Improvement of Dscam homophilic binding affinity throughout Drosophila evolution
1 State Key Laboratory of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences, 15 Datun Road, Beijing 100101, China
2 Bioinformatics Division, TNLIST/Department of Automation, Tsinghua University, Beijing 100084, China
3 Department of Computer Science and Engineering, Hong Kong University of Science and Technology, Clearwater Bay, Kowloon, Hong Kong, China
4 School of Life Sciences, Tsinghua University, Beijing 100084, China
5 Current address: Department of Neuroscience, The University of Texas at Southwestern Medical Center, Dallas, TX, USA
6 Department of Electrical, Computer and Biomedical Engineering, University of Pavia, via Ferrata 1, Pavia, 27100, Italy
BMC Evolutionary Biology 2014, 14:186 doi:10.1186/s12862-014-0186-zPublished: 27 August 2014
Drosophila Dscam1 is a cell-surface protein that plays important roles in neural development and axon tiling of neurons. It is known that thousands of isoforms bind themselves through specific homophilic interactions, a process which provides the basis for cellular self-recognition. Detailed biochemical studies of specific isoforms strongly suggest that homophilic binding, i.e. the formation of homodimers by identical Dscam1 isomers, is of great importance for the self-avoidance of neurons. Due to experimental limitations, it is currently impossible to measure the homophilic binding affinities for all 19,000 potential isoforms.
Here we reconstructed the DNA sequences of an ancestral Dscam form (which likely existed approximately 40 ~ 50 million years ago) using a comparative genomic approach. On the basis of this sequence, we established a working model to predict the self-binding affinities of all isoforms in both the current and the ancestral genome, using machine-learning methods. Detailed computational analysis was performed to compare the self-binding affinities of all isoforms present in these two genomes. Our results revealed that 1) isoforms containing newly derived variable domains exhibit higher self-binding affinities than those with conserved domains, and 2) current isoforms display higher self-binding affinities than their counterparts in the ancient genome. As thousands of Dscam isoforms are needed for the self-avoidance of the neuron, we propose that an increase in self-binding affinity provides the basis for the successful evolution of the arthropod brain.
Our data presented here provide an excellent model for future experimental studies of the binding behavior of Dscam isoforms. The results of our analysis indicate that evolution favored the rise of novel variable domains thanks to their higher self-binding affinities, rather than selection merely on the basis of simple expansion of isoform diversity, as that this particular selection process would have established the powerful mechanisms required for neuronal self-avoidance. Thus, we reveal here a new molecular mechanism for the successful evolution of arthropod brains.