|
|
||||||||


*
Laboratory of Immunovirology, Pediatric Research Center and Department of Microbiology and Immunology, University of Montreal and Ste.-Justine Hospital, Montreal, Quebec, Canada;
Department of Pediatrics, National Jewish Center for Immunology and Respiratory Medicine, Denver, CO;
Department of Otorhinolaryngology, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia; and
§
Divisions of Experimental Oncology, and Medical Oncology and AIDS, Centro di Riferimento Oncologico, Aviano (PN), Italy
| Abstract |
|---|
|
|
|---|
5%) of the healthy, EBV-seropositive
individuals were positive for them; however, their OD values were much
lower than those of NPC patients. These studies demonstrate, for the
first time, the potential significance of LMP-1-specific Abs for the
diagnosis and prognosis of EBV-associated malignancies, especially of
NPC. | Introduction |
|---|
|
|
|---|
EBV has been causally associated with many human tumors, e.g.: nonkeratinizing nasopharyngeal carcinoma (NPC), which is prevalent in Southeast Asia, North Africa and in certain indigenous populations of North America; endemic Burkitts lymphoma in Africa and with the classic Hodgkins disease (HD) (2, 3). Recently, EBV has been found associated with T cell lymphomas, NK cell granulomas of the nasal septum, gastric tumors, and B cell lymphomas of AIDS patients (1, 2, 3). In the latter patients, EBV also causes oral hairy leukoplakia characterized by the localized lesions in the oral cavity, especially on tongue, in which intensive viral replication is taking place (2). In vitro, EBV infects and readily immortalizes human B cells into continuously growing lymphoblastoid cell lines (2).
Latent membrane protein-1 (LMP-1) is one of the limited number of EBV Ags that are expressed in the latent EBV infection. This infection is exemplified by EBV-immortalized B cells (1). In the latently infected cells, EBV exists as an episome that replicates once with each cell division and expresses only a few genes, i.e., LMP-1, -2A, -2B, six EBV nuclear Ags, and two small polyadenylated EBV RNAs (EBERs) (1, 3). In addition to its role in latent infection (see below), LMP-1 (hereafter referred to as LMP) also seems to play a role in lytic EBV infection by acting as an antiapoptotic protein and thereby delaying the cell death until all the viral structural and nonstructural proteins are expressed, and virions bud off the infected cells (1, 4). Because LMP has been detected in virions, it may also play a role in the viral infection process (5).
LMP is a 63-kDa type III protein which carries six hydrophobic
transmembrane regions and occurs as aggregated patches spanning the
surface of EBV-infected and/or-immortalized cells. It resembles
TNFR family of proteins (6). Its intracellular
carboxy-terminal cytosolic region interacts with signaling molecules
that are involved in the pathways activated by the TNFR family members
which include CD40 (3, 7, 8, 9, 10, 11). By this virtue, LMP acts as
a constitutively activated CD40 receptor in B cells. LMP transforms
rodent fibroblasts and is essential for the EBV-induced immortalization
of B cells (1, 3, 12). It induces activation of NF-
B,
c-JNK, and AP-1, and the expression of antiapoptotic genes
Bcl2, Mcl-1, and A20 in different cell
types (4, 13, 14). In LMP-transfected cells, it induces
several phenotypic changes characteristic of the EBV-infected cells,
e.g., CD23, epidermal growth factor receptor, certain adhesion
molecules, etc. (3, 13, 14). Recently, LMP was shown to
induce the expression of cyclin D2 and hyperphosphorylation of pRb,
rendering cells unresponsive to the growth-inhibitory effects of
TGF-ß1 (15). These studies strongly suggest a role of
LMP-1 in the EBV-associated lymphoproliferative diseases and
malignancies. This is further supported by the consistent expression of
this protein in many EBV-associated tumors.
