Characterization of volatiles in tea leaf cuticular wax during white tea processing across distinct climatic regions

Abstract

Climate change significantly impacts tea plant (Camellia sinensis) physiology, particularly the cuticular wax layer on leaves, which serves as a critical barrier against environmental stress. This study investigates how climatically distinct growing regions modify the composition and morphology of tea leaf cuticular wax, and thereby affect volatile organic compounds (VOCs) during white tea processing. Through gas chromatography–mass spectrometry (GC–MS) analysis and sensory evaluation, we demonstrate that climate-induced wax modifications correlate with volatile profiles shifts, ultimately impacting tea fragrance. The results showed that most of the differential components in cuticular wax of samples between Guizhou (GZ) and Shandong (SD) Province were related to stress tolerance, indicating the influence of climate differences on chemical property of tea leaves. Among these compounds, β-N-acetylglucosamine, D-arabitol, L-arabitol and tiglic acid were identified throughout the whole processing, with the first two components showing higher content in SD samples. The differential volatiles in tea leaves between GZ and SD samples were mostly enriched in pathways related to cell survival proliferation and environmental adaption, some of which also participated in modulating flavor-related components. The results of sensory evaluation showed that GZ samples were found to have stronger floral and fruity aroma and weaker green flavor than SD samples, which was in accordance with previous study. Our findings highlight the ecological and industrial implications of wax-mediated aroma changes, offering insights for adapting tea cultivation to climatic variability.

Highlights

• •Climate alters tea leaf cuticular wax composition, impacting stress barrier function.
• •Climate-induced wax changes modify volatile organic compounds (VOCs) during white tea processing.
• •Key stress-related wax compounds vary significantly with growing region.
• •Climate differences shift tea VOCs, affecting sensory profiles (floral/fruity vs. green aroma).
• •Wax-mediated aroma changes highlight climate adaptation needs for tea quality and cultivation.

1. Introduction

Tea, renowned for its delicate flavor and rich healthcare efficacy, has gained significant attention from researchers and consumers. Due to its popularity and the demand of consumers, the introduction of tea plants to Shandong Province (35°35′N–37°09′N) was conducted from the 1950s and finally succeeded in 1965. It is known that tea trees thrive in the humid and rainy southern regions, making them unsuitable for cultivation in the dry and cold northern areas. Hence, it is easy to understand that abiotic stress including drought and low temperature in northern China would induce physicochemical changes in tea plants and thus affect tea flavor and aroma. Specifically, tea produced in northern China has higher brewing resistance due to thicker leaves and more water extracts (Wang et al., 2015). In addition, higher amino acids, soluble sugars and volatile flavor compounds in northern tea is also beneficial to unique tea flavor compared with southern tea (Wang et al., 2015). The Arabidopsis wax-overproducing mutant dewax displays higher resistance to both dehydration and freezing, confirming the positive effect of high cuticular wax accumulation on abiotic stress (Rahman et al., 2021). The cuticle, composed of cutin and cuticular wax, covers the surface of plants and is in direct contact with the external environment, which is an important barrier for tea plants to cope with biotic and abiotic stresses.

Investigations have confirmed that total wax content and wax component varies on the latitudinal gradient in response to different environmental factors (temperature, arid index, precipitation, ultraviolet irradiation, etc.) (Chen et al., 2023; Guo et al., 2015; Guo, Gao, et al., 2016; Li et al., 2019). In general, the total content and thickness of cuticular wax would be increased to reduce non-stomatal water loss and maintain leaf temperature in order to adapt to high latitude areas with cold, dry stress and elevated ultra-violet (UV)-B radiation (Cameron et al., 2006; Guo, Gao, et al., 2016; Guo, Xu, et al., 2016; He et al., 2019; Jenks et al., 2001; Seo et al., 2011). In addition to the total wax content, the constituent of wax is also sensitive to climatic conditions and is easily altered in response to different abiotic stress (Bueno et al., 2019; Celano et al., 2006). In general, latitude is significantly positively correlated with total wax coverage and the contents of alkanes and esters (Chen et al., 2023). In detail, aldehydes and alkanes, contributing to plant drought tolerance, show great increase trend under drought stress in cuticular wax of many plants, including alfalfa, Arabidopsis, wheat and sorghum (He et al., 2022; Kosma et al., 2009; Ni et al., 2012; Sanjari et al., 2021; Seo et al., 2011). Low temperature (4 °C) treatment increases the content of acids and thus leads to an increase in the total wax content in Thellungiella salsuginea, forming a cold-resistant barrier structure and reducing the risk of freeing of cell (He et al., 2019). While alkanes and primary alcohols decrease due to the exposure to low temperature (He et al., 2019; Rahman et al., 2021). In summary, latitude or environmental factors leads to variation in the content and composition of cuticular wax in the case of many plants.

