Submitted to: Plant Genetic Resources Newsletter
Publication Type: Peer reviewed journal
Publication Acceptance Date: 2/1/2004
Publication Date: 4/1/2006
Citation: Johnson, R.C., Bradley, V.L., Evans, M.A. 2006. Effect of seed sampling method during regeneration on genetic population structure and growth of model ryegrass populations. Plant Genetic Resources Newsletter. 144:1-7. Interpretive Summary: Germplasm managers must balance limitations in resources with the need to maintain the genetic integrity of accessions. This research showed that decisions regarding plant population number for seed regeneration at gene banks should consider the potential for genetic drift and allele loss based on the number of randomly mating plants rather just than on the population of plants grown. If resources are available then larger populations and techniques such as paired sampling to control male and female mating, and independent counting and storage of seeds per plant, should be employed to reduce the probability for genetic change. In many cases this may not be practical, so preparing a balanced sample; that is, a sample with an equal number of seeds from each plant is an option that will delay population changes compared to bulked seeds from plants that vary in seed production. Balanced sampling reduces the variation in seeds per plant and will also reduce genetic changes during seed regeneration cycles, but is labor intensive. However, sampling an equal number of inflorescences per plant is a way to reduce the variation in seed production among plants and thus reduce genetic drift with minimal additional inputs.
Technical Abstract: Regeneration is a critical part of managing germplasm collections and has an important and longstanding effect on the genetic integrity of heterogenetic accessions. This study was undertaken to determine how sampling methods affect genetic structure, development, and morphology in three model annual ryegrass accessions over three regeneration cycles. Harvested seed was either bulked (seeds combined proportionally according to seeds per plant) or balanced (seeds per plants equalized) over R1, R2, and R3 regeneration cycles. Analysis of eight isozymes and a total of 26 loci were completed on populations of 88 plants for each accession and sampling method. A field study measuring heading and anthesis dates, biomass, and leaf characterisitics was also completed. For both isozymes and field data, the original populations of each accession were compared to regeneration populations. For isozyme comparisons the R2 and R3 bulked populations differed from the original populations, but for balanced samples only the R3 population differed from the original. The R2 and R3 bulked regeneration populations had higher biomass, penultimate leaf length and leaf area than original populations, but the balanced regeneration populations did not differ from the original populations for any developmental or morphological factor measured. The results show that at both the isozyme and whole plant level, variation in seeds per plant associated with bulking accelerated changes in regeneration populations. Balanced sampling, however, reduced the rate of those changes. Since balanced sampling reduces the variation in seeds per plant, the effective population size (Ne) would be higher in balanced then bulked samples, leading to fewer changes over regeneration cycles. When resources are limiting balanced sampling maybe impractical, but we show that Ne can be improved with minimal additional inputs by sampling an equal number of four or more inflorescences per plant.