Identification of An7 as a positive awn regulator from two wild rice species

Abstract

The awn is a bristle-like appendage that protrudes from the seed tip and plays a critical role in preventing feed damage and spreading habitats in many grass species, including rice. While all wild species in the Oryza genus have awns, this trait has been eliminated in domesticated species due to its obstructive nature to agricultural processes. To date, several genes involved in awn development have been identified in wild rice, Oryza rufipogon and Oryza barthii which are ancestral species of cultivated rice in Asia and Africa, respectively. However, the responsible genes for awn development have not been identified in other wild rice species even though multiple QTLs have been reported previously. In this study, we identified An7 gene responsible for awn development in two wild rice species, Oryza glumaepatula and Oryza meridionalis. An7 encodes a cytochrome P450 enzyme and is homologous to D2/CYP90D2, a known brassinosteroid biosynthesis enzyme in rice. The identification of An7 provides insight into a distinct molecular mechanism underlying awn development that occurs in geographically separated environments.

Introduction

Wild rice retains a variety of morphological, physiological, and ecological characteristics that make it an important genetic resource. For example, Oryza longistaminata, a wild African rice species, has strong bacterial blast resistance and Xa21 gene contributes to this ability (Song et al. 1995). The O. longistaminata allele of Xa21 has been introduced into a cultivated rice variety to accomplish the high blast resistance (Nguyen et al. 2018). On the contrary, traits considered to be detrimental in agriculture have been selected and eliminated during the rice domestication. Awn is the one example of eliminated traits in rice. Awn is a bristle-like appendage at the tip of lemma and helps to avoid feeding damage and seed dispersal. All wild rice species have awns, however, cultivated species have been lost this trait. Several genes responsible for awn development have been selected during rice domestication. In Asia, loss of function of three genes were identified as the main causes for the loss of awn during rice domestication. An-1 (Luo et al. 2013), which is also detected as Regulator of Awn Elongation 1 (RAE1) (Furuta et al. 2015) encoding bHLH transcription factor, LABA1 which encodes cytokinin activating enzyme (Hua et al. 2015), and RAE2 (Bessho-Uehara et al. 2016), which is also annotated as GAD1 (Jin et al. 2016) encoding OsEPFL1 signal peptide were selected during Asian rice domestication. On the other hand, in Africa, single gene named RAE3 which encodes an E3 ubiquitin ligase was identified as the main responsible gene for the loss of awn (Bessho-Uehara et al. 2023). Among these genes, An-1/RAE1 and RAE2/GAD1 genes are commonly conserved in most wild rice species according to comparative analysis of several chromosome segment substitution lines (CSSLs) (Bessho-Uehara et al. 2021). Through the genetic analysis conducted by Ikemoto et al., it was revealed that An-1/RAE1 and RAE2/GAD1 complement awn phenotype around 70% of O. rufipogon in the O. sativa genetic background (Ikemoto et al. 2017). However, interestingly, the backcross line carrying An-1/RAE1 and RAE2/GAD1 genes from O. sativa (dys-functional alleles) in O. rufipogon background, it was observed that the awn length remained long, approximately 70% of that of O. rufipogon (Ikemoto et al. 2017). These findings indicate An-1/RAE1 and RAE2/GAD1 possess large effects on awn elongation, vice versa, multiple other genes with small effects that work in concert in awn elongation.

More than ten QTLs related to rice awn have been identified (Cai and Morishima 2002, Fawcett et al. 2013), but only some responsible genes have been identified and the molecular mechanism elucidated (Bessho-Uehara et al. 2016, 2023, Furuta et al. 2015, Hua et al. 2015, Jin et al. 2016, Luo et al. 2013, Toriba and Hirano 2014). To understand the comprehensive molecular mechanism of awn development, we need to identify more responsible genes of awnness in rice.

Previous studies on genes associated with awn development have mainly relied on analyses using domesticated rice and its wild ancestor species. However, since there are 22 wild rice species, all of which possess long awns, it was hypothesized that new genes involved in awn elongation could be discovered. Previous researches using inbred lines between O. sativa and either O. meridionalis or O. glumaepatula suggested the involvement of genes named An7 and An8 in awn elongation (Kurakazu et al. 2001, Matsushita et al. 2003a, 2003b). The gene An8 on chromosome 4 is presumed to be An-1/RAE1 according to its locus position. However, An7 on chromosome 5 has not been identified. In this study, we employed inbred lines to narrow down the An7 gene to shed light on its contribution to the complex genetic control of awn development.

Materials and Methods

Plant materials and growth conditions

O. sativa ssp. japonica ‘Taichung 65 (T65)’, O. glumaepatula acc. IRGC105668, O. meridionalis acc. W1625, and two lines retrieved from chromosome segment substitution lines (CSSLs) were employed for evaluation of awn phenotype. Two CSSL lines, Glu-IL115 and Mer-IL116 were used to map An7 gene that controls awn development. Glu-IL and Mer-IL are CSSLs that carry chromosome segments from O. glumaepatula and O. meridionalis, respectively, in T65 genetic background (Yoshimura et al. 2010). Seeds of all lines are provided from Kyushu University. Glu-IL115 harbors chromosome 1 (25,071,234-31,948,582 bp) and chromosome 5 (5,284,924-21,426,063 bp) segments substitutions from the donor parent, O. glumaepatula acc. IRGC105668. Mer-IL116 harbors a chromosome 5 (2,881,341-16,811,499 bp) segment substitution from the donor parent, O. meridionalis acc. W1625. These plant materials were grown either in the greenhouse at Nagoya University, Furo-cho, Nagoya, Aichi or in the research field of Nagoya University, Togo, Aichi, Japan with conventional agricultural calendar under natural conditions. Seedlings were first grown in the greenhouse for about 30 days and then transplanted in the greenhouse or the field. The transgenic plants were grown in isolated greenhouses under long-day conditions until the 10-leaf stage, and then transferred to short-day conditions until flowering.

