A Model for Angiogenesis in HPV-Mediated Cervical Neoplasia

Janice Matthews-Greer, PhD^{a, b}

Arrigo DeBenedetti, PhD^c

Angela Tucker, MS^d

Susanna Dempsey^a

Destin Black^d

Elba Turbat-Herrera, MD^{a, d, e}

^aDepartment of Pathology

^bDepartment of Pediatrics

^cDepartment of Biochemistry

^dDepartment of Obstetrics and Gynecology

Louisiana State University Health Sciences Center

1501 Kings Highway

Shreveport, LA 71130-3932

 

^eDepartment of Pathology

Overton Brooks Veterans Medical Center, Shreveport, LA

 

 

KEY WORDS: HPV, HPV E7, HPV E6, HPV 16, HPV 18, cervical dysplasia, cervical cancer, angiogenesis, VEGF, eIF4E

ABSTRACT

Almost all cervical cancers are associated with human papillomaviruses (HPVs) belonging to the class of “high (cancer)-risk” HPVs (e.g., HPV 16, HPV 18). This is thought to be due to an increased expression of HPV oncogenes (E6 and E7) when HPV is inserted into the host genome in addition to higher binding affinities of E6 and E7 for tumor suppressor gene products, p53 and the retinoblastoma protein (pRB), respectively–all properties of these high-risk types. Angiogenesis has been studied in a variety of cancers, but none that are so intimately associated with an oncogenic virus. Cervical cancer progression and recurrence are associated with an increase in angiogenesis and angiogenesis-promoting factors, such as vascular endothelial growth factor F (VEGF). Our objective was to determine the efficacy of using angiogenesis as a marker for cervical cancer and to determine what role, if any, HPV infection might have in this process. We found that, like other cancers, such as breast carcinoma, another angiogenesis-promoting protein, eukaryotic translation initiation factor 4E (eIF4E) is increased in cervical neoplasia. However, we also found elevations in eIF4E and VEGF expression and increased mean vessel counts (MVC) in cervical dysplasia as well as in carcinoma. Moreover, HPV-infected cell lines transfected with eIF4E produce increased amounts of E7 oncoprotein as compared with nontransfected cell lines. We conclude that HPV plays a role in angiogenesis in cervical neoplasia. We propose eIF4E to be a marker for early cervical cancer. In addition, we propose that the HPV E7 protein plays a role in the eIF4E/c-myc cascade and, therefore, is involved in the early events of angiogenesis. A model is suggested for such an interaction between eIF4E and HPV E7.

INTRODUCTION

Cervical cancer is a leading cause of morbidity and mortality worldwide. Widespread screening in affluent countries has reduced the incidence of death due to the disease, but nonablative treatment options are limited. Like hepatitis B virus-associated liver cancer, cervical cancer has an infectious etiology. Almost all cervical cancers are associated with persistent infection with certain high-risk HPV types. These oncogenic HPVs can integrate into the host genome. Their ability to cause cancer is believed to be due to both an increased expression of HPV oncogene (E7 and E6) mRNA when HPV is integrated into the host DNA1, 2 as well as an increased affinity of the high-risk HPV types for the products of tumor suppressor genes.3 E7 binds to and inactivates pRB; E6 binds to and degrades p53.4

eIF4E is rate-limiting for the translation of mRNAs with a high degree of secondary structure.5 Examples of this class of “weak mRNAs” include those that play a critical role in growth and differentiation,6 such as c-myc,7 the pro-angiogenic VEGF,8 and that reported by Stacey et al,9 HPV E7. In turn, eIF4E is one of the few transcriptional targets of c-myc,10 creating a positive feedback loop between eIF4E and c-myc (eIF4E increases c-myc translationally, whereas
c-myc increases eIF4E transcriptionally). eIF4E’s role in malignancy appears to be the action of increasing translation efficiency of oncogene transcripts, resulting in an overexpression.11 This strongly suggests that eIF4E’s elevation may be a harbinger of malignancy.

The association between angiogenesis, angiogenic factors, and tumor dissemination has been studied extensively in a variety of tumors.7,8,11–16 However, little is known about the contribution of oncogenic viruses in this process, particularly viruses such as HPV, which is deemed causally associated with cervical cancer. Further study is necessary to elucidate the role of angiogenesis during HPV-associated neoplasia. These data could lead to new screening paradigms as well as new treatment modalities for these cancers.

