The APC/C^TAD1-WIDE LEAF 1-NARROW LEAF 1 pathway controls leaf width in rice

Abstract

Leaf morphology is one of the most important features of the ideal plant architecture. However, the genetic and molecular mechanisms controlling this feature in crops remain largely unknown. Here, we characterized the rice (Oryza sativa) wide leaf 1 (wl1) mutant, which has wider leaves than the wild-type due to more vascular bundles and greater distance between small vascular bundles. WL1 encodes a Cys-2/His-2-type zinc finger protein that interacts with Tillering and Dwarf 1 (TAD1), a co-activator of the anaphase-promoting complex/cyclosome (APC/C) (a multi-subunit E3 ligase). The APC/C^TAD1 complex degrades WL1 via the ubiquitin-26S proteasome degradation pathway. Loss-of-function of TAD1 resulted in plants with narrow leaves due to reduced vascular bundle numbers and distance between the small vascular bundles. Interestingly, we found that WL1 negatively regulated the expression of a narrow leaf gene, NARROW LEAF 1 (NAL1), by recruiting the co-repressor TOPLESS-RELATED PROTEIN and directly binding to the NAL1 regulatory region to inhibit its expression by reducing the chromatin histone acetylation. Furthermore, biochemical and genetic analyses revealed that TAD1, WL1, and NAL1 operated in a common pathway to control the leaf width. Our study establishes an important framework for understanding the APC/C^TAD1–WL1–NAL1 pathway-mediated control of leaf width in rice, and provides insights for improving crop plant architecture.

The APC/C^TAD1–WL1–NAL1 pathway-mediated control of leaf width in rice is crucial for regulating the leaf architecture in crop plants.

Introduction

Since leaves are important for photosynthesis, leaf width is a critical plant architectural component. Appropriate leaf width has great biological significance in terms of improving the absorption and conversion efficiency of light energy (Ort et al., 2015). Therefore, in-depth studies on the regulatory mechanisms of leaf width are not only important theoretically, but also practically for improving rice (Oryza sativa) yield. Leaves develop from the shoot apical meristem (SAM), which contains specialized rapidly dividing cells on its periphery that develop into leaf primordia first, then alter their plane of division and establish a 3D axis along the proximal/distal, adaxial/abaxial, and medial/lateral axes (Fleming, 2005; Moon and Hake, 2011). Auxin helps regulate the leaf width, via two narrow leaf (NAL) genes, NAL1 and NAL7, which influence the auxin polar transport and synthesis, respectively. NAL1 encodes a plant-specific protein with largely unknown biochemical functions. Mutation of NAL1 significantly reduced the polar auxin transport capacity, thereby altering the distribution pattern of vascular tissues and reducing the leaf width (Qi et al., 2008; Chen et al., 2012). In contrast, NAL7 encodes a flavin-containing monooxygenase, which shares sequence homology with YUCCA and is involved in auxin biosynthesis. The nal7 mutant showed reduced auxin content and a NAL phenotype (Fujino et al., 2008). Leaf width in rice is also closely related to leaf vascular development. NAL2 and NAL3 are paralogs that encode the identical OsWOX3A (WUSCHEL-related homeobox 3A) transcriptional activators. The significantly reduced number of small veins between the adjacent large veins in the nal2 nal3 mutant, along with the abnormal arrangement of xylem and phloem, resulted in narrow leaves. The phenotypes of these mutants indicate that NAL2 and NAL3 significantly regulate vascular bundle development (Cho et al., 2013; Ishiwata et al., 2013). Narrow and rolled leaf 1 (NRL1) encodes the cellulose synthase-like protein D4 (OsCSLD4). Mutation of NRL1 reduced the number of longitudinal veins in leaves, thereby producing phenotypes with reduced leaf width (Li et al., 2009; Yoshikawa et al., 2013). Additionally, leaf width is related to polarity. SHALLOT-LIKE1 (SLL1) encodes a SHAQKYF class MYB transcription factor belonging to the KAN family. Mutation of SLL1 resulted in narrow and abaxialized leaves (Zhang et al., 2009). SEMI-ROLLED LEAF 2 (SRL2) encodes a newly identified plant-specific protein of unknown biochemical function, with the srl2 mutant showing narrow incurved leaves due to the defective development of the abaxial sclerenchymatous cells (Liu et al., 2016; Ma et al., 2017).

The Cys-2/His-2-type (C2H2) zinc finger protein family is an important class of eukaryotic transcription factors implicated in diverse cellular processes involving plant development and plant stress response (Sun et al., 2010). DROUGHT AND SALT TOLERANCE (DST) encodes a novel C2H2 zinc finger transcription factor that is important in the plant stress response. DST negatively regulates stomatal closure by directly modulating the genes related to H₂O₂ homeostasis via a novel signal transduction pathway. Loss-of-DST function promotes stomatal closure and reduces stomatal density, resulting in enhanced drought and salt tolerance in rice (Huang et al., 2009). DST also regulates drought and salt tolerance by interacting with the co-activator DST Co-activator 1 to regulate the expression of PEROXIDASE 24 PRECURSOR, which encodes an H₂O₂ scavenger (Cui et al., 2015). DST can also regulate the expression of a leucine-rich repeat-RLK gene named LEAF PANICLE 2 under drought stress (Wu et al., 2015). Additionally, DST is important in the biological development, where it positively and directly regulates the expression of CYTOKININ OXIDASE 2 (OsCKX2), thereby causing increased cytokinin (CK) accumulation in the SAM and enhanced seed production. A DST mutant showed an increased grain number per panicle (Li et al., 2013). Further studies showed that DST is phosphorylated upon interacting with OsMPK6, thereby enhancing its capacity to activate the transcription of OsCKX2 (Guo et al., 2020). Generally, DST plays a dual role in regulating stress and developmental responses. However, how DST helps regulate leaf width is largely unknown.

In this study, we isolated a wide leaf 1 (wl1) mutant with an increased leaf width. We report that the wl1 phenotype is due to a novel allelic mutation in DST. Although the drought and salt tolerance function of WL1/DST in rice are well-reported (Huang et al., 2009; Li et al., 2013; Cui et al., 2015), how the encoded protein controls leaf width has not been studied in detail. Our study establishes an important genetic and molecular framework for elucidating the anaphase-promoting complex/cyclosome (APC/C)^TAD1–WL1–NAL1 pathway-mediated control of leaf width in rice, thereby suggesting that it is crucial for regulating the leaf architecture in crop plants.

Results

Phenotypic identification of the wl1 mutant

We isolated the rice mutant wl1 from an ethyl methane sulfonate (EMS)-mutagenized population of the O. sativa subsp. indica cultivar Xinong 1B. Compared with the wild-type (WT) Xinong 1B, we found that the most remarkable feature of the wl1 mutant was its increased leaf width throughout the growth phase (Figure 1, A–F), after measuring both the leaf width and length of the WT and wl1 at the same time points during the growth period. The leaves of wl1 plants were significantly wider than those of the WT throughout the growth period (Figure 1G; Supplemental Table S1). However, the wl1 leaf length did not vary significantly from that of the WT (Figure 1H; Supplemental Table S1).

