Skip Navigation


Carcinogenesis Advance Access originally published online on March 14, 2007
Carcinogenesis 2007 28(8):1839-1848; doi:10.1093/carcin/bgm055
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrowOA All Versions of this Article:
28/8/1839    most recent
bgm055v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Google Scholar
Right arrow Articles by Charalambous, C. T.
Right arrow Articles by Wilson, J. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Charalambous, C. T.
Right arrow Articles by Wilson, J. B.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org
The online version of this article has been published under an open access model. Users are entitled to use, reproduce, disseminate, or display the open access version of this article for non-commercial purposes provided that: the original authorship is properly and fully attributed; the Journal and Oxford University Press are attributed as the original place of publication with the correct citation details given; if an article is subsequently reproduced or disseminated not in its entirety but only in part or as a derivative work this must be clearly indicated. For commercial re-use, please contact journals.permissions@oxfordjournals.org

Latent membrane protein 1-induced EGFR signalling is negatively regulated by TGF{alpha} prior to neoplasia

Chrystalla T. Charalambous, Adele Hannigan, Penelope Tsimbouri, Gordon M. McPhee and Joanna B. Wilson*

Division of Molecular Genetics, Biomedical and Life Sciences, University of Glasgow, Glasgow G11 6NU, UK

* To whom correspondence should be addressed. Tel: +141 330 5100; Fax: +141 330 4878; Email: Joanna.Wilson{at}bio.gla.ac.uk


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 References
 
The latent membrane protein 1 (LMP1) of Epstein–Barr virus (EBV) is an oncoprotein expressed in several EBV-associated malignancies. We have utilised mice expressing the Cao strain of LMP1 in epithelia to explore the consequences of expression in vivo, specifically the changes that occur prior to neoplasia, in the hyperplastic but degenerating tissue. Epidermal growth factor receptor (EGFR) ligands (transforming growth factor {alpha} (TGF{alpha}), heparin-binding EGF-like growth factor and epiregulin) are constitutively induced by LMP1, leading to EGFR phosphorylation but also down-regulation, degradation or turn-over, with the appearance of cleaved EGFR fragments. This is accompanied by down-regulation of Akt and activation of caspase-3 and p38 mitogen-activated protein kinase (MAPK). Surprisingly, removal of TGF{alpha} (using the null strain) does not ameliorate the LMP1-induced phenotype, but instead accelerates the deterioration. Consistent with this, EGFR is reduced less rapidly and MAPK/ERK kinase (MEK) and extracellular-signal-regulated kinase (ERK) are initially activated in the null background, suggesting that TGF{alpha} or excess of the ligands together act to divert phosphorylated EGFR into a cleavage pathway. In addition, LMP1 leads to the activation of c-Jun N-terminal kinase 2 (JNK2) followed by JNK1 in the effected tissue. Specific AP1 family members FosB, Fra-1 and JunB are constitutively induced and serum response factor, AP1 and nuclear factor {kappa}B (incorporating p65) are activated in the transgenic tissue compared with wild-type. This system allows the analysis of early events resulting from the expression of a viral oncogene with broad impact in the signalling milieu and the attempts at homeostasis in the responding tissue. It reveals what regulatory circuits are in place in a normal tissue, thus facilitating further prediction of causative events in carcinogenic progression.

Abbreviations: EBV, Epstein–Barr virus; EDTA, ethylenediaminetetraacetic acid; EGFR, epidermal growth factor receptor; EPR, epiregulin; HB-EGF, heparin-binding EGF-like growth factor; JNK, c-Jun N-terminal kinase; LMP1, latent membrane protein 1; MAPK, mitogen-activated protein kinase; NF-{kappa}B, nuclear factor {kappa}B; NPC, nasopharyngeal carcinoma; NSC, non-transgenic sibling controls; SDS, sodium dodecyl sulphate; SRE, serum response element; TRE, TPA response element; WT, wild-type


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 References
 
Epstein–Barr virus (EBV) is a human herpesvirus associated with several tumours of lymphoid and epithelial origin, including Burkitt's lymphoma and nasopharyngeal carcinoma (NPC). The latent membrane protein 1 (LMP1) of EBV has been shown to activate multiple signalling pathways in its action as the primary viral oncoprotein. It shows transforming activity in several cell types (1–4) is required for B-cell immortalisation in culture by the virus (5), causes epidermal hyperplasia and papillomatosis in transgenic mice (6,7) and predisposes transgenic mice expressing the protein in B-cells to lymphoma (8). LMP1 is expressed in EBV-associated Hodgkin's disease tumour cells and NPC cells and is thought to play a central role in the genesis of these tumours.

Studies in B-cells particularly and to some extent in epithelial cells have shown that LMP1 stimulates the nuclear factor {kappa}B (NF-{kappa}B) pathway from the C-terminal-activating regions 1 and 2 of the protein (9–15). In doing so, several NF-{kappa}B-regulated genes are effected including anti-apoptotic genes (bcl-2, A20, Mcl-1 and c-IAP2) and genes involved in angiogenesis, invasion and metastasis (MMP9, COX-2, VEGF, IL-8 and Id-1). LMP1 can activate c-Jun N-terminal kinase (JNK) via the C-terminal-activating region 2 domain (16–18) leading to c-Jun phosphorylation and assembly of the AP1 complex. LMP1 has also been shown to activate p38 MAPK leading to IL-6 and IL-8 up-regulation (19) and IP-10 mRNA stabilisation (20). In addition, the Janus kinase/signal transducer and activation of transcription cascade is also activated by LMP1 through direct binding of Janus kinase 3 to the proline-rich region in C-terminal-activating region 3 (21,22), with subsequent induction of c-Myc (23). Furthermore, LMP1 activation of the PI3K/Akt pathway has been described through recruitment of the p85 subunit of PI3K either directly or via an as yet unidentified linker (24). Finally, in rat fibroblasts, LMP1 has been shown to increase ERK1/2 (p42/p44 MAPK) phosphorylation in a Ras-dependent manner (25) and an LMP1-induced increase in hypoxia-inducible factor-1 activity has been shown to be dependent upon ERK1/2 activity (26).

The NPC-derived Cao variant of LMP1 (used in these studies) is more frequently encountered in NPCs from endemic regions (27). It has several differences compared with the B95-8 strain, including point mutations, small deletions and an amplification of the central repeats. It has been shown to activate NF-{kappa}B and JNK pathways to different degrees in different cell types relative to the B95-8 variant (28–31).

One target of NF-{kappa}B induced by LMP1 expression is the epidermal growth factor receptor (EGFR or ErbB1) (32,33) and up-regulation of EGFR correlates with LMP1 levels in NPC specimens (34,35). EGFR is a member of a family of tyrosine kinase receptors (including ErbB2, 3 and 4). Ligand binding leads to homo- or heterodimerisation, tyrosine phosphorylation and activation (36). Pathways activated by EGFR include the Ras/MAPK (via Shc or Grb2 adaptor proteins), the Janus kinase/signal transducer and activation of transcription and PI3K/Akt cascades. Receptors in the family (reported for EGFR, ErbB3 and ErbB4) can also translocate to the nucleus upon induction via their specific ligands and act to regulate transcription (37–39). Also, inducible expression of LMP1 in HNE2 cells can lead to the nuclear translocation of EGFR in a dose-dependent manner, and thereby increase the number of cells entering and progressing through the cell cycle (40). Approximately 30% of human cancers show elevated levels of EGFR and it has been correlated with poor prognosis and decreased disease-free and overall survival rates.

