Volatile metabolomics highlights tea trichome's positive contribution to aroma and quality of white tea

Abstract

Tea trichomes, rich in secondary metabolites, are hypothesized to significantly influence white tea aroma, yet their biochemical contributions remain unclear. This study investigated the volatile profiles of tea trichome and tea body across six white tea samples using headspace solid-phase microextraction combined with gas chromatography-mass spectrometry (HS-SPME-GC-MS) and multivariate statistical analysis. Comparative analysis revealed significant differences in the volatile profile between tea trichome and tea body in all tested samples, with major differences lying in the content of volatile components. Sixteen potential key aroma components with relative odor activity value (ROAV) ≥ 1 and 39 aroma modifiers (0.1<ROAV<1) were identified in both tea trichome and tea body, predominantly associated with green, fruity, sweet, floral, woody and herbal aroma. In brief, trans-β-ionone, β-damascone, linalool, non-8-enal and (E,Z)-2,6-nonadienal were identified as key aroma components in both tea trichome and tea body, determining the basic aroma characteristic of white tea. Aroma modifiers were also affirmed, including β-myrcene, 2-ethyl-3,5-dimethyl-pyrazine, (E)-2-hexenal, dehydro-β-ionone, (Z)-6-nonenal, (E,E)-2,4-decadienal and benzeneacetaldehyde, which were shared by all samples in tea trichome and tea body. Notably, most of these key and modified aroma components showed higher relative contents and ROAVs in tea trichome than those in tea body. These data reveals that tea trichome positively enhance white tea's aroma quality, notably intensifying fruity, floral, sweet, green, and woody notes. This study systematically clarifies the biochemical specificity of tea trichome and their positive impact on white tea aroma, highlighting their critical role in aroma formation through differential accumulation of aroma contributors.

Graphical abstract

Highlights

• •Volatile profile in tea trichome differs greatly from that in tea body.

• •trans-β-Ionone, β-damascone and linalool were identified as key aroma components.

• •β-Myrcene, (E)-2-hexenal and benzeneacetaldehyde were identified as aroma modifiers.

• •The majority of aroma contributors showed higher content in tea trichome than in tea body.

• •Tea trichomes positively enhance the volatile profile and aroma quality of white tea.

1. Introduction

White tea, a delicately processed tea category, is highly valued for its unique aroma, mild taste, and health-promoting properties (Zhou et al., 2023). The most renowned types of white tea are silver needle (Baihao Yinzhen, bud only) and white peony (Baimudan, one bud and one or two leaves). A key characteristic of white tea, especially in high-quality types made from buds and tender leaves such as silver needle, is the abundance of fine trichomes or "tea hairs", which cover the buds and tender leaves (Cao et al., 2020; Song et al., 2017; B. Sun et al., 2020). Despite their microscopic size, tea trichomes are believed to play a significant role in the chemical composition and overall quality of white tea, which has sparked scientific interest in understanding their contributions (Li et al., 2020a; P. Li et al., 2020b).

Silver needle, composed entirely of young buds, is densely covered with trichomes, which is one of its defining features (Z. Chen et al., 2019a). These trichomes are thought to contribute significantly to its pekoe, floral, and sweet fragrance (Liu et al., 2024). Similarly, premium white peony (Mudan Wang, the best white peony), which comprises both bud and leaf, is covered with tea trichomes on its surface, too (Liu et al., 2024). Pekoe aroma is also one of the representative flavor characteristics of premium white peony with slight difference in its intensity compared with silver needle (C. Huang et al., 2021). Existing studies have demonstrated that trichomes in tea leaves contain lipophilic substances, phenolic compounds, and enzymes, all of which could potentially participate in the biosynthesis and transformation of flavor-related compounds (P. Li et al., 2020c; X. Wang et al., 2021). In terms of trichome-rich tea like white tea, trichomes may contribute to the pekoe flavor by storing or facilitating the release of aroma precursors and volatiles during processing and brewing (S. Wang et al., 2020; Yin et al., 2018). During tea processing and brewing, the trichomes can interact with environmental factors such as light, temperature, and moisture, to influence the release of aroma components. The similarities and differences in trichome density and structure between silver needle and premium white peony thus provides a unique model for studying the role of tea trichomes in volatile profiles and aroma formation.

