Objective
Alcohol (ethanol) consumption during pregnancy is linked to congenital heart defects that are associated with fetal alcohol syndrome. Recent reports have associated ethanol exposure with the Wnt/β-catenin pathway. Therefore, we defined whether ethanol affects Wnt/β-catenin signaling during cardiac cell specification.
Study Design
Pregnant mice on embryonic day 6.75 during gastrulation were exposed by an intraperitoneal injection to a binge-drinking dose of ethanol. Folic acid supplementation of mouse diet was tested for the prevention of ethanol-induced cardiac birth defects.
Results
Acute ethanol exposure induced myocardial wall changes and atrioventricular and semilunar valve defects, which was determined by echocardiography on embryonic day 15.5. A high folate diet prevented the ethanol-induced cardiac defects. Ethanol exposure in avian embryos suppressed 2 key Wnt-modulated genes that are involved in cardiac induction; folic acid rescued normal gene expression.
Conclusion
Folic acid supplementation alone or with myoinositol prevented alcohol potentiation of Wnt/β-catenin signaling that allowed normal gene activation and cardiogenesis.
Fetal alcohol syndrome (FAS) is characterized by cardiac defects, fetal growth restriction, neurodevelopmental delays, and craniofacial malformations. In a recent epidemiologic study, periconceptional alcohol use was associated with cardiac birth defects, specifically conotruncal (outflow) defects and transposition of the great arteries. The FAS-related outflow tract congenital birth defects are similar to those that we reported in animal models that were exposed acutely to the therapeutic drug lithium or to an elevated dose of the metabolic intermediary homocysteine. We demonstrated that a single exposure to lithium or to homocysteine during gastrulation of vertebrate development induced cardiac valve defects by modulating canonical Wnt/β-catenin signaling. We provided evidence that folic acid (folate) alone or in combination with myoinositol can rescue the adverse effects on cardiac development that are induced by lithium, homocysteine, and Wnt3A. It had been reported that ethanol suppresses Wnt/β-catenin signaling in bone marrow stromal cells, which provides further credence to the possibility of alcohol interacting with the canonical Wnt pathway during cardiogenesis. Because of the similarities between cardiac birth defects that are associated with FAS and those induced by lithium, homocysteine, or canonical Wnt exposure, we investigated whether alcohol (ethanol) impacts Wnt/β-catenin signaling and can be protected by folate.
The mechanism by which alcohol induces congenital defects is not known, nor is it known how early in gestation the damage occurs. The gastrulation stage of gestation is targeted specifically in our studies, because it is an especially vulnerable period in pregnancy during which cardiac, neural crest, and neural cell fates are being specified. The cardiac effects of alcohol exposure during this early period of development have not been defined, specifically in relation to the canonical Wnt pathway, which is an important regulatory pathway in cardiac progenitor cell fate specification. The specification signaling cascade that involves canonical Wnt/β-catenin is active during gestation when a woman may be unaware she is pregnant. In the human, our targeted experimental window extrapolates to 16-18 days after conception. We also analyzed whether there is a potential for folate or folate/myoinositol to protect the embryo from ethanol’s adverse effects on development.
Materials and Methods
Avian model
White leghorn chick ( Gallus gallus ; Charles River Laboratories, Wilmington, MA) and quail ( Coturnix coturnix ; Strickland Farms, Lake Park, GA) embryos were used because of accessibility to determine precisely the embryonic stage. Heart development in the avian embryo is similar to that in the mouse and human. Stage 4 embryos were incubated on an agar-albumin medium for 8 or 24 hours at 38°C. The teratogenic ethanol dose was determined to be 25-30% ethanol. Control embryos were incubated on physiologic saline solution in agar-albumin medium. For avian folate (Sigma Chemical Company, St. Louis, MO) rescue experiments, folate was added to the agar-albumin at a dose of 10 μg/mL, which had been determined previously. Myoinositol (Sigma Chemical Company) was used at 277 mmol/L (stock solution). For in ovo exposure, ethanol was injected into the blunt-end egg airspace at doses of 10% or 30% in physiologic saline solution.
Mouse model
The C57Bl/6 mouse strain (Jackson Laboratories, Bar Harbor, ME) was used. Mice handling was performed according to protocols that have been approved by the University of South Florida Institutional Animal Care and Utilization committee. Morning of the vaginal plug was defined as embryonic day 0.5. Pregnant mice were administered a binge drinking level of ethanol (30%; ie, intraperitoneal injections of 306 μL of 2.9 g ethanol/kg maternal weight that were administered at 3 pm and at 6 pm on embryonic day 6.75); 1 intraperitoneal injection of ethanol (3.5 g/kg or 370 μL), or a higher dose of 4.5 g/kg. Only the 2 intraperitoneal exposure doses on embryonic day 6.75 led to healthy adult mice and induced defects in embryonic cardiovascular function. The 2-intraperitoneal dose regimen was used throughout these analyses. On embryonic day 15.5, the circulations of the pregnant mice and embryos were examined in utero by Doppler ultrasonography.
