Because TILLING is a PCR-based technique, only genes for which the DNA sequence is known can be screened for mutations. In 2003, an International Tomato Genome Sequencing Project was launched to sequence the euchromatic regions of the tomato genome with the goal of fully annotating the estimated 35,000 tomato genes (Mueller etal. 2005b) (see Sect. 1.18). According to SGN, sequencing of the tomato genome is 17% complete as of February 2007. Designing PCR primers for TILLING any genomic region in tomato will be greatly simplified once the genome has been fully sequenced. Until then, some excellent sources for tomato genetic information include unigenes, and genomic and mRNA sequences at the NCBI (http://www.ncbi.nlm.nih.gov), and an EST collection at the SGN website (http://www.sgn.cornell.edu). NCBI contains over 270,000 nucleotide sequences from both domesticated and wild tomato species and 226,728 S. lycopersicum ESTs and mRNAs clustered into 16,978 unigenes. SGN currently has 239,593 ESTs from S. lycopersicum assembled into 34,829 unigenes.
In addition to helping investigators characterize known tomato genes, TILLING can also be used to assign function to putative tomato homologs of genes that are well characterized in other species. In this case, the investigator can clone novel tomato genes by homology and then use TILLING to select tomato plants containing mutations in the putative homologs for phenotypic study. In particular, the fully sequenced Arabidopsis genome proves useful for identifying tomato homologs of a variety of well-characterized plant genes. Van der Hoeven et al. (2002) compared a large subset of tomato and Arabidopsis ESTs and found identifiable homologs in 70% of tomato unigenes. In addition, the splice junctions in Arabidopsis genes are often conserved in tomato homologs. The online program, COnsensus-DEgenerate Hybrid Oligonucleotide Primers (CODEHOP) can be used to design degenerate primers to conserved regions ofprotein families to increase the likelihood of obtaining functional homologs (http://bioinformatics.weizmann.ac.il/blocks/ codehop.html). The first step in this process is using the Block Maker program (http://bioinformatics. weizmann.ac.il/blocks/blockmkr/www/make_blocks. html) to create "blocks" of protein homology (Henikoff et al. 2000), which are then reverse translated by CODEHOP and used to design degenerate primers for amplifying genomic DNA (Rose etal. 2003).
Once the DNA sequence for a tomato gene target is known, another online program known as CODDLE (COdons Optimized to Discover Deleterious LEsions) can be employed to predict which conserved areas of the gene would be most severely affected by mutation and also to calculate the types of alterations expected depending on the particular mutagen used (Till et al. 2003). For instance, only the codons for the amino acids tryptophan (TGG), arginine (CGA) and glutamine (CAA, CAG) can be mutated to stop codons (TGA, TAG, TAA) using EMS as a mutagen. For ENU, only the codons for the amino acids leucine (TTA, TTG), tyrosine (TAT), cysteine (TGT), lysine (AAA, AAG) or arginine (AGA) can be mutated to stop codons. Regions of a gene with the highest abundance of codons susceptible to conversion to stops are better targets for primer design if a stop mutation is desired. CODDLE thus predicts which regions of a gene will have the most severe impact on gene function and designs TILLING primers using the software Primer3 (Rozen and Skaletsky 2000).
Alternatively, if the investigator is interested in recovering all mutations in a gene regardless of severity, Primer3 can be used directly to design primers to introns or untranslated regions flanking each exon for maximum coverage of the coding region (http://frodo.wi.mit.edu/cgi-bin/primer3/ primer3_www.cgi). In the absence of genomic sequence, cDNA and EST sequences from NCBI and SGN can be used to design primers to obtain intron sequence for TILLING primer design.
Was this article helpful?