Studies of a rice sterile mutant sstl from the TRIM collection
- 818 Downloads
Rice (Oryza sativa) is one of the main crops in the world, and more than 3.9 billion people will consume rice by 2025. Sterility significantly affects rice production and leads to yield defects. The undeveloped anthers or abnormal pollen represent serious defects in rice male sterility. Therefore, understanding the mechanism of male sterility is an important task. Here, we investigated a rice sterile mutant according to its developmental morphology and transcriptional profiles.
An untagged T-DNA insertional mutant showed defective pollen and abnormal anthers as compared with its semi-sterile mutant (sstl) progeny segregates. Transcriptomic analysis of sterile sstl-s revealed several biosynthesis pathways, such as downregulated cell wall, lipids, secondary metabolism, and starch synthesis. This downregulation is consistent with the morphological characterization of sstl-s anthers with irregular exine, absence of intine, no starch accumulation in pollen grains and no accumulated flavonoids in anthers. Moreover, defective microsporangia development led to abnormal anther locule and aborted microspores. The downregulated lipids, starch, and cell wall synthesis-related genes resulted in loss of fertility.
We illustrate the importance of microsporangia in the development of anthers and functional microspores. Abnormal development of pollen grains, pollen wall, anther locule, etc. result in severe yield reduction.
KeywordsAnther development Microspores Transcriptomic analysis Semi-sterile Sterility
early pollen mitosis stage
- GM filed
genetically modified plants isolation field
late pollen mitosis stage
mature pollen stage
Rice No Pollen 1
programmed cell death
pollen mother cells
primary parietal cells
scanning electron microscopy
single nucleotide polymorphism
secondary parietal layers
SQUAMOSA PROMOTER-BINDING PROTEIN-LIKE
fertile plants of sstl progeny segregates
sterile plants of sstl progeny segregates
semi-sterile plants of sstl progeny segregates
Taiwan Agricultural Research Institute
TAPETUM DEGENERATION RETARDATION
transmission electron microscope
Taiwan Rice Insertional Mutagenesis
vacuolated pollen stage
young microspore stage
Rice (Oryza sativa) is one of the most important staple crops in the world and is also an economical and cultivated food in Asia. One of the major limitations in rice yield production is sterility. Several studies provided strong evidence that male sterility was a serious defect in rice production and led to significantly decreased rice production (Thangasamy et al. 2011; Wang et al. 2013; Guo et al. 2013; Pu et al. 2017) or even almost complete loss (Liu et al. 2010; Ko et al. 2014; Yang et al. 2014; Li et al. 2016).
Male sterility is associated with a series of anther development steps and/or pollen maturation at the rice reproductive stage. Studies of the developmental morphology of pollen revealed several important steps, such as cell differentiation into microspores and circular anther locules, the formation of four specialized cell layers in the anther, meiosis of the pollen mother cell, microspores with a well-developed pollen wall, microspore mitosis, and intine establishment of mature pollen (Raghavan 1988; Zhang et al. 2011; Walbot and Egger 2016). Previous studies concluded that the process of pollen maturation in rice was initiated in the L1, L2 and L3 layers of the stamen primordia, and central archesporial (AR) cells were generated from the mitotic division in the L2 layer to develop the primary parietal cells (PPCs) and primary sporogenous cells (PSCs) (Zhang et al. 2011; Kelliher et al. 2014; Yang et al. 2018). After AR cells divide and differentiate to PSCs and PPCs, PSCs form as sporogenous cells and later generate microspores; PPCs divide and develop into secondary parietal layers (SPCs). The outer layer of SPCs becomes the endothecium and middle layer, and the inner layer becomes the tapetal cell (Zhang and Wilson 2009).
Receptor protein kinase (RPK) is important in regulating tapetal cells and microsporocytes during sporophytic generation. In Arabidopsis, a putative leucine-rich repeat receptor protein kinase (LRR-RPK) gene, EXCESS MICROSPOROCYTES1 (EMS1)/EXTRA SPOROGENOUS CELLS (EXS), participates in early division of the anther and microsporocyte cells to control tapetum and microsporocyte development (Cai and Zhang 2018; Verma 2019). TAPETUM DETERMINANT1 (TPD1) was found as a cysteine-rich protein ligand of EMS1. TPD1 interacts with the EMS1 LRR domain to generate precursors of tapetal cells by promoting cell division of inner SPCs; the developmental tapetum layer determines the number of sporogenous cells by suppressing proliferation (Huang et al. 2016). A rice protein binding cassette with the LRR-RLK protein MULTIPLE SPOROCYTE1 and its ligand TPD1-like protein (OsTDL1A) is involved in PPC division to contribute SPCs for the middle layer and establish the tapetum. The anther with double mutated OSTDL1A and MULTIPLE SPOROCYTE1 genes illustrated a lack of middle layers and tapetum and increased the number of microsporocytes in the early development of pollen (Yang et al. 2016). On mitosis, primary sporogenous cells divide into secondary sporogenous cells and become pollen mother cells (PMCs). PMCs pass through meiosis to generate microspores. At the same time, programmed cell death (PCD) of the tapetum layer occurs during meiosis. Several regulators respond to microspore generation, and PCD of tapetal cells such as a rice basic helix-loop-helix (bHLH) protein, TAPETUM DEGENERATION RETARDATION (TDR), controls tapetum degeneration to contribute to microspore development (Li et al. 2006). Gibberellic acid-inducible transcription factor GAMYB upregulates TDR1 expression, and a frame shift with a C deletion of the allele gamyb-4 did not go through normal meiosis or PCD of tapetum in rice (Liu et al. 2010). The bHLH142 that conjugates with TDR1 turned on the downstream gene ETERNAL TAPETUM1 (EAT1) to organize microspore development. With T-DNA insertion into bHLH142, the tapetal PCD and microspore development were defective in anthers of rice loss-of-function bhlh142 (Ko et al. 2014).
