IL (Interleukin)-1α Promotes Nuclear Factor-κB and AP-1-induced IL-8 Expression, Cell Survival, and Proliferation in Head and Neck Squamous Cell Carcinomas¹

IL-1⁵ is a cytokine released after stress or injury that mediates a variety of protective responses by the cell and the organism (1). IL-1 can induce the activation of immediate-early transcription factors and genes that promote the survival and proliferation of cells (2) and cytokines that mediate protective inflammatory and angiogenesis responses (3, 4). Previously, we reported that HNSCCs constitutively express IL-1α and a repertoire of IL-1-inducible cytokines, including IL-6, IL-8, and GM-CSF (5). These cytokines have been detected in the supernatants of cell lines, in tumor specimens, and in serum from patients with HNSCC (5, 6). Several of these cytokines seem to contribute to the malignant phenotype of SCCs and other cancers (6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17). IL-1α has been reported to directly enhance the proliferation of transformed human keratinocytes and cervical SCC cell lines (10). IL-8 has been shown to act as an autocrine growth factor in HNSCC, melanoma, and bronchogenic carcinoma, and IL-8 and GM-CSF have been reported to have proinflammatory and angiogenic activities that enhance tumor growth and metastatic potential (14, 15, 16, 17). However, the nature and biological importance of the mechanism(s) that initiate the coactivation of this repertoire of cytokines in HNSCC and other cancers has not been well defined.

In considering the possible mechanisms responsible for coexpression of IL-1α, IL-6, IL-8, and GM-CSF, we noted that the promoter region of all four cytokine genes share recognition sites for immediate-early transcription factors NFκB/Rel A and AP-1, and found that differences in cytokine expression correlated with activation of these transcription factors (18, 19, 20). Additional studies to define the contribution of NFκB/Rel A and AP-1 in the activation of the IL-8 gene in HNSCC demonstrated that mutation of the NFκB and AP-1 site within the IL-8 promoter reduced IL-8 reporter gene activity (18). The expression of a mutant inhibitor-κB in human or murine SCC cells inhibited the activation of NFκB and the expression of cytokines (19, 20). Furthermore, inactivation of NFκB in HNSCC inhibited cell survival and growth in vivo, indicating that the activation of NFκB plays an important role in the malignant phenotype of HNSCC (19). Activation of NFκB has also been detected and shown to prevent apoptosis in lymphomas and breast and prostate cancer (21), but the factor(s) responsible for the activation of NFκB in HNSCC and these other cancers has not been defined. Because IL-1 has been shown to be an inducer of NFκB activation and IL-8 expression (2, 3, 4), we hypothesized that the expression of IL-1α by HNSCC could play an important role in the autocrine activation of transcription factors NFκB and AP-1, the expression of cytokines such as IL-8, and the promotion of cell survival and growth.

Previously we explored whether IL-1α could promote the coexpression of IL-8 and GM-CSF in HNSCC, and we found that increased concentrations of recombinant IL-1α or the expression of IL-1α after transfection was able to induce IL-8 and GM-CSF production by HNSCC cell lines (5). In contrast, the addition of anti-IL-1α neutralizing antibody and IL-1RA blocked the cytokine production induced by exogenous IL-1α but had little effect on constitutive expression (5). In this study, we examined the effects of modulating exogenous and endogenously expressed IL-1α on the activation of NFκB, AP-1, and the expression of the proangiogenic cytokine IL-8; and on the survival and proliferation of HNSCC cell lines. We demonstrated that exogenous and endogenous IL-1α contributes to the transcriptional activation of NFκB and AP-1, to the expression of IL-8, and to the cell survival and growth of HNSCC in vitro.

Goat antihuman IL-1α, goat IgG control, and recombinant IL-1RA were purchased from R&D Systems, (Minneapolis, MN). Plasmid pIgk-luc (immunoglobulin κB-luc containing two copies of the NFκB binding site upstream of the minimal promoter fused with the luciferase gene) was described previously (22). IL-8 promoter reporter gene plasmids, in which expression of a luciferase gene is directed by a wild-type or a mutant IL-8 promoter, were described previously (23). These plasmids contain the 133-bp sequence upstream from the transcription start site of the IL-8 gene. Constructs were used which contain point mutations in the putative NFκB and AP-1 sites (23). PCMVLacZ consists of a LacZ gene inserted between the CMV promoter and bovine growth hormone poly(A) signal sequence in pcDNA3 (Invitrogen, Carlsbad, CA).

