Mutagenesis in Rice: The Basis for Breeding a New Super Plant

Abstract

The high selection pressure applied in rice breeding since its domestication thousands of years ago has caused a narrowing in its genetic variability. Obtaining new rice cultivars therefore becomes a major challenge for breeders and developing strategies to increase the genetic variability has demanded the attention of several research groups. Understanding mutations and their applications have paved the way for advances in the elucidation of a genetic, physiological, and biochemical basis of rice traits. Creating variability through mutations has therefore grown to be among the most important tools to improve rice. The small genome size of rice has enabled a faster release of higher quality sequence drafts as compared to other crops. The move from structural to functional genomics is possible due to an array of mutant databases, highlighting mutagenesis as an important player in this progress. Furthermore, due to the synteny among the Poaceae, other grasses can also benefit from these findings. Successful gene modifications have been obtained by random and targeted mutations. Furthermore, following mutation induction pathways, techniques have been applied to identify mutations and the molecular control of DNA damage repair mechanisms in the rice genome. This review highlights findings in generating rice genome resources showing strategies applied for variability increasing, detection and genetic mechanisms of DNA damage repair.

Keywords: DNA repair; Oryza sativa L; functional genomics; genetic variability; mutagenesis; mutation detection; random mutations; targeted mutations.

Figure 1

Mutant distribution according to Food and Agriculture Organization of the United Nations/International Atomic Energy Agency – Mutant Variety Database (FAO/IAEA-MVD, 2019). (A) Distribution of mutants by species; (B) Distribution of mutants by country (map).

Figure 2

Scheme for obtaining a mutant population in rice. Mutagenesis is performed with a chemical or physical seed treatment given to an M₀ population. The M₁ plants originate from each treated M₀ seed. Seeds from M₁ plants will form the M₂ population in which DNA analyses will be performed to find mutations. Seeds from M₂ plants will form the M₃ population and the next populations will be obtained in the same way. adapted from (Tai, 2017)

Figure 3

Mutagens applied in rice mutagenesis according to Food and Agriculture Organization of the United Nations/International Atomic Energy Agency – Mutant Variety Database (FAO/IAEA-MVD, 2019). (A) Chemical mutagens; (B) Chemical and physical agents; (C) Physical mutagens.

Figure 4

Genome editing using MN, ZFN, TALEN and CRISPR strategies which result in the DNA double strand break that will be repaired by homologous recombination (HR) or nonhomologous end-joining (NHEJ). HR requires the introduction of a donor DNA plasmid or a PCR product corresponding to the donor DNA. For HR should be inserted a donor DNA together with the editing system. In this hypothetical scheme a gene associated with seed yield in rice will be edited. (A) Meganuclease (MN): The DNA binding domain of the meganucleases are engineered to recognize specific target sequences. (B) Zinc finger nuclease (ZFN): Zinc finger transcription factors present the C2H2 motif with DNA recognition capability. The ZFN system function requires the fusion of ZF motifs with the restriction enzyme Fok I. Fok I presents a DNA binding domain that recognizes a 5-’GGATG-3’ sequence and a nonspecific DNA cleavage domain. For the zinc finger nuclease (ZFN) system, Fok I was engineered, and the DNA binding domain was removed. In this sense, the Fok I nonspecific DNA cleavage domain was used to construct a hybrid nuclease, through binding to the ZF binding domain. (C) Transcription activator-like effector nucleases (TALEN): The TALE DNA binding domain is formed by monomers, each monomer recognizing only one nucleotide. Each monomer is composed by 34 amino acids, that are repeated in the other monomers, except the hypervariable amino acids located at positions 12 and 13 of each monomer, which are determinants for binding to a specific nucleotide. These amino acids are known as repeat variable di-residues (RVDs). The nonspecific Fok I DNA cleavage domain, was used to construct a hybrid nuclease, through the binding with TALE DNA binding domain, forming the TALEN system. (D) Clustered regularly interspaced short palindromic repeats (CRISPR): In this system, a guide RNA (gRNA) (crRNA + tracrRNA) directs a restriction enzyme to a specific DNA sequence that will be cleaved. P, promoter; PMS, Plant marker selection; BMS, Bacterial marker selection; DBD, DNA binding domain. The plasmid introduction into rice plants can be performed directly by electroporation and biobalistic methods or indirectly via Agrobacterium tumefaciens. In this last case it is necessary to add the right and left borders in the plasmid containing the editing system.

Figure 5

Mutation detection by TILLING technique (Till et al., 2003). The wild-type and mutant DNAs are subjected to a polymerase chain reaction (PCR) using gene specific primers. The amplified DNA is combined in the same tube, amplified mutant DNA and amplified wild-type DNA. In a second tube, the wild-type DNA is added two times. Both tubes are subjected to denaturation and renaturation cycles, forming, in a mutant case, an heteroduplex, a hybrid double stranded with a mutant strand and a wild-type strand, and in a wild-type case, a homoduplex is formed with a wild-type double stranded DNA. After, an endonuclease (CEL I) treatment is performed in both DNAs (hybrid and no hybrid), the nuclease is able to find and recognize the mismatch pairs in the hybrid DNA and cleave the double strand DNA. Both DNAs are subjected to electrophoresis using a molecular marker (MM), the heteroduplex (Ht) digested by the endonuclease and the undigested homoduplex (Hm).

Figure 6

MutMap (Abe et al., 2012), MutMap Gap (Takagi et al., 2013) and MutMap+ (Fekih et al., 2013) techniques. MutMap was developed to identify mutations with polymorphic single nucleotide (SNPs) between a mutant and wild-type genotypes. Both plants are crossed and the F₁ generation is self-pollinated (⊗). DNAs are extracted from the wild-type and mutant F₂ to perform DNA sequencing followed by SNP mapping to find mutations linked to the phenotype. The MutMap Gap was developed when there is no available sequence of the desired genotype for mutation. The desired genotype (WT-X) is sequenced and aligned to the reference genome (a sequenced genotype) and the polymorphic SNPs are found and changed between WT-X and the reference genome to produce a desired reference genome. After the MutMap is followed. The MutMap+ was developed to avoid crosses between wild-type and mutant genotypes. The heterozygous plants in M₁ and M₂ generation are self-pollinated (⊗) and the homozygous M₃ plants are subjected to DNA sequencing for SNP mapping. After, an alignment with a reference genome is performed in order to detect phenotype associated polymorphic SNPs.

Figure 7

History of mutation induction in rice. Stadler’s pioneering study in inducing mutation using X-ray in barley, maize and wheat (Stadler, 1928; Stadler 1930), demonstrated the possibility in creating genetic variability through mutagenesis. It began a new era where it was possible to apply this tool in the most diverse organisms. The development of the first rice mutant in China, in 1957 (ISAAA, 2019), marks one of the important milestones in the rice mutation history. In USA (1977), the first commercial mutant variety of rice was approved (Rutger et al., 1977). After, the insertion and expression of T-DNA in rice was also reported (Raineri et al., 1990). In 1993, the autonomous element Ac, a component of the maize transposon system, was inserted into the rice genome (Shimamoto et al., 1993). In 1996, the use of retrotransposons to induce mutations in rice was reported (Hirochika et al., 1996). In 2002, one of the rice mutants that revolutionized agriculture was developed in USA. Imidazolinone herbicide-resistant rice cultivars are applied in invasive red rice control (reviewed in Sudianto et al., 2013). In 2002, draft genome sequences of the japonica Nipponbare (Goff et al., 2002) and indica 9311 (Yu et al., 2002) first appeared. In 2005, the map-based complete rice genome sequence was published, becoming the gold standard of crop genomes (IRGSP, 2005). Recently, genome editing technologies have been widely used in various organisms, including plants. The first techniques applied to rice mutagenesis used a non-specific nuclease (Fok I) associated to a DNA-specific domain, comprehend the TALEN (Transcription-like effectors nucleases) (Li et al., 2012; Shan et al., 2013) and ZFN (Zinc-finger nucleases) (Cantos et al., 2014; Jung et al., 2018). CRISPR/Cas system has been widely used to induce targeted genomic editing and has been applied in rice since 2013 (Shan et al., 2013; Jiang et al., 2013, Miao et al., 2013; Feng et al., 2013; Shan et al., 2014). Nowadays, according to FAO/IAEA records, 823 rice mutants have been officially registered since the first genotype developed in China in 1957.

Figure 8

Strategies to generate variability in rice. The improved cultivars need to fulfill requirements, i.e., be adapted to the growing environments and please consumer taste. With the long-time selection for desired traits, the genetic variability is reduced, and genotypes decrease their value due to adaptation (i.e., environments change) or consumer taste (i.e., consumers change their habits) losses. To overcome this problem, the strategies of random or targeted mutations are applied to increase rice genome variations, generating new populations with different traits that can be selected according to the aim of the breeder. With this, the aim is to produce a super rice, which responds to environmental and market demands.