Figure 1.

Phenotypic characteristics of the wl1 mutant. A–F, Phenotypes of the WT (left) and wl1 (right) at the seedling stage (A and D), tillering stage (B and E), and heading stage (C and F). G and H, Leaf blade width (G) and length (H) of the WT and wl1 plants for the 1–13th leaves produced during the total growth period. Values are means ± sd (n = 10 leaves). The P-values were determined using one-way ANOVA test as compared with Xinong 1B. ^*P < 0.05, while ^**  P < 0.01. Bars, 10 cm in (A–C); 1 cm in (D–F).

To investigate the developmental sequence leading up to the wide leaf phenotype in wl1, we performed histological analyses of the SAM, leaf primordia, and leaf blades. The size and shape of the SAM of wl1 were not significantly different from those of the WT (Figure 2, A, B, and I). However, the area and width of the P1, P2, and P3 leaf primordia in the wl1 mutant were greater than those in the WT (Figure 2, C, D, J, and L). Subsequent histological analysis of the fifth fully expanded leaves at the seedling stage revealed that the wl1 leaves were significantly wider than those of the WT (Figure 2, E and F). Notably, the wl1 showed a 22.1% greater distance between the small vascular bundles than in the WT (Figure 2K). There were significantly more small and large vascular bundles in wl1 leaves than in those of the WT throughout the whole growth season (Figure 2, G and H). There were no visible changes in the shape of the midrib or the large and small vascular bundles (Supplemental Figure S1). Therefore, the distinctly wide leaf of wl1 plants originated at the leaf primordium developmental stage and mainly depended on the number of vascular bundles and the distances between the small vascular bundles.

Figure 2.

Histological analysis of WT and the wl1 mutant. A and B, Longitudinal sections of the shoot apex of WT (A) and wl1 (B) plants at seedling stage. C and D, Transverse sections of the shoot apex of WT (C) and wl1 (D) plants at seedling stage. E and F, Transverse sections of the fifth fully expanded leaves of WT (E) and wl1 (F) plants at the seedling stage. G–L, Comparison of the number of small vascular bundles (SVs) (G) and large vascular bundles (LVs) (H), the area of SAM (I), the area of the primordium (J), the distance between two SVs (K), and the width of the primordium (L) of WT and wl1. Values are means ± sd (n = 10 leaves). The P-values were determined using one-way ANOVA test as compared with Xinong 1B. ^*P < 0.05, ^**P < 0.01, and ns represents no significance. P1, the first primordium; P2, the second primordium; P3, the third primordium. Bars, 100 µm.

Molecular cloning and identification of the WL1 gene

We adopted a map-based cloning strategy to isolate the WL1 gene. Genetic analysis of the F₂ progeny revealed that the segregation ratio of the WT and individual mutant plants showed a goodness-of-fit to 3:1 (544 of 1,648 were mutant individuals; χ² = 0.037 < χ²_0.05 = 3.84), indicating that a single recessive gene controls the wl1 trait. Using the 544 mutant plants in the F₂ population, we narrowed down the location of the WL1 gene to a 53.31-kb region between the markers ID56 and ID65 on the long arm of chromosome 3. In this region, there were six annotated genes based on the Michigan State University Rice Genome Annotation Project (Figure 3A). In the wl1 mutant, we identified a single-nucleotide substitution from C to T in the annotated gene LOC_Os03g57240, which converted a glutamine codon into a termination codon, resulting in premature termination of protein translation (Figure 3A). Further analysis showed that WL1 is an allelic gene of DST, which encodes a C2H2 zinc finger protein that regulates the drought and salt stress responses, as well as the grain number per panicle in rice (Huang et al., 2009; Li et al., 2013). We found that the WL1 protein was nuclear localized (Supplemental Figure S2), and its homologs were abundant and highly conserved in both monocotyledons and dicotyledons (Supplemental Figure S3).

Figure 3.

Map-based cloning and expression pattern of WL1. A, Mapping of the WL1 locus (triangle) and the molecular locus of LOC_Os03g57240 in the wl1 and dst mutants. B, Comparison of 2-month-old plants and leaf (the widest part of the latest fully expanded leaf) (insets) of the WT, wl1, and the transgenic lines of WL1-C (complementation) and WL1-RNAi (RNA interference). C, Expression analysis of WL1 in the leaves of the WT and RNAi lines using RT–qPCR. ACTIN (LOC_Os03g50885) was used as an internal control. Values are means ± sd (n = 3 biological replicates). The P-values were determined using Tukey’s HSD test. Lowercase letters indicate significant differences (P < 0.05). D, Leaf width of WT, wl1, the transgenic lines WL1-C and WL1-RNAi, ZF802, and dst. Values are means ± sd (n = 10 leaves). The P-values were determined using Tukey’s HSD test. Lowercase letters indicate significant differences (P < 0.05). E, Expression analysis of WL1 in the root, stem, leaf, leaf sheath, and spikelet of the WT. F–I, Expression analysis of WL1 in the SAMs and leaf primordia of the WT using in situ hybridization. ACTIN (LOC_Os03g50885) was used as an internal control. Values are means ± sd (n  = 3 biological replicates). The P-values were determined using Tukey’s HSD test. Lowercase letters indicate significant differences (P < 0.05). VB, vascular bundle; Xy, xylem; Ph, phloem. Bars, 10 cm in (B); 200 µm in (F–H); 50 µm in (I).

To verify whether the mutation of LOC_Os03g57240 is linked to the mutant phenotype, we performed a complementation experiment by introducing a 3,776-bp WT DNA fragment containing LOC_Os03g57240 into the wl1 mutant. All positive transgenic lines harboring the full-length WL1 transgene showed completely WT phenotypes with normal leaf width (Figure 3, B and D). Next, we used the RNAi suppression test to silence WL1 in the WT plants, which resulted in a significantly reduced quantity of WL1 transcripts, with its leaf width being similar to that of the wl1 mutant (Figure 3, B–D). Furthermore, dst, an allelic mutant of wl1 derived from ZheFu802 (ZF802), which has a single nucleotide A insertion in the exon that results in a coding frame shift, also exhibits a wide leaf phenotype (Figure 3, A and B). Overall, these results confirmed that LOC_Os03g57240/DST corresponds to the WL1 gene.

