SINGLE-SEED RESPIRATION PATTERNS REVEAL MULTIPLE SUBPOPULATIONS IN SEED LOTS
Bello, P.H.N.1, Marks, P.2 and Bradford, K.J.1
1Department of Plant Sciences, Seed Biotechnology Center, University of California, Davis, USA
2Aginnovation LLC, Lodi, CA, USA
Contact: Pedro Bello, email@example.com
Germination time courses can be analyzed to quantify seed responses to diverse environmental and physiological conditions such as temperature, water potential, oxygen, hormones, dormancy, aging, and other factors. However, obtaining sufficiently detailed data on germination timing of seed populations requires repeated observations at frequent intervals, which is labor-intensive and often impractical. Respiration is among the earliest metabolic processes initiated following hydration of a dry seed, and respiration rates have been linked with seed quality. Instruments have been developed to monitor oxygen consumption of individual seeds at frequent (hourly) intervals following imbibition, providing complete respiratory time courses for populations of individual seeds in an automated manner. Conversion of these oxygen consumption curves into population oxygen depletion (POD) curves results in time courses that closely resemble germination time courses. Median rates of germination and of respiration were linearly correlated across diverse conditions for lettuce, tomato and radish seeds. In addition, the population-based threshold models that have been used to analyze germination rates can be applied directly to the analysis and quantification of POD time courses as well. Furthermore, having data for every seed in a measured population provides sufficiently detailed information to enable the identification of sub-populations of seeds having distinct respiratory behavior within a single seed lot. Testing of many commercial seed lots of maize, for example, revealed that they are often composed of multiple subpopulations with widely differing respiration characteristics. Subpopulations of seeds in lot can arise from many sources, such as location on the mother plant or in the fruit, maturity at harvest or dispersal, age or storage conditions, and physical mixtures. Modeling these subpopulations separately and summing their individual contributions allows apparently complex germination/respiration time courses to be accurately predicted. The existence of multiple subpopulations can often explain cases in which the assumption of a single normal distribution of seed sensitivity thresholds does not match well to the data. As subpopulations can vary in the medians and variances of their germination behavior, and can respond independently to germination-influencing factors, a multi-population threshold model can describe a wide array of apparently non-normal germination distributions that have been observed.