The same thing happens to tea plants in response to abiotic stress, too. The difference is that leaves of tea plants can be made into tea for people to drink and thus flavor is an important evaluation factor for the whole quality. As it is proved that leaves with higher cuticular wax loading would produce more aroma than those with lower cuticular wax loading, making it a positive correlation between cuticula wax and tea aroma formation (Zuo et al., 2023). Therefore, we propose a hypothesis that regional climate differences alter tea cuticular wax composition, which affects aroma substance accumulation, leading to distinct tea aroma quality.

Till now, about 100 kinds of compounds have been identified in cuticular wax of tea plants, including acids, 1-alkanols, aldehydes, alkanes, esters, fatty acids, glycols, sterols, etc. (Zhu et al., 2018; Zhu et al., 2022). Some of these compounds either act as aromatic precursors or aroma compounds with specific flavor, contributing to the formation of tea flavor. Moreover, some of these compounds are already to confirmed to be altered by climate change, such as fatty acids (n-hexadecanoic acid and n-nonanoic acid), alkanes (n-nonacosane), primary alcohols (n-triacotan-1-ol and n-dotriacontane-1-ol), some esters (glycol esters and phthalate esters) and sterols (Chen et al., 2020; Zhang et al., 2021; Zhu et al., 2022). However, there is no studies digging into the effects of changed cuticular wax caused by climate differences on tea flavor.

Thus, this study aims to determine whether cuticular wax composition varies with climatic differences and its potential effects on tea flavor, integrating volatile metabolomics, full-spectrum metabolomic and wax metabolomics and sensory evaluation. Raw materials were collected from Fudingdabai (Camellia sinensis L.) planted in Shandong (35°10′52.0″N, 118°50′11.5″E) and Guizhou Province (27°45′4.0″N, 107°27′26.9″E) and made into white tea following traditional crafts (Fig. 1). Next, we comprehensively analyzed the cuticular wax volatiles (extracted from the cuticular wax) and volatile profiles of fresh leaves (extracted from tea leaves using a different method), manufactured tea leaves and made tea, respectively. Combined with sensory evaluation results, the effects of cuticular wax caused by climate difference on volatiles and flavor were demonstrated, which would be beneficial to explaining the difference of tea flavor in south and north China.

Fig. 1.

Sampling during white tea processing.

2. Materials and methods

2.1. White tea sample collection

Fudingdabai (Camellia sinensis L.) is a famous tea cultivar widely planted in China and widely planted in China. In this study, a bud and two leaves were plucked from Fudingdabai (Camellia sinensis L.) planted in Junan, Shandong and Meitan, Guizhou Province (the comparison of the climates and soil conditions of these two tea production regions are presented in Table 1 and Table 2) and then produced into white tea with the same crafts. In general, fresh tea leaves were subjected to traditional processing as follows: fresh tea leaves were withered indoor for about 48 h, and then dried at 70–80 °C to a moisture content of approximately 6 %. Samples were taken every eight hours, as shown in Fig. 1. All the samples were freeze-dried and then stored in ultra-low temperature refrigerator.

Table 1.

The climate differences between Guizhou and Shandong province.

Climate factor	Guizhou	Shandong
Coordinates	27°45′4.0″N, 107°27′26.9″E	35°10′52.0″N, 118°50′11.5″E
Altitude	900–1100 m	100–200 m
Annual average temperature	15.0 °C	12.5 °C
Effective accumulative temperature (≥10 °C)	4500–5000 °C	4200–4500 °C
Diurnal temperature variation	7–9 °C	10–12 °C
Annual precipitation	1100–1200 mm	800–900 mm
Air humidity	75 %–90 %	55 %–85 %

Table 2.

The soil conditions between Guizhou and Shandong province.