EGFR has several ligands including EGF, TGF{alpha} and amphiregulin that preferentially bind EGFR and heparin-binding EGF-like growth factor (HB-EGF), epiregulin (EPR), epigen and betacellulin. The ligands are first synthesised as a transmembrane presursor (proform) and are then cleaved to produce the mature soluble form and, in the case of HB-EGF at least, the membrane remnant can also translocate to the nucleus to modulate the cell cycle (41). TGF{alpha} plays an important role in embryonic development, wound healing, angiogenesis and keratinocyte proliferation and migration. However, aberrant expression can promote skin tumorigenesis and there are several animal models that show this (42–45).

Transgenic mice expressing LMP1 (of B95-8 and Cao strains) in the epidermis display a hyperplastic phenotype (6,7). The ears are particularly affected in the L2LMP1CAO mice (where expression is highest) with a phenotype from birth that gradually advances with age and has been categorised into visibly recognisable stages (st)—st1: mild hyperplasia with increased vascularisation, transiting between 4 and 8 weeks of age to st2: moderate hyperplasia and further vascularisation; st3 (by 2–3 months old): severe hyperplasia with ulcerative dermatitis; st4 (~6 months old): extensive hyperplasia and ulcerative dermatitis and tissue degeneration with frequent keratoacanthoma and st5 (variable, on average by 10 months old): severe hyperplasia with extensive tissue degeneration and occasional squamous cell carcinoma (7). In the FVB strain, the mice also develop papillomas on the dorsal skin. The phenotype of these mice is reminiscent of transgenic mice over-expressing TGF{alpha} (42) indeed it was found that LMP1 induced TGF{alpha} expression and EGFR phosphorylation in the affected tissues (7). This led to the hypothesis that LMP1 might mediate the phenotype, in part through up-regulation of TGF{alpha} and as such TGF{alpha} signalling would represent a therapeutic target in the treatment of LMP1 expressing EBV-associated epithelial disorders. In order to explore this possibility (in the absence of a suitable viable EGFR dominant-negative mouse model), mice null for TGF{alpha} have been used in cross-breed studies with L2LMP1CAO mice and MAPK and Akt signalling pathways have been examined in the tissues through the advancing phenotype.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 References
 
Transgenic lines
Transgenic L2LMP1CAO mice of the line 117 (7) were generated in the strain B6D2F1 and have since been backcrossed to FVB. Mice used in these studies were at backcross 6 or more (>98% FVB). L2LMP1CAO line 117 mice were cross-bred with TGF{alpha}-null mice (46) in a mixed strain background (FVB/C57Bl/6/129) to produce mice heterozygous for TGF{alpha}. These progeny were crossed together to generate TGF{alpha}-null, heterozygous or wild-type (WT) (LMP1 positive and negative) keeping the mixed strain constant and these siblings were monitored for phenotypic changes once a week. The proportion of LMP1/TGF{alpha}-null mice generated from all the crosses was approximately half the expected number for Mendelian frequencies (P = 0.0005 from Chi-square analysis, not seen in the LMP1-negative group), suggesting a reduction in embryonic fitness in this group. However, there was no apparent increase in postnatal mortality in this group of mice compared with the LMP1CAO transgenic and TGF{alpha}-null mice separately.

Western blotting
Proteins were extracted from ear tissue by polytron disruption of snap-frozen samples using Ripa buffer [20 mM Tris–HCl (pH 7.5), 150 mM NaCl, 5 mM ethylenediaminetetraacetic acid (EDTA) (pH 8.0), 1.5% NP40, 0.1% sodium dodecyl sulphate (SDS), 0.5% deoxycholic acid], with protease inhibitors (such as complete-mini Roche, 20 µl/ml) and phosphatase inhibitors (such as 10 µl/ml Sigma cocktail, cat.# P5726) freshly added and incubated for 15 min on ice. Following centrifugation, 16 000g for 10 min at 4°C, protein concentration of the supernatant was determined using the Bradford assay or 2D-Quant (GE Healthcare). Protein extract (100 µg) was mixed with sample loading buffer [to a final concentration: 62 mM Tris–HCl (pH 6.8), 10% glycerol, 5% 2-mercaptoethanol, 2% SDS, 0.002% bromophenol blue], heated at 95°C for 5 min, then subjected to SDS–polyacrylamide gel electrophoresis and blotted onto Immobilon P membrane. Electroblotting, blot probing, washing and detection were carried out as described previously (6,7,47).

Antibodies used for western and/or electrophoretic mobility shift assay were directed against the following cellular proteins: EGFR (cat.# 2232), phospho(tyr845)-EGFR (cat.# 2231), MEK1/2 (cat.# 9122), phospho(ser217/221)-MEK1/2 (cat.# 9121), ERK1/2 (p42/44) (cat.# 9102), phospho-ERK1/2 (cat.# 9101), JNK (cat.# 9252), phospho(thr183/tyr185)-JNK (cat.# 9251), Akt (cat.# 9272), phospho(ser473)-Akt (cat.# 9271), c-Jun (cat.# 9165), caspase-3 (cat.# 9662), asp175-cleaved caspase-3 (cat.# 9664), p38 (cat.# 9212) (Cell Signalling, USA) Fra-1 (cat.# sc-183), Fra-2 (cat.# sc-604), JunB (cat.# sc-46X), c-Fos (cat.# sc-253G), NF-{kappa}B p65 (cat.# sc-372), NF-{kappa}B p50 (cat.# sc-1190), HB-EGF (cat.# sc-1414), EPR (cat.# sc-25231), (Santa Cruz Biotechnology), phospho(thr180/tyr182)-p38 (cat.#V1211, Promega, UK) and GAPDH (6C5 cat.# H86504M, Biodesign, USA). These antibodies were used in the primary detection at a dilution of 1/1000 (1/2000 for GAPDH) in blocking buffer {PBST [PBS, 0.1% (v/v) Tween-20] with 5% [w/v] non-fat milk powder}. Secondary horseradish peroxidase-conjugated antibodies directed against immunoglobulin G of the relevant species (rabbit: sc-2030, mouse: sc-2031 and goat: sc-2020; Santa Cruz Biotechnology) were used at a dilution of 1/4000.

Western blots for reprobing were stripped of antibody by agitation at 50°C for 1 h in 2% SDS, 62.5 mM Tris (pH 6.8), 100 mM 2-mercaptoethanol and washed twice in PBST for 10 min.

Densitometry of autoradiograph bands was carried out using Kodak 1D or MACBAS software on imported images. As such, comparisons between samples are relative, but the values are not necessarily linear.

Electrophoretic mobility shift assay
Nuclear extracts were prepared from tissues by polytron disruption of snap-frozen samples using NE buffer [20 mM HEPES (pH 7.9), 0.4 M NaCl, 1 mM EDTA (pH 8.0) and 1 mM DTT with protease inhibitors: 2% (v/v) aprotinin, 1 mM PMSF, 1% (v/v) vanadate and 1x Sigma cocktail] and incubated for 10 min on ice. Following centrifugation, 16 000g for 15 min at 4°C, the protein concentration of the supernatant was determined using the Bradford assay. Binding reactions were performed for 30 min on ice [or for serum response element (SRE): 15 min on ice followed by 15 min at room temperature] using 10 µg (for SRE and NF-{kappa}B) or 5 µg [for Sp1 and TPA response element (TRE)] of nuclear extract and 0.2–1 ng probe in binding buffer [SRE: 10 mM HEPES (pH 7.9), 2 mM EDTA (pH 8.0), 1 M NaCl, 1 mg/ml bovine serum albumin and 10% (v/v) glycerol; Sp1: 30 mM HEPES (pH 7.9), 2 mM EDTA (pH 8.0), 1 M NaCl, 1 mg/ml bovine serum albumin and 10% (v/v) glycerol; TRE/NF-{kappa}B: 100 mM Tris (pH 7.5), 500 mM NaCl, 10 mM DTT, 10 mM EDTA (pH 8.0) and 50% (v/v) glycerol]. The probes were end labelled, double-stranded oligonucleotides incorporating the factor binding site (forward and reverse single-stranded oligos from Sigma Genosys, UK) using either T4 Polynucleotide kinase and {gamma}32P dATP for SRE or Klenow and {alpha}32P dCTP for Sp1, TRE and NF-{kappa}B incubated in the appropriate buffers and column purified (NICKTM Amersham, USA). To one aliquot of each sample, excess (SRE: 50x; Sp1: 100x; TRE: 200x and NF-{kappa}B: 50x) unlabelled oligonucleotide (competitor) was added for 10 min on ice prior to incubation with the probe. The reaction products were electrophoresed through a 6% non-denaturing acrylamide gel, which after completion was dried and exposed to film.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 References
 
LMP1CAO activates ERK but leads to down-regulation of Akt
Transgenic L2LMP1CAO mice (7) were used in all of the studies described herein and compared with non-transgenic sibling controls (NSC), thus always matching age and strain. We have shown previously that through the advancing LMP1-induced transgenic phenotype, levels of total EGFR declined compared with NSC, whereas phosphorylated EGFR and fragments thereof increased, suggestive of increasing EGFR activation and cleavage with age (7). We next explored the status of pathways downstream of EGFR in this system. Levels of total MEK1/2 and total ERK1/2 in the ear tissue were similar between L2LMP1CAO transgenic and NSC mice and remained relatively constant with advancing phenotype and age (Figure 1A). The activation status of MEK (by phosphorylation), although a little variable, was not consistently enhanced in L2LMP1CAO samples compared with NSC samples. In contrast, higher levels of the activated, phosphorylated ERK1 and ERK2 were detected in L2LMP1CAO samples compared with controls. In both L2LMP1CAO and NSC samples, ERK1/2 showed increasing activation with age (Figure 1A), but activation was higher in the transgenic samples at all stages (quantified by stage in Figure 4). Thus, LMP1 leads to ERK1 and 2 activation in the disease process, but this does not reflect the activation status of MEK.


Figure 1
View larger version (43K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. Protein extracts (80 µg in A, 100µg in B and C) from ears of transgene-positive animals of line L2LMP1CAO.117 of phenotypic stages 1, 2, 3 and 5 (st1, st2, st3 and st5) as indicated, compared with NSC (c1, c2, c3 and c5), from biological replicates (different mice) in each category, were separated by 10% SDS–polyacrylamide gel electrophoresis and western blotted. (A) The blot was probed with anti-sera directed against phosphorylated JNK1/2/3, total JNK1/2/3 (46 D and 54 kDa, respectively), phosphorylated MEK1/2, total MEK1/2 (45 kDa), phophorylated ERK1/2, total ERK1/2 (42 and 44 kDa), phosphorylated (ser473) Akt, total Akt (60 kDa) and GAPDH (36 kDa) as indicated. Ponceau staining of the blot indicative of protein loading is also shown for an abundant protein at 70 kDa. Note: MEK1/2 is detected as a doublet in this blot (arrows) and the phospho-MEK band equates with the top band of this doublet. (B) Blots were probed with anti-sera directed against phosphorylated p38, pan-Fos (recognising c-Fos, Fra-1, Fra-2 and FosB), Fra-1 specific, Fra-2 specific, c-Jun and JunB, as indicated. (C) The blot was probed with anti-sera directed against phosphorylated p38 (also recognising a 45 kDa unidentified protein in transgenic samples, possibly a related MAPK), total p38 (also recognising a 39 kDa unidentified protein in transgenic samples, possibly another of the p38 isoforms) and GAPDH.

 
EGF/ErbB receptors can also activate the PI3K–Akt cascade through binding (or docking via Gab1) of the p85 subunit of PI3K (48) and LMP1 has been shown to activate PI3K in HeLa cells (24). Activation of Akt was examined in the transgenic tissues by the phosphorylation status of Akt upon serine (ser473). Whereas the levels of total Akt in the samples from NSC mice remained constant with increasing age (c1, c2, c5, Figure 1A), the transgenic tissues showed a dramatic decline from st1 to st5 samples. Similarly, a constant basal activation of phospho-Akt was detected in normal tissue, whereas in the transgenic samples, the reduction in levels of phospho-Akt with age and stage reflected the total levels.

LMP1CAO activates p38 and differentially regulates JNK members
In order to investigate whether LMP1CAO leads to the activation of p38 MAPK and JNK in this system, levels of the phosphorylated (activated) products were examined. Activation of JNK1 (p46) and JNK2/3 [p54: probably JNK2 in this context as JNK3 shows restricted expression to brain, heart and testes (49)] showed a complex inverse relationship. Levels of total protein remained constant at all ages and stages in transgenic and NSC samples (Figure 1A). Activation of JNK1 was not apparent in NSC samples, whereas low levels of the phosphorylated form could be detected in early stage transgenic samples and clearly at st5 (Figure 1A). Conversely, phosphorylated JNK2 was only detectable in tissues from older NSC mice (c5) in contrast to detection in younger transgenic samples (st1 and st2) but not at all in the advanced stage transgenic samples (st5) (Figure 1A). Thus, it seems that in normal mice, only JNK2 becomes gradually activated with age in this tissue. With the expression of LMP1 and the progressive phenotype, JNK2 may be initially activated, but over time, JNK1 becomes activated instead.

Transgenic samples showed an induction in the level of p38 MAPK protein (4-fold at stage 1 and 1.6-fold at stage 5) as well as a strong activation of p38 compared with NSC with the phosphorylated form being barely or not detectable in NSC samples (Figure 1B and C).

LMP1CAO expression induces particular members of the AP1 family
Activated ERK1/2 stimulates several downstream targets including p90RSK, C-Myc and Elk-1. Elk-1 (a member of the Ets family of transcription factors) forms a ternary complex with serum response factor and binding to the SRE leads to transcriptional activation of linked genes such as c-fos, fosB and jun-B. Therefore, steady-state protein levels of members of the fos and jun gene families were examined. Levels of FosB, Fra-1 and JunB were considerably elevated in L2LMP1CAO transgenic samples from the early st1 onwards compared with NSC (Figure 1B). In contrast, c-Fos, c-Jun and Fra-2 showed no induction.

