| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
University of California San Francisco Comprehensive Cancer Center and Department of Surgery, University of California, San Francisco 94143-0808 [R. R. R., J. M. A.]; Department of Pathology, Harvard Medical School, Boston, Massachusetts 02115 [S. D., K. M.]; and McArdle Laboratory for Cancer Research, University of Wisconsin Medical School, Madison, Wisconsin 53706 [T. B., P. F. L.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
There are >100 types of HPVs, but only a subset of "high-risk" viral types induce cervical carcinogenesis. HPVs contain two genes in their early region, E6and E7, and high-risk viral types encode E6 and E7 oncoproteins with an enhanced affinity for host cellular targets (1 , 4) . Whereas transcriptional activity has been associated with both E6 and E7, as these oncoproteins are not DNA binding transcription factors, the major thrust of investigation has centered on protein-protein interactions with cellular targets (5, 6, 7) . In many instances, association of the cellular target with a HPV oncoprotein leads to target protein destabilization and ultimate destruction (8, 9, 10) . In other instances, viral oncoprotein binding may activate the host cellular protein (11) . Several studies support a functional role for the E6and E7oncogenes in tumorigenesis. Both oncogenes are expressed consistently in cervical cancer tissues obtained from patients (2) . Cell transfection of high-risk HPV16 or 18 E6and E7oncogenes transforms established cell lines and immortalizes primary cells (12, 13, 14, 15) . Ubiquitous or targeted expression of E6 and E7 has been shown to produce benign tumors or cancers in transgenic mice (16, 17, 18, 19, 20, 21) . The potential importance of E6 and E7 function in cervical carcinogenesis has also spawned vaccine development targeting viral oncoprotein epitopes (22, 23, 24, 25) . These vaccine efforts are a parallel approach to prophylaxis based on immunization against viral capsid antigens (26) . Rational drug design is also being applied to target the enzymatic functions of E6 or more recently E7 (27, 28, 29) . Thus, the importance of HPV oncogenes for viral infection, carcinogenesis, and potential therapy mandates a complete understanding of their molecular, cellular, and tissue biology in cervical epithelial cells.
However, despite insights into viral oncogene function, the discrete biology of the individual E6 and E7 oncoproteins in cervical epithelial cells is unclear. Therefore, we combined our expertise in the induction of cervix cancer in HPV transgenic mice by chronic estrogen administration (30 , 31) with transgenic mice engineered to express either HPV E6 or E7 oncoprotein individually (32) , to determine the discrete effects of each HPV oncogene in cervical carcinogenesis. Surprisingly, E7 alone was sufficient to produce both high-grade cervical dysplasia and invasive cervical malignancies. In mice engineered for expression of E6 oncoprotein alone, only low-grade cervical dysplasia was evident without additional neoplastic progression within the 6-month treatment interval of the study. Coexpression of E6 and E7 in double-transgenic mice revealed that E6 modulated the malignant phenotype produced by E7, in that cervical cancers were larger and more extensive. As such, this study demonstrates the potent cocarcinogenic activity of the HPV16 E7 oncoprotein with estrogen in the mouse cervix. HPV16 E6 appears to have predominant efficacy as a modulator of E7 estrogen cervical cocarcinogenesis.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Hormone Treatment.
One-month virgin female transgenic and nontransgenic mice were anesthetized with halothane, s.c. implanted in the dorsal skin with continuous release pellets delivering 0.05 mg 17ß-estradiol over 60 days (Innovative Research of America, Sarasota, FL). Mice were treated with hormone for 1, 3, or 6 months; the longer treatment times required an additional insertion of one or two pellets (31)
. Groups of 4–7 transgenic and nontransgenic mice were treated for 1 and 3 months. Larger groups comprising 20 (K14-E6, K14-E7, K14-E6:E7, and nontransgenic) or 40 (K14-HPV16) transgenic mice were treated with 17ß-estradiol for 6 months. Data from the K14-HPV16 transgenic mice were pooled from two concurrent studies.
Tissue Procurement, Histopathological Analysis, and Determination of Cancer Size.
Mice were anesthetized with 2.5% Avertin and perfused through the aorta with 3.75% paraformaldehyde. The entire reproductive tract including parametrial, retroperitoneal soft tissue and lymph nodes were dissected, placed in an embedding cassette (Fisher) with one sponge for compression, and postfixed in 3.75% paraformaldehyde overnight at 4°C. In-cassette postfixation produced a linear reproductive tract facilitating subsequent sectioning of the entire organ in one plane (Fig. 1)
. The posterior vaginal wall was removed leaving an intact rim of vulvar tissue, and a 3–5 mm end piece of plastic pipette tip was inserted into the vaginal opening to keep the vagina open during subsequent processing steps. Tissues were washed in 1x PBS, and dehydrated through graded alcohols and xylene using a Leica TP1050 vacuum tissue processor (Leica Microsystems, Inc., Bannockburn, IL). After pipette tip removal, the reproductive tract was embedded with the cut vaginal wall surface oriented downward. The entire reproductive tract was serially sectioned, and 10–15 sections were collected at 75 µm intervals for H&E staining and immunohistochemistry.
|
HPV16 E6/E7 mRNA Transgene Expression.
