Excitation and inhibition onto central courtship neurons biases Drosophila mate choice
- University of California, Berkeley, United States
Figures
Figure 1 with 1 supplement
F and M cells comprise distinct chemosensory neuron classes.
(A) F cells (PPK23+ PPK25+) respond to female pheromones whereas M cells (PPK23+ PPK25-) respond to male pheromones; n = 7 bristles. The female pheromone mix contained 7,11-heptacosadiene and 7,11-nonacosadiene. The male pheromone mix contained 7-tricosene and cis-vaccenyl acetate. (B) GCaMP6s marks two PPK23 cells per bristle (left). CD8::tdTomato marks the PPK25 cell (middle). Maximum ΔF of both PPK23 cells to female (red) or male pheromones (blue) (right). (C) vGlut (magenta) is expressed in one PPK23 cell (green) under each bristle. Transgenic flies with ppk23-LexA, lexAop-CD2::GFP, vGlut-Gal4, UAS-CD8::tdTomato were used for cell labeling. (D) vGlut (green) and PPK25 (magenta) are expressed in the same cell under each bristle. Flies contained vGlut-LexA::VP16, lexAop-CD2::GFP, ppk25-Gal4, UAS-CD8::tdTomato. (E) Axons from F cells in the legs (green) do not project to the central brain (left) but instead terminate in the six leg neuromeres of the ventral nerve cord (VNC, right). Brains are counterstained with nc82 (magenta) to show neuropil. (F) Expression of dTRPA1 in F cells promoted male-female courtship upon heat-evoked neural activation; n = 11–21/condition. Number of unilateral wing extensions per 10-minute trial was recorded. (G) M cells from the legs project to the SEZ in the brain (arrow shows fibers entering from the cervical connective) and VNC. Other SEZ axons come from the proboscis, with fibers entering from the labellar nerve. (H) dTRPA1-mediated activation of M cells suppressed male-female courtship. n = 10–12/condition. Number of unilateral wing extensions per 10-minute trial was recorded. Scale bars, 5 (B), 10, (C, D) 25 (G, SEZ) or 50 μm (E, G, VNC). Data are Mean ± SEM. Kruskal-Wallis test, Dunn’s post-hoc (A) or 2-way ANOVA, Bonferroni post-hoc (F, H). **p< 0.01. See also Figure 1—figure supplement 1, on selectively targeting F or M cells.
Figure 1—figure supplement 1
Approach to selectively target F or M cells.
(A) vGlut-QF (green) is expressed in PPK25 cells (magenta), similar to vGlut-LexA in Figure 1D. Flies were vGlut-QF2, QUAS-Gal80, ppk25-Gal4, UAS-CD8::tdTomato. (B) (left), Ablating F cells (ppk25-Gal4) using diphtheria toxin leaves M cells intact, with one GFP-positive cell/bristle rather than two. This strategy was used to label M cells in Figure 1G. (middle), Suppressing Gal4-dependent expression in F cells (ppk23-Gal4, UAS-CD8::tdTomato, vGlut-LexA, lexAop-Gal80) leaves expression intact in M cells. This strategy was used to activate M cells in Figure 1H. (right), Suppressing LexA-dependent expression in F cells (ppk23-LexA, lexAop-CD2::GFP, vGlut-QF2, QUAS-Gal80) leaves expression intact in M cells. This strategy was used to activate M cells in Figures 2B, 4B, and 5E. Different approaches were used to accommodate chromosome locations of transgenes. Scale bars, 10 μm.
Figure 2
F cells activate courtship-promoting P1 neurons.
(A) Male-specific P1 neurons (green) are located in the protocerebrum. Scale bar, 50 μm. Flies contained P1-Gal4DBD, P1-Gal4AD, UAS-CD8::GFP. (B) ATP-mediated stimulation of F+M cells (ppk23-LexA, lexAop-P2X2), F cells (vGlut-LexA) but not M cells (ppk23-LexA, lexAop-P2X2, vGlut-QF2, QUAS-Gal80) triggered calcium increases in P1 neurons; n = 5–8/condition. Mock is no P2X2. Traces on the left show averaged △F/F with mean in black and SEM shaded. Arrows indicate stimulus. Schematics show cells monitored with GCaMP6s (green) and connections tested. Data are also displayed as bar graph (right). Differences in expression levels of ppk23-LexA and vGlut-LexA may contribute to different response magnitudes of F+M versus F cell stimulation. Mean ± SEM of maximum △F/F. Kruskal-Wallis test, Dunn’s post-hoc to mock, *p<0.05, **p<0.01, ***p<0.001.