Microspores are produced after PMC meiosis, and young microspores are released from the tetrads later. The rice Wax-deficient anther1 (Wda1) defective mutant showed a significant defect of very-long-chain fatty acids in the anther epidermis and pollen wall that led to a severe absence of pollen (Jung et al. 2006). Thus, the fatty acid of pollen wall conformation plays an important role in microspore maturation.
The pollen wall is composed of the outer pollen wall exine and inner wall intine. Transmission electron microscope (TEM) observation revealed that exine consists of the tectum, bacula, and nexine together as a bridge-like structure of the pollen outer wall (Shi et al. 2015). The pollen walls of microspores are constructed during tapetum layer degradation and again indicate the importance of lipid metabolism (Li et al. 2011). Transcriptomic analysis revealed that transcripts encoding cytochrome P450, acyltransferases, and lipid transfer proteins are produced from the tapetum secretion stage to support fatty acid accumulation and transportation to build exine (Huang et al. 2009). Without PCD, tapetum did not degenerate to generate fatty acid components such as sporopollenin for exine formation. Rice PERSISTENT TAPETAL CELL1 (PTC1), which affects fatty acid synthesis in the exine structure, promotes tapetum degradation to generate sporopollenin for pollen wall development (Li et al. 2011). Scanning electron microscopy (SEM) revealed that the tapetal cell also produces orbicules during tapetum degeneration, which surround the surface of the tapetum layer, and orbicules provided the fatty acid product for assembly of sporopollenin for exine (Daku et al. 2016; Ruggiero and Bedini 2018). A rice lipid transfer protein OsC6 involved in lipid transfer from tapetum to exine is regulated by bHLH, TDR, and GAMYB. The truncated osc6 mutation resulted in defective orbicules and disordered exine structure in the pollen wall (Zhang et al. 2010). Thus, a well-developed exine is an indicator of maturation of microspore fertility during anther development.
The biosynthesis of sporopollenin is modified by several enzymes such as in-chain fatty acid hydroxylase (AtCYP703A2; OsCYP703A3), fatty acid ω-hydroxylase (AtCYP704B1; OsCYP704B2), polyketide synthase (OsPKS1), tetraketide reductase (OsTKPR1), fatty acid anther-specific acyl-CoA synthetase (ACOS; OsACOS12), and reductase (OsDPW) in fatty acid metabolism (Li et al. 2010, 2016; Choi et al. 2011; Wang et al. 2013; Yang et al. 2014; Daku et al. 2016). These enzymes participate in a series of steps to convert fatty acid acyl-CoA to the final sporopollenin product, which is then transported by ABC transporters such as ABCG26/WBC27, OsABCG3 and lipid transfer protein OsC6 from the tapetum to exine for pollen wall utilization (Zhang et al. 2010; Choi et al. 2011; Chang et al. 2018). The intine of the pollen wall consists of carbohydrate polysaccharides, which are converted from UDP-glucose (Shi et al. 2015). The intine formation of rice is controlled by Golgi-localized GLYCOSYLTRANSFERASE1 (OsGT1), which transfers UDP-glucose for polysaccharide composition requirement, and also regulates starch and protein synthesis in pollen grain maturation to determine rice fertility (Moon et al. 2013). The polysaccharides of intine also regulate water content in pollen dehydration and disperse pressure when mature pollen spores expand with internal content accumulation (Xu et al. 2016).
