Regulation des Zelloberflächenmoleküls CD83 durch das Epstein-Barr Virus und Analyse seiner Funktion

EBV is a γ-herpes virus which is able to infect human resting
B-cells and to transform them into permanently growing
lymphoblastoid cell lines (LCLs). EBNA2 (Epstein-Barr virus nuclear
antigen 2) is one of the first viral proteins expressed after in
vitro infection and interacts with different cellular proteins like
RBP-Jκ and PU.1. The EBNA2 protein acts as a transcriptional
activator of the viral Latent Membrane Proteins 1 and 2 (LMP1 and
LMP2) and the viral nuclear genes EBNA1, EBNA3A, -3B, -3C, EBNA-LP.
Additionally EBNA2 is also able to transactivate cellular genes
like CD21, CD23 or c-myc. To study the different EBNA2 target genes
and the function of EBNA2 a LCL was established (ER/EB2-5 cells,
Kempkes et al., 1995) harboring an estrogen-inducible EBNA2. In the
presence of estrogen the ER/EBNA2 fusion protein (estrogen receptor
binding domain) is located in the nucleus were EBNA2 can
transactivate its target genes, whereas in the absence of estrogen
the ER/EBNA2 fusion protein is kept in the cytoplasm and therefore
inactive. The cells proliferate in the presence of estrogen and
they arrest in the absence resulting in a phenotype similar to
resting B-lymphocytes. By using the ER/EB2-5 cell line I could
clearly show that the cell surface molecule CD83, belonging to the
immunoglobuline superfamily (Zhou et al., 1992), is upregulated
after the activation of EBNA2. By using a derivative ER/EB2-5 cell
line that constitutively expressed LMP1 I could show that CD83 is
still expressed even in the absence of functional EBNA2 suggesting
that LMP1, the viral target gene of EBNA2, is responsible for the
induction of CD83. Therefore I analysed the activation of the CD83
promoter by LMP1. LMP1 is a transmembrane protein with a short
intracellular N-terminus, 6 hydrophobic transmembrane domains and a
long intracellular C-terminus, containing C-terminal activator
regions CTAR1, 2 and 3. The different CTAR regions are responsible
for activating genes via NF-κB, ATF, AP1 and STAT signaling
pathways. For the activation of its target genes LMP1 uses the same
signaling molecules (TRAF, TRADD) as family members of the TNF-R
family (CD40, TNF-R1, TNF-R2). The CD83 promoter was activated by
LMP1 as shown by promoter luciferase reporter assays in 293-T
cells. The induction was not observed in the absence of a NF-κB
binding site in a CD83 promoter mutant. Furthermore LMP1 mutants
which are mutated in the binding regions for TRAF2 (CTAR1) or TRADD
(CTAR2) are not able to transactivate the CD83 promoter. By
co-transfection of LMP1 and dominant/negative IκB the CD83 promoter
could not be activated because of inactivation of NF-κB. These
experiments clearly demonstrate that the CD83 promoter is
transactivated by LMP1 via NF-κB. Additionally to the regulation of
CD83 I was also interested in the functional role of CD83. Until
now only little is known about the function of CD83. CD83 seems to
have a specific role in the decision to single positive CD4+
T-cells in the thymus (Fujimoto et al., 2002). I have tested a
possible co-stimulatory function of CD83 to CD4+ T-cells by
retroviral expression of CD83 in non-professional antigen
presenting cells (RCC). Indeed CD83 expression increased the CD4+
response in comparison to CD80 or GFP retroviral infected RCC
cells. In mixed lymphocyte reactions this co-stimulatory effect
could not be clearly demonstrated although a soluble CD83-Ig showed
a small inhibitory influence. The identification of a CD83 ligand
molecule could give new insights into the function of CD83.
Therefore a CD83-Ig fusion protein as well as a CD83-tetramer
construct were generated and used to screen for a potential ligand
of CD83. First results showed that the CD83-Ig fusion protein and
the CD83-tetramer construct bound to CD4+ and to CD8+ T-cells of
isolated PBMCs as well as to activated T-cells in a culture of
mixed T-cell populations.