The interactions between the L. pneumophila phagosome and monocyte lysosomes were investigated by prelabeling the lysosomes with thorium dioxide, an electron-opaque colloidal marker, and by acid phosphatase cytochemistry. Phagosomes containing live L. pneumophila did not fuse with secondary lysosomes at 1 h after entry into monocytes or at 4 or 8 h after entry by which time the ribosome-lined L. pneumophila replicative vacuole had formed. In contrast, the majority of phagosomes containing formalin-killed L. pneumophila, live Streptococcus pneumoniae, and live Escherichia coli had fused with secondary lysosomes by 1 h after entry into monocytes. Erythromycin, a potent inhibitor of bacterial protein synthesis, at a concentration that completely inhibits L. pneumophila intracellular multiplication, had no influence on fusion of L. pneumophila phagosomes with secondary lysosomes. However, coating live L. pneumophila with antibody or with antibody and complement partially overcame the inhibition of fusion. Also activating the monocytes promoted fusion of a small proportion of phagosomes containing live L. pneumophila with secondary lysosomes. Acid phosphatase cytochemistry revealed that phagosomes containing live L. pneumophila did not fuse with either primary or secondary lysosomes. In contrast to phagosomes containing live bacteria, the majority of phagosomes containing formalin-killed L. pneumophila were fused with lysosomes by acid phosphatase cytochemistry. The capacity of L. pneumophila to inhibit phagosome-lysosome fusion may be a critical mechanism by which the bacterium resists monocyte microbicidal effects.

Although M. tuberculosis was identified as an agent of infectious disease by Koch in 1884, it has proved difficult to produce a safe and effective vaccine against it. In this article, Marcus A. Horwitz explains the need to replace the BCG vaccine, outlines the search for new vaccine designs, and points the way to the likely future of TB vaccines.

Legionella pneumophila, the causative agent of Legionnaires' disease, is a Gram-negative bacterium and a facultative intracellular parasite that multiplies in human monocytes and alveolar macrophages. In this paper, mutants of L. pneumophila avirulent for human monocytes were obtained and extensively characterized. The mutants were obtained by serial passage of wild-type L. pneumophila on suboptimal artificial medium. None of 44 such mutant clones were capable of multiplying in monocytes or exerting a cytopathic effect on monocyte monolayers. Under the same conditions, wild-type L. pneumophila multiplied 2.5-4.5 logs, and destroyed the monocyte monolayers. The basis for the avirulent phenotype was an inability of the mutants to multiply intracellularly. Both mutant and wild-type bacteria bound to and were ingested by monocytes, and both entered by coiling phagocytosis. Thereafter, their intracellular destinies diverged. The wild-type formed a distinctive ribosome-lined replicative phagosome, inhibited phagosome-lysosome fusion, and multiplied intracellularly. The mutant did not form the distinctive phagosome nor inhibit phagosome-lysosome fusion. The mutant survived intracellularly but did not replicate in the phagolysosome. In all other respects studied, the mutant and wild-type bacteria were similar. They had similar ultrastructure and colony morphology; both formed colonies of compact and diffuse type. They had similar structural and secretory protein profiles and LPS profile by PAGE. Both the mutant and wild-type bacteria were completely resistant to human complement in the presence or absence of high titer anti-L. pneumophila antibody. The mutant L. pneumophila have tremendous potential for enhancing our understanding of the intracellular biology of L. pneumophila and other parasites that follow a similar pathway through the mononuclear phagocyte. Such mutants also show promise for enhancing our understanding of immunity to L. pneumophila, and they may serve as prototypes in the development of safe and effective vaccines against intracellular pathogens.

Previous studies have shown that L. pneumophila multiplies intracellularly in human monocytes and alveolar macrophages within a membrane-bound cytoplasmic vacuole studded with ribosomes. In this paper, the formation of this novel vacuole is examined. After entry into monocytes, L. pneumophila resides in a membrane-bound vacuole. During the first hour after entry, vacuoles containing L. pneumophila are found surrounded by smooth vesicles fusing with or budding off from the vacuolar membrane and by mitochondria closely apposed to the vacuolar membrane. By 4 h, vacuoles are found less frequently surrounded by these cytoplasmic organelles, but now ribosomes and rough vesicles are found gathered about the vacuole. By 8 h, the ribosome-lined vacuole has formed. Erythromycin, at concentrations that completely inhibit the intracellular multiplication of L. pneumophila, has no effect on vacuole formation. Formalin-killed L. pneumophila also reside in a membrane-bound vacuole after entry into monocytes. In contrast to the situation with live L. pneumophila, cytoplasmic organelles are not found surrounding vacuoles containing formalin-killed L. pneumophila at any time after entry. Formalin-killed bacteria are rapidly digested, and by 4 h, few remain intact. The L. pneumophila-containing vacuole has certain features in common with other intracellular organisms that inhibit phagosome-lysosome fusion; these organisms may share a common mechanism for vacuole formation and inhibition of phagosome-lysosome fusion.

