Cytolytic Granule-mediated Apoptosis

First Published in R&D Systems' 2005 Catalog

Contents

Cytotoxic T lymphocytes (CTLs) and Natural Killer (NK) cells play critical roles in the immune response to viruses and cellular transformation. This results from their ability to recognize and initiate apoptosis in the affected cells. CTLs and NK cells both utilize the release of cytolytic lysosomal granules containing proteins capable of entering the cytoplasm of the target cell and initiating cascades of events that ultimately result in its controlled destruction.1, 2 This minireview briefly outlines the processes that trigger CTL granule release and highlights the functions of several granule components critical for mediating the activities of cytotoxic cells.

Immunological Synapse Formation

Activated T cells rely on the ability of their T cell receptor (TCR) to recognize antigen on the surface of target cells. In the case of CD4+ T cells, recognition occurs via antigen-bound MHC II, while CD8+ CTLs recognize antigen displayed by MHC I. The TCR of both CD4+ and CD8+ T cells consists of a clonally variable alpha/beta transmembrane heterodimer, and a CD3 complex consisting of delta epsilon, gamma epsilon, and zeta zeta membrane-spanning subunits.3 CD3 cytoplasmic domains contain immunoreceptor tyrosine activation motifs (ITAMs) that become phosphorylated by members of the Src family (e.g. Lck) upon TCR ligation.4 This leads to the recruitment of proximal signaling kinases like ZAP-70 and adaptor proteins that initiate a cascade of downstream signaling events including activation of Rho family GTPases, Ras/MAPK, PLC gamma, and Ca2+ mobilization.5, 6, 7, 8, 9 The actin cytoskeleton reorganizes, and the Golgi and microtubule-organizing center (MTOC) reorient toward the target cell.10, 11, 12 A ring of adhesion proteins, including the integrin alphaLbeta2 (LFA-1), interact with the receptor ICAM-1 on the target cell, surrounding a signaling domain containing the TCR/CD4 or TCR/CD8 complex.13, 14, 15 This close association between T cell and target is known as the immunological synapse (Figure 1).

Figure 1. Ligation of the CTL TCR by antigen induces the activation of signaling cascades that result in the polarization of the Golgi and MTOC, and the docking and release of lytic granules.
View Larger Image 		Figure 1. Ligation of the CTL TCR by antigen induces the activation of signaling cascades that result in the polarization of the Golgi and MTOC, and the docking and release of lytic granules. Surrounded by adhesion molecules that include alphaLbeta2/LFA-1 and ICAM-1 are separate signaling and secretory domains.

Although similar in many ways, there are also key differences between the synapse formed by active CTLs and CD4+ T cells. A CD4+ cell interaction with its target may last many hours, prolonging signaling cascades necessary for gene activation, proliferation, and differentiation, while a CTL might only require minutes to initiate an irreversible apoptosis cascade in its target.14 The antigen/MHC I recognition complex is a potent inducer of CTL cytotoxic activity. Remarkably, as little as one TCR/MHC I interaction may be all that is necessary to elicit a cytolytic response.16 Upon TCR ligation, cytolytic granules move via polarized microtubules to the region of the MTOC where they dock prior to release. The mechanisms regulating the docking, fusion, and release process are not well understood. However, there is evidence to suggest important roles for the small GTPase Rab27A, and Munc13-4, a member of a family of proteins known for their roles in priming neuronal synaptic vesicles for fusion and release.17, 18, 19, 20 In the CTL synapse, vesicle release appears to occur in a secretory domain distinct from the signaling domain containing the TCR complex.14, 15 The close opposition and highly organized topography of the CTL synapse allows for the focused release of the highly toxic components of secretory granules in a process that likely avoids damaging nearby cells. Tight control is critical since the contents of just a small number of granules are enough to elicit apoptosis in the target cell.14

Protein Function
Granzyme-A, B, C, D, E, F, G, H, K, and M Serine proteases with roles in Caspase-dependent and -independent apoptosis
Perforin Membrane perturbing protein that mediates Granzyme entry into target cell cytoplasm
Calreticulin Binds Perforin and may inhibit Perforin-mediated damage to the effector cell
Cathepsin B Appears on the cell surface following degranulation and may offer protection from Perforin-mediated self-destruction
Cathepsin C Processes Granzyme pro-enzymes
Granulysin Human membrane perturbing microbicidal protein that can initiate Cytochrome c release and apoptosis
Serglycin Proteoglycan that non-covalently binds Granzymes
H+ ATPase Granule acidification
Chemokines: RANTES/CCL5, IP-10/CXCL10, MIP-1 alpha/CCL3, and MIP-1 beta/CCL4 Small proteins capable of mediating chemotaxis and/or cell activation
Fas Ligand/TNFSF6 Potent apoptosis-inducing TNF superfamily member
Table 1. Cytolytic granule contents and their function.

