Changes in the expression pattern of OsWUS negatively regulate plant stature and panicle development in rice

Abstract WUSCHEL (WUS) and WUSCHEL-RELATED HOMEOBOX (WOX) encode transcription factors and play important roles in regulating the formation and maintenance of shoot and floral meristems. OsWUS have distinct functions in meristem development with slightly tuned expression. However, the mechanisms regulating the specific expression of OsWUS need to be further explored. In this study, an abnormal expression mutant of OsWUS, called Dwarf and aberrant panicle 1 (Dap1) was used. In order to identify the causal gene in Dap1, high-efficiency thermal asymmetric interlaced (hiTAIL)-PCR and co-segregation analysis were performed. We surveyed the growth and yield traits in Dap1 and wild type. Changes in gene expression between Dap1 and wild type were determined by RNA-seq. The Dap1 mutant is due to the T-DNA inserted at 3,628-bp upstream of the translation start codon of OsWUS. Plant height, tiller numbers, panicle length, the number of grains per main panicle, and the number of secondary branches was significantly reduced in the Dap1 mutant. The expression of OsWUS was markedly increased in Dap1 mutant plants compared to the wild type, which might be due to a disruption in the genomic sequence integrity. Simultaneously, the expression levels of gibberellic acid-related genes and genes involved in panicle development were significantly changed in the Dap1 mutant. Our results suggest that OsWUS is a precise regulatory element, its specific spatio-temporal expression pattern is critical for its function, and both loss-of-function and gain-of-function mutations lead to abnormal plant growth.


Introduction
WUSCHEL (WUS), the prototypical member of the WUSCHEL-RELATED HOMEOBOX (WOX) gene family, encodes a homeodomain transcription factor that regulates the formation and maintenance of shoot-and floral meristems (FMs) (Laux et al. 1996;Mayer et al. 1998). The development of shoot apical meristems (SAMs) and FMs in plants is regulated by the WUS-CLV3 signaling pathway (Mjomba et al. 2016). WUS was originally identified in Arabidopsis (AtWUS) as a central regulator of cell fate in shoot and FMs. The WUS protein is produced in the organizing center (OC) domain of apical meristems (AM) and is transported to the stem cells of the central zone through plasmodesmata to promote the proliferation of stem cells (Yadav et al. 2011). AtWUS is a potential target involved in the CLAVATA (CLV3) pathway to regulate the fate of stem cells in meristems; the expression of CLV3 negatively regulates AtWUS expression in the OC (Brand et al. 2000). The homeotic gene AGAMOUS (AG) is required for the formation and differentiation of FMs. AtWUS cooperates with the floral identity protein LEAFY to activate AG in the center of flowers, and the WUS and AG loop is important to control FM determinacy (Lenhard et al. 2001;Lohmann et al. 2001).
A single WUS ortholog was identified in rice based on phylogenetic analyses and was found to be expressed in young leaf primordia, preferentially in the lateral leaf margins (Nardmann and Werr 2006). OsWUS encodes a protein of 290 amino acids that contains a homeobox domain (HD) (Mjomba et al. 2016). OsWUS, also known as TILLERS ABSENT1 (TAB1), is required for axillary meristem development. The tab1 mutant produces no tillers during vegetative growth, but a single panicle with short branches is formed in tab1 plants in the reproductive phase, and most of the spikelets show morphological defects (Tanaka et al. 2015). Another OsWUS mutant, MONOCULM 3 (moc3) also produces no tillers because the formation of tiller buds is disrupted. Importantly, moc3 is likely a female sterile mutant, because staining moc3 pollen grains showed them to be viable (Lu et al. 2015). STERILE AND REDUCED TILLERING 1 (Srt1), another mutant of OsWUS, produces a nearly full-length WUS peptide that is missing 7 amino acids in the HD. The Srt1 mutant plants also showed a reduction in the number of tillers and complete female sterility. This report further demonstrated that the HD of OsWUS has a potential function in rice (Mjomba et al. 2016). The low-tillering mutant decreased culm number 1 (dc1) was identified as a loss-of-function mutant of OsWUS by its increased apical dominance compared to wild type (WT). The auxin action-associated gene ABERRANT SPIKELET AND PANICLE 1 (ASP1) and OsWUS are both involved in the outgrowth of the rice tiller bud (Xia et al. 2020).
Plant hormones (phytohormones) are low molecular weight signal molecules produced in plants that are active at extremely low concentrations. Phytohormones control essentially every aspect of growth and development by regulating gene expression, transcription, and cell division and differentiation. Cytokinins, a class of plant hormones that actively promote cell division in plant shoots and roots, are involved in the regulation of lateral bud growth and apical dominance. Cytokinins are required for the formation, maintenance, and growth of shoot meristems and act through a complex signaling network (Ferguson and Beveridge 2009). The WUS gene regulates tiller development in association with cytokinin signaling (Lu et al. 2015;Wang et al. 2017). In moc3 plants, the expression levels of 2-component cytokinin response regulator genes (A-type and B-type), including OsRR1, OsRR9, OsRR10, and ORR4 were found to be significantly upregulated (Ito and Kurata 2006;Lu et al. 2015). Gibberellins (GA) play important roles in a variety of growth and developmental processes, including seed germination, stem elongation, flower development, apical dominance, plant stature, tillering, and root development (Ogas et al. 1999;Yamaguchi and Kamiya 2000;Oikawa et al. 2004;Lo et al. 2008;Huang et al. 2010). GA 20-oxidase (GA20ox) is a key enzyme that catalyzes the penultimate steps in GA biosynthesis (Lo et al. 2008). OsGA20ox2 (SD1) is a member of the rice GA20ox gene family known as the "green revolution gene", and loss-of-function mutations at this locus cause semidwarfism (Sasaki et al. 2002;Oikawa et al. 2004). A major catabolic pathway for GA is initiated by a 2-β hydroxylation reaction catalyzed by GA 2-oxidase (GA2ox). High expression levels of OsGA2ox6 may cause decreased levels of bioactive GA, which could explain the dwarf phenotype in rice (Huang et al. 2010).
In this study, we isolated and characterized an abnormal expression mutant of OsWUS that we call Dwarf and aberrant panicle 1 (Dap1). A T-DNA insertion 3,628-bp upstream of the translation start codon of OsWUS alters the expression pattern of OsWUS resulting in shorter plants with reduced panicle length and secondary branch number in the Dap1 mutant. Transcriptomic analyses showed that the expression of GA-related genes and genes involved in panicle development and flowering regulation associated with OsWUS was significantly changed in the Dap1 mutant. This uncoordinated gene expression may contribute to the phenotype of Dap1 mutant plants.

