1. Introduction

Many studies have demonstrated that invasive alien plants exerted tremendous impacts on species diversity, ecosystem structure and functioning (Vilà et al., 2011; Waller et al., 2020; Han et al., 2024). However, it is unclear whether these impacts would be exacerbated by other global change factors such as elevated atmospheric CO₂ concentration (eCO₂) as few related studies have been conducted (Xiao et al., 2013). The concentration of atmospheric CO₂ has increased rapidly since the Industrial Revolution and will continue to increase with globalization in future (IPCC, 2014). The eCO₂ may magnify the impacts of invasive plants on ecosystems because they are more positive responses to the eCO₂ than native plants (Lei et al., 2012; Liu et al., 2017). Thus, it is essential to elucidate the effects of invasive plants on ecosystems in relation to eCO₂.

Soil nematodes are the most abundant animal group on the earth, and the important composition of soil biota, which play an important role in determining nutrient cycling, energy flow, and many other ecosystem processes (Yeates, 2003; Yeates et al., 2009; Shao et al., 2016). Nematodes are often used to evaluate soil quality (Bardgett and van Der Putten, 2014). It has been demonstrated that invasive alien plants can influence soil nematodes, and consequently soil quality (Reid and Emery, 2018; Zhang et al., 2018). For example, Qin et al. (2019) found that invasive Ambrosia artemisiifolia significantly increased soil nematode density. Reid and Emery (2018) found that the invasive Gypsophila paniculata reduced the heterogeneity of soil nematodes at small spatial scales. Invasive alien plants may affect soil nematodes by altering the availability of resources through rhizosphere (exudates) effects and litter inputs (Zhang et al., 2018).

Enrichment of atmospheric CO₂ concentration often promotes invasions of alien plants, which may aggravate the impacts of invasive plants on soil nematodes (Kao-Kniffin and Balser, 2007; Liu et al., 2017). In addition to increasing the inputs of leaf and root litters and root exudates into soil, eCO₂ may also affect characteristics of these plant-derived resources for invasive alien plants, and therefore influencing soil nematodes. For example, eCO₂ decreased nitrogen, but increased cellulose, hemicellulose and lignin contents of litter (Hager et al., 2016; Rai et al., 2020). Xiao et al. (2013) found that eCO₂ increased proportion of bacteria-feeding nematodes, but decreased the proportion of herbivorous nematodes in soil of the invasive Chromolaena odorata. However, most of the related studies were conducted in only one growing season (Xiao et al., 2013; Lu et al., 2022), which may not well elucidate the effects of invasive plants on soil nematodes, as the responses of the invasive plants to eCO₂ were influenced by experimental duration (Blumenthal et al., 2022). Thus, a long-time experiment (lasting over one growing season) is urgently required.

Relative to aboveground parts, plant root responds more intensively to eCO₂, which may lead to the greater effects of plants on soil nematodes under the eCO₂. For example, Zhang et al. (2010) found that eCO₂ resulted in a 40% increase in the root biomass of Leymus chinensis, whereas the aboveground biomass increased only by 9%. Day et al. (2013) found that eCO₂ had significantly greater effects on belowground relative to aboveground parts of a plant. Furthermore, through a meta-analysis, Dong et al. (2021) revealed that eCO₂ increased efflux amount of total dissolved organic carbon by 31%, and efflux rates of soluble-sugars, carboxylates and citrate by 47%, 111%, and 16% in root exudates, respectively. An increased allocation of resources to roots may amplify the effects of below relative to aboveground of invasive alien plants on soil nematodes under eCO₂. To the best of our knowledge, however, no study has been conducted to test the effects of different plant parts (above- vs. belowground) of invasive plants on soil nematodes under eCO₂.

To assess the effects of invasive alien plants on soil nematodes under different CO₂ concentrations, we cultivated an annual invasive alien plant Xanthium strumarium and two annual phylogenetically related natives in open-top chambers (OTCs) with ambient (aCO₂) and elevated (eCO₂) atmospheric concentrations over three consecutive years. We also tested the roles of different plant-derived resources in determining the effects using leaf litters, root litters, and root exudates of these species collected under aCO₂ and eCO₂. We hypothesize that (1) species and CO₂ concentrations can significantly affect soil nematode community composition and structure, and the effects of eCO₂ may be greater for invasive relative to native species as the invasive species are often more responsive to the eCO₂ (Song et al., 2009; Liu et al., 2017). (2) With the increasing in experimental duration, the effects of species and eCO₂ on soil nematodes may become greater due to the increase in the input of plant-derived resources, especially for the invader, and thus result in the difference between invasive and native species in their effects on soil nematodes. (3) Root litter and exudates may play more important role than leaf litter in influencing soil nematodes, and therefore, the magnitude of the belowground effects may increase more greatly with increasing atmospheric CO₂ concentration, especially for the invasive species. This study will contribute to revealing effects of invasive alien plants on soil ecosystems under the background of continuous increase of atmospheric CO₂ concentration and its related mechanisms.