Granule Contents and Their Functions

The cytolytic granules contain an array of proteins designed to activate pro-enzymes, perturb the target cell membrane, act as molecular chaperones, protect the CTL from its own cytolytic mechanisms, and initiate apoptosis in the target cell (Table 1).

Membrane perturbation

Perforin

Perforin is a major component of the granule and is important for mediating the entry of Granzymes into the host cell cytoplasm. Emphasizing its importance, CTLs from Perforin-deficient mice are compromised in granule-mediated cytolytic activity, and are susceptible to intracellular pathogens and the formation of tumors.21, 22 Perforin is activated by cleavage of its C-terminus in a process that is not completely understood but appears to require an acidic environment.23

How Perforin works has been a long-standing question among immunobiologists. It polymerizes in the presence of Ca2+ and is capable of forming pores in lipid membranes.24 This activity led to the classical view that cytotoxic components like Granzymes enter the target cell by passive diffusion through these pores. Although plausible, most evidence now suggests that other mechanisms exist. Only at high concentration does Perforin form a pore with the diameter necessary to allow Granzyme B (~32 kDa) to enter the cytosol.25 Lower, sublytic concentrations that do not allow molecules of 8-13 kDa to pass still mediate Granzyme nuclear translocation, chromatin condensation, and DNA fragmentation.25, 26, 27, 28 Furthermore, Granzyme B may exist in a large, ~300 kDa complex with the proteoglycan serglycin, making it even more unlikely that it enters the cytosol through a Perforin pore.29, 30, 31 Although the precise mechanism remains to be elucidated, several lines of evidence suggest that Perforin acts to release Granzymes into the cytosol following uptake by endocytosis. Disrupting vesicle trafficking with Brefeldin A blocks Perforin-mediated access of Granzyme B to the cytosol.25 In addition, it has been shown that Granzyme B can be taken up by cells in a Perforin-independent manner and only later when Perforin is added is Granzyme B freed to elicit its downstream effects.26, 32 The idea that Perforin might free Granzymes from the endosomes is also supported by the observation that endosmolytic agents, including adenovirus and the bacterial toxin Listeriolysin O, can substitute for Perforin in allowing cytosolic access to Granzyme B.25, 32 Granzyme B has saturable binding sites on the cell surface, and although the evidence is conflicting, the Mannose-6-Phosphate receptor has been implicated in Granzyme B receptor-mediated endocytosis.32, 33, 34, 35 Dynamin, a protein with roles in vesicle trafficking events, appears to be important for Granzyme B entry, especially when Granzyme B is complexed with serglycin in what may be its more physiologically relevant form.35, 36, 37

Granulysin

Granulysin is a cytolytic molecule related to the Saposin-like family, proteins involved in lipid and/or membrane degradation.38, 39 It is found in the granules of human CTLs and NK cells where it exists in 9 and 15 kDa forms.40, 41, 42, 43 It is known for its ability to act as anti-microbial and anti-tumor agent.39 In vitro, it exhibits anti-microbial activity against a broad spectrum of pathogens including bacteria, parasites, and fungi.44 For example, it kills the pathogen M. tuberculosis in infected macrophages in a Perforin-dependent manner and is required for CTL-mediated killing of the fungi C. neoformans.44, 45 Granulysin also displays lytic activity on cells.46 In Jurkat T cells, it stimulates elevations of intracellular Ca2+ that lead to the opening of Ca2+-dependent K+ channels and a decrease in intracellular K+.47 It also elicits mitochondrial membrane potential changes (Delta Psim), and the release of Cytochrome c, Apoptosis Inducing Factor (AIF), and reactive oxygen species (ROS) from the mitochondria.47, 48, 49

Protection from Self-destruction

Given the potency of CTL-mediated lytic mechanisms, processes must be in place for protection from self-destruction. Some studies suggest that CTLs may be generally resistant to granule-mediated cytotoxicity. For instance, Proteinase Inhibitor 9 (PI9; SPI6 in mouse) is an endogenous Serpin family protease inhibitor with the potential to suppress the activity of Granzyme B in CTLs and other cell types.50, 51, 52, 53 However, it has clearly been shown, that CTLs are not entirely exempt from granule-mediated cytotoxicity suggesting that other protective mechanisms may be in place.54, 55

Cathepsin B

A membrane-associated form of the cysteine protease Cathepsin B is found in the cytolytic granule. Upon granule fusion and release of its contents, it is transferred to the surface of the CTL.56 Because Cathepsin B can cleave Perforin, it may form a barrier, offering the effector cell protection from Perforin-mediated cytotoxicity.56

