Of note, hepatocytes pulsed in vivo and in vitro with α-GalCer can activate iNKT cells to secrete IL-4 and not IFN-γ [28]. Thus, although not essential, MI-503 mouse hepatocytes could play a role in iNKT cell activation in actively sensitized wild-type mice. There may simply be a network of CD1d+ cells (e.g. dendritic cells, Kuppfer cells or NKT cells themselves) that activate iNKT cells in vivo, as suggested here and elsewhere, via presentation of rapidly accumulating stimulatory lipids after sensitization [28, 32]. Dendritic cells have recently been shown to be

able to potentiate iNKT cell activation in a CD1d-dependent manner even in the context of low levels of lipid antigen [33]. Important questions remain pertaining to the stimulatory hepatic lipids observed here. It is unclear whether the accumulation of stimulatory lipids is the result of an increase in the quantity Staurosporine order of stimulatory hepatic lipids, a change in the quality of pre-existing hepatic lipids or a combination. A quantitative difference would imply migration of lipids from an extra-hepatic site, perhaps the skin at the site of sensitization. A qualitative difference would be mediated by chemical or structural modification of lipids native to

the liver. Although our extracts are sensitive to lipase (N. Dey, K. Lau, M. Szczepanik, P.W. Askenase, unpublished observations), the identity of these lipids is as yet unknown. This determination remains for further studies collaborating with glycolipid biochemists. The lipids may represent a subset of endogenous skin-derived self-lipids that have particular iNKT cell–activating potential. They may be released from the skin following sensitization. Alternatively, these may be hepatic lipids that Urocanase are somehow modified following skin sensitization to provide increased stimulation to iNKT cells. Finally, exogenous glycolipids derived from the host skin microbiota may be involved. While the finding of accumulating stimulatory

hepatic lipids begins to clarify the mystery of rapid iNKT cell response after sensitization, whether the entire role of iNKT cells in CS has been defined remains unclear. For example, we have observed using ELISA assays that serum IFN-γ levels peak approximately 1 day after sensitization in mice (unpublished observations), a finding that remains unexplained in terms of both mechanism and relevance. iNKT cells could potentially account for this. This and other described immune activities of iNKT cells, such as cytotoxicity and influence on regulatory T cells [34], remain unexplored in CS. The implications of these data for other diseases are also unclear and should be investigated further. Finally, these and related data on iNKT cell biology may have implications for a multitude of clinical diseases. For example, IL-4-producing iNKT cells may be therapeutic (e.g. NAFLD) or detrimental (e.g.