Cellular iron is needed for cell survival and hydroxylation of hypoxia-inducible factor-1α (HIF-α) by prolyl hydroxylases (PHD). One mechanism of iron uptake is mediated by the cell surface transferrin receptor (TfR). Because iron is required for cell growth and suppression of HIF-α levels, we tested the effects of the two anti-TfR monoclonal antibodies (mAb) E2.3 and A27.15 on growth of breast cancer cells and induction of HIF-α and hypoxia-regulated genes. Treatment with both mAbs together synergistically inhibited cell proliferation in a dose-responsive manner by up to 80% following 8 days of exposure, up-regulated HIF-1α and HIF transcription targets, down-regulated TfR expression, and down-regulated cellular labile iron pool by 60%. Because combined treatment with anti-TfR mAbs resulted in the up-regulation of the hypoxia pathway, which may increase tumor angiogenesis, we analyzed the effects of ascorbate on cell viability and HIF-1α levels in cells treated with both anti-TfR mAbs together, as ascorbate has been shown to be required by PHD enzymes for full catalytic activity. Ascorbate at physiologic concentrations (25 μmol/L) suppressed HIF-1α protein levels and HIF transcriptional targets in anti-TfR mAb-treated cells but did not suppress the antiproliferative effect of the mAbs. These results indicate that the addition of ascorbate increased the activity of the PHD enzymes in down-regulating HIF but not the proliferation of iron-starved anti-TfR mAb-treated cells. The use of anti-TfR mAbs and ascorbate in inhibiting both cell proliferation and HIF-1α and angiogenesis under normoxic conditions may be of therapeutic use. (Cancer Res 2006; 66(5): 2749-56)

Hypoxia-inducible factor (HIF) is an α,β heterodimeric transcription factor, which directs a broad range of responses in hypoxic cells (1). HIF-α is a constitutive nuclear localized subunit, which binds to available HIF-α (2). In the presence of oxygen, two prolyl sites within a central degradation domain of HIF-α are hydroxylated by a set of closely related Fe(II) and 2-oxoglutarate-dependent oxygenases [prolyl hydroxylases (PHD) 1-3], which leads to HIF-α degradation via the pVHL E3 ubiquitin ligase complex and the 26S proteosome (3). A second hydroxylation-dependent control of HIF on an asparaginyl residue in the COOH-terminal activation domain by another Fe(II) and 2-oxoglutarate-dependent oxygenase, factor-inhibiting hypoxia, inhibits HIF-α transcriptional activity by preventing interaction with the p300/CBP coactivator (4). These enzymes are also known to be dependent on iron and ascorbate for full catalytic activity (3). Limiting oxygen levels or the availability of iron with iron chelators allows HIF-α to escape proteolysis and become transcriptionally active, which leads to up-regulation of genes involved in angiogenesis, glucose metabolism, and pH regulation (5). The HIF transcription cascade has been shown to contribute to tumor progression and metastasis and plays an important part in the malignant phenotype, contributing to increased angiogenesis, enhanced glycolysis, and other properties that promote growth (1).

In addition to iron involvement in hydroxylation of HIF-α, iron plays critical roles in electron transport and cellular respiration, cell proliferation and differentiation, and regulation of gene expression and DNA synthesis (68). Growing cells require iron to maintain activity of ribonucleotide reductase, which synthesizes deoxynucleotides for DNA synthesis. Iron in the serum is bound to the iron transport protein transferrin, and uptake of iron by cells is mediated by a cell surface transferrin receptor (TfR; ref. 9). Although TfR expression is relatively limited in normal tissues, many tumor cells display high levels on their surface (1012). Therefore, TfR could be a relevant target for antibody-based therapies against tumors. Several authors have reported therapeutic approaches based on this idea using anti-TfR antibodies to kill malignant cells (1315). It has been shown that combinations of monoclonal antibodies (mAb) against TfR on human tumor cells have antiproliferative effects in vitro and in vivo (1618). However, targeting TfR, by reducing iron uptake, could reduce PHD activity, as PHDs are dependent on iron for full catalytic activity in suppressing HIF-1α levels. This raised whether the labile iron pool (LIP) regulated by TfR affects the HIF system in cells treated with anti-TfR mAbs. This could contribute to a self-limitation for therapeutic purposes. Because ascorbate is required by PHDs for catalytic activity and can enhance their activity, we wanted to assess if ascorbate could restore full PHD activity, even with low iron pools, and whether this would counteract the antitumor effects. Here, we report the effects of anti-TfR mAbs A27.15 and E2.3 on growth inhibition of human epithelial breast cancer cell lines and induction of HIF-1α and HIF transcriptional response and the effects of ascorbate on cells treated with anti-TfR mAbs.

Cell culture. Breast carcinoma MDA-468, MDA-231, and MCF-7 cell lines were obtained from the Cancer Research UK Cell Service and maintained in DMEM supplemented with 10% fetal bovine serum, 2 μmol/L l-glutamine, 50 IU/mL penicillin, and 50 μg/mL streptomycin sulfate. l-Ascorbic acid sodium salt was obtained from Sigma (Poole, United Kingdom). Anti-TfR mAbs A27.15 and E2.3 were provided as a kind gift by The Salk Institute, and isotype IgG1 antibody (Abcam, Cambridge, United Kingdom) was used as control.

Cell viability by 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt. Cells were seeded at 2.5 × 103 per well (100 μL) in 96-well plates 24 hours before experimental treatments. Cells were treated with 0.05 to 250 μg/mL anti-TfR mAbs A27.15 and E2.3 alone or in combination in triplicates. Cell viability was measured at days 2, 4, 6, and 8 by measuring metabolic conversion (by viable cells) of the dye 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt (MTS) Cell Titer 96 AQueous One Solution Cell Proliferation Assay (Promega, Southampton, United Kingdom). In each well of a 96-well plate, MTS (20 μL) was added, and plates were incubated for 2 to 4 hours in cell culture incubator. MTS assay results were read in a 96-well format plate reader by measuring absorbance at 490 nm.

Cell viability by particle Coulter counter. Cells were seeded at 2.5 × 104/mL (1 mL in 24-well plates) 24 hours before experimental treatments. Cells were treated with 10 and 50 μg/mL anti-TfR mAbs A27.15 and E2.3 alone or in combination. Cell viability was measured at days 1, 2, 3, and 4. Cells were washed in PBS, trypsinized with 200 μL trypsin/EDTA, and resuspended to a final volume of 1 mL with cell medium. Cell numbers were counted with a Coulter particle size analyzer (Beckman Coulter, High Wycombe, United Kingdom).

Vascular endothelial growth factor ELISA. Vascular endothelial growth factor (VEGF) secretion into the culture medium was measured using DuoSet ELISA Development Human VEGF Immunoassay (R&D Systems, Minneapolis, MN) following the manufacturer's protocol and 3,3′,5,5′-Tetramethylbenzidine Liquid Substrate System for ELISA (Sigma). VEGF ELISA assay results were read in a 96-well format plate reader by measuring absorbance at 450 nm with correction at 540 nm.

Western blotting. Cells were homogenized in lysis buffer (6.2 mol/L urea, 10% glycerol, 5 mmol/L DTT, 1% SDS plus protease inhibitors). Whole-cell extract (40 μg) was heat denatured at 95°C, separated by 8% or 10% SDS-PAGE, and transferred to polyvinylidene difluoride membrane. Primary antibodies were mouse anti-HIF-1α mAb (BD Transduction Laboratories, Lexington, KY), mouse anti-TfR mAb (PharMingen, Oxford, United Kingdom), mouse anti-CAIX M75 mAb (a gift from Dr. J. Pastorek, Institute of Virology, Slovak Academy of Sciences, Bratislava, Slovak Republic; ref. 19), mouse anti-PHD3 mAb (Abcam), mouse anti-BNIP3 (Sigma), and mouse anti-β-tubulin mAb (Sigma). Immunoreactivity was visualized with horseradish peroxidase–linked goat anti-mouse serum and chemiluminescence.

Cellular LIP assay. LIP in cells was measured using a modified Darbari et al. (20) and Epsztejn et al. (21) protocol based on the fluorescent probe calcein. This probe binds iron rapidly, stoichiometrically, and reversibly while forming fluorescence-quenched calcein-iron complexes. For LIP assay, cells were washed with loading medium serum and bicarbonate-free 2% RPMI with 20 mmol/L HEPES and 1 mg/mL bovine serum albumin (BSA) and incubated for 10 minutes at 37°C with loading medium containing 250 nmol/L calcein-AM (Molecular Probes, Cambridge, United Kingdom), an acetomethoxy precursor of calcein. Once inside the cells, calcein-AM is hydrolyzed by endogenous esterase into the highly negatively charged green fluorescent calcein, the fluorescence of which is quenched on binding to iron and retained in the cytoplasm. Excess calcein-AM was washed four times with loading medium, and cells were maintained at room temperature in loading medium until use. Just before measurement, loading medium was replaced with 2 mL prewarmed HBS [20 mmol/L HEPES, 150 mmol/L NaCl (pH 7.3)] containing 10 mmol/L glucose, and fluorescence was monitored at an excitation of 488 nm and emission of 517 nm using a fluorescence plate reader for 10 minutes to establish a stable baseline signal. Estimation of calcein-iron by fluorescence was obtained by adding 100 μmol/L 2,2′-bipyridyl (BIP) in ethanol, a highly permanent and high-affinity iron binding chelator, and fluorescence was measured as before for 20 minutes. Ethanol alone was used as control instead of BIP. The method for assessing LIP reflecting the concentration of cellular labile iron was done by calculating the fractional increase in fluorescence signal (ΔF) given as fluorescence arbitrary units elicited by the addition of BIP compared with background fluorescence and ethanol control-treated cells in a given number of cells.

CAIX flow cytometry. Mouse monoclonal anti-human CAIX antibody M75 was used for flow cytometric analysis. mAb M75 (10 μL) was added to tube containing a 50 μL aliquot of 0.5 × 105 to 1 × 105 cells. This working volume of mAb was set up from preliminary tests on cell lines. An isotype IgG (DAKO, Ely, United Kingdom) was used as a negative control. Cells and antibody were gently mixed and incubated at 4°C for 30 minutes. FITC-conjugated goat anti-mouse immunoglobulin antiserum (10 μL; DAKO) was added after washing with 2 mL PBS-0.1% BSA-4 mmol/L EDTA solution. After another 30-minute incubation in the dark at 4°C, cells were washed with 2 mL PBS-BSA-EDTA solution. Samples were analyzed by flow cytometry (XL Beckman, High Wycombe, United Kingdom). A typical viable cell area was gated and 1 × 104 events were counted. The isotype negative control was used to define the threshold of the background staining. Results were expressed as ratio between the median value of anti-CAIX mAb and isotype control stained sample.

