agronomy The Positive Effects of Increased Light Intensity on Growth and Photosynthetic Performance of Tomato Seedlings in Relation to Night Temperature Level

: Light and temperature are related to the growth and development of plants as well as their energy consumption in plant factories. However, most of the studies to date have focused on light and temperature extremes, while the adaptive responses and underlying mechanisms of plants to non-stress light intensity (LI) and night temperature (NT) largely remain elusive. Here, we investigated the growth and physiological responses of tomatoes grown under three LI regimes of 250 (L L ), 300 (L M ), and 350 (L H ) µ mol m − 2 s − 1 , respectively, combined with two NT conditions. The results revealed that increased LI comprehensively improved tomato growth and physiological status at lower NT levels, but the growth stimulations induced by increasing LI were limited by higher NT. In addition, the lower NT at L M and L H conferred a relatively better endogenous physiological condition and signiﬁcantly promoted tomato growth, but the higher NT signiﬁcantly accelerated shoot growth at L L , indicating a compensation of higher NT for low light induced growth restriction. Taken together, the current study suggests that the adaptation mechanism of tomato plants to higher NT varied with LI levels, and higher LI plus lower NT would be an effective strategy to improve tomato growth.


Introduction
Plant factories can avoid unfavorable external environmental factors during crop production, so growing and nourishing vegetables is very stable throughout the year, letting an uninterrupted supply of vegetables at a constant price, regardless of weather conditions. However, in an artificially controlled plant factory, one of the biggest challenges is the massive energy consumption due to the maintenance and regulation of light and temperature environments [1]. Light and temperature are the two most crucial environmental stimuli that regulated plant growth and development [2], and not only serve as cues to initiate multiple developmental processes, but are also required for optimal performances of biochemical and physiological cascades in the normal plant life cycle [3]. Thus, understanding the responses in the growth and development of the plants to different light and temperature levels is an essential prerequisite for deciphering appropriate strategies to regulate environmental parameters in plant factories.
The effects of sole light or temperature on plant growth and development, including light intensity (LI) and night temperature (NT) have been studied extensively. Plants have and energy utilization of cherry tomato seedlings, accompanied with an understanding the interactive regulatory mechanism between LI and NT for tomato seedling growth.

Plant Materials and Growth Conditions
Cherry tomato (Lycopersicon esculentum Mill. 'Qianxi') seeds were sterilized and incubated at 28 • C and then sown and grown in pots (8-cm diameter) containing a mixture of peat and vermiculite (3:1, v/v). The seedlings have grown to two true-leaf stages in an open ambient climate under natural light condition before being subjected to six combinations of three LI levels (250, 300, and 350 µmol m −2 s −1 , recorded as L L , L M , and L H , respectively) and two NT regimes [(15 ± 1) • C and (18 ± 1) • C, recorded as NT L and NT H , respectively] for 30 days. Herein, a combination of red (peak at 660 nm), blue (peak at 445 nm), and green (peak at 550 nm) LEDs light (Opt-run Biotechnology Co., Nanjing, China) was used as a light source, and the full width at half maximum of all LEDs was 30 nm. The photosynthetic photon flux density (PPFD) ratio of red to blue LEDs was 1:1, and 30 µmol m −2 s −1 of green LEDs was precisely added. Each treatment contained 40 uniform seedlings, and 6 groups were exposed at a day temperature of (26 ± 1) • C with 12 h photoperiod (lighting from 08:00 to 20:00). In addition, the CO 2 concentration and relative humidity were the same as outdoor.

Growth Parameters and Photosynthetic Pigment Measurements
Tomato seedlings were collected for the measurement of growth indices along with subsequent chemical investigations. At least 3 young plants were randomly selected for each treatment. Plant height, stem diameter, and dry weight were measured in accordance with the standard procedure. Leaf area was measured by the disc method. In brief, 1-cm 2 partial leaves were cut from the intact leaves with a puncher to measure the dry weight, and the number of 1-cm 2 partial leaves was recorded as N. Leaf area was determined according to the formula: leaf area = (Dry weight of all leaves × N)/Dry weight of N pieces 1-cm 2 partial leaves. The health index was determined using the following formula: (Stem diameter/Plant height + Root dry weight/Shoot dry weight) × Plant dry weight.
The chlorophyll (Chl) and carotenoid (Car) contents were extracted from 0.1 g of fresh leaves and immersed in 10 mL of 80% acetone (v/v) until the leaves turned white, and then the optical density was measured using a UV-1200 spectrophotometer (MAPUDA instrument, Shanghai, China) and the photosynthetic pigment contents were calculated as described by Fan et al. [20].

Determination of Carbon and Nitrogen Nutritional Status
The carbon and nitrogen nutritional status of plants were estimated by non-structural carbohydrates (NSC), defined as the sum of soluble sugar and starch, and by main nitrogenous compounds including soluble protein and free amino acid, respectively. Briefly, the starch and soluble sugar contents were measured using the anthrone method, and the sucrose concentration was calculated by the resorcinol method in accordance with the method of Li et al. [21]. Soluble protein was extracted from fresh samples and determined using Coomassie Brilliant Blue G-250 [21]. Free amino acid was determined according to the published method of Li et al. [21].

