|
|
Mol Cancer Ther. Author manuscript; available in PMC 2009 September 14. Published in final edited form as: doi: 10.1158/1535-7163.MCT-07-2046. | PMCID: PMC2743084 NIHMSID: NIHMS58928 |
Differential bortezomib sensitivity in head and neck cancer lines corresponds to proteasome, NF-κB and AP-1 related mechanisms *Zhong Chen,1 Justin L. Ricker,1 Pramit S. Malhotra,2 Liesl Nottingham,1 Lorena Bagain,1 Tin Lap Lee,1 Ning T Yeh,1 and Carter Van Waes1 1 Tumor Biology Section, Head and Neck Surgery Branch, National Institute on Deafness and Other Communication Disorders, National Institutes of Health, Bethesda, MD 2 Dept. of Otolaryngology – Head & Neck Surgery, University of Minnesota Hospitals, 420 Delaware St, S.E., Mayo Building, Box 396, Minneapolis, MN Head and neck squamous cell carcinomas (HNSCC) exhibit constitutive activation of transcription factors NF-κB and AP-1, which are modulated by the proteasome, and promote resistance to cell death. HNSCC show variable sensitivity to the proteasome inhibitor bortezomib in vitro, as well as in murine xenografts and patient tumors in vivo, and the mechanisms are not well understood. To address this question, the sensitivities of nine HNSCC cell lines to bortezomib were determined using MTT assays, and the potential relationship between the sensitivity and bortezomib effects on biological processes were examined in HNSCC lines of differential bortezomib sensitivity. The most sensitive cell line (UM-SCC-11B) underwent cell death at 10–9M in vitro and tumor regression at a maximally tolerated dose of bortezomib in a murine xenograft model. The differential sensitivity between UM-SCC-11A and -11B cells corresponded to differences in the extent of suppression of proteasome activity, ubiquitinated protein degradation, and NF-κB and AP-1 activation. Lower concentrations of bortezomib transiently increased NF-κB and sustained AP-1 activation in UM-SCC-11A cells. AP-1 reporter activity and cell density of UM-SCC-11A were suppressed when bortezomib was combined with JNK and p38 kinase pathways inhibitors. Thus, the differential sensitivities to bortezomib corresponded to dissimilar effects on the proteasome, NF-κB and AP-1 activities. Inhibition of JNK and p38 pathways blocked AP-1 activity and enhanced the anti-tumor effects. These findings revealed molecular mechanisms of bortezomib sensitivity and resistance which are under development as biomarkers for clinical trials in patients with HNSCC. Keywords: Bortezomib (PS-341/VELCADE™), proteasome, NF-κB, AP-1, head and neck cancer Constitutive activation of the NF-kappaB (NF-κB) and Activator Protein-1 (AP-1) signal transduction pathways have been identified as prominent events promoting tumor progression of hemopoietic and solid malignancies ( 1– 4), including head and neck squamous cell carcinomas (HNSCC, 5–7). We have shown that NF-κB is frequently co-activated with AP-1, and promotes the proliferative, prosurvival and proangiogenic cancer phenotype ( 6, 7). Targeting NF-κB and AP-1, or up-stream signal transduction pathways by genetic or chemical inhibitors, has been shown to effectively suppress the tumor phenotype in vitro, as well as inhibit tumor growth in preclinical animal models in vivo ( 4– 5, 7– 11). Subsequent clinical studies have correlated NF-κB and AP-1 pathways, as well as their targeted biomarkers, with worse prognosis ( 12– 15). Thus, aberrant activation of NF-κB and AP-1 are critical signal transduction pathways promoting the aggressive tumor phenotype and survival of HNSCC. Bortezomib (VELCADE™/PS-341) has been developed in recent years for molecular targeting and inhibition of the proteasome, a complex which mediates the turnover of many intracellular proteins, including those controlling cell signaling, survival, and cell cycle regulation ( 16, 17). Bortezomib selectively inhibits proteasome activity, which is required for activation of NF-κB and degradation of components of AP-1 and other signal pathways involved in the pathogenesis of cancer ( 16– 18). Bortezomib can inhibit the NF-κB pathway through its inhibitory effects on degradation of ubiquitinated Inhibitor-κB (IκB), which binds and sequesters NF-κB in the cytoplasm, inhibiting its nuclear localization and binding to the promoters of target genes ( 11, 16, 17, 19– 21). The protein components of AP-1 family members are also degraded through the proteasome system ( 18, 21). The inhibitory activity of bortezomib has been demonstrated against a spectrum of cancer cells in culture ( 19– 29) and in animal models ( 11, 30– 32), including suppression of NF-κB and other signal transcription pathways ( 11, 16, 17, 19– 32), with induction of cell apoptosis and cell cycle arrest ( 19, 20, 22– 35). The molecular and clinical effects of bortezomib and potential mechanisms of variable activity have been most extensively studied in multiple myeloma (MM) and certain other hematopoietic malignancies ( 20, 22– 25, 33– 35), but to a lesser extent in solid cancers ( 8, 11, 19, 26– 32). In clinical trials ofcarcinomas and solid tumors, lower response rates and greater heterogeneity in responsiveness to bortezomib monotherapy was observed compared to MM ( 36– 38), and combination of bortezomib with other anti-cancer agents has been undertaken in an effort to achieve significant anticancer effects in vivo ( 8, 12, 37, 39). Bortezomib demonstrates anti-tumor and radiosensitizing effects in HNSCC cell lines and SCC animal models which exhibit constitutively activated NF-κB ( 4– 11, 32), and these responses are associated with inhibition of NF-κB, its target genes and expected cytopathic effects (11 and 32). The anti-tumor effects of bortezomib against HNSCC in vitro and in murine models, and its suppressive effects against radiation induced NF-κB activation ( 39), led us to develop a phase I clinical trial, to investigate the optimal dose, schedule, toxicity and anti-tumor effects of combination therapy of bortezomib and radiation in patients with HNSCC. In this trial, heterogeneity in response to the combination therapy has also been observed, with 5/17 evaluable patients treated to date demonstrating objective responses (8, Van Waes C, unpublished data). Identification of molecular mechanisms for these differences in sensitivity, and markers for selection of therapy with bortezomib and/ or additional agents, is