It is suggested that the onset of angiogenesis in cervical cancer occurs very early during premalignant stages17 and that an HPV oncoprotein may be responsible for this process.18 Angiogenesis also is reported to be predictive of cervical cancer recurrence.19 These reports prompted us to believe it feasible that VEGF and other angiogenic markers would be present in early HPV lesions. Since HPV E7 upregulates c-myc, which upregulates eIF4E, which in turn increases VEGF (and angiogenesis), we hypothesized that HPV E7 must upregulate angiogenesis in the infected cervix. Our objectives were both to investigate the potential for angiogenic markers for screening cervical neoplasia and to determine what role, if any, HPV E7 oncoprotein might have in the angiogenesis process.

MATERIALS AND METHODS

Immunohistochemistry

For eIF4E staining, a 1:200 dilution of polyclonal rabbit anti-eIF4E (produced in the laboratory of ADB) was used as a primary antibody. Staining intensity was analyzed using Optimas Image Analysis software. The numbers of positively stained cells in five high-power fields (40X) for each section were counted. VEGF, HPV E6, and HPV E7 proteins were labeled with monoclonal antibodies purchased commercially (rabbit anti-human VEGF from BioGenex Laboratories, Inc. San Ramon, CA; mouse anti-HPV-16/18 E6 from Chemicon International, Inc., Temecula, CA; and mouse anti-HPV-16 E7 from Zymed Laboratories, So. San Francisco, CA). (HPV 16 is the most common type of HPV that we see in our population; however, it is recognized that tissues infected with HPV types other than HPV 16 may yield false negative results.) Primary antibody dilutions were 1:100. Positive controls were CaSki cells (HPV-infected); negative controls were sections incubated with phosphate buffered saline in place of primary antibody. Staining intensities within the epithelium were scored manually by two independent observers. Granular staining was graded as 0231 and assigned corresponding numerical counts (e.g., 1 for 11). The two counts obtained by the two observers for each section were averaged, and these averages were totaled for each diagnosis (e.g., normal, dysplasia, or carcinoma). The reported mean stain intensity was the average of these total counts.

Mean Vessel Counts

At LSUHSC-S, we perform a high number of procedures for the diagnosis and treatment of cervical disease. All of the tissues removed are archived within the Department of Pathology. From these, we chose paraffin blocks collected from women with the full range of progressive histologic changes—normal, dysplastic, and cancer—to measure angiogenesis. For measuring mean vessel counts, sections were stained for CD34 using rabbit anti-CD34 (BioGenex) to enhance visualization of newly formed vessels.20 Neovascularization was quantitated by two investigators working in tandem and reported as the MVC using a Chalkley UV/DF/TR eyepiece (Olympus America, Inc., Melville, NY) as described previously.21 The sections were scanned first on low power (10X) to find a representative area, prior to insertion of the Chalkley. The eyepiece provides a random arrangement of dots. Any capillary vessel touching one of the dots is counted. Each MVC represents the mean count from 3 separate analyses.

Transfection of Cell Lines with an eIF4E Vector

Immortalized cells, HeLa (H) containing integrated HPV 18, SiHa (S) containing integrated HPV 16, and HPV-negative cervical carcinoma cells, C-33A (C), were obtained from the American Type Culture Collection (Manassas, VA) and maintained as recommended. The eIF4E vector (E) is a bacterial-mammalian cell shuttle vector previously described22 and maintained in the laboratory of ADB. For eIF4E DNA transfer, cells cultured to 60% confluence in MEM with 10% (v/v) FBS were transfected with 20 to 30 µg of plasmid DNA using the Gene Porter II reagent (Gene Therapy Systems, Inc., San Diego, CA). Selection and maintenance was with Geneticin (Gibco Invitrogen Corporation, Carlsbad, CA), 250 µg/mL and 50 µg/mL, respectively. eIF4E-transfected HeLa, SiHa, and C33a cell lines are designated as HE, SE, and CE, respectively.

Gel Electrophoresis with Amplified Western Blotting

Cell lysates, prepared from cell lines as described,9 were assayed for total protein concentration using Bio-Rad Protein Assay (Bio-Rad Laboratories, Hercules, CA). Standardized aliquots of 30 µg/mL were subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and the gels either silver stained (Bio-Rad Silver Stain Kit) or transferred to polyvinylidene difluoride membrane (PVDF) for amplified Western blots. Blots were reacted with anti-HPV-18 E7 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) or anti-c-myc antibody (Santa Cruz) and detected using the amplified RAD^TM (Bio-Rad) protocol as described in the manufacturer’s insert.