The Determinations of Malondialdehyde (MDA) and Ascorbic Acid (AsA) Contents, and Antioxidant Enzyme Activities
The MDA content was determined using a thiobarbituric acid solution according to Li et al. [21]. For the measurements of the antioxidants, the radical scavenger AsA was extracted by 4.5% aqueous phosphoric acid and quantified as described [22]. Antioxidant enzymes including superoxide dismutase (SOD), catalase, and peroxidase were quantified by previously described procedures [21].

Photosynthesis and Chl Fluorescence Traits Measurements
The net photosynthetic rate (P N ), stomatal conductance (G S ), and intercellular CO 2 concentration (C i ) of the fully expanded third leaves were measured during the daytime between 9:30 and 11:30 using a photosynthesis measurement system (LI-6400XT; LI-COR, Lincoln, NE, USA). Measurements were performed under a PPFD of 250, 300, and 350 µmol m −2 s −1 provided by a mixture of red and blue LEDs for data measurements of L L , L M, and L H treatments, respectively. The leaf temperature, ambient CO 2 concentration (C a ), and relative humidity were 26 • C, (390 ± 10) µmol mol −1 , and 60-70%, respectively. Stomatal limitation (L S ) was estimated by the following formula: The Chl fluorescence parameters were also measured using LI-6400XT. The initial Chl fluorescence yield (Fo) was determined in standard modulated excitation intensity, and a 0.8 s saturation pulse was applied to measure the maximum fluorescence yield (Fm) after the plants were kept in darkness for 30 min. The maximal photochemistry efficiency of PSII (Fv/Fm), where Fv = Fm − Fo, the minimal (Fo') and maximum (Fm') fluorescence level during illumination were also estimated. The fluorescence traits of light-adapted leaves were evaluated when the steady state fluorescence (Fs) level was reached. The actual photochemical efficiency of PSII (PhiPS2) was determined as (Fm' − Fs)/Fm', photochemical fluorescence quenching (qP) was calculated as (Fm' − Fs)/(Fm' − Fo'), non-photochemical fluorescence quenching (NPQ) was recorded as (Fm − Fm')/Fm', and efficiency of open PSII center (Fv'/Fm') was determined as (Fm' − Fo')/Fm'. The electron flow through PSII (ETR) was calculated as follows: ETR = 0.5 × PhiPS2 × PPFD × 0.84. The relative restriction in photosynthetic performance (L PFD ) of seedlings was calculated as described by Schreiber et al. [23] as follows: L PFD = 1 − qP × Fv'/Fm'/0.83. The instantaneous fluorescence decline ratio in light (R FD ) was calculated as follows: R FD = (Fm − Fs)/Fs. The excitation pressure (E P ) of PSII, measured as the relative redox state of QA, was calculated from: E P = 1 − qP.
The distribution coefficient of excitation energy of PSI (α) was calculated as qP/(1 + qP), and the distribution coefficient of excitation energy of PSII (β) was calculated as 1/(1 + qP). The imbalance of excitation energy distribution between two photosystems is expressed by β/α − 1. According to the method of Demmig-Adams et al. [24]. The fraction of absorbed light energy used for photochemical reaction (P) was determined as (Fv'/Fm') × qP; the fraction of non-photochemical dissipation (E X ) was determined as E P × (Fv'/Fm'); the fraction of antenna thermal dissipation (D) was estimated from D = 1 − Fv'/Fm'. Three plants were randomly measured in each treatment, and each plant was repeated twice to take the average value.

Measurement of Dark Respiration Rate and Photorespiration Rate
The fully expanded third leaves were used to measure the light response curve and CO 2 response curve using LI-6400XT. The measurement procedures and fitting models were referred to in a previous study [25]. During the measurements of the light response curve, leaf temperature was 25 • C, CO 2 concentration was 390 µmol mol −1 , and the PPFD gradients were set as 15 levels (1600, 1400, 1200, 1000, 800, 600, 400, 200, 150, 125, 100, 75, 50, 25, 0 µmol m −2 s −1 ). The P N was measured after 2 min adaptation to each PPFD level. During the measurements of the CO 2 response curve, the leaf temperature was 25 • C, and the CO 2 concentration gradients were set as 9 levels (1200, 1000, 800, 600, 400, 200, 150, 100, 50 µmol mol −1 ). The P N was measured after 2 min adaptation to each CO 2 concentration. Using a mechanistic model to fit the two response curves and finally calculate dark respiration rate and photorespiration rate (http://photosynthetic.sinaapp. com/calc.html, accessed on 6 December 2021) [25].

Measurement of Leaf Anatomy and Stomatal Density
The fully expanded leaf of each treatment with the same leaf position was cut vertically with two razor blades pressed together into thin slices, and the slices were then gently put on a glass slide, added a drop of distilled water, and covered with another glass slide. The temporary slices were used to observe the anatomical structure by an optical microscope (DP71; Olympus Inc., Tokyo, Japan), and cross-sections of leaves were measured for leaf thickness, upper epidermis, palisade tissue, spongy tissue, and lower epidermis thickness. Twelve images from three plants per treatment were analyzed.
For the observation of stomata, the imprinting method was adopted according to Yao et al. [26]. A colorless and transparent nail polish was applied evenly to the lower epidermis of the leaf samples, allowed it to solidify for 8 min at room temperature, and subsequently, dried coating layers were torn with tweezers. After the coating layers were unfolded in the glycerol layer on glass slides and squeezed to remove air bubbles, filter paper was used to drain the surrounding glycerin. Temporary slides of epidermal fingerprints were processed using an optical microscope (DP71; Olympus Inc., Tokyo, Japan) connected to an imaging analysis system. The number of stomata per field of view in the leaf epidermis was used to calculate the stomatal density, and stomatal length, width, and area were also measured. Eight images per leaf, one leaf per plant, and three plants per treatment were analyzed.

