• Latest News
• About RevGenUK
  • Advisory Panel
• Models
• TILLING
• Links
• Publications
• Client Portal

M3 genotyping

Identification of homozygous mutants in the M3 generation

1. From the sequencing data it can usually be determined whether the mutation in the TILLING plant of interest is homozygous or heterozygous.
2. If the mutation is heterozygous 20 M3 seed from this individual are sown. This should statistically result in approximately 6 homozygous individuals if normal segregation is occurring
3. For genotyping the individuals from a segregating line, DNA from each of the 20 M3 individuals is isolated. Several SNP detection methods can be used to determine zygosity of the individuals. We found restriction enzyme analysis the quickest and most robust technique, but it is not applicable to all sites. Alternative methods include minisequencing and microsatellite based genotyping.
4. Using single nucleotide polymorphism (SNP) detection methods homozygous lines can be identified. We routinely keep all individuals that are homozygous for the mutation, in addition 2 heterozygous individuals and 2 individuals that are homozygous WT for the mutation. This spectrum should be sufficient to facilitate comparisons in the M4 generation of WT and mutant phenotype and to test for co-segregation of genotype and phenotype. The additional harvested seed also allows the maintenance of the line for future screens.
5. If feasible a non-destructive phenotypic screen can be carried out on the M3 plants. It is advisable that genotyping of the plants is carried out first and that the plants maintained for M4 seed production. We therefore avoid all growth conditions that limit plant growth such as low nutrient supply used to promote symbiotic phenotypes.
6. If the M2 plant is homozygous for the mutation only 4 seed are sown, to provide seed-producing plants so that sufficient individuals for detailed phenotyping can be obtained. When quantitative phenotypes are studied, it may be advisable to obtain sibling seed homozygous wild type for the mutation under study, to control for the influence of background mutations. If no siblings are immediately available, a backcross can be initiated that will deliver a population segregating for the mutation of interest.

Controlling the effect of background mutations

EMS causes multiple mutations per genome, and we have estimated our TILLING population has around 1300 on average. Therefore it is important to design phenotyping experiments in such a way that the influence of background mutations can be controlled. This issue needs different measures depending on the type of data that is desired from the mutants.

A) Strong phenotype, qualitative phenotyping only
When the phenotype conferred by the mutation is strong, for example leads to a non-nodulating or non-starch accumulating plant and this level of qualitative data is sufficient, the following genetic tests can be performed to ensure the mutation under study is causing the phenotype.
1. Analyse co-segregation between genotype and phenotype in a population segregating the mutation under study. For this, 93 plants plus three controls (homozygous wild-type; homozygous mutant; heterozygous individual), should be grown and the phenotype of each plant scored. DNA is prepared from the plants in a 96 well format and the genotype of each plant determined. If complete co-segregation occurs, the mutation causing the phenotype is within less than 1cM of the mutation genotyped.
2. Obtain multiple mutant alleles in the candidate gene. If more than one independent allele results in a similar phenotype, the statistical chance is greatly increased that the phenotype observed is indeed caused by the mutation in the candidate gene.
3. Test complementation between two independent mutant alleles of the same gene. Homozygous mutant individuals representing two different alleles are crossed and the resulting F1 is genotyped. If no complementation occurs, the statistical chance that both plants both plants carry a second site mutation responsible for the phenotype is dramatically reduced. Crossing success should be monitored by genotyping for heterozygosity of both mutations in the F1.

B) Weak phenotype, or quantitative data required
The large number of background mutations often causes a change in general plant fitness affecting physiological parameters and growth behaviour. If the quantitative effect of a particular mutation on, for example, plant growth is to be determined, larger populations and statistical methods need to be applied to control the influence of background mutations.
1. Compare phenotype of siblings with similar mutational load. Quantitative phenotyping can be done on a population of siblings segregating the mutation of interest. Correlation between mutation and phenotype can be established. In addition, several homozygous WT and mutant lines can be grown and the phenotype compared between these lines. More than one line from each is necessary for better representation of the background mutational load in a particular family.
2. Backcrossing. The classical, but for most projects prohibitively lengthy procedure is to do a series of backcrossing steps to statistically 'purify' the line from background mutations. One backcrossing step reduces the mutation load by 50%, so to reach an asymptotic level of reduction one would need a minimum of 6 backcrosses, which exceeds the duration of an average research grant.