2. Materials and methods 2.1. Study site and plant species

This study was conducted at Tianzhushan experimental station (41°50′9″ N, 123°34′33″ E; 53 m a.s.l.) of Shenyang Agricultural University, located in Shenyang, Liaoning Province, Northeast China. The site is in the warm-temperate continental monsoon climate. The mean annual temperature is 8.5 ℃, mean annual precipitation is 698.5 mm (mainly in June to September), mean annual sunshine is 2628.3 h, and annual frost-free season is 171 d (Chen et al., 2022; Lu et al., 2022).

The invasive plant Xanthium strumarium was selected as the study material. It is an annual herb in Compositae, and native to North America. It was first found in Beijing, China in 1991, and now has become a harmful invasive species in north China (Han et al., 2024). Our previous study showed that X. strumarium has higher plasticity in response to increasing soil nutrient (Wang et al., 2022). Phylogenetically related invasive and native plants generally share more similar biological and ecological characteristics than non-related ones, and thus are often compared to elucidate the invasiveness and ecological impacts of invasive plants (Feng, 2008; Wang et al., 2022). Thus, the invader was compared with its native congeneric Xanthium sibiricum and confamilial Bidens biternata in this study. The details of these species were described by Lu et al. (2022).

2.2. Experimental design and sampling 2.2.1. Effects of the plants under different CO₂ concentrations on soil nematode community 2.2.1.1. Plant cultivation

The cultivation experiment was conducted annually for three consecutive years in four CO₂ open-top growth chambers (OTCs; QiuShi environment, Hangzhou, China), two for ambient (aCO₂) and the other two for elevated atmospheric CO₂ concentration (eCO₂) treatments. The available area was 9 m² in each OTC. The CO₂ concentrations were automatically controlled around 400 μmol mol⁻¹ and 800 μmol mol⁻¹ in the OTCs as aCO₂ and eCO₂ treatments, respectively (Lu et al., 2022). For each of the three species under each CO₂ treatment, 24 plants were grown in the first year, and 8 plants were harvested annually. In the first year, the growth substrate was a mixture of 70% forest topsoil and 30% river sands, in which the contents of organic matter, total carbon, total nitrogen, total phosphorus, available nitrogen and available phosphorus were 5.89 mg g⁻¹, 4.10 mg g⁻¹, 0.60 mg g⁻¹, 0.42 mg g⁻¹, 32.15 mg kg⁻¹ and 11.22 mg kg⁻¹, respectively. One plant was cultivated in one pot (15 cm × 30 cm × 18 cm in bottom diameter × caliber × height) which contained 6 kg of substrate. In total, 72 plants (4 replicates × 3 species × 2 CO₂ concentrations × 3 years) were cultivated in the first year. Two of 4 replicates were grown in one of the two OTCs for each species and CO₂ treatment. In the middle of May, the seedlings were transplanted into OTCs, the air temperature was 28/18 (day/night) ℃, and air humidity was kept at 60 (±10)% (Lu et al., 2022). The pots were watered every day without water stress. At the end of the growing season, the naturally dead roots were remained intact in pots, and aboveground (leaf and stem) litters were collected and put on the surface of their source pots, respectively. In the second and third year, the seedlings of each species were grown in the pots, in which the same species had be grown under the same CO₂ concentration in the previous year. The methods for control of CO₂ concentration, seedling preparation, and cultivation were described in detail by Lu et al. (2022).

2.2.1.2. Soil collection

In August of each study year (after 90 d treatments), four pots of each species under each CO₂ treatment (two pots from each of the two OTCs with the same CO₂ concentration) were randomly selected to collect soil for measurement of nematodes. In total, 24 pots (4 replicates × 3 species × 2 CO₂ concentrations) were harvested in each study year. To exclude surface effects (often with algae), the topsoil (0.5 cm) was removed for each pot. Then ≈200 g soil was collected in the center, sieved through a 10 mesh sieve, mixed evenly, and divided into two part: one was stored at 4 ℃ for nematode separation and the other was dried at room temperature for measuring soil characteristics.