Observation and Measurement of Chloroplast Ultrastructure
Leaf samples were collected after illumination for approximately 3 h, and the ultrastructure of chloroplast was examined using the methods of Yao et al. [26]. Briefly, the leaves were immerged in 0.2 mol L −1 of phosphate buffer (pH 7.2) containing 2.5% glutaraldehyde, followed by vacuum-pumping. The pre-treated samples were fixed for 2 h in 1% osmium tetroxide, dehydrated in a series of graded ethanol, immersed in acetone, and embedded in Epon-812 epoxide resin. Ultrathin sections (100 nm thickness), which were sectioned with an ultra-microtome (Powertome XL, RMC Products; Boeckeler Instruments Inc., Tucson, AZ, USA), were put on copper grids (one-slot 1 mm) and stained with 2% uranyl acetate and lead citrate. The sections were observed and photographed with an electron microscope (H-7650; Hitachi Ltd., Tokyo, Japan). Eight images per leaf, one leaf per plant, and three plants per treatment were analyzed.

Statistical Analysis
The experiment was repeated three times. Statistical analyses were carried out using Statistical Product and Service Solutions for Windows, ver. 19.0 (SPSS Inc., Chicago, IL, USA). Two-way ANOVA was performed using Fisher's LSD test to assess the interactive effects of LI and NT and differences between means were tested using Duncan's multiple range test (p < 0.05). The correlations among the parameters used in the study were determined by Pearson's correlation coefficient. Differences were considered to be statistically significant for p-values below 0.05.

Growth Traits
F-test indicated that LI and NT as well as their interaction significantly affected tomato seedling growth (Figures 1 and 2A-E). At the NT L level, an increase in LI significantly promoted plant growth and increased plant height, stem diameter, leaf area, health index, and dry weight of tomato seedlings. In contrast, at the NT H level, the increased LI from L L to L M did not promote growth, but the significant promotion still remained in biomass accumulation and health index when increasing LI from L L to L H or from L M to L H . Under L L conditions, higher NT greatly increased plant height, stem diameter, and shoot and plant dry weights. Conversely, a lower NT significantly propelled tomato growth under L M and L H conditions. The healthiest tomato seedlings were observed under L H NT L conditions. These results indicated that the growth promotion by increasing LI is limited by higher NT and the influence of NT on tomato growth is significantly affected by the LI levels. LHNTL conditions. These results indicated that the growth promotion by increasing LI is limited by higher NT and the influence of NT on tomato growth is significantly affected by the LI levels.   LHNTL conditions. These results indicated that the growth promotion by increasing LI is limited by higher NT and the influence of NT on tomato growth is significantly affected by the LI levels.   Abbreviations of different treatments were as described in Figure 1. The symbol "**" and "***" indicate no significance at p < 0.05, significant level at p < 0.01 and p < 0.001 by Duncan's test, respectively. Different letters on bars for the same parameter indicate statistically significant differences (p < 0.05).

Photosynthetic Traits, Dark Respiration Rate (R d ) and Photorespiration Rate (R p )
F-test showed that LI had a significant effect on P N , C i , and L S , while the LI and NT, as well as their interaction, did not significantly affect the G S ( Figure 3A-D). Under the same NT condition, the P N and L S at L M and L H levels were greater than those at L L levels, while C i had an opposite result. The P N , L S , and C i at L M and L H levels had no significant difference at NT L , while, at NT H , the C i decreased but the L S increased significantly at L M level compared with L H level. At the same LI level, NT had no significant influence on photosynthetic traits. The correlation analysis showed that the P N had no significant positive linear correlation with tomato growth parameters (Table S1).

Photosynthetic Traits, Dark Respiration Rate (Rd) and Photorespiration Rate (Rp)
F-test showed that LI had a significant effect on PN, Ci, and LS, while the LI and NT, as well as their interaction, did not significantly affect the GS ( Figure 3A-D). Under the same NT condition, the PN and LS at LM and LH levels were greater than those at LL levels, while Ci had an opposite result. The PN, LS, and Ci at LM and LH levels had no significant difference at NTL, while, at NTH, the Ci decreased but the LS increased significantly at LM level compared with LH level. At the same LI level, NT had no significant influence on photosynthetic traits. The correlation analysis showed that the PN had no significant positive linear correlation with tomato growth parameters (Table S1).
We also measured the Rd and Rp using the method of Ye et al. [26] to determine rates of respiration in the dark and in the light (Table S2). The data showed that the values of Rp were much larger than Rd, and Rd at NTL was higher than that at NTH, and conversely, the Rp at NTL was lower than that at NTH. The differences of Rd at LL and LH were more obvious than that at LM. At NTL level, the Rd was the greatest under LL conditions, while the Rd was the greatest under LM conditions at NTH level. The highest Rp was found under two LM treatments.  and L s , stomatal limitation (D) of cherry tomato seedlings. Abbreviations of different treatments were as described in Figure 1. The symbol "NS" and "**" indicate no significance at p < 0.05, and significant level at p < 0.01 by Duncan's test, respectively. Different letters on bars for the same parameter indicate statistically significant differences (p < 0.05).
We also measured the R d and R p using the method of Ye et al. [26] to determine rates of respiration in the dark and in the light (Table S2). The data showed that the values of R p were much larger than R d , and R d at NT L was higher than that at NT H , and conversely, the R p at NT L was lower than that at NT H . The differences of R d at L L and L H were more obvious than that at L M . At NT L level, the R d was the greatest under L L conditions, while the R d was the greatest under L M conditions at NT H level. The highest R p was found under two L M treatments.

Photosynthetic Pigments
To better define how LI and NT jointly affect photosynthesis, we measured photosynthetic pigments. F-test showed that LI and its interaction with NT had no significant effect on Chl and Car accumulations ( Figure 4). Under NT L conditions, the Chl a and Chl (a + b) contents in L L and L M conditions were significantly lower than those in L H condition; the Chl a/b increased with an increase of LI ( Figure 4). Under NT H conditions, LI alterations had no significant influence on photosynthetic pigment contents and Chl a/b. NT plays a great role in Chl and Car accumulations, and elevated NT significantly increased the contents of Chl and Car under L L conditions; under L M condition, elevated NT also significantly increased the contents of Chl a and Chl (a + b); while under L H condition, elevated NT had no significant effect on Chl and Car contents, but significantly decreased the Chl a/b. The results show that enhancing pigments contents induced by improved LI is closely related to lower NT, while the promotion of high NT on photosynthetic pigment are limited by high LI. The correlation analysis showed that the photosynthetic pigments had no significant linear correlation with the P N , but Chl a/b was highly positive linear correlated with tomato growth parameters shown in Table S1.

Photosynthetic Pigments
To better define how LI and NT jointly affect photosynthesis, we measured photosynthetic pigments. F-test showed that LI and its interaction with NT had no significant effect on Chl and Car accumulations ( Figure 4). Under NTL conditions, the Chl a and Chl (a + b) contents in LL and LM conditions were significantly lower than those in LH condition; the Chl a/b increased with an increase of LI ( Figure 4). Under NTH conditions, LI alterations had no significant influence on photosynthetic pigment contents and Chl a/b. NT plays a great role in Chl and Car accumulations, and elevated NT significantly increased the contents of Chl and Car under LL conditions; under LM condition, elevated NT also significantly increased the contents of Chl a and Chl (a + b); while under LH condition, elevated NT had no significant effect on Chl and Car contents, but significantly decreased the Chl a/b. The results show that enhancing pigments contents induced by improved LI is closely related to lower NT, while the promotion of high NT on photosynthetic pigment are limited by high LI. The correlation analysis showed that the photosynthetic pigments had no significant linear correlation with the PN, but Chl a/b was highly positive linear correlated with tomato growth parameters shown in Table S1. Effect of different light intensity (LI) and night temperature (NT) on photosynthetic pigment content of cherry tomato seedlings. Car, carotenoid; Chl, chlorophyll. The unit of photosynthetic pigments is (mg/g). Abbreviations of different treatments were as described in Figure 1. The symbol "NS", "**" and "***" indicate no significance at p < 0.05, significant level at p < 0.01 and p < 0.001 by Duncan's test, respectively. Different letters on bars for the same parameter indicate statistically significant differences (p < 0.05).

Chl Fluorescence Parameters
We measured Chl fluorescence parameters to better understand the light energy utilization and distribution of tomato seedlings grown under different LI and NT conditions. F-test showed that LI, as well as LI and NT interactions, significantly affected most of the Chl fluorescence parameters (Table 1). Under NTL conditions, PhiPS2 and ETR increased while the LPFD decreased with an increase of LI, and the highest NPQ and Fv'/Fm' occurred in LM conditions; the D decreased significantly in LM and LH conditions compared to that in LL condition. Under NTH conditions, PhiPS2 and qP tended to decrease while LPFD Figure 4. Effect of different light intensity (LI) and night temperature (NT) on photosynthetic pigment content of cherry tomato seedlings. Car, carotenoid; Chl, chlorophyll. The unit of photosynthetic pigments is (mg/g). Abbreviations of different treatments were as described in Figure 1. The symbol "NS", "**" and "***" indicate no significance at p < 0.05, significant level at p < 0.01 and p < 0.001 by Duncan's test, respectively. Different letters on bars for the same parameter indicate statistically significant differences (p < 0.05).

Chl Fluorescence Parameters
We measured Chl fluorescence parameters to better understand the light energy utilization and distribution of tomato seedlings grown under different LI and NT conditions. F-test showed that LI, as well as LI and NT interactions, significantly affected most of the Chl fluorescence parameters (Table 1). Under NT L conditions, PhiPS2 and ETR increased while the L PFD decreased with an increase of LI, and the highest NPQ and Fv'/Fm' occurred in L M conditions; the D decreased significantly in L M and L H conditions compared to that in L L condition. Under NT H conditions, PhiPS2 and qP tended to decrease while L PFD increased with an increase of LI, while NPQ significantly increased; the ETR first increased and then decreased, the ETR at L M level was significantly higher than that at L H level; the Agronomy 2022, 12, 343 9 of 17 E P , β, β/α − 1 and E X at L H level were significantly higher than those at other two light levels. Compared with NT L , NT H significantly promoted PhiPS2 but decreased L PFD at L L level, while NT H significantly decreased PhiPS2 and Fv'/Fm' at L M and L H levels at L H level; NT H also significantly decreased the qP and ETR under L H condition. Besides, under L H condition, NT H resulted in greater E P , β and β/α − 1, L PFD as well as higher E X and D, but caused a lower α and p, showing that the excitation pressure and energy distribution of PSII were increased by the combination of higher LI and NT.  The symbol "NS", "*", "**" and "***" indicate no significance at p < 0.05, significant level at p < 0.05, p < 0.01 and p < 0.001 by Duncan's test, respectively. Different letters within the same parameter indicate statistically significant differences (p < 0.05).
The correlation analysis showed that Chl fluorescence parameters were significantly correlated with the P N . Therein, the L PFD , Fv/Fm, NPQ, E P , β, β/α − 1, and E X were strongly positively correlated with the P N , while the P N was highly negatively correlated with the qP and α. In addition, the Fv'/Fm' and ETR were significantly positively correlated with the growth parameters shown in Figure 2, while the D was significantly negatively correlated with growth parameters (Table S1).

Leaf Anatomical Structure and Stomatal Traits
Leaf growth has a feedback effect on photosynthesis. F-test showed that LI and NT as well as their interaction did not significantly affect stomatal parameters and leaf thickness but affected the thicknesses of palisade tissue, spongy tissue, and lower epidermis with varying significant levels ( Figure 5A-C, Figures S1 and S2A-D). The stomatal density reduced at L M level but significantly increased at L H level compared with L L level regardless of NT; while the stomatal length increased with an increasing LI under NT L condition, and the stomatal length of the L M NT L treatment was significantly longer than that of the L L NT L treatment. Under NT L condition, the stomatal width and lower epidermis thickness had a trend to increase with increasing LI, and the palisade tissue and PTT/SPT in the L M Agronomy 2022, 12, 343 10 of 17 condition were significantly greater than that in the other two light levels. Under NT H conditions, the stomatal width had a trend to decrease with increasing LI, and the thickness of leaf and palisade tissue decreased with the increase of LI. The effects of NT on stomatal and leaf development depend on LI. At L L levels, elevated NT significantly increased the stomatal length but slightly reduced the stomatal width. At L M level, the thickness of palisade tissue and PTT/SPT decreased while upper epidermis thickness decreased with an increase of NT; while under L H condition, a much thinner leaf, sponge tissue, and lower epidermis were observed at NT H levels.
Agronomy 2022, 12, x FOR PEER REVIEW 10 of 17 regardless of NT; while the stomatal length increased with an increasing LI under NTL condition, and the stomatal length of the LMNTL treatment was significantly longer than that of the LLNTL treatment. Under NTL condition, the stomatal width and lower epidermis thickness had a trend to increase with increasing LI, and the palisade tissue and PTT/SPT in the LM condition were significantly greater than that in the other two light levels. Under NTH conditions, the stomatal width had a trend to decrease with increasing LI, and the thickness of leaf and palisade tissue decreased with the increase of LI. The effects of NT on stomatal and leaf development depend on LI. At LL levels, elevated NT significantly increased the stomatal length but slightly reduced the stomatal width. At LM level, the thickness of palisade tissue and PTT/SPT decreased while upper epidermis thickness decreased with an increase of NT; while under LH condition, a much thinner leaf, sponge tissue, and lower epidermis were observed at NTH levels. Abbreviations of different treatments were as described in Figure 1. The symbol "NS", "*", "**" and "***" indicate no significance at p < 0.05, significant level at p < 0.05, p < 0.01 and p < 0.001 by Duncan's test, respectively. Different letters on bars for the same parameter indicate statistically significant differences (p < 0.05).
The correlation analysis showed that the PTT/SPT was significantly positively correlated with the PN, while spongy tissue thickness was significantly negatively correlated with the PN; the lower epidermis thickness was highly positively correlated with growth parameters shown in Figure 2. In addition, leaf thickness was significantly correlated with most of the Chl fluorescence parameters (Table S1).  Figure 1. The symbol "NS", "*", "**" and "***" indicate no significance at p < 0.05, significant level at p < 0.05, p < 0.01 and p < 0.001 by Duncan's test, respectively. Different letters on bars for the same parameter indicate statistically significant differences (p < 0.05).
The correlation analysis showed that the PTT/SPT was significantly positively correlated with the P N , while spongy tissue thickness was significantly negatively correlated with the P N ; the lower epidermis thickness was highly positively correlated with growth parameters shown in Figure 2. In addition, leaf thickness was significantly correlated with most of the Chl fluorescence parameters (Table S1).

Carbon and Nitrogen Nutrition Status
The LI and NT as well as their interaction didn't significantly affect soluble protein content but affected the accumulations of carbohydrates and free amino acid ( Figure 6A-C). Under NT L conditions, the content of free amino acid, soluble sugar, and NSC contents were significantly higher at L M and L H levels, when compared to those at L L level, and the contents of sucrose, starch, and NSC also remained at a significantly high level at L H under NT H condition. Under L M conditions, free amino acid content decreased significantly at NT H conditions, and the contents of soluble sugar, starch and NSC also decreased with an increase of NT under L M and L H conditions, which indicated that higher NT hindered the accumulation of carbon and nitrogen in the plant at relatively high LI levels. Compared with other treatments, L H NT L gained higher contents of free amino acid, soluble sugar, starch, and NSC. The results showed that higher LI plus lower NT increased the accumulation of carbohydrates and free amino acids.

Carbon and Nitrogen Nutrition Status
The LI and NT as well as their interaction didn't significantly affect soluble protein content but affected the accumulations of carbohydrates and free amino acid ( Figure 6A-C). Under NTL conditions, the content of free amino acid, soluble sugar, and NSC contents were significantly higher at LM and LH levels, when compared to those at LL level, and the contents of sucrose, starch, and NSC also remained at a significantly high level at LH under NTH condition. Under LM conditions, free amino acid content decreased significantly at NTH conditions, and the contents of soluble sugar, starch and NSC also decreased with an increase of NT under LM and LH conditions, which indicated that higher NT hindered the accumulation of carbon and nitrogen in the plant at relatively high LI levels. Compared with other treatments, LHNTL gained higher contents of free amino acid, soluble sugar, starch, and NSC. The results showed that higher LI plus lower NT increased the accumulation of carbohydrates and free amino acids. Figure 6. Effect of light intensity (LI) and night temperature (NT) on the carbon and nitrogen nutrition status including soluble protein (A), free amino acid (B) and sugar (C) contents of cherry tomato seedlings. NSC, non-structural carbohydrates. Abbreviations of different treatments were as described in Figure 1. The symbol "NS", "*", "**" and "***" indicate no significance at p < 0.05, significant level at p < 0.05, p < 0.01 and p < 0.001 by Duncan's test, respectively. Different letters on bars for the same parameter indicate statistically significant differences (p < 0.05).
The correlation analysis showed that the contents of soluble sugar, starch, NSC, and free amino acid were significantly correlated with growth parameters shown in Table S1, while the sucrose content was significantly positively correlated with the PN. In addition, Figure 6. Effect of light intensity (LI) and night temperature (NT) on the carbon and nitrogen nutrition status including soluble protein (A), free amino acid (B) and sugar (C) contents of cherry tomato seedlings. NSC, non-structural carbohydrates. Abbreviations of different treatments were as described in Figure 1. The symbol "NS", "*", "**" and "***" indicate no significance at p < 0.05, significant level at p < 0.05, p < 0.01 and p < 0.001 by Duncan's test, respectively. Different letters on bars for the same parameter indicate statistically significant differences (p < 0.05).
The correlation analysis showed that the contents of soluble sugar, starch, NSC, and free amino acid were significantly correlated with growth parameters shown in Table S1, while the sucrose content was significantly positively correlated with the P N . In addition, the nitrogen and carbon nutrition status were also significantly correlated with most of the Chl fluorescence parameters (Table S1).

Chloroplast Ultrastructure
Electron microscopy revealed an increase in LI, regardless of NT, significantly increased starch granules accumulation and gradually altered chloroplast shape (Figure 7). The chloroplasts exhibited a long shuttle shape under the two lower LI conditions but some chloroplasts have shown a similar trapezoid shape under L H conditions. Besides, increased LI plus lowered NT level induced larger and more starch granules.
Agronomy 2022, 12, x FOR PEER REVIEW 12 of 17 the nitrogen and carbon nutrition status were also significantly correlated with most of the Chl fluorescence parameters (Table S1).

Chloroplast Ultrastructure
Electron microscopy revealed an increase in LI, regardless of NT, significantly increased starch granules accumulation and gradually altered chloroplast shape (Figure 7). The chloroplasts exhibited a long shuttle shape under the two lower LI conditions but some chloroplasts have shown a similar trapezoid shape under LH conditions. Besides, increased LI plus lowered NT level induced larger and more starch granules.

Antioxidants and MDA Content
We measured MDA and antioxidants enzymes activity to evaluate oxidative damage and antioxidant ability. F-test showed that LI and NT as well as their interactions had no significant effect on CAT activity and MDA content, but affected the activities of SOD, POD, and AsA content ( Table 2). At the same NT level, SOD and POD activities were increased with an increase of LI, and their activities under LHNTL were the highest; the AsA content at LM and LH levels was significantly higher than those of LL level under NTL condition, but that at LM level was significantly lower than at LL and LH levels under NTH condition. At the same LI level, the SOD activity significantly decreased with an increase of NT, while the POD activity also reduced with an increase of NT at LH level. The results showed that increasing NT or decreasing LI would partly reduce the antioxidant capacity, and increasing LI plus lowering NT would significantly improve the antioxidant capacity.   Abbreviations of different treatments were as described in Figure 1. Scale bar is 5 µm.

Antioxidants and MDA Content
We measured MDA and antioxidants enzymes activity to evaluate oxidative damage and antioxidant ability. F-test showed that LI and NT as well as their interactions had no significant effect on CAT activity and MDA content, but affected the activities of SOD, POD, and AsA content ( Table 2). At the same NT level, SOD and POD activities were increased with an increase of LI, and their activities under L H NT L were the highest; the AsA content at L M and L H levels was significantly higher than those of L L level under NT L condition, but that at L M level was significantly lower than at L L and L H levels under NT H condition. At the same LI level, the SOD activity significantly decreased with an increase of NT, while the POD activity also reduced with an increase of NT at L H level. The results showed that increasing NT or decreasing LI would partly reduce the antioxidant capacity, and increasing LI plus lowering NT would significantly improve the antioxidant capacity.
The correlation analysis showed that the activities of SOD and POD and AsA content were strongly positively correlated with growth parameters shown in Table S1, soluble sugar, starch, NSC, free amino acid, and lower epidermis thickness. The SOD activity was also positively correlated with the ETR and NPQ, and the activity of POD was positively correlated with the ETR. In addition, the AsA was highly negatively correlated with the D (Table S1). Abbreviations of different treatments were as described in Table 1. AsA, ascorbic acid; CAT, catalase; FM, fresh mass; MDA, malondialdehyde; POD, peroxidase; SOD, superoxide dismutase. The symbol "NS", "*", "**" and "***" indicate no significance at p < 0.05, significant level at p < 0.05, p < 0.01 and p < 0.001 by Duncan's test, respectively. Different letters within the same columns indicate statistically significant differences (p < 0.05).

