The Endogenous Retinoic Acid Receptor Pathway Is Exploited by Mycobacterium tuberculosis during Infection, Both In Vitro and In Vivo.

Retinoic acid (RA) is a fundamental vitamin A metabolite involved in regulating immune responses through the nuclear RA receptor (RAR) and retinoid X receptor. While performing experiments using THP-1 cells as a model for Mycobacterium tuberculosis infection, we observed that serum-supplemented cultures displayed high levels of baseline RAR activation in the presence of live, but not heat-killed, bacteria, suggesting that M. tuberculosis robustly induces the endogenous RAR pathway. Using in vitro and in vivo models, we have further explored the role of endogenous RAR activity in M. tuberculosis infection through pharmacological inhibition of RARs. We found that M. tuberculosis induces classical RA response element genes such as CD38 and DHRS3 in both THP-1 cells and human primary CD14+ monocytes via a RAR-dependent pathway. M. tuberculosis-stimulated RAR activation was observed with conditioned media and required nonproteinaceous factor(s) present in FBS. Importantly, RAR blockade by (4-[(E)-2-[5,5-dimethyl-8-(2-phenylethynyl)-6H-naphthalen-2-yl]ethenyl]benzoic acid), a specific pan-RAR inverse agonist, in a low-dose murine model of tuberculosis significantly reduced SIGLEC-F+CD64+CD11c+high alveolar macrophages in the lungs, which correlated with 2× reduction in tissue mycobacterial burden. These results suggest that the endogenous RAR activation axis contributes to M. tuberculosis infection both in vitro and in vivo and reveal an opportunity for further investigation of new antituberculosis therapies.

Heme oxygenase-1 inhibition promotes IFNγ- and NOS2-mediated control of Mycobacterium tuberculosis infection.

Mycobacterium tuberculosis (Mtb) infection induces pulmonary expression of the heme-degrading enzyme heme oxygenase-1 (HO-1). We have previously shown that pharmacological inhibition of HO-1 activity in experimental tuberculosis results in decreased bacterial loads and unexpectedly that this outcome depends on the presence of T lymphocytes. Here, we extend these findings by demonstrating that IFNγ production by T lymphocytes and NOS2 expression underlie this T-cell requirement and that HO-1 inhibition potentiates IFNγ-induced NOS2-dependent control of Mtb by macrophages in vitro. Among the products of heme degradation by HO-1 (biliverdin, carbon monoxide, and iron), only iron supplementation reverted the HO-1 inhibition-induced enhancement of bacterial control and this reversal was associated with decreased NOS2 expression and NO production. In addition, we found that HO-1 inhibition results in decreased labile iron levels in Mtb-infected macrophages in vitro and diminished iron accumulation in Mtb-infected lungs in vivo. Together these results suggest that the T-lymphocyte dependence of the therapeutic outcome of HO-1 inhibition on Mtb infection reflects the role of the enzyme in generating iron that suppresses T-cell-mediated IFNγ/NOS2-dependent bacterial control. In broader terms, our findings highlight the importance of the crosstalk between iron metabolism and adaptive immunity in determining the outcome of infection.