Regulation of Fas(Apo-1/CD95)- and perforinmediated lytic pathways of primary cytotoxic T lymphocytes by the protooncogene bcl-2

Cytotoxic T cells (CTL) induce cell death of their target cells either by the surface interaction between Fas ligand and Fas or by the release of perforin and granzymes. Both lytic pathways induce apoptosis yet it is not known whether identical or distinct apoptotic pathways are activated. The protooncogene bcl-2 is known to protect various hematopoietic cells from apoptosis induced by diverse agents. Here we show that overexpression of the Bcl-2 protein in the murine mastocytoma line P815 or in concanavalin A-activated splenocytes suppresses apoptotic cell death induced by allospecific primary cytotoxic T lymphocytes (CTL) in which only the Fas lytic pathway was functional. Bcl-2 also reduced target cell killing induced by CTL whose lytic activity was dependent on the perforin/granzyme pathway only. These data provide evidence that, in the target cells studied here, both perforin/granzyme and Fas apoptotic pathways are modulated by Bcl-2 and suggest that these two pathways converge at a step prior to Bcl-2 inhibition.     

A role for endobrevin/VAMP8 in CTL lytic granule exocytosis

CTL clear virus‐infected cells and tumorigenic cells by releasing potent cytotoxic enzymes stored in preformed lytic granules. The exocytosis process includes polarization of lytic granules toward the immunological synapse, tethering of lytic granules to the plasma membrane and finally fusion of lytic granules with the plasma membrane to release cytotoxic enzymes. Although much is known about the molecular machineries necessary for the earlier steps in lytic granule exocytosis, the molecular machinery governing the final step in the fusion process has not been identified. Here, we show using control and VAMP8 KO mice that VAMP8 is localized to the CTL lytic granules. While the immunological synapse and granule polarization appears normal in both VAMP8 KO and control CTL, CTL‐mediated killing was reduced for the Vamp8^–/– CTL. Analysis of lytic enzyme secretion demonstrated that granzyme A and granzyme B secretion is significantly compromised in VAMP8^–/– CTL, while the levels of the lytic enzymes in the cells are unaffected. Our results clearly show that VAMP8 is one of the v‐SNARE that regulate the lytic ability of CTL by influencing the ability of the lytic granules to fuse with the plasma membrane and release its contents.     

Mechanism of lymphocyte-mediated cytolysis: functional cytolytic T cells lacking perforin and granzymes.

Involvement of the lytic protein perforin (c. 65,000 MW) and of granule proteases (granzymes) in cell lysis induced by cytolytic T lymphocytes (CTL) has been suggested, but is still controversial. For example, in vivo-primed peritoneal exudate CTL (PEL) have been found to express perforin and granzyme activity in amounts comparable to those found in non-lytic lymphocytes, although PEL are the most potent of all CTL. Exploiting several cloned CTL hybridomas developed in this laboratory and newly available molecular probes for detecting perforin, granzymes, protein and mRNA, we now directly demonstrate killer T lymphocytes which kill effectively and specifically, but are free from perforin, lytic granules and granzymes, all three of which have been postulated to be involved in lymphocyte-mediated killing. The CTL hybridomas are completely devoid of perforin and granzymes prior to, during, and after activation by antigen, mitogen or interleukin-2 (IL-2). The induction of lytic granules, perforin, and granzymes in the in vivo-primed PEL, but not in the cloned CTL hybridomas, upon cultivation in IL-2, further suggests the involvement of these constituents in antigen/lymphokine-induced CTL activation and differentiation rather than directly in their cytocidal activity. Together, these findings support a perforin- and granzyme-independent CTL lytic mechanism.     

Inactivation and proteolytic degradation of perforin within lytic granules upon neutralization of acidic pH.

