Earlier studies have shown that ascorbic acid (vitamin C) inhibits bortezomib-induced cytotoxicity against cancer cells in vitro. However, the clinical significance of vitamin C on bortezomib treatment is unclear. In this study, we examined whether daily oral intake of vitamin C inhibits antimultiple myeloma (MM) activities of bortezomib. Vitamin C, at orally achievable concentrations, inhibited in vitro MM cell cytotoxicity of bortezomib and blocked its inhibitory effect on 20S proteasome activity. Specifically, plasma collected from healthy volunteers taking 1 g/day vitamin C reduced bortezomib-induced MM cell death in vitro. This antagonistic effect of vitamin C against proteasome inhibitors is limited to the boronate class of inhibitors (bortezomib and MG262). In vivo activity of this combination treatment was then evaluated using our xenograft model of human MM in SCID (severe combined immune-deficient) mice. Bortezomib (0.1 mg/kg twice a week for 4 weeks) significantly inhibits in vivo MM cell growth, which was blocked by oral vitamin C (40 mg/kg/day). Therefore, our results for the first time show that vitamin C can significantly reduce the activity of bortezomib treatment in vivo; and importantly, suggest that patients receiving treatment with bortezomib should avoid taking vitamin C dietary supplements.
The increasing use of antioxidative dietary supplements in cancer patients has renewed concerns about their potential impact on tumor cells and therapies.1 Besides their ability to reduce treatment-related side effects2 and increase the sensitivity of tumor cells to chemotherapeutic agents,3 antioxidative agents can also decrease the efficacy of treatments by reducing their induction of oxidative stress.4 However, to date, experimental data and clinical studies have not definitively shown impact on patient's outcome.1, 5, 6
Vitamin C (ascorbic acid), a potent water-soluble antioxidant in plant and animal cells, is among the most common dietary supplement in cancer patients. The median level of vitamin C in the plasma after oral ingestion ranges between 60–100 μmol/l, but higher levels (220 μmol/l) are achievable.7 Moreover, vitamin C can accumulate as ascorbic acid, reaching millimolar intracellular concentrations depending on the characteristics of the tissue.8, 9 Although epidemiological studies support a role for vitamin C in preventing cancer,10 its influence on both tumor and activity of chemotherapeutic agents remains controversial. Several preclinical studies suggest that vitamin C can increased the efficacy of cancer agents;11, 12 however, more recent studies have shown that dehydroascorbic acid, the oxidant form of vitamin C, can antagonize the effect of conventional antineoplastic agents by preserving the stability of mitochondrial membrane potential in cancer cells.13
Bortezomib is the first proteasome inhibitor approved by the Food and Drug Administration for the treatment of relapsed refractory, relapsed and newly diagnosed multiple myeloma (MM).14, 15, 16 Moreover, combination therapies of bortezomib with doxil17, 18 are approved by the Food and Drug Administration, and with heat-shock protein 90,19 AKT20 and histone deacetylase (HDAC) inhibitors21, 22, 23, 24 are being evaluated in phase III trials. Despite the encouraging results observed with the proteasome inhibitor alone or in combination, resistance develops in most cases. Importantly, earlier studies have shown that vitamin C blocks the activity of bortezomib in vitro by its direct binding to bortezomib.25, 26, 27 In addition, production of reactive oxygen species by bortezomib28 is reduced by vitamin C.29 However, the clinical significance of vitamin C intake in MM patients receiving bortezomib treatment remains unclear. Our study confirms that vitamin C as a dietary supplement can inhibit anti-MM activity of bortezomib in vivo.
Materials and methods
MM1S cells were kindly provided by Dr Steven Rosen (Northwestern University, Chicago, IL, USA); OPM1 cells by Dr P Leif Bersagel (Mayo Clinic, Scottsdale, AZ, USA); and OPM2 cells by Dr Edward Thompson (University of Texas Medical Branch, Galveston, TX, USA). RPMI8226, U266 and H929 cells were obtained from American Type Culture Collection (Rockville, MD, USA). MM cell lines were cultured in RPMI1640 medium supplemented with 10% heat-inactivated fetal bovine serum (Sigma Chemical Co., St Louis, MO, USA), 2 μM L-glutamine and 100 U/ml penicillin (Life technologies, Inc., Grand Island, NY, USA).
Vitamin C and proteasome inhibitors
Vitamin C and L-acetyl cysteine were purchased from Sigma-Aldrich (St Louis, MO, USA). Vitamin C was dissolved in sterile distilled water and prepared immediately before use. Vitamin C for human study was purchased from commercial sources (Vitamin C, tablet 1000 mg, from CVS Pharmacy Inc., Woonsocket, RI, USA). Bortezomib was purchased from Millennium Pharmaceuticals Inc. (Cambridge, MA, USA). MG132, MG262 and lactacystin were obtained from Calbiochem (EMD Biosciences, San Diego, CA, USA). NPI-0052 was obtained from Nereus Pharmaceuticals (San Diego, CA, USA).