We investigated WL1 expression using reverse transcription–quantitative PCR (RT–qPCR) analysis, and found this gene to be expressed in various tissues, including the roots, stems, leaves, leaf sheaths, and spikelets (Figure 3E). For a more detailed analysis of its spatial expression pattern, we conducted an in situ hybridization (Figure 3, F–I), and we found WL1 signals being mainly concentrated in the SAM and leaf primordium (Figure 3, F–I). However, despite observing strong signals in both the SAM (Figure 3F) and the P1 primordium, the signal was scattered in the latter (Figure 3G). In the P2 and P3 primordia, the WL1 signals were concentrated at the margin of the primordia and the procambial cells (Figure 3G), potentially indicating the sites of differentiation of the fundamental cells into the procambial cells. In the P4 primordium, WL1 signals were largely restricted to the vascular bundles, with no preference for either the xylem or phloem (Figure 3, H and I). Therefore, these results suggested that WL1 may be involved in leaf development.

WL1 interacts with TAD1

To analyze the molecular mechanism of the WL1-mediated regulation of leaf width, we first searched for proteins that may interact with the WL1 protein to regulate leaf development. A yeast two-hybrid (Y2H) assay showed that WL1 interacted with TAD1 (Figure 4A), a co-activator of the APC/C, which is a multi-subunit E3 ligase that regulates the growth of rice tillers and roots (Lin et al., 2012; Xu et al., 2012; Lin et al., 2020). We verified the interaction between WL1 and TAD1 using an in vitro GST pull-down assay. We expressed both glutathione-S-transferase-tagged WL1 (GST-WL1) and His-tagged TAD1 (His-TAD1) in Escherichia coli, and found that GST-WL1, but not GST, could pull down His-TAD1 (Figure 4B). These were consistent with the results we obtained from the bimolecular fluorescence complementation assay (BiFC) in the leaf epidermal cells of Nicotiana benthamiana in vivo. We fused the WL1 and TAD1 to the C-terminal half (WL1-cYFP) and the N-terminal half (TAD1-nYFP) of a YFP, respectively. The YFP signal was mainly restricted to the nuclei when WL1-cYFP was co-transformed with TAD1-nYFP (Figure 4C). These results confirm the direct interaction between WL1 and TAD1 both in vitro and in vivo. Furthermore, RT–qPCR results showed that TAD1 also shared the same spatial expression pattern as WL1 (Figure 4D). In situ hybridization results showed that TAD1 was expressed in the SAM, P1 and P2 primordium, the margin of the P3 primordium, the procambial cells, and the vascular bundles with no preference for either xylem or phloem, which is similar to that of WL1 (Figure 4, E–H). Thus, the expression pattern of TAD1 was consistent with that of WL1.

Figure 4.

WL1 interaction with TAD1. A, WL1 interacts with TAD1 in yeast cells, as shown by a Y2H assay. B, WL1 interacts with TAD1 in vitro, as detected by a GST pull-down assay. C, WL1 interacts with TAD1 in the nuclei of N. benthamiana leaf cells, as measured by a BiFC assay. D, Expression analysis of TAD1 in the root, stem, leaf, leaf sheath, and spikelet of the WT. ACTIN (LOC_Os03g50885) was used as an internal control for RT–qPCR. Values are means ± sd (n = 3 biological replicates). The P-values were determined using Tukey's HSD test. Lowercase letters indicate significant differences (P < 0.05). E–H, Expression analysis of TAD1 in the SAMs and leaf primordia of the WT using in situ hybridization. Bars, 25 μm in (C); 200 µm in (E–G); 50 µm in (H).

WL1 acts as a degradable substrate for APC/C^TAD1

TAD1 encodes a rice homolog of Cdh1, which is an activator of the APC/C (Lin et al., 2012). Based on this information, we conducted a series of biochemical experiments to examine whether the APC/C^TAD1 complex mediates WL1 degradation via the ubiquitin-26S proteasome degradation pathway. An in vitro ubiquitination assay demonstrated that WL1 was poly-ubiquitinated more efficiently by the WT total protein extracts of the LS (Lansheng) than by those of the tad1 plants (Figure 5A), thereby indicating that TAD1 was involved in the ubiquitination of WL1. A cell-free degradation assay showed that the LS total protein extracts only rapidly degraded the purified recombinant GST-WL1 protein, but not the GST control. However, its degradation was compromised in the tad1 mutant extracts (Figure 5B). Moreover, MG132 (carbobenzoxy-Leu-Leu-leucinal), a 26S proteasome-specific inhibitor, blocked the degradation of GST-WL1 in the total protein extracts (Figure 5B). Additionally, the WL1 protein level in tad1 was significantly higher than that in LS (Figure 5C). Taken together, these results support the hypothesis that WL1 is a degradable substrate for APC/C^TAD1.

Figure 5.

WL1 acts as a degradable substrate for APC/C^TAD1. A, In vitro ubiquitination assay of GST-WL1 by the LS and tad1 plant extracts. The larger panel of GST-WL1 was confirmed to be polyubiquitinated GST-WL1 by blotting with anti-polyubiquitin antibody and the small panel of GST-WL1 was confirmed to be the unubiquitinated GST-WL1 protein by blotting with anti-WL1 antibody. B, Cell-free degradation assay shows the stability of GST-WL1 protein in WT and tad1 plant extracts with or without 40 mM of proteasome inhibitor MG132. C, Immunoblot analysis showing the quantities of WL1 protein in leaves of LS and tad1 at seedling stage (top). ACTIN bands showing approximately equal loading of total proteins (bottom). D, Comparison of the 2-month-old plants and leaf (the widest part of the latest fully expanded leaf) (insets) of LS, tad1, WL1-RNAi lines, and tad1 WL1-RNAi lines. E, Expression of WL1 in the leaves of LS, tad1, WL1-RNAi lines, and tad1 WL1-RNAi lines. Values are means ± sd (n = 3 biological replicates). ACTIN (LOC_Os03g50885) was used as an internal control for RT–qPCR. F, Leaf width of LS, tad1, WL1-RNAi lines, and tad1 WL1-RNAi lines. Values are means ± sd (n = 10 leaves). The P-value was determined using Tukey’s HSD test. Lowercase letters indicate significant differences (P < 0.05). Bars, 10 cm.

To further understand the genetic interaction between WL1 and TAD1, we analyzed the leaf width of the tad1 mutant. As hypothesized, the tad1 mutant obtained by EMS mutagenesis of LS, showed a NAL phenotype with a 18.63% reduced leaf width as compared to that of the LS (Figure 5, D and F). Furthermore, we generated RNAi lines of WL1 in the LS and tad1 backgrounds. Phenotypic analyses demonstrated that the WL1-RNAi lines in the LS background (WL1-RNAi) displayed a wide leaf phenotype, which was significantly wider by 53.8% (#1) and 61.30% (#2) than in the WT (LS) (Figure 5, D–F). Concurrently, the WL1-RNAi lines in the tad1 background (tad1 WL1-RNAi) had a leaf phenotype that was 23.05% (#1) and 24.11% (#2) wider than that of tad1, thereby indicating that the WL1-RNAi partially rescued the NAL phenotype of tad1 (Figure 5, D–F). We also performed histological analyses of the leaf blades in LS, tad1, WL1-RNAi, and tad1 WL1-RNAi plants, and found that both the number of vascular bundles and the distance between the small vascular bundles in tad1 and WL1-RNAi were significantly lower and higher than that of LS, respectively, while those in tad1 WL1-RNAi were significantly higher than that of tad1 (Supplemental Figure S4). Therefore, these results suggest that TAD1 and WL1 may at least partly operate via a common pathway to control leaf width by influencing the development of vascular bundles.

