| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Carcinogenesis |
Robertson Laboratory, Division of Molecular Genetics, Institute of Biomedical and Life Sciences, University of Glasgow, Glasgow, United Kingdom [D. C., J. M., J. B. W.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
NPC incidence is greatest in Southern China (up to 4/104/year), of intermediate occurrence (3–10/105/year) among native Alaskans and Inuits, indigenous Southeast Asian populations and among certain populations of North Africa and Arabia. Most other populations have a low incidence of NPC (<1/105/year; Ref. 2 ). The demographic pattern of high NPC incidence along with the anthropological relationship between Southeast Asian and Inuit and native Alaskan populations suggests that there might be genetic loci associated with NPC susceptibility, and there is some evidence to support this idea (3) . In addition, environmental and dietary factors may influence the risk of developing the disease. In particular, the high levels of carcinogenic volatile nitrosamines in the salted fish and other preserved foods eaten by people of high incidence groups has been cited as a potential risk factor (4, 5, 6, 7) . As such, environmental/dietary and genetic factors in addition to EBV infection are likely contributors to NPC risk.
Viral gene expression in the malignant epithelial cells of NPC includes the EBER transcripts (8) and consistent expression of EBV nuclear antigen-1 (9) in the absence of the other nuclear antigens. In addition, LMP1 has been detected in approximately two-thirds of NPC specimens tested, whereas LMP1 mRNA has been detected by reverse transcription-PCR assays in the great majority of NPC biopsies, (10, 11, 12) . LMP2 transcripts are also consistently detected in NPC cells (11 , 13) . In addition, the multiply spliced rightward BamHI-A transcripts (BARTs/CSTs) are abundant in NPC cells (14, 15, 16, 17) and capable of encoding several proteins including RPMS1, A73, BARFO and RK-BARFO (18, 19, 20) .
Genetic deletion studies have revealed that LMP1 is one of the latent genes required for in vitro B-cell immortalization by EBV (21 , 22) . Moreover, LMP1 is oncogenic in nonlymphoid cells, first demonstrated by its ability to induce growth transformation in certain immortalized rodent fibroblast cell lines (23, 24, 25) . Importantly, this oncogenic action extends to epithelial cells. Expression of LMP1 in the epidermis of transgenic PyLMP1 mice induces hyperplasia, an early step in the carcinogenic process (26) . Furthermore, in cultured carcinoma cell lines, heterologous expression of LMP1 leads to reduced serum requirements, loss of anchorage dependence, increased invasive capacity, and, in some cases, inhibition of terminal differentiation (27, 28, 29, 30, 31) . Moreover, growth characteristics of NPC tumors have been correlated with LMP1 expression levels. Detectable LMP1 protein is linked with the expression of EGFR and Ki67 in NPC biopsies (32) , and LMP1-positive NPC tumors appear to grow faster and more expansively than LMP1-negative NPC tumors (33) .
In B-cells, LMP1 has been shown to act similarly to a constitutively active CD40 receptor in prolonging cell survival and supporting proliferation (34, 35, 36)
. Like CD40, LMP1 binds to tumor necrosis factor receptor-associated factors 1, 2, 3 and 5 (37, 38, 39, 40)
. In addition, LMP1 associates with the tumor necrosis factor receptor-associated factor-associated death domain proteins TRADD and RIP (41
, 42)
and to JAK3 (43)
. Through these interactions, LMP1 causes the activation of NF
B, JNK, and STAT proteins (43, 44, 45, 46, 47)
, which ultimately results in a plethora of changes in cellular gene expression. In epithelial cell lines, LMP1 can induce the expression of CD40 and CD54 (both expressed in NPC cells; Ref. 48
, 49
), as well as the antiapoptotic gene A20, EGFR, and IL6 (50, 51, 52)
. In vivo, transgenic epidermal expression of LMP1 leads to up-regulation of the proliferative cytokeratin marker, K6 (26)
.
