Developmental origin of health and disease-Evidence and time for action.

Abstract

The hypothesis that environmental exposures in early life affect later health and risk for disease has been conceptualized as the Developmental Origin of Health and Disease (DOHAD).1 Several epidemiological studies linking nutritional status in childhood to later health gave rise to this hypothesis, which later was expanded to include intrauterine nutritional status and growth. The field is getting more attention as potential underlying physiological mechanisms are emerging. Epigenetics explains how gene expression may be altered without altering the DNA sequence, and provides a mechanistic explanation for the interaction between gene expression and environmental cues. Epigenetic modifications in the developing fetus may therefore link maternal health, including the pre-conception period, to lifetime health of the offspring. This has underscored the importance of healthier lifestyle, adequate nutrition, and physical and mental well-being among adolescents and women of fertile age.2 However, interventions have been difficult to evaluate in humans due to several inevitable confounding factors. Furthermore, developmental physiology is a complex field to explore, and caution is required, as interventions in pregnant women or adolescents may have future implications for the offspring that are unforeseen. The current evidence of DOHAD is based on combined data from several types of studies. The main body of evidence is generated from retrospective studies investigating different time periods, but not the entire lifespan from conception to old age. Perinatal outcomes are related to maternal health before or during pregnancy. Several such studies suggest an association between perinatal outcomes and the development of non-communicable diseases, such as diabetes, cancer and cardiovascular disease.1, 3 Importantly, studies of siblings and of abrupt changes in nutritional access link maternal health and nutritional status before or during pregnancy to the life course health of the offspring.4 On the other hand, prospective studies investigating the effect of the intrauterine environment for the offspring usually include pregnant women in the first or second trimester of pregnancy and report only early neonatal outcomes. The need for prospective longitudinal data across generations is being met by recent studies that recruit women before pregnancy and follow the offspring up to adolescence and reproductive age. Experimental data aiming to provide mechanistic understanding of phenotypic effects on fetal organ structure, growth and function suggest a causal effect of maternal diet, and metabolic or other types of stress on offspring outcomes in later life. Several studies have demonstrated that the composition and content of maternal diet (energy density, content of protein, antioxidant or methyl donors, etc.) may potentially have epigenetic effects on fetal phenotype.3 Experimental animal studies using in vitro fertilization and healthy surrogate mothers demonstrated altered phenotype in offspring when the gametes originated from nutritionally challenged mice.5 Moreover, human studies linking epigenetic traits to fetal exposure, searching for mechanistic links to later risk for disease, are emerging. Based on these observations, several large and well-designed interventional studies have been launched, aiming to improve maternal metabolic health during pregnancy, perinatal outcome and later risk of disease in the offspring. At first glance, the results have been somewhat disappointing, even when the interventions resulted in marked improvement of the maternal metabolic state.6 There are several potential explanations for this, illustrating the complexity of this intergenerational risk cycle. First, most of the studies started interventions well into the second trimester of pregnancy because approximately half of all pregnancies are unplanned and women of reproductive age are not frequent users of health services and are therefore less accessible through classical study recruitment strategies. Lifestyle changes are difficult to implement and it takes time to achieve planned goals. Thus, interventions in pregnancy may have been started too late to have a proper impact. Second, gametogenesis and early stages of embryogenesis are periods particularly susceptible for epigenetic alterations. In addition, maternal metabolic factors in early pregnancy probably affect placentation and thus placental function throughout pregnancy. In sum, the effect of an intervention depends both on the nature and the timing of the exposure. This is reflected in human observational studies demonstrating different effects of famine during early or late stages in pregnancy.4 Moreover, it is established that epigenetic patterns are tissue- and cell-specific and epigenetic changes in humans are not easily monitored due to the inaccessibility of adequate samples; in addition, data from experimental studies on animal are not directly transferable to humans. To complicate the picture even more, epigenetic modifications may be time-limited and reversible. Consequently, interventions may have other outcomes than expected or monitored and physiological conditioning may not necessarily result in measurable outcomes at birth or in early childhood. Indeed, follow up of children after interventional studies indicate that there might be effects on metabolic measurements in later childhood.7 Inversely, an outcome may be seemingly beneficial at birth but associated with later risk of disease. This is illustrated by the effects of bariatric surgery, where this exposure might have both positive (decreased risk of gestational diabetes and having large for gestational age fetus) and negative (shorter gestation and increased risk of having small-for-gestational age fetus) effects on neonatal outcomes.8 Interventional studies of micronutrient status and folate supplementation in human pregnancy revealed some important issues.9 First, an exposure (folate) may be beneficial at birth (higher birthweight in a population with frequent small for gestational age neonates) but may confer long-term risk (insulin resistance) in childhood. Second, it may be beneficial to some outcomes (neural tube defects), but not for others (insulin resistance). Third, the results are diverging depending on the population studied. Fourth, the effects may be influenced by other factors (such as vitamin A, homocysteine and vitamin B12 status). Fifth, the effects may be dependent on timing and dosage, eg both high and low folate status may confer risks. Finally, following the DOHAD concept, the time of observation from exposure to outcome is long and most studies will struggle to differentiate intrauterine effects from residual confounding from life itself. We inherit epigenetic marks not just from our life in utero, or early childhood, but also from the genome and postnatal environment. A mismatch between the pre- and postnatal environment is particularly linked to later disease. The epigenome is not a constant trait, but namely a way for cells, tissues and metabolic systems to adjust the genetic expression throughout life. In fact, the epigenome between monozygotic twins is increasingly different with age and diverse life exposures.3 Interventional generational studies have been launched and will provide us with important insight in the years to come.2 In summary, there are challenges that hamper the opportunities to study the concept of DOHAD mechanistically, and conclusions from interventional studies may not be transferable between different populations. Nevertheless, the sum of evidence, although fragmented, supports the DOHAD concept. The ideal studies needed to advise and survey the consequences of potential interventions must span generations and record even unforeseen outcomes. However, as a parallel to the global climate crisis, it is not a wise strategy to wait for more evidence before acting. Rather, the focus should be on the simplest and safest solutions. Targeted efforts to ensure good education, balanced nutrition and a healthy lifestyle in childhood and adolescence should be governmental priorities to promote healthy reproductive life and long-term health of the next generation.