Food odors trigger Drosophila males to deposit a pheromone that guides aggregation and female oviposition decisions

  1. Chun-Chieh Lin
  2. Katharine A Prokop-Prigge
  3. George Preti
  4. Christopher J Potter  Is a corresponding author
  1. Johns Hopkins University School of Medicine, United States
  2. Monell Chemical Senses Center, United States
  3. University of Pennsylvania, United States
9 figures, 3 videos and 2 additional files


Figure 1 with 8 supplements
Identification of an apple cider vinegar odor induced post-stimulus aggregation behavior mediated by males.

(A, B) Schematic of behavior setup and experimental design. (C) Fly tracking for 5-min of 25 male and 25 female wild-type flies. Flies are highly attracted to apple cider vinegar food odor, which gives rise to a post-stimulus aggregation behavior in the absence of exogenous odorants (right). (D) The lack of flies with apple cider vinegar stimulation (left) led to a lack of a post-stimulus aggregation (right). (E) Humidified air vs dry air is attractive (left), but does not lead to a post-stimulus aggregation (right). (F) Post-stimulus response summary (p = 0.0002 and 0.00001 for apple cider vinegar only and humidified air + WT, respectively; t-test, n = 3–6 per trial) (G) Definition of attraction index, A.I. Error bars indicate ±s.e.m. throughout. (H) Schematic of 4-quadrant arena using different populations of depositor and detector flies. (I) Different depositor fly populations (females + males, females only, males only) were used as pheromone sources and assayed for post-stimulation aggregation by female + male detector flies (p = 0.003; t-test, n = 3–5 per combination).
Figure 1—source data 1

Source data for bar graphs shown in Figure 1.
Figure 1—figure supplement 1
Basic characterization of the four-field olfactometer.

Fly tracking for 5-min of 25 male and 25 female wild-type flies to dry air (left) and insect repellent citronellal (right). Example single fly tracks within the dataset are shown color-coded from start (blue) to end (red) of a continuous track.
Figure 1—figure supplement 2
Colormap of all fly trajectories from 0 min to 7 min.

Apple cider vinegar was applied at 0.5 min. Right, attraction index over time is shown.
Figure 1—figure supplement 3
Four-field behavioral control experiments.

(A) Wild-type flies stimulated with apple cider vinegar, humidified air, and apple cider vinegar with humidified air perfusion in the other 3 quadrants (apple cider vinegar +3 humidified air). (B) Orco mutant flies responding to apple cider vinegar, humidified air and apple cider vinegar +3 humidified air.
Figure 1—figure supplement 4
Summary of four-field olfactometer controls.

Response summary of wild-type and orco mutant flies to apple cider vinegar, humidified air and apple cider vinegar +3 humidified air. n = 4–6 for each condition ±s.e.m.
Figure 1—figure supplement 5
Time-course of aggregation pheromone responses.

Recordings of post-stimulus aggregation in 5-min intervals up to 25 min post clean air perfusion.
Figure 1—figure supplement 6
Post-stimulus aggregation induced by various concentrations of apple cider vinegar.

Post-stimulus aggregation was assayed using different dilutions of apple cider vinegar.
Figure 1—figure supplement 7
Post-stimulus aggregation induced by additional food odors, but not by an attractive odorant.

Quantification of post-stimulus aggregation by food-odors (apple cider vinegar, banana and yeast), and by ethyl acetate, an attractive component of apple cider vinegar (p < 0.001 for post-stimulus phase ethyl acetate; t-test, n = 4–5 per odor stimulation).
Figure 1—figure supplement 8
The aggregation pheromone is similarly attractive to virgin or mated males and females.

Post-stimulus aggregation to natural pheromone is similar among virgin males, virgin females, virgin male + virgin females (1:1) and premixed males + females (1:1).
Figure 2 with 4 supplements
Post-stimulus aggregation requires Orco-dependent olfactory signaling.

(A) Diagram indicating the different genetic components required for gustatory or olfactory-based pheromone detection. (B) Post-stimulus aggregation responses by gustatory receptor (Poxn; p = 0.2258; t-test) and ppk-23 mutants (p = 0.0951; t-test, n = 4–5 per trail). (C) Post-stimulus aggregation responses by olfactory receptor (orco), and Ionotropic receptor (Ir8, Ir25a) mutants (Ir8a−/−;Ir25a−/−: p = 0.1524; orco−/− : p < 0.001; Ir8a−/−;;orco−/−: p = 0.004; t-test, n = 4–6 per genotype). Wild-type flies were used as pheromone depositors.
Figure 2—source data 1

Source data for bar graphs shown in Figure 2.
Figure 2—figure supplement 1
Ablation of the Ir64a + acid-sensing neurons increases post-stimulus attraction.

Post-stimulus attraction responses of flies are increased when acid-sensing olfactory neurons are removed (Ir64a-Gal4/UAS-hid)(p = 0.021; t-test, n = 3–4 per trial ±s.e.m.).
Figure 2—figure supplement 2
Repulsion of orco mutant flies to the aggregation pheromone is likely mediated by acid sensing of residual apple cider vinegar.

Post-stimulus repulsive behavior is abolished in orco mutant flies assayed with neutralized apple cider vinegar (pH = 7.0).
Figure 2—figure supplement 3
Acidity of apple cider vinegar is not required for post-stimulus responses.

Post-stimulus responses using neutralized apple cider vinegar (pH = 7).
Figure 2—figure supplement 4
Mutating components of the cis-vaccenyl acetate pheromone pathway does not disrupt post-stimulus responses.

Quantification of post-stimulus aggregation in mutants of key components in the cVA signaling pathway (p = 0.4405, one-way ANOVA test, n = 3 for each trial). Error bars indicate ±s.e.m. throughout.
Figure 3 with 4 supplements
9-Tricosene is a food-odor induced pheromone.

(A) Schematic of pheromone extract paint experiment. (B) Hexane extracts of the pheromone quadrant were used to paint the letter ‘E’ onto the glass plate. Shown are traces of naïve new flies in the painted arenas by deposited pheromone extract (apple cider vinegar + flies) or control (humidified air + flies). The blue to red color trace indicates a single fly track from start to end of tracking. (C) GC–MS results of hexane extracts from quadrants stimulated by apple cider vinegar-only, humidified air and flies, and apple cider vinegar with flies. Peak #2 is (Z)9-tricosene. 9-Tricosene exhibited a 2.8 fold enrichment on the glass plates upon food-odor stimulation. (D) Olfactory behavioral response of flies to 0.1% 9-tricosene. (E) Dose-response curve of 9-tricosene for mediating attraction. (F) Traces of flies in response to 9-tricosene deposited in an ‘E’ pattern on the glass plate. In this context, flies appear to be repelled by a concentrated 9-tricosene pattern and prefer to trail the 9-tricosene pheromone border. The behavioral differences between (B) and (F) maybe modulated by additional pheromone components present in the hexane extracts, or reflect that 9-tricosene trailing occurs only over a narrow odor range.
Figure 3—figure supplement 1
Single fly trajectories of painted ‘E’ experiment.

The blue to red color trace indicates a single fly track from the start to end of one continuous track. See also Video 2.
Figure 3—figure supplement 2
The aggregation pheromone is heat-sensitive.

To determine effects of heat on pheromone stability, the arena was maintained at 32°C for the entire testing period (red). The wild-type response at 25°C is shown for comparison.
Figure 3—figure supplement 3
Additional examples of flies responding to a 9-tricosene ‘E’ pattern.

Traces of flies in response to 9-tricosene deposited in an ‘E’ pattern on the glass plate.
Figure 3—figure supplement 4
Apple cider vinegar stimulation of oenocyte-less males leads to a reduction in post-stimulus aggregation responses.

Post-stimulus behavior is reduced when using mixtures of oenocyte-negative males (UAS-hid; PromE(800)-Gal4,Tub-Gal80 ts) with wild-type females as the source of the aggregation pheromone (p = 0.0021; t-test, n = 4 per trial ±s.e.m.).
Figure 4 with 3 supplements
Electrophysiological results identify Or7a as the receptor for 9-tricosene.

(A) Electroantennography (EAG) traces of wild-type and orco−/− flies stimulated with 100% 9-tricosene. (B) EAG response summaries of different 9-tricosene concentrations in different sexes of wild-type and orco−/− flies (n = 5–7 per stimulation). (C) Single sensillum recording (SSR) in all orco-positive antennal and maxillary palp sensilla. n = 3–6 per sensillum. (D) SSR traces showing responses to 9-tricosene stimulation in ab4 (9-tricosene responsive), ab3 empty neuron (halo/halo;Or22a-Gal4), and ab3 rescue (halo/halo;Or22a-Gal4/UAS-Or7a) sensilla. (E) SSR response summary to 9-tricosene of native ab4 and rescued ab3 sensilla in different sexes. n = 7–8 per sensillum. (F, G) SSR trace responses and quantitative summary of ab4 sensillum of Or7a mutant flies stimulated with 100% 9-tricosene and geosmin (n = 4). Error bars indicate ±s.e.m. throughout.
Figure 4—source data 1

Source data for line and bar graphs in Figure 4.
Figure 4—figure supplement 1
Fly odors can stimulate ab4A neurons.

(A) SSR trace of ab4 responding to fly odors after apple cider vinegar stimulation. Flies were housed in a glass vial and stimulated with apple cider vinegar to induce pheromone deposition. (B) Box plot of changes in ab4 spike frequency to vials containing flies stimulated with dry air, or vials containing flies that had been stimulated with apple cider vinegar (p = 0.03; t-test, n = 5 per condition).
Figure 4—figure supplement 2
Response of ab4 sensillum to cVA.

(A) SSR response traces and (B) activity summary of WT ab4 sensillum to different dilutions of cVA (n = 4 per concentration ±s.e.m.).
Figure 4—figure supplement 3
Generation of Or7a mutant.

(A) Schematic of constructs and crosses utilized for accelerated homologous recombination of the Or7a locus (See Materials and methods for details). (B) PCR verification of Or7a mutant. Primer locations are diagramed in (A).
Figure 5 with 3 supplements
Or7a neurons are necessary for the behavioral response to naturally deposited aggregation pheromone and 9-tricosene.

(A) Immunostaining of Or7a-expressing neurons innervating the DL5 glomerulus in the antennal lobe (Or7a-Gal4/UAS-mCD8GFP). (B, C) Four-field behavior responses of WT, Or7a mutant, Or7a-neuron ablated, and control OrX-neuron ablated flies (Or83c, Or43a and Or88a-Gal4 x UAS-hid) to naturally deposited pheromone (p = 0.0012 and 0.006 comparing WT to Or7a neurons ablated and Or7a−/− flies; t-test, n = 4–6 per experiment). (D, E) Behavioral response of WT, Or7a mutant, and Or7a neuron ablated flies to 9-tricosene (0.1%) (p < 0.001; t-test; n = 4–5 per trial). Error bars indicate ±s.e.m. throughout.
Figure 5—figure supplement 1
Whole-animal Or7a expression pattern.

Shown is the expression profile of Or7a-GAL4 in whole animals (males and females), and in dissected brain and ventral nerve cord. Or7a-GAL4 expression was determined by using a strong 20xUAS-mCD8GFP reporter. GFP expression was not detected in tissues outside of the antennal olfactory neurons.
Figure 5—figure supplement 2
Or7a-ablation experiments for the deposited pheromone and 9-tricosene.

Post-stimulus aggregation response of wild-type or Or7a neuron ablated flies (Or7a-Gal4/UAS-DTI) to natural pheromone (p = 0.0014; t-test, n = 4–5 per stimulation) and 0.1% 9-tricosene (p = 0.0049; t-test, n = 4–5 per stimulation). Error bars indicate ±s.e.m.
Figure 5—figure supplement 3
Olfactory assays for 7-tricosene.

Four-field behavior response (left) and summary (right) of wild-type flies to 0.1% 7-tricosene (n = 4 per odor ±s.e.m.).
Figure 6 with 2 supplements
9-Tricosene modulates female oviposition site selection.

