Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

The antiepidermal growth factor receptor monoclonal antibody cetuximab/C225 reduces hypoxia-inducible factor-1 alpha, leading to transcriptional inhibition of vascular endothelial growth factor expression

Abstract

We have previously shown that the antiepidermal growth factor receptor monoclonal antibody cetuximab (C225; Erbitux), which was recently approved for the treatment of metastatic colorectal cancer, has antiangiogenic properties, inhibiting vascular endothelial growth factor (VEGF) secretion in culture and in animal models. Here, we have furthered the study by demonstrating that cetuximab reduces cellular levels of hypoxia-inducible factor-1 alpha (HIF-1α), a transcriptional regulator of VEGF expression, in A431 epidermoid carcinoma cells under both normoxic and hypoxic culture conditions. Expression of a constitutively active Ras in A431 cells rendered cellular resistance to the cetuximab-mediated reduction of the HIF-1α level. Cell lines with naturally occurring phosphatase and tensin homologue deleted on chromosome 10 mutations or deletions were also resistant to cetuximab-mediated reduction of the HIF-1α level. Pharmacologic inhibition of phosphatidylinositol 3-kinase with LY294002 reduced the HIF-1α level in both normoxic and hypoxic A431 cells, whereas inhibition of the mitogen-activated protein kinase kinase by PD98059 reduced the level of HIF-1α only in normoxic A431 cells. In addition, cetuximab reduced the cellular level of HIF-1α in the presence of a proteasome inhibitor, lactacystin, indicating that cetuximab acts mainly at the level of protein synthesis. The reduction of HIF-1α in response to cetuximab treatment was accompanied by transcriptional inhibition of VEGF expression, measured by a luciferase assay in A431 cells transfected with a vector containing the VEGF hypoxia response element. Taken together, our results indicate that the previously demonstrated inhibition of VEGF by cetuximab occurs at the level of transcription in response to a reduced level of HIF-1α and justify further testing of therapeutic strategies that combine cetuximab with approaches inhibiting the function of VEGF or the VEGF receptor.

Introduction

The epidermal growth factor (EGF) receptor is a 170-kDa transmembrane glycoprotein with intrinsic tyrosine kinase activity (Ullrich and Schlessinger, 1990) that is often highly expressed in a variety of human tumors of epithelial origin, including nonsmall-cell lung cancer and breast, head and neck, gastric, colorectal, esophageal, prostate, bladder, renal, pancreatic and ovarian cancers (Fan and Mendelsohn, 1998; Mendelsohn and Baselga, 2000). EGF receptor signaling causes increased proliferation, decreased apoptosis, enhanced tumor cell motility and neoangiogenesis, as demonstrated by many laboratory and clinical studies performed over the past 20 years. Several therapeutic strategies to inhibit EGF receptor signaling have been devised, including the use of monoclonal antibodies, such as cetuximab (C225; Erbitux), which has recently been approved to treat irinotecan-refractory metastatic colorectal cancer (Cunningham et al., 2004) and small molecule inhibitors, such as gefitinib (ZD1839; Iressa), which is now approved for the treatment of nonsmall-cell lung cancer after failure of both platinum-based and docetaxel chemotherapies (Cohen et al., 2004). The mechanisms of the antitumor activities of these agents have been well explored, with a large body of experimental evidence revealing that cetuximab and the small molecule inhibitors prevent ligand-induced receptor activation and subsequently inhibit downstream signaling of the EGF receptor, resulting in cell cycle arrest, promotion of apoptosis and inhibition of angiogenesis (Mendelsohn and Baselga, 2000).

With regard to the antiangiogenesis activity, we and others have shown that treatment of human cancer cell lines in vitro (in normoxic culture) and in vivo with cetuximab or gefitinib reduces the production of vascular endothelial growth factor (VEGF), a key factor promoting tumor angiogenesis (Petit et al., 1997; Perrotte et al., 1999; Ciardiello et al., 2000; Milas et al., 2000; Shaheen et al., 2001a; Huang et al., 2002a, 2002b; Karashima et al., 2002). Together with the demonstrated antiproliferative and proapoptotic effects, this antiangiogenic activity of cetuximab and gefitinib is believed to contribute to their overall antitumor activity in vivo (Mendelsohn and Baselga, 2000). Despite these advances, the mechanisms of the reduction of VEGF production by these treatments have not been clearly explored.

An important physiologic stimulus leading to increased levels of VEGF is hypoxia (i.e., insufficient oxygen supply to tissue). Hypoxia is very common in solid tumors and generally occurs at a distance greater than 100 μm from functional blood vessels (Folkman and Shing, 1992; Hanahan and Folkman, 1996; Helmlinger et al., 1997). A hallmark of hypoxia is the dramatic increase in the cellular level of the transcriptional factor hypoxia-inducible factor-1 (HIF-1) (Semenza and Wang, 1992; Wang et al., 1995), which plays a pivotal role in the cellular response to the stress of hypoxia and regulates more than 60 downstream target genes, including VEGF (Forsythe et al., 1996). HIF-1 is composed of an expression-inducible α subunit and a constitutively expressed β subunit. Under hypoxic conditions, the level of HIF-1α increases as a result of decreased ubiquitination and degradation (Pugh et al., 1997; Huang et al., 1998). The molecular basis for this regulation is the O2-dependent hydroxylation of proline residues in HIF-1α (residues 402 and 564) by the oxygen sensor enzymes (HIF-1 prolyl hydroxylases, PHD-1–3) that were recently discovered (Ivan et al., 2001; Jaakkola et al., 2001). Prolyl hydroxylation of HIF-1α is required for its association with the von Hippel–Lindau tumor suppressor protein, which is the recognition partner of an E3 ubiquitination ligase in targeting HIF-1α for proteasomal degradation (Stebbins et al., 1999; Ohh et al., 2000).

