Abstract Blister blight, as one of the most threatening and damaging disease worldwide, mainly infects young organs and tissues seriously affecting tea growth and quality. In this study, the spread of pathogen on tea leaves were examined by toluidine blue staining, scanning electron microscope and transmission electron microscope analysis. The composition and abundance of fungal community on leaf tissues were firstly analyzed. Sensory evaluation and metabolites analysis indicated that diseased tea leaves had strong sweet taste and soluble sugars contributed significantly to the taste, while metabolites showing bitter and astringent taste (caffeine, catechins) were significantly decreased. According to the biological functions of differential metabolites, sugars including 7 monosaccharides (d-xylose, d-arabinose, d-mannose, d-glucuronic acid, glucose, d-galactose and d-fructose), 2 disaccharide (sucrose and maltose) and 1 trisaccharide (raffinose) were the main differential sugars increased in content (>2 fold change), which was of great significance to sweet taste of diseased tea. Introduction Tea, as a beverage, is popular for its healthy benefits and pleasant flavor. Tea plant is a perennial economic crop which is susceptible to various kinds of destructive foliar diseases due to the warm and humid growth environment in tea plantations that provides a suitable microclimate for the breeding of pathogens ([29]Baby, 2002). Among these diseases, blister blight caused by Exobasidium vexans Massee is considered as the most serious disease in the world that mainly damages tender leaves, buds and young fruits, resulting in yield loss and quality decrease ([30]Punyasiri et al., 2005, [31]Sen et al., 2020). At the tea plant experimental plots of Sri Lanka, the disease first occurs through the formation of appressoria and penetration of the cuticle, after penetration visible translucent spots are formed, and with the development of spots, characteristic circular blisters appear on the surface of young tissues, finally resulting in the infected tea leaves distorted and infected stems broken off at the point of infection ([32]Punyasiri et al., 2005). Blister blight disease occurring in almost all the tea plantation regions of Asia, has become the major problem on tea plants that causes serious economic losses ([33]Sinniah et al., 2016, [34]Baby, 2002, [35]Guo et al., 2005). It is reported that blister blight disease is more serious in alpine tea gardens in southwest and south of China. In epidemic years, the incidence of blister blight in tea growing regions located in southwest of China can reach 40–50 %, and even reach 90 % in severe cases ([36]Chen and Sun, 2013, [37]Sen et al., 2020). It was estimated to cause about 33 % crop loss in Sri Lanka ([38]De Silva et al., 1992), 25 % in Indonesia and as high as 50 % in South India, respectively, in tea fields in the absence of control measures ([39]Radhakrishnan and Baby, 2004). Besides yield loss, blister blight disease negatively affects the growth of tea trees, resulting in the quality decrease of processed tea which is even exceeded the economic threshold level 35 % ([40]Ponmurugan et al., 2016, [41]Radhakrishnan and Baby, 2004). Tea has abundant non-volatile and volatile metabolites such as tea polyphenols, amino acids components, caffeine, soluble sugars, lipids, and aromatic compounds, which are proved to be closely associated with the flavor and quality of tea ([42]Cao et al., 2021). Some quality related metabolites such as caffeine, epicatechin (EC) and theanine (Thea) also have been reported to have potential effect against various kinds of viral or fungal diseases ([43]Sharma et al., 2021, [44]Punyasiri et al., 2005). The development of pathogen in various stages showed different effect on pathogenesis related protein, anti-oxidative enzymes and flavonoid pathway in tea, suggesting the possible role of some chemical compounds (i.e., reactive oxygen species, anthocyanins, lignins, catechins) and other synthesized compounds in acting as antifungal agents in different tea cultivars ([45]Nisha et al., 2018). A large amount of studies have validated that various kinds of biotic (ie, disease infection, insect attack) and abiotic (ie, mechanical damage, environment conditions) stresses can remarkably influence the quality and flavor of tea by significantly changing the composition and amount of metabolites, especially affect the aromatic compounds in tea plants ([46]Chen et al., 2017, [47]Cho et al., 2017, [48]Mei et al., 2017, [49]Zeng et al., 2019). Tea processed from blister blight infected fresh tea leaves is fragile and shows obvious bitter taste, and the level of chemical components related to tea quality has decreased significantly, especially tea polyphenols and catechins ([50]Guo et al., 2005, [51]Jayaswall et al., 2016). [52]Premkumar et al. (2008) reported that after the infection of blister blight