Evaluation of morphological phenotypes

Panicles of the parental lines, CSSLs and transgenic lines were harvested after seed maturation for evaluation. Five to ten plants were randomly selected for phenotypic evaluation. The length of the awn was taken as the distance from the tip of lemma to the end of awn. The length of the awn was measured by a ruler on the first spikelet located on the primary branch of each panicle. The average values of at least five panicles were calculated for each plant. Five hundred grains were used for measuring grain length and grain width by ImageJ (version 1.51n). Fully filled grains were used to measure 1,000-grain weight, and five replicates were performed. The statistical phenotypic data was analyzed using student t-test for significant difference.

DNA extraction and genotyping of the An7 mapping population using SSR markers

For linkage analysis of An7, we used 960 BC₄F₃ plants produced by inbreeding BC₄F₂ plants of Glu-IL115 and Mer-IL116 whose chromosome 5 substitution regions are heterozygous, respectively. Segregation ratio of BC₄F₃ was evaluated by chi-square test to calculate the deviation from Mendelian ratios. Genomic DNA from the mapping population was extracted following the TPS method (Hattori et al. 2009). The polymerase chain reaction (PCR) analysis was performed in a 10 μL reaction mixture containing 50 mM KCl, 10 mM Tris-HCl (pH 9.0), 1.5 mM MgCl₂, 200 μM of each dNTP, 0.2 μM of each primer, 0.75 U of Taq polymerase (Takara, Otsu, Japan), and approximately 10 ng of template DNA in a thermal cycler (cat. 9700N8050200, Applied Biosystems, Foster City, CA, USA). The PCR program was 95°C for 5 min for initial denaturation, followed by 35 cycles at 95°C for 30 s, 55°C for 30 s and 72°C for 30 s, with no final extension. PCR products were run in 3% agarose gels in 0.5× Tris-borate-EDTA (TBE) buffer containing ethidium bromide. Primer sequence information of single sequence repeat (SSR) markers is shown in Table 1.

Table 1.Primers used in this study

Purpose	Primer name	Sequence	Physical position (bp)
An7 rough mapping marker	RM6320F	CGCTACGGAAGGGTAATAATGC	471,659
	RM6320R	AGCGTGGGAAGAAGGATACACC	471,838
	RM592F	ACCTCACCCGAATTACTGTGATATGC	2,796,922
	RM592R	GTTGAATTGCACGCGACTCTGG	2,797,118
	RM3328F	ATTGTCGAAGCAGCAAGAAAGG	5,266,943
	RM3328R	GACAATTGGAGTGTTGGACATGG	5,267,025
	RM3419F	TGCTGCTATTCCTCAAGACAAACC	5,284,902
	RM3419R	CTTGGTGAAACAGTGCTCTCTGG	5,285,075
	RM18055F	AGATCTCCTCTCAGAGTCTACCG	5,877,862
	RM18055R	CACTGAGTATAATCCCTGCAACC	5,878,157
	RM18059F	CGGAGGGTATGATCCGAAGAGG	5,965,605
	RM18059R	GCTGTGGTTGCTGCTGTTGTAGC	5,965,757
	6.097F	TGTGATGCTCTCCTATGCTC	6,097,290
	6.097R	GTTACCGAAAAGCAGTGGAC	6,097,394
	6.139F	CGATTCAGTGACCAATTTGC	6,138,951
	6.139R	TATGCGGATCTCTTCTCCTT	6,139,132
	6.172F	ACGTATAGCCGCTACTTTGG	6,172,945
	6.172R2	CCGGTTTTCGTCTATAACAT	6,173,074
	6.213F	GACTCGTGATGAACGAACTC	6,213,247
	6.213R	GTGCGCTATATATCTGCAAC	6,213,352
	6.223F	CAAGCTTGCAAGTAATTACG	6,222,952
	6.223R	CATTTTAATCAATTCCCTTTTG	6,223,012
	6.24F	GTAAGAATGTTGGTATCCAAC	6,246,212
	6.24R	GATTAAACCTATGTCCATGC	6,246,314
	6.324F	GGATCCACGCGCCTGGA	6,323,632
	6.324R	GACCCAACCAAGCAGAGAC	6,323,692
	RM7293F	CCTAGGGGATCCAAGATGTC	7,526,972
	RM7293R	GCACGGATCTACATACATGC	7,526,829
	RM5844F	AACGTGGCATCCATGTTAGTACC	9,149,506
	RM5844R	AGCTAGGAGCCATTGTCGAAGG	9,149,696
	RM5140F	GGCACTCGTATTTCTCAACTTCTCC	13,541,662
	RM5140R	GGGTGTATCAGGAGTACAGGTTGC	13,541,917
	RM18382F	GGAATTAAATGTGCGGGAATGC	14,824,645
	RM18382R	TGTAAGTACAAATCCGGCACCTATGG	14,824,842
	RM18550F	CCGAGATTGCAAGAATGAAGACAACC	18,080,743
	RM18550R	GCTGGCCATCGAGATGTTCG	18,080,963
	RM18551F	CGTGGCGAATAAATAGGCGAGAGG	18,081,391
	RM18551R	TGACCCTCTCCTGCCACCTACG	18,081,491
An7 fine mapping marker O. glumaepatula	6.254_4_Hinf1_F	CAATGTGTCGGCAATGTCGC	6,254,284
	6.254_4_Hinf1_R	CAACCCGACCCATATTAGCC	6,254,536
	6.258_3_Msp1_F	AGTATGGGCCGCACATTAGG	6,258,155
	6.258_3_Msp1_R	TGCAAAATCAAGCAGGCAACA	6,258,313
	6.264F	GACTGGTTCTAATTGCGAGC	6,264,178
	6.264R	CCCTTAGCGCTGCATCAAAT	6,264,275
	6.267F	GGGCTTTTGCTTTAAAGAAG	6,267,114
	6.267R	GATAGAAACAGCTACAATGG	6,267,227
	6.291F	GAAAGCACGCAGCATATATAC	6,291,470
	6.291R	GCTCAAATAGCTAGTATTCCTAGTTC	6,291,548
	6.306F	CTTGCCACTGGTCAGACTAG	6,205,606
	6.306R	GTGAATGTAGGTTGTATGTGTTAAAG	6,305,670
An7 fine mapping marker O. meridionalis	RM18058F	CGGTCCTCCACCTCAAGTCC	5,873,620
	RM18058R	CCCAGGGTAATGCAAGTGTAGC	5,873,639
	6.264F	GACTGGTTCTAATTGCGAGC	6,264,178
	6.264R	CCCTTAGCGCTGCATCAAAT	6,264,275
	RM18080F	GCGAAACTCATCGACTGATCG	6,323,956
	RM18080R	GGTGATCACATCATTGTCCATCC	6,323,979
	RM18076F	ACATGTCCCTGCCCGTAAACC	6,332,999
	RM18076R	GTCCACTCCAGCTCCTGCTTCC	6,333,162
	RM18107F	CGTATGGACTTGCCTTGAGTCG	6,827,017
	RM18107R	TCCAATCTGCCAAGCTTTACACC	6,827,311
An7 candidate cloning	LOC_Os05g11120_Glu-IL115_F	TACTGTCGGCTGACCATGTTTTCTT
	LOC_Os05g11120_Glu-IL115_R	ACCCTTGTTTGGGACTTTAGGGACT
	LOC_Os05g11130_Glu-IL115_F	GCACCTTACCGATGTTCACCATCTACAT
	LOC_Os05g11130_Glu-IL115_R	TGTCCGCCTTGCTATTCTGTACACATTT
	LOC_Os05g11140_Glu-IL115_F	ACAGTCTAGACCCGGGATGGACGACGTCGACAGCG
	LOC_Os05g11140_Glu-IL115_R	TGGCTGCAGGTCGACGTCATGACACCGTTCTTCCA
	LOC_Os05g11160_Glu-IL115_F	ACACTATCGATCCACATTTGTTTTGAA
	LOC_Os05g11160_Glu-IL115_R	TATGTCTTATCATCAACCCCATAACCA
An7 sequence primer	GLUM06720ORF seq R7.2-	TGCTCAGTACCATTGCAGTG
	GLUM06720ORF seq R7-	AGTATGAGTACCATTTTCGTC
	GLUM06720ORF seq F6-	GTTTGTGGCTTGGTGATGTAG
	GLUM06720ORF seq R6-	TTACTACATCGATCCGTTTC
	GLUM06720ORF seq F5-	AGTTGGTGCTTACTCTATCC
	GLUM06720ORF seq R5-	CTCCATTTCTTTTACGTTGTC
	GLUM06720ORF seq F4-	CTGTTAGCAAAACAACACTAC
	GLUM06720ORF seq R4-	TCCAGCTTTTGGAAGAGAAG
	GLUM06720ORF seq F3-	CTTTGTTCACTATTGGTTGAC
	GLUM06720ORF seq R3-	GACACTTATAAACGGACAAACG
	GLUM06720ORF seq F2-	GAATATGAGGAGTAGCTTGC
	GLUM06720ORF seq R2-	CATCCTCTTCTTAGCCTAGATG
	GLUM06720ORF seq F1-	TACCCACATGACAACCTCTC
	GLUM06720ORF seq R1-	GTGCATGGGTCAACAGAATAG
	GLUM06720ORF seq R1	ATTAGCTTAAGGAATCAGGACTTGGTC
	GLUM06720ORF seq F2	GATATTCCAAGATTCACCCTACCGTATG
An7 qRT primer	An7qRT_F	AGACGCTGAGTTCATCTCC
	An7qRT_R	TCCCCCATCAGCTCCATC
	UBQ_F	GAGCCTCTGTTCGTCAAGTA
	UBQ_R	ACTCGATGGTCCATTAAACC
RAE3 CRISPR gRNA	RAE3_CRISPR1_gRNA1	TGGAACCCATGGCGGCATTTCGG
	RAE3_CRISPR1_gRNA2	CACCACGGCGCGGTCGACGCCGG
	RAE3_CRISPR1_gRNA3	GGGATGCTTCGAGGCCGCCAAGG
	RAE3_CRISPR2_gRNA1	TGGAACCCATGGCGGCATTTCGG
	RAE3_CRISPR2_gRNA2	TGGGTTCCACGTCGAGTGCGTGG
	RAE3_CRISPR2_gRNA3	CTCGAGTGCGCGGTGTGCCTCGG

