Next Article in Journal
Effect of a Soil-Applied Humic Ameliorative Amendment on the Yield Potential of Switchgrass Panicum virgatum L. Cultivated under Central European Continental Climate Conditions
Previous Article in Journal
Predicting Long-Term Effects of Alternative Management Practices in Conventional and Organic Agricultural Systems on Soil Carbon Stocks Using the DayCent Model
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Agronomy
Volume 13
Issue 4
10.3390/agronomy13041094
agronomy-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Genetic Analysis of S5 Regulating the Hybrid Sterility between Indica and Japonica Subspecies in Rice

by
Jie Guo
1,2,3,
Xinqiao Zhou
1,2,3,
Ke Chen
1,2,3,
Chanjuan Ye
1,2,3,
Juan Liu
1,2,3,
Kangli Sun
4,
Guang Tang
4,
Shaokui Wang
4,
Guiquan Zhang
4,
Youding Chen
1,2,3,
Dagang Chen
1,2,3,* and
Chuanguang Liu
1,2,3,*
1
Key Laboratory of Genetics and Breeding of High Quality Rice in Southern China (Co-Construction by Ministry and Province), Ministry of Agriculture and Rural Affairs, Guangzhou 510640, China
2
Guangdong Key Laboratory of New Technology in Rice Breeding, Guangzhou 510640, China
3
Guangdong Rice Engineering Laboratory/Rice Research Institute, Guangdong Academy of Agricultural Sciences, Guangzhou 510640, China
4
Guangdong Provincial Key Laboratory of Plant Molecular Breeding, South China Agricultural University, Guangzhou 510640, China
*
Authors to whom correspondence should be addressed.
Agronomy 2023, 13(4), 1094; https://doi.org/10.3390/agronomy13041094
Submission received: 24 February 2023 / Revised: 24 March 2023 / Accepted: 3 April 2023 / Published: 11 April 2023
(This article belongs to the Special Issue Study of Targeted Genome Editing for Crop Improvement)
Browse Figures
Versions Notes

Abstract

:
Hybrid sterility is the major obstacle to the utilization of inter-subspecific heterosis in hybrid rice breeding. The S5 locus, composed of three adjacent genes ORF3, ORF4, and ORF5, plays a crucial role in regulating indica/japonica hybrids’ female sterility. Through a series of crosses involving 38 parents, three alleles of S5, ORF3+ORF4−ORF5n, ORF3+ORF4+ORF5n, and ORF3−/ORF4−/ORF5n, all could be regarded as wide-compatibility alleles, and when crossed with indica or japonica rice, they all showed significantly high fertility. Then, in order to explore the genes’ function, we further knocked out genes by using CRISPR/Cas9-based genome editing. Our results demonstrate that the ORF3+ was not just the protector in the killer-protector system, and knocking out ORF3 of the indica allele seriously affected the rice’s normal development. We observed the concrete enhancing hybrid spikelet fertility from the crosses between the ORF4+ knockout japonica materials with indica varieties. By conducting the comparative RNA-Seq analysis of young spikelets, we found that the ORF4+/ORF4− could modulate the hybrid fertility by affecting the expressions of genes related to the function of the Golgi apparatus. This study indicated that knocking out the ORF4+ of the japonica allele or using the alleles carrying ORF5n would provide effective approaches to overcome indica/japonica hybrid female sterility in rice breeding.

1. Introduction

Rice is one of the most important crops in the world. With the rapid growth of the world population and the dramatically decreasing cultivated lands globally, high yield is the unchangeable theme and challenge for rice breeding in the world [1]. The Asian cultivated rice (Oryza sativa L.) is divided into two subspecies, known as Xian/indica and Geng/japonica [2,3]. Due to their greater genetic diversity, indica/japonica crosses contained stronger heterosis than intra-subspecific crosses, mainly in plant height, effective spike, grain weight, and other essential agronomic traits [4,5]. So, the indica-japonica interspecific hybrid rice, known as fifth generation (5G) rice, could be represented as a potential approach to increasing rice productivity [6,7,8]. However, this heterosis of inter-subspecific crosses is hardly used in breeding programs because of the serious hybrid sterility [9,10,11].
Hybrid sterility is the common form of postzygotic reproductive isolation [9,12]. To unveil the molecular mechanism of hybrid sterility among indica/japonica subspecies remained a hugely complicated work. Subsequently, about 50 loci controlling inter-specific or inter-subspecific hybrid sterility have been identified or mapped, among which 12 of them have been cloned [7,11,13,14]. However, in indica/japonica hybrids, only three loci Sa, Sc, pf12, and DPL1/DPL2 were associated with hybrid pollen sterility [15,16,17,18], and three loci S5, HSA1, and S7 relating to hybrid embryo sac sterility [19,20,21] were cloned. Several genetic models have been proposed to explain the genetic mechanisms of hybrid sterility, among them, two major genetic models focusing on F1 hybrid sterility were more influential, the one-locus sporo-gametophytic interaction model, and the duplicate gametic lethal model [22,23,24]. Previous reports of five cloned loci conformed to the one-locus sporo-gametophytic interaction model but they have unique molecular mechanisms [7].
Among all these hybrid sterility loci, S5 is a major reproductive barrier regulator in cultivated rice. Chen et al. [19] cloned this gene and found that it encodes an aspartic protease, hereby proposing a tri-allelic system. Their research showed that the two different nucleotides (C819A and C1412T) in S5-i and S5-j have caused hybrid sterility as a heterozygous pair, while the S5-n allele with a 136 bp DNA-sequence deletion triggered the malfunction of genes conferring fertility with either allele [25,26,27,28]. Based on these results, Yang et al. [29] proposed a killer-protector system in indica/japonica hybrid rice. It indicated that the gene S5 contained three tightly linked open reading frames (ORFs). From protein function prediction, the ORF3 encodes a heat shock protein Hsp70, while the ORF4 encodes a membrane protein and the ORF5 encodes an aspartate protease. The genetic evidence showed that the ORF5+ in combination with the ORF4+ acts as a “killer” and selectively eliminates the female gametes without the “protector” ORF3+. In contrast, the ORF3+ prevented the gametes from being killed and subsequently rescued the hybrid fertility [29]. In addition, another study [30] showed the processes underlying a reproductive barrier induced by the S5 gene at the transcriptomic level, revealing that the interaction of different ORF combinations induced endoplasmic reticulum (ER) stress, and then led to programmed cell death (PCD), eventually resulted in abortive female gametes. So, diverse ORF combinations of S5 locus in hybrids in different genetic backgrounds would show a different level of hybrid sterility [30,31]. In summary, S5 gene is the key reason responsible for sterility of indica/japonica subspecies hybrids.
The discovery of wide-compatible genes (WCGs) and wide-compatible varieties (WCVs) provides the possibility to overcome interspecific hybrid sterility. The S5 is the first identified WCG, and the S5-n allele showed wide compatibility and could produce normal or higher fertility whether crossed with indica or japonica rice [29,31]. It is an essential resource to dispatch the sterility of indica/japonica subspecies hybrids. In order to subdue the indica/japonica hybrid sterility, the S5-n allele could be crossed with neutral alleles of some hybrid pollen sterility loci to develop wide compatibility lines (WCLs), which had compatibility with both indica and japonica varieties [32,33,34]. Furthermore, the S5-n allele could be pyramided with pollen sterility allele S-i in the japonica genetic background to develop indica-compatible japonica lines (ICJLs), which when crossed with indica rice could produce hybrid rice with normal or near normal fertility [35,36].
The clustered regularly interspaced short palindromic repeats (CRISPR) based gene editing technology, such as CRISPR/Cas9 system, have become powerful tools for genetic improvement and molecular breeding of crops [37,38,39]. It is capable of rapidly generating novel neutral or advanced alleles at the hybrid sterility loci to alter the expression level of the gene, which subsequently leads to conquering the inter-subspecies reproductive barrier [16,40]. The S5 gene is a vital locus associated with hybrid sterility [19,29]. In this study, we analyzed the genetic effect of the S5 genes on hybrid sterility by using indica or japonica testers and Huajingxian74 (HJX74) single-segment substitution lines (SSSLs). Moreover, we created new alleles of ORF3 and ORF4 to overcome the S5-induced hybrid sterility by knockout the allele of ORF4 in the japonica genotype. Finally, we performed RNA-sequencing (RNA-seq) to quantify the gene expression levels in hybrids F1 (ORF4+/ORF4+) in comparison to the hybrids F1 (ORF4+/ORF4−). Our study provided significant implications for understanding the molecular mechanisms of hybrid sterility and promoted the application of heterosis between indica and japonica cultivars in the future.

2. Materials and Methods

2.1. Plant Materials

The plant material included 38 rice varieties, four breeding lines HJX74-SSSLs, and one pyramiding line TISL-Dbc-Gde. (A) HXJ74, a recipient of the SSSLs and an elite variety planted widely in South China, was used as the donor of genetic background. (B) Three SSSLs carrying S5 locus in the substituted segments were selected from the SSSL library in HJX74 genetic background, and the genotypes and chromosome substitution segments were analyzed in previous research [34]. (C)TISL-Dbc-Gde was a pyramiding line that carried the S-i allele at the Sb and Sc loci from Dijiaowujian and the S-i allele at the Sd and Se loci from Guangluai4, and it was also a japonica line with a genotype and phenotype similar to Taichung 65 (T65) [36]. (D) Other testers were selected from a previously reported study [36]. Among them, 35 varieties were used for sequencing the coding sequence (CDS) of the S5 locus, and 38 varieties were used as testers to analyze the genetic effect of the S5 locus. The cultivars belong to typical indica or typical japonica based on available information. All rice plants were cultivated under natural growth conditions in Guangzhou, China.

2.2. Plasmid Construction and Plant Transformation

Three vectors were constructed in this paper. (A) The full length of ORF4 cDNA (Os06g0213000) was cloned from rice elite cultivar HJX74 into a plant binary vector pCAMBIA1300 under the control of the ubiquitin (UBI) promoter to generate the ORF4 overexpression construct Ubi:ORF4. Then the vector was transferred into Agrobacterium tumefaciens strain EHA105 by electroporation and consequently delivered into 10-06 as previously described [41]. (B) The ORF4 gene (Os06g0213000) in SSSL 10-06 was used as a reference to identify proper editing target sites. Two gRNAs, 5′-ACTACCTACATCACCTGCCG-3′, and 5′-CCGCGGCAGCTACGCGGCTT-3′ located within the 1st exon in ORF4 were selected using the CRISPR2 website (http://crispr.hzau.edu.cn/CRISPR2/, accessed on 1 May 2018). These gRNAs were cloned separately in the pBWA-Hu-cas9pl-U3 vector using this protocol [42]. The ORF4-gRNA1 and Cas9 expression cassettes were cloned into pCAMBIA1300 to construct the intermediate vector. Then, the ORF4-gRNA2 was cloned into this vector. Finally, the single gene ORF4 knock-out expression vector pBWA-Hu-cas9pl-U3-ORF4 plasmid was obtained. It was transferred into EHA105 and subsequently delivered into HJX74 SSSL 10-06 and TISL-Dbc-Gde, respectively. (C) The ORF3 (Os06g0212900) and ORF4 (Os06g0213000) gene in SSSL 14-06 was used as a reference, and the CRISPR/Cas9 gene plasmids for ORF3 and ORF4 were constructed as [43]. Six target site sequences of ORF3 GAACCAGTGATCGAAATCGATGG, GATTTCGATCACTGGTTCCGTGG, and CAGAGGGAACAGCAGCAGCTCGG) and ORF4 (CACCTGCCGCGGCAGCTACGCGG, CCGCGGCAGCTACGCGGCTTCGG, and GCTGCCGCGGCAGGTGATGTAGG) were cloned into the single guide RNA (sgRNA) expression cassette pBWA-Hu-cas9pl-U3 accordingly. Then, all integrated sgRNA expression cassettes were amplified and cloned into the CRISPR/Cas9 vector to construct pBWA-Hu-cas9pl-U3-ORF3-ORF4 plasmids. Then, this plasmid was transferred into EHA105 by electroporation and later delivered into HJX74 SSSL 14-06. All the primer sequences used in this study are listed in Table S9. All transgenic plants were cultivated in natural conditions in Guangzhou, China.

