The interplay of auxin and brassinosteroid signaling tunes root growth under low and different nitrogen forms

Abstract

The coordinated signaling activity of auxin and brassinosteroids (BRs) is critical for optimal plant growth and development. Nutrient-derived signals regulate root growth by modulating the levels and spatial distribution of growth hormones to optimize nutrient uptake and assimilation. However, the effect of the interaction of these two hormones and their signaling on root plasticity during low and differential availability of nitrogen (N) forms (⁠${\text{NH}}_4^{+}$/${\mathrm{NO}}_3^{-}$⁠) remains elusive. We demonstrate that root elongation under low N (LN) is an outcome of the interdependent activity of auxin and BR signaling pathways in Arabidopsis (Arabidopsis thaliana). LN promotes root elongation by increasing BR-induced auxin transport activity in the roots. Increased nuclear auxin signaling and its transport efficiency have a distinct impact on root elongation under LN conditions. High auxin levels reversibly inhibit BR signaling via BRI1 KINASE INHIBITOR1. Using the tissue-specific approach, we show that BR signaling from root vasculature (stele) tissues is sufficient to promote cell elongation and, hence, root growth under LN condition. Further, we show that N form-defined root growth attenuation or enhancement depends on the fine balance of BR and auxin signaling activity. ${\text{NH}}_4^{+}$ as a sole N source represses BR signaling and response, which in turn inhibits auxin response and transport, whereas ${\mathrm{NO}}_3^{-}$ promotes root elongation in a BR signaling-dependent manner. In this study, we demonstrate the interplay of auxin and BR-derived signals, which are critical for root growth in a heterogeneous N environment and appear essential for root N foraging response and adaptation.

Introduction

Plant growth relies on the integrated activity of hormonal, developmental, and environmental signaling cascades (Gruber et al., 2013; Guan, 2017; Pandey et al., 2020). The heterogeneous distribution of nutrients in the rhizosphere and their ability or inability to make complexes with the soil particles often makes them unavailable for the plants (López-Bucio et al., 2003; Miller and Cramer, 2005; Thorup-Kristensen, 2006). Among the nutrients, nitrogen (N) deficiency or its low availability to plants is a common problem due to its inert behavior and high leaching activity (Nishimura and Itani, 2013; Vidal et al., 2020). The high diffusion rates of major N forms nitrate/ammonium (⁠${\mathrm{NO}}_3^{-}$/${\text{NH}}_4^{+}$⁠) in the environment lead to altered ${\mathrm{NO}}_3^{-}$/${\text{NH}}_4^{+}$ ratio or N deficiency (Hachiya and Sakakibara, 2017). To overcome N deficiency, plant roots undergo adaptive morphological changes such as an increased primary root (PR) and lateral root (LR) length to exploit the N from the rhizosphere. These changes in root system architecture (RSA) are also known as N foraging responses (Miller and Cramer, 2005; Nishimura and Itani, 2013). The ${\text{NH}}_4^{+}$/${\mathrm{NO}}_3^{-}$ have a distinct impact on root elongation. ${\mathrm{NO}}_3^{-}$ as a sole N form promotes PR elongation, whereas high ${\text{NH}}_4^{+}$ has an inhibitory effect on it (Qin et al., 2011; Hachiya et al., 2012; Nishimura and Itani, 2013). In the past few years, the increased use of chemical fertilizers to supplement the plant N demand have caused severe environmental issues including eutrophication (Tian and Niu, 2015; Gaudinier et al., 2018). Therefore, increasing N use efficiency by modulating the root traits would be an important strategy to reduce the use of fertilizers.

The phytohormones such as auxin and brassinosteroids (BRs) regulate several developmental and defense-related processes in the plants (Vert et al., 2008; Chaiwanon and Wang, 2015; Vragović et al., 2015; Singh et al., 2018). Plant adaptation to a differential or low availability of nutrients is linked with a change in biosynthesis and signaling of growth hormones such as auxin and BRs (Malamy and Ryan, 2001; Singh et al., 2014; Pandey et al., 2020; Zhang et al., 2021). Auxin perception by its co-receptor complex, TRANSPORT INHIBITOR RESISTANT1 (TIR1)/AUXIN SIGNALING F-BOX and AUXIN/INDOLE-3-ACETIC ACID proteins (Aux/IAAs) promote the ubiquitin-mediated degradation of corepressors, Aux/IAAs leading to the activation of auxin-responsive transcriptional effectors (Salehin et al., 2015). The N-terminus F box domain and a large leucine-rich repeat domain of TIR1 together form an auxin-binding pocket. Mutations in the F box domain of TIR1 lead to reduced auxin response and altered auxin-mediated growth processes (Yu et al., 2015).

The cellular auxin concentration and distribution along the root axis are important to modulate root developmental plasticity (Grieneisen et al., 2007). For example, auxin biosynthesis and its simultaneous tissue and cell-to-cell gradient by the plasma membrane (PM)-localized efflux and influx transporter family proteins, PIN-FORMED (PINs) and AUXIN1/LIKE-AUX1 (AUX1/LAX) determine the root growth and cell expansion (Grieneisen et al., 2007; Kitakura et al., 2011). Auxin effect in a dose-dependent manner has been shown to influence root growth by altering the transcriptional and posttranscriptional activity of PINs in a tissue-specific manner. High auxin represses PIN proteins posttranscriptionally (Vieten et al., 2005). The ${\text{NH}}_4^{+}$ as a sole N source represses auxin levels and genes involved in its distribution (PIN1, PIN2, and AUX1) in the roots suggesting the possible interaction of these genes in maintaining the auxin gradient during the differential availability of N forms (Zou et al., 2012; Li et al., 2018; Ötvös et al., 2021). In addition to auxin transport, the auxin biosynthesis gene, TRYPTOPHAN AMINOTRANSFERASE RELATED2 has been shown to regulate the LR initiation/elongation during N deficiency (Ma et al., 2014). The role of auxin has been well documented in LR organogenesis/elongation under N deficiency; however, its impact on PR elongation under low N (LN) availability is still cryptic.

Auxin biosynthesis and signaling have been shown to reprogram RSA by coordinating with BRs in a cell and tissue-specific manner (Bao et al., 2004; Chaiwanon and Wang, 2015; Vragović et al., 2015). For example, PR elongation depends on the spatiotemporal activity of BRs in the root tissues/zones, whereas high auxin represses cell elongation by inhibiting the BR signaling in the elongation zone (EZ) cells (Chaiwanon and Wang, 2015). The BRs are perceived by the LEUCINE-RICH REPEAT RECEPTOR-LIKE KINASES, BRASSINOSTEROID INSENSITIVE1 (BRI1), and its homologs, BRI1-LIKE1 (BRL1) and 3 (BRL3) in complex with the BRI1-ASSOCIATED RECEPTOR KINASE1/SOMATIC EMBRYOGENESIS RECEPTOR KINASE3. The binding of BRs with the receptor complex initiates an intracellular phosphorylation relay cascade that phosphorylates the negative regulator of the BR signaling pathway, BRI1 KINASE INHIBITOR1 (BKI1). Phosphorylation of BKI1 leads to its dissociation from the BRI1 receptor complex, thus activating BR signaling transcriptional effectors, BRASSINAZOLE RESISTANT1 (BZR1) and BRASSINOSTEROID INSENSITIVE1-ETHYLMETHANESULFONATE-SUPPRESSOR1 (BES1)/BZR2 (Li and Chory, 1997; Yin et al., 2002; Caño-Delgado et al., 2004; Wang and Chory, 2006; Jaillais et al., 2011). BZR1 and (BES1)/BZR2 regulate several growth-related processes in the plants (He et al., 2002; Wang et al., 2002; Tian et al., 2018; Kim et al., 2019). Optimal BR signaling activity in a cell and tissue-specific manner is necessary for coherent growth. High BRs repress root elongation by promoting early cell differentiation and premature cell cycle exit (González-García et al., 2011; Hacham et al., 2011). Targeted BRI1 signaling from hair and nonhair cells of the epidermis promotes and inhibits cell elongation, respectively (Fridman et al., 2014). Recently, BRs biosynthesis and signaling components are shown to modulate root growth under mild N deficiency where PR and LR elongation depend on BR signaling and response (Jia et al., 2019, 2020). The antagonistic or synergistic activity of auxin and BR signaling cascades regulates various plant developmental features; however, their coordinated activity in mediating PR growth in response to low or differential availability of N forms is still enigmatic.