LMP1CAO activates serum response factor, AP1 and NF-{kappa}B
In order to explore this further, binding of factors to the SRE was examined (using Sp1 binding as indicator of sample quality and quantity). Extracts from L2LMP1CAO tissues of st2/3 revealed specific binding activity to an SRE oligo, whereas none was evident in NSC samples (Figure 2A). Specific binding to the AP1/TRE was readily detected in the L2LMP1CAO transgenic tissues and barely detected in control samples (Figure 2A, compare samples 359 and 375 in relation to Sp1 binding). Similarly, specific binding to the NF-{kappa}B binding site was detected only in transgene-positive samples. Supershift analysis of representative samples revealed that the complex binding to the NF-{kappa}B site in the transgenic samples contained the p65 (RelA) NF-{kappa}B form and p50 may be a minor component (Figure 2B).


Figure 2
View larger version (68K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2. Protein extracts (5 µg for TRE and Sp1 and 10 µg for SRE and NF-{kappa}B) from individual ear tissue samples (mouse identity number given) from transgene-positive (+) and NCS (–) mice were incubated with a labelled oligonucleotide (as indicated), with an excess or without, competing unlabelled oligonucleotide (TRE: 200x, SRE: 50x, Sp1: 100x, NF-{kappa}B: 50x) and separated by 6% polyacrylamide gel electrophoresis. Extracts from the cell lines NIH3T3 (N) or A431 were used as positive controls and labelled oligonucleotide without extract (o) is shown. (A) Specific bands inhibited by the addition of competitor are indicated with arrows. Note, NSC extract 351 shows similar integrity with respect to Sp1 binding as do transgene-positive extracts 347 and 349 (used in the SRE assay) and NSC extract 375 shows similar integrity with respect to control Sp1 binding as does transgene-positive extract 359 (used additionally in the TRE assay). (B) Protein extracts from transgene-positive (347) and NSC (375) ear tissue samples were used. Addition of antibody to p50 or to p65 is indicated (+). Specific bands (a and b) and non-specific band (c) are indicated. Note, addition of anti-p65 shifts band (a) to a more slowly migrating complex (arrow) and addition of anti-p50 does not lead to complete shifting of bands (a) or (b) but does reveal a more slowly migrating complex (arrow).

 
Absence of TGF{alpha} does not improve the phenotype of L2LMP1CAO transgenic mice
Given that LMP1 leads to increased steady-state levels of TGF{alpha} while ERK and serum response factor are activated, combined with the observation that the LMP1-induced phenotype strongly resembles that of TGF{alpha}-over-expressing mice, the hypothesis that LMP1 mediates this phenotype through TGF{alpha} was explored. The approach taken was of complete removal of TGF{alpha} expression through cross-breeding L2LMP1CAO transgenic mice with TGF{alpha}-null animals, generating LMP1CAO transgenic, TGF{alpha}-null and heterozygous mice (LMP1/TGF{alpha}-null, LMP1/TGF{alpha}-het, respectively). Papillomas develop on the dorsal skin of L2LMP1CAO line 117 mice in the FVB strain but not the C57Bl/6 strain (7). Mice resulting from the TGF{alpha} cross, siblings in the mixed strain background (FVB/C57Bl/6/129) did not develop dorsal papillomas (as expected in this strain). TGF{alpha}-null mice show crinkly whiskers from birth and develop wavy fur (46). This phenotype was evident and not altered in mice also expressing LMP1.

A cohort of LMP1/TGF{alpha}-null, LMP1/TGF{alpha}-het, LMP1/WT (TGF{alpha} wild type) and 36 NSC were monitored for ear epidermal phenotype development (Figure 3). In order to compare the development of the phenotype, it was categorised every week for each mouse according to the phenotypic stages as described above and previously (7). The NSC mice did not develop any ear phenotype as expected. LMP1/TGF{alpha}-null mice advanced from a st1 ear phenotype to a st2 between 4 and 9 weeks of age, whereas LMP1/WT showed this advance between 7 and 9 weeks. In both groups, progression from st2 to st3 was completed by 15 weeks of age. LMP1/TGF{alpha}-null mice started showing a st4 phenotype from 13 weeks of age, with 50% at this stage by 15 weeks. In contrast, advancement to st4 did not start in LMP1/WT mice until 20 weeks (Figure 3). Mice in the LMP1/TGF{alpha}-null group began to show st5 phenotype at 20 weeks, whereas LMP1/WT mice did not progress to st5 in the time frame of this study. LMP1/TGF{alpha}-het mice showed the same pattern of phenotype development as LMP1/WT mice (not shown). While the study is subjective in nature, it is nevertheless clear that the LMP1-induced phenotype is not negated or even alleviated in the TGF{alpha}-null background. In contrast, the data are suggestive that the phenotype advances more rapidly in the TGF{alpha}-null background. Consistent with this, proliferating cell nuclear antigen staining was just as abundant in LMP1/TGF{alpha}-null ear epidermis (not shown) as it was in a WT background (7). This is contrary to what would be expected if the hypothesis was correct that the LMP1-induced ear phenotype is mediated via up-regulation of TGF{alpha} alone. Conversely, the data suggest that the phenotype may be tempered by TGF{alpha} induction.


Figure 3
View larger version (53K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3. A cohort of 8 LMP1/TGF{alpha}-null (above), 9 LMP1/TGF{alpha}-het (not shown), seven LMP1/WT (below) siblings were monitored for ear epidermal phenotype development from 4 to 26 weeks of age. The advancement of the phenotype from st1 to st5 over time is shown as the proportion of mice in each genetic group.

 
EGFR is down-regulated more slowly in the absence of TGF{alpha}
In order to explore if EGFR is indirectly activated by LMP1 in the absence of its ligand TGF{alpha}, expression and activation of EGFR were examined in tissues from LMP1/TGF{alpha}-null mice. The constitutive signalling from LMP1 in conjunction with tissue homeostatic responses results in a dynamic equilibrium and this was investigated through a detailed analysis of tissues at different stages using three biological replicates in each group to allow for sample to sample variation (Figure 4). As shown previously, levels of total EGFR are reduced in the ear tissue of L2LMP1CAO mice compared with the age-matched controls in a WT background, such that it is barely detectable at st2 and not at st5. However, in a TGF{alpha}-null background, higher EGFR levels are observed at st1 and st2 compared with a WT background, but this declines by st5 to below NSC levels (Figure 4). Thus, in the absence of TGF{alpha}, LMP1 initially leads to increased EGFR levels delaying the subsequent down-regulation of EGFR. With the down-regulation of EGFR in the L2LMP1CAO samples by stage 5, in a reciprocal fashion, phosphorylated EGFR antibody-reactive fragments appear (a 48 kDa product is detected by the anti-phospho-tyr845 EGFR antibody, as noted previously (7), tyrosine 845 being indicative of full receptor activation), which are evident in both TGF{alpha}-null and WT backgrounds (Figure 4C), but not seen in the NSC samples.


Figure 4
View larger version (91K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4. Protein extracts (80 µg) from ears of transgene-positive animals of line L2LMP1CAO.117 of phenotypic stages 1, 2 and 5 (st1, st2 and st5) and NSC (c1, c2 and c5) shown in (A), (B) and (C), respectively, from three biological replicates in each category were separated by either 7.5 or 10% SDS–polyacrylamide gel electrophoresis and western blotted. Mice were either in a WT background or null for TGF{alpha} as indicated. The blots were sequentially probed (as indicated) with anti-sera directed against total EGFR, phospho-(tyr845)-EGFR, phospho-(ser473)-Akt, total Akt, phosphorylated MEK1/2, total MEK1/2, phophorylated ERK1/2, total ERK1/2 and ß-tubulin (for 7.5% gels—upper panels B and C) or GAPDH (10% gels) as loading controls. The average relative band intensity for each triplicate (normalised against the appropriate ß-tubulin or GAPDH) is indicated below the blot, adjusted to NSC in the WT background = 1, where possible. Note, where bands were barely detectable, relative levels could not be assessed; where two bands are relevant (e.g. ERK1/2), these were summed.

