Nonsteroidal anti-inflammatory treatment prevents delayed effects of early life stress in rats

Abstract

Background: Early developmental insults can cause dysfunction within parvalbumin (PVB)-containing interneurons in the prefrontal cortex. The neuropsychiatric disorders associated with such dysfunction might involve neuroinflammatory processes. Cyclooxygenase-2 (COX-2) is a key mediator of inflammation and is therefore a potential target for preventive treatment. Here, we investigated whether the developmental trajectories of PVB expression and COX-2 induction in the prelimbic region of the prefrontal cortex are altered after maternal separation stress in male rats.

Methods: Male rat pups were separated from their mother and littermates for 4 hours/day between postnatal Days 2 and 20. Western blotting and immunohistochemistry were used to analyze PVB and COX-2 expression in the prefrontal cortex and hippocampus. A separate cohort of animals was treated with a COX-2 inhibitor during preadolescence and analyzed for PVB, COX-2, and working memory performance.

Results: We demonstrate that maternal separation causes a reduction of PVB and an increase in COX-2 expression in the prefrontal cortex in adolescence, with concurrent working memory deficits. Parvalbumin was not affected earlier in development. Prophylactic COX-2 inhibition preadolescence prevents PVB loss and improves working memory deficits induced by maternal separation.

Conclusions: These data are the first to show a preventive pharmacological intervention for the delayed effects of early life stress on prefrontal cortex interneurons and working memory. Our results suggest a possible mechanism for the relationship between early life stress and interneuron dysfunction in adolescence.

Figure 1

Measurement of parvalbumin (PVB) changes over development in control (CON) and maternal separation (MS) animals with Western blot analysis. Top: graphic representation of relative PVB at two ages (juvenile [Juve], postnatal day [P]25 and adolescent [Adolesc], P40) after CON or MS rearing in prefrontal cortex (PFC) (A) or hippocampus (HIP) (B) tissue; *p<.05 different from CON; @p<.05 different from juveniles. Bottom: representative Western blots from the PFC (A) or HIP (B) showing all groups tested. Means ± SE presented. OC, optical density.

Figure 2

Top: graphic representation of PVB-immunoreactivity at two ages after CON or MS rearing; *p<.05 different from CON; @p<.05 different from Juve. Bottom: representative PVB-immunoreactive cells within the PFC of Adolesc. Both images taken from the same location and layer (5) within the PFC. 40×: bar: 33 μm. Means ± SE presented. Abbreviations as in Figure 1.

Figure 3

Cyclooxygenase-2 (COX-2) levels during adolescence after CON or MS conditions. (A) Western blot analysis of COX-2 in the prelimbic PFC of CON and MS animals; *p<.05. (B) Western blot analysis of COX-2 in the HIP of CON and MS animals. A representative Western blot from each region is shown below each graph. Due to different concentrations of proteins in HIP, COX-2 and actin required different exposure times, necessitating two separate images. Means ± SE presented. Abbreviations as in Figure 1.

Figure 4

Effects of preadolescent treatment with NS-398 (NS) on PVB (A) and COX-2 (B) levels in the PFC during adolescence after CON or MS conditions. Means SE presented. A representative Western blot from each region is shown below each graph; *p<.05; (C) linear regression analysis shows a direct relationship between COX-2 and PVB levels; R=−.613; p=.02. Note: 2–3 subjects/group were not included in regression analysis, due to lack of within-subject measurement of both proteins. Veh, vehicle; other abbreviations as in Figures 1 and 3.

Figure 5

Problems with working memory in MS animals were detected with the win-shift test, regardless of delay. (A) Total errors made after each delay; (B) total errors made by each group collapsed across delay. @p<.05 different from CON; *p<.05 different from Veh. Means ± SE presented. Abbreviations as in Figures 1 and 4.