| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Cancer Research Campaign, Institute for Cancer Studies, University of Birmingham, Edgbaston, Birmingham, United Kingdom;
Department of Medicine, University of Wales College of Medicine, Heath Park, Cardiff, United Kingdom; and
Department of Rheumatology, University of Birmingham, The Medical School, Birmingham, United Kingdom
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B, up-regulation of cell surface markers
ICAM-1/CD54 and Fas/CD95, and growth inhibition. Aggregation was seen
at lower doses than those required to elicit the other effects. By
contrast, low responder (LR) lines showed no detectable response to
CD40 mAbs, while their responses to sCD40L were limited to activation
of NF-B and up-regulation of CD95 only. However, in transfection
experiments, LMP1 uniformly induced the full spectrum of phenotypic
effects in both HR and LR lines. We conclude that some BL cell lines
show a highly restricted response to CD40 ligation but remain fully
susceptible to LMP1. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Multiple effects of CD40 engagement have been described in in vitro systems after ligation with either specific mAb or recombinant CD40L in the soluble or the membrane-bound form. In resting human B cells, CD40 ligation can act to augment the proliferation and Ig secretion induced by anti-Ig 3, 11, 12 . On its own, CD40 stimulation of resting B cells induces the formation of homotypic aggregates and up-regulation of several cell surface molecules such as CD23, CD54, and CD95 13, 14, 15 . Addition of IL-4 to such cultures leads to rapid B cell proliferation and a pattern of Ig secretion that is skewed toward IgE; other cytokines can cause preferential skewing toward other Ig isotypes in this system 3, 16, 17, 18 . Parallel studies with cycling GC B cells isolated from tonsillar preparations are complicated by the tendency of these cells to undergo rapid spontaneous apoptosis in vitro. However, it is clear that CD40 ligation and, to a lesser extent, surface Ig cross-linking can provide a short term survival signal to cultured GC cells 3, 19 . In combination with specific ILs, such treatments can either promote GC cell transition to a memory B cell phenotype or prolong cell proliferation 12, 20 .
In addition to work involving freshly isolated B cells, a number of studies have reproduced several aspects of these responses to CD40 ligation using established human B cell lines, often of Burkitt’s lymphoma (BL) origin. Thus, enhanced cell aggregation, up-regulation of specific cell surface markers such as CD23 and CD54, and activation of the HLA class I Ag processing pathway were observed in a number of different lines 21, 22, 23, 24 . Furthermore, where cells could be induced into apoptosis by serum deprivation or by exposure to agents such as TGF-ß or anti-IgM, then CD40 ligation protected against cell death 21, 25, 26 . As in the GC cell system, survival was associated with the up-regulation of anti-apoptotic proteins such as A20 and members of the Bcl-2 family 21, 27, 28 . In addition, several (but not all) of the studies with cell lines also reported growth inhibitory effects of CD40 ligation, a finding that was not anticipated from the work with freshly isolated B cells 29, 30, 31 .
Our interest in these phenomena stem from the parallels that exist
between CD40-mediated effects and those induced in B cells by the
EBV-encoded latent membrane protein 1 (LMP1). Thus, vectored expression
of LMP1 in BL cell lines has been shown to lead to cell aggregation,
up-regulation of several cell-surface markers including CD54,
activation of HLA class I processing, and protection from apoptosis
through up-regulation of A20 and/or Bcl-2 32, 33, 34, 35, 36 . Furthermore, a
membrane-proximal domain within the cytoplasmic C-terminus of LMP1
(CTAR1) can interact with a set of TRAF proteins that are also known to
associate with CD40. One early effect of TRAF aggregation in both
systems is activation of the transcription factor NF-B 4, 5, 37, 38 . Finally, in assays with stably transfected clones, LMP1 has been
shown to bring about a slowing of BL cell growth; this cytostatic
effect occurs without any induction of cell death 39 .
Given the extensive studies of LMP1-induced effects in the BL cell system and the availability of a wide panel of well-characterized lines of this type, we elected to screen these lines for their responses to CD40 ligation. Ligation was mediated either through mAb or through purified trimeric CD40L. The work reveals a clear division of BL lines into high responders (HR) showing the full range of responses and low responders (LR) showing only a limited range; in contrast, all lines were fully responsive to vectored expression of LMP1.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The EBV-negative BL lines, Ramos, Louckes, DG75, BL2, BL40,
BL41, and L3055, and the EBV-positive BL lines, Chep, Mutu-I, Eli, Wan,
Rael, and Akata, have been described in previous studies 23, 35, 37, 40 . Sal is a recently established, EBV-positive BL line (A.B.R.,
unpublished observation). All lines had cytogenetic markers
characteristic of BL and had retained a tumor biopsy-like phenotype on
serial passage. Some assays used lymphoblastoid cell lines (LCLs),
generated by EBV transformation of normal human B cells in vitro, as
LMP1-positive controls. Cells were maintained in RPMI 1640 medium
supplemented with 2 mM glutamine, gentamicin, 1 mM sodium-pyruvate,
-thioglycerol, and 10% (v/v) FCS. The LMP1-transfected clones of
Ramos and Akata were established using a tetracycline-regulated vector
system and were maintained in culture medium containing 1.5 mg/ml G418,
0.5 mg/ml hygromycin B, and 1 µg/ml tetracycline. Tetracycline
withdrawal induced LMP1 expression as previously described 39 .