Effects of anti-TfR mAb on cell viability. Cell viability was measured by Cell Titer 96 MTS assay. Addition of anti-TfR mAb E2.3 to human breast cancer cell line MDA-468 had very little effect in reducing cell proliferation, whereas anti-TfR mAb A27.15 resulted in a moderate concentration-dependent reduction in cell proliferation following 4 days of treatment (Fig. 1A). In contrast, exposure to equal concentrations (10-250 μg/mL) of both anti-TfR mAbs together resulted in concentration-dependent synergistic reduction in cell proliferation following 4 days of treatment, with up to 80% inhibition in cell proliferation following 8 days of exposure to both anti-TfR mAbs at 250 μg/mL. IgG1 control did not inhibit cell proliferation following 8 days of treatment (Fig. 1B). Synergistic inhibition of cell proliferation was also observed when MDA-231 and MCF-7 cell lines were treated for 8 days with equal concentrations (10-250 μg/mL) of both anti-TfR mAbs together (Fig. 1C).

Figure 1.

Effect of anti-TfR mAbs on viability of cancer cell lines. Cell viability was assayed by MTS. MDA-468 cells (A) were incubated with increasing concentration of anti-TfR mAbs or IgG1 antibody (B) control for the number of days indicated. Cell viability of breast cancer cell line MDA-231 and MCF-7 (C) was also tested following treatment with anti-TfR mAbs for 8 days. Combined mAb treatment included equal concentration of A27.15 and E2.3 mAbs (X μg/mL mAb = X μg/mL A27.15 and X μg/mL E2.3). Results were plotted in duplicate and dose-response curves were calculated using Graph Prism software. Bars, SE.

Figure 1.

Effect of anti-TfR mAbs on viability of cancer cell lines. Cell viability was assayed by MTS. MDA-468 cells (A) were incubated with increasing concentration of anti-TfR mAbs or IgG1 antibody (B) control for the number of days indicated. Cell viability of breast cancer cell line MDA-231 and MCF-7 (C) was also tested following treatment with anti-TfR mAbs for 8 days. Combined mAb treatment included equal concentration of A27.15 and E2.3 mAbs (X μg/mL mAb = X μg/mL A27.15 and X μg/mL E2.3). Results were plotted in duplicate and dose-response curves were calculated using Graph Prism software. Bars, SE.

Close modal

Effects of anti-TfR mAbs on HIF-1α, HIF-1α transcription targets, and TfR expression. To assess whether TfR transported iron contributes to PHD function and regulation of HIF-1α, VEGF, a downstream target gene of HIF-1α, was measured by ELISA by collecting cell medium from MDA-468 cells treated with anti-TfR mAbs (Fig. 2A and B). Addition of anti-TfR mAb E2.3 or A27.15 resulted in a moderate concentration-dependent induction of VEGF after 4 days. In contrast, exposure to equal concentrations (0.5-250 μg/mL) of both anti-TfR mAbs together resulted in concentration-dependent synergistic induction of VEGF after 4 days compared with theoretical combined value of VEGF pg/mL when cells were treated with A27.15 or E2.3 alone.

Figure 2.

Effect of anti-TfR mAbs on HIF-1α, HIF-1α transcription targets, and TfR expression in MDA-468 and MDA-435 cells. MDA-468 cells were treated with increasing concentration of anti-TfR mAbs in triplicates for 4 days, and cell medium was used for the VEGF ELISA assay. Dose-response curves were plotted as (A) total VEGF (pg) and (B) VEGF (pg) standardized to the day 4 MTS viability values. Results were plotted in duplicate and dose-response curves were calculated using Graph Prism. Bars, SE. Theoretical combined VEGF expression induced by anti-TfR mAbs was calculated as the combined VEGF expression induced by A27.15 and E2.3 alone above the PBS control. For Western blot analysis of MDA-468 (C) and MDA-231 (D), protein extracts from cells treated with anti-TfR mAbs were prepared at the indicated times. Isotype IgG1 antibody was used as control. Proteins on Western blots were probed with antibodies against HIF-1α, CAIX, PHD3, BNIP3, and TfR and against β-tubulin as control.

Figure 2.

Effect of anti-TfR mAbs on HIF-1α, HIF-1α transcription targets, and TfR expression in MDA-468 and MDA-435 cells. MDA-468 cells were treated with increasing concentration of anti-TfR mAbs in triplicates for 4 days, and cell medium was used for the VEGF ELISA assay. Dose-response curves were plotted as (A) total VEGF (pg) and (B) VEGF (pg) standardized to the day 4 MTS viability values. Results were plotted in duplicate and dose-response curves were calculated using Graph Prism. Bars, SE. Theoretical combined VEGF expression induced by anti-TfR mAbs was calculated as the combined VEGF expression induced by A27.15 and E2.3 alone above the PBS control. For Western blot analysis of MDA-468 (C) and MDA-231 (D), protein extracts from cells treated with anti-TfR mAbs were prepared at the indicated times. Isotype IgG1 antibody was used as control. Proteins on Western blots were probed with antibodies against HIF-1α, CAIX, PHD3, BNIP3, and TfR and against β-tubulin as control.