Soil characterization	Guizhou (Guo et al., 2024; Liu et al., 2020)	Shandong (http://www.junan.gov.cn/info/1277/78083.htm/. Accessed October 1, 2024.)
Soil type	Acidic yellow soil	Slightly acidic brown soil
pH value	5.01	5.6
Soil organic matter	22.84 g/kg (above-average); abundant total nitrogen, available nitrogen and phosphorus; potassium content is moderate (135.61 mg/kg)	16.8 g/kg (moderately low); relatively low available nitrogen, abundant available phosphorus, low level of available potassium.
Soil texture	Sticky and heavy with poor aeration; the deep soil has strong water capacity.	Loose and friable with good aeration; poor water and nutrient retention capacity
Mineral elements	Rich in manganese and zinc, with a significant deficiency of calcium and magnesium	a common deficiency of calcium, magnesium, sulfur, boron, and molybdenum

2.2. Standards and reagents

Purified water was obtained from Hangzhou Wahaha Group Co., Ltd. (Hangzhou, China). Methanol was obtained from Thermo Fisher Scientific (USA). n-Hexane was obtained from Shanghai Anpel Experimental Technology Co. Ltd. (Shanghai, China). Chloroform was obtained from Shanghai Taitan Technology Co. Ltd. (Shanghai, China). Methyl nonanoate, methyl decanoate, methyl dodecanoate, methyl myristate, methyl stearate, methyl arachidate, methyl behenate and methyl tetracosanoate were obtained from Nu-Chek-Prep., Inc. (Elysian, Minnesota, USA). Methyl octanoate and methyl palmiate were obtained from LGC standards (Charleston, SC, USA). D-Luciferin free acid, L-2-chlorophenylalanine and L-valine-d8 were purchased from Shanghai Yuanye Bio-Technology Co., Ltd. (Shanghai, China), Shanghai Hengchuang Bio-Technology Co. Ltd. (Shanghai, China), and Shanghai Haoyuan Bio-Technology Co. Ltd. (Shanghai, China), respectively.

2.3. GC–MS analysis of cuticular wax volatiles during white tea processing

2.3.1. Extraction of cuticular wax volatiles and derivation reactions

Weigh approximately 1 g of the sample and add 5 mL of chloroform for extraction at 60 °C for 5 min. After extraction, discard the sample and collect the extract in a 4 mL glass bottle. Transfer the extract to the 4 mL glass bottle multiple times for nitrogen blowdown; perform two transfers, each with 1.5 mL of the extract. After nitrogen blowdown, place the sample in a dry environment for derivatization. First, add 150 μL of chloroform to redissolve the sample, vortex for 1 min, and then sonicate for 5 min. Next, add 80 μL of BSTFA reagent and place the sample in a 70 °C environment for a 1-h reaction. After derivatization, let the sample stand at room temperature for 30 min. Transfer the liquid to a 1.5 mL centrifuge tube and centrifuge at room temperature for 10 min (12,000 rpm, 25 °C). Take 150 μL of the supernatant and transfer it to a sample vial containing a glass insert for GC–MS metabolomics analysis. The quality control (QC) sample is prepared by mixing equal volumes of extracts from all samples.

2.3.2. GC–MS analysis

GC–MS analysis is performed on an Agilent 7890B gas chromatograph with a 5977A mass spectrometer (Agilent, Santa Clara, CA, USA). A DB-5MS capillary column (30 m × 0.25 mm × 0.25 μm film thickness, Agilent J&W Scientific, Folsom, CA, USA) is applied. Helium (purity >99.999 %) is employed as carrier gas with a constant flow rate of 1 mL/min. The temperature of injection port is set at 260 °C with splitless mode. The injection volume is 1 μL. Solvent delay time is 5 min. The oven temperature is initially set at 80 °C and held for 2 min; then it is programmed to increase at 15 °C/min to 260 °C and held for 10 min. Then it rises at 5 °C/min to 315 °C and held for 10 min.

Mass spectrometer conditions: ionization mode, EI; ion source temperature, 230 °C; quadrupole temperature, 150 °C; electron energy, 70 eV; full scan mode, mass scan range: 50–650 amu (amu).

2.4. GC–MS analysis of leaves during white tea processing

2.4.1. Extraction of volatile metabolites in leaves

Weigh 60 mg of the sample and place it in a 1.5 mL centrifuge tube along with two small steel beads. Add 600 μL of methanol–water (V:V = 7:3, containing mixed internal standards at 4 μg/mL). Pre-cool the mixture in a −40 °C freezer for 2 min, then grind it in a grinder at 60 Hz for 2 min. Perform ultrasonic extraction in an ice water bath for 30 min, and allow the sample to stand at −40 °C for 2 h. Centrifuge the sample at 13,000 rpm for 10 min at 4 °C, and transfer 100 μL of the supernatant to a glass derivatization vial. Dry the sample using a centrifugal concentrator. Add 80 μL of pyridine solution containing hydroxylamine hydrochloride (15 mg/mL) to the glass derivatization vial and incubate it in a shaking incubator at 37 °C for 60 min for oximation. Afterwards, add 50 μL of BSTFA derivatization reagent and 20 μL of n-hexane, followed by 10 μL of ten internal standards (C8/C9/C10/C12/C14/C16/C18/C20/C22/C24, all prepared in chloroform) to the vial. React the sample at 70 °C for 60 min. After removing the sample, let it stand at room temperature for 30 min before performing GC–MS metabolomics analysis. The quality control (QC) sample is prepared by mixing equal volumes of extracts from all samples.

2.4.2. GC–MS analysis

GC–MS analysis is performed on an Agilent 7890B gas chromatograph with a 5977 A mass spectrometer (Agilent, Santa Clara, CA, USA). A DB-5MS capillary column (30 m × 0.25 mm × 0.25 μm film thickness, Agilent J&W Scientific, Folsom, CA, USA) is applied. Helium (purity >99.999 %) is employed as carrier gas with a constant flow rate of 1 mL/min. The temperature of injection port is set at 260 °C with splitless mode. The injection volume is 1 μL. Solvent delay time is 5 min. The oven temperature is initially set at 60 °C and held for 0.5 min; then it is programmed to increase at 8 °C/min to 210 °C; subsequently, it rises at 15 °C/min to 270 °C; and finally, at 20 °C/min to 305 °C, where it is held for 5 min.

Mass spectrometer conditions: ionization mode, EI; ion source temperature, 230 °C; quadrupole temperature, 150 °C; electron energy, 70 eV; full scan mode, mass scan range: 50–500 amu (amu).

2.5. Sensory analysis

In this study, white tea samples made in Shandong and Guizhou were submitted to quantitative descriptive analysis (QDA) following previous methods (Qi et al., 2018). In brief, a panel of ten professional assessors (5 males and 5 females, aged from 27 to 33 years old) were trained with white tea so as to recognize, describe and discriminate different aroma traits of different white tea samples. Five aroma traits were chosen to generalize the aroma characteristics of white tea in this study. Briefly, 3.0 g of tea sample was infused with 150 mL boiling water and maintained for 5 min before being assessed. Every assessor was required to grade the intensity of every aroma trait on a five-point scale (0, not perceptible; 3, moderate; 5, strong). Each sample was required to be assessed three times. Results of formal QDA were visualized in radar chart. The ANOVA results (analysis of variance) of aroma intensity between Shandong and Guizhou white tea were marked with asterisks.

2.6. Data preprocessing

MS-DIAL is used to perform qualification and quantification. The basic information including retention time, peak area, names, mass-to-charge ratio, is acquired after the pretreatment of MS-DIAL. Qualification of volatiles in tea leaves is based on LuMet-GC 5.0 database.

3. Results and discussion

3.1. Sensory quality evaluation of white tea samples from Shandong and Guizhou province

Sensory evaluation of white tea aroma was conducted based on QDA method by a panel of professional assessors. The most frequently occurring attributes in the evaluation of white tea samples included pekoe, sweet, floral, fruity and green. A significant difference was observed between GZ and SD white tea samples in their aroma intensity, indicating considerable variation in their aroma quality (Fig. 2). In detail, both GZ and SD white tea were characterized with moderate pekoe and slight sweet flavor. The aroma intensity of floral, fruity and green showed extremely significant difference between them, with stronger floral and fruity aroma in GZ white tea than SD white tea. While green flavor in GZ sample was weaker than in SD sample. Research has shown that newly produced white tea is prone to have green smell due to some volatiles with low-boiling point such as (Z)-3-hexenol and (E)-2-hexenal (Qi et al., 2018; Qin et al., 2024). In some way, intense green flavor is not conducive to the overall aroma quality, which needs a period of time for transformation to enhance its aroma. Combined with the national standards for tea sensory evaluation, it suggests that of GZ white tea has more pleasant aroma quality than SD white tea.

Fig. 2.

Aroma profile of white tea samples from Guizhou and Shandong Province. Asterisks (**) indicates extremely significant differences.

3.2. Analysis of cuticular wax profiles throughout the withering process in tea samples from Shandong and Guizhou province

Fresh tea leaves are submitted to withering and drying process to make white tea. Due to its unique processing craft, white tea undergoes the minimal mechanical damage, thereby preserving the integrity of its cuticular wax to the greatest extent compared with other teas. It is therefore very suitable for investigating the differences of cuticular wax from different regions. In order to understand the components of cuticular wax during white tea withering, we profiled the cuticular wax extracted from the manufactured tea leaves which were sampled every eight hours with a gas chromatography–mass spectrometry (GC–MS). A total of 187 compounds were identified in cuticular wax of Shandong and Guizhou samples, including alkanes, fatty acids, alcohols, esters, lipids, monoterpenoids, etc. (Fig. 3C). The very-long-chain volatiles including fatty acids and their derivates, alcohols, alkanes and esters make up more than half of volatiles and lots of these volatiles were not commonly identified in tea leaves. The rest volatiles with less than 12C were mostly identified as aroma contributors of tea aroma such as linalool oxide, nerol, linalool, benzyl alcohol, etc. Principal component analysis (PCA) and partial least squares discrimination analysis (PLS-DA) summarized the dynamic change of volatiles in cuticular wax during withering tea withering. The results showed similar dynamic change of cuticular wax profiles in manufactured tea samples from two provinces, indicating similar role of cuticular wax in its volatile profiles. In addition, the overview of cuticular wax components were similar in SD and GZ manufactured white tea samples. However, great differences in cuticular wax composition were discovered in tea samples from two provinces with totally different climates between two provinces, indicating great effects of environment on cuticular wax.

Fig. 3.

Analysis of cuticular wax during withering process. A. GC–MS traces of cuticular waxes during withering process; B. Chain length distributions of cuticular wax in volatiles; C. Volatile categories in cuticular wax; D. Heatmap of identified volatiles during withering process.

3.2.1. Most cuticular volatiles showed higher level in GZ samples than in SD samples

Both PCA and HCA results (Fig. 4A, G) showed great differences between SD and GZ samples at each sampling time, indicating great variations caused by climate changes or planted areas. In detail, higher levels of most volatiles in GZ samples than SD samples during the whole withering process were discovered as shown in heatmap analysis (Fig. 4G). OPLS-DA model (Fig. 4B, C) was used to characterize the dynamic changes of cuticular wax volatiles during withering, showing obvious spatial distribution characteristics at different withering sampling points in both GZ and SD samples. Although similar metabolic trends of GZ and SD samples during withering process were discovered, K-means clustering analysis (Fig. 4E, F) of volatiles showed differences in their variation trends during withering between GZ and SD samples. The results showed that cuticular wax volatiles in GZ and SD samples were grouped into 3 and 5 sub classes, respectively. In accordance with the results of heatmap, more volatiles (132/187) in GZ samples showed continuous increasing trend during withering (Fig. 4E sub classes 1, 2) in comparison with SD samples. In addition, 55 of 187 volatiles in GZ samples (Fig. 4E sub classes 3) showed obvious decrease during withering in comparison with SD samples (Fig. 4F sub classes 1 & 4).

Fig. 4.

A. PCA results of all tea samples based on cuticular wax volatiles. B. OPLS-DA results of GZ tea samples based on cuticular wax volatiles. C. OPLS-DA results of GZ tea samples based on cuticular wax volatiles. D. The number of significant differences between GZ and SD samples at each sampling time. E. K-mean clustering analysis of volatiles in GZ samples during withering. F. K-mean clustering analysis of volatiles in SD samples during withering. G. Heatmap analysis of volatiles in cuticular wax: volatiles marked in green are those with less than 12 carbon atoms; volatiles marked in red are those with more than 12 carbon atoms; H. Venn diagram of volatiles in each comparison group with different withering times. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

As for the difference between them, the comparison in Fig. 4D displayed higher level of most volatiles in GZ samples than SD samples at each sampling point. It was also reflected in heatmap results (Fig. 4G). In addition, most long-chain volatiles in GZ samples in heatmap (Fig. 4G) showed higher content than SD samples. Specifically, most long-chain volatiles showed a higher increase in GZ samples than in SD samples. However, some common aroma volatiles identified in tea, such as nerolidol and myristic acid showed higher content in SD cuticular than in GZ samples. To sum up, there are great differences in cuticular wax profiles between GZ and SD samples, which could be attributed to the climate difference in Shandong and Guizhou province. Lower humidity, lower annual average temperature and larger diurnal temperature variation in Shandong production region may act as an environmental stress, inducing tea plants to synthesize cuticular wax with different compositions to reduce non-stomatal water loss and protect tissues.

