Figure 1. Increase of nuclear factor–κB (NF-κB) luciferase reporter activity by paclitaxel in the oral cancer cell line CA-9-22. Cells were transfected with the NF-κB luciferase reporter construct. After 24 hours of incubation with varying paclitaxel concentrations, cells were assayed for luciferase activity. The results are expressed as relative activity of NF-κB (SE), normalized to B-galactosidase activity. Similar results were obtained in 3 independent experiments. DMSO indicates dimethylsulfoxide; RLU, relative luciferase units.
Figure 2. Electrophoretic mobility shift assay (EMSA) of nuclear factor–κB (NF-κB) and Oct-1 binding activity in CA-9-22 cells. A, Augmented NF-κB binding activity in CA-9-22 cells incubated with 10nM paclitaxel is shown in lane 4 when compared with control (lane 1). Competition with unlabeled wild-type probe (lanes 2 and 5). The major nuclear proteins in CA-9-22 extracts were super-shifted (SS) with anti-p65 antibody (lanes 3 and 6). Decreased NF-κB binding activity in CA-9-22 cells incubated with 40-mg/mL indomethacin (Indo) and paclitaxel + Indo is shown in lanes 8 and 11. Competition with unlabeled wild-type probe is shown in lanes 8 and 11. The nuclear extracts were SS with anti-p65 antibody (lanes 9 and 12). Similar results were obtained in 3 independent experiments. B, Quantification of NF-κB binding activity by densitometry analysis. C, EMSA of Oct-1 binding of CA-9-22 cell extracts. Nuclear extracts from the experiment depicted in panel A were incubated with labeled Oct-1 probe (lines 1, 3, 5, and 7). Competition with a 100-fold excess of unlabeled Oct-1 probe is shown in lanes 2, 4, 6, and 8. Similar results were obtained in 3 independent experiments.
Figure 3. Indomethacin augments chemosensitivity in paclitaxel-treated cells. CA-9-22 cells were treated with 40-mg/mL indomethacin or DMSO for 72 hours. MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] cell proliferation assays were carried out as described in the “Methods” section. DMSO indicates dimethylsulfoxide.
Figure 4. Indomethacin-enhanced apoptosis in CA-9-22 cells (Apoptosis index). CA-9-22 cells were treated with 10nM paclitaxel, 40-mg/mL indomethacin, preincubation with 40-mg/mL indomethacin and 10nM paclitaxel, or dimethylsulfoxide (DMSO) for 24 hours. Apoptosis was quantified by flow cytometry. The percentage of apoptotic cells for each group was obtained and compared with control. Cells treated with staurosporine (sts) were used as a positive control for apoptosis.
Figure 5. Decrease of nuclear factor–κB (NF-κB) luciferase reporter activity by stable transfection of inhibitor kappa B-alpha (IκBα) mutant in CA-9-22 cells. The NF-κB luciferase reporter gene assay was performed on CA-9-22 cells stably transfected with dominant negative IκBα or with the empty vector. These experiments were repeated in triplicate with similar results.
Figure 6. Expression of mutant inhibitor kappa B-alpha (IκBα) increases cell sensitivity to paclitaxel. CA-9-22–dominant negative IκBα and empty vector cells treated with varying concentrations of paclitaxel (1.0nM, 2.5nM, 5.0nM, and 10.0nM) for 72 hours. Cell proliferation was evaluated by MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay. Data are expressed as mean (SE) of 3 independent experiments. DMSO indicates dimethylsulfoxide.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Caicedo-Granados EE, Wuertz BR, Marker PH, Lee GS, Ondrey FG. The Effect of Indomethacin on Paclitaxel Sensitivity and Apoptosis in Oral Squamous Carcinoma Cells: The Role of Nuclear Factor–κB Inhibition. Arch Otolaryngol Head Neck Surg. 2011;137(8):799–805. doi:10.1001/archoto.2011.131
Author Affiliations: Molecular Oncology Program, Department of Otolaryngology (Drs Caicedo-Granados and Ondrey and Ms Wuertz), and Department of Medicine, Hematology, Oncology, and Transplantation (Mr Marker), University of Minnesota, Minneapolis; and Department of Otolaryngology, Children's Hospital Boston, Harvard Medical School, Boston, Massachusetts (Dr Lee).
