Identification of a spliced gene from Kaposi's sarcoma-associated herpesvirus encoding a protein with similarities to latent membrane proteins 1 and 2A of Epstein-Barr virus
M. Glenn et al., Identification of a spliced gene from Kaposi's sarcoma-associated herpesvirus encoding a protein with similarities to latent membrane proteins 1 and 2A of Epstein-Barr virus, J VIROLOGY, 73(8), 1999, pp. 6953-6963
Kaposi's sarcoma-associated herpesvirus (KSHV) or human herpesvirus 8 (HHV-
8) is a novel herpesvirus implicated as the causative agent of Kaposi's sar
coma (KS), primary effusion lymphoma, and some cases of multicentric Castle
man's disease. KSHV persists in the majority of KS spindle (endothelial tum
or) cells and lymphoid cells in a latent form, with only a limited set of v
iral genes expressed in a tissue-specific manner. Here, we report the ident
ification of a family of alternatively-spliced transcripts of approximately
7.5 kb expressed in latently infected body cavity-based lymphoma (BCBL) ce
ll lines which are predicted to encode membrane proteins with similarities
to the LMP2A and LMP1 proteins of Epstein-Barr virus. In two highly diverge
nt sequence variants of the right end of the KSHV genome, alternative splic
ing of eight exons located between KSHV ORF 75 and the terminal repeats yie
lds transcripts appropriate for proteins with up to 12 transmembrane domain
s, followed by a hydrophilic C-terminal, presumably cytoplasmic, domain. Th
is C-terminal domain contains several YxxI/L motifs reminiscent of LMP2A an
d a putative TRAF binding site as in LMP1. In latently (persistently) infec
ted BCBL cells the predominant transcript utilizes all eight exons, whereas
in phorbol-ester-induced cells, a shorter transcript, lacking exons 4 and
5, is also abundant. We also found evidence for an alternative use of exon
1. Transfection of an epitope-tagged cDNA construct containing all exons in
dicates that the encoded protein is localized on cell surface and intracell
ular membranes, and glutathione S-transferase pull-down experiments indicat
e that its cytoplasmic domain, like that of LMP1, interacts with TRAF1, -2,
and -3. Two of 20 KS patients had antibodies to the hydrophilic C-terminal
domain, suggesting that the protein is expressed in vivo.