The Conserved Histidine 106 of Large Thioredoxin Reductases Is Likely to Have a Structural Role but Not a Base Catalyst Function

CORE Metadata, citation and similar papers at core.ac.uk

Provided by Elsevier - Publisher Connector

FEBS 29213 FEBS Letters 579 (2005) 745–748

The conserved histidine 106 of large thioredoxin reductases is likely to have a structural role but not a base catalyst function

Judit Jacob, R. Heiner Schirmer, Stephan Gromer* Biochemiezentrum der Universita¨t Heidelberg (BZH), Im Neuenheimer Feld 504, D-69120 Heidelberg, Germany

Received 26 November 2004; revised 3 January 2005; accepted 4 January 2005

Available online 11 January 2005

Edited by Hans Eklund

of thioredoxin reductases exist: small TrxRs (subunit M ap- Abstract The catalytic activity of selenocysteine-containing r thioredoxin reductases can be mimicked by cysteine-variants if prox. 35 kDa) primarily present in bacteria, fungi, and plants the local environment at the C-terminal redox center supports and large TrxRs (subunit Mr approx. 55 kDa) present e.g. in thiol activation. This concept of a linear catalytic site was chal- higher eukaryotes. They differ in many aspects, such as the lenged by structural data suggesting that the invariant residue substrate spectrum, which is far broader for large than for His106 functions as a base catalyst for the dithiol-disulphide ex- small TrxRs [1]. change reaction between enzyme and substrate. As reported here, A key feature of the proposed mechanism of large thiore- 106 we changed His to asparagine, glutamine, and phenylalanine doxin reductases is the relatively flexible C-terminal tail which in various C-terminal mutants of Drosophila melanogaster thio- is responsible for the transport of electrons from the buried N- redoxin reductase. The catalytic activity dropped considerably, terminal redox-center near the flavin to bulky substrates at the yet pH-profiles did not reveal differences, rendering a function surface. This requires a second redox-active site located on the for His106 as a base catalyst unlikely. Interestingly, the phenylalanine-mutants, designed as negative controls were the most ac- C-terminal tail [3–5]. tive mutants which suggests rather a structural role of His106. The particular design of the second redox-center present in 2005 Federation of European Biochemical Societies. Published large thioredoxin reductases differs between species: In mam- by Elsevier B.V. All rights reserved. malian TrxRs it is formed by a sequentially directly adjacent cysteine–selenocysteine-pair, whereas in the enzyme of Diptera Keywords: Thioredoxin reductase; Base catalyst; Mobile two cysteines form this motif [6,7]. Astonishingly, these cys- catalytic site; Thiol activation; Histidine functional analysis teine-variants are almost as active as their selenocysteine coun- terparts, despite the significantly higher pKa of cysteines and thus a reduced tendency to form the anions required for disul- 1. Introduction phide exchange reactions. Recent experiments strongly suggest that indeed sequentially neighbouring residues activate the thi- The thioredoxin system – formed by thioredoxin reductase ols [8]:InDrosophila melanogaster TrxR (containing the flexi- 1 (TrxR; EC 1.8.1.9 ), NADPH and its naming substrate thio- ble terminal sequence SCCS) hydroxyl groups of the flanking redoxin (Trx) – is an important constituent of the intracellular serines form hydrogen bonds with the thiols, lowering their redox milieu. It interacts with many different metabolic path- pKa and thus increase reactivity. ways such as DNA-synthesis, selenium metabolism and the It was, however, also suggested that a base catalyst facili- antioxidative network, and it exhibits significant species differ- tates the reaction between enzyme and substrate – with the ences. This renders the system an attractive and currently highly conserved His106, 2 located in close proximity to the extensively studied chemotherapeutic target in many fields of C-terminal Cys–Cys-pair [9], being the prime candidate (Fig. medicine – ranging from infectious diseases to cancer therapy 1A and B). This function of His106 was postulated in analogy (see Ref. [1] for review). to His464 which catalyzes the intramolecular dithiol-disulphide + TrxR (thioredoxin-S2 + NADPH + H thioredoxin- exchange between two cysteine pairs within the dimeric en- + (SH)2 + NADP ) is an NADPH-dependent flavoenzyme zyme (Fig. 1A and B, [10–13] and own observations, unpub- belonging to a family of homodimeric pyridine nucleotide- lished). disulphide oxidoreductases which includes enzymes like gluta- The objective of this work was to verify this hypothesis and thione reductase, lipoamide dehydrogenase, trypanothione thus provide further insight into the mechanistic details of reductase, and mercuric ion reductase [2]. Two distinct classes large thioredoxin reductases – a prerequisite for a rational, mechanism-based drug design. *Corresponding author. Fax: +49 6221 54 5586. E-mail address: [email protected] (S. Gromer). 2. Materials and methods 1 Thioredoxin reductase was formerly designated 1.6.4.5. Cloning, expression and purification of the 15 DmTrxR-1 mutants Abbreviations: Dm, Drosophila melanogaster; Sec, selenocysteine; and wild-type enzyme were performed applying established proce- (Dm)Trx, (Drosophila melanogaster) thioredoxin; (Dm)TrxR, (Dro- dures, as were enzyme kinetics (K , k ), pH-activity profiles, circular sophila melanogaster) thioredoxin reductase; -SCCS, -SCCG, -GCCS M cat and -GCCG, indicate the respective sequence of the C-terminal tetrapeptide of Dm TrxR in one letter code; DTNB, 5,50-dithio-bis- 2 The numbering of residues in the text refers to Drosophila melano- (2-nitrobenzoic acid) gaster TrxR-1 (AF301144).