The EBV-infected subjects mount a vigorous antiviral cellular and humoral immune responses (reviewed in Refs. 16, 17). These responses, particularly cellular ones, are thought to keep the viral replication under control in healthy subjects. The detection of humoral responses (i.e., Abs) against various viral Ags has been very useful for the diagnosis and prognosis of EBV-associated disease conditions. LMP-specific cellular immune responses have been well documented (16, 17). However, information on anti-LMP Abs is scanty despite the fact that this protein is expressed on the surface of EBV-infected/immortalized cells and may be targeted by such Abs. The detection of these Abs may be helpful not only in the diagnosis and prognosis of EBV-associated diseases but may also provide insights about their pathogenetic processes. We report here our studies on the detection, analysis, and significance of LMP-specific Abs in different EBV-associated diseases and show for the first time that these Abs can readily be demonstrated in the sera of patients with EBV-associated malignancies and that anti-LMP IgA may be of prognostic value in patients with advanced stages of NPC.
| Materials and Methods |
|---|
|
|
|---|
LMP-1 was expressed in an EBV-genome negative human B cell line BJA-B derived from Burkitts lymphoma (18). For this purpose, a pGEM2 (Promega, Madison, WI)-based expression plasmid pIgLMP-1 (provided by Dr. Nancy Raab-Traub, University of North Carolina, Chapel Hill, NC) was transfected into these cells by electroporation as described earlier (19). This plasmid has an Ig promoter/enhancer 5' to two tandemly arranged copies of the 3.0-kb LMP-1 genomic sequences and carries a neomycin resistance marker for selection of transfected eukaryotic cells. The control plasmid (pIgNeo) was obtained from pIgLMP-1 after deleting the LMP sequences. The transfected cells were cloned by limiting dilution in the selection medium, i.e., RPMI 1640 containing 10% FBS and 1 mg per ml of geneticin (Life Technologies, Burlington, ON, Canada). The transfected cell clones were screened for LMP-1 expression by immunofluorescence using S12 (provided by Dr. Elliott Kieff, Harvard Medical School, Boston, MA) and CS1-4 (Dako, Glostrup, Denmark) Abs. The selected clones were maintained in the selection medium, and their lysates were used to detect anti-LMP-1 Abs in the sera (see below).
Serum samples
Serum samples for the detection of anti-LMP Abs were obtained after informed consent from the following groups of individuals: category 1, healthy EBV seronegatives (HN), 3; category 2, healthy EBV seropositives (HI), 20; category 3, IM, 15; category 4, chronic EBV infection (CEI), 25; category 5, HD, 15; category 6, nonkeratinizing NPC, 86.
These EBV-associated disease conditions have been defined in our recent publications (20, 21). We further divided the NPC patients into different stages based on tumor size (T), lymph node involvement (N), the presence of hematogenous metastases (M), patient death or tumor remission, and AJC stages as described (22).
Radioimmunoprecipitation (RIPA) and Western blots
RIPA was conducted as described by us earlier (19).
Briefly, 5 x 106 cells were labeled with
250 µCi [35S]methionine (NEN/DuPont), lysed
in RIPA buffer in the presence of protease inhibitors, and
immunoprecipitated using 2 µl of the undiluted ascites fluid of S12
Ab or 40 µl of the serum sample. The immunoprecipitates were resolved
on 10% SDS-PAGE and detected by autoradiography. For Western blots,
5 x 106 cells were lysed in 50 mM Tris-HCl
(pH 6.8), 2% SDS and sonicated for 15 s. The lysates were
clarified by centrifugation (14,000 x g for 30 min at
4°C), and their protein concentrations were determined with a
commercial kit (Bio-Rad Laboratories, Hercules, CA). Forty micrograms
of the proteins were resolved on 10% SDS-PAGE, electroblotted onto
nylon membranes (Immobilon, Millipore, Bedford, MA). The membranes were
blocked in 5% skim milk powder in PBS and then incubated with
appropriately diluted S12, CS1-4, or serum samples. After washings,
blots were developed using alkaline phosphatase (AP)-conjugated
secondary Abs (anti-mouse or anti-human IgG; both from Promega)
and chromogenic substrates 5-bromo-4-chloro-3-indolyl phosphate and
nitroblue tetrazolium (Promega). For detection of LMP-1-specific IgA
Abs, AP-conjugated rabbit anti-human IgA (
-chain specific; Dako;
at 1:1000 dilution) was used in Western blots.
Indirect immunofluorescence
For this purpose, cells were fixed and premeabilized using the Cytofix/Cytoperm Kit (PharMingen, San Diego, CA). They were incubated with CS1-4 or S12 for 45 min at 4°C, washed, and stained with FITC-conjugated goat anti-mouse IgG. The stained cells were examined under fluorescent microscope or analyzed by flow cytometry using FACScan.
ELISA protocol
For the quantitation of anti-LMP-1 Abs, a LMP-1-specific
ELISA was developed using a protocol described by us earlier
(23) after some modifications. Briefly, plastic ELISA
plates (Corning, Corning, NY; catalog number 28805) were coated (at
4°C overnight) with 50 µl of the cell lysates (protein content, 100
µg/ml) from LMP-1-expressing or control vector-transfected BJA-B
cells. These lysates were obtained as described above for Western
blots. The wells were washed three times with wash buffer containing
0.1% gelatin (Bio-Rad) and 0.05% Tween 20 in PBS and blocked with
0.2% gelatin in PBS (200 µl/well) at 4°C overnight. After washing,
the wells were incubated with 50 µl of the serum dilution and
incubated at room temperature for 2 h. After three subsequent
washings with the wash buffer, 50 µl of 1:7500 diluted AP-conjugated
anti-human IgG (heavy and light chain, Promega) was added. After
incubation at room temperature for 2 h with the secondary Ab, the
plates were washed extensively with the wash buffer, and colors were
developed by using 50 µl p-nitrophenyl phosphate (Sigma,
St. Louis, MO; catalog number N-9389, 1 mg/ml) dissolved in 0.1 M
glycine buffer (pH 10.4) containing 1 mM MgCl2
and 1 mM ZnCl2. The reactions were stopped 30 min
later with the addition of 50 µl of 3 N NaOH, and OD was measured at
405
in an automatic ELISA reader (Easy Reader EAR 400AT, STL
Labinstruments, Salzburg, Austria).
| Results |
|---|
|
|
|---|
The cell clones that we used as a source of LMP-1 in these studies
could be easily stained and detected by flow cytometry using
anti-LMP Abs (CS14 or S12) and FITC-conjugated secondary Abs
after permeabilization (Fig. 1
). These
Abs, however, were unable to stain these cells by indirect membrane
immunofluorescence without permeabilization (data not shown). With
these Abs, LMP-1 could also be demonstrated in these cells by RIPA and
Western blots (Fig. 2
and data not
shown). Following this, we used sera from EBV-seronegative and healthy
EBV-seropositive individuals and from NPC patients in the RIPA and
Western blots to see whether they contained anti-LMP Abs. As shown
in Figs. 3
and
4, several NPC sera showed positivity for
these Abs. However, only a single sample of 20 EBV-seropositive healthy
sera showed positivity for anti-LMP Abs. Sera from EBV-seronegative
individuals were always negative both in RIPA and Western blots (data
not shown). These data strongly suggested the presence of anti-LMP
humoral responses in EBV-infected individuals. Interestingly, sera from
CEI and IM patients were negative for these Abs. Table I
depicts the percent positivity of sera
for anti-LMP-1 Abs for various EBV-associated disease conditions.
