DNA methylation is the most studied epigenetic modification and occurs at the CpG dinucleotide sequence. CpG methylation can regulate gene expression through its effect on chromatin state, as well as interfere with accessibility of transcription factor binding sites. It is thought that hypermethylation of CpGs, especially when they are located within the promoter region or closed to transcription starting site, is associated with silencing of the gene, whereas hypomethylation has the opposite effect. DNA methylation has an important role in processes such as tumorigenesis and changes in methylation can contribute to genetic instability as well as aberrant gene silencing including inactivation of tumor-suppressor genes. DNA methyltransferases (DNMTs) are responsible for the transfer of a methyl group from the universal methyl donor, S-adenosyl-methionine (SAM), to carbon-5 (C5) of cytosine. In mammals, three families of DNMTs have been shown to be involved in DNA methylation. DNMT1 is responsible for the maintenance of genomic DNA methylation patterns; it requires a hemimethylated DNA to propagate existing methylation patterns from the old strand to the newly synthesized strand. DNMT2 transfers methyl groups to RNA. The third family represents the de novo methyltransferases consisting of DNMT3A, DNMT3B and DNMT3L. DNMT3A and DNMT3B possess methyltransferase activity and are responsible for the establishment of de novo DNA methylation patterns. DNMT3L, a closely related homologue of DNMT3A/B that lacks methyltransferases activity, and recruits the methyltransferases to specific DNA regions by binding to the unmethylated lysine 4 of histone H3 in the nucleosome. DNMT3A and DNMT3B are important for mammalian development, since loss of DNMT3A results in lethality at 4 weeks age loss of DNMT3B is embryonically lethal (Okano et al. 1999). It has been shown that deletion of DNMT3A inhibits the earliest stage of intestinal tumor development in Apc mouse model (Weis et al. 2015). In contrast, somatic inactivation mutations in human DNMT3A are found in acute myeloid leukemia, which suggests a tumor suppressor role. Furthermore, genetic variants in human DNMT3A and –B loci have been associated with an increased risk of inflammatory bowel disease (Franke et al. 2010, Jostins et al. 2012). Several studies suggest that epigenetic mechanisms including DNA methylation might have a crucial role in IBD pathogenesis (Ventham et al. 2013). Furthermore, genome-wide DNA methylation profiling of colonic intestinal epithelial cells, from children diagnosed with IBD, revealed altered DNA methylation patterns suggesting a role for DNA methylation (Kraiczy et al. 2015). However, the exact mechanisms that drive changes in methylation in IBD remain elusive. The interplay of epigenetic processes and the intestinal microbiota may play an important role for intestinal development and homeostasis. A complex and dynamic union of microorganisms inhabits the mammalian gastrointestinal tract and contributes to several aspects of host physiology including metabolism, maturation of the immune system, cellular homeostasis and behaviour. Previous studies have established that the microbiota regulates a large proportion of the intestinal epithelial transcriptome in the adult host, but microbial effects on DNA methylation and gene expression during early postnatal development are still poorly understood. A conditional knockout mouse has been generated that lacks Dnmt3a in IECs and baseline characterization has been performed. We hypothesize that DNMT3A and B may play an important role in the lifelong stability of DNA methylation patterns in IECs under physiological conditions. The previous project has mainly investigated the function of the respective isoforms in vitro and has described important functions in gene regulation beyond DNA methylation. The subsequent project will focus on the effects of DNMT3A in intestinal epithelial cells in in vivo models of intestinal inflammation and carcinogenesis.
The applicant´s group focuses on the identification of targets genes by silencing of DNMT3A and –B and functional follow-up study analysis. We performed RNA sequencing and 850k methylation chip array in Caco-2 cells transfected for 72h with siRNA targeting DNMT3A and a non-targeting control
siRNA. Analysis of the RNA-seq data revealed, that approximately 1000 genes were differentially expressed when DNMT3A was knocked-down. Surprisingly, we observed no difference in DNA methylation between the two groups. Gene ontology analysis of the differentially expressed genes revealed several functions that relate to cell adhesion and repair, hinting to a possible function in relation to IBD pathogenesis.
Since intestinal microbiota plays an important role in intestinal homeostasis, DNMT3A expression levels were analysed in germ-free and conventional-raised mice at week 1,4,12 and 16. We detected generally higher levels of Dnmt3a during W1 compared to W4 or W12/16 and increased expression in CONV-R compared to GF mice intestinal epithelial cells. These data suggest a link between microbiota and DNMT3A mediated methylation.
3-year research plan for the doctoral student
Year 1 – In vivo work, baseline characterization of inducible KO Dnmt3A mice: For this purpose we will breed floxed conditional Dnmt3A mice with inducible villin-Cre-ERT2 mice, where recombinase expression can be induced by tamoxifen treatment. Tamoxifen will be added to the drinking water in three experimental groups: (i) directly after weaning, (ii) at 4 weeks of age and (iii) at 16 weeks of age representing the infant, juvenile and adult state. 5 animals will be investigated 1 week and 8 weeks after induced deletion. Histological appearance, epithelial gene expression patterns, 850k EPIC chip array and 16s rDNA microbiome analysis will be analyzed.
Year 2 – DSS- inflammation/CAC intervention experiments – Depending on the outcome of experiment 1 we will select appropriate conditions (e.g. adult time point, constitutive or inducible Dnmt3a KO) and focus on the effect of de novo DNA methylation on intestinal inflammation and inflammation-induced colorectal cancer. After induced deletion, we will perform acute and chronic DSS colitis, which represents a classical model for intestinal epithelial injury. Mice of the respective genotypes will be subjected to either a singular one week treatment with 2.5% DSS in the drinking water (acute colitis) or 3 repeated 5d cycles of 1% DSS in drinking water to induce chronic colitis followed by distilled drinking water for 2 weeks (group size n=5, males). The protocol for AOM/DSS tumor induction is similar to the chronic DSS model with an additional injection of azoxymethane (AOM) 12.5 mg/kg 5 days before first DSS administration. The control genotypes receive distilled water only. The consumption of DSS solution per cage will be measured throughout the experiment, and the fluid will be changed daily. Fecal samples will be collected before treatment and after sacrifice. The intestinal tract, mesenteric lymphnodes, spleen and liver will be carefully dissected and partially stored frozen (intestinal epithelial scrapings and whole tissue) or fixated for histological analysis. RNA levels of epithelial cytokines (e.g. IL-8, TNF-), antimicrobial peptides, pro-regenerative factors (e.g. TFF3, Reg3b) and candidates derived from the in vitro experiments will be determined by qPCR in induced and non-induced mice. Global patterns of DNA methylation in IECs will be derived by 850k EPIC chip array in selected samples (e.g. only chronic DSS) depending on outcome.
Year 3 – High risk agenda: Dnmt3a ko mice in GF vs Conv-R condition or NASH model – Dnmt3A ko mice (constitutive or inducible) will be raised either in the presence or absence of a microbiota. Fecal samples of mice at different weeks of ages (week1, week 4 and week 16) will be collected. Body weight, histological appearance, epithelial gene expression patterns, 850k EPIC chip array and 16s rDNA microbiome analysis will be analyzed. Alternatively, a model of diet-induced liver inflammation (NASH, non-alcoholic steatohepatitis) could be employed, as the inflammatory tone of the intestinal epithelial layer is known to potentially modify hepatic inflammatory responses.