Leaf Gas Exchange of Tomato Depends on Abscisic Acid and Jasmonic Acid in Response to Neighboring Plants under Different Soil Nitrogen Regimes

High planting density and nitrogen shortage are two important limiting factors for crop yield. Phytohormones, abscisic acid (ABA), and jasmonic acid (JA), play important roles in plant growth. A pot experiment was conducted to reveal the role of ABA and JA in regulating leaf gas exchange and growth in response to the neighborhood of plants under different nitrogen regimes. The experiment included two factors: two planting densities per pot (a single plant or four competing plants) and two N application levels per pot (1 and 15 mmol·L−1). Compared to when a single plant was grown per pot, neighboring competition decreased stomatal conductance (gs), transpiration (Tr) and net photosynthesis (Pn). Shoot ABA and JA and the shoot-to-root ratio increased in response to neighbors. Both gs and Pn were negatively related to shoot ABA and JA. In addition, N shortage stimulated the accumulation of ABA in roots, especially for competing plants, whereas root JA in competing plants did not increase in N15. Pearson’s correlation coefficient (R2) of gs to ABA and gs to JA was higher in N1 than in N15. As compared to the absolute value of slope of gs to shoot ABA in N15, it increased in N1. Furthermore, the stomatal limitation and non-stomatal limitation of competing plants in N1 were much higher than in other treatments. It was concluded that the accumulations of ABA and JA in shoots play a coordinating role in regulating gs and Pn in response to neighbors; N shortage could intensify the impact of competition on limiting carbon fixation and plant growth directly.


Introduction
Neighboring plant competition and soil mineral shortages resulting from an increasing planting density are common biotic and abiotic stresses, limiting crop growth and production [1,2]. Nitrogen (N), as one of the most essential macronutrients for plant growth and development, plays a significant role in carbon metabolism, protein synthesis, and plant hormone synthesis [3]. Plants may detect and interact with neighbors in different manners [4]. Recent data argue that phytohormones, such as ethylene and abscisic acid (ABA), play a crucial role in the modulation of many processes of plant development, including response to neighbors [5,6]. Transcriptomic analysis of the whole-genome expression showed that nutrient deficiency (mainly N) upregulated genes of ABA biosynthesis in competing Arabidopsis plants [7]. However, the combined effects of the two factors on endogenous hormones remain unclear, particularly on ABA. Plants growing in high planting canopies usually

Plant Growth and Leaf Gas Exchange
Two-way ANOVA showed that competition, nitrogen, and their interaction could significantly impact the growth indexes (Table 1). Compared with a single tomato plant being grown in one pot (P1), competition from neighbors (P4) did not affect shoot length under N15 at the end of our experiment (P = 0.193), but significantly decreased the shoot length under N1 (P = 0.004). However, P4 statistically reduced the stem diameter and leaf area of tomato plants under both N15 and N1 applications. Generally speaking, with the presence of neighbors, these indexes were much lower in the N1-treated plants than in N15-treated plants. Similarly, the dry biomass of competing plants was 38.5% and 25.3% lower than for plants grown singly under N15 and N1, respectively ( Figure S1). Data represents mean ± SE. * and ** within a column indicate a significant difference between treatments at the P < 0.05 or P < 0.01 level by two-way ANOVA, Duncan's multiple-range test, n = 6.
The net photosynthesis rate (Pn), stomatal conductance (gs) and transpiration rate (Tr) were significantly affected by competition and nitrogen application. Pn was suppressed by neighbor competition as compared to the singly grown plants (P = 0.002), and low nitrogen aggravated the suppression of Pn in grouped plants (P = 0.000) (Figure 2A). The response of gs and Tr to competitors was similar to Pn, and much lower under low N application ( Figure 2B, C). Intercellular CO2