Within NPC patients, the percentage of anti-LMP Ab-positive sera
showed a clear increasing trend in higher stages of the tumor (Table II
). An overwhelmingly high percentage of
the dead NPC patients was positive for anti-LMP Abs (Table II
). In
the case of HD, all EBER and/or LMP-positive tumors were positive for
these Abs (Table III
). One (of six)
EBER-negative HD patient also contained these Abs. Taken together,
these results illustrate the presence of anti-LMP humoral
responses, especially in patients with EBV-associated NPC and
HD.
|
|
|
|
|
|
Although anti-LMP Abs were detected by RIPA and/or Western
blots, in many cases the revealed bands were faint, showed background
darkness, and were not amenable to quantitative analyses. Therefore, to
quantitate these Abs in the sera, we developed ELISA using lysates from
LMP-expressing and control BJA-B cells. First, we determined the
suitability of these lysates for ELISA using CS1-4 or S12 Abs. Both the
LMP-1-specific Abs gave higher OD as compared with control Abs when
lysates from LMP-1-expressing cells were used, whereas the OD of these
Abs did not differ from those of control Abs when lysates from the
control-transfected cells were used, suggesting specificity of the
ELISA (data not shown). We then used pooled sera from EBV-seronegative
subjects, EBV-seropositive healthy subjects, and from each
EBV-associated disease separately in the ELISA. The results of the
ELISA are depicted in Fig. 5
. The highest
ODs were observed in NPC sera followed by the sera from HD patients.
The OD values from these two diseases were significantly
(p
0.05) higher than those of HI, IM, and
CEI sera at all dilutions tested (Fig. 5
).
|
The NPC sera were classified as described in Materials and
Methods. To determine whether these anti-LMP-1 Abs differed in
various stages of this tumor, all sera from each category were tested
individually in ELISA and average OD values for each category of serum
were calculated. As shown in Fig. 6
,
AD, in each category, a trend toward higher average OD
with increase in tumor size
(T0
T4), increased nodal
involvement (N0
N3) and
with metastasis is evident. Higher average ODs were also seen in sera
from dead patients as compared with the patients with remissions.
Furthermore, these OD values also increased in advanced stages of the
tumor. These results strongly suggest a prognostic value of these
Abs.
|
Because earlier studies from this and other laboratories have
shown that Abs of IgA isotype against several EBV Ags are of diagnostic
and prognostic value, especially for NPC (16, 20, 24), we
did Western blots using AP-conjugated human
-chain-specific
secondary Abs to detect these Abs in these sera. Fig. 7
depicts a typical Western blot, and the
results from several blots are shown in Table IV
. Only NPC sera (50%) were found to be
positive for anti-LMP-1 IgA Abs. The percent positivity for this Ab
in NPC sera also tended to increase in advanced stages of the tumor
(Table V
). These data collectively show
that anti-LMP IgA Abs are pathognomonic for NPC and have prognostic
value.
|
|
|
| Discussion |
|---|
|
|
|---|
Although the presence of anti-LMP Abs in normal sera was reported earlier (27), our results show that only a fraction (1 in 20; 5%) of healthy EBV-seropositive individuals are positive for these Abs. Concerning HD, we found one patient positive for anti-LMP Abs despite the absence of EBV genome in its tumor (Reed-Sternberg) cells. This confirms an earlier report (31) that, unlike Burkitts lymphoma, the presence or absence of anti-EBV Abs is not predictive of the EBV status of the tumor cells in HD. Interestingly, sera from IM and CEI patients were negative for these Abs. More importantly, two-thirds or more sera from patients with EBV-associated malignancies (NPC and HD) were positive for these Abs. Furthermore, the OD values of these Abs in ELISA also correlated positively with advanced stages of the tumor in NPC patients. IgA Ab titers for several EBV Ags, e.g., early Ag, viral capsid Ag, gp350/220, have been documented to have both diagnostic and prognostic significance in NPC (2, 24). We also found that anti-LMP Abs of IgA isotype were detectable only in the sera of NPC patients, suggesting pathognomonicity of these Abs to this malignant condition. Collectively, these results show the potential significance of anti-LMP Abs for the diagnosis and prognosis of EBV-associated malignancies.