Taiwan Rice Insertional Mutagenesis (TRIM) was initiated in 2002. In this mutant population, 93,000 rice mutated lines were created by T-DNA insertion by using a japonica variety, Tainung 67 (Wu et al. 2017). The reverse genetics approach provided a large-scale population of T-DNA mutagenesis screening for comparing phenotypic characterization and rice genomic function. The TRIM population was used in two studies on male-sterile mutants. T-DNA insertion into bHLH142 led to no response to PCD of tapetum, and this gene was regulated by the early tapetum developing-related genes UNDEVELOPED TAPETUM1 (UDT1) and GAMYB, then was responsible for downstream TDR1 and EAT1 regulation (Ko et al. 2014). Another TRIM mutant study indicated that a rice SUMO E3 ligase SIZ1 participated in controlling endothecium development during anther dehiscence to manipulate the fertility of rice. Mutant siz1 showed the inability of anther dehiscence and no cavity of two adjacent locules; thus, it released no pollen grains during anthesis. In addition, this mutant had a shorter plant with few tillers and low seed set rate (Thangasamy et al. 2011).
Here we report a morphological characterization of a male sterile mutant from the TRIM collection and investigate the mechanism of sterility. We illustrate the importance of microspore fertility in anther development and the functional biosynthesis of microspores. We also performed a comparative transcriptomic analysis between fertile and sterile anthers to prove that fatty acid biosynthesis pathways participate in the morphology of pollen wall formation.
Plant materials and growth condition
T-DNA insertional mutant line M0037841 with 30 T2 seeds were obtained from the TRIM collection and grown at the genetically modified plant isolation field (GM field) in the Taiwan Agricultural Research Institute (TARI) for agronomic trait evaluation. Leaf samples from individual plants of the T2 population were harvested and frozen under liquid nitrogen for genomic DNA extraction experiments. Progeny from the semi-sterile line of M0037841 were both grown in the TARI GM qualified field or a growth chamber under day 28 °C/night 25 °C, 16 h of light and 80% humidity at Academia Sinica, Taipei. Yield component analysis including panicle number, grain number, fertility rate, and grain weight was performed by the semi-sterile segregates in the TARI GM field. The anther samples for floral morphology observation, pollen viability assay, gene expression analysis, phenotypic characterization of cross-sections, TEM and SEM were collected from plants grown in the growth chamber.
Genomic DNA extraction and PCR-based genotyping
Genomic DNA was extracted from leaves by using the Wizard Genomic DNA purification kit (Promega Corporation, Madison, WI, USA). All T-DNA mutagenesis plants from T2 were identified by T-DNA tagging regions, a T-DNA construct pTAG8, such as hygromycin phosphotransferase gene (HPT), GUS gene, and CaMV35S enhancer with PCR-based genotyping assay. The PCR reaction in 20 μL volume involved 50 ng genomic DNA, 0.2 μm primers and Taq DNA Polymerase 2× Master Mix RED (Ampliqon, Herlev, Denmark) for amplification. The sequences of primers are in Additional file 1: Table S1.
Pollen viability and germination analysis
Mature pollen grains were collected before anthesis in the fertile, semi-sterile and sterile types from semi-sterile progeny segregates and stained with iodine–potassium iodide (I2–KI) solution for pollen viability. Pollen viability was evaluated by using pollen shape and staining for classification into four grades as dark black (+++), partial black (++), orange (+) and yellow (−). More than 2000 individual pollen grains were counted under a Zeiss AxioImager Z1 microscope. More than 1000 pollen grains per sample were co-cultured with 15% sucrose (Sigma-Aldrich, St. Louis, MO, USA), 5% starch (Sigma-Aldrich, St. Louis, MO, USA), and 0.005% orthoboric acid (Merck, Darmstadt, Germany) at 28 °C for 30 min for detecting pollen tube germination rate.
Morphological analysis of anther characterization
Several reproductive growth parameters of development were revealed by kernel size, spikelet length, spikelet width, anther color, and anther length for 5 anther stages, young microspores stage (YM), vacuolated pollen stage (VP), early pollen mitosis stage (EpM), late pollen mitosis stage (LpM), and mature pollen stage (Mp) (Zhang and Wilson 2009; Huang et al. 2009). These five stages were equal to the anther stages from Zhang and Wilson (2009) reported in “stage 9”, “stage 10”, the first mitotic division of microspore in “stage 11”, the second mitosis of microspore in “stage 12”, and the spherical mature pollen grains in “stage 13”. Anther samples of all developmental stages were fixed in 2.5% glutaraldehyde, 4% paraformaldehyde, and 0.1 M sodium K-phosphate buffer (pH 7.0) at 4 °C for 24 h. Then, after three rinses with 0.1 M K-phosphate buffer (pH 7.0) for 20 min each, samples were transferred to 1% OsO4 in 0.1 M K-phosphate (pH 7.0) for 4 h at room temperature. The samples were washed again with 0.1 M K-phosphate buffer (pH 7.0) as for the previous rinsed steps. Samples were dehydrated with an ethanol series and propylene oxide, and embedded in Spurr resin. The cross sections were prepared in 1-μm sections and stained with toluidine blue for anther transverse section detection and anther locule area investigation under a Zeiss AxioImager Z1 microscope. The ultrathin sections in 70–90 nm were stained with uranyl acetate and lead citrate for TEM imaging by using a Philips CM 100 TEM system at 80 kV. For cryo-SEM, anthers and pollen grains at Mp stage were frozen in liquid nitrogen and placed into the sample preparation chamber at − 160 °C. After the temperature reached − 85 °C and sublimation/etching proceeded for 15 min, samples were coated with platinum particles at − 130 °C. Samples were moved to a cryo-stage SEM chamber and examined by using a Cryo-SEM system at temperature − 190 °C (Quanta 200 SEM/cryo system Quorum PP2000TR FEI).
RNA extraction and transcriptomic analysis
Anther samples were collected from YM or LpM stages for total RNA isolation by using TRIzol reagent (Thermo Fisher Scientific, Wilmington, DE, USA), and 5 μg total RNA was prepared for RNA sequencing. For RNA sequencing, single-end libraries were created and sequenced on an Illumina HiSeq 2000 system in 90-bp read length for 10 million reads per sample. Sequence data were deposited in the NCBI gene expression omnibus (accession no. GSE129579). RNA sequencing data were evaluated by RNA-seq analysis, differential expression of genes and heat map drawing with the CLC Genomics Workbench 11 (CLC Bio, QIAGEN Bioinformatics, Aarhus, Denmark). The RNA-seq transcripts were annotated by using the RAP-DB database (https://rapdb.dna.affrc.go.jp/index.html) and compared with the Rice Expression Profile Database (http://ricexpro.dna.affrc.go.jp/). Anther-specific transcriptomic genes were also confirmed by Rice Anther Expression Plots (https://www.cpib.ac.uk/anther/riceindex.html). EXPath Tool (http://expathtool.itps.ncku.edu.tw/), Kyoto Encyclopedia of Genes and Genomes (KEGG) (http://www.genome.jp/kegg/), and MapMen (https://mapman.gabipd.org/) were used for gene function prediction and analysis.
Gene expression analysis
Total RNA from YM and LpM were converted to cDNA with use of oligo (dT) primers by the SuperScript III First-Strand Synthesis System (Thermo Fisher Scientific, Wilmington, DE, USA). Real-time quantitative RT-PCR was performed with the SYBR Green-based quantitative PCR method by mixing 20 ng cDNA, 0.25 μm primers and fluorescence Power SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA) for gene expression detection. The gene expression was calculated by the 2−ΔΔCt method with OsUBI, OsACT11, OsUBQ5, and eEF-1α used as internal control genes. All primer sequences are in Additional file 1: Table S1.
A semi-sterile mutant from TRIM collection
Sterility is one of the important regulators of the reproductive stage in rice yield production. To study this phenomenon of sterility, we selected the T-DNA insertional mutant line M0037841 from the TRIM collection. Thirty T2 seeds of M0037841 were grown in a GM field in TARI. The T2 population in M0037841 was identified without co-segregation between the phenotype of sterility and the T-DNA insertional event of the genome (Additional file 2: Table S2).
Yield component analysis of sstl progeny segregates
Thus, the sterile characterization of sstl-s was not caused by a T-DNA insertion, and sstl-s still inherited the sterility trait from the progenitor semi-sterile sstl.
Sterility from sstl-s pollen viability
Pollen viability and pollen tube germination analyses were performed to determine the severity of defective pollen in sstl-s. Because mature pollen grains contained abundant starch to provide energy for supporting pollen maturation and pollen tube germination (Lee et al. 2016), we studied pollen viability by using I2–KI staining. Figure 2k showed a significant difference between mature pollen grains of SSTL-F (left) and sstl-s (right) in pollen viability, as indicated by staining color and pollen spherical shape. The starch accumulation of these pollen grains was illustrated as dark black (+++), partial black (++), orange (+) and yellow (−) after I2–KI staining (Fig. 2l). The proportion of dark-black mature pollen in SSTL-F, sstl-ss, and sstl-s was 96%, 64%, and 0.35%, respectively, revealing low pollen viability in the sterile mutant (Fig. 2l). Besides, sstl-ss showed 18% partial black pollen, with none in SSTL-F and sstl-s. The proportion of dark-black pollen in semi-sterile sstl-ss was about 60% and agreed well with half fertile seeds per panicle in sstl-ss (Table 1 and Fig. 2l). In addition, examination of pollen-tube germination of fertile and sterile pollen grains confirmed an 81.7% germination rate in SSTL-F but only 0.26% in sstl-s (data not shown).
The defects in sstl-s pollen reveal that its anther locules did not undergo anther dehiscence because the pollen sacs did not expand and pollen grains did not form a spherical shape. Almost zero percentage of pollen viability and pollen germination rates were consistent with the sterility of the mutant sstl-s as well as pollen-defective morphology.