Transgenic (n = 6 for each genotype; K14-HPV16, K14-E7, and K14-E6 transgenic mice) and nontransgenic (n = 4) mice were treated with 17ß-estradiol for 1 month. After sacrifice by cervical dislocation and thoracotomy, the uterus, cervix, and vagina were rapidly removed. The uterine horns were separated and discarded from the cervix, vagina, and vulva, which were maintained as an intact tissue block (see cervico-vaginal region, Fig. 1
). This distal reproductive tract block was snap frozen in liquid nitrogen and stored at -80°C. Total RNA was extracted after a 90-s homogenization in TRIzol (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. DNA was removed using AMBION DNA free kit (Ambion, Austin, TX). RNA samples were quantified by UV spectroscopy, and 500 ng of total RNA was used for reverse transcription, as described previously (16)
. After inactivation of reverse transcriptase by heating at 95°C for 5 min, transgene E6/E7 expression was determined by real-time TaqMan RT-PCR (34
, 35)
using PCR primers 5' and 3' to the E6* splice site and a TaqMan probe specific for the 3' splice acceptor side of the E6* splice site. This primer and TaqMan probe set detected both full-length E6 and E6* mRNA. Primer and TaqMan probe sequences were: E6 forward, (5') GCACAGAGCTGCAAACAACTATACA; E6 reverse, (5') GCTTTTTGTCCAGATGTCTTTGCT; TaqMan probe, 6-carboxyfluorescein-TGCGACGTGAGGTGTATTAACTGTCAAAAGC-6-carboxy-tetramethyl-rhodamine-3' (Integrated DNA Technologies, Coralville, IA). PCR was conducted in triplicate, on an ABI PRISM 7700 instrument (Applied Biosystems, Foster City, CA), in 50-µl reaction volumes containing: 1x PCR buffer, 5.5 mM MgCl2, 0.9 µM of each primer, 200 µM dNTPs, 200 nM of probe, and 0.025 units/µl Taq Gold (Applied Biosystems). PCR cycling conditions consisted of 95°C for 15 min, followed by 40 cycles of 95°C for 30 s and 60°C for 60 s. Using histone 3.3A as an endogenous control gene, relative expression levels for E6 ORF expression were calculated as 2-(Ct E6 –Ct histone 3.3A), as described previously (35)
.
Determination of Reproductive Tract Proliferation Using BrdUrd Incorporation.
Labeling of S phase cells by BrdUrd was performed as described previously (19
, 36)
. An intermediate magnification image of the entire cervical-lower uterine transformation zone (see box, Fig. 1
) was captured, and all of the S phase epithelial cells within this region were counted. The entire squamous epithelium of the transformation zone, from the squamo-columnar junction to the opening of the cervical canal, but not including the canal, was counted on five to six adjacent x10 fields. Analysis for the number of BrdUrd-positive cells was conducted on 4–7 animals per genotype and time point.
p53 Immunohistochemistry.
Five µm paraffin-embedded sections were processed as described above and endogenous peroxidase activity quenched with 3% H2O2 in methanol for 30 min, followed by antigen retrieval (Retrievit pH 2; Biogenex, San Ramon, CA) in a 95°C water bath for 35 min. After room temperature equilibration, sections were washed in PBS and blocked at with 5% albumin:5% goat serum in PBS for 30 min. An avidin and biotin block was performed for 15 min each per the manufacturer’s instructions (Avidin/Biotin blocking kit; Vector Laboratories). The p53 antibody (CM 5 clone; Novacastra) was applied overnight at 4°C and diluted 1:2000 in 5% albumin:5% goat serum. After incubation with a biotinylated goat-antirabbit (1:200; Vector Laboratories) antibody for 45 min, antibody detection and visualization was performed by sequential applications of the Vector Elite ABC kit, and the chromagen 3,3 diaminobenzadine (Sigma). Sections were counterstained in Gills #1 Hematoxylin (Sigma).
Determination of Apoptosis.
For analysis of DNA fragmentation in the squamous epithelium of reproductive tracts, the Trevigen DermaTacs Apoptag kit (Trevigen, Gaithersburg, MD) was used, which incorporates BrdUrd into sites of double-strand DNA breakage. The experiments were performed according to the manufacturer’s instructions, except that reagent incubation times were extended, in addition to a less stringent blocking for endogenous peroxidase. Briefly, sections were digested with proteinase K (1:50) for 1 h at 37°C, followed by endogenous peroxidase quenching in 1% H2O2 in PBS for 5 min. After equilibration in 1x terminal deoxynucleotidyltransferase buffer, sections were incubated at ambient temperature for 1 h each with terminal deoxynucleotidyltransferase labeling mix and then the anti-BrdUrd antibody. Sections were subsequently incubated with peroxidase-conjugated streptavidin for 20 min, developed with Tacs blue label for 15 min, and counterstained for 10 min using Counterstain Red C (DermaTacs kit).
Determination of Centrosome Copy Number.
Formalin-fixed, paraffin-embedded 5-µm tissue sections were deparaffinized in xylene for 30 min, rehydrated in alcohols and water as described above, and boiled in 10 mM citrate buffer (pH 6.0) for 30 min in a microwave oven. After a room temperature dH2O wash, sections were treated with Digest-All 3 (Zymed) for 10 min at 37°C, blocked with 10% normal donkey serum in PBS for 15 min, and incubated overnight with a 1:50 dilution of a polyclonal rabbit antipericentrin antibody (BAbCO) at 4°C. Antibody detection was performed using a 1:1000 dilution of a rhodamine red-labeled donkey antirabbit antibody (Jackson Immunoresearch) for 2 h at 37°C. Sections were counterstained and mounted with 4',6-diamidino-2-phenylindole-Vectashield (Vector Laboratories). Cells were analyzed using a Leica DMLB Epifluorescence microscope equipped with a Sony DKC5000 digital camera system.
