Tumor viruses and the ubiquitin–proteasome system

Modification of proteins by covalent attachment of ubiquitin or ubiquitin-related polypeptides and the degradation of some of the conjugates by the proteasome participate in the regulation of fundamental cellular processes (reviewed in Glickman and Ciechanover, 2002). It is therefore not surprising that many pathogens have devoted a considerable part of their genome to the production of proteins that mimic, block or redirect the activity of the ubiquitin–proteasome system in order to modify the cellular environment and protect the infected cells from the host immune attack. Many examples are provided by DNA tumor viruses and retroviruses that exploit the ubiquitin-dependent proteolytic machinery to promote their replication, inhibit apoptosis and manipulate the host immune responses (Figure 1).

Figure 1.

Manipulation of the ubiquitin proteasome system by human tumor viruses. Proteins are targeted for proteasomal degradation by conjugation of a polyubiquitin tree, which requires the activity of a ubiquitin activase (E1), a ubiquitin conjugase (E2) and a ubiquitin ligase (E3). The polyubiquitinated protein is then bound to the proteasome that unfolds the substrate and cleaves the protein in small peptide fragments. Ubiquitin is recycled through the activity of deubiquitinating enzymes (DUB). Different components of the ubiquitin–proteasome system are targeted by the oncogenic proteins of HPV, HBV, HTLV-1, KSHV and EBV

Full figure and legend (186K)

The first striking evidence for the ability of tumor viruses to capture the ubiquitin–proteasome system for their own purposes was provided by the demonstration that human papilloma viruses (HPV) associated with anogenital malignancies utilize the nonstructural proteins E6 and E7 to orchestrate the destruction of two negative regulators of cell proliferation, the tumor suppressors pRB and p53 (reviewed in Munger and Howley, 2002). It was later shown that reprogramming of ubiquitin-dependent proteolysis underlies the transforming ability of Tax, the oncogenic protein of the human T-cell leukemia virus (HTLV)-1. Tax binds to p105, the precursor of the p50 subunit of NF-B, and to two subunits of the 20S proteasome, HsN3 and HC9 (Rousset et al., 1996). Targeting to the proteasome promotes the proteolytic maturation of p105 and the establishment of an NF-B-activated, IL-2-dependent autocrine loop that sustains T-cell proliferation (Hemelaar et al., 2001). It appears that also the hepatitis B virus (HBV) X-antigen uses the proteasome to achieve transformation since its function as an AP-1 transcriptional coactivator is critically dependent on sequences involved in binding to the C6-I subunit of the 20S proteasome and the 19S ATPase S4 (Zhang et al., 2000). These interactions may be instrumental for the ability of X-antigen to inhibit the turnover of c-Jun, a well-known substrate of the ubiquitin–proteasome system (Zhang et al., 2000).

Large DNA tumor viruses exploit the ubiquitin–proteasome system to manipulate the host immune response. This is often achieved through the activity of dedicated viral immune-evasion proteins that are dispensable for replication in vitro but contribute to virulence in vivo. Proteins containing RING finger domains, the signature of many ubiquitin ligases, trigger the cytosolic or lysosomal degradation of MHC class I products. A related zinc-finger structure, termed plant homeodomain (PHD), is found in the Kaposi sarcoma herpesvirus/human herpesvirus-8 (KSHV/HHV-8) proteins K3 and K5 (Fruh et al., 2002; Lorenzo et al., 2002). These viral proteins cause the destruction of MHC class I, with K5 downregulating B7.2 and ICAM-1 as well (Coscoy and Ganem, 2001). The PHD domain of K5 was shown to mediate ubiquitination in vitro.

The first identified human tumor virus, Epstein–Barr virus (EBV), is involved in the pathogenesis of a broad spectrum of malignancies of lymphoid and epithelial cell origin (reviewed in Kieff, 1996). This relatively large virus has evolved multiple strategies for parasitizing the complex life cycle of its primary target, the human B lymphocyte; many of these strategies involve modulation of the ubiquitin–proteasome system. With the remarkable exception of EBV-associated lymphomas arising in immunosuppressed patients, genetic alterations of the host cell environment are required for the autonomous growth of EBV-associated malignancies. Recent evidence suggests that modulation of the ubiquitin–proteasome system by cellular oncogenes may contribute to the malignant phenotype. These findings will be reviewed in the context of their significance for EBV persistence and oncogenesis.

EBV latency and oncogenesis

In spite of its association with malignancies, EBV is a largely nonpathogenic virus widespread in all human populations, with over 90% of the adults being life-long carriers. Clinically silent primary infection is common during early childhood, while delay of the first encounter until adolescence or adulthood is often accompanied by a benign self-limiting lymphoproliferative disease known as glandular fever, or infectious mononucleosis (IM) (reviewed in Rickinson and Kieff, 1996). B lymphocytes are the primary targets of EBV infection and the site of virus persistence in healthy carriers. These cells are largely nonpermissive for virus replication but readily express a restricted set of viral genes that are collectively known as the 'latency' genes to distinguish them from the 'lytic' genes that characterize the productive virus cycle. The latency genes encode for six nuclear and three membrane-associated proteins known as the EBV nuclear antigens (EBNA)-1, -2, -3 (or 3A), -4 (or 3B), -5 (or LP) and -6 (or 3C) and the latent membrane proteins (LMP)-1, LMP-2A and LMP-2B (reviewed in Kieff, 1996). These proteins interfere with the cellular control of proliferation and differentiation, which underscores the unique capacity of the virus to induce growth transformation and adapt to various stages of the B-cell life cycle.

Four EBV latency programs have been identified based on the expression of different combinations of viral genes in cells corresponding to different stages of B-cell activation/differentiation (Table 1). Healthy EBV carriers have between 1 and 50 EBV-infected B lymphocytes per million cells in the peripheral blood (Chen et al., 1995; Babcock et al., 1998,1999). These are long-lived memory cells that may express a putative Latency 0, characterized by complete silencing of the viral genome, or Latency I, in which the membrane protein LMP-2A is expressed alone or together with the nuclear antigen EBNA-1 (Chen et al., 1995; Miyashita et al., 1997). In the absence of effective immune surveillance, as observed in vitro or in vivo in immunosuppressed patients, the EBV-infected B cells express a different viral program called Latency III that is characterized by the regular detection of all nine latency proteins. This latency program is associated with autonomous B-cell proliferation, as exemplified by the establishment of EBV-carrying lymphoblastoid cell lines (LCLs) upon infection of B lymphocytes in vitro. The 'growth program' is probably required to expand the pool of infected cells before the establishment of effective immunity, increasing thereby the likelihood of access to the memory B-cell compartment. An intermediate form of latency characterized by the expression of EBNA-1 and the three LMPs (Latency II) has been identified in EBV-infected B lymphocytes that home to the germinal centers of lymphoid follicles (Babcock et al., 2000; Babcock and Thorley-Lawson, 2000). The latent membrane proteins regulate B-cell activation and apoptosis and this is therefore defined as a 'rescue program' that allows survival and maturation of the EBV-infected lymphoblasts into memory cells (reviewed in Thorley-Lawson, 2001). Since very few EBV-infected lymphocytes are present in the blood of healthy carriers at any given time, our knowledge of viral gene expression in these cells rests exclusively on the detection of viral transcripts by highly sensitive PCRs while evidence for protein expression is lacking. However, the existence of discrete latency programs is strongly supported by studies of EBV-associated malignancies (Table 1). Thus, Latency III is expressed in the immunoblast-like cells of EBV carrying lymphoproliferative disorders arising in organ and bone marrow transplant recipients and HIV patients (Thomas et al., 1991), while Latency I is found in EBV carrying Burkitt's lymphomas (BLs) that are phenotypically similar to memory B lymphocytes (Rowe et al., 1986). In line with the germinal center origin of Hodgkin's disease (HD) lymphomas, cells from EBV-positive HD express a Latency II program (reviewed in Niedobitek et al., 2000). Latency II is also expressed in non-B-cell tumors of both hematopoietic and epithelial cell origin, including T-cell lymphomas, NK cell lymphomas and hemophagocytic syndrome lymphomas, nasopharyngeal carcinoma (NPC) and lymphoepitheliomas of the stomach, thymus and salivary glands (reviewed in Knecht et al., 2001; Dolcetti and Masucci, 2003).

