The availability of the B73 inbred reference genome sets the stage for high-throughput functional characterization of maize genes on a whole-genome scale. Among the 39 324 protein-coding genes predicted, the vast majority are untapped due to the lack of suitable high-throughput reverse genetic resources. We have generated a gene-indexed maize mutant collection through ethyl methanesulfonate mutagenesis and detected the mutations by combining exome capture and next-generation sequencing. A total of 1086 mutated M plants were sequenced, and 195 268 CG>TA-type point mutations, including stop gain/loss, missplice, start gain/loss, and various non-synonymous protein mutations as well as 4610 InDel mutations, were identified. These mutations were distributed on 32 069 genes, representing 82% of the predicted protein-coding genes in the maize genome. We detected an average of 180 mutations per mutant line and 6.1 mutations per gene. As many as 27 214 mutations of start codons, stop codons, or missplice sites were identified in 14 101 genes, among which 6232 individual genes harbored more than two such mutations. Application of this mutant collection is exemplified by the identification of the ent-kaurene synthase gene, which encodes a key enzyme in the gibberellin biosynthesis pathway. This gene-indexed genome-wide mutation collection provides an important resource for functional analysis of maize genes and may bring desirable allelic variants for genetic breeding in maize.

No abstract