Allelopathic effects and composition of aqueous extracts from different parts of Galinsoga parviflora Cav. on Medicago sativa L. and Avena sativa L

Preparation of aqueous extracts from roots, stems, leaves and whole plants of it G. parviflora

In 2022, seeds of G. parviflora were collected from M. sativa fields in Qujing City of China (103.21°E, 25.18°N), which were germinated in petri dishes at room temperature for 72 h after rinsing with water. Twelve germinated seeds were planted in a pot with a diameter of 30 cm containing moist loam soil, and kept in artificial climate chambers at 25/20 °C (light/dark) for a 16-h photoperiod with a photosynthetic photon flux density of approximately 270 µmol m⁻² s⁻¹. During the cultivation process, the soil was watered moderately every five days to ensure moisture.

Approximately 45 d after planting, nearly all G. parviflora were in flowering phase, which for most species was the period of maximum allelopathic activity (Singh, Singh & Singh, 2015). Meanwhile, the fresh roots, stems, leaves and whole plants of G. parviflora were washed, dried and stored at 4 °C after crushing to a particle size of less than 0.5 mm, respectively. Accurate 50 g of the above sample from different parts was added into 500 mL distilled water and then was oscillated for 48 h at the speed of 150 rpm min⁻¹ at 4 °C. After 15 min of ultrasound, the mixture was repeatedly filtrated by vacuum with 0.45-µm membrane to obtain an extraction solution with a final concentration of 0.100 g mL⁻¹. Subsequently, the extracts from different parts of G. parviflora were diluted with distilled water to solutions with the concentration of 0.025, 0.050, 0.075 and 0.100 g mL⁻¹ (Dai et al., 2022). The range of concentration was based on a preliminary study, which showed that extracts exhibited a promoting or slight inhibitory effect on recipient plants at the lowest concentration (0.025 g mL⁻¹).

Seed germination test

The varieties of M. sativa and A. sativa tested were Zhongmu No.3 and Qinghai No.444, respectively, obtained by Beijing Best Grass Industry Co., Ltd in 2023. The seeds were washed for three times, then sterilized with 75% alcohol for one minute, and cleaned with sterile water for five times to remove the alcohol. A total of 50 disinfected seeds of M. sativa and A. sativa were placed in a square plastic culture dish with a size of 12 × 12 cm (Nantong Mingcheng Experimental Instrument Co., Ltd). Subsequently, eight mL of different concentrations of extracts from the roots, stems, leaves and whole plants of G. parviflora were added. The sealed culture dish was placed under the condition of 25 °C temperature, 15000 lx light intensity, and 12 h/d light duration for cultivation.

During the cultivation, the number of germinated seeds was counted every day, and four mL of corresponding solutions were periodically supplemented to each culture dish every other day. The control group was treated with equal volume of sterile water. The experiments were conducted with four replicates per dose. The seed that produced at least one mm radicle was considered as germinated (ISTA, 2020; Pamplona et al., 2020). The last count was conducted when no new germinated for three consecutive days. The germination rate (GR) and germination potential (GP) of M. sativa or A. sativa were calculated according to the following formulas (Sun et al., 2021).

GR = Number of germinated seeds within 7 days/total number of test seeds

GP = Number of germinated seeds within 3 days/total number of test seeds.

Seedling growth test

Ten newly sprouted seeds were put in a culture dish containing above eight mL of extracts, and the method of cultivation was same as mentioned above. Similarly, four mL of the corresponding solution was added to each culture dish every other day, and 48 mL of solution was needed for each treatment during the cultivation. After 21 days of cultivation, ten seedlings were randomly selected from each dish, gently wiped the water on the surface, and their height and fresh weight were measured (García-Locascio, Valenzuela & Cervantes-Avilés, 2024). The experiments were conducted with four replicates per dose.

Determination of chlorophyll content

The leaves in the section of seedling growth test were removed and washed with distilled water to determine the content of chlorophyll. Accurately 200 mg leave tissues in different treatment with 10 mL of 95% ethanol was placed in a dark room for 48 h, and was shaken for three times during the extraction. The absorbance of extracted solution was measured at 665 nm and 649 nm with 95% ethanol as the blank. The content of chlorophyll was calculated based on the following formula (Dai et al., 2022).

Chlorophyll a (mg L⁻¹) = 13.95 A665 nm − 6.88 A649 nm

Chlorophyll b (mg L⁻¹) = 24.96 A649 nm − 7.32 A665 nm

Chlorophyll content (mg g⁻¹) = (Concentration of chlorophyll × total volume dilution of the extract)/fresh weight of the samples.

Analysis of synthetical allelopathic effects

The response index (RI) of extracts to recipient plants was calculated according to the following formula. In this equal, C and T represented the values of control group and treatment group respectively. If the RI value was positive, the effect of extracts was promoted. On the contrary, it was inhibited. The absolute value of RI reflected the magnitude of the allelopathy. The synthetical allelopathic effects (SAE) represented the arithmetic mean of the RI of five measurement index of M. sativa or A. sativa (Dai et al., 2022), including GR, GP, the height and fresh weight of seedings and the content of chlorophyll.

RI = 1−C/T (T≥C)

RI = T/C−1 (T < C).

Identification of allelopathic substances of G. parviflora

In order to further elucidate the mechanism of allelopathic effect, the components of the extracts from different parts of G. parviflora were identified. Accurately 100 mg of freshly thawed powders was extracted with one mL pre-cooling extraction solution (water-acetonitrile-isopropyl alcohol mixed solution 1:1:1, v/v/v). After 30 min ultrasound at 4 °C, the mixture was centrifuged at 12,000 rpm for 10 min (X3R, Thermo Fisher Scientific, USA), set still for 1 h to precipitate protein at −20 °C, and then centrifuged 10 min with 12,000 rpm at 4 °C. Subsequently, supernatant was transferred to a centrifuge tube and dried in vacuum. The dried supernatant was redissolved with 200 µL of 30% acetonitrile (Sinopharm Chemical Reagent Co., Ltd, EC number: 200-835-2) solution, centrifuged 15 min with 14,000 rpm at 4 ° C. Finally, the supernatant was filtered through and analyzed by ultra high performance liquid chromatography (UHPLC, Vanquish; Thermo Fisher Scientific, Waltham, MA, USA) and high resolution mass spectrum (HRMS, Q Exactive HFX; Thermo Fisher Scientific) (Ghirardo et al., 2020). The composition analysis was conducted by UHPLC with a Waters HSS T3 column (100 × 2.1 mm, 1.8 µm) with a temperature of 40 °C. Mobile phase A was Mili-Q water (0.1% formic acid) (Sinopharm Chemical Reagent Co., Ltd, EC number: 200-579-1), and phase B was acetonitrile (0.1% formic acid). The original proportion of phase A/phase B was 100:0 (v/v) in the first 12 min, changed to 5:95 (v/v) within 12-13 min, and returned to original 100:0 (v/v) from 13.1 min to the end (17 min). Data of HRMS were recorded by the Full-ms-ddMS2 acquisition methods. The ESI source parameters were set as follows: sheath gas pressure, 40 arb; aux gas pressure, 10 arb; spray voltage, +3000 v/-2800 v; temperature, 350 °C; and ion transport tube temperature, 320 °C. The scanning range of primary MS was (scan m/z range) 70–1050 Da, with a primary resolution of 70,000 and secondary resolution of 17,500.

Determination of herbicidal activity of key allelopathic substances

According to the identification of extracts from G. parviflora, the content of 1,4-cyclohexanedicarboxylic acid (CHDA) and trehalose ranked the top two. At the same time, they were also the highest in all phenolic acids or polysaccharides, and were selected to explore the herbicidal activity of key allelopathic substances. Technical-grade CHDA (EC number: 214-068-6) and trehalose (EC number: 202-739-6) were purchased in Shanghai Yuanye Bio-Technology Co., Ltd. The stock solution with a concentration of 1.0 g mL⁻¹ was prepared by dissolving the above substance in distilled water and stored at 4 °C for later use. Subsequently, distilled water was added to dilute the CHDA solution to seven concentrations, namely 1.0, 0.1, 1.0 × 10⁻², 5.0 × 10⁻³, 1.0 × 10⁻³, 1.0 × 10⁻⁴ and 1.0 × 10⁻⁵ g mL⁻¹. Due to the lower phytotoxicity of trehalose, only five solutions with higher concentrations mentioned above were prepared.

The seeds of D. sanguinalis and A. retroflexus to determine herbicidal activity were collected from M. sativa fields in Qingdao City, China (120.08°E, 36.45°N) in September 2022. To break dormancy of the two weeds, the harvested seeds were soaked in 98% concentrated sulfuric acid (Sinopharm Chemical Reagent Co., Ltd, EC number: 231-639-5) for two minutes, and cleaned with sterile water for three times before use. Meanwhile, the above-mentioned M. sativa and A. sativa were also selected to evaluate safety of the two allelopathic substances. The method for seed germination testing was consistent with that mentioned earlier.

Statistical analysis

Data was submitted to Shapiro–Wilk’s normality test. When the assumptions of normality were met, the Dunnett’s test (P < 0.05) at the 5% level of significance in one-way analysis of variance (ANOVA) was carried out to compare the RI values under five concentrations (C0-C4) and four parts (P1-P4) in the same indicator. Meanwhile, the two-way ANOVA was also conducted on the RI values of each indicator to analyze effects of part (P), concentration (C), and their interactions (P × C). In terms of herbicidal activity, the differences between different concentrations of CHDA or trehalose solutions were compared by one-way ANOVA. Statistical analysis was performed using SPSS software (v. 20.0, IBM, USA), and the data was expressed as mean ± standard deviation (SD).

Effects of extracts from different parts of G. parviflora on the germination of M. sativa and A. sativa

Both GR and GP were used to evaluate the allelopathic effects of the extracts on the germination of M. sativa and A. sativa (Fig. 1). The extracts of G. parviflora displayed varied degrees of inhibitory effects, and the concentrations of extracts and parts of G. parviflora exhibited significant effects on the GR of M. sativa and A. sativa (P < 0.001, Figs. 1A, 1B). The allelopathic effects of extracts gradually strengthened with the increase of concentration. When the concentration was 0.075 g mL⁻¹, the GR of M. sativa significantly decreased under the extracts from various parts of G. parviflora (Fig. 1A). The inhibitory effect was the strongest at the concentration of 0.100 g mL⁻¹, and the GR of M. sativa was only 48% treated by extracts from the whole plants (Fig. 1A). Compared with M. sativa, A. sativa showed a higher tolerance to extracts of G. parviflora, the higher concentration (0.100 g mL⁻¹) of corresponding extracts were required to significantly inhibit its germination (Fig. 1B). In addition, the allelopathic effects of extracts from different parts of G. parviflora with the same concentration were different. There was no significant difference in the effect of extracts from different parts of G. parviflora on the GR of M. sativa and A. sativa at the lowest concentration (0.025 g mL⁻¹), and the differences gradually emerged accompanied by the increase in concentration. Overall, the allelopathic ability from weak to strong was roots < stems < leaves < whole plants, and the corresponding RI values at the highest concentration were −0.07, −0.14, −0.39, and −0.49, respectively (Fig. 1A). Nevertheless, there was no significant difference in allelopathic effects of extracts from the leaves and other two parts on A. sativa (Fig. 1B).

The concentrations of extracts, parts of G. parviflora and their interactions also exhibited significant effects on GP of M. sativa (P < 0.001, Fig. 1C) and A. sativa (P < 0.01, Fig. 1D), which was highly correlated with GR, with a Pearson correlation coefficient of 0.962. Similarly, the allelopathic effect of extracts on GP gradually increased with increasing concentration, and all different parts of the extracts showed significant difference on the GP of M. sativa at the concentration of 0.100 g mL⁻¹ (Fig. 1C). The influence of different parts of extracts from G. parviflora on the GP of recipient plants was also different. The inhibitory effect of extracts from the whole plants of G. parviflora was the strongest, the RI values of M. sativa and A. sativa at the highest concentration were −0.71 and −0.56, respectively (Figs. 1C, 1D). On the contrary, the inhibitory effect of root extracts was the weakest, and no significant inhibitory effect on the GP of recipient plant seeds at various concentrations was observed (Fig. 1).

Effects of extracts from different parts of G. parviflora on the seedling growth of M. sativa and A. sativa

The concentrations and parts also had significant effects on the seedling growth of M. sativa and A. sativa (P < 0.001, Fig. 2). The extracts of G. parviflora exhibited promoting growth at low concentrations and inhibiting growth at high concentrations on the seedling height of M. sativa (Fig. 2A), which was different from the allelopathic effect on the germination of recipient plants (Fig. 1). Almost all extracts with the concentration of 0.025 g mL⁻¹ promoted the height of M. sativa seedlings. It was not until the concentration reached 0.075 g mL⁻¹ that the extracts from various parts began to show significant inhibitory effects on the height of M. sativa. The inhibitory was most significant at the concentration of 0.100 g mL⁻¹ (Fig. 2A). However, the promotion effect of extracts from various parts on the growth of A. sativa was not significant at low concentration, and the inhibitory effect only became apparent when the concentration of the extracts was 0.075 g mL⁻¹ (Fig. 2B). Similarly, the allelopathic effect of root extracts on the height of M. sativa and A. sativa plants was significantly lower than that from other parts of G. parviflora, and the values of RI at the concentration of 0.100 g mL⁻¹ was −0.08 and −0.40, respectively. In addition, the sensitivity of M. sativa and A. sativa to extracts from different parts of G. parviflora was also varied. For instance, the inhibitory effect of stem extracts on A. sativa height was far stronger than that of leaf extracts at the concentration of 0.100 g mL⁻¹, with corresponding RI values of −0.64 and −0.49, while no significant difference was observed on the height of M. sativa (Figs. 2A, 2B).

In addition to height, fresh weight was also used in this study to evaluate the allelopathic effect of the extracts on the seedling growth of recipient plants (Figs. 2C, 2D). Compared with height, the inhibitory effect of the extracts of G. parviflora on the fresh weight of seedlings was weaker. For example, all concentrations of stem extracts showed no significant inhibitory effect on the fresh weight of M. sativa. The extracts from all parts of G. parviflora at the concentration of 0.025 g mL⁻¹ or 0.050 g mL⁻¹ exhibited a positive effect on the fresh weight of M. sativa, which was similar to the results of plant height (Fig. 2C). Only extracts from the whole plants of G. parviflora exhibited a significant inhibitory effect on the fresh weight of A. sativa at the highest concentration, while the extracts from other parts did not (Fig. 2D). This phenomenon once again indicated the differences in allelopathic effects of different parts from G. parviflora.

Effects of extracts from different parts of G. parviflora on the chlorophyll content of M. sativa and A. sativa

The extracts of G. parviflora exhibited varied degrees of allelopathic effects on the chlorophyll content of M. sativa, and the allelopathic effects was also affected by the concentrations (C) of extracts, and parts (P) of G. parviflora (P < 0.001, Fig. 3A). The RI values of roots, stems, leaves, and whole plants extracts at the concentration of 0.100 g mL⁻¹ on the chlorophyll content were −0.18, −0.47, −0.41 and −0.51 (Fig. 3A), respectively, which was basically consistent with the results of germination and seedling growth mentioned earlier (Figs. 1–2). However, the inhibitory effect of extracts on the chlorophyll content of A. sativa was relatively weaker, and there was no significant difference between the extracts of different concentrations (Fig. 3B). This might be because inhibiting the chlorophyll synthesis in A. sativa required a concentration of G. parviflora extracts exceeding 0.100 g mL⁻¹.

Synthetical allelopathic effects of extracts from different parts of G. parviflora on M. sativa and A. sativa

The SAE values of extracts from different parts of G. parviflora were calculated based on the above five indicators and shown in Fig. 4. The concentrations of extracts, parts of G. parviflora and their interactions significantly affected the SAE of M. sativa and A. sativa (P < 0.001, Fig. 4). All extracts with concentrations exceeding 0.050 g mL⁻¹ exhibited inhibitory effects on both recipient plants (SAE<0), and the allelopathic effect gradually increased with increasing concentrations. The allelopathic effects of the extracts from different parts of G. parviflora varied greatly. The effect of roots was the weakest, while the whole plants was the strongest, and the SAE values at the concentration of 0.100 g mL⁻¹ to M. sativa was −0.12 and −0.40, respectively (Fig. 4A). The allelopathic effects of the extracts from stems and leaves were moderate, and their intensity on the two types of forage was opposite. According to the values of SAE at the low concentration, M. sativa was generally more sensitive to extracts than A. sativa (Fig. 4).

Identification of allelopathic substances of G. parviflora

To analyze the mechanism of the allelopathic effects of extracts from different parts of G. parviflora on recipient plants, untargeted metabolomics detection was conducted. The number of chemical components detected in the extracts of the roots, stems, leaves, and whole plants of G. parviflora is 305, 310, 308, and 313, respectively, including acids, sugars, glycosides, terpenoids, etc. The composition and content in the extracts from different parts varied wildly (Fig. 5), and detailed information, such as the chemical name, type, formula, CAS number, fragmentation score, retention time (t_R) of the extracts from the whole plants were listed in Table 1. Obviously, the component with the highest content in the extracts from the whole plants, stems and leaves was CHDA, with relative contents of 29.65%, 34.69% and 37.39%, respectively. However, this component was rarely found in root extracts, and the highest content was occupied by 18.49% trehalose. The trehalose content detected in extracts from the whole plants and stems was only second to CHDA, with the values of 8.76% and 6.36%, respectively. The content of trehalose in leaf extracts was also 4.89%, which indicated that G. parviflora was a plant rich in trehalose. In addition, components such as gossypin, succinic acid and 9S,10S,11R-trihydroxy-12Z-octadecenoic acid were commonly present in the extracts from various parts of G. parviflora (Fig. 5).

Determination of herbicidal activity of key allelopathic substances

The effects of CHDA and trehalose on the GR of D. sanguinalis, A. retroflexus, M. sativa and A. sativa were shown in Table 2. The CHDA significantly inhibited the germination of D. sanguinalis starting from the concentration of 1.0  × 10⁻⁴ g mL⁻¹ with a GR of 60%. And CHDA showed a hormesis effect on the germination of A. retroflexus, with a GR increased by 24% at the concentration of 1.0  × 10⁻³ g mL⁻¹ and almost no germination at the concentration of 5.0  × 10⁻³ g mL⁻¹. In terms of M. sativa and A. sativa, CHDA had no significant effect on their GRs at low concentrations, while reduced GRs to 49% and 59% at the highest concentration (1.0 g mL⁻¹), respectively. Meanwhile, trehalose does not affect the germination of the four tested plants (Table 2).