Objective To investigate new strategies to intensify chemosensitivity in head and neck squamous cell carcinoma.
Design Oral squamous carcinoma cells were examined for nuclear factor–κB (NF-κB) activation and binding activity by paclitaxel, an agent currently used in head and neck cancer chemotherapy. Electromobility shift assays were used to assess the effect of indomethacin on NF-κB binding activity. Cell proliferation assays were used to study cell sensitivity to paclitaxel. To examine whether cytotoxicity could be increased by specifically inhibiting NF-κB, a dominant negative cell line, inhibitor kappa B-alpha (IκBα), was stably expressed in CA-9-22 cells.
Results Paclitaxel possessed the capacity to functionally activate NF-κB, as demonstrated by luciferase reporter gene assays and electromobility shift assay. Indomethacin was able to inhibit paclitaxel-mediated NF-κB activation and promote apoptosis of paclitaxel-treated cells at 24 hours. Indomethacin augmented the paclitaxel cell-killing effect. The dominant negative IκBα cell line exhibited increased chemosensitization to paclitaxel by 2- to 10-fold.
Conclusions Paclitaxel has the capacity to activate NF-κB in oral squamous carcinoma cells. Indomethacin can reverse this activation to decrease cell proliferation and increase apoptosis. Treatment strategies that combine paclitaxel with indomethacin may have therapeutic benefits attributable to paclitaxel chemosensitization through NF-κB inhibition.
Head and neck squamous cell carcinoma (SCC) accounts for approximately 5% of all malignant neoplasms in the United States.1 Locoregional control of disease remains the most significant barrier in improving long-term survival after recurrence. Survival for advanced and recurrent disease has not improved in approximately 3 decades despite extraordinary advances in multimodality therapy, imaging, and surgical techniques.2,3 Patient survival after head and neck SCC recurrence averages 10% at 1 year.4 At present, many accepted cytotoxic chemotherapeutic strategies for this malignancy are not mechanistically based on the molecular biological characteristics of these tumors.
Therapies outside the accepted chemotherapeutic strategies based on molecular mechanisms of head and neck cancer produce clinical results of varying effectiveness. Head and neck SCC is a malignancy characterized by multiple immune defects in both humoral and cell-mediated immunity.5-8 It is unknown whether these defects are primarily a result of tumor burden or existing comorbidities in patients with head and neck cancer resulting from malnutrition secondary to dysphagia, smoking, or excessive alcohol consumption. However, it has been known for over 2 decades that some of these immune defects are reversible pharmacologically with nonsteroidal anti-inflammatory drugs (NSAIDs) and some patients with advanced recurrent tumors respond to NSAID therapies.9 Nonsteroidal anti-inflammatory drugs have been shown to decrease cell proliferation in different cell lines through NF-κB down-regulation.10 Sulindac, an NSAID that is structurally related to indomethacin, inhibits cyclooxygenase activity to prevent prostaglandin synthesis. In addition, sulindac and its metabolites (sulindac sulfide and sulindac sulfone) inhibit NF-κB pathway activation.11 Multiple prostaglandin synthesis inhibitors have also been studied in patients with aerodigestive cancer. They reverse some of the immune defects encountered in these malignant neoplasms; however, molecular mechanisms for these effects are poorly understood. In addition, they have not improved survival in patients when delivered as a single-agent therapy.12,13
Squamous aerodigestive cancer is clinically resistant to the effects of most chemotherapy agents, but platinum compounds and taxanes have been demonstrated to result in significant durable responses in advanced head and neck cancer when combined with radiation.14-16 Dose-limiting toxic effects are encountered under these treatment regimens, when used as primary treatments, but neither agent, alone or in combination, results in acceptable durable response after cancer recurrence.17 Chemotherapy resistance to both platinum and taxanes remains a significant barrier to their efficacy in advanced and recurrent head and neck cancer.18 Therefore, strategies that may alter chemotherapy resistance would be very attractive in attenuating this disease process.