The nonhypoxic induction of HIF-1α was first demonstrated in 1997, when it was found that cells expressing the v-Src oncogene have increased expression of HIF-1, VEGF and enolase 1 under both hypoxic and nonhypoxic conditions (Jiang et al., 1997). Later, this nonhypoxic induction of HIF-1α was demonstrated in cells stimulated with many growth factors and cytokines including insulin and insulin-like growth factor-1 (IGF-1), IGF-2, EGF, platelet-derived growth factor, fibroblast growth factor-2, transforming growth factor-β, hepatocyte growth factor, tumor necrosis factor-α, interleukin-1β, angiotensin-2 and thrombin (Zelzer et al., 1998; Feldser et al., 1999; Hellwig-Burgel et al., 1999; Richard et al., 2000; Thornton et al., 2000; Zhong et al., 2000; Gorlach et al., 2001; Tacchini et al., 2001). Activation of oncogenes other than v-Src, such as HER2 and Ras (Chen et al., 2001; Laughner et al., 2001), or certain signaling intermediates, such as nitric oxides (Palmer et al., 2000; Sandau et al., 2000; Sheta et al., 2001) and inactivation of tumor suppressor genes, such as phosphatase and tensin homologue deleted on chromosome 10 (PTEN) (Zundel et al., 2000), have also been shown to upregulate the level of HIF-1α. This receptor- or oncogene-mediated HIF-1α regulation has now been found to occur via two well-known cell signaling pathways: the phosphatidylinositol 3-kinase (PI3-K)/Akt pathway and the mitogen-activated protein kinase (MAPK) kinase (MEK)/MAPK pathway that control protein synthesis and transactivation function of HIF-1α (Minet et al., 2000; Zhong et al., 2000; Blancher et al., 2001; Fukuda et al., 2002; Treins et al., 2002; Sang et al., 2003).

In the current study, we found reduced levels of HIF-1α after cetuximab treatment in A431 cells cultured in both normoxic and hypoxic conditions. To further elucidate the molecular mechanism, we examined the same activity of cetuximab in A431 cells that were engineered to be independent of EGF receptor signaling due to expression of a constitutively active Ras and in other cancer cells that contain naturally occurring loss-of-function mutations or deletions of the PTEN gene. Cetuximab appears to act mainly by reducing the synthesis of the HIF-1α protein in both normoxic and hypoxic conditions. The reduction of the level of HIF-1α was accompanied by transcriptional inhibition of VEGF expression, as determined using a luciferase promoter reporter assay.

Results

Reduction of the cellular level of HIF-1α after cetuximab treatment

Previous reports showing that cetuximab inhibited VEGF production in various tumor cell lines did not examine the changes in the cellular level of HIF-1α, a key transcriptional regulator of VEGF, after treatment. Figure 1 shows the levels of HIF-1α after cetuximab treatment in A431 cells that were cultured in normoxic (21% O2) or hypoxic (1% O2) conditions. Compared with the baseline level of HIF-1α in normoxic conditions, the level in A431 cells cultured in hypoxic conditions showed a marked increase (Figure 1a). The presence of cetuximab in the culture medium markedly reduced both the baseline level of HIF-1α in normoxic A431 cells and the upregulated level of HIF-1α in hypoxic A431 cells (Figure 1a).

Figure 1
figure1

Reduction of the levels of HIF-1α by cetuximab in cells cultured under normoxic and hypoxic conditions. (a) A431 cells were left untreated or were treated with 20 nM cetuximab in serum-free medium for 16 h in a cell culture incubator filled with 21% O2 (normoxic; N) or 1% O2 (hypoxic; H), as described in Materials and methods. (b) A431 cells were left untreated or were treated with 20 nM cetuximab for 16 h in serum-free and normoxic culture medium containing 0, 50, or 100 μM DFO. After cell lysis, equal amounts of protein lysates were assayed for the levels of HIF-1α by Western blot analysis. The level of β-actin was also measured as a protein-loading control

Deferoxamine (DFO), an iron-chelating agent, has been used experimentally to mimic the effects of hypoxia to increase cellular level of HIF-1α and HIF-1α transcriptional activity (Blancher et al., 2000; Alvarez-Tejado et al., 2001; Bacus et al., 2002). An advantage of using DFO in experimental studies is that, within a dose range, exposure of cells to DFO leads to a dose-dependent increase of the HIF-1α level in normoxic conditions that may mimic the hypoxia caused by various levels of oxygen deficiency. Indeed, we found that exposure of cells to DFO resulted in marked and DFO dose-dependent increases in the levels of HIF-1α in A431 cells (Figure 1b). Similar to the effect seen with the cells cultured in the hypoxia chamber, the presence of cetuximab in the culture medium reduced both the baseline level of HIF-1α (without DFO treatment) and the DFO-elevated level of HIF-1α in A431 cells (Figure 1b).

It has been reported that the PI3-K/Akt and MEK/MAPK pathways regulate the level and function of HIF-1α in various types of cells (Richard et al., 1999; Minet et al., 2000; Zhong et al., 2000; Jiang et al., 2001). These two pathways are known to be regulated by EGF receptor tyrosine kinase in a variety of cell models. To explore the signal transduction pathway that leads to the reduced level of HIF-1α after cetuximab treatment, we first compared the levels of phosphorylated Akt and phosphorylated MAPK in normoxic and DFO-induced ‘hypoxic’ cells with or without cetuximab treatment. We found that the increase in the HIF-1α level after exposure of the cells to DFO was not accompanied by any increase in the phosphorylated level of Akt or MAPK in A431 cells, despite others having reported cell-type-specific increases in the levels of phosphorylated Akt and MAPK in their studies (Alvarez-Tejado et al., 2001; Beitner-Johnson et al., 2001).

Similar to the findings with HIF-1α (Figure 1b), the presence of cetuximab in the culture medium reduced the levels of both phosphorylated Akt and MAPK in A431 cells under normoxic (without DFO treatment) or ‘hypoxic’ (induced by 50 or 100 μM DFO) conditions (Figure 2). As expected, the inhibition of Akt and MAPK phosphorylation after cetuximab treatment was not accompanied by changes in the total levels of Akt and MAPK.