2.3. PCR Amplification and DNA Analysis

The genomic DNA was extracted from fresh-frozen leaves of each plant as described by Zheng et al. [44] with minor alterations. For DNA sequencing, the PCR was conducted using the method with minor modifications [45]. Two primers (ORF3-G and ORF4-G) listed in Table S9 were used to amplify sequences with the target sites accordingly to identify the genotype of the transgenic plants. All the sequence data were assembled and aligned by using the Lasergene Seqman software (DNASTAR Lasergene 7.0, Madison, WI, USA).

2.4. Fertility Examination

Three panicles per plant were harvested to examine spikelet fertility and scored as the ratio of the number of filled grains to the total spikelet. These mature flowers were collected from the upper one-third portion of the panicle and fixed in FAA solution (ethanol, formaldehyde, and acetic acid at a ratio of 89:6:5) to examine pollen fertility. More than 300 pollen grains were scanned by microscope using the I2-KI (containing 0.1% (w/v) iodine and 1% (w/v) potassium iodide) staining method randomly per plant.

2.5. Agronomic Traits and Statistical Analysis

All data on agronomic traits were collected from 10 plants per line. Multiple repeated mean values of each trait were statistically analyzed. The data are presented as the means ± standard deviation (SD) and transferred by arcsin−1 prior if they were provided as a percentage. Student’s t-test and Dunnett’s t-test were used to calculate the significant differences. The Chi-square (χ2) test was used to detect the distorted segregation of three genotypes in F2 populations from the Mendelian ratio of 1:2:1. The results were considered statistically significant when p < 0.05 (*) and extremely significant when p < 0.01 (**). The least significant difference (LSD) method or Duncan’s method was used for multiple-range tests among multiple groups. SPSS version 17 (Chicago, IL, USA), Microsoft Office Excel 2016 (Redmond, WA, USA), and Origin Pro 9.0 (Northampton, MA, USA) were used for data statistical analysis and figure making.

2.6. Sample Collection and Total RNA Extraction

Panicles were separately collected from three hybrids, 04-06/14-06 (as controls), 04-06/10-06 and 14-06/10-06 at the different panicle developed stages, such as 3.0–4.5-mm-long young panicles (meiosis stages, Sp1), 10.2–11.5-mm-long young panicles (Sp2), and mature panicles before heading (Sp3), correspondingly named Sp1-ck, Sp1-1, Sp1-2, Sp2-ck, Sp2-1, Sp2-2, Sp3-ck, Sp3-1, Sp3-2 with two replicates. Collected samples were frozen in liquid nitrogen and stored at −80 °C. Then, RNA was extracted using a Promega RNA extraction kit (Promega, Beijing, China) according to the manufacturer’s instructions. Total RNA was qualified and quantified using a NanoDrop and Agilent 2100 bioanalyzer (Thermo Fisher Scientific, Waltham, MA, USA).

2.7. Library Construction and RNA-Seq Analysis

In this paper, 1 μg RNA was used as raw material for the RNA sample preparations. Sequencing libraries were generated using NEBNext UltraTM RNA Library Prep Kit for Illumina (NEB, Ipswich, MA, USA) following the manufacturer’s recommendation with index codes adding to attribute sequences to each sample. The sequencing data was filtered through in-house perl scripts, the reads containing sequencing adapter; low-quality read, and reads with ploy-N were removed, afterward clean reads were obtained and stored in FASTQ format [46] The Q20, Q30, and GC content of the clean reads were shown in Table S5. The rice reference genome and annotation files were downloaded from NCBI (https://www.ncbi.nlm.nih.gov/assembly/GCF_001433935.1/ accessed on 5 June 2020). The clean reads were mapped to the reference genome using HISAT2 (v2.0.4) [47] and aligned to the reference coding gene using Bowtie2 (v2.2.5) [48].

2.8. DEG Identification and Analysis

Differential expression analysis of two biological replicates per group was performed using the DESeq2 (v1.4.5, R Foundation for Statistical Computing, Vienna, Austria). The genes with an adjusted Q ≤ 0.05 were assigned as differentially expressed. The heatmap was drawn by pheatmap (v1.0.8, Raivo Kolde, Boston, MA, USA) according to the gene expression in different samples. To take insight into the functions of differentially expressed genes (DEGs), Gene Ontology (GO, http://www.geneontology.org/ accessed on 2 August 2021) and Kyoto Encyclopedia of Genes and Genomes (KEGG, https://www.kegg.jp/ accessed on 1 August 2021) enrichment analysis of annotated DEGs was performed by Phyper (https://en.wikipedia.org/wiki/Hypergeometric_distribution/ accessed on 3 July 2021) based on Hypergeometric test. The significant levels of terms and pathways were corrected by Q value with a rigorous threshold (Q value ≤ 0.05) by Bonferroni [49]. The model of the protein was developed by using SWISS-MODEL (https://swissmodel.expasy.org/ accessed on 4 May 2021) and AlphaFold Protein Structure Database (https://www.alphafold.ebi.ac.uk/ accessed on 4 May 2021).

2.9. Real-Time Quantitative PCR Analysis

In order to confirm the genetic effect of S5, Real-time quantitative PCR (qRT-PCR) reactions were performed with three biological replicates using RNA samples from young panicles of SSSL carrying different genotypes. To validate the DEGs identified in the RNA-seq analysis, qRT-PCR analysis also was conducted on the selected genes. Real-time PCR reactions were performed using the SYBR® Premix Ex TaqTM Kit (Takara Biomedical Technology, Beijing, China) on an ABI PRISM 7900HT platform (Applied Biosystems, Waltham, MA, USA), according to the manufacturer’s instruction. Relative gene expression levels were calculated using the 2−ΔΔCt method [50]. The primers used for qRT-PCR are listed in Table S9.

3. Results

3.1. Nucleotide Variations and Genetic Effect of S5 Locus

Samples of 35 Oryza accessions (13 indica cultivars, 22 japonica cultivars) were taken to investigate the natural variations of ORF3, ORF4, and ORF5 at the S5 locus. The Coding sequences in ORF3-5 amplified by primers in Table S9 were obtained for each accession. Our results showed that two variants of ORF3 (one SNP and one InDel), four variants of ORF4 (three SNPs and one InDels), and seven variants of ORF5 (five SNPs and two InDels) in the coding regions (Figure 1A, Figures S1–S3). The analysis of all genetic sequences at the S5 locus, ORF3− and ORF3+, ORF4− and ORF4+, ORF5−, ORF5+, and ORF5n, was distinguished by predicted proteins according to previous reports [29]. Six allele combinations were found in these 35 selected varieties, ORF3+/ORF4+/ORF5+, ORF3+/ORF4–/ORF5+, ORF3+/ORF4−/ORF5n, ORF3+/ORF4+/ORF5n, ORF3−/ORF4−/ORF5n, and ORF3–/ORF4+/ORF5–, and the last three allele combinations were only discovered in 16 japonica varieties (Table S1).
To assay the genetic effect of the S5 locus, these Oryza accessions were regarded as parental testers in pairs to develop 62 crosses. In addition, the genotypes of the S5 locus for 15 other varieties were obtained from a previously reported study [36]. The spikelet fertility in all F1 hybrids varied from 3.28% to 92.48%, mainly due to the different alleles at the S5 locus in the crossing parental. The hybrids in heterozygous of S5 with common indica/japonica genotypes (ORF3+ORF4−ORF5+/ORF3−ORF4+ORF5−) showed full sterility, varied from 3.28% to 8.21%, with an average of 5.50%. The distorted segregation from the Mendelian segregation ratio was observed in the selected F2 populations. In contrast, the average spikelet fertility of the hybrids in the genotype of ORF3−ORF4+ORF5−/ORF3+ORF4+ORF5+ was significantly higher, reached to 39.51% (Figure 1B, Table S2). Combined with the genotypic analysis of the parents, this result may be related to the allelic interactions (+/−) of the ORF4 gene. In contrast, three different wide-compatibility alleles (ORF3+/ORF4−/ORF5n, ORF3+/ORF4+/ORF5n, and ORF3−/ORF4−/ORF5n) could rescue the fertility to 68.89%, 79.67%, and 80.29%, when crosses with either japonica or indica genotypes, and no segregation distortion was observed in the F2 offspring from these crosses (Figure 1B, Table S2). Therefore, the presence of ORF5n may have no impact on hybrid embryo sac sterility whatever alleles of ORF3 or ORF4 it combined with. Three wide-compatibility alleles could rescue the reproductive isolation between indica and japonica. Besides, the fertility of these hybrids was lower than expected, possibly due to the lower pollen fertility controlled by pollen sterility genes.

3.2. Allelic Interactions of ORF4 in the HJX74-SSSLs

Three SSSLs carrying S5 locus in the substituted segments were selected from the SSSL library in HJX74 genetic background, and the genotypes and chromosome substitution segments were analyzed in previous research [34]. In the S5 locus, the genotypes of 04-06, 10-06, and 14-06 were ORF3+ORF4−ORF5+/ORF3+ORF4−ORF5+, ORF3−ORF4+ORF5−/ORF3−ORF4+ORF5− and ORF3+ORF4+ORF5+/ORF3+ORF4+ORF5+ respectively [34]. The ORF3+ORF4−ORF5+ is the typical genotype in indica rice. Three ORFs all showed low chromatin accessibility during the spikelet development periods via Assay for Transposase-Accessible Chromatin with high throughput sequencing (ATAC-seq) (Figure S4). Then, we evaluated the expression of ORF3, ORF4, and ORF5 in young panicles. The results showed that the transcript levels of ORF3 in 04-06 were the highest, showing that ORF3+ and ORF3− have divergent functions. However, the transcript levels of ORF4 were obviously lower in 04-06 and 10-06 (Figure 2A). This result might due to the genetic structure of ORF4− (Figure S5). Three different ORF combinations at the S5 locus in F1 hybrids were obtained from the crosses between the SSSLs. The results indicated that the F1 hybrids in all genotypes showed normal pollen fertility, the fertility rates range from 92.18% to 95.02%. The genotypes of ORF3+ORF4+ORF5+/ORF3+ORF4−ORF5+ as control showed normal spikelet fertility. However, we found that the spikelet fertility in the ORF3+ORF4+ORF5+/ORF3−ORF4+ORF5− genotype was 76.55%, higher than 68.72% in the ORF3+ORF4−ORF5+/ORF3−ORF4+ORF5− genotype, which was only different in the genotype of ORF4, ORF4+/ORF4+ or ORF4+/ORF4− (Figure 2B,C). Interestingly, we discovered that the remarkable distorted segregation was detected both in two F2 populations, but the population of the japonica genotype was significantly lower in ORF3+ORF4−ORF5+/ORF3−ORF4+ORF5− population (Figure 2D). Taken together, the allele interaction of ORF4+/ORF4− could lead to lower fertility in F1 hybrids of indica/japonica crosses.