Figure 3
PPN1 neurons are courtship-promoting neurons in proximity to PP23 axons and P1 fibers.
(A) PPN1 neurons have cell bodies in the third leg neuromere of the VNC and send projections to the six leg neuromeres and wing neuromere of the VNC and to the ventrolateral protocerebrum of the brain. R56C09-Gal4 drives expression of UAS-CD8::GFP exclusively in the pair of PPN1 neurons. (B-C) PPN1 has dendrites in the VNC (B, DenMark, magenta) and axons in the ventrolateral protocerebrum (C, syt-GFP, green). B and C are from the same animal containing R56C09-Gal4, UAS-DenMark, UAS-synaptotagmin-GFP. (D) Activation of PPN1 with dTRPA1 causes increased male-female courtship at 30°C; mean ± SEM, n = 16–30/condition, **p<0.01 (2-way ANOVA, Bonferroni post-hoc). (E) Overlap is observed in the VNC between PPN1 dendrites (green) and incoming PPK25 axons (magenta). R56C09-LexA, lexAop-CD2::GFP, ppk25-Gal4, UAS-CD8::tdTomato flies were used. F. Overlap between PPN1 (green) and P1 (magenta) in the anterior ventrolateral protocerebrum (50 μm collapsed Z-stack). Scale bars, 25 μm (F) 50 μm (A-E).
Figure 4
F cells activate courtship-promoting PPN1 neurons which activate P1.
(A) Calcium imaging of PPN1 axons while activating different sensory classes revealed that PPN1 is activated by F+M and F cell stimulation but not M cell stimulation; n = 5–7/condition. Mock is no P2X2. Arrows indicate stimulus. (B) Chrimson-mediated activation of PPN1 triggers calcium increases in P1 by GCaMP6s calcium imaging; n = 5–6/condition. Arrows indicate stimulus. Schematics show cells monitored with GCaMP6s (green) and connections tested. Data are Mean ± SEM. Kruskal-Wallis test, Dunn’s post-hoc to mock (A) or Mann-Whitney test (B). *p<0.05, **p<0.01.
Figure 5 with 1 supplement
M cells and F cells activate courtship-suppressing mAL neurons.
(A) Example image of a GCaMP6s ΔF heat map in fru-LexA neurons upon P2X2-mediated activation of F+M cells. (B) mAL neurons labeled by the intersection of R43D01-Gal4 and fru-LexA connect the SEZ and protocerebrum. (C) Activating mAL neurons with dTRPA1 suppresses courtship toward females; n = 10/condition. (D) Silencing mAL neurons with tetanus toxin or knocking down vGAT with R43D01-Gal4 induces male-male chaining; n = 8–10 groups/condition, 6–9 males per group. For C and D, >> means >stop>. Chaining index represents the fraction of time 3 or more males were courting over the 10-min trial. (E) P2X2-mediated stimulation of either F or M cells activates mAL neurons by GCaMP calcium imaging. Arrows indicate stimulus. (F) Maximum ΔF/F in mAL cell bodies; n = 5–9/condition. Mock is no P2X2. Scale bars, 50 μm (A, B). Data are Mean ± SEM, 2-way ANOVA, Bonferroni post-hoc (C), Mann-Whitney test (D), or Kruskal-Wallis test, Dunn’s post-hoc to mock (F). *p<0.05, ***p<0.001. See also Figure 5—figure supplement 1, for anatomical characterization of mAL neurons.
Figure 5—figure supplement 1
Characterization of mAL neurons.
(A) mAL neurons labeled by R43D01-Gal4 (green, CD8::GFP) are immunopositive for GABA (magenta). (B) mAL neurons labeled by R43D01-Gal4 (green, CD8::GFP) express FruM (magenta). (C) mAL neurons have presynaptic termini (green, nsyb-GFP) in the higher brain and SEZ. (D) mAL neurons have dendrites (green, Dscam17.1-GFP) primarily in the SEZ. (E) M cell axons (magenta, myr::mCherry) interdigitate with mAL fibers (green, CD8::GFP) in the SEZ. (F) A representative image of the mAL axonal tract in the higher brain before and after 2-photon-mediated lesioning. (G) R43D01-LexA labels mAL neurons and some olfactory sensory neurons (green, CD2::GFP). This line was used in Figure 6E. Scale bars, 5 μm (A), 10 μm (B, F), 25 μm (E), 50 μm (C, D, G).
Figure 6
mAL neurons functionally and behaviorally inhibit P1 neurons.
(A) Overlap between mAL (green) and P1 (magenta) in the superior lateral protocerebrum (collapsed 97-μm stack). Scale bar is 25 μm. (B) Knockdown of GABAA receptor Rdl in P1 neurons induces male-male chaining; n = 10/condition. (C) Chrimson-mediated activation of P1 neurons causes wing extension (wing ext.) in solitary males, which is suppressed by co-activation of neurons expressing R43D01-Gal4. (D). Lesioning mAL axons increases GCaMP response in P1 upon P2X2-mediated stimulation of F+M cells. (E) Chrimson-mediated activation of mAL neurons causes P1 hyperpolarization, as detected by increased ArcLight fluorescence. Arrow indicates laser. F. Maximum ΔF/Fin P1; n = 10–11/condition. Data are Mean ± SEM. Mann-Whitney test (B, F), Fisher’s exact test (C), or paired t-test (D). *p<0.05, **p<0.01, ***p<0.001.
Figure 7
Schematic of courtship-promoting and courtship-inhibiting circuits activated by F and M cells.
F cells on the leg express PPK23, PPK25, and VGlut, and respond to female pheromones. M cells on the leg express PPK23 and respond to male pheromones. The M cell neurotransmitter is unknown. F cells activate PPN1, a class of projection neuron with cell bodies and dendrites in the VNC and long-range axonal projections to the ventrolateral protocerebrum. PPN1 axons are in close proximity to P1 fibers, and PPN1 activation causes activation of P1. M cells on the leg activate GABAergic mAL neurons, which connect the SEZ and superior lateral protocerebrum. mAL axons interdigiate with P1 fibers, and mAL acitvation causes hyperpolarization of P1, likely through the GABA-A receptors containing the Rdl subunit. F cells also provide an inhibitory drive onto P1 via mAL. The contact between F cells and mAL is not direct (dotted line). Other connections may not be monosynaptic.
Tables
Table 1
Genotypes of flies used for experiments in this study.
| Figure panel | Genotype |
| Figure 1A | w-/y; ppk23-LexA/ppk25-Gal4; lexAop-GCaMP6S/UAS-CD8::tdTomato |
| Figure 1B | same as Figure 1A |
| Figure 1C | w-/y; ppk23-LexA/vGlut-Gal4; lexAop-CD2::GFP/UAS-CD8::tdTomato |
| Figure 1D | w-/y; vGlutMI04979-LexA::QFAD/ppk25-Gal4; lexAop-CD2::GFP/UAS-CD8::tdTomato |
| Figure 1E | w-/y; UAS-CD8::GFP/+; ppk25-Gal4/+ |
| Figure 1F | see Figure |
| Figure 1G | w-/y; ppk23-LexA/ppk25-Gal4; lexAop-CD2::GFP/UAS-DTI |
| Figure 1H | see Figure |
| Figure 1—figure supplement 1A | w-/y; vGlutMI04979-QF2/ppk25-Gal4; QUAS-mCD8::GFP/UAS-CD8::tdTomato |
| Figure 1—figure supplement 1B | (left) w-/y; ppk23-LexA/ppk25-Gal4; lexAop-CD2::GFP/UAS-DTI (middle) w-/y; vGlutMI04979-LexA::QFAD/ppk23-Gal4; QUAS-Gal80/UAS-CD8::tdTomato (right) w-/y; vGlutMI04979-QF2/ppk23-LexA; QUAS-Gal80/lexAop-CD2::GFP |
| Figure 2A | w-/y; P1-Gal4-AD/UAS-CD8::GFP;P1-Gal4-DBD/+ |
| Figure 2B | F cell stim: w-/y; vGlutMI04979-LexA::QFAD/lexAop-P2X2; R71G01-Gal4/UAS-GCaMP6S M cell stim: w-/y; ppk23-LexA, lexAop-P2X2/ vGlutMI04979-QF2; R71G01-Gal4, UAS-GCaMP6S/QUAS-Gal80 |
| Figure 3A | w-/y; UAS-CD8::GFP/+; R56C09-Gal4/+ |
| Figure 3B–C | w-/y; +/+; R56C09-Gal4/UAS-DenMark, UAS-synaptotagmin-GFP |
| Figure 3D | see Figure |
| Figure 3E | w-/y; R56C09-LexA/ppk25-Gal4; lexAop-CD2::GFP/UAS-CD8::tdTomato |
| Figure 3F | w-, UAS-CD8::tdTomato/y; P1-Gal4-AD/R56C09-LexA; P1-Gal4-DBD/lexAop-CD2::GFP |
| Figure 4A | w-/y; UAS-CD8::GFP/+; R56C09-Gal4/+ |
| Figure 4B | F cell stim: UAS-CD8::tdTomato/y; vGlutMI04979-LexA::QFAD/lexAop-P2X2; R56C09-Gal4/UAS-GCaMP6S M cell stim: UAS-CD8::tdTomato/y; ppk23-LexA, lexAop-P2X2/ vGlutMI04979-QF2; R56C09-Gal4, UAS-GCaMP6S/QUAS-Gal80 F+M cell stim: UAS-CD8::tdTomato/y; ppk23-LexA/lexAop-P2X2; R56C09-Gal4/UAS-GCaMP6S |
| Figure 4C | w-, UAS-CD8::tdTomato/y; P1-Gal4-AD/R56C09-LexA; P1-Gal4-DBD/lexAop-CD2::GFP |
| Figure 4D | w-/y; UAS-GCaMP6S/R56C09-LexA; R71G01-Gal4/lexAop-Chrimson |
| Figure 5A | w-/y; ppk23-LexA/lexAop-P2X2; fru-Gal4/UAS-GCaMP6S |
| Figure 5B | w-/y; UAS>stop>CD8::GFP/lexAop-FLPL; R43D01-Gal4/fru-LexA |
| Figure 5C | see Figure |
| Figure 5D | see Figure |
| Figure 5E | F cell stim: w-/y; vGlutMI04979-LexA::QFAD/lexAop-P2X2; R43D01-Gal4/UAS-GCaMP6S M cell stim: w-/y; ppk23-LexA, lexAop-P2X2/ vGlutMI04979-QF2; R43D01-Gal4, UAS-GCaMP6S/QUAS-Gal80 |
| Figure 5F | same as Figure 5e and F+M cell stim: ppk23-LexA/lexAop-P2X2; R43D01-Gal4/UAS-GCaMP6S |
| Figure 5—figure supplement 1A | w-/y; UAS-CD8::GFP/+; R43D01-Gal4/+ |
| Figure 5—figure supplement 1B | w-/y; UAS-CD8::GFP/+; R43D01-Gal4/+ |
| Figure 5—figure supplement 1C | w-/y; UAS>stop>nsyb-GFP 19a/lexAop-FLPL; R43D01-Gal4/fru-LexA |
| Figure 5—figure supplement 1D | w-/y; UAS>stop>Dscam17.1-GFP 19a/lexAop-FLPL; R43D01-Gal4/fru-LexA |
| Figure 5—figure supplement 1E | w-/y; UAS-CD8::GFP/ppk23-LexA; R43D01-Gal4/lexAop-myr::mCherry |
| Figure 5—figure supplement 1F | w-/y; ppk23-LexA/lexAop-P2X2; R43D01-Gal4/R71G01-Gal4, UAS-GCaMP6S |
| Figure 5—figure supplement 1G | w-/y; R43D01-LexA/+; lexAop-CD2::GFP/+ |
| Figure 6A | w-, UAS-CD8::tdTomato/y; P1-Gal4DBD/R43D01-LexA; P1-Gal4AD/lexAop-CD2::GFP |
| Figure 6B | see figure, genotype includes UAS-Dicer (X) |
| Figure 6C | see Figure |
| Figure 6D | w-/y; ppk23-LexA/lexAop-P2X2; R71G01-Gal4, UAS-GCaMP6S/R43D01-Gal4 |
| Figure 6E | w-/y; R43D01-LexA/UASArcLight; R71G01-Gal4/lexAop-Chrimson |
| Figure 6F | same as Figure 6E |
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)
Excitation and inhibition onto central courtship neurons biases Drosophila mate choice
eLife 4:e11188.
https://doi.org/10.7554/eLife.11188
