Original Article

Oncogene (2007) 26, 4908–4917; doi:10.1038/sj.onc.1210298; published online 19 February 2007

The Shb signalling scaffold binds to and regulates constitutive signals from the Epstein–Barr virus LMP2A membrane protein

L V Matskova1, C Helmstetter1, R J Ingham2, G Gish3, C K Lindholm4, I Ernberg1, T Pawson3 and G Winberg1

  1. 1Microbiology and Tumor Biology Center, Karolinska Institutet, Stockholm, Sweden
  2. 2Department of Biology, University of Victoria, Victoria, British Columbia, Canada
  3. 3Samuel Lunenfeld Research Institute, Mount Sinai Hospital, Toronto, Ontario, Canada
  4. 4Ludwig Institute for Cancer Research, Lausanne Branch, University of Lausanne, Epalinges, Switzerland

Correspondence: Dr LV Matskova, Microbiology Tumor and Cell Biology (MTC), Karolinska Institutet, Nobels vag 16, Box 280, Stockholm SE-171 77, Sweden. E-mail: ludmat@ki.se

Received 17 July 2006; Revised 11 December 2006; Accepted 22 December 2006; Published online 19 February 2007.



The Epstein–Barr virus latency-associated membrane protein LMP2A has been shown to activate the survival kinase Akt in epithelial and B cells in a phosphoinositide 3-kinase-dependent fashion. In this study, we demonstrate that the signalling scaffold Shb associates through SH2 and PTB domain interactions with phosphorylated tyrosine motifs in the LMP2A N-terminal tail. Additionally, we show that mutation of tyrosines in these motifs as well as shRNA-mediated downregulation of Shb leads to a loss of constitutive Akt-activation in LMP2A-expressing cells. Furthermore, utilization by Shb of the LMP2A ITAM motif regulates stability of the Syk tyrosine kinase in LMP2A-expressing cells. Our data set the precedent for viral utilization of the Shb signalling scaffold and implicate Shb as a regulator of LMP2A-dependent Akt activation.


Epstein–Barr virus, latency-associated membrane protein 2A (LMP2A), Syk, Shb, AIP4, Akt



The Epstein–Barr virus (EBV) LMP2A membrane protein can act in at least some respects as a surrogate immunoreceptor. In B cells, it can deliver signals that permit survival of B-cell receptor (BCR)-negative cells in the peripheral circulation (Caldwell et al., 1998; Merchant et al., 2000). These signals were shown to depend on a constitutive activation of the PI3K (phosphoinositide 3-kinase)–Akt signalling pathway (Scholle et al., 2000, Swart et al., 2000) mediated by the LMP2A ITAM motif. Depending on its expression level, LMP2A can direct B-cell differentiation either to the B1 or the follicular (B2) and marginal zone B cells (Casola et al., 2004). How this qualitative difference in B-cell development depends on the LMP2A expression level is not known, but regulation of the Syk tyrosine kinase by interaction with the LMP2A ITAM motif may be of key importance. This interaction not only activates the Syk tyrosine kinase but also juxtaposes it with Nedd4/AIP4 HECT-domain E3 ubiquitin ligases that are constitutively bound by LMP2A (Ikeda et al., 2000; Winberg et al., 2000). The Syk homolog Zap70 in T cells can also bind to the LMP2A ITAMs (Ingham et al., 2005). The mechanism by which LMP2A activates PI3K has not been defined biochemically. Although Syk can mediate PI3K activation in B cells, it is not clear whether the association between Syk and LMP2A leads to both PI3K activation and Syk ubiquitination or whether LMP2A inactivates Syk and another interacting protein mediates PI3K activation. If so, then such a protein must utilize the LMP2A ITAM motif, which was shown to be required for LMP2A-mediated Akt-activation (Scholle et al., 2000; Swart et al., 2000; Morrison and Raab-Traub, 2005). One such regulatory protein is the signalling scaffold Shb, which is recruited to the ITAM SH2-binding motifs of the CD3zeta upon T-cell receptor (TCR)-stimulation (Welsh et al., 1998). The proline-rich N-terminal domain of Shb interacts with the Src-homology 3 (SH3) domains of PI3K (Karlsson et al., 1995). Overexpression of Shb results in activation of both the PI3K–Akt and extracellular regulated kinase pathways (Welsh et al., 2002), whereas downregulation of Shb with siRNA results in a loss of PI3K activation (Holmqvist et al., 2004). Thus, it is conceivable that binding of Shb to LMP2A might provide an alternate mechanism for activating the PI3K–Akt pathway.

In this study, we first show that Shb interacts with LMP2A in EBV-transformed B cells. We then confirm and characterize the interaction biochemically using the epithelial HEK293 cell line, and lastly we provide biochemical evidence that Shb regulates LMP2A-mediated signalling through the PI3K–Akt pathway.

Our data support the conclusion that Shb is recruited to the tyrosine phosphorylated N-terminal domain of LMP2A by an interaction of its SH2 domain with the LMP2A ITAM and by an interaction of its PTB domain with a PTB-binding motif at tyrosine 64 of LMP2A. Mutation of LMP2A tyrosines 64, 74 or 85 impairs the capacity of LMP2A to activate Akt. Conversely, downregulation of either Syk or Shb by specific small hairpin RNAs (shRNAs) leads to a loss of LMP2A-mediated constitutive Akt activation.

We also find that LMP2A expression in HEK293 cells leads to a reduction of the steady-state expression level of Syk protein, whereas downregulation of LMP2A using shRNA against LMP2A stabilizes Syk. ShRNA-mediated downregulation of Shb leads to extensive destabilization of Syk in LMP2A positive cells, but has no effect on Syk levels in control cells. Surprisingly, Syk levels are not stabilized by shRNA-mediated downregulation of AIP4 in LMP2A positive cells, which suggests that ubiquitin ligases other than AIP4 may contribute to the regulation of Syk stability in LMP2A positive cells.



The Shb signalling scaffold interacts with LMP2A in EBV-transformed B cells

In Figure 1a, we show that antibody against LMP2A immunoprecipitates Shb from three EBV-positive lymphoblastoid cell lines: JAC-1, BK and CBMI Ral-Sto, but not from the EBV-negative B-cell line Ramos-NUT. In panel b we show that the reciprocal immunoprecipitation, using Shb antibody, allows detection of LMP2A in the CBMI Ral-Sto LCL, but not in Ramos-NUT. This demonstrates an interaction of endogenous Shb with virally expressed LMP2A in EBV-infected B cells with the latency III phenotype. It appears that the larger of the two isoforms of Shb preferentially binds to LMP2A, as in Figures 1 and 3c, whereas both the 56 and 67 kDa isoforms are detectable in whole-cell lysates (Figures 5 and 6).

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Endogenous Shb associates with virally expressed LMP2A in EBV-transformed lymphoblastoid cells. (a) Immunoprecipitates of LMP2A from the EBV positive lymphoblastoid cell lines (LCL) Jac-1, BK and CBMI-Ral-Sto and from the EBV negative Ramos NUT B cell line were probed with rabbit antiserum against Shb. (b) Reciprocal immunoprecipitates with the Shb antiserum were probed with the 14B7 mAb against LMP2A.

