The facultative intracellular pathogen, Legionella pneumophila, multiplies within and kills human monocytes and alveolar macrophages. We show that L. pneumophila strain Philadelphia-1 infects, multiplies within and kills the promyelocyte HL-60 cell line after its differentiation into macrophage-like cells. The characteristics of the interaction between L. pneumophila and differentiated HL-60 cells closely resemble those between L. pneumophila and human peripheral blood monocytes. With both cell types, C receptors and serum C mediate attachment of L. pneumophila, which are taken up by coiling phagocytosis. The replicative phagosome is lined with ribosomes; intracellular multiplication is iron-dependent; and replicating bacteria ultimately destroy the host cell. As in human monocytes, an avirulent mutant derivative of L. pneumophila Philadelphia-1, 25D, does not replicate in and is not cytopathic for differentiated HL-60 cells. Differentiated HL-60 cells therefore provide a convenient and faithful model for the study of L. pneumophila-mononuclear phagocyte interaction.

The legionnaires' disease bacterium, Legionella pneumophila, is a facultative intracellular parasite. Its interaction with phagocytes has characteristics in common with several other intracellular parasites. Critical aspects of L. pneumophila intracellular multiplication are evasion of lysosomal host cell defenses and the presence of a nutritionally appropriate environment. Following phagocytosis, wild-type L. pneumophila multiply within a specialized phagosome which does not fuse with secondary lysosomes. Mutants which have lost the ability to grow within phagocytes no longer cause disease in animals, indicating that the capacity to multiply intracellularly is important for pathogenesis. One such mutant, 25D, has been shown to be defective in inhibiting phagosome-lysosome fusion. This phagolysosomal environment is not conducive to Legionella growth. We report the isolation of a region of the L. pneumophila genome (icm, intracellular multiplication) which restores the capacity of 25D to multiply in human macrophages. The complemented mutants also regain the capacity to interfere with phagosome-lysosome fusion and to cause lethal pneumonia in guinea pigs.

We have examined whether a molecule that is capable of inducing immune protection, the major secretory protein (MSP) of Legionella pneumophila, is required for virulence in a guinea pig model of Legionnaires' disease. To do so, we have compared the virulence in guinea pigs of an isogenic pair of L. pneumophila, Philadelphia 1 strain, one of which produces MSP (MSP+) and one of which does not (MSP-). Both the MSP- strain and the MSP+ strain of L. pneumophila are highly virulent for guinea pigs, inducing similar signs and progression of illness. Both strains are lethal and have comparable LD50s and LD100s. Both strains multiply in the lungs of guinea pigs at a similar rate, and both strains produce indistinguishable pathological lesions in the lungs. Both strains maintain a stable phenotype with guinea pig passage, i.e., the MSP- strain does not regain the capacity to secrete MSP and the MSP+ strain retains its capacity to secrete MSP after lung passage. Although vaccination with MSP induces strong protective immunity in the guinea pig against lethal aerosol challenge with L. pneumophila, this protective immunogen is not required in its intact proteolytically active form for the expression of virulence by the intracellular pathogen L. pneumophila. This demonstrates that a protective immune response need not necessarily be directed against a virulence determinant and suggests that any molecule that allows the host immune system to detect and act against an intracellularly sequestered pathogen may potentially serve as a protective immunogen against such a pathogen.