In transgenic modeling, the accessibility of the mouse skin makes it an ideal tissue in which to study the carcinogenic process. Moreover, the mouse skin model of multistage carcinogenesis has been well characterized with respect to genetic and epigenetic changes (reviewed in Ref. 53 ). Using chemical carcinogens, the stages of initiation, promotion to a benign tumor (papilloma), malignant conversion, and metastatic invasion have been defined. The most commonly used tumor induction protocol involves a single treatment with the mutagen DMBA and then repeated treatments with the tumor promoter TPA. In virtually all DMBA-initiated tumors, activation of the H-ras gene has been identified as an initiating event (54 , 55) , and cyclin D1 has been shown to be a critical target of activated ras in this system (56 , 57) . TPA treatment then promotes the clonal survival and outgrowth of initiated cells to form an overt lesion. The hyperplastic response of the epidermis to TPA is thought to be mediated in part through sustained direct activation of PKC (58) , but, in addition, the enhanced keratinocyte proliferation rate is linked to the accompanying inflammatory skin response (59) . Subsequent loss of p53 contributes to malignant conversion, and deletion of the INK4 locus is associated with transformation to a highly invasive tumor phenotype. Several of these steps have been modeled by engineering transgenic strains of mice with activated oncogenes or deleted tumor suppressor genes (60, 61, 62, 63, 64, 65) .
Transgenic mice expressing LMP1 in the epidermis displayed the phenotype of epidermal hyperplasia with hyperkeratosis that did not progress to tumor formation, demonstrating that LMP1 is insufficient for carcinogenesis (26)
. The phenotype of LMP1-expressing mice is markedly similar to that displayed by mice expressing activated H-ras, TGF
, or (to some extent) v-fos in the epidermis, all of which demonstrate increased epidermal proliferation (61
, 62 , 64
, 66
, 67)
. These observations raise a number of questions relating to the action of LMP1. First, does LMP1 induce hyperplasia by increasing epidermal proliferation, inhibiting differentiation, or both? Second, can LMP1 act as a classical initiator and elicit tumors similar to those observed by TPA promotion of H-ras transgenic mice (68)
; or can LMP1 act as a tumor promoter in the manner of v-fos (69)
. Alternatively, LMP1 may act as either initiator or promoter, depending upon the context, as with TGF, or possibly act at a later stage to promote carcinogenic progression. To address the role of LMP1 in inducing hyperplasia in vivo, the action of LMP1 in the transgenic mouse epidermis was examined. We show that LMP1 acts by inducing epithelial cell proliferation. Furthermore, to assess the synergistic molecular events associated with LMP1 in carcinogenesis, PyLMP1 mice were treated with a classical regime of chemical carcinogens, revealing that LMP1 can act as a tumor promoter.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
All blots were hybridized with [-32P]dCTP-labeled, randomly primed DNA probe fragments (Prime-it II kit; Stratagene) and washed under stringent conditions [0.1x SSC-0.1% SDS (w/v) at 68°C] before autoradiography. Northern blot signals were quantified by phosphorimage analysis; and a radiolabeled GAPDH probe was used to normalize for loading, dividing sample-specific phosphorimage values with the GAPDH value of the same sample. Short cDNA fragments primarily consisting of the 3' noncoding region were used as probes for specific murine cytokeratin sequences (K1, K6, K10, and K14 described in Ref. 26
). The K17 probe was a 3' 255-bp cDNA fragment (71)
. For the transgene probe, a 3-kb HindIII-EcoRI LMP1 fragment from the PyLMP1 transgene was used.
BrdUrd Labeling and Immunodetection.
The protocol used is essentially as described (72)
. One h before being killed, BrdUrd (50 µg/g body weight) in 0.9% saline was injected i.p. into mice. Dorsal skin, tongue, and tail tissues were collected (as well as tissue for transgene status determination), fixed in 10% neutral buffered formalin (20 mM NaH2PO4; 46 mM Na2HPO4; and 4% formaldehyde, v/v), and embedded in paraffin. Seven-µm sections were cut and subjected to immunohistochemical staining using a murine anti-BrdUrd antibody (Sigma Chemical Co.), then an antimouse IgG biotinylated second antibody, and finally the Vectastain elite ABC complex (which contains avidin and biotinylated horseradish peroxidase, and which allows the avidin: biotinylated horseradish peroxidase to bind to the biotinylated secondary antibody). Positively stained nuclei (brown) were then detected using DAB (3,3'-diaminobenzidine tetrahydrochloride; Sigma Chemical Co.), and negative nuclei were counterstained with hematoxylin (blue). The number of BrdUrd-labeled nuclei in the interfollicular areas, per field of vision (x312.5) were counted for 10 fields of vision/section, and a mean value for the tissue was obtained. Tissues from at least three different mice in each group were counted in this way, where the mean value of these for the tissue type is referred to as the labeling index. All counts were conducted blind, and transgene status determined subsequently.
Immunofluorescent Staining.
Immunofluorescent examination of PyLMP1 transgenic epidermis was conducted essentially as described previously (9)
. Seven-µm sections of snap-frozen epidermis were taken and stored at -20°C on gelatin-coated slides. Sections were blocked for 30 min in 1x goat wash (1.3 M NaCl; 70 mM Na2HPO40.2H2O; 30 mM NaH2HPO4.H2O; 0.5% sodium azide, v/v; 10% goat serum, v/v; 2% Triton X-100, v/v; and 0.5% Tween 20, v/v). The primary antibody used was rabbit polyclonal anti-loricrin with secondary FITC-conjugated goat antirabbit IgG (Sigma Chemical Co.). Sections were mounted with Vectashield (Vector Laboratories) and viewed using a Leitz orthoplan microscope at x312.5.
Chemical Carcinogen Administration.
The protocol used is essentially as described (73)
. Mice of PyLMP1 line 53 were backcrossed (from C57Bl/6 strain) into the chemical carcinogen-sensitive FVB strain for this study and mice from a minimum of three backcrosses were used (average, 87.5% FVB at backcross 3). Chemical treatment was initiated with mice 8 weeks of age. Hair from the mouse dorsal region was removed by shaving 24 h before the initial chemical administration. Thereafter, the dorsal skin was carefully and regularly shaved. Twenty-five µg DMBA was applied topically to the dorsal skin in 200 µl of acetone. One week later, mice were treated with the tumor promoter TPA, twice weekly, with 200 µl of 5 x 10-5 M TPA for 20 weeks (regime 1) or, alternatively, not treated further (regime 3). TPA treatment for 20 weeks without DMBA initiation was also conducted (regime 2). DMBA treatment followed by only 4 weeks of TPA treatment constituted regime 4. The number of benign and malignant tumors on each mouse was recorded weekly during treatment and beyond. The benign tumors were categorized according to size, and the number of lesions in size categories 2–4 were counted [category 1, <0.2 mm (not counted); category 2, 0.2–0.5 cm; category 3, 0.5–1.0 cm; and category 4, >1.0 cm diameter]. Analysis of the data from each papilloma size category, from the total papilloma count, and from the total malignant carcinoma count was facilitated by use of the Minitab statistical analysis package. Normal distribution of the data were confirmed using Rankits plots, and differences between the two groups were analyzed for significance using the two-sample t test. All mice were of the same background strain, with negative siblings of the transgenic mice used as controls, and housed and fed under the same conditions in a conventional facility. Lesion counts were conducted blind of transgenic status.
Mice were killed if the tumor load became excessive; if any tumor inhibited movement, caused discomfort, or was located at a natural orifice; if any individual tumor exceeded 20 mm in diameter or became ulcerated; or at the termination of the experiment. Decision to remove an animal from study because of these criteria was taken blind of transgenic status. Sections of tumor samples were collected for culture and snap frozen or fixed in 10% neutral buffered formalin (20 mM NaH2PO4; 46 mM Na2HPO4; and 4% formaldehyde, v/v) for paraffin embedding and histopathological analysis.
Development of Carcinoma Cell Lines.