The Growth Promotion Induced by Increasing LI Was Obviously Limited by Higher NT
The present study showed that continuously increased LI, especially at lower NT level, comprehensively improved plants growth quality and physiological traits (Figures 1, 2, 5 and 6, Table 1), which is consistent with the previous results that increased LI allows better performances of Chinese red radish and rape seedlings [26,27]. In this study, however, the promotion induced by the increased LI was obviously limited by higher NT. At NTH conditions, an increase in LI only slightly boosted the health index and dry weights of seedlings ( Figure 2).
Under both NT L and NT H conditions, an increase of LI boosted carbohydrate accumulation ( Figure 6C). High levels of NT are considered to be one of the leading environmental factors resulting in lower yield in cotton, which has been attributed to a negative effect of respiration on carbohydrate accumulation [9]. Our result confirmed the point that the tomato growth limitation under higher NT was strongly correlated with lower carbohydrate accumulation and stronger photorespiration but not dark respiration (Tables S1 and S2). Higher NT significantly decreased starch accumulation, increased R p , changed the shape, and reduced quantity of starch granule, leading to an obvious reduction in NSC, especially under L M and L H conditions ( Figures 6C and 7 and Table S2).
Previous results showed that the leaves became thinner in higher temperature environments [28]. It was supported by our findings that when the NT increased, the leaves' thickness gradually became less, which was significantly obvious under high LI.

Reversible Inactivation of PSII Reaction Center and Photorespiration Caused Tomato Growth Limitation Induced by Higher NT at L H Condition
The Chl is the fundamental material of plant photosynthesis, participating in the absorption, transmission, and distribution of light energy [29]. Under natural conditions, the light environments are continuously fluctuating resulting in unstable light interception of plant leaves. One of the physiological mechanisms that allows plants to quickly adapt to the fluctuation of both light intensity and light quality is regulating the Chl a/b ratio in leaves [26,30]. Higher Chl a/b is a typical character of plants grown under high LI conditions [26]. The current study showed similar results that, under NT L conditions, with the increase of LI, the Chl a/b ratio of the treatments gradually increased (Figure 4). The Chl a/b ratio was also affected by light quality (mainly the blue and red light). Exactly, blue light boosted the formation of Chl a, and thus led to a higher Chl a/b ratio, whereas red light induced more Chl b and lower Chl a/b ratio [31]. In this study, the NTs affected the contents of Chl a and Chl b, as well as the Chl a/b ratio. Especially at L H condition, the increasing NT significantly reduced the Chl a/b ratio of tomato leaves (Figure 4). It was deduced that the tomato plants grown under L H NT H might absorb more red light than blue light compared to the plants grown under L H NT L . The plant morphological differences between L H NT H and L H NT L treatments might be related to the light energy absorbed by plants under the two treatments, because the absorption of blue light excites the electrons to a higher energy state than red light [32].
Correlation analysis showed that Chl a/b was not only highly positively correlated with morphology and dry weight of tomato plants, but also correlated with the Fv'/Fm', ETR, and D (Table S1). Within the pigment molecules, the absorbed light energy excites electrons to a higher state, which is then transferred to the proteins located on the thylakoid membrane, therefore entering into the electron transport chain. In this study, elevated NT at L H level caused a significant reduction in PhiPS2, qP, Fv'/Fm', and ETR, but a significant increase in the excitation pressure of PSII (E P ), implying a reduction in the utilization of excitation energy through photosynthetic electron transport chain and the excessive accumulation of light energy in PSII. In addition, L H NT H also led to more distribution of excitation energy at PSII (higher β) and less excitation energy distribution at PSI (less α), resulting in an imbalance of excitation energy distribution between the two photosystems, which was indicated by a larger β/α − 1 (Table 1). High PSII excitation energy pressure (namely high E P and high β) would induce reversible inactivation of the reaction center, and even destroy the structure of PSII and thylakoid membrane, so as to hinder electron transfer and seriously restrict the efficient operation of photosynthetic apparatus. However, the excess excitation energy in the L H NT H treatments did not seem to destroy the photosynthetic apparatus but tended to cause relative reversible inactivation of the PSII reaction center, because of no differences in Fv/Fm, R FD , and MDA content between the L H NT L and L H NT H treatments (Tables 1 and 2). In addition, under NT H conditions, a higher photorespiration rate indicated that photorespiration also consumed excess energy in order to protect the reaction center of PSII from being damaged by strong light. These reasons ultimately led to relative limitation for photosynthetic function (reflected by the highest L PFD ) and lower light energy utilization, and resulted in a growth inhibition of plants in the L H NT H treatment (Table 1) [33,34].

