Flavor enhancement of summer-autumn black tea through synergistic processing with mixed fruit juice: a metabolomics analysis

Abstract

Tea processed from the leaves harvested in the summer and autumn is often characterized by a pronounced bitterness and astringency, coupled with an aroma profile that lacks complexity. This study aimed to enhance the flavor of summer-autumn black tea by incorporating a mixed fruit juice (Siraitia grosvenorii and Clausena anisum-olens (Blanco) Merr.) during processing. Its impact was comprehensively assessed based on sensory attributes, physicochemical components, and key volatile and non-volatile substances. The results demonstrated that the addition of the mixed fruit juice significantly enhanced the sensory flavor of the black tea, specifically intensified fruity, woody, and spicy aroma notes. A total of 17 key aroma compounds with OAVs >1 were identified, primarily contributing to honey, floral, woody, and spicy aromas. Metabolomics analysis further revealed that the juice addition significantly elevated the levels of theaflavins, mogrol, apigenin, vitexin, cosmosiin, γ-aminobutyric acid, and coumarin derivatives (P < 0.001).

Highlights

• •A reference for improving the quality of summer-autumn tea.
• •Mixed fruit juice significantly enhanced summer-autumn black tea flavor.
• •Molecular basis for flavor enhancement through the addition of mixed juice revealed.
• •Addition of mixed juice drives the consumption of surrounding agricultural products.

1. Introduction

Black tea, a fully fermented tea variety, is rich in bioactive compounds and widely appreciated by consumers globally for its distinctive flavor profile and health benefits (Assad et al., 2023). It accounts for 70–80% of global tea consumption and ranks as China's second-largest tea export category (Wang, Lei, et al., 2024; Wang, Liu, et al., 2024). In recent years, demand for high-quality black tea has steadily increased in both domestic and international markets, with aroma and taste serving as key quality indicators (Ma et al., 2025). According to “Tea Sensory Evaluation Method” (GB/T 23776–2018), aroma and taste contribute 25% and 30%, respectively, to the total score in Gongfu black tea evaluation. However, during summer and autumn, intense sunlight and high temperatures promote vigorous carbon metabolism in tea plants—favoring the accumulation of polyphenols and anthocyanins—while suppressing nitrogen metabolism, which limits the synthesis of amino acids, vitamins, and aromatic compounds. As a result, summer and autumn teas, which constitute approximately 60% of annual tea production, often exhibit quality deficiencies such as bitterness, reduced freshness, and weak fragrance (Guo et al., 2025; Wang et al., 2021), significantly affecting plantation productivity.

Clausena anisum-olena (Blauco) Merr. (commonly known as Shanhuangpi), a perennial evergreen shrub or small tree belonging to the Rutaceae family, is primarily distributed in Guangxi, Yunnan, and Guangdong provinces of China (Luo et al., 2023; Qin et al., 2017). Its fruit possesses a pleasantly sour taste and a unique aroma, and is rich in amino acids including aspartic acid, γ-aminobutyric acid, and alanine (Yang & Huang, 2011). Additionally, it contains abundant vitamins, pectin, organic acids, polyphenols, and antioxidants, contributing to various health benefits such as neuroprotection against Alzheimer's disease and tumor inhibition (Fu et al., 2011; Prasad et al., 2009). Notably, both the peel and leaves of Shanhuangpi are rich in volatile aroma compounds, including myristicin, terpinolene, β-myrcene, 3-carene, and β-bisabolene (Zhou et al., 2008). The application of Shanhuangpi in new-style tea beverages continues to rise and has become a new internet-famous product. Siraitia grosvenorii, a herbaceous plant belonging to the Cucurbitaceae family, is a traditional medicinal and edible herb with an annual output value exceeding 10 billion yuan. It contains triterpenoid saponins, polysaccharides, flavonoids, amino acids, and lignans, which contribute to its anti-inflammatory, antioxidant, and immunomodulatory properties (Wu et al., 2022), indicating considerable potential for further development. The primary triterpenoids isolated from Siraitia grosvenorii are mogrol and mogrosides, which share a common mogroside aglycone structure but differ in the number of glucose units attached—increased glycosylation correlates with greater sweetness (Wu et al., 2022). Mogrosides are natural sweeteners approximately 300 times sweeter than sucrose and non-caloric, making them ideal sugar substitutes for individuals with diabetes or obesity who must limit sugar intake. The incorporation of Siraitia grosvenorii into summer and autumn tea formulations may effectively mitigate bitterness and astringency without compromising the beverage's health-promoting attributes.

Given the rich flavor profile, distinctive aroma, and multiple pharmacological properties of both Shanhuangpi (SHP) and Siraitia grosvenorii (SG), this study proposes a novel approach to enhance summer and autumn black tea by incorporating fresh pulp, peel, and leaves of these fruits to produce a fruit-scented black tea with improved sensory attributes and health benefits. This strategy aims to address the common defects of bitterness, astringency, and weak aroma in off-season teas. HS-SPME-GC–MS and UPLC-MS/MS were employed to analyze changes in key aroma and nutritional components, providing a scientific foundation for the high-value utilization of summer and autumn tea resources, the promotion of underutilized fruit species, and the comprehensive development of functional tea products.

2. Materials and methods

2.1. Materials and chemicals

Fresh tea leaves: Fresh leaves of Yunnan large-leaf tea cultivar, were harvested in June 2024 from the tea plantation of the South Subtropical Agricultural Science Research Institute, Guangxi Academy of Agricultural Sciences. The collection was selectively conducted, with the plant materials comprising mainly one bud with three or four leaves.

Instruments and chemicals: High-precision digital refractometer, Phenolphthalein (indicator), Anhydrous ethanol, Sodium hydroxide (NaOH), Microplate reader, Low-temperature centrifuge, pH meter, amino acid content determination kit, plant soluble sugar content determination kit, plant total phenol determination kit (Suzhou Mengxi Biomedical Technology Co., Ltd.), and standard of caffeine (Henan Wanjia Standard Material Center, purity ≥98%), theaflavin (Shanghai yuanye Bio-Technology Co., Ltd., purity ≥98%).

2.2. Tea sample processes

The processing of the tea samples was conducted by the Tea Processing Research Center of the South Subtropical Agricultural Science Research Institute in Chongzuo, China. The specific process flow is detailed below:

Preparation of Shanhuangpi (SHP) pulp and peel concentrates: Deseeded, frozen SHP (1 kg) was separated into peel and pulp. Each component was individually combined with 2 L of water. The mixtures were brought to a boil and then simmered for 40 min under constant stirring. Subsequently, they were filtered through 300-mesh nylon gauze, yielding two separate concentrates (peel and pulp). These concentrates were stored at 4 °C before use.

Preparation of Siraitia Grosvenorii (SG) juice concentrate: 50 G of SG was added to 2 L of water. The mixtures were heated to boiling over high heat, then simmered for 40 min with constant stirring. After boiling, filter the mixture through a 300-mesh nylon gauze to obtain the concentrated juice. The final concentrate was stored at 4 °C before use.