WL1 directly represses NAL1 expression

WL1 encodes a C2H2 zinc finger protein that contains a conserved C2H2 zinc finger domain and two ERF-associated amphiphilic repression (EAR) motifs (Supplemental Figure S5A). Previous studies and our Y2H results have revealed that WL1 is a transcriptional activator (Huang et al., 2009; Li et al., 2013) (Supplemental Figure S5B). Here, we used a dual-luciferase reporter (DLR) system to investigate the transcriptional-regulation activity of WL1. The positive control VP16 showed a relatively high luciferase (LUC) activity, whereas the VP16–WL1 fusion protein displayed comparatively much lower activity than the positive control (Supplemental Figure S5C). Therefore, these findings demonstrated that WL1 might also act as a transcriptional repressor.

To identify the target genes of WL1, we used RT–qPCR to analyze the expression of some genes related to leaf width regulation. Interestingly, the NAL1 gene, which positively regulates leaf width growth by influencing the development of vascular bundles (Qi et al., 2008), was significantly upregulated and downregulated in wl1 and tad1, respectively, as compared to the WT (Xinong 1B and LS) (Figure 6, A and, B). The C2H2 zinc finger domain can bind to the cis-element called the DST-binding sequence (DBS) (TGNTANN(A/T)T) (Huang et al., 2009; Li et al., 2013). Sequence analysis revealed the presence of DBS in the NAL1 promoter region (Figure 6C). We used chromatin immunoprecipitation (ChIP) assays to test whether the WL1 protein could bind to the DBS region of the NAL1 promoter. Chromatin isolated from young WT seedlings was immuno-precipitated with the anti-WL1 antibody and then subjected to quantitative real-time PCR (qPCR) analysis. WL1 could indeed bind stably to the F2 site, which contained the TGGTATCTT sequence (Figure 6, C and D), suggesting that WL1 might directly regulate NAL1. Next, we studied the effect of WL1 on the expression of the firefly LUC gene reporter, using the DBS-like motif region of NAL1 as a promoter for transient expression assays in rice protoplasts. When compared with the negative control Pro35Smini:LUC reporter (Reporter 1), LUC activity was significantly increased with the ProNAL1-Pro35Smini:LUC reporter (Reporter 2). Co-expression of the ProNAL1-Pro35Smini:LUC reporter with Pro35S:WL1 (Effector 1 + Reporter 2) significantly decreased the LUC activity, whereas there were no significant LUC activity changes, when we co-expressed the ProNAL1-Pro35Smini:LUC reporter with Pro35S:mutated WL1 (mWL1) (Effector 2 + Reporter 2) (Figure 6, E and F). For detailed analysis of the NAL1 expression pattern, we examined the transverse and longitudinal SAM sections for NAL1 signals via in situ hybridization. We observed strong signals in the SAM, P1 and P2 primordium, the margin of the P3 primordium and the procambial cells, and the vascular bundles with no preference for either the xylem or phloem (Figure 6, G–J), thus sharing similarity with that of WL1. Therefore, these results indicated that WL1 repressed NAL1 expression by binding to the DBS-like motif in its promoter.

Figure 6.

WL1 directly regulates NAL1 expression. A and B, Expression analysis of NAL1 in the fully expanded leaves of 2-week-old WT, wl1 (A) and LS, tad1 plants (B) using RT–qPCR. ACTIN (LOC_Os03g50885) was used as an internal control. Values are means ± sd (n = 3 biological replicates). The P-values were determined using one-way ANOVA test as compared with Xinong 1B and LS. ^**P < 0.01. C, Distribution of the potential binding sites in the NAL1 promoter and genomic regions. Blue bars (F1–F4) represent the DNA fragments amplified in the ChIP assays. D, ChIP-qPCR for the F1–F4 sites of NAL1 with the anti-WL1 antibody. ChIP enrichment, as compared with the input sample, was tested by qPCR. Values are means  ± sd (n = 3 biological replicates). The P-values were determined using one-way ANOVA test. ^**P < 0.01. ns represents no significance. E and F, WL1 repressed NAL1 expression in vivo. Protoplasts were extracted from 2-week-old rice plants, and were transformed with Pro35Smini:LUC (Reporter 1), Pro35Smini:LUC plus Pro35S:WL1 (Reporter 1 + Effector 1), Pro35Smini:LUC plus Pro35S:mWL1 (Reporter 1 + Effector 2), ProNAL1-Pro35Smini:LUC (Reporter 2), ProNAL1-Pro35Smini:LUC plus Pro35S:WL1 (Reporter 2 + Effector 1), or ProNAL1-Pro35Smini:LUC plus Pro35S:mWL1 (Reporter 2  + Effector 2). The P-values were determined using Tukey’s HSD test. Lowercase letters indicate significant differences (P < 0.05). G–J, Expression analysis of NAL1 in the SAMs and leaf primordia of the WT using in situ hybridization. Bars, 200 µm in (G–I); 50 µm in (J).

WL1 interacts with the OsTPR corepressors to downregulate the acetylation levels of histone at NAL1

WL1 contains an EAR motif (LxLxL and DLNxxP), at the C- and N-terminal ends, respectively (Supplemental Figure S5A). Proteins with EAR motifs are hypothesized to interact with the TPR corepressors. There are three TPR proteins in rice, namely OsTPR1 (LOC_Os01g15020), OsTPR2 (LOC_Os08g06480), and OsTPR3 (LOC_Os03g14980) (Zhuang et al., 2020). To further dissect the molecular mechanism of WL1-mediated regulation of NAL1 expression, we examined the interaction between WL1 and OsTPRs. First, we performed a Y2H assay to determine whether WL1 and OsTPRs physically interact. The results demonstrated that the full-length WL1 protein, the C-terminal EAR motif and the N-terminal EAR motif of WL1 protein, and the mWL1 protein, all interacted with the three OsTPR proteins in the yeast cells (Figure 7A) (Supplemental Figure S6). Second, we conducted a BiFC assay to establish whether WL1 could interact with OsTPRs in N. benthamiana leaves. The yellow fluorescent protein signal was present in the nuclei of epidermal cells of N. benthamiana leaves co-expressing the WL1-cYFP fusion protein with nYFP-OsTPR1, nYFP-OsTPR2, or nYFP-OsTPR3 (Figure 7B). Thus, WL1 interacted with OsTPRs in vivo. It is well-documented that TPR can interact with histone deacetylases to regulate the histone acetylation level in the target regions (Zhuang et al., 2020). Therefore, we further detected the histone acetylation level in NAL1 by ChIP-qPCR using the chromatin immunoprecipitated by the anti-H3K9ac antibodies. The acetylation levels of H3K9 at the P4, P6, P7, and P9–P11 sites, which are largely in the promoter and first intron of NAL1, were significantly increased in the wl1 mutant as compared with those in the WT (Figure 7, C and D). Therefore, these results support the hypothesis that WL1 recruits OsTPR corepressors to inhibit NAL1 expression by reducing the histone acetylation levels on the chromosomes at NAL1.

