From July 15th to 30th, 2020, the leaf assimilation rate (An, μmol m^-2 s^-1) was recorded using a CIRAS-3 portable photosynthesis system (PP Systems, USA). The measurements were taken at a stable temperature of 25°C, with a CO₂ concentration set at 400 μmol mol^-1, a flow rate of 500 μmol s^-1, and a photosynthetic photon flux density (PPFD) of 1600 μmol m^-2 s^-1 within the leaf chamber. For each plant, the gas exchange in leaves was gauged on the second and third leaves from the shoot apex, with the recordings done between 8:00 am and 4:00 pm, ensuring six replicates for consistency.

From July 15th to 30th, the dynamics of the chlorophyll a O-J-I-P kinetic curve were measured using a Handy-PEA continuous excitation chlorophyll fluorimeter from Hansatech Instruments, adhering to Strasser’s method and conducted at ambient temperatures (Strasser and Govindjee, 1992; Strasser et al., 2000). Initially, leaf samples underwent a 30-minute dark adaptation phase, using clips to avoid the midrib, ensuring uniform darkening. The rapid fluorescence kinetics were tracked from 10 ms to 1 s, with FO, FJ, FI, and FP denoting fluorescence intensities at 20 ms, 2 ms, 30 ms, and 300 ms, respectively. The K peak at 300 ms on the kinetic curve was particularly noted (Strasser, 1997; Strasser et al., 2000; Ronde et al., 2004; Strasser et al., 2004; Li et al., 2009). PI _abs, indicative of energy conservation efficiency from excitation to electron acceptor reduction, was calculated (Strasser et al., 2004; Strauss et al., 2006). A comprehensive fluorescence kinetics analysis was conducted, entailing various normalization and kinetic differential calculations, with key parameters and formulas detailed in Table 1 (Strasser et al., 2000; Cun et al., 2022) (n = 6).

Further, the effective quantum yield of PSII (φPSII), the non-photochemical quenching coefficient (NPQ), and the electron transport rate (ETR, μmol e^-1 s^-1 m^-2) were measured using an IMAGING PAM M-series from Walz, with a 30-minute dark acclimation preceding the measurements to assess the PSII quantum efficiency of the plants (n = 6).

Following the measurement of chlorophyll fluorescence parameters, two leaves from each plant were harvested, instantly frozen in liquid nitrogen, and preserved at -80°C for subsequent biochemical analyses. Another two leaves were heat-fixed at 105°C for 30 minutes and then dehydrated to constant mass at 65°C for the determination of leaf dry weight and leaf mass per unit area (LMA, g m^-2) (n = 6).

Accurately weigh a certain amount of ground plant samples (2-3 mg) into tin foil cups, tightly wrap them, record the mass, and place them in the automatic sample introduction tray of the instrument for analysis using a stable isotope ratio mass spectrometer (Elementar, Isoprime 100, UK) to determine the N content (N_m, g/kg) in the leaves, stems, and roots. Leaf N content per unit leaf area (N_area, g m^-2) was calculated by N_m × LMA (n = 6).

Chlorophyll a and chlorophyll b concentrations were quantified from 0.1 g of leaf tissue in ethanol extracts, with absorbance readings taken at 645 nm and 663 nm using a UVmini-1240 spectrophotometer (Shimadzu, Japan) (Wellburn, 1994; Wei et al., 2022). The total chlorophyll content per unit area (Chl_area, mg m^-2) was then calculated by multiplying chlorophyll a+b with the leaf mass per unit area (LMA) (n = 6).

The content of reactive oxygen species (ROS) and hydrogen peroxide (H₂O₂), were determined by corresponding ELISA kits (Shanghai Jining Biology Co., Ltd., China). Conduct sample processing according to the instructions of the kits, the concentrations were measured using an enzyme reader (Bio-imark, BIO-RAD, USA) at the corresponding wavelengths (ROS at 450 nm and H₂O₂ at 240 nm).

