Skip to main content
NCBI home page
The Plant Cell logoLink to The Plant Cell
. 2022 May 30;34(9):3319–3338. doi: 10.1093/plcell/koac161

Alternative splicing of REGULATOR OF LEAF INCLINATION 1 modulates phosphate starvation signaling and growth in plants

Meina Guo 1,#, Yuxin Zhang 2,#, Xianqing Jia 3, Xueqing Wang 4, Yibo Zhang 5, Jifeng Liu 6, Qingshen Yang 7, Wenyuan Ruan 8,✉, Keke Yi 9,✉
PMCID: PMC9421462  PMID: 35640569

Abstract

Phosphate (Pi) limitation represents a primary constraint on crop production. To better cope with Pi deficiency stress, plants have evolved multiple adaptive mechanisms for phosphorus acquisition and utilization, including the alteration of growth and the activation of Pi starvation signaling. However, how these strategies are coordinated remains largely unknown. Here, we found that the alternative splicing (AS) of REGULATOR OF LEAF INCLINATION 1 (RLI1) in rice (Oryza sativa) produces two protein isoforms: RLI1a, containing MYB DNA binding domain and RLI1b, containing both MYB and coiled-coil (CC) domains. The absence of a CC domain in RLI1a enables it to activate broader target genes than RLI1b. RLI1a, but not RLI1b, regulates both brassinolide (BL) biosynthesis and signaling by directly activating BL-biosynthesis and signaling genes. Both RLI1a and RLI1b modulate Pi starvation signaling. RLI1 and PHOSPHATE STARVATION RESPONSE 2 function redundantly to regulate Pi starvation signaling and growth in response to Pi deficiency. Furthermore, the AS of RLI1-related genes to produce two isoforms for growth and Pi signaling is widely present in both dicots and monocots. Together, these findings indicate that the AS of RLI1 is an important and functionally conserved strategy to orchestrate Pi starvation signaling and growth to help plants adapt to Pi-limitation stress.

Alternative splicing of REGULATOR OF LEAF INCLINATION 1 is an important strategy that orchestrates phosphate (Pi) starvation signaling and growth to help plants adapt to Pi-limitation stress.

In a Nutshell.

Background: Phosphate (Pi) limitation represents a primary constraint on crop production. To better cope with Pi deficiency stress, plants have evolved multiple adaptive mechanisms for phosphorus acquisition and utilization, including the alteration of growth and the activation of Pi starvation signaling.

Question: Despite significant advances in understanding the mechanisms underlying the regulation of Pi-associated shoot architecture, Pi-signaling, and Pi-homeostasis under Pi deficiency stress in plants, how these strategies are coordinated remains largely unknown.

Findings: We provide evidence that the alternative splicing (AS) of REGULATOR OF LEAF INCLINATION 1 (RLI1) produces two protein isoforms in rice: RLI1a (containing MYB DNA binding domain) and RLI1b (containing both MYB and coiled-coil [CC] domains). Both RLI1a and RLI1b can modulate Pi starvation signaling. However, the absence of the CC domain in RLI1a enables it to directly activate a broader range of target genes than RLI1b, including genes involved in both brassinolide biosynthesis and signaling, thereby modulating plant growth. We also found that the AS of RLI1-related genes for both Pi signaling and growth is a widely present mechanism in plants.

Next steps: Given the functionally conserved strategy to orchestrate Pi starvation signaling and growth in plants, we aim to create smart crops with ideal shoot architecture and high phosphorus utilization efficiency by molecular design breeding.

Introduction

Phosphorous (P) is one of the most important macronutrients for plant growth and productivity. Despite its high abundance in soil, P is one of the most limiting elements for plants due to the low solubility and high adsorption capacity of orthophosphate (inorganic phosphate, Pi), the preferred form absorbed by plants (Holford, 1997; Schachtman et al., 1998). To cope with low Pi availability in soil, plants employ a series of adaptive responses for increasing Pi uptake and mobilization (reviewed in Vance et al., 2003; Franco-Zorrilla et al., 2004; Chiou and Lin, 2011; Wu et al., 2013; Puga et al., 2017). These adaptive response strategies usually introduce external morphological alterations and internal molecular changes in plants.

The typical morphological changes in response to low Pi stress are changes in root and shoot architecture. Typically, root length and density increase under Pi deficiency stress to increase the interaction area between the roots and soil for better Pi acquisition (López-Bucio et al., 2003; Péret et al., 2011). Additionally, the root angles are modulated to increase the presence of roots in the topsoil to help the plant better explore Pi derived from organic decomposition (Lynch and Brown, 2001; Williamson et al., 2001; López-Bucio et al., 2002). Meanwhile, shoot growth is repressed by stress. For instance, in rice (Oryza sativa), Pi starvation induces P-type CYCLIN 4 (CYCP4) expression to inhibit shoot cell division (Xu et al., 2020). Shoot architecture is also altered to make full use of P preserved in plants or to relieve the adverse impact of Pi-deficiency stress. In rice, the leaf blades become erect, reducing the photon capture area of leaves, thus decreasing the photosynthetic capacity and reducing Pi consumption to help the plant withstand Pi deficiency stress (Dietz and Foyer, 1986; Natr, 1992; Ruan et al., 2019).

To better adapt to the crisis of Pi deprivation, the composition of multiple internal molecules change, such as Pi-transporters, Pi-scavenging enzymes and membrane lipids (reviewed in Chiou and Lin, 2011; Wu et al., 2013; Puga, et al., 2017). The levels of phytohormones such as cytokinin (CK), brassinolide (BL), auxin (IAA), and strigolactones are also altered; these phytohormones are critical for helping the plant adapt to low Pi stress (reviewed by Chiou and Lin, 2011).

Processes that help plants adapt to low Pi stress are regulated by a sophisticated Pi-monitoring system involving Pi-homeostasis and Pi-signaling (Chiou and Lin, 2011; Wu et al., 2013; Pugga et al., 2018). PHOSPHATE STARVATION RESPONSE proteins (PHRs, MYB DNA binding domain, and coiled-coil [CC] contained transcription factors) and SPX domain-containing proteins (SPXs for Syg1/Pho81/XPR1) are key components of this regulatory system (Chiou and Lin, 2011; Wu et al., 2013; Puga et al., 2017). PHRs, such as AtPHR1/AtPHL1/AtPHL3 in Arabidopsis thaliana and OsPHR1/2/3/4 in rice, directly activate the expression of the Pi starvation-induced (PSI) genes containing PHR1 binding sites (P1BS: GNATATNC) in their promoters (Rubio et al., 2001; Zhou et al., 2008; Bustos et al., 2010; Guo et al., 2015; Sun et al., 2016; Ruan et al., 2017). Enhancing the expression of PHRs genes results in Pi over-accumulation in shoots and induces Pi-starvation signaling even under Pi-sufficient conditions. In contrast, the malfunction of PHRs leads to the repression of Pi-starvation signaling and the disruption of Pi homeostasis. The activity of PHRs is negatively regulated by SPXs proteins via physical interactions (Puga et al., 2014; Wang et al., 2014). PHRs and SPXs coordinately regulate low Pi stress-induced root hair elongation (Zhou et al., 2008; Bustos et al., 2010; Puga et al., 2014; Wang et al., 2014; Guo et al., 2015). We recently reported that SPX DOMAIN GENE 1 (SPX1) and REGULATOR OF LEAF INCLINATION 1 (RLI1, RLI1a in this study) modulate shoot architecture (leaf inclination) in response to Pi-deficiency stress in rice (Ruan et al, 2018). Despite significant developments in understanding the regulation of Pi-associated shoot architecture, Pi-signaling, and Pi-homeostasis, how they are coordinately regulated in rice is largely unknown.

Here, we provide evidence that alternative splicing (AS) of RLI1 produces two protein isoforms; RLI1a (solely containing an MYB DNA binding domain) and RLI1b (with both MYB and CC domains) to modulate Pi-associated shoot architecture, Pi-signaling, and Pi-homeostasis in rice. The absence of the CC domain in RLI1a enables it to activate a broader range of target genes than RLI1b on a genome-wide scale. RLI1a regulates both BL biosynthesis and signaling, as well as Pi starvation signaling and Pi-homeostasis, whereas RLI1b mainly affects Pi starvation signaling and Pi-homeostasis. We also found that the AS mechanism of RLI1 is widespread among dicots and monocots. Therefore, our results indicate that the AS of RLI1 is an important strategy to modulate Pi starvation signaling and growth in plants.

Results

Alternative pre-mRNA splicing produces two RLI1 isoforms

We previously showed that RLI1, a GOLDEN2, ARR-B, and Psr1 (GARP) subfamily transcription factor without a CC domain, regulates leaf inclination in response to Pi availability in rice (Ruan et al., 2018). Genome annotation and publicly available RNA-Sequencing data indicated that RLI1 has two different transcript variants. The retention of the fifth intron leads to an early stop codon in RLI1a, which encodes an MYB transcription factor, while the splicing of the fifth intron in RLI1b enables it to encode an MYB-CC protein (Supplemental Figure S1). We hypothesized that RLI1 modulates both shoot architecture and Pi signaling.

To test this hypothesis, we first determined whether these two transcript variants and their corresponding protein isoforms were actually present in rice. Two pairs of primers (F/Ra and F/Rb) were designed to amplify the possible variants (Figure 1A). RT-PCR analysis showed that both RLI1a and RLI1b transcript variants were present in rice (Figure 1C and Supplemental Figure S1). We raised polyclonal antibodies against the N-terminal fragment of RLI1, which could detect both RLI1a and RLI1b in vivo, and performed immunoblot analysis. Two specific bands between 35 and 45 kD were detected in wild-type (WT) Nipponbare (NIP) rice, but not in the rli1 mutant. We determined that the two bands were RLI1a (∼37 kD) and RLI1b (∼42 kD), respectively (Figure 1D). Together, these results demonstrate that RLI1 has two transcript variants and can produce RLI1a and RLI1b isoforms in rice.

Figure 1.

Figure 1

Open in a new tab

RLI1 generates two transcripts and their corresponding protein isoforms. A, Gene structure of RLI1 and the primers used to identify two different RLI1 splicing variants. Boxes indicate exons, solid lines indicate introns, dotted lines indicate splicing sites of introns, and the box with dots indicates the intron for RLI1b. “ATG” indicates the start codon and “TAG” indicates the stop codon. F is the forward primer. Ra is the reverse primer for RLI1a amplification. Rb is the reverse primer for RLI1b amplification. bp, base pairs. B, Protein structures of RLI1a and RLI1b. MYB indicates the DNA binding domain of RLI1a and RLI1b. CC indicates the CC domain. C, RT-PCR analysis of the different transcript isoforms of RLI1. D, Immunoblot analysis of the protein isoforms of RLI1a and RLI1b. The nuclear proteins of the WT (NIP) and rli1-1 were isolated for immunoblot assays with anti-RLI1 antibody. Coomassie brilliant blue (CBB) staining indicates that similar amounts of proteins were loaded. * indicates nonspecific bands.