Anther morphology of sstl-s
Transcriptomics analysis of the sstl-s mutant
The transcriptomics analysis also revealed that the sterile sstl-s anther genes contained low expression of biosynthesis pathway genes such as xanthine dehydrogenase (Os03g0429800), which participates in purine metabolism; the catalytic cycle of cytochrome P450 (Os01g0854800 and Os08g0131100), involved in secondary metabolites synthesis for exine formation; class III peroxidase (Os11g0210100 and Os03g0234900), which uses electrons in ox-reduction reaction to manipulate phenolic compound synthesis and secondary metabolites; GDSL esterase (Os05g0209600, Os05g0518300, and Os06g0156600), related to lipolytic enzymes and thus controlling pollen wall structure; fructosyl transferase (Os02g0106100), participating in starch composition of mature pollen grains; and the polygalacturonases-like 9 (Os05g0542800), involved in carbohydrate metabolism of polysaccharides for intine utilization. We used Rice Anther Expression Plots analysis to confirm that these significant transcripts of pollen development were expressed during anther-specific stages (Additional file 6: Figure S3). For instance, genes that appear from later meiosis to uninucleate microspore stage were consistent with our data in the YM stage. No Pollen 1 (Os10g0524500), xanthine dehydrogenase (Os03g0429800), and cytochrome P450 86A7-2 (Os01g0854800) were specifically upregulated in YM only, which was identical to the pattern in Rice Anther Expression Plots analysis. Os02g0106100, Os05g0542800, Os11g0210100, and Os09g0388400, significant genes in LpM stage, also responded to the same stage of bicellular pollen and mature pollen stage as in Rice Anther Expression Plots.
Somaclonal variation might cause the sstl trait
On cosegregation of the T-DNA insertion and sterile phenotype of two TRIM mutants M0024091 and M0017896, SUMO E3 ligase SIZ1 and bHLH142 were found to control male sterility via anther dehiscence and PCD of tapetum (Thangasamy et al. 2011; Ko et al. 2014). However, in the current study, the phenotype (sterility) and genotype (T-DNA insertion) did not cosegregate in the TRIM mutant M0037841. Previous studies have shown that rice T-DNA insertional mutants, such as Tos17 and TRIM, show low tagging efficiency (Droc et al. 2013; Wei et al. 2016b; Wu et al. 2017). The transformation was frequently used to study both forward and reverse genetics in knockdown/knockout mutants, overexpression lines, and T-DNA activation tagging lines, etc. In rice transformation, the procedure included repetitive tissue culture steps such as rice embryogenic callus generation, antibiotic selection of callus, and seedling regeneration. This long-term cullus growth period may lead to somaclonal variation including single nucleotide polymorphism (SNP), indels, chromosome doubling, and chromosome translocations (Wei et al. 2016a). From 600 to 2000 SNPs/indels per plant may occur during rice regeneration and transformation in the TRIM mutants (Wei et al. 2016a). Thus, this M0037841 semi-sterile sstl mutation might be also caused by somaclonal variations during transformation.
From the developmental morphology of sstl progeny segregates, the completely sterile sstl-s line inherited the sterility and showed defects in anther locule and pollen grain. This kind of somaclonal variation might be applied to crop improvement with stable heritability and also may speed up the breeding process and replace traditional crossbreeding (Jain 2001). Additional file 2: Table S2 illustrates that 6 T-DNA insertional mutant lines have normal fertility. From TRIM database information, we already know that the T-DNA insertion site of M0037841 locates in the exon of a Mei2-like protein (Os02g0517531). Mei2-like of Arabidopsis is a regulator of meiosis and expressed in gametes of pollens (Kaur et al. 2006). The rice genome contains 6 OsMei2-like genes (Anderson et al. 2004). Thus, M0037841 might maintain fertility because of the gene abundance.