Figure 7.

WL1 interacts with the OsTPR corepressors to lower histone acetylation levels on the chromosomes at NAL1. A, WL1 interacts with OsTPR1, OsTPR2, and OsTPR3 in yeast cells, as shown by a Y2H assay. B, WL1 interacts with OsTPR1, OsTPR2, and OsTPR3 in the nuclei of N. benthamiana leaf cells, as demonstrated by a BiFC assay. C, Distribution of potential binding sites in the promoter and open reading frame regions (P1–P14) of NAL1 and the mutated sites of nal1 and nal1-c1, nal1-c2. Blue bars represent the DNA fragments amplified in the ChIP assays. D, ChIP-qPCR for the P1–P14 sites of NAL1 with anti-H3K9ac antibody in leaves of the WT and wl1. Values are means ± sd (n = 3 biological replicates). The P-values were determined using one-way ANOVA test as compared with Xinong 1B. ns represents no significance. ^**P < 0.01. E, Comparison of the plants and leaf of the WT, wl1, nal1-c1, and wl1 nal1-c2. F, Comparison of the 2-month-old plants and leaf width (the widest part of the latest fully expanded leaf) (insets) of WT, wl1, nal1-c1, and wl1 nal1-c2 plants. Values are means ± sd (n = 10 leaves). The P-values were determined using Tukey’s HSD test. Lowercase letters indicate significant differences (P < 0.05). Bars, 10 cm.

To further analyze the genetic interaction between WL1 and NAL1, we generated the loss-of-function mutant nal1-c1 using CRISPR/Cas9 technology in the WT (Xinong 1B). The nal1-c1 mutant had a 5-bp deletion in the NAL1 CDS, resulting in a reading frameshift that generated a premature stop codon (Figure 7C). We observed a narrower leaf (38.08% reduced width compared with the WT) as expected in nal1-c1 (Figure 7, E and F). Next, we used clustered regularly interspaced short palindromic repeats (CRISPR/Cas9) technology to knock out NAL1 in the wl1 background and generate the wl1 nal1-c2 double mutant. The nal1-c2 mutant had the same 5-bp deletion in NAL1 as nal1-c1 (Figure 7C). The wl1 nal1-c2 double mutant showed a narrower leaf phenotype than wl1, with a 11.89% reduced leaf width, thus indicating that the nal1-c2 mutation suppressed the leaf width phenotype of wl1. Furthermore, we performed histological analyses of the leaf blades in wl1, nal1-c1, and wl1 nal1-c2, and found that the number of small and large vascular bundles and the distance between the small vascular bundles in nal1-c1 was significantly lower than that of WT (Xinong 1B), while the number of small vascular bundles and the distance between them in wl1 nal1-c2 was significantly lower than that of wl1 (Supplemental Figure S7). Furthermore, we analyzed the genetic relationship between WL1 and NAL1 via the dst and nal1 mutants, which were the allelic mutants of wl1 and nal1-c1 (nal1-c2), respectively, and both of which were derived from ZF802. The previous results showed that dst presented wide leaves (Figure 3, A, B, and D). The nal1 mutant, which has a 30-bp deletion in the NAL1 exon, had narrow leaves (Figure 7C; Supplemental Figure S8, A and B) (Qi et al., 2008). Similarly, the dst nal1 double mutant also had narrower leaves than dst plants (Supplemental Figure S8, A and B). Additionally, we conducted histological analyses of the leaf blades in dst, nal1, and dst nal1 and found that the number of small and large vascular bundles and the distance between the small vascular bundles in dst and nal1 plants were significantly higher and lower than in the WT (ZF802), respectively, while the number of small and large vascular bundles in dst nal1 were significantly lower than in dst (Supplemental Figure S8, C–F). Overall, the above results suggest that WL1 and NAL1 may partly function antagonistically via a common pathway to control the leaf width growth by influencing the development of vascular bundles in rice.

Discussion

Leaf morphology is one of the most important components of an ideal plant architecture that critically influences photosynthesis in rice. A good leaf shape can optimize the canopy structure, increase the photosynthetic area, and ultimately boost grain yield. Although the molecular mechanism of leaf development has been well-studied in the model dicot Arabidopsis thaliana (oval leaves with reticulate veins) (Yasui et al., 2018), contrastingly, it is limited in agronomically important monocots like rice, which have long and narrow leaves with parallel veins. In this study, we isolated a wl1 mutant, which exhibited an increased leaf width. We reported that the wl1 phenotype was induced by a mutation in DST, which encodes a C2H2 zinc finger protein. DST regulates drought and salt tolerance as well as grain number per panicle in rice (Huang et al., 2009; Li et al., 2013; Cui et al., 2015). However, how DST helps regulate leaf width is not well-studied.

To elucidate the involvement of WL1 in the control of leaf width, we demonstrated that the APC/C^TAD1–WL1–NAL1 pathway promoted leaf growth, suggesting that this pathway can potentially increase the leaf width in key crop plants. TAD1 is an activator of the APC/C complex and regulates the development of tillers and roots (Lin et al., 2012; Xu et al., 2012; Lin et al., 2020). In this study, the loss-of-function of TAD1 caused the NAL phenotype, thereby indicating that TAD1 is a positive regulator of leaf width in rice. Contrastingly, the wl1 mutant and the RNAi lines showed a wide leaf, thus indicating that WL1 is a negative regulator of this phenotype. Biochemical analyses indicated that TAD1 interacts with WL1 and mediates its degradation via the ubiquitin–26S proteasome degradation pathway. Additionally, genetic analyses suggested that TAD1 and WL1 act via a common pathway to regulate leaf width. Therefore, we hypothesize that TAD1 and WL1 act as a module that regulates leaf width. Furthermore, we provided evidence that WL1 directly binds to the promoter or other regulatory regions of the NAL1 gene, and then recruits the corepressor OsTPRs. These repressors inhibit the NAL1 expression by reducing the histone acetylation levels on the region of the chromosome harboring NAL1, thus indicating that NAL1 is a target gene for WL1. Taken together, our genetic and biochemical analyses revealed that the APC/C^TAD1–WL1–NAL1 module positively regulates leaf width in rice. Based on these results, we proposed a working model of how the APC/C^TAD1–WL1–NAL1 module controls leaf width in rice (Figure 8). TAD1 activates the APC/C E3 ubiquitin ligase, which targets WL1 for degradation, while WL1 negatively regulates NAL1 expression, and consequently reduces widthwise expansion of the leaf (Figure 8).