Immunoprecipitation of Cell Lysates with Specific Antisera

Immunoprecipitation was performed using either anti-eIF4E or anti-HPV-16 E7-agarose conjugated antisera (Santa Cruz) as suggested by the manufacturer (Santa Cruz). Cell lysates were prepared from 100-mm plates of cells at 90% confluence (18 to 24 hours) in RIPA buffer containing PMSF, aprotinin, and sodium orthovanadate using a syringe with a 21-gauge needle. After incubation, a volume of lysate equaling 100 µg total protein/mL (as determined by Bradford reagent from Sigma-Aldrich Corp., St. Louis, MO) was treated with either 15 µL of anti-eIF4E or 5 µL of anti-HPV-16 E7-agarose (Santa Cruz). After 1 hour, 10 µL of protein G-agarose was added to the lysate containing anti-eIF4E. Lysates were then incubated overnight. To the pellets, 40 µL of electrophoresis sample buffer was added and equal volumes loaded onto the gel. Separated proteins on SDS-PAGE gels were either silver-stained (for eIF4E detection) or transferred to nitrocellulose, reacted with anti-HPV-16 E7 (Santa Cruz) followed by a secondary antibody (anti-mouse-HRP, Santa Cruz) incubation, and visualized by chemiluminescent development of X-ray film.

RESULTS

Markers of angiogenesis are present in cervical dysplasia as well as carcinoma. As seen in Table 1, mean vessel counts were significantly increased in early cervical neoplasia. Normal cervical tissues had an average MVC of 6.91, compared with dysplastic or cervical carcinoma tissues with a combined average of 9.52. Sections demonstrating dysplasia also yielded higher MVC counts than normal tissues for that cohort. Because angiogenesis was found in dysplastic tissues, these were tested for additional markers of angiogenesis. In sections immunostained with anti-eIF4E (Figure 1), eIF4E was found to be upregulated in dysplastic cervix (Table 2). VEGF was also significantly increased in dysplastic tissues as compared with normal; however, no difference was found between VEGF stain intensities of dysplastic and cancerous cervix (Table 3).

There was an increase in HPV E7 in cervical dysplasia as compared to normal cervical tissue. Potential correlation was found between these markers of angiogenesis and HPV oncoprotein (E6 and E7) expression in tissue investigated. Like mean vessel counts and stain intensities for eIF4E and VEGF cervical dysplasias had significantly increased levels of HPV E7 staining over that of normal tissues (Table 3). Stain intensity for HPV E6 was significantly increased in carcinoma but not in dysplasia as compared with normal. However, E6 staining differences between carcinoma and dysplasia were not statistically significant. HPV E7 staining (Figure 2) was significantly greater in both dysplasia and carcinoma as compared with normal (Table 3).

HPV E7 is elevated in cell lines overexpressing eIF4E. If E7 were increased by eIF4E, then both a mechanism for early angiogenesis in cervical neoplasia as well as an additional explanation for the causal relationship between HPV and squamous cell carcinoma could be surmised. To investigate the possibility that eIF4E might upregulate HPV E7 expression, HeLa, SiHa, and C33a cell lines were transfected with an eIF4E DNA vector. Transfected cell lines were confirmed to express the geneticin-resistance protein by SDS-PAGE (Figure 3A), and to overexpress eIF4E (Figure 3B and 3E) as reported previously for transfected HeLa cells.22 And, as expected from previous reports indicating that elevated eIF4E generates an increase in c-myc protein,6 these transfected cell lines show an elevation in c-myc by amplified Western blot (Figure 3C). C-33A cells, transfected and nontransfected, failed to react with anti-E7 (data not shown). Analysis of transfected HeLa cell lysates (Figure 3D) and transfected SiHa cell lysates (Figure 3F) shows an elevation of HPV E7 protein in those cell lines as determined by amplified blot and immunoprecipitation, respectively.

CaSki cell lines could not be transfected successfully with eIF4E. Although the eIF4E-transfected C33A, HeLa, and SiHa cells appear stable, eIF4E transfection of CaSki cells repeatedly resulted in rapid cell death.

