A pheromone receptor mediates 11-cis-vaccenyl acetate-induced responses in Drosophila

Abstract

Insect pheromones elicit stereotypic behaviors that are critical for survival and reproduction. Defining the relevant molecular mechanisms mediating pheromone signaling is an important step to manipulate pheromone-induced behaviors in pathogenic or agriculturally important pests. The only volatile pheromone identified in Drosophila is 11-cis-vaccenyl acetate (VA), a male-specific lipid that mediates aggregation behavior. VA activates a few dozen olfactory neurons located in T1 sensilla on the antenna of both male and female flies. Here, we identify a neuronal receptor required for VA sensitivity. We identified two mutants lacking functional T1 sensilla and show that the expression of the VA receptor is dramatically reduced or eliminated. Importantly, we show misexpression of this receptor in non-T1 neurons, normally insensitive to VA, confers pheromone sensitivity at physiologic concentrations. Sensitivity of T1 neurons to VA requires LUSH, an extracellular odorant-binding protein (OBP76a) present in the sensillum lymph bathing trichoid olfactory neuron dendrites. Here, we show LUSH are also required in non-T1 neurons misexpressing the receptor to respond to VA. These data provide new insight into the molecular components and neuronal basis of volatile pheromone perception.

Figure 1.

tod¹ and tot¹ mutants lack T1 sensilla. A, Representative raw traces of neuronal activity from trichoid sensilla of w¹¹¹⁸ (wild-type controls), tod¹, and tot¹. Single-unit recordings from the trichoid sensilla in the trichoid zone were obtained after stimulation with 11-cis-vaccenyl acetate (1%), known to selectively activate T1 neurons. The arrow represents onset, and the gray bars above each trace represent the stimulus duration (300 ms). B, The number of identified T1 or non-T1 functional types in the trichoid zone in independent trials. For tod¹, n = 69; for tot¹, n = 59; for wild-type controls, n ≥ 2000 recordings. T1 sensilla are lacking in the two mutants. Error bars represent SEM. For each data set, the statistical significance of the difference was tested using ANOVA for independent observations. ***p < 0.01 was considered significant between wild-type and two mutants animals for non-T1 sensilla in the trichoid zone.

Figure 2.

Scanning electron microscopy of trichoid sensilla from w¹¹¹⁸, tod¹, and tot¹. Trichoid sensilla, identified by their long slender morphology and sharp tips, are distributed on the anterior surface of the antenna regions in w¹¹¹⁸ (A), tod¹ (B), and tot¹ (C) animals. Scale bar, 10 μm. Number of trichoid sensilla (D) and large basiconic (E) were quantified for each genotype. n = 6 antenna for each genotype. Error bars represent SEM. ***p < 0.01 was considered significant. There is a significant difference between wild type and tod¹, but not wild type and tot¹.

Figure 3.

Expression of Or67d is defective in tod¹ and tot¹ mutants. A, Schematic diagrams of Or83b and Or67d gene fragments and the relative location of the PCR primers. Primers were chosen to span an intron to discriminate genomic contamination from cDNA products. The thick line denotes the exon sequences, and the thin line denotes the intron region of each gene fragment. The expected cDNA size of the RT-PCR products was 296 bp for Or67d and 269 bp for Or83b. The expected genomic DNA size of the RT-PCR products was 368 bp for Or67d and 630 bp for Or83b. The arrows indicate priming sites for each specific primer sets. B, Or67d in tod¹ and tot¹ mutants flies is absent or dramatically reduced (long arrow). Or83b was used as a positive control for RNA integrity and RT quality (short arrow). C, RNA in situ hybridization with antisense probes to Or83b (left panels; red colors) and Or67d (right panels; green colors) in w¹¹¹⁸, tod¹, and tot¹ mutants. The bottom panel shows antisense labeling in antennas misexpressing Or67d under control of the ELAV promoter. Or83 and Or67d signals were amplified with TSA-Plus Cy5 (red) or TSA-Plus Fluorescence systems (green).

Figure 4.

Non-T1 neurons misexpressing Or67d display VA-induced, dose-dependent activity. Typical single-unit recordings of non-T1 and T1 sensillum at different concentrations of VA in w¹¹¹⁸ controls and flies misexpressing Or67d. Action potentials from single sensilla were recorded with different stimulus concentrations of VA from 0.1 to 100%. The gray bars above each group represent duration of VA application (300 ms). The graph shows dose–response curves for wild-type T1 neurons (diamonds), non-T1 neurons misexpressing Or67d (squares), and w¹¹¹⁸ non-T1 (circles). Note dose-dependent VA activation in non-T1 neurons from flies misexpressing Or67d, but not in the wild-type non-T1 neurons. The relative overexpression of Or67d in the T1 sensilla of the pELAV-Gal4; UAS Or67d animals did not alter VA sensitivity compared with wild-type T1 neurons. The experimental data were fitted with a logistic function, derived from the Hill equation (Hill, 1910). The statistical means and SEs were calculated using the values obtained from at least six independent experiments. Bottom panels, Representative raw traces from trichoid sensilla of flies misexpressing Or67d in the lush¹ mutant background. VA responses are absent in both T1 and non-T1 sensilla. Each data point represents the mean ± SEM.