Nuclear factor–κB is an early response gene associated with SCC progression.19 Its overexpression has been described in multiple solid organ tumors, and NF-κB has been described as an important factor in tumor cell apoptosis resistance in several tumor types including fibrosarcomas and breast cancer.20,21 Consequently, strategies to reduce NF-κB through multiple mechanisms, such as gene therapy and proteosome inhibition, have been suggested as potential therapy for associated apoptosis resistance.22-24 Nuclear factor–κB activation is associated with many pathophysiologic conditions associated with aerodigestive SCC progression including the production of proinflammatory and proangiogenic cytokines responsible for tumor growth and metastases. Nuclear factor–κB is also induced during periods of cellular stress secondary to bacterial pathogens and radiation.25-27 As an injury-response gene, its induction during periods of cellular stress likely represents a survival tactic, which triggers apoptosis resistance.
In the present study, we hypothesized that paclitaxel may induce transient or stable induction of functional NF-κB as a stress response. We also hypothesized indomethacin treatment of head and neck cancer cells may prevent paclitaxel-mediated NF-κB induction, increase cell sensitivity to taxol, and enhance apoptosis.
The oral SCC cell line used for this study, CA-9-22, was a gift from Toshio Kuroki, MD.28,29 Cells were maintained in minimal essential medium supplemented with 10% fetal bovine serum, L-glutamine (5.8 mg/mL) and penicillin-streptomycin (50 μg/mL) at 37°C in 5% carbon dioxide as adherent monolayer cultures. Paclitaxel obtained from ICN Pharmaceuticals Inc, Costa Mesa, California, was reconstituted in dimethylsulfoxide (DMSO) at a stock concentration of 20mM and stored at –20°C. Indomethacin [1-(p-chlorobenzoyl)-5-methoxy-2-methylindole-3-acetic acid] was obtained from Sigma-Aldrich Co, St Louis, Missouri, and reconstituted in DMSO at a stock concentration of 20 mg/mL. The vehicle, DMSO, was added to all experimental controls at equivalent concentrations to treatment groups.
The pIgκB-Luc reporter construct contains 3 immunoglobulin G-κ chain NF-κB binding sites and the luciferase gene.30 The pCMV-Lac-Z reporter construct was cotransfected as an internal standard as previously described.31 Subclonfluent CA-9-22 cells were plated in 12-well plates and transfected with luciferase pIgκB-Luc reporter vector (2 μg per well) and pCMV-Lac-Z reporter vector (0.4 μg per well) in Opti-MEM medium containing 10 mg/mL of LipofectAMINE (Invitrogen Corporation, Carlsbad, California). After 4 hours, transfection medium was removed and cells were placed in complete media overnight. Cells were treated in serum-free media for 24 hours with paclitaxel or vehicle alone. The relative luciferase activity was determined with the Dual Light reporter gene assay (Applied Biosystems, Carlsbad) according to the manufacturer's instructions using a Tropix TR717 dual injector plate luminometer (Berthold Technologies, Oak Ridge, Tennessee).
Extracts from cells in log-phase growth were prepared according to the methods of Dignam et al32 and Lee et al33 with the following modifications: Cells were grown in 150 cm2 flasks to 60% to 80% confluence. A cell suspension was prepared by Trypsin-EDTA treatment, and the cells were washed twice in cold phosphate-buffered saline (PBS) and pelleted at 500 g. The cell pellets were incubated on ice for 5 minutes with lysis buffer (10mM TRIS hydrochloride, 60 mM potassium chloride, 1mM EDTA, 0.5% NP-40, and fresh 0.1M dithiothreitol) and protease inhibitor. The lysates were then microfuged for 5 minutes at 500 g. The supernatant was collected and saved as cytoplasmic extract. Next, the cell pellet was resuspended in a volume of nuclear extract buffer equal to the volume of the cell pellet (20mM TRIS hydrochloride, 420mM sodium chloride, 0.2mM EDTA, 25% glycerol, fresh 0.1M dithiothreitol, and phenylmethylsulfonyl fluoride), incubated on ice for 10 minutes, and centrifuged at high speed for 10 minutes. The nuclear extracts were aliquoted and stored at −80°C. Protein concentrations were determined in duplicate by using the bicinchoninic acid modification of the biuret reaction (Pierce Protein Assay Kit; Pierce Chemical Company, Rockford, Illinois). Bovine serum albumin was used as a standard, and absorbance was read at 580 nm. Correlation coefficients for the functions were greater than 0.95 in all experiments.