Figure 2
figure2

Inhibition of Akt and MAPK phosphorylation by cetuximab in cells cultured under normoxic and DFO-induced hypoxic conditions. A431 cells were left untreated or were treated with 20 nM cetuximab for 16 h in serum-free medium without DFO or containing 50 or 100 μM DFO. Cell lysates were prepared and analysed for the levels of total and phosphorylated Akt and total and phosphorylated MAPK by Western blot analysis with appropriate antibodies

Potential effect of oncogenic mutation of Ras in cetuximab-mediated reduction of HIF-1α

To further elucidate the signal transduction pathway involved in the cetuximab-mediated reduction of the HIF-1α level, we next sought to determine whether expression of a constitutively active Harvey Ras (Ras G12V), which allows cells to grow independent of upstream signaling including that of the EGF receptor, could abolish the ability of cetuximab to reduce the baseline level and DFO-elevated level of HIF-1α. Figure 3a (upper panel) shows the results of high expression of His-tagged RasG12V in stably transfected A431 cells (A431-Ras). There was no substantial difference in the baseline level of HIF-1α between the control-vector transfected (A431V) and RasG12V-transfected (A431-Ras) cells. However, compared with the results from A431V cells, in which cetuximab reduced both the baseline and DFO-elevated levels of HIF-1α, cetuximab failed to reduce the baseline and DFO-elevated levels of HIF-1α in A431-Ras cells (Figure 3a, lower panel). This result indicates that oncogenic mutation of Ras conferred cellular resistance to cetuximab-mediated reduction of HIF-1α, although whether this resistance is mediated via the same pathway that is blocked by cetuximab or via a novel pathway remains to be determined.

Figure 3
figure3

Resistance to cetuximab-mediated reduction of the HIF-1α level in cells expressing a constitutively active Ras. (a) A431 cells transfected with a control vector (A431V) or an expression vector containing constitutively active RasG12 V (A431-Ras) were left untreated or were treated with 20 nM cetuximab for 16 h in serum-free medium without DFO or containing 100 μM DFO. Cell lysates were prepared and analysed for the levels of His-tagged Ras and HIF-1α by Western blot analysis with appropriate antibodies. The level of β-actin was also measured as a protein-loading control. (b) A431 cells were exposed to 20 μM LY294002, 40 μM PD98059, or vehicle control for 16 h in serum-free medium without DFO or containing 100 μM DFO. Cell lysates were prepared and analysed for the levels of HIF-1α, total and phosphorylated Akt and total and phosphorylated MAPK by Western blot analysis with appropriate antibodies. The level of β-actin was also measured as a protein-loading control

We next used specific pharmacologic inhibitors to dissect the two key downstream pathways that are inhibited upon exposure of A431 cells to cetuximab (i.e., the PI3-K/Akt and MEK/MAPK signaling pathways) to determine the degree of their involvement in the regulation of HIF-1α levels. Figure 3b shows that LY294002, a specific inhibitor of PI3-K and PD98059, a specific inhibitor of MEK, each markedly inhibited phosphorylation of their specific targets without affecting the phosphorylation of the other agent's target. The inhibition of the PI3-K/Akt pathway by LY294002 reduced the level of HIF-1α in A431 cells under both normoxic and DFO-induced ‘hypoxic’ conditions (Figure 3b, bottom two panels), whereas the inhibition of the MEK/MAPK pathway by PD98059 at the given dose only moderately reduced the level of HIF-1α under normoxic conditions but had no such effect in DFO-induced ‘hypoxic’ conditions.

Failure of cetuximab to reduce the HIF-1α level in PTEN-mutant cells

Because inhibition of the PI3-K/Akt pathway plays an important role in the cetuximab-induced reduction of HIF-1α levels, we examined whether cells with a constitutively active PI3-K/Akt pathway might therefore be resistant to that reduction. We performed experiments in two cell lines, the breast carcinoma cell line MDA468 and the prostate carcinoma cell line PC3, which contain loss-of-function mutations (MDA468 cells) or deletion (PC3 cells) of the PTEN gene, leading to constitutive activation of the PI3-K/Akt pathway in these cells (Vlietstra et al., 1998; Lu et al., 1999). Figure 4 shows that, in contrast to the results seen in A431 cells, exposure of MDA468 and PC3 cells to cetuximab did not reduce the level of phosphorylated Akt under either normoxic or DFO-induced ‘hypoxic’ conditions, despite marked inhibition of MAPK phosphorylation in MDA468 cells and moderate inhibition of MAPK phosphorylation in PC3 cells after cetuximab treatment (Harper et al., 2002; Shimada et al., 2004).

Figure 4
figure4

Failure of cetuximab to reduce the HIF-1α level in PTEN-mutant cells. Two cell lines with naturally occurring loss-of-function PTEN mutations (MDA468) or deletions (PC3) were left untreated or were treated with 20 nM cetuximab for 16 h in serum-free medium without DFO or containing 50 or 100 μM DFO. Cell lysates were prepared and analysed for the levels of HIF-1α, total and phosphorylated Akt and total and phosphorylated MAPK by Western blot analysis with appropriate antibodies. The level of β-actin was also measured as a protein-loading control

As we predicted, treatment of MDA468 and PC3 cells with cetuximab failed to reduce the levels of HIF-1α under either normoxic or DFO-induced ‘hypoxic’ conditions (Figure 4, lower panels). These observations suggest that the PTEN mutation within these cells confers resistance to cetuximab-mediated inhibition of the PI3-K/Akt pathway and thereby contributes to the maintenance of a constitutive level of HIF-1α in these cells.

Reduced HIF-1α protein synthesis by cetuximab

We further examined whether the cetuximab-mediated reduction in HIF-1α was attributable to increased HIF-1α proteasomal degradation or reduced HIF-1α protein synthesis. To answer this question, we performed an experiment to evaluate whether pharmacologic inhibition of the ubiquitin/proteasome degradation pathway with a proteasome inhibitor, lactacystin, would prevent the reduction of HIF-1α after cetuximab treatment. Figure 5 shows the results of a Western blot analysis for the level of HIF-1α in A431 cells that were cultured overnight with DFO followed by the addition of cetuximab with or without lactacystin for 8 h. As expected, the presence of cetuximab in the culture medium reduced the level of HIF-1α induced by DFO to approximately 50% of that in the culture without cetuximab (Figure 5a and b). Cetuximab also reduced the DFO-induced increase in the level of HIF-1α in the presence of lactacystin in the culture medium. This indicates that cetuximab acts primarily by reducing HIF-1α protein synthesis; although the possibility that cetuximab increases HIF-1α proteasomal degradation cannot be completely ruled out. Of note, lactacystin also led to a higher level of HIF-1α in untreated cells (without cetuximab), an expected result because the inhibition of the proteasome would naturally lead to reduced degradation of HIF-1α in hypoxic cells.