Plant Growth and Leaf Gas Exchange
Two-way ANOVA showed that competition, nitrogen, and their interaction could significantly impact the growth indexes (Table 1). Compared with a single tomato plant being grown in one pot (P1), competition from neighbors (P4) did not affect shoot length under N15 at the end of our experiment (p = 0.193), but significantly decreased the shoot length under N1 (p = 0.004). However, P4 statistically reduced the stem diameter and leaf area of tomato plants under both N15 and N1 applications. Generally speaking, with the presence of neighbors, these indexes were much lower in the N1-treated plants than in N15-treated plants. Similarly, the dry biomass of competing plants was 38.5% and 25.3% lower than for plants grown singly under N15 and N1, respectively ( Figure S1). Data represents mean ± SE. * and ** within a column indicate a significant difference between treatments at the p < 0.05 or p < 0.01 level by two-way ANOVA, Duncan's multiple-range test, n = 6.
The net photosynthesis rate (P n ), stomatal conductance (g s ) and transpiration rate (T r ) were significantly affected by competition and nitrogen application. P n was suppressed by neighbor competition as compared to the singly grown plants (p = 0.002), and low nitrogen aggravated Plants 2020, 9,1674 4 of 14 the suppression of P n in grouped plants (p = 0.000) (Figure 2A). The response of g s and T r to competitors was similar to P n , and much lower under low N application ( Figure 2B,C). Intercellular CO 2 concentration (C i ) in competing plants significantly increased in comparison to single plants under N15 conditions (p = 0.022), but significantly decreased as compared to single plants under N1 treatment (p = 0.04) ( Figure 2D).  Figure 2D).

Hormonal Response
Competition and the interaction among two factors significantly affected the changes of ABA in shoots, roots and the ratio of shoot-to-root ABA ratio ( Table 2). As shown in Figure 3A, the presence of neighbors significantly increased ABA in shoots, which was almost three times higher in N15 application (P = 0.000) and 1.4 times higher in N1 application (P = 0.006), respectively. The concentration of shoot ABA in N15P4-treated plants was higher than in N1P4-treated plants (P = 0.009). Interestingly, the concentration of root ABA in N1-treated plants was higher than in N15treated plants. As a result, competition induced an increase in the shoot-to-root ABA ratio in N15treated plants (P = 0.014). In contrast, N shortage did not affect the ratio between single and grouped plants due to the remarkable root ABA accumulation ( Figure 3A). Changes of JA in shoots, roots, and the shoot-to-root ratio were significantly affected by the neighboring competition, nitrogen, and their interaction ( Table 2). Competition significantly increased the concentration of JA in shoots regardless of N treatments (P = 0.019 in N15 and P = 0.000 in N1, respectively), and soil N shortage resulted in almost 2.3-fold higher shoot JA as compared to that of a single plant ( Figure 3B). The concentrations of JA in the roots of grouped plants did not increase in N15-treated plants, but increased dramatically in N1-treated plants in response to neighbors (P = 0.001). Therefore, competition significantly increased the shoot-to-root JA ratio in N15 and N1 plants (P = 0.012 in N15 and P = 0.000 in N1, respectively), and the ratio of JA in N1 was lower than in N15-treated plants.

Hormonal Response
Competition and the interaction among two factors significantly affected the changes of ABA in shoots, roots and the ratio of shoot-to-root ABA ratio ( Table 2). As shown in Figure 3A, the presence of neighbors significantly increased ABA in shoots, which was almost three times higher in N15 application (p = 0.000) and 1.4 times higher in N1 application (p = 0.006), respectively. The concentration of shoot ABA in N15P4-treated plants was higher than in N1P4-treated plants (p = 0.009). Interestingly, the concentration of root ABA in N1-treated plants was higher than in N15-treated plants. As a result, competition induced an increase in the shoot-to-root ABA ratio in N15-treated plants (p = 0.014). In contrast, N shortage did not affect the ratio between single and grouped plants due to the remarkable root ABA accumulation ( Figure 3A). Changes of JA in shoots, roots, and the shoot-to-root ratio were significantly affected by the neighboring competition, nitrogen, and their interaction ( Table 2). Competition significantly increased the concentration of JA in shoots regardless of N treatments (p = 0.019 in N15 and p = 0.000 in N1, respectively), and soil N shortage resulted in almost 2.3-fold higher shoot JA as compared to that of a single plant ( Figure 3B). The concentrations of JA in the roots of grouped plants did not increase in N15-treated plants, but increased dramatically in N1-treated plants in response to neighbors (p = 0.001). Therefore, competition significantly increased the shoot-to-root JA ratio in N15 and N1 plants (p = 0.012 in N15 and p = 0.000 in N1, respectively), and the ratio of JA in N1 was lower than in N15-treated plants.

