High Nitrate or Ammonium Applications Alleviated Photosynthetic Decline of Phoebe bournei Seedlings under Elevated Carbon Dioxide

Phoebe bournei is a precioustimber species and is listed as a national secondary protection plant in China. However, seedlings show obvious photosynthetic declinewhen grown long-term under an elevated CO2 concentration (eCO2). The global CO2 concentration is predicted to reach 700 μmol·mol−1 by the end of this century; however, little is known about what causes the photosynthetic decline of P. bournei seedlings under eCO2 or whether this photosynthetic decline could be controlled by fertilization measures. To explore this problem, one-year-old P. bournei seedlings were grown in an open-top air chamber under either an ambient CO2 (aCO2) concentration (350 ± 70 μmol·mol−1) or an eCO2 concentration (700 ± 10 μmol·mol−1) from June 12th to September 8th and cultivated in soil treated with either moderate (0.8 g per seedling) or high applications (1.2 g per seedling) of nitrate or ammonium. Under eCO2, the net photosynthetic rate (Pn) of P. bournei seedlings treated with a moderate nitrate application was 27.0% lower than that of seedlings grown under an aCO2 concentration (p < 0.05), and photosynthetic declineappeared to be accompanied by a reduction of the electron transport rate (ETR), actual photochemical efficiency, chlorophyll content, ribulose-1,5-bisphosphate carboxylase/oxygenase (rubisco), rubisco activase (RCA) content, leaf thickness, and stomatal density. The Pn of seedlings treated with a high application of nitrate under eCO2 was 5.0% lower than that of seedlings grown under aCO2 (p > 0.05), and photosynthetic declineoccurred more slowly, accompanied by a significant increase in rubisco content, RCA content, and stomatal density. The Pn of P. bournei seedlings treated with either a moderate or a high application of ammonium and grown under eCO2 was not significantly differentto that of seedlings grown under aCO2—there was no photosynthetic decline—and the ETR, chlorophyll content, rubisco content, RCA content, and leaf thickness values were all increased. Increasing the application of nitrate or the supply of ammonium could slow down or prevent the photosynthetic declineof P. bournei seedlings under eCO2 by changing the leaf structure and photosynthetic physiological characteristics.


Introduction
A range of human activities such as fossil fuel combustion, deforestation, and industrial development are contributing to the rise of global CO 2 concentration [1]. CO 2 , which is one of the main greenhouse gases in Earth's atmosphere, has increased in concentration from 280 µmol·mol −1 before the industrial revolution to more than 400 µmol·mol −1 at present. CO 2 concentration is predicted to reach 700 µmol·mol −1 by the end of this century, which is causing concern among the international community [2][3][4]. An elevated CO 2 (eCO 2 ) concentration not only increases the greenhouse effect of this and concentrations can alleviate the photosynthetic decline of P. bournei seedlings under eCO 2 and to elucidate the mechanism behind these traits. To investigate this problem, we grew one-year-old P. bournei seedlings in an open-top air chamber (OTC) under either an ambient CO 2 (aCO 2 ) concentration or an eCO 2 concentration, applied different amounts of nitrate and ammonium to the soil, and studied the effect of nitrogen form and concentration on gas exchange and chlorophyll fluorescence parameters, which are important indicators of photosynthetic performance [36,37]. Chlorophyll has the function of absorbing and transferring light energy, which is essential for photosynthesis [38].We explored the mechanism of nitrogen regulation from the aspects of photosynthetic chlorophyll content, which can reflect the photosynthetic capacity of plant leaves and has been shown to be significantly positively correlated with the net photosynthetic rate (Pn) [39].We also determined rubisco and rubisco activase (RCA) content because rubisco is the core enzyme in the Calvin cycle and RCA increases the proportion of active rubisco in photosynthetic cells [40]. The response of plants to eCO 2 is basically mediated by leaf photosynthesis, which is closely related to changes in leaf structure and chemical composition [41]. Furthermore, leaf anatomical structures best reflect the effects of environmental factors on plants and the adaptation strategies of plants to the environment [42]. Therefore, we also investigated the influence of the different treatments on leaf structure. These data should provide a theoretical basis and technical guidance for P. bournei cultivation in the future.

