To the Editor:
Pleckstrin is exclusively expressed in human hemopoietic cells and is induced during differentiation. Pleckstrin protein contains two copies of the prototypic pleckstrin homology (PH) domain and has been shown to be hyperphosphorylated and appears to have a role in signal transduction.1 Recently we reported that the coordinate expression of the Epstein-Barr virus (EBV) encoded EBNA3, 4 and 6 proteins (EBNA 3 family) lead to the upregulation of pleckstrin protein in the transfected Burkitt's lymphoma (BL) cell line dG75.2 The present study was undertaken to determine if pleckstrin was upregulated in EBV+ cells and to evaluate its phosphorylation status.
Isogenic cell pairs of BL cells or the corresponding lymphoblastoid cell line expressing either the EBNA1 protein alone (Mutu-I, BL29) or the full set of EBV latency proteins (Mutu-III, IARC167) were used. The cell phenotype and the expression of EBV genes were analyzed by fluorescence-activated cell sorter and immunoblot analysis. Equal numbers of exponentially growing cells were radiolabeled with 32P and the cell lysates subjected to immunoprecipitation using rabbit antipleckstrin serum1 followed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (Fig 1A). To determine the amount of total pleckstrin protein in the radiolabeled cells, aliquots of the cell lysates were analyzed in parallel by immunoblot using the rabbit antipleckstrin serum (Fig 1B). Analysis of the polyacrylamide gels by silverstaining and of the nitrocellulose filters by staining with Ponceaus (Sigma, Castle Hill, NSW, Australia) and with an anti-βII microglobulin antibody confirmed that each sample contained similar amounts of protein.
Pleckstrin protein was highly expressed in each of the cell lines expressing the EBV-latency antigens (Fig 1B, lanes 2 and 4). However, the amount of 32P associated with this induced pleckstrin was not markedly increased over the basal level of 32P-pleckstrin found in the cells expressing only the EBNA1 protein (Fig 1A, lanes 1 through 4). This is presented graphically in Fig 1C where it can be seen that the EBV-induced pleckstrin contained less radiolabeled phosphate per amount of protein compared to the pleckstrin present in the cells expressing only EBNA1. The EBV-induced hypophosphorylated pleckstrin protein could be due to either the induction of unphosphorylated pleckstrin (diluting the small pool of endogenous hyperphosphorylated pleckstrin) or the presence of a large pool of hypophosphorylated pleckstrin molecules. In contrast to these results, both the phosphorylation status and amount of pleckstrin expressed in the EE346-G7 cells (transfected with the EBNA 3 gene family) was higher than in either the control E-C3 (transfected with the expression vector only) or the parental dG75 cell line (Fig 1A and B, lanes 5 through 7). The ratio of radiolabeled pleckstrin to total pleckstrin protein in EE346-G7 cells was similar to the control transfectants E-C3 (Fig 1C) indicating that these EBNA gene products, while being able to upregulate pleckstrin, are unlikely to be involved in controlling the phosphorylation status of this protein. Therefore, it is possible that in the latently infected B cell other EBV-encoded gene products may, either directly or indirectly, interfere with pathways involved in the phosphorylation of pleckstrin.
Recent data support the idea that phosphorylation regulates the activity of pleckstrin. Abrams et al1 showed that phosphorylated pleckstrin could inhibit agonist-induced phosphoinositide hydrolysis and that the N-terminal PH domain of pleckstrin was crucial for this interaction. Moreover, a recent report provided evidence for a link between impaired pleckstrin phosphorylation in platelets and abnormal signal transduction.3 Given this data, it is possible that EBV-induced pleckstrin could interfere with cellular signaling pathways, involving phosphoinositide, unless the induced pleckstrin was kept in a hypophosphorylated form. Indeed, two putative G protein-coupled peptide receptors (which would mediate their activity via phosphoinositide) were found to be induced in BL cells after EBV infection.4 It is possible that EBV upregulates pleckstrin expression, not to interfere with G protein-coupled signaling, but to modulate other cellular signaling processes or cytoskeletal organization, which could be mediated by protein-protein interactions via the PH and DEP5 domains present in pleckstrin. Regulation of these cellular processes might be important in EBV-induced transformation or viral persistence in B lymphocytes.
The authors are indebted to A. Rickinson for the gift of the Mutu cell clones and thank K. Krauer, S.L. Silins, R. Khanna, and L. Morrison for helpful discussions and excellent technical assistance. N. Kienzle was a fellow of the German Infektionsforschung/AIDS-Stipendiumsprogramm, DKFZ, Heidelberg, Germany. Supported by Grants from the National Health and Medical Research Council and the Queensland Cancer Fund, Australia.