Comparative capacity of four antifungal agents to stimulate murine macrophages to produce tumour necrosis factor alpha: an effect that is attenuated by pentoxifylline, liposomal vesicles, and dexamethasone.

The efficacy and toxicity of certain antifungal agents may be related to their ability to induce the production of cytokines by mononuclear phagocytes. The capacity of incremental concentrations of fluconazole, 5-fluorocytosine (5-FC), amphotericin B (AmB), and liposomal AmB (LAB) to stimulate murine peritoneal and RAW 264.7 macrophages to secrete tumour necrosis factor alpha (TNF alpha) after 3, 6 and 24 h incubation was assessed by L929 cytotoxic bioassay. Fluconazole (2.5-40 mg/L) and 5-FC (25-100 mg/L) did not have a stimulatory effect. However, AmB (0.25-10 mg/L) elicited TNF alpha production by macrophages. This response was concentration-dependent, and peak TNF alpha levels were detected between 3 and 6 h. This effect was attenuated by incorporation of AmB into liposomal vesicles and by pretreating macrophages with pentoxifylline or dexamethasone. AmB I mg/L in combination with 1 x 10(6) cfu of Candida albicans stimulated peritoneal macrophages to produce similar quantities of TNF alpha as AmB alone, and two- to four-fold more TNF alpha than C. albicans alone. Thus, this study suggests that: (1) the immunomodulatory activity and toxicities of AmB, in part, may be attributed to the capacity of this drug to stimulate macrophages to secrete TNF alpha, (2) the TNF alpha that is produced by macrophages in response to AmB may have clinical relevance even in the face of C. albicans infection, and (3) the failure of fluconazole, 5-FC, and LAB to elicit a TNF alpha response may explain their improved side-effect profiles.