Paternal Retrieval Behavior Regulated by Brain Estrogen Synthetase (Aromatase) in Mouse Sires that Engage in Communicative Interactions with Pairmates

Abstract

Parental behaviors involve complex social recognition and memory processes and interactive behavior with children that can greatly facilitate healthy human family life. Fathers play a substantial role in child care in a small but significant number of mammals, including humans. However, the brain mechanism that controls male parental behavior is much less understood than that controlling female parental behavior. Fathers of non-monogamous laboratory ICR mice are an interesting model for examining the factors that influence paternal responsiveness because sires can exhibit maternal-like parental care (retrieval of pups) when separated from their pups along with their pairmates because of olfactory and auditory signals from the dams. Here we tested whether paternal behavior is related to femininity by the aromatization of testosterone. For this purpose, we measured the immunoreactivity of aromatase [cytochrome P450 family 19 (CYP19)], which synthesizes estrogen from androgen, in nine brain regions of the sire. We observed higher levels of aromatase expression in these areas of the sire brain when they engaged in communicative interactions with dams in separate cages. Interestingly, the number of nuclei with aromatase immunoreactivity in sires left together with maternal mates in the home cage after pup-removing was significantly larger than that in sires housed with a whole family. The capacity of sires to retrieve pups was increased following a period of 5 days spent with the pups as a whole family after parturition, whereas the acquisition of this ability was suppressed in sires treated daily with an aromatase inhibitor. The results demonstrate that the dam significantly stimulates aromatase in the male brain and that the presence of the pups has an inhibitory effect on this increase. These results also suggest that brain aromatization regulates the initiation, development, and maintenance of paternal behavior in the ICR male mice.

Keywords: brain aromatase; communicative interaction; immunoreactivity; mouse; parental behavior; paternal care.

Figure 1

Schematic showing the different housing conditions and retrieval experiments in the paternal behavior test. One adult virgin male (S0, blue) and female (pink) were co-housed and maintained until parturition. They were housed together with their biological pups (S1 and S5) in their home cage (black circle). The sire was separated from his pups and pairmate in the home cage (S6-1) or a new cage (red circle; S6-2) for 10 min. The sire was also separated from his pups but placed along with his mate dam in a new cage (S2 and S6-3). Retrieval by the sires during 10 min was examined for five selected pups out of the litters, which were placed in a remote area away from the nest (yellow; S3). The pups in the nest represented retrieval (S4-2) and those outside the nest represented no retrieval of pups to the nest (S4-1) by the sires, retrievers, and non-retrievers. At step S6, the mice were maintained for 30 min before being sacrificed.

Figure 2

Time course of the aromatase immunofluorescence intensity after social stimulation. (A) Immunofluorescence was measured in the mPOA regions of sires at 0, 10, and 30 min after isolation together with their pairmate dams in new cages. Three mice were used for each time point. The intensity was measured in 182–421 cells. One-way ANOVA followed by Bonferroni's post-hoc test: F_{(2, 849)} = 219.82, ^***P < 0.001. (B) Representative images at 0, 10, and 30 min, respectively.

Figure 3

Photomicrographs of coronal sections showing aromatase immunoreactivity in nine areas of the brains of sires under various housing conditions. Larger images with lower magnification (left pane) and three smaller images with higher magnifications (right panels) are shown for each brain area on the left. The four housing conditions are indicated at the top. Green (GFP) and blue (DAPI) represent aromatase immunoreactivity and nuclei, respectively. The dashed line indicates the boundary of the third ventricle (3v), fasciculus retroflexus (fr), medial lemniscus (ml), and anterior commissure, anterior part (aca). Scale bars, 100 and 10 μm, respectively.

Figure 4

Statistical analysis of the semiquantitative determination of aromatase expression in different brain regions of sires. The total intensity of the aromatase signal was measured in each aromatase-expressing cell. The intensity was significantly higher in sires from the S6-3 group, compared with those in the S6-2, S6-1, and S5 groups. The intensity was significantly higher in sires from the S6-2 group, compared with sires from the S5 group in the mPOA, VTA, NAcc, VP, VMH, CA3, and LSD, but lower in the AMY and PFC regions. One-way ANOVA results: P < 0.001 in all areas among four conditions (n = 642–1430, see text); ^aP < 0.05 vs. S5, ^bP < 0.001 vs. S5, ^cP < 0.05 vs. S6-1, ^dP < 0.01 vs. S6-1,^eP < 0.001 vs. S6-1,^fP < 0.001 vs. S6-2.