Figure 8.

Proposed model for the APC/C^TAD1–WL1–NAL1 module-mediated control of leaf width. TAD1 activates APC/C E3 ubiquitin ligase activity and targets WL1 for degradation. WL1 negatively regulates NAL1 expression, and consequently regulates the leaf width.

Previous studies have identified a variety of NAL mutants in rice, including the auxin-related mutants nal1, fib, nal7, and tdd1 (Fujino et al., 2008; Qi et al., 2008; Sazuka et al., 2009; Yoshikawa et al., 2014); the CSLD-related mutant nrl1 (Li et al., 2009) and the polarity-related mutants sll1 and lf1 (Zhang et al., 2009, 2021). Therefore, the molecular mechanisms of the NAL trait have been extensively studied. However, studies on mutants with wide leaf traits along with their underlying molecular mechanisms are scarce. The D62 gene encodes a protein of the GRAS family, which are GA signaling inhibitors. Mutations in this gene produce dwarf plants with more erect, wider, shorter, and darker leaves (Tong et al., 2009). Osa-miR319a and Osa-miR319b belong to the miR319 gene family. Overexpression of Osa-miR319a and Osa-miR319b in rice led to increased leaf width. The number of small vascular bundles in the Osa-miR319-overexpressing plants was significantly higher than in the WT, thereby indicating that the wide leaves of the Osa-miR319-overexpressing plants resulted from a greater number of small vascular bundles and increased number of cells between them (Yang et al., 2013). Here, we identified a novel wide leaf mutant with an increase in the area and width of leaf primordia, the distances between the small vascular bundles, and the numbers of small and large vascular bundles, as compared with the WT. Therefore, WL1 is likely to be important in regulating leaf width development by modifying the number of large vascular bundles and the distances between the small vascular bundles. Besides the altered leaf width, the wl1 mutant displayed a larger panicle with a significantly higher number of grains per panicle (Supplemental Figure S9). Thus, we propose WL1 as a prime candidate for further studies to support the genetic improvement of the rice leaf phenotypes and grain production.

Materials and methods

Plant materials

The rice (O. sativa) mutant wl1 was isolated from an EMS-mutagenized population of O. sativa subsp. indica cultivar Xinong 1B. Xinong 1B was used as the WT reference for phenotypic observation and in situ hybridization analyses. All plant materials were grown in the experimental fields of the Rice Research Institute of Southwest University in Chongqing and Hainan, China.

Map-based cloning of WL1

The wl1 mutant was crossed with “Jinhui 10” (bred by the Rice Research Institute, Southwest University), and 544 F₂ plants exhibiting the mutant phenotype were selected for gene mapping. Gene mapping was conducted using the simple sequence repeat markers obtained from the publicly available rice databases, Gramene (http://www.gramene.org) and the Rice Genomic Research Program (http://rgp.dna.affrc.go.jp/E/publicdata/caps/index.html). The insertion/deletion markers were developed by comparing the genomic sequences of Xinong 1B with those of Jinhui 10 in our laboratory. The primer sequences used in gene mapping and candidate gene analyses are provided in Supplemental Data Set 1.

Vector construction and plant transformation

For the complementation test, a 3,776-bp fragment containing the LOC_Os03g57240 genomic fragment was cloned into the binary vector pCAMBIA1301. The resulting recombinant plasmids were introduced into wl1 using the Agrobacterium tumefaciens-mediated transformation method as described previously (Zhang et al., 2017). To generate the RNAi construct, a 264 bp WL1 complementary DNA (cDNA) was amplified and inserted into the vector pTCK303 to obtain the intermediate vector. The resulting recombinant plasmids were transformed into the WT (Xinong 1B), LS, and tad1 mutant, respectively, using the Agrobacterium-mediated transformation method (Zhang et al., 2017). For the knockout assay, the 20-bp target sequence (5′-TGATGCTCCTCGGAGCGTCT-3′) from the NAL1 genomic fragment for sgRNA targeting was cloned into the CRISPR/cas9 expression vector to generate the CRISPR/cas9 construct pCRISPR-NAL1 (Ma et al., 2015). The resulting recombinant plasmid was introduced into the WT (Xinong 1B) and the wl1 mutants, using the A. tumefaciens-mediated transformation method (Zhang et al., 2017). The primer sequences used are listed in Supplemental Data Set 1.

RNA isolation and RT–qPCR analysis

Total RNA of rice was isolated from the roots, stems, leaf sheaths, leaves, and young panicles using the RNAprep Pure Plant RNA Purification Kit (Tiangen, Beijing, China). First-strand cDNA was synthesized from 2 µg of purified total RNA using oligo(dT)₁₈ primers in a 25 µL reaction volume with the SuperScript III Reverse Transcriptase Kit (Invitrogen, Carlsbad, CA, USA). The resulting cDNA (0.5 µL) was used as a template for PCR amplification with gene-specific primers (Supplemental Data Set 1). RT–qPCR analysis was performed with the CFX Connect Real-Time System (Bio-Rad, Berkeley, CA, USA) and an SYBR premix Ex Taq II Kit (TaKaRa, Kyoto, Japan), and ACTIN as an internal control. A minimum of three replicates were analyzed to produce the mean values of the expression levels of each gene.

Microscopy analysis

The aerial tissues containing the SAMs and leaf primordia from WT and wl1, and leaves of LS, tad1, WL1-RNAi, tad1 WL1-RNAi, WT, wl1, nal1-c1, wl1 nal1-c2, ZF802, dst, nal1, and dst nal1 were collected and immediately fixed in formaldehyde alcohol acetic acid solution (3.7% formaldehyde, 50% ethanol, 0.9-M glacial acetic acid) and maintained at 4°C overnight. The samples were dehydrated in a series of ethanol solutions, then substituted with xylene, and finally embedded in paraffin (Sigma, St Louis, MO, USA). The samples were subsequently sectioned into 8 µm thickness, and mounted onto poly-L-Lys-coated glass slides, de-paraffinized with xylene, and finally rehydrated through a series of ethanol solutions. The sections were stained with 1% w/v safranine (Amresco, Framingham, MA, USA) and 1% w/v Fast Green (Amresco), dehydrated through a series of ethanol solutions, cleared with xylene, and covered with a coverslip. Light microscopy observation of the sections was performed using an Eclipse E600 microscope (Nikon, Tokyo, Japan).

