Background

Artemisia sp. plants belong to the Compositae families Anthemideae and Artemisiinae [1]. There have been estimated total of 344 species and 69 varieties of Artemisia in the world; Asia contains the most types with 269 species and 60 varieties [2]. There are 187 species and 46 varieties of Artemisia have been found throughout China [3]. Artemisia is tenacious and can grow at high altitude and in extremely arid areas [4, 5] which are widely distributed in Tibet and are the major species in desert steppe and grassland. Survey results showed that 57 species and 6 varieties of Artemisia species distributed in Tibet, accounting for a quarter of the national Artemisia species [6]. A. wellbyi has been shown to be of higher nutritional quality containing higher crude protein and crude fat content than other herbs found in Tibetan grasslands [7], and has the potential to become a supplementary grass seed for ecological restoration of grasslands in the Tibetan plateau [8]. A. sieversiana is mainly used for hay, a reserve feed for cattle and sheep in winter [9], and an important source of animal feed in Tibet [10]. In addition, A. sieversiana plants can be used as high-quality roughage after silage [11]. A. annua extracts added to feed can promote animal growth, improve the body's disease resistance, and improve animal production performance [12].

Artemisia contain a large class of medicines widely used traditionally by Tibetans. Traditional Chinese medicinal practitioners believe that this genus has antibacterial and anti-inflammatory and has wide range of health beneficial properties [13,14,15]. It is widely used in malaria, hepatitis, cancer, inflammation, infection and other diseases. In 2015, Tu Youyou was awarded Nobel Prize in Physiology or Medicine for her discovery of the antimalarial sesquiterpenoid artemisinin from A. annua. Since then, artemisinin and Artemisia have attracted worldwide attention. In recent years, Xiao [16] used a variety of chromatographic methods to separate and purify the compounds from the aqueous fraction of the aerial portions of A. annua, and identified 15 compounds based on the physicochemical properties and NMR spectral data. Wang [17] et al. studied the chemical constituents of the whole plant of A. annua by chromatography with silica gel matrix and HPLC, and purified 17 compounds from the ethyl acetate extract from the ethanolic extract of A. annua and Zhong [18] et al. extracted.and isolated 6 flavonoids from A. annua.

Artemisia produces many medicinally important secondary metabolites that have antimicrobial and antioxidant activities. External application of artesunate can inhibit S. aureus, D. Bacillus, B. subtilis, P. aeruginosa. Three kinds of extracts of A. annua (petroleum ether extract I; chloroform extract II; ethanol extract III) have antifungal effects; the antifungal activity of extract III is close to that of clinical routine antifungal drug [19]. Artemisia essential oils have strong antibacterial effects on S. aureus, S. epidermidis, E. coli, and Streptococcus, and have strong antioxidant effects [20].

While the metabolites from A. annua have been well characterized, the metabolites from other Artemisia sp. and particularly those from Tibet have not been thoroughly identified. Presently, the types of metabolites of Artemisia plants and the differences in metabolites among these plants are not clear. In addition, studies on the metabolites of the Artemisia genus have been limited, with low sensitivity, and relatively poor qualitative and quantitative accuracy [21,22,23]. Widely targeted metabolomics integrates the advantages of untargeted metabolites and targeted metabolite detection techniques to achieve high throughput, high sensitivity and broad coverage. We used metabolomic analysis to identify the metabolites from A. sieversiana, A. wellbyi, and A. annua and elucidate the differential metabolite species. The antibacterial and antioxidant capabilities were also evaluated on the three Artemisia species from Tibet. This study will provide new evidence for the potential medicinal use of the three Tibetan Artemisia species and lay the foundation for further exploration of the active constituents, their metabolic pathways, and pharmacological mechanisms of action.

Results

Qualitative and quantitative analysis of the metabolites

The primary metabolites and secondary metabolites in the samples were identified by UPLC-MS. 1109 metabolites were identified from 3 species of Artemisia, including 79 amino acids and their derivatives, 73 nucleotides and their derivatives, 101 organic acids, 155 lipids, and 168 phenolic acids, 227 flavonoids, 40 lignans and coumarin, 86 alkaloids, 56 terpenes, and 124 others (Supplementary 1). Metabolic profiles differed by Artemisia species. Total ion chromatograms of the metabolite analysis were shown in Fig. 1.