3.3. CRISPR/Cas9-Mediated Editing of ORF4 Gene in HJX74-SSSL

Previously, we collected 171 accessions of O. sativa and used them to identify the genotypes of the S5 gene in previous studies, but the combination ORF3−ORF4−ORF5−/ORF3−ORF4−ORF5− was not detected in O. sativa [36]. The ORF4 acted as an essential gene to assist the ORF5+ in eliminating gametes purposefully to control hybrid fertility. The ORF4 encodes a transmembrane structural domain protein, and ORF4− lost its function because of an 11 bp deletion, resulting in the premature termination of the encoded protein and leading to the loss of the transmembrane structural domain [29].
To obtain new alleles of the S5 locus, two adjacent targets sites of ORF4 located in exon 1 were selected and constructed into the CRISPR/Cas9 gene editing vectors, and transformed into the inbred HJX74-SSSL 10-06 (Figure 3A). Six T0 generation plants were amplified by PCR-specific primers, and positive plants were screened. Thus, two ORF4 knockout mutant lines E403-9 and E403-12 were obtained. The E403-9 harbored a 1-base pair(bp) insertion in the T1 editing site and a 41-bp deletion in the T2 editing site, while E403-12 harbored a 1 bp insertion of the sequence A in the T1 editing site and a 2 bp insertion of the sequence A in T2 editing site. Subsequently, the protein sequence analysis showed that the 1 bp base “A” insertion led to the early termination of the ORF4+ protein. In comparison with the reference ORF4+ protein, 234 amino acids (aa), the E403-9 and E403-12 were both predicted to produce the same truncated proteins of 103 amino acids, named ORF4(−) (Figure 3B, Figure S6). In addition, several agronomic traits of edited lines were investigated, and no significant difference between SSSL 10-16 and two gene editing mutants was found in the majority of investigated traits (Figure S7).
Hence, we examined the expression of three ORFs at the S5 locus in the young spikes of E403-9 mutants. The results revealed that the expression level of the ORF4 gene was significantly decreased in the two mutants in comparison to the control SSSL 10-06. There was no significant difference in the expression level of the ORF3 gene, while the expression of the ORF5 was significantly lower in mutants than that of the WT plant. Our results suggested that the editing of ORF4 caused a significant reduction in the expression of both ORF4 and ORF5 (Figure 3C).

3.4. ORF4(−) Improved Indica/Japonica Hybrid Fertility in HJX74 Background

To analyze the effect of ORF4(−) on hybrid sterility, two mutants, E403-9 and E403-12, were crossed with HJX74, SSSL 04-06, and 14-06. The F1 hybrid from HJX74/E403-9 and 04-06/E403-9 (ORF3+ORF4−ORF5+/ORF3−ORF4(−)ORF5−) showed normal spikelet fertility, compared with 65.39% of F1 hybrid from HJX74/10-06. While the F1 hybrid of 14-06/E403-9 (ORF3+ORF4+ORF5+/ORF3−ORF4(−)ORF5−) showed lower spikelet fertility, about 69.46%, compared with 78.51% of the 14-06/10-06 plant. (Figure 3D,E, Table S3). In addition, no distorted segregation of the S5 locus was detected in the F2 population of HJX74/E403-9. Previous studies showed that the majority of current indica rice varieties owned the genotype ORF3+ORF4−ORF5+/ORF3+ORF4−ORF5+ at the S5 locus, while the genotype of japonica rice was ORF3−ORF4+ORF5−/ORF3−ORF4+ORF5− [29,36]. In conclusion, our results revealed that ORF4 knockout in the japonica genotype might significantly enhance the hybrid spikelet fertility.

3.5. Overexpression of ORF4 Gene in HJX74-SSSL

Previous studies showed that the expression of ORF3 and ORF5 genes had no significant effect on the fertility of hybrid F1 lines [29,30,51]. In this study, we constructed an ORF4 overexpression vector and transferred it into HJX74-SSSL 10-06 background to obtain positive plant line E343. The results exhibited that the E343 had significantly higher expression of ORF4, ORF3, and ORF5 genes in young spikelets in comparison to 10-06 (Figure 3C). Then, we used E343 crossing with the 04-06 and 14-06 to obtain hybrid F1. Analysis of the fertility revealed that the combinations 04-06/E343 and 14-06/E343 showed normal pollen fertility, and lower spikelet fertility, 77.52% and 79.98%, respectively (Figure 3D,E). In sum, our results confirmed that overexpression of the ORF4 gene in japonica rice parents significantly improved the fertility of hybrid F1 of ORF4+/ORF4− combinations but had no significant effect on ORF4+ORF4/+ combinations.

3.6. CRISPR/Cas9-Mediated Editing of ORF3 and ORF4 Gene in HJX74-SSSL

To obtain new indica genotypes and confirm the function of the ORF4 gene, three target sites were selected in exon 4 of the ORF3 gene and exon 1 of the ORF4 gene respectively to develop two CRISPR/Cas9-based editing constructs and introduced in the HJX74-SSSL 14-06 (Figure 4A). In the T2 generation, we selected three homozygous mutant lines for further study. Sequence alignments of these two homozygous mutants Y184-2 and Y184-3 against that of the control 14-06 further indicated that 1 bp deletion appeared at the first target site, inducing gene frameshift of ORF3 resulting in a deletion of amino acid residues and no difference in ORF4. However, compared to the control 14-06, mutant Y184-4 existed 4 bp deletion between the PAM sites in the ORF3 gene which resulted in a premature translation but remained intact in the ORF4 gene. This allele of ORF3 was named ORF3(−β) (Figure 4B). In the T3 generation, we focused on three homozygous plants (Y184-18, Y184-22, and Y184-6) with different editing patterns in both ORF3 and ORF4 proteins. However, only the last ten amino acids were altered in the C termino of the ORF3 protein in Y184-6, we suspected that it might not affect the function of ORF3 (Figure 4B, Figures S8 and S9).
The expression level of the three genes at the S5 locus was analyzed in the young spikelet of the six mutants by qRT-PCR. Compared with it in WT, the expression of the gene clusters in the double knockout gene mutants Y184-18 and Y184-22 significantly declined. However, the expression of ORF3 in double knockout gene mutant Y184-6 was no obvious changes. In contrast, the expression of ORF4 and ORF5 in the mutants Y184-4 were significantly down-regulated, and its relative expression of ORF3 was up-regulated (Figure 4C). Previous reports discussed that the promoter region of ORF4 and ORF5 overlapped and frequently co-expressed [29]. In addition, lower expressions of ORF3 and ORF5 were shown in Y184-2 and Y184-3 (Figure 4C). In conclusion, based on the mutation sites and the expression of ORFs, the allele of S5 in Y184-18 and Y184-22 was named ORF3(−)ORF4(−)ORF5+, while it in Y184-6 was named ORF3(−γ)ORF4(−γ)ORF5+. The allele of S5 in Y184-4 was named ORF3+ORF4(−β)ORF5+, while it in Y184-2 and Y184-3 was named ORF3(−β)ORF4+ORF5+. The result also indicated that ORF3(−γ), ORF3(−), ORF3(−β) and ORF3+ alleles may have diverse functions.
Besides fertility, we also investigated other major relevant agronomic traits of 14-06 and six mutants. As expected, we discovered that no significant difference was present between 14-06 and Y184-6, Y184-4 in the majority of traits. In comparison to the 14-06, the panicle number, grain length, 1000 grain weight, and plant height of Y184-2, Y184-3, and Y184-22 consistently decreased to a relatively lower status. In total, the results indicated the editing of ORF3 would affect plant growth and development (Figure S10).

3.7. Strong Influence on Hybrid Sterility by Editing of ORF3 and ORF4 in HJX74 Background

To analyze the function of new alleles, we chose six mutants crossed with HJX74-SSSL 10-06 and mutants E403-9. We analyzed the fertility of mutants in heterozygous or homozygous genotypes and the hybrid F1 lines. The results suggested that the mutants in the homozygous genotype had normal fertility, except for mutant Y184-2 and Y184-3 (genotype of ORF3(−β)ORF4+ORF5+/ORF3(−β)ORF4+ORF5+), which had lower spikelet fertility, 33.24%, and 14.55%, but normal pollen fertility (Figure 5, Table S4). These data indicated that unlike common ORF3+, ORF3(−β) showed no function on hybrid sterility, and could not prevent the gametes from being eliminated.
Furthermore, the spikelet fertility of most mutants in the heterozygous genotype showed about 25% abortive, while the mutant Y184-4 (genotype of ORF3+ORF4(−β)ORF5+/ORF3+ORF4+ORF5+) and Y184-6 (genotype of ORF3(−γ)ORF4(−γ)ORF5+/ORF3+ORF4+ORF5+) showed normal spikelet fertility, revealing the ORF3+ and ORF3(−γ) acted as the same role and prevented the gametes from being killed.
In this case, we identified the spikelet fertility of 14-06/10-06 hybrid (genotype of ORF3+ORF4+ORF5+/ORF3−ORF4+ORF5−), served as control, was lower than normal level, about 78.51%. The hybrids F1 from the combinations Y184-2/10-06 (genotype of ORF3(−β)ORF4+ORF5+/ORF3−ORF4+ORF5−), Y184-2/E403-9 (genotype of ORF3(−β)ORF4+ORF5+/ORF3−ORF4(−)ORF5−) and Y184-18/10-06 (genotype of ORF3(−)ORF4+ORF5+/ORF3−ORF4+ORF5−) showed the lowest spikelet fertility 19.89%, 21.56% and 25.14%, respectively (Figure 5, Table S4). It revealed that the sequence difference in ORF3(−) and ORF3(−β) probably destroyed the ORF3+ protein structure resulting in functional changes in the hybrid sterility in the case of combining with japonica gametes.
Besides, compared with 14-06/10-06 hybrid (ORF3+ORF4+ORF5+/ORF3−ORF4+ORF5−, 78.51%) and 13-06/10-06 (ORF3+ORF4−ORF5+/ORF3−ORF4+ORF5−, 66.50%), the hybrid F1 of Y184-4/10-06 showed semi-sterility 49.62% in genotype ORF3+ORF4(−β)ORF5+/ORF3−ORF4+ORF5−, remarkably less than the controls. Previous research showed 14-06/E403-12 produce hybrid F1 with 65.25% fertility in genotype ORF3+ORF4+ORF5+/ORF3−ORF4(−)ORF5− (Figure 5, Table S4). In conclusion, all these results revealed there might exist a correlation between ORF4+, ORF4−and ORF5+ when indica-allele crossed with japonica-allele. This existence would regulate the hybrid fertility and ORF4 knockout in the indica allele would destroy this correlation leading to reduce the hybrid fertility.
In addition, the spikelet fertility of Y184-4/E404-9 in the genotype of ORF3+ORF4(−β)ORF5+/ORF3−ORF4(−)ORF5− and Y184-18/E403-9 or Y184-22/E403-9 in the genotype of ORF3(−)ORF4(−)ORF5+/ORF3−ORF4(−)ORF5− was normal, about 88.45%, 87.19% and 90.25% respectively (Figure 5, Table S4). These results revealed that the ORF4 knockout in two parents might hugely restore the hybrid F1 spikelet fertility.

3.8. Konckout the ORF4 Gene to Overcome Hybrid Sterility Caused by S5-Dependent in Indica/Japonica Background

The TISL-Dbc-Gde was a pyramiding line that carried the S-i allele at the Sb and Sc loci from Dijiaowujian and the S-i allele at the Sd and Se loci from Guangluai4, and it was also a japonica line with a genotype and phenotype similar to T65 [36]. In this study, we introduced ORF4-sgRNA/Cas9 editing constructs into the pyramiding line TISL-Dbc-Gde. We obtained three homozygous mutants, T304-3, T304-6, and T304-12. We identified the ORF4 gene premature terminated in all mutants by sanger sequencing, named ORF4 (−t) (Figure 6A, Figure S11). All these mutants have excluded transgenic elements of the Cas9 vectors. To further explore whether mutations in the ORF4 gene in the japonica background affect the main agronomic traits, we measured the main agronomic traits of three mutant lines, including plant height, 1000-seed weight, etc. The results showed almost no significantly different between TISL-Dbc-Gde and mutant lines (Figure S12).
To evaluate the effect of ORF4 (−t), the mutants were test-crossed with a set of testers including typical indica varieties Nanjing11 (NJ11), Minghui63 (MH63), and 9311. The genotype of the S5 locus in NJ11 and MH63 was ORF3+ORF4−ORF5+/ORF3+ORF4−ORF5+, while its genotype in 9311 was ORF3+ORF4+ORF5+/ORF3+ORF4+ORF5+. Compared with TISL-Dbc-Gde, three edited lines showed higher spikelet fertility of F1 hybrids when tested with NJ11 or MH63, with an average of 87.83% or 91.99% (Figure 6C,D). In the F2 populations from the T304-3/NJ11 and T304-3/MH63 crosses, no distorted segregation at the S5 gene was found, and the three genotypes’ segregated ratio were 56:113:52 and 63:142:59. Thus, these results indicated that the ORF4 knockout japonica lines could produce F1 hybrids with high spikelet fertility when it crossed with indica rice.

