- Email: [email protected]
Insulin-like growth factor-1 receptor and ligand targeting in head and neck squamous cell carcinoma
Cancer Letters 248 (2007) 269–279 www.elsevier.com/locate/canlet
Insulin-like growth factor-1 receptor and ligand targeting in head and neck squamous cell carcinoma q Mark G. Slomiany a, Leigh Ann Black a, Megan M. Kibbey a, Melissa A. Tingler a, Terry A. Day b, Steven A. Rosenzweig a,* a
b
Department of Cell and Molecular Pharmacology and Experimental Therapeutics and Hollings Cancer Center, Medical University of South Carolina, 173 Ashley Avenue Charleston, SC 29425, USA Department of Otolaryngology, Medical University of South Carolina, 173 Ashley Avenue Charleston, SC 29425, USA Received 30 May 2006; received in revised form 31 July 2006; accepted 2 August 2006
Abstract IGF-1 receptor (IGF-1R) signaling is associated with increased tumorigenesis of epithelial cancers. In light of recent epidemiological studies correlating high circulating levels of IGF-1 with increased risk of second primary tumors (SPTs) of the head and neck, we examined IGF system and epidermal growth factor receptor (EGFR) expression in human head and neck squamous cell carcinoma (HNSCC) matched pairs and a cross-section of HNSCC cell lines. Employing the latter, we demonstrated that IGF-1 stimulated S-phase transition in a PI 3-K/Akt and Erk-dependent manner in 5 of 8 cell lines, with Erk activation being dependent upon EGFR kinase activity. IGF-1 stimulated thymidine incorporation was inhibited by treatment with IGFBP-3, the IGF-1R tyrosine kinase inhibitor NVP-AEW541, or the EGFR tyrosine kinase inhibitor AG1478. Combining IGFBP-3 with NVP-AEW541 or AG1478 abrogated IGF-1 responses at 10-fold lower doses than either compound alone. These results demonstrate the potential for co-targeting the IGF system and EGFR in HNSCC. Ó 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Insulin-like growth factor-1 receptor; Insulin-like growth factor binding protein-3; Epidermal growth factor; Head and neck squamous cell carcinoma; Receptor tyrosine kinase inhibitors; NVP-AEW541; AG1478
Abbreviations: Akt, protein kinase B (pkB) a-isoform; BCA, bicinchoninic acid; BLOTTO, bovine lacto transfer technique optimizer; BSA, bovine serum albumin; Erk, extracellular-signal-regulated kinase; FBS, fetal bovine serum; HRP, horse radish peroxidase; HNSCC, head and neck squamous cell carcinoma; IGF-1, insulin-like growth factor 1; IGFBP, IGF binding protein; MAPK, MEK/mitogen-activated protein kinase; PBS, phosphate-buffered saline; PI 3-K, phosphatidylinositol 3-kinase; rhIGFBP-2 or -3, recombinant human IGFBP-2 or -3; TBS, Tris-buffered saline; [3H]Thymidine, tritiated thymidine. q This work was supported, in part, by Grant CA78887 from the National Institutes of Health and Department of Defense grant to Hollings Cancer Center, (N6311601MD10004) to S.A.R. and (N6311602MD200) to T.A.D. M.G.S. was supported by an Abney Foundation Research Scholarship and M.M.K. was supported by National Research Service Award 5F30DE015249 from the NIDCR and by the Dental Medicine Scientist Training Program, Colleges of Dental Medicine and Graduate Studies, Medical University of South Carolina. * Corresponding author. Tel.: +1 843 792 5841; fax: +1 843 792 2475. E-mail address: [email protected] (S.A. Rosenzweig). 0304-3835/$ - see front matter Ó 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2006.08.004
270
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
1. Introduction Head and neck squamous cell carcinoma (HNSCC) accounts for more than 90% of all pharyngeal and oral cavity tumors. With nearly 8000 deaths a year nationally, it constitutes approximately 4% of all cancers in the United States and is one of the six most frequent cancers worldwide. Although tobacco and alcohol are primary risk factors, it is clear that additional factors contribute to this cancer [1]. On this basis, novel molecular markers for HNSCC are being sought. Epidermal growth factor receptor (EGFR) overexpression has been reported in a variety of human cancers, including breast, lung, colon, prostate, brain, ovary, and HNSCC [2–5]. Indeed, targeting the EGFR represents one of the newest therapies for HNSCC in many years [6]. Although targeting EGFRs using EGFR monoclonal antibodies or EGFR tyrosine kinase inhibitors have yielded promising results in preclinical studies, limited anti-tumor effects have been obtained when administered to cancer patients [3,4,7,8]. Compelling evidence exists for a role of the IGF/IGF-IR system in human neoplasia. Accordingly, interfering with IGF-1R-signaling may provide an attractive strategy for treating some cancers, including HNSCC. The biologic actions of IGF-1 and IGF-2 are initiated by their interaction with the IGF-1 receptor (IGF-IR). The six IGFBPs primarily act as IGF-1/2 antagonists that bind their ligands with high affinity and block their access to the IGF-1R [9]. In certain cases however, preincubation of cells with IGFBP-3 before IGF-1 treatment leads to the accumulation of cell-bound forms of IGFBP-3 with lowered affinity for IGF, which may enhance the IGF-1/IGF-1R interaction. However, this mechanism has never been proven explicitly (reviewed in [10]). In addition, epidemiologic studies have shown that high circulating levels of IGF-1 are associated with increased risk of second primary tumors (SPTs) of the head and neck [11] breast [12], prostate [13], lung [14], and colorectal [15] cancer, whereas high IGFBP3 concentrations are inversely associated with a risk of these cancers [16–18]. Moreover, methylation of the IGFBP-3 promoter, resulting in decreased local expression of IGFBP-3, has been associated with poor prognosis in non-small-cell
lung cancer [19]. Accordingly, the IGFBPs or IGFBP-mimetics may serve as therapeutics or lead compounds for development of small molecule IGF antagonists. Similarly, several experimental approaches to inhibiting primary tumor growth, including IGF-1R blocking antibodies, dominant negative mutants, antisense cDNA, and oligonucleotides to downregulate IGF-1R expression have demonstrated the therapeutic potential of interfering with IGF-1R-mediated signaling in vivo [20]. NVP-AEW541 represents the first small molecule IGF-1R tyrosine kinase inhibitor with proven in vivo anti-tumor activity for potential therapeutic application [20]. Targeting the IGF-1R may impact downstream signaling by other receptors. Accumulating evidence suggests that activation of heterologous receptors, including the IGF-1R [21], can induce EGFR transactivation or at a minimum require EGFR kinase activity for Erk activation. This can proceed through indirect means, via receptor activated shedding of the EGFR ligands including HB-EGF [21] and amphiregulin [22] or by direct means, such as IGF-1R/EGFR complex formation [21]. IGF-1R/EGFR cross-talk has been described in normal human mammary epithelial cells [23], as well as for tamoxifenresistant MCF-7 cells where increased sensitivity to the proliferative effects of IGF-1/2 following estradiol or tamoxifen treatment were blocked by the IGF-1R tyrosine kinase inhibitor AG1024 or the anti-IGF-1R monoclonal antibody, a IR-3 [24]. In this report, we surveyed 12 matched tumor/ normal pairs of human HNSCC specimens for the presence of EGFR and IGF system components. All specimens expressed EGFRs and IGF-1Rs. Of the eight pairs demonstrating IGFBP-2 immunoreactivity, expression was lower in tumor versus normal in seven of the samples (Table 1 and Fig. 1). Examination of 8 established human HNSCC cell lines revealed IGF1R, IGFBP-2, IGFBP-3, and IGF-1/2 expression. IGF-1 stimulated S-phase transition in 5 out of 8 cell lines. IGF-1 stimulated thymidine incorporation could be inhibited by treatment with IGFBP-3, NVP-AEW541 or AG1478. The combination of IGFBP-3 with either receptor tyrosine kinase inhibitor resulted in a 10-fold greater inhibition than that obtained with either reagent alone, providing a rationale for inhibiting the IGF-1R in EGF-sensitive-HNSCC.
Table 1 Summary of IGF system component and EGFR expression in HNSCC tumor–normal matched pairs Sample #
1240
1256 1341 1285 1454 1548 1787 1957 1964 1967 1988
Normal Tumor Normal Tumor Normal Tumor Normal Tumor Normal Tumor Normal Tumor Normal Tumor Normal Tumor Normal Tumor Normal Tumor Normal Tumor Normal Tumor
Site
Grade/description
Stagea
Tongue Moderately differentiated
Pt2
pN0
pMx
Moderate to poor differentiation
pT4a
pN2
pMx
Moderately differentiated
pT2
pN2b
pMx
Well differentiated
pT2
pN0
pMx
Well differentiated
pT1
pN0
pMx
Moderate to poor differentiation
pT2
pNx
pMx
Moderately differentiated
pT2
pN3
pMx
Poorly differentiated
pT3
pN0
pMx
Well differentiated
pT2
pN2b
pMx
Moderately differentiated
pT2
pN0
pMx
Moderate to poor differentiation
pT2
pN2b
pMx
Moderately differentiated
pT3
pN0
pMx
Tongue Tongue Hypopharynx Floor of mouth Tongue Oropharynx Tongue Oropharynx Tongue Oral cavity (mandible) Tongue
EGFR
IGF1R
p p p p p p p p p p p p p p p p p p p p p p p p
p p p p p p p p p p p p p p p p p p p p p p p p
IGFBP 2
3
5
++ +++ ++ + ++ + ++ + ++ ++ ++ + ++ +
+ + + + ++ + + + + + + +++ ++ + + +++ ++ ++ +
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
1246
Condition
Normal and tumor samples (100 lg) were processed for EGFR, IGF-1R, and IGFBP immunoreactivity. For IGFBPs, +, minimal optical density (OD); ++, moderate OD; and +++, high OD. a Cancer staging was defined according to standard procedures (http://www.upmccancercenters.com/cancer/headneck/staging.html).
271
272
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
Fig. 1. IGFBP-2 expression in HNSCC tumor matched pairs. Normal and tumor samples (100 lg) were processed for IGFBP and b-actin immunoreactivity as described in Section 2.
