SAPK2 contributes to rice yield by modulating nitrogen metabolic processes under drought stress CURRENT

Background The sucrose non-fermenting 1-related kinases 2 (SnRK2s) play roles in stress responses in A. thaliana and rice ( Oryza sativa L.). Osmotic stress/ABA–activated protein kinase 2 (SAPK2) is a member of SnRK2s subclass II in rice, but its function in rice yield under drought stress is unclear. Results Compared with wild-type (Oryza.Sativa L.spp.japonica, WT) plants, the sapk2 rice mutant lines were shorter and produced fewer grains per panicle, smaller grains and lower grain yield. Subsequent analysis suggested that SAPK2 considerably influences the nitrogen, phosphorus, and potassium contents of rice grains. The examination of rice seedling growth and development under nutrient-deprived conditions (−N, −K, and −P) proved that SAPK2 can significantly affect rice seedling growth and root development in hydroponic cultures lacking N and K. Moreover, the NO 3 − influx rate and nitrate concentration analysis indicated that SAPK2 promotes nitrate uptake and assimilation by regulating nitrate-related transporters. These results suggest that SAPK2 could enhance grain production by regulating nitrogen utilization efficiency under drought stress. Our work provided insights to breeding drought tolerant rice with high nutrient uptake.

4 plant stress responses, and the interaction between ABA and other classical stress-related hormones enables plants to rapidly respond and properly adapt to drought stress (Brodribb and McAdam 2017).
The ABA-dependent signaling pathways are critical for the expression of genes responsive to various stresses, especially osmotic stress. SnRK2 family members are protein kinases which promote ABA responses (Hrabak et al. 2003).
A recent study confirmed that the overexpression of SAPK9 may significantly enhance drought tolerance, while also increasing the grain yield under drought condition (Dey et al. 2016). In previous research, we found that the sapk2 mutants exhibited an ABA-insensitive phenotype during the germination and post-germination stages and sensitive phenotype to drought stress, with lower survival rates than the wide-type plants. The mutants also exhibited greater water loss, lower proline and soluble sugar contents, higher proportions of fully open stomata, higher ROS levels, and lower antioxidant enzyme activities (Lou et al. 2017). These results indicate that SAPK2 may be useful for improving crop yields under drought conditions. However, the effect of SAPK2 on the productivity under drought stress remains unclear.
We found that the sapk2 mutant showed lower grains yield and lower nitrogen, phosphorus, and potassium contents of rice grains than the WT under drought stress. Moreover, the sapk2 mutant 5 exhibited weaker seedling growth and root development in hydroponic cultures lacking N and K. The NO 3 − influx rate and nitrate concentration analysis indicated that SAPK2 promotes nitrate uptake and assimilation by regulating nitrate-related transporters. Our work provided insights to breeding drought tolerant rice with high nutrient uptake.

Generation of transgenic rice lines
We employed knock-out mutant lines (sapk2, sapk2-1 and sapk2-7) which we built previously by the CRISPR/Cas9 system and over-expression lines (OE, OES2-1 and OES2-2) (Lou et al. 2018 Rice (Oryza sativa L. japonica.) was used for transformation. The primers used in this study were listed in the Additional file 2: Table S1.

Plant cultivation and agronomic traits analysis
For basic agronomic traits analysis, rice plants were grown in the paddy field from March to August at the rice experimental station of Xishuangbanna Tropical Botanical Garden. Ten plants at a spacing of 16.5 cm × 26.5 cm were planted in a row and 5 rows of each line were planted. At reproductive stage, 20 plants of each line were randomly chosen to detect agronomic traits. The grain number per panicle was measured as the total number of grains per plant divided by the number of panicles per plant.
The 1000-grain weight was calculated as the weight of the total grains per plant and divided by the grain number, then converted to 1000-grain weight. Grain yield per plant was measured as the weight of total grains per plant.
For reproductive-stage drought stress, transplanted experiments were maintained as described by 6 Sandhu et al. (2016). The drought stress was initiated at 32 days after transplanting. After the inception of the stress, the soil water potential was measured using tensiometers (30 cm depth). The plots in the reproductive stage drought stress treatments were rewatered when the soil water potential dropped to −50 to −70 kPa (tensiometer). The decline in water