Despite the importance of aroma in consumer preference and marketability, research on the specific contributions of tea trichomes to aroma formation remains limited. Most studies have focused on the general chemical composition of white tea or the influence of processing techniques, while the role of trichomes has been relatively overlooked. In addition, comparisons between different white tea types in this context are rare, leaving a gap in the understanding of the contribution of tea trichome to the nuanced aromatic differences between teas like silver needle and premium white peony tea. This study aims to investigate the influence of tea trichomes on the volatile profiles of silver needle and premium white peony tea. Headspace solid-phase microextraction combined with gas chromatography-mass spectrometry (HS-SPME-GC-MS) was conducted to elucidate the specific volatiles in tea trichomes and their potential contributions to the overall volatile profile of white tea. Furthermore, the study will explore the potential correlation between tea trichome and white tea aroma. Understanding the role of tea trichomes in aroma formation not only deepens our knowledge of the biochemistry of white tea but also provides valuable insights for tea industry.

2. Materials and methods

2.1. Materials

Two types of white tea including silver needle and premium white peony made of three varieties: Camellia sinensis cv. Fuandabaicha (abbreviated as FAB), Camellia sinensis cv. Fudingdabaicha (abbreviated as FDB) and Camellia sinensis cv. Fudingdahaocha (abbreviated as FDH) were chosen as experimental materials due to the hairy characteristic. All these tea samples were manufactured using traditional craft during the same period in spring in 2022 in Fujian Province.

The separation of tea leaves and tea trichome (Fig. 1) followed the method described in previous literature (Ye et al., 2010): The dried tea samples were ground into fine powder using a grinder (DFY-500C, Wenling Linda Machinery Co., Ltd, Zhejiang, China). Then, the ground tea powder was sieved through a 60-mesh sifter to isolate tea trichome from tea powder. In brief, tea trichome would remain on the top of sifter and the powder of tea body would pass through the sifter.

Fig. 1.

The white fuzz covered on the surface of tea is tea trichome.

A. FAB premium white peony sample (FABMDW); B. FDB premium white peony sample (FDBMDW); C. FDH premium white peony sample (FDHMDW); D. FAB silver needle sample (FABYZ); E. FDB silver needle sample (FDBYZ); F. FDH silver needle sample (FDHYZ).

2.2. Sensory quality evaluation

The sensory evaluation of six white tea samples was performed referring to the national tea evaluation standard (GB/T 23776-2018). A panel of six professional assessors (3 male and 3 female assessors, aged 25–35 years old with more than five years of sensory evaluation experience) were invited to evaluate the overall quality of these samples. In brief, 3.0 g of tea samples were infused with 150 mL boiling water in specialized white porcelain cup. After 5min of brewing, tea liquor was filtered to a bowl for sensory evaluation. According to the national standard, a 100-point grading system was employed to quantitatively assess the sensory attributes of six white tea samples, with scores as follows: appearance (25 %), liquor color (10 %), aroma (25 %), taste (30 %) and infused tea leaves (10 %). This study focused on the aroma and appearance attributes according to the main purpose of our research.

2.3. Extraction of VOCs from tea trichome and tea body

Briefly, 500 mg of the powder was transferred immediately to a 20 mL head-space vial (Agilent, Palo Alto, CA, USA), containing NaCl saturated solution (26.47 %, China National Medicines Corporation Ltd. Beijing, China) to inhibit any enzyme reaction. In addition, 20 μL internal standard solution (3-hexanone-2,2,4,4-d4,10 μg/mL, BioBioPha, Yunnan, China) with a concentration of 10 μg/mL was added in the vial. The vials were sealed using crimp-top caps with TFE-silicone headspace septa (Agilent). At the time of SPME analysis, each vial was placed in 60 °C for 5 min, then a 120 μm DVB/CWR/PDMS fibre (Agilent) was exposed to the headspace of the sample for 15 min at 60 °C.

2.4. GC-MS analysis

After sampling, desorption of the VOCs from the fibre coating was carried out in the injection port of the GC apparatus (Model 8890; Agilent) at 250 °C for 5 min in the splitless mode. The identification and quantification of VOCs was carried out using an Agilent Model 8890 GC and a 7000D mass spectrometer (Agilent), equipped with a 30 m × 0.25 mm × 0.25 μm DB-5MS (5 % phenyl-polymethylsiloxane) capillary column. Helium was used as the carrier gas at a linear velocity of 1.2 mL/min. The injector temperature was kept at 250 °C. The oven temperature was programmed from 40 °C (3.5 min), increasing at 10 °C/min to 100 °C, at 7 °C/min to 180 °C, at 25 °C/min to 280 °C, hold for 5 min. Mass spectra was recorded in electron impact (EI) ionisation mode at 70 eV. The quadrupole mass detector, ion source and transfer line temperatures were set at 150, 230 and 280 °C, respectively. The selected ion monitoring (SIM) mode was used for the identification and quantification of analytes.

2.5. Identification of volatiles in tea trichome and tea body

Based on multi-species data, literature references, some standard substances, and retention indices, a self-constructed database was established, containing determined retention time (RT) and qualitative and quantitative ions for SIM mode for accurate scanning. For each compound detected in this study, one quantitative ion and 2–3 qualitative ions were selected. All ions to be detected in each group were analyzed in specific time windows according to their elution order. If the detected retention time matched the standard reference and the selected ions appeared in the sample's mass spectrum after background subtraction, the substance was identified (Yuan et al., 2022). The quantitative ions were then used for integration and calibration to improve the accuracy of quantification.

2.6. Relative odor activity value analysis

The relative odor activity value (ROAV) analysis is a method that quantitatively assessed the contribution of volatile components to aroma quality by integrating their odor thresholds (Ren et al., 2025). In general, the ROAV of the volatile component that had the greatest contribution to the overall aroma was taken as 100, with the ROAVs of the other components calculated based on the following equation (Pang et al., 2025; Zhang et al., 2020):

$$\text{ROAV}=\frac{{\mathrm{C}}_{\mathrm{i}}}{{\mathrm{C}}_{\max }}\times \frac{{\mathrm{T}}_{\max }}{{\mathrm{T}}_{\mathrm{i}}}\times 100$$

Where corner mark i and max referred to the detected volatile components and the components with the maximum ROAV, respectively. C _i and C _max denoted the relative content of VOC i and max present in tea samples (%) which was relatively quantified by peak area normalization method. T _max and T _i indicated the threshold of volatile component i and max according to the published literatures (A. Sun et al., 2025).

According to previous study, volatile components with ROAV ≥1 were considered as the potential key aroma contributors, while those with 0.1<ROAV<1 were considered as modified volatile components. The higher the ROAV, the greater its contribution to the overall aroma. Others were regarded to have limited contribution to the overall aroma quality (A. Sun et al., 2025; Zeng et al., 2024).

2.7. Data processing

Multivariate statistical analysis, including principal component analysis (PCA), hierarchical clustering analysis (HCA), orthogonal partial least squares-discriminant analysis (OPLS-DA) and variable importance projection (VIP) were performed using the Metware Cloud, a free online platform for data analysis (https://cloud.metware.cn).

3. Results and discussion

3.1. Volatile profiles of tea trichome and tea body as analyzed by GC-MS

After peak alignment and quality control, entities obtained in tea trichome and tea body were submitted to unsupervised PCA and supervised OPLS-DA to investigate the differences between them. The results of PCA (Fig. 2A–F) allowed for visualization of differences between tea trichome and tea body, with the first two principals explaining almost 80 % of total variance in all six comparisons (tea trichomes vs tea body). Clear separation of tea trichome and tea body indicated markedly different volatile profiles between them regardless of variety and grade, which was consistent with previous research (J. Li et al., 2020a). Heatmap (Fig. 2G) was then conducted to visualize the overall difference in content of volatiles between tea trichome and tea body, suggesting remarkable difference of volatile profile of tea trichome and tea body as influenced by variety and grade. OPLS-DA model (Fig. S1) was also established to verify its reliability. All R²Y and Q² values of OPLS-DA model exceeded 0.98 and 0.86, respectively, indicating high accuracy and stability of these models. In another way, these models reliably explained the difference in volatiles between tea trichome and tea body across six samples.

Fig. 2.

Principal component analysis of volatiles in six sample: A. FABMDW; B. FDBMDW; C. FDHMDW; D. FABYZ; E. FDBYZ; F. FDHYZ. G. Heatmap of volatiles in tea trichome and tea body.

3.2. Identification and relative quantification of volatiles in tea trichome and tea body

A total of 1540 volatiles were finally identified in tea trichome and tea body according to self-created database, which was constructed based on standard substances, RI comparison and literature references. Among them, 1245 volatiles were with |ΔRI| ≤ 10. These volatiles could be grouped into the 13 subclasses: terpenoids, esters, ketones, heterocyclic compounds, alcohols, aldehydes, hydrocarbons, acids, aromatics, phenols, amines, ethers and others (Fig. 3A). Among these identified volatiles, terpenoids, esters, ketones, heterocyclic compounds and alcohols were dominant in terms of volatile number, accounting for 19.61 %, 17.98 %, 12.14 %, 10.52 % and 7.79 %, respectively. In general, volatiles in tea trichome and tea body shared great similarities in their subclasses with minor difference among varieties and tenderness. The result of relative quantitative analysis showed higher total peak area of volatile components in tea trichome than in tea body among five samples except FDBYZ sample (Fig. 3B).

Fig. 3.

A. Pie char of volatiles identified in six samples; B. Total peak area of volatiles in tea trichome and tea body in six samples, ∗p < 0.05, ∗∗p < 0.01.

Relative content of different types of volatiles in six samples: C. FABMDW; D. FDBMDW; E. FDHMDW; F. FABYZ; G. FDBYZ; H. FDHYZ.

