Tumor cells generate substantial amounts of reactive oxygen species (ROS), engendering the need to maintain high levels of antioxidants such as thioredoxin (Trx)- and glutathione (GSH)-dependent enzymes. Exacerbating oxidative stress by specifically inhibiting these types of ROS-scavenging enzymes has emerged as a promising chemotherapeutic strategy to kill tumor cells. However, potential redundancies among the various antioxidant systems may constrain this simple approach. Trx1 and thioredoxin reductase 1 (Txnrd1) are upregulated in numerous cancers, and Txnrd1 has been reported to be indispensable for tumorigenesis. However, we report here that genetic ablation of Txnrd1 has no apparent effect on tumor cell behavior based on similar proliferative, clonogenic, and tumorigenic potential. This finding reflects widespread redundancies between the Trx- and GSH-dependent systems based on evidence of a bypass to Txnrd1 deficiency by compensatory upregulation of GSH-metabolizing enzymes. Because the survival and growth of Txnrd1-deficient tumors were strictly dependent on a functional GSH system, Txnrd1−/− tumors were highly susceptible to experimental GSH depletion in vitro and in vivo. Thus, our findings establish for the first time that a concomitant inhibition of the two major antioxidant systems is highly effective in killing tumor, highlighting a promising strategy to combat cancer. Cancer Res; 70(22); 9505–14. ©2010 AACR.

Although the activated oncogene is indispensable for the maintenance of the malignant phenotype (1, 2), reliance on various pathways counteracting the oncogenesis-associated cellular stresses is commonly associated with tumorigenesis, collectively termed as “nononcogene addiction” (3, 4), which could be an Achilles' heel for cancer. Higher reactive oxygen species (ROS) production, constitutive oxidative stress, and overstrained antioxidant defense systems differentiate tumor cells from their untransformed counterpart (5), as the event of transformation is followed by increased production of ROS (69). This unique feature of tumor cells can be exploited for “selective toxicity” using the redox modifiers like l-buthionine sulfoximine (BSO), ascorbic acid, arsenic trioxide, imexon, phenethyl isothiocyanate, and motexafin gadolinium that selectively kill the tumor cells by perturbing the redox homeostasis (10).

The cellular redox homeostasis is mainly maintained by the thioredoxin (Trx)- and glutathione (GSH)-dependent systems. The Trx system, which consists of Trx, thioredoxin reductase (Txnrd), and Trx-dependent peroxidase (peroxiredoxins), is vital for antioxidant defense, DNA synthesis, and redox-regulated signal transduction, gene expression, and apoptosis (11). Whereas both Trxs and Txnrds are indispensable for embryonic development in mammals (1216), cytosolic Trx (Trx1) and cytosolic Txnrd (Txnrd1) are essential for cell proliferation (13, 17), as they provide reducing equivalents to ribonucleotide reductase and are thus involved in the maintenance of the deoxynucleotide triphosphate pool (18, 19).

The critical involvement in sustaining high proliferation rates (13, 14, 16) marks the Trx1/Txnrd1 system as a promising drug target for cancer therapy (20). In fact, Trx1 expression level positively correlates with high proliferation capacity, low apoptosis, elevated metastasis (21), increased drug resistance (22), and decreased patient survival (23). By contrast, the role of Txnrd1 in cancer is less clear (24) and could be tumor origin dependent (20). Txnrd1 inactivation by chemical inhibition (25) or small interfering RNA (siRNA)–mediated knockdown (2628) inhibits self-sufficiency of tumor cells, reverts the malignant phenotype, and sensitizes tumor cells toward the oxidative stress–inducing agents H2O2 and ionizing radiation (29). However, the lack of genetic models had thus far precluded to unequivocally assign a crucial role to Txnrd1 in tumorigenesis because of the intertwined functions of Trx1 and Txnrd1. Moreover, the degree of redundancies among various antioxidant systems in cancer has remained obscure. Using a genetic approach, we show that targeted disruption of Txnrd1 has hardly any effect on proliferation and clonogenic or tumorigenic potential of transformed cells. Furthermore, we identify distinct alterations in GSH metabolism, which compensate for the lack of Txnrd1, rendering knockout tumor cells highly susceptible to GSH deprivation.

Materials

Unless stated otherwise, all chemicals were purchased from Sigma-Aldrich. pBJ3Ω c-myc and pUC EJ6.6 H-rasv12 were kindly provided by Dr. Hartmut Land. For gene transfer, the lentivirus vector system was used as described (30). Antibodies against human Trx and Txnrd1 were kindly provided by Dr. Vadim Gladyshev (Redox Biology Center). The Tat-Cre protein was a generous gift from Dr. W. Hammerschmidt (Helmholtz Zentrum München). Antibodies against glutathione S-transferases (GST), heme oxygenase-1 (HMOX1), and aldehyde oxidase 1 (AOX1) were kindly provided by Dr. Dolph Hatfield (NIH).

Isolation of mouse embryonic fibroblasts, establishment of transformed Txnrd1−/− cell lines, and reconstitution of Txnrd1 expression

Mouse embryonic fibroblasts (MEF), isolated from conditional Txnrd1 knockout mouse embryos (13), were transformed by cotransducing them with c-myc– and H-rasv12–expressing lentiviruses. In vitro deletion of Txnrd1 was achieved by treating the cells with Tat-Cre fusion protein (31). For reconstitution, the Txnrd1−/− cells were stably transfected with the tetracycline-inducible expression vector pRTS-1-SF-Txnrd1 (32). For induction, doxycycline (Dox), a semisynthetic tetracycline that is active at lower concentration and more stable than tetracycline, was used throughout this study.

MTT assay

Cell proliferation and cytotoxicity were measured by the MTT assay as described (ref. 33; Supplementary Data).

Western blot

SDS-PAGE and Western blotting were performed as described previously (30).

Cell cycle analysis

The cell cycle analysis was done by the propidium iodide (PI) method (34), and data were analyzed by ModFitLT V3.0 software (Supplementary Data).

Soft agarose assay

Five hundred cells per well were plated in 0.3% soft agarose prepared in cell culture medium in a six-well cell culture plate and allowed to grow for 10 days. Colonies were fixed in methanol, stained with 0.5% crystal violet, and counted.

Quantitative reverse transcription-PCR

Quantitative reverse transcription-PCR (RT-PCR) was carried out by using the LightCycler FastStart DNA MasterPLUS SYBR Green I kit in combination with the LightCycler 1.5 System (Roche Diagnostic). Expression was normalized against 18S RNA and presented as fold change with respect to the parental cells.

Measurement of GSH concentration by high-performance liquid chromatography

GSH concentrations were measured by the isocratic high-performance liquid chromatography as described (35).

Determination of GSH reductase activity

GSH reductase (GR) activity was measured by monitoring the decrease in absorbance at 340 nm wavelength resulting from consumption of NADPH (36) in a 96-well plate (Supplementary Data).