Close modal

We next wanted to determine if anti-TfR mAbs can also up-regulate HIF-1α in normoxic MDA-468 cells (Fig. 2C). Although the iron chelator desferrioxamine is well known to do this (22), the role of the normal uptake and pools for transferrin has not been studied. Treatment of MDA-468 with 50 μg/mL anti-TfR mAb A27.15 or E2.3 had a moderate effect in increasing HIF-1α protein levels following 24 hours or 4 days of treatment, whereas equal concentration of anti-TfR mAbs A27.15 and E2.3 at 50 μg/mL together significantly up-regulated HIF-1α following 24 hours or 4 days of treatment. Levels of CAIX, PHD3, and BNIP3, downstream targets of HIF-1α in MDA-468 cells, were also higher in cells treated with combined anti-TfR mAbs compared with cells treated with A27.15 mAb or E2.3 mAb alone. Similar results were also observed when MDA-231 cell line was treated with anti-TfR mAbs (Fig. 2D). HIF-1α and BNIP were up-regulated following treatment with combined anti-TfR mAbs.

It has been shown that TfR can be up-regulated by HIF-1α via the hypoxia response element in the TfR gene (23) and that TfR can also be negatively regulated post-transcriptionally by intracellular iron through iron-responsive elements in the 3′-untranslated region of the TfR mRNA (8). We therefore analyzed the effect of anti-TfR mAb on TfR expression in MDA-468 cells. Anti-TfR mAb A27.15 or E2.3 (50 μg/mL) had very little effect in changing the level of TfR expression (Fig. 2C), whereas treatment with both 50 μg/mL anti-TfR mAbs A27.15 and E2.3 together down-regulated TfR expression. Similar results were also observed when MDA-231 cell line was treated with anti-TfR mAbs (Fig. 2D). TfR was down-regulated following treatment with combined anti-TfR mAbs.

Effect of anti-TfR mAb on LIP. Because combined anti-TfR mAbs inhibited proliferation and up-regulated HIF-1α, we tested to see if this was due to a reduction in cellular LIP. MDA-468 cells were treated with combined 50 μg/mL anti-TfR mAbs A27.15 and E2.3 in triplicates for 24 hours in six-well plates. After the addition of calcein-AM and stabilization in the fluorescence readings (usually 20 averaged time points over a 10-minute period), the control cells responded to the addition of BIP by a discrete increase in fluorescence signal ΔF = 35 due to the release of calcein from iron (Fig. 3A), showing a detectable LIP, whereas anti-TfR mAb-treated cells responded with a smaller increase in fluorescence signal ΔF = 13 (Fig. 3B), showing lower level of cellular LIP. Cells treated with ethanol as control instead of BIP only produced a very small increase in fluorescence signal. From the mean values of ΔF minus the ethanol control ΔF from four samples treated or not treated with anti-TfR mAb, the addition of anti-TfR mAbs significantly (P < 0.05, Student's t test) reduced the cellular LIP by 60% (Fig. 3C).

Figure 3.

Cellular LIP assay of anti-TfR mAb-treated MDA-468 cells. Control cells (A) and 50 μg/mL anti-TfR mAbs A27.15 and E2.3 24 hour–treated cells (B) were incubated with HBS medium containing calcein-AM. After fluorescence signal stabilization (fluorescence is given as arbitrary units on Y axis), the iron chelator BIP (100 μmol/L) was added, causing an increase in fluorescence. Ethanol (EtOH) was used as control instead of BIP. The increase in fluorescence was calculated as the difference between the mean of several high fluorescent points in the BIP/ethanol-treated cells and the mean of several high fluorescent background points before the addition of BIP/ethanol given as ΔF. The mean ΔF values of four samples treated with or without anti-TfR mAbs minus the ΔF value of the ethanol control were calculated (C) and compared using Student's t test.

Figure 3.

Cellular LIP assay of anti-TfR mAb-treated MDA-468 cells. Control cells (A) and 50 μg/mL anti-TfR mAbs A27.15 and E2.3 24 hour–treated cells (B) were incubated with HBS medium containing calcein-AM. After fluorescence signal stabilization (fluorescence is given as arbitrary units on Y axis), the iron chelator BIP (100 μmol/L) was added, causing an increase in fluorescence. Ethanol (EtOH) was used as control instead of BIP. The increase in fluorescence was calculated as the difference between the mean of several high fluorescent points in the BIP/ethanol-treated cells and the mean of several high fluorescent background points before the addition of BIP/ethanol given as ΔF. The mean ΔF values of four samples treated with or without anti-TfR mAbs minus the ΔF value of the ethanol control were calculated (C) and compared using Student's t test.

Close modal

Effects of ascorbate and anti-TfR mAb on cell viability. Because we have shown that combining anti-TfR mAbs inhibits MDA-468 breast carcinoma cell proliferation and up-regulates HIF-1α by lowering the cellular LIP, we next tested to see if ascorbate can reverse the antiproliferative effects of anti-TfR mAb. Breast cancer cell line MDA-468 cells were treated with 10 or 50 μg/mL anti-TfR mAbs A27.15 and E.2.3 in the presence or absence of 25 μmol/L ascorbate, which was added every 24 hours due to its short half-life, for 4 days. Cell viability was measured using cell particle counter (Fig. 4A). Untreated and ascorbate-treated cells expanded by a factor of 9 and 7, respectively, after 4 days, whereas cells exposed to combined 10 or 50 μg/mL anti-TfR mAbs A27.15 and E2.3 had a reduction in cell proliferation. The addition of ascorbate did not reverse the antiproliferative effect of anti-TfR mAbs.

Figure 4.

Effects of ascorbate on MDA-468 cells treated with anti-TfR mAbs. Cells were treated for 4 days with 25 μmol/L ascorbate and 10 or 50 μg/mL combined anti-TfR mAbs A27.15 and E2.3. Cell viability was analyzed with the particle Coulter counter (A), expression of CAIX was analyzed by FACS (B), and expression of VEGF was analyzed by ELISA (C). Results were plotted in triplicate. Bars, SE. *, P < 0.05, statistical significance between treated and control samples (Student's t test).

Figure 4.

Effects of ascorbate on MDA-468 cells treated with anti-TfR mAbs. Cells were treated for 4 days with 25 μmol/L ascorbate and 10 or 50 μg/mL combined anti-TfR mAbs A27.15 and E2.3. Cell viability was analyzed with the particle Coulter counter (A), expression of CAIX was analyzed by FACS (B), and expression of VEGF was analyzed by ELISA (C). Results were plotted in triplicate. Bars, SE. *, P < 0.05, statistical significance between treated and control samples (Student's t test).

Close modal

Effects of ascorbate and anti-TfR mAb HIF. Because ascorbate is required by PHD enzymes for full catalytic activity in normoxic cells (24), we tested to see if ascorbate can down-regulate HIF-1α in MDA-468 cells treated with combined anti-TfR mAbs. Using the same day 4 cells, ascorbate significantly blocked the up-regulation of CAIX and VEGF in cells treated with combined anti-TfR mAbs analyzed by fluorescence-activated cell sorting (FACS) and ELISA, respectively (Fig. 4B and C). This shows that ascorbate at physiologic concentrations suppressed HIF transcription targets but did not suppress the antiproliferative effect of the combined anti-TfR mAbs.

We analyzed the time course effect of ascorbate on HIF-1α in MDA-468 cells treated with anti-TfR mAbs for up to 4 days (Fig. 5A). Treatment of MDA-468 breast carcinoma cells with both anti-TfR mAbs A27.15 and E2.3 together at 50 μg/mL induced the expression of HIF-1α protein (Fig. 5A). In the presence of 25 μmol/L ascorbate, which was added every 24 hours, levels of HIF-1α induced by anti-TfR mAbs were inhibited completely at 8 hours and 24 hours, whereas on day 4 the levels of HIF-1α were suppressed but not inhibited completely.

Figure 5.

Western blot analysis of anti-TfR mAb and ascorbate-treated breast cancer cell lines. MDA-468 (A) and MDA-231 (B) cells were treated with ascorbate and combined anti-TfR mAbs A27.15 and E2.3. Isotype IgG1 antibody was used as control. Protein extracts were prepared at the indicated times and analyzed by Western blotting. Proteins on Western blots were probed with antibodies against HIF-1α, CAIX, and BNIP3 and against β-tubulin as control.

Figure 5.

Western blot analysis of anti-TfR mAb and ascorbate-treated breast cancer cell lines. MDA-468 (A) and MDA-231 (B) cells were treated with ascorbate and combined anti-TfR mAbs A27.15 and E2.3. Isotype IgG1 antibody was used as control. Protein extracts were prepared at the indicated times and analyzed by Western blotting. Proteins on Western blots were probed with antibodies against HIF-1α, CAIX, and BNIP3 and against β-tubulin as control.

Close modal

We analyzed the expression of HIF-1α target genes by Western blotting. CAIX and BNIP3, which are both up-regulated in cells treated with anti-TfR mAbs, remained at basal levels in cells treated with anti-TfR mAb in combination with ascorbate, consistent with the effects observed on HIF-1α protein levels and data from CAIX FACS analysis and VEGF ELISA. The same effect was observed with the MDA-231 cell line (Fig. 5B). Combined mAb treatment up-regulated HIF-1α and HIF-α target gene BNIP, whereas the addition of ascorbate prevented the up-regulation of HIF-1α and HIF-α target genes by anti-TfR mAbs.

There has been a growing interest in the potential usefulness of iron deprivation as a component of cancer therapy. There are currently three main agents in clinical studies. The first is based on gallium nitrate, which interferes with the release of iron from endocytic vesicles, the second is iron chelation, and the third treatment is based on the use of mAbs or small molecules against transferrin or the TfR (2533). In this study, we showed that two different anti-TfR mAbs A27.15 and E.2.3 against the external domain of TfR synergistically inhibited proliferation of cultured breast tumor cells in vitro and can block the growth of s.c. MDA-231 tumors in vivo when 1.5 mg combined anti-TfR mAb (0.75 mg A27.15 and 0.75 mg E2.3) was i.p. injected daily for 15 days into BALB/c severe combined immunodeficient mice (data not shown). Both these anti-TfR mAbs and other anti-TfR mAbs have been shown previously to have synergistic antiproliferative effects on human cell lines cells in vitro and inhibit the growth and induce tumor regressions of established s.c. tumors of CCRF-CEM leukemia cells in nude mice (17, 34). Breast cancer cell lines have also been shown to express high levels of TfR compared with normal breast epithelial cells and are more sensitive to iron depletion, which leads to inhibition of cell proliferation (3537). The synergistic inhibition of proliferation induced by two anti-TfR mAbs is likely related to the ability of the mAbs to bind to different epitopes in TfR, resulting in receptor cross-linking, and block in iron uptake, which is required for DNA synthesis (9, 38). In addition, following treatment with combined anti-TfR mAbs, expression of TfR was down-regulated or degraded as a result of TfR cross-linking, which may also contribute to iron deprivation (17, 34).