In both tea trichome and tea body, terpenoids, esters, ketones and heterocyclic compounds constitute the predominant volatile components, collectively accounting for almost 60 % of the total volatile content in each sample. Terpenoids were the most abundant in both volatile number and relative content. Thereinto, trans-β-ionone, linalool, β-myrcene and β-damascone had been identified as important aroma volatiles in white tea and they were present in high relative abundances, which was in accordance with previous study (H. Chen et al., 2023; Q. Chen et al., 2019b; H. Wu et al., 2022). Alcohols, ketones and aldehydes were also important volatiles in tea. In this study, benzaldehyde, benzyl alcohol and 2-hexenal were important aroma contributors and they were at higher levels in both tea trichome and tea body in every sample.

3.3. Volatile profile differences between tea trichomes and tea body

3.3.1. Potential key aroma volatiles in tea trichome and tea body were identified

ROAV is a measure used in the field of flavor chemistry to evaluate the contribution of individual volatile compound to the overall flavor profile of product (Zhu et al., 2024). According to previous study, volatile components with ROAV ≥1 were considered as the potential key aroma contributors, while those with 0.1<ROAV<1 were considered as modified aroma components (A. Sun et al., 2025; Zeng et al., 2024). The magnitude of a compound's ROAV directly correlated with its influence on the overall aroma quality, with higher values indicating more significant sensory impact (W. Huang et al., 2022). This approach facilitated the identification and prioritization of key aroma-active volatiles that significantly shaped the olfactory characteristics of the tea samples under investigation.

In this study, the ROAV of volatiles with |ΔRI| ≤ 10 was calculated when their odor thresholds were available through literature retrieval. trans-β-Ionone was found to have the highest ROAV (set as 100) in all samples, indicating that it was among the most important aroma contributors in both tea body and tea trichome quality of white tea. In general, about 11–15 potential key aroma components with ROAV ≥1 were identified in tea body and tea trichome of six samples, respectively (Table S7). Notably, tea body and tea trichome of the same white tea sample exhibited nearly identical profiles of potential key aroma contributors as shown in supplementary material (Table S7). Thereinto, trans-β-ionone, non-8-enal, β-damascone, linalool and (E,Z)-2,6-nonadienal were identified in all six tea samples, suggesting their considerable role in the aroma formation of white tea. Considering their aroma characteristic, they might be the cornerstone of sweet, floral and green aroma quality, consistent with the previous research results (Liu et al., 2024; H Wu et al., 2024).

There were about 19–31 modified aroma components with 0.1<ROAV<1 identified in tea body and tea trichome of six samples, respectively (Table S7). Similarly, tea body and tea trichome exhibited nearly identical profiles of modified aroma components, whereas tea trichome demonstrated a greater number and higher ROAVs of these components (Table S7). Among these aroma modifiers, β-myrcene, 2-ethyl-3,5-dimethyl-pyrazine, (E)-2-hexenal, dehydro-β-ionone, (Z)-6-nonenal, (E,E)-2,4-decadienal and benzeneacetaldehyde were shared by all samples in tea trichome and tea body. They had significant impact on modifying the aroma of white tea in all samples. Additionally, the results suggested that tea trichome had more influence on the overall aroma due to higher ROAVs and more modified components. On the whole, the differences in key and modified aroma components lied in mostly ROAVs rather than volatile composition.

In summary, potential key aroma components with ROAV ≥1 in both tea body and tea trichome contributed to the dominant aroma quality, such as trans-β-ionone, non-8-enal, β-damascone, linalool and (E,Z)-2,6-nonadienal. While aroma modifiers with 0.1<ROAV<1 contributed to regulating the balance of the overall aroma quality, such as β-myrcene, 2-ethyl-3,5-dimethyl-pyrazine, (E)-2-hexenal, dehydro-β-ionone, (Z)-6-nonenal, (E,E)-2,4-decadienal. Our results revisited and re-evaluated previous research results, demonstrating that tea trichome contribute to the overall volatile profile considering the content of trichome and the content of volatiles in trichome (J. Li et al., 2020b).

3.3.2. Limited contribution of differential volatiles between tea trichome and tea body to the overall aroma quality

The primary distinction between tea trichome and tea body was in the content of volatiles with minor differences in their composition. Relative quantitative analysis revealed significant variations in the relative content of volatiles across six samples (Fig. 3B–H). Generally, the total peak area of volatile components was higher in tea trichome than in tea body, except for the FDBYZ sample. Notably, in FDBMDW, FABYZ and FDHYZ samples, the total peak area of volatile components was significantly or extremely higher (p < 0.05 or p < 0.01) in tea trichome than tea body. While in FDBYZ sample, the total peak area of volatile components was significantly or extremely lower (p < 0.05 or p < 0.01) in tea trichome than tea body. No significant differences were observed in the other samples. In terms of each subclass of volatiles, aromatics and acids were present at higher levels in tea trichome compared to tea body. Conversely, alcohols, esters and ethers exhibited lower contents in tea trichomes than in tea body across all examined samples. No consistent pattern was observed in the content of other volatiles between tea trichome and tea body.

