Fig. 1.

Transgenic expression ofamn⁺ rescues the memory defect ofamn^28A mutants. A, Structure of the amnesiac locus. This schematic diagram details molecular aspects of theamnesiac gene as per Feany and Quinn (1995) with corrections from Moore et al. (1998). The 744 bp rescuing fragment described in this paper extends from 164 bp upstream of the start codon (−164) to 40 bp past the end of the open reading frame (+580). The primary transcription unit encodes a putative 180 amino acid protein with sequence features characteristic of neuropeptide precursors. These include a 25 amino acid hydrophobic domain at the N terminus, which may function as a signal sequence, and four pairs of basic residues (black boxes), which may function as protease cleavage sites. The three putative processed peptides consist of (1) a 24 amino acid N-terminal fragment with homology to the growth-hormone releasing hormone (GHRH) and to the mammalian pituitary adenylyl cyclase-activating peptide precursor (PACAP), (2) a 32 amino acid fragment with homology to mature PACAP-38 neuropeptide, and (3) a distal 57 amino acid fragment. The amn^28A mutation was produced by the insertion of a pGawB P-element 100 bp upstream of the translation start site and a second 1.4 kb insertion at position +340 (arrows). Theamn^X8 mutation is a deletion of the genomic region extending from −100 to somewhere between +451 and +587 bp. B, Genetic strategy to express GAL4-driven transgenes. Endogenous enhancer elements near the amngene presumably drive expression of theamn^28A GAL4 protein, encoded within the pGawB P-element. When expressed alone, this GAL4 protein has no effect. When expressed in flies that also carry aUAS–amn⁺ transgene, however, GAL4 binds to its UAS recognition element and drives the expression of AMN⁺. In the absence of GAL4 theUAS–amn⁺ transgene is not expressed.C, Rescue of adult olfactory memory.Memory retention was quantified in wild-type (+), mutant (amn^28A), and transgenic (UAS–amn⁺/+ andamn^28A; UAS–amn⁺/+) males immediately (0 min) or 3 hr (180 min) after training. At both time points the memory retention in transgenic males with no GAL4 expression (UAS–amn^+/+) is not significantly different from that in wild-type flies (p = 0.96 and p = 0.44) and is significantly higher than that in mutant amn^28A males (p < 0.001 and p < 0.001). In contrast, memory retention inamn^28A mutants expressing anamn⁺ transgene (amn^28A;UAS–amn⁺/+) is significantly higher than that in mutantamn^28A males (p < 0.001 and p = 0.001) and is not significantly different from that inUAS–amn⁺/+ (control) flies (p = 0.06 and p = 0.69). These data indicate that GAL4-induced expression of theUAS–amn⁺ transgene is sufficient to rescue the memory defect of amn^28Amutants. The mean ± SEM PI is plotted for males of each genotype;n = 8 PIs per group. The p values were derived from planned pairwise comparisons (α′ = 0.013) after one-way ANOVAs were done separately at each time point.