Keywords: Developmental Origins of Health and Disease (DOHaD), Fetal Programming, Maternal Nutrition, Maternal Exercise, DNA Methylation, Epigenetics, One-Carbon Metabolism, Gestational Diabetes Mellitus (GDM), Childhood Metabolic Syndrome, Placental Function, Insulin Resistance
Reviewed by A1C Medical Team
Rethinking Origins: From Genes to Programming
Childhood metabolic syndrome and cardiometabolic diseases are becoming more common with alarming risk of future increase in number of cases. The first possible reason is due to increased cases of Inherited or inborn metabolic disorders (IMDs). IMDS have become a representative on a large and heterogenous group of rare diseases, particularly in children. Heterogeneity within this group of scarce disorders is attributed to variants in genes, which are also responsible in encoding certain enzymes or cofactors. Consequently, it may lead to dysfunction of at least one metabolic pathway; leading to impairment of multiple organs since infancy and formative years. Several metabolic defects which may occur in IMDs include defects in metabolism of amino acids, fatty acids, organic acids, urea cycles, metal, lysosomes, steroids, carbohydrates, and so on. Most of these metabolic errors are monogenic disorders; predominantly autosomal recessive of X-linked. A wide range of genetic heterogeneity and clinical phenotypes can be found in IMDs Clinical manifestations depend on the main endocrine systems and organs involved (Table 1); thus, alleged IMDs must be screened particularly in pediatric patients with multiple endocrinopathies.1 Another study conducted in 232,561 newborns in Shanghai also showed that the total incidence rate of inborn error of metabolism was 1:4153 with the initial positive rate was 0.66%.2 These findings highlighted the importance in addressing the possible reasons behind the sharp increase in IMDs over time among younger populations; especially to obtain more meaningful and favorable outcomes of therapy.
Table 1. Main endocrine systems involved and the clinical manifestation in various IMDs.1,2
| Main Endocrine Organs Involved | Types of IMDs |
| Central precocious puberty | Mitochondrial diseases |
| Central hypothyroidism | Mitochondrial diseases |
| Growth hormone deficiency | Mitochondrial diseases |
| Hypogonadotropic hypogonadism | Mitochondrial diseasesHemochromatosis |
| Secondary adrenal insufficiency | Mitochondrial diseases |
| Hypothyroidism | Fabry diseaseCystinosisCongenital disorders of glycosylationMitochondrial diseases |
| Multinodular goiter | Mitochondrial diseases |
| Congenital thyroid malformation | Mitochondrial diseases |
| Papillary carcinoma | Mitochondrial diseases |
| Diabetes mellitus | HemochromatosisAceruloplasminemiaMitochondrial diseasesGlycogen Storage Disease type I, III, and XIRoger’s syndromeAltstrom syndrome |
| Diabetic ketoacidosis | Organic acid disorders |
| Glucose intolerance | Cystinosis |
| Hypoparathyroidism | Mitochondrial diseasesWilson’s diseasesMitochondrial Trifunctional Protein deficiency |
| Parathyroid hypoplasia | Mitochondrial Trifunctional Protein deficiency |
| Adrenal insufficiency | X-linked adrenoleukodystrophySmith-Lemli-Opitz syndromeMitochondrial diseasesCongenital disorders of glycosylationGlycerol kinase deficiencyZellweger spectrum disorders |
| Adrenal hypoplasia | Glycerol kinase deficiency |
| Subclinical adrenal insufficiency | Fabry disease |
| Hypergonadotropic hypogonadism | X-linked adrenoleukodystrophyClassic galactosemiaCongenital disorders of glycosylationCystinosis (males)Fabry diseaseMitochondrial diseases |
| Hypogonadotropic hypogonadism | Mitochondrial diseasesHemochromatosis |
| Polycystic Ovary Syndrome | Glycogen Storage Disease type I |
As the number of new cases keep rising, demands for early diagnosis of IMDs have also been increasing. Early diagnosis of IMDs can be achieved through identification of early symptoms and newborn screening. One of the most common newborn screening methods is tandem mass spectrometry, which is able to inform several metabolite profiles with a single experimental test. Zhang, et al.2 demonstrated that the positive predictive value of this spectrometry-based method was 4.71%, with the most common IMDs recognized were hyperephenylalaninemia and short-chain acyl-CoA dehydrogenase. In symptomatic patients, the detection rate was 0.37%. Tandem mass spectrometry was also able to observed hotspot mutations through genetic sequencing.2 Another study in Italy showed the role of Expanded Newborn Screening (ENS) in early identification of many IMDs, especially, without early clinical onset (e.g. aminoacidopathies) or IMDS with clinical onset occurring later than the first week (e.g. congenital adrenal hyperplasia). This study also emphasized the importance of early clinical suspicion by healthcare providers (Table 2), as well as appropriate sample collection timing, sample transportation, and sample analysis with follow-ups without delay.3 Challenges in performing diagnosis of IMDs can also arise from the absence of specific metabolic biomarkers, the presence of variants of uncertain significance, common overlaps between biomarkers, and time-consuming diagnostic processes, as well as lack of visualization of the relationships between individual metabolic biomarkers and the enzymes involved. Therefore, integration between clinical data, available biomarker information, and metabolic reaction models must be performed in a framework, which is specifically developed for individualized patient approach.4
Table 2. Potential early symptoms and laboratory findings of main classes of IMDs.3
| Disease Category | IMDs | Early Clinical Symptoms | Early Laboratory Findings | Onset Timing |
| Urea Cycle Disorders (UCDs) | Carbamoyl phosphate synthetase 1 (CPS1) deficiency, Ornithine transcarbamylase (OTC) deficiency, citrullinemia type 1 (ASS1), argininosuccinic aciduria (ASL), hyperargininemia (ARG1) | Progressive lethargy, recurrent cyclic vomiting, increased breathing frequency, seizure, hypotonia, hyperammonemic encephalopathy. | Severe hyperammonemia (more than 200-500 umol/L), respiratory alkalosis, low plasma citrulline levels, increased orotic aciduria. | Around 24-72 hours of life. |
| Organic Acidemias (OAs) | Methylmalonic acidemia (MMA), propionic acidemia (PA), isovaleric acidemia (IVA), glutaric acidemia type 1 (GA-1) | Recurrent vomiting, progressive lethargy, hypotonia, seizure, respiratory failure, metabolic encephalopathy | Metabolic acidosis with high anion gap, lactic acidosis, ketoacidosis, moderate hyperammonemia (less than 200-300 umol/L), increased acylcarnitines. | Around day 2-5 after the start of feeding. |
| Fatty Acid Oxidation Disorders (FAODs) | Medium-chain acyl-CoA dehydrogenase deficiency (MCAD), very long chain acyl-CoA dehydrogenase deficiency (VLCAD), long-chain 3-hydroxyacyl-CoA dehydrogenase deficiency (LCHAD), carnitine palmitoyltransferase I/II deficiency (CPT1/CPT2) | Progressive lethargy, hypoglycemia, hypotonia, liver failure, severe cardiomyopathy, rhabdomyolysis. | Hypoglycemia, low keton levels, altered serum acylcarnitines, increased ammonia. | Between day 2-7 after prolonged disorder or fasting. |
| Disorders of carbohydrate metabolism | Classic galactosemia (GALT), galactokinase deficiency (GALK), epimerase deficiency (GALE) | Milk feeding intolerance, prolonged jaundice, recurrent vomiting, diarrhea, hepatomegaly, liver failure. | Increased direct and indirect bilirubin levels, elevated serum galactose, generalized aminoaciduria, low urinary osmolality. | Within the first two days after the start of milk feeding. |
| Lysosomal storage disorders with neonatal onset | Mucopolysaccharidosis (MPS) type I, Hurler syndrome, Pompe disease, GM1 gangliosidosis, Niemann-Pick disease type A | Severe hypotonia, cardiomyopathy (especially in Pompe disease), hepatosplenomegaly, dysmorphisms. | Elevated tissue glycogen (in Pompe disease), enzyme deficiency, increased urinary glycosaminoglycans (in MPS type 1). | Between the first days and the first weeks of life. |
| Disorders of cysteine and homocysteine metabolism | Homocystinuria (Cystathione Beta-Synthase deficiency), cystinuria, nephropathy cystinosis | Progressive lethargy, feeding difficulties, growth delay, progressive renal impairment (in cystinosis) | Increased serum and urinary homocysteine levels, lysosomal cystine accumulation. | Between the first days and the first months in life. |
The second possible reason is due to health consequence of urbanization in metabolic syndrome since childhood era. Metabolic syndrome is a cluster of cardiometabolic risk factors, which include insulin resistance, central obesity, low levels of High-Density Lipoproteins (HDL), hypertension, and increased triglycerides (TG). Aside from genetic and epigenetic risk factors, increased sedentary lifestyle, unfavorable nutritional intake, increased visceral adiposity, and vitamin D deficiency may also contribute to the increasing trend of childhood metabolic syndrome, worldwide. Although the overall prevalence of metabolic syndrome in children is relatively low, there has been increasing prevalence of overweight adolescents until 4-8 fold increase. The prevalence of metabolic syndrome, worldwide, in 2020 was 2.8% in children and 4.8% in adolescents The prevalence was significantly higher in overweight children (2.8% to 29.3%) and in obese children (10% to 66%).5,6 Rising prevalence of metabolic diseases in younger population has been observed in certain ethnic groups with higher predisposition and heritability. For instance, a study by Amer, etc.7 showed the increasing prevalence of pediatric metabolic syndrome from 11.8% to 20.6% within one decade in almost 8000 children aged 12-18 years old. The increment was more evident in boys compared to girls. Also, in the 2019 cohort, the boys tended to have significantly higher Body Mass Index (BMI), systolic blood pressure, glucose, and HDL cholesterol levels.7
Development of metabolic syndrome in childhood itself was mainly affected by genetic factors, oxidative stress, vascular injury, altered adipokine secretion, and activation of inflammatory pathways. The Fat Mass and Obesity-Associated (FTO) gene on chromosome 16 have been known as a playing a role in energy balance and body weight regulation. The development of obesity, including in children, is closely related to A/A phenotype of the risk allele rs9939609 (T/A). Increased FTO transcription in carriers of the A allele of rs9939609 has been shown to increase the risk of metabolic syndrome in children. FTO variants have been suggested to disrupt the methylation status of FTO target messenger Ribonucleic Acids (mRNAs) and other non-coding RNAs. As a result, there will be an imbalance between energy intake and expenditure, as well as elevated appetite and lowered satiety.8,9 Another gene which has been closely associated with development of metabolic syndrome is CETP gene on chromosome 16. Cholesteryl ester transfer protein (CETP) stimulates the exchange of cholesteryl esters from HDL or LDL to triglycerides-rich lipoproteins. Consequently, concentration of HDL cholesterol will be reduced and small-sized Low-Density Lipoprotein (LDL) particles will be generated. A significant interaction has also been observed between the rs11774572 polymorphism and CETP-TaqIB, located between GATA binding protein 4 (GATA4) and retinitis pigmentosa (RP1) genes in cholesterol metabolism. GATA4 has a role in encoding a transcription factor which mediates the transport of cholesterol and phytosterols; while RP1 genes play a role in changing plasma HDL cholesterol and triglycerides concentrations.10 Another gene which has been associated with high TG levels is the Apolipoprotein A5 (APOA5), particularly the rs662799. APOA5 has been proposed as an inhibitor of the activation of lipoprotein lipase.11 In Asian and Caucasian adolescents, TCF7L2 gene polymorphisms, particularly rs7903146 risk allele, also have higher fasting insulin concentrations, disrupted insulin sensitivity, and greater insulin resistance.12 Although usually established at conception, genetic risk factors in the form of single nucleotide polymorphisms, environmental factors will also strongly influence the baseline of metabolic syndrome during growth. Sedentary behavior, high-fat diets, insufficient sleep, and systemic inflammation are deemed as the strong contributors to the development of obesity as the early part of metabolic syndrome, as well as the progression of insulin resistance in children.8
Another major contributor in the development of childhood metabolic syndrome is early life exposure, especially maternal behavior during pregnancy. Associated gestational conditions; such as gestational diabetes or hypertension or hyperlipidemia, may increase the risk of early development of metabolic syndromes. The understanding about the effect of poor prenatal environment in the womb on the elevated cardiovascular risk in adults has been proposed initially by David Barker who discovered a strong correlation between maternal undernutrition during pregnancy and increased risk of cardiovascular diseases. This Developmental Origins of Health and Diseases (DOHaD) theory stated that high infant mortality and poor living conditions could increase the risk of ischemic heart disease mortality. The theory was initially proposed from a discovery of strong geographical association between infant deaths in 1921-1925 and death rates attributed to ischemic heart disease in 1968-1978 in England and Wales.13 Further observational study indicated the possible involvement of hypertension and low birthweight to higher mortality risk. The risk of non-fatal cardiovascular disease was increased by 50% for the birthweight less than 2267 grams, in comparison to normal birthweight. In addition, girls with low birthweight showed higher tendency to develop hypertension.14
There are several proposed mechanisms of DOHaD theory. The first one is the drastic change of functions of the hypothalamus, pituitary, and adrenal glands during fetal development. The hypothalamic-pituitary-adrenal (HPA) axis mainly regulates intra-uterine homeostasis. Prenatal stress can cause disruption of HPA axis; resulting into increased glucocorticoid expression, disrupted neurodevelopment, as well as excessive glucocorticoid release in utero. The second mechanism is the influence of epigenetic modifications. DNA methylation and histone modification strongly affect birthweight and postnatal condition, especially tissue maturation. For instance, hypermethylation in CYCLIN T2 promoter region in adipose-derived stem cells was found in adult male patients with low birthweight. Moreover, epigenetic changes attributed to prenatal undernutrition can also be transmitted to the offspring. Also, in comparison to other organs, development of cardiomyocyte is highly influenced by the epigenome status. Particularly, suppression of myocardial genes is mainly attributed to DNA methylation and demethylation of H3K27me3 and de novo methylation by DNA methyltransferase 3A/B. The third mechanism is post-translational modification, usually in the form of histone modification. Normally, post-translational modification consists of phosphorylation, acetylation, and ubiquitination. Neonatal insufficient ketogenesis can induce independent starvation and affect histone modification through protein acetylation and hyperacetylation of mitochondrial proteins; triggering nutritional adaptation mechanism; leading to severe hepatosteatosis during the neonatal period; as well as development of non-communicable diseases throughout adolescence and adulthood.13,15
Maternal glucose level has a close relationship with the incidence of obesity in newborns. Up-regulation of genes involved in lipid metabolism was more prominent in the gene expression profiles in placentas from with gestational diabetes compared to women with normal glucose tolerance. A study by Boney, et al.16 showed that the children from obese females also had 2.0-fold increased risk of metabolic syndrome and higher likelihood of also suffering from obesity during adolescence. The same study also performed a multivariate analysis, demonstrating a significant association between the development of insulin resistance and gestational diabetes maternal status. Another multinational study conducted in 23,316 pregnant women underwent a 75-grams oral glucose tolerance test at 24-32 weeks of gestation showed that there was a significantly stronger association between glucose level with birth wight, sum of skinfolds, and body fat percentage at > 90th percentile, compared to hemoglobin A1c.17
In the formation of subcutaneous adipose tissue since fetal period, the progenitor cells and adipocyte populations set the foundation on fat distribution and metabolic health throughout childhood until adulthood. Adverse intrauterine environment; worsened by obesity or diabetes; can induce higher fat accumulation of fetal white adipose tissue. Moreover, high-sugar and high-fat environment can trigger differentiation of stem cells into adipocytes through a premature terminal differentiation process. This process will increase the susceptibility of offspring to suffer from obesity, hinder the plasticity of white adipose tissue and lower its ability to adapt and regulate metabolism of energy.8 Normally, white adipose tissue works as an energy storage in the form of triglycerides packed in lipid droplets of adipocytes. White adipose tissue also forms and secretes adipokines with multiple functions in lipid metabolism. For examples, leptin, which is an adipokine produced by adipocytes and able to regular hunger and satiety by acting on its cognate receptors in the hypothalamus. Another commonly recognized adipokine is adiponectin, which is also secreted from adipocytes in subcutaneous depots, able to increase the sensitivity of liver and muscle to insulin, increase fatty acid oxidation, and lower influx of fatty acids. Most of the white adipose tissue is located on the subcutaneous adipose tissue, instead of visceral adipose tissue. Upon the presence of obesogenic environment, there will also be an expansion of white adipose tissue by inducing hypertrophy of existing adipocytes and hyperplastic expansion of new adipocytes, which were generated from the differentiation of adipose-derived stem cells. These white adipose tissue depots and their resident progenitor pools have been established since fetal life; especially because high rate of growth in the final gestational weeks can occur simultaneously with rapid elevation of fat accumulation, particularly in subcutaneous adipose tissue. Therefore, throughout this critical window of white adipose tissue, maternal obesity of diabetes during pregnancy can perturb intrauterine milieu; resulting into higher risk of metabolic syndrome and obesity occurrence in the offsprings. More limited white adipose tissue plasticity due to perturbation in fetal white adipose tissue can also become a predisposing factor to early onset adiposopathy.18
The Plastic Epigenome and Its Critical Windows
Epigenetic change is a change of gene function which can be inherited and can be found without any modifications to the deoxyribonucleic acid (DNA) sequence. As essential diagnostic instruments, epigenetic biomarkers are consistently recognized to improve workup and management of diseases, including prognostication and individualized treatment. Epigenetic modifications consist of DNA methylation, histone modification and non-coding RNAs. These modifications have a role in regulating gene expression with dynamic and reversible characteristics; thus, making them a potential therapeutic target. DNA methylation is a covalent chemical modification, in which, there is an addition of a methyl group to cytosine’s 5-carbon position. DNA methylation leads to chromatin conformation, which may affect transcriptional silence, genomic imprinting, inactivation of X-chromosome, suppression of transposable element, and genomic stability. The second epigenetic change is histone modification. Histone modification is considered as the most important epigenetic regulator for alteration of chromatin structure. A number of post-translational modifications is experienced by histone proteins, particularly the N-terminal tails of H3 and H4. Post-translational modifications consist of methylation, acetylation, phosphorylation, and ubiquitination. Several specialized enzymes play a role as “writers” (e.g. histone acetyltransferases), “readers” (e.g. chromodomain-containing proteins), and “erasers” (e.g. histone deacetylases) in these processes. Meanwhile, non-coding RNAS are transcripts which are not able to encode proteins, yet are able to affect transcriptional and post-transcriptional processes to regulate the structure of chromatin. There are three main classes of non-coding RNAs: small RNAs, which play a role in breaking down mRNA or halting translation; long non-coding RNAs, which play a role as a framework or regulator of chromatin structure; and circular RNAs, which play a role as a modulator of transcription and translation, as well as acting as microRNA sponges. Non-coding RNAs, in particular, have been noticed as a potential epigenetic biomarker due to their relative stability, abundance, and direct engagement in regulatory networks.19
Epigenetic changes in the uterine environment have been proposed as a background of non-communicable diseases incidence on the offspring. The initial paradox proposed by David Barker was the fact that the least prosperous citizens of a wealthy nation suffered the most from rising rates of cardiovascular diseases. This led to further studies about the effect of adverse fetal environment followed by abundant food in adulthood, and its association with development of chronic non-communicable diseases. The study, which will be known later as the Dutch Hunger Winger study, showed that intrauterine food restriction (as little as 400-800 calories daily) resulted into detrimental effects on subsequent adult health. Interestingly, these long-lasting consequences did not necessarily relate to changes of birth weight. Even babies with normal birth weights, whose mothers were exposed only during early gestation, grew up to have higher rates of obesity, abnormal lipid profiles, and cardiovascular diseases. The general concept from this finding is due to the timing of exposure, whether it matches the critical windows during development or not. Several hypotheses showed that alterations in placental growth due to any transient exposure to deprivation of or excess can affect fetus, even at later stages; especially because the placenta mainly provides oxygen and nutrients to the fetus, as well as adapts to fetal needs through hormonal changes. In addition, fetal adaptations to prenatal famine may become maladaptive when the affected offsprings are exposed to an environment of plenty after birth.20 A study by Eichenauer and Ehlert21 also demonstrated that adult patients with prenatal undernutrition in adverse environmental conditions have increased risk of altered DNA methylation of the IGF2 gene. Impaired DNA methylation of IGF2 gene is associated with fetal development and birth weight, as well as its long-term impact on higher risk of metabolic syndrome.
