Cu is an essential trace element for bodies. Besides playing important roles in several biological activities including electron transfer and scavenging free radical (35), Cu also involves in embryonic development (36). In Menkes disease, an X-linked recessive genetic disease characterized by Cu metabolism disorder, a higher CHDs prevalence (4.2%) was found as compared with general populations (about 1%), suggesting a possible involvement of Cu dysregulation in cardiac development (37).

Excessive Cu exposure is associated with increased risks of CHDs (Figure 1). A case–control study showed that the offspring of mothers with hair Cu levels ≥17.77 μg/g had a 6-fold risks of CHDs, as compared with that of mothers with hair levels of 5.61–17.77 μg/g (OR = 5.70, 95% CI: 2.58, 12.61) (38). Similarly, a specific pattern of cardiac malformation syndrome including double-outlet right ventricle, pulmonary hypoplasia and ventricular spetal defect has been detected in offspring of hamsters that exposed to excessive Cu during pregnancy (29, 39). The abnormal morphological structures and impaired physiological functions of embryonic heart induced by Cu exposure have also been found in several model organisms such as red sea bream (Pagrus major) (40), rare minnow (Gobiocypris rarus) (28), and marine medaka (Oryzias melastigma) (41). Although the specific mechanisms are largely unknown, it has been proposed that impaired anti-oxidant system and altered development-related gene expression patterns may be responsible for Cu-induced cardiac malformation (28).

However, Cu is also an essential trace element for cardiac development, which has been studied in several animal models with Cu deficiency (Figure 1). Uriu-Adams’s group cultured rodent embryos from dams fed with Cu-deficient diet in Cu-deficient media, and found high risks of heart abnormalities in these embryos (26, 35, 42). Subsequent studies revealed an impaired anti-oxidant defense system and increased reactive oxygen species (ROS) levels in these Cu-deficient embryos, and supplementation of anti-oxidant agents such as glutathione peroxidase (GPx) and Cu-Zn superoxide dismutase (CuZnSOD) rescued the heart abnormalities (42). The increased ROS levels induced by Cu-deficiency can be partially explained by the dysfunction of cytochrome-c oxidase (COX), the terminal enzyme in the respiratory chain (43). The deficiency of Cu, the pivotal component of COX, led to an impaired COX activity, which could result in leakage of electrons and reaction intermediates and elevated ROS levels in mitochondria (42, 44). Similarly, in offspring of rat dams fed with Cu-deficient diet, the COX subunit 1 (COX-1) and COX-4 activities were dampened in the late postnatal development in cardiac mitochondria (45). Another possible mechanisms on Cu-deficiency-induced heart abnormalities comes from the increased protein nitration and low bioavailability of nitric oxide (NO) (26, 35). Under the circumstance of Cu deficiency, the elevated O₂^•- reacts with NO rapidly to form peroxynitrite (ONOO⁻), leading to increased protein nitration and dysregulated protein functions (35). The large consumption of NO results in a decreased NO bioavailability, which may induce cardiac abnormalities (26). Moreover, the decreased phosphorylation levels of endothelial NO synthase (eNOS) at Ser1177 in Cu-deficient embryos may further exaggerate the low bioavailability of NO (26).

Cu can pass through the placental barrier through the active transport of two Cu-ATPases (ATP7A and ATP7B), therefore, the fetus may also be affected by maternal Cu status (46). However, in a recent case–control study conducted by Yang et al., no association was detected between dietary or supplemental Cu intake in mothers during pregnancy and CHDs occurrence in offspring (47). The results may be interpreted that few pregnant women in the study population had deficient or excessive Cu status. Therefore, it is still unclear whether it is necessary to supplement Cu in Cu-deficient pregnant women in order to prevent CHDs in offspring.

Zinc (Zn) is an essential trace element for human, and exerts important roles in regulating immunity and maintaining male reproductive function (48, 49). Most importantly, Zn is also closely associated with cardiac development.

Results from epidemiological studies regarding Zn and CHDs seem to be equivocal, though limited evidence show potential benefit of Zn on heart development. For instance, higher serum Zn levels were associated with decreased odds of isolated ventricular septum defects (VSDs) in children (50). A recent study evaluated the association between maternal dietary or supplementary Zn intake during pregnancy and CHDs risks in offspring (47). The results showed that higher intake of Zn was associated decreased risks of CHDs in offspring. In addition, mothers whose total Zn intake during pregnancy reached the recommended nutrient intakes (RNIs) (i.e., 9.5 mg/d) had lower risks of CHDs in offspring, as compared with those did not reach. These results were consistent with a previous study showed that Zn supplementation in mice rescued cardiac malformation induced by diabetes through modulating oxidative stress (25). However, some studies did not found similar results. A study showed that serum Zn concentrations in CHDs fetus and their mothers were significantly higher than those in controls (51). Several studies demonstrated that no significant difference of maternal Zn levels existed between CHDs and controls (38, 52). Sadoh et al. measured serum Zn levels in 41 children with CHDs and 41 children without CHDs, but found no significant difference between the two groups (53).