Plant materials
The Dap1 mutant was identified from a T-DNA insertion population created from the japonica rice variety 'Zhonghua11' (ZH11). All rice materials used in this study were grown in the field under normal growth conditions in Guangzhou City, China.

T-DNA flanking sequence analysis
The T-DNA flanking sequences in the Dap1 mutant were amplified by high-efficiency Thermal Asymmetric Interlaced (hiTAIL)-PCR (Liu and Chen 2007). The PCR products were purified and sequenced.

Genotyping of mutant plants
To genotype the mutant plants, total DNA was isolated from T 2 -generation plants generated from a heterozygous T 1 mutant plant. A T-DNA-specific primer (P1) and primers flanking the insertion site (P2 and P3) were used for PCR amplification (Supplementary Table 1). The primer pair P1/P2 directs amplification of a DNA fragment that includes part of 5′ end of the T-DNA and part of the upstream region of OsWUS, and only amplifies the transgene. Primer pair P2/P3 amplifies a 587-bp fragment of the upstream region of OsWUS from WT genomic DNA.

Analysis of cis-regulatory elements in the upstream region of OsWUS
For analysis of the cis-regulatory elements in the sequence upstream of the T-DNA insertion site, we extracted a 500-bp fragment (from −4,128 to −3,629 bp) in the upstream region of OsWUS to detect possible cis-acting elements using New PLACE software (https:// www.dna.affrc.go.jp/PLACE/?action = newplace) (Higo et al. 1999).

