Solitude and serotonin: juvenile isolation alters the covariation between social behavior and cFos expression by serotonergic neurons

Abstract

Variation in the mutual responsiveness of social partners to each other can be reflected in behavioral suites that covary with neural activity in ways that track the salience or valence of interactions. Juvenile social isolation alters social behavior and neural activity during social interaction, but whether and how it alters the covariation between behavior and neural activity has not been as well explored. To address this issue, four classes of experimental subjects: isolated males, socially housed males, isolated females, and socially housed females, were paired with an opposite-sex social partner that had been socially housed. Social behaviors and c-Fos expression in the serotonergic dorsal raphe nucleus (DRN) were then measured in subjects following the social interactions. Relative to social housing, postweaning isolation led to a decrease in the density of neurons double-labeled for tryptophan hydroxylase and c-Fos in the dorsomedial subdivision of the DRN, regardless of sex. Vocal and non-vocal behaviors were also affected by isolation. In interactions with isolated males, both ultrasonic vocalization (USVs) and broadband vocalizations (squeaks) increased in conjunction with greater male investigation of females. Neural and behavioral measures also correlated with each other. In the isolated male group, the density of double-labeled neurons in the dorsomedial DRN was negatively correlated with USV production and positively correlated with a principal component of non-vocal behavior corresponding to greater defensive kicking by females and less investigation and mounting behavior. This correlation was reversed in direction for socially housed males, and for isolated males versus isolated females. These findings confirm that the dynamics of social interactions are reflected in c-Fos activation in the dorsomedial DRN, and suggest an altered responsiveness of serotonergic neurons to social interaction following social isolation in males, in parallel with an altered male response to female cues.

Keywords: dorsal raphe nucleus; opposite sex interaction; postweaning; serotonin; social competence; social isolation; social plasticity; ultrasonic vocalization.

Figure 1

Top: Diagram of experimental design showing the three social treatment groups (A–C) that result in four subject groups: 1. isolated males (light blue), 2. isolated females (light green), 3. socially housed males (dark blue), and 4. socially housed females (dark green), each interacting with a socially housed partner. Some behavioral measurements that could not be ascribed to a single individual, such as call number and nose-to-nose investigation, are the same for social males and females in the same specific interaction. Bottom: timeline of experimental manipulations.

Figure 2

Antibody labeling. (A) Double label of antibodies to tryptophan hydroxylase (TPH) and cFos in the DRN. Arrows indicate neurons with both labels. (B) Sample images showing TPH label from each of the four subject groups. White arrowheads indicate double-labeled neurons. AQ, aqueduct; PAG, periaqueductal gray; DRD, dorsomedial dorsal raphe; DRV, dorsoventral dorsal raphe; DRL, dorsolateral dorsal raphe; PDR, posterodorsal raphe; MLF, median longitudinal fasciculus. Images have been enhanced in brightness and contrast for the figure. Insets: white arrowheads indicate double-labeled neurons, yellow arrowheads indicate neurons labeling for TPH but not c-Fos, and white arrows indicate c-Fos-labeled nuclei. (C) Densities of neurons labeled for antibodies to c-Fos and TPH in four subregions of the dorsal raphe nucleus. The four subject groups of individually versus socially housed females and males are represented for each subregion. *p < 0.05.

Figure 3

Differences in calls across the three different social groups. (A) Spectrograms showing examples of a BBV (left) and harmonic and non- harmonic USVs (right). (B) The numbers of harmonic USVs, but not non-harmonic USVs, are higher in groups with individually housed males. Values are shown as log transformations to better compare call types. (C) Numbers of harmonic versus nonharmonic USVs are correlated in the three social groups. Harmonic and nonharmonic calls are significantly correlated in all groups. (D) The numbers of BBVs are higher in groups with individually housed males. Identical values are slightly displaced from each other to distinguish all measurements. *p < 0.05; **p < 0.001.

Figure 4

Plots of call ratio (numbers of harmonic/non-harmonic calls) for the four subject groups versus numbers of neurons expressing the double label, across individual mice. (A) Isolated males, (B) isolated females, (C) social males, and (D) social females (dark green). Note the difference in Y-axis values across groups. Y-axis values for groups (C,D) are the same for individual males and females that were in the same social interaction.

Figure 5

Pie charts showing proportions of time spent performing different behaviors by males (top row) and females (bottom row) during an opposite-sex social interaction. The three social treatment conditions of isolated males (left), isolated females (middle) and socially housed males and females (right) are shown. Females in the isolated male treatment and males in the isolated female treatment were socially housed. “nose-nose” = nose-to-nose investigation; “anogen” = anogenital investigation.

Figure 6

Box-and-whiskers plots of behaviors shown by males and females during an opposite-sex social interaction. (A) Isolated males show a lower total duration of digging (left panel) and a higher number of mounts (right panel) than the socially housed males in the other two treatment groups. (B) Isolated females also show a lower total duration of digging (left panel) than the socially housed females in the other treatment groups. Although numbers of kicks by females are higher in both of the isolated groups than the social group (right panel), this difference is not significant. *p value of less than 0.05, **p value of less than 0.01.

Figure 7

Comparison of the density of double labeled neurons in the DRD and behavior during the social interaction. (A) Plot of the first versus second dimensions of a categorical PCA of males and female behaviors. Open blue circles = the isolated male group. Open gray circles = the isolated female group. Filled gray circles = the social group. (B) Comparison of the densities of double-labeled neurons in the DRD versus dimension 2 scores of the categorical PCA in males. Values correlate positively across isolated males (open circles) but negatively across socially housed males (closed circles). (C) Comparison of the densities of double-labeled neurons in the DRD versus dimension 2 scores of the categorical PCA in females. Values are negatively correlated in isolated females.