Functionality of Risk Loci for Hematological Malignancies

Genome-wide association (GWA) studies have been instrumental in identifying genetic variation influencing cancer risk. GWA studies conducted on lymphoid leukemias by the "Host Research Group" have identified ten risk loci for chronic lymphocytic leukemia (CLL) and four for acute lymphoblastic leukemia (ALL). All identified risk alleles were shown to confer a small effect on CLL and ALL risk, but significantly impacting on tumor risk in concert. Elucidating the genetic and biological basis of specific associations is challenging, as the single nucleotide polymorphisms (SNPs) shown in GWA studies to be associated with cancer risk are generally not directly causal but are correlated with functional variants. Moreover, several risk loci map to gene deserts, lacking identifiable genes or predicted transcripts in closer proximity. Chromosome Conformation Capture (3C) experiments are increasingly providing evidence that differential short or long range enhancer-promoter interactions functionally underlie disease associations. While 3C suffers from the limitation that only interactions that have been considered a priori will be detected, the proposal is to develop a truly agnostic method to detect genome-wide chromatin interactions with the risk loci as predefined test regions at high resolution. This will be achieved by combining HiC, a second-generation version of 3C, with a sequence capture step targeting genomic regions defining associations. Functional interactions between regulatory elements and genes will be subsequently validated using conventional 3C. Combining data from genome-wide interactions with conventional fine-mapping of the association signals, in- silico prediction of regulatory elements and functional studies including reporter gene assays will allow the mechanism underlying the 14 leukemia risk loci to be established. During the "visiting phase" of the project, the method will be set up in the "Host Laboratory" and subsequently applied onto the 14 leukemia risk loci in ALL and CLL cell lines and primary patient material. The "return phase" will be crucial for applying the technique on investigating the basis of inherited risk for Myeloproliferative Neoplasms (MPN), a group of myeloid disorders the "Home Research Group`s" work is dedicated to. The strategy of mapping genome-wide chromatin interactions with risk loci for hematological malignancies in an unbiased way, followed by in-silico prediction of regulatory elements and functional studies will give critical insight into the mechanistic effects of risk variants for lymphoid and myeloid leukemias.

Single-letter genetic variations within parts of the genome can increase cancer risk through effects on distant genes. DNA sequences within gene deserts (regions devoid of genes) can regulate gene activity elsewhere by forming DNA loops across large distances. In recent years genome-wide association studies (GWAS) have identified inherited risk loci on the DNA for many types of cancers. These risk loci are often located in gene deserts and typically comprise DNA variants commonly present in the population. Causal variants within the loci impact on regulatory elements. It is now recognized that such regulatory elements can influence the expression of multiple target genes by forming DNA loops across relatively large distances. Therefore, elucidating the biological basis of a cancer-associated DNA locus by revealing its target genes is challenging. Physical interactions between regulatory elements at risk loci and their target genes can be identified by methods based on chromosome conformation capture (e.g. 3C, 4C, 5C, Hi-C). Since multiple targets are possible for multiple elements within a single cancer risk locus, an agnostic method of detection is required. In the visiting phase of the Erwin-Schroedinger-Fellowship, we developed a novel enhancement of Hi-C using target DNA sequence enrichment, capture Hi-C (cHi-C), allowing for state of the art characterization of chromatin interactomes (collectivity of all detectable interactions) at high resolution. In a pilot study we applied cHi-C to 14 risk loci for bowel cancer to depict long-range interactions with the loci. Furthermore, we integrated our interactome data obtained from cHi-C with publicly available information on regulatory features at risk loci and target elements, allowing us to gain insight into the functional basis of association at these loci. We subsequently expanded our chromatin interaction studies to 14 leukemia risk loci. As initially proposed for the return phase of the fellowship, we applied the methodologies and workflows established during the visiting phase for research on inherited risk for myeloproliferative neoplasms (MPN), a group of chronic blood cancers. Besides expanding population-based studies for discovery and characterization of genetic associations in MPN, we initiated functional studies on MPN risk loci. Depiction of cHi-C chromatin interactomes for six genomic loci implicated in MPN pathogenesis represents the core of these studies. The projects mentioned herein have the potential to exert clinical impact on prognosis and treatment of bowel cancer, lymphoid blood cancers and MPN. Knowledge on genetic susceptibility can influence clinical management and therapy. Moreover, identification of genes and pathways implicated in cancer represents the basis for developing specific drugs targeting the involved pathways.