Social isolation impairs adult neurogenesis in the limbic system and alters behaviors in female prairie voles

Abstract

Disruptions in the social environment, such as social isolation, are distressing and can induce various behavioral and neural changes in the distressed animal. We conducted a series of experiments to test the hypothesis that long-term social isolation affects brain plasticity and alters behavior in the highly social prairie vole (Microtus ochrogaster). In Experiment 1, adult female prairie voles were injected with a cell division marker, 5-bromo-2'-deoxyuridine (BrdU), and then same-sex pair-housed (control) or single-housed (isolation) for 6 weeks. Social isolation reduced cell proliferation, survival, and neuronal differentiation and altered cell death in the dentate gyrus of the hippocampus and the amygdala. In addition, social isolation reduced cell proliferation in the medial preoptic area and cell survival in the ventromedial hypothalamus. These data suggest that long-term social isolation affects distinct stages of adult neurogenesis in specific limbic brain regions. In Experiment 2, isolated females displayed higher levels of anxiety-like behaviors in both the open field and elevated plus maze tests and higher levels of depression-like behavior in the forced swim test than controls. Further, isolated females showed a higher level of affiliative behavior than controls, but the two groups did not differ in social recognition memory. Together, our data suggest that social isolation not only impairs cell proliferation, survival, and neuronal differentiation in limbic brain areas, but also alters anxiety-like, depression-like, and affiliative behaviors in adult female prairie voles. These data warrant further investigation of a possible link between altered neurogenesis within the limbic system and behavioral changes.

Fig. 1

Chronic social isolation in adulthood affected long-term cell survival (assessed by BrdU-labeling) and proliferation (assessed by Ki67-labeling) in the brains of female prairie voles. Compared to pair-housed females (control, n=9), single-housed females (isolation, n=9) showed fewer BrdU-labeled cells in the amygdala (AMY) (A–C), dentate gyrus (DG) of the hippocampus (D–F), and ventromedial hypothalamus (VMH) (G), but not in the medial preoptic area (H, K). In addition, the number of BrdU-labeled cells in the AMY and DG was significantly higher than that in the medial preoptic area (MPOA) and VMH (K). Socially isolated females (n=8) also showed fewer Ki67-labeled cells in the DG (J) and MPOA, compared to pair-housed females (n=8; L). Finally, the number of Ki67-labeled cells in the AMY (I) was similar to the number in the DG, but higher than that in the MPOA and VMH (L). opt: optic tract; BlA: basolateral, CeA: central, CoA: cortical, and MeA: medial nuclei of the AMY. *p<0.05 and **p<0.01. Alphabetic characters represent the results of the post-hoc test. Error bars represent SEM. Scale bar=500 μm.

Fig. 2

Stacked confocal microscopy images illustrating labeled cells in the granular cell layer of the hippocampus (A–E) and amygdala (AMY) (F–I). Cells were labeled for BrdU (red), the mature neuronal marker NeuN (green), or both (yellow). The white box in Panel A indicates the area of the DG with high magnification (B–D; stack thickness=10 μm). Arrows indicate cells double-labeled for BrdU/NeuN. Arrowheads indicate the double-labeled cell shown in the magnified image in the DG (E) and AMY (I; stack thickness =8 μm). 3D co-localization of BrdU and NeuN is demonstrated using views along the y–z axis (right) and x–z axis (below). Scale bar=20 μm (A–D and F–H) or 5 μm (E and I). (J) Isolated females (n=8) had significantly lower percentages of BrdU-ir cells that co-labeled with NeuN in the hippocampal granular cell layer (GCL) and AMY, compared to control females (n=10). *p<0.05. Error bars represent SEM.

Fig. 3

Photo images illustrating TUNEL-labeled cells in the AMY (A) and the hippocampus (B). White boxes in Panels A and B indicate cells visualized under high magnification. opt, optic tract. Scale bar=250 μm (A and B) or 25 μm (insets in A and B). (C) Number of TUNEL-labeled cells in the AMY and DG between pair-housed (control, n=8) and single-housed (isolation, n=9) females. (D) Ratio of TUNEL-ir/Ki67-ir cells in the AMY and DG between control and isolated females. *p<0.05 and **p<0.01. Error bars represent SEM.

Fig. 4

Chronic social isolation in adulthood affected anxiety- and depression-like behaviors of female prairie voles. In the open field test, single-housed females (isolation, n=13) made fewer center entries (A) and spent less time in the center and more time in the corners of the apparatus (B), compared to the pair-housed females (control, n=13). The two groups did not differ in locomotor activity (C). In the elevated plus maze test, single-housed females (n=11) did not differ from control females in the latency to enter the open arm (n=13, D). While the ratio of open to total arm duration tended to be higher in control females (E), isolation females spent more time in the closed arms than control females (F). The two groups did not differ in the total number of arm entries (G). In the forced swim test, single-housed females (n=15) had a shorter latency to immobility (H) and a higher number of immobility bouts (I) than control females (n=11). Single-housed females also spent more time immobile than control females (J), while locomotion (K) did not differ between the two groups. *p<0.05 and **p<0.01. Error bars represent SEM.

Fig. 5

In the two-chambered social affiliation test, single-housed females (isolation, n=11) spent more time in the cage containing a conspecific young adult female and less time in the empty cage than control females (n=11; A). Single-housed females also tended to spent more time in direct body contact with the conspecific compared to the control females (B). No group differences were found in locomotor activity (C). The social recognition test consisted of three 5-minute exposures to the same juvenile (T1, T2, and T3), followed by a 5-minute exposure to a new juvenile (New). The inter-exposure interval was 30 min. Single-housed (n=11) and control females (n=7) did not differ in the frequency and duration of olfactory investigation of the juvenile female (D, E). For both treatment groups, trial 3 (T3) differed significantly from the other trials for frequency and duration of olfactory investigation. Furthermore, close pursuit duration was higher in single-housed as compared to control females (F). *p<0.05 and **p<0.01. Error bars represent SEM.