Calreticulin

Also found in the granules is the Ca2+-binding protein Calreticulin.57 In the granule, Calreticulin co-localizes with Perforin, but upon granule release and exposure to higher Ca2+ concentrations, the two likely dissociate.58 Calreticulin can suppress Perforin-mediated lytic activity, and in this capacity may protect the effector cell against Perforin effects on the granule or cell membrane.58, 59, 60

Others

Fas Ligand

In addition to granule-mediated cytotoxic mechanisms, CTLs are well known for utilizing the activation of "death receptor" pathways.61, 62, 63 Death receptors are activated by members of the tumor necrosis factor (TNF) superfamily including TNF-alpha (TNFSF2), Fas Ligand (TNFSF6), and TRAIL (TNFSF10).64 Activation of death receptors leads to a cascade of intracellular events including Cytochrome c release from mitochondria and activation of apoptosis-mediating Caspases.65 For instance, Fas Ligand is a secreted or membrane-associated cytokine that binds to its death receptor Fas (TNFRSF6). This leads to the recruitment of the death domain (DD)-containing adaptor protein FADD and the activation of pro-apoptotic initiator Caspases.65, 66, 67 Reports suggest that Fas Ligand is found in the granules of CTLs and that it makes its way to the membrane upon granule release where it could contribute to target cell killing.68, 69 Sequestering Fas Ligand to the secretory granule and targeting its cell surface expression could protect bystander cells from its cytotoxic effects.

Chemokines

Chemokines make up a large family of small secreted proteins known to mediate the chemotaxis of leukocytes and other cell types.70 Among many sources during an immune response, chemokines are constitutively secreted by CTLs, stimulating the activation and recruitment of leukocytes to regions of inflammation.71 Some studies also suggest that chemokines exist in the lytic granule where their secretion may be regulated by TCR recognition of antigen. For instance, RANTES/CCL5 and MIP-1 alpha/CCL3 have been localized to the cytolytic granules of HIV-specific CTLs, and RANTES/CCL5 and IP-10/CXCL10 are found in the granules of CTLs localized to inflammatory skin lesions.72, 73 Chemokines may also be stored in a compartment separate from the lytic granule but still secreted upon antigen ligation.74 Although the function of antigen-regulated secretion of chemokines at the immune synapse is unclear, it has been suggested that it could provide a positive feedback loop, enhancing T cell activation or acting as a mechanism for self recruitment.73, 74, 75

Granzymes

The Granzymes are serine protease family members structurally related to Chymotrypsin and exhibit the characteristic catalytic triad His-Asp-Ser at the active site.76 They are key components of the lytic granule and are responsible for initiating many of the apoptosis-associated events that occur in cells targeted for destruction by cytotoxic cells. They are produced as zymogens and are activated by cleavage of an N-terminal pro-peptide by Cathepsin C (also known as Dipeptidyl Peptidase I), a protease localized to the granule.77 CTLs from Cathepsin C-deficient mice contain normal levels of Granzymes A and B, but the enzymes retain their pro region and are inactive.78 There are at least 12 Granzymes. Five of these, Granzymes A, B, H, K, and M, have been described in humans.7980 Granzymes A to G, J, K, M, and N are found in rodents, while human Granzyme H has no defined rodent homolog. Granzymes A and B are by far the most widely studied members of the family, while the others, known as the "orphan Granzymes," are less understood.

Figure 2. Granzyme A targets components of the SET complex, releasing NM23-H1 to elicit single-stranded DNA nicks.
View Larger Image 		Figure 2. Granzyme A targets components of the SET complex, releasing NM23-H1 to elicit single-stranded DNA nicks. Other Granzyme A substrates include core and H1 histones and the nuclear Lamins A through C.
Granzyme A

Granzyme A, the most abundantly expressed Granzyme in CTLs, initiates Perforin-dependent events that lead to apoptosis (Figure 2).81 It exists as a disulfide-linked homodimer and works in an independent yet synergistic manner with Granzyme B to initiate apoptosis.82, 83, 84 It is a tryptase that cleaves synthetic substrates with a P1 arginine or lysine and is able to induce many of the typical cellular characteristics associated with apoptosis including membrane blebbing, chromatin condensation, and Delta Psim.81, 85, 86, 87 Although the precise mechanisms underlying these apoptotic effects continue to be elucidated, they appear to occur independent of Caspases and Cytochrome c and are unaffected by anti-apoptotic Bcl-2.87 In addition, Granzyme A does not induce typical oligonucleosomal DNA fragmentation but elicits distinctive single stranded DNA nicks.87, 88 Recently, a series of studies has uncovered Granzyme A targets responsible for this unique form of DNA damage. The SET complex is a large (270-420 kDa) endoplasmic reticulum-associated multi-protein oligomer that causes Granzyme A-dependent DNA nicks in isolated nuclei.89, 90 It translocates from the ER to the nucleus upon treatment with Perforin and Granzyme A or when a cell is targeted by CTLs.90 SET components include the DNA repair enzyme APE, the PP2A inhibitor pp32, the DNA bending protein HMG-2, the tumor metastases suppressor NM23-H1, and its inhibitor SET. APE, HMG-2, and SET all act as substrates for Granzyme A.90, 91, 92 Specifically, the cleavage of SET by Granzyme A releases NM23-H1 from repression, allowing this Mg2+-dependent DNase to cause the characteristic DNA nicks.

Histones and DNA come together to form the nucleosome, the unit particle of chromatin responsible for the packaging of DNA. DNA coils around an octomeric core of nucleosomal/core histones and is further packed into higher order molecular structures by linker H1 histones. Both the core and H1 histones are targets of Granzyme A.93 Perforin-dependent cellular loading of Granzyme A leads to complete degradation of H1 and cleavage of core histones into approximately 16 kDa fragments. The disruption of the nucleosome likely enhances the exposure of chromatin to DNases and facilitates apoptotic processes.93

Apoptotic events include a breakdown of the nuclear lamina. The lamina is composed of a meshwork of intermediate filaments termed Lamins A through C. They are localized to the inner surface of the nuclear envelope, providing structural support, regulating the arrangement of membrane pores, and interacting with and regulating chromatin.94, 95 They are targeted by several different apoptosis-related enzymes including some Caspases, Granzyme A, and Granzyme B.96, 97, 98, 99 Granzyme A is shown to cleave Lamins A through C, while Granzyme B cleaves Lamin B.96 The Perforin-dependent loading of Granzyme A, and to a lesser extent Granzyme B, results in the dissolution of the lamina in intact cells.96

Figure 3. Granzyme B entry into the cytosol unleashes a cascade of events contributing to apoptosis.
View Larger Image 		Figure 3. Granzyme B entry into the cytosol unleashes a cascade of events contributing to apoptosis. Substrates include Bcl-2 family members Mcl-1 and BID, leading to the downstream release of Cytochrome c from the mitochondria and formation of the apoptosome. Also released from the mitochondria are the IAP suppressors SMAC/Diablo and HTRA2/Omi, and mediators of chromatin condensation and DNA fragmentation, AIF and Endo G. Granzyme B may also directly target Lamin B, Procaspase-3, Procaspase-8, and ICAD. The endogenous Serpin PI9 (mouse SPI6), and several viral proteins (US3, L4-100kDa, CrmA) are known to suppress Granzyme B activity.

Granzyme B

Granzyme B is the most widely studied member of the Granzyme family, and it is a potent activator of Caspase-dependent and independent apoptosis (Figure 3). It is an Aspase, cleaving after aspartic acid residues, although hydrolysis is dependent upon extended substrate interactions with an optimal P4-P2 substrate sequence determined to be Ile/Val, Glu/Met/Gln, Pro/Xaa.100, 101

Granzyme B has an increasing array of cellular substrates that contribute to apoptosis. For instance, it initiates a Caspase-dependent pathway via cleavage of the Bcl-2 family member BID. Cleaved BID (tBID) then translocates to the mitochondria initiating Bax/Bak-mediated Delta Psim and a release of Cytochrome c.102, 103, 104, 105 Granzyme B may also indirectly initiate BID cleavage by targeting and activating Procaspase-8.106 Although cleavage of BID by Granzyme B and Caspase-8 occur at a slightly different position, both proteases appear to have similar effects on BID activity.107 Some studies suggest that Granzyme B might also have pro-apoptotic effects on the mitochondria independent of BID, Bax, or BAK.108 For instance, cleavage of anti-apoptotic Bcl-2 family member Mcl-1 by Granzyme B may release pro-apoptic BIM from repression and lead to Cytochrome c release.109 In addition, Granzyme B is shown to elicit BID-independent Delta Psim and nuclear fragmentation via direct activation of Procaspase-3.110

Once released from the mitochondria, Cytochrome c triggers the formation of the apoptosome and elicits an intensely studied cascade of events that lead to apoptosis.65, 67, 111, 112 Briefly, Apoptosis Protease-activating Factor 1 (APAF-1) oligomerizes in a Cytochrome c and ATP-dependent manner. Seven APAF-1 molecules, associating via their Caspase-recruitment Domains (CARD), recruit and activate CARD domain-containing Procaspase-9. The multimolecular megadalton complex that forms is termed the apoptosome.113, 114, 115, 116 Active Caspase-9 is then free to cleave and activate downstream apoptosis-related effector Caspases including Caspase-3 and -7.117 Among the "death substrates" of effector Caspases is ICAD (or DFF45), an inhibitor of DNase Caspase-activating deoxyribonuclease (CAD).118 When cleaved, ICAD releases CAD from repression, freeing it to translocate to the nucleus and initiate oligonucleosomal DNA fragmentation. Highlighting its role in Caspase-independent apoptosis, Granzyme B also directly targets ICAD.119, 120 Other Granzyme B pro-apoptosis substrates include PARP, NUMA, Filamin, CD3 zeta, U1-70kD, and DNA PKcs.121

In addition to Cytochrome c, other mediators of apoptosis are also released from the mitochondria. For instance, AIF and Endonuclease G (Endo G) are freed, allowing them to translocate to the nucleus and mediate chromatin condensation and DNA fragmentation.122, 123, 124 Further enhancing the apoptosis cascade, HTRA2/Omi and SMAC/Diablo are released from the mitochondria in response to Granzyme B.125, 126 These proteins inhibit the activity of members of the inhibitors of apoptosis (IAP) family, releasing pro-apoptotic Caspases from suppression.126, 127, 128, 129, 130

Several viruses have developed mechanisms to circumvent Granzyme B-mediated apoptosis, thus emphasizing its importance in the defense against intracellular pathogens. The adenovirus assembly protein, L4-100 kDa, is a Granzyme B inhibitor that protects infected cells from apoptosis.131 Interestingly, L4-100 kDa appears to suppress only human Granzyme B and not rodent, despite the conserved nature of the Granzyme.132 Herpes simplex virus 1 (HSV-1)-infected cells are protected from CTL-mediated apoptosis via the activity of US3, a viral protein kinase with several apoptosis-related substrates including Granzyme B.133, 134, 135, 136, 137 US3-mediated phosphorylation of Granzyme B suppresses its cleavage of BID in vitro and potentially contributes to HSV-1 evasion of the CTL-mediated immune response.138 Poxvirus-encoded cytokine-response modifier A (CrmA) is a Serpin family inhibitor of serine proteases. It is a well-known suppressor of death receptor activated Caspase-8 and CTL-mediated apoptosis.139, 140, 141 CrmA has been shown to bind Granzyme B as well, inhibiting its proteolytic activity.142

  Granzyme C Granzyme K Granzyme M
Chromation Condensation Yes Not Determined Yes
DNA Fragmentation No Yes No
DNA Nicks Yes Not Determined Not Determined
Cytochrome c Release Yes No No
Caspase-dependent No No No
Inhibition by Bcl-2 Not Determined Yes No
Mitochondria Delta Psim; Swelling Delta Psim;ROS No
Table 2. Orphan Granzymes with known Cytolytic functions

Orphan Granzymes

Relative to Granzymes A and B, very little is known about the activities of the remaining members of the Granzyme family.143 To date, only Granzymes C, K, and M are shown to have potential cytolytic activities (Table 2). Granzyme C has thus far been described only in rodents. In vitro, it induces several features of apoptosis.144 It induces single-stranded DNA nicks, although in contrast to Granzyme A nicks, Granzyme C-mediated nicks are TUNEL-positive.144 It directly induces swelling and Delta Psim in isolated mitochondria, as well as Perforin-dependent Delta Psim and Cytochrome c release in intact cells. Interestingly, there is no evidence for Granzyme C-mediated Caspase activation despite the release of Cytochrome c.144 Granzyme K is a tryptase, and among the Granzymes, it is most closely related to Granzyme A.76 In vitro, rodent Granzyme K induces Perforin-dependent cell death that is accompanied by the production of ROS, Delta Psim, and DNA fragmentation but is independent of Cytochrome c release.86, 145 Human Granzyme M expression is restricted primarily to NK cells, gamme delta T cells, and CD3+CD56+ T cells.146 It appears to prefer substrates with a P1 hydrophobic aa such as Met, Leu, or Pro at P2, and Ala, Ser, or Asp at P3.147, 148, 149 It has been shown to stimulate Perforin-dependent cell death with a time-course similar to, or faster than, Granzyme B.150 Although much remains to be learned about its activity, it appears to function via novel mechanisms that differ from the other Granzymes. Although some chromatin condensation is observed with Perforin/Granzyme M-treatment in vitro, cell death appears to be independent of Caspase activation, mitochondrial perturbation, perceivable DNA fragmentation, and Cytochrome c release and is unaffected by anti-apoptotic Bcl-2.150

