Pregnancy-induced oxidative stress and inflammation are not associated with impaired maternal neuronal activity or memory function

Abstract

Pregnancy is associated with neural and behavioral plasticity, systemic inflammation, and oxidative stress, yet the impact of inflammation and oxidative stress on maternal neural and behavioral plasticity during pregnancy is unclear. We hypothesized that healthy pregnancy transiently reduces learning and memory and these deficits are associated with pregnancy-induced elevations in inflammation and oxidative stress. Cognitive performance was tested with novel object recognition (recollective memory), Morris water maze (spatial memory), and open field (anxiety-like) behavior tasks in female Sprague-Dawley rats of varying reproductive states [nonpregnant (nulliparous), pregnant (near term), and 1-2 mo after pregnancy (primiparous); n = 7 or 8/group]. Plasma and CA1 proinflammatory cytokines were measured with a MILLIPLEX magnetic bead assay. Plasma oxidative stress was measured via advanced oxidation protein products (AOPP) assay. CA1 markers of oxidative stress, neuronal activity, and apoptosis were quantified via Western blot analysis. Our results demonstrate that CA1 oxidative stress-associated markers were elevated in pregnant compared with nulliparous rats (P ≤ 0.017) but there were equivalent levels in pregnant and primiparous rats. In contrast, reproductive state did not impact CA1 inflammatory cytokines, neuronal activity, or apoptosis. Likewise, there was no effect of reproductive state on recollective or spatial memory. Even so, spatial learning was impaired (P ≤ 0.007) whereas anxiety-like behavior (P ≤ 0.034) was reduced in primiparous rats. Overall, our data suggest that maternal hippocampal CA1 is protected from systemic inflammation but vulnerable to peripartum oxidative stress. Peripartum oxidative stress elevations, such as in pregnancy complications, may contribute to peripartum neural and behavioral plasticity.NEW & NOTEWORTHY Healthy pregnancy is associated with elevated maternal systemic and brain oxidative stress. During postpregnancy, brain oxidative stress remains elevated whereas systemic oxidative stress is resolved. This sustained maternal brain oxidative stress is associated with learning impairments and decreased anxiety-like behavior during the postpregnancy period.

Keywords: anxiety-like behavior; cognition; inflammation; oxidative stress; parity.

Figure 1.

Circulating inflammatory cytokine profiles in healthy rats with various reproductive experiences: plasma concentrations of proinflammatory (TNF-α, IL-6, IL-1β, IL-17A) and anti-inflammatory (IL-10) cytokines. NULLI, nulliparous (n = 14–16); PREG, gestational day 21 (n = 6 or 7); PRIMI, primiparous (2 mo postpregnancy, n = 6 or 7). One-way ANOVA with Tukey’s multiple comparisons test, means ± SD. Outliers removed before analysis: TNF-α (A): NULLI = 2 (n = 14), PREG = 1 (n = 6); IL-6 (B): NULLI = 1 (n = 15); IL-1β (C): NULLI = 1 (n = 15); IL-17A (D): PRIMI = 1 (n = 6); IL-10 (E): NULLI = 2 (n = 14), PREG = 1 (n = 6).

Figure 2.

Inflammatory cytokine profiles in dorsal hippocampal CA1 of healthy rats with various reproductive experiences: concentrations of proinflammatory [TNF-α (A), IL-6 (B), IL-1β (C), IL-17A (D)] and anti-inflammatory [IL-10 (E)] cytokines within the CA1 of the dorsal hippocampus. NULLI, nulliparous (n = 10 or 11); PREG, gestational day 21 (n = 5 or 6); PRIMI, primiparous (2 mo postpregnancy, n = 5 or 6). One-way ANOVA with Tukey’s multiple comparisons test, means ± SD. Outliers removed before analysis: IL-6 (B): PREG = 1 (n = 5), NULLI = 2 (n = 10); IL-1β: NULLI = 2 (n = 10), PRIMI = 1 (n = 5); IL-17A: NULLI = 1 (n = 11); IL-10: NULLI = 2 (n = 10).

Figure 3.

Circulating oxidative stress and cell activity in CA1 of dorsal hippocampus. A: plasma concentrations of advanced oxidation protein products (AOPP, measure of oxidative stress). B: Western blot image of enzyme-mediated cleavage of Spectrin (270 kDa), resulting in cleaved fragments at 150 kDa (calpain mediated) and 120 kDa (caspase-3 mediated) in CA1. C: quantification of calpain-mediated cleavage of Spectrin (measure of oxidative stress) in CA1. D: quantification of caspase 3-mediated cleavage of Spectrin (measure of apoptosis) in CA1. E and F: protein expression of early growth response protein-1 (Egr-1, neuronal activation marker) in CA1. NULLI, nulliparous (n = 12–16); PREG, gestational day 21 (n = 6); PRIMI, primiparous (2 mo postpregnancy, n = 6). Total Spectrin and cleavage products were first normalized to β-actin, and a proportion of cleaved product to total Spectrin was analyzed. Protein expression shown as % of NULLI. PREG rats were compared with age-matched NULLI rats run in parallel (lane 1 in B and E), and PRIMI rats were compared with age-matched NULLI rats run in parallel (lane 3 in B and E). NULLI control rats were collapsed into 1 group for group comparisons. One-way ANOVA with Tukey’s multiple comparisons test, means ± SD. No outliers removed before analysis.

Figure 4.

Recollective memory during novel object recognition behavior task. A: latency to initial contact with novel object. B: total contacts with novel object. C: total distance traveled during test. D: latency to initial movement during test. NULLI, nulliparous (n = 12–16); PREG, gestational day 20 (n = 7 or 8); PRIMI, primiparous (2 mo postpregnancy, n = 7). One-way ANOVA with Tukey’s multiple comparisons test, means ± SD. Outliers removed before analysis: latency to initial movement (D): NULLI = 4 (n = 12), PREG = 1 (n = 7).

Figure 5.

Learning indexes and spatial memory during Morris water maze. A: latency to target on learning days 2–4 of behavior task. B: learning index calculated for each rat based on latency to target on learning days 2–4. A lower learning index represents better learning. C: spatial memory assessed via latency to target on day 5 of behavior task. D: pathlength to target on learning days 2–4 of behavior task. E: learning index calculated for each rat based on pathlength to target on learning days 2–4. A lower learning index represents better learning. F: spatial memory assessed via pathlength to target on day 5 of behavior task. NULLI, nulliparous (n = 7 or 8); PRIMI, primiparous [1 wk after weaning (postpregnancy day 35), n = 7]. Two-way repeated-measures ANOVA with Sidak’s multiple comparisons test (A and D) or unpaired t test (B, C, E, and F), means ± SE. *P < 0.05 vs. PRIMI day 4; #P < 0.05 vs. NULLI day 2. No outliers removed before analysis.

Figure 6.

Anxiety-associated behavior during open field behavior task. A: total entries into the center of the open arena. B: duration in seconds in center of open area. C: distance traveled in the center of the open arena divided by the total distance traveled during behavior task. D: plasma concentrations of stress-associated steroid hormone corticosterone. NULLI, nulliparous (n = 16); PREG, gestational day 20 (n = 7 or 8); PRIMI, primiparous (2 mo postpregnancy, n = 7). One-way ANOVA with Tukey’s multiple comparisons, means ± SD. No outliers removed before analysis.