The enzyme activities of ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco), the reduced form of nicotinamide adenine dinucleotide phosphate (NADPH), alternative oxidase (AOX), and ascorbate peroxidase (APX) were assessed using specific ELISA kits provided by Shanghai Jining Biology Co., Ltd., China. The sample preparation was executed in strict adherence to the kit protocols. The concentrations were then quantified using an enzyme reader (Bio-imark, BIO-RAD, USA), with absorbance measurements taken at the respective wavelengths for each enzyme (Rubisco at 340 nm, NADPH at 450 nm, AOX at 450 nm, and APX at 290 nm).

Data processing and visual analysis were conducted using R software version 4.0.4 (RStudio, USA, [https://www.r-project.org/]) (R Core Team, 2021). To assess significant differences among treatment groups, the “Fisher’s LSD” function in the “agricolae” package was applied (P< 0.05). For correlation analysis, the “pearson” function in the “gpairs” and “ggpmisc” packages was utilized, and the “ggplot2” package was employed for creating graphics. The statistical tests included one-way ANOVA to evaluate the impact of various N forms and drought stress on plant parameters, and two-way ANOVA for investigating the effects of nitrogen forms (N), drought stress (D), and their interaction. The Pearson correlation coefficient was used to explore relationships between different plant parameters. All datasets were initially tested for normal distribution using the Kolmogorov-Smirnov test and for variance homogeneity using Levene’s test.

Significant variations were observed in chlorophyll content due to the combined effects of different N forms and drought stress. Under moderate (MD) and heavy drought (HD) conditions, chlorophyll a (Chla) and chlorophyll b (Chlb) contents in L. chinensis leaves were notably higher in plants treated with nitrate (NO₃) and a mix of ammonium nitrate (NH₄NO₃) compared to those treated with no nitrogen (N0) and ammonium (NH₄) ( Figures 1A, B , p< 0.05). Leaf mass per unit area (LMA) showed no significant difference between NO₃ and NH₄NO₃ plants under HD, but it was significantly greater than in N0 plants ( Figure 1C , p< 0.05). Under light drought (LD), the chlorophyll content per unit area (Chl_area) in leaves was significantly elevated in NO₃ plants relative to N0, NH₄, and NH₄NO₃ plants (p< 0.05), showing increases of 59.46%, 26.87%, and 19.21%, respectively. Under MD and HD conditions, Chl_area remained significantly higher in NO₃ and NH₄NO₃ plants compared to N0 and NH₄ plants ( Figure 1D , p< 0.05). Under HD, nitrogen content per unit area (N_area) in NO₃ plants was 73.61% and 25.70% higher than in N0 and NH₄ plants, respectively ( Figure 1E , p< 0.05). Across all drought treatments (LD, MD, HD), the leaf assimilation rate (An) was significantly elevated in NO₃-treated plants compared to N0, NH₄, and NH₄NO₃ treatments ( Figure 1F , p< 0.05). Particularly under HD, An in NO₃-treated plants was 111.49%, 44.82%, and 8.84% higher than in N0, NH₄, and NH₄NO₃-treated plants, respectively.

Drought stress triggers an excessive build-up of reactive oxygen species (ROS) and hydrogen peroxide (H₂O₂) in L. chinensis leaves. Under moderate (MD) and heavy drought (HD) conditions, the ROS levels in plants treated with ammonium (NH₄), nitrate (NO₃), and a combination of both (NH₄NO₃) were considerably lower than those in plants without nitrogen supplementation (N0). The highest ROS accumulation, reaching 364.75 μmol L^-1, was observed in N0-treated L. chinensis leaves ( Figure 2A , p< 0.05). As the severity of drought stress increases, so does the accumulation of H₂O₂. Under light drought (LD), the H₂O₂ content in NO₃-treated plants was just 10.39 U ml^-1. However, under MD and HD conditions, the H₂O₂ levels in N0 plants rose to 27.31 U ml^-1, significantly surpassing those in NH₄, NO₃, and NH₄NO₃ plants ( Figure 2B , p< 0.05).

Different N forms and drought stress exert a significant effect on the potential activity of PSII (Fv/Fo) and the maximum photochemical efficiency of photosystem II (Fv/Fm) (p< 0.01). Both Fv/Fo and Fv/Fm values declined in leaves subjected to MD and HD drought conditions ( Figure 3 ). In NH₄NO₃-treated plants, Fv/Fo decreased by 16% under MD and by 27.7% under HD-treated plants compared to LD-treated plants ( Figure 3A ; p< 0.05). For NO₃-grown plants, Fv/Fm experienced a reduction of 8.2% under MD and 28.6% under HD ( Figure 3B ), while in NH₄NO₃-treated plants, the decreases were 6.7% and 22.73% under MD and HD, respectively ( Figure 3B ).

Significant variations were observed in the effective quantum yield of PSII (φPSII), the electron transport rate of PSII (ETR), and the non-photochemical quenching coefficient (NPQ) in plants subjected to moderate (MD) and heavy drought (HD) treatments ( Figure 4 , p< 0.05). Both φPSII and ETR were considerably higher in plants grown with nitrate (NO₃) and a mixture of ammonium nitrate (NH₄NO₃) under MD and HD conditions ( Figures 4A, B ; p< 0.05). In NO₃ and NH₄NO₃ plants under HD stress, φPSII values decreased by 40.2% and 55.4%, respectively ( Figure 4A ). Similarly, ETR values in NO₃ and NH₄NO₃ plants under HD diminished by 51.8% and 52.1%, respectively ( Figure 4B ). However, HD-treated plants showed no significant change in NPQ when treated with NH₄, NO₃, and NH₄NO₃, with reductions of 55.5%, 52.8%, and 55.5% respectively ( Figure 4C ).

Under moderate (MD) and heavy drought (HD) treatments, the ΔV_OJ and ΔV_OI values in L. chinensis leaves treated with no nitrogen (N0) increased, with ΔV_IP also rising under HD conditions ( Figure 5A ). In leaves treated with ammonium (NH₄), ΔV_OI exhibited a decreasing trend under MD treatment ( Figure 5B ). Drought stress led to noticeable relative disaggregation in the antenna complex within the L-band of L. chinensis leaves ( Figures 5E -a, F -a). Under drought, leaves showed positive differences in the K-band curve, with similar peak values under both MD and HD conditions ( Figures 5E -b, F -b). The H-band under HD treatment indicated positive differences, while it shifted from positive to negative under MD treatment ( Figure 5E -c). Positive differences were observed in the G-band for both MD and HD treatments, with higher peak values under MD ( Figure 5E -d). Under HD treatment, the H-band displayed positive differences, whereas under MD treatment, it showed negative differences ( Figure 5F -c). The G-band under HD treatment had positive differences, but under MD treatment, the peak values were negatively different ( Figure 5F -d). For leaves treated with NO₃ and NH₄NO₃, ΔV_OJ, ΔV_OK, and ΔV_OI showed increasing trends during the O-J-I-P transition under HD treatment, and ΔV_IP also increased ( Figures 5C, D ). Both HD and MD treatments caused positive differences in the L-band, with higher peak values under HD ( Figures 5G -a, H -a). Under drought, NO₃-treated leaves displayed positive differences in the K-band ( Figure 5G -b), while NH₄NO₃-treated leaves also showed positive differences ( Figure 5H -b). In NO₃-treated leaves, the H-band had negative differences under drought ( Figure 5G -c), but in NH₄NO₃-treated leaves, it showed positive differences ( Figure 5H -c). The early peak of the G-band under HD treatment exhibited positive differences, with later peak amplitude decreasing ( Figure 5G -d), whereas the G-band in NH₄NO₃-treated leaves tended toward positive differences ( Figure 5H -d).