Conclusion

Cytolytic lymphocytes are important for defense against intracellular pathogens and formation of tumors. The mechanisms utilized by these cells to recognize and initiate apoptosis in targeted cells are complex. Great strides have been made in recent years, defining many of the molecular mechanisms responsible for their activities, although many questions remain. For instance, the long-held notion that Perforin forms pores in the target membrane continues to seem less likely. Granzymes appear to be taken up by endocytosis, although how Perforin mediates Granzyme access to the cytosol remains a mystery. Many recent studies have greatly advanced our understanding of Granzyme A and B function. Several substrates have been identified that explain many of their effects on cells. Conversely, very little is known about the remaining members of the Granzyme family. It appears that those known to have cytolytic activity (Granzymes C, K, and M) have unique mechanisms with distinct effects on the target cell. Almost nothing is known about their natural substrates or their physiological importance in granule-mediated apoptosis. Whether other members of the Granzyme family have cytolytic activity also awaits further studies. The increasing availability of new tools to investigate these proteins will undoubtedly add to our understanding of their roles in cell-mediated cytotoxicity.

References

  1. Raja, S.M. et al. (2003) Curr. Opin. Immunol. 15:528.
  2. Lieberman, J. (2003) Nat. Rev. Immunol. 3:361.
  3. Call, M.E. & K.W. Wucherpfennig (2004) Mol. Immunol. 40:1295.
  4. van Oers, N.S. et al. (1996) J. Exp. Med. 183:1053.
  5. Iwashima, M. et al. (1994) Science 263:1136.
  6. Salojin, K.V. et al. (2000) J. Biol. Chem. 275:5966.
  7. Williams, B.L. et al. (1999) EMBO J. 18:1832.
  8. Kane, L.P. et al. (2000) Curr. Opin. Immunol. 12:242.
  9. Pitcher, L.A. & N.S. van Oers (2003) Trends Immunol. 24:554.
  10. Geiger, B. et al. (1982) J. Cell Biol. 95:137.
  11. Kupfer, A. & G. Dennert (1984) J. Immunol. 133:2762.
  12. Kuhn, J.R. & M. Poenie (2002) Immunity 16:111.
  13. Monks, C.R. et al. (1998) Nature 395:82.
  14. Stinchcombe, J.C. et al. (2001) Immunity 15:751.
  15. Stinchcombe, J.C. & G.M. Griffiths (2003) Semin. Immunol. 15:301.
  16. Sykulev, Y. et al. (1996) Immunity 4:565.
  17. Haddad, E.K. et al. (2001) J. Cell Biol. 152:835.
  18. Stinchcombe, J.C. et al. (2001) J. Cell Biol. 152:825.
  19. Zhang, B. & D. Ginsburg (2003) Cell 115:372.
  20. Feldmann, J. et al. (2003) Cell 115:461.
  21. Kagi, D. et al. (1994) Nature 369:31.
  22. van den Broek, M.E. et al. (1996) J. Exp. Med. 184:1781.
  23. Uellner, R. et al. (1997) EMBO J. 16:7287.
  24. Young, J.D. et al. (1986) Cell 44:849.
  25. Browne, K.A. et al. (1999) Mol. Cell. Biol. 19:8604.
  26. Shi, L. et al. (1997) J. Exp. Med. 185:855.
  27. Jans, D.A. et al. (1996) J. Biol. Chem. 271:30781.
  28. Trapani, J.A. et al. (1998) Cell Death Differ. 5:488.
  29. Metkar, S.S. et al. (2002) Immunity 16:417.
  30. Galvin, J.P. et al. (1999) J. Immunol. 162:5345.
  31. Raja, S.M. et al. (2002) J. Biol. Chem. 277:49523.
  32. Froelich, C.J. et al. (1996) J. Biol. Chem. 271:29073.
  33. Motyka, B. et al. (2000) Cell 103:491.
  34. Dressel, R. et al. (2004) J. Biol. Chem. 279:20200.