Table 1 - EBV latency programs and their expression in B lymphocytes and EBV-associated malignancies.

Full table (51K)

The growth program: LMP-1, a viral oncogene with an unusual processing signal

The Latency III program is potentially very dangerous for the infected host since the uncontrolled proliferation of virus infected cells could be lethal, as illustrated by the highly malignant B-cell lymphomas arising in patients suffering from severe congenital or iatrogenic immune deficiencies. To contain the hazard, the viral proteins associated with the growth program induce a plethora of B-cell activation markers, adhesion and costimulatory molecules and enhance the activity of various components of the antigen presentation machinery, rendering the infected cells highly immunogenic and therefore easily attacked by T-cell-mediated immune responses. LMP-1 plays a major role in driving this phenotype by acting as a master switch that controls the proliferation, survival and immunogenicity of EBV-infected cells.

LMP-1 is the only EBV protein with recognized oncogenic activity. Transfection of LMP-1 into mouse or human fibroblasts and epithelial cells confers tumorigenicity in immunosuppressed animals (Hu et al., 1993; Takanashi et al., 1999; Yang et al., 2000), and LMP-1 transgenic mice develop hyperproliferations and lymphomas (Kulwichit et al., 1998; Curran et al., 2001). LMP-1 contains a short N-terminal cytoplasmic domain followed by six membrane-spanning domains and a large cytoplasmic C-terminal domain that is involved in signaling (reviewed in Kieff, 1996). Through its transmembrane domain LMP-1 forms multimers that localize to lipid rafts together with LMP-2A and a variety of cellular proteins (Rothenberger et al., 2002). LMP-1 contributes to B-cell transformation by acting as a constitutive CD40-like receptor and inducing expression of NF-B through activation of the tumor necrosis factor-receptor associated factor (TRAF) signaling pathway (Mosialos et al., 1995; Kieser et al., 1999). Unlike CD40, LMP-1 lacks an extracellular ligand-binding domain and does not target TRAF2 and -3 for proteasomal degradation, which results in their constitutive activation and sustained activity (Brown et al., 2001). Relevant to this finding may be the observation that LMP-1 does not bind TRAF6, a RING domain ubiquitin ligase that, together with Ubc13/Uev1A, catalyses the synthesis of polyubiquitin chains linked through Lys63 (Deng et al., 2000). LMP-1 is a short-lived protein (Moorthy and Thorley-Lawson, 1990) and is degraded by ubiquitin-dependent proteolysis (Aviel et al., 2000). Interestingly, the single internal lysine residue of the B95.8 LMP-1 is not required for ubiquitination, while modification of the N-terminus results in full stabilizations, suggesting that this is the site of ubiquitin conjugation. This unusual ubiquitination linkage has been described for only three cellular proteins, MyoD (Breitschopf et al., 1998), HPV E7 (Reinstein et al., 2000) and more recently p21 (Bloom et al., 2003). Intriguingly, this very rare modification was shown to be involved also in the degradation of the second EBV membrane protein, LMP-2A (Ikeda et al., 2002). AIP4/Itchy, a member of the Nedd4 ubiquitin ligase family, was shown to ubiquitinate LMP-2A and regulate its activity (Ikeda et al., 2000,2003). It remains to be seen whether the same ligase can also target LMP-1.

LMP-1 orchestrates the interaction of the infected cells with the environment through constitutive activation of NF-B. Two of the major traits of malignant transformation, angiogenesis and invasiveness, are regulated by NF-B via transcriptional activation of cytokines and cytokine receptors that act on the tumor cells and the surrounding stroma. The most intriguing example of this pleiotropic function of LMP-1 comes from Hodgkin's lymphoma where the malignant Hodgkin/Reed–Sternberg cells constitute only a small fraction of the total tumor mass, the remaining being composed of a mixture of nonmalignant lymphocytes, plasma cells, granulocytes and macrophages (reviewed in Herbst, 1996). Similar lymphoid infiltrates are found in all LMP-1-expressing malignancies of epithelial cell origin, which explains their histological definition as lymphoepitheliomas (Iezzoni et al., 1995). Several components of the antigen presentation pathway are also upregulated in LMP-1-expressing cells including the transporters associated with antigen processing (TAPs) (Rowe et al., 1995; Frisan et al., 1996) and the interferon--regulated subunits of the proteasome core and regulatory particles, which results in enhanced enzymatic activity and changed cleavage specificity (Frisan et al., 1998). A more direct mode of regulation is suggested by the finding that LMP-1 contains two peptides with strong homology to an immunosuppressive peptide found in the retrovirus-encoded transmembrane protein p15E (Dukers et al., 2000). Recombinant peptides corresponding to these sequences strongly inhibit CTL and natural killer (NK) cell activity in vitro. Proteasomal processing of LMP-1 may play a role in the generation of these suppressive peptides since their length is within the range of the fragments produced by the proteasome.

The rescue program: LMP-2A and the capture of cellular ubiquitin ligases

Although LMP-2A and its N-terminus truncated form LMP-2B are not required for B-cell transformation in vitro (Longnecker et al., 1993), their regular expression in the most restricted forms of latency suggests that these viral proteins fulfill an important function in the biology of EBV infection in vivo. Latently infected B lymphocytes are the primary site of virus re-activation, a process that must be tightly regulated since virus production is associated with cell death. The events that lead to disruption of latency are poorly understood, but studies with tumor cell lines suggest that triggering of the B-cell receptor (BCR) plays a critical role (Rowe et al., 1992). LMP-2A regulates this process through its ability to inhibit BCR signaling by interfering with the activity of BCR-associated tyrosine kinases (Miller et al., 1994a,1994b). By preventing exit from latency, LMP-2A may rescue antigen-triggered EBV-carrying memory B cells and allow their homing to the germinal centers of lymphoid follicles where LMP-1 may further counteract apoptosis in the absence of survival signals delivered by BCR ligation (reviewed in Masucci and Ernberg, 1994; Thorley-Lawson, 2001).