In the DOHaD hypothesis, epigenetic modification has been recognized as a possible precursor of alterations in developmental programming. In this hypothesis, maternal malnutrition (overnutrition or undernutrition) during early pregnancy can alter the development of the offspring throughout gestation, in which, the alterations can be associated to epigenetic changes. These epigenetic changes participate into the expression, activity level, and silencing of certain genes. DNA methylation, for instance, has a critical role in the embryonal development and cell lineage specification. Consequently, DNA methylation contributes a lot in regulating differential gene expression. High levels of DNA methylation at the cytosine-phosphate-guanine (CpG) site, especially at the promoter regions, will prevent transcription factors from binding to the DNA; resulting into inhibition of gene expression. As another epigenetic change, histone modification also has a crucial role in embryonal development since histone acetylation is strongly associated to gene expression and an open chromatin condition. On the other hand, histone deacetylation has a role in gene silencing; while histone methylation can induce or reduce gene transcription, as well as assembling and compacting heterochromatin. The last epigenetic changes contributing to development of germ cells and embryo are non-coding RNAs. Long non-coding RNAs are closely related to various molecular processes; for example: splicing, regulation of transcriptional and post-transcriptional processes, and modification of chromatins.22
Maternal Nutrition as Molecular Information
Maternal nutrition during pregnancy period plays a critical role in the health outcome of the offspring; especially in terms of molecular changes related to the fetal growth and development. Methyl donor nutrients; such as folate, choline, vitamin B12, methionine, and betaine; can directly affect epigenetic changes, particularly DNA methylation. As has been explained above, DNA methylation plays a vital role in maintaining normal cellular function, regulating gene expression, and embryonic development. Long-lasting outcome on the offspring can occur when changes in the methylation level occur in the promoter region of certain genes. Maternal intake of methyl-donor nutrients itself is important since methyl donor nutrients act as substrates for methylation reaction in neurodevelopmental processes. The general term of this nutrient is one-carbon metabolism nutrient, in which it gives one carbon unit for biosynthesis of DNA, lipid, protein, as well as epigenetic modifications. One-carbon metabolism comprises folate and methionine cycles. In these cycles, folate, choline, methionine, and betaine acts as methyl donors; while vitamin B2, vitamin B6, and vitamin B12 play a role as coenzymes. In addition, S-adenosylmethionine acts as a functional biomarker of one-carbon metabolism.23 A study involving 397 pregnant women at 24-32 gestational weeks in five Chinese cities showed that almost 10% of the subjects suffered from vitamin B12 deficiencies; while elderly pregnant women (aged at least 35-year-old) had significantly lower serum methionine. Statistically significant correlations were also observed between serum folate and serum choline, as well as serum folate and serum betaine. Positive association was also exhibited between plasma S-adenosyl methionine (SAM) and betaine. These findings highlighted the lower levels of vitamin B12 during mid-to-late gestational weeks. In addition, association between SAM and betaine also supported the importance of supplementation of not only folate supplements, but also other one-carbon metabolism-related nutrients.24
Unfortunately, underconsumption of one-carbon metabolism nutrients often occurs simultaneously excessive folate or other unmetabolized folic acids (UMFA). Pregnant women or women of childbearing age who regularly consume combination of supplementation and food fortification are more prone to have higher folate intake compared to the age-adjusted tolerable upper intake level. From the National Health and Nutrition Examination Survey, it was revealed that 95% of American females had significantly higher UMFA levels, especially among supplement consumers. In Addition, a study conducted on Canada showed that over 97% of pregnant women had detectable plasma UMFA.25-27 In a dose-dependent manner, folic acid supplementation will cause higher circulating UMFA. Recommended dietary folate equivalents for breastfeeding and pregnant women were 500 ug and 600 ug, respectively. As for general adults, the tolerable upper intake level of folate supplementation was 1000 ug daily. Excessive folate intake has been linked a number of deleterious pregnancy and offspring outcomes; such as reduced birth length, lower birth weight, higher incidence of plagiocephaly position, and metabolic dysfunction in the late adulthood.25 A prospective study by Huang, et al.28 in 326 pregnant women also demonstrated that excessive consumption of folic acid during pre-pregnancy and early pregnancy had higher incidence of Gestational Diabetes Mellitus (GDM), especially in those taking folic acid supplementation for more than 90 days compared to the participants who consumed folic acid supplementation for < 60 days. The study also revealed a positive correlation between lipid profiles with duration of folic acid supplementation, as well as GDM. Similarly, an animal study by Morakinyo, et al.29 showed that high folic acid supplementation significantly elevated triglyceride and insulin levels; while at the same time significantly diminished insulin sensitivity and adiponectin expression in adult offspring Sprague-Dawley rats.
Another concern that has been raised is about the impact of maternal malnutrition during pregnancy on the occurrence of developmental mismatch. Maladaptive responses attributed to more abundant than predicted post-natal nutritional environment has been associated with increased cardiovascular risk. Elevated cardiometabolic risk has been connected, particularly with poor fetal growth and rapid post-natal weight gain. This hypothesis has been studied by Ong, et al.30 in 1,247 children; showing higher blood pressure and body fat percentage in children with fetal growth deceleration. The same study also found that children with rapid post-natal weight gain had higher body fat percentage, higher Homeostatic Model Assessment of Insulin Resistance (HOMA-IR) values, and higher blood pressure. In addition, children who experienced both fetal growth deceleration and rapid post-natal weight gain also had higher ectopic fat and higher HOMA-IR value. The same group also had the highest blood pressure compared to the children who only had either fetal growth deceleration or rapid post-natal weight gain. The findings of this study highlighted the fact that rapid weight gain during the first two years of life, with or without fetal growth, had a strong relationship with higher cardiometabolic risk in early childhood. The risks of insulin resistance, ectopic lipid accumulation, and high blood pressure were stronger if there was a mismatch between poor fetal growth and rapid post-natal weight gain.30
The relationship between various maternal diet models and their possible metabolic outcome on the offspring through epigenetic modifications is summarize in Table 3.22
Table 3. Relationship between maternal diet models and epigenetic changes in the offspring.22
| Maternal Diet | Epigenetic Changes | Possible Effects on the Offspring | Possible Outcome on the Offspring |
| Dutch famine (undernutrition) | DNA methylation | Increased methylation of PIM3, PFKFB3, METTL8 | Higher risk of metabolic diseases |
| Low Protein Diet (6% protein) | DNA methylation | Reduced methylation of G6PC | Hyperglycemia in adulthood |
| Low Protein Diet (9% protein) | Histone acetylation and methylation | Acetylation of H3, H4; increased methylation of H3k4; reduced GLUT4 expression | Insulin tolerance, metabolic syndrome |
| Low Protein Diet (8% protein) | Histone methylation | Increased methylation of H3K9, decreased Cyp7a1 | Metabolic diseases, long term hypercholesterolemia |
| Low Protein Diet (6.5% protein) | DNA methylation | Increased methylation of CYCS promotor | Disruption of mitochondrial energy metabolism and production |
| High Fat Diet (45% fat) | DNA methylation | Increased methylation in Ephb2 and Fgf21 | Non-alcoholic steatohepatitis (especially in males) |
| High Fat Diet (60% fat) | DNA methylation | Increased methylation of Leptin and Pomc promotors | Insulin resistance |
| High Fat Diet (34% fat) | DNA methylation | Increased methylation of Pomc and InsR promotors in the hypothalamus | Metabolic syndrome |
| High Fat Diet (35% fat) | Histone acetylation | Hyperacetylation of H3K14, H3K9 and H3k18 on the promotors of GPT2 and RDH12 | Obesity, Non-Alcoholic Fatty Liver Disease |
| High Fat Diet (maternal obesity) | Dysregulation of miRNA | Downregulation in miR-181a | Cerebrovascular and cardiovascular diseases |
Maternal Exercise As A Hormetic Signal
Maternal exercise during pregnancy is an essential part of lifestyle behaviors to ensure appropriate weight gain and gestational weight gain for the mother and the offsprings, respectively. The risks of delivery by Caesarean section and post-partum weight gain were significantly higher in cases of excessive gestational weight gain. In addition, more physical activities are strongly related to more benefits on physical and mental health. In uncomplicated pregnancies, physical activity does not have any significant relation with miscarriage, stillbirth, fetal abnormalities, preterm birth, preterm pre-labor rupture of membranes, or even neonatal mortality.31 A meta-analysis by Beetham, et al.32 in low-risk women without any chronic disease or cervical insufficiency or other contraindications to physical exercise; showed that vigorous intensity maternal exercises, in which women were reported to achieve at least 50% of maximum heart rate, showed no significant difference in birth weight and the number of babies who were small for gestational age. A small increase, however, was show in the gestational age at delivery, with an average of 39+4 weeks in group with vigorous intensity maternal exercise, compared to 39+3 weeks in non-exercising control group. A significantly reduced risk was also observed in babies of mothers in group with vigorous intensity maternal exercise. In terms of maternal weight gain, higher percentage of women with excessive weight gain was observed in non-exercising control group (53%) compared to light intensity group (35%) and vigorous intensity group (31%).32
