- Email: [email protected]
Human glutaredoxin 3 can bind and effectively transfer [4Fe–4S] cluster to apo-iron regulatory protein 1
Biochemical and Biophysical Research Communications xxx (2015) 1e5
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Human glutaredoxin 3 can bind and effectively transfer [4Fee4S] cluster to apo-iron regulatory protein 1 Haiyan Xia a, Binghua Li a, Zhou Zhang a, Qi Wang b, Tong Qiao b, Kuanyu Li a, * a b
Jiangsu Key Laboratory for Molecular Medicine, Medical School of Nanjing University, Nanjing, China Department of Vascular Surgery, Nanjing Drum Tower Hospital, Medical School of Nanjing University, Nanjing, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 August 2015 Accepted 17 August 2015 Available online xxx
Glutaredoxin 3 (GLRX3) is a member of monothiol glutaredoxins with a CGFS active site that has been demonstrated to function in cellular iron sensing and trafficking via its bound ironesulfur cluster. Human GLRX3 has been shown to form a dimer that binds two bridging [2Fee2S] clusters with glutathione (GSH) as a ligand, assembling a compound 2GLRX3e2[2Fee2S]e4GSH. Each iron of the ironesulfur clusters is bound to the thiols of the cysteines, one of which is from the active site of GLRX3, the other from the noncovalently bound GSH. Here, we show that the recombinant human GLRX3 isolated anaerobically from Escherichia coli can incorporate [4Fee4S] cluster in the absence of GSH, revealed by spectral and enzymatic analysis. [4Fee4S] cluster-containing GLRX3 is competent for converting iron regulatory protein 1 (apo-IRP1) into aconitase within 30 min, via intact ironesulfur cluster transfer. These in vitro studies suggest that human GLRX3 is important for cytosolic FeeS protein maturation. © 2015 Elsevier Inc. All rights reserved.
Keywords: Glutaredoxin 3 Glutathione Ironesulfur cluster Iron regulatory protein 1
1. Introduction Glutaredoxins (GLRXs) are redox proteins widely present in both prokaryotes and eukaryotes [1]. Depending on the number of active site cysteine residues, the proteins are generally divided into two groups, monothiol (C-X-X-S) and dithiol (C-X-X-C). Dithiol GLRXs have been proposed to serve as oxidative stress sensors, in contrast, monothiol GLRXs mainly act as scaffolds for [2Fee2S] cluster delivery and regulators of cellular iron homeostasis [2e4]. In monothiol human GLRX3, there is an N-terminal Trx-like domain and two tandem Grx-like domains [5]. Each of the two Grx domains has a conserved C-G-F-S active site and is supposed to incorporate a [2Fee2S] cluster with another GLRX3 monomer [6]. The cluster is ligated by the cysteine residues of the active site and the thiols of non-covalently bound glutathione (GSH) molecules as a ligand. However, the crystal structure of GLRX3 protein has not been revealed thus far.
Abbreviations: GLRX, glutaredoxin; GSH, glutathione; IRP1, iron regulatory protein 1; IRE, iron responsive element; IPTG, isopropyl thio-b-D-galactoside; EDTA, ethylenediaminetetraacetic acid; DTT, dithiothreitol; SDSePAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis. * Corresponding author. 22 Hankou Road, Nanjing, 210093 PR China. E-mail address: [email protected] (K. Li).
GLRX3 is located in the cytosol and switches between the free monomer and [2Fee2S] cluster-bound dimer, which plays a role in cellular iron sensing and trafficking [7]. The cytosolic iron regulatory protein 1 (IRP1) also serves as a cellular iron sensor. When the cellular iron increases, IRP1 is bound to the [4Fee4S] and turns into cytosolic aconitase. In contrast, if cellular iron supply is limited or ironesulfur cluster synthesis is defective, IRP1 is dissociated from the [4Fee4S] cluster, and binds to iron responsive elements (IREs) of target mRNAs of some iron related proteins, which facilitates cellular iron uptake and reduces iron storage [8,9]. In this work, we found that isolated recombinant GLRX3 prepared anaerobically from Escherichia coli contained ironesulfur clusters which could be transferred to the apo-protein IRP1 within 30 min to generate the aconitase activities. Two possible explanations were proposed: first, two [2Fee2S] clusters converted to one [4Fee4S] cluster and transferred to apo-IRP1 during the process; second, some GLRX3 protein might directly bind [4Fee4S] cluster in addition to [2Fee2S] cluster reported previously and deliver [4Fee4S] cluster to apo-IRP1. Spectral and enzymatic results confirmed the second hypothesis that GLRX3 could directly bind [4Fee4S] cluster, which raised the possibility of a role in the maturation of [4Fee4S] cluster containing proteins such as IRP1.
http://dx.doi.org/10.1016/j.bbrc.2015.08.073 0006-291X/© 2015 Elsevier Inc. All rights reserved.
Please cite this article in press as: H. Xia, et al., Human glutaredoxin 3 can bind and effectively transfer [4Fee4S] cluster to apo-iron regulatory protein 1, Biochemical and Biophysical Research Communications (2015), http://dx.doi.org/10.1016/j.bbrc.2015.08.073
2
H. Xia et al. / Biochemical and Biophysical Research Communications xxx (2015) 1e5
2. Materials and methods 2.1. Cloning of human GLRX3 For overexpression of human GLRX3 in E. coli strain BL21 (DE3) (NEB, Ipswich, MA), PCR products were generated with primers P1 (50 -TTTTAGGATCCATGGCGGCGGGGGCGGC-30 ) and P2 (50 -CCCGGTA CCTTAATTTTCTCCTCTCAG-30 ). The BamH I-Kpn I fragment of GLRX3 PCR product was cloned into the pQE-80L vector (Qiagen, Valencia, CA) to generate pQE80LeGLRX3-his for overexpression in BL21 (DE3). The construct was verified by DNA sequencing. 2.2. Protein expression and purification For overexpression of human GLRX3, plasmid pQE80LeGLRX3his was transformed into BL21 (DE3) cells, which were cultured in 1-liter flask containing 500 ml of LB medium supplemented with ampicillin (100 mg/L) in a shaker at 37 C/200 rpm until the optical density at 600 nm reached 0.8. Overexpression of GLRX3 was induced by the addition of isopropyl thio-b-D-galactoside (IPTG) to a final concentration of 0.5 mM. After that, bacteria were cultured overnight at 16 C and harvested by centrifugation. The cell pellets were resuspended in lysis buffer (50 mM sodium phosphate, pH 8.0, 300 mM NaCl, and 10 mM imidazole), followed by sonication. Soluble GLRX3 protein was purified from the clarified extract by affinity chromatography using nickelenitrilotriacetic acid (NieNTA) superflow beads (Qiagen). The protein was eluted with 250 mM imidazole and further dialyzed with Centrifugal Filtration Device (Millipore, Shanghai, China). The purity of the recombinant GLRX3 protein was checked by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDSePAGE) after the purification. Apo-GLRX3 was obtained by incubation of as-purified holo-GLRXs with EDTA and potassium ferricyanide as described previously [10]. Apo-IRP1 was overexpressed and purified in Pichia pastoris under
the control of the strong alcohol oxidase promoter (AOX1) using the method described by Li [11]. The protein concentrations were determined with Bradford assay. 2.3. FeeS cluster transfer from holo-GLRX3 to apo-IRP1 FeeS cluster transfer assays were performed in a 25 mM TriseHCl, 40 mM KCl buffer (pH 7.4) that contained 0e1 mg/ml of GLRX3 overexpressed E. coli lysates, 5 mM of dithiothreitol (DTT), and 1 mM of apo-IRP1. For control experiments, E. coli lysates transformed with empty plasmid were used. All transfer reactions were incubated in an anaerobic chamber for 0.5 h. Conversion of apo-IRP1 to the [4Fee4S]-IRP1 form was evaluated using the in-gel aconitase assay as described previously [11]. Next, 10 mg of freshly prepared FeeS cluster containing GLRX3 protein was incubated with 1 mM of apo-IRP1, 5 mM DTT under strictly anaerobic conditions, in the presence or absence of 2 mM GSH. For control experiments, holo-GLRX3 or apo-IRP1 was omitted from the mixture. The reaction was carried out for 0.5e3 h as indicated in the text, after which aconitase activities of the mixture were determined using the in-gel aconitase assay. 2.4. In vitro reconstitution of FeeS clusters on human GLRX3 in the presence or absence of GSH In the presence or absence of 3 mM GSH, 50 mM apo-GLRX3 was incubated with 1 mM Ferric ammonium citrate, 2 mM sodium sulfide, 5 mM DTT, 25 mM TriseCl, pH 8.0 and 40 mM KCl in a final volume of 1.5 ml under strictly anaerobic conditions for 3 h. Excess reagents were removed by loading the reaction mixture onto a prepacked HiTrap Q column (GE Healthcare, Shanghai) preequilibrated with 25 mM TriseCl, pH 8.0. Proteins were then eluted with a step gradient from 100 to 600 mM NaCl in 25 mM TriseCl (pH 8.0). The eluted fractions were collected and aliquots
Fig. 1. Recombinant human GLRX3 purified from E. coli contained FeeS cluster. (A) His-tagged GLRX3 was purified from E. coli BL21 (DE3) by affinity chromatography using prepacked NieNTA column. When his-tagged GLRX3 bound the column during the purification, the light blue NieNTA resin (left) turned brownish (right). (B) Eluate fractions (F4eF7) are brownish. (C) The brownish fractions contained purified GLRX3, revealed by Coomassie stain. The protein shows a molecular weight of 39 kDa, consistent with the predicted size. (D) UVevisible absorption spectrum of the purified GLRX3 (normalized to the absorbance at 280 nm) exhibited additional bands at 324, 418, 520 and 580 nm besides the protein band at 280 nm, suggesting FeeS cluster-bound GLRX3. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.).
Please cite this article in press as: H. Xia, et al., Human glutaredoxin 3 can bind and effectively transfer [4Fee4S] cluster to apo-iron regulatory protein 1, Biochemical and Biophysical Research Communications (2015), http://dx.doi.org/10.1016/j.bbrc.2015.08.073
H. Xia et al. / Biochemical and Biophysical Research Communications xxx (2015) 1e5
were analyzed using SDSePAGE followed by Coomassie stain. The fractions were then incubated with 1 mM of apo-IRP1 for 30 min. Aconitase activity was determined using the in-gel aconitase assay to estimate activation of apo-IRP1. 2.5. Spectroscopic analyses UVevisible absorption spectra and kinetic data were recorded with a UV5600-PC scanning spectrophotometer (Metash, Shanghai). All samples for spectroscopic studies were prepared in an anaerobic chamber. 3. Results 3.1. Recombinant human GLRX3 purified from E. coli contains ironesulfur cluster Human GLRX3 was cloned and the protein was heterologously expressed in E. coli BL21 (DE3). Purification by affinity chromatography using NieNTA resin yielded a satisfactory amount of soluble GLRX3 protein, with a brownish color (Fig. 1A, B). SDSePAGE analysis showed a molecular weight of 39 kDa, in agreement with the predicted size (Fig. 1C). The expression of GLRX3 protein was further confirmed by Western blot with antibody against hexahistidines (data not shown). UVevisible absorption spectrum of GLRX3 exhibited additional bands at 324, 418, 520 and 580 nm besides the protein band at 280 nm (Fig. 1D), quite distinct from
3
that observed in [2Fee2S] cluster-bound protein, for instance GLRX2, whose crystallographic structure has already been revealed [12,13]. 3.2. Human GLRX3 may transfer ironesulfur cluster to apo-protein IRP1 As GLRX3 was shown to be involved in FeeS cluster trafficking [3,7], the potential transfer of the cluster to apo-IRP1 was studied in vitro in this study. E. coli lysates containing overexpressed GLRX3 were incubated anaerobically with apo-IRP1 for 0.5 h, followed by an in-gel assay to detect if apo-IRP1 gained the aconitase activity by receiving FeeS cluster from GLRX3. Compared with the control lysates transformed with empty plasmid, lysates containing overexpressed GLRX3 resulted in a marked increase in IRP1 aconitase activity in a lysateedose-dependent manner (Fig. 2A). Further, purified brownish GLRX3 protein was incubated with apo-IRP1 in the presence or absence of GSH from 0.5 to 3 h. IRP1 quickly gained aconitase activities within 0.5 h without GSH. In the presence of 2 mM GSH, markedly lower aconitase activities of IRP1 was detected at each time point from 0.5 to 3 h (Fig. 2B), suggesting that free GSH may have an inhibitory effect on ironesulfur cluster transfer from holo-GLRX3 to apo-IRP1. Surprisingly, purified brownish GLRX3 was able to quickly convert apo-IRP1 to gain the aconitase activity, which requires a [4Fee4S] cluster. Two possible explanations were proposed: firstly, two [2Fee2S] units generated one [4Fee4S] cluster to be transferred to apo-IRP1 during the
Fig. 2. FeeS cluster of GLRX3 was transferred to apo-IRP1 to gain the aconitase activity in vitro. (A) Human GLRX3-overexpressed E. coli BL21 (DE3) lysates were incubated with apoIRP1 in buffer (pH 7.4) containing 5 mM DTT, 25 mM TriseHCl, 40 mM KCl in an anaerobic chamber for 0.5 h. For control lysate, E. coli transformed with empty plasmid were used in the reaction. Transfer of FeeS cluster was analyzed with an in-gel aconitase assay. (B) Purified brownish GLRX3 protein (10 mg from fraction 5 of Fig. 1) was incubated with apo-IRP1 in the presence or absence of 2 mM GSH. For control, GLRX3 or IRP1 protein was omitted in the reaction. As indicated, GLRX3 quickly transferred the bound FeeS cluster to apo-IRP1 within 30 min and GSH blocked the transfer.