Control group black tea (CK) processing: The processing of fresh leaves (15 kg) involved successive steps of withering (10 h at room temperature), rolling (1 h), and fermentation (4 h), followed by drying. The drying was conducted first at 100 °C for 30 min, with a subsequent phase at 80 °C for 150 min.

Treatment group black tea (GC) processing: The processing protocol was consistent with the control, with the exception of incorporating specific additives before the rolling and drying stages. Before rolling, a mixture of fruit juice concentrates (SG: 0.12 L; SHP peel: 0.18 L; SHP pulp: 0.30 L) was added per 15 kg of withered leaves. The proportion with the highest score obtained after sensory evaluation based on different proportions of mixed juice (Table 1).

Table 1.

Sensory evaluation results of mixed fruit juices (Siraitia grosvenorii and Shanhuangpi) incorporated at varying proportions during processing.

Tea sample: 15 kg fresh leaves of Camellia sinensis var. assamica (Yunnan large-leaf species)	Appearance of dry tea (25%)		Tea infusion						Infused leaves (10%)		Total Score
			Color (10%)		Aroma (25%)		Taste (30%)
	Comments	Score	Comments	Score	Comments	Score	Comments	Score	Comments	Score
During rolling, add 75 mL Siraitia grosvenorii, 75 mL Shanhuangpi peel and 75 mL Shanhuangpi pulp.	Tight and knotted strips, slightly moist.	85	Bright red.	89	Sweet with fruity notes.	87	Mellow and smooth.	85	Red, relatively intact.	83	85.7
During rolling, add 75 mL Siraitia grosvenorii, 112.5 mL Shanhuangpi peel and 150 mL Shanhuangpi pulp.	Tight and knotted strips, slightly moist.	85	Bright red.	89	Sweet and fruity.	90	Mellow and smooth.	85	Red, relatively intact.	82	86.35
During rolling, add 112.5 mL Siraitia grosvenorii, 150 mL Shanhuangpi peel and 225 mL Shanhuangpi pulp.	Tight and knotted strips, slightly moist.	85	Bright red.	89	Sweet and fruity.	90	Sweet with a hint of acidity.	82	Red, relatively intact.	82	85.45
During rolling, add 150 mL Siraitia grosvenorii, 225 mL Shanhuangpi peel and 300 mL Shanhuangpi pulp.	Tight and knotted strips, slightly moist.	86	Bright red.	89	Sweet with fruity notes.	88	Sweet with fruity acidity.	80	Red, relatively intact.	82	84.6
During rolling, add 187.5 mL Siraitia grosvenorii, 300 mL Shanhuangpi peel and 375 mL Shanhuangpi pulp.	Tight and knotted strips, slightly moist.	85	Bright red.	88	Sweet and fruity.	90	Mellow with astringency.	80	Red, relatively intact.	81	83.03
During rolling, add 120 mL Siraitia grosvenorii, 180 mL Shanhuangpi peel, and 300 mL Shanhuangpi pulp. During drying, add 700 g fresh Shanhuangpi leaves.	Tight and knotted strips, dark black with oily lustre.	88	Bright red.	90	Intense and high - pitched fruity notes with sweet undertones.	91	Mellow and smooth, with prominent fruity aroma.	91	Red, relatively even and intact.	90	90.05

Note: Shanhuangpi, fruit of Clausena anisum-olens (Blanco) Merr.

2.3. Sensory evaluation

Appearance of dry tea, color, aroma, taste of tea infusion and infused leaves of the finished tea products made by the processes were evaluated according the Chinese national standard procedure for evaluating tea leaves (GB/T 23776, 2018). Tea leaves were brewed with boiling water for 5 min in a tea-to-water ratio of 1:50. After filtration, the color, aroma, taste of tea infusions were evaluated in order.

2.4. Analysis of biochemical components

Amino acids, tea polyphenols, soluble sugar and other biochemical indicators were tested and analyzed in strict accordance with the requirements of the kit; water extracts were tested and analyzed with reference to the national standard Determination of Tea Extracts (GB/T 8305–2013); The determination of theaflavins and caffeine refers to the national standard “Determination of theaflavins in tea by high performance liquid chromatography” (GB/T 30483 2013).

2.5. Analysis of aroma components

2.5.1. Solution preparation

Standard solutions were prepared as follows and stored at 4 °C before use. Internal Standard: 2-Octanol was dissolved in a 50% ethanol aqueous solution to a final concentration of 10 mg/L. n-Alkanes Standard: n-Alkanes (C7-C40) were dissolved in hexane to a final concentration of 1 mg/L.

2.5.2. Volatiles extraction

A 1 g sample was weighed into a 20 mL headspace vial, followed by the addition of 10 μL of internal standard solution. The vial was incubated at 60 °C for 10 min. Prior to extraction, the SPME fiber was conditioned at 270 °C for 10 min. It was then exposed to the headspace of the vial at 60 °C for 15 min to adsorb volatiles. After extraction, the fiber was transferred to the GC injection port and desorbed at 250 °C for 5 min. Finally, the fiber was re-conditioned at 270 °C for 10 min. For the n-alkanes standard (C7-C40), 10 μL of the solution was transferred to a separate 20 mL headspace vial and analyzed following the same procedure as the tea samples.

2.5.3. GC × GC-TOF MS analysis

Chromatographic conditions: LECO Pegasus BT 4D (LECO, St. Joseph, MI, USA) GC × GC-TOF MS chromatographic system was prepared by Agilent 8890 A gas chromatograph (Agilent Technologies, Palo Alto, CA, USA), a dual-stage jet modulator, and a split/splitless injection module,The mass spectrum system is a high-resolution TOF mass spectrum detector. The separation system was a DB-Heavy Wax (30 m × 250 μm × 0.5 μm) (Agilent, USA) column. High purity helium was used as the carrier gas at a constant flow rate of 1.0 mL/min. Chromatographic column DB-Heavy Wax (30 m × 250 μm × 0.5 μm): the initial temperature was 50 °C, held for 2 min, and then increased to 240 °C at a rate of 5 °C/min, held for 5 min.

MS conditions: LECO Pegasus BT 4D MS detector (LECO, St. Joseph, MI, USA), MS transmission line temperature of 250 °C, ion source temperature of 250 °C, acquisition rate of 10 spectra/s, electron impact source of 70 eV, The detector voltage was 2014 V, and the mass spectrum was scanned from m/z 35 to 550.

