The present studies examined production of the cytokines tumor necrosis factor alpha (TNF-ac), interleukin-lo (IL-1|3), and IL-6 by human monocyte-derived macrophages exposed to Pneumocystis carinii in vitro and the impact of concurrent macrophage infection with human immunodeficiency virus type 1 (HIV-1) on these cytokine responses. Macrophages were infected with the HIV-1 BaL monocytotropic strain for 10 to 14 days and then exposed to P. carinii. At various times following P. carinii treatment, culture supernatants were harvested to assess the cytokine profile. Addition of P. carinii to HIV-uninfected macrophages resulted in augmented production of IL-6, TNF-a, and IL-1" protein. By contrast, in HIV-infected macrophages exposed to P. carinii, only the release of IL-6 was increased compared with that for HIV-uninfected macrophages, while the levels of TNF-o and IL-1p decreased. This altered response was confirmed at the molecular level for TNF-ot mRNA. Preventing physical contact between P. carinii and macrophages by a membrane filter inhibited all cytokine release. Substituting P. carinii with a preparation of P. carinii 95to 115-kDa major membrane glycoprotein A yielded a response similar to that obtained by addition of intact P. carinii. These results suggest that HIV-1 infection of human macrophages modulates cytokine responses to P. carinii.

S Su ur rf fa ac ct ta an nt t p pr ro ot te ei in n--A A l le ev ve el ls s i in nc cr re ea as se e d du ur ri in ng g P Pn ne eu um mo oc cy ys st ti is s c ca ar ri in ni ii i p pn ne eu um mo on ni ia a i in n t th he e r ra at t There was a severalfold increase in SP-A protein and mRNA levels in uninfected glucocorticoid-treated rats. However, contrary to what has been reported with the surfactant-associated lipids, SP-A mRNA and protein levels in P. carinii-infected animals were significantly higher than those found in the uninfected, immunosuppressed animals.Our results demonstrate that SP-A increases, probably as a result of elevated mRNA levels, in immunosuppressed rats with P. carinii infection and are consistent with our findings in HIV-positive patients with P. carinii pneumonia.

Surfactant lipids are not only important to the physiologic function of the lungs, but may also influence disease processes like Pneumocystis pneumonia, in which the interaction of host-defense cells with pathogen occurs within the confines of the surfactant-rich alveolar hypophase. In the present studies the lipid profile of bronchoalveolar lavage fluid (BALF) was characterized in subjects with AIDS-related lung diseases including Pneumocystis pneumonia. BALF lipid and total protein measurements were made in 43 subjects with acquired immune deficiency syndrome (AIDS)-related lung disease and compared with those made in 50 normal human immunodeficiency virus (HIV)-seronegative controls. The AIDS patient samples contained significantly greater amounts of total cholesterol, phosphatidylglycerol (PG), and protein than the control samples; in contrast to previous observations in rodent P. carinii infection, no differences were seen in total phospholipid (PL) or phosphatidylcholine (PC) in the two groups. The proportions of several of these lipids were deranged in BALF obtained from the patient group: PG/PL and PC/cholesterol differed significantly from normal samples. In the subset of patients with AIDS-related Pneumocystis pneumonia, no correlation was apparent between discrete BALF lipids and clinical indices reflective of disease severity. Using these measurements to approximate the lipid composition of the alveolar microenvironment in AIDS-related lung disease, we performed experiments in which normal human alveolar macrophages were exposed to exogenous liposomal lipids and then challenged with P. carinii. The ingestion but not binding of P. carinii by macrophages was diminished as a result of lipid exposure.(ABSTRACT TRUNCATED AT 250 WORDS)

The alveolar macrophage (AM), as a representative human tissue macrophage, was used in an in vitro system to examine the anti-human immunodeficiency virus type-1 (HIV-1) activity of zidovudine (AZT) and granulocyte-macrophage colony-stimulating factor (GM-CSF). AMs were infected with the IIIB strain of HIV-1 and exposed to AZT (1 mumol/L), GM-CSF (30 U/mL), a combination of AZT (1 mumol/L)/GM-CSF (30 U/mL), or medium control. At 10 or 20 days post-infection, phytohemagglutinin (PHA)-stimulated peripheral blood mononuclear leukocytes (PBMLs) were added to the AM cultures as stimulated target cells. AZT effectively suppressed HIV replication and prevented transfer/amplification in target PBMLs as long as the drug was maintained in the medium. GM-CSF neither suppressed nor augmented HIV replication. The combination of AZT/GM-CSF was comparable with AZT alone in suppressing both the initial infection of AMs and the transfer to target PBMLs as long as the agents were maintained in the cultures. However, when the drugs were removed at the same time that PHA-stimulated PBMLs were added to the culture, the combination of AZT/GM-CSF was found to be more effective than AZT alone in preventing the transfer/amplification of HIV in the target lymphocytes. These results suggest that (1) AZT is effective in inhibiting HIV-1 infection in mononuclear phagocytes; (2) GM-CSF neither inhibits nor augments the replication of the IIIB strain of HIV in human AMs; and (3) the combination of AZT and GM-CSF may have an enhanced anti-HIV-1 activity compared with AZT alone. Clinical trials with the two agents in combination appear warranted.

No abstract