(A) Quantification and positions of eggs laid over ∼23 hr in the 1% agarose arena with apple cider vinegar-only control or naturally deposited aggregation pheromone (A: p = 0.4836; n = 10; A′: p < 0.001; n = 9; one-way ANOVA test) (B) Quantification and positions of eggs laid over ∼23 hr in the agarose arena in blank control or with 9-tricosene (yellow, 0.001%) (B: p = 0.9499; n = 11; p < 0.001; t-test, n = 9 per trial, One-way ANOVA test). (C) The effect of 9-tricosene on female oviposition site selection was assayed in Or7a neuron ablated flies (C: UAS-hid/+, p < 0.001, n = 9 per trial; C′: Or7a-Gal4/UAS-hid, p = 0.384; n = 9, One-way ANOVA test). (D) 9-Tricosene guided oviposition site selection assayed in Or7a−/− mutant flies (p = 0.69; n = 9, One-way ANOVA test). (E) Positional recording throughout the 23 hr course of female oviposition behavior with a 9-tricosene hybrid gel (p = 0.1; n = 6, One-way ANOVA test). (F) Oviposition site selection using a 7-tricosene hybrid gel. (p = 0.28; n = 8, One-way ANOVA test). (G) Box plots indicating the total number of eggs laid in AD (p = 0.0021 comparing WT and Or7a > hid, p = 0.0001 comparing WT and Or7a−/−, p = 0.68 comparing Or7a > hid and Or7a−/ ; t-test ; n = 9–11). In all panels, colored dots indicate actual egg locations. Different colors represent different experiment trials. Error bars indicate ±2.5 s.e.m. throughout. Data points not within this range are plotted as circles.
Figure 6—figure supplement 1
Schematic of hybrid 9-tricosene gel construction.

Concentration of 9-tricosene in the agarose gel is ∼0.001%.
Figure 6—figure supplement 2
Oviposition guidance of Or7a-neuron ablated flies to 9-tricosene.

(A) Egg laying preference summary to the 9-tricosene quadrant in control (UAS-DTI/+, p = 0.0028, 0.0028 and 0.0353 to 9-tricosene quadrant; t-test, n = 8) and (B) Or7a neuron ablated flies (Or7a-Gal4/UAS-DTI, p = 0.1349; one-way ANOVA test, n = 11).
Figure 7 with 1 supplement
E2-hexenal modulates oviposition site selection.

(A) The effects of 9-tricosene guidance on egg-laying using a 3-well spot plate (34 × 85 mm) containing 9-tricosene (10−4 dilution) in a 1% agarose gel (yellow, 0.001%; p < 0.001; t-test, n = 12) or control 1% agarose gel (blue) (p = 0.53, One-way ANOVA test, n = 17). (B) Egg laying preference of Or7a−/− mutant flies in the 3-well spot 9-tricosene egg laying assay (p = 0.69, One-way ANOVA test, n = 15). (C) Egg laying preference of w1118 flies in a 3-well spot egg laying assay (10−6 dilution) (p = 1.73x10-4, t-test, n = 10). (D) Egg laying preference of Or7a−/− mutant flies in a 3-well spot E2-hexenal egg laying assay (p = 0.68, One-way ANOVA test, n = 10). For box plots (AD), error bars indicate ±2.5 s.e.m. Data points not within this range are plotted as circles. (E) Attraction of wild-type w1118 or Or7a mutant flies to E2-hexenal as determined in the four-field olfactory assay (p = 2.4x10−4 for comparing 10−6 and 10−5 E2-hexenal dilutions; p = 0.0149 for comparing WT and Or7a−/− at 10−5 E2-hexenal dilution; t-test,. n = 4–6 for each condition).
Figure 7—source data 1

Source data for box plots and bar graphs in Figure 7.
Figure 7—figure supplement 1
Oviposition selection to 9-tricosene in a surrounding presence of E2-hexenal odors.