To assess the growth of MM cells, the rate of DNA synthesis was measured as described earlier.30 Briefly, MM cells (3 × 105 cells per well) were incubated in 96-well culture plates in the presence of vitamin C or of plasma from healthy donors for 24 or 48 h, with or without proteasome inhibitors. Cells were pulsed with 0.5 uCi/well of [3H] thymidine ([3H]-TdR; Perkin Elmer, Boston, MA, USA) during the last 6 h of culture, harvested onto glass filters with an automatic cell harvester (Cambridge Technology, Cambridge, MA, USA), and counted using an LBK Betaplate scintillation counter (Wallac, Gaithersburg, MD, USA).
Multiple myeloma cells were harvested and lysed using radioimmunoprecipitation assay lysis buffer (5 mM EDTA (ethylenediaminetetraacetic acid), 2 Mm Na3VO4, 5 mM NaF and 1 mM PMSF (phenylmethanesulphonylfluoride or phenylmethylsulphonyl fluoride)) as described before.7 The whole cell lysates were subjected to SDS-polyacrylamide gel electrophoresis, transferred to nitrocellulose membranes (BIO-RAD Laboratories, Hercules, CA, USA) and immunoblotted with antiubiquitin (Cell Signaling, Beverly, MA, USA), α-tubulin or antiactin antibodies (Santa Cruz Biotech, Santa Cruz, CA, USA).
20S proteasome inhibitor assay
Cell lysates from RPMI8226 cells were assessed for 20S proteasome activity, based on the rate of the chymotryptic subunit cleavage of a fluorescent molecule 7-amino-4-methylcoumarin (AMC) labeled pentapeptide, as described earlier.31
Quantification of vitamin C
Vitamin C levels in the plasma from healthy donors and mice were measured using a colorimetric assay kit, according to the manufacturer's instruction (Ferric Reducing Ascorbate Assay Kit, BioVision, CA, USA).
Vitamin C-rich plasma culture
An initial study to assess pharmacokinetics of ascorbic acid was carried out in healthy volunteers receiving 1 g/day of vitamin C orally for 4 consecutive days (15–20 mg/kg/day). Peripheral blood was collected on the 4th day at 0, 4 and 6 h post-treatment under the auspices of an institutional review board-approved protocol. Plasma was separated, and vitamin C level was measured as described above. In proliferation assays, RPMI8226 cells were cultured for 24 h in 96-well plates with plasma, in the presence or absence of bortezomib.
Subsequently, we carried out a dose-escalation study in healthy volunteers receiving vitamin C supplements for 7 days. The dose of vitamin C was increased every other day from 250 to 500 and 1000 mg. Blood samples were collected at baseline and 4 h after the last vitamin C intake. Plasma was separated, and vitamin C level was measured as described above.
Xenograft murine model of human MM
Fox-chase SCID (severe combined immune-deficient) mice (18–20 weeks old) were purchased from Charles River Laboratories (Wilmington, MA, USA). All animal studies were conducted according to protocols approved by the Animal Ethics Committee of the Dana-Farber Cancer Institute. An initial pharmacokinetic study to evaluate the peak level of vitamin C was carried out in a cohort of six mice. Vitamin C (40 mg/kg) was given orally for 4 days, and blood samples were collected 2 or 4 h after the last intake for measurement of vitamin C plasma levels. We next examined the effect of vitamin C on bortezomib treatment in vivo. Mice were irradiated (200 rad), and then 5 × 106 RPMI8226 cells were inoculated subcutaneously in the right flank. When tumor was measurable, mice were assigned into three cohorts (four mice each) receiving vehicle i.v. saline solution (0.9% NaCl following the schedule of bortezomib), bortezomib i.v. (0.1 mg/kg on days 1, 4, 8, 11, 14, 17, 22 and 26), or bortezomib plus vitamin C (40 mg/kg/day). Caliper measurement of the longest perpendicular tumor diameters was carried out every alternate day to estimate the tumor volume, using the following formula: 4π/3 × (width/2)2 × (length/2), representing the three-dimensional volume of an ellipse. Mice were killed at the end of the first cycle of treatment, 1 h after the 8th injection, or when tumor reached 2 cm3; plasma and tumor from mice were immediately collected and evaluated.
Statistical significance of differences in tumor growth observed between treated and untreated groups was determined using unpaired Student's two-tailed t-tests; results are presented as mean±s.e. Statistical significance of differences in proteasome activity between bortezomib versus vitamin C plus bortezomib treatment was determined using two-way analysis of variance test; all statistical analyses were carried out using Graph Pad Prism software (Graph Pad Software Inc., San Diego, CA, USA).
Vitamin C blocks bortezomib-induced growth inhibition in vitro
We first determined the effect of vitamin C on growth of MM cell lines. MM1S, OPM1, RPMI8226, U266, OPM2 and H929 MM cells were cultured with increasing doses (30–500 μM) of vitamin C for 24 h. As shown in Figure 1a, the growth of MM cell lines was minimally inhibited by vitamin C, even at high micromolar concentrations. We next examined whether vitamin C inhibits bortezomib-induced growth inhibition in vitro. RPMI8226 cells were cultured with bortezomib (5–10 nM) in the presence (30–500 μM) or absence of vitamin C. Consistent with earlier studies,25, 27 vitamin C almost completely blocked bortezomib-induced growth inhibition in a dose-dependent fashion (Figure 1b). For example, 7.5 nM of bortezomib induced 95% inhibition of [3H] thymidine uptake, whereas 125 μM vitamin C reduced its inhibition to 30%. To determine whether the antagonistic effect of vitamin C against bortezomib treatment was associated with a block in proteasome inhibition, we next examined polyubiquitinated protein levels and 20S proteasome activity. Bortezomib significantly increased the levels of polyubiquitinated protein, which was blocked in the presence of vitamin C (125 μM; Figure 1c). Moreover, vitamin C significantly (P<0.05) blocked the inhibition of the 20S proteasome activity triggered by bortezomib (Figure 1d).
Vitamin C blocks peptide boronate class proteasome inhibitors
To test the hypothesis that the inhibition of the activity of bortezomib is due to a direct binding between vitamin C and boronic acid, we next examined whether vitamin C could also inhibit other classes of proteasome inhibitors. Although vitamin C markedly blocked MG262 (peptide boronate) and bortezomib-induced growth inhibition in a dose-dependent fashion (Figure 2a), it did not abrogate the effect of NPI0052 (data not shown), lactacystin or MG132 (peptide aldehyde; Figure 2b). On the other hand, L-acetyl cysteine (LNAC), most commonly used as an antioxidant, blocked the growth inhibitory effect of all of these proteasome inhibitors (Figure 2c). These results indicate that vitamin C selectively blocks peptide boronate class of proteasome inhibitors, and that oxidative stress plays a major role in proteasome inhibitor-induced growth inhibition in MM.
Plasma from healthy donors taking vitamin C dietary supplements inhibits bortezomib activity
We next examined the relevance of vitamin C plasma levels achieved after dietary supplements on bortezomib-induced MM cell growth inhibition. A peak concentration of 150 μM vitamin C was observed 4 h after oral intake of 1 g vitamin C tablet, which decreased to baseline levels after 6 h (Figure 3a). Although vitamin C-enriched plasma alone did not trigger cytotoxicity, it markedly reduced bortezomib-induced growth inhibition. For example, bortezomib alone triggered 76% growth inhibition; however, plasma at 2 and 4 h after vitamin C intake decreased its inhibition to 56% (Figure 3a). This result suggests that vitamin C as a dietary supplement blocks bortezomib-induced cytotoxicity in MM cells in vitro.
We next evaluated the impact of lower doses of vitamin C supplements to assess whether there is a dose that does not affect bortezomib-induced cytotoxicity. Plasma samples were collected from healthy volunteers taking increasing doses of vitamin C supplements (250, 500 and 1000 mg). The average vitamin C plasma level measured was 57 umol/l (range: 38–79 umol/l), 75 umol/l (range: 64–82 umol/l), 101 umol/l (range:84–115 umol/l) and 115 umol/l (range: 96–153 umol/l) at baseline and 4 h after ingestion of 250, 500 and 100 mg of vitamin C (Figure 3b). Subsequently, RMPI8226 cells were cultured in 10% plasma collected from healthy volunteers who are not taking vitamin C supplements in the presence of different doses of vitamin C (31, 62, 125, 250 500 μM) and bortezomib (10 nM). As shown in Figure 3c, the reduction of bortezomib induced-cytotoxicity invereses linearly with the dose of vitamin C and reaches a plateau at levels higher than 125 μM of vitamin C. Our results suggest that even levels of ascorbic acid present at baseline in healthy volunteers (range:38–79 umol/l) can reduce bortezomib-induced cytotoxicity (90 versus 63% of growth inhibition, with 0 and 62 μM of vitamin C, respectively). Moreover, 125 μM of vitamin C levels, attained 4 h after ingestion of 500 mg of vitamin C supplements, strongly inhibits bortezomib cytotoxicity (90 versus 22% of growth inhibition, with 0 and 125 μM of vitamin C).