Further analysis of the JIP-test parameters revealed that drought stress significantly impacted the photosynthetic characteristics of PSII after the application of NO₃ and NH₄NO₃ ( Figure 6 ; p< 0.05; Supplementary Figure 1 ). S_m (indicating multiple-turnover Q_A reduction events), V _j (relative variable fluorescence at the J-step), and REo/CSo (Reduction of end acceptors at PSI electron acceptor side per CS at t=0) decreased in N0 and NH₄-grown plants treated with MD and HD ( Figures 6A, B ). However, V _j (relative variable fluorescence at the J-step) showed no significant difference in plants treated with different N forms under HD ( Figure 6C ). PI _abs (performance index for energy conservation from photons absorbed by PSII antenna to the reduction of Q_B) decreased in N0-grown plants treated with LD and HD ( Figures 6D ). NO₃-grown plants treated with HD exhibited significantly higher S_m and REo/CSo ( Figures 6A, B ; p< 0.05).

The correlation analysis indicated a strong correlation between PI _abs and An with respect to N forms. To further evaluate the N forms supplied to L. chinensis, a model based on the parameters PI _abs and An was developed. The PI _abs values increased linearly with the An level, revealing a significant positive linear correlation under NO₃ treatment (R^2 = 0.86) ( Figure 6E ). To elucidate the improvement mechanism of photosynthesis in NO₃-grown plants treated with HD, the relationship between REo/CSo and PI _abs was analyzed. The correlation analyses revealed highly active relationships between REo/CSo and PI _abs in NO₃ and NH₄NO₃-grown plants ( Figure 6F ). This suggests that the up-regulation of PSII overall activity is primarily due to the improved energetic connectivity of PSII units and OEC state, as well as the increased PSII electron transport efficiency in NO₃-treated plants under drought stress. Clearly, NO₃ ^- contributes to L. chinensis plants coping with drought stress by enhancing photosynthetic capacity.

The phenomenological models of energy fluxes through the cross sections (CS) of the leaves of L. chinensis under different N forms and drought stress are presented in Figure 7 and Supplementary Figure 2 . Deprivation of N0 or NH₄ significantly diminished the energy fluxes in the plants compared to those supplied with NO₃ or NH₄NO₃ under drought stress. Additionally, NO₃ or NH₄NO₃ resulted in a significant decrease in the energy absorption by a cross section of the leaves (ABS/CS), energy trapping (TR/CS), the electron transport flux (ET/CS), and energy dissipation (DI/CS) under HD compared with LD treated, but the decrease in all the aforementioned parameters was significantly lower than those observed under N0 and NH₄ supplied ( Figures 7C, F, I, L ). Under HD treated, the TRo/RC increased in NO₃ and NH₄NO₃ treatments, with TRo/RC changes resembling those of ABS/RC; ETo/RC increased in N0 and NO₃ treatments and decreased in NH₄ and NH₄NO₃ treatments; DIo/RC decreased in NH₄ treatment, and significantly increased in NO₃ and NH₄NO₃ treatments ( Supplementary Figures 2C, F, I, L ).

The interaction between different N forms and drought stress profoundly influences the activities of Rubisco, NADPH, and AOX. Notably, under HD treatment, Rubisco activity in NO₃ plants exhibited a remarkable increase, surpassing N0, NH₄, and NH₄NO₃ plants by 51.27%, 31.01%, and 29.64%, respectively ( Figure 8A , p< 0.05). Moreover, during MD and HD treatments, NADPH and APX activities in NH₄, NO₃, and NH₄NO₃ plants, while not significantly different from each other, were significantly higher compared to N0 plants ( Figures 8B, C , p< 0.05). Interestingly, under HD treatment, AOX activity in NO₃ and NH₄NO₃ treated plants significantly decreased that in N0 and NH₄ plants ( Figure 8D , p< 0.05).