Double-stranded DNA oligonucleotide probes for NF-κB and Oct-1 were synthesized commercially (Promega Corp, Madison, Wisconsin). The consensus sequences used were 5′-AGTTGAGGGGACTTTCCCAGGC-3′ for NF-κB and 5′-TGTCGAATGCAAATCACTAGAA-3′ for Oct-1. Consensus oligonucleotide probes were labeled with T4 polynucleotide kinase (Promega Corp) and [γ-32P]-ATP 6000 Ci/mmoL (GE Healthcare, Piscataway, New Jersey).
Binding reactions were performed using 5 μg of nuclear protein incubated for 25 minutes at 20°C in buffer containing 20mM Hepes (pH 7.9), 4.6mM magnesium chloride, 63mM potassium chloride, 11% glycerol, 1mM dithiothreitol, 1 μg of poly-dIdC (GE Healthcare), and 60 000 cpm 32P-labeled nucleotide probe. For unlabeled probe competitions, extracts were preincubated with a 100-fold excess of each unlabeled probe before addition of labeled probe. The binding complexes were resolved on 5% native polyacrylamide gels in 0.25× TBE at 4°C and run for 90 minutes at 200 V. The gels were dried and imaged with a Cyclone Phosphor Imager (PerkinElmer Inc, San Jose, California). OptiQuant software (PerkinElmer Inc) was used to quantify the radioactive signals. All reactions were performed with at least 3 separate nuclear extract preparations.
Antibody supershift analysis was performed using 1 μg of a p-65 NF-κB subunit antibody (Rockland Laboratories, Gilbertsville, Pennsylvania) 30 minutes prior to the binding reaction.
Cells were plated at a density of 7.5 × 103 per well in 96 well plates. The next day, serum-free media containing 40 mg/mL of indomethacin was added to each well. After 24 hours, indomethacin was removed and serum-free media containing 1nM to 20nM paclitaxel was placed into each well and incubated for an additional 72 hours. Control groups were treated with serum-free media containing similar DMSO concentrations. Cell proliferation was determined by MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay (Roche, Indianapolis, Indiana) per manufacturer instructions and read at 560 nm on a plate spectrophotometer.
ApopTag direct In Situ apoptosis detection kit (Millipore, Billerica, Massachusetts) was used to assess apoptosis following drug treatments. Briefly, CA-9-22 cells were grown and treated as described in the previous subsection. (1μM) Staurosporine treatment for 24 hours was used as a positive control. From each treated group, 3 × 106 cells were fixed in 1% paraformaldehyde in PBS (pH 7.4) on ice. Cells were washed in ice-cold PBS, spun down and resuspended in 70% ice-cold ethanol, and kept at –20°C. The ApopTag direct In situ apoptosis detection kit was used to label the 3′-OH DNA termini in situ with fluorescein-labeled and unlabeled nucleotides following the manufacturer's protocol. Appropriate control samples (cells stained with ApopTag fluorescein only, cells stained with propidium iodide only, and unstained cells) were used to set up the electronic compensation and quadrant statistics. Next, cells were incubated for 15 minutes at 37°C in propidium iodide solution (1 mg/mL) containing ribonuclease (1 μg/mL). Cells were analyzed on a Becton Dickinson FACSCalibur (Becton Dickinson, San Jose, California) flow cytometer, with a gated mode for forward light scatter pulse, side scatter pulse height for analysis of cell cycle fractions, and an ungated mode for detection of cells with subG1 DNA content. Resulting histograms were evaluated with Cell Quest pro software (Becton Dickinson). Comparison between control, indomethacin, paclitaxel, and indomethacin + paclitaxel was based on 5000 events counted.
ApopTag direct In Situ apoptosis detection kit was also used to detect apoptosis using immunohistochemical analysis. Slides were blindly analyzed and cells counted by a senior investigator (F.G.O.). Four hundred cells per specimen were counted and an apoptosis index was established. The slides were scored using an inverted microscope (Nikon Eclipse TE 200; Nikon Corporation, Tokyo, Japan). The number of positive cells per ×20 field of view was scored. Statistics were determined by using 1-way analysis of variance (Graph Pad Prism software v3.0; GraphPad Software Inc, La Jolla, California)
The complementary DNA plasmid, pCMX IκBαM, contains a mutation at S36 of the NH2-terminus and a C-terminus PEST (proline [P], glutamic acid [E], serine [S], and threonine [T]) sequence mutation, thereby inhibiting phosphorylation and translocation of NF-κB. This plasmid and the pCMX empty vector were a gift from Inder M. Verma, PhD (Salk Institute, La Jolla, California).34 The pCMV plasmid containing the neomycin resistance gene is described by Brown et al35 Subconfluent CA-9-22 cells were cotransfected with either the pCMX-IκBαM vector (2-μg per well) and pCMV–neomycin resistance vector (2-μg per well) or the empty vector (2-μg per well) and pCMV–neomycin resistance vector (2-μg per well). LipofectAMINE (10 mg/mL) (Life Technologies Inc) in Opti-MEM medium was used for 4 hours. After 72 hours of recovery in complete media, selection media containing 750-μg/mL G-418 sulfate (Life Technologies Inc) yielded neomycin-resistant clones that were established as subcultures. Cells were screened for NF-κB activity by luciferase reporter assay.
Two-tailed t tests were used to compare the means in all in vitro experiments, and P < .05 was considered statistically significant. GraphPad software was used to perform the analysis.
To determine if paclitaxel has an effect on NF-κB promoter activity, we transiently transfected CA-9-22 cells with a prototypic NF-κB consensus sequence in a luciferase reporter plasmid. Cells were treated at different paclitaxel concentrations for 24 hours, and luciferase reporter assays were performed. CA-9-22 cells underwent a significant dose-dependent increase in NF-κB functional activity after 5nM, 10nM, and 20nM paclitaxel treatment compared with the untreated control (Figure 1). We observed a maximum increase in NF-κB reporter gene activation at 20 nM (P < .001). At concentrations greater than 20nM, paclitaxel was too cytotoxic to determine levels of NF-κB activation (data not shown). Each experiment was conducted 3 times with similar results.
To examine whether functional activation of NF-κB by paclitaxel occurs through p50/p65 NF-κB heterodimer interactions, we performed electrophoretic mobility shift assays on CA-9-22 cells. DNA-binding activity of NF-κB increased when CA-9-22 cells were incubated for 1 hour with 10nM paclitaxel (Figure 2A, lane 4) compared with control (Figure 2A, lane 1). Densitometric analysis confirmed a 2-fold increase in paclitaxel-mediated DNA binding (Figure 2B). Experiments were repeated 3 times with similar results.
Preincubation with indomethacin followed by a 1-hour paclitaxel treatment of CA-9-22 cells suppressed this effect (Figure 2A, lane 10 vs lane 4). Densitometric analysis confirmed that indomethacin + paclitaxel decreased DNA binding activity compared with paclitaxel alone (Figure 2B). Supershift analysis for p65-containing heterodimers is demonstrated in lanes 3, 6, 9, and 12. This confirmed that functional activation of NF-κB by paclitaxel is contributed to by p50-p65 heterodimers. Nuclear factor–κB binding activity did not increase when cells were treated with indomethacin alone (Figure 2, lane 7). Competition experiments with unlabeled wild-type NF-κB oligonucleotides confirmed that the binding activity detected in Figure 2 lanes 2, 5, 8, and 11 was specific for NF-κB. Equivalence of loading and nuclear extract integrity was confirmed by demonstration of similar Oct-1 binding (Figure 2C).
To investigate whether indomethacin potentiates paclitaxel-mediated decreases in cell proliferation, we performed MTT assays on CA-9-22 cells pretreated with 40-mg/mL indomethacin for 24 hours followed by paclitaxel treatment. Preincubation with indomethacin inhibited cell proliferation at all concentrations of paclitaxel. The effect was most profound at 1nM to 10nM paclitaxel. Indomethacin alone decreased proliferation by approximately 15%. Cell proliferation was inhibited by 59% in the 1nM paclitaxel group compared with control (Figure 3). Cell growth inhibition by preincubation with indomethacin was statistically significant at all paclitaxel concentrations (P ≤ .001).
Tumor cell apoptosis resistance is a common mechanism for cancer therapy failure and is incompletely understood. Chemotherapy agents, radiation, and other stimuli have the capacity to activate NF-κB, and this activation strongly suppresses the apoptotic potential of these stimuli in vitro.36,37
To determine whether the paclitaxel chemosensitivity conferred by preincubation with indomethacin occurred through apoptotic mechanisms, we analyzed CA-9-22 cells via TUNEL (terminal deoxynucleotidyltransferase-mediated dUTP nick-end labeling) assay and flow cytometry. Indomethacin and paclitaxel were associated with a 41% and 71% increase, respectively, in apoptosis over control cells. The combination of the agents was associated with a 293% increase over controls, while the staurosporine positive control showed a 261% increase over controls (Figure 4). This was similar to a second experiment in which indomethacin and paclitaxel augmented apoptosis by 57% and 108%, respectively, over controls, whereas both agents caused a 760% increase in apoptosis, as confirmed by immunohistochemical analysis (data not shown).
To further investigate effects of paclitaxel-mediated NF-κB activation, we stably transfected an IκBα dominant negative plasmid into CA-9-22 cells.38 Nuclear factor–κB reporter gene activity was reduced up to 5-fold in CA-9-22 cells, which expressed the dominant negative IκBα compared with the empty vector control (Figure 5) (P < .001).
We also examined the effect of paclitaxel on cell proliferation in dominant negative IκBα CA-9-22 cells. MTT assays were performed using CA-9-22 wild-type, CA-9-22 dominant negative IκBα cells and empty vector control. CA-9-22 dominant negative IκBα cells treated with paclitaxel showed a marked decrease in cell proliferation compared with control (Figure 6). We found that inhibition of cell proliferation of CA-9-22 dominant negative IκBα cells was statistically significant at 5nM paclitaxel compared with control, and the largest difference between the cell lines was noted with 10nM paclitaxel.
In this study, we provide evidence that paclitaxel is capable of functionally activating NF-κB in a dose-dependent manner in upper aerodigestive carcinoma cells. Nuclear factor–κB luciferase reporter and electrophoretic mobility shift assay experiments in CA-9-22 cells incubated with paclitaxel showed an increase in functional activation and an increase in NF-κB binding activity. The functional activation of NF-κB was maximal when cells were incubated with 20nM paclitaxel, a steady-state serum concentration achieved in a recent head and neck cancer clinical trial.39 However, other recent publications point out that peak plasma paclitaxel levels can range from 228nM at 135 mg/M2 over 24 hours, (ovarian cancer) to 349nM at 100 mg/M2 over 1 hour (metastatic esophageal cancer).40,41 We have found that concentrations of paclitaxel greater than 20nM are 100% cytotoxic in our cell systems. Furthermore, it has recently been shown that paclitaxel dosing regimens are based on empirical, rather than physiologic, effects.42 Therefore, our finding of maximal effect at 20nM has interesting implications. First, it may be that in vivo, paclitaxel and indomethacin combinations can be optimized for an indomethacin effect while avoiding the adverse effects of paclitaxel, particularly in patients with recalcitrant disease whose performance status may contraindicate high-dose chemotherapy. Second, the optimal dosing of indomethacin with paclitaxel is still unknown and will require careful monitoring of pharmacokinetic parameters.