desirable. In this study, we identified a bortezomib sensitive cell line, UM-SCC-11B, and a cell line of isogenic origin from the same patient, UM-SCC-11A, which demonstrated relatively less sensitivity to bortezomib, similar to other members of a nine UM-SCC line panel. Between the two cells lines, we observed significant differences in their response to treatment, in terms of proteasome inhibition, the accumulation of ubiquitinated proteins and corresponding effects on activation of transcription factors NF-κB and AP-1. Activation of AP-1 inhibited by JNK and p38 antagonists sensitized the more resistant line to the effects of bortezomib. These findings suggest that differences in proteasome-dependent effects on NF-κB and AP-1 may contribute to the differential sensitivity of HNSCC to bortezomib. Understanding such molecular differences involved in the cellular responses to bortezomib could provide biomarkers to guide us in improving the selection and potential combinations of treatment to be used with bortezomib. Head and Neck Squamous Cell Carcinoma Lines Human UM-SCC cell lines were derived from patients with SCC arising from sites in the upper aerodigestive tract at the University of Michigan, Ann Arbor, MI, following informed consent, as described previously (40, 41 Supplemental Table 1). The cell lines established from each patient specimen are designated by a numeric designation, and where isolates from two different time points or anatomical sites were obtained from the same patient, the designation includes an alphabetical suffix (i.e., “A” or “B”). The cell lines used were maintained in Eagle’s minimal essential media supplemented with 10% fetal bovine serum and penicillin/streptomycin. Reagents Proteasome inhibitor bortezomib was obtained from Millennium Inc (Boston, MA) under a Materials Cooperative Research and Development Agreement ( 8, 11). TNF-α was purchased from R&D Systems (Minneapolis, MN). Rabbit polyclonal antibodies against β-tubulin (H-235, sc-9104) and ubiquitin (FL-76, sc-9133), and goat anti-rabbit IgG-HRP (sc-2004) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The chemical inhibitor for JNK, SP600125, was obtained from Biomol (Plymouth Meeting, PA); p38 inhibitor, SB203580, was obtained from Calbiochem/EMD Biosciences (San Diego, CA); and inhibitors for MEK, UO126 and PD98059, were obtained from Promega (Madison, WI) and Calbiochem/EMD Biosciences (San Diego, CA), respectively. Measurement of Cell Proliferation and Viability by Trypan Blue Exclusion Cell proliferation was assessed by MTT assay. 5X103 cells were plated in a 96 well microtiter plate in quadruplicates and treated by bortezomib the next day. MTT labeling reagent (Boehringer Mannheim, Indianapolis, IN) was added at day 1, 3, 5 after treatment, and colorimetric optical density (O. D.) were measured at 570 nm by a microplate autoreader (Biotek, Winooski, VT). For trypan blue exclusion assay, 2X105 UM-SCC-11A cells and 3X105 UM-SCC-11B cells were plated in T-25 flask, and exposed to bortezomib 48 hours after plating. Cells were harvested, stained with trypan blue dye, and visually counted under microscope at different time points. Apoptosis and Cell Cycle Analysis by Flow Cytometry UM-SCC-11A and-11B cells were plated in T-25 flask and treated with bortezomib at different concentrations as indicated. Cells were harvested at 12, 24 and 96 hours after treatment and 105 cells were collected for DNA cell cycle and apoptosis analysis using cycleTEST Plus DNA reagent kit (Becton Dickinson, San Jose, CA). Cellular DNA was labeled with propidium iodide, and DNA content of 10,000 cells in each sample was measured by flow cytometry (FACScan, Becton Dickinson, San Jose, CA). Percentage of apoptotic and viable cells in each phase of cell cycle were analyzed by ModFit LT software (Verity Software House, Topsham, Maine). SCC Xenograft Tumor Model BALB/c SCID mice (4–6 week old) were obtained from the National Cancer Institute, Frederick Cancer Research and Development Center (Frederick, MD), and housed in a pathogen-free animal facility. 1.5x10 7 UM-SCC-11A or -11B cells were injected subcutaneously over the flanks of immunodeficient BALB/c SCID mice ( 11, 32). When ~0.3 cm 3 palpable tumors were developed, 2.0 mg/kg/dose of bortezomib (maximum tolerated dose tested previously) was administered by intraperitoneal injections on a Monday/Wednesday/Friday schedule weekly for a total of 3 doses. Tumor size was measured weekly with an engineer caliper by an independent animal care personnel, and presented as tumor volume that resulted from 2(tumor width) X tumor length. Terminal Deoxynucleotidyl Transferase (TdT)-Mediated dUTP Digoxigenin Nick-End Labeling (TUNEL) SCC xenograft tumor specimens were harvested at different time points after bortezomib treatment, and thin frozen sections were processed by American Histo Lab (Gaithersburg, MD). TUNEL assay was carried on the specimens according to the manufacturer’s instructions (Chemicon International, Inc., Temecula, CA). Measurement of Proteasome Activity Proteasome activity was measured according to the protocol by Elliott et al ( 42). 10 μg of protein lysate was incubated with the synthetic peptide substrate LLVY-AMC (Bachem, San Carlos, CA) in each well of the 96 well plate in triplicates, and the release of the fluorophore AMC was measured by Wallac Victor 2 1420 multilabel counter at 37 oC (λex=380nm, and λem=450nm). The kinetics of proteasome inhibitory activity was measured for five repetitive times after 10-second shaking and 4 minute incubation. NF-κ B Binding Activity To detect NF-κB p65 binding activity, cell nuclear extracts were obtained using Nuclear Extraction kit (Active Motif, Carlsbad, CA). 20 μg of nuclear extracts were used in each reaction in triplicates. The binding activity was evaluated using TransAM kits for NF-κB family members according to manufacturer’s protocol (Active Motif). The optical density was measured at wavelength of 450 nm by a microplate reader (Biotek, Winooski, VT). Reporter Gene Assays 5XNF-κB- and 7XAP-1-luciferase reporter gene constructs were purchased from Stratagene (La Jolla, CA). Murine dominant negative IκBαM plasmid (pCMX IκBαM) was kindly provided by Dr. Inder M Verma (Salk Institute, La Jolla, CA, Ref 7). Cells were plated at 5X10 4/well in the 24 well plate in triplicates, and the next day were transfected with 0.3 μg of plasmid DNA plus Effectene Transfection Reagents for 3 hours at 37°C following manufacturer’s suggestions (Qiagen, Hilden, Germany). After transfection, the cells were grown in EMEM plus 10% FBS and exposed to bortezomib. The cell lysates were harvested and reporter gene activities were assayed using the Dual-Light Luciferase Reporter Gene Assay System (Tropix, Bedford, MA). The chemoluminescence activity was measured by a Monolight 2010 luminometer (Analytical Luminescence Lab, San Diego, CA). Isolation of Whole-cell Extract Cells were plated in 100-mm tissue culture dishes and allowed to grow until 80–90% confluent, and the whole-cell lysates were isolated using nuclear extract kit (Active Motif, Carlsbad, CA). Cultured cells were washed with ice-cold PBS/phosphatase inhibitors and scraped, and then were lysed in 100 μl lysis buffer and pipetted up and down several times. The lysates were incubated for 10 min on ice followed by centrifuged at 14,000g for 20 min at 4°C, and the supernatants were collected and stored at −80°C. Total protein concentrations were determined using the Pierce BCA protein assay method. Western Blot Analysis 20 μg whole-cell lysates were mixed with Laemmli loading buffer (containing β-mercaptoethanol) and heated at 100°C for 5 min. The samples were loaded onto 10% Tris-Glycine pre-cast gels and electrophoresed at 140V for 90 min. The proteins were transferred to 0.45 μm nitrocellulose membranes (Invitrogen, Carlsbad, CA) for 2 hr at 20V at room temperature using the Invitrogen Gel Blot Module. Immunoblotting was conducted according to manufacture’s specifications. Primary antibodies were diluted in 5% nonfat powdered milk prepared from TBS-T and diluted at 1:500, and the secondary goat anti-rabbit IgG-HRP were used at 1:1000 dilution. Each blot was incubated with Pierce Super Signal West Pico substrate and exposed to Kodak X-OMAT film. Differences in Sensitivity of UM-SCCCell Lines to Bortezomib We screened nine UM-SCC cell lines and normal human keratinocytes for their sensitivity to bortezomib by MTT cell proliferation assay, and identified UM-SCC-11B as the most sensitive cell line (, Supplemental Table 1). The IC 50 value for bortezomib in UM-SCC-11B was 0.37 nM, whereas a modest sensitivity was observed among cell lines, -1, 5, -6, 11A, -22A and -22B, where the IC 50swere ~0.54–1.1 nM. The most resistant cell lines identified were UM-SCC-9 and-38, where the IC 50 values were 7.19 and 8.37 nM. While UM-SCC-11B and the other UM-SCC studied demonstrated greater activation and dependence on NF-κB activation than UM-SCC-9, the degree of sensitivity to bortezomib among the cell lines was not strictly proportional to NF-κB activation (, and Ref 6, 7, 43), suggesting that other factors could contribute to the differential sensitivity observed. Cultured human normal keratinocytes also showed a relatively low IC 50 (0.48 nM), which was comparable to the tumor cells, while exhibiting minimal activation of NF-κB in culture ( 43). To further investigate the potential mechanisms of the differential sensitivity of UM-SCC cell lines to bortezomib, we elected to study separate isolates of the same primary hypopharyngeal HNSCC which, interestingly, showed different alterations in sensitivity to bortezomib and other agents before (UM-SCC-11A) and after chemotherapy with cisplatin (UM-SCC-11B, Supplemental Table 1, Ref. 40, 41). After cisplatin chemotherapy, -11B cells showed elevated constitutive and EGF induced NF-κB and STAT3 levels, and resistance to EGFR inhibitor, gefitinib ( 7, 44, 45), but greater sensitivity to bortezomib in vitro and in murine xenografts ( 11). When we compared the cell viability of the two cell lines by trypan blue exclusion assay after treatment with bortezomib, we confirmed our previous observation that UM-SCC-11B cells were more sensitive to bortezomib than -11A cells, in terms of induction of cytotoxicity and reduction of viable cells (). Anti-Tumor Activity of Bortezomib In Vivo The anti-tumor activity of bortezomib in UM-SCC-11A and -11B was further evaluated in a xenograft tumor model in SCID mice ( 11, 32). An HNSCC xenograft tumor model established by subcutaneous injection of tumor cells was used in this study in order to study the potential differences in response to bortezomib, as the rate of growth of tumors when implanted orthotopically causes premature death caused by obstruction of swallowing or respiration. However, this model represents many molecular characteristics similar to other HNSCC xenografts implanted in orthotopic murine models ( 46). At maximal tolerated doses, we observed tumor regression of UM-SCC-11B tumors after three injections (, right panel), but no effect upon -11A tumors (, left panel). Tumor specimens were also harvested after bortezomib treatment. Bortezomib treatment induced -11B tumor cell nuclear condensation and tissue degradation (, upper right panel), as well as TUNEL labeling consistent with apoptosis (, lower right panel), when compared with the vehicle controls ( left panels). However, the morphology of -11A tumors after treatment were the same as the vehicle controls, and there was no TUNEL labeling observed in the 11A tumor samples treated with bortezomib (data not shown). Bortezomib Induced Cell Cycle Blockade in UM-SCC-11A and Cell Apoptosis in UM-SCC -11B Cells To further determine if bortezomib induced differential effects on cell cycle and/or apoptosis in UM-SCC-11A and -11B cells, DNA cell cycle analysis was performed by flow cytometry. UM-SCC-11A cells treated with 10–8 M bortezomib demonstrated a significant cell cycle arrest in G2/M phase between 12–24 hours after treatment (). In contrast, UM-SCC-11B cells, treated with the same concentration of bortezomib, showed a slight alteration of cell cycle distribution at 12 hours, and a markedly increased sub-G0/G1 DNA fragmentation peak within 24 hours after treatment, consistent with apoptosis (). At the lower concentration (10–9 M), no significant cell cycle alteration or apoptosis were observed in UM-SCC-11A cells (), but apparent changes of cell cycle were observed in -11B cells (). Differential Inhibition of Proteasome Activity by Bortezomib in UM-SCC-11A and -11B Cells We next explored if the different sensitivity to bortezomib observed in UM-SCC-11A and -11B was potentially