N plays a pivotal role in shaping key components like Rubisco, chlorophyll, and photosynthetic proteins, underscoring its significance in plant photosynthesis (Li et al., 2013; Mu and Chen, 2021). The efficient absorption and utilization of N can profoundly impact the photosynthetic N use efficiency of plants, thereby influencing overall photosynthetic efficiency (Evans and Clarke, 2019). Existing evidence indicates a positive correlation between the photosynthetic efficiency of L. chinensis and increasing photosynthetic N use efficiency (PNUE) (Wei et al., 2022). Photosynthesis is intricately linked to the N and chlorophyll contents in leaves (Rogers et al., 2020). Elevated levels of N_area and Chl_area have been associated with enhanced light capture (Hikosaka, 2004). Our findings align with previous research (Liang et al., 2020), demonstrating that under drought conditions, N0 and NH₄ treatments led to reduced LMA, chlorophyll, and N_area, resulting in diminished light capture and suppressed photosynthesis ( Figure 1 ). Consequently, maintaining higher chlorophyll content through the application of NO₃ and NH₄NO₃ treatments proved effective in sustaining the photosynthetic performance of leaves under drought stress.

Plants can also utilize the NPQ mechanism to dissipate an excess of absorbed light energy under sufficient sunlight, thereby avoiding the formation of ROS that can exert detrimental effects on the photosynthetic apparatus (Souza et al., 2022). When the light energy absorbed by the antenna pigments exceeds the capacity of the photosynthetic apparatus, then photochemical inhibition occurs in both PSI and PSII (Sonoike, 2011). Furthermore, it has been reported that E. adenophorum (Spreng.) and P. notoginseng (Burkill.) enhance electron transport to adapt to fluctuations in light and improve the efficiency of light energy utilization (Feng et al., 2009; Cun et al., 2022). In the present study, we also analyzed the impact of drought stress on the PSII reaction centers in L. chinensis leaves and found that in plants treated with NO₃ and NH₄NO₃, Fv/Fm decreased to a lesser extent under drought stress. The reduction in leaf pigment content during drought stress did not significantly reduce Fv/Fo and Fv/Fm ( Figure 3 ). These results suggest that the changes in leaf pigment content had a minimal impact on Fv/Fm, thus indicating that the reduction in pigment synthesis may be a pathway by which plants reduce light absorption to establish photoprotection mechanisms (Galmés et al., 2007). In this study, we observed significant differences in Fv/Fm in the presence of different N forms and under drought stress conditions, thus indicating that Fv/Fm is more sensitive to drought stress when subjected to different N treatments. φPSII, ETR and NPQ data all showed that plants treated with NO₃ and NH₄NO₃ exhibited a smaller reduction under drought stress ( Figure 4 ). We speculate that the reduction in PSII activity during PSII photoinhibition leads to a reduction in ETR. Similar findings have also been reported in L. chinensis plants grown under N0 and NH₄ conditions (Zhong et al., 2019; Wei et al., 2022).

Under drought stress, the PSII reaction centers can undergo reversible inactivation, where they can absorb light energy but do not transfer the absorbed energy to the electron transport chain (Banks, 2018). When the donor side of PSII is damaged, there are changes in chlorophyll a fluorescence in the O-K interval (within a very short time in the L band), thus reflecting the grouping level of PSII within thylakoid membranes and the probability of redistribution of excitation energy between them ( Figure 5 ). The movement of PSII antenna complexes can occur, and any changes in these structures during the growth and development of plants can result in changes in the efficiency of interaction between antenna complexes, thus influencing energy transfer within the process of vegetation growth. Changes in the L band (50-300 μs) can reveal energy transfer between PSII antenna complexes (Strasser et al., 2000; Oukarroum and Strasser, 2004; Strauss et al., 2006). Under drought stress and various N treatments, the difference in the L band was positive, with smaller variations for NO₃ and NH₄NO₃ treatments, thus indicating only mild damage to the leaves of L. chinensis plants. The K band (5-20 ms) reflects the photochemical reduction of Q_A ^- and the partial re-oxidation of Q_A ^- by PQ via Q_B ^- (Tsimilli-Michael and Strasser, 2008). Changes in the K band at approximately 300 μs directly reflect changes in the oxygen-evolving complex (OEC), especially the Mn-complex on the donor side of PSII (Strasser, 1997; Yusuf et al., 2010), thus indicating the functionality of electron transfer on the donor and acceptor sides of PSII. Under HD and MD drought stress treatments, the difference in the K band was positive for all N treatments, although the trends of difference for plants treated with NO₃ and NH₄NO₃ were lower. This is because electron transfer from the donor side was relatively fast, the OEC was not completely inactivated, and electron withdrawal from the acceptor side was accelerated, thereby alleviating damage to the donor side under drought stress (Guha et al., 2013). The H band (2-30 ms) reflects the dynamic process of a reduction in the PQ pool between two chlorophyll fluorescence sites (Harley et al., 1992; Strasser et al., 2010). Under drought stress, the H band was negative for the NO₃ treatment; these finding suggest that under drought stress, the NO₃ treated leaves were beneficial for increasing the PQ pool, promoting proton or electron transfer, and further facilitating the transfer of energy and substances. Changes in the G band (30-300 ms) reflect a reduction in the electron acceptor pool at the PSI terminal, and the rate of decrease determines the variation in the peak of the G band. As the electron acceptor pool increases, the rate of reaching the maximum value slows down, thus resulting in a transient negative value in the G band. When the electron acceptor pool at the PSI terminal decreases, the transient rate accelerates, thus resulting in a positive peak in the differential fluorescence curve (Küpper et al., 2019). Under drought stress, leaves reduce the relative proportion of the more accessible acceptor, and this reduction accelerates, thus resulting in a positive peak in the G band. Conversely, an increase in the more accessible acceptor leads to a negative value. We found that the application of NO₃ ^--N and NO₃ ^--N: NH₄ ⁺-N slowed down the reduction in the more accessible acceptors.

In the N0 and NH₄ treatments, drought stress increased the flow of electrons from the leaf area to the PSI end (REo/CSo); however, due to the restriction of photosynthetic electron transfer, the amount of light energy transferred from the reaction centers to PSI (REo/RC) also increases, thus resulting in a reduced capacity of the leaf to consume excess light energy. However, under the NO₃ and NH₄NO₃ treatments, the opposite effect was observed. Under the NO₃ treatment, the electron acceptor on the acceptor side of PSII (S_m) increased, thus indicating the enhanced probability of electron transfer from captured light energy to Q_A ^- (Strasser et al., 2010). When analyzing PI _abs, we observed that the application of different N forms modified the response of L. chinensis to drought stress ( Figure 6 , Supplementary Figure 1 ). Moreover, PI _abs, identified as the most sensitive parameter in the O-J-I-P kinetic curve, serves as an efficient approach for assessing and determining the resilience of plants to stress (Cun et al., 2022). The absolute value of the slope (K) is 9.37 (N0), 10.49 (NH₄), 11.63 (NO₃), and 11.85 (NH₄NO₃), respectively ( Figure 6E ). A significant positive linear correlation was observed between PI _abs and An ( Figure 6E ). It is evident that L. chinensis tends to utilize NO₃ ^- to enhance photosynthetic capacity to tolerate drought stress. This aligns with our previous research results (Wei et al., 2022). Additionally, a significant positive linear correlation was noted between PI _abs and REoCSo ( Figure 6F ), further confirming that the most crucial determinant of PSII loss of function is the damage to OEC centers (Chen et al., 2016).