0014-5793/$30.00 2005 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2005.01.001 746 J. Jacob et al. / FEBS Letters 579 (2005) 745–748

0 Fig. 1. (A) Partial sequence alignment of large TrxRs and glutathione reductases. The conserved residues His106 (red) and His464 (blue) are highlighted. Alignments were done using the online-tool Clustal W 1.82 (www.ebi.ac.uk/clustalw). hTrxR-1 = human TrxR-1 (NCBI accession number: s66677), hTrxR-2 = human TrxR-2 (aad51324), SsTrxR = pig TrxR (aaf78791.1), mTrxR = mouse TrxR-1 (np056577), CeTrxR = C. elegans TrxR (aad41826), AgTrxR = Anopheles gambiae TrxR (cad30858), DmTrxR-1 = D. melanogaster TrxR-1 (aag25640), hGR = human glutathione reductase 0 (rdhuu) (B) 3D-model of His106 (red) and His464 (blue) in D. melanogaster TrxR based on the structural information provided by protein data base (www.rcsb.org) ﬁle 1H6V of a Sec498 ﬁ Cys rat TrxR mutant (subunits E and F) [9]. The C-terminal tail of one subunit is shown in green, with the adjacent cysteines highlighted in yellow. The second subunit, containing His106, is shown in dark gray. Also highlighted in yellow are the N-terminal cysteines and FAD (orange). The structure shown here was created using H. J. BernsteinÕs RasMol V 2.7.2.1.1 (www.bernstein-plus-sons.com/ software/rasmol). (C) Structural side-chain comparison of histidine and the amino acids used in the mutants of His106: glutamine, asparagine and phenylalanine.