due to intrinsic differences in proteasome inhibition. We tested the proteasome activity in both cell lines after bortezomib treatment at different time points, and found that strong inhibition was observed by four hours after bortezomib treatment. As shown in , at a lower dose of bortezomib (10–9 M), there was a slight but significant stimulation of proteasome activity in UM-SCC-11A cells, but inhibition in -11B cells (). At higher concentrations, we saw stronger inhibition of proteasome activity in UM-SCC -11B cells when compared with -11A cells (). In addition, we examined the accumulation of ubiquitinated proteins, an effect of proteasome inhibition, in the two cell lines following treatment with 10–8 M bortezomib at the different time points. More significant accumulation of ubiquitylated proteins was observed in UM-SCC-11B cells than in -11A cells (), consistent with stronger proteasome inhibition. Thus, UM-SCC-11A and -11B cells appear to exhibit differences in effect of bortezomib on proteasome inhibition and accumulation of ubiquitinated proteins. Bortezomib Effects on NF-κ B and AP-1 Activity To explore if differences in sensitivity could reflect differences in NF-κB and/or AP-1 activation previously detected in HNSCC, we examined basal and bortezomib induced activity. UM-SCC-11A and -11B cells exhibited high basal levels of constitutively activated NF-κB as demonstrated by NF-κB p65 binding () and reporter activities (Chen Z, data not shown). UM-SCC-11A cells exhibited higher transfection efficiency and absolute luciferase activity than -11B cells, however, after normalization to β-gal activity, UM-SCC-11B showed two-fold higher reporter activity than -11A cells (Chen Z, data not shown), consistent with the difference in binding activity. To examine if the differential sensitivity to bortezomib observed in UM-SCC-11A and -11B cells were related to their NF-κB activity, NF-κB-luciferase reporter assays were conducted after cells were exposed to 10–8M bortezomib at different time points (). Bortezomib transiently increased NF-κB activity in -11A cells at the early time points (4 and 8 hours after treatment), followed by relatively weaker NF-κB inhibition, at 16 and 24 hours after treatment (). Bortezomib reduced NF-κB reporter activity in -11B cells by 4 hours, and reached statistical significance to nearly complete inhibition by 16–24 hours after treatment (). Next, we asked whether NF-κB activity in the cells could be stimulated and inhibited by the classical mechanisms, such as TNF-α, or overexpression of a dominant negative IκBαM by transient transfection. In both cell lines, TNF-α was able to stimulate NF-κB reporter activity, and expression of the dominant negative IκBαM was able to completely suppress NF-κB reporter activity (). Our laboratory previously identified another transcription factor AP-1, which is constitutively co-activated and promotes proliferation and the production of proangiogenic cytokines in HNSCC ( 6, 7). Since the induction of AP-1 occurs in response to many types of cellular stress, and degradation of proteins such as cJun that comprise AP-1 occurs via the proteasome ( 18), we tested if bortezomib affected AP-1 activity in these cells. As shown in , bortezomib significantly increased AP-1 activity, as early as 4 hours after treatment in UM-SCC-11A cells, but not in the UM-SCC-11B cells (). Blockade of Up-stream Signaling Pathways Activating AP-1 Enhanced Bortezomib Effects in UM-SCC-11A Cells We next examined if the constitutive and bortezomib-inducible AP-1 activity by bortezomib could contribute to the relative resistance of UM-SCC-11A cells. Since multiple signal transduction pathways upstream of AP-1 may contribute to AP-1 activation, several chemical inhibitors for these pathways were tested, including specific JNK inhibitor SP600125, p38 inhibitor SB203580, and MEK/ERK inhibitors UO126 and PD98059 ( 27). We titrated these chemical inhibitors to sub-optimum dosage for inhibition of MTT in -11A cells, and combined them with bortezomib at 10 –8M (). We found the combination of bortezomib with JNK inhibitor SP600125 showed significant synergistic inhibitory effects on cell proliferation of -11A cells (, left panel). Other inhibitors, such as SB203580 for the p38 pathway ( right panel), and UO126 and PD98059 for the ERK pathway showed lesser combined activity (data not shown). Consistent with this, the combination of bortezomib and the JNK inhibitor SP600125 showed more significant inhibition on AP-1 reporter activity in -11A cells than either agent alone (). Our data support the hypothesis that activation of JNK and AP-1 pathways could contribute to bortezomib resistance in UM-SCC-11A cells. In this study, sensitivity to bortezomib was determined in a panel of nine UM-SCC lines and cultured normal human keratinocytes, and the role of NF-κB and other potential mechanisms in bortezomib sensitivity were examined. The pharmacological concentrations for the IC 50 ranged between 0.37 and 8.37nM (), consistent with the range of sensitivities for other solid and hematopoietic malignancies ( 11, 19– 28, 30– 35), rather than the higher IC 50 reported previously for HNSCC lines ( 29). While UM-SCC-11B and most other UM-SCC cell lines studied exhibited greater NF-κB activation and bortezomib sensitivity than UM-SCC-9 and non-malignant keratinocytes, the differential sensitivity to bortezomib among the cell lines in this study was not strictly proportional to the baseline differences in activation of NF-κB (, Ref 6, 8, 43), indicating that other NF-κB-independent mechanism(s) potentially affected by proteasome inhibition may contribute to the differential sensitivities ( 16– 18). Comparison of UM-SCC-11A and -11B cell lines isolated from the same patient before and after chemotherapy with cisplatin displayed different sensitivities and response both in vitro and in vivo (, Supplemental Table 1), suggesting that differences in mechanism(s) intrinsic to these cell lines could contribute to their differential sensitivities to bortezomib. Comparison of baseline and post-treatment levels of BCL-2/BCL-xL and HSP27, previously implicated in intrinsic differences in sensitivity of MM ( 25, 33), did not substantially differ between the lines (data not shown). While UM-SCC-11A and B both showed similarly high levels of NF-κB activation, they