2.5.4. Qualitative and quantitative, OAV and ACI analysis

The results of GC–MS data were retrieved by NIST11 spectrum library and compared with the mass spectrum obtained from the experiment. After analyzing the spectrum, the substances with matching degree greater than 90% were found as the identification standard. The concentration ($C_x$) of each volatile substance was calculated relative to the internal standard; the OAV value of the volatile substance is calculated according to formula (1); the ACI value of the characteristic aroma substance is calculated according to formula (2).

$$\mathrm{OAV}=\frac{C_{\chi }}{{\mathit{OT}}_{\chi }}$$ (1)

$$\mathrm{ACI}\%=\frac{O_x}{\sum n\mathit{On}}\times 100$$ (2)

where $C_{\chi }$ is the mass concentration of volatile component X/ (μg/Kg); ${\mathit{OT}}_{\chi }$ is the aroma threshold of volatile component X in water/ (μg/Kg); $O_x$ is the OAV of volatile component X; $\sum n\mathit{On}$ is the sum of OAV of all key volatile components.

2.6. Metabonomics analysis

The non-volatile components were profiled using a Waters Acquity I-Class PLUS UPLC system coupled to a Xevo G2-XS QTOF high-resolution mass spectrometer. Tea samples (50 mg) were extracted with 1000 μL of a methanol:acetonitrile:water (2:2:1, v/v/v) solution containing a 20 mg/L internal standard. The mixture was vortex-mixed, homogenized with steel balls (45 Hz, 10 min), and sonicated in an ice-water bath. Following incubation at −20 °C for 1 h and centrifugation (12,000 rpm, 4 °C, 15 min), 500 μL of supernatant was dried under vacuum. The residue was reconstituted in 160 μL of 50% aqueous acetonitrile, vortexed, sonicated, and centrifuged again. Finally, 120 μL of the supernatant was collected for analysis, with a pooled QC sample created from 10 μL of each sample extract.

Raw data from MassLynx V4.2 were processed by Progenesis QI software for peak picking and alignment. Metabolites were identified using Progenesis QI against the METLIN database, public databases, and a self-built Baimaike database, with mass accuracies of 100 ppm for precursor ions and 50 ppm for fragment ions.

2.7. Statistical analysis

Principal component analysis (PCA), hierarchical cluster analysis (HCA) and OPLS-DA were performed using SIMCA (version 14.1) software. The variable important in projection (VIP) was extracted from the orthogonal projection latent structure discriminant analysis (OPLS-DA) model, and the score plot and permutation test plot were drawn (Wen et al., 2023). One-way analysis of variance (ANOVA), least significant difference (LSD) and correlation analysis were conducted by SPSS (version 29.0.1.0).

3. Results analysis

3.1. Effects of mixed fruit juice addition on the sensory and physicochemical components of summer-autumn black tea

The sensory evaluation results of the two black tea types are presented in Table 2a. The GC group samples, prepared with mixed fruit juice, exhibited a rich and pronounced fruity aroma, a sweet fragrance, and a mellow taste with distinct fruity notes, resulting in good palatability and a higher overall score of 90.61. In contrast, the CK group samples demonstrated a comparatively weaker aroma and taste, with a less substantial body, leading to a lower score of 85.15.

Table 2.

Effects of Mixed Fruit Juice Addition on the Sensory and Physicochemical Components of Summer-Autumn Black Tea.

a) Sensory Quality Analysis of CK and GC Black Tea
Sample	Appearance (25%)		Soup Color (10%)		Aroma (25%)		Taste (30%)		Infused Leaves (10%)		Comprehensive Score
	Comments	Score	Comments	Score	Comments	Score	Comments	Score	Comments	Score
CK	Tight and knotty strips, dark brown with moderate gloss	85.33	Bright red	89.67	Sweet aroma	85.67	Mellow	84.67	Mixed leaves, relatively even and intact	80.33	85.15
GC	Tight and knotty strips, dark brown with good gloss	88.00	Bright red	90.33	Rich and high-pitched fruity aroma with sweet notes	92.33	Mellow with distinct fruity flavor	91.67	Relatively even red color, relatively even and intact	90.00	90.61
b) Changes in Physicochemical Components of CK and GC Black Tea
	Total Phenolic(mg/g)		Amino Acid(mg/g)		Soluble Sugar(mg/g)		Theaflavin(mg/g)			Caffeine(mg/g)
CK	134.34 ± 1.48		32.88 ± 1.64		58.68 ± 1.10		832.79 ± 5.08			36.26 ± 0.10
GC	132.19 ± 7.09		28.26 ± 1.47*		60.21 ± 0.97		903.26 ± 1.97**			37.71 ± 0.17**

Note: CK, Control group black tea; GC, Treatment group black tea.“**”indicates P < 0.01,“*”indicates P < 0.05.

The effects of mixed fruit juice addition on the physicochemical components of summer and autumn black tea are presented in Table 2b. The contents of total phenols and soluble sugars exhibited no significant differences between the two groups. In contrast, the amino acid content in the mixed fruit juice-treated group (GC) was significantly lower than that in the control group (CK) (P < 0.05). Meanwhile, the GC group demonstrated extremely significant increases in theaflavin and caffeine contents compared to the CK group (P < 0.01).

3.2. Effects of mixed fruit juice addition on the aroma components of summer-autumn black tea

3.2.1. Qualitative analysis of aroma components

According to the peak retention time, peak area value and other related parameters of these aroma components, the aroma fingerprint of fruity black tea (Fig. 1A) was established, and there were significant differences between the aroma fingerprints of CK and GC groups. PCA showed (Fig. 1B) that the GC group of fruity black tea samples and the CK group of black tea samples had a clear separation trend in the first principal component (PC1 = 49%). This result indicates that the CK group, there were significant differences in the aroma components in the GC group. A total of 316 volatile metabolites were detected, including 75 alkanes, 41 alkenes, 33 alcohols, 32 esters, 30 ketones, 19 aldehydes, 13 pyrazines, 11 aromatics, 7 phenols, 5 furans, 3 acids, 2 ethers, and 45 others. The relative content of each component is shown in Fig. 1D. Compared with CK group, the relative content of alcohols in GC group decreased from 55% to 44%, alkanes decreased from 19% to 11%, and aldehydes decreased from 4% to 1%; The relative content of olefins in GC group increased from 0.79% to 24%, ethers from 0.02% to 5.76%, aromatics from 2.6% to 5%, and other substances did not change significantly.

Fig. 1.

Changes in volatile components of CK and GC black tea. (A) Fingerprint profiles of volatile components. (B) PCA plot of volatile components. (C) Composition of total volatile metabolites. (D) Stacked plot of relative contents of volatile components in each group. CK: Control group black tea; GC: Treatment group black tea.

