 Hello everyone and thank you for your interest in stem cells. My name is Keith Tanaway and I'm the first author of this paper. Many of you will find the most exciting aspect of this study to be that we used baby teeth to identify epigenetic differences in an autism spectrum disorder. So while the title is Dental Pulp Stem Cells Model Early Life and in Printed DNA Methylation Patterns, some of us like to think of this as the epigenetic tooth fairy paper. DNA methylation is an epigenetic mark that can affect and reflect gene regulation. Our lab is very interested in identifying DNA methylation changes in neurodevelopmental disorders. So we looked at stem cells to model changes in methylation during differentiation. Unfortunately, we found that embryonic stem cells and induced peripotent stem cells don't accurately reflect early life methylation. Early methylated domains, or PMDs, are a key feature of the early life epigenetic landscape. We used whole genome by sulfite sequencing, which assays methylation at almost every CPG in the genome to show that oocytes, pre-implantation embryos, and placenta all have PMDs covering 30 to 40% of their genomes. In contrast, embryonic stem cells and induced peripotent stem cells had no PMDs. So we decided to examine global methylation patterns in a different type of stem cell, specifically dental pulp stem cells, or DPSCs, which are derived from baby teeth that can be gathered from kids without invasive procedures like drawing blood or collecting tissue. If you take a look at this PCA plot from Figure 1C of the paper, you can see that ESCs and IPSEs cluster closer to the differentiated tissues of brain and liver than they do to early life tissues such as placenta. Interestingly, we found that DPSCs have a much closer methylation pattern to placenta than most other cell types we assayed. This led us to investigate DPSCs as a potential model for methylation in neurodevelopmental disease states. In particular, we looked at Duped 15Q syndrome, which occurs due to a duplication of part of chromosome 15. Specifically, Duped 15Q patients will have two extra copies of the maternal 15Q 1,1 to 1,3 region. This yields a 3 to 1 ratio of maternal to paternal alleles in Duped 15Q syndrome as opposed to the normal 1 to 1 ratio. Looking at Figure 3B, if you just focus in on the imprint in control region called the PWSIC, you can see how this allele ratio changes the methylation pattern. At the PWSIC, the maternal allele is methylated, while the paternal allele is unmethylated. So a typical individual, which has one copy of each allele, will have about 50% methylation. However, an individual with Duped 15Q syndrome usually has a 3 to 1 ratio, resulting in about 75% methylation. All of this epigenetic information has been retained in the DPSCs. In my other recent publication examining DNA methylation in postmortem Duped 15Q brain samples, we found a hypomethylated cluster over the entire imprinted locus. While the DPSCs are trending towards hypomethylation, they are not yet significantly decreased. However, we believe that the DPSCs model an earlier stage of epigenetic changes found in Duped 15Q syndrome. Specifically, if you look at the SNR116 region, it already has evidence of hypomethylation. Other support that DPSCs are a useful epigenetic model of Duped 15Q can be found in Figure 3e. These vendograms show that genes previously observed to be differentially methylated in Duped 15Q brain significantly overlap with those identified in Duped 15Q DPSCs. The overlapping gene lists include several known autism candidate genes and those with known functions at neural synapses, which are all indicated in Table 1 of the manuscript. In conclusion, our results show that DPSCs may be a useful model for investigating DNA methylation changes in autism spectrum disorders. At this point, I'd like to invite you to read the rest of the paper and draw your own conclusions. All of the sequencing data and code are freely available, and we look forward to hearing any questions or comments you may have.