Image and Statistical Analysis.
Images were captured using a SPOT CCD camera (Diagnostics Instruments, Sterling Heights, MI) attached to a LEICA DMRXA microscope. Figures were composed using Adobe PhotoShop 6.0 (San Jose, CA). Numerical data are presented as the mean ± SD, and were analyzed statistically by ANOVA, two-tailed unpaired Student’s t test and 
2 analysis, where appropriate.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. With the exception of one uterine adenocarcinoma, all of the reproductive tract cancers in the K14-E7-transgenic mice originated in squamous epithelium (see next paragraph). Similar to previous work, 83% of K14-HPV16-transgenic mice developed reproductive tract squamous cancer (Table 1
; Refs. 30
, 31
). In stark contrast, none of the K14-E6 or nontransgenic mice developed reproductive tract malignancies.
|
. To determine a distinct biology for each HPV transgenic genotype, the complete reproductive tract from ovaries to vulva was harvested, embedded, cut, and stained in one tissue plane, and the entire organ was analytically divided into discrete anatomical regions (Fig. 1)
. Reproductive tract squamous carcinomas in K14-E7 transgenic mice were dispersed throughout the cervico-vaginal epithelium, with a biased incidence favoring localization in the lower reproductive tract squamous epithelium (vulva, vagina, and outer cervix) compared with the cervical-uterine transformation zone at the junction of the upper cervix and lower uterus (Table 1
; Fig. 1
). In contrast, reproductive tract malignancy in K14-HPV16-transgenic mice was more frequently detected at the cervical transformation zone (Table 1)
. The number of mice with multiple cancers and the mean number of reproductive tract cancers per mouse were greater in K14-E7 compared with K14-HPV16-transgenic mice (Table 1)
. However, cervical cancer size was significantly less in the K14-E7 compared with either K14-HPV16 or K14-E6:E7 transgenic counterparts (Table 1)
.
Development of a Histopathological Grading System for Transgenic Mouse Cervical Squamous Carcinogenesis.
The marked discrepancy in the extent of carcinogenic progression in estrogen-treated K14-E7 versus K14-E6-transgenic mice motivated us to determine differences in histological progression according to genotype. Therefore, we developed a system to determine the degree of dysplastic or carcinogenic progression in the mouse cervix, initially using estrogen-treated K14-HPV16 transgenic mice in which histopathological progression to cervical malignancy was first demonstrated (Refs. 30
, 31
; Fig. 2
). This grading system was based on the established criteria for classification of human cervical neoplasia or malignancy accounting for differences between the mouse model and patients. One major difference was that high-grade dysplastic mouse cervical epithelium always retained multiple suprabasal cell layers with largely intact terminal differentiation. In contrast, high-grade dysplasia or CIS in the human consisted of replacement of the entire extent of squamous epithelium by immature basal-type squamous cells, each of which are almost entirely composed of enlarged, anaplastic nuclei with scant cytoplasm (41)
. As such, the grade of neoplastic progression in mouse squamous epithelium was determined by the increase in nuclear:cytoplasmic ratio in individual squamous epithelial cells, the frequency of such cells within the squamous epithelium, and the morphology of the interface between squamous epithelium and the underlying vaginal or cervical stroma.
|
, compare panels 1 and 2). Interspersed in the basal and immediate suprabasal cell layers were individual cells with an increased nuclear:cytoplasmic ratio and nuclear atypia (Fig. 2
, black arrowhead). CIN II lesions contained cells with additional increases in nuclear size, degree of anaplasia, frequency, and distribution of dysplastic cells in the suprabasal layers of the squamous epithelium (Fig. 2
, panel 3). Moreover, the basal aspect of the squamous epithelium was thrown into papillary folds projecting into the underlying vaginal or cervical stroma (Fig. 2
, panel 3). CIN III lesions contained abundant anaplastic cells, some with pronounced increases in nuclear size (Fig. 2
, panel 4). CIS demonstrated a pronounced degree of remodeling and undulation of the epithelial-stromal border (Fig. 2
, panel 5), and most of the cellular features of well-differentiated squamous carcinoma, with retention of an intact basement membrane without evidence of microinvasion on serial sections. Invasive, well-differentiated squamous cancers consisted of invading nests of cancer cells arranged in part in keratin pearls (Ref. 42
; Fig. 2
, panel 6). In contrast with transgenic mice, continuous estrogen treatment for 6 months in nontransgenic mice produced squamous epithelial hyperplasia similar to persistent estrus, with an increase in the number of squamous epithelial cell layers, occasional basal cell mitotic figures, and preservation of differentiated suprabasal keratinocytes (Fig. 2
, panel 1).
Distribution of Severity of Histopathological Lesions According to Expressed HPV ORF.
