Mice lacking the perforin gene were generated by using targeted gene disruption in embryonal stem cells. When infected with lymphocytic choriomeningitis virus (LCMV), perforin-less (-/-) mice showed clear signs of having mounted an immune response based on activation of CD8 T cells but were unable to clear the LCMV infection. This failure to eliminate virus was accompanied by a failure to generate spleen cells capable of lysing LCMV-infected fibroblasts in vitro. Spleen cells from LCMV-infected -/- mice were able to lyse hematopoietic target cells after exposure to phorbol 12-myristate 13-acetate and ionomycin, provided the target cells expressed the Fas antigen. Spleen cells from -/- mice also responded to alloantigen in mixed leukocyte culture by blastogenesis and proliferation. The resulting cells were able to lyse hematopoietic target cells, although not as well as spleen cells from +/+ littermates sensitized in the same manner. However, lysis by -/- cells was again seen only if the target cells expressed Fas antigen. We conclude that perforin-less -/- mice retain and express the Fas lytic pathway as expressed in vitro but that this pathway is insufficient to clear an LCMV infection in vivo.

No abstract

Abstract. Extracellular ATP is shown here to induce programmed cell death (or apoptosis) in thymocytes and certain tumor cell lines. EM studies indicate that the ATP-induced death of thymocytes and susceptible tumor cells follows morphological changes usually associated with glucocorticoid-induced apoptosis of thymocytes. These changes include condensation of chromatin, blebbing of the cell surface, and breakdown of the nucleus. Cytotoxicity assays using doublelabeled cells show that ATP-mediated cell lysis is accompanied by fragmentation of the target cell DNA. DNA fragmentation can be set off by ATP but not the nonhydrolysable analogue ATP'yS nor other nucleoside-5'-triphosphates. ATP-induced DNA fragmentation but not ATP-induced 5tCr release can be blocked in cells pretreated with inhibitors of protein or RNA synthesis or the endonuclease inhibitor, zinc; whereas pretreatment with calmidazolium, a potent calmodulin antagonist, blocks both DNA fragmentation and 51Cr release. The biochemical and morphological changes caused by ATP are preceded by a rapid increase in the cytoplasmic calcium of the susceptible cell. Calcium fluxes by themselves, however, are not sufficient to cause apoptosis, as the poreforming protein, perforin, causes cell lysis without DNA fragmentation or the morphological changes associated with apoptosis. Taken together, these results indicate that ATP can cause cell death through two independent mechanisms, one of which, requiring an active participation on the part of the cell, takes place through apoptosis.

CD8+ T lymphocytes have been implicated in the protective immune response against human and murine tuberculosis. However, the functional role that this cell subset plays during the resolution of infection remains controversial. In this study, we demonstrate the presence of Mycobacterium tuberculosis-specific CD8+ CTL in the lungs and lung-draining lymph nodes of mice infected with M. tuberculosis via the aerosol or i.v. route. These cells expressed perforin in vivo and specifically recognized and lysed M. tuberculosis-infected macrophages in a perforin-dependent manner after a short period of in vitro restimulation. The efficiency of lysis of infected macrophages was dependent upon the time allowed for interaction between macrophage and M. tuberculosis bacilli. Recognition of infected targets by CD8+ CTL was β2-microglobulin and MHC class I dependent and was not CD1d restricted. The presented data indicate that CD8+ T cells contribute to the protective immune response during M. tuberculosis infection by exerting cytotoxic function and lysing infected macrophages.

The role of mast cells and mast-cell-derived factors in natural cytotoxic reactions was investigated. Cultured and freshly isolated murine mast cells are shown to be cytotoxic to WEHI-164 and YAC-1 targets in 18-hr viability assays but not in 4-hr assays. Here, we describe a cytotoxic factor in murine mast cells that is immunologically related to tumor necrosis factor (TNF). This TNF-like factor lyses WEHI-164 cells with a slow time course requiring 16-20 hr for the lytic reaction to complete. Antibodies specific for human and murine TNF and human lymphotoxin partially block mast cell lysis of WEHI-164 cells. These antibodies react on immunoblots with one major mast cell protein band of 50 kDa. Immunoblot analysis shows this factor in cloned and uncloned cultured mouse mast cells and in mature "connective tissue-type" mast cells freshly purified from rat or mouse peritoneal cavities. The amount of this factor is greatly enhanced in cells that have been stimulated with a combination of phorbol ester/concanavalin A or bacterial lipopolysaccharide. Subcellular fractionation analysis of mast cells with Percoll gradients reveals two pools of TNF-related cytotoxic activity that are associated with free cytosolic material and granule fractions. In contrast to cytotoxic T lymphocytes and natural killer cells, granuleenriched fractions of mast cells do not contain any hemolytic activity. The localization of the TNF-like molecule in mast cell granules may play a strategical role in the rapid delivery of this mediator to the target cell membrane following cell surface stimulation and degranulation.