displayed greater differences in the kinetics and extent of proteasome inhibition, ubiquitination, and AP-1 activation (, ). Inhibitors of JNK and p38 inhibited activation of AP-1 and sensitized cells to bortezomib (), indicating that bortezomib-induced signal activation of these kinase pathways both contribute to these differences in sensitivity, and may serve as potential targets for combined therapy in HNSCC. The difference in IC 50 for bortezomib between UM-SCC-11A and -11B cells in vitro was less than one log in the concentration range, but such a difference corresponded with a significantly altered drug activity for -11A or -11B xenograft tumor growth in vivo (). At the maximally tolerated dosage in this preclinical xenograft tumor model, bortezomib induced -11B tumor regression () and cell apoptosis by TUNEL staining (), but had no inhibitory effect on -11A tumors at the maximally tolerated dose ( and data not shown), indicating that such a dosage of bortezomib did not achieve sufficient intratumoral concentrations to induce regression of UM-SCC-11A tumors in vivo. Such differences in the response to bortezomib by the two cells lines in vitro and in vivo are consistent with differences in clinical responses in our ongoing phase I trial of combination therapy of bortezomib with re-irradiation in patients with recurrent HNSCC ( 8, 47), where objective responses have been observed in a subset 5/17 evaluable patients treated with 0.6 mg/m 2 with different schedules of re-irradiation ( 8, 47). The limited responses observed in our clinical trial could partially be explained by the expected activity of the relatively low dose of bortezomib (0.6mg/m 2), studied thus far with different schedules of re-irradiation ( 8, 47), where the bortezomib serum concentration estimated from previous pharmacokinetic studies would fall in the lower concentration range ( 47– 49). When limiting concentrations are present, increased interstitial pressures have been implicated in poorer bioavailability and relatively lower response rates in solid malignancies compared to hematopoietic malignancies ( 48, 49). In addition, although the IC 50 for UM-SCC lines overlapped that of keratinocytes, clinically significant skin or mucosal toxicity of bortezomib other than mild rash or mucositis has not been observed in our study, or even with higher doses used in major phase II clinical trials ( 8, 16, 17, 36– 38, 47– 49). We next investigated proteasome-dependent effects of bortezomib in the paired UMSCC-11A and -11B lines to examine the hypothesis that differences in kinetics of proteasome inhibition and/or ubiquitination could contribute to the differential response and potentially serve as indicators of sensitivity. Our studies show that bortezomib inhibits the proteasome to a lesser extent and with a slower kinetics in -11A cells when compared with -11B cells (), consistent with such a hypothesis. Previously, bortezomib-induced inhibition of proteasome activity was reported to peak in peripheral blood cells or tumor specimens one hour after delivery of the drug ( 8, 28). However, we found bortezomib-induced proteasome inhibition reached its peak in HNSCC at about four hours after delivery even in vitro, and the peak effect was sustained for ~8 hours (Chen Z, unpublished observation), similar to observations in a murine SCC transplantation model ( 50). Consistent with these differences, accumulation of ubiquitinated proteins and the extent of inhibition of NF-κB (dependent on IκB ubiquitination and degradation ( 16, 17) after bortezomib treatment were significantly lower in -11A cells, when compared with -11B cells (). In addition, we observed that, at the lower dose of bortezomib (10 –9M), the proteasome activity in -11A was stimulated instead of inhibited (), suggesting the possibility that proteasome activity was modified by bortezomib-inducible mechanisms that remain to be defined. These differential effects of bortezomib and low dose stimulation of proteasome activity observed in -11A and -11B cells could potentially occur upstream of or at the proteasome level, possibly due to differences in cellular drug uptake or metabolism, ubiquitinases or chaperones affecting stability, or to basal and inducible levels of proteasome components ( 16– 18), which warrant investigation in future studies. In the past, investigations have been conducted to identify the molecular mechanisms underlying sensitivity of different cancers to bortezomib, particularly in MM. Based on these studies, there remains controversy regarding if and how much of bortezomib’s anti-cancer effect is dependent on NF-κB activity. In MM, NF-κB dependent expression of IL-6, ICAM and VEGF by bone marrow stroma and endothelium has been implicated in response and resistance to bortezomib ( 51, 52), but such a microenvironment is not identical to that of HNSCC or other solid cancers which arise outside the bone marrow. Further, we did not detect differences in other NF-κB related genes such as BCL-2 or HSP27 implicated in intrinsic resistance of MM (data not shown). Our data in UM-SCC cells, as well as others from some solid tumors, suggest that NF-κB activity is an important, but not the only factor determining bortezomib’s anti-cancer effects ( 19– 29, 32– 35). Previously, our laboratory showed that murine and human HNSCC exhibit increased constitutive NF-κB activation and that specific inhibition of NF-κB increases cell death ( 5– 11). At concentrations that inhibit NF-κB activation, bortezomib was able to block cell proliferation and induce cell death in a panel of murine and human SCC cells from the skin and aerodigestive tract, as well as tumorigenesis in murine HNSCC xenograft animal models ( 11). In addition, specific inhibition of NF-κB by an inducible IκBα signal phosphorylation mutant as well as bortezomib was associated with significant reduce proinflammatory and proangiogenic cytokines, Gro-1 and VEGF production in vitro and suppression of angiogenesis in vivo ( 10, 11). Furthermore, among nine UM-SCC cell lines screened in this study, UM-SCC-9 and -38 are the most resistant lines to bortezomib and exhibit relatively low constitutive NF-κB activity (), while UM-SCC-11B is the most sensitive cell line and exhibits high basal NF-κB binding and reporter activity (6, 43 and this study, Chen Z, data not shown). Bortezomib’s