Dr. Patrick Stover discusses the competition between genome synthesis and DNA methylation for the use of available folate and one-carbons.
The Balancing Act of Folate Metabolism
So one of the primary focus of our laboratory is trying to link how changes in metabolism affect specific disease outcomes. And we’re most interested in neural tube defects and colon cancer.
So if you look at the epidemiology literature, whether it’s the nutritional epidemiology literature, where they look at how folate deficiency is associated with diseases, or you look at the genetic epidemiology literature, looking at how various genetic variants are associated with diseases, what you see in the whole field of folate metabolism is that disruptions in metabolism, either caused genetically or by nutrition, are tightly correlated with birth defects and cancer risk. Specifically, colon cancer shows some of the best data for a linkage between the vitamin metabolism and the disease.
We are interested in understanding what’s that black box in the middle. What is the specific pathway that folate’s involved in? And the two leading candidates are genome synthesis, genome stability, which relates back to nucleotide biosynthesis, or methylation.
And there’s a long literature that goes back decades that shows, in fact, that these two pathways compete in folate metabolism. That is, if you increase the flux towards generating one-carbons to methylating chromatin, to methylate DNA and histones, that comes at the expense of one carbons that can be used for nucleotide biosynthesis.
Vice versa, when you shunt towards nucleotide biosynthesis those one-carbons, you don’t have as much methylation capacity to methylate the genome. So we’re interested in this fork in the road. How is a decision made to send one-carbons for DNA synthesis or to build up methylation capacity to regulate the genome? And does this decision point play a role in these folate-related pathologies?
So we make transgenic mice, where we push and pull that decision, that fork in the road between genome stability and methyl group biosynthesis, and we then see how changing the flux one way or the other affects specific outcomes. And we look at neural tube defects and we look at colon cancer, and we do this in mouse models.
Nucleotide Biosynthesis and Disease
So one of the most interesting findings the last couple years that we’ve had, if you look at the bias in the literature, people believe that neural tube defects and these cancers were really an epigenetic phenomenon. That is, disregulation of methyl group biosynthesis and then subsequently chromatin methylation somehow were involved in these disorders.
What we have found is that it’s actually nucleotide biosynthesis, at least in the models that we have made in my laboratory. What we’ve been able to show is that if you disrupt the synthesis of thymidylate. And thymidylate is one of the four bases in DNA. It’s also a very interesting base, because it’s the only one that’s not required for DNA synthesis. We require A, G and C for DNA synthesis, but T isn’t essential.
If you don’t have sufficient T base, you can misincorporate U into DNA. So you have urea being incorporated into DNA. Folate is needed to convert the U base to the T base. And if one carbon metabolism isn’t as efficient as it should be, and there’s not sufficient T, DNA polymerases don’t discriminate between T and U, and instead of having an AT base pair, you will have an AU base pair.
And we’ve been able to show in our laboratory that we can create mice, either through genetic disruption or through combinations of diet and genetic disruptions, where we can increase uracil content and DNA up to 50-fold.
When we do this, we see a neural tube defect phenotype. Now it’s long been known that folate can prevent about 70% of human neural tube defects. These are disorders that happen very early in pregnancy, development disorders, between day 23 and 28 in humans. You have closure of the neural tube. So the neural epithelium has to proliferate, it has to migrate, and it has to differentiate to form a neural tube.
If during this critical window, before a woman normally even knows she’s pregnant, between day 23 and 28 of pregnancy, if you don’t have sufficient proliferation, migration and differentiation of the neural epithelium, the neural tube will not close in particular regions. And if it doesn’t close in that critical window, it will remain open throughout gestation and that individual will be paralyzed from the site of the lesion down.
So if it happens in the cranial region, that’s always lethal. If it happens more in the trunk region, you have disorders such as spina bifida, where you have then paralysis from the site of the lesion down. These are common. They’re about 1 in 1,000 births. They’re debilitating. They’re expensive.
But the good news is in the 1970s– or rather in the 1990s– clinical trials showed that 70% of these can be prevented with folic acid. We don’t know the mechanisms. We’ve made the first mouse model of folate responsive neural tube defects. That is, if we genetically make a mouse that disrupts folate metabolism leading to uracil incorporation in DNA, they get neural tube defects. If we give them elevated levels of folate, we can lower that level of uracil misincorporation in DNA and we prevent the neural tube defect.
So that’s been a major discovery in our laboratory the past couple years. We’ve also shown that that same mouse that accumulates more uracil in DNA is more susceptible to intestinal tumors. So we get both phenotypes in this mouse related to uracil. We get neural tube defects and susceptibility to colon cancer.
Now this doesn’t mean, though, that epigenetics isn’t involved. Because we’ve gone on to look at the mechanism of how this pathway that converts the U base to the T base, how that functions. And if you look in the textbooks, it will show that that folate pathway is actually present in the cytoplasm.
“So therefore, every time during replication, if there’s an A on the template, you get a choice. The metabolism is right there. You can put in a U, or you can use folate and put in a T.”
What we’ve shown the past couple years is it’s present in the cytoplasm, but during times of DNA synthesis, either in S phase or in DNA damage, that entire pathway moves into the nucleus, associates with the replication fork, and you actually convert U to T at the replication fork on site, on demand.
So it’s always been known that T is the limiting nucleotide for DNA synthesis. And it was assumed with the other three nucleotides it was made in the cytoplasm. What we’ve shown is it is exclusively made in the nucleus at the replication fork and made on demand.
So therefore, every time during replication, if there’s an A on the template, you get a choice. The metabolism is right there. You can put in a U, or you can use folate and put in a T.
So we’re interested in that decision whether to put in a U, whether to put in a T, and whether or not that decision then relates to these outcomes of both birth defects and colon cancer.
Interestingly, we found that associated with this metabolic pathway is one of the histone demethylases, KDM-1. KDM-1 was recently shown to also use folate. And this was done out of Connie Wagner’s lab, about a year and a half ago. So not only do you need folate to methylate chromatin, both histones and DNA, but now we see that some of the histone demethylases require folate.
And now we’re showing an interaction now between DNA synthesis and some of these one carbons that are generated from histone demethylation, linking DNA synthesis to some of these epigenetic signatures. And so we’re trying to understand how, again, this interaction then relates to these folate-related pathologies.