Correlation analysis between filamentous fungi and chemical compositions in a pu‐erh type tea after a long‐term storage

Abstract Storage environment caused the difference between Jinhua Pu‐erh tea (JPT) and General Pu‐erh tea. In this study, fungal flora and chemical compositions were analyzed. The results showed that storage environment caused significant (p < .05) differences of theaflavins (TF), theabrownins (TB), tea polyphenols (TP), and water‐soluble sugars (WSS), and a highly significant (p < .01) difference of thearubigins (TR). Aspergillus niger, Aspergillus pallidofulvus, Aspergillus sesamicola, Penicillium manginii, and Aspergillus tamarii were isolated from Pu‐erh teas and identified based on colony characteristics and ITS, β‐tubulin, and calmodulin gene sequences, respectively. A. pallidofulvus, A. sesamicola, and P. manginii were dominant fungi in JPT and generated macroscopic yellow cleistothecia after a long‐term storage. Correlation analysis showed that dominant fungi exhibited significantly (p < .05 or p < .01) positive or negative corrections with TF, TB, TP, WSS, TR, and gallic acid. This study revealed dominant fungi including A. pallidofulvus, A. sesamicola, and P. manginii and their effects on given chemical compositions.

Pu-erh tea could be stored in natural condition to make pu-erh aged tea after a long-term storage for years, and quality increases with aging time (Xu et al., 2019).
The macroscopic yellow cleistothecia called "golden flowers" are traditionally judged as a quality standard of Fu-zhuan tea . Due to the difference in processing techniques, golden flowers were hardly found in solid-state fermentation of pu-erh tea.
However, in a particular storage environment, the golden flowers similar to that in Fu-zhuan tea could appear in Pu-erh tea occasionally named as Jinhua pu-erh tea (JPT) with higher sensory quality, which was chosen in this work to investigate the correlation between filamentous fungi and chemical compositions in the storage. In this paper, five filamentous fungal strains were isolated from pu-erh tea and identified as Aspergillus niger, Aspergillus pallidofulvus, Aspergillus sesamicola, Penicillium manginii, and Aspergillus tamari based on ITS, β-tubulin, and calmodulin gene sequences, respectively. A. pallidofulvus, A. sesamicola, and P. manginii were dominant fungi and generated macroscopic yellow cleistothecia after a long-term storage in relative humid environment. Dominant fungi improved tea quality in aroma and taste, and had significant (p < .05) impacts on specific chemical compounds during the storage.

| Samples
Pu-erh tea samples: JPT and General Pu-erh tea (GPT) were processed and offered by Kunming Dapu Tea CO. Ltd, Yunnan Province, China. Both pu-erh teas were made from identical puerh ripen tea (loose tea) after they were autoclaved, compressed, F I G U R E 1 The processing and storage of pu-erh tea type. Pu-erh tea used in this paper was made from the same batch of sun-dried green tea leaves with identical processing. Due to the different storage environments, the Pu-erh tea used were divided into Jinhua Pu-erh tea and General Pu-erh tea according to appearance and dried, and were stored at different natural conditions in Kunming, Yunnan Province, China. The processing and storage of Pu-erh teas are shown in Figure 1. After a long-term storage for about 15 years, macroscopic golden cleistothecia appeared on the surface and in the inside of the Pu-erh tea stored at relative humid environment that was defined as JPT; GPT stored at relative dry environment was defined as control without visible golden flowers.

| Isolation and calculation of filamentous fungi
Tea samples were used to isolate the fungi using potato dextrose agar (PDA) medium, and they were counted by dilution-plating method (Wang, Peng, & Gong, 2011). The colony-forming units were calculated by per dry weight gram of tea leaves after cultivation at 30°C for 2 days (Wang, Gong, Chisti, & Sirisansaneeyakul, 2015).

| Fungal identification
The colony morphological characteristics and conidia structure of isolated filamentous fungi were observed after cultivation on PDA medium and Czapek yeast extract agar medium at 25°C for 7 days (Zhou, Ma, Wang, & Xia, 2018). The filamentous fungus grew aerobically as pure cultures in 20 ml of Czapek Dox medium in 125-ml shake flasks at 30°C, 250 rpm, for 2 days. Fresh cells were obtained by centrifugation at 1,700 g for 5 min and freeze-dried at −80°C (Wang et al., 2015;Zhao et al., 2013). DNA was extracted by using SP Fungal DNA Kit. The fungal primers ITS1 and ITS4, Bt2a and Bt2b, and CF1L and CF4 were used in the PCR to amplify the ITS, β-tubulin, and calmodulin regions, respectively (Abe et al., 2008).
The PCR procedure was as follows: pre-degeneration at 95°C for 5 min, degeneration at 94°C for 1 min, annealing at 54°C for 1 min, extension at 72°C for 1.5 min, with 35 cycles, and extension at 72°C for 10 min (Abe et al., 2008). It was stored at 10°C in the end of reaction process.
The received sequences were sent to GenBank of NCBI to seek similar sequences of type strain by using BLASTn (Wang et al., 2015).
Multiple sequence alignment was carried out by using Clustal X for Windows. The evolution distance was calculated through a Kimura 2-parameter of the MEGA 4.0 Soft. Neighbor-joining method was used to establish phylogenetic trees.

| Sensory evaluation
The sensory panel composed of seven panelists selected from 10 professional tea tasters, and the selection was based on evaluation TA B L E 1 Sensory evaluation and score of General pu-erh tea (control) and Jinhua pu-erh tea Note: Score was present by mean value ± SD of three replications. Total score was the summation of products of factors and score coefficient. Total score = 25%*(a) + 10%*(b) + 25%*(c) + 30%*(d) + 10%*(e).
*There is significant difference in p < .05 level, and **a highly significant difference in p < .01 level using the independent t test of SPSS 20.0.
performance of consistency and reliability (Qin et al., 2013). Sensory evaluation was based on five factors, including appearance (a), liquor color (b), aroma (c), taste (d), and infused leaves (e) according to China National Institute of Standardization (CNIS) GB/T 23776-2018 (Gong et al., 2018). The evaluation procedures were as follows: A total of 100-150 g of tea samples were prepared for the evaluation of appearance; 3 g of samples was infused in 150 ml boiled water for 5 min; and the liquor color, aroma, taste, and infused leaves were evaluated after they were infused, respectively. Total score was estimated as follows: Total score = 25%*(a) + 10%*(b) + 25%*(c) + 30%*(d) + 10%*(e).

| Determination of chemical compositions
Moisture content and water extract content in each pu-erh tea sample were developed according to CNIS GB/T 8304-2013  and GB/T 8305-2013 (Zhou, Xu, Lu, Wang, & Sha, 2013), respectively. Contents of tea polyphenols (TP) and total free amino acids were determined by using the spectra photometric method based on FeSO 4 and the ninhydrin assay (Liang, Zhang, & Lu, 2005), respectively. The main tea pigments, including theaflavins (TF), thearubigins (TR), and theabrownins (TB), were analyzed using the spectra photometric method described by Wang et al. (2011).
Water-soluble sugar (WSS) content was determined by visible spectrophotometer method with anthraquinone on 620 nm (Liang et al., 2005

| Statistical analysis
Each pu-erh tea sample was measured in three replications. All data are presented as mean value ± standard deviation (SD Note: The content of total catechins was the summation of C, EC, EGC, ECG, and EGCG. All data were present by mean value ± SD of three replications. *There is a significant difference at p < .05 levels, and **A highly significant difference at p < .01 levels using the independent t test of SPSS 20.0.

| Quality characteristic analysis of different puerh teas
Quality characteristics were calculated by sensory evaluation based on five sensory factors. Significant difference analysis was carried out by using independent t test of SPSS 20.0.
Results of sensory evaluation in GPT (control) and JPT are shown in Table 1. Due to both pu-erh teas made from identical pu-erh ripen tea (loose tea) after they were autoclaved, compressed, and dried, quality characteristics in appearance, liquor color, and infused leaves were similar with no significant differences (p > .05). However, due to the difference in storage environment, particularly the appearance of macroscopic golden cleistothecia, JPT had better aroma (p < .05) and taste (p < .01).
Therefore, storage environment had significant (p < .05) impact on aroma and taste.

| Biological identification of filamentous fungi
In this paper, about nine different fungal strains were isolated from pu-erh teas. Five filamentous fungi could be detected with a high abundance and named as GPTS1, JPTS1, JPTS2, JPTS3, and GPTS2, respectively, which were superior in numbers. Others were occasionally detected in pu-erh teas, which belonged to the genera

F I G U R E 2 (Continued)
in Table 3. Strain GPTS1 was identified as A. niger based on colony characteristics and conidial structure, which has been described in our previous study (Zhou et al., 2018

| Distribution of fungal consortium in puerh teas
The distribution and difference of fungal consortium are presented in in JPT had significant (p < .05) increase due to a high moisture content ( Figure 3b). Particularly, A. pallidofulvus and A. sesamicola had a significant (p < .05) or a highly significant (p < .01) increase in JPT, respectively. Only A. niger had a significant (p < .05) decrease in JPT.

