Stress, brains and bairns: Reviews from the 4th International Conference on the Parental Brain

The 4th International Conference on the Parental Brain was held in Edinburgh from 1st to 4th September 2010. The aim of the meeting was to discuss the recent developments in the neurobiological and behavioural adaptations occurring in preparation for parenthood, the interactions between parents and new offspring (before and after birth) and the impact of sub-optimal conditions (e.g. maternal stress) on offspring quality of life. The meeting had four main themes: maternal adaptations in pregnancy and lactation, parental behaviour, early-life programming and perinatal influences on mental health. The latter two topics will be the focus of this themed issue of Stress. The contributions include a collection of review articles and original research reports from delegates who presented their work at the 4th International Conference on the Parental Brain and encompass research from both basic and clinical perspectives.

The perinatal period is a time of enhanced plasticity and as such adverse experiences in early life can induce long-lasting or permanent changes in the physiology and behaviour of the offspring. It is becoming increasingly clear that the way in which an individual responds to stress is dependent upon several factors, including experience in early life, and, moreover, that an organism's phenotype depends not solely on their inherited genes or environmental influences but rather on an interaction between the genes and the environment. Postnatal adversity or stress is often associated with a hyperactive hypothalamo–pituitary–adrenal (HPA) axis, causing elevated circulating glucocorticoids throughout life. It is the deleterious consequence of the raised glucocorticoids that has been hypothesised to underpin observed abnormalities in mood and cognition in later life. The review by Murgatroyd and Spengler (2011) describes the role of elevated arginine vasopressin (AVP) in the paraventricular nucleus of the hypothalamus, which appears causal to HPA axis hyperactivity in mice subjected to postnatal stress. In an elegant series of experiments, they describe changes in epigenetic marks on the AVP promoter enhancer, which underpin the altered expression levels of AVP, namely the promoter region of the AVP gene is less methylated in mice that have experienced early-life adversity (separation from mother) resulting from decreased binding of MeCP2 to this region (MeCP2 binding couples DNA methylation with transcriptional repression). The work of these authors is reviewed alongside other studies on glucocorticoid programming and the potential role of the epigenome mediating the long-lasting programming of the HPA axis and the behaviour in both rodent models and humans subjected to developmental stress.

Although the concept of developmental origins of adult disease was generated from human epidemiological evidence, the mechanisms underpinning effects of stress and glucocorticoid programming have mostly been studied in animal models. However, recently, Raikkonen and colleagues (2011) have taken advantage of the popularity of eating liquorice in Finnish populations to address glucocorticoid programming in humans. Liquorice is a potent inhibitor of the glucocorticoid-metabolising enzyme 11β-hydroxysteroid dehydrogenase type 2 (11β-HSD2), which is highly expressed in the placenta, inactivates cortisol and is an essential barrier protecting the developing foetus from high maternal glucocorticoids. Studies in rodents have shown that if this barrier is inhibited or knocked out, birth weights are reduced and offspring show increased anxiety- and depressive-like symptoms as adults. Here, Raikkonen et al. review a series of studies which show that children born to women who eat significant quantities of liquorice throughout pregnancy have long-term abnormalities in behaviour and physiology, highlighting the importance of 11β-HSD2 in glucocorticoid programming in human populations.

Following a review of the literature on the consequences of prenatal stress on offspring behaviour in human populations, Weinstock (2011) concludes that the most vulnerable period for the stressful event to take place is within the window of 12–20 weeks of pregnancy. Hence modelling this equivalent period in rats, Weinstock and colleagues have shown that prenatal stress adversely affects mood and cognition in offspring in a gender-specific manner, which is dependent on the type of stressor, the timing of presentation and the timing of assessment of the behaviour (juvenile or adult). Intriguingly, the behavioural results correlate well with size or dendritic arborisation within the hippocampus, but the mechanisms generating these structural changes (and others such as altered hippocampal neurogenesis) still need to be determined.