By using pharmacological approaches, transcriptomics, available auxin and BR signaling pathways mutants, and targeted BR signaling lines in Arabidopsis (Arabidopsis thaliana) (Col-0), we demonstrate that fine-tuning of auxin and BR signaling is essential for mediating N foraging response during differential availability of N forms. We show that low nitrogen (LN; 0.05-mM NH₄NO₃ and 0.05-mM KNO₃) promotes PR elongation by modulating BRs-dependent auxin distribution in the roots. LN-grown roots show a reduced response to exogenous auxin as compared to sufficient N (SN; 9.2-mM NH₄NO₃ and 10-mM KNO₃). By using auxin signaling and transporter mutants, we demonstrate that an increase in PR length under LN condition depends on active nuclear auxin signaling and control of its distribution by PINs. Additionally, our transcriptomic data show that LN regulates a subset of auxin and BRs-regulated genes, and BKI1 is among them, which is repressed by LN. We show that high auxin or blocking auxin transport increases BKI1 in N-dependent manner in root vasculature/stele tissues. By using the tissue-specific approach, we demonstrate that stele/vasculature BRI1 signaling showed enhanced PR growth under LN condition. Further, we uncover that distinct N forms modulate BR signaling which in turn regulates auxin levels and transport. The ${\text{NH}}_4^{+}$ as a sole N form inhibits auxin by repressing BR signaling, whereas ${\mathrm{NO}}_3^{-}$ as a sole N source promotes root elongation in BRs dependent manner. Taken together, our study provides insight into the mechanisms underlying the root responses to LN and fluctuating availability of N forms via intertwined signaling responses of BRs and auxin.

Results

Root elongation in response to LN depends on auxin signaling and transport

To understand the interaction of LN and auxin on root elongation, wild-type, tir1-1 mutant, tir1-1 complemented line (tir1:pTIR1-gTIR1-VENUS, abbreviated tir1:pTIR1:gTIR1) and a strong TIR1 mutant line (pTIR1:tir1E12K-GUS, abbreviated tir1E12K from here onwards) which has an amino acid substitution in the F-box domain of TIR1, were grown on SN and LN medium (Yu et al., 2015). LN promotes PR elongation as described (Jia et al., 2019). Interestingly, tir1-1 and tir1E12K seedlings failed to promote PR elongation in response to LN. However, tir1:pTIR1:gTIR1 showed PR elongation similar to wild-type under LN condition (Figure 1, A and B). This observation suggested that LN-mediated PR growth depends on the TIR1. To understand how does LN modulate auxin signaling and response, wild-type seedlings were grown on SN and LN medium along with the increasing concentration of auxins, 2,4-Dichlorophenoxyacetic acid (2,4-D) and Indole-3-acetic acid (IAA). Given the fact that exogenous auxin inhibits root growth in a concentration-dependent manner (Simon et al., 2013; Fendrych et al., 2018;), SN-grown roots along with 2,4-D or IAA showed PR growth inhibition as compared to LN-grown roots under similar conditions (Figure 1C;Supplemental Figure S1, A and B). Following the reduced response of PR to 2,4-D under LN, these roots showed reduced meristem size and cell numbers as compared to SN-grown roots under similar conditions (Supplemental Figure S1, C and D). However, LN-grown roots were largely insensitive to 2,4-D-mediated cell elongation inhibition (Figure 1D;Supplemental Figure S1E). To understand the mechanisms underlying LN-dependent reduced auxin response of PR, we analyzed the auxin sensor protein, DII-VENUS that reflects the nuclear auxin signaling and distribution (Brunoud et al., 2012). As described, the nuclear DII-VENUS signal was reduced in 2,4-D-treated roots independent of N availability (Figure 1, E and F). Notably, LN-grown roots showed reduced nuclear DII-VENUS signal at the root tip (meristem zone [MZ] to elongation [EZ]/differentiation [DZ] zone) as compared to SN-grown roots indicating high nuclear auxin signaling under this condition (Figure 1, E and F). Our physiological data suggest that reduced response to 2,4-D and IAA in LN-grown roots is due to an already high nuclear auxin signaling which in turn led to the lower response to exogenous auxins (2,4-D or IAA) (Figure 1D;Supplemental Figure S1, A and B). This observation prompted us to look at the auxin distribution and transport characterizing PIN proteins dynamics at the root tip. The levels and polarity of PIN proteins are critical that determine the directional auxin flow among the cells to establish an auxin gradient which is necessary for cell expansion (Grieneisen et al., 2007; Barbosa et al., 2018). LN increases PIN1 levels in the PR vasculature from MZ to EZ (Supplemental Figure S1, F–H). Similarly, PIN2 that directs auxin from MZ to EZ in the epidermal cells showed increased PIN2 levels under this condition (Figure 1, G–I). High auxin or long-term auxin treatment inhibits PIN proteins levels (Vieten et al., 2005). In agreement, short-term (1 day) and long-term (8 days) 2,4-D treatment inhibited PIN2 levels in SN-grown roots. LN-grown roots maintain PIN2-GFP levels suggesting that N availability regulates directional auxin flow among the tissues by maintaining PINs protein levels for synchronized growth (Figure 1, G–I). We do not observe a change in marker line, UBQ10prom::CITRINE-2xPH(FAPP1) (abbreviated P21Y from here onwards) under similar conditions suggesting a change in GFP fluorescence is specific to PIN2 levels (Figure 1J). To understand whether enhanced PIN protein levels are associated with PR elongation under LN condition by maintaining cellular auxin gradient, we used a polar auxin transport inhibitor, 2,3,5-triiodobenzoic acid (TIBA). High concentration of TIBA (1-µM) inhibited PR elongation in SN-grown roots. In contrast, LN-grown roots under similar conditions showed a reduced response to PR growth inhibition whereas lower concentrations of it (50–500-nM) does not affect PR growth (Figure 1K;Supplemental Figure S2A). In line, auxin influx mutant, aux1-7 failed to promote PR elongation in response to LN. These findings further provided independent evidence that auxin transport and its distribution are necessary for root elongation in response to LN. Taken together, we demonstrate that LN modifies polar auxin transport and nuclear auxin signaling to regulate PR growth.

Figure 1

LN regulates root growth by modulating auxin signaling and transport. A, Primary root (PR) length of 11-day-old wild-type (Col-0), tir1-1, tir1:pTIR1:gTIR1, and tir1E12K-GUS seedlings grown on sufficient nitrogen (SN) and low nitrogen (LN) medium for 8 days (n = 3 biological replicates, 13 or more roots each treatment). B, Representative images of 11-day-old wild-type, tir1-1, tir1:pTIR1:gTIR1, and tir1E12K-GUS seedlings as grown in (A) (scale bar: 1 cm). C, PR length of 11-day-old wild-type seedlings grown on SN and LN medium for 8 days to increasing concentration of 2,4-Dichlorophenoxyacetic acid (2,4-D) as indicated (n = 4 biological replicates, 18 or more seedlings each treatment, numbers on the top of each bar represent percent root growth inhibition with respect to untreated control in each condition). D, Cortical cell length of the wild-type roots grown on SN and LN medium along with 25-nM 2,4-D as in (C) (n = 3 biological replicates, six roots from each treatment). E and F, DII-VENUS signal in the 11-day-old seedling roots grown on SN and LN medium for 8 days along with 25-nM 2,4-D (n = 3 biological replicates, scale bar: 100 µm, six or more roots each treatment). G, PIN2-GFP signal in 11-day-old roots grown on SN and LN medium for 8 days along with 25-nM 2,4-D (n = 3 biological replicates, scale bar 100 µm). H, PIN2-GFP (upper) and P21Y (UBQ10prom::CITRINE-2xPH(FAPP1) (P21Y) (lower) signal in the epidermal cells of the roots as grown in (G) (n = 3 biological replicates, 6 or more roots each treatment, scale bar: 50 µm). I, Total PIN2-GFP signal intensity from the epidermal cells of transition/elongation zone (TEZ) as grown in (G) (n ≥ 4 cells from each root, 6 or more roots each treatment, n = 3 biological replicates). In the scattered plot, dots represent single data point and error bars represent mean ± se. J, Quantification of total P21Y signal from TEZ epidermal cells as grown in H (n ≥ 4 cells from each root, 6 or more roots, n = 3 biological replicates). Dots in scattered plot represent single data point and error bars represent mean ± se. K, PR length of 11-day-old wild-type seedlings grown on SN and LN medium for 8 days along with the increasing concentration of polar auxin transport inhibitor, TIBA. Percent PR growth inhibition was represented with respect to untreated roots grown on SN and LN condition (n = 3 biological replicates, 16 or more roots each treatment, error bars represent mean ± se, numbers shows % PR growth inhibition). L, PR length of wild-type and aux1-7 in response to SN and LN. Seedlings were grown as in (A) (n = 4 biological replicates, 24 or more roots each treatment). Different letters in each graph indicate significant differences among the treatments according to one-way ANOVA and post-hoc Tukey’s test at P < 0.05. In the box and whisker plots, the box represents interquartile range (IQR), whiskers represent minimum and maximum value and horizontal line represents median value.

N-dependent auxin distribution alters BKI1 to regulate root growth

To unravel how does LN promote root growth by modulating auxin signaling activity at the root tip, we performed RNA-sequencing (RNA-seq) of the roots grown on SN and LN conditions for 4 days. Data show that 2,065 genes were differentially regulated between SN and LN conditions (Supplemental Figure S3A). The log2 fold change of differentially expressed genes (DEGs) was shown as SN versus LN ratio (Supplemental Figure S3A; Supplemental Table S1). Hierarchical clustering and correlation analysis of DEGs highlighted the pathways involved in hormone signaling and response, cell wall organization, and N metabolism (Supplemental Figure S3B). In accordance that LN regulates auxin, RNA-seq also highlighted the genes involved in auxin signaling and response (1.9% of total DEGs) such as IAAs, SMALL AUXIN-UPREGULATED RNAs and auxin-responsive transcription factors (ARF9, ARF11, and PLETHORA 5) (Supplemental Figure S3C). Auxin and BRs together regulate many growth and development-related processes in the plants; therefore, we hypothesized whether auxin and BR signaling pathways integrate to promote root growth under LN condition. Toward this end, we found that 0.6% of total DEGs which are involved in BRs biosynthesis, signaling, and responses were differentially regulated between SN and LN condition. Among them, BKI1 was repressed by LN (Figure 2, A and B; Supplemental Figure S3D). BKI1 is known to inhibit BR signaling by repressing the BRI1 receptor; therefore, we investigated the role of BKI1 in response to LN and its possible regulation by auxin.

Figure 2

Polar auxin transport modulates BKI1 expression levels in an N-dependent manner. A, Normalized counts of BKI1. B, Quantitative reverse transcription PCR (RT-qPCR) validation of BKI1 transcripts in the wild-type roots treated with SN and LN (n = 3 biological replicates, ***P < 0.001, indicate the significant difference with the two-tailed Student’s t test, error bars represent mean ± se). C, Colour coded confocal images of 11-day-old 35S:BKI1-YFP roots grown on SN and LN medium for 8 days (scale bar: 100 µm, n = 6 roots, 4 biological replicates). (D) Western blot analysis of BKI1-YFP protein in shoots and roots of transgenic seedlings as grown in (C). E, RT-qPCR analysis of BR-responsive genes in wild-type and 35S:BKI1-YFP roots as grown in (C) (n = 2 biological replicates, *P < 0.05, **P < 0.01, ***P < 0.001, indicate significant difference with two-tailed Student’s t test, error bars represent mean ± se). F, Analysis of BKI1-YFP protein levels in the roots grown on SN and LN medium for 8 days along with 5-µM TIBA for 6 h (n = 3 biological replicates, 7 or more roots each treatment, scale bar: 100 µm). G and H, Quantification of BKI1-YFP signal in the root stele/vasculature tissues up to the TEZ (G) and at the PM of epidermal cells from the TEZ (H) (n = 3 biological replicates, six or more roots each treatment). I, PR length of 11-day-old wild-type, 35S:BKI1-YFP and bki1-1 seedlings grown on SN and LN medium along with the increasing concentrations of 2,4-D as indicated (n = 3 biological replicates, 13 or more roots each treatment). In the box and whisker plots, the box represents IQR, whiskers represent minimum and maximum value and horizontal line represents median value. J, Effect of eBL and 2,4-D treatment on PIN2-GFP (upper) and P21Y (UBQ10prom::CITRINE-2xPH(FAPP1) (lower). Epidermal cells from TEZ in the 11-day-old roots grown on SN and LN medium for 8 days along with 25-nM 2,4-D and 10-nM eBL for 24 h. Long-term (8 days) 2,4-D treatment represses PIN2-GFP PM signal in an N-dependent manner (n = 5 roots, three biological replicates, scale bar: 50 µm). K, Total PIN2-GFP signal intensity from TEZ root cells as grown in (J). PIN2-GFP signal from three to four cells from each root was quantified (n = 3 biological replicates, five roots each treatment). In the scattered plot, dots represent single data point, horizontal bars represent mean and error bars represent mean ± se. Different letters in each graph indicate significant difference according to one-way ANOVA and post-hoc Tukey’s test at P < 0.05.

Given the fact that BKI1 protein levels and its cellular localization are critical for BR signaling output (Jaillais et al., 2011; Wang et al., 2017; Zhang et al., 2017; Singh et al., 2018), we also tested the possibility whether BKI1 protein levels are modulated by LN. Toward this end, 35S: BKI1-YFP seedlings were grown on SN and LN media and BKI1-YFP protein levels were analyzed. Indeed, LN-grown roots showed reduced BKI1-YFP protein at the root tip as compared to SN-grown roots suggesting that both transcripts and protein levels of BKI1 are regulated by N availability (Figure 2, C and D). To test whether reduced levels of BKI1 under LN are associated with enhanced BR response, we tested the known BRs responsive genes in the root tissues of wild-type and 35S: BKI1-YFP seedlings grown on SN and LN conditions (Chaiwanon and Wang, 2015). Indeed, LN enhanced the expression of BRs responsive genes in the 35S: BKI1-YFP line similar to wild-type (Figure 2E). Exogeneous epibrassinolide (eBL)-treated roots grown on SN condition were used as a positive control. Further, LN-grown roots showed increased levels of BZR1-CFP. In agreement that high BR signaling enhances nuclear (N)/cytoplasmic (C) ratio of BZR1 (Chaiwanon and Wang, 2015), LN-grown roots showed increased N/C ratio similar to eBL-treated roots (Supplemental Figure S4, A–D). Additionally, LN-grown roots showed reduced transcript levels of the BRs biosynthesis genes (CPD, DWF4, and BR6OX2) similar to eBL-treated roots, further supporting our claim that LN increases BR signaling and response that represses BRs biosynthesis genes in a feedback manner (Supplemental Figure S4, E–H). Low BRs enhance BKI1 accumulation at the PM thus, enhancing its association with the BRI1 receptor to inhibit the receptor complex activity. In contrast, high BRs promote its dissociation from the PM, thereby releasing its inhibitory effect on BR signaling output (Wang and Chory, 2006; Jaillais et al., 2011). To understand the BKI1 regulation by auxin and LN, BKI1-YFP seedlings were grown on SN and LN along with 2,4-D and IAA for 8 days. Long-term (8 days) 2,4-D or IAA treatment enhances BKI1-YFP in SN-grown roots as compared to LN-grown (Supplemental Figure S5, A and B). SN-grown roots along with 2,4-D for 8 days also affected cell morphology; therefore, we used a high concentration of 2,4-D (5-µM) for 8 h (Supplemental Figure S5C). BKI1-YFP fluorescence was enhanced in epidermal cells and root vasculature (stele) tissues within 8 h of 2,4-D treatment. However, an increase in vasculature (stele) signal was more prominent (Supplemental Figure S5, C–F). Our data show that LN promotes root growth by regulating the directional auxin flow at the root tip, therefore, we tested the effect of TIBA on BKI1 levels. Interestingly, both short-term (6 h) and long-term (8 days) TIBA-treated roots showed enhanced BKI1-YFP signal intensity in N dependent manner (Figure 2F;Supplemental Figure S6A). In our conditions, TIBA treatment does not affect the PM marker, LTI6b-GFP expression levels under similar conditions (Supplemental Figure S6, B–E). Moreover, TIBA-treated roots grown on SN medium showed enhanced BKI1 levels in the root tip vasculature tissues (from MZ to EZ) as compared to LN-grown roots (Figure 2, F–H). TIBA treatment also elevated PM accumulation of BKI1-YFP. However, an increase in the BKI1-YFP signal at the PM was similar in both SN and LN conditions (Figure 2H). To determine the effect of auxin-dependent increased BKI1 levels on root elongation, wild-type, 35S: BKI1-YFP and bki1-1 seedlings were grown on SN and LN conditions along with 2,4-D. Wild-type seedlings show increased PR length in response to LN with and without 2,4-D as compared to SN-grown roots (Figure 2 I). In contrast, 35S: BKI1-YFP roots grown under LN condition with 2,4-D showed increased root elongation inhibition (39% inhibition in 50-nM 2,4-D as compared to wild-type, 16%) and bki1-1 (Figure 2I). In addition to 2,4-D, 35S: BKI1-YFP roots treated with IAA also showed an increase in PR growth inhibition (Supplemental Figure S6F). LN-grown bki1-1 mutant does not show a significant increase in PR length plausibly due to enhanced BR signaling that inhibits root growth by promoting early cell differentiation and premature cell cycle exit. In agreement, bki1-1 roots showed longer cells and reduced meristem size independent of N availability (Supplemental Figure S7, A–C). A similar response on root length, cell elongation, and MZ length were observed in 35S:BRI1-GFP (high BR signaling) seedlings as compared to pBRI1:BRI1 GFP and wild-type, further supporting the hypothesis that enhanced BR signaling negatively affected PR growth under LN condition (Supplemental Figure S7, A–C). Additionally, eBL treatment (24 h) enhanced PIN2-GFP levels in 2,4-D-treated roots grown on SN condition (Figure 2, J and K), whereas 2,4-D treatment does not affect PIN2 levels in LN-grown roots due to already enhanced BR signaling under this condition (Figure 2, J and K). SN-grown roots treated with eBL showed increased levels of PIN2-GFP as reported (Retzer et al., 2019). Based on our conclusions that LN increases BR signaling, roots grown under this condition showed increased levels of PIN2-GFP signal (Figure 2, J and K). P21Y seedlings grown under similar conditions were used as a control (Supplemental Figure S7D). These seedlings were largely insensitive to treatments, except under eBL conditions, where P21Y levels were slightly reduced (Supplemental Figure S7D). Altogether, we show that LN modifies polar auxin transport and activation of nuclear auxin signaling which suggests that LN promotes the accumulation of auxin at the root tip. Such an increase in auxin modifies BKI1 dynamics in N dependent manner. High auxin or inhibiting auxin transport represses BR signaling thus, root growth by increasing BKI1 levels.