Full figure and legend (35K)

Shb associates with tyrosine phosphorylated LMP2A but not with the unphosphorylated N-terminal tail of LMP2A

We wished to determine whether the association of Shb with LMP2A might resemble the previously established, phosphorylation-dependent binding of Shb to the CD3zeta ITAM following TCR stimulation (Welsh et al., 1998). In this experiment we made use of the 3Tm chimeric transmembrane protein where the LMP2A N-terminal signalling domain was fused to the CD38 exodomain (Winberg et al., 2000). The LMP2A domain in this fusion protein is not detectably tyrosine phosphorylated in HEK 293 cells unless the CD38 exodomain is crosslinked with specific antibody (Dianzani et al., 1994; Winberg et al., 2000). We first observed that antibody to Shb readily immunoprecipitates LMP2A from two cell lines (1C2 and C4) stably expressing the wt LMP2A, which is constitutively phosphorylated on tyrosine (Figure 2a, lanes 2 and 3) but not from the 3Tm cell line, expressing the unphosphorylated CD38 fusion protein (Figure 2a, lane 4).

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Shb and the Shb SH2 domain associate with LMP2A in a phosphotyrosine-dependent manner. (a and b) The association of Shb with LMP2A as a function of LMP2A tyrosine phosphorylation. (c and d) The association of the Shb SH2 domain with LMP2A and its dependence on tyrosine phosphorylation of the LMP2A signalling domain. (a) Immunoprecipitates of LMP2A were probed with the LMP2A mAb 14B7 to show expression of LMP2A in the cell lines used (top panel). Immunoprecipitates were carried out with 14B7 mAb covalently coupled to Sepharose CL-4B to reduce IgH chains in the subsequent immunoblots. The next panel shows a reprobe with antibody to phosphotyrosine demonstrating that the wt LMP2A is constitutively tyrosine phosphorylated, whereas the LMP2A-CD38 fusion protein in the 3Tm cell line is not (Winberg et al., 2000). In the third panel, immunoprecipitates of Shb were probed with LMP2A antibody. The constitutively tyrosine-phosphorylated wt LMP2A in the 1C2 and C4 cell lines (lanes 2 and 3) was brought down by the Shb antibody, whereas the N-terminal domain of LMP2A in the chimeric CD38-LMP2A protein was not detected (lane 4). The last panel shows control immunoprecipitates with normal rabbit serum (NRS). IgH denotes the immunoglobulin heavy chain. (b) The CD38 exodomain of the 3Tm fusion protein was crosslinked either with IB-4 mAb (three top panels) or with OX34 mAb against the rat CD2 exodomain (bottom panel). Since CD2 is absent from 3Tm cells, the OX34 antibody is an irrelevant antibody. After adding the primary crosslinking IB4 and the secondary F(ab)'2 antibody, the cells were either lysed directly (0 min) or incubated 10 min at 37°C before lysis. The primary crosslinking antibody was here utilized as primary antibody for the immunoprecipitation. Thus, Protein G Sepharose was added to the lysates. In the top panel, the immunoprecipitates were probed with the 14B7 mAb against LMP2A. The second panel shows a reprobe with phosphotyrosine antibody (pY), whereas the third and bottom panels were probed with rabbit antiserum against Shb. (c) Glutathione-Sepharose-bound GST protein and a Glutathione-Sepharose-immobilized GST-Shb SH2 fusion protein were compared for their ability to interact with wt LMP2A in cell lysates. C4 and 1C2 are independent HEK 293 cell clones expressing the wt LMP2A. (d) the 3Tm cell line and 293P vector control cells were subjected to crosslinking with the IB4 mAb as described in Materials and methods. Pull-down assays using the GST-Shb SH2 fusion protein were immunoblotted with anti-phosphotyrosine antibody (upper panel) and reprobed with the 14B7 mAb against LMP2A (lower panel). Trace bands of LMP2A at the 0 min time point in (d) likely reflect that a low level of LMP2A tyrosine phosphorylation takes place during handling of the cells before lysis.

Full figure and legend (121K)

Next, we sought to establish whether crosslinking the CD38 exodomain with specific agonistic antibody would stimulate the association of Shb with the LMP2A N-terminal signalling domain. In Figure 2b, we show first that the CD38-LMP2A N-terminal fusion protein underwent tyrosine phosphorylation during a 10-min incubation at 37°C in the presence of agonistic antibody against the CD38 exodomain (Figure 2b, two top panels). Concomitantly with this, Shb association with the CD38-LMP2A fusion protein was greatly stimulated (Figure 2b), whereas control CD2-specific antibody failed to bring down Shb. Traces of Shb binding are detectable also in the absence of a 37°C incubation period after adding the crosslinking antibody (panel b, 0 min time point). This probably reflects a low level of LMP2A tyrosine phosphorylation that may take place during handling of the cells before lysis.

As Shb binding to LMP2A thus appeared to depend on LMP2A tyrosine phosphorylation, we decided to test whether Shb might associate with LMP2A by an interaction involving its SH2 domain. We hypothesized that the Shb SH2 domain might utilize the tyrosine phosphorylated ITAM SH2-binding motifs of LMP2A. In Figure 2c, we first show that the Shb SH2 domain could bring down wt LMP2A in a GST fusion protein-binding assay, whereas the GST protein alone did not. Next, we tested whether induction of LMP2A tyrosine phosphorylation would cause binding of the Shb SH2 domain to the N-terminal tail of LMP2A. In Figure 2d, we show that the GST-Shb SH2 fusion protein captured the chimeric CD38-LMP2A protein within 5 min after crosslinking with CD38 antibody (bottom panel), and that the captured fusion protein was tyrosine phosphorylated (top panel). Under the same conditions, the GST control protein failed to bind LMP2A (data not shown). It cannot be excluded that the association between Shb and the crosslinked LMP2A-CD38 chimera might include further interactions, in addition to the phosphotyrosine-dependent interaction that we deteced using the GST-Shb SH2 domain as a probe.

