Development and improvement of ‘functional neural cellomics’ to elucidate the structure‐function relationships of neural networks of Caenorhabditis elegans

Background A nervous system generates complex behaviors. However, relationships between neural computation and behaviors have hardly been elucidated. Optogenetics is a powerful technique for investigating mechanisms of neural computation. However, conventional optogenetics has three problems: (1) a hypothesis is required before an experiment; (2) the throughput is low because we should generate appropriate transgenic animals to test each hypothesis; (3) single‐cell analysis is difficult because few single‐cell‐specific promoters are known. To solve these problems, we have developed a new methodology ‘functional neural cellomics’ which enables comprehensive elucidation of relationships between behaviors and neural networks of Caenorhabditis elegans . Method The core technology of functional neural cellomics is the stochastic labeling of opsin in the neural network of C. elegans . The stochastic labeling of opsin was implemented by applying the Cre‐ lox system. We designed a genetics circuit in which two sets of lox variants, lox2272 and loxP sequences, are inserted alternately downstream of pan‐neuronal F25B3.3 promoter. In addition, mCherry and a transcription factor, QF2w, are interposed between these lox sequences. After the induction of Cre recombinase by heat shock, a Cre‐ lox recombination event occurs exclusively either between lox2272 sequences or between loxP sequences. If Cre is allowed to act on lox2272 sequences, QF2w is expressed. Then, QF2w induces expression of ChR2::GFP. In this way, we can easily acquire a C. elegans library in which ChR2 is labeled in a stochastic manner in each animal. Results & Discussion By inducing Cre recombinase, we confirmed that ChR2::GFP was labeled in a stochastic manner in each animal. To verify the feasibility of functional neural cellomics, we tried to demonstrate that neurons involved in the egg‐laying behavior of C. elegans could be identified in a high‐throughput manner. The C. elegans library was irradiated with blue light to activate ChR2, and we found some animals laid eggs in a blue‐light‐dependent manner. We identified neurons producing ChR2 using a confocal microscope, and confirmed that ChR2::GFP was expressed in HSNs (hermaphrodite‐specific neurons) in the egg‐laying individuals, but was not in the non‐egg laying individuals [1]. In the above experiment, we found that the labeling rate of opsin was about 30%. If opsin is labeled in too many neurons, it becomes difficult to distinguish the function of individual neurons. To enable precise control of the labeling rate, we generated a library of randomized lox sequences by PCR. We evaluated excision efficiencies of the lox variants by using NGS, and successfully identified lox variant sequences which showed excision efficiencies ranging from 0.01% to 100%. Using these lox variants, we will be able to improve functional neural cellomics for comprehensive intervention in neural networks and high‐throughput identification of neural functions. Support or Funding Information This work was supported by PRESTO, JST (grant No. JPMJPR16F1), JSPS KAKENHI (grant No. JP17K19452), and Kyoto University Live Imaging Center.[1] Aoki et. al. Sci Rep 8 , 10380 ( 2018 )