Gene editing by use of clustered regularly interspaced short palindromic repeats (CRISPR) has become a powerful tool for crop improvement. However, a common bottleneck in the application of this approach to grain crops, including rice (Oryza sativa), is efficient vector delivery and calli regeneration, which can be hampered by genotype-dependent requirements for plant regeneration. Here, methods for Agrobacterium-mediated and biolistic transformation and regeneration of indica rice were optimized using CRISPR-Cas9 gene-editing of the submergence tolerance regulator SUBMERGENCE 1A-1 gene of the cultivar Ciherang-Sub1. Callus induction and plantlet regeneration methods were optimized for embryogenic calli derived from immature embryos and mature seed-derived calli. Optimized regeneration (95%) and maximal editing efficiency (100%) were obtained from the immature embryo-derived calli. Phenotyping of T1 seeds derived from the edited T0 plants under submergence stress demonstrated inferior phenotype compared to their controls, which phenotypically validates the disruption of SUB1A-1 function. The methods pave the way for rapid CRISPR-Cas9 gene editing of recalcitrant indica rice cultivars.
Global climate change has increased the number of severe flooding events that affect agriculture, including rice production in the U.S. and internationally. Heavy rainfall can cause rice plants to be completely submerged, which can significantly affect grain yield or completely destroy the plants. Recently, a major effect submergence tolerance QTL during the vegetative stage, qSub8.1, which originated from Ciherang-Sub1, was identified in a mapping population derived from a cross between Ciherang-Sub1 and IR10F365. Ciherang-Sub1 was, in turn, derived from a cross between Ciherang and IR64-Sub1. Here, we characterize the qSub8.1 region by analyzing the sequence information of Ciherang-Sub1 and its two parents (Ciherang and IR64-Sub1) and compare the whole genome profile of these varieties with the Nipponbare and Minghui 63 (MH63) reference genomes. The three rice varieties were sequenced with 150 bp pair-end whole-genome shotgun sequencing (Illumina HiSeq4000), followed by performing the Trimmomatic-SOAPdenovo2-MUMmer3 pipeline for genome assembly, resulting in approximate genome sizes of 354.4, 343.7, and 344.7 Mb, with N50 values of 25.1, 25.4, and 26.1 kb, respectively. The results showed that the Ciherang-Sub1 genome is composed of 59–63% Ciherang, 22–24% of IR64-Sub1, and 15–17% of unknown sources. The genome profile revealed a more detailed genomic composition than previous marker-assisted breeding and showed that the qSub8.1 region is mostly from Ciherang, with some introgressed segments from IR64-Sub1 and currently unknown source(s).