Optimization of a high-throughput phenotyping method for chain-forming phytoplankton species
Abstract
Modern equipment facilitates phenotyping of hundreds of strains of unicellular organisms by culturing and monitoring growth in microplates. However, in the field of phytoplankton ecology, automated monitoring of growth is not often done and this method has not been tested for many species. To meet the demand for a high-throughput technique for monitoring growth of chain-forming phytoplankton species, we have assessed and optimized a method commonly used for other microorganisms. Skeletonema marinoi is a pelagic chain-forming diatom, and we have acquired growth patterns in four different treatments (i.e., low and high light, low and high nutrient concentrations) when cultured in multi-well plates. Due to the unexpected heterogeneity in growth rates and maximum cell densities observed between wells (spatial) and runs (temporal), a set of models was fitted to the obtained phenotypic data to correct for these biases. Models were tested for robustness on two replicate multi-strain experiments including 23 different strains. Using the model accounting for temporal and spatial bias, we could reliably determine changes in growth rate caused by nutrient treatments as well as differences in cell density as a response to nutrient availability and light treatment. This method can facilitate high-throughput phenotyping of hundreds of strains, which is often a bottleneck in characterizing the ecology and capacity for adaptation of chain-forming phytoplankton.
