Assessing the association between hypoxia during craniofacial development and oral clefts

Abstract Objectives To evaluate the association between hypoxia during embryo development and oral clefts in an animal model, and to evaluate the association between polymorphisms in the HIF-1A gene with oral clefts in human families. Material and Methods The study with the animal model used zebrafish embryos at 8 hours post-fertilization submitted to 30% and 50% hypoxia for 24 hours. At 5 days post-fertilization, the larvae were fixed. The cartilage structures were stained to evaluate craniofacial phenotypes. The family-based association study included 148 Brazilian nuclear families with oral clefts. The association between the genetic polymorphisms rs2301113 and rs2057482 in HIF-1A with oral clefts was tested. We used real time PCR genotyping approach. ANOVA with Tukey's post-test was used to compare means. The transmission/disequilibrium test was used to analyze the distortion of the inheritance of alleles from parents to their affected offspring. Results For the hypoxic animal model, the anterior portion of the ethmoid plate presented a gap in the anterior edge, forming a cleft. The hypoxia level was associated with the severity of the phenotype (p<0.0001). For the families, there was no under-transmitted allele among the affected progeny (p>0.05). Conclusion Hypoxia is involved in the oral cleft etiology, however, polymorphisms in HIF-1A are not associated with oral clefts in humans.


Introduction
Oral clefts are a common birth defect seen worldwide 16  Hypoxia is an environmental factor involved in the etiology of different birth defects 27 . Oral clefts are a possible hypoxia-induced birth defects. Some evidence in human populations suggests an association between these two conditions, such as maternal cigarette smoking, acardiac twining, and living in very high altitudes (>6500 feet from sea level) 6,7,10,13 . Data on birth defects in the highlands of South America showed a significantly higher risk for oral clefts when compared to the population living at sea level 6 .
Many studies assessed the association between maternal cigarette smoking during pregnancy and oral clefts. A meta-analysis of these studies suggested a positive association between oral clefts and maternal smoking 12 , as well as passive maternal smoking.
Tobacco smoke contains thousands of compounds, some of which are known to have deleterious effects during embryonic development. Among smoke products, nicotine is considered to be the main teratogenic substance that alters and delays embryonic development 21 . The carbon monoxide in tobacco smoke is another mechanism that may cause oral clefts due to maternal smoking, since it reduces the availability of oxygen to the fetus, causing hypoxia 9,13 . This study had two main objectives: 1) to evaluate the influence of the hypoxia condition in the craniofacial formation; and 2) to evaluate the association between oral cleft in humans and HIF-1A. Therefore, we used zebrafish as a model to evaluate the association between hypoxia during embryo development and oral clefts; and we evaluated the association between genetic polymorphisms in HIF-1A with oral clefts in human families.

Material and methods
Animal model and hypoxic conditions  We performed a pilot study to establish a protocol that allows the study of the association between hypoxic conditions during embryonic development and clefts in zebrafish. Sample size was also estimated according to the results observed in the pilot study. We repeated the experimental condition described above three times with 10 embryos in each group to assure that the results were consistent.
The hypoxia experiment was performed with two experimental hypoxic condition groups: 30% hypoxic condition and 50% hypoxic condition. We created a control group for comparison. We used 105 samples (35 per group).
Embryos at 8 hpf (right before neural crest cell migration) the embryos were submitted to 30% and 50% hypoxia treatment until 32 hpf developmental stage (stage where the face structures are forming).
After the hypoxia treatment, the embryos were transferred to a normoxic condition until reaching 5 days post-fertilization (dpf).
For the hypoxia treatment, the oxygen level was reduced by bubbling nitrogen gas into the water.
The oxygen levels were monitored using a dissolved

Family-based association study
This study was approved by the Human Ethics We collected saliva samples for DNA extraction.
The genomic DNA was extracted from buccal cells based on the previously published protocol 11 . The

Family-based study
We included 530 subjects in this study (148 oral cleft nuclear families). In 78 families, both parents were included in the study (complete triads); in 61 families, only the mother was available, and in 9 families, only the father was available (70 dyads). One  Table 1 presents the characteristics of the sample.

Discussion
Some experimental in vivo studies 4,14,17 and observational studies 6,9,10,13 suggest that hypoxia in the first trimester can increase the risk of oral clefts.
To the best of our knowledge, our study is the first to observe that hypoxia during embryonic development can cause oral clefts in zebrafish. Zebrafish are an important model in the study of developmental biology and organogenesis 1,3,7,18 . For this reason, we decided to use zebrafish embryos as a model to evaluate the effect of severe levels of hypoxia on the palate development.
We were able to note that hypoxia caused clefts in all treated larvae. In addition, we observed that the level of hypoxia was directly related to the severity of the cleft. In the experimental group with 50% hypoxia, there were deeper clefts than in the group with 30% hypoxia. Zebrafish are a valuable model to study jaw development due to having the same skeletal elements as higher vertebrates 25 , however, it is important to consider the extent to which mechanisms for palatogenesis are conserved across the species.
The mammalian palate development is a process in which primary and secondary palatal shelves develop as outgrowths from the medial nasal and maxillary prominences, respectively, remodel, and fuse to form the intact roof of the oral cavity 8 . The zebrafish palate development consists of a series of bones in the roof of the mouth that separate the oral cavity from the brain.
The development of this structure does not involve palatal shelf formation, but instead the condensation of cranial neural crest-derived mesenchyme 5 .
Some studies propose that oral clefts could result from interference with neural crest cell migration 2 .
In our study, we performed the hypoxia treatment before the migration of cranial neural crest cells. In   are not associated to oral clefts etiology.

Authors of a study performed with 2524 mothers
who gave birth to babies with an oral cleft, hypothesize that the severity of oral cleft might increase due to hypoxia resulting from gestational bleeding 19 . Our study, with an animal model, suggested that higher levels of hypoxia are associated with a more severe cleft phenotype. However, our familial study did not observe any preferential under-transmitted allele according to the cleft type.
We examined a gene-environment interaction to verify if polymorphisms in HIF-1A are associated with oral clefts in a subgroup that had history of maternal smoking during pregnancy, which is a condition that causes oral clefts due to the reduced oxygen. In fact, animal studies show that the administration of nicotine during pregnancy decreases uterine blood flow 29 . We did not find an association. However, the sample size of those who reported maternal smoking during pregnancy might be too small for this type of analysis. Polymorphisms in genes that express nicotinic receptors could be involved in oral clefts, since studies report the effects nicotinic receptors on the uterine arteries 29 .
Despite the association between hypoxia and oral cleft being reported in literature, genes and their function in neural crest induction, specification and differentiation started recently. The genetic circuits that coordinate this complex developmental process in humans is still largely unknown. Our data from the family analysis should be interpreted with caution.
Although our results did not show an association between polymorphisms in HIF-1A and oral clefts, we can hypothesize that other polymorphisms in HIF-1A are associated with oral clefts. Since all mothers lived at sea level during pregnancy, this could be the reason we were not able to detect any association.

Conclusion
Our results support the hypothesis that hypoxia plays an important role in the oral cleft etiology. However, we were not able to find an association between polymorphisms in HIF-1A and oral clefts in humans.