 
The consequences of this are that both MEK and ERK activation are considerably higher in LMP1/TGF{alpha}-null mice compared with LMP1/WT mice at st1. Although the difference has gone by st 2, the increased activation of these MAPKs at an early stage are consistent with a more rapidly advancing L2LMP1CAO phenotype in TGF{alpha}-null background compared with a WT background.

Levels of total and phosphorylated Akt decline with increasing age and stage in the L2LMP1CAO transgenic samples compared with NSC in a WT background as noted previously (Figures 1A and 4C). In the TGF{alpha}-null background, total Akt levels are lower than WT samples, but at an early age (Figure 4B), Akt activation is still evident in the L2LMP1CAO samples (mirroring the EGFR levels). At the later st5, no Akt can be detected in the TGF{alpha}-null samples.

The activation of EGFR and the MAPK pathway by LMP1 in the absence of TGF{alpha} suggested that other ligands might also be induced by LMP1. Whereas amphiregulin and EGF expression could not be detected (not shown), HB-EGF showed considerable induction and EPR was detectably induced in transgenic tissues, independently of TGF{alpha} status (Figure 5A). In particular, the cell-associated C-terminal remnant of HB-EGF (41) showed constitutive induction in transgenic tissues of all stages, whereas the mature form was more readily detected in the more advanced stages. In conjunction with this, the MAPK-regulated gene JunB shows constitutive induction by LMP1 also in the TGF{alpha}-null background (Figure 5B).


Figure 5
View larger version (78K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5. Protein extracts from ears of transgene-positive animals of line L2LMP1CAO.117 of phenotypic stages 1–5 (st1–st5) compared with NSC (c1–c4) as indicated were separated by SDS–polyacrylamide gel electrophoresis and western blotted. Mice were either in a WT background or null for TGF{alpha} as indicated. (A) 15% polyacrylamide gel electrophoresis. The blot was probed with an antibody directed against HB-EGF (top) and stripped and reprobed with an antibody against EPR (below). Note, (top) a non-specific band at 30 kDa is evident in all lanes. Mature HB-EGF is 14–22 kDa (depending upon the degree of glycosylation), whereas the cleaved C-terminal cell-associated remnant is ~5 kDa (arrows). EPR migrates between 5 and 19 kDa (glycosylation dependent) and is indicated (arrow). Protein marker sizes are given in kilo Dalton. (B) 10% polyacrylamide gel electrophoresis. The blot was successively probed with anti-sera directed against cleaved, activated caspase-3 (cl-caspase-3), total caspase-3 and JunB.

 
Apoptosis accompanies the LMP1 phenotype
Pathways known to promote proliferation have been activated in the L2LMP1CAO phenotypic tissues and hyperplasia and increased proliferating cell nuclear antigen staining result (7). Oncogenes can induce apoptosis or senescence in premalignant tissue (50); therefore, the levels of total caspase-3 and its active, cleaved form were examined. Both forms were found to be elevated in the L2LMP1CAO transgenic ears compared with NSC samples, increasing with advancing stage in WT and TGF{alpha}-null backgrounds (Figure 5B). This suggests that apoptosis is ongoing in the transgenic tissues, which may underlie the degenerative nature of the phenotype.


   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
References
 
We have used a model system to explore the action of LMP1 of the Cao strain of EBV in the early stages of carcinogenesis, how this proceeds in vivo and how it is modulated in the absence of one of the key receptor ligands induced by LMP1 (summarised in Figure 6). The LMP1-induced phenotype steadily worsens with age in a predictable manner in this model, each stage reflecting a steady-state composite of signalling cascades which result in hyperplasia as well as apoptosis, leading to an advancing degenerative phenotype from which progression to papilloma/keratoacanthoma and carcinoma can emerge.


Figure 6
View larger version (33K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6. The signalling proteins p38 MAPK, JNK1/2, NF{kappa}B and ERK1/2 are activated by LMP1 in the phenotypic transgenic skin tissue leading to activation of the transcription factors AP1, NF{kappa}B (incorporating the p65 component and possibly p50) and serum response factor (upper panel). (The Janus kinase/signal transducer and activation of transcription pathway, known to be activated by LMP1, has not been examined in this system.) Consequently, AP1 family members FosB, Fra-1 and JunB and EGFR ligands TGF{alpha}, HB-EGF and EPR become constitutively induced. (HB-EGF-C may signal independently of EGFR activation through promyelocytic leukaemia zinc finger (41), dotted arrow). As a result, EGFR becomes down-regulated (lower panel), possibly through expressional repression and/or by proteolytic cleavage to the detected phosphorylated fragments (such as the 48 kDa fragment). MEK shows little activation and Akt is down-regulated, possibly via the EGFR fragments. In the absence of TGF{alpha} (in the null mice), initially EGFR is up-regulated in the LMP1 transgenic tissue (st1) and MEK, ERK and Akt become activated. Over time, but still prior to neoplastic conversion (by st5), EGFR becomes down-regulated through LMP1 action, as in the WT background (dashed arrow).

 
It has recently become established that oncogene-induced senescence or apoptosis precedes malignant neoplasia (50), which is abundantly apparent in this model. Indeed by st5, if malignant progression has not occurred, the tissue degeneration accompanying the severe hyperplasia leads to the ear pinnae becoming mere knubs.

Several oncogenic viruses lead to the stimulation of EGFR signalling in their action to promote cellular proliferation and EBV is no exception, mediated by the action of LMP1 in epithelial cells. In both NPC biopsies and cultured epithelial cells, LMP1 expression has been correlated with up-regulation of EGFR and it was recently shown that LMP1 affects the nuclear accumulation of EGFR and thereby transcriptional regulation of certain cell cycle genes (40). In this study and previously, we have found that LMP1 induces TGF{alpha} expression and leads to a decrease in steady-state EGFR levels compared with age-matched controls, with the appearance of phospho-EGFR and proteolytic fragments correlating with a worsening phenotype (7). Surprisingly, in the absence of TGF{alpha}, the worsening phenotype was not ameliorated but even accelerated, correlating with the initially higher EGFR levels in the LMP1 transgenic tissues compared with non-transgenic. This could suggest that LMP1 induces EGFR expression in vivo (as has been observed in cell lines and tumours), but in pre-neoplastic tissue, the stimulated signalling through TGF{alpha} promotes a regulatory feedback loop to down-regulate EGFR or alternatively increases turnover. Lower levels of EGFR were observed in the phenotypic skin of TGF{alpha}-over-expressing transgenic mice (51) and thus the reverse might be expected for TGF{alpha}-null mice. However, through time, even in the TGF{alpha}-null background, EGFR levels reduce in the LMP1 transgenic samples compared with non-transgenic (st5). Nevertheless, the detection of phosphorylated, cleaved EGFR fragments are suggestive of active EGFR degradation as might be predicted from the constitutive induction of EGFR ligands in the LMP1 transgenic samples (52). A working model for these observations would be that excessive signalling through EGFR via up-regulation of several ligands (and possibly initially of the receptor) leads to a rapid and sustained activation and degradation of EGFR (Figure 6). In the absence of one ligand, the turnover is slower to reach equilibrium. This hypothesis would imply that a similar result would be obtained in HB-EGF or EPR-null backgrounds and that there is no special role for TGF{alpha}. Alternatively, there could be a unique role played by TGF{alpha} in the modulation of its receptor. Indeed, stimulation of EGFR by different ligands has been shown to lead to activation of subtly different pathways and to different degrees (53). A prediction of this model is that a key hurdle during carcinogenic progression would be to nullify the negative feedback loops existent on EGFR expression and activity, such that ligand-mediated activation becomes a constitutive, positive growth stimulus. Consistent with this idea is the observation that increased serum levels of TGF{alpha} in NPC patients were correlated with recurrence of disease and inversely with survival (54).

Akt activation reflects the absolute levels of EGFR through the different stages in transgenic versus NSC samples, being initially high in the LMP1/TGF{alpha}-null samples (st1) and then declining in all of the LMP1 samples. This suggests that the EGFR phosphorylated fragments detected in LMP1-positive samples at st5 are not involved in Akt activation and reduction in levels of EGFR leads to loss of Akt signalling and possibly overall Akt levels. This in turn would lead to a loss of protection from apoptosis, which is observed both in the degenerative phenotype and progressive activation of caspase-3. Moreover, the process could be cyclic and accelerate cell death, as caspase-dependent cleavage of EGFR/ErbB proteins could produce apoptosis-promoting fragments. Tyrosine kinase domain fragments of ErbB were found to strongly induce apoptosis (an amino acid 651-803 fragment was found to be effective for EGFR) (55) and we detected phosphorylated fragments (48 kDa for Tyr 845) at st 5 in both WT and TGF{alpha}-null backgrounds in the LMP1 transgenic samples.

Activation of the stress associated MAPK p38 along with caspase-dependent apoptosis and activation of Akt and cell survival have been shown previously to display a reciprocal correlation (56,57). Consistent with this, we observed activation of p38 in the transgenic tissue. Furthermore, phosphorylation of EGFR by the p38 pathway has been shown to lead to receptor internalisation into endosomes and thereby reduce access to the receptor by growth factors (58). Whether this mechanism is acting in the transgenic tissue is a subject for further investigation.

If at all, MEK is only marginally activated in the transgenic tissue (compared with NSC), but in the absence of TGF{alpha}, like Akt, MEK is clearly activated in response to LMP1-induced signalling at an early stage. At st1, ERK activation mirrors MEK, being high in the LMP1/TGF{alpha}-null samples. However, unlike MEK, through time ERK activation continues to be higher in the transgenic samples than NSC in a WT background, despite decreasing EGFR levels (st5). This could suggest that the turnover and degradation of EGFR maintains a signal to ERK that is not via MEK or that ERK signalling is independent of EGFR at this stage [ERK activation independent of MEK has been noted in other cell types, via Rac/cdc2 or the protein kinase C pathway (59,60)].

In addition to TGF{alpha}, we show that LMP1 induces the expression of other EGFR ligands, specifically HB-EGF and to lesser degree EPR. The cell-associated cleaved remnant of HB-EGF (HB-EGF-C), which has been shown to function as an intracellular signal independently of EGFR, by binding to a transcriptional repressor (41), is strongly induced in all stages in the transgenic tissues, providing a novel route by which LMP1 could influence gene expression. Like the LMP1-induced expression of TGF{alpha}, the induced level of HB-EGF-C is constant across the phenotypic stages, indicating that this is directly mediated by LMP1 and not modulated over time in response to the changes in the signalling milieu.

Similarly, expression of specific members of the AP1 gene family, namely, JunB, Fra-1 and FosB, show a constant induction across the phenotypic stages, again suggestive that this results directly from LMP1 signalling and is not influenced over time. It is interesting to note that c-Jun, c-Fos and Fra-2, while expressed, show no induction in the transgenic tissue and thus induction of AP1 family members is specifically targeted.

JNK1 and JNK2 show induced activation in the early stage transgenic ear tissue, but whereas JNK2 activation declines to below levels seen in NSC tissues in the later stages, JNK1 activation increases. JNK2 has been shown to be critical in papilloma induction during chemical carcinogenesis (61,62). Thus, the observed stimulation of JNK2 by LMP1 in these studies is consistent with our earlier observations that LMP1 leads to an increase in papilloma load upon chemical carcinogen treatment (63).

Taken together, the data show that LMP1CAO induces proliferative pathways including MAPK and NF-{kappa}B cascades, with stable induction of several EGFR ligands and members of the AP1 family. As a consequence, apoptotic and stress reaction pathways become increasingly activated resulting in a deteriorating phenotype of combined hyperplasia and concomitant tissue degeneration. We have demonstrated previously that p16INK4a becomes induced in the later stages in the transgenic tissue, but is not expressed in carcinomas arising in LMP1-expressing mice (7,47). Thus, loss of p16INK4a expression is likely to be a progression factor in this model and consistent with what is observed in NPC. A further critical step in the progression to carcinoma would be the inhibition of apoptosis, which is possibly being mediated in part through EGFR degradation products and the p38 MAPK pathway. A further hypothesis predicted from this model is that during carcinogenic progression, there must be a switch that releases EGFR induction and activation by its ligands from homeostatic, negative regulatory loops, allowing constitutive activation of the positive growth pathways.

This disease model allows an investigation into the pre-neoplastic stages of carcinogenesis induced by a widely acting viral oncoprotein and reveals how the tissue responds in homeostatic control. In particular, these studies demonstrate some of the complexities in EGFR signalling. The model can be used to reveal what factors must be overcome during progression to carcinoma and thereby provide insight into how to revert the process. Furthermore, it demonstrates that disruption of a particular pathway (in this case removal of TGF{alpha}) might shift the homeostatic response (while in place) in an unexpected way, thus providing a cautionary note in the use of therapeutic agents that target single pathway points. Therefore, if the homeostatic loops are intact in a tissue, use of such a therapeutic approach would be deleterious; however, if negative feedback loops are disrupted during carcinogenesis (as hypothesised), then therapeutic approaches through blockade of EGFR signalling could be effective. The model reveals the central role played by EGFR and its ligands in EBV-associated, LMP1-expressing carcinoma, with particular significance for the understanding and treatment of NPC.


   Acknowledgments
 
C.C. was supported by a faculty of Biomedical and Life Sciences scholarship and an Overseas Research Student award. The work was partly funded by the Wellcome Trust (069113/Z/02). We thank Mark Drotar, Donald Campbell and Elizabeth Hill for assistance with tissue collection and genetic screening and Colin Chapman for assistance with the animal studies.