However, the new findings in our study are that blocking the uptake of iron via the TfR resulted in the up-regulation of VEGF. This was mediated through the stabilization of HIF-1α via inhibiting proline hydroxylation by PHDs as a result of reduction in the cellular LIP. Many others have shown that the iron chelator desferrioxamine can also up-regulate VEGF (39), but this could be through direct removal of iron from intracellular compartments or enzymes. However, desferrioxamine is a poor iron chelator with poor antiproliferative activity and iron chelation efficacy (40). This study shows that extracellular iron transported through the TfR is used by the PHDs, showing a new and rapidly responsive physiologic pathway of HIF-1α response. Thus, the induction of erythropoietin in iron deficiency may be related to response to low iron transport rather than effects of low oxygen from anemia. In addition, there are insufficient iron stores or mechanisms to release iron to supply PHDs if the external source is removed.

Although the treatment of tumors with potent iron chelators or anti-TfR mAb results in HIF-1α stabilization and VEGF expression and could promote angiogenesis, iron depletion is well known to cause cell cycle arrest and apoptosis (4144). Without constant supply of iron, the conversion of ribonucleotides to deoxyribonucleotides by ribonucleotide reductase is inhibited and DNA synthesis is prevented (44). Up-regulation of HIF-1α by anti-TfR or iron chelators may also induce apoptosis via induction of BNIP3 (7, 45), a proaptotic Bcl-2 member (46). However, in an in vivo setting, up-regulation of angiogenic factors and CAIX may increase tumor vasculature and resistance to acidic stress (47).

Because ascorbate has been shown to have an effect on down-regulating the expression and activity of the HIF system (24), we decided to test if ascorbate had an effect on HIF-1α and downstream targets of HIF-1α and proliferation of cells treated with anti-TfR mAbs. Here, we showed that ascorbate at physiologic concentration down-regulated HIF-1α expression and HIF-1α transcription of VEGF, CAIX, and BNIP3 in cells treated with anti-TfR mAbs.

The mechanism of action is not completely understood. Ascorbate is thought to increase the activity of PHDs by increasing the availability of Fe(II) either free in solution or within the enzyme active site by reducing Fe(III) to Fe(II) (3, 48). In the absence of ascorbate, PHDs can rapidly catalyze the hydroxylation of proline. However, under these conditions, the enzyme becomes inactivated as the enzyme-bound Fe(II) is rapidly converted to Fe(III). Enzyme-bound Fe(III) can be reduced by ascorbate, thus reactivating the enzyme. Although ascorbate may stimulate the activity of PHDs by reducing iron from Fe(III) to Fe(II), the stimulation of oxygenase activity by ascorbate may occur through other mechanisms (e.g., by promoting completion of uncoupled cycles; ref. 49). It is believed that one role of ascorbate is to function as a surrogate-reducing substrate to rescue the enzyme in the event of uncoupled production of a ferryl [Fe(IV)] intermediate. Ascorbate may also have a role in the binding of prime substrate to the enzyme (50).

Whichever mechanism used by PHDs, our findings indicate that ascorbate has large effects on the HIF system through promotion of HIF hydroxylase activity, even when cells are starved of iron when treated with combined anti-TfR mAb. This implies that, without ascorbate or iron, PHD enzyme activity is suboptimal, allowing stabilization of HIF-1α and up-regulation of downstream targets in normoxic conditions. More surprisingly, ascorbate did not rescue cells from antiproliferative effects of anti-TfR mAbs, which suggests separate iron pools regulating these processes or different iron thresholds between PHDs and enzymes required for cell proliferation and survival.

The proapoptotic BNIP3 in anti-TfR-treated cells were shown to be down-regulated in cells treated with ascorbate, which suggests that BNIP3 plays no role in apoptosis in anti-TfR mAb-treated cells. It has been reported that ascorbate can induce apoptosis in melanoma cells, which seems to be initiated by a reduction of TfR expression, resulting in a down-regulation of iron uptake followed by an induction of apoptosis (51). Here, in breast cancer cell line MDA-468, ascorbate alone did not induce cell death or down-regulate TfR expression (data not shown).

Novel iron chelators that have high permeability, such as desferrithiocin analogues, tachpyridine, O-trensox, pyridoxal isonicotinoyl hydrazone analogues, and Triapine, have antiproliferative effect and cause a marked cell cycle arrest and apoptosis in vitro and in vivo (41, 43, 44). Consideration should be given to combining these agents with ascorbate in their clinical trials. Pharmacodynamic monitoring of plasma VEGF should be possible to evaluate the effects.

In conclusion, combination of anti-TfR mAbs inhibits tumor cell proliferation and induces HIF response, which can be blocked by physiologic concentration of ascorbate. Vitamin C deficiency is common in cancer patients (52). Clinical use of anti-TfR mAbs is likely to induce the HIF response through iron depletion, and effects may be more pronounced in patients because of the vitamin C deficiency. Our demonstration of the separate iron regulation involved in the HIF and proliferation response due to iron depletion by blocking TfR suggests that administration of anti-TfR mAb combined with physiologic doses of ascorbate may be beneficial.

Grant support: Cancer Research UK.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Prof. Peter Ratcliffe (Wellcome Institute of Human Genetics, Oxford, United Kingdom) for discussions, The Salk Institute for providing us with anti-TfR mAbs, and all members of Adrian Harris laboratory, especially Dr. Helen Knowles, for support and developing the LIP assay.