Both the Shb SH2 and PTB domains participate in binding to LMP2A

Shb has two domains that can bind to phosphotyrosine motifs. Previous studies have determined the optimal binding specificities of the SH2 (pY-T/V/I-X-L, Karlsson et al., 1995) and PTB (D-D-X-pY; Welsh et al., 1998) domains respectively using degenerate peptide libraries. The LMP2A ITAM-binding motif (pY74QPL-(X)7-pY85LGL) and a putative Shb PTB-binding motif (EDPpY64) are close to the predicted optimal-binding motifs. Therefore, we tested the binding of the Shb SH2 and PTB domains using GST fusion protein-binding assays to a set of tyrosine mutants of LMP2A. The results in Figure 3a show that mutations in the ITAM tyrosines (Y74 and Y85, lanes 4, 5 and 6) reduce the binding of the Shb SH2 domain to LMP2A, whereas the Y64F mutant (lane 3) binds the SH2 domain as wt. Conversely, the Y64F mutant (lane 3) as well as the Y64/74F double mutant (lane 5) show a clearly reduced binding of the Shb PTB domain, suggesting that the Y64 motif is the major PTB-binding motif in LMP2A. We cannot exclude, however, that the sequence PDGY23 might also contribute to Shb binding. These results indicate that both the SH2 and PTB domains mediate binding of Shb to LMP2A. These and other relevant deletion mutants of LMP2A were previously shown to be tyrosine phosphorylated in HEK293 (Scholle et al., 2001). Our experimental data in Figure 3 confirm this finding. Phenylalanine substitutions at LMP2A Y60, Y64 and Y101 are also tyrosine phosphorylated in B cells (Swart et al., 1998). As a control for the effect of the mutants selected in Figure 3a, we tested the binding of the Shb SH2 or PTB domain GST fusion proteins to the LMP2A Y112F. In Figure 3b we show that the Y112F mutant (lane 3) LMP2A binds both the Shb SH2 and PTB domains indistinguishably from the wt LMP2A (lane 2). In panel b, we also confirm that the LMP2A Y112F mutant is tyrosine phosphorylated (Scholle et al., 2001). Shb interactions with the LMP2A protein complex appear not to be limited to the two domains tested in Figure 3, since all the LMP2A tyrosine mutants used in Figure 3 are detectable in immunoprecipitates of endogenous Shb in HEK293 cells (not shown).

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Interactions of the Shb SH2 and PTB domains with tyrosine mutants of LMP2A. Lysates from 107 control HEK293 cells, cells stably expressing wt LMP2A or the LMP2A tyrosine mutants Y64F, Y74F, Y64/74F and Y74/85F (a) and lysates from cells expressing FLAG epitope-labelled wt or tyrosine mutant Y112F LMP2A (b) were incubated with Glutathione-Sepharose immobilized GST-Shb SH2, GST-Shb PTB or GST control protein. In each GST pull-down assay, 20 mM phosphotyrosine was included to reduce nonspecific association of the SH2 and PTB domains to unphosphorylated target motifs. Immunoblots of the different GST pull-downs (top three panels) were probed with the 14B7 antibody against LMP2A (a) or with the FLAG-M2 antibody (b). To compare the expression levels of the LMP2A mutants, whole-cell lysates (WCL) from 105 cells of each cell line was probed with the 14B7 antibody and reprobed with antibody against beta-actin to control for protein loading in the lysates. LMP2A immunoprecipitates from similar lysates as above were also probed with anti-phosphotyrosine antibody to determine the tyrosine phosphorylation of the mutants.

Full figure and legend (109K)

Shb and AIP4 regulate Syk stability in LMP2A positive but not in control cells

Downregulation of the Syk tyrosine kinase in normal cells has been attributed to ubiquitination of the activated kinase by the Cbl RING-domain E3 ubiquitin ligase (Ota et al., 2000; Rao et al., 2001). Results from our group (Winberg et al., 2000) and others (Ikeda et al., 2000) show that AIP4 and Nedd4-like HECT-domain E3-ligases are recruited by LMP2A and subsequently catalyse ubiquitination of Syk. Since our above data suggest that Shb binding to the LMP2A ITAM motifs might regulate the binding of Syk to LMP2A, we undertook to compare the effects of AIP4 and Shb on the steady-state expression levels of Syk in LMP2A-expressing and control HEK 293 cells. To this end, we constructed both LMP2A expressing and control cell lines where AIP4 and Shb are downregulated using stably expressed specific shRNAs. As controls, we used shRNAs against LMP2A and GFP.

In Figure 4 we show that downregulation of Shb in LMP2A expressing HEK293 cells resulted in a further destabilization of Syk (lane 3), compared to LMP2A expressing cells (lane 5), whereas the Syk levels in LMP2A negative control cells (lane 1) and control cells with downregulated Shb (lane 2) were unaffected. Utilization of LMP2A ITAMs by Shb might reduce the rate of Syk binding to LMP2A. Conversely, downregulation of Shb might allow an increased rate of binding, activation and subsequent degradation of Syk following exposure to LMP2A-associated ubiquitin ligases.

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Downregulation of Shb or AIP4 reduces steady-state levels of Syk protein in LMP2A-expressing but not in control cells. ShRNAs against LMP2A, Shb and AIP4 were stably expressed in control or LMP2A-expressing HEK293 cells as indicated in the table at the top. Immunoblots were probed with antibody against Syk and then reprobed with antibody against the Nedd4 and AIP4 proteins, Shb and LMP2A, as indicated on the left side of each panel. To normalize for protein loading, filters were reprobed for italic gamma-tubulin. Three independent experiments were used for densitometric scans of the Syk, italic gamma-tubulin and LMP2A immunoblots as described in Materials and methods. The graph shows the average valuesplusminusone standard deviation (s.d.). The numbers above each column represent the Syk pixel densities normalized to italic gamma-tubulin (top graph) or LMP2A (bottom graph). The AIP4 shRNA sequences were designed to specifically suppress AIP4 mRNAs. The rabbit antiserum used to detect AIP4 was raised against conserved sequences in the WW-domains, and so detect also the Nedd4 proteins. Thus, the remaining bands in the AIP4 knockdown cells (lanes 6 and 7) likely represent Nedd4 proteins unaffected by the AIP4 siRNAs.

Full figure and legend (85K)

Unexpectedly, Syk was also significantly destabilized by downregulating AIP4 in LMP2A-expressing cells (lane 6). This effect was not observed in control cells where AIP4 was downregulated (lane 7). This would be in line with the established role of AIP4 as an E3 ubiquitin ligase regulating Cbl stability (Courbard et al., 2002). Downregulation of AIP4 would be expected to lead to stabilization of both Cbl and LMP2A (Ikeda et al., 2002). A greater rate of Syk recruitment by the stabilized LMP2A protein in addition to more available Cbl might result in more rapid degradation of the activated Syk. This indicates that AIP4 has an indirect effect on Syk levels, possibly mediated by regulating Cbl stability, in addition to its previously demonstrated activity as a ubiquitin ligase for LMP2A-bound Syk. The reduction in Syk levels associated with downregulation of Shb or AIP4 was significant whether gamma-tubulin or LMP2A was used for the normalization (Figure 4, inserted graphs). In Figure 7, we present a hypothetical model for how the interactions between LMP2A, Shb, AIP4 and Cbl might influence Syk stability in LMP2A expressing cells.

Downregulation of Shb impairs LMP2A-mediated Akt activation

To understand whether the LMP2A-dependent destabilization of Syk that was observed in cells with downregulated Shb (Figure 4) would also impact on the capacity of LMP2A to activate Akt, we compared the effects of downregulating LMP2A and Syk to those of Shb downregulation in LMP2A-expressing cells. The results in Figure 5 demonstrate that shRNA-mediated downregulation of Shb (lanes 11–14) as well as of Syk (lanes 9–10) and LMP2A itself (lanes 7–8) reduce the ability of LMP2A to activate the PI3K–Akt pathway. Shb downregulation, in addition, results in a concomitant reduction of steady-state levels of Syk (lanes 11–14).

Figure 5.
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

RNA interference against Shb inhibits LMP2A-mediated Akt activation. The effect of shRNA-mediated reduction of the expression levels of LMP2A, Syk and Shb on LMP2A-mediated Akt activation was tested by immunoblotting with a rabbit mAb against a Akt phosphoserine 473 peptide on whole-cell lysates (WCL) of HEK293 control cells and wtLMP2A-expressing HEK293 cells (clone LMP2A2-31) expressing shRNA against LMP2A, Syk or Shb. Two independent clones expressing shRNA against Shb are shown (lanes 11–12 and 13–14). LMP2A-positive cells expressing shRNA against EGFP was used as an irrelevant shRNA control. The presence or absence of the different proviruses is indicated with + or - respectively for each cell line in the table at the top. Each cell line was either grown in 5% FBS or shifted to 0.5% FBS for 72 h before lysis as indicated by the gray bars above each lane. After probing the immunoblot with antibody against phosphoserine 473 of Akt (pS473), densitometry was carried out on three independent experiments as described in Figure 4. For each pair of samples, the pixel density value at 0.5% FBS was first normalized to the value at 5% FBS. In the graph, the average pixel density value at 0.5%plusminus1 s.d. is shown as a percentage of the 5% FBS value. The reprobes for Akt protein as well as beta-actin show the protein loading. Reprobes for LMP2A, Syk and Shb show the effect of shRNA-mediated downregulation of the corresponding transcripts. For Syk, the effect of the specific shRNA inhibition pertains to lanes 9 and 10. Reduced Syk expression in lanes 3–6 and 11–14 as compared to lanes 1–2 and 7–8 may reflect the effects of LMP2A expression and Shb downregulation, respectively, on steady-state expression levels of Syk.

Full figure and legend (97K)

These results indicate that constitutive PI3K–Akt activation is lost because Syk is destabilized as a result of Shb downregulation, as argued above. This would suggest that Shb is a regulator of Syk-mediated activation of the PI3K–Akt pathway in LMP2A-expressing cells. It cannot be excluded that Shb also contributes to Syk-mediated PI3K activation, for example by recruitment of PI3K to its proline-rich N-terminal domain (Karlsson et al., 1995).

Densitometric measurements of activated Akt were performed in cells grown at 0.5% fetal bovine serum (FBS) and the pixel density values were normalized to the Akt-activation in the same cells grown at 5% FBS. The major growth factor in FBS is platelet-derived growth factor, which was shown to activate Akt by direct recruitment of PI3K to the phosphorylated platelet-derived growth factor receptor beta tyrosines Y740 and Y751 (Franke et al., 1995). Thus, serum stimulates the PI3K–Akt pathway independently of LMP2A.

Tyrosine residues involved in Shb binding to LMP2A control constitutive Akt activation

The LMP2A ITAM tyrosines Y74 and Y85 were previously shown to bind the Syk tyrosine kinase (Longnecker et al., 1991) and mutations of these tyrosines were shown to influence Syk binding to LMP2A as well as LMP2A-mediated Akt activation and survival signalling (Scholle et al., 2000; Swart et al., 2000). Here, we confirm the negative effect of the Y74F and Y85F mutations on the ability of LMP2A to activate Akt (Figure 6, lanes 11 and 12) and extend this observation by showing that the Y64F mutation suppresses Akt activation by LMP2A, either alone (lanes 5 and 6) or in combination with the Y74F mutation (lanes 9 and 10). As LMP2A recruitment and activation of Syk is established as a major mechanism for the LMP2A-mediated constitutive activation of Akt, we also included shRNAs against LMP2A and Syk (lanes 13–16) as controls. Activated Akt was detected using antibody specific for phosphoserine 473 of Akt (pS473). Cells coexpressing LMP2A and shRNA directed against enhanced green fluorescent protein (EGFP) were included as an shRNA vector control (lanes 17–18).

Figure 6.
Figure 6 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Akt-activation by LMP2A tyrosine mutants affecting Shb binding to LMP2A. Cell lines expressing LMP2A with the point mutations Y64F, Y74F, Y64/74F or Y74/85F were compared to control HEK293 cells, HEK293 cells expressing wt LMP2A and cells stably coexpressing LMP2A and shRNAs directed against LMP2A or Syk. LMP2A positive cells coexpressing shRNA against EGFP served as a shRNA vector control. As described in Figure 5, Akt activation was tested in duplicate cultures at 5 or 0.5% FBS. Densitometry of immunoblots for Akt pS473 and normalizations were carried out as described for Figure 5. In the graph, the mean and standard deviations of three experiments are plotted for the 0.5% sample for each cell line. The Akt protein immunoblot serves as a control for protein loading.

Full figure and legend (73K)



Several immunoreceptors (BCR, TCR, Fcitalic gammaR, Fcalt epsilonR, myeloid, dendritic and NK cell receptors) rely on recruitment of Syk or Zap70 tyrosine kinases to the twin tyrosines of the ITAM motifs in associated adaptor proteins (CD79alpha and beta, CD3delta, italic gamma, alt epsilon and zeta, FcRitalic gamma and DAP12) to activate signalling after ligand recognition (Fodor et al., 2006). Apart from immunoreceptor signalling, Syk and Zap70 participate in numerous other cellular processes. Activation of Syk has been detected after stimulation of integrin beta chains, although this association does not appear to require the tandem SH2-binding motifs of ITAMs (Koretzky et al., 2006). Binding of Syk to the ITAMs of FcRitalic gamma and DAP12 was shown to regulate the development of osteoclasts (Fodor et al., 2006) and SLP76 and Syk participate in the developmental separation of blood and lymphatic vascular epithelial cells, perhaps dependent on an as yet unidentified ITAM-containing receptor (Koretzky et al., 2006). The observation that Syk acts as a tumor suppressor gene in breast carcinoma (Coopman et al., 2000) has raised the question whether LMP2A might contribute to the tumorigenic development in nasopharyngeal carcinoma through its effects on Syk (Lu et al., 2006).

ITAM motifs are present in a number of viral proteins (Grande et al., 2006). LMP2A and bovine leukemia virus gp 30 were first shown to issue ITAM-dependent signals in T cells (Beaufils et al., 1993). Some functions of the ITAM-containing viral proteins may relate to tumorigenicity, such as survival signalling by activation of the PI3K–Akt pathway by LMP2A and KSHV K1 (Scholle et al., 2000; Swart et al., 2000; Tomlinson and Damania, 2004) and stimulation of cellular migration and invasiveness by LMP2A and MMTV env (Katz et al., 2005; Lu et al., 2006).