Multiple sequence alignment and phylogenetic tree construction

Protein sequences were obtained with Basic Local Alignment Search Tool in PHYTOZOME, using an EXPECT value threshold of 10⁻⁵ (https://phytozome-next.jgi.doe.gov/blast-search). A phylogenetic tree was constructed using MEGA version 5.0 (Tamura et al., 2011). The tree was constructed using the maximum-likelihood method based on the Jones, Taylor, Thornton matrix-based model with the lowest Bayesian information criteria scores (Tamura et al., 2011). Each node’s bootstrap support values from 500 replicates were included next to the branches.

Subcellular localization

The full-length WL1 coding region was amplified and cloned into the expression cassette 35S-GFP-NOS (pAN580) to generate the GFP-WL1 and WL1-GFP fusion vectors. Then, GFP, GFP-WL1, and WL1-GFP plasmids were inserted into rice protoplasts as previously described (Zhang et al., 2017). After overnight incubation at 28°C, GFP fluorescence was observed with a confocal laser scanning microscope (Olympus, Tokyo, Japan). The primers used are listed in Supplemental Data Set 1.

In situ hybridization

For the WL1 probe, its cDNA was amplified and labeled using the DIG RNA Labeling Kit (Roche, Basel, Switzerland). Pretreatment of sections, hybridization, and immunological detection was performed as described previously (Zhang et al., 2017). The primers used are listed in Supplemental Data Set 1.

Transcriptional activation assay

The transcriptional activity of WL1 in rice protoplasts was analyzed using the DLR assay system. The DLR assay system used the GloMax 20-20 luminometer (Promega, Madison, WI, USA) to measure the relative LUC activity (Zhuang et al., 2020). The full length and truncated coding frames of WL1 were fused to the GAL4 DNA-binding domain (BD) driven by the 35S promoter. The transcriptional activator VP16 was used as a positive control, whereas GAL4-BD was the negative control. VP16, BD-WL1, VP16-WL1, and GAL4-BD effectors were transiently expressed in rice protoplasts. The primers used are listed in Supplemental Data Set 1.

ChIP-qPCR

The WT and wl1 plants were subjected to ChIP analysis. Leaves were collected to isolate nuclear extracts. The EpiQuik Plant ChIP Kit (P-2014-48, Epigentek, New York, USA), anti-WL1 antibody, and anti-Histone H3 (acetyl K9) antibody (ChIP grade; ab10812, Abcam, Cambridge, U K) were used in the ChIP assays. Immunoglobulin G (IgG) antibody was included as a negative control. All PCR experiments were conducted using 40 cycles of 95°C for 5 s, 60°C for 30 s, and 72°C for 30 s. The reaction mixtures contained 10 pmol of each primer and 1 mL of DNA from the ChIP, control or input DNA diluted 20-fold (per biological replicate) as template. A minimum of three biological repeats (1 g of leaf sample), each with three technical repeats, were subjected to statistical analysis. Experimental procedures for ChIP-qPCR were performed as described previously (Zhang et al., 2017). The primer sequences are listed in Supplemental Data Set 1.

Transient expression assay

The NAL1 and the 35S promoters were amplified and cloned into the pGreenII0800-LUC double-reporter vector. To generate the 2xPro35S:WL1 constructs, the full-length WL1 coding region was amplified using PCR and cloned into the pAN580 vector downstream of the two CaMV35S promoters. The constructs were then transformed into rice protoplasts as previously described (Zhang et al., 2021). After overnight incubation at 28°C, LUC and Renilla (REN) LUC activities were measured using the Dual Luciferase Assay Kit (Promega, Madison, WI, USA), and analyzed using the Luminoskan Ascent Microplate Luminometer (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s instructions. The results were calculated based on the ratios of LUC/REN (Zhuang et al., 2020). A minimum of six transient assay measurements were carried out for each sample. The primers used are listed in Supplemental Data Set 1.

Y2H

The Y2H assays were performed using the Matchmaker Gold Yeast Two-Hybrid System (Clontech, Mountain View, CA, USA). The full-length and truncated coding regions of WL1 were amplified and cloned into the yeast expression vector pGADT7 (Clontech) and pGBKT7 (Clontech) to produce pGADT7-WL1, pGADT7-N-WL1, pGADT7-C-WL1, pGADT7-mWL1 and pGBKT7-WL1, pGBKT7-N-WL1, pGBKT7-C-WL1, pGBKT7-mWL1, respectively. The full-length coding regions of TAD1, OsTPR1, OsTPR2, and OsTPR3 were cloned into the vector pGBKT7 (Clontech) to produce pGBKT7-TAD1, pGBKT7-OsTPR1, pGBKT7-OsTPR2, and pGBKT7-OsTPR3. The Y2H Gold yeast strain was used in the Y2H assays. pGADT7-T1pGBKT7-lam and pGADT7-T1pGBKT7-53 served as the negative and positive controls, respectively. According to the manufacturer’s instructions, these plasmids were co-transformed into the Y2H Gold yeast strain in an activation domain–BD coupled manner (Yeast Protocols Handbook, PT3024-1; Clontech). The primers used are listed in Supplemental Data Set 1.

BiFC

The vectors pXY104 and pXY106 harboring the fragments encoding the C- and N-terminal halves of YFP (cYFP and nYFP), respectively, were used to generate the constructs for the BiFC assays. The WL1 CDS was fused to the fragment encoding the C-terminus of YFP, whereas the CDS fragments encoding TAD1, OsTPR1, OsTPR2, or OsTPR3 were fused to the fragment encoding the N-terminus of YFP. All these vectors were transformed into the Agrobacterium strain GV3101. The Agrobacterium cultures harboring the constructs expressing the nYFP and cYFP fusion proteins were mixed in a 1:1 ratio and then transformed into N. benthamiana leaves via agroinfiltration. The BiFC signals were detected using confocal microscopy, as described above. The primers used are listed in Supplemental Data Set 1.

GST pull-down assay using purified proteins

The GST pull-down assays were performed as previously described, with minor modifications (Zhang et al., 2021). The WL1 and TAD1 CDS fragments were cloned into the pGEX-4T-1 and pET32a vectors, respectively. The recombinant GST-WL1 and His-TAD1 fusion proteins were expressed in E. coli (BL21). The His-TAD1 and GST-WL1 fusion proteins were purified according to the manufacturer’s protocol (New England Biolabs, Ipswich, MA, USA). The bait protein GST-WL1 was incubated with GST beads (GE Healthcare, Chicago, IL, USA) in non-EDTA buffer (140-mM NaCl, 2.7-mM KCl, 10-mM Na₂HPO₄, 1.8-mM KH₂PO₄, pH 7.4) at 4°C for 2 h, and then washed thrice with the same buffer. The beads were re-suspended in 1 mL non-EDTA buffer before 50–100 ng His-TAD1 prey protein was added to the solution. The mixture was then incubated at 4°C for 1 h and rinsed thrice. Proteins were eluted with sodium dodecyl sulfate (SDS) loading buffer and subjected to immunoblot analysis. The prey and bait proteins were detected using the anti-His (Cat. no. 2,365; CST, MA, USA; 1:1,000 dilution) and anti-GST (Cat. no. 2,622; CST, MA, USA; 1:1,000 dilution) antibodies, respectively. The primers used are listed in Supplemental Data Set 1.