RNA-seq analysis
RNA-seq analysis was conducted using total RNA extracted from leaves of Dap1 mutant and WT plants at the booting stage with three biological replicates. Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. RNA quality was assessed on an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA) and visually checked using RNase-free agarose gel electrophoresis. After total RNA extraction, the mRNA fraction was enriched with Oligo (dT) beads. The enriched mRNA was fragmented into short fragments using fragmentation buffer and then reverse transcribed into cDNA using the NEBNext Ultra RNA Library Prep Kit for Illumina (New England Biolabs, Ipswich, MA, USA). The purified double-stranded cDNA fragments were end repaired, a single base was added to the 3′ ends, and the fragments were ligated to Illumina sequencing adapters. The ligation reaction was purified with AMPure XP Beads (1.0×) and the DNA fragments were subjected to size selection by agarose gel electrophoresis and amplified using PCR. The resulting cDNA library was sequenced on an Illumina NovaSeq 6000 instrument by Gene Denovo Biotechnology Co. (Guangzhou, China).
In order to gain clean reads, the raw reads were filtered using Fastp by removing low-quality reads. We used the Bowtie 2 to remove the read of the rRNA on the comparison without mismatch and used the retained unmapped read for subsequent transcriptome analysis. The resulting high-quality clean reads were mapped to Nipponbare (http://plants.ensembl.org/Oryza_sativa/ Info/Index) reference genome using HISAT 2. The resulting alignment was used Stringtie to reconstruct transcripts. The fragment counts of each gene were normalized by kb of transcript per million mapped reads to obtain the fragment per kilobase million (FPKM). Gene expression levels were estimated using FPKM values by the RSEM.
Gene annotation was performed using the NCBI nonredundant (Nr), Gene Ontology (GO) (http://www.geneontology.org/), and Kyoto Encyclopedia of Genes and Genomes pathway (KEGG) (https://www.kegg.jp/) databases. Differential expression analysis of mutant versus WT was performed using cuffdiff program, differentially expressed genes (DEGs) were identified using the software edgeR in pair-wise comparisons. DEGs among the samples were estimated by referring to the standard of false discovery rate (FDR) < 0.05 & | log2 (fold change) | > 1. Raw sequencing data have been uploaded in the NCBI Gene Expression Omnibus under the accession number PRJNA853805. Gene expression data are available at GEO with the accession number: GSE228728.

RNA extraction and quantitative reverse-transcription PCR
Young panicles at booting stages from Dap1 mutant and the wild type (ZH11) were sampled for qRT-PCR analysis. Young inflorescences (inflorescences length <3 cm) were collected and frozen at −80°C. For RNA extraction, each sample was grounded in liquid nitrogen. One microgram of RNA was subjected for first-strand cDNA synthesis with HiScript III RT SuperMix for qPCR (Vazyme, China). The cDNA products were used for qRT-PCR. Actin1 (Os03g0718100) was used as an internal reference. Three independent biological replicates were performed. All the primers used for qRT-PCR analysis are listed in Supplementary Table 2. The qRT-PCRs were performed using the 7500 Real-Time PCR System (ABI, USA) and ChamQ Universal SYBR qPCR Master Mix (Vazyme, China). The qRT-PCR was conducted with an initial denaturation at 95°C for 3 min followed by 40 cycles of 95°C for 15 s, and 60°C for 30 s.

Statistical analysis
Significance was determined by Student's t-test using Statistical Product and Service Solutions software (v19.0, IBM, USA). Two-sided tests were performed for homoscedastic matrices.

Phenotype of the Dap1 mutant
The Dap1 mutant was identified from a T-DNA insertion population created using the japonica rice variety 'Zhonghua11' (ZH11). In the T 2 generation, the Dap1 mutant and wild-type phenotypes segregated consistent with a 2:1 ratio (heterozygous Dap1:WT = 29:13, χ 2 = 0.107, P > 0.05), and no homozygous Dap1 seedling was identified in the T 2 progeny (see below), suggesting that the mutant phenotype is controlled by a single dominant gene. The heterozygous Dap1 mutant was selected for further phenotypic analysis. The growth and development of heterozygous Dap1 mutant plants are slower than that of WT plants (Fig. 1a). Heading date of the Dap1 mutants was delayed by more than 16 d compared with the WT plants ( Fig. 1c). At the heading stage, the heterozygous Dap1 mutant plants showed a 49.2% reduction in plant height and a severe enclosed-panicle phenotype compared to WT plants ( Fig. 1a, b and d). Tiller numbers of Dap1 mutant plants were decreased by 11.1% compared with WT plants (Fig. 1e). Furthermore, panicle length was significantly reduced by 59.8%, and the number of grains per main panicle in heterozygous Dap1 plants was 71.8% lower than in WT plants ( Fig. 1f and g). We found no significant differences in primary branch numbers between the mutant and WT (Fig. 1h); however, the number of secondary branches was significantly reduced by 98.6% (Fig. 1i) in the Dap1 plants.