Our discovery that the signalling scaffold Shb binds to LMP2A and participates in the regulation of LMP2A-mediated constitutive signals indicates that recruitment of this adaptor protein may occur in a more direct way and at a more proximal position than has been suggested for SLP65 (Engels et al., 2001). The fact that Shb is recruited to the CD3zeta upon TCR stimulation (Welsh et al., 1998) and associates with both SLP76 and Zap70 in T cells (Lindholm et al., 2002) is suggestive of an analogous function in LMP2A-expressing cells.

It is not excluded that Shb may participate in the regulation of Syk or Zap70-mediated signalling from ITAM-containing adaptor proteins other than those associated with the BCR and TCR. The arrangement of Shb PTB and SH2-binding motifs in LMP2A is strongly reminiscent of a similar motif in CD79alpha but putative Shb PTB-binding motifs precede the SH2-binding motifs in several ITAM-containing adaptors (Table 1). Previous work implicated PTB- as well as SH2-domain interactions in the binding of Shb to CD3zeta (Welsh et al., 1998). The finding that LMP2A coopts both Shb and Syk in the regulation of ITAM-mediated activation of the PI3K–Akt pathway may thus exemplify a more general mechanism for regulation of ITAM-mediated signalling in the cell, as the recruitment of Shb to the CD3zeta during TCR activation suggests (Karlsson et al., 1995; Welsh et al., 1998). In contrast, the presence of proline-rich motifs that bind the WW-domains of Nedd4/AIP4 HECT-domain E3 ubiquitin ligases in close proximity to an ITAM motif appears to be a unique feature of the gamma-herpesviral homologs of LMP2A.

The effect of AIP4 downregulation on Syk stability in LMP2A-expressing cells (Figure 5) suggests that additional mechanisms may influence the metabolism of Syk in the LMP2A complex. Previous work has shown that Cbl is recruited to the autophosphorylated Y323 of Syk (Lupher et al., 1998) and that Cbl is targeted by the Nedd4/AIP4 family of E3 ubiquitin ligases (Magnifico et al., 2003). The regulation of CblC by AIP4 is an important part of the mechanism for downregulation of the epidermal growth factor receptor signaling complex (Ettenberg et al., 2001; Courbard et al., 2002). The interactions of Cbl with Syk and AIP4 may explain why we detect Cbl in immunoprecipitates of LMP2A (data not shown). A hypothesis would be that the LMP2A complex coordinates the activities of AIP4 and Cbl in a similar fashion to regulate the stability of the activated Syk. If so, then downregulation of AIP4 might relieve the control of Cbl activity leading to enhanced degradation of Syk (Figure 7).

Figure 7.
Figure 7 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Model of LMP2A-mediated regulation of the Syk tyrosine kinase. (a) The protein complex assembled by wt LMP2A is shown. Black arrows indicate ubiquitination pathways. Syk turnover is determined by the rate of Cbl binding to the autophosphorylated tyrosine 323 (pY323) of Syk (Lupher et al., 1998). Syk binding to LMP2A ITAMs is balanced by Shb occupancy of the ITAM SH2-binding motifs, whereas Cbl activity towards Syk (arrow 1; Rao et al., 2001) is checked by AIP4 (arrow 2; Courbard et al., 2002; Magnifico et al., 2003). AIP4 ubiquitinates Syk (arrow 3; Ikeda et al., 2000; Winberg et al., 2000) and LMP2A (arrow 4; Ikeda et al., 2002). (b) Changes induced by shRNA mediated downregulation of Shb are shown. In the absence of Shb, Syk occupancy of the ITAMs is unrestricted. The rate of Syk autophosphorylation increases, which facilitates its ubiquitination by Cbl and accelerates Syk turnover. (c) AIP4 downregulation leads to stabilization of its targets, LMP2A and Cbl, both of which contribute to increased Syk turnover.

Full figure and legend (52K)

Our demonstration of Shb recruitment to LMP2A provides evidence for an additional level of regulation of LMP2A constitutive signals. By competing for ITAM SH2-binding motifs in LMP2A, Shb might regulate Syk access to LMP2A and thus also its rate of activation, autophosphorylation and downregulation by Cbl and AIP4. Binding of Shb to LMP2A might also provide additional docking sites for functional protein interactions, including PI3K.


Materials and methods

Cell lines, expression constructs and DNA transfections

Ramos-NUT is an EBV-negative germinal center B cell line. CBMI-RAL-STO is a lymphoblastoid cell line (LCL) established from cord blood with the Rael strain of EBV and Jac-1 and BK are LCL from peripheral blood of healthy individuals after infection with the B-95-8 strain of EBV. The following HEK 293 (Graham et al., 1977; Shaw et al., 2002) cell lines were used: 293T, which expresses the SV40 T-antigen; 293P, which is a pLXPOP vector control (Winberg et al., 2000); C4 and IC2 express the LMP2A gene from the B95-8 strain of EBV (Laux et al., 1988), cloned in the pLXPOP vector; LMP2A2-31 stably expresses LMP2A from the pLNPOX vector, derived from the pLNCX vector (a gift of AD Miller, Genbank M28247) by insertion of the poliovirus 5'-UTR ribosome entry site (PO) in place of the CMV intermediate early promoter.

The 3Tm is a HEK293 clone stably expressing a chimerical transmembrane protein containing the first 175 amino acids (aa) of LMP2A fused to the extracellular part of the lymphocyte marker CD38 (Zubiaur et al., 1997). Full-length wild type (wt) and 4 times Flag-tagged wt LMP2A (Winberg et al., 2000; Ingham et al., 2005) were based on the pSP64-23TP plasmid, a kind gift from Dr P Farrell (Laux et al., 1988). Vectors LMP2A point mutations Y64F, Y74F, Y64/74F and Y74/85F were constructed in the pLXPOP vector using overlap recombination PCR. Full-length retroviral constructs with the wild-type (wt) LMP2A and one point mutation, Y112F, was constructed similarly by subcloning singly Flag-tagged wt and mutant LMP2A into the p3XFLAG-CMV-7.1 vector (Sigma-Aldrich, St Louis, MO, USA), thus obtaining constructs with a fourfold tandemly repeated Flag-epitope (4 times Flag, 4F) at the N-terminus. All these retroviral constructs were stably expressed in HEK293 cells. A plasmid containing the full-length wt Shb for expression in eukaryotic cells as well as bacterially expressed GST fusion protein containing the SH2 (aa 488–597) or PTB (aa 840–1740) domains of Shb were generous gifts from Michael Welsh (Welsh et al., 1994, 1998).

For protein expression in HEK 293 cells, 5 times 106–1 times 107 cells in 85 mm plates were transfected with 5–10 mug plasmid DNA by a modified PEI technique (Abdallah et al., 1996) using 25 kDa polyethyleneimine (PEI, Aldrich cat. 40,872-7, Sigma Aldrich Chemie GmbH, Steinheim, Germany). Transfected cells were lysed at 36 h post-transfection in 1% NP40 lysis buffer as described (Winberg et al., 2000).

