The Biochemistry of GIGANTEA: A Circadian Clock-Controlled Regulator of Photoperiodic Flowering in Plants

Abstract

In many organisms, important biological and developmental processes such as growth and reproduction occur in time with particular seasons of the year. Many species recognise seasonal changes through the measurement of changes in day length (photoperiodism) by comparison to an internal oscillator called the circadian clock. Genes involved in photoperiodic processes such as flowering can be studied using the model plant Arabidopsis thaliana (Arabidopsis). One Arabidopsis gene central to photoperiodism is GIGANTEA (GI). GI mutants show a delay in flowering in inductive day lengths, leading to a longer phase of vegetative growth. Expression of GI is regulated by the circadian clock and is cyclical, peaking in the evening. The GI protein is large, plant-specific, nuclear localised, and has no homology to any other proteins. GI accumulates during the day and drops at night due to predicted proteosomal degradation. While it has been forty years since the first mutant alleles of GI were described much is still unknown about the molecular mechanism of GI action. The objective of this work was to characterise the GI protein, to give a greater understanding of the role of GI in floral induction. Soluble affinity-tagged full-length GI was expressed in Escherichia coli (E. coli) and was stabilised by the addition of the detergent N-dodecyi-β-D-maltoside (DDM) to storage and purification buffers. Stabilised GI was purified using a variety of chromatographic methods, and characterised using a selection of biochemical techniques including circular dichroism, and dynamic light scattering. This showed that purified GI contained secondary structure, but was polydisperse in solution. Limited proteolytic digests and mass spectrometry were used to identify possible GI domains. This led to the identification of a predicted 46 kDa amino terminal GI domain. A variety of potential GI domains were targeted using either a sequence-based bioinformatics approach or an experimentally-derived approach, and were expressed in E.coli. While most of these domains remained intractably insoluble, the 46 kDa amino-terminal domain showed partial solubility when expressed at 10°C and lysed in buffer containing DDM. This domain was expressed from two plant species (Arabidopsis and Rye Grass) and large-scale expression and purification protocols developed. In vivo characterisation of the amino terminal GI domain in Arabidopsis plants suggested that while the protein product of the domain may be unstable, a region of the domain could have a dominant negative effect on flowering time in wild type plants. Arabidopsis plants were also used to characterise the light responses of full-length GI protein. This showed that GI protein accumulated to high levels in plants exposed to blue, white and red light, with blue-light plants showing the highest levels of GI accumulation. GI protein was then shown to accumulate less in plants lacking the CRY1 and CRY2 blue light photoreceptors than plants with functional CRY1 and CRY2 genes.