Ectomycorrhizal (ECM) functional traits related to nutrient acquisition are impacted by nitrogen (N) deposition. However, less is known about whether these nutrient-acquisition traits associated with roots and hyphae differentially respond to increased N deposition in ECM-dominated forests with different initial N status. We conducted a chronic N addition experiment (25 kg N ha-1 yr-1) in two ECM-dominated forests with contrasting initial N status, i.e. a Pinus armandii forest (with relatively low N availability) and a Picea asperata forest (with relatively high N availability), to assess nutrient-mining and -foraging strategies associated with roots and hyphae under N addition. We show that nutrient-acquisition strategies of roots and hyphae differently respond to increased N addition. Root nutrient-acquisition strategies showed a consistent response to N addition, regardless of initial forest nutrient status, shifting from organic N mining toward inorganic N foraging. In contrast, the hyphal nutrient-acquisition strategy showed diverse responses to N addition depending on initial forest N status. In the Pinus armandii forest, trees increased belowground carbon (C) allocation to ECM fungi thus enhancing hyphal N-mining capacity under increased N availability. By comparison, in the Picea asperata forest, ECM fungi enhanced both capacities of P foraging and P mining in response to N-induced P limitation. In conclusion, our results demonstrate that ECM fungal hyphae exhibit greater plasticity in nutrient-mining and -foraging strategies than roots do in response to changes in nutrient status induced by N deposition. This study highlights the importance of ECM associations in tree acclimation and forest function stability under changing environments.

Study sites and the N-addition manipulation

The study sites were located at Tudiling Giant Panda National Park nearby the Maoxian Ecological Station of the Chinese Academic of Sciences (31º41´N, 103º53´E) in Maoxian country, Sichuan Province, China. According to the long-term meteorological monitoring data of the station, the mean annual temperature, precipitation, and evaporation are 8.9℃, 920mm, and 796 mm, respectively. A chronic N-addition experiment was conducted in two adjacent ECM-dominated coniferous forests, a Pinus armandii forest and a Picea asperata forest, to disentangle the effects of chronic N deposition on the structure and function of these forests. These two forests were planted in the 1970s under the auspices of a local reforestation program. The soils in both forests are classified as a Cambic Umbrisol. Compared with the Pinus armandii forest, the Picea asperata forest has higher concentrations of soil organic C, total N, dissolved inorganic N (DIN, NH₄⁺+NO₃^-), DIN: plant-available soil phosphorus (Av. P) ratio and net N-mineralization rate but lower Av. P concentration (p < 0.05, Table S1). These results suggest that the Pinus armandii forest has relatively low N availability, but high soil P availability compared to the Picea asperata forest, providing a chance to whether nutrient-acquisition strategies of roots and hyphae respond differently to N addition due to changes in forest N and P status.

In April 2017, three replicated blocks of two N-treatment conditions (the control with 0 kg N ha^-1 yr^-1 and N addition with 25 kg N ha^-1 yr^-1) were randomly established in each plantation, and each block was at least 20 m apart. Two 10 m × 10 m plots separated by 10-m wide buffer strips were established in each block. The ammonium nitrate (NH₄NO₃) was divided into 6 doses and sprayed to each plot monthly from May to October (i.e. 41.6 g N per month) in each year. The N-fertilizer application has been going on since 2017. 

Ingrowth core installation and collection

To partition the C inputs from roots and hyphae, three types of ingrowth cores (inner diameter 6 cm, height 15 cm) were set up in each plot in May 2017 as per Phillips et al (2012) and Zhang et al (2018): one with 2-mm mesh allowing the penetration of roots and hyphae (R-cores), one with a 48-μm mesh only permitting the penetration of hyphae (H-cores), and one with a 1-μm mesh that prevents the ingrowth of roots and hyphae (EH-cores). The natural abundance of ¹³C has been widely used to estimate changes in soil C stock. Commonly, C₃ plants are grown in soils with organic matter derived from C₄ plants (Kuzyakov & Domanski, 2000; Wallander et al., 2011; Keller et al., 2021). Six sets of ingrowth-cores with 2-mm, 48-µm and 1-µm mesh-size were filled with homogeneous maize-grown soil and installed in the topmost mineral horizon in each plot. These cores were used to quantify the amount of root- and hypha-derived C input. The C isotopic signature (δ¹³C) in the maize-grown soil was -24.29 ± 0.04‰, which was significantly enriched in ¹³C compared with the C₃ roots (-27.14 ~ -27.85%) and hyphae (-26.79 ~ -28.55%) in two forests. Additionally, six sets of ingrowth-cores of different mesh sizes were filled with homogeneous native mineral soil (0-15 cm) collected from each plot in each forest and installed in the same soil horizon to assess the effects of roots and hyphae on N- and P-acquisition enzyme activities. Before filling the cores, all C₄ and native soils were sieved through a 5-mm mesh after removing visible roots. Each ingrowth core was at least 20cm apart. As the top of the cores were tied tightly (preventing entry of fresh litter), the C source in the R-cores was mainly derived from roots, hyphae, and litter leachates, and that of the H-cores was from hyphae and litter leachates, while the EH-cores received C only from litter leachates. To block the entry of new C derived from saprophytic hyphae outside the cores, we spread a 2-mm-thick layer of silica sand (0.36–2.0 mm, 99.6% silicon dioxide (SiO₂) around the cores. In a preliminary experiment, we found that more than 80% of the fungal taxa entering silica-sand-blocked cores from outside belong to ectomycorrhizal fungi (relative abundance at trophic-Mode level, % ), while saprophytic mycorrhizal fungi only account for 4-15% (Unpublished data). This result again confirmed that silica sand can effectively block the entry of external saprophytic fungal hyphae in a previous study (Hagenbo et al., 2017). In addition, we used mesh-cores (inner diameter 6 cm, height 15 cm, 48 µm mesh size) containing 135 g acid-washed silica sand to harvest and quantify hyphal biomass (Wallander et al. 2001; Guo et al., 2021).