A T-DNA insertion located 3,628 bp upstream of the OsWUS ATG is associated with the Dap1 phenotype
To identify the causal gene in the Dap1 mutant, we performed hiTAIL-PCR (Liu and Chen 2007) to recover genomic sequences flanking the T-DNA insertion. Sequencing analysis revealed that the T-DNA was positioned 3,628-bp upstream of the ATG of OsWUS (Os04g0663600), and 4,345-bp downstream of the TGA of a predicted gene called Os04g0663700 (Fig. 2a). The T-DNA insertion also caused the deletion of a 77-bp fragment (from −3,705 bp to −3,629 bp) in the upstream region of OsWUS.
We next performed co-segregation analysis between the T-DNA insertion and the mutant phenotype in the T 2 progeny using specific primers that anneal to the T-DNA (P1) and flanking genomic DNA sequences (P2 and P3). The 29 plants expressing the mutant phenotype were confirmed to be heterozygous (+//−) for the T-DNA insertion, while the other 13 plants with the normal phenotype were wild type (−//−) (heterozygous:WT = 2:1, χ 2 2:1 = 0.107, P > 0.05); however, no homozygous (+//+) plants were identified in the T 2 -generation plants (Fig. 2b). These results demonstrate that the Dap1 mutant trait is due to a single T-DNA insertion and that it is inherited in a dominant fashion, and probably results from disruption or activation of the flanking gene(s).
To determine the causal gene for Dap1, we analyzed the expression of the 2 genes (Os04g0663700 and OsWUS) that flank the T-DNA insertion in leaves from the mutant and WT plants at the booting stage. The results showed that there were no significant differences in the expression levels of Os04g0663700 between WT and Dap1 plants (Fig. 2c). However, the expression level of OsWUS in Dap1 leaves was significantly higher than in WT leaves (Fig. 2d). The distinct expression patterns of OsWUS in the Dap1 mutant compared to WT coincide with the mutant phenotype, which mainly affects plant stature and panicle development.

Analysis of cis-regulatory elements in the upstream region of OsWUS
We reasoned that some cis-regulatory element(s) are probably present near the T-DNA insertion site between the OsWUS and Os04g0663700 genes, and that they are essential for the specific pattern of OsWUS expression. Accordingly, we analyzed the upstream sequence of OsWUS using the online software New PLACE. The results showed that the 500-bp region upstream of the T-DNA insertion site contained a large number of cis-acting regulatory elements probably associated with precise spatial and temporal regulation of OsWUS expression. These included 2 GATA-box motifs, 7 GTGA motifs, 2 RY repeats, 4 CAAT-boxes, 8 CACTFTPPCA1 motifs, and 2 OSE2ROOTNODULE elements (Fig. 3). These results suggest that the altered expression pattern of OsWUS in the Dap1 mutant might be due to a disruption in the genomic sequence integrity upstream of the gene.