  35. Trapani, J.A. et al. (2003) J. Cell Biol. 160:223.
  36. Veugelers, K. et al. (2004) Blood 103:3845.
  37. Wiejak, J. & E. Wyroba (2002) Cell. Mol. Biol. Lett. 7:1073.
  38. Munford, R.S. et al. (1995) J. Lipid Res. 36:1653.
  39. Clayberger, C. & A.M. Krensky (2003) Curr. Opin. Immunol. 15:560.
  40. Gansert, J.L. et al. (2003) J. Immunol. 170:3154.
  41. Pena, S.V. et al. (1997) J. Immunol. 158:2680.
  42. Pena, S.V. & A.M. Krensky (1997) Semin. Immunol. 9:117.
  43. Hanson, D.A. et al. (1999) Mol. Immunol. 36:413.
  44. Stenger, S. et al. (1998) Science 282:121.
  45. Ma, L.L. et al. (2002) J. Immunol. 169:5787.
  46. Gamen, S. et al. (1998) J. Immunol. 161:1758.
  47. Okada, S. et al. (2003) J. Immunol. 171:2556.
  48. Kaspar, A.A. et al. (2001) J. Immunol. 167:350.
  49. Pardo, J. et al. (2001) J. Immunol. 167:1222.
  50. Hirst, C.E. et al. (2003) J. Immunol. 170:805.
  51. Bird, C.H. et al. (1998) Mol. Cell. Biol. 18:6387.
  52. Buzza, M.S. et al. (2001) Cell. Immunol. 210:21.
  53. Phillips, T. et al. (2004) J. Immunol. 173:3801.
  54. Kupfer, A. et al. (1986) J. Exp. Med. 163:489.
  55. Hanon, E. et al. (2000) Immunity 13:657.
  56. Balaji, K.N. et al. (2002) J. Exp. Med. 196:493.
  57. Dupuis, M. et al. (1993) J. Exp. Med. 177:1.
  58. Andrin, C. et al. (1998) Biochemistry 37:10386.
  59. Fraser, S.A. et al. (2000) J. Immunol. 164:4150.
  60. Fraser, S.A. et al. (1998) Biochem. Cell Biol. 76:881.
  61. Kagi, D. et al. (1994) Science 265:528.
  62. Lowin, B. et al. (1994) Nature 370:650.
  63. Ju, S.T. et al. (1994) Proc. Natl. Acad. Sci. USA 91:4185.
  64. Ware, C.F. (2003) Cytokine Growth Factor Rev. 14:181.
  65. Jiang, X. & X. Wang (2004) Annu. Rev. Biochem. 73:87.
  66. Muzio, M. et al. (1998) J. Biol. Chem. 273:2926.
  67. Lawen, A. (2003) Bioessays 25:888.
  68. Bossi, G. & G.M. Griffiths (1999) Nat. Med. 5:90.
  69. Kojima, Y. et al. (2002) Biochem. Biophys. Res. Commun. 296:328.
  70. Wong, M.M. & E.N. Fish (2003) Semin. Immunol. 15:5.
  71. Conlon, K. et al. (1995) Eur. J. Immunol. 25:751.
  72. Wagner, L. et al. (1998) Nature 391:908.
  73. Iijima, W. et al. (2003) Am. J. Pathol. 163:261.
  74. Catalfamo, M. et al. (2004) Immunity 20:219.
  75. Taub, D.D. et al. (1996) J. Leukoc. Biol. 59:81.
  76. Sattar, R. et al. (2003) Biochem. Biophys. Res. Commun. 308:726.
  77. McGuire, M.J. et al. (1993) J. Biol. Chem. 268:2458.
  78. Pham, C.T. & T.J. Ley (1999) Proc. Natl. Acad. Sci. USA 96:8627.
  79. Smyth, M.J. et al. (2001) J. Leukoc. Biol. 70:18.
  80. Trapani, J.A. (2001) Genome Biol. 2:REVIEWS3014.
  81. Lieberman, J. & Z. Fan (2003) Curr. Opin. Immunol. 15:553.
  82. Garcia-Sanz, J.A. et al. (1990) J. Immunol. 145:3111.
  83. Masson, D. et al. (1986) FEBS Lett. 208:84.
  84. Nakajima, H. et al. (1995) J. Exp. Med. 181:1037.
  85. Shi, L. et al. (1992) J. Exp. Med. 175:553.
  86. Shi, L. et al. (1992) J. Exp. Med. 176:1521.
  87. Beresford, P.J. et al. (1999) Immunity 10:585.
  88. Shresta, S. et al. (1999) Immunity 10:595.
  89. Beresford, P.J. et al. (2001) J. Biol. Chem. 276:43285.
  90. Fan, Z. et al. (2003) Cell 112:659.
  91. Fan, Z. et al. (2002) Mol. Cell. Biol. 22:2810.
  92. Fan, Z. et al. (2003) Nat. Immunol. 4:145.