GST fusion protein binding assay

All GST fusion proteins were produced in Escherichia coli BL-21 cells (CGSC, New Haven, CO, USA), purified and used as described (Winberg et al., 2000; Matskova et al., 2001). Specific proteins were detected on the membranes using standard procedures for immunoblotting (Harlowe and Lane, 1988). After washing, filters were treated with a luminol reagent (ECL, Amersham Pharmacia Biotech) and images were captured at 2 times 2 binning in a Peltier-cooled Fuji Luminiscent Image Analyzer LAS-1000-CH in a Intelligent Dark Box II Imaging System (Fuji Photo Film Co., Ltd, Japan) and densitometry was performed directly on the 8 bit digital images using the Image J program (NIH, Bethesda, MD, USA).


The rat anti-LMP2A monoclonal antibodies (mAbs) 8C3, 14B7 and 4E11 (Fruehling et al., 1996) were purchased from Ascenion GmbH Munich, Germany. The anti-rat CD2 hybridoma OX34 was purchased from American Type Culture Collection and was purified as described (Matskova et al., 2001). The purified agonistic anti-human CD38 mAb (hybridoma IB4) was a generous gift from Fabio Malavasi (Silvennoinen et al., 1996; Zubiaur et al., 1997). Rabbit serum against the Shb PTB domain was generously provided by Michael Welsh. Preimmune and hyperimmune rabbit serum against the WW domains of the human AIP4 protein was previously described (Winberg et al., 2000). Goat polyclonal antibody against actin (I-19, sc1616), gamma-tubulin (D-10, sc-17788), mouse mAb against phosphotyrosine, directly conjugated to horseradish peroxidase, clone PY99 (sc-7020), Akt mAb (B-1, sc-5298) and rabbit polyclonal antibody against Akt phosphoserine 473 (sc-7985-R) were all from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Rabbit mAb against Akt phosphoserine 473 (clone SK703) was from Upstate (Charlottesville, VA, USA). Affinity-purified anti-FLAG M2 mAb (F1804) was from Sigma (St Louis, MO, USA). Goat anti-mouse IgG (H+L), AffiniPure F(ab)'2 fragment was purchased from Jackson Immunoresearch Laboratories, Inc., West Grove, PA, USA. For use in immunoprecipitation, the LMP2A and CD2 Mabs were covalently coupled to CNBr-activated Sepharose CL4-B (Amersham Pharmacia Biotech, Uppsala, Sweden) as described (Coligan et al., 2001).

Immunochemical assays

Immunoprecipitations and immunoblotting experiments were performed using standard techniques (Harlowe and Lane, 1988). Where appropriate, preclearing was performed with preimmune serum (AIP4, Shb). Immunoprecipitations of LMP2A and CD2 were carried out using immobilized primary antibodies. Antigens captured in solution were immobilized on Protein A or Protein G-Sepharose CL-4B (Amersham Pharmacia Biotech). Stripping of filters was carried out in 6 M guanidinium-HCl, 62.5 mM Tris-HCl, pH 6.8, 100 mM beta-mercaptoethanol for 30 min at 37°C, followed by several washes in PBS.

Antibody-mediated crosslinking of exodomains in transmembrane fusion proteins

Cells grown in 85 mm dishes were serum starved in Iscove's modified Dulbecco's medium (IMDM), 0.5% FBS for 4 h. Following one wash with ice-cold PBS, 5 mug of the IB4 anti-CD38 mAb in 2 ml of IMDM, 0.5% FBS was adsorbed to the cells for 10 min on ice and excess primary antibody was removed by washing with PBS. Crosslinking of the primary antibody was performed by addition of 20 mug of goat anti-mouse IgG (H+L) F(ab)'2 fragment (AffiniPure, Jackson Immunoresearch Laboratories Inc.), for 10 min (Silvennoinen et al., 1996; Zubiaur et al., 1997). The cells were then incubated at 37°C for the specified times. Unstimulated control cells and cells stimulated with F(ab)'2-fragment alone were incubated for 30 min at 37°C. After the indicated time (0, 1, 5, 10 or 30 min), cells were washed once with ice-cold PBS and lysed in 1% NP-40 lysis buffer (Winberg et al., 2000). As a control, crosslinking with the mAb against the rat CD2 exodomain was performed using the same protocol.

shRNA interference with Shb and LMP2A expression

To reduce LMP2A, Shb, Syk and AIP4 protein levels, we used a human H1 RNA pol III promoter-based shRNA vector, pBINNS2 (Kunath et al., 2003), to generate siRNA that target the EBV LMP2A mRNA, human Shb, Syk or AIP4 mRNAs in mammalian cells. A control shRNA vector targeting the EGFP protein was also made. Preparation of LMP2A, Shb, Syk, AIP4 and EGFP retroviral-shRNA vectors was as follows.

For the silencing of LMP2A (Genbank: M87779) and Shb (Genbank: NM003028), the sense- and antisense-strand oligonucleotides 5'-tcgagGGACTACAAGGCATTTACGTTCTcaagagaAGAAC

The sense- and antisense-strand oligonucleotides for Syk shRNA were 5'-tcgagAATCAAGTTTGACCAGTGCAGTTcaagagaAACTG
tcttgAACTGCACTGGTCAAACTTGATTc-3', respectively, and the corresponding oligonucleotides for AIP4 shRNA were 5'-tcgagAACCAGTTGGACTCAAGGATTTAcaagagaTAAAT

The control shRNA vector directed towards EGPF was prepared using the oligonucleotides, 5'-tcgagAACCACTACCTGAGCACCCAGTTcaagagaAACTG

The vectors were prepared by ligating the annealed oligonucleotides into the XhoI and BglII of pBINNS2. The pBINNS2 vector is derived from incorporation by PCR of the human H1 RNA pol III promoter (Kunath et al., 2003) into the EcoRI and XhoI site of a self-inactivating murine stem cell virus (pMSCVpuro) plasmid modified through deletion of the 3' LTR. The authenticity of the resulting vectors was confirmed by DNA sequencing. Stably shRNA expressing cell lines were prepared by retroviral infection of target HEK293 or HEK293 LMP2A2-31 (G418 resistant) cells with virus-containing supernatant from amphotropic PA317 cells transfected 36–40 h previously with retroviral vector DNA using Lipofectin (Invitrogen, Paisley, UK) or PEI-mediated transfection (see above). After 4 days of selection in 1.25 mug/ml of puromycin (P8833, Sigma-Aldrich Chemie Gmbh, Germany) resistant cell clones were grown up and tested for downregulation of the corresponding protein. About 12 clones of each construct were tested by immunoblotting for downregulation of the targeted gene. For the shEGFP clones, individual cell clones were transfected with EGFP-N1 plasmid and the ratio of EGFP-positive to negative cells was scored by fluorescence microscopy after 18 h after transfection. Clones with at least a 90% reduction of EGFP positive cells were used as shRNA vector controls.