We also determined CA-9-22 cells preincubated with indomethacin prevented NF-κB induction when treated with paclitaxel, and this correlated with increased tumor cell chemosensitivity to paclitaxel by 2- to 10-fold. This was demonstrated in several experiments by decreased paclitaxel dosing to achieve a given level of decreased cell proliferation in cells preincubated with indomethacin compared with controls. Interestingly, this decrease was observed at paclitaxel concentrations routinely achieved in the serum of patients. It is also important to note that even the lowest concentration of paclitaxel, 1nM, achieved a 59% decrease in proliferation after indomethacin treatment. Furthermore, electromobility shift assays provide evidence of the elevated binding activity of NF-κB after paclitaxel incubation and the abrogation of NF-κB binding activity when indomethacin preincubation was used.
We also obtained evidence that CA-9-22 cell apoptosis was intensified when paclitaxel incubation was preceded by 24 hours of indomethacin incubation, which coincides with a decrease in NF-κB binding via electrophoretic mobility shift assay. In contrast, incubation with indomethacin alone caused only slight increases in the apoptosis index. In support of the role and specificity of NF-κB in the augmentation of cell chemoresistance to paclitaxel, we demonstrated expression of a dominant negative mutant inhibitor-κB in the human oral squamous cells was associated with decreased rates of proliferation after paclitaxel treatment compared with NF-κB–containing control cells.
The up-regulation of NF-κB is commonly described as a potential apoptosis resistance mechanism in a variety of cancers. In fact, the constitutive up-regulation of NF-κB is described as a vital contributing feature of both Hodgkin disease as well as certain breast cancers.43 This is secondary to endogenous defects in IκBα signaling. In preclinical studies, SCC of the head and neck has been associated with inducible and constitutive up-regulation of NF-κB and downstream genes associated with proliferation, invasion, and angiogenesis.44,45 In addition, SCC of the head and neck is a malignancy with high levels of chemoresistance, as demonstrated by the universal failure of all chemotherapy for cancer recurrence. Therefore, strategies to treat this malignancy based on altering apoptosis resistance are very attractive. Phase 1 studies of proteosome inhibitor PS-341 are based on preliminary studies demonstrating its efficacy in down-regulating NF-κB signaling in head and neck cancer.46
In the present study, we hypothesized that paclitaxel may transiently induce NF-κB and that this increase of activation has the potential to induce apoptosis resistance during paclitaxel treatment. It is possible that failed chemotherapy for recurrent head and neck cancer may involve a number of mechanisms including alteration in glutathione S-transferase enzyme activation, induction of multidrug resistance gene, or NF-κB induction by chemotherapeutic agents.47,48 The ability of a simple Food and Drug Administration–approved agent such as indomethacin for chemosensitization of paclitaxel in head and neck cancer would be a very attractive therapeutic option for salvage therapy for upper aerodigestive cancers. In addition, this strategy may be useful in other malignant conditions in which NF-κB overexpression contributes to the malignant phenotype.
Correspondence: Frank G. Ondrey, MD, PhD, Molecular Oncology Program, Department of Otolaryngology, University of Minnesota, 420 Delaware St SE, MMC 396, Minneapolis, MN 55455 (email@example.com).
Submitted for Publication: January 8, 2011; final revision received May 16, 2011; accepted May 20, 2011.
Author Contributions: Dr Ondrey had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Caicedo-Granados and Ondrey. Acquisition of data: Caicedo-Granados, Wuertz, Marker, Lee, and Ondrey. Analysis and interpretation of data: Caicedo-Granados, Wuertz, Marker, and Ondrey. Drafting of the manuscript: Caicedo-Granados and Ondrey. Critical revision of the manuscript for important intellectual content: Caicedo-Granados, Wuertz, Marker, Lee, and Ondrey. Statistical analysis: Ondrey. Obtained funding: Ondrey. Administrative, technical, and material support: Caicedo-Granados, Wuertz, Marker, and Ondrey. Study supervision: Caicedo-Granados and Ondrey.
Financial Disclosure: None reported.
Funding/Support: This work was supported by the American Cancer Society Institutional Research Cancer Grant IRG-58-001-40-IRG-45; the NIH National Research Service Award Institutional Research Training Grant T32-DC00059; and the Lion's Multiple District 5M Hearing Foundation of Minnesota.