Memory of temperature in the seasonal control of flowering time: An unexplored link between meteorology and molecular biology

Abstract

Organismal response to a specific environmental signal should have a property that is sensitive to the useful cue but insensitive to other erroneous noises. Because adaptive characteristics of an organism have been shaped by natural selection, their functions should be evaluated in the natural habitats. In the study of flowering time, recent emergence of in natura systems biology i.e. combined applications of molecular biology and statistical modeling in natural habitats, bridges biometeorology and molecular biology. Especially, expression of genes with known function can be analyzed by conventional time-series analyses that have been frequently used in biometeorological studies. Here, by comparing knowledge from (1) meteorology, (2) molecular biology (3) in natura systems biology, and (4) biometeorology, we showed how signal properties in natural temperature fluctuation correspond to mechanistic property of flowering-time regulation. Meteorological data indicate that seasonal signal in temperature is a pattern over multiple weeks or months with short-term serious noises. It predicts that molecular machinery of flowering-time control should refer past temperatures for 4–6 weeks or longer. Molecular biology on the regulation of a key flowering repressor, FLC, in Arabidopsis thaliana revealed that molecular machinery of vernalization response is a representative example that corresponds to the meteorological properties of the seasonal temperature signal. In natura systems biology has revealed that the machinery of vernalization response serves as memory of past temperature that extract seasonal signal from natural complex fluctuation. The referring periods of past temperatures suggested by meteorology, molecular biology and in natura systems biology correspond to the lengths of proceeding periods during which temperature affect first flowering dates of diverse plants. Therefore, we should not take yearly variation of flowering date as a mealy passive developmental effect of preceding temperatures. It includes more active responses of plants that have been evolved to control flowering-time in the noisy temperature fluctuations in the natural habitats.