Compared to NO₃ ^--N and the mixture of NH₄ ⁺-N: NO₃ ^--N, the application of NH₄ ⁺-N significantly inhibited the growth of L. chinensis under drought stress. The leaf electron transport pathway model (also referred to as the phenomenological energy flux model) and the thylakoid membrane model are widely used to analyze the effects of biotic or abiotic stresses on plants (Pan et al., 2018; Faseela et al., 2019). In response to drought stress, plant leaves exhibit a reduction in the absorbed light energy per unit area (ABS/CSm) (Faseela et al., 2019), thus indicating that drought stress may lead to the degradation or inactivation of reaction centers, or changes in the structure or degradation of antenna pigments, thus resulting in a reduction in the captured light energy. In the present study, we found that a reduction in ABS/CSm inhibits the excitation energy used to reduce Q_A ^- (TRo/CSm) per unit area, as well as the reduction energy (ETo/CSm) entering the additional electron transfer chain (Wang F. et al., 2019). Furthermore, drought stress led to a reduction in thermal dissipation per unit leaf area (DIo/CSm) and a reduction in the activity of the reaction centers per leaf cross-section (RC/CSm) (Hu et al., 2023). These results indicated that L. chinensis leaves activate defense mechanisms when subjected to drought stress, including a reduced leaf area, weakened transpiration, and impeded dissipation of excess excitation energy within the leaf (Guo et al., 2020). We found that the plants treated with NO₃ and NH₄NO₃ were able to alleviate the reduction in ABS/CSm, TRo/CSm, and ETo/CSm ( Figure 7 , Supplementary Figure 2 ). Furthermore, drought stress reduced the number of active reaction centers (RC) per unit leaf area, thereby promoting the efficiency of additional active reaction centers within the leaf, reducing leaf area, and improving the efficiency of excess excitation energy dissipation (Wang B. et al., 2019; Wang F. et al., 2019). In L. chinensis leaves treated with NO₃ and NH₄NO₃, the absorbed light energy per active reaction center (ABS/RC) and the excitation energy used to reduce Q_A ^- (TRo/RC) increased, thereby increasing thermal dissipation (DIo/RC) (Pandey et al., 2023).

During drought stress, the oxidation of Rubisco by RuBP in the photosynthetic carbon oxidation cycle of C₃ plants forms a major alternative sink of electrons, thus maintaining partial oxidation of the PSII acceptor and preventing PSII photoinactivation when CO₂ concentrations decrease (Faizan et al., 2023). The activity of the AOX enzyme is known to be closely related to the accumulation of ROS under drought stress, which can be detrimental to plants (Bartoli et al., 2004). Plants are known to induce the generation of ROS when subjected to environmental stress. However, during the process of evolution, plants have developed rapid and appropriate responses to prevent the formation of ROS induced by environmental stress during growth, development, and defense processes (Huang et al., 2021; Mittler et al., 2022). In most C₃ plants, ROS and H₂O₂ are generated by the oxidation of malate in the peroxisome via the photosynthetic carbon oxidation cycle pathway. Plants usually activate defense mechanisms, including antioxidant enzymes, photoprotection, and the respiratory electron transport chain, to maintain osmotic pressure and balance energy transfer (Ahanger et al., 2017; Huang et al., 2021). In the present study, the levels of ROS in plants treated with NO₃ and NH₄NO₃ under drought stress were significantly lower than those in L. chinensis plants treated with N0 and NH₄, thus indicating that L. chinensis plants treated with N0 and NH4 were more sensitive to drought stress. The application of NO₃ ^--N and NH₄NO₃ activated defense mechanisms in plants and alleviated the cell damage caused by drought stress ( Figure 2 ). The AOX pathway can also serve as an antioxidant mechanism for plant tolerance to environmental stress (Li et al., 2021). In the ascorbate-glutathione cycle pathway, APX catalyzes the reaction between ascorbic acid and H₂O₂, and NADPH uses electron transfer with glutathione as an intermediary to reduce H₂O₂ to H₂O, thereby eliminating the toxicity caused by excessive levels of H₂O₂ (Gupta et al., 2018).

In this study, we analyzed the activity of AOX in plants and found that the AOX activity under NO₃ and NH₄NO₃ treatments was significantly higher than that under N0 and NH₄ treatments and drought stress ( Figure 8 ). Thus, the addition of NO₃ ^--N and a mixture of NO₃ ^--N and NH₄ ⁺-N can activate AOX, consume the excessive reducing power caused by drought, maintain the electron transport capacity of L. chinensis leaves, and reduce oxidative damage. When subjected to drought stress, the application of NO₃ ^--N and a mixture of NH₄ ⁺-N and NO₃ ^--N enhanced the energy consumption efficiency of the remaining active reaction centers to dissipate energy within the electron transfer chain, thereby mitigating damage to the leaves caused by drought stress. These findings demonstrated that the addition of NO₃ ^--N optimized the stability of the light-harvesting and electron transfer systems in L. chinensis leaves under drought stress, thus maximizing the PSII quantum yield.