Commercial preparations of lipoteichoic acid contain endotoxin that contributes to activation of mouse macrophages in vitro

Lipoteichoic acids (LTA), cell wall components of gram-positive bacteria, h ave been reported to induce various Inflammatory mediators and to play a ke y role in gram-positive-microbe-mediated septic shock. In a large number of these studies, investigators used commercially available LTA purified from a variety of gram-positive bacteria, including Staphylococcus aureus, Baci llus subtilis, and Streptococcus sanguis, We report here that, although the se commercially available LTA could be readily shown to stimulate productio n of nitric oxide (NO) in RAW 264.7 mouse macrophages, the activity was dra matically inhibited by polymyxin B, a relatively specific inhibitor of endo toxin biological activity. One-step purification of the commercially availa ble S. aureus LTA using hydrophobic interaction chromatography resulted in two well-separated peak fractions, one highly enriched fur LTA and a second highly enriched for endotoxin, The LTA-enriched fractions did not induce p roduction of NO in RAW 264.7 macrophages, although they caused a dose-depen dent induction of NO in the presence of low concentrations of gamma interfe ron (IFN-gamma) (which by itself induced little NO), regardless of the pres ence of polymyxin B, In contrast, the endotoxin-enriched fractions by thems elves inhibited in high levels of NO in RAW 264.7 macrophages but activity was almost completely inhibited in the presence of polymyxin B, Consistent with these findings, our data also indicate that commercial LTA preparation s from S, aureus, B, subtilis, and S. sanguis were not able to induce NO fr om lipopolysaccharide-hyporesponsive C3H/HeJ mouse peritoneal macrophages, but in the presence of IFN-gamma, these LTA preparations were able to induc e relatively high levels of NO from C3H/HeJ macrophages, These results indi cate that commercially available LTA can contain contaminating and potentia lly significant levels of endotoxin that can be expected to contribute to t he putative macrophage-stimulating effects of LTA as assessed by NO product ion. The fact that the purified LTA, by itself, was not able to induce sign ificant levels of NO secretion in RAW 264.7 macrophages supports the conclu sion that caution in attributing high-level biological activity to this mic robial cell wall constituent should be exercised.