Transcriptomic changes between the Dap1 mutant and wild type
To understand how OsWUS affects plant growth and panicle development, we examined its expression pattern using qRT-PCR. Expression pattern of OsWUS in roots, stems, leaves, and inflorescences of wild type were analyzed at the booting stage. The expression of OsWUS in leaves was relatively higher than that in other tissues (Fig. 4a). To elucidate the mechanism underlying the phenotype in the Dap1 mutant, RNA-seq analysis was conducted to evaluate the changes in gene expression induced by the T-DNA insertion in the upstream region of OsWUS. Total RNA was extracted from leaves sampled from Dap1 and WT (ZH11) plants at the booting stage. Three biological replicates were conducted for each sample. DEGs between the Dap1 mutant and WT were estimated by referring to the standard of FDR < 0.05, |log2 fold change| > 1. A total of 925 genes in leaves were found to be differentially expressed in Dap1, with the expression of 532 genes significantly up-regulated and 393 genes down-regulated ( Fig. 4b and c and Supplementary File 1). The DEGs identified between the Dap1 mutant and WT plants were divided into 5 categories based on KEGG pathway enrichment; "metabolism", "genetic information processing", "environmental information processing", "organismal systems", and "cellular processes". "Metabolism" represented the largest group, within which "carbohydrate metabolism" represented the largest subgroup (Fig. 4d). From GO enrichment analysis, the DEGs identified between the Dap1 mutant and WT were classified into the 3 main ontologies "biological process" (BP), "molecular function" (MF), and "cellular component" (CC). The genes in the terms "metabolic process" and "cellular process" in BP, "binding" and "catalytic activity" in MF, and "cell", "cell part", and "membrane" in CC showed more changes in expression (Fig. 5). Short stature and abnormal panicle development are the 2 main characteristics observed in Dap1 mutant plants. We selected genes related to plant stature and panicle  (Arora et al. 2007;Lee et al. 2008). The expression level of OsMADS50 in Dap1 was higher than in the WT (ZH11), while the expression levels of OsMADS55, OsMADS62, OsMADS26, and OsMADS34 were lower than in WT. The expression of other important flowering regulatory genes such as RICE FLOWERING-LOCUS T 1 (RFT1), HEADING DATE 7 (GHD7), and SHORT PANICLE 1 (SP1) was also significantly lower than in WT (Fig. 6). Changes in the expression of these genes may contribute to the aberrant panicle phenotype in Dap1 plants. Plant height is regulated by various factors, and gibberellic acid (GA) is one of the most important determinants of plant height (Sasaki et al. 2002). GA20-oxidase (GA20ox) is a key enzyme that catalyzes the late steps of gibberellin biosynthesis (Sasaki et al. 2002;Asano et al. 2007). The rice "Green Revolution" mutant gene semi-dwarf1 (sd1), which has been widely used in breeding, encodes GA20-oxidase. The sd1 mutant has a semi-dwarf phenotype due to the increased expression of GA20ox2 (Sasaki et al. 2002). GA 2-oxidase plays a key role in the GA catabolic pathway through 2β-hydroxylation and regulates plant growth by inactivating endogenous bioactive gibberellins (Oikawa et al. 2004). The upregulation of the OsGA2ox6 gene may cause dwarfing by decreasing the levels of bioactive GA in the H032 mutant (Huang et al. 2010). Notably, the expression levels of OsGA20ox2 (SD1) and OsGA2ox6 were both found to be up-regulated in Dap1 plants (Fig. 6). Plant cytochrome P450s are heme-binding enzymes with monooxygenase activities that are involved in a wide range of biosynthetic reactions. Plant cytochrome P450s constitute a large class of diverse enzymes that have played crucial roles in plant evolution. Due to their large number and substrate plasticity, the activities of plant P450s are central to many aspects of primary metabolism, including the biosynthesis of plant hormones and growth regulators such as gibberellins, jasmonic acid, auxin, and brassinosteroids (Nelson et al. 1996;Luo et al. 2006). Some of the genes encoding cytochrome P450 monooxygenases have been found to be involved in the regulation of plant height, such as ELONGATED UPPERMOST INTERNODE1 (Eui1) and the CYP96 family (Luo et al. 2006;Ramamoorthy et al. 2011;Xie et al. 2018). The reduced expression level of OsCYP96B4 in Dap1 plants may also contribute to the dwarf phenotype (Fig. 6).