disease, there were remarkable reduction in the amount of tea polyphenols, catechins, sugar, nitrogen, proteins, amino acids and other substances in infected tea leaves. [53]Mur et al. (2015) analyzed the chemical compounds including caffeine, flavan-3-ol, flavone and flavonol in blister blight infected tender tea leaves by using high performance liquid chromatography with online photodiode array detection and electrospray ionization-tandem mass spectrometry (HPLC-PDA-ESI/MS), and found that kaempferol and quercetin glucosides, kaempferol triglycosides and some catechin-class antioxidants were increased, while the level of caffeine and apigenin and myricetin glycosides were remarkably reduced as disease progressed. However, limited information are still available on the tea quality affected by blister blight disease. In this study, the development and spread of blister blight on tea leaves, and the composition and abundance of fungal community on leaf tissues were fully investigated. After the infection of the disease, major metabolite differences in healthy and diseased tea leaves were studied, centring on the metabolites that contributed to the tea taste, which is helpful to improve the understanding on the influence of blister blight on tea quality. Materials and methods Materials During 2018 to 2020, tea leaves displaying blister blight symptoms were observed in high mountain tea plantations above 1000 m altitude located in Quxian county, Dazhou city, Sichuan province, China (30°85′N, 106°94′E). The typical blister blight symptoms were collected to show the symptom development. To analyze the characteristics of the blister blight disease, more than 100 fresh diseased tea shoots showing blister blight symptoms were harvested at April 2020, and used for microbial diversity analysis and microscopic analysis of pathogens including tissue section observation, transmission electron microscopy (TEM) and scanning electron microscopy (SEM) analysis. To investigate the effect of blister blight on tea quality and flavor, diseased tea shoots with one bud and two leaves (500 g) were sampled, while healthy tea leaves were used as a control. All the experiment had three replicates. Tea samples used for metabolites profile analysis were fixed by microwave (2–3 min), dried in a tea dryer machine (1 h, 80℃), finally milled and stored in a − 80℃ freezer. Microscopic analysis Toluidine blue staining (TBS) Tea leaf tissues (1 × 1 cm) were firstly fixed with FAA (Formaldehyde-acetic acid–ethanol) (Solarbio, Beijing, China), then dehydrated with ethanol, embedded in paraffin and routinely sliced. The slices were placed in xylene I (Sinaopharm Group Chemical Reagent Co. LTD) for 20 min, xylene II (Sinaopharm Group Chemical Reagent Co. LTD) for 20 min, anhydrous ethanol I (Sinaopharm Group Chemical Reagent Co. LTD) for 5 min, anhydrous ethanol II (Sinaopharm Group Chemical Reagent Co. LTD) for 5 min, 75 % alcohol for 5 min, and washed with double distilled water (ddH[2]O). The dehydrated slices were stained in toluidine blue solution (Wuhan Google Biotechnology Co. LTD) for 6 min, then washed with ddH[2]O, finally the qualified slices were dried in an oven. Transparent sealing: the dried slices were treated in xylene (Sinaopharm Group Chemical Reagent Co. LTD) for 5 min, then taken out to dry, finally sealed with neutral gum for microscopic examination. Scanning electron microscopy (SEM) Tea leaf tissue blocks (3 mm^2) were harvested within 3 min and washed with phosphate buffer saline (PBS) (Servicebio, Wuhan, China) gently, then immediately immersed in electron microscopy fixative (Servicebio, Wuhan, China) at room temperature for 2 h, finally transferred into 4 °C for preservation. After fixation, tissue blocks were rinsed with 0.1 M phosphate buffer (PB, pH 7.4) for 3 times (15 min each), transferred into 1 % osmic acid (Ted Pella Inc.) for 1–2 h at room temperature, then washed in 0.1 M PB (pH 7.4) for 3 times (15 min each). Leaf tissue blocks were dehydrated with different ethanol concentrations (30 %, 50 %, 70 %, 80 %, 90 %, 95 %, 100 % and 100 %) in sequence for 15 min each, finally were placed in isoamyl acetate (Sinaopharm Group Chemical Reagent Co. LTD) for 15 min. After 0.1 M PB (pH 7.4) wash, the samples were dried with critical point dryer (Quorum, United Kingdom), then were attached to metallic stubs by using carbon stickers and sputter-coated with gold for 30 s, finally were observed with SEM (Hitachi, Japan). Transmission electron microscope (TEM) Sampled fresh tea leaves were cut into 1 mm^3 tissue blocks within 3 min, then quickly fixed in the electron microscopy fixative (Servicebio, Wuhan, China) at room temperature for 2 h, finally stored at 4 ℃. After fixation, leaf tissue blocks were rinsed with 0.1 M PB (pH 7.4) for 3 times (15 min each). Post-fixation: tissue blocks were fixed in 1 % osmic acid (Ted Pella Inc.) that prepared in 0.1 M PB (pH 7.4) at