The experimental field was located in Yuxi vegetable research and development center in Xiguan Village, Qitang town, Bishan District, Chongqing. The basic physical and chemical properties of the soil were 1.04 g·kg-1 total nitrogen, 143.93 mg·kg-1 alkali-hydrolyzed nitrogen, 1.12 g·kg-1 total phosphorus, 131.53 mg·kg-1 available phosphorus, 18.76 g·kg-1 total potassium, 237.02 mg·kg-1 available potassium, 36.10 g·kg-1 organic matter, with pH of 5.19 and cation exchange capacity (CEC) of 21.73cmol·kg-1. The pig manure, chicken manure and biogas residue used in the experiment were purchased from a large farm nearby. The biogas residue was fermented from the same pig manure sampling site. The contents of tetracycline antibiotics are shown in table 1. The vegetable planted was Brassica juncea var. gemmifera (Lee et Lin).
Table 1 Contents of tetracycline antibiotics in three organic fertilizers
Types
|
TC/mg·kg-1
|
OTC/mg·kg-1
|
CTC/ mg·kg-1
|
Pig manure
|
81.08
|
23.94
|
80.92
|
Chicken manure
|
28.06
|
527.16
|
0.89
|
Biogas residue
|
2.90
|
41.37
|
0.16
|
Experimental design
A total of 9 treatments were set up in the experiment, including low amount of pig manure (1350 kg·667m-2), LPM; medium amount of pig manure is (2700 kg·667m-2), MPM; high amount of pig manure (4050 kg·667m-2), HPM; low amount of chicken manure (450 kg·667m-2), LCM; medium amount of chicken manure (900 kg·667m-2), MCM; high amount of chicken manure (1350 kg·667m-2), HCM; inorganic NPK fertilizer (50 kg·667m-2), CM; biogas residue (2700 kg·667m-2), BR; No fertilizer, CK. The plot area of the test field was 9 m2, each treatment was repeated for three times, with random block group arrangement. Irrigation, fertilizer, pests and diseases in the test area shall be routinely managed. When harvesting brassica juncea var. gemmifera, soil samples with depth of 0-20cm were collected using "S" sampling method for determination of soil TCs. The content of TCs in edible part of brassica juncea var. gemmifera was determined.
Determination of tetracycline antibiotics in soil and chicken manure
The samples were frozen at -20℃ for 3 days, freeze-dried by a freeze-dryer, ground and screened via a 1mm sieve, and stored at -20℃ for preparation. Soil sample extraction: 2.0000g soil sample was accurately weighed and added into a 50ml centrifuge tube, followed by addition of 10ml of extract, mixed by swirling and shaking for 1min, and then underwent ultrasonic treatment for 10min. The supernatant was collected after centrifugation at 4000r•min-1 for 10min. After that, the supernatant was extracted twice with 8ml and 6ml of extract, respectively. The supernatant extracted three times was combined, and then evaporated to a volume of about 12ml at 40℃. Solid phase extraction: SPE column was activated with 5ml of methanol and 5ml of ultra-pure water in turn, and the extraction solution passed through the column at a flow rate of 1ml·min-1. Then, the columns were washed with 5ml of ultrapure water and 5ml of 5% methanol solution, vacuumed for 20min, and eluted with 5ml of 0.01mol·L-1 methanol oxalate solution to collect eluent. Subsequently, nitrogen was blown to near dry at 40℃, the volume was fixed to 1ml with methanol, and the eluent was stored in the automatic sampler after passing a 0.22μm filter membrane. Sample extraction of chicken manure: 1.0000 g of pig manure/chicken manure sample were weighed and added in a 50ml centrifugal tube, followed by addition of the extraction solution and extract (the process is the same as the soil sample). After that, the extracts were combined, with addition of 5ml of n-hexane for lipid removal, and then evaporated to a volume of about 12ml at 40℃. The SPE column and SAX column were activated with 5ml of methanol and 5ml of ultra-pure water in turn, and the extracted solution passed through the column at a flow rate of 1ml·min-1. The treatment process of other samples was the same as that of soil samples.
The pretreatment method was mainly referred to the method proposed by Jacobsen et al. (2004), with some optimization. Samples of chicken manure were determined by high performance liquid chromatography (HPLC) and soil samples were determined by liquid chromatography tandem mass spectrometry (LC-MS/MS).