Control and LMP1-transfected clones of Sal were established using the
pSG5-Neo and pSG5-Neo/LMP1 vectors, respectively (see below), and
transfectants were selected in culture medium containing 2 mg/ml G418.
Abs and reagents
The murine IgG1 mAbs used as surrogate ligands in this work
included two CD40-specific reagents, G28.5 41 and mAb89 42 , and two
control reagents, the 3 integrin-specific mAb 1A3 (ICRF
Biotherapeutics and Hybridoma Development Unit, South Mimms, U.K.) and
the CD44-specific mAb BU52 (provided by D. Hardie, Department of
Immunology, University of Birmingham, Birmingham, U.K.). All were used
as DE-52-purified Abs from hybridoma supernatant, and protein
concentrations were determined using the Bio-Rad protein assay
(Hercules, CA). An HPLC-purified preparation of trimeric human soluble
CD40L (sCD40L) was provided by Immunex (Seattle, WA). Bound Ab levels
were determined using a phycoerythrin (PE)-labeled goat anti-mouse
IgG (Caltag, Burlingame, CA).
CD40 expression was monitored using the G28.5 mAb followed by PE-labeled goat anti-mouse IgG (Caltag). Cell surface changes on cultured cells were detected using CD54- and CD23-specific IgG mAbs that were FITC labeled (Serotec, Oxford, U.K.) or unlabeled IgM mAbs specific for CD95 (CH11, Upstate Biotechnology, Lake Placid, NY) and CD54 (BU72, Department of Immunology, University of Birmingham), followed by PE-labeled goat anti-mouse IgM (Caltag). Rat CD2 expression of transfected cells was detected using an FITC-labeled rat-CD2-specific mouse mAb, OX34. LMP1 expression was monitored using the CS1–4 IgG mAbs 35 followed by Texas Red-labeled goat anti-mouse IgG (Southern Biotechnology Associates, Birmingham, AL).
Assays of CD40-mediated effects
Cells were exposed to known concentrations of CD40-specific or
control reagents. Cell aggregation was assessed visually by microscopic
observation over a 3-day period. Cell surface changes were monitored
after 48 h by immunofluorescent staining as described above.
NF-B activation was assessed by bandshift assays using 10 µg of
nuclear extract prepared after 3 h. For this, a probe derived from
the HIV long terminal repeat was 32P-labeled and
used essentially as previously described 43 . To assess cell
proliferation, microplate wells were seeded at 3 x
105 cells/well to which CD40-specific or control reagents
were added at known concentrations. Triplicate cultures of each type
were harvested on days 1, 2, and 3, and the viable cell number from a
known volume was determined by flow cytometric analysis using a
dedicated XL program (Coulter, Hialeah, CA) following addition of
propidium iodide to identify dead cells. Evidence of cell death was
also obtained by staining Permeafix-treated (Dako, Carpenteria, CA)
cells with 7-amino-actinomycin D (7-AAD) and determining the proportion
of the subdiploid cells by flow cytometric analysis of the resulting
cell cycle profile. In some experiments, cells were transiently
transfected (as described below) with a CD40 expression vector,
pcDNA3-CD40 44 , and levels of CD40, CD54, and CD95 expression on
transfected cells were monitored by double staining for these
individual markers and rat-CD2 using flow cytometry.
Assays of LMP1-induced effects
In transient transfection assays cells were transfected at
2 x 107/ml in 0.5 ml of culture medium by
electroporation using a Bio-Rad apparatus at 960 µF and a
predetermined optimal voltage (270–310 V depending on the cell line).
The electroporation cuvette contained 2 µg of pSG5-LMP1 or pSG5
control DNA 37 , 2 µg of pSG5-rat-CD2 45 , and 2 µg DNA of the
NF-B reporter construct 3Enh.B.Con A-luc; the latter construct
contains three tandem NF-B binding sites from the Ig promoter
linked to a minimal conalbumin promoter and to the luciferase gene
46 . Cells were harvested after 48 h, and transfection efficiency
was monitored by immunofluorescent staining for rat-CD2. Cell surface
changes induced by LMP1 were monitored by double staining with CD54-
and CD95-specific IgM mAbs, respectively, followed by anti-mouse
IgM-PE, then analyzing the rat-CD2-positive population for expression
of the second marker. NF-B activation was determined using a
standard luciferase assay, and results were normalized for transfection
efficiency. Levels of LMP1 expression were determined by double
staining of fixed cell smears with LMP1-specific IgG mAbs CS1–4 plus
CD54-specific IgM mAb BU72 followed by Texas Red-labeled anti-IgG
plus FITC-labeled anti-IgM, then staining was quantitated using a
Bio-Rad MR500–600 confocal microscope.
Stable LMP1 transfectants were established using either the tetracycline-regulated vector system as previously described 39 or a G418-selectable pSG5 derivative. For this, the G418 expression cassette was removed from pSV2neo as a BamHI fragment, blunt ended, and cloned into the blunt-ended XbaI site of pSG5 to yield pSG5-Neo. For the construction of pSG5-Neo-LMP1, the LMP1 gene was removed from pSG5-LMP1 37 as an EcoRI fragment and cloned into the EcoRI site of pSG5-Neo. Cell proliferation of LMP1-positive and -negative clones was assessed as described above. Levels of LMP1 expression were monitored by Western blotting using the CS1–4 mAbs 35 .
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The initial sets of experiments used a panel of 14 BL cell lines (seven EBV positive and seven EBV negative), selected for their retention of a biopsy-like phenotype in vitro, i.e., growth as single cells and generally low to undetectable expression of surface markers such as CD23, CD54, and CD95. Note that in EBV-positive BL lines of this kind, the virus exhibits a form of latency like that of the original tumor where LMP1 is not expressed 40 . Each of the lines was first assessed for cell aggregation following CD40 ligation over a period of 3 days. For this, we used three CD40-specific reagents, namely two murine IgG1 mAbs (G28.5 and mAb 89) and an HPLC-purified preparation of soluble trimeric CD40 ligand (sCD40L). Each agonist was tested over a 100-fold concentration range (0.1–10 µg/ml), and all three gave an essentially similar pattern of results.
The BL lines clearly fell into one of two categories based on their
aggregation response. Typical results are illustrated in Fig. 1
from cultures exposed to mAb G28.5.
Several lines, here represented by Mutu-I, showed aggregation across
the full range of concentrations of CD40 mAbs and CD40L tested; the
aggregation occurred within 8 h and was maintained over the 3-day
assay period. Other lines, here represented by Rael, showed no
detectable aggregation at any dose tested. As a control, IgG1 mAbs
directed against other cell surface markers (for example, mAb 1A3
against 3 integrin; Fig. 1) were used, and these did not
induce cell aggregation in any of the cell lines tested at any
concentration. This control excludes the possibility of Fc-Fc receptor
interactions as mediators of the aggregation induced by CD40-specific
mAbs.
|
summarizes the aggregation
response of all 14 BL lines to all three CD40-specific stimuli and
highlights the consistency of the response patterns observed. Note that
differences in response were not related to the EBV status of the
lines. Furthermore, Table I also records the level of CD40 cell surface
expression on each line, as determined by indirect immunofluorescent
staining and expressed as the mean fluorescence intensity (MFI) ± SD
of three independent assays. Clearly, all lines are CD40 positive, and
levels of surface expression are not related to CD40 responder status.
Based on these initial experiments, we grouped the BL lines into CD40
high responder (HR; aggregating) and CD40 low responder (LR;
nonaggregating) categories. This nomenclature is used throughout the
rest of the manuscript.
|
We next examined the effects of CD40 ligation on cell
proliferation. Cells were seeded at 3 x 105/ml and
exposed to a 0.1–10 µg/ml concentration range of sCD40L, mAb G28.5,
and an isotype-matched control mAb 1A3. Proliferation was assessed by
counting viable cell numbers over the next 3 days. As illustrated in
Fig. 2 for three representative HR lines
(Akata, Mutu-I, and Ramos), 10 µg/ml concentrations of sCD40L or mAb
G28.5 caused a substantial inhibition of cell growth compared with that
seen in untreated cultures or cultures containing 10 µg/ml control
mAb 1A3. In contrast, none of the LR lines showed any detectable effect
on cell growth by CD40 ligation; this is illustrated in Fig. 2 for the
representative LR lines, L3055, Rael, and Sal. Note that in the above
growth-inhibited HR cell cultures, the cells remained viable. No
evidence of apoptotic cell death could be observed either
microscopically or by flow cytometric analysis of the DNA content
quantified by 7-amino-actinomycin D staining (data not shown).
|
using mAb G28.5. In every case the inhibitory effect was already
maximal at a mAb concentration of 1 µg/ml, whereas at 0.1 µg/ml
there was either no effect on cell growth (Akata and Ramos), or
suboptimal inhibition occurred (Mutu-I). In the same assays sCD40L gave
a similar pattern of dose dependence, as did the other CD40-specific
mAb 89 (data not shown). These experiments therefore showed that the
growth inhibitory effect of CD40 ligation was restricted to HR lines,
i.e., those lines showing an aggregation response. Interestingly,
induction of growth inhibition required a higher concentration of the
CD40 stimulus than was required to induce aggregation.
|
In parallel experiments all HR and LR lines were examined for two
reported downstream effects of CD40 signaling in B cells, namely
increases in cell surface expression of CD54 (ICAM-1) and CD95 (Fas).