Recent studies have shed some light on the mechanism of action of LMP-2A. The amino-terminal domain of this viral protein contains eight tyrosines that associate with the cellular protein tyrosine kinases Lyn and Syk via SH2-phosphotyrosine interactions and five proline-rich regions, three of which possess the PxxP core consensus sequence required for interacting with SH3 domains and two of which possess the PPxY core consensus sequence (PY motif) required for interacting with proteins containing WW modules (reviewed in Longnecker et al., 2000). In screening for proteins interacting with the PY domains, two laboratories have independently identified members of the Nedd4 ubiquitin ligase family (Ikeda et al., 2000; Winberg et al., 2000). These include AIP4/Itchy, WWP2/AIP2 and KIAA0439, all containing a C-terminal HECT domain (Homologous to E6-AP C-Terminus). Binding of these ubiquitin ligases to LMP-2A correlates with ubiquitination of Lyn and Syk and accelerated degradation of Lyn. Ubiquitination of Syk does not appear to affect its turnover, although accelerated degradation of a small pool of activated Syk cannot be excluded. Recent findings suggest that interference with Syk phosphorylation, rather than its ubiquitination, may be important for the action of LMP-2A. Phosphorylation of Syk causes constitutive activation of the Syk substrate SLP-65 (SH2 domain-containing leukocyte protein-65), which in turn induces the formation of a ternary complex containing the ubiquitin ligase Cbl, C3G and the proto-oncogene CrkL (CT10 regulator of kinase Like) (Engels et al., 2001). Cbl-b ubiquitinates Syk, which is important for its function as a negative regulator of BCR signaling (Sohn et al., 2003). As discussed in the previous section, LMP-2A is specifically ubiquitinated by AIP4 and WWP2, and this modification is required for modulation of BCR signaling. Although much of the involved interactions remain obscure, these data confirm that the expression of scaffold proteins that recruit components of the ubiquitin–proteasome pathway is a common viral strategy for selective activation/inactivation of cellular substrates.

The hiding program: EBNA-1 and the blockade of proteasomal processing

EBNA-1 is the only EBV protein expressed in all EBV-associated malignancies. This 'necessity' of EBNA-1 is explained by its specific role in the EBV life cycle. EBNA-1 binds to the dyad symmetry and family of repeat sequences in the origin of plasmid replication (oriP) region of the viral genome and coordinates the replication of the viral episomes in parallel with cellular DNA and their partitioning during cell division. EBNA-1 is also a transcriptional regulator that acts on the three EBNA promoters: Wp and Cp and its own latent promoter Qp (reviewed in Leight and Sugden, 2000). The identified functional domains of EBNA-1, including a nuclear localization signal and dimerization and DNA-binding domains, reside in the C-terminal half of the protein while most of the N-terminal half is occupied by a long repetitive sequence exclusively composed of Gly and Ala residues (GAr) that varies in length between different EBV isolates from approximately 60 to more than 300 amino acids (Falk et al., 1995).

The regular and in some cases exclusive expression of EBNA-1 in EBV-associated malignancies has spurred an intensive search for specific T-cell responses that could be selectively boosted in cancer patients. These efforts were initially thwarted by the failure to isolate EBV-specific CTLs capable of recognizing cells infected with recombinant vaccinia or adenovirus vectors expressing EBNA-1 (reviewed in Rickinson et al., 1992). Later studies have demonstrated that EBNA-1-specific T cells exist but they are either of the CD4 type (reviewed in Paludan and Munz, 2003), or recognize antigen-presenting cells fed with exogenous EBNA-1 (Blake et al., 1997) while the endogenous protein is not processed for MHC class I restricted presentation. This peculiar feature of EBNA-1 is due to GAr since removal of the domain resulted in accelerated protein turnover and efficient presentation of peptide epitopes derived from the endogenous protein (Levitskaya et al., 1995; Blake et al., 1997). In line with these findings, deletion of GAr reconstituted the capacity of EBNA-1 to trigger specific rejection responses in a mouse tumor model (Mukherjee et al., 1998). The capacity of EBNA-1 to escape immune recognition may offer an important advantage in the establishment of EBV latency by permitting the continuous expression of this essential protein, whereas other highly immunogenic viral products must be downregulated in order to prevent elimination of the viral reservoir.

Elucidation of the mechanism of action of the GAr is particularly interesting since this is the first example of a protein domain that can block antigen presentation and this feature could be exploited in immune and gene therapy settings. Using an in vitro processing assay, it was shown that the GAr is a specific inhibitor of ubiquitin-dependent proteolysis and acts as a transferable element that, when expressed in the context of other viral or cellular proteasome substrates, abrogates or severely inhibits their degradation (Levitskaya et al., 1997). Several features of this stabilization signal were revealed using a set of GAr chimeras involving IB (Sharipo et al., 1998), p53 (Heessen et al., 2002) and green fluorescent protein (GFP)-based proteasome substrates in mammalian (Dantuma et al., 2000a) and budding yeast (Heessen et al., in press). It was found that the activity of the GAr is independent of its location in the target protein (Levitskaya et al., 1997; Sharipo et al., 1998; Dantuma et al., 2000a; Heessen et al., 2002) and is not restricted by the type of ubiquitin ligase involved in substrate modification (Sharipo et al., 1998; Dantuma et al., 2000a; Heessen et al., 2002). Fusions of the GAr to GFP-based reporters that are targeted for degradation with different efficiencies (Dantuma et al., 2000b) showed that the GAr counteracts the degradation signal in a length-dependent manner (Dantuma et al., 2000a). Indeed, EBNA-1 itself could be targeted for ubiquitin-dependent proteolysis using a strong degradation signal (Dantuma et al., 2000a). The only restriction for the GAr activity appears to be the requirement for a sufficiently long stretch of Ala or similar small hydrophobic residues, preferably interspersed by one, two or three Gly residues that may act by increasing solubility (Sharipo et al., 2001). These findings and the demonstration that ubiquitinated GAr-containing IB cannot form stable complexes with the proteasome (Sharipo et al., 1998) but still interacts with the S5a ubiquitin-binding subunit of the 19S cap (Heessen et al., 2002; Leonchiks et al., 2002) suggest that the highly hydrophobic GAr may compete for a putative recognition site in the proteasome. This possibility is substantiated by the demonstration that a synthetic GAr peptide inhibits the degradation of biotinylated lysozyme in vitro (Leonchiks et al., 2002). Recent findings using a modified GAr that allows chemical crosslinking suggest that the GAr interacts with at least one subunit of the 19S cap (MG Masucci, unpublished observations). Preliminary evidence identifies the interacting subunit as a member of the ATPases that are involved in substrate unfolding and regulate access to the proteolytic chamber in the 20S. If confirmed, this finding supports a model where the GAr interferes with the interaction of the ubiquitinated substrate with the 19S, perhaps by promoting the premature release of the substrate, before efficient unfolding may occur, or its rapid refolding.