Measurement of nitrate, phosphorus, potassium and total nitrogen content and nitrate influx
For free NO 3 analysis, 0.1 g plant materials from different genotypes were homogenized by grinding in cold extraction buffer [50 mM Tris-HCl (pH 7.0), 10 mM imidazole, and 0.5% (w/v) bmercaptoethanol]; the homogenates were then centrifuged at 12, 000 g for 20 min at 4 ℃ and the supernatant was collected. Free NO 3 in the supernatant was determined by using the Griess method (Walther et al. 1999), the absorbance at 540 nm was determined and NO 3 contents were calculated from the standard curve of KNO 3 .
NO 3 − influx was calculated as the difference in NO 3 − content between the plants under control conditions and drought stress conditions in an hour.
For total nitrogen, phosphorus and potassium content analysis, 50 seeds from different genotypes were homogenized by grinding in Liquid nitrogen. Then 0.1 g seed materials were measured using 7 Agilent 7700e ICP -MS.
At least three independent biological experiments were conducted. One representative result was displayed here.

RNA extraction and qRT-PCR analysis
To detect the transcript level of SAPK2 under different nutrient-deprived conditions, 14 DAG seedlings were transferred to different nutrient-deprived solutions for 24 hours. To detect the transcript level of target genes under drought stress, seedlings were transferred into half-strength liquid medium supplemented with 25% PEG6000 (m/v) for 24 hours. All samples were collected at the right time.
For the qRT-PCR analysis, we used the same method as described (Jiang et al. 2016). Total RNA was isolated from whole seedlings using the TriZol reagent (Invitrogen). The cDNAs were obtained by using Superscript II in accordance with manufacturer's instructions (Invitrogen). The qRT-PCR analysis was performed using SYBR Premix Ex Taq kit (Takara).
At least three independent biological experiments were conducted (three independent samples were conducted for each experiment and three technological replications in every independent experiment). One representative result was displayed here. Gene-specific primers used in qRT-PCR analysis were listed in Additional file 1: Table S1.

Statistical analysis
The experiments were arranged in a completely randomized design with at least three replicates for each treatment. Excel 2010 was used for making charts. Two-tailed Student's t tests were performed using the SPSS 10 software (IBM, Inc.). "* and **" indicate significance at P < 0.05 and P < 0.01, respectively. The data represent mean ± standard error (SE) of three independent experiments.

Mutations to SAPK2 decrease plant height and grain yield
Under drought conditions, the two sapk2 lines were shorter than the WT control plants (Fig. 1a-c). To further investigate the SAPK2 roles influencing rice yield, we analyzed OE lines (OES2-1 and OES2-2) and sapk2 knock-out mutant lines (sapk2-1 and sapk2-7). An analysis of plants exposed to reproductive stage drought stress (RS) indicated that the two sapk2 lines had substantially more 8 tillers than the WT plants ( Fig. 1a, d), whereas there were no significant differences in the two OE lines ( Fig. 1b-d). However, although they had more tillers, the sapk2 mutant plants produced fewer grains per plant than WT under drought condition (Fig. 1f). A subsequent investigation of the regulatory effect of SAPK2 on the number of effective tillers revealed that the sapk2 mutant lines had considerably fewer effective tillers than WT, but there were no significant differences in the OE lines ( Fig. 1e). This result further confirmed that the sapk2 mutant produces fewer grains than WT under drought condition.
The number of grains per panicle is one of the three key factors determining rice grain yield (Xing and Zhang 2010). Thus, we investigated the SAPK2 roles related to panicle and grain development by analyzing the number of grains per panicle, the seed setting rate, the grain length and width, the 1,000-grain weight and the grain yield per plant in response to RS. Under RS, the panicles and grains of the sapk2 mutant lines were smaller than those of WT ( Fig. 2a). Additionally, the grain number per panicle of the sapk2 mutant lines was 75% and 60.8% of that of WT and OE lines (Fig. 2b). The setting rate of OE lines did not differ from that of WT, whereas the setting rate of the sapk2 mutant lines decreased significantly (76.9% of that of WT) (Fig. 2c). Compared with WT, the grains of OE lines were significantly longer, whereas there was no significant difference in the grain length of the sapk2 mutant lines (Fig. 2d). In contrast, the grains of the sapk2 mutant lines were significantly thinner than WT, whereas the grain width of OE lines was not significantly different ( Fig. 2e). Moreover, the 1,000grain weight and grain yield per plant of the sapk2 mutant lines were much lower than WT and OE plants (Fig. 2f, g). These results implied that SAPK2 influences panicle and grain sizes in rice.
Overall, our observations indicated that knocking out of SAPK2 significantly decreases rice plant height and grain yield per plant under drought condition. Additionally, overexpressing SAPK2 does not appear to enhance rice plant growth or grain production.

Mutations to SAPK2 decrease nitrate, phosphorus, and potassium contents in rice grains
under reproductive-stage drought Umezawa et al. (2004) reported that the expression of SnRK2.8, which is a homolog of rice SAPK2, is down-regulated by potassium deprivation. This down-regulation is associated with a substantial decrease in the growth of A. thaliana under nutrient-deprived conditions. As mentioned earlier, SAPK2 influences rice panicle and grain sizes. To clarify the mechanisms by which SAPK2 influences panicle and grain sizes in rice, we measured total nitrogen, nitrate, phosphorus, and potassium contents of seeds. Under RS conditions, the seeds of the sapk2 mutant lines had lower total nitrogen, nitrate, phosphorus, and potassium content than WT, with the biggest difference observed for the total nitrogen ( Fig. 3a-c, SFig 1). However, the total nitrogen, nitrate, phosphorus, and potassium content were relatively consistent between the OE and WT plants (Fig. 3a-c). These findings confirmed that in rice, the seed nitrate, phosphorus, and potassium contents are largely affected by SAPK2. Therefore, we hypothesized that SAPK2 influences panicle and grain sizes by modulating metabolic processes of N, P, and K.

SAPK2 affects seedling growth and root development in response to N and K deprivation
To further validate our hypothesis, we investigated the effects of knocking out and overexpressing presented). A previous study proved that the root morphology influences plant interactions with soil nitrates, making it important for N absorption (Hachiya and Sakakibara 2017). Accordingly, we examined the root development of the SAPK2 transgenic lines. In the sapk2 mutant lines, root growth was inhibited, resulting in roots than WT seedlings (Fig. 4b, f). In contrast, the root phenotypes of the OE and WT plants were similar (Fig. 4b, f; In Fig. 4f, OE phenotype is not presented). The root and shoot dry weights of the sapk2 mutant lines were significantly lower than WT, but there were no significant differences in the OE lines (Fig. 4c, d). Similarly, compared with WT, the sapk2 mutant lines had fewer roots, whereas there were no significant differences in the OE lines (Fig. 4e).
The effects of the K-deprived conditions were similar to those of the N-deprived conditions. For example, the sapk2 mutant seedlings produced weaker culms and had lower root and shoot dry weights than WT (SFig. 2a-f). In contrast, the exposure to P-deprived conditions did not result in any significant differences in the seedling growth and root development of the WT, OE and sapk2 seedlings (SFig. 3a-f).
These findings suggested that SAPK2 can significantly influence rice seedling growth and root development in hydroponic cultures under N-and K-deprived conditions.