3.9. Comparative Transcriptome Analysis of Young Spikelet in Different Hybrids by RNA-Seq

To better understand the effects of the S5 locus in the transcription levels, the hybrid F1 lines from three hybrid combinations, 04-06/10-06 (genotype of ORF3+OR4−ORF5+/ORF3−ORF4+ORF5−), 14-06/10-06 (genotype of ORF3+OR4+ORF5+/ORF3−ORF4+ORF5−), and 04-06/14-06 (genotype of ORF3+OR4−ORF5+/ORF3+ORF4+ORF5+, as control), were selected at different periods of spikelet development, young spikelet length 3.0–4.5 cm (Sp1-1, Sp1-2, Sp1-ck), young spikelet length 10.2–11.5 cm (Sp2-1, Sp2-2, Sp2-ck), and the day before heading (Sp3-1, Sp3-2, Sp3-ck) for transcriptome analysis. Totally 1237.04 M clean reads were obtained. The Q30 base percentage was 95.00% (Table S5). The DEGs in different spikelet development stages were screened out based on the expression of fold change ≥ 2 and p-value ≤ 0.05. Then we verified the expression of six randomly selected DEGs by qRT-PCR, and the results showed similar expression trends between RNA-Seq data and transcriptome analysis (Figure S14).
In the Sp1 stage, the oocyte meiosis stage, genes related to ER stress as well as PCD such as OsBip5 [52] and OsAP25 [53] were not significantly differentially expressed in Sp1-1 and Sp1-2 compared with Sp1-ck, revealing the ER stress and PCD did not contribute to S5-dependent fertility. These results were probably due to the presence of ORF3+ genes in both hybrid combinations (Figure S13, Table S6). This conclusion was consistent with the results previous [30]. In the Sp2 stage, the young spikelet development, 1938 DEGs (495 up- and 1443 down-regulated) were identified in the Sp2-1 vs. Sp2-ck comparison (Figure 3B); while 1232 DEGs (175 up- and 1057 down-regulated), were identified in the Sp2-2 vs. Sp2-ck comparison. Furthermore, in the Sp3 stage, a total of 6294 and 4330 genes were differentially expressed in 04-06/10-06, 14-06/10-06 compared with 04-06/14-06, respectively. Among them, 3382 and 2296 genes were up-regulated, while 2912 and 2034 genes were down-regulated respectively at this stage (Figure 7A,B).
In general, the number of DEGs in Sp2-1/Sp2-ck and Sp2-2/Sp2-ck was 804 while it in S3-1/Sp3-ck and Sp3-2/Sp3-ck was 3212. Subsequently, to explore the metabolic regulatory network of rice hybrid sterility, KEGG enrichment analysis of these 804 DEGs and 3212 DEGs showed that these DEGs were enriched in these pathways relative to the ck, such as metabolic pathways, global and overview maps pathway, secondary metabolite biosynthesis pathway, phenylpropanoid biosynthesis synthesis, photosynthesis, TGF-beta signaling pathway, starch and sucrose metabolism and flavonoids biosynthesis (Figure S15). Interestingly, the secondary metabolism processes are important in plant stress response including phenylpropanoids and flavonoids metabolism [54,55], suggesting that the ORF interaction at the S5 locus might affect the metabolism and the activities in two hybrids involved in reducing fertility.
In particular, some genes related to fertility were selected from these DEGs. Among them, the expression of genes related to ER stress was significantly up-regulated in 04-06/10-06 and 14-06/10-06 compared with the control 04-06/14-06, such as OsHSP70, and OsBip5. However, there was no significant difference in PCD-related genes except OsAP37 [56]. The genes related to oocyte meiosis and phenylpropanoid biosynthesis were also significantly different from the control, with some genes significantly up-regulated in expression at the Sp2 stage and significantly down-regulated at the Sp3 stage (Figure 7C, Table S7). These results indicated that ORF interaction at the S5 locus induced the ER stress but not led to premature PCD, and it might produce some phenylpropanoids and flavonoids to response the stress and up-regulated oocyte meiosis genes expression to induce embryo-sac-producing abortive female gametes.

3.10. Analysis of Genes Differentially Expressed in Hybrids F1 (ORF4+/ORF4+) Compare to Hybrids F1 (ORF4+/ORF4−)

The ORF4 gene at the S5 locus encodes a transmembrane structural domain protein [29]. The hybrid F1 spikelet fertility from the cross of 14-06/10-06 was higher than it of 04-06/10-06. Generally speaking, these two combinations had the same background. To further understand the molecular mechanism of hybrid sterility, we performed a transcriptomic analysis of these two hybrids. Only 6, 107, and 984 DEGs (2, 48, and 543 up-regulated and 4, 59, and 441 down-regulated) were identified in the 04-06/10-06 vs. 14-06/10-06 in three different stages (Figure 8A). The GO standardized classification system was used to analyze these DEGs and classify 1097 DEGs into three categories: biological process, molecular function, and cellular components. Under molecular function, cellular process, binding and catalytic activity were the largest subcategories. The membrane and membrane part were two of the predominant subcategories in the cellular component category. Major subcategories under the biological process were the metabolic process, response to stimulus, and biological regulation (Figure 8B).
Later on, we analyzed these differentially expressed DEGs in the three periods (Figure S16). The genes related to glyoxylate and dicarboxylic acid metabolism, such as RBCS multigene family were significantly higher expressed compared with the hybrid F1 from 14-06/10-06 combination in Sp3 stage, leading to an accumulation of Rubisco holoenzyme [57,58]. Some genes related to glycines, serine and threonine metabolisms, and photosynthesis, such as OsGDCH, OsLFNR1, and the genes encoded by Ribosome protein were remarkable up-regulation (Figure 8C). In addition, comparing with the hybrid F1 from 14-06/10-06 combination, there were several noteworthy down-regulated genes such as F3H, flavanone 3-hydroxylase that is involved in flavonoid biosynthesis to response to stress [59], and ANS, important structural gene to response to abiotic stress [60]. ORF4+ localized to the plasma membrane and Golgi. The expression of CESA7 and BC7, encoding cellulose synthase catalytic subunit [61] were significantly decreased in the hybrid F1 from 04-06/10-06 combination. These genes may play crucial roles in hybrid sterility caused by ORF4+/ORF4− and ORF5+.