Please cite this article in press as: H. Xia, et al., Human glutaredoxin 3 can bind and effectively transfer [4Fee4S] cluster to apo-iron regulatory protein 1, Biochemical and Biophysical Research Communications (2015), http://dx.doi.org/10.1016/j.bbrc.2015.08.073
4
H. Xia et al. / Biochemical and Biophysical Research Communications xxx (2015) 1e5
process; secondly, some of the GLRX3 protein might directly incorporate [4Fee4S] cluster and deliver the cluster to apo-IRP1. 3.3. In vitro reconstituted human GLRX3 in the presence of GSH was not able to transfer the FeeS cluster to apo-IRP1 GLRX3 was proposed to be a [2Fee2S]-bridged homodimer with GSH as a ligand [14,15]. To assess the types of FeeS clusters that GLRX3 can bind, we incubated apo-GLRX3 with iron, sulfur, DTT and GSH in an anaerobic chamber for 3 h to reconstitute FeeS clusters on GLRX3 as described previously [14]. The reaction mixture was applied to a prepacked HiTrap Q column, and proteins were eluted with increasing concentrations of NaCl from 100 to 600 mM. SDSePAGE analysis showed that GLRX3 was eluted in three fractions (E1, E2, and E4 in Fig. 3A). Of them, E1 and E2 were supposed to be a [2Fee2S] cluster-loaded form of GLRX3, with absorption bands at 320 nm and 420 nm besides at
Fig. 3. In vitro reconstituted FeeS cluster on GLRX3 in the presence of GSH was not capable of converting apo-IRP1 to aconitase within 30 min. (A) FeeS cluster was reconstituted on apo-GLRX3 in buffer (pH 8.0) with 1 mM Ferric ammonium citrate, 2 mM sodium sulfide, 5 mM DTT, 3 mM GSH, 25 mM TriseCl, and 40 mM KCl for 3 h in an anaerobic chamber. After removal of excess reagents by HiTrap Q column, the protein was eluted by increasing concentrations of NaCl from 100 to 600 mM in 25 mM Tris, pH 8.0. The collected eluate fractions were analyzed by SDSePAGE and Coomassie stain. Ctrl: purified GLRX3, FT: flow-through from the HiTrap Q column. E1, E2, E3, E4 and E5 represented eluates from sequential elution with 100, 200, 300, 400, and 600 mM NaCl in 25 mM Tris, pH 8.0. (B) UVevisible absorption spectrum of E2 fraction showed additional bands at 320 nm and 425 nm besides the protein band at 280 nm, suggesting [2Fee2S]-bound GLRX3 protein. (C) The eluate fractions from (A) were incubated with 1 mM of apo-IRP1 for 0.5 h under a strictly anaerobic condition, followed by an in-gel aconitase assay. No aconitase activity was detectable in all reactions. Pos.: positive band of aconitase.
280 nm (Fig. 3B, here shows only E2. The spectrum of E1 fraction was similar to that of E2, but the signal was weaker than E1 fraction), while E4 did not present any additional features besides of its absorption peak at 280 nm (not shown). No eluate fractions, however, could convert apo-IRP1 to a functional aconitase within 30 min (Fig. 3C). 3.4. In vitro reconstituted human GLRX3 in the absence of GSH incorporated a [4Fee4S] cluster To further confirm that GLRX3 may bind [4Fee4S] cluster, we performed the cluster reconstitution experiment under anaerobic conditions in the absence of GSH and examined if the FeeS cluster of GLRX3 could be transferred to apo-IRP1 by an in-gel aconitase assay. The results showed that the fraction (E2) eluted with buffer containing 200 mM NaCl greatly made apo-IRP1 gain the aconitase activity (Fig. 4A). To identify whether the [4Fee4S] cluster transferred to apo-IRP1 was provided by GLRX3, we analyzed the HiTrap-Q eluate fractions with SDSePAGE and found that E2 fraction contained GLRX3 protein (Fig. 4B). The UVevisible absorption spectrum of E2 exhibited an absorption shoulder around 410 nm which is typical characteristic of a [4Fee4S] cluster (Fig. 4C). Our results provided the evidence that human GLRX3 may directly bind [4Fee4S] cluster to convert apo-IRP1 to gain aconitase activity in vitro.
Fig. 4. In vitro reconstituted FeeS cluster on GLRX3 in the absence of GSH was capable of quickly converting apo-IRP1 to aconitase. Methods were the same as in Fig. 3, but in absence of GSH. (A) Fraction E2 was highly competent to make apo-IRP1 gain aconitase activity within 30 min. (B) Coomassie stain showed GLRX3 in the eluated fractions. (C) UVevisible spectrum of E2 fraction (normalized to A280) showed the characteristic absorption of [4Fee4S] at 410 nm. Ctrl, FT, E1eE5, and Pos. are defined same as in Fig. 3.
Please cite this article in press as: H. Xia, et al., Human glutaredoxin 3 can bind and effectively transfer [4Fee4S] cluster to apo-iron regulatory protein 1, Biochemical and Biophysical Research Communications (2015), http://dx.doi.org/10.1016/j.bbrc.2015.08.073
H. Xia et al. / Biochemical and Biophysical Research Communications xxx (2015) 1e5
5
4. Discussion
References
GLRX3 functions as a key regulator of cellular iron homeostasis [2,16]. Depletion of GLRX3 is associated with defects in all ironrequiring reactions, including the synthesis of ironesulfur clusters, heme, and di-iron centers [7,17]. Its important role in intracellular iron sensing and trafficking is regulated via dis-/association of ironesulfur cluster [18]. The cluster of human GLRX3 was interpreted as a [2Fee2S] cluster on the basis of the UVevisible absorption spectra [6,19] which were similar to that observed in our study. Our results provided additional evidence that human GLRX3 is capable of binding [4Fee4S] cluster in addition to [2Fee2S] cluster reported previously [6]. In vitro FeeS cluster reconstitution on GLRX3 with or without GSH resulted in totally different spectra, former one with two strong bands at 320 and 420 nm, latter one with one band at 410 nm. Recombinant GLRX3 isolated anaerobically from E. coli displayed a complicated spectrum, similar to that observed in another monothiol glutaredoxin, Saccharomyces cerevisiae Grx5, which may bind [2Fee2S], [4Fee4S], and [3Fee4S] clusters [4]. The spectrum in our study might reflect a mix of [2Fee2S] and [4Fee4S] clusters-loaded GLRX3, not excluding the existence of [3Fee4S] cluster. The quickly gained aconitase activity of IRP1 by incubation with assembled or purified brownish GLRX3 within 30 min also supports the [4Fee4S] cluster-bound GLRX3. Although [2Fee2S]-loaded scaffold protein may transfer FeeS cluster to [4Fee4S]-binding target protein, for instance IRP1 or APS reductase [20,21], it would take longer time to reach the high activity. We used the [2Fee2S]-bound human GLRX2 as a control under the same conditions as for GLRX3, it took 2 h for apo-IRP1 to gain the high aconitase activity (data not shown). Free GSH inhibited cluster transfer from [4Fee4S]-GLRX3 to apo-IRP1, probably due to that GSH facilitates the conversion to [2Fee2S] cluster containing form of FeeS protein and lowers the efficiency of cluster transfer to target proteins [22,23]. Interestingly, a very dense smeared enzymatic band just below the IRP1 aconitase was detected when E. coli lysate containing overexpressed GLRX3 was incubated with apo-IRP1 to evaluate aconitase activity by the in-gel assay (Fig. 2A). The band was likely endogenous aconitase of E. coli or a component that also reacted with the substrates in the assay to, finally, reduce MTT to formazan dyes. This band was dramatically diminished when purified GLRX3 was used to be assayed (Fig. 2B). Even though, this band might represent a target protein, which was copurified with human GLRX3 through interaction. On the basis of previous gene knockout studies in mice, GLRX3 is essential for embryogenesis [24]. Although depletion of human GLRX3 only mildly influences mitochondrial iron metabolism, it specifically affects cytosolic FeeS proteins, including IRP1 which senses iron concentration and regulates cellular iron uptake and storage [7,8,25]. The regulation of iron homeostasis by IRP1 also requires the function of FeeS cluster [8]. Here, we show that GLRX3 is competent for maturation of IRP1 in vitro. If this is the case in vivo, GLRX3, likely, transfers the [4Fee4S] cluster by interaction with apo-IRP1 or other target proteins for maturation. Future work will focus on the investigation of interaction between GLRX3 and IRP1 in vivo and how GLRX3 and IRP1 may work together to regulate cellular iron homeostasis.
[1] N. Rouhier, J. Couturier, M.K. Johnson, J.P. Jacquot, Glutaredoxins: roles in iron homeostasis, Trends Biochem. Sci. 35 (2010) 43e52. [2] E.M. Hanschmann, J.R. Godoy, C. Berndt, C. Hudemann, C.H. Lillig, Thioredoxins, glutaredoxins, and peroxiredoxinsemolecular mechanisms and health significance: from cofactors to antioxidants to redox signaling, Antioxid. Redox Signal. 19 (2013) 1539e1605. [3] T. Iwema, A. Picciocchi, D.A. Traore, J.L. Ferrer, F. Chauvat, L. Jacquamet, Structural basis for delivery of the intact [Fe2S2] cluster by monothiol glutaredoxin, Biochemistry 48 (2009) 6041e6043. [4] B. Zhang, S. Bandyopadhyay, P. Shakamuri, S.G. Naik, B.H. Huynh, J. Couturier, N. Rouhier, M.K. Johnson, Monothiol glutaredoxins can bind linear [Fe3S4]þ and [Fe4S4]2þ clusters in addition to [Fe2S2]2þ clusters: spectroscopic characterization and functional implications, J. Am. Chem. Soc. 135 (2013) 15153e15164. [5] H. Li, C.E. Outten, Monothiol CGFS glutaredoxins and BolA-like proteins: [2Fe2S] binding partners in iron homeostasis, Biochemistry 51 (2012) 4377e4389. [6] P. Haunhorst, C. Berndt, S. Eitner, J.R. Godoy, C.H. Lillig, Characterization of the human monothiol glutaredoxin 3 (PICOT) as ironesulfur protein, Biochem. Biophys. Res. Commun. 394 (2010) 372e376. [7] P. Haunhorst, E.M. Hanschmann, L. Brautigam, O. Stehling, B. Hoffmann, U. Muhlenhoff, R. Lill, C. Berndt, C.H. Lillig, Crucial function of vertebrate glutaredoxin 3 (PICOT) in iron homeostasis and hemoglobin maturation, Mol. Biol. Cell 24 (2013) 1895e1903. [8] J. Wang, G. Chen, C. Filebeen, K. Pantopoulos, Insights on regulation and function of the iron regulatory protein 1 (IRP1), Hemoglobin 32 (2008) 109e115. [9] P. Lipinski, R.R. Starzynski, The role of iron regulatory proteins (IRPs) in the regulation of systemic iron homeostasis: lessons from studies on IRP1 and IRP2 knockout mice, Postepy. Hig. Med. Dosw. 60 (2006) 322e330. [10] M.C. Kennedy, H. Beinert, The state of cluster SH and S2- of aconitase during cluster interconversions and removal. A convenient preparation of apoenzyme, J. Biol. Chem. 263 (1988) 8194e8198. [11] K. Li, W.H. Tong, R.M. Hughes, T.A. Rouault, Roles of the mammalian cytosolic cysteine desulfurase, ISCS, and scaffold protein, ISCU, in ironesulfur cluster assembly, J. Biol. Chem. 281 (2006) 12344e12351. [12] C.H. Lillig, C. Berndt, O. Vergnolle, M.E. Lonn, C. Hudemann, E. Bill, A. Holmgren, Characterization of human glutaredoxin 2 as ironesulfur protein: a possible role as redox sensor, Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 8168e8173. [13] C. Johansson, K.L. Kavanagh, O. Gileadi, U. Oppermann, Reversible sequestration of active site cysteines in a 2Fe-2S-bridged dimer provides a mechanism for glutaredoxin 2 regulation in human mitochondria, J. Biol. Chem. 282 (2007) 3077e3082. [14] H. Li, D.T. Mapolelo, N.N. Dingra, S.G. Naik, N.S. Lees, B.M. Hoffman, P.J. RiggsGelasco, B.H. Huynh, M.K. Johnson, C.E. Outten, The yeast iron regulatory proteins Grx3/4 and Fra2 form heterodimeric complexes containing a [2Fe-2S] cluster with cysteinyl and histidyl ligation, Biochemistry 48 (2009) 9569e9581. [15] H. Li, D.T. Mapolelo, N.N. Dingra, G. Keller, P.J. Riggs-Gelasco, D.R. Winge, M.K. Johnson, C.E. Outten, Histidine 103 in Fra2 is an ironesulfur cluster ligand in the [2Fe-2S] Fra2-Grx3 complex and is required for in vivo iron signaling in yeast, J. Biol. Chem. 286 (2011) 867e876. [16] S. Labbe, M.G. Khan, J.F. Jacques, Iron uptake and regulation in Schizosaccharomyces pombe, Curr. Opin. Microbiol. 16 (2013) 669e676. [17] L. Ojeda, G. Keller, U. Muhlenhoff, J.C. Rutherford, R. Lill, D.R. Winge, Role of glutaredoxin-3 and glutaredoxin-4 in the iron regulation of the Aft1 transcriptional activator in Saccharomyces cerevisiae, J. Biol. Chem. 281 (2006) 17661e17669. [18] U. Muhlenhoff, S. Molik, J.R. Godoy, M.A. Uzarska, N. Richter, A. Seubert, Y. Zhang, J. Stubbe, F. Pierrel, E. Herrero, C.H. Lillig, R. Lill, Cytosolic monothiol glutaredoxins function in intracellular iron sensing and trafficking via their bound ironesulfur cluster, Cell Metab. 12 (2010) 373e385. [19] H. Li, D.T. Mapolelo, S. Randeniya, M.K. Johnson, C.E. Outten, Human glutaredoxin 3 forms [2Fe-2S]-bridged complexes with human BolA2, Biochemistry 51 (2012) 1687e1696. [20] M. Wollenberg, C. Berndt, E. Bill, J.D. Schwenn, A. Seidler, A dimer of the FeS cluster biosynthesis protein IscA from cyanobacteria binds a [2Fe2S] cluster between two protomers and transfers it to [2Fe2S] and [4Fe4S] apo proteins, Eur. J. Biochem. 270 (2003) 1662e1671. [21] M.A. Uzarska, R. Dutkiewicz, S.A. Freibert, R. Lill, U. Muhlenhoff, The mitochondrial Hsp70 chaperone Ssq1 facilitates Fe/S cluster transfer from Isu1 to Grx5 by complex formation, Mol. Biol. Cell 24 (2013) 1830e1841. [22] L. Wang, B. Ouyang, Y. Li, Y. Feng, J.P. Jacquot, N. Rouhier, B. Xia, Glutathione regulates the transfer of ironesulfur cluster from monothiol and dithiol glutaredoxins to apo ferredoxin, Protein Cell 3 (2012) 714e721. [23] W. Qi, J.A. Cowan, Mechanism of glutaredoxin-ISU [2Fe-2S] cluster exchange, Chem. Commun. 47 (2011) 4989e4991. [24] N.H. Cheng, W. Zhang, W.Q. Chen, J. Jin, X. Cui, N.F. Butte, L. Chan, K.D. Hirschi, A mammalian monothiol glutaredoxin, Grx3, is critical for cell cycle progression during embryogenesis, FEBS J. 278 (2011) 2525e2539. [25] N. Wilkinson, K. Pantopoulos, IRP1 regulates erythropoiesis and systemic iron homeostasis by controlling HIF2alpha mRNA translation, Blood 122 (2013) 1658e1668.
Competing interest All of the authors of our manuscript declare that we have no competing interests. Acknowledgments This work was supported by grants from the National Natural Science Foundation of China (No. 31071085, 31371060).
