A Bidirectional Circuit Switch Reroutes Pheromone Signals in Male and Female Brains

摘要

class="head no_bottom_margin" id="sec1title">IntroductionMany species exhibit sexually dimorphic behaviors, typically as part of their reproductive repertoire. These behaviors, which often have a substantial unlearned component, provide highly tractable paradigms to explore the genetic and neural circuit basis of behavior (). As potent releasers of specific dimorphic behavior, sex pheromones are particularly experimentally advantageous (). Nevertheless, even here the neural mechanisms underlying differential processing within the brain remain largely unknown ().Several models have been proposed for how pheromones can elicit different behavior in males and females. One model is exemplified by classic work on the attraction of male silkmoths to bombykol (). Here, one sex expresses a pheromone receptor, while the other is blind to this cue. However, this peripheral change cannot account for situations in which a common stimulus produces behavior in both sexes. These are likely due to circuit differences within the brain. For instance, in mice, only males show courtship behavior toward females, but after ablation of the vomeronasal organ females show female-directed courtship (). This leads to a second model in which both sexes can express male behaviors, but these are normally repressed in females by sex-specific circuits downstream of pheromone detection. However, these circuit differences remain unknown, because the relevant receptors and downstream pathways have yet to be identified.A simpler paradigm is offered by analogous results in flies and mice, in which a monomolecular pheromone can activate identified sensory neurons in both sexes (). In the mouse, the male pheromone ESP1 activates V2Rp5 sensory neurons in both sexes but produces distinct patterns of immediate-early gene expression in deeper brain nuclei (). ESP1 triggers lordosis in females, but no effect on male behavior has been reported.In Drosophila, the male pheromone 11-cis-vaccenyl acetate (cVA) stimulates courtship in females but decreases courtship and increases aggression in males (). Because both first- and second-order olfactory neurons show similar cVA responses in males and females (), it is likely that some circuit difference deeper in the brain results in sex-specific behavioral output. Two further studies have characterized downstream elements of this pathway. used an elegant tracing approach based on sequential photoactivation of green fluorescent protein to identify candidate third- and fourth-order neurons, some of which were shown to be cVA responsive in males. However, they were unable to characterize these neurons anatomically or functionally in females, so the presence or nature of any circuit dimorphism remained unclear. In a parallel study, used a genetic mosaic technique to carry out an exhaustive analysis of sexually dimorphic neurons in male and female brains. In the olfactory system they found two groups of third-order neurons, present in both sexes, that appeared to be differentially connected, suggesting a precise circuit hypothesis for differential pheromone processing in male and female brains (A).class="figpopup" href="/pmc/articles/PMC3898676/figure/fig1/" target="figure" rid-figpopup="fig1" rid-ob="ob-fig1">Figure 1Sex-Specific Pheromone Responses in fru+ LHNs(A and B) Abstract circuit model for sexually dimorphic behavior (A), and circuit model for cVA processing in females and males (B).(C) Targeted in vivo whole-cell recording setup, with odor delivery and photoionization detector (PID). A dye-filled neuron is shown.(D–F) Z projections of female and male neuroblast clones on a reference brain; the ventral lateral horn is marked with a white circle. Insets show spatial relationship between LHN dendrites and DA1 PN axon terminals (ochre). Cell numbers for cluster aSP-f: 23.2 ± 2.6 in males versus 18.6 ± 5.0 in females; aSP-g: 13.4 ± 0.89 versus 13.4 ± 4.97; aSP-h: 5.0 ± 0.8 versus 5.0 ± 0.5.(G–I) Single aSP-f, aSP-g, and aSP-h LHNs filled during patch-clamp recording and traced (magenta or green lines) compared with volume-rendered DA1 PNs (pale magenta or pale green).(J–L) Physiological data for aSP-f, aSP-g, and aSP-h LHNs. These three panels are arranged in a 3-row × 2-column grid. The top row shows averaged current clamp recordings for each LHN shown in (G), (H), and (I) (cell 1). Row 2 shows raster plots for the same neurons. Row 3 shows raster plots for an additional neuron (cell 2).(M–O) Summary of cVA responses. Each dot is one neuron, colored red for significant cVA responses (adjusted p < 0.01, see the href="#sec4" rid="sec4" class=" sec">Experimental Procedures); nonsignificant responses are black. Response counts: aSP-f: 1/34 female and 20/37 male neurons; aSP-g: 11/15 female and 1/17 male neurons; aSP-h: 2/8 female and 12/14 male.See href="#mmc1" rid="mmc1" class=" supplementary-material">Table S1C for statistics. Scale bars, 25 μm. Pale red bars in (J)–(L) mark 500 ms odor presentation. See the href="#sec4" rid="sec4" class=" sec">Experimental Procedures for odorant abbreviations and href="/pmc/articles/PMC3898676/figure/figs1/" target="figure" class="fig-table-link figpopup" rid-figpopup="figs1" rid-ob="ob-figs1" co-legend-rid="lgnd_figs1">Figure S1 for additional data.