Vitamin C abrogates in vivo anti-MM activity of bortezomib
To further investigate the clinical relevance of the antagonistic effect of vitamin C against bortezomib, we next carried out in vivo studies of this combination in our xenograft mouse model of human MM. Mice were treated with vitamin C (40 mg/kg/day) orally, approximately 0.04% of their average diet, which does not affect intracellular levels of vitamin C and results in only transient fluctuations in blood concentrations.32 The plasma level of vitamin C peaked 2 h after vitamin C intake (Figure 4a), suggesting that kinetics of vitamin C in mice is different than in man (Figure 3a).
Mice were then treated with bortezomib, as described earlier. As vitamin C alone did not appreciably alter tumor growth, we compared cohorts of mice treated with vehicle control (•), bortezomib (▪) and bortezomib plus vitamin C (▴). Bortezomib significantly inhibited tumor growth (P=0.0004); however, vitamin C plus bortezomib treatment showed similar tumor growth as the control group (•; Figure 4b). For example, after the first cycle of therapy, mice treated with bortezomib alone had a significantly lower tumor volume compared with mice receiving bortezomib plus vitamin C supplement (137±39.12 versus 507.5±100.6 mm3; P=0.014), indicating that vitamin C antagonized anti-MM activity of bortezomib in vivo. Tumor tissues were collected to confirm the block of proteasome inhibition in the bortezomib plus vitamin C-treated cohort. As shown in Figure 4c, the increase in ubiquitinated proteins induced by bortezomib alone was completely blocked when vitamin C was added to bortezomib. These data confirm that vitamin C as a dietary supplement can significantly reduce the efficacy of bortezomib treatment in vivo.
Recent studies have shown that vitamin C can inhibit the in vitro activity of bortezomib through a direct binding between the hydroxyl group of the antioxidant agent and the boronic acid of the proteasome inhibitor,25, 26, 27 thereby reducing the affinity of the proteasome inhibitor for the chymotrypsin-like subunit of the proteasome. In this study, we show that vitamin C at physiological concentration blocks bortezomib both in vitro and in vivo.
After confirming the low toxic profile of vitamin C alone on MM cell lines, we showed that the inhibition on bortezomib by vitamin C is dose dependent. We also showed that this inhibitory effect of vitamin C was restricted to proteasome inhibitors containing a peptide boronic acid, consistent with earlier studies in cancer cell lines.25, 26, 27 In contrast, L-acetyl cysteine, another common antioxidant dietary supplement, inhibits cell death induced by proteasome inhibitors regardless of their chemical structure.33 These data should be carefully interpreted, as the uptake of ascorbic acid is less efficient and reproducible in vitro than in vivo.34 Nevertheless, these data suggest that the activity of bortezomib can be attenuated either by direct blocking by Vitamin C or by inhibition of the downstream pathway by L-acetyl cysteine.
Most importantly, plasma collected from donors receiving a daily supplement of vitamin C reduced the activity of bortezomib level in vitro, both its proteasome inhibitor activity and cytotoxicity. Our data suggest that even a baseline value of vitamin C present in healthy donors can inhibit the activity of bortezomib. Moreover, our in vitro findings were confirmed in vivo using our xenograft mouse model of MM: the efficacy of Bortezomib in inhibiting human MM cell growth and prolonging host survival were reduced when mice received bortezomib with vitamin C supplements. Although the precise mechanism and the kinetics of vitamin C binding to bortezomib in vivo is not completely known, our data nevertheless confirms that clinically relevant levels of vitamin C block bortezomib-induced cytotoxicity. We, therefore, suggest that vitamin C supplements should be avoided in patients receiving proteasome inhibitor therapy. In particular, we recommend that patients who are under treatment with bortezomib should not take more than 500 mg of vitamin C. Specifically, it should be avoided at least 12 h before and after bortezomib treatment as the half-life of oral ascorbic acid in plasma is 10 h.7, 35
The transport and the accumulation of vitamin C depend on several mechanisms, including diet as well as modulating factors, such as proton pump inhibitors,36 which increase gastric pH and reduce the stability and absorption of vitamin C; or dexamethasone, which induces sodium dependent vitamin C transporters and increases the accumulation of ascorbic acid, as shown in an osteoclastic cell model.37 Importantly, our study shows that vitamin C supplementation decreases the efficacy of bortezomib treatment and outcome in vivo even at relatively low concentrations. Besides vitamin C, other natural agents carrying a hydroxyl groups, such as flavonoid compounds as quercetin,2 bind and inhibit the activity of bortezomib in vitro. Taken together, these studies suggest that antioxidant supplements should be avoided in patients receiving boronic acid proteasome inhibitor therapy.
Conflict of interest
The authors declare no conflict of interest.
About this article
Preclinical evaluation of the antitumor activity of bortezomib in combination with vitamin C or with epigallocatechin gallate, a component of green tea
Cancer Chemotherapy and Pharmacology (2011)