Differential effects on TNF alpha production by pharmacological agents with varying molecular sites of action

This study describes the activation conditions for tumor necrosis factor-alpha (TNF alpha) production in myelomonocytic U937 cells and human primary peripheral blood monocytes in response to lipopolysaccharide (LPS) and/or phorbol 12-myristate 13-acetate (PMA). PMA itself induced only low levels of TNF alpha production with delayed kinetics (e.g. 0.758 +/- 0.128 ng/ml from U937 cells after 48 h) while LPS induced greater levels of TNF alpha production in less time (e.g. 2.083 +/- 0.96 ng/ml from monocytes in 24 h). Pharmacological agents with various molecular sites of action were used to validate the two systems, with the protein serine-threonine kinase inhibitors staurosporine and Ro-31-8220, the protein tyrosine kinase inhibitor herbimycin A (HBA) and dexamethasone exhibiting the greatest potency (IC50S 5-350 nM). In contrast to the effect on TNF alpha production, PMA induced strong phosphorylation/activation of p42/p44mapk in monocytes by 10 min determined in a mobility shift assay, while LPS was a weaker inducer. Additionally, staurosporine (to LPS and PMA) and HBA (to LPS only) inhibited the activation of these mitogen-activated protein kinase (MAPK) isoforms at doses 10-100 fold higher than those required to inhibit maximal TNF alpha production. These data indicate the involvement of the p42/p44mapk signalling pathway in LPS-induced pro-inflammatory cytokine production but suggest that other signalling pathways are also implicated in this phenomenon.