Novel interplay involving the histone demethylase LSD1 in vivo – LSD1

We are using Drosophila as a model to study the role of dLsd1 in chromatin homeostasis and gene expression in vivo. We took advantage of a library of transgenic RNAi lines available at Harvard Medical School to set up a screen to look for novel genetic interactors of dLsd1. An advanatge of using transgenic RNAi is the possibility to disrupt the activity of genes with a spatial and temporal resolution difficult to achieve using classical genetic methods. Using this library, we had the possibility to screen hundreds of genes in an unbiased way in the context of a whole organism and we found several novel dLsd1 genetic interactors. We are now deciphering the interplay between dLsd1 and its interactors using a biochemical approach. In addition, we are planning to perform a highthrough-put analysis to identify where dLsd1 binds in the genome and how this binding is affected by its genetic interactors. Importantly, we will do these experimentsin vivo in specific developmental contexts, such as ovaries and wing discs.

The genetic screen gave me the opportunity to discover novel and exciting modulators and downstream targets of dLsd1, potentially expanding the understanding of the functions of these enzymes in cancer. The next step now is to integrate dLsd1 role with that of its interactors. These studies will provide a detailed understanding of when, where and how dLsd1 is recruited to specific loci in specific developmental contexts. Importantly an appreciation of these epigenetic mechanisms in flies will aid our understanding of parallel phenomena in humans and how their deregulation can lead to cancer development. We are indeed planning to test the conservation of these mechanisms in humans in collaborations with other researchers working on different types of cancers.

LSD1 role in cancer cells appears to be highly context dependent, therefore, a careful systemic analysis of the different biological roles of LSD1 is needed, especially considering its enormous potential of being used as a drug target.The experiments proposed are aimed at gaining a better understanding of the function of dLsd1 in vivo and might provide key insights for the development of selective epigenetic strategies to kill cancer cells.

Di Stefano L and Dyson N. The emerging roles for histone demethylases in the modulations of signaling pathways. Biomolecular Concepts. 2013; 4(1): 13–27

Di Stefano L and Dyson N. The complex roles of histone demethylases in vivo. Cell Cycle. 2011 Jul 1;10 (13): 2049-50.

Mulligan P, Yang F, Di Stefano L, Ji J, Nishikawa J, Wang Q, Kulkarni M, Najafi-Shoushtari H, Mostosvalsky R, Gygi S, Gill G, Dyson N and Näär A. A SIRT-LSD1 co-repressor complex regulating Notch target gene expression and development. Mol Cell. 2011 Jun 10;42(5):689-99.

Di Stefano L, Walker J, Burgio G, Corona D, Mulligan P, Näär A and Dyson N. Functional antagonism between histone H3K4 demethylases in vivo. Genes and Development. 2011 Jan 1; 25 (1): 17-28.

The modENCODE Consortium. Identification of functional elements and regulatory circuits in Drosophila by large scale data integration. Science. 2010 Dec 24; 330 (6012):1787-97.

Morris E, Ji J, Yang F, Di Stefano L, Herr A, Moon NS, Kwon EJ, Haigis KM, Näär A and Dyson N. E2F1 represses beta-catenin transcription and is antagonized by both pRb and CDK8. Nature. 2008 Sep 25; 455(7212): 552-6.

Moon NS, Di Stefano L, Morris E, Patel R, White K and Dyson N. E2F and p53 induce apoptosis independently during Drosophila development but intersect in the context of DNA damage. PLoS Genet. 2008 Aug 8;4(8).

Di Stefano L, Ji J, Moon NS, Herr A and Dyson N. Mutation of Drosophila Lsd1 disrupts H3-K4 methylation resulting in tissue specific defects during development. Curr Biol. 2007 May 1; 17(9): 808-12.

Qiao H*, Di Stefano L*, Tian C, Li YY, Yin YH, Qian XP, Pang XW, Li Y, McNutt MA, Helin K, Zhang Y, Chen WF. Human TFDP-3, a novel DP protein, inhibits DNA binding and trans-activation by E2F. J Biol Chem. 2007 Jan 5; 283 (1) 454-66. *these authors equally contributed to the paper.

Histone modifications are instrumental for the control of gene expression and chromatin architecture and a strict regulation of histone marks is critical for normal development. Defects in chromatin organization and gene expression due to aberrant activity of chromatin modifying enzymes have been implicated in tumorigenesis. Our current knowledge regarding histone methylation largely stems from the study of histone methyltransferases, and only recently the discovery of histone demethylases showed that histone methylation is reversible supporting a more dynamic role of histone methylation. The first histone lysine demethylase to be identified was LSD1/KDM1. Since its discovery, LSD1 is emerging as a key regulator of chromatin. Although its activity is potentially widespread, evidence suggests that LSD1 performs pathway-specific functions. In addition, several studies have implicated LSD1 in tumorigenesis and there is a growing interest in LSD1 as a drug target. However, despite the tremendous progress in understanding the dynamic regulation of histone methylation by LSD1, the biological function of LSD1 is just beginning to be uncovered. We still have a limited knowledge of LSD1 pathway-specific functions and a genome-wide study of LSD1 role in a whole organism is lacking.
I have taken advantage of the streamlined network of proteins present in Drosophila, to demonstrate that dLsd1 activity is conserved in flies and to find novel functions of dLsd1 during development. Using a genetic approach, I found a surprising antagonism between the two histone demethylases dLsd1 and Lid. Investigation of the basis for this antagonism revealed that Lid opposes the functions of dLsd1 in promoting heterochromatin spreading. Moreover, these data unveiled a role for dLsd1 in Notch signaling in Drosophila. In the proposed work (Task 1), I plan to take advantage of the vast array of tools available in Drosophila to identify novel pathways involving dLsd1. Specifically, I am carrying out an in vivo RNAi screen and I have already found several enhancers and suppressors of a dLsd1 loss of function phenotype. This approach has the key advantages of being unbiased and to take place in the context of a whole animal. This is crucial because it could unveil context dependent interaction important for development, which might be lost in cells grown tissue culture. The screen will be followed by a detailed functional characterization of key genetic interactions, with a special focus on transcription factors, chromatin regulators and proteins involved in signaling and RNA metabolism.
In the second approach (Task 2), I will use ChIP-Seq technologies and transcriptome analysis to extend the characterization of dLsd1 mutant flies. The goal is to understand how dLsd1 and its interactors control gene transcription and chromatin homeostasis in specific developmental contexts.
An important aspect of this strategy is the idea that the mechanisms that enhance or suppress dLsd1 mutant phenotypes are conserved between flies and mammals. In a first attempt to verify this, we already found that this is true for the interplay between dLsd1 and the Notch pathway. In Task 3, I plan to take a few new key interactions discovered in aim 1 and to test whether this newly acquired information is conserved in mammals and can be exploited to target tumor cells. I will start by investigating the role of the human orthologs of dLsd1 and Lid in heterochromatin formation in mammalian cells. Heterochromatin is deregulated in many human diseases, including cancer, but a detailed understanding of how this state is established and maintained at specific loci and how its misregulation contributes to tumorigenesis is still missing. The experiments proposed are aimed at gaining a better understanding of these processes and might provide key insights for the development of selective epigenetic strategies to kill cancer cells.