Fig. 1: Flies avoid feeding and laying eggs on ammonia-, urea-, and putrescine-containing media via bitter-sensing GRNs.

a Tip recording performed on the taste sensilla (solid colored lines are S6 and dashed colored lines are L4) from the labella of control (w¹¹¹⁸) flies with indicated concentrations of ammonium sulfate [(NH₄)₂SO₄], ammonium chloride (NH₄Cl), urea [CO(NH₂)₂], and putrescine [NH₂(CH₂)₄NH₂] (n = 10). b Frequencies of the action potentials elicited from S6 sensilla using 100 mM of the indicated chemicals in control (w¹¹¹⁸), UAS-Kir2.1/+, and each neuron-ablated fly, including Gr5a-GAL4, Gr33a-GAL4, Gr66a-GAL4, ppk23-GAL4, and ppk28-GAL4 with UAS-Kir2.1 (n = 10–13). c Representative sample traces from panel (b). d Two-way choice feeding assay showing the preferences of control (w¹¹¹⁸) flies upon presentation of 100 mM sucrose alone versus 100 mM sucrose laced with the indicated concentrations of (NH₄)₂SO₄, NH₄Cl, CO(NH₂)₂, and NH₂(CH₂)₄NH₂ (n = 6). e Ovipositional preference assay of control (w¹¹¹⁸), UAS-Kir2.1/+, and bitter-sensing GRN-ablated flies (Gr66a-GAL4/UAS-Kir2.1) between 100 mM sucrose food and food containing 100 mM sucrose laced with 50 mM (NH₄)₂SO₄, NH₄Cl, CO(NH₂)₂, and NH₂(CH₂)₄NH₂, respectively (n = 6). The line colors indicate each chemical, as in panel (a). f Average percentage of flies exhibiting the proboscis extension response (PER) after applying 100 mM of the indicated chemical stimulus to the labellum of the control (w¹¹¹⁸), UAS-Kir2.1/+, and two aversive GRN-ablated flies (ppk23-GAL4/UAS-Kir2.1 and Gr66a-GAL4/UAS-Kir2.1). A pretest was conducted using 100 mM sucrose alone. The taste stimuli consisted of 100 mM sucrose and 100 mM of (NH₄)₂SO₄, NH₄Cl, CO(NH₂)₂, and NH₂(CH₂)₄NH₂, respectively (n = 6). All error bars represent the SEMs. Multiple comparisons were conducted using single-factor ANOVA coupled with Scheffe’s post hoc test. Asterisks indicate statistical significance compared with the control (^*P < 0.05, ^**P < 0.01).