2.2.2. Effects of litters (leaf and root) and root exudates on soil nematode community 2.2.2.1. Litter collection

From September to October of the second year of the cultivation experiment, leaf and root litter was collected from four severely aged or dead plants (pots) for each species under each CO₂ treatment, respectively. Fine root (< 2 mm) litter was sieved from the soil, and rinsed carefully with tap water. The leaf and root litter from each plant under each CO₂ treatment was dried at 60 ℃ to constant weight, and randomly divided into two parts: one was grinded for measuring litter characteristics, and the other for evaluating its effects on soil nematodes.

2.2.2.2. Root exudates collection

In the spring of the fourth year of the study, root exudates was collected using the nutrient solution culture method for each of the studied species under each CO₂ treatment. Firstly, the seedlings of each species were grown in sand with diluted Hoagland nutrient solution (0.5 concentration) under each CO₂ treatment. When being ≈10 cm in tall, 96 seedlings of each species under each CO₂ treatment were transplanted into eight hydroponic boxes (30 cm × 20 cm × 15 cm; 12 seedlings per box), and 3 L of diluted Hoagland nutrient solution was added in each of the boxes. For each box, air was continuously suppled using a pump, and the nutrient solution was suppled every two days to keep the nutrient solution at 3 L. One week later, we rinsed plant roots of each box with distilled water, grew the seedlings in a new container filled with diluted Hoagland nutrient solution for 3 d, and collected the solution (containing root exudates), which was stored at 4 ℃. The collection process was repeated for three times. The solution was filtered with absorbent cotton and filter papers, adsorbed with macroporous resin (D101; eluted with 90% ethanol) and eluted with 95% ethyl alcohol for three times. Finally, the solution was concentrated to 240 mL using rotary evaporator (RE-5210A, Shanghai Yarong, China) at 45 ℃.

2.2.2.3. Effects of litters and exudates on soil nematodes

This experiment was conducted in a greenhouse in the early winter of the fourth year of the study. The same growth substrate (3 kg) that was used to grow the seedlings in the first year of the experiment was put into pots (10 cm × 15 cm × 15 cm). According to previous observation of leaf and root litters production of the studied species under the two CO₂ treatments (≈2 : 1), 5.0 g leaf litter and 2.5 g root litter (due to limited fine root litter) were added into the pots, respectively. Based on the size of seedlings used for exudates collection and the volume of the test substrate, 35 mL root exudates (had be diluted 5 times) from each species under each CO₂ treatment were added into the pots, respectively. Leaf litter was laid on the soil surface, root litter was buried in the soil at a depth of 5 cm. 35 mL diluted root exudates was sprayed on the soil surface once every two days. Litters and control (CK; without treatment of any plant-derived resources) treatments were sprayed with 35 mL distilled water once every two days. In total, 76 pots (3 species × 2 CO₂ concentrations × 3 additives × 4 replicates + 4 CK) were cultivated in the greenhouse (day/night temperature was ≈25/10 ℃). Seventy days later, the topsoil (0.5 cm) and the undecomposed litters were removed for each pot, and then ≈300 g soil was collected to determine soil nematodes (≈200 g stored at 4 ℃) and soil characteristics (≈100 g dried at room temperature), respectively.

2.3. Measurements 2.3.1. Soil characteristics

Soil total carbon (TC) and nitrogen (TN) contents were measured using the elemental analyzer (EA 3000, Euro Vector, Italy). Soil organic matter content (SOM) was measured using the potassium dichromate titration method (acidic-digestion), and soil available nitrogen content (AN) was measured using the alkaline hydrolysis diffusion method (acidic titration). Soil available phosphorus (AP) and total phosphorus (TP) contents were measured using the molybdenum-antimony anti-colorimetric method (alkaline-digestion). The concrete methods were described in our previous study (Lu et al., 2022).

2.3.2. Litter characteristics

Nitrogen (N) and carbon (C) contents of leaf and root litters were measured using the elemental analyzer (EA 3000, Euro Vector, Italy), and cellulose and lignin contents of the litters were measured according to Bhaskara et al. (1999).

