Long Interspersed Nuclear Elements 1 (LINE-1) retroelements hypomethylation in aging and cancer is a central issue of my group’s research. A long-standing open question is how LINE-1 hypomethylation might arise in urothelial cancer (UC). A clue came from our observation that DNA methylation of genes like ODC1, AHCY and MTHFR was altered in early UC tissue samples. Lars Erichsen, at that time a grad student in my group, pointed out that the enzymes encoded by these genes influence methyl group and polyamine metabolism. Based on these findings we developed and published in our first article the PrimeEpiHit hypothesis (Erichsen et al. Scientific Reports, 2018). This hypothesis postulates that altered DNA methylation of methyl group and polyamine metabolism genes disturbs methyl group metabolism in the proliferating early UC cells leading to LINE-1 hypomethylation, reactivation, DNA strand breaks, genomic instability and thereby tumor progression. To validate this idea, we decided on investigating the consequences of limited ODC1 activity. As an optimal model system, we could use primary uroepithelial cells for our experiments, since the methodology to culture these cells is well established at our university in the Bladder Cancer Research Group in the urology department, led by Michèle Hoffmann and Günter Niegisch. Indeed siRNA-mediated ODC1 interference in these cells rapidly leads to LINE-1 hypomethylation and further hallmarks of UC as described in our present paper (Erichsen et al. Scientific Reports, 2020). Our findings hence may help to solve a long-standing enigma. In addition, our experimental approach allowing to induce LINE-1 hypomethylation at will in vitro is also applicable to address the question how LINE-1 hypomethylation influences the initiation and progression of other cancer entities like colorectal, lung, prostate, breast, hepatocellular and gastric carcinoma, chronic lymphocytic leukemia and others. Furthermore, manipulating ODC1 activity is well known as a potential means to specifically target cancer cells (Fig. 1). Finally, I wish to warmly thank my close and reliable collaboration partners at the university hospitals in Basel, Düsseldorf, Curitiba, Zürich and at the Biodonostia Health Research Institute bioinformatic center, Spain, for their crucial contributions. Not least, we thank the scientific reports team for the excellent handling of our manuscript.
You may also be interested in...
Machine learning on H&E slides could help identify patients that would benefit from additional therapy after surgery
Deep learning for diagnosis of Acute Promyelocytic Leukemia via recognition of genomically imprinted morphologic features