Maternal physical exercise has been highly recommended as safe to be performed regularly for pregnant women and their offsprings in uncomplicated pregnancies. It is recommended for women to keep their physical fitness throughout the pregnancy to tackle the problems which may arise from elevated weight gain, forward drift in the gravitational center, and more prominent lumbar lordosis. Healthy pregnant women with uncomplicated pregnancies should perform at least 20-30 minutes of moderate-intensity physical exercise daily for at least three days week, with aerobic and resistance training being performed, alternately. Gradual increase of frequency and duration of the sessions can be performed in the second trimester. It is also recommended to slowly continue doing regular physical exercises after uncomplicated deliveries, especially since several exercises can be beneficial for post-partum care; such as lowering post-partum weight gain, decreasing the risk of deep vein thrombosis, and lowering the risk of urinary incontinence (i.e. pelvic floor exercises). Practical recommendations for maternal physical activities based on maternal age and fitness level are listed on Table 4.31
| Maternal Age | Fitness Level/Body Mass Index | Range of Heart Rate (beats/minute) | Frequency of Exercises |
| Less than 20 years old | – | 140-155 | Starting from three times weekly. Progression can be made to four times weekly or with accumulation of duration of 150 minutes of moderate intensity physical exercise weekly. In any physical activity session, always perform warm-up and cool-down phases. |
| 20-29 years old | LowActiveBMI more than 25 kg/m2 | 129-144135-150102-124 | |
| 30-39 years old | LowActiveBMI more than 25 kg/m2 | 128-144130-145101-120 |
Activities that are not recommended to be conducted during pregnancy are sports with high risk of falls (e.g. non-stationary cycling, gymnastics) or with unpredictable risks related to shock forces or low oxygen saturation (e.g. sky diving) or with high risk of fetal decompression disease (scuba diving). Maternal exercises are absolutely contraindicated if the pregnant woman suffers from significant cardiovascular disease, restrictive lung disease, incompetent cervix, intra-uterine growth restriction, history of multiple gestation with high risk of pre-term labor, ongoing vaginal hemorrhages, placenta previa after 26 weeks of gestation, pre-eclampsia, eclampsia, pregnancy-induced high blood pressure, pre-term labor, pre-term labor rupture of membranes, significant anemia, and other uncontrolled comorbidities. Increased awareness to the risk of hypoglycemia must also be adopted in maternal exercises with longer duration (more than 45 minutes) or vigorous-intensity exercises. Several safety measures that can be conducted in these conditions include ensuring adequate hydration and caloric intake before performing physical exercise and educating the patients to perform physical exercises in a cool environment.31
Physiologically, during pregnancy, there will be changes in maternal adaptation to metabolize nutrients. Development of fetus and placenta can cause higher maternal energy expenditure. During the early stages of pregnancy, there will be several vital hormones secreted by placenta; which are human placental lactogen (hPL), estrogen, progesterone, and prolactin. All of these hormones will increase maternal blood glucose by reducing hepatic glycogen storage in and inducing more glucose release from the liver. As a result, there will be elevated blood sugar levels; thus, stimulating pancreatic beta cells to secrete more insulin. Therefore, during the final stages of pregnancy, there will be higher risk of insulin resistance; shown by reduced peripheral insulin sensitivity and higher plasma glucose levels of the mother. Decreased insulin sensitivity also occurs in muscle and fat; indicating impaired ability of the mother to utilize glucose optimally. Most of the glucose will be transported to the fetus to ensure the nutritional requirement of the fetus.33
Significant changes also happen in maternal lipid metabolism. In the early stages of pregnancy, fat synthesis will be prioritized to prepare energy reserves; shown by higher estrogen, progesterone, and insulin levels. Elevation of these hormones will induce lipid deposition and lipogenesis. Maternal hyperphagia is also be enhanced by increased progesterone, prolactin, and hPL; which also contributes to fat synthesis. In the second trimester, rising insulin levels will induce lipogenesis and lower fatty acid oxidation; thus, also causing more maternal fat accumulation. In contrast, in the later stages of pregnancy, catabolic state is more prominent; indicated by higher lipid mobilization attributed to lower activity of lipoprotein lipase. As a result, there will be more free fatty acid levels. Some of these free fatty acids will then undergo beta-oxidation in the liver; while some other will be re-esterified to form triglycerides. These mechanisms lead to significant increase of plasma lipid levels in the mother during late pregnancy. High lipid accumulation can trigger metabolic complications during pregnancy while low levels of total cholesterol, LDL, and triglycerides have also been associated with higher risk of small for gestational age offsprings.33
Other prominent adaptive change is found in maternal protein metabolism. During pregnancy, there will be a positive nitrogen balance to fulfill the physiological requirements for both the mother and the fetus. In the early and middle pregnancy stages, maternal plasma amino acid will be decreased, which will activate the phosphoinositil 3-kinase/serine threonine kinase/mTOR pathway in the liver; thus, inducing more protein synthesis. In the third trimester, catabolic state will be prioritized more; indicated by elevated expression of amino acid transporters to assist the transfer of amino acids to the fetus. Protein metabolism is beneficial, especially for improving maternal muscle reserve, insulin sensitivity, uterine artery responses to vasodilators, as well as reducing severity of infectious diseases.33
Integration of exercise into prenatal care has been proven as a low risk and modifiable lifestyle with a large number of benefits to improve metabolic health for both the mother and the offsprings. Aside from enhancing skeletal muscle function, maternal exercise can also induce carbohydrate and lipid metabolisms and improve gut microbiota dysbiosis. Maternal exercise can regulate plasma glucose effectively by enhancing insulin sensitivity. Physical exercise can activate the Glucose Transporter isoform 4 (GLUT4) to optimize glucose uptake and utilization in skeletal muscle. This mechanism is triggered by 5’-adenosine monophosphate-activated protein kinase (MAPK) by stimulating phosphorylation of TBC1 domain family, member 1 (TBC1D1) and TBC1D4 to support translocation of GLUT4 to the plasma membrane. Moreover, aerobic exercise can also decrease physiological increment of insulin during middle to late stages of pregnancy.33,34 Improved insulin sensitivity by maternal exercise can also be achieved through modulation of inflammatory cytokines. Physical exercise can reduce pro-inflammatory cytokines that can disturb insulin signal transduction through interleukin-6 (IL-6) modulation.33 A randomized trial in 46 pregnant women with overweight and obese nutritional status demonstrated that in early gestational weeks, high IL-6 levels were significantly associated with lower fasting insulin levels and lower first phase insulin response. After 12-weeks of moderate-to-vigorous maternal exercise, IL-6 was reduced by almost 20%. This change was observed along with increased first-phase insulin secretion; suggesting the beneficial effects of moderate-to-vigorous maternal exercise on insulin sensitivity.35
Maternal exercise can also mitigate metabolic dysfunction through optimization of lipid metabolism and attenuation of oxidative stress. Exercise can improve lipid metabolism by lowering free fatty acids levels by increasing beta-oxidation process through microRNA-122/Peroxisome Proliferator-Activated Receptor-Beta (PPAR-beta) signaling axis and modulation of maternal-fetal circulating cytokines; such as: interleukin-1-beta, IL-6, interleukin-8 (IL-8), and interleukin-10 (IL-10). Maternal exercise can also increase the utilization of free fatty acids through up-regulation of Fatty Acid Transporter Protein 4 (FATP4) in placenta.33 A study by Acosta-Manzano, et al.36 in 2025 demonstrated the influence of maternal physical exercise during pregnancy on metabolic markers in 88 pregnant females. The results showed that physical exercise contributed to lower maternal total cholesterol and LDL cholesterol gains, especially in subjects who also experienced decrease in IL-8 levels; implying the role of exercise in lipid metabolism regulation through IL-8.36 Attenuation of oxidative stress can also be yielded by performing regular maternal exercise. High oxidative stress; indicated by increased placental HSP70 and decreased Glutamate-Cysteine Ligase Modifier (GCLM), is strongly related to insulin-resistance. Performing moderate-to-vigorous physical exercise can reduce HSP70 and increase GCLM. Moderate exercise can also cause acute increase in placental and serum Apelin levels. High Apelin level can lower Nicotinamide Adenine Dinucleotide Phosphate (NADPH) oxidase activity and stimulate more antioxidant enzymes expression. In addition, Apelin can also decrease endothelial oxidative stress and risk of pre-eclampsia. Resistance training, in particular, can also increase mitochondrial respiratory capacity and protein expression in skeletal muscle; thus, also improving mitochondrial biogenesis and oxidative capacity associated with insulin sensitivity.33
Aside from metabolic adaptations, the role of maternal physical exercise during pregnancy in regulating fetal development can also be explained through epigenetic regulatory mechanisms. The biggest role in maintaining gene expression and metabolic functions was placed on placental DNA methylation. Elevated serpin family A member 3 (SERPINA3) expression, which has a role in encoding an acute phase plasma protein, is significantly related to decreased methylation in its 5’ region. High SERPINA3 expression is vital for fetal development since it contributes to the development of pre-eclampsia and fetal growth restriction.33 A study by Li, et al.37 further elucidated the finding by showing that physical exercise during pregnancy stimulated demethylation of the Klf4 gene promoter in fetal pre-adipocytes by regulating SERPINA3C protein. Consequently, adipose inflammation was inhibited and glucose tolerance was improved. Other genes that may be influenced by maternal exercise, such as TIMP2 and CLINT1, are also known to be affected by placental angiogenesis and fetal substance transport.33