The expression pattern of floral development-related genes in inflorescences changed in Dap1 mutant
To determine the effect mechanism for Dap1, we further analyzed the expression alterations of OsWUS in the mutant and wild-type inflorescences (inflorescence length < 3 cm) at booting stages by qRT-PCR. The expression level of OsWUS in Dap1 inflorescences was significantly higher than WT (Fig. 7a). We further examined the expression of genes related to panicle development in inflorescences. The expression level of RFT1 in Dap1 inflorescences was comparable to WT (Fig. 7b). OsMADS5, OsMADS26, and OsMADS34 expressed in Dap1 inflorescences were elevated compared to WT, while the expression levels of OsMADS55 were lower than WT (Fig. 7c-f). Simultaneously, the expression of SP1 in Dap1 inflorescences was significantly lower than WT (Fig. 7g). TERMINAL FLOWER 1 (TFL1)/CENTRORADIALIS (CEN)-like genes play important roles in determining plant architecture, mainly by controlling the timing of phase transition (Nakagawa et al. 2002). Expression level of RCN1 (rice TFL1/CEN homologs) in Dap1 inflorescences was increased compare to WT (Fig. 7h). Previous studies have shown that the WUS gene regulates growth and development in association with cytokinin signaling (Lu et al. 2015;Wang et al. 2017). CYTOKININ OXIDASE/DEHYDROGENASE 9 (OsCKX9) encodes a cytokinin oxidase to catalyze the degradation of cytokinin, functions as a primary strigolactone-responsive gene to regulate rice tillering, plant height, and panicle size (Duan et al. 2019). We found that the expression level of OsCKX9 in Dap1 inflorescences was significantly increased (Fig. 7i).

