ± 0.23F332.47 ± 1.6437.38 ± 3.8233.69 ± 3.1840.06 ± 0.2444.30 ± 1.0345.98 ± 1.88F427.30 ± 1.1736.68 ± 2.1035.82 ± 0.8735.78 ± 0.1541.46 ± 4.9141.67 ± 0.23F533.62 ± 1.1741.82 ± 0.1925.53 ± 1.1637.84 ± 0.2443.35 ± 2.3342.24 ± 0.23F632.76 ± 0.9435.98 ± 0.3830.50 ± 1.1633.63 ± 4.5441.14 ± 1.5538.22 ± 2.11F731.61 ± 0.9435.98 ± 0.7619.15 ± 0.5840.64 ± 1.6732.59 ± 3.3639.66 ± 0.47F813.22 ± 1.4114.25 ± 2.8612.06 ± 1.1615.50 ± 0.2411.39 ± 0.5212.93 ± 3.05F931.61 ± 3.2841.12 ± 0.7620.57 ± 0.5838.01 ± 0.4839.24 ± 2.5850.29 ± 1.17F1027.01 ± 0.9435.98 ± 1.1425.18 ± 0.2937.72 ± 0.7241.14 ± 2.0745.69 ± 1.17Cholestyramine73.231 ± 2.36475.023 ± 1.85380.167 ± 1.256

Previous researches have demonstrated that several components accumulated during the flowering process, including theabrownin, polysaccharides, and statins, among others, possess the ability to augment the lipid-lowering effects of Fu brick tea (Li, Liu, et al., 2023; Wang, Zhao, et al., 2021; Yang et al., 2023). Furthermore, correlation analysis (Fig. 4g) revealed positive associations between bile salt binding rates and a suite of non-volatile metabolites (echinulin, flavoglaucin, fuzhuanin A, teadenol A, betaine, reserpic acid). Statistically significant correlations (p < 0.05) were identified for fuzhuanin A (r = 0.37–0.62), teadenol A (r = 0.32–0.59), and betaine (r = 0.30–0.51). Although these findings establish a clear association between specific metabolites and in vitro lipid-lowering activity, whether this activity translates into physiologically relevant effects in vivo remains to be elucidated. The complex physiological environment may substantially influence the bioavailability and bioefficacy of these compounds. Accordingly, future investigations employing suitable animal models or clinical trials are warranted to validate their functional efficacy and to delineate the underlying mechanisms of action.

Conclusion

4

In summary, this study provides a comprehensive insight of the chemical and functional transformations that occur during the flowering process of Shaanxi Fu brick tea. Specifically, the levels of moisture and water extract in most Fu brick tea samples increased after flowering, whereas the total polyphenols and free amino acids decreased. Notably, the levels of most individual catechins and amino acids decreased significantly, whereas caffeine and total flavonoid levels remained relatively unchanged. Metabolomic analysis indicated a substantial decrease in the majority of volatile compounds and a significant increase in non-volatile compounds after flowering. These dynamic compositional shifts collectively underpin the development of the characteristic flavour profile and the modulation of functional properties associated with finished Fu brick tea. Assessment of biological activity further demonstrated that while antioxidant capacity decreased in most samples after flowering, bile acid-binding activity was enhanced in the majority AFs. Of particular significance, the flowering process induced the accumulation of several microbial-derived metabolites, including echinulin, flavoglaucin, fuzhuanin A, and teadenol A. Correlation analyses implicated these compounds as positive contributors to both antioxidant and hypolipidaemic activities, suggesting that they serve as potential functional markers of flowering-driven bioactivity enhancement. Furthermore, the synergetic effects of these functional chemicals and their interactions with microbiota should be investigated, and the mechanism behind the bioavailability and in vivo regulation of the aforementioned biochemicals are also worth in-depth study. These findings underscore the profound influence of flowering process on the chemical composition and biological activities of Fu brick tea, thereby furnishing a scientific rationale for subsequent production and development of Fu brick tea.

CRediT authorship contribution statement

Wanjun Ma: Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Honger Yao: Writing – original draft, Software, Formal analysis, Conceptualization. Li Zhao: Validation, Software, Investigation. Haipeng Lv: Writing – review & editing, Supervision, Resources, Project administration, Methodology, Funding acquisition, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.