A Neural Basis for Control of Cichlid Female Reproductive Behavior by Prostaglandin F2α

Abstract

In most species, females time reproduction to coincide with fertility. Thus, identifying factors that signal fertility to the brain can provide access to neural circuits that control sexual behaviors. In vertebrates, levels of key signaling molecules rise at the time of fertility to prime the brain for reproductive behavior [1-11], but how and where they regulate neural circuits is not known [12, 13]. Specifically, 17α,20β-dihydroxyprogesterone (DHP) and prostaglandin F2α (PGF2α) levels rise in teleost fish around the time of ovulation [10, 14, 15]. In an African cichlid fish, Astatotilapia burtoni, fertile females select a mate and perform a stereotyped spawning routine, offering quantifiable behavioral outputs of neural circuits. We show that, within minutes, PGF2α injection activates a naturalistic pattern of sexual behavior in female A. burtoni. We also identify cells in the brain that transduce the prostaglandin signal to mate and show that the gonadal steroid DHP modulates mRNA levels of the putative receptor for PGF2α (Ptgfr). We use CRISPR/Cas9 to generate the first targeted gene mutation in A. burtoni and show that Ptgfr is necessary for the initiation of sexual behavior, uncoupling sexual behavior from reproductive status. Our findings are consistent with a model in which PGF2α communicates fertility status via Ptgfr to circuits in the brain that drive female sexual behavior. Our targeted genome modification in a cichlid fish shows that dissection of gene function can reveal basic control mechanisms for behaviors in this large family of species with diverse and fascinating social systems [16, 17].

Figure 1. PGF_2α activates female reproductive behavior in A. burtoni

a, Natural progression of spawning behavior. After ovulation, females lay eggs during circling. B–I, Analysis of all reproductive assays. B, C, G, Males quiver, lead, and attack similarly with vehicle and PGF_2α-injected females. D–F, I, Vehicle- and PGF_2α-injected females show similar following and pot entry (D–E), but females show circling behavior more rapidly and frequently after PGF_2α injection (F, I). H, Body weight-normalized ovary mass does not differ between groups. *, P = 0.0018, Mantel-Cox test; **, P = 0.0010, Mann-Whitney U-test. Mean ± SEM. n = 28 females per treatment. J, Ethograms reveal that naturally spawning and a reproductive subset of PGF_2α-injected females exhibit quantitatively similar transitions between reproductive behaviors. Vehicle-treated females were matched for normalized ovary mass, but rarely circle. Diameters of circles are proportional to count of each behavior; weights of arrows are proportional to fraction followed by second behavior. Transitional probabilities do not differ between naturally spawning and PGF_2α-injected females; all transitions P > 0.05, Mann-Whitney U-test. n = 5 assays per group. See also Figure S1; Movies S1, S2.

Figure 2. Ptgfr is expressed in regions active during spawning, and rises at the time of spawning

A, B, Ptgfr mRNA is expressed in the POA and in scattered cells of the VL. C–H, POA and VL express cFos mRNA after being allowed to spawn naturally when exposed to a male (C, D). Females exposed to a courting male that did not spawn did not show cFos expression (E, F). Scale bars, 100 μm; n = 6–11 per group. See also Figures S2, S3; Table S1.

Figure 3. Progestin signaling upregulates Ptgfr mRNA expression in POA

A–D, Ptgfr mRNA levels rise in the POA around the time of spawning. Females with small (A) or large (B) ovary mass but that did not spawn had less Ptgfr mRNA expression than females that spawned 30 minutes prior (C). *, P = 0.0009 by Kruskal-Wallis test and p < 0.05 by Dunn’s post-hoc test; n = 6–11 females per group. E, Progesterone receptor is expressed in the Ptgfr+ compartment of the POA. F–H, Treatment of ovariectomized (OVX) females with 17α,20β-dihydroxyprogesterone (DHP) results in a rise in Ptgfr mRNA levels. *, P = 0.0286 by Mann-Whitney test; n = 4 females/group. Mean ± SEM; scale bar, 100 μm.

Figure 4. CRISPR/Cas9-mediated mutation of Ptgfr results in failure to spawn

A, Schematic for generation of biallelic Ptgfr mutants. sgRNA, single guide RNA. B, Three Ptgfr alleles encoding a large deletion or frameshift mutations were analyzed in F₁ females. The protospacer-adjacent motif is underlined, and Cas9 cut site indicated by arrowhead. C–D, Ptgfr^Δ/Δ females did not initiate circling behavior in response to PGF_2α injection *, P = 0.031, Mantel-Cox test; n = 8–13 females/genotype. e, Ovary size was not different from WT males. F–H, males did not show different levels of courtship (F, G) or aggression (H) toward Ptgfr^Δ/Δ females. Mean ± SEM. I, Group-housed Ptgfr^Δ/Δ females were not found to carry offspring despite comprising 42% of the population. **, P < 0.0001, Fisher’s exact test, n = 93 group housed fish, n = 31 mouthbrooders. See also Figure S4, Table S2.