However, several animal studies showed a pivotal role of Zn in embryonic heart development (Figure 2). In 1966, Hurley et al. demonstrated a high incidence of abnormalities in the embryonic hearts of Zn-deficient rats (54). Further studies have illustrated that cardiac malformations induced by Zn deficiency were mainly involved in the great vessels, the outflow tract and the development of the atrium and ventricle, which may possibly due to the anomalous distribution, amount, and function of cardiac neural crest cells (cNCC) during embryonic development (55, 56). Indeed, rat cNCC cells cultured in Zn-deficient media exhibited decreased cell viability, and elevated oxidative stress levels and active caspase-3 expressions (57). Another study used connexin-43 (Cx43) and HNK-1 as biomarkers of cNCC and found abnormal amount and distribution of Cx43 and HNK-1 in the embryonic hearts of Zn-deficient rats (56). In addition, the gene expression patterns of α-myosin heavy chain (α-MHC) and cardiac troponin I (cTnI), both of which are important in embryonic cardiac development, were significantly altered in the Zn-deficient rat fetus hearts (55). Zn deficiency also decreased expression levels of metallothionein-1 (MT-1) and zinc transporter-1 (ZnT-1) in rat placentas, which may contribute to the cardiac malformation in Zn-deficient fetus, although the direct evidence is still lacking (58). Zn deficiency-induced cardiac abnormalities is also mediated by dysregulated sumoylation and desumoylation during heart development (41). Zn-deficient fetus demonstrated decreased small ubiquitin-related modifier protein (SUMO)-1 levels and increased SUMO-specific protease (SENP)-5 levels in the embryonic hearts at embryonic day (E) 10.5. Further in vitro analysis showed that Zn deficiency induced human induced pluripotent stem cells (hiPSC)-derived cardiomyocytes (hiPSC-CMs) apoptosis, and inhibited cell viability and differentiation, and these adverse effects were mediated through SENP-5 overexpression.

Few studies have investigated the cardiac teratogenicity of excessive Zn (Figure 2). Zhu et al.’s study showed that Zn exposure induced cardiac morphological abnormalities and cardiac dysfunction in embryonic rare minnow (Gobiocypris rarus) (28). The study also detected increased expression of vezf1, a pivotal regulator in embryonic heart and vascular development, in Zn exposure embryos, which may partially explain the cardiac teratogenicity of Zn.

High prevalence of Fe deficiency has been evident in patients with cyanotic CHDs, which may be explained by the increased Fe consumption due to excessive erythropoiesis in response to hypoxia (59–61). However, the roles of iron in the embryonic heart development as well as the pathophysiology of CHDs are still largely unknown.

A case–control study conducted by Yang et al. showed that low levels of Fe intake, Fe supplementation, and Fe status in mothers were associated with increased risks of CHDs in offspring (27). Similarly, animal studies also found that Fe deficiency increased risks of cardiac abnormalities. In 2006, Andersen et al. found that cultured rat embryos with Fe deficiency showed poor yolk sac circulation and decreased heart size (62). A recent study further unveiled the essential role of Fe in cardiac development (63). The results showed that maternal Fe deficiency led to increased risks of ventricular septal defects (VSDs), atrioventricular septal defects (AVSD), thin ventricular myocardium in embryos. These CHDs phenotypes in Fe deficient embryos were associated with malrotation of the cardiac outflow tract (OFT), abnormal cardiac cushions, as well as abnormal aortic arch. Further analysis showed that Fe deficiency resulted in elevated retinoic acid (RA) signaling in second heart field (SHF) and OFT, leading to ectopic activation of cardiac transcription factor GATA4 and premature differentiation of SHF cardiac progenitor cells. Most importantly, these CHDs phenotypes can be rescued by RA signaling abrogation or dietary supplementation of Fe during mid-gestation.

Conversely, a recent case–control study demonstrated that high maternal hair Fe concentrations (≥52.95 μg/g) were associated with 2.87-fold risks of fetus CHDs, as compared with pregnant mothers with hair Fe of 43.15–52.95 μg/g (aOR = 2.87, 95% CI: 1.54, 5.37) (64). These results suggest a potential cardiac teratogenic effect of excessive Fe. Moreover, it is also reasonable to deduce that excessive Fe supplementation may increase the risks of fetus CHDs in pregnant women, though direct evidence is still lacking. However, other biomarkers that reflex body Fe status such as haemoglobin and ferritin were not evaluated in the same study.

Se plays important roles in multiple biological processes such as antioxidant defense, cellular signaling, and protein folding (65). Se deficiency has been closely correlated with cardiovascular diseases, including Keshan disease, myocardial infarction, and coronary heart disease (66). However, the role of Se in cardiac development and CHDs is still largely unknown. A case–control study conducted by Ou et al. showed that maternal blood Se collected during middle and late gestation was significantly lower in CHDs infant compared with controls (67). After adjusting several confounding factors, the authors demonstrated that Se exposure >199.67 μg/l was associated with 75% decreased risks of CHDs (adjusted odds ratio (aOR) = 0.25, 95% CI: 0.08, 0.77), compared with Se levels <178.12 μg/L. Similarly, a recent study found that both dietary and supplement intakes during pregnancy were inversely associated with lower risks of CHDs in offspring (47). However, Guo et al. utilized hair to assess Se exposure in pregnant women, and found that high Se exposure was associated with increased risks of CHDs in offspring (68). One possible explanation is the U-shape effect of Se on human health, and the specific role of Se on CHDs may dependent on baseline Se status (69, 70).

Mn is an essential trace metal that implicates in various biological functions such as anti-oxidative stress, metabolic processes, and regulating endocrine (71). However, both Mn deficiency and excessive Mn exposure can lead to increased ROS as well as neurological impairments (72). An epidemiological study showed that mothers with hair Mn >3.01 μg/g has increased risks of CHDs in fetus, as compared with mothers with hair Mn of 0.11–3.01 μg/g (aOR = 2.68, 95% CI: 1.44, 4.99) (64). Conversely, Zhu et al.’s study showed that maternal dietary supplementation of organic Mn reversed oxidative stress and apoptosis induced by heat stress in chick embryonic hearts (73). However, the study did not research the alterations of morphology or functions of embryonic hearts.

Cd is a toxic non-essential metal that can result in various types of cancer, osteoporosis, as well as liver and renal diseases (30). Moreover, the teratogenic effect of Cd during cardiac development has been evident. A case–control study evaluated maternal hair Cd levels and CHDs occurrence in offspring showed that high maternal Cd levels (≥ 25.85 ng/g) were associated with a 1.96-fold increased risks of CHDs as compared with low Cd levels (≤7.23 ng/g, aOR = 1.96, 95% CI: 1.24, 3.09) (74). Similarly, in zebrafish (Danio rerio) embryos exposed to 10 μM CdCl₂, some larvae showed hypertrophy of the ventricle and pericardium (75). In addition, Cd exposure during embryos also affected cardiovascular physiology in zebrafish, as evident in reduced heart rate and abnormally elevated heart contractility (75). In chick embryos, Cd exposure increased the myocardial tissue area of the right ventricle, though scarcely gross cardiac malformation was detected (75). Further mechanical analysis showed that these pathological changes were associated with increased cell proliferation and upregulated expression of cell circle related genes such as Cdk1, Cdk6, CycA, CycD, and CycE. Since right ventricular hypertrophy has been identified as an independent risk factor of heart failure and cardiovascular death (76), it is speculated that Cd exposure during cardiac development may increased risks of cardiovascular diseases in adulthood. Subsequent study conducted by Hudson et al. may partially confirm this hypothesis, which showed that maternal Cd exposure in mice increased heart weight at birth as well as risks of hypertension in adulthood in offspring (77). However, it is needed to point out that Cd can hardly be detected in offspring of mothers that exposed to Cd, which may be explained by the presence of placental barrier (77). Further study demonstrated that the cardiac alterations may be secondary to the altered essential trace element profiles induced by maternal Cd exposure (77).

As is a toxic metalloid that naturally occurred via geogenic processes in the aquifer (78). It is estimated that 200 million people are exposed to high levels of As in the world, which has raised public health concerns (31). Epidemiological studies associated maternal exposure to As in drinking water with increased risks of CHDs in offspring (79–81). Moreover, a case–control study evaluating maternal hair As levels and CHDs in fetus also showed similar results (74). Experimental studies also confirmed that As exposure may affect cardiac development as well as heart function in embryos. Li et al. assessed the developmental toxicity of As utilizing zebrafish (Danio rerio) embryos, and the results showed that As exposure significantly decreased the amount of myocardium in the ventricle, and delayed cardiac looping (82). As exposure also dampened heart rate in the zebrafish embryos in time-and dose-dependent manners. Hematoxylin and eosin (HE) staining conducted by the same group also showed increased pericardial cavity as well as elongated atrium and ventricle in As exposed zebrafish embryos, and these effects are potentially mediated via suppressed expression of Dvr1, the pivotal regulator of dorsal mesendoderm activity (83). Maternal As exposure in rats increased the risks of CHDs (especially ventricular septal defects (VSDs) and atrial septal defects (ASDs)) in rats, and these effects could be reversed by folic acid supplementation (84). A series studies conducted by Camenisch et al. showed that As and its toxic metabolite (i.e., monomethylarsonous acid) can impair epithelial-mesenchymal transition (EMT) during cardiac development, potentially through affecting TGF-β/Smad signalings (85–87).

Pb is a toxic element which shows deleterious systematic effects on human body, and its exposure remains an important public health problem (88). It has been reported that Pb exposure is associated with increased risks of cardiovascular diseases such as hypertension and atherosclerosis (89). In particular, prenatal Pb exposure in pregnant woman may affect cardiac development in offspring. For example, a case–control study conducted by Salehi et al. showed significantly elevated blood Pb concentrations in mothers of children with CHDs (90). In another case–control study, Wang et al. evaluated maternal plasma Pb levels in 303 CHD cases and 303 controls, and found that each unit concentration of Pb was associated with a 2.74-fold (95%CI: 1.00, 7.57) increased risks of CHDs after adjusting for several potential confounders (91). Consistent with these results, Ou et al. found that high levels of maternal blood Pb (blood Pb >3.04 μg/dl) were associated increased risks of CHDs in fetus, as compared with mothers with low blood Pb levels (<2.15 μg/dL) (aOR = 12.09, 95% CI: 2.81, 51.97) (67). Moreover, a study that utilizing maternal hair Pb as a biomarker of Pb exposure also detected Pb as a harmful factor for CHDs (92). These results showed a significant and positive association between Pb exposure and CHDs, however, there is still a lack of studies that exploring the causal effect of Pb on cardiac development and the underlying mechanisms.

Ni is a ubiquitous metal in the earth, and has been associated with allergy and carcinoma in human-beings (30). A previous case–control study conducted by our group first associated maternal Ni exposure with fetus CHDs (93). The results showed that maternal hair Ni >0.7216 ng/mg was associated with 2.672-fold risks of CHDs in offspring as compared with mothers with hair Ni <0.4111 ng/mg (aOR = 2.672, 95% CI: 1.623, 4.399). Laboratory-based studies further support the cardiac teratogenic effect of Ni. In marine medaka (Oryzias melastigma) embryos, Ni exposure induced cardiovascular anomalies as well as dysregulated expression of cardiac development-related genes including ATPase, smyd1, cox2, and bmp4 (94). Similarly, Ni exposure in zebrafish (Danio rerio) embryos led to abnormally increased expression of cardiac gene including gata4 and nkx2.5 (95). One possible explanation for the cardiac teratogenesis role of Ni is that excessive Ni can impair the electrophysiology of Ni²⁺-sensitive T-type Ca²⁺ channels in the embryonic hearts (96, 97).

Ba is a widespread heavy metal in nature, and its poisoning can lead to impairment of gastrointestinal, cardiovascular, and musculoskeletal systems (98). One potential mechanism is that Ba can inhibit the potassium inward rectifier channels (IRCs), thus affecting all types of muscle (99). Particularly, a case–control study that conducted by our group linked prenatal Ba exposure with CHDs in offspring (100). In this study, we measured maternal hair Ba concentrations in 399 CHDs and 490 controls, and found that Ba exposure was associated with increased risks of CHDs and major subtypes (i.e., septal defects, right ventricular outflow track obstruction, left ventricular outflow track obstruction, and anomalous pulmonary venous return) in offspring in a dose–response manner.

Cr is a heavy metal that may play roles in regulating the metabolism of glucose, lipid, and protein (101). However, there is no convincing evidence that can confidently conclude the essential role of Cr in human beings (102). Conversely, Cr exposure has been associated with multiple health hazards such as allergy, impaired reproductive system, and cancer (103). Cr exposure during pregnancy has been associated with poor fetal biometric parameters such as abdominal circumference and estimated fetal weight, suggesting the embryonic toxicity of Cr (104). Ou et al. explored the association between maternal blood Cr levels during pregnancy and CHDs in offspring, but did not detect significant result (67).

Hg is a toxic heavy metal that is hazardous to human health (105). Long term exposure to a low level Hg also leads to cardiovascular, reproductive, and developmental toxicity (106). A recent case–control study evaluated Hg levels in maternal plasma found that each unit concentration of Hg was associated with 2.88-fold risks of CHDs in offspring (aOR = 2.88, 95% CI: 1.22, 6.77) (91). However, there is still a lack of study that validate the causal effects as well as the specific mechanisms between Hg and cardiac development.