4. Discussion

Hybrid sterility is the major obstacle to the utilization of the strong vigor in indica/japonica hybrids. S5 was identified as the major locus contributing to inter-subspecific hybrid spikelet sterility [62,63,64]. It was a complex gene cluster, composed of three adjacent genes, ORF3, ORF4, and ORF5 [29]. Here, we explored the functions of three ORFs and analyzed the genetic effect of different ORF combinations on hybrid sterility via knocking out of these ORFs with CRISPR/Cas9-based genome editing or by transcriptomic analysis.
The S5 locus conformed to the one-locus sporo-gametophytic interaction model [65]. In the hybrids, the S5-n was regarded as a wide-compatibility allele and does not produce female sterility in combination with S5-i or S5-j alleles in the triallelic system [19] and further studies [19,25,27,28,65]. A killer-protector system was put forward, and it showed ORF5+ together with ORF4+ could selectively eliminate the female gametes without protector ORF3+ [29]. Previous studies showed that four different alleles containing ORF5n were all identified in Oryza sativa L. [31,36,66]. Three common alleles of them were used to cross with S5-i and S5-j alleles in this paper, and the higher fertility of F1 hybrids and no segregation distortion in the F2 population indicated that the presence of ORF5n has no effect on embryo sac sterility regardless of the alleles of ORF3 and ORF4. For instance, the hybrid with ORF3+ORF4+ORF5+/ORF3−ORF4−ORF5n showed higher fertility than it with ORF3+ORF4+ORF5+/ORF3−ORF4+ORF5−. The ORF5+, ORF4+, and ORF3+ all existed in these two hybrids, but they showed different fertility, which was not completely consistent with previous studies [29]. So, according to the one-locus sporo-gametophytic interaction model, ORF3+ORF4−ORF5n, ORF3+ORF4+ORF5n, and ORF3−/ORF4−/ORF5n might be the S5-n as previous [17], and they all could be regarded as wide-compatibility alleles. The superior wide compatible varieties were useful in obtaining an F1 hybrid with high fertility crossed with indica or japonica varieties [67,68]. Therefore, the germplasm carrying ORF5n would be an important resource to break the reproductive barrier between indica and japonica subspecies in rice breeding.
The HSP70s in plants were encoded by multiple gene families. The ORF3+ encoded a heat shock protein and localized to the ER [29], and most closely resembles the BiP (the luminal binding protein) protein. Five BiP genes were previously identified and the expression of all these genes could be up-regulated by ER stress [52,69,70]. The ORF3+ was regarded as a protector because it may help to fold the signal proteins to suppress the ORF5+ and ORF4+ induced ER stress and PCD [29,30]. Here, we provided some evidence supporting this point. We created new alleles of ORF3 with CRISPR/Cas9-based genome editing technology. ORF3+ and ORF3−, ORF3(−), and ORF3(−β) had several amino acids difference at the C terminal, but showed two different functions on hybrid sterility. without ORF3+, the hybrid in genotype ORF3(−β)ORF4+ORF5+/ORF3(−β)ORF4+ORF5+, ORF3(−β)ORF4+ORF5+/ORF3−ORF4+ORF5−, ORF3(−β)ORF4+ORF5+/ORF3−ORF4(−)ORF5−, and ORF3(−)ORF4(−)ORF5+/ORF3−ORF4+ORF5− showed normal pollen fertility but lowest spikelet fertility. These results confirmed the function of ORF3 in the killer-protector system [29]. Furthermore, we found that all the ORF3-knockout mutants in the indica background showed dwarf, sterility, and low yield. However, the japonica varieties grow normally in the genotype ORF3−ORF4+ORF5−/ORF3−ORF4+ORF5−. These results indicated there may exist a gene in the indica background that interacts with the ORF3+ gene, and knocking out ORF3+ might cut off the relationship and affect plant growth and development. ORF3 was an important gene that not only could protect the gametes to produce hybrid offspring with normal fertility, but also protect plant growth normally in indica background.
The ORF4+ was localized to the plasma membrane and Golgi, and it could help ORF5+ selectively eliminate the female gametes [29]. Typical japonica varieties carry the ORF3−/ORF4+/ORF5− haplotype, while typical indica varieties carry the ORF3+/ORF4−/ORF5+ haplotype. Here, we generated a new hybrid-compatible allele by knocking out the ORF4 in the japonica genotype. The hybrid in ORF3+ORF4−ORF5+/ORF3−ORF4(−)ORF5− genotype showed normal spikelet fertility from CRISPR/Cas9 knockout mutants crossed with typical indica varieties, showing ORF4(−) did break down the reproductive barrier induced by S5 in the hybrids. Besides, the hybrid F1 in the genotypes of ORF3+ORF4+ORF5+/ORF3−ORF4+ORF5− showed higher fertility than it in ORF3+ORF4−ORF5+/ORF3−ORF4+ORF5− genotype, which was only different in the interaction of ORF4+/ORF4−. The comparative RNA-Seq analysis of young spikelets was then conducted. Compared with the hybrid F1 from 14-06/10-06 combination, F3H, and ANS were dramatically down-regulated which were important in Flavonoid biosynthesis to respond to abiotic stress [59,60]. Meanwhile, the expression of CESA7 and BC7 encoded cellulose synthase catalytic subunit was also significantly decreasing in the hybrid F1 from 04-06/10-06 combination, which was related to the formation or function of the Golgi apparatus [61].
The three ORFs of the S5 locus together regulated inter-specific hybrid sterility in rice. In the indica/japonica hybrid, ORF3+ may help the folding of the signal proteins to suppress the ER stress and PCD induced by ORF5+ and ORF4+, and ORF4+ may be a receptor to receive a signal from ORF5+. Without ORF4+ in both parents would not induce female gamete abortion. If ORF4+ and ORF4− both existed, the genes related to the function of the Golgi apparatus were dramatically down-regulated and may affect the Golgi apparatus’s reprocessing of signal proteins from ER, inducing the lower fertility of hybrid in ORF4+/ORF4− combinations.
The CRISPR/Cas9 system was a powerful and highly efficient genome editing tool [37,38,39]. It could rapidly generate new neutral alleles at the hybrid sterility loci. Shen et al. [16] used CRISPR/Cas9 system to knock out one or two of the three Sc-i copies and rescued the expression level to overcome Sc-induced hybrid sterility. Xie et al. [40] created neutral Sa alleles by knocking out the SaF and SaM alleles by CRISPR/Cas9 system, which could overcome the male sterility caused by the Sa gene in hybrids. In this paper, TISL-Dbc-Gde was the pyramiding line in the japonica background. Knocking out ORF4 in TISL-Dbc-Gde showed wide compatibility and could produce high pollen fertility and spikelet fertility in their F1 hybrids with indica rice.
In conclusion, our study analyzed the function of three ORFs of S5. The allele carrying ORF5n showed wide compatibility and could produce high fertility in their F1 hybrids with both indica and japonica alleles. The ORF3+ was an important gene, and knockout in indica background could affect rice normal development. The allele interaction of ORF4 could regulate hybrid fertility by affecting the expressions of genes related to the function of the Golgi apparatus. Finally, we provide effective approaches to overcome indica/japonica hybrid sterility by knocking out the ORF4+ in the japonica genotype or using the alleles carrying ORF5n. These results of the present study would have laid a good foundation for the utilization of inter-subspecific heterosis in hybrid rice breeding.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/agronomy13041094/s1, Figure S1: Alignment of the protein sequences encoded by ORF3 in 35 varieties; Figure S2: Alignment of the protein sequences encoded by ORF4 in 35 varieties; Figure S3: Alignment of the protein sequences encoded by ORF5 in 35 varieties; Figure S4: The chromatin accessibility of the three genes during different spikelet development periods via ATAC-seq; Figure S5: Structure of ORF3+, ORF3−, ORF4+ and ORF4−; Figure S6: Sequence alignment of the sgRNA target region amplified by marker ORF4−G in two different ORF4 gene edited lines compared with the HJX74 SSSL 10-06; Figure S7: Phenotype of HJX74-SSSL 10-06 as control and two ORF4 gene edited mutants E403-9 and E403-12 in T2 generation, and ORF4 over-expression mutants E 343 and E342; Figure S8: Sequence alignment of the sgRNA target region amplified by marker ORF3-G in six different ORF3-ORF4 genes edited lines; Figure S9: Sequence alignment of the sgRNA target region amplified by marker ORF4-G in six different ORF3-ORF4 genes edited lines; Figure S10: Phenotype of HJX74-SSSL 14-06 as control and six ORF3-ORF4 genes edited mutants in T2 generation; Figure S11: Sequence alignment of the sgRNA target region amplified by marker ORF4-G in three different ORF4 gene edited lines compared with the TISL-Dbc-Gde; Figure S12: Phenotype of TISL-Dbc-Gde as control and three ORF4 genes edited mutants in T2 generation; Figure S13: Transcriptome analysis of the panicles of the hybrids from 04-06/10-06 and 14-06/10-06 (sterility) vs. 04-06/14-06 (fertility) at meiosis stage; Figure S14: The comparisons of key genes expressions between qRT-PCR and RNA-Seq data; Figure S15: Differentially expressed genes enriched GO categories and KEGG pathway (the spikelet of hybrids F1 from 04-06/10-06 and 14-06/10-06 vs. 04-06/14-06 in Sp2 and Sp3 stage); Figure S16: Differentially expressed genes enriched KEGG pathway (the spikelet of hybrids F1 from 04-06/10-06 vs. 14-06/10-06 in three development stage); Table S1: Genotype of ORF3, ORF4, and ORF5 at S5 locus for 38 parents used in diallel crosses; Table S2: Pollen fertility and spikelet fertility in hybrids F1 and the F2 distorted segregation of the S5 locus in a diallel set of crosses involving 38 parents (%±SD, n = 10); Table S3: The spikelet fertility in hybrids F1 from the crosses of indica-SSSLs with ORF4+ overexpression mutations or ORF4+ edited mutations in the japonica-SSSL 10-06 background; Table S4: The spikelet fertility in hybrids F1 from the crosses of gene-edited mutations in the SSSL 14-06 background with HJX74-SSSL 10-06 or ORF4-edited mutations in 10-06 background; Table S5: Sequencing data quality assessment and reads mapping with reference genome; Table S6: The differentially expressed genes in combination from 04-06/14-06 (fertility) and 04-06/10-06 or 14-06/10-06 (sterility) at meiosis stage; Table S7: The differentially expressed genes in combination from 04-06/14-06 (fertility) and 04-06/10-06 or 14-06/10-06 (sterility) at different period of immature inflorescences; Table S8: The differentially expressed genes in combination from 04-06/10-06 (low) and 14-06/10-06 (high) at different period of immature inflorescences; Table S9: Primer sequences used in this paper.

Author Contributions

Conceptualization and methodology, C.L., D.C., J.G. and G.Z.; validation, G.T., X.Z. and K.C.; formal analysis and investigation, C.Y., J.L. and K.S.; resources, S.W. and G.Z.; data curation, C.L., X.Z. and D.C.; writing—original draft preparation, J.G.; writing—review and editing, C.L., D.C. and Y.C.; supervision, Y.C. funding acquisition, J.G. and C.L. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the National Natural Science Foundation of China (grant number 31801449), the Natural Science Foundation of Guangdong Province, China (grant number 2018A030313867), the Special fund for scientific innovation strategy-construction of high-level Academy of Agriculture Science (grant number R2019PY-QF007), and the Guangdong Key Laboratory of New Technology in Rice Breeding (grant number 2020B1212060047).

Data Availability Statement

The raw data of RNA-seq reads were deposited in the NCBI database under the accession number PRJNA912986.

