Experiments were conducted at the Forest Ecosystem State Positioning Observation Station in the Three Gorges Reservoir area (110°54′E, 30°53′N, 375 m altitude), Zigui County, Hubei Province, China (Figure 1). P. massoniana is one of the main coniferous species in this area, and the proportion of total area and total stock is as high as 48.8 and 64.2% (Ge et al., 2017), and the N deposition is 30 kg N hm^–1 a^–1 (Ding et al., 2020).

We designed a two-factor randomized block treatment. The first factor was EMF, which were Sg, Pt, and control (CK), respectively, EMF were provided by the Institute of Forest Ecological Environment and Nature Conservation, Chinese Academy of Forestry Sciences. The second factor was N addition treatment, which was divided into four levels according to the annual atmospheric N deposition in this area: 0 kg N hm^–1 a^–1, 30 kg N hm^–1 a^–1 (normal deposition), 60 kg N hm^–1 a^–1 (moderate deposition), and 90 kg N hm^–1 a^–1 (excessive deposition). There were a total of 12 treatments, and each treatment had 100 pots, with a total of 1,200 pots. Before the N addition, the seedlings grew for 3 months to ensure the symbiotic relationship between seedlings and EMF. After that, N addition was conducted once a month, and the NH₄NO₃ solution of 0, 0.714, 1.428, and 2.143 g/L was added each time, respectively.

In March 2021, 1,200 one-year-old P. massoniana seedlings were randomly assigned to greenhouse, and each was planted in a separate pot (20 cm in diameter and 15 cm in height), and filled with 3 kg of soil. The soil was collected from the P. massoniana forest stands within 2 km of the experimental site and stand by after high-temperature sterilization. The basic physical and chemical properties of the experimental soil were as follows: total soil N was 1.01 g kg^–1, total soil P was 0.54 g kg^–1, total soil K was 1.53 g kg^–1, available soil N was 47.28 mg kg^–1, available soil P was 8.97 mg kg^–1, available soil K 90.17 mg kg^–1, organic matter content was 13.97 g kg^–1, and pH was 5.97.

The ¹³C–CO₂-labeling took place under natural daylight conditions (cloudy, partly sunny) on 12th October 2021. In total, 180 seedlings were randomly selected for each treatment, labeled with ¹³CO₂ in the morning (7:00–11:00). The ¹³CO₂ pulse labeling was conducted in a homemade Plexiglas chamber. To avoid gas leakage, the chamber was placed in a concave groove and sealed with water. The tiny hole left by the injection was covered with scotch tape to avoid gas leakage. To guarantee a uniform distribution of ¹³CO₂, two fans were installed inside the chamber and were used to mix the air thoroughly during the labeling.¹³CO₂ was produced by 1 N ¹³C–Na₂CO₃ solution and 1 N H₂SO₄ solution. Using an LI-8100 CO₂ infrared gas analyzer (LI-COR, Inc., American) to continuously monitor the concentration of carbon dioxide in the chamber. When the initial CO₂ levels had declined to 100 μl L^–1, H₂SO₄ was added to the vial to increase the CO₂ concentration to 400 μl L^–1. Once all of the ¹³C–Na₂CO₃ was used and the total CO₂ concentration decreased below 400 μl L^–1, H₂SO₄ was added to the flask with Na₂CO₃ to maintain the CO₂ concentration to 400 μl L^–1. This process was repeated approximately 8 times before the Plexiglas chamber was removed (Figure 2).

Leaves, branches, shoot, root, and soil samples were collected on the labeling day before (coded as 0 day) and the following 1, 5, 10, 21, and 30 days after labeling. Plant aboveground parts of all species were harvested and pooled as shoot samples by clipping at the soil surface. Soil cores (5 cm in diameter) were taken to 15-cm depth. All roots and soil in the cores were carefully extracted and sieved with a 2-mm sieve. Soil samples passed through the sieve were air-dried for total C and ¹³C analysis. The sampled roots were carefully washed by wet sieving through a 0.5-mm sieve to remove attached soil and debris. Shoot and root samples were oven-dried at 65°C for 48 h. The δ¹³C and C contents were determined using an isotope ratio mass spectrometer (Delta V Advantage) coupled with an elemental analyzer (Flash 2000 EA-HT) (Thermo Fisher Scientific Inc., Waltham, MA, United States).

NaOH was used to absorb CO₂ produced by soil respiration in a certain period of time (Deng et al., 2019). At the end of labeling, a closed breathing chamber containing NaOH solution (1 mol L^–1) (7.5 cm in diameter and 9 cm in height) was inserted into the basin soil to absorb CO₂ produced by soil respiration on the 1st, 4th, 5th, 11th, and 9th day after labeling, and the amount of CO₂ uptake was determined by HCl (1 mol L^–1).

Photosynthetic rates were measured from 9:00 to 11:00 am before collected plants. The top-positioned leaves from one pot were used to establish photosynthetic rates using a portable closed gas-exchange system (LI-6400; Li-Cor, Lincoln). Thirteen light conditions (2,000, 1,600, 1,200, 1,000, 800, 400, 200, 160, 120, 80, 40, 20, and 0 μmo m^–2 s^–1) of PPFD were provided using a red–blue LED light source. The concentrations of CO₂ in ambient air entering the leaf chamber and leaf-to-air vapor pressure deficit were maintained at 350 μl L^–1 and 1.1 kPa, respectively, during the measurement period.

All C isotope values were reported using the isotope delta values (δ¹³C), that is, ¹³C/¹²C, were expressed in per mil (‰) in relation to the international reference standard:

where R_sam and R_std are the ¹³C/¹²C ratios of the sample and standard, respectively.

The atomic percentage of ¹³C in each organ (F_i, %), was as follows:

The total amount of C in each organ (C_i, g) was determined as follows:

where W_i is the biomass in each organ, and T_i is the percentage of C in each organ.

The total amount of fixed ¹³C in each organ (¹³C_i, g) was determined as follows:

where F_n is the ¹³C atomic percentage in the unlabeled sample.

The ¹³C allocation ratio in each organ at each sampling time (P_i, %) was as follows:

where ¹³C_F is the sum of the ¹³C in the whole plant.

The ¹³CO₂ respiration rate in soil (VCO₂ (soil), mg plant^–1 day^–1) was as follows:

where mCO₂ is the amount of CO₂ absorbed by NaOH, A is the area of the breathing chamber, B is the area of the mouth of the pot, and Δt is the time to absorb CO₂.

The variation of the partitioning of the A¹³C_t (%) in each organ of the plant under different treatment after pulse labeling is given by:

where ¹³C_t is the total amount of ¹³C in different parts of plants, t is different parts, and ¹³C_total,1 is the total ¹³C reserved 1 day after pulse lebeling.

Soil retention and plant respiration [A¹³C_other(mg)] is given by:

Statistical analysis was performed using the SPSS 16.0 software package (SPSS, Chicago, IL, United States). A two-factor analysis of variance (ANOVA) followed by Duncan’s multiple comparisons was used. Data shown are the mean standard error (SE). Two-way ANOVA was used to test the effects of the different labeling times and sampling times on the variables under evaluation. All figures were drawn using Origin software 2021 (Origin Lab Corp., Northampton, MA, United States).

The mean value of the net photosynthetic rate at different periods before collecting plants was calculated. Across the N-gradient fertilization, we observed that the net photosynthetic rate of P. massoniana inoculated with EMF was higher than CK, it increased steadily with N addition to a peak at N60, and declined at the highest levels. The net photosynthetic rate of P. massoniana inoculated with Sg was better than Pt at N0 and N30. In contrast, Pt was better than Sg at N60 and N90. This indicates that the net photosynthetic rate of P. massoniana inoculated with EMF was better than that of non-inoculated and was also affected by the concentration of N addition (Table 1).

The variation trend of δ¹³C in different organs of P. massoniana under different treatments was as follows: the δ¹³C sharply declined in leaves after the end of pulse labeling and decreased slowly until reaching the peak in roots, stem, and branches on day 5. During 1–30 days after pulse labeling, the value of the δ¹³C in leaves inoculated with Sg, Pt, and CK decreased by 383.23, 368.76, and 319.70‰, respectively, under N0 (Supplementary Table 1). With increasing N concentration, the δ¹³C in leaves initially increased before decreasing and reached the highest value at N60. The δ¹³C produced by photosynthesis in the leaves completed the transport from the leaves to roots, stem, and branches 1–5 days after the end of labeling. N addition, mycorrhizal symbiosis, and interaction of N and EMF could all increase the output of photosynthetic products from the source leaves during the period of pulse labeling, and the output was the highest at Sg + N60 (Figure 3).

On the first day after pulse labeling, the δ¹³C in roots, stem, and branches was treated with N addition, mycorrhizal symbiosis, and interaction of N and EMF and was higher than those of the CK, indicating that the treatment could promote the transport of photosynthetic products from source to sink during the marking period. The variation trend of δ¹³C in root was gradually increased after the end of labeling, and the value of EMF reached the maximum on the tenth day (Pt + N60) (Supplementary Table 2). During the 1st–5th days, the N treatment value increased before stabilizing, whereas the CK treatment decreased gradually; the minimum value appeared on the tenth day. In the branches and stems, the δ¹³C of all treatments reached the maximum on the fifth day, and all reached the maximum value under N60 treatment, except for the CK (Supplementary Tables 3, 4). The δ¹³C reached the maximum value on the fifth day, then decreased sharply, before stabilizing (Figure 3).

Nitrogen addition, mycorrhizal symbiosis, and interaction of N and EMF increased the rate of ¹³CO₂ release from soil respiration and delayed the occurrence of the maximum value. The specific performance was as follows: the CK treatment reached the maximum value on the first day after the end of labeling, whereas under N addition, mycorrhizal symbiosis, and interaction of N and EMF, it reached the maximum value on the fifth day after the end of labeling. After reaching the maximum value, the ¹³CO₂ release rate of soil respiration decreased gradually and then stabilized (Supplementary Table 5). Under all of the N addition concentrations, the soil respiration release rate of mycorrhizal plants was significantly higher than that of nonmycorrhizal plants. The amount of ¹³CO₂ accumulated in soil increased sharply within 10 days after labeling and then stabilized. It can be seen that the fixed ¹³C was exported within 0–10 days. During this period, the amount of ¹³CO₂ accumulated in soil inoculated with EMF was higher than that of uninoculated plants at the same level of N addition (Figure 4 and Supplementary Table 6).

During the labeling period, ¹³C produced by photosynthesis of Pinus was finally distributed in three parts: plant, soil respiration, and other parts. On the 30th day after the end of labeling, ¹³C remaining in CK-treated plants accounted for 46.85% of the total fixed amount during the labeling period, 16.71% of ¹³C consumed through soil respiration, and 36.44% of ¹³C in other parts. At the same level of N application, the retention of ¹³C in mycorrhizal seedlings increased, and Pt was higher than Sg. The two kinds of EMF were the largest at N60, and Sg and Pt accounted for 71.78 and 81.44%, respectively. Mycorrhizal symbiosis also reduced the proportion of ¹³C released from soil respiration. With the increase of N application concentration, the ¹³C retained by plants increased, reached the maximum distribution ratio at N60 and decreased at N90. Compared with CK, when inoculated with Sg and Pt in N60, the retention ratio of ¹³C in plant increased by 24.93 and 34.59%, respectively. The results showed that the N addition, mycorrhizal symbiosis, and interaction of N and EMF increased, whereas N application decreased the recent photosynthate retention ratio in the plant (Figure 5).

The distribution proportion of ¹³C in leaves decreased with increasing labeling time, whereas increased with the increase of labeling time. EMF increased the output of ¹³C from leaves and decreased the distribution ratio of ¹³C in leaves. On the 30th day after the end of labeling, the distribution ratio of ¹³C inoculated with Sg and Pt in leaves increased by 8.83 and 7.71%, respectively, compared with CK. The distribution ratio of ¹³C in roots inoculated with Sg and Pt increased by 7.62 and 4.80% under N0, respectively. Within 1–10 days after the completion of ¹³C transfer, the increase of ¹³C distribution ratio of ¹³C with EMF was higher than that of the CK under N0 and N30, but lower than that of the CK under N60 and N90 concentrations. However, the ¹³C distribution ratio in roots initially increased and then decreased with N addition. The distribution ratio was highest at N60, indicating that the N application could promote the recent photosynthate consumption of Pinus roots (Figure 6).

The transport of C compounds produced by photosynthesis is affected by EMF and environmental N concentration. Högberg traced ¹³C compounds produced by photosynthesis in pine forests and found that ¹³C compounds were quickly transferred to mycorrhizal mycelium and redistributed, but excessive N application significantly reduced their distribution to the underground part of plant (Högberg et al., 2010). In our study, under the treatment of EMF, δ¹³C in leaves decreased sharply after labeling and greatly slowed down on the 5th day. δ¹³C in stems and branches increased to the highest value on the 5th day after labeling, whereas δ¹³C in roots continued toward the peak on the 10th day, and δ¹³C was significantly higher than that of the CK. Under the concentration of N0 and N30, the effect of Sg inoculation was better than that of Pt. Contrarily, the effect of Pt inoculation was better than that of Sg under N60 and N90, indicating that the transport rate of photosynthate from source leaf to root was affected by EMF. Inoculation with EMF delayed the process when δ¹³C peaked in roots, which was consistent with the results of Högberg et al. (2010) and Leake et al. (2001). It is well-known that since EMF can form a good symbiotic relationship with the host plant, it can promote the growth of itself and host plants by expanding the root absorption area for obtaining more nutrients (Kayama and Yamanaka, 2014; Zhang et al., 2017). When the growth of trees was limited due to low soil nutrient content, EMF could help trees obtain growth-limiting nutrients, such as C, N, and P, through more mycelia formed by symbiosis with tree roots. This improves the photosynthetic C assimilation ability of host plants (Liu, 2017; Plett et al., 2020).

In the ¹⁴C labeling experiment of Pinus sylvestris var. mongolica, it was found that ¹⁴C photosynthetic products were quickly transported to the root within 1 day after the end of labeling and transferred to the external mycelium within 3 days. The higher the mycelium density, the higher the ¹⁴C photosynthate (Wu et al., 2001). EMF was a strong C sink, and sufficient nutrients could increase the amount of photosynthetic products and the distribution of photosynthate from source to sink of Pinus, whereas the transport of C compounds from root to aboveground was conducted through the soil outside the hypha (Simard et al., 1997; Gorka et al., 2019). In our study, both kinds of EMF could promote the amount of δ¹³C in various organs of Pinus, which was significantly higher than that of the CK. However, there was a difference between the two kinds of EMF, which may be due to the difference in the number of mycelium formed by the young root and the number of mycelium that continued to grow after the symbiosis of different EMF and host plants. This resulted in differences in nutrient absorption capacity due to different root richness. EMF delays the time when δ¹³C reaches the peak value in roots, which may be because photosynthetic products can promote the extension of hypha. These are used by EMF for its own growth and extension, and δ¹³C can be transferred between mycelium. It is seldom transported to the shoot.

The availability of soil N influences the morphological characteristics of soil. It has been found that high N supply promotes plant root development and leads to larger root biomass (Kraiser et al., 2011; Meng et al., 2016). In our study, δ¹³C in roots, stems, branches, and leaves of P. massoniana initially increased before decreasing with increasing concentration of N application. All ratios reached the peak on the 5th day after the end of labeling, reaching the highest at N60 and declined at N90, which was consistent with the results of Pinus tabulaeformis (Wang et al., 2013) and poplar (Meng et al., 2018). The results showed that in the range of medium concentration N deposition, N addition promoted the photosynthesis of Pinus and increased the transport of δ¹³C to the sink organs, which was mainly attributed to the fact that when N was sufficient, the source leaves transported enough photosynthates to the sink tissue to ensure the growth of roots, so that the roots synthesized C and N compounds beneficial to plant growth and formed a benign C and N cycle (Schlüter et al., 2012). When the concentration of N application exceeded this range, it began to decrease, and the effect could be weakened when the N concentration was excessive after inoculation with EMF.

Pulse labeling experiment had become a common method to study assimilation, distribution, and respiration in physiology and ecology (Bowling et al., 2008). Epron et al. (1999) found that when photosynthetic products were synthesized, the release from soil respiration was mainly concentrated in the early stage after photosynthate synthesis. In our study, during 1 to 10 days after labeling, the release of ¹³C from photosynthesis in soil respiration increased sharply, before stabilizing, indicating that the ¹³C products consumed by soil respiration were mainly the recent products of plant photosynthesis. N addition, mycorrhizal symbiosis, and interaction of N and EMF all increased the total release of photosynthates from soil respiration. At the same the N application concentration, the ¹³C produced by soil respiration of P. massoniana inoculated with EMF was higher than that of uninoculated plants, but with increasing N concentration, the ¹³C produced by soil respiration initially increased before decreasing, with the highest at middle N concentration (N60), and decreased at the highest N application concentration (N90). The results showed that the C emission from soil increased with the increase of photosynthate synthesis, and N addition and EMF could improve the photosynthetic capacity of plants, thus affecting the process of C cycle in plants.