The fungal flora, particularly the dominant fungi found in JPT including
A. pallidofulvus, A. sesamicola, and P. manginii, caused significant differences of given chemical components in JPT after a long-term storage.
Correlation analysis was conducted later to investigate the relationship between each isolated filamentous fungus and chemical components, that is TF, TR, TB, TP, GA, and WSS, respectively.

| Effects of isolated filamentous fungi on chemical compositions
In this paper, A. niger, A. pallidofulvus, A. sesamicola, A. tamarii, and P. manginii were isolated from pu-erh teas. Particularly, A. pallidofulvus, A. sesamicola, and P. manginii were dominating fungi in JPT, which generate yellow or golden color cleistothecia in the storage.
The relationships between each isolated filamentous fungus and chemical compound were evaluated by bivariate correlation analysis using SPSS 20.0 for Windows. The results are presented in Table 4.

| D ISCUSS I ON
Solid-state fermentation plays an important role in the processing of pu-erh tea (Abe et al., 2008). Fungi have profound impact on substance metabolisms and show correlation with quality formation of pu-erh tea (Zhao et al., , 2019 Zhang et al., 2016;Zhao et al., 2015;Zhou et al., 2018). In this study, five filamentous fungi were isolated from pu-erh teas and identified as A. niger, A. pallidofulvus, A. sesamicola, A. tamarii, and P. manginii, respectively. The detection of A. niger and Aspergillus tamari in pu-erh tea storage has a close connection with solidstate fermentation of pu-erh tea.
The strain generating yellow color cleistothecia is termed as the "golden flower fungus" (Mo, Zhu, & Chen, 2008;Mao, Wei, Teng, Huang, & Xia, 2017). The species within the genera Aspergillus, Eurotium, and Penicillium are main fungal taxa isolated from the postfermentation of dark tea (Mo et al., 2008). Aspergillus cristatus and Eurotium cristatum are golden flower fungi involved in the fermentation of Chinese fu-zhuan tea (Ge et al., 2016;Zou et al., 2014). Numerous macroscopic yellow cleistothecia similar to that in other dark tea also could appear in pu-erh tea after a long-term F I G U R E 3 Distribution (a) and differences (b) of isolated filamentous fungi in pu-erh tea type. GPT, General pu-erh tea. JPT, Jinhua puerh tea. About nine different kinds of fungal strains were isolated from pu-erh tea samples based on PDA medium, and those colony-forming units were calculated. Five filamentous fungi were isolated and identified. Aspergillus pallidofulvus, Aspergillus sesamicola, and P. manginii were dominant fungi in count and generated macroscopic golden cleistothecia after a long-term storage in relative humid environment. Others were insignificant fungal strains in count, which belonged to the genera Aspergillus, Penicillium, and Saccharomyces according to the colony morphological characteristics and conidia structures. * indicates there is a significant difference at p < .05 level, and ** indicates a highly significant difference at p < .01 level using the independent t test of SPSS 20.0  *A significant correlation at p < .05 level, **A highly significant correlation at p < .01 level.
storage, which was defined as JPT. With GPT made from the same pu-erh ripen tea (loose tea) with identical processing as the control, through comparison of fungal consortium, A. pallidofulvus, A. sesamicola, and P. manginii were dominant fungi found in JPT and generate the appearance of macroscopic yellow cleistothecia after a long-term storage, which had a significantly (p < .05) positive correlation with moisture content in pu-erh tea cake.
The individual numbers of fungi in pu-erh tea decreased significantly after solid-state fermentation through autoclaving, compressing, and drying (Tian et al., 2013). The fungal flora is maintained at a relatively low level. Due to a long-term storage for years, dominant fungi also had significant impact on chemical compositions during the storage, which caused significant (p < .05) differences of TF, TB, TP, and WSS contents, and a highly significant (p < .01) difference of TR between JPT and GPT (

| CON CLUS IONS
Our present work describes fungal diversity and differences in chemical compositions between GPT and JPT to study effects of dominant fungi in pu-erh tea storage. Due to the appearance of macroscopic yellow cleistothecia, JPT had better aroma (p < .05) and taste (p < .01) ( Table 1) and had significant (p < .05) differences of TF, TB, TP, and WSS contents, and a highly significant (p < .01) difference of TR compared with the control (  (Table 4).
Particularly, A. pallidofulvus and P. manginii exhibited a highly significantly positive correlation (p < .01) with WSS and a significantly (p < .05) positive correlation with TF, while A. sesamicola showed a highly significantly negative correlation (p < .01) with TB and a significantly (p < .05) positive correlation with GA, respectively. Those results indicated that the isolated filamentous fungi had significant impact on given chemical compositions in the storage, which provides reference for the application of A. pallidofulvus, A. sesamicola, and P. manginii in tea.

ACK N OWLED G M ENTS
The authors are very grateful for the financial supports from

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.

AUTH O R CO NTR I B UTI O N S
B. Zhou and C. Ma designed the research and wrote the paper. C.
Ma and C. Zheng participated in the performance of the research.
X. Liu conducted the statistical analysis. T. Xia contributed to the writing and data analysis. All authors read and approved the final manuscript.

E TH I C A L A PPROVA L
The study did not involve any human or animal testing.