Spatial Learning Is Impaired in Male Pubertal Rats Following Neonatal Daily but Not Randomly Spaced Maternal Deprivation

Abstract

Severe early life stress has long been associated with neuropsychological disorders in adulthood, including depression, schizophrenia, post-traumatic stress disorder, and memory dysfunction. To some extent, all of these conditions involve dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis and reduced negative feedback inhibition of cortisol release in adulthood. However, the time course for mental health and hormonal outcomes across life stages and the attributes of early life stress that direct the behavioral and biological alterations is not fully understood. We designed our studies to compare outcomes of the two most common maternal deprivation schedules on cognitive ability prior to adulthood. We exposed rat pups to daily or randomly spaced maternal separation bouts within the first 3 weeks of life and examined cognitive performance, neurotrophic signaling, and stress and immune system markers during puberty. We found that the daily separation schedule impaired spatial learning while the randomly spaced schedule did not alter maze performance relative to normally reared control animals. Animals that underwent daily separation showed a tendency for reduced body weight compared to the randomly spaced condition, but there were no differences in adrenal weight. Thymus weight normalized by body weight was increased following daily separation compared to random separation and control conditions. Plasma corticosterone levels measured after behavior testing did not differ amongst experimental groups and there was no impact of TrKB receptor inhibition. Combined, the results show that different early life stress schedules produce different behavioral and biological outcomes when measured at puberty. Combined with prior findings from more mature animals, the results presented here suggest that daily neonatal stress produces varied alterations in spatial cognition at different life stages with a transient learning deficit at puberty preceding a more persistent and a progressive memory impairment through adulthood and into aging.

Keywords: brain derived neuronal factor; corticosterone; glucocorticoid receptor; hippocampus; maternal deprivation; spatial learning; tyrosine receptor kinase B.

FIGURE 1

Event timeline. The onsets and durations for all experimental events are depicted. All animals were handled briefly on P23–28. Mock injections were performed on P34–39. Drug (K252a) or ACSF was infused 1 h prior to the 24-HP. Tissues were harvested after the 24-HP. IMM, Immediate probe trial performed on the same day as training.

FIGURE 2

Image of a sagittal brain section showing cannula placement. Red arrows designate the most ventral extents of the cannulae tracts. The triangular notch in the right hemisphere designates brain orientation.

FIGURE 3

MWM and depiction of spatial analyses. The escape platform is just barely visible in the Goal quadrant (with dotted circle). Spatial cues can be seen attached to curtains that surround the pool. Boxed letters indicate the pseudo-randomized starting locations used during training and probe trials. Bold boxes indicate potential start locations for probe trials for the platform location shown in the figure. Dashed gray lines indicate the boundaries of the four quadrants (Goal, two Adjacent, and Opposite). Distance to platform (DtP) analysis is conducted by noting the linear distance between the center of the rat and the platform center (D) in video snapshots separated by 1 s (t_x).

FIGURE 4

Summary of MD effects on escape learning in the MWM. (A) Litter mean escape latency in seconds for each training block across MD conditions. Sample sizes: CON = 11; P2–11 = 4; RAN = 6. (B) Bar graphs showing litter mean quadrant dwell time for each MD condition during the IMM probe trial. Asterisks denote significant Chi Square tests for quadrant (^∗∗∗p < 0.001, ^∗∗p < 0.01). (C) Each point marks the litter’s average distance between the animal and the platform location (DtP) across the first fifteen seconds of the IMM probe for each MD condition. The asterisk denotes significant RMANOVA main effect of MD condition (p < 0.05). The pound symbol (#) denotes significant Tukey effect of MD condition (p < 0.05). Sample sizes: CON = 11; P2–11 = 4; RAN = 6. The open circle on the Y-axis marks the starting location distance for all animals (120 cm).

FIGURE 5

Summary of MD effects on spatial memory in the 24 h probe trial. (A) Quadrant dwell time distributions of litter means across MD and drug conditions. (B) Litter distance to platform location across MD and drug conditions. Sample sizes: ACSF (CON = 10; P2–11 = 4; RAN = 6); K252a (CON = 10; P2–11 = 3; RAN = 5). The open circle on the Y-axis marks the starting location distance for all animals (120 cm).

FIGURE 6

Mean plasma CORT levels collected after the 24-HR probe trial across MD and drug conditions. Dots represent individual litter means. (A) CORT levels for animals treated with ACSF before the 24-HR probe trial. (B) CORT levels for animals treated with K252a before the 24-HR probe trial. Sample sizes: ACSF: CON = 4, P2–11 = 4, RAN = 5; K252a: CON = 6, P2–11 = 2, RAN = 4.

FIGURE 7

Body and tissue weights across MD groups. Dots represent individual litter means. (A) Body weights at time of surgery (P30) compared across MD condition with litter as a random factor. Body weight sample sizes: CON = 58; P2–11 = 24; RAN = 33. (B) Adrenal weight normalized by body weight across MD condition. Adrenal weight sample sizes: CON = 58; P2–11 = 24; RAN = 33. (C) Thymus weight normalized by body weight across MD condition. Thymus weight sample sizes CON = 58; P2–11 = 24; RAN = 33; RAN = 10. Asterisks denote significance (^∗∗∗p < 0.001, ^∗∗p < 0.01, ^∗p < 0.05).

FIGURE 8

Table listing results from prior studies of MD with daily separation side-by-side with the current study. The Prior Reports column shows the consensus of research studies utilizing a daily (Daily) separation schedule with assays performed in adulthood (Anisman et al., 1998; Huot et al., 2002; Heim et al., 2004; Bartolomucci et al., 2005; Brunson et al., 2005; Greisen et al., 2005; Ladd et al., 2005; Reber et al., 2007; Chourbaji et al., 2010; O’Sullivan et al., 2011; Lee et al., 2012; McEwen et al., 2015; Holubová et al., 2016; Bolton et al., 2017; Chaby et al., 2017; Récamier-Carballo et al., 2017). The Current Study columns summarize the results from the P2–11 and RAN separation schedules. Arrows indicate direction of change in each parameter relative to non-stressed controls. No Change indicates that there is no difference between stressed and non-stressed controls. No Consensus indicates that there is no clear agreement in the literature. TBD indicates “to be determined.”