To obtain key information responsible for the difference between tea trichome and tea body, differential VOCs were determined by VIP (VIP >1), fold change (FC) ≥ 2 or FC ≤ 0.5. A total of 56, 111, 99, 175, 102 and 128 differential VOCs were identified between tea trichome and tea body in FABMDW, FABYZ, FDBMDW, FDBYZ, FDHMDW and FDHYZ samples, respectively (Tables S1–6). Among the differential VOCs identified, some were present in both tea trichome and tea body, whereas others were exclusive to either tea trichome or tea body. The differential VOCs between tea trichome and tea body exhibited significant variability across six samples examined. Notably, the vast majority of these differential volatiles were not potential key aroma volatiles of tea in previous studies and were with ROAV ≤0.1 (Table S7) in this study, indicating their limited contribution to the overall aroma quality. It might suggest that the spatial differences in volatiles between tea trichome and tea body may not directly dominate the overall tea aroma quality. However, potential synergistic effects of these compounds needed further investigation to fully elucidate their roles.

3.3.3. Flavor omics analysis of differential VOCs between tea trichome and tea body

Even though, the ROAVs of these differential VOCs were below 0.1, the potential synergistic effects of these compounds might be influential to the overall aroma quality. They were screened and annotated with their aroma characteristics according to online database (https://www.thegoodscentscompany.com/index.html; https://flavornet.org/flavornet.html). A total of 136, 230, 249, 345, 250 and 280 flavor characteristics were annotated to differential VOCs identified in the comparison between tea trichome and tea body of FABMDW, FABYZ, FDBMDW, FDBYZ and FDHMDW samples. For each comparison between tea trichome and the tea body of six samples, the top ten aroma attributes with the highest number of differential VOCs were selected for the construction of a flavor wheel. As illustrated in Fig. 4, the most frequently annotated flavor attributes in the comparison of the tea trichome and tea body of the FABMDW sample were identified as sweet, floral, green, minty, and herbal. In the case of the FDBMDW sample (Fig. 4B), the predominant flavor attributes of the differential volatile organic compounds (VOCs) included green, fruity, waxy, sweet, and woody. Conversely, the FDHMDW sample (Fig. 4C) exhibited the most annotated flavor attributes as sweet, fruity, green, balsamic, and herbal. A similar trend was observed across the three silver needle samples (Fig. 4D, E, F). Notably, the most annotated flavor characteristics of these VOCs were sweet, floral, fruity green, herbal and balsamic, indicating a significant variation in these flavor characteristics attributed to the differences between tea trichome and tea body. These results corroborated the earlier finding, providing additional evidence for the potential contribution of trichomes to the pekoe and sweet aroma of white tea (Liu et al., 2024).

Fig. 4.

Flavor wheel of differential VOCs between tea trichome and tea body among six samples: A. FABMDW; B. FDBMDW; C. FDHMDW; D. FABYZ; E. FDBYZ; F. FDHYZ.

3.4. Tea grade significantly influences the volatile profile of tea trichomes

Different subtypes of white tea vary in their metabolites and flavor, as reported by previous investigations (L. Chen et al., 2024; Z. Chen et al., 2019a; C. Li et al., 2020c; Yang et al., 2018). The differences among different subtypes of white tea lie in the harvest time and tenderness of harvested leaves, with buds and younger leaves containing a higher abundance of tea trichome (J. Li et al., 2020a; Zheng et al., 2016). Moreover, the presence and density of trichome are considered as key features contributing to high-quality white tea (Yue et al., 2018). Despite this difference, the buds in premium white peony display morphological and flavor characteristics that closely resemble those of silver needle. Occasionally, these similar buds may be selectively separated and marketed as silver needle. Consequently, sensory evaluations revealed a shared pekoe aroma between them, accompanied by subtle variations in their fresh and floral notes (Table 1), which was in accordance with previous study (H. Chen et al., 2023). Based on these findings, it was proposed that the differences in tea trichome between silver needle and premium white peony may contribute to the variations in volatiles responsible for aroma quality to some extent, which was confirmed by the following chemometrics analysis.

Table 1.

Sensory evaluation of appearance and aroma characteristics of six samples.