room temperature for 7 h away from light, then washed three times with 0.1 M PB (pH 7.4), 15 min each time. Tissue dehydration: treated tissues were dehydrated in 30 %, 50 %, 70 %, 80 %, 95 %, 100 % alcohol in sequence, 1 h each time, and then in ethanol: acetone = 3: 1, ethanol: acetone = 1: 1 and ethanol: acetone = 1: 3, for 0.5 h each, respectively, finally in pure acetone for 1 h. EMBed 812 (SPI, USA) was used for osmotic embedding, then the embedded plates were placed in 65℃ oven to polymerize for 48 h. Treated blocks were cut into thin slices (60–80 nm) with ultra-microtome (Leica, Germany). The copper mesh slices were dyed in the dark with uranium acetate saturated alcohol solution (2 %) for 8 min, rinsed in ethanol (70 %) for 3 times and in ddH[2]O for 3 times, dyed in lead citrate solution (2.6 %, without CO[2]) for 8 min, washed with ddH[2]O for 3 times, finally put into a copper mesh box to dry overnight at room temperature. The prepared slices were observed with TEM (Hitachi, Japan). Microscopic examination Diseased tea leaves exhibiting blister blight with abundant basidiospore were observed with a U-TV0.5XC-3 microscope (Olympus, Japan). The shapes and sizes of basidiospore were recorded by measuring at least 30 randomly selected basidiospore. Microbial diversity analysis To analysis the composition of microbial communities in diseased lesion tissues showing blister blight in Quxian, Sichuan province, China, genomic DNA of lesion tissues was extracted by using HiPure Soil DNA Kits (Magen, Guangzhou, China) based on the manufacturer’s protocols. ITS gene region was used for the identification of fungal communities. Gene region of ITS2 was amplified with primer pair ITS3_KYO2 (5′-GATGAAGAACGYAGYRAA-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′). The amplified products were purified and then quantified using Nanodrop procedures. Sequencing was conducted by Genedenovo Inc. (Guangzhou, China) by using Illumina Hiseq 2500 PE250 platform (Illumina, San Diego, CA, USA). The α-diversity indices such as Chao1, Simpson, and Shannon were quantified to show the fungal diversity in diseased tea leaves according to the Operational Taxonomic Units (OTUs) richness in QIIME (version 1.9.1) ([54]Caporaso et al., 2010). The species distribution river map of the fungal composition was plotted using R project ggplot2 package (version 2.2.1). The heatmap of species abundance was constructed based on pheatmap package (version 1.0.12) in R project. Functional group (guild) of the microbial communities were inferred by FUNGuild (version 1.0). The phylogenetic tree was constructed by Neighbor-Joining (NJ) method in MEGA6. The distances of evolutionary were computed by the method of Maximum Composite Likelihood. Sensory evaluation Quantitative descriptive analysis was conducted by a sensory panel with seven well-trained panelists (4 males and 3 females, 40–55 years old) from the Department of Tea Science, Southwest University, China. All the panelists had been informed of the sensory evaluation, and agreed to take part in this research and use their information. The rights and privacy of all the panelists were protected during the research, and ethical approval for the involvement of human subjects in this study was granted by Southwest university research ethics committee. Total 3.0 g tea sample was weighed, and brewed with 150 mL boiling water for 5 min, then tea infusion was used for taste evaluation. Taste evaluation included the common characteristics such as bitterness, astringency, umami and sweetness ([55]Cao et al., 2021). Sensory evaluation was conducted by the national sensory evaluation method for tea GB/T 23776–2018 ([56]Gong et al., 2018). Values of taste attributes were scored by a 10-point scale and 5 intensity were evaluated in given tea samples: 0–2 represents extremely weak, 2–4 represents weak, 5–6 represents moderate, 7–8 represents strong, and 9–10 represents extremely strong. The highest and lowest scores for each attribute were removed and the mean value was used for the evaluation. Sugars analysis by GC–MS The content and composition of sugars were analyzed using gas chromatography-mass spectrometry (GC/MS) ([57]Chen et al., 2023). Tea sample (20 mg) was mixed with 500 μL methanol: isopropanol: water (3: 3: 2, V/V/V) solution, vortexed for 3 min, ultrasound for 30 min, and then centrifuged at 12,210 g (3 min, 4℃). Agilent 8890 gas chromatograph coupled to a 5977B mass spectrometer with a DB-5MS column (J&W Scientific, USA) was used for GC/MS analysis. Helium was the carrier gas with a flow rate of 1 mL/min. Injections were made with a split ratio 5:1 and the total injection volume for each sample was 1 μL. The oven temperature was set at 170℃ for 2 min, then increased to 240℃ at 10 ℃/min, finally raised to 280 °C at 5 °C/min, raised to 310 °C at 25 °C/min and kept for 4 min. Selective ion monitoring mode was used for sample analysis. The