The immune system is very effective in fighting infections but seems not to be as efficient in recognizing and destroying cancer cells. In the cancer setting, tumor antigen uptake and presentation by APCs to antigen-specific T-cells often occurs in the absence of inflammation resulting therefore in tolerance induction. It is plausible therefore that by converting APCs from a non-inflammatory to an inflammatory phenotype through ligation of TLRs we may well overcome immune tolerance and tip the balance towards productive tumor antigen-specific T-cell responses. Among all the TLR-ligands identified to date, flagellin is the only one with a strictly proteinic nature, characteristic that make it a suitable candidate for cloning and transfection into tumor cells to generate novel tumor cell based vaccines. In this study, we first evaluated whether treatment with purified flagellin could prevent tolerance induction in vivo. Naìˆve CD4+ T-cells (2.5x106) specific for a MHC class II-restricted epitope of influenza hemagglutinin (HA) were adoptively transferred intravenously into BALB/c mice, 24 hours after mice were given either a tolerogenic dose of HA-peptide (200 mg), or a combination of this high dose of peptide together with Flagellin (10 mg I.V.). Two weeks later animals were sacrificed and antigen-specific CD4+ T-cell responses towards the cognate antigen evaluated in vitro. As expected clonotypic T-cells isolated from animals treated with high dose peptide were fully tolerant, in sharp contrast with those isolated from flagellin treated animals that displayed normal responses in terms of cytokine production and proliferation. Surprisingly, this preservation of T-cell function following in vivo treatment with flagellin was not observed when animals were treated with high dose HA-peptide in the presence of the TLR4 ligand, LPS. To better understand the mechanism(s) by which flagellin, and not LPS, preserved the responsiveness of antigen-specific CD4+ T-cells to cognate antigen presented by APCs, we assessed the phenotypic characteristics and the cytokine profile of macrophages and DCs treated in vitro with these TLRs ligands. Although LPS-treated APCs produce higher levels of IL-12, relative to flagellin-treated APCs, the production of this pro-inflammatory cytokine was accompanied by a parallel induction of the anti-inflammatory cytokine, IL-10. Interestingly, flagellin-treated APCs produced IL-12 but were unable to produce IL-10. This effect was dependent on ligation of TLR5, since it was not observed when RAW264.7 cells -which lack TLR5- were treated with flagellin. In vivo studies further confirmed our observations since IL-10 was not detected in the serum of animals treated with flagellin, but it was present in significant amounts in LPS-treated animals. This inhibitory effect of flagellin on IL-10 production was seen even when APCs were stimulated in vitro with strong inducers of IL-10. Given the above properties of flagellin, we generated two novel approaches to be used in the formulation of tumor cell based vaccines: 1) A bystander cell line transfected with the fliC gene from Salmonella typhimurim for flagellin expression (B78H1-Flagellin) and 2) Microspheres beads coated with flagellin. Both vaccination strategies are being currently studied in the in vivo and in vitro settings

Author notes

Corresponding author