Closing the neural tube: gene-gene interactions of TET1 and the interplay with nutrient metabolism

The etiology of congenital malformations, particularly neural tube defects (NTDs), is complex, and involves a combination of genetic and environmental factors affecting each other. Understanding the relationship between intrinsic genetic factors and extrinsic environmental factors has been challenging. Here in this thesis, I describe my efforts to improve our understanding of the epigenome-gene-environment interaction model. Of particular interest is the DNA demethylase TET1; Tet1 null embryos develop a NTD with different rates dependent on the genetic background. Its capabilities as a dioxygenase demethylating DNA places TET1 as an ideal environmental sensor during development. Therefore, I hypothesize that early in development, TET1 is an important mediator of the interaction between the epigenome, genome and environment, to properly coordinate embryonic development. This thesis is assembled of one manuscript published, and two manuscripts currently in preparation:

In the first manuscript, I dissect the catalytic vs non-catalytic functions of TET1 in neural induction. Tet1 null mouse embryonic stem cells (ESC) shift from neural to mesoderm and endoderm fates, while ESCs expressing an endogenously mutated catalytically inactive TET1, form ectoderm. Further, TET1-dependent differential methylated regions are located at enhancers, but there is an uncoupling between DNA methylation and chromatin accessibility. Additionally, peri-gastrulation TET1 antagonizes non-catalytically PRC2 at neuroectoderm gene promoters, attenuated by Wnt signaling regulated by TET1 catalytic activity. These results show that the function of TET1 is highly dependent on the genomic context and developmental timepoint.

In the second manuscript, first we evaluate the percentage of embryos with NTDs in Tet1 null embryos on different genetic backgrounds. We define a quantitative trait locus (QTL) by analyzing single nucleotide variants (SNVs) in crosses with a high-penetrant and low-penetrant strain. Further, due to this QTL, embryonic stem cells derived from the high-penetrant strain do not activate the same signaling pathways as cells from the low-penetrant strain. Second, we assess how maternal folic acid (FA) modulation affects Tet1 null embryos. Tet1 null embryos are non-responsive to FA supplementation or deficiency, possibly due to downregulation of one of the FA transporters. However, supplementation of FA in Tet1 null embryos leads to hypermethylation and results in downregulation of neurotransmitter genes. These results reveal a strong interdependency between the epigenome and the genome that regulates cellular responsiveness to FA supplementation and individual susceptibilities to congenital disorders.

In the last manuscript, I evaluate whether deficiency of vitamin C (VitC), an important co-factor for dioxygenases and TET1 catalytic activity, results in a similar phenotype as loss of Tet1 does. Depending on the genetic background, VitC deficiency results in gross embryonic malformations and developmental delays, but no NTDs. Further, Vitamin C deficiency combined with loss of TET1 results in exacerbation of the delays and malformations, suggesting that the phenotypes observed are not a simple result of disruption of the catalytic function of TET1. Additionally, we detect an increase in locus specific 5mC, and a global increase in H3K9me2. These results highlight the role of VitC in modulating the fetal epigenome and gene-environment interactions during early development.

In conclusion, the role of TET1 in development is multifaceted and nuanced, involving both its non-catalytic function and its 5mC oxidizing function. The impact of TET1 and other dioxygenases on NTDs is influenced by genetic variation and environmental factors, which can affect susceptibility to congenital malformations and disease later in life.