BR signaling from the root vasculature (stele) tissues shows enhanced root elongation under LN condition

We show that high auxin or inhibiting polar auxin transport modulate BR signaling negative regulator, predominantly in the inner tissues (root vasculature/stele tissues; Figure 2, F and G; Supplemental Figure S5, A–F). Therefore, we speculated that BR signaling in inner tissues might be necessary for LN response. To test the hypothesis, we used the lines with the targeted expression of BR receptor, BRI1 in the atrichoblast cells of the epidermis (pGL2:BRI1), endodermis (pSCR:BRI1), and root vasculature (stele) (pSHR:BRI1) in a bri1-116 mutant background. In our conditions, by using a strong bri1-116 and bri1 null allele, we do not observe a significant change in PR length; however, it shows a slight nonsignificant increase in PR length under LN (Supplemental Figure S8, A and B). pSCR:BRI1 line was largely insensitive to LN whereas pGL2:BRI1 showed a slight increase in PR length. Interestingly, as speculated, BR signaling from inner cells in pSHR:BRI1 line showed enhanced PR elongation in response to LN as compared to pGL2:BRI1 and pSCR:BRI1 lines (Figure 3A). Further, pSHR:BRI1 and pGL2:BRI1 lines also show a reduced response to 2,4-D as compared to the pSCR:BRI1 line (Figure 3A). PR elongation under LN condition in pSHR:BRI1 and wild-type roots correlated with the increased cortical cell length (Figure 3B;Supplemental Figure S8, C and D). Taken together, we show that inner cell BR signaling modulates root growth under LN condition.

Figure 3

BR signaling from root vasculature (stele) tissues shows enhanced PR elongation under LN condition. A, PR length of 11-day-old wild-type, and cell layer-specific targeted BRI1 transgenic lines, bri1-116;pGL2:BRI1 (atrichoblast cells of epidermis), bri1-116;pSCR:BRI1 (endodermis), and bri1-116; pSHR:BRI1 (stele/vasculature) grown on SN and LN medium along with 25-nM 2,4-D for 8 days (12 or more roots each treatment, n = 4 biological replicates). B, Cortical cell length of wild-type, bri1-116; pGL2:BRI1, bri1-116; pSCR:BRI1 and bri1-116; pSHR:BRI1 seedlings as grown in (A) (8 roots for each line, n = 3 biological replicates). Different letters indicate significant difference according to one-way ANOVA and post-hoc Tukey’s test at P < 0.05. In the box and whisker plots, the box represents IQR, whiskers represent minimum and maximum value, dots represent data points and horizontal line represents median value.

N forms tune root growth by regulating BR signaling activity

We demonstrate that LN enhances auxin distribution to promote PR growth in a BR-dependent manner. Therefore, we explored the potential role of BRs in the conditions where ${\mathrm{NO}}_3^{-}$/${\text{NH}}_4^{+}$ sources are differentially distributed, either in combination (SN) or in the form of individual N sources (⁠${\mathrm{NO}}_3^{-}$/${\text{NH}}_4^{+}$⁠). First, to know whether N forms mediated root growth depends on BR signaling and response, we treated BKI1-YFP seedlings with SN, ${\text{NH}}_4^{+}$⁠, and ${\mathrm{NO}}_3^{-}$ as sole N forms. 35S:BKI1-YFP roots show an increase in PR length in response to ${\mathrm{NO}}_3^{-}$ similar to wild-type as compared to ${\text{NH}}_4^{+}$ and SN-grown roots (Figure 4A). Further, ${\text{NH}}_4^{+}$ as a sole N form enhanced BKI1-YFP protein at the root tip in all the tissues, whereas P21Y and LTI6b-GFP markers remain unchanged (Figure 4, B–D; Supplemental Figure S9, A–E). However, bki1-1 mutant showed a reduced response to NH₄⁺-mediated root growth inhibition as compared to wild-type (Supplemental Figure S9F). In our conditions, we do not observe any significant change in BKI1 levels under SN and ${\mathrm{NO}}_3^{-}$ conditions, indicating that ${\mathrm{NO}}_3^{-}$-mediated root elongation is plausibly independent of BKI1 (Figure 4, B and C; Supplemental Figure S9B). These findings suggested that root growth inhibition in response to ${\text{NH}}_4^{+}$ as a sole N form could be due to reduced BR signaling under this condition. Further, root growth sensitivity assay to increasing concentration of BRs biosynthesis inhibitor, brassinazole (BRZ; Asami et al., 2000), showed reduced root growth inhibition to BRZ in ${\mathrm{NO}}_3^{-}$-grown roots. In contrast, NH₄⁺-grown roots showed full inhibition at 0.25-µM, BRZ concentration, and no further inhibition was observed at higher BRZ concentrations (Figure 4E). In line with the BRZ assay, the nuclear BZR1-CFP signal (N/C ratio) was reduced in NH₄⁺-grown roots and enhanced in the roots grown on ${\mathrm{NO}}_3^{-}$ medium as compared to SN condition. Exogenous eBL treatment further increases BZR1-CFP levels (N/C ratio) in NH₄⁺-grown roots suggesting that N forms regulate BR signaling and response (Supplemental Figure S9, G–I). In agreement, bri1-116 null mutant blocks root response to ${\mathrm{NO}}_3^{-}$ and failed to promote PR elongation. In contrast, NH₄⁺-grown bri1-116 mutant show root growth inhibition similar to wild–type, suggesting that ${\mathrm{NO}}_3^{-}$-mediated root growth depends on BRI1 signaling activity (Figure 4F).

Figure 4

N forms modulate BR signaling to regulate root growth. A, PR length of 11-day-old wild-type and 35S:BKI1-YFP harboring transgenic plants grown on SN, ${\text{NH}}_4^{+}$⁠, and ${\mathrm{NO}}_3^{-}$ containing medium for 8 days (n = 3 biological replicates, 15 or more roots each treatment). B, Localization of BKI1-YFP protein in response to ${\text{NH}}_4^{+}$ and ${\mathrm{NO}}_3^{-}$ as a sole N form. Seven-day-old seedlings were transferred to SN, ${\text{NH}}_4^{+}$⁠, and ${\mathrm{NO}}_3^{-}$ containing medium for 4 days (n = 3 biological replicates, six roots each treatment, scale bar: 100 µm). C, Quantification of BKI1-YFP signal in the root vasculature (stele) tissues corresponding to two to three cortical cells above the MZ (n = 3 biological replicates, 13 or more roots each treatment). In the scattered plot, dots represent single data point, horizontal bars represent mean and error bars represent mean ± se. D, Expression of UBQ10prom::CITRINE-2xPH(FAPP1) (P21Y) marker as grown in B (n = 3 biological replicates, five roots each treatment, scale bar: 100 µm). E, Root sensitivity to increasing concentration of BR biosynthesis inhibitor, brassinazole (BRZ). Three days post germinated (dpg) wild-type seedlings were transferred to SN, ${\text{NH}}_4^{+}$⁠, and ${\mathrm{NO}}_3^{-}$ medium along with BRZ for 8 days, and PR length was measured. Numbers indicate % root growth inhibition compared with respective untreated control roots (n = 3 biological replicates, 13 or more roots each treatment, error bars represent mean ± se). F, Root length of 11-day-old wild-type and bri1-116 mutant grown on SN, ${\text{NH}}_4^{+}$⁠, and ${\mathrm{NO}}_3^{-}$ containing medium for 8 days (n = 3 biological replicates, 13 or more roots each treatment). Note: bri1-116 mutant blocks root elongation in response to ${\mathrm{NO}}_3^{-}$⁠. G, Recovery of PR length by epibrassinolide (eBL) in the 11-day-old roots grown on SN, ${\text{NH}}_4^{+}$⁠, and ${\mathrm{NO}}_3^{-}$ containing medium for 6 days. Seedlings treated with 10-nM eBL for 24 h on 10th day (Note: eBL-treated roots show PR growth under ${\text{NH}}_4^{+}$ condition, n ≥ 12 roots, three biological replicates). H, Representative images of 11-day-old wild-type roots showing MZ (arrowheads) grown on SN, ${\text{NH}}_4^{+}$⁠, and ${\mathrm{NO}}_3^{-}$ medium for 6 days along with 10-nM eBL for 24 h (n = 3 biological replicates, 9 or more roots, scale bar: 100 µm). I and J, Meristem cell numbers (I) and cortical cell length (J) of the wild-type roots as grown in (G). Different letters indicate significant difference according to one-way ANOVA and post-hoc Tukey’s test at P < 0.05. In the box and whisker plots, the box represents IQR, whiskers represent minimum and maximum value, dots represent data points and horizontal line represents median value.

Consistent with our findings, short-term (24 h) eBL treatment promoted PR elongation in the seedlings grown on ${\text{NH}}_4^{+}$ as a sole N source (Figure 4G). However, there was no change in the root length under long-term (8 days) eBL treatment due to premature cell cycle exit and early cell differentiation which in turn inhibit overall root elongation as described (González-García et al., 2011; Hacham et al., 2011) (Supplemental Figure S9J). In agreement with the PR elongation in short-term eBL-treated roots (24 h), both meristem cell numbers and cell length were increased in NH₄⁺-grown roots (Figure 4, H–J). Taken together, we show that in a heterogeneous N environment, N forms regulate root growth by modulating BR signaling.

N form-dependent BR signaling activity tunes root growth by modulating nuclear auxin signaling and distribution

Recently, the function of auxin has been shown to modulate root growth by regulating root meristematic activity and cell elongation in N forms specific manner (Liu et al., 2013; Ötvös et al., 2021). Therefore, we asked whether BRs could modulate auxin transporter (PIN2-GFP) and nuclear auxin signaling reporter (DII-VENUS) in N forms-specific manner. Toward this end, PIN2-GFP and DII-VENUS seedlings were grown on SN, ${\text{NH}}_4^{+}$⁠, and ${\mathrm{NO}}_3^{-}$ medium along with 2,4-D, IAA, and eBL. As described (Liu et al., 2013; Li et al., 2018), PIN2-GFP levels were repressed in ${\text{NH}}_4^{+}$-grown roots as compared to SN- and ${\mathrm{NO}}_3^{-}$-grown roots (Figure 5, A and B). However, the P21Y and LTI6b-GFP markers were unaffected by N forms suggesting that change in PIN2-GFP is specific to N forms (Figure 5C;Supplemental Figure S9, C–E). Interestingly, eBL, 2,4-D, and IAA treatment enhanced PIN2-GFP levels in ${\text{NH}}_4^{+}$-grown roots within 24 h (Figure 5, A and B). In agreement with the previous report (Vieten et al., 2005), SN-grown roots with 2,4-D/IAA showed reduced PIN2-GFP levels. A similar effect of 2,4-D/IAA on PIN2-GFP levels was observed in ${\mathrm{NO}}_3^{-}$-grown roots. Under SN condition, eBL enhances PIN2 levels, whereas it does not affect its levels in ${\mathrm{NO}}_3^{-}$-grown roots (Figure 5, A and B). In line with PIN2 levels, ${\text{NH}}_4^{+}$ as a sole N source increases DII-VENUS signal at the root tips, whereas ${\mathrm{NO}}_3^{-}$ as a sole N represses it. Indeed, eBL represses DII-VENUS fluorescence at the root tip in SN- and ${\text{NH}}_4^{+}$-grown roots similar to 2,4-D and IAA (Figure 5, D and E). However, we do not observe a reduction in DII-VENUS fluorescence in ${\mathrm{NO}}_3^{-}$-grown roots similar to PIN2 levels in eBL treated roots suggesting the involvement of other pathways (Camut et al., 2021; Figure 5, D and E). Similarly, ${\text{NH}}_4^{+}$ repressed DR5-GUS reporter expression at the root tip whereas eBL, 2,4-D, and IAA treatment further enhanced the reporter expression (Figure 5F). Long term (6 days) ${\text{NH}}_4^{+}$ treatment also repressed the genes involved in auxin distribution (PIN2, PIN3, PIN4, and AUX1) whereas eBL increases their transcription within 24 h (Supplemental Figure S9K). Based on these findings, our data suggested that coordinated activity of auxin and BR signaling modulates root growth in N forms-specific manner.

Figure 5

BR signaling activity modulates PIN2 levels and nuclear auxin signaling in an N form-dependent manner. A, Effect of N forms (⁠${\text{NH}}_4^{+}$/${\mathrm{NO}}_3^{-}$⁠) and BR on PIN2-GFP levels. The PIN2-GFP seedlings were grown on SN medium for 5 days and transferred to either combined N sources (SN) or ${\text{NH}}_4^{+}$ and ${\mathrm{NO}}_3^{-}$ containing medium as a sole N form for 6 days along with 10-nM eBL, 50-nM 2,4-D, and 60-nM IAA for 24 h (n = 3 biological replicates, five or more roots each treatment, scale bar: 100 µm). Lower in (A) shows PIN2-GFP signal in epidermal cells of transition zone (scale bar: 50 µm). B and C, Total PIN2-GFP (B) and P21Y (C) signal intensity in the roots as grown in (A). GFP signal was quantified from the transition zone cells from Z stack images. About four to five cells with clearly visible boundaries by using PI channel were quantified from each root (n = 3 biological replicates, five or more roots each treatment). In the scattered plot, dots represent single data point, horizontal bars represent mean and error bars represent mean ± se. D, Analysis of DII-VENUS in the roots treated with 10-nM eBL, 50-nM 2,4-D and 60-nM IAA for 24 h as grown in (A) (scale bar: 100 µm). E, Quantification of nuclear DII-VENUS signal intensity in the first 8–10 cells of epidermis layer from CEI (n = 3 biological replicates, five or more roots each treatment). In the scattered plot, dots represent single data point, horizontal bars represent mean and error bars represent mean ± se. F, DR5-GUS expression in the roots as grown in (A) along with 10-nM eBL, 50-nM 2,4-D, and 60-nM IAA for 24 h. Note: BL, 2,4-D, and IAA-treated roots show recovery of GUS reporter gene in the roots grown in ${\text{NH}}_4^{+}$ containing medium (n = 3 biological replicates, five or more roots each treatment, scale bar: 100 µm). Different letters in the graphs show significant difference according to one-way ANOVA and post-hoc Tukey’s test at P < 0.05.

To determine the effect of exogenous auxins and BR signaling-induced auxins on root growth and cell elongation, wild-type roots were treated with eBL, 2,4-D, and IAA for 24 h. Based on PIN2-GFP levels under the same condition, roots grown on ${\text{NH}}_4^{+}$ medium along with 2,4-D and IAA also showed PR elongation like eBL treated roots suggesting that auxin is necessary for root growth under this condition (Figure 6A). The synergistic and antagonistic activity of auxin and BRs in different tissues has been shown to modulate cell division and elongation. For example, high auxin in EZ represses BR signaling transcription factor, BZR1, thus, inhibiting BRs dependent cell elongation processes (Chaiwanon and Wang, 2015). However, regulation of meristematic cell divisions and its size depends on high auxin levels in the stem cell niche (surrounding cells; Dello Ioio et al., 2008). To understand the effect of exogenous auxins and BR induced auxin signaling on meristematic cell numbers and cell elongation, we quantified the meristem cell numbers and mature cell length in 2,4-D/IAA and eBL treated roots grown on SN, ${\text{NH}}_4^{+}$⁠, and ${\mathrm{NO}}_3^{-}$ medium. SN- and ${\mathrm{NO}}_3^{-}$-grown roots showed cell elongation inhibition in response to 2,4-D and IAA (Figure 6B). The ${\text{NH}}_4^{+}$ as a sole N source shows mild (50% roots) to severe root phenotype and inhibited meristem size, cell numbers, and cortical cell length as compared to SN- and ${\mathrm{NO}}_3^{-}$-grown roots (Supplemental Figure S10A). Regardless, the severity of the phenotype, eBL increases the mature cell length in ${\text{NH}}_4^{+}$-grown roots to a level of SN-grown roots. eBL-treated roots also showed a partially increase in the meristem size and cell numbers (as compared to mild phenotype in ${\text{NH}}_4^{+}$-grown roots; Figure 6B;Supplemental Figure S10, A and B). Interestingly, 2,4-D or IAA treated roots showed recovery only in meristem cell numbers and failed to promote cell elongation in NH₄⁺-grown roots, highlighting the role of BR signaling induced auxin distribution that determines the cell elongation (Figure 6, C–E; Supplemental Figure S10B). In accordance that 2,4-D/IAA promotes meristem cell numbers in NH₄⁺-grown roots, DR5-GUS reporter was also enhanced in NH₄⁺-grown roots along with 2,4-D or IAA (Supplemental Figure S10C). Altogether, we demonstrated that ${\mathrm{NO}}_3^{-}$/${\text{NH}}_4^{+}$ N forms regulate root growth by modulating auxin and BR signaling activity in a heterogeneous N environment as an adaptive strategy.

Figure 6

N form-dependent BR signaling-induced auxin activity tunes root growth. A, PR length recovery of wild-type seedlings within 24 h of hormone treatments (10-nM eBL, 50-nM 2,4-D, and 60-nM IAA). Five days SN-grown seedlings were transferred to SN, ${\text{NH}}_4^{+}$⁠, and ${\mathrm{NO}}_3^{-}$ containing medium for 6 days (n = 3 biological replicates, 13 or more roots each treatment, different letters in the graphs indicate significant difference according to one-way ANOVA and Tukey’s test at P < 0.05). B, Cortical cell length of the roots as grown in (A). C, Representative images of root meristems of wild-type seedlings as grown in (A). Arrow head indicates transition of meristem cells to elongation zone (n = 4 biological replicates, six roots each treatment, scale bar: 100 µm). D, Meristem cell number in wild-type roots as grown in (A). Note- 2,4-D, IAA- and BL-treated roots grown on ${\text{NH}}_4^{+}$ medium show an increase in meristem cell numbers. E, Cell number versus cell length from CEI (n = 3 biological replicates, arrows indicate transition from MZ to TEZ). F, Proposed model showing the fine tuning of BR and auxin signaling pathways by N forms. The arrows represent positive regulation and the bars represent inhibition. Different letters in the graphs show significant difference according to one-way ANOVA and post-hoc Tukey’s test at P < 0.05. In the box and whisker plots, the box represents IQR, whiskers represent minimum and maximum value, dots represent data points and horizontal line represents median value.

Discussion

Coordinated activity of auxin and BR in the cells and tissues regulates several developmental features of the plants that enable them to grow in balance (Vert et al., 2008; González-García et al., 2011; Oh et al., 2014;Meier et al., 2020). In this study, we show that LN increases radial auxin distribution at the root tip which in turn represses BKI1, thereby enhancing BR signaling to promote PR elongation. N in the form of ${\text{NH}}_4^{+}$ as a sole N source represses BR signaling whereas BR signaling activity is necessary for ${\mathrm{NO}}_3^{-}$-dependent root elongation (Figure 6F). Due to the heterogeneous nature of the soils, the major inorganic N forms (⁠${\mathrm{NO}}_3^{-}$/${\text{NH}}_4^{+}$⁠) are differentially distributed and often unavailable for uptake. Plants tend to exploit N from the soils to achieve growth by reprogramming their RSA to meet the N demand. The N in the form of ${\mathrm{NO}}_3^{-}$ as a sole N form promotes root elongation whereas ${\text{NH}}_4^{+}$ represses it by attenuating the auxin responses (Qin et al., 2011; Ötvös et al., 2021). We show that the root developmental reprogramming during LN or under the specific N forms depends on BR signaling activity and response. The ${\mathrm{NO}}_3^{-}$-promoted root growth depends on BR signaling and ${\text{NH}}_4^{+}$ represses it to attenuate root growth as an adaptive response.

The spatiotemporal activity of BR and auxin signaling pathways have been shown to modify the RSA under environmental stresses (Ma et al., 2014; Singh et al., 2018). For example, high auxin inhibits cell elongation potentially by modulating the cell wall acidification that depends on the auxin gradient along the root axis (Barbez et al., 2017). We demonstrate that LN-grown roots showed a reduced response to exogenous auxins (2,4-D/IAA) and TIBA. Moreover, longer PR length under this condition is associated with the increased mature cell length. The auxin binding to the TIR1 receptor is necessary for mediating the downstream signaling events to drive auxin-mediated growth processes (Salehin et al., 2015; Yu et al., 2015). Indeed, the PR growth in response to LN depends on the TIR1. The tir1 and E12K mutants showed insensitivity to PR growth under LN. These findings suggested that auxin binding to its receptor and its subsequent signaling are necessary for root growth under LN. In agreement that fine-tuning of auxin biosynthesis and its subsequent cellular distribution is critical for the transition of cell division to elongation (Ötvös et al., 2021), we observe increased levels of PIN1 and PIN2 in response to LN. In line, aux1-7 mutant was unable to promote PR growth in response to LN, further supporting that root elongation under LN depends on auxin distribution and transport. A balance of auxin and BR responses in root zones and tissues is necessary to maintain a dynamic balance of cell division, elongation, and differentiation for optimal growth (Chaiwanon and Wang, 2015; Singh and Savaldi-Goldstein, 2015; Vragović et al., 2015). Recently, it was shown that the PR and LR elongation in response to mild N deficiency depends on BRs biosynthesis and signaling (Jia et al., 2019, 2020). In agreement, our data also showed that root elongation in response to LN depends on a balance between auxin and BR signaling activities. LN repressed the transcripts of BRs biosynthesis genes such as CYP90D1 within 4 days of treatment whereas 8 days LN-grown roots further showed the reduced levels of CPD, DWF4, and BR6OX2 suggesting high BR signaling under this condition. In agreement, BZR1 levels were increased in LN-grown roots, leading to feedback repression of BRs biosynthesis genes due to high BR signaling (Jung et al., 2010). Therefore, the reduced PR growth inhibition in response to 2,4-D/IAA, and TIBA under LN could be due to the BR signaling-induced auxin distribution which is necessary for cell elongation and root growth.