The role of exercise in promoting DNA demethylation can also be seen other molecules involved in DNA methylation. A study by Kusuyama, et al.38 showed that maternal exercise activated vitamin D receptor signaling in placenta. As a result, secretion of superoxide dismutase 3 (SOD) was also increased; promoting DNA demethylation of certain metabolic genes which can improve hepatic glucose homeostasis of the offspring. SOD3 was also able to neutralize the negative effects of high fat diet by the mother. Additionally, SOD3 was able to stabilize H3K4me3 and inhibited carbonylation of WDR82. Both mechanisms preserved histone-mediated epigenetic regulation of metabolic genes in the offsprings. Moreover, physical exercises during pregnancy can also induce placental nutrient transfer through downregulation of Let-7f miRNA expression and upregulation of placental IGF1/IGF1R protein expression. Upregulation of IGF expression is associated with lower risk of fetal growth restriction.33 An animal study by Wang, et al.39 also highlighted the contribution of exercise-induced extracellular vesicle miR-27a to enhanced insulin sensitivity of skeletal muscle. Another study by Zhou, et al.40 in mice demonstrated that maternal exercise during three weeks before pregnancy and throughout pregnancy significantly halted the increment in body weight even after the mice were being fed by maternal high fat diet. Additionally, maternal exercise intervention also ameliorated glucose metabolism abnormalities in the midlife of adult offspring. Similarly, significantly higher fasting serum insulin and HOMA-IR levels were also seen from the findings. Further transcriptomic analysis also revealed that physical exercise during pregnancy repaired diet-induced dysregulation of genes in the liver, which were also involved in glucose and lipid metabolism. Maternal exercise also normalized three main miRNAs in the liver, which normally had roles in regulating cholesterol biosynthesis and epigenetic changes.40
The impact of maternal exercise to metabolic health of offsprings can be classified into the impact on metabolic health, glucose metabolism, and obesity-related metabolism. Generally, pregnant mothers with normal pre-pregnancy weight and regular maternal exercise have significantly lower risks in post-natal complications; such as cyanosis, respiratory distress, or meconium aspiration. On improving maternal glycemic control and reducing GDM incidence, maternal exercise can increase maternal insulin sensitivity; thus, disrupting intergenerational transmission of metabolic dysfunction.33 Previous animal studies on rodent demonstrated that glucose tolerance and hyperinsulinemia in the offsprings could be prevented with treadmill activities at 30% VO2Max with 30-minutes duration for each session, three times weekly. Reduced in fat mass percentages was also observed, especially in male offsprings.41-42 The impact of maternal exercise on obesity-related metabolism of the offsprings has been described as dose-dependent. Modest, yet consistent body weight decrease in offsprings with high risk of obesity has been associated with moderate-intensity maternal exercise, which was initiated as early as the 14th week of gestation, with duration of at least 150 minutes weekly. Aside from reduction in body weight, reduction subscapular skinfold thickness has also been exhibited from routine maternal exercise. These effects are also persistent into adulthood. An animal study by Quiclet, et al.43 performed an animal study with Wistar rats to evaluate the impact of maternal submaximal exercise on the offsprings with high fat or high sucrose diet for ten weeks. The study found higher insulin sensitivity in groups with maternal exercise; determined by the ratio of the level of phosphorylated protein kinase B over total protein kinase B in the liver. In addition, maternal exercise also increased muscle mitochondrial affinity for pyruvate and palmitoyl coenzyme A in the offsprings. Changes in these kinetic properties of the mitochondria were linked to changes in plasma free fatty acids, body weight, and fat mass. Not only from moderate or vigorous intensity exercise, but positive metabolic impact has also been demonstrated by low intensity maternal exercise. Another study on rats performed by Ribeiro, et al.44 exhibited that low intensity exercise (30% of VO2max) during pregnancy and lactation significantly improved glucose tolerance, as well as significantly lowered fat pad storages and fasting glucose. These positive effects on body weight and adiposity were also observed across the lifespan; indicating the persisting metabolic effect which extended beyond post-natal environment. Additionally, aerobic exercise elevated blood flow and contributed to better oxygen and nutrient delivery; resulting into a better overall fetal growth rate. Another interesting finding from this study is the involvement of improved autonomic nervous system activity to VO2max balance. Autonomic nervous system itself can be involved in fatty acid metabolism, triggered by physical exercises; through the stimulation of cardiac activity and increased of blood flow by sympathetic nervous system; as well as reduction in fat pad stores, influenced by sympathoadrenal axis.44
Several trials with human subjects have also examined the impact of maternal exercise during pregnancy to the metabolic outcome of the offsprings. Results from GESTAFIT randomized controlled trial showed that upper-body muscle strength maternal exercise initiated at the 16th week of gestation was significantly associated with higher birth weight. Greater maternal flexibility at the 16th week of gestation and maternal cardiorespiratory fitness were also significantly related with higher oxygen saturation in the umbilical cord blood. This trial also showed possible association between cardiorespiratory fitness in pregnancy and lower risk of caesarean sections.45 Another single-center randomized controlled trial evaluated the impact of maternal exercise, which consisted of combination between 35-minutes duration of moderate intensity endurance activities and 25-minutes duration of strength training for three times weekly. The results demonstrated a significantly lower incidence of GDM and significantly lower systolic blood pressure in pregnant women who underwent maternal exercise. The authors from this study, however, also addressed the possibility of low adherence to the training protocol due to pregnancy-related symptoms; for instance: nausea, exhaustion, time management problems, and limited previous physical training experiences.46 A pilot trial involving 19 pregnant females also addressed the importance of continuous social support to maintain maternal physical exercises during pregnancy. The results showed that, in comparison to obese pregnant women, women with lean body weight had larger mean social support score at the 23rd week of gestation. Nevertheless, improvement of these findings was observed after applying interventions through mobile application which enabled the users to track their caloric intake, physical activity, gestational weight gain, as well providing educational information continuously.47
The Disrupted Milieu and Its Mediators: From Preconception, Pregnancy, and Lactation Stages
Exposure to maternal hyperglycemia and obesity throughout pregnancy can be a major contributor to the development of non-communicable metabolic disorders in the offsprings, particularly through epigenetic changes. It has been well-established that epigenetic modifications can be inherited with persistent variations in genetic regulations. The most stable and well-understood epigenetic alteration is DNA methylation of sequences in promoter regions. In relation to maternal metabolic environment and nutritional intake, detrimental health effects related to fetal programming that can elevate the risk of type 2 diabetes mellitus and obesity in adulthood have been known to be significantly associated with epigenetic modifications. A recent study by Franzago, et al.48 assessed DNA methylation levels in the promoter regions of MC4R and LPL genes from Caucasian mother and infant pairs. MC4R is a seven transmembrane, G-protein-coupled receptor, which plays a prominent role in energy homeostasis and body weight regulation by controlling metabolism and hypothalamic food intake. Peptide ligands of MC4R are related to feeding and energy expenditure. Meanwhile, lipoprotein lipase has been widely known as a major contributor in lipid homeostasis by hydrolyzing triglyceride; thus, also having a role in transferring free fatty acids from maternal lipoproteins to the fetus. The results highlighted a significantly lower MC4R DNA methylation in the neonates of pregnant women with GDM, as well as significantly higher LPL DNA methylation on the fetal side of the placenta in pregnant women with obesity. These results suggested possible involvement of LPL DNA methylation in regulating placental lipid transfer to the fetus. In addition, LPL DNA methylation was also significantly associated with maternal total cholesterol.48
One of the main reasons on why fetal development is a sensitive period on long-term developmental programming is the role of placenta in metabolic and epigenetic modifications. Placenta has a critical role in regulating fetal growth and development. The availability of nutrients crossing into the fetal circulation through placenta is determined by the signaling of receptors for insulin, IGF-1, and leptin located on the maternal-fetal interface to induce amino acid transporter activity in trophoblast cells. In placenta, histone modifications act as the primary mediators of imprinting in several genomic regions. Genomic enhancers, promoters, and silenced regions are related to histone marks H3K4me1/H3K27ac, H3K4me3, and H3K27me3. These epigenetic controls of gene expression in the placenta may contribute to gender differences in placental function; including biases in sensitivity towards insults. Another well-studied epigenetic regulator of placental function is miRNA. miRNAs have a role in epigenetical regulation of genes involved in adaptive and innate immune responses during pregnancy; controlling angiogenic, oncogenic, and anti-apoptotic genes in the first trimester of pregnancy; as well as stimulating cell differentiation in late pregnancy. The most interesting aspect of miRNAs signaling in placenta is their potential to be biomarker since they are able to be detected in maternal blood.49
As a highly active endocrine organ, placenta secrets a lot of hormones in the host bloodstream; including major protein hormones for metabolic functions. Leptin is a peptide hormone, produced by syncytiotrophoblast and extravillous trophoblasts. During the first and second trimester, the concentration of leptin reaches its peak drastically until near week 28. Afterwards, the concentration of leptin will decrease rapidly after delivery. Leptin receptor LepRb is expressed strongly in the hypothalamus and placenta. Through its binding to leptin, it will stimulate ERK and JAK2-STAT5 pathways. Leptin is important for lipolysis and decline of triglyceride and cholesterol levels in placenta. Leptin has also been known to be capable in increasing host-derived amino acid transport, increasing secretion of human chorionic gonadotropin, and inhibiting synthesis of progesterone and human placental growth hormone. Other metabolic protein hormones secreted by placenta are insulin-like growth factors (IGF-I and IGF-II). IGF-1 is produced by syncytiotrophoblast and cytotrophoblast cells throughout the pregnancy; while IGF-II is secreted by cytotrophoblast cells and extravillous trophoblasts in the first trimester of pregnancy. IGF-II, specifically, can bind to insulin receptors; implying its role in development of GDM. IGFs contribute to the proliferation of cytotrophoblast cells through PI3K and MAPK signal transduction. IGFs can also facilitate amino acid transport through the placenta.50