Liquid phase conditions: 40 column temperatures, flow rate of 0.4 ml·min-1, injection volume of 2 samples, gradient elution procedure: 1-3 min, 15-40% B phase.3-4 min, 40%-95% B phase;4-5 min, 95% B phase;5-5.1 min, 15%B phase;5.1-8 min, 15% B phase.
Mass spectrometry conditions: ion source: electrospray ion source (ESI); Scanning method: positive ion scanning; Monitoring mode: multi-reflection monitoring mode (MRM). Dry gas: N2; Dry temperature: 400℃; Dry air velocity: 15 L·min-1; Atomizer pressure: 30 psi; Capillary electrical pulse pressure: 4500 V. The LLOQ of tetracycline, oxytetracycline and aureomycin was 00.007 μg·mL-1,0.014 μg·mL-1,0.021 μg·mL-1, respectively. The recoveries of three antibiotics were determined by standard additions method. The average recovery rate of OTC, TC and CTC was 58.3%, 57.1% and 87.4%, respectively, with all variation coefficients less than 11.04%.
Determination of TCs content in plants
Plant samples were analyzed by HPLC-MS/MS. The specific methods are as follows:
Liquid phase: the column temperature was 40℃, the flow rate was 0.4 ml·min-1, the injection amount was 2μL, the gradient elution procedure: 1-3 min, 15-40% B phase; 3-4 min, 40%-95% B phase; 4-5 min, 95% B phase; 5-5.1 min, 15%B phase; 5.1-8 min, 15% B phase.
Mass spectrometry conditions: ion source: electrospray ion source (ESI); Scanning method: positive ion scanning; Monitoring mode is multi-reflection monitoring mode (MRM). Dry gas: N2; Dry temperature: 400℃; Dry air velocity: 15 L·min-1; Atomizer pressure: 30 psi; Capillary electrical pulse pressure: 4500 V. The LLOQ of tetracycline, oxytetracycline and aureomycin were 7 μg·kg-1,14 μg·kg-1,21 μg·kg-1, respectively. The recoveries of three antibiotics were determined by standard additions method. The average recovery rate of OTC, TC and CTC was 76.3%, 77.0% and 87.4%, respectively, which met the analysis requirements.
16S rRNA gene sequencing method
Based on Illumina sequencing technology platform (Genepioneer Biotechnologies, Nanjing, China), 16S rRNA gene sequencing was carried out by PE250 sequencing method. The bacterial 16S rRNA gene V4/V5 region was selected for amplification and sequencing, and 420bp amplification fragments were obtained using primers515F (5’- TGCCAGCMGCCGCGG- 3’) and 907R9 (5’- CCGTCAATTCMTTTRAGTTT- 3’). The coupling was added, and 2×250bp paired-end data was obtained by sequencing based on NovaSeq platform. By splicing, a longer sequence could be obtained, so as to conduct 16S analysis. Raw Data obtained by sequencing was spliced and filtered to obtain Clean Data. Operations such as chimera removal, combining and clustering of data were carried out using Usearch software. In Usearch clustering, Reads were sorted from large to small in terms of abundance, and OTU clustering was conducted according to the similarity level of 97%, then chimeric was removed, and Operational Taxonomic Units (OTUs) were obtained. Next, Reads of each sample were randomly flattened, with corresponding OTU sequences being extracted. The RDP classifier bayesian algorithm was used to plot the dilution curve of Alpha diversity index on the Qiime platform, and the reasonable flattening parameters were selected according to the dilution curve. Qiime was used to analyze the flattened OTU. First, a Read was extracted from each OTU as a representative sequence, which was then compared with the RDB (Relational Database) Database to annotate each OTU. According to the number of sequences in each OTU, the OTU abundance table was obtained, based on which subsequent analysis was carried out.
Data analysis
Duncan's multiple comparison was used to analyze the significant difference between different treatments (P<0.05). Based on the OUT abundance obtained by 16S rRNA sequencing, principal component analysis was used to compare the difference of bacterial community under different treatments. Pearson correlation analysis was carried out to calculate the correlation coefficient between the relative abundance of TCs and that of major phylum level bacteria. CANOCO 4.5 was used to make the graph. PCA, correlation analysis and variance analysis were performed by SPSS Statistics 21.0. Heatmap was plotted by R3.1.0.