Background Methotrexate (MTX) remains a cornerstone of treatment in multiple forms of inflammatory arthritis, lupus and vasculitis. The anti-inflammatory effects of MTX are more likely to result from an increase in intracellular and extracellular adenosine concentration, which are produced after inhibition of the enzyme AICAR transformylase by MTX, which converts the nucleotide analog AICAR (5-aminoimidazole-4-carboxamide ribonucleotide), also known as ZMP to formyl-AICAR, resulting in the accumulation of this metabolite. Another function of AICAR is the activation of the AMP-dependent kinase (AMPK). AMPK is a highly conserved trimeric protein kinase complex that exists in essentially all eukaryotic cells and is a crucial cellular energy sensor. In mammals, AMPK is activated by an increasing cellular ADP/ATP ratio secondary to metabolic stress (glucose deprivation, hypoxia, and ischemia) or accelerate ATP consumption. Once activated by decreased intracellular energy status, AMPK will promote ATP production by switching on catabolic and turning off anabolic biosynthetic pathways. We hypothesize that AMPK activation mediates a major portion of the anti-inflammatory effects of MTX and that this may account for the efficacy of MTX in rheumatic diseases. A better understanding of the molecular targets of methotrexate may allow the development of novel anti-inflammatory drugs.
Methods We investigated the role of the anti-inflammatory effect of methotrexate via AMPK in human monocytes-derived macrophages (MDM) and mouse bone marrow-derived macrophages (BMDM) along with AICAR and A769662 (well knows as AMPK activators) and compound C, a selective ATP-competitive inhibitor of AMPK. AMPK phosphorylation and total AMPK were measured by Western blotting. Cells were then stimulated with LPS or TNF-α, and production of pro-inflammatory cytokines were measured in the supernatant using a Luminex multiplex assay technique. We also generated AMPKα1 deficient macrophages in order to test if these are resistant to the anti-inflammatory effects of MTX.
Results MTX induced AMPK phosphorylation in a time and dose-dependent manner, with effects comparable to the synthetic AMPK activator A769662 and AICAR both in hMDM and BMDM. MTX-induced AMPK activation was associated with a reduction in the production of pro-inflammatory cytokines (IL-6, IL-1 β, and TNF-α) in response to LPS and TNF stimulation. Compound C is able to partially reverse the effects of MTX on LPS and TNF -induced cytokine production, suggesting that AMPK activation is responsible for these anti-inflammatory effects. Folic acid is not able to revert the MTX activation of AMPK in hMDM and BMDM.
Conclusion Methotrexate is able to induce AMPK activation in both human and mouse macrophages, and suppress pro-inflammatory cytokines in a manner dependent on AMPK activity. These results have been confirmed genetically in macrophages deficient in AMPK subunits and models of chronic inflammation and diseases such as serum transfer arthritis. Our findings raise the possibility that some anti-inflammatory effects of MTX are mediated by AMPK, suggest that AMPK may be a target for the action of current ‘antimetabolite’ anti-inflammatory agents and a target for the development of new anti-inflammatory drugs.