dichroism-analysis and absorbance spectra recordings. Details can be netic results (KM- and kcat-values differed by less than ±5%). found in this articleÕs online supplement, which is available at the Neither the absorbance-spectra (which showed typical flavo- FEBS-Letters homepage. The enzyme species are designated according protein characteristics) nor the circular dichroism spectra re- to their C-terminal tetrapeptide sequence (Gly–Cys–Cys–Gly = GCCG, Gly–Cys–Cys–Ser = GCCS, Ser–Cys–Cys–Gly = SCCG and vealed any significant differences between the mutants and Ser–Cys–Cys–Ser = SCCS = wild-type (wt)), and, furthermore, accord- wild-type enzyme, indicating an identical overall structure. If ing to the respective mutation at position 106. His106 indeed acted as an important base catalyst, one would expect a significant drop in activity for all three mutants, with the highest loss of activity for the phenylalanine mutant. This 3. Results and discussion was, however, not the case: Even though all three mutants were less active in thioredoxin reduction (kcat) than the His- The catalytic reactions on the Si-side of the flavin in large containing wild-type enzyme (Fig. 2A), the phenylalanine mu- thioredoxin reductases are essentially dithiol-disulphide ex- tants, originally designed as negative controls, were the most change reactions. In the course of these reactions a more reac- active ones (Fig. 2). The KM-values for thioredoxin remained tive intermediate, a thiolate or selenolate, has to be formed at essentially unaltered (Fig. 2A), indicating that the formation the C-terminal redox center of thioredoxin reductase [1,14]. of the Michaelis-complex is not affected. Moreover, no shift The question, why the catalytic activity of cysteine containing in the pH-profile (not shown, pH-optima at around 7.5–8.0) TrxRs, such as those from D. melanogaster or Anopheles gam- was observed for any of the mutants [8]. biae, almost matches that of their selenocysteine counterpart The exchange of the polar serines for glycines led to a signif- was addressed recently [8]. The hydroxyl groups of residues icant loss of activity. Thus these glycine-mutants should be sequentially adjacent to the redox-active C-terminal Cys– even more sensitive to an exchange of His106 for Asn, Gln or 106 Cys-pair seem to lower the thiols’ pKa, facilitating thiolate even Phe if histidine acted as a base catalyst. Yet, this was formation. not the case either. We changed His106 to asparagine, glutamine, and phenylal- In order to investigate the possibility that His106 might serve anine, respectively. All three amino acids share some structural as a base catalyst for substrates other than thioredoxin we also features with histidine (Fig. 1C), however, they are either no analyzed DTNB (5,50-Dithio-bis-(2-nitrobenzoic acid)), a low (Phe) or poorer potential base catalysts (Asn, Gln) than His molecular weight compound [17], with the SCCS mutant series [15]. Asparagine and glutamine were chosen, as relative phylo- (Fig. 2B 3). In this case the asparagine and glutamine mutant genetic substitution frequencies indicate, that histidine is more showed 15–20% reduced activities compared to the wild-type likely substituted for glutamine (ratio 1.2) or asparagine (ratio 1.3) than for glutamate (ratio 0.8) or aspartate (ratio 0.9); the ratio for His Phe is 0.7 [16]. 3 DmTrxR, as well as Plasmodium falciparum and Anopheles gambiae We obtained pure, yellow-colored, enzymatically active pro- 106 TrxR are rapidly inactivated at higher concentrations of DTNB. This teins for wild-type and all 15 mutants of His . A number of leads to the higher standard deviation values as indicated by the error mutants were cloned twice independently with reproducible ki- bars in Fig. 2B compared to Fig. 2A. J. Jacob et al. / FEBS Letters 579 (2005) 745–748 747

A B

Fig. 2. (A) Kinetic parameters of mutagenized DmTrxR species using thioredoxin as a substrate. The bottom line of the Figure shows various C- terminal linear catalytic sites, the wild-type SCCS in the left panel and 3 mutants thereof in the other panels. The line above indicates varying amino 106 acids at position 106 for each panel, His representing the wild-type. Dark vertical columns show the kcat-values and white columns the KM-values (Dm Trx-2, GHOST-assay). Error bars indicate ± SD. (B) Kinetic parameters of mutagenized DmTrxR species using 5,50-Dithio-bis-(2-nitrobenzoic acid) as a substrate. The symbols are as in (A). Only the catalytic site mutant Dm TrxR-SCCS with diﬀerent residues at position 106 (His, Asn, Gln, Phe) is shown.

enzyme, whereas the phenylalanine mutant did not show a de- supported by the finding that Cys490 fi Sec mutants of crease in activity at all. DmTrxR exhibit kcat-values essentially identical to those of Taken together, these data show that His106 does not serve wild-type (SCCS) enzyme [8]. as an important, if at all, base catalyst in large TrxRs. How- From a more general biochemical point of view, it is note- ever, the fact that all substitutions at position 106 led to a worthy that the substitution His fi Phe should probably be significant decrease in (TrxS2-)reduction activity, indicates used more frequently in mutational studies. Our results indi- that the highly conserved His106 is indeed of importance cate that other negative control substitutes, such as the com- for large thioredoxin reductases. As acid–base functionality monly used alanine, might lead to misinterpretation, as apparently plays a subordinate role, the highest degree of histidine may be underestimated as a structural component un- deviation from the imidazoleÕs structure is present in aspar- der these conditions. agine and glutamine, whereas the aromatic ring of phenylalanine is sterically most closely related to His (Fig. 1C). This Acknowledgement: The authors like to thank Dr. Manuela Lopez de la view is reflected by the obtained kinetic data (Fig. 2A and Paz and Dr. Mayte Pastor (EMBL Heidelberg) for active support with the CD-spectra and Irene Ko¨nig for excellent technical assistance. This B). The overall structure of the enzymes – as judged by work was supported by a grant of the Deutsche Forschungsgemeins- KM-determination, absorbance- and CD-spectrometry – re- chaft (GR 2028/1-1). mained unaltered, yet, obviously the local structural environment has changed. Cavity creation, as occasionally observed with alanine mutation [18], is a possible, yet not very likely explanation for the differences between the mutants, as Appendix A. Supplementary data asparagine is larger and more rigid and glutamine, even though more flexible, is much bulkier than alanine (Fig. Supplementary data associated with this article can be found, 1C) [16]. in the online version, at doi:10.1016/j.febslet.2005.01.001. From the data provided by the rat mutant enzyme structure [9], His106 fits well into a grove formed by a hydropho- References bic and a hydrophilic circle in the enzyme structure. As the currently available structural data of large TrxRs still leave [1] Gromer, S., Urig, S. and Becker, K. (2004) The thioredoxin room for interpretations, it is difficult to predict which struc- system – from science to clinic. Med. Res. Rev. 24, 40–89. tural interactions and features render His106 so important, [2] Williams Jr., C.H. (1992) Lipoamide dehydrogenase, glutathione particularly as the C-terminal tail is very flexible and thus reductase, thioredoxin reductase, and mercuric ion reductase – a family of flavoenzyme transhydrogenases (Mu¨ller, F., Ed.), difficult to study by x-ray diffraction analysis [19]. However, Chemistry and Biochemistry of Flavoenzymes, vol. 3, pp. 121– our data strongly suggest that it is indeed primarily a struc- 211, CRC Press, Boca Raton, FL. tural feature that has led to the conservation of His106. This [3] Zhong, L., Arne´r, E.S.J., Ljung, J., A˚ slund, F. and Holmgren, A. is in line with recent reports suggesting that histidine might (1998) Rat and calf thioredoxin reductase are homologous to serve more frequently as a structural component of proteins glutathione reductase with a carboxyl-terminal elongation containing a conserved catalytically active penultimate selenocysteine than previously thought [20]. residue. J. Biol. Chem. 273, 8581–8591. 106 Now that His has basically been dropped from the list of [4] Arscott, L.D., Gromer, S., Schirmer, R.H., Becker, K. and candidates for a base catalyst in large TrxR, two options re- Williams Jr., C.H. (1997) The mechanism of thioredoxin main: Either there is no real base catalyst, or other residues reductase from human placenta is similar to the mechanisms of lipoamide dehydrogenase and glutathione reductase and or helix-dipoles are committed to that task. The notion that in- is distinct from the mechanism of thioredoxin reductase deed no base catalyst is involved in the intermolecular dithiol- from Escherichia coli. Proc. Natl. Acad. Sci. USA 94, 3621– disulphide exchange reaction between enzyme and substrate is 3626. 748 J. Jacob et al. / FEBS Letters 579 (2005) 745–748

