Information source and search strategy

We conducted a systematic scoping review to outline key (patho-) physiological mechanisms, examine different health outcomes, identify evidence gaps, and review different types of evidence [14]. A systematic scoping review follows the steps of a systematic review but with a broader scope [14]. We followed PRISMA for systematic reviews [15] and Joanna Briggs Institute for systematic scoping reviews guidelines [14].

EMBASE and MEDLINE via PubMed databases were systematically searched to identify studies published from January 2000 to August 2020. We followed PRISMA guidelines for systematic reviews (S1 Appendix). Search terms were: (cortisol) OR (glucocorticoids) OR (stress) AND (fetal development) OR (prenatal) OR (perinatal programming) AND (cardiovascular disease) OR (metabolic disease) as Medical Subject Headings and Embase Subject Headings terms and title/abstract terms (S2 Appendix). No protocol has been published. Studies were selected by two independent reviewers. First, we searched databases, eliminated duplicates, and selected publications by screening titles and abstracts. Second, we assessed studies with full-text for eligibility. Third, to identify further relevant evidence, reference lists of included studies were manually searched.

Eligibility criteria

To increase our comprehensiveness, we included different study designs. Articles included were literature reviews and reports, interventional studies (randomized controlled trials, non-randomized controlled clinical trials), observational studies (cohort studies), and experimental studies. We included peer-reviewed articles with full-text published over the past 20 years (January 2000 until August 2020), and human and animal model studies in English and German. Eligible for review were publications examining the impact of maternal “stress” on in utero programming of adverse cardio-metabolic outcomes. As already mentioned, “stress” was defined from a pathophysiological point of view as a condition of threatened homeostasis caused by intrinsic or extrinsic stressors and redressed by a range of physiologic and behavioral responses that restore the optimal balance [3, 4]. We included environmental, physiological, and emotional stressors. Conference abstracts, letter, notes, and comments were excluded.

Data extraction and synthesis

Study characteristics (authors, title, year of publication, design, objective), methods (e.g. intervention details, sample, patient characteristics), outcomes, main findings, and conclusions were extracted. Studies were grouped according to their key findings (mechanisms and outcomes) and our goals: (patho-) physiological mechanisms in perinatal cardio-metabolic programming (1), and the impact on short and long-term cardio-metabolic outcomes in offspring (2).

In accordance with the guidelines for (systematic) scoping reviews, we aimed at providing an overview of the existing evidence and reviewing different types of evidence regardless of quality [14]. Therefore, a formal quality assessment was not performed.

Overview

Search identified n = 2.634 citations. After removing duplicates, we screened n = 2.439 titles/abstracts. Excluding n = 2.388 ineligible articles based on inclusion and exclusion criteria, n = 51 publications remained. After removing n = 10 unsuitable full-text articles and identifying n = 4 studies [16–19] via manual research of reference lists, we finally included n = 45 publications in our synthesis. The search and selection process is displayed in PRISMA flow diagram (Fig 1). We identified n = 16 reviews, n = 12 cohort studies, n = 11 experimental animal model studies, n = 2 non-randomized controlled clinical trials, n = 1 randomized controlled trial, and n = 3 reports. An overview of the studies is provided in S3 Appendix.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. PRISMA flow diagram.

https://doi.org/10.1371/journal.pone.0245386.g001

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 [20].

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 [8]. 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 [11]. 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 [11]. 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 [7].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2.

https://doi.org/10.1371/journal.pone.0245386.g002

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 [6]. 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 [2]. In this way prenatal stress/excess glucocorticoid exposure links fetal development to adverse adult health outcomes [27].

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 [22]. Hypersecretion of cortisol can permanently modify HPA function, causing an increased secretion of cortisol which in turn impacts development of the fetal HPA axis [28]. 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 [29]. 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 [25]. 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 [7]. 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 [35]. 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 [34].

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 [36]. 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 [37]. 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 [38]. 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 [31].

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 [34]. 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 [42]. Glucocorticoids stimulating PGC-1α and inhibiting Pdx-1 gene expression might lead to β-cells dysfunction [43]. 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 [43].

Adverse cardio-metabolic outcomes induced by maternal prenatal “stress”

Maternal stress that increases glucocorticoids is a known contributor for the development of obesity in the offspring [9, 18, 44]. Hence, glucocorticoid physiology has a potential impact on multiple targets of perinatal programming related to body composition and obesity risk [9].