Sample	Appearance	Score	Aroma	Score
FAB premium white peony	Whole shoot, fat bud, with base leaf, greyish green, slightly tippy	91	Pekoe scent, with strong honey flavor, lasting	94
FAB silver needle	Nearly thick and straight, tippy, dark greyish green, uniform, clean	91	Pekoe scent, a little floral	90
FDB premium white peony	Whole shoot, fat bud, tippy, greyish green, uniform, clean	92	Pekoe scent with strong honey flavor, a little floral, lasting	95
FDB silver needle	Fat bud, tippy, silvery, uniform, clean	95	Pekoe scent with honey flavor	94
FDH premium white peony	Whole shoot, fat bud, tippy, greyish green, uniform, clean	92	Pekoe scent with honey flavor, a little floral, lasting	95
FDH silver needle	Fat bud, tippy, white hair, nearly uniform, nearly clean	92	Strong pekoe scent	93

The trichomes of premium white peony exhibited higher total volatile peak area than tea bodies. Notably, silver needle and premium white peony displayed statistically significant differences (p < 0.01) in two out of three comparisons. (Fig. 5A). To analyze the overall volatiles differences in tea trichome between silver needle and premium white peony, PCA was conducted based on all volatiles identified. The first two principal components explained over 80 % of total variance in three comparisons (Fig. 5B, C, D). The clear distinction between silver needle and premium white peony indicated significant differences in volatile profile between them. To further find out key VOCs in tea trichome between silver needle and premium white peony, VIP (VIP >1), p < 0.05, and |Log₂FC| ≥ 1.0 were used as the parameters to filter information. A total of 281, 494 and 243 volatiles were identified as main contributors responsible for the difference in trichome between silver needle and premium white peony made by FAB, FDB, FDH, respectively. The majority of these differential VOCs showed higher relative content in trichome of premium white peony than silver needle. This discrepancy may be attributed to the fact that the premium white peony samples utilized in this study were made by the first-harvested leaves in 2022.

Fig. 5.

A. Total peak area of volatiles in tea trichome; B. PCA of tea trichome in silver needle (FABYZT) and premium white peony (FABMDWT); C. PCA of tea trichome in silver needle (FDBYZT) and premium white peony (FDBMDWT); D. C. PCA of tea trichome in silver needle (FDHYZT) and premium white peony (FDHMDWT).

To further evaluate the contribution of these differential VOCs to the overall flavor quality, the ROAV of them were calculated. Only a few volatiles with ROAV >0.1 were filtered, indicating limited contribution of tea trichome volatiles to the differences of tea aroma between premium white peony and silver needle. Particularly, three volatiles with ROAV ≥1, including smoky non-8-enal, floral linalool and roasted 2-thiophenemethanethiol, were identified as potential key aroma volatiles among differential volatiles in the FDH comparison. Moreover, non-8-enal, together with several modifiers such as fruity and grassy 1-octen-3-ol, green and melon (Z)-5-octen-1-ol, earthy 3-methoxy-2,5-dimethylpyrazine and waxy nonanoic acid were identified among differential volatiles in the FDB comparison. In FAB comparison, only three of 281 differential volatiles were certified to be aroma contributors with 0.1<ROAV<1, including grassy hexanal, musty 2,3-diethyl-5-methyl-pyrazine, fatty and melon 2,4-nonadienal. Interestingly, all these key and modified aroma components exhibited higher ROAVs in premium white peony than those in silver needle. However, the effective differential aroma contributors in tea trichome between premium white peony and silver needle were too limited to draw any conclusion. The main differences in tea trichome between two grades of tea might be the appearance.

3.5. Potential effects of tea trichome on the overall aroma quality of tea

Unlike the other categories of tea, white tea undergoes the least mechanical damaging so that the trichome on the surface of leaves are preserved to the maximum extent (Y. Liang et al., 1993). Pekoe aroma or "haoxiang" of white tea is often believed to be closely related with tea trichome, which is resulted by aldehydes and alcohols (Cao et al., 2020; Q. Chen et al., 2020; H. Lin et al., 2024). In this study, tea trichome was separated in a different way from previous study so that tea trichome was retained to the maximum extent (J. Li et al., 2020a). The results showed that tea trichome accounted for 10 % or more in entire silver needle and premium white peony samples (Table S11). Furthermore, our results revealed a substantial presence of volatiles in tea trichome, suggesting that trichome potentially play an influential role in determining tea aroma.