Primary cells were established from malignant carcinomas by removing the tumor into explant medium (DMEM containing 87 µg/ml penicillin, 117 µg/ml streptomycin, and 2.9 µg/ml Fungizone). The tumor was then transferred to warmed culture medium (DMEM containing 20% FCS, v/v; 2% L-glutamine, v/v; 12 µg/ml penicillin; 97 µg/ml streptomycin; and 1.2 µg/ml Fungizone) in which it was diced with a sterile scalpel. The small pieces were allowed to adhere to a tissue culture flask briefly before the careful addition of 2.5 ml of culture medium. The samples were incubated at 37°C in 5% CO2 for 7 days before an additional 2.5 ml of culture medium were added. After an additional 7 days, visible outgrowth of adherent cells from the tumor segments could be observed. The cells were trypsinized, washed, and reseeded in culture medium into 25-cm3 culture flask until confluent. Thereafter, the immortal cultures were maintained in culture medium with 10% FCS.
Western Blotting.
Protein was extracted from cells using RIPPA buffer [150 mM NaCl, 0.5% NP40 detergent, 0.5% sodium deoxycholate, 0.1% SDS, 5 mM DTA, and 20 mM Tris (pH 7.5)] with protease inhibitors (protease inhibitor cocktail, P2714; Sigma Chemical Co.). Forty µg of protein from each extract was electrophoretically separated by SDS-PAGE (8%) minigels, at 180 V for 40 min. Proteins from the gel were electroblotted onto Millipore Immobilon-P membranes at 65 V for 1 h. The membranes were blocked overnight at 4°C in 15 ml of blocking buffer (0.4% casein, 0.1% Tween 20, and 0.02% NaN3 in PBS), then incubated for 1 h at room temperature with a 1:20 dilution of S12 anti-LMP1 monoclonal antibody (in blocking buffer). The membranes were washed three times for 5 min each in PBS-Tween, and incubated for 1 h at room temperature in secondary antibody [1:1000 dilution of alkaline phosphatase conjugated antimouse IgG (Sigma Chemical Co.) in blocking buffer]. Antibody binding was visualized using the CDP-Star detection system (detailed in Ref. 74
).
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Expression Levels of Markers of Differentiation and Proliferation.
Up-regulation of K6 protein and mRNA was previously noted in the skin of transgenic LMP1 pups and adults of both lines (26)
. Several postnatal time points were examined revealing that the up-regulation of both RNA and protein was apparent from 1 day postnatal, with maximal RNA levels reached by day 10. Thus, additional Northern analyses were focused around day 10. Northern blots of total RNA samples from skins of transgenic and control siblings of line 5 mice at postnatal days 8 and 10 were probed with sequences for the murine cytokeratin genes K1, K6, K10, K14, and K17 and then with GAPDH to control for loading (Fig. 1)
. Signals were quantified by phosphorimage analysis and adjusted against the loading control, and transgene and control samples were compared. Samples from several mice of these ages have been compared, and the range of ratios is presented in summary in Table 1
.
|
|
.
|
In Vivo Proliferation.
To address directly whether cellular proliferation was affected in the transgenic epidermis in vivo, cellular DNA synthesis was measured by BrdUrd nuclear incorporation in epidermal cells after i.p. injection. Eight transgenic mouse pups of line 53 and eight control siblings were injected at 1 week of age (when the phenotype is most obvious in the dorsal skin). Three transgenic adults, when there is no apparent phenotype in the dorsal skin, and three control siblings were also tested.
Dorsal skin, tongue, and tail tissues were taken from pups, dorsal skin was taken from adults, and several sections of each tissue were analyzed for BrdUrd incorporation (Fig. 2)
. Ten fields of view were counted for interfollicular positively stained keratinocytes for each tissue section, and the mean value of these for the section was determined. The section values for each tissue were collated for the groups of mice and the mean for each transgenic or control group calculated as the labeling index. The labeling indices for the sample groups are presented graphically in Fig. 3
. In dorsal skin from pups 1 week of age, epidermal cell proliferation is 2-fold higher in the transgenic compared with the control samples. The labeling indices of 22.94 and 11.22, respectively, show a highly significant difference (two-sample t test; 95% confidence limit; P = 0.0035). As expected, proliferation in the adult epidermis is much lower than in pups; nevertheless the 2-fold difference between transgenic and control samples is maintained, with labeling indices of 4.24 and 2.1, respectively, with a very high significant difference (P = 0.0004). Similarly, transgenic animals showed increased proliferation in tail epidermis and tongue epithelium compared with controls (Fig. 3)
. For 1-week tail epidermis, transgenic and control labeling indices were 56.0 and 17.13, respectively, showing a significant 3-fold elevation of proliferation in the transgenic samples (P = 0.026). For 1-week tongue epithelium, where the hyperplastic phenotype is less evident, the labeling indices for transgenic and control samples were 26.75 and 17.16, showing a 1.6-fold elevation in the LMP1 transgenics (P = 0.02).
|
.
Chemical Carcinogen Treatment.
To ascertain whether LMP1 can act to augment chemical carcinogens or whether it can substitute for either an initiator or a promoter, three treatment regimes were initially conducted using PyLMP1.53 transgenic mice and their negative sibling controls. Treatment 1 involved the classical single initiating treatment with DMBA and then 20 weeks of promoting TPA-treatment (twice/week). Thirty-two transgenic mice and 26 control siblings were treated under this regime and monitored for up to 55 weeks from first treatment. An artificial limit was placed on carcinoma load in view of animal welfare (see "Materials and Methods"), and as such, mice were removed from the study as required, from week 17 onwards. Regime 2 involved 20 weeks of TPA treatment (twice/week) in the absence of DMBA initiation. Transgenic mice and control siblings placed on this regime were monitored for 46 weeks. Regime 3 involved a single DMBA treatment. Transgenic and control mice placed on this regime were monitored for 35 weeks. Lesion formation was counted weekly, and the papillomas were categorized by size (see "Materials and Methods"). Conversion of a papillomatous lesion to a carcinoma was recorded, and confirmation of lesion status by histopathology was conducted after the final sample collection.
All mice in regime 1 developed papillomas (Table 2)
. The mean numbers of papillomas/mouse/week in size categories 2–4 and the conversion to carcinoma are shown graphically in Fig. 4
. Lesions appeared earlier in the transgenic group, at week 5, compared with the control mice, in which lesions first appeared at week 7. Transgenic mice showed a marked increase in the number of small papillomas (size 2, 0.2–0.5 mm) formed compared with controls (Fig. 4)
. The difference became significant at week 13 of the study (two-sample t test; 95% confidence; 56 degrees of freedom; P = 0.017 at week 13), showing increasing significance for the remainder of the study (e.g., P = 0.0007 at week 25 and P = 0.0001 at week 26).
|
|
, supported by the statistical analysis showing no significant difference in this category. However, surprisingly, control mice developed significantly more papillomas in the size 4 category (>1.0 cm) than the transgenic mice. The difference became significant at week 16 of the study (P = 0.013) and continued for the remainder of the study. Combining the opposing sets of data, more lesions in total formed with time in the transgenic group compared with the controls, but this did not show consistent significant difference until after cessation of the carcinogen treatment at week 21.
The mean number of carcinomas formed per mouse (2.69 for transgenic and 2.81 for control siblings) does not show significant difference (P = 0.86); however, carcinoma load was given an artificial limit in view of animal welfare. Animals were removed from the study because of lesion load from week 17 in the transgenic group and from week 20 in the controls. Consequently, conversion rates of papilloma to carcinoma have been calculated at the week 31 values (allowing time for conversion), when 
65% of animals remained under study (in both groups). At week 31, the mean percentage of carcinomas/total lesions was 6% for the transgenic mice and 12% for the controls, without showing significant difference (P = 0.14). This indicates that LMP1 did not enhance conversion of papillomatous lesions to carcinomas, although more small papillomas were induced.