The importance of EBNA-1 stability in the context of viral immune escape is questioned by the demonstration that analogous repeat domains found in baboons and rhesus macaques -herpesviruses do not confer protection from CTL lysis in autologous systems (Blake et al., 1999). Although the failure to block antigen processing fully may be explained by the fact that the EBNA-1 homologs have short GAr-like sequences and contain, in addition to Gly and Ala, Ser residues which affects their capacity to influence proteolysis (Sharipo et al., 2001), these findings point to a nonimmunologic role for EBNA-1 stability. Resistance to proteolysis may be required for the capacity of EBNA-1 to act as a transcriptional regulator either alone or together with interacting cellular proteins. Indeed, using both affinity chromatography and TAP-tagging approaches, Holowaty et al. (2003b) have recently shown that EBNA-1 interacts with the isopeptidase USP7, also known as herpesvirus-associated ubiquitin-specific protease, HAUSP. This nuclear enzyme was first identified by virtue of its interaction with the ICP0 protein of herpes simplex virus type 1, which is required for efficient initiation of the HSV-1 lytic cycle (Everett et al., 1997). ICP0 is an E3 ubiquitin ligase that promiscuously activates gene expression and induces the destruction of specific cellular proteins (Boutell et al., 2002). The demonstration that EBNA-1 binds to USP7 confirms that this cellular protein is a common target of herpesviruses, but the significance of this binding in the context of EBV infection remains unclear. USP7 is probably not involved in the turnover of EBNA-1 but may still regulate its activity since disruption of USP7 binding enhanced the capacity of EBNA-1 to replicate oriP-containing plasmids. This may be achieved through regulation of EBNA-1 ubiquitination, which may affect its capacity for protein–protein interaction (Spence et al., 2000; Muller et al., 2001). Alternatively, EBNA-1 may bring USP7 to oriP where it could act on specific cellular substrates. It is also possible that binding to EBNA-1 may interfere with the physiologic function of USP7. USP7 was shown to bind and deubiquitinate p53, resulting in p53 stabilization and p53-dependent growth arrest and apoptosis (Boutell and Everett, 2003). By sequestering USP7, EBNA-1 may destabilize p53 with important consequences for both B-cell immortalization and the development of EBV-associated tumors. This possibility was recently substantiated by the demonstration that a synthetic peptide corresponding to residues 395–450 of EBNA-1 is able to compete with a p53 peptide for binding to USP7 (Holowaty et al., 2003a).

Contribution of cellular oncogenes: c-Myc, a substrate and modulator of the ubiquitin–proteasome system

Since malignancies expressing one or more EBV antigens usually arise in relatively immunocompetent individuals, specific changes must be required to promote evasion from the host immune control. This concept is clearly illustrated by BL, a classical example of how multiple phenotypic characteristics associated with a single genetic alteration may contribute to promote tumor growth. BL is a highly malignant B-cell tumor characterized by the presence of chromosomal translocations involving the c-myc gene on chromosome 8 and one of the immunoglobulin heavy or light chain genes on chromosomes 14, 2 and 22 (reviewed in Klein, 1994). EBV carrying BL is the most frequent childhood malignancy in endemic malaria areas of subtropical Africa and Papua New Guinea and occurs at a 100-fold higher frequency in immunosuppressed HIV-infected adults. A genetically and phenotypically similar but EBV-negative variant of the tumor is found all over the world suggesting that EBV infection acts as a cofactor in the pathogenesis of this malignancy. Only EBNA-1 is usually detected in the EBV positive tumors (Rowe et al., 1986), although tumors expressing EBNA-1 together with the high molecular weight EBNA-3, -4 and -6 were recently described (Kelly et al., 2002). In all cases, the viral EBNA-2 and LMP-1 are not expressed and, as a result, EBV-positive BL cells do not express B-cell activation markers, adhesion and costimulatory molecules and grow as single cell suspensions rather than in clumps (Rowe et al., 1986). Characteristic for the tumor cells are also the low levels of MHC class I and selective loss of certain class I alleles, HLA A11 in particular (Andersson et al., 1991), which may contribute to the poor allostimulatory capacity of these cells in mixed lymphocyte cultures (Avila-Carino et al., 1987). In addition, BL cells are deficient in their ability to present endogenous antigens via the MHC class I pathway (Frisan et al., 1996), which correlates with downregulation of the interferon--inducible subunits of the proteasome and the peptide transporters TAP1 and TAP2 (Rowe et al., 1995; Frisan et al., 1998).

Studies of in vitro EBV-transformed cell lines that recapitulate the BL cell phenotype through regulated expression of EBNA2/LMP-1 and c-Myc have shown that overexpression of c-Myc in the absence of LMP-1 is directly responsible for the altered immunogenicity of BL cells (Frisan et al., 1996,1998; Staege et al., 2002), which also correlates with alterations of ubiquitin-dependent proteolysis (Gavioli et al., 2001). BL cells were shown to be resistant to treatment with doses of proteasome inhibitors that are readily toxic for normal lymphoblasts. Furthermore, the turnover of short- and long-lived protein proceeded virtually undisturbed in BL cells in the presence of high doses of proteasome inhibitors, suggesting that other enzymatic activities participate in this critical cellular function. Indeed, deubiquitinating enzymes and the serine protease tripeptidyl peptidase (TPP)-II were shown to be highly expressed in tumor-derived BL lines and were upregulated upon overexpression of c-Myc in genetically normal cells.

The upregulation of deubiquitinating enzymes is likely to be important for the survival of BL cells since polyubiquitin chains that are generated during substrate degradation are highly toxic. In addition, deubiquitinating enzymes may also play a role in the turnover of specific substrates. A remarkably large number of putative ubiquitin isopeptidases have been identified in the human genome and some of the characterized enzymes appear to have distinct substrate specificity (reviewed in Wilkinson, 2000). Four types of enzymes that hydrolyze ubiquitin bonds are presently known. These include C-terminal hydrolases (UCH) that are involved in the production of ubiquitin from polyubiquitin precursors and the recycling of activated ubiquitin from illegitimate conjugates; ubiquitin-specific proteases (USP) that trim and disassemble polyubiquitin conjugates; a Zn²⁺-dependent isopeptidase, Rpn11/POH1, that was found to be associated with the proteasome (Cope et al., 2002; Yao and Cohen, 2002); and a new family of cysteine proteases known as Otubains (Borodovsky et al., 2002; Balakirev et al., 2003). We have recently initiated studies aiming to characterize the expression of isopeptidases in BL cells using a set of suicide substrates that allow enzymatic labeling (Borodovsky et al., 2002). Several consistent differences were observed in a panel of BL and LCL lines. In particular, UCH-L1 was regularly overexpressed in the tumor-derived cell lines, while some isopeptidases, notably USP7, appeared to be expressed at higher levels in LCL cells. UCH-L1, which is also known as PGP9.5, is an abundant neuronal enzyme and constitutes 1–2% of the brain proteins. Mutations of the UCH-L1 gene have been associated with familial Parkinson's disease (Lansbury and Brice, 2002) and with other neurodegenerative disorders characterized by the formation of protein aggregates, including spinocerebellar ataxia (Fernandez-Funez et al., 2000) and Huntington's disease (Naze et al., 2002). In addition, high levels of UCH-L1 have been detected in a variety of human malignancies including neuroblastoma (Yanagisawa et al., 1998), colon carcinoma (Yamazaki et al., 2002), non-small-cell-lung carcinoma (Brichory et al., 2001; Sasaki et al., 2001), pancreatic carcinoma (Tezel et al., 2000), prostate and breast carcinomas (Schumacher et al., 1994; Aumuller et al., 1999) and renal carcinoma (D'Andrea et al., 1997), and appear to be associated with the more malignant and invasive forms of these tumors. The physiological targets of UCH-L1 are unknown. In vitro, UCH-L1 acts as a C-terminal hydrolase and catalyses the hydrolysis of C-terminal ubiquityl esters and amides (Larsen et al., 1998). In addition, recent evidence suggests that UCH-L1 may have the unique capacity to act both as a deubiquitinating enzyme and as a ubiquitin conjugase, depending on its level of expression and capacity to form enzymatically active homodimers (Liu et al., 2002).

