Cytology And Molecular Analysis On Sterile Mechanism Of SaNa-1A CMS In Brassica Napus

Heterosis is an important way to improve rapeseed yield. Cytoplasm male sterile (CMS), as an ideal system regulating pollination, has greatly contributed to the improvement in rapeseed output. According to the origin of cytoplasm, CMS in rapeseed was classified into two types. The first type was derived from mutation or interspecies crosses during natural reproduction. The second type was caused by nucleus substitution or mitochondrial gene recombination during distant hybridization or protoplast fusion between different species.Somatic hybridization is effective in creating new CMS lines from potential mitochondrial genome recombination between two parent lines. Previously, a sterile line (SaNa-1A) containing 38 chromosomes was selected from the BC3 progenies of Brassica napus-Sinapis alba somatic hybrids by using B. napus cv. Yangyou6 as a recurrent parent. The corresponding maintainer line (SaNa-1B) was selected by crossing SaNa-1A with different varieties of B. napus. Cytology and molecular analysis on sterile mechanism of SaNa-1A were as follows:(1) Differences from the morphology of floral organs between CMS and the maintainer lines were observed. Anthers in sterile flowers were visually smaller than that in fertile flowers, and wrinkles on petals of CMS line were observed. Anthers in sterile flowers were shriveled with no pollen attached on the surface. No significant difference was observed on other organs between two rapeseed lines.(2) By comparing different anther developmental stages between sterile and maintainer lines through microscope observation, we found few anthers in sterile lines that failed to form pollen sacs, even though most of the pollen sacs in sterile lines were normally formed during early development. Morphological structures of different cells at the pollen mother cell (PMC) stage of sterile lines were similar to that of the maintainer line. Tetrads in sterile lines were covered with thick callose, and tapetum cells were severely vacuolated during the tetrad stage. In the early uninucleate stage, microspores in sterile line were vacuolated with few inclusions; tapetum was condensed and separated with the intermediate cells. In the late uninucleate stage, microspores were adhered together and embraced with the tapetum cells, which were located in the center of pollen sac. Hence, tapetum and microspores dissolved together, and no mature pollen was developed and released. The results mentioned above revealed that anther development in SaNa-1A was abnormal since the tetrad stage, and the development of microspores ceased during the uninucleate stage. The main reason for sterile was the obstacle in the degradation of tapetum cells, which hindered the necessary nutrition that microspores needed.(3) Comparison of ultra-structure of developing anthers in sterile and maintainer lines revealed an abnormality in sterile lines in the subcellular structure of PMCs since PMC stage. For instance, the volume of PMC and size of nucleus and nucleolus in sterile lines were relatively smaller than that in maintainer lines. PMCs in sterile line were observed with many vacuoles, fragmentation of endoplasmic reticulum, and abnormalities in mitochondria structure. In the tetrad stage, few PMCs were identified with meiosis; however, tetrads were vacuolated; nucleus and nuclear membrane started to degrade and vanish; the mitochondria and endoplasmic reticulum were destroyed; and tetrads were covered with thick callose. Although vacuolizations were observed in tapetum cells, they did not degrade as expected, and callose enzymes could not be secrete. Degradation of tapetum started since the uninucleate stage, and the tapetosome containing sporopollenin appeared. In the uninucleate stage, microspores were covered with accumulated sporopollenin, but organelles such as mitochondria and nucleus vanished with no inclusions. Then, the microspores were degraded together with tapetum cells. These results illustrate that deficiency happened since the beginning of anther development in sterile lines. The abnormality in tapetum cell development and delay in its degradation further inhibited the development of microspores. As such, anther abortion in sterile line occurred.(4) Physiological and biochemical analysis of developing anther and mitochondria in sterile and maintainer lines revealed that reactive oxygen species (ROS) content in sterile lines was higher than that in the maintainer line before abortion, which were significantly higher and accumulated afterward. Analysis on enzymatic activity of antioxidant enzyme system responsible for ROS cleaning revealed that the superoxide dismutase (SOD) activity increased and then decreased with anther development in sterile lines, and the SOD activity in sterile lines was lower than in maintainer lines throughout the anther development. However, the peroxidase (POD) activity in sterile lines increased with anther development, which was higher than in the maintainer line. These results indicated that SOD and POD responses were induced with the accumulation of ROS in aborted anther of SaNa-lA. Obviously, inhibition of SOD activity reduced the efficiency of ROS elimination. POD, as an enzyme with multiple functions, could inhibit the synthesis of plant auxins and could clear oxygen-free radicals. Thus, increase of POD activity in anther of sterile lines, as a response to ROS accumulation, could negatively regulate anther development. Over-accumulation of ROS in sterile line and deficiency in antioxidant enzyme system aggravated the oxidization of membrane lipids, resulting in the accumulation of malonaldehyde (MDA) in anthers. The MDA content in sterile lines was higher than in maintainer lines, which was toxic to the cells.Meanwhile, we compared the F1F0-ATPase activity, ATP content in mitochondria, and cytochrome oxidase (COX) activity in anther between sterile and maintainer lines and found that the hydrolytic activity of F1F0-ATPase was significantly downregulated with the anther development of sterile line, which was lower than in maintainer lines. Similar changes on ATP accumulation were observed in the mitochondria. COX activity in anther of sterile lines was also significantly downregulated than in maintainer lines, but increased at the uninucleate stage. These results indicate that the great accumulation of ROS in sterile lines affected anther development. Hence, the abnormalities in the structure and function of terminal oxidase, which participated in the electron transport chain of mitochondrial membrane, occurred by affecting the activity of COX and F1F0-ATPase, which inhibited ATP biosynthesis. Deficiency of proline accumulation in sterile lines also affected anther development.(5) By using the TdT-mediated dUTP Nick-End Labeling method to identify the programmed cell death (PCD) during anther development, we identified that tapetum cells in normal anthers undergo PCD since the tetrad stages and maximized it at the uninucleate stage. Then, the hybridization signals expanded outward since the uninucleate stage and prepared for the release of pollens. However, few hybridization signals were identified in the tetrads of sterile lines, indicating that PCD began. Meanwhile, PCD was not obvious in the tapetum cells of sterile lines and was identified in tapetum cells since the uninucleate stage. The clear hybridization signal around the microspores of the maintainer line during the uninucleate stage denotes great accumulation of sporopollenin. Furthermore, few accumulation of sporopollenin was identified in the microspores of sterile lines. These results validated the cytology features mentioned above and proved that the abnormity of tapetum degradation and anther development in sterile lines were the main reasons causing abortion in Sterile line.(6) RNA-seq analysis on the key stage of anther development was carried out in sterile line SaNa-1A and fertile line by using high-throughput sequencing. In total,9440 differential expressed genes (DEGs) were identified using the de novo assembled data. Bioinformatic analysis on these DEGs revealed that 18 of them related to the tricarboxylic acid cycle (TCA) in the mitochondria, electron transport chain, and oxidative phosphorylation. Most of the interested DEGs were downregulated in sterile lines compared with that in maintainer lines. For instance, the expression levels of ATP6 and ATP9 in the bud of sterile lines were relatively low, which might be one of the primary causes of anther abortion. Meanwhile,13 DEGs regulating anther development, except for TPD1, were downregulated in sterile lines compared with that in maintainer lines. The abnormal expressional pattern of these genes directly caused the different developments of anthers in SaNa-1A and SaNa-1B. Referring to the previous reports, we suspect the molecular mechanism of anther abortion in SaNa-1A should be the mutation in ATPase subunit on mitochondria genome, resulting in a new mosaic gene (open reading frame, ORF) and dysfunction of FoFi-ATPase as well as the sequential regulation on other genes functioned on electron transport chain, oxidative phosphorylation, TCA cycle, and glycometabolism. Thus, structural and functional distortion occurred in the mitochondria genome, and the signals would be transferred into the nucleus, causing the expressional changes in nucleus gene AG and the downstream genes. Finally, anther abortion in SaNa-1A occurred.Moreover,12 pentatricopeptide repeat proteins putatively related to the fertility recovery were identified. Further analysis on the expressional patterns of these candidate genes, which are related to CMS, would be helpful to the dissection of their regulation on downstream genes as well as to the support of the discovery of fertility restore gene.