|ADDO-QUAYE, CHARLES - Purdue University
|BUESCHER, ELIZABETH - Purdue University
|BEST, NORMAN - Purdue University
|CHAIKAM, VIJAY - Purdue University
|DILKES, BRIAN - Purdue University
Submitted to: G3, Genes/Genomes/Genetics
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/20/2016
Publication Date: 2/1/2017
Citation: Addo-Quaye, C., Buescher, E., Best, N., Chaikam, V., Baxter, I.R., Dilkes, B. 2017. Forward genetics by sequencing EMS variation-induced inbred lines. G3, Genes/Genomes/Genetics. 7(2):413-425. doi: 10.1534/g3.116.029660.
Interpretive Summary: DNA sequencing technologies have advanced to the point where sequencing of large genomes is relatively cheap and easy. This change enables novel methods for cloning genes identified in screens of mutant populations and germplasm . In this report we have shown that we can use these new sequencing technologies to clone a gene without having to perform a genetic cross. The keys to this approach are a careful selection of the lines from the pedigree to be sequenced, using the sequence of other lines to rule out error prone positions on the genome, and the type of mutations that the mutant population contains. We show that with this approach we have narrowed down the possible mutations for a dwarf phenotype to one and only one change in a known hormone biosynthesis gene. We also show that the other lines that are sequenced will be resources for easier cloning of other mutants.This new approach will greatly speed up the process for identifying the important underlying genetic components for a desired trait and will allow breeders to more rapidly develop elite lines for commercialization.
Technical Abstract: The dramatic increase in throughput of sequencing techniques enables gene cloning through pre-existing forward genetics approaches. We show that it also brings with it the potential to change the crossing designs and approach of forward genetics. To achieve this for eukaryotic organisms with complex genomes, the false positive rate of variant discovery and inherent error rates of sequencing and alignment must be accounted for by experimental design and informatics. We sequenced five lines from three pedigrees of an EMS mutagenized Sorghum bicolor population, including a pedigree segregating a recessive dwarf mutant. By comparing the sequences of the lines, we were able to identify error prone positions in the genome and eliminate them as possible causes of phenotypic variation. The spectrum of EMS induced mutations allowed us to further limit our search to changes in protein coding sequence where one and only one mutation was included in the region of interest. We took multiple approaches to defining the locus responsible for the dwarf mutation. For example, only one region contained homozygous EMS mutant alleles in dwarfs, homozygous reference sequence in homozygous wild-type plants and were heterozygous in the segregating parent. This region contained a single non-synonymous change that was homozygous in the dwarf mutants and not present in the other lines. By so doing we identified the coding sequence impacts that were private to the dwarf mutants and then performed a simple PCR test of cosegregation to confirm, by linkage and segregation, the position of the genetic locus. This also identified one, and only one, coding sequence difference for the dwarf mutation. This SNP caused a premature stop codon in the sorghum orthorlog of the giberillic acid biosynthetic enzyme ent-kaurene oxidase from rice, Arabidopsis and Pea, and segregated with the dwarf phenotype. Application of exogenous giberillic acid was able to rescue the mutant phenotype. As mapping and positional cloning were done without performing a single outcross we introduced no segregation variance and have identified an approach that can be used in systems that lack the facile human-directed crossing life history traits tha tcharacterize genetic models. We propose a more general approach for the discovery of chemically induced mutations that inverts the historical approach of using recombination to define a locus and then sequencing the coding regions within that region. Essentially inverting the typical approach and sequencing the genome of the organism first, we could then eliminate all positions that could not be responsible for phenotypic impact, in this case dwarfism. In addition the mutagenized lines that lack any detected phenotypic alterations are available as resources for mapping with a known marker set in a line that is phenotypically identical to the sequenced reference genome in sorghum. BTx623.