Why are most brain cancers so difficult to treat leaving the unfortunate patients who suffer from them in dire circumstances so often?
The short answer is that we don’t know why.
This lack of understanding is like a wall standing between the cures and us for these patients.
We have to climb or knockdown this wall.
Brain cancers are a diverse group, but lessons about one or more types can sometimes teach us about others. There may also be patterns that transcend different types of brain tumors and may even include other kinds of cancers as well.
What are some established patterns that are at work in many kinds of brain tumors and indeed other tumors?
- Faulty DNA repair
- Impaired apoptosis
- Abnormally active or inactive signaling pathways
- Altered cellular metabolism
- Cell cycle
A relatively newer, but intriguing pattern emerging is related to a type of protein called histones.
What are histones?
Histones are DNA binding proteins that basically allow the cell to obtain information from the DNA. The presence or absence of histones has profound consequences for how DNA “talks” to cell behaviors. In addition histones are post-translationally modified (e.g. methylated, acetylated, phosphorylated, etc etc) in many ways creating a “code” that also influences DNA functions ranging from gene transcription to chromatin structure to chromosome segregation. All of these things operate like a histone-dependent machine in normal cells that controls most aspects of cell behavior via translating the information in DNA into a “language” that the cellular machinery can understand.
An emerging concept in the brain cancer field as well as in the study of other tumors is that monkey wrenches in histones also contribute to cancer.
While histones have been known for a long time to be associated with cancer, only recently have we obtained a clearer understanding of the specifics, although may open questions remain. In the case of a tumor that arises in the back of the brain in the cerebellum in children called “medulloblastoma” a seminal 2009 paper from Michael Taylor’s awesome lab provided evidence of convergent histone methylation pathways altered in medulloblastoma (see table above from their paper outlines these convergent genetic events). There are many other papers on the role of histones and epigenetics more broadly in cancer.
Just this year histones were implicated in brain cancer at another level with a handful of very exciting papers on the role of an interest histone variant called H3.3 in glioblastoma, the most deadly of brain cancers (see below).
The papers indicate that histone H3.3 is mutated on specific residues preferentially in pediatric glioblastoma. The molecular mechanisms involved are still being worked out, but this is a fascinating development.
Taken more generally, these studies and so many more raise the possibility of an even larger scale convergence of cancerous events that involve histones in brain cancer and in many other cancers.
This is fascinating stuff and adds some context to the lively debate about functional genes post ENCODE. Linking DNA, histones, proteomics and the key systems that seem to break down in cancer. Environmental factors damaging these systems over time trigger communication with the genome altering their output which in turn alters the proteomic systems responsible for function.
It is in proteomic systems where the impact of cancer manifests in the nuts and bolts machinery that drive biological function. These systems are complex and governed by the rules of complex systems science. This would seem the logical starting point rather than genetics to establish how these systems function; the role of histones and then work back to the genetics if indeed that is necessary.
Before now this was not possible but this is where the field of network pharmacology enters providing tools that can interpret and analyse proteomic systems and cope with their complexity for the first time.
Lets hope the promise illustrated in the this article and the correlation of histone differences with brain cancers does not get lost in the long grass of an obsession of assuming correlating genetic differences are cancer causes or the logical thread to cancer treatment. Alternatively analysis using pathway biology techniques incapable of interpreting the networks within complex proteomic systems.