Conflict of Interest Statement: None declared.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 

  1. Wang D, et al. An EBV membrane protein expressed in immortalized lymphocytes transforms established rodent cells. Cell (1985) 43:831–840.[CrossRef][ISI][Medline]
  2. Baichwal VR, et al. Transformation of Balb 3T3 cells by the BNLF-1 gene of Epstein-Barr virus. Oncogene (1988) 2:461–467.[ISI][Medline]
  3. Dawson CW, et al. Epstein-Barr virus latent membrane protein inhibits human epithelial cell differentiation. Nature (1990) 344:777–780.[CrossRef][Medline]
  4. Fåhraeus R, et al. Morphological transformation of human keratinocytes expressing the LMP gene of Epstein-Barr virus. Nature (1990) 345:447–449.[CrossRef][Medline]
  5. Kaye KM, et al. Epstein-Barr virus latent membrane protein 1 is essential for B-lymphocyte growth transformation. Proc. Natl Acad. Sci. USA (1993) 90:9150–9154.[Abstract/Free Full Text]
  6. Wilson JB, et al. Expression of the BNLF-1 oncogene of Epstein-Barr virus in the skin of transgenic mice induces hyperplasia and aberrant expression of keratin 6. Cell (1990) 61:1315–1327.[CrossRef][ISI][Medline]
  7. Stevenson D, et al. Epstein-Barr virus latent membrane protein 1 (CAO) up-regulates VEGF and TGF{alpha} concomitant with hyperlasia, with subsequent up-regulation of p16 and MMP9. Cancer Res. (2005) 65:8826–8835.[Abstract/Free Full Text]
  8. Uchida J, et al. Mimicry of CD40 signals by Epstein-Barr virus LMP1 in B lymphocyte responses. Science (1999) 286:300–303.[Abstract/Free Full Text]
  9. Mitchell T, et al. Stimulation of NF-{kappa}B-mediated transcription by mutant derivatives of the latent membrane protein of Epstein-Barr virus. J. Virol. (1995) 69:2968–2976.[Abstract]
  10. Huen DS, et al. The Epstein-Barr virus latent membrane protein-1 (LMP1) mediates activation of NF-{kappa}B and cell surface phenotype via two effector regions in its carboxy-terminal cytoplasmic domain. Oncogene (1995) 10:549–560.[ISI][Medline]
  11. Devergne O, et al. Association of TRAF1, TRAF2, and TRAF3 with an Epstein-Barr virus LMP1 domain important for B-lymphocyte transformation: role in NF-{kappa}B activation. Mol. Cell Biol. (1996) 16:7098–7108.[Abstract]
  12. Sandberg M, et al. Characterization of LMP-1's association with TRAF1, TRAF2, and TRAF3. J. Virol. (1997) 71:4649–4656.[Abstract]
  13. Izumi KM, et al. The Epstein-Barr virus oncoprotein latent membrane protein 1 engages the tumor necrosis factor receptor-associated proteins TRADD and receptor-interacting protein (RIP) but does not induce apoptosis or require RIP for NF-{kappa}B activation. Mol. Cell Biol. (1999) 19:5759–5767.[Abstract/Free Full Text]
  14. Luftig M, et al. Epstein-Barr virus latent infection membrane protein 1 TRAF-binding site induces NIK/IKK alpha-dependent noncanonical NF-kappaB activation. Proc. Natl Acad. Sci. USA (2004) 101:141–146.[Abstract/Free Full Text]
  15. Thornburg NJ, et al. LMP1 signaling and activation of NF-kappaB in LMP1 transgenic mice. Oncogene (2006) 25:288–297.[ISI][Medline]
  16. Eliopoulos AG, et al. Activation of the cJun N-terminal kinase (JNK) pathway by the Epstein-Barr virus-encoded latent membrane protein 1 (LMP1). Oncogene (1998) 16:1731–1742.[CrossRef][ISI][Medline]
  17. Kieser A, et al. Epstein-Barr virus latent membrane protein-1 triggers AP-1 activity via the c-Jun N-terminal kinase cascade. EMBO J. (1997) 16:6478–6485.[CrossRef][ISI][Medline]
  18. Wan J, et al. Elucidation of the c-Jun N-terminal kinase pathway mediated by Estein-Barr virus-encoded latent membrane protein 1. Mol. Cell Biol. (2004) 24:192–199.[Abstract/Free Full Text]
  19. Eliopoulos AG, et al. Activation of the p38 mitogen-activated protein kinase pathway by Epstein-Barr virus-encoded latent membrane protein 1 coregulates interleukin-6 and interleukin-8 production. J. Biol. Chem. (1999) 274:16085–16096.[Abstract/Free Full Text]
  20. Vockerodt M, et al. The Epstein-Barr virus oncoprotein latent membrane protein 1 induces expression of the chemokine IP-10: importance of mRNA half-life regulation. Int. J. Cancer (2005) 114:598–605.[CrossRef][ISI][Medline]
  21. Gires O, et al. Latent membrane protein 1 of Epstein-Barr virus interacts with JAK3 and activates STAT proteins. EMBO J. (1999) 18:3064–3073.[CrossRef][ISI][Medline]
  22. Zhang L, et al. Multiple signal transducers and activators of transcription are induced by EBV LMP-1. Virology (2004) 323:141–152.[CrossRef][ISI][Medline]
  23. Chen H, et al. A positive autoregulatory loop of LMP1 expression and STAT activation in epithelial cells latently infected with Epstein-Barr virus. J. Virol. (2003) 77:4139–4148.[Abstract/Free Full Text]
  24. Dawson CW, et al. Epstein-Barr virus latent membrane protein 1 (LMP1) activates the phosphatidylinositol 3-kinase/Akt pathway to promote cell survival and induce actin filament remodeling. J. Biol. Chem. (2003) 278:3694–3704.[Abstract/Free Full Text]
  25. Roberts ML, et al. Activation of a ras-MAPK-dependent pathway by Epstein-Barr virus latent membrane protein 1 is essential for cellular transformation. Virology (1998) 240:93–99.[CrossRef][ISI][Medline]
  26. Wakisaka N, et al. Epstein-Barr virus latent membrane protein 1 induces synthesis of hypoxia-inducible factor 1 alpha. Mol. Cell Biol. (2004) 24:5223–5234.[Abstract/Free Full Text]
  27. Hu LF, et al. Isolation and sequencing of the Epstein-Barr virus BNLF-1 gene (LMP1) from a Chinese nasopharyngeal carcinoma. J. Gen. Virol. (1991) 72:399–409.[Abstract/Free Full Text]
  28. Dawson CW, et al. Identification of functional differences between prototype Epstein-Barr virus-encoded LMP1 and a nasopharyngeal carcinoma-derived LMP1 in human epithelial cells. Virology (2000) 272:204–217.[CrossRef][ISI][Medline]
  29. Blake SM, et al. The transmembrane domains of the EBV-encoded latent membrane protein 1 (LMP1) variant CAO regulate enhanced signalling activity. Virology (2001) 282:278–287.[CrossRef][ISI][Medline]
  30. Miller WE, et al. The NPC derived C15 LMP1 protein confers enhanced activation of NF- {kappa}B and induction of the EGFR in epithelial cells. Oncogene (1998) 16:1869–1877.[CrossRef][ISI][Medline]
  31. Johnson RJ, et al. The 30-base-pair deletion in Chinese variants of the Epstein-Barr virus LMP1 gene is not the major effector of functional differences between variant LMP1 genes in human lymphocytes. J. Virol. (1998) 72:4038–4048.[Abstract/Free Full Text]
  32. Miller WE, et al. Epstein-Barr virus LMP1 induction of the epidermal growth factor receptor is mediated through a TRAF signaling pathway distinct from NF-{kappa}B activation. J. Virol. (1997) 71:586–594.[Abstract]