orthogonal projections to latent structure-discriminant analysis (OPLS-DA) were applied to distinguish the significantly differential metabolites between different tea samples. Significantly changed metabolites between healthy and diseased tea leaves were determined by absolute Log2FC (fold change), variable importance in project (VIP) and P value (Student’s t test). When the fold change ≥ 2 or ≤ 0.5, P value ≤ 0.5 and VIP ≥ 1 between two groups, the metabolites difference was considered significant. The metabolites annotated in Kyoto Encyclopedia of Genes and Genomes (KEGG) compound database ([58]https://www.kegg.jp/kegg/compound/), were then mapped into metabolic pathway based on database search ([59]https://www.kegg.jp/kegg/pathway.html). The P value represented the metabolites enrichment in a pathway, and P ≤ 0.05 was indicative of significant enrichment. Catechins, caffeine and amino acids analysis by HPLC Catechins and caffeine contents were analyzed by using high performance liquid chromatography (HPLC) (Agilent 1200, Agilent Technologies, Santa Clara, CA, USA) ([60]Chen et al., 2023). Prepared tea infusions were analyzed by an Agilent Eclipse XD8 C18 column (250 mm × 4.6 mm i.d., 5 μm; Thermo Electron Corporation, Waltham, MA, USA). The volume of injected sample was 5 μL with 0.9 mL/min flow rate. The wavelength of UV detection was 278 nm. The composition and content of amino acids were analyzed with HPLC system ([61]Chen et al., 2023). The HPLC conditions were as follows: mobile phase A was 4 mM sodium acetate (pH = 5.5); mobile phase B was 80 % acetonitrile solution; column temperature was 35 °C; flow rate was set at 0.9 mL/min. The wavelength of UV detection was 360 nm. The results were expressed as mean value ± standard deviation (SD). The t test of independent sample was used to calculate the statistically significant difference between different groups (P ≤ 0.05) by IBM SPSS Statistics 20.0 (SPSS Inc., Chicago, USA). Results and discussions Disease symptoms on tea caused by blister blight disease As an important tea foliar disease, blister blight mainly infected young organs and tissues including tender leaves, buds, petioles and stems ([62]Fig. 1), directly affecting tea industry qualitatively and quantitatively. The damage to tender shoots and leaves was the most serious, while mature leaves and stems were not susceptible to blister blight disease. The symptoms caused by blister blight were firstly described in details by [63]Petch (1923). Blister blight usually has a short life cycle of 11–28 days due to the different climatic conditions in different locations ([64]Sen et al., 2020). In this study, the initial symptoms of blister blight disease usually infected tender leaves that caused light green, yellow-green, light yellow or slightly red tiny water-soaked translucent spots after 5–15 days of infection, then the spots gradually developed into well-defined circular and larger lesions with the diameter of 1–13 mm ([65]Fig. 1A). Simultaneously, the spots were sunken into slightly concaves, and on the downside, they bulged correspondingly, finally resulting in typical blister lesions, but there were also a few lesions that bulged towards the upper of the leaves ([66]Fig. 1A). The concave upper surface of the blister was smooth and shiny, and the convex surface was generally thickened with powdery coating, then become pure white velvety due to the intensive growth of basidiospores ([67]Fig. 1B). At the most serious stage, the adjacent blisters might fuse into large irregular lesions, and diseased tender shoots became distorted or irregularly rolled. At late stage, the mature basidiospores were released and spread by airflow, and the blisters turned dark brown or purple-red ulcer-like and dried up, or even form holes ([68]Fig. 1A), and might be infected by other saprophytic fungi. When young stems and petioles were infected, the diseased tissues showed obvious swelling, which turned to dark brown withered spots in the later stage ([69]Fig. 1C). The damage to stems was more serious, as the infected stems bent over and broke off at the affected spot, affecting the growth of tea plants. Fig. 1. [70]Fig. 1 [71]Open in a new tab Symptoms of blister blight disease on tea plants. (A) Tea leaves infected with blister blight disease at different stages. (B) Blisters on lower surface of tea leaf. (C) Blisters on stems. Blister blight is a low temperature and high humidity type disease, and the mycelium is latent in the living tissue of the diseased leaves for over-wintering and over-summering. When the climatic conditions were suitable, the basidiospores fell to the water droplets on the young leaves or new shoots of the tea trees with the wind, germinated and invaded the tea leaves in the water environment, and then the basidiospores reproduced in large numbers, so repeatedly infected, thus causing large yield and economic loss ([72]Sinniah