Plant Materials and Fertilizer
One-year-old P. bournei container seedlings grown from seeds of the same mother plant were used as the experimental material. In January 2019, seedlings (average height, 9.4 cm; average ground diameter, 2.73 mm) were selected and transplanted to pots (diameter 15.5 cm, height 14.0 cm) containing 1.5 kg of yellow soil, with one seedling per pot. The nutrient content of the soil was as follows: total nitrogen, 0.84g·kg −1 ; alkali nitrogen, 28.3 mg·kg −1 ; total phosphorus, 0.23 g·kg −1 ; available phosphorus, 7.8 mg·kg −1 ; total potassium, 21.6 g·kg −1 ; available potassium, 207.3 mg·kg −1 . Seedlings were allowed to recover from transplanting for four months before treatments were applied. The nitrogen fertilizers used in the experiment were calcium nitrate and ammonium sulfate, the phosphate fertilizer was superphosphate, and the potassium fertilizer was potassium chloride.

Experimental Treatments
An OTC was used to simulate eCO 2 conditions. The OTC was a regular octagonal prism with sides of 1 m in length, a diameter of 2.42 m, and a height of 1.7 m, and comprised a 4 cm × 4 cm square steel frame and high-light-transmission 8-mm tempered glass. To accumulate CO 2 and slow the rate of CO 2 gas loss, the top opening of the OTC was inclined 45 degrees inward. The CO 2 concentration in the gas chamber was controlled by a CO 2 sensor device (GMP-220, Vaisala, Finland). The collected data were transmitted to a computer that controlled the switch of the solenoid valve of each gas chamber by running the programmed program to control the CO 2 concentration in the gas chamber within the target concentration range. Seedlings were treated with either a natural aCO2 concentration of 280-450 µmol·mol −1 or a controlled eCO 2 concentration of 700 µmol·mol −1 , and with either a moderate level of nitrate (NO 3 − ; 0.8 g per seedling), a high level of nitrate (iNO 3 − ; 1.2 g per seedling), a moderate level of ammonium (NH 4 + ; 0.8 g per seedling), or a high level of ammonium (iNH 4 + ; 1.2 g per seedling), i.e., eight treatments in total, each treatment consisting of 45 replicates, totaling to 360 pots. The nitrogen applications were divided into three equal quantities and applied on 3 June, 3 July, and 2 August. The CO 2 treatments began on 12 June and endedon8 September. Phosphorus and potassium were also applied to seedlings on 3 June, 3 July, and 2 August. The phosphorus and potassium application rate was based on the moderate level of nitrogen application (i.e., N:P 2 O 5 :K 2 O = 1:1:1). The relevant indicators for each treatment were assessed at the end of CO 2 treatment.

Determination of Gas Exchange
Three seedlings were selected from each treatment, and each of the three upperpart functional leaves were labeled. The net photosynthetic rate (Pn), stomatal conductance (Gs), intercellular CO 2 concentration (Ci), and transpiration rate (Tr) of these leaves were measured between 9:00 a.m. and 11:00 a.m. on a sunny day using an Li-6400 portable photosynthesis measurement system (LI-COR Biosciences, Lincoln, NE, USA). The LiCor chamber temperature and humidity measured by the LiCor were 30.7 ± 1.8 • C and 48.6 ± 2.4%. A small CO 2 cylinder was used to control the CO 2 concentration of the leaf chamber at 350 µmol·mol −1 in theaCO 2 gas chamber and at 700 µmol·mol −1 in the eCO 2 chamber. Photon flux was set to 1200 µmol·mol −1 using a red and blue light source.

