Sulforaphane and its Target Macrophage Migration Inhibitory Factor in Acute and Chronic Inflammation

Sulforaphane (SFN) is a naturally occurring chemopreventive agent with putative anti-inflammatory capabilities. SFN binds to and covalently inhibits the tautomerase domain of the pro-inflammatory protein Macrophage Migration Inhibitory Factor (MIF). MIF is a protein expressed by most cells that can function in intracellular signaling pathways, has chemokine-like functions, and induces cytokine production in immune cells. There is no known in vivo substrate for MIF tautomerase; however, studies suggest that this enzymatic activity is still biologically relevant. Upregulation of MIF is a characteristic seen in many diseases, including cancer, obesity, and rheumatoid arthritis. Given the role of MIF in such a wide variety of diseases with immunological components, we set out to examine the effects of MIF in both acute and chronic disease models with the intent of evaluating the effect of SFN as an inhibitor of MIF in these models.
We hypothesized that in the context of the acute inflammatory response, SFN administration would inhibit the tautomerase activity of MIF and thereby reduce the pro-inflammatory cellular response to an acute inflammatory stimulus. In vitro experiments demonstrated the ability of SFN to inhibit pro-inflammatory signaling through reduction of TNFα secretion, NF-κB activation, and NO production. SFN also blocked both MIF-dependent and CSF-1-dependent migration of bone marrow-derived macrophages (BMDM) in a Transwell assay. Examination of SFN treatment in the murine air pouch model of acute inflammation showed that while SFN does inhibit MIF tautomerase activity in the pouch lining, it does not decrease the number of infiltrating immune cells as hypothesized. We did observe a shift in immune cell populations in the SFN pre-treated mice compared to LPS-induced mice, such that the percentage of CD11bintGR1lo cells decreased with SFN pre-treatment while the percentage of CD11bhiGR1hi cells increased. The absolute numbers of these populations in the SFN pre-treated group were not significantly different from the LPS alone group and, in fact, trended toward inducing more cells than LPS alone. The F4/80+ macrophage population was not significantly altered. These data suggest that SFN does not have the anticipated anti-inflammatory activity in vivo, and may actually have a pro-inflammatory function in the murine air pouch model of acute inflammation.
We concurrently studied the effects of MIF on the immune response to LPS in the air pouch model. We hypothesized that addition of rmMIF, but not tautomerase-dead rmMIF, into the air pouch would rescue the immune response to LPS in MIF–/– mice. However, in our hands, MIF–/– mice unexpectedly did not have a less robust cellular response to LPS compared to WT, as had been previously published. We observed equivalent numbers of infiltrating immune cells in both genotypes in response to LPS. This complication with the model prevented us from moving forward with examination of the role of MIF tautomerase activity in acute inflammation.
Finally, we investigated the role of MIF in the chronic immune response to obesity. We hypothesized that genetic ablation of MIF would result in fewer visceral white adipose tissue (vWAT) infiltrating immune cells and concurrent protection from the development of insulin resistance. Using a high fat diet-induced model of obesity in MIF–/– and MIF WT mice, we demonstrated that that absence of MIF does not result in differences in weight gain or visceral white adipose tissue mass. Immune cell infiltration into vWAT was not dependent upon MIF. Finally, endpoint analysis of metabolic outcomes in obese MIF–/– mice showed that the absence of MIF did not have an effect on blood glucose or blood insulin levels, nor did it change the course of the development of insulin resistance compared to WT controls. Had there been a demonstrable MIF-dependent difference in this model, we would have followed up with studies examining the effect of SFN treatment of obese MIF WT mice. However, our data led us to conclude that global MIF expression is not a critical factor for the development of obesity and related metabolic complications in this model.