After histopathological analysis, lesions that were scored as carcinomas were revealed to be predominantly of the squamous cell type. Two of 37 carcinomas from the LMP1-positive group were of the more malignant spindle cell type, with 1 of 31 in the control group, indicating that LMP1 does not lead to an increased progression to the more aggressive spindle cell carcinoma.
Cell lines were established in culture from several of the carcinomas that developed in the transgenic and control mice. LMP1 expression in the transgenic carcinoma cell lines was confirmed (Fig. 5)
.
|
Because the PyLMP1 mouse phenotype shows some similarities with the phenotype of mice transgenic for oncogenic H-ras, regime 2 (20 weeks TPA treatment only) was conducted to address the ability of LMP1 to functionally substitute for oncogenic ras. After 46 weeks of monitoring (including the 20-week treatment period), no (0 of 11) LMP1-positive mice developed lesions, whereas one control animal (1 of 17) developed a single papilloma, at week 29 of the study, which did not progress to carcinoma (Table 2)
. Therefore, LMP1 is not able to substitute for the action of oncogenic ras in this system.
To investigate whether LMP1 could substitute for TPA-induced tumor promotion, regime 3 (a single DMBA treatment) was conducted. After 35 weeks of monitoring, four (4 of 14) LMP1-positive mice developed papillomas (1, 1, 2, and 5 papillomas, in each case) whereas no controls (0 of 12) developed lesions (Table 2)
. This result suggests that LMP1 may be able to act as a weak tumor promoter.
To investigate this further, a new regime was conducted. Weak (or second-stage) promoter activity can be revealed after minimal TPA treatment (77
, 78)
. Regime 4 involved a single DMBA treatment and then 4 weeks of TPA treatment (twice/week). Seven transgenic and six control mice were placed on this regime and monitored for up to 70 weeks (although mice were removed from the study, as described above, from week 31). In this treatment regime, control mice remained papilloma free for 52 weeks; subsequently two mice in this group developed size 1 (<0.2 cm) papillomatous lesions (at the end of the study period), one of which regressed after a few weeks (Fig. 6
; Table 2
). In contrast, 7 of 7 LMP1-expressing mice formed papillomas after the minimal TPA treatment, some of which converted to carcinoma. Mice in the transgenic group developed papillomas from week 9 of the study period, all mice having lesions by week 12. Expression of LMP1 in the epidermis is therefore able to substitute for prolonged TPA promotion to induce lesion formation. As such, LMP1 acts as a weak (or second-stage) tumor-promoting agent in this system.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A transient induction of the RNA levels of differentiation markers K1 and K10 was observed in the transgenic skin compared with controls at a time when the juvenile phenotype is maximal. This may underlie the hyperkeratotic aspect of the phenotype. LMP1 has been shown to activate JNK, which regulates transcriptional control through the AP-1 complex. Functional AP-1 sites have been shown to activate the transcription of several epithelial differentiation-specific genes, including K1 and involucrin (79, 80, 81, 82)
. In addition, both LMP1 expression and CD40 ligation in epithelial cell lines result in activation of NFB (52)
. Activation of NFB in epidermal suprabasal cells is necessary for growth inhibition and differentiation (83)
. As such, LMP1 may be able support differentiation in vivo while promoting proliferation.
We have gone on to show that expression of LMP1 in the transgenic mouse epidermis can act in a tumor-promoting fashion, to augment the action of chemical carcinogens. This is consistent with the ability of LMP1 to induce proliferation. TPA treatment of mouse skin similarly induces proliferation and is thought to act in part by binding to PKC and activating it (mimicking diacylglycerol). Activated PKC is proposed in turn to activate AP-1 (among other substrates) by the dephosphorylation of sites on c-Jun, which are inhibitory to DNA binding. This is mediated by PKC phosphorylation and inhibition of glycogen synthase kinase-3ß, a c-Jun kinase (84) . It is tempting to speculate that LMP1 may bring about a similar effect by the activation of the JNK pathway.