1
Harris AL. Hypoxia—a key regulatory factor in tumour growth.
Nat Rev Cancer
2002
;
2
:
38
–47.
2
Wang GL, Jiang BH, Rue EA, et al. Hypoxia-inducible factor 1 is a basic-helix-loop-helix-PAS heterodimer regulated by cellular O2 tension.
Proc Natl Acad Sci U S A
1995
;
92
:
5510
–4.
3
Schofield CJ, Ratcliffe PJ. Oxygen sensing by HIF hydroxylases.
Nat Rev Mol Cell Biol
2004
;
5
:
343
–54.
4
Elkins JM, Hewitson KS, McNeill LA, et al. Structure of factor-inhibiting hypoxia-inducible factor (HIF) reveals mechanism of oxidative modification of HIF-1α.
J Biol Chem
2003
;
278
:
1802
–6.
5
Maxwell PH. HIF-1's relationship to oxygen: simple yet sophisticated.
Cell Cycle
2004
;
3
:
156
–9.
6
Cooper CE, Lynagh GR, Hoyes KP, et al. The relationship of intracellular iron chelation to the inhibition and regeneration of human ribonucleotide reductase.
J Biol Chem
1996
;
271
:
20291
–9.
7
Le NT, Richardson DR. Iron chelators with high antiproliferative activity up-regulate the expression of a growth inhibitory and metastasis suppressor gene: a link between iron metabolism and proliferation.
Blood
2004
;
104
:
2967
–75.
8
Casey JL, Koeller DM, Ramin VC, et al. Iron regulation of transferrin receptor mRNA levels requires iron-responsive elements and a rapid turnover determinant in the 3′ untranslated region of the mRNA.
EMBO J
1989
;
8
:
3693
–9.
9
Trowbridge IS, Shackelford DA. Structure and function of transferrin receptors and their relationship to cell growth.
Biochem Soc Symp
1986
;
51
:
117
–29.
10
Gatter KC, Brown G, Trowbridge IS, et al. Transferrin receptors in human tissues: their distribution and possible clinical relevance.
J Clin Pathol
1983
;
36
:
539
–45.
11
Shindelman JE, Ortmeyer AE, Sussman HH. Demonstration of the transferrin receptor in human breast cancer tissue. Potential marker for identifying dividing cells.
Int J Cancer
1981
;
27
:
329
–34.
12
Keer HN, Kozlowski JM, Tsai YC, et al. Elevated transferrin receptor content in human prostate cancer cell lines assessed in vitro and in vivo.
J Urol
1990
;
143
:
381
–5.
13
Ng PP, Dela Cruz JS, Sorour DN, et al. An anti-transferrin receptor-avidin fusion protein exhibits both strong proapoptotic activity and the ability to deliver various molecules into cancer cells.
Proc Natl Acad Sci U S A
2002
;
99
:
10706
–11.
14
Trowbridge IS, Domingo DL. Anti-transferrin receptor monoclonal antibody and toxin-antibody conjugates affect growth of human tumour cells.
Nature
1981
;
294
:
171
–3.
15
Domingo DL, Trowbridge IS. Transferrin receptor as a target for antibody-drug conjugates.
Methods Enzymol
1985
;
112
:
238
–47.
16
Sauvage CA, Mendelsohn JC, Lesley JF, et al. Effects of monoclonal antibodies that block transferrin receptor function on the in vivo growth of a syngeneic murine leukemia.
Cancer Res
1987
;
47
:
747
–53.
17
White S, Taetle R, Seligman PA, et al. Combinations of anti-transferrin receptor monoclonal antibodies inhibit human tumor cell growth in vitro and in vivo: evidence for synergistic antiproliferative effects.
Cancer Res
1990
;
50
:
6295
–301.
18
Domingo DL, Trowbridge IS. Characterization of the human transferrin receptor produced in a baculovirus expression system.
J Biol Chem
1988
;
263
:
13386
–92.
19
Pastorek J, Pastorekova S, Callebaut I, et al. Cloning and characterization of MN, a human tumor-associated protein with a domain homologous to carbonic anhydrase and a putative helix-loop-helix DNA binding segment.
Oncogene
1994
;
9
:
2877
–8.
20
Darbari D, Loyevsky M, Gordeuk V, et al. Fluorescence measurements of the labile iron pool of sickle erythrocytes.
Blood
2003
;
102
:
357
–64.
21
Epsztejn S, Kakhlon O, Glickstein H, et al. Fluorescence analysis of the labile iron pool of mammalian cells.
Anal Biochem
1997
;
248
:
31
–40.
22
Wang GL, Semenza GL. Desferrioxamine induces erythropoietin gene expression and hypoxia-inducible factor 1 DNA-binding activity: implications for models of hypoxia signal transduction.
Blood
1993
;
82
:
3610
–5.
23
Lok CN, Ponka P. Identification of a hypoxia response element in the transferrin receptor gene.
J Biol Chem
1999
;
274
:
24147
–52.
24
Knowles HJ, Raval RR, Harris AL, et al. Effect of ascorbate on the activity of hypoxia-inducible factor in cancer cells.
Cancer Res
2003
;
63
:
1764
–8.
25
Donfrancesco A, Deb G, Dominici C, et al. Effects of a single course of deferoxamine in neuroblastoma patients.
Cancer Res
1990
;
50
:
4929
–30.
26
Seidman AD, Scher HI, Heinemann MH, et al. Continuous infusion gallium nitrate for patients with advanced refractory urothelial tract tumors.
Cancer
1991
;
68
:
2561
–5.
27
Buss JL, Greene BT, Turner J, et al. Iron chelators in cancer chemotherapy.
