Mitochondrial genome editing of <i>WA352</i> via <scp>mitoTALENs</scp> restore fertility in cytoplasmic male sterile rice

Abstract

Cytoplasmic male sterility (CMS) is the inability of plants to produce functional pollen due to nucleo-cytoplasmic interaction (Wang et al., 2021). The CMS line is widely used in the three-line (CMS, maintainer and restorer) system to produce hybrid seeds with heterosis (Chen and Liu, 2014). Rice is one of the three major food crops in the world, and CMS-based hybrid rice technology has greatly improved the yield, among which the wild-abortive type CMS (CMS-WA) line is the most widely used (Tang et al., 2014). WA352, also named WA352c, is a CMS-associated gene of CMS-WA rice. Luo et al. (2013) transferred WA352 with mitochondrial transit signal into ZH11, indirectly proving that WA352 can cause CMS. orf352 in CMS-RT102 rice, which has only five nucleotide differences from WA352 and thus lead to four amino acid substitutions (Tang et al., 2017). Omukai et al. (2021) deleted orf352 via TALEN with mitochondrial localization signal (MLS; mitoTALENs), but the mutants only restored pollen viability and did not restore seed setting. In this study, we utilized mitoTALENs to directly edit WA352 in CMS-WA type rice to study the pollen viability and seed setting rate of edited offspring. WA352 is a chimeric gene mainly composed of orf284 (100% nucleotide identities), unique region, orf224 (81%) and orf288 (97%) (Luo et al., 2013, Figure 1a). We designed a target in the unique region and constructed the WA352::mitoTALEN vector (Figure 1a, Figure S1). The vector was transferred into an indica rice CMS-WA line Jin23A via Agrobacterium-mediated transformation, and we obtained 17 independent T0 plants. First, we amplified an introduced hygromycin resistance gene (HPT) and the coding region of WA352 by PCR. The results showed that all T0 plants were vector positive and all achieved deletion of WA352 (Figure 1b). Previous study has shown that mitoTALENs can lead to the deletion of target and its vicinity (Omukai et al., 2021). To analyse the extent of the deletion, we conducted PCR detection on eight regions (R1-8) around WA352 (Figure 1c). Agarose gel electrophoresis results showed that #12 and #13 achieved larger deletion around WA352 than the other T0 plants. We conducted PacBio HiFi sequencing on Jin23A, #4, #13 and #15. We assembled the mitochondrial genome of Jin23A, mapped the reads of #4, #13 and #15 to the contig where WA352 related, respectively, and visualized the results with Integrative Genomics Viewer (IGV) (Figure 1d). Compared with Jin23A, we found deletions of 142 bp, 849 bp and 210 bp around the WA352 target in the #4, #13 and #15 (Figure 1d, Figure S2). Kazama et al. (2019) show that mitoTALEN-introduced DNA double-strand breaks (DSB) are repaired by ectopic homologous recombination. We preliminarily confirmed this conclusion by detecting region 9 in Figure 1c. To analyse the repair mechanism of DSB, we assembled the mitochondrial genomes of #4, #13 and #15. Through comparison and analysis with Jin23A, we found that homologous recombination events occurred at the left-side and right-side free ends, respectively. The same recombination event mediated by 559 bp repeats occurred at the left-side free end of #4, #13 and #15, while three different types of recombination events occurred at the right-side free end of #4, #13 and #15, respectively (Figure 1e). For example, a homologous recombination mediated by 52 bp repeats occurred at the right-side free end of #4 (type 1), a 19 bp repeat at the right-side free end of #13 (type 2), and a 56 bp repeat at the right-side free end of #15 (type 3) (Figure 1e). Sequence-specific primers were used to detect the results of recombination at the left-side and right-side ends by PCR and Sanger sequencing (Figure 1f, Figure S3). We also predicted several new genes at 3′-recombination sites via Open Reading Frame Finder (ORFfinder) (Table S3). These results illustrated that the deletion of WA352 mediated by mitoTALENs is produced by causing a DSB first and then homologous recombination with the same or similar sequence as the template at the free end. Next, we conducted statistical analysis on pollen viability and seed setting rate of T0 plants. Compared with Jin23A, the mutants showed different degree of restoration of pollen viability and seed setting (Figure 1g–i, Tables S1 and S2). To analyse the genetic stability, we obtained T1 plants derived from the self-pollination of T0 plants. We amplified WA352 of T1 plants by PCR, and agarose gel electrophoresis results showed that no plant could amplify WA352 (Figure 1j). Meanwhile, we conducted PCR analysis on nine regions around WA352 in T1 plants obtained from self-pollination of #4 and #15, which all achieved the deletion of WA352 (Figure 1k). Finally, the seed setting rate of #1 to #17 T1 plants were statistically analysed (Figure 1l). These results indicated recovery of pollen viability and seed setting in mutant offspring, confirming the genetic stability of mitoTALENs-mediated gene editing offspring. In summary, we successfully utilized mitoTALENs to edit rice CMS-associated gene WA352 and obtained T0 plants that restored pollen viability and seed setting, which directly confirmed that WA352 was the cause of CMS. The differences in the mutant phenotypes caused by the editing results of WA352 and orf352 may be related to the differences in the nucleus of the receptor materials where they are located, which requires further analysis. In addition, we found that the WA352-disrupted mitochondrial genome could be stably inherited in the next generation, even lacking mitoTALENs expression cassette. Our results are consistent with those of previous studies that mitoTALEN-mediated DSBs are repaired by ectopic homologous recombination, which can lead to the deletion of hundreds of bases around the target. Our results confirm that mitoTALENs can effectively realize the functional study of plant mitochondrial genes, which will lay a good research foundation for crop organelle gene editing breeding research in future. We thank Prof. Yaoguang Liu (South China Agricultural University), Prof. Tomohiko Kazama (Kyushu University), Prof. Kinya Toriyama (Tohoku University), Dr Yi Zou, and Dr Wenchuang He for their assistance in experiment and data analysis. This work was supported by the Science Technology and Innovation Commission of Shenzhen Municipality (RCYX20200714114538196), the R&D program of Shenzhen (KCXFZ20211020164207012) and the Innovation Program of the Chinese Academy of Agricultural Sciences. The authors declare no conflict of interest. Z.W. designed the study. S.J., S.A. and H.M. provided technical assistance. J.Z. performed the experiments and prepared the manuscript. L.N. and S.Z. provided assistance with data analysis. The data that supports the findings of this study are available in the supplementary material of this article. Appendix S1 Methods. Figure S1 Nucleotide sequences of the WA352 coding region. Figure S2 Three types of deletion results of WA352 and its surrounding sequences in #4, #13, and #15 T0 plants induced by mitoTALENs. Figure S3 Verification of recombination sequences in #4, #13, and #15 T0 plants. Table S1 Statistics on the percentage of darkly stained and unstained pollen in T0 plants. Table S2 Statistics on the percentage of filled and unfilled spikelets of T0 plants. Table S3 Chimeric loci generated by recombination at the 3′-site. Table S4 Primers used in this study. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts