doi:10.1016/j.freeradbiomed.2006.04.031 
Copyright © 2006 Elsevier Inc. All rights reserved.
Original Contribution
Selenite induces apoptosis in sarcomatoid malignant mesothelioma cells through oxidative stress
Gustav Nilsonnea, Xiaojuan Suna, Christina Nyströma, Anna-Klara Rundlöfa, Aristi Potamitou Fernandesa, Mikael Björnstedta and Katalin Dobra
, a, 
aDepartment of Laboratory Medicine, Karolinska University Hospital, F-46, Karolinska Institutet, S-141 86 Huddinge, Stockholm, Sweden
Received 24 January 2006;
revised 4 April 2006;
accepted 5 April 2006.
Available online 10 May 2006.
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Abstract
Malignant mesothelioma cells differentiate into sarcomatoid or epithelioid phenotypes. The sarcomatoid cell type is more resistant to chemotherapy and gives a worse prognosis. We have investigated whether selenite alone and in combination with doxorubicin induced apoptosis in variously differentiated mesothelioma cells. Selenite in concentrations that could potentially be administered to patients strongly inhibited the growth of the sarcomatoid mesothelioma cells (IC50 = 7.5 μM), whereas epithelioid cells were more sensitive to doxorubicin. Benign mesothelial cells remained largely unaffected. Selenite potentiated doxorubicin treatment. Apoptosis was the dominating mode of cell death. The toxicity of selenite was mediated by oxidative stress. Furthermore the activity of the thioredoxin system was directly dependent on the concentration of selenite. This offers a possible mechanism of action of selenite treatment. Our findings suggest that selenite is a promising new drug for the treatment of malignant mesothelioma.
Keywords: Mesothelioma; Phenotype; Drug resistance; Apoptosis; Selenium; Thioredoxin reductase; Free radicals
Abbreviations: AB, human AB serum; ASK-1, apoptosis signal regulating kinase-1; DCF, 2′,7′-dichlorodihydrofluorescein diacetate; DCFH-DA, 5(6)-carboxy-2′,7′-dichlorodihydrofluorescein diacetate; DMSO, dimethyl sulfoxide; DTNB, 5,5′-dithiobis(nitrobenzoic acid); EDTA, ethylenediaminetetraacetic acid; FACS, fluorescence-activated cell sorting; FCS, fetal calf serum; GPx, glutathione peroxidase; GR, glutathione reductase; GSH, glutathione (reduced); GSSG, glutathione (oxidized); Hepes, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; PI, propidium iodide; ROS, reactive oxygen species; SEP15, selenoprotein-15; Trx1, thioredoxin-1; TrxR1, thioredoxin reductase 1
Fig. 1. Phase-contrast photomicrographs depicting (A) the STAV-AB mesothelioma cells with epithelioid morphology and (B) the STAV-FCS mesothelioma cells with sarcomatoid morphology (bar, 50 μm).
 |
Fig. 2. Effects of selenite and doxorubicin on cell viability. Treatment with doxorubicin shows a dose- and time-dependent inhibitory effect on (A) epithelioid but not (B) sarcomatoid mesothelioma cells. Asterisks denote a statistically significant difference from the cells treated with the lowest dose (0.33 μM). (C) The dose–response curve for selenite indicates sensitivity of both mesothelioma phenotypes after 28 h, the effects being more pronounced in the sarcomatoid cells. Asterisks denote a statistically significant difference between phenotypes. (D) The response to 7.5 μM selenite was similar in three other mesothelioma cell lines and two adenocarcinoma cell lines, whereas the benign mesothelial cells remained unaffected. Asterisks denote a statistically significant difference from both populations of benign mesothelial cells. The combination of selenite and doxorubicin inhibited both (E) epithelioid and (F) sarcomatoid mesothelioma cells. Asterisks denote a statistically significant difference from the untreated control (not shown in the diagram). Error bars show the standard error of the mean.
 |
Fig. 3. Selenite induces apoptosis in mesothelioma cells. The morphological examination and FACS analysis of apoptosis correlate. Morphological determination of apoptosis in relation to concentration of selenite is shown in (A) epithelioid and (B) sarcomatoid cells. Asterisks denote a significant difference from untreated cells. The proportion of apoptotic cells, as estimated by the pyknotic index, was dose-dependent, with higher sensitivity to selenite in sarcomatoid cells than in epithelioid cells. (C, D) Apoptosis could also be demonstrated with a similar dose dependence by FACS analysis. The y axis depicts the fluorescence intensity of PI and the x axis the log of fluorescence intensity of Annexin V. Dose- and time-dependent increase in the apoptotic cells can be observed in both phenotypes (C, D). (C) The epithelioid cell population shows signs of early apoptosis (lower right quadrant) in a substantial proportion of cells only after 52 h (lower row), whereas (D) the sarcomatoid cells are apoptotic already after 28 h (upper row). The cells in the upper right quadrant are late apoptotic and correspond very well to the proportion of pyknotic cells in morphological examination in both the epithelioid and the sarcomatoid phenotypes.
Fig. 4. Selenite treatment decreases (A) the TrxR1 activity and (B) the GPx activity, in cells of the two mesothelioma phenotypes. The cells were grown for 48 h in culture medium and supplemented with different concentrations of selenite for 48 h.
Fig. 5. ROS generation by selenite in malignant mesothelioma cells. Representative micrographs showing (A) untreated controls of sarcomatoid STAV-FCS cells, (B) STAV-FCS cells treated with 5 μM selenite, (C) untreated controls of epithelioid STAV-AB cells, and (D) STAV-AB cells treated with 5 μM selenite. In the STAV-FCS cells, selenite generates ROS. The converse effect appears in the STAV-AB cells.
Fig. 6. (A) 10 μM selenite caused thiol oxidation in both epithelioid and sarcomatoid cells. Asterisk denotes a statistically significant difference compared to untreated cells of the same phenotype. (B) The toxic effect of 7.5 μM selenite was abrogated by 1 mM ascorbic acid, indicating that the toxicity of selenite is mediated by oxidative stress. Asterisks denote a significant difference between cells treated with selenite and ascorbic acid and cells treated with only selenite.
Fig. 7. Representative photomicrographs showing immunocytochemical staining for Trx1 (left column) and TrxR1 (right column) in cells exfoliated into pleural effusions. (A, B) Malignant mesothelioma and (C, D) adenocarcinoma cells demonstrated reactivity to both epitopes, whereas (E, F) benign mesothelial cells remained negative. Preincubation of antibodies with excess amounts of (G) Trx1 or (H) TrxR1 protein completely abolished all binding of antibodies to the tumor cells (bar, 50 μm).
Fig. 8. Immunohistochemical staining for (A, C) Trx1 and (B, D) TrxR1 in vivo in biphasic mesothelioma, demonstrating that the enzyme is present in native tumor tissue. The biphasic mesothelioma tissue specimen show marked reactivity to Trx1 and TrxR1 in the epithelial tumor component, whereas the spindle-shaped sarcomatoid tumor components showed less staining (bar, 50 μm). The images are representative of nine sampled tumor specimens.
 |
Fig. 9. Hypothetical model for the cytotoxic effects of selenium. The metabolic product selenide (HSe−) will redox cycle with thiols (R-SH) to generate massive nonstoichiometric amounts of reactive oxygen species (ROS). The redox cycling is increased by free thiols and consumes NADPH. The formation of ROS will cause oxidative stress that damages the cell and is a direct cause of apoptosis. It will also oxidize R-SH into disulfides (R-SS-R), thereby creating new covalent bonds within and between proteins. The tertiary structure of the affected proteins will change and they may be unable to maintain their function. Thioredoxin 1 (Trx) reduces disulfides back into thiol groups in a redox cycle that is catalyzed by TrxR1 and consumes NADPH. Apoptosis signal regulating kinase-1 (ASK-1) is a regulatory protein, which is inactive when bound to reduced Trx. The consequence of selenium mediated oxidative stress with the appearance of multiple disulfides and the parrallel decrease in TrxR1 activity will be a loss of Trx/ASK-1 binding resulting in apoptosis.
Abstract
Purchase PDF (973 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (12)
PDF (94 K)