The other enzyme that was found to be upregulated in c-myc-overexpressing cells, TPPII, is an evolutionarily conserved cytosolic serine protease of the subtilisin family that removes tripeptides from the free N-terminus of oligopeptides (Balow et al., 1983,1986; Balow and Eriksson, 1987; Tomkinson et al., 1997). The 138 kDa core subunit of TPPII forms large oligomeric complexes of more than 1000 kDa that are detected in the cytoplasm and associate with the plasma membrane of most cell types (Balow et al., 1983,1986; Rose et al., 1996; Geier et al., 1999). It has been proposed that TPPII may act downstream of the proteasome and accelerate the production of free amino acids from longer precursors by generating tripeptide intermediates that are easily degraded by other cellular exopeptidases (Wang et al., 2000). In addition, TPPII recognizes specific cellular substrates, as it is the main cholecystokinin-inactivating enzyme in the rat brain (Rose et al., 1996) and regulates apoptotic responses by promoting the maturation of procaspase-1 in macrophages infected with the enteropathogenic bacterium Shigella flexneri (Hilbi et al., 2000). Recent evidence suggest that TPPII may participate in antigen processing, being directly involved in the production of antigenic peptides in cells with impaired proteasome activity (Kessler et al., 2003; Seifert et al., 2003). Some of these functions of TPPII are likely to be dependent on its weak endopeptidase activity (Geier et al., 1999; Seifert et al., 2003), which may allow the generation of longer peptides from intact proteins or polypeptide precursors. TPPII was found to be overexpressed in cells adapted to grow in the presence of lethal concentrations of proteasome inhibitors (Glas et al., 1998; Geier et al., 1999; Deng et al., 2000) and was shown to rescue transfected cells from acute intoxication with these inhibitors (Glas et al., 1998; Geier et al., 1999; Wang et al., 2000), suggesting that it may compensate or substitute for some of the vital functions of the proteasome.

The upregulation of TPPII and isopeptidases appears to be a direct consequence of c-Myc overexpression and does not require selection in vivo. The combined effect of these changes could contribute to the deregulation of a wide variety of cellular proteins involved in the control of cell cycle and apoptosis. Like many other transcription factors, the c-Myc oncoprotein is a short-lived substrate of the ubiquitin–proteasome system. Most importantly, the same protein domains regulate the turnover of c-Myc and its transcriptional activation function (Salghetti et al., 1999; Bahram et al., 2000). The half-life of wild-type c-Myc is increased between two- and sixfold in BL (Gregory and Hann, 2000), suggesting that modulation of c-Myc ubiquitination by DUBs may play an important role in determining its turnover. In addition, recent findings suggest a more direct role of ubiquitination in the regulation c-Myc function. The F-box protein Skp2 was shown to interact with c-Myc in vivo and promote its ubiquitination and degradation (Kim et al., 2003; von der Lehr et al., 2003). Skp2 binds to Myc-responsive regions of specific promoters only in the presence of c-Myc, suggesting that Skp2 is recruited to target promoters as a coactivator of c-Myc-induced transcription. The E3 ligase activity of Skp2 appears to be essential for this coactivator function. Conceivably, SCFSkp2 may ubiquitinate and degrade negative regulators of transcriptions. Another possibility is that ubiquitin modifications of c-Myc or other substrates may play a nonproteolytic function by mediating, for example, protein–protein interactions.

Concluding remarks

The unraveling of aberrant ubiquitin/proteasome-dependent proteolysis in EBV-infected cells and the newly gained understanding of how manipulation of the system may assist the virus during latency and oncogenesis suggest new strategies of intervention for EBV-associated pathologies. Interference with the interaction between LMP-2A and BCR-associated tyrosine kinases may release the blockade of productive infection and promote the elimination of virus-infected cells either directly or indirectly, through the exposure of numerous highly immunogenic epitopes that could target the infected cells for destruction by CTLs. Induction of EBNA-1 processing may also sensitize the infected cells to CTL-mediated rejection since EBNA-1-specific precursors appear to be present in virtually all EBV carriers. However, reversing the protective effect of the GAr in living cells may prove very difficult. It may be easier to attempt the destruction of EBNA-1-specific protein–protein interactions especially if, as appears to be the case for USP7/HAUSP, this may interfere with cellular functions that control proliferation and apoptosis. Finally, modulation of components of the ubiquitin–proteasome system and/or associated proteolytic machineries by specific inhibitors may have a profound impact on the growth and immunogenicity of EBV-associated malignancies as exemplified by the reconstitution of sensitivity to CTL lysis in BL cells treated with proteasome inhibitors. Changes in cell phenotype, immunogenicity or growth potential may also be achieved through functional knockout of critical genes by RNA interference.