Generation of Txnrd1-deficient B-cell lymphoma mice

To achieve B cell–specific deletion of Txnrd1, conditional Txnrd1 knockout mice (Txnrd1fl/fl; ref. 13) were bred with B cell–specific Cre-expressing mice (mb-1 Cre; ref. 37). To address the role of Txnrd1 in c-myc–driven B-cell lymphoma, Txnrd1+/fl;mb-1 Cre mice were crossed with λ-myc mice (38). The breeding scheme is shown in Fig. 3A.

Subcutaneous implantation of transformed cells into C57BL/6 mice

For s.c. tumor transplantation experiments, 8-week-old C57BL/6 mice were purchased from Janvier. Transformed cells (1 × 105) in a final volume of 100 μL were injected s.c. into the mice. After 12 days of tumor growth, mice were sacrificed, and the tumor mass was determined. Mice were kept under standard conditions with food and water ad libitum (Ssniff). All animal experiments were performed in compliance with the German animal welfare law and have been approved by the institutional committee on animal experimentation and the government of Oberbayern.

Treatment of Txnrd1−/− tumor-bearing mice with BSO

Transformed Txnrd1−/− cells (1 × 105) and the parental Txnrd1fl/fl cells were implanted s.c. into C57BL/6 mice. Tumors were allowed to settle for 3 days. Then, BSO (20 mmol/L) was provided in drinking water for 10 days. BSO-containing water was changed twice a week. At the end of the experiment, mice were sacrificed, and the tumor mass was determined.

Transformation and in vitro deletion of Txnrd1 in conditional Txnrd1 knockout MEFs

To address the supposedly vital role of Txnrd1 in oncogenesis by genetic means, we generated immortalized MEF cell lines from conditional Txnrd1 knockout mouse embryos (two Txnrd1fl/fl and one Txnrd1+/fl; fl, floxed). This became necessary as we had invariably failed to establish MEFs directly from Txnrd1−/− embryos (13). The gene targeting strategy previously used for targeted disruption of Txnrd1 is depicted in Fig. 1A (13). We used Txnrd1fl/fl and Txnrd1+/fl cells as parental cell lines to exclude potential clonal effects during isolation and subsequent culturing. Moreover, the floxed Txnrd1 allele was shown to behave like wild-type Txnrd1 allele (13). Therefore, we considered the Txnrd1fl/fl and Txnrd1+/fl cells as best available controls for the following experiments. The procedure of establishing transformed Txnrd1−/− cell lines is schematically depicted in Fig. 1B. Transformation of the immortalized fibroblasts was achieved by the well-established cooperative action of the two oncogenes c-myc and H-rasv12 (39). Outgrowing single-cell colonies in soft agar were individually picked and expanded to establish the clonally transformed cell lines. Expression of c-myc and H-rasV12 in these clones was confirmed by immunoblotting (Fig. 1C). Transformed Txnrd1−/− cell lines were established by treatment of the transformed cells with Tat-Cre protein followed by single-cell cloning (see Supplementary Data for detailed procedure). Knockout of Txnrd1 in the resulting clones was confirmed by PCR, RT-PCR, and immunoblotting (Fig. 1D). Transformed Txnrd1fl/fl parental cells and knockout cells derived from them are called Txnrd1fl/fl and Txnrd1−/− in the following, respectively.

Figure 1.

Disruption of Txnrd1 in MEFs by Cre ex vivo. A, schematic representation of the generation of conditional Txnrd1 knockout mice (13). Exon 15 of Txnrd1, harboring the COOH terminal catalytic center, was flanked by loxP sites (white triangles), which can be targeted by Cre recombinase. Primer pairs used for genotyping of mice and cells are depicted as black and gray arrows. B, experimental outline of how transformed Txnrd1−/− cell lines were established. C, in vitro transformation of MEFs. MEFs were transduced with c-myc– and H-rasv12–expressing lentiviruses. The event of transformation was only noticed when cells were cotransduced with c-myc and H-rasV12 lentiviruses in the form of outgrowing colonies in soft agar. Single-cell clones were picked and expanded to establish clonal cell lines. Overexpression of c-Myc and H-rasV12 was confirmed by immunoblotting (bottom). D, genotype of transformed MEF cell lines. Transformed MEF cell lines were treated with Tat-Cre protein. Successful deletion of the gene was verified by PCR using the deletion-specific primer pairs (as depicted in A). The 66-bp band corresponds to the wild-type allele, the 130-bp band corresponds to the floxed allele, and the 320-bp band is obtained after successful deletion of the loxP-flanked allele. Deletion in these cell lines was further confirmed by semiquantitative RT-PCR. RT-PCR amplification using primer pair Txnrd1 13-15 yielded no product, confirming the deletion of Txnrd1 at the mRNA level. The upstream primer pair Txnrd1 59-60 generated products in all cases, which indicated the production of truncated mRNA transcripts confirming our previous report (13). Western blot results using a Txnrd1-specific antibody showed that no truncated Txnrd1 protein was generated.

Figure 1.

Disruption of Txnrd1 in MEFs by Cre ex vivo. A, schematic representation of the generation of conditional Txnrd1 knockout mice (13). Exon 15 of Txnrd1, harboring the COOH terminal catalytic center, was flanked by loxP sites (white triangles), which can be targeted by Cre recombinase. Primer pairs used for genotyping of mice and cells are depicted as black and gray arrows. B, experimental outline of how transformed Txnrd1−/− cell lines were established. C, in vitro transformation of MEFs. MEFs were transduced with c-myc– and H-rasv12–expressing lentiviruses. The event of transformation was only noticed when cells were cotransduced with c-myc and H-rasV12 lentiviruses in the form of outgrowing colonies in soft agar. Single-cell clones were picked and expanded to establish clonal cell lines. Overexpression of c-Myc and H-rasV12 was confirmed by immunoblotting (bottom). D, genotype of transformed MEF cell lines. Transformed MEF cell lines were treated with Tat-Cre protein. Successful deletion of the gene was verified by PCR using the deletion-specific primer pairs (as depicted in A). The 66-bp band corresponds to the wild-type allele, the 130-bp band corresponds to the floxed allele, and the 320-bp band is obtained after successful deletion of the loxP-flanked allele. Deletion in these cell lines was further confirmed by semiquantitative RT-PCR. RT-PCR amplification using primer pair Txnrd1 13-15 yielded no product, confirming the deletion of Txnrd1 at the mRNA level. The upstream primer pair Txnrd1 59-60 generated products in all cases, which indicated the production of truncated mRNA transcripts confirming our previous report (13). Western blot results using a Txnrd1-specific antibody showed that no truncated Txnrd1 protein was generated.

