Sexually dimorphic neurons in the ventromedial hypothalamus govern mating in both sexes and aggression in males

Abstract

Sexual dimorphisms in the brain underlie behavioral sex differences, but the function of individual sexually dimorphic neuronal populations is poorly understood. Neuronal sexual dimorphisms typically represent quantitative differences in cell number, gene expression, or other features, and it is unknown whether these dimorphisms control sex-typical behavior exclusively in one sex or in both sexes. The progesterone receptor (PR) controls female sexual behavior, and we find many sex differences in number, distribution, or projections of PR-expressing neurons in the adult mouse brain. Using a genetic strategy we developed, we have ablated one such dimorphic PR-expressing neuronal population located in the ventromedial hypothalamus (VMH). Ablation of these neurons in females greatly diminishes sexual receptivity. Strikingly, the corresponding ablation in males reduces mating and aggression. Our findings reveal the functions of a molecularly defined, sexually dimorphic neuronal population in the brain. Moreover, we show that sexually dimorphic neurons can control distinct sex-typical behaviors in both sexes.

Figure 1. Visualizing PR+ neurons in the mouse brain

(A) Generating the PR^PL allele. ACN is a self-excising neomycin selection cassette (Bunting et al., 1999). Orange rectangles are exons and the red line in the 3’ exon denotes the stop codon. (B) PCR to detect homologous recombination at the PR locus. Primers used to detect integration of the 5’ (F1, R1) and 3’ (F2, R2) arms of the targeting vector. ACN precludes detection of the 3’ recombination event in ES cells. (C, D) No difference between WT and PR^PL/PL females in litter size and frequency. (E–G) No difference in titers of sex hormones between WT and PR^PL/PL adults. (H–M) Boxed areas in Nissl-stained coronal sections (Paxinos and Franklin, 2003) through the adult brain depict locations of the regions shown in panels to the right. PR expression in PR^PL/+female as labeled by β-gal activity mirrors expression of PR mRNA in adjacent sections. Scale bars = 50 µm. Mean ± SEM; n ≥ 12/genotype (C–G); n = 3 (H–M). See also Figure S1 and Table S3.

Figure 2. Sexual dimorphism in PR-expression in the adult brain

Boxed areas in Nissl-stained coronal sections through the adult brain depict regions of PR^PL/PLmice labeled for β-gal activity in the panels to the right. (A–L) More PR+ cells in the female AVPV/POA, VMHvl, and arcuate nucleus. (M–X) More PR+ cells in the male basal forebrain, BNSTmpm, and MeApd. (Y) Representation of sexually dimorphic PR expression in different brain regions as projected on to a mid-sagittal section. c, caudal, d, dorsal, r, rostral, v, ventral. Scale bars = 50 µm (C, K) and 100 µm (G, O, S, W). Inset scale bars = 25 µm. Mean ± SEM; n ≥ 4/sex; *p < 0.04, **p < 0.01. See also Figure S2 and Table S1.

Figure 3. PR+ VMHvl neurons project in a sexually dimorphic manner

(A) Strategy to visualize projections of PR+ neurons. (B–E) Lenti-lxlplap targeted to the VMH infects cells in PR^Cre/+ and WT mice as visualized by EGFP+ cells. Only a few cells are PR+ in this region so there is no apparent difference in the number of EGFP+ cells in PR^Cre and WT mice. PLAP+ soma and local arbors of VMHvl neurons are only observed in PR^Cre mice. (F–U) Boxed areas in Nissl-stained coronal sections depict regions shown in panels to the right. Lenti-lxlplap targeted to the VMHvl of adult PR^Cre/+ mice labels PLAP+ soma and local arbors of VMHvl neurons (F–I). The lentiviral titer limits the number of infected Cre+ neurons and does not highlight the sex difference in the number of these neurons. The variable multiplicity of infection can lead to apparent size differences in PLAP-labeled soma. However, there is no sex difference in the soma size of these neurons (Figure S2G). PR+ VMHvl neurons project to the AVPV, POA, and PAG (J–U). There are more PLAP+ projections to the AVPV in females (J–M). (V, W) Schematic summarizing projections of PR+ VMHvl neurons. No difference in anatomical extent of projections in different regions, but female AVPV receives more innervation from these neurons. Scale bars = 100 µm (C), 50 µm (H,P,T), 25 µm (L). Mean ± SEM; n ≥ 7/sex; *p < 0.001. See also Figure S3, Tables S2, S3.

Figure 4. Genetic strategy to ablate neurons in a Cre-dependent manner

(A) Intramolecular cleavage of endogenous pro-caspase-3 by upstream caspases activates caspase-3, which then induces apoptosis. This intramolecular cleavage site has been replaced by a TEV linker domain (black bar) in inactive taCasp3 (pro-taCasp3) such that only TEV protease activates taCasp3, which then induces apoptosis. (B) Viral strategy to ablate PR+ neurons conditionally. (C, D) Cell death 1 week following infection of Cre:EGFP+ HEK293T cells with AAV-flex-taCasp3-TEVp. n = 3 experiments. (E, F) Ablation of PR+ VMHvl neurons in a PR^PL/Cre but not PR^PL/+ female injected with AAV-flex-taCasp3-TEVp. n ≥ 10/experimental group. Scale bar = 100 µm (C, D) and 25 µm (E, F). See also Figure S4.

Figure 5. PR+ VMHvl neurons regulate female sexual receptivity

(A) Experimental design to test the role of PR+ VMHvl neurons in female behaviors. Mating was tested with ovariectomized females primed to be in estrus. Other behaviors were tested with gonadally-intact females. (B–J) PR^Cre and control females were injected with AAV-flex-taCasp3-TEVp and tested for sexual behavior with WT males. (B) PR^Cre females spend more time rejecting male mating attempts, walking away when the male approaches. (C–E) PR^Cre females display lower receptivity index (mounts leading to intromission/total mounts) and reduced number and duration of lordosis events. (F) PR^Cre females spend more time moving about and being unreceptive during intromission. (G) Fewer than 20% of PR+ neurons remain in the VMHvl of PR^Cre females, who reject male mating attempts more than control females. (H) Males sniff and initiate mating equivalently with PR^Cre and WT females but ejaculate in fewer assays with PR^Cre females. (I) Males mount PR^Cre females more but without a corresponding increase in intromission. (J) Males mount PR^Cre females longer, but intromit for shorter duration. (K, L) Ablation of PR+ VMHvl neurons in PR^Cre females results in loss of Cckar expression. Mean ± SEM; n ≥ 10/experimental group (B–J); n = 3 (K, L); *p < 0.02, ** p < 0.005. Scale bar = 50 µm. See also Figure S5, Table S3, Movies S1, S2.

Figure 6. PR+ VMHvl neurons regulate male sexual behavior

(A) Experimental design to test the role of PR+ VMHvl neurons in male behaviors. (B–G) PR^Cre and control males were injected with AAV-flex-taCasp3-TEVp and tested for mating, ultrasonic vocalizations toward male or female intruders, and territory marking. (B) PR^Cre males intromit females in fewer assays. (C, D) PR^Cre males mount and intromit females less and have shorter bouts of intromissions. (E) Both PR^Cre and control males emit more vocalizations to females. (F, G) No difference between PR^Cre and control males in the number and distribution of urine marks. % marks in center = 100*(# urine marks not abutting cage perimeter/# of all urine marks). Mean ± SEM; n ≥ 24/experimental group (B–D, F,G), n ≥ 5/experimental group (E); *p < 0.008, **p < 0.001. See also Figures S6 and S7.

Figure 7. PR+ VMHvl neurons regulate male aggression

(A–F) PR^Cre and control resident males were injected with AAV-flex-taCasp3-TEVp targeted to the VMHvl and tested for aggression toward a WT male intruder. (A) All residents sniff intruders equivalently, but PR^Cre males attack less. (B–D) When PR^Cre males fight, they attack less, for a shorter duration, and with longer intervals between attacks. (E) PR^Cre males bite less. (F) Fewer than 20% of PR+ neurons remain in the VMHvl of PR^Cre males, who attack intruders less. Mean ± SEM; n ≥ 24/experimental group; *p < 0.04, **p ≤ 0.009. (G, H) Ablation of PR+ VMHvl neurons in a PR^PL/Cre male injected with AAV-flex-taCasp3-TEVp. Scale bar = 25 µm. See also Figure S7.