While there is a consensus that early-life adversity is generally accompanied by an altered reactivity of the HPA axis throughout life in most species, there is conflicting data on the initiation of this increase in small children. Hunter et al. (2011) carried out a systematic review of the literature concerning baseline and stressed cortisol concentrations in children aged 0–5 years who have been exposed to prenatal or postnatal adversity. As there are a plethora of types of adversity, differences in their timing and the age at which cortisol samples were collected, it is not surprising that it is difficult to tease out a common thesis. However, a general conclusion can be made that adversity does generate altered HPA reactivity in young children but the direction of the change is age dependent: neonates may exhibit decreased reactivity, often no differential effects are observed around 2 months of age, but from approximately 12 months of age onwards there is hyperactivity of the HPA axis (increased cortisol) when compared with children not exposed to adverse stressors. The authors discuss the limitations of such studies in children but it is essential to pursue ways to overcome this and provide crucial knowledge on the trajectory of adverse programming of the HPA axis in children.

The first original research report on this issue is by Hellgren et al. (2011). In this study the authors have provided evidence that measuring skin conductance in pregnant women given a cold pressor test may provide a non-invasive test for assessing sympathetic activity, which in turn will predict the timing of labour initiation. The sympathetic response to stress continuously decreases towards labour, hence this simple test could potentially be applied to obtain unique information on the proximity of labour.

In rodents, non-human primates and humans, exposure to prenatal stress typically results in increased fearful or anxiety-like behaviour in the offspring, though often the effects can be sex dependent. The central actions of corticotropin-releasing hormone (CRH) play a key role in mediating anxious behavioural responses, in particular CRH expressing neurones in the central nucleus of the amygdala (CeA) are considered especially important. Two different receptors mediate the actions of CRH. These receptors have generally opposing roles, such that activation of the CRH type 1 receptor (CRH-R1), by its principal ligand (CRH), typically increases anxious behaviour, whereas activation of the CRH type 2 receptor (CRH-R2) by its main ligands (urocortins II and III) dampens anxiety-type behaviour. In this issue, Brunton et al. (2011) examine whether differences in CRH-R1 and CRH-R2 mRNA expression in the amygdaloid complex may underlie sex differences in anxiety-like behaviour in rats born to mothers exposed to repeated social stress during the last week of pregnancy. In prenatally stressed male rats that show heightened anxiety-like behaviour, they report increased mRNA expression for the ‘pro-anxiogenic’ receptor, CRH-R1, in the CeA and basolateral amygdala (BLA), concomitant with a reduction in mRNA expression for the so-called pro-anxiolytic receptor, CRH-R2, in the basomedial amygdala. Conversely, CRH-R1 mRNA expression in the CeA and the BLA is unaltered and CRH-R2 mRNA is increased in prenatally stressed female offspring, which do not show anxiety-like behaviour under this maternal stress paradigm. The authors propose that differential expression of the type 1 and type 2 CRH receptors in the amygdaloid complex may explain the sex differences in anxiety-type behaviour in these prenatally stressed offspring.

The neuroactive metabolite of progesterone, allopregnanolone, plays an important role in central nervous system development, promoting neuronal growth in early development. Moreover, allopregnanolone is neuroprotective, promotes cognitive function and has ‘anti-stress’ and ‘anti-anxiety’ actions. Gestational stress has been shown to reduce allopregnanolone levels in the maternal hippocampus, and blocking allopregnanolone production during late pregnancy results in reduced hippocampal levels of allopregnanolone and impaired cognitive performance in the offspring. In this issue, Paris et al. (2011) investigate the effect of immune challenge (infection may account for up to 40% of preterm births in women) during late pregnancy in rats on birth outcomes, cognition and anxiety-type behaviour in the offspring and steroid levels in the offspring brain. They report that immune challenge during pregnancy reduced gestational length and the number of viable pups born. Moreover, both male and female prenatally stressed offspring demonstrated deficits in cognitive performance which was associated with perturbed central progesterone metabolism and reduced oestradiol levels in the hippocampus, medial prefrontal cortex and hypothalamus. Whether altered steroid content/production and/or metabolism in the prenatally stressed offspring brain directly underlies deficits in cognition remains unclear, but further investigation in this area is warranted.

