Department of Biology, Brandeis University, Waltham, MA 02254, USA
The period (per) gene of Drosophila melanogaster is a central component of the fly's circadian clock. Mutations in this gene can abolish, shorten, or lengthen locomotor-activity and eclosion rhythms. Expression of the per gene products fluctuates with 24 hour cycle durations, and these periods are changed in the mutants in concert with the behavior, indicating that the PER protein regulates its own transcription.
To monitor RNA and protein expression temporally, many individuals have to be sacrificed at a given time, making it difficult and laborious to perform high time-resolution experiments. Also, differences amongst individuals remain undetected, and it has previously been impossible to monitor gene expression in individual behaviorally rhythmic or arrhythmic flies.
The luciferase reporter allows gene expression to be monitored in living individuals and with high time-resolution over a time span of 4-7 days (Brandes et al. 1996, Neuron, 16: 687-692). To test the suitability of luciferase for tracking per expression in vivo we constructed two fusions between both genes, containing either the per promoter region (per-luc), or in addition to these regulatory sequences per DNA coding for the first 2/3 of the PER protein (BG-luc), fused to the luciferase cDNA. Flies transformed with these constructs were put in microtiter plates containing food mixed with the substrate luciferin. These plates were analyzed in an automated scintillation counter measuring the light-output of each fly every hour. Luciferase expression of both transgenes turned out to be rhythmic, although the phases were different: per-luc transgenics showed maximal expression around ZT18 (6 hours after the lights go off in a 12hr: 12hr light-dark cycle), whereas BG-luc expression consistently peaked 2 hours later. This difference indicates an influence of per coding sequences on the phase of RNA transcription. Expression analysis using standard molecular techniques (RNase Protections) verified these findings: per promoter-driven RNA accumulated earlier and cycled with a lower amplitude compared to BG-luc RNA, which oscillated identical to the endogenous per transcript, showing that per coding sequences are necessary to achieve proper mRNA cycling of this gene.
This demonstrated that monitoring luciferase expression in vivo, reflected RNA expression of reporter constructs rendering extensive molecular analyses unnecessary and making it possible now, to perform experiments that more closely connect behavior and gene expression.
Currently we use the luciferase system to isolate new clock mutants and to identify putative clock-controlled genes (i.e. ones that are rhythmically expressed) by screening the bioluminescence cycling or non-cycling of mutagenized flies, carrying different luciferase constructs.