Photodynamic therapy (PDT) involves the treatment of tumors in the presence of sensitizer, light, and oxygen, causing energy-dependent cytotoxicity. A vascular effect that causes hemorrhagic tumor necrosis has been described with PDT, but its basis remains undefined. To investigate the possible role of tumor necrosis factor (TNF) production in the generation of such a vascular effect and/or a direct tumor effect, we treated thioglycollate-elicited murine macrophages with PDT, and we measured the possible production of TNF using the L929 assay. An energy-dependent production of TNF by macrophage treated with PDT, stimulated or unstimulated with endotoxin, was demonstrated, and TNF production was inhibited at the highest treatment energy levels. These data represent the first description of cytokine production by PDT-treated macrophages, and may serve as another mechanism of PDT cytotoxicity in vivo, either directly by TNF-mediated tumor necrosis, or indirectly by vascular effects on tumor vessels.

mRNA from lungs of mice exposed to high-dose oxygen (greater than 95%) for 3 days demonstrated increased expression of the genes for tumor necrosis factor (TNF), interleukin-1, and interleukin-6 compared with mRNA from lungs of mice exposed to room air. Daily treatment of mice exposed to high-dose oxygen with an antibody to TNF improved survival compared with mice receiving a similar dose of control immunoglobulin G. Pretreatment of mice with repetitive sublethal intraperitoneal doses of recombinant human TNF for 3 days or a single intravenous dose followed by exposure to high-dose oxygen afforded a significant survival advantage compared with high-dose oxygen-exposed mice pretreated with vehicle or interleukin-1. The repetitive intraperitoneal TNF pretreatment reduced the development of interstitial pneumonitis, pulmonary edema, and lung weight gain associated with oxygen toxicity and enhanced expression of the gene for the free radical protective enzyme manganous superoxide dismutase in lung tissue, a gene that is augmented as mice are exposed to high-dose oxygen. Furthermore a single intravenous dose of TNF 24 h after oxygen exposure was still protective. The results suggest that the toxicity of oxygen therapy can be partially ameliorated by either treatment with anti-TNF antibody or pretreatment and early treatment with TNF. These findings are consistent with the hypothesis that oxygen exposure induces TNF, which causes part of the toxicity of high-dose oxygen, and that pretreatment or early treatment with TNF induces the gene for an enzyme that recently has been shown to be very effective in protecting mice from the toxicity of oxygen.

No abstract