Nitrogen addition, mycorrhizal symbiosis, and interaction of N and EMF all delayed the time when the rate of soil respiration and release of ¹³CO₂ reached the peak value, indicating that treatments could enhance the recent release of photosynthetic products from soil respiration in P. massoniana. Soil respiration was highest at medium N concentration (N60) (Pt > Sg), which may be because the photosynthetic C input in the underground was closely related to the growth rate of roots. Suitable N concentration and EMF improved the soil environment of plant roots (Ge et al., 2015). Plants had sufficient nutrition and root activity to transport more C and enhance root respiration during the growing period. Simultaneously, N addition and EMF promoted the ability of photosynthesis of Pinus and made plants produce more photosynthates, which also increased the plant consumption of recent photosynthate transformation and respiration. In addition, N addition and EMF inoculation reduced the retention time of recent photosynthates in source leaves, thus increasing the release rate of recent photosynthates from soil respiration. However, the rate of ¹³CO₂ release from soil respiration began to decrease with excessive N concentration, but in the mycorrhizal plants it was significantly higher than that of nonmycorrhizal plants.

There are three ways for plants to distribute new photosynthetic products, including absorption and utilization of plant organs, retention of soil organic matter through soil respiration or transformation into soil, and atmospheric release in the form of CO₂ (Wu et al., 2010). During the whole tracer period, N addition, mycorrhizal symbiosis, and interaction of N and EMF all increased the retention ratio of photosynthate in plants, and the retention ratio of plants inoculated with Pt at N60 was the highest. The results showed that EMF was beneficial to the allocation of photosynthetic products in the plant during N deposition in a certain range. On the 1st to 5th day after labeling, with the increase of N addition, the output rate of ¹³C in leaves inoculated with EMF increased with increasing photosynthesis, which was consistent with the overall increment of root, stem, and branch absorption. This phenomenon was consistent with the C distribution from source leaves to sink organs of P. tabulaeformis as observed by Wang and Liu (2014). The amount of C allocated to the root mainly depends on the amount produced by plant photosynthesis (Farrar and Jones, 2000). A certain range of N application and EMF promoted the production of more photosynthetic products in Pinus, which increased the photosynthetic products allocated to the root system. The root system had enough nutrients to maintain the growth of itself and the plant (Zong et al., 2018), and the symbiosis of hypha and root increased the absorption and transport area of the plant. This can distribute more C to the deeper soil, whereas excessive N concentration would lead to the imbalance of nutrient elements in the soil, thereby reducing the distribution of photosynthetic products to the root system. Under the suitable N application rate, the ¹³C content in leaves and roots of Pinus inoculated with EMF increased significantly, indicating that the C distribution in the roots and leaves with the most active metabolism was controlled by N application rate and EMF “pull” (Farrar and Jones, 2000). When the aboveground C sink was sufficient, the mycorrhiza had enough “pulling force” to distribute ¹³C to each sink organ, and the gradient difference of N concentration led to the difference of C distribution. At the highest N addition (90 kg N hm^–1 a^–1), the favorable effect on the allocation of photosynthetic products began to decline, but the negative effect could be attenuated by EMF.

Previous studies have shown that plants can better maintain their growth under adequate nutrient supply and suitable growth environment, increase the biomass of aboveground and belowground parts, and enhance photosynthesis to produce more C compounds and distribute them to various organs (Reynolds and Thornley, 1982; Zong et al., 2018). However, N deficiency increases the distribution of photosynthate to the roots of different plant types, shrubs, herbs, or woody plants (Cronin and Lodge, 2003; Vanninen and Mäkelä, 2005; Grechi et al., 2007). This was because N deficiency reduces the transport of cytokinin from root to stem, reduces the rate of cell division, and reduces the transport of C from phloem to stem, which causes C to accumulate around the phloem, increasing pressure on leaves. Meanwhile, root cells continue to divide, resulting in a pressure difference between source leaf and sink root, and relatively more photosynthetic products are allocated to the root (Ping et al., 2010). But adequate N supply and inoculation of EMF enabled the plant to produce more photosynthetic products, whereas larger root area and biomass also ensured the distribution of photosynthetic products to all parts. In our study, it was found that EMF could promote the distribution of newly fixed photosynthetic products in plants, and inoculation of EMF slowed down the decrease of photosynthetic products caused by high N deposition. Therefore, inoculation of EMF and a certain range of N deposition (less than 60 kg N hm^–1a^–1) could promote the allocation of photosynthetic products in Pinus.