- Email: [email protected]
Impaired Angiogenic Balance and Suppression of Tumorigenicity in HeLa Cells Chronically Exposed to Interferon-α
Biochemical and Biophysical Research Communications 277, 410 – 416 (2000) doi:10.1006/bbrc.2000.3690, available online at http://www.idealibrary.com on
Impaired Angiogenic Balance and Suppression of Tumorigenicity in HeLa Cells Chronically Exposed to Interferon-␣ O. Lo´pez-Ocejo,* S. E. Perea,† M. Bequet-Romero,† M. J. Aran˜a,† and P. Lo´pez Saura‡ *Division of Vaccine, †Division of Pharmaceuticals, and ‡Division of Clinical Trials, Center for Genetic Engineering and Biotechnology, P.O. Box 6162, C. Havana, Cuba
Received September 13, 2000
We have previously reported that IFN␣-chronic treatment for 41 days induced a partial phenotype reversion on HeLa cells along with a down-regulation of HPV18 mRNA levels. However, tumorigenicity of these cells in nude mice was unchanged. Interestingly, after 1 year of IFN␣-chronic exposition, HeLa cells failed to induce s.c. tumors when injected into nude mice. In such experimental conditions both HPV18 DNA integration pattern and viral DNA copy number present in HeLa cells remained intact in the nontumorigenic phenotype cells. As result of the treatment with IFN␣, HeLa cells rendered more resistant to lysis mediated by activated natural killer cells in vitro. Furthermore, IFN␣-chronic treatment was able to induce VEGF and decrease bFGF mRNA expression, suggesting a potential effect on the angiogenic behavior of these tumoral cells. Thus, long-term treatment of HeLa cells with IFN␣ can accomplish a reversion of the malignant phenotype by a sequential multistep mechanism, in which the antiangiogenic effect of IFN␣ could be one of the contributing events. © 2000 Academic Press Key Words: angiogenesis; interferon; tumorigenicity; HPV; VEGF; bFGF.
Human papillomavirus (HPV) has been etiologically linked to specific human epithelial carcinomas (1). However, since most warts do not progress to cancer and malignant transformation occurs only after a long latency period, probably infection with HPV is necessary but not sufficient for the development of HPVrelated epithelial malignancies (2). Among the HPVs infecting the anogenital tract, several types, particularly HPV16 and 18, show a strong association with carcinomas of the cervix and induce cervical intraepithelial neoplasias that have a high tendency to malignant progression (3). The oncogenic potential of HPV16 and 18 has been attributed primarily to two early expressed genes, E6 and E7. On in vitro systems, contin0006-291X/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
uous expression of E7 and E6 major transforming proteins are required for maintenance of the proliferative and malignant phenotype of tumor virus-infected cells (4). In HPV-immortalized cells, chromosomal instability, aberrations and mutations are features commonly found and E7 oncogene has been identified as the gene responsible for chromosomal changes (5, 6). E6 and E7 transcription is controlled by host cell factors and modification of such control mechanisms could contribute to an increase in their expression (7). Recently, increasing numbers of cellular proteins, which negatively or positively influence E6/E7 gene expression, have been described. They could be summarized into three groups: (i) cytokines as intercellular mediators of viral gene regulation, (ii) intracellular signal mediators, and (iii) transcription factors controlling the HPV promoter-enhancer region. Among the first group, interferon’s (IFN) are regulatory proteins with potent biological activities. Effects of IFN on cell function include promotion of differentiation, inhibition of cell proliferation and amplification of the immune effector-cell functions. IFNs also modulate other important events for tumor establishment and progression including angiogenesis. Colposcopic studies of cervical cancer patients have revealed that since the earliest stages a frank invasion of tortuous capillaries can be observed in the affected zone. IFNs can inhibit angiogenesis along or in cooperation with retinoic acid in HPV associated lesions, but the pathway leading to this effect is still unclear. Even when IFN could directly affect endothelial cell proliferation, it might be also involved in the regulation of angiogenic factors expression by tumor cells modulating the angiogenic switch. IFNs have shown to regulate HPV expression transcriptionally in both HPV-immortalized cells and cervical cancer established cell lines. On SiHa cells, IFN␣ differentially reduces the HPV16 E6/E7 transcript levels (8). On HeLa cells, HPV18 E6 and E7 mRNA down-
410
Vol. 277, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
regulation by IFN␣ is mediated by selective suppression of endogenous viral transcription. Regulatory cellular flanking regions to viral integration sites appear to mediate such IFN inhibitory effect (9, 10). In HPV-related diseases including cervical intraepithelial neoplasia (11), condylomata acuminata (12), laryngeal papillomatosis (13), and cervical carcinomas (14), IFN␣ therapy has been successfully employed. In our experience, IFN␣ therapy was able to modulate the HPV16 mRNA expression levels in cervical carcinoma patients after 15 days of treatment (15). In vitro long-term treatment studies have established that both natural and recombinant IFN␣ produced a partial phenotype reversion of HeLa cells with the concomitant inhibition of HPV18 mRNA. Nevertheless, despite the phenotype changes accumulated by 41 days of IFN␣ chronic treatment, the capacity of IFN-treated cells to form tumors in immunosuppressed hosts remained unchanged (16). In the present study, the effects of longer treatment of HeLa cells with low concentration and nonantiproliferative, of both natural IFN␣ and recombinant IFN␣2b, were investigated. We demonstrated that both IFN␣ preparations suppress tumorigenicity of HeLa cells after one year of exposition. Molecular characterization of the status of viral DNA in the nontumorigenic HeLa cells demonstrated that the HPV18 sequences present in control HeLa cells were retained in both IFN treated cells. IFN long-term treated HeLa cells also showed an impairment of the angiogenic factor balance characterized by induction of VEGF and inhibition of bFGF mRNA expression, as well as an increase resistance to activated-NK mediated cytotoxicity. MATERIALS AND METHODS Cell culture and reagents. Human cervical carcinoma (HeLa) cells were grown in Dulbecco’s modified Eagle’s medium supplemented with 50 g/mL gentamycin and 10% fetal calf serum (GIBCO/BRL) at 5% CO 2 and 37°C. Cantell type human natural IFN␣ (IFN␣n) and recombinant IFN␣2b were purchased from HeberBiotec (Havana, Cuba). Growth was assessed by seeding approximately 6 ⫻ 10 4 cells/mL in 25 cm 2 plastic flasks. One day after seeding and thereafter every 3 days medium was replaced by fresh medium containing 200 IU/mL of either IFN␣ preparation. Cells were subcultured weekly and parallel cultures were maintained on medium alone as control. Tumorigenicity assay. After 1 year of IFN␣ treatment, HeLa cells were harvested by trypsinization, washed once with phosphatebuffered saline (PBS), and suspended in 1 mL PBS. The tumorigenic potential of the cell lines was evaluated by injection into 6- to 8-weekold female BALB/c athymic mice. Subcutaneous injections of 2 ⫻ 10 6 cells/mouse (DT50 ⱖ 50%) were accomplished with a 23-gauge needle and tumor volume was measured every 7 days. Eight mice were injected with each cell line. DNA analysis. For Southern blot analyses, approximately 30 g high molecular weight DNA, obtained by cesium chloride gradient (17) from the 1-year IFN-treated and untreated HeLa cells, were digested with BamHI or PstI and electrophoresed through a 0.8%
agarose gel. The resolved DNA was transferred to a Hybond-N nylon membrane (Amersham International) and hybridized with a DNA probe radiolabeled, as described previously (16, 18). The HPV18 DNA probe (whole genome) cloned into pBR322, was a gift from Professor Harald zur Hausen. HPV18 DNA copy number was determined by dotting 2.5, 1.0, 0.5, and 0.1 g of cell high molecular weight DNA into Hybond-N membranes. Simultaneously, a standard curve containing 100, 50, 25, 12.5, and 0 pg of HPV18 genome was dotted. Filters were hybridized as indicated for Southern blot analysis. Radioactivity associated with every dot was analyzed in a beta counter equipment (LKB). The HPV18 DNA copy number in each sample was determined by interpolation into the reference curve of the values for HeLa-treated and nontreated cells DNA. The values obtained for HPV-DNA copy numbers for each experimental group were average and the one-way ANOVA test applied show nonsignificant differences among test conditions. NK cell cytolysis assay. For cytolysis assays, Ficoll–Paqueseparated peripheral blood mononuclear cells from healthy donors were incubated for 3 h with 500 IU/mL IFN␣2b. The 1-year treated and untreated control cells were used as target. These target cells were labeled with 51Cr (5 Ci/mL for 6 h). Effector and target cells (5 ⫻ 10 4) were mixed at different effector:target cell ratios in 96-well polystyrene plates, and after a 4-h incubation without IFN␣ at 37°C in 5% CO 2, 100 l of the supernatant from each well was harvested and counted in a gamma counter. NK cell killing was determined calculating the percentage of NK cell-induced release of 51Cr from the target cells, as previously described (19). Results presented represent means ⫾ standard errors of the means (SEM) of at least three separated experiments. The mean percentage spontaneous release from all target cells was less than 10%. The results were compared using an ANOVA test and a Bonferroni post-test in order to analyzed the origin of the differences found (Graph Pad Prism software). RNA isolation and Northern blot analysis. Total RNA was isolated from cultured cells using the Trizol reagent (GIBCO/BRL/Life Science Technologies), essentially as described by the manufacturer. Approximately 20 g total RNA were fractionated by electrophoresis in a 1% agarose gel containing 6.6 mol/L formaldehyde. RNA was transferred to a Hybond-N nylon membrane and immobilized by UV cross-linking for 4 min. Filters were hybridized with 32P-labeled 520-bp human VEGF cDNA and 1100-bp human bFGF cDNA, which were generous gifts from Dr. G. Breier (Max-Planck Institut, 61231 Bad Nauheim, Germany) and Dr. P. Dell’Era (Universita degli Studi di Brescia, Italy), respectively. We used human glyceraldehyde phosphate dehydrogenase (GAPDH) cDNA probe (gift from Dr. Bryan Williams, Cleveland Clinic Foundation, Cleveland, OH) in order to normalize results. Levels of mRNA were quantitated by densitometry analysis using a Scanlet Jet Plus and the Molecular Analyst software (Bio-Rad).
RESULTS The tumorigenicity of HeLa cells (2 ⫻ 10 6/inoculum) after 1 year of IFN treatment using the experimental conditions described above was analyzed first. In contrast to controls, IFN-treated HeLa cells failed to produce any local tumors (Table 1). Even after 28 additional weeks of observation, IFN-treated cells injected mice did not have tumor. Thus, chronic IFN␣ treatment for 1 year led to a nontumorigenic state in HeLa cells. Since we previously observed that HPV18 mRNA was maintained down-regulated after chronic IFN␣ exposition (16), we examined the physical state of HPV18 DNA in HeLa cells after regular trypsinization
411
Vol. 277, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1
Tumorigenicity of HeLa Cells Group
Tumor incidence
Tumor diameter (mm)
HeLa Control HeLa IFN␣n HeLa IFN␣2b
5/8 0/8 0/8
7.3 ⫾ 1.4 0 0
Note. HeLa Control, HeLa IFN␣n, and HeLa IFN␣2b cells (2 ⫻ 10 6/mouse) were injected s.c. into the lateral flanks of nude mice. Tumor incidence and diameter were determined 30 days later.
and subculture for 1 year. Results indicated that IFN␣ chronic treatment for 365 days did not affect the HPV18 DNA integration pattern as determined by Southern blot (Fig. 1). A quantitative analysis of the dot blot shown in Fig. 2 revealed that 1-year IFN␣nand IFN␣2b-treated HeLa cells exhibited similar number of HPV18 DNA copies to that observed in non-IFNtreated control cells (⬇11 copies per cell). Next, we evaluated the influence of long-term IFN␣ treatment on the relative susceptibility of HeLa cells to NK cell-mediated cytotoxicity. As shown in Fig. 3, a significant increment (P ⬍ 0.01) in resistance to IFN␣-activated NK cells was observed for HeLa cells treated with both IFN␣n and IFN␣2b compared to nontreated controls, for all effector:target ratios as confirmed by the Bonferroni test. To examine possible contribution of this treatment to the inhibition of the angiogenic process, we analyzed
FIG. 2. Analysis of HPV18 viral DNA copy number on HeLa cells exposed to IFN␣ for 1 year. (A) Represents a reference curve with different amounts of HPV18 DNA dotted as follows: lane 1, 100 pg; lane 2, 50 pg; lane 3, 25 pg; lane 4, 12.5 pg; lane 5, 6.25 pg; lane 6, 3.12 pg; lane 7, 1.56 pg. (B) High-molecular-weight DNA isolated from HeLa cells treated or not with IFN was dotted as follows: lane 1, 2.5 g; lane 2, 1.0 g; lane 3, 0.5 g; lane 4, 0.1 g. Hybond-N filters were subsequently fixed by UV irradiation and hybridized under the experimental conditions used in Southern blot analysis. Radioactivity associated with different dots was analyzed in a beta counter equipment (LKB). Estimation of HPV18 viral DNA copy number on HeLa cells was carried out according to the reference curve obtained in A.
the mRNA levels of the tumor angiogenesis factors in HeLa cells. As shown in Fig. 4, IFN␣-chronic treatments were able to induce an increase of VEGF and a decrease of bFGF mRNA expression, causing in this way an impairment of angiogenic balance. Analysis of bFGF mRNA expression demonstrated the characteristic 7.4 and 4.4 kb transcripts whereas the usual pattern of 3.4, 1.8, and 1.4 kb transcripts were detected in the case of VEGF. FIG. 1. Analysis of HPV18 DNA integration on HeLa cells chronically exposed to IFNs-␣ for 1 year. 30 g of high-molecular-weight DNA isolated from HeLa cells treated or not with IFN was digested with PstI (P), BamHI (B), or not digested (lanes without label) and further analyzed by Southern blot. The molecular weight pattern was obtained by digesting phage with HindIII. Southern hybridization was performed at 65°C with 32P-labeled HPV18 DNA in 0.5 M sodium phosphate buffer, pH 7, containing 1% BSA for 20 h. Filters were washed several times with 0.04 M sodium phosphate buffer, pH 7, plus 1 mM EDTA. Autoradiograms were developed after 72 h.
DISCUSSION The effect of IFN on the phenotype of various transformed cells has been previously described (20, 21). For instance, studies on murine transformed cell lines have shown that a long-term low-dose mouse IFN treatment can revert cells transformed with bovine papillomavirus to a nontransformed phenotype by elimination of
412
Vol. 277, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. IFN-activated NK cell cytotoxicity vs HeLa cells. Effector:target cell ratio indicates the amount of IFN-activated peripheral mononuclear cells per target cell. Target cell killing is expressed as the percentage (mean of 3 experiments ⫾ SEM) of NK cell-induced release of incorporated 51Cr from target cells in 4-h cytolysis assays. The P values are the result of ANOVA tests at each ratio. The Bonferroni post-test applied to the different groups showed that both IFN-treated cell were more resistant then controls to NK lysis.
viral genomes (22) and reduce expression of the transforming ras oncogene in NIH 3T3 cultures transfected with the viral Ha-MuSV(ras) and Mo-MuSV(mos) oncogenes (23). Other changes like cytoskeleton reorganization, cell shape modification and decrease of cell membrane mobility, have been also associated to phenotypic reversion by IFN (24) although in most cases this effect is rather partial since cells keep the capacity to produce tumors in immunosuppressed hosts (25). In earlier studies, the reversion of HeLa cells transformed phenotype has been achieved using antisense HPV18 RNA-expression vectors (26). In such findings, antisense HPV18 transfected HeLa cells showed inhibition of growth rate, low anchorage independence, inhibition of growth in low serum concentration, and reduced plating efficiency. Accordingly similar results were reached by treating HeLa cells with IFN␣ for 41 days, indicating a partial phenotypic reversion with a parallel HPV18 mRNA down-regulation. Other IFNinduced changes included reduction of the mitotic index and cell pleomorfism, low capacity for forming colonies in semisolid medium and increase of the 2⬘-5⬘ oligoadenylate synthetase gene expression (16). However, after that IFN␣ incubation period, HeLa cells were equally tumorigenic in athymic mice, suggesting
that inhibition of viral expression could be determinant for the observed phenotypic changes but not sufficient for suppressing the tumorigenic potential. This work demonstrates that treatment for one year of HeLa cells with IFN␣ inhibits their tumorigenic properties in nude mice. At the same time HPV18 DNA sequences were not modified neither by continuous in vitro passages nor by a possible IFN-induced cytogenetic change after one year of chronic exposition, confirming that HPV18 mRNA down-regulation by IFN␣ is mediated by epigenetic mechanisms. Long-term treatment of HeLa cells with IFN␣ reduces sensitivity to NK cell cytotoxicity. This effect could be independent of HPV downregulation, because particularly, E7 mRNA expression do not influence the resistant induced by IFN to NK cytotoxicity, in contrast to Adenovirus E1A protein (27). However this point requires further investigation. The switch to a nontumorigenic state in cervical carcinoma cell lines containing HPV genome has been previously observed after the reintroduction of a normal chromosome 11 into HeLa and SiHa cells using somatic cell hybrids (28, 29). Studies of the mechanism involved in such tumorigenicity suppression have demonstrated that the 55 kDa regulatory subunit of protein phosphatase 2A activates the HPV16 long control region in human cells with a deletion in the short arm of chromosome 11 (30). Those and other findings lead to the hypothesis that loci 11p13 could contain a tumor suppressor like factor (31). Decreased expression of E6 and E7 oncogenes was required for the nontumorigenic phenotype of hybrids between HeLa cells and immortalized rodent fibroblasts in nude mice (32). The fact that 1-year IFN␣-treated HeLa cells lost completely the capacity to produce tumor in nude mice, confirms that new phenotypic changes besides those described after 41 days of treatment (16) took place. Nevertheless these data do not discard that such phenotype reversion could be earlier achieved. The antiangiogenic effect of IFN, which has been demonstrated in vitro and in vivo on HeLa and other HPVtransformed cell lines, could explain at least in part these findings (33). Recently, it was reported that IFN␣ is able to decrease bFGF expression in human carcinoma cells, and this inhibitory effect increase with the duration of the treatment (34, 35). Accordingly a significant reduction of bFGF mRNA levels on HeLa cells chronically treated with IFN␣ was found in this paper. These results are consistent with the antiangiogenic properties already described for IFNs. The VEGF mRNA upregulation observed in our system, also agrees with previous data. (36, 37). The synergistic interaction of VEGF and bFGF on inducing the angiogenic event has been widely described. VEGFdependent angiogenesis in certain settings might require the presence of a second positive regulator, such as bFGF, which increases VEGF receptor expression
413
Vol. 277, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. IFN␣-chronic treatments induce VEGF and inhibit bFGF mRNA expression in HeLa cells. HeLa cells were incubated for 1 year in medium (lane 1) or medium containing 200 IU of IFN␣n (lane 2), or IFN␣2b (lane 3) per mL. (A) Northern blot analysis. One representative experiment of three is shown. (B) VEGF:bFGF ratio after densitometric scanning of the filters.
and hence mediated signal transduction (38). In our case, VEGF mRNA induction could not necessarily account for an induction of angiogenesis, given the lower levels of bFGF mRNA detected after IFN treatment. In addition, it has been demonstrated that the local variation in the concentrations of these growth factors in the tumor microenvironment may modulate the adhesion of cytotoxic lymphocytes to tumor-associated endothelial cells. VEGF produced by tumor cells upregulates cellular adhesion molecules (CAMs), particularly, intracellular and vascular CAMs (ICAM-1 and VCAM-1) on tumor-associated endothelial cells and facilitates NK cell adhesion, whereas bFGF released by tumor cells prevents this interaction (39). Consequently, it has been suggested that the production of bFGF in wound repair or within tumors may provide varying degrees of protection to growing vessels by inhibiting adhesion of circulating lymphocytes to endothelial cells that have been exposed to VEGF or other adhesion-promoting angiogenesis factors. Thus, for the complete tumorigenicity suppression of these cells by long-term treatment with IFNs-␣, other cell mechanisms activated or regulated by IFN␣ should be simultaneously produced besides viral expression abrogation. On the other hand, previous results in juvenile testicular hemangiomas have demonstrated that lack of bFGF expression is related to an inadequate efficacy of IFN␣ treatment (40). So, it might be helpful to assess the dependence of tumors on VEGF/bFGF balance in order to use IFN␣ therapy rationally. Treatment with IFN␣ seems to be adjusted to one of the general strategies to develop antiangiogenic therapy for clinical use (41). This strategy consists in inhi-
bition of angiogenic molecules release from tumor cells; besides being based on the long-term use of inhibitors with low toxicity. In summary, we demonstrated that the IFN␣ longterm treatment suppresses tumorigenicity of HeLa cells. The induction of nontumorigenic phenotype by IFN␣ is mediated by a multistep mechanism that includes down-regulation of HPV18 mRNA and a potential antiangiogenic effect. These results suggest a beneficial effect of the sustained presence of IFN␣ in human tumors, specifically on cervical cancers associated to HPV infection. ACKNOWLEDGMENTS This research was partly supported by grants from UNIDO No. 92/050, CRP/CUB 91-04, and SAREC Project S/2.CUB.00.
REFERENCES
414
1. Lorincz, A. T., Temple, G. F., Kurman, F. J., Jenson, A. B., and Lancaster, W. D. (1987) Oncogenic association of specific human papillomavirus types with cervical neoplasia. J. Natl. Cancer Inst. 79, 671– 677. 2. Mansur, C. P., and Androphy, E. J. (1993) Cellular transformation by papillomavirus oncoproteins. Biochim. Biophys. Acta 1155, 323–345. 3. zur Hausen, H. (1991) Viruses in human cancer. Science 254, 1167–1173. 4. Wanatabe, S., Kanada, T., and Yoshike, K. (1989) Human papillomavirus type 16 transformation of primary human embryonic fibroblasts requires expression of open reading frames E6 and E7. J. Virol. 63, 965–969. 5. Drews, R. E., Chan, V. T. W., and Schnipper, L. E. (1992) Oncogenes result in genomic alterations that activate a transcription-
Vol. 277, No. 2, 2000
6.
7.
8.
9.
10.
11. 12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
ally silent, dominantly selectable reporter gene (neo). Mol. Cell Biol. 12, 198 –206. Hashida, T., and Yasumoto, S. (1991) Induction of chromosome abnormalities in mouse and human epidermal keratinocytes by the human papillomavirus type 16 E7 oncogene. J. Gen. Virol. 72, 1569 –1577. zur Hausen, H. (1989) Papillomaviruses in anogenital cancer as a model to understand the role of viruses in human cancers. Cancer Res. 49, 4677– 4681. De Marco, F., and Marcante, M. L. (1993) HPV-16 E6 –E7 differential transcription induced in SiHa cervical cancer cell line by interferons. J. Biol. Reg. Homeostasis 7, 15–21. Nawa, A., Nishiyama, Y., Yamamoto, N., Maeno, K., Goto, S., and Tomoda, Y. (1990) Selective suppression of human papilloma virus type 18 mRNA level in HeLa cells by interferon. Biochem. Biophys. Res. Commun. 170, 793–799. Perea, S. E., Lo´pez-Ocejo, O., Garcı´a-Milia´n, R., and Aran˜a, M. J. (1995) Interferon-alpha elicits downregulation of human papillomavirus 18 mRNA in HeLa cells by selective repression of endogenous viral transcription. J. Interferon Cytokine Res. 15, 495–501. Strander, H. (1989) The action of interferons on virus-associated human neoplasms. Cancer Surv. 8, 755–792. Reichman, R. C., and Strike, D. G. (1989) Pathogenesis and treatment of human genital papillomavirus infections: A review. Antiviral Res. 11, 109 –118. Haglund, S., Lundquist, P. G., Kantell, K., and Strander, H. (1985) Interferon therapy in juvenile laryngeal papillomatosis. Arch. Otolaryngol. 107, 327–332. Lippman, S. M., Kavanagh, J. J., Paredes-Espinosa, F., Delgadillo-Madruen˜o, F., Paredes-Casillas, P., Hong, W. K., Holdener, E., and Krakoff, I. H. (1992) 13-cis-Retinoic acid plus interferon alpha-2a: Highly active systemic therapy for squamous cell carcinoma of the cervix. J. Natl. Cancer Inst. 84, 241–245. Garcı´a-Milia´n, R., Rı´os, M. A., Amigo´, M., Dı´as, D., Guilar, O., Silveira, M., Aran˜a, M. J., and Perea, S. E. (1996) Modulation of human papillomavirus type 16 mRNA in cervical invasive carcinoma patients by interferon-alpha therapy. J. Interferon Cytokine Res. 16, 709 –713. Lo´pez-Ocejo, O., Perea, S. E., Reyes, A., Vigoa, L., and Lo´pezSaura, P. (1993) Partial phenotypic reversion of HeLa cells by long-term interferon-alpha treatment. J. Interferon Res. 13, 369 –375. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1989) Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Feinberg, A. P., and Vogelstein, B. (1983) A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132, 6 –13. Apezteguı´a, I., and Perea, S. E. (1989) Procedimiento para el marcaje de ce´lulas tumorales con Cr 51. Interfero´n Biotecnol. 6, 77–79. Takano, S., and Nagano, Y. (1984) Differentiation in mouse leukemia cells mediated by an immune interferon preparation. J. Interferon Res. 4, 383–388. Dahl, H. (1983) Human interferon and cell growth inhibition. VII. Reversibility of interferon activities. J. Interferon Res. 3, 387–393. Turek, L. P., Byrne, J. C., and Lowy, D. R. (1982) Interferon induces morphologic reversion with elimination of extrachromosomal viral genomes in bovine papillomavirus-transformed mouse cells. Proc. Natl. Acad. Sci. USA 79, 7914 –7918. Samid, D., Chang, E. H., and Friedman, R. M. (1985) Development of transformed phenotype induced by a human ras onco-
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
415
gene is inhibited by interferon. Biochem. Biophys. Res. Commun. 126, 509 –516. Bourgeade, M. F., Rousset, S., Paulin, D., and Chany, C. (1981) Reorganization of the cytoskeleton by interferon in MSVtransformed cells. J. Interferon Res. 1, 323–332. Brouty-Boye, D., Maunoury, M. T., and Lowy, R. (1988) Effects of interferons alpha,beta,gamma alone or in combination on the phenotype of human colon carcinoma cells. Anticancer Res. 8, 197–204. Shillitoe, E. J., and Steele, C. (1992) Inhibition of the transformed phenotype of carcinoma cells that contain human papillomavirus. Ann. NY Acad. Sci. 660, 286 –287. Routes, J. M., and Ryan, S. (1995) Oncogenicity of human papillomavirus- or adenovirus-transformed cells correlates with resistance to lysis by natural killer cells. J. Virol. 69, 7639 –7647. Saxon, P. J., Srivatsan, E. S., and Stanbridge, J. (1986) Introduction of human chromosome 11 via microcell transfer controls tumorigenic expression of HeLa cells. EMBO J. 5, 3461–3466. Koi, M., Morita, H., Yamada, H., Saboh, H., Barrett, J. C., and Oshimura, H. (1989) Normal human chromosome 11 suppresses tumorigenicity of human cervical tumor cell line SiHa. Mol. Carcinogen. 2, 12–21. Smits, P. H. M., Smits, H. L., Minnaar, R. P., Hemmings, B. A., Mayer-Jaekel, R. E., Schuurman, R., Der Noordaa, J. V., and Schegget, J. T. (1992) The 55 kDa regulatory subunit of protein phosphatase 2A plays a role in the activation of the HPV16 long control region in human cells with a deletion in the short arm of chromosome 11. EMBO J. 11, 4601– 4606. Ter Schegeet, J., and Van Der Noordaa, J. (1994) Protein phosphatase 2A and the regulation of human papillomavirus gene activity. In Current Topics in Microbes and Immunity, Vol. 186, pp. 121–129. Springer-Verlag Berlin/Heidelberg. Bosch, F. X., Schwarz, E., Boukamp, P., Fusenig, N. E., Bartsch, D., and zur Hausen, H. (1990) Suppression in vivo of human papillomavirus type 18 E6-E7 gene expression in nontumorigenic HeLa X fibroblast hybrid cells. J. Virol. 64, 4743– 4754. Majewsky, S., Szmurlo, A., Marczak, M., Jablonska, S., and Bollag, W. (1994) Synergistic effect of retinoids and interferon alpha on tumor-induced angiogenesis: Anti-angiogenic effect on HPV-harboring tumor-cell lines. Int. J. Cancer 57, 81– 85. Singh, R. K., Gutman, M., Bucana, C. D., Sanchez, R., Llansa, N., and Fidler, I. J. (1995) Interferons alpha and beta downregulate the expression of basic fibroblast growth factor in human carcinomas. Proc. Natl. Acad. Sci. USA 92, 4562– 4566. Dinney, C. P. N., Bielenberg, D. R., Perrotte, P., Eve, B. Y., Bucana, C. D., and Fidler, I. J. (1998) Inhibition of basic fibroblast growth factor expression, angiogenesis, and growth of human bladder carcinoma in mice by systemic interferon-alpha administration. Cancer Res. 58, 808 – 814. Cohen, T., Nahari, D., Cerem, L. W., Neufeld, G., and Levi, B. Z. (1996) Interleukin 6 induces the expression of vascular endothelial growth factor. J. Biol. Chem. 271, 736 –741. Der, S. D., Zhou, A., Williams, B. R. G., and Silverman, R. H. (1998) Identification of genes differentially regulated by interferon alpha, beta, or gamma using oligonucleotide arrays. Proc. Natl. Acad. Sci. USA 95, 15623–15628. Pepper, M. S., Mandriota, S. J., Vassalli, J. D., Orci, L., and Montesano, R. (1996) Angiogenesis-regulating cytokines: Activities and interactions. In Current Topics in Microbes and Immunity (Gunthert, U., and Birchmeier, W., Eds.), Vol. 213, pp. 31– 67, Springer-Verlag, Berlin/Heidelberg. Melder, R. J., Koenig, G. C., Witwer, B. P., Safabakhsh, N., Munn, L. L., and Jain, R. K. (1996) During angiogenesis, vascular endothelial growth factor and basic fibroblast growth factor
Vol. 277, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
regulate natural killer cell adhesion to tumor endothelium. Nat. Med. 2, 992–997. 40. Uchida, K., Takahashi, A., Miyao, N., Takeda, K., Tsutsumi, H., Satoh, M., and Tsukamoto, T. (1997) Juvenile hemangioma of
the testis: Analysis of expression of angiogenic factors. Urology 49, 285–286. 41. Folkman, J., and Ingber, D. (1992) Inhibition of angiogenesis. Semin. Cancer Biol. 3, 89 –96.
416
Impaired Angiogenic Balance and Suppression of Tumorigenicity in HeLa Cells Chronically Exposed to Interferon-␣ O. Lo´pez-Ocejo,* S. E. Perea,† M. Bequet-Romero,† M. J. Aran˜a,† and P. Lo´pez Saura‡ *Division of Vaccine, †Division of Pharmaceuticals, and ‡Division of Clinical Trials, Center for Genetic Engineering and Biotechnology, P.O. Box 6162, C. Havana, Cuba
Received September 13, 2000
We have previously reported that IFN␣-chronic treatment for 41 days induced a partial phenotype reversion on HeLa cells along with a down-regulation of HPV18 mRNA levels. However, tumorigenicity of these cells in nude mice was unchanged. Interestingly, after 1 year of IFN␣-chronic exposition, HeLa cells failed to induce s.c. tumors when injected into nude mice. In such experimental conditions both HPV18 DNA integration pattern and viral DNA copy number present in HeLa cells remained intact in the nontumorigenic phenotype cells. As result of the treatment with IFN␣, HeLa cells rendered more resistant to lysis mediated by activated natural killer cells in vitro. Furthermore, IFN␣-chronic treatment was able to induce VEGF and decrease bFGF mRNA expression, suggesting a potential effect on the angiogenic behavior of these tumoral cells. Thus, long-term treatment of HeLa cells with IFN␣ can accomplish a reversion of the malignant phenotype by a sequential multistep mechanism, in which the antiangiogenic effect of IFN␣ could be one of the contributing events. © 2000 Academic Press Key Words: angiogenesis; interferon; tumorigenicity; HPV; VEGF; bFGF.
Human papillomavirus (HPV) has been etiologically linked to specific human epithelial carcinomas (1). However, since most warts do not progress to cancer and malignant transformation occurs only after a long latency period, probably infection with HPV is necessary but not sufficient for the development of HPVrelated epithelial malignancies (2). Among the HPVs infecting the anogenital tract, several types, particularly HPV16 and 18, show a strong association with carcinomas of the cervix and induce cervical intraepithelial neoplasias that have a high tendency to malignant progression (3). The oncogenic potential of HPV16 and 18 has been attributed primarily to two early expressed genes, E6 and E7. On in vitro systems, contin0006-291X/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
uous expression of E7 and E6 major transforming proteins are required for maintenance of the proliferative and malignant phenotype of tumor virus-infected cells (4). In HPV-immortalized cells, chromosomal instability, aberrations and mutations are features commonly found and E7 oncogene has been identified as the gene responsible for chromosomal changes (5, 6). E6 and E7 transcription is controlled by host cell factors and modification of such control mechanisms could contribute to an increase in their expression (7). Recently, increasing numbers of cellular proteins, which negatively or positively influence E6/E7 gene expression, have been described. They could be summarized into three groups: (i) cytokines as intercellular mediators of viral gene regulation, (ii) intracellular signal mediators, and (iii) transcription factors controlling the HPV promoter-enhancer region. Among the first group, interferon’s (IFN) are regulatory proteins with potent biological activities. Effects of IFN on cell function include promotion of differentiation, inhibition of cell proliferation and amplification of the immune effector-cell functions. IFNs also modulate other important events for tumor establishment and progression including angiogenesis. Colposcopic studies of cervical cancer patients have revealed that since the earliest stages a frank invasion of tortuous capillaries can be observed in the affected zone. IFNs can inhibit angiogenesis along or in cooperation with retinoic acid in HPV associated lesions, but the pathway leading to this effect is still unclear. Even when IFN could directly affect endothelial cell proliferation, it might be also involved in the regulation of angiogenic factors expression by tumor cells modulating the angiogenic switch. IFNs have shown to regulate HPV expression transcriptionally in both HPV-immortalized cells and cervical cancer established cell lines. On SiHa cells, IFN␣ differentially reduces the HPV16 E6/E7 transcript levels (8). On HeLa cells, HPV18 E6 and E7 mRNA down-
410
Vol. 277, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
regulation by IFN␣ is mediated by selective suppression of endogenous viral transcription. Regulatory cellular flanking regions to viral integration sites appear to mediate such IFN inhibitory effect (9, 10). In HPV-related diseases including cervical intraepithelial neoplasia (11), condylomata acuminata (12), laryngeal papillomatosis (13), and cervical carcinomas (14), IFN␣ therapy has been successfully employed. In our experience, IFN␣ therapy was able to modulate the HPV16 mRNA expression levels in cervical carcinoma patients after 15 days of treatment (15). In vitro long-term treatment studies have established that both natural and recombinant IFN␣ produced a partial phenotype reversion of HeLa cells with the concomitant inhibition of HPV18 mRNA. Nevertheless, despite the phenotype changes accumulated by 41 days of IFN␣ chronic treatment, the capacity of IFN-treated cells to form tumors in immunosuppressed hosts remained unchanged (16). In the present study, the effects of longer treatment of HeLa cells with low concentration and nonantiproliferative, of both natural IFN␣ and recombinant IFN␣2b, were investigated. We demonstrated that both IFN␣ preparations suppress tumorigenicity of HeLa cells after one year of exposition. Molecular characterization of the status of viral DNA in the nontumorigenic HeLa cells demonstrated that the HPV18 sequences present in control HeLa cells were retained in both IFN treated cells. IFN long-term treated HeLa cells also showed an impairment of the angiogenic factor balance characterized by induction of VEGF and inhibition of bFGF mRNA expression, as well as an increase resistance to activated-NK mediated cytotoxicity. MATERIALS AND METHODS Cell culture and reagents. Human cervical carcinoma (HeLa) cells were grown in Dulbecco’s modified Eagle’s medium supplemented with 50 g/mL gentamycin and 10% fetal calf serum (GIBCO/BRL) at 5% CO 2 and 37°C. Cantell type human natural IFN␣ (IFN␣n) and recombinant IFN␣2b were purchased from HeberBiotec (Havana, Cuba). Growth was assessed by seeding approximately 6 ⫻ 10 4 cells/mL in 25 cm 2 plastic flasks. One day after seeding and thereafter every 3 days medium was replaced by fresh medium containing 200 IU/mL of either IFN␣ preparation. Cells were subcultured weekly and parallel cultures were maintained on medium alone as control. Tumorigenicity assay. After 1 year of IFN␣ treatment, HeLa cells were harvested by trypsinization, washed once with phosphatebuffered saline (PBS), and suspended in 1 mL PBS. The tumorigenic potential of the cell lines was evaluated by injection into 6- to 8-weekold female BALB/c athymic mice. Subcutaneous injections of 2 ⫻ 10 6 cells/mouse (DT50 ⱖ 50%) were accomplished with a 23-gauge needle and tumor volume was measured every 7 days. Eight mice were injected with each cell line. DNA analysis. For Southern blot analyses, approximately 30 g high molecular weight DNA, obtained by cesium chloride gradient (17) from the 1-year IFN-treated and untreated HeLa cells, were digested with BamHI or PstI and electrophoresed through a 0.8%
agarose gel. The resolved DNA was transferred to a Hybond-N nylon membrane (Amersham International) and hybridized with a DNA probe radiolabeled, as described previously (16, 18). The HPV18 DNA probe (whole genome) cloned into pBR322, was a gift from Professor Harald zur Hausen. HPV18 DNA copy number was determined by dotting 2.5, 1.0, 0.5, and 0.1 g of cell high molecular weight DNA into Hybond-N membranes. Simultaneously, a standard curve containing 100, 50, 25, 12.5, and 0 pg of HPV18 genome was dotted. Filters were hybridized as indicated for Southern blot analysis. Radioactivity associated with every dot was analyzed in a beta counter equipment (LKB). The HPV18 DNA copy number in each sample was determined by interpolation into the reference curve of the values for HeLa-treated and nontreated cells DNA. The values obtained for HPV-DNA copy numbers for each experimental group were average and the one-way ANOVA test applied show nonsignificant differences among test conditions. NK cell cytolysis assay. For cytolysis assays, Ficoll–Paqueseparated peripheral blood mononuclear cells from healthy donors were incubated for 3 h with 500 IU/mL IFN␣2b. The 1-year treated and untreated control cells were used as target. These target cells were labeled with 51Cr (5 Ci/mL for 6 h). Effector and target cells (5 ⫻ 10 4) were mixed at different effector:target cell ratios in 96-well polystyrene plates, and after a 4-h incubation without IFN␣ at 37°C in 5% CO 2, 100 l of the supernatant from each well was harvested and counted in a gamma counter. NK cell killing was determined calculating the percentage of NK cell-induced release of 51Cr from the target cells, as previously described (19). Results presented represent means ⫾ standard errors of the means (SEM) of at least three separated experiments. The mean percentage spontaneous release from all target cells was less than 10%. The results were compared using an ANOVA test and a Bonferroni post-test in order to analyzed the origin of the differences found (Graph Pad Prism software). RNA isolation and Northern blot analysis. Total RNA was isolated from cultured cells using the Trizol reagent (GIBCO/BRL/Life Science Technologies), essentially as described by the manufacturer. Approximately 20 g total RNA were fractionated by electrophoresis in a 1% agarose gel containing 6.6 mol/L formaldehyde. RNA was transferred to a Hybond-N nylon membrane and immobilized by UV cross-linking for 4 min. Filters were hybridized with 32P-labeled 520-bp human VEGF cDNA and 1100-bp human bFGF cDNA, which were generous gifts from Dr. G. Breier (Max-Planck Institut, 61231 Bad Nauheim, Germany) and Dr. P. Dell’Era (Universita degli Studi di Brescia, Italy), respectively. We used human glyceraldehyde phosphate dehydrogenase (GAPDH) cDNA probe (gift from Dr. Bryan Williams, Cleveland Clinic Foundation, Cleveland, OH) in order to normalize results. Levels of mRNA were quantitated by densitometry analysis using a Scanlet Jet Plus and the Molecular Analyst software (Bio-Rad).
RESULTS The tumorigenicity of HeLa cells (2 ⫻ 10 6/inoculum) after 1 year of IFN treatment using the experimental conditions described above was analyzed first. In contrast to controls, IFN-treated HeLa cells failed to produce any local tumors (Table 1). Even after 28 additional weeks of observation, IFN-treated cells injected mice did not have tumor. Thus, chronic IFN␣ treatment for 1 year led to a nontumorigenic state in HeLa cells. Since we previously observed that HPV18 mRNA was maintained down-regulated after chronic IFN␣ exposition (16), we examined the physical state of HPV18 DNA in HeLa cells after regular trypsinization
411
Vol. 277, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1
Tumorigenicity of HeLa Cells Group
Tumor incidence
Tumor diameter (mm)
HeLa Control HeLa IFN␣n HeLa IFN␣2b
5/8 0/8 0/8
7.3 ⫾ 1.4 0 0
Note. HeLa Control, HeLa IFN␣n, and HeLa IFN␣2b cells (2 ⫻ 10 6/mouse) were injected s.c. into the lateral flanks of nude mice. Tumor incidence and diameter were determined 30 days later.
and subculture for 1 year. Results indicated that IFN␣ chronic treatment for 365 days did not affect the HPV18 DNA integration pattern as determined by Southern blot (Fig. 1). A quantitative analysis of the dot blot shown in Fig. 2 revealed that 1-year IFN␣nand IFN␣2b-treated HeLa cells exhibited similar number of HPV18 DNA copies to that observed in non-IFNtreated control cells (⬇11 copies per cell). Next, we evaluated the influence of long-term IFN␣ treatment on the relative susceptibility of HeLa cells to NK cell-mediated cytotoxicity. As shown in Fig. 3, a significant increment (P ⬍ 0.01) in resistance to IFN␣-activated NK cells was observed for HeLa cells treated with both IFN␣n and IFN␣2b compared to nontreated controls, for all effector:target ratios as confirmed by the Bonferroni test. To examine possible contribution of this treatment to the inhibition of the angiogenic process, we analyzed
FIG. 2. Analysis of HPV18 viral DNA copy number on HeLa cells exposed to IFN␣ for 1 year. (A) Represents a reference curve with different amounts of HPV18 DNA dotted as follows: lane 1, 100 pg; lane 2, 50 pg; lane 3, 25 pg; lane 4, 12.5 pg; lane 5, 6.25 pg; lane 6, 3.12 pg; lane 7, 1.56 pg. (B) High-molecular-weight DNA isolated from HeLa cells treated or not with IFN was dotted as follows: lane 1, 2.5 g; lane 2, 1.0 g; lane 3, 0.5 g; lane 4, 0.1 g. Hybond-N filters were subsequently fixed by UV irradiation and hybridized under the experimental conditions used in Southern blot analysis. Radioactivity associated with different dots was analyzed in a beta counter equipment (LKB). Estimation of HPV18 viral DNA copy number on HeLa cells was carried out according to the reference curve obtained in A.
the mRNA levels of the tumor angiogenesis factors in HeLa cells. As shown in Fig. 4, IFN␣-chronic treatments were able to induce an increase of VEGF and a decrease of bFGF mRNA expression, causing in this way an impairment of angiogenic balance. Analysis of bFGF mRNA expression demonstrated the characteristic 7.4 and 4.4 kb transcripts whereas the usual pattern of 3.4, 1.8, and 1.4 kb transcripts were detected in the case of VEGF. FIG. 1. Analysis of HPV18 DNA integration on HeLa cells chronically exposed to IFNs-␣ for 1 year. 30 g of high-molecular-weight DNA isolated from HeLa cells treated or not with IFN was digested with PstI (P), BamHI (B), or not digested (lanes without label) and further analyzed by Southern blot. The molecular weight pattern was obtained by digesting phage with HindIII. Southern hybridization was performed at 65°C with 32P-labeled HPV18 DNA in 0.5 M sodium phosphate buffer, pH 7, containing 1% BSA for 20 h. Filters were washed several times with 0.04 M sodium phosphate buffer, pH 7, plus 1 mM EDTA. Autoradiograms were developed after 72 h.
DISCUSSION The effect of IFN on the phenotype of various transformed cells has been previously described (20, 21). For instance, studies on murine transformed cell lines have shown that a long-term low-dose mouse IFN treatment can revert cells transformed with bovine papillomavirus to a nontransformed phenotype by elimination of
412
Vol. 277, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. IFN-activated NK cell cytotoxicity vs HeLa cells. Effector:target cell ratio indicates the amount of IFN-activated peripheral mononuclear cells per target cell. Target cell killing is expressed as the percentage (mean of 3 experiments ⫾ SEM) of NK cell-induced release of incorporated 51Cr from target cells in 4-h cytolysis assays. The P values are the result of ANOVA tests at each ratio. The Bonferroni post-test applied to the different groups showed that both IFN-treated cell were more resistant then controls to NK lysis.
viral genomes (22) and reduce expression of the transforming ras oncogene in NIH 3T3 cultures transfected with the viral Ha-MuSV(ras) and Mo-MuSV(mos) oncogenes (23). Other changes like cytoskeleton reorganization, cell shape modification and decrease of cell membrane mobility, have been also associated to phenotypic reversion by IFN (24) although in most cases this effect is rather partial since cells keep the capacity to produce tumors in immunosuppressed hosts (25). In earlier studies, the reversion of HeLa cells transformed phenotype has been achieved using antisense HPV18 RNA-expression vectors (26). In such findings, antisense HPV18 transfected HeLa cells showed inhibition of growth rate, low anchorage independence, inhibition of growth in low serum concentration, and reduced plating efficiency. Accordingly similar results were reached by treating HeLa cells with IFN␣ for 41 days, indicating a partial phenotypic reversion with a parallel HPV18 mRNA down-regulation. Other IFNinduced changes included reduction of the mitotic index and cell pleomorfism, low capacity for forming colonies in semisolid medium and increase of the 2⬘-5⬘ oligoadenylate synthetase gene expression (16). However, after that IFN␣ incubation period, HeLa cells were equally tumorigenic in athymic mice, suggesting
that inhibition of viral expression could be determinant for the observed phenotypic changes but not sufficient for suppressing the tumorigenic potential. This work demonstrates that treatment for one year of HeLa cells with IFN␣ inhibits their tumorigenic properties in nude mice. At the same time HPV18 DNA sequences were not modified neither by continuous in vitro passages nor by a possible IFN-induced cytogenetic change after one year of chronic exposition, confirming that HPV18 mRNA down-regulation by IFN␣ is mediated by epigenetic mechanisms. Long-term treatment of HeLa cells with IFN␣ reduces sensitivity to NK cell cytotoxicity. This effect could be independent of HPV downregulation, because particularly, E7 mRNA expression do not influence the resistant induced by IFN to NK cytotoxicity, in contrast to Adenovirus E1A protein (27). However this point requires further investigation. The switch to a nontumorigenic state in cervical carcinoma cell lines containing HPV genome has been previously observed after the reintroduction of a normal chromosome 11 into HeLa and SiHa cells using somatic cell hybrids (28, 29). Studies of the mechanism involved in such tumorigenicity suppression have demonstrated that the 55 kDa regulatory subunit of protein phosphatase 2A activates the HPV16 long control region in human cells with a deletion in the short arm of chromosome 11 (30). Those and other findings lead to the hypothesis that loci 11p13 could contain a tumor suppressor like factor (31). Decreased expression of E6 and E7 oncogenes was required for the nontumorigenic phenotype of hybrids between HeLa cells and immortalized rodent fibroblasts in nude mice (32). The fact that 1-year IFN␣-treated HeLa cells lost completely the capacity to produce tumor in nude mice, confirms that new phenotypic changes besides those described after 41 days of treatment (16) took place. Nevertheless these data do not discard that such phenotype reversion could be earlier achieved. The antiangiogenic effect of IFN, which has been demonstrated in vitro and in vivo on HeLa and other HPVtransformed cell lines, could explain at least in part these findings (33). Recently, it was reported that IFN␣ is able to decrease bFGF expression in human carcinoma cells, and this inhibitory effect increase with the duration of the treatment (34, 35). Accordingly a significant reduction of bFGF mRNA levels on HeLa cells chronically treated with IFN␣ was found in this paper. These results are consistent with the antiangiogenic properties already described for IFNs. The VEGF mRNA upregulation observed in our system, also agrees with previous data. (36, 37). The synergistic interaction of VEGF and bFGF on inducing the angiogenic event has been widely described. VEGFdependent angiogenesis in certain settings might require the presence of a second positive regulator, such as bFGF, which increases VEGF receptor expression
413
Vol. 277, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. IFN␣-chronic treatments induce VEGF and inhibit bFGF mRNA expression in HeLa cells. HeLa cells were incubated for 1 year in medium (lane 1) or medium containing 200 IU of IFN␣n (lane 2), or IFN␣2b (lane 3) per mL. (A) Northern blot analysis. One representative experiment of three is shown. (B) VEGF:bFGF ratio after densitometric scanning of the filters.
and hence mediated signal transduction (38). In our case, VEGF mRNA induction could not necessarily account for an induction of angiogenesis, given the lower levels of bFGF mRNA detected after IFN treatment. In addition, it has been demonstrated that the local variation in the concentrations of these growth factors in the tumor microenvironment may modulate the adhesion of cytotoxic lymphocytes to tumor-associated endothelial cells. VEGF produced by tumor cells upregulates cellular adhesion molecules (CAMs), particularly, intracellular and vascular CAMs (ICAM-1 and VCAM-1) on tumor-associated endothelial cells and facilitates NK cell adhesion, whereas bFGF released by tumor cells prevents this interaction (39). Consequently, it has been suggested that the production of bFGF in wound repair or within tumors may provide varying degrees of protection to growing vessels by inhibiting adhesion of circulating lymphocytes to endothelial cells that have been exposed to VEGF or other adhesion-promoting angiogenesis factors. Thus, for the complete tumorigenicity suppression of these cells by long-term treatment with IFNs-␣, other cell mechanisms activated or regulated by IFN␣ should be simultaneously produced besides viral expression abrogation. On the other hand, previous results in juvenile testicular hemangiomas have demonstrated that lack of bFGF expression is related to an inadequate efficacy of IFN␣ treatment (40). So, it might be helpful to assess the dependence of tumors on VEGF/bFGF balance in order to use IFN␣ therapy rationally. Treatment with IFN␣ seems to be adjusted to one of the general strategies to develop antiangiogenic therapy for clinical use (41). This strategy consists in inhi-
bition of angiogenic molecules release from tumor cells; besides being based on the long-term use of inhibitors with low toxicity. In summary, we demonstrated that the IFN␣ longterm treatment suppresses tumorigenicity of HeLa cells. The induction of nontumorigenic phenotype by IFN␣ is mediated by a multistep mechanism that includes down-regulation of HPV18 mRNA and a potential antiangiogenic effect. These results suggest a beneficial effect of the sustained presence of IFN␣ in human tumors, specifically on cervical cancers associated to HPV infection. ACKNOWLEDGMENTS This research was partly supported by grants from UNIDO No. 92/050, CRP/CUB 91-04, and SAREC Project S/2.CUB.00.
REFERENCES
414
1. Lorincz, A. T., Temple, G. F., Kurman, F. J., Jenson, A. B., and Lancaster, W. D. (1987) Oncogenic association of specific human papillomavirus types with cervical neoplasia. J. Natl. Cancer Inst. 79, 671– 677. 2. Mansur, C. P., and Androphy, E. J. (1993) Cellular transformation by papillomavirus oncoproteins. Biochim. Biophys. Acta 1155, 323–345. 3. zur Hausen, H. (1991) Viruses in human cancer. Science 254, 1167–1173. 4. Wanatabe, S., Kanada, T., and Yoshike, K. (1989) Human papillomavirus type 16 transformation of primary human embryonic fibroblasts requires expression of open reading frames E6 and E7. J. Virol. 63, 965–969. 5. Drews, R. E., Chan, V. T. W., and Schnipper, L. E. (1992) Oncogenes result in genomic alterations that activate a transcription-
Vol. 277, No. 2, 2000
6.
7.
8.
9.
10.
11. 12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
ally silent, dominantly selectable reporter gene (neo). Mol. Cell Biol. 12, 198 –206. Hashida, T., and Yasumoto, S. (1991) Induction of chromosome abnormalities in mouse and human epidermal keratinocytes by the human papillomavirus type 16 E7 oncogene. J. Gen. Virol. 72, 1569 –1577. zur Hausen, H. (1989) Papillomaviruses in anogenital cancer as a model to understand the role of viruses in human cancers. Cancer Res. 49, 4677– 4681. De Marco, F., and Marcante, M. L. (1993) HPV-16 E6 –E7 differential transcription induced in SiHa cervical cancer cell line by interferons. J. Biol. Reg. Homeostasis 7, 15–21. Nawa, A., Nishiyama, Y., Yamamoto, N., Maeno, K., Goto, S., and Tomoda, Y. (1990) Selective suppression of human papilloma virus type 18 mRNA level in HeLa cells by interferon. Biochem. Biophys. Res. Commun. 170, 793–799. Perea, S. E., Lo´pez-Ocejo, O., Garcı´a-Milia´n, R., and Aran˜a, M. J. (1995) Interferon-alpha elicits downregulation of human papillomavirus 18 mRNA in HeLa cells by selective repression of endogenous viral transcription. J. Interferon Cytokine Res. 15, 495–501. Strander, H. (1989) The action of interferons on virus-associated human neoplasms. Cancer Surv. 8, 755–792. Reichman, R. C., and Strike, D. G. (1989) Pathogenesis and treatment of human genital papillomavirus infections: A review. Antiviral Res. 11, 109 –118. Haglund, S., Lundquist, P. G., Kantell, K., and Strander, H. (1985) Interferon therapy in juvenile laryngeal papillomatosis. Arch. Otolaryngol. 107, 327–332. Lippman, S. M., Kavanagh, J. J., Paredes-Espinosa, F., Delgadillo-Madruen˜o, F., Paredes-Casillas, P., Hong, W. K., Holdener, E., and Krakoff, I. H. (1992) 13-cis-Retinoic acid plus interferon alpha-2a: Highly active systemic therapy for squamous cell carcinoma of the cervix. J. Natl. Cancer Inst. 84, 241–245. Garcı´a-Milia´n, R., Rı´os, M. A., Amigo´, M., Dı´as, D., Guilar, O., Silveira, M., Aran˜a, M. J., and Perea, S. E. (1996) Modulation of human papillomavirus type 16 mRNA in cervical invasive carcinoma patients by interferon-alpha therapy. J. Interferon Cytokine Res. 16, 709 –713. Lo´pez-Ocejo, O., Perea, S. E., Reyes, A., Vigoa, L., and Lo´pezSaura, P. (1993) Partial phenotypic reversion of HeLa cells by long-term interferon-alpha treatment. J. Interferon Res. 13, 369 –375. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1989) Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Feinberg, A. P., and Vogelstein, B. (1983) A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132, 6 –13. Apezteguı´a, I., and Perea, S. E. (1989) Procedimiento para el marcaje de ce´lulas tumorales con Cr 51. Interfero´n Biotecnol. 6, 77–79. Takano, S., and Nagano, Y. (1984) Differentiation in mouse leukemia cells mediated by an immune interferon preparation. J. Interferon Res. 4, 383–388. Dahl, H. (1983) Human interferon and cell growth inhibition. VII. Reversibility of interferon activities. J. Interferon Res. 3, 387–393. Turek, L. P., Byrne, J. C., and Lowy, D. R. (1982) Interferon induces morphologic reversion with elimination of extrachromosomal viral genomes in bovine papillomavirus-transformed mouse cells. Proc. Natl. Acad. Sci. USA 79, 7914 –7918. Samid, D., Chang, E. H., and Friedman, R. M. (1985) Development of transformed phenotype induced by a human ras onco-
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
415
gene is inhibited by interferon. Biochem. Biophys. Res. Commun. 126, 509 –516. Bourgeade, M. F., Rousset, S., Paulin, D., and Chany, C. (1981) Reorganization of the cytoskeleton by interferon in MSVtransformed cells. J. Interferon Res. 1, 323–332. Brouty-Boye, D., Maunoury, M. T., and Lowy, R. (1988) Effects of interferons alpha,beta,gamma alone or in combination on the phenotype of human colon carcinoma cells. Anticancer Res. 8, 197–204. Shillitoe, E. J., and Steele, C. (1992) Inhibition of the transformed phenotype of carcinoma cells that contain human papillomavirus. Ann. NY Acad. Sci. 660, 286 –287. Routes, J. M., and Ryan, S. (1995) Oncogenicity of human papillomavirus- or adenovirus-transformed cells correlates with resistance to lysis by natural killer cells. J. Virol. 69, 7639 –7647. Saxon, P. J., Srivatsan, E. S., and Stanbridge, J. (1986) Introduction of human chromosome 11 via microcell transfer controls tumorigenic expression of HeLa cells. EMBO J. 5, 3461–3466. Koi, M., Morita, H., Yamada, H., Saboh, H., Barrett, J. C., and Oshimura, H. (1989) Normal human chromosome 11 suppresses tumorigenicity of human cervical tumor cell line SiHa. Mol. Carcinogen. 2, 12–21. Smits, P. H. M., Smits, H. L., Minnaar, R. P., Hemmings, B. A., Mayer-Jaekel, R. E., Schuurman, R., Der Noordaa, J. V., and Schegget, J. T. (1992) The 55 kDa regulatory subunit of protein phosphatase 2A plays a role in the activation of the HPV16 long control region in human cells with a deletion in the short arm of chromosome 11. EMBO J. 11, 4601– 4606. Ter Schegeet, J., and Van Der Noordaa, J. (1994) Protein phosphatase 2A and the regulation of human papillomavirus gene activity. In Current Topics in Microbes and Immunity, Vol. 186, pp. 121–129. Springer-Verlag Berlin/Heidelberg. Bosch, F. X., Schwarz, E., Boukamp, P., Fusenig, N. E., Bartsch, D., and zur Hausen, H. (1990) Suppression in vivo of human papillomavirus type 18 E6-E7 gene expression in nontumorigenic HeLa X fibroblast hybrid cells. J. Virol. 64, 4743– 4754. Majewsky, S., Szmurlo, A., Marczak, M., Jablonska, S., and Bollag, W. (1994) Synergistic effect of retinoids and interferon alpha on tumor-induced angiogenesis: Anti-angiogenic effect on HPV-harboring tumor-cell lines. Int. J. Cancer 57, 81– 85. Singh, R. K., Gutman, M., Bucana, C. D., Sanchez, R., Llansa, N., and Fidler, I. J. (1995) Interferons alpha and beta downregulate the expression of basic fibroblast growth factor in human carcinomas. Proc. Natl. Acad. Sci. USA 92, 4562– 4566. Dinney, C. P. N., Bielenberg, D. R., Perrotte, P., Eve, B. Y., Bucana, C. D., and Fidler, I. J. (1998) Inhibition of basic fibroblast growth factor expression, angiogenesis, and growth of human bladder carcinoma in mice by systemic interferon-alpha administration. Cancer Res. 58, 808 – 814. Cohen, T., Nahari, D., Cerem, L. W., Neufeld, G., and Levi, B. Z. (1996) Interleukin 6 induces the expression of vascular endothelial growth factor. J. Biol. Chem. 271, 736 –741. Der, S. D., Zhou, A., Williams, B. R. G., and Silverman, R. H. (1998) Identification of genes differentially regulated by interferon alpha, beta, or gamma using oligonucleotide arrays. Proc. Natl. Acad. Sci. USA 95, 15623–15628. Pepper, M. S., Mandriota, S. J., Vassalli, J. D., Orci, L., and Montesano, R. (1996) Angiogenesis-regulating cytokines: Activities and interactions. In Current Topics in Microbes and Immunity (Gunthert, U., and Birchmeier, W., Eds.), Vol. 213, pp. 31– 67, Springer-Verlag, Berlin/Heidelberg. Melder, R. J., Koenig, G. C., Witwer, B. P., Safabakhsh, N., Munn, L. L., and Jain, R. K. (1996) During angiogenesis, vascular endothelial growth factor and basic fibroblast growth factor
Vol. 277, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
regulate natural killer cell adhesion to tumor endothelium. Nat. Med. 2, 992–997. 40. Uchida, K., Takahashi, A., Miyao, N., Takeda, K., Tsutsumi, H., Satoh, M., and Tsukamoto, T. (1997) Juvenile hemangioma of
the testis: Analysis of expression of angiogenic factors. Urology 49, 285–286. 41. Folkman, J., and Ingber, D. (1992) Inhibition of angiogenesis. Semin. Cancer Biol. 3, 89 –96.
416