3.2.2. Many cuticular wax showed increasing trend during withering process in both GZ and SD samples

To elucidate dynamic changes in cuticular wax components during withering, all identified components were submitted to PCA and HCA analysis. The results exhibited great differences in cuticular wax components throughout withering process in both SD and GZ samples. To elucidate the variation trends of these compounds during withering process, K-means analysis was applied to cluster volatile components based on their dynamic change (Fig. 4E, F). Several different variation trends were defined among 187 components. In the case of SD and GZ samples, a total of 110 (Fig. 4F sub-classes 2 and 5) and 132 (Fig. 4E sub-classes 1 and 2) components, respectively, showed the highest content at the end of withering stage, indicating positive influence of withering on potential aroma-related components in terms of cuticular wax volatiles. Further analysis showed that SD and GZ shared 90 components with an upward trend during withering, indicating more common features in dynamic changes of cuticular wax from different planted areas. Half of these shared increased volatiles were with more 12C. Most of these long-chain volatiles increased during withering in comparison with volatiles with less 12C in both GZ and SD samples. Besides very-long-chain volatiles, many common volatiles in tea leaves were also identified. It is worth noting that several characteristic aroma components including phenylethyl alcohol, geraniol, nerol, linalool and linalool oxide, were also found to be abundant in cuticular wax of fresh leaves and increased at the end of withering. They contribute mainly flowery, fruity and sweet aroma and are positively correlated with flavor quality. Thus, it could be proposed that aroma quality could be influenced by cuticular wax considering its volatile composition.

3.3. Analysis of metabolic profiles throughout the withering process in tea samples from Shandong and Guizhou province

A total of 296 metabolites were identified in GZ and SD samples, undergoing dynamic change during withering. The results in PCA and heatmap showed great differences in these metabolites at each sampling point between SD and GZ samples. Nevertheless, variation trend during the whole withering process showed similarity in both samples from two provinces.

3.3.1. Dynamic change of metabolic profiles during withering process

To more intuitively present the variation trends of metabolites during the processing, K-means clustering analysis method was employed. Metabolites with similar variation trends during the processing could be grouped in the same subclass by an unsupervised clustering algorithm. As shown in Fig. 5C, metabolites in SD samples could be grouped into 3 subclasses according to their changing trend of content. Subclass 1 (Fig. 5C) contained 82 components and they showed minor changes during withering and decreased sharply after drying process.

Fig. 5.

A. PCA results of all tea samples based on volatile metabolites. B. PLS-DA results of all tea samples based on volatile metabolites. C. K-mean clustering analysis of volatiles in SD samples during withering. D. K-mean clustering analysis of volatiles in GZ samples during withering. E. Heatmap analysis of volatiles in tea leaves.

Thereinto, linalool, benzyl alcohol and 2-phenylethanol are important contributors to tea aroma and they are also involved in the synthesis of other aroma components, which explains the content fluctuation during white tea process (Zhou et al., 2023). In addition, saccharides and acids are also main components in this subclass, including monosaccharide and organic acids. They are also important flavor contributors to tea quality. Subclass 2 (Fig. 5C) showed continuous decrease throughout the entire process and contained 87 components, including organic acids, saccharides and alcohols. Among these components, phytol, geraniol and 2-hexenol are important aroma sources in tea and the decrease of them could be attributed to further transformation to other aroma volatiles. Subclass 3 exhibited an increasing trend during the entire process and contained 127 metabolites (Fig. 5C). About 20 amino acids are grouped in this subclass, and their content peaks after drying. Moreover, more fatty acids and saccharides are also grouped in this subclass, indicating accumulation of these compounds during white tea process.