To additionally elucidate differences in the biology of E6 and E7 functions in the estrogen-treated cervix, we applied our CIN grading system to determine the extent of neoplastic progression in transgenic mice expressing the E6 or E7 oncoproteins individually and in combination (Fig. 3)
. Invasive cancer or high-grade dysplastic lesions (CIN III and CIS) occurred in 100% of K14-E7, K14-E6:E7, and K14-HPV16 transgenic mice, whereas dysplastic lesions in K14-E6 transgenic mice did not progress beyond CIN II (Fig. 3A)
. Histologically, CIN III and CIS lesions in K14-E7 (Fig. 3
, panel 3) and K14-E6:E7 (Fig. 3
, panel 4) transgenic mice were similar to those observed in K14-HPV16 mice, except that the degree of nuclear anaplasia and increase in nuclear:cytoplasmic ratio was greater in transgenic mice expressing the entire early region (see Fig. 2
, panels 4 and 5). In K14-E6 transgenic mice, the cervical and vaginal squamous epithelium was thicker compared with nontransgenic control mice, with occasional dysplastic cells with increased nuclear size and anaplasia (see arrow Fig. 3
B, panel 2). The epithelial-stromal border of K14-E6 transgenic mice was more undulating compared with the linear border in nontransgenic controls; however, the marked papillary stromal projections of squamous epithelium characteristic of K14-E7 or K14-HPV16 transgenic mice (Fig. 2
, panel 3) was absent in K14-E6 transgenic mice. Cancer formation and growth also appeared to distinctively vary according to which HPV ORFs were expressed in the estrogen-treated cervix. Cervical and vaginal cancers in K14-E7 transgenic mice were microinvasive, small, and multifocal (Fig. 3
B, panel 5). Cervical malignancies in K14-E6:E7 and K14-HPV16 transgenic mice consisted of multiple nests of invasive squamous cells dispersed over a large area of cervical stromal tissue (Fig. 3
B, panel 6). The distinctive biological differences in cancer growth in transgenic mice expressing E7 versus both E6 and E7 support a role for E6 as a late-stage progression factor for cervical cancer in this model.
|
. The mean number of S phase squamous epithelial cells progressively increased at each time point in K14-HPV16 and K14-E7 compared with either K14-E6 transgenic mice or nontransgenic controls (Fig. 4A)
. Moreover, S phase cells were present in multiple cell layers in the squamous epithelia of both K14-E7 (Fig. 4
B, panel 5) and K14-HPV16 (Fig. 4
B, panel 7) transgenic mice, particularly abundant in each papillary projection of squamous epithelium into the cervical (and vaginal) stroma. In contrast, BrdUrd-positive cells were restricted to the basal and immediate suprabasal cell layers in cervical squamous epithelium of K14-E6 transgenic mice and nontransgenic controls (Fig. 4
B, panels 1 and 2).
|
Real-time RT-PCR Analysis of E6/E6* Transgene Expression.
One explanation for the lack of neoplastic progression in the K14-E6 transgenic mice could have been "insufficient" transgene expression in cervical epithelium. To determine transgene mRNA expression, quantitative real-time RT-PCR was performed using PCR primers and a TaqMan probe amplifying both full-length and E6* mRNA on total RNA isolated from the lower reproductive tract tissue block ("Materials and Methods"). E6 ORF mRNA expression of homozygous K14-E6 transgenic mice was 13–24-fold elevated compared with heterozygous K14-E7 and HPV transgenic mice (Fig. 5)
. Development of complete carcinogenesis in K14-E7 and K14-HPV transgenic mice despite a threshold level of transgene expression highlights the cocarcinogenic potency of the E7 oncoprotein in conjunction with estrogen in cervical squamous epithelium. Thus, lack of carcinogenic progression in the K14-E6 transgenic mice cannot be explained by insufficient transgene expression.
|
, panels 3 and 1, respectively), with an additional induction of p53 protein in cervical cancers from this transgenic lineage (Fig. 6
, panel 4). However, the intensity of p53 immunostaining in these cancers remained much less than that from islet cell carcinomas derived from transgenic mice expressing SV40 T-antigen behind a rat insulin promoter (Fig. 5
, panel 6). These data are consistent with coordinate p53 up-regulation by E7 oncoprotein (46)
, rather than stabilization and accumulation of a mutant p53 protein (47)
. In marked contrast to p53 induction in K14-E7 transgenic mice, there was no detectable p53 immunostaining in CINI/II lesions in K14-E6 or invasive transformation zone squamous cancers in K14-HPV16 transgenic mice (Fig. 6
, panels 2 and 5) consistent with E6-E6AP mediated p53 protein destruction (48
, 49)
.
|
Centrosome Abnormalities in Cervical Neoplasia and Cancers in Transgenic Mice.
Centrosome abnormalities have been reported previously in a wide array of human cancers including cervical neoplasia (37)
. In particular, both E6 and E7 viral oncoproteins have been shown to induce an increase in centrosome number albeit via different mechanisms (51
, 52)
. However, one mechanism for the differential carcinogenic progression in K14-E6 transgenic mice versus transgenic counterparts expressing functional E7 could be differential induction of centrosome abnormalities in the latter compared with the former. To determine centrosome numbers, immunofluorescent staining for pericentrin, a protein component of the centrosome, was performed on deparaffinized tissue sections from transgenic and nontransgenic mice treated with estrogen for 6 months (Fig. 7
A, top and bottom panels). This analysis yielded several surprising results. First, premalignant lesions from each group of transgenic mice evidenced an increase in centrosome number compared with nontransgenic controls (Fig. 7B)
. As such, CINII lesions in K14-E6, and CINIII/CIS lesions in K14-E7 or K14-HPV16 transgenic mice each displayed a similar 6–8% incidence of cells with centrosome number elevations compared with 2.5% frequency in estrogen-treated nontransgenic mice. Second, the frequency of elevated centrosome copy number markedly increased in cervical cancers from estrogen-treated K14-HPV16 transgenic mice compared with either precursor CIN III/CIS lesions in mice of the same genotype or to malignant cells in cancers evident in K14-E7 transgenic counterparts (17 versus 9.5%; Fig. 7B
). Whereas the differences between centrosome abnormalities in premalignant lesions in each genotype of transgenic mice were not significantly different from nontransgenic controls, the difference between centrosome number in cancers of HPV transgenic mice compared with nontransgenic controls was significantly different. These data plus the trend in the comparison between the K14-HPV versus the K14-E7 transgenic mice support a hypotheses that E6 functions as a late-stage, malignant progression factor, in combination with other carcinogenic effectors, such as E7, or chemical carcinogens (see "Discussion").
