Figure 2.

Activity of Pdf neurons is required for LTM maintenance. A–D, Temporal activation of Pdf neurons compensates for DD-dependent LTM impairment. Pdf-GAL4/+, UAS-TrpA1/+, and Pdf-GAL4/UAS-TrpA1 males were used. For MI, asterisks indicate a comparison between F₁ (GAL4/UAS) and GAL4 control flies, and hash marks indicate a statistical comparison between F₁ and UAS control flies. A, All experiments were performed at PT. B, On Days 2 and 3 after 7 h conditioning, flies were kept in DD. C, On Days 2 and 3 after 7 h conditioning, flies were kept at 34°C during the period between CT 0 and CT 8. D, On Days 2 and 3 after 7 h conditioning, flies were kept at 34°C during the period between CT 12 and CT 20. E, F, UAS-Kir2.1/Pdf-GAL4; tub-GAL80^ts/+ flies were used. UAS-Kir2.1/tub-GAL80^ts flies were used as the control. E, All experiments were performed at PT (25°C). F, Flies were kept at RT (32°C) for 48–72 h after 7 h conditioning. G–I, Confocal section images of whole brain and Pdf neurons. R61G12-LexA/LexAop2-mCD8::GFP (G), R14F03/UAS-mCD8::GFP (H), and R61G12-LexA/LexAop-FLPL; UAS>STOP>Kir2.1::eGFP /R14F03 flies (I) were used. Scale bars: 50 or 10 μm. Green, GFP; magenta, Pdf. J, l-LNv-specific silencing impairs LTM. R61G12-LexA/LexAop-FLPL; UAS>STOP>Kir2.1/R14F03 flies were used. For MI, asterisks indicate a comparison between l-LNv-silenced flies and the LexA/LexAop control (R61G12-LexA/LexAop-FLPL), and the hash mark indicates a comparison between l-LNv-silenced flies and the GAL4/UAS control (UAS>STOP>Kir2.1/R14F03). The permutation test was used (LexA/LexAop control vs l-LNv-silenced flies, p = 0.0011; GAL4/UAS control vs l-LNv-silenced flies, p = 0.0113). A–F, J, The Mann–Whitney U test was used for comparisons of CI. The permutation test with 10,000 random permutations was used for comparisons of MI. *p < 0.05, **p < 0.01, ***p < 0.001, #p < 0.05, ##p < 0.01. NS, Not significant; N, sample size in each box.