We harvested ingrowth cores in August 2019 and August 2020, respectively. Two sets of cores filled with C₄ soil and native soil were randomly collected from each plot at each sampling date. In total, four sets of cores filled with C₄ soil and native soil were collected during the experimental period, and the remaining in-growth cores in the plots were reserved for a follow-up study. Cores were transported to the laboratory using an ice box. The C₄ and native soil inside the cores with the same size of mesh in each plot were sieved through a 2-mm mesh and thoroughly mixed to form a composite sample for analysis, respectively. The total number of soil samples per treatment was six for C₄ soils and six for native soils for two sampling dates (2019 and 2020) in each forest. The C₄ soils were air-dried for the determination of soil organic C (SOC) and δ¹³C. Subsamples of the native soils inside the cores were stored at 4 °C for the analyses of soil nutrient availability and extracellular enzyme activities within 48 h of sampling.

Nutrient-foraging traits of roots and hyphae

Fine root/hyphal biomass, root morphology, and the relative abundance of ECM fungi with different hyphal exploration types were determined to access the changes in nutrient-foraging strategies of roots and hyphae under N addition. Fine roots inside the 2000-µm mesh cores were manually picked out and washed thoroughly, and then scanned at 600 dpi with images analyzed using WinRHIZO (Regent Instruments, Inc., Quebec, Canada) to record root diameter, accumulative root length and root-branching numbers. After scanning, the root samples were oven-dried at 60°C for 48 h to determine fine root biomass. Specific root length (SRL) was calculated as total root length per unit root dry mass (cm g^-1). The hyphae were collected using the suspension-filtration method (Wallander et al., 2004), subsequently lyophilized and stored at -20 ℃ (Guo et al., 2021). To quantify the hyphal biomass in silica sand samples, the fungal-specific biomarker (ergosterol) was extracted as per Wallander & Nylund (1992) and quantified using high-performance liquid chromatography (HPLC) (Dionex Ultimate 3000, Thermo Fisher Scientific, Waltham, USA). In our previous study, we characterized the response of ECM fungal community composition to N addition in the same forests (Guo et al., 2021). Details of soil DNA extraction, amplicon barcoding, and sequence data processing were described in Appendix S1 and Guo et al. (2021). In total, 32 and 27 ECM fungal genera were identified in the Pinus armandii and Picea asperata forests, respectively. Among them, 16 ECM fungal genera were shared between the two forests which accounted for 81.7% and 99.6% of the ECM fungal community in each forest. In this study, we reclassified ECM fungal hyphal exploration types into contact-short distance (C-S), contact-medium distance (C-M) type and medium-long distance (M-L) types, based on Agerer (2001, 2006) and Tedersoo and Smith (2013) to reflect changes in hypha foraging strategy. 

Nutrient-mining traits of roots and hyphae

The amount of root- and hypha-derived C inputs as well as the root- and hypha-effect on N- and P-acquisition enzyme activities were determined to access the changes of nutrient-mining strategies of roots and hyphae under N addition. Briefly, the C and N concentrations and δ¹³C of the roots, hyphae, and C₄ soil in cores were determined on a continuous flow CN analyzer (Flash EA 2000; Thermo Fisher Scientific, Bremen, Germany) coupled to an isotope mass spectrometer (model DeltaV; Thermo Fisher Scientific, Bremen, Germany). The C isotope ratio values were expressed with the delta notation (δ):

δ¹³C‰ = [(R_sample/R_standard -1) ×1000]

where R_sample and R_standard represent the ¹³C/¹²C ratio of samples and standard, respectively, and R_standard is referenced to the Vienna Pee Dee Belemnite (VPDB) (0.13‰). The ¹³C isotopic signature and C concentration of C4 soil in each type of ingrowth core are shown in Table S2. Root- and hypha-derived C inputs into each core were quantified using a two-source isotopic mixing model as per Panzacchi et al. (2016):

f_new= (δ¹³C_R/H - δ¹³C_EH)/(δ¹³C_root/hyph - δ¹³C_EH)

New C input = SOC_{R /H} × f_new,

where δ¹³C_R/H is the δ¹³C of the soil in the R- or H-cores, δ¹³C_EH is the δ¹³C of the soil in the EH-cores, and δ¹³C_root/hypha is the δ¹³C of the root or hypha (Table S3). SOC_R/H is the %C of the soil in the R- or H-cores. The amount of new C originating from the root was calculated by the difference between new C inputs in the R-cores and that in the H-cores.

The activities of two N-acquisition enzymes (leucine aminopeptidase (LAP) and 1,4-β-N-acetyl-glucosaminidase (NAG)) and one P-acquisition enzyme (acid phosphatase, ACP) were measured to evaluate the mineralization of soil organic N and soil organic P, respectively. The enzyme activities were assayed by using a microplate fluorometric method (Saiya-Cork et al., 2002). To assess the influences of roots and hyphae on soil nutrient mineralization, the root-effect (RE) and hypha-effect (HE) were calculated as the ratios of enzymatic parameters in the R-cores to its corresponding values in the H-cores (R_core/H_core) and the ratios of enzymatic parameters in the M-cores to its corresponding values in the EH-cores (H_core/EH_core), assuming that the root- and hypha-effect are additive in the R-core (Finzi et al., 2015).