Consistent with the role of BRs in regulating the root growth under LN, BKI1 levels were repressed in TIBA-treated roots, predominantly in the inner tissues under this condition. Thus, leading to the activation of BR signaling and root elongation. Local BRs biosynthesis and its spatial distribution in the root zones have been shown to regulate cell elongation and optimal root growth (Vukašinović et al., 2021). The BR signaling from the outer tissues (epidermis) regulates meristematic cell divisions by modulating the auxin levels in the inner cell files (Vragović et al., 2015). Recently, the dual function of BR signaling in mediating the auxin response from specific cell files has been demonstrated (Ackerman-Lavert et al., 2021). Using the tissue-specific approach, it was shown that pGL2:BRI1, pSCR:BRI1, and pSHR:BRI1 lines in the bri1-116 mutant and pGL2:BRI1 and pSHR:BRI1 lines in the bri1brl1brl3 triple mutant were able to partially rescue the PR growth and cell length of the respective mutants (Hacham et al., 2011; Kang et al., 2017). This partial rescue could result from a supra-optimal response to BR signaling that inhibits root length (as in pGL2:BRI1; Fridman et al., 2014) or from a reduced responsiveness, as apparent in lower magnitude of gene activation in the inner tissues in response to the hormone (Vragović et al., 2015). Here, we used these lines and found that BR signaling in the root vasculature (stele) largely triggers PR growth and cell elongation under LN condition compared to pGL2:BRI1 and pSCR:BRI1 lines, highlighting the tissue specificity of BR signaling under adverse environmental conditions. The environment-dependent effect of BRI1 in the stele could be due to the mechanisms regulating its subcellular localization or regulating downstream signaling components to enhance the sensitivity of the pathway by a hormone (Geldner et al., 2007; Irani et al., 2012). Therefore, targeted BR signaling in the inner cells could be an important strategy for improving growth under limited N availability. Nonetheless, the mechanisms behind BRs modulation of N-responsive genes to adjust plant N metabolism are still vague. The nitrate transporter, NRT1.1 regulates plant N balance and LR growth by modulating auxin transport during N deficiency (Krouk et al., 2010; Jian et al., 2018). Therefore, it would be interesting to investigate whether BRs are also involved in regulating NRT1.1 activity to improve plant growth and N metabolism. Based on our findings, we propose that LN modulates auxin and BR signaling activity to promote root elongation in a LN environment.

The major inorganic N sources (⁠${\mathrm{NO}}_3^{-}$/${\text{NH}}_4^{+}$⁠) vary in the soils from low to high levels due to high diffusion rates. The contrasting effects of ${\mathrm{NO}}_3^{-}$ and ${\text{NH}}_4^{+}$ on root growth kinetics have been shown to alter auxin levels and distribution in the root zones (Ötvös et al., 2021). In agreement, our data also support that N forms as a sole N source regulate auxin levels and its transport as assessed by using nuclear auxin input marker (DII-VENUS), auxin reporter (DR5-GUS), and auxin efflux carrier protein (PIN2-GFP) lines. Here, we provided the evidence that N forms mediated auxin response and distribution depends on BR signaling activity. Exogenous application of BR restores PIN2-GFP levels as well as DII-VENUS/DR5-GUS levels within 24 h in NH₄⁺-grown roots. Given the fact that fine-tuning of these hormones is critical for meristem maintenance and cell elongation, we showed that BR (eBL) promotes meristem cell number and cell elongation in NH₄⁺-grown roots, whereas auxin (2,4-D/IAA) was able to rescue only meristem activity. Therefore, coordination of these two hormones in distinct tissues is critical for optimal root growth. High leaching and less reactivity of ${\mathrm{NO}}_3^{-}$ ions over the ${\text{NH}}_4^{+}$ allow the plant to promote deep RSA for increased N exploitation (Nishimura and Itani, 2013). In contrast, ${\text{NH}}_4^{+}$ as a sole N represses it as a survival strategy to avoid ${\text{NH}}_4^{+}$ toxicity. Here, we demonstrate that the low and individual N forms regulate BR signaling activity to achieve such adaptations. Despite the major roles of auxin/BRs in regulating the RSA in a heterogeneous N environment, it would be interesting to understand how does inner cell BR signaling modulates the genes involved in plant N response and provides a promising hypothesis for future work. Taken together, this study demonstrates that root growth inhibition or elongation in response to N sources depends on the activity of BRs and auxin responses that allows the plant to adapt in heterogeneous N environment.

Conclusions

Nutrient-derived signals modify RSA, which is tightly linked with the hormonal responses. Our study provides mechanistic insights into the role of auxin and BRs in regulating PR growth under fluctuating N conditions. Nonetheless, the targeted manipulation of these two hormonal signaling pathways in a tissue-specific manner could be pivotal for developing plants with improved nutrient use efficiency. In a heterogeneous environment, where N is limited due to increased ${\mathrm{NO}}_3^{-}$ leaching, physiological adaptations such as PR and LR elongation is necessary for N exploitation from the rhizosphere and BRs appear essential in modulating these growth processes. In contrast, plants inhibit PR growth by inhibiting BRs during ${\text{NH}}_4^{+}$ toxicity as an adaptive strategy.

In summary, this work highlights contrasting effects of LN and N forms on root plasticity that depends on auxin and BR signaling pathways. The mechanism explored in this study may provide unique advancement towards improving the RSA for increased N foraging response and adaptation under differential availability of N forms.

Materials and methods

Plant material, chemical treatments, and quantification of PR and LR length

Arabidopsis (A. thaliana; Col-0) seeds were sterilized and sown on half strength Murashige and Skoog (1/2 MS) agar containing plates as described (Fridman et al., 2014). Following 2 days of dark incubation at 4°C, plates were placed vertically in a growth chamber (16-h light/8-h dark cycle with the light intensity of 120 µmol photons m⁻² s⁻¹). Three days after germination, seedlings were transferred to 1/2 MS medium with altered N salts (Singh et al., 2018). SN medium contains 9.2-mM NH₄NO₃ and 10-mM KNO₃ and the low LN medium contains 0.05-mM NH₄NO₃ and 0.05-mM KNO₃. The pH of the medium was adjusted to 5.8. After transferring to the modified 1/2 MS medium, seedlings were allowed to grow for 8 days unless indicated in the figure legends. To make the 1/2 MS medium with individual N sources, N salts from the SN medium were omitted with the specific N forms. For ${\text{NH}}_4^{+}$⁠, 1-mM NH₄SO₄ and for ${\mathrm{NO}}_3^{-}$⁠, 10-mM KNO₃ (comparable to SN medium) (Sigma Aldrich) was used unless it is indicated. For chemical, BRZ 220 (Asami et al., 2000), eBL (Sigma - Aldrich), 2,4-D (Sigma - Aldrich), IAA (Sigma - Aldrich), and TIBA (Sigma - Aldrich) treatments, modified 1/2 MS medium was mixed with indicated concentration of each chemical while pouring the plates. The PR length was quantified using ImageJ software presented in the graphs. A minimum of three independent biological replicates were performed unless indicated in figure legends. The following transgenic lines and mutants were used in this study; tir1 (CS3798), tir1:pTIR1-gTIR1-VENUS (CS69661), P21Y (UBQ10prom::CITRINE-2xPH(FAPP1)/pB7m34GW) (CS2105612; Simon et al., 2014), LTI6b–GFP (Cutler et al., 2000), bri1-116; pGL2:BRI1-GFP, bri1-116; pSCR:BRI1-GFP, bri1-116; pSHR:BRI1-GFP, bri1-116 (Hacham et al., 2011), pBZR1:BZR1-CFP (Wang et al., 2002), pPIN1:PIN1-GFP (Kumar Meena et al., 2019), 35S:BKI1-YFP (Wang and Chory, 2006), pPIN2:PIN2-GFP, DII-VENUS (Brunoud et al., 2012), DR5-GUS (Savaldi-Goldstein et al., 2008),bki1-1(Jiang et al., 2015), aux1-7 (CS3074), pTIR1:tir1E12K-GUS (CS69663), pBRI1:BRI1-GFP (Geldner et al., 2007), bri1, bri1:pBRI1:BRI1-GFP, and 35S:BRI1-GFP (Holzwart et al., 2018).

RNA extraction, transcriptome sequencing, and gene expression analysis

For RNA extraction, 7-day-old seedlings were transferred to an appropriate medium (SN and LN) for 4 days (for RNA sequencing) and total RNA from the roots was isolated using RNA extraction kit (Sigma Aldrich). Sequencing was performed on Illumina HiSeq 2500 System to obtain 50 bp single-end reads from three independent biologicals. Quality was assessed using fastp (FASTQ) tool with the default setting. The HISAT2 algorithm was used to map the read counts with the TAIR10 assembly of the A. thaliana genome. The number of reads that mapped or aligned with a particular gene were used to normalize the read counts, model-dependent P-value, and false discovery rate (0.05 or less) estimation based on multiple hypothesis testing using DESeq2 package. Genes were prioritized with |log2FC| > 1 and differentially regulated genes SN versus LN were represented (Supplemental Table S1). For pathway enrichment, a web-based tool was used as described (Supek et al., 2011; Ge et al., 2020). For heat map, MultiExperiment Viewer program was used (Howe et al., 2011). Gene expression analysis was performed in a real-time PCR machine (CFX96 touch; Bio-Rad, Hercules, CA, USA) using SYBR green master mix (Applied Biosystems, Waltham, MA, USA) and normalized against the Arabidopsis At5g15400 gene. Primers used in the study are given in Supplemental Table S2.

Confocal imaging, cellular analysis, and fluorescence signal quantification

For visualization of propidium iodide (PI), VENUS, YFP, GFP, CITRINE, and CFP fluorescence, Leica TCS-SP8 microscope was used (Leica microsystems, Wetzlar, Germany). PI was observed at excitation and emission wavelengths of 561 and 615 nm, respectively. For growth profiling, Z stack images (3-µm intervals) were captured and Fiji-ImageJ tool was used for cell numbers and length quantification from cortex/endodermis initial (CEI) to elongation/differentiation zone. For NH₄⁺-grown roots, cellular analysis was performed only in the roots showing mild phenotype since severe phenotype under this condition, meristem was exhausted and cell numbers could not be quantified. VENUS, YFP, and GFP fluorophores were excited at 514 and 488 nm, respectively, and emission was detected at 500–530 nm. CITRINE was analyzed as described (Simon et al., 2014). CFP was excited at a wavelength of 434 nm and emission was detected at 474 nm. Laser intensity for all experiments was kept between 15% and 20% for all the fluorescent proteins. To quantify the GFP/YFP signal intensity, Fiji-ImageJ tool was used and the signal intensity was normalized with the area and represented as mean signal intensity in arbitrary units (a.u). For vasculature/inner cell signal intensity, the mean YFP/GFP fluorescence signal upto the second cell (cortex) above the meristem was quantified from the QC cells in the stele tissues and normalized with the area as described in figure legends. PIN2-GFP, P21Y, and LTI6b-GFP epidermal signals were quantified from Z-stack images from the transition/elongation zone (TEZ) cells. Nuclear DII-VENUS signal was quantified in the first 8–10 cells of epidermis layer from CEI. Minimum three independent biological replicates were performed unless indicated in the figure legends.

Histochemical GUS analysis

GUS staining was performed as described (Hacham et al., 2011). Eleven-day-old seedlings grown on an appropriate N-containing medium (SN, ${\mathrm{NO}}_3^{-}$⁠, and ${\text{NH}}_4^{+}$⁠) were treated with eBL, IAA, or 2,4-D as indicated. Images were captured in a light microscope (Nikon, Tokyo, Japan).

Protein extraction and western blotting

Shoot and root tissues from 11-day-old seedlings grown on SN and LN medium for 8 days were harvested, crushed in liquid N, and 300 mg powder was dissolved in extraction buffer (50-mM Tris–HCl, pH 7.4, 150-mM NaCl, 0.05% Tween-20 (v/v), 1-mM EDTA, and 1% protease inhibitor cocktail [w/v]). The extracted samples were centrifuged (15,000 rpm for 30 min) and equal amounts of supernatant (30 µL) were mixed with Laemmli buffer and separated on a 10% sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE) followed by transfer (300 mA for 1 h 30 min) to a PVDF membrane (Millipore, Burlington, MA, USA). For detection of YFP, anti-GFP antibody (Cloud-Clone Corp, 2:10,000 dilutions) were used followed by incubation with secondary antibodies for 1 h (anti-Rabbit-HRP, GE Healthcare, Chicago, IL, USA). The membrane was exposed using the Clarity Max Western ECL substrate (Bio-Rad) in a ChemiDoc (Bio-Rad) for 100–200 s. Membrane was stripped (200-mM glycine, 0.1% SDS [w/v], 1% Tween-20 [v/v]) for 10 min (2 times) and used for detection of actin using anti-actin antibody (2:10,000 dilutions; Sigma, St Louis, MO, USA) followed by incubating with secondary antibody (anti-mouse-HRP, 1:10,000 dilutions, GE Healthcare) for 1 h.

Statistical analysis

One-way analysis of variance was used to compare the root length and cell length followed by post-hoc Tukey’s test at P < 0.05. In the measurements where variances were unequal (Figures 1, A, D, J, and K, 2, I, and K, 3A, 4A, and 5, B, C, and E; Supplemental Figures S1B and S6F) post-hoc Tukey’s test was performed on ranked data.

Accession numbers

AT5G42750 (BKI1), AT4G39400 (BRI1), AT1G75080 (BZR1), AT1G73590 (PIN1), AT5G57090 (PIN2), AT1G70940 (PIN3), AT2G01420 (PIN4), AT2G38120 (AUX1), and AT3G62980 (TIR1).

Supplemental data

The following materials are available in the online version of this article.

Supplemental Figure S1. LN-grown roots show a reduced response to exogenous auxins.

Supplemental Figure S2. TIBA represses PR length in an N-dependent manner.

Supplemental Figure S3. RNA sequencing analysis of the roots grown on SN and LN medium.

Supplementary Figure S4. LN promotes BR signaling and response.

Supplemental Figure S5. N and auxin-dependent regulation of BKI1.

Supplemental Figure S6. TIBA regulates BKI1 in an N-dependent manner.

Supplemental Figure S7. Enhanced BR signaling under LN negatively affects PR length in bki1-1 and 35S:BRI1-GFP lines.

Supplemental Figure S8. BR signaling from inner cell files promotes root growth and cell elongation in response to LN.

Supplemental Figure S9. N source-specific regulation of BR signaling activity and response.

Supplemental Figure S10. Effect of BR and 2,4-D on meristem size and cell length in the roots grown on SN, ${\text{NH}}_4^{+}$⁠, and ${\mathrm{NO}}_3^{-}$ medium.

Supplemental Table S1. Differentially regulated genes (SN vs LN).

Supplemental Table S2. Primers used in the study.

A.P.S. conceived the research plans and supervised the experiments. A.P.S., L.L.D., A.P., and S.G. interpreted the results. A.P.S. wrote the manuscript with contributions from all the authors. L.L.D., A.P., and S.G. performed the experiments. All authors read, corrected, and approved the manuscript.

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (https://academic.oup.com/plphys/pages/general-instructions) is Amar Pal Singh ([email protected]).

Acknowledgments

We thank Prof. Sigal Savaldi-Goldstein (Technion, Israel) and Dr. Aashish Ranjan (NIPGR) for their critical suggestion to the manuscript. We thank Prof. T. Asami (University of Tokyo) for providing BRZ, ABRC for providing seeds (tir1, tir1:pTIR1-gTIR1-VENUS, aux1-7, and tir1E12K), Prof. Dr. Sebastian Wolf (Heidelberg University/Eberhard Karls University, Germany) for providing bri1 and bri1:pBRI1-BRI1-GFP and Dr. Ashverya Laxmi (DBT-NIPGR) for pPIN2:PIN2-GFP and tir1-1 seeds. Authors acknowledge DBT-NIPGR central instrumentation and confocal microscopy facility for help in various experiments. Authors are thankful to the DBT e-library Consortium (DeLCON) for providing access to e-resources.

Funding

This study was supported by the Department of Biotechnology (DBT) India for the Har Govind Khorana-Innovative Young Biotechnologist Award grant (BT/11/IYBA/2018/02) and DST-SERB, India for the grant (ECR/2018/000526) to A.P.S. A.P., S.G., and L.L.D. acknowledge University Grants Commission of India (UGC, India) and Council of Scientific & Industrial Research, India (CSIR, India) for Junior/Senior Research Fellowship (JRF/SRF) fellowship, respectively.

Conflict of interest statement. The authors declare no conflict of interest.

Author notes

†

Senior author