Determination of Chlorophyll Fluorescence
Chlorophyll fluorescence was determined using a modulated chlorophyll fluorescence imaging system (Imaging-Pam mini version, Walz, Germany). After seedlings had been treated in the dark for 1h, the initial fluorescence (Fo) and maximum fluorescence (Fm) in the dark-adapted state of the labeled leaves were determined. Next, the fluorescence kinetic curve was determined to calculate the electron transport rate (ETR), actual photochemical efficiency (Y(II)), photochemical quenching (qP), and non-photochemical quenching (NPQ) in which the qP is the quenching of chlorophyll fluorescence production due to photosynthesis and NPQ is the quenching of chlorophyll fluorescence production due to heat dissipation. The maximum quantum efficiency of photosystem II (Fv/Fm) was determined by calculating (Fm -Fo)/Fm. The Fo and Fm are initial fluorescence and maximum fluorescence in the light-adapted state respectively, the effective photochemical efficiency of photosystem II (Fv /Fm )was determined by calculating (Fm -Fo )/Fm .

Determination of Chlorophyll and Key Photosynthetic Enzyme content
The upper functional leaves (0.2 g) of five seedlings in each treatment were sampled. The chlorophyll content was determined by ethanol extraction UV spectrophotometry and calculated according to the following formula: chlorophyll a = 12.7 × OD 663 − 2.69 × OD 645 ; chlorophyll b = 22.9 × OD 645 − 4.68 × OD 663 [43], OD is optical density. Rubisco and RCA content were measured using enzyme-linked immunosorbent assay kits produced by Shanghai Keshun Biological Technology Co., Ltd., Shanghai, China.

Determination of Leaf Anatomy and Stomatal Density
To observe the leaf anatomy, labeled leaves were fixed in FAA (5% formaldehyde, 5% glacial acetic acid, 90% ethanol of 70%), embedded in paraffin wax and then sectioned to obtain transverse sections (10 µm thick), which were stained with safranin-fast green. These sections were dehydrated and their anatomy was observed using an optical microscope of Nikon Eclipse (Nikon, Tokyo, Japan) and photographed using a Nikon DS-U3 (Nikon, Tokyo, Japan) [44]. Case Viewer software was used to measure the following anatomical structure indicators: cell tightness rate (CTR) = palisade tissue thickness/leaf thickness; scattered rate (SR) = spongy tissue thickness/leaf thickness [45]. To calculate stomatal density, labeled fresh leaves were sprayed with gold and observed and photographed using a scanning electron microscope (Carl Zeiss, Jena, Germany).

Statistical Analysis
Three-way analysis of variance (ANOVA) was employed to compare the treatment effects, and the treatments means were evaluated by Tukey's multiple-range test (p < 0.05). All data analyses were

Gas Exchange
Under eCO 2 , the Pn, Gs, and Tr of seedlings treated with a moderate amount of nitrate were significantly lower (27.0%, 20.2%, and 22.5% lower, respectively) than that of seedlings grown under aCO 2 ( Figure 1). However, compared with seedlings treated with a moderate amount of nitrate, seedlings treated with a high level of nitrate showed a significant increase in Pn, Gs, and Tr under both eCO 2 and aCO 2 . Furthermore, the Pn, Gs, and Tr values obtained under aCO 2 and eCO 2 conditions were not significantly different (i.e., the adverse effects of eCO 2 on the gas exchange of P. bournei seedlings were alleviated when seedlings were treated with a high level of nitrate) ( Figure 1). Seedlings that received applications of ammonium showed increased Pn, Gs, and Tr under eCO 2 compared with seedlings grown under aCO 2 ; however, the Pn of seedlings that received moderate and high applications of ammonium were only 8.3% and 4.4% higher, respectively, than those grown under aCO 2 and the difference was not significant. By contrast, the Gs of seedlings that received moderate and high applications of ammonium were 21.9% and 23.5% higher, respectively, under eCO 2 than those grown under aCO 2 , indicating that under eCO 2 , ammonium applications were beneficial for gas exchange ( Figure 1).

