MADS-box Transcription Factor OsMADS25 Regulates Root Development through Affection of Nitrate Accumulation in Rice

MADS-box transcription factors are vital regulators participating in plant growth and development process and the functions of most of them are still unknown. ANR1 was reported to play a key role in controlling lateral root development through nitrate signal in Arabidopsis. OsMADS25 is one of five ANR1-like genes in Oryza Sativa and belongs to the ANR1 clade. Here we have investigated the role of OsMADS25 in the plant’s responses to external nitrate in Oryza Sativa. Our results showed that OsMADS25 protein was found in the nucleus as well as in the cytoplasm. Over-expression of OsMADS25 significantly promoted lateral and primary root growth as well as shoot growth in a nitrate-dependent manner in Arabidopsis. OsMADS25 overexpression in transgenic rice resulted in significantly increased primary root length, lateral root number, lateral root length and shoot fresh weight in the presence of nitrate. Down-regulation of OsMADS25 in transgenic rice exhibited significantly reduced shoot and root growth in the presence of nitrate. Furthermore, over-expression of OsMADS25 in transgenic rice promoted nitrate accumulation and significantly increased the expressions of nitrate transporter genes at high rates of nitrate supply while down-regulation of OsMADS25 produced the opposite effect. Taken together, our findings suggest that OsMADS25 is a positive regulator control lateral and primary root development in rice.


Introduction
Nitrogen (N) is one of the essential macronutrients required by plants for normal growth and development and is frequently the major limiting factor for crop yields [1]. For higher plants the major source of N is usually in the form of nitrate (NO 3 -) [2,3]. One of the most important functions of NO 3 is to provide nitrogen for synthesis of amino acids and other forms of organic N [4]. In addition, NO 3 acts as a signal to regulate many metabolic and developmental processes such as transcription and translation, energy transfer, protein accumulation, cytokinin transport, seed germination, and plant growth and development [1,[5][6][7][8]. The NO 3 content of aerobic soils can vary markedly in time and space [2,3,9], requiring plants to evolve sophisticated signaling and transport processes to enable them to adjust to these variations. Rice is one of the most important food crops and N deficiency is considered as an important limiting factor affecting its productivity [2,3]. Although under flooded conditions rice mainly takes up N in the form of ammonium, NO 3 still contributes 15-40% of the total N absorbed by the rice crop [10,11]. Furthermore, NO 3 enhances the uptake and assimilation of ammonium by rice plants [11]. In plants, NO 3 is mainly taken up by two mechanisms, namely the high affinity uptake system (HATS) and low affinity uptake system (LATS) [12]. It has been suggested that two families of membrane proteins, the nitrate transporter 1 /peptide transporter family (NRT1/PTR) and nitrate transporter 2 (NRT2), are involved in NO 3 uptake by plants [12][13][14]. NRT1/PTR family is named unified NPF according to the phylogenetic relationship of these proteins [15]. NRT2 proteins are the high affinity nitrate transporters while most of the NPF family is low affinity nitrate [16][17][18][19].
A properly developed root system is essential to ensure the optimum uptake of water and mineral nutrients by plants [20,21]. As the key component of root system, lateral root (LR) initiation and development is affected by the combined actions of gene regulation, hormone and environmental signals such as light, water and nutrient [22][23][24][25]. In Arabidopsis, the ANR1 and AtABF3 genes are reported to be involved in distinct NO 3 signaling pathways regulating LR development [26,27]. AtNPF6.3 functions upstream of ANR1 in regulating LR elongation, apparently in its role as a NO 3 sensor [28,29]. Exogenous application of NO 3 has been reported to affect the expression of ANR1, a MADS-box transcription factor in regulating LR numbers and LR elongation [26]. ANR1 expression is induced by nitrate deprivation and constitutive over-regulation of ANR1 in roots of transgenic Arabidopsis increases LR growth while having no direct effect on LR density or primary root growth [30,31]. The root phenotype of the ANR1 overexpressing lines mainly depends on the presence of NO 3 -, suggesting that there other components involve in NO 3 --dependent signaling pathway.
Although NO 3 regulation of LR development has been extensively studied in Arabidopsis, little is known about this process in rice. miR444 has been reported to target four ANR1-like homologous genes (OsMADS23, OsMADS27a, OsMADS27b and OsMADS57) to regulate root development in rice [32][33][34][35][36]. miR444a regulates the NO 3 --signaling pathway in rice roots as well as regulating NO 3 accumulation and the response to phosphate starvation [37].
OsMADS25 is one of the five ANR1-like homologues in rice [32,33]. Previous studies reported that the expression of OsMADS25 is significantly induced by NO 3 -, salt and osmotic stress [38,39]. It was also reported that OsMADS25 is active in the central cylinder of the root and respond to auxin treatment [39]. To gain further insight into the possible regulatory functions of OsMADS25 in control root development in rice, we have investigated its regulatory role in root development through NO 3 regulation.