2.3.3. Nematode isolation and identification

Soil nematodes were isolated using the wet sieving - sucrose centrifugation method, which was improved from sucrose centrifugal flotation method (Mao et al., 2004; Liu et al., 2008). In brief, 100 g of fresh soil was fully dispersed using 3 L distilled water in a 4 L container, after standing for 3 min, the suspension was filtered using a triple-deck screen (40-80-400 mesh from top to bottom). The remaining soil in the container was fully dispersed again using 3 L distilled water, and the suspension was filtered again using the method described above. These processes were repeated for 3 times. We then washed the triple-deck screen using distilled water from top to bottom, took out the lowest 400 mesh screen, and washed the residue (containing nematodes) into a 100 mL centrifuge tube using 60 mL distilled water, and finally centrifuged at 2000 rpm for 5 min. The supernatant was poured out, and 10 mL sucrose solution (454 g L⁻¹) was added into the centrifuge tube, stirred evenly with glass rod, and centrifugated again at 2000 rpm for 2 min. The supernatant was filtered using a 500 mesh sieve, and then washed the nematodes from sieve into a test tube with 10 mL distilled water, which was stored at 4 ℃ for 2–3 d. We then carefully sucked up the upper layer water using suction tube until the volume of the solution decreased to 2 mL, and killed the nematodes by water-bath at 65 ℃ for 3 min. Finally, the nematodes were fixed with 5 mL of 5% formalin solution, put into a grided Petri dish, and counted under stereoscopic microscope (ind. 100 g⁻¹ dry soil). We randomly selected 100 individuals, identified them up to genus, and also classified them into four trophic groups: herbivores (He), bacterivores (Ba), fungivores (Fu), and omnivores-predators (OP) (Yeates et al., 1993; Ferris et al., 2001, http://nemaplex.ucdavis.edu/index.htm).

2.4. Statistical analyses

Linear mixed models (LMM) were used to analyze the effects of plant species, CO₂ concentrations, cultivation years, and their interactions on nematode abundance (total and each trophic group), and the Shannon-diversity. The OTC number was treated as a random factor (two of the four replicates from one of the two OTCs with the same CO₂ concentration). These analyses were conducted using lmer() in R package lmerTest (Kuznetsova et al., 2020). PCoA and PERMANOVA (based on bray–curtis distance) were used to analyze the differences of nematode community among different planting years, plant species and CO₂ concentrations, respectively. The PCoA and PERMANOVA were conducted using capscale() and adonis() in R package vegan, respectively. Effects of soil characteristics on soil nematode community were analyzed using rda() and rdacca.hp() in R packages vegan and rdacca.hp, respectively (Lai et al., 2022, Oksanen et al., 2020). Correlations between soil characteristics and each nematode trophic group were analyzed using mantel() in R package vegan (Oksanen et al., 2020), and the correlation between soil characteristics and nematode abundance was analyzed using Pearson's correlation method.

Three-way ANOVA was used to analyze the effects of additives (leaf litter, root litter, and root exudates), species status (invasive vs. native), CO₂ concentrations and their interactions on soil nematode abundance (total and each trophic group). Independent samples t-test were used to analyze the differences in nematode abundance between aCO₂ and eCO₂, invasive and native species, and each additive treatment and CK, respectively. PCoA and PERMANOVA (based on bray–curtis distance) were also used to analyze the differences in nematode community among leaf litter, root litter, and root exudate input treatments, and the differences between species status and CO₂ concentrations. The proportional variation of soil nematode community explained by soils and each soil parameter were analyzed using rad() and rdacca.hp() in R packages vegan and rdacca.hp, respectively (Lai et al., 2022, Oksanen et al., 2020). All these analyses were performed in R 4.0.5 (R Core Team, 2021).

3. Results 3.1. Effects of three plant species on soil nematodes under different CO₂ concentrations

A total of 22 genera of nematodes were identified in the soil that were trained by the invasive and native plants under both aCO₂ and eCO₂ for one, two and three years: 7 genera of herbivores, 8 genera of bacterivores, 2 genera of fungivores, and 5 genera of omnivores-predators (Table S1). Among them, herbivorous nematodes accounted for the largest proportion of the total number of soil nematodes (48.5%, 59.4%, and 63.5% in first, second, and third year, respectively). The dominant genera were Rhabditis (bacterivores), Dorylaimellus (herbivores), and Helicotylenchus (herbivores) in soil collected in the first, second, and third year of the cultivation experiment, respectively. Total nematode abundance was significantly affected by CO₂ concentrations, plant species, experimental duration, and their interactions (Table 1). Plant species, experimental duration, and their interactions also significantly affected total nematode abundance regardless of CO₂ concentration (Fig. 1a and b).

Under aCO₂, total nematode abundance and that of the dominant trophic group increased significantly with increasing experimental duration under all plant species, especially under the invasive plant Xanthium strumarium (Fig. 1a, Fig. S1a). In the first and second years of the cultivation experiment, total nematode abundance in the soil trained by X. strumarium were all significantly lower than those trained by the native plant X. sibiricum, but similar with and higher than those trained by the native plant Bidens biternata, respectively. In the third year of the cultivation experiment, total nematode abundance in the soil trained by the X. strumarium was significantly higher than that trained by each of the natives. Under eCO₂, total nematode abundance increased significantly with increasing experimental duration under X. strumarium, changed little under X. sibiricum, but decreased under B. biternata (Fig. 1b). Compared with total nematode abundance in soil trained by the two native species, that in the soil trained by X. strumarium was significantly lower in the first, but higher in the second and third years of the study. The eCO₂ treatment resulted in a 151% increase in total nematode abundance in soil trained by X. strumarium in the second year, and a 245% increase in soil trained by B. biternata in the first year of the study (Fig. 1a and b). For X. sibiricum, however, eCO₂ decreased total nematode abundance in the second and third years of the study by 41% and 42%, respectively.

In general, Shannon-diversity of soil nematodes was not affected by CO₂ concentrations, plant species, and cultivation years, but significantly affected by the interactions between CO₂ concentrations and cultivation years, and between plant species and cultivation years, respectively (Table 1; Fig. 1c and d). Under aCO₂, Xanthium strumarium had lower Shannon-diversity than that of the two natives in the first year (Fig. 1c). Under eCO₂, compared with Bidens biternata, X. strumarium had significantly higher Shannon-diversity in the first year, but lower in the third year (Fig. 1d). The invader had significantly lower Shannon-diversity than X. sibiricum in the second year.

When ignore other factors, soil nematode community composition was significantly affected by CO₂ concentrations (R² = 0.04, P = 0.015), plant species (Xanthium strumarium vs. Bidens biternata; R² = 0.17, P < 0.001), and cultivation years (Fig. 2a, R² = 0.61, P < 0.001), respectively. The composition of soil nematode community was significantly affected by CO₂ concentration for the two native species, but was not for X. strumarium with different experimental duration (Table 2). Soil nematode community composition was different between invasive X. strumarium and these two natives under each study year, respectively (Table 2). Soil nutrient characteristics explained 30.2% (adjusted 23.7%) of the variation of soil nematode community composition (Fig. 2b). Among soil parameters, SOM showed the largest proportion of explanation (occupied 31.7%) on nematode community composition, and had strong correlations with abundances of herbivorous (dominant) and bacterivorous groups (Fig. 2c).

3.2. Effects of leaf and root litters and root exudates on soil nematodes

A total of 17 genera of nematodes were identified in the soil treated with leaf and root litter and root exudates of invasive and native plants under both aCO₂ and eCO₂: 4 genera of herbivores, 7 genera of bacterivores, 3 genera of fungivores, and 3 genera of omnivores-predators. Bacterivores accounted for the largest proportion of the total number of nematodes (67.3%, 61.0%, and 66.1% in leaf litter, root litter, and root exudates treatments, respectively; Table S2). Bacterivorous Chiloplacus and Rhabditis were the dominant genera in soil treated with leaf and root litter, respectively. The dominant genera in soil treated with root exudates were bacterivorous Chiloplacus and fungivorous Aphelenchoides. CO₂ concentration (except for fungivores and omnivores-predators), plant species, plant-derived resources addition, and their interactions all significantly affected total nematode abundance and that of each trophic group (Table 3).

Leaf litter of two natives grown under aCO₂ significantly decreased total nematode abundance (Fig. 3a). Total nematode abundance was increased, not changed, and decreased by the addition of leaf litters of Xanthium strumarium, X. sibiricum, and Bidens biternata grown under eCO₂, respectively (Fig. 3b). Total nematode abundance was significantly higher in soil treated with leaf litter of the invader relative to the two natives grown under either aCO₂ or eCO₂. For the two Xanthium species rather than B. biternata, total nematode abundance was significantly higher when treated with leaf litter collected under eCO₂ relative to that under aCO₂.

Root litter of all three species grown under both aCO₂ and eCO₂ all significantly increased total nematode abundance (Fig. 3). Total nematode abundance was significantly higher in soil treated with root litter of Xanthium strumarium relative to Bidens biternata collected under both eCO₂ and aCO₂. In contrast, total nematode abundance was significantly lower in soil treated with root litter of X. strumarium relative to X. sibiricum grown under eCO₂, but not under aCO₂. For the two Xanthium species rather than B. biternata, total nematode abundance was significantly higher in soil treated with root litter collected under eCO₂ relative to under aCO₂.

Root exudates of Xanthium strumarium collected under both aCO₂ and eCO₂ significantly increased total nematode abundance, but those of the two natives under aCO₂ rather than eCO₂ decreased the total nematode abundance (Fig. 3). Total nematode abundance was significantly higher in soil treated with root exudates of invader relative to its native congener under both CO₂ concentrations, and relative to Bidens biternata under aCO₂. For B. biternata rather than the two Xanthium species, total nematode abundance was significantly higher when treated with root exudates collected under eCO₂ relative to aCO₂. It appeared that effects of root litter on nematode abundance was greater than those of leaf litter and root exudates for all species under both CO₂ concentrations, especially considering the fact that the amount of added root litter was lesser than that of leaf litter (2.5 vs. 5 g).

In general, soil nematode community composition was significantly different among treatments of leaf litter, root litter and root exudates of invasive and native species under different CO₂ concentrations (Fig. 4a; R² = 0.50, P < 0.001). CO₂ concentration and plant species all significantly affected nematode community composition under the treatments of three plant-derived additions (for leaf litter, R² = 0.90; for root litter, R² = 0.68; for root exudates, R² = 0.76; all P < 0.001).

Soil available nitrogen (AN) and SOM influenced nematode community composition the most greatly (mainly via bacterivores and herbivores), which explained the variation by 15.5% and 15.0%, respectively (Fig. 4b and c). When treated with root litter, SOM influenced nematode community composition the most greatly (mainly via omnivores-predators), which explained the variation by 10.9% (Fig. 4d and e). When treated with root exudates, soil total nitrogen (TN) and available phosphorus (AP) influenced nematode community composition the most greatly (mainly via omnivores-predators), which explained the variation by 7.2% and 7.0%, respectively (Fig. 4f and g).

4. Discussion

It is widely known that invasive alien plants have caused enormous threats to natural ecosystems, and these impacts will be exacerbated by the changes of other global factors such as elevated atmospheric CO₂ (eCO₂). Soil nematodes, an important component of soil biota in determining ecosystem functions (Yeates, 2003; Yeates et al., 2009), were affected by the invasion of alien plants (Reid and Emery, 2018; Qin et al., 2019) and eCO₂ (Xiao et al., 2013). However, the effects of invasive plants and CO₂ concentration on soil nematodes were seldomly studied using long-term experiments, which impede the understanding of the effects of these factors on the soil nematodes under global change (Elgersma et al., 2011). In a three consecutive years experiment, our results showed that soil nematode community composition was significantly affected by the invasive plant (Xanthium strumarium vs. Bidens biternata; R² = 0.17, P < 0.001), CO₂ concentration (R² = 0.04, P = 0.015), and experimental duration (Fig. 1a, R² = 0.61, P < 0.001), respectively. In addition, plant species, CO₂ concentration, experimental duration, and their interactions all significantly influenced soil total nematode abundance and that of each trophic group (in most cases) (Fig. 1). This study provides clear evidence for the interactive effects of invasive alien plants and eCO₂ on the soil ecosystem.

Our results showed that soil nutrient characteristics explained 30.2% of the variation of nematode community (Fig. 2b), suggesting that effects of these plant species on nematode community may be due to their effects on soil available resources. In our addition experiment, we indeed found that the root litter of Xanthium strumarium significantly increased soil organic matter (SOM) concentration (Table S4), which had strong correlations with the abundances of herbivores (dominant trophic group) and bacterivorous, respectively (Fig. 2c).

4.1. Effects of plant species, CO₂ concentration, study duration, and their interactions on soil nematode abundance

Our results indicate that plant species, CO₂ concentration, experimental duration, and their interactions all significantly influenced total soil nematode abundance and that of each trophic group (in most cases). These interactions suggest that effects of invasive and native species on soil nematodes will be different with changes of atmospheric CO₂ concentration and experimental duration. Thus, it is pivotal to take into account of the effects of environmental conditions and experimental duration when explore effects of invasive plants on soil nematodes. Similarly, it is also vital to consider the effects of plant species and experimental duration when investigating the effects of environmental changes, such as CO₂ enrichment on soil nematodes as the results were dependent on species and experimental duration (Table 1).

Consistent with our hypothesis, both species and CO₂ concentration had significant effects on total nematode abundance and that of the dominant trophic group (herbivores), and the effects of the eCO₂ were greater for the invasive Xanthium strumarium than for the natives (Fig. 1 and Fig. S1). For example, in the second year of the study, eCO₂ increased total soil nematode abundance by 151% under the invader, much higher than that under the natives. Therefore, total soil nematode abundance was higher in the soil trained by X. strumarium than by the two natives under eCO₂, but not under aCO₂ in the second year of the study (Fig. 1). The greater positive effects of the invader under eCO₂ may be due to its greater responses to the eCO₂, which in turn input more plant-derived resources into the soil (Yeates et al., 1999; Zhou et al., 2022). In a meta-analysis, Liu et al. (2017) found that invasive plants had significantly higher growth performance than natives under eCO₂. Greater biomass production by X. strumarium was also found under eCO₂ in this study (data not shown).

Also consistent with our hypothesis, with increasing experimental duration the invasive plant increased total nematode abundance and that of the dominant trophic group more greatly than the natives under aCO₂. Our results showed that Xanthium strumarium increased total nematode abundance by 412% in the third relative to the first year of the study, much higher than the native X. sibiricum (131%) and Bidens biternata (148%) did. Under eCO₂, X. strumarium increased total nematode abundance by 477% in the third relative to the first year of the study, while X. sibiricum did not influence, and B. biternata even decreased the total nematode abundance by 41%. These results suggest that the differences in soil nematode abundance between the invasive and native species would increase with increasing experimental duration. We indeed found that X. strumarium had significantly higher total nematode abundance and that of the dominant trophic group than the two natives in the third year of the study under both aCO₂ and eCO₂. The differences in the effects between the invasive and native plants on soil nematode abundance with increasing experimental duration may be associated with the inter-group differences in their growth rate (Pyšek and Richardson, 2007; van Kleunen et al., 2010). Invasive plants with higher growth rate may input more litters and exudates into soil than native plants with increasing duration (Castro-Díez et al., 2014; Jo et al., 2017), exerting greater effects on the soil nematode abundance. These results indicate that it is critical to take into account the effects of experimental duration when testing the influences of invasive plants on soil nematodes.

With increasing experimental duration, the effects of eCO₂ on soil nematodes were different among the studied species, which was not fully consistent with the hypothesis. For example, eCO₂ significantly increased total nematode abundance in soil under Xanthium strumarium in the second year of the study, while decreased it under the native X. sibiricum in the second and third years of the study. For native Bidens biternata, eCO₂ increased total soil nematode abundance in the first year, and had no effects in the second and third years of the study (Fig. 1). With increasing study duration, the interspecific differences of effects on soil nematode abundance under different CO₂ concentrations may be associated with the interspecific differences of leaf and root litter and/or root exudates that input into the soil. With increasing experimental duration, invasive relative to native plants may produce more litter (also more root exudates), and the litter of the formers may be decomposed more quickly due to their lower structural and defensive compounds (Müller-Schärer et al., 2004; Huang et al., 2020), and these inter-group differences may be amplified by the eCO₂ as the greater responses of the invasives (Liu et al., 2017). Thus, invasive relative to native plants may increase soil nematodes more greatly with increasing experimental duration, especial under the eCO₂ (Table S3; Yeates et al., 1999). The negative effects of native X. sibiricum on the soil nematodes under eCO₂ may be associated with its greater production of toxic and/or defensive compounds (Dhakshinamoorthy et al., 2014). It has been found that eCO₂ increased phenolic syntheses in Cymbopogon citratus and C. proximus (Almuhayawi et al., 2021), and in Austrodanthonia bipartita and Bothriochloa macra (Johnson and Hartley, 2018).

4.2. Effects of plant-derived resources on soil nematode abundance

Consistent with cultivation experiment, plant-derived resources additions, plant species, CO₂ concentration, and their interactions indeed affected total soil nematode abundance (Table 3). Our results showed that root litter promoted total soil nematode abundance more greatly than leaf litter and root exudates under both aCO₂ and eCO₂ (Fig. 3). The greater effects of root litter relative to leaf litter and root exudates may be due to their differences in chemical contents. Many studies have found that leaf litter and root exudates had higher concentrations of toxic compounds such as soluble phenolics than root litter (Xia et al., 2015; Weinhold et al., 2022). In addition, the higher carbon concentration in root litter relative to leaf litter and root exudates may also contribute to the higher soil nematode abundance (Li et al., 2021). In our study, the positive effects of root litter on soil nematodes may be underestimated because only 2.5 g root litter (only half of the amount of leaf litter) were added into the experimental soil.

Consistent with the plant-derived resource addition experiment, Xanthium strumarium had the higher total soil nematode abundance than the native Bidens biternata under both CO₂ concentrations in the second and third years of the study. These results were also consistent with our observation that invader produced more fine roots (litter) than B. biternata. The positive effects of the addition of leaf litter and root exudates (except under eCO₂) were also greater for X. strumarium than for B. biternata (Fig. 3), which also partially explained the higher total nematode abundance in soil trained by X. strumarium than by B. biternata.

Effects of additions of plant-derived resources (litter and exudates) on total soil nematode abundance were higher under eCO₂ than those under aCO₂ in most cases, which contributed to the higher total nematode abundance in the soil trained by Xanthium strumarium (in the second year of the cultivation experiment) and Bidens biternata (in the first year of the cultivation experiment) under eCO₂ than those under aCO₂. However, different effects were found between the soil trained by X. sibiricum and the soil into which X. sibiricum-derived resources were added (Fig. 1, Fig. 3), which may be affected by other factors.

4.3. Effects of plant species, CO₂ concentration, study duration, and their interactions on soil nematode diversity

In contrast to effects on abundance, our results showed that plant species, CO₂ concentration, and experimental duration did not affect Shannon-diversity of soil nematodes (generic level). Consistent with our results, some previous studies also found that invasive alien plants did not affect soil nematode diversity (Renčo et al., 2019; Jurová et al., 2020). However, other studies found that the diversity of soil nematodes was significantly affected by the invasion of exotic plants (Fitoussi et al., 2016; Porazinska et al., 2022). In a meta-analysis, Zhou et al. (2022) found that eCO₂ decreased genetic richness of nematodes. Different results among these studies indicate that diversity of soil nematodes was influenced by many other factors, such as resources availabilities in substrate of the study (Thakur et al., 2019). Plant species and CO₂ concentration had significant interactive effects with study duration on Shannon-diversity of soil nematodes, although they did not affect Shannon-diversity of soil nematodes (Table 1; Fig. 1c and d). These results indicate that the effects of plant species and CO₂ concentration on soil nematode diversity would be different with different experimental duration.

We could not simply extrapolate our results into other invasive and native species because only one invasive and two native species were included in our study. Enrichment of CO₂ decreased soil nematode abundance under the native plant Xanthium sibiricum, but not under native Bidens biternata in the second and third years of the study (Fig. 1). Effects of plants on soil nematodes under eCO₂ may be species specific, especially over a long period of time. Species with higher growth rate may be more responsive to eCO₂ (Liu et al., 2017), and hence have greater effects on soil nematodes with increasing duration. Thus, studies with more species are urgently needed to explore their effects on soil nematodes under climate changes in future. In addition, our study did not consider effects of soil microorganisms on soil nematodes. Fungivores and bacterivores can interact with fungi and bacteria in soil (Carrascosa et al., 2014). Soil nematodes may cascade the changes of soil fungi and bacteria, which have important roles in nutrient transformation and energy transfer within soil ecosystems.

5. Conclusions

Our results showed that plant species, CO₂ concentration, and their interactions had significant effects on total soil nematode abundance and that of the dominant trophic group (herbivores), and these effects were dependent on the experimental duration. However, the Shannon-diversity of soil nematode was not affected by these factors. Under ambient CO₂, the invasive plant Xanthium strumarium and its phylogenetically related natives increased total nematode abundance and that of the dominant trophic group with increasing experimental duration, respectively, and the effects of the invader were greater than those of the two natives. With increasing experimental duration, eCO₂ increased soil total nematode abundance and that of the dominant trophic group under the invader, but not under the two natives (no change or even decrease). The effects of root litter on soil nematode abundance were greater than those of leaf litter and root exudates. Our study indicates that the invasion of alien plant species would affect soil nematode community composition by increasing abundance rather than diversity, and these effects would be aggravated by eCO₂. In addition, the effects of invasive species and CO₂ concentrations on soil nematode would vary with increasing duration.

Acknowledgements

We are grateful to Chun-Feng Hu and Dan Li for their assistance during plant cultivation and harvesting. This work was supported by the National Key R & D Program of China (2023YFC2604500), the National Natural Science Foundation of China (32171662, 32471753 and 32171666) and the Natural Science Foundation of Liaoning (2020-MS-199).

CRediT authorship contribution statement

Xiu-Rong Lu: Writing – review & editing, Writing – original draft, Visualization, Software, Methodology, Investigation. Ming-Chao Liu: Writing – original draft, Visualization, Software, Methodology, Investigation. Wei-Wei Feng: Methodology, Investigation. Bo Qu: Methodology. Jing-Kuan Wang: Writing – review & editing, Conceptualization. Yu-Long Feng: Writing – review & editing, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.pld.2024.12.002.