To make vectors that express human IL-1α under control of a CMV promoter, a BamHI DNA fragment containing the complete coding sequence of human IL-1α was isolated from phIL1AcDNA (ATCC no. 65259) and ligated into pcDNA3 (Invitrogen) that was predigested by BamHI. Recombinants were screened for clones with inserts in sense orientation, yielding pcDNAhIL-1α (IL-1α). The construct lacking a signal sequence for secretion-encoding intracellular expression of IL-1RA (pcDNAicIL1-RA) was a kind gift from the laboratory of Dr. Steve Haskill, University of North Carolina, Chapel Hill, NC (24).

UM-SCC-9 and -11B lines were derived from human SCC of the upper aerodigestive tract, with informed consent, at the University of Michigan (Ann Arbor, MI). The cell lines were from patients exhibiting an aggressive clinical course who developed local or regional recurrence, and who died within two years of diagnosis (5, 6). The cell lines established from single isolates of a patient specimen are designated by a numeric designation; where established from isolates from two time points, the designation includes an alphabetical suffix, i.e., “A” or “B.” The cell lines were maintained in Eagle’s minimal essential media supplemented with 10% fetal bovine serum and penicillin/streptomycin. For ELISA, 12 ml of fresh medium were added to tumor cell lines when 60–80% confluent in 75-cm² tissue culture flasks, and cells were collected after 48 h. Supernatants were centrifuged at 1500 × g to remove particles, and aliquots were stored frozen at −80° until use in ELISA. Cell number in the culture flasks was determined to standardize the quantity of cytokine secreted per 10⁶ tumor cells.

ELISA kits for specific cytokines were purchased from R&D Systems and Endogen (Cambridge, MA) and used according to the manufacturer’s protocol. Each sample was tested in duplicate in each experiment. After development of the colorimetric reaction, the absorbance (OD) at 450 nm was quantitated by an 8-channel spectrophotometer (Bio-Tek Systems, Winooski, VT), and the OD readings were converted to pg/ml of cytokines based on standard curves analyzed in each assay. If the OD readings exceeded the linear range of the standard curves, the ELISA assay was repeated using serial dilutions of the supernatants. Each sample was tested by at least two independent ELISA measurements, and the data were calculated from the mean of the two to four tests for each sample. The lower limit of sensitivity of these ELISAs for detecting IL-1α and IL-8 was found to be 4 and 10 pg/ml, respectively.

Cell extracts were made according to the method of Beg et al. (25), using a procedure from the laboratory of Dr. U. Siebenlist (National Institute of Allergy and Infectious Diseases, NIH) with minor modifications. Briefly, 1 × 10⁶ cells were rinsed with PBS and harvested from tissue culture flasks by gentle scrapping. After spinning down the cells and removing the PBS, an equivalent volume of lysis buffer [50 mm Tris (pH 7.4), 100 mm NaCl, 50 mm NaF, 30 mm sodium PP_i, and 0.5% NP40) was added to resuspend the cell pellet. A protease inhibitor cocktail tablet (Complete, Mini; Boehringer, Mannheim, Germany) was added per 10 ml of lysis buffer before use. Then samples were spun at 18,000 × g for 20 min at 4°C. Supernatants were aliquoted, snap frozen, and stored at −80°C. Protein concentrations were determined using a BCA protein assay kit following the manufacturer’s instructions (Pierce, Rockford, Illinois).

