Hoffman, O A

Abstract Recent studies suggest that TNF-alpha plays a central role in host defenses during Pneumocystis carinii pneumonia. To determine whether P. carinii directly stimulates TNF-alpha secretion, rat alveolar macrophages were cultured in the presence of purified P. carinii. Whereas unstimulated alveolar macrophages released only 13.0 +/- 2.7 pg/ml of TNF-alpha into the medium, macrophages stimulated with P. carinii released 108.2 +/- 20.4 pg/ml of TNF-alpha after overnight culture (p = 0.0001). Maximal TNF-alpha release was observed after 8 h of incubation and required a P. carinii: macrophage ratio of at least 2.5:1. Autoclaved P. carinii were also able to trigger TNF-alpha release from macrophages albeit at a reduced level. In view of recent evidence that P. carinii is phylogenetically related to the fungi and contains a beta-glucan-rich cell wall, we hypothesized that TNF-alpha release might in part be mediated by this cell wall component. Preincubation of macrophages with particulate beta-glucan derived from Baker’s yeast resulted in complete inhibition of TNF-alpha release in response to P. carinii. In addition, digestion of P. carinii with zymolyase, a preparation containing predominantly beta-glucanase activity, substantially reduced the ability of P. carinii to cause TNF-alpha release from macrophages. In a similar manner, macrophages incubated with P. carinii in the presence of laminariheptaose, an oligosaccharide that binds to macrophage beta-glucan receptors, also displayed decreased TNF-alpha release. Interestingly, TNF-alpha release may not be completely linked to the adherence of the organism to macrophages. Particulate beta-glucan significantly reduced P. carinii adherence to macrophages and also impaired TNF-alpha release. However, yeast mannan also significantly reduced P. carinii adherence to macrophages and also impaired TNF-alpha release. However, yeast mannan also significantly reduced P. carinii adherence but had no effect on TNF-alpha release. These data demonstrate that P. carinii can directly stimulate the secretion of TNF-alpha from alveolar macrophages and that this effect is largely mediated by beta-glucan components of the P. carinii cell wall.

beta-Glucans are polymers of D-glucose which represent major structural components of fungal cell walls. It was shown previously that fungi interact with macrophages through beta-glucan receptors, thereby inducing release of tumor necrosis factor alpha (TNF-alpha). Additional studies demonstrated that vitronectin, a host adhesive glycoprotein, binds to fungi and enhances macrophage recognition of these organisms. Since vitronectin contains a carbohydrate-binding region, we postulated that vitronectin binds fungal beta-glucans and subsequently augments macrophage TNF-alpha release in response to this fungal component. To study this, we first determined the release of TNF-alpha from alveolar macrophages stimulated with fungal beta-glucan. Maximal TNF-alpha release occurred with moderate concentrations of beta-glucan (100 to 200 micrograms/ml), whereas higher concentrations of beta-glucan (> or = 500 micrograms/ml) caused apparent suppression of the TNF-alpha activity released. This suppression of TNF-alpha activity by high concentrations of beta-glucan was mediated by the particulate beta-glucan binding soluble TNF-alpha, through the lectin-binding domain of the cytokine, rendering the TNF-alpha less available for measurement. Next, we assessed the interaction of vitronectin with beta-glucan. Binding of 125I-vitronectin to particulate fungal beta-glucan was dose dependent and specifically inhibitable by unlabeled vitronectin. Furthermore, treatment of beta-glucan with vitronectin substantially augmented macrophage TNF-alpha release in response to this fungal component. These findings demonstrate that fungal beta-glucan can directly modulate TNF-alpha release from macrophages. Further, these studies indicate that the host adhesive glycoprotein vitronectin specifically binds beta-glucan and augments macrophage cytokine release in response to this fungal element.

Pneumocystis carinii attaches to alveolar epithelial cells during the development of pneumonia. Adhesive proteins found within the alveolar space have been proposed to mediate P. carinii adherence to lung cells. Vitronectin (Vn), a 75-kDa glycoprotein present in the lower respiratory tract, has substantial cell-adhesive properties and might participate in the host-parasite interaction during P. carinii pneumonia. To address whether Vn binds to P. carinii, we studied the interaction of radiolabeled Vn with purified P. carinii organisms. Vn bound to P. carinii, occupying an estimated 5.47 x 10(5) binding sites per organism, with an affinity constant, Kd, of 4.24 x 10(-7) M. Interestingly, the interaction of Vn with P. carinii was not mediated through the Arg-Gly-Asp cell-adhesive domain of Vn. Addition of Arg-Gly-Asp-Ser (RGDS) peptides did not inhibit binding. In contrast, Vn binding to P. carinii was substantially inhibited by the addition of heparin or by digesting the organisms with heparitinase, suggesting that P. carinii may interact with the glycosaminoglycan-binding domain of Vn. To determine whether Vn might enhance P. carinii attachment to lung epithelial cells, we studied the binding of 51Cr-labeled P. carinii to cultured A549 lung cells. Addition of Vn resulted in significantly increased binding of P. carinii to A549 cells, whereas a neutralizing anti-Vn serum substantially reduced the binding of P. carinii to A549 cells. These data suggest that Vn binds to P. carinii and that Vn might provide an additional means by which P. carinii attaches to respiratory epithelial cells.

Abstract Pneumocystis carinii is an opportunistic organism that causes severe lung injury in immunocompromised hosts. Macrophage responses to P. carinii are poorly defined. Arachidonic acid (AA) and its metabolites are potent mediators of inflammation and have been implicated in host response to microorganisms. We therefore examined the production of eicosanoids from rat and rabbit alveolar macrophages stimulated with purified P. carinii. [14C]AA-labeled rabbit macrophages released 8.50 ± 1.33% of the incorporated [14C]AA after 90 min in response to P. carinii (P = 0.0001 compared with unstimulated controls). In contrast, a similar number of rat alveolar macrophages exhibited a smaller but significant response to P. carinii, releasing 3.84 ± 1.54% of their [14C]AA after 90 min (P = 0.001 compared with control). We further determined that P. carinii stimulated substantial production of prostaglandin E2 and concurrently a small amount of leukotriene B4 release from alveolar macrophages. To further investigate whether serum opsonization of P. carinii enhances these alterations in AA metabolism, we assessed the effect of P. carinii immune serum on P. carinii-induced AA release. P. carinii opsonized with this antiserum caused significantly greater AA release from rat alveolar macrophages than either unopsonized P. carinii or organisms opsonized with nonimmune serum. Previous studies suggest that P. carinii interacts with macrophage β-glucan and mannose receptors. However, incubation of macrophages with P. carinii in the presence of either soluble β-glucan or α-mannan failed to alter the release of AA from macrophages in response to P. carinii. Macrophage release of eicosanoids represents a potentially important host inflammatory response to P. carinii infection.