Affiliative behavior, ultrasonic communication and social reward are influenced by genetic variation in adolescent mice

Abstract

Social approach is crucial for establishing relationships among individuals. In rodents, social approach has been studied primarily within the context of behavioral phenomena related to sexual reproduction, such as mating, territory defense and parental care. However, many forms of social interaction occur before the onset of reproductive maturity, which suggests that some processes underlying social approach among juvenile animals are probably distinct from those in adults. We conducted a longitudinal study of social investigation (SI) in mice from two inbred strains to assess the extent to which genetic factors influence the motivation for young mice to approach one another. Early-adolescent C57BL/6J (B6) mice, tested 4-6 days after weaning, investigated former cage mates to a greater degree than BALB/cJ (BALB) mice, irrespective of the sex composition within an interacting pair. This strain difference was not due to variation in maternal care, the phenotypic characteristics of stimulus mice or sensitivity to the length of isolation prior to testing, nor was it attributable to a general difference in appetitive motivation. Ultrasonic vocalization (USV) production was positively correlated with the SI responses of mice from both strains. Interestingly, several USV characteristics segregated with the genetic background of young mice, including a higher average frequency and shorter duration for the USVs emitted by B6 mice. An assessment of conditioned place preference responses indicated that there was a strain-dependent difference in the rewarding nature of social contact. As adolescent mice aged, SI responses gradually became less sensitive to genetic background and more responsive to the particular sex of individuals within an interacting pair. We have thus identified a specific, genetic influence on the motivation of early-adolescent mice to approach one another. Consistent with classical theories of motivation, which propose a functional relationship between behavioral approach and reward, our findings indicate that reward is a proximal mechanism through which genetic factors affect social motivation during early adolescence.

Figure 1

Social approach among adolescent mice. (a) The SI responses of test mice were quantified during 5-min tests throughout adolescent development. Weaning age and the average age of reproductive maturity in females is illustrated for mice from both strains. (b) Photograph of a B6 mouse investigating a former cage mate after 24 hrs of social isolation.

Figure 2

Differences in social investigation between adolescent mice as a function of genetic background, age and sex. SI responses of BALB and B6 test mice during adolescent development for (a) males approaching females, (b) females approaching males, (c) males approaching males and (d) females approaching females. All data are presented as the mean±SEM (* P<0.05, ** P<0.01, *** P<0.001 for BALB vs. B6 mice; ^# P<0.05, ^## P<0.01 for PD 25/26 vs. PD 45/46 mice).

Figure 3

Strain-dependent differences in social investigation as a function of maternal care, stimulus mouse characteristics and length of social isolation. (a) Following 24 hours of social isolation, on PD 30/31, male B6 mice investigated a familiar female stimulus mouse for a greater duration then age-matched BALB males. This strain-dependent pattern was also expressed by (b) male mice that had been raised by a mother of the alternate strain, (c) male mice approaching a female from the alternate strain and (d) male mice approaching a same-strain female after 8–10 days of social isolation. All data are presented as the mean±SEM (** P<0.01, *** P<0.001).

Figure 4

Approach behaviors of adolescent mice towards a novel olfactory stimulus and a food source. After 24 hrs of social isolation, (a) BALB and B6 mice investigated a lemon-scented cotton ball for a similar amount of time. Following complete food deprivation during a 24-hr social isolation period, (b) BALB males consumed more food (standard lab chow) than mice from the other groups during a 10-min period. All data are presented as the mean±SEM (* P<0.05 compared to BALB females and B6 mice of both sexes).

Figure 5

Production and sonographic characteristics of ultrasonic vocalizations during the social interactions of early-adolescent mice. USV emission was positively associated with the SI responses of early-adolescent (a) B6 and (b) BALB mice. (c) USV production was selectively modulated during social interactions that involved a male test mouse from the BALB strain. (d–d′) Emission rates were similar for mice from both strains when USVs were detected. However, compared to BALB mice, the USVs of B6 mice occurred at (e–e′) higher average frequencies and lasted for (f–f′) shorter durations. Data in Figures d–f and d′–f′ are presented as frequency distributions of the raw acoustic signal that was collected during SI tests on PD 30/31. A portion of the data (<0.5% of the sample from each strain) is not illustrated to keep the abscissa of each distribution within a reasonable size for presentation. Data in Figure 5c are presented as the mean±SEM (* P<0.05, ** P<0.01 for BALB vs. B6 mice).

Figure 6

Classification of ultrasonic vocalizations into distinct categories. Representative sonograms and descriptive statistics for different types of USV that were emitted by (a–d) BALB and (a′–d′) B6 mice (see Results for classification criteria). Descriptive statistics (mean±std. dev.) are given for the duration of each call type, as well as the beginning and ending dominant frequency. The USV traces are arbitrarily aligned to the 100-msec demarcation on the abscissa of each sonogram. Intensity changes within each representative vocalization are color-coded. (a″–d″) The production of each USV subtype was strain-dependent (* P<0.05, *** P<0.001 for BALB vs. B6 mice).

Figure 7

Sonographic characteristics of punctuated ultrasonic vocalizations in relation to the genetic background of mice. (a) B6 mice emitted more punctuated USVs than BALB mice. Four sonographic traces illustrate the varied form of punctuated USVs emitted by early-adolescent mice: (b₁) a BALB punctuated USV with 2 pitch-jumps, (b₂) a B6 punctuated USV with 2 pitch-jumps and 2 harmonics corresponding to the 1^st and 3^rd elements of the vocalization, (b₃) a B6 punctuated USV with 1 pitch-jump and 1 harmonic corresponding to the 1^st element of the vocalization, and (b₄) a B6 punctuated USV with 1 pitch-jump. (c) 141 BALB and (d) 140 B6 punctuated USVs were graphed as scatter-plots to illustrate the degree of internal frequency modulation present within individual ultrasonic syllables. Data points lying off the diagonal of each plot represent pitch-jumps. (e) Line graph depicting the number of punctuated USVs emitted during a 5-min interaction between mice that expressed the most punctuated USVs of each strain. The arrow and arrowheads denote pairs of mice that emitted a similar number of punctuated USVs. (f–f′) Data re-plotted from Figures 7c and 7d, respectively. (g) The pitch-jumps of punctuated USVs from one extreme BALB pair. (g₁′–g₄′) The pitch-jumps of four B6 pairs that produced punctuated USVs at a rate comparable to the BALB pair illustrated in Figure 7g. (h–h′) Data from Figures 7f and 7f′ with pitch-jumps from Figure 7g′ subtracted and the pitch-jumps from Figures 7g₁′–g₄′ added, respectively. (i) The total number of pitch-jumps within individual punctuated USVs for each strain graphed as a relative-density histogram.

Figure 8

Strain-dependent variation in the social conditioned place preference responses of early-adolescent mice. Unconditioned mice from both strains expressed a preference for the paper bedding (control bars) and this natural bias obscured any conditioning effect that might have resulted from pairing social interaction with paper bedding (social plus paper bars). However, there was a robust, strain-dependent SCPP response for B6 mice when the less-preferred aspen bedding was paired with social enrichment (social plus aspen bars). Preference scores were calculated as the time each mouse spent in the aspen bedding-lined environment minus the duration in the paper bedding-lined compartment of the place-preference arena. All data are presented as the mean±SEM (*** P<0.001 for BALB vs. B6 mice).