Gas Exchange
Under eCO2, the Pn, Gs, and Tr of seedlings treated with a moderate amount of nitrate were significantly lower (27.0%, 20.2%, and 22.5% lower, respectively) than that of seedlings grown under aCO2 ( Figure 1). However, compared with seedlings treated with a moderate amount of nitrate, seedlings treated with a high level of nitrate showed a significant increase in Pn, Gs, and Tr under both eCO2 and aCO2. Furthermore, the Pn, Gs, and Tr values obtained under aCO2 and eCO2 conditions were not significantly different (i.e., the adverse effects of eCO2 on the gas exchange of P. bournei seedlings were alleviated when seedlings were treated with a high level of nitrate) ( Figure  1). Seedlings that received applications of ammonium showed increased Pn, Gs, and Tr under eCO2 compared with seedlings grown under aCO2; however, the Pn of seedlings that received moderate and high applications of ammonium were only 8.3% and 4.4% higher, respectively, than those grown under aCO2 and the difference was not significant. By contrast, the Gs of seedlings that received moderate and high applications of ammonium were 21.9% and 23.5% higher, respectively, under eCO2 than those grown under aCO2, indicating that under eCO2, ammonium applications were beneficial for gas exchange ( Figure 1).

Chlorophyll Fluorescence
Under eCO2, ETR and Y(II) of seedlings treated with moderate nitrate applications were significantly lower than under aCO2; however, additional nitrate increased ETR and Y(II) regardless of CO2treatment (Figure 2). The ETR and Y(II) of seedlings applied with ammonium were higher stomatal conductance; Ci, intercellular CO 2 concentration; Tr, transpiration rate. Data points represent means ± standard deviation (SD); n = 9. Different lowercase letters represent significant differences between treatments (p < 0.05) according to Tukey's test.

Chlorophyll Fluorescence
Under eCO 2 , ETR and Y(II) of seedlings treated with moderate nitrate applications were significantly lower than under aCO 2 ; however, additional nitrate increased ETR and Y(II) regardless  (Figure 2). The ETR and Y(II) of seedlings applied with ammonium were higher under eCO 2 than under aCO 2 . The qP values of seedlings treated with high nitrate applications were significantly higher than those treated with moderate nitrate or ammonium applications regardless of CO 2 treatment. By contrast, the qP of seedlings treated with high ammonium was only significantly higher (36.59% higher) than those treated with moderate ammonium or with moderate nitrate applications when grown under eCO 2 ( Figure 2). The different CO 2 concentrations did not significantly affect the Fv/Fm, Fv /Fm , and NPQ values of seedlings ( Figure 2). Forests2018, 9, x; doi:FOR PEER REVIEW www.mdpi.com/journal/forests under eCO2 than under aCO2. The qP values of seedlings treated with high nitrate applications were significantly higher than those treated with moderate nitrate or ammonium applications regardless of CO2 treatment. By contrast, the qP of seedlings treated with high ammonium was only significantly higher (36.59% higher) than those treated with moderate ammonium or with moderate nitrate applications when grown under eCO2 ( Figure 2). The different CO2 concentrations did not significantly affect the Fv/Fm, Fv , /Fm , , and NPQ values of seedlings ( Figure 2).

Chlorophyll Content
The chlorophyll content of seedlings treated with moderate nitrate applications decreased when grown under elevated CO 2 conditions (Table 1). The chlorophyll a, b, and a + b content of seedlings that received moderate applications of nitrate decreased under eCO 2 by 19.3%, 20.0%, and 18.6%, respectively. By contrast, the chlorophyll a, b and a + b content of seedlings grown under eCO 2 that received high applications of nitrate were not significantly different, indicating that high levels of nitrate could alleviate the negative effects of eCO 2 on chlorophyll content (Table 2). Seedlings treated with ammonium showed higher levels of chlorophyll a, b and a + b synthesis under eCO 2 than seedlings grown under aCO 2 , particularly when treated with high levels of ammonium. Although the chlorophyll a/b of seedlings was not significantly affected by the different CO 2 concentrations, chlorophyll a/b levels were significantly higher when treated with moderate nitrogen applications rather than high nitrogen applications.

Content of Key Photosynthetic Enzymes
The rubisco and RCA content of seedlings treated with a moderate amount of nitrate were significantly lower (21.48% and 38.81% lower, respectively) in seedlings grown under eCO 2 compared with those grown under aCO 2 . By contrast, the rubisco and RCA content in seedlings treated with high levels of nitrate increased by 22.30% and 2.00%, respectively, under eCO 2 compared with seedlings under aCO 2 (Figure 3). The rubisco and RCA content of seedlings that received high levels of ammonium were significantly higher than those that received moderate ammonium application, but were not significantly different under aCO 2 or eCO 2 conditions (Figure 3). Forests2018, 9, x; doi:FOR PEER REVIEW www.mdpi.com/journal/forests received high levels of ammonium were significantly higher than those that received moderate ammonium application,but were not significantly different under aCO2 oreCO2 conditions ( Figure  3).

Leaf Anatomical Structure and Stomatal Density
We assessed the upper and lower epidermis, palisade tissue, spongy tissue structure and stomata ( Figure 4). The leaf thickness and stomatal density values of seedlings treated with moderate nitrate applications and grown under eCO2 were significantly lower (12.44% and 18.90% lower, respectively) than those of seedlings grown under aCO2 (Table 3). High nitrate applications had a positive effect on stomatal density; however, high nitrate applications did not have a significant effect on leaf thickness under eCO2. The leaf thickness of seedlings that were treated with ammonium and grown under eCO2 was not significantly different to those grown under aCO2 (Table  3). Moderate application of ammonium did not significantly affect stomatal density; however, high levels of ammonium significantly increased stomatal density under eCO2 compared with seedlings grown under aCO2( Table 3). The different nitrogen and CO2 treatments did not significantly affect palisade tissue thickness and CTR. applications of nitrate or moderate (NH 4 + ) or high (iNH 4 + ) applications of ammonium. Data points represent means ± SE; n = 5. Different letters beside data points represent significant differences between treatments (p < 0.05) according to Tukey's test.

Leaf Anatomical Structure and Stomatal Density
We assessed the upper and lower epidermis, palisade tissue, spongy tissue structure and stomata ( Figure 4). The leaf thickness and stomatal density values of seedlings treated with moderate nitrate applications and grown under eCO 2 were significantly lower (12.44% and 18.90% lower, respectively) than those of seedlings grown under aCO 2 (Table 3). High nitrate applications had a positive effect on stomatal density; however, high nitrate applications did not have a significant effect on leaf thickness under eCO 2 . The leaf thickness of seedlings that were treated with ammonium and grown under eCO 2 was not significantly different to those grown under aCO 2 (Table 3). Moderate application of ammonium did not significantly affect stomatal density; however, high levels of ammonium significantly increased stomatal density under eCO 2 compared with seedlings grown under aCO 2 ( Table 3). The different nitrogen and CO 2 treatments did not significantly affect palisade tissue thickness and CTR. received high levels of ammonium were significantly higher than those that received moderate ammonium application,but were not significantly different under aCO2 oreCO2 conditions ( Figure  3). . Rubisco and rubisco activase (RCA) content of Phoebe bournei seedlings grown under ambient (aCO2) or elevated CO2 (eCO2) concentrations and that received moderate (NO3 -) or high (iNO3 -) applications of nitrate or moderate (NH4 + ) or high (iNH4 + ) applications of ammonium. Data points represent means ± SE; n = 5. Different letters beside data points represent significant differences between treatments (p < 0.05) according to Tukey's test.

Leaf Anatomical Structure and Stomatal Density
We assessed the upper and lower epidermis, palisade tissue, spongy tissue structure and stomata ( Figure 4). The leaf thickness and stomatal density values of seedlings treated with moderate nitrate applications and grown under eCO2 were significantly lower (12.44% and 18.90% lower, respectively) than those of seedlings grown under aCO2 (Table 3). High nitrate applications had a positive effect on stomatal density; however, high nitrate applications did not have a significant effect on leaf thickness under eCO2. The leaf thickness of seedlings that were treated with ammonium and grown under eCO2 was not significantly different to those grown under aCO2 (Table  3). Moderate application of ammonium did not significantly affect stomatal density; however, high levels of ammonium significantly increased stomatal density under eCO2 compared with seedlings grown under aCO2( Table 3). The different nitrogen and CO2 treatments did not significantly affect palisade tissue thickness and CTR.

Discussion
Although the photosynthetic rate of various plants grown under eCO 2 has been shown to increase in the short-term, generally, under long-term eCO 2 , the stimulation effect weakens but the rates under eCO 2 remain higher than controls [19]. The growth response of trees to eCO 2 manifests in many physiological aspects in addition to photosynthetic efficiency. Although long-term eCO 2 has been shown to decrease the Gs of Liquidambar styraciflua, stomatal density was unaffected [46]. However, a survey of 100 species and 122 observations showed an average reduction in stomatal density of 14.3% with CO 2 enrichment [47]. Elevated CO 2 and limited nitrogen nutrition can restrict excitation energy dissipation in photosystem II of Betula platyphylla var. japonica [48] and expression of rubisco and RCA in Pinus ponderosa [49] and Piceaabies [50]. In this study, the photosynthetic efficiency of seedlings treated with a moderate amount of nitrate decreased significantly under eCO 2 . In addition, the Gs, ETR, chlorophyll content, rubisco and RCA content, and leaf structure-related indexes of seedlings were reduced, indicating that under eCO 2 and moderate application of nitrate, the anatomical structure and photosynthetic physiological state of seedlings changed, causing photosynthesis to decrease. However, we found that there was no CO 2 effect under high applications of nitrate. By contrast, the Gs, ETR, Y (II), qP, and chlorophyll content did not change significantly, and rubisco and RCA content and stomatal density increased significantly, indicating that high nitrate applications under eCO 2 could help to improve the photosynthetic physiological and structural characteristics of P. bournei seedlings to sustain stimulation of photosynthesis.
Plant growth is governed by the balance of carbon assimilation and respiration. The joint influence of low nitrogen and excess light had an adverse effect on plant growth; in contrast, no adverse physiological responses were observed for plants under either nitrogen limitation or high light intensity conditions [25]. Similarly, in this study, P. bournei seedlings that received moderate nitrate applications under eCO 2 showed the phenomenon of photosynthetic decline, which is consistent with the findings reported by Han et al. [35]. High nitrate applications improved the physiological and structural characteristics of leaves of P. bournei seedlings and reduced photosynthetic decline. A possible explanation for this is that adequate nitrogen supply involved in carbohydrate catabolism and ATP production, which was in line with the sugar availability and high photosynthetic activity and so facilitated plants to cope with adverse environments such as abiotic stress. High applications of nitrate compensated for the shortage of an external nitrogen supply to some extent and adjusted the state of the carbon and nitrogen supply balance, which resulted in improved leaf structure and physiological characteristics and alleviated the degree of photosynthesis decline. Nitrogen is the basic element for enzyme synthesis in plants. About 40% of available nitrogen might be in rubisco [51]. Changes in nitrogen supply had a significant effect on rubisco and RCA content levels, and ultimately on the photosynthetic decline of P. bournei seedlings. Another possible explanation for the photosynthetic decline of P. bournei seedlings under eCO 2 is that eCO 2 affects the metabolism and absorption of nitrate in plants [52,53], which results in the photosynthetic decline of P. bournei seedlings, whereas higher applications of nitrate alleviated the negative impact of eCO 2 on nitrate metabolism and absorption to a certain extent so as to slow down photosynthetic decline. A study by Pettersson and McDonald [26] showed that an insufficient supply of nitrogen could reduce the expression of enzymes involved in photosynthesis, and limit the synthesis of proteins, leading to photosynthetic decline. Furthermore, eCO 2 concentrations can have certain inhibitory effects on the plant's photosynthetic rate, chlorophyll content, photosynthetic electron transport and potential activity of PSII, and its negative effects can be effectively alleviated by nitrogen application [54].
One involves the first biochemical step of nitrate assimilation, the conversion of nitrate to nitrite in the cytoplasm of leaf mesophyll cells. Photorespiration stimulates the export of malate from chloroplasts and increases the availability of nicotinamide adenine dinucleotide hydride in the cytoplasm that powers this first step of nitrate assimilation [22]. The eCO 2 conditions can reduce the photo respiration of leaves [55] and limit plant nitrate uptake in favor of ammonium uptake because eCO 2 conditions inhibit plant photorespiration and thus affect the transport of malate, which is ultimately detrimental to the assimilation of nitrate [56,57]. Under eCO 2 , the Pn of seedlings treated with ammonium was increased compared with those grown under aCO 2 , photosynthetic decline did not occur, and leaf physiological and structural indicators (including ETR, Y(II), and the content of chlorophyll and key photosynthetic enzymes) were increased, which demonstrated that the application of ammonium under eCO 2 was more beneficial for the growth of P. bournei seedlings compared withaCO 2 . However, leaf physiological and structural indicators of seedlings treated with a moderate amount of nitrate decreased significantly under eCO 2 ; the reason for this finding may be that eCO 2 affects photorespiration, which in turn affects the assimilation of nitrate but has no effect on the assimilation of ammonium. Silva et al. [58] reported that coffee trees grown under eCO 2 and treated with ammonium showed better growth than those treated with nitrate. Some studies [32,33,59] have also indicated that ammonium treatment was superior to nitrate treatment under eCO 2 conditions because the increase in CO 2 concentration affected plant stomata conductance and water transpiration, which in turn affected the movement of nitrate in the soil with water, and ultimately affected the absorption of nitrate by the root system. This may also be the reason that ammonium had a more positive effect on seedlings than nitrate under eCO 2 in our experiment. High applications of ammonium had a better effect on P. bournei seedlings than moderate applications. The reason for this may be that an unsaturated increase in fertilizer supply promoted growth. Under the same CO 2 concentration, the photosynthesis of P. bournei seedlings that received an ammonium treatment was weaker than that of seedlings that were treated with nitrate, which may be due to a preference of P. bournei seedlings for this form of nitrogen. Although nitrogen limitation is a major bottleneck for many terrestrial ecosystems, phosphorus limitation issues are also critical for several ecosystems and species [60]. Phosphorus affects plant responses to eCO 2 [61,62]. Also, eCO 2 results in lower Zn and Fe assimilation rates [63]. Therefore, further studies that consider other elements could be recommended in the future.

Conclusions
The results of this experiment indicate that the photosynthetic efficiency of P. bournei seedlings treated with a moderate amount of nitrate decreased under eCO 2 conditions and that photosynthetic decline occurred. Increasing the dosage of nitrate or the supply of ammonium fertilizer could regulate the physiological and structural characteristics of P. bournei seedlings and alleviate or avoid photosynthetic decline, which would be beneficial for the photosynthesis of P. bournei seedlings. Therefore, we suggest that increasing the application of nitrate or applying ammonium may be an effective way of improving the growth of P. bournei seedlings under eCO 2 levels in the future.