Apoptosis and growth inhibition of squamous carcinoma cells treated with interferon-alpha, IFN-beta and retinoic acid are associated with induction of the cyclin-dependent kinase inhibitor p21.

ARTICLE

INTRODUCTION

Type I IFNs are a group of glycoproteins with a wide range of effects on cells of different lineages [1, 2]. In particular, IFN-alpha and IFN-ß have an antiviral, anti-proliferative, and immunomodulating activity [3] that results in an anti-tumoral effect in vitro and in vivo against a variety of cancers including squamous cell carcinoma [4-6].

The essential role of retinoids in regulating normal squamous cell differentiation is well established. Retinoids are derivatives of retinol (vitamin A) that exert a wide variety of effects on vertebrate development, cellular differentiation and homeostasis [7, 8]. The use of retinoids in squamous cell carcinoma of the skin has been evaluated in both preclinical and clinical studies [9, 10]. Recently, impressive clinical responses achieved with RA plus IFN-alpha in acute promyelocytic leukemia, and 13-cis-RA plus IFN-alpha in squamous cell carcinoma of the skin and cervix, have rekindled interest in retinoids-interferon combined therapy [4, 11].

All type I IFNs bind to the type I IFN receptor and elicit a common set of signalling events. IFN-alpha and IFN-ß activate transcription of interferon-stimulated genes (ISG) mainly through the assembly and translocation, from the cytoplasm to the nucleus, of the transcription factor ISGF3 (interferon-stimulated gene factor) [12]. The generation of ISGF3 requires two receptor-associated tyrosine kinases of the Jak family, Tyk2 and Jak1. ISGF3 recognizes the interferon-specific response element (ISRE) within the regulatory sequence of type I IFN-target genes. In addition, the transcription factor GAF (IFN-alpha-activated factor), able to bind gamma activation sequence (GAS) within the promoter of IFN-alpha-target genes, is formed at lower efficiency during IFN-alpha signalling and plays a role in the regulation of gene expression by type I IFN [13-15].

While ISGF-3 and GAF are responsible for the initial transmission of the IFN signal to the nucleus, the proper regulation of the broad range of genes induced by the IFNs involves other transcription factors such as IRF-1. IRF-1 and ISGF-3 have been shown to bind overlapping sequences in the promoters of many IFN-alpha/ß-inducible genes [16]. In addition, to regulate the IFN system,
IRF-1 manifests tumor-suppressive activities and is also required for the induction of apoptosis [14, 17].

Retinoids exert their effects through a different pathway that involves two classes of ligand-dependent transcription factors, the retinoic acid receptors (RARs) and the retinoid X receptors (RXRs), both members of the nuclear receptor superfamily. Etherodimers RAR/RXR recognize the specific retinoic acid response elements (RAREs) of target genes. Interaction with additional transcription factors and mechanisms of protein-protein coregulation are probably involved [18-20]. It is unclear how interferons and retinoids cross-talk in the regulation of their anti-tumoral effects. Some investigations have recently illustrated common target genes involved in mediating these events [21-25].

In order to understand how the combination of IFNs and retinoids might work in SCC, we have previously demonstrated that the combination of IFN-alpha2b and RA affects proliferation and gene expression of human cervical SCC [21]. Inhibition of cell proliferation was associated with apoptosis and IRF-1 gene expression correlated with the phenomenon of RA-and IFN-induced apoptosis [22].

While most extensive clinical experience has been performed with IFN-alpha, it is well established that IFN-ß exerts more powerful anti-proliferative effects than IFN-alpha in several in vitro and in vivo models [5, 26, 27]. An increasing number of recent clinical studies focus on the use of IFN-ß against human carcinomas. Since several clinical trials have already shown the possibility of using IFN-ß both alone and in combination with chemotherapeutic or chemopreventive agents including retinoids, in the treatment of breast carcinoma [5] it might be of interest to explore the possibility of using IFN-ß alone or in combination, in different carcinoma systems.

The present study examines the anti-proliferative effect of recombinant IFN-ß both alone and in combination with RA on the human cervical SCC lines ME180 and SiHa, in order to establish whether IFN-ß is more active than IFN-alpha in this model.

In vitro analyses have shown that IFN-ß exerts anti-proliferative effects, which correlate to the induction of apoptosis at lower doses than IFN-alpha, in both cell lines. Combined treatment with RA increases the growth inhibitory effect of the single agents in ME180, whereas in SiHa it does not. The precise molecular mechanism of this interaction is unclear. Our study suggests that IRF-1, a transcription factor which belongs to the IFN machinery, and the CDK inhibitor p21 might be involved in cellular growth inhibition and in the induction of apoptosis due to IFN and RA in SCC.

MATERIALS AND METHODS

Cell cultures

Two human epidermal cell lines were used: ME180 and SiHa, both obtained from the American Type Culture Collection (Rockville, MD).

MEI80, isolated from an omental metastasis of a rapidly spreading cervical carcinoma, was maintained in McCoy's 5a medium supplemented with 10% fetal bovine serum, previously inactivated at 56° C for 30 min.

SiHa, established from fragments of a primary tissue sample of an undifferentiated squamous carcinoma of the cervix, was maintained in modified MEM supplemented with 10% heat inactivated fetal bovine serum, 1 mM sodium pyruvate, and 1 X nonessential amino acids. Cells were grown to approximately 85-90% of confluence in a humidified atmosphere of 5% CO₂ at 37° C.

RA (Sigma, St. Louis, MO) was added to the medium from a stock solution of 10^{­ 2} M in dimethylsulfoxide (DMSO) to the final concentration. Cells treated with the same volume of DMSO were used as a control in all experiments performed.

Human recombinant IFN-alpha2b (INTRON A; 2 x 10⁸ IU/mg of protein; Shering Corp) was added to the medium from stock solution of 10⁶ IU/ml to the final concentration. Human recombinant IFN-ß (Rebif; 3 x 10⁸ IU/mg of protein; ARES-SERONO) was added to the medium from a stock solution of 10⁴ IU to the final concentration.

To measure cell proliferation, cells were plated in duplicate in 35 mm tissue culture plates at an initial density of 2 or 3 x 10⁵ cells/dish. Twenty-four hours after cell seeding, appropriate dilutions of RA and/or IFN-alpha2b and/or IFN-ß were added to the medium and the cells were grown in the absence or presence of supplements, and DMSO used as a control for RA. DMSO (0,1%) did not affect proliferation of the cells. At the end of the incubation period, the cells were detached after prior washing with 100 mM EDTA followed by a 10 min exposure at 37° C to a solution of 0.1% trypsin-2 mM EDTA in PBS (pH 7.2) and suspended repeatedly to give a single-cell suspension.

Cells were counted using a hemocytometer. Cell mortality was evaluated by the trypan blue dye exclusion method.

In addition, we performed the BrdU incorporation assay. Cells were treated with IFNs for 64 hours and incubated with 20 µM bromodeoxyuridine (BrdU; Sigma) for the last 24 hours. The number of positive cells was determined by fixing for 20 min with 95% ethanol/5% acetic acid, treating for 10 min with 1.5 M HCl, and staining with an anti-BrdU monoclonal antibody (Amersham) followed by a rhodamine conjugated goat anti-mouse antibody (Cappel).

DNA fragmentation analysis

DNA fragmentation was analyzed by a modification of the method reported previously [28]. Fragmented DNA normalized on cell number was electrophoresed in a 1.5% agarose gel in 0,05 M Tris base, and 1 mM EDTA pH 8 and visualized by ethidium bromide staining.

Morphological analysis

For Hoechst 33258 fluorescence staining, detached cells were first collected by centrifugation (5 min) and resuspended in PBS. An aliquot of these (2 x 10⁵ in 40 µl) was seeded on polylysine-coated coverslips for 15 min and fixed with 3% formaldehyde in PBS (pH 7.4) for 10 min at room temperature. The cells adhering to the dish were fixed and processed using the same methods.

After washing, the cells were permeabilized with 0.5% Triton X 100 (Sigma) in PBS for 5 min at room temperature and, after washing, all samples were stained with Hoechst dye, then mounted with glycerol-PBS (2:1) and observed with a Nikon microphot fluorescence microscope. Quantitative evaluation of apoptotic cells by Hoechst staining was performed by counting at least 500 cells at high magnification (x 500).

These analyses were carried out a) in the entire cell population (detaching and adhering cells using a policeman) and b) by counting detached cells and adhering cells separately. However, in consideration of the negative results achieved from adhering cells, only data obtained from detached cells have been reported.

Statistical analysis: the comparison of the results obtained in the same cell line by different treatments versus control cells was performed using the Student's t test. A p value lower than 0.05 was considered significant.

RNA isolation and Northern blot analysis

Total cellular RNA was isolated and purified by the guanidine thiocyanate-cesium chloride method [22], quantitated by absorbance at 260 nm. Thirty µg of each RNA sample were analyzed by agarose-formaldehyde gel electrophoresis and transferred to Hybond^TM-N membranes. The membranes were then prehybridized at 65° C for 1 hour in Church's buffer (0.5 m NaPi pH 6.8; 7% SDS) with 100 mg/ml calf thymus DNA and hybridized for 24 hours at 65° C with 1.5 x 10⁶ dpm/ml random primed ³²P-radiolabeled human 2-5A synthetase cDNA (1.32 Kb EcoRI insert subcloned in pBR) (9-21 cDNA) [22], human p21 cDNA (2-1 Kb subcloned in pCEP) [29, 30] and human glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The blots were sequentially washed for 10-15 min each at 65° C in 2 x SCC-0.1% SDS, 1 x SCC-0.1% SDS and 0.1 x SCC-0.1% SDS and exposed at ­ 80° C to X-ray film.

RNase protection analysis

A 400-bp SmaI restriction fragment was derived from the human IRF₁ cDNA clone, pUC28-8, subcloned in pBS/KS + vector (Stratagene, Madison, WI) and used as a template to generate a ³²P-labeled anti-sense riboprobe following transcription by T7 RNA polymerase using an EcoRI linearized template. A 316-bp SacI-BamHI restriction fragment of pTRI-glyceraldehyde 3-phosphate dehydrogenase (GAPDH)-human cDNA clone (Ambion, J.N.C., Austin, Texas) was used as a template to generate a ³²P-labeled antisense riboprobe following transcription by T₃ RNA polymerase using HindIII linearized template. It was included in each reaction as an internal control. Briefly, the ³²P-labeled probes (25 x 10⁴ cpm) were hybridized for 12-16 hours with 10 µg of total cellular RNA at 55° C. The samples were then digested with RNase A and RNase T1 (Boehringer-Mannheim), extracted and ethanol-precipitated before being separated on 8% polyacrylamide gels containing 8 M urea. ³²P-labeled, sized markers were also run on the gels. The gels were then dried and exposed to X-ray film for 1 to 3 days.