Please cite this article in press as: H. Xia, et al., Human glutaredoxin 3 can bind and effectively transfer [4Fee4S] cluster to apo-iron regulatory protein 1, Biochemical and Biophysical Research Communications (2015), http://dx.doi.org/10.1016/j.bbrc.2015.08.073
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Human glutaredoxin 3 can bind and effectively transfer [4Fee4S] cluster to apo-iron regulatory protein 1 Haiyan Xia a, Binghua Li a, Zhou Zhang a, Qi Wang b, Tong Qiao b, Kuanyu Li a, * a b
Jiangsu Key Laboratory for Molecular Medicine, Medical School of Nanjing University, Nanjing, China Department of Vascular Surgery, Nanjing Drum Tower Hospital, Medical School of Nanjing University, Nanjing, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 August 2015 Accepted 17 August 2015 Available online xxx
Glutaredoxin 3 (GLRX3) is a member of monothiol glutaredoxins with a CGFS active site that has been demonstrated to function in cellular iron sensing and trafficking via its bound ironesulfur cluster. Human GLRX3 has been shown to form a dimer that binds two bridging [2Fee2S] clusters with glutathione (GSH) as a ligand, assembling a compound 2GLRX3e2[2Fee2S]e4GSH. Each iron of the ironesulfur clusters is bound to the thiols of the cysteines, one of which is from the active site of GLRX3, the other from the noncovalently bound GSH. Here, we show that the recombinant human GLRX3 isolated anaerobically from Escherichia coli can incorporate [4Fee4S] cluster in the absence of GSH, revealed by spectral and enzymatic analysis. [4Fee4S] cluster-containing GLRX3 is competent for converting iron regulatory protein 1 (apo-IRP1) into aconitase within 30 min, via intact ironesulfur cluster transfer. These in vitro studies suggest that human GLRX3 is important for cytosolic FeeS protein maturation. © 2015 Elsevier Inc. All rights reserved.
Keywords: Glutaredoxin 3 Glutathione Ironesulfur cluster Iron regulatory protein 1
1. Introduction Glutaredoxins (GLRXs) are redox proteins widely present in both prokaryotes and eukaryotes [1]. Depending on the number of active site cysteine residues, the proteins are generally divided into two groups, monothiol (C-X-X-S) and dithiol (C-X-X-C). Dithiol GLRXs have been proposed to serve as oxidative stress sensors, in contrast, monothiol GLRXs mainly act as scaffolds for [2Fee2S] cluster delivery and regulators of cellular iron homeostasis [2e4]. In monothiol human GLRX3, there is an N-terminal Trx-like domain and two tandem Grx-like domains [5]. Each of the two Grx domains has a conserved C-G-F-S active site and is supposed to incorporate a [2Fee2S] cluster with another GLRX3 monomer [6]. The cluster is ligated by the cysteine residues of the active site and the thiols of non-covalently bound glutathione (GSH) molecules as a ligand. However, the crystal structure of GLRX3 protein has not been revealed thus far.
Abbreviations: GLRX, glutaredoxin; GSH, glutathione; IRP1, iron regulatory protein 1; IRE, iron responsive element; IPTG, isopropyl thio-b-D-galactoside; EDTA, ethylenediaminetetraacetic acid; DTT, dithiothreitol; SDSePAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis. * Corresponding author. 22 Hankou Road, Nanjing, 210093 PR China. E-mail address: [email protected] (K. Li).
GLRX3 is located in the cytosol and switches between the free monomer and [2Fee2S] cluster-bound dimer, which plays a role in cellular iron sensing and trafficking [7]. The cytosolic iron regulatory protein 1 (IRP1) also serves as a cellular iron sensor. When the cellular iron increases, IRP1 is bound to the [4Fee4S] and turns into cytosolic aconitase. In contrast, if cellular iron supply is limited or ironesulfur cluster synthesis is defective, IRP1 is dissociated from the [4Fee4S] cluster, and binds to iron responsive elements (IREs) of target mRNAs of some iron related proteins, which facilitates cellular iron uptake and reduces iron storage [8,9]. In this work, we found that isolated recombinant GLRX3 prepared anaerobically from Escherichia coli contained ironesulfur clusters which could be transferred to the apo-protein IRP1 within 30 min to generate the aconitase activities. Two possible explanations were proposed: first, two [2Fee2S] clusters converted to one [4Fee4S] cluster and transferred to apo-IRP1 during the process; second, some GLRX3 protein might directly bind [4Fee4S] cluster in addition to [2Fee2S] cluster reported previously and deliver [4Fee4S] cluster to apo-IRP1. Spectral and enzymatic results confirmed the second hypothesis that GLRX3 could directly bind [4Fee4S] cluster, which raised the possibility of a role in the maturation of [4Fee4S] cluster containing proteins such as IRP1.
http://dx.doi.org/10.1016/j.bbrc.2015.08.073 0006-291X/© 2015 Elsevier Inc. All rights reserved.
Please cite this article in press as: H. Xia, et al., Human glutaredoxin 3 can bind and effectively transfer [4Fee4S] cluster to apo-iron regulatory protein 1, Biochemical and Biophysical Research Communications (2015), http://dx.doi.org/10.1016/j.bbrc.2015.08.073
2
H. Xia et al. / Biochemical and Biophysical Research Communications xxx (2015) 1e5
2. Materials and methods 2.1. Cloning of human GLRX3 For overexpression of human GLRX3 in E. coli strain BL21 (DE3) (NEB, Ipswich, MA), PCR products were generated with primers P1 (50 -TTTTAGGATCCATGGCGGCGGGGGCGGC-30 ) and P2 (50 -CCCGGTA CCTTAATTTTCTCCTCTCAG-30 ). The BamH I-Kpn I fragment of GLRX3 PCR product was cloned into the pQE-80L vector (Qiagen, Valencia, CA) to generate pQE80LeGLRX3-his for overexpression in BL21 (DE3). The construct was verified by DNA sequencing. 2.2. Protein expression and purification For overexpression of human GLRX3, plasmid pQE80LeGLRX3his was transformed into BL21 (DE3) cells, which were cultured in 1-liter flask containing 500 ml of LB medium supplemented with ampicillin (100 mg/L) in a shaker at 37 C/200 rpm until the optical density at 600 nm reached 0.8. Overexpression of GLRX3 was induced by the addition of isopropyl thio-b-D-galactoside (IPTG) to a final concentration of 0.5 mM. After that, bacteria were cultured overnight at 16 C and harvested by centrifugation. The cell pellets were resuspended in lysis buffer (50 mM sodium phosphate, pH 8.0, 300 mM NaCl, and 10 mM imidazole), followed by sonication. Soluble GLRX3 protein was purified from the clarified extract by affinity chromatography using nickelenitrilotriacetic acid (NieNTA) superflow beads (Qiagen). The protein was eluted with 250 mM imidazole and further dialyzed with Centrifugal Filtration Device (Millipore, Shanghai, China). The purity of the recombinant GLRX3 protein was checked by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDSePAGE) after the purification. Apo-GLRX3 was obtained by incubation of as-purified holo-GLRXs with EDTA and potassium ferricyanide as described previously [10]. Apo-IRP1 was overexpressed and purified in Pichia pastoris under
the control of the strong alcohol oxidase promoter (AOX1) using the method described by Li [11]. The protein concentrations were determined with Bradford assay. 2.3. FeeS cluster transfer from holo-GLRX3 to apo-IRP1 FeeS cluster transfer assays were performed in a 25 mM TriseHCl, 40 mM KCl buffer (pH 7.4) that contained 0e1 mg/ml of GLRX3 overexpressed E. coli lysates, 5 mM of dithiothreitol (DTT), and 1 mM of apo-IRP1. For control experiments, E. coli lysates transformed with empty plasmid were used. All transfer reactions were incubated in an anaerobic chamber for 0.5 h. Conversion of apo-IRP1 to the [4Fee4S]-IRP1 form was evaluated using the in-gel aconitase assay as described previously [11]. Next, 10 mg of freshly prepared FeeS cluster containing GLRX3 protein was incubated with 1 mM of apo-IRP1, 5 mM DTT under strictly anaerobic conditions, in the presence or absence of 2 mM GSH. For control experiments, holo-GLRX3 or apo-IRP1 was omitted from the mixture. The reaction was carried out for 0.5e3 h as indicated in the text, after which aconitase activities of the mixture were determined using the in-gel aconitase assay. 2.4. In vitro reconstitution of FeeS clusters on human GLRX3 in the presence or absence of GSH In the presence or absence of 3 mM GSH, 50 mM apo-GLRX3 was incubated with 1 mM Ferric ammonium citrate, 2 mM sodium sulfide, 5 mM DTT, 25 mM TriseCl, pH 8.0 and 40 mM KCl in a final volume of 1.5 ml under strictly anaerobic conditions for 3 h. Excess reagents were removed by loading the reaction mixture onto a prepacked HiTrap Q column (GE Healthcare, Shanghai) preequilibrated with 25 mM TriseCl, pH 8.0. Proteins were then eluted with a step gradient from 100 to 600 mM NaCl in 25 mM TriseCl (pH 8.0). The eluted fractions were collected and aliquots
Fig. 1. Recombinant human GLRX3 purified from E. coli contained FeeS cluster. (A) His-tagged GLRX3 was purified from E. coli BL21 (DE3) by affinity chromatography using prepacked NieNTA column. When his-tagged GLRX3 bound the column during the purification, the light blue NieNTA resin (left) turned brownish (right). (B) Eluate fractions (F4eF7) are brownish. (C) The brownish fractions contained purified GLRX3, revealed by Coomassie stain. The protein shows a molecular weight of 39 kDa, consistent with the predicted size. (D) UVevisible absorption spectrum of the purified GLRX3 (normalized to the absorbance at 280 nm) exhibited additional bands at 324, 418, 520 and 580 nm besides the protein band at 280 nm, suggesting FeeS cluster-bound GLRX3. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.).