2. Experimental procedures 2.1. Materials and reagents SCC-9, SCC-25, and FaDu cells were obtained from the ATCC (Manassas, VA). UM-SCC-1, -5, -12, -14A, and -22A were generous gifts from Dr. Thomas E. Carey (University of Michigan, Department of Otorhinolaryngology). Snap frozen matched pair HNSCC biopsy samples were obtained from the Hollings Cancer Center Tissue Bank with IRB approval. Fetal bovine serum (FBS) was purchased from Atlas Biologicals (Fort Collins, CO). DMEM and wortmannin were purchased from Sigma (St. Louis, MO). AG1478 was purchased from Calbiochem (La Jolla, CA). NVP-AEW541 was kindly provided by Novartis (Basel, Switzerland). IGF-1 was generously provided by Genentech, Inc. (San Francisco, CA). rhIGFBP-3 (N109D), IGFBP-2, -3, and -5 polyclonal antibodies, IGF-1 and IGF-2 mouse monoclonal antibodies, and EGFR clone LA22 antibody were from Upstate, Inc. (Charlottesville, VA). IGFBP-2 was expressed and purified from CHO cells as previously described [25]. IGF-1R b-subunit (H-60) antibody was obtained from Santa Cruz Biotechnologies, Inc. (Santa Cruz, CA). ActiveÒ IGF-1 and IGF-2 ELISA kits were purchased from Diagnostic Systems Laboratory Inc. (Webster, TX). Re-Blot Plus (mild and strong) Western blot stripping solutions were purchased from Chemicon (Temecula, CA). Western blot and immunohistochemistry specific phospho-Akt(Ser473) polyclonal antibodies and Akt antibody were obtained from Cell Signaling Technology (Beverly, MA). Erk antibody was obtained from BD Transduction Laboratories (San Jose, CA). NeutravidinÒ-horse radish peroxidase (neutravidin-HRP) and BCA reagent were obtained from Pierce (Rockford, IL). Phospho-44/42 MAPK antibody was purchased from New England Biolabs
(Beverly, MA). ECL Western blotting detection reagent, [125I]IGF-1, and [3H]thymidine were purchased from GE Healthcare Bio-Sciences Corp. (Piscataway, NJ). All other chemicals were of reagent grade or higher. TM
2.2. Tissue culture SCC [26], UM-SCC [27], and FaDu [28] cells were cultured as previously described and maintained at 37 °C in a humidified 5% CO2–95% air incubator. For receptor binding assays, cells were seeded in 24-well plates, grown overnight to 80% confluency, and serum-starved (FBS was eliminated in all experiments) for 24 h before the indicated treatment. For whole cell lysates, secretion, and cell cycle studies, cells were seeded in six-well plates. Following pretreatment for 2 h with fresh serum free medium containing inhibitors dissolved in ethanol, cells were treated with ethanol vehicle, inhibitors, IGF-1, IGFBP-2, or IGFBP-3. For the survey of IGF-1/2 and IGFBP secretion in Table 1, cells were seeded in 10 cm dishes to 80% confluency, plates were rinsed three times with PBS, and fresh serum free medium added. After 24 h, conditioned medium was dialyzed, concentrated, and processed for ELISA, ligand blot, and immunoblot analysis. 2.3. Receptor binding assays HNSCC cells serum starved for 24 h, were washed with HMS± (25 mM Hepes, 104 nM NaCl, 5 mM MgCl2, and 0.01% soybean trypsin inhibitor, containing 0.2% BSA. Triplicate wells were treated with increasing concentrations of IGF-1 (0–10 lM IGF-1) and 20,000 cpm of [125I]IGF-1 for 30 min at 37 °C. Cells were rinsed with HMS±, dissolved in 2 N NaOH, and radioactivity was quantified by gamma spectrometry using a Compugamma spectrometer (LKB-Wallac,
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
273
2.5. Cell cycle analysis
Turku, Finland). Assays were repeated a minimum of four times.
Following a 24 h treatment, HNSCC cells were harvested, fixed in 70% EtOH, stained with propidium iodide, and analyzed by flow cytometry on a FACSCalibur instrument (BD Biosciences, Franklin Lakes, NJ), in the Hollings Cancer Center Flow Cytometry Core Facility.
2.4. ELISA, immunoblot, and ligand blot analysis HNSCC matched pair samples were sonicated in lysis buffer and subjected to immunoblot analysis. Conditioned medium from HNSCC cell lines was analyzed by ELISA for IGF-1 and -2 according to the manufacturer’s instructions (DSL, Inc, Webster, TX). Whole cell lysates for Akt/Erk immunoblots were incubated for 15 min before being prepared using a modified RIPA buffer containing 50 mM Tris–HCl, pH 7.4, 1% Triton X-100, 150 mM NaCl, 10 mM EDTA, 1 mM PMSF, 10 lg/ml aprotinin and leupeptin, 2 mM sodium orthovanadate, and 10 mM NaF. Protein content was determined by BCA assay, and 100 lg aliquots were solubilized in SDS sample buffer. IGFBPs in conditioned medium were quantified by immunoblot after a 12 h incubation. Conditioned medium was precipitated in 10% trichloroacetic acid (TCA), the resultant pellet washed with acetone and solubilized in SDS sample buffer. Lysates and conditioned medium so collected were resolved on 10% and 12.5% non-reducing polyacrylamide gels, respectively, transferred to nitrocellulose and subjected to ligand (in the case of IGFBPs in Table 2) [29] and immunoblot analysis. Immunoblots were blocked, probed, and visualized as previously described [30], with either 1 lg/ml phospho-Akt (Ser 473) polyclonal, 1 lg/ml Akt polyclonal, 1:2000 phospho-p44/42 MAPK monoclonal, 1:5:000 Erk polyclonal,1:10,000 bactin monoclonal, or 1:1000 IGFBP-2, -3, or -5 polyclonal antibodies. To reprobe immunoblots, antibodies were removed from the nitrocellulose via the application of Re-Blot Plus-Mild stripping solution according to the manufacturers instructions (Amersham Biosciences, Piscataway, NJ).
2.6. [3H]Thymidine incorporation SCC-9 cells were plated in 48-well dishes and grown to 60–80% confluency in FBS supplemented growth medium. Wells were washed three times with PBS and replaced with SFM for 24 h. Cells were subsequently treated for 2 h with inhibitors followed by treatment for 21 h. Medium was then replaced and [3H]thymidine incorporation assay was performed as described [31]. 2.7. Statistical analysis Data are expressed as the mean between independent experiments ± standard error. Statistical differences in multiple experiments between multiple conditions were analyzed by ANOVA with a one way analysis of variance (95% confidence interval) and Bonferroni Post Test.
3. Results 3.1. The IGF-1 system in HNSCC matched pairs IGF system components and EGFR expression profiles of 12 HNSCC matched pairs are shown in Table 1. Immunoblot analysis revealed IGF-1R and EGFR expression in all samples, whereas IGFBP-3 immunoreactivity was detected in 10/12 normal and 9/12 tumor samples. Interestingly, of the eight pairs demonstrating IGFBP-2 immunoreactivity, expression was lower in
Table 2 Summary of IGF system component expression in HNSCC cell lines Cell line
SCC-9 SCC-25 FaDu UM-SCC-1 UM-SCC-5 UM-SCC-12 UM-SCC-14A UM-SCC-22A
Site
Tongue Tongue Hypopharynx Floor of mouth Supraglottis Larynx Floor of mouth Hypopharynx
IGF-a
IGF-1Rb
1
2
++ ++ ++ + + ++ ++ +
+++ +++ + ++ +++ ++
169,400 ± 14,504 346,900 ± 35,068 531,633 ± 40,130 285,366 ± 76,410 171,500 ± 11,107 40,000 ± 10,209 69,625 ± 5,112 29,500 ± 2,771
IGFBP 2 p p p p p p p p
3 p p p p p p p p
5 p p p p p
Following 24 h of serum starvation, conditioned medium from subconfluent HNSCC cell lines was processed for IGF-1/2 and IGFBP immunoreactivity, while cells were assayed for IGF-1R expression. (For IGFs +, 10 pg/100,000 cells; ++, 100 pg/100,000 cells; and +++, 1000 pg/100,000 cells). Values shown are representative of three or more independent experiments for each cell line. a IGF-1 and -2 quantified by specific ELISAs. b Receptor numbers based on 125I-IGF-1 binding studies and Scatchard analysis.
274
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
tumor versus normal in seven of the samples (Table 1 and Fig. 1). None of the pairs exhibited detectable IGFBP-5 immunoreactivity. Blots were not assayed for 1GFBP-1, -4, or -6. In addition, none of the pairs exhibited detectable IGF-1 or IGF-2 immunoreactivity (data not shown). 3.2. The IGF-1 system in HNSCC cell lines To further examine IGF system representation in squamous cell carcinoma, 8 established HNSCC cell lines were examined for IGF-1R, IGFBP, IGF-1, and IGF-2 expression (Table 2). IGF-1 binding analyses revealed that all eight cell lines expressed IGF-1Rs. Based on Scatchard analyses, six of these cell lines overexpressed the IGF-1R, based on a normal physiologic range of 10,000–30,000 receptors/cell [32]. IGF-1 and IGF-2 were detected in the conditioned medium from all cell lines, except for FaDu and SCC-5 cells. Ligand blot and immunoblot analyses revealed that all eight cell lines expressed IGFBP-2 and 3, whereas IGFBP-5 was only present in 5 out of 8 cell lines. 3.3. Effect of IGF-1 and IGFBPs on S-phase entry To determine whether the proliferative effects of IGF-1 correlated with receptor number, serum starved HNSCC cells were treated with IGF-1 (10 nM) for 24 h and harvested for cell cycle analysis. As shown in Fig. 2A, IGF-1 stimulated a significant (p < 0.05) increase in S-phase entry in 5 of the cell lines. The three cell lines showing greatest response, SCC-9, UMSCC-5, and UMSCC-22A, were used in subsequent analyses. To ascertain whether the IGFBPs could modulate IGF-1R activity, cells were treated with recombinant IGFBP-2 or IGFBP-3 and S-phase entry was evaluated (Fig. 2B). Addition of both IGFBP-2 and IGFBP-3 completely blocked IGF-1 stimulated S-phase entry, while having no effect on this parameter when added alone.
Fig. 2. Effect of IGF-1 and IGFBP-3 on S-phase entry in HNSCC cell lines. Subconfluent, serum-starved (24 h) HNSCC cells were treated for 24 h with 10 nM IGF-1 (A) or 1 nM IGF-1 and IGFBP-2 or IGFBP-3 (B). Cells were harvested, stained, and analyzed by flow cytometry. Error bars represent standard deviation between three individual analyses by flow cytometry. Significant differences were observed (*p < 0.05, compared to unstimulated). (A and B) Representative of three independent experiments.
3.4. Role of PI 3-kinase and Erk pathways in S-phase entry Immunoblot analysis of whole cell lysates revealed the presence of constitutively active Akt (phosphoAkt) in UMSCC-1, FaDu, and SCC-25 cells (Fig. 3A). Constitutively active Erk was detected in all but the UMSCC-22A cell line. IGF-1 addition increased Akt activity in all cell lines and Erk activity among all cell lines except the two that showed the highest basal pErk levels (SCC-25 and UMSCC-12). SCC-9, UMSCC-5, and UMSCC-22A cells were responsive to EGF stimulation; a representative analysis of SCC-9 cells showing robust Erk activation in response to EGF is shown in Fig. 3B (UMSCC-5 and UMSCC-22A data not shown). IGF-1 and EGF both stimulated [3H]thymidine incorporation. Coupled with our previous
findings in SCC-9 cells that demonstrate IGF-1 stimulated S-phase entry to be Erk-dependent [33], our current findings further support a role for Erk activation in HNSCC proliferation (Fig. 3C). Inhibition of EGFR tyrosine kinase activity with AG1478 completely abrogated IGF-1 stimulated Erk activity in SCC-9, (Fig. 4A), UMSCC-5 (Fig. 4B), and UMSCC-22A (Fig. 4C) cells. Significantly, AG1478 had no effect on IGF-1 stimulated Akt activity whereas NVP-AEW541 inhibited IGF-1 stimulated Akt and Erk activity without affecting basal Erk activity. We next tested the effect of these inhibitors on IGF-1 stimulated S-phase entry. AG1478 reduced the basal level of S-phase in unstimulated cells below control values
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
275
Fig. 3. IGF-1 signaling and EGFR activity in HNSCC cell lines. Subconfluent, serum-starved (24 h) HNSCC cells (A) and SCC-9 cells (B) were treated for 15 min as indicated. Whole cell lysates were processed for immunoblot analysis. (C) SCC-9 cells were plated in 48-well dishes and grown to 70% confluency. They were serum starved for 24 h, treated as indicated for 21 h, and processed for [3H]thymidine incorporation. Error bars represent standard deviation between three independent experiments. Significant differences were observed (*p < 0.05, compared to unstimulated). (A–C) Representative of three or more independent experiments.
(Fig. 4D). IGF-1 treatment significantly (p < .05) increased S-phase entry (Fig. 4D), which was inhibited by NVP-AEW541 treatment. NVP-AEW541 treatment reduced the basal level of S-phase entry to control (UMSCC-5) or below (SCC-9 and UMSCC-22A) control levels and completely abrogated IGF-1 stimulation. 3.5. Combined effects of IGFBP-3 and receptor tyrosine kinase inhibitor treatment Treatment of SCC-9 cells with 100 nM IGFBP-3 reduced IGF-1 (10 nM) stimulated [3H]thymidine incorporation by 57.3 ± 7.3% (where a 100% reduction would equal a final value of 0) (Fig. 5A and B). Treatment of IGF-1 stimulated cells with 0.1 or 1 lM NVP-AEW541, reduced [3H]thymidine incorporation by 60.9 ± 2.0% and 96.5 ± 0.5%, respectively; both values were significantly (p < 0.001) below basal levels (Fig. 5A). Co-addi-
Fig. 4. Effect of IGF-1R and EGFR kinase inhibitors on IGF-1 stimulated signaling pathways and S-phase entry in HNSCC cell lines. Subconfluent, serum-starved (24 h) SCC-9 (A), UMSCC-5 (B), and UMSCC-22A (C) cells were pretreated for 2 h with inhibitors as indicated before addition of 10 nM IGF-1 plus inhibitor for 15 min. Cells were harvested for immunoblot analysis (A–C). (D) SCC-9, UMSCC-5, and UMSCC-22A cells were plated in 48-well dishes and grown to 70% confluency. Following 24 h of serum starvation, they were pretreated with inhibitors for 2 h as indicated, before treatment for 21 h with inhibitors and 10 nM IGF-1. Cells were harvested, stained, and analyzed by flow cytometry. Error bars represent standard deviation between three independent experiments. Significant differences were observed (*p < 0.05, **p < 0.01, compared to unstimulated unless indicated by brackets). (A–D) Representative of three or more independent experiments.
276
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
Fig. 5. Effect of IGF-1R and EGFR tyrosine kinase inhibition and IGFBP-3 on IGF-1 stimulated [3H]thymidine incorporation. (A and B) SCC-9 cells plated in 48-well dishes were grown to 70% confluency and serum starved for 24 h. After a 2 h pretreatment with inhibitor, cells were treated as indicated for 21 h and processed for [3H]thymidine incorporation. Error bars represent standard deviation between three independent experiments. Significant differences were observed (*p < 0.05, **p < 0.01, ***p < 0.001, compared to unstimulated unless indicated by brackets; indicates an insignificant difference). (A and B) Representative of three or more independent experiments.
tion of 10 nM IGFBP-3 with 0.1 lM NVP-AEW541 did not further reduce [3H]thymidine incorporation (data not shown). Application of 0.01 lM NVP-AEW541 or 10 nM IGFBP-3 reduced IGF-1 stimulated [3H]thymidine incorporation by 2.8 ± 3.4% and 18.1 ± 4.3%, respectively. However, when they were combined we obtained an enhanced inhibitory action, resulting in a 41.5 ± 4.8% reduction in [3H]thymidine incorporation. This combined dosage was as effective as 100 nM IGFBP-3 in reducing [3H]thymidine incorporation. Similarly, 250 nM AG1478 reduced IGF-1 stimulated [3H]thymidine incorporation by 72.4 ± 2.3%, a value significantly (p < 0.001) below basal levels (Fig. 5B). However, AG1478 (25 nM) plus IGFBP-3 (10 nM), each of which individually reduced IGF-1 stimulated [3H]thymidine incorporation 31.8 ± 2.3% and 18.1 ± 4.3%, respectively, caused a com-
bined reduction of 61.4 ± 4.74%. This combination was as effective as 100 nM IGFBP-3 or 250 nM AG1478 alone, significantly reducing [3H]thymidine incorporation (p < 0.001) below basal levels.
4. Discussion Signaling by the IGF-1R is upregulated in many solid tumors, contributing to the malignant phenotype by enhancing resistance to EGF receptor and HER2 inhibitors, protection from apoptosis, increased tumor cell motility, invasiveness, and metastasis [34]. EGFR overexpression has been reported in a number of cancers including head and neck [2–5]. On this basis, the IGF-1R is a
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
promising therapeutic target for innovative molecularly targeted therapies in several malignancies. Considering high IGFBP-3 concentrations are inversely associated with the risk of some cancers [16–18], IGFBP-3 may provide an additional opportunity to inhibit the IGF/IGF-1R system. Not surprisingly, an array of in vitro, in vivo, and clinical findings point to IGFBP-3 as an anticancer molecule [35]. In a small survey of HNSCC specimens, we observed a lower expression of IGFBP-2 in tumor versus normal samples. This has the potential to contribute to autocrine/paracrine IGF-1/2 signaling in HNSCC, contributing to increased tumorigenicity. Although little is known about IGFBP-2 regulation, methylation of promoter CpG islands was recently reported for IGFBP-3, accounting for reduced IGFBP-3 levels in NSCLC [19]. Alternatively, reduced IGFBP-2 levels may reflect increased proteolysis by matrix metalloproteinases as reported for IGFBP-3 [36] and IGFBP-5 [37]. Analysis of eight human HNSCC cell lines revealed that all lines secrete IGFs, IGF-1Rs, and are responsive to IGF-1, supporting a role for autocrine IGF-1R action in HNSCC. IGFBP-2 and IGFBP-3 each abrogated IGF-1 stimulated HNSCC cell proliferation consistent with the IGF-1-dependent action of IGFBPs [9]. IGF-1 stimulated PI 3-K/Akt and/or Erk pathways in the 8 HNSCC cell lines, with the latter depending upon EGFR kinase activity. This is in agreement with studies on Ca9-22 cells where EGFR kinase activity was required for IGF-1 induced Erk activity [21]. The ability of the EGFR kinase inhibitor AG1478 to abrogate IGF-1 stimulated [3H]thymidine incorporation was consistent with our previous studies demonstrating that Erk activity was required for IGF1 stimulated SCC-9 cell S-phase entry [33]. The decrease in [3H]thymidine incorporation below control levels obtained by combining IGFBP-3 with AG1478 suggests that basal [3H]thymidine incorporation is, in part, the result of an autocrine IGF loop as seen in MCF-7 cells [25]. In addition, we have previously shown [33] that IGF-1R signaling in HNSCC cells induces VEGF signaling, which in turn acts in an autocrine stimulatory manner. Autocrine effects by VEGF or other RTKs may cause a basal level of cell activity which would be unaffected by IGFBP-3 addition but could be reduced by NVP-AEW541. The
277
nearly complete lack of [3H]thymidine incorporation observed with the IGF-1R kinase inhibitor NVP-AEW541 corroborates this view. NVP-AEW541 treatment in conjunction with IGFBP-3 led to enhanced attenuation of IGF-1 stimulated [3H]thymidine incorporation. The combination of AG1478 and IGFBP-3 caused a synergistic reduction in [3H]thymidine incorporation with 10-fold or lower concentrations of the two providing an inhibitory effect equivalent to either component alone at high dosages. In light of these findings, co-targeting of the IGF system and EGFR, employing IGF-1 antagonists and pharmacological inhibitors, appears promising. Such combination therapies offer specificity advantages inherent at lower concentrations of small molecule receptor tyrosine kinase inhibitors, as well as reduced side effects. Additionally, pharmacological inhibitors rarely exert 100% influence over their respective target. By directing antagonists to IGF-1, inherent limitations of any single inhibitor may be overcome. Similarly, when compared to antibody-based therapies, IGFBP-3 may provide an advantage by being a natural, small, diffusible protein having dual specificity against both IGF-1 and IGF-2. As such, our data support a role for IGFBP-3 as a therapeutically relevant molecule to be considered in the design of innovative therapeutic regimens in treating patients with HNSCC. Acknowledgements We wish to acknowledge the Hollings Cancer Center Tissue Bank for human specimens and the Hollings Cancer Center Flow Cytometry Core Facility for cell cycle analysis. References [1] M.T. Canto, S.S. Devesa, Oral cavity and pharynx cancer incidence rates in the United States, Oral. Oncol. 38 (2002) 610–617. [2] D.S. Solomon, R. Brandt, F. Ciardiello, N. Normanno, Epidermal growth factor-related peptides and their receptors in human malignancies, Crit. Rev. Oncol. Hematol. 19 (1995) 183–232. [3] D.E. Davies, S.G. Chamberlin, Targeting the epidermal growth factor receptor for therapy of carcinomas, Biochem. Pharmacol. 51 (1996) 1101–1110. [4] V. Rusch, J. Mendelsohn, E. Dmitrovsky, The epidermal growth factor receptor and its ligands as therapeutic targets in human tumors, Cytokine Growth Factor Rev. 7 (1996) 133–141.
278
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
[5] J.R. Grandis, A. Chakraborty, Q. Zheng, M.F. Melhem, D.J. Tweardy, Downmodulation of TGF-alpha protein expression with antisense oligonucleotides inhibits proliferation of head and neck squamous carcinoma but not normal mucosal epithelial cells, J. Cell Biochem. 69 (1998) 55–62. [6] L.J. Wirth, R.J. Haddad, N.I. Lindeman, X. Zhao, J.C. Lee, V.A. Joshi, C.M. Norris Jr., M.R. Posner, Phase I study of gefitinib plus celecoxib in recurrent or metastatic squamous cell carcinoma of the head and neck, J. Clin. Oncol. 23 (2005) 6976–6981. [7] J. Baselga, Monoclonal antibodies directed at growth factor receptors, Ann. Oncol. 11 (Suppl. 3) (2000) 187–190. [8] K. Garber, Tyrosine kinase inhibitor research presses on despite halted clinical trial, J. Natl. Cancer Inst. 92 (2000) 967–969. [9] S.A. Rosenzweig, What’s new in the IGF-binding proteins?, Growth Horm IGF Res. 14 (2004) 329–336. [10] S.M. Firth, R.C. Baxter, Cellular actions of the insulin-like growth factor binding proteins, Endocr. Rev. 23 (2002) 824– 854. [11] X. Wu, H. Zhao, K-A. Do, M.M. Johnson, Q. Dong, W.K. Hong, M.R. Spitz, Serum levels of insulin growth factor (IGF-1) and IGF-binding protein predict risk of second primary tumors in patients with head and neck cancer, Clin. Cancer Res. 10 (2004) 3988–3995. [12] S.E. Hankinson, W.C. Willet, G.A. Colditz, D.J. Hunter, D.S. Michaud, B. Deroo, B. Rosner, F.E. Speizer, M. Pollak, Circulating concentrations of insulin-like growth factor-1 and risk of breast cancer, Lancet 351 (1998) 1393–1396. [13] J.M. Chan, M.J. Stampfer, E. Giovannucci, P.H. Gann, J. Ma, P. Wilkinson, C.H. Hennekens, M. Pollak, Plasma insulin-like growth factor-I and prostate cancer risk: a prospective study, Science 279 (1998) 563–566. [14] H. Yu, M.R. Spitz, J. Mistry, J. Gu, W.K. Hong, X. Wu, Plasma levels of insulin-like growth factor-1 and lung cancer risk: a case-control study, 91 (1999) 151–156. [15] J. Ma, M.N. Pollak, E. Giovannucci, J.M. Chan, Y. Tao, C.H. Hennekens, M.J. Stampfer, Prospective study of colorectal cancer risk in men and plasma levels of insulinlike growth factor (IGF)-1 and IGF-binding protein-3, J. Natl. Cancer Inst. 91 (1999) 620–625. [16] V.A. Papadimitrakopoulou, E.N. Brown, D.D. Liu, A.K. El-Naggar, J.J. Lee, W.K. Hong, H-Y. Lee, The prognostic role of loss of insulin-like growth factor-binding protein-3 expression in head and neck carcinogenesis, Cancer Let. (2005). [17] R.G. Rosenfeld, G. Lamson, H. Pham, Y. Oh, C. Conover, D.D. De Leon, S.M. Donovan, I. Ocrant, L. Giudice, Insulin-like growth factor-binding proteins, Recent Prog. Horm. Res. 46 (1990) 99–159, discussion 159–163. [18] R.F. Krywichi, D. Yee, The insulin-like growth factor family of ligands, receptors, and binding proteins, Breast Cancer Res. Treat. 22 (1992) 7–19. [19] Y.S. Chang, L. Wang, Y-A. Suh, L. Mao, S.J. Karpen, F.R. Khuri, W.K. Hong, H-Y. Lee, Mechanisms underlying lack of insulin-like growth factor-binding protein-3 expression in non-small-cell lung cancer, Oncogene 23 (2004) 6569–6580. [20] C. Garcia-Echeverria, M.A. Pearson, A. Marti, T. Meyer, J. Mestan, J. Zimmermann, J. Gao, J. Brueggen, H.G. Capraro, R. Cozens, D.B. Evans, D. Fabbro, P. Furet, D.G. Porta, J. Liebetanz, G. Martiny-Baron, S. Ruetz, F.
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28] [29]
[30]
[31]
[32]
[33]
[34]
Hofmann, In vivo antitumor activity of NVP-AEW541- A novel, potent, and selective inhibitor of the IGF-IR kinase, Cancer Cell. 5 (2004) 231–239. A. Kuribayashi, K. Kataoka, T. Kurabayashi, M. Miura, Evidence that basal activity, but not transactivation, of the epidermal growth factor receptor tyrosine kinase is required for insulin-like growth factor 1-induced activation of extracellular signal-regulated kinase in oral carcinoma cells, Endocrinology 145 (2004) 4976–4984. Q. Zhang, S.M. Thomas, V.W.Y. Lui, S. Xi, J.M. Siegfried, H. Fan, T.E. Smithgall, G.B. Mills, J.R. Grandis, Phosphorylation of TNF-{alpha} converting enzyme by gastrinreleasing peptide induces amphiregulin release and EGF receptor activation, Proc. Natl. Acad. Sci. USA 103 (2006) 6901–6906. T. Ahmad, G. Farnie, N.J. Bundred, N.G. Anderson, The mitogenic action of insulin-like growth factor I in normal human mammary epithelial cells requires the epidermal growth factor receptor tyrosine kinase, J. Biol. Chem. 279 (2004) 1713–1719. J.M. Knowlden, I.R. Hutcheson, D. Barrow, J.M.W. Gee, R.I. Nicholson, Insulin-like growth factor-1 receptor signaling in tamoxifen-resistant breast cancer: A supporting role to the epidermal growth factor receptor, Endocrinology 146 (2005) 4609–4618. M.M. Kibbey, M.J. Jameson, E.M. Eaton, S.A. Rosenzweig, Insulin-like growth factor binding protein-2: contributions of the C-terminal domain to insulin-like growth factor-1 binding, Mol. Pharmacol. 69 (2006) 833–845. J.G. Rheinwald, M.A. Beckett, Tumorigenic keratinocyte lines requiring anchorage and fibroblast support cultured from human squamous cell carcinomas, Cancer Res. 41 (1981) 1657–1663. T.E. Carey, K.A. Kimmel, D.R. Schwartz, D.E. Richter, S.R. Baker, C.J. Krause, Antibodies to human squamous cell carcinoma, Otolaryngol. Head Neck Surg. 91 (1983) 482– 491. S.R. Rangan, A new human cell line (FaDu) from a hypopharyngeal carcinoma, Cancer 29 (1972) 117–121. S.A. Robinson, S.A. Rosenzweig, Synthesis and characterization of biotinylated forms of insulin-like growth factor-1: topographical evaluation of the IGF-1/IGFBP-2 and IGFBP-3 interface, Biochemistry 43 (2004) 11533–11545. M.G. Slomiany, S.A. Rosenzweig, IGF-1 induced VEGF and IGFBP-3 secretion correlates with increased Hif-1 alpha expression and activity in retinal pigment epithelial cell line D407, Invest. Ophthalmol. Vis. Sci. 45 (2004) 2838–2847. S.A. Robinson, S.A. Rosenzweig, Paradoxical effects of the phage display-derived peptide antagonist IGF-F1-1 on insulin-like growth factor-1 receptor signaling, Biochem. Pharmacol. 72 (2006) 53–61. M.A. Steller, C.H. Delgado, C.J. Bartels, C.D. Woodworth, Z. Zou, Overexpression of the insulin-like growth factor-1 receptor and autocrine stimulation in human cervical cancer cells, Cancer Res. 56 (1961) 1761–1765. M.G. Slomiany, L.A. Black, M.M. Kibbey, T.A. Day, S.A. Rosenzweig, IGF-1 induced vascular endothelial growth factor secretion in head and neck squamous cell carcinoma, Biochem. Biophys. Res. Commun. 342 (2006) 851–858. S. Benini, M. Zuntini, M.C. Manara, P. Cohen, G. Nicoletti, P. Nanni, Y. Oh, P. Picci, K. Scotlandi, Insulin-like growth
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279 factor binding protein 3 as an anticancer molecule in Ewing’s Sarcoma, Int. J. Cancer. 119 (2006) 1046–1049. [35] O. Ali, P. Cohen, K.W. Lee, Epidemiology and biology of insulin-like growth factor binding protein-3 (IGFBP-3) as an anti-cancer molecule, Horm. Metab. Res. 35 (2003) 726–733. [36] S.I. Miyamoto, K. Yano, S. Sugimoto, G. Ishii, T. Hasebe, Y. Endoh, K. Kodama, M. Goya, T. Chiba, A. Ochiai, Matrix metalloproteinase-7 facilitates insulin-like growth
279
factor bioavailability through Its proteinase activity on insulin-like growth factor binding protein 3, Cancer Res. 64 (2004) 665–671. [37] E. Hemers, C. Duval, C. McCaig, M. Handley, G.J. Dockray, A. Varro, Insulin-like growth factor binding protein-5 Is a target of matrix metalloproteinase-7: Implications for epithelial-mesenchymal signaling, Cancer Res. 65 (2005) 7363–7369.
Insulin-like growth factor-1 receptor and ligand targeting in head and neck squamous cell carcinoma q Mark G. Slomiany a, Leigh Ann Black a, Megan M. Kibbey a, Melissa A. Tingler a, Terry A. Day b, Steven A. Rosenzweig a,* a
b
Department of Cell and Molecular Pharmacology and Experimental Therapeutics and Hollings Cancer Center, Medical University of South Carolina, 173 Ashley Avenue Charleston, SC 29425, USA Department of Otolaryngology, Medical University of South Carolina, 173 Ashley Avenue Charleston, SC 29425, USA Received 30 May 2006; received in revised form 31 July 2006; accepted 2 August 2006
Abstract IGF-1 receptor (IGF-1R) signaling is associated with increased tumorigenesis of epithelial cancers. In light of recent epidemiological studies correlating high circulating levels of IGF-1 with increased risk of second primary tumors (SPTs) of the head and neck, we examined IGF system and epidermal growth factor receptor (EGFR) expression in human head and neck squamous cell carcinoma (HNSCC) matched pairs and a cross-section of HNSCC cell lines. Employing the latter, we demonstrated that IGF-1 stimulated S-phase transition in a PI 3-K/Akt and Erk-dependent manner in 5 of 8 cell lines, with Erk activation being dependent upon EGFR kinase activity. IGF-1 stimulated thymidine incorporation was inhibited by treatment with IGFBP-3, the IGF-1R tyrosine kinase inhibitor NVP-AEW541, or the EGFR tyrosine kinase inhibitor AG1478. Combining IGFBP-3 with NVP-AEW541 or AG1478 abrogated IGF-1 responses at 10-fold lower doses than either compound alone. These results demonstrate the potential for co-targeting the IGF system and EGFR in HNSCC. Ó 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Insulin-like growth factor-1 receptor; Insulin-like growth factor binding protein-3; Epidermal growth factor; Head and neck squamous cell carcinoma; Receptor tyrosine kinase inhibitors; NVP-AEW541; AG1478
Abbreviations: Akt, protein kinase B (pkB) a-isoform; BCA, bicinchoninic acid; BLOTTO, bovine lacto transfer technique optimizer; BSA, bovine serum albumin; Erk, extracellular-signal-regulated kinase; FBS, fetal bovine serum; HRP, horse radish peroxidase; HNSCC, head and neck squamous cell carcinoma; IGF-1, insulin-like growth factor 1; IGFBP, IGF binding protein; MAPK, MEK/mitogen-activated protein kinase; PBS, phosphate-buffered saline; PI 3-K, phosphatidylinositol 3-kinase; rhIGFBP-2 or -3, recombinant human IGFBP-2 or -3; TBS, Tris-buffered saline; [3H]Thymidine, tritiated thymidine. q This work was supported, in part, by Grant CA78887 from the National Institutes of Health and Department of Defense grant to Hollings Cancer Center, (N6311601MD10004) to S.A.R. and (N6311602MD200) to T.A.D. M.G.S. was supported by an Abney Foundation Research Scholarship and M.M.K. was supported by National Research Service Award 5F30DE015249 from the NIDCR and by the Dental Medicine Scientist Training Program, Colleges of Dental Medicine and Graduate Studies, Medical University of South Carolina. * Corresponding author. Tel.: +1 843 792 5841; fax: +1 843 792 2475. E-mail address: [email protected] (S.A. Rosenzweig). 0304-3835/$ - see front matter Ó 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2006.08.004
270
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
1. Introduction Head and neck squamous cell carcinoma (HNSCC) accounts for more than 90% of all pharyngeal and oral cavity tumors. With nearly 8000 deaths a year nationally, it constitutes approximately 4% of all cancers in the United States and is one of the six most frequent cancers worldwide. Although tobacco and alcohol are primary risk factors, it is clear that additional factors contribute to this cancer [1]. On this basis, novel molecular markers for HNSCC are being sought. Epidermal growth factor receptor (EGFR) overexpression has been reported in a variety of human cancers, including breast, lung, colon, prostate, brain, ovary, and HNSCC [2–5]. Indeed, targeting the EGFR represents one of the newest therapies for HNSCC in many years [6]. Although targeting EGFRs using EGFR monoclonal antibodies or EGFR tyrosine kinase inhibitors have yielded promising results in preclinical studies, limited anti-tumor effects have been obtained when administered to cancer patients [3,4,7,8]. Compelling evidence exists for a role of the IGF/IGF-IR system in human neoplasia. Accordingly, interfering with IGF-1R-signaling may provide an attractive strategy for treating some cancers, including HNSCC. The biologic actions of IGF-1 and IGF-2 are initiated by their interaction with the IGF-1 receptor (IGF-IR). The six IGFBPs primarily act as IGF-1/2 antagonists that bind their ligands with high affinity and block their access to the IGF-1R [9]. In certain cases however, preincubation of cells with IGFBP-3 before IGF-1 treatment leads to the accumulation of cell-bound forms of IGFBP-3 with lowered affinity for IGF, which may enhance the IGF-1/IGF-1R interaction. However, this mechanism has never been proven explicitly (reviewed in [10]). In addition, epidemiologic studies have shown that high circulating levels of IGF-1 are associated with increased risk of second primary tumors (SPTs) of the head and neck [11] breast [12], prostate [13], lung [14], and colorectal [15] cancer, whereas high IGFBP3 concentrations are inversely associated with a risk of these cancers [16–18]. Moreover, methylation of the IGFBP-3 promoter, resulting in decreased local expression of IGFBP-3, has been associated with poor prognosis in non-small-cell
lung cancer [19]. Accordingly, the IGFBPs or IGFBP-mimetics may serve as therapeutics or lead compounds for development of small molecule IGF antagonists. Similarly, several experimental approaches to inhibiting primary tumor growth, including IGF-1R blocking antibodies, dominant negative mutants, antisense cDNA, and oligonucleotides to downregulate IGF-1R expression have demonstrated the therapeutic potential of interfering with IGF-1R-mediated signaling in vivo [20]. NVP-AEW541 represents the first small molecule IGF-1R tyrosine kinase inhibitor with proven in vivo anti-tumor activity for potential therapeutic application [20]. Targeting the IGF-1R may impact downstream signaling by other receptors. Accumulating evidence suggests that activation of heterologous receptors, including the IGF-1R [21], can induce EGFR transactivation or at a minimum require EGFR kinase activity for Erk activation. This can proceed through indirect means, via receptor activated shedding of the EGFR ligands including HB-EGF [21] and amphiregulin [22] or by direct means, such as IGF-1R/EGFR complex formation [21]. IGF-1R/EGFR cross-talk has been described in normal human mammary epithelial cells [23], as well as for tamoxifenresistant MCF-7 cells where increased sensitivity to the proliferative effects of IGF-1/2 following estradiol or tamoxifen treatment were blocked by the IGF-1R tyrosine kinase inhibitor AG1024 or the anti-IGF-1R monoclonal antibody, a IR-3 [24]. In this report, we surveyed 12 matched tumor/ normal pairs of human HNSCC specimens for the presence of EGFR and IGF system components. All specimens expressed EGFRs and IGF-1Rs. Of the eight pairs demonstrating IGFBP-2 immunoreactivity, expression was lower in tumor versus normal in seven of the samples (Table 1 and Fig. 1). Examination of 8 established human HNSCC cell lines revealed IGF1R, IGFBP-2, IGFBP-3, and IGF-1/2 expression. IGF-1 stimulated S-phase transition in 5 out of 8 cell lines. IGF-1 stimulated thymidine incorporation could be inhibited by treatment with IGFBP-3, NVP-AEW541 or AG1478. The combination of IGFBP-3 with either receptor tyrosine kinase inhibitor resulted in a 10-fold greater inhibition than that obtained with either reagent alone, providing a rationale for inhibiting the IGF-1R in EGF-sensitive-HNSCC.
Table 1 Summary of IGF system component and EGFR expression in HNSCC tumor–normal matched pairs Sample #
1240
1256 1341 1285 1454 1548 1787 1957 1964 1967 1988
Normal Tumor Normal Tumor Normal Tumor Normal Tumor Normal Tumor Normal Tumor Normal Tumor Normal Tumor Normal Tumor Normal Tumor Normal Tumor Normal Tumor
Site
Grade/description
Stagea
Tongue Moderately differentiated
Pt2
pN0
pMx
Moderate to poor differentiation
pT4a
pN2
pMx
Moderately differentiated
pT2
pN2b
pMx
Well differentiated
pT2
pN0
pMx
Well differentiated
pT1
pN0
pMx
Moderate to poor differentiation
pT2
pNx
pMx
Moderately differentiated
pT2
pN3
pMx
Poorly differentiated
pT3
pN0
pMx
Well differentiated
pT2
pN2b
pMx
Moderately differentiated
pT2
pN0
pMx
Moderate to poor differentiation
pT2
pN2b
pMx
Moderately differentiated
pT3
pN0
pMx
Tongue Tongue Hypopharynx Floor of mouth Tongue Oropharynx Tongue Oropharynx Tongue Oral cavity (mandible) Tongue
EGFR
IGF1R
p p p p p p p p p p p p p p p p p p p p p p p p
p p p p p p p p p p p p p p p p p p p p p p p p
IGFBP 2
3
5
++ +++ ++ + ++ + ++ + ++ ++ ++ + ++ +
+ + + + ++ + + + + + + +++ ++ + + +++ ++ ++ +
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
1246
Condition
Normal and tumor samples (100 lg) were processed for EGFR, IGF-1R, and IGFBP immunoreactivity. For IGFBPs, +, minimal optical density (OD); ++, moderate OD; and +++, high OD. a Cancer staging was defined according to standard procedures (http://www.upmccancercenters.com/cancer/headneck/staging.html).
271
272
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
Fig. 1. IGFBP-2 expression in HNSCC tumor matched pairs. Normal and tumor samples (100 lg) were processed for IGFBP and b-actin immunoreactivity as described in Section 2.
2. Experimental procedures 2.1. Materials and reagents SCC-9, SCC-25, and FaDu cells were obtained from the ATCC (Manassas, VA). UM-SCC-1, -5, -12, -14A, and -22A were generous gifts from Dr. Thomas E. Carey (University of Michigan, Department of Otorhinolaryngology). Snap frozen matched pair HNSCC biopsy samples were obtained from the Hollings Cancer Center Tissue Bank with IRB approval. Fetal bovine serum (FBS) was purchased from Atlas Biologicals (Fort Collins, CO). DMEM and wortmannin were purchased from Sigma (St. Louis, MO). AG1478 was purchased from Calbiochem (La Jolla, CA). NVP-AEW541 was kindly provided by Novartis (Basel, Switzerland). IGF-1 was generously provided by Genentech, Inc. (San Francisco, CA). rhIGFBP-3 (N109D), IGFBP-2, -3, and -5 polyclonal antibodies, IGF-1 and IGF-2 mouse monoclonal antibodies, and EGFR clone LA22 antibody were from Upstate, Inc. (Charlottesville, VA). IGFBP-2 was expressed and purified from CHO cells as previously described [25]. IGF-1R b-subunit (H-60) antibody was obtained from Santa Cruz Biotechnologies, Inc. (Santa Cruz, CA). ActiveÒ IGF-1 and IGF-2 ELISA kits were purchased from Diagnostic Systems Laboratory Inc. (Webster, TX). Re-Blot Plus (mild and strong) Western blot stripping solutions were purchased from Chemicon (Temecula, CA). Western blot and immunohistochemistry specific phospho-Akt(Ser473) polyclonal antibodies and Akt antibody were obtained from Cell Signaling Technology (Beverly, MA). Erk antibody was obtained from BD Transduction Laboratories (San Jose, CA). NeutravidinÒ-horse radish peroxidase (neutravidin-HRP) and BCA reagent were obtained from Pierce (Rockford, IL). Phospho-44/42 MAPK antibody was purchased from New England Biolabs
(Beverly, MA). ECL Western blotting detection reagent, [125I]IGF-1, and [3H]thymidine were purchased from GE Healthcare Bio-Sciences Corp. (Piscataway, NJ). All other chemicals were of reagent grade or higher. TM
2.2. Tissue culture SCC [26], UM-SCC [27], and FaDu [28] cells were cultured as previously described and maintained at 37 °C in a humidified 5% CO2–95% air incubator. For receptor binding assays, cells were seeded in 24-well plates, grown overnight to 80% confluency, and serum-starved (FBS was eliminated in all experiments) for 24 h before the indicated treatment. For whole cell lysates, secretion, and cell cycle studies, cells were seeded in six-well plates. Following pretreatment for 2 h with fresh serum free medium containing inhibitors dissolved in ethanol, cells were treated with ethanol vehicle, inhibitors, IGF-1, IGFBP-2, or IGFBP-3. For the survey of IGF-1/2 and IGFBP secretion in Table 1, cells were seeded in 10 cm dishes to 80% confluency, plates were rinsed three times with PBS, and fresh serum free medium added. After 24 h, conditioned medium was dialyzed, concentrated, and processed for ELISA, ligand blot, and immunoblot analysis. 2.3. Receptor binding assays HNSCC cells serum starved for 24 h, were washed with HMS± (25 mM Hepes, 104 nM NaCl, 5 mM MgCl2, and 0.01% soybean trypsin inhibitor, containing 0.2% BSA. Triplicate wells were treated with increasing concentrations of IGF-1 (0–10 lM IGF-1) and 20,000 cpm of [125I]IGF-1 for 30 min at 37 °C. Cells were rinsed with HMS±, dissolved in 2 N NaOH, and radioactivity was quantified by gamma spectrometry using a Compugamma spectrometer (LKB-Wallac,
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
273
2.5. Cell cycle analysis
Turku, Finland). Assays were repeated a minimum of four times.
Following a 24 h treatment, HNSCC cells were harvested, fixed in 70% EtOH, stained with propidium iodide, and analyzed by flow cytometry on a FACSCalibur instrument (BD Biosciences, Franklin Lakes, NJ), in the Hollings Cancer Center Flow Cytometry Core Facility.
2.4. ELISA, immunoblot, and ligand blot analysis HNSCC matched pair samples were sonicated in lysis buffer and subjected to immunoblot analysis. Conditioned medium from HNSCC cell lines was analyzed by ELISA for IGF-1 and -2 according to the manufacturer’s instructions (DSL, Inc, Webster, TX). Whole cell lysates for Akt/Erk immunoblots were incubated for 15 min before being prepared using a modified RIPA buffer containing 50 mM Tris–HCl, pH 7.4, 1% Triton X-100, 150 mM NaCl, 10 mM EDTA, 1 mM PMSF, 10 lg/ml aprotinin and leupeptin, 2 mM sodium orthovanadate, and 10 mM NaF. Protein content was determined by BCA assay, and 100 lg aliquots were solubilized in SDS sample buffer. IGFBPs in conditioned medium were quantified by immunoblot after a 12 h incubation. Conditioned medium was precipitated in 10% trichloroacetic acid (TCA), the resultant pellet washed with acetone and solubilized in SDS sample buffer. Lysates and conditioned medium so collected were resolved on 10% and 12.5% non-reducing polyacrylamide gels, respectively, transferred to nitrocellulose and subjected to ligand (in the case of IGFBPs in Table 2) [29] and immunoblot analysis. Immunoblots were blocked, probed, and visualized as previously described [30], with either 1 lg/ml phospho-Akt (Ser 473) polyclonal, 1 lg/ml Akt polyclonal, 1:2000 phospho-p44/42 MAPK monoclonal, 1:5:000 Erk polyclonal,1:10,000 bactin monoclonal, or 1:1000 IGFBP-2, -3, or -5 polyclonal antibodies. To reprobe immunoblots, antibodies were removed from the nitrocellulose via the application of Re-Blot Plus-Mild stripping solution according to the manufacturers instructions (Amersham Biosciences, Piscataway, NJ).
2.6. [3H]Thymidine incorporation SCC-9 cells were plated in 48-well dishes and grown to 60–80% confluency in FBS supplemented growth medium. Wells were washed three times with PBS and replaced with SFM for 24 h. Cells were subsequently treated for 2 h with inhibitors followed by treatment for 21 h. Medium was then replaced and [3H]thymidine incorporation assay was performed as described [31]. 2.7. Statistical analysis Data are expressed as the mean between independent experiments ± standard error. Statistical differences in multiple experiments between multiple conditions were analyzed by ANOVA with a one way analysis of variance (95% confidence interval) and Bonferroni Post Test.
3. Results 3.1. The IGF-1 system in HNSCC matched pairs IGF system components and EGFR expression profiles of 12 HNSCC matched pairs are shown in Table 1. Immunoblot analysis revealed IGF-1R and EGFR expression in all samples, whereas IGFBP-3 immunoreactivity was detected in 10/12 normal and 9/12 tumor samples. Interestingly, of the eight pairs demonstrating IGFBP-2 immunoreactivity, expression was lower in
Table 2 Summary of IGF system component expression in HNSCC cell lines Cell line
SCC-9 SCC-25 FaDu UM-SCC-1 UM-SCC-5 UM-SCC-12 UM-SCC-14A UM-SCC-22A
Site
Tongue Tongue Hypopharynx Floor of mouth Supraglottis Larynx Floor of mouth Hypopharynx
IGF-a
IGF-1Rb
1
2
++ ++ ++ + + ++ ++ +
+++ +++ + ++ +++ ++
169,400 ± 14,504 346,900 ± 35,068 531,633 ± 40,130 285,366 ± 76,410 171,500 ± 11,107 40,000 ± 10,209 69,625 ± 5,112 29,500 ± 2,771
IGFBP 2 p p p p p p p p
3 p p p p p p p p
5 p p p p p
Following 24 h of serum starvation, conditioned medium from subconfluent HNSCC cell lines was processed for IGF-1/2 and IGFBP immunoreactivity, while cells were assayed for IGF-1R expression. (For IGFs +, 10 pg/100,000 cells; ++, 100 pg/100,000 cells; and +++, 1000 pg/100,000 cells). Values shown are representative of three or more independent experiments for each cell line. a IGF-1 and -2 quantified by specific ELISAs. b Receptor numbers based on 125I-IGF-1 binding studies and Scatchard analysis.
274
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
tumor versus normal in seven of the samples (Table 1 and Fig. 1). None of the pairs exhibited detectable IGFBP-5 immunoreactivity. Blots were not assayed for 1GFBP-1, -4, or -6. In addition, none of the pairs exhibited detectable IGF-1 or IGF-2 immunoreactivity (data not shown). 3.2. The IGF-1 system in HNSCC cell lines To further examine IGF system representation in squamous cell carcinoma, 8 established HNSCC cell lines were examined for IGF-1R, IGFBP, IGF-1, and IGF-2 expression (Table 2). IGF-1 binding analyses revealed that all eight cell lines expressed IGF-1Rs. Based on Scatchard analyses, six of these cell lines overexpressed the IGF-1R, based on a normal physiologic range of 10,000–30,000 receptors/cell [32]. IGF-1 and IGF-2 were detected in the conditioned medium from all cell lines, except for FaDu and SCC-5 cells. Ligand blot and immunoblot analyses revealed that all eight cell lines expressed IGFBP-2 and 3, whereas IGFBP-5 was only present in 5 out of 8 cell lines. 3.3. Effect of IGF-1 and IGFBPs on S-phase entry To determine whether the proliferative effects of IGF-1 correlated with receptor number, serum starved HNSCC cells were treated with IGF-1 (10 nM) for 24 h and harvested for cell cycle analysis. As shown in Fig. 2A, IGF-1 stimulated a significant (p < 0.05) increase in S-phase entry in 5 of the cell lines. The three cell lines showing greatest response, SCC-9, UMSCC-5, and UMSCC-22A, were used in subsequent analyses. To ascertain whether the IGFBPs could modulate IGF-1R activity, cells were treated with recombinant IGFBP-2 or IGFBP-3 and S-phase entry was evaluated (Fig. 2B). Addition of both IGFBP-2 and IGFBP-3 completely blocked IGF-1 stimulated S-phase entry, while having no effect on this parameter when added alone.
Fig. 2. Effect of IGF-1 and IGFBP-3 on S-phase entry in HNSCC cell lines. Subconfluent, serum-starved (24 h) HNSCC cells were treated for 24 h with 10 nM IGF-1 (A) or 1 nM IGF-1 and IGFBP-2 or IGFBP-3 (B). Cells were harvested, stained, and analyzed by flow cytometry. Error bars represent standard deviation between three individual analyses by flow cytometry. Significant differences were observed (*p < 0.05, compared to unstimulated). (A and B) Representative of three independent experiments.
3.4. Role of PI 3-kinase and Erk pathways in S-phase entry Immunoblot analysis of whole cell lysates revealed the presence of constitutively active Akt (phosphoAkt) in UMSCC-1, FaDu, and SCC-25 cells (Fig. 3A). Constitutively active Erk was detected in all but the UMSCC-22A cell line. IGF-1 addition increased Akt activity in all cell lines and Erk activity among all cell lines except the two that showed the highest basal pErk levels (SCC-25 and UMSCC-12). SCC-9, UMSCC-5, and UMSCC-22A cells were responsive to EGF stimulation; a representative analysis of SCC-9 cells showing robust Erk activation in response to EGF is shown in Fig. 3B (UMSCC-5 and UMSCC-22A data not shown). IGF-1 and EGF both stimulated [3H]thymidine incorporation. Coupled with our previous
findings in SCC-9 cells that demonstrate IGF-1 stimulated S-phase entry to be Erk-dependent [33], our current findings further support a role for Erk activation in HNSCC proliferation (Fig. 3C). Inhibition of EGFR tyrosine kinase activity with AG1478 completely abrogated IGF-1 stimulated Erk activity in SCC-9, (Fig. 4A), UMSCC-5 (Fig. 4B), and UMSCC-22A (Fig. 4C) cells. Significantly, AG1478 had no effect on IGF-1 stimulated Akt activity whereas NVP-AEW541 inhibited IGF-1 stimulated Akt and Erk activity without affecting basal Erk activity. We next tested the effect of these inhibitors on IGF-1 stimulated S-phase entry. AG1478 reduced the basal level of S-phase in unstimulated cells below control values
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
275
Fig. 3. IGF-1 signaling and EGFR activity in HNSCC cell lines. Subconfluent, serum-starved (24 h) HNSCC cells (A) and SCC-9 cells (B) were treated for 15 min as indicated. Whole cell lysates were processed for immunoblot analysis. (C) SCC-9 cells were plated in 48-well dishes and grown to 70% confluency. They were serum starved for 24 h, treated as indicated for 21 h, and processed for [3H]thymidine incorporation. Error bars represent standard deviation between three independent experiments. Significant differences were observed (*p < 0.05, compared to unstimulated). (A–C) Representative of three or more independent experiments.
(Fig. 4D). IGF-1 treatment significantly (p < .05) increased S-phase entry (Fig. 4D), which was inhibited by NVP-AEW541 treatment. NVP-AEW541 treatment reduced the basal level of S-phase entry to control (UMSCC-5) or below (SCC-9 and UMSCC-22A) control levels and completely abrogated IGF-1 stimulation. 3.5. Combined effects of IGFBP-3 and receptor tyrosine kinase inhibitor treatment Treatment of SCC-9 cells with 100 nM IGFBP-3 reduced IGF-1 (10 nM) stimulated [3H]thymidine incorporation by 57.3 ± 7.3% (where a 100% reduction would equal a final value of 0) (Fig. 5A and B). Treatment of IGF-1 stimulated cells with 0.1 or 1 lM NVP-AEW541, reduced [3H]thymidine incorporation by 60.9 ± 2.0% and 96.5 ± 0.5%, respectively; both values were significantly (p < 0.001) below basal levels (Fig. 5A). Co-addi-
Fig. 4. Effect of IGF-1R and EGFR kinase inhibitors on IGF-1 stimulated signaling pathways and S-phase entry in HNSCC cell lines. Subconfluent, serum-starved (24 h) SCC-9 (A), UMSCC-5 (B), and UMSCC-22A (C) cells were pretreated for 2 h with inhibitors as indicated before addition of 10 nM IGF-1 plus inhibitor for 15 min. Cells were harvested for immunoblot analysis (A–C). (D) SCC-9, UMSCC-5, and UMSCC-22A cells were plated in 48-well dishes and grown to 70% confluency. Following 24 h of serum starvation, they were pretreated with inhibitors for 2 h as indicated, before treatment for 21 h with inhibitors and 10 nM IGF-1. Cells were harvested, stained, and analyzed by flow cytometry. Error bars represent standard deviation between three independent experiments. Significant differences were observed (*p < 0.05, **p < 0.01, compared to unstimulated unless indicated by brackets). (A–D) Representative of three or more independent experiments.
276
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
Fig. 5. Effect of IGF-1R and EGFR tyrosine kinase inhibition and IGFBP-3 on IGF-1 stimulated [3H]thymidine incorporation. (A and B) SCC-9 cells plated in 48-well dishes were grown to 70% confluency and serum starved for 24 h. After a 2 h pretreatment with inhibitor, cells were treated as indicated for 21 h and processed for [3H]thymidine incorporation. Error bars represent standard deviation between three independent experiments. Significant differences were observed (*p < 0.05, **p < 0.01, ***p < 0.001, compared to unstimulated unless indicated by brackets; indicates an insignificant difference). (A and B) Representative of three or more independent experiments.
tion of 10 nM IGFBP-3 with 0.1 lM NVP-AEW541 did not further reduce [3H]thymidine incorporation (data not shown). Application of 0.01 lM NVP-AEW541 or 10 nM IGFBP-3 reduced IGF-1 stimulated [3H]thymidine incorporation by 2.8 ± 3.4% and 18.1 ± 4.3%, respectively. However, when they were combined we obtained an enhanced inhibitory action, resulting in a 41.5 ± 4.8% reduction in [3H]thymidine incorporation. This combined dosage was as effective as 100 nM IGFBP-3 in reducing [3H]thymidine incorporation. Similarly, 250 nM AG1478 reduced IGF-1 stimulated [3H]thymidine incorporation by 72.4 ± 2.3%, a value significantly (p < 0.001) below basal levels (Fig. 5B). However, AG1478 (25 nM) plus IGFBP-3 (10 nM), each of which individually reduced IGF-1 stimulated [3H]thymidine incorporation 31.8 ± 2.3% and 18.1 ± 4.3%, respectively, caused a com-
bined reduction of 61.4 ± 4.74%. This combination was as effective as 100 nM IGFBP-3 or 250 nM AG1478 alone, significantly reducing [3H]thymidine incorporation (p < 0.001) below basal levels.
4. Discussion Signaling by the IGF-1R is upregulated in many solid tumors, contributing to the malignant phenotype by enhancing resistance to EGF receptor and HER2 inhibitors, protection from apoptosis, increased tumor cell motility, invasiveness, and metastasis [34]. EGFR overexpression has been reported in a number of cancers including head and neck [2–5]. On this basis, the IGF-1R is a
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
promising therapeutic target for innovative molecularly targeted therapies in several malignancies. Considering high IGFBP-3 concentrations are inversely associated with the risk of some cancers [16–18], IGFBP-3 may provide an additional opportunity to inhibit the IGF/IGF-1R system. Not surprisingly, an array of in vitro, in vivo, and clinical findings point to IGFBP-3 as an anticancer molecule [35]. In a small survey of HNSCC specimens, we observed a lower expression of IGFBP-2 in tumor versus normal samples. This has the potential to contribute to autocrine/paracrine IGF-1/2 signaling in HNSCC, contributing to increased tumorigenicity. Although little is known about IGFBP-2 regulation, methylation of promoter CpG islands was recently reported for IGFBP-3, accounting for reduced IGFBP-3 levels in NSCLC [19]. Alternatively, reduced IGFBP-2 levels may reflect increased proteolysis by matrix metalloproteinases as reported for IGFBP-3 [36] and IGFBP-5 [37]. Analysis of eight human HNSCC cell lines revealed that all lines secrete IGFs, IGF-1Rs, and are responsive to IGF-1, supporting a role for autocrine IGF-1R action in HNSCC. IGFBP-2 and IGFBP-3 each abrogated IGF-1 stimulated HNSCC cell proliferation consistent with the IGF-1-dependent action of IGFBPs [9]. IGF-1 stimulated PI 3-K/Akt and/or Erk pathways in the 8 HNSCC cell lines, with the latter depending upon EGFR kinase activity. This is in agreement with studies on Ca9-22 cells where EGFR kinase activity was required for IGF-1 induced Erk activity [21]. The ability of the EGFR kinase inhibitor AG1478 to abrogate IGF-1 stimulated [3H]thymidine incorporation was consistent with our previous studies demonstrating that Erk activity was required for IGF1 stimulated SCC-9 cell S-phase entry [33]. The decrease in [3H]thymidine incorporation below control levels obtained by combining IGFBP-3 with AG1478 suggests that basal [3H]thymidine incorporation is, in part, the result of an autocrine IGF loop as seen in MCF-7 cells [25]. In addition, we have previously shown [33] that IGF-1R signaling in HNSCC cells induces VEGF signaling, which in turn acts in an autocrine stimulatory manner. Autocrine effects by VEGF or other RTKs may cause a basal level of cell activity which would be unaffected by IGFBP-3 addition but could be reduced by NVP-AEW541. The
277
nearly complete lack of [3H]thymidine incorporation observed with the IGF-1R kinase inhibitor NVP-AEW541 corroborates this view. NVP-AEW541 treatment in conjunction with IGFBP-3 led to enhanced attenuation of IGF-1 stimulated [3H]thymidine incorporation. The combination of AG1478 and IGFBP-3 caused a synergistic reduction in [3H]thymidine incorporation with 10-fold or lower concentrations of the two providing an inhibitory effect equivalent to either component alone at high dosages. In light of these findings, co-targeting of the IGF system and EGFR, employing IGF-1 antagonists and pharmacological inhibitors, appears promising. Such combination therapies offer specificity advantages inherent at lower concentrations of small molecule receptor tyrosine kinase inhibitors, as well as reduced side effects. Additionally, pharmacological inhibitors rarely exert 100% influence over their respective target. By directing antagonists to IGF-1, inherent limitations of any single inhibitor may be overcome. Similarly, when compared to antibody-based therapies, IGFBP-3 may provide an advantage by being a natural, small, diffusible protein having dual specificity against both IGF-1 and IGF-2. As such, our data support a role for IGFBP-3 as a therapeutically relevant molecule to be considered in the design of innovative therapeutic regimens in treating patients with HNSCC. Acknowledgements We wish to acknowledge the Hollings Cancer Center Tissue Bank for human specimens and the Hollings Cancer Center Flow Cytometry Core Facility for cell cycle analysis. References [1] M.T. Canto, S.S. Devesa, Oral cavity and pharynx cancer incidence rates in the United States, Oral. Oncol. 38 (2002) 610–617. [2] D.S. Solomon, R. Brandt, F. Ciardiello, N. Normanno, Epidermal growth factor-related peptides and their receptors in human malignancies, Crit. Rev. Oncol. Hematol. 19 (1995) 183–232. [3] D.E. Davies, S.G. Chamberlin, Targeting the epidermal growth factor receptor for therapy of carcinomas, Biochem. Pharmacol. 51 (1996) 1101–1110. [4] V. Rusch, J. Mendelsohn, E. Dmitrovsky, The epidermal growth factor receptor and its ligands as therapeutic targets in human tumors, Cytokine Growth Factor Rev. 7 (1996) 133–141.
278
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279
[5] J.R. Grandis, A. Chakraborty, Q. Zheng, M.F. Melhem, D.J. Tweardy, Downmodulation of TGF-alpha protein expression with antisense oligonucleotides inhibits proliferation of head and neck squamous carcinoma but not normal mucosal epithelial cells, J. Cell Biochem. 69 (1998) 55–62. [6] L.J. Wirth, R.J. Haddad, N.I. Lindeman, X. Zhao, J.C. Lee, V.A. Joshi, C.M. Norris Jr., M.R. Posner, Phase I study of gefitinib plus celecoxib in recurrent or metastatic squamous cell carcinoma of the head and neck, J. Clin. Oncol. 23 (2005) 6976–6981. [7] J. Baselga, Monoclonal antibodies directed at growth factor receptors, Ann. Oncol. 11 (Suppl. 3) (2000) 187–190. [8] K. Garber, Tyrosine kinase inhibitor research presses on despite halted clinical trial, J. Natl. Cancer Inst. 92 (2000) 967–969. [9] S.A. Rosenzweig, What’s new in the IGF-binding proteins?, Growth Horm IGF Res. 14 (2004) 329–336. [10] S.M. Firth, R.C. Baxter, Cellular actions of the insulin-like growth factor binding proteins, Endocr. Rev. 23 (2002) 824– 854. [11] X. Wu, H. Zhao, K-A. Do, M.M. Johnson, Q. Dong, W.K. Hong, M.R. Spitz, Serum levels of insulin growth factor (IGF-1) and IGF-binding protein predict risk of second primary tumors in patients with head and neck cancer, Clin. Cancer Res. 10 (2004) 3988–3995. [12] S.E. Hankinson, W.C. Willet, G.A. Colditz, D.J. Hunter, D.S. Michaud, B. Deroo, B. Rosner, F.E. Speizer, M. Pollak, Circulating concentrations of insulin-like growth factor-1 and risk of breast cancer, Lancet 351 (1998) 1393–1396. [13] J.M. Chan, M.J. Stampfer, E. Giovannucci, P.H. Gann, J. Ma, P. Wilkinson, C.H. Hennekens, M. Pollak, Plasma insulin-like growth factor-I and prostate cancer risk: a prospective study, Science 279 (1998) 563–566. [14] H. Yu, M.R. Spitz, J. Mistry, J. Gu, W.K. Hong, X. Wu, Plasma levels of insulin-like growth factor-1 and lung cancer risk: a case-control study, 91 (1999) 151–156. [15] J. Ma, M.N. Pollak, E. Giovannucci, J.M. Chan, Y. Tao, C.H. Hennekens, M.J. Stampfer, Prospective study of colorectal cancer risk in men and plasma levels of insulinlike growth factor (IGF)-1 and IGF-binding protein-3, J. Natl. Cancer Inst. 91 (1999) 620–625. [16] V.A. Papadimitrakopoulou, E.N. Brown, D.D. Liu, A.K. El-Naggar, J.J. Lee, W.K. Hong, H-Y. Lee, The prognostic role of loss of insulin-like growth factor-binding protein-3 expression in head and neck carcinogenesis, Cancer Let. (2005). [17] R.G. Rosenfeld, G. Lamson, H. Pham, Y. Oh, C. Conover, D.D. De Leon, S.M. Donovan, I. Ocrant, L. Giudice, Insulin-like growth factor-binding proteins, Recent Prog. Horm. Res. 46 (1990) 99–159, discussion 159–163. [18] R.F. Krywichi, D. Yee, The insulin-like growth factor family of ligands, receptors, and binding proteins, Breast Cancer Res. Treat. 22 (1992) 7–19. [19] Y.S. Chang, L. Wang, Y-A. Suh, L. Mao, S.J. Karpen, F.R. Khuri, W.K. Hong, H-Y. Lee, Mechanisms underlying lack of insulin-like growth factor-binding protein-3 expression in non-small-cell lung cancer, Oncogene 23 (2004) 6569–6580. [20] C. Garcia-Echeverria, M.A. Pearson, A. Marti, T. Meyer, J. Mestan, J. Zimmermann, J. Gao, J. Brueggen, H.G. Capraro, R. Cozens, D.B. Evans, D. Fabbro, P. Furet, D.G. Porta, J. Liebetanz, G. Martiny-Baron, S. Ruetz, F.
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28] [29]
[30]
[31]
[32]
[33]
[34]
Hofmann, In vivo antitumor activity of NVP-AEW541- A novel, potent, and selective inhibitor of the IGF-IR kinase, Cancer Cell. 5 (2004) 231–239. A. Kuribayashi, K. Kataoka, T. Kurabayashi, M. Miura, Evidence that basal activity, but not transactivation, of the epidermal growth factor receptor tyrosine kinase is required for insulin-like growth factor 1-induced activation of extracellular signal-regulated kinase in oral carcinoma cells, Endocrinology 145 (2004) 4976–4984. Q. Zhang, S.M. Thomas, V.W.Y. Lui, S. Xi, J.M. Siegfried, H. Fan, T.E. Smithgall, G.B. Mills, J.R. Grandis, Phosphorylation of TNF-{alpha} converting enzyme by gastrinreleasing peptide induces amphiregulin release and EGF receptor activation, Proc. Natl. Acad. Sci. USA 103 (2006) 6901–6906. T. Ahmad, G. Farnie, N.J. Bundred, N.G. Anderson, The mitogenic action of insulin-like growth factor I in normal human mammary epithelial cells requires the epidermal growth factor receptor tyrosine kinase, J. Biol. Chem. 279 (2004) 1713–1719. J.M. Knowlden, I.R. Hutcheson, D. Barrow, J.M.W. Gee, R.I. Nicholson, Insulin-like growth factor-1 receptor signaling in tamoxifen-resistant breast cancer: A supporting role to the epidermal growth factor receptor, Endocrinology 146 (2005) 4609–4618. M.M. Kibbey, M.J. Jameson, E.M. Eaton, S.A. Rosenzweig, Insulin-like growth factor binding protein-2: contributions of the C-terminal domain to insulin-like growth factor-1 binding, Mol. Pharmacol. 69 (2006) 833–845. J.G. Rheinwald, M.A. Beckett, Tumorigenic keratinocyte lines requiring anchorage and fibroblast support cultured from human squamous cell carcinomas, Cancer Res. 41 (1981) 1657–1663. T.E. Carey, K.A. Kimmel, D.R. Schwartz, D.E. Richter, S.R. Baker, C.J. Krause, Antibodies to human squamous cell carcinoma, Otolaryngol. Head Neck Surg. 91 (1983) 482– 491. S.R. Rangan, A new human cell line (FaDu) from a hypopharyngeal carcinoma, Cancer 29 (1972) 117–121. S.A. Robinson, S.A. Rosenzweig, Synthesis and characterization of biotinylated forms of insulin-like growth factor-1: topographical evaluation of the IGF-1/IGFBP-2 and IGFBP-3 interface, Biochemistry 43 (2004) 11533–11545. M.G. Slomiany, S.A. Rosenzweig, IGF-1 induced VEGF and IGFBP-3 secretion correlates with increased Hif-1 alpha expression and activity in retinal pigment epithelial cell line D407, Invest. Ophthalmol. Vis. Sci. 45 (2004) 2838–2847. S.A. Robinson, S.A. Rosenzweig, Paradoxical effects of the phage display-derived peptide antagonist IGF-F1-1 on insulin-like growth factor-1 receptor signaling, Biochem. Pharmacol. 72 (2006) 53–61. M.A. Steller, C.H. Delgado, C.J. Bartels, C.D. Woodworth, Z. Zou, Overexpression of the insulin-like growth factor-1 receptor and autocrine stimulation in human cervical cancer cells, Cancer Res. 56 (1961) 1761–1765. M.G. Slomiany, L.A. Black, M.M. Kibbey, T.A. Day, S.A. Rosenzweig, IGF-1 induced vascular endothelial growth factor secretion in head and neck squamous cell carcinoma, Biochem. Biophys. Res. Commun. 342 (2006) 851–858. S. Benini, M. Zuntini, M.C. Manara, P. Cohen, G. Nicoletti, P. Nanni, Y. Oh, P. Picci, K. Scotlandi, Insulin-like growth
M.G. Slomiany et al. / Cancer Letters 248 (2007) 269–279 factor binding protein 3 as an anticancer molecule in Ewing’s Sarcoma, Int. J. Cancer. 119 (2006) 1046–1049. [35] O. Ali, P. Cohen, K.W. Lee, Epidemiology and biology of insulin-like growth factor binding protein-3 (IGFBP-3) as an anti-cancer molecule, Horm. Metab. Res. 35 (2003) 726–733. [36] S.I. Miyamoto, K. Yano, S. Sugimoto, G. Ishii, T. Hasebe, Y. Endoh, K. Kodama, M. Goya, T. Chiba, A. Ochiai, Matrix metalloproteinase-7 facilitates insulin-like growth
279
factor bioavailability through Its proteinase activity on insulin-like growth factor binding protein 3, Cancer Res. 64 (2004) 665–671. [37] E. Hemers, C. Duval, C. McCaig, M. Handley, G.J. Dockray, A. Varro, Insulin-like growth factor binding protein-5 Is a target of matrix metalloproteinase-7: Implications for epithelial-mesenchymal signaling, Cancer Res. 65 (2005) 7363–7369.