A sugar rush can fuel many things. It can power the late-night experiments demanded by reviewer number 3 or it can drive tumor evolution. Fueled by both these factors, new insight into the linked metabolic-epigenetic mechanisms of metastasis comes at you from a collaborative effort led by the lab of Andrew Feinberg in the Center for Epigenetics at Johns Hopkins University School of Medicine.
Pancreatic ductal adenocarcinoma (PDAC) is a highly lethal cancer that is all too common. The mechanisms of its metastasis remain a mystery since both primary tumors and far-flung secondary (metastatic) tumors share the same driver mutations. By examining matched primary and metastatic PDAC lesions, the team was able study tumor evolution in extensively sequenced samples that had no metastasis-specific driver mutations. Thus, they had the perfect samples to study the epigenomic mechanisms of cancer’s spread, which they confirmed in cell lines.
Immunohistochemistry of histone post-translational modifications (PTMs) revealed global epigenomic reprogramming during the evolution of distant metastases. The team then characterized the specific locations of the modifications by launching an integrated assault on the epigenomic landscape. They used ChIP-seq to analyze histone PTMs (H3K9me2, H3K9me3, and H3K27me3 as marks of heterochromatin and H3K27ac and H3K36me3 as marks of euchromatin), whole-genome bisulfite sequencing (WGBS) for DNA methylation, and RNA-seq for gene expression.
Here’s what they found:
- The epigenomic reprogramming is targeted to thousands of large-scale euchromatin and heterochromatin domains.
- The large heterochromatin domains are large organized chromatin (H3)K9-modications (LOCKs) domains and their partially overlapping large DNA-hypomethylated blocks.
- They also came across a familiar and interesting hybrid feature, resembling LOCK euchromatin islands (LOCK-EIs). There are highly localized reciprocal changes to H3K27ac and H3K9me2, within promoters, coupled to similar reciprocal changes of H3K36me3 and H3K27me3, in gene bodies, within LOCK genes.
- The team also found reprogramming in a unique subset of very large LOCK domains.
- A connection between metabolism and histone modifications, where distant metastases co-evolve a dependence on the oxidative branch of the pentose phosphate pathway (oxPPP).
- This lets tumors binge on glucose to feed the metabolic sweet tooth that is critical to their growth in their new environment.
- This mechanism was confirmed by inhibiting phosphogluconate dehydrogenase (PGD), a key enzyme in the oxPPP, either with RNAi or pharmacologically, which reversed the reprogrammed chromatin and gene expression, as well as the tumorigenesis.
Overall, these findings detail a non-genetic form of natural selection critical to tumor evolution, where linked metabolic-epigenetic programs are selected for and allow distant metastases to take advantage of their new glucose rich environments.
Offering a humble peek behind the scientific curtain, Feinberg concludes, “I would actually like to say something about peer review. We all complain about it but we are also the reviewers. In this case, the amount of work and resources doubled in response to reviewers’ criticisms, but in the end I thank them for that. The additional work made a much more convincing case that the epigenetic changes drive distant cancer metastasis and that they are linked to a metabolic pathway that we might be able to manipulate therapeutically, increasing the chances that this work will benefit patients’ lives.”
Catch all the evolutionary links over at Nature Genetics, January 2017