High-throughput discovery of genetic determinants of circadian misalignment

Circadian systems provide a fitness advantage to organisms by allowing them to adapt to daily changes of environmental cues, such as light/dark cycles. The molecular mechanism underlying the circadian clock has been well characterized. However, how internal circadian clocks are entrained with regular daily light/dark cycles remains unclear. By collecting and analyzing indirect calorimetry (IC) data from more than 2000 wild-type mice available from the International Mouse Phenotyping Consortium (IMPC), we show that the onset time and peak phase of activity and food intake rhythms are reliable parameters for screening defects of circadian misalignment. We developed a machine learning algorithm to quantify these two parameters in our misalignment screen (SyncScreener) with existing datasets and used it to screen 750 mutant mouse lines from five IMPC phenotyping centres. Mutants of five genes (Slc7a11, Rhbdl1, Spop, Ctc1 and Oxtr) were found to be associated with altered patterns of activity or food intake. By further studying the Slc7a11^tm1a/tm1a mice, we confirmed its advanced activity phase phenotype in response to a simulated jetlag and skeleton photoperiod stimuli. Disruption of Slc7a11 affected the intercellular communication in the suprachiasmatic nucleus, suggesting a defect in synchronization of clock neurons. Our study has established a systematic phenotype analysis approach that can be used to uncover the mechanism of circadian entrainment in mice.

Synchronization to environmental changes such as day and night cycles and seasonal cycles is critical for survival. Organisms have therefore evolved a specialized circadian system to anticipate and adapt to daily changes in the environment. Loss of synchrony between the internal circadian clock and environment day and night changes is responsible for jet lag, but may also promote sleep disorders, metabolic disorders and many diseases. The availability of large amounts of mouse data from the International Mouse Phenotype Consortium provides new opportunities to identify novel genetic components of mouse behaviour and metabolism. In this study, we performed a high-throughput identification of genetic components of circadian misalignment by developing a machine learning-based algorithm. By analyzing the indirect calorimetry parameters from more than 2000 C57BL/6N mice and mice from 750 mutant lines, we identified 5 genes involved in circadian misalignment of activity and feeding behaviour. Further analyzing genetic knock-out mice for one of these genes, we were able to validate our screening method by functional studies. Our systemic analysis thus paves the way for searching the genetic determinants for circadian misalignment.