Please cite this article in press as: H. Xia, et al., Human glutaredoxin 3 can bind and effectively transfer [4Fee4S] cluster to apo-iron regulatory protein 1, Biochemical and Biophysical Research Communications (2015), http://dx.doi.org/10.1016/j.bbrc.2015.08.073
H. Xia et al. / Biochemical and Biophysical Research Communications xxx (2015) 1e5
were analyzed using SDSePAGE followed by Coomassie stain. The fractions were then incubated with 1 mM of apo-IRP1 for 30 min. Aconitase activity was determined using the in-gel aconitase assay to estimate activation of apo-IRP1. 2.5. Spectroscopic analyses UVevisible absorption spectra and kinetic data were recorded with a UV5600-PC scanning spectrophotometer (Metash, Shanghai). All samples for spectroscopic studies were prepared in an anaerobic chamber. 3. Results 3.1. Recombinant human GLRX3 purified from E. coli contains ironesulfur cluster Human GLRX3 was cloned and the protein was heterologously expressed in E. coli BL21 (DE3). Purification by affinity chromatography using NieNTA resin yielded a satisfactory amount of soluble GLRX3 protein, with a brownish color (Fig. 1A, B). SDSePAGE analysis showed a molecular weight of 39 kDa, in agreement with the predicted size (Fig. 1C). The expression of GLRX3 protein was further confirmed by Western blot with antibody against hexahistidines (data not shown). UVevisible absorption spectrum of GLRX3 exhibited additional bands at 324, 418, 520 and 580 nm besides the protein band at 280 nm (Fig. 1D), quite distinct from
3
that observed in [2Fee2S] cluster-bound protein, for instance GLRX2, whose crystallographic structure has already been revealed [12,13]. 3.2. Human GLRX3 may transfer ironesulfur cluster to apo-protein IRP1 As GLRX3 was shown to be involved in FeeS cluster trafficking [3,7], the potential transfer of the cluster to apo-IRP1 was studied in vitro in this study. E. coli lysates containing overexpressed GLRX3 were incubated anaerobically with apo-IRP1 for 0.5 h, followed by an in-gel assay to detect if apo-IRP1 gained the aconitase activity by receiving FeeS cluster from GLRX3. Compared with the control lysates transformed with empty plasmid, lysates containing overexpressed GLRX3 resulted in a marked increase in IRP1 aconitase activity in a lysateedose-dependent manner (Fig. 2A). Further, purified brownish GLRX3 protein was incubated with apo-IRP1 in the presence or absence of GSH from 0.5 to 3 h. IRP1 quickly gained aconitase activities within 0.5 h without GSH. In the presence of 2 mM GSH, markedly lower aconitase activities of IRP1 was detected at each time point from 0.5 to 3 h (Fig. 2B), suggesting that free GSH may have an inhibitory effect on ironesulfur cluster transfer from holo-GLRX3 to apo-IRP1. Surprisingly, purified brownish GLRX3 was able to quickly convert apo-IRP1 to gain the aconitase activity, which requires a [4Fee4S] cluster. Two possible explanations were proposed: firstly, two [2Fee2S] units generated one [4Fee4S] cluster to be transferred to apo-IRP1 during the
Fig. 2. FeeS cluster of GLRX3 was transferred to apo-IRP1 to gain the aconitase activity in vitro. (A) Human GLRX3-overexpressed E. coli BL21 (DE3) lysates were incubated with apoIRP1 in buffer (pH 7.4) containing 5 mM DTT, 25 mM TriseHCl, 40 mM KCl in an anaerobic chamber for 0.5 h. For control lysate, E. coli transformed with empty plasmid were used in the reaction. Transfer of FeeS cluster was analyzed with an in-gel aconitase assay. (B) Purified brownish GLRX3 protein (10 mg from fraction 5 of Fig. 1) was incubated with apo-IRP1 in the presence or absence of 2 mM GSH. For control, GLRX3 or IRP1 protein was omitted in the reaction. As indicated, GLRX3 quickly transferred the bound FeeS cluster to apo-IRP1 within 30 min and GSH blocked the transfer.
Please cite this article in press as: H. Xia, et al., Human glutaredoxin 3 can bind and effectively transfer [4Fee4S] cluster to apo-iron regulatory protein 1, Biochemical and Biophysical Research Communications (2015), http://dx.doi.org/10.1016/j.bbrc.2015.08.073
4
H. Xia et al. / Biochemical and Biophysical Research Communications xxx (2015) 1e5
process; secondly, some of the GLRX3 protein might directly incorporate [4Fee4S] cluster and deliver the cluster to apo-IRP1. 3.3. In vitro reconstituted human GLRX3 in the presence of GSH was not able to transfer the FeeS cluster to apo-IRP1 GLRX3 was proposed to be a [2Fee2S]-bridged homodimer with GSH as a ligand [14,15]. To assess the types of FeeS clusters that GLRX3 can bind, we incubated apo-GLRX3 with iron, sulfur, DTT and GSH in an anaerobic chamber for 3 h to reconstitute FeeS clusters on GLRX3 as described previously [14]. The reaction mixture was applied to a prepacked HiTrap Q column, and proteins were eluted with increasing concentrations of NaCl from 100 to 600 mM. SDSePAGE analysis showed that GLRX3 was eluted in three fractions (E1, E2, and E4 in Fig. 3A). Of them, E1 and E2 were supposed to be a [2Fee2S] cluster-loaded form of GLRX3, with absorption bands at 320 nm and 420 nm besides at
Fig. 3. In vitro reconstituted FeeS cluster on GLRX3 in the presence of GSH was not capable of converting apo-IRP1 to aconitase within 30 min. (A) FeeS cluster was reconstituted on apo-GLRX3 in buffer (pH 8.0) with 1 mM Ferric ammonium citrate, 2 mM sodium sulfide, 5 mM DTT, 3 mM GSH, 25 mM TriseCl, and 40 mM KCl for 3 h in an anaerobic chamber. After removal of excess reagents by HiTrap Q column, the protein was eluted by increasing concentrations of NaCl from 100 to 600 mM in 25 mM Tris, pH 8.0. The collected eluate fractions were analyzed by SDSePAGE and Coomassie stain. Ctrl: purified GLRX3, FT: flow-through from the HiTrap Q column. E1, E2, E3, E4 and E5 represented eluates from sequential elution with 100, 200, 300, 400, and 600 mM NaCl in 25 mM Tris, pH 8.0. (B) UVevisible absorption spectrum of E2 fraction showed additional bands at 320 nm and 425 nm besides the protein band at 280 nm, suggesting [2Fee2S]-bound GLRX3 protein. (C) The eluate fractions from (A) were incubated with 1 mM of apo-IRP1 for 0.5 h under a strictly anaerobic condition, followed by an in-gel aconitase assay. No aconitase activity was detectable in all reactions. Pos.: positive band of aconitase.
280 nm (Fig. 3B, here shows only E2. The spectrum of E1 fraction was similar to that of E2, but the signal was weaker than E1 fraction), while E4 did not present any additional features besides of its absorption peak at 280 nm (not shown). No eluate fractions, however, could convert apo-IRP1 to a functional aconitase within 30 min (Fig. 3C). 3.4. In vitro reconstituted human GLRX3 in the absence of GSH incorporated a [4Fee4S] cluster To further confirm that GLRX3 may bind [4Fee4S] cluster, we performed the cluster reconstitution experiment under anaerobic conditions in the absence of GSH and examined if the FeeS cluster of GLRX3 could be transferred to apo-IRP1 by an in-gel aconitase assay. The results showed that the fraction (E2) eluted with buffer containing 200 mM NaCl greatly made apo-IRP1 gain the aconitase activity (Fig. 4A). To identify whether the [4Fee4S] cluster transferred to apo-IRP1 was provided by GLRX3, we analyzed the HiTrap-Q eluate fractions with SDSePAGE and found that E2 fraction contained GLRX3 protein (Fig. 4B). The UVevisible absorption spectrum of E2 exhibited an absorption shoulder around 410 nm which is typical characteristic of a [4Fee4S] cluster (Fig. 4C). Our results provided the evidence that human GLRX3 may directly bind [4Fee4S] cluster to convert apo-IRP1 to gain aconitase activity in vitro.
Fig. 4. In vitro reconstituted FeeS cluster on GLRX3 in the absence of GSH was capable of quickly converting apo-IRP1 to aconitase. Methods were the same as in Fig. 3, but in absence of GSH. (A) Fraction E2 was highly competent to make apo-IRP1 gain aconitase activity within 30 min. (B) Coomassie stain showed GLRX3 in the eluated fractions. (C) UVevisible spectrum of E2 fraction (normalized to A280) showed the characteristic absorption of [4Fee4S] at 410 nm. Ctrl, FT, E1eE5, and Pos. are defined same as in Fig. 3.
Please cite this article in press as: H. Xia, et al., Human glutaredoxin 3 can bind and effectively transfer [4Fee4S] cluster to apo-iron regulatory protein 1, Biochemical and Biophysical Research Communications (2015), http://dx.doi.org/10.1016/j.bbrc.2015.08.073
H. Xia et al. / Biochemical and Biophysical Research Communications xxx (2015) 1e5
5
4. Discussion
References
GLRX3 functions as a key regulator of cellular iron homeostasis [2,16]. Depletion of GLRX3 is associated with defects in all ironrequiring reactions, including the synthesis of ironesulfur clusters, heme, and di-iron centers [7,17]. Its important role in intracellular iron sensing and trafficking is regulated via dis-/association of ironesulfur cluster [18]. The cluster of human GLRX3 was interpreted as a [2Fee2S] cluster on the basis of the UVevisible absorption spectra [6,19] which were similar to that observed in our study. Our results provided additional evidence that human GLRX3 is capable of binding [4Fee4S] cluster in addition to [2Fee2S] cluster reported previously [6]. In vitro FeeS cluster reconstitution on GLRX3 with or without GSH resulted in totally different spectra, former one with two strong bands at 320 and 420 nm, latter one with one band at 410 nm. Recombinant GLRX3 isolated anaerobically from E. coli displayed a complicated spectrum, similar to that observed in another monothiol glutaredoxin, Saccharomyces cerevisiae Grx5, which may bind [2Fee2S], [4Fee4S], and [3Fee4S] clusters [4]. The spectrum in our study might reflect a mix of [2Fee2S] and [4Fee4S] clusters-loaded GLRX3, not excluding the existence of [3Fee4S] cluster. The quickly gained aconitase activity of IRP1 by incubation with assembled or purified brownish GLRX3 within 30 min also supports the [4Fee4S] cluster-bound GLRX3. Although [2Fee2S]-loaded scaffold protein may transfer FeeS cluster to [4Fee4S]-binding target protein, for instance IRP1 or APS reductase [20,21], it would take longer time to reach the high activity. We used the [2Fee2S]-bound human GLRX2 as a control under the same conditions as for GLRX3, it took 2 h for apo-IRP1 to gain the high aconitase activity (data not shown). Free GSH inhibited cluster transfer from [4Fee4S]-GLRX3 to apo-IRP1, probably due to that GSH facilitates the conversion to [2Fee2S] cluster containing form of FeeS protein and lowers the efficiency of cluster transfer to target proteins [22,23]. Interestingly, a very dense smeared enzymatic band just below the IRP1 aconitase was detected when E. coli lysate containing overexpressed GLRX3 was incubated with apo-IRP1 to evaluate aconitase activity by the in-gel assay (Fig. 2A). The band was likely endogenous aconitase of E. coli or a component that also reacted with the substrates in the assay to, finally, reduce MTT to formazan dyes. This band was dramatically diminished when purified GLRX3 was used to be assayed (Fig. 2B). Even though, this band might represent a target protein, which was copurified with human GLRX3 through interaction. On the basis of previous gene knockout studies in mice, GLRX3 is essential for embryogenesis [24]. Although depletion of human GLRX3 only mildly influences mitochondrial iron metabolism, it specifically affects cytosolic FeeS proteins, including IRP1 which senses iron concentration and regulates cellular iron uptake and storage [7,8,25]. The regulation of iron homeostasis by IRP1 also requires the function of FeeS cluster [8]. Here, we show that GLRX3 is competent for maturation of IRP1 in vitro. If this is the case in vivo, GLRX3, likely, transfers the [4Fee4S] cluster by interaction with apo-IRP1 or other target proteins for maturation. Future work will focus on the investigation of interaction between GLRX3 and IRP1 in vivo and how GLRX3 and IRP1 may work together to regulate cellular iron homeostasis.
[1] N. Rouhier, J. Couturier, M.K. Johnson, J.P. Jacquot, Glutaredoxins: roles in iron homeostasis, Trends Biochem. Sci. 35 (2010) 43e52. [2] E.M. Hanschmann, J.R. Godoy, C. Berndt, C. Hudemann, C.H. Lillig, Thioredoxins, glutaredoxins, and peroxiredoxinsemolecular mechanisms and health significance: from cofactors to antioxidants to redox signaling, Antioxid. Redox Signal. 19 (2013) 1539e1605. [3] T. Iwema, A. Picciocchi, D.A. Traore, J.L. Ferrer, F. Chauvat, L. Jacquamet, Structural basis for delivery of the intact [Fe2S2] cluster by monothiol glutaredoxin, Biochemistry 48 (2009) 6041e6043. [4] B. Zhang, S. Bandyopadhyay, P. Shakamuri, S.G. Naik, B.H. Huynh, J. Couturier, N. Rouhier, M.K. Johnson, Monothiol glutaredoxins can bind linear [Fe3S4]þ and [Fe4S4]2þ clusters in addition to [Fe2S2]2þ clusters: spectroscopic characterization and functional implications, J. Am. Chem. Soc. 135 (2013) 15153e15164. [5] H. Li, C.E. Outten, Monothiol CGFS glutaredoxins and BolA-like proteins: [2Fe2S] binding partners in iron homeostasis, Biochemistry 51 (2012) 4377e4389. [6] P. Haunhorst, C. Berndt, S. Eitner, J.R. Godoy, C.H. Lillig, Characterization of the human monothiol glutaredoxin 3 (PICOT) as ironesulfur protein, Biochem. Biophys. Res. Commun. 394 (2010) 372e376. [7] P. Haunhorst, E.M. Hanschmann, L. Brautigam, O. Stehling, B. Hoffmann, U. Muhlenhoff, R. Lill, C. Berndt, C.H. Lillig, Crucial function of vertebrate glutaredoxin 3 (PICOT) in iron homeostasis and hemoglobin maturation, Mol. Biol. Cell 24 (2013) 1895e1903. [8] J. Wang, G. Chen, C. Filebeen, K. Pantopoulos, Insights on regulation and function of the iron regulatory protein 1 (IRP1), Hemoglobin 32 (2008) 109e115. [9] P. Lipinski, R.R. Starzynski, The role of iron regulatory proteins (IRPs) in the regulation of systemic iron homeostasis: lessons from studies on IRP1 and IRP2 knockout mice, Postepy. Hig. Med. Dosw. 60 (2006) 322e330. [10] M.C. Kennedy, H. Beinert, The state of cluster SH and S2- of aconitase during cluster interconversions and removal. A convenient preparation of apoenzyme, J. Biol. Chem. 263 (1988) 8194e8198. [11] K. Li, W.H. Tong, R.M. Hughes, T.A. Rouault, Roles of the mammalian cytosolic cysteine desulfurase, ISCS, and scaffold protein, ISCU, in ironesulfur cluster assembly, J. Biol. Chem. 281 (2006) 12344e12351. [12] C.H. Lillig, C. Berndt, O. Vergnolle, M.E. Lonn, C. Hudemann, E. Bill, A. Holmgren, Characterization of human glutaredoxin 2 as ironesulfur protein: a possible role as redox sensor, Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 8168e8173. [13] C. Johansson, K.L. Kavanagh, O. Gileadi, U. Oppermann, Reversible sequestration of active site cysteines in a 2Fe-2S-bridged dimer provides a mechanism for glutaredoxin 2 regulation in human mitochondria, J. Biol. Chem. 282 (2007) 3077e3082. [14] H. Li, D.T. Mapolelo, N.N. Dingra, S.G. Naik, N.S. Lees, B.M. Hoffman, P.J. RiggsGelasco, B.H. Huynh, M.K. Johnson, C.E. Outten, The yeast iron regulatory proteins Grx3/4 and Fra2 form heterodimeric complexes containing a [2Fe-2S] cluster with cysteinyl and histidyl ligation, Biochemistry 48 (2009) 9569e9581. [15] H. Li, D.T. Mapolelo, N.N. Dingra, G. Keller, P.J. Riggs-Gelasco, D.R. Winge, M.K. Johnson, C.E. Outten, Histidine 103 in Fra2 is an ironesulfur cluster ligand in the [2Fe-2S] Fra2-Grx3 complex and is required for in vivo iron signaling in yeast, J. Biol. Chem. 286 (2011) 867e876. [16] S. Labbe, M.G. Khan, J.F. Jacques, Iron uptake and regulation in Schizosaccharomyces pombe, Curr. Opin. Microbiol. 16 (2013) 669e676. [17] L. Ojeda, G. Keller, U. Muhlenhoff, J.C. Rutherford, R. Lill, D.R. Winge, Role of glutaredoxin-3 and glutaredoxin-4 in the iron regulation of the Aft1 transcriptional activator in Saccharomyces cerevisiae, J. Biol. Chem. 281 (2006) 17661e17669. [18] U. Muhlenhoff, S. Molik, J.R. Godoy, M.A. Uzarska, N. Richter, A. Seubert, Y. Zhang, J. Stubbe, F. Pierrel, E. Herrero, C.H. Lillig, R. Lill, Cytosolic monothiol glutaredoxins function in intracellular iron sensing and trafficking via their bound ironesulfur cluster, Cell Metab. 12 (2010) 373e385. [19] H. Li, D.T. Mapolelo, S. Randeniya, M.K. Johnson, C.E. Outten, Human glutaredoxin 3 forms [2Fe-2S]-bridged complexes with human BolA2, Biochemistry 51 (2012) 1687e1696. [20] M. Wollenberg, C. Berndt, E. Bill, J.D. Schwenn, A. Seidler, A dimer of the FeS cluster biosynthesis protein IscA from cyanobacteria binds a [2Fe2S] cluster between two protomers and transfers it to [2Fe2S] and [4Fe4S] apo proteins, Eur. J. Biochem. 270 (2003) 1662e1671. [21] M.A. Uzarska, R. Dutkiewicz, S.A. Freibert, R. Lill, U. Muhlenhoff, The mitochondrial Hsp70 chaperone Ssq1 facilitates Fe/S cluster transfer from Isu1 to Grx5 by complex formation, Mol. Biol. Cell 24 (2013) 1830e1841. [22] L. Wang, B. Ouyang, Y. Li, Y. Feng, J.P. Jacquot, N. Rouhier, B. Xia, Glutathione regulates the transfer of ironesulfur cluster from monothiol and dithiol glutaredoxins to apo ferredoxin, Protein Cell 3 (2012) 714e721. [23] W. Qi, J.A. Cowan, Mechanism of glutaredoxin-ISU [2Fe-2S] cluster exchange, Chem. Commun. 47 (2011) 4989e4991. [24] N.H. Cheng, W. Zhang, W.Q. Chen, J. Jin, X. Cui, N.F. Butte, L. Chan, K.D. Hirschi, A mammalian monothiol glutaredoxin, Grx3, is critical for cell cycle progression during embryogenesis, FEBS J. 278 (2011) 2525e2539. [25] N. Wilkinson, K. Pantopoulos, IRP1 regulates erythropoiesis and systemic iron homeostasis by controlling HIF2alpha mRNA translation, Blood 122 (2013) 1658e1668.
Competing interest All of the authors of our manuscript declare that we have no competing interests. Acknowledgments This work was supported by grants from the National Natural Science Foundation of China (No. 31071085, 31371060).
Please cite this article in press as: H. Xia, et al., Human glutaredoxin 3 can bind and effectively transfer [4Fee4S] cluster to apo-iron regulatory protein 1, Biochemical and Biophysical Research Communications (2015), http://dx.doi.org/10.1016/j.bbrc.2015.08.073