[5] Gromer, S., Wissing, J., Behne, D., Ashman, K., Schirmer, R.H., [12] Rietveld, P., Arscott, L.D., Berry, A., Scrutton, N.S., Deonarain, Flohe´, L. and Becker, K. (1998) A hypothesis on the catalytic M.P., Perham, R.N. and Williams Jr., C.H. (1994) Reductive and mechanism of the selenoenzyme thioredoxin reductase. Biochem. oxidative half-reactions of glutathione reductase from Escherichia J. 332, 591–592. coli. Biochemistry 33, 13888–13895. [6] Kanzok, S.M., Fechner, A., Bauer, H., Ulschmid, J.K., Mu¨ller, [13] Karplus, P.A. and Schulz, G.E. (1987) Refined structure of H.M., Botella-Munoz, J., Schneuwly, S., Schirmer, R. and glutathione reductase at 1.54 A˚ resolution. J. Mol. Biol. 195, 701– Becker, K. (2001) Substitution of the thioredoxin system for 729. glutathione reductase in Drosophila melanogaster. Science 291, [14] Bauer, H., Massey, V., Arscott, L.D., Schirmer, R.H., Ballou, D. 643–646. and Williams Jr., C.H. (2003) The mechanism of high Mr [7] Bauer, H., Gromer, S., Urbani, A., Schno¨lzer, M., Schirmer, R.H. thioredoxin reductase from Drosophila melanogaster. J. Biol. and Mu¨ller, H.M. (2003) Thioredoxin reductase from the malaria Chem. 278, 33020–33028. mosquito Anopheles gambiae – Comparisons with the orthologous [15] Kass, I.J. and Sampson, N.S. (1998) Evaluation of the role of enzymes of Plasmodium falciparum and the human host. Eur. J. His447 in the reaction catalyzed by cholesterol oxidase. Biochem- Biochem. 270, 4272–4281. istry 37, 17990–18000. [8] Gromer, S., Johansson, L., Bauer, H., Arscott, L.D., Rauch, S., [16] Schulz, G.E. and Schirmer, R.H. (1978) Principles of protein Ballou, D.P., Williams Jr., C.H., Schirmer, R.H. and Arne´r, E.S. structure, Springer, New York. (2003) Active sites of thioredoxin reductases: why selenoproteins? [17] Holmgren, A. (1977) Bovine thioredoxin system. Purification of Proc. Natl. Acad. Sci. USA 100, 12618–12623. thioredoxin reductase from calf liver and thymus and studies of its [9] Sandalova, T., Zhong, L., Lindqvist, Y., Holmgren, A. and function in disulfide reduction. J. Biol. Chem. 252, 4600–4606. Schneider, G. (2001) Three-dimensional structure of a mamma- [18] Xu, B., Hua, Q.X., Nakagawa, S.H., Jia, W., Chu, Y.C., lian thioredoxin reductase: implications for mechanism and Katsoyannis, P.G. and Weiss, M.A. (2002) A cavity-forming evolution of a selenocysteine-dependent enzyme. Proc. Natl. mutation in insulin induces segmental unfolding of a surrounding Acad. Sci. USA 98, 9533–9538. alpha-helix. Protein Sci. 11, 104–116. [10] Untucht-Grau, R., Schulz, G.E. and Schirmer, R.H. (1979) [19] Schiering, N., Kabsch, W., Moore, M.J., Distefano, M.D., Walsh, The C-terminal fragment of human glutathione reductase C.T. and Pai, E.F. (1991) Structure of the detoxification catalyst contains the postulated catalytic histidine. FEBS Lett. 105, mercuric ion reductase from Bacillus sp. strain RC607. Nature 244–248. 352, 168–172. [11] Fujiwara, N., Fujii, T., Fujii, J. and Taniguchi, N. (2001) Roles of [20] Safarian, S., Moosavi-Movahedi, A.A., Hosseinkhani, S., Xia, Z., N-terminal active cysteines and C-terminal cysteine-selenocysteine Habibi-Rezaei, M., Hosseini, G., Sorenson, C. and Sheibani, N. in the catalytic mechanism of mammalian thioredoxin reductase. (2003) The structural and functional studies of His119 and His12 in J. Biochem. (Tokyo) 129, 803–812. RNase A via chemical modification. J. Protein Chem. 22, 643–654.