Analyzing Enhancer Expression In Pan-Cancer Patients
A human body contains about 37 trillion (3.7 X 1013) cells  that originated from a single zygote. This developmental process is precisely controlled by thousands of genes to ensure each cell follows the proper cell fate during replication.
However, no machine has a zero failure rate. When this cellular machinery is disrupted, cells might lose control and grow unlimitedly which eventually leads to cancer. The huge number of cancer-related genes make the pathology extremely complicated. For example, tumors with the same histology can have distinct mutations and respond differentially to a given therapy, rendering precision medication highly desired.
In order to fully understand this molecular complexity in tumorigenesis, the Cancer Genome Atlas (TCGA) project was launched in 2005 to characterize >11,000 tumors of 33 common cancer types at six molecular levels, including DNA mutation, gene copy number aberration, DNA methylation, RNA transcription, protein expression, and microRNA expression. However, a lot of molecular mechanisms has recently emerged as critical contributors to tumorigenesis since the project was designed, such as chromatin modification, stromal environment, and enhancer activation. Thanks to the advancements in sequencing technology and bioinformatics, researchers have been able to decipher a lot of these missing information from the TCGA dataset.
In a recent publication in Cell, researchers from the University of Texas MD Anderson Cancer Center developed a bioinformatic pipeline to capture the enhancer activities from the RNA transcription data format of the TCGA project . Enhancers are a group of regulatory DNA sequences without protein products. In response to biological signals, an enhancer usually approaches to its targeted gene physically and regulate the gene’s transcription. Thus, enhancers controlling the expression of cancer-related genes can significantly contribute to tumorigenesis. By analyzing ~16,000 enhancers across >9,000 TCGA patients of 33 cancer types, the researchers obtained a global landscape of enhancer activation in cancers. The major findings are summarized below.
Firstly, although the thousands of TCGA tumors were highly heterogeneous in terms of tissue of origins and tumor grades, their global enhancer activation patterns surprisingly fall into only three categories representing the diverse mutational mechanisms determined by chromatin organization of the cancer cells (Figure 1; This is the Figure 2I of the original publication). About 10% of the tumors are normal-like. They had low mutation rate, low gene copy number aberration, and little abnormally activated enhancer. 40% of the rest tumors were driven by hypermutation. These tumors usually had low copy number aberration and low enhancer activation levels. The last category showed a strong pattern of enhancer activation associated with the pervasive copy number changes across the whole genome.
Secondly, thousands of enhancers were found to be associated with the patients’ clinical outcome or disease subtype. The proportion of enhancer with prognostic power is comparable to that of the protein-coding genes, indicating the overall significance of enhancers during tumorigenesis. What’ more, the majority of these enhancers tend to have recurrent prognostic power in multiple cancer types. For example, an enhancer of the gene SYK consistently shows strong associations to patient-survival-time of up to six different cancer types, making it a promising prognostic indicator.
Thirdly, dozens of causal relationships between enhancer activation and key cancer gene misregulation were discovered in this study, including the gene PD-L1 which is wildly expressed by cancer cells to disguise themselves as normal cells to escape from the immune system. Immunotherapy masking PD-L1 molecules on the surface of the cancer cells can efficiently help the immune system to eliminate these cancer cells. When the enhancer of PD-L1 was genetically disrupted, the expression level of PD-L1 substantially decreased by nearly 90% in a human cultured cell line.
These findings are described in the article entitled A Pan-Cancer Analysis of Enhancer Expression in Nearly 9000 Patient Samples, recently published in the journal Cell. This work was conducted by Han Chen, Chunyan Li, Xinxin Peng, Zhicheng Zhou, and John N. Weinstein, and Han Liang from The Cancer Genome Atlas Research Network.
- Bianconi, E., et al., An estimation of the number of cells in the human body. Ann Hum Biol, 2013. 40(6): p. 463-71.
- Ding, L., et al., Perspective on Oncogenic Processes at the End of the Beginning of Cancer Genomics. Cell, 2018. 173(2): p. 305-320 e10.
- Schuster-Bockler, B. and B. Lehner, Chromatin organization is a major influence on regional mutation rates in human cancer cells. Nature, 2012. 488(7412): p. 504-7.
- Liu, X.S. and E.R. Mardis, Applications of Immunogenomics to Cancer. Cell, 2017. 168(4): p. 600-612.
- Chen, H., et al., A Pan-Cancer Analysis of Enhancer Expression in Nearly 9000 Patient Samples. Cell, 2018. 173(2): p. 386-399 e12.
- Chen, L. and X. Han, Anti-PD-1/PD-L1 therapy of human cancer: past, present, and future. J Clin Invest, 2015. 125(9): p. 3384-91.