Preparation of samples
S. suberectus Dunn decoction pieces (batch number: 210801, origin: Guangxi) was purchased from Chinese herbal medicine market of Zhangshu City, Jiangxi Province. Decoction pieces with reddish brown, large number of catheter holes and similar size were selected as samples. The samples were divided into three equal groups and each group was parallel three times. According to the accelerated experimental method (Liang 2019), the samples were placed in an environment with a temperature of 28℃ and relative air humidity of 95%. The samples at first day were taken as samples at the early stage of mildew (A), The samples at fifth day were taken as samples at the middle stage of mildew (B), and the samples at ninth day was taken as samples at the late stage of mildew (C). The samples were dipped into sterile water and shaken at speed of 2000 rmp for 1.5h, then enriching fungi of the extract with 0.22-µm detachable filter, the filters were collected and stored at -80℃ for further use. The remaining samples of each group were pulverized into powder and screened by 40 mesh, then stored at -80℃ for further use.
Extraction of genomic DNA from samples
Fungi were collected from detachable filters, DNA of samples was extracted with DNA kits, and the purity and concentration of DNA were determined by 1% agarose gel electrophoresis.
ITS library construction and high-throughput sequencing
Using the DNA of the sample as a template, ITS1F: 5'-CTTGGTCATTTAGAGGAAGTAA-3' (5µmol/L) and ITS2R: 5'-GCTGCGTTCTTCATCGATGC-3' (5µmol/L) were used for PCR amplification of the ITS region of samples. The PCR amplification volume was 20 µl: 5×FastPfu Buffer, 4 µl; dNTPs (2.5mmol/L), 2µl; Forward Primer (5µmol/L), 0.8µl; Reverse Primer (5 µmol/L), 0.8 µl; FastPfu Polymerase, 0.4µg; BSA, 0.2 µl; DNA, 10 ng; Fill with double steaming water to 20 µl; The sequence parameters of PCR reaction were 95°C and 3 min. Qualified purified samples were analysis by high-throughput sequencing with platform of Illumina NovaSeq6000.
Sequencing data processing and statistical analysis
Raw data preprocessing
At first, the Trimmomatic v0.33 software (Bolger et al. 2014) was used to filter Raw Reads, and the Cutadapt 1.9.1 software (Martin 2011) was used to identify and remove primer sequences to obtain Clean Reads, which did not contain primer sequences. Finally, the dada2 method (Callahan et al. 2016) in the QIIME 2020.6 software (Bolyen et al. 2019) was used for denoising, double-ended sequences were spliced and chimeric sequences were removed to obtain the final effective data (Non-chimeric Reads)
OTU clustering and annotation
The Uparse software (Segata et al. 2011) was used to cluster all sample non-chimeric reads, and the sequences were grouped into operational taxonomic units (OTUs) with 97% identity by default. The blast method (Altschu et al. 1997) in the QIIME 2020.6 software (Bolyen et al. 2019) was used to annotate each sequence by species classification, and the Unite fungal database [27] was used for classification annotation. The Shannon diversity index dilution curve was used to analyze the sequencing depth, and α-diversity, species composition, Principal ordinates analysis (PCoA) and Non-MetricMulti-Dimensional Scaling (NMDS) were used to evaluate the diversity structure of fungal community in the S. suberectus Dunn decoction pieces during the process of mildew.