Experimental Procedures Plant material and growth conditions
Oryza sativa L. cv. Nipponbare was used as the wild type for both physiological and genetic transformation experiments. Rice sterilization, growth conditions and measurements were performed according to our previously reported study [38]. For NO 3 treatments, rice plants were grown in hydroponic culture or on Gelzan plates with modified 1/2 Murashige and Skoog salts in which KNO 3 and NH 4 NO 3 were replaced by KCl or KNO 3 [37,40]. To prepare cultures of different nitrate concentration, 0 mM KNO 3 , 0.2 mM KNO 3 and 10 mM KNO 3 were added to N-free medium [37].
For NH 4 + treatments, rice seedlings were germinated and grown on Gelzan plates with modified 1/2 Murashige and Skoog salts in which KNO 3 and NH 4 NO 3 were replaced by KCl or NH 4 Cl, respectively. To prepare cultures of different ammonium concentration, 0 mM NH 4 Cl, 0.5 mM NH 4 Cl and 5 mM NH 4 Cl were added to N-free medium [41].
To examine the NO 3 response of overexpressing OsMADS25 lines in Arabidopsis, surfacesterilized seeds were sown in 10×10 cm rectangular Petri dishes on medium containing 1% agar, 0.6%(w/v) sucrose, 1/50×B5 salts and 1 mM KCl and 1 mM glutamine was used to replace the nitrogen source of KNO 3 and (NH 4 ) 2 SO 4 [31]. 7-day-old seedlings were transferred to fresh plates containing (in addition to 1 mM glutamine): 0 mM KNO 3 , 0.2 mM KNO 3 , 2 mM KNO 3 or 10 mM KNO 3 as N source [31]. Images taken at growth intervals of 16 d (no NO 3 -), 14 d (0.2 and 2 mM NO 3 -), and 13 d (10 mM NO 3 -) were used to analyze different root parameters [26,31]. To investigate whether overexpression of OsMADS25 affected early seedling development, surface-sterilized seeds were sown in 10×10 cm rectangular Petri dishes including 0, 0.2, 2 or 10 mM KNO 3 [31]. After 2 d at 4°C, the plates were kept vertically at 22°C under the 16 h-light/8 h-dark light regime. Primary root length and the length of the first lateral root emerged were determined from images taken at 6 d, 8 d, 10 d, 12 d and 14d after sowing.

Gene constructs and generation of transgenic plants
For the overexpression construct, a full-length OsMADS25 cDNA was PCR-amplified and digested with restriction enzymes Sal I and Sma I for cloning into the pSB130-actin-NOS vector (a generous gift of Jumin Tu, Zhejiang University, China). To construct the RNA interference (RNAi) vector, a 259-bp cDNA fragment of OsMADS25 was amplified and inserted into the BamH I and Kpn I sites (for the reverse insert) and the Sac I and Spe I sites (for the forward insert) in the pTCK303 vector [42]. These constructs were transformed into rice using Agrobacterium tumefaciens EHA105 as previously described [43]. All transgenic lines were first selected based on the expression level of OsMADS25 and further confirmed by their phenotype.
To construct 35S::OsMADS25 for transformation into Arabidopsis, the 684 bp of the ORF were amplified by PCR and digested with Sal I and Not I and cloned into pENTR-1A vector. Subsequently the construct was recombined into pH2GW7 using the Gateway 'LR reaction' [44]. The binary vector construct was introduced into Agrobacterium strain GV3101 and Arabidopsis Col-0 plants were transformed by employing the floral dip method [45].
For the OsMADS25::YFP fusion, the 684 bp of the ORF was amplified by PCR and introduced in frame, after the YFP reporter gene of the 35S-pCAMBIA1300-YFP vector (a generous gift of Jumin Tu, Zhejiang University, China). Then the prepared construct was introduced into Agrobacterium tumefaciens strain EHA105. The recombinant constructs and free YFP (under the control of the 35S promoter) were introduced into epidermal cells of tobacco (Nicotiana tabacum) leaves which expressed the red fluorescent protein (RFP)-H2B by agroinfiltration [46,47], and the epithelial tissue was examined using a scanning confocal laser microscope (Zeiss LSM 510) with a filter set for YFP fluorescence (514 nm for excitation and 525 nm for emission) and for RFP-fusion proteins using 543 nm laser lines. Primers used in the assembly of these constructs are listed in Table 1.