Studies on humans indicate that maternal anxiety can also have persistent negative influences on infant neurodevelopment, including delayed motor development, poorer cognitive performance and attention deficits. Blair et al. (2011) have explored the impact of prenatal maternal state anxiety and pregnancy-specific anxiety (fears/worries about the pregnancy, delivery and health of her baby) on child temperament. They report that higher maternal pregnancy-specific anxiety in early pregnancy (13–17 weeks) is associated with increased negative temperament (such as sadness and frustration) in the children at 2 years of age, while generalised measures of state anxiety are not. These data suggest that during the prenatal period, signals of anxiety specific to the pregnancy are transmitted to the foetus and these have long-term consequences for the foetus. This study has far-reaching implications given that children assessed as having a difficult temperament in early childhood are more likely to develop behavioural problems and psychiatric disorders later in life.

The study by Buss et al. (2011) investigates whether associations exist between maternal affective state during pregnancy (pregnancy-specific anxiety, general state anxiety and depression) and cognitive performance in pre-pubertal children (age 6–9 years). They report that pregnancy-specific anxiety during gestation is a strong predictor of executive function in the children. High levels of pregnancy-specific anxiety in the mothers were associated with lower visuospatial working memory performance in both boys and girls, while inhibitory control (the ability to resolve conflicts when competing information is present) was impaired only in girls. The underlying mechanisms involved are not known, though may involve impaired neurodevelopment of prefrontal cortical structures since maternal pregnancy-specific anxiety is associated with volume reductions in grey matter in the prefrontal cortex in these children, consistent with studies from rats which report abnormal development (reduced neurotrophin, axonal outgrowth and synaptogenesis) of the prefrontal cortex.

Postpartum mood can also affect offspring development. For example, reports suggest that the children of mothers suffering postpartum depression may have impaired emotional and cognitive development. There is evidence that social conflict or exposure to chronic stress may play a role in the development of depression including postpartum depression. In this issue, Nephew and Bridges (2011) have employed a chronic social stress model in lactating rats to evaluate the effects on maternal behaviour and development of the offspring. They report that chronic social stress during lactation significantly increases self-grooming (a behavioural indicator of anxiety) in the dams, reduces maternal care and has long-lasting effects on pup growth. While further assessments of this paradigm on anxiety- and depression-like behaviour are necessary, this model may prove to be a useful tool for studying the effects of stress-induced alterations in mood in the postpartum period on the mother and her offspring.

Maternal care is an important neurodevelopmental modulator and as such maternal deprivation has long-lasting or even permanent consequences for brain function and behaviour of the offspring. In particular, maternal deprivation or separation has been associated with increased stress reactivity. The study by Lomanowska et al. (2011) examines the effects of artificial rearing (pups are reared without a mother and fed via a feeding tube) on basal and stress-induced corticosterone levels. They report that artificial rearing does not alter basal corticosterone secretion, but does result in increased responsiveness of the HPA axis to stressful stimuli in the first few postnatal days; however, this effect does not persist. The authors conclude that the long-term effects of artificial rearing are not mediated by sustained levels of elevated corticosterone and are likely to instead be mediated by other factors, such as the loss of tactile stimulation by maternal licking and grooming.

The studies described in this special issue advance our understanding of the negative impacts of stress during critical periods of development on future health. Given the wider socio-economic implications these adverse consequences on health have for humans, further research in this area seems critical. New developments in this field will be revealed at the next meeting in the Parental Brain series to be held in Regensburg, Germany, in 2013 (http://www.uni-r.de/ParentalBrain2013/).

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.

Notes

^*Bairn [bεən (Scot) bern]: n Scots, a child or baby.