Tumor-selective coronaviruses as potential anti-cancer agents

Abstract

Despite much progress in the treatment of certain types of cancer, generally the successes of cancer therapy are still largely insufficient and new treatment options are therefore urgently needed. Oncolytic virotherapy may offer an unconventional approach to selectively eradicate cancer cells, while leaving the normal tissues essentially unaffected. Several viruses are currently being analysed for their capability to kill cancer cells. So far, coronaviruses had not been explored for use in cancer virotherapy, despite the several features that make them attractive for this purpose. The aim of the research described in this thesis - made possible by the support of the Dutch Cancer Society (KWF Kankerbestrijding) - was to explore whether the non-human coronaviruses FIPV and MHV have the potential to function as oncolytic agents.

Because FIPV and MHV cannot infect human cells due to the absence of the natural receptors for these viruses, we first examined whether these coronaviruses have the potential to kill human cancer cells once the entry barrier is alleviated by artificially expressing the appropriate receptor on their surface (Chapter 2). The results show that FIPV and fMHV efficiently kill cancer cells, once they gain access to these cells. In the subsequent chapters we focussed on the targeting of the non-human coronaviruses to receptors present on human cancer cells. We aimed at indirect targeting by using bispecific adapter proteins to direct the viruses to specific receptors. In chapter 3 we generated bispecific single-chain antibodies that bind on the one hand to the FIPV spike protein and on the other to the human epidermal growth factor receptor (EGFR). Chapter 4 comprises the analysis of adapters based on the ectodomain of the MHV receptor fused to a cancer cell receptor-binding ligand. Altogether chapters 3 and 4 show that FIPV, fMHV, and MHV can be targeted to the human EGF receptor via different bispecific adapter proteins, resulting in infection of human cancer cells. A limitation of the indirect targeting approach in the context of oncolytic virotherapy is that the infection is dependend on added adapter. To avoid this inconvenience, the adapter genes were incorporated into the MHV genome to obtain genetically targeted coronaviruses (Chapter 5). Indeed, an MHV expressing a His-tag-adapter protein proved to be self-targeting: it infected human target cells expressing a His-tag receptor, spreading through the culture and efficiently killing the cell monolayer. In Chapter 6 we explored the use of antibodies to redirect viruses to Fc receptors expressed on certain types of acute myeloid leukemia (AML). Upon combining with certain antiviral antibodies, MHV gained access to the otherwise refractory Fc-receptor-expressing leukemic THP-1 cells. This novel example of indirect targeting may have implications for the development of selective viral vector-mediated gene transfer approaches for the treatment of AML.

In conclusion, animal coronaviruses can be retargeted in different ways to human cancer cells and are able to infect and destroy such cells in vitro. The next steps will be to manufacture non-human coronaviruses targeted to relevant tumor targets and to analyze their oncolytic capacity in appropriate in vivo tumor models. At this point we would like to propose coronavirus as a potential novel oncolytic agent.