Correlations of gs, Pn, Hormones, and Leaf N Content
In N15-treated plants, gs negatively responded to the ABA in shoots, roots, and the shoot-to-root ratio, and the Pearson's correlation coefficient (R 2 ) increased in N1-treated plants (Table 3). Although there was no significant relationship between gs and root JA or shoot-to-root JA ratio in N15-treated plants, this relationship significantly increased in N1-treated plants; the P values were both equal to 0.000. The correlations of Pn with ABA and JA were significant in N15-treated plants, and R 2 was higher in N1 plants. gs decreased linearly with increasing shoot ABA concentration in response to neighboring competition in both N15 and N1 ( Figure 4A). The absolute value of slope of gs to ABA in N1-treated plants was higher than in N15-treated plants. gs also decreased with increasing shoot JA as compared to a single plant. However, the slope of gs to JA did not increase in N1-treated plants (Fig. 4B). gs increased linearly with the increase in leaf N content, and the R 2 was very high and significant at the P < 0.01 level, being 0.92 ( Figure 4C). Table 3. Relationships of gs and Pn to ABA and JA concentrations in shoots, roots and shoot-to-root ratio by Pearson's correlation coefficient analysis under different N applications.

Correlations of g s , P n , Hormones, and Leaf N Content
In N15-treated plants, g s negatively responded to the ABA in shoots, roots, and the shoot-to-root ratio, and the Pearson's correlation coefficient (R 2 ) increased in N1-treated plants ( Table 3). Although there was no significant relationship between g s and root JA or shoot-to-root JA ratio in N15-treated plants, this relationship significantly increased in N1-treated plants; the p values were both equal to 0.000. The correlations of P n with ABA and JA were significant in N15-treated plants, and R 2 was higher in N1 plants. g s decreased linearly with increasing shoot ABA concentration in response to neighboring competition in both N15 and N1 ( Figure 4A). The absolute value of slope of g s to ABA in N1-treated plants was higher than in N15-treated plants. g s also decreased with increasing shoot JA as compared to a single plant. However, the slope of g s to JA did not increase in N1-treated plants ( Figure 4B). g s increased linearly with the increase in leaf N content, and the R 2 was very high and significant at the p < 0.01 level, being 0.92 ( Figure 4C). Table 3. Relationships of g s and P n to ABA and JA concentrations in shoots, roots and shoot-to-root ratio by Pearson's correlation coefficient analysis under different N applications.

Correlations of Chlorophyll Index and Leaf nitrogen Content
The effects of competition and nitrogen on chlorophyll index (as measured by soil-plant analysis development, SPAD), were significant at the P < 0.01 level, whereas the interaction among these two factors was insignificant ( Table 4). The SPAD of grouped plants decreased in both the N15-and N1treated plants (Fig. 5A). SPAD in N1-treated plants was much lower than in N15. Competition did not significantly affect the contribution of stomatal limitation (Ls). In contrast, nitrogen and the interaction between these two factors could substantially affect Ls and the limit of non-stomatal conductance (Lns) ( Table 4). The Ls in grouped plants was lower than in single plants per pot under N15 conditions (P = 0.022), but Ls increased in N1P4 as compared with the N1P1 treatment ( Figure  5B) (P = 0.040). The Lns was much higher in N1P4 than in other treatments. Table 4. Effects of competition and nitrogen application on chlorophyll index (SPAD), the contribution of stomatal limitation (Ls) and non-stomatal limitation (Lns) by two-way variance analysis.

Correlations of Chlorophyll Index and Leaf nitrogen Content
The effects of competition and nitrogen on chlorophyll index (as measured by soil-plant analysis development, SPAD), were significant at the p < 0.01 level, whereas the interaction among these two factors was insignificant ( Table 4). The SPAD of grouped plants decreased in both the N15-and N1-treated plants ( Figure 5A). SPAD in N1-treated plants was much lower than in N15. Competition did not significantly affect the contribution of stomatal limitation (L s ). In contrast, nitrogen and the interaction between these two factors could substantially affect L s and the limit of non-stomatal conductance (L ns ) ( Table 4). The L s in grouped plants was lower than in single plants per pot under N15 conditions (p = 0.022), but L s increased in N1P4 as compared with the N1P1 treatment ( Figure 5B) (p = 0.040). The L ns was much higher in N1P4 than in other treatments. Table 4. Effects of competition and nitrogen application on chlorophyll index (SPAD), the contribution of stomatal limitation (L s ) and non-stomatal limitation (L ns ) by two-way variance analysis.