Defective events start in microsporangia in the sstl-s mutant
From the cytological observation of anther morphology, we investigated the microspores of the sterile mutant sstl-s from YM to Mp stages and discovered that the early stage of microspore development is unusual and defective (Fig. 3f–o). The expression of the later pollen preference gene HAD-superfamily hydrolase-like (Os09g0388400) is extremely high in fertile SSTL-F, which indicates a severe defect in this expression in the mutant sstl-s. Significant downregulation of HAD-superfamily hydrolase-like may occur because sstl-s has empty defective pollen or even no microspores produced in anther locules (Fig. 3f–o). Moreover, the mutant sstl-s has abnormal and defective anthers in slender twig-like or even “X”-like shapes during microsporangia development stage (Additional file 4: Figure S2). A recent study of the rice SQUAMOSA PROMOTER-BINDING PROTEIN-LIKE 6 (SPL6) mutant spl6-1 also showed a cell death phenotype in the panicle and the distorted anther in a “X”-like shape, signaling damaged microsporangia development (Wang et al. 2018). OsSPL6 directly repressed inositol-requiring enzyme 1 (IRE1) expression as a transcriptional repressor. Also, overexpressed IRE1 was an endoplasmic reticulum stress sensor resulting in a cell death phenotype of panicle (Wang et al. 2018). Un-developmental microsporangia of the spl6-1 mutant led to sterility, including white spikelet and yield loss; the mutant also showed a significant reduction in panicle length, spikelet number, seed setting rate, and 1000 grain-weight. However, the spl6-1 mutant had no effect on plant height and number of panicle branches and thus was similar to sstl-s. In Arabidopsis, the expression of SPL1 and SPL12 increased heat tolerance during the reproductive stage and promoted seed production under heat stress. Without functional SPL1 and SPL12, the inflorescences of Arabidopsis showed dead cells and seeds were aborted (Chao et al. 2017). From morphological characterization, the microsporangia development of sstl-s has been destroyed, similar to the anthers of the spl6-1 mutant with an “X”-shape of early developmental anthers. From the genome sequence data, we confirm that sstl-s has no mutation in OsSPL6 (data not shown). The defective pollen and the “X”-like defective anther of sstl-s imply that microsporangia development is defective; thus, no sporogenous cells are produced in abnormal anther locules during early anther development.
Several biosynthetic pathways are affected in sstl-s
We also investigated several genes related to male-sterile phenotype. Rice No Pollen 1 (NP1) (Os10g0524500) of fertile pollen is expressed up to threefold that of the sterile sstl-s (Fig. 6b and Additional file 5: Table S3). NP1 regulates anther size and Ubisch body formation to contribute a pollen exine structure and manipulate fertility in pollen grains (Liu et al. 2017). The np1-4 mutant has an irregular pollen granule, reductive cutin, and less cuticular wax in exine formation (Liu et al. 2017). Rice lipid hydrolysis-related genes GDSLs are upregulated in mature pollen of SSTL-F (Fig. 6b–c and Additional file 5: Table S3). One GDSL was confirmed from Brassica rapa, known as extracellular lipase 6 (BrEXL6), which functions as a pollen development-related gene. Another case was ZmMs30, a maize GDSL, which was found essential for male fertility because the knockout ZmMs30 mutant showed a disordered pollen wall and defective pollen grains (Ji et al. 2017; An et al. 2019).
Here we report an untagged T-DNA insertional mutant M0037841 that has defective pollen and abnormal anthers from its sstl progeny segregaters. The abnormal anther locule is related to the damaged microsporangia and results in no microspore generation. The defective anther locule of sstl-s also illustrates the importance of the anther tapetum, which provides lipid-like sporopollenin for pollen maturation. Transcripts of sterile sstl-s are downregulated in several secondary-metabolism biosynthesis pathways and lead to irregular exine and absent intine in sstl-s. This information reveals the defective pollen grains of sstl-s and coincides with the irregular pollen wall and loss of pollen viability. Comparative transcriptomics analysis illustrates that the mechanism of male sterility is related to biosynthetic pathways for cell wall, lipids, secondary metabolism, and starch accumulation. We illustrate the importance of microsporangia development and male fertility of microspores. A series of morphological observations of mutant sstl-s anther demonstrated that the defective pollen grains are caused by unproduced intine and unfunctional exine of the pollen wall, then the defective pollen of sstl-s do not germinate and become sterile.
We thank Ms. Lie-Hong Wu for maintenance of greenhouse plants, and Laura Smales (BioMedEditing, Toronto, Canada) for English editing.
CLC and YICH are responsible for the experimental design. All sample collection was prepared by CLC. JCS and YCT were in charge of T2 population sample management. HHH and MHL performed the rice cultivation in TARI. CLC performed molecular analyses of PCR-genotyping, gene expression and bioinformatics analysis of transcriptomic data. YCL performed pollen staining analysis. WNJ prepared TEM samples for observation. CLC observed samples and took the photographs. CLC consolidated results and wrote the manuscript. YIH and MCC edited the manuscript. All authors read and approved the final manuscript.
This work was supported by Academia Sinica, Taiwan to YICH.
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no competing interests.
- An X, Dong Z, Tian Y, Xie K, Wu S, Zhu T, Zhang D, Zhou Y, Niu C, Ma B, Hou Q, Bao J, Zhang S, Li Z, Wang Y, Yan T, Sun X, Zhang Y, Li J, Wan X (2019) ZmMs30 encoding a novel GDSL lipase is essential for male fertility and valuable for hybrid breeding in maize. Mol Plant 12:343–359. https://doi.org/10.1016/j.molp.2019.01.011 CrossRefPubMedGoogle Scholar
- Anderson GH, Alvarez NDG, Gilman C, Jeffares DC, Trainor VCW, Hanson MR, Veit B (2004) Diversification of genes encoding Mei2-Like RNA binding proteins in plants. Plant Mol Biol 54:653–670. https://doi.org/10.1023/b:plan.0000040819.33383.b6 CrossRefPubMedGoogle Scholar
- Daku RM, Rabbi F, Buttigieg J, Coulson IM, Horne D, Martens G, Ashton NW, Suh DY (2016) PpASCL, the Physcomitrella patens anther-specific chalcone synthase-like enzyme implicated in sporopollenin biosynthesis, is needed for integrity of the moss spore wall and spore viability. PLoS ONE 11(1):e0146817. https://doi.org/10.1371/journal.pone.0146817 CrossRefPubMedPubMedCentralGoogle Scholar
- Droc G, An G, Hsing YI, Hirochika H, Pereira A, Undaresan CS, Han CD, Upadhyaya N, Ramachandran S, Comai L, Leung H, Guiderdoni E (2013) Mutant resources for the functional analysis of the rice genome. In: Wing R, Zhang Q (eds) Genetics and genomics of rice, vol 5. Springer, Berlin, pp 81–115CrossRefGoogle Scholar
- El-Kereamy A, Bi YM, Ranathunge K, Beatty PH, Good AG, Rothstein SJ (2012) The Rice R2R3-MYB transcription factor OsMYB55 is involved in the tolerance to high temperature and modulates amino acid metabolism. PLoS ONE 7(12):e52030. https://doi.org/10.1371/journal.pone.0052030 CrossRefPubMedPubMedCentralGoogle Scholar
- Gutiérrez-Luna FM, Hernández-Domínguez EE, Valencia-Turcotte LG, Rodríguez-Sotres R (2018) Review: “Pyrophosphate and pyrophosphatases in plants, their involvement in stress responses and their possible relationship to secondary metabolism”. Plant Sci 267:11–19. https://doi.org/10.1016/j.plantsci.2017.10.016 CrossRefPubMedGoogle Scholar
- Jung KH, Han MJ, Lee DY, Lee YS, Schreiber L, Franke R, Faust A, Yephremov A, Saedler H, Kim YW, Hwang I, An G (2006) Wax-deficient anther1 is involved in cuticle and wax production in rice anther walls and is required for pollen development. Plant Cell 18:3015–3032. https://doi.org/10.1105/tpc.106.042044 CrossRefPubMedPubMedCentralGoogle Scholar
- Ko SS, Li MJ, Sun-Ben KuM, Ho YC, Lin YJ, Chuang MH, Hsing HX, Lien YC, Yang HT, Chang HC, Chan MT (2014) The bHLH142 transcription factor coordinates with TDR1 to modulate the expression of EAT1 and regulate pollen development in rice. Plant Cell 26:2486–2504. https://doi.org/10.1105/tpc.114.126292 CrossRefPubMedPubMedCentralGoogle Scholar
- Lee S-K, Eom J-S, Hwang S-K, Shin D, An G, Okita TW, Jeon J-S (2016) Plastidic phosphoglucomutase and ADP-glucose pyrophosphorylase mutants impair starch synthesis in rice pollen grains and cause male sterility. J Exp Bot 67:5557–5569. https://doi.org/10.1093/jxb/erw324 CrossRefPubMedPubMedCentralGoogle Scholar
- Li N, Zhang DS, Liu HS, Yin CS, Li XX, Liang WQ, Yuan Z, Xu B, Chu HW, Wang J, Wen TQ, Huang H, Luo D, Ma H, Zhang DB (2006) The Rice Tapetum Degeneration Retardation gene is required for tapetum degradation and anther development. Plant Cell 18:2999–3014. https://doi.org/10.1105/tpc.106.044107 CrossRefPubMedPubMedCentralGoogle Scholar
- Li H, Pinot F, Sauveplane V, Werck-Reichhart D, Diehl P, Schreiber L, Franke R, Zhang P, Chen L, Gao Y, Liang W, Zhang D (2010) Cytochrome P450 family member CYP704B2 catalyzes the & #x03C9;-hydroxylation of fatty acids and is required for anther cutin biosynthesis and pollen exine formation in rice. Plant Cell 22:173–190. https://doi.org/10.1105/tpc.109.070326 CrossRefPubMedPubMedCentralGoogle Scholar
- Li H, Yuan Z, Vizcay-Barrena G, Yang C, Liang W, Zong J, Wilson ZA, Zhang D (2011) PERSISTENT TAPETAL CELL1 encodes a PHD-finger protein that is required for tapetal cell death and pollen development in rice. Plant Physiol 156:615–630. https://doi.org/10.1104/pp.111.175760 CrossRefPubMedPubMedCentralGoogle Scholar
- Li Y, Li D, Guo Z, Shi Q, Xiong S, Zhang C, Zhu J, Yang Z (2016) OsACOS12, an orthologue of Arabidopsis acyl-CoA synthetase5, plays an important role in pollen exine formation and anther development in rice. BMC Plant Biol 16:256. https://doi.org/10.1186/s12870-016-0943-9 CrossRefPubMedPubMedCentralGoogle Scholar
- Li-Beisson Y, Shorrosh B, Beisson F, Andersson MX, Arondel V, Bates PD, Baud S, Bird D, Debono A, Durrett TP, Franke RB, Graham IA, Katayama K, Kelly AA, Larson T, Markham JE, Miquel M, Molina I, Nishida I, Rowland O, Samuels L, Schmid KM, Wada H, Welti R, Xu C, Zallot R, Ohlrogge J (2013) Acyl-lipid metabolism. The Arabidopsis book, vol 11. American Society of Plant Biologists, Washington, p e0161. https://doi.org/10.1199/tab.0161 CrossRefGoogle Scholar
- Moon S, Oo MM, Kim B, Koh H-J, Oh SA, Yi G, An G, Park SK, Jung KH (2018) Genome-wide analyses of late pollen-preferred genes conserved in various rice cultivars and functional identification of a gene involved in the key processes of late pollen development. Rice 11:28. https://doi.org/10.1186/s12284-018-0219-0 CrossRefPubMedPubMedCentralGoogle Scholar
- Ning L, Lin Z, Gu J, Gan L, Li Y, Wang H, Miao L, Zhang L, Wang B, Li M (2018) The initial deficiency of protein processing and flavonoids biosynthesis were the main mechanisms for the male sterility induced by SX-1 in Brassica napus. BMC Genom 19:806. https://doi.org/10.1186/s12864-018-5203-y CrossRefGoogle Scholar
- Pu CX, Han YF, Zhu S, Song FY, Zhao Y, Wang CY, Zhang YC, Yang Q, Wang J, Bu SL, Sun LJ, Zhang SW, Zhang SQ, Sun DY, Sun Y (2017) The rice receptor-like kinases DWARF AND RUNTISH SPIKELET1 and 2 repress cell death and affect sugar utilization during reproductive development. Plant Cell 29:70–89. https://doi.org/10.1105/tpc.16.00218 CrossRefPubMedPubMedCentralGoogle Scholar
- Segami S, Tomoyama T, Sakamoto S, Gunji S, Fukuda M, Kinoshita S, Mitsuda N, Ferjani A, Maeshima M (2018) Vacuolar H+-pyrophosphatase and cytosolic soluble pyrophosphatases cooperatively regulate pyrophosphate levels in Arabidopsis thaliana. Plant Cell 30:1040–1061. https://doi.org/10.1105/tpc.17.00911 CrossRefPubMedPubMedCentralGoogle Scholar
- Walbot V, Egger RL (2016) Pre-meiotic anther development: cell fate specification and differentiation. Annu Rev Plant Biol 67:365–395. https://doi.org/10.1146/annurev-arplant-043015-111804 CrossRefPubMedGoogle Scholar
- Wei FJ, Kuang LY, Oung HM, Cheng SY, Wu HP, Huang LT, Tseng YT, Chiou WY, Hsieh-Feng V, Chung CH, Yu SM, Lee LY, Gelvin SB, Hsing YIC (2016a) Somaclonal variation does not preclude using rice transformants for genetic screening. Plant J 85:648–659. https://doi.org/10.1111/tpj.13132 CrossRefPubMedGoogle Scholar
- Wei FJ, Tsai YC, Hsu YM, Chen YA, Huang CT, Wu HP, Huang LT, Lai MH, Kuang LY, Lo SF, Yu SM, Lin YR, Hsing YIC (2016b) Lack of genotype and phenotype correlation in a rice T-DNA tagged line is likely caused by introgression in the seed source. PLoS ONE 11(5):e0155768. https://doi.org/10.1371/journal.pone.0155768 CrossRefPubMedPubMedCentralGoogle Scholar
- Yang X, Wu D, Shi J, He Y, Pinot F, Grausem B, Yin C, Zhu L, Chen M, Luo Z, Liang W, Zhang D (2014) Rice CYP703A3, a cytochrome P450 hydroxylase, is essential for development of anther cuticle and pollen exine. J Integr Plant Biol 56:979–994. https://doi.org/10.1111/jipb.12212 CrossRefPubMedGoogle Scholar
- Yang T, Wang L, Li C, Liu Y, Zhu S, Qi Y, Liu X, Lin Q, Luan S, Yu F (2015) Receptor protein kinase FERONIA controls leaf starch accumulation by interacting with glyceraldehyde-3-phosphate dehydrogenase. Biochem Biophys Res Commun 465:77–82. https://doi.org/10.1016/j.bbrc.2015.07.132 CrossRefPubMedGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.