Neuropeptides Modulate Female Chemosensory Processing upon Mating in Drosophila

Abstract

A female's reproductive state influences her perception of odors and tastes along with her changed behavioral state and physiological needs. The mechanism that modulates chemosensory processing, however, remains largely elusive. Using Drosophila, we have identified a behavioral, neuronal, and genetic mechanism that adapts the senses of smell and taste, the major modalities for food quality perception, to the physiological needs of a gravid female. Pungent smelling polyamines, such as putrescine and spermidine, are essential for cell proliferation, reproduction, and embryonic development in all animals. A polyamine-rich diet increases reproductive success in many species, including flies. Using a combination of behavioral analysis and in vivo physiology, we show that polyamine attraction is modulated in gravid females through a G-protein coupled receptor, the sex peptide receptor (SPR), and its neuropeptide ligands, MIPs (myoinhibitory peptides), which act directly in the polyamine-detecting olfactory and taste neurons. This modulation is triggered by an increase of SPR expression in chemosensory neurons, which is sufficient to convert virgin to mated female olfactory choice behavior. Together, our data show that neuropeptide-mediated modulation of peripheral chemosensory neurons increases a gravid female's preference for important nutrients, thereby ensuring optimal conditions for her growing progeny.

Fig 1. Mating state modulates the perception of polyamines.

(A) Virgin flies are less attracted to a high polyamine concentration of 1 mM (10 ppm) as compared to mated flies. Olfactory preference index of Canton S mated (♀) and Canton S virgin (☿) females in the T-maze assay. Violet and green bars represent putrescine and cadaverine, respectively. (n = 8, 60 mated (♀) or virgin (☿) flies/trial). (B) Mated females, unlike virgins, preferred relatively high concentrations of polyamines, naturally present in fermenting fruit (1 mM or 10ppm). By contrast, virgin females were most attracted to very low concentrations of polyamine. Line graph shows dose-dependent olfactory preference index of mated (♀) and virgin (☿) flies to polyamines. (n = 8 ± SEM, 60 mated (♀) or virgin (☿) flies/trial). (C) Virgin flies show no preference between polyamines (putrescine or cadaverine, 1 mM) and 1% low melting agarose, and deposited their low number of eggs on either site of the assay. (n = 8, 60 mated (♀) or virgin (☿) flies/trial). (D) Polyamine preference appears to correlate with the female’s egg-laying activity. Graphs show olfactory preference index of females 2 d post-mating (♀, 2 d), virgin females (☿), and females 14 d post-mating (♀, 14 d) to 10 ppm of polyamine. (n = 8, 60 mated (♀, 2 d), virgin (☿) and mated (♀, 14 d) female flies/trial). (E) Mated but sterile ovoD1 mutant females (ovoD1/+, Canton S) show similar attraction to polyamine odor compared to wildtype controls (+/+, Canton S). (F) Mated sex peptide receptor mutant (SPR^-/-) female flies display a significantly reduced attraction to polyamine odor (n = 8, 60 flies/trial). (G) Oviposition preference index of mated sex peptide receptor mutant (SPR^-/-) females. Mated SPR mutant females show indifference to polyamines. (n = 8, 60 mated (♀) flies/ trial). (H) Olfactory preference for 10 ppm polyamine of SPR^-/- male flies is comparable to control males. Box plots show median and upper/lower quartiles (n = 8, 60 flies/trial). All p-values were calculated via two-way ANOVA with the Bonferroni multiple comparison post-hoc test, with the exception of (E), where p-values were calculated with an unpaired t-test (ns > 0.05, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001).

Fig 2. GPCR signaling in chemosensory neurons modulates female perception.

(A) Knockdown of SPR in IR76b polyamine chemosensory neurons using RNAi (IR76b-Gal4;UAS-SPRi) significantly reduces olfactory preference to 10 ppm putrescine or cadaverine in mated females as compared to mated controls. (n = 8, 60 mated (♀) flies/trial). (B) The effect of SPRi in IR76b neurons is mating state-dependent, as knockdown of SPR (IR76b-Gal4;UAS-SPRi) does not further decrease the olfactory attraction of virgin females to polyamines compared to control virgins. (n = 8, 60 mated (♀) or virgin (☿) flies/trial). (C) Oviposition avoidance of a 1 mM polyamine/agarose substrate compared to a plain agarose substrate is strongly reduced upon knockdown of SPR in IR76b neurons (IR76b-Gal4;UAS-SPRi) (n = 8, 60 mated (♀) flies/trial). (D) Knockdown of SPR in IR41a neurons (IR41a-Gal4;UAS-SPRi) leads to a similar decrease in attraction to the odor of polyamine (10 ppm) in the T-maze as compared to knockdown of SPR with IR76b-Gal4, suggesting that SPR is required in olfactory neurons to enhance the attraction of mated females to the polyamine odors (n = 8, 60 mated (♀) flies/trial). (E) Knockdown of SPR in IR41a neurons (IR41a-Gal4;UAS-SPRi) did not affect oviposition behavior, and female behavior remained like their genetic controls. This result is consistent with the lack of expression of IR41a in taste neurons. (F) Re-expression of SPR using either IR41a-Gal4 or IR76b-Gal4 neurons fully rescued the SPR mutant phenotype of mated females in olfaction behavior to 10 ppm polyamines. (n = 8, 60 flies/trial). (G) Re-expression of SPR in IR76b taste neurons using IR76b-Gal4 fully rescued the SPR mutant phenotype of mated females in oviposition behavior. Conversely, re-expression of SPR in IR41a olfactory neurons or GR66a bitter taste neurons did not rescue oviposition preference behavior. (n = 8, 60 flies/trial). Box plots show median and upper/lower quartiles. All p-values were calculated via two-way ANOVA with the Bonferroni multiple comparison post-hoc test (ns > 0.05, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001).

Fig 3. Mating increases sensitivity of taste neurons through SPR.

(A) Scheme of the SEZ in vivo calcium imaging setup (top). Illustration of the SEZ area showing the innervation pattern of IR76b taste neuron axons (bottom). ROI 1 and ROI 2 delineate the regions of interest (ROI) used for quantification of the relative change in GCaMP-fluorescence (%ΔF/F). (B) Representative images of SEZ imaging of IR76b-Gal4; UAS-GCaMP6f mated and virgin female flies stimulated with distilled water (0 mM), 1 mM putrescine (1 mM), 10 mM putrescine (10 mM), and 100 mM putrescine (100 mM), respectively. (C) IR76b taste neuron terminals of mated females show a significantly increased response to putrescine after mating. While the response is highly significant at 1–6 h post-mating, it remains only a trend at 1 wk post-mating (n = 7). (D) Females at 1–6 h post-mating show higher IR76b taste neuron responses. GCaMP6f-fluorescence peak responses were quantified (in %ΔF/F) in the ROI 1 area. Flies were stimulated with increasing concentrations of putrescine (n = 7). (E) IR76b taste neurons of the same females as in (D) show no difference in the ROI 2 area. (F) Average response trace of the ROI 1 area (n = 7). The gray bar illustrates the stimulation period. The dark colored line in the middle presents the average value, and the light shade presents the SEM. (G) Representative images of IR76b GRN axons in the SEZ of test (IR76b-Gal4,UAS-SPRi;UAS-GCaMP5) and control (IR76b-Gal4;UAS-GCaMP5) females at 1–6 h post-mating. Flies were stimulated with distilled water (0 mM) and 10 mM putrescine (10 mM). (H) Quantification of peak responses (in %ΔF/F) of IR76b axon terminals of IR76b-Gal4,UAS-SPRi;UAS-GCaMP5 and control (IR76b-Gal4;UAS-GCaMP5) females at 1–6 h post-mating (n = 8). Box plots show median and upper/lower quartiles, and whiskers show minimum/maximum values. (I) Average response trace of ROI 1 and ROI 2 area of IR76b axons in the SEZ of IR76b>SPRi and control females (n = 8). All p-values were calculated using an unpaired t-test (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001).

Fig 4. SPR decreases sensitivity of olfactory neurons to polyamines after mating.

(A) Schematic diagram of a fly brain and its antennal appendages with olfactory sensory neurons (OSNs). OSNs project into the antennal lobe (AL), where they innervate a specific glomerulus (green). Projection neurons (PN) send the information mainly to two higher brain centers, the mushroom body (MB) and the lateral horn (LH) (top). Illustrative confocal image stack showing the IR41a and IR76b OSN innervation in the AL (bottom). VC5 is the glomerulus innervated by the polyamine-responding IR41a/IR76b sensory neurons. (B) Illustration of the in vivo calcium imaging setup. (C–E) In vivo calcium imaging of IR41-Gal4;UAS-GCaMP6f flies stimulated with water and 10 ppm putrescine, respectively. Mated females’ OSN axon terminals show a significant reduction in their sensitivity to putrescine at 1–6 h post-mating. (C) Quantification of peak ΔF responses (in %ΔF/F) in virgin and mated females. Boxes show median and upper/lower quartiles, and whiskers show minimum/maximum values. *p < 0.05, unpaired t test (n = 8). (D) Representative pseudo-color images showing the response to water and 10 ppm putrescine in virgin and mated flies at 1–6 h and 1 wk post-mating. (E) Average response trace (in %ΔF/F) of the VC5 glomerulus peak response at 1–6 h and 1 wk post-mating compared to traces from virgin females. The dark colored line in the middle presents the average value and the light shade presents the SEM. (F–H) In vivo calcium imaging of test (IR41a-Gal4,UAS-SPRi;UAS-GCaMP5) and control (IR41a-Gal4;UAS-GCaMP5) mated female flies. OSN axon terminals of IR41a>SPRi females show significantly enhanced responses to putrescine compared to control females. (F) Representative pseudo-color images showing the response to water and 10 ppm putrescine in IR41a>SPRi and control females, respectively. (G) Average response trace of the VC5 glomerulus in IR41a>SPRi and control females at 1–6 h post-mating for 10 ppm putrescine. (E,G) The gray column represents the 0.5 s stimulation period. Dark colored line is the average response and the light shade is the SEM. (H) Quantification of peak ΔF responses (in %ΔF/F) in IR41a>SPRi and control females for 0 ppm, 6 ppm, 8 ppm, and 10 ppm putrescine, respectively (n = 7 ± SEM). All p-values were calculated using an unpaired t test (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001).

Fig 5. Myoinhibitory peptides regulate polyamine sensitivity in the context of mating.

(A) Loss of sex peptide (SP) in the sperm of the male does not significantly affect chemosensory attraction of mated females to 10 ppm of polyamines. Wild-type (wt) Canton S females mated to wild-type or sex peptide mutant (SP⁰) males do not show a significantly altered level of attraction to the odor of putrescine or cadaverine. (n = 8, 60 flies/trial). (B) SP⁰ male-mated Canton S females lay their low numbers of eggs on either site of the oviposition assay and show no preference behavior. (n = 8, 60 flies/trial). (C) Myoinhibitory peptide (MIP) expression in the AL and SEZ regions in the female brain. In the AL, the glomerulus innervated by IR41a OSNs is displayed (IR41a-Gal4;UAS-mCD8GFP). Note that MIP staining is detected in close proximity to IR41a axon terminals. In the SEZ, anti-MIP staining (green) localizes close to IR76b neuron axons and axon terminals (magenta) consistent with MIPs being secreted by IR76b neurons (IR76b-QF;QUAS-mtd-tomato) (see arrowheads). (D,E) MIPs modulate olfactory attraction to polyamines selectively in mated females but not males. RNAi-mediated knockdown of MIPs with four different RNAi transgenic lines in IR76b neurons (IR76b-Gal4;UAS-MIPi) selectively reduces the olfactory preference of mated females but not of males to 10 ppm of putrescine (D) or 10 ppm of cadaverine (E) (n = 8, 60 flies/trial). (F) The effect of MIP knockdown (IR76b-Gal4;UAS-MIPi) depends on the mating state of the female, as the low attraction of virgin females to 10 ppm polyamine odor was not further reduced in virgin females with MIP knockdown compared to virgin controls without RNAi against MIPs. Box plots show median and upper/lower quartiles (n = 8, 60 flies/trial). (G) Knockdown of MIPs in IR76b neurons abolishes oviposition preference to 1 mM putrescine and cadaverine using three different MIPi transgenic lines (IR76b-Gal4;UAS-MIPi). Females laid their eggs on either side of the assay. All box plots show median and upper/lower quartiles (n = 8, 60 flies/trial). All p-values were calculated via two-way ANOVA with the Bonferroni multiple comparison post-hoc test (ns > 0.05, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001).

Fig 6. MIP expression is increased in the AL upon mating.

(A) SPR and MIP expression analysis before and after mating of antenna and brain of virgin or mated females. Quantitative PCR (n = 3 genetic variants with 200 females per n and condition) of the antenna and brain of virgin and mated flies reveals that SPR expression upon mating increases upon mating ~10-fold in the antenna and ~3-fold in the brain. Graph displays 2^∆∆CT ± SEM (see Materials and Methods for details). (B) Quantification of MIP protein expression in the AL. Mated flies show a small but significant increase of MIP expression in the AL. n = 20 flies per group. *p = 0.0113, unpaired t test. (C) Overexpression of SPR under the control of the IR41a enhancer (IR41a-Gal4;UAS-SPR) or IR76b enhancer (IR76b-Gal4;UAS-SPR) in virgin females increases their attraction to polyamine odor in the T-maze assay (n = 8). (D) Overexpression of MIP under the control of the IR41a enhancer (IR41a-Gal4;UAS-MIP) or IR76b enhancer (IR76b-Gal4;UAS-MIP) in virgin females induces a strongly increased attraction to polyamine odor in olfactory T-maze assay (n = 8). (E,F) No egg-laying preference was observed in virgin females overexpressing SPR or MIP under the control of the IR41a enhancer (IR41a-Gal4;UAS-SPR or IR41a-Gal4;UAS-MIP) or IR76b enhancer (IR76b-Gal4;UAS-SPR or IR76b-Gal4;UAS-MIP) in oviposition assays compared to controls (n = 8). Virgin females overexpressing SPR or MIP laid very few eggs, similar to control virgins, which results in the high variability observed in the data. (G–I) In vivo calcium imaging of presynaptic terminals of OSNs in the AL expressing IR41a-Gal4,UAS-SPR;UAS-GCaMP5 or IR41a-Gal4;UAS-GCaMP5 (control). Virgin females overexpressing SPR in IR41a OSNs show significantly suppressed calcium signals to putrescine compared to virgin control females. (G) Representative pseudo-color images showing the response to 0 ppm and 10 ppm putrescine in SPR-overexpressing and control virgin females, respectively. (H) Average activity trace of the VC5 glomerulus in SPR-overexpressing and control virgin females for 10 ppm putrescine. (I) Quantification of peak ΔF responses in SPR-overexpressing (n = 7) and control (n = 8) females for 0 ppm and 10 ppm putrescine. Boxes show median and upper/lower quartiles, and whiskers show minimum/maximum values. All p-values were calculated via two-way ANOVA with the Bonferroni multiple comparison post-hoc test (ns > 0.05, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001) except for Fig 6B and 6I, where p-values were calculated via an unpaired t-test (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001).

Fig 7. SPR/MIP signaling in chemosensory neurons adjusts female preference behavior upon mating.

Model for mating-state-dependent modulation of olfactory and gustatory polyamine preference. (Left side) Upon mating, increased amounts of SPR in the sensory neuron suppress the output of IR41a/IR76b olfactory neurons, thereby increasing the female’s preference for higher concentrations of polyamine. (Right side) Mating increases the SPR amount in gustatory sensory neurons, and in contrast to the situation in OSNs increases the presynaptic output of IR76b taste neurons. This change increases the mated female’s preference for polyamine taste. In conclusion, mating increases SPR expression in chemosensory neurons and by two different cellular mechanisms enhances the mated female’s preference for beneficial polyamines.