  93. Zhang, D. et al. (2001) J. Biol. Chem. 276:3683.
  94. McKeon, F.D. et al. (1986) Nature 319:463.
  95. Stuurman, N. et al. (1998) J. Struct. Biol. 122:42.
  96. Zhang, D. et al. (2001) Proc. Natl. Acad. Sci. USA 98:5746.
  97. Alam, A. et al. (1999) J. Exp. Med. 190:1879.
  98. Slee, E.A. et al. (2001) J. Biol. Chem. 276:7320.
  99. Rao, L. et al. (1996) J. Cell Biol. 135:1441.
  100. Thornberry, N.A. et al. (1997) J. Biol. Chem. 272:17907.
  101. Harris, J.L. et al. (1998) J. Biol. Chem. 273:27364.
  102. Wang, G.Q. et al. (2001) J. Exp. Med. 194:1325.
  103. Heibein, J.A. et al. (2000) J. Exp. Med. 192:1391.
  104. Heibein, J.A. et al. (1999) J. Immunol. 163:4683.
  105. Sutton, V.R. et al. (2000) J. Exp. Med. 192:1403.
  106. Medema, J.P. et al. (1997) Eur. J. Immunol. 27:3492.
  107. Luo, X. et al. (1998) Cell 94:481.
  108. Thomas, D.A. et al. (2001) Proc. Natl. Acad. Sci. USA 98:14985.
  109. Han, J. et al. (2004) J. Biol. Chem. 279:22020.
  110. Metkar, S.S. et al. (2003) J. Cell Biol. 160:875.
  111. Saelens, X. et al. (2004) Oncogene 23:2861.
  112. van Gurp, M. et al. (2003) Biochem. Biophys. Res. Commun. 304:487.
  113. Zou, H. et al. (1999) J. Biol. Chem. 274:11549.
  114. Hu, Y. et al. (1998) J. Biol. Chem. 273:33489.
  115. Acehan, D. et al. (2002) Mol. Cell 9:423.
  116. Jiang, X. & X. Wang (2000) J. Biol. Chem. 275:31199.
  117. Slee, E.A. et al. (1999) J. Cell Biol. 144:281.
  118. Liu, X. et al. (1997) Cell 89:175.
  119. Thomas, D.A. et al. (2000) Immunity 12:621.
  120. Sharif-Askari, E. et al. (2001) EMBO J. 20:3101.
  121. Trapani, J.A. & V.R. Sutton (2003) Curr. Opin. Immunol. 15:533.
  122. Daugas, E. et al. (2000) FASEB J. 14:729.
  123. Widlak, P. et al. (2001) J. Biol. Chem. 276:48404.
  124. Li, L.Y. et al. (2001) Nature 412:95.
  125. Sutton, V.R. et al. (2003) Immunity 18:319.
  126. Goping, I.S. et al. (2003) Immunity 18:355.
  127. Du, C. et al. (2000) Cell 102:33.
  128. Verhagen, A.M. et al. (2000) Cell 102:43.
  129. Nachmias, B. et al. (2004) Semin. Cancer Biol. 14:231.
  130. Verhagen, A.M. et al. (2001) Genome Biol. 2:REVIEWS3009.
  131. Andrade, F. et al. (2001) Immunity 14:751.
  132. Andrade, F. et al. (2003) Mol. Cell. Biol. 23:6315.
  133. Ogg, P.D. et al. (2004) Virology 319:212.
  134. Sloan, D.D. et al. (2003) J. Immunol. 171:6733.
  135. Cartier, A. et al. (2003) Exp. Cell Res. 291:242.
  136. Hagglund, R. et al. (2002) J. Virol. 76:743.
  137. Munger, J. & B. Roizman (2001) Proc. Natl. Acad. Sci. USA 98:10410.
  138. Cartier, A. et al. (2003) Cell Death Differ. 10:1320.
  139. Tewari, M. et al. (1995) J. Biol. Chem. 270:22705.
  140. Tewari, M. & V.M. Dixit (1995) J. Biol. Chem. 270:3255.
  141. Zhou, Q. et al. (1997) J. Biol. Chem. 272:7797.
  142. Quan, L.T. et al. (1995) J. Biol. Chem. 270:10377.
  143. Grossman, W.J. et al. (2003) Curr. Opin. Immunol. 15:731.
  144. Johnson, H. et al. (2003) Blood 101:3093.
  145. MacDonald, G. et al. (1999) J. Exp. Med. 189:131.
  146. Sayers, T.J. et al. (2001) J. Immunol. 166:765.
  147. Kelly, J.M. et al. (1996) Immunogenetics 44:340.
  148. Smyth, M.J. et al. (1995) Biochem. Biophys. Res. Commun. 217:675.
  149. Rukamp, B.J. et al. (2004) Arch. Biochem. Biophys. 422:9.
  150. Kelly, J.M. et al. (2004) J. Biol. Chem. 279:22236.