Mapping the Chromatin Landscape of Cancers
A betrayal of our own cells against us, cancer is a devastating disease that affects over 18 million people world-wide . With more than 200 different types of cancer, most with diverse modes of action and variable outcomes among people, a deeper understanding of how different cancers function and progress is vital . In 2006 the National Institute of Health (NIH) implemented The Cancer Genome Atlas (TCGA) as an effort to better understand the genetic basis of cancer though genomic analysis . Many regions of the genome that do not encode genes contain sequences that are important for regulating gene expression. A way to identify these regulatory regions is to ask how accessible the chromatin is. Having more open and accessible chromatin allows for the binding of transcription factors and other regulators, thus controlling the expression of the target genes of these regulatory elements. Changes or mutations in these regulatory regions can adversely affect cells and lead to cancer . Using samples from TCGA, Corces et al mapped the chromatin accessibility landscape of 23 different types of human cancers. In doing so, they not only identified global patterns of accessible chromatin in the cancers studied, but also found new regulatory elements and non-coding mutations associated with specific cancers that affect chromatin accessibility, providing deeper insight into the molecular mechanisms and altered regulation underlying specific cancers.
With 410 tumor samples, Corces et al performed an Assay for Transposase-Accessible Chromatin using sequencing (ATAC-seq)  across 23 different types of cancer. In addition to identifying the same regulatory elements that were found in earlier datasets like the Roadmap DNase-seq dataset , they also identified potential new regulatory sites. They were able to group the cancers into 18 distinct clusters based on similarities in their ATAC-seq profiles. Interesting observations were revealed in these 18 clusters including some cancers splitting into different clusters (such as basal and non-basal breast cancer), clustering together of cancers of different types but which originated from the same tissue, and cancers across different tissues clustering together perhaps due to similar cell types (like those for squamous cell types). Looking for patterns of chromatin accessibility that were specific to each cluster, they found that places of high chromatin accessibility correlated with an enrichment in transcription factor binding sites and known single nucleotide polymorphisms (SNPs) that are associated with that particular type of cancer, supporting the idea that these accessible regions are important regulatory regions playing crucial roles in carcinogenesis.
In a proof of principle, the researchers showed that with a large enough sample size of patients with a certain type of cancer, they could identify subgroups of patients with different chromatin accessibility profiles. Focusing on kidney renal papillary cell carcinoma (KIRP), they saw that one subgroup of patients had a high level of chromatin accessibility near the MECOM gene which correlated with its increased expression by RNA-seq. They saw that KIRP cancer patients that overexpressed MECOM had a decreased likelihood of survival. This finding indicates that these cancer chromatin accessibility profiles have the potential to identify subgroups of patients with different molecular mechanisms and outcomes within the same type of cancer.
Corces et al also took advantage of their deep ATAC-sequencing to ask where transcription factors were binding. By correlating gene expression with transcription factor footprinting, they identified both activating and repressive transcription factors in different types of cancer. This also enabled them to link regulatory elements with the genes they are predicted to regulate for 8552 protein-coding genes. They found that most regulatory elements were located one or more genes away from the gene they are predicted to regulate. They validated their findings through a CRISPRi approach that silenced the regulatory elements of BCL2 and SRC genes respectively and resulted in a decrease in expression of each of these genes. They were also able to do this for peak-to-gene links in immune cells versus cancer cells, which often evade the immune system.
Common causes of cancer are mutations that occur during the lifetime of the individual. To find mutations in regulatory elements, Corces et al combined their ATAC-seq results with whole genome sequencing data from 35 patients across 10 cancer types. Looking at bladder cancer, they found that one patient had a mutation upstream of the FGD4 gene that resulted in greater chromatin accessibility at that location and higher FGD4 gene expression than in patients without this upstream mutation. Differential motif analysis identified NKX transcription factor motifs as being the most enriched for binding to the mutated regulatory sequence. In fact, the mutant sequence created a completely new NKX2-8 transcription factor binding site, which likely leads to the increase in chromatin accessibility and thus gene expression of the FGD4 gene. Bladder cancer patients with higher expression of FGD4 had decreased survival compared to those with lower FGD4 expression, indicating that this mutation could be driving this particular type of cancer. This finding shows that by combining the authors’ new ATAC-seq data with whole-genome sequencing, cancer-causing mutations in regulatory elements can be uncovered leading to a deeper mechanistic understanding of different cancers.
The results of this study provide a rich resource for scientists studying cancer by providing a detailed map of chromatin accessibility across 23 different cancer types, which will hopefully lead to a more complete understanding of the drivers of specific types of cancer. Future cancer patients will benefit from this information through the development of precision medicine based on both their cancer type and their chromatin accessibility profile. This study takes us one step closer to understanding how our cells go rouge and gives hope for how to stop them.
Original article: Corces MR*, Granja JM*, Shams S, Louie BH, Seoane JA, Zhou W, Silva TC, Groeneveld C, Wong CK, Cho SW, Satpathy AT, Mumbach MR, Hoadley KA, Robertson AG, Sheffield NC, Felau I, Castro MAA, Berman BP, Staudt LM, Zenklusen JC, Laird PW, Curtis C; Cancer Genome Atlas Analysis Network, Greenleaf WJ, Chang HY (2018). The chromatin accessibility landscape of primary human cancers. Science, 362 (6413): eaav1898. DOI: 10.1126/science.aav1898. *contributed equally
 World Health Organization: International Agency for Research on Cancer (2018). All cancers fact sheet. Globocan 2018, http://gco.iarc.fr/today/data/factsheets/cancers/39-All-cancers-fact-sheet.pdf
 Buenrostro JD, Giresi PG, Zaba LC, Chang HY, Greenleaf WJ (2013). Transposition of native chromatin for fast and sensitive epigenomic profiling of open chromatin, DNA-binding proteins and nucleosome position. Nat Methods, 10 (12): 1213-8. DOI: 10.1038/nmeth.2688.