  1. Abdallah B, Hassan A, Benoist C, Goula D, Behr JP, Demeneix BA. (1996). A powerful nonviral vector for in vivo gene transfer into the adult mammalian brain: polyethylenimine. Hum Gene Ther 7: 1947–1954. | PubMed | ISI | ChemPort |
  2. Beaufils P, Choquet D, Mamoun RZ, Malissen B. (1993). The (YXXL/I)2 signalling motif found in the cytoplasmic segments of the bovine leukaemia virus envelope protein and Epstein–Barr virus latent membrane protein 2A can elicit early and late lymphocyte activation events. EMBO J 12: 5105–5112. | PubMed | ISI | ChemPort |
  3. Caldwell RG, Wilson JB, Anderson SJ, Longnecker R. (1998). Epstein–Barr virus LMP2A drives B cell development and survival in the absence of normal B cell receptor signals. Immunity 9: 405–411. | Article | PubMed | ISI | ChemPort |
  4. Casola S, Otipoby KL, Alimzhanov M, Humme S, Uyttersprot N, Kutok JL et al. (2004). B cell receptor signal strength determines B cell fate. Nat Immunol 5: 317–327. | Article | PubMed | ISI | ChemPort |
  5. Coligan JE, Dunn BM, Ploegh HL, Speicher DW, Wingfield PT. (2001). Current Protocols in Protein Science. John Wiley & Sons, Inc.: New York.
  6. Coopman PJ, Do MT, Barth M, Bowden ET, Hayes AJ, Basyuk E et al. (2000). The Syk tyrosine kinase suppresses malignant growth of human breast cancer cells. Nature 406: 742–747. | Article | PubMed | ISI | ChemPort |
  7. Courbard JR, Fiore F, Adelaide J, Borg JP, Birnbaum D, Ollendorff V. (2002). Interaction between two ubiquitin-protein isopeptide ligases of different classes, CBLC and AIP4/ITCH. J Biol Chem 277: 45267–45275. | Article | PubMed | ISI | ChemPort |
  8. Dianzani U, Funaro A, DiFranco D, Garbarino G, Bragardo M, Redoglia V et al. (1994). Interaction between endothelium and CD4+CD45RA+ lymphocytes. Role of the human CD38 molecule. J Immunol 153: 952–959. | PubMed | ISI | ChemPort |
  9. Engels N, Merchant M, Pappu R, Chan AC, Longnecker R, Wienands J. (2001). Epstein–Barr virus latent membrane protein 2A (LMP2A) employs the SLP-65 signaling module. J Exp Med 194: 255–264. | Article | PubMed | ISI | ChemPort |
  10. Ettenberg SA, Magnifico A, Cuello M, Nau MM, Rubinstein YR, Yarden Y et al. (2001). Cbl-b-dependent coordinated degradation of the epidermal growth factor receptor signaling complex. J Biol Chem 276: 27677–27684. | Article | PubMed | ISI | ChemPort |
  11. Fodor S, Jakus Z, Mócsai A. (2006). ITAM-based signaling beyond the adaptive immune response. Immunology Lett 104: 29–37. | Article | ISI | ChemPort |
  12. Franke TF, Yang SI, Chan TO, Datta K, Kazlauskas A, Morrison DK et al. (1995). The protein kinase encoded by the Akt proto-oncogene is a target of the PDGF-activated phosphatidylinositol 3-kinase. Cell 81: 727–736. | Article | PubMed | ISI | ChemPort |
  13. Fruehling S, Lee SK, Herrold R, Frech B, Laux G, Kremmer E et al. (1996). Identification of latent membrane protein 2A (LMP2A) domains essential for the dominant-negative effect on B-lymphocyte surface immunoglobulin signal transduction. J Virol 70: 6216–6226. | PubMed | ISI | ChemPort |
  14. Grande SM, Ross SR, Monroe JG. (2006). Viral immunoreceptor-associated tyrosine-based activation motifs: potential players in oncogenesis. Future Oncol 2: 301–310. | Article | PubMed | ChemPort |
  15. Graham FL, Smiley J, Russell WC, Nairn R. (1977). Characteristics of a human cell line transformed by DNA from human adenovirus type 5. J Gen Virol 36: 59–74. | PubMed | ISI | ChemPort |
  16. Harlowe E, Lane D. (1988). Antibodies, A Laboratory Manual. Cold Spring Harbor Laboratory Press: Cold Spring Harbor, New York.
  17. Holmqvist K, Cross MJ, Rolny C, Hagerkvist R, Rahimi N, Matsumoto T et al. (2004). The adaptor protein Shb binds to tyrosine 1175 in vascular endothelial growth factor (VEGF) receptor-2 and regulates VEGF-dependent cellular migration. J Biol Chem 279: 22267–22275. | Article | PubMed | ISI | ChemPort |
  18. Ikeda M, Ikeda A, Longan LC, Longnecker R. (2000). The Epstein–Barr virus latent membrane protein 2A PY motif recruits WW domain-containing ubiquitin-protein ligases. Virology 268: 178–191. | Article | PubMed | ISI | ChemPort |
  19. Ikeda M, Ikeda A, Longnecker R. (2002). Lysine-independent ubiquitination of Epstein–Barr virus LMP2A. Virology 300: 153–159. | Article | PubMed | ISI | ChemPort |
  20. Ingham RJ, Raaijmakers J, Lim CS, Mbamalu G, Gish G, Chen F et al. (2005). The Epstein–Barr virus protein, latent membrane protein 2A, co-opts tyrosine kinases used by the T cell receptor. J Biol Chem 280: 34133–34142. | Article | PubMed | ISI | ChemPort |
  21. Karlsson T, Songyang Z, Landgren E, Lavergne C, Di Fiore PP, Anafi M et al. (1995). Molecular interactions of the Src homology 2 domain protein Shb with phosphotyrosine residues, tyrosine kinase receptors and Src homology 3 domain proteins. Oncogene 10: 1475–1483. | PubMed | ISI | ChemPort |
  22. Katz E, Lareef MH, Rassa JC, Grande SM, King LB, Russo J et al. (2005). MMTV Env encodes an ITAM responsible for transformation of mammary epithelial cells in three-dimensional culture. J Exp Med 201: 431–439. | Article | PubMed | ISI | ChemPort |
  23. Koretzky GA, Abtahian F, Silverman MA. (2006). SLP76 and SLP65: complex regulation of signalling in lymphocytes and beyond. Nat Revs Immunol 6: 67–78. | Article | ChemPort |
  24. Kunath T, Gish G, Lickert H, Jones N, Pawson T, Rossant J. (2003). Transgenic RNA interference in ES cell-derived embryos recapitulates a genetic null phenotype. Nat Biotechnol 21: 559–561. | Article | PubMed | ISI | ChemPort |
  25. Laux G, Perricaudet M, Farrell PJ. (1988). A spliced Epstein–Barr virus gene expressed in immortalized lymphocytes is by circularization of the linear viral genome. EMBO J 7: 769–774. | PubMed | ISI | ChemPort |