SAPK2 influences the NO 3 − influx rate and nitrate concentration under drought stress condition
To explore the potential mechanism underlying the effects of SAPK2 on rice seedling growth and root development under N-deprived condition, we investigated the NO 3 − influx rate and nitrate concentration of the WT, OE, and sapk2 plants under control and drought conditions. Under control conditions, there were no significant differences among the WT, OE, and sapk2 plants (Fig. 5a, c).
However, in response to drought stress, the NO 3 − influx rate and nitrate concentration significantly decreased in the WT, OE, and sapk2 plants (Fig. 5a-d). Additionally, the rate of NO 3 − influx into the roots was lower for the sapk2 mutant lines than for WT (Fig. 5b), suggesting that knocking out of SAPK2 weakens the nitrate uptake by the roots. Regarding the sapk2 mutant lines, we also detected a lower rate of NO 3 − influx into the leaf sheath and leaf blade, implying that SAPK2 promotes the translocation of NO 3 − from the roots to the leaf sheath (Fig. 5b). Moreover, the root, leaf sheath, and leaf blade nitrate concentrations were consistent with the NO 3 − influx rates in the different lines (Fig.   5d). These results demonstrated that SAPK2 enhances nitrate influx and increases the nitrate concentration by promoting the translocation of nitrate from the roots to the leaf sheath.
Nitrogen use efficiency is an important trait for the development of sustainable agricultural production . Plants have diverse transporters facilitating N uptake and internal distribution (Rentsch et al. 2007). In higher plants, members of the NPF family (previously called the PTR/NRT1 family) can take up and translocate nitrate or small peptides. Of the rice NPF family members, only a few have been studied. For example, OsNPF7.2, which encodes a positive regulator of nitrate influx and concentration, helps control the allocation of nitrate between the roots and shoots . In rice, the peptide transporter OsNPF7.3 (OsPTR6) mediates the transport of organic N from the leaves to the grains and increases the grain yield (Fang et al., 2017).
Additionally, OsNPF6.5 (OsNRT1.1B) is predominantly expressed in the root hairs, epidermis, and stellar cells adjacent to the xylem in roots. The osnrt1.1b mutant is reportedly defective in both nitrate uptake and root-to-shoot nitrate transport, suggesting that OsNRT1.1B is involved in nitrate uptake and transport . Down-regulating OsNRT2.3a expression impairs the loading of nitrate into the xylem and inhibits plant growth under low-nitrate conditions, implying OsNRT2.3a contributes to the long-distance transport of nitrate from the roots to the shoots (Tang et al., 2012).
The silencing of OsNPF2.4 diminishes the low-affinity nitrate acquisition by roots, disrupts the Kcoupled root-to-shoot nitrate transport, and inhibits the redistribution of nitrate from old leaves to Nstarved roots or young leaves .
To further validate our hypothesis that SAPK2 influences panicle and grain sizes by modulating N metabolic processes, we determined the expression levels of genes crucial for the absorption, transport, and assimilation of nitrate among the WT, OE, and sapk2 plants cultured under control and drought conditions. The OsNPF7. 2,OsNPF7.3,OsNPF5.6,OsNPF2.2,OsNRT2.3a,and OsNPF2.4 expression levels were significantly lower in the sapk2 mutant lines than in WT under drought condition ( Fig. 6a-f). However, the opposite expression patterns were detected for the OE lines ( Fig.   6a-f). These results implied that SAPK2 promotes nitrate uptake and assimilation by regulating nitrate-related transporters.

Mutations to SAPK2 decrease grain yield
Drought is one of the most important environmental stresses affecting the productivity of most field crops (Gray and Brady 2016). Elucidating the complex mechanisms underlying the drought resistance of crops will accelerate the development of new varieties with enhanced drought resistance.
Improving the grain yield is the primary aim for most rice breeders. The rice yield potential is determined by the biomass and harvest index.
To identify additional rice grain yield-related genes affected by drought stress, we functionally characterized SAPK2 by examining sapk2 mutant lines exposed to drought stress. Under RS condition, the sapk2 mutant plants produced fewer effective tillers and grains per plant than the WT and OE plants. Rice grain yield is determined by the following three traits: number of panicles, number of grains per panicle, and grain weight (Xing and Zhang 2010). Consequently, we were interested in whether SAPK2 can be used to improve rice yields under RS conditions. We also investigated the SAPK2 roles associated with panicle and grain development in response to RS conditions. Specifically, the panicle size, number of grains per panicle, grain size, setting rate, 1,000-grain weight and grain yield per plant were significantly lower for the sapk2 mutant lines than WT ( Fig. 2a-g). These results indicated that SAPK2 increases rice yield under RS condition by influencing panicle and grain size.

