Hsp70 in immunotherapy : a potential vector in cancer and viral vaccines

Abstract: New efficient vaccines against cancers and infectious diseases are in demand. In 1993 heat shock protein 70 (hsp70) purified from tumour was shown to elicit tumour protection in mice. It was also shown that the protection was not mediated by the hsp70 itself, but by tumour-specific peptides bound to the molecule. In this thesis I have studied hsp70 in two different tumour systems and in one viral system to find out if this chaperone has the capacity to act as a useful vector in cancer and viral vaccines. The first tumour system studied was the methylcholanthrene-(MC-) induced fibrosarcomas. These tumours express tumour specific antigens that have not been characterised. T cell cultures from mice immunised with different MC tumours were tested for recognition of the tumours in vitro, as well as for recognition of peptides eluted from the different tumours. The recognition in vitro correlated with the tumour protection in vivo and was specific for each tumour. The recognition was mediated by CD8+ T cells, as depletion of CD4+ T cells or using T cells from CD4-/- mice showed the same pattern of tumour recognition. Hsp70 was purified from MC tumours or from liver tissue and used to immunise mice. Upon challenge with live tumour cells it was clear that mice immunised with hsp70 from tumours were significantly protected from tumour outgrowth. Mice immunised with hsp70 from liver tissue were not significantly protected, but the tumours in these mice grew slower than in the buffer control mice. This delay can at least in part be explained by the ability of hsp70 to activate macrophages as shown here. T cells from the tumour-purified hsp70-immunised mice were stimulated in vitro by the cognate tumour, but not by other MC tumours, to produce TNF-[alpha]. Three hsp70-immunisations failed to elicit antibodies to hsp70. Hsp70 was also purified from two human melanoma tumours and pulsed on monocytes. Specific T cell clones were incubated with the pulsed monocytes to study if tumour specific peptides could be presented by the MHC class I or II or the hsp70-pulsed monocytes. Four out of five clones were stimulated by the pulsed monocytes in a manner that was dependent both on the presence of antigenic peptides bound to hsp70 and on the expression of the restricting HLA molecule by the monocytes. It was not necessary however to have the hsp70 purified from a tumour expressing the T cell restricting HLA molecule, clearly indicating that hsp70 is mediating cross-priming. The last study was performed in the Lymphocytic Choriomeningitis Virus (LCMV) system, using recombinant hsp70. A known LCMV specific CTL epitope was mixed with hsp70 in vitro under peptide binding conditions. Mice immunised with this mixture and challenged with virus had 10-100-fold lower viral titres after infection, and CTLs from these mice could kill peptide-pulsed target cells as well as LCMV-infected cells. As hsp70 was effective in eliciting protection in vivo and CTLs in vitro, has in itself adjuvant effects and at the same time does not elicit neutralising antibodies, hsp70 proves to be a promising candidate as a vector in future vaccines.