(A) Schematic of the 3-choice spot plate oviposition assay in which the center well contains 9-tricosene (yellow) and the surrounding wells contain differing concentrations of E2-hexenal (green). (B) Box plots of the number of eggs laid per well per day in the center 9-tricosene well, or in surrounding wells that contain high (10−6, 10−7), medium (10−8) or low (10−9, 10−10) concentrations of E2-hexenal. (C) The potency of 9-tricosene relative to differing E2-hexenal concentrations in guiding egg-laying decisions. Values > 1 (dotted-line) highlight concentrations of E2-hexenal that are more potent as oviposition cues than 9-tricosene.
Figure 8 with 2 supplements
Odorants that activate Or7a guide oviposition site selection.

(A) Summary graph of Or activities induced by 7 different odorants as detected by single sensillum recordings. All odorant responses are from (Hallem and Carlson, 2006). ++++, spikes ≥200; +++, spikes ≥150; ++, spikes ≥100; +, spikes ≥50; 0, spikes ≥0; -, spikes ≤0. (B) Oviposition-guidance preference for each odorant as assayed in the 3-choice assay. An oviposition preference index (OPI) was calculated as: (# of eggs laid in odor well – average # of eggs laid in control wells) / (# of eggs laid in odor well +average # of eggs laid in control wells) See Figure 8—Source data 1. ++++, OPI ≥0.8; +++, OPI ≥0.5; ++, OPI ≥0.2; +, OPI = 0–0.2; -. OPI = 0∼-0.2; --, OPI ≤ −0.2; ---, OPI ≤ −0.5. n = 5–9 for each odor concentration. n.d., not determined.
Figure 8—source data 1

Source data for values graphically represented in Figure 8.
Figure 8—figure supplement 1
Oviposition preferences by different Or7a agonists and control odorants in a 3-well egg-laying assay.

(A-C) Odorants that activate Or7a guide positive oviposition preferences. (D-F) Odorants that do not activate Or7a demonstrate negative or neutral oviposition preferences. The final odorant concentrations in the 1% agarose gel are listed. p-values are calculated by using one-way ANOVA test.
Figure 8—figure supplement 2
Optogenetic activation of Or7a neurons in egg-laying assay.

Oviposition preference was determined in a 3-well egg-laying assay in which the center well was illuminated by red (627 nm wavelength) light. The red light activated the Or7a neurons of flies fed the co-factor all-trans-retinal (ATR), but not the same genotype of flies not fed all-trans-retinal. Genotype: Or7a-GAL4; UAS-CsChrimson. p = 0.022 by t-test. n = 8 for (−) all-trans-retinal and 10 for (+) all-trans-retinal condition.
Model of food-odor induced pheromonal behavioral responses.

Upon exposure to food odors, male Drosophila melanogaster deposit the pheromone 9-tricosene. 9-Tricosene functions via Or7a odorant receptors to guide aggregation and oviposition site-selection decisions. Activation of Or7a by other odorants may also guide similar behaviors.


Video 1
Tracking flies stimulated with apple cider vinegar in four-quadrant behavior assay.

Odor is supplied to the bottom right quadrant at time 30 s. The graph at the bottom reflects the attraction index over time. Related to Figure 1.
Video 2
Tracking flies responding to post-stimulus aggregation pheromone in four-quadrant behavior assay.

The arena has been rotated 90° counter-clockwise with the aggregation pheromone deposits located at the top right quadrant. The graph at the bottom reflects the attraction index over time. Related to Figure 1.
Video 3
Tracking flies responding to hexane extract of the post-stimulus aggregation pheromone painted as an ‘E’ pattern.

Related to Figure 3.

Additional files

Supplementary file 1

Volatile male-specific and male-enriched pheromones detected by GC–MS under control and experimental conditions.
Source code 1

Source Matlab and Arduino scripts for fly tracker analyses and optogenetic stimulations.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Chun-Chieh Lin
  2. Katharine A Prokop-Prigge
  3. George Preti
  4. Christopher J Potter
Food odors trigger Drosophila males to deposit a pheromone that guides aggregation and female oviposition decisions
eLife 4:e08688.