Determination of Aflatoxin B1 content
5.0 g of sample powder was accurately weighed and mixed with 25 ml of 70% methanol for oscillating extraction at 200 rpm for 10 min, proper amount of extract was transfer into a centrifuge tube, centrifuge at 5000 rmp for 10 min, and the middle layer of supernatant was extract. 0.1 ml of extract mixed with 9 ml of sample diluent solution to obtain the test solution (dilution coefficient K = 50). The measured liquid will be detected by ELISA method. According to the regression equation of the AFB1 standard curve (1):
$$Y=-0.3017X+0.7245 ({R}^{2}=0.9229)$$
1
the corresponding logarithmic value lgC of concentration (C) can be obtained, then calculated its antilog to obtain the AFB1 concentration (C) of test solution, the content of AFB1 in the sample (W) can be calculated according to the Eq. (2):
W. AFB1 content of samples (µg/kg)
C. AFB1 content of test solution (µg/kg)
K. Dilution coefficient of test solution
Determination of total flavonoids content
Referring to the extraction method of total flavonoids from S. suberectus Dunn by Pan, G.H. (Pan et al. 2022). The 0.3 g of sample powder was mixed with 25 ml of 50% ethanol solution, which was extracted by using ultrasonic oscillation at 300W, 45Hz and 50℃ for 1.5 h, then the solution was filtered and centrifuged at 8000 rmp for 10 min, the supernatant was the extract. The content of total flavonoids of extract was detected by aluminum nitrate colorimetric method (Lee et al. 2011). 1 ml of the extract was placed into 25 ml volumetric flask, mixed with 1 ml of 5% sodium nitrite, shaken well and left for 6 min, mixed with 1 ml of 10% aluminum nitrate, shaken well and left for 6 min, mixed with 10 ml of 1 mol/L sodium hydroxide solution, filled with 50% ethanol to the mark, then shaken well and left for 15 min. A510nm were measured by UV-spectrophotometer, and the concentration of total flavonoids in the samples were calculated according to the regression equation of rutin standard curve (3):
\(Y=13.175X+0.0012 ({R}^{2}\)=0.9985) (3)
Determination of protocatechuic acid, catechin and epicatechin content
Preparation and content detection of extract
Referring to the method of detecting flavonoids content in the S. suberectus Dunn by Du, J.S.(Du et al. 2017). 1.0 g of sample powder was accurately weighed and mixed with 10 ml of 80% methanol. Ultrasonic extraction was conducted at a frequency of 40Hz and power output of 500W for one hour at a temperature of 50℃, followed by filtration. The filtrate was collected and centrifuged with speed of 12,000 rmp for 10 min at 4℃. The extract was obtained by diluting the supernatant 10 times and filtered with 0.22-µm organic membranes.
Drawing of standard curves
The standard products of 4.90 mg of catechin, epicatechin and protocatechuic acid were weighed and dissolved with a small amount of 80% methanol respectively, then transferred into 5 ml volumetric flasks, filled with 80% methanol to produce 0.98 mg/ml standard stock solutions. The stock solution of each standard was accurately measured and diluted with 80% methanol to prepare seven different concentrations of mixed standard solutions, of which the mass concentrations of catechin standards were 196, 98, 49, 24.5, 12.25, 6.125, 3.0625 µg/ml, the mass concentrations of epicatechin standards were 392, 196, 98, 49, 24. 5, 12.25, 6.125 µg/ml, and the mass concentrations of protocatechuic acid standards were 196, 98, 49, 24.5, 12.25, 6.125, 3.0625 µg/ml. The mixed standard solution was filtered through 0.22-µm filter and then detected according to the chromatographic conditions in 2.7.3. The standard curves were drawn with the concentration (µg/ml) as the horizontal coordinate and the peak area as the vertical coordinate, and the regression equations were calculated respectively as follows: Eq. (4) represents catechin; Eq. (5) represent epicatechin, Eq. (6) represent protocatechuic acid:
\(Y=9814.6X-8090.3 ({R}^{2}=0.999\)) (4)
$$Y=10985X-2632.8 ({R}^{2}=0.999)$$
5
$$Y=22458X-12893 ({R}^{2}=0.999)$$
6
Chromatographic conditions
Columns: Phenomenex Kinetex C18 (250 mm × 4.6 mm, 5 µm); Mobile phase: acetonitrile (A), 0.1% formic acid solution (B); Gradient elution (0 ~ 5 min, 9% ~ 11% A; 5 ~ 10 min,11% ~ 12% A; 10 ~ 30 min, 12% ~ 16% A; 30 ~ 39 min, 16% A; 39 ~ 50 min, 16% ~ 19% A); Detection wavelength: 278 nm; Flow rate: 0.7 ml/min; Column temperature: 25 ℃; Sample size: 10 µl.