Highlighted paper: Liu, McCormack and Sheen (2012) Targeted parallel sequencing of large genonmic regions for identifying mutations in Arabidopsis. Plant Methods 8:12
Kun-hsiang Liu, Matthew McCormack and Jen Sheen from Harvard have developed a PCR-based method of identifying mutations in Arabidopsis. It is cheaper and easier than traditional methods of identifying mutations, using bench-top PCR and a new user-friendly method of bioinformatics analysis using web-based resource Galaxy. Liu et al. estimate that using their method to identify a mutation mapped to a 550kb genomic region will cost roughly US$500, a fraction of the usual ten thousand dollar cost of currently used methods of mutant identification.
Liu et al. tested the new method of identifying mutations by searching for new nitrogen response genes. They generated an Arabidopsis thaliana line in which LUCIFERASE was driven by the promoter for nitrogen response marker NIR. Using EMS-mutagenesis, the team made 25 000 mutant NIR:LUC lines and identified seedlings that were nitrate insensitive (nis) or showed nitrate constitutive response (ncr).
When the lines were made, the phenotypes were identified and a second generation was grown. Three second generation lines – ncr1, nis1 and nis2 – were selected for further investigation.
Liu et al. used their novel TPSeq method to locate the mutations causing the ncr1, nis1 and nis2 phenotypes. They started by using existing simple sequence length polymorphism (SSLP) and cleaved amplified polymorphic sequences (CAP) molecular markers to narrow down the genomic regions that would be sequenced to search for the mutation. Using the method outlined in Jenta et al. (2002), Liu et al. narrowed down the sites of the three mutations to three non-overlapping genomic regions of between 400 – 700 kb.
From the three clones, the team made many PCR-amplified DNA fragments of around 7kb. These small fragments covered 99.7% of the targeted regions. They were separated by agarose gel electrophoresis and the non-specific PCR products were removed. The specific PCR products were re-suspended to a standard concentration. Liu et al. made three pools of PCR fragments, representing the three regions of DNA where the mutations were expected to be found. The fragments in the pools were sheared to 200bp, and ligated to adaptors for NGS in an Illumina HiSeq 2000 genome analyser.
The NGS output was 8.5Gb of sequence. Liu et al used web-based resource Galaxy to analyse the data. The method they used is written in detail as a simple step by step guide in the paper, so anyone can use it.
The most expensive step in the protocol, and the step most labs will not be able to do in-house, is Illumina sequencing. The cost of this is reduced compared to other methods, as it is possible here to sequence targeted genomic fragments from dozens of mutants in just one lane. I haven’t had a personal recommendation of this method, but the paper is persuasive and the methods are clear. If you regularly do forward genetics this method may become invaluable.
