UCSF

High mobility group box 1 (HMGB1) is a nuclear protein first discovered nearly 30 years ago. It was described to aid in transcription by forming complexes with transcription factors. Further studies demonstrated that HMGB1 has the unique ability to act as an extracellular ligand for toll-like receptor (TLR) family proteins and the Receptor for Advanced Glycation End products (RAGE) when released from necrotic cells. Extracellular HMGB1 activates the NFkappaB and MAPK signaling pathways through these receptors. It is thought that a primary function of HMGB1 is to act as an "alarmin" or danger signal to activate immune cells in the presence of tissue damage. The prevailing model has suggested that HMGB1 overexpression in cancer is due to its ability to activate MAPK signaling, an important pathway in several cancers.

We have investigated MAPK-independent functions of HMGB1 in cancer by utilizing cell lines with activating mutations in this pathway in combination with an shRNA-mediated knockdown approach. Using this technique, we have discovered a previously unknown role for HMGB1 in promoting tumor cell growth and migration. We show that HMGB1 remodels the SEMA3A genomic locus to a more silenced state involving histone deacetylation and decreased CCCTC-binding factor (CTCF) occupancy, and thus promotes tumor cell migration.

In another line of investigation, we have studied the mechanism of release of HMGB1 during Adenovirus infection. Oncolytic viruses are viruses that have been engineered to infect and replicate specifically in tumor but not normal cells. One of the greatest challenges in the field has been the immunogenicity of the viruses. A healthy immune system can eliminate the virus before it is able to spread throughout the tumor. Because HMGB1 is a critical component of immune activation, we investigated how Adenovirus induces its release. We found that poly-ADP-ribose polymerase (PARP) was required for the release of HMGB1 during Adenovirus infection.

We have discovered novel roles for HMGB1 in tumor biology and helped uncover modes of its regulation during Adenovirus infection. It is hoped that these findings will aid in the design of new HMGB1-targeted cancer therapies as well as increase the efficacy of existing oncolytic virus therapies.