RLI1a and RLI1b respond differently to external Pi availability

Since RLI1a and RLI1b are actually present in planta, we examined their transcript and protein levels in response to Pi-deficiency stress. We previously showed that the transcription of RLI1 and RLI1 promoter activity were repressed by Pi-deficiency stress (Ruan et al., 2018). Here, we found that the transcription of RLI1a but not RLI1b was strongly repressed under −P conditions (Figure 2A). We examined whether the protein levels of RLI1a and RLI1b respond to external Pi-availability and found that the RLI1a protein levels were significantly lower under −P conditions than under +P conditions (Figure 2, B and C). However, RLI1b protein levels significantly increased under −P conditions (Figure 2, B and C). These results suggest that Pi starvation might stabilize RLI1b.

Figure 2.

Figure 2

Open in a new tab

RLI1a and RLI1b respond differently to Pi-deficiency stress. A, qRT-PCR analysis of the expression levels of RLI1a and RLI1b in WT plants grown under +P (200 μM Pi) and −P (0 μM Pi) conditions. Ten-day-old plants grown under +P conditions were transferred to +P and −P conditions for another 10 days. Total RNA was isolated from shoots for qRT-PCR analysis with specific primers for RLI1a and RLI1b (Primers are listed in Supplemental Data Set S2). Values represent means ± sd three replicates. B, Immunoblot analysis of RLI1a and RLI1b protein levels in WT plants grown under +P and −P conditions. The shoots of 20-day-old plants were used to isolate nuclear proteins for immunoblotting with anti-RLI1 antibody. CBB staining indicates that similar amounts of proteins were loaded. * indicates nonspecific bands. C, Quantification of the relative protein levels in (B). The protein level under +P conditions was set as 1. D, Cell-free degradation assay of RLI1a and RLI1b in the extracts of WT plants grown under +P and −P conditions. RLI1a-HIS or RLI1b-HIS (100 ng) was incubated at different time points at 28°C with protein extracts from leaves (20 mg total proteins) of WT grown under +P and −P conditions. RLI1a-HIS or RLI1b-HIS was detected by immunoblotting using anti-HIS antibody. CBB staining indicates that similar amounts of proteins were loaded. E, The relative amounts of RLI1a-HIS or RLI1b-HIS remaining after incubation in WT extracts were calculated and plotted on a log graph. The protein levels of RLI1a-HIS or RLI1b-HIS were normalized with that at “0” min. Data represent means ± sd (n = 3, three individual analyses). Data significantly different from the corresponding controls are indicated. Significantly lower: ##P < 0.01; significantly higher: **P < 0.01; ns, no significant difference (Student’s t test).

To further confirm this notion, we conducted a cell-free degradation analysis of the RLI1a-HIS and RLI1b-HIS recombinant proteins in the total protein extracts from plants grown under +P and −P conditions. Immunoblotting showed that RLI1a-HIS was degraded with a half-life of ∼30 min when incubated with +P extract (Figure 2, D and E). However, when incubated with −P extract, the degradation of RLI1a-HIS was much faster, with a half-life of ∼10 min (Figure 2, D and E). This indicates that RLI1a is more stable under +P conditions. Unlike RLI1a-HIS, RLI1b-HIS displayed a degradation half-life of ∼6 and ∼20 min in +P and −P extracts, respectively. These results suggest that RLI1b-HIS is more stable under −P conditions (Figure 2, D and E).

Together, these results indicate that the transcript and protein levels of RLI1a and RLI1b respond to external Pi availability differently.

RLI1a, but not RLI1b, mediates the regulation of leaf inclination

Given the finding that RLI1a and RLI1b respond differently to Pi deficiency stress and that RLI1 (RLI1a in this study) functions in leaf inclination in response to external Pi availability (Ruan et al., 2018), we hypothesized that RLI1a and RLI1b might play different roles in regulating leaf inclination. To test this hypothesis, we examined the shoot architecture of RLI1a (a-Ox-3, a-Ox-7, and a-Ox-10; Ruan et al., 2018) and RLI1b (b-Ox-1, b-Ox-9, and b-Ox-12, in this study) overexpression lines (Supplemental Figures S3 and S4). Consistent with a previous report (Ruan et al., 2018), the shoot architecture of RLI1a-Ox lines was much looser than that of WT plants (Figure 3, A and C). However, there was no obvious difference in shoot architecture between RLI1b-Ox lines and WT plants (Figure 3, A–C). In keeping with this result, the leaf inclination angle of RLI1a-Ox was significantly greater than that of WT, but RLI1b-Ox plants were similar to WT (Figure 3, B and D), indicating that only RLI1a, but not RLI1b regulates the leaf inclination in rice.

Figure 3.

Figure 3

Open in a new tab

RLI1a, but not RLI1b, regulates leaf inclination. A, Phenotypic performance assays of the WT, RLI1a overexpression lines (a-Ox-3/7/10), and RLI1b overexpression lines (b-Ox-1/9/12) under hydroponic culture conditions. B, Leaf inclination assays of WT, a-Ox, and b-Ox in (A). The third leaves from the top of the main tiller were used to analyze leaf inclination. Values represent the means ± sd of six independent plants. Data significantly different from the corresponding controls are indicated. **P < 0.01; ns: no significant difference (Student’s t test). C, Phenotypic performance of WT, irl1, a-Ox-3, and b-Ox-1 plants grown in soil. Scale bars, 10 cm. D, Leaf inclination assays of WT, irl1, a-Ox-3, and b-Ox-1 plants in (C). The second leaves from the top were used to measure leaf inclination. Values represent the means ± sd of ten independent plants. E, Longitudinal sections of the lamia joints of the second top leaf from WT, rli1, RLI1a-Ox-3, and RLI1b-Ox-1 plants. Scale bars, 50 μm. F, Lamina joint cell length. Values represent means ± sd (n = 30–50). Different letters above the bars indicate significant differences between groups. Statistics, one-way ANOVA with post-hoc Tukey’s test (P < 0.05).

Our previous study showed that RLI1 (RLI1a in this study) affects lamina joint cell elongation to regulate leaf inclination in response to external Pi availability (Ruan et al., 2018). We therefore measured the lamina joint cell length in longitudinal sections of both RLI1a-Ox and RLI1b-Ox plants The cell length of RLI1a-Ox-3 was significantly greater than those of the other plants, while there was no significant difference between RLI1b-Ox-1 and WT (Figure 3, E and F). These results demonstrate that, unlike RLI1a, RLI1b does not affect lamina joint cell elongation to modulate leaf inclination (Figure 3, E and F). Furthermore, the leaf inclination of RLI1b-Ox plants was also comparable to that of WT under −P conditions (Supplemental Figure S5). These results indicate that PSI leaf erectness results from the inhibition of RLI1a but not RLI1b.

In summary, we demonstrated that RLI1a, but not RLI1b, regulates shoot architecture in rice.

Both RLI1a and RLI1b regulate Pi accumulation and Pi signaling

Given the finding that RLI1b contains both MYB and CC domains, which shows high similarity to PHR2, we hypothesized that RLI1b, not RLI1a, regulates Pi accumulation in rice. To test this notion, we analyzed Pi accumulation in both RLI1a and RLI1b overexpression lines. The RLI1b-Ox lines displayed leaf tip necrosis, a typical symptom of Pi toxicity in rice. Unexpectedly, RLI1a-Ox also showed leaf tip necrosis (Figures 3A and 4A). To determine whether the leaf tip necrosis in these overexpression lines was due to Pi over accumulation, we grew these plants under both Pi-sufficient and Pi-deficient conditions and measured the Pi content in planta. The leaf tip necrosis symptom of these over-expression lines almost disappeared when grown under Pi-deficient conditions. Measurements of Pi contents in these overexpression lines corroborated the finding that both RLI1a-Ox and RLI1b-Ox plants accumulate higher shoot Pi than WT under Pi-sufficient conditions, while Pi-deficiency treatment significantly alleviated these differences (Figure 4, B and C).

Figure 4.

Figure 4

Open in a new tab

Both RLI1a and RLI1b regulate Pi-homeostasis and Pi-starvation signaling. A, Phenotypic performance of the WT, RLI1a overexpression lines (a-Ox-3/7/10), and RLI1b overexpression lines (b-Ox-1/9/12) leaves. The plants were grown under Pi sufficient (HP, 200 μM Pi) and deficient (HP, 5 μM Pi) conditions for 30 days. The third leaves from the top of the main tiller were used for analysis. Scale bars, 1 cm. B and C, Pi measurement of WT, a-Ox, and b-Ox grown under HP and LP conditions. FW indicates fresh weight. Values represent means ± sd three replicates. D, Relative expression levels of IPS1, SPX1, VPE2, PT2, and PT10 in WT, RLI1a-Ox, and RLI1b-Ox plants. Values represent the means ± sd of three replicates. The expression levels of the detected genes in WT were set as 1. The RT-qPCR primers are listed in Supplemental Data Set S2. Data significantly different from the corresponding controls are indicated. Significantly higher: *P < 0.05 **P < 0.01 (Student’s t test).

Since both RLI1a and RLI1b regulate Pi accumulation in rice, we analyzed the expression of PSI genes in WT, RLI1a-Ox and RLI1b-Ox plants. The Pi-starvation signaling and Pi uptake related genes IPS1, SPX1, VACUOLAR PHOSPHATE EFFLUX TRANSPORTER 2 (VPE2), and the Pi transporter genes PT2 and PT10 were significantly induced in RLI1a-Ox and RLI1b-Ox plants even under Pi-sufficient conditions (Figure 4D), suggesting that both RLI1a and RLI1b positively regulate Pi starvation signaling and Pi homeostasis. Taken together, these results demonstrate that both RLI1a and RLI1b function in regulating Pi accumulation and Pi starvation signaling in rice.

RLI1a has broader targets than RLI1b

Since RLI1a and RLI1b have an identical MYB binding domain (DNA binding domain) and play different as well as similar roles in rice, we reasoned that the presence of CC domain might affect the DNA binding ability of RLI1b. We previously showed that RLI1a could not form a homodimer (Ruan et al., 2018). RLI1a binds to the R1BS cis-element in its target genes (RLI1a binding site, NAKATNCN; Ruan et al., 2018). Given that the CC domain commonly plays an important role in protein dimerization (Bustos et al., 2010), this observation suggests that RLI1b forms a homodimer, which affects its DNA binding properties. To investigate this notion, we first tested the protein-protein interactions between RLI1a, RLI1b, and PHR2 in a bimolecular fluorescence complementation assay in Nicotiana benthamiana leaf epidermal cells. The results showed that RLI1b interacts with itself and with PHR2, while RLI1a does not (Supplemental Figure S6).

We reasoned that the different dimerization properties of RLI1 isoforms might affect their DNA binding ability. We examined this issue using an electrophoretic mobility shift assay (EMSA). Both RLI1a and RLI1b bound to R1BS (RLI1a binding site, NAKATNCN) and P1BS (PHR1 binding site, GNATATNC) cis-elements in vitro (Figure 5, A and B). However, RLI1a displayed a higher affinity for R1BS and RLI1b for P1BS (Figure 5B). To verify these differences in DNA affinity, we performed a yeast-one hybrid (Y1H) assay. Indeed, RLI1a had a higher DNA binding affinity for the R1BS motif, while RLI1b had a higher affinity for the P1BS motif (Figure 5C).

Figure 5.

Figure 5

Open in a new tab

RLI1a and RLI1b have different motif recognition properties. A, WebLogos of the R1BS and P1BS motifs. B, EMSA showing that both RLI1a and RLI1b can bind to the R1BS and P1BS cis-elements. DNA fragments containing R1BS and P1BS were incubated with RLI1a-His and RLI1b-His recombinant proteins as indicated. 6xHis proteins were used as negative controls. C, Y1H analysis of the DNA binding ability of RLI1a and RLI1b with the R1BS and P1BS motifs. pLacZ2u was used as a negative control. β-Galactosidase activity was measured to indicate the DNA binding ability of RLI1a and RLI1b with the R1BS and P1BS motifs. Values represent means ± sd of four replicates. (**P < 0.01; Student’s t test). D, Analysis of the transcriptional activation properties of RLI1a and RLI1b in N. benthamiana leaves. A LUC reporter driven by the artificial promoters 4×R1BS-min35S and 4×P1BS-min35S was used to test the transcriptional activity of RLI1a and RLI1b. The different combinations were co-transformed in N. benthamiana leaves via A. tumefaciens-mediated transfection and observed 2 days later; empty vectors with 4×R1BS-min35S and 4×P1BS-min35S were used as the negative control. E, Assays of RLI1a and RLI1b transactivation of the artificial promoters 4×R1BS-min35S and 4×P1BS-min35S using the dual-LUC transient transactivation assay system in N. benthamiana leaves. Data represent means ± SD (n = 4, four repeats in different N. benthamiana leaves in one experiment) (**P < 0.01; Student’s t test).

To further confirm the different DNA binding affinities between RLI1a and RLI1b in planta, we generated constructs harboring two luciferase (LUC) reporters driven by an artificial promoter of four tandem copies of R1BS or P1BS fused to the minimal 35S promoter from CaMV (4×R1BS-min35S:LUC or 4×P1BS-min35S:LUC) and co-injected them into N.benthamiana leaf epidermal cells to monitor the DNA binding ability of RLI1a and RLI1b. After normalization to the internal control, the luminescence signal of leaves co-injected with 35S-RLI1a/4×P1BS-min35S:LUC was lower than that of 35S-RLI1b/4×P1BS-min35S:LUC. In contrast, the 35S-RLI1a/4×R1BS-min35S:LUC signal was more intense than that of 35S-RLI1b/4×R1BS-min35S:LUC (Figure 5, D and E). These results indicate that RLI1b has higher DNA binding affinity to P1BS than RLI1a, while RLI1a has higher DNA binding affinity to R1BS than RLI1b.

Given that RLI1a and RLI1b display different DNA binding affinities to R1BS and P1BS, we hypothesized that RLI1a and RLI1b might preferentially target different genes in vivo. To test this hypothesis, we performed Chromatin Immunoprecipitation followed by high-throughput Sequencing (ChIP-Seq) of 35S-RLI1a-FLAG and 35S-RLI1b-FLAG seedlings.

The ChIP-Seq data showed that RLI1a bound to more DNA fragments in vivo compared with RLI1b, with 11,417 DNA binding peaks by RLI1a and only 3,415 peaks by RLI1b (Figure 6, A and B). Further candidate target genes were associated with peaks within 3k bp upstream to 1k bp downstream of their coding sequences. As expected for transcription factors, both RLI1a and RLI1b frequently bound near the transcription start sites (TSSs) of genes (Supplemental Figure S7). RLI1a targeted 6,716 genes, while RLI1b only targeted 2,884 genes in the rice genome (Figure 6C). For these genes, a set of 2,047 genes were common targets of both RLI1a and RLI1b, while 4,669 and 837 genes were only bound by RLI1a and RLI1b, respectively (Figure 6C). These results demonstrate that RLI1a has a broader range of target genes than RLI1b in vivo.

Figure 6.

Figure 6

Open in a new tab

RLI1a has broader targets than RLI1b in vivo. A, Peak numbers of RLI1a and RLI1b at different chromosome positions. The nuclei were isolated from shoots of transgenic plants harboring RLI1a-FLAG and RLI1b-FLAG for ChIP-seq assay with a CUT&Tag kit (NovoNGS). B, The total number of peaks targeted by RLI1a and RLI1b. Error bars are SD (n = 2). C, Venn diagram representing the overlap of RLI1a- and RLI1b-targeted genes. The numbers represent the number of genes directly targeted by RLI1a or RLI1b. The BR-biosynthesis and -signaling related genes or Pi-homeostasis and -signaling related genes are listed.

To verify the ChIP-seq data, we selected the common target genes of both RLI1a and RLI1b for further ChIP-qPCR analysis, that is, IPS1, SPX1, and VPE2. Both RLI1a-FLAG and RLI1b-FLAG were significantly enriched at the promoters of IPS1, SPX1, and VPE2 (Supplemental Figure S8), indicating that the ChIP-seq data were reliable.

To determine the possible pathways associated with RLI1a and RLI1b, we performed Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis of the target genes. KEGG analysis of the common target subset and genes only targeted by RLI1a revealed a significant overrepresentation of genes involved in the plant hormone signal transduction pathway (false discovery rate [FDR] = 0.000125; Supplemental Data Set S1). BL biosynthesis genes (such as D11, DWF4, and CYPD903) and BL-signaling genes (such as GSK3 and BZR1) were included in this set (Figure 6C and Supplemental Figure S9). Although the KEGG pathway brassinosteroid (BR) biosynthesis was not significantly enriched among these genes (FDR = 0.549), four of the nine BL biosynthesis-related genes were enriched among RLI1a targets. Consistent with the notion that RLI1a and RLI1b regulate Pi-signaling and Pi-uptake, some typical Pi starvation response pathways, including metabolic pathways, photosynthesis, and oxidative phosphorylation (Secco et al., 2013), were enriched in both the common RLI1a and RLI1b-targeted genes (Supplemental Data Set S1). Even though we did not find a Pi-signaling or Pi-homeostasis specific KEGG pathway term in the subset of genes (targeted by both RLI1a and RLI1b or only by RLI1b), a class of Pi starvation signaling and homeostasis related genes including IPS1, SPX1, SPX3, SQD2, PAP1, VPE1, VPE2, PT1, and miR827 were found in the set of both RLI1a- and RLI1b-targeted genes (Figure 6C and Supplemental Figure S9).

Together these results demonstrate that RLI1a and RLI1b have different DNA binding preferences, enabling them to target similar and/or different downstream genes.

RLI1a directly regulates BL biosynthesis and signaling to affect growth

Given that RLI1a but not RLI1b mediates the regulation of leaf inclination, and many BL-biosynthesis and signaling genes were detected in the RLI1a ChIP-Seq data, we hypothesized that RLI1a directly regulates the expression of BL-related genes to modulate BL contents and signaling, thus affecting rice growth. To test this hypothesis, we designed specific primers for the promoters of BL-biosynthesis genes (D11, DWF4, and BZR1) and used them for ChIP-qPCR analysis (Supplemental Figure S9). Similar to the positive control, a significant enrichment of RLI1a-FLAG but not RLI1b-FLAG was detected at the D11, DWF4 and BZR1 promoters (Figure 7A), suggesting that RLI1a, but not RLI1b, binds to these promoters and might affect D11, DWF4, and BZR1 expression in vivo. To further verify this notion, we used a LUC reporter driven by the D11 promoter to monitor the effect of RLI1a and RLI1b. Nicotianabenthamiana leaves injected with D11-pro-LUC or co-injected D11-pro-LUC with 35S-RLI1b only displayed basal luminescence signal. However, leaves co-injected with D11-pro-LUC and 35S-RLI1a showed intense luminescence (Figure 7B), suggesting that RLI1a directly activates the expression of genes related to BL biosynthesis and signaling. Consistently, the expression levels of D11, DWF4, and BZR1 were significantly induced in the RLI1a-Ox lines, but not in RLI1b-Ox plants (Figure 7C).

Figure 7.

Figure 7

Open in a new tab

RLI1a regulates brassinosteroid biosynthesis to affect leaf inclination. A, ChIP-qPCR analysis of the levels of RLI1a and RLI1b on the promoters of D11, DWF4, BZR1, and BU1. Chromatin immunoprecipitation was performed with an anti-FLAG antibody; ChIP without anti-FLAG was used as the negative control. Amounts of the immunoprecipitated genomic fragments were quantified by qPCR. The fold enrichments were calculated as follows: for each examined gene, the number of DNA fragments from each sample was normalized to the constitutively expressed ACTIN in each sample, and subsequently, the ACTIN-normalized values for RLI1a-Ox-3 without antibody were calculated. Shown are the means and ±sd of three ChIP experiments. B, RLI1a has higher transcriptional activating ability than RLI1b for D11 genes. A LUC reporter driven by the D11 promoter was used to test the transcriptional activity of RLI1a and RLI1b in N. benthamiana leaf epidermal cells. C, Relative expression levels of D11, DWF4, BZR1, and BU1 in WT, rli1-1, RLI1a-Ox, and RLI1b-Ox plants. Values represent the means ± sd of three replicates. The expression levels of the detected genes in WT were set as 1. The RT-qPCR primers are listed in Supplemental Data Set S2. D, Measurement of brassinosteroid (BL) levels in the shoots of WT, rli1-1, RLI1a-Ox-3, and RLI1b-Ox-1 plants. E and F, RLI1a-Ox-3 is more sensitive to exogenous eBL-24 treatments than RLI1b-Ox-1. Germinated WT, rli1-1, RLI1a-Ox, and RLI1b-Ox seeds were sown on nets floating on nutrient solution with or without exogenous brassinosteroid (BL, 2 μM) or the BL biosynthesis inhibitor BRZ (5 μM). Fourteen-day-old plants were used to further analysis. The lamina joint of the first leaf was used to take images (E) and to quantify the leaf inclination angles (F). G, Fold changes in leaf angle of WT, rli1-1, RLI1a-Ox, and RLI1b-Ox plants after treatment with BRZ and BL. The leaf angle of each plant was used to calculate fold change with the formula Log2[Mock/+BRZ/+BL versus Mock]. Data significantly different from the corresponding controls are indicated. Significantly higher: *P < 0.05; **P < 0.01; ns, no significant difference (Student’s t test). Different letters above the bars indicate significant differences between groups. Statistics, two-way ANOVA with post-hoc Tukey’s test (P < 0.05).

Since RLI1a directly activates the expression of BL-biosynthesis genes such as D11 and DWF4, we also measured the BL contents in RLI1a-Ox, RLI1b-Ox and rli1 plants. The BL content of RLI1a-Ox plants was almost three times as high as that of WT plants (Figure 7D). In contrast, the BL content of rli1 was lower than that of WT plants (Figure 7D). Consistent with the BL biosynthesis and signaling defect, the biomass of the rli1 mutant was significantly lower than that of WT. In line with the finding that RLI1b has little effect on the expression of BL-biosynthesis genes, the BL content of RLI1b-Ox was comparable to that of the WT (Figure 7D).

Consistent with the differences in BL content between lines and the notion that RLI1a rather than RLI1b activates BL signaling, the leaf inclination angle of RLI1b-Ox was comparable to that of WT under control conditions and under BL and brassinazole (BRZ, a specific BL biosynthesis inhibitor) treatments (Figure 7, E and F and Supplemental Figure S10). Also, RLI1a-Ox still displayed a larger leaf angle than WT following the inhibition of BL biosynthesis by BRZ. Conversely, the compromised RLI1a-induced BL signaling activation led to reduced sensitivity to BL treatment in the rli1 mutant compared with the WT (Figure 7, E and F). To further confirm the role of RLI1a in regulating BL biosynthesis and signaling, we also tested the root growth of the plants in response to BL and BRZ treatments. The BRZ and BL treatments induced comparable levels of root growth inhibition in WT and RLI1b-Ox. However, compared with the WT, RLI1a-Ox was more sensitive to BL treatment and rli1 was less sensitive to BL treatment and more sensitive to BRZ treatment (Supplemental Figure S10). These results highlight the importance of RLI1a in BL biosynthesis and signaling.

Together, these results demonstrate that RLI1a directly regulates the expression of BL-biosynthesis genes to modulate BL-homeostasis and signaling, thus affecting rice growth.

RLI1 and PHR2 cooperatively regulate plant adaption of Pi-deficiency stress

Given that RLI1b shares high similarity with PHR2 (a central regulator of Pi responses in rice) and that SPXs interact with both RLI1a and PHR2 to inhibit their transcriptional activity (Wang et al., 2014; Ruan et al., 2018), we reasoned that PHR2 could coordinate with RLI1 to regulate Pi starvation signaling to help plants adapt to Pi-deficiency stress. To investigate this issue, we first examined the interaction between RLI1b and SPXs by performing a LUC complementation imaging (LCI) assay in N. benthamiana leaf epidermal cells. Both RLI1a and RLI1b interacted with SPXs in the assay (Supplemental Figure S11). We then analyzed the growth phenotypes of WT, phr2, rli1, and rli1 phr2 plants under +P and −P conditions (Figure 8A and Supplemental Figure S12). Similar to a previous report (Guo et al., 2015), the biomass of the phr2 mutant was reduced under −P conditions compared with the WT and rli1, while the reduction in shoot biomass under Pi-deficiency stress was further aggravated in the rli1 phr2 double mutant (Figure 8B). Additionally, we observed that Pi-deficiency stress resulted in leaf senescence symptoms in rli1 phr2, which were more extreme than in the other plants, including the WT and each single mutant (Figure 8A). Furthermore, the Pi concentration in shoots was significantly lower in rli1 phr2 plants than in the other plants grown under both Pi-sufficient and Pi-deficient conditions (Figure 8C). These results indicate that RLI1 and PHR2 redundantly regulate Pi accumulation and plant growth in response to Pi-deficiency stress.

Figure 8.

Figure 8

Open in a new tab

RLI1 and PHR2 co-regulate plant adaptation to Pi starvation. A, Phenotypic performance of WT, phr2, rli1, and rli1 phr2 plants grown under (200 μM Pi) and −P (0 μM Pi) hydroponic culture conditions. Fourteen-day-old plants gown under +P conditions were transferred to +P or −P conditions for another 20  days before analysis. Scale bars, 10 cm. B, Biomass of plants in (A). DW, dry weight. Error bars are sd (n = 6). C, Pi concentration of WT, phr2, rli1, and rli1 phr2 plants grown under HP (200 μM Pi) and LP (5 μM Pi) hydroponic culture conditions. Error bars are sd (n = 3). D, Relative expression of Pi starvation response genes in WT, phr2, rli1, and rli1 phr2 plants grown under +P and −P hydroponic culture conditions. Ten-day-old plants grown under +P conditions were transferred to −P conditions for another 10 days. Total RNA was isolated from roots for RT-qPCR analysis with specific primers for IPS1, SPX1, PT10, Pre-miR827, SQD2, RNS4, SPX-MFS1, and PHO2. The gene expression level in WT/+P was set as 1. The RT-qPCR primers are listed in Supplemental Data Set S2. Error bars are sd (n = 3). Different letters above the bars indicate significant differences between groups. Statistics, one-way ANOVA with post-hoc Tukey’s test (P < 0.05).

Since RLI1 and PHR2 function redundantly based on physiological and morphological analysis, we examined the effects of the rli1 and phr2 mutations on Pi starvation signaling under −P conditions via qRT-PCR. The expression of PSI genes (IPS1, SPX1, PT10, miR827, SQD2, and RNS4) was obviously repressed in the phr2 mutant and further repressed in the rli1 phr2 double mutant (Figure 8D). In contrast, the expression levels of Pi starvation-repressed genes SPX-MFS1 and PHO2 were higher in the rli1 phr2 mutant than WT under −P conditions (Figure 8D).

Together, these results demonstrate that RLI1 and PHR2 coordinately regulate Pi-starvation signaling, Pi-homeostasis, and plant growth in response to Pi-deficiency stress.

AS of RLI1 homologous genes is widespread among dicots and monocots

Given the finding that AS of RLI1 can produce two functional protein isoforms to modulate plant growth, Pi-signaling, and Pi-homeostasis to enhance the adaptation of rice to low Pi stress, we asked whether this AS mechanism of RLI1 is functionally conserved among different plant species. To test this notion, we built a phylogenic tree of MYB-CC proteins from 20 representative land plants and found an RLI1-specific clade in angiosperms (Supplemental Figure S13 and Figure 9A). Structural analysis of genes in this clade showed that most genes were predicted to undergo similar types of AS (Figure 9B). To verify this notion, we designed specific primers for AT5G06800 (AtRLI1) in Arabidopsis, Glyma.03G143600 in soybean (Glycine max), and GRMZM2G035370 in maize (Zea mays) to amplify their specific transcripts. Similar to rice RLI1, two transcript isoforms encoding putative MYB and MYB-CC proteins were detected in Arabidopsis, soybean, and maize (Figure 9C). These results demonstrate that the AS of RLI1-like genes is conserved in both dicots and monocots.

Figure 9.

Figure 9

Open in a new tab

AS of RLI1 homologous genes is widespread among dicots and monocots. A, Phylogenetic analysis of RLI1 homologous proteins from different dicot and monocot species. The full-length protein sequences were used to build the tree. The supporting values are shown on the branches in the following order: SH-aLRT test/bootstrap value. B, Genomic structures of RLI1 homologous gene splicing variants and identification of primers. C, RT-PCR analysis of the splice variants of AT5G06800, Glyma.03G143600, and GRMZM2G035370 in Arabidopsis, soybean, and maize, respectively. The shoots and roots of 10-day-old plants grown under normal conditions were harvested for total RNA isolation and used for RT-PCR analysis with specific primers. MC indicates with the MYB and CC domain, M- indicates only with MYB domain. The AT1g49240 (AtACTIN8), Glyma.12G020500 (GmCons4), and ZmTUB4 genes were used as the internal control in Arabidopsis, soybean, and maize, respectively. D–F, AS of AT5G06800 (AtRLI1) regulates Pi accumulation and BR responses. Phenotypic performance assays of WT and overexpression lines of AT5G06800.1 (M-Ox) and AT5G06800.2 (MC-Ox) without and with BL treatment (D). Scale bars, 1 cm. The primary root length (E) and Pi concentrations of shoots (F) were measured.

To test whether the functions of the protein isoforms encoded by these transcripts are conserved in different species, we generated transgenic overexpression lines for AtRLI1a (with a MYB domain, AtRLI1a-Ox) and AtRLI1b (with MYB and CC domains, AtRLI1b-Ox; Figure 9D). Similar to their rice homologs, overexpressing AtRLI1a and AtRLI1b enhanced Pi accumulation in Arabidopsis (Figure 9F). Analysis of root growth responses showed that, like in rice, AtRLI1a-Ox was much more sensitive to BL treatment than the WT, whereas the BL sensitivity of AtRLI1b-Ox was comparable to that of the WT (Figure 9, D and E). In addition, AtRLI1b has similar dimerization properties to AtPHR1, as it can interact with AtPHR1 and itself, as revealed by a LCI assay in N. benthamiana leaf epidermal cells (Supplemental Figure S14). These results suggest that, like RLI1 and PHR2 in rice, AtRLI1 and AtPHR1 co-regulate Pi-signaling and Pi-homeostasis in Arabidopsis.

Together, these results suggest that the AS of RLI1 and its homologous genes to modulate plant growth and Pi signaling is a widespread mechanism in both dicots and monocots.

Discussion

Plants are continuously exposed to a wide range of abiotic stresses. Under Pi-deficiency stress, multiple strategies for morphological, physiological, and molecular adaptation will be activated in plants (Chiou and Lin, 2011; Wu et al., 2013). However, how these strategies are coordinated remains largely unknown. In this study, we determined that the AS of RLI1 is an important adaptative mechanism in plants that modulates Pi starvation signaling, Pi-homeostasis, and plant growth under Pi-deficiency stress.