Curr Top Med Chem
2004
;
4
:
1623
–35.
28
Dayani PN, Bishop MC, Black K, et al. Desferoxamine (DFO)-mediated iron chelation: rationale for a novel approach to therapy for brain cancer.
J Neurooncol
2004
;
67
:
367
–77.
29
Kemp JD. Iron deprivation and cancer: a view beginning with studies of monoclonal antibodies against the transferrin receptor.
Histol Histopathol
1997
;
12
:
291
–6.
30
Trowbridge IS. Transferrin receptor as a potential therapeutic target.
Prog Allergy
1988
;
45
:
121
–46.
31
Moura IC, Lepelletier Y, Arnulf B, et al. A neutralizing monoclonal antibody (mAb A24) directed against the transferrin receptor induces apoptosis of tumor T lymphocytes from ATL patients.
Blood
2004
;
103
:
1838
–45.
32
Brown JX, Buckett PD, Wessling-Resnick M. Identification of small molecule inhibitors that distinguish between non-transferrin bound iron uptake and transferrin-mediated iron transport.
Chem Biol
2004
;
11
:
407
–16.
33
Li H, Qian ZM. Transferrin/transferrin receptor-mediated drug delivery.
Med Res Rev
2002
;
22
:
225
–50.
34
Kovar J, Naumann PW, Stewart BC, et al. Differing sensitivity of non-hematopoietic human tumors to synergistic anti-transferrin receptor monoclonal antibodies and deferoxamine in vitro.
Pathobiology
1995
;
63
:
65
–70.
35
Faulk WP, Hsi BL, Stevens PJ. Transferrin and transferrin receptors in carcinoma of the breast.
Lancet
1980
;
2
:
390
–2.
36
Pahl PM, Horwitz MA, Horwitz KB, et al. Desferri-exochelin induces death by apoptosis in human breast cancer cells but does not kill normal breast cells.
Breast Cancer Res Treat
2001
;
69
:
69
–79.
37
Yang DC, Jiang XP, Elliott RL, et al. Inhibition of growth of human breast carcinoma cells by an antisense oligonucleotide targeted to the transferrin receptor gene.
Anticancer Res
2001
;
21
:
1777
–87.
38
Trowbridge IS, Collawn J, Jing S, et al. Structure-function analysis of the human transferrin receptor: effects of anti-receptor monoclonal antibodies on tumor growth.
Curr Stud Hematol Blood Transfus
1991
;
58
:
139
–47.
39
Beerepoot LV, Shima DT, Kuroki M, et al. Up-regulation of vascular endothelial growth factor production by iron chelators.
Cancer Res
1996
;
56
:
3747
–51.
40
Simonart T, Boelaert JR, Andrei G, et al. Desferrioxamine enhances AIDS-associated Kaposi's sarcoma tumor development in a xenograft model.
Int J Cancer
2002
;
100
:
140
–3.
41
Lovejoy DB, Richardson DR. Iron chelators as anti-neoplastic agents: current developments and promise of the PIH class of chelators.
Curr Med Chem
2003
;
10
:
1035
–49.
42
Richardson DR, Tran EH, Ponka P. The potential of iron chelators of the pyridoxal isonicotinoyl hydrazone class as effective antiproliferative agents.
Blood
1995
;
86
:
4295
–43065.
43
Richardson DR, Hefter GT, May PM, et al. Iron chelators of the pyridoxal isonicotinoyl hydrazone class. III. Formation constants with calcium(II), magnesium(II) and zinc(II).
Biol Methods
1989
;
2
:
161
–7.
44
Giles FJ, Fracasso PM, Kantarjian HM, et al. Phase I and pharmacodynamic study of Triapine, a novel ribonucleotide reductase inhibitor, in patients with advanced leukemia.
Leuk Res
2003
;
27
:
1077
–83.
45
Le NT, Richardson DR. Competing pathways of iron chelation: angiogenesis or anti-tumor activity: targeting different molecules to induce specific effects.
Int J Cancer
2004
;
110
:
468
–9.
46
Ray R, Chen G, Vande Velde C, et al. BNIP3 heterodimerizes with Bcl-2/Bcl-X(L) and induces cell death independent of a Bcl-2 homology 3 (BH3) domain at both mitochondrial and nonmitochondrial sites.
J Biol Chem
2000
;
275
:
1439
–48.
47
Robertson N, Potter C, Harris AL. Role of carbonic anhydrase IX in human tumor cell growth, survival, and invasion.
Cancer Res
2004
;
64
:
6160
–5.
48
de Jong L, Albracht SP, Kemp A. Prolyl 4-hydroxylase activity in relation to the oxidation state of enzyme-bound iron. The role of ascorbate in peptidyl proline hydroxylation.
Biochim Biophys Acta
1982
;
704
:
326
–32.
49
Myllyla R, Majamaa K, Gunzler V, et al. Ascorbate is consumed stoichiometrically in the uncoupled reactions catalyzed by prolyl 4-hydroxylase and lysyl hydroxylase.
J Biol Chem
1984
;
259
:
5403
–5.
50
Wilmouth RC, Turnbull JJ, Welford RW, et al. Structure and mechanism of anthocyanidin synthase from Arabidopsis thaliana.
Structure (Camb)
2002
;
10
:
93
–103.
51
Kang JS, Cho D, Kim YI, et al. Sodium ascorbate (vitamin C) induces apoptosis in melanoma cells via the down-regulation of transferrin receptor dependent iron uptake.
J Cell Physiol
2005
;
204
:
192
–7.
52
Mayland CR, Bennett MI, Allan K. Vitamin C deficiency in cancer patients.
Palliat Med
2005
;
19
:
17
–20.