Figure 5
figure5

Decreased protein synthesis of HIF-1α after cetuximab treatment. (a) A431 cells were cultured with or without 20 nM cetuximab for 8 h in serum-free medium containing 100 μM DFO with or without 10 μM lactacystin (LCN), as indicated in the figure. Cells cultured in medium without any additions served as the control. After the treatment, cell lysates were prepared and analysed for levels of HIF-1α by Western blot analysis. The level of β-actin was also measured as a protein-loading control. (b) The density of each band in (a) was analysed by an image analyser (FluorChem 8000, San Leandro, CA, USA). The band density of the cells treated with DFO only was set as 1 arbitrary unit

Transcriptional inhibition of VEGF expression by cetuximab

To determine whether the reduction of the HIF-1α level after cetuximab treatment leads to functional inhibition of HIF-1α, we examined the transcriptional activity of HIF-1α in A431 cells that were transiently transfected with the luciferase reporter gene construct pBI-GL-V6L, which contains six tandem repeats of the hypoxia response element from the human VEGF gene (Post and Van Meir, 2001). Compared with the results from normoxic cells with or without cetuximab treatment, a substantial increase in luciferase activity was observed in cells cultured in hypoxic conditions, either mimicked by exposure to DFO (a 24.5-fold increase) or induced environmentally in a 1% O2 chamber (a 44-fold increase) (Figure 6). Cetuximab remarkably reduced the increase in luciferase activity induced by DFO (from a 24.5-fold to an 11.8-fold increase) or by 1% O2 (from a 44-fold to an 18.4-fold increase). There was also a considerable reduction of luciferase activity for cells cultured in normoxic conditions after cetuximab treatment compared with untreated cells; however, it should be noted that the absolute value of the luciferase activity was low.

Figure 6
figure6

Transcriptional inhibition of VEGF expression after cetuximab treatment. A431 cells were transiently transfected with the pBI-GL-V6L vector for 24 h with FuGENE-6, as described in Materials and methods. The cells were left untreated or were treated with 20 nM cetuximab for 16 h in three culture conditions: regular culture (normoxic, 21% O2), normoxic culture medium containing 100 μM DFO, or hypoxic culture (1% O2). After the treatment, luciferase reporter activity was measured in each group with the method described in Materials and methods. Relative luciferase activity was determined by standardizing the readings of untreated cells in normoxic culture (21% O2) to 1 after normalizing values to the total protein concentration of each sample. The results are shown as the means and standard deviations of triplicate wells

Discussion

This work expands on the findings of previous studies reported by us and others demonstrating that blockade or inhibition of the EGF receptor by monoclonal antibodies or small molecular inhibitors leads to reduced expression of VEGF and inhibition of angiogenesis (Petit et al., 1997; Perrotte et al., 1999; Ciardiello et al., 2000; Milas et al., 2000; Shaheen et al., 2001a; Huang et al., 2002a, 2002b; Karashima et al., 2002). It was shown earlier that VEGF is transcriptionally regulated by HIF-1 (Forsythe et al., 1996). In the current study, we found that cetuximab reduces the level of HIF-1α, leading to transcriptional inhibition of VEGF expression. Inhibition of proteasomal degradation with lactacystin did not alter the rate of HIF-1α reduction cetuximab treatment, suggesting that cetuximab mainly acts by inhibiting protein synthesis. The level of HIF-1α reduction by cetuximab treatment was more apparent in cells cultured in hypoxic conditions than in the same cells cultured in normoxic conditions. This result was expected because HIF-1α has a much longer half-time under hypoxic conditions than under normoxic conditions and is thus more sensitive to inhibition of protein synthesis in hypoxia than in normoxia. In contrast, because the levels of HIF-1α in hypoxic cells resulted mainly from reduced degradation rather than increased expression, hypoxic cells would be less sensitive to the stimulation of HIF-1α synthesis by EGF. This may explain the findings of another group that, although EGF increased VEGF production in the supernatant of cells cultured in normoxic conditions, it had no effect when the same cells were grown in hypoxic conditions (Clarke et al., 2001). Overall, our data support the concept that EGF receptor-mediated signaling is important in mediating HIF-1α and the corresponding VEGF levels in both normoxic and hypoxic conditions.

The mechanism of reducing HIF-1α synthesis by cetuximab is most likely through inhibiting the PI3-K/Akt pathway, which is consistent with findings from several laboratories demonstrating that cell signal transduction leading to activation the PI3-K/Akt pathway stimulates HIF-1α protein synthesis (Zhong et al., 2000; Blancher et al., 2001; Fukuda et al., 2002; Treins et al., 2002). These previous studies reported that EGF induces HIF-1α expression by stimulating the PI3-K/Akt signaling in prostate cells (Zhong et al., 2000). In addition to EGF stimulation, activation of PI3-K/Akt signaling by other growth factors, such as heregulin, platelet-derived growth factor, or insulin, also increased HIF-1α expression, suggesting a common role for this pathway in mediating the regulation of HIF-1α by growth factors (Jiang et al., 2001; Laughner et al., 2001; Fukuda et al., 2002; Treins et al., 2002; Zhang et al., 2003).

Previous studies have suggested that inhibition of MAPK alone may not alter the levels of HIF-1α (Laughner et al., 2001), although MAPK may enhance HIF-1α transcriptional activity through phosphorylation of p300/CBP (CREB-binding protein), a co-activator of HIF-1α (Arany et al., 1996; Richard et al., 1999; Minet et al., 2000; Sang et al., 2003; Ruas et al., 2005). In the current study, we found that, despite the fact that inhibition of MAPK by the MEK inhibitor PD98059 moderately reduced the level of HIF-1α in normoxic A431 cells, the reduction was less than that produced by the inhibition of PI3-K by LY294002 and no reduction of HIF-1α by PD98059 was observed in hypoxic cells, which are supposed to be more sensitive to approaches leading to inhibition of HIF-1α protein synthesis. However, because the inhibition of MAPK in A431 cells by the dose of PD98059 used in the current study was incomplete and because a higher dose of PD98059 led to nonspecific inhibition of other kinases including PI3-K (data not shown), the definitive role of the MAPK pathway remains to be determined. As an important downstream pathway of EGF receptor signaling, the MEK/MAPK pathway is often inhibited after cetuximab treatment in sensitive cell lines, shown by reduced levels of phosphorylated MAPK in our previous reports with other types of cells (Albanell et al., 2001; Liu et al., 2001; Liang et al., 2003) and in our current study with A431, PC3 and MDA468 cells. Because inhibition of the MEK/MAPK pathway and inhibition of the PI3-K/Akt pathway often coexist, it is generally difficult to separate the effect of MAPK inhibition from that of PI3-K/Akt inhibition on the level of HIF-1α after cetuximab treatment. However, the mutation of PTEN in MDA468 cells, which renders the PI3K/Akt pathway insensitive to cetuximab, allowed us to determine a definitive role of MEK/MAPK inhibition in the reduction of HIF-1α by cetuximab. Despite the importance of inhibiting the MEK/MAPK pathway in mediating the antiproliferative effect of cetuximab on cell growth, our findings that inhibition of this pathway was not accompanied by reduction of HIF-1α suggest that the pathway does not contribute to the antiangiogenic activity of cetuximab.

In the current study, we demonstrated that the activity of cetuximab to reduce the level of HIF-1α protein in A431 cells was abolished by expression of a constitutively active Ras. We also found that cells with naturally occurring loss-of-function PTEN mutations (MDA468 cells) or deletion (PC3 cells) were resistant to the cetuximab-mediated reduction of the level of HIF-1α. This observation may provide further mechanistic explanation for the result of recent studies reporting that a loss of PTEN counteracted the antitumor action of EGF receptor tyrosine kinase inhibitor ZD1839 (Bianco et al., 2003; She et al., 2003). It has been shown previously that experimental expression of wild-type PTEN in human prostate cancer cells or glioma cells reduced the level of HIF-1α or inhibited HIF-1α-mediated gene regulation (Zhong et al., 2000; Zundel et al., 2000).

Our findings may have clinical implications for patients who are treated with cetuximab. Despite the fact that recent clinical studies have demonstrated the antitumor activity of cetuximab or gefitinib in patients, particularly in a small percentage of patients whose tumors contain the activating mutations of the EGF receptor; the latter have been linked to much more favorable responses to gefitinib treatment (Lynch et al., 2004; Paez et al., 2004). Still, large numbers of patients whose tumors express or even highly express the EGF receptors do not show favorable responses (Fukuoka et al., 2003; Kris et al., 2003; Cunningham et al., 2004; Giaccone et al., 2004; Herbst et al., 2004). The lack of clinical responses to the EGF receptor-directed therapy may be caused by multiple intrinsic and extrinsic resistance mechanisms. Ras oncogenic mutation (gain-of-function) or PTEN mutational inactivation (loss-of-function) may contribute to these resistance mechanisms. Our data suggest that these types of resistance mechanism may be circumvented by supplementation of the anti-EGF receptor therapy with additional approaches targeting HIF-1α or VEGF, one of its major targeted genes. In deed, several recent studies have explored the combined use of anti-EGF receptor inhibitors with VEGF or VEGF receptor inhibitors (Shaheen et al., 2001b, 2002; Jung et al., 2002). We and others have shown that a combination of cetuximab and DC101, a monoclonal antibody to VEGFR2, resulted in less growth and more apoptosis in several human cancer cell lines than did either of the agents alone (Shaheen et al., 2001b, 2002; Jung et al., 2002). Another alternative strategy is to develop novel inhibitors that target the EGF receptor family and the VEGF receptor simultaneously. An example of this approach is AEE788, a potent inhibitor of both EGF and VEGF receptor tyrosine kinase family members, inhibiting the proliferation of EGF- and VEGF-stimulated human umbilical vein endothelial cells and cells of several tumor models (Traxler et al., 2004). It should be noted that the inhibition of HIF-1α by cetuximab would contribute not only to the blockade of tumor angiogenesis but also to the transcriptional inhibition of more than 60 genes that are activated by HIF-1α and other potential gene products not yet found to be associated with HIF-1α regulation. Thus, it would be interesting to explore whether a combination of cetuximab and novel HIF-1α inhibitors may also lead to an exciting therapy for solid tumors, especially highly hypoxic solid tumors.

In conclusion, the results of our study expand our understanding of the antiangiogenic mechanisms of cetuximab to include its ability to inhibit HIF-1α expression via the PI3-K/Akt signaling pathway and to decrease HIF-1α protein synthesis. Evaluation of the therapeutic responses to the blockade of growth factor-mediated signaling in hypoxic tumors will provide greater insight into strategic planning and a better combined use of anticancer agents for more effective therapeutic intervention.

Materials and methods

Antibodies and reagents

The human–mouse chimeric anti-EGF receptor antibody cetuximab has been described previously (Fan and Mendelsohn, 1998; Mendelsohn and Baselga, 2003) and was provided by ImClone Systems Inc. (New York, NY, USA). Antibodies directed against total Akt, ser473-phosphorylated Akt and phosphorylated MAPK p42/p44 were obtained from Cell Signaling Technology Inc. (Beverly, MA, USA). The anti-MAPK (Erk2) antibody was purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA). The anti-HIF-1α antibodies used for Western blotting and immunoprecipitation studies were purchased from BD Transduction Laboratories (San Diego, CA, USA) and Affinity BioReagents (Golden, CO, USA), respectively. The anti-His G antibody was purchased from Upstate Biotechnology (Charlottesville, VA, USA). The MEK inhibitor PD98059 and the PI3-K inhibitor LY294002 were both purchased from CalBiochem Corp. (San Diego, CA, USA) and the proteasome inhibitor lactacystin was purchased from AG Scientific Inc. (San Diego, CA, USA). All other reagents were purchased from Sigma-Aldrich (St Louis, MO, USA).