3.2.2. Differential analysis of aroma components

Through the difference analysis of all aroma components, and taking VIP > 1, P < 0.05 as the screening criteria, 92 kinds of differential substances were screened (Table 3), including 10 kinds of alcohols, 7 kinds of aldehydes, 10 kinds of ketones, 8 kinds of esters, 1 kind of ethers, 1 kind of acids, 6 kinds of phenols, 2 kinds of furans, 13 kinds of alkanes, 19 kinds of alkenes, 7 kinds of pyrazines and 8 kinds of others. Compared with CK group, 66 kinds of aroma components were significantly increased in GC group. Among them, 32 compounds were the unique aroma components of GC group, and 26 compounds were significantly reduced, of which 13 were the unique aroma components of CK group. At the same time, GC group increased significantly in 6 phenolic compounds and 19 alkenes. Among the 66 new aroma components, the main aroma characteristics were woody, fruity, floral, spicy and so on. In addition, when | log2FC | > 5, eight aroma compounds were screened out, which were p-cymene-8-ol, p-methylacetophenone, 3-methylbutyrolactone, myristicin, p-cresol, α-phellandrene, myrcene and p-mentha-1,3,8-triene. Among them, only the content of 3-methylbutyrolactone was significantly decreased in GC group, while the other seven were significantly increased (P < 0.05).

Table 3.

Differential aroma components identified in CK and GC black tea and its Odor descriptions.