Close modal

Txnrd1 deficiency does not impair the proliferation of transformed MEFs

Several previous reports, including our unsuccessful attempts to establish Txrnd1−/− cell lines directly from knockout embryos (13), indicated an indispensable role for Txnrd1 in proliferation (13, 17) and tumorigenesis (2528). Yet, by the procedure described above, we succeeded in generating Txnrd1−/− cell lines. A comparative analysis revealed that there was no difference in the proliferation rate of Txnrd1fl/fl and Txnrd1−/− cells (Fig. 2A). Because the cytosolic Trx system has been implicated in cell proliferation, for example, by providing reducing equivalents to ribonucleotide reductase (18, 19), we hypothesized that cell cycle phases might be altered in Txnrd1−/− cells. But there was no alteration in cell cycle distribution of Txnrd1−/− cells (G0-G1, 33.85 ± 4.7%; S, 52.33 ± 0.8%; G2-M, 13.81 ± 3.8%) compared with the Txnrd1fl/fl cells (G0-G1, 34.55 ± 1.1%; S, 48.76 ± 2.3%; G2-M, 16.68 ± 3.5%; Fig. 2B). This is in line with earlier reports that knockdown of Txnrd1 has no effect on cell cycle progression under normal culture conditions (27), whereas under serum-deprived conditions, transition through S phase is delayed (28).

Figure 2.

Loss of Txnrd1 has no effect on proliferation, cell cycle progression, clonogenicity, and tumorigenicity of transformed cells. A, Txnrd1−/− cells showed equal proliferation propensity as Txnrd1fl/fl cells measured by MTT assay. Data are representative of two independent experiments performed in quadruplicates. B, PI staining of asynchronously growing Txnrd1−/− cells showed comparable cell cycle distribution profile as Txnrd1fl/fl cells. The pie diagram represents the pooled data from two independent experiments (mean ± SD in %). C, similar numbers of colonies were formed by Txnrd1−/− cells (153 ± 21) and Txnrd1fl/fl cells (215 ± 9) per 500 cells seeded in soft agar. The experiment was done in triplicate and is representative of two independent experiments. D, s.c. transplantation of 1 × 105 transformed Txnrd1−/− and Txnrd1fl/fl cells into the flanks of C57BL/6 mice gave rise to tumors of comparable masses (0.57 ± 0.18 versus 0.59 ± 0.25, respectively; mean ± SD) after 12 d of tumor growth (n = 11; total number of tumor analyzed).

Figure 2.

Loss of Txnrd1 has no effect on proliferation, cell cycle progression, clonogenicity, and tumorigenicity of transformed cells. A, Txnrd1−/− cells showed equal proliferation propensity as Txnrd1fl/fl cells measured by MTT assay. Data are representative of two independent experiments performed in quadruplicates. B, PI staining of asynchronously growing Txnrd1−/− cells showed comparable cell cycle distribution profile as Txnrd1fl/fl cells. The pie diagram represents the pooled data from two independent experiments (mean ± SD in %). C, similar numbers of colonies were formed by Txnrd1−/− cells (153 ± 21) and Txnrd1fl/fl cells (215 ± 9) per 500 cells seeded in soft agar. The experiment was done in triplicate and is representative of two independent experiments. D, s.c. transplantation of 1 × 105 transformed Txnrd1−/− and Txnrd1fl/fl cells into the flanks of C57BL/6 mice gave rise to tumors of comparable masses (0.57 ± 0.18 versus 0.59 ± 0.25, respectively; mean ± SD) after 12 d of tumor growth (n = 11; total number of tumor analyzed).

Close modal

siRNA-mediated knockdown (2628) or chemical inhibition (25) of Txnrd1 has been reported to cause the reversal of malignant phenotype and loss of clonogenicity and tumorigenicity. At variance to this expectation, Txnrd1−/− cells were able to form colonies in soft agar to the same extent as Txnrd1fl/fl cells (153 ± 21 versus 215 ± 9; Fig. 2C). Moreover, Txnrd1−/− cells formed tumors of comparable mass (0.57 ± 0.18 g) on s.c. implantation into C57BL/6 mice like the Txnrd1fl/fl cells (0.59 ± 0.25 g; Fig. 2D). Similar results were obtained with another pair of cell lines (data not shown). This indicated that the loss of Txnrd1 had no effect on clonogenicity and tumorigenicity of transformed cells.

Txnrd1 is dispensable for c-myc–driven B-cell lymphomagenesis

Previously, we found that Txnrd1 is a target gene of the oncogene c-myc (40). To address the role of Txnrd1 in B-cell lymphomagenesis, we crossed the conditional Txnrd1 knockout mice (13) with λ-myc mice (38), which develop lymphoma of immature B-cell origin. B cell–specific deletion of Txnrd1 was achieved by crossing Txnrd1fl/fl mice with B cell–specific Cre-expressing mice (mb-1 Cre mouse; ref. 37). The mb-1–driven Cre is expressed at the very early pro-B-cell stage and is therefore the best available pan-B cell–specific Cre mouse strain. The breeding strategy and genotyping of the mice are depicted in Fig. 3A. To our surprise, there was no effect of Txnrd1 deletion on lymphoma development and survival in diseased mice as shown in Fig. 3B. Ten of 11 Txnrd1fl/fl;Cre+,myc+ mice developed tumors (median survival period of 117 ± 39.4 days) with comparable incidence with Txnrd1fl/fl;Cre,-myc+ mice (9 of 11 succumbed to the disease with a median survival of 108 ± 69.7 days; Fig. 3B). Txnrd1 knockout was confirmed by semiquantitative RT-PCR using primer pairs binding in exons 13 and 15 of murine Txnrd1 (Fig. 3C). Having shown that loss of Txnrd1 has no effect on tumor growth in two different tumor models, fibrosarcoma and myc-driven B-cell lymphoma, we hypothesized a compensatory upregulation of a yet unrecognized pathway as a plausible explanation.

Figure 3.

Txnrd1 is dispensable for B-cell lymphomagenesis. A, mice breeding scheme to generate Txnrd1-deficient B-cell lymphoma. Txnrd1fl/fl mice were crossed with B cell–specific Cre-expressing (mb-1 Cre) mice and λ-myc–expressing mice. The resulting mice were then intercrossed to generate mice harboring four transgenic alleles. Genotyping of mice was confirmed by PCR using primer pairs specific for respective transgenes. B, Txnrd1 deletion had no effect on the survival of mice succumbing from c-myc–induced lymphomas. The survival time of lymphoma-bearing Txnrd1fl/fl;Cre+;λ-myc mice (10 of 11) was comparable with that of Txnrd1fl/fl;Cre;λ-myc mice (9 of 11) with a median survival of 117 ± 39.4 d (108 ± 69.7 d in control mice; right). C, expression of Txnrd1 was analyzed in tumors by semiquantitative RT-PCR using primer pairs binding in the deleted region. 18S rRNA was used as the product of a housekeeping gene. Faint amplification products were observed in knockout tumors, which most likely derive from contaminating stroma and blood cells.

Figure 3.

Txnrd1 is dispensable for B-cell lymphomagenesis. A, mice breeding scheme to generate Txnrd1-deficient B-cell lymphoma. Txnrd1fl/fl mice were crossed with B cell–specific Cre-expressing (mb-1 Cre) mice and λ-myc–expressing mice. The resulting mice were then intercrossed to generate mice harboring four transgenic alleles. Genotyping of mice was confirmed by PCR using primer pairs specific for respective transgenes. B, Txnrd1 deletion had no effect on the survival of mice succumbing from c-myc–induced lymphomas. The survival time of lymphoma-bearing Txnrd1fl/fl;Cre+;λ-myc mice (10 of 11) was comparable with that of Txnrd1fl/fl;Cre;λ-myc mice (9 of 11) with a median survival of 117 ± 39.4 d (108 ± 69.7 d in control mice; right). C, expression of Txnrd1 was analyzed in tumors by semiquantitative RT-PCR using primer pairs binding in the deleted region. 18S rRNA was used as the product of a housekeeping gene. Faint amplification products were observed in knockout tumors, which most likely derive from contaminating stroma and blood cells.

Close modal

Upregulation of GSH metabolizing enzymes is observed under Txnrd1 deficiency

To study the compensatory mechanisms rendering Txnrd1 dispensable for cell proliferation and tumor development, we used the Txnrd1-deficient transformed MEF cell lines for the subsequent analyses. Studies with mitochondrial Txnrd (Txnrd2) revealed that Txnrd2−/− cells rapidly die in response to GSH depletion by BSO, a highly specific and irreversible inhibitor of the rate limiting enzyme γ-glutamylcysteine synthetase (γ-GCS) in GSH biosynthesis (15). We therefore treated Txnrd1−/− and Txnrd1fl/fl cells with increasing concentrations of BSO. Like Txnrd2−/− cells, Txnrd1−/− cells readily died on treatment with 10 μmol/L BSO (3.06 ± 1.12% viability), whereas the viability of Txnrd1fl/fl cells was not affected even at 3-fold higher concentrations of BSO (100.48 ± 5.55% viability at 30 μmol/L BSO; Fig. 4A). We thus conclude that GSH is essential for proliferation and survival of Txnrd1−/− cells.

Figure 4.

Txnrd1 null cells are highly sensitive to GSH deprivation due to compensatory upregulation of the GSH system. A, depletion of GSH by BSO for 72 h caused massive cell death of Txnrd1−/− cells. Depicted is the percentage of viability representative of three independent experiments with similar results (mean ± SD). B, quantitative RT-PCR analyses revealed higher expression of key enzymes of the GSH-dependent pathway (Gclc, 3.17 ± 1.5; Gclm, 7.88 ± 1.76; Gsr, 2.46 ± 0.23-fold increased in Txnrd1−/− cells compared with Txnrd1fl/fl cells). Pooled data from two independent experiments; mean ± SD (Gclc, catalytic; Gclm, modifier subunit of γ-GCS). C, total (GSH + GSSG, 209.239 ± 68.257 versus 81.211 ± 19.870 μmol/L/mg protein), reduced (GSH, 186.814 ± 59.941 versus 74.159 ± 17.455 μmol/L/mg protein), and oxidized (GSSG, 11.212 ± 4.474 versus 3.525 ± 2.637 μmol/L/mg protein) GSH were >2.5-fold higher in Txnrd1−/− cells compared with Txnrd1fl/fl cells; yet, the GSH/GSSG ratio did not change appreciably. Pooled data from four independent experiments; mean ± SD. D, GR activity was measured by the decrease in absorbance at 340 nm wavelength reflecting the consumption of NADPH (left). Txnrd1−/− cells showed ∼2.5-fold (2.053 ± 0.36 units/mg protein) higher GR activity than Txnrd1fl/fl cells (0.87 ± 0.27 units/mg protein). Right, the pooled data from three independent experiments; mean ± SD. E, immunoblotting against various phase II enzymes revealed that various GSTs and HMOX1, but not AOX1, were upregulated in Txnrd1−/− cells compared with parental cells.

Figure 4.

Txnrd1 null cells are highly sensitive to GSH deprivation due to compensatory upregulation of the GSH system. A, depletion of GSH by BSO for 72 h caused massive cell death of Txnrd1−/− cells. Depicted is the percentage of viability representative of three independent experiments with similar results (mean ± SD). B, quantitative RT-PCR analyses revealed higher expression of key enzymes of the GSH-dependent pathway (Gclc, 3.17 ± 1.5; Gclm, 7.88 ± 1.76; Gsr, 2.46 ± 0.23-fold increased in Txnrd1−/− cells compared with Txnrd1fl/fl cells). Pooled data from two independent experiments; mean ± SD (Gclc, catalytic; Gclm, modifier subunit of γ-GCS). C, total (GSH + GSSG, 209.239 ± 68.257 versus 81.211 ± 19.870 μmol/L/mg protein), reduced (GSH, 186.814 ± 59.941 versus 74.159 ± 17.455 μmol/L/mg protein), and oxidized (GSSG, 11.212 ± 4.474 versus 3.525 ± 2.637 μmol/L/mg protein) GSH were >2.5-fold higher in Txnrd1−/− cells compared with Txnrd1fl/fl cells; yet, the GSH/GSSG ratio did not change appreciably. Pooled data from four independent experiments; mean ± SD. D, GR activity was measured by the decrease in absorbance at 340 nm wavelength reflecting the consumption of NADPH (left). Txnrd1−/− cells showed ∼2.5-fold (2.053 ± 0.36 units/mg protein) higher GR activity than Txnrd1fl/fl cells (0.87 ± 0.27 units/mg protein). Right, the pooled data from three independent experiments; mean ± SD. E, immunoblotting against various phase II enzymes revealed that various GSTs and HMOX1, but not AOX1, were upregulated in Txnrd1−/− cells compared with parental cells.

Close modal

We hypothesized that upregulation or increased activity of GSH-metabolizing enzymes may compensate for the loss of Txnrd1 and thus determined the expression and activity of various key components of the GSH-dependent system in Txnrd1−/− cells. Txnrd1−/− cells exhibited a 3-fold (3.17 ± 1.5) increase in the transcript of the catalytic subunit (Gclc), an 8-fold (7.88 ± 1.76) increase of the modifier subunit (Gclm) of γ-GCS, and a 2.5-fold (2.46 ± 0.23) increase in GSH reductase (Gsr) transcripts compared with Txnrd1fl/fl cells (Fig. 4B).

