Personal tools

Information zum Seitenaufbau und Sprungmarken fuer Screenreader-Benutzer: Ganz oben links auf jeder Seite befindet sich das Logo der JLU, verlinkt mit der Startseite. Neben dem Logo kann sich rechts daneben das Bannerbild anschließen. Rechts daneben kann sich ein weiteres Bild/Schriftzug befinden. Es folgt die Suche. Unterhalb dieser oberen Leiste schliesst sich die Hauptnavigation an. Unterhalb der Hauptnavigation befindet sich der Inhaltsbereich. Die Feinnavigation findet sich - sofern vorhanden - in der linken Spalte. In der rechten Spalte finden Sie ueblicherweise Kontaktdaten. Als Abschluss der Seite findet sich die Brotkrumennavigation und im Fussbereich Links zu Barrierefreiheit, Impressum, Hilfe und das Login fuer Redakteure. Barrierefreiheit JLU - Logo, Link zur Startseite der JLU-Gießen Direkt zur Navigation vertikale linke Navigationsleiste vor Sie sind hier Direkt zum Inhalt vor rechter Kolumne mit zusaetzlichen Informationen vor Suche vor Fußbereich mit Impressum

Document Actions

Human thioredoxin reductases

Human thioredoxin reductases: selenoproteins as potential drug targets

(Research Grant within the DFG Research Focus “Selenoproteins,” SPP 1087; Be 1540/6-3)


Human thioredoxin reductase (hTrxR) is a redox-active enzyme equipped with a selenocysteine residue essential for catalysis. The enzyme and its physiological substrate thioredoxin play a key role in cellular redox regulation, antioxidant defense, (tumor) cell proliferation, and apoptosis. Thus hTrxR represents a most promising target molecule for the development of anticancer drugs and an excellent tool for studying the function of redox-regulatory cellular networks. Over the last several years we studied hTrx structurally and mechanistically in comparison to other TrxRs, e.g. from Drosophila melanogaster. We demonstrated that the enzyme is potently inhibited by a range of chemical compounds including organic gold compounds, terpyridine-platinum complexes, cisplatin derivatives, and palladium complexes. Cytostatic activity of the compounds was demonstrated on different highly malignant cell lines and in a glioblastoma animal model. Within the next two years we would like to continue/complete the functional and structural work on enzyme(-ligand) complexes, including the characterization of human thioredoxin glutathione reductase (TGR) as a novel selenoprotein, and test the most promising hTrxR inhibitors in animal models. The drugs' mechanism of action (hTrxR inhibition versus DNA intercalation; interaction with the Sec-residue) will be characterized. Additionally we aim to study the effects of a highly efficient hTrxR inhibition on (tumor) cells by employing glioblastoma-specific microarrays, 2-dimensional gel electrophoresis, SELDI analysis, and apoptotic markers.

State of the art

Thioredoxin reductases. Mammalian TrxR as well as the TrxR from Drosophila melanogaster and other eukaryotes are members of the pyridine nucleotide-disulfide oxidoreductase family of dimeric flavoenzymes with high homology to glutathione reductases. Thioredoxin reductase (TrxR; EC reduces the 12 kDa disulfide protein thioredoxin (Trx) to its dithiol-containing form:

NADPH + H+ + Trx-S2    -------------   NADP+ + Trx-(SH)2

Reduced Trx in turn provides reducing equivalents for a whole range of intra- and extracllular processes. In contrast to their low molecular weight bacterial counterparts, mammalian TrxRs were shown to be capable of reducing a large variety of different substrates. Low Mr compounds such as lipoamide, lipoic acid, hydrogen peroxide, lipid hydroperoxides, S-nitrosoglutathione (but not glutathione disulfide), ascorbate, alloxane, Ellman's disulfide (DTNB), ebselen (diselenide), methylseleninate, ubiquinone, and others serve as well as substrates, as do proteins such as calcium-binding protein 1 and 2, NK-lysine, protein disulfide isomerase, plasma glutathione peroxidase, and of course thioredoxin. The functions of the thioredoxin system have been described in recent, comprehensive reviews. In addition to the cytosolic hTrxR1, a second protein located in mitochondria has been described in mammals. Human TrxR1 is transcribed by a house-keeping-type promoter and is regulated post-transcriptionally. Heterogeneity concerning the 5' UTR in mammalian and rat TrxR mRNA results from alternative first exon splicing and is also responsible for the differentiation between TrxR1 mRNA and TrxR2 mRNA. Mitochondrial TrxR2 has been shown to participate in the regulation of protein tyrosine phosphorylation and cell growth as a component of the mitochondria-specific H2O2-eliminating system that includes peroxiredoxin III and thioredoxin 2.

Thioredoxins. The multifunctionality of the mammalian TrxR1 is further augmented by the numerous intra- and extracellular activities of its main product, reduced thioredoxin. Even though closely related to its E. coli counterpart, mammalian Trx has gained a magnitude of new – mainly (redox) regulatory – functions during evolution. Apart from its "classical" function as an electron shuttle for ribonucleotide reductase, it modulates the activity of transcription factors such as NF-Y, Pax-8, TTF-1, and NF-κB; serves as a glucocorticoid receptor-activating factor; plays a role in protein biosynthesis; has an impact on the structure of export proteins; regulates the activity of other enzymes; serves as an antioxidant; and can act extracellularly as an autocrine growth factor. Most organisms contain different Trx isoforms, which act in different compartments with different specificity. Recently a splice variant of hTrx-1 (Delta3Trx-1) was characterized that has no catalytic activity but competes with full-length Trx for the interaction with TrxR. For the E. coli thioredoxin system, properties similar to those of molecular chaperones have been described, and for yeast the functional role of thioredoxins in the protection from both oxidative and reductive stress has been shown.


Thioredoxin glutathione reductase. In mammals three types of TrxRs have been characterized: the cytosolic form hTrxR1 and the mitochondrial hTrxR2, which are expressed ubiquitously and exhibit 52% homology, as well as a new type of pyridine nucleotide disulfide reductases, the thioredoxin glutathione reductase (TGR), with high homology to TrxR1 and TrxR2. This hybrid enzyme has been detected in Mus musculus, Schistosoma mansoni, and in Echinococcus granulosus. The E. granulosus TGR was purified from larval protoscoleces, where it seemed to be the predominant pyridine nucleotide disulfide reductase. Two trans-spliced mRNA variants derived from a single gene that differ solely in their N-terminus were isolated. They code for mitochondrial and cytosolic isoforms of TGR. In S. mansoni a cytosolic form of TGR exists and seems to completely replace TrxR and GR in adult worms. In mice, the TGR was isolated from the testes and was found in the microsomal fraction. This location distinguished the enzyme from TrxR1 and Grx, located in the cytosol, TrxR2, located in the mitochondria, and GR, located in both cytosolic and mitochondrial fractions. Mouse TGR exhibits strong sequence homology to a partial sequence of human TGR that was formerly described as TrxR3 (TR2).

Mouse TGR, a homodimer of 66 kDa subunits exhibits specificity for both redox systems and displays TrxR, GR and glutaredoxin activities. This unusual substrate specificity is achieved by the fusion of the glutaredoxin and TrxR domains. The enzyme contains a selenocysteine residue encoded by the TGA codon. The proposed electron flow within TGR includes several redox centers and the flexible C-tail as in hTrxR: NADPH FAD dithiol/disulfide center C-terminal Sec-containing center Cys within the Grx domain downstream substrate. Biochemical characterization of S. mansoni TGR indicates that the Grx and TrxR domains of the enzyme can function either coupled or independently from each other.


The thioredoxin system of Drosophila melanogaster. In the fruit fly the absence of a genuine glutathione reductase was demonstrated in 2001. GR-activity is likely to be compensated by the thioredoxin system, since thioredoxin was shown to reduce GSSG in a chemical reaction. This reaction was also established in other organisms. The larval death of TrxR null mutations in Drosophila supports the lack of a second redox system of similar efficacy. Additional analysis of the redox state of thioredoxins and glutaredoxins in yeast mutants has shown that thioredoxins are maintained independently of the glutathione system. This constellation enables cells to survive in conditions under which the GSH-glutaredoxin system is oxidized. The effects of the  overexpression of TrxR, superoxide dismutase, and catalase on longevity of Drosophila has recently been studied. Drosophila melanogaster was shown to possess a selenium-free TrxR containing a Cys-Cys pair as a second redox center in its C-terminal extension. The catalytic mechanism of this enzyme was recently elucidated and showed that the enzyme indeed oscillated between the EH2 and the EH4 electron state, further proving the importance of the C-terminal redox pair. In the meantime two different thioredoxins with distinct functions as well as different thioredoxin-dependent peroxidases have been described in Drosophila. A single Drosophila gene, TrxR1, encodes non-selenocysteine-containing cytoplasmic and mitochondrial TrxR isoforms, which were studied functionally and biochemically.


Human thioredoxin reductase as a selenoenzyme. Mammalian TrxRs and TrxRs of other eukaryotes contain selenocysteine as the penultimate amino acid. The species-specific usage of the selenocysteine insertion sequence (SECIS)-element makes a recombinant expression of the mammalian enzyme in E. coli difficult – a challenge that has been tackled by different groups.

Various approaches including site-directed mutagenesis, selenium depletion, and selective digestion or alkylation proved that the selenocysteine is essential for the catalytic mechanism of the mammalian enzyme and that the transfer of electrons takes place from the active site disulfide to the C-terminal redox center. The first – and so far only – three-dimensional structure of a mammalian TrxR was solved by Sandalova et al. in 2001. The overall structure of the studied SeCys498Cys rat TrxR mutant was found to be similar to GR including FAD- and NADPH-binding residues. Interestingly all residues directly interacting with GSSG in GR were conserved in rat TrxR, although the enzyme does not turn over GSSG. The flow of electrons through the dimeric TrxR is – based on the structure – possible without larger conformational changes, and the C-terminal extension extends the electron transport chain and prevents the TrxR from acting as a GR. On the basis of the redox properties of the selenocysteine in hTrxR, Sun et al. postulated in 1999 that oxidation of this residue contributes to redox signaling and, specifically, to mediating responses to oxidative stress. As recently reported by Mostert et al. in 2003, a loss of TrxR activity after selenium depletion is responsible for the induction of the stress response enzyme heme oxygenase in liver. Selenium overexposure tested in the rat model had an adverse effect on TrxR mRNA levels and activity and implicated tissue-specific differences in regulation. In contrast selenium supplementation of a human endothelial cell line induced TrxR and GPx activities and protected from oxidative damage. In accordance with the emerging importance of the selenium-containing C-terminal extension of hTrxR, Novoselov & Gladyshev postulated in 2003 that these extensions of proteins represent a general mechanism of evolution of a new protein function.


The human thioredoxin system: growth promotion versus apoptosis. Trx is secreted by normal and neoplastic cells through an unusual secretory pathway and acts in reduced form as an autocrine growth factor. In 2000 Soderberg et al. revealed secretion of TrxR from monocytes via the classical Golgi pathway. Many tumor cells are known to have manifoldly increased Trx and TrxR levels. One might speculate that enhanced DNA synthesis and defense against the host's immune system are major causes. The involvement of the thioredoxin system in oncogenesis and tumorogenesis and its potential as a target for anticancer therapy were recently confirmed by Lincoln et al. in 2003. Furthermore mitochondrial TrxR has been demonstrated to participate in the regulation of cell proliferation, and Trx-1 regulates the expression of breast cancer-specific genes. Trx-1 overexpression results in increased vascular endothelial growth factor production and enhanced tumor angiogenesis. Since hTrxR is moreover inhibited by various anticancer components, the enzyme is generally considered to promote cell viability.

However, several recent studies have suggested that TrxR1 may also be capable of promoting apoptosis. The tumor-suppressing protein p53 is thioredoxin-dependently regulated via its cysteines near the p53-DNA interface, and activated p53 is (down-)regulated via TrxR in mamalian cells. TrxR plays a critical role in IFN retinoid-mediated tumor growth suppression in vivo, and electrophilic prostaglandins and lipid aldehydes repress p53 by impairing hTrxR, thus suggesting a role of hTrxR in suppressing carcinogenesis. As reported by Song and Lee in 2003, metabolic oxidative stress may activate apoptosis signal-regulating kinase 1 glutaredoxin- and/or thioredoxin-dependently. Redox-regulatory and anti-apoptotic functions of thioredoxin were shown to depend on the S-nitrosylation of Trx at Cys69, and glutathionylation of Trx was suggested to regulate Trx functions. However, the two diverse effects of the thioredoxin system – growth promotion and induction of apoptosis – will have to be studied in further detail.


Human TrxR as a drug target. The broad functional spectrum of mammalian TrxRs, their involvement in a magnitude of cellular processes, and therefore their potential importance in a variety of diseases render them highly interesting candidates for drug development (for a recent review of the literature please see Gromer et al., 2003). To simultaneously decrease DNA synthesis, antioxidant defense, and autocrine growth, stimulation of a tumor cell by the inactivation of one enzyme is indeed a tempting approach. A number of clinically and experimentally used chemotherapeutic drugs have already been shown to effectively inhibit hTrxR. A prominent example is the inhibition of TrxR by the cytostatic agent BCNU. Most interestingly BCNU is one of the very few cytostatic agents successfully employed in the treatment of glioblastoma. In 1998 Hofman et al. reported that analogs of the TrxR-interface domain are effective inhibitors of the enzyme. This might render TrxR a potential target for dimerization inhibitors, a principle that has already been demonstrated to be effective for glutathione reductase. In order to also facilitate efficient and cost-effective inhibitor screening in high throughput systems, alternative substrates of hTrxR and the isofunctional enzyme of the malaria parasite Plasmodium falciparum have been developed.

A range of different compounds have been described to inhibit hTrxR and to have antitumor effects; these compounds include gold(I)-compounds such as aurothioglucose and auranofin, trivalent arsenicals such as methyl-AS, water-soluble organotellurium compounds, as well as naphthoquinones and cis-diamminedichloroplatinum(II), which have been shown to inhibit hTrxR, Trx, and glutaredoxin. Furthermore (2,2':6',2'-terpyridine)platinum(II) complexes have been found to almost stoichiometrically target hTrxR, which is inhibited efficiently in vitro. In vivo the proliferation of different gliobastoma cell lines could be strongly inhibited by these compounds. Apart from its great potential as a target for the rational design of cytostatic and antirheumatic drugs, hTrxR represents an important reference protein for specific drug design against other members of the pyridine nucleotide-disulfide oxidoreductase family, e.g. against glutathione reductase or trypanothione reductase.


Previous contributions

This group has experience with disulfide reductases, including human glutathione reductase (GR), thioredoxin reductase (TrxR), and lipoamide dehydrogenase, the corresponding enzymes from the malaria parasite Plasmodium falciparum, as well as disulfide reductases from Drosophila melanogaster and E. coli. We developed an effective method to purify mg quantities of hTrxR from human placenta and showed that isolated hTrxR possesses one selenium per subunit. In cooperation with Professor Charles Williams, Ann Arbor, MI, we studied the reductive half-reaction of the enzyme (NADPH + H+ + Trx-S2 Trx-(SH)2 + NADP+), which substantiated the importance of the C-terminal Cys-Sec-pair in catalysis. Steady-state kinetic experiments on hTrxR resulted in the typical model of a ping-pong mechanism. Our group has purified mouse TrxR and has characterized it as a selenium-dependent homodimeric flavoenzyme that can be inhibited by the cytostatic agent BCNU.

