Oncogene (2007) 26, 4979–4986 & 2007 Nature Publishing Group All rights reserved 0950-9232/07 $30.00 www.nature.com/onc ORIGINAL ARTICLE Deregulation of c-Myc in primary effusion lymphoma by Kaposi’s sarcoma herpesvirus latency-associated nuclear antigen D Bubman1, I Guasparri2 and E Cesarman3 1Department of Pathology and Laboratory Medicine, Weill Graduate School of Medical Sciences of Cornell University, New York, NY, USA; 2Department of Pathology and Laboratory Medicine, Weill Graduate School of Medical Sciences of Cornell University, New York, NY, USA and 3Department of Pathology and Laboratory Medicine, Weill Medical College of Cornell University, New York, NY, USA Primary effusion lymphoma (PEL) is a rare subtype of Introduction non-Hodgkin’s lymphoma, which is associated with infec- tion by Kaposi’s sarcoma herpesvirus (KSHV)/human Kaposi’s sarcoma herpesvirus (KSHV), or human herpesvirus-8. The c-Myc transcription factor plays an herpesvirus-8, is a lymphotrophic g2-herpesvirus that is important role in cellular proliferation, differentiation the etiologic agent of Kaposi’s sarcoma (Moore and and apoptosis. Lymphomas frequently have deregulated Chang, 1995), a subset of multicentric Castleman’s c-Myc expression owing to chromosomal translocations, disease (Soulier et al., 1995) and primary effusion amplifications or abnormal stabilization. However, no lymphoma (PEL) (Cesarman et al., 1995a). PEL is a structural abnormalities were found in the c-myc oncogene rare subtype of non-Hodgkin’s lymphoma that is most in PEL. Given that c-Myc is often involved in lympho- common in human immunodeficiency virus-infected magenesis, we hypothesized that it is deregulated in PEL. individuals, usually presenting as lymphomatous effu- We report that PEL cells have abnormally stable c-Myc sions in serous cavities. The expression of KSHV genes protein. The turnover of c-Myc protein is stringently is highly restricted in latently infected PEL cells. The few regulated by post-transcriptional modifications, including latent KSHV proteins expressed subvert various cellular phosphorylation of c-Myc threonine 58 (T58) by glycogen pathways in order to increase the proliferation and synthase kinase-3b (GSK-3b). Our data show that the survival of PEL cells. Most PEL cases are co-infected impaired c-Myc degradation in PEL cells is associated with Epstein–Barr virus (EBV), albeit with very with a significant underphosphorylation of c-Myc T58. restricted EBV latent gene expression (Horenstein The KSHV latency-associated nuclear antigen (LANA) is et al., 1997). Although KSHV is thought to be essential responsible for this deregulation. Overexpression of for PEL development, it is not sufficient, and EBV and LANA in human embryonic kidney 293 or peripheral other cellular factors are likely to play an important role blood B cells leads to post-transcriptional deregulation of (Fan et al., 2005). c-Myc protein. Conversely, when LANA is eliminated c-Myc is a major controller of cellular proliferation, from PEL cells using RNA interference, GSK-3b- and its deregulation is a common cause of malignancies mediated c-Myc T58 phosphorylation is restored. The (Pelengaris et al., 2002a). It is a basic helix–loop–helix- presence of c-Myc and LANA in GSK-3b-containing leucine-zipper transcription factor that regulates the complexes in PEL cells further confirms the significance transcription of numerous specific E-box-containing of these interactions in naturally KSHV-infected cells. genes as a heteromeric complex with a partner protein Oncogene (2007) 26, 4979–4986; doi:10.1038/sj.onc.1210299; Max (Henriksson and Luscher, 1996; Grandori and published online 19 February 2007 Eisenman, 1997). Constitutive expression of c-Myc suppresses exit from the cell cycle, interferes with cell Keywords: KSHV; HHV-8; PEL; c-Myc; LANA differentiation and can cause apoptotic cell death (Pelengaris et al., 2002b). Additionally, elevated expres- sion of c-Myc contributes to tumor formation in transgenic mice (Pelengaris and Khan, 2003). Frequent c-Myc deregulation occurs in lymphomas owing to chromosomal translocations and gene ampli- fication. The proto-oncogene is translocated to the immunoglobulin loci in nearly all Burkitt’s lymphomas (BL) (Dalla-Favera et al., 1982; Taub et al., 1982) and Correspondence: Associate Professor E Cesarman, Department of amplified in 16% of diffuse large B-cell lymphomas Pathology and Laboratory Medicine, Weill Medical College of Cornell (DLBCLs) (Rao et al., 1998). Mutations can result from University, 1300 York Ave, Room C410, New York, NY, USA. E-mail: [email protected] the germinal center somatic hypermutation machinery Received 17 June 2006; revised 12 August 2006; accepted 12 August in B cells (Gaidano et al., 2003) and can be found in the 2006; published online 19 February 2007 translocated alleles of primary tumors and cell lines c-Myc deregulation in KSHV-associated lymphomas D Bubman et