  26. Lindholm CK, Henriksson ML, Hallberg B, Welsh M. (2002). Shb links SLP-76 and Vav with the CD3 complex in Jurkat T cells. Eur J Biochem 269: 3279–3288. | Article | PubMed | ISI | ChemPort |
  27. Longnecker R, Druker B, Roberts TM, Kieff E. (1991). An Epstein–Barr virus protein associated with cell growth transformation interacts with a tyrosine kinase. J Virol 65: 3681–3692. | PubMed | ISI | ChemPort |
  28. Lu J, Lin WH, Chen SY, Longnecker R, Tsai SC, Chen CL et al. (2006). Syk tyrosine kinase mediates Epstein–Barr virus latent membrane protein 2A-induced cell migration in epithelial cells. J Biol Chem 281: 8806–8814. | Article | PubMed | ISI | ChemPort |
  29. Lupher Jr ML, Rao N, Lill NL, Andoniou CE, Miyake S, Clark EA et al. (1998). Cbl-mediated negative regulation of the Syk tyrosine kinase. A critical role for Cbl phosphotyrosine-binding domain binding to Syk phosphotyrosine 323. J Biol Chem 273: 35273–35281. | Article | PubMed | ISI | ChemPort |
  30. Magnifico A, Ettenberg S, Yang C, Mariano J, Tiwari S, Fang S et al. (2003). WW domain HECT E3s target Cbl RING finger E3s for proteasomal degradation. J Biol Chem 278: 43169–43177. | Article | PubMed | ISI | ChemPort |
  31. Matskova L, Ernberg I, Pawson T, Winberg G. (2001). C-terminal domain of the Epstein–Barr virus LMP2A membrane protein contains a clustering signal. J Virol 75: 10941–10949. | Article | PubMed | ISI | ChemPort |
  32. Merchant M, Caldwell RG, Longnecker R. (2000). The LMP2A ITAM is essential for providing B cells with development and survival signals in vivo. J Virol 74: 9115–9124. | Article | PubMed | ISI | ChemPort |
  33. Morrison JA, Raab-Traub N. (2005). Roles of the ITAM and PY motifs of Epstein–Barr virus latent membrane protein 2A in the inhibition of epithelial cell differentiation and activation of beta-catenin signaling. J Virol 79: 2375–2382. | Article | PubMed | ISI | ChemPort |
  34. Ota S, Hazeki K, Rao N, Lupher Jr ML, Andoniou CE, Drukeri B et al. (2000). The RING finger domain of Cbl is essential for negative regulation of the Syk tyrosine kinase. J Biol Chem 275: 414–422. | Article | PubMed | ISI | ChemPort |
  35. Rao N, Ghosh AK, Ota S, Zhou P, Reddi AL, Hakezi K et al. (2001). The non-receptor tyrosine kinase Syk is a target of Cbl-mediated ubiquitylation upon B-cell receptor stimulation. EMBO J 20: 7085–7095. | Article | PubMed | ISI | ChemPort |
  36. Scholle F, Bendt KM, Raab-Traub N. (2000). Epstein Barr virus LMP2A transforms epithelial cells, inhibits cell differentiation, and activates Akt. J Virol 74: 10681–10689. | Article | PubMed | ISI | ChemPort |
  37. Scholle F, Longnecker R, Raab-Traub N. (2001). Analysis of the phosphorylation status of Epstein-Barr virus LMP2A in epithelial cells. Virology 291: 208–214. | Article | PubMed | ISI | ChemPort |
  38. Shaw G, Morse S, Ararat M, Graham FL. (2002). Preferential transformation of human neuronal cells by human adenoviruses and the origin of HEK 293 cells. FASEB J 16: 869–871. | PubMed | ISI | ChemPort |
  39. Silvennoinen O, Nishigaki H, Kitanaka A, Kumagai M, Ito C, Malavasi F et al. (1996). CD38 signal transduction in human B cell precursors. Rapid induction of tyrosine phosphorylation, activation of syk tyrosine kinase, and phosphorylation of phospholipase C-gamma and phosphatidylinositol 3-kinase. J Immunol 156: 100–107. | PubMed | ISI | ChemPort |
  40. Swart R, Fruehling S, Longnecker R. (1998). Tyrosines 60, 64 and 101 of Epstein–Barr virus LMP2A are not essential for blocking B cell signal transduction. Virology 263: 485–495. | Article | ISI | ChemPort |
  41. Swart R, Ruf IK, Sample J, Longnecker R. (2000). Latent membrane protein 2A-mediated effects on the phosphatidylinositol 3-Kinase/Akt pathway. J Virol 74: 10838–10845. | Article | PubMed | ISI | ChemPort |
  42. Tomlinson CC, Damania B. (2004). The K1 protein of Kaposi's sarcoma-associated herpesvirus activates the Akt signaling pathway. J Virol 78: 1918–1927. | Article | PubMed | ISI | ChemPort |
  43. Welsh M, Mares J, Karlsson T, Lavergne C, Bréant B, Claesson-Welsh L. (1994). Shb is a ubiquitously expressed Src homology 2 protein. Oncogene 9: 19–27. | PubMed | ISI | ChemPort |
  44. Welsh M, Songyang Z, Frantz JD, Trub T, Reedquist KA, Karlsson T et al. (1998). Stimulation through the T cell receptor leads to interactions between SHB and several signaling proteins. Oncogene 16: 891–901. | Article | PubMed | ISI | ChemPort |
  45. Welsh N, Makeeva N, Welsh M. (2002). Overexpression of the Shb SH2 domain-protein in insulin-producing cells leads to altered signaling through the IRS-1 and IRS-2 proteins. Mol Med 8: 695–704. | PubMed | ISI | ChemPort |
  46. Winberg G, Matskova L, Chen F, Plant P, Rotin D, Gish G et al. (2000). Latent membrane protein 2A of Epstein–Barr virus binds WW domain E3 protein-ubiquitin ligases that ubiquitinate B cell tyrosine kinases. Mol Cell Biol 20: 8526–8535. | Article | PubMed | ISI | ChemPort |
  47. Zubiaur M, Izquierdo M, Terhorst C, Malavasi F, Sancho J. (1997). CD38 ligation results in activation of the Raf-1/mitogen-activated protein kinase and the CD3-zeta/zeta-associated protein-70 signaling pathways in Jurkat T lymphocytes. J Immunol 159: 193–205. | PubMed | ISI | ChemPort |


We thank Michael Welsh for generous gifts of reagents and for stimulating discussions. Fabio Malavasi kindly contributed the agonistic anti CD38 antibody IB4. We thank W Rod Hardy for providing the pMSCVpuro vector. This work was supported by grants from The Swedish Research Council, Grant K2002-16X-14227-01A and The Swedish Children's Cancer Foundation, PROJ01/18, to GW and LM, and from the Canadian Institutes for Health Research (CIHR) to TP. RJI is a Fellow of the Leukemia & Lymphoma Society. CKL is supported in part by a fellowship from the Wenner-Gren Foundation.