Paradoxically, expression of LMP1 in the epidermis appears to inhibit the expansion of the benign papillomas (to the larger size lesions), although more small lesions have formed. Similarly, the numbers of carcinomas developing in each group were equivalent, despite the greater abundance of small papillomas in the transgenic group. These seemingly opposite activities may be explained in the ability of LMP1 to activate pathways involved in proliferation (such as the JNK pathway) and in epithelial differentiation (such as the NFB pathway). Whereas LMP1 may provide the proliferative environment which promotes lesion formation, it could simultaneously present a hurdle to lesion expansion by maintaining cell differentiation or senescence. A prediction of this theory would be that the factors (or the loss thereof) which lead to the inhibition of differentiation or senescence could overcome the hurdle and possibly cooperate with LMP1 in lesion expansion or progression.
There are striking similarities between mice expressing TGF in the epidermis (62
, 66)
and the PyLMP1 mice. In addition to similarities in pup phenotype and the diminution of phenotype in adults, the mice in both series show an increase in BrdUrd mitotic index in the epidermis (64
, 85)
. However, TPA treatment (in the absence of initiating DMBA) of transgenic mice overexpressing TGF in the basal layer induced papillomas in the transgenic mice (86
, 87)
, which is in contrast with our findings with the LMP1 mice. However, the inability of LMP1 to induce benign papillomas upon TPA treatment alone may be a function of expression level or the precise site of expression of the transgene. We are currently investigating whether the tumor-promoting activity of LMP1 may be mediated through the TGF/EGFR pathway.
The importance of the inflammatory response in tumor promotion has been revealed using TNF-null mice (88)
. TPA treatment of TNF-null mice does not result in tumor promotion. Again, it is tempting to suggest that LMP1 may be activating inflammatory signaling pathways through interaction with TRAFs and TRADD to induce lesion formation.
In the viral context in vivo, as a latent and lytic protein, LMP1 may promote cellular proliferation to expand the viral pool and support differentiation to facilitate the lytic cycle. In conjunction with other viral proteins and/or cellular oncogenic mutations that inhibit or redirect differentiation, a potential consequence is tumorigenesis.
NPC has a complex etiology with a possible genetic predisposition, potential contribution by dietary carcinogens, and a viral component. In the transgenic system we have analyzed here, the genetic background of the mice is critical in that different strains of mice show markedly different responses to chemical carcinogens. We have concentrated our efforts on mice bred to a background of genetic predisposition to chemical carcinogens. We have studied one viral component, LMP1, and found that it induces epithelial proliferation in vivo and renders the transgenic mice highly sensitive to chemical carcinogens. Minimizing the carcinogen treatment such that no lesions are formed in control mice nevertheless leads to carcinogenesis in the PyLMP1 mice. LMP1 acts early in the carcinogenic process, increasing the frequency of lesion formation, but can also act to inhibit expansion of the benign lesions, and it does not enhance progression of the lesions. This leads to the hypothesis that LMP1 may be an early factor in the genesis of EBV-associated NPC, which would be consistent with the observation that LMP1 is expressed in preinvasive NPC lesions (89) . Moreover, the expression of LMP1 in EBV-infected cells may increase the risk of lesion formation associated with dietary carcinogens (7) . Whether LMP1 continues to be a tumorigenic factor in later stages remains to be elucidated.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 The work was supported by grants from the Royal Society and from the Cancer Research Campaign. 
2 To whom requests for reprints should be addressed, at Institute of Biomedical and Life Sciences, Division of Molecular Genetics, Robertson Building, University of Glasgow, 54 Dumbarton Road, Glasgow, G11 6NU, United Kingdom. Phone: 44-141-330-5108; Fax: 44-141-330-4878; E-mail: Joanna.Wilson{at}bio.gla.ac.uk
3 The abbreviations used are: NPC, nasopharyngeal carcinoma; LMP, latent membrane protein; EGFR, epidermal growth factor receptor; JNK, c-Jun NH2-terminal kinase; NFB, nuclear factor B; DMBA, dimethylbenzanthracene; TPA, 12-O-tetradecanoylphorbol-13-acetate; PKC, protein kinase C; TGF, transforming growth factor; BrdUrd, bromodeoxyuridine; Py, polyomavirus; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; AP-1, activator protein-1.