In our recent studies, an inhibitor of vacuolar-type H(+)-ATPase, concanamycin A (CMA) has been shown to neutralize acidic pH in vacuolar organelles, including lytic granules, and to decrease the perforin content markedly. In the present paper, we have further investigated the role of acidification in perforin storage by using CMA. In CD8+ cytotoxic T-lymphocyte (CTL) clones, the amount of perforin decreased rapidly at 30-90 min but no more decrease occurred at 90-120 min after the addition of CMA. Since exposure to actinomycin D, cycloheximide, or brefeldin A failed to reduce the perforin content, the perforin decrease in CMA-treated cells seems to be largely due to a reduction in the perforin already stored in lytic granules, rather than to the inhibition of the de novo synthesis or the intracellular glycoprotein transport of perforin. Diisopropylfluorophosphoridate (DFP) markedly antagonized the decrease in the perforin content in CMA-treated cells, while other protease inhibitors, i.e. antipain, E-64, leupeptin, pepstatin A and phenylmethylsulphonyl fluoride, did not. Nevertheless, DFP hardly reversed the abrogation of the killing activity by CMA. Indeed, the lytic granules prepared from DFP plus CMA-treated cells showed only a marginal level of haemolytic activity. In cell-free experiments using perforin-enriched granule fractions, acidic pH completely blocked the perforin activity. Under the acidic conditions, perforin was more resistant to an inactivation by calcium when exposed to calcium prior to the haemolysis test. Thus, these data suggest that perforin is primarily inactivated, possibly in a calcium-dependent manner, and is subsequently hydrolysed by DFP-sensitive proteases in the lytic granules at neutral pH. We conclude that acidic pH plays an essential role to maintain the integrity of perforin within the lytic granules.     

Cytotoxic T lymphocyte perforin and Fas ligand working in concert even when Fas ligand lytic action is still not detectable

The reason(s) why individual cytotoxic T lymphocytes (CTL) possess a fast-acting, perforin/granzyme-mediated, as well as a much slower, Fas ligand (FasL) -driven killing mechanism is not clear, nor is the basis for wide variations in killing activity exhibited by individual CTL, ranging from minutes to hours. We show that perforin expression among individual, conjugated CTL varies widely, which can account for the heterogeneity in killing speeds exhibited by individual CTL. Despite a 2-hr lag in FasL-based killing, CTL lytic action is enhanced when the two mechanisms operate in concert. This is explained by finding that the two pathways in fact are jump-started simultaneously with the lag in FasL lytic action reflecting pre-lytic caspase-8 activation and BH3-interacting domain (BID) cleavage. The complementary action of the two lytic pathways, co-expressed at varying levels among individual CTL, facilitates the lytic action of late-stage poor perforin-expressing CTL, ensuring optimal cytocidal action throughout the CTL response.     

The cytolytic T lymphocyte and its mode of action

While the binding step of cytolytic T lymphocyte (CTL) target cell interaction resulting in conjugate formation is a well-characterized event, there seems to be more than one mechanism whereby lymphocytes kill the target. In recent years, infliction of complement (C)-like "holes" (I.D. 10-20 nm) on the target cell membrane, believed to be produced by the Ca2+-dependent lytic protein(s) perforin/cytolysin of secretory lytic granule origin has been proposed to be the mechanism of lymphocytotoxicity. More recent evidence, however, suggests that Ca2+-dependent exocytosis of lytic granules (where detectable) is not involved in lymphocyte-mediated cytolysis. Furthermore, neither formation of C-like "holes" in targets exposed to CTL, nor higher-than-background levels of lytic granules, perforin or BLT-esterases, have been detected in highly potent, peritoneal exudate CTL (PEL) derived directly from the animal or in cytocidal PEL-hybridomas. Hence exocytosis of perforin and formation of the above pores may apply to certain effector cells, particularly those grown in vitro in IL-2, but not to in vivo primed CTL such as PEL. On the other hand, work from this laboratory with Ca2+ probes has shown that lysis induced by CTL such as PEL-not involving lytic granules, perforin or formation of the above "holes"-is preceded by a marked prelytic elevation of cytosolic Ca2+ in the target. CTL-induced target cell membrane perturbation--a direct result of receptor-mediated effector-to-target interaction or through a membrane-bound or secreted effector component(s)--may be responsible for triggering the prelytic influx of Ca2+ from external sources, or its mobilization from internal stores in the target. We propose that CTL-induced, persistent elevation of cytosolic Ca2+, above a critical level, rather than formation of 10-20 nm pores, is responsible for the catastrophic prelytic events observed in the target, such as bleb formation, metabolic exhaustion and DNA degradation, ultimately leading to lysis.     

Morphological and functional analysis of PTA granules in human NK cells

 Human natural killer (NK) cells contain unique granules with parallel tubular arrays (PTA granules) of approximately 30 nm diameter that can be seen only by electron microscopy. In order to clarify the role of PTA granules in NK cell-mediated cytolysis we examined these structures with regard to frequency and expression of lytic proteins (perforin, granzymes). NK cells (CD3⁻, CD16⁺, CD56⁺) were obtained from heparinized blood of healthy donors and enriched by double-step negative selection using mAb coupled to magnetic beads. PTA granules were found in 31.3% of freshly separated NK cells. When NK cells were cultivated, even in the presence of various stimulating agents (rhIL-2, rhIL-4, rhIL-6, rhIL-12, GM-CSF, rhIFNα, anti-CD16 mAb, dexamethasone), PTA granules disappeared and transformed into conventional primary lysosomes. By immune electron microscopy using antibodies directed against perforin and granzyme B we observed distinct immuno-reactivity in the tubules and in the tubule-associated faintly electron-dense matrix of PTA granules. Immuno-labelling for perforin and granzyme B was also found in the fine granular matrix of primary lysosomes. Finally, we tested the distribution of chondroitin 4-sulfate which is suggested to inactivate lytic proteins. Immuno-reactivity for chondroitin 4-sulfate was detected only in the moderately electron-dense matrix but not in the tubules of PTA granules. These observations indicate that perforin and granzyme B are stored in an inactive form in PTA tubules due to highly ordered paracrystalline protein folding. As soon as the tubules decay the lytic proteins are kept in an environment of chondroitin 4-sulfate for inactivation. Accepted: 24 March 1997     

Prevention of antigen-induced microtubule organizing center reorientation in cytotoxic T cells by modulation of protein kinase C activity

Lysis of target cells (TC) by cytotoxic T lymphocytes (CTL) is achieved by directional exocytosis of cytolytic molecules-perforin and granzymes. They are stored within lytic granules which can be readily released following antigenic stimulation. Secretion of lytic molecules appears to be controlled by protein kinase C (PKC) activity, since specific modulators of PKC activity abolish the lysis of TC. We have examined the effect of PKC modulation on some of the earliest events in the perforin/granzyme-mediated cytotoxicity. De novo synthesis of perforin mRNA, required for the refilling of granules and sustained cytotoxicity, seems to be unaltered in the presence of PKC modulators. Immunofluorescent studies of CTL-TC conjugates revealed that PKC modulation impairs reorientation of the microtubule organizing center toward the contact point with the TC, which accounts for the specific direction of lytic granules exocytosis. Thus, it appears that PKC regulates exocytosis of lytic granules by governing microtubule reorganization, one of the initial steps in perforin/granzyme-mediated cytotoxicity.      

Role of calcium influx in cytotoxic T lymphocyte lytic granule exocytosis during target cell killing.

One mechanism cytotoxic T lymphocytes use to kill targets is exocytosis of cytotoxic agents from lytic granules, a process that requires Ca(2+) influx. We investigated the role of Ca(2+) influx in granule exocytosis using TALL-104 human leukemic cytotoxic T cells triggered via a bispecific antibody containing an anti-CD3 F(ab') to kill Raji B lymphoma cells. Using a novel fluorescence method, we detected target-directed release of approximately 15% of lytic granules during killing. Consistent with previous work, we observed sustained CTL Ca(2+) gradients during killing, but gradients reflect the behavior of Fura-2 in granules. Rapid imaging experiments suggest that Ca(2+) channels are not polarized during killing, indicating that Ca(2+) influx does not direct granule reorientation. Furthermore, we find that Ca(2+) acts via a high-affinity interaction to promote granule exocytosis.     