We have used the cryosection immunogold technique to study the composition of the Mycobacterium tuberculosis phagosome. We have used quantitative immunogold staining to determine the distribution of several known markers of the endosomal-lysosomal pathway in human monocytes after ingestion of either M. tuberculosis, Legionella pneumophila, or polystyrene beads. Compared with the other phagocytic particles studied, the M. tuberculosis phagosome exhibits delayed clearance of major histocompatibility complex (MHC) class I molecules, relatively intense staining for MHC class II molecules and the endosomal marker transferrin receptor, and relatively weak staining for the lysosomal membrane glycoproteins, CD63, LAMP-1, and LAMP-2 and the lysosomal acid protease, cathepsin D. In contrast to M. tuberculosis, the L. pneumophila phagosome rapidly clears MHC class I molecules and excludes all endosomal-lysosomal markers studied. In contrast to both live M. tuberculosis and L. pneumophila phagosomes, phagosomes containing either polystyrene beads or heat-killed M. tuberculosis stain intensely for lysosomal membrane glycoproteins and cathepsin D. These findings suggest that (a) M. tuberculosis retards the maturation of its phagosome along the endosomal-lysosomal pathway and resides in a compartment with endosomal, as opposed to lysosomal, characteristics; and (b) the intraphagosomal pathway, i.e., the pathway followed by several intracellular parasites that inhibit phagosome-lysosome fusion, is heterogeneous.

Legionella pneumophila, the etiologic agent of Legionnaires' disease, is phagocytized in an unusual way and multiplies in human mononuclear phagocytes in a novel phagosome. As a first step toward understanding these L. pneumophila-phagocyte interactions, we have studied the envelope of L. pneumophila Philadelphia 1 strain. We isolated cell envelopes by treating whole bacterial cells with lysozyme and EDTA to convert them to spheroplasts, then lysing the spheroplasts osmotically or sonically. We resolved the cell envelopes into two membrane fractions by isopycnic centrifugation. We localized NADH oxidase to the fraction of buoyant density 1.145, which we designated cytoplasmic membrane, and lipopolysaccharide (LPS) to the fraction of density 1.222, which we designated outer membrane. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) revealed that the L. pneumophila outer membrane contains a single major protein species migrating at 28,000 mol wt; this is the major protein of the bacterium. The cytoplasmic membrane also contains a single major protein species migrating at 65,000 mol wt. Surface iodination of the bacteria and agglutination and immunofluorescence studies with rabbit antibody produced against the purified major outer membrane protein (MOMP) revealed that this protein is exposed at the cell surface. We isolated LPS from L. pneumophila membranes by SDS-EDTA treatment. The pattern obtained by subjecting the LPS to SDS-PAGE and staining the gel with silver nitrate suggests that L. pneumophila LPS might be atypical. We studied patient serologic responses to cell envelope components of L. pneumophila Philadelphia 1, a serogroup 1 organism. Sera from patients with evidence of infection with serogroup 1 L. pneumophila contained large amounts of antibody to this strain. Few of these antibodies recognized the MOMP of L. pneumophila. In contrast, greater than 98% of these antibodies were directed against the LPS. This indicates that LPS is the dominant serogroup antigen and the major antigen responsible for the reactivity of patient sera in the indirect fluorescent antibody assay, currently the principal diagnostic assay for Legionella infection.

Previous studies have demonstrated that the Mycobacterium tuberculosis phagosome in human monocyte-derived macrophages acquires markers of early and late endosomes, but direct evidence of interaction of the M. tuberculosis phagosome with the endosomal compartment has been lacking. Using the cryosection immunogold technique, we have found that the M. tuberculosis phagosome acquires exogenously added transferrin in a time-dependent fashion. Near-maximal acquisition of transferrin occurs within 15 min, kinetics of acquisition consistent with interaction of the M. tuberculosis phagosome with early endosomes. Transferrin is chased out of the M. tuberculosis phagosome by incubation of the infected macrophages in culture medium lacking human transferrin. Phagosomes containing latex beads or heat-killed M. tuberculosis, on the other hand, do not acquire staining for transferrin. These and other findings demonstrate that M. tuberculosis arrests the maturation of its phagosome at a stage at which the phagosome interacts with early and late endosomes, but not with lysosomes. The transferrin endocytic pathway potentially provides a novel route for targeting antimicrobials to the M. tuberculosis phagosome.

We have studied the capacity of a selected fraction of Mycobacterium tuberculosis extracellular proteins (EP) released into broth culture by mid-logarithmic-growth-phase organisms to induce cell-mediated immune responses and protective immunity in a guinea pig model of pulmonary tuberculosis. Guinea pigs infected with M. tuberculosis by aerosol but not uninfected control guinea pigs exhibit strong cell-mediated immune responses to EP, manifest by dose-dependent cutaneous delayed-type hypersensitivity and splenic lymphocyte proliferation. Guinea pigs immunized subcutaneously with EP but not sham-immunized control guinea pigs also develop strong cell-mediated immune responses to EP, manifest by dose-dependent cutaneous delayed-type hypersensitivity and splenic lymphocyte proliferation. EP is nonlethal and nontoxic to guinea pigs upon subcutaneous immunization. Guinea pigs immunized with EP and then challenged with aerosolized M. tuberculosis exhibit protective immunity. In five independent experiments, EP-immunized guinea pigs were consistently protected against clinical illness, including weight loss. Compared with EP-immunized guinea pigs, sham-immunized control guinea pigs lost 12.9 +/- 2.0% (mean +/- SE) of their total weight. EP-immunized guinea pigs also had a 10-fold reduction in viable M. tuberculosis bacilli in their lungs and spleens (P = 0.004 and 0.001, respectively) compared with sham-immunized control animals. In the two experiments in which some guinea pigs died after aerosol challenge, EP-immunized animals were protected from death. Whereas all 12 (100%) EP-immunized guinea pigs survived challenge with aerosolized M. tuberculosis, only 6 of 12 (50%) sham-immunized control guinea pigs survived challenge (P = 0.007, Fisher exact test). This study demonstrates that actively growing M. tuberculosis cells release immunoprotective molecules extracellularly, that a subunit vaccine against tuberculosis is feasible, and that extracellular molecules of M. tuberculosis are potential candidates for a subunit vaccine.