Higher NT at L M Condition Limited Tomato Growth but Exerted No Excess Excitation Pressure on PSII
Unlike L H levels, elevated NT at L M levels did not lead to an imbalance of excitation energy distribution between the two photosystems and excess excitation pressure of PSII, but merely significantly reduced the PhiPS2 and Fv'/Fm', and increased the quotient of absorbed luminous energy used for antenna heat dissipation (D) ( Table 1). This result suggested that the enhancement of PSII antenna heat dissipation, in order to maintain the normal excitation pressure of PSII in plants, was one of the main mechanisms of selfprotection of photosynthetic apparatus during photosynthesis when increased the NT under L M condition.
A previous study indicated that AsA also stimulated electron flow and photochemical quenching [35]. In this study, AsA was positively correlated to Fv'/Fm', and when increased NT under L M , AsA content was also reduced by 75.12%, implying that the restriction in the utilization of excitation energy under L M NT H condition was in relation to the reduction in the AsA content (Tables 2 and S1). In addition, AsA and SOD confer higher free radical scavenging ability in plants [36]. At L M levels, the increased NT significantly reduced AsA content and SOD activity, indicating that increasing NT weakened the antioxidant capacity of plants, thus somewhat limiting the plant growth.
Moreover, significantly less free amino acid was observed under L M NT H than that under L M NT L , indicating that plants could respond to NT alteration via nitrogen metabolism. The limitation of tomato growth at L M NT H is also partly due to some NADPH and ATP consumed in nitrogen reduction [23]. In addition, under L M NT H condition, stronger pho-torespiration increased the inclusion losses in the leaf and slowed down the growth process of plants.

Elevated NT Performed Compensatory Effect on Plant Growth under Insufficient Light Conditions
In contrast to L M and L H levels, elevated NT significantly reduced the relative restriction of photosynthetic performance (L PFD ) and could comprehensively improve tomato shoot growth at L L levels ( Figure 1 and Table 1). The dry weight of shoot and plant grown under higher NT was both significantly increased by about 50% compared to those under lower NT (Figure 1). Fan et al. [4] indicated that red and blue composite light of 300 µmol m −2 s −1 was fit for the culture of young tomato plants, so red, blue, and green composite light of 250 µmol m −2 s −1 in this study was slight insufficient for tomato growth. However, under L L NT H condition, the R d was largely lower than that under L L NT L condition, while the R p was only slightly higher than that under L L NT L condition, suggesting that elevated NT decreased the energy consumption for respiration. The results implied that elevated NT had a compensatory effect on plant growth and the photosynthesis process when LI was not sufficient for tomato growth.
To Cope with the shortage of light, free amino acid, and AsA contents were significantly increased to create a relatively better endogenous environment for energy consumption, electron transfer, and radical scavenging ability, and plants grown under L L NT H improved Chl and Car contents ( Figures 2C, 3 and 6B, Table 2) [23,36]. An increase of Chl content broadened the light absorption bands (mainly at green and the far-red regions), so we deduced that the plants grown under higher NT might absorb more green light than that grown lower NT at L L level [24]. The green light sensory systems adjusted growth and development interacting with red and blue sensors [37], and green light can penetrate further into the leaf than red or blue light to promote leaf photosynthesis more efficiently [38]. Correspondingly, at L L levels, the PhiPS2 of the higher NT treatment was significantly increased (Table 1). Similarly, at L M levels, higher NT may display a similarly positive effect, but it was overwhelmed by other limiting factors.

Conclusions
In summary, different combinations of LI and NT induced differential growth response in tomato seedlings, and the promotion induced by increased LI was limited by higher NTs. The responsive mechanism of tomato plants to higher NT varied with LI levels. At L H level, higher NT inhibited tomato growth via reducing the leaf area, changing the Chl a/b and chloroplast shape, regulating excitation energy utilization and distribution, and carbon metabolism; while, reducing antioxidant capacity mainly responsible for growth constraints caused by higher NT at L M level. Whereas, elevated NT has a growth compensation for tomato seedlings under the lowest LI, and higher NT improved tomato growth mainly by improving light energy utilization in low light and physiological indicators, primarily by increasing stomatal length, pigments, free amino acid, and AsA contents, and decreasing R d . Comprehensive analysis showed that higher LI plus lower NT would be an effective strategy to improve tomato growth, while lower LI plus higher NT might be worth considering as a strategy for tomato production in hot season in view of energy consumption cost in a controlled environment.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/agronomy12020343/s1, Figure S1: Electron microscopy of stomata in lower epidermis of cherry tomato grown under different light intensity (LI) and night temperature (NT), Figure S2: Effect of light intensity (LI) and night temperature (NT) on the stomata characteristics, upper epidermis thickness and PTT/SPT of cherry tomato seedlings, Table S1: Correlation analysis among growth, photosynthetic, leaf development and antioxidants parameters, Table S2: Fitted photosynthetic parameters via light/CO 2 response curves under different light intensity and night temperature conditions. Author Contributions: Conceptualization, X.L.; Data curation, Z.C. and X.L.; Formal analysis, J.S., Z.C., M.W. and X.L.; Funding acquisition, X.L.; Investigation, J.S., A.Z., Y.W. and X.L.; Methodology, X.L.; Supervision, X.L.; Validation, M.S.J. and X.L.; Writing-original draft, Z.C. and X.L. All authors have read and agreed to the published version of the manuscript.