Western blot analysis

Whole cell lysates from ME180 (40 µg) cells were prepared in lysis buffer (0.5% NP-40, 10% glycerol, 50 mM Tris-HCl pH 8, 0.1 mM EDTA, 150 mM NaCl; 1 mM DTT, 0.4 mM phenylmethylsulfonyl fluoride, 3 µg/ml aprotinin, 1 µg/ml leupeptin, 0.5 µg/ml pepstatin, 50 mM NaF and 1 mM sodium ortho-vanadate were freshly added to the buffer before each use), electrophoresed on SDS-polyacrylamide gel and transferred to nitrocellulose for 60 min at 100 V with a Bio-Rad transblot. Western blot detection was performed using rabbit polyclonal antibody against IRF-1 (Santa Cruz Biotechnology), rabbit polyclonal antibody against p21 (WAF1) (Santa Cruz, Biotechnology), and mouse monoclonal antibody against ß-tubulin (ICN Biomedicals) developed with reagents for ECL (Amersham). Protein concentration was determined by the Bio-Rad protein assay.

RESULTS

Anti-proliferative effects

It is well established that both IFNs and retinoids exert anti-proliferative effects in several in vitro models [31, 32]. Previously, we have shown that both RA and IFN-alpha2b inhibit proliferation of ME180 cells and that combined treatment exerts an increased growth inhibitory effect [21].
IFN-alpha2b was also a potent inhibitor of SiHa cell proliferation; conversely, RA was not. Agents administered in combination were as effective, in growth inhibition, as IFN-alpha2b alone in this cell line. The growth inhibition is associated with cell detachment and apoptosis [22]. In particular, two different scenarios can be observed in the two cell lines. The first was described in ME180 cells [22] where both type I IFN and RA inhibit cell proliferation and exert proapoptic effects (with increased activity with their combination). The second is observed in the SiHa cell line where RA: i) does not exert proapoptic effect but increase cell adhesion properties of the cell and counteracts the increase in detachment induced by IFN-alpha2b, suggesting a possible effect of this drug in the reduction of invasiveness, and ii) does not counteract, but indeed increases, the proapoptic effects of IFN-alpha2b on the remaining detaching cells [22, 33]. Different growth features in terms of cell to cell and cell-substrate interactions could explain the different response to RA of SiHa with respect to ME180 [33].

Figure 1A, B shows the effect of IFN-ß. IFN-ß inhibited proliferation of ME180 and SiHa in a dose -and time-dependent manner. Growth inhibition was already evident after 48 hours of treatment. At this time cell detachment began to become evident at a concentration as low as 25 IU/ml for ME180 and at a concentration of 5 IU/ml for SiHa. The strongest effects were observed when IFN-ß was administered at a concentration of 50-100 IU/ml in both cell lines after 72 hours of treatment. It was clear that at equal concentrations of 100 U/ml, both cell lines were more sensitive to IFN-ß than IFN-alpha2b; in fact 100 U/ml of IFN-ß are as effective in growth inhibition as 2,000 IU/ml of IFN-alpha2b (Figure 1C, D). Combined treatment with RA (10^{­ 6} M) and IFN-ß (25 IU/ml) significantly increased the growth inhibitor effect of single agents in ME180 after 48 hours (Figure 1E); conversely, as expected, this regimen was as effective as IFN-ß alone in SiHa (Figure 1F).

The stronger effect of IFN-ß with respect to IFN-alpha2b has been confirmed by BrdU incorporation analyses. Both cell lines were treated as described in Materials and Methods. Figure 1 bis shows a clear, greater reduction of proliferating positive cells by IFN-ß than by IFN-alpha treatment.

DNA fragmentation analysis

Several studies have revealed the involvement of apoptosis in IFN-alpha and RA-inhibition of cellular proliferation [22, 34, 35]. Since the formation of internucleosomal endonucleolitic DNA degradation is considered as a biochemical marker of apoptosis, DNA fragmentation analysis was carried out to determine whether IFN-ß and IFN-ß plus RA could induce programmed cell death. As shown in Figure 2a, IFN-ß used as a single agent was able to induce formation of the typical pattern of DNA degradation in ME180 cells after 48 hours of treatment in a dose-dependent manner. When IFN-ß was administered at a dose of 5 IU/ml the phenomenon was still evident, in contrast to IFN-alpha2b where the same concentration was not effective (data not shown). At equal concentrations of 25 IU/ml, IFN-ß induced a more evident effect than IFN-alpha2b. We have studied the effect of the combined treatment of RA and IFN-ß maintaining a constant (1 mM) RA concentration and using IFN-ß at concentrations of 5 or 25 IU/ml. All combined treatments appeared to increase RA- and IFN-ß-induced DNA fragmentation. In addition, the combined treatments with 25 IU/ml of IFN-ß produced more DNA ladders than combined treatments performed with 25 IU/ml
of IFN-alpha2b. DNA ladders were also evident after IFN-ß treatment in SiHa cells (Figure 2b), while combined treatment with RA was as effective as IFN-ß alone. It clearly appeared that IFN-ß, at a concentration of 25 IU/ml, induced a more evident pattern of DNA degradation than IFN-alpha2b at the same concentration.

Data are in agreement with the pattern of growth inhibition observed.

Percentage of apoptotic cell

To evaluate the percentage of apoptotic cells, we performed analysis of nuclear bodies by Hoechst fluorescence staining. No significant apoptosis was detected in cells adhering to the substrate (not shown) suggesting that cell detachment-associated apoptosis (i.e., anoikis [36]) is induced in our system. Figure 3a, b shows the percentage of apoptosis in detached cells after treatment of SiHa (a) and ME180 (b) carcinoma cells with RA, different doses of IFN-alpha2b and IFN-ß, or their association. The apoptotic cell death induced by IFN-alpha2b and ß is a dose-dependent phenomenon. In fact, a higher percentage of apoptotic cells was observed at a concentration of 2,000 IU/ml (ME180: IFN-alpha2b 23.98%, IFN-ß 80.70%; SiHa: IFN-alpha2b 10.98%, IFN-ß 30.28%). Both cell lines were more sensitive to IFN-ß than to IFN-alpha2b. Administration of 10^{­ 6} M of RA was capable of inducing apoptosis only in ME180 cells (35.30%), whereas in SiHa this percentage (2.01%) was not significantly increased with respect to control cells (1.30%) in agreement with the DNA fragmentation analysis and the anti-proliferative effect. On the other hand, combined treatments of both IFNs with RA significantly increased the percentage of apoptosis in detached cells of both cell lines; the synergistic effect was more evident using IFN-ß. These data, in particular those regarding cell proliferation, clearly indicated that the ME180 cell line was more sensitive to RA/IFNs combined treatments as compared to SiHa cell line and also underlined that IFN-ß was more active as compared to IFN-alpha2b.

Induction of 2-5A synthetase mRNA expression

To evaluate the molecular mechanism of the effect of IFN-ß and its combination with RA, we analysed the expression of IFN-induced genes that have been suggested to have a role in the growth inhibitory action of IFNs [37, 38].

We first analysed the 2-5A synthetase gene. This codes for an enzyme able to synthesize 2-5A oligomers which specifically activate a 2-5A-dependent ribonuclease to cleave cellular ribosomal and messenger RNAs. Northern blot analysis was performed using human cDNA 9-21 [39] as a probe. The expression of this gene was clearly induced with similar kinetics in both cell lines (data not shown). Figure 4 shows mRNA expression after 15 hours of treatment with IFN-ß (25 IU/ml) or RA (1 mM) in both cell lines. When RA was combined with IFN-ß, an enhancement in the expression of the 2-5A synthetase gene was observed in ME180 cell lines, whereas combined treatment was as effective as IFN treatment in the SiHa cell line (Figure 4 bis).

Induction of IRF1 expression

IRF-1, a transcription factor which belongs to the IFN machinery, has been considered one of the potential target genes for the anti-proliferative function of IFNs and it has recently been demonstrated to manifest tumor suppressor activities and to participate in the regulation of apoptosis in specific systems [14]. IRF-1 and ISGF-3 have been shown to bind overlapping sequences in the promoters of many IFN-alpha/ß-inducible genes [16].

Previous results, concerning the inhibition of cell proliferation exerted by IFN-alpha2b and/or RA, strengthened the correlation between induction of IRF-1 and cell growth inhibition [21, 22]. Analysis of IRF-1 gene expression was carried out in ME180 and SiHa cell lines treated with IFN-ß and/or RA for the indicated times. The RNase protection analysis showed the induction of IRF1 gene expression by IFN-ß. Significant upregulation was observed at 1 hour and 3 hours of treatment in both cell lines when IFN-ß was used at doses as low as 25 IU/ml. On the other hand, IFN-alpha2b exerted the same effect only when administered at a dose of 2,000 IU/ml (see 22). Combined treatment with RA (1 mM) slightly increased the IFN-ß-induced expression after 3 hours of treatment in ME180 cell lines, whereas no increase of IFN-ß induction was observed in the SiHa cell line (Figure 5 bis).

As already demonstrated [22], RA induces IRF-1-gene expression in ME180 and not in SiHa cell lines.

In addition, we performed Western blot analysis on extracts from ME180 cells responsive to both agents treated with IFN-ß or alpha2b, to analyse the IRF-1 expression at the protein level (Figure 6). Upregulation of IRF-1 messenger RNA was maximum at 1 hour. The protein level was maximum at 5 hours persisting significantly up to 48 hours. IFN-ß induced upregulation of IRF-1 at lower doses than IFN-alpha2b. The kinetics of induction by RA appeared to be delayed with respect to that of type I IFN. Combined treatment with RA increased IFN-alpha2b and -ß upregulation at 5 hours of treatment that subsequently reach a plateau.

