CRISPR/Cas9‐mediated genome editing of <i>MaACO1</i> (aminocyclopropane‐1‐carboxylate oxidase 1) promotes the shelf life of banana fruit

Abstract

Banana is a fruit with high nutrient content and high economic importance. It is regarded as a main staple food in developing countries. As a typical climacteric fruit, banana will ripen and decay in one week after exogenous ethylene induction. The short shelf life of banana largely limits its storage, transportation and marketing and causes great postharvest loss (Krishnakumar and Thirupathi, 2014). The shelf life of banana is closely related to ethylene production, which is the first factor considered for developing postharvest preservation technology. A reduction of endogenous ethylene production or impaired ethylene signal transduction by genetic modification might be highly efficient methods to delay the ripening process (Elitzur et al., 2016). In climacteric fruit, ethylene synthesis is controlled by the transcription of 1-aminocyclopropane-1-carboxylic acid (ACC) synthase (ACS) and ACC oxidase (ACO) genes. ACO is responsible for converting ACC into ethylene by a reduction reaction. In banana, Inaba et al.(2007) showed that MA-ACO1 (GenBank No. X91076) might play a more important role in banana fruit ripening than MA-ACO2 (GenBank No. X95599) according to expression analysis. Another two ACO homologs, named Mh-ACO1 (GenBank No. AF030411) and Mh-ACO2 (GenBank No. U86045), show high induction by ethylene and specific expression in flower and fruit tissues and also imply their close involvement on banana fruit ripening (Do et al., 2005; Huang et al., 1997; Liu et al., 1999). A recent report indicated that there are 18 ACC oxidase homologous genes on Musa acuminata genome (Xia et al., 2016). Further analysis indicated that the encoded sequences of MA-ACO2 and Mh-ACO1 partially matched the annotated sequence of Ma00_t04770.1 and Ma06_t14370.1, respectively. MA-ACO1 and Mh-ACO2 are Ma07_t19730.1 paralogues; their nucleotide sequences exhibit greater than 98% similarity with Ma07_t19730.1. To determine the role of MaACOs in fruit ripening, we performed RNA-seq of mature green fruit under ethephon or 1-methylcyclopropene (1-MCP) treatment. The results showed that among the MaACO gene family, Ma07_t19730.1 and Ma03_t02700.1 showed high transcript abundance under natural conditions in both pulp and peel tissues (day 0), and the highest transcript abundance of Ma07_t19730.1 and Ma03_t02700.1 occurred in pulp and peel tissues, respectively, in all of the examined MaACOs (Figure 1a, day 0). Moreover, the transcripts of these ACO genes were strongly induced by ethephon and were inhibited by 1-MCP to a greater extent in Ma07_t19730.1 in both the pulp and peel (Figure 1a), which suggests that Ma07_t19730.1 might play a more important role than Ma03_t02700.1 in banana fruit ripening. Xia et al.(2016) reported that RNAi of Mh-ACO2, the Ma07_t19730.1 paralogue, in Musa acuminata (AAA group, cv. Grand Nain) has strong effects on the expression of genes involved in ethylene signalling in ripening banana fruits. To investigate whether Ma07_t19730.1 is responsible for ethylene synthesis in Musa acuminata (AAA group, cv. Brazilian) and affects fruit ripening postharvest, MaACO1 (Ma07_t19730.1)-disrupted mutants were developed using the Clustered Regularly Interspaced Short Palindromic Repeats/CRISPR-associated endonuclease 9 (CRISPR/Cas9) system of banana that our group recently established (Hu et al., 2017). We selected one sgRNA ‘-5’-CCTCATGGATGAAGTGGAGAAGG-3’-’ that specifically targets the second exon of MaACO1 (Figure 1b). The sgRNA driven by an OsU6b promoter was introduced into the pYLCRISPR/Cas9Pubi-H vector (Ma et al., 2015) to obtain a pCRISPR/Cas9-ACO1T vector (Figure 1c). The embryogenic cell suspension (ECS) induced from Musa acuminata (AAA group, cv. Brazilian) was used to perform the genetic transformation (Hu et al., 2013). Six independent lines were regenerated from rooting medium with hygromycin B. After the integration of the vector cassette fragment was checked using PCR, healthy plantlets were transplanted and cultivated in a greenhouse until fruit maturation. During the whole growth cycle, the MaACO1-disrupted lines exhibited similar growth and development as the wild type except for slightly lower height. For fruit traits, the length and weight of a single fruit finger in MaACO1-disrupted lines were about 15% shorter and 5%−14% lighter, respectively, than those of the wild type, and results in a little bit lower yield in MaACO1-disrupted lines (Figure 1d,e). To determine the impact of MaACO1 on fruit ripening, two of six mutant lines were evaluated in the field. We compared the ripening process of the wild-type and two MaACO1-disrupted lines during postharvest storage at room temperature. Wild-type fruit completely ripened 21 days after harvest, whereas the ripening process of MaACO1-disrupted fruit was greatly delayed in the ripening transition period postharvest (Figure 1f). The wild-type fruit was yellow with brown speckles at day 21, whereas MaACO1-disrupted fruit remained yellow or green with no speckles (little damage), even at day 60 (Figure 1f). Importantly, when the banana fruit of the wild-type and MaACO1-disrupted plants was treated with ethephon, they exhibited a similar response to ethephon and ripened normally (Figure 1g). Moreover, the ripening process of the MaACO1-disrupted fruit was delayed for about 1–2 days compared to that of wild-type fruit, implying improved shelf life for the MaACO1-disrupted fruit after ethephon treatment. In addition, the TSS, total sugar and total acid contents of these two lines were also comparable for completely ripe fruits, and the MaACO1-disrupted fruit has more vitamin C than that of wild type (Figure 1i), suggesting that MaACO1 strongly affects banana shelf life and is an ideal target gene to edit for improved shelf life. To identify the mutation pattern, six generated lines were used for deep amplicon sequencing with PCR products of the target regions amplified from genomic DNA. Indels occurring at the Cas9 cleavage sites were considered mutations. The mutant lines exhibited different indel patterns at different cleavage sites (Figure 1h). Four mutant lines are three allelic genes disrupted mutants, and 98.43% and 98.61% of the reads had deletion mutations in the MaACO1-disrupted line 1# and line 2#, respectively (Figure 1h). These data suggested that the target site of MaACO1 was efficiently edited. As MaACO1 is mainly responsible for ethylene synthesis, we monitored the ethylene production of one wild-type line and one MaACO1-disrupted line (Figure 1j,k). Under natural ripening conditions, high amounts of ethylene were produced 18–21 days postharvest in wild-type fruits (Figure 1j), whereas, in the MaACO1-disrupted line, ethylene production was strongly delayed and reduced (Figure 1j). Under ethephon treatment, ethylene production exhibited a similar trend between the wild-type fruits and MaACO1-disrupted lines, but in the wild-type line, more ethylene was produced at day 3 compared with the MaACO1-disrupted line (Figure 1k). In summary, we created several MaACO1-disrupted plants with different editing patterns using the CRISPR/Cas9 system. The mutant fruits exhibited reduced ethylene synthesis and extended shelf life under natural ripening conditions. Moreover, MaACO1-disrupted fruit was also sensitive to ethephon and ripened normally after ethephon treatment. In addition, the vegetative growth, lifecycle and fruit quality of the MaACO1-disrupted line were comparable to those of wild-type plants except for slightly lower height and yield. These data suggest that MaACO1 is an ideal target for creating fruit with a long shelf life using the CRISPR/Cas9-mediated editing system. The application of newly created germplasm will greatly reduce the postharvest losses and will increase the economic value of the banana industry by improving the shelf life of banana fruit. This work was supported by grants from National Key R&D Project (2019YFD1000900), National Natural Science Foundation of China (31772267, 31772289), Project from Guangzhou Municipal Science and Technology Bureau (201904020033), The program of the Common Technical Innovation Team of Guangdong Province on Preservation and Logistics of Agricultural Products (2019KJ145), Discipline team-building projects of Guangdong Academy of Agricultural Sciences in the 13th Five-Year Period (201803XX), and a special fund for scientific innovation strategy construction of high level Academy of Agriculture Science (R2017PY-JX002), and Key-Area Research and Development Program of Guangdong Province (2018B020202005). The authors declare no conflict of interest. C.H, G.Y and F.B designed the experiments; Q.Y, C.L, F.B and T.D provided genomics and bioinformatics analysis; C.H, W.H and T-X.D performed the genetic transformation of banana; O.S and G.D are responsible for plant cultivation; C.H, S.L, F.B and H.G performed the physiological and molecular experiments; C.H, G.Y and F.B supervised the project and wrote the manuscript.