Figure 4. DN2 neurons are sufficient to generate daytime TPR in LD.

(A-B) Clk9M-Gal4; pdf-Gal80 is selectively expressed in DN2s. (A)Clk9M-Gal4::UAS-nlsGFP (nuclear GFP) flies were stained with anti-GFP (green) and anti-PDF (blue). The PDF antibody labels the s-LNv (arrowhead) and l-LNvs neurons. Clk9M-Gal4 labels both DN2 (arrow) and s-LNv neurons (arrowhead). 2 DN2 cells are observed in Clk9MGal4::UAS-nGFP flies, but GFP staining was frequently weaker in one of them.

(B) Clk9M-Gal4, pdf-Gal80/ UAS-cd8GFP flies are stained with anti-GFP and anti-PER. The PER antibody labels circadian pacemaker cells in the brain. pdf-Gal80 eliminates the expression in s-LNv neurons. Clk9M-Gal4, pdf-Gal80/ UAS-cd8GFP is selectively expressed in the DN2 neurons. Upper and lower pictures show different Z-stacks of the same brain.

(C–F) PER expression in DN2 neurons is sufficient to rescue daytime TPR of per⁰¹ in LD. The line graphs (C, E) show preferred temperatures of w¹¹¹⁸ and rescue fly lines in the null mutant per⁰¹ background during the daytime. The same LD daytime data from Figures 1B, and 2F were used for w¹¹¹⁸ and per⁰¹, respectively. PER expression in Clk9M-Gal4, pdf-Gal80 (E, blue) is sufficient to rescue the per⁰¹ mutant phenotype for daytime TPR (ZT1–12), while PER expression in Clk9M-Gal4 (C, red) partially rescues it. The bar graphs (D, F) show the daytime TPR (ZT1–12) for each genotype. ANOVA was performed on the following groups; w¹¹¹⁸, per⁰¹;;UAS-per/+, per⁰¹; Clk9M-Gal4/+, and per⁰¹; Clk9M-Gal4/+;UAS-per/+ (D), as well as w¹¹¹⁸, per⁰¹;;UAS-per/+, per⁰¹; Clk9MGal4/+; pdf-Gal80/+, and per⁰¹; Clk9M-Gal4/+; pdf-Gal80/UAS-per (F), Tukey-Kramer test comparison to ZT1–3 (C, E) and each control fly line (D, F). ***P<0.001, **P<0.01 or *P<0.05.