Antibody preparation and immunoblot analysis

The WL1 CDS was cloned into the pGEX-4T-1 vector. The purified GST-WL1 fusion protein was injected into rabbits to produce polyclonal antibodies against WL1. Total protein extracts from the LS and tad1 young seedlings were isolated using a Plant Total Protein Extraction Kit (Sangon Biotech, Shanghai, China). Immunoblots were performed using primary antibodies against WL1 and ACTIN (K800001M, Solarbio, 1:2,000 dilution). After incubating with secondary antibodies HRP-labeled Goat Anti-Mouse IgG (H + L) epizyme LF101 (Cat. no. LF101; epizyme; 1:2,000 dilution) and HRP-labeled Goat Anti-Rabbit IgG (H + L) epizyme LF102 (Cat. no. LF102; epizyme; 1:2,000 dilution), respectively, for 1 h, the immunoblot signal was visualized using the Immobilon Western HRP substrate (ZOMANBIO, ZD310A-1).

Cell-free degradation assay

The cell-free degradation assay was performed as previously reported (Lin et al., 2012; Xu et al., 2012). Total proteins from the tad1 and WT (LS) seedlings were extracted with the 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) and adjusted to equal concentrations with the degradation buffer. MG132 (T510313-0001, Sangon Biotech, Shanghai, China) was selectively added to various in vitro degradation assays, as indicated. To degrade the purified GST and GST-WL1 proteins, equal amounts of the purified proteins (∼100 ng) were incubated in 100-μL aliquots of rice total protein extracts (∼500 μg) at 28°C for each assay. An equal amount of solvent for each reagent was used as the mock control. The extracts were then incubated at 28°C and the samples were collected at the indicated intervals to determine the GST and GST-WL1 protein abundance via immunoblots using anti-GST antibody (Cat. no. 2622; CST; 1:1,000 dilution). After incubating with a secondary antibody (Cat. no. 7074P2; CST, MA, USA; 1:2,000 dilution) for 1 h, the immunoblot signal was visualized using the Immobilon Western HRP substrate (ZOMANBIO, ZD310A-1).

In vitro ubiquitination assay

The in vitro ubiquitination assay was performed as previously reported (Lin et al., 2012; Xu et al., 2012). Purified GST-WL1 protein bound to Glutathione Sepharose (P2020, Solarbio) was incubated at 28°C with equal amounts of the crude protein extracts of rice seedlings in a buffer (25-mM Tris–HCl, pH 7.5, 10-mM MgCl₂, 5-mM dithiothreitol, 10-mM NaCl, 10-mM ATP, and 40-μM MG132). After incubation at 28°C for the indicated intervals, both the GST-WL1 and polyubiquitinated GST-WL1 (GST-WL1-(Ub)n) fusion proteins dissolved in SDS–PAGE loading buffer, were added. The samples were run on sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS–PAGE) gels. Immunoblots were performed with antibodies against the polyubiquitin tail (Cat. no. 3,936S; CST, MA, USA; 1:1,000 dilution) or WL1 (Rabbit polyclonal antibody against WL1 produced by YOUKE BIOTECH, Shanghai, China; 1:1,000 dilution). The upshifted bands of GST-WL1 in Figure 5A were confirmed to be the polyubiquitinated GST-WL1 by blotting with anti-polyubiquitin antibodies (Cat. no. 3,936S; CST, MA, USA; 1:1,000 dilution).

Statistical analysis

Quantitative data for gene expression, LUC activity, and statistical data were examined for statistically significant differences using one-way analysis of variance （one-way ANOVA） test or Tukey’s honestly significant difference （HSD） test as described in the corresponding figure legends. The results are shown in Supplemental Data Set 2. For analysis, fully expanded leaves from different growth stages were randomly selected and analyzed as biological replicates. One-way ANOVA tests were performed using Excel (2021) and Tukey’s HSD tests were performed using the SPSS version 26.0 Statistical package (IBM, Armonk, NY, USA).

Accession numbers

Sequence data from this article can be accessed in the GenBank database under the following accession numbers: WL1 (DST), NAL1, TAD1, ACTIN, OsTPR1, OsTPR2, and OsTPR3 are GQ178286, EU093963, AK070642, AB047313, AP014957, AK111830, and AP014959, respectively. Locus identifications in the Rice Genome Annotation Project Database are as follows: WL1 (LOC_Os03g57240), NAL1 (DST) (LOC_ Os04g52479), TAD1 (LOC_Os03g03150), ACTIN (LOC_Os03g50885), OsTPR1 (LOC_Os01g15020), OsTPR2 (LOC_ Os08g06480), and OsTPR3 (LOC_Os03g14980).

Supplemental data

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

Supplemental Figure S1. Histological analysis of the midribs and vascular bundles in the fifth fully expanded leaves at the seedling stage of the WT and wl1.

Supplemental Figure S2. Analysis of the subcellular localization of WL1 protein in rice protoplasts.

Supplemental Figure S3. Phylogenetic tree for the WL1 protein.

Supplemental Figure S4. Histological analysis of the fifth fully expanded leaves at the seedling stage of LS, tad1, WL1-RNAi, and tad1 WL1-RNAi.

Supplemental Figure S5. Analysis of the transcriptional activation of WL1.

Supplemental Figure S6. WL1 interacts with OsTPR1, OsTPR2, and OsTPR3 in yeast cells as shown by Y2H assays.

Supplemental Figure S7. Histological analysis of the sixth fully expanded leaves at the seedling stage of WT, wl1, nal1-c1, and wl1 nal1-c2.

Supplement Figure S8. Phenotypes of ZF802, dst, nal1, and dst nal1.

Supplemental Figure S9. Yield-related agronomic traits in the WT and wl1 mutant.

Supplemental Table S1. Leaf width and length statistics of wild-type and wl1 mutant throughout the growth period.

Supplemental Data Set 1. Sequences of the oligonucleotide primers and probes used in this study.

Supplemental Data Set 2. Statistical analysis used in this study.

Supplemental Data Set 3. Text file of the alignment used for the phylogenetic analysis shown in Supplemental Figure S3.

 

T.Z. and G.H.H. conceived and designed the experiments. J.Y., W.W.X., Y.Z., W.Q.S., L.Y., P.Y., G.L.Y., Q.N.D., X.F.Z., Z.F.H., and Y.X. conducted the experiments. T.Z., J.Y., W.W.X., X.C.S., Y.F.L., F.M.Z., and Y.H.L. analyzed the data, and T.Z., G.H.H., and J.Y. wrote the manuscript.

The authors 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) are: Ting Zhang (tingzhang@swu.edu.cn) and Guanghua He (heghswu@163.com).