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
One notable finding in our study was the remarkable potency of the HPV16 E7 oncogene as an estrogen cocarcinogen in the mouse cervix. E7 expression in conjunction with estrogen was sufficient to induce the requisite components of carcinogenic progression including hyperproliferation, inhibition of epithelial differentiation leading to high-grade dysplasia, insufficient compensatory induction of apoptosis, and centrosome copy number elevations (53) . Hyperproliferation could be explained by a repertoire of E7 functions shown previously to dysregulate the cell cycle including: E7 binding, destabilization, and consequent destruction of pRB, p107, and p130, sequestration of Mi2 histone deacetylase, induction of c-Myc, and abrogation of transforming growth factor ß inhibitory signaling (5) . It is important to keep in mind that estrogen itself increased reproductive tract squamous epithelial proliferation (54) . Our data suggested that E7 complemented and amplified an underlying proliferative stimulus propagated by the estrogen alone. One molecular explanation for this finding was induction of c-Jun/AP-1 activity by estrogen and HPV16 E7. In particular, HPV16/18 E7 has been shown to increase c-Jun protein levels and transcriptional activity (11) , and ligand-bound estrogen receptor has also been shown to activate AP-1 mediated transcription by indirect "tethered" binding to AP-1 complexes on DNA (55) . In addition to hyperproliferation, the development of neoplastic progression via high-grade dysplasia and ultimately malignancy in K14-E7 transgenic mice was likely facilitated by a combination of p21 inactivation, which uncoupled squamous epithelial terminal differentiation from proliferation (44) , and centrosome copy number dysregulation, which produced multipolar mitotic spindles, abnormal nuclear divisions, and potentially, cellular multinucleation and aneuploidy (52 , 56 , 57) . Estrogen itself could induce direct DNA damage via its catechol metabolites (58, 59, 60) , which could have been the potential "initiating" step in our model of cervical carcinogenesis.
The absence of concomitant apoptosis, at any stage of cervical neoplastic progression in estrogen-treated K14-E7 transgenic mice, was also surprising. Previous studies demonstrated coordinate induction of apoptosis in both embryonic lens and retina, and the epidermis of transgenic mice with E7 expression targeted to each of these epithelia (38
, 50)
. Whereas induction of apoptosis in embryonic tissue by E7 could have been because of increased sensitivity of developing epithelia to E7 functions, the differential apoptosis in skin compared with cervix merits additional discussion. In the skin, apoptosis was primarily detected in the granular and stratum corneum layers. These epidermal layers contained nucleated cells in the transgenic mice (parakeratosis) compared with anucleate cells in nontransgenic controls (38
, 50) . As such, apoptosis in the skin model could have been because of persistent E7 expression in keratinocytes, in which the terminal differentiation program was already activated. However, the granular and corneal squamous epithelial cell layers do not exist in the cervix. Therefore, lack of susceptible target cells could be one mechanism underlying the lack of apoptosis in the cervix of estrogen-treated K14-E7 transgenic mice. In addition, the inability to detect apoptosis in the basal layer of cervical squamous epithelium in these same transgenic mice could have been because of paracrine elaboration of survival factors, such as insulin-like growth factor I, epidermal growth factor, platelet-derived growth factor, and transforming growth factor 
, from the subjacent stroma (61, 62, 63, 64, 65, 66)
. These survival factors could have balanced the unopposed expression of E7 in the vaginal and cervical basal squamous epithelial cells of estrogen-treated K14-E7 transgenic mice. Importantly, prior work in the skin has also demonstrated functional p53 in the squamous epithelium of K14-E7 mice. Specifically, coordinate induction of p53 and its target p21 was maintained in the keratinocytes within irradiated transgenic skin (40)
. As such, the lack of apoptosis in the estrogen-treated cervices of K14-E7 transgenic mice was likely not because of failure of p53 function. Conversely, our immunohistochemical analysis of p53 expression in cervical carcinomas in K14-E7 transgenic mice failed to demonstrate stabilization of p53 protein to a level consistent with independent mutation and oncogenic gain of function.
