Inhibition of release of tumor necrosis factor alpha (tnf-α) from human vascular tissue and smooth muscle cells by steroids

We found that bacterial lipopolysaccharide (LPS) stimulates release of TNF-α from human vascular tissue and smooth muscle cells (SMC). Here, we tested the hypothesis that release could be inhibited by pretreatment with corticosteroids. With IRB approval and patient consent, excess saphenous vein and internal mammary artery were obtained during coronary bypass. SMC were cultured from these vessels. Confluent SMC in 48 well culture dishes were exposed to 20 μg/ml LPS following pretreatment for 18 hr with 0.1, 1.0, and 10.0 μM dexamethasone (DEX). At 1, 3, 6, 18, and 24 hr the medium was removed and analyzed for biologically active TNF-α by the L929 cell cytotoxicity assay, SMC exposed to LPS but not treated with DEX served as controls. In control SMC, LPS stimulated TNF-α release in a time dependent manner from (indetectable levels at zero time to a peak of 18.0 ± 1.8 units/mg cell protein at 6 hr (n=8). DEX inhibited release at all time points. For instance at 6 hr, TNF-α was 5.6 ± 1.5, 4.6 ± 2.3 and 0.2 ± 0.2 units/mg for cells pretreated with 0.1, 1.0, and 10.0 μM DEX respectively (p < 0.05 vs. control). Additionally, time dependent LPS stimulated release of TNF-α was determined in segments of vascular tissue taken from 5 patients who received 1 g methlyprednisolone (MP) by i.v. injection during induction of anesthesia and compared to release from patients who had not received MP. In patients treated with MP, LPS did not stimulate release from tissues incubated for up to 24 hr. In contrast, LPS induced release of TNF-α from tissues of untreated patients (e.g., 733 ± 33 units/g tissue at 6 hr, n=9 ) These results confirm that human vascular tissue is a source of TNF-α and that release can be inhibited by corticosteriods.