Exploring plant adaptation strategies to phosphorus limitation induced by nitrogen addition: foliar phosphorus allocation and root functional traits analysis in two dominant subalpine tree species

Abstract

Plants adapt to the limitation of soil phosphorus (P) induced by nitrogen (N) deposition through a complex interaction of various root and leaf functional traits. In this study, a pot experiment was conducted to explore the effects of different levels of N addition (control, low N [LN]: 25 kg N ha⁻¹ yr⁻¹, high N [HN]: 50 kg N ha⁻¹ yr⁻¹) on tree growth, leaf nutrient content, foliar P fractions and root characteristics of two dominant tree species, the pioneer species Salix rehderiana Schneid and the climax species Abies fabri (Mast.) Craib, in a subalpine forest in southwestern China. The results demonstrated that LN addition had a minimal impact on leaf N and P contents. Conversely, HN addition significantly decreased the leaf P content in both species. Salix rehderiana exhibited more pronounced increases in specific root length and specific root area under P deficiency triggered by HN addition when compared with A. fabri. In contrast, A. fabri showed weaker morphological responses to N addition but had a higher proportion of foliar P to metabolic P, as well as higher root exudates rate and root phosphatase activity in response to HN addition. Abies fabri employs a synergistic approach by allocating a greater amount of leaf P to metabolite P and extracting P from the soil through P-mobilizing exudates and root phosphatase activity, while S. rehderiana exhibits higher flexibility in modifying its root morphology in response to P limitation induced by HN addition. This study provides insights into subalpine tree species adaptation to N-induced P limitation, emphasizing its significance for guiding forest management and conservation in the context of global climate change.

摘要

探究植物对氮添加诱导的磷缺乏的适应策略：两种亚高山优势树种叶片磷分配及根系功能性状分析

植物通过各种根系和叶片功能性状的复杂相互作用来适应氮(N)沉降诱导的土壤磷(P)限制。本文通过控制实验研究不同水平N添加对西南亚高山森林两种主要树种——先锋树种川滇柳(Salix rehderiana Schneid)和顶级群落树种峨眉冷杉(Abies fabri (Mast.) Craib)的生长、叶片养分含量、叶片P组分以及根系功能性状的影响。不同N水平处理包括对照、低N (25 kg N ha⁻¹ yr⁻¹)和高N (50 kg N ha⁻¹ yr⁻¹)。实验结果表明，低N添加对这两种树种的叶片N和P含量无显著影响，而高N添加下两种树种的叶片P含量均显著下降。与峨眉冷杉相比，川滇柳在面临由高N添加引发的P缺乏时，其比根长(SRL)和比根面积(SRA)的增加更为显著。相反，峨眉冷杉在N添加下根系形态变化不明显，但在高N添加下，其叶片P向代谢P的分配比例较高，并且其根系分泌碳含量与根系磷酸酶的活性显著升高。这些发现表明，两种树种在适应高N添加引起的P缺乏时采取了不同的适应策略。峨眉冷杉通过地上-地下协同作用，优化了叶片P向代谢P的分配，同时借助磷活化分泌物和根系磷酸酶的活性从土壤中获取P，而川滇柳则通过调整其根系形态来应对高N添加导致的P缺乏。本研究深化了对亚高山树种如何适应N添加诱导的P缺乏现象的认知，同时也为科学指导森林管理和保护工作提供了重要依据。

INTRODUCTION

Since the Industrial Revolution, human activities such as the use of mineral fertilizers and biomass combustion have led to a rapid increase in the input of nitrogen (N) and phosphorus (P) into the biosphere. The deposition of N and P reached 220 and 12 Tg a⁻¹, respectively, in the year 2000 (Penuelas et al. 2013), and it is expected to continue to rise in the future. The continuous and imbalanced inputs of N and P can lead to significant stoichiometric imbalances in ecosystems over time, potentially shifting terrestrial ecosystems from N to P limitation (Goswami et al. 2017). Moreover, N addition usually stimulates plant growth, inevitably increasing the demand for P (Dong et al. 2022). In addition, continued input of exogenous N typically results in soil acidification (Cai et al. 2021; Zhang et al. 2023), which could further aggravate P limitation (Li et al. 2016; Vitousek et al. 2010). While there is growing evidence of N enrichment-induced P limitation in various terrestrial ecosystems (Yin et al. 2021; Zhang et al. 2022), it remains unclear how plants can adapt to such P deficiency, which hinders our ability to predict vegetation dynamics following future N enrichment (Terrer et al. 2020).

Plants employ different strategies to acquire P from the soil, including morphological and physiological adaptations. Common root pathways to enhance P foraging involve allocating more carbon to absorptive fine roots, possessing a higher specific root length (SRL), having thinner roots and increasing the frequency and length of root hairs (Eissenstat et al. 2015). In addition to these morphological adaptations, plants can also employ physiological strategies or P-mining strategies to enhance P mobilization, such as releasing specific compounds, including carboxylates, phosphatases and protons, into the rhizosphere to extract P from insoluble sources (Wen et al. 2022). However, the response of tree growth to N deposition shows inconsistent trends, including positive effects (Netzer et al. 2019), negative effects (Penuelas et al. 2013) and insignificant alterations (Schwede et al. 2018). Some studies have also shown contrasting effects of N inputs on plant root dynamics (Li et al. 2015; Wang et al. 2019a). These contradictory findings may be attributed to (i) the inherent nutrient use strategies of different plant species (Baez and Homeier 2018; Bardgett et al. 2014), (ii) the form, amount and duration of N inputs (Tian et al. 2016) and (iii) abiotic environmental conditions (Schulte-Uebbing and de Vries 2018). These inconsistencies suggest that the impact of N addition on fine roots remains uncertain and requires further research to enhance our understanding of resource acquisition strategies (Peng et al. 2017; Wang et al. 2019b).

Leaves are considered to be the most sensitive organs when it comes to the availability of plant nutrients (Hidaka and Kitayama 2011). Shifting the allocation of P in leaves is a crucial mechanism for plants to adapt to changes in soil P availability (Hidaka and Kitayama 2011). The fractionation of foliar P can be classified into inorganic phosphate (Pi) and organic P fractions, such as metabolite P, structural P, nucleic acid P and residual P. Pi is a significant portion of leaf P and is typically stored in the vacuole. The metabolite P fraction consists mainly of carbon metabolism intermediates, such as phosphorylated sugars, adenosine -diphosphate (ADP) and adenosine-triphosphate (ATP). Nucleic acid P is the main component of organic P in plants, constituting more than 50% of the foliar organic P pool. Within nucleic acid P, ribosomal Ribonucleic Acid (rRNA) makes up as much as 85% of the total. The main component of structural P is phospholipids, which are primarily located in the plasmalemma and organelle membranes. The residual fraction is uncharacterized and may include phosphorylated proteins that play a role in regulating cellular processes. Soil properties have a greater influence on the various forms of foliar P than on the overall foliar P content, as stated by Veneklaas et al. (2012). Therefore, the fractions of P in leaves can serve as an indicator of how plants adapt to changes in nutrient availability. The allocation of P in leaves is prioritized to cope with limited P availability by decreasing the allocation of P to non-metabolite P fractions such as nucleic acid and lipid P. This helps to mitigate the direct impact of P restriction on photosynthesis (Zhang et al. 2021). Nevertheless, there have been relatively few studies examining the effect of N addition on foliar P fractions (Lei et al. 2021; Yin et al. 2021). Furthermore, the coordination or trade-offs between root morphology, P-mobilizing exudates and leaf P allocation in influencing P acquisition across different species or in response to N addition are still not well understood (Yin et al. 2021). This lack of understanding hinders our ability to accurately predict vegetation dynamics in response to N enrichment.

The alpine coniferous forest in Southwest China plays a crucial role in the ecological protection of the upper reaches of the Yangtze River. It is the second-largest forest area in China and is particularly susceptible to N deposition. Forest ecosystems in this region have a relatively low critical threshold of N deposition load (50 kg hm⁻²) (Yu et al. 2019a). This makes the region an ideal natural experiment to study the effects of N deposition on forest tree growth and nutrient limitation. Salix rehderiana Schneid, an early-successional species, and Abies fabri Mast. Craib, a late-successional species, are both extensively distributed in the subalpine regions of Southwest China, exerting significant ecological influences on local forest ecosystems. Salix rehderiana is characterized by greater SRL and thinner root diameters compared with the late-successional species A. fabri (Song et al. 2017; Yu et al. 2019b). Abies fabri demonstrates a conservative nutrient utilization strategy, whereas S. rehderiana shows an acquisitive approach to nutrient acquisition. However, there is limited empirical evidence regarding the performance and plasticity differences between S. rehderiana and A. fabri in response to N addition.

This study aims to explore the variations in root morphology and physiology displayed by these two tree species in the context of N supplementation. Additionally, we aim to investigate whether these species employ distinct strategies for P acquisition from the soil under N addition conditions. Based on prior research, our hypotheses are: (i) Different levels of N addition may induce P deficiency in plants to some extent, as increased soil N availability typically promotes plant growth, thereby increasing plant demand for P in N-limited forests (Li et al. 2016); (ii) S. rehderiana and A. fabri seedlings exhibit distinctive aboveground and belowground P acquisition strategies to alleviate P limitations induced by N addition, reflecting their specific physiological and ecological characteristics. The findings of this study will establish a theoretical basis for the precise management of nutritional requirements for S. rehderiana and A. fabri seedlings and their plantations in the context of N deposition.

MATERIALS AND METHODS

Study site and experimental design

This study took place at the Gongga Mountain Alpine Ecosystem Observation and Experiment Station (29°34ʹ N, 101°59ʹ E), positioned on the southeastern edge of the Tibetan Plateau, with an elevation of 3000 m above sea level. The station experiences an annual mean temperature of 4.2 °C, annual precipitation of 1947 mm and relative humidity of 90.2%. The region where premature glacier retreat occurs is situated around 2 km from the station, and the climatic conditions in that area are akin to those at the station.

A total of 96 seedlings, comprising 48 S. rehderiana and 48 A. fabri, were carefully selected. Uniform-sized seedlings (1–3 years old) with approximately the same tree height (15–30 cm) were obtained from a nearby nursery. They were planted individually in plastic pots with a diameter of 30 cm and a height of 25 cm. The pots were filled with soil obtained from the nearby forests, which had a soil organic matter content of 28.0 ± 1.1 g kg⁻¹, total N content of 0.52 ± 0.04 g kg⁻¹ and total P content of 0.26 ± 0.02 g kg⁻¹. The experimental design was completely randomized and included two factors: two species (A. fabri and S. rehderiana) and three levels of N supply: control, 25 kg N ha⁻¹ yr⁻¹ (low N, LN) and 50 kg N ha⁻¹ yr⁻¹ (high N, HN). The N addition levels were determined based on the measured ambient atmospheric N deposition rate of 26.45 kg N ha⁻¹ yr⁻¹ reported by Zhu et al. (2015) and the critical value of N deposition load in this region, which is 50 kg N ha⁻¹ yr⁻¹, as stated by Yu et al. (2019a). Based on the desired N fertilization levels, N was applied at 0, 3.47 and 6.94 g N yr⁻¹ for each pot, respectively. Each treatment had four replicates, with each replicate comprising four pots, resulting in a total of 96 pots. The selected N fertilization level was based on the local N deposition rate and the widely used method to double the local deposition rate to simulate additional N deposition. For the control treatment, water was sprayed over the pots instead of the N solution. The experiment was conducted in an open field site and was exposed to natural rain conditions. NH₄NO₃ was used as the N source for the fertilized treatments. The pots were watered every 4 days to maintain the expected soil moisture at a constant level of 60% water saturation. All pots received the same amount of water, and the same routine management practices were followed during experiment time. The N treatments were performed from 10 May to 10 October every year from 2016 to 2018. To implement N addition, a solution of NH₄NO₃ was uniformly sprayed onto the pots monthly throughout the growing season, from May to October. Measurements of various morphological, physiological and biochemical parameters were done at the end of the experiments.

Soil characters

Following root shaking, the soil tightly adhered to the roots was removed by brushing, and collected as rhizosphere soils (Phillips and Fahey 2006). The pH of the rhizosphere soils was measured using a pH meter. Soil available N (AN) and available P (AP) were measured using a continuous flow auto-analyzer (Smartchem 200, Amassurance, Italy) after extracting the soil with 2 mol L⁻¹ KCl for AN and 0.5 mol L⁻¹ NaHCO₃ for AP.

Determination of leaf N, P concentration and leaf acid phosphatase activity

Dried leaf samples were finely ground into powder and filtered through a mesh with a pore diameter of approximately 275 µm. The semi-micro Kjeldahl method (Mitchell 1998) was used to determine the concentrations of N in the leaves. To determine the concentration of P, persulfate oxidation followed by colorimetric analysis was employed (Schade et al. 2003). Determination of leaf acid phosphatase: grind 0.2 g of fresh leaves and subsequently add 5 mL of an ice-cold acetic acid–sodium acetate buffer solution with a molarity of 0.20 mol L⁻¹ to the resulting mixture. The grinding solution was centrifuged at a low temperature for 10 min. 0.05 mL of the supernatant enzyme solution, 0.45 mL of the buffer and 4.50 mL of 5 mmol L⁻¹p-nitrophenyl phosphate were mixed and incubated in a dark water bath at 30 °C for 30 min. The reaction was terminated by adding 2 mL of 2 mol L⁻¹ sodium hydroxide. The absorbance was measured at 405 nm, and the blank was adjusted to zero using a no-enzyme reaction (Ushio et al. 2015).

Foliar P fraction measurements

Foliar P can be categorized into inorganic P (Pi) and organic P, which includes sugar P, lipid P, nucleic acid P and residual P. In the study conducted by Hidaka and Kitayama (2013), the organic P fractions were extracted sequentially. The acetic acid extraction method (Yan et al. 2019) was used to extract foliar Pi, which was then measured using a molybdenum blue-based method developed by Ames (1966). To determine the four fractions of organic P in a freeze-dried and ground foliar sample, a 0.5-g foliar sample was mixed with 15 mL of a CMF (chloroform:methanol:formic acid = 12:6:1) solution in a 50-mL centrifuge tube. The liquid was then extracted twice using a total of 19 mL of a CMW (chloroform:methanol:water = 1:2:0.8) solution, with 9.5 mL of chloroform-washed water added to each extract. The final solvent used was a CMW solution, which separated the extract into an upper layer rich in sugars and nutrients, and a bottom layer rich in lipids. The upper layer was transferred to a new tube to determine the sugar and nutrient P fraction. The bottom layer was used to determine the lipid P fraction. In the new tube, a 5-mL volume of methanol solution was added to the material and placed in a vacuum dryer for 48 h to remove dissolved chloroform and methanol. The aqueous layer was then refrigerated at 4 °C for 1 h, and a trichloroacetic acid (TCA) solution was prepared. A cold TCA solution was added to the tube and after 1 h, the material was shaken and then centrifuged. The supernatant was prepared for the determination of the sum of Pi and metabolite P. Pi was subtracted from the sum to obtain the metabolite P fraction. Finally, the remaining residue after extraction with TCA was mixed with a TCA solution and extracted in a hot water bath. Aliquots were then centrifuged and taken for analysis of nucleic acid P. The residue remaining from the hot TCA final extraction was the residual P fraction. The determination method for all foliar P fractions was similar to that of foliar total P, and the quantities of the fractions were expressed on a dry mass basis.

Root trait measurements

The roots were carefully extracted from the soil and cleaned using deionized water. Our samples were comprised exclusively of first to third-order roots, as classified by Pregitzer et al. (2002). Consequently, we categorized these roots based on their role as absorptive fine roots, directly facilitating nutrient uptake (McCormack et al. 2015). The roots were then scanned using a desktop scanner at a resolution of 400 dpi (Epson Expression1600, Japan). The scanned images were analyzed using WinRHIZO Pro root analysis software (Regent Instruments Inc., Quebec, Canada) to determine the root diameter, length, surface area and volume. After the scanning process, the harvested roots were dried at 75 °C for 72 h. The dried roots were then weighed to calculate SRL (root length divided by dry mass), specific root area (SRA, root area divided by dry mass) and root mass density (RMD, dry mass divided by root volume in g cm⁻³). The activity of root phosphatase (APase) was determined using the methodology described by Ishidzuka (1999). To assess root APase activity, the hydrolysis of p-nitrophenyl phosphatase (p-NPP) to p-nitrophenol (p-NP) was measured after incubating roots and soil at 37 °C for 1 h. Phosphatase activity was calculated as the amount of p-NPP cleaved per gram of root per hour, after adjusting for control and sorption effects.

Root exudation collection

We used a modified in-situ static collection device based on the method described by Phillips et al. (2008) to collect root exudates. First, the fine roots were carefully cleaned and placed in tubes containing acid-washed glass beads. These tubes also contained a carbon-free nutrient solution consisting of 0.1 mmol L^-1 KH₂PO₄, 0.2 mmol L^-1 K₂SO₄, 0.2 mmol L^-1 MgSO₄ and 0.3 mmol L^-1 CaCl₂. To protect the roots from sunlight and excessive heat, the tubes were wrapped with aluminum foil. To ensure that the root exudates were collected under field-like conditions, the roots were incubated in the tubes for 24 h before the nutrient solutions were collected. The collected nutrient solutions were then analyzed for the total organic carbon of the root exudates using a total organic C/total N analyzer (Multi N/C 2100, Analytik Jena AG, Jena, Germany). After the collection of root exudates, the roots from the tubes were also collected to determine root morphological characteristics, such as root mass and length. Finally, the root carbon exudation rate was calculated by dividing the total organic carbon content of the root exudates by both the root mass and the root length.

Statistical analysis

Before conducting the analysis of variance (ANOVA), we checked the data to ensure that they met the assumptions of normality and homogeneity of variance. We used a general linear model (GLM) procedure with type III sum of squares to analyze the effects of species, N addition and their interactions on root functional traits and plant growth parameters through a two-way ANOVA. To determine significant differences between treatments, we conducted Tukey’s post-hoc test with a significance level of 0.05. The analyses were carried out using the general linear ANOVA model (GLM) in SPSS 19.5 (SPSS Inc., Chicago, IL). To examine the multivariate ordination of the sixteen functional traits for the two species, we conducted a principal component analysis (PCA) using the ‘prcomp’ function in R. This analysis allowed us to explore the relationships and patterns among the different functional traits.

RESULTS

Soil chemical traits

The results showed that the LN treatment did not have a significant effect on pH, while the pH decreased under HN treatment (Fig. 1a). Furthermore, as the level of N addition increased, the AN content showed a significant increase, while the AP content decreased (Fig. 1b and c). In terms of the activity of root APase, the LN treatments had no significant effect on it, but the HN treatment increased the activity of root APase in A. fabri (Fig. 1d).

Leaf chemical traits and leaf P allocation

The addition of N had no significant effect on the leaf N content of either species (P > 0.05, Fig. 2a). However, the HN treatment significantly decreased the leaf P content in S. rehderiana and A. fabri by 37.6% and 32.30%, respectively, compared with the control (P < 0.05, Fig. 2b). The HN treatment significantly increased the N:P ratio by 58.1% in the leaves of S. rehderiana (Fig. 2c). Furthermore, the HN treatment significantly increased the activity of acid phosphatase in the leaves of A. fabri by 56.8%, compared with the control (P < 0.05, Fig. 2d).

N addition did not have significant effects on metabolic P, nucleic acid P, structural P and residual P in S. rehderiana (Fig. 3a–d). However, in A. fabri, the HN treatment led to an increase in leaf metabolic P (Fig. 3d). Additionally, the HN treatment resulted in a decrease in leaf nucleic acid P and structural P in A. fabri (Fig. 3a and b). There was an increase in residual P in both species under the HN treatment, but it was not statistically significant (Fig. 3c). The HN treatment resulted in a significant reduction in Pi levels in both species (Fig. 3e).

Root characteristics

In S. rehderiana, the growth relative to initial biomass (GRIB) increased in the LN treatment. However, GRIB decreased in the HN treatment in A. fabri (Fig. 4a). The HN treatment significantly increased root:shoot ratio (RS) in S. rehderiana but had no effects on A. fabri (Fig. 4b). Salix rehderiana had a lower average root diameter ranging from 0.19 to 0.50 mm, while A. fabri tended to have higher average root diameter ranging from 0.50 to 0.83 mm. The HN treatment significantly increased SRL and SRA (P < 0.01) but decreased the average root diameter in S. rehderiana (Fig. 4c–e). Compared with the control, the HN treatment tended to decrease RMD in S. rehderiana (Fig. 4f). RS, root diameter, SRL, SRA and RMD showed little response to N addition in A. fabri. There were significant species variations in the root exudation rate in response to N addition. With increasing soil N availability, the carbon exudation rate per biomass and the carbon exudation rate per length in A. fabri increased significantly under HN treatment (Fig. 4g and h). N addition had little effect on rhizosheath carboxylates in S. rehderiana. Furthermore, compared with the control, HN treatment increased the root exudation rate per mass by 53.3%, and the root exudation rate per length by 66.5% in A. fabri.

Multivariate coordination

PCA analysis based on 16 plant traits showed clear differentiation between A. fabri and S. rehderiana in all N treatments (Fig. 5), indicating contrasting responses in root functional traits to varying soil N availability. In the N addition treatment, the first two trait axes of the PCA explained 33.3% and 31.6% of the total variation, respectively. The two carbon exudate traits of the root had high scores on the first axis, while the root morphology-related parameters had high scores on the second axis (Fig. 5). Salix rehderiana exhibited a distribution along the axis of root morphological parameters, while A. fabri was mainly distributed along the axis of root exudates, root APase activity and leaf P fraction (Fig. 5).

DISCUSSION

HN addition, but not LN addition, induces P limitation in S. rehderiana and A. fabri

Leaf nutrient concentration and stoichiometric ratios are important factors for predicting and determining nutrient limitation in tree growth. In this study, N addition did not have a significant impact on leaf N levels. This could be ascribed to plant acclimatization to N enrichment in the soil, leading to a deficiency in additional N absorption from the soil and thus no significant modification in leaf N concentrations (Lu et al. 2018). In this study, HN addition reduced leaf P content in both S. rehderiana and A. fabri (Fig. 2), indicating insufficient P supply to meet plant growth needs (Gusewell 2004). In addition, HN addition resulted in a significant rise in N:P ratio in both species to differing extents. Specifically, the N:P ratio of leaves of S. rehderiana increased from 12.7 under controlled conditions to 20.1 following the application of high levels of N, whereas the N:P ratio of A. fabri needles rose from 14.3 under controlled conditions to 16.8 (although not significantly). The results suggest that the introduction of N plays a crucial role in driving the observed P limitation trend in both species within the region, as explained by Zhang et al. (2022), which supports our Hypothesis i. Our findings, which demonstrate P deficiency induced by N enrichment, align with earlier N addition studies (Yin et al. 2021). Nevertheless, caution is required when extrapolating the findings from these studies to the highly complex field conditions, as demonstrated in our pot experiment. More experiments are needed to confirm whether N triggers P limitation in field conditions. The N-induced P limitation could be attributed to various factors, such as increased growth outpacing nutrient accumulation, low natural availability of P in the soil, N addition exacerbating soil acidification and reducing AP (Yin et al. 2021; Zhang et al. 2022). Moreover, the activity of leaf acid phosphatase is essential for regulating P cycling and indicating the severity of P limitation. Alterations in leaf acid phosphatase activity represent a crucial physiological mechanism through which plants respond to P limitation (Yin et al. 2021; Zhang et al. 2022). The findings of this study demonstrated a significant increase in leaf acid phosphatase activity in A. fabri seedlings when exposed to HN levels, leading to improved acquisition of AP. Specifically, under HN supplementation, plant acid phosphatase activity rises, facilitating the conversion of organic P into inorganic P through hydrolysis, thereby enhancing the efficiency of plant utilization of accessible P.

In this study, A. fabri demonstrated strong regulation of its N:P ratio while S. rehderiana exhibited more flexibility in its N:P stoichiometry. Previous studies have shown that species with high levels of stoichiometric N:P homeostasis are more resistant to environmental changes (Yu et al. 2010). However, our findings suggest that S. rehderiana had increased resistance to P deficiency due to its higher degree of stoichiometric N:P flexibility, compared with A. fabri. When soil P levels are low, it becomes energetically demanding for plants to maintain a constant N:P ratio as nutrient uptake becomes more challenging (Sardans and Penuelas 2012). It is possible that S. rehderiana benefited from the increase in soil AN and was able to take up more N due to its greater homeostatic flexibility. This could explain that S. rehderiana was able to increase its biomass even under N addition-induced P limitation. On the other hand, A. fabri was more sensitive to the P limitation caused by N addition, leading to a decline in growth. Abies fabri invests a considerable amount of energy in taking up low accessible P at the expense of plant growth to maintain a less flexibility of N:P ratio. These results emphasized that a greater ability to maintain N:P homeostasis may become inefficient when access to soil P becomes limited, as the energy required to maintain stoichiometric homeostasis is costly for plants (Schindler 2003). In contrast, S. rehderiana can tolerate wider variations in tissue N:P ratio (i.e. stoichiometric flexibility) and may allocate more carbon to root growth to improve N uptake.

Effects of HN addition on foliar P allocation and adaptation mechanisms

In our study, the reduction in foliar P concentration in A. fabri under HN addition was found to be correlated with a decrease in the concentrations of leaf Pi, structural P and nucleic acid P. According to Veneklaas et al. (2012), the level of Pi in leaves can be used as a measure of soil inorganic P accessibility. Under a low P environment, reduced P investment in the lipid P fraction through lipid remodeling and/or replacement by lipids has been demonstrated as an important adaptation mechanism for plants to decrease foliar P demand (Prodhan et al. 2019). The leaf metabolic P increased while nucleic acid P decreased in A. fabri under HN addition, indicating that the species acclimated to N addition-induced P limitation by reducing their allocation of P to non-metabolite foliar P fractions (Zhang et al. 2021). The primary form of nucleic acid P is predominantly found in ribonucleic acid (RNA), particularly in rRNA (Lambers et al. 2011). Therefore, less P was allocated to rRNA to restrain protein synthesis for growth and photosynthesis (Zhang et al. 2021). Residual P is an insoluble P compound primarily comprised of phosphorylated proteins, which plays essential roles not only in P uptake, distribution and remobilization, but also in many metabolic processes, such as carbohydrate metabolism, phytohormones and signal transduction (Hidaka and Kitayama 2011). Therefore, in this study, the tendency of an increase in the residual P in both species under N addition may be associated with improved plant adaptation to P deficiency.

