Subfunctionalization of Ethylene Receptors and Homology Modeling of Cytosolic Domains in <i>Arabidopsis thaliana</i>

Ethylene is a gaseous phytohormone that initiates and modulates several mechanisms related to growth and development in plants through a family of five disulphide-linked receptor dimers. Although the ethylene receptors are very similar in their structures, they have diverse functions with both overlapping and non-overlapping roles. Silver ions are able to support ethylene binding to the receptors but it is also interesting to note that ethylene responses are blocked in the presence of silver. A part of the present study identified that ETR1 receiver domain has little or no role in mediating responses to silver ions, supported by data obtained from end point analysis and analyzing growth kinetics of dark grown Arabidopsis seedlings. However, previous data suggested that these receptors are important for other responses. This led to an interest in studying the structural aspects that lead to the sub-functionalization of ethylene receptors. The current study mainly focuses on looking at the structures of these domains for a better understanding of their physiological roles. As information regarding the crystal structures of different domains of ethylene receptors is only limited to ETR1 catalytic domain and receiver domain, we predicted the three dimensional protein structure using knowledge-based prediction, homology modeling.

The models generated for receiver domains showed similar tertiary structure for ETR2 and EIN4 receiver domains as compared to that of ETR1 crystal structure. The models created for kinase domains suggested that although the sub families function through different kinases, structurally they were similar to sensor histidine kinases. ERS2 had been an exception for this and the DHp domain of ERS2 is yet to be characterized well. The models predicted for GAF domains suggested that GAF domains mostly have conserved alpha helices and the models generated either long loops or very short beta strands against beta strands in the templates. We could predict the approximate structures of the different receptor domains and compared each of the predicted structures of each domain in all the receptor isoforms for a better understanding of how conformational changes in the structure result in different physiological functions.