The inability of E6 to affect neoplastic progression beyond low-grade dysplasia was another surprising finding of this study considering the repertoire and functions of cellular targets of this viral oncoprotein (7) . E6 binding to the calcium-binding protein, E6BP (67 , 68) , has been shown to inhibit keratinocyte terminal differentiation in cell culture and could have been a potential mechanism for uncoupling terminal differentiation from proliferation. E6 complex formation with either the human homologue of the Drosophila discs-large protein (69) or paxillin (70) could each disrupt basal cell polarity potentially increasing proliferation and facilitating dysplasia by permitting escape of these proliferating cells into the upper stratified squamous epithelial layers. E6-mediated activation of the human telomerase reverse transcriptase component of telomerase (71 , 72) could also have contributed to in vivo immortalization of squamous epithelial cells, allowing for persistence of transformed cells and their subsequent progression to malignancy. However, the lack of induction of cervical epithelial cell proliferation and high-grade dysplasia, suggests that either these cellular targets do not interact with E6 in the tissue and hormonal context of this model or that the murine cervix is resistant to their dysregulation despite perturbation by the oncoprotein. Certainly the lack of a robust neoplastic phenotype cannot be explained by either lack of reproductive tract E6 expression or even function, as both p53 expression was decreased and centrosome copy number elevated in these transgenic mice. However, our results clearly demonstrated that these particular E6 functions by themselves were insufficient to support neoplastic progression in the face of the inability of the oncoprotein by itself to affect cervical epithelial proliferation and differentiation.
Similarly, the discordance in neoplastic potency between HPV E7 and E6 in the face of similar elevations in centrosome number in this model of cervical carcinogenesis reinforced our previous conclusions regarding the relevance of this parameter of genomic instability, per se, as a predictor of ultimate neoplastic progression. Our current model is that centrosome elevations arise by two different mechanisms in E6- or E7-expressing cells (56
, 57)
. In E6-expressing cells, centrosome abnormalities were detected in polynucleated cells, many of which were senescent. These data were consistent with the detection of occasional cells with markedly increased nuclear:cytoplasmic ratio in the cervical epithelium of estrogen-treated K14-E6 transgenic mice (Fig. 3)
. In contrast, centrosome abnormalities in E7-expressing cells were detected in proliferative diploid cells before additional features of genomic instability (57)
, suggesting that E7 itself is an independent driving force for the induction of genomic stability in cells with the potential for neoplastic progression. The current study is the first determination that this concept of parallel alteration in centrosome number control indeed occurs in the cervix expressing either HPV oncogene and that these centrosome abnormalities are associated with different neoplastic consequences.
However, our results with individual expression of E7 or E6 in the estrogen-treated cervix contrasted sharply to that seen in the skin of the same transgenic mice. In skin, both E7 and E6 were potent inducers of hyperproliferation (Fig. 4
; Refs. 32
, 33
). Both oncogenes dysregulated terminal differentiation and allowed escape of proliferating cells to the upper epidermal layers. Most strikingly, the carcinogenic potential of these oncogenes was reversed in skin. There, E7, alone or in combination with two-stage chemical carcinogens, induced predominantly benign papillomas, possessing a low frequency of malignant conversion (32
, 39)
, whereas epidermal cancer was the predominant lesion in transgenic mice expressing E6 alone (33)
or in combination with chemical carcinogens (39)
. Additional examination of the data from these experiments also supported a role for E6 as a late-stage malignant progression factor in epidermis, reminiscent of our current results in the mouse cervix, albeit when both HPV oncogenes were coexpressed. In skin, the combination of E6 and chemical carcinogens increased the frequency of spindle-cell compared with well-differentiated epidermal carcinomas. Spindle-cell cancer is a more advanced stage of malignant progression in mouse skin carcinogenesis (73)
. Despite this similarity, the overall discordance in solo carcinogenic potency of E7 and E6 in mouse skin and cervix highlights the importance of cell-type, tissue, and hormonal context in the modeling of human disease by transgenic models.
However, despite our compelling demonstration of distinctive E7 and E6 functions in the mouse cervix, these viral oncogenes are invariably coexpressed in human HPV disease (2) . Both our current and prior work is concordant with a role for E6 as a late-stage malignant progression factor in both cervix and skin of double E6:E7 transgenic mice (39) . In the cervix, E6 and E7 coexpression contributes to the formation of large, extensively invasive cancers in double-transgenic mice, and the enhanced chromosomal copy number dysregulation in the malignancies in these mice compared with counterparts expressing E7 alone.
In summary, this study highlights the distinctive functions of the E7 and E6 oncogenes in cervical carcinogenesis with the caveats that several species-specific and technological differences exist between our model and human cervical neoplastic or malignant disease. Collectively, our data suggest that both induction of squamous epithelial proliferation and perturbation of centrosome copy number contribute to progression to invasive cervical cancer. Moreover, whereas E6 appears to have a low neoplastic penetrance alone, it facilitates the genesis of aggressively malignant cervical cancers in conjunction with E7. Future work will genetically dissect discrete E7 and E6 oncoprotein functions to elucidate the mechanisms underlying their cooperation in cervical carcinogenesis.
|
FOOTNOTES |
|---|
1 Supported by NIH Grants R01-CA71398, NO1-CN-95115, NCI R01 CA 66980, and NCI R01 CA22443. 
2 To whom requests for reprints should be addressed, at Cancer Genetics Program, University of California San Francisco Comprehensive Cancer Center, 2340 Sutter Street, Room N419, San Francisco, CA 94143-0808. Phone: (415) 502-3292; Fax: (415) 476-8218; E-mail: jarbeit{at}cc.ucsf.edu
3 The abbreviations used are: HPV, human papillomavirus; ORF, open reading frame; ttl, translation termination linker; RT-PCR, reverse transcription-PCR; BrdUrd, bromodeoxyuridine; CIN, cervical intraepithelial neoplasia; CIS, carcinoma in situ; AP-1, activator protein.
Received 3/10/03. Revised 6/ 9/03. Accepted 6/10/03.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
.. Genes Dev., 15: 2520-2532, 2001.- and ß-subunits in the mouse uterus and vagina: potential mediators of estrogen action. Endocrinology, 136: 2325-2340, 1995.[Abstract]
precursors in the mouse uterus during the periimplantation period and after steroid hormone treatments.. Biol. Reprod., 50: 481-491, 1994.[Abstract]
, and epidermal growth factor receptor localization in the baboon (Papio anubis) uterus during the menstrual cycle and early pregnancy.. J. Soc. Gynecol. Investig., 1: 277-284, 1994.[Medline]
This article has been cited by other articles:
|
|
 |
|
  S.-H. Chung and P. F. Lambert Prevention and treatment of cervical cancer in mice using estrogen receptor antagonists PNAS, November 17, 2009; 106(46): 19467 - 19472. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. W. Sepkovic, J. Stein, A. D. Carlisle, H. B. Ksieski, K. Auborn, and H. L. Bradlow Diindolylmethane Inhibits Cervical Dysplasia, Alters Estrogen Metabolism, and Enhances Immune Response in the K14-HPV16 Transgenic Mouse Model Cancer Epidemiol. Biomarkers Prev., November 1, 2009; 18(11): 2957 - 2964. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. R. Garbow, A. C. Santeford, J. R. Anderson, J. A. Engelbach, and J. M. Arbeit Magnetic Resonance Imaging Defines Cervicovaginal Anatomy, Cancer, and VEGF Trap Antiangiogenic Efficacy in Estrogen-Treated K14-HPV16 Transgenic Mice Cancer Res., October 15, 2009; 69(20): 7945 - 7952. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-K. Shin, S. Balsitis, T. Brake, and P. F. Lambert Human Papillomavirus E7 Oncoprotein Overrides the Tumor Suppressor Activity of p21Cip1 in Cervical Carcinogenesis Cancer Res., July 15, 2009; 69(14): 5656 - 5663. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. Katzenellenbogen, P. Vliet-Gregg, M. Xu, and D. A. Galloway NFX1-123 Increases hTERT Expression and Telomerase Activity Posttranscriptionally in Human Papillomavirus Type 16 E6 Keratinocytes J. Virol., July 1, 2009; 83(13): 6446 - 6456. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. F. Jabbar, L. Abrams, A. Glick, and P. F. Lambert Persistence of High-Grade Cervical Dysplasia and Cervical Cancer Requires the Continuous Expression of the Human Papillomavirus Type 16 E7 Oncogene Cancer Res., May 15, 2009; 69(10): 4407 - 4414. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. L. Reimers, W. F. Anderson, P. S. Rosenberg, D. E. Henson, and P. E. Castle Etiologic Heterogeneity for Cervical Carcinoma by Histopathologic Type, Using Comparative Age-Period-Cohort Models Cancer Epidemiol. Biomarkers Prev., March 1, 2009; 18(3): 792 - 800. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-H. Chung, K. Wiedmeyer, A. Shai, K. S. Korach, and P. F. Lambert Requirement for Estrogen Receptor {alpha} in a Mouse Model for Human Papillomavirus-Associated Cervical Cancer Cancer Res., December 1, 2008; 68(23): 9928 - 9934. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Shai, H. C. Pitot, and P. F. Lambert p53 Loss Synergizes with Estrogen and Papillomaviral Oncogenes to Induce Cervical and Breast Cancers Cancer Res., April 15, 2008; 68(8): 2622 - 2631. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Strati and P. F. Lambert Role of Rb-Dependent and Rb-Independent Functions of Papillomavirus E7 Oncogene in Head and Neck Cancer Cancer Res., December 15, 2007; 67(24): 11585 - 11593. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Huh, X. Zhou, H. Hayakawa, J.-Y. Cho, T. A. Libermann, J. Jin, J. Wade Harper, and K. Munger Human Papillomavirus Type 16 E7 Oncoprotein Associates with the Cullin 2 Ubiquitin Ligase Complex, Which Contributes to Degradation of the Retinoblastoma Tumor Suppressor J. Virol., September 15, 2007; 81(18): 9737 - 9747. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. H. Lu, J. D. Wright, B. Belt, R. D. Cardiff, and J. M. Arbeit Hypoxia-Inducible Factor-1 Facilitates Cervical Cancer Progression in Human Papillomavirus Type 16 Transgenic Mice Am. J. Pathol., August 1, 2007; 171(2): 667 - 681. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Pyeon, M. A. Newton, P. F. Lambert, J. A. den Boon, S. Sengupta, C. J. Marsit, C. D. Woodworth, J. P. Connor, T. H. Haugen, E. M. Smith, et al. Fundamental Differences in Cell Cycle Deregulation in Human Papillomavirus-Positive and Human Papillomavirus-Negative Head/Neck and Cervical Cancers Cancer Res., May 15, 2007; 67(10): 4605 - 4619. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. H. Storrs and S. J. Silverstein PATJ, a Tight Junction-Associated PDZ Protein, Is a Novel Degradation Target of High-Risk Human Papillomavirus E6 and the Alternatively Spliced Isoform 18 E6 J. Virol., April 15, 2007; 81(8): 4080 - 4090. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. P. Matthews, A. M. Birkholz, A. R. Baker, C. M. Perella, G. R. Beck Jr., M. R. Young, and N. H. Colburn Dominant-Negative Activator Protein 1 (TAM67) Targets Cyclooxygenase-2 and Osteopontin under Conditions in which It Specifically Inhibits Tumorigenesis Cancer Res., March 15, 2007; 67(6): 2430 - 2438. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Shai, T. Brake, C. Somoza, and P. F. Lambert The Human Papillomavirus E6 Oncogene Dysregulates the Cell Cycle and Contributes to Cervical Carcinogenesis through Two Independent Activities Cancer Res., February 15, 2007; 67(4): 1626 - 1635. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Balsitis, F. Dick, N. Dyson, and P. F. Lambert Critical Roles for Non-pRb Targets of Human Papillomavirus Type 16 E7 in Cervical Carcinogenesis Cancer Res., October 1, 2006; 66(19): 9393 - 9400. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J P A Baak, A-J Kruse, S J Robboy, E A M Janssen, B van Diermen, and I Skaland Dynamic behavioural interpretation of cervical intraepithelial neoplasia with molecular biomarkers J. Clin. Pathol., October 1, 2006; 59(10): 1017 - 1028. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Strati, H. C. Pitot, and P. F. Lambert Identification of biomarkers that distinguish human papillomavirus (HPV)-positive versus HPV-negative head and neck cancers in a mouse model PNAS, September 19, 2006; 103(38): 14152 - 14157. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Gammoh, H. S. Grm, P. Massimi, and L. Banks Regulation of Human Papillomavirus Type 16 E7 Activity through Direct Protein Interaction with the E2 Transcriptional Activator J. Virol., February 15, 2006; 80(4): 1787 - 1797. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L.-H. Yih, Y.-Y. Tseng, Y.-C. Wu, and T.-C. Lee Induction of Centrosome Amplification during Arsenite-Induced Mitotic Arrest in CGL-2 Cells Cancer Res., February 15, 2006; 66(4): 2098 - 2106. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Balsitis, F. Dick, D. Lee, L. Farrell, R. K. Hyde, A. E. Griep, N. Dyson, and P. F. Lambert Examination of the pRb-Dependent and pRb-Independent Functions of E7 In Vivo J. Virol., September 1, 2005; 79(17): 11392 - 11402. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Genther Williams, G. L. Disbrow, R. Schlegel, D. Lee, D. W. Threadgill, and P. F. Lambert Requirement of Epidermal Growth Factor Receptor for Hyperplasia Induced by E5, a High-Risk Human Papillomavirus Oncogene Cancer Res., August 1, 2005; 65(15): 6534 - 6542. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Walker, H. Adle-Biassette, P. Madelenat, D. Henin, and T. Lehy Increased Apoptosis in Cervical Intraepithelial Neoplasia Associated with HIV Infection: Implication of Oncogenic Human Papillomavirus, Caspases, and Langerhans Cells Clin. Cancer Res., April 1, 2005; 11(7): 2451 - 2458. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Brake and P. F. Lambert Estrogen contributes to the onset, persistence, and malignant progression of cervical cancer in a human papillomavirus-transgenic mouse model PNAS, February 15, 2005; 102(7): 2490 - 2495. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Matsumoto, G. R. Leggatt, J. Zhong, X. Liu, R. L. de Kluyver, T. Peters, G. J. P. Fernando, A. Liem, P. F. Lambert, and I. H. Frazer Impaired Antigen Presentation and Effectiveness of Combined Active/Passive Immunotherapy for Epithelial Tumors J Natl Cancer Inst, November 3, 2004; 96(21): 1611 - 1619. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Munger, A. Baldwin, K. M. Edwards, H. Hayakawa, C. L. Nguyen, M. Owens, M. Grace, and K. Huh Mechanisms of Human Papillomavirus-Induced Oncogenesis J. Virol., November 1, 2004; 78(21): 11451 - 11460. [Full Text] [PDF]
|
|
|
|
|
|
K. Kawamura, H. Izumi, Z. Ma, R. Ikeda, M. Moriyama, T. Tanaka, T. Nojima, L. S. Levin, K. Fujikawa-Yamamoto, K. Suzuki, et al. Induction of Centrosome Amplification and Chromosome Instability in Human Bladder Cancer Cells by p53 Mutation and Cyclin E Overexpression Cancer Res., July 15, 2004; 64(14): 4800 - 4809. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Longworth and L. A. Laimins Pathogenesis of Human Papillomaviruses in Differentiating Epithelia Microbiol. Mol. Biol. Rev., June 1, 2004; 68(2): 362 - 372. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. Balsitis, J. Sage, S. Duensing, K. Munger, T. Jacks, and P. F. Lambert Recapitulation of the Effects of the Human Papillomavirus Type 16 E7 Oncogene on Mouse Epithelium by Somatic Rb Deletion and Detection of pRb-Independent Effects of E7 In Vivo Mol. Cell. Biol., December 15, 2003; 23(24): 9094 - 9103. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Brake, J. P. Connor, D. G. Petereit, and P. F. Lambert Comparative Analysis of Cervical Cancer in Women and in a Human Papillomavirus-Transgenic Mouse Model: Identification of Minichromosome Maintenance Protein 7 as an Informative Biomarker for Human Cervical Cancer Cancer Res., December 1, 2003; 63(23): 8173 - 8180. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. R. Lowy and M. L. Gillison A New Link Between Fanconi Anemia and Human Papillomavirus-Associated Malignancies J Natl Cancer Inst, November 19, 2003; 95(22): 1648 - 1650. [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 |