Divergent root traits and foliar nutrient allocation strategies of S. rehderiana and A. fabri in response to N addition

Our research reveals significant variation in root functional traits within species. In comparison to A. fabri, S. rehderiana exhibited thinner roots, lower RS and RMD and higher SRL and SRA across all treatments. Salix rehderiana displayed thinner roots with lower RMD, indicating a lower construction cost per unit root length, facilitating the more efficient exploration of a larger soil volume (Wen et al. 2020). On the other hand, A. fabri exhibited a high root diameter and higher RMD, suggesting a conservative strategy. Moreover, A. fabri showed greater amounts of carboxylates and higher activity of root APase. These findings indicated that there were two distinct plant strategies between the two species. We present a framework illustrating the distinct strategies employed by A. fabri and S. rehderiana to alleviate N addition-induced P deficiency (Fig. 6). Salix rehderiana exhibits a preference for modifying root growth to enhance soil exploration efficiency in response to limited P resources, which is considered a ‘P-scavenging’ strategy relying less on P-mobilizing exudates (Wen et al. 2020). On the other hand, A. fabri exhibited greater plasticity in P-mobilizing exudates and root phosphatase activities compared with root morphology, which expresses a typical ‘P-mining strategy’ that increases soil P availability by mobilizing P through processes such as ligand exchange or chelation of cations bound to precipitated phosphates (Lambers et al. 2015). Furthermore, A. fabri exhibited greater plasticity in leaf P allocation. These findings suggest that A. fabri employs a synergistic strategy of aboveground leaf P allocation and belowground P mining to alleviate P deficiency. This represents a complementary approach to enhance P exploration in the soil (Yin et al. 2021). The distinct patterns of foliar nutrient allocation and root traits in these two species suggest that they have evolved different adaptive strategies in response to soil P availability, which aligns with their differences in growth strategies. These findings explain the distinct responses of S. rehderiana and A. fabri to N addition. Given the wide distribution and diverse ecological functions of various mycorrhizas, mycorrhizal type is an essential factor for further exploring the response of plant growth to N addition. Future studies should evaluate the reasons behind the species-specific response to N addition caused by different mycorrhizal types through isolating and identifying the mycorrhizal type.

CONCLUSIONS

This study provides evidence that HN addition leads to P limitation in the growth of subalpine wood species. Salix rehderiana and A. fabri employ different adaptive strategies in response to P limitation induced by HN addition, with A. fabri employing a synergistic strategy of leaf P allocation and belowground P mining to alleviate P deficiency, while S. rehderiana displayed greater plasticity in terms of root morphology to explore soil P. These findings have significant implications for forest management and restoration, especially in nutrient-poor ecosystems.

Funding

The research was supported by Open Foundation of the Key Laboratory of Natural Resource Coupling Process and Effects (2023KFKTA005, 2023KFKTB012) and by the Science and Technology Research Program of Institute of Mountain Hazards and Environment, Chinese Academy of Sciences (IMHE-ZDRW-06).

Authors’ Contributions

B.L.D., Y.S. and Y.F.T. designed the experiments; Y.S. and Y.F.T. conducted most of the experiments; Y.H. and M.Y.L. performed data and statistical analyses; X.Y.L. provided study guidance and interpretation of data and experimental results; Y.S. and Y.F.T. prepared the figures; B.L.D. and X.Y.L. co-wrote and edited the manuscript.

Conflict of interest statement. The authors declare that they have no conflict of interest.

REFERENCES

Author notes