Fig. 1
figure 1

Total ion chromatograms of the metabolite analysis. A QC_MS_TIC-N. B QC_MS_TIC-P

Full size image

Sample quality control and statistical analysis

The results showed that the contribution rate of principal component 1 (PC1) was 49.57%, and PC2 was 42.58%, and the three groups of samples were separated in the two-dimensional diagram (Fig. 2). The differences in metabolites between the three Artemisia sp. are shown in the PCA results.

Fig. 2
figure 2

Principal component analysis. A All samples principal component analysis. B AS vs AA principal component analysis (PCA). C AS vs AW principal component analysis (PCA). D AW vs AA principal component analysis (PCA)

Full size image

Differences in accumulation patterns of metabolites from the three Artemisia sp. were analyzed by clustering heatmaps (Fig. 3). The heat map analysis showed the differences in substances within the plants that were grouped into 4 clusters. The metabolites in cluster 1 were the highest in AW group, medium in AS group, and AA group. Metabolites in cluster 2 were highest in AS group, moderately present in AW group, and lowest in AA group. The different biological replicates also found to be clustered together, both cluster analysis and PCA showed that metabolites were significant different in the three Artemisia sp.

Fig. 3
figure 3

Cluster Analysis of Differential Metabolites. A All samples heatmap. B AS vs AA heatmap. C AS vs AW heatmap. D AW vs AA heatmap

Full size image

OPLS-DA analysis of the differential grouping

OPLS-DA was used to analyze the AW, AS, and AA groups in pairs to generate a score map. All the Q2 of the comparison groups were all higher than 0.9, indicating that the constructed model was suitable. According to the OPLS-DA score plot significant separation occurred in the different comparison groups. As shown in Fig. 4, the OPLS-DA model produced two principal components and the contribution rate of PC1 is 76%, and the contribution rate of PC2 is 6%. The difference between the two groups of samples is highly significant. Among the evaluation parameters of the OPLS-DA model, the indicators R2X = 0.878, R2Y = 1, Q2Y = 0.993 were all greater than 0.5 and Q2Y > 0.9, suggesting that the OPLS-DA model was correctly constructed, the prediction was reliable, and the differential metabolites could be screened according to the VIP value analysis.

Fig. 4
figure 4

OPLS-DA score plot. A AS vs AA OPLS-DA. B AS vs AW OPLS-DA. C AS vs AW OPLS-DA

Full size image

Differential metabolite screening

The results of differential metabolites can be shown using Volcano and Wayn maps. Volcano plots visually demonstrated the overall distribution of different metabolites and the results are shown in Fig. 5 showing significant differences between the three Artemisia species. The visual display of specific metabolites and their differences were used for functional analysis of metabolic pathways, with upregulation in red, and downregulation in green and no changes in gray. There were 449 different metabolites of different species identified by the multivariate statistical analyses (Fig. 6).

Fig. 5
figure 5

Volcano plots of the differential metabolites. A AS vs AA Volcano plot. B AS vs AW Volcano plot. C AS vs AW Volcano plot

Full size image
Fig. 6
figure 6

Wayne diagram of the differential metabolite. A All samples Wayne diagram. B AS vs AA Wayne diagram. C AS vs AW Wayne diagram. D AW vs AA Wayne diagram

Full size image

Analysis of the differential metabolites

Table 1 shows the top ranking of 20 differentially expressed metabolic components in the fold change of the distinct metabolites in A. sieversiana and A. annua. Compared with A. annua, clear differences could be seen in A. sieversiana regarding the contents of 2,6-Dimethoxybenzoic acid, Blumeatin, Luteolin-6-C-glucoside, Ethyl maltol, Luteolin-8-C-glucoside, 4,5-Epoxyartemisinic Acid, ageconyflavone B, Fraxidin and Chrysosplenetin.

Table 1 Significant analysis results of different metabolites (AS vs AA)
Full size table

Table 2 shows the top 20 differential metabolites in the samples of A. sieversiana and A. wellbyi. Compared to A. sieversiana, A. wellbyi showed higher levels of 2-Phenylphenol, Reynosin, Rhamnetin, Methyl Cinnamate, Phenyl acetate, 4-Hydroxyacetophenone, Phloretin-4'-O-glucoside (Trilobatin), and Hispidulin-7-O-(6-O-p-Coumaroyl) Glucoside.

Table 2 Significant analysis results of different metabolites (AS vs AW)
Full size table

Table 3 shows the top 20 differentially expressed metabolites in A. wellbyi and A. annua. Compared with A. annua, A. wellbyi contains more 4,5-Epoxyartemisinic Acid, Dihydro Artemisinin-D3, 2,6-Dimethoxybenzoic acid, Dihydro-epi-arteannuin B, 2-Hydroxy-3-phenylpropanoic acid, 1-O-Vanilloyl-D-Glucose.

Table 3 Results of significant analysis of differential metabolites (AW vs AA)
Full size table

The medicinal important metabolites from different species were assayed and compared by three plants (Fig. 7). We identified a total of 227 flavonoids from the three Artemisia species, accounting for 20.4% of the metabolite species. These included flavonoids such as Luteolin, Quercetin, Kaempferol, and Apigenin, which were enriched in AW. Fifty-six terpenoid metabolites were identified, including sesquiterpenoids with important pharmacological effects, such as Artemisinine, Arteannuin A, Artemisinin B, and Dihydroartemisinin. Artemisinin was found in the highest content in A. annua, followed by A. sieversiana. Coumarin, Isoscopoletin and Scoparone were in highest levels in A. wellbyi. Salicylic acid was in the highest level in A. sieversian, while Vanillic acid was found in the highest levels in A. wellbyi.

Fig. 7
figure 7

Peak area integration of the main medicinal functional metabolites of the three Artemisia species. A Artemisinin metabolites. B Flavonoid metabolites. C Phenolic acid metabolites. D Coumarin-class metabolites

Full size image

KEGG enrichment analysis

The differential metabolites of AS group and AA group were mainly enriched in the Purine metabolism pathway, 2-Oxocarboxylic acid metabolism pathway, and Tryptophan metabolism pathway. The differential metabolites of AS group and AW group were mainly enriched in the Purine metabolism pathway, as well as Tryptophan metabolism pathway. The differential metabolites in AW group and AA group were mainly enriched in the 2-Oxocarboxylic acid metabolism pathway, Purine metabolism pathway, and Phenylpropanoid biosynthesis pathway. In these comparison groups, some metabolic pathways overlap, such as Purine metabolism pathway, Tryptophan metabolism pathway, 2-Oxocarboxylic acid metabolism pathway (Fig. 8). Diterpenoid Biosynthesis metabolic pathway related to differential metabolites and bioactive components (Fig. 9).

Fig. 8
figure 8

KEGG enrichment bubble chart on specific characteristics of the important metabolites. A AS vs AA KEGG enrichment bubble chart. B AS vs AW KEGG enrichment bubble chart. C AS vs AW KEGG enrichment bubble chart

Full size image
Fig. 9
figure 9

Diterpenoid Biosynthesis

Full size image

Antibacterial activity of plant extracts

Table 4 shows that each extraction partition of A. sieversiana extract inhibits E. coli, Salmonella, Streptococcus, S. aureus, P. mirabilis, B. cereus, and P. aeruginosa differently. When the mass concentration of each extract was 200 mg/mL, Petroleum ether had better inhibitory effect on these 7 kinds of bacteria and Petroleum ether had the strongest inhibitory effect on S. aureus (p < 0.01).

Table 4 The inhibition zone diameters of A. sieversiana extract against tested bacteria (mm)
Full size table

Similarly, each extraction partition of the A. wellbyi extract has different degrees of inhibition to E. coli, Salmonella, Streptococcus, S. aureus, P. mirabilis, B. cereus, and P. aeruginosa (Table 5). When the mass concentration of each extract was 200 mg/mL, from the perspective of the inhibition degree of each organic relative to various bacteria, the petroleum ether had better inhibitory effect on these 7 kinds of bacteria and petroleum ether had the strongest inhibitory effect on Streptococcus (p < 0.001).

Table 5 The inhibition zone diameters of A. wellbyi extract against tested bacteria (mm)
Full size table

In addition, each extraction part of A. annua extract has different degrees of inhibition to E. coli, Salmonella, Streptococcus, S. aureus, P. mirabilis, B. cereus, and P. aeruginosa (Table 6). Petroleum ether had better inhibitory effect on these 7 kinds of bacteria and Petroleum ether had the strongest inhibitory effect on P. mirabilis (p < 0.001).

Table 6 The inhibition zone diameters of A. annua extract against tested bacteria (mm)
Full size table

Antioxidant activity of plant extracts

The ethyl acetate extraction of A. sieversiana plant showed the strongest antioxidant capacity, followed by n-butanol extraction and petroleum ether extraction. In contrast, the dichloromethane extraction of A. annua plant had the strongest antioxidant capacity, followed by petroleum ether partition and n-butanol partition. The dichloromethane extraction of A. wellbyi had the strongest antioxidant capacity, followed by the n-butanol extraction and the ethyl acetate extraction. The water extracts of the three plants had the weakest antioxidant capacity. Among the three species, the dichloromethane extraction of A. annua has the strongest antioxidant capacity (Fig. 10A).

Fig. 10
figure 10

Antioxidant activity of the three Artemisia species. A FRAP assay. B Scavenging capacity of DPPH free radicals. C Scavenging activity against ABTS free radical. D Determination of hydroxyl radical scavenging capacity

Full size image

DPPH is a stable free radical, soluble in polar solvents such as methanol and ethanol, and has a large absorption at 515 nm. When antioxidants are added to the DPPH solution, a decolorization reaction occurs, so the change in absorbance can be used to quantify the antioxidant capacity of antioxidants with Trolox as a control system. The petroleum ether part of A. sieversiana had the strongest scavenging ability to DPPH free radicals, followed by methylene chloride and n-butanol, and the ethyl acetate part had the weakest scavenging ability to DPPH free radicals; The petroleum ether part of A. annua plant had the strongest scavenging ability to DPPH free radical, followed by ethyl acetate part, and the dichloromethane part has the weakest scavenging ability to DPPH free radical; The scavenging ability of DPPH free radical was the strongest in the dichloromethane part of A. wellbyi followed by the n-butanol part and the ethyl acetate part, and the water extract had the weakest scavenging ability on DPPH free radical. Among the three species, the petroleum ether part of A. sieversiana has the strongest scavenging ability to DPPH free radicals (Fig. 10B).

The dichloromethane extraction of A. sieversiana plant had the strongest scavenging ability to ABTS free radical, followed by petroleum ether partition, and the ethyl acetate part had the weakest scavenging ability to ABTS free radical. The petroleum ether part of A. annua plant had the strongest scavenging ability to ABTS free radicals, followed by the methylene chloride part, and the n-butanol part had the weakest scavenging ability to ABTS free radicals. The ethyl acetate part of A. wellbyi plant had the strongest scavenging ability to ABTS free radical, followed by n-butanol part, and the petroleum ether part had the weakest scavenging ability to ABTS free radical. Among the three species, the dichloromethane site of A. sieversiana had the strongest scavenging ability to ABST free radicals (Fig. 10C).

H2O2/Fe2+ generates hydroxyl radicals through the Fenton reaction, and salicylic acid can effectively capture the generated hydroxyl radicals and react with them to form a colored substance, 2,3-dihydroxybenzoic acid. After the substance is removed, the colored substances will be reduced, so that the ability of the sample to scavenge hydroxyl radicals can be judged according to the value of the absorbance value. The dichloromethane part of A. sieversiana plant had the strongest scavenging ability to hydroxyl radicals, followed by water extract, and the petroleum ether part has the weakest scavenging capacity to hydroxyl radicals; The water extract of A. annua plant had the strongest scavenging ability to hydroxyl free radicals, followed by the dichloromethane part, and the petroleum ether part had the weakest scavenging ability; The water extract of A. wellbyi has the strongest scavenging ability, followed by ethyl acetate, and petroleum ether had the weakest scavenging ability. Among the three plants, the water extract of A. serrata had the strongest scavenging ability (Fig. 10D).

Discussion

In this study, we used widely targeted metabolomics to analyze the primary and secondary metabolites of three Artemisia species collected from Tibet, and identified 1109 metabolites in 10 categories. This compares to the total of 535 metabolites identified using non-targeted metabolomics to analyze three species of Artemisia [24]. The main metabolites identified here were flavonoids, phenolic acids, lipids, amino acids and their derivatives, organic acids, alkaloids, and terpenes. The important pharmacologically active compounds are flavonoids, phenolic acids, artemisinins and coumarin compounds.

The metabolites of three Artemisia species were identified by widely-targeted metabolomics technology, and a total of 227 flavonoids were obtained. Flavonoids were the most abundant metabolites, accounting for 20.4% of the total metabolites. Flavonoids are widely present in Artemisia and are an important class of natural organic compounds [25]. Zhang [26] et al. identified 10 flavonoids from A. sphaerocephala. Among them, the representative quercetin has a wide range of pharmacological effects in antioxidant, anti-inflammatory and antibacterial, anti-tumor [27,28,29]. In addition to these documented compounds, we also detected 42 sesquiterpenoids from these 3 species of Artemisia, such as artemisinin, artemisinin A, artemisinin B, artemisinic acid, dihydroartemisinic acid, etc. Previous studies have found that artemisinin compounds such as ATS can inhibit βIL-1, IL-6, IL-17α and other inflammatory cytokines, suggesting that they play the roles of anti-inflammatory, anti-angiogenesis, inhibiting autoimmune arthritis and treating rheumatoid arthritis [30]. Phenolic acids have significant effects in anti-inflammatory, anti-allergic, vascular protection, antioxidant activity, anti-tumor, anti-bacterial and fungal and liver protection [31] Coumarin compounds have good physiological and pharmacological activities in antiviral, antifungal, anti-tumor, and anti-inflammatory aspects [32].

The antibacterial experiments of the three A. species showed that the different polar solvent extracts from the three A. species had strong antibacterial activities. Zohra [33] et al. used an aqueous extract of A. annua against 3 Gram-negative bacteria and 3 Gram-positive bacteria were evaluated for bacteriostatic activity. Although the antibacterial activity of ACAE is lower than that of ampicillin, at a concentration of only 50 mg/mL, it has the strongest inhibition zone (13 mm) with good inhibition against S. aureus. The study by Darwish [34] et al. showed that the methanol extract of this plant has high antibacterial activity. Widely targeted metabolomics results revealed the presence of derivatives such as flavonoids, terpenoids, phenols and alkaloids. In addition, alkaloids, flavonoids, phenols, and terpenes in various plant extracts have all been shown to be effective antibiotics. Our results are also consistent with these studies showing that these 3 Artemisia species have efficacy against clinical pathogens.

The antioxidant activity test showed that the three species of Artemisia have strong antioxidant capacity in vitro. The antioxidant activity experiments of A. sieversiana essential oil by Li [35] et al. showed that the IC50 of A. annua essential oil on DPPH free radicals, ABTS+ and hydroxyl radicals were lower than vitamin C, indicating that A. sieversiana essential oil had strong in vitro antioxidant capacity, which is stronger than VC. This flavonoid purified product has scavenging ability for hydroxyl radicals, antioxidant activity to grease, that is stronger than citric acid. It has stronger antioxidant activity on vegetable oils than ascorbic acid, and slightly weaker than ascorbic acid on animal fats and oils. The residue of A. annua is rich in flavonoids and has strong antioxidant activity, which is a natural antioxidant. Our findings are consistent with some studies [36] showing that 3 Artemisia species have antioxidant effects, and their antioxidant properties may be related to the phenolic and flavonoid content of Artemisia.

We found that A. sieversiana and A. wellbyi collected from Tibet, are likely to have the same antibacterial and antitumor properties as widely reported A. annua. It has great potential medicinal value in pharmacological effects such as antiviral and anti-inflammatory [37, 38], therefore, we have reason to believe that Artemisia sp. have an extensively application prospect on medicine and feed additives in Tibet in future.

Conclusions

This study identified and quantified the metabolites from three Artemisia species collected from Tibet using widely targeted metabolomics technology. The types of screened and identified differential metabolites were mainly flavonoids, phenolic acids, artemisinins and coumarins. The antibacterial experiments showed that the three Artemisia species had strong antibacterial activities against B. subtilis, E. coli, S. aureus, P. mirabilis and P. aeruginosa. The antioxidant activity test showed that the three species of Artemisia have strong antioxidant capacity in vitro, these wide ranges of beneficial effects suggest great potential for these components for future therapeutic applications.

Materials and methods

Plant materials

Samples of A. sieversiana, A. wellbyi and A. annua were collected from Jinbei, Caina Township, Qushui County, Lhasa City, Tibet Autonomous Region in July 2020 (east longitude 90°53′ 58.60", north latitude 29°26′ 6.03", elevation 3581 m). Official permits for collection of these native plants were not required because these plants are not included in the list of national key protected plants, however permission for collections was obtained from the Lhasa Forestry and Grassland Administration. The formal identification of the plant material was performed by Professor Zhaoyang Chang of College of Life Science, Northwest A&F University based on morphological characters. The specimens of A. sieversiana, A. wellbyi, and A. annua have been deposited at Herbarium, Institute of Botany, Chinese Academy of Sciences (voucher # PE01890226, PE01890481, PE01997408, respectively). Sample collection of Plants were from each 10 m × 10 m sampling site; 3 plants were collected diagonally with a total of 9 plants/site. All samples were dried, crushed, passed through a 40-mesh sieve (with an aperture of 0.425 mm), put into a paper bag, and stored in a desiccator at room temperature for later use. One g each of 9 samples were wrapped in tin foil, snap frozen in liquid nitrogen for storage, transported in dry ice to Biomarker Technology Co., Ltd. for analysis.

Chemical reagents and instruments

Methanol (Merck, Germany), acetonitrile (Merck, Germany), formic acid (Merck, Germany), pipette (Thermo company, USA), freeze dryer (Scientz company, Germany), grinder (Retsch company, Germany), UPLC (SHIMADZU, Japan), Tandem mass spectrometry (ABI, USA), column (Agilent, Germany), industrial alcohol, petroleum ether, ethyl acetate, dichloromethane, n-butanol, dimethyl sulfoxide, peptone, beef extract, agar powder, sodium chloride, etc. Pressure steam sterilizer (Shanghai Boxun), electronic balance PTX-FA110 (Sartorius, Germany), ultra-clean workbench (Suzhou purification), refrigerator (FRESTECH SC-208A), water bath, microwave oven (Foshan Midea), rotary steamer.

Widely targeted metabolomics experiment

Metabolite extraction

The main processing steps are as follows: the biological samples are freeze-dried in vacuum (Scientz-100F), and the dried samples are ground in a grinder (MM 400, Retsch) at 30 Hz for 1.5 min to powder. Dissolve 100 mg of powder sample in 1.2 mL of 70% methanol, mix it with vortex every 30 min for 30 s each time for 6 times, and store it at 4 °C overnight. The sample was centrifuged at 12,000 rpm for 10 min. Suck the supernatant through the hole diameter of 0.22 μ M and stored for UPLC-MS/MS analysis.

The LC/MS system for metabolomics analysis is composed of Waters Acquity I-Class PLUS ultra-high performance liquid tandem Waters Xevo G2-XS QT of high resolution mass spectrometer. The column used is purchased from Waters Acquity UPLC HSS T3 column (1.8 μm 2.1*100 mm). Positive ion mode: mobile phase A: 0.1% formic acid aqueous solution; mobile phase B: 0.1% formic acid acetonitrile. Negative ion mode: mobile phase A: 0.1% formic acid aqueous solution; mobile phase B: 0.1% formic acid acetonitrile. Injection volume 1μL.

Waters Xevo G2-XS QTOF high resolution mass spectrometer can collect primary and secondary mass spectrometry data in MSe mode under the control of the acquisition software (MassLynx V4.2, Waters). In each data acquisition cycle, dual-channel data acquisition can be performed on both low collision energy and high collision energy at the same time. The low collision energy is 2 V, the high collision energy range is 10-40 V, and the scanning frequency is 0.2 s for a mass spectrum. The parameters of the ESI ion source are as follows: Capillary voltage: 2000 V (positive ion mode) or -1500 V (negative ion mode); cone voltage: 30 V; ion source temperature: 150 °C; desolvent gas temperature 500 °C; backflush gas flow rate: 50L/h; Desolventizing gas flow rate: 800L/h.

Antibacterial assays

Extraction of active components from plants

The dried plant samples (100 ɡ) were ground, soaked in industrial alcohol, extracted under reflux at 60 °C for 3 h, concentrated by rotary evaporation under reduced pressure in water bath held at 55 °C. The total ethanol extracts were obtained by heating and drying for 12 h. Thirty g of the ethanol extract was dissolved in 800 mL water, and was extracted in 1-L flask with petroleum ether, chloroform, ethyl acetate and n-butanol. The extracts were condensed under vacuum and dried at 55 °C. The extracts of each extraction were dissolved in dimethyl sulfoxide separately, prepared into a solution in a concentration of 200 mg/mL, filtered through a 0.22 μm filter, and stored at 4 °C for later use.

Medium preparation

Nutrient agar medium (1000 ml containing 3.0 g beef extract, 10.0 ɡ peptone, NaCl 5.0 ɡ and 20 ɡ agar (pH 7.2–7.4).

Bacterial liquid preparation

E. coli, Salmonella (G−), P. mirabilis (G−), B. cereus (G+), S. aureus (G+), Streptococcus (G+), P. aeruginosa, (G−). The above strains were provided by the Microbiology Laboratory of Northwest A&F University. Under aseptic conditions, single colonies of activated E. coli, Salmonella, Streptococcus, S. aureus, P. mirabilis, B. cereus, and P. aeruginosa after activation were picked with an inoculation loop and inoculated into the liquid medium, respectively, at a constant temperature of 37 °C. Cultivated for 24 h. Use sterilized liquid medium to adjust the concentration of each bacterial solution equivalent to 0.5 McFarland turbidity standard (about 1.5 × 108 CFU/mL) for use.

Agar well diffusion assay

A sterilized filter paper sheet, 6 mm dia. was soaked in a drug extract for 1 h. The bacterial suspension to be tested (0.5 mL) was spread evenly onto the surface of the solid medium. The filter paper of the extract was layered onto the bacteria-containing plate under sterile conditions; a sterile filter paper soaked in DMSO was used as the negative control, and the ceftazidime drug sensitive tablet was used as the positive control. The leaching solution treatment and control were incubated (37 °C, 12 h); inhibition zone of the filter paper was measured. The experiment was repeated 3 times. If the diameter of the inhibition zone is greater than or equal to 18 mm, the bacteria is rated as highly sensitive, 12 to 18 mm as moderately sensitive, 7 to 12 mm as low sensitivity, and less than 7 mm as insensitive.

Determination of antioxidant activity

FRAP assay

Sample (1 ml) was mixed with 2.5 mL of phosphate buffer (0.2 moL/L, pH 6.6) and 2.5 mL of 1% K3Fe(CN)6, and heated in a water bath (50 °C, 20 min). After incubation, 2.5 mL of 10% (w/v) trichloroacetic acid was added and the samples were centrifuged (15,000 × ɡ, 10 min) and 2.5 mL of the supernatant was aspirated and mixed with 2.5 mL of H2O and 0.5 mL of 0.1% FeCl3. The absorbance was measured with a spectrophotometer (700 nm).

Scavenging capacity of DPPH free radicals

To 2.0 mL of each solution to be tested, 2 mL 0.04 mg/mL of DPPH solution (with ethanol as a solvent) is added, mixed.

Scavenging activity against ABTS free radical

The ABTS working solution was carefully transferred into the first reagent tube, the tube cover was rotated, shaken well, and placed at temperature for 14–16 h. Take 10 μL ABTS and dilute the diluent (first 20 diluted in pure water) and record the dilution ratio. Fully mix, microabsorption values were measured at 734 nm.

Determination of hydroxyl radical scavenging capacity

Two mL of the samples to be tested was mixed with 1.4 mL 6 mmol/L H2O2, then 0.6 mL 20 mmol/L sodium salicylate and 2 mL 1.5 mmol/L ferrous sulfate added. Samples were thoroughly mixed; for the blank group, the sample was replaced with ultrapure water. The absorbance value of H2O2 was measured in the same way and recorded as A background. OH free radicals. The formula for calculating the clearance rate = [A blank—(A sample—A background)] / A blank × 100%.

Data analysis

After identification of distinct compounds and pathway analysis using KEGG database, t-test was used to calculate the difference significance p-value of each compound. R package was used for the OPLS-DA modeling was performed using R package and the reliability of the model was tested with 200 times permutation.