Balasubramanian et al. revealed in an animal model study that prenatal maternal stress, that increases glucocorticoids, in combination with a high-fat diet increases the offspring’s risk for obesity in adulthood [44]. Moreover, prenatal stress (arising from socioeconomic or psychosocial factors that activate the HPA axis causing hypersecretion of cortisol) and a postnatal diet high in fat increases not only the offspring’s susceptibility to diet-induced obesity but also leads to secondary adverse metabolic consequences. Male and female offspring affected by hypersecretion of cortisol and a high-fat diet are similarly showing hyperleptinemia and hyperinsulinemia at weaning [45]. Moreover, several transgenic mouse models were developed to study the influence of increased maternal stress on adverse offspring health outcomes. The corticotropin-releasing factor receptor-2 deficient mouse is such a model of heightened stress sensitivity [46]. Chronic stress exposure (corticosterone levels) in these mice is linked to an increased intake of palatable foods. During exposure to chronic variable stress, transgenic mice consumed a greater proportion of their calories in form of a high-fat diet compared to non-stressed controls [47]. In transgenic mice selectively overexpressing 11β-HSD1 in adipose tissue, excess glucocorticoid exposure to the fetus produces visceral obesity. Moreover, these mice exhibited insulin resistance and diabetes [17].

In humans, maternal distress caused by flood exposure (self-reported psychological reaction to the flood) during an early phase of pregnancy was associated with a greater increase of the child’s BMI from age 2.5 to 4, as well as total adiposity at the age of 2.5. These results suggest that early gestation is a sensitive period for psychological maternal stress having a great impact on the development of childhood adiposity [48]. Furthermore, adolescents exposed to maternal psychosocial stress during intrauterine life consistently exhibit a higher BMI and body fat (%) with concomitant primary insulin resistance, and a lipid profile comparable to the metabolic syndrome [9]. Fetal exposure to glucocorticoids is able to program central adiposity in later life. Maternal corticotropin-releasing hormone levels found in the late second trimester of gestation, predicted child adiposity at 3 years of age. Thus, excess fetal exposure to glucocorticoids programs increased relative adiposity and central obesity in the offspring of prenatally stressed mothers [16].

Alterations in glucose-insulin metabolic function can also be induced by exposing the mother to excess glucocorticoid hormones. It is therefore possible that elevated circulating concentrations of stress hormones may elevate glucose levels in maternal and fetal circulation contributing to the development of insulin resistance [10].

Animal studies suggest that prenatal glucocorticoid exposure, either induced by hormonal treatment or by maternal stress during pregnancy, programs adverse metabolic alterations such as insulin resistance, hyperglycemia, and hyperinsulinemia [49, 50]. The aim of Lesage et al. (2004) was to investigate the consequences of prenatal stress (maternal glucocorticoids) on in utero growth restriction and a following increased risk to develop metabolic disorders [50]. Prenatal maternal stress reduced birth, adrenal and pancreas weight as well as plasma corticosterone and glucose levels in rat fetuses at term. In male rats 24 months of age, prenatal stress induced hyperglycemia and glucose intolerance. Adverse cardio-metabolic programming in the aged offspring could be linked to restricted fetal growth and elevated levels of circulating glucocorticoids [50].

Dancause et al. (2013) measured the effects of physiological prenatal stress caused by the Quebec Ice Storm in 1998 among a subsample of n = 32 exposed adolescents. Severity of stress was positively associated with insulin secretion, an early feature of insulin resistance suggesting that prenatal stress is an independent predictor of metabolic outcomes in adolescence [49]. Besides, physiological prenatal stress was associated with shorter length at birth and development of childhood adiposity which are predictors for an increased risk of cardio-metabolic diseases in adult life [49].

Furthermore, Igosheva et al. showed in an animal model study that stress exposure (a regimen of heat, light and restraint stress) early in development can have long-term effects on arterial blood pressure in adulthood, with potential sex differences [51]. Differences in cardiovascular function between prenatally stressed and control rats were especially evident when animals were challenged by acute restraint stress. In general, the detected effects were more obvious in female than in male animals [51]. In addition, van Dijk et al. revealed that prenatal maternal psychosocial stress predisposes for cardiovascular disease. Multiple psychosocial stressors during pregnancy were associated with higher systolic and diastolic blood pressure and mean arterial pressure in children (age 5–7) [19] and maternal depressive and anxiety symptoms were associated with lower offspring diastolic blood pressure and wider brachial artery diameter (age 10–12) [52].