Plasmid construction and generation of transgenic plants

For complementation test, the An7 candidate genes, LOC_Os05g11120, LOC_Os05g11130, and LOC_Os05g11160, were amplified by PCR using Glu-IL115 genomic DNA as templates and LOC_Os05g11140 was amplified by PCR using Glu-IL115 complementary DNA as a template. Primers were designed in a region about 3-kb upstream and 2-kb downstream to contain the promoter and terminator for LOC_Os05g11120, LOC_Os05g11130, and LOC_Os05g11160. The amplified fragment was introduced into the pCAMBIA1380 vector and written as own promoter::An7-candidate gene ^Glu-IL115/pCAMBIA1380. Insertion of the fragments was confirmed by Sanger sequencing using ABI3730 (Thermo Fisher Scientific). For LOC_Os05g11140, its amplified fragment was integrated into the ΩpCAMBIA1380 vector and written as Ubiquitin::An7-candidate gene ^Glu-IL115/ΩpCAMBIA1380. These constructs were introduced into callus of T65, which is an awnless line, using Agrobacterium tumefaciens (EHA105)-mediated transformation method (Hiei et al. 1994). The line transformed empty vector of pCAMBIA1380 is used as a vector control. For producing RAE3 mutation in Glu-IL115, a construct in pMgPoef4_129-2A-GFP vector backbone with gRNA targets designed on three locations of RAE3 was generated according to the previous manuscript (Mikami et al. 2015). Using A. tumefaciens (EHA105)-mediated transformation, the construct was introduced into callus of Glu-IL115, which is an awnled line. pMgPoef4_129-2A-GFP, a vector without guide RNA sequences, was used as a vector control. Primers used to make constructs are listed in Table 1.

Gene annotation, sequencing, and amino acid alignment

The reference genome of O. glumaepatula and O. meridionalis was retrieved as GCA_000576495.1 (O. glumaepatula acc. GEN1233_2) and GCA_000338895.2 (O. meridionalis acc. W2112) from Gramene (https://www.gramene.org/). The short reads of IRGC105668 (O. glumaepatula) and W1625 (O. meridionalis) were retrieved from DRR057991 and DRR058014 respectively, registered in NCBI SRA database. MEGANTE (Numa and Itoh 2014) was used for gene prediction in O. glumaepatula and O. meridionalis. The Rice Genome Annotation Project (http://rice.uga.edu/) was used for O. sativa gene annotation. Four candidate genes were amplified with Prime Star GXL (Takara) using T65 and Glu-IL115 genomic DNA as templates. Then, PCR products are sequenced with BigDye Terminator (Thermo Fisher Scientific) and ABI3730 (Thermo Fisher Scientific). Based on the CDS sequence we acquired, amino acid sequence was predicted comparing with the CDS sequence of LOC_Os05g11130. The multiple sequence alignment of An7 protein sequences in O. sativa, O. glumaepatula and O. meridionalis were performed by MUSCLE program using Geneious software (version 9.0) with default settings.

Visualization of local genome synteny around the candidate region of An7

To visualize differences in genome structure around the candidate region of An7, the genome sequences of O. sativa, O. glumaepatura and O. meridionalis were compared. For the candidate region of An7, a specific segment (positions 6,254,283 to 6,305,695 bp) on chromosome 5 of O. sativa was retrieved from Nipponbare genome sequence IRGSP-1.0, distributed via the RAP-DB (https://rapdb.dna.affrc.go.jp/). The corresponding genomic regions in O. glumaepatula and O. meridionalis, aligning with the An7 candidate region in O. sativa, were identified based on BLAST hits of the first and last 100 bp of the O. sativa subset sequence. The reference sequences of O. glumaepatura and O. meridionalis were obtained as described above. The blastn program of bl2seq was performed using GenomeMatcher ver. 2.3 (Ohtsubo et al. 2008) with default parameter. The local genomic sequence corresponding to the candidate region of An7 was visualized using GenomeMatcher, with adjustments made to exclude lines that connected sequences showing relatively low similarity (less than 70%).

Constructing phylogenetic tree

Amino acid sequences of An7 homologues from various plant species were identified through reciprocal best-match BLAST searches using Phytozome v13 database (https://phytozome-next.jgi.doe.gov). Accession numbers and locus identifications were derived from Phytozome v13. Amino acid sequences were aligned using the MUSCLE program implemented in Geneious software version 6.1.6 (Biomatters; http://www.geneious.com/) with default parameters. The phylogenetic tree was constructed by the maximum likelihood method using PhyML 3.0 (Guindon et al. 2010). Branch nodes were labeled with a cutoff at 500, using bootstrap values calculated from 1000 replications.

RNA isolation and quantitative Real-Time (qRT) PCR

For qRT PCR analysis of An7, young panicles (length is less than 5 mm) and spikelets which are in the stage of before heading (panicle length is around 10 cm) of T65, Glu-IL115 and Mer-IL116 were used. Total RNA was extracted by RNeasy Plant Mini Kit (QIAGEN), then first-strand cDNA synthesis was performed using the Omniscript RT Kit (QIAGEN). StepOneTM Real-Time PCR system (Applied Biosystems) was used to analyze the expression levels of An7. Relative expression levels of the target genes were normalized to the levels of endogenous ubiquitin transcripts (LOC_Os02g06640). The experiment was repeated three times and the Comparative CT method (ΔΔCT Method) was used to calculate the relative expression levels of the target genes. The error bars display the calculated maximum (RQMax) and minimum (RQMin) expression levels that represent standard error of the mean expression level (RQ value).

Results

Phenotypic evaluation of CSSLs and their parents

Previous studies using the inbred lines between O. sativa and O. glumaepatura or O. meridionalis had shown that An7 is located on short arm of chromosome 5 of the respective wild rice species (Kurakazu et al. 2001, Matsushita et al. 2003a, 2003b). It was estimated that the genes on chromosome 5 detected in these two species were identical but with different alleles according to the overlap of gene location. Prior to identification of An7 in wild rice O. glumaepatula and O. meridionalis, awn phenotype of CSSLs and their parents was evaluated. O. glumaepatula acc. IRGC105668 formed a 56.9 mm long awn on average, while O. sativa ssp. japonica ‘T65’ formed no awn (Fig. 1A, 1B, 1D). The CSSL Glu-IL115, which had a substitution of a portion of its chromosomes 1 and 5 with those of O. glumaepatula, formed an awn of approximately 10 mm in length (Fig. 1C, 1D). However, other CSSLs, namely Glu-IL103 and Glu-IL107, which harbored a segment substitution in the corresponding region on chromosome 1 from O. glumaepatula, did not form awns. This led us to conclude that the segment on chromosome 5 from O. glumaepatula was responsible for awn elongation in Glu-IL115. W1625 which is an accession of O. meridionalis formed awns that reached 80.4 mm long (Fig. 1E, 1F, 1H), while the CSSL, Mer-IL116, which had a substitution of a segment of its chromosome 5 with that of O. meridionalis, formed an awn of approximately 25.2 mm in length (Fig. 1G, 1H). These results indicated that both alleles of An7 have positive effect on awn development.

Fig. 1.

Plant materials used in this study and their awn phenotype. (A–D) Phenotypic comparison of awns in South American wild rice, O. glumaepatula acc. IRGC105668 (A), O. sativa ssp. japonica ‘T65’ (B) and Glu-IL115 (C). Awn length of O. glumaepatula, T65, and Glu-IL115 (D). (E–H) Phenotypic comparison of awns in Oceanian wild rice, O. meridionalis acc. W1625 (E), T65 (F) and Mer-IL116 (G). Awn length of O. glumaepatula, T65, and Mer-IL116 (H). Twelve rectangles below each photo represent the rice chromosomes with colors indicating species; blue, O. glumaepatula; green, O. meridionalis; gray, O. sativa. Left panels display whole plants (bars = 50 cm) and right panels show seeds and awns (bars = 1 cm). Awn length data were measured from 6 to 15 seeds, with error bars representing mean ± SD.

Mapping of An7 as responsible gene for awn elongation in O. glumaepatula and O. meridionalis

To investigate the inheritance pattern of An7, we observed an awn phenotype in a randomly selected population of approximately 100 individuals from the BC₄F₃ generation. Based on the results of Glu-IL115 and Mer-IL116, the segregation of awned versus awnless individuals followed a ratio of 72:24 and 73:21, respectively. These ratios showed a good fit to the expected 3:1 ratio, as indicated by the chi-square test, suggesting a dominant inheritance pattern for An7.

To map An7 in O. glumaepatula and O. meridionalis, we used 960 lines from the BC₄F₃ populations of Glu-IL115 and Mer-IL116, respectively. Through genotyping and phenotyping, we identified 62 recombinant lines with awn and 20 recombinant lines without awn in the Glu-IL115 progenies, which narrowed down the candidate region of An7 in O. glumaepatula to approximately 52 kb on the short arm of chromosome 5 (Fig. 2A). Similarly, in the Mer-IL116 progenies, we found 93 awned recombinant lines and 54 awnless recombinant lines, resulted in the candidate region of An7 in O. meridionalis to approximately 954 kb on the short arm of chromosome 5 (Fig. 2B). These candidate regions overlapped, indicating that An7 in O. glumaepatula and O. meridionalis may be identical. Within the 52 kb candidate region in Glu-IL115, there were 5 genes, LOC_Os05g11120, LOC_Os05g11130, LOC_Os05g11140, LOC_Os05g11150, and LOC_Os05g111600, located in Nipponbare (O. sativa ssp. japonica) genome in Rice Genome Annotation Project (http://rice.uga.edu/, searched on March 29, 2022). Among them, four genes were predicted within the 52 kb candidate region in O. glumaepatula by MEGANTE (https://megante.dna.affrc.go.jp/, searched on March 29, 2022). One of 5 genes annotated in O. sativa, LOC_Os05g11150, was not predicted in O. glumaepatula. Considering the result of awn phenotype segregation and the functional presence of An7 in O. glumaepatula, LOC_Os05g11150 was excluded as a candidate gene for An7 in O. glumaepatula. Thus, we narrowed down the candidate genes for An7 into four genes in O. glumaepatula (Fig. 2A).

Fig. 2.

Mapping of An7. (A, B) Mapping results of An7 using progenitor of Glu-IL115 (A) and Mer-IL116 (B). White bars represent chromosome 5 of T65 (O. sativa). Blue and green bar represent chromosome segment of O. glumaepatula and O. meridionalis, respectively. Black vertical lines denote DNA marker locations, and the numbers corresponding to marker names. Blue arrows indicate four genes annotated within 52 kb of candidate region in Glu-IL115 by MEGANTE.

By comparing the regional genomic synteny, we aligned the genomic sequence of 52-kb candidate region among O. sativa, O. glumaepatura and O. meridionalis (Fig. 3A). We found few SNPs between O. glumaepatula and O. sativa in the candidate region, in addition to a 10-kb deletion between LOC_Os05g11120 and LOC_Os05g11130. On the contrary, there are many in-dels between O. meridionalis and O. sativa in this region. To identify the functional An7 allele from Glu-IL115 that induces awn elongation in the genetic background of O. sativa ssp. japonica ‘T65’, a complementation test was conducted. Genomic fragments, including the promoter and terminator regions of each gene, were cloned from the genomic DNA of Glu-IL115 (Fig. 3B). Among the transgenic lines, only those carrying LOC_Os05g11130^Glu-IL115 showed complementation of the awn phenotype with an average awn length of 5.8 mm (Fig. 3C, 3D). Based on these results, we concluded that LOC_Os05g11130^Glu-IL115 is the responsible gene for An7.

Fig. 3.

Complementation test of An7 candidate genes. (A) Regional synteny around the An7 candidate region on chromosome 5 compared among O. glumaepatula, O. sativa, and O. meridionalis. A color scale for percent identity shown on the right. Lower case letters, a to d, indicate genes; a, LOC_Os05g11120; b, LOC_Os05g11130; c, LOC_Os05g11140; d, LOC_Os05g11160. (B) Genomic regions utilized for the complementation test. Pentagons indicate gene body and black horizontal lines above pentagons reoresenting promoter and terminator regions of each gene. (C) Awn lengths of the transgenic lines used for complementation test. We used T65 as the recipient for the transgenes. Error bars represent mean ± SD. (D) Awn phenotypes of complementation test lines. VC, vector control; a to d indicate the genes described above. Scale bars = 10 mm.

An7 encodes cytochrome P450 enzyme and locates in the sister clade of D2

According to the sequence, An7 encodes cytochrome P450 CYP90D3 gene. To elucidate the An7 mutation for loss-of-function in O. sativa, we compared the amino acid sequence of An7 among O. sativa, O. glumaepatula and O. meridionalis. Two non-synonymous O. sativa specific mutations are present (Fig. 4). Considering that almost the entire portion of An7 encodes a cytochrome P450 domain (14L to 474F), these two mutations could cause a decrease in affinity for its substrate. According to the phylogenetic tree of An7 and its homologues, An7 is located in CYP90D3 clade (Fig. 5), which is next to CYP90D2 clade including D2, which is previously reported brassinosteroid (BR) biosynthesis enzyme in rice (Hong et al. 2003). These results suggested that An7 carries cytochrome P450 domain and might be involved in BR biosynthesis.

Fig. 4.

Comparative analysis of the protein sequence of An7. A comparison of the amino acid sequences of An7 between O. sativa, O. glumaepatula and O. meridionalis. The sequence positions are indicated by numbers representing amino acid locations in An7. Non-synonymous variations, as compared with wild rice and O. sativa, are marked with red arrows.

Fig. 5.

Phylogenetic tree of An7 homologues across various plant species. Phylogenetic tree was generated using the An7 homologues from several plant species, A. thaliana (AT), O. sativa (LOC_Os), O. glumaepatula, O. meridionalis, Sorghum bicolor (Sobic), Zea mays (Zm), Brachypodium distachion (Bradi), and Glycine max (Glyma). The tree was constructed using the maximum likelihood method with PhyML, with 1000 bootstrap trials. Bootstrap values at key nodes are displayed. Scale bar indicates number of amino acid changes per branch length. An7 is marked with a blue circle, and D2 which is previously reported as brassinosteroid biosynthesis enzyme is marked with an orange circle.

The functional alleles of An7 are highly expressed in young panicle

To examine the expression pattern of An7, we performed quantitative Real-Time (qRT) PCR analysis using O. sativa and CSSLs (Glu-IL115 carrying O. glumaepatura allele of An7, and Mer-IL116 carrying O. meridionalis allele of An7). A previous study reported that CYP90D3 is expressed specifically in the root of O. sativa (Hong et al. 2003). We collected floral organs such as young panicle (length <5 mm) and spikelet (after heading) in order to examine the relationship between An7 expression and awn development. No expression of An7 in O. sativa was detected in young panicle whereas An7 was expressed in both CSSLs. In addition, the relative expression level of An7 in young panicle of Mer-IL116 was approximately 90 times higher than that of Glu-IL115 (Fig. 6). This result indicated that An7 allele of O. glumaepatula is highly expressed in comparison to that of O. sativa, but is significantly lower than that of O. meridionalis in young panicles. There are no expressions in spikelets of all three lines. This result suggested that An7 functions in young panicle to elongate awn in both O. glumaepatura and O. meridionalis, and this expression could be diminished in the later stage.

Fig. 6.

Expression pattern of An7 in O. sativa and CSSLs. Relative expression levels of An7 in spikelets and young panicle of T65 (O. sativa allele), Glu-IL115 (O. glumaepatula allele) and Mer-IL116 (O. meridionalis allele). OsUBI was used as an internal control. Spikelet and panicle samples are denoted as ‘spi’ and ‘pan’, respectively. Error bars represent the calculated maximum (RQMax) and minimum (RQMin) expression levels, indicating the standard error of the mean expression level (RQ value).

An7 functions cooperatively with RAE3 for awn elongation in rice

The CSSL line, Glu-IL115 carries functional alleles of both An7 and RAE3. Since RAE3 encodes an E3 ubiquitin ligase and works cooperatively with An-1/RAE1 and RAE2/GAD1 for awn elongation (Bessho-Uehara et al. 2023), there were two hypotheses regarding the relationship between An7 and RAE3. First, An7 may work independently for awn elongation with RAE3. Second, An7 may work cooperatively with RAE3 for awn elongation. To understand the relationship between An7 and RAE3, we generated a knockout line of RAE3 in Glu-IL115 background using CRISPR/Cas9 system (Fig. 7A). The RAE3 knockout line showed an awnless phenotype even An7 is functional (Fig. 7B), indicating that An7 works cooperatively with RAE3 for awn elongation in rice.

Fig. 7.

Knockout mutant of RAE3 in the background of Glu-IL115 results in an awnless phenotype. (A) Awn phenotype and graphical genotype of the control (Glu-IL115, left) and RAE3 CRISPR line (right, indicated as rae3). Scale bars represent 1 cm. (B) Awn length of the control and RAE3 CRISPR line (indicated as rae3). Awn length were measured from 15 or more seeds, with error bars representing mean ± SD. An asterisk indicates P < 0.05 by student’s t-test.

An7 has the potential to affect grain phenotype pleiotropically

The d2 mutant, which is deficient in the biosynthesis of active BR, castasterone, presents a short grain phenotype (Hong et al. 2003). To clarify whether An7 effects on grain phenotype, we measured grain size of An7 complementation line and vector control (VC). The result indicated that the grain length of An7 complementation line was longer than VC (Fig. 8A), but the grain width was not significantly different between both lines (Fig. 8B). This result suggested that An7 could produce much BR for awn elongation and contribute pleiotropically to grain length elongation. To clarify the effect of An7 on grain yield, we examined the grain size and yield in T65, Glu-IL115 and Mer-IL116 (Fig. 8C) due to the less enough number of seeds for measuring 1,000-grain weight of An7 complementation lines. The grain lengths of Glu-IL115 and Mer-IL116 were significantly longer than T65 (Fig. 8D), but the grain widths of Glu-IL115 and Mer-IL116 were significantly narrower than T65 (Fig. 8E). The 1,000-grain weight of two CSSL lines decreased compared with T65 (Fig. 8F).

Fig. 8.

Grain phenotype of O. sativa and An7 functional CSSLs. (A, B) Grain length (A) and grain width (B) of Vector Control (VC) and An7 complementation line. (C) Grain characteristics of T65, Glu-IL115 and Mer-IL116. Scale bar, 10 mm. (D–F) Grain length (D), grain width (E), and 1,000-grain weight (F) of T65, Glu-IL115 and Mer-IL116. Data for grain length and width were measured from 500 seeds, and 1,000-grain weight data were measured over five trials. Error bars represent mean ± SD. Asterisks indicate P < 0.05 by student’s t-test.

Discussion

Understanding the regulation of awn development in rice is crucial for agricultural improvement. Awn presence plays a key role in deterring feeding damage from wild boars, particularly in the mountainous regions of Japan (Kanbe et al. 2011). On the other hand, the absence of awns in cultivated varieties facilitates more efficient harvesting and processing. Previous studies have identified several key genes associated with awn development, including An-1/RAE1, RAE2/GAD1, LABA1 and RAE3, which play a role in awn regulation in wild rice species such as O. rufipogon and O. barthii, which are ancestors of domesticated rice species (Bessho-Uehara et al. 2016, 2023, Furuta et al. 2015, Hua et al. 2015, Jin et al. 2016, Luo et al. 2013). Despite these discoveries, there remains a significant gap in the research on this topic. Specifically, the genetic basis for awn regulation in other wild rice species remains unexplored. To address this knowledge gap, we investigated unidentified genetic loci involved in awn regulation. We focused on a range of wild rice species with the AA genome group that could be crossed with O. sativa which have no awns. A previous study, which evaluate several CSSLs crossed with five wild rice species belonging to AA genome species and fixed its background to O. sativa indicated that at least four loci related to awn length have not been identified (Bessho-Uehara et al. 2021). Among these loci, we identified the An7 gene associated with awn elongation on chromosome 5 in O. glumaepatula and O. meridionalis.

An7 encodes a CYP90D3 gene. Its homologue, D2/OsCYP90D2, has been reported as a brassinosteroid (BR) biosynthesis enzyme involved in catalyzing specific steps in the late BR biosynthesis pathway, including the conversion of 6-deoxoteasterone to 3-dehydro-6-deoxoteasterone and teasterone to 3-dehydroteasterone (Hong et al. 2003). Considering the phenotype of d2 mutant, which displays a semi-dwarf phenotype and shorter grain length (Hong et al. 2003), D2 contributes to BR biosynthesis in multiple organs and resulted in a specific morphological phenotype. The grain lengths of An7 complementation lines and both CSSLs carrying functional An7 are longer than T65 (O. sativa). This suggested that An7 may function pleiotropically not only to promote awn elongation but also grain length. However, it is important to note that due to the presence of multiple genes, besides An7, located on the substituted region of chromosome 5 in each CSSL, these may influence the final grain shape and yield. Consequently, based on the measurement of 1,000-grain weight in CSSLs alone, we cannot conclude whether functional An7 negatively impacts yield.

There are two possibilities that An7 allele of O. sativa to lose its function for awn elongation. One is the SNPs in the coding sequence of An7 causing loss-of function as indicated in Fig. 4, and the other is a change in expression pattern compared to O. glumaepatula and O. meridionalis. A previous study showed that An7, which is annotated as CYP90D3, is expressed in root specifically in O. sativa (Hong et al. 2003). In young panicles, An7 alleles of O. glumaepatula and O. meridionalis were highly expressed whereas An7 allele of O. sativa was not. Some structural differences in the promoter sequence of An7 in O. sativa compared to O. glumaepatula and O. meridionalis have the potential to alter the expression pattern of An7. Moreover, higher expression level of An7 in Mer-IL116 than that of Glu-IL115 was consistent with the awn length (Fig. 1D, 1H). Since the plant height of Glu-IL115 and Mer-IL116 were the same as T65 (O. sativa), the expression pattern of An7 would be restricted at awn and lemma/palea.

Several studies have provided support that implicates BR biosynthesis in awn development across different species. In barley, for instance, mutations in BR biosynthesis and receptor genes result in shorter awns (Dockter et al. 2014). Similarly, An-1/RAE1 gene in rice, which encodes a bHLH-type transcription factor regulating BR response, was reported to positively regulate awn elongation (Furuta et al. 2015, Luo et al. 2013). In green foxtail (Setaria viridis), the bristleless1 (Bsl1) gene has been identified as being responsible for bristle formation, which is an altered branch in inflorescences and shaped like awn by regulating BR biosynthesis at boundary regions (Yang et al. 2018). Taken together, these studies suggest BR is a positive regulator for awn and bristle development. Future experiments, such as quantification of endogenous BR in An7 complementation line, or BR treatment to the spikelets on awnless varieties would allow us to verify the relationship between BR and awn development in rice.

Deciphering the genetic interactions underlying complex traits like awn length is crucial for a comprehensive understanding of their molecular mechanisms. Over ten QTLs have been reported for awn development in previous research (Cai and Morishima 2002, Fawcett et al. 2013). In this complex landscape, our study has added value by identifying An7 as a new regulator that works cooperatively with RAE3. Taken together, these findings suggest that RAE3 plays a critical role in awn elongation, even in species such as O. glumaepatula.

Author Contribution Statement

MM, RM, MA and KBU designed the study. MM, RM and SA performed all the experiments. YS, YY, HY and AY provide the materials and performed sequence analysis. MM, RM and KBU wrote the manuscript. All authors have read and approved the final manuscript.

 Acknowledgments

This work was funded by JSPS KAKENHI (grant no. 20H05912 to MA, 22H04978 to MA, 21K15115 to KBU, 20J01187 to MM, 19K16163 to MM), the SATREPS program (no. JPMJSA1706 to MA) of the JST and JICA, and JST-Mirai Program (grant no. JPMJMI20C8). Great thankful to the National Bio Resource Project (NBRP) for providing the materials.

Literature Cited

•  Bessho-Uehara,  K.,  D.R.  Wang,  T.  Furuta,  A.  Minami,  K.  Nagai,  R.  Gamuyao,  K.  Asano,  R.B.  Angeles-Shim,  Y.  Shimizu,  M.  Ayano et al. (2016) Loss of function at RAE2, a previously unidentified EPFL, is required for awnlessness in cultivated Asian rice. Proc Natl Acad Sci USA 113: 8969–8974.
•  Bessho-Uehara,  K.,  Y.  Yamagata,  T.  Takashi,  T.  Makino,  H.  Yasui,  A.  Yoshimura and  M.  Ashikari (2021) Exploring the loci responsible for awn development in rice through comparative analysis of all AA genome species. Plants 10: 725.
•  Bessho-Uehara,  K.,  K.  Masuda,  D.R.  Wang,  R.B.  Angeles-Shim,  K.  Obara,  K.  Nagai,  R.  Murase,  S.  Aoki,  T.  Furuta,  K.  Miura et al. (2023) Regulator of Awn Elongation 3, an E3 ubiquitin ligase, is responsible for loss of awns during African rice domestication. Proc Natl Acad Sci USA 120: e2207105120.
•  Cai,  H. and  H.  Morishima (2002) QTL clusters reflect character associations in wild and cultivated rice. Theor Appl Genet 104: 1217–1228.
•  Dockter,  C.,  D.  Gruszka,  I.  Braumann,  A.  Druka,  I.  Druka,  J.  Franckowiak,  S.P.  Gough,  A.  Janeczko,  M.  Kurowska,  J.  Lundqvist et al. (2014) Induced variations in brassinosteroid genes define barley height and sturdiness, and expand the Green Revolution genetic toolkit. Plant Physiol 166: 1912–1927.
•  Fawcett,  J.A.,  T.  Kado,  E.  Sasaki,  S.  Takuno,  K.  Yoshida,  R.P.  Sugino,  S.  Kosugi,  S.  Natsume,  C.  Mitsuoka,  A.  Uemura et al. (2013) QTL map meets population genomics: An application to rice. PLoS One 8: e83720.
•  Furuta,  T.,  N.  Komeda,  K.  Asano,  K.  Uehara,  R.  Gamuyao,  R.B.  Angeles-Shim,  K.  Nagai,  K.  Doi,  D.R.  Wang,  H.  Yasui et al. (2015) Convergent loss of awn in two cultivated rice species Oryza sativa and Oryza glaberrima is caused by mutations in different loci. G3 (Bethesda) 5: 2267–2274.
•  Guindon,  S.,  J.F.  Dufayard,  V.  Lefort,  M.  Anisimova,  W.  Hordijk and  O.  Gascuel (2010) New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0. Syst Biol 59: 307–321.
•  Hattori,  Y.,  K.  Nagai,  S.  Furukawa,  X.J.  Song,  R.  Kawano,  H.  Sakakibara,  J.  Wu,  T.  Matsumoto,  A.  Yoshimura,  H.  Kitano et al. (2009) The ethylene response factors SNORKEL1 and SNORKEL2 allow rice to adapt to deep water. Nature 460: 1026–1030.
•  Hiei,  Y.,  S.  Ohta,  T.  Komari and  T.  Kumashiro (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.
•  Hong,  Z.,  M.  Ueguchi-Tanaka,  K.  Umemura,  S.  Uozu,  S.  Fujioka,  S.  Takatsuto,  S.  Yoshida,  M.  Ashikari,  H.  Kitano and  M.  Matsuoka (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.
•  Hua,  L.,  D.R.  Wang,  L.  Tan,  Y.  Fu,  F.  Liu,  L.  Xiao,  Z.  Zhu,  Q.  Fu,  X.  Sun,  P.  Gu et al. (2015) LABA1, a domestication gene associated with long, barbed awns in wild rice. Plant Cell 27: 1875–1888.
•  Ikemoto,  M.,  M.  Otsuka,  P.T.  Thanh,  P.D.T.  Phan,  R.  Ishikawa and  T.  Ishii (2017) Gene interaction at seed-awning loci in the genetic background of wild rice. Genes Genet Syst 92: 21–26.
•  Jin,  J.,  L.  Hua,  Z.  Zhu,  L.  Tan,  X.  Zhao,  W.  Zhang,  F.  Liu,  Y.  Fu,  H.  Cai,  X.  Sun et al. (2016) GAD1 encodes a secreted peptide that regulates grain number, grain length, and awn development in rice domestication. Plant Cell 28: 2453–2463.
•  Kanbe,  T.,  T.  Itani,  T.  Matsui,  K.  Nabata,  E.  Nara and  K.  Ishizaki (2011) The cultivation characteristics of long awned landrace rice cultivar “Shishikuwazu” in Niigata prefecture, Japan and prospects for the genetic dissection of long awn. Crop Science Society 231: 36 (in Japanese).
•  Kurakazu,  T.,  Sorbizal and  A.  Yoshimura (2001) RFLP mapping of genes for awn on chromosomes 4 and 5 in rice using Oryza meridionalis introgression lines. Rice Genet Newsl 18: 28–30.
•  Luo,  J.,  H.  Liu,  T.  Zhou,  B.  Gu,  X.  Huang,  Y.  Shangguan,  J.  Zhu,  Y.  Li,  Y.  Zhao,  Y.  Wang et al. (2013) An-1 encodes a basic Helix-Loop-Helix protein that regulates awn development, grain size, and grain number in rice. Plant Cell 25: 3360–3376.
•  Matsushita,  S.,  T.  Kurakazu,  K.  Doi and  A.  Yoshimura (2003a) Mapping of genes for awn in rice using Oryza meridionalis introgression lines. Rice Genet Newsl 20: 17–18.
•  Matsushita,  S.,  P.L.  Sanchez,  Sobrizal,  K.  Doi and  A.  Yoshimura (2003b) Identification of new alleles of awnness genes, An7 and An8, in rice using Oryza glumaepatula introgression lines. Rice Genet Newsl 20: 19–20.
•  Mikami,  M.,  S.  Toki and  M.  Endo (2015) Comparison of CRISPR/Cas9 expression constructs for efficient targeted mutagenesis in rice. Plant Mol Biol 88: 561–572.
•  Nguyen,  H.T.,  Q.H.  Vu,  T.  Van Mai,  T.T.  Nguyen,  L.D.  Vu,  T.T.  Nguyen,  L.V.  Nguyen,  H.T.T.  Vu,  H.T.  Nong,  T.N.  Dinh et al. (2018) Marker-assisted selection of Xa21 conferring resistance to bacterial leaf blight in indica rice cultivar LT2. Rice Sci 25: 52–56.
•  Numa,  H. and  T.  Itoh (2014) MEGANTE: A web-based system for integrated plant genome annotation. Plant Cell Physiol 55: e2.
•  Ohtsubo,  Y.,  W.  Ikeda-Ohtsubo,  Y.  Nagata and  M.  Tsuda (2008) GenomeMatcher: A graphical user interface for DNA sequence comparison. BMC Bioinformatics 9: 376.
•  Song,  W.Y.,  G.L.  Wang,  L.L.  Chen,  H.S.  Kim,  L.Y.  Pi,  T.  Holsten,  J.  Gardner,  B.  Wang,  W.X.  Zhai,  L.H.  Zhu et al. (1995) A receptor kinase-like protein encoded by the rice disease resistance gene, Xa21. Science 270: 1804–1806.
•  Toriba,  T. and  H.Y.  Hirano (2014) The DROOPING LEAF and OsETTIN2 genes promote awn development in rice. Plant J 77: 616–626.
•  Yang,  J.,  S.  Thames,  N.B.  Best,  H.  Jiang,  P.  Huang,  B.P.  Dilkes and  A.L.  Eveland (2018) Brassinosteroids modulate meristem fate and differentiation of unique inflorescence morphology in Setaria viridis. Plant Cell 30: 48–66.
•  Yoshimura,  A.,  H.  Nagayama,  Sobrizal,  T.  Kurakazu,  P.L.  Sanchez,  K.  Doi,  Y.  Yamagata and  H.  Yasui (2010) Introgression lines of rice (Oryza sativa L.) carrying a donor genome from the wild species, O. glumaepatula Steud. and O. meridionalis Ng. Breed Sci 60: 597–603.