References

1. Andersson ML, Stam NJ, Klein G, Ploegh HL and Masucci MG. (1991). Int. J. Cancer, 47, 544–550. | PubMed | ChemPort |
2. Aumuller G, Renneberg H, Leonhardt M, Lilja H and Abrahamsson PA. (1999). Prostate, 38, 261–267. | Article | PubMed | ISI | ChemPort |
3. Aviel S, Winberg G, Masucci MG and Ciechanover A. (2000). J. Biol. Chem., 275, 23491–23499. | Article | PubMed | ISI | ChemPort |
4. Avila-Carino J, Torsteinsdottir S, Ehlin-Henriksson B, Lenoir G, Klein G, Klein E and Masucci MG. (1987). Int. J. Cancer, 40, 691–697. | PubMed |
5. Babcock GJ, Decker LL, Freeman RB and Thorley-Lawson DA. (1999). J. Exp. Med., 190, 567–576. | Article | PubMed | ISI | ChemPort |
6. Babcock GJ, Decker LL, Volk M and Thorley-Lawson DA. (1998). Immunity, 9, 395–404. | Article | PubMed | ISI | ChemPort |
7. Babcock GJ, Hochberg D and Thorley-Lawson AD. (2000). Immunity, 13, 497–506. | Article | PubMed | ISI | ChemPort |
8. Babcock GJ and Thorley-Lawson DA. (2000). Proc. Natl. Acad. Sci. USA, 97, 12250–12255. | Article | PubMed | ChemPort |
9. Bahram F, von der Lehr N, Cetinkaya C and Larsson LG. (2000). Blood, 95, 2104–2110. | PubMed | ISI | ChemPort |
10. Balakirev MY, Tcherniuk SO, Jaquinod M and Chroboczek J. (2003). EMBO Rep., 4, 517–522. | Article | PubMed | ChemPort |
11. Balow RM and Eriksson I. (1987). Biochem. J., 241, 75–80. | PubMed |
12. Balow RM, Ragnarsson U and Zetterqvist O. (1983). J. Biol. Chem., 258, 11622–11628. | PubMed |
13. Balow RM, Tomkinson B, Ragnarsson U and Zetterqvist O. (1986). J. Biol. Chem., 261, 2409–2417. | PubMed | ISI | ChemPort |
14. Blake N, Lee S, Redchenko I, Thomas W, Steven N, Leese A, Steigerwald-Mullen P, Kurilla MG, Frappier L and Rickinson A. (1997). Immunity, 7, 791–802. | Article | PubMed | ISI | ChemPort |
15. Blake NW, Moghaddam A, Rao P, Kaur A, Glickman R, Cho YG, Marchini A, Haigh T, Johnson RP, Rickinson AB and Wang F. (1999). J. Virol., 73, 7381–7389. | PubMed |
16. Bloom J, Amador V, Bartolini F, DeMartino G and Magano M. (2003). Cell, 115, 71–82. | Article | PubMed | ISI | ChemPort |
17. Borodovsky A, Ovaa H, Kolli N, Gan-Erdene T, Wilkinson KD, Ploegh HL and Kessler BM. (2002). Chem. Biol., 9, 1149–1159. | Article | PubMed | ISI | ChemPort |
18. Boutell C, Dasis S and Everett RD. (2002). J. Virol., 76, 841–850. | Article | PubMed | ISI | ChemPort |
19. Boutell C and Everett R. (2003). J. Biol. Chem., 278, 36596–36602. | Article | PubMed | ISI | ChemPort |
20. Breitschopf K, Bengal E, Ziv T, Admon A and Ciechanover A. (1998). EMBO J., 17, 5964–5973. | Article | PubMed | ISI | ChemPort |
21. Brichory F, Beer D, Le Naour F, Giordano T and Hanash S. (2001). Cancer Res., 61, 7908–7912. | PubMed | ISI | ChemPort |
22. Brown KD, Hostager BS and Bishop GA. (2001). J. Exp. Med., 193, 943–954. | Article | PubMed | ISI | ChemPort |
23. Chen F, Zou J-Z, di Renzo L, Winberg G, Hu L-F, Klein E, Klein G and Ernberg I. (1995). J. Virol., 69, 3752–3758. | PubMed | ISI | ChemPort |
24. Cope GA, Suh GS, Aravind L, Schwarz SE, Zipursky SL, Koonin EV and Deshaies RJ. (2002). Science, 298, 608–611. | Article | PubMed | ISI | ChemPort |
25. Coscoy L and Ganem D. (2001). J. Clin. Invest., 107, 1599–1606. | PubMed | ISI | ChemPort |
26. Curran JA, Laverty FS, Campbell D, Macdiarmid J and Wilson JB. (2001). Cancer Res., 61, 6730–6738. | PubMed |
27. D'Andrea V, Malinovsky L, Berni A, Biancari F, Biassoni L, Di Matteo FM, Corbellini L, Falvo L, Santoni F, Spyrou M and De Antoni E. (1997). G. Chir., 18, 521–524. | PubMed | ISI |
28. Dantuma N, Heesseen S, Lindsten K, Jellne M and Masucci MG. (2000a). Proc. Natl. Acad. Sci. USA, 97, 8381–8385. | Article | PubMed | ChemPort |
29. Dantuma N, Lindsten K, Glas R, Jellne M and Masucci MG. (2000b). Nat. Biotechnol., 18, 538–543. | Article | PubMed | ISI | ChemPort |
30. Deng L, Wang C, Spencer E, Yang L, Braun A, You J, Slaughter C, Pickart C and Chen ZJ. (2000). Cell, 103, 351–361. | Article | PubMed | ISI | ChemPort |
31. Dolcetti R and Masucci MG. (2003). J. Cell. Physiol., 196, 207–218. | Article | PubMed | ISI | ChemPort |
32. Dukers DF, Meij P, Vervoort MB, Vos W, Scheper RJ, Meijer CJ, Bloemena E and Middeldorp JM. (2000). J. Immunol., 165, 663–670. | PubMed | ISI | ChemPort |
33. Engels N, Merchant M, Pappu R, Chan AC, Longnecker R and Wienands J. (2001). J. Exp. Med., 194, 255–264. | Article | PubMed | ISI | ChemPort |
34. Everett RD, Meredith M, Orr A, Cross A, Kathorai M and Parkinson J. (1997). EMBO J., 16, 1519–1530. | Article | PubMed | ISI | ChemPort |
35. Falk K, Gratama JW, Rowe M, Zou JZ, Khanim F, Young LS, Oosterveer MAP and Ernberg I. (1995). J. Gen. Virol., 76, 779–790. | PubMed | ISI | ChemPort |
36. Fernandez-Funez P, Nino-Rosales ML, de Gouyon B, She WC, Luchak JM, Martinez P, Turiegano E, Benito J, Capovilla M, Skinner PJ, McCall A, Canal I, Orr HT, Zoghbi HY and Botas J. (2000). Nature, 408, 101–106. | Article | PubMed | ISI | ChemPort |
37. Frisan T, Levitsky V, Polack A and Masucci MG. (1998). J. Immunol., 160, 3281–3289. | PubMed | ISI | ChemPort |
38. Frisan T, Zhang QJ, Levitskaya J, Coram M, Kurilla MG and Masucci MG. (1996). Int. J. Cancer, 68, 251–258. | Article | PubMed |
39. Fruh K, Bartee E, Gouveia K and Mansouri M. (2002). Virus Res., 88, 55–69. | Article | PubMed | ChemPort |