Acknowledgments

We thank BioRun biotechnology Inc. Ltd. (Wuhan, China) for performing the transgenic experiments.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Khush, G.S.; Gupta, P. Strategies for increasing the yield potential of cereals: Case of rice as an example. Plant Breed. 2013, 132, 433–436. [Google Scholar] [CrossRef]
  2. Garris, A.J.; Tai, T.H.; Coburn, J.; Kresovich, S.; McCouch, S. Genetic structure and diversity in Oryza sativa L. Genetics 2005, 169, 1631–1638. [Google Scholar] [CrossRef] [Green Version]
  3. Wang, X.; Guan, Z.; Gong, Z.; Yan, J.; Yang, G.; Liu, Y.G.; Yin, P. Crystal structure of WA352 provides insight into cytoplasmic male sterility in rice. Biochem. Biophys. Res. Commun 2018, 501, 898–904. [Google Scholar] [CrossRef] [PubMed]
  4. Yang, G.P.; Maroof, M.A.; Xu, C.G.; Zhang, Q.; Biyashev, R.M. Comparative analysis of microsatellite DNA polymorphism in landraces and cultivars of rice. Mol. Gen. Genet. 1994, 245, 187–194. [Google Scholar] [CrossRef] [PubMed]
  5. Ying, Y.G.; Wang, B.X.; Chen, T.M.; Yang, B.; Liu, Y.; Xu, B.; Sun, Z.G.; Chi, M.; Li, J.; Li, J.M.; et al. Heterosis analysis of yield and quality characters of two-line indica-japonica hybrid combination. Jiangsu Agric. Sci. 2021, 49, 59–63. [Google Scholar]
  6. Cheng, S.H.; Zhuang, J.Y.; Fan, Y.Y.; Du, J.H.; Cao, L.Y. Progress in research and development on hybrid rice: A super-domesticate in China. Ann. Bot. 2007, 100, 959–966. [Google Scholar] [CrossRef] [PubMed]
  7. Zhang, G.Q. Prospects of utilization of inter-subspecific heterosis between indica and japonica rice. J. Integr. Agric. 2020, 19, 1–10. [Google Scholar] [CrossRef]
  8. Zhang, G.Q. The Next Generation of Rice: Inter-subspecific indica-japonica hybrid rice. Front. Plant Sci. 2022, 13, 857896. [Google Scholar] [CrossRef]
  9. Kato, A. On the affinity of rice varieties as shown by the fertility of rice plants. Centr. Agric. Inst. Kyushu Imp. Univ. 1928, 2, 241–276. [Google Scholar]
  10. Ikehashi, H.; Araki, H. Varietal screening of compatibility types revealed in F1 fertility of distant crosses in rice. Jpn. J. Breed. 1984, 34, 304–313. [Google Scholar] [CrossRef] [Green Version]
  11. Ouyang, Y.D.; Zhang, Q.F. The molecular and evolutionary basis of reproductive isolation in plants. J. Genet. Genom. 2018, 45, 613–620. [Google Scholar] [CrossRef]
  12. Ouyang, Y.D.; Liu, Y.G.; Zhang, Q.F. Hybrid sterility in plant: Stories from rice. Curr. Opin. Plant Biol. 2010, 13, 186–192. [Google Scholar] [CrossRef] [PubMed]
  13. Ouyang, Y.; Chen, J.J.; Ding, J.H.; Zhang, Q.F. Advances in the understanding of inter-subspecific hybrid sterility and wide-compatibility in rice. Chin. Sci. Bull. 2009, 54, 2332–2341. [Google Scholar] [CrossRef]
  14. Li, J.; Zhou, J.; Zhang, Y.; Yang, Y.; Tao, D. New insights into the nature of interspecific hybrid sterility in rice. Front. Plant Sci. 2020, 11, 555572. [Google Scholar] [CrossRef] [PubMed]
  15. Zhou, P.; Wang, Z.; Zhu, X.; Tang, Y.; Ye, L.; Yu, H.; Li, Y.; Zhang, N.; Liu, T.; Wang, T.; et al. A minimal genome design to maximally guarantee fertile inter-subspecific hybrid rice. Mol. Plant 2023, 16, 726–738. [Google Scholar] [CrossRef]
  16. Shen, R.; Wang, L.; Liu, X.; Wu, J.; Jin, W.; Zhao, X.; Xie, X.; Zhu, Q.; Tang, H.; Li, Q.; et al. Genomic structural variation-mediated allelic suppression causes hybrid male sterility in rice. Nat. Commun. 2017, 8, 1310. [Google Scholar] [CrossRef] [Green Version]
  17. Mizuta, Y.; Harushima, Y.; Kurata, N. Rice pollen hybrid incompatibility caused by reciprocal gene loss of duplicated genes. Proc. Natl. Acad. Sci. USA 2010, 107, 20417–20422. [Google Scholar] [CrossRef] [Green Version]
  18. Long, Y.; Zhao, L.; Niu, B.; Su, J.; Wu, H.; Chen, Y.; Zhang, Q.; Guo, J.; Zhuang, C.; Mei, M.; et al. Hybrid male sterility in rice controlled by interaction between divergent alleles of two adjacent genes. Proc. Natl. Acad. Sci. USA 2008, 105, 18871–18876. [Google Scholar] [CrossRef] [Green Version]
  19. Chen, J.; Ding, J.; Ouyang, Y.; Du, H.; Yang, J.; Cheng, K.; Zhao, J.; Qiu, S.; Zhang, X.; Yao, J.; et al. A triallelic system of S5 is a major regulator of the reproductive barrier and compatibility of indica-japonica hybrids in rice. Proc. Natl. Acad. Sci. USA 2008, 105, 11436–11441. [Google Scholar] [CrossRef] [Green Version]
  20. Yu, Y.; Zhao, Z.; Shi, Y.; Tian, H.; Liu, L.; Bian, X.; Xu, Y.; Zheng, X.; Gan, L.; Shen, Y.; et al. Hybrid sterility in rice (Oryza sativa L.) involves the tetratricopeptide repeat domain containing protein. Genetics 2016, 203, 1439–1451. [Google Scholar] [CrossRef] [Green Version]
  21. Kubo, T.; Takashi, T.; Ashikari, M.; Yoshimura, A.; Kurata, N. Two tightly linked genes at the hsa1 locus cause both F1 and F2 hybrid sterility in rice. Mol. Plant 2016, 9, 221–232. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  22. Kitamura, E. Genetic studies on sterility observed n hybrids between distantly related varieties of rice, Oryza sativa L. Rep. Chugoku Exp. Stat. 1962, 8, 141–205. [Google Scholar]
  23. Oka, H.I. Genic analysis for the sterility of hybrids between distantly related varieties of cultivated rice. J. Genet. 1957, 55, 397–409. [Google Scholar] [CrossRef]
  24. Oka, H.I. Analysis of genes controlling F1 sterility in rice by the use of isogenic lines. Genetics 1974, 77, 521–534. [Google Scholar] [CrossRef]
  25. Wei, C.M.; Wang, L.; Yang, Y.X.; Chen, Z.X.; Shahid, M.Q.; Li, J.Q.; Liu, X.D.; Lu, Y.G. Identification of an S5n allele in Oryza rufipogon Griff. and its effect on embryo sac fertility. Chin. Sci. Bull. 2010, 55, 1255–1262. [Google Scholar] [CrossRef]
  26. Du, H.; Ouyang, Y.; Zhang, C.; Zhang, Q. Complex evolution of S5, a major reproductive barrier regulator, in the cultivated rice Oryza sativa and its wild relatives. New Phytol. 2011, 191, 275–287. [Google Scholar] [CrossRef]
  27. Tong, J.F.; Li, Y.H.; Yang, Y.X.; Shahid, M.Q.; Lu, Y.G. Molecular evolution of rice S5n and functional comparison among different sequences. Chin. Sci. Bull. 2011, 56, 2016–2024. [Google Scholar] [CrossRef] [Green Version]
  28. Peng, H.; Shahid, M.Q.; Li, Y.H.; Tong, J.F.; Zhou, X.W.; Liu, X.D.; Lu, Y.G. Molecular evolution of S5 locus and large differences in its coding region revealed insignificant effect on indica/japonica embryo sac fertility. Plant Syst. Evol. 2015, 301, 639–655. [Google Scholar] [CrossRef]
  29. Yang, J.; Zhao, X.; Cheng, K.; Du, H.; Ouyang, Y.; Chen, J.; Qiu, S.; Huang, J.; Jiang, Y.; Jiang, L.; et al. A killer-protector system regulates both hybrid sterility and segregation distortion in rice. Science 2012, 337, 1336–1340. [Google Scholar] [CrossRef] [PubMed]
  30. Zhu, Y.F.; Yu, Y.M.; Cheng, K.; Ouyang, Y.D.; Wang, J.; Gong, L.; Zhang, Q.H.; Li, X.H.; Xiao, J.H.; Zhang, Q. Processes underlying a reproductive barrier in indica-japonica rice hybrids revealed by transcriptome analysis. Plant Physiol. 2017, 174, 1683–1696. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  31. Mi, J.; Li, G.; Xu, C.; Yang, J.; Yu, H.; Wang, G.; Li, X.; Xiao, J.; Song, H.; Zhang, Q.; et al. Artificial selection in domestication and breeding prevents speciation in rice. Mol. Plant 2020, 13, 650–657. [Google Scholar] [CrossRef]
  32. Mi, J.; Li, G.; Huang, J.; Yu, H.; Zhou, F.; Zhang, Q.; Ouyang, Y.; Mou, T. Stacking S5-n and f5-n to overcome sterility in indica-japonica hybrid rice. Theor. Appl. Genet. 2016, 129, 563–575. [Google Scholar] [CrossRef] [PubMed]
  33. Ma, X.; Wei, X.; Gao, G.; Jiang, H. Development and evaluation of improved lines based on an elite rice variety 9311 for overcoming hybrid sterility in rice. Mol. Breed. 2020, 40, 102. [Google Scholar] [CrossRef]
  34. Guo, J.; Li, Y.; Xiong, L.; Yan, T.; Zou, J.; Dai, Z.; Tang, G.; Sun, K.; Luan, X.; Yang, W.; et al. Development of wide-compatible indica lines by pyramiding multiple neutral alleles of indica-japonica hybrid sterility loci. Front. Plant Sci. 2022, 13, 890568. [Google Scholar] [CrossRef] [PubMed]
  35. Chen, L.; Zhao, Z.; Liu, X.; Liu, L.; Jiang, L.; Liu, S.; Zhang, W.; Wang, Y.; Liu, Y.; Wan, J. Marker-assisted breeding of a photoperiod-sensitive male sterile japonica rice with high cross-compatibility with indica rice. Mol. Breed. 2011, 27, 247–258. [Google Scholar] [CrossRef]
  36. Guo, J.; Xu, X.; Li, W.; Zhu, W.; Zhu, H.; Liu, Z.; Luan, X.; Dai, Z.; Liu, G.; Zhang, Z.; et al. Overcoming inter-subspecific hybrid sterility in rice by developing indica-compatible japonica lines. Sci. Rep. 2016, 6, 26878. [Google Scholar] [CrossRef] [Green Version]
  37. Afzal, S.; Sirohi, P.; Singh, N.K. A review of CRISPR associated genome engineering: Application, advances and future prospects of genome targeting tool for crop improvement. Biotechnol. Lett. 2020, 42, 1611–1632. [Google Scholar] [CrossRef]
  38. Gao, C. Genome engineering for crop improvement and future agriculture. Cell 2021, 184, 1621–1635. [Google Scholar] [CrossRef] [PubMed]
  39. Zhang, R.X.; Li, B.B.; Yang, Z.G.; Huang, J.Q.; Sun, W.H.; Bhanbhro, N.; Liu, W.T.; Chen, K.M. Dissecting plant gene functions using CRISPR toolsets for crop improvement. J. Agric. Food Chem. 2022, 70, 7343–7359. [Google Scholar] [CrossRef] [PubMed]
  40. Xie, Y.; Niu, B.; Long, Y.; Li, G.; Tang, J.; Zhang, Y.; Ren, D.; Liu, Y.G.; Chen, L. Suppression or knockout of SaF/SaM overcomes the Sa-mediated hybrid male sterility in rice. J. Integr. Plant Biol. 2017, 59, 669–679. [Google Scholar] [CrossRef] [Green Version]
  41. Hiei, Y.; Komari, T.; Kubo, T. Transformation of rice mediated by Agrobacterium tumefaciens. Plant Mol Biol 1997, 35, 205–218. [Google Scholar] [CrossRef] [PubMed]
  42. Mao, Y.F.; Zhang, H.; Xu, N.F.; Zhang, B.T.; Gou, F.; Zhu, J.K. Application of the CRISPR-Cas system for efficient genome engineering in plants. Mol. Plant 2013, 6, 2008–2011. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  43. Ma, X.; Zhang, Q.; Zhu, Q.; Liu, W.; Chen, Y.; Qiu, R.; Wang, B.; Yang, Z.; Li, H.; Lin, Y.; et al. A robust CRISPR/Cas9 system for convenient, high-efficiency multiplex genome editing in monocot and dicot plants. Mol. Plant 2015, 8, 1274–1284. [Google Scholar] [CrossRef] [PubMed]
  44. Zheng, K.; Ning, H.; Bennett, J.; Khush, G.S. PCR-based marker-assisted selection in rice breeding. IRRI Discuss. Pap. 1995, 12. [Google Scholar]
  45. Panaud, O.; Chen, X.; McCouch, S.R. Development of microsatellite markers and characterization of simple sequence length polymorphism (SSLP) in rice (Oryza sativa L.). Mol. Gen. Genet. 1996, 252, 597–607. [Google Scholar] [CrossRef] [PubMed]
  46. Li, R.Q.; Li, Y.R.; Kristiansen, K.; Wang, J. SOAP: Short oligonucleotide alignment program. Bioinformatics 2008, 24, 713–714. [Google Scholar] [CrossRef] [Green Version]
  47. Kim, D.; Langmead, B.; Salzberg, S.L. HISAT: A fast spliced aligner with low memory requirements. Nat. Methods 2015, 12, 357–360. [Google Scholar] [CrossRef] [Green Version]
  48. Langmead, B.; Salzberg, S.L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 2012, 9, 357–359. [Google Scholar] [CrossRef] [Green Version]