Involvement of granzyme B and perforin gene expression in the lytic potential of human natural killer cells

Natural killer (NK) cells (CD3-) or large granular lymphocytes (LGL) spontaneously kill K562 targets but are unable to kill Daudi cells in the absence of IL-2 stimulation. IL-4 is reported to prevent or inhibit the IL-2-driven lymphokine-activated killer (LAK) generation in NK cells. Therefore, we wished to determine whether the antagonistic effect of IL-4 on IL-2-induced LAK activity might regulate the expression of genes encoding proteins involved in lysis, such as perforin, the pore-forming protein, or which are associated with lysis, such as granzymes A and B. By using in situ hybridization, we showed that, in addition to inducing LAK activity, IL-2 stimulation increased the amount of perforin and granzyme B mRNA at the single-cell level in 40 to 100% of the total CD3- LGL cell population. In addition, our results indicated that the stimulatory effect of IL-2 can be downregulated by IL-4 for both LAK activity and granzyme B and perforin gene expression. Here again, a decrease in the amount of specific mRNA per cell was noted. These findings suggest that modulation of the lytic machinery via lymphokines might be associated with regulation of the lytic potential of NK cells.     

Lysis of bystander target cells after triggering of human cytotoxic T lymphocytes

Human cloned cytotoxic T lymphocytes (CTL) specific for class I HLA antigens were used to investigate whether triggering of CTL leads to killing of innocent bystander target cells. After triggering of CTL by their specific target cells, lysis of bystander targets was detected in a 7-h cytotoxicity assay. Considerable differences were found in the susceptibility of various target cells to this type of lysis. Targets susceptible to this bystander lysis were also susceptible to lysis by CTL triggered by F(ab')2 fragments of an anti-T3 monoclonal antibody, whereas other targets were resistant to both types of cytotoxicity. Triggering of CTL by oxidized target cells or via a T3-independent activation pathway led to bystander lysis detectable already after 4 h. Bystander lysis was considerably enhanced under conditions that facilitated a non-specific cell contact between CTL and bystander target. We conclude that a function besides antigen recognition of the T cell receptor on CTL is to direct killing to the target cell. This directing, however, is incomplete and destruction of innocent bystanders can be detected under appropriate conditions.     

Perforin-dependent cytotoxic mechanism in killing by CD8 positive T cells in ginbuna crucian carp, Carassius auratus langsdorfii

T cell-mediated cytotoxicity occurs via pathways based on perforin or Fas mechanisms. Perforin is a protein present in the cytoplasmic granules of CD8⁺ cytotoxic T lymphocytes and is secreted to form pores on target cell membranes. In fish, although the involvement of perforin in cytotoxicity have been suggested for several species, perforin-mediated cytotoxicity of CD8α⁺ lymphocyte in conjunction with expression of the perforin gene has not been reported. In order to investigate the killing mechanism of CD8α⁺ lymphocytes by perforin-mediated pathway in fish, we measured apoptosis of target cells triggered by CD8α⁺ lymphocytes, performed cytotoxic assays in the presence or absence of perforin inhibitor; concanamycin A and EGTA, and analysed the expression of perforin1, perforin2 and perforin3 isotypic genes in ginbuna crucian carp. In the present study, we found that CTLs attached with target cells. CTL should have direct contact with target cells to kill them. Approximately 50% of target cells were positive for annexin V after co-cultured with CD8α⁺ lymphocytes, indicating the induction of apoptotic cell death. Concanamycin A, which induces depolymerization of perforin resulting in lytic function, suppressed the cytotoxicity of CD8α⁺ cells in a dose-dependent manner. In addition, cytotoxicity mediated by CD8α⁺ lymphocytes were significantly suppressed by the addition of the Ca²⁺-chelating agents EGTA or EGTA-Mg²⁺, and the addition of Ca²⁺ restored the killing mechanism of target cells. We further found enhanced expression of perforin1 but not perforin2 or perforin3 in CTLs from allo-sensitized fish. The present study has demonstrated that ginbuna CTLs kill target cells through perforin-mediated pathway, suggesting that perforin-mediated pathway is conserved throughout vertebrate.     

Involvement of granzyme B and perforin gene expression in the lytic potential of human natural killer cells

Natural killer (NK) cells (CD3⁻) or large granular lymphocytes (LGL) spontaneously kill K562 targets but are unable to kill Daudi cells in the absence of IL-2 stimulation. IL-4 is reported to prevent or inhibit the IL-2-driven lymphokine-activated killer (LAK) generation in NK cells. Therefore, we wished to determine whether the antagonistic effect of IL-4 on IL-2-induced LAK activity might regulate the expression of genes encoding proteins involved in lysis, such as perforin, the pore-forming protein, or which are associated with lysis, such as granzymes A and B. By using in situ hybridization, we showed that, in addition to inducing LAK activity, IL-2 stimulation increased the amount of perforin and granzyme B mRNA at the single-cell level in 40 to 100% of the total CD3⁻ LGL cell population. In addition, our results indicated that the stimulatory effect of IL-2 can be downregulated by IL-4 for both LAK activity and granzyme B and perforin gene expression. Here again, a decrease in the amount of specific mRNA per cell was noted. These findings suggest that modulation of the lytic machinery via lymphokines might be associated with regulation of the lytic potential of NK cells.     

Switch from perforin-expressing to perforin-deficient CD8+ T cells accounts for two distinct types of effector cytotoxic T lymphocytes in vivo

Although CD8⁺ cytotoxic T lymphocytes (CTL) exhibit both Fas ligand (FasL) -based and perforin-based lytic activities, the accepted hallmark of a fully active CTL remains its perforin killing machinery. Yet the origin, rationale for possessing both a slow-acting (FasL) and a fast-acting (perforin) killing mechanism has remained enigmatic. Here we have investigated perforin expression in CTL directly involved in acute tumour (i.e. leukaemias EL4 and L1210) allograft rejection occurring within the peritoneal cavity. We show that at the height of the immune response, the majority of conjugate-forming CD8⁺ CTL express high levels of perforin messenger RNA and protein, and kill essentially via perforin. Later however, coinciding with complete rejection, fully cytocidal CTL emerge which exhibit a stark decrease in perforin and now kill preferentially via constitutively expressed FasL. Although late in emergence, and persistent, these powerful CTL are neither effector-memory nor memory CTL. This finding has implications for the monitoring of anti-transplant responses in clinical settings, based on assessing perforin expression in graft infiltrating CD8⁺ T cells. The results show that as the immune response progresses in vivo, targeted cellular suicide mainly prunes high perforin-expressing CD8⁺ cells, resulting in the gradual switch in effector CTL, from mostly perforin-based to largely Fas/FasL-based killers. Hence, two kinds of CD8⁺ CTL have two killing strategies.     

Midline 1 directs lytic granule exocytosis and cytotoxicity of mouse killer T cells

Midline 1 (MID1) is a microtubule‐associated ubiquitin ligase that regulates protein phosphatase 2A activity. Loss‐of‐function mutations in MID1 lead to the X‐linked Opitz G/BBB syndrome characterized by defective midline development during embryogenesis. Here, we show that MID1 is strongly upregulated in murine cytotoxic lymphocytes (CTLs), and that it controls TCR signaling, centrosome trafficking, and exocytosis of lytic granules. In accordance, we find that the killing capacity of MID1^?/? CTLs is impaired. Transfection of MID1 into MID1^?/? CTLs completely rescued lytic granule exocytosis, and vice versa, knockdown of MID1 inhibited exocytosis of lytic granules in WT CTLs, cementing a central role for MID1 in the regulation of granule exocytosis. Thus, MID1 orchestrates multiple events in CTL responses, adding a novel level of regulation to CTL activation and cytotoxicity.     

