Mothers’ exposure to stress during pregnancy could have long-term detrimental effects on their children’s diets, and thereby on health conditions related to diet – such as increased levels of obesity and obesity-related diseases – according to new research from Michele Belot, professor in the Cornell University Department of Economics.
“Being exposed to stressful events when pregnant seems to impact the dietary preferences and diet of the children in a negative way, and for reasons that are actually aside from what the mother is eating herself,” says Belot, who has a joint appointment in the School of Industrial and Labor Relations and College of Arts and Sciences.
“So that means that we need to think about how to help pregnant women manage stress in a way that could be beneficial for the mother and also for the child.”
In their paper published in Nutrition, Belot and her co-authors found that higher than average stress during pregnancy is linked with significantly less healthy food preferences for their children, as well as a weaker preference for sour and bitter foods.
“Stress during pregnancy could have long-term detrimental effects on the next generation in terms of a less healthy diet and subsequent health implications associated with these effects, such as higher rates of obesity and obesity-related diseases,” wrote the authors, which include Nicoli Vitt (University of Bristol), Martina Vecchi (Penn State) and Jonathan James (University of Bath).
“As a consequence, we advocate for more research into understanding the sources of maternal stress and the extent to which these can be altered. Prenatal care and preconception counseling could be critical to develop preventive strategies to improve public health.”
For the study, the researchers selected 213 mothers of low socioeconomic status living in the area of Colchester, United Kingdom, with children aged between 2 and 12 years old. Their stress levels during pregnancy were assessed using retrospective self-reporting.
Specifically, they asked whether mothers experienced one or more of the following life events during the pregnancy with their child: death of close family member or close friend, changes or difficulties in their relationship, legal issues, changes or difficulties in their family life, health issues, changes or difficulties in their or their spouse’s employment, financial issues, changes in their habits, other potentially stressful events.
For each stressor, mothers were asked to rank how stressful the event was on a scale of one to 10.
The researchers then examined the healthiness of the diet for each participant’s youngest child. They also studied the child’s preferences for the five basic tastes—sour, salty, umami, bitter and sweet.
Their results found “a sizable negative association between in utero exposure to stress and the healthiness of food preferences and diet of the child.” More specifically, they found in utero exposure to stress to be negatively associated with the child’s liking of sour and bitter foods.
This effect does not appear to be channeled through the mother’s own diet either during or after the pregnancy.
“I’m a great proponent of policies geared towards supporting people with well-being and mental health programs to try and reduce the stress that people are facing on a daily basis,” says Belot. “Stress is not good for many reasons, and it’s something I’ve now written a few papers about.
“Pregnant mothers from low socioeconomic backgrounds might not have a stable workplace providing wonderful mental health and well-being programs. So, it might be beneficial to run them through community based programs.
“In the UK there are already many community and local organizations that do try to support low-income families, so adding mental health and well-being programs could be really beneficial. We should really try to have more ways to help mothers and to support mothers during their pregnancy.”
In general, there are indications that maternal prenatal glucocorticoids play a key role in underlying mechanism in trans-generational programming of cardio-metabolic diseases. From a pathophysiological point of view, the maternal risk factor stress was clearly linked to adverse cardio-metabolic offspring outcomes such as obesity, insulin resistance, diabetes, metabolic syndrome, cardiovascular disease (hypertension), hyperglycemia, restricted fetal growth as well as reduced birth, adrenal, and pancreas weight (Fig 2).
The evaluation of the studies is affected by differences in the nature of the stress applied and its timing during pregnancy. Additionally, the effects of stress vary depending on the sex and age of the offspring .
In humans, prenatal maternal stress can be caused by exposure to both severe stressors and milder forms of psychosocial stress during pregnancy. Milder forms include stress experienced in daily life, but also pregnancy specific anxiety, e.g. the fear of giving birth to a handicapped child, or the mother suffering from prenatal depression .
Studies investigating the influence of maternal stress need to be carefully tested on potential confounding factors, which can cause the reported association to be misleading. Several maternal characteristics are considered to be confounders including age, educational level, ethnicity, tobacco and alcohol consumption .
Findings between studies can also vary considerably depending on the methodological approach used . Some measures of maternal prenatal stress are based on mother’s self-reports, which might be biased by either over- or underreporting . Maternal prenatal stress is converted into physiological signals that are subsequently transmitted to the fetus. By measuring clinical parameters, e.g. the mother’s heart rate or electrical conductance of the skin, it is possible to evaluate the maternal physiological arousal as a parameter of emotionality .
Descriptions of the children’s problems also often rely on a mother’s self-report, which makes it challenging to know whether the results simply show that anxious mothers characterize their children as being more difficult than non-anxious mothers do . If human studies detect an association between physiological maternal prenatal stress and an adverse health outcome, the association may be mainly confounded by postnatal stress. Moreover, it can potentially be a result of shared genes or childrearing practices .
Since stress-related outcomes are likely dependent on a critical time-point during pregnancy, the timing of stress assessment is also an important factor in finding potential effects of prenatal stress on later disease development [28, 48]. Development of organs and tissues occurs over the course of gestation. Specific outcomes of stress-related perinatal programming are both dependent on the timing of stress exposure and the developmental stage of each individual organ system [34, 46].
Thus, vulnerable periods differ for the diverse outcomes because of different developmental stages of the various organs . In that way, a mother’s stress experience during sensitive periods of cell and tissue growth may permanently alter the baby’s tissue structure and function, causing far-reaching effects which may persist throughout life .
Examples of extreme stressors are the exposure to (natural) disasters, e.g. earthquakes [22, 48]. These kind of disasters provide unique opportunities to investigate the effects of prenatal stress on offspring’s health outcomes independently from potential confounding factors. These disasters randomly affect women independent of their maternal and socioeconomic characteristics. Moreover, since the dates of such events are well-known, timing of stress exposure during pregnancy can be exactly determined .
Animal models are widely used to investigate the effect of maternal prenatal stress on a variety of different health outcomes in later life. A great advantage of laboratory animals is that stress responses can be reliably induced by a wide range of experimental methods. By exposing pregnant dams to stressful events (e.g. restraint) psychological stress is transferred to the fetus and thus produces effects on the offspring .
Even though the use of animal models provides benefits, substantial physiological differences between humans and animal models have to be taken into account [2, 33]. For example, the placental 11β-HSD2 expression declines towards term in mice, whereas it rises in humans. On this account the mouse might not be an ideal model of this biology . Furthermore, the type of prenatal stress applied to animals is different to stressors affecting human pregnancy .
In animal studies, stressors are external events that can be controlled in terms of duration, frequency, and intensity [20, 45, 46], so that confounding factors can be successfully avoided . Likewise, animal studies can be designed to make sure that the origin of stress-related effects is prenatal rather than postnatal, it is for example possible to cross-foster prenatally stressed pups to control dams after birth .
Compared to that, women who are psychologically stressed before pregnancy are more likely to be stressed postnatal too. Thus, it is difficult to distinguish social influences after birth from pregnancy effects that are transmitted biologically . It also remains to be seen if an “optimal stress level” may beneficially affect offspring`s health .
It should also be mentioned that other uterine stress factors such as chronic stress, for example exposure to environmental contaminants, may or may not lead to an increased HPA axis, which can lead to cardiometabolic changes in the offspring .
Other systematic reviews and meta-analyzes support our findings. For instance, Lamichhane et al. (2019)  reported that there might be a direct association between psychological prenatal maternal stress and obesity in offspring. According to a meta-analysis by Burgueño et al. (2019) , prenatal maternal stress (increased release of glucocorticoids by activation of the HPA axis) was associated with increased BMI of their offspring. A meta-analysis by Lima et al. (2018)  found a significant association between antenatal stress exposure (adversely affected and hyperactivated HPA axis and sympathetic nervous system) and increased rates of low birth weight. Other reviews reported on the key role of glucocorticoids in fetal development and programming disease later in adult life [57, 58].
The key role of the hypothalamic-pituitary-adrenotrophic (HPA) axis in trans-generational programming of cardio-metabolic diseases
In a nutshell, the stimulation of maternal hypothalamic-pituitary-adrenotrophic (HPA) axis might play a key role modifying in utero milieu leading to cardio-metabolic diseases in the offspring later in life. Since there are no direct neural connections between the mother and the fetus, e.g. maternal psychological functioning has to be translated into physiological effects. At least three mechanisms are postulated how maternal psychological stress might have a direct impact on the fetus: alteration in maternal behaviors (e.g., substance abuse), reduction in blood flow leading to fetal deprivation of oxygen and nutrients, and transport of stress-related neurohormones from the mother to the fetus through the placenta .
During gestation the placenta functions as an exchange site between the mother and the developing fetus [11, 21]. This unique organ is supposed to modulate and filter signals from the maternal to the fetal milieu . From a neuroendocrine and epigenetic point of view, there is evidence suggesting that the placenta is highly susceptible to maternal distress and thereby serves as a key mechanistic link between maternal distress and adverse offspring outcomes. Prenatal maternal distress may act through the placenta and in this way affect the growth and development of the fetus . Maternal cortisol may be passed to the fetus through the placenta, generating a cascade of effects with potential impact on fetal development and offspring’s future health [2, 22–24].
High levels of maternal glucocorticoids may cause altered gene expression profiles in placental tissues (Fig 2). One of these target genes is 11β-HSD2 encoding the enzyme 11β-hydroxysteroid dehydrogenase type 2. 11β–HSD2 is highly expressed in the placenta and functions as a protective feto-placental barrier to maternal glucocorticoids [8, 11]. Although glucocorticoids are able to cross the placenta, their levels are significantly lower in the fetus than in the mother because 11β–HSD2 can inactivate glucocorticoids. It catalyzes the rapid conversion of active cortisol to biologically inactive cortisone and thereby protects glucocorticoid sensitive tissues in the fetus from high levels of circulating maternal stress hormones [7, 11]. 10–20% of maternal glucocorticoids reach the fetus in its intact form indicating that this barrier is apparently incomplete. Compared to the fetus, maternal glucocorticoid levels are much higher. Only modest changes in the expression of placental 11β–HSD2 may profoundly influence fetal glucocorticoid exposure . It is additionally postulated that reduced levels of placental 11β–HSD2 enable greater glucocorticoid signaling within the placenta itself, which can indirectly impact fetal development by changing placental function .
In mammals, glucocorticoids play a crucial role during fetal development. They are essential for growth regulation, tissue development and maturation of various organs. This action is critical to prepare the fetus for extra-uterine existence [6, 8]. Glucocorticoids exert their most potent effects during late gestation by stimulating surfactant production by the lung. For this reason, synthetic glucocorticoids are widely used to treat women at risk of preterm labor where immaturity of the lung highly impacts neonatal viability [6, 25, 26]. Antenatal glucocorticoid therapy can reduce birth weight which is associated with numerous long-term effects on the offspring’s health including an increased risk of cardio-metabolic diseases in adult life [6, 26].
Since 11β-HSD2 allows passage of a small amount of active glucocorticoids from the mother to the fetus, elevations in maternal cortisol can still double the amount of intact cortisol reaching the fetus [8, 22, 27]. Excess glucocorticoids reduce placental 11β-HSD2 expression and/or activity leading to an increased transport of active cortisol across the placenta. This is correlated with a lower birth weight, higher blood pressure and glucose intolerance in later life . As an endproduct, one of the body’s major stress responsive system, glucocorticoids have been proposed to play a crucial role in stress-related perinatal programming of cardiovascular and metabolic disorders . In this way prenatal stress/excess glucocorticoid exposure links fetal development to adverse adult health outcomes .
The HPA axis is a key part of the neuroendocrine stress response. In humans and many other species, HPA axis is activated in reaction to physical as well as psychological stress . Hypersecretion of cortisol can permanently modify HPA function, causing an increased secretion of cortisol which in turn impacts development of the fetal HPA axis . Evidence comes from studies in humans where high levels of prenatal psychosocial stress during pregnancy were associated with altered HPA axis functioning in adult life. Psychosocial stress was defined as multi-component construct that involves negative life events, appraisal of the stress, and symptoms such as anxiety. Impaired regulation of the offspring’s HPA axis regulation is linked to several adverse health outcomes including the metabolic syndrome . Besides chronic stress conditions, maternal exposure to synthetic glucocorticoids and nutrient restriction can also significantly influence HPA function, leading to increased fetal exposure to glucocorticoids. Both is linked to cardiovascular diseases, insulin resistance and diabetes in later life [30, 31]. Birth weight of rats was reduced after prenatal exposure to the synthetic glucocorticoid dexamethasone or the 11β-HSD2 inhibitor carbenoxolone. During adulthood, the offspring display increased HPA axis activity, hypertension and hyperglycaemia . Additionally, rodent experiments revealed that programming of HPA function can be passed to subsequent generations. Treatment of a pregnant rat with the synthetic glucocorticoid dexamethasone reduced offspring’s birthweight and glucose tolerance. These effects were transmitted to the grandchild generation without further exposure of the F1 generation [6, 32, 33].
Glucocorticoids exert their effects on the developing fetus by binding to glucocorticoid receptors (GRs) which subsequently act as transcription factors to alter gene expression levels . GRs are expressed in most fetal tissues, including a high expression level in the placenta [27, 34] of the fetus to either high levels of stress or glucocorticoids can permanently affect GR expression . In both animal and human studies, excessive production of endogenous glucocorticoids as well as their exogenous administration have been shown to cause effects on many systems such as diabetogenic and hypertensive effects .
Prenatal glucocorticoid administration to women having an increased risk of preterm delivery is associated with higher systolic and diastolic blood pressures in adolescents 14 years of age . Antenatal glucocorticoid therapy is also linked to higher plasma insulin concentrations in subjects 30 years of age, which might result in insulin resistance later in life . A prospective cohort study by Goedhart et al. analyzing n = 2.810 pregnant women indicated that high maternal cortisol levels measured in the morning were negatively associated to offspring birth weight and positively to being born small for gestational age . In the Hertfordshire study, high cortisol levels in a cohort of elderly men and women were related to low recorded birth weights and a variety of cardiovascular risk factors, e.g. increased plasma glucose and triglyceride levels and increased systolic blood pressure [26, 39]
Furthermore, studies in several animal models have shown that excessive exposure to prenatal glucocorticoid excess reduces birth weight and subsequently causes hypertension and hyperglycemia in later life of the offspring [12, 34]. In rats, the correlation between low birth weight and adulthood diseases such as high blood pressure could be related to an adverse glucocorticoid environment in utero .
Administration of cortisol to sheep early in pregnancy elevates blood pressure in adult offspring [40, 41]. In rats, prenatal exposure to glucocorticoids during the last week of pregnancy was able to produce permanent effects by causing long-lasting changes in blood glucose and insulin levels . Moreover, glucocorticoid signaling is important in fetal β-cell development. Increased glucocorticoids have shown to be associated with a decreased β-cell mass in rat pups . Glucocorticoids stimulating PGC-1α and inhibiting Pdx-1 gene expression might lead to β-cells dysfunction . In addition, glucocorticoids may influence obesity by blocking AMP-activated protein kinase and (in-)directly activating SREBP-1c gene expression. They may influence hypertension by imbalancing vasoactive factors .
reference link : https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0245386#sec015
Original Research: Closed access.
“Maternal stress during pregnancy and children’s diet: Evidence from a population of low socioeconomic status” by Nicolai Vitt et al. Nutrition