In GZ samples, metabolites with similar trends during processing were grouped into 5 subclasses. Components in subclass 1, 2 and 5 (Fig. 5D) showed higher content in made tea than in fresh leaves and they contained 142 components. Thereinto, subclass 2 showed continuous increase during the whole processing and contained 74 components (Fig. 5D). Many amino acids and some aroma volatiles were grouped in subclass 2, such as L-aspartic acid, L-asparagine and benzyl alcohol. Components grouped in subclass 1 and 5 showed fluctuations during processing and finally they increased in the late stage of withering and drying. Subclass 3 showed continuous decrease during the whole processing and contained 107 components (Fig. 5D), including some organic acids and saccharides. Subclass 4 contained 45 components and exhibited rising trend at first and then decreased in the late withering stage and drying process, leading to lower content in made tea than in fresh leaves. Quite a few aroma volatiles, such as linalool, geraniol and 2-phenylethanol, were grouped in this subclass, which might be attributed to the enzymatic or nonenzymatic transformation.

3.3.2. Differences in metabolite profiles between Shandong and Guizhou Province tea samples

Heatmap analysis (Fig. 5E) of metabolite profile during white tea processing showed big difference between SD and GZ samples, indicating great influence on metabolites by climate contrast. To analyze the difference between SD and GZ samples during white tea processing at each sampling point, PCA analysis was conducted (Fig. 6). Clear distinction was obtained between SD and GZ samples at each sampling point as shown in Fig. 6, with the first two principal components explained over 65 % of the total variation in each comparison. In addition, volcano plot (Fig. 6) was constructed to visualize the difference between SD and GZ samples during white tea processing at each sampling point. Differential metabolites with p < 0.05 and log2 (fold change) > |0.263| were screened out to investigate the reasons for the difference between SD and GZ samples. A total of 57, 115, 100, 67, 84, 106, 102 and 97 differential components were identified in the comparison of SD and GZ samples at each sampling point during white tea processing, respectively (Fig. 7).

Fig. 6.

The effect of different regions on metabolites in tea leaves during withering at each sampling time point: A. fresh leaves; B. at 8 h of withering; C. at 16 h of withering; D. at 24 h of withering; E. at 32 h of withering; F. at 40 h of withering; G. at 48 h of withering; H. made tea.

Note: The red dots represent significantly up-regulated differential metabolites in the experimental group, the blue dots represent significantly down-regulated differential metabolites, and the gray dots represent metabolites with no significant differences. Each point in the Figure represents a metabolite, the horizontal coordinate is the log2 (FC) value of the comparison between the two groups, the vertical coordinate is the -log10 (p-value) value, the red point is the significantly up-regulated differential metabolite (p < 0.05, and FC > 1), the blue point is the significantly down-regulated differential metabolite (p < 0.05, and FC < 1). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 7.

A. The number of differential volatile metabolites during white tea processing; Chord diagrams of KEGG enrichment analysis in leaves of SD and GZ samples at each sampling time point: B. fresh leaves; C. at 8 h of withering; D. at 16 h of withering; E. at 24 h of withering; F. at 32 h of withering; G. at 40 h of withering; H. at 48 h of withering; I. made tea.

It was noted that most of differential components were related to stress tolerance to some extent, which was in coincidence with the actual case. Notably, four differential components were identified throughout the whole processing, including β-N-acetylglucosamine, D-arabitol, L-arabitol and tiglic acid. According to previous reports, they might play important role in response to drought and cold stress by regulating cell wall structure, osmotic pressure, antioxidant reaction and membrane stability, respectively (Boulanger et al., 2014; Chi et al., 2020). The synergistic effects of these substances may further enhance tea plant's stress resistance. Both the content of β-N-acetylglucosamine and D-arabitol showed higher content in SD samples, while the content of L-arabitol and tiglic acid were reversed. The variation in the content of these metabolites in tea leaves reflects their adaptability to different climatic conditions. Tea plants in Shandong Province with low temperature and drought conditions might enhance cell wall stability and osmotic regulation ability by increasing the contents of β-N-acetylglucosamine and D-arabitol, while tea plants in Guizhou Province with higher temperature and higher precipitation can cope with oxidative stress and lipid metabolism needs by increasing the contents of L-arabitol and tiglic acid. High humidity combined with a cloudy and foggy environment promotes the accumulation of amino acids and aromatic substances, which endows Guizhou tea with floral and fruity aroma. Moreover, the mid-to-high altitude (900-1100 m) of Guizhou causes tea trees to grow slowly and retain strong tenderness in their buds and leaves, promoting the gradual accumulation of aromatic compounds.

It was noted that more differential volatiles showed higher content in SD samples than in GZ samples in the late stage of white tea processing (Fig. 6A). These differential volatiles were mostly enriched in the following metabolic pathways (Fig. 6B–I): pentose and glucuronate interconversions, galactose metabolism, ABC transporters, D-amino acid metabolism, aminoacyl-tRNA biosynthesis pathway. The common essence of these pathways is to integrate carbon and nitrogen metabolism, material transport and biosynthesis to form the basic network that maintains cell survival, proliferation and environmental adaptation. Specifically, pentose and glucuronate interconversions contribute to the production of nucleotide precursors and detoxification intermediates, which may influence tea aroma complexity. Galactose metabolism provides energy and structural sugars, potentially affecting tea polysaccharide content and sensory properties. ABC transporters facilitate the translocation of metabolites, including volatile organic compounds (VOCs), across membranes, directly impacting aroma profile development. D-Amino acid metabolism and aminoacyl-tRNA biosynthesis are critical for protein turnover and secondary metabolite synthesis, further modulating flavor-related compounds. This interconnected metabolic activity suggests that the divergence in volatile profiles between SD and GZ samples arises from differential regulation of these pathways, likely driven by regional raw material traits or processing conditions (e.g., withering duration, temperature).

3.4. Potential correlation between cuticular wax and sensory quality

The chemical profiling of tea cuticular wax revealed distinct compositional variations in cuticular wax of tea leaves from SD and GZ province, predominantly characterized by alkanes (C4–C21), lipids, benzenoids and organic oxygen compounds (Fig. 2B, C). Notably, GZ white tea with higher sensory scores (e.g., floral and fruity intensity) exhibited a significantly elevated proportion of long-chain alkanes (C29–C33), suggesting their potential role in modulating aroma retention. This is in accordance with previous studies proposing that hydrophobic wax layers may act as resources for volatile organic compounds (VOCs), thereby prolonging fragrance release during tea infusion.

Intriguingly, fatty acid esters (e.g., hexacosyl palmitate) showed a negative correlation with astringency. We hypothesize that ester-dominated wax structures might reduce water permeability, limiting the leaching of polyphenols (particularly catechins) into the infusion phase, thereby mitigating bitter perception. These findings imply that wax composition and physical structure collectively influence tea sensory profiles through both chemical partitioning and mechanical barrier effects.

However, the causality between wax composition and sensory attributes requires further validation. Quantitative analysis of key aroma contributors and their contribution to the specific sensory attributes should be conducted in the future research for a direct chemical components-sensory link. Moreover, whether specific wax components directly interact with taste receptors or indirectly alter metabolite bioavailability remains unclear. Future studies integrating in vitro biosensor assays and dynamic infusion modeling could elucidate these mechanisms.

4. Conclusion

In conclusion, the dynamic change of volatiles in cuticular wax during white tea processing influenced by climate differences along with the aroma differences between white tea sample in two provinces (Guizhou and Shandong Province) were investigated. We have found that volatiles in cuticular wax and leaves are influenced greatly by climate in two provinces. A total of 296 metabolites were identified in GZ and SD white tea samples, with significant variation during processing. In particular, most of differential components in cuticular wax between GZ and SD samples were related to stress tolerance, indicating the influence of climate differences on chemical property of tea leaves. Among these compounds, β-N-acetylglucosamine, D-arabitol, L-arabitol and tiglic acid were identified throughout the whole processing, with the first two components showing higher content in SD samples. The differential volatiles in tea leaves between GZ and SD samples were mostly enriched in pathways related to cell survival proliferation and environmental adaption, some of which also participated in modulating flavor-related components. The results of sensory evaluation showed that GZ samples were found to have stronger floral and fruity aroma and weaker green flavor than SD samples, which was in accordance with previous study. The findings of this study not only reveal the basic characteristics of cuticular wax in the Fudingdabaicha tea cultivar, but more importantly, preliminarily demonstrate that its cuticular wax metabolism exhibits strong plasticity in response to environmental factors. The in-depth understanding of the variation in cuticular wax volatiles could guide customized processing craft, such as adjusting withering, spreading, fixation and fermentation based on the fresh leaves' cuticular wax profile to better bring out its target floral, fruity or sweet aroma. Since our existing findings only establish a correlative relationship between specific wax compounds and aroma volatiles, further research on the casual relationship between cuticular wax and aroma quality would be the future research directions —such as whether wax compounds convert to aroma substances via enzymatic hydrolysis, or regulate aroma precursor synthesis.

CRediT authorship contribution statement

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

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.

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.

Data availability

Data will be made available on request.