These results were corroborated by measuring the intracellular GSH concentrations and GR activity in Txnrd1−/− cells. Total [GSH + oxidized GSH (GSSG)] and reduced GSH levels were 2.7-fold higher in Txnrd1−/− cells than in Txnrd1fl/fl cells (209.24 ± 68.26 versus 81.21 ± 19.87 μmol/L/mg protein for total and 186.81 ± 59.94 versus 74.16 ± 17.45 μmol/L/mg protein for reduced GSH). Likewise, GSSG was 3-fold increased (11.21 ± 4.47 versus 3.52 ± 2.64 μmol/L/mg protein) in Txnrd1−/− cells compared with Txnrd1fl/fl cells (Fig. 4C). However, this apparent increase in total (GSH + GSSG), reduced (GSH), and oxidized (GSSG) did not alter the GSH/GSSG redox couple considerably as the ratio of GSH/GSSG remained virtually constant in knockout and wild-type cells (16.6 versus 21.4). Additionally, Txnrd1−/− cells exhibited ∼2.5-fold higher GR activity than the Txnrd1fl/fl cells (2.05 ± 0.36 versus 0.87 ± 0.27 units/mg protein; Fig. 4D).

Reconstitution of Txnrd1 expression in knockout cells reverts the compensatory upregulation of GSH-metabolizing enzymes

To provide definitive proof that the increases in GSH levels and GR activity are indeed caused by the loss of Txrnd1 and are not due to clonal variations, Strep-FLAG–tagged wild-type mouse Txnrd1 (SF-Txnrd1) was expressed in Txnrd1−/− cells from the tetracycline-inducible expression vector pRTS-1 (32). Dox-inducible expression of SF-Txnrd1 was confirmed by immunoblotting (Fig. 5A). As illustrated in Fig. 5B, Txnrd1−/− cells survived treatment with 20 μmol/L BSO only when Dox (1 μg/mL) was added to the cell culture medium. This indicates that, in the absence of Txnrd1, survival and proliferation of cells relied on the GSH system.

Figure 5.

Reconstitution of Txnrd1 reverts the compensatory increase in GSH levels and GR activity. A, Strep-FLAG–tagged Txnrd1 (SF-Txnrd1) was ectopically expressed from a Dox-inducible expression vector. Dox-dependent expression of SF-Txnrd1 was confirmed by immunoblotting using a FLAG-specific antibody. B, proliferation of Txnrd1−/− cells after BSO (20 μmol/L) treatment in the presence (1 μg/mL) or absence of Dox. Only in the presence of Dox, Txnrd1−/− cells survived the BSO treatment. Data are representative of two independent experiments performed in quadruplicates; mean ± SD. C, Dox treatment caused a 3-fold reduction in total (164.143 ± 5.046 versus 444.308 ± 47.868 μmol/L/mg protein) and reduced (156.343 ± 1.023 versus 420.661 ± 41.042 μmol/L/mg protein) GSH compared with untreated cells. This was accompanied by a 3-fold reduction in GSSG levels (3.900 ± 3.034 versus 11.823 ± 3.413 μmol/L/mg protein); yet, the GSH/GSSG ratio remained unaffected before or after Dox treatment. D, Dox-treated cells revealed an ∼50% reduction in GR activity (1.4 ± 0.35 versus 2.8 ± 0.3 units/mg protein; mean ± SD).

Figure 5.

Reconstitution of Txnrd1 reverts the compensatory increase in GSH levels and GR activity. A, Strep-FLAG–tagged Txnrd1 (SF-Txnrd1) was ectopically expressed from a Dox-inducible expression vector. Dox-dependent expression of SF-Txnrd1 was confirmed by immunoblotting using a FLAG-specific antibody. B, proliferation of Txnrd1−/− cells after BSO (20 μmol/L) treatment in the presence (1 μg/mL) or absence of Dox. Only in the presence of Dox, Txnrd1−/− cells survived the BSO treatment. Data are representative of two independent experiments performed in quadruplicates; mean ± SD. C, Dox treatment caused a 3-fold reduction in total (164.143 ± 5.046 versus 444.308 ± 47.868 μmol/L/mg protein) and reduced (156.343 ± 1.023 versus 420.661 ± 41.042 μmol/L/mg protein) GSH compared with untreated cells. This was accompanied by a 3-fold reduction in GSSG levels (3.900 ± 3.034 versus 11.823 ± 3.413 μmol/L/mg protein); yet, the GSH/GSSG ratio remained unaffected before or after Dox treatment. D, Dox-treated cells revealed an ∼50% reduction in GR activity (1.4 ± 0.35 versus 2.8 ± 0.3 units/mg protein; mean ± SD).

Close modal

Next, we asked whether reexpression of Txnrd1 in Txnrd1−/− cells may reduce the augmented GSH levels and GR activity (Fig. 5C and D). The total (444.30 ± 47.86 versus 164.14 ± 5.05 μmol/L/mg protein) and reduced (420.66 ± 41.04 versus 156.34 ± 1.02 μmol/L/mg protein) GSH concentrations in untreated Txnrd1−/− cells were 2.7-fold higher than in Dox-treated cells. Moreover, Dox treatment caused a 3-fold reduction in GSSG levels (3.90 ± 3.03 μmol/L/mg protein) compared with untreated cells (11.82 ± 3.41 μmol/L/mg protein; Fig. 5C). As observed before, when GSH and GSSG levels in knockout and control cells were compared, the GSH/GSSG ratio remained virtually unchanged on reexpression of Txnrd1 (35.5 in untreated cells versus 40.0 in Dox-treated cells). The drop in GSH levels was paralleled by an ∼50% reduction in GR activity after Dox addition (2.8 ± 0.3 versus 1.4 ± 0.3 units/mg protein; Fig. 5D). The reversal of augmented GSH levels and GR activity on reintroduction of Txnrd1 further lends support to our hypothesis that Txnrd1−/− cells upregulate the GSH-dependent pathway as a compensatory mechanism for the missing Txnrd1 function.

Upregulation of phase II enzymes in Txnrd1−/− cells

Sengupta and colleagues reported that targeted removal of the gene encoding selenocysteine-specific tRNA (Trsp) in liver leads to compensatory upregulation of phase II response genes including GSTs, HMOX1, and AOX1 (41). Given that Txnrd1 is a selenoprotein, we asked whether Txnrd1 disruption may also affect phase II gene expression in a manner similar to that observed in the selenoprotein-deficient liver. In fact, compensatory upregulation of phase II enzymes (GST and HMOX1) was observed in Txnrd1−/− cells, but not of AOX1 (see Fig. 4E).

Txnrd1−/− tumors are highly susceptible to pharmacologic GSH deprivation

If the growth of Txnrd1−/− tumor cells is dependent on the GSH-dependent system also in vivo, then pharmacologic intervention of GSH synthesis and/or metabolism might be of substantial therapeutic relevance. To test this, Txnrd1−/− tumor-bearing mice (see also Fig. 2D) were treated with 20 mmol/L BSO in the drinking water (Fig. 6A). As shown in Fig. 6B and C, Txnrd1−/− tumors were highly susceptible to BSO treatment in vivo and only gave rise to very small tumors (0.062 ± 0.05 g) compared with Txnrd1−/− tumors (0.48 ± 0.42 g) in untreated mice. The absence of Txnrd1 in these tumors was confirmed by Western blotting (Fig. 6D). Interestingly, no effect of BSO treatment was observed in Txnrd1fl/fl tumor-bearing mice (0.40 ± 0.20 g treated versus 0.38 ± 0.32 g untreated tumors). These studies suggest that the growth of Txnrd1-deficient tumors relies on a functional GSH-dependent system and that targeting Txnrd1 alone may not be sufficient to restrain tumor growth.

Figure 6.

Txnrd1-deficient tumors are highly susceptible to pharmacologic GSH deprivation. A, experimental outline for BSO administration in tumor-bearing mice. Transformed cells (1 × 105) were implanted s.c. in C57BL/6 mice and allowed to settle for 3 d. After 3 d, mice were provided with BSO (20 mmol/L) in drinking water for 10 d. Mice were sacrificed on day 13, and the tumor mass was determined. B and C, Txnrd1−/− tumors were highly susceptible to BSO treatment and showed an 8-fold reduction in tumor mass on BSO treatment (0.06 ± 0.05 g) compared with untreated tumors (0.49 ± 0.43 g). Txnrd1fl/fl tumors were resistant to BSO treatment (0.40 ± 0.20; untreated tumors, 0.38 ± 0.32 g). n = 10 (mean ± SD). D, the absence of Txnrd1 was confirmed by immunblotting using a Txnrd1-specific antibody.

Figure 6.

Txnrd1-deficient tumors are highly susceptible to pharmacologic GSH deprivation. A, experimental outline for BSO administration in tumor-bearing mice. Transformed cells (1 × 105) were implanted s.c. in C57BL/6 mice and allowed to settle for 3 d. After 3 d, mice were provided with BSO (20 mmol/L) in drinking water for 10 d. Mice were sacrificed on day 13, and the tumor mass was determined. B and C, Txnrd1−/− tumors were highly susceptible to BSO treatment and showed an 8-fold reduction in tumor mass on BSO treatment (0.06 ± 0.05 g) compared with untreated tumors (0.49 ± 0.43 g). Txnrd1fl/fl tumors were resistant to BSO treatment (0.40 ± 0.20; untreated tumors, 0.38 ± 0.32 g). n = 10 (mean ± SD). D, the absence of Txnrd1 was confirmed by immunblotting using a Txnrd1-specific antibody.

Close modal

Compelling evidence established that Txnrd1 and in particular Trx1 are upregulated in numerous cancers (21). Given the procancerous function of the Trx system (2123), both Txnrd1 and Trx1 are suitable candidates for cancer therapy. Here, we addressed two important questions: (a) can Txnrd1 deficiency be compensated by intrinsic mechanisms and (b) is the combined use of drugs simultaneously targeting Trx- and the GSH-dependent systems of potential pharmacologic value to combat cancer?

We show here for the first time that Txnrd1 is dispensable for aggressive tumors and provide evidence for a widespread redundancy between the two major antioxidant systems, the GSH- and Trx-dependent systems, in tumor growth. Although Txnrd1 has been widely regarded as an Achilles' heel for cancer, we failed to recapitulate the findings of previous reports using a well-defined genetic model system (13). Surprisingly, there was no difference in the proliferation, cell cycle distribution, clonogenicity, or tumorigenicity of c-myc– and H-rasv12–transformed Txnrd1−/− cells. Moreover, we could reproduce these results in a c-myc–driven B-cell lymphoma mouse model in which we could not find any effect of loss of Txnrd1 on lymphoma incidence and survival of the mice. Furthermore, the increased susceptibility of Txnrd1−/− cells toward GSH depletion indicated that the proliferation and survival of Txnrd1−/− cells was strictly dependent on the GSH system. The increased susceptibility of tumor cells toward the combined treatment with arsenic (III) oxide that has been reported to inhibit Txnrd and BSO (25) can be explained in the light of findings from our present study. A complex cross-talk between the Trx- and GSH-dependent systems has been reported in yeast (42), Drosophila melanogaster (43), and Arabidopsis thaliana (44); however, the compensatory upregulation of the GSH-dependent system was unprecedented in the context of mammalian tumor growth.

In Txnrd1−/− cells, a 2.5- to 3-fold upregulation in total, reduced, and oxidized GSH was observed, which was reversed on Txnrd1 reexpression. Remarkably, the GSH/GSSG redox couple and thus the redox potential of GSH/GSSG remained unaltered in knockout cells, which may be due to the corresponding increase in GR activity. By upregulating GR level and activity, Txnrd1−/− cells effectively maintain the steady-state redox potential and thereby avoid oxidative stress. Moreover, upregulation of GSH-synthesizing enzymes and phase II enzymes provides an additional mechanism for the protection of Txnrd1−/− cells against oxidative stress.

It should be stressed that cells with a genetic defect in Txnrd1, as reported here, had a better chance to adapt to Txnrd1-independent growth in vitro and in vivo than wild-type cells in which the enzyme was downregulated by siRNA or targeted by Txnrd inhibitors. This may in part explain the discrepancies between our findings and previous reports. Yet, only the genetic system provides the ultimate answer whether compensatory mechanisms exist that may bypass Txnrd1 inhibition and may favor tumor relapse after an initial response to chemotherapy.

Augmented expression of GSH-metabolizing enzymes and phase II enzymes is the probable adaptation mechanism by which Txnrd1−/− cells protect themselves from cell death. This induction is presumably mediated through the nuclear factor E2-related factor 2 (Nrf2) and the Kelch-like ECH associated protein 1 (Keap1) system (45). On oxidation of Keap1, Nrf2 is liberated from the cytosolic Keap1/Nrf2 complex and translocates into the nucleus. Nrf2 induces gene expression by binding to the “antioxidant response element” found in many phase II gene promoters including Gclc, Gclm, and GR (Gsr1) and GSTs (46). Indeed, knockout of all selenoproteins in liver triggers an Nrf2-dependent upregulation of antioxidant enzymes including HO-1, GCLC, and GSTP1 and combined liver-specific disruption of Nrf2 and Trsp causes hepatocyte dysfunction and impaired survival of compound mutant mice (47). Gene expression profiling of mice lacking liver selenoprotein synthesis revealed independently that many phase II enzymes are induced by selenoprotein loss (41). Hence, one may hypothesize that induction of a fraction of phase II enzymes on Trsp removal is mediated in part by Txnrd1 deficiency.

Although cancer cells are susceptible to pharmacologic ROS insults, an upregulation of antioxidant capacity as adaptive response to increased intrinsic oxidative stress might lead to emergence of drug resistance (48). A compensatory upregulation of the GSH-dependent system and phase II enzymes on Txnrd1 deletion and the high susceptibility of Txnrd1-deficient tumors toward pharmacologic inhibition of GSH suggest that simultaneous inhibition of more than one antioxidant systems is particularly efficient in killing tumor cells. Indeed, the redox modifiers like phenethyl isothiocyanate (8) and Motexifen gadolinium (49) have been reported to effectively kill tumor cells by causing the preferential accumulation of ROS, oxidative damage in mitochondria, inactivation of redox-sensitive molecules, and massive cell death.

Simultaneous inhibition of more than one antioxidant system for cancer therapy becomes even more important with the growing evidence that the two key redox systems display widespread redundant functions (35, 50). Recently, we showed that the cystine/cysteine cycle along with Txnrd1 cooperatively rescue GSH deficiency (35). Our data thus rule out a previously alleged indispensable role of Txnrd1 in cancer development. In light of our findings, the use of redox modifiers in cancer treatment should be exercised with great care due to the extensive cross-talk among the key redox systems.

No potential conflicts of interest were disclosed.

We thank Drs. V. Gladyshev, W. Hammerschmidt, and H. Land for providing the reagents as listed in Materials and Methods; Dr. Dolph Hatfield and Brad Carlson for the generous help for providing the antibodies against the phase II enzymes; Dr. Michael Reth for providing the mb-1 Cre mice; and Drs. Matilde Maiorino and Fulvio Ursini for fruitful discussions. P.K. Mandal thanks Dr. Derrick Rossi for his extended cooperation.

Grant Support: Deutsche Forschungsgemeinschaft (DFG) CO 291/2-1 (M. Conrad), Deutsche Krebshilfe e.V. 10-1979-Bo4 (G.W. Bornkamm), and DFG-Priority Programme SPP1190 (M. Conrad and H. Beck).

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.

1
Sharma
SV
,
Settleman
J
. 
Oncogene addiction: setting the stage for molecularly targeted cancer therapy
.
Genes Dev
2007
;
21
:
3214
31
.
2
Weinstein
IB
,
Joe
A
. 
Oncogene addiction
.
Cancer Res
2008
;
68
:
3077
80; discussion 80
.
3
Solimini
NL
,
Luo
J
,
Elledge
SJ
. 
Non-oncogene addiction and the stress phenotype of cancer cells
.
Cell
2007
;
130
:
986
8
.
4
Luo
J
,
Solimini
NL
,
Elledge
SJ
. 
Principles of cancer therapy: oncogene and non-oncogene addiction
.
Cell
2009
;
136
:
823
37
.
5
Schumacker
PT
. 
Reactive oxygen species in cancer cells: live by the sword, die by the sword
.
Cancer Cell
2006
;
10
:
175
6
.
6
Kim
JH
,
Chu
SC
,
Gramlich
JL
, et al
. 
Activation of the PI3K/mTOR pathway by BCR-ABL contributes to increased production of reactive oxygen species
.
Blood
2005
;
105
:
1717
23
.
7
Sattler
M
,
Verma
S
,
Shrikhande
G
, et al
. 
The BCR/ABL tyrosine kinase induces production of reactive oxygen species in hematopoietic cells
.
J Biol Chem
2000
;
275
:
24273
8
.
8
Trachootham
D
,
Zhou
Y
,
Zhang
H
, et al
. 
Selective killing of oncogenically transformed cells through a ROS-mediated mechanism by β-phenylethyl isothiocyanate
.
Cancer Cell
2006
;
10
:
241
52
.
9
Szatrowski
TP
,
Nathan
CF
. 
Production of large amounts of hydrogen peroxide by human tumor cells
.
Cancer Res
1991
;
51
:
794
8
.
10
Engel
RH
,
Evens
AM
. 
Oxidative stress and apoptosis: a new treatment paradigm in cancer
.
Front Biosci
2006
;
11
:
300
12
.
11
Gromer
S
,
Urig
S
,
Becker
K
. 
The thioredoxin system-from science to clinic
.
Med Res Rev
2004
;
24
:
40
89
.
12
Nonn
L
,
Williams
RR
,
Erickson
RP
,
Powis
G
. 
The absence of mitochondrial thioredoxin 2 causes massive apoptosis, exencephaly, and early embryonic lethality in homozygous mice
.
Mol Cell Biol
2003
;
23
:
916
22
.
13
Jakupoglu
C
,
Przemeck
GK
,
Schneider
M
, et al
. 
Cytoplasmic thioredoxin reductase is essential for embryogenesis but dispensable for cardiac development
.
Mol Cell Biol
2005
;
25
:
1980
8
.
14
Matsui
M
,
Oshima
M
,
Oshima
H
, et al
. 
Early embryonic lethality caused by targeted disruption of the mouse thioredoxin gene
.
Dev Biol
1996
;
178
:
179
85
.
15
Conrad
M
,
Jakupoglu
C
,
Moreno
SG
, et al
. 
Essential role for mitochondrial thioredoxin reductase in hematopoiesis, heart development, and heart function
.
Mol Cell Biol
2004
;
24
:
9414
23
.
16
Bondareva
AA
,
Capecchi
MR
,
Iverson
SV
, et al
. 
Effects of thioredoxin reductase-1 deletion on embryogenesis and transcriptome
.
Free Radic Biol Med
2007
;
43
:
911
23
.
17
Soerensen
J
,
Jakupoglu
C
,
Beck
H
, et al
. 
The role of thioredoxin reductases in brain development
.
PLoS ONE
2008
;
3
:
e1813
.
18
Camier
S
,
Ma
E
,
Leroy
C
,
Pruvost
A
,
Toledano
M
,
Marsolier-Kergoat
MC
. 
Visualization of ribonucleotide reductase catalytic oxidation establishes thioredoxins as its major reductants in yeast
.
Free Radic Biol Med
2007
;
42
:
1008
16
.
19
Laurent
TC
,
Moore
EC
,
Reichard
P
. 
Enzymatic synthesis of deoxyribonucleotides: IV. Isolation and characterization of thioredoxin, the hydrogen donor from Escherichia coli B
.
J Biol Chem
1964
;
239
:
3436
44
.
20
Arner
ES
,
Holmgren
A
. 
The thioredoxin system in cancer
.
Semin Cancer Biol
2006
;
16
:
420
6
.
21
Lincoln
DT
,
Ali Emadi
EM
,
Tonissen
KF
,
Clarke
FM
. 
The thioredoxin-thioredoxin reductase system: over-expression in human cancer
.
Anticancer Res
2003
;
23
:
2425
33
.
22
Yokomizo
A
,
Ono
M
,
Nanri
H
, et al
. 
Cellular levels of thioredoxin associated with drug sensitivity to cisplatin, mitomycin C, doxorubicin, and etoposide
.
Cancer Res
1995
;
55
:
4293
6
.
23
Raffel
J
,
Bhattacharyya
AK
,
Gallegos
A
, et al
. 
Increased expression of thioredoxin-1 in human colorectal cancer is associated with decreased patient survival
.
J Lab Clin Med
2003
;
142
:
46
51
.
24
Lechner
S
,
Muller-Ladner
U
,
Neumann
E
, et al
. 
Thioredoxin reductase 1 expression in colon cancer: discrepancy between in vitro and in vivo findings
.
Lab Invest
2003
;
83
:
1321
31
.
25
Lu
J
,
Chew
EH
,
Holmgren
A
. 
Targeting thioredoxin reductase is a basis for cancer therapy by arsenic trioxide
.
Proc Natl Acad Sci U S A
2007
;
104
:
12288
93
.
26
Gan
L
,
Yang
XL
,
Liu
Q
,
Xu
HB
. 
Inhibitory effects of thioredoxin reductase antisense RNA on the growth of human hepatocellular carcinoma cells
.
J Cell Biochem
2005
;
96
:
653
64
.
27
Yoo
MH
,
Xu
XM
,
Carlson
BA
,
Gladyshev
VN
,
Hatfield
DL
. 
Thioredoxin reductase 1 deficiency reverses tumor phenotype and tumorigenicity of lung carcinoma cells
.
J Biol Chem
2006
;
281
:
13005
8
.
28
Yoo
MH
,
Xu
XM
,
Carlson
BA
,
Patterson
AD
,
Gladyshev
VN
,
Hatfield
DL
. 
Targeting thioredoxin reductase 1 reduction in cancer cells inhibits self-sufficient growth and DNA replication
.
PLoS ONE
2007
;
2
:
e1112
.
29
Smart
DK
,
Ortiz
KL
,
Mattson
D
, et al
. 
Thioredoxin reductase as a potential molecular target for anticancer agents that induce oxidative stress
.
Cancer Res
2004
;
64
:
6716
24
.
30
Seiler
A
,
Schneider
M
,
Forster
H
, et al
. 
Glutathione peroxidase 4 senses and translates oxidative stress into 12/15-lipoxygenase dependent- and AIF-mediated cell death
.
Cell Metab
2008
;
8
:
237
48
.
31
Peitz
M
,
Pfannkuche
K
,
Rajewsky
K
,
Edenhofer
F
. 
Ability of the hydrophobic FGF, basic TAT. peptides to promote cellular uptake of recombinant Cre recombinase: a tool for efficient genetic engineering of mammalian genomes
.
Proc Natl Acad Sci U S A
2002
;
99
:
4489
94
.
32
Bornkamm
GW
,
Berens
C
,
Kuklik-Roos
C
, et al
. 
Stringent doxycycline-dependent control of gene activities using an episomal one-vector system
.
Nucleic Acids Res
2005
;
33
:
e137
.
33
Mosmann
T
. 
Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays
.
J Immunol Methods
1983
;
65
:
55
63
.
34
Fried
J
,
Perez
AG
,
Clarkson
BD
. 
Flow cytofluorometric analysis of cell cycle distributions using propidium iodide. Properties of the method and mathematical analysis of the data
.
J Cell Biol
1976
;
71
:
172
81
.
35
Mandal
PK
,
Seiler
A
,
Perisic
T
, et al
. 
System x(c)- and thioredoxin reductase 1 cooperatively rescue glutathione deficiency
.
J Biol Chem
2010
;
285
:
22244
53
.
36
Mavis
RD
,
Stellwagen
E
. 
Purification and subunit structure of glutathione reductase from bakers' yeast
.
J Biol Chem
1968
;
243
:
809
14
.
37
Hobeika
E
,
Thiemann
S
,
Storch
B
, et al
. 
Testing gene function early in the B cell lineage in mb1-Cre mice
.
Proc Natl Acad Sci U S A
2006
;
103
:
13789
94
.
38
Kovalchuk
AL
,
Qi
CF
,
Torrey
TA
, et al
. 
Burkitt lymphoma in the mouse
.
J Exp Med
2000
;
192
:
1183
90
.
39
Land
H
,
Parada
LF
,
Weinberg
RA
. 
Tumorigenic conversion of primary embryo fibroblasts requires at least two cooperating oncogenes
.
Nature
1983
;
304
:
596
602
.
40
Schuhmacher
M
,
Kohlhuber
F
,
Holzel
M
, et al
. 
The transcriptional program of a human B cell line in response to Myc
.
Nucleic Acids Res
2001
;
29
:
397
406
.
41
Sengupta
A
,
Carlson
BA
,
Weaver
JA
, et al
. 
A functional link between housekeeping selenoproteins and phase II enzymes
.
Biochem J
2008
;
413
:
151
61
.
42
Muller
EG
. 
A glutathione reductase mutant of yeast accumulates high levels of oxidized glutathione and requires thioredoxin for growth
.
Mol Biol Cell
1996
;
7
:
1805
13
.
43
Cheng
Z
,
Arscott
LD
,
Ballou
DP
,
Williams
CH
 Jr.
The relationship of the redox potentials of thioredoxin and thioredoxin reductase from Drosophila melanogaster to the enzymatic mechanism: reduced thioredoxin is the reductant of glutathione in Drosophila
.
Biochemistry
2007
;
46
:
7875
85
.
44
Reichheld
JP
,
Khafif
M
,
Riondet
C
,
Droux
M
,
Bonnard
G
,
Meyer
Y
. 
Inactivation of thioredoxin reductases reveals a complex interplay between thioredoxin and glutathione pathways in Arabidopsis development
.
Plant Cell
2007
;
19
:
1851
65
.
45
Itoh
K
,
Wakabayashi
N
,
Katoh
Y
, et al
. 
Keap1 represses nuclear activation of antioxidant responsive elements by Nrf2 through binding to the amino-terminal Neh2 domain
.
Genes Dev
1999
;
13
:
76
86
.
46
Thimmulappa
RK
,
Mai
KH
,
Srisuma
S
,
Kensler
TW
,
Yamamoto
M
,
Biswal
S
. 
Identification of Nrf2-regulated genes induced by the chemopreventive agent sulforaphane by oligonucleotide microarray
.
Cancer Res
2002
;
62
:
5196
203
.
47
Suzuki
T
,
Kelly
VP
,
Motohashi
H
, et al
. 
Deletion of the selenocysteine tRNA gene in macrophages and liver results in compensatory gene induction of cytoprotective enzymes by Nrf2
.
J Biol Chem
2008
;
283
:
2021
30
.
48
Trachootham
D
,
Alexandre
J
,
Huang
P
. 
Targeting cancer cells by ROS-mediated mechanisms: a radical therapeutic approach?
Nat Rev Drug Discov
2009
;
8
:
579
91
.
49
Hashemy
SI
,
Ungerstedt
JS
,
Zahedi Avval
F
,
Holmgren
A
. 
Motexafin gadolinium, a tumor-selective drug targeting thioredoxin reductase and ribonucleotide reductase
.
J Biol Chem
2006
;
281
:
10691
7
.
50
Tan
SX
,
Greetham
D
,
Raeth
S
,
Grant
CM
,
Dawes
IW
,
Perrone
GG
. 
The thioredoxin-thioredoxin reductase system can function in vivo as an alternative system to reduce oxidized glutathione in Saccharomyces cerevisiae
.
J Biol Chem
2010
;
285
:
6118
26
.