SAPK2 affects seedling growth and root development in response to N and K deprivation
A previous study revealed that SRK2C/SnRK2.8 may mediate the phosphorylation of enzymes involved in metabolic processes (Umezawa et al., 2004). To investigate whether the decrease in the grain yield of the sapk2 mutant lines under RS conditions is related to nutrient metabolism, we measured the seed total nitrogen, nitrate, phosphorus, and potassium content. The data indicated the sapk2 mutant seeds had lower total nitrogen, nitrate, phosphorus, and potassium content than that in WT and OE lines seeds, especially for total nitrogen content (Fig. 3a-c and SFig. 1).
The roots, which are responsible for the uptake of water and nutrients from the soil, are vital for plant growth, development, and fitness. Additionally, ABA controls several key steps associated with lateral root initiation as well as meristem activation and elongation (De Smet et al. 2003;Ding and De Smet 2013). Root development is directly affected by environmental factors. Furthermore, the plant root system architecture is plastic and dynamic; enabling plants to respond to environmental changes, which then promotes root growth and development to avoid water deficit stress in the early stages of drought stress. Nitrogen availability is a major determinant of plant growth and crop productivity 13 (Hachiya and Sakakibara 2017;Li et al. 2017). Plants use inorganic forms in natural soils, including nitrates, nitrites, and ammonium, and nitrate is the major form of N in most aerated soils. Of these available N sources, researchers have mainly focused on nitrate and ammonium because these are often present in natural and cropland soils at much higher levels than the other N sources (Miller and Cramer, 2004).
To further validate the relationship between SAPK2 and N metabolism, we investigated the effects of knocking out and overexpressing SAPK2 on rice seedling growth and development in hydroponic cultures under N-deprived conditions. The sapk2 mutant seedlings under N-deprived conditions had weaker culms than WT, with inhibited root growth, significantly lower root and shoot dry weights, and fewer roots (Fig. 4a-f). However, we did not find significant difference between WT and OE plants ( Fig. 4a-e).
Earlier investigations confirmed the importance of K for enzyme activities and ionic homeostasis in plants (Shabala and Pottosin 2014;Ahmad and Maathuis 2014). Additionally, we previously determined that SAPK2 regulates the expression of genes related to Na + and K + homeostasis, including OsSOS1, OsNHX1, OsHKT1;1, and OsHKT1;5 (lou et al. 2018). In this study, the sapk2 mutant seedlings deprived of K produced weaker culms and had lower root and shoot dry weights than WT and OE plants (SFig. 2).
Furthermore, we also investigated SAPK2 effects on rice seedling growth and development in hydroponic cultures under P-deprived conditions and found there was no significant difference between sapk2 mutant, WT and OE plants (SFig. 3).
Therefore, above results presented here confirm that SAPK2 influences panicle and grain sizes via its effect on N and P metabolic processes and is positive regulator in response to N and K deprived conditions.
However, the SAPK2 expression was down-regulated in response to N, P, and K deprivation in the roots (Fig. 3d-f). We think the reason for this result is related to how SAPK2 works. SAPK2 is a member of SnRK2 subclass II in rice. In the absence of ABA or osmotic stress, SAPK2is inhibited by ABA receptors OsPYL/RCAR5 and induces complex formation with PP2C30 through dephosphorylating.
14 Under ABA or osmotic stress, ABA accumulates and binds to its receptors OsPYL/RCAR5, subsequently inhibit PP2C30 activity, resulting in the release of SAPK2 from inhibition. Activated SAPK2 phosphorylate downstream effectors to mediate stress responses (Kim et al. 2012). Current models suggest that ABA, which is induced by drought stress, inhibits the activity of OsPP2C49 to release the kinase activity of SAPK2 for further activation of OsbZIP23 through phosphorylation. OsbZIP23 could directly bind to the promoters of OsPP2C49 and positively regulates the expression of OsPP2C49, which in turn negatively regulates the ABA signaling (Zong et al. 2016). This regulatory mechanism of SAPK2 mainly occurs at the protein level, but does not change much at SAPK2 transcription level.
However, there may be a negative regulatory mechanism at SAPK2 transcription level after this response, resulting in a decrease in SAPK2 transcription level.