inhibitory effect corresponded to differences in the extent of inhibition on NF-κB activity by the reporter gene assay (). In UM-SCC-11B cells, bortezomib induced a stronger and earlier inhibition of NF-κB reporter activity (); by contrast, in UM-SCC-11A cells, the NF-κB activity was stimulated early and less inhibited by the drug (). Together, our present and previous data are consistent with the hypothesis that NF-κB is important but not the only determinant of bortezomib activity in HNSCC. In addition to the differential effects of bortezomib on NF-κB activity, UM-SCC-11A cells, relative to -11B, showed significant enhancement of AP-1 reporter activity (), another pro-proliferative and pro-survival factor often co-activated in HNSCC ( 6, 7). Consistent with the finding of constitutive activation of AP-1 identified in UM-SCC cell lines ( 6, 7), we have shown that a subset of the tumor cell lines, including UM-SCC-11A and -11B, over-express gene clusters with an increased prevalence of predicted AP-1 binding motifs in their promoter region, when compared with the binding frequencies of other gene promoters ( 41). Since AP-1 includes JUN/FOS family protein complexes ( 2, 3), and proteasome inhibitors can potentially activate AP-1 via decreased turnover of JUN/FOS, one mechanism potentially contributing to differences in AP-1 induction in HNSCC may be due to the greater accumulation of JUN/FOS available to accept signals from upstream signal pathways ( 18). Previously, our laboratory obtained evidence for aberrant activation of signal pathways upstream of AP-1, including the intermediate kinase pathways, MEK/MAPK ( 7), and the membrane growth factor receptors, such as IL-1R ( 5), EGFR ( 7, 44, 45 ) and hepatocyte growth factor (HGF)/c-Met ( 53). Our experiments using inhibitors of MAPK pathway kinases JNK, p38 or MEK/ERK (, Ref 7) support the hypothesis that JNK and p38 are major pathways activating AP-1 and promoting proliferation of UM-SCC-11A following bortezomib treatment. This observation is supported by our phase I clinical investigation, where bortezomib alone significantly suppressed nuclear localization of the phosphorylated active form of the p65 subunit of NF-κB, but not ERK, JUN or STAT3 phosphorylation by immunohistochemitry in patients’ tumor specimens ( 48). Our data are also consistent with more recent reports that blockage of MAPK pathways can significantly enhance cytotoxicity and apoptosis in hematopoietic malignancies ( 23– 25, 34). p38, JNK and c-JUN activation have been studied in MM, and the effects of p38 pathway on bortezomib sensitivity in MM ( 23– 25, 34) are similar to HNSCC as described in this study. However, increased c-Jun and JNK activation was associated with cell apoptosis induced by bortezomib ( 22, 27) in MM, which is opposite what we observed in HNSCC and presented in this paper. Such differences in c-JUN and JNK functions could be tissue specific, as suggested by Podar et al ( 22). In HNSCC cells, activation of JNK and AP-1 pathways has been consistently related to pro-proliferative, anti-apoptosis and angiogenesis functions ( 7, 44, 45), which is different from observations drawn from MM. Based on the this and previous studies, we have worked to develop biomarkers including the molecules involved in NF-κB and AP-1 pathways in our clinical trial with bortezomib combined with radiation therapy ( 8, 11, 48). We have obtained biopsies from a subset of HNSCC patients with accessible tumors both pre- and post-treatment, taken 24 hours after the initial bortezomib treatment but before the initiation of radiotherapy. The intensity of baseline NF-κB/REL, ERK, STAT3 nuclear staining, apoptosis (TUNEL) and proliferation (Ki-67) was quantified by immunohistochemistry, and effects of bortezomib on phospho-RelA, RelB, c-Rel, p105/p50, p100/p52, phospho-ERK1/2 and phospho-STAT3 was evaluated. HNSCC tumor specimens showed increased baseline nuclear staining for all five NF-κB subunits, p-ERK1/2, and p-STAT3 when compared with normal mucosa. In biopsied tumor specimens obtained 24 hours post-bortezomib from these patients, apoptosis was detected by TUNEL staining in association with inhibition of nuclear phospho-RELA subunit of NF-κB in 3/4 patients, but there was no inhibition or increase in staining of nuclear phospho-ERK1/2 and/or cJUN in these patients who subsequently progressed, and in a patient with activated nuclear phospho-ERK1/2 and cJUN in the absence of NF-κB who showed no apoptosis or clinical response ( 8, 48). We conclude that the bortezomib dosage tested in this trial inhibits activation of subunits of the canonical NF-κB pathway, however, it did not block nuclear activation of the non-canonical NF-κB or other prosurvival signal pathways, such as ERK or JUN, which may contribute to early or delayed progression after partial responses observed in patients with HNSCC following bortezomib. The data from the clinical study are consistent with our findings in this study, and will be presented elsewhere ( 48). In conclusion, distinct differences in sensitivity to bortezomib were observed in UM-SCC-11A and -11B cells, including in xenograft models (–). Our data suggest that the mechanisms underlying these differences may exist at various levels of the regulation, from differences in the extent of proteasome inhibition through variable effects on NF-κB and AP-1 transcription factor activity ( and ). We showed that bortezomib inhibited proteasome and NF-κB activities in UM-SCC-11A cells to a lesser extent when compared with -11B cells ( and ), and increased AP-1 activity in UM-SCC-11A cells was observed after treatment with bortezomib (). Blockage of JNK and other MAPK activities could reduce AP-1 activity and synergize with bortezomib anti-tumor activity in UM-SCC-11A cells (). This pre-clinical study suggests that assays comparing the effects of bortezomib in pre- and post-treatment biopsies on proteasome activity, NF-κB and AP-1 signal activation, proliferation, and apoptosis could be useful to develop biomarkers for clinical trials of cancers, in which heterogeneous responses to bortezomib have been observed. These biomarkers could be relevant for determining optimal regimens for more effective therapy. Inhibition of JNK and other MAPK pathways, which block AP-1 activity and enhance bortezomib anti-tumor effects, may be one such combination meriting further investigation. We would like to express our appreciation to Millennium Pharmaceuticals, Inc (Boston, MA) for providing bortezomib for this study, and Drs. Mark Rolfe and William Riordan, for providing the proteasome activity assay protocol. | | HNSCC | head and neck squamous cell carcinoma | | | NF-kB | nuclear factor-kappaB | | | AP-1 | activator protein-1 | | | MTT | 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide | | | JNK | c-Jun amino-terminal kinases | | | IκB | inhibitor kappaB, TNF-α: Tumor necrosis factor-alpha | | | TUNEL | terminal deoxynucleotidyl transferase (TdT)-mediated dUTP digoxigenin nick-end labeling, BCA: bicinchoninic acid, (Pierce protein assay) | | | MM | multiple myeloma, MEK: MAP/ERK kinase, MAPK/ERK: mitogen-activated protein kinases/extracellular signal-regulated kinases | | | Gro1 | Gro1 oncogene/CXCL1/KC | | | PI3K | phosphatidylinositol 3-kinase | | | VEGF | vascular endothelial growth factor |
1. Aggarwal BB. Nuclear factor-κB: The enemy within. Cancer Cell. 2004;6:203–8. [PubMed] 2. Shaulian Karin M. AP-1 as a regulator of cell life and death. Nature Cell Biol. 2002;4:E131–6. [PubMed] 3. Eferl R, Wagner EF. AP-1: A double-edeged sword in tumorigenesis. Nature Rev Cancer. 2003;3:859–68. [PubMed] 4. Van Waes C. Nuclear factor-kappaB in development, prevention, and therapy of cancer. Clin Cancer Res. 2007;13:1076–82. [PubMed] 5. Allen CT, Ricker JL, Chen Z, Van Waes C. Role of activated nuclear factor-kappaB in the pathogenesis and therapy of squamous cell carcinoma of the head and neck. Head Neck. 2007;29:959–71. [PubMed] 6. Ondrey FG, Dong G, Sunwoo J, et al. Constitutive activation of transcription factors NF- (kappa)B, AP-1, and NF-IL6 in human head and neck squamous cell carcinoma cell lines that express pro-inflammatory and pro-angiogenic cytokines. Mol Carcinog. 1999;26:119–29. [PubMed] 7. Bancroft CC, Chen Z, Yeh J, et al. Effects of pharmacologic antagonists of epidermal growth factor receptor, PI3K and MEK signal kinases on NF-kappaB and AP-1 activation and IL-8 and VEGF expression in human head and neck squamous cell carcinoma lines. Int J Cancer. 2002;99:538–48. [PubMed] 8. Van Waes C, Chang AA, Lebowitz PF, et al. Inhibition of nuclear factor-kappaB and target genes during combined therapy with proteasome inhibitor bortezomib and reirradiation in patients with recurrent head-and-neck squamous cell carcinoma. Int J Radiat Oncol Biol Phys. 2005;63:1400–12. [PubMed] 9. Duffey DC, Chen Z, Dong G, et al. Expression of a dominant-negative mutant inhibitor-kappaBalpha of nuclear factor-kappaB in human head and neck squamous cell carcinoma inhibits survival, proinflammatory cytokine expression, and tumor growth in vivo. Cancer Res. 1999;59:3468–74. [PubMed] 10. Loercher A, Lee TL, Ricker JL, et al. Nuclear Factor-kB is an important modulator of the altered gene expression profile and malignant phenotype in squamous cell carcinoma. Cancer Res. 2004;64:6511–23. [PubMed] 11. Sunwoo JB, Chen Z, Dong G, et al. Novel proteasome inhibitor PS-341 inhibits activation of nuclear factor-kappa B, cell survival, tumor growth, and angiogenesis in squamous cell carcinoma. Clin Cancer Res. 2001;7:1419–28. [PubMed] 12. Druzgal CH, Chen Z, Yeh NT, et al. A pilot study of longitudinal serum cytokine and angiogenesis factor levels as markers of therapeutic response and survival in patients with head and neck squamous cell carcinoma. Head Neck. 2005;27:771–84. [PubMed] 13. Allen C, Duffy S, Teknos T, Islam M, Chen Z, Albert PS, Wolf G, Van Waes C. Nuclear Factor-kappaB-Related Serum Factors as Longitudinal Biomarkers of Response and Survival in Advanced Oropharyngeal Carcinoma. Clin Cancer Res. 2007;13:3182–3190. [PubMed] 14. Zhang PL, Pellitteri PK, Law A, et al. Overexpression of phosphorylated nuclear factor-kappa B in tonsillar squamous cell carcinoma and high-grade dysplasia is associated with poor prognosis. Mod Pathol. 2005;18:924–32. [PubMed] 15. Chung CH, Parker JS, Ely K, et al. Gene Expression Profiles Identify Epithelial-to-Mesenchymal Transition and Activation of Nuclear Factor-{kappa}B Signaling as Characteristics of a High-risk Head and Neck Squamous Cell Carcinoma. Cancer Res. 2006;66:8210–8. [PubMed] 16. Adams J. The development of proteasome inhibitors as anticancer drugs. Cancer Cell. 2004;5:417–21. [PubMed] 17. Adams J, Palombella VJ, Sausville EA, et al. Proteasome inhibitors: a novel class of potent and effective antitumor agents. Cancer Res. 1999;59:2615–22. [PubMed] 18. Treier M, Staszewski LM, Bohmann D. Ubiquitin-dependent c-Jun degradation in vivo is mediated by the delta domain. Cell. 1994;78:787–98. [PubMed] 19. An J, Sun Y, Fisher M, Rettig MB. Maximal apoptosis of renal cell carcinoma by the proteasome inhibitor Bortezomib is nuclear factor-KB dependent. Mol Cancer Ther. 2004;3:727–36. [PubMed] 20. Dai Y, Rahmani M, Grant S. Proteasome inhibitors potentiate leukemic cell apoptosis induced by the cyclin-dependent kinase inhibitor flavopiridol through a SAPK/JNK-and NF-κB-depedent process. Oncogene. 2003;22:7108–22. [PubMed] 21. Hipp MS, Urbich C, Mayer P, et al. Proteasome inhibition leads to NF-kappaB-independent IL-8 transactivation in human endothelial cells through induction of AP-1. Eur J Immunol. 2002;32:2208–2. [PubMed] 22. Podar K, Raab MS, Tonon G, et al. Up-Regulation of c-Jun Inhibits Proliferation and Induces Apoptosis via Caspase-Triggered c-Abl Cleavage in Human Multiple Myeloma. Cancer Res. 2007;67:1680–1688. [PubMed] 23. Tai YT, Fulciniti M, Hideshima T, et al. Targeting MEK induces myeloma cell cytotoxicity and inhibits osteoclastogenesis. Blood. 2007;110:1656–63. [PubMed] 24. Kojima K, Konopleva M, Samudio IJ, Ruvolo V, Andreeff M. Mitogen-activated protein kinase kinase inhibition enhances nuclear proapoptotic function of p53 in acute myelogenous leukemia cells. Cancer Res. 2007;67:3210–9. [PubMed] 25. Navas TA, Nguyen AN, Hideshima T, et al. Inhibition of p38a MAPK enhances proteasome inhibitor-induced apoptosis of myeloma cells by modulating Hsp27, Bcl-XL, Mcl-1 and p53 levels in vitro and inhibits tumor growth in vivo. Leukemia. 2006;20:1017–1027. [PubMed] 26. Johnson TR, Stone K, Nikrad M, et al. The proteasome inhibitor PS-341 overcomes TRAIL resistance in Bax and caspase 9-negative or Bcl-xL overexpressing cells. Oncogene. 