Correlations of Chlorophyll Index and Leaf nitrogen Content
The effects of competition and nitrogen on chlorophyll index (as measured by soil-plant analysis development, SPAD), were significant at the P < 0.01 level, whereas the interaction among these two factors was insignificant ( Table 4). The SPAD of grouped plants decreased in both the N15-and N1treated plants (Fig. 5A). SPAD in N1-treated plants was much lower than in N15. Competition did not significantly affect the contribution of stomatal limitation (Ls). In contrast, nitrogen and the interaction between these two factors could substantially affect Ls and the limit of non-stomatal conductance (Lns) ( Table 4). The Ls in grouped plants was lower than in single plants per pot under N15 conditions (P = 0.022), but Ls increased in N1P4 as compared with the N1P1 treatment ( Figure  5B) (P = 0.040). The Lns was much higher in N1P4 than in other treatments. Table 4. Effects of competition and nitrogen application on chlorophyll index (SPAD), the contribution of stomatal limitation (Ls) and non-stomatal limitation (Lns) by two-way variance analysis.

SPAD Ls L ns
Competition (C) 1 0.000 ** 0.959 ns 0.010 ** Nitrate application (N) 1 0.000 ** 0.001 ** 0.000 ** C × N 1 0.907 ns 0.002 ** 0.258 * Error 20 * represents significant difference at the P < 0.05 level; ** represents significant difference at the P < 0.01 level; ns represents no significant difference between treatments by wo-way ANOVA, Duncan's multiple-range test, n = 6. SPAD decreased linearly with the decreasing leaf N concentration in response to neighbors in both N15 and N1, and the slope was higher in N1 than in N15 ( Figure 6A). The relationship of P n to SPAD was similar to that of SPAD to leaf N concentration, and the slope increased in N1 as compared to N15 ( Figure 6B). SPAD decreased linearly with the decreasing leaf N concentration in response to neighbors in both N15 and N1, and the slope was higher in N1 than in N15 ( Figure 6A). The relationship of Pn to SPAD was similar to that of SPAD to leaf N concentration, and the slope increased in N1 as compared to N15 ( Figure 6B).

Discussion
There is a common consensus that plants increase the length of the stem and petiole to gain more light intensity in response to the presence of neighbors [11]. Likewise, in this study, the shoot length of grouped plants did not decrease in sufficient N supply condition (Table 1). These results suggested that plants might suffer from neighboring competition under an elevated planting density. However, N deficiency seriously impeded plant growth, especially for grouped plants.

Effects of Neighbor Competing and Nitrogen Availability on Leaf Gas Exchange and Abscisic Acid (ABA)
In competing plants gs and Tr decreased, but the shoot water potential was unaffected as all plants were well-watered (almost 70-75% total sand moisture capacity, TMC) during the experiment, so the leaf hydraulic signals might not drive the closure of stomata. Nevertheless, there was a negative relationship between gs and ABA (Table 3). Previous studies have reported that competitioninduced stomatal closure was associated with the accumulation of ABA in shoots due to the increased transport of ABA from root to shoot under the increased planting density [6,29,30]. Our data also supported this view because more ABA accumulated in shoots than in roots in competing plants, resulting in a 1.7-fold increase in the shoot-to-root ABA ratio ( Figure 3A). The importance of root-toshoot ABA transport was supported by data showing that enhanced root-to-shoot ABA transport decreased gs in ABA-deficient flacca mutant scions grafted with WT rootstock compared to flacca rootstocks [30]. Experiments with fluridone, an inhibitor of ABA transport, showed that the absence of ABA accumulation in shoots prevented growth inhibition in competing plants [31]. However, the shoot-to-root ABA ratio did not increase in response to neighbors under N deficiency conditions and this ratio was much lower in N1-treated plants than in N15-treated plants, implying that low N availability was able to decrease ABA transport from root to shoot ( Figure 3A). One of the reasons

Discussion
There is a common consensus that plants increase the length of the stem and petiole to gain more light intensity in response to the presence of neighbors [11]. Likewise, in this study, the shoot length of grouped plants did not decrease in sufficient N supply condition (Table 1). These results suggested that plants might suffer from neighboring competition under an elevated planting density. However, N deficiency seriously impeded plant growth, especially for grouped plants.