40. Gavioli R, Frisan T, Vertuani S, Bornkamm GW and Masucci MG. (2001). Nat. Cell. Biol., 3, 283–288. | Article | PubMed | ISI | ChemPort |
41. Geier E, Pfeifer G, Wilm M, Lucchiari-Hartz M, Baumeister W, Eichmann K and Niedermann G. (1999). Science, 283, 978–981. | Article | PubMed | ISI | ChemPort |
42. Glas R, Bogyo M, McMaster JS, Gaczynska M and Ploegh HL. (1998). Nature, 392, 618–622. | Article | PubMed | ISI | ChemPort |
43. Glickman MH and Ciechanover A. (2002). Physiol. Rev., 82, 373–428. | PubMed | ISI | ChemPort |
44. Gregory MA and Hann SR. (2000). Mol. Cell. Biol., 20, 2423–2435. | Article | PubMed | ISI | ChemPort |
45. Heessen S, Dantuma N, Tessarz P, Jellne M and Masucci MG: FEBS Lett., 555, 397–404.
46. Heessen S, Leonchiks A, Isseva N, Sharipo A, Selivanova G, Masucci MG and Dantuma N. (2002). Proc. Natl. Acad. Sci. USA, 99, 1532–1537. | PubMed |
47. Hemelaar J, Bex F, Booth B, Cerundolo V, McMichael A and Daenke S. (2001). J. Virol., 75, 11106–11115. | Article | PubMed | ChemPort |
48. Herbst H. (1996). Semin. Cancer Biol., 7, 183–189. | PubMed |
49. Hilbi H, Puro RJ and Zychlinsky A. (2000). Infect. Immun., 68, 5502–5508. | Article | PubMed | ISI | ChemPort |
50. Holowaty MN, Sheng Y, Nguyen T, Arrowsmith C and Frappier L. (2003a). J. Biol. Chem., 278, 47753–47761. | Article | PubMed | ISI | ChemPort |
51. Holowaty MN, Zeghouf M, Wu H, Tellam J, Athanasopoulos V, Greenblatt J and Frappier L. (2003b). J. Biol. Chem., 278, 29987–29994. | Article | PubMed | ISI | ChemPort |
52. Hu LF, Chen F, Zheng X, Ernberg I, Cao SL, Christensson B, Klein G and Winberg G. (1993). Oncogene, 8, 1575–1583. | PubMed | ISI | ChemPort |
53. Iezzoni JC, Gaffey MJ and Weiss LM. (1995). Am. J. Clin. Pathol., 103, 308–315. | PubMed | ISI | ChemPort |
54. Ikeda A, Caldwell RG, Longnecker R and Ikeda M. (2003). J. Virol., 77, 5529–5534. | Article | PubMed | ISI | ChemPort |
55. Ikeda M, Ikeda A, Longan LC and Longnecker R. (2000). Virology, 268, 178–191. | Article | PubMed | ISI | ChemPort |
56. Ikeda M, Ikeda A and Longnecker R. (2002). Virology, 300, 153–159. | Article | PubMed | ISI | ChemPort |
57. Kelly G, Bell A and Rickinson A. (2002). Nat. Med., 8, 1098–1104. | Article | PubMed | ISI | ChemPort |
58. Kessler B, Hong X, Petrovic J, Borodovsky A, Dantuma NP, Bogyo M, Overkleeft HS, Ploegh H and Glas R. (2003). J. Biol. Chem., 278, 10013–10021. | Article | PubMed | ISI | ChemPort |
59. Kieff E. (1996). Fields Virology, 3rd edn, Vol. 2, Fiels BN, Knipe DM, Howley PM et al. (eds). Lippincott, Raven Publishers: Philadelphia, pp. 2343–2396.
60. Kieser A, Kaiser C and Hammerschmidt W. (1999). EMBO J., 18, 2511–2521. | Article | PubMed | ISI | ChemPort |
61. Kim SY, Herbst A, Tworkowski KA, Salghetti SE and Tansey WP. (2003). Mol. Cell, 11, 1177–1188. | Article | PubMed | ISI | ChemPort |
62. Klein G. (1994). Antibiot. Chemoth., 46, 110–116.
63. Knecht H, Berger C, Rothenberger S, Odermatt BF and Brousset P. (2001). Oncology, 60, 289–302. | Article | PubMed | ISI | ChemPort |
64. Kulwichit W, Edwards RH, Davenport EM, Baskar JF, Godfrey V and Raab-Traub N. (1998). Proc. Natl. Acad. Sci. USA, 95, 11963–11968. | Article | PubMed | ChemPort |
65. Lansbury Jr PT and Brice A. (2002). Curr. Opin. Cell Biol., 14, 653–660. | PubMed | ISI | ChemPort |
66. Larsen CN, Krantz BA and Wilkinson KD. (1998). Biochemistry (Moscow), 37, 3358–3368.
67. Leight ER and Sugden B. (2000). Rev. Med. Virol., 10, 83–100. | Article | PubMed | ISI | ChemPort |
68. Leonchiks A, Stavropoulou V, Sharipo A and Masucci MG. (2002). FEBS Lett., 522, 93–98. | PubMed |
69. Levitskaya J, Coram M, Levitsky V, Imreh S, Steigerwald-Mullen PM, Klein G, Kurilla MG and Masucci MG. (1995). Nature, 375, 685–688. | Article | PubMed | ISI | ChemPort |
70. Levitskaya J, Sharipo A, Leonchiks A, Ciechanover A and Masucci MG. (1997). Proc. Natl. Acad. Sci. USA, 94, 12616–12621. | Article | PubMed | ChemPort |
71. Liu Y, Fallon L, Lashuel HA, Liu Z and Lansbury Jr PT. (2002). Cell, 111, 209–218. | Article | PubMed | ISI | ChemPort |
72. Longnecker R, Merchant M, Brown ME, Fruehling S, Bickford JO, Ikeda M and Harty RN.. (2000). Exp. Cell Res., 257, 332–340. | Article | PubMed | ISI | ChemPort |
73. Longnecker R, Miller CL, Miao XQ, Tomkinson B and Kieff E. (1993). J. Virol., 67, 2006–2013. | PubMed | ISI | ChemPort |
74. Lorenzo ME, Jung JU and Ploegh HL. (2002). J. Virol., 76, 5522–5531. | Article | PubMed | ISI | ChemPort |
75. Masucci MG and Ernberg I. (1994). Trends Microbiol., 2, 125–130. | Article | PubMed | ChemPort |
76. Miller CL, Lee JH, Kieff E, Burkhardt AL, Bolen JB and Longnecker R. (1994a). Infect. Agents Dis., 3, 128–136.
77. Miller CL, Lee JH, Kieff E and Longnecker R. (1994b). Proc. Natl. Acad. Sci. USA, 91, 772–776. | PubMed | ChemPort |
78. Miyashita EM, Yang B, Babcock GJ and Thorley-Lawson DA. (1997). J. Virol., 71, 4882–4891. | PubMed | ISI | ChemPort |
79. Moorthy R and Thorley-Lawson DA. (1990). J. Virol., 64, 829–837. | PubMed |
80. Mosialos G, Birkenbach M, Yalamanchili R, VanArsdale T, Ware C and Kieff E. (1995). Cell, 80, 389–399. | Article | PubMed | ISI | ChemPort |
81. Mukherjee S, Trivedi P, Dorfman DM, Klein G and Townsend A. (1998). J. Exp. Med., 187, 445–450. | Article | PubMed | ISI | ChemPort |