DISCUSSION

Previous reports indicate neovascularization and markers of angiogenesis (eIF4E and VEGF) are increased in various carcinomas not cervical in origin.7,8,12,14,16,23 A notable difference between those cancers and cervical cancer is the presence of an oncogenic virus in our patient population. Believing that one cannot discount the potential contribution of HPV to the process of angiogenesis, we began our investigation of both cervical dysplasias and carcinomas. We conclude from these data that HPV, particularly HPV E7 protein, might contribute to early neoplastic events in cervical tissue. eIF4E is increased not just in cancer but also in early neoplasia in cervical tissue (Table 2). Angiogenesis (measured as MVC in Table 1 and VEGF staining intensity in Table 3) is also an early event. HPV is present in these early lesions (Table 3). These experiments demonstrate that cervical dysplasias had significantly increased levels of eIF4E, VEGF, and HPV E7 over that of normal tissues (Tables 2 and 3). Staining intensity for HPV E6 was significantly increased in carcinoma but not in dysplasia (albeit the comparison between carcinoma and dysplasia was not significant and the sample size is small). In contrast to breast carcinoma, our data suggest eIF4E expression is increased significantly, not just in carcinoma, but also in dysplastic tissue. Others also have postulated that HPV infection increases the neovascularization of tissue.24 This may be due to HPV infection, particularly E7 expression. HPV E7 staining (Table 3) was significantly greater in both dysplasia and carcinoma as compared with normal. Thus, if E7 were increased by eIF4E, or the alternative, if eIF4E were increased by E7, then an explanation of these data could be surmised. To investigate this experimentally, we first addressed the possibility that eIF4E could increase HPV E7 expression in cell lines.

We find that eIF4E-overexpressing (transfected), HPV-infected cells produce higher amounts of E7 than cells that are not transfected. This is demonstrated in two cell lines, one infected with HPV 18 (HeLa cells) and another infected with HPV 16 (SiHa cells) as measured by Western blots (Figure 3D) and immunoprecipitation (Figure 3F), respectively. Thus, it appears that E7 can be upregulated by eIF4E. These data are corroborated by the report of Stacey et al9 showing E7 to be dependent upon eIF4E for efficient translation. In that report, E6 also was shown to be upregulated by eIF4E; however, inasmuch as E6 lacks the mechanistic connection for acting to upregulate eIF4E (described in the model in Figure 4), we believe that E7 may hold the key to our observations. Although the eIF4E-transfected C33A, HeLa, and SiHa cells appear stable, eIF4E transfection of CaSki cells repeatedly resulted in rapid cell death. We interpret this rapid cell death as a potential validation of our hypothesis. In our model, greatly increased levels of HPV E7 should result in apoptosis. (The HPV copy number of CaSki is greatly increased over that of SiHa and is higher than in HeLa cells.)

The E7 proteins of high-risk HPVs complex with pRb, leading to pRb inactivation. In so doing, E7 disrupts the ability of pRb to inhibit transcription factor E2F (its normal substrate), thus releasing E2F.25 Once E2F is released, it is free to induce transcription of growth-related proteins such as c-myc. Thus, E7 increases c-myc transcriptionally. The relationship between E7 and c-myc overexpression is reported in a variety of studies. c-myc has been demonstrated to be significantly increased during cervical carcinogenesis and in primary keratinocytes transfected with HPV oncogenes.13,26–28 c-myc is positively correlated to cell proliferation in precancerous lesions.18 Oral keratinocytes transfected with HPV-16 DNA have higher levels of c-myc mRNAs compared with normal cells,29,30 and transformed cells requiring E7 for the transformed phenotype can be maintained without E7 if an increase in c-myc protein is present.31