Figure 5

Photomicrographs of coronal sections showing aromatase immunoreactivity in the mPOA of sires in three different housing conditions. Each panel shows representative images of a confocal laser-scanning microscope. Cells with aromatase immunoreactivity (red) around the nucleus (blue) in the mPOA of sires in the nursing cage with whole family (S5, A), sires with their pairmate dams in old cages (S6-0, B) or new cages (S6-3, C) after removing pups for 10 min (B,C). (D) Statistical analysis of the semiquantitative determination of aromatase expression in different isolation conditions. N = 10–16 areas in the mPOA in three mice each. One-way ANOVA followed by Bonferroni's post-hoc test: F_{(2, 33)} = 47.45, P < 0.001; ^***P < 0.001 from values of sires at S5. ^#P < 0.01 from values of sire at S6-0.

Figure 6

Time course of the aromatase immunofluorescence-positive cell number in sires treated with letrozole or not. (A) By means of a confocal laser-scanning microscope, the number of cells with aromatase immunoreactivity (red) around the nucleus (blue) was counted in the mPOA regions of sires at 0, 10, and 30 min after isolation together with their pairmate dams in new cages. Five to seven mice were used for each time point. One-way ANOVA followed by Bonferroni's post-hoc test: F_{(2, 12)} = 20.76, P < 0.0001; F_{(2, 18)} = 21.85, P = 0.1860. ^***P < 0.001 from values at 0 min. ^#P < 0.001 from values in the presence of letrozole (+Let), two-tailed Student's t-test. (B,C) Representative images of the mPOA in sires treated with (Let) or without (con) of letrozole, respectively.

Figure 7

Inhibition of the co-housing-induced paternal retrieval behavior by an aromatase inhibitor. The percentage of male parental behaviors was measured based on the retrieval responses after co-habitation with the mother for a 10-min separation period. The couples were housed with their pups for 4 days in their nurturing cages and then removed together from their home cage. Experimental sires were then injected intraperitoneally with letrozole (1 mg/kg of body weight). After 30 or 60 min, the sires were removed along with mates to allow habituation in new cages for 10 min. The sires were returned to the original home with his pups and the retrieval behavior was monitored. N = 10 family units in each group. One-way ANOVA followed by Bonferroni's post-hoc test detected significant differences between time points: F_{(2, 27)} = 11.92, P < 0.001, n = 10 in each group. ^*P < 0.05, ^***P < 0.001.

Figure 8

Effect of letrozole on parental retrieval behavior development. The capacity for pup retrieval by first-time sires (A) and primiparous dams (B) was measured daily from day 1 to day 5 after parturition. The sires (n = 31) were administered peritoneally each day with letrozole (1 mg/kg of body weight, n = 57–67) or PBS (n = 30–31). The dams were also treated in a similar manner with letrozole (n = 26–31) or PBS (n = 5–9). After 1 h, the retrieval behavior per 10 min was measured, which was repeated for 5 days. The number of pups retrieved was counted and expressed as a percentage. Fisher's exact test, two-tailed:^aP < 0.05, ^bP < 0.02 vs. day 1; ^cP < 0.05,^dP < 0.01, ^eP < 0.001 between treated and untreated sires.

Figure 9

Neural model showing the circuits that may regulate the paternal retrieval behavior. This figure is based on Sheehan et al. (2004), Zhong et al. (2014), Akther et al. (2014), Numan et al. (2005), Rilling and Young (2014), Numan (2015), and Bridges (2015). The figure shows how aromatase-expressing neurons relay signals related to social stimuli, which convert the sensory information received by sires to elicit the retrieval behavior. The sensory inputs are processed in the prefrontal cortex (PFC), CA3 region of the hippocampus (CA3), and medial amygdala (AMY), which are grouped. The output signals from the three regions are transmitted to the ventromedial hypothalamus (VMH), medial preoptic area (mPOA), and lateral septum dorsal (LSD), followed by the ventral tegmental area (VTA), and nucleus accumbens (NAcc). The final region is the ventral pallidum (VP), which induces motion. Information related to neurotransmitters and excitatory or inhibitory regulators is omitted for simplicity.