|Home | About | Journals | Submit | Contact Us | Français|
We have shown previously that exposure to anticancer drugs can trigger the activation of human epidermal receptor (HER) survival pathways in colorectal cancer (CRC). In this study, we examined the role of ADAMs (a desintegrin and metalloproteases) and soluble growth factors in this acute drug resistance mechanism.
In vitro and in vivo models of CRC were assessed. ADAM-17 activity was measured using a fluorometric assay. Ligand shedding was assessed by ELISA or Western blotting. Apoptosis was assessed by flow cytometry and Western blotting.
Chemotherapy (5-Fluorouracil, 5-FU) treatment resulted in acute increases in TGF-α-, amphiregulin- and heregulin-ligand shedding in vitro and in vivo that correlated with significantly increased ADAM-17 activity. siRNA-mediated silencing and pharmacological inhibition confirmed that ADAM-17 was the principal ADAM involved in this pro-survival response. Furthermore, overexpression of ADAM-17 significantly decreased the effect of chemotherapy on tumour growth and apoptosis. Mechanistically, we found that ADAM-17 not only regulated phosphorylation of HERs, but also increased the activity of a number of other growth factor receptors, such as IGF-1R and VEGFR.
Chemotherapy acutely activates ADAM-17 which results in growth factor shedding, growth factor receptor activation and drug resistance in CRC tumours. Thus, pharmacological inhibition of ADAM-17 in conjunction with chemotherapy may have therapeutic potential for the treatment of CRC.
Resistance to chemotherapy is a major barrier in the treatment of cancer. Recent studies including our own have shown that exposure to anticancer drugs or ionizing radiation can activate stress pathways, which trigger activation of multiple signalling pathways, such as those regulated by the Human Epidermal Receptor (HER) tyrosine kinase family (1-5).
The HER family of receptor tyrosine kinases and their ligands are important regulators of tumour cell proliferation, angiogenesis and metastasis (6-8). There are four receptors in the ErbB family: EGFR (HER1 or ErbB1), HER2 (neu or ErbB2), HER3 (ErbB3) and HER4 (ErB4). HER1 is activated by binding of the HER1-specific ligands: epidermal growth factor, EGF; transforming growth factor-α, TGF-α; amphiregulin, AREG or ligands with dual specificity (heparin-binding EGF, HB-EGF; betacellulin; epiregulin, EREG) to the ectodomain of HER1 (9, 10). HER1 and its ligand TGF-α constitute one of the best defined autocrine loops in human tumours (6, 11), and their co-expression correlates with aggressive disease and poor prognosis in several types of tumours, including colorectal cancer (CRC). Recently, high AREG and EREG mRNA expression levels in Kras wild type colorectal primaries have been correlated with response and survival benefit following treatment with cetuximab and irinotecan in advanced CRC (12).
HER ligands are synthesized as transmembrane precursors that can be cleaved by cell surface proteases, in particular members of the ADAM (a desintegrin and metalloprotease) family. ADAM-mediated ligand shedding results in enhanced juxtacrine and paracrine signalling (13). ADAMs are synthesized as inactive precursors containing a prodomain that blocks the activity of the catalytic domain. During transit through the secretory pathway, the prodomain of ADAMs is removed by furin-like proprotein convertases (14). Several studies have demonstrated that members of the ADAM family, such as ADAM-9, ADAM-10, ADAM-12, ADAM-15 and ADAM-17 may be involved in regulating HER1 activation via proteolytic processing of HER1-ligand precursors (1, 3, 15, 16) and that these metalloprotease-dependent mechanisms can be activated by cellular stress (1). Of these, ADAM-17 has been suggested to be the major HER ligand ‘sheddase’. Studies by various groups have shown that ADAM-17 deficiency abrogates the shedding of TGF-α, HB-EGF, epiregulin, amphiregulin and heregulin (17-22).
The aims of this study were to investigate whether exposure to chemotherapy treatment results in increased HER ligand shedding and whether this survival response was associated with resistance to chemotherapy treatment. We have also investigated the mechanism by which chemotherapy triggers HER ligand shedding, in particular the role of ADAM proteases in regulating this survival response.
All chemicals and reagents of Analar grade were obtained from BDH Laboratory Supplies (Poole, BH15 1TD, England) unless otherwise stated. GI254023X and GW280264X were provided by GlaxoSmithKline (GSK, Stevenage, Herts, SG1 2NY, UK). A 10mM working solution of GI254023X and GW280264X in DMSO was prepared, aliquotted and stored at −70°C. Oxaliplatin was obtained from Sanofi-Synthelabo (Surrey, UK). A 1mM stock solution was prepared in injection water and stored at 4°C. 5-Fluorouracil (5-FU) was purchased from Sigma Chemical Co. (St. Louis, MO). A 10mM stock solution in 1×PBS was prepared and stored at 4°C. SN-38 was obtained from Abatra (Xi'an, China) and a 2mM solution was prepared in DMSO and stored at 4°C.
All tissue culture material was obtained from Invitrogen (Paisley, Scotland), unless otherwise stated. HCT116 and HCT116-p53null CRC cells were kindly provided by Bert Vogelstein (Johns Hopkin University, Baltimore, MD) and maintained in McCoy's 5A medium. LoVo CRC cells, supplied by AstraZeneca, were grown in Dulbecco's Modified Eagle Medium (DMEM). The RKO and H630 CRC cells were provided by the National Cancer Institute (Bethesda, MD) and maintained in DMEM. All medium was supplemented with 10% dialysed foetal calf serum (FCS), 50μg/mL penicillin-streptomycin (P/S), 2mM L-glutamine and 1mM sodium pyruvate (Invitrogen). All cells were grown in a humidified atmosphere with 5% CO2 at 37°C.
Flow cytometry was performed as previously described (5).
Cell pellets were resuspended in 100μL 1× binding buffer. 5μL Annexin V stain was added to each sample along with 5μL PI stain (50μg/mL – 1:20 dilution in PBS of stock), and samples were incubated in the dark at room temperature for 15 minutes. After incubation, 320μl 1× binding buffer was added to each sample prior to analysis on the EPICS XL Flow Cytometer.
Western blot analysis was carried out as previously described (5). Immunodetections were performed using anti-EGFR (Clone 13, Pharmingen, BD Biosciences, San Jose, CA), anti-heregulin (R&D systems, Abingdon UK), anti-amphiregulin (R&D systems) and anti-IGF-1 (Santa Cruz Biotechnology, Santa Cruz, CA) mouse monoclonal antibodies in conjunction with a horseradish peroxidase (HRP)-conjugated anti-mouse secondary antibody (Amersham). Anti-phospho-EGFR (Tyr1068, Calbiochem), anti-ADAM-17 (Pharmigen, BD Biosciences), anti-phospho-IGF-1R (Calbiochem), anti-phospho-VEGFR1 (Calbiochem), anti-phospho-VEGFR2/3 (Calbiochem) and anti-PDGFRβ (Calbiochem) rabbit polyclonal antibodies were used in conjunction with an HRP-conjugated anti-rabbit secondary antibody (Amersham). Equal loading was assessed using β-tubulin (Sigma) mouse monoclonal primary antibodies. The Super Signal® chemiluminescent system (Pierce, Rockford, IL) or ECL-plus® (Amersham) were used for detection.
ADAM-9, -10, -12, -15, -17 and SC (scramble control) siRNAs were obtained from Dharmacon (Lafayette, co). HCT116, HCT116-p53null, LoVo, RKO and H630 CRC cells were seeded out in the appropriate media, without P/S. Twenty-four hours after seeding, siRNA transfections were performed on sub-confluent cells incubated in unsupplemented Optimem using the oligofectamine reagent (both from Invitrogen) according to the manufacturer's instructions. After 4 hours, 3×growth medium was added; cells were treated with oxaliplatin, 5-FU or SN-38 1 hour following this.
The plasmid encoding the hemagglutinin (HA)-tagged full length mouse ADAM-17 (HA-ADAM-17) was obtained as a kind gift from Dr. Atanasio Pandiella (Instituto de Microbiología Bioquímica and Centro de Investigación del Cáncer) and has been previously described (23). The pcDNA 3.1 empty vector (EV) was purchased from Invitrogen. HCT116 cells were co-transfected with 10μg of HA-ADAM-17 construct or pcDNA 3.1 (empty vector) and 1μg of a construct expressing a puromycin resistance gene (pIRESpuro3-Clontech) using GeneJuice transfection reagent (Novagen). Stably transfected cells were selected and maintained in medium supplemented with 1μg/ml puromycin (Life Technologies Inc.).
Female BALB/c severe combined immunodeficient (SCID) mice were maintained under sterile and controlled environmental conditions (22°C, 50 ± 10% relative humidity, 12h/12h light/dark cycle, autoclaved bedding), with food and water ad libidum. Following 14 days of quarantine, mice were included in our protocol. The experiment was carried out in accordance with the Animals (Scientific Procedures) Act, 1986. To determine tumour volume, two axes of the tumours were measured using digital Vernier callipers. Tumour volumes were calculated using the following formula: (longest tumour diameter) × (shortest tumour diameter)2 × 0.5. HCT116 xenografts mouse models were established by sc. inoculation of 2 × 106 HCT116 cells into the right and left flanks using Matrigel (BD Biosciences). Tumours were allowed to grow until they reached approximately 200mm3 (day 14) at which point the first group received placebo (PBS) and the second group 75mg/kg 5-FU by intraperitoneal injection. Twenty-four hours after treatment, six mice were sacrificed in each of the two treatment groups. Whole blood was collected from the axillary vessels and tumours were harvested and snap frozen in liquid nitrogen. For analysis of the effect of ADAM-17 overexpression, HA-ADAM-17 or pcDNA 3.1 cells were sc. implanted using Matrigel. Tumours were allowed to grow until they reached size of approximately 50-100mm3 (day 8) at which point the first group received placebo (PBS) and the second group received chemotherapy [(5-FU; 15mg/kg, day 8-12 and day 15-19) and (oxaliplatin, 2mg/kg, days 8 and 15)] and each treatment group contained eight animals. The tumours were measured 3 times a week in two dimensions using a calliper. The statistical significance was analysed using the unpaired two-tailed Student's t test.
An equal number of cells were plated into 24 well plates and incubated for 24 hours. Cells were treated in each experiment for a particular period. Conditioned medium or serum was collected in vitro and in vivo respectively and TGF-α or VEGF levels were analysed according to the ELISA kit instructions (Calbiochem).
Excised tumours from HCT116 xenografts were homogenized in RIPA buffer using an IKA labortechnik homogenizer. After centrifugation of tissue homogenates, the supernatants were transferred to a new tube and protein concentration was determined. The InnoZyme TACE Activity Kit (Calbiochem) was used to measure ADAM-17 (TACE) activity in 500μg of each protein sample. Pure ADAM-17 was used as a positive control (+TACE).
Activity of a panel of receptor tyrosine kinases (RTK) was detected using an antibody-based array from R&D systems. Antibodies against 42 different RTKs were prespotted in duplicate on nitrocellulose membranes, and cell lysates from empty vector control cells and ADAM-17 overexpressing cells were incubated with the membrane. Thereafter, a pan-anti-phosphotyrosine antibody was used to detect the phosphorylated tyrosine on activated RTKs.
Two-way ANOVA test was used to determine the significance of change in levels of apoptosis between different treatment groups. All changes in levels of apoptosis that are described as significant had p values that were <0.05 (* denotes p<0.05; ** denotes p<0.01; *** denotes p<0.001). The nature of the interaction between GW280264X and chemotherapy was determined by calculating the combination index (CI) according to the method of Chou and Talalay (24). CI values of <0.3, 0.3 < CI < 0.7, 0.7 < CI < 0.85, 0.85 < CI < 1, 1, and >1 indicate very strong synergism, strong synergism, moderate synergism, slight synergism, an additive interaction, and antagonism respectively. CI values were calculated from isobolograms generated using the CalcuSyn software program (Microsoft Windows).
In view of our previous data showing that cytotoxic chemotherapy activates an EGFR-mediated survival response in CRC cells (5), we examined whether 5-FU treatment induced the release of the prototypical EGFR ligand TGF-α in culture medium of CRC cells. 5-FU-based regimens (FOLFOX: 5FU/LV + oxaliplatin; FOLFIRI: 5FU/LV + irinotecan (CPT-11)) represent the cornerstone of treatment of patients with advanced CRC. Twenty-four hours following treatment with 5-FU, we found a dose-dependent increase in TGF-α ligand shedding in HCT116 cells (Fig. 1A upper panel). In addition, shedding of other HER ligands such as the EGFR ligand amphiregulin (sAREG) and the HER3 ligand heregulin (sHeregulin) also increased significantly following treatment with 5-FU in HCT116 cell line (Fig. 1A lower panel). Importantly, these effects were also observed in vivo, with statistically significant acute increases in human TGF-α shedding into the circulation observed following treatment of HCT116 CRC xenografts-bearing mice with 5-FU for 24 hours (Fig. 1B upper panel). In addition, the serum levels of amphiregulin and heregulin were also increased in the 5-FU treatment group compared to control (Fig. 1B lower panel). The relevance of HER ligand shedding for drug resistance was demonstrated by co-treating HCT116 cells with recombinant TGF-α, AREG, EGF or heregulin and 5-FU. Treatment with each HER ligand significantly protected cells from 5-FU-induced apoptosis and apoptosis induced by oxaliplatin and the active metabolite of irinotecan, SN-38 (Fig. 1C left and right panel and Supplementary Fig. S1). Taken together, these data suggest that induction of TGF-α, amphiregulin and heregulin ligand shedding is an acute mechanism of drug resistance in colorectal cancer cells.
One mechanism by which EGFR can be activated is via ADAM-mediated ligand shedding. Treatment with 5-FU had no effect on TGF-α mRNA or AREG mRNA expression and resulted only in a moderate ~ 2-fold increase in heregulin mRNA expression level (Supplementary Fig. S2). These results indicate that increased shedding of TGF-α, AREG and heregulin following 5-FU treatment are predominantly due to posttranslational mechanisms. With this in mind, we investigated whether a number of different ADAM family members play a significant role in regulating chemotherapy-induced EGFR activation and TGF-α shedding by using gene specific siRNAs directed against ADAM-9, -10, -12, -15 and -17 (Fig. 2A and 2B). Using quantitative real-time PCR, we found a decrease of ~70-80% in ADAM-9, -10, -12, -15 and -17 mRNA expression following transfection with each siRNA in the LoVo and HCT116 cell lines (Supplementary Fig. S3). Specific down-regulation of ADAM-17 protein was also confirmed by Western blotting (Fig. 2A). Interestingly, we found that the increased TGF-α shedding and EGFR activity following 5-FU treatment was only abrogated by ADAM-17 gene silencing, while silencing of the other ADAMs (ADAM-9, -10, -12 and -15) did not significantly affect 5-FU-induced TGF-α shedding and EGFR activation (Fig. 2A upper panel and 2B). We next directly investigated the effect of chemotherapy treatment on ADAM-17 activity. We found that treatment with 5-FU significantly increased ADAM-17 activity in HCT116 cells (Fig. 2C). Importantly, this was also observed in vivo, with a significant increase in ADAM-17 activity observed in HCT116 xenografts exposed to 75mg/kg 5-FU for 24 hours (Fig. 2D). These results correlated strikingly with the effect of chemotherapy treatment on HER ligand shedding in vitro and in vivo (Fig. 1).
In order to rule out the possibility that the effects of ADAM-17 were specific to the Kras mutant and p53 wild type HCT116 and LoVo models, we extended these studies into three further CRC cell line models: a p53 null HCT116 daughter cell line, the Braf mutant and p53 wild type RKO cell line, and the Kras wild type/Braf wild type and p53 mutant H630 cell line. We found that silencing of ADAM-17 attenuated 5-FU-induced EGFR-activation in all three of these cell lines, indicating that these effects are not p53-, Kras- or Braf dependent (Fig. 3A). Importantly, we found that shedding of TGF-α following 5-FU treatment was ADAM-17-dependent in HCT116, LoVo, RKO and H630 cell line using ADAM-17 siRNA (Fig. 3B). In addition, 5-FU-induced AREG and heregulin shedding was completely abrogated when ADAM-17 was silenced in HCT116, LoVo and H630 cell lines (Fig. 3C).
We also determined the role of ADAM-17 in regulating an EGFR-mediated survival response following SN-38 and oxaliplatin (Supplementary Fig. S4). We found that silencing of ADAM-17 inhibited SN-38-induced TGF-α shedding and EGFR activation in HCT116, LoVo, RKO and H630 cell lines (Supplementary Fig. S4 upper and lower panels). Similar data were obtained with oxaliplatin. Consistent with our earlier studies (5), oxaliplatin failed to increase EGFR activity in HCT116 cell line, and no significant increase in TGF-α shedding following oxaliplatin was observed in this cell line.
Taken together, these results indicate that colorectal cancer cells respond to chemotherapy by increasing ADAM-17 activity, which further regulates HER-ligand shedding and activation of EGFR.
We next investigated the effect of ADAM-17 inhibition on chemotherapy-induced apoptosis in the CRC cell lines models, using ADAM-17 gene silencing. A significant supra-additive/synergistic increase in apoptosis was observed in ADAM-17 silenced CRC cells co-treated with 5-FU (Fig. 4A). Similar results were obtained when ADAM-17 siRNA was combined with SN-38 or oxaliplatin in this cell line panel (Supplementary Fig. S5A).
We next assessed the effect of pharmacological inhibition of ADAM-17 using a small molecule dual ADAM-10/17 inhibitor, GW280264X, and compared this with a specific ADAM-10 inhibitor, GI254023X. Physiologically relevant doses that have been used in previous publications were used (25). We found that GW280264X inhibited 5-FU-induced ADAM-17 activity, TGF-α ligand shedding and EGFR activity in both HCT116 and LoVo cells (Fig. 4B and 4C). In contrast, the ADAM-10 specific inhibitor GI254023X had only a marginal effect on 5-FU-induced TGF-α shedding and EGFR activation (Supplementary Fig. S6). In addition, when GW280264X was combined with 5-FU, SN-38 or oxaliplatin in HCT116 and LoVo cells, significant supra-additive/synergistic increases in apoptosis were observed (Fig. 4D and Supplementary Fig. S5B and S5C). In contrast, GI254023X had no effect on chemotherapy-induced apoptosis (Fig. 4D and Supplementary Fig. S5B and S5C). We next determined the level of synergy between GW280264X and chemotherapy using the method of Chou and Talalay to calculate combination index (CI) values. For the combination of GW280264X with 5-FU and SN-38, CI values below 0.7 were observed for majority of concentrations, indicative for strong synergism in HCT116 and LoVo cells (Supplementary Fig. S7A and S7B). The combination of GW280264X with oxaliplatin was also synergistic for the majority of drug concentrations with most CI values between 0.5 and 1.0 (Supplementary Fig. S7A and S7B). These results further highlight the importance of ADAM-17 as a key mediator of resistance to chemotherapy, and strongly suggest that ADAM-17-targeted agents may be novel drugs for use in conjunction with existing chemotherapy regimens in patients with colorectal cancer.
To complement our gene silencing and inhibitor studies, we developed HCT116 cell line models that stably overexpress HA-tagged ADAM-17. Two stable clones with different levels of ADAM-17 overexpression were identified: HA-ADAM17 3 (AD3) and HA-ADAM17 4 (AD4) (Fig. 5A). ADAM-17 activity in clones AD3 and AD4 were approximately 2.5- and 35-fold higher compared to empty vector (EV) (Fig. 5A right upper panel). Importantly, TGF-α-, AREG and heregulin shedding (Fig. 5A left panel and right lower panel), EGFR and HER3 activation (Fig. 5A) in the ADAM-17 overexpressing clones were significantly increased compared to the empty vector (EV) cell line, with the more highly ADAM-17 overexpressing clone (AD4) exhibiting greater TGF-α shedding and EGFR activation. In vitro, ADAM-17 overexpression protected cells from 5-FU-induced apoptosis (Fig. 5B and Supplementary Fig. S8B) and also apoptosis induced by oxaliplatin and SN-38 (Supplementary Fig. S8A and S8B).
We next determined the effect of increased expression of ADAM-17 on the growth of human HCT116 xenografts and evaluated their response to combined treatment with 5-FU and oxaliplatin (Fig. 5C). The growth of ADAM-17 overexpressing CRC xenografts was more rapid than control xenografts. Importantly, overexpression of active ADAM-17 in vivo attenuated the antitumor activity of 5-FU/oxaliplatin combination treatment (Fig. 5C). Western blot analysis confirmed that the ADAM-17 overexpressing xenografts expressed HA-tagged ADAM-17 (Fig. 5D left panel). Furthermore, ADAM-17 activity levels were significantly higher (~ 4-fold) in ADAM-17 overexpressing tumours, and this was associated with increased EGFR phosphorylation in these tumours compared with the EV xenografts (Fig. 5D). Consistent with our previous findings, we found that ADAM-17 activity levels increased significantly in EV controls and ADAM-17-overexpressing HCT116 xenografts following treatment with 5-FU and oxaliplatin (Fig. 5D right panel). Moreover, chemotherapy-induced ADAM-17 activity levels was associated with increased EGFR phosphorylation (Fig. 5D left panel). In addition, chemotherapy-induced caspase-3 activation was abrogated in the ADAM-17 overexpressing CRC xenografts compared to EV xenografts. Collectively, these data indicate that ADAM-17 activity regulates the sensitivity of colorectal cancer tumours to standard chemotherapy treatment, further indicating that combining ADAM-17 inhibitors with chemotherapy could be a potential novel strategy for the treatment of colorectal cancer.
In order to further investigate the mechanisms of ADAM-17-mediated resistance to chemotherapy treatment, we assessed the phosphorylation status of 42 receptor tyrosine kinases in HCT116 cells, 24 hours following transient transfection with ADAM-17, using a human phospho-receptor tyrosine kinase array kit (Fig. 6A). In addition to EGFR, we found increased phosphorylation levels of VEGFR2/3, IGF-1R, PDGFR, the Ephrin receptors and developmental tyrosine kinase (DTK) in ADAM-17 overexpressing HCT116 cells. In contrast, phosphorylation levels of hepatocyte growth factor receptor (pHGFR) was reduced following transient transfection with ADAM-17 (Fig. 6A and Supplementary Fig. S9). The VEGFRs, PDGFR and IGF-1R are key regulators of colorectal tumour angiogenesis, lymphangiogenesis, tumour growth and proliferation and have emerged as important targets in colorectal cancer (26-28). So, we next validated our array results for these receptors by Western blotting analysis, using phospho-specific VEGFR1, VEGFR2/3, IGF-1R and PDGFR antibodies that reflect the activation state of the receptors. We found that VEGFR2/3, VEGFR1 and IGF-1R activity was significantly increased in the stable ADAM-17 overexpressing cell lines AD3 and AD4 compared to the EV cell line (Fig. 6B). Activity of PDGFRβ was only slightly increased in AD clones compared to the EV cell line. Moreover, we found that shedding of the VEGF and IGF-1 ligands was significantly increased in cells stably over-expressing ADAM-17 (Fig. 6C). We subsequently assessed whether VEGF and IGF-1 shedding occurred following chemotherapy treatment and if this was also regulated by ADAM-17. We found that 5-FU treatment resulted in increased VEGF and IGF-I ligand shedding and that this was abrogated in the presence of the ADAM10/17 inhibitor GW280264X (Fig. 6D). Taken together, these results indicate that through its ability to shed multiple growth factors, ADAM17 is an important regulator of several key survival pathways following treatment of colorectal cancer cells with chemotherapy.
We previously reported that colorectal cancer (CRC) cells respond acutely to chemotherapy by activating a human epidermal receptor (HER)-mediated survival response and are thereby sensitized to HER inhibitors (5). In light of this, we investigated the mechanisms by which HERs are activated in response to chemotherapy in CRC cells. Initially, we assessed shedding of TGF-α and other HER ligands following exposure to 5-FU treatment both in vitro and in vivo. We found that 5-FU treatment resulted in statistically significant increases in human TGF-α, amphiregulin and heregulin shedding both in culture medium of HCT116 cells and serum of mice bearing human HCT116 xenografts. Furthermore, addition of exogenous EGFR ligands to the culture medium resulted in decreased 5-FU-induced cell death, demonstrating the functional relevance of HER-ligand shedding following chemotherapy treatment. Hagan et al. showed that radiation therapy can increase shedding of TGF-α in serum of patients treated for hormone-refractory prostate cancer (2). Our study is the first to show increased HER-ligand shedding in the context of chemotherapy treatment in colorectal cancer.
Several reports have indicated that different ADAMs such as ADAM-9, -10, -12, -15 and -17 can induce EGFR activation by cleaving the ectodomain of six ligands of EGFR (13), resulting in their shedding and ability to activate the receptor in an autocrine and paracrine manner. Hence, we determined the effect of ADAM-9, 10, 12, 15 and 17 gene silencing on TGF-α shedding and EGFR activity following 5-FU treatment. In both HCT116 and LoVo cell lines, complete inhibition of 5-FU-induced TGF-α shedding and EGFR activation was only observed following ADAM-17 silencing. More importantly, we found that ADAM-17 activity was potently up-regulated following 5-FU treatment in CRC cells and that chemotherapy (the clinically relevant 5-FU and oxaliplatin combination) significantly increased ADAM-17 activity in human HCT116 xenograft models. These results correlated with the results of the TGF-α ELISA both in vitro and in vivo. Moreover, we found that ADAM-17 regulated 5-FU-, SN-38- and oxaliplatin-induced HER ligand shedding and EGFR activation in a broad panel of CRC cell lines, irrespective of p53, Kras or Braf mutational status. Furthermore, a synergistic activation of apoptosis was observed in vitro when ADAM17 siRNA or the small molecule ADAM10/17 inhibitor GW280264X was combined with chemotherapy treatment in this panel CRC cells. In contrast, the specific ADAM-10 inhibitor GI254023X had no effect on chemotherapy-induced apoptosis. Taken together, our findings suggest that chemotherapy treatment results in acute up-regulation in ADAM-17 activity, which promotes EGFR ligand shedding and an EGFR-mediated pro-survival response following chemotherapy treatment. Thus, targeting ADAM-17 in combination with chemotherapy could represent an important treatment strategy for patients with metastatic CRC.
To complement our gene silencing and small molecule inhibitor studies, we further examined the importance of ADAM-17 activity as a mediator of resistance to chemotherapy treatment using ADAM-17 overexpressing HCT116 cell lines. ADAM-17 overexpressing clones showed increased ADAM-17 activity, TGF-α-, amphiregulin and heregulin ligand shedding and EGFR/HER3 activation. Moreover, we showed for the first time that ADAM-17 can also regulate shedding of other growth factors such as IGF-1 and VEGF and subsequently regulates activity of their respective receptors IGF-1R and VEGFR. Of note, the enhanced tumour growth of AD4 PBS xenografted mice compared to EV PBS xenografted mice may be the result of ADAM-17 regulating growth factor shedding and activity of multiple survival receptors that promote xenografts growth. It may be that ADAM-17 should be added to a growing list of non-oncogenes that could be exploited as an anti-cancer drug target (29). Importantly, the clones with increased ADAM-17 activity levels had a decreased response to chemotherapy treatment compared to the empty vector clones (EV). Moreover, overexpression of ADAM-17 protected HCT116 xenografts from the growth-inhibitory effects of chemotherapy and abrogated chemotherapy-induced apoptosis in vivo. Collectively, these results further indicate that ADAM-17 is an important regulator of chemotherapy-resistance and suggests that targeting this ADAM in conjunction with chemotherapy may have therapeutic potential for the treatment of CRC tumours.
Although a number of studies have shown additive interactions when chemotherapy was combined with the ADAM10/17 inhibitor INC3619 in lung cancer, breast cancer and head and neck xenograft models (6, 30), no underlying mechanism behind this interaction was provided. Our data show for the first time that chemotherapy treatment can result in potent increases in ADAM-17 activity in colorectal cancer cells and that this protease thereby regulates resistance to chemotherapy treatment. Blocking this survival response, using small molecule ADAM10/17 inhibitors, resulted in synergistic increase in apoptosis, and this was irrespective of p53, Kras or Braf mutational status. One other study has investigated the role of ADAM-17 in colorectal cancer and only examined the interaction between an ADAM-17 inhibitor and EGFR targeted agents (31); the current study is the first to examine the interaction between ADAM-17 and cytotoxic chemotherapy in CRC. We have previously shown that the death receptor ligand TRAIL can increase TGF-α shedding and HER1/HER2 activity and that this may be regulated by ADAM-17 in colorectal cancer, indicating that activation of ADAM-17 may be a common pro-survival response following treatment with a range of cytotoxic and apoptosis-inducing agents (32). Merchant et al. showed that ADAM-17 is overexpressed in primary and metastatic CRC compared with normal colonic epithelium, further highlighting the importance of ADAM-17 as a potential target in CRC (31).
In conclusion, our findings provide strong evidence that CRC tumours respond to chemotherapy by activating ADAM-17, which results in increased growth factor shedding and activation of growth factor receptor-mediated pro-survival response. Furthermore, we provide strong evidence that enhanced ADAM-17 activity and HER ligand shedding results in resistance to chemotherapy treatment in CRC tumours. Moreover, therapies targeting ADAM-17 (and thereby the activity of multiple receptor tyrosine kinases, such as EGFR, HER3, IGF-1R and VEGFR) in conjunction with existing chemotherapy treatments (FOLFOX, FOLFIRI) may enhance response rates in patients with advanced CRC and thereby improve survival rates compared to those obtained with combined EGFR mAB inhibition (cetuximab)/chemotherapy treatment (33).
Resistance to chemotherapy is a major barrier in the treatment of colorectal cancer. In this study, we show that cytotoxic chemotherapy treatment results in an acute increase in ADAM-17 activity in vitro and in vivo. Blocking ADAM-17 activity, using siRNA or a small molecule inhibitor, significantly increased apoptosis following chemotherapy treatment. We further show that overexpression of ADAM-17 increases activity of the human epidermal receptors (HER) and other pro-survival receptors such as IGF-1R and VEGFR and that this results in resistance to chemotherapy treatment in CRC tumours. Thus, targeting ADAM-17 in conjunction with existing chemotherapy treatments may enhance response rates in patients with advanced CRC by blocking the activity of multiple pro-survival receptors.
Grant support: Cancer Research UK and Ulster Cancer Foundation. We thank GlaxoSmithKline for supplying us with GW280264X and GI254023X. We thank Atanasio Pandiella for providing us with the ADAM-17 construct.
Grant support: Cancer Research UK, Medical Research Council, and Ulster Cancer Foundation.
Conflict of interest
P.G. Johnston: shareholdings, Fusion Antibodies, GlaxoSmithKline (GSK); consultancy, AstraZeneca, Pfizer, Roche Pharmaceuticals, Merck, Amgen, Bristol Myers Squibb, Ortho Biotech; contracted research, AstraZeneca, Amgen. The other authors disclosed no potential conflicts of interest.