82. Muller S, Hoege C, Pyrowolakis G and Jentsch S. (2001). Nat. Rev. Mol. Cell Biol., 2, 202–210. | Article | PubMed | ISI | ChemPort |
83. Munger K and Howley PM. (2002). Virus Res., 89, 213–228. | Article | PubMed | ISI | ChemPort |
84. Naze P, Vuillaume I, Destee A, Pasquier F and Sablonniere B. (2002). Neurosci. Lett., 328, 1–4. | PubMed |
85. Niedobitek G, Baumann I, Brabletz T, Lisner R, Winkelmann C, Helm G and Kirchner T. (2000). J. Pathol., 191, 394–399. | PubMed |
86. Paludan C and Munz C. (2003). Curr. Mol. Med., 3, 341–347. | PubMed | ChemPort |
87. Reinstein E, Scheffner M, Oren M, Ciechanover A and Schwartz A. (2000). Oncogene, 19, 5944–5950. | Article | PubMed | ChemPort |
88. Rickinson AB and Kieff E. (1996). Virology, Fields BN, Knipe DM and Howley PM (eds). Lippincott-Raven: Philadelphia, pp. 2397–2446.
89. Rickinson AB, Murray RJ, Brooks J, Griffin H, Moss DJ and Masucci MG. (1992). Cancer Surv., 13, 53–80. | PubMed | ISI | ChemPort |
90. Rose C, Vargas F, Bourgeat P and Schwartz JC. (1996). Neuropeptides, 30, 231–235. | PubMed |
91. Rothenberger S, Rousseaux M, Knecht H, Bender FC, Legler DF and Bron C. (2002). Cell. Mol. Life Sci., 59, 171–180. | Article | PubMed | ISI | ChemPort |
92. Rousset R, Desbois C, Bantignies F and Jalinot P. (1996). Nature, 381, 328–331. | Article | PubMed | ISI | ChemPort |
93. Rowe DT, Rowe M, Evan GI, Wallace L, Farrell P and Rickinson AB. (1986). EMBO J., 5, 2599–2607. | PubMed |
94. Rowe M, Khanna R, Jacob CA, Argaet V, Kelly A, Powis S, Belich M, Croom-Carter D, Lee S, Trowsdale J, Moss DJ, Rickinson AP and Burrows SR. (1995). Eur. J. Immunol., 25, 1374–1384. | Article | PubMed | ISI | ChemPort |
95. Rowe M, Lear A, Croom-Carter D, Davies A and Rickinson A. (1992). J. Virol., 66, 122–131. | PubMed | ISI | ChemPort |
96. Salghetti SE, Kim SY and Tansey WP. (1999). EMBO J., 18, 717–726. | Article | PubMed | ISI | ChemPort |
97. Sasaki H, Yukiue H, Moriyama S, Kobayashi Y, Nakashima Y, Kaji M, Fukai I, Kiriyama M, Yamakawa Y and Fujii Y. (2001). Jpn. J. Clin. Oncol., 31, 532–535. | Article | PubMed | ChemPort |
98. Schumacher U, Mitchell BS and Kaiserling E. (1994). DNA Cell Biol., 13, 839–843. | PubMed | ISI | ChemPort |
99. Seifert U, Maranon C, Shmueli A, Desoutter JF, Wesoloski L, Janek K, Henklein P, Diescher S, Andrieu M, de la Salle H, Weinschenk T, Schild H, Laderach D, Galy A, Haas G, Kloetzel PM, Reiss Y and Hosmalin A. (2003). Nat. Immunol., 4, 375–379. | Article | PubMed | ISI | ChemPort |
100. Sharipo A, Imreh M, Leonchiks A, Brändén CI and Masucci MG. (2001). FEBS Lett., 499, 137–142. | Article | PubMed | ISI | ChemPort |
101. Sharipo A, Imreh M, Leonchiks A, Imreh S and Masucci MG. (1998). Nat. Med., 4, 939–944. | Article | PubMed | ISI | ChemPort |
102. Sohn HW, Gu H and Pierce SK. (2003). J. Exp. Med., 197, 1511–1524. | Article | PubMed | ISI | ChemPort |
103. Spence J, Gali RR, Dittmar G, Sherman F, Karin M and Finley D. (2000). Cell, 102, 67–76. | Article | PubMed | ISI | ChemPort |
104. Staege MS, Lee SP, Frisan T, Mautner J, Scholz S, Pajic A, Rickinson AB, Masucci MG, Polack A and Bornkamm GW. (2002). Proc. Natl. Acad. Sci. USA, 99, 4550–4555. | Article | PubMed | ChemPort |
105. Takanashi M, Li J, Shirakata M, Mori S and Hirai K. (1999). Arch. Virol., 144, 241–257. | PubMed |
106. Tezel E, Hibi K, Nagasaka T and Nakao A. (2000). Clin. Cancer Res., 6, 4764–4767. | PubMed | ISI | ChemPort |
107. Thomas J, Allday M and Crawford D. (1991). Adv. Cancer Res., 57, 329–380. | PubMed |
108. Thorley-Lawson DA. (2001). Nat. Rev. Immunol., 1, 75–82. | Article | PubMed | ChemPort |
109. Tomkinson B, Hansen M and Cheung WF. (1997). FEBS Lett., 405, 277–280. | PubMed |
110. von der Lehr N, Johansson S, Wu S, Bahram F, Castell A, Cetinkaya C, Hydbring P, Weidung I, Nakayama K, Nakayama KI, Soderberg O, Kerppola TK and Larsson LG. (2003). Mol. Cell, 11, 1189–1200. | Article | PubMed | ISI | ChemPort |
111. Wang EW, Kessler BM, Borodovsky A, Cravatt BF, Bogyo M, Ploegh HL and Glas R. (2000). Proc. Natl. Acad. Sci. USA, 97, 9990–9995. | Article | PubMed | ChemPort |
112. Wilkinson KD. (2000). Semin. Cell Dev. Biol., 11, 141–148. | Article | PubMed | ISI | ChemPort |
113. Winberg G, Matskova L, Chen F, Plant P, Rotin D, Gish G, Ingham R, Ernberg I and Pawson T. (2000). Mol. Cell. Biol., 20, 8526–8535. | Article | PubMed | ISI | ChemPort |
114. Yamazaki T, Hibi K, Takase T, Tezel E, Nakayama H, Kasai Y, Ito K, Akiyama S, Nagasaka T and Nakao A. (2002). Clin. Cancer Res., 8, 192–195. | PubMed | ISI | ChemPort |
115. Yanagisawa TY, Sasahara Y, Fujie H, Ohashi Y, Minegishi M, Itano M, Morita S, Tsuchiya S, Hayashi Y, Ohi R and Konno T. (1998). Tohoku J. Exp. Med., 184, 229–240. | PubMed |
116. Yang X, Sham JS, Ng MH, Tsao SW, Zhang D, Lowe SW and Cao L. (2000). J. Virol., 74, 883–891. | Article | PubMed | ISI | ChemPort |
117. Yao T and Cohen RE. (2002). Nature, 419, 403–407. | Article | PubMed | ISI | ChemPort |
118. Zhang Z, Torii N, Furusaka A, Malayaman N, Hu Z and Liang TJ. (2000). J. Biol. Chem., 275, 15157–15165. | Article | PubMed | ChemPort |

Acknowledgements

This research was supported by grants from the Swedish Cancer Society, the Swedish Foundation of Strategy Research, the Petrus och Augusta Hedlunds Stiftelsen, and the Karolinska Institute, Stockholm, Sweden.