SAPK2 influences the NO 3 − influx rate and nitrate concentration
Nitrogen uptake, assimilation, and recycling in plant roots reportedly determine plant development and productivity (Yamaya and Kusano, 2014). Plant growth under natural conditions is often limited by N availability. Therefore, plants have developed transport and signaling mechanisms for their specific N sources (Kiba and Krapp, 2016). To further validate the relationship between SAPK2 and nitrogen absorption and utilization by investigating the effects of silencing and overexpressing SAPK2 on rice. Under drought stress conditions, in the sapk2 mutant lines, we detected a lower rate of NO 3 − influx into roots and also into the leaf sheath and leaf blade (Fig. 5b). Besides, the detected nitrate concentration was consistent with the rate of NO 3 − influx in different lines (Fig. 5d). These results demonstrated that SAPK2 enhanced nitrate influx and concentration by promoting nitrate uptake by roots and translocation of nitrate from roots to leaf sheath.
In this study, we analyzed the expression of several nitrate transporter genes (OsNPF7.2,OsNPF7.3,OsNPF5.6,OsNPF2.2,OsNRT2.3a,and OsNPF2.4). Under drought conditions, the expression levels of these six genes were significantly lower in the sapk2 mutant lines than in the WT plants in response to drought stress ( Fig. 6a-f). Furthermore, OsNPF6.5 regulates the number of rice tillers and promotes grain production . The overexpression of OsNPF7.2 and OsNPF7.3 increases the production of tillers, panicles per plant, and filled grains per panicle, while also increasing the grain N content Fang et al., 2017). The osnpf2.2 (OsPTR2) mutants are defective in longdistance nitrate transport, with repressed nitrate unloading from the xylem, resulting in plant growth retardation and abnormal grain filling . Our study revealed that the sapk2 mutant lines produced significantly fewer tillers than the WT plants (Fig. 1d). Moreover, the sapk2 mutant seeds had a lower total nitrogen content than WT seeds under RS condition (Fig. 3a).
On the basis of our results, we conclude that under drought stress, SAPK2 promotes nitrate uptake and assimilation and influences the number of tillers and the number of grains per panicle by regulating nitrate-related transporters.

Conclusions
In this study, we examined knock-out mutant lines and SAPK2-overexpressing lines. We found that the sapk2 mutant showed lower grains yield and lower nitrogen, phosphorus, and potassium contents of rice grains than the WT under drought stress. Moreover, the sapk2 mutant exhibited weaker seedling growth and root development in hydroponic cultures lacking N and K. The NO 3 − influx rate and nitrate concentration analysis indicated that SAPK2 promotes nitrate uptake and assimilation by regulating nitrate-related transporters. These results suggest that SAPK2 could enhance grain production by regulating nitrogen utilization efficiency under drought stress. Our work provided insights to breeding drought tolerant rice with high nutrient uptake.      shown as means ± SD (n = 20) from three replicates. A student's t-test was used to generate P values; "**" indicate significance at P < 0.01.

Supplementary Files
28 This is a list of supplementary files associated with this preprint. Click to download.