Folate rescue mouse diet
Animal diet that was supplemented with 10.5 mg/kg or 6.2 mg/kg folate was prepared by Harlan Laboratories (Madison, WI): These doses are based on human population trials for rescue of craniofacial anomalies. Control mice received the baseline diet that contained 3.3 mg folate/kg, which is a dose that maintains the health of the pregnant dam but does not rescue the cardiac defects. The calculations for folate level in the folate-supplemented diets were based on the metabolic body weight of mice because of the body weight difference between humans and mice. For mice, it is calculated as body weight 0.75 (Harlan Laboratories). Dams were assigned randomly to an experimental group that was supplemented with the high 10.5 mg/kg or moderate 6.2 mg/kg dose of folate or to a control group that received normal diet (3.3 mg/kg). On embryonic day 0.5, pregnant mice that had been assigned to the experimental groups were placed on the folate supplemented diets. On embryonic day 6.75, experimental pregnant females received ethanol or the control dams received physiologic saline solution by intraperitoneal injection. One group of control pregnant mice remained uninjected (the untreated control).
Doppler ultrasonography
On embryonic day15.5, Doppler ultrasonographic (echocardiograph) examinations were performed with a 40-mHz transducer, as previously described with Vevo 770 (VisualSonics, Toronto, CA) instrumentation.
In situ hybridization
Control and experimental chick embryos were analyzed with digoxigenin-labeled riboprobes with alkaline phosphatase detection.
Histologic findings and microscopy
The avian embryos were processed for immunohistochemistry according to published protocol. The monoclonal antibody MF20 (developed by D.A. Fischman) used to localize sarcomeric myosin was obtained from the Developmental Studies Hybridoma Bank, developed under the auspices of the National Institute of Child Health and Human Development and maintained by the University of Iowa, Department of Biology (Iowa City, IA). Mouse embryonic heart sections also were stained with hematoxylin-eosin.
Results
Alcohol exposure of avian embryos during gastrulation
Avian embryos that were exposed to ethanol in vitro at Hamburger and Hamilton staging stage 4 displayed morphologic defects of delayed growth and severe cardiac malformations ( Figure 1 , A, control, and B-D, ethanol exposed; Figure 2 ). After 24 hours, the early cardiac malformations included (1) wide hearts at embryonic middle and cardia bifida (ie, bilateral heart fields not fusing, (2) cardiac tissue situated anterior to the head that signified a truncation of the neural tube because of an inhibition of convergence-extension, and (3) left-looping hearts arising from abnormal development of the second heart field (SHF). These defects most likely would be embryonic lethal.
Using smaller quail eggs at stage 4, we analyzed ethanol exposure effects in ovo by microinjecting 10%, 25%, or 30% ethanol into the airspace at the blunt end of the egg. Ten percent ethanol had no adverse effects; however, 25%/30% ethanol produced cardiac valve anomalies in 68% of the embryos (32% normal). For both in vitro and in ovo regimens, the administration of folate (10 μg/mL) concurrently with ethanol resulted in cardiac protection; 51% of the embryos displayed normal heart development. An additive protective effect occurred with the administration of folate/myoinositol that resulted in 62% normal embryos, which was close to the control level of 67% for this stage ( Figures 3 and 4 ).
We next examined whether ethanol alters the gene expression patterns of 2 Wnt-modulated genes that are critical in cardiac cell specification: Hex is a homeodomain gene that is involved in primary heart field specification that signifies the left ventricle and part of the right; Islet 1 is a marker of the SHF that gives rise to the outflow part of the right ventricle that includes the tricuspid valve and the conotruncal region and its derivatives. Hex and Islet 1 expression in control embryos is shown in Figure 5 , A and E, respectively (sense negative control, Figure 5 , B and F). In comparison, ethanol exposure suppressed the canonical Wnt-modulated gene expression of Hex ( Figure 5 , C) and of Islet 1 ( Figure 5 , G). With folate supplementation, both gene expression patterns were normalized ( Figure 5 , D and H) and appeared often at higher intensity levels than that seen in control embryos. In summary, folate suppressed ethanol inhibition and protected the cardiac progenitor cell expression of 2 early genes critical in cardiac induction.
Mouse embryo ethanol exposure and folate protection
We next examined whether folate protects mouse hearts from the known adverse effects of an exposure to a binge-drinking dose of ethanol that targeted a 3-hour window of gastrulation. Animal diet supplemented with the high 10.5 mg/kg diet or the moderate 6.2 mg/kg dietary dose of folate was fed to pregnant females starting on the morning of the vaginal plug. The pregnant mice were injected intraperitoneally with ethanol or control saline solution on embryonic day 6.75 (gastrulation) at a level that is used commonly as a model of human binge drinking. We used the 2 injections straddling our previously defined target window, because 1 intraperitoneal injection of 3.5 g/kg did not produce any abnormalities. One intraperitoneal injection of 4.5 g/kg ethanol resulted in cardiac abnormalities, but the pregnant mouse appeared sickly. By administering 2 intraperitoneal injections of 2.9 g/kg each on embryonic day 6.75, the pregnant dam remained healthy, but embryonic/fetal abnormalities were prevalent. Therefore, the 2–intraperitoneal injection regimen was used in all of our subsequent studies that used the mouse model. More than 1 week later after ethanol exposure, 87% of the mouse embryos had cardiac defects. The myocardial wall often appeared thinner; in these hearts, trabeculation was decreased in comparison with control hearts. With acute exposure on embryonic day 6.75, predominantly semilunar (aortic and pulmonary) valve regurgitation was noted (77%), with a lower percentage of embryos displaying atrioventricular (tricuspid) valve defects. Placental function was also compromised. Effects on heart and placental function were identified noninvasively by monitoring blood flow patterns with echocardiography on embryonic day15.5 ( Tables 1 and 2 ).