Induction of CDKi p21

The CDKi p21, encoded by the Cip1 gene, can associate with different CDK cyclin complexes and inhibits the kinase activity that is required for cell cycle progression. p21 gene expression can be transcriptionally regulated by the tumor suppressor p53 [29] and has been proposed to be involved in the control of cellular senescence as well as neoplasia [30]. The existence of an IRF-1-dependent pathway of DNA damage-induced apoptosis, distinct from the p53-mediated apoptotic pathway, has recently been proposed, probably involving a regulation of p21 gene expression by IRF-1. The p21 promoter contains three potential IRF-1 binding sites [14, 17]. In order to start analysing the possible role of p21 in ME180 growth inhibition and induction of apoptosis, we performed Northern blot and immunoblot analysis to evaluate p21 expression in extracts of ME180 cells. A basal level of expression was observed in control cells that appeared increased when either serum starvation was performed or the culture reached confluency (data not shown), indicating that p21 gene expression is cell cycle regulated in our system. To evaluate the p21 expression after treatment with IFN-alpha2b, ß or RA, cells were seeded at low density. Under this condition, IFN-alpha2b (2,000 IU/ml), RA (1 mM), and IFN-ß (25 IU/ml) treatment appeared to increase the p21 mRNA (Figure 7 bis) as well as protein expression with different kinetics of induction (Figure 8). The protein level appeared increased after 18 hours of IFN-alpha or -ß treatment and decreased to the control level thereafter. On the other hand, RA treatment increased protein expression at each time point tested.

DISCUSSION

Combined therapy with retinoids and IFNs of certain hematologic malignancies and SCC is known to improve their individual anti-tumor effectiveness [40]. Increasing evidence suggests that the balance between cell proliferation and cell death plays a central role in the maintenance of normal tissue homeostasis. Dysregulation of apoptosis may lead to an altered number of cells within the tissue and, finally, to malignant transformation [41]. Recent evidence indicates oncogenes and tumor suppressor genes are able to regulate the susceptibility of tumor cells to undergo apoptotic cell death [17, 42]. The recognition of apoptosis as a mechanism of action of many chemotherapeutic agents, including interferons and retinoids, leads to novel experimental approaches aimed at stimulating apoptotic pathways in order to improve therapeutic response [41].

With the objective of exploring how the combination of IFNs and retinoids might work in cervical SCC, and which molecular mechanisms might be responsible for their effectiveness, we previously examined the effects of IFN-alpha2b and RA on two human cell lines from squamous carcinoma of the cervix: ME180 and SiHa [21, 22].

The goal of this investigation was to analyse the antiproliferative effects of IFN-ß in vitro and its combination with RA with respect to IFN-alpha2b against SCC. Several studies have already demonstrated that IFN-ß has a much stronger inhibitory growth effect than IFN-alpha in several in vitro and in vivo models [6, 43, 44]. Although it was assumed that IFN-alpha and IFN-ß interact with the same receptor, there is accumulating evidence that significant differences exist between the effects of IFN-alpha and IFN-ß in vitro and in vivo [3, 45, 46], but the mechanisms of signal generation specific for the different type I IFNs have not been identified. Recent work suggests the existence of a gene that is selectively induced by IFN-ß, but not IFN-alpha [47]. Evidence suggest the possibility that interactions induced by IFN-ß at the receptor level selectively regulate the expression of genes involved in IFN-ß-specific biological responses [45, 46]. Our studies on ME180 and SiHa cell lines provide evidence that this squamous carcinoma model is more susceptible to IFN-ß than IFN-alpha in terms of growth inhibition. In fact, lower doses of IFN-ß than IFN-alpha2b are effective in growth inhibition in both cell lines. It was clear that RA cooperates with IFN-ß with respect to the growth-inhibitory effect in ME180. On the other hand, combined treatment is as effective as IFN alone on SiHa cells. Although combined treatment does not increase inhibitory effects in the SiHa cell line, RA does not interfere with IFN's ability to inhibit growth. The inhibition of proliferation induced by IFN-ß and/or RA correlates with the induction of apoptosis.

Since we had previously correlated 2-5A synthetase and IRF-1 gene expression with growth inhibition and induction of apoptosis exerted by IFN-alpha2b and RA in our system [21, 22], we analysed these IFN-target genes to observe any difference in their expression after IFN-ß in single or combined treatment with RA. In fact recently, an ISRE-ISGF3-independent pathway for IFN-alpha and -ß induction of the IRF1 gene has been described [48].

The 2-5A synthetase system is a well known mediator of cellular responses to IFNs. Moreover, interesting correlations exist between 2-5A-dependent RNase and the fundamental control of cell growth and differentiation [49]. Our analyses show that IFN-ß can induce 2-5A gene expression at doses significantly lower than IFN-alpha2b in the cell system analysed. The presence of RA in combination with IFN-ß increases the transcriptional rate of 2-5A synthetase gene expression in comparison to IFN-ß alone in ME180 while in SiHa it does not. Upregulation of gene expression correlates with growth inhibition observed in both cell lines.

IRF-1 was first identified as a regulator of type I IFN genes, and its involvement in the regulation of some IFN-inducible genes, such as the 2-5A synthetase, has been also proved [50]. More recently, it has also been proposed that IRF-1 manifests tumor suppressor activity, activating a set of genes, the products of which are necessary for the negative regulation of cell growth, analogously to the tumor suppressor genes p53 [17]. The existence of an IRF-1-dependent pathway of DNA damage-induced apoptosis in T-lymphocytes, distinct from the p53-mediated apoptotic pathway, has recently been reported. Expression of the IRF-1 gene itself appears regulated during the normal cell cycle [51]. The induction of IRF-1 might account for the ability of IFNs to inhibit cell proliferation. Recent work has proposed its analogous role in RA inhibition of cell growth [22, 23]. In light of these hypotheses, we investigated the effect of IFN-ß and IFN-ß plus RA on IRF-1 gene expression in both cell lines. We had already shown that RA alone affected IRF-1 gene expression in ME180 but not in SiHa, in agreement with growth inhibition and apoptosis data [22].

Our analyses revealed that IFN-ß was more effective in inducing IRF-1 gene expression than IFN-alpha2b in both cell lines. IRF-1 mRNA expression is transiently upregulated by IFN-ß in both cell lines, while combined treatment with RA is more effective only in ME180 which is in full agreement with cell growth inhibition analyses. IRF-1 protein remained stable up to 48 hours and correlated with the time course for induction of apoptosis. Thus, taken together 2-5A synthetase and IRF-1 analyses strengthen the correlation between induction of these two IFN-target genes and cell growth control. Even though our studies on ME180 show that the presence of RA in combination with IFN-ß increases the transcription of 2-5A synthetase and IRF-1 gene expression in comparison to IFN-ß alone, the question of how the complex transcriptional machinery of IFN-responsive gene can be influenced by the presence of RA still remains unanswered.

Interestingly, in ME180, treatment with IFN-alpha2b, IFN-ß and RA also increases the level of CDKi p21/WAF1/CIP1 protein. These analyses parallel with Northern blot data. Evidence is accumulating that the CDKi p21 is a target of extra- and intracellular signals that regulate growth, differentiation and apoptosis [52, 53]. Several agents that induce cell differentiation, including RA and IFNs, have been shown to up-regulate p21 expression at the mRNA or protein level [52-54]. It has been demonstrated that the gene is transcriptionally regulated by p53 and MyoD [29]. In addition, it has recently been suggested that transcriptional induction of the gene is dependent also upon IRF-1 [17].

Taken together, these results show that IFN-ß is clearly more effective than IFN-alpha in inducing cell growth inhibition and apoptosis in squamous carcinoma cells and also show a relevant in vitro anti-proliferative activity of IFN-ß and RA, supporting their use in further clinical investigations. The anti-proliferative effects and apoptosis phenomenon, due to IFN-ß and RA, are coupled with an increased induction of genes thought to be involved in cell cycle control such as IRF1 and CDKi p21.

More studies are required to define the roles of IRF-1 and p21 in cross-talking between RA and IFN in light of the possibility that IFNs and retinoids may interact in the induction of differentiation and G1 cell cycle arrest [55]. Further analysis of the p21 promoter could elucidate the mechanism which underlies its upregulation, by both IFN-alpha and -ß as well as by RA in our system.

CONCLUSION

Acknowledgements

We are very grateful to Dr. T. Taniguchi and Dr. K. Sugiyama for providing us with IRF-1 cDNA and Dr. B. Vogelstein for p21 reagents. We thank Sabrina Tocchio and Giulia Pacetto for editorial assistance. This work was supported by grants from the Consiglio Nazionale delle Ricerche Progetto Finalizzato Applicazioni Cliniche della Ricerca Oncologica and n. 97.03974.CT04 and MURST 40%.

REFERENCES

1. Petska S, Langer A J, Zoon K, Samuel C. 1987. Interferons and their action. Annu. Rev. Biochem. 56: 727.

2. Romeo G, Affabris E, Battistini A, Rossi G B. 1992. Cell and animal effects of IFNs: natural occurences, differentiation effects and mechanisms. In: Baron S, Coppenhaver D H,
Dianzani F, Flaishmann W R Jr, Hughes T K Jr, Klimpel G R, Nielsen D W, Stanton G J, Tyring S K, eds. Interferon: principles and medical applications. Galveston, TX: Texas Press: 271-287.

3. Kalvakolanu D V, Borden E C I. 1996. An overview of the interferon system: signal transduction and mechanisms of action. (Review) Cancer Investigation 14: 25.

4. Lippman S M, Lotan R, Schleuniger U. 1997. Retinoid-interferon therapy of solid tumors. (Review) Int. J. Cancer 70: 481.

5. Recchia F, Sica G, De Filippis S, Discepoli S, Rea S, Torchio P, Frati L. 1995. Interferon-ß, retinoids, and tamoxifen in the treatment of metastatic breast cancer: a phase II study.
J. Interferon Cytokine Res. 15: 605.

6. Lindner D J, Borden E C, Kalvakolanu D V. 1997. Synergistic anti-tumor effects of a combination of interferons and retinoic acid on human tumor cells in vitro and in vivo. Clin. Cancer Res 3: 931.

7. Gudas L J, Sporn M B, Roberts A B. 1994. Cellular biology and biochemistry of the retinoids. In: Sporn M B, Roberts A B, Goodman D S, eds. The Retinoids: 443-520, New York, Raven Press.

8. Sporn M B, Roberts A B, Goodman D S. 1994. Eds The Retinoids. New York, Raven Press.

9. Lotan R. 1994. Suppression of squamous cell carcinoma growth and differentiation by retinoids. (Review) Cancer Res. (suppl.) 54: 1987s.

10. Lotan R. 1996. Retinoids in cancer chemoprevention. (Review) FASEB J. 10: 1031.

11. Nason-Burchenal K, Gandini D, Botto M, Allopenna J, Seale J R, Cross N C, Goldman J M, Dmitrosky E, Pandolfi P P. 1996. Interferon augments PML and PML/RAR alpha expression in normal myeloid and acute promyelocytic cells and cooperates with all-trans retinoic acid to induce maturation of a retinoids-resistant promyelocytic cell line. Blood 88: 3926.

12. Levy D E, Kessler D S, Pine R, Darnell J E. 1989. Cytoplasmic activation of ISGF3, the positive regulator of interferon-alpha-stimulated transcription, reconstituted in vitro. Genes Dev. 3: 1362.

13. Darnell J E Jr, Kerr I M, Stark G R. 1994. Jak-STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins. (Review) Science 264: 1415.

14. Taniguchi T, Harada H, Lamphier M. 1995. Regulation of the interferon system and cell growth by the IRF transcription factors. (Review) J. Cancer Res. Clin. Oncol. 121: 516.

15. Ihle J N. 1996. STATs: signal transducers and activators of transcription. (Review) Cell 84: 331.

16. Veals S A, Schindler C, Leonard D, Fu X Y, Aebersold R, Darnell J E Jr, Levy D E. 1992. Subunit of an alpha-interferon-responsive transcription factor is related to interferon regulatory factor and Myb families of DNA-binding proteins. Mol. Cell Biol. 12: 3315.

17. Tanaka N, Ishihara M, Lamphier M S, Nozawa H, Matsuyama T, Mak T W, Aizawa S, Tokino T, Oren M, Taniguchi T. 1996. Cooperation of the tumour suppressor IRF-1 and p53 in response to DNA damage. Nature 382: 816.

18. Leid M, Kastener P, Chambon P. 1992. Multiplicity generates diversity in the retinoic acid signalling pathways. (Review) Trends Biochem. Sci. 17: 427.

19. Chen J Y, Penco S, Ostrowski J, Balaguer P, Pons M, Starrett J E, Reczek P, Chambon P. 1995. RAR-specific agonist/antagonist which dissociate transactivation and AP-1 tranespression inhibit anchorage-independent cell proliferation. EMBO J. 14: 1187.

20. Chen J Y, Clifford J, Zusi C, Starrett J, Tortolani D, Ostrowski J, Reczek P R, Chambon P, Gronemeyer H. 1996. Two distinct action of retinoid-receptor ligands. Nature 382: 819.

21. Lancillotti F, Giandomenico V, Affabris E, Fiorucci G, Romeo G, Rossi G B. 1995. interferon-alpha-2b and retinoic acid combined treatment affect proliferation and gene expression of human cervical carcinoma cells. Cancer Res. 55: 3158.

22. Giandomenico V, Lancillotti F, Fiorucci G, Percario Z A, Rivabene R, Malorni W, Affabri E, Romeo G. 1997. Retinoic acid and IFN inhibition of cell proliferation is associated with apoptosis in squamous carcinoma cell lines: role of IRF-1 and TGase II-dependent pathways. Cell Growth Diff. 8: 91.

23. Matikainen S, Ronni T, Hurme M, Pine R, Julkunen I. 1996. Retinoic acid activates interferons regulatory factor-1 gene expression in myeloid cells. Blood 88: 114.

24. Kolla V, Lindner D J, Xiao W, Borden E C, Kalvakolanu D V. 1996. Modulation of interferon (IFN)-inducible gene expression by retinoic acid. Up-regulation of STAT1 protein in IFN-unresponsive cells. J. Biol. Chem. 271: 10508.

25. Gianni M, Terao M, Fortino I, Li Calzi M, Viggiano V, Barbui T, Rambaldi A, Garattini E. 1997. STAT1 is induced and activated by all-trans retinoic acid in acute promyelocytic leukemia cells. Blood 89: 1001.

26. Lokshin A, Mayotte J E, Levitt M L. 1995. Mechanism of interferon beta-induced squamous differentiation and programmed cell death in human non-small-cell lung cancer cell lines. J. Natl. Cancer Inst. 87: 206.

27. Kopf J, Hanson C, Delle U, Weimarck A, Stierner U. 1996. Action of interferon-alpha and -ß on four human melanoma cell lines in vitro. Anti-cancer Res. 16: 791.

28. Laurent-Crawford A G, Krust B, Muller S, Rivière Y, Rey-Cuillé M A, Bèchet J M, Montagnier L, Hovanessian A G. 1986. The cytopathic effect of HIV is associated with apoptosis. Virology 185: 829.

29. el-Deiry W S, Tokino T, Velculescu V E, Levy D B, Parsons R, Trent J M, Lin D, Mercer W E, Kinzler K W, Vogelstein B. 1993. WAF1, a potential mediator of p53 tumor suppression. Cell 75: 817.

30. el-Deiry W S, Tokino T, Waldma T, Oliner J D, Velculescu V E, Burrell M, Hill D E, Healy E, Rees J L, Hamilton S R, Kinzler K W, Volgelstein B. 1995. Topological control of p21 WAF1/CIP1 expression in normal and neoplastic tissues. Cancer Res. 55: 2910.

31. Toma S, Monteghirfo S, Tasso P, Nicolo G, Spadini N, Palumbo R, Molina F. 1994. Anti-proliferative and synergistic effect on interferon alpha-2a, retinoids and their association in established human cancer cell lines. Cancer Lett. 82: 209.

32. Eisenhauer E A, Lippman S M, Kavanagh J J, Parades-Espinoza M, Arnold A, Hong W K, Massimini G, Schleuniger U, Bollag W, Holdener E E, et al. 1994. Combination 13-cis-retinoic acid and interferon alpha-2a in the therapy of solid tumors. (Review) Leukemia 8: 1622.

33. Matarrese P, Giandomenico V, Fiorucci G, Rivabene R, Straface E, Romeo G, Affabris E, Malorni W. Anti-proliferative activity of interferon-alpha and retinoic acid in SiHa carcinoma cells: the role of cell adhesion. Int. J. Cancer (in press).

34. Rodriguez Villanueva J, McDonnel T J. 1995. Induction of apoptotic cell death in non-melanoma skin cancer by interferon-alpha. Int. J. Cancer 61:110.

35. Zhang L X, Mills K J, Dawson M I, Collins S J, Jetten A M. 1995. Evidence for the involvement of retinoic acid receptor RAR alpha-dependent signaling pathway in the induction of tissue transglutaminase and apoptosis by retinoids. J. Biol. Chem. 270: 6022.

36. Frisch S M, Vuori K, Ruoslahti E, Chan-Hui P Y. 1996. Control of adhesion-dependent cell survival by focal adhesion kinase. J. Cell Biol. 134: 793.

37. Wells V, Mallucci L. 1985. Expression of the 2-5A system during the cell cycle. Exp. Cell Res. 159: 27.

38. Yaffe A, Schwarz Y, Hacohen D, Kinar Y, Nir U, Salzberg S. 1996. Inhibition of 2-5A synthetase expression by antisense RNA interferes with interferon-mediated anti-viral and anti-proliferative effects and induces anchorage-independent cell growth. Cell Growth Differ. 7: 969.

39. Merlin G, Chabath J, Benech P, Metz R, Revel M. 1983. Molecular cloning and sequence of partial cDNA for interferon-induced (2'-5')oligo(A) synthetase mRNA from human cells. Proc. Natl. Acad. Sci. USA 80: 4904.

40. Hoffmann W, Schiebe M, Hirlne P, Souchon R, Clemens M, Adamietz I, Bamberg M. 1997. 13-cis retinoic acid and interferon-alpha ± irradiation in the treatment of squamous-cell carcinomas. Int. J. Cancer, 70: 475.

41. Kalemkerian G P, Ramnath N. 1996. Retinoids and apoptosis in cancer therapy. Apoptosis 1: 11.

42. McDonnell T J, Meyn R E, Robertson L E. 1995. Implications of apoptotic cell death regulation in cancer therapy. (Review) Semin. Cancer Biol. 6:53.

43. De Marco F, Marcante M L. 1993. HPV-16 E6-E7 differential transcription induced in SiHa cervical cancer cell line by interferons. J. Biol. Regul. Homeost. Agents 7: 15.

44. Arbaje Y M, Bittner G, Yingling J M, Storer B, Schiller J H. 1993. Anti-proliferative effects of interferons-alpha and -ß in combination with 5-fluorouracil, cisplatin, and cis- and trans-retinoic acid in three human lung carcinoma cell lines. J. Interferon Res. 13: 25.

45. Abramovich C, Shulman L M, Ratovitski E, Harroch S, Tovey M, Eid P, Revel M. 1994. Differential tyrosine phosphorylation of the IFNAR chain of the type I interferon receptor and of an associated surface protein in response to IFN-alpha and IFN-ß. EMBO J. 13: 5871.

46. Platanias L C, Uddin S, Domanski P, Colamonici O R. 1996. Differences in interferon-alpha and -ß signaling. Interferon-ß selectively induces the interaction of the alpha and ß subunits of the type I interferon receptor. J. Biol. Chem. 271: 23630.

47. Rani M R S, Foster G R, Leung S, Leaman D, Stark G R, Ransohoff R M. 1996. Characterization of ß-R1, a gene that is selectively induced by interferon-ß (IFN-ß) compared with
IFN-alpha. J. Biol. Chem. 271: 22878.

48. Haque S J, Williams B R. 1994. Identification and characterization of an interferon (IFN)-stimulated response element-IFN-stimulated gene factor 3-independent signaling pathway for IFN-alpha. J. Biol. Chem. 269: 19523.

49. Hassel B A, Zhou A, Sotomayor C, Maran A, Silverman R H. 1993. A dominant negative mutant of 2-5A-dependent RNase suppresses anti-proliferative and anti-viral effects of interferon. EMBO J. 12:3297.

50. Pine R. 1992. Constitutive expression of an ISGF2/IRF1 transgene leads to interferon-independent activation of interferon-inducible genes and resistance to virus infection. J. Virol. 66: 4470.

51. Vaughan P S, van Wijnen A J, Stein J L, Stein G S. 1997. Interferon regulatory factors: growth control and histone gene regulation it's not just interferon anymore. (Review) J. Mol. Med. 75: 348.

52. Sheikh M S, Chen Y Q, Smith M L, Fornace A J Jr. 1997. Role of p21 Waf1/Cip1/Sdi1 in cell death and DNA repair as studied using a tetracycline-inducible system in p53-deficient cells. Oncogene 14: 1875.

53. Subramaniam P S, Johnson H M. 1997. A role for the cyclin-dependent kinase inhibitor p21 in the G1 cell cycle arrest mediated by the type 1 interferons. J. Interferon Cytokine Res. 17: 11.

54. Liu M, Iavarone A, Freedman L P. 1996. Transcriptional activation of the human p21^WAF/CIP1 gene by retinoic acid receptor. Correlation with retinoid induction of U937 cell differentiation. J. Biol. Chem. 271: 31723.

55. Jiang H, Lin J, Su Z Z, Collart F R, Huberman E, Fisher P B. 1994. Induction of differentiation in human promyelocytic HL-60 leukemia cells activates p21, WAF1/CIP1 expression in the absence of p53. Oncogene 9: 3397.