Where responses were observed, pilot studies using different
concentrations of mAb G28.5 or sCD40L again showed that doses of 1 and
10 µg/ml were indistinguishable in their effects. Furthermore, the
lower dose of 0.1 µg/ml produced either no detectable change or a
suboptimal response. Note that isotype-matched control mAbs never
produced any detectable changes in CD54 and CD95 expression compared
with untreated cultures. The illustrative data in Fig. 4 were obtained in experiments using 1
µg/ml doses of mAbs or sCD40L.
|
A shows the MFI ± SD from three independent
experiments in which cell surface levels of CD54 were determined after
48 h in the presence of the CD40-specific stimuli or of control
mAb 1A3. With the single exception of Akata, a line that has previously
been reported to be deficient in CD54 up-regulation in a variety of
circumstances 40 , all HR lines showed a clear CD40-induced response
to both mAb G28.5 and sCD40L. The level of CD54 induction was 2- to
3-fold in lines with appreciable baseline expression (e.g., Ramos and
Chep) and up to 10-fold in lines with very low baseline levels (e.g.,
Mutu-I). In contrast, the four LR lines illustrated showed no
significant response to either mAb G28.5 or sCD40L regardless of their
baseline levels of CD54 expression; the fifth LR line, BL40, showed no
response to mAb G28.5, but there was a weak (<2-fold) increase in CD54
staining in response to sCD40L (data not shown).
The corresponding findings for CD95 are shown in Fig. 4B.
Here all HR lines including Akata show clear up-regulation of CD95 from
low baseline levels in response to either mAb G28.5 or sCD40L. In the
same experiments, LR lines did not show any detectable change in their
low baseline surface expression of CD95 in response to mAb G28.5.
However, with the exception of Rael, all LR lines did show a small but
significant increase (Student’s t test,
p < 0.05) in CD95 expression following exposure to
sCD40L.
CD40 ligation-induced activation of NF-B
We next examined one of the more proximal reported effects of CD40
ligation, namely activation of NF-B. For this purpose, we measured
NF-B activity in nuclear extracts prepared from CD40-stimulated
cells in a bandshift assay using a radiolabeled oligonucleotide probe
containing a defined NF-B binding sequence from the HIV long
terminal repeat 43 . Specific retardation of the probe was visualized
by autoradiography. Pilot studies again showed that 1 µg/ml of mAb
G28.5 or sCD40L was sufficient to induce an optimal response in this
assay. The optimal stimulation time was approximately 3 h, with
detectable NF-B levels beginning to fall by 12 h
poststimulation.
All lines were tested in this way, and Fig. 5 shows representative data from a single
experiment; similar results were obtained in two additional independent
experiments. Note that nuclear extracts of all the BL lines tested
produced two retarded bands on the gel, both of which appeared to be
specific in that they were competed out only by the unlabeled cognate
oligonucleotide probe and not by irrelevant competitor sequences (data
not shown). All HR lines showed increased NF-B binding activity in
response to both mAb G28.5 and sCD40L, as illustrated in Fig. 5 by the
results from Ramos, Mutu-I, and Akata. In contrast, LR lines showed no
detectable (or occasionally a very marginal) NF-B response to mAb
G28.5 stimulation. However, in four of the five LR lines tested
(illustrated by L3055 and Sal), we did observe a significant increase
in NF-B activity in response to sCD40L. Interestingly Rael, the only
line that was nonresponsive to sCD40L in terms of CD95 induction (see
Fig. 4B), also gave no detectable NF-B response (Fig. 5).
|
presents an overall summary of
the data for all 14 BL lines tested in each of the different CD40
response assays described above; each parameter was tested in a minimum
of three independent experiments. Table I highlights the consistency of
the HR and LR response patterns observed in a range of assays.
|
In light of these clear differences in BL cell responsiveness to CD40 ligation, we used a similar approach to compare HR and LR lines in their responsiveness to EBV-encoded LMP1 expressed in the cell lines by transient transfection. In these experiments cells were electroporated with the pSG5-LMP1 expression vector or with the parental pSG5 vector as a control, in each case alongside a marker vector expressing rat-CD2. Immunofluorescent staining of the cells for surface expression of rat-CD2 thus provided both an indicator of transfection efficiency and a means of identifying the transfected subpopulation. Four HR lines (Ramos, DG75, Akata, and Eli) and two LR lines (Sal and Rael) were sufficiently transfectable (i.e., producing 8–20% rat-CD2-positive cells) to allow their study in this way.
We first examined the effects of LMP1 expression on NF-B activation.
In this case bandshift analysis was not feasible because the
transfected cells formed only a small proportion of the whole
population. Instead, we assayed NF-B activity using a cotransfected
reporter construct expressing luciferase under the control of a
NF-B-responsive promoter 46 . Fig. 6
presents the results of three independent experiments as the mean
luciferase counts ± SD following normalization for transfection
efficiency. We observed significant activation of NF-B in all four
HR lines and in both LR lines tested; the effects were reproducible and
independent of the baseline NF-B levels in each line.
|
A) and CD54 (Fig. 7B), in each case analyzing only the rat-CD2-positive
population. We observed consistent up-regulation of CD95 in all four HR
lines and in both LR lines (Fig. 7A). CD54 was up-regulated
in three of four HR lines (the exception again being Akata-BL) and
again in both LR lines (Fig. 7B). In these experiments,
transfection per se did not affect the baseline levels of either cell
surface marker. Thus, in control cotransfections of rat-CD2 and pSG5
vector, the rat-CD2-positive cells showed the same CD95 and CD54
expression levels as the rat-CD2-negative cells in the same culture.
Fig. 8 shows the results of confocal
microscopic analysis following double staining for LMP1 and CD54 in one
of the LR lines, Rael, transiently transfected with either the pSG5 or
the pSG5-LMP1 vector; EBV-transformed LCL cells were simultaneously
analyzed as a positive control. The data show firstly that CD54
up-regulation is restricted to LMP1-transfected cells, and secondly
that LMP1 levels produced by transient transfection with the pSG5-LMP1
construct do not exceed normal LCL-like levels. In an extension of
these experiments, we transiently transfected Rael cells with the
rat-CD2 vector and a vector, pcDNA3-CD40, that is known to express a
functional CD40 molecule 44 . Subsequent exposure of the cells to
CD40L did not result in any detectable CD54 or CD95 up-regulation in
the rat-CD2-positive population (data not shown).
|
|
presents the cell proliferation data
from one representative set of experiments comparing LMP1-expressing vs
control cell populations. LMP1 expression was associated with a slowing
of cell growth not only in the HR cell backgrounds Ramos and Akata, but
also in the LR line Sal. We consistently observed this result in assays
comparing a total of four control and four LMP1-positive clones
established on the Sal background.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The HR group of lines was easily identified by virtue of the homotypic
aggregation invariably induced by a wide range of agonist
concentrations. Every one of these lines also showed other
well-documented features of the CD40-induced response, namely
activation of NF-B and specific changes in cell surface phenotype
such as up-regulation of CD95 and (except for Akata) of CD54. Finally,
all the HR lines displayed clear evidence of growth inhibition
following CD40 ligation. This is the first systematic analysis of such
growth effects in BL lines and adds to earlier reports of CD40-mediated
growth inhibition in particular B cell lines of murine and human origin
29, 31 . In the present study a marked growth inhibition of HR lines
was consistently observed in the absence of detectable cell death by
apoptosis. In this respect the results mirrored the growth inhibition
induced by CD40 ligation in human epithelial cell lines 43 and were
unlike those observed in another study following high dose CD40L
treatment of human/mouse heterohybridomas in vitro 30 .
It is significant that in our experiments the lowest dose of either CD40 mAb or sCD40L (0.1 µg/ml) was sufficient to induce optimal aggregation in all HR lines, whereas other aspects of the response required at least 1 µg/ml for optimal induction. Hence, aggregation per se was distinguishable from other effects, in particular from the up-regulation of ICAM-1/CD54, an adhesion molecule that in CD40-stimulated normal B cells can mediate homotypic aggregation through binding to its counter-receptor, LFA-1 13, 23 . This distinction is further emphasized by results from the HR line Akata, which is universally negative for detectable CD54 expression under any condition of CD40 ligation and yet showed a clear aggregation response. Interactions between LFA-1 and alternative ligands such as ICAM-2 and ICAM-3 can also mediate adhesion 48 . However, LFA-1/ICAM interactions do not appear to be important in the present context, since attempts to block CD40-induced aggregation in the HR lines using mAbs to either CD18 (the ß subunit of LFA-1) or CD54 produced only a weak transient effect, whereas in parallel experiments they achieved complete disaggregation of EBV-transformed lymphoblastoid lines (N.V.H., unpublished observations). This is reminiscent of other reports that LFA-1 blocking Abs led to only partial inhibition of CD40-mediated cell aggregation in B cell systems 30, 45 . Together the data imply that other receptor-ligand interactions must be responsible for the bulk of the CD40-induced homotypic aggregation in the BL system. Another candidate pathway, the CD23-CD21 interaction 45 , is also unlikely to be involved, because CD23 was virtually undetectable on most lines, and CD40 ligation up-regulated CD23 expression on only one of the 14 BL lines tested, Ramos (data not shown).
It is also important to note that aggregation and growth inhibition were differentially dependent on the CD40 agonist dose, with only aggregation observable at the lowest concentration tested. Hence, the observed effects on cell growth at higher agonist doses are not nonspecific consequences of cell aggregation per se. This is further supported by the fact that one can induce aggregation in cells treated with the control mAbs (such as 1A3) by adding anti-mouse IgG as a secondary cross-linking Ab, yet this never impaired the growth of any of the HR lines tested (N.V.H., unpublished observations). Finally, although cell surface phenotypic changes occurred at similar agonist doses as did growth inhibition, the data from the HR line Akata clearly show that growth inhibition can occur in the complete absence of at least one feature of surface phenotypic change, i.e., CD54 up-regulation.
The LR group of lines, initially identified by a complete absence of
CD40-induced aggregation, showed a much more limited range of
CD40-induced responses when tested in the full range of assays. Thus,
neither CD40 mAb nor sCD40L had any detectable effect on cell growth or
on cell surface expression of CD54. However, most (but not all) LR
lines did show positive responses in the CD95 and NF-B induction
assays, but usually only following exposure to sCD40L. The limited
nature of CD40-induced effects in LR lines was observed despite
relatively strong CD40 surface expression (Table I). Furthermore,
addition of secondary cross-linking anti-mouse IgG to CD40
mAb-treated LR lines did cause aggregation (as did cross-linking of
anti-3 mAb-treated cells; see above), but, again, this did not
induce any of the other parameters of the CD40-induced response in
these lines (N.V.H., unpublished observations). Such data imply that
the different responses observed between LR and HR lines cannot be
explained by quantitative differences in CD40 expression and/or in CD40
cross-linking, but, rather, reflect qualitative differences in
engagement of downstream signaling pathways.
The results with LR lines were also interesting in that they provided
an example in which the CD40L-induced response was distinct from that
induced by CD40 mAb. Indeed, the results observed with sCD40L
highlighted an apparent correlation between the induction of CD95
expression and the activation of NF-B; both aspects of the
CD40L-induced response were seen in four of five LR lines tested, but
both were absent in the fifth line, Rael. It is possible that NF-B
plays a role in CD95 induction, and this would be consistent with the
reported presence of NF-B response elements in the CD95 promoter
region 49 . A corollary of these results is that NF-B activation
(at least as measured by the bandshift assay used here) and also CD95
induction can occur independently of any effect on the CD54 expression
level or on cell growth. In this context there were no obvious
quantitative differences in the levels of NF-B induced by sCD40L in
HR and LR lines. Certainly, the NF-B activation seen in L3055 (Fig. 5) and in BL40 (data not shown) was at least as marked as that observed
in several HR lines (e.g., Ramos; Fig. 5). It will be interesting to
determine how another facet of the CD40 response in BL cells, namely
protection from apoptosis, compares in HR and LR lines. The reported
ability of sCD40L to rescue the line L3055 from ionomycin- and
anti-IgM-induced apoptotic cell death suggests that such rescue may
be part of the limited LR cell response 25, 27 . However, the point
needs to be examined in a proper comparative study, and this will
require standardization of the various apoptosis-inducing protocols to
one that is similarly effective across a range of lines 21, 27 .
Given the differences in response between HR and LR lines to CD40
ligation, we were interested to determine whether a similar dichotomy
could be observed in the response to LMP1. This was particularly
important because perusal of the literature showed that almost all
reports of LMP1-induced phenotypic effects in BL cells had examined
lines that we identified as HR in the present study 32, 33, 34, 37, 39 ;
studies on representative LR lines, therefore, became particularly
significant. The final sets of experiments clearly show that in
transient transfection assays LMP1 induces NF-B, CD95, and CD54
equally well in HR and LR cell backgrounds. The capacity of LMP1 to
induce CD95 in BL lines has not been shown previously and provides yet
another parallel between LMP1- and CD40-mediated effects. Note that one
of the LR lines used in this part of the work, Rael, was uniformly
unresponsive to any form of CD40 ligation and yet (along with a second
LR line, Sal) showed the full spectrum of changes following LMP1
transfection. We were concerned that the use of transient transfection
to study LMP1 responsiveness might be a source of artifact, especially
if the LMP1 levels thus produced were atypically high. However, this
was not the case; LMP1 expression in transiently transfected Rael cells
was below LCL-like levels (Fig. 8). Moreover, in a converse experiment,
Rael cells were transiently transfected with a functional CD40
expression vector but did not acquire responsiveness to CD40L (data not
shown). Furthermore, in stable LMP1 transfectants on the LR cell
background, Sal, where there was a significant slowing of growth
compared with that in controls, LMP1 was again expressed at or below
LCL-like levels. These observations make it clear that LMP1 is capable
of inducing the full range of phenotypic effects in cells that show
grossly impaired responses to CD40 ligation. This may reflect the fact
that CD40 signaling is ligand dependent, whereas LMP1 functions
constitutively. In addition, however, it is likely that the two
molecules signal through overlapping but nonidentical combinations of
TRAFs. Firstly, even in the immediate vicinity of their PxQxT
TRAF-binding motif, CD40 and LMP1 show little sequence homology 37, 38 , implying that the two domains may be functionally distinct.
Secondly, each molecule appears to have other unique TRAF-interacting
domains that, again, may induce differential effects. Thus, CD40
contains a domain interacting with TRAF6, whereas the CTAR2 domain of
LMP1 can use TRADD (TNFR-associated death domain protein) as an
adapter molecule that provides an additional link to TRAF pathways 50, 51 .
The present findings imply that in at least a proportion of BL tumors, CD40 is expressed at the tumor cell surface, but CD40-mediated signaling may be impaired. It will be important to understand the molecular basis of the differential responsiveness to CD40 ligation in HR and LR lines and to determine whether such differences are apparent between freshly isolated tumor cell preparations as well as between tumor-derived lines. The possibility remains that since reception of a CD40-mediated signal by B cells in the absence of cognate Ag can lead to cell elimination in vivo 9 , then the selection of cells with a lesion in the CD40 pathway may in some cases be a contributory step in BL pathogenesis.
|
Acknowledgments |
|---|
B bandshifts. We
also thank Dr. Phil Gallimore for assisting with the photographs and D.
Hardie for performing the confocal microscopy. |
Footnotes |
|---|
2 Address correspondence and reprint requests to Prof. A. B. Rickinson, Cancer Research Campaign, Institute for Cancer Studies, University of Birmingham, Edgbaston, Birmingham, B15 2TA, U.K. E-mail address:
3 Abbreviations used in this paper: CD40L, CD40 ligand; TRAF, TNF receptor-associated factor; GC, germinal center; BL, Burkitt’s lymphoma; LMP1, latent membrane protein 1; HR, CD40 high responder; LR, CD40 low responder; LCL, lymphoblastoid cell line; sCD40L, trimeric soluble CD40L; PE, phycoerythrin; MFI, mean fluorescence intensity.
Received for publication July 13, 1998. Accepted for publication December 16, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B mobilization and induction of IL-6 production. J. Immunol. 155:4588.[Abstract]
B by TNF receptor 2 and CD40. Science 269:1424.B activation and to induction of cell surface markers. J. Virol. 68:5602.B. J. Biol. Chem. 267:24157.B and cell surface phenotype via two effector regions in its carboxy-terminal cytoplasmic domain. Oncogene 10:549.[Medline]
B activation. Mol. Cell. Biol. 16:7098.[Abstract]
B pathway involving TNF receptor-associated factors. Oncogene 14:2899.[Medline]
B and increases human immunodeficiency virus replication in human monocytes and T lymphocytes. J. Virol. 67:6596.B activation. Oncogene 15:1851.[Medline]
B. Proc. Natl. Acad. Sci. USA 94:12592.This article has been cited by other articles:
|
|
 |
|
  L. Terrin, J. Dal Col, E. Rampazzo, P. Zancai, M. Pedrotti, G. Ammirabile, S. Bergamin, S. Rizzo, R. Dolcetti, and A. De Rossi Latent Membrane Protein 1 of Epstein-Barr Virus Activates the hTERT Promoter and Enhances Telomerase Activity in B Lymphocytes J. Virol., October 15, 2008; 82(20): 10175 - 10187. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Le Clorennec, T.-S. Ouk, I. Youlyouz-Marfak, S. Panteix, C.-C. Martin, J. Rastelli, E. Adriaenssens, U. Zimber-Strobl, J. Coll, J. Feuillard, et al. Molecular Basis of Cytotoxicity of Epstein-Barr Virus (EBV) Latent Membrane Protein 1 (LMP1) in EBV Latency III B Cells: LMP1 Induces Type II Ligand-Independent Autoactivation of CD95/Fas with Caspase 8-Mediated Apoptosis J. Virol., July 1, 2008; 82(13): 6721 - 6733. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Le Clorennec, I. Youlyouz-Marfak, E. Adriaenssens, J. Coll, G. W. Bornkamm, and J. Feuillard EBV latency III immortalization program sensitizes B cells to induction of CD95-mediated apoptosis via LMP1: role of NF-{kappa}B, STAT1, and p53 Blood, March 1, 2006; 107(5): 2070 - 2078. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Hill, S. J. Youde, S. Man, G. R. Teale, A. J. Baxendale, A. Hislop, C. C. Davies, D. M. Luesley, A. M. Blom, A. B. Rickinson, et al. Activation of CD40 in Cervical Carcinoma Cells Facilitates CTL Responses and Augments Chemotherapy-Induced Apoptosis J. Immunol., January 1, 2005; 174(1): 41 - 50. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. P. Lee, J. M. Brooks, H. Al-Jarrah, W. A. Thomas, T. A. Haigh, G. S. Taylor, S. Humme, A. Schepers, W. Hammerschmidt, J. L. Yates, et al. CD8 T Cell Recognition of Endogenously Expressed Epstein-Barr Virus Nuclear Antigen 1 J. Exp. Med., May 17, 2004; 199(10): 1409 - 1420. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. N. D'Souza, L. C. Edelstein, P. M. Pegman, S. M. Smith, S. T. Loughran, A. Clarke, A. Mehl, M. Rowe, C. Gelinas, and D. Walls Nuclear Factor {kappa}B-Dependent Activation of the Antiapoptotic bfl-1 Gene by the Epstein-Barr Virus Latent Membrane Protein 1 and Activated CD40 Receptor J. Virol., February 15, 2004; 78(4): 1800 - 1816. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. G. P. Atkinson, H. J. Coope, M. Rowe, and S. C. Ley Latent Membrane Protein 1 of Epstein-Barr Virus Stimulates Processing of NF-{kappa}B2 p100 to p52 J. Biol. Chem., December 19, 2003; 278(51): 51134 - 51142. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G Siegler, E Kremmer, R Gonnella, and G Niedobitek Epstein-Barr virus encoded latent membrane protein 1 (LMP1) and TNF receptor associated factors (TRAF): colocalisation of LMP1 and TRAF1 in primary EBV infection and in EBV associated Hodgkin lymphoma Mol. Pathol., June 1, 2003; 56(3): 156 - 161. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Paludan, K. Bickham, S. Nikiforow, M. L. Tsang, K. Goodman, W. A. Hanekom, J.-F. Fonteneau, S. Stevanovic, and C. Munz Epstein-Barr Nuclear Antigen 1-Specific CD4+ Th1 Cells Kill Burkitt's Lymphoma Cells J. Immunol., August 1, 2002; 169(3): 1593 - 1603. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N J Gallagher, A G Eliopoulos, A Agathangelo, J Oates, J Crocker, and L S Young CD40 activation in epithelial ovarian carcinoma cells modulates growth, apoptosis, and cytokine secretion Mol. Pathol., April 1, 2002; 55(2): 110 - 120. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Mehl, M. Jones, M. Rowe, and P. Brennan Characterization of a CD40-Dominant Inhibitory Receptor Mutant J. Immunol., December 1, 2001; 167(11): 6388 - 6393. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. W. Tong, M. H. Papayoti, G. Netto, D. T. Armstrong, G. Ordonez, J. M. Lawson, and M. J. Stone Growth-inhibitory Effects of CD40 Ligand (CD154) and Its Endogenous Expression in Human Breast Cancer Clin. Cancer Res., March 1, 2001; 7(3): 691 - 703. [Abstract] [Full Text]
|
|
|
|
|
|
N. S. Andersen, J. K. Larsen, J. Christiansen, L. B. Pedersen, N. S. Christophersen, C. H. Geisler, and J. Jurlander Soluble CD40 ligand induces selective proliferation of lymphoma cells in primary mantle cell lymphoma cell cultures Blood, September 15, 2000; 96(6): 2219 - 2225. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Munz, K. L. Bickham, M. Subklewe, M. L. Tsang, A. Chahroudi, M. G. Kurilla, D. Zhang, M. O'Donnell, and R. M. Steinman Human CD4+ T Lymphocytes Consistently Respond to the Latent Epstein-Barr Virus Nuclear Antigen EBNA1 J. Exp. Med., May 8, 2000; 191(10): 1649 - 1660. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. R. Posner, L. A. Cavacini, M. P. Upton, K. C. Tillman, E. R. Gornstein, and C. M. Norris Jr. Surface Membrane-expressed CD40 Is Present on Tumor Cells from Squamous Cell Cancer of the Head and Neck in Vitro and in Vivo and Regulates Cell Growth in Tumor Cell Lines Clin. Cancer Res., August 1, 1999; 5(8): 2261 - 2270. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or in the presence of 10 µg/ml control mAb 1A3 (
) or 10
µg/ml sCD40L (
). Cells were counted 1, 2, and 3 days after
plating. The data are the mean of three experiments; SDs were within
10% of the mean in every case.
) or CD40 mAb G28.5 at 0.1 µg/ml
(
). The data represent the
mean of three experiments; SDs were within 15% of the means in every
case.
) or induced (
and 
) or the pSG5-Neo-LMP1 vector (