al 4980 derived from BL and DLBCL. These mutations can be Results in regulatory elements and affect transcription (Cesar- man et al., 1987) or occur in coding regions (e.g., Myc c-Myc is stabilized in PEL cell lines Box 1 and threonine 58 (T58)), leading to impaired Previous studies evaluating c-myc in PEL did not find proteasome-mediated degradation of c-Myc protein translocations or mutations in the regulatory region of (Rabbitts et al., 1983; Bhatia et al., 1993; Yano et al., the proto-oncogene (Arvanitakis et al., 1996; Gaidano 1993). et al., 1996; Nador et al., 1996). However, analyses of Because of its transforming potential, c-Myc is a mutations in the coding region of c-Myc have not been short-lived protein with a strictly regulated half-life of reported. Therefore, we sequenced exon 2 of c-Myc about 30 min (Hann and Eisenman, 1984; King et al., DNA from a PEL cell line, BC-3 and found no 1986). Fine tuning of protein stability is achieved mutations in the Myc Box 1. As abnormal stabilization through phosphorylation of two specific residues in is a mechanism for c-Myc deregulation, we sought to Myc box 1 located in the N-terminal transactivation examine c-Myc protein turnover in PEL cell lines. BC-1, domain, T58 and serine 62 (S62) and subsequent BC-3, BC-5 and BCBL-1 cells were treated with a proteasomal degradation (Dominguez-Sola and protein synthesis inhibitor, cycloheximide and c-Myc Dalla-Favera, 2004). These phosphorylation events degradation was monitored using immunoblot analysis. take place in an orderly manner during the G1 stage BL cell lines Daudi and JD38 were previously reported of the cell cycle through the stimulation of Ras by to have a normal and a prolonged half-life of c-Myc, growth factors and its activation of the Raf-MAPK/ respectively (Gregory and Hann, 2000), and were used ERK kinase-extracellular signal-regulated kinase and as a negative and a positive control. Figure 1 shows that phosphatidylinositol 30-kinase/Akt pathways. Phos- the steady-state level of c-Myc declined rapidly at 30 min phorylation of c-Myc on S62 is necessary for its in Daudi as reported, but decayed at a much slower pace phosphorylation by glycogen synthase kinase-3b in the PEL cell lines. All PEL cells exhibited a similar or (GSK-3b) on T58, which is crucial for c-Myc degrada- more stable turnover profile to that of JD38 (positive tion in late G1 (Gregory and Hann, 2000; Sears et al., control), which has a deregulation in c-Myc stability. 2000). Subsequent dephosphorylation of S62 by protein We also looked at the levels of LANA protein in PEL phosphatase 2A (PP2A), is required for a proper cells, which went down slightly at the 2-h time point. recognition of c-Myc by the proteasomal machinery Immunoblot for actin confirmed equal protein loading. (Yeh et al., 2004). Toextend the cycloheximide experiments pulse chase Although aberrant c-Myc appears to be important for studies were performed, which confirmed that PEL cells lymphomagenesis, previous studies failed to detect have abnormally stable c-Myc (data not shown). From structural abnormalities in the c-myc gene in PELs these experiments we concluded that KSHV-infected (Arvanitakis et al., 1996; Gaidano et al., 1996; Nador PEL cell lines have an abnormally stable c-Myc protein. et al., 1996). Therefore, we wanted to determine whether Although c-Myc appears to have a shorter half-life it is deregulated in this malignancy, and if so, identify in the KSHV þ /EBV þ cell lines BC-1 and BC-5 than in the KSHV gene responsible for the deregulation. A the KSHV þ /EBVÀ cell lines BC-3 and BCBL-1, possible candidate is the latency-associated nuclear further studies are needed toascertain the effect of antigen (LANA). LANA is one of the few latent genes EBV on c-Myc stability in PEL cells. expressed in KSHV-infected PEL cell lines. It tethers viral DNA to host chromosomes to ensure efficient viral LANA enhances c-Myc stability in HEK 293 cells replication (Ballestas et al., 1999). In addition, LANA To test the hypothesis that LANA is responsible for associates with various proteins including p53 and pRb c-Myc deregulation, we transfected human embryonic and through these interactions stimulates pathways kidney (HEK) 293 cells with a FLAG-tagged c-Myc and involved in survival and proliferation (Friborg et al., empty vector or LANA-expressing vectors and per- 1999; Radkov et al., 2000). Recently, it was found that formed protein stability studies. Figure 2a and b shows LANA promotes stabilization of b-catenin in PEL cells that transfection of LANA significantly increased the by inhibiting its negative regulator, GSK-3b (Fujimuro stability of transfected c-Myc-FLAG protein. We et al., 2003). GSK-3b phosphorylates numerous sub- observed the same effect when looking at endogenous strates, including more than a dozen transcription c-Myc stability in LANA-transfected cells (data not factors, such as activator protein-1 (AP-1), cAMP shown).