AS can increase proteome diversity by generating multiple transcripts from a single gene and is thought to be an energy-saving strategy in eukaryotes. We found that external Pi-availability modulated the AS of RLI1 to affect the abundance of the two transcript variants. These variants encode two protein isoforms, with RLI1b having both MYB and CC domains and RLI1a only harboring a MYB domain. Pi-deficiency stress can reduce the amounts of RLI1a transcript, but without obvious effects on the RLI1b variant, suggesting that some unknown spliceosome components might be regulated by Pi-deficiency stress. Indeed, some RNA splicing regulators of Ser/Arg (SR) protein-coding genes are regulated by Pi-deficiency stress, and their loss-of-function results in Pi overaccumulation (Dong et al., 2018). In addition, we found that RLI1b protein levels significantly increased under low Pi stress, while RLI1a was repressed under these conditions (Figure 4), perhaps due to the differences in translation efficiency or protein stability of these isoforms. Therefore, the mRNA and protein levels of RLI1a and RLI1b are finely regulated by Pi availability.

Even though the two isoforms contain an identical DNA binding domain, the absence of the dimerization domain enables RLI1a to bind to broader targets in the rice genome. This might be explained by the following mechanisms: according to a DNA recognition model for GARP transcription factors, the dimerization and dissociation of GARP family proteins could affect their binding affinity to the target cis-element (Yanagisawa, 2013). The monomer form of RLI1a perfectly binds to all six core bases of the R1BS (ATATTC) motif to form a stable RLI1a-R1BS complex. In contrast, the formation of RLI1b dimers enables this protein to recognize the eight core bases of P1BS (GnATATnC) to form a stable RLI1b–P1BS–RLI1b complex, but with insufficient binding to the supporting bases in R1BS (Supplemental Figure S15). Furthermore, based on the finding that the positions of nucleosomes and their release of transcription factor binding motif regions shape the binding and regulatory landscape of transcription factors (Zhou and O’Shea, 2011), it is possible that the nucleosomes status of RLI1a target genes makes them more easily accessible for the smaller protein (RLI1a) but not the larger proteins (RLI1b or the RLI1b dimer).

We demonstrated that RLI1a directly regulates the expression of BL biosynthesis and signaling-related genes to affect the BL pathway in rice (Figure 7). BL has profound effects on plant growth (Nolan et al., 2020). Tong et al. (2014) showed that lower doses of BL treatment promoted root elongation, while higher doses of BL treatment inhibited root elongation and increased leaf inclination angle to modulate shoot architecture in rice. In addition, mutants with defects in BL biosynthesis show retarded growth and compact shoot architectures (Choe et al., 1998; Hong et al., 2003; Tanabe et al., 2005). Here, we found that the rli1 mutant had lower BL contents than the WT and showed retarded growth. Due to the defects in activating both BL biosynthesis and signaling, the rli1 mutant was less sensitive to BL treatment and hypersensitive to BRZ treatment compared with the WT based on its leaf inclination angles and root growth responses (Figure 7E and Supplemental Figure S9). Consistently, the RLI1a overexpression lines showed high BL contents and a large leaf inclination phenotype, which was partially suppressed by treatment with a BL biosynthesis inhibitor (Figure 7E and Supplemental Figure S9). These results indicate that Pi-deficiency stress inhibits RLI1a activity to fine-tune BL biosynthesis and signaling, thereby modulating plant growth and architecture to help the plant adapt to Pi-deficiency stress. BL signaling is involved in plant adaptation to Pi-deficiency stress. BRI1-EMS-SUPPRESSOR 1 (BZR1/BES1), which are central regulators of BL signaling, regulate root development under Pi-deficiency stress in Arabidopsis (Singh et al., 2014). Here, we showed that the rice gene BZR1 was a target of RLI1a (Figure 7, A and B). These findings demonstrate that Pi availability modifies the AS of RLI1 to regulate BL biosynthesis and signaling to affect plant growth.

We also demonstrated that the cooperation of RLI1 and PHR2 is essential for plants to overcome Pi-deficiency stress. Similar to PHR2, SPXs interact with both RLI1 isoforms to inhibit their transcriptional activity for Pi signaling and growth (Supplemental Figure S11; Ruan et al., 2018; Zhang et al., 2021). We further showed that RLI1 and PHR2 redundantly regulate Pi starvation signaling (Figure 8D). Considering the role of RLI1 in BL signaling and the impact of RLI1 and PHR2 on Pi starvation signaling, the rli1 phr2 double mutant displayed serious defects in growth and in Pi signaling and accumulation (Figure 8). The malfunction of both RLI1 and PHR2 even resulted in rapid leaf senescence upon long term Pi-deficiency stress (Figure 8C), perhaps due to the overactivation or inactivation of some important defense processes in the double mutant, similar to the case of the at-phr1 at-phl1 mutant in Arabidopsis (Castrillo et al., 2017). However, we cannot rule out the possibility that the accumulation of some other nutrition, such as nitrate, is defective in the mutant, since RLI1 (RLI1b) is known to play a significant role in regulating nitrate homeostasis in rice (Zhang et al., 2021). Therefore, RLI1 is an important regulator of both plant growth and Pi starvation signaling.

We further demonstrated that the AS mechanism of RLI1 is widely present in both dicots and monocots (Figure 9C). Indeed, a recent transcriptome study showed that the deficiency of the mineral nutrients Fe, Zn, Cu, Mn, and P affects AS events in rice (Dong et al., 2018). However, the specific physiological meaning and molecular mechanisms of the association of these nutrients with AS events remain unclear. Here, we showed that the AS of RLI1-related genes in Arabidopsis also created an MYB-only isoform, which modulates BL signaling and affects Pi accumulation, whilst the MYC-CC isoform plays a conserved role in regulating Pi accumulation (Figure 9). Together, these results suggest that the AS of RLI1 might be a crucial adaptation for plants to overcome Pi-deficiency stress, since a certain amount of RLI1 mRNA could generate two protein isoforms that play different roles under Pi-sufficient and Pi-deficient conditions. In fact, it is well known that AS is an important way to overcome biotic and abiotic stress (Rigo et al., 2019). AS of JASMONATE ZIM-DOMAIN PROTEIN 10 (JAZ10) and BES1 fine-tune the jasmonate and BR signaling pathways, respectively, during biotic stress in Arabidopsis (Chung and Howe, 2009; Jiang et al., 2015). The clock component CIRCADIAN CLOCK ASSOCIATED 1, a GARP transcription factor, coordinates circadian rhythms and cold acclimation via two functionally antagonistic transcript isoforms, with one containing both a MYB DNA binding domain and a dimerization domain and the other only containing a dimerization domain (Seo et al., 2012). Therefore, AS of transcription factor genes to create different isoforms with similar and broader target genes is also an important strategy for plants to adapt to fluctuating environmental conditions.

We propose a working model illustrating how AS of RLI1 and RLI1 protein stability cooperate with PHR2 to regulate Pi-signaling, Pi-homeostasis, and plant growth to enhance adaptation of Pi-deficiency stress (Figure 10). Under Pi-sufficient conditions, the AS of RLI1 produces comparable RLI1a and RLI1b transcripts. The lower protein stability of RLI1b and PHR2 under Pi-sufficient conditions (Figure 2;Guo et al., 2022), together with the higher protein stability of RLI1a and its lower affinity to the P1BS motif, lead to the moderate activation of PSI genes and promote BL biosynthesis and signaling. These processes help plants efficiently maintain suitable Pi and BR contents for growth under Pi-sufficient conditions. Under Pi-deficiency conditions, the suppression of RLI1 transcription (Figure 1A) and the AS of RLI1 help maintain RLI1b and reduce RLI1a transcript levels. The stabilization of RLI1b and PHR2 under Pi-deficient conditions leads to the activation of PSI genes to promote Pi uptake and utilization. In addition, the reduction in RLI1a transcript levels and lower RLI1a protein stability under Pi-deficiency stress attenuates the activation of BL biosynthesis and signaling genes, which modulates plant growth, including the induction of erect shoot architecture. This erect shoot architecture help plants reduce the consumption of Pi. Therefore, the AS of RLI1 produces two variants that cooperate with PHR2 to regulate Pi starvation signaling and growth to help plants adapt to Pi availability. Understanding the molecular and physiological mechanism mediated by the AS of RLI1 opens the possibility to breed smart crops with ideal shoot architecture and higher Pi use efficiency in the future.

Figure 10.

Figure 10

Open in a new tab

Working model of the role of RLI1 in regulating plant growth and Pi signaling to help plants adapt to Pi-deficiency. The thickness of the lines represents the degree of effects. The different sizes of protein diagrams represent protein levels. RLI1a targets downstream genes as a monomer. RLI1b and PHR2 target downstream genes as a dimer. A detailed description of the model can be found in the “Discussion.”

Materials and methods

Plant material and growth conditions

All rice (O.sativa) plants used in this study were derived from the Japonica variety NIP. The RLI1b overexpression lines RLI1b-Ox-1, RLI1b-Ox-9, and RLI1b-Ox-12 were obtained by transforming 35S-RLI1b-FLAG into WT NIP. The rli1 mutant and RLI1a overexpression lines (RLI1a-Ox-3, RLI1a-Ox-7, and RLI1a-Ox-9) were generated in our previous study (Ruan et al., 2018). The rli1 mutant was crossed with the phr2 T-DNA insertion mutant (Guo et al., 2015), and the rli1 phr2 double mutant was identified from an F2 population. WT Arabidopsis (Col-0), maize (B73), and soybean (Hongyuan) plants were used in this study. The overexpression lines AtRLI1a-Ox (lines 1 and 2) and AtRLI1b-Ox (lines 1 and 2) were obtained by transforming 35S-AtRLI1a-FLAG and 35S-AtRLI1b-FLAG into Arabidopsis. The Pi-sufficient (HP, 200 μM Pi), Pi-deficient (LP, 10 μM Pi), and Pi-starvation (−P, 0 μM Pi) treatments were performed as previously described (Ruan et al., 2019) using standard rice culture solution (Yoshida et al., 1976). The nutrient solution was adjusted to pH 5.5 using 1 M NaOH and replaced every 3 days during treatment. The experiments were performed in a greenhouse with a 12-h-day (30°C)/12-h-night (22°C) photoperiod, with 250 μmol m−2 s−1 photon densities and 60% relative humidity.

RNA extraction, RT-PCR, and qPCR

Total RNA was isolated from plant tissues using an RNA extraction kit (Promega, Madison, WI, USA) following the manufacturer’s instructions. The cDNA was synthesized with a Moloney Murine Leukemia Virus Reverse Transcriptase cDNA Synthesis Kit (Promega) according to the manufacturer’s instructions. Quantitative reverse transcription PCR (RT-qPCR) was performed as previously described (Zhou et al., 2008). The rice ACTIN gene was used as an internal control. Three biological replicates (different plants sampled at different times) were performed per gene.

Development of RLI1 polyclonal antibodies

Polyclonal rabbit RLI1 antibody was acquired using an N-terminal fragment of RLI1 (N-HDCHFGSPLCDPSPAPHLLSSA-COOH, positions 14–35) to immunize rabbits and purified before use. The immunization and purification processes were performed by the GenScript Company.

Development of RLI1b, AtRLI1a, and AtRLI1b overexpression lines

The 35S-RLI1b-FLAG, 35S-AtRLI1a-FLAG, and 35S-AtRLI1b-FLAG vectors were constructed by cloning the open reading frame of RLI1b, AtRLI1a, and AtRLI1b into the modified binary expression vector pF3PZPY122 after the cauliflower mosaic virus 35S promoter. The 35S-RLI1b-FLAG construct was transformed into mature embryos from NIP seeds via Agrobacterium tumefaciens (strain EHA105)-mediated transformation as described previously (Hiei et al., 1994). The vectors 35S-AtRLI1a-FLAG and 35S-AtRLI1b-FLAG were transformed into Arabidopsis by the floral dip method as described (Clough and Bent, 1998).

EMSA

To obtain the recombinant RLI1a-HIS and RLI1b-HIS proteins, the full-length coding sequences of RLI1a and RLI1b were amplified and cloned into the pET29b vector (Promega), respectively. RLI1a-HIS and RLI1b-HIS were expressed and purified from Escherichia coli (strain BL21(DE3); Novagen) according to the manufacturer’s instructions (Qiagen, Hilden, Germany). The biotin-labeled R1BS and P1BS probes were synthesized by BGI. The probe sequences are listed in Supplemental Data Set S2. The EMSA was performed using a Light Shift Chemiluminescent EMSA Kit (Thermo Fisher Scientific Waltham, MA, USA) according to the manufacturer’s instructions.

Gel blot analysis

The antibodies used for immunoblotting were anti-FLAG-HRP (M2, Sigma, St. Louis, MO, USA; 1:5,000), goat anti-rabbit-IgG-HRP (Sigma; 1:10,000), and goat anti-mouse-IgG-HRP (Sigma, 1:10,000). Total plant proteins were extracted using extraction buffer (containing 25 mM Tris–HCl [pH 7.5], 10 mM NaCl, 4 mM phenylmethylsulfonyl fluoride [PMSF], 20 mM MG132, and protease inhibitor cocktail). Proteins were visualized using an Immobilon kit (Millipore, Burlington, MA, USA) under standard conditions. To reduce nonspecific bands, nuclear proteins were used for immunoblot analysis to detect endogenous RLI1 proteins. Nuclear protein was isolated from the samples using Plant Nuclei Isolation/Extraction Kit (Sigma-Aldrich, St. Louis, MO, USA) following the procedure Semi-Pure Preparation of Nuclei.

To detect FLAG-tagged proteins RLI1a-FLAG and RLI1b-FLAG, total proteins were extracted from the samples and 20 µg proteins were loaded for each sample. After switching on the current (set to 100 V for 8 × 8 × 0.1 cm gel), the samples were electrophoresed until the bromophenol blue dye has migrated approximately to the bottom of the gel. The gels were blotted onto nitrocellulose membranes, blocked for 3 h with 5% skim milk, and incubated with anti-FLAG-HRP antibody for ∼10 h. The washed membranes were incubated with 5% skim milk containing anti-rabbit-IgG-HRP for 1.5 h. Finally, the washed membranes were incubated with HRP detecting solution for 5 min before imaging with a ChemiDoc Touch imaging system (Bio-Rad Image Lab).

ChIP-Seq and ChIP-qPCR analysis

ChIP-Seq analysis was performed as described previously (Tao et al., 2020). Briefly, ∼1–2 g shoot tissues were harvested from 15-day-old seedlings and used to isolate total nuclei with a Plant Nuclei Isolation/Extraction Kit (Sigma-Aldrich) following the procedure High-Pure Preparation of Nuclei. These nuclei were used for ChIP-Seq with a NovoNGS CUT&Tag High-Sensitivity Kit (for Illumina) (Novoprotein). The purified PCR products were subjected to Illumina sequencing or used for ChIP-qPCR analysis. The sequencing data were processed as described in Kaya-Okur et al. (2019). In brief, all generated reads from each rice individual were mapped against the rice reference genome (IRGSP1.0) using Bowtie2 (version 2.1.0) (Langmead and Salzberg, 2012). All peaks were called using MACS3 (Zhang et al., 2008) with default parameters. KEGG analysis was performed using KEGG API. Significantly enriched KEGG items were filtered by P-value < 0.01 and FDR < 0.05. All diagrams were generated using R scripts (available upon request).

Measurement of leaf inclination angle

Leaf inclination angle was measured as described previously (Ruan et al., 2018).

LUC activity assay

For the transient transcriptional activation assays, a firefly LUC reporter driven by the D11 promoter (∼2 kb) or four tandem copies of R1BS or P1BS fused to the minimal 35S promoter from CaMV (4×R1BS-min35S or 4×P1BS-min35S) were used to test the differences in DNA binding ability of RLI1a and RLI1b. These promoters were cloned into pGreen-0800-LUC to generate the D11-Pro:LUC, 4×R1BS-min35S:LUC, and 4×P1BS-min35S:LUC constructs. The Renilla LUC gene under the control of the CaMV 35S promoter in the pGreen-0800-LUC vector was used as an internal control. pSoup and the reporter plasmids were co-transformed into A.tumefaciens (EHA105). Different combinations of 35S:RLI1a, 35S:RLI1b, 4×R1BS-min35S:LUC, 4×P1BS-min35S:LUC, D11-pro:LUC were co-transformed into N. benthamiana leaf epidermal cells by A.tumefaciens-mediated transformation. After two days, 10 μM luciferin was injected into the leaves 5 min before observation and imaging. Leaves con-injected with pF3PZPY122 and 4×R1BS-min35S:LUC or 4×P1BS-min35S:LUC were used as the negative control. To measure LUC activity, the dual-LUC transient transactivation assay system was employed according to the manufacturer’s protocol (Dual-LUC Reporter Assay System; Promega).

LCI assay

The LCI assay for detecting protein interactions was performed in N. benthamiana leaves as described previously (Ruan et al., 2019). In brief, full-length SPX1, SPX2, PHR2, RLI1a, and RLI1b were separately fused with the N- and C-terminal parts of the LUC reporter gene LUC (LUCN and LUCC), generating the vectors SPX1-LUCN, SPX2-LUCN, PHR2-LUCN, RLI1a-LUCN, RLI1b-LUCN SPX1-LUCC, SPX2-LUCC, PHR2-LUCC, RLI1a-LUCC, and RLI1b-LUCC. The vectors were co-infiltrated into N. benthamiana leaves, and LUC activity was analyzed 48 h after infiltration using NightSHADE LB 985 (Berthold). In total, three biological replicates were performed with similar results.

Cell-free degradation assay

The cell-free degradation assays were performed as previously described (Ruan et al., 2019). In brief, the shoots of WT seedlings grown under +P and −P conditions were harvested and ground in liquid nitrogen for total protein extraction with degradation buffer (25 Mm Tris–HCl [pH 7.5], 10 mM NaCl, 10 mM MgCl2, 4 mM PMSF, 5 mM dithiothreitol, and 10 mM ATP). The total proteins were adjusted to equal concentrations in degradation buffer for each assay. The purified RLI1a-HIS and RLI1b-HIS proteins were incubated with the +P and −P protein extracts at 28°C for 0, 10, 20, 40, and 60 min. The HIS-tagged proteins were detected by HIS-antibody (TransGenBiotech) diluted 1:5,000.

Y1H assay

To determine the DNA binding affinity of RLI1a and RLI1b with the R1BS and P1BS motifs, artificial promoters containing four tandem copies of R1BS or P1BS were constructed in the pLAZ2u (Clontech, Mountain View, CA, USA) vector backbone to generate reporter gene constructs, and the coding sequence of RLI1a or RLI1b was cloned into pB42AD to generate the activator constructs. Yeast strain EGY48 was transformed with the activator and reporter gene constructs and grown on synthetic medium lacking urea and tryptophan (Clontech). The β-galactosidase activity was measured as previously reported (Ruan et al., 2015).

Preparation of stem sections and cell length measurements

The preparation of sections and cell length measurement were conducted as described previously (Ruan et al., 2018). Briefly, lamina joint segments were embedded in 3% agar and sectioned with a vibratome (Leica VT 1000 S). The images of lamina joint autofluorescence were taken under a microscope (Leica DM6000M). The cell lengths were measured using microscope analysis software (Leica DM6000M).

Measurement of Pi levels

Measurements of Pi concentrations in plants were performed as described previously (Zhou et al., 2008). Briefly, freshly harvested plant tissues were ground in liquid nitrogen, and ∼0.4 g tissue powder was collected. The Pi was extracted from the samples by incubation in 1 mL 5 M H2SO4 for 1 h before centrifugation (12,000 rpm/10 min). In total, 0.2 mL supernatant was used to measure Pi via the molybdenum blue method (Murphy and Riley, 1962).

Measurement of BL levels

BL contents were detected by MetWare (http://www.metware.cn/) using the AB Sciex QTRAP 6500 LC–MS/MS platform. Briefly, ∼5 g shoot tissues were harvested from 21-day-old WT, RLI1a-Ox, RLI1b-Ox, and rli1 plants, frozen in liquid nitrogen, and stored at −80°C. The samples (0.2 g fresh weight) were ground into a powder in liquid nitrogen and extracted with 4 mL acetonitrile. Internal standards were added to the plant samples before extraction. The supernatants were collected after centrifugation, and 800 µL of 4-(N, N-dimethylamino) phenylboronic acid (4-DMABPA) was added to the resulting solutions. The reaction solution was vortexed, incubated at 75°C for 1 h, and evaporated to dryness under a nitrogen gas stream. The sample was redissolving in 400 µL acetonitrile and filtered through a 0.22 μm filter for LC–MS analysis.

Identification of RLI1 proteins and phylogenetic analysis

Genome and gene annotation data for all 20 selected plant species were collected from the Phytozome database (Goodstein et al., 2012), NCBI Assembly, and PLAZA (Van Bel et al., 2018). The RLI1 proteins were identified using a similar method to that used in our previous study (Wang et al., 2021). In brief, all protein sequences were identified by InterProScan version 5 (Jones et al., 2014), and sequences with C-terminal MYB and CC domains were marked as candidate RLI1 proteins. The candidate RLI1 proteins were validated by BLASTP with an e-value cutoff at 1e-10. To construct a phylogenic tree of all RLI1 proteins from selected plant species, the full-length RLI1 protein sequences were aligned using MAFFT version 7.429 (Katoh and Standley, 2013). The maximum likelihood gene phylogeny was reconstructed by IQ-TREE2 (Minh et al., 2020) with JTT + F + I + G4 as the best-fitting model. The ultrafast bootstrap method with 1,000 replicates was conducted in IQ-TREE2 to obtain the supporting values for each internal node of the tree. Finally, the genetic trees were visualized and colored using iTOL (Letunic and Bork, 2021). The sequence alignments and phylogenetic trees files have been submitted to Figshare (https://doi.org/10.6084/m9.figshare.19778509).

Statistical analysis

To determine significant differences among groups, a Student’s two-tailed test and one-way ANOVA with post-hoc Tukey’s test were used for all experiments. The ANOVA and T test tables are provided in Supplemental Files S1 and S2.

Accession numbers

Sequence data from this article can be found at Phytozome (https://phytozome-next.jgi.doe.gov/) and the Rice Genome Annotation Project (http://rice.uga.edu/) under the following accession numbers: RLI1 (Os04g56990), SPX1 (Os06g40120), SPX2 (Os02g10780), BU1 (Os06g12210), D11 (Os04g39430), DWF4 (Os03g12660), CYP90D3 (Os01g10040), P450 related (Os11g04710), GSK3 (Os06g06090), BZR1 (Os07g39220), IPS1 (Os03g05334), SPX1 (Os06g40120), SPX2 (Os02g10780), SPX3 (Os06g03860), SQD2 (Os03g15840), VPE1 (Os04g46880), VPE2 (Os08g06010), PT1 (Os03g05620), and ACTIN (Os03g50885). miR827 (MI0010490), PHO2 (Os05g48390), PT10 (Os06g21950), RNS4 (Os09g36680), SPX-MFS1 (Os04g48390), AT1G49240 (AT-ACTIN8), Glyma.12G020500 (Cons4), AT5G06800 (AtRLI1), Glyma.03G143600 (GmRLI1), and GRMZM2G035370 (ZmRLI1).

Data from this article can be found in the miRNA Database under accession number MI0010490 (OsmiR827). ChIP-Seq data from this article can be found in the GenBank/EMBL data libraries under BioSample accession number of SAMN22799474, SAMN22799475, SAMN22799476, SAMN22799477, SAMN22799478, SAMN22799479, SAMN22799480, and SAMN22799481. The sequence alignments and phylogenetic trees files have been submitted to Figshare (https://doi.org/10.6084/m9.figshare.19778509).

Supplemental data

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

Supplemental Figure S1. Amino acid sequence alignment of RLI1a and RLI1b.

Supplemental Figure S2. RLI1b shares similar domains with PHR2 and AtPHR1.

Supplemental Figure S3. Identification of the RLI1a and RLI1b overexpression lines.

Supplemental Figure S4. Phenotypic analysis of RLI1a-Ox and RLI1b-Ox plants under low Pi conditions.

Supplemental Figure S5. Leaf inclination analysis of WT, rli1, RLI1a-Ox, and RLI1b-Ox under +P and −P conditions.

Supplemental Figure S6. RLI1b interacts with PHR2 and with itself but not with RLI1a.

Supplemental Figure S7. Count of peaks in the TSS regions.

Supplemental Figure S8. ChIP-qPCR analysis of the enrichment levels of RLI1a-FLAG and RLI1b-FLAG on the IPS1, SPX1, and VPE2 promoters.

Supplemental Figure S9. Genome browser views of RLI1a-FLAG and RLI1b-FLAG occupancy at Pi- and BL-related genes.

Supplemental Figure S10. RLl1a overexpression lines are more sensitive to exogenous eBL-24 treatment than the WT.

Supplemental Figure S11. RLI1a and RLI1b interact with SPX1 and SPX2, as does PHR2.

Supplemental Figure S12. Identification of the rli1 phr2 double mutant.

Supplemental Figure S13. Phylogram of MYB-CC proteins from different species.

Supplemental Figure S14. AtRLI1b interacts with AtPHR1.

Supplemental Figure S15. The predicated DNA binding forms of RLI1a and RLI1b with R1BS and P1BS motifs.

Supplemental Data Set S1. KEGG analysis of RLI1a and RLI1b target genes.

Supplemental Data Set S2 . Primers used in this study.

Supplemental File S1. ANOVA tables.

Supplemental File S2. T test tables.

Supplementary Material

koac161_Supplementary_Data
Click here for additional data file. (2.3MB, zip)

Acknowledgments

We thank Professor Danhua-Jiang kindly for help with Arabidopsis transformation. We also thank Professors Qingyu-Wu and Zhe-Yan for providing the maize (B73) and soybean (Hongyuan) seeds, respectively.

Funding

This work was funded by the National Key Research and Development Program of China (No. 2021YFF1000402) and the National Natural Science Foundation of China (Nos. 32130096, 31972493 and 31772386), and K.Y. and W.R. were supported by the Innovation Program of Chinese Academy of Agricultural Sciences. M.G. and X.J. were supported by the China Postdoctoral Science Foundation (Nos. 2019M650913 and 2021M693447).

Conflict of interest statement. None declared.

Contributor Information

Meina Guo, Key Laboratory of Plant Nutrition and Fertilizer, Ministry of Agriculture, Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 10081, China.

Yuxin Zhang, Key Laboratory of Plant Nutrition and Fertilizer, Ministry of Agriculture, Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 10081, China.

Xianqing Jia, Key Laboratory of Plant Nutrition and Fertilizer, Ministry of Agriculture, Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 10081, China.

Xueqing Wang, Key Laboratory of Plant Nutrition and Fertilizer, Ministry of Agriculture, Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 10081, China.

Yibo Zhang, Key Laboratory of Plant Nutrition and Fertilizer, Ministry of Agriculture, Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 10081, China.

Jifeng Liu, Hebei Wotu Seed Co. Ltd., Handan 057550, China.

Qingshen Yang, Hebei Wotu Seed Co. Ltd., Handan 057550, China.

Wenyuan Ruan, Key Laboratory of Plant Nutrition and Fertilizer, Ministry of Agriculture, Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 10081, China.

Keke Yi, Key Laboratory of Plant Nutrition and Fertilizer, Ministry of Agriculture, Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 10081, China.

 

M.G., Y.Z., and W.R. performed most of the described experiments. Y.Z. and X.J. performed the ChIP-Seq and phylogenetic analysis. K.Y. and W.R. designed the research. X.W. and YB.Z. constructed the vectors and purified the proteins. M.G., W.R., Y.Z., X.J., X.W., YB.Z., J.L., Q.Y., and K.Y. analyzed the data. K.Y., W.R., and M.G. wrote the manuscript. All authors reviewed 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/plcell) is: Keke Yi (yikeke@gmail.com/yikeke@caas.cn).

References

  1. Bustos R, Castrillo G, Linhares F, Puga MI, Rubio V, Pérez-Pérez J, Solano R, Leyva A, Paz-Ares J (2010) A central regulatory system largely controls transcriptional activation and repression responses to phosphate starvation in Arabidopsis. PLoS Genet 6: e1001102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Castrillo G, Teixeira PJ, Paredes SH, Law TF, de Lorenzo L, Feltcher ME, Finkel OM, Breakfield NW, Mieczkowski P, Jones CD, et al. (2017) Root microbiota drive direct integration of phosphate stress and immunity. Nature 543: 513–518 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Chiou TJ, Lin SI (2011) Signaling network in sensing phosphate availability in plants. Annu Rev Plant Biol 62: 185–206 [DOI] [PubMed] [Google Scholar]
  4. Choe S, Dilkes BP, Fujioka S, Takatsuto S, Sakurai A, Feldmann KA (1998) The DWF4 gene of Arabidopsis encodes a cytochrome P450 that mediates multiple 22 alpha-hydroxylation steps in brassinosteroid biosynthesis. Plant Cell 10: 231–243 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Chung HS, Howe GA (2009) A critical role for the TIFY motif in repression of jasmonate signaling by a stabilized splice variant of the JASMONATE ZIM-domain protein JAZ10 in Arabidopsis. Plant Cell 21: 131–145 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Clough SJ, Bent AF (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J 16: 735–743 [DOI] [PubMed] [Google Scholar]
  7. Dietz K, Foyer C (1986) The relationship between phosphate status and photosynthesis in leaves; reversibility of the effects of phosphate deficiency on photosynthesis. Planta 167: 376–381 [DOI] [PubMed] [Google Scholar]
  8. Dong C, He F, Berkowitz O, Liu J, Cao P, Tang M, Shi H, Wang W, Li Q, Shen Z, et al. (2018) Alternative splicing plays a critical role in maintaining mineral nutrient homeostasis in rice (Oryza sativa). Plant Cell 30: 2267–2285 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Franco-Zorrilla JM, González E, Bustos R, Linhares F, Leyva A, Paz-Ares J (2004) The transcriptional control of plant responses to phosphate limitation. J Exp Bot 55: 285–293 [DOI] [PubMed] [Google Scholar]
  10. Goodstein DM, Shu S, Howson R, Neupane R, Hayes RD, Fazo J, Mitros T, Dirks W, Hellsten U, Putnam N, et al. (2012) Phytozome: a comparative platform for green plant genomics. Nucleic Acids Res 40: D1178–D1186 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Guo M, Ruan W, Zhang Y, Zhang Y, Wang X, Guo Z, Wang L, Zhou T, Paz-Ares J, Yi K (2022) A reciprocal inhibitory module for Pi and iron signaling. Mol Plant 15: 138–150 [DOI] [PubMed] [Google Scholar]
  12. Guo M, Ruan W, Li C, Huang F, Zeng M, Liu Y, Yu Y, Ding X, Wu Y, Wu Z, et al. (2015) Integrative comparison of the role of the PHOSPHATE RESPONSE1 subfamily in phosphate signaling and homeostasis in rice. Plant Physiol 168: 1762–1776 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hiei Y, Otis S, Komari T, Kumashiro T (1994) Efficient transformation of rice (Oryza sativa L.) mediated by Agrobacterium and sequence analysis of the boundaries of the T-DNA. Plant J 6: 271–282 [DOI] [PubMed] [Google Scholar]
  14. Holford ICR (1997) Soil phosphorus: its measurement, and itsuptake by plants. Aust J Soil Res 35: 227–239 [Google Scholar]
  15. Hong Z, Ueguchi-Tanaka M, Umemura K, Uozu S, Fujioka S, Takatsuto S, Yoshida S, Ashikari M, Kitano H, Matsuoka M (2003) A rice brassinosteroid-deficient mutant, ebisu dwarf (d2), is caused by a loss of function of a new member of cytochrome P450. Plant Cell 15: 2900–2910 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Jiang J, Zhang C, Wang X (2015) A recently evolved isoform of the transcription factor BES1 promotes brassinosteroid signaling and development in Arabidopsis thaliana. Plant Cell 27: 361–374 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Jones P, Binns D, Chang HY, Fraser M, Li W, McAnulla C, McWilliam H, Maslen J, Mitchell A, Nuka G, et al. (2014) InterProScan 5: genome-scale protein function classification. Bioinformatics 30: 1236–1240 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Katoh K, Standley DM (2013) MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol Biol Evol 30: 772–780 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kaya-Okur HS, Wu SJ, Codomo CA, Pledger ES, Bryson TD, Henikoff JG, Ahmad K, Henikoff S (2019) CUT&Tag for efficient epigenomic profiling of small samples and single cells. Nat Commun 10: 1930. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Langmead B, Salzberg SL (2012) Fast gapped-read alignment with Bowtie 2. Nat Methods 9: 357–359 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Letunic I, Bork P (2021) Interactive tree of life (iTOL) v5: an online tool for phylogenetic tree display and annotation. Nucleic Acids Res 49: W293–W296 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. López-Bucio J, Cruz-Ramírez A, Herrera-Estrella L (2003) The role of nutrient availability in regulating root architecture. Curr Opin Plant Biol 6: 280–287 [DOI] [PubMed] [Google Scholar]
  23. López-Bucio J, Hernández-Abreu E, Sánchez-Calderón L, Nieto-Jacobo MF, Simpson J, Herrera-Estrella L (2002) Phosphate availability alters architecture and causes changes in hormone sensitivity in the Arabidopsis root system. Plant Physiol 129: 244–256 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lynch JP, Brown KM (2001) Topsoil foraging-an architectural adaptation of plants to low phosphorus availability. Plant Soil 237: 225–237 [Google Scholar]
  25. Minh BQ, Schmidt HA, Chernomor O, Schrempf D, Woodhams MD, von Haeseler A, Lanfear R (2020) IQ-TREE 2: new models and efficient methods for phylogenetic inference in the genomic era. Mol Biol Evol 37: 1530–1534 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Murphy J, Riley JP (1962) A modified single solution method for determination of phosphate in natural waters. Anal Chim Acta 26: 31–36 [Google Scholar]
  27. Natr L (1992) Mineral nutrients—a ubiquitous stress factor for photosynthesis. Photosynthetica 27: 271–294 [Google Scholar]
  28. Nolan TM, Vukašinović N, Liu D, Russinova E, Yin Y (2020) Brassinosteroids: multidimensional regulators of plant growth, development, and stress responses. Plant Cell 32: 295–318 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Péret B, Clément M, Nussaume L, Desnos T (2011) Root developmental adaptation to phosphate starvation: better safe than sorry. Trends Plant Sci 16: 442–450 [DOI] [PubMed] [Google Scholar]
  30. Puga MI, Mateos I, Charukesi R, Wang Z, Franco-Zorrilla JM, de Lorenzo L, Irigoyen ML, Masiero S, Bustos R, Rodríguez J, et al. (2014) SPX1 is a phosphate-dependent inhibitor of Phosphate Starvation Response 1 in Arabidopsis. Proc Natl Acad Sci USA 111: 14947–14952 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Puga MI, Rojas-Triana M, de Lorenzo L, Leyva A, Rubio V, Paz-Ares J (2017) Novel signals in the regulation of Pi starvation responses in plants, facts and promises. Curr Opin Plant Biol 39: 40–49 [DOI] [PubMed] [Google Scholar]
  32. Rigo R, Bazin J, Crespi M, Charon CL (2019) Alternative splicing in the regulation of plant-microbe interactions. Plant Cell Physiol 60: 1906–1916 [DOI] [PubMed] [Google Scholar]
  33. Ruan W, Guo M, Cai L, Hu H, Li C, Liu Y, Wu Z, Mao C, Yi K, Wu P, Mo X (2015) Genetic manipulation of a high-affinity PHR1 target cis-element to improve phosphorous uptake in Oryza sativa L. Plant Mol Biol 87: 429–440 [DOI] [PubMed] [Google Scholar]
  34. Ruan W, Guo M, Wang X, Guo Z, Xu Z, Xu L, Zhao H, Sun H, Yan C, Yi K (2019) Two RING-finger ubiquitin E3 ligases regulate the degradation of SPX4, an internal phosphate sensor, for phosphate homeostasis and signaling in rice. Mol Plant 12: 1060–1074 [DOI] [PubMed] [Google Scholar]
  35. Ruan W, Guo M, Wu P, Yi K (2017) Phosphate starvation induced OsPHR4 mediates Pi-signaling and homeostasis in rice. Plant Mol Biol 93: 327–340 [DOI] [PubMed] [Google Scholar]
  36. Ruan W, Guo M, Xu L, Wang X, Zhao H, Wang J, Yi K (2018) An SPX-RLI1 module regulates leaf inclination in response to phosphate availability in rice. Plant Cell 30: 853–870 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Rubio V, Linhares F, Solano R, Martín AC, Iglesias J, Leyva A, Paz-Ares J (2001) A conserved MYB transcription factor involved in phosphate starvation signaling both in vascular plants and in unicellular algae. Genes Dev 15: 2122–2133 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Schachtman DP, Reid RJ, Ayling SM (1998) Phosphorus uptake by plants: from soil to cell .Plant Physiol 116: 447–453 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Secco D, Jabnoune M, Walker H, Shou H, Wu P, Poirier Y, Whelan J (2013) Spatio-temporal transcript profiling of rice roots and shoots in response to phosphate starvation and recovery. Plant Cell 25: 4285–4304 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Seo PJ, Park MJ, Lim MH, Kim SG, Lee M, Baldwin IT, Park CM (2012) A self-regulatory circuit of CIRCADIAN CLOCK-ASSOCIATED1 underlies the circadian clock regulation of temperature responses in Arabidopsis. Plant Cell 24: 2427–2442 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Singh AP, Fridman Y, Friedlander-Shani L, Tarkowska D, Strnad M, Savaldi-Goldstein S (2014) Activity of the brassinosteroid transcription factors BRASSINAZOLE RESISTANT1 and BRASSINOSTEROID INSENSITIVE1-ETHYL METHANESULFONATE- SUPPRESSOR1/BRASSINAZOLE RESISTANT2 blocks developmental reprogramming in response to low phosphate availability. Plant Physiol 166: 678–688 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Sun L, Song L, Zhang Y, Zheng Z, Liu D (2016) Arabidopsis PHL2 and PHR1 act redundantly as the key components of the central regulatory system controlling transcriptional responses to phosphate starvation. Plant Physiol 170: 499–514 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Tanabe S, Ashikari M, Fujioka S, Takatsuto S, Yoshida S, Yano M, Yoshimura A, Kitano H, Matsuoka M, Fujisawa Y, et al. (2005) A novel cytochrome P450 is implicated in brassinosteroid biosynthesis via the characterization of a rice dwarf mutant, dwarf11, with reduced seed length. Plant Cell 17: 776–790 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Tao X, Feng S, Zhao T, Guan X (2020) Efficient chromatin profiling of H3K4me3 modification in cotton using CUT&Tag. Plant Methods 16: 120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Tong H, Xiao Y, Liu D, Gao S, Liu L, Yin Y, Jin Y, Qian Q, Chu C (2014) Brassinosteroid regulates cell elongation by modulating gibberellin metabolism in rice. Plant Cell 26: 4376–4393 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Van Bel M, Diels T, Vancaester E, Kreft L, Botzki A, Van de Peer Y, Coppens F, Vandepoele K (2018) PLAZA 4.0: an integrative resource for functional, evolutionary and comparative plant genomics. Nucleic Acids Res 46: D1190–D1196 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Vance CP, Uhde-Stone C, Allan DL (2003) Phosphorus acquisition and use: critical adaptations by plants for securing a nonrenewable resource. New Phytol 157: 423–447 [DOI] [PubMed] [Google Scholar]
  48. Wang L, Jia X, Zhang Y, Xu L, Menand B, Zhao H, Zeng H, Dolan L, Zhu Y, Yi K (2021) Loss of two families of SPX domain-containing proteins required for vacuolar polyphosphate accumulation coincides with the transition to phosphate storage in green plants. Mol Plant 14: 838–846 [DOI] [PubMed] [Google Scholar]
  49. Wang Z, Ruan W, Shi J, Zhang L, Xiang D, Yang C, Li C, Wu Z, Liu Y, Yu Y, et al. (2014) Rice SPX1 and SPX2 inhibit phosphate starvation responses through interacting with PHR2 in a phosphate-dependent manner. Proc Natl Acad Sci USA 111: 14953–14958 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Williamson LC, Ribrioux SP, Fitter AH, Leyser HM (2001) Phosphate availability regulates root system architecture in Arabidopsis. Plant Physiol 126: 875–882 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Wu P, Shou H, Xu G, Lian X (2013) Improvement of phosphorus efficiency in rice on the basis of understanding phosphate signaling and homeostasis. Curr Opin Plant Biol 16: 205–212 [DOI] [PubMed] [Google Scholar]
  52. Xu L, Wang F, Li R, Deng M, Fu M, Teng H, Yi K (2020) OsCYCP4s coordinate phosphate starvation signaling with cell cycle progression in rice. J Integr Plant Biol 62: 1017–1033 [DOI] [PubMed] [Google Scholar]
  53. Yanagisawa S (2013) Characterization of a nitrate-inducible transcriptional repressor NIGT1 provides new insights into DNA recognition by the GARP family proteins. Plant Signal Behav 8: e24447. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Yoshida S, Forno DA, Cock JH, Gomez KA (1976) Laboratory Manual for Physiological Studies of Rice, Ed 3. International Rice Research Institute, Manila, The Philippines [Google Scholar]
  55. Zhang Z, Li Z, Wang W, Jiang Z, Guo L, Wang X, Qian Y, Huang X, Liu Y, Liu X, et al. (2021) Modulation of nitrate-induced phosphate response by the MYB transcription factor RLI1/HINGE1 in the nucleus. Mol Plant 14: 517–529 [DOI] [PubMed] [Google Scholar]
  56. Zhang Y, Liu T, Meyer CA, Eeckhoute J, Johnson DS, Bernstein BE, Nusbaum C, Myers RM, Brown M, Li W, et al. (2008) Model-based analysis of ChIP-Seq (MACS). Genome Biol 9: R137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Zhou J, Jiao F, Wu Z, Li Y, Wang X, He X, Zhong W, Wu P (2008) OsPHR2 is involved in phosphate-starvation signaling and excessive phosphate accumulation in shoots of plants. Plant Physiol 146: 1673–1686 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Zhou X, O’Shea EK (2011) Integrated approaches reveal determinants of genome-wide binding and function of the transcription factor Pho4. Mol Cell 42: 826–836 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

koac161_Supplementary_Data
Click here for additional data file. (2.3MB, zip)

Articles from The Plant Cell are provided here courtesy of Oxford University Press

RESOURCES