6 Laverty and Wilson, unpublished observations.
Received 9/ 8/00. Accepted 7/10/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B activation by latent infection membrane protein 1, the Epstein- Barr virus transforming protein. Proc. Natl. Acad. Sci. USA, 93: 11085-11090, 1996.B-activating site and the sole TRAF3 binding site of LMP-1 defines two distinct signaling motifs. J. Biol. Chem., 272: 19777-19784, 1997.B. Proc. Natl. Acad. Sci. USA, 94: 12592-12597, 1997.B activation. Mol. Cell. Biol., 19: 5759-5767, 1999.B and cell surface phenotype via two effector regions in its carboxy-terminal cytoplasmic domain. Oncogene, 10: 549-560, 1995.[Medline]
B-mediated transcription by mutant derivatives of the latent membrane protein of Epstein-Barr virus. J. Virol., 69: 2968-2976, 1995.[Abstract]
B. J. Biol. Chem., 267: 24157-24160, 1992.B pathway involving TNF receptor-associated factors. Oncogene, 14: 2899-28916, 1997.[Medline]
in complex with phorbol ester. Cell, 81: 917-924, 1995.[Medline]
in the epidermis of transgenic mice elicits hyperplasia, hyperkeratosis, and spontaneous, squamous papillomas. Cell Growth Differ., 4: 1071-1082, 1993.[Abstract]
promotes skin tumorigenesis via H-ras-dependent and H-ras-independent pathways. Cell Growth Differ., 5: 385-394, 1994.[Abstract]
. Cancer Res., 56: 4413-4423, 1996. during epidermal development and differentiation. Genes Dev., 5: 714-727, 1991.B function in transgenic epithelial tissue demonstrate a growth inhibitory role for NF-B. Proc. Natl. Acad. Sci. USA, 95: 2307-2312, 1998. and v-fos cooperation in transgenic mouse epidermis induces aberrant keratinocyte differentiation and stable, autonomous papillomas. Oncogene, 10: 279-289, 1995.[Medline]
bypasses the need for c-Ha-ras mutations in mouse skin tumorigenesis. Mol. Cell. Biol., 12: 4643-4653, 1992. 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., 10: 15-22, 1994.[Medline]
are resistant to skin carcinogenesis. Nat. Med., 5: 828-831, 1999.[Medline]
This article has been cited by other articles:
|
|
 |
|
  C. T. Charalambous, A. Hannigan, P. Tsimbouri, G. M. McPhee, and J. B. Wilson Latent membrane protein 1-induced EGFR signalling is negatively regulated by TGF{alpha} prior to neoplasia Carcinogenesis, August 1, 2007; 28(8): 1839 - 1848. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Soni, T. Yasui, E. Cahir-McFarland, and E. Kieff LMP1 Transmembrane Domain 1 and 2 (TM1-2) FWLY Mediates Intermolecular Interactions with TM3-6 To Activate NF-{kappa}B J. Virol., November 1, 2006; 80(21): 10787 - 10793. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Stevenson, C. Charalambous, and J. B. Wilson 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., October 1, 2005; 65(19): 8826 - 8835. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Macdiarmid, D. Stevenson, D. H. Campbell, and J. B. Wilson The latent membrane protein 1 of Epstein-Barr virus and loss of the INK4a locus: paradoxes resolve to cooperation in carcinogenesis in vivo Carcinogenesis, July 1, 2003; 24(7): 1209 - 1218. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Rundhaug Matrix Metalloproteinases, Angiogenesis, and Cancer: Commentary re: A. C. Lockhart et al., Reduction of Wound Angiogenesis in Patients Treated with BMS-275291, a Broad Spectrum Matrix Metalloproteinase Inhibitor. Clin. Cancer Res., 9: 00-00, 2003. Clin. Cancer Res., February 1, 2003; 9(2): 551 - 554. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
, LMP1 +ve; 
, controls.