Lactation period and its associated epigenetic changes have also been widely investigated on preventing childhood metabolic syndrome. A meta-analysis by Harder, et al.51 showed an inverse dose-dependent association between breastfeeding duration and risk of overweight in the offsprings.51 As a body fluid with abundant bioactive components, human breast milk has a dynamic consistency and composition, which vary according to the time it is secreted and each lactation meal. Human breast milk also has epigenetic regulators, including miRNAs. These miRNAs are often encapsulated in exosomes or extracellular vesicles along with other proteins or small molecules. These exosomes or extracellular vesicles protect the miRNAs from degradation, as well as assist their transport from one cell to another and transmit epigenetic information to the offsprings. miRNAs from human breast milk are also capable in passing through the intestinal barrier and assembled into Argonaute-2 complexes, which have a major role in post-transcriptional gene regulation. A study by Chondrogianni, et al.52 demonstrated that miR-30b and miR-let-7a had positive association with maternal adiposity or increased body weight throughout pregnancy. miRNAs in human breast milk have also been suggested to influence the development of infants more directly compared to miRNAS detected in plasma or urine. miR-30b is one of the major contributors in stimulating adipogenesis through differentiation of adipocytes and energy metabolism. Meanwhile, miR-let-7a contributes to lipid metabolism by regulating polyunsaturated and saturated fatty acids.52
Hormonal components of breastmilk also provide significant supports for child metabolism, appetite control, energy balance, hormonal regulation, body composition, and accumulation of fat. Among all metabolic targets of hormonal components in breastmilk, the most prominent roles are played by leptin and adiponectin. Leptin regulates appetite and energy expenditure through its actions on hypothalamus. Meanwhile, adiponectin increased insulin sensitivity and oxidation of fatty acids. Higher concentrations of adiponectin in human milk can contribute to higher insulin sensitivity through adiponectin receptor signaling pathways, which results into activation of AMP-activated protein kinase. As a consequence, there will also be more fatty acid oxidation and glucose peripheral uptake. These signaling pathways can also exert anti-inflammatory effect to reduce pro-obesity cytokines. Other hormonal components in breastmilk which may influence infant feeding behavior and body composition is ghrelin. As an orexigenic hormone, ghrelin can regulate pancreatic beta cell development, glucose homeostasis, and sustain metabolic maturation.52
Another factor which may strongly influence early programming of metabolic conditions in the offspring is the impact of maternal dietary patterns and gut microbiota composition. Gut microbiota has also been associated significantly with DNA methylation mechanisms which can potentially contribute to diabetes. Regulation of several methyl-donor nutrients; such as 6-methyltetrahydrofolate; has been connected to composition of gut microbiota; such as Lactobacillus and Bifidobacteria. These microbiomes have also been involved in folate production. Other evidences have also suggested the benefits of modifying maternal microbiome to prevent adverse epigenetic modifications and fetal phenotypes. Unfavorable microbial balance is often associated to high-fat and high-carbohydrate diet. In addition, antibiotic administration and the presence of infections can also influence gut microbiota composition. The most important Phila, which often undergo imbalance initially in gut dysbiosis, are Firmicutes and Bacteroidetes. Particularly, in gut dysbiosis, Firmicutes are more often to be found with lower amount of Bacteroidetes. Higher amount of Bacteroidetes can be induced by elevated population of biliary anaerobes and diets consisting of red meat and dairy products. The imbalance can also be affected by various intrinsic factors; such as intestinal permeability, pH level, and production of mucus. Higher amount of Lactobacillus, Bacteroidetes, and Prevotella was also observed in infants with history of vaginal birth deliveries; while lower quantities of Lactobacillus, Bacteroidetes, and Bifidobacteria were observed in infants with history of Cesarean section deliveries. These findings will influence the recommendation of administering probiotics in prevention of metabolic syndrome.53
Physical exercise has also been demonstrated as a contributor to overall systemic metabolism through breastmilk composition. Potential mechanistic roles of human breastmilk composition included the presence of macronutrients, micronutrients, metabolic hormones, adipokines, miRNAs, and inflammatory mediators. Changes in the breastmilk metabolome remnants with high presence of several acylcarnitines can affect branched-chain amino acid metabolism, especially in individuals with type 2 DM and obesity. In addition, maternal obesity is strongly related to changes in the concentration of human milk oligosaccharides. Human milk oligosaccharides have significant roles in regulating growth of the infants, especially during the first five years of life. Another metabolite which has been shown to have correlation with maternal BMI and infant body weight is 5-methyladenosine. Along with mannose, lyxitol, and shikimic acid; 5-methyladenosine can be used to predict infant adiposity on the first six months of life. All of these bioactive agents in the breastmilk can be modified by physical exercise. Acute increase of 12,13-dihydroxy-9Z-octadenoic acid, which plays a major role in regulating brown adipose tissue fuel uptake and thermogenesis, can be found in pregnant women after exercise. Higher amount of 3’sialylllactose, which is a human milk oligosaccharide acting as a mediator in improving metabolic and cardiovascular health of the offsprings until adulthood, was also shown after exercise training.54
Clinical Layer: The Technological Layer of Maternal Lifestyle Management
Continuously emerging knowledge and availability of artificial intelligence (AI) and wearable technology has offered innovative and applicable solutions to improve diabetes management, including in pregnancy. By possessing the ability of computer system to do a wide range of assignments which usually needs human intelligence, AI can be utilized for data analysis, pattern identifications, predictions build-up, as well as integrating all of them to make clinical decision making, predict therapeutic outcomes, and build individualized treatment approaches. Meanwhile, the usage of wearable devices; for instance: Continuous Glucose Monitoring (CGM) devices or smartwatches; becomes important in real-time data record of physiological parameters in diabetic patients. Bridging AI algorithms and wearable devices provides many potentials for future studies. A study by Fraser, et al.55 in 2025 showed that digital biomarkers with smartphone-based photoplethysmography can be utilized for non-invasive diabetes detection. Additionally, improvement in diagnostic accuracy and ability to predict interstitial glucose trends were also established by application of deep neural networks and machine learning frameworks. Optimization of insulin dosing can also be performed using reinforcement learning and fuzzy logic models. Several limitations, however, still arose from differences in the interpretability of complex AI models and limited generalizability due to selection bias. Another trial in pregnant women also demonstrated improvement of lean body weight after applying interventions through mobile application which enabled the users to track their caloric intake, physical activity, gestational weight gain, as well as sharing educational information regularly.47
Wearable health tracking device have also been proven to be an effective monitoring strategy for maternal and fetal well-being throughout pregnancy. Wearable health tracking devices can be used for monitoring gestational weight gain and maternal physical activity. The interpretation of sensor data is typically enhanced by AI algorithm to detect anomalies, identify patters, and create personalized diagnostics. In general, wearable sensors can be classified into biopotential sensors, pressure and inertial sensors, acoustic sensors, electrodermal sensors, and multimodal systems. Integration with mobile applications can provide an easily accessible and user-friendly environment for both the mother and the infant. Aside from tracking body weight, nutrients intake, and physical activity; mobile-based platforms can also provide educational content, assist in goal settings and progress monitoring; as well as give opportunity to socially connect with other users for peer support. From social point of view, wearable health-tracking devices can overcome healthcare discrepancies by enabling remote diagnostics for vulnerable population in resource-limited areas, as well as facilitating early diagnosis and timely intervention before life-threatening complications occur. Challenges that still need to be addressed are long-term wearability, personal data security, and possible risk of AI error. Nevertheless, clinical decision making still needs to be integrated with examinations from doctors and AI should not be relied upon exclusively to replace direct history taking and physical examination. Gradual establishment of multilayer protection mechanism should be applied by compulsory review from medical professionals even after AI has established a diagnosis.56,57
Another application of modern technology in optimizing metabolic health of the offsprings is by integrating maternal lifestyle and infant’s metabolic trajectories through multi-omics data. A study by Stratakis, et al.58 in 2025 exhibited the ability of specific integrated multi-omics profiles; consisting of DNA methylation, miRNAs, transcript clusters, proteins and metabolites; to provide comprehensive information about critical pregnancy phases and identify potential risk factors management strategies on future metabolic consequences of the offsprings. In addition, multi-omics clustering approach can also assist in determining prenatal factors associated with metabolic dysfunction and childhood obesity beyond only relying upon body weight and/or body mass index. For instance, miRNA expression, transcriptomics, and proteomics were proven to be critical for clustering profiles related to metabolic dysfunction.
Conclusion
Maternal metabolic conditions; including nutrition and physical activity; represents the most powerful and dose-dependent environmental influences on the developing fetal epigenome. During gestational period, nutritional exposures and exercise-related metabolic signals can modulate various epigenetic mechanisms; including histone modifications, DNA methylation, and non-coding RNA expression; thereby, shaping gene regulation during critical periods of fetal development. These epigenetic adaptations have been associated with long-term effects on metabolic health, cardiac function, and non-communicable disease susceptibility across the lifespan. As a result, pregnancy shall be viewed as a compressed longevity intervention, in which maternal behaviors exert biological effects that extend until far beyond birth and may influence health trajectories for more than seven decades in the second human being. This perspective challenges traditional deterministic view of inheritance and underscores a central principle of developmental epigenetics: hereditary is not fixed, but continuously negotiated through dynamic interactions between genetic predisposition and environmental exposure.
References
1) Tagi VM, Fiori L, Montanari C, Tonduti D, Ferrario M, Gambino M, et al. Endocrine system disturbances in children with inherited metabolic diseases: A narrative review. Front Endocrinol. 2025; 16: 1717675. DOI: 10.3389/fendo.2025.1717675.
2) Zhang X, Ji W, Wang Y, Zhou Z, Guo J, Tian G. Comparative analysis of inherited metabolic diseases in normal newborns and high-risk children: Insights from a 10-year study in Shanghai. Clinica Chimica Acta. 2024; 558: 117893. DOI: 10.1016/j.cca.2024.117893.
3) Montanari G, Candela E, Baronio F, Ferrari V, Biasucci G, Lanari M, et al. Early-onset inherited metabolic diseases: When clinical symptoms precede newborn screenings – Insights from Emilia-Romagna (Italy). Children. 2025; 12: 464. DOI: 10.3390/children12040464.
4) Slenter DN, Hemel IMGM, Evelo CT, Bierau J, Willighagen EL, Steinbucsch LKM. Extending inherited metabolic disorder diagnostics with biomarker interaction visualizations. Orphanet J Rare Dis. 2023; 18: 95. DOI: 10.1186/s13023-023-02683-9.
5) Noubiap JJ, Nansseu JR, Lontchi-Yimagou E, Nkeck JR, Nyaga UF, Ngouo AT, et al. Global, regional, and country estimates of metabolic syndrome burden in children and adolescents in 2020: A systematic review and modelling analysis. Lancet Child Adolesc Health. 2022; 6: 158-170. DOI: 10.1016/s2352-4642(21)00374-6.
6) Friend A, Craig L, Turner S. The prevalence of metabolic syndrome in children: A systematic review of the literature. Metab Syndr Relat Disord. 2013; 11: 71-80. DOI: 10.1089/met.2012.0122.
7) Amer OE, Sabico S, Khattak MNK, Alnaami AM, Aljohani NJ, Alfawaz H, et al. Increasing prevalence of pediatric metabolic syndrome and its components among Arab youth: A time-series study from 2010-2019. Children. 2021; 8: 1129. DOI: 10.3390/children8121129.
8) Zhang B, Shi H, Cai W, Yang B, Xiu W. Metabolic syndrome in children and adolescents: Definitions, epidemiology, pathophysiology, interventions, and challenges. Front Endocrinol. 2025; 16: 1512642. DOI: 10.3389/fendo.2025.1512642.
9) Szkup M, Owczarek AJ, Schneider-Matyka D, Brodowski J, Loj B, Grochans E, et al. Associations between the components of metabolic syndrome and the polymorphisms in the peroxisome proliferator-activated receptor gamma (PPAR-gamma), the fat mass and obesity-associated (FTO), and the melanocortin-4 receptor (MC4R) genes. Aging (Albany NY). 2018; 10: 72-82. DOI: 10.18631/aging.101360.
10) Bideyan L, Lopez Rodriguez M, Priest C, Kennelly JP, Gao Y, Ferrari A, et al. Hepatic GATA4 regulates cholesterol and triglyceride homeostasis in collaboration with LXRs. Genes Dev. 2022; 36: 1129-1144. DOI: 10.1101/gad.350145.122
11) Xu C, Bai R, Zhag D, Li Z, Zhu H, Lai M, et al. Effects of APOA5-1131T>C (rs662799) on fasting plasma lipids and risk of metabolic syndrome: Evidence from a case-control study in China and a meta-analysis. PLoS One. 2013; 8: e56216. DOI: 10.1371/journal.pone.0056216.
12) Ferreira MC, da Silva MER, Fukui RT, Arruda-Marques MDC, Dos Santos RF. TCF7L2 correlation in both insulin secretion and postprandial insulin sensitivity. Diabetol Metab Syndrome. 2018; 10: 1-7. DOI: 10.1186/s13098-018-0338-1.
13) Arima Y, Fukuoka H. Developmental origins of health and disease theory in cardiology. J Cardiology. 2020; 76: 14-17. DOI: 10.1016/j.jjcc.2020.02.003.
14) Jarvelin MR, Sovio U, King V, Lauren L, Xu B, McCarthy MI, et al. Early life factors and blood pressure at age 31 years in the 1966 northern Finland birth cohort. Hypertension. 2004; 44: 838-846.
15) Tikanmaki M, Tammelin T, Vaarasmaki M, Sipola-Leppanen M, Miettola S, Pouta A, et al. Prenatal determinants of physical activity and cardiorespiratory fitness in adolescence – Northern Finland Birth Cohort 1986 study. BMC Public Health. 2017; 17: 346.
16) Boney CM, Verma A, Tucker R, Vohr BR. Metabolic syndrome in childhood: Association with birth weight, maternal obesity, and gestational diabetes mellitus. Pediatrics. 2005; 115(3): e290-296. DOI: 10.1542/peds.2004-1808.
17) Lowe LP, Metzger BE, Dyer AR, Lowe J, McCance DR, Lappin TRJ, et al. Hyperglycemia and Adverse Pregnancy Outcome (HAPO) study. Diabetes Care. 2012; 35: 574-580. DOI: 10.2337/dc11-1687.
18) Scheidl TB, Brightwell AL, Easson SH, Thompson JA. Maternal obesity and programming of metabolic syndrome in the offspring: Searching for mechanisms in the adipocyte progenitor pool. BMC Medicine. 2023; 21: 50. DOI: 10.1186/s12916-023-02730-z.
19) Bencik I, Martinovic LS, Mladenic T, Ostojic S, Pavlic SD. From DNA methylation and histone modifications to non-coding RNAs: Evaluating tools for epigenetic research. Appl Sci. 2025; 15: 9940. DOI: 10.3390/app15189940.
20) Schulz LC. The Dutch Hunger Winter and the developmental origins of health and disease. PNAS. 2010; 107(39): 16757-16758. DOI: 10.1073/pnas.1012911107.
21) Eichenauer H, Ehlert U. The association between prenatal famine, DNA methylation, and mental disorders: A systematic review and meta-analysis. Clinical Epigenetics. 2023; 15: 152. DOI: 10.1186/s13148-021-01557-y.
22) Peral-Sanchez I, Hojeij B, Ojeda DA, Steegers-Theunissen RPM, Willamie-Morawek S. Epigenetics in the uterine environment: How maternal diet and ART may influence the epigenome in the offspring with long-term health consequences. Genes. 2022; 13, 31. DOI: 10.3390/genes13010031.
23) Villamor E, Rifas-Shiman SL, Gillman MW, Oken E. Maternal intake of methyl-donor nutrients and child cognition at 3 years of age. Paediatr Perinat Epidemiol. 2012; 26(4): 328-335. DOI: 10.1111/j.1365-3016.2012.01264.x.
24) Zhang R, Wu X, Lu L, Hu R, Teng Y, Pan L, et al. Assessment of blood one-carbon metabolism indexes during mid-to-late pregnancy in 397 Chinese pregnant women. Front Nutr. 2024; 11: 1348930. DOI: 10.3389/fnut.2024.1348930.
25) Fardous AM, Heydari AR. Uncovering the hidden dangers and molecular mechanisms of excess folate: A narrative review. Nutrients. 2023; 15: 4699. DOI: 10.3390/nu15214699.
26) Fazili Z, Sternberg MR, Potischman N, Wang CY, Storandt RJ, Yeung L, et al. Demographic, physiologic, and lifestyle characteristics observed with serum total folate differ among folate forms: Cross-sectional data from fasting samples in the NHANES 2011-2026. J Nutr. 2019; 150: 851-860.
27) Plumptre L, Masih SP, Ly A, Aufreiter S, Sohn KJ, Croxford R, et al. High concentrations of folate and unmetabolized folic acid in a cohort of pregnant Canadian women and umbilical cord blood. Am J Clin Nutr. 2015; 102: 848-857.
28) Huang L, Yu X, Li L, Chen Y, Yang Y, Yang Y, et al. Duration of periconceptional folic acid supplementation and risk of gestational diabetes mellitus. J Drug Dev. 2019; 6: 321-329. DOI: 10.6133/apjcn.201906_28(2).0014.
29) Morakiyo AO, Samuel TA, Awobajo FO, Oludare GO, Mofolorunso A. High-dose perinatal folic-acid supplementation alters insulin sensitivity in Sprague-Dawley rats and diminishes the expression of adiponectin. J Diet Suppl. 2018; 14-26, DOI: 10.1080/19390211.2018.1426076.
30) Ong YY, Sadananthan SA, Aris IM, Tint MT, Yuan WL, Huang JY, et al. Mismatch between poor fetal growth and rapid postnatal weight gain in the first 2 years of life is associated with higher blood pressure and insulin resistance without increased adiposity in childhood: The GUSTO cohort study. Int J Epidemiol. 2020; 49(5): 1591-1603. DOI: 10.1093/ije.dyaa143.
31) Lee R, Thain S, Tan KH, Ang SB, Tan EL, Tan B, et al. Guidelines on physical activity and exercise in pregnancy. Perinatal Society of Singapore. 2020.
32) Beetham KS, Giles C, Noetel M, Clifton V, Jones JC, Naughton G. The effects of vigorous intensity exercise in the third trimester of pregnancy: A systematic review and meta-analysis. BMC Pregnancy and Childbirth. 2019; 19: 281. DOI: 10.1186/s12884-019-2441-1.
33) He Z, Yuan Y, Zhang L, Niu W, Liu L, Chen B, Wang X. The metabolic implications of maternal exercise: Effects on pregnant women and their offspring. Front Nutr. 2026; 13: 1741714. DOI: 10.3389/fnut.2026.1741714.
34) Florez-Opazo M, McGee SL, Hargreaves M. Exercise and GLUT4. Exerc Sport Sci Rev. 2020; 48: 110-118. DOI: 10.1249/jes.0000000000000224.
35) van Poppel MNM, Peinhaupt M, Eekhoff MEW, Heinemann A, Oostdam N, Wouters MGAJ, et al. Physical activity in overweight and obese pregnant women is associated with higher levels of proinflammatory cytokines and with reduced insulin response through interleukin-6. Diabetes Care. 2014; 37: 1132-1139. DOI: 10.2337/dc13-2140.
36) Acosta-Manzano P, Flor-Alemany M, van Poppel MNM, Coll-Risco I, Segura-Jimenez V, Stanford KI, et al. Concurrent exercise training during pregnancy is related to more favourable maternal lipid levels when IL-8 increases. J Sports Sci. 2025; 43(3): 308-322. DOI: 10.1080/02640414.2025.2456384.
37) Li Y, Li RY, Zhu JY, Cen M, Mu WJ, Luo Hy, et al. Maternal exercise prevents metabolic disorders in offspring mice through SERPINA3C. Nat Metab. 2025; 7: 401-420. DOI: 10.1038/s42255-024-01213-6.
38) Kusuyama J, Alves-Wagner AB, Conlin RH, Makareqicz NS, Albertson BG, Prince NB, et al. Placental superoxide dismutase 3 mediates benefits of maternal exercise on offspring health. Cell Metab. 2021; 33: 939-956. DOI: 10.1016/j.cmet.2021.03.004.
39) Wang D, Zhang X, Li Y, Jia L, Zhai L, Wei W, et al. Exercise-induced browning of white adipose tissue and improving skeletal muscle insulin sensitivity in obese/non-obese growing mice: Do not neglect exosomal miR-27a. Front Nutr. 2022; 9: 940673. DOI: 10.3389/fnut.2022.940673.
40) Zhou L, Li S, Zhang Q, Yu M, Xiao X. Maternal exercise programs glucose and lipid metabolism and modulates hepatic miRNAs in adult male offspring. Front Nutr. 2022; 9: 853197. DOI: 10.3389/fnut.2022.853197.
41) Ribeiro TA, Tofolo LP, Martins IP, Pavanello A, de Oliveira JC, Prates KV, et al. Maternal low intensity physical exercise prevent obesity in offspring rats exposed to early overnutrition. Sci Rep. 2017; 7: 7634. DOI: 10.1038/s41598-017-07395-2.
42) Carter LG, Lewis KN, Wilkerson DC, Tobia CM, Ngo Tenlep SY, Shridas P, et al. Perinatal exercise improves glucose homeostasis in adult offspring. Am J Physiol Endocrinol Metab. 2012; 303: E1061-1068. DOI: 10.1152/ajpendo.00213.2012.
43) Quiclet C, Dubouchaud H, Berthon P, Sanchez H, Vial G, Siti F, et al. Maternal exercise modifies body composition and energy substrates handling in male offspring fed a high-fat/high-sucrose diet. J Physiol. 2017; 595.23: 7049-7062. DOI: 10.1113/JP274739.
44) Ribeiro TA, Tofolo LP, Martins IP, Pavanello A, de Oliveira JC, Prates KV, et al. Maternal low intensity physical exercise prevents obesity in offspring rats exposed to early overnutrition. Scientific Reports. 2017; 7: 7634. DOI: 10.1038/s41598-017-07395-2.
45) Baena-Garcia L, Coll-Risco I, Ocon-Hernandez O, Romero-Gallardo L, Acosta-Manzano P, May L, Aparicio VA. Association of objectively measured physical fitness during pregnancy with maternal and neonatal outcomes. The GESTAFIT project. PLoS ONE. 2020; 15(2): e0229079. DOI: 10.1371/journal.pone.0229079.
46) Garnaes KK, Morkved S, Salvesen O, Moholdt T. Exercise training and weight gain in obese pregnant women: A randomized controlled trial (ETIP trial). PLoS Med. 2016; 13(7): e1002079. DOI: 10.1371/journal.pmed.1002079.
47) Perera M, Hawk GS, Nagpal TS, Tinius RA. Social support for exercise from pregnancy to postpartum and the potential impact of a mobile application: A randomized controlled pilot trial in Southern United States. Preventive Medicine Reports. 2023; 36: 102485. DOI: 10.1016/j.pmedr.2023.102485.
48) Franzago M, Borelli P, Di Nicola M, Cavallo P, D’Adamo E, Di Tizio L, et al. From mother to child: Epigenetic signatures of hyperglycemia and obesity during pregnancy. Nutrients. 2024; 16: 3502. DOI: 10.3390/nu16203502.
49) Nugent BM, Bale TL. The omniscient placenta: Metabolic and epigenetic regulation of fetal programming. Front Neuroendocrinol. 2015; 39: 28-37. DOI: 10.1016/j.yfrne.2015.09.001.
50) Ahmadi SM, Perez ML, Guardia CM. Secretion of placental peptide hormones: Functions and trafficking. Front Endocrinol. 2025; 16: 1584303. DOI: 10.3389/fendo.2025.1584303.
51) Harder T, Bergmann R, Kallischnigg G, Plagemann A. Duration of breastfeeding and risk of overweight: A meta-analysis. Am J Epidemiol. 2005; 162(5): 397-403. DOI: 10.1093/aje/kwi222.
52) Dheir NA, Radwan H, Hasan H, Papandreou D, Zeb F, Abdelrahim DN, et al. Breastfeeding and prevention of childhood obesity: A narrative review of behavioral, hormonal, and microbiome programming. Front Nutr. 2026; 13: 1800487. DOI: 10.3389/fnut.2026.1800487.
53) Faienza MF, Urbano F, Anaclerio F, Moscogiuri LA, Konstantinidou F, Stuppia L, et al. Exploring maternal diet-epigenetic-gut microbiome crosstalk as an intervention strategy to counter early obesity programing. Curr Issues Mol Biol. 2024; 46: 4358-4378. DOI: 10.3390/cimb46050265.
54) Moholdt T, Stanford KI. Exercised breastmilk: A kick-start to prevent childhood obesity. Trends Endocrinol Metab. 2024; 35(1): 23-30. DOI: 10.1016/j.tem.2023.08.019.
55) Fraser RA, Walker RJ, Campbell JA, Ekwunife O, Egede LE. Integration of artificial intelligence and wearable technology in the management of diabetes and prediabetes. NPJ Digit Med. 2025; 8: 687. DOI: 10.1038/s41746-025-02036-9.
56) Ahmedova A, Veleva N, Stoev S, Petkova-Gueorguieva E, Yanachkova V, Lebanova H. Wearable health-tracking technologies in pregnancy: Capabilities, clinical evidence, and implementation challenges. Pharmacia. 2026; 73: 1-10. DOI: 10.3897/pharmacia.73.e184239.
57) Liu L, Pu Y, Fan J, Yan Y, Liu W, Luo K, et al. Wearable sensors, data processing, and artificial intelligence in pregnancy monitoring: A review. Sensors. 2024; 24: 6426. DOI: 10.3390/s24196426.
58) Stratakis N, Anguita-Ruiz A, Fabbri L, Maitre L, Gonzalez JR, Andrusaityte S, et al. Multi-omics architecture of childhood obesity and metabolic dysfunction uncovers biological pathways and prenatal determinants. Nature Communications. 2025; 16:654. DOI: 10.1038/s41467-025-56013-7.