Histones in cancer pathogenesis: defining the role of H3.3 mutations in bone and cartilage tumours

Lay Summary

Two group leaders in UCL Cancer Institute, Paolo Salomoni who works on brain tumours, and Adrienne Flanagan who has worked on bone tumours and other bone diseases for 20 years, have joined forces because they and others have shown that the same genes are mutated (altered) in brain and bone tumours. One family of such mutated genes is known as histones. DNA is packed in the nucleus and this structure is known as chromatin. Histones are involved in coding for the basic building blocks of chromatin. The recent finding that histones are mutated in cancer has changed the way that scientists think about how human cancer develops.

This proposal aims to define how histone mutations alter the normal development of the bone/cartilage and how these changes can lead to cancer. We also want to define similarities and differences in the mechanisms by which these mutant histones drive bone/cartilage and brain tumours. Overall, this work will increase our understanding of how cancer develops: it is likely that the information will have an impact beyond bone and brain tumours. The findings in turn will help identifying novel targets for drug development, which may be valuable for treating more than one tumour type.

Epigenetics are emerging as an important contributor to cancer development. Epigenetic reprogramming caused by IDH1 and IDH2 mutations have been implicated previously in brain and cartilaginous tumours. Recently, driver mutations in the histone 3.3 (H3.3) variant have been identified in over 50% of paediatric glioblastoma multiforme cases and we have also found that these mutations are found in ~60% of conventional central and dedifferentiated chondrosarcomas: the mutations in both these brain and bone tumours lead to modifications of the epigenetic landscape. The identification of direct alterations of the basic building blocks of chromatin represents a paradigm shift in our understanding of cancer epigenetics.
Very recently, we have also reported that H3.3 variants are mutated in primary central bone (H3F3A in 92% of giant cell tumours and H3F3B in 95% of chondroblastoma), thus suggesting that direct alterations of histones are more common than previously thought.

Several questions remain unanswered: are histone mutations transforming?
In which cell type(s) and at which developmental/differentiation stage(s) do the mutations exert their transforming potential?
What are the molecular changes caused by deposition of mutant histones?
Do H3.3 mutations work in part through a non-cell autonomous mechanism?
This proposal will address these questions. In particular, we aim to determine the biological and molecular consequences of bone/cartilage tumour-associated H3.3 mutations. Furthermore, we have set out how we will correlate the epigenetic and transcriptome changes imposed by H3.3 mutations across the different tumour types (bone and brain). Overall, this study will contribute to explaining the mechanisms by which the H3.3 mutations bring about the disease phenotypes. Our finding will provide opportunities for developing targeted therapies, which may be valuable for treating more than one tumour type.

Who will benefit from this research?
-Researchers in the field of cancer research
-Clinical scientists and clinicians in the field of oncology
-Brain cancer patients

Impact on basic research and translational potential

This study will increase our understanding of pathogenesis of bone/cartilage tumours and findings generated over the course of the study will potentially be relevant also for other tumours where mutations in histone genes and their chaperones are found. Specifically, it has the potential to define the transforming role of histone mutants in the osteogenic/chondrogenic lineages and provide cues into the underlying mechanisms in musculoskeletal health and disease. Furthermore, these discoveries may help understanding the function of H3.3 mutations also in glioblastoma multiforme, thus extending the impact of this work.

Our work could identify novel biomarkers for predicting disease behaviour which are needed if we are to improve disease outcome: ~50% of giant cell tumours of bone recur and provide clinicians with challenging treatment decisions. Although recurrence in some cases may be related to tumour size at presentation, and anatomical site, some tumours appear to behave in a more aggressive fashion ab initio but histological assessment does not allow objective prediction of such tumour.
Surgery is currently the main treatment for GCT and this can result in severe morbidity. Although there is recent evidence from clinical trials that a RANKL inhibitor (demosumab) shows good response for such tumours, this needs further testing. Furthermore, the mechanism by which this drug works is not entirely clear: our research will shed more light on this. Furthermore, this research will have a wider impact than bone tumours as this drug is also being used for metastatic carcinoma (breast and prostate) to bone.

The research could also identify targets for therapeutic intervention. It is conceivable that the timescale for transition to clinical studies could be relatively small. Finally, discoveries produced through this study can increase our understanding of fundamental mechanisms involving the regulation of chromatin function, transcription and genomic stability.

Beyond cancer, this research may also provide insight into the role of epigenetics in the normal development of the skeleton and thereby explain some of developmental disorders, and may also provide insight into the common diseases of bones and joints (osteoporosis and osteoarthritis) as there is evidence that a component of these disorders is brought about disturbed crosstalk between osteoclasts and stromal cells which express RANKL.

This study will also increase our understanding of brain cancer pathogenesis. Glioblastoma multiforme (GBM) is one of the first causes of cancer death in the active adult population and the first cause of cancer death in children. No cure is available and the median overall survival is below 1.5 years. Research in this field is urgently needed, as the societal as well as economical impacts are very much substantial.

Finally, by combining the research strengths of Salomoni and Flanagan, the funding will provide added value as the discoveries produced through this study can have an impact on our understanding of the pathogenesis of both brain and musculoskeletal cancers sharing similar genetic and epigenetic features.

Dissemination (see communication plan)