  49. Abdi, H. The Bonferonni and Šidák Corrections for Multiple Comparisons. Encycl. Meas. Stat. 2007, 3, 2007. [Google Scholar]
  50. Livak, K.J.; Schmittgen, T.D. Analysis of relative gene expression data using real-time quantitative PCR and the 2−ΔΔCt Method. Methods 2001, 25, 402–408. [Google Scholar] [CrossRef]
  51. Ji, Q.; Zhang, M.; Lu, J.; Wang, H.; Lin, B.; Liu, Q.; Chao, Q.; Zhang, Y.; Liu, C.; Gu, M.; et al. Molecular basis underlying the S5-dependent reproductive isolation and compatibility of indica/japonica rice hybrids. Plant Physiol. 2012, 158, 1319–1328. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  52. Hayashi, S.; Wakasa, Y.; Takahashi, H.; Kawakatsu, T.; Takaiwa, F. Signal transduction by IRE1-mediated splicing of bZIP50 and other stress sensors in the endoplasmic reticulum stress response of rice. Plant J. 2012, 69, 946–956. [Google Scholar] [CrossRef] [PubMed]
  53. Niu, N.; Liang, W.; Yang, X.; Jin, W.; Wilson, Z.A.; Hu, J.; Zhang, D. EAT1 promotes tapetal cell death by regulating aspartic proteases during male reproductive development in rice. Nat. Commun. 2013, 4, 1445. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  54. Dixon, R.A.; Paiva, N.L. Stress-induced phenylpropanoid metabolism. Plant Cell 1995, 7, 1085–1097. [Google Scholar] [CrossRef]
  55. Winkel-Shirley, B. Biosynthesis of flavonoids and effects of stress. Curr. Opin. Plant Biol. 2002, 5, 218–223. [Google Scholar] [CrossRef]
  56. Mao, C.; Ding, J.; Zhang, B.; Xi, D.; Ming, F. OsNAC2 positively affects salt-induced cell death and binds to the OsAP37 and OsCOX11 promoters. Plant J. 2018, 94, 454–468. [Google Scholar] [CrossRef] [Green Version]
  57. Ogawa, S.; Suzuki, Y.; Yoshizawa, R.; Kanno, K.; Makino, A. Effect of individual suppression of RBCS multigene family on Rubisco contents in rice leaves. Plant Cell Environ. 2012, 35, 546–553. [Google Scholar] [CrossRef] [PubMed]
  58. Suzuki, Y.; Makino, A. Availability of rubisco small subunit up-regulates the transcript levels of large subunit for stoichiometric assembly of its holoenzyme in rice. Plant Physiol. 2012, 160, 533–540. [Google Scholar] [CrossRef] [Green Version]
  59. Chen, S.; Sun, B.; Shi, Z.; Miao, X.; Li, H. Identification of the rice genes and metabolites involved in dual resistance against brown planthopper and rice blast fungus. Plant Cell Environ. 2022, 45, 1914–1929. [Google Scholar] [CrossRef]
  60. Reddy, I. Rice flavonoid pathway genes, OsDfr and OsAns, are induced by dehydration, high salt and ABA, and contain stress responsive promoter elements that interact with the transcription activator, OsC1-MYB. Plant Sci. 2004, 166, 1505–1513. [Google Scholar]
  61. Zhang, B.; Deng, L.; Qian, Q.; Xiong, G.; Zeng, D.; Li, R.; Guo, L.; Li, J.; Zhou, Y. A missense mutation in the transmembrane domain of CESA4 affects protein abundance in the plasma membrane and results in abnormal cell wall biosynthesis in rice. Plant Mol. Biol. 2009, 71, 509–524. [Google Scholar] [CrossRef] [PubMed]
  62. Song, X.; Qiu, S.Q.; Xu, C.G.; Li, X.H.; Zhang, Q. Genetic dissection of embryo sac fertility, pollen fertility, and their contributions to spikelet fertility of intersubspecific hybrids in rice. Theor. Appl. Genet. 2005, 110, 205–211. [Google Scholar] [CrossRef] [PubMed]
  63. Wang, G.W.; He, Y.Q.; Xu, C.G.; Zhang, Q. Identification and confirmation of three neutral alleles conferring wide compatibility in inter-subspecific hybrids of rice (Oryza sativa L.) using near-isogenic lines. Theor. Appl. Genet. 2005, 111, 702–710. [Google Scholar] [CrossRef] [PubMed]
  64. Li, G.; Li, X.; Wang, Y.; Mi, J.; Xing, F.; Zhang, D.; Dong, Q.; Li, X.; Xiao, J.; Zhang, Q.; et al. Three representative inter and intra-subspecific crosses reveal the genetic architecture of reproductive isolation in rice. Plant J. 2017, 92, 349–362. [Google Scholar] [CrossRef] [Green Version]
  65. Ikehashi, H.; Araki, H. Genetics of F1 sterility in remote crosses of rice. In Rice Genetics I; IRRI: Manila, Philippines, 1986; pp. 119–130. [Google Scholar]
  66. Ouyang, Y.; Li, G.; Mi, J.; Xu, C.; Du, H.; Zhang, C.; Xie, W.; Li, X.; Xiao, J.; Song, H.; et al. Origination and establishment of a trigenic reproductive isolation system in rice. Mol. Plant 2016, 9, 1542–1545. [Google Scholar] [CrossRef] [Green Version]
  67. Kallugudi, J.; Singh, V.J.; Vinod, K.K.; Krishnan, S.G.; Nandakumar, S.; Dixit, B.K.; Ellur, R.K.; Bollinedi, H.; Nagarajan, M.; Kumar, A.; et al. Population dynamics of wide compatibility system and evaluation of intersubspecific hybrids by indica-japonica hybridization in rice. Plants 2022, 11, 1930. [Google Scholar] [CrossRef]
  68. Priyadarshi, R.; Arremsetty, H.P.S.; Singh, A.K.; Khandekar, D.; Ulaganathan, K.; Shenoy, V.; Sinha, P.; Singh, V.K. Marker-assisted improvement of the elite maintainer line of rice, IR 58025B for wide compatibility (S5(n)) gene. Front. Plant Sci. 2018, 9, 1051. [Google Scholar] [CrossRef] [Green Version]
  69. Oono, Y.; Wakasa, Y.; Hirose, S.; Yang, L.; Sakuta, C.; Takaiwa, F. Analysis of ER stress in developing rice endosperm accumulating beta-amyloid peptide. Plant Biotechnol. J. 2010, 8, 691–718. [Google Scholar] [CrossRef]
  70. Ohta, M.; Wakasa, Y.; Takahashi, H.; Hayashi, S.; Kudo, K.; Takaiwa, F. Analysis of rice ER-resident J-proteins reveals diversity and functional differentiation of the ER-resident Hsp70 system in plants. J. Exp. Bot. 2013, 64, 5429–5441. [Google Scholar] [CrossRef] [Green Version]
Agronomy 13 01094 g001 550
Figure 1. Genetic effect of ORF3, ORF4, and ORF5 at the S5 locus in different indica/japonica crosses. (A) Schematic drawing of polymorphisms in CDS of three genes at S5 locus. Black boxes and white boxes represent the exons and introns respectively. Single nucleotide polymorphisms are indicated by lines with solid circles in the gene model. Insertions/deletions (InDels) are indicated by lines with a solid block in the gene model. (B) Spikelet fertility of F1 hybrids in different genotypes at the S5 locus (left) and the ratio of genotypes at the S5 locus in the F2 populations from different indica/japonica crosses (right). +−n/−+−, +−n/+++, +−n/+−+, −−n/−+−, −−n/+++, −−n/+−+, ++n/−+−, ++n/+++, ++n/+−+, +−+/−+−, and +++/−+− represented the hybrid F1 in the genotype of ORF3+/ORF4−/ORF5n/ORF3−/ORF4+/ORF5−, ORF3+/ORF4−/ORF5n/ORF3+/ORF4+/ORF5+, ORF3+/ORF4−/ORF5n/ORF3+/ORF4−/ORF5+, ORF3−/ORF4−/ORF5n/ORF3−/ORF4+/ORF5−, ORF3−/ORF4−/ORF5n/ORF3+/ORF4+/ORF5+, ORF3−/ORF4−/ORF5n/ORF3+/ORF4−/ORF5+, ORF3+/ORF4+/ORF5n/ORF3−/ORF4+/ORF5−, ORF3+/ORF4+/ORF5n/ORF3+/ORF4+/ORF5+, ORF3+/ORF4+/ORF5n/ORF3+/ORF4−/ORF5+, ORF3+/ORF4−/ORF5+/ORF3−/ORF4+/ORF5−, and ORF3+/ORF4+/ORF5+/ORF3−/ORF4+/ORF5− respectively. P1P1, Genotype of S5 in female parents. P1P2, Heterozygous genotype. P2P2, Genotype of S5 in male parents. Error bars represent SD, n = 10. ** Significantly different at 0.01 probability level.
Figure 1. Genetic effect of ORF3, ORF4, and ORF5 at the S5 locus in different indica/japonica crosses. (A) Schematic drawing of polymorphisms in CDS of three genes at S5 locus. Black boxes and white boxes represent the exons and introns respectively. Single nucleotide polymorphisms are indicated by lines with solid circles in the gene model. Insertions/deletions (InDels) are indicated by lines with a solid block in the gene model. (B) Spikelet fertility of F1 hybrids in different genotypes at the S5 locus (left) and the ratio of genotypes at the S5 locus in the F2 populations from different indica/japonica crosses (right). +−n/−+−, +−n/+++, +−n/+−+, −−n/−+−, −−n/+++, −−n/+−+, ++n/−+−, ++n/+++, ++n/+−+, +−+/−+−, and +++/−+− represented the hybrid F1 in the genotype of ORF3+/ORF4−/ORF5n/ORF3−/ORF4+/ORF5−, ORF3+/ORF4−/ORF5n/ORF3+/ORF4+/ORF5+, ORF3+/ORF4−/ORF5n/ORF3+/ORF4−/ORF5+, ORF3−/ORF4−/ORF5n/ORF3−/ORF4+/ORF5−, ORF3−/ORF4−/ORF5n/ORF3+/ORF4+/ORF5+, ORF3−/ORF4−/ORF5n/ORF3+/ORF4−/ORF5+, ORF3+/ORF4+/ORF5n/ORF3−/ORF4+/ORF5−, ORF3+/ORF4+/ORF5n/ORF3+/ORF4+/ORF5+, ORF3+/ORF4+/ORF5n/ORF3+/ORF4−/ORF5+, ORF3+/ORF4−/ORF5+/ORF3−/ORF4+/ORF5−, and ORF3+/ORF4+/ORF5+/ORF3−/ORF4+/ORF5− respectively. P1P1, Genotype of S5 in female parents. P1P2, Heterozygous genotype. P2P2, Genotype of S5 in male parents. Error bars represent SD, n = 10. ** Significantly different at 0.01 probability level.
Agronomy 13 01094 g001
Agronomy 13 01094 g002 550
Figure 2. Genotypes and genetic effects of three genes at the S5 locus in HJX74 SSSLs (A) Expression level of three adjacent genes at S5 locus in young panicles of three SSSLs. The reference gene was Profilin. (B) Pollen grains stained by I2-KI solution (top) and spikelet fertility in the panicles (bottom) in F1 hybrids. Scale bar in pollen grains, 30 μm. Scale bar in panicles, 1 cm. The arrow points to the empty grain. (C) Pollen fertility and spikelet fertility in hybrids F1 from the crosses of the SSSLs carrying the S5 gene in different genotypes. ORF3, ORF4, and ORF5 were abbreviated as “3”, “4”, and “5”. Error bars represent SD, n = 10. (D) The ratio of genotypes at the S5 locus in the F2 populations from the crosses of SSSLs with S5 gene and testers. P1P1, Genotype of the left SSSL. P1P2, Heterozygous genotypes. P2P2, Genotypes of the right SSSL. **, significantly different at 0.01 probability level.
Figure 2. Genotypes and genetic effects of three genes at the S5 locus in HJX74 SSSLs (A) Expression level of three adjacent genes at S5 locus in young panicles of three SSSLs. The reference gene was Profilin. (B) Pollen grains stained by I2-KI solution (top) and spikelet fertility in the panicles (bottom) in F1 hybrids. Scale bar in pollen grains, 30 μm. Scale bar in panicles, 1 cm. The arrow points to the empty grain. (C) Pollen fertility and spikelet fertility in hybrids F1 from the crosses of the SSSLs carrying the S5 gene in different genotypes. ORF3, ORF4, and ORF5 were abbreviated as “3”, “4”, and “5”. Error bars represent SD, n = 10. (D) The ratio of genotypes at the S5 locus in the F2 populations from the crosses of SSSLs with S5 gene and testers. P1P1, Genotype of the left SSSL. P1P2, Heterozygous genotypes. P2P2, Genotypes of the right SSSL. **, significantly different at 0.01 probability level.
Agronomy 13 01094 g002
Agronomy 13 01094 g003 550
Figure 3. Knockout of ORF4 in SSSL S5-10 improved the hybrid fertility (A) The sites of target sequences in ORF4. The black boxes represent the exons. Two target sites are located on the first exon of the ORF4 gene. The target site is located in the box. The PAM sequence is underlined and shown in red. (B) Identification of mutations in ORF4 by sequencing of the target site in T2 generation. The positions highlighted in red show the targeted mutations and deletions indicated by the red dash “−”. ORF4 (−) indicated the ORF4 gene loss of function. (C) qRT-PCR of ORF3, ORF4, and ORF5 expression levels of RNA isolated from young panicles in different lines. 10-06, wild-type. E343, ORF4+ overexpression plant in the 10-06 background. E403-9, ORF4+ edited mutant line in the 10-06 background. Profilin was the reference gene. Results were presented relative to the expression level of three ORFs in 10-06. * Significantly different at 0.05 probability level. (D) The panicles in F1 hybrids from three crosses. Scale bar in panicles, 1 cm. (E) Spikelet fertility of hybrids F1 from the crosses of indica-SSSLs with E343 or E403-9. The genotypes of the parents in the crosses were in parentheses, ORF3, ORF4, and ORF5 were abbreviated as “3”, “4”, and “5”. Error bars represent SD, n = 10.
Figure 3. Knockout of ORF4 in SSSL S5-10 improved the hybrid fertility (A) The sites of target sequences in ORF4. The black boxes represent the exons. Two target sites are located on the first exon of the ORF4 gene. The target site is located in the box. The PAM sequence is underlined and shown in red. (B) Identification of mutations in ORF4 by sequencing of the target site in T2 generation. The positions highlighted in red show the targeted mutations and deletions indicated by the red dash “−”. ORF4 (−) indicated the ORF4 gene loss of function. (C) qRT-PCR of ORF3, ORF4, and ORF5 expression levels of RNA isolated from young panicles in different lines. 10-06, wild-type. E343, ORF4+ overexpression plant in the 10-06 background. E403-9, ORF4+ edited mutant line in the 10-06 background. Profilin was the reference gene. Results were presented relative to the expression level of three ORFs in 10-06. * Significantly different at 0.05 probability level. (D) The panicles in F1 hybrids from three crosses. Scale bar in panicles, 1 cm. (E) Spikelet fertility of hybrids F1 from the crosses of indica-SSSLs with E343 or E403-9. The genotypes of the parents in the crosses were in parentheses, ORF3, ORF4, and ORF5 were abbreviated as “3”, “4”, and “5”. Error bars represent SD, n = 10.
Agronomy 13 01094 g003
Agronomy 13 01094 g004 550
Figure 4. CRISPR/Cas9-mediated editing of ORF3 and ORF4 gene in HJX74-SSSL (A) Target sites in ORF3 or ORF4 for the CRISPR/Cas9 editing. The black boxes represent the exons. Two target sites are located on the four exons of the ORF3 or the first exon of ORF4. The target sites were located in frames. (B) Sequencing of the CRISPR/Cas9-targeted sites in ORF3 (left) and ORF4 (right). The PCR fragments containing the targeted sites were directly sequenced. The wild-type sequences in 14-06 were as references. The positions highlighted in red show the targeted mutations and deletions indicated by the red dash “−”. (C) qRT-PCR analysis of ORF3, ORF4, and ORF5 in young panicles of different gene-edited lines. Profilin was the reference gene. Results were presented relative to the expression level of three ORFs in 14-06. * Significantly different at 0.05 probability level.
Figure 4. CRISPR/Cas9-mediated editing of ORF3 and ORF4 gene in HJX74-SSSL (A) Target sites in ORF3 or ORF4 for the CRISPR/Cas9 editing. The black boxes represent the exons. Two target sites are located on the four exons of the ORF3 or the first exon of ORF4. The target sites were located in frames. (B) Sequencing of the CRISPR/Cas9-targeted sites in ORF3 (left) and ORF4 (right). The PCR fragments containing the targeted sites were directly sequenced. The wild-type sequences in 14-06 were as references. The positions highlighted in red show the targeted mutations and deletions indicated by the red dash “−”. (C) qRT-PCR analysis of ORF3, ORF4, and ORF5 in young panicles of different gene-edited lines. Profilin was the reference gene. Results were presented relative to the expression level of three ORFs in 14-06. * Significantly different at 0.05 probability level.
Agronomy 13 01094 g004
Agronomy 13 01094 g005 550
Figure 5. Knockout of both ORF3 and ORF4 by CRISPR/Cas9 resulted in lower hybrid fertility (A) Spikelet fertility in the panicles in F1 hybrids from six crosses. Scale bar in panicles, 1 cm. (B) Spikelet fertility in hybrids F1 from the crosses of gene-edited mutations in the SSSL 14-06 background with 10-06. The genotypes of the parents in the crosses were in parentheses, ORF3, ORF4, and ORF5 were abbreviated as “3”, “4”, and “5”. Error bars represent SD, n = 10.
Figure 5. Knockout of both ORF3 and ORF4 by CRISPR/Cas9 resulted in lower hybrid fertility (A) Spikelet fertility in the panicles in F1 hybrids from six crosses. Scale bar in panicles, 1 cm. (B) Spikelet fertility in hybrids F1 from the crosses of gene-edited mutations in the SSSL 14-06 background with 10-06. The genotypes of the parents in the crosses were in parentheses, ORF3, ORF4, and ORF5 were abbreviated as “3”, “4”, and “5”. Error bars represent SD, n = 10.
Agronomy 13 01094 g005
Agronomy 13 01094 g006 550
Figure 6. ORF4 knockout overcame the S5-dependent hybrid sterility in indica/japonica background (A) Identification of mutations in ORF4 by sequencing of the target site in T2 generation. The positions highlighted in red show the targeted mutations and deletions indicated by the red dash “–”. ORF4 (−t) indicated the ORF4 gene loss of function. (B) qRT-PCR of ORF3, ORF4, and ORF5 expression levels in RNA isolated from young panicles of different lines. Profilin was the reference gene. Results were presented relative to the expression level in TISL-Dbc-Gde. * Significantly different at 0.05 probability level. (C) Spikelet fertility in the panicles in F1 hybrids from three crosses. Scale bar in panicles, 1 cm. (D) Spikelet fertility in hybrids F1 from the crosses of TISL-Dbc-Gde with indica testers Minghui63 (MH63), Nanjing11 (NJ11), and 9311. Capital letters indicate statistical differences at 0.01 probability level, n = 10.
Figure 6. ORF4 knockout overcame the S5-dependent hybrid sterility in indica/japonica background (A) Identification of mutations in ORF4 by sequencing of the target site in T2 generation. The positions highlighted in red show the targeted mutations and deletions indicated by the red dash “–”. ORF4 (−t) indicated the ORF4 gene loss of function. (B) qRT-PCR of ORF3, ORF4, and ORF5 expression levels in RNA isolated from young panicles of different lines. Profilin was the reference gene. Results were presented relative to the expression level in TISL-Dbc-Gde. * Significantly different at 0.05 probability level. (C) Spikelet fertility in the panicles in F1 hybrids from three crosses. Scale bar in panicles, 1 cm. (D) Spikelet fertility in hybrids F1 from the crosses of TISL-Dbc-Gde with indica testers Minghui63 (MH63), Nanjing11 (NJ11), and 9311. Capital letters indicate statistical differences at 0.01 probability level, n = 10.
Agronomy 13 01094 g006
Agronomy 13 01094 g007 550
Figure 7. Transcriptome analysis of the panicles of WT (fertility) and F1 hybrids (sterility) (A) Number of the up- and down-regulated DEGs at Sp2 and Sp3 stages. (B) The Venn diagram shows the numbers of shared and unique DEGs. (C) The ratio of the expression of genes selected in the DEGs. The expression level based on log2 (FPKM + 0.01) is represented by a heat map. Sp2-ck, Sp2-1 and Sp2-2 indicated the florets in length ranges of 10.2–11.5 mm of the hybrid F1 from 04-06/14-06, 04-06/10-06, and 14-06/10-06 respectively. Sp3-ck, Sp3-1, and Sp3-2 indicated the florets one day before heading off the hybrid F1 from 04-06/14-06, 04-06/10-06, and 14-06/10-06 respectively.
Figure 7. Transcriptome analysis of the panicles of WT (fertility) and F1 hybrids (sterility) (A) Number of the up- and down-regulated DEGs at Sp2 and Sp3 stages. (B) The Venn diagram shows the numbers of shared and unique DEGs. (C) The ratio of the expression of genes selected in the DEGs. The expression level based on log2 (FPKM + 0.01) is represented by a heat map. Sp2-ck, Sp2-1 and Sp2-2 indicated the florets in length ranges of 10.2–11.5 mm of the hybrid F1 from 04-06/14-06, 04-06/10-06, and 14-06/10-06 respectively. Sp3-ck, Sp3-1, and Sp3-2 indicated the florets one day before heading off the hybrid F1 from 04-06/14-06, 04-06/10-06, and 14-06/10-06 respectively.
Agronomy 13 01094 g007
Agronomy 13 01094 g008 550
Figure 8. Analysis of genes differentially expressed in hybrids F1 (ORF4+/ORF4+) compared with hybrids F1 (ORF4+/ORF4−). (A) Number of the up- and down-regulated DEGs of hybrid F1 in different genotypes of ORF4. (B) The significant GO annotations of DEGs during three spikelet development stages. (C) Heat map of expression change of genes selected in the DEGs. The color in each frame represents the level of expression change based on the log2 fold. Sp1-1 and Sp1-2 indicated the florets in length ranges of 3.0–4.5 mm of the hybrid F1 from 04-06/10-06 and 14-06/10-06 respectively; Sp2-1 and Sp2-2 indicated the florets in length ranges of 10.2–11.5 mm of the hybrid F1 from 04-06/10-06 and 14-06/10-06 respectively; Sp3-1 and Sp3-2 indicated the florets one day before heading of the hybrid F1 from 04-06/10-06 and 14-06/10-06 respectively.
Figure 8. Analysis of genes differentially expressed in hybrids F1 (ORF4+/ORF4+) compared with hybrids F1 (ORF4+/ORF4−). (A) Number of the up- and down-regulated DEGs of hybrid F1 in different genotypes of ORF4. (B) The significant GO annotations of DEGs during three spikelet development stages. (C) Heat map of expression change of genes selected in the DEGs. The color in each frame represents the level of expression change based on the log2 fold. Sp1-1 and Sp1-2 indicated the florets in length ranges of 3.0–4.5 mm of the hybrid F1 from 04-06/10-06 and 14-06/10-06 respectively; Sp2-1 and Sp2-2 indicated the florets in length ranges of 10.2–11.5 mm of the hybrid F1 from 04-06/10-06 and 14-06/10-06 respectively; Sp3-1 and Sp3-2 indicated the florets one day before heading of the hybrid F1 from 04-06/10-06 and 14-06/10-06 respectively.
Agronomy 13 01094 g008
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Guo, J.; Zhou, X.; Chen, K.; Ye, C.; Liu, J.; Sun, K.; Tang, G.; Wang, S.; Zhang, G.; Chen, Y.; et al. Genetic Analysis of S5 Regulating the Hybrid Sterility between Indica and Japonica Subspecies in Rice. Agronomy 2023, 13, 1094. https://doi.org/10.3390/agronomy13041094

AMA Style

Guo J, Zhou X, Chen K, Ye C, Liu J, Sun K, Tang G, Wang S, Zhang G, Chen Y, et al. Genetic Analysis of S5 Regulating the Hybrid Sterility between Indica and Japonica Subspecies in Rice. Agronomy. 2023; 13(4):1094. https://doi.org/10.3390/agronomy13041094

Chicago/Turabian Style

Guo, Jie, Xinqiao Zhou, Ke Chen, Chanjuan Ye, Juan Liu, Kangli Sun, Guang Tang, Shaokui Wang, Guiquan Zhang, Youding Chen, and et al. 2023. "Genetic Analysis of S5 Regulating the Hybrid Sterility between Indica and Japonica Subspecies in Rice" Agronomy 13, no. 4: 1094. https://doi.org/10.3390/agronomy13041094

APA Style

Guo, J., Zhou, X., Chen, K., Ye, C., Liu, J., Sun, K., Tang, G., Wang, S., Zhang, G., Chen, Y., Chen, D., & Liu, C. (2023). Genetic Analysis of S5 Regulating the Hybrid Sterility between Indica and Japonica Subspecies in Rice. Agronomy, 13(4), 1094. https://doi.org/10.3390/agronomy13041094

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop