Comparative transcript profiling of the fertile and sterile flower buds of pol CMS in B. napus
© An et al.; licensee BioMed Central Ltd. 2014
Received: 25 November 2013
Accepted: 26 March 2014
Published: 3 April 2014
Skip to main content
© An et al.; licensee BioMed Central Ltd. 2014
Received: 25 November 2013
Accepted: 26 March 2014
Published: 3 April 2014
The Polima (pol) system of cytoplasmic male sterility (CMS) and its fertility restoration gene Rfp have been used in hybrid breeding in Brassica napus, which has greatly improved the yield of rapeseed. However, the mechanism of the male sterility transition in pol CMS remains to be determined.
To investigate the transcriptome during the male sterility transition in pol CMS, a near-isogenic line (NIL) of pol CMS was constructed. The phenotypic features and sterility stage were confirmed by anatomical analysis. Subsequently, we compared the genomic expression profiles of fertile and sterile young flower buds by RNA-Seq. A total of 105,481,136 sequences were successfully obtained. These reads were assembled into 112,770 unigenes, which composed the transcriptome of the bud. Among these unigenes, 72,408 (64.21%) were annotated using public protein databases and classified into functional clusters. In addition, we investigated the changes in expression of the fertile and sterile buds; the RNA-seq data showed 1,148 unigenes had significantly different expression and they were mainly distributed in metabolic and protein synthesis pathways. Additionally, some unigenes controlling anther development were dramatically down-regulated in sterile buds.
These results suggested that an energy deficiency caused by orf224/atp6 may inhibit a series of genes that regulate pollen development through nuclear-mitochondrial interaction. This results in the sterility of pol CMS by leading to the failure of sporogenous cell differentiation. This study may provide assistance for detailed molecular analysis and a better understanding of pol CMS in B. napus.
A very important source of vegetable oil worldwide, Brassica napus has extremely high oil production efficiency , and pol cytoplasm has been widely used in most of cultivated breeds . Cytoplasmic male sterility (CMS) prevents self-pollination through pollen abortion, enabling the use of heterosis in hybrid crops for genetic improvement [3, 4]. The sterility mechanism of pol has been examined in many studies, in which the orf224/atp6 and Rfp genes have mainly been analyzed. Results have shown that the sterility is caused by chimeric mitochondrial genes regulated by nuclear genes [2, 5–8].
In recent years, high-throughput sequencing methods such as Illumina SOLEXA, ABI SOLiD and Roche 454 have observably increased the efficiency and reduced the cost of sequencing, making the study of transcriptomes and even genome levels easier and more feasible . Nowadays, the transcriptomes of many higher plants have been sequenced for different purposes, including Arabidopsis thaliana[10, 11], Myrica rubra, Fagopyrum, Citrus sinensis, Carthamus tinctorius, and even Brassica napus for fertility studies . By RNA-Seq, researchers can obtain almost all of the expressed genes, especially genes with very low abundance. Therefore, genes with abundant expression differences and interesting pathways can be analyzed exhaustively . Additionally, RNA-Seq also has great advantages in the identification of new genes and SNPs, and even in genome-wide association studies (GWAS) [18–20]. In polyploid plants, it is used to study the fate of duplicated genes as well, such as in soybean and bread wheat [21, 22].
In this work, fertile and sterile flower buds of pol CMS with a length of 0–1 mm were sequenced using the Illumina high-throughput sequencing platform, representing the first study of the pol CMS genome at the transcriptome level. The aim of this work was to identify the differences between the fertile and sterile buds at the transcriptional level, and find out the different bioprocesses involved and their related functions. These results will help the elucidation of the sterility molecular mechanism, and assist the breeding of B. napus.
Summary of de novo transcriptome
> = 2000
The development of pollen is a fundamental and complex process in flowering plants. It is essential for propagation and evolution. As a model plant, Arabidopsis thaliana has been well studied for this bioprocess. Therefore, all of the unigenes identified here were annotated to the TAIR database (http://www.arabidopsis.org/). Subsequently, unigenes annotated to 19 genes from AG to MS1, which are considered to regulate pollen development and act before meiosis, were extracted for analysis [25–29]. The results showed that among the 19 genes, some were significantly down-regulated in sterile buds, such as BnNZZ/SPL, BnSPL8, BnEXS/EMS1 and BnER, but no gene was dramatically up-regulated (Additional file 2). Additionally, most of these genes did not show significant differences in expression between the fertile and sterile buds.
The FPKM (Fragments per Kilobase of transcript per Million mapped reads) value was calculated to test the expression levels of the unigenes. We noticed that in some basic bioprocesses these unigenes showed high expression levels in both the fertile and sterile buds. For example, unigene78635 and unigene78636, which were involved in RNA transport, showed very high expression in both fertile buds (FPKM = 14,059.7 and FPKM = 9,451.37) and sterile buds (FPKM = 12,954.3 and FPKM = 8,286.57). Additionally, some unigenes related to metabolism, such as unigene10390 (FPKM > 4,300), showed high abundance as well.
In addition, GO analysis was performed using Blast2GO . After enrichment analysis, only 9 GO terms showed significant differences. Among these terms, four (“oxidoreductase activity”, “cellular aromatic compound metabolic process”, “secondary metabolic process” and “cell wall”) were associated with the up-regulated unigenes and five (mainly hydrolase, pyrophosphatase and nucleoside-triphosphatase activity, and “nuclear lumen”) were associated with down-regulated unigenes (Figure 6B). In the molecular function category, only oxidoreductase and hydrolase activities were different between the up- and down-regulated unigenes (Figure 6C).
In particular, orf224, which is believed to be a key ORF of pol CMS in the mitochondrial genome , showed a 4-fold transcript increase in sterile buds compared with fertile buds. Atp6, a chimeric gene of orf224, also showed minor up-regulation.
Here, the transcriptome reads of fertile and sterile buds were acquired using the Illumina sequencing platform. Then, because of the unavailability of the genome and transcriptome of B. napus at present, we assembled the transcriptome of the buds for further research. Altogether, 112,770 unigenes were obtained, but only 1,148 unigenes (1.02% of all unigenes) showed notable differences in expression, indicating that though the development of buds is a complicated and polygenic process, changes in a relatively small number of genes can transform the trait observably.
In the present study, unigenes that were annotated as direct regulatory genes of pollen development showed higher abundance in fertile buds. Interestingly, unigene28529 (BnNZZ/SPL) was the first unigene that showed a difference in the regulatory network of anther development. Unigene19392 (BnAG), which acts as upstream gene of unigene28529, showed no difference in expression . Therefore, it can be presumed that the lack of Rfp emancipates the male-sterile gene in mitochondria. Through a series of bioprocesses, this male-sterile gene leads to serial inhibition of the downstream genes by regulating BnNZZ/SPL, a pivotal regulator of sporogenesis [33, 34]. Consequently, the fertile buds become sterile before the formation of sporogenous cells. It has been reported that nuclear-mitochondrial interaction results in CMS in previous studies [35, 36]. These genes were inhibited rather than silenced, which consequently led to the formation of only a few pollen sacs in the sterile buds.
Although orf224 and atp6 were both up-regulated in sterile buds, the change in orf224 expression was twice that of atp6, which might have been caused by the truncation of most of the orf224 transcripts in the presence of Rfp. Nevertheless, what has been discussed above indicates that energy supply was much more plentiful in the fertile samples. We speculate that even though the orf224/atp6 gene was normally transcribed, the translation of it was defective, or the translated protein could not form a functional advanced structure. After all, the existence of integral orf224 led to the insufficiency of ATPase protein 6, which caused the energy deficiency in the sterile buds. In other studies, the candidate cytoplasmic male sterile genes have also had a close relationship with ATPase genes, such as atp2, atp6[37, 38].
Comparative profiling was performed between the two transcriptomes. It was found that metabolism was still the principal activity in buds. Unigenes that involved in pyrophosphatase and nucleoside-triphosphatase activity were down-regulated, suggesting an energy deficiency in sterile buds. The depauperate bud activity was elucidated by the weakening of transcription and translation, which is similar to the results of an earlier study of Arabidopsis[39, 40]. Different distributions of energy and materials may strongly influence fertility, and the regulation of sterility is very complicated. At the transcriptional level, mRNA plays an important role; besides that, small RNA also participates in the process and makes the regulation network more complete [41, 42].
In this study, the different bioprocesses and genes that regulate pollen development were analyzed through a high-throughput sequencing approach. Combined with the investigation of the orf224/atp6 gene region, we propose a speculative model for the sterility and restoration mechanism of pol CMS. These results will promote the study of sterility mechanisms in more detail and the cloning of restorer genes. Additionally, the assembled B. napus unigenes obtained in this study will contribute to other transcriptomic and genomic studies.
Fertile and sterile flower buds of pol CMS in B. napus were used in this study. They were near-isogenic lines (NIL) and were both cultivated in the same experimental plot in Huazhong Agriculture University (Wuhan, Hubei Province, China). Buds with a length of 0–1 mm were stripped from three different plants for transcriptomic profiling. All fertile buds were gathered together (as were sterile buds), snap-frozen in liquid nitrogen and kept at -80°C for further use. Total RNA was isolated according to the instructions of the TRIzol kit (Invitrogen, USA) and purified using a mRNA purification kit (Promega, China) following the manufacturer’s protocol.
The mRNA was reverse transcribed by PowderscriptTM II (Takara, China), and then double-stranded cDNA was amplified according to the user manual of the SMARTTM cDNA construction kit. Finally, double-stranded cDNA was purified using a DNA purification kit (QIAGEN, Germany) to generate high quality cDNA. Approximately 10 μg of sheared cDNA was then prepared for Illumina sequencing according to the manufacturer’s protocols. Libraries were prepared from a 300–500 bp size-selected fraction following adapter ligation and agarose gel separation. The libraries were sequenced using a paired-end read protocol with 100 bp of data collected per run on the Illumina Hiseq 2000.
Clean reads were obtained from raw data by filtering the adaptor sequences and the low-quality sequences using SolexaQA . Then, the pair-end clean reads were de novo assembled into contigs by Trinity. After that, TGI Clustering tools were used to obtain the non-redundant unigenes .
Functional annotation of unigenes was performed using BLASTX searches against the NR, Swiss-Prot and COG databases (E-value < 1e-5). The GO annotations for the unigenes were determined using Blast2GO . Then, the results were submitted to WEGO to obtain the GO classification graph. The raw reads were deposited in the NCBI SRA (Short Read Archive) with the accession number SRA069852. The assembled unigenes are shown in Additional file 4.
The reads of the fertile and sterile samples were mapped back to our de novo assembling results separately using Tophat . Then, the differentially expressed unigenes were obtained using Cufflinks . The results were submitted to KAAS and Blast2GO for further interpretation.
First-strand cDNA was synthesized using the RevertAid First Strand cDNA Synthesis Kit (Thermo, USA), from 1,200 ng total RNA. Gene-specific primers were designed based on the selected unigene sequences (Additional file 5). Reactions were performed with the SYBR Green Realtime PCR Master Mix (TOYOBO, Japan) in a Bio-Rad CFX96 instrument. Three biological replicates for each sample and three technical replicates were performed and the relative expression level was calculated using the 2-ΔΔCt method. The actin gene was used to normalize the gene expression.
Sterile and fertile flower buds were fixed and embedded in resin using Technovit Embedding Kits (Germany). Then, semi-thin (2.5 μm) sections were obtained using an automatic microtome (Microm HM 360, Thermo). The sections were stained with 0.1% toluidine blue O for 30–60 sec at room temperature and observed with a Nikon Eclipse 80i microscope (Nikon, Japan). Images of the anthers in different stages were captured with a Nikon DS-Ri1 camera (Nikon, Japan).
The project was submitted to NCBI BioProject with BioProject ID: PRJNA193477 (http://www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA193477). The raw reads were deposited in NCBI SRA (Short Read Archive) with the accession number SRA069852 (http://www.ncbi.nlm.nih.gov/sra/?term=SRA069852). The assembled unigenes were deposited in the Additional file 4: De novo assembled transcripts sequences. The DGEs and their annotations were deposited in the Additional file 3: Expressions and annotations of the differentially expressed unigenes.
This project was supported by the Natural Science Foundation of China (Grant No. 31371488) and the National High Technology Research and Development Program of China (Grant No. 2012AA101107).
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.