Maternal stress and vulnerability to depression: coping and maternal care strategies and its consequences on adolescent offspring

Abstract

Depressive mothers often find mother-child interaction to be challenging. Maternal stress may further impair mother-child attachment, which may increase the risk of negative developmental consequences. We used rats with different vulnerability to depressive-like behavior (Wistar and Kyoto) to investigate the impact of stress (maternal separation-MS) on maternal behavior and adolescent offspring cognition. MS in Kyoto dams increased pup-contact, resulting in higher oxytocin levels and lower anxiety-like behavior after weaning, while worsening their adolescent offspring cognitive behavior. Whereas MS in Wistar dams elicited higher quality of pup-directed behavior, increasing brain-derived neurotrophic factor (BDNF) in the offspring, which seems to have prevented a negative impact on cognition. Hypothalamic oxytocin seems to affect the salience of the social environment cues (negatively for Kyoto) leading to different coping strategies. Our findings highlight the importance of contextual and individual factors in the understanding of the oxytocin role in modulating maternal behavior and stress regulatory processes.

Fig. 1. Schematic timeline summarizing the experimental procedures.

PND = Postnatal day.

Fig. 2. Maternal separation impacts maternal behavior of mothers with different vulnerabilities to depression.

A MS increased maternal behavior in both strains. However, Wistar MS dams increased B affiliative behavior (high-quality behaviors) and Kyoto MS dams increased C non-affiliative contact time. D Wistar showed higher weight gain than Kyoto strain, but it was not impacted by MS. Data were analyzed by two-way ANOVA (2 × 2 factorial design) with Bonferroni comparisons, except for D which was analyzed with a linear mixed model (strain and MS as fixed effects and litter as random effect). E Latency to first contact was higher in the Kyoto strain. A, B, and C were data arcsine transformed. E, F, and G data were log-transformed. n = 8 (controls Wistar and Kyoto); n = 7 (MS Wistar and Kyoto, except for PRT: n = 6 for MS Wistar). Results are expressed as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001. MS = maternal separation; PND = postnatal day.

Fig. 3. Maternal separation affects mothers’ emotional behavior.

During MS period (using EPM), A, B no differences were found regarding anxiety. MS increased D Kyoto total arm entries and increased Wistar E number of rearing and F number of head dipping. G example of Wistar and Kyoto control groups tracking in OF (Smart v3.0, Panlab, Barcelona, Spain; https://www.panlab.com/en/). After weaning (using OF), MS increased Kyoto H time in center area and J number of rearing, indicating decreased anxiety levels. FST confirmed Kyoto depressive profile since Kyoto strain showed higher time K immobility and lower L climbing comparing with Wistar. MS did not impact depression-like behaviors. Data were analyzed by two-way ANOVA (2 × 2 factorial design) with Bonferroni comparations. A, C, H, K, L Data were arcsine transformed. B, D, E, F Data were log-transformed. I, J Data were square-root transformed. n = 8 (control and MS Wistar); n = 7 (control and MS Kyoto). Results are expressed as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001. MS = maternal separation.

Fig. 4. Maternal separation disturbed mothers’ neurobehavior.

After weaning, MS increased Kyoto A oxytocin levels, in agreement with the decrease of anxiety observed. E Kyoto showed high levels of TNF than Wistar strain. Quantification of A oxytocin (pg/ml) levels in hypothalamus by ELISA; OXTR in B PFC, C hippocampus, and D amygdala by Western blotting; E TNF (pg/ml), F IL-6 (pg/ml), and (G) IL-1β (pg/ml) levels in hippocampus by ELISA. Data were analyzed by two-way ANOVA (2 × 2 factorial design) with Bonferroni comparisons. A–C, F, G data were log-transformed. n = 5, except for B MS Kyoto, and C, D for control Wistar: n = 4. Results are expressed as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001. MS = maternal separation.

Fig. 5. Maternal separation worsened Kyoto adolescent offspring’s cognitive performance.

MS increased Kyoto A latency to reach the platform, showing learning impairment. Regarding probe day (B–G) MS slightly affected both strains. H MS-induced poor performance in working memory. J VGlut expression was higher in Kyoto strain. K MS led to an increase of BDNF level on Wistar, which may be related to the increase of affiliative behavior received by their mothers. M MS altered IL1-β levels on Kyoto that can be associated with cognitive impairments presented. Quantification of I PSD95 and J VGlut in hippocampus by Western blotting; K BDNF (pg/ml), L IL-6 (pg/ml), and M IL-1β (pg/ml) levels in hippocampus by ELISA. Generalized linear models with 3 factors (strain: Wistar/Kyoto, MS: control/MS, and sex: male/female) and their interactions were used. Time was added as a fourth factor for learning and working memory results. A, F–M used models with normal distribution and link function = identity, C–E used Poisson distribution with link function = log. A, C–G n = 22. I–M n = 16, except for I n = 13 for control Wistar and Kyoto, n = 11 for MS Wistar and Kyoto; K n = 12 for control Wistar and MS Kyoto, L n = 15 for control Wistar and n = 14 for MS Kyoto, M n = 15 for MS Kyoto. Results are expressed as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001. MS = maternal separation; PND = postnatal day.

Fig. 6. Graphical summary of main findings.

Our results showed that mothers with different depression-like vulnerabilities facing maternal stress (induced by MS) adopted different strategies. Although both strains increased the maternal behavior in response to MS, non-depressive-like mothers (Wistar) exhibited a higher quality maternal behavior (affiliative), showing a more adaptive response and more pro-social behavior (more active strategy). On the other hand, depressive-like mothers (Kyoto) displayed a more passive/defensive strategy (non-adaptive response) in response to MS. The higher levels of oxytocin observed in Kyoto mothers that experienced MS seem to be part of the stress-response mechanism, which amplifies the negative perception of the environment (as insecure), leading to a more defensive behavior by keeping the pups very close to them and away from outsiders (out-group anti-social behavior). This increase in oxytocin expression seems to decrease the anxiety-like state in Kyoto dams in response to MS. Furthermore, the quality of maternal behavior observed in non-depressive-like mothers (Wistar) after MS seems to protect the cognitive performance of adolescent rats from the negative effects of MS, leading to an increase of BDNF expression in the offspring hippocampus. On the other hand, depressive-like adolescents showed lower resilience to MS effects, exhibiting worst performance in the cognitive tests and alteration in the IL-1β levels.