Journal of Inorganic Biochemistry 148 (2015) 57–61
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry journal homepage: www.elsevier.com/locate/jinorgbio
Photoinduced reduction of the medial FeS center in the hydrogenase small subunit HupS from Nostoc punctiforme Patrícia Raleiras, Leif Hammarström, Peter Lindblad, Stenbjörn Styring ⁎, Ann Magnuson ⁎⁎ Department of Chemistry, Ångström Laboratory, Uppsala University, Box 523, SE-75120 Uppsala, Sweden
a r t i c l e
i n f o
Article history: Received 19 December 2014 Received in revised form 22 March 2015 Accepted 24 March 2015 Available online 14 April 2015 Keywords: Hydrogenase Iron–sulfur Electron transfer Flash photolysis Electron paramagnetic resonance
a b s t r a c t The small subunit from the NiFe uptake hydrogenase, HupSL, in the cyanobacterium Nostoc punctiforme ATCC 29133, has been isolated in the absence of the large subunit (P. Raleiras, P. Kellers, P. Lindblad, S. Styring, A. Magnuson, J. Biol. Chem. 288 (2013) 18,345–18,352). Here, we have used flash photolysis to reduce the iron-sulfur clusters in the isolated small subunit, HupS. We used ascorbate as electron donor to the photogenerated excited state of Ru(II)-trisbipyridine (Ru(bpy)3), to generate Ru(I)(bpy)3 as reducing agent. Our results show that the isolated small subunit can be reduced by the Ru(I)(bpy)3 generated through flash photolysis. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Hydrogenases are metalloenzymes that catalyze the reversible oxidation of H2: þ
−
H2 ⇌2H þ 2e :
Hydrogenases belong to either one of three categories: the [NiFe] hydrogenases, containing an inorganic nickel–iron complex in the catalytic center, the [FeFe] hydrogenases, containing a binuclear iron complex, and the [Fe] hydrogenases, containing a mononuclear iron center [1]. The [NiFe] hydrogenases consist of two protein subunits, referred to as the large and small subunit respectively [2,3]. The large subunit contains the active site where H2 oxidation or production is catalyzed by the [NiFe] complex. The typical electron transfer motif in the small subunit is three iron– sulfur (FeS) clusters: a proximal [4Fe–4S] cluster closest to the active site, a medial [3Fe–4S] and a distal [4Fe–4S] cluster. The three FeS clusters are aligned so that they connect the protein surface with the active site in the large subunit, and it is generally believed that electron transfer in the small subunit goes via this array of FeS clusters. Depending on whether the hydrogenase catalyzes proton reduction or hydrogen oxidation, the electrons are transported to, or away from the active site. ⁎ Corresponding author. ⁎⁎ Corresponding author. Fax: +46 18 471 6844. E-mail addresses: [email protected] (S. Styring), [email protected] (A. Magnuson).
http://dx.doi.org/10.1016/j.jinorgbio.2015.03.018 0162-0134/© 2015 Elsevier Inc. All rights reserved.
Cyanobacteria are phototrophic microorganisms that can produce H2 from solar energy and water. They are therefore attractive targets for efforts to improve their productivity via genetic engineering. All cyanobacteria with known gene sequence possess at least one hydrogenase copy. In cases where more than one copy exists within a genus, they usually differ in metabolic purpose and activity [4]. In nitrogen-fixing strains, an uptake hydrogenase is usually present which recycles H2 that is evolved in the nitrogenase reaction. Other, bidirectional, hydrogenases are known to perform both hydrogen evolution and uptake under different metabolic conditions. Only one cyanobacterial hydrogenase, the bidirectional hydrogenase from Synechocystis PCC 6803, has so far been isolated and characterized [5]. The filamentous, heterocystous cyanobacterium Nostoc punctiforme ATCC 29133 has only one hydrogenase, the [NiFe] uptake hydrogenase HupSL. We have previously isolated the small subunit, HupS, in the absence of the large subunit by expressing it in Escherichia coli as a fusion protein, from now on referred to as f–HupS [6]. This enabled us to characterize the FeS centers in the small subunit without spectroscopic or magnetic interference from the active site. Since HupSL is a H2-oxidizing enzyme, the electron transfer in HupS is expected to be directed away from the active site. Due to the linear arrangement of the FeS clusters in HupS, the electron transfer route from the active site is believed to be first to the proximal cluster, then to the medial and distal clusters, and finally to the native redox partner. The relative reduction potentials of the three FeS clusters have been suggested to play an important role in steering the electron transfer directionality in known NiFe hydrogenases [7,8]. The reduction potentials of the proximal and distal [4Fe–4S] clusters have been determined to −290 to −360 mV [7,9–13]. The medial [3Fe–4S] cluster on the other
58
P. Raleiras et al. / Journal of Inorganic Biochemistry 148 (2015) 57–61
hand, is more oxidizing with a potential of −70–+100 mV in known cases. It might thus act as an electron trap in the electron transfer chain [14,15]. It remains unclear whether electron transfer proceeds via the medial cluster, or if the electron tunnels the ca. 17 Å distance from the proximal to the distal cluster avoiding the medial cluster entirely. All previous investigations of electron transfer reactions in NiFe hydrogenases have been performed in the intact enzyme including the large subunit. The spectroscopic and redox properties of the small subunit have therefore been affected by the presence of the nearby active site. Our aim with this study is to investigate if the FeS clusters in the isolated f-HupS are accessible for photo-induced reduction from the surface of the protein. Flash photolysis using ruthenium-based photosensitizers, has been used for investigating electron transfer pathways in a number of metalloproteins [16–18] and recently to initiate turnover in hydrogenases [19–21]. We used ascorbate as electron donor to the photogenerated excited state of Ru(II)-trisbipyridine (Ru(bpy)3), to generate Ru(I)(bpy)3 as reducing agent, and monitored photoreduction of f-HupS using electron paramagnetic resonance (EPR) spectroscopy. Our results show that the isolated f-HupS can be reduced by flash photolysis. 2. Materials and methods 2.1. Protein expression and purification The HupS protein was heterologously expressed and isolated as a fusion protein, f-HupS, with NusA, in E. coli BL21(DE3) as described previously [6]. Briefly: E. coli BL21(DE3) (Novagen), carrying the pET431HupS plasmid vector, were grown aerobically for about 20–24 h in autoinduction medium ZYP-5052 [22]. The cells were then collected by centrifugation, washed once in buffer W (100 mM Tris–HCl pH 7.5 containing 150 mM NaCl) and frozen at −20 °C until further use. The cells were broken by sonication after suspension in buffer W containing a glucose/glucose oxidase/catalase mixture and a protease inhibitor. A soluble fraction was obtained after addition of avidin, DNAse I and RNAse A and centrifugation of the resulting crude extract at 184,000 ×g. Protein purification was carried out by loading the soluble fraction on a Strep-Tactin column (IBA) in a glove box (MBraun) under an argon atmosphere. The Strep-tagged fusion protein, f-HupS, was eluted with three bed volumes of buffer W containing 5 mM desthiobiotin. f-HupS was aliquoted either directly into EPR tubes (150 μL each), capped with rubber septa, removed from the glove box and frozen (and kept frozen) in liquid nitrogen; or into 2 mL screw-cap tubes capped with rubber septa, removed from the glove box and frozen at −80 °C. Fractions containing the highest concentration of pure f-HupS were used in EPR experiments. All solutions used in the purification step were deoxygenated by purging for at least 15 min with N2 prior to use. 2.2. Sample preparation and flash photolysis EPR samples were prepared from f-HupS that had either been directly added to EPR tubes immediately after purification in the glove box, or from stock solutions kept anaerobic outside the glove box as detailed above. In the latter case, transfer to the EPR tubes was performed under a flow of argon gas. Ru(bpy)3Cl2 (20 mM) and sodium ascorbate (2 M) stock solutions were prepared in buffer W, purged for at least 15 min with N2 and kept in the dark until further use. Ru(bpy)2+ and 3 ascorbate were added anaerobically (under an argon gas flow) to each EPR tube, to a final concentration of 1.7 mM and 170 mM, respectively. All manipulations were at this stage performed under dim red light, chosen to avoid photoreactions by the sensitizer and electron donor. The final protein concentration in the EPR tubes was 13.6 μM. The samples were subjected to a train of laser flashes at room temperature and then immediately frozen at 77 K. Flashes were provided by a Nd:
YAG laser (Spectra Physics, USA) at 532 nm, 850 mJ/pulse and a frequency of 1 Hz, or in some cases 5 Hz. 2.3. EPR spectroscopy Samples were investigated by continuous wave X-band EPR directly as purified, after addition of Ru(bpy)23 + and sodium ascorbate in the dark, and after flash photolysis. EPR measurements were performed on a Bruker ELEXYS E500 spectrometer using an ER049X SuperX microwave bridge, in a Bruker SHQ0601 cavity equipped with an Oxford Instruments continuous flow cryostat. Measurement temperature was 7 K, using an ITC 503 temperature controller (Oxford Instruments) and liquid helium as coolant. Signal processing and quantification was performed using the Xepr software package (Bruker). 3. Results and discussion To achieve photoreduction of the isolated f-HupS protein, we used Ru(bpy)2+ as photosensitizer and sodium ascorbate as electron donor. 3 Scheme 1 illustrates the reactions involved in this system. When the photosensitizer is excited by a laser flash (Scheme 1, (1)) in the presence of an excess amount of ascorbate (ca. 100 times the concentration of Ru(bpy)23 +), the excited state Ru(bpy)23 + ⁎ is reductively quenched by ascorbate (Scheme 1, (2)), generating Ru(bpy)+ 3 [23,24]. The reduction potential of Ru(bpy)+ 3 in aqueous solution at neutral pH is about −1.3 V vs. the standard hydrogen electrode (SHE) [25], which is theoretically well below the potentials of the FeS clusters in HupS and should therefore be sufficient to reduce all three clusters. The Ru(bpy)+ 3 then transfers an electron to the f-HupS protein (Scheme 1, (3)) where one of the FeS clusters is reduced. There is sufficient electron donor in the reaction mix to allow for this reaction cycle to occur several times on continued flashing. FeS clusters are known to display changes in the UV-visible (UV– VIS) absorption spectrum upon reduction [26–28]. However, the presence of three FeS clusters with overlapping absorption spectra in the same protein makes the observation and assignment of reduction in the different clusters difficult. On the other hand, the spectral signatures of FeS clusters as observed by EPR spectroscopy are distinguishable from each other. In fact, we have previously observed the EPR signatures of all three FeS clusters in the isolated f-HupS [6]. We therefore utilized EPR spectroscopy to monitor the photoreduction of FeS clusters in f-HupS. Fig. 1A shows the EPR spectrum of the medial, [3Fe–4S] cluster before and after flash photolysis of f-HupS at room temperature. When f-HupS is isolated from the cell culture, the medial cluster is in the oxidized form in a variable portion (15–45%) of the proteins, indicated by the presence of the medial cluster EPR signature (Fig. 1A, top spectrum). When Ru(bpy)23 + and ascorbate were added to the protein in the dark, the EPR spectrum was virtually identical to that of the as-purified protein (not shown). This shows that, in the absence of light, the photosensitizer and in particular the electron donor were incapable of reducing the FeS clusters on their own, on the timescale of the experiment including freezing of the sample. We then exposed samples containing f-HupS, Ru(bpy)2+ 3 and ascorbate to a train of 150 laser flashes at 532 nm, which efficiently excites Ru(bpy)23 +. The EPR signal from the [3Fe–4S] cluster then decreased in amplitude (Fig. 1A, middle spectrum). Fig. 1B shows a plot of the decrease in EPR signal intensity in five individual samples given different numbers of flashes. It shows how the EPR signal from the medial cluster decreased successively with higher number of flashes. After ca. 20 flashes, the signal amplitude was reduced by 50%, and the signal was completely absent after 190 flashes. This suggests that the reduction yield per flash is in the 5% range. A difference spectrum, obtained by subtracting the top and middle spectra in Fig. 1A, shows the EPR signal which had disappeared with a total number of 150 flashes (Fig. 1A, bottom spectrum). This g = 2.023 signal is typical of the oxidized [3Fe–4S]
P. Raleiras et al. / Journal of Inorganic Biochemistry 148 (2015) 57–61
59
Scheme 1. Reaction mechanism for light-induced reduction of the fusion protein f-HupS, using a photosensitizer (Ru(bpy)2+ 3 ) and an electron donor (sodium ascorbate). See text for details.
medial cluster [6]. Its disappearance after flash photolsyis shows that the medial cluster was photoreduced. The procedure of flashing and freezing the samples at room temperature introduces idle time when the ascorbate in the solution might reduce the FeS clusters on its own, competing with the photochemical reduction. In a control sample we therefore let the ascorbate react with the protein at room temperature in the dark, in the presence of Ru(bpy)2+ 3 , for a total time of 2.5 min. This corresponds to the time it takes to thaw a sample, flash it at 5 Hz up to 190 flashes, and then freeze it. The EPR spectrum taken in the dark control after 2.5 min showed no discernible change in the EPR signal intensity from the [3Fe–4S] cluster (Fig. 1B, square). In contrast, in a sample that had been given about 190 flashes at 5 Hz, the [3Fe–4S] EPR signal had disappeared completely. It can thus be concluded that the observed reduction of the medial cluster is entirely due to photochemistry, and that the Ru(bpy)2+ or ascorbate 3 in the solution did not reduce f-HupS in the dark. The interesting question now arises whether it is also possible to photoreduce either one or both of the proximal and distal [4Fe–4S] clusters. One can anticipate that this is much more difficult, given the lower redox potentials of these. The proximal and distal clusters are considerably more reducing than the medial cluster, with reduction potentials of − 290 to − 360 mV [7,9–13]. Following the reduction of the medial cluster, we therefore analyzed the EPR spectra for signs of reduction of also the proximal and distal clusters. However, we did not observe any of the EPR signatures typical of the reduced [4Fe–4S] species, which we had observed previously after chemical reduction [6]. In our previous study of the isolated f-HupS protein, we could observe the reduction of all three FeS clusters by addition of 1–2 mM dithionite. At neutral pH the potential of a deoxygenated 1 mM dithionite solution is approximately −0.5 V [29]. It is thus close to the supposed potentials of the [4Fe–4S] clusters, but reducing enough for accomplishing complete reduction of both the proximal and distal cluster. That we obtained different results with the two reduction protocols was unexpected. However, it is conceivable that the method of flashing individual samples, as presented in Fig. 1B, leads to a slow or inefficient reaction, making the protein reduction incomplete. Therefore, in a separate experiment, we did serial flash-photolysis during several freezethaw cycles, where each sample was given several trains of flashes on different occasions. This successive photochemical reduction resulted in the same rate and degree of reduction of the medial [3Fe–4S] cluster EPR signal as when we flashed parallel samples individually (Fig. 1C, filled symbols). Also for this experimental series we kept a control sample in the dark for 15 min at room temperature, to simulate the total amount of
time a sample has spent at room temperature during several cycles of thawing and flashing up to 150 flashes. In the control sample, the EPR signal from the oxidized [3Fe–4S] cluster decreased by about 31% (Fig. 1C, filled square) after 15 min. However, the flash photolysis reduced the cluster to the same level considerably faster, with a comparable reduction yield already after a few flashes (Fig. 1C). Importantly, under these first 150 flashes we were again unable to observe any reduction of the [4Fe–4S] clusters (not shown). We therefore attempted to reduce the protein further, by continuing to flash the samples where all of the [3Fe–4S] clusters already had been photoreduced, since the reaction mixture contains enough electron donor to sustain photoreduction for many more cycles of flash photolysis. In spite of the application of up to 400 flashes, we did not observe any further reactions beyond the reduction of the medial [3Fe–4S] cluster. Thus, even with the serial flashing protocol, and prolonged flashing, we did not observe any EPR signals that could be attributed to the reduced [4Fe–4S] clusters. It is clear that it is possible to completely reduce the medial cluster by flash photolysis using the Ru/ascorbate system at pH 7.5. The lack of photoreduction of the [4Fe–4S] clusters might hypothetically be a result of instability of the protein during the experiment, leading to loss of metal and protein integrity. We have previously observed that the protein is unstable after exposure of the protein to oxygen and we developed our experimental protocol to minimize this problem [6]. During all steps of protein isolation and sample handling in the present study, we have therefore avoided air exposure, e.g. preparing samples in a glovebox. It is thus unlikely that oxygen can cause protein instability under our experimental conditions. Another possibility is that prolonged flash photolysis with Ru(bpy)2+ 3 could be detrimental to the protein. To investigate this possibility we made two control experiments. In the first one, we tested the reversibility of the photoinduced reduction. f-HupS samples that had been subjected to flash photolysis were first desalted by gel fitration to remove excess Na-ascorbate. Then a stoichiometric amount of K3Fe(CN)6 (ferricyanide) was added to oxidize all [3Fe–4S] centers. The EPR spectrum obtained after this showed that at least 80% of the medial clusters could be reoxidized by ferricyanide, similar to earlier results (not shown but cf. [6]). In our second control experiment, we investigated if [4Fe–4S] clusters, that have been pre-reduced, would undergo degradation by flash photolysis. Therefore we first reduced the f-HupS protein using sodium dithionite as in our previous study [6]. This chemically reduced sample displayed an EPR spectrum with typical features of the reduced FeS clusters, which we had observed previously (not shown, but cf. ref. [6]). We then added Ru(bpy)2+ and gave the sample 200 laser flashes 3
60
P. Raleiras et al. / Journal of Inorganic Biochemistry 148 (2015) 57–61
at room temperature. In the EPR spectrum which then resulted (Fig. S1), the amplitude of the signal from the reduced [4Fe–4S] clusters was slightly diminished, but otherwise the spectrum looked essentially the same. We then thawed the sample, and let it stand in the dark at room temperature for 30 min, and then froze it again. The EPR spectrum which was recorded after this was nearly identical to the spectrum recorded directly after the first reduction with dithionite (Fig. S1), clearly demonstrating the integrity of the low potential FeS clusters. Thus, we conclude that the protein survives flash photolysis in the presence of a ruthenium photosensitizer, and that the FeS clusters are intact after such a treatment. We can therefore exclude protein damage as explanation for the lack of EPR signals from reduced [4Fe–4S] clusters after flash photolysis. In the f-HupS construct which we use for making the HupS protein soluble in the absence of the large subunit (HupL), the NusA solubilization protein is fused to HupS close to the proximal [4Fe–4S] cluster. It is possible that the NusA protein thus covers the “entry point” for electrons to the proximal cluster, preventing Ru(bpy)+ 3 from donating electrons there. If this is the case, it is possible that the flash photolysis protocol could lead to a situation where both the medial and the distal FeS clusters are reduced, while the proximal cluster remains oxidized. In that case it is clear that the EPR signal from the oxidized medial cluster should disappear, which is what we observe (Fig. 1). In contrast, it is not clear what the EPR signature from the distal cluster would look like. We have earlier studied the fully reduced protein, i.e. where all three FeS clusters have been reduced by dithionite [6]. Then we could observe both EPR signatures from the proximal and distal [4Fe–4S] clusters. As we do not observe any similar [4Fe–4S] EPR signals after flash photolysis, it is possible that the distal [4Fe–4S] cluster in a partially reduced system is EPR silent [30], or is displaying EPR features that we have yet to discover. No cyanobacterial uptake hydrogenase has so far been structurally characterized. As previously discussed [6] there are important differences in the amino acid sequences which distinguish the cyanobacterial uptake hydrogenases from the more well-studied classical NiFe hydrogenases in e.g. the Desulfovibrio family [6,31,32]. It is thus possible that the cluster composition might be different from previously known hydrogenases. The clusters might differ in structure or be differently located. If for example the [3Fe–4S] cluster, which is reduced by flash photolysis, would be located closer to the protein surface than expected, the other two clusters (the [4Fe–4S] clusters) might then be relatively more difficult to reduce with our flash protocol. In summary, it is clearly possible to completely reduce the medial FeS cluster by flash photolysis using Ru(bpy)2+ and sodium ascorbate 3 at pH 7.5. This is a useful result, which opens the possibility to study electron transfer and partial reduction of the FeS clusters in the purified f-HupS protein. These studies are helpful in advancing the knowledge of the HupS subunit from the biotechnologically interesting organism N. punctiforme, where the uptake hydrogenase HupSL is present in the anaerobic environment of the heterocysts. Fig. 1. A, X-band EPR spectra of f-HupS. Top spectrum: The EPR signal from the oxidized medial [3Fe–4S] cluster in f-HupS as purified; middle spectrum: after giving the sample 150 laser flashes at room temperature; bottom spectrum: difference spectrum between the top and middle spectra. The remaining EPR spectrum after 150 flashes is different from the EPR signal from the medial cluster and does not reflect the FeS centers. B, Flashing of individual samples: Circles show the normalized EPR signal intensity from the oxidized medial cluster as a function of the number of flashes given to five individual samples. Square: the signal intensity in a control sample which has been standing in the dark at room temperature, for a period of 2.5 min (equal to the time the sample flashed 190 times had been at room temperature during flashing). C, Serial flashing: Filled symbols show the signal intensity of the oxidized medial cluster from three separate samples, each having been flashed multiple times during successive freeze–thaw cycles. The filled square shows the signal intensity in a control sample that was let standing in the dark at room temperature during 15 min, which is the approximate time a sample has spent at room temperature in total, after about 150 flashes including several freeze–thaw-flash cycles. The open symbols show the same data as shown in Fig. 1B for comparison. All experiments were performed in the presence of Ru(bpy)2+ and sodium ascorbate. EPR 3 conditions in all experiments: 2 mW applied microwave power, T = 7 K.
4. Conclusions We have shown that it is possible to inject electrons into the isolated small subunit, f-HupS using flash photolysis. The protein remains intact in spite of intense flashing in the presence of a photosensitizer. This opens up the possibility to continue flash photolysis studies in this protein and eventually in its mutants. Presently we have achieved controlled reduction of the medial, high potential FeS cluster, while we could not photoreduce the proximal and distal [4Fe–4S] clusters, judging from the absence of EPR signatures from these clusters. However, one possibility remains that the distal cluster might be invisible by EPR under these circumstances. We have nevertheless a straightforward method to semi-reduce f-HupS to a redox state that is otherwise hard to achieve chemically, opening possibilities for future studies of electron
P. Raleiras et al. / Journal of Inorganic Biochemistry 148 (2015) 57–61
transfer and magnetic interaction between clusters in this particular redox state. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jinorgbio.2015.03.018. Abbreviations Bpy f-HupS
bipyridine recombinant NusA-HupS fusion protein
Acknowledgments The Swedish Energy Agency (11674-5) and the Knut and Alice Wallenberg Foundation (2011.0067) are gratefully acknowledged for their financial support. References [1] W. Lubitz, H. Ogata, O. Rüdiger, E. Reijerse, Chem. Rev. 114 (2014) 4081–4148. [2] H. Ogata, Y. Mizoguchi, N. Mizuno, K. Miki, S.-I. Adachi, N. Yasuoka, T. Yagi, O. Yamauchi, S. Hirota, Y. Higuchi, J. Am. Chem. Soc. 124 (2002) 11628–11635. [3] J. Fritsch, P. Scheerer, S. Frielingsdorf, S. Kroschinsky, B. Friedrich, O. Lenz, C.M.T. Spahn, Nature 479 (2011) 249–252. [4] P. Tamagnini, E. Leitão, P. Oliveira, D. Ferreira, F. Pinto, D.J. Harris, T. Heidorn, P. Lindblad, FEMS Microbiol. Rev. 31 (2007) 692–720. [5] F. Germer, I. Zebger, M. Saggu, F. Lendzian, R.D. Schulz, J. Appel, J. Biol. Chem. 284 (2009) 36462–36472. [6] P. Raleiras, P. Kellers, P. Lindblad, S. Styring, A. Magnuson, J. Biol. Chem. 288 (2013) 18345–18352. [7] M. Rousset, Y. Montet, B. Guigliarelli, N. Forget, M. Asso, P. Bertrand, J.C. FontecillaCamps, E.C. Hatchikian, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 11625–11630. [8] C.C. Page, C.C. Moser, X.C. Chen, P.L. Dutton, Nature 402 (1999) 47–52. [9] R. Cammack, D. Patil, R. Aguirre, E.C. Hatchikian, FEBS Lett. 142 (1982) 289–292. [10] R. Cammack, K.K. Rao, J. Serra, M.J. Llama, Biochimie 68 (1986) 93–96.
61
[11] M. Teixeira, I. Moura, A.V. Xavier, J.J.G. Moura, J. Legall, D.V. Dervartanian, H.D. Peck, B.H. Huynh, J. Biol. Chem. 264 (1989) 16435–16450. [12] P.J. Silva, E.C.D. van den Ban, H. Wassink, H. Haaker, B. de Castro, F.T. Robb, W.R. Hagen, Eur. J. Biochem. 267 (2000) 6541–6551. [13] C.M. Cordas, I. Moura, J.J.G. Moura, Bioelectrochemistry 74 (2008) 83–89. