Functional trait variations of the invasive plant Alternanthera Philoxeroides and the native plant Ludwigia peploides under nitrogen addition

In natural environments, native plant communities often face varying degrees of invasion pressure from exotic plants^1,2,3,4. During the establishment phase, invasive plants must overcome resistance from native vegetation, yet they often succeed in occupying niches due to their superior competitive abilities⁴. In the expansion phase, competitive advantages directly determine the degree of dominance by invasive species and the severity of their ecological impacts, where intense interspecific competition may even lead to the exclusion of native species⁴. These competitive dynamics are fundamentally driven by differences in functional traits among species and are maintained through niche differentiation and mechanisms of competitive coexistence^5,6,7.

Plants optimize key physiological processes such as photosynthetic carbon acquisition, structural support, and nutrient uptake by adjusting biomass allocation patterns to organs like roots and leaves⁸. Variations in root and leaf morphology, along with tissue nutrient content (e.g., nitrogen concentration), directly reflect plant strategies and efficiency in resource utilization^8,9,10. Furthermore, secondary metabolites serve as critical biochemical traits, not only providing defensive compounds but also significantly influencing plant adaptation to biotic and abiotic environments^5,8. Non-structural carbohydrates (NSCs), as labile carbon reserves, play a central role in plant growth and environmental adaptation by supplying carbon skeletons for development, fueling respiratory metabolism, and participating in osmotic regulation, among other processes¹⁰. Previous research has shown that invasive plants often exhibit greater leaf mass fraction, specific root length (SRL), and root nitrogen concentration, as well as release higher amounts of allelochemicals, compared to co-occurring native species^5,8,9,10.

However, most existing studies in invasion ecology focus on either “pre-invasion” or “post-invasion” community states^{1,9,11,12,13,14}, often overlooking the continuum of invasion degree and failing to systematically analyze how different invasion degrees differentially affect the functional traits of invasive and native plants. This knowledge gap hinders the identification of key invasive traits that determine competitive outcomes^8,9,10.

Moreover, as a critical component of global change, biological invasions often interact synergistically with anthropogenic drivers such as nitrogen deposition, climate warming, and rising atmospheric CO₂ levels^15,16. Research indicates that the success of invasive plants in establishment and expansion closely depends on resource availability and plasticity in their functional traits¹². In recent decades, global fossil fuel combustion and synthetic fertilizer use have led to a significant increase in atmospheric nitrogen deposition^17,18. China has become the world’s third-largest nitrogen deposition hotspot after Europe and North America, with southern regions experiencing an annual deposition rate of up to 63.53 kg N ha⁻¹ yr⁻¹^5,19. As a key nutrient for plant growth, nitrogen availability profoundly influences plant performance and adaptive strategies during invasion^{16,20,21,22,23,24,25}. For instance, in high-nutrient microenvironments, invasive plants often enhance their competitiveness through rapid adjustments in root and leaf trait plasticity⁵. Nevertheless, there remains a scarcity of systematic research on the responses of above- and below-ground traits in invasive and native plants under varying nitrogen levels and invasion degree.

Previous studies have found that aboveground and belowground components of plants are closely interconnected: plants compete for light through above-ground organs (e.g., leaves) while simultaneously vying for nutrients and water via below-ground structures (e.g., roots)⁴. Recent studies have highlighted the critical importance of coordinated root and leaf functional traits in responding to heterogeneous above- and below-ground resource availability^26,27. Such coordination enhances a plant’s capacity to either optimize acquisition of limited resources or minimize demand for specific resources²⁸. For instance, research has demonstrated systematic correlations between root morphological traits and leaf traits associated with nutrient utilization^26,29. This reflects an integrated whole-plant strategy where above- and below-ground components function synergistically to adapt to environmental variables including temperature, light intensity, and nutrient/water availability³⁰. Despite their well-established role in plant resource-use strategies³¹, current understanding of how inter-organ trait coordination responds to environmental changes remains fragmented.

Our earlier study observed that, in a 60-day short-term experiment, although invasive and native plant communities exhibited differential responses in total and root biomass, root morphology, and exudate composition to nitrogen addition, neither showed a significant preference for ammonium vs. nitrate nitrogen⁸. Since leaves are central organs in above-ground resource competition⁴, the physiological and morphological responses of leaves to nitrogen addition were not sufficiently addressed in previous work. Therefore, this study systematically investigates the effects of nitrogen addition on whole-plant biomass, root and leaf morphology, and root exudates in the invasive species Alternanthera philoxeroides and the native species Ludwigia peploides.

Based on the above background, this study establishes three nitrogen addition levels (i.e., control, low, and high nitrogen treatments) and five invasion scenarios (i.e., no invasion, early invasion, mid-invasion, dominant invasion, and native species migration period). We hypothesize that: (1) nitrogen addition will promote the growth of both A. philoxeroides and L. peploides, but increasing invasion degree will suppress the growth of L. peploides; (2) nitrogen addition and invasion degree will significantly alter key root and leaf functional traits in both species; (3) nitrogen addition and invasion degree can regulate the invasion success of A. philoxeroides by modulating root traits, leaf traits, and their interactions, whereas this pattern will not be observed in L. peploides.

Variation in total biomass

Both invasion degree and nitrogen level significantly affected the individual plant biomass of A. philoxeroides and L. peploides, although their interaction was not significant (Fig. 1). Under the same invasion degree, both nitrogen addition treatments significantly increased total individual biomass of A. philoxeroides compared to the control (Fig. 1a). But only high nitrogen significantly increased the total individual biomass of L. peploides (Fig. 1b). Under the same nitrogen treatment, the total biomass of A. philoxeroides showed an increasing trend with the degree of invasion, while the opposite was true for L. peploides (Fig. 1a, b).

Variation in root traits

As shown in Fig. 2, the invasion degree and nitrogen level, as well as their interaction, significantly affected all root trait indices of A. philoxeroides (except for RMF and SRL) and L. peploides (except for RD).

For A. philoxeroides, under different invasion degree, both nitrogen application treatments tended to decrease RMF, RD, and SRL (at early invasion (25%) and mid-invasion (50%)) compared to the CK. In contrast, both nitrogen levels increased root starch, NSC, nitrogen content, and SRL (at dominant invasion (75%) and native species migration period (100%)) (Fig. 2a-f). For L. peploides, across invasion degree, RMF gradually decreased with increasing nitrogen application, while RD showed the opposite trend (Fig. 2g, h). High nitrogen significantly reduced SRL, whereas the effect of low nitrogen varied depending on invasion degree (Fig. 2i). Both nitrogen treatments significantly decreased root starch and NSC content but markedly increased root nitrogen content (Fig. 2j-l).

Compared to the native species migration period (100%), the presence of L. peploides significantly increased root starch and NSC content in A. philoxeroides (Fig. 2d, e). Under the CK treatment, the presence of L. peploides significantly reduced RMF and RD but increased SRL in A. philoxeroides, while its effect on root nitrogen content was minimal (Fig. 2a-c, f). Under both nitrogen application levels, the presence of L. peploides significantly decreased RMF (at mid-invasion (50%) and dominant invasion (75%)) and increased SRL, root starch, and NSC content in A. philoxeroides (Fig. 2a, c-e). Under low nitrogen, the presence of L. peploides significantly reduced root nitrogen content but increased RD (Fig. 2b, f). Under high nitrogen, it significantly increased RD (at early invasion (25%) and mid-invasion (50%)) and root nitrogen content (at mid-invasion (50%) and dominant invasion (75%)) in A. philoxeroides (Fig. 2b, f).

Compared to the absence of A. philoxeroides invasion, the presence of A. philoxeroides showed no significant effect on the RD of L. peploides (Fig. 2h). Under the CK conditions, A. philoxeroides presence exhibited a decreasing trend in RMF, root starch, NSC and nitrogen content of L. peploides, while no significant effect was observed on SRL (Fig. 2g, i-l). Under both nitrogen application treatments, A. philoxeroides presence had minimal effects on root starch and NSC content of L. peploides (Fig. 2j, k). With low nitrogen treatment, A. philoxeroides presence showed limited influence on RMF but significantly increased SRL and root nitrogen content of L. peploides (Fig. 2g, i, l). Under high nitrogen treatment, A. philoxeroides presence significantly reduced RMF of L. peploides while demonstrating minor effects on SRL and root nitrogen content (Fig. 2g, i, l).

Changes in root-secreted secondary metabolites

The results of the Adonis analysis revealed that invasion degree significantly affected the composition of root exudates in both plant species, whereas nitrogen application showed no significant effect on their root exudate profiles (Fig. 3a and c).

For A. philoxeroides, compared to the native species migration period (100%), the presence of L. peploides significantly reduced phenolic compound content in root exudates and showed a decreasing trend for terpenoids (except under both nitrogen treatments) and alkanes content. In contrast, organic acids and amides exhibited opposite trends, while alkaloids content remained unaffected (Fig. 3b). For L. peploides, compared to the absence of A. philoxeroides invasion, the presence of A. philoxeroides showed minimal effects on the content of phenolic compounds, alkanes, alkaloids, organic acids (under both nitrogen treatments), and amides (under CK and low nitrogen treatments) in its root exudates. However, a decreasing trend was observed for terpenoids content, while an increasing trend was noted for organic acids (under CK treatment) and amides (under high nitrogen treatment) (Fig. 3d).

The Pearson correlation analysis revealed significant relationships between root traits and total biomass for both species (Fig. 4). For A. philoxeroides, RD, SRL, and root-secreted amides content showed significant negative correlations with total biomass, whereas root nitrogen concentration, phenolics, and alkanes content in root exudates exhibited significant positive correlations (Fig. 4). For L. peploides, RMF, and root-secreted terpenoids content showed significant positive correlations with total biomass, whereas SRL, root starch and root-secreted organic acids content exhibited significant negative correlations (Fig. 4).

Changes in leaf traits

As shown in Fig. 5, invasion degree, nitrogen level, and their interaction significantly affected all leaf trait indices of A. philoxeroides (except LMF and SLA) and L. peploides (except SLA).

Compared to the CK, nitrogen addition exhibited an increasing trend in LMF for both A. philoxeroides and L. peploides, with a more pronounced effect in A. philoxeroides (Fig. 5a, f). Both nitrogen levels significantly enhanced leaf nitrogen concentration in both species (Fig. 5e, j) but markedly reduced leaf starch (Fig. 5c, h) and NSC content (Fig. 5d, i). Differently, nitrogen application showed a trend of reducing the SLA of A. philoxeroides, but significantly increased the SLA of L. peploides (Fig. 5b, g).

Compared with the native species migration period (100%), the presence of L. peploides non-significant effects the SLA of A. philoxeroides, and had a moderate effect on the LMF under nitrogen treatment. However, it significantly increased the leaf starch and NSC content under CK treatment, and significantly reduced the LMF under CK treatment (at early invasion (25%) and mid-invasion (50%)) and NSC content and leaf nitrogen content under high nitrogen treatment (Fig. 5a-e). Compared to the absence of A. philoxeroides invasion, the presence of A. philoxeroides had minimal effects on LMF and leaf nitrogen content of L. peploides, but significantly reduces its SLA (except at early invasion (25%)), the starch and NSC content in its leaves vary depending on the invasion degree of A. philoxeroides (Fig. 5f-j).

Changes in leaf-secreted secondary metabolites

Results from the Adonis analysis revealed that invasion degree, nitrogen level, and their interaction significantly influenced the leaf exudate composition of A. philoxeroides (Fig. 6a). In contrast, only nitrogen level exhibited a significant effect on the leaf exudate profile of L. peploides (Fig. 6c).

For A. philoxeroides, nitrogen application showed an increasing trend in leaf terpenoids content compared to the CK, while demonstrating decreasing trends for organic acids and alkaloids (under high nitrogen treatment). No significant effects were observed on phenols, amides, alkanes, or alkaloids (under low nitrogen treatment) (Fig. 6b). In contrast, for L. peploides, nitrogen application had no significant effect on leaf terpenoids, amides, and alkanes relative to CK. Minor effects were observed on phenols and alkaloids, while an increasing trend was noted for organic acids content (Fig. 6d).

For A. philoxeroides, compared to the native species migration period (100%), the presence of L. peploides exhibited minimal effects on leaf terpenoids and amides content, no significant effects on leaf phenols and alkaloids content (under both CK and low nitrogen treatments), a decreasing trend in organic acids content, but significant increases in alkanes content, significant reductions in alkaloids content (under high nitrogen treatment) (Fig. 6b).

The Pearson correlation analysis revealed that the biomass of A. philoxeroides was significantly positively correlated with its LMF, leaf nitrogen concentration, and terpenoids content secreted by leaves, whereas its SLA, leaf starch, NSC, and alkanes, amides secreted by leaves showed a significant negative correlation with total biomass (Fig. 7). For L. peploides, its individual biomass was significantly positively correlated with its SLA, leaf nitrogen concentration, and terpenoids content secreted by its leaves, whereas its leaf starch, NSC, and phenols, amides secreted by leaves showed a significant negative correlation with total biomass (Fig. 7).

Correlations among total biomass, root traits, and leaf traits

A. philoxeroides exhibited 56 significant root-leaf trait correlations, whereas L. peploides showed 50 significant root-leaf trait correlations (Fig. 8a, b). The PLS-PM model showed that nitrogen application level (path coefficient = 0.340, P < 0.001) and invasion degree (path coefficient = 0.249, P < 0.05) had a significant positive effect on the total biomass of A. philoxeroides, while nitrogen application level (P < 0.001) and invasion degree (P < 0.001) had a significant negative effect on the root traits of A. philoxeroides; At the same time, nitrogen application level (P < 0.001) and invasion degree (P < 0.05) also had a significant negative effect on the leaf traits of A. philoxeroides, but subsequently, changes in leaf traits promoted the growth of A. philoxeroides (Fig. 9a). There was a significant negative correlation between the invasion degree (P < 0.001) and the total biomass of L. peploides; There was also a significant negative correlation between nitrogen application level and its root traits (P < 0.001) and leaf traits (P < 0.001), but there was a significant positive correlation between invasion degree and its root traits (P < 0.001) and leaf traits (P < 0.01) (Fig. 9b).

While the 90-day experimental period cannot fully capture the long-term dynamics of invasion processes, the short-term response patterns observed in this study remain highly significant. These early adaptive changes may represent critical first steps toward successful invasion.

Nitrogen addition was likely to facilitate the displacement of L. peploides by A. philoxeroides in plant communities experiencing mid-invasion (50%) and dominant invasion (75%)

Biomass serves as a direct measure of plant growth performance and resource-use efficiency^28,30. Nitrogen enrichment generally promotes biomass production of invasive species and native species, invasive plants typically exhibit greater sensitivity to nitrogen enrichment compared to their native counterparts^{24,28,32,33,34}, a pattern consistently observed in our study (Fig. 1a, b). As a critical nutrient regulating plant growth and productivity, nitrogen availability often determines the degree of biomass accumulation by fulfilling metabolic demands^16,21,25,35. Compared to the native L. peploides, A. philoxeroides was a fast-growing, opportunistic perennial herb that typically exhibits faster phenotypic adjustments to nutrient addition or shifts in nutrient availability¹³.

Notably, under nitrogen addition, A. philoxeroides outproduced L. peploides in biomass accumulation within mid-invasion (50%) and dominant invasion (75%) communities (Fig. 1a, b). In coexisting communities, A. philoxeroides and L. peploides occupy analogous ecological niches, and the high overlap of ecological niches may lead to fierce competition between invasive species and closely related native species occupying the same domain range⁵. Previous studies have shown that invasive plants can suppress native species by forming high-density canopies that intercept light, fiercely compete for nutrients, and even secrete allelopathic substances². Therefore, as the relative abundance of A. philoxeroides increases, its ability to compete for space and nutrients in the community becomes stronger, and the growth of most native plants is inhibited, resulting in a decrease in individual biomass¹². Atmospheric nitrogen deposition levels have increased significantly in recent years (cite source), with demonstrated impacts on plant communities in highly invaded ecosystems³⁶. Our findings reveal that the early-stage invasive L. peploides exhibited superior growth performance compared to A. philoxeroides under these conditions (Fig. 1a, b). This temporal advantage suggests that early detection and intervention may represent the most effective strategy for mitigating A. philoxeroides invasion, particularly in the context of rising nitrogen availability.

Variation in root traits

Plant roots, as critical organs for soil nutrient absorption, play a vital role in mediating plant-plant interactions and acquiring soil resources²⁷. Their morphological and architectural traits can also predict a plant’s ability to tolerate competitors⁵. In this experiment, A. philoxeroides exhibited a higher SRL (Fig. 2c, i). Compared to the CK, nitrogen application tended to reduce its RD, whereas low nitrogen had no effect on L. peploides RD, and high nitrogen even increased it (Fig. 2b, h). Generally, thinner roots (but with higher SRL) may possess narrower water-conducting vessels and faster turnover rates, enhancing soil exploration and exploitation efficiency^37,38. This trait reduces the cost of root proliferation in fluctuating environments, thereby improving nutrient absorption efficiency^37,38.

Differences in root trait changes were also observed between the two plant species at different invasion stages. In this experiment, compared to the native species migration period (100%), the presence of L. peploides had a minor effect on the root nitrogen content of A. philoxeroides under the CK treatment. However, under low nitrogen treatment, L. peploides significantly reduced the root nitrogen content of A. philoxeroides, while a slight increasing trend was observed under high nitrogen treatment (Fig. 2f). Conversely, compared to plots without A. philoxeroides invasion, the presence of A. philoxeroides tended to decrease the root nitrogen content of L. peploides under CK treatment, significantly increased it under low nitrogen treatment, and had a negligible effect under high nitrogen treatment (Fig. 2l). These patterns may be linked to changes in root exudate composition between the two species. Generally, higher root nutrient concentrations may correlate with greater abundance of soil-borne pathogens and root-feeding insects^39,40. Consequently, plants may allocate additional resources to defense structures, including the secretion of secondary metabolites into roots for chemical protection⁴¹. Among the secondary metabolites detected in root exudates, we observed that the presence of L. peploides tended to elevate the secretion of organic acids and amides by A. philoxeroides roots compared to the native species migration period (100%) (Fig. 3b). In contrast, the presence of A. philoxeroides had minimal effects on the composition of L. peploides root exudates (Fig. 3d). Given that organic acids and amides are known allelochemicals which can affect the growth of neighboring plants⁸. Our finding suggests a potential feedback mechanism: the presence of L. peploides induces A. philoxeroides to release such compounds, likely to negatively impacting the growth of the native species itself.

Further, physiological traits serve as sensitive indicators of plant environmental responses, typically detectable earlier than morphological traits⁴². Among these, non-structural carbohydrates (NSC, including soluble sugars and starch) availability profoundly influences plant growth and long-term survival^43,44,45,46. In this experiment, nitrogen application and the presence of L. peploides significantly increased starch and NSC content in A. philoxeroides roots compared to both the CK and native species migration period (100%) (Fig. 2d, e). In contrast, nitrogen fertilization significantly reduced starch and NSC content in L. peploides roots relative to CK. Furthermore, compared to plots without A. philoxeroides invasion, the invasive species’ presence showed a tendency to decrease starch and NSC content in L. peploides roots under CK treatment, with minimal effects under nitrogen addition (Fig. 2j, k). These findings suggest divergent resource allocation strategies: A. philoxeroides allocates substantial resources to root storage, while L. peploides prioritizes rapid growth.

Variation in leaf traits

Leaves, as the primary photosynthetic organs of plants, exhibit higher metabolic activity than roots and stems⁴⁴. Typically, increased leaf biomass exerts strong control over aboveground resource acquisition⁴⁷. Traits such as high SLA and elevated leaf nitrogen content were generally associated with enhanced photosynthetic capacity^25,37,48. Additionally, the mobilization of starch and NSC stored in leaves can increase leaf respiration rates, thereby meeting the elevated carbohydrate demands of maintenance respiration⁴⁴. Collectively, these adaptive traits improve a plant’s ability to absorb and utilize resources under changing environmental conditions, ultimately supporting greater aboveground growth. In this study, we found that nitrogen application enhanced the photosynthetic capacity of both study species compared to the CK, as evidenced by increased LMF, higher leaf nitrogen content, and reduced starch and NSC concentrations in leaves (Fig. 5). These results align with prior research demonstrating that nitrogen fertilization generally promotes leaf photosynthesis and plant growth⁴⁴.

We observed distinct patterns in leaf trait modifications of A. philoxeroides compared to L. peploides under varying invasion scenarios. Generally, elevated leaf nitrogen content alters plant interactions with nutrient-rich organisms and increases palatability to herbivores^39,40. In our experiments, the presence of L. peploides significantly reduced leaf nitrogen content in A. philoxeroides relative to the native species migration period (100%) (Fig. 5e). Conversely, A. philoxeroides invasion showed minimal effects on L. peploides leaf nitrogen content (Fig. 5j). These differential responses may reflect variations in leaf exudate composition between the species. Compared to the native species migration period (100%), L. peploides presence exerted limited effects on terpenoids, amides, phenols, and alkaloids (under CK and low nitrogen treatments) in A. philoxeroides leaves, while showing a tendency to reduce organic acids and significantly decreasing alkaloids content under high nitrogen treatment (Fig. 6b). Notably, A. philoxeroides invasion did not alter the exudate profile of L. peploides leaves (Fig. 6d). The observed reduction in leaf nitrogen content, coupled with limited allelochemical secretion, we hypothesize that an adaptive strategy in A. philoxeroides to minimize herbivory pressure.

The correlation between root and leaf functional traits and their adaptive strategies

Trait correlations were considered to reflect either trade-offs or synergistic optimization in resource allocation to meet fundamental survival requirements^38,49. Plants exhibiting stronger root-leaf trait coordination may demonstrate greater growth success and survival when exposed to environmental variability^30,50. This study revealed that under different nitrogen addition and invasion degree treatments, most root-leaf traits of both plant species exhibited significant correlations, though the strength and patterns of these correlations differed markedly. Compared to the native species L. peploides (involving 50 root-leaf trait correlations), A. philoxeroides demonstrated stronger root-leaf trait integration (involving 56 root-leaf trait correlations) (Fig. 8). This divergent trait correlation pattern may confer important adaptive value: it not only helps plants minimize negative impacts in unfavorable environments but also effectively enhances their capacity for survival, growth, and reproduction³⁰. The differences in trait correlation strategies between A. philoxeroides and L. peploides reflect the diverse manifestations of ecological trait plasticity during environmental adaptation among different species.

Results of the Partial Least Squares Path Modeling (PLS-PM) revealed that nitrogen addition and invasion degree can modulate the growth of A. philoxeroides through regulating root traits, leaf traits, and their interactions – a pattern not observed in L. peploides (Fig. 9a, b). Specifically, nitrogen addition and invasion degree not only directly increased the per-plant biomass of A. philoxeroides, but also indirectly promoted its biomass accumulation by altering leaf traits—such as increasing the leaf mass fraction and leaf nitrogen content, while reducing leaf starch content (Fig. 9a). In contrast, nitrogen addition had no significant effect on the per-plant biomass of L. peploides, whereas its biomass significantly decreased with increasing invasion degree. Furthermore, under conditions of nitrogen addition and invasion degree, the root traits, leaf traits, and their interrelationships in L. peploides did not exhibit significant regulatory effects on its total biomass (Fig. 9b). Previous research has indicated that trait plasticity and trait coordination play crucial roles in plant adaptation to environmental changes and may facilitate niche expansion³⁰. Our findings further corroborate that, compared to the native species L. peploides, A. philoxeroides exhibits greater phenotypic plasticity in response to environmental variation. This advantage was likely attributable to its more tightly coordinated root-leaf trait relationships. This enhanced trait integration may be a key mechanism underlying the ecological success of this invasive species in heterogeneous habitats.

In summary, a critical invasion mechanism of successful alien species lies in their superior trait values compared to co-occurring native species, enabling them to outcompete native flora and facilitate establishment in recipient habitats^8,28. Notably, variations in root and leaf traits result from complex interactions between multiple biotic and abiotic factors^26,51. Consequently, comprehensive quantification of interspecific differences in these traits is essential for elucidating the mechanisms underlying plant invasions.

In conclusion, we observed that A. philoxeroides and L. peploides displayed differing belowground and aboveground trait responses under varying invasion scenarios and nitrogen treatments. Nitrogen addition promoted growth in both species, whereas invasion pressure had a more pronounced negative effect on L. peploides. These species-specific patterns appear linked to differential adjustments in root and leaf trait expression. Overall, compared to the native plant L. peploides, the invasive plant A. philoxeroides had more advantages in root and leaf traits under environmental treatment, and the correlation between root and leaf traits was stronger. These findings suggest that under progressive atmospheric nitrogen deposition, A. philoxeroides may progressively displace L. peploides, particularly in communities experiencing mid-invasion (50%) to dominant invasion (75%) invasion degrees. However, the limited soil types and plant species used in this study constrain the generalizability of our findings. Future experimental designs should incorporate more naturalistic scenarios to enhance ecological relevance.

Study species

This study selected the invasive plant Alternanthera philoxeroides and the native plant Ludwigia peploides as research subjects. A. philoxeroides and L. peploides both exhibit rapid expansion capability through clonal growth and can quickly adapt to environmental changes by adjusting their above- and below-ground traits^46,52. In natural ecosystems, these two species often co-occur over broad geographical ranges, sharing similar habitat types such as rice paddies, wetlands, canals, ponds, and ditches⁵². A. philoxeroides, native to South America, was now widely distributed across many regions worldwide and has become one of the most aggressive invasive alien species in China, causing significant ecological and economic impacts in China and numerous other countries⁴⁶. In contrast, L. peploides is native to Zhejiang, Fujian, and eastern Guangdong in China and serves as a dominant native species in subtropical to tropical regions of the country⁵².

Experimental design

In April 2022, 300 seedlings of A. philoxeroides and 300 seedlings of L. peploides were collected at the Liangzihu national field ecological research station of Wuhan University (N30°05–30°18, E114°21–114°39). They were then cultured in a aquarium tanks (100 × 30 × 50 cm, L × W × H) filled with 30 cm deep lake sediment (TC, 31.22 mg·g⁻¹; TN, 4.09 mg·g⁻¹; TP, 2.27 mg·g⁻¹). These two types of clonal plants were grown under greenhouse conditions (The mean annual temperature is 25 °C with an average annual sunshine duration of 1,810 h) for one year.

On May 1, 2023, we selected 150 ramets each of A. philoxeroides and L. peploides from the pre-cultured seedlings, choosing individuals with biomass (approximately 1.9 g) and height (approximately 15 cm). These selected plants were then transplanted into stainless steel pots (70 cm inner diameter × 20 cm height) according to experimental treatments. Each pot was filled with approximately 15 cm of lake sediment (TC: 31.22 mg·g⁻¹; TN: 4.09 mg·g⁻¹; TP: 2.27 mg·g⁻¹). The experimental pots were randomly arranged on the outdoor cement platform at Liangzi Lake Ecological Station, which featured an open, flat terrain without obstructions and received ample sunlight.

Following established methodologies¹⁴, we employed a space-for-time substitution approach to simulate the progressive invasion process of A. philoxeroides. Five invasion scenarios were established: the total biomass of 4 plants per pot was strictly controlled within the range of 7.40–7.60 g. (1) no invasion period (0%): composed exclusively of 4 L. peploides seedlings; (2) early invasion period (25%): consisting of 1 A. philoxeroides and 3 L. peploides seedlings; (3) mid-invasion period (50%): containing 2 A. philoxeroides and 2 L. peploides seedlings; (4) dominant invasion period (75%): comprising 3 A. philoxeroides and 1 L. peploides seedling; (5) native species migration period (100%): represented by 4 A. philoxeroides seedlings. Throughout the experiment, any spontaneously occurring rare weeds in the pots were manually removed. Furthermore, according to previous research^1,8,16,23, each invasion treatment was coupled with three nitrogen addition levels: 0, 6, and 12 g N·m⁻²·yr⁻¹, designated as N0 (control, CK), N6 (low nitrogen), and N12 (high nitrogen) treatments, respectively. The N6 level represents the current average nitrogen deposition rate recorded in certain regions of China, while N12 corresponds to potential future high nitrogen deposition scenarios^1,16,23. Starting in May 7, 2023, artificial nitrogen addition (NH₄NO₃) to the pots on a weekly basis according to the designated nitrogen deposition levels. The experiment was conducted for 90 days, concluding on August 5, 2023. The complete experimental design was illustrated in Fig. 10. Although the space-for-time substitution approach has inherent limitations, this study implemented strict controls to ensure consistent total biomass and plant numbers per pot, which helps partially compensate for the constraints of short-term observations.

Sampling collect and trait measurement

In accordance with established methodologies, we collected a comprehensive dataset of above- and below-ground functional traits for both plant species. These trait parameters are widely recognized in ecological research as key indicators of plants’ acquisition, utilization, and conservation strategies for critical resources^{41,53,54,55,56,57,58}. Specifically, each stainless steel pot was fully submerged in water, and soil particles were removed under running tap water (water pressure: 0.2 MPa; duration: 5 min per sample). This process yielded intact plant specimens completely free of adhering soil particles.

Randomly selected 4 intact leaves from the uppermost canopy, and collected 3 intact root segments representing the complete root system architecture^8,31. Arranged samples on A4-sized acrylic trays with minimal overlap, scanned using a calibrated scanner (600 dpi, Epson 1680, Seiko Epson Corporation) following standardized calibration procedures, and analyzed using WinRHIZO software (Regent Instruments, Quebec, Canada) to determine leaf area (LA), average root diameter (RD), and total root length (RL)⁵⁶. The scanned leaves and roots were dried to a constant mass in a 70 ℃ oven, then weighed, and these data were used to determine the specific leaf area (SLA) and specific root length (SRL)^18,29,59.

After these measurements, the residual parts of the plant material were divided into three parts: roots, stems, and leaves. Then place it in a 70 ℃ oven to dry until a constant weight was reached, and weigh it. We separately measured the total biomass of individual plants of A. philoxeroides and L. peploides. The leaf mass fraction (LMF) was calculated as the ratio of the leaf mass to the total mass³²; The root mass fraction (RMF) was calculated as the ratio of the root mass to the total mass²⁸.

Collection and quantification of secondary metabolites in roots and leaves

Following previous research methods^60,61, randomly select 3–6 mature and intact leaves from harvested plants of weigh 1 g, and immediately grind them in a 10 ml sterile centrifuge tube. Add 5 ml of ethyl acetate for extraction and immerse for 72 h. Similarly, for collected belowground plant tissues and randomly select 2–5 intact root systems of weigh 1 g, and repeat the aforementioned steps. Preserve the obtained extracts in the dark at a low temperature (-20℃). Subsequently, filter the excess cellular debris using a 0.45 μm syringe filter, concentrate using a rotary evaporator (RE-52AA), dissolve in 1 ml of ethyl acetate, transfer to a GC-MS sample vial, and store in the dark at a low temperature (-40℃) until analysis⁶². The ethyl acetate extract of each sample was analysed by GC-MS (GCMS-QP 2020NX, SHIMADZU, Japan)⁶³. The GC injector temperature was 250 °C. The oven temperature was maintained at 45 °C, then increased from 45 °C to 150 °C at 10 °C/min, and then increased to 250 °C at 15 °C/min for 10 min. The transfer line temperature was set to 250 °C. Helium was used as the carrier gas at a flow rate of 1 mL/min. The MS source was operated in electron impact (EI) mode at 70 eV. The MS was scanned from 45 to 450 m/z. For the GC–MS data of root exudates, we normalized the peak area of each compound with the sum of all peak area for allidentified metabolites for each sample. Relative peak area was then used to calculate compound-specific concentrations^{8,60,61,62,63}.

Determination of non-structural carbohydrates and C, N, P elements

The total carbon and nitrogen concentrations in leaf and root plant tissues were analyzed using an organic elemental analyzer (Elementar, UNICUBE, Germany) via the dry combustion method. Total phosphorus content was measured using the molybdenum antimony anti-colorimetric method⁶⁴. For the extraction and determination of soluble sugars and starch, 80% anhydrous ethanol and ethyl anthraquinone acetate reagents were utilized, respectively⁴⁵. The sum of soluble sugar and starch content was considered the concentration of non-structural carbohydrates (NSC)⁶⁵.

Data analysis

Since the data from each pot were not completely independent, prior to statistical analysis, we calculated the mean values of measurements for each plant species within individual pots. This approach ensures the validity of our statistical analysis.

We employed two-way ANOVA to examine the effects of nitrogen addition, invasion degree, and their interaction on the total biomass, root traits, and leaf traits of A. philoxeroides and L. peploides. Means were compared using Duncan’s multiple range test. P-values ≤ 0.05 were considered statistically significant. Prior to analyses, we tested whether the assumptions of an ANOVA, homogeneity of variances and normally distributed residuals were achieved. The homogeneity of variances for all the studied parameters was evaluated by Levene’s test and the distribution of the residuals was assessed by Kolmogorov-Smirnov test. When necessary, logarithmic, reciprocal, or square root transformations were applied to meet assumptions.

Based on their chemical properties, plant tissue metabolites were classified into six categories: phenols, alkaloids, amides, alkanes, organic acids, and terpenoids. Each value represents the total concentration of all compounds within a given category. Principal Component Analysis (PCA) was performed to visualize differences in exudate composition between the two plant species under varying nitrogen levels and invasion degree. Adonis analysis was used to test for significant differences between the nitrogen level and invasion degree. At the same time, two-way ANOVA was performed for each compound using the “dplyr” and “agricolae” packages in R 4.2.3 (R, 2022) to analyze the significant differences in the relative concentrations of the compounds, with nitrogen addition and invasion degree as independent variables.

To examine the relationships between root/leaf functional traits and total biomass, we conducted Pearson correlation analyses separately for A. philoxeroides and L. peploides, assessing the associations between their respective root/leaf traits and total biomass. Furthermore, we quantified pairwise correlations between root and leaf traits under different nitrogen levels and invasion degree using Pearson’s correlation coefficients. This approach allowed us to investigate how these two species coordinate their root-leaf trait relationships in response to environmental changes.

To further investigate the potential relationships among plant biomass, root traits, and leaf traits under different environmental factors, we employed partial least squares path modeling (PLS-PM) to evaluate the direct and indirect effects of invasion degree and nitrogen level on root traits, leaf traits, and total biomass in A. philoxeroides and L. peploides. The model was constructed using the “innerplot” function from the “plspm” package in R software (4.2.3). Hypothesized pathways were defined a priori based on ecological theory, with invasion degree and nitrogen levels as exogenous variables, and root traits, leaf traits, and total biomass as endogenous variables. All variables were standardized (mean = 0, SD = 1) to ensure comparability of path coefficients. Non-normality was addressed using the package’s robust weighting algorithm.

All statistical analyses were performed using SPSS (SPSS Inc.) and R software (4.2.3), while figures were generated using Origin (Version 9.0, OriginLab Co.) and R software (4.2.3)⁶³.