Cells and cell culture

The epidermoid carcinoma cell line A431, the breast carcinoma cell line MDA468 and the prostate carcinoma cell line PC3 have all been previously described (Fan et al., 1993a, 1993b; Karashima et al., 2002). All the three cell lines were maintained in Dulbecco's-modified Eagle's medium containing 10% fetal bovine serum, 2 mM glutamine, 100 U/ml penicillin and 100 μg/ml streptomycin. Cells were incubated in a humidified atmosphere of 95% air and 5% CO2 at 37°C (normoxic conditions). For hypoxic stimulation, cells were placed in an airtight chamber that was flushed with a gas mixture of 5% CO2 and 95% N2. The oxygen concentration inside the chamber was maintained at 1% and maintained using the Pro-Ox O2 regulator (Model 110; BioSpherix, Redfield, NY, USA). The hypoxic chamber was placed in the same 37°C incubator where parallel groups of cells were incubated under normoxic culture conditions. Alternatively, cells were treated with DFO for 16–24 h to mimic hypoxic conditions.

For stable expression of the constitutively active Harvey Ras (G12V), A431 cells were transfected with pcDNA3.1RasG12V or pcDNA3.1 backbone vector with the FuGENE-6 transfection kit (Roche Diagnostics Corp., Indianapolis, IN, USA) and selected with G418 as previously described (Jin et al., 2003). The expression of Ras G12V was measured by Western blot analysis with anti-His G-tag antibodies in selected pooled cells.

Western blot analysis

Cells were lysed in a lysis buffer containing 50 mM Tris (pH 7.4), 150 mM NaCl, 0.5% NP-40, 50 mM NaF, 1 mM Na3VO4, 1 mM phenylmethylsulfonyl fluoride, 25 μg/ml leupeptin and 25 μg/ml aprotinin and clarified by centrifugation (14 000 g for 20 min at 4°C). Equal amounts of protein lysate, as determined by the Pierce Coomassie Plus colorimetric protein assay method (Pierce, Rockford, IL, USA), were separated by SDS-polyacrylamide gel electrophoresis, blotted onto nitrocellulose and probed with the indicated primary antibodies. The signals were visualized using the Enhanced chemiluminescence detection kit (Amersham Biosciences, Piscataway, NJ, USA).

Transient transfection and luciferase assay

The pBI-GL-V6L construct, which contains six copies of the VEGF hypoxia response element, has been previously described (Post and Van Meir, 2001). A431 cells were transiently transfected with the pBI-GL-V6L construct using the FuGENE-6 transfection kit. After a 24-h transfection period, the cells were washed twice with phosphate-buffered saline and cultured with or without DFO (100 μM) or in the hypoxic chamber for an additional 16 h in serum-free medium in the presence or absence of 20 nM cetuximab. The cells were then harvested and lysed in a lysis buffer (0.2 M Tris-HCl (pH 8.0) and 0.1% Triton X-100). The luciferase assay was performed by adding luciferase substrate solution (0.5 mM D-luciferin, 0.25 mM coenzyme A, 20 mM Tris HCl, 4 mM MgSO4, 0.1 mM EDTA, 30 mM DTT and 0.5 mM ATP) to the samples and immediately measuring for luciferase activity using a multiplate luminometer (Berthold Detection Systems, Oak Ridge, TN, USA). Arbitrary luciferase activity units were normalized to the amount of protein in each sample. The protein concentration was determined using the Pierce BCA protein assay kit.

References

  1. Albanell J, Codony-Servat J, Rojo F, Del Campo JM, Sauleda S, Anido J, Raspall G, Giralt J, Rosello J, Nicholson RI, Mendelsohn J and Baselga J . (2001). Cancer Res., 61, 6500–6510.

  2. Alvarez-Tejado M, Naranjo-Suarez S, Jimenez C, Carrera AC, Landazuri MO and del Peso L . (2001). J. Biol. Chem., 276, 22368–22374.

  3. Arany Z, Huang LE, Eckner R, Bhattacharya S, Jiang C, Goldberg MA, Bunn HF and Livingston DM . (1996). Proc. Natl. Acad. Sci. USA, 93, 12969–12973.

  4. Bacus SS, Altomare DA, Lyass L, Chin DM, Farrell MP, Gurova K, Gudkov A and Testa JR . (2002). Oncogene, 21, 3532–3540.

  5. Beitner-Johnson D, Rust RT, Hsieh TC and Millhorn DE . (2001). Cell Signal., 13, 23–27.

  6. Bianco R, Shin I, Ritter CA, Yakes FM, Basso A, Rosen N, Tsurutani J, Dennis PA, Mills GB and Arteaga CL . (2003). Oncogene, 22, 2812–2822.

  7. Blancher C, Moore JW, Robertson N and Harris AL . (2001). Cancer Res., 61, 7349–7355.

  8. Blancher C, Moore JW, Talks KL, Houlbrook S and Harris AL . (2000). Cancer Res., 60, 7106–7113.

  9. Chen C, Pore N, Behrooz A, Ismail-Beigi F and Maity A . (2001). J. Biol. Chem., 276, 9519–9525.

  10. Ciardiello F, Bianco R, Damiano V, Fontanini G, Caputo R, Pomatico G, De Placido S, Bianco AR, Mendelsohn J and Tortora G . (2000). Clin. Cancer Res., 6, 3739–3747.

  11. Clarke K, Smith K, Gullick WJ and Harris AL . (2001). Br. J. Cancer, 84, 1322–1329.

  12. Cohen MH, Williams GA, Sridhara R, Chen G, McGuinn Jr WD, Morse D, Abraham S, Rahman A, Liang C, Lostritto R, Baird A and Pazdur R . (2004). Clin. Cancer Res., 10, 1212–1218.

  13. Cunningham D, Humblet Y, Siena S, Khayat D, Bleiberg H, Santoro A, Bets D, Mueser M, Harstrick A, Verslype C, Chau I and Van Cutsem E . (2004). N. Engl. J. Med., 351, 337–345.

  14. Fan Z, Baselga J, Masui H and Mendelsohn J . (1993a). Cancer Res., 53, 4637–4642.

  15. Fan Z, Masui H, Altas I and Mendelsohn J . (1993b). Cancer Res., 53, 4322–4328.

  16. Fan Z and Mendelsohn J . (1998). Curr. Opin. Oncol., 10, 67–73.

  17. Feldser D, Agani F, Iyer NV, Pak B, Ferreira G and Semenza GL . (1999). Cancer Res., 59, 3915–3918.

  18. Folkman J and Shing Y . (1992). J. Biol. Chem., 267, 10931–10934.

  19. Forsythe JA, Jiang BH, Iyer NV, Agani F, Leung SW, Koos RD and Semenza GL . (1996). Mol. Cell. Biol., 16, 4604–4613.

  20. Fukuda R, Hirota K, Fan F, Jung YD, Ellis LM and Semenza GL . (2002). J. Biol. Chem., 277, 38205–38211.

  21. Fukuoka M, Yano S, Giaccone G, Tamura T, Nakagawa K, Douillard JY, Nishiwaki Y, Vansteenkiste J, Kudoh S, Rischin D, Eek R, Horai T, Noda K, Takata I, Smit E, Averbuch S, Macleod A, Feyereislova A, Dong RP and Baselga J . (2003). J. Clin. Oncol., 21, 2237–2246.

  22. Giaccone G, Herbst RS, Manegold C, Scagliotti G, Rosell R, Miller V, Natale RB, Schiller JH, Von Pawel J, Pluzanska A, Gatzemeier U, Grous J, Ochs JS, Averbuch SD, Wolf MK, Rennie P, Fandi A and Johnson DH . (2004). J. Clin. Oncol., 22, 777–784.

  23. Gorlach A, Diebold I, Schini-Kerth VB, Berchner-Pfannschmidt U, Roth U, Brandes RP, Kietzmann T and Busse R . (2001). Circ. Res., 89, 47–54.

  24. Hanahan D and Folkman J . (1996). Cell, 86, 353–364.

  25. Harper ME, Goddard L, Glynne-Jones E, Assender J, Dutkowski CM, Barrow D, Dewhurst OL, Wakeling AE and Nicholson RI . (2002). Prostate, 52, 59–68.

  26. Hellwig-Burgel T, Rutkowski K, Metzen E, Fandrey J and Jelkmann W . (1999). Blood, 94, 1561–1567.

  27. Helmlinger G, Yuan F, Dellian M and Jain RK . (1997). Nat. Med., 3, 177–182.

  28. Herbst RS, Giaccone G, Schiller JH, Natale RB, Miller V, Manegold C, Scagliotti G, Rosell R, Oliff I, Reeves JA, Wolf MK, Krebs AD, Averbuch SD, Ochs JS, Grous J, Fandi A and Johnson DH . (2004). J. Clin. Oncol., 22, 785–794.

  29. Huang LE, Gu J, Schau M and Bunn HF . (1998). Proc. Natl. Acad. Sci. USA, 95, 7987–7992.

  30. Huang SM, Li J, Armstrong EA and Harari PM . (2002a). Cancer Res., 62, 4300–4306.

  31. Huang SM, Li J and Harari PM . (2002b). Mol. Cancer Ther., 1, 507–514.

  32. Ivan M, Kondo K, Yang H, Kim W, Valiando J, Ohh M, Salic A, Asara JM, Lane WS and Kaelin Jr WG . (2001). Science, 292, 464–468.

  33. Jaakkola P, Mole DR, Tian YM, Wilson MI, Gielbert J, Gaskell SJ, Kriegsheim A, Hebestreit HF, Mukherji M, Schofield CJ, Maxwell PH, Pugh CW and Ratcliffe PJ . (2001). Science, 292, 468–472.

  34. Jiang BH, Agani F, Passaniti A and Semenza GL . (1997). Cancer Res., 57, 5328–5335.

  35. Jiang BH, Jiang G, Zheng JZ, Lu Z, Hunter T and Vogt PK . (2001). Cell Growth Differ., 12, 363–369.

  36. Jin W, Wu L, Liang K, Liu B, Lu Y and Fan Z . (2003). Br. J. Cancer, 89, 185–191.

  37. Jung YD, Mansfield PF, Akagi M, Takeda A, Liu W, Bucana CD, Hicklin DJ and Ellis LM . (2002). Eur. J Cancer, 38, 1133–1140.

  38. Karashima T, Sweeney P, Slaton JW, Kim SJ, Kedar D, Izawa JI, Fan Z, Pettaway C, Hicklin DJ, Shuin T and Dinney CP . (2002). Clin. Cancer Res., 8, 1253–1264.

  39. Kris MG, Natale RB, Herbst RS, Lynch Jr TJ, Prager D, Belani CP, Schiller JH, Kelly K, Spiridonidis H, Sandler A, Albain KS, Cella D, Wolf MK, Averbuch SD, Ochs JJ and Kay AC . (2003). JAMA, 290, 2149–2158.

  40. Laughner E, Taghavi P, Chiles K, Mahon PC and Semenza GL . (2001). Mol. Cell. Biol., 21, 3995–4004.

  41. Liang K, Ang KK, Milas L, Hunter N and Fan Z . (2003). Int. J Radiat. Oncol. Biol. Phys., 57, 246–254.

  42. Liu B, Fang M, Lu Y, Mendelsohn J and Fan Z . (2001). Oncogene, 20, 1913–1922.

  43. Lu Y, Lin YZ, LaPushin R, Cuevas B, Fang X, Yu SX, Davies MA, Khan H, Furui T, Mao M, Zinner R, Hung MC, Steck P, Siminovitch K and Mills GB . (1999). Oncogene, 18, 7034–7045.

  44. Lynch TJ, Bell DW, Sordella R, Gurubhagavatula S, Okimoto RA, Brannigan BW, Harris PL, Haserlat SM, Supko JG, Haluska FG, Louis DN, Christiani DC, Settleman J and Haber DA . (2004). N. Engl. J. Med., 350, 2129–2139.

  45. Mendelsohn J and Baselga J . (2000). Oncogene, 19, 6550–6565.

  46. Mendelsohn J and Baselga J . (2003). J. Clin. Oncol., 21, 2787–2799.

  47. Milas L, Mason K, Hunter N, Petersen S, Yamakawa M, Ang K, Mendelsohn J and Fan Z . (2000). Clin. Cancer Res., 6, 701–708.

  48. Minet E, Arnould T, Michel G, Roland I, Mottet D, Raes M, Remacle J and Michiels C . (2000). FEBS Lett., 468, 53–58.

  49. Ohh M, Park CW, Ivan M, Hoffman MA, Kim TY, Huang LE, Pavletich N, Chau V and Kaelin WG . (2000). Nat. Cell Biol., 2, 423–427.

  50. Paez JG, Janne PA, Lee JC, Tracy S, Greulich H, Gabriel S, Herman P, Kaye FJ, Lindeman N, Boggon TJ, Naoki K, Sasaki H, Fujii Y, Eck MJ, Sellers WR, Johnson BE and Meyerson M . (2004). Science, 304, 1497–1500.

  51. Palmer LA, Gaston B and Johns RA . (2000). Mol. Pharmacol., 58, 1197–1203.

  52. Perrotte P, Matsumoto T, Inoue K, Kuniyasu H, Eve BY, Hicklin DJ, Radinsky R and Dinney CP . (1999). Clin. Cancer Res., 5, 257–265.

  53. Petit AM, Rak J, Hung MC, Rockwell P, Goldstein N, Fendly B and Kerbel RS . (1997). Am. J. Pathol., 151, 1523–1530.

  54. Post DE and Van Meir EG . (2001). Gene Ther., 8, 1801–1807.

  55. Pugh CW, O'Rourke JF, Nagao M, Gleadle JM and Ratcliffe PJ . (1997). J. Biol. Chem., 272, 11205–11214.

  56. Richard DE, Berra E, Gothie E, Roux D and Pouyssegur J . (1999). J. Biol. Chem., 274, 32631–32637.

  57. Richard DE, Berra E and Pouyssegur J . (2000). J. Biol. Chem., 275, 26765–26771.

  58. Ruas JL, Poellinger L and Pereira T . (2005). J. Cell Sci., 118, 301–311.

  59. Sandau KB, Faus HG and Brune B . (2000). Biochem. Biophys. Res. Commun., 278, 263–267.

  60. Sang N, Stiehl DP, Bohensky J, Leshchinsky I, Srinivas V and Caro J . (2003). J. Biol. Chem., 278, 14013–14019.

  61. Semenza GL and Wang GL . (1992). Mol. Cell. Biol., 12, 5447–5454.

  62. Shaheen RM, Ahmad SA, Liu W, Reinmuth N, Jung YD, Tseng WW, Drazan KE, Bucana CD, Hicklin DJ and Ellis LM . (2001a). Br. J. Cancer, 85, 584–589.

  63. Shaheen RM, Ahmad SA, Liu W, Reinmuth N, Jung YD, Tseng WW, Drazan KE, Bucana CD, Hicklin DJ and Ellis LM . (2001b). Br. J. Cancer, 85, 584–589.

  64. She QB, Solit D, Basso A and Moasser MM . (2003). Clin. Cancer Res., 9, 4340–4346.

  65. Sheta EA, Trout H, Gildea JJ, Harding MA and Theodorescu D . (2001). Oncogene, 20, 7624–7634.

  66. Shimada K, Nakamura M, Ishida E, Kishi M, Matsuyoshi S and Konishi N . (2004). Mol. Carcinogen., 39, 1–9.

  67. Stebbins CE, Kaelin Jr WG and Pavletich NP . (1999). Science, 284, 455–461.

  68. Sweeney P, Karashima T, Kim SJ, Kedar D, Mian B, Huang S, Baker C, Fan Z, Hicklin DJ, Pettaway CA and Dinney CP . (2002). Clin. Cancer Res., 8, 2714–2724.

  69. Tacchini L, Dansi P, Matteucci E and Desiderio MA . (2001). Carcinogenesis, 22, 1363–1371.

  70. Thornton RD, Lane P, Borghaei RC, Pease EA, Caro J and Mochan E . (2000). Biochem. J., 350 (Part 1), 307–312.

  71. Traxler P, Allegrini PR, Brandt R, Brueggen J, Cozens R, Fabbro D, Grosios K, Lane HA, McSheehy P, Mestan J, Meyer T, Tang C, Wartmann M, Wood J and Caravatti G . (2004). Cancer Res., 64, 4931–4941.

  72. Treins C, Giorgetti-Peraldi S, Murdaca J, Semenza GL and Van Obberghen E . (2002). J. Biol. Chem., 277, 27975–27981.

  73. Ullrich A and Schlessinger J . (1990). Cell, 61, 203–212.

  74. Vlietstra RJ, van Alewijk DC, Hermans KG, van Steenbrugge GJ and Trapman J . (1998). Cancer Res., 58, 2720–2723.

  75. Wang GL, Jiang BH, Rue EA and Semenza GL . (1995). Proc. Natl. Acad. Sci. USA, 92, 5510–5514.

  76. Zelzer E, Levy Y, Kahana C, Shilo BZ, Rubinstein M and Cohen B . (1998). EMBO J., 17, 5085–5094.

  77. Zhang SX, Gozal D, Sachleben Jr LR, Rane M, Klein JB and Gozal E . (2003). FASEB J., 17, 1709–1711.

  78. Zhong H, Chiles K, Feldser D, Laughner E, Hanrahan C, Georgescu MM, Simons JW and Semenza GL . (2000). Cancer Res., 60, 1541–1545.

  79. Zundel W, Schindler C, Haas-Kogan D, Koong A, Kaper F, Chen E, Gottschalk AR, Ryan HE, Johnson RS, Jefferson AB, Stokoe D and Giaccia AJ . (2000). Genes Dev., 14, 391–396.

Download references

Acknowledgements

We thank Drs Erwin G Van Meir and Dawn E Post (Department of Neurosurgery, Winship Cancer Institute, Emory University, Atlanta, GA, USA) for providing the pBI-GL-V6L construct and Mr David Galloway (Department of Scientific Publications, MD Anderson Cancer Center, Houston, TX, USA) for editorial assistance. This work was supported in part by a generous grant from the Breast Cancer Research Foundation (New York, NY, USA).

Author information

Affiliations

Authors

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Luwor, R., Lu, Y., Li, X. et al. The antiepidermal growth factor receptor monoclonal antibody cetuximab/C225 reduces hypoxia-inducible factor-1 alpha, leading to transcriptional inhibition of vascular endothelial growth factor expression. Oncogene 24, 4433–4441 (2005). https://doi.org/10.1038/sj.onc.1208625

Download citation

Keywords

  • HIF-1α
  • VEGF
  • EGF receptor
  • Akt
  • Ras
  • PTEN

Further reading

Search

Quick links