Selecting and executing specific behaviors in response to various environmental stimuli is a fundamental function of the nervous system. In this review, we introduce larval Drosophila as a model to study this process. We first review recent technological innovations in this system that allow cellular-level functional dissection of the central circuits, such as systematic single-neuron genetic targeting, all-optical neurophysiology and connectomics. We then introduce our recent studies on the structure and function of neural circuits that regulate forward and backward locomotion of the larvae. Taken together with the statistical analyses of the populational activity dynamics and mathematical modeling, system level understanding of the operation of the neural circuits is becoming possible.

Protein design holds promise for applications such as control of cells, therapeutics, new enzymes and protein-based materials. Recently, rational design of protein molecules has made a great progress, guided by the consistency principle proposed by Nobuhiro Gō in 1983: local and non-local interactions consistently favor the same folded conformation. We discovered a set of rules for designing ideal protein structures stabilized by consistent local and non-local interactions. The rules enabled the de novo design of amino acid sequences having the funnel-shaped energy landscapes toward the desired target structures. Various ideal protein structures have been created using the rules.

Raman microscope is not widely used in life sciences. This fact is attributed to the fundamental weaknesses of Raman imaging, i.e., signal weakness and difficulty in molecular identification upon a cell spectrum. Signal weakness leads to long data acquisition time in Raman imaging. We discuss the recent advances in Raman microscopy for accelerating data acquisition speed. Difficulty in spectrum interpretation is attributable to spectral overlapping of vast kinds of molecules existing in a biological sample. We present Raman tags that have characteristic spectra in the silent region, where little intrinsic cellular signal appears. We also present super-resolution techniques that can aid the spectral complexity.

Mutations in the gene coding Cu/Zn-superoxide dismutase (SOD1) are known to cause amyotrophic lateral sclerosis (ALS), a neurodegenerative disease with no cures. SOD1 is a highly stable enzyme where copper and zinc ions bind and a disulfide bond forms, but is also known to accumulate as misfolded forms in spinal motoneurons of ALS. A key to understand such pathological changes in SOD1 is the contribution of metal binding as well as disulfide formation to the conformational stability of SOD1. In this review, I will summarize mechanisms of SOD1 misfolding in ALS where the metal binding and/or disulfide formation go awry.

A cutinase-like enzyme from a thermophilic isolate, Saccharomonospora viridis AHK190, Cut190, has the ability to depolymerize polyethylene terephthalate (PET). The catalytic activity and thermal stability of Cut190 are increased by Ca²⁺ binding. The structural analysis of Cut190 mutants in complex with metal ions and substrates elucidated the reaction mechanism regulated by Ca²⁺. The metal ion-binding properties, analyzed using isothermal titration calorimetry were correlated with the effects on Cut190 activity and stability, which could be improved using protein engineering. The Cut190 mutant will be used for PET chemical recycling.