Rhus typhina L., an exotic tree species indigenous to North America, was introduced to China as a horticultural and greening ornamental plant in 1959 (Hu et al. 2022). Due to its high ecological risk and potent reproductive capacity, it is now considered as one of the most destructive invasive tree species in China (Hu et al. 2022; Xu et al. 2023). To construct a native plant community, we selected four plant species from the warm temperate zone of China that commonly co-occur with R. typhina, including one herbaceous species (Chenopodium album L.), one shrub (Vitex negundo L.), and two trees (Rhus chinensis Mill. and Acer truncatum Bunge). Given the challenges of cultivating herbaceous plants with consistent growth and development, we dug up C. album seedlings with similar growth time directly after the other three plants (V. negundo, R. chinensis, and A. truncatum) had completed their seedling cultivation. Seeds of V. negundo, R. chinensis, and A. truncatum were collected from Mashan, Jimo, Qingdao, Shandong, China in the winter of 2020. We classified each species as invasive or native according to Alien Invasive Flora of China (Ma 2020) and the database Flora of China (www.efloras.org).

In mid-April 2021, we prepared sufficient quantities of seeds of the three selected plant species. The following treatments were performed to improve their germination rate. First, seeds of R. chinensis were treated with a baking soda solution at 70 °C for 10 minutes to remove their waxy skins and subsequently cleaning with sterile deionized water. Second, seeds of A. truncatum were subjected to a 4 °C treatment for 24 hours. Third, seeds of both R. chinensis and A. truncatum, as well as those of V. negundo, were soaked in fresh water for 24 hours, replacing the water every 12 hours. Soaked seeds were then evenly spread into trays with three layers of gauze on the bottom, and two layers of gauze were placed over the seed surface to improve germination. During this process, sterile deionized water was sprayed into the trays daily to maintain moisture levels. After approximately 30% of the seeds had germinated, we selected strong and uniform seedlings of each species and transferred them to cell trays for further culture. Finally, seedlings of C. album with 4–5 leaves were collected from the campus of Qingdao Agricultural University, Shandong, China, until three plant seedlings had developed 4–5 leaves.

There are two primary rationales for using the root aqueous extracts from R. typhina to study the allelopathic effects: (1) allelopathic effects of R. typhina are mainly mediated by phenolic substances, most of which, especially polyphenols, can be dissolved in water (Djurdjević et al. 2012; Xu et al. 2023), and (2) in nature, the root exudation pathway is one of the primary methods by which plants release allelochemicals (Hassan and Mohamed 2020; Inderjit et al. 2021). Fully mature R. typhina roots were randomly selected from Qingdao Agricultural University, Shandong, China, and gently washed before being cut into small fragments. We added 1000 ml of sterilized deionized water per 100 g of fresh roots at approximately 25 °C for 48 hours, stirring every six hours. The root residues were filtered out to obtain 100 g L^{- 1} mother liquor of the R. typhina root aqueous extract. Sterilized deionized water was then added in various ratios to the mother liquor to achieve different concentrations of the root aqueous extract (10, 20, 30, and 40 g L^{- 1}). The root extract concentration of 20 g L^{- 1} represented the normal condition with R. typhina’s allelopathic effects (hereafter referred to as NRE) (Xu et al. 2023). Since the concentration of allelochemicals in the soil during the initial period of invasion was low (Zhang et al. 2020), 10 g L^{- 1} was designated as the low concentration of the root extract treatment (hereafter referred to as LRE). During the establishment of R. typhina, allelochemicals in the soil gradually accumulated (Zhang et al. 2020), so 30 and 40 g L^{- 1} were classified as medium and high concentration of the root extract treatments (hereafter referred to as MRE and HRE, respectively). The R. typhina root aqueous extracts were refrigerated at approximately 4 °C (for no more than one week), while sterilized deionized water served as the control (0 g L^{- 1}; simulating conditions without allelopathic effects; hereafter referred to as CK).

The AMF inoculum comprised a blend of three common AMF strains – Funneliformis mosseae, Glomus versiforme, and Rhizophagus intraradices – in a 1:1:1 volume ratio. As different AMF types have varying colonization rates in diverse plant species, the AMF mixtures can ensure consistent colonization rates across diverse plant species. In the AMF treatments, each pot was inoculated with the AMF mixtures containing approximately 8 × 10³ viable propagules. In the NoAMF treatments, an equivalent amount of sterilized AMF inoculum was added to prevent substrate effects.

The experiment was performed at Qingdao Agricultural University, situated in Qingdao, China (36°31'N, 120°39'E). The region has a temperate monsoon climate with an annual temperature of approximately 12.7 °C, mean annual precipitation of around 821.8 mm, and average yearly sunshine duration of about 2,541.1 h. In the summer of 2021, we conducted a 70-day experiment in an artificially-controlled greenhouse maintained at an average temperature of 29.4 °C and a relative humidity of 77%, using a digital display humidity thermometer, DL-WS20 (Hangzhou Gsome Technology Co., China). The plastic pots used in this study were 20.0 cm in height and 20.0 cm in diameter, and were filled with soil materials consisting of autoclaved field soil and grass ash mixed at a 1:1 volume ratio.

In each plastic pot, we planted four different species of seedlings (C. album, V. negundo, R. chinensis, and A. truncatum), with four to five leaves per seedling. One seedling for each plant species was transplanted into one of the four equal sections of the cross lines past the center of the circle within each pot, resulting in a total of four plants per pot. The control treatment consisted of no AMF and root aqueous extracts of R. typhina. Overall, we established 100 experimental pots, including two AMF treatments (with and without AMF) and five concentrations (0 g L^{– 1}, CK; 10 g L^{– 1}, LRE; 20 g L^{– 1}, NRE; 30 g L^{– 1}, MRE; and 40 g L^{– 1}, HRE). Each treatment had ten replicates (n = 10), and all individuals of each species and treatments were randomly distributed. Over the entire experiment, we added 50 ml of the appropriate concentration solution of the root aqueous extracts from R. typhina once every two days in the five concentration treatments, for a total of thirty times. Weeding and pest control measures were implemented as usual and the greenhouse was adequately ventilated. All pots were placed randomly within the greenhouse and rotated normally to minimize any potential effects of environmental differences during the experiment.

Physiological parameters measured in the native plant community included specific leaf area (SLA), total chlorophyll concentration (Chl), F_v/F_m, and leaf nitrogen (LN), which can characterize the degree of leaf photosynthetic capacity (Wang et al. 2020). Growth parameters included plant height, crown area (CA), relative growth rate (RGR), and total aboveground biomass (TAB). Plant height and CA characterize the ability of the native plant community to acquire light (Wang et al. 2020; Wang et al. 2022b). RGR describes the rate of plant growth. At the conclusion of the experiment, the intertwining of root systems in each plastic pot prevented the measurement of below-ground biomass for each plant. Consequently, we measured the total aboveground biomass (TAB) of all four plants to represent the community’s growth. Additionally, the AMF colonization rate was used to characterize the quantity of mycorrhizal symbiont formation under different conditions.

Prior to the treatment, the aboveground component of five seedlings from each indigenous species with 4–5 true leaves were oven-dried at 105 °C for 0.5 h to inactivate them. The drying process was continued for an additional 48 h at 80 °C, and then the weight was measured. The mean value was regarded as the initial biomass for each species.

After 60 days, when the herbaceous flora had reached the reproductive stage, the maximum growth index of all native plants was determined. The vertical height of the plant from the base to the terminal bud was gauged as the plant height. The CA was calculated using the diamond-shaped area formula (Wang et al. 2022b; Guo et al. 2023b).

CA = 0.5 × a × b

where a indicates the length of the plant’s maximum horizontal extension, and b signifies the length perpendicular to a on the plant’s maximum horizontal extension plane.

For each plant, chlorophyll fluorescence parameters were evaluated during sunny and cloudless weather conditions from 8:30 to 11:30 using Pocket PEA (Hansatech Instruments Ltd., UK). The top-to-bottom first fully developed leaf was selected, and after undergoing dark treatment for 0.5 h, the maximum fluorescence value (F_m) and variable fluorescence value (F_v) were evaluated. The maximum photochemical efficiency (F_v/F_m) of PSII was then calculated (Hu et al. 2022; Xing et al. 2022).

The second or third fully expanded and healthy leaf from the apical meristem of each plant was selected, cleaned with sterilized deionized water, and the leaf area was measured using a portable leaf area meter (Yaxin-1241, Yaxin Inc., Beijing, China). The leaves were inactivated to inactivation at 105 °C for 0.5 h and dried at 85 °C for 24 h in a drying oven until they reached constant weight to obtain their dry weights. The SLA was calculated using the formula (Wang et al. 2022b; Guo et al. 2023a).

SLA = leaf area ÷ dry weight

The total chlorophyll content was determined using the ethanol extraction method. Four fully mature leaves near the top of each plant (one leaf per plant) were collected for each treatment and chopped to approximately 0.2 g fresh weight. The leaves were then submerged in 10 ml of 95% alcohol until completely whitened. The absorbance of chlorophyll in the solution at 665 nm and 649 nm wavelengths was measured using a UH5300 UV/VIS spectrophotometer (Hitachi, Inc., Tokyo, Japan). The concentrations of chlorophyll a, b, and total chlorophyll were calculated using the following formulas (Wang et al. 2022b; Guo et al. 2023a):

Chlorophyll a concentration = 13.95 × A₆₆₅ – 6.88 × A₆₄₉

Chlorophyll b concentration = 24.96 × A₆₄₉ – 7.32 × A₆₆₅

Chlorophyll content

= Chlorophyll concentration

× volume of extraction solution / fresh weight of sample

Chl = Chlorophyll a content + Chlorophyll b content

The aboveground parts of the plants were harvested, washed, classified according to species, inactivated at 105 °C for 0.5 h in the oven, and dried in the oven at 80 °C for 48 h. The total aboveground biomass of all species in the same pot was measured and the RGR of native plants was calculated using the formula (Wang et al. 2022b).

RGR = (ln X₁ – ln X₂) ÷ ∆T

where X₂ and X₁ denote the final and initial biomass, respectively, and ∆T is the duration of the experiment.

For LN measurements, dried leaves of each species were ground and weighed to 0.5 g using the Kai-style digestion method (K9860, Hanon, Shandong, China).

Roots were collected from 70-day plants. Fine roots (15 g) were excised from each plant, washed, and stained with acid fuchsin to determine the percentage of mycorrhizal colonization by the method of Biermann and Linderman using a BX50 Olympus microscope (Olympus Optical Co., Ltd., Tokyo, Japan) (Biermann and Linderman 1981).

The Community-Weighted Trait (CWT) method was utilized to evaluate the relevant physiological parameters (leaf nitrogen, total chlorophyll concentration, F_v/F_m, and specific leaf area) and growth parameters (height, crown area, and relative growth rate) of the native plant community in this study (Wang et al. 2020). CWT was calculated according to the following formula:

$$CWT=\sum _{i=1}^S{P}_i\times X_i$$

where S is the total number of species and P_i represents the relative abundance of species i in the native community. P_i = n_i ÷ N, n_i and N represent the plant number of species i and the number of plants of all species in the native plant community, separately. X_i denotes the average value of morphological and physiological characteristics of species i.

To assess differences in growth and physiological traits among different treatments, we conducted a two-way analysis of variance (ANOVA). Prior to the ANOVA, we performed variance homogeneity tests for each group of data and transformed unevenly distributed data. We used Duncan’s test with a significance level of α ≤ 0.05. For each concentration of allelochemicals, we performed linear regressions to determine the relationship between AMF colonization rates and the concentration of root aqueous extracts from R. typhina. To evaluate the correlation level of the main traits for the native plant community, we generated Pearson product-moment correlation coefficients among the physiological and growth parameters. We used the IBM SPSS Statistics 25.0 software package (IBM Corporation, Armonk, NY, USA) to conduct all analyses. We created all figures using Origin 2021 (OriginLab Co., Northampton, MA, USA).

To gain an understanding of how growth and physiological parameters of the native plant community were influenced by allelopathy and AMF, and how these parameters were moderated, structural equation modeling (SEM) was utilized. A conceptual model, based on theoretical interactions among variables, was established to serve as our formal hypothesis. To test our hypotheses statistically, the variance-covariance matrix implied by the conceptual model was compared with the observed variance-covariance matrix, and maximum likelihood estimation was employed to estimate model coefficients. Model fit was considered acceptable if the χ² test was non-significant, the root mean square error of approximation (RMSEA) was low (RMSEA < 0.01), and the comparative fit index (CFI) was high (CFI > 0.90). SEM was conducted using the lavaan and semPlot package in R 4.2.3 (R Core Team 2023).

Physiological parameters such as LN and Chl were significantly affected by allelopathy (p < 0.001, Table 1). In detail, LN values were improved by the NRE and MRE treatments, increasing by 32% and 28% respectively compared to the control (Table 1, Fig. 2A), whereas Chl was higher in the NRE treatment than the LRE, MRE, and HRE treatments (p = 0.037, Table 1, Fig. 2B), which resulted in a 34%, 46%, and 57% advance respectively. Furthermore, the allelopathic effects had negligible effect on F_v/F_m and SLA (Table 1, Fig. 2C, D).

The AMF inoculation significantly affected only LN, which was depressed within the LRE, NRE, MRE, and HRE treatments (p < 0.001, Table 1, Fig. 1A), leading to a 15%, 35%, 28%, and 29% reduction compared to the corresponding NoAMF treatments respectively. According to the results of the two-way ANOVA, inoculating AMF did not significantly affect Chl (Table 1), while multiple comparisons showed that Chl was reduced by 31% in the NRE treatment compared to the corresponding NoAMF treatments (Fig. 2B). Additionally, all other physiological parameters, including SLA and F_v/F_m, were not impacted by AMF (Table 1, Fig. 2C, D).

Figure 1.

Responses of AMF colonization rate to the different concentrations of allelochemicals (n = 5). Five different concentrations of Rhus typhina L. root aqueous extracts were used, which were 0, 10, 20, 30, and 40 g L^{- 1}, respectively. The proportion of variance is explained by R². The regression line was significant at the p < 0.05 level.

Figure 2.

Responses of physiological parameters of native plant community to the allelopathy and its interactions with AMF (n = 10). The boxes represent the range of maximum and minimum values, with a bold horizontal line indicating the average value. Colors indicate different AMF treatments (orange, AMF; purple, NoAMF). The physiological parameters include leaf nitrogen (LN) (A) total chlorophyll concentration (Chl) (B) F_v/F_m (C) and specific leaf area (SLA) (D). There were five different concentrations of the root aqueous extracts of Rhus typhina including 0 g L^{− 1} (CK), 10 g L^{− 1} (LRE), 20 g L^{− 1} (NRE), 30 g L^{− 1} (MRE), and 40 g L^{– 1} (HRE). The different letters indicate a significant difference (p ≤ 0.05) with Duncan’s test.

Most physiological trait values were not influenced by the interaction between allelopathic effects and AMF treatments, such as Chl, F_v/F_m, and SLA (Table 1, Fig. 2B–D). Nevertheless, the interaction between allelopathic effects and AMF had a significant effect on LN of the native plant community (Table 1). The LRE, NRE, and HRE treatments decreased LN by 13%, 14%, and 30% respectively, compared to the control (p < 0.001, Fig. 2A).

The two-way ANOVA indicated that only RGR was altered by allelopathic effects (p = 0.043, Table 1). Specifically, the LRE treatments raised RGR by 25% compared to the control (Table 1, Fig. 3D). The allelopathic effects had no significant effects on plant height, CA, and TAB (Table 1, Fig. 3A–C). However, according to the multiple comparisons, TAB was significantly improved by the HRE treatment, increasing 40% by compared to the control (Table 1, Fig. 3C). Further analysis of four individual species found that the HRE treatment significantly improved the aboveground biomass of only C. album (Suppl. material 1: fig. S1A).

Figure 3.

Responses of the growth parameters of native plant community to the allelopathy and its interactions with AMF (n = 10). The boxes represent the range of maximum and minimum values, with a bold horizontal line indicating the average value. Colors indicate different AMF treatments (orange, AMF; purple, without AMF). The growth parameters include height (A) crown area (CA) (B) total aboveground biomass (TAB) (C) and relative growth rate (RGR) (D). There were five different concentrations of the root aqueous extracts of Rhus typhina including 0 g L^{− 1} (CK), 10 g L^{− 1} (LRE), 20 g L^{− 1} (NRE), 30 g L^{− 1} (MRE), and 40 g L^{– 1} (HRE). The different letters indicate a significant difference (p ≤ 0.05) with Duncan’s test.

Inoculation with AMF increased TAB by 57% in the LRE treatment and decreased it by 29% in the HRE treatment compared to the NoAMF treatment (p = 0.006, Table 1, Fig. 3C). Overall, plant height was not influenced by AMF while the LRE treatment increased plant height by 34% compared to the corresponding NoAMF treatments (Table 1, Fig. 3A). Moreover, AMF treatments did not significantly change all other growth parameters, including CA and RGR (Table 1, Fig. 3B, D).

The interaction between allelopathic effects and AMF significantly affected on TAB and the LRE and NRE treatments increased TAB by 48% and 45% respectively, compared to the control (p = 0.005, Table 1, Fig. 3C). Meanwhile, the LRE treatment resulted in a 27% advance in plant height compared to the control (Fig. 3A). Nevertheless, the CA and RGR were not influenced by the interaction between allelopathic effects and AMF treatments (Table 1, Fig. 3B, D).

We examined 28 pairs of growth and physiological traits in the native community using a Pearson correlation test. Nine pairs for the NoAMF treatments and 10 pairs for the AMF treatments were significantly correlated at p ≤ 0.05 (Fig. 4). The six correlated pairs of the NoAMF treatments were also correlated in the AMF treatments. The different three correlated pairs of the NoAMF treatments were RGR and SLA, RGR and Chl, RGR and F_v/F_m, which were positively correlated (Fig. 4). The different four correlated pairs of the AMF treatments were CA and Chl, CA and F_v/F_m, which were negatively correlated, while LN and F_v/F_m, RGR and TAB were positively correlated (Fig. 4).

Figure 4.

Matrix of Pearson product-moment correlation coefficients for the growth and physiological traits of the native community across different AMF treatments (n = 10). Traits include leaf nitrogen (LN), total chlorophyll concentration (Chl), F_v/F_m, specific leaf area (SLA), plant height (H), crown area (CA), total aboveground biomass (TAB), and relative growth rate (RGR). Significant correlations are denoted by bold font and asterisks: *** P ≤ 0.001, ** P ≤ 0.01, and * P ≤ 0.05.

The SEMs were well-ftted to the data (χ² = 4.507, df = 11, P = 0.953, CFI = 1.000, RMSEA = 0.000). Allelopathy had direct negative associations with relative growth rate, total chlorophyll concentration, specific leaf area, leaf nitrogen and F_v/F_m but positive associations with height, crown area and total aboveground biomass (Fig. 5). AMF had a negative direct association with all plant traits except height and total aboveground biomass (Fig. 5). Although leaf nitrogen had a positive association with relative growth rate and total aboveground biomass, the association was not statistically signifcant (p = 0.455; p = 0.470, Fig. 5). Further, AMF had a negative direct association with all plant traits except height and total aboveground biomass (Fig. 5). Total aboveground biomass was also positively linked to relative growth rate, total chlorophyll concentration and specific leaf area, but negatively related to crown area and leaf nitrogen. Also, growth parameters had a strong and positive relationship with height, crown area, relative growth rate and total aboveground biomass. In turn, physiological parameters had direct positive associations with total chlorophyll concentration, specific leaf area, leaf nitrogen and F_v/F_m (Fig. 5). Finally, growth parameters had a direct association with physiological parameters (Fig. 5).

Figure 5.

The structural equation models relating the growth and physiological parameters of the native plant community. Green lines indicate positive relationships while red lines indicate negative relationships. Line thickness depicts the magnitude of path coefficients. Significant effects are indicated by asterisks: *** P ≤ 0.001, ** P ≤ 0.01, and * P ≤ 0.05.

The colonization rates of mycorrhizal in the native plant community were negatively correlated with the concentrations of root aqueous extracts from R. typhina (p < 0.001; R² = 0.7828, Fig. 1). The AMF colonization rates in the LRE, NRE, MRE, and HRE treatments were significantly reduced by 25%, 27%, 33%, and 35% respectively compared to the control (Fig. 1).

Allelopathy, as a plants’ defense mechanism, could alter the physical and chemical properties of the soil conditions (Lorenzo et al. 2013). The normal concentration of allelochemicals can produce stress on the growth of native plants. In response to the stress, the native plant community allocated or produced more LN and Chl to improve the degree of leaf photosynthetic capacity and organic synthesis rate. This observation was consistent with previous studies that compared the photosynthetic properties of native plant species in plots invaded and non-invaded by alien species (Wang et al. 2020). As reported, the ability of sunlight acquisition and the degree of leaf photosynthetic capacity of native plant species were prominently improved under invasion by Solidago canadensis and Erigeron annuus (Wang et al. 2020). In our experiments, when the concentration of allelopathic compounds was above a certain threshold value, the stress no longer influenced the native plant community.

Allelochemicals have long been considered as a mechanism by which invasive species eliminate natives (Bais et al. 2003; Gruntman et al. 2016; Hassan and Mohamed 2020). Nevertheless, few studies have found that allelochemicals of invasive plants have neutral or positive effects on native plants (Gross 2003; Zhang et al. 2020). Moreover, it is important to note that the source heterogeneity of allelochemicals can affect the allelopathic effects (Zhang et al. 2020). In this study, we investigated the allelopathic effects of root aqueous extracts of R. typhina on native plant communities. In conclusion, the low concentrations of allelochemicals promoted RGR, whereas the medium and normal concentrations of allelochemicals fostered LN in native plant communities, contradicting our first hypothesis. In another study, the leaf aqueous extracts of R. typhina decreased all seed germination indices and radicle length of the native plant Lactuca sativa L. (Xu et al. 2023). Although disturbances from competition among plants were excluded when using the extracts to explore the allelopathic effects of invasive plants, the different simulating ways might make the conclusions inconsistent (Callaway et al. 2008; Zhang et al. 2020). Since the allelopathic effects of R. typhina were simulated only by the root aqueous extracts in our study, it is reasonable that our findings differed from those of experiments involving individual plants of R. typhina or using other simulations.

Additionally, we found that the total aboveground biomass of the native plant community was significantly higher in the HRE treatment than in the other extract treatments. We analyzed the allelopathic effects on the aboveground biomass of four species (C. album, A. truncatum, R. chinensis, and V. negundo) that constituted the native plant community separately. Only the aboveground biomass of C. album increased significantly with the HRE treatment, while the aboveground biomass of the remaining three plants remained unchanged statistically. A previous study suggests that under the presence of allelopathic compounds, distinct plant species exhibited varying germination and growth rates (McEwan et al. 2010). Therefore, we speculate that these results might be due to a superior defense of C. album against allelochemicals.

AMF can facilitate host plants growth and provide them with competitive advantages (Antunes et al. 2008; Delavaux et al. 2017; Adomako et al. 2019). However, our results suggest that AMF induced LN decline of the native community under the influence of allelopathic effects of R. typhina. The presence of allelopathic effects was considered as an environmental stress. In response to this stress, the degree of leaf photosynthetic capacity of the native plant community was reduced when AMF existed. Furthermore, under allelopathic effects alone on the native plant community, the response of LN was inconsistent with the results obtained from the combined effects of allelopathy and AMF. The Pearson product-moment correlation coefficients among the growth and physiological traits indicated that CA and F_v/F_m had a negative correlation, while LN and F_v/F_m, CA and TAB positively correlated. In the presence of AMF, low concentrations of allelochemicals hindered LN but promoted TAB in the native plant community. This suggests that native plants prioritize absorbing a greater amount of resources from the soil to resist environmental stress rather than enhancing the degree of leaf photosynthetic capacity in response to mild allelopathy. Based on the resource optimization hypothesis (Agren and Franklin 2003), we speculate that the host plant allocates more available resources to promote the growth of belowground parts. The inoculated AMF can combine with the host plant roots to generate mycorrhizal symbionts that can obtain additional water and nutrients from the soil, contributing to the growth and development of the native plant community (Agren and Franklin 2003).

While the significance of AMF alone in soil has long been recognized in relation to plant invasions, few studies have examined the influence of AMF on allelopathic effects in the context of communities (Pellegrino et al. 2015; Bennett and Klironomos 2019). Our results showed that the TAB was differently affected by the AMF treatment among the growth parameters treated with different strengths of the allelopathic effects. Specifically, AMF treatments significantly promoted TAB of the native plant community in the LRE treatment. Although few empirical experiments have investigated the effects of soil microbes on plant allelopathy, available experiments have found that soil microbes can counteract allelopathy by degrading the secondary metabolites secreted by the plants, which could reduce or completely eliminate allelopathy (Saia and Jansa 2022). This function might have mitigated the negative effects of allelopathy on AMF.

The high concentration of allelochemicals transformed the promotive effect of AMF on TAB into the inhibitory effect. Our results suggest that the high concentration of allelochemicals had a neutral impact on the native plant community. Therefore, this phenomenon can be attributed to two plausible reasons: (1) the allelopathic effects from invasive plants can influence patterns of soil nutrient cycling and the soil nutrient environment, altering the effect of AMF on the host (Zhang et al. 2017; de Vries et al. 2020); (2) allelopathy can decline the AMF colonization rates, influencing the mycorrhizal beneficial symbiosis and reducing the uptake and utilization of water and nutrients from the soil by native plants (Zhang et al. 2017; de Vries et al. 2020). Invasive and native hosts have different preferences for AMF and high concentrations of allelochemicals from R. typhina with strong toxicity may filter the AMF in the soil and accumulate beneficial fungal species to promote their own growth (Antunes et al. 2008; Inderjit and van der Putten 2010; Inderjit et al. 2021), which could also explain the reduction of AMF colonization rates. The filtering could result in a reduction of mycorrhizal communities helpful for the growth of native plant communities (Antunes et al. 2008).