A TILLING population with a high mutation frequency will maximize efficient mutation detection and minimize the number of individuals that need to be screened to find mutations. For example, screening an 1 Kb gene in a population of 5,000 individuals with an average mutation frequency of 1 mutation per 250 Kb yields about 20 mutations in Arabidopsis (Greene et al. 2003). There is an approximately 60% chance that one of these 20 mutations will be a knockout (Henikoff et al. 2004). Increasing the population size to 12,000 individuals would raise the probability of finding a knockout mutation to 95%. But, if the mutation frequency were much lower, such as 1 mutation per 1,000 Kb, then approximately 20,000 individuals would need to be screened to find 20 mutations. In any case, a high mutation frequency must be bal anced against detrimental effects such as lethality and sterility that are inherent in mutagenic treatments.
Chemical mutagenesis has been used for decades to increase available genetic diversity in plants and animals for forward screens and it has also been used successfully for generating TILLING populations. There are a number of chemicals available for the induction oflarge quantities ofsingle base changes throughout the genome. Among the most commonly used are alkylating agents such as EMS, which induces primarily G/C to A/T transitions, and N-ethyl-N-nitrosourea (ENU), which induces mainly A/T to G/C transitions and A/T to T/A transversions (Kodym and Afza 2003). Mutagenic treatments that may work well in one species are not usually transferable to another, and even different varieties of the same crop may react differently to the same mutagenic treatment (Wu et al. 2005). As a result, the most effective chemical mutagen and treatment regime for a TILLING population must be determined empirically.
Seeds are the preferred starting material for chemical mutagenesis in tomato. The seeds are incubated in aqueous solutions of EMS or other mutagen and the optimal treatment parameters, including concentration of the mutagen and duration of the treatment, are determined. After treatment, seeds are washed extensively and sown. Lethality and delays in germination are often used as early indicators of the mutagenic effectiveness of the treatment. There is some correlation between the degree of lethality of treated seeds and the mutation frequency of the population although this correlation may vary considerably between plant species and even between varieties of the same species (Caldwell etal. 2004; Wu etal. 2005). For example, a high mutation frequency of 1 per 24 Kb can be achieved with little lethality in wheat (Slade et al. 2005). Conversely, a high degree of lethality was seen in a rice population with a mutation frequency of only of 1 per 1,000Kb (Wu etal. 2005). Therefore, factors other than the extent of DNA damage may affect viability of M1 seeds. For instance, it has been noted that different batches of mutagen can have different cytotoxic effects due to impurities (Loppes 1968) that could also contribute to lethality.
The most straightforward way to evaluate the effectiveness of different mutagenic treatments is to measure the mutation frequency directly by TILLING. For this purpose, small pilot populations of M1 plants (800 to 1,000 plants per treatment) can be grown from treated seeds. Mi plants are genetically chimeric since mutagenesis of the seed results in mutations in many different cells, but only the mutations in germ cells will be heritable and recovered in subsequent generations. Before committing to bring these pilot M1 populations to maturity, one can measure the mutation frequencies in the growing M1 seedlings by TILLING the pilot population over a small region of the genome. Although it has not been demonstrated that the mutation frequency in the somatic tissue of the M1 population can be used to predict the precise mutation frequency of the M2 population, it has been observed that a low mutation frequency at the M1 stage indicates a low mutation frequency at the M2 stage. Promising M1 pilot populations can be expanded to the M2 stage to both directly measure the mutation frequency in the M2 population by TILLING and to assess the effects of the treatment on fertility in order to determine the optimal treatment.
Was this article helpful?