Effects of Neighbor Competing and Nitrogen Availability on Leaf Gas Exchange and Abscisic Acid (ABA)
In competing plants g s and T r decreased, but the shoot water potential was unaffected as all plants were well-watered (almost 70-75% total sand moisture capacity, TMC) during the experiment, so the leaf hydraulic signals might not drive the closure of stomata. Nevertheless, there was a negative relationship between g s and ABA (Table 3). Previous studies have reported that competition-induced stomatal closure was associated with the accumulation of ABA in shoots due to the increased transport of ABA from root to shoot under the increased planting density [6,29,30]. Our data also supported this view because more ABA accumulated in shoots than in roots in competing plants, resulting in a 1.7-fold increase in the shoot-to-root ABA ratio ( Figure 3A). The importance of root-to-shoot ABA transport was supported by data showing that enhanced root-to-shoot ABA transport decreased g s in ABA-deficient flacca mutant scions grafted with WT rootstock compared to flacca rootstocks [30]. Experiments with fluridone, an inhibitor of ABA transport, showed that the absence of ABA accumulation in shoots prevented growth inhibition in competing plants [31]. However, the shoot-to-root ABA ratio did not increase in response to neighbors under N deficiency conditions and this ratio was much lower in N1-treated plants than in N15-treated plants, implying that low N availability was able to decrease ABA transport from root to shoot ( Figure 3A). One of the reasons might be the lower T r in N1-treated plants ( Figure 2C). Transpiration pull was considered to be the main power for plants to transport water, hormones and other nutrients from the roots [32]. N supply was expected to have a major effect on leaf gas exchange, particularly on g s and T r [33,34]. Another reason may be more ABA being produced in the roots under N deficiency conditions. The results of more ABA accumulating in N-deprived plants were in agreement with the data of Oka et al. [35], who showed that low nitrogen availability promoted ABA accumulation. N deficiency has been shown to upregulate zeaxanthin epoxidase (TaZEP), a key ABA synthesis-related enzyme and increase the endogenous ABA content [36]. Interestingly, with no increase in ABA export from roots, g s in nitrogen-deficient plants showed a fast decrease in response to the accumulation of ABA in shoots of competing plants ( Figure 4A). Moreover, the relationship between g s and ABA in shoots and roots, and the shoot-to-root ratio of N1-treated plants, was higher than that of N15-treated plants (Table 3). Thus, it is plausible that different mechanisms might exist in the response of g s to competitors under low nitrogen conditions. One of the reasons for the high sensitivity of g s to ABA under N deficiency might be that N deprivation alkalized the xylem sap, a moderator of the ABA signaling process [29]. The increased planting density increased the pH of the xylem sap and soil solution as more nitrate was absorbed from the same volume of soil [6,37]. Furthermore, it has been shown that g s decreased with decreasing leaf N content [26,38,39]. A positive relationship between g s and leaf N content was also observed in this study ( Figure 4C), which suggested that N might have a direct influence on the stomatal movement (Figure 7). Although the mechanisms of N in the regulation of g s are not investigated in depth in this study, it is plausible that N deficiency sensitizes g s to ABA in response to neighbor competition (Figure 7). might be the lower Tr in N1-treated plants ( Figure 2C). Transpiration pull was considered to be the main power for plants to transport water, hormones and other nutrients from the roots [32]. N supply was expected to have a major effect on leaf gas exchange, particularly on gs and Tr [33,34]. Another reason may be more ABA being produced in the roots under N deficiency conditions. The results of more ABA accumulating in N-deprived plants were in agreement with the data of Oka et al. [35], who showed that low nitrogen availability promoted ABA accumulation. N deficiency has been shown to upregulate zeaxanthin epoxidase (TaZEP), a key ABA synthesis-related enzyme and increase the endogenous ABA content [36]. Interestingly, with no increase in ABA export from roots, gs in nitrogen-deficient plants showed a fast decrease in response to the accumulation of ABA in shoots of competing plants ( Figure 4A). Moreover, the relationship between gs and ABA in shoots and roots, and the shoot-to-root ratio of N1-treated plants, was higher than that of N15-treated plants (Table 3). Thus, it is plausible that different mechanisms might exist in the response of gs to competitors under low nitrogen conditions. One of the reasons for the high sensitivity of gs to ABA under N deficiency might be that N deprivation alkalized the xylem sap, a moderator of the ABA signaling process [29]. The increased planting density increased the pH of the xylem sap and soil solution as more nitrate was absorbed from the same volume of soil [6,37]. Furthermore, it has been shown that gs decreased with decreasing leaf N content [26,38,39]. A positive relationship between gs and leaf N content was also observed in this study ( Figure 4C), which suggested that N might have a direct influence on the stomatal movement (Figure 7). Although the mechanisms of N in the regulation of gs are not investigated in depth in this study, it is plausible that N deficiency sensitizes gs to ABA in response to neighbor competition (Figure 7).

Effects of Neighbor Competing and Nitrogen Availability on Leaf Gas Exchange and Jasmonic Acid (JA)
To understand the role of JA in response to neighbor competition, changes of JA in shoots and roots were also investigated in our experiment. We found that the concentration of JA in shoots increased in competing plants ( Figure 3B). The data presented here were in agreement with previous studies showing that the concentration of JA in shoots was able to increase in response to water stress [17,40]. Shoot-to-root JA ratio increased simultaneously in grouped plants, implying that neighbors' presence enabled increased JA transport from root to shoot (Figure 7). The importance of root to shoot JA transport has been stressed by De Ollas et al. [19], who found that attenuated root to shoot JA transport improved the leaf water status under soil drying conditions. However, there is a knowledge gap about the mechanisms involved in the control of JA allocation under increasing density. Future

Effects of Neighbor Competing and Nitrogen Availability on Leaf Gas Exchange and Jasmonic Acid (JA)
To understand the role of JA in response to neighbor competition, changes of JA in shoots and roots were also investigated in our experiment. We found that the concentration of JA in shoots increased in competing plants ( Figure 3B). The data presented here were in agreement with previous studies showing that the concentration of JA in shoots was able to increase in response to water stress [17,40]. Shoot-to-root JA ratio increased simultaneously in grouped plants, implying that neighbors' presence enabled increased JA transport from root to shoot (Figure 7). The importance of root to shoot JA transport has been stressed by De Ollas et al. [19], who found that attenuated root to shoot JA transport improved the leaf water status under soil drying conditions. However, there is a knowledge gap about the mechanisms involved in the control of JA allocation under increasing density. Future research should focus on the effects of neighboring competition on the root to shoot JA transport. By contrast, some studies showed that low R/FR ratio light, an early warning sign of crowding, would inhibit the expression of JA-induced genes and promote disease symptoms in Arabidopsis [41] and tomatoes [42]. Although we did not measure the effect of JA on plant disease symptoms, different results might be attributable to different plant materials, cultivars and growth conditions. At the same time, the present Plants 2020, 9, 1674 9 of 14 work showed that the g s of the grouped plants decreased with increasing shoot JA, indicating that g s negatively responded to JA (Table 3, Figure 4B). Previous studies have shown that JA-induced stomatal closure was closely related to cytoplasmic alkalinization in guard cells, the production of reactive oxygen species (ROS) and NO, activation of K + -efflux, and slowing anion channels [43][44][45]. Therefore, the role of JA in inducing stomatal closure response to neighbors was similar to that of ABA, indicating that there might be a synergistic rather than an antagonistic interaction between ABA and JA in regulating stomatal movement in response to biotic stress (Figure 7). Suhita et al. [39] reported that the stomata of ABA-insensitive mutants were able to close in response to the exogenous application of JA. In addition, the sensitivity of g s to drying decreased in def-1 self-grafts (which fail to accumulate JA in shoots during drying) as compared to def-1 scions grafted onto WT rootstocks [19]. Since the role of JA in regulating ABA biosynthesis has been addressed [21], the coordinated action of ABA and JA may be due to their converged signal transduction pathways at the level of intracellular Ca 2+ modulation [45]. However, our data cannot fully explain this interaction. In order to reveal the interaction of ABA and JA, grafted mutant plants should be used in the future. Moreover, as shown in Table 3, tomato seedlings displayed a negative correlation between JA in shoots and P n , such as −0.791 and −0.977 in N15 and N1 conditions, respectively. The negative role of JA on photosynthesis might be due to the down-regulation of the rubisco enzyme by JA [46]. These results suggest the important role of JA accumulation in the modulation of leaf gas exchange to adapt to competitors. N deficiency decreased the concentration of JA in shoots and shoot-to-root ratio as compared to the N15 treatment ( Figure 3B), in agreement with the findings of Jang et al. [47], who found that elevated N application increased JA concentrations in shoots of rice cultivars as compared with no N supply. However, unlike the g s to shoot ABA, the slope of g s to shoot JA of grouped plants under N deficiency did not change as compared to in N-sufficient conditions ( Figure 4B), suggesting that N would not change the sensitivity of g s to JA in response to neighbors. Although the insensitivity of g s to JA in the face of N shortage was not investigated in depth in this study, grouped plants had the lowest g s under N deficiency conditions.

