UAS regulatory sequences govern expression of the shakB N isoform cDNA.

Expression of shakB^N.Scer\UAS via Scer\GAL4^elav-C155, which expresses in all neurons from neural birth through adulthood, causes nearly a complete rescue of the shakB² electroretinogram phenotype. The on-transient of the rescued animals is approximately 91% of the wild-type level compared to approximately 39% in shakB² mutants. The off-transient in rescued animals is approximately 76% of the wild-type level, compared to approximately 23% in shakB² mutants. Expression of shakB^N.Scer\UAS via retina-expressing drivers, such as Scer\GAL4^ogre.PC, Scer\GAL4^GMR.PF, and Scer\GAL4^unspecified does not rescue the shakB² electroretinogram phenotype. Expression of shakB^N.Scer\UAS via Scer\GAL4^unspecified does not rescue the off-transient of shakB², although there is weak rescue of the on-transient. When these experiments are repeated in a wild-type background, wild-type electroretinograms are observed, indicating that the negative rescue results are not caused by exogenous shakB(N) interfering with endogenous innexins. Heat-shock of shakB² mutants in which shakB^N.Scer\UAS expression is induced by Scer\GAL4^hs.PB at an adult stage are indistinguishable from flies that are not heat-shocked, indicating that adult expression of shakB(N) is not sufficient for rescue of the shakB² phenotype. Flies heat-shocked during the first two-thirds of pupal development show significant rescue of off-transients to 60% of wild-type values. This represents a 2.7x increase over the off-transient values for shakB² without rescue constructs and is comparable to rescue observed with Scer\GAL4^elav-C155. In addition, these animals exhibit rescue up to five days after heat-shock is terminated.