Granzyme B: a natural born killer

Summary: A main pathway used by cytotoxic T lymphocytes (CTLs) and natural killer cells to eliminate pathogenic cells is via exocytosis of granule components in the direction of the target cell, delivering a lethal hit of cytolytic molecules. Amongst these, granzyme B and perforin have been shown to induce CTL‐mediated target cell DNA fragmentation and apoptosis. Once released from the CTL, granzyme B binds its receptor, the mannose‐6‐phosphate/insulin‐like growth factor II receptor, and is endocytosed but remains arrested in endocytic vesicles until released by perforin. Once in the cytosol, granzyme B targets caspase‐3 directly or indirectly through the mitochondria, initiating the caspase cascade to DNA fragmentation and apoptosis. Caspase activity is required for apoptosis to occur; however, in the absence of caspase activity, granzyme B can still initiate mitochondrial events via the cleavage of Bid. Recent work shows that granzyme B‐mediated release of apoptotic factors from the mitochondria is essential for the full activation of caspase‐3. Thus, granzyme B acts at multiple points to initiate the death of the offending cell. Studies of the granzyme B death receptor and internal signaling pathways may lead to critical advances in cell transplantation and cancer therapy.     

Human granzyme B is essential for DNA fragmentation of susceptible target cells

We have partially characterized the granules of the human NK cell line, YT-INDY, and assessed granule-mediated lysis and DNA fragmentation of assorted targets. Biochemical studies demonstrated significant quantities of granzyme B (asp-ase) and a heretofore undescribed chymase but no tryptase (i.e., granzyme A or 3) or distinct met-ase. YT-INDY expressed mRNA for granzyme B, perforin and CCPX. The existence of perforin was confirmed by immunoblot. The granules lysed both human and murine NK-sensitive and NK-resistant targets. YTindependent lytic pathway is associated with the granules. In addition, 4-(2-aminoethyl) benzenesulfonylfluoride hydrochloride (AEBSF), an inhibitor that selectively blocked the chymase and 3,4-dichloroisocoumarin (DCI), an inhibitor that inactivated both chymase and asp-ase activities, marginally affected lysis. By gel electrophoresis and ¹²⁵I-labeled deoxyuridine release assay, only murine cells (SP2/0 and YAC-1) underwent DNA fragmentation, and cleavage was completely inhibited by DCI, whereas EGTA, AEBSF and aurintricarboxylic acid (ATA) had no effect. The results, therefore, underscore the central role of granzyme B in granule-mediated DNA fragmentation, emphasize that the protease acts via an ATA-resistant endonuclease pathway and stress that nucleolysis does not invariably accompany granule-mediated cytolysis. Finally, ATA inhibited the asp-ase activity of isolated but not granule-associated granzyme B. ATA, therefore, is not a specific endonuclease inhibitor and results obtained with ATA should be viewed cautiously.     

MID2 can substitute for MID1 and control exocytosis of lytic granules in cytotoxic T cells

We have recently shown that the E3 ubiquitin ligase midline 1 (MID1) is upregulated in murine cytotoxic lymphocytes (CTL), where it controls exocytosis of lytic granules and the killing capacity. Accordingly, CTL from MID1 knock‐out (MID1^?/?) mice have a 25–30% reduction in exocytosis of lytic granules and cytotoxicity compared to CTL from wild‐type (WT) mice. We wondered why the MID1 gene knock‐out did not affect exocytosis and cytotoxicity more severely and speculated whether MID2, a close homologue of MID1, might partially compensate for the loss of MID1 in MID1^?/? CTL. Here, we showed that MID2, like MID1, is upregulated in activated murine T cells. Furthermore, MID1^?/? CTL upregulated MID2 two–twenty‐fold stronger than CTL from WT mice, suggesting that MID2 might compensate for MID1. In agreement, transfection of MID2 into MID1^?/? CTL completely rescued exocytosis of lytic granules in MID1^?/? CTL, and vice versa, knock‐down of MID2 inhibited exocytosis of lytic granules in both WT and MID1^?/? CTL, demonstrating that both MID1 and MID2 play a central role in the regulation of granule exocytosis and that functional redundancy exists between MID1 and MID2 in CTL.