It has been suggested previously that the onset of angiogenesis in cervical cancer occurs very early during premalignant stages17 and that an HPV oncoprotein may be responsible for this process.18 Angiogenesis also is reported to be predictive of cervical cancer recurrence.19 These data support our findings. It is logical to expect that VEGF and other angiogenic markers would be present in early HPV lesions. We find increased staining intensity for VEGF and HPV E7 in addition to increased mean vessel counts in early lesions (CIN 1), albeit the number of patients within the subcategories of dysplasia precludes comparison between CIN 1 and later stages of dysplasia (i.e., CIN 2/3). Further immunohistochemical studies, paired with HPV typing, on additional dysplastic blocks are ongoing to investigate potential differences among CIN subcategories. The early events of angiogenesis are difficult to dissect, but it is known that eIF4E increases the translation of both VEGF and c-myc. In addition, c-myc protein upregulates the transcription of eIF4E overexpressing cells and E7 is a likely candidate for the class of weak mRNA that requires excess eIF4E for efficient translation. Taking these into account, and knowing that HPV E7 upregulates the transcription of c-myc, we propose that HPV E7 acts very early in the process of cervical neoplasia in the initiation and promotion of angiogenesis. It is likely that both (a) HPV E7 increases the expression of eIF4E via the increase in c-myc mRNA and that (b) eIF4E increases E7 protein synthesis. Both of these actions occur in vivo to augment the initiation of angiogenesis, and both are less likely in low-risk HPV infections due to the reduction in the amount of E7 protein expressed in those infections. We believe we have found evidence for eIF4E’s upregulation of E7 and are currently investigating the effect of E7 on the level of eIF4E. We find it possible that both events are occurring. We propose a model (Figure 4) in which eIF4E upregulates E7 translationally and E7 upregulates eIF4E transcriptionally through its action on c-myc. The resultant increase in c-myc increases eIF4E, which further amplifies the pathway. All of this culminates into an early increase in angiogenesis through the action of eIF4E’s action on VEGF.

CONCLUSION

We began by investigating the possibility of using eIF4E and other angiogenesis markers as a potential diagnostic tool for HPV-associated neoplastic progression. We found these to be increased in both carcinoma and dysplasia. We are intrigued by the possibility that HPV E7 may contribute to angiogenesis and provide evidence that E7 is increased by eIF4E overexpression. Although it is difficult to delineate any single angiogenic stimulus for all cancers, we believe HPV to be directly involved in the cervix. We propose that E7 also effects an increase in eIF4E through the action of c-myc. Since HPV E7 upregulates c-myc, which upregulates eIF4E, and since eIF4E, in turn, increases VEGF (and angiogenesis), HPV E7 logically would upregulate angiogenesis in the infected cervix. If substantiated, novel therapies for cervical neoplasia could result. Then the contribution of HPV and other oncogenic viruses should be investigated in other cancers, particularly those in which HPV has been found. It is possible that elevated MVCs may mark early HPV-induced neoplastic changes, not just in cervix, but in all HPV-induced carcinomas (e.g., penile, head and neck, vulva, vagina, anus).

REFERENCES

 1. Herber R, Liem A, Pitot H, Lambert PF: Squamous epithelial hyperplasia and carcinoma in mice transgenic for the Human papillomavirus type 16 E7 oncogene. J Virol 70:1873–1881, 1996.

 2. Sanchez-Perez AM, Soriano S, Clarke AR, Gaston K: Disruption of the human papillomavirus type 16 E2 gene protects cervical carcinoma cells from E2F-induced apoptosis. J Gen Virol 78:3009–3018, 1997.

 3. Barbosa MS: The oncogenic role of human papillomavirus proteins. Crit Rev Oncogen 7:1–18, 1996.

 4. Arends MJ, Buckley CH, Wells M: Aetiology, pathogenesis, and pathology of cervical neoplasia. J Clin Path 51:96–103, 1998.

 5. Lazaris-Karatzas A, Sonenberg N: The mRNA 5’ cap-binding protein, eIF-4E, cooperates with v-myc or E1A in the transformation of primary rodent fibroblasts. Molec Cell Biol 12: 1234–1238, 1992.

 6. Kleijn M, Scheper GC, Voorma HO, Thomas AAM: Regulation of translation initiation factors by signal transduction. Eur J Biochem 253:531–544, 1998.

 7. DeBenedetti A, Harris AL: eIF4E expression in tumors: Its possible role in progression of malignancies. Intl J Biochem Cell Biol 31:59–72, 1999.

 8. Crew JP, Fuggle S, Bicknell R, et al: Eukaryotic initiation factor-4E in superficial and muscle invasive bladder cancer and its correlation with vascular endothelial growth factor expression and tumour progression. Br J Cancer 82:161–166, 2000.

 9. Stacey SN, Jordan D, Williamson AJK, et al: Leaky scanning is the predominant mechanism for the translation of human papillomavirus type 16. J Virol 74:7284–7297, 2000.

10. Raught B, Gingras A: Human papillomavirus type 18 E1 protein is translated from polycistronic mRNA by a discontinuous scanning mechanism. J Virol 73:3062–3070, 1999.

11. DeFatta RJ, Turbat-Herrera EA, Li BD, et al: Elevated expression of eIF4E is confined in early breast cancer lesions: Possible role of hypoxia. Int J Cancer 80:516–522, 1999.

12. Anthony B, Carter P, De Benedetti A: Overexpression of the proto-oncogene/translation factor 4E in breast-carcinoma cell lines. Int J Cancer 65:858–863, 1996.

13. Bourhis J, Le MG, Barrois M, et al: Prognostic value of c-myc proteo-oncogene overexpression in early invasive carcinoma of the cervix. J Clin Oncol 8:1789–1796, 1990.

14. Nathan CA, Liu L, Li BD, et al: Detection of the proto-oncogene eIF4E in surgical margins may predict recurrence in head and neck cancer. Oncogene 15:579–584, 1997.

15. Nathan CA, Franklin S, Abreo RW, et al: Analysis of surgical margins with the molecular marker eIF4E: A prognostic factor in patients with head and neck cancer. J Clin Oncol 17:2909–2914, 1999.

16. Sorrells DL, Black DR, Meschonat C, et al: Detection of eIF4E gene amplification in breast cancer by competitive PCR. Ann Surg Oncol 5: 232–237, 1998.

17. Smith-McCune K, Zhu Y-H, Hanahan D, Arbeit J: Cross-species comparison of angiogenesis during the premalignant stages of squamous carcinogenesis in the human cervix and K14-HPV 16 transgenic mice. Cancer Res 57:1294–1300, 1997.

18. Dellas A, Schultheiss E, Leivas MR, et al: Association of p27Kip1, cyclin E and c-myc expression with progression and prognosis in HPV-positive cervical neoplasms. Anticancer Res 18:3991–3998, 1998.

19. Wiggins DL, Granai CO, Steinhoff MM, Calabresi P: Tumor angiogenesis as a prognostic factor in cervical carcinoma. Gynecol Oncol 56:353–356, 1995.

20. Fina L, Molgaard HV, Robertson D, et al: Expression of the CD34 gene in vascular endothelial cells. Blood 75:2417–2426, 1990.

21. Weidner N, Carroll PR, Flax J, et al: Tumor angiogenesis correlates with metastasis in invasive prostate carcinoma. Am J Pathol 143: 401–409, 1993.

22. De Benedetti A, Rhoads RE: Overexpression of eukaryotic protein synthesis initiation factor 4E in HeLa cells results in aberrant growth and morphology. Proc Natl Acad Sci USA 87:8212–8216, 1990.

23. Li BD, Liu L, Dawson M, DeBenedetti A: Overexpression of eukaryotic initiation factor 4E (eIF4E) in breast carcinoma. Cancer 79:2385–2390, 1997.

24. Doldi N, Origoni M, Bassan M, et al: Vascular endothelial growth factor expression in human vulvar neoplastic and non-neoplastic tissues. J Reprod Med 41:844–848, 1996.

25. Wattre P, Bert V, Hober D: Apoptosis and human viral infections. Ann Biol Clin 54:189–197, 1996.

26. Pei XF: The human papillomavirus E6/E7 genes induce discordant changes in the expression of cell growth regulatory proteins. Carcinogenesis 17:1395–1401, 1996.

27. Pei XF, Sherman L, Sun YH, Schlegel R: HPV-16 E7 protein bypasses keratinocyte growth inhibition by serum and calcium. Carcinogenesis 19:1481–1486, 1998.

28. Sastre-Garau X, Favre M, Couturier J, Orth G: Distinct patterns of alteration of myc genes associated with integration of human papillomavirus type 16 or type 45 DNA in two genital tumours. J Gen Virol 81:1983–1993, 2000.

29. Kim MS, Shin KH, Baek JH, et al: HPV-16, tobacco-specific N-nitrosamine, and N-methyl-N’-nitro-N-nitrosoguanidine in oral carcinogenesis. Cancer Res 53:4811–4816, 1993.

30. Shin KH, Min BM, Cherrick HM, Park NH: Combined effects of human papillomavirus-18 and N-methyl-N’-nitro-N-nitrosoguanidine on the transformation of normal human oral keratinocytes. Molec Carcinogen 9:76–86, 1994.

31. Pim D, Banks L: Loss of HPV-16 E7 dependence in cells transformed by HPV-16 E7 plus EJ-ras correlates with increased c-myc expression. Oncogene 6:589–594, 1991.

 

 

 

Table 1. Mean Vessel Counts of Tissue Sections Stained with CD34*

Tissue Histology                   MVC                     Range                       n                      P Value^†

Normal                                      6.91                    5.0–10.0                      6

Dysplasia                                 9.62                    7.0–11.3                    10                         .008

Cancer                                      9.32                    6.0–13.0                      8                         .017

*Three counts were performed for each slide. This mean was then averaged for all of the sections within that category.

^†2-sample t-test, as compared with normal.

 

 

 

Table 2. Cumulative Staining Intensities for Tissues Reacted with anti-4E

                                         Count                 Mean                  Standard Error                 P Value*

Normal                               10                       14.5                              7.0

Dysplasia                          10                     140.8                            23.0                            <.0001

Carcinoma                         10                     165.1                            12.4                            <.0001

*2-sample t-test, as compared with normal.

 

Table 3. Stain Intensity Means^a as Measured by Immunohistochemistry

  Stain       Normal (n)       Dysplasia^b (n)       P Value^c        Carcinoma (n)        P Value^c      P Value^d

VEGF        1.22  (9)              2.43 (7)              .0091               2.33  (9)              .0064          .8027^e

    E6           0.2 (10)             0.75 (10)               .09^e                1.45 (10)              .0003            .07^e

    E7           0.5 (10)             1.60 (10)             .0026              2.35 (10)            .000001           .03

^aMean stain intensity was determined by calculating the average intensities as judged by two independent researchers using a scale of 0–3+.

^bDysplasia grades included 3 with CIN I, 2 with CIN II, and 5 with CIN III, including 3 with CIS.

^cMean stain intensity of dysplastic or carcinoma sections versus normal mean stain intensity using 2-sample t-test. Values less than or equal to .5 are considered significant differences.

^dMean stain intensities compared between dysplastic and carcinoma tissues using 2-sample t-test. Values less than or equal to .5 are significant.

^eNot significant.

 

 

Figure 1. Tissues Stained with Anti-4E: (A) normal cervical tissue; (B) cervical dysplasia;
(C) invasive carcinoma.

 

 

Figure 2. Representative Immunohistochemistry with antibody to HPV: (A) invasive carcinoma stained with anti-HPV E6 given a score of 11; (B) invasive carcinomas stained with anti-HPV E7 given a score of 31. Note the deeper staining of the section given a 31 score.

 

 

 

Figure 3. Transfected Cell Lysates. (A) Silver-Stained SDS PAGE Gel of eIF4E-Transfected (H/E) and Non-Transfected (H) HeLa Cell Lysates. A 33-KD band corresponding to the Geneticin resistance protein is shown at the arrowhead. (B) Amplified Western Blot with eIF4E-Transfected (C/E) and Non-Transfected (C) Cell Lysates Reacted with Anti-eIF4E Antisera. A 25-KD band corresponding to eIF4E is shown at the arrowhead. (C) Amplified Western Blot with eIF4E-Transfected (H/E) and Non-Transfected (H) Cell Lysates Reacted with Anti-c-myc Antisera. A 62-KD band corresponding to c-myc protein is shown at the arrowhead. (D) Amplified Western Blot with eIF4E-Transfected (H/E) and Non-Transfected (H) Cell Lysates Reacted with anti-HPV-18 E7 Serum. Molecular weight standards are found and well designated with an asterisk. Note the elevated E7 band in the H/E columns (represented by the arrowhead). (E) Silver-Stained SDS-PAGE of Immunoprecipitation with Anti-eIF4E of eIF4E-Transfected (S/E) and Non-Transfected (S) Cell Lysates. Overexpression of 4E is indicated by the arrowhead in the well containing the S/E lysate as compared to the S lysate. (F) Chemiluminescence-Developed X-Ray Film of Immunoprecipitation with Anti-HPV-16 E7 of eIF4E-Transfected (S/E) and Non-Transfected (S) Cell Lysates. The increase in E7 in the eIF4E-transfected cells is noted by the arrowhead.

 

 

Figure 4. Model for HPV-mediated Angiogenesis in Cervical Neoplasia. Solid lines indicated published information. Dashed lines indicate hypothesized events. 4E is eIF4E; E7 is HPV E7.