A total of 16 volatiles with ROAV exceeding 1 and 39 volatiles with 0.1<ROAV<1 were found in tea trichome and tea body, among which were annotated the most with green, fruity, sweet, floral, woody and herbal aroma. Similarly, absolute majority of them were shared by both tea trichome and tea body, indicating non-negligible contribution of tea trichome to the overall volatile profile. To be more specific, those potential key aroma components identified in white tea, such as trans-β-ionone, β-damascone, linalool, 2,4-undecadienal were identified with ROAV much greater than 1, some of which showed higher ROAVs in tea trichome (Fig. S7), confirming its significant role in determining the overall aroma quality. Furthermore, considering the proportion of tea trichome in tea leaves, the contribution of it to the overall aroma quality was equivalent to or greater than that of tea body due to more functional aroma components with higher ROAVs in tea trichome. It is worth noting that ROAV in this study serves as a comparative screening tool rather than a substitute for OAV for initial selection of potential odorants. GC-MS absolute quantification faces inherent limitations rooted in three technical barriers: availability of authentic reference standards, sample matrix interference, and insufficient sensitivity for trace analytes. Once these limitations are overcome, absolute quantitation will become more feasible, thus enabling the screening of key aroma components through OAV analysis.

The flavor omics analysis further investigated the flavor characteristics of differential VOCs in tea trichome and tea body along with potential contribution of tea trichome to the overall aroma quality. The differential VOCs between tea trichome and tea body across were evenly distributed most samples, except for FDBMDW and FDBYZ samples, which exhibited significant varietal differences. A flavor omics association network was constructed based on the top 10 flavor characteristics with the highest number of annotated differential VOCs to explore the impact of tea trichome volatiles on the aroma profile of white tea (Fig. 6). The size of each volatile circle corresponded to its relative importance, with larger circles indicating greater contributions. Notably, several VOCs, including citronellal, (E,E)-2,4-nonadienal, (E)-6,10-dimethyl-5,9-undecadien-2-one, β-damascone, heptanal and (E,E)-2,4-heptadienal, displayed prominent larger circles and higher levels in tea trichome, which were reported to be vital contributors to sweet floral, fruity, milky aroma (Hao et al., 2023; Q. Lin et al., 2021). Given their aromatic characteristics, they were believed to play a crucial role in distinguishing tea trichome from tea body, as well as contributing to the overall aroma quality of white tea. The findings of this study further corroborate the results of previous research while extending these insights in several new directions (Liu et al., 2024).

Fig. 6.

Flavor network based on differential VOCs between tea trichome and tea body of samples: A. FABMDW; B. FDBMDW; C. FDHMDW; D. FABYZ; E. FDBYZ; F. FDHYZ.

Note: The larger the pink circle, the more differential metabolites are annotated to the corresponding sensory flavor attribute, indicating the greater importance of that sensory flavor attribute. The larger the green, purple and red circle, the more sensory flavor attributes are annotated to the corresponding differential metabolite, indicating the greater importance of that differential metabolite. When the number of differential metabolites annotated to a specific sensory flavor attribute exceeds 50, only the top 50 differential metabolites with the highest VIP values are displayed.

Silver needle and premium white peony were known for their pekoe, fresh, fruity and floral aroma. Geraniol, linalool and benzoic acid methyl ester were identified as main contributors to the sweet and floral aroma of silver needle and white peony (Feng et al., 2022; Z. Liang et al., 2023). To elucidate the correlation between differential VOCs and aroma characteristics, Sankey diagrams were constructed using the top 10 aroma attributes and their corresponding top 10 VOCs. Notably, attributes such as sweet, fruity, green, herbal, woody, floral, and fresh were consistently observed across all three comparisons (silver needle trichome vs premium white peony trichome), highlighting the similarities in tea trichome among different varieties. As illustrated in Fig. 7, the differential VOCs were predominantly present at higher levels in premium white peony compared to silver needle. This observation may be attributed to the later harvest time of the first-harvested premium white peony, which allows for an extended accumulation of volatiles prior to being plucked. Thereinto, linalool and benzoic acid methyl ester were reported to be vital contributors to pekoe, fruity and floral aroma (Z. Liang et al., 2023), which showed either higher content in premium white peony or no significant difference between silver needle and premium white peony. Furthermore, sensory analysis corroborated these findings, revealing a richer and more diverse aroma profile in premium white peony than in silver needle. Based on the above results, it is presumed that the enhanced aroma profile is partially influenced by the presence of tea trichomes.

Fig. 7.

Sankey diagram of differential volatiles in tea trichome between premium white tea made of A. FAB; B: FDB; C: FDH variety.

The left column represents sensory flavor characteristics and the number of volatiles annotated, while the right column indicates differential VOCs. Red streamlines denote the flow direction of upregulated (up) differential VOCs, whereas blue streamlines represent the flow direction of downregulated (down) differential VOCs. The height of the boxes in the left column is determined by the number of differential VOCs displayed in the right column; the more numerous the VOCs, the taller the box. When a sensory flavor characteristic corresponds to more than ten differential metabolites, only the top 10 with the highest VIP values are shown.