The inhibition of hTrxR by the antirheumatic gold compounds auranofin (Ki = 4 nM) and aurothioglucose (Ki = 20 nM) has been studied in detail. A mathematical model describing the complex type of inhibition of these extremely efficient inhibitors has been proposed. The conditions for denaturation and reactivation of P. falciparum and human TrxR have been established, and studies with dimerization inhibitors have been conducted. In cooperation with Dr. Elisabeth Davioud, Pasteur Institute Lille, we developed and tested alternative low molecular weight substrates for hTrxR and PfTrxR. The structure of these compounds, which have been patented, is based on DTNB and facilitates inhibitor screening.

Thioredoxin reductase of the malarial parasite Plasmodium falciparum has been studied in detail. The expression of PfTrxR has been enabled by optimizing the codon usage of the gene in the C-terminal extension. PfTrxR does not contain selenium. This difference between the human and the parasite enzyme represents a promising starting point for specific antiparasitic drug design.

The effects of physiological NO-carrier molecules, namely S-nitrosoglutathione, dinitrosyl-dithiol-iron complexes, and peroxynitrite on human GR, an enzyme very closely related to hTrxR, were described in detail. The modified enzyme species were crystallized, and the type of inhibition was elucidated at 1.7 Å resolution. The studies indicated that oxidation must be considered a major reaction induced by NO-carrier molecules and that different NO-carriers mediate distinct chemistry.

Based on the absence of a typical GR in Drosophila melanogaster, we studied the potential of the DmTrxR1/DmTrx1 system to reduce glutathione disulfide and determined a chemical reaction between reduced Trx and GSSG with a rate constant k2 of 170 M-1 s-1. This value allows high GSSG fluxes in D. melanogaster cells.

Apart from NO-donors, comparative inhibition studies on disulfide reductases include work on BCNU (carmustin), methylene blue, isoalloxazine derivatives, and paraquat. We identified hTrxR as a major target of (2,2´:6´,2´´-terpyridine)platinum(II) complexes, which had been synthesized by Prof. Gordon Lowe, Oxford. The platinum complexes were found to inhibit the proliferation of glioblastoma and head-and-neck squamous carcinoma cell lines dose-dependently in the lower micromolar range. In these cells they had an immediate dose-dependent effect on TrxR activity. GR activity was found to be induced by the treatment. First studies on bioavailability showed that the terpyridine-platinum complexes also reacted – although to a lower extent – with other thiol-containing molecules such as serum albumin and glutathione. Therefore the respective HSA and GSH complexes were tested and found to inhibit hTrxR very efficiently with IC50 values in the lower nanomolar range.

With hTrxR isolated from human placenta, crystals of up to 600 x 300 x 100 µm were reproducibly obtained. A synchrotron dataset obtained by the group of Andrew Karplus in Berkeley demonstrated a diffraction down to 4 Å (30% completeness). Complete data were obtained to a resolution of 4.5 Å.

In cooperation with the groups of Heiner Schirmer, BZH, Heidelberg, and Stephan Schneuwly, Department of Zoology, Regensburg, we demonstrated the absence of glutathione reductase in the fruit fly Drosophila melanogaster. We identified and characterized in detail a non-selenium-dependent thioredoxin reductase in the insect and provided evidence that a thioredoxin system supports glutathione disulfide reduction in this organism. Our data suggest that antioxidant defense in Drosophila – and probably in related insects – differs fundamentally from other organisms. These studies were carried out in the first funding period of the selenoprotein priority program.