The LMP-specific mAbs CS1-4 and S12 are known to recognize intracellular parts of LMP (32, 33), and LMP Ab-positive sera stained our LMP-expressing cells in indirect immunofluorescence assays only when these cells were first permeabilized. These Abs and sera, however, were unable to stain these cells without prior permeabilization, suggesting that these Abs recognized only intracellular parts of the protein. Similar observations were made earlier by Rowe et al. (27). These data suggest that extracellular parts of LMP (the three 8- to 12-amino acid-long turns) are not immunogenic and are not targeted by human anti-LMP Abs. This is further supported buy our observations that LMP-expressing cells are not killed by NK cells in Ab-dependent cellular cytotoxicity (ADCC) assays as described by us for anti-gp350/220 Abs (Ref. 20 and data not shown). Thus, it appears that anti-LMP humoral response does not play a role in the control of EBV-infected/immortalized cells. Therefore, LMP-expressing malignant cells are not under pressure to accumulate mutations to evade this response. This also reinforces the commonly held notion that anti-EBV cellular immune responses are more important to control EBV infections (16, 17).
LMP is a transforming protein that mimics (but is not identical with) a constitutively activated CD40 receptor (1, 3, 7, 9, 10, 14). It activates several cellular genes, acts as a classical oncogene and is the frequently expressed EBV protein in several EBV-associated malignancies. The expression of LMP-1 in 100% cases of rarely detected preinvasive NPC lesions, and monoclonality of EBV genomes in these tumors strongly suggest a causative role of EBV and this EBV protein in this tumor (34). LMP inhibits differentiation of epithelial cells and induces the expression of several antiapoptotic genes, e.g., bcl-2, A20, Mcl-1, depending on the cell type (3, 35). Antisense oligos to the first five codons of LMP induce apoptosis in EBV-immortalized B cells (36). Taken together, these observations strongly suggest a role for LMP in EBV-induced tumorigenicity. However, there is no clear in vivo evidence for such a role for this protein. On the contrary, LMP per se is cytostatic or even cytotoxic for human cells (37, 38), and LMP-positive tumors have better prognosis than LMP-negative tumors (39, 40). LMP is targeted by EBV-specific CTL and its expression in Burkitts lymphoma makes them susceptible to lysis by these CTL (39). It also enhances expression of MHC class I Ags and costimulatory molecules, e.g., CD80, CD86, and consequently increases the presentation of endogenous and exogenous Ags to the immune system (41, 42, 43). LMP-1-expressing Burkitts lymphomas are less tumorigenic than their LMP-negative counterparts in nude and SCID mice (39). Recently, Cherney et al. (44) demonstrated that expression of LMP augments nonspecific antitumor immunity by inducing the secretion of cytokines and chemokines. The EBV-induced tumors, however, may have evolved strategies to overcome these responses by accumulating mutations in LMP. An LMP gene cloned from an NPC patient was reported to be more tumorigenic and less immunogenic in mice than its B95-8 counterpart (45). Mutations in the promoter region and in the C-terminal cytoplasmic region of LMP have been documented by several workers (46, 47, 48, 49). A characteristic 30-bp mutation in the C-terminal part has attracted much attention (49, 50, 51). However, there is not yet convincing in vivo evidence of the role of this or any other LMP mutation in increased tumorigenicity. Our results suggest that anti-LMP Abs recognize hidden intracellular parts of LMP and do not mediate ADCC and therefore may not be driving the accumulation of these mutations in EBV-associated tumors. Instead, these mutations may enable tumors to escape LMP-specific CTL, render this protein unable to stimulate nonspecific antitumor immunity, or reduce its cytotoxic/cytostatic effects. Because human anti-LMP-1 Abs recognized only intracellularly located parts of the protein, the immune system may be exposed to LMP-1 on lysis of LMP-expressing EBV-infected cells and/or by virolysis of released virions. The titers of anti LMP Abs, therefore, may reflect the magnitude of this exposure. Our data also suggest that anti-LMP Abs increase in advanced stages of NPC. The increase in anti-LMP Abs in advanced NPC stages may simply reflect an enhanced exposure to LMP in these patients. Alternatively, these patients may be mounting disproportionally higher humoral responses as compared with the CTL responses. Because we have not determined LMP-specific cellular immune responses in these patients, further studies will be needed to address these issues.
This quantitation and analysis of anti-LMP Abs using ELISA underscores the potential usefulness of these Abs in the diagnosis and prognosis of EBV-associated malignancies, particularly NPC.
|
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Ali Ahmad, Laboratory of Immunovirology, Ste.-Justine Hospital, 3175 Côte Ste.-Catherine, Montreal, Quebec, H3T 1C5, Canada. E-mail address: ![]()
3 Abbreviations used in this paper: IM, infectious mononucleosis; CEI, chronic EBV infection; LMP-1, latent membrane protein-1; NPC, nasopharyngeal carcinoma; EBER, EBV RNA; HI, healthy EBV seropositives; RIPA, radioimmunoprecipitation assay; AP, alkaline phosphatase; ADCC, Ab-dependent cellular cytotoxicy. ![]()
Received for publication November 2, 1999. Accepted for publication December 17, 1999.
| References |
|---|
|
|
|---|
B activation. J. Virol. 71:586.[Abstract]
B activation. Mol. Cell. Biol. 16:7098.[Abstract]
B activation and to induction of cell surface markers. J. Virol. 68:5602.This article has been cited by other articles:
![]() |
W.-L. Hsu, J.-Y. Chen, Y.-C. Chien, M.-Y. Liu, S.-L. You, M.-M. Hsu, C.-S. Yang, and C.-J. Chen Independent Effect of EBV and Cigarette Smoking on Nasopharyngeal Carcinoma: A 20-Year Follow-Up Study on 9,622 Males without Family History in Taiwan Cancer Epidemiol. Biomarkers Prev., April 1, 2009; 18(4): 1218 - 1226. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Kobayashi, T. Nagato, M. Takahara, K. Sato, S. Kimura, N. Aoki, M. Azumi, M. Tateno, Y. Harabuchi, and E. Celis Induction of EBV-Latent Membrane Protein 1-Specific MHC Class II-Restricted T-Cell Responses against Natural Killer Lymphoma Cells Cancer Res., February 1, 2008; 68(3): 901 - 908. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Tedeschi, E. Pin, D. Martorelli, E. Bidoli, A. Marus, C. Pratesi, M. T. Bortolin, S. Zanussi, E. Vaccher, R. Dolcetti, et al. Serum Antibody Response to Lytic and Latent Epstein-Barr Virus Antigens in Undifferentiated Nasopharyngeal Carcinoma Patients from an Area of Nonendemicity Clin. Vaccine Immunol., April 1, 2007; 14(4): 435 - 441. [Abstract] [Full Text] [PDF] |
||||
![]() |
C.-L. Tsai, H.-P. Li, Y.-J. Lu, C. Hsueh, Y. Liang, C.-L. Chen, S. W. Tsao, K.-P. Tse, J.-S. Yu, and Y.-S. Chang Activation of DNA Methyltransferase 1 by EBV LMP1 Involves c-Jun NH2-Terminal Kinase Signaling Cancer Res., December 15, 2006; 66(24): 11668 - 11676. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Ahmad, S. T. A. Sindhu, E. Toma, R. Morisset, and A. Ahmad Elevated Levels of Circulating Interleukin-18 in Human Immunodeficiency Virus-Infected Individuals: Role of Peripheral Blood Mononuclear Cells and Implications for AIDS Pathogenesis J. Virol., November 13, 2002; 76(24): 12448 - 12456. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Xu, A. Ahmad, and J. Menezes Preferential Localization of the Epstein-Barr Virus (EBV) Oncoprotein LMP-1 to Nuclei in Human T Cells: Implications for Its Role in the Development of EBV Genome-Positive T-Cell Lymphomas J. Virol., March 19, 2002; 76(8): 4080 - 4086. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Woulfe, D. G. Munoz, J. De Keyser, and M. Lousa Status epilepticus with neuron-reactive serum antibodies: Response to plasma exchange Neurology, November 14, 2000; 55(9): 1421 - 1421. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |