To understand the neural information processing navigating animal behaviors, integration of the in-vivo observation of neuronal activity and the computational modeling using observation data is quite useful. This strategy gives us important information to characterize the spatio-temporal dynamics of the activity in each single neuron and their roles on information processing through synaptic integration. Recent studies have succeeded to apply the integrated analysis into C. elegans sensory neurons [Kato et al., 2014]. We have focused on a C. elegans chemosensory circuit and are conducting an integrational analysis to understand the neural information processing during salt-chemotaxis behavior. By using calcium-sensitive fluorescent proteins, in this study, we monitored the activities of the ASE salt-sensing neurons and their postsynaptic interneurons, and using these data, we have developed a novel simulation model to estimate their activities in-silico.The calcium-sensitive fluorescent proteins GECO were expressed in either ASE neurons or their postsynaptic neurons such as AIY, and the transgenic animals were stimulated by the decrease or increase of NaCl concentration. To monitor and collect the patterns of neuronal activities close to worm's living conditions, we applied various types of stimuli to the transgenic animals, such as long or short duration of stimulation, tiny or large concentration change, and the rapidly-flickering concentration change. These imaging analyses imply several characteristics about one of the salt-sensing ASE neurons ASER. First, the ASER neuronal activity is described by a leaky integrator which includes an exponential decay occurred after initial activation process. Second, the activity of the ASER neuron has a nonlinear relationship with the concentration change of NaCl. Thirdly, the ASER can response reliably to rapidly-flickering stimuli as the temporal dynamics observed in the AWC neuron [Kato et al., 2014]. For the computational modeling of the ASER neuronal activity, we have developed a kind of 'leaky integrate-and-fire model', in which neuronal cells are described as multi-compartments. We confirmed that this model reproduces the neuronal activity of the ASER neuron in agreement well with imaging data. Our integrated strategy will help to understand how neurons are responded and how these responses do contribute to the spatio-temporal dynamics of the postsynaptic neuronal activity, as worms sense various types of stimulation.

Affiliations:
- Life Sci. & Bioeng., Grad. Sch. of Life & Env. Sci., Univ. of Tsukuba, Ibaraki, Japan
- Biomedical RI., AIST, Ibaraki, Japan