Effects of Neighbor Competing and Nitrogen Availability on P n
Under the N sufficiency condition, the limitation of g s to photosynthesis of grouped plants was lower than of plants grown singly. However, grouped plants grown in N deficiency conditions showed higher L s values. Thus, one reason for the decline in photosynthesis of grouped plants in N insufficiency might be the closure of stomata due to the low CO 2 supply for carboxylation [48]. The N1-treated plants had a higher L ns value, especially for grouped plants, indicating that photosynthesis was simultaneously limited by non-stomatal factors, such as the significant decrease of SPAD accompanied by a decline in leaf N content in this study ( Figure 6A). Moreover, the slope of SPAD to P n in N1 was higher than in N15-treated plants, suggesting that P n in N-deficiency plants was more sensitive to SPAD ( Figure 6B). Our findings were consistent with those of other studies [26,49], which found that the N supply had a positive effect on SPAD values, as N was a major source for chlorophyll synthesis [50]. Hence, it was not surprising that plant growth parameters such as the stem diameter and dry biomass of group plants were severely inhibited by low nitrogen availability (Table 1, Figure S1).

Plant Materials and Experimental Design
The indoor experiment was conducted in a controlled environment room at the experimental station of the Institute of Farmland Irrigation (35 • 18 N, 113 • 54 E; 81 m altitude). Tomato seeds (Solanum lycopersicum L., cv. Helan108) were germinated in a nursery seedling plate with substrate (sphagnum peat, Pindstrup Mosebrug A/S, Ryomgaard, Denmark). When tomato seedlings were 16 days old, those with two true leaves were transplanted into pots (height 15 cm, diameter 16 cm) and grown in a controlled environment room with a 14-h photoperiod (6:00 to 20:00), provided by helium and sodium lamps with 400 µmol·m −2 ·s −1 photosynthetic active radiation (PAR) at 25-30 • C with 45-55% relative air humidity. Each pot was filled with 3.5 kg sterile sand. Two experimental factors with two levels (i.e., planting density and nitrogen rate) were designed in this study. The planting density levels were maintained by transferring a single plant (P1) or four plants (P4) into one pot; the two levels of nitrogen rate were 1 mmol·L −1 (N1) and 15 mmol·L −1 (N15), respectively. Chloride was used as a counter-ion to regulate the K + in the N1 solution. KOH solution was used to adjust the pH to 6.0 ± 0.2 of both in the N15 and N1 solutions. After transplanting, each pot was watered with 600 mL of 1/2 complete Hoagland solution (1 and 15 mmol·L −1 , respectively). From 2 DAT (days after transplanting, DAT) up to harvest at 14 DAT, all pots were weighed in the morning and irrigated with 20 mL half-strength Hoagland solution depending on the nitrogen treatments, with additional distilled water to reach a level of 70-75% total sand moisture capacity (TMC). To prevent soil evaporation, pots were covered with aluminum foil.

Measurement of Plant Development, SPAD and Leaf N Content
At 14 DAT, plant growth parameters like shoot length, stem diameter, biomass and leaf area were measured with six replications of each treatment. The shoot length of tomato seedlings was measured using a ruler. A vernier caliper was used to measure the stem diameter 1 cm from cotyledon. Leaf area was recorded by the LI-3000C Area Meter (LI-COR Inc., Lincoln, NE, USA). The middle part of the fully expanded leaf was chosen to measure the chlorophyll index (SPAD) using a Minolta SPAD 502 chlorophyll meter (Minolta Camera Co., Ltd., Osaka, Japan). The dry weight was obtained after oven drying at 75 • C for 24 h. Leaf N concentrations were determined with a continuous flow autoanalyzer (Bran + Luebbe, Hamburg, Germany).