Double-stranded DNA oligonucleotide probes for NFκB, AP-1, and OCT-1 were synthesized commercially (Promega, Madison WI). The consensus sequences used were: NFκB, 5′-AGTTGAGGGGACTTTC-CCAGGC-3′; AP-1, 5′-CGCTTGATGAGTCAGCCGGAA-3′; and OCT-1, 5′-TGTCGA-ATGCAAATCACTAGAA-3′. Oligonucleotide probes were labeled using T4 polynucleotide kinase (Promega) and [γ-³²P]ATP (6000 Ci/mmol; Amersham, Arlington Heights, IL). EMSA was performed as described previously (19), with minor modifications. Briefly, 5–10 μg of whole cell extracts were incubated with 1 μg of polydI-dC (Pharmacia Biotech, Piscataway, NJ) alone or with unlabeled wild-type or mutant oligonucleotide in 20 μl of buffered binding mixture [20 mm HEPES (pH 7.9), 5 mm MgCl, 60 mm KCl, 1 mm DTT, 0.1% NP40, and 10% Glycerol] for 10 min at room temperature. Then a 20,000-cpm ³²P-labeled probe was added, and the reaction mixture was incubated for another 30 min at room temperature. Each reaction mixture was loaded immediately onto a 5% nondenaturing polyacrylamide gel made in 0.25× Tris-borate EDTA. Gels were run at 200 V for 2 h. After drying, gels were subject to autoradiography.

UM-SCC-9 and -11B cells were plated at 2 × 10⁵ cells/well in a six-well plate and allowed to adhere overnight. For transfections, 2 μg of −133 IL-8-luc or pIgk-luc reporter DNA were added in 100 μl of OptiMEM medium, and mixed with 5 μl of Lipofectin (Life Technologies, Inc., Gaithersburg, MD) in 100 μl OptiMEM medium. pCMVLacZ or pRSVLacZ (0.2 μg) was cotransfected to confirm transfection efficiency. The mixture was incubated for 30 min, and 0.8 ml of OptiMEM was added to the DNA/Lipofectin solution and gently mixed. Then 1 ml of the solution was added to each well and the plates were incubated at 37° for 5 h. Then the transfection media was replaced with Eagle’s minimal essential medium plus 10% serum, and incubation of the cells was continued for 48 h. The cells were harvested and reporter gene activities were assayed using the Dual-Light Luciferase and β-Galactosidase Reporter Gene Assay System (Tropix, Bedford, MA), and chemiluminescence was measured by a Monolight 2010 luminometer (Analytical Luminescence Lab, San Diego, CA).

UM-SCC-9 or -11B cells (5 × 10³) were plated in a 96-well microtiter plate and incubated overnight. Then the cells were washed twice with PBS and exposed to 250 pg/ml of IL-1α in serum-free KGM. Cell density was determined using an MTT cell proliferation assay (Boehringer Mannheim, Indianapolis, IN). MTT-labeling reagent was added at days 1, 3, and 5 after IL-1α treatment, and colorimetric ODs were measured at 570 nm by a microplate reader (Bio-Tek Systems).

Previously we showed that differences in IL-1α expression were associated with corresponding differences in IL-8 expression among HNSCC cell lines, and that the addition of recombinant IL-1α was able to stimulate IL-8 production by a low cytokine-producer cell line, UM-SCC-9 (5). To confirm whether IL-1α of HNSCC origin is functional and can induce IL-8 secretion by UM-SCC-9, we compared the abilities of recombinant IL-1α and an IL-1α-containing supernatant from the UM-SCC-11B cell line to induce IL-8 secretion (Fig. 1). Although UM-SCC-9 cells can produce 30–50 pg IL-1α/10⁶ cells over 5 days (19), the amount of IL-1α in control cultures of UM-SCC-9 cells by 24 h was below the 4-pg/ml threshold for detection by ELISA, as shown in Fig. 1,A. IL-1α was detected after the addition of 500 pg/ml of rIL-1α, rIL-1α-plus-control antibody, or rIL-1α-plus-IL-1RA, but IL-1α was completely neutralized when rIL-1α-plus-anti-IL-1 antibody was added. Fig. 1,B shows that, with low basal IL-1α expression, UM-SCC-9 produces a relatively low amount of IL-8 (106 pg/ml), when compared with the 1200 pg/ml of IL-8 detected after the addition of 500 pg/ml of recombinant IL-1α (Fig. 1,B). IL-1α-induced IL-8 production was completely inhibited by anti-IL-1 antibody or IL-1RA (Fig. 1,B). Anti-IL-1 antibody and IL-1RA did not effectively block the constitutive production of IL-8 by UM-SCC-9 cells. Next we tested a 48-h-conditioned medium from UM-SCC-11B cells, which contained ∼500 pg/ml of IL-1α by ELISA. In a pilot experiment, we tested IL-1α stimulatory activity for IL-8 induction in 100- or 50%-conditioned supernatant and found that 100%-conditioned supernatant showed stronger inducibility of IL-8 (data not shown). As shown in Fig. 1,C, 24 h after the addition of UM-SCC-11B supernatant to UM-SCC-9 cell cultures, we were able to detect ∼300 pg/ml IL-1α, which was completely neutralized when anti-IL-1 antibody was added. Fig. 1,D shows that UM-SCC-11B-conditioned medium (initially containing ∼500 pg/ml of IL-1α) induced a level of IL-8 production that is ten times higher (>12,000 pg/ml after subtraction of the IL-8 in UM-SCC-11B supernatant alone) than that induced (1,200 pg/ml) by an equivalent amount of rIL-1α. Fig. 1 D shows that neutralization by anti-IL-1 antibody and IL-1R antagonist, but not by the control antibody, results in complete inhibition of inducible IL-8 production, indicating that IL-1α in UM-SCC-11B cell-conditioned supernatant is functional and can induce IL-8 production by UM-SCC-9 cells. These results suggest that, once IL-1α induces IL-8 expression, another factor(s) in the conditioned supernatants of UM-SCC-11B may enhance further the expression of IL-8 by UM-SCC-9 cells.

We examined whether IL-8 production by UM-SCC-9 cells could be modulated by increases or decreases in endogenous IL-1α production. UM-SCC-9 cells were transiently transfected with plasmid DNA encoding the gene sequence for IL-1α and an icIL-1RA construct that is expressed intracellularly, and the cell culture supernatants were collected 48 h posttransfection. IL-1α-transfected UM-SCC-9 cells exhibited increased IL-1α (Fig. 2,A) and IL-8 (Fig. 2,B) production in supernatants by ELISA. In contrast, expression of an icIL-1RA construct that is expressed intracellularly resulted in significant inhibition of IL-1α (Fig. 2,A) and IL-8 (Fig. 2 B) expression. These results indicate that the expression of IL-1α by UM-SCC-9 cells can regulate the expression of IL-8 through an autocrine mechanism.

Previously, we showed that NFκB and AP-1 are activated in and contribute to IL-8 expression in HNSCC (18, 19). To define the contribution of NFκB and AP-1 transcription factor activation in IL-1α-induced IL-8 production in UM-SCC-9 cells, we collected extracts from UM-SCC-9 cells that were exposed to IL-1α and incubated these extracts with oligonucleotides encoding the binding sequence of NFκB, AP-1, or control OCT-1. Fig. 3 shows that IL-1α increased both NFκB (A) and AP-1 (B) binding to the oligonucleotides within 30 min, and induction was increased further at 1 h. We confirmed that NFκB and AP-1 activation could be induced by TNF-α as a positive control. Competition of binding by unlabeled oligonucleotides indicated that the DNA binding detected is specific, and the integrity of the extracts and loading was confirmed by constitutive OCT-1 binding.

The role of IL-1α in the induction of NFκB- and AP-1-dependent expression of IL-8 was examined using IL-8 promoter luciferase reporter gene assays. A set of IL-8 promoter constructs with mutations introduced in the NFκB and AP-1 binding sites were used to examine the relative contribution of these transcription factors to IL-1α-induced activation. Stimulation of UM-SCC-9 cells with IL-1α for 18 h resulted in increased reporter activity of the minimal IL-8 promoter containing both the NFκB and AP-1 transcription factor binding sites (−133; Fig. 3 C). The basal and IL-1α-inducible activation of IL-8 reporter activity in UM-SCC-9 was almost completely abolished with the mutation of the NFκB site (P < 0.05). Mutation in AP-1 binding sites partially decreased IL-1α-inducible IL-8 reporter activity. Although the CMV promoter in the β-Gal reporter used for determination of transfection efficiency has a putative NFκB binding site that could potentially be responsive to IL-1-induced activation, we observed a variation of <10% in CMV-β-Gal activity between constitutive and IL-1-stimulated samples in this and two independent experiments, which did not meet statistical significance.

To examine further the effects of endogenous IL-1α on NFκB activity, we tested NFκB reporter gene activity after the transfection of IL-1α and IL-1RA. A modest increase in NFκB reporter activity was observed at 72 h after transfection of IL-1α (Fig. 4), whereas transfection of IL-1RA resulted in a significant decrease in NFκB reporter activity (Fig. 4). Intracellular IL-1RA suppressed >80% of NFκB luciferase activity when compared with vector-transfected control (P < 0.05). These results provide evidence that endogenous IL-1α expression contributes to constitutive activation of NFκB in the UM-SCC-9 and -11B cell lines.

To examine whether endogenous IL-1α contributes to survival, UM-SCC-9 cells were cotransfected with a plasmid encoding CMV-β-Gal as an indicator for survival, along with plasmids encoding either IL-1α or intracellular IL-1RA. As shown in Fig. 5, transfecting cells with the plasmid with the IL-1α sequence resulted in increased β-Gal activity in cell cultures when compared with vector transfection. Conversely, transfecting cells with icIL-1RA significantly decreased β-Gal activity in the low IL-1α-producing UM-SCC-9 cells (Fig. 5,A) and had a more modest effect upon the high IL-1α-producing UM-SCC-11B cells (Fig. 5,B). Antagonism of IL-1α resulting from icIL-1RA demonstrated a greater effect on survival of the low IL-1-producing UM-SCC-9 cells when compared with the high IL-1-producing UM-SCC-11B cell line. Because the CMV promoter in the β-Gal reporter used for the determination of activity in Fig. 5 has a putative NFκB binding site that potentially also could be responsive to icIL-1 RA, it is possible that the effect seen results from the lower level of NFκB activity. To address this possibility, we conducted an additional independent experiment comparing the effect of the expression of icIL-1RA on CMV-Lac Z and RSV-LacZ activity, because the RSV promoter lacks a putative NFκB site. A reduction in RSV-LacZ as well as CMV-Lac Z reporter activity was observed, which is consistent with a decrease in the survival of cells cotransfected with icIL-1RA (data not shown).

Because IL-1α has been reported to stimulate the growth of human keratinocytes or transformed cervical carcinoma cells in vitro (10), we investigated whether exogenous addition of rIL-1α protein is able to stimulate cell growth of UM-SCC lines. Under serum-free conditions, a modest increase in cell growth of UM-SCC-9 in a 5-day MTT assay was detected with the addition of rIL-1α (250 pg/ml; Fig. 6,A). UM-SCC-11B cells, which already produce ∼500 pg/ml IL-1α, grew under serum-free conditions without the addition of exogenous IL-1α, and the addition of the same concentration of IL-1α to UM-SCC-11B cells resulted in little additional increase in cell growth (Fig. 6 B). We observed no additional significant increase in growth stimulation when either cell line was treated with 500 pg/ml of IL-1α nor a decrease exceeding 10–15% after the addition of up to 500 ng/ml of anti-IL-1 antibody or rIL-1RA (data not shown). Because the limited effect of recombinant anti-IL-1 and IL-1RA on the endogenous proliferative rate was observed in the same concentration range that inhibited the IL-8 response of the UM-SCC cell lines to exogenous IL-1α, the results suggest that exogenous IL-1RA is a relatively inefficient inhibitor of the autocrine growth-promoting effects of endogenously produced IL-1α in UM-SCC cells.

Previously we reported that human HNSCCs constitutively express IL-1α and a repertoire of proinflammatory and proangiogenic cytokines that are potentially IL-1-inducible, including IL-6, IL-8, and GM-CSF (5, 6). We found that coexpression of these cytokines correlated with activation of immediate-early transcription factors NFκB/Rel A and AP-1 and demonstrated that inactivation of NFκB/Rel A inhibited the expression of this repertoire of cytokines as well as cell survival and growth (18, 19, 20). Because IL-1 has been shown to modulate activation of these immediate-early pathways and cytokines in response to injury, we explored the effects of modulating exogenous and endogenously expressed IL-1α on the activation of NFκB and AP-1 expression of proangiogenic cytokine IL-8 and upon the survival and proliferation of HNSCC cell lines. Herein, we provide evidence that IL-1α contributes to the expression of IL-8 through the transcriptional activation of NFκB and AP-1 and also promotes cell survival and the growth of HNSCC cell lines in vitro.

In this study, we showed that HNSCC cell lines produce functional IL-1α in their culture supernatants by demonstrating their ability to induce IL-8 production and the ability to neutralize this response by anti-IL-1α antibody (Fig. 1). We also showed IL-1α produced by cells transfected with the IL-1α gene is functional in terms of IL-8 induction (Fig. 2 and Ref. 5). IL-1α was found to induce both NFκB and AP-1 DNA binding, which are potential activators of IL-8 in UM-SCC-9 cells (Fig. 3, A and B). We established that mutation of the NFκB site abrogated IL-8 reporter activity, whereas mutation of AP-1 partially inhibited IL-8 reporter activity (Fig. 3 C). These results are consistent with previous evidence showing that recombinant IL-1α can activate transcription factor NFκB and AP-1 and synergistically induce IL-8 promoter activity in other cell types (23, 25) and IL-8 production in HNSCC (5).

Transfection of cells with icIL-1RA significantly inhibited IL-8 production (Fig. 2) as well as NFκB activity (Fig. 4). These findings provide evidence that endogenous IL-1α is an important inducer of constitutive activation of NFκB and expression of IL-8 in HNSCC. Constitutive activation of NFκB has been detected in other tumors that express IL-1α. In breast cancer, constitutive activation of NFκB was found in three of three estrogen receptor-negative breast cancer cell lines that produced functional IL-1α constitutively (26, 27). The demonstration that exogenous and endogenous IL-1α can constitutively activate NFκB and IL-8 expression in HNSCC provides evidence that IL-1 is one of the important factors involved in the transcriptional activation of proangiogenic cytokine genes by HNSCC.

The observation that UM-SCC-11B supernatant containing IL-1α is significantly more potent than a similar amount of rIL-1α suggests that other factors produced by these tumors may also contribute to the activation of transcription and to the expression of IL-8. Whereas IL-1β has similar activity, and icIL-1RA could conceivably inhibit either IL-1α- and IL-1β-induced cell activation, we have not detected IL-1β in culture supernatants in any HNSCC cell lines tested (5, 6). Overexpression of other factors and receptors that can potentially activate NFκB and AP-1 pathways have been detected in HNSCC, including EGF receptor (28, 29), the hepatocyte growth factor receptor (c-Met; Ref. 30),⁶ and platelet-derived growth factor receptor (31). In cervical SCC, IL-1 has been shown to induce the expression of amphiregulin, a growth factor that binds the EGF receptor and promotes proliferation (10). Inhibition of the EGF receptor has been shown to have antiproliferative effects on HNSCC in vitro and in clinical trials, but it is not cytotoxic (29). It will be interesting to determine whether activation of the NFκB and AP-1 pathway by IL-1 contributes to proliferation through the expression of EGF family ligands or receptors. Alternatively, IL-1α, EGF, and other factors together may contribute to the activation of these pathways, indicating that inhibitors that target only one component of upstream activation may yield limited effects.

IL-8 induced in response to autocrine or exogenous sources of IL-1α is likely to be an important mediator of the malignant phenotype in human HNSCC. IL-8 and its CXC chemokine homologues have been shown to contribute to increased growth and metastasis of SCCs and other cancers through both autocrine and host-dependent mechanisms. IL-8 and its closely related homologues in the GRO family were shown originally to promote directly the proliferation of melanoma through an autocrine mechanism involving CXC receptor-2 (32). Recently we showed that the addition of rIL-8 can promote directly the proliferation of the UM-SCC-9 cells used in the present study (33). IL-8 and GRO chemokines have also been shown to promote chemotaxis of neutrophils and endothelial cells during angiogenesis (34). Treatment of mice bearing tumors with IL-8 and GRO cytokine-neutralizing antibodies have been reported to inhibit angiogenesis and the growth of lung, melanoma, and prostate xenografts in vivo (15). Recently we established that the murine homologue of human IL-8 and GRO-1 promotes growth, metastasis, angiogenesis, and inflammation in a homologous animal model of SCC through host responses involving CXC receptor-2 (35). Luca et al. (16) demonstrated that transfection of IL-8 cDNA into human melanoma cells can also increase in vivo tumor growth and metastasis in xenograft models. Thus, proinflammatory and proangiogenic cytokines regulated by IL-1 can have important effects in the pathogenesis of cancer.

The finding that expression of IL-1 promoted and IL-1RA markedly reduced the number of transfectants that coexpress the β-Gal reporter provides evidence that IL-1 expression may contribute to the survival of the HNSCC cells. The finding that expression of IL-1RA markedly reduced NFκB promoter activity and survival is consistent with our previous data indicating that NFκB activation is important for survival (19). Inhibition of NFκB activation by a dominant negative mutant inhibitor-κB decreased cell survival (19) and the resistance of SCC to cell death induced by TNF-α or radiation (36, 37). The cytoprotective effects of IL-1 against radiation have been reported previously and are consistent with the activation of antiapoptotic pathways (38). NFκB can inhibit cell death induced by TNF or cytotoxic therapies and can induce a variety of cytoprotective genes that mediate antiapoptotic signals (38, 39, 40, 41). Our results in human HNSCC are corroborated by a recent report by Vale et al. (42), who found that inhibition of IL-1 by IL-1RA resulted in decreased survival of foci of Raf- and NFκB-activated NIH 3T3 cells. Furthermore, we have obtained evidence that murine SCC cells that overexpress IL-1 and demonstrate increased activation of NFκB show increased expression of the prosurvival candidate inhibitor of apoptosis-1 (43). We have yet to determine directly whether the effects of IL-1 on survival of HNSCC cell lines are mediated by the antiapoptosis effects of NFκB and which NFκB-dependent genes may be involved. Development of an inducible system that will enable inducible overexpression of icIL-1RA and dominant-negative inhibitor-κB is under way to examine further the role of IL-1 and NFκB and to identify prosurvival genes regulated by these factors.

Expression of IL-RA in UM-SCC-9 was found to have potent inhibitory effects on the constitutive activation of NFκB, IL-8 expression, and cell survival, providing evidence that endogenous IL-1α expression is an important factor in the activation of prosurvival transcription factors and genes. The inhibitory effect observed with the expression of a construct of IL-1RA lacking leader sequence containing the signal for secretion was greater than that observed after the addition of high concentrations of exogenous IL-1RA (Figs. 2,A, 5, and 7 and Ref. 5). These findings suggest that IL-1 activity may not be limited to a receptor binding loop that involves extracellular secretion and suggests that IL-1α may mediate its autocrine effects in HNSCC via membrane or intracellular IL-1 receptor binding or actions that are relatively inaccessible to exogenous IL-1RA. IL-1α is produced intracellularly as IL-1α precursor, which is fully active (2). In normal cells, IL-1α precursor is released only during cell death and is cleaved by extracellular proteases to become mature IL-1α (2). In transformed cells, IL-1α can be released from cells without cell death (2), and we have shown that most HNSCCs secrete IL-1α spontaneously (5, 6). However, evidence has accumulated that is consistent with the hypothesis that IL-1α may have intracellular functions. Soluble IL-1 has been shown to bind to surface IL-1R and be internalized (44). IL-1α also has been detected as an intranuclear protein, where it may either directly bind to DNA or form complexes with intracellular IL-1R, which then binds to DNA (45). We have shown that intracellular inhibition of IL-1α activity has potent inhibitory effects on the constitutive activation of NFκB, IL-8 expression, and cell survival. Determination of the mechanisms by which IL-1α activates the transcription and expression of prosurvival genes in HNSCC and other cancers that express IL-1 may provide new approaches for anticancer therapy.

We thank Drs. Keith Brown and Ulrich Siebenlist of the National Institute of Allergy and Infectious Diseases (NIAID), NIH, Bethesda, MD for the generous gift of the plasmid IgkB-Luc and Steve Haskill, University of North Carolina, Chapel Hill, NC, for pcDNA icIL-1RA. We also thank Drs. James Battey (NIH) and James Mulshine (NIH) for their helpful suggestions.