Discussion
OsWUS, a rice ortholog of Arabidopsis WUSCHEL (WUS), plays a crucial role in tiller bud formation and development in rice (Lu et al. 2015;Tanaka et al. 2015;Mjomba et al. 2016). There are currently 4 OsWUS loss-of-function mutants reported in rice, they are monoculm 3 (moc3), tillers absent 1 (tab1), sterile and reduced tillering 1 (Srt1), and decreased culm number1 (dc1). moc3 and tab1 are 2 null allelic mutants of OsWUS, which result in partial N-terminal peptides of OsWUS. The Srt1 mutant protein has a deletion of 7 amino acids in the conserved HD of OsWUS, while the dc1 mutation results in a truncated protein that lacks the WUS-box and EAR motif at the C-terminus. These 4 mutants show similar plant phenotypes, such as no or fewer tillers, malformed spikelets, and female sterility. In addition, Srt1 plants show an opposite effect on panicle development compared to the other 3 mutants, and the panicles of Srt1 are larger than those of WT (Lu et al. 2015;Tanaka et al. 2015;Mjomba et al. 2016;Xia et al. 2020). These observations confirm that the loss of OsWUS function has a serious impact on AM formation and its subsequent outgrowth. TAB1/OsWUS plays an important role in the formation of AMs by regulating OSH1 expression (Tanaka et al. 2015). Tanaka and Hirano (2020) further showed that TAB1 is required for maintaining stem cells during axillary meristem development, and FON2 (FLORAL ORGAN NUMBER2) negatively regulates stem cell fate by restricting TAB1. MOC3/ OsWUS is a vital regulator of tiller formation in rice and is able to directly bind the promoter of FLORAL ORGAN NUMBER1 (FON1) and subsequently activate the expression of FON1, the homolog of CLAVATA1 (Shao et al. 2019). In dc1 plants, OsWUS and the auxin action-associated gene ASP1 are both involved in the outgrowth of tiller buds in rice, and a transcription factor that putatively binds to ORYZA SATIVA HOMEOBOX 1 (OSH1) also participates in the regulation (Xia et al. 2020).
In this study, we have identified a T-DNA insertion mutant of OsWUS that differs from the previously reported mutants. In the Dap1 mutant, the T-DNA insertion is flanked by 2 genes, OsWUS and Os04g0663700. There was no significant difference in the expression level of Os04g0663700 between Dap1 and the WT (Fig. 2c). However, the expression level of OsWUS in Dap1 leaves was significantly higher than in leaves of WT plants (Fig. 2d). This indicated that the T-DNA insertion mainly affects the expression of OsWUS, resulting in a different plant phenotype compared to the WT (Fig. 1a). Alteration or disruption of regulatory sequences can change gene expression patterns and produce developmental phenotypes. For example, the multiple copies of the AGATAT element present in the proximal promoter region of the maize ZmWUS1-B gene cause its overexpression, and lead to major rearrangements of inflorescence meristems and mis-regulation of key stem cell regulators (Chen et al. 2021). Sequence analysis showed that the 500-bp region upstream of the T-DNA insertion site harbors 2 RY elements (Fig. 3). The RY elements are binding sites for B3 domain-containing proteins involved in transactivation. The rice B3 domain-containing transcription factor OsGD1, a homolog of Arabidopsis VAL proteins, suppresses OsLFL1 expression by binding to the RY element in the OsLFL1 promoter and is associated with seedling development and GA homeostasis in rice (Guo et al. 2013). Moreover, the B3 transcriptional repressors OsGD1 and OsVAL2 bind to the RY-containing cis-silencing element (SE1) in the first intron of the rice GA-deactivating enzyme gene Eui1 and recruit a trans epigenome reader that represses Eui1 expression and modulates GA homeostasis. Deletion of SE1 elevates the expression of Eui1 in the dEui1 mutant, thus leading to GA deficiency and dwarfism (Xie et al. 2018). The Dap1 mutant exhibited a dwarf phenotype and abnormal panicle development, which is similar to gd1 and dEui1. In summary, we conclude that the altered expression pattern of OsWUS in Dap1 is probably due to a disruption in the genomic sequence integrity upstream of the gene. Analysis of RNA-seq data identified a total of 925 genes that were differentially expressed in leaves between the Dap1 mutant and WT ( Fig. 4a and b). In the KEGG pathway enrichment analysis, "metabolism" represented the largest group, in which "carbohydrate metabolism" represented the largest subgroup (Fig. 4c). From the GO enrichment analysis, the largest number of DEGs were in the "metabolic process" term in the "BP" category, "binding" in "MF", and "cell" in "CC" (Fig. 5). The panicle architecture of Dap1 plants was characterized by short, malformed panicle with little seed set (Fig. 1b). Several genes related to floral development were abnormally expressed in Dap1 plants. OsMADS50 plays an important role in regulating flowering time in rice. Inhibiting the expression causes late flowering, while ectopic expression of OsMADS50 causes early flowering (Lee et al. 2004). OsMADS34 [also called PANICLE PHYTOMER2 (PAP2)] is induced in the SAM during the transition from vegetative to reproductive development and is also induced when glumes are initiated during spikelet development (Kobayashi et al. 2012). The decrease in the expression of OsMADS34 in Dap1 may affect the initial development of the panicle. At the same time, OsMADS34 acts immediately downstream of RFT1 (Pasriga et al. 2019). RFT1 is a major florigen, and its function is to induce reproductive development of the SAM. The expression of both RFT1 and OsMADS34 was also reduced in the Dap1 mutant (Fig. 6). GHD7 has a key role in photoperiodic flowering by regulating the putative Ehd1-Hd3a pathway and has positive regulatory effects on an array of traits in rice, including the number of grains per panicle, plant height, and heading date under long-day conditions (Xue et al. 2008). Although rice is a short-day plant, its domestication led to the Ghd7-Ehd1-Hd3a/RFT1 pathway for adaptation to long-day conditions (Peng et al. 2021). We showed that GHD7 and RFT1 are both down-regulated in Dap1 leaves (Fig. 6), and these may contribute to the delayed flowering phenotype in Dap1 mutants ( Fig. 1a and c).