qRT-PCR analysis
Total RNA was isolated using the RNAiso Plus reagent from rice roots and shoots according to the manufacturer's instructions. The first-strand cDNA was synthesized using the Reverse Transcriptase M-MLV from 2 μg total RNA in a 25 μl reaction, and diluted 4-fold with RNasefree water. Quantitative real-time RT-PCR was performed using SYBR Premix Ex Taq II as described in previous study [38,48]. Expression of OsActin (Os03g0718100) was used as a reference to normalize expression of the other genes. Three biological replicates were performed for each RT-PCR experiment. Primers used for qRT-PCR are listed in Table 1.

Measurement of tissue nitrate concentrations
Nitrate concentration was determined according to the method previously reported [37,49]. Samples of root and shoot tissue (~2 g) were collected and immersed in 10 ml deionized water and heated at 100°C for 20 min. After cooling to room temperature, deionized water was added to the suspension to a 25 ml final volume. The suspension was centrifuged at 7000 g for 15 min and 0.1 ml supernatant was mixed with 0.4 ml 5% (w/v) salicylic acid in concentrated H 2 SO 4 . After 20 min at room temperature, 9.5 ml 8% (w/v) NaOH was added slowly into the mixture and after cooling again to room temperature and the absorbance of the samples was measured for absorbance readings at 410 nm wave length.

Statistics
The results were analyzed by means of ANOVA for significance by IBM SPSS Statistics 21. Student's t-test was analyzed to evaluate the significant difference between treatments at the probability at either 5% (P<0.05 with significant level Ã ) or at 1% (P<0.01 with significant level ÃÃ ) as we previously described [50,51].

Subcellular localization of OsMADS25
To investigate the intracellular localization of OsMADS25, its cDNA sequence was fused to the coding sequence of yellow fluorescent protein (YFP) under the control of the 35S promoter. The 35S:: YFP:: OsMADS25 construct and free YFP (under the control of the 35S promoter) alone were expressed in tobacco epidermal leaf cells which expressing RFP:H2B nuclear marker. The yellow fluorescent signals from the free YFP and YFP:: OsMADS25 were overlapped with the red fluorescent signal from the RFP-H2B (Fig 1). The confocal images in Fig 1, showed that fluorescence associated with expression of the YFP-OsMADS25 fusion protein was detected in both nucleus and cytoplasm.

Effect of OsMADS25-overexpression on root development in Arabidopsis
Since OsMADS25 is not only closely related to ANR1 but is also inducible by nitrate [38,39], we investigated its possible role in NO 3 regulation of root architecture, initially using Arabidopsis as a model system. We created 35S::OsMADS25 overexpression transgenic lines and selected three representative transgenic lines that showed significantly higher OsMADS25 Table 1. Primer sequences used in this study.
Primer name Sense primer (5'->3') Anti-sense primer (5'->3')   per unit primary root length to minimize effects that might arise from differences in the rate of PR growth [31]. Overexpression of OsMADS25 in Arabidopsis significantly increased shoot and root fresh weight in the presence and absence of NO 3 -, but this positive effect was much stronger in the presence of NO 3 - (Fig 3D and 3E).
To investigate whether OsMADS25 overexpression in Arabidopsis affected early seedling development, transgenic lines (OE25-18, OE25-22 and OE25-23) were germinated and grown on medium containing four different concentrations of nitrate. Primary root length and the . Surface-sterilized seeds were sown in 10×10 cm rectangular Petri dishes on medium without nitrate and 7-d-old seedlings were transferred to fresh plates containing various concentrations of nitrate. Images were taken at different time intervals for measurement of root parameters. Errors indicate standard deviation (SD; n = 12). A Student's t-test was employed to calculate the significant difference between treatments at the probability of 5% (* p< 0.05) or 1% (**, P< 0.01).

OsMADS25 regulates root development through the NO 3 regulation in rice
Previous studies reported that the expression of OsMADS25 was significantly regulated by NH 4 + [38], we further investigated whether OsMADS25 regulates primary and lateral root development in response to ammonium. We tested the effects of OsMADS25 overexpression and its RNAi lines on root architecture under different concentrations of ammonium in rice. As shown in Fig 6, the numbers or length of adventitious roots, the PR length and the numbers or length of lateral roots were not significantly affected by either up-or down-regulation of OsMADS25 in the presence or absence of ammonium in comparison to the control line. Since OsMADS25 overexpression was able to promote LR growth and development in Arabidopsis (Fig 3), thus, we also want to know whether overexpression of OsMADS25 in rice would produce the similar effect. We created actin::OsMADS25 overexpression transgenic lines and obtained fifteen independent transgenic lines, from which we chose two representative transgenic lines (OEOsMADS25-3 and OEOsMADS25-18) and these two lines both showed significantly higher OsMADS25 expression levels compared to wild type (Fig 2B). We also generated sixteen OsMADS25-interrupting lines and chose two typical transgenic lines (RiOs-MADS25-3 and RiOsMADS25-4) which significantly suppressed the expressions of  OsMADS25 (Fig 2B). Homozygous seeds of overexpressing lines (OEOsMADS25-3 and OEOs-MADS25-18) and RNAi lines (RiOsMADS25-3 and RiOsMADS25-4) and the wild type were germinated and grown on agar plates for 8 d on 10 mM NO 3 or for 7 d on either 0.2 mM NO 3 or no nitrate. As shown in Figs 7 and 8, the LR lengths (per unit PR length) of the OEOs-MADS25-3 and OEOsMADS25-18 lines were significantly greater than those of the wild type, while those of RiOsMADS25-3 and RiOsMADS25-4 lines were significantly shorter under both high nitrate and low nitrate. The same effects, but on a smaller scale, were seen in the cases of PR length (about 10% increase and 15% decrease), LR number (about 20% increase and 20% decrease) and lateral root length (about 20% increase and 25% decrease) in the presence of nitrate, but the numbers or length of adventitious roots (AR) were not significantly affected by either up-or down-regulation of OsMADS25. No significant differences between wild type and transgenic lines were seen in the absence of nitrate. These results indicated that OsMADS25 positively regulates the primary and lateral root growth in NO 3 --regulation pathway in rice.

Effect of OsMADS25 overexpression and its down-regulation by RNAi on nitrate accumulation and shoot growth in rice
To examine the effect of over-expressing and down-regulating the expression of OsMADS25 on nitrate content and plant growth, Homozygous seeds of both wild type and transgenic plants (OEOsMADS25-3, OEOsMADS25-18, RiOsMADS25-3 and RiOsMADS25-4) were germinated and seedlings grown hydroponically for 14 d in medium containing 0, 0.2 or 10 mM KNO 3 . As shown in Figs 9A and 10, in the presence of nitrate, both shoot and primary root growth in OEOsMADS25-3 and OEOsMADS25-18 plants were significantly increased compared with the wild type, while down-regulation of OsMADS25 produced the opposite effect. However, no significant differences were observed in these parameters in the absence of nitrate. As seen in Fig 9B, at 10  higher nitrate content in the shoot and root while OsMADS25-interferringing plants produced significantly lower nitrate content in shoot and root than the wild plants. However, at 0.2 mM NO 3 supply, the nitrate content in the root not in the shoot of OsMADS25-overexpressing lines were significantly increased and nitrate content in root of OsMADS25 down-regulating plants was significantly decreased in comparison to the WT plants and there were no significant differences between these lines when nitrate was excluded from the medium. As the increased nitrate content in the OsMADS25-overexpressing lines and decreased nitrate accumulation in the OsMADS25-interfering plants under 10 mM KNO 3 , we investigated the expressions of four nitrate transporter genes. As shown in Fig 9C, the mRNA abundance of four nitrate transporter genes (OsNRT1;2, OsNRT2;1, OsNRT2;3a and OsNRT2;4) were significantly decreased in the both shoots and roots of RiOsMADS25-3 and RiOs-MADS25-4 lines compared with the wild type, while over-expression of OsMADS25 significantly increased the gene expressions of these four transporters.

Discussion
OsMADS25 may regulate plant growth and development and nitrate accumulation through NO 3 -

signaling pathway
The ANR1 MADS box gene has previously been identified as having a key role in the regulation of LR development by external nitrate [26,31]. Five ANR1-like genes have been identified in rice: OsMADS23, OsMADS25, OsMADS27, OsMADS57 and OsMADS61 [33]. Four of these (OsMADS23, OsMADS27a, OsMADS27b, and OsMADS57) are miR444a targets [32,33,35,36], although only the first three of these are reported to be expressed in roots [37,39]. It has been reported that miR444a acts as a negative regulator of the NO 3 --signaling pathway to modify LR development, leading to the suggestion that it acts by controlling the expression of its ANR1related targets in rice [37]. However the existence of additional targets of miR444a leaves open the possibility that other miR444a targets could be involved [35,36]. In this study, we observed that overexpression of OsMADS25 significantly promoted LR and PR growth in rice seedlings at early developmental stages, while interference with OsMADS25 expression had the opposite effect and that these responses were strongest in the presence of nitrate (Fig 7). These findings are similar to what was observed in ANR1-overexpressing lines of Arabidopsis, except that ANR1 overexpression specifically affected the LRs. In addition, we also found that there were no significant differences between wild type and transgenic seedlings under various concentrations of ammonium in rice (Fig 6). Previous studies reported that nitrate could act as a nutrient source as well as a signal to regulate gene expression, plant growth and development [5,37,52]. There observations indicate that the effect of altered OsMADS25 expression on root architecture was not accounted for nutrient regulation but for nitrate signaling. These results confirmed that OsMADS25 is a key transcriptional factor that controls root growth and development in rice and Arabidopsis by NO 3 -. Although miR444a-overexpression in rice affected adventitious root development [37], OsMADS25-overexpressing in rice had no significant effect on adventitious root development, which may indicate that they regulate rice root development in a different way (Fig 7).
In addition to its role in LR growth, the involvement of ANR1 in shoot growth and nitrate accumulation has been investigated in Arabidopsis [31]. Previous results showed that shoot fresh weight was increased in ANR1-overexpressing lines, with evidence that this was likely to be a secondary effect of the larger root system [31]. Here we found that both shoot and PR growth were decreased in OsMADS25-downregulated rice lines compared with the wild type (Fig 9) and that the RNAi lines showed a significant reduction in nitrate accumulations and in expressions of nitrate transporters (Fig 9). Previous studies have indicated that ANR1 acts as a positive regulator to promote the expression of the NRT2.1 -a high-affinity nitrate transporter in Arabidopsis [30]. Therefore it seems likely that OsMADS25 is also a positive regulator of nitrate accumulation in rice and that the nitrate content increases in shoot and root in the overexpressing line confirmed this role. In conclusion, this study characterized a novel MADSbox transcription factor OsMADS25, which plays a key role in regulation primary and lateral root development in rice. Overexpression of OsMADS25 significantly increased lateral and primary root growth in the presence of high (10 mM) or low concentration of nitrate (0.2 mM). These results suggested that OsMADS25 might positively regulate root and shoot development through NO 3 --regulation pathway in rice. Furthermore, overexpression of OsMADS25 altered the expressions of nitrate transporter genes, thus leading to increase nitrate accumulation under 10 mM NO 3 -. Further study will be needed to explore the molecular mechanism in details to explain the role of OsMADS25 in regulating shoot and root development through NO 3 --signaling pathway.