COMMENTARY
COMMENTARY
Systems approach to metalbased pharmacology Isolda Romero-Canelón and Peter J. Sadler1 Department of Chemistry, University of Warwick, Coventry CV4 7AL, United Kingdom
At least 20 elements of the periodic table are essential to humans, but understanding their biochemistry presents many challenges. The genetic codes for metals such as iron, copper, and zinc are readily identified, but the same cannot be said for vanadium, chromium, nickel, or tin, metals frequently in mineral supplements. The genetic codes for metals are not just for the elements, they are for particular species of the elements. The metal oxidation state, types and numbers of coordinated ligands, as well as the coordination geometry are crucial for recognition by specific uptake, transport, and delivery proteins. Understanding and controlling the biochemistry of metal ions requires speciation on timescales of nanoseconds to years (kinetics of ligand substitution), consideration of metal–ligand affinities (thermodynamics, equilibrium constants), as well as redox potentials (1, 2). Medicines use not only essential, but also nonessential elements. For example, bismuth, the heaviest stable element, has been used for the treatment of infectious diseases and gastrointestinal conditions for over three
centuries. Preparations such as bismuth subsalicylate (Pepto-Bismol), colloidal bismuth subcitrate (CBS) (De-Nol), and ranitidine bismuth citrate (Pylorid) are widely used today. However, why is bismuth much more toxic to microbes such as Helicobacter pylori compared with eukaryotic cells? In PNAS, Hong et al. (3) show that a systems pharmacology approach can provide new insights, an approach that, if applied more generally, will greatly advance the rational design of metallodrugs. The speciation of bismuth itself is exceptionally challenging: only one natural isotope (209Bi), a quadrupolar nucleus that is of little use for NMR, with no unpaired spins, no useful metal-centered electronic transitions, and a highly variable and fluxional coordination sphere with coordination numbers of 3–9. These features and the strong acidity of Bi3+OH2 (pKa ∼1.5) lead to complicated stoichiometries for its compounds (“sub” salts). Bi3+ is a borderline class “a/b,” “‘hard/soft’” metal ion, binding strongly not only to (hard) oxygen ligands, but also to (soft) sulfur, especially thiolates (4).
Fig. 1. Involvement of GSH, some GSH-dependent enzymes, and membrane pumps in cellular reactions of As(III), Pt(II), Bi(III), and Au(I) drugs and essential Cu(I). Lysosomal deposits (gold aurosomes) or intranuclear inclusion bodies (Bi) are curious features also seen for some other metals. Pt(II) exchanges ligands more slowly and its reactions have some different features. These pathways can detoxify metals; their inhibition might be built into, e.g., an anticancer strategy. Speciation of metal ions in intact cells is a major challenge. GSH also has redox functions in cells.
www.pnas.org/cgi/doi/10.1073/pnas.1503858112
There is an abundance of thiolate sulfur in cells. Most cells contain 2–14 mM glutathione, the tripeptide γ-L-glutamyl-L-cysteinylglycine (GSH), not only in the cytoplasm but also in other organelles, e.g., mitochondria (5). Hong et al. have shown that uptake of Bi into HK-2 human proximal tubular cells and bacteria (Escherichia coli) occurs by passive diffusion and is biphasic, the second phase mirroring exponential growth of HK-2, but not E. coli. Moreover, Bi uptake greatly increases GSH levels (approximately four times). Bi uptake is “self-propelled” by a feedback loop. Stimulation of GSH biosynthesis by an external supply of cystine as a cysteine precursor led to a remarkable accumulation of 34 mM Bi in HK-2 cells. In contrast, on depletion of GSH using L-buthionine sulfoximine, an inhibitor of γ-glutamyl-cysteine synthase (glutamate-cysteine ligase), the first step in GSH synthesis, CBS gave rise to HK-2 cell degradation and nucleus malformation. Coordination of Bi3+ to GSH in cells, forming Bi(SG)3, triggers other events. This (conjugation) step may be aided by GSTs, which constitute up to 10% of cytosolic proteins. The wide family of multidrug resistanceassociated proteins (MRPs) mediate ATPdependent transport of GSH conjugates, so decreasing cellular drug accumulation (conferring resistance). MRPs also pump conjugates into perinuclear vesicles, where Bi is seen as dense black deposits, probably Bi sulfide (inclusion bodies). The nature of Bi (as well as Ag and Pb) “intranuclear inclusion bodies” has long been a puzzle (6, 7). Importantly, detoxification of Bi in human cells is not so effective in bacteria where GSH is less prevalent. Where does the sulfide come from? —via the action of cysteine desulfurase on cysteine as it does for ferredoxins? Perhaps it arises from cleavage of the γ-glutamyl residue of Bi(SG)3 by γ-glutamyltransferase (γ-GT), which initiates mercapturic acid synthesis. In many anaerobic bacteria, GSH is absent and cysteine is the major thiol. Hence Bi drugs can be effective and selective against Author contributions: I.R.-C. and P.J.S. wrote the paper. The authors declare no conflict of interest. See companion article on page 3211 in issue 11 of volume 112. 1
To whom correspondence should be addressed. Email: p.j.sadler@ warwick.ac.uk.
PNAS | April 7, 2015 | vol. 112 | no. 14 | 4187–4188
pathogens that cause syphilis and dysentery, for example. Many other soft metal ions have a high affinity for thiolate sulfur and GSH sequestration pathways might be expected to be involved in their metabolism too. These include anticancer drugs based on As(III) and Pt(II), antiarthritic Au(I) drugs, as well as the essential metal ion Cu(I) (Fig. 1). The As(III) drug Trisenox [As2O3, As(OH)3 in aqueous solution], the drug of choice for promyelocytic leukemia, forms GSH conjugates that have been related to poor patient outcome. These GSH-adducts are GST-substrates easily detoxified by cancer cells (8). In contrast, the drug candidate Zinapar (darinaparsin, ZIO-101), dimethylarsenic conjugated to glutathione, takes advantage of the excessive cystine/cysteine requirement of cancer cells. This As(III) complex appears to be transformed by γ-GT at the cell surface with formation of the dimethylarsino-cysteine adduct (DMAC), which is readily taken up via cystine/cysteine importers. Hence increased cellular accumulation of As from ZIO-101 occurs. This complex might be favored for multidrug-resistant tumors where chemoresistance is mediated by high GSH levels (9). In the case of Pt(II)-based anticancer drugs, GSH may be responsible for the inactivation of compounds such as cisplatin, cis-diamminedichloroplatinum(II), or its derivatives carboplatin and oxaliplatin. Indeed, at least 60% of cellular Pt is present as GSH-adducts and increased levels of intracellular GSH are directly linked to the development of resistance to Pt(II) drugs in lung and ovarian cancers (10). For ovarian cancer, proteomic studies have related differentially expressed GSTs to Pt chemoresistance. Both cisplatin and carboplatin generate adducts in which Pt is bound to the CXXC copper-binding motif of methyl-CpG binding domain protein 2 (MBD2), and then excreted by ATP7B, a Cu exporter (11). Injectable Au(I) thiolate drugs such as Myochrysine (aurothiomalate) and the oral drug Ridaura [auranofin; (tetraacetylthioglucose)Au(PEt3)] are used for treating rheumatoid arthritis. It has long been known that gold accumulates in lysosomes in cells as dense deposits (aurosomes), widely believed to be sulfur-rich (12), but they are not well characterized. In view of the observations on Bi by Hong et al. (3), it now seems to be the time to reexamine aurosomes. Myochrysine
4188 | www.pnas.org/cgi/doi/10.1073/pnas.1503858112
direct binding and subsequent inhibition of GST (15). Interestingly, a member of the GST family, TDR1, activates antiparasitic antimonial prodrugs by deglutathionylation of Sb(V)-glutathione adducts with reduction and release of Sb(III) (16). There is a link to the metabolism of essential metal ions. GSH stimulates the uptake of copper via the transporter Ctr1. Initial formation of [CuI(SG)3]2− may be followed by delivery to P1B-type ATPases and superoxide dismutase mediated by the dynamic transfer of Cu(I) between cysteine thiolates of chaperones Atox1 and CCS in the cytoplasm (11). The field of bioinorganic chemistry and its applications in medicine are fast-growing
areas of research that use the rich structural diversity of metallic scaffolds for drug design. Metal-based drugs can provide in-cell chemical reactions that are not accessible to purely organic drugs, accompanied by multitargeting that can combat resistance (17). However, there is an urgent need for further development of methods to elucidate the spatial and temporal speciation of metallodrugs. This requires a multidisciplinary effort with contributions like those of Hong et al. (3) to understand fully the implications of the metal center in the metabolism and pharmacokinetic/dynamic properties of candidate drugs. It is evident from this brief commentary that progress in the design of metallodrugs requires a systems pharmacology approach. Not only do we need a good understanding of the kinetics and thermodynamics of metal– ligand exchange and redox reactions, but also of the responses of cells in terms of attack on multiple targets, self-propelled transport, and delivery processes related to feedback loops (Fig. 1). For metallodrugs, it will often not be appropriate to describe their mechanisms of action in terms of a single target site. This “polypharmacology” offers potential advantages because it is now becoming apparent that resistance to singletarget drugs can readily develop. The introduction of systems pharmacology approaches to unraveling the mechanism of action of metallodrugs should lead to advances in their design and in their use so as to avoid unwanted side effects.
1 Chellan P, Sadler PJ (2015) The elements of life and medicines. Philos Trans A Math Phys Eng Sci 373(2037):20140182. 2 Barry NPE, Sadler PJ (2013) Exploration of the medical periodic table: Towards new targets. Chem Commun (Camb) 49(45): 5106–5131. 3 Hong Y, Lai Y-T, Chan GC-F, Sun H (2015) Glutathione and multidrug resistance protein transporter mediate a self-propelled disposal of bismuth in human cells. Proc Natl Acad Sci USA 112(11):3211–3216. 4 Sun H (2011) Biological Chemistry of Arsenic, Antimony and Bismuth (Wiley, Chichester, UK). 5 Marí M, et al. (2013) Mitochondrial glutathione: Features, regulation and role in disease. Biochim Biophys Acta 1830(5): 3317–3328. 6 Wachstein M (1949) Studies on inclusion bodies; acid-fastness of nuclear inclusion bodies that are induced by ingestion of lead and bismuth. Am J Clin Pathol 19(7):608–614. 7 Fowler BA, Goyer RA (1975) Bismuth localization within nuclear inclusions by x-ray microanalysis. Effects of accelerating voltage. J Histochem Cytochem 23(10):722–726. 8 Sui M, Zhang Z, Zhou J (2014) Inhibition factors of arsenic trioxide therapeutic effects in patients with acute promyelocytic leukemia. Chin Med J (Engl) 127(19):3503–3506. 9 Garnier N, et al. (2014) The novel arsenical darinaparsin is transported by cystine importing systems. Mol Pharmacol 85(4): 576–585.
10 Sawers L, et al. (2014) Glutathione S-transferase P1 (GSTP1) directly influences platinum drug chemosensitivity in ovarian tumour cell lines. Br J Cancer 111(6):1150–1158. 11 Dolgova NV, Nokhrin S, Yu CH, George GN, Dmitriev OY (2013) Copper chaperone Atox1 interacts with the metal-binding domain of Wilson’s disease protein in cisplatin detoxification. Biochem J 454(1): 147–156. 12 Ghadially FN (1979) The aurosome. J Rheumatol Suppl 5:45–50. 13 Reglinski J, et al. (1997) Myocrisin-mediated oxidative stress. Clin Chim Acta 268(1-2):85–99. 14 Casini A (2012) Exploring the mechanisms of metal-based pharmacological agents via an integrated approach. J Inorg Biochem 109:97–106. 15 De Luca A, Hartinger CG, Dyson PJ, Lo Bello M, Casini A (2013) A new target for gold(I) compounds: Glutathione-S-transferase inhibition by auranofin. J Inorg Biochem 119(I):38–42. 16 Fyfe PK, Westrop GD, Silva AM, Coombs GH, Hunter WN (2012) Leishmania TDR1 structure, a unique trimeric glutathione transferase capable of deglutathionylation and antimonial prodrug activation. Proc Natl Acad Sci USA 109(29): 11693–11698. 17 Romero-Canelón I, Sadler PJ (2013) Next-generation metal anticancer complexes: Multitargeting via redox modulation. Inorg Chem 52(21):12276–12291.
inhibits GSH-peroxidase in vivo, which contributes to increased oxidative stress (13). Auranofin exerts its antiinflammatory activity by inhibiting mitochondrial thioredoxin reductase (TrxR) (14); nonetheless, its anticancer activity can be enhanced by
Hong et al. show that a systems pharmacology approach can provide new insights, an approach that, if applied more generally, will greatly advance the rational design of metallodrugs.
Romero-Canelón and Sadler
COMMENTARY
Systems approach to metalbased pharmacology Isolda Romero-Canelón and Peter J. Sadler1 Department of Chemistry, University of Warwick, Coventry CV4 7AL, United Kingdom
At least 20 elements of the periodic table are essential to humans, but understanding their biochemistry presents many challenges. The genetic codes for metals such as iron, copper, and zinc are readily identified, but the same cannot be said for vanadium, chromium, nickel, or tin, metals frequently in mineral supplements. The genetic codes for metals are not just for the elements, they are for particular species of the elements. The metal oxidation state, types and numbers of coordinated ligands, as well as the coordination geometry are crucial for recognition by specific uptake, transport, and delivery proteins. Understanding and controlling the biochemistry of metal ions requires speciation on timescales of nanoseconds to years (kinetics of ligand substitution), consideration of metal–ligand affinities (thermodynamics, equilibrium constants), as well as redox potentials (1, 2). Medicines use not only essential, but also nonessential elements. For example, bismuth, the heaviest stable element, has been used for the treatment of infectious diseases and gastrointestinal conditions for over three
centuries. Preparations such as bismuth subsalicylate (Pepto-Bismol), colloidal bismuth subcitrate (CBS) (De-Nol), and ranitidine bismuth citrate (Pylorid) are widely used today. However, why is bismuth much more toxic to microbes such as Helicobacter pylori compared with eukaryotic cells? In PNAS, Hong et al. (3) show that a systems pharmacology approach can provide new insights, an approach that, if applied more generally, will greatly advance the rational design of metallodrugs. The speciation of bismuth itself is exceptionally challenging: only one natural isotope (209Bi), a quadrupolar nucleus that is of little use for NMR, with no unpaired spins, no useful metal-centered electronic transitions, and a highly variable and fluxional coordination sphere with coordination numbers of 3–9. These features and the strong acidity of Bi3+OH2 (pKa ∼1.5) lead to complicated stoichiometries for its compounds (“sub” salts). Bi3+ is a borderline class “a/b,” “‘hard/soft’” metal ion, binding strongly not only to (hard) oxygen ligands, but also to (soft) sulfur, especially thiolates (4).
Fig. 1. Involvement of GSH, some GSH-dependent enzymes, and membrane pumps in cellular reactions of As(III), Pt(II), Bi(III), and Au(I) drugs and essential Cu(I). Lysosomal deposits (gold aurosomes) or intranuclear inclusion bodies (Bi) are curious features also seen for some other metals. Pt(II) exchanges ligands more slowly and its reactions have some different features. These pathways can detoxify metals; their inhibition might be built into, e.g., an anticancer strategy. Speciation of metal ions in intact cells is a major challenge. GSH also has redox functions in cells.
www.pnas.org/cgi/doi/10.1073/pnas.1503858112
There is an abundance of thiolate sulfur in cells. Most cells contain 2–14 mM glutathione, the tripeptide γ-L-glutamyl-L-cysteinylglycine (GSH), not only in the cytoplasm but also in other organelles, e.g., mitochondria (5). Hong et al. have shown that uptake of Bi into HK-2 human proximal tubular cells and bacteria (Escherichia coli) occurs by passive diffusion and is biphasic, the second phase mirroring exponential growth of HK-2, but not E. coli. Moreover, Bi uptake greatly increases GSH levels (approximately four times). Bi uptake is “self-propelled” by a feedback loop. Stimulation of GSH biosynthesis by an external supply of cystine as a cysteine precursor led to a remarkable accumulation of 34 mM Bi in HK-2 cells. In contrast, on depletion of GSH using L-buthionine sulfoximine, an inhibitor of γ-glutamyl-cysteine synthase (glutamate-cysteine ligase), the first step in GSH synthesis, CBS gave rise to HK-2 cell degradation and nucleus malformation. Coordination of Bi3+ to GSH in cells, forming Bi(SG)3, triggers other events. This (conjugation) step may be aided by GSTs, which constitute up to 10% of cytosolic proteins. The wide family of multidrug resistanceassociated proteins (MRPs) mediate ATPdependent transport of GSH conjugates, so decreasing cellular drug accumulation (conferring resistance). MRPs also pump conjugates into perinuclear vesicles, where Bi is seen as dense black deposits, probably Bi sulfide (inclusion bodies). The nature of Bi (as well as Ag and Pb) “intranuclear inclusion bodies” has long been a puzzle (6, 7). Importantly, detoxification of Bi in human cells is not so effective in bacteria where GSH is less prevalent. Where does the sulfide come from? —via the action of cysteine desulfurase on cysteine as it does for ferredoxins? Perhaps it arises from cleavage of the γ-glutamyl residue of Bi(SG)3 by γ-glutamyltransferase (γ-GT), which initiates mercapturic acid synthesis. In many anaerobic bacteria, GSH is absent and cysteine is the major thiol. Hence Bi drugs can be effective and selective against Author contributions: I.R.-C. and P.J.S. wrote the paper. The authors declare no conflict of interest. See companion article on page 3211 in issue 11 of volume 112. 1
To whom correspondence should be addressed. Email: p.j.sadler@ warwick.ac.uk.