Determination of Leaf Gas Exchange Parameters
Leaf gas exchange parameters, including P n , g s , T r and the intercellular CO 2 concentration (C i ), were measured, with six replicates, on the same leaves used for SPAD measurement, using a LI-6400XT portable photosynthesis system (LI-COR Inc. Lincoln, NE, USA) equipped with a red and blue light source chamber. Each measurement was taken between 9:00 and 13:00. On the leaf chamber, PAR was set to 400 µmol·m −2 ·s −1 , which was near the environmental light intensity; the sample CO 2 concentration was maintained at 400 µmol·mol −1 using a CO 2 mixer. The block leaf temperature was 28 • C.
The limitation of g s to photosynthesis was calculated as follows, L s = 1 − C i /C a [51], and the limit of non-stomatal conductance was then calculated as L ns = C i /g s , where C a is the ambient CO 2 concentration (C a = 400 µmol·m −2 ·s −1 ).

Root and Shoot Sap Collection
After the measurement of leaf gas exchange, tomato seedlings were cut off at 1-2 cm above the sand surface. Then, root sap was collected from the cut surface and placed into a centrifuge tube in liquid nitrogen. Plant shoots were placed into the pressure chamber Model 3115 Pressure Extractor (Soil Moisture Equipment Corp., Santa Barbara, CA, USA) with the stem stump outside the chamber. After cleaning the cut surface with pure water, the pressure was increased slowly, and we noted the pressure when the first drop of solution occurred. The negative pressure was regarded as the shoot water potential. To collect the sap solution, 0.3-0.5 MPa pressure was continuously applied to the shoot. Then, 2-mL samples of sap collected from some tomatoes of each treatment were immediately placed into centrifuge tubes in liquid nitrogen. Sap solutions were stored at −80 • C for hormone analysis.

ABA and JA Measurement
ABA and JA concentrations were determined according to the method described by You et al. [52] with little modification. First, 2 mL of extracted sap sample were placed into a microcentrifuge tube containing 5 mL extraction buffer composed of isopropanol/hydrochloric acid and 8 µL 1 µg mL −1 deuterated internal standard solutions. Then, 10 mL dichloromethane were added and sap samples were shaken at 4 • C for 30 min and then centrifuged at 13,000× g for 5 min at 4 • C. The supernatant was carefully removed and the lower organic phase was dried by N 2 in shade conditions and dissolved in 400 µL methanol (containing 0.1% methane acid), then filtered with a 0.22-mm filter membrane. The purified sample was then subjected to high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS), fitted with a POROSHELL120 SB-C18 (Agilent Technologies Inc., Santa Clara, CA, USA) column (2.1 mm × 150 mm; 2.7 mm), at 30 • C. The solvent gradient used was 100% A (99.9% methanol: 0.1% CHOOH) to 100% B (99.9% H 2 O: 0.1% CHOOH) over 15 min. The injection volume was 2 µL. MS conditions were as follows: the spray voltage was 4500 V; the pressure of the air curtain, nebulizer, and aux gas were 15, 65, and 70 psi, respectively; and the atomizing temperature was 400 • C. The ABA and JA concentrations were calculated with reference to the peak area of the deuterated internal standard.

Statistical Analysis
Statistical differences among different treatments were assessed by two-way analysis of variance using SPSS 16.0 (IBM SPSS Statistics, Chicago, IL, USA). Multiple comparisons between all treatments were analyzed using a Duncan's test. All data are presented as the mean ± standard error (SE). Different letters were considered significant at the p < 0.05 and p < 0.01 level. Data fitting and graphical presentation were carried out in Origin-Pro 2017 (Origin Lab, Northampton, MA, USA). The general linear regression model was used to fit the relationships between parameters under different N application levels.

Conclusions
Endogenous JA in the shoots and shoot-to-root ratio are affected by the presence of neighbors, JA works synergistically with ABA in the regulation of stomatal movement and carbon fixation under elevated planting density. Negative correlations between JA or ABA and both g s and P n increase under N deficiency conditions. At the same time, a soil nutrient shortage somewhat increased the sensitivity of g s to ABA in response to neighbors, whereas N shortage does not increase the g s to JA in response to neighbors. Stomatal limitations and non-stomatal limitations contribute to the decrease of P n in N shortage conditions. Future studies should focus on how neighbors would affect the root to shoot JA transport or the interaction between ABA and JA. However, our results shed new light on the important role of ABA, JA, and their orchestrated signaling in the regulation of plant responses to neighbor competition in a complex growth environment.