  33. Tao YG, et al. Nuclear factor kappa B (NF-{kappa}B) dependent modulation of Epstein-Barr virus latent membrane protein 1 (LMP1) in epidermal growth factor receptor (EGFR) promoter activity. Virus Res. (2004) 104:61–70.[CrossRef][ISI][Medline]
  34. Zheng X, et al. Expression of Ki67 antigen, epidermal growth factor receptor and Epstein-Barr virus-encoded latent membrane protein (LMP1) in nasopharyngeal carcinoma. Eur. J. Cancer B Oral Oncol. (1994) 30B:290–295.[CrossRef]
  35. Sheen TS, et al. Epstein-Barr virus-encoded latent membrane protein 1 co-expresses with epidermal growth factor receptor in nasopharyngeal carcinoma. Jpn. J. Cancer Res. (1999) 90:1285–1292.[CrossRef][ISI][Medline]
  36. Jorissen RN, et al. Epidermal growth factor receptor: mechanisms of activation and signalling. Exp. Cell Res. (2003) 284:31–53.[CrossRef][ISI][Medline]
  37. Lin SY, et al. Nuclear localization of EGF receptor and its potential new role as a transcription factor. Nat. Cell Biol. (2001) 3:802–808.[CrossRef][ISI][Medline]
  38. Offterdinger M, et al. c-erbB-3: a nuclear protein in mammary epithelial cells. J. Cell Biol. (2002) 157:929–939.[Abstract/Free Full Text]
  39. Williams CC, et al. The ERBB4/HER4 receptor tyrosine kinase regulates gene expression by functioning as a STAT5A nuclear chaperone. J. Cell Biol. (2004) 167:469–478.[Abstract/Free Full Text]
  40. Tao Y, et al. Nuclear accumulation of epidermal growth factor receptor and acceleration of G1/S stage by Epstein-Barr-encoded oncoprotein latent membrane protein 1. Exp. Cell Res. (2005) 303:240–251.[CrossRef][ISI][Medline]
  41. Nanba D, et al. Proteolytic release of the carboxy-terminal fragment of proHB-EGF causes nuclear export of PLZF. J. Cell Biol. (2003) 163:489–502.[Abstract/Free Full Text]
  42. Vassar R, et al. Transgenic mice provide new insights into the role of TGF-{alpha} during epidermal development and differentiation. Genes Dev. (1991) 5:714–727.[Abstract/Free Full Text]
  43. Wang XJ, et al. Epidermal expression of transforming growth factor-alpha in transgenic mice: induction of spontaneous and 12-O-tetradecanoylphorbol-13-acetate-induced papillomas via a mechanism independent of Ha-ras activation or overexpression. Mol. Carcinog. (1994) 10:15–22.[ISI][Medline]
  44. Jhappan C, et al. Transgenic mice provide genetic evidence that transforming growth factor-{alpha} promotes skin tumorigenesis via H-ras-dependent and H-ras-independent pathways. Cell Growth Differ. (1994) 5:385–394.[Abstract]
  45. Shibata MA, et al. Enhanced sensitivity to tumor growth and development in multistage skin carcinogenesis by transforming growth factor-{alpha}-induced epidermal growth factor receptor activation but not p53 inactivation. Mol. Carcinog. (1997) 18:160–170.[CrossRef][ISI][Medline]
  46. Mann GB, et al. Mice with a null mutation of the TGF alpha gene have abnormal skin architecture, wavy hair, and curly whiskers and often develop corneal inflammation. Cell (1993) 73:249–261.[CrossRef][ISI][Medline]
  47. Macdiarmid J, et al. The latent membrane protein 1 of Epstein-Barr virus and loss of the INK4a locus: paradoxes resolve to cooperation in carcinogenesis in vivo. Carcinogenesis (2003) 24:1209–1218.[Abstract/Free Full Text]
  48. Rodrigues GA, et al. A novel positive feedback loop mediated by the docking protein Gab1 and phosphatidylinositol 3-kinase in epidermal growth factor receptor signaling. Mol. Cell Biol. (2000) 20:1448–1459.[Abstract/Free Full Text]
  49. Bogoyevitch MA, et al. Targeting the JNK MAPK cascade for inhibition: basic science and therapeutic potential. Biochim. Biophys. Acta (2004) 1697:89–101.[Medline]
  50. Collado M, et al. The power and the promise of oncogene-induced senescence markers. Nat. Rev. Cancer (2006) 6:472–476.[CrossRef][ISI][Medline]
  51. Dominey AM, et al. Targeted overexpression of transforming growth factor-{alpha} in the epidermis of transgenic mice elicits hyperplasia, hyperkeratosis, and spontaneous, squamous papillomas. Cell Growth Differ. (1993) 4:1071–1082.[Abstract]
  52. Wiley HS. Trafficking of the ErbB receptors and its influence on signaling. Exp. Cell Res. (2003) 284:78–88.[CrossRef][ISI][Medline]
  53. Saito T, et al. Differential activation of epidermal growth factor (EGF) receptor downstream signaling pathways by betacellulin and EGF. Endocrinology (2004) 145:4232–4243.[Abstract/Free Full Text]
  54. Yu Y, et al. The significance of serum soluble intercellular adhesion molecule 1 and transforming growth factor alpha in patients with nasopharyngeal carcinoma. Arch. Otolaryngol. Head Neck Surg. (2004) 130:1205–1208.[Abstract/Free Full Text]
  55. Tikhomirov O, et al. Identification of proteolytic fragments from ErbB-2 that induce apoptosis. Oncogene (2005) 24:3906–3913.[CrossRef][ISI][Medline]
  56. Berra E, et al. The activation of p38 and apoptosis by the inhibition of Erk is antagonized by the phosphoinositide 3-kinase/Akt pathway. J. Biol. Chem. (1998) 273:10792–10797.[Abstract/Free Full Text]
  57. Tikhomirov O, et al. Ligand-induced, p38-dependent apoptosis in cells expressing high levels of epidermal growth factor receptor and ErbB-2. J. Biol. Chem. (2004) 279:12988–12996.[Abstract/Free Full Text]
  58. Zwang Y, et al. p38 MAP kinase mediates stress-induced internalization of EGFR: implications for cancer chemotherapy. EMBO J. (2006) 25:4195–4206.[CrossRef][ISI][Medline]
  59. Bapat S, et al. Peroxynitrite activates mitogen-activated protein kinase (MAPK) via a MEK-independent pathway: a role for protein kinase C. FEBS Lett. (2001) 499:21–26.[CrossRef][ISI][Medline]
  60. Barry OP, et al. Constitutive ERK1/2 activation in esophagogastric rib bone marrow micrometastatic cells is MEK-independent. J. Biol. Chem. (2001) 276:15537–15546.[Abstract/Free Full Text]
  61. Chen N, et al. Suppression of skin tumorigenesis in c-Jun NH(2)-terminal kinase-2-deficient mice. Cancer Res. (2001) 61:3908–3912.[Abstract/Free Full Text]
  62. She QB, et al. Deficiency of c-Jun-NH(2)-terminal kinase-1 in mice enhances skin tumor development by 12-O-tetradecanoylphorbol-13-acetate. Cancer Res. (2002) 62:1343–1348.[Abstract/Free Full Text]
  63. Curran JA, et al. Epstein-Barr virus encoded latent membrane protein-1 induces epithelial cell proliferation and sensitizes transgenic mice to chemical carcinogenesis. Cancer Res. (2001) 61:6730–6738.[Abstract/Free Full Text]
Received October 5, 2006; revised February 23, 2007; accepted March 2, 2007.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available