OsCSLD1 Mediates NH4+-Dependent Root Hair Growth Suppression and AMT1;2 Expression in Rice (Oryza sativa L.)

Root hairs play crucial roles in the roots, including nutrient uptake, water assimilation, and anchorage with soil, along with supporting rhizospheric microorganisms. In rice, ammonia uptake is mediated by a specialized ammonium transporter (AMT). AMT1;1, AMT1;2, and AMT1;3 have been extensively studied in relation to nitrogen signaling. Cellulose synthase-like D1 (CSLD1) is essential for cell expansion and is highly specific to root hair cells. csld1 mutants showed successful initiation but failed to elongate. However, when nitrogen was depleted, csld1 root hairs resumed elongation. Further experiments revealed that in the presence of ammonium (NH4+), csld1 roots failed to elongate. csld1 elongated normally in the presence of nitrate (NO3−). Expression analysis showed an increase in root hair-specific AMT1;2 expression in csld1. CSLD1 was positively co-expressed with AMT1;2 changing nitrogen concentration in the growth media. CSLD1 showed increased expression in the presence of both ammonium and nitrate. Methylammonium (MeA) treatment of CSLD1 overexpression lines suggests that CSLD1 does not directly participate in nitrogen transport. Further studies on the root hair elongation mutant sndp1 showed that nitrogen assimilation is unlikely to depend on root hair length. Therefore, these results suggest that CSLD1 is closely involved in nitrogen-dependent root hair elongation and regulation of AMT1;2 expression in rice roots.


Introduction
Root hairs are tube-like outgrowths of root epidermal cells, a functionally essential part of the root system in plants that serve as the interface for nutrient uptake, water assimilation, and anchorage with soil. Root hairs aid these functions by effectively increasing the contact surface of the plant with the soil-particle interface [1]. Root hair cells are characterized by the release of organic compounds (exudates) into the rhizosphere, which can shape the microbial community around the plant root [2]. Root hairs develop from a group of specific root epidermal cells called trichoblasts, in contrast to atrichoblasts, which are not involved in their formation [3]. However, the arrangements of these cells are species-dependent and can be classified into three types. Type I, such as in rice, where trichoblasts and atrichoblasts are arranged randomly to facilitate a plastic response under alternating surrounding conditions. Type II patterning can be seen in grasses such as Brachypodium, where trichoblasts are derived from asymmetric cell division, larger cells become non-hair-forming cells and smaller cells live to produce root hair. Type III patterning is widely seen in Brassicaceae species such as Arabidopsis, where the hair-forming and non-hair-forming cells are organized as cell files inter-spread with other cell files [1,4,5]. However, cell patterning and cell fate is defined by positional information, not by cell lineage [6]. Root hair development is regulated by genetic and environmental factors. Root hair development occurs in three steps: root hair cell specification, initiation, and elongation. The process of cell growth or elongation is an expansion of the cell wall relative to the protoplast. During expansion, essential components such as cellulose and hemicellulose are synthesized at the plasma membrane. Pectin and xyloglucan were synthesized in the In Arabidopsis, CAP1 has shown to regulate root hair tip growth by regulating cytoplasmic Ca 2+ gradients along root hair cells. Mutants of CAP1 did not produce normal root hairs on Murashige and Skoog (MS) medium. However, they resumed regular root hair growth when NH4 + was depleted [35]. Moreover, AMT1;2 expression was increased under phosphorus deficiency which indicates a relationship between nutrient availability and AMT1s [36]. Root-hair-specific oscsld1 mutants also exhibited regular root hair initiation and failed to elongate root hair under regular conditions. In this study, we established a relationship between NH4 + and NO3 − -mediated CSLD1-dependent root hair elongation in rice (Oryza sativa L.).

NH4 + Suppresses Root Hair Elongation of csld1
Since csld1 mutants successfully initiated root hairs in the seminal roots and only failed to elongate root hairs [25], csld1 mutants were grown without NH4 + and/or NO3 − to examine any effect of CSLD1 on nitrogen-mediated root hair elongation. Initially, root hairs failed to elongate in half-MS media and without CaCl2 ( Figure 1A and B) but csld1 root hairs demonstrated an increase in length when the supply of NO3 − was depleted (Figure 1C). When NH4 + was absent, the roots of csld1 recovered their length to the wild-type root hairs ( Figure 1D). To verify the effect of NH4 + on csld1 root hairs, mutants were grown in NH4 + -deficient media. In the absence of NH4 + alone, root hair length of clsd1 significantly increased relative to the standard half-MS media counterpart. However, the wild-type roots did not show significant differences between standard and modified half-MS media ( Figure  2A,B). Investigation of the expression levels of CSLD genes by qRT-PCR revealed that in wild-type roots, expression of CSLD3 and CSLD4 expression was increased in the presence of NH4 + . In csld1 mutants, the expression of CSLD2, CSLD3, and CSLD4 were significantly increased under normal conditions. In the presence of NH4 + , the csld1 mutants did not show any significant differences in expression in any other CSLD genes ( Figure 2C). To verify the effect of NH 4 + on csld1 root hairs, mutants were grown in NH 4 + -deficient media. In the absence of NH 4 + alone, root hair length of csld1 significantly increased relative to the standard half-MS media counterpart. However, the wild-type roots did not show significant differences between standard and modified half-MS media (Figure 2A,B). Investigation of the expression levels of CSLD genes by qRT-PCR revealed that in wild-type roots, expression of CSLD3 and CSLD4 expression was increased in the presence of NH 4 + . In csld1 mutants, the expression of CSLD2, CSLD3, and CSLD4 were significantly increased under normal conditions. In the presence of NH 4 + , the csld1 mutants did not show any significant differences in expression in any other CSLD genes ( Figure 2C). in WT and oscsld1 mutants in MS and modified MS medium without NH4 + . Scale bars = 500 μm. Ba graphs show mean values ± standard error of means (n = 9). Different letters indicate significan differences between groups according to Tukey's HSD test (p < 0.05).
To distinguish the effects of different nitrogen sources on root hair elongation, csld1 mutants were grown in KNO3 and NH4NO3. In the presence of KNO3, csld1 produced roo hairs that were similar in lengths that of the wild-type root hairs. On the other hand, when NH4NO3 was added to the media, csld1 mutants showed significant reduction in root hai length ( Figure 3A,B). It is noteworthy that the reduction of root hair length in csld1 mu tants in the presence of NH4NO3 was similar to the root hair length of the wild-type roo hair grown in KCl. in WT and oscsld1 mutants in MS and modified MS medium without NH 4 + . Scale bars = 500 µm. Bar graphs show mean values ± standard error of means (n = 9). Different letters indicate significant differences between groups according to Tukey's HSD test (p < 0.05).
To distinguish the effects of different nitrogen sources on root hair elongation, csld1 mutants were grown in KNO 3 and NH 4 NO 3 . In the presence of KNO 3 , csld1 produced root hairs that were similar in lengths that of the wild-type root hairs. On the other hand, when NH 4 NO 3 was added to the media, csld1 mutants showed significant reduction in root hair length ( Figure 3A,B). It is noteworthy that the reduction of root hair length in csld1 mutants in the presence of NH 4 NO 3 was similar to the root hair length of the wild-type root hair grown in KCl.
Following the observation from KNO 3 and NH 4 NO 3 treatments, wild-type and mutant seedlings were grown in NH 4 Cl or KNO 3 to distinguish the effect of both NH 4 + and NO 3 − . In the presence of NH 4 + csld1 mutants showed significant reduction in root hair length when compared with the wild-type root hairs. However, in the presence of NO 3 − , both wild-type and csld1 root hairs elongated up to twice that of the wild-type root hairs supplemented with NH 4 + and showed significant difference ( Figure 4A,B), suggesting that csld1 mutants are not sensitive to NO 3 − but NH 4 + and NO 3 − have more influence on root hair elongation than NH 4 + . To distinguish the effects of different nitrogen sources on root hair elongation, csld1 mutants were grown in KNO3 and NH4NO3. In the presence of KNO3, csld1 produced root hairs that were similar in lengths that of the wild-type root hairs. On the other hand, when NH4NO3 was added to the media, csld1 mutants showed significant reduction in root hair length ( Figure 3A,B). It is noteworthy that the reduction of root hair length in csld1 mutants in the presence of NH4NO3 was similar to the root hair length of the wild-type root hair grown in KCl.   Following the observation from KNO3 and NH4NO3 treatments, wild-type and mutant seedlings were grown in NH4Cl or KNO3 to distinguish the effect of both NH4 + and NO3 − . In the presence of NH4 + csld1 mutants showed significant reduction in root hair length when compared with the wild-type root hairs. However, in the presence of NO3 − , both wild-type and csld1 root hairs elongated up to twice that of the wild-type root hairs supplemented with NH4 + and showed significant difference ( Figure 4A,B), suggesting that csld1 mutants are not sensitive to NO3 − but NH4 + and NO3 − have more influence on root hair elongation than NH4 + .

CSLD1 and AMT1:2 Show Close Relationship in Activity under Different NH4 + Concentrations
Differential expression of root specific ammonium transporters such as AMT1;2 under different nutrient states were observed in previous studies. To identify any relationship between CSLD1 and the three AMT1s, the expression levels of AMT1 genes were examined in WT and csld1 mutants. In csld1 mutants, only AMT1;2 showed significant increase in expression compared with the wild-type, indicating a negative relationship between AMT1;2 and CSLD1 ( Figure 5A).
Under nitrogen starvation, Arabidopsis root hairs show an increased length [37], suggesting a possible significant correlation between nitrogen signaling and csld1 expression. Expression analyses for CSLD1 and AMT1;2 were performed to determine their expression levels under different nitrogen levels. When seedlings were transplanted to 1 mM NH4 + during nitrogen starvation, the initial expression levels of CSLD1 and AMT1;2 were similar. However, AMT1;2 expression was significantly higher than CSLD1 3 h after transfer. After 6 h, both CSLD1 and AMT1;2 expression levels decreased, while AMT1;2 expression was maintained at higher levels and CSLD1 expression was significantly reduced

CSLD1 and AMT1:2 Show Close Relationship in Activity under Different NH 4 + Concentrations
Differential expression of root specific ammonium transporters such as AMT1;2 under different nutrient states were observed in previous studies. To identify any relationship between CSLD1 and the three AMT1s, the expression levels of AMT1 genes were examined in WT and csld1 mutants. In csld1 mutants, only AMT1;2 showed significant increase in expression compared with the wild-type, indicating a negative relationship between AMT1;2 and CSLD1 ( Figure 5A). was applied to 1/10 Johnson's solution, GUS expression was increased in trichoblasts relative to the controls (KCl), indicating the induction of CSLD1 expression in the presence of NH4 + and NO3 − ( Figure 6).  Under nitrogen starvation, Arabidopsis root hairs show an increased length [37], suggesting a possible significant correlation between nitrogen signaling and csld1 expression. Expression analyses for CSLD1 and AMT1;2 were performed to determine their expression levels under different nitrogen levels. When seedlings were transplanted to 1 mM NH 4 + during nitrogen starvation, the initial expression levels of CSLD1 and AMT1;2 were similar. However, AMT1;2 expression was significantly higher than CSLD1 3 h after transfer. After 6 h, both CSLD1 and AMT1;2 expression levels decreased, while AMT1;2 expression was maintained at higher levels and CSLD1 expression was significantly reduced ( Figure 5B). However, CSLD1 and AMT1;2 expression levels increased simultaneously, with significantly (but only diverging in one point i.e., 48 h) increasing expression levels when seedlings were transferred from high to low NH 4 + levels ( Figure 5C). To visualize the expression of CSLD1 in trichoblasts in the presence of NH 4 + , GUS expression patterns were observed under CSLD1::Ds/CSLD1::Ds (csld1) background. When NH 4 Cl or KNO 3 was applied to 1/10 Johnson's solution, GUS expression was increased in trichoblasts relative to the controls (KCl), indicating the induction of CSLD1 expression in the presence of NH 4 + and NO 3 − ( Figure 6).

Ammonium Uptake Is not Directly Correlated with Root Hair Length
Similar to oscsld1 root hairs, ossndp1 mutants also exhibited reduced root hair length caused by the depolarization of root hair growth orientation ( Figure 7A) [12]. In order to examine the impact of root hair length and roles of OsCSLD1 and OsSNDP1 mutants related to nitrogen transport, each mutant was treated with the toxic analog of ammonia, methyl ammonium (MeA) in MS media. In the absence of MeA, both csld1 and sndp1 seminal root lengths were similar but significantly shorter than the wild-type. However, in the presence of MeA, csld1 seminal roots resisted the effects of MeA. On the other hand, WT and sndp1 seminal roots were susceptible to MeA and showed reduction in root length ( Figure 7B,C). When examined for ammonium concentration in roots in MS media, shorter sndp1 roots showed similar concentrations to that of the wild-type. The csld1 roots contained less ammonium than the WT and sndp1 ( Figure 7D). of CSLD1 and AMT1;2 when transferred from nitrogen starvation to 1 mM NH4NO3 and (C) when transferred from high (10 mM) to low (1 mM) concentration of NH4NO3. Bar graphs show mean values ± standard error of means (n = 3). Different letters indicate significant differences between groups according to Tukey's HSD test (p < 0.05).

CSLD1 Is not Directly Involved in Ammonium Uptake but Involved in Ammonium Response
CSLs are characterized to directly involve in cell wall biosynthesis and patterning. On the other hand, AMTs are the primary transporters of ammonia and nitrate. In order to verify the effect of NH 4 + or NO 3 − on root hair growth in AMT1 RNAi mutants in which AMT1;1, AMT1;2, and AMT1;3 were suppressed, seedlings were germinated and grown in half-MS medium with 0.1 mM NH 4 + or NO 3 − ( Figure 8A). Mutants showed significantly strong reductions in root hair length and density ( Figure 8B,C). In contrast, seedlings grown in 0.1 mM NO 3 − failed to show any significant difference in root hair length and density ( Figure 8D-E).
To further describe the role of CSLD1 in ammonium signaling, CSLD1:OX ( Figure 9A) seedlings were grown in methylammonia (MeA). CSLD1:OX lines were sensitive to MeA showing reduction in length and csld1 seminal roots showed no response to MeA when compared with wild-type seminal roots ( Figure 9B,D). However, AMT1;2:OX lines showed a dramatic reduction in seminal root length relative to the wild-type in the presence of MeA ( Figure 9C,E). This observation was further validated when csld1 and CSLD1:OX seedlings were grown in MeA where csld1 seminal roots showed relatively long lengths under increasing MeA concentration ( Figure S1). This indicates a strong resistance of csld1 for MeA and enhanced susceptibility of CSLD1:OX.
methyl ammonium (MeA) in MS media. In the absence of MeA, both csld1 and sndp1 seminal root lengths were similar but significantly shorter than the wild-type. However, in the presence of MeA, csld1 seminal roots resisted the effects of MeA. On the other hand, WT and sndp1 seminal roots were susceptible to MeA and showed reduction in root length ( Figure 7B,C). When examined for ammonium concentration in roots in MS media, shorter sndp1 roots showed similar concentrations to that of the wild-type. The csld1 roots contained less ammonium than the WT and sndp1 ( Figure 7D).

CSLD1 Is not Directly Involved in Ammonium Uptake but Involved in Ammonium Response
CSLs are characterized to directly involve in cell wall biosynthesis and patterning. On the other hand, AMTs are the primary transporters of ammonia and nitrate. In order to verify the effect of NH4 + or NO3 − on root hair growth in AMT1 RNAi mutants in which AMT1;1, AMT1;2, and AMT1;3 were suppressed, seedlings were germinated and grown  To further describe the role of CSLD1 in ammonium signaling, CSLD1:OX ( Figure  9A) seedlings were grown in methylammonia (MeA). CSLD1:OX lines were sensitive to MeA showing reduction in length and csld1 seminal roots showed no response to MeA
Bar graphs show mean values ± standard error of means (n = 10). Different letters indicate significant differences between groups according to Tukey's HSD test (p < 0.05).

Discussion
Among all the mineral necessities plants have developed to adopt, nitrogen plays a key role in all kinds of plant growth and development [38][39][40][41][42]. Among the major N sources in soil, such as nitrate and ammonia, the former tends to be most abundant in aerobic environments, and the latter is abundant in flooded conditions [43]. Higher levels of ammonia in anaerobic soils are highly toxic to plants. Rice plants have developed a well-regulated equilibrium of direct ammonia uptake mechanisms using ammonium transporters (AMT) to avoid the toxic effects of ammonia accumulation [44].
Initial experiments with csld1, NH4 + , and NO3 − suggested that ammonium ions significantly suppressed root hair elongation and nitrate ions enhanced the elongation of root hair even in the presence of ammonium ions, indicating that CSLD1 promotes root elongation in the presence of NH4 + rather than NO3 − (Figures 1 and 2A,B). A similar observation was observed in Arabidopsis Atrop11 CA mutants [45]. CSLDs are essential for root hair and pollen tube growth in Arabidopsis and synthesize polysaccharides essential for the cell wall structure in tip-growing cells [24]. Expression analysis of other CSLD genes suggested that CSLD1, CSLD3, and CSLD4 showed increased activity in the absence of NH4 + . In csld1 mutants, CSLD2, CSLD3, and CSLD4 showed increased expression but reduced expression when NH4 + was absent ( Figure 2C). Based on these results, it is highly likely that CSLD1 promotes the expression of other CSLDs when NH4 + was absent and suppresses other CSLDs when ammonia is present. Moreover, it is possible that CSLD1 does not conduct a regulative process alone. Root morphology can be shaped by the direct and indirect effects of NH4 + and NO3 − [46]. When NH4 + and NO3 − were present in the media, csld1 mutants showed significant reduction in root hair length. When NO3 − was present, root hair length was similar to the wild-type. When NH4 + showed strong suppression of root hair length on clsd1 mutants, wild-type root hair also showed significant reduction in

Discussion
Among all the mineral necessities plants have developed to adopt, nitrogen plays a key role in all kinds of plant growth and development [38][39][40][41][42]. Among the major N sources in soil, such as nitrate and ammonia, the former tends to be most abundant in aerobic environments, and the latter is abundant in flooded conditions [43]. Higher levels of ammonia in anaerobic soils are highly toxic to plants. Rice plants have developed a well-regulated equilibrium of direct ammonia uptake mechanisms using ammonium transporters (AMT) to avoid the toxic effects of ammonia accumulation [44].
Initial experiments with csld1, NH 4 + , and NO 3 − suggested that ammonium ions significantly suppressed root hair elongation and nitrate ions enhanced the elongation of root hair even in the presence of ammonium ions, indicating that CSLD1 promotes root elongation in the presence of NH 4 + rather than NO 3 − (Figures 1 and 2A,B). A similar observation was observed in Arabidopsis Atrop11 CA mutants [45]. CSLDs are essential for root hair and pollen tube growth in Arabidopsis and synthesize polysaccharides essential for the cell wall structure in tip-growing cells [24]. Expression analysis of other CSLD genes suggested that CSLD1, CSLD3, and CSLD4 showed increased activity in the absence of NH 4 + . In csld1 mutants, CSLD2, CSLD3, and CSLD4 showed increased expression but reduced expression when NH 4 + was absent ( Figure 2C). Based on these results, it is highly likely that CSLD1 promotes the expression of other CSLDs when NH 4 + was absent and suppresses other CSLDs when ammonia is present. Moreover, it is possible that CSLD1 does not conduct a regulative process alone. Root morphology can be shaped by the direct and indirect effects of NH 4 + and NO 3 − [46]. When NH 4 + and NO 3 − were present in the media, csld1 mutants showed significant reduction in root hair length. When NO 3 − was present, root hair length was similar to the wild-type. When NH 4 + showed strong suppression of root hair length on csld1 mutants, wild-type root hair also showed significant reduction in length but not up to the degree of csld1 root hair length (Figures 3 and 4A,B). It is evident that both NH 4 + and NO 3 − produce opposite signals for root hair elongation where NH 4 + results in suppressing signals and NO 3 − results in promoting signals. Both signals converge into the same regulatory pathway to elongate root hair in a dose-dependent manner.
AMT1s act as sensors for ammonium sources available in the media [47]. AMT1;1, AMT1;2, and AMT1;3 are the most essential physiological and morphological responses to external N signals. Overexpression of AMT1;1 presented improved growth and yield [33]. In contrast, overexpression of AMT1;3 reduced the performance [48]. Current experiments with AMT1 mutants revealed that the mutants had poorly elongated roots in the presence of NH 4 + . However, these mutants did not respond to NO 3 − . Csld1 mutants showed increased expression of AMT1;2, suggesting that CSLD1 is a negative regulator of root hair-specific AMT1;2 expression. The expression dynamics of CSLD1 and AMT1;2 showed similar patterns. When plants were exposed to NH 4 + after N starvation, both CSLD1 and AMT1;2 showed higher activity indicating CSLD1 and AMT1;2 coupled response in progress. Similar patterns were observed for CSLD1 and AMT1;2 when plants were transferred from high-to low-N medium, where the expression progressively increased up to 72 h after transfer.
Experiments on roots with GUS staining revealed that when presented with NH 4 + or NO 3 − , CSLD1 showed activity limited to trichoblast cells. This observation is in agreement with previous studies in regular conditions, CSLD1 is expressed in root hair cells [25]. Unlike CSLD1, the short root hair mutant sndp1 [12] showed hindered root hair elongation when grown in MeA media and csld1 showed strong resistance to MeA treatment relative to wild-type. This suggests that CSLD1 is involved in the ammonia response process. It is noteworthy that sndp1 had ammonium concentrations similar to those of the wild-type. csld1 had lower ammonium concentrations, implying that root hair length does not affect ammonium content or ammonium absorption capacity of root hair.
The toxic ammonia analog, methyl-ammonium (MeA), can be used to identify ammonium transport [49,50]. Overexpression of ammonium-inducible AMT1;2 is highly sensitive to MeA. CSLD1 overexpressing lines showed higher degree of susceptibility to MeA and csld1 mutants showed significant resistance to MeA, indicating that CSLD1 is not a part of ammonium uptake (Figure 9). These observations were further validated when csld1 and CSLD1:OX lines were grown in the presence of MeA. In this case, csld1 seminal roots remained longer than wild-type and CSLD1:OX roots when exposed to increasing concentrations of MeA ( Figure S1). The results from these experiments suggest that CSLD1 is involved in the response to NH 4 + -triggered root hair elongation. These results also indicate that, unlike AMT1s, CSLD1 is not directly involved in N signaling in rice. Further studies are necessary to explore the possibility of CSLD1 interacting with AMT1s and the role of N signals in the dynamics of AMT1s and CSLD1 related to root hair elongation. Current experiments suggest that the length of root hair is not related to nitrogen uptake in in vitro conditions. However, the conditions in soil can be different and root hair length may play a significant role in nitrogen uptake in soil. Studies must be also done to explore possible tools increase the nitrogen use efficiency by varying root hair length.

Mutant Materials
A large-scale Ds transposon population was generated by regenerating gene trap Ds starter lines transferred by Ac via tissue culture [51]. Previously characterized csdl1 and sndp1 seeds from the gene trap mutants were kindly provided by Han's lab, Gyeongsang University, Republic of Korea [12,25].

Plant Materials and Growth Conditions
The japonica cultivar Dongjin (WT; Wild-type), csld1, 35s::CSLD1, three 5 AMT1 RNAi lines (3-1 and 5-2), and sndp1 were used in the experiments. Rice seeds were surfacesterilized with 0.05% SPORTEX and germinated for three days in the dark. Uniformly germinated seedlings were selected and cultured hydroponically in different nutrient solutions [ 1 4 MS (Murashige and Skoog), 1 4 KB (Kimura B), and 1 4 NS (Nutrient Solution)] containing NH 4 + or NO 3 − as the sole nitrogen source. Media compositions were obtained from previous studies [46,52,53]. Hydroponic nutrient solutions were replaced with fresh media every two days for two weeks. Roots were examined after culturing for 14 days in a growth chamber under the following conditions: 16/8 h light/dark, light intensity 280 µmol m −2 s −1 , temperature 26 • C/18 • C, and 70% humidity. To measure the expression levels of three OsAMT1 (1;1, 1;2, and 1;3) and four CSLD (CSLD1, CSLD2, CSLD3, and CSLD4) genes, WT and csld1, seedlings were grown hydroponically in 1 4 nutrient medium supplemented with 0.5 mM NO 3 − or 0.5 mM NH 4 + for seven days. Total cellular RNAs were extracted from roots. For the media shift assays, germinated seeds were cultured in modified 1 4 NS containing 0.1 mM NH 4 NO 3 for seven days. The samples were transferred and cultured in a nutrient solution containing no NH 4 NO 3 for nitrogen starvation and for moving low to high nitrogen medium. Germinated seeds were first grown in 0.1 mM NH 4 NO 3 , moved to 1 mM, and 10 mM NH 4 NO 3 containing media after seven days. The same solutions were replaced with fresh medium every two days.

Isolation of Full-Length cDNA of CSLD1 and Agrobacterium Transformation
The 4.0-kb full-length OsCSLD1 cDNA was isolated by PCR amplification using primers OsCSLD1-5#SpeI (ACTAGTATGGCGTCGAAGGGCATCCTC-AAG) and OsCSLD1-3#SpeI (ACTAGTCCAGGGGAAAGAGAAGGATCC-TCC). The PCR product was ligated into a pGEM-T vector (Promega; Madison, WI, USA) and sequenced. A 3.4-kb fragment was excised from the vector by SpeI digestion and ligated into the corresponding site of pCAMBIA1302. Full-length OsCSLD1 cDNA was fused with GFP at its 3 end and expressed under the cauliflower mosaic virus 35S promoter and nopaline synthase 3 terminator. Calli of rice (Dongjin) were transformed with Agrobacterium LEA4404 carrying pCAMBIA (35S::OsCSLD1). Rice calli (Dongjin) were transformed with T-DNA carrying the hygromycin phosphotransferase gene as previously described method [54], with slight modifications.

Construction of AMT1 RNAi Vector
To generate AMT1;1 RNAi transgenic plants, 5 and 3 fragments of the AMT1;1 ORF were amplified using the following primer sets: Ri5-F (gagctcggtaccctcgccgcgcacgtcatccag) and Ri5-R (gaattcctgcaggcatgtgcttgaggccgaaga) and Ri3-F (gagctcggtaccctcgcggcgcacatcgtgcag) and Ri3-R (gaattcctgcagttacacttggttgttgctgtt), respectively. The PCR products were digested and cloned into EcoRI and SacI sites for sense orientation insertion and KpnI and XhoI sites for antisense orientation insertion in a pBluscript-catalase intron vector. After sequencing, whole inserts were cloned into SacI and KpnI sites of PGA1611 binary vector.

Generation of Transgenic Overexpression Plants
To generate plants overexpressing AMT1;2, ORFs were cloned into BamHI and HindIII restriction sites of the pGA1611 binary vector to produce pGA1611-AMT1;2, in which AMT1;2 was expressed under the control of the UBIQUITIN promoter. Rice calli were transformed with pGA1611-AMT1;2 using Agrobacterium-mediated transformation.

Histochemical Analysis
β-Glucuronidase (GUS) activity was visualized by incubating 5-bromo-4-chloro-3indoyl glucuronide solution (1 mg mL − 1 ). The solution contained 50 mM sodium phosphate buffer (pH 7.0), 10 mM EDTA, 0.1% Triton, 2 mM potassium ferrocyanide, and 200 mg/mL chloramphenicol. Samples were incubated at 37 • C for two days in the dark and dehydrated in a 30 to 70% graded ethanol series. GUS-stained samples were dehydrated and embedded in paraffin wax using the method described above. For toluidine blue staining, roots grown on 0.5 X MS medium were stained with 0.1% aqueous toluidine blue for 2 min and washed with distilled water. The roots on the slides were inspected under a light microscope [25].

RNA Extraction and qRT-PCR
Total cellular RNA was isolated using a Qiagen RNeasy Mini Kit (Qiagen, Valencia, CA, USA) following the manufacturer's instructions. RNA concentration was measured using a NanoDrop ND-1000 spectrophotometer. The samples were treated with RQ-RNase-free DNase (Promega, Madison, WI, USA). An RNaseH (Toyobo, https://www. toyobo-global.com/) reverse transcription kit was used to synthesize cDNA, according to the manufacturer's instructions (Promega). qRT-PCR was performed using iQ SYBR Green Supermix (Bio-Rad, Hercules, CA, USA). Amplification and quantification were performed with gene-specific primers using CFX Manager software (Bio-Rad), and values were normalized against internal UBIQ1. Three biological and two technical replicates were used for each analysis [46]. All the primers used for qRT-PCR are listed in Table S1.

Root Hair Length Measurements
Five-day-old roots of rice plants grown in 1/2 MS solution were observed. The root images were captured with an Olympus SZX12 stereomicroscope system. Images were processed and analyzed using the ImageJ software version 1.53k (National institute of health, Bethesda, MD, USA, http://rsbweb.nih.gov/ij/).

Ammonium Content Determination
Determination of NH 4 + content in the roots by enzymatic digestion was performed using an F-kit (Roche) following the manufacturer's instructions [56].

Statistical Analysis
All experiments were completely randomized and at least eight replicates were used for the quantitative experiments. Data shown in the figures are expressed in average ±SD (standard deviations) of each group. Quantitative data were analyzed using the JMP 16.0.0 software package (SAS Institute, Cary, NC, USA). All statistical comparisons were performed using the "Fit Y by X" function. A one-way ANOVA followed by Tukey HSD was used to visualize the significance among the experimental groups means at the probability level of p < 0.05.

Conclusions
CSLD1 mediates ammonia-triggered root hair elongation in rice. Moreover, CSLD1 also suppresses root hair-specific AMT1;2 and coexpressed with AMT1;2 when nitrogen concentration regimes change. The expression of CSLD1 is confined to trichoblast cells and occurs under both ammonia and nitrate signals. Results with short root hair mutants, such as sndp1, showed that root hair length did not affect the absorption of nitrogen in root hairs. Further experiments using MeA suggested that CSLD1 does not directly participate in ammonium absorption. However, it is noteworthy that the mutants and overexpression lines of CSLD1 showed minor differences in seminal root length. These results suggest that NH 4 + can elongate root hair via CSLD1. Further studies are required to understand the mechanism of ammonia-triggered CSLD1-mediated root hair elongation.