2003;22:4953–63. [PubMed] 27. Yang Y, Ikezoe T, Saito T, Kobayashi M, Koeffler HP, Taguchi H. Proteasome inhibitor PS-341 induces growth arrest and apoptosis of non-small cell lung cancer cells via the JNK/c-JUN/AP-1 signaling. Cancer Sci. 2004;95:176–80. [PubMed] 28. Yu J, Tiwari S, Steiner P, Zhang L. Differential apoptotic response to the proteasome inhibitor Bortezomib [VELCADE, PS-341] in Bax-deficient and p21-deficient colon cancer cells. Cancer Biol Ther. 2003;2:694–9. [PubMed] 29. Fribley A, Zeng Q, Wang CY. Proteasome Inhibitor PS-341 Induces Apoptosis through Induction of Endoplasmic Reticulum Stress-Reactive Oxygen Species in Head and Neck Squamous Cell Carcinoma Cells. Mol Cell Biol. 2004;24:9695–704. [PubMed] 30. Williams S, Pettaway C, Song R, Papandreou C, Logothetis C, McConkey DJ. Differential effects of the proteasome inhibitor Bortezomib on apoptosis and angiogenesis in human prostate tumor xenografts. Mol Cancer Ther. 2003;2:835–43. [PubMed] 31. Nawrocki ST, Bruns CJ, Harbison MT, et al. Effects of the proteasome inhibitor PS-341 on apoptosis and angiogenesis in orthotopic human pancreatic tumor xenografts. Mol Cancer Ther. 2002;1:1243–53. [PubMed] 32. Duan J, Friedman J, Nottingham L, Chen Z, Ara G, Van Waes C. Nuclear factor-kappaB p65 small interfering RNA or proteasome inhibitor bortezomib sensitizes head and neck squamous cell carcinomas to classic histone deacetylase inhibitors and novel histone deacetylase inhibitor PXD101. Mol Cancer Ther. 2007;61:37–50. [PubMed] 33. Chauhan D, Li G, Shringarpure R, et al. Blockade of Hsp27 Overcomes Bortezomib/Proteasome Inhibitor PS-341 Resistance in Lymphoma Cells. Cancer Res. 2003;63:6174–7. [PubMed] 34. Hideshima T, Podar K, Chauhan D, et al. p38 MAPK inhibition enhances PS-341 (Bortezomib)-induced cytotoxicity against multiple myeloma cells. Oncogene. 2004;23:8766–76. [PubMed] 35. Minderman H, Zhou Y, O'loughlin KL, Baer MR. Bortezomib activity and in vitro interactions with anthracyclines and cytarabine in acute myeloid leukemia cells are independent of multidrug resistance mechanisms and p53 status. Cancer Chemother Pharmacol. 2007;60:245–55. [PubMed] 36. Kondagunta GV, Drucker B, Schwartz L, et al. Phase II trial of Bortezomib for patients with advanced renal cell carcinoma. J Clin Oncol. 2004;22:3720–25. [PubMed] 37. Fanucchi MP, Fossella FV, Belt R, et al. Randomized phase II study of bortezomib alone and bortezomib in combination with docetaxel in previously treated advanced non-small-cell lung cancer. J Clin Oncol. 2006;24:5025–33. [PubMed] 38. Alberts SR, Foster NR, Morton RF, et al. PS-341 and gemcitabine in patients with metastatic pancreatic adenocarcinoma: a North Central Cancer Treatment Group (NCCTG) randomized phase II study. Ann Oncol. 2005;16:1654–61. [PubMed] 39. Van Waes C, Sunwoo JB, DeGraff W, Mitchell JB. Radiosensitization and proteasome inhibition. In: Adams J, editor. Cancer Drug Discovery and Development: Proteasome Inhibitors in Cancer Therapy. Humana Press; Totowa, NJ: 2004. pp. 123–31. 40. Krause C, Carey T, Ott R, Hurbis C, McClatchey K, Regezi J. Human squamous cell carcinoma. Establishment and characterization of new permanent cell lines. Arch Otolaryngol. 1982;107:703–10. [PubMed] 41. Yan B, Yang X, Lee TL, et al. Genome-wide identification of novel expression signatures reveal distinct patterns and prevalence of binding motifs for p53, NF-kappaB and other signal transcription factors in head and neck squamous cell carcinoma. Genome Biol. 2007;8(5):R78. [PubMed] 42. Elliott PJ, Soucy TA, Pien CS, Adams J, Lightcap ES. Assays for proteasome inhibition. Methods Mol Med. 2003;85:163–72. [PubMed] 43. Yu M, Yeh J, Van Waes C. Protein kinase casein kinase 2 mediates inhibitor-kappaB kinase and aberrant nuclear factor-kappaB activation by serum factor(s) in head and neck squamous carcinoma cells. Cancer Res. 2006;66:6722–31. [PubMed] 44. Lee TL, Yeh J, Van Waes C, Chen Z. Epigenetic modification of SOCS-1 differentially regulates STAT3 activation in response to interleukin-6 receptor and epidermal growth factor receptor signaling through JAK and/or MEK in head and neck squamous cell carcinomas. Mol Cancer Ther. 2006;5(1):8–19. [PubMed] 45. Pernas FG, Allen CT, Winters ME, et al. Proteomic signatures of epidermal growth factor receptor and survival signal pathways correspond to gefitinib sensitivity in head and neck cancer. 99th Annual Meeting of American Association for Cancer Res; San Diego, CA. April 12–16th, 2008. 46. Zhang X, Liu Y, Gilcrease MZ, et al. A Lymph Node Metastatic Mouse Model Reveals Alterations of Metastasis-Related Gene Expression in Metastatic Human Oral Carcinoma Sublines Selected from a Poorly Metastatic Parental Cell Line. CANCER. 2002;95:1663–1672. [PubMed] 47. Allen CT, Chang AA, Chen Z, et al. Bortezomib-induced apoptosis with limited clinical response is accompanied by inhibition of canonical but not alternative NF-κB subunits or other prosurvival signal pathways activated in head and neck cancer. Clinical Cancer Research (accepted). 2008 48. Russo A, Fratto ME, Bazan V, et al. Targeting apoptosis in solid tumors: the role of bortezomib from preclinical to clinical evidence. Expert Opin Ther Targets. 2007;11(12):1571–1586. [PubMed] 49. Hamilton AL, Eder JP, Pavlick AC, et al. Proteasome inhibition with bortezomib (PS-341): a phase I study with pharmacodynamic end points using a day 1 and day 4 schedule in a 14-day cycle. J Clin Oncol. 2005;23:6107–16. [PubMed] 50. Poff JA, Traughber B, Allen C, et al. Pulsed-high intensity focused ultrasound enhances apoptosis and growth inhibition of squamous cell carcinoma xenografts by proteasome inhibitor bortezomib. Radiology. 2008 (in press). 51. Hideshima, Anderson Preclinical studies of novel targeted therapies. Hematol Clinics North Am. 2007;21:1071–91. 2007. 52. Hideshima T, Akiyama M, Hayashi T, et al. Targeting p38 MAPK inhibits multiple myeloma cell growth in the bone marrow milieu. Blood. 2003;101(2):703–5. [PubMed] 53. Dong G, Chen Z, Li ZY, Yeh NT, Bancroft CC, Van Waes C. Hepatocyte growth factor/scatter factor-induced activation of MEK and PI3K signal pathways contributes to expression of proangiogenic cytokines interleukin-8 and vascular endothelial growth factor in head and neck squamous cell carcinoma. Cancer Res. 2001;61:5911–8. [PubMed]
|
The publisher's final edited version of this article is available free at 