Expression pattern of genes related to floral development in inflorescences were further determined. The expression levels of the MADS-box family genes in inflorescences were significantly changed between Dap1 and WT (Fig. 7). OsMADS5 and OsMADS34 expressed in Dap1 inflorescences were higher than that in WT (Fig. 7c and e). OsMADS5 and OsMADS34 play similar functions in limiting branching and promoting the transition to spikelet meristem identity. The number of primary branches per panicle in OsMADS5 and OsMADS34 overexpression lines was comparable to WT plants; however, the number of secondary branches and spikelets decreased significantly (Zhu et al. 2022). Here, we observed similar phenotypes in Dap1 mutant ( Fig. 1h and i). The expression of OsMADS34 in leaves was reduced (Fig. 6), but this expression in inflorescences was significantly increased in the Dap1 mutant (Fig. 7e). It seems that the mutation delays the initial vegetative to inflorescence meristem transition ( Fig. 1a and c), but later accelerate the transition from branching to spikelet meristems, resulting in less secondary branching ( Fig. 1b and i). The architecture of the rice inflorescence, which is determined mainly by the number and length of primary and secondary inflorescence branches. SP1 encodes a PTR family transporter and determines rice panicle size (Li et al. 2009). OsCKX9 plays a critical role in regulating rice shoot architecture, and OsCKX9-overexpressing transgenic plants showed significant decreases in plant height and panicle size (Duan et al. 2019). In 35S::RCN1 transgenic rice plants, the delay of transition to the reproductive phase was observed and the transgenic plants exhibited a more branched, denser panicle morphology (Nakagawa et al. 2002). We found that expression level of OsCKX9 and RCN1 both increased in Dap1 plants ( Fig. 7h and i).
The T-DNA insertion in the Dap1 mutant causes the overexpression of OsWUS in leaves and panicles, and coordinated regulation of these genes contributes to the abnormal panicle development in Dap1 plants.
In previous reports, the expression of OsWUS in association with cytokinin signaling was shown to regulate tiller development in rice (Lu et al. 2015;Wang et al. 2017). In our study, we first found that the overexpression of OsWUS can regulate plant stature in association with gibberellic acid biosynthesis. GA 20-oxidase (GA20ox) is a 2-oxoglutarate-dependent dioxygenase and is a key enzyme in the biosynthesis of gibberellins (Oikawa et al. 2004). The mutant gene semi-dwarf1 (sd1), which is known as the "Green Revolution" gene, encodes GA 20-oxidase and is one of the most important genes used in rice breeding. The sd1 mutation affects the late steps in gibberellin metabolism, leading to an accumulation of the initial substrate of GA20ox, low gibberellin levels, and a semi-dwarf phenotype (Sasaki et al. 2002;Spielmeyer et al. 2002). Gibberellin (GA) 2-oxidase plays a key role in the GA catabolic pathway through 2β-hydroxylation (Huang et al. 2010). GA2ox6 is localized to the nucleus and cytoplasm, and the mRNA can be detected in young seedling leaves and transiently at high levels during the active tillering stage, which mainly affects plant stature (Lo et al. 2008;Huang et al. 2010). In our study, the expression levels of OsGA20ox2 (SD1) and OsGA2ox6 in the Dap1 mutant were both elevated compared to WT (Fig. 6). In addition, expression of OsCYP96B4 was lower in Dap1 plants than in WT (Fig. 6). A point mutation in the SRS2 domain of CYP96B4 resulted in a dwarf phenotype in the sd37 mutant (Zhang et al. 2014). Another CYP96B4 mutant, dss1, also shows a dwarf phenotype, and this phenotype is regulated by finetuning the GA-to-ABA balance (Tamiru et al. 2015). In addition, the bioactive forms of GA (GA 1 and GA 4 ) could not be detected in dss1 mutant, but the levels of the GA precursors (GA 19 and GA 53 ) were significantly reduced (Tamiru et al. 2015). These results show that the deficiency of OsCYP96B4 leads to the decrease of accumulation of bioactive GAs. Accumulation of bioactive GAs triggers GA signaling pathways and inhibits GA biosynthesis, whereas deficiency of bioactive GA upregulates GA biosynthetic enzymes and inhibits GA catabolic enzymes via feedback regulation (Fleet and Sun 2005). Therefore, the feedback regulation of deficiency of bioactive GA may contribute to higher expression of OsGA20ox2 (SD1) in Dap1 mutants. Fig. 7. Expression levels of floral development related genes in inflorescences by qRT-PCR analyses at booting stage. Actin1 as an internal reference. WT, wild type. Inflorescence length was <3 cm. a) OsWUS; b) RFT1; c) OsMADS5; d) OsMADS26; e) OsMADS34; f) OsMADS55; g) SP1; h) RCN1; i) CKX9. Data are presented as the mean ± SD; n = 3; * P < 0.05; ** P < 0.01, according to Student's t-test.
ConclusionsIn this study, we identified a T-DNA insertion mutant that causes the overexpression of OsWUS, which resulted in some phenotypes similar to that of a loss-of-function OsWUS mutant. Furthermore, our results indicate that OsWUS is an essential panicle development regulator in rice, and the abnormal expression of OsWUS mutation is associated with the gibberellin signaling pathway to regulate plant stature.

Data availability
The data that support the findings of this study are available in the Supplementary material of this article. Supplemental data are available at G3 online. Supplementary Files 1 and 2 contain DEGs between the Dap1 mutants and WT plants by RNA-seq analysis. Raw sequencing data have been uploaded in the NCBI Gene Expression Omnibus under the accession number PRJNA853805. Gene expression data are available at GEO with the accession number: GSE228728.
Supplemental material available at G3 online.