4. Conclusion

This study investigated the volatile profile of tea trichome and its potential role in shaping the overall aroma quality of silver needle and premium white peony tea. Comparative analysis revealed significant differences in the volatile profile between tea trichome and tea body in all six white tea samples, with major differences lying in the content of volatile components. In brief, trans-β-ionone, β-damascone, linalool, non-8-enal and (E,Z)-2,6-nonadienal were identified as potential key aroma components in both tea trichome and tea body, determining the basic aroma characteristic of white tea. Greater number of aroma modifiers were also affirmed, including β-myrcene, 2-ethyl-3,5-dimethyl-pyrazine, (E)-2-hexenal, dehydro-β-ionone, (Z)-6-nonenal, (E,E)-2,4-decadienal and benzeneacetaldehyde, which were shared by all samples in tea trichome and tea body. Interestingly, the majority of these key and modified aroma components showed higher relative contents and ROAVs in tea trichome than those in tea body. Considering the larger surface area of tea trichome and higher ROAVs of key and modified aroma components, it was supposed to release aroma volatiles rapidly in the first stage of brewing, forming the initial olfactory profile. While aroma volatiles in tea body would gradually be liberated to extend the aroma duration due to the differences in physical properties between tea body and tea trichome. Premium white peony contained significantly higher levels of differential VOCs than silver needle, primarily contributing sweet, fruity, green, herbal, and woody notes. This difference likely resulted from premium white peony's later harvest time, allowing extended winter growth for greater volatile accumulation. Additionally, tea trichome enhanced premium white peony's aroma complexity, yielding a more diverse profile than silver needle.

The study reveals that tea trichome positively enhance white tea's aroma quality, notably intensifying fruity, floral, sweet, green, and woody notes. These findings indicate that trichome density could be a valuable criterion for selecting and breeding tea varieties with superior aroma characteristics. Considering that tea's holistic sensory profile emerges from synergistic interactions between volatile compounds and LC-MS-detected non-volatile metabolites, integrated metabolomics approaches are warranted to comprehensively map the complete metabolic landscape of tea trichomes.

Ethical statement

The authors confirm that this study complies with the ethical principles outlined in the World Medical Association's Declaration of Helsinki for human experimentation. Although national regulations do not mandate formal ethical approval for sensory assessment studies, and no institutional ethics committee oversees such protocols in this jurisdiction, the research team implemented rigorous safeguards to protect participants' rights and welfare. These measures included: 1) Full disclosure of study objectives, procedures, and potential risks prior to participation; 2) Voluntary engagement with explicit verbal consent documentation; 3) Strict confidentiality protocols prohibiting unauthorized data sharing; 4) Freedom to withdraw from the study at any stage without penalty; 5) Prohibition of coercive recruitment practices. All experimental procedures were conducted in accordance with these established ethical standards.

Abbreviations

HS-SPME-GC-MS	Headspace solid-phase microextraction combined with gas chromatography–mass spectrometry
VOCs	Volatile organic compounds
ROAV	Relative odor activity value
OT	Odor threshold
FAB	Camellia sinensis cv. Fuandabaicha
FDB	Camellia sinensis cv. Fudingdabaicha
FDH	Camellia sinensis cv. Fudingdahaocha
FABMDW	Premium white peony sample of Camellia sinensis cv. Fuandabaicha
FDBMDW	Premium white peony sample of Camellia sinensis cv. Fudingdabaicha
FDHMDW	Premium white peony sample of Camellia sinensis cv. Fudingdahaocha
FABYZ	Silver needle sample of Camellia sinensis cv. Fuandabaicha
FDBYZ	Silver needle sample of Camellia sinensis cv. Fudingdabaicha
FDHYZ	Silver needle sample of Camellia sinensis cv. Fudingdahaocha
SIM	Selective ion monitoring
RT	Retention time
RI	Retention index
PCA	Principal component analysis
HCA	Hierarchical clustering analysis
OPLS-DA	Orthogonal partial least squares-discriminant analysis
VIP	Variable importance projection
FC	Fold change

CRediT authorship contribution statement

Dandan Qi: Conceptualization, Data curation, Funding acquisition, Methodology, Visualization, Writing – review & editing, Writing – original draft. Min Lu: Data curation, Funding acquisition, Formal analysis, Validation. Yali Shi: Formal analysis, Visualization. Xingmin Zhang: Methodology. Lin Yue: Methodology. Houzhen Jia: Data curation. Chunwang Dong: Supervision. Huimin Liu: Supervision. Changbo Yuan: Funding acquisition, Resources, Supervision.

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.

Handling Editor: Professor Aiqian Ye

Appendix A. Supplementary data

The following is the Supplementary data to this article.

Data availability

Data will be made available on request.