et al., 2016, [73]Mur et al., 2015, [74]Sen et al., 2020). It has been reported that a white powder coating on blister can produce two million spores in 24 h which represents an enormous consumption of metabolites and energy of tea ([75]Baby, 2002). In Quxian county, Sichuan province of China, blister blight disease mainly occurred in high mountain tea plantations above 1000 m altitude, infection began to appear when the temperature was above 13℃, and with the increase of temperature and humidity, the disease aggravated. Higher temperature coupled with less rainfall and low humidity during August in summer resulted in the disappearance of blister blight disease, and during December to January in winter, blister blight disease also disappeared when the temperature was lower than 12℃ ([76]Table S1). As shown in [77]Table S1, In the main tea growing areas worldwide, the occurrence period and damage degree of blister blight was closely associated with the different climatic conditions, therefore, the disease incidence were remarkably different in different tea plantations such as Sichuan, Guizhou, Anhui, Yunan, Zhejiang and Hunan province of China, and Sri Lanka, India, etc ([78]Wang et al., 2013, [79]Liu et al., 2021, [80]Liu, 2017, [81]Shi et al., 2016, [82]Sinniah et al., 2016, [83]Mur et al., 2015, [84]Ran et al., 2021). The relationships between rainfall, temperature and humidity to the intensity of blister blight show a strong linear regression pattern, which strongly supports that blister blight intensity decreases with reduced intensity of rainfall, rising temperatures and low humidity ([85]Mur et al., 2015). Although the crop loss changes with the nature of tea varieties and geographical locations, there are no cultivars that are completely resistant to blister blight disease in China, Sri Lanka, India or elsewhere. However, tolerant and susceptible tea varieties showed various degrees of physiological and biochemical changes during the infection of blister blight. Prevention and control of blister blight disease in early stage has not been highly effective due to the lack of suitable biological, chemical and cultural methods. So far, a few fungicides (ie. copper oxychloride, nickel chloride hexahydrate, ergosterol biosynthesis inhibitors) were found to be effective in blister blight disease control ([86]Baby, 2002, [87]Sen et al., 2020). However, for tea quality, the chemical elicitors may be an more eco-friendly approach for disease control, which can improve the innate immunity of tea plants by significantly increasing the level of defense molecule ([88]Sen et al., 2020). The development and spread of disease on tea leaf The pathogen Exobasidium usually infected young tissues and harvestable tea shoots resulting in serious crop loss, and the normal physiological metabolism and growth of tea was negatively affected by the infection of Exobasidium ([89]Fig. 1). To reveal the development and spread of the pathogen E. vexans on tea leaves, TBS, SEM and TEM analysis were applied. The structure of tea leaves from top to bottom was upper epidermis, palisade tissue, sponge tissue and lower epidermis. As shown in [90]Fig. 2A, in healthy tea leaves, the cells in the upper epidermis were closely arranged without gap between cells; palisade tissue was composed of closely arranged cylindrical cells, which was perpendicular to the upper epidermal cells; below the palisade tissue was sponge tissue that was consisted of loosely arranged parenchyma cells with irregular cell morphology and large cell gap; the lower epidermis was made up of a layer of flat cell, which was densely covered with villi and stomata. Generally, the pathogen of blister blight disease invaded through the stomata of epidermal cells and was stained dark blue ([91]Fig. 2A). With the proliferation of pathogens, basidiospores grew in clusters by consuming nutrients in tea tissue cells ([92]Mur et al., 2015), broke through the epidermis, and a large number of hyphae expanded into the cells of sponge tissue, resulting in the necrosis of tea leaf cells and the destruction of leaf tissue structure ([93]Fig. 2A). Fig. 2. [94]Fig. 2 [95]Open in a new tab The development and spread of blister blight disease occurring on tea leaves. (A) Tissue structure of tea leaves analyzed by toluidine blue staining. up-EP: up-epidermis; ST: spongy tissue; PT: palisade tissue; down-EP: down-epidermis. The arrows indicate the hyphae of pathogens. bar = 20 μm. (B and C) Scanning electron microscope and transmission electron microscope analysis. CK: healthy tea leaves;CB: infected tea leaves;GC: guard cell;St: stoma;MC: mesophyll cells;Ch: chloroplast; V: vacuole; CW: cell wall; Gr: grana lamella; SL: stroma lamella; M: mitochondria; Ba: basidiospore, basidia. The arrows indicate the basidiospore and hyphae of pathogens. (For interpretation of the references to colour in this figure legend, the reader is referred to