PNAS | April 7, 2015 | vol. 112 | no. 14 | 4187–4188
pathogens that cause syphilis and dysentery, for example. Many other soft metal ions have a high affinity for thiolate sulfur and GSH sequestration pathways might be expected to be involved in their metabolism too. These include anticancer drugs based on As(III) and Pt(II), antiarthritic Au(I) drugs, as well as the essential metal ion Cu(I) (Fig. 1). The As(III) drug Trisenox [As2O3, As(OH)3 in aqueous solution], the drug of choice for promyelocytic leukemia, forms GSH conjugates that have been related to poor patient outcome. These GSH-adducts are GST-substrates easily detoxified by cancer cells (8). In contrast, the drug candidate Zinapar (darinaparsin, ZIO-101), dimethylarsenic conjugated to glutathione, takes advantage of the excessive cystine/cysteine requirement of cancer cells. This As(III) complex appears to be transformed by γ-GT at the cell surface with formation of the dimethylarsino-cysteine adduct (DMAC), which is readily taken up via cystine/cysteine importers. Hence increased cellular accumulation of As from ZIO-101 occurs. This complex might be favored for multidrug-resistant tumors where chemoresistance is mediated by high GSH levels (9). In the case of Pt(II)-based anticancer drugs, GSH may be responsible for the inactivation of compounds such as cisplatin, cis-diamminedichloroplatinum(II), or its derivatives carboplatin and oxaliplatin. Indeed, at least 60% of cellular Pt is present as GSH-adducts and increased levels of intracellular GSH are directly linked to the development of resistance to Pt(II) drugs in lung and ovarian cancers (10). For ovarian cancer, proteomic studies have related differentially expressed GSTs to Pt chemoresistance. Both cisplatin and carboplatin generate adducts in which Pt is bound to the CXXC copper-binding motif of methyl-CpG binding domain protein 2 (MBD2), and then excreted by ATP7B, a Cu exporter (11). Injectable Au(I) thiolate drugs such as Myochrysine (aurothiomalate) and the oral drug Ridaura [auranofin; (tetraacetylthioglucose)Au(PEt3)] are used for treating rheumatoid arthritis. It has long been known that gold accumulates in lysosomes in cells as dense deposits (aurosomes), widely believed to be sulfur-rich (12), but they are not well characterized. In view of the observations on Bi by Hong et al. (3), it now seems to be the time to reexamine aurosomes. Myochrysine
4188 | www.pnas.org/cgi/doi/10.1073/pnas.1503858112
direct binding and subsequent inhibition of GST (15). Interestingly, a member of the GST family, TDR1, activates antiparasitic antimonial prodrugs by deglutathionylation of Sb(V)-glutathione adducts with reduction and release of Sb(III) (16). There is a link to the metabolism of essential metal ions. GSH stimulates the uptake of copper via the transporter Ctr1. Initial formation of [CuI(SG)3]2− may be followed by delivery to P1B-type ATPases and superoxide dismutase mediated by the dynamic transfer of Cu(I) between cysteine thiolates of chaperones Atox1 and CCS in the cytoplasm (11). The field of bioinorganic chemistry and its applications in medicine are fast-growing
areas of research that use the rich structural diversity of metallic scaffolds for drug design. Metal-based drugs can provide in-cell chemical reactions that are not accessible to purely organic drugs, accompanied by multitargeting that can combat resistance (17). However, there is an urgent need for further development of methods to elucidate the spatial and temporal speciation of metallodrugs. This requires a multidisciplinary effort with contributions like those of Hong et al. (3) to understand fully the implications of the metal center in the metabolism and pharmacokinetic/dynamic properties of candidate drugs. It is evident from this brief commentary that progress in the design of metallodrugs requires a systems pharmacology approach. Not only do we need a good understanding of the kinetics and thermodynamics of metal– ligand exchange and redox reactions, but also of the responses of cells in terms of attack on multiple targets, self-propelled transport, and delivery processes related to feedback loops (Fig. 1). For metallodrugs, it will often not be appropriate to describe their mechanisms of action in terms of a single target site. This “polypharmacology” offers potential advantages because it is now becoming apparent that resistance to singletarget drugs can readily develop. The introduction of systems pharmacology approaches to unraveling the mechanism of action of metallodrugs should lead to advances in their design and in their use so as to avoid unwanted side effects.
1 Chellan P, Sadler PJ (2015) The elements of life and medicines. Philos Trans A Math Phys Eng Sci 373(2037):20140182. 2 Barry NPE, Sadler PJ (2013) Exploration of the medical periodic table: Towards new targets. Chem Commun (Camb) 49(45): 5106–5131. 3 Hong Y, Lai Y-T, Chan GC-F, Sun H (2015) Glutathione and multidrug resistance protein transporter mediate a self-propelled disposal of bismuth in human cells. Proc Natl Acad Sci USA 112(11):3211–3216. 4 Sun H (2011) Biological Chemistry of Arsenic, Antimony and Bismuth (Wiley, Chichester, UK). 5 Marí M, et al. (2013) Mitochondrial glutathione: Features, regulation and role in disease. Biochim Biophys Acta 1830(5): 3317–3328. 6 Wachstein M (1949) Studies on inclusion bodies; acid-fastness of nuclear inclusion bodies that are induced by ingestion of lead and bismuth. Am J Clin Pathol 19(7):608–614. 7 Fowler BA, Goyer RA (1975) Bismuth localization within nuclear inclusions by x-ray microanalysis. Effects of accelerating voltage. J Histochem Cytochem 23(10):722–726. 8 Sui M, Zhang Z, Zhou J (2014) Inhibition factors of arsenic trioxide therapeutic effects in patients with acute promyelocytic leukemia. Chin Med J (Engl) 127(19):3503–3506. 9 Garnier N, et al. (2014) The novel arsenical darinaparsin is transported by cystine importing systems. Mol Pharmacol 85(4): 576–585.
10 Sawers L, et al. (2014) Glutathione S-transferase P1 (GSTP1) directly influences platinum drug chemosensitivity in ovarian tumour cell lines. Br J Cancer 111(6):1150–1158. 11 Dolgova NV, Nokhrin S, Yu CH, George GN, Dmitriev OY (2013) Copper chaperone Atox1 interacts with the metal-binding domain of Wilson’s disease protein in cisplatin detoxification. Biochem J 454(1): 147–156. 12 Ghadially FN (1979) The aurosome. J Rheumatol Suppl 5:45–50. 13 Reglinski J, et al. (1997) Myocrisin-mediated oxidative stress. Clin Chim Acta 268(1-2):85–99. 14 Casini A (2012) Exploring the mechanisms of metal-based pharmacological agents via an integrated approach. J Inorg Biochem 109:97–106. 15 De Luca A, Hartinger CG, Dyson PJ, Lo Bello M, Casini A (2013) A new target for gold(I) compounds: Glutathione-S-transferase inhibition by auranofin. J Inorg Biochem 119(I):38–42. 16 Fyfe PK, Westrop GD, Silva AM, Coombs GH, Hunter WN (2012) Leishmania TDR1 structure, a unique trimeric glutathione transferase capable of deglutathionylation and antimonial prodrug activation. Proc Natl Acad Sci USA 109(29): 11693–11698. 17 Romero-Canelón I, Sadler PJ (2013) Next-generation metal anticancer complexes: Multitargeting via redox modulation. Inorg Chem 52(21):12276–12291.
inhibits GSH-peroxidase in vivo, which contributes to increased oxidative stress (13). Auranofin exerts its antiinflammatory activity by inhibiting mitochondrial thioredoxin reductase (TrxR) (14); nonetheless, its anticancer activity can be enhanced by
Hong et al. show that a systems pharmacology approach can provide new insights, an approach that, if applied more generally, will greatly advance the rational design of metallodrugs.
Romero-Canelón and Sadler