Name	CAS Number	CK(μg/Kg)	GC(μg/Kg)	Log2FC	P.value	VIP	Regulation	Odor Description
1-Penten-3-ol	616–25-1	75.74 ± 10.00	–	–	–	–	Down	Fruity, vegetable, horseradish
2-Ethyl-1-hexanol	104–76-7	7.95 ± 1.00	–	–	–	–	Down	Rose floral
2-Heptanol	543–49-7	28.31 ± 3.70	96.52 ± 3.69	1.77	0.00	1.42	Up	Citrus, oily/fatty
(E)-2-Hexen-1-ol	928–95-0	129.26 ± 15.18	185.00 ± 13.89	0.52	0.01	1.31	Up	Green, fruity
(Z)-2-Penten-1-ol	1576-95-0	95.40 ± 16.15	484.78 ± 29.07	2.35	0.00	1.42	Up	Fruity aroma
3,7-Dimethyl-1,5,7-octatrien-3-ol	29,957–43-5	8.23 ± 0.67	–	–	–	–	Down	Sweet floral
α-Terpineol	98–55-5	4.83 ± 0.49	9.49 ± 0.64	0.97	0.00	1.40	Up	Citrus, lilac floral, woody
p-Cymen-8-ol	1197-01-9	1.35 ± 0.14	274.18 ± 28.74	7.67	0.00	1.43	Up	–
Eudesmol	6750-60-3	–	4.01 ± 0.61	–	–	–	Up	Eucalyptus,herbaceous, green
(−)-Eudesmol	77,171–55-2	–	0.26 ± 0.10	–	–	–	Up	–
5-Formyl-7-methoxy-1,3-benzodioxole	5780-07-4	–	0.17 ± 0.03	–	–	–	Up	–
2-Methylbutanal	96–17-3	102.62 ± 12.91	62.73 ± 3.29	−0.71	0.00	1.37	Down	Coffee, roasted, floral
4-Ethylbenzaldehyde	4748-78-1	0.30 ± 0.03	–	–	–	–	Down	Almond aroma
4-Propylbenzaldehyde	28,785–06-0	0.42 ± 0.05	–	–	–	–	Down	–
Hexanal	66–25-1	32.07 ± 3.79	50.29 ± 4.49	0.65	0.01	1.34	Up	Grassiness and apple fragrance
2-Pyrrolecarbaldehyde	1003-29-8	6.61 ± 0.71	11.50 ± 0.91	0.80	0.00	1.38	Up	Roasted, coffee
(E)-2-Hexenal	6728-26-3	31.30 ± 3.06	61.15 ± 4.60	0.97	0.00	1.40	Up	Fruity, green leaves, fresh aroma
2,2,6-Trimethylcyclohexanone	2408-37-9	13.93 ± 1.88	30.57 ± 7.87	1.13	0.01	1.32	Up	Honey aroma
2,3-Pentanedione	600–14-6	8.41 ± 1.14	5.82 ± 0.54	−0.53	0.02	1.27	Down	Caramel fragrance
2-Methyltetrahydrofuran-3-one	3188-00-9	11.71 ± 1.16	4.83 ± 0.63	−1.28	0.00	1.40	Down	–
2-Pyrrolidone	616–45-5	0.75 ± 0.10	1.56 ± 0.13	1.06	0.00	1.39	Up	–
3,4-Dimethyl-1,5-dihydro-2H-pyrrol-2-one	4030-22-2	17.02 ± 2.94	46.81 ± 4.61	1.46	0.00	1.40	Up	–
(3E,5E)-3,5-Octadien-2-one	30,086–02-3	–	1.26 ± 0.16	–	–	–	Up	Apple fragrance
Butyl phenyl ketone	1009-14-9	1.66 ± 0.15	–	–	–	–	Down	–
Methyl heptenone	110–93-0	12.96 ± 1.61	17.79 ± 0.42	0.46	0.01	1.30	Up	Fruity notes, fresh and clean scent
p-Methylacetophenone	122–00-9	0.21 ± 0.03	10.80 ± 0.84	5.68	0.00	1.43	Up	–
Piperitone	491–09-8	–	0.36 ± 0.08	–	–	–	Up	–
(−)-Bornyl acetate	5655-61-8	0.78 ± 0.18	–	–	–	–	Down	Green, sweet with coolness, herbal aroma
3-Methylbutyrolactone / Valerolactone	1679-49-8	1.01 ± 0.12	0.02 ± 0.02	−5.81	0.00	1.41	Down	–
Ethyl palmitate	628–97-7	0.13 ± 0.04	0.46 ± 0.08	1.80	0.00	1.36	Up	Oily, creamy
Methyl acetate	79–20-9	22.14 ± 4.48	34.13 ± 3.82	0.62	0.03	1.23	Up	–
Methyl palmitate	112–39-0	0.86 ± 0.15	2.05 ± 0.30	1.26	0.00	1.37	Up	Iris, oily
Ethyl butyrate	123–86-4	–	3.19 ± 0.57	–	–	–	Up	Pineapple aroma
Triisobutyl orthoformate	16,754–49-7	3.59 ± 0.93	–	–	–	–	Down	–
Trans-2-hexenyl 2-methylbutyrate	–	–	8.75 ± 0.85	–	–	–	Up	–
Myristicin	607–91-0	0.21 ± 0.03	1136.20 ± 164.37	12.39	0.00	1.43	Up	Spicy, woody
N,N′-Dimethylterephthalic acid	5176-82-9	0.21 ± 0.05	0.32 ± 0.04	0.65	0.03	1.21	Up	–
2,4-Di-tert-butylphenol	96–76-4	10.76 ± 1.28	26.47 ± 4.20	1.30	0.00	1.39	Up	Phenolic aroma
4-Vinylphenol / p-Vinylphenol	2628-17-3	0.23 ± 0.03	0.40 ± 0.04	0.84	0.00	1.37	Up	–
Carvacrol	499–75-2	–	2.58 ± 0.29	–	–	–	Up	Coriander, spice, thyme
Methyleugenol	93–15-2	–	13.74 ± 2.10	–	–	–	Up	Woody, clove, spicy
p-Cresol	106–44-5	0.11 ± 0.01	4.11 ± 0.38	5.23	0.00	1.43	Up	Floral
Phenol	108–95-2	1.75 ± 0.12	2.96 ± 0.25	0.75	0.00	1.39	Up	Phenolic
2,3-Dihydrofuran	1191-99-7	–	1.65 ± 0.39	–	–	–	Up	–
2-Pentylfuran	3777-69-3	34.40 ± 4.31	–	–	–	–	Down	Sweet
2,5-Dimethylundecane	17,301–22-3	–	38.70 ± 3.75	–	–	–	Up	–
2,6-Dimethylnonane	17,302–28-2	103.92 ± 14.14	48.99 ± 4.66	−1.08	0.00	1.39	Down	–
2-Methylheptane	592–27-8	2.39 ± 0.37	1.58 ± 0.13	−0.59	0.02	1.28	Down	–
3-Ethylpentane	617–78-7	–	0.53 ± 0.12	–	0.00	1.43	Up	–
5-Ethyldecane	17,302–36-2	13.89 ± 1.22	39.88 ± 3.33	1.52	0.00	1.42	Up	–
5-Methyltridecane	25,117–31-1	2.20 ± 0.20	8.20 ± 0.83	1.90	0.00	1.42	Up	–
3-Methyldecane	13,151–34-3	2.47 ± 0.26	5.44 ± 0.48	1.14	0.00	1.40	Up	–
2,4-Dimethylheptane	2213-23-2	30.87 ± 4.46	17.74 ± 1.40	−0.80	0.00	1.35	Down	–
n-Dodecane	112–40-3	134.61 ± 9.75	394.92 ± 26.97	1.55	0.00	1.42	Up	–
n-Octane	111–65-9	64.63 ± 11.21	43.57 ± 3.86	−0.57	0.03	1.24	Down	–
n-Tetradecane	629–59-4	12.52 ± 0.77	50.54 ± 6.13	2.01	0.00	1.42	Up	–
n-Tridecane	629–50-5	0.84 ± 0.04	–	–	–	–	Down	–
3-Methyltridecane	6418-41-3	8.91 ± 0.70	56.58 ± 8.33	2.67	0.00	1.42	Up	–
(−)-α-Pinene	7785-26-4	–	61.91 ± 4.85	–	–	–	Up	Warm resinous aroma and fresh pinewood scent
(E)-4-Methoxy-6-propenyl-1,3-benzodioxole	18,312–21-5	–	12.31 ± 1.94	–	–	–	Up	–
1-Decene	872–05-9	–	4.58 ± 0.44	–	–	–	Up	–
2-Ethyl-1-octene	51,655–64-2	4.97 ± 0.76	8.48 ± 0.78	0.77	0.01	1.34	Up	–
3-Isopropenyl-5,5-dimethylcyclopentene		–	3.40 ± 0.21	–	–	–	Up	–
α-Muurolene	489–28-1	–	3.18 ± 0.78	–	–	–	Up	–
α-Bergamotene	17,699–05-7	–	33.72 ± 3.51	–	–	–	Up	Lemon-like fruity aroma
α-Curcumene	644–30-4	–	2.87 ± 0.57	–	–	–	Up	Sichuan pepper fragrance, ginger-like sweet aroma
α-Phellandrene	99–83-2	1.07 ± 0.27	362.60 ± 38.41	8.40	0.04	1.31	Up	Orange aroma
α-Thujene	2867-05-2	–	8.31 ± 0.71	–	–	–	Up	Floral, woody
β-Bisabolene	495–61-4	–	236.70 ± 35.89	–	–	–	Up	Woody, citrus, floral and fruity
Trans-β-Farnesene	18,794–84-8	–	25.24 ± 4.37	–	–	–	Up	Woody, fruity, grassy, sweet
5-Decene	19,689–19-1	–	1.87 ± 0.20	–	–	–	Up	–
Terpinolene	586–63-0	–	1120.89 ± 1895.63	–	–	–	Up	Fresh woody, citrus
Myrcene	123–35-3	18.20 ± 2.22	921.34 ± 75.64	5.66	0.00	1.43	Up	Peppery, spicy
β-Ocimene	3779-61-1	–	53.81 ± 5.56	–	–	–	Up	Herbaceous sweet
p-α-Dimethylstyrene	1195-32-0	–	710.32 ± 84.91	–	–	–	Up	–
p-Mentha-1,3,8-triene	18,368–95-1	0.10 ± 0.09	47.32 ± 4.90	8.83	0.02	1.38	Up	Turpentine aroma, camphorwood scent, grassy note, woody fragrance
β-Sesquiphellandrene	20,307–83-9	–	6.39 ± 0.97	–	–	–	Up	Herbal aroma, woody scent, green note
2,3-Dimethylpyrazine	5910-89-4	4.22 ± 0.45	2.45 ± 0.17	−0.79	0.00	1.38	Down	Caramel fragrance
2,5-Dimethylpyrazine	123–32-0	27.31 ± 2.05	37.96 ± 6.43	0.48	0.04	1.19	Up	Nutty aroma
2-Ethylpyrazine	13,925–00-3	14.70 ± 1.69	7.49 ± 0.66	−0.97	0.00	1.39	Down	Nutty aroma
2-Ethyl-3-methylpyrazine	15,707–23-0	3.34 ± 0.32	–	–	–	–	Down	Corn-like fragrance
2-Ethyl-6-methylpyrazine	13,925–03-6	9.42 ± 0.99	2.98 ± 0.23	−1.66	0.00	1.42	Down	Nutty aroma
2-Methylpyrazine	109–08-0	42.32 ± 4.10	22.02 ± 1.82	−0.94	0.00	1.40	Down	Cocoa and nutty-like aroma
2,5-Diethylpyrazine	13,238–84-1	2.20 ± 0.20	–	–	–	–	Down	–
1,1,5-Trimethyl-1,2-dihydronaphthalene		–	3.41 ± 0.29	–	–	–	Up	–
2-Acetylpyrrole	1072-83-9	69.39 ± 3.51	54.03 ± 6.18	−0.36	0.02	1.24	Down	–
N,4,5-Trimethylbenzene-1,2-diamine	17,978–55-1	–	0.72 ± 0.62	–	–	–	Up	–
Acetoin	116–09-6	54.12 ± 6.72	171.87 ± 10.06	1.67	0.00	1.42	Up	–
Cosmene	460–01-5	–	5.67 ± 0.40	–	–	–	Up	–
Elemicin	487–11-6	–	5.02 ± 0.88	–	–	–	Up	Woody, green, herbal
Indole	120–72-9	–	0.24 ± 0.02	–	–	–	Up	Floral
p-(1-Propenyl)toluene	–	3.67 ± 0.42	–	–	–	–	Down	–

Note: In the Regulation, “Up” and “Down” refer to the changes in the GC group compared to the CK group; “--” indicates that the data was not found or detected; “**”indicates P. value <0.01,“*”indicates P. value <0.05.

3.2.3. Analysis of key aroma contributing compounds

The thresholds of 92 different compounds listed in Table 3 were searched, and the thresholds of 52 aroma compounds were found, and the OAV and ACI were calculated (Table S1).

OAV is a method to comprehensively evaluate the aroma contribution of each aroma component in a sample by combining the content of the aroma component and the odor threshold, and the degree of aroma contribution can be compared by the ACI value (Feng et al., 2022). It is generally believed that volatile compounds with OAV ≥ 1 contribute more to the overall aroma. Volatile compounds with OAV > 10 have been identified as important aroma components (Guo et al., 2022; Hu et al., 2022). The OAV and ACI of each aroma component were calculated by consulting the threshold values of aroma components in water in the relevant literature, and the aroma characteristics of the described aroma components were recorded. A total of 17 aroma components with OAV > 1 were identified from the two samples, and the aroma components with OAV > 10 were further screened. As shown in Table S1, there were 9 and 15 aroma components with OAV > 1 in CK and GC groups, respectively, and there were 2 and 6 aroma components with OAV > 10 in CK and GC groups, respectively.

In CK group, 2,2,6-trimethylcyclohexanone (OAV = 139.315, ACI = 52.256%) contributed the most to the aroma of black tea in CK group, followed by 2-methylbutyraldehyde (OAV = 102.615, ACI = 38.490%), hexanal (OAV = 7.126, ACI = 2.673%), 2-pentylfuran (OAV = 5.733, ACI = 2.150%),These substances have the characteristics of flower fragrance, honey fragrance, apple fragrance and sweet fragrance respectively, which are the standard aroma characteristics of ordinary black tea. In GC, 2,2,6-trimethylcyclohexanone (OAV = 305.659, ACI = 57.748%) contributed the most to the aroma of black tea in GC group, and the content of trimethylcyclohexanone in GC group was significantly higher than that in CK group, followed by 2-methylbutanal (OAV = 62.732, ACI = 11.Myrcene (OAV = 61.422, ACI = 11.604%), Myristicin (OAV = 37.873, ACI = 7.155%), Acetoin (OAV = 12.277, ACI = 2.319%), Hexanal (OAV = 11.176, ACI = 2.111%). Myrcene, myristicin and acetoin were the only aroma compounds in GC group, among which myrcene and myristicin were the main contributors to the “fruity aroma” of fruity black tea with the characteristics of spicy and woody aroma.

There were many common key aroma compounds in the two groups of black tea, such as 2,2,6-trimethylcyclohexanone, 2-methylbutyraldehyde, hexanal, etc., but due to the different OAV and ACI values of the key aroma components, the black tea of CK group and GC group showed different aroma characteristics. The OAV of 2-methylbutyraldehyde in the CK group was higher than that in the GC group, which made the black tea in the CK group not only have the characteristics of floral aroma, but also have the characteristics of honey aroma; the OAV of 2,2,6-trimethylcyclohexanone and hexanal in the GC group was higher than that in the CK group, which made the black tea in the GC group not only have the characteristics of spicy aroma, woody aroma, but also have the characteristics of floral aroma and apple aroma. The aroma characteristics of CK and GC groups were consistent with the sensory evaluation.

3.3. Effects of mixed fruit juice addition on the metabonomics of summer-autumn black tea

3.3.1. Qualitative analysis of metabolites

Metabolite profiles of two black teas, CK and GC, were displayed by heat mapping (Fig. 3A, B), and the results showed that the 3 replicate samples of each group were clustered together with good uniformity and high reliability. A total of 915 metabolites were identified in fruity black tea, including 172 flavonoids and isoflavones and 144 alkaloids, 143 terpenoids, 66 amino acids and short peptides, 63 polyketones, 46 fatty acids, 45 carbohydrates, 41 coumarins, 37 lignans, 33 cinnamic acids and derivatives, 27 phenolic acids and 98 other types of compounds (Fig. 2A). The average relative contents of 12 metabolites of the two black teas are shown in Fig. 2B. The relative contents of flavonoids and isoflavones of the two black teas are the highest, followed by amino acids and short peptides, alkaloids, cinnamic acid and its derivatives. The relative contents of flavonoids and isoflavones in GC group were significantly higher than those in CK group, while the relative contents of carbohydrates and lignans in GC group were significantly lower than those in CK group (P < 0.05). The remaining nine compounds showed no significant differences between the two black tea groups.

Fig. 3.

Analysis of differential metabolites. (A and B) Systematic clustering heatmap of 42 differential non-volatile metabolites in CK vs. GC; (C) Pathway enrichment analysis of differential metabolites in CK vs. GC.

Fig. 2.

Changes in non - volatile components of CK and GC black tea. (A) Composition of total non - volatile metabolites; (B) Stacked plot of relative contents of non - volatile components in each group; (C) PCA of total non - volatile metabolites; (D) OPLS - DA of total non - volatile metabolites; (E) Permutation test of OPLS - DA; (F) Validation plot of OPLS - DA model.

3.3.2. Multivariate analysis of metabolites in CK and GC black tea

PCA and OPLS-DA were used in this study to discriminate between all samples based on all metabolites. The PCA results (Fig. 2C) showed that PC1 and PC2 explained 47.1% and 17.2% of the total variables, respectively, indicating a difference between the two black teas, CK and GC. In order to further explore the differential metabolic components in the two black teas, a multivariate statistical analysis method with supervised pattern recognition, OPLS-DA model (biological repetition≥3), was established to obtain VIP. Among the prediction parameters of the OPLS-DA evaluation model, R2X and R2Y represent the explanation rate of the established model to the X and Y matrices, respectively, and Q represents the prediction ability of the model.The closer the values of these three indicators are to unity, the greater the model's stability and reliability (Yun et al., 2021). As shown in Fig. 2D, the abscissa represents the score of the major metabolite, reflecting the difference between groups. The ordinate represents the score of the orthogonal component, indicating the within-group difference. Clearly, the OPLS-DA score plot showed significant between-group differences in metabolites between the CK and GC samples. The permutation test shows that the model is not overfitted (all the red Q2 points are lower than original blue R2 points from left to right, and the intercept of the Q2 regression line is less than 0) (Fig. 2E). From Fig. 2F, it can be seen that Q2 is 0.952, R2Y is 1, P < 0.05, indicating that the model is reliable. The differential metabolites between CK and GC black tea samples can be screened according to VIP value analysis.

3.3.3. Analysis of characteristic differential compounds

VIP score can be used as a potential indicator for screening differential metabolites (Feng et al., 2024). It is generally believed that VIP > 1 indicates that the compound has a higher contribution rate to differential discrimination, that is, the higher the VIP, the more important the compound is to the sample. Based on VIP > 1.47 and P ≤ 0.05, 41 different compounds were identified, including amino acids and short peptides (6), flavonoids (6), polyketones (2), alkaloids (7), carbohydrates (6), terpenoids (2), fatty acids (4), cinnamic acid and its derivatives (5), and other compounds (3), and regarded them as characteristic metabolites between the CK and GC groups (Table 4). In order to better visualize the metabolic differences among different black teas, cluster analysis was performed on 41 metabolites (Fig. 3A) to explore their interrelationships and relative contents, and the CK and GC group samples showed different clustering patterns, which was consistent with the results of multivariate statistical analysis.

Table 4.

Differential metabolites identified in black tea of CK and GC groups by UPLC-MS/MS.

Compounds	CAS	RT	VIP	P.Value	Fold Change
Amino acids and short peptides (6 types)
4-Aminobutyric acid	56-12-2	47.92	1.49	0.000	1.78
Histidine	71-00-1	52.11	1.50	0.000	0.64
Ac-DL-Trp-OH	87-32-1	279.45	1.47	0.000	0.50
Aurantiamide acetate	56,121-42-7	386.34	1.49	0.000	2.92
Cilastatin	82,009-34-5	240.77	1.48	0.000	0.60
Nepetalinic acid	–	315.60	1.50	0.000	3.21
Flavonoids (6 types)
Apigenin	520-36-5	341.24	1.49	0.000	1.18
Vitexin	3681-93-4	266.83	1.49	0.000	1.27
Eriocitrin	13,463-28-0	235.25	1.47	0.001	1.53
Cosmosiin	578-74-5	266.64	1.49	0.000	1.24
Theaflavin	1,011,850	294.66	1.49	0.000	1.32
4′,5,7-Trihydroxy-3,6-dimethoxyflavone	22,697-65-0	326.81	1.48	0.001	1.21
Polyketides (2 types)
3-Butylidenephthalide	551-08-6	46.66	1.49	0.000	0.76
Tanshinone IIA	568-72-9	451.15	1.49	0.005	1.26
Alkaloids (7 types)
Creatinine	60-27-5	50.58	1.47	0.001	0.60
N-Acetylputrescine	18,233-70-0	50.73	1.50	0.000	0.59
Anabasine	13,078-04-1	47.96	1.47	0.000	0.58
Higenamine	106,032-53-5	192.51	1.48	0.000	0.30
Pyropheophorbide-a	24,533-72-0	522.31	1.50	0.000	0.59
N-Formylcytisine	53,007-06-0	193.49	1.47	0.001	0.51
Methylharmine	21,236-68-0	231.68	1.48	0.000	0.23
Carbohydrates (6 types)
1-Methyladenosine	15,763-06-1	50.44	1.49	0.000	0.69
1-Deoxynojirimycin	19,130-96-2	47.06	1.49	0.000	0.55
Pseudouridine	1445-07-4	49.00	1.48	0.000	1.75
Aristeromycin	19,186-33-5	47.61	1.50	0.000	1.73
Palatinose	13,718-94-0	49.97	1.49	0.000	0.67
5′-Methylthioinosine	17,298-58-7	223.66	1.49	0.000	0.77
Terpenoids (2 types)
Lamiidoside	64,597-22-4	265.86	1.48	0.001	0.52
16,17-Dihydroxy-7-kauranone	145,701-00-4	401.38	1.48	0.000	1.46
Fatty acids (4 types)
Traumatic acid	6402-36-4	358.84	1.49	0.000	1.68
Dodecanedioic acid	693-23-2	369.77	1.48	0.000	1.34
Cyclopentaneacetic acid	39,924-52-2	370.92	1.49	0.000	1.85
Pinellic acid	97,134-11-7	362.30	1.49	0.000	1.30
Cinnamic acid and its derivatives (5 types)
Coumarin derivatives	53,771-63-4	344.46	1.49	0.000	14.99
Coniferaldehyde	20,649-42-7	280.89	1.49	0.000	1.68
5-Caffeoylshikimic acid	73,263-62-4	248.61	1.48	0.001	0.59
Elemicin	487-11-6	367.79	1.50	0.001	30.91
1-Caffeoylquinic acid	1241-87-8	217.90	1.47	0.000	0.59
Others (3 types)
N-Trimethyllysine	19,253-88-4	48.45	1.50	0.000	0.82
Hispidin	56,070-89-4	284.78	1.49	0.000	0.50
Asperphenamate	63,631-36-7	410.89	1.49	0.000	1.73

Note: In the Fold Change, refer to the changes in the GC group compared to the CK group; “***”indicates P. value <0.001, “**”indicates P. value <0.01.

The screened differential metabolites are also the main components that determine the quality of black tea, including its flavor and nutritional value. The relative contents of all flavonoids (6 species), fatty acids (4 species) were significantly increased and alkaloids (7 species) were significantly decreased in GC group compared with CK group (P < 0.01). In addition, 4-Aminobutyric Acid, Aurantiamide acetate and Nepetalinic acid were significantly increased compared with CK group (P < 0.001). In addition to the above metabolites, 21 other metabolites, including amino acids and short peptides (3), polyketones (2), carbohydrates (6), terpenes (2), cinnamic acid and its derivatives (5), and other compounds (3), still play an important role in distinguishing the two black teas. Among them, Asperphenamate, Pseudouridine, Coniferaldehyde, Aristeromycin, Elemicin, Coumarin deriv, Tanshinone LLA, 16,17-Dihydroxy-7-kauranone increased significantly (P < 0.01) compared with CK group, while the other 13 compounds showed the opposite trend. A comprehensive analysis of 41 characteristic metabolites showed that, 4-Aminobutyric acid, Nepetalinic acid, Theaflavin, Aristeromycin, Coumarin deriv, Elemicin and Hispidin contributed most. Because they have higher VIP values above 1.49 and significant fold change.

3.3.4. Metabolic pathway analysis

Significantly different metabolites were further localized to metabolic pathways by performing KEGG enrichment analysis, as shown in Fig. 3C. In the comparison of CK and GC, we observed that different metabolites were mainly involved in Flavone and flavonol biosynthesis, Flavonoid biosynthesis, beta − Alanine metabolism. In addition, other metabolic pathways were identified, including Histidine metabolism, Aminoacyl − tRNA biosynthesis, Nucleotide metabolism, and Phenylpropanoid biosynthesis.

4. Discussion

Black tea, as the most consumed tea in the world, accounts for 80% of the total consumption of tea in the world. Its liquor color is bright red, its taste is mellow, and its aroma is fragrant, so it is deeply loved by consumers (Liu et al., 2021). Aroma and taste are important evaluation indexes for the classification of black tea. Aroma, as the result of the comprehensive action of different types and proportions of volatile organic compounds, can affect the intuitive judgment of the drinker on the quality of tea products (Liu et al., 2018). In this study, mixed fruit juice of Siraitia grosvenorii and Shanhuangpi was added during the fermentation of black tea, so as to enrich the aroma and taste qualities of summer-autumn black tea. The metabolic fingerprint of the fruit-scented black tea is established by HS-SPME-GC–MS and UPLC-MS/MS, and the differences in non-volatile and volatile components of the fruit-scented black tea are explored.

Through volatile metabolomics analysis, a total of 316 aroma components were identified, among which 92 aroma components showed obvious differences. These substances mainly included olefins, alkanes, alcohols, aldehydes, ketones and esters. It is worth noting that compared with the CK group, the relative content of olefins in the GC group increased by 30 times; the content of ethers increased by 288 times, among which myristicin changed most significantly. Myristicin exists in a variety of foods, herbs and essential oils (He et al., 2024) and has multiple pharmacological activities, including anti-inflammatory, antibacterial and anti-proliferative activities (Zhu et al., 2019). The changes in these aroma substances are closely related to the addition of Shanhuangpi fruits. In our previous study, it was found that fresh Shanhuangpi fruits contain rich aroma-active olefins and ethers, among which myristicin is present (Luo et al., 2023).

Although the most dominant aroma compound in two black tea groups was 2,2,6-trimethylcyclohexanone (CK-ACI: 52.256; GC-ACI: 57.748), characterized by honey-like aroma (Wang, Lei, et al., 2024; Wang, Liu, et al., 2024), the ACI of only two compounds—2,2,6-trimethylcyclohexanone and 2-methylbutanal—in the CK group reached as high as 90.75%. In contrast, the GC group developed a complex aroma profile dominated by honey-like notes, supplemented by floral, spicy, and woody notes, with the ACI of its top five key aroma compounds also reaching 90.75%, which was generally consistent with the results of sensory evaluation.

A total of 915 metabolites were identified via non-targeted non-volatile metabolomics analysis, from which 41 differential metabolites were screened. The analysis revealed that all six flavonoids showing significant changes were markedly increased in the GC group. Among them, vitexin and cosmosiin are both glycoside derivatives of apigenin, exhibiting antiviral, antioxidant, and anticancer activities in vitro and in vivo (An et al., 2015). Eriocitrin, a disaccharide derivative widely distributed in citrus fruits, possesses strong antioxidant capacity and is commonly used as an antioxidant in beverages, foods, and wines (Li et al., 2019; Yao et al., 2022). Given the similarity in nutritional composition between wampee fruits and citrus fruits, eriocitrin is presumed to originate from wampee. Notably, the content of theaflavins—regarded as the “golden molecules” of black tea—was significantly elevated in the GC group. This not only enhanced the thickness, mellow taste and freshness of the tea infusion (He, 2017; Miao et al., 2021), but also strengthened its health-promoting functions such as hypolipidemic and antioxidant effects (Liu et al., 2022; Tan et al., 2019; Wu et al., 2018). The increase in theaflavins is speculated to result from exogenous active components enhancing the activity of polyphenol oxidase and improving cell membrane permeability (Xue et al., 2019). In addition, the contents of coumarin derivatives and elemicin, both derived from wampee (Qin et al., 2017; Wang et al., 2010), surged by 14.99-fold and 30.91-fold in the GC group, respectively. It is noteworthy that while mogrol (CAS No. 88930–15-8), derived from Siraitia grosvenorii, does not fall within the predefined VIP screening criteria (with a VIP value of 1.322), its concentration has nonetheless markedly increased by 13.84-fold (P. value = 0.0000858), rendering it a pivotal differential metabolite. Meanwhile, the content of γ-aminobutyric acid (GABA), which exerts neuroprotective effects and improves liver function (Silva et al., 2019), reached 1.78 times that of the CK group, further enriching the functional value of the product.

In brief, regarding volatile metabolites, the GC group of black tea exhibits a richer aroma compared to the CK group, characterized predominantly by a honey aroma, complemented by floral, spicy and woody aromas. As for non-volatile metabolites, the GC group exhibits notably higher levels of multiple health-functional active ingredients compared to the CK group, particularly theaflavin, γ-aminobutyric acid, coumarin deriv and elemicin. The odor activity value (OAV), utilized in research, functions as an evaluation tool for the perceived intensity of aroma compounds, facilitating the swift identification of potential key aroma compounds. However, the thresholds employed in this method are based on reported literature, and the neglect of matrix effects, interactions between aroma substances, and the non-linear response characteristics of human olfaction may introduce certain deviations in the results. When confirming the aroma contribution of a compound, further validation through recombination (addition and omission tests) experiments may be considered. Furthermore, this study also has some limitations. On the one hand, a single-factor control experiment was not conducted, making it impossible to analyze the impact of each added component on black tea individually; on the other hand, the added concentrated juices were all prepared by boiling at high temperatures, potentially inactivating the rich enzyme systems in the pulp and peel and preventing them from participating in the fermentation process of black tea. In future studies, attention should be focused on the impact of exogenous enzyme systems and exogenous active ingredients in fruits on the key quality components during black tea processing, so as to improve the quality of summer-autumn teas.

5. Conclusion

This study investigated the effects of adding mixed fruit juice (from Siraitia grosvenorii and fruit of Clausena anisum-olens (Blanco) Merr.) during processing on the aroma and overall quality of summer-autumn black tea. A comprehensive evaluation was conducted through sensory assessment, biochemical composition analysis, and both volatile and non-volatile metabolomics. The addition of mixed fruit juice significantly enhanced the sensory quality of the summer-autumn black tea, concurrently increasing theaflavin and caffeine contents, whereas the total free amino acid content decreased significantly. The volatile compounds exhibited a 30-fold surge in the relative content of alkenes and a remarkable 288-fold increase in ethers, predominantly myristicin. Odor activity value (OAV) analysis pinpointed 17 potential key aroma compounds, primarily characterized by honey, floral, woody, and spicy notes. Non-volatiles including apigenin, vitexin, cosmosiin, γ-aminobutyric acid, coumarin deriv, and elemicin were significant increased. The results not only provides practical guidance for the production of high-quality summer-autumn black tea products but also establishes a theoretical foundation and actionable strategy for the high-value utilization of underutilized summer and autumn tea resources.

CRediT authorship contribution statement

Baogui Liu: Writing – original draft, Investigation, Formal analysis, Data curation. Mengfan Niu: Writing – original draft, Methodology, Formal analysis. Yunxian Wang: Investigation. Renyuan Jia: Investigation. Hongchun Cui: Investigation. Yunxiong Zhao: Methodology, Investigation, Conceptualization. Jingjing Yang: Investigation. Zongyou Wei: Investigation. Guangzhi Liang: Investigation. Zhihui Feng: Writing – review & editing, Formal analysis. Junfeng Yin: Supervision, Funding acquisition. Lianfeng Luo: Supervision, Funding acquisition, Conceptualization.

Ethical statements

Ethical permission was not required. No vulnerable populations were involved, and no personally identifiable information was collected or disclosed. The rights and privacy of the participants were fully protected, participation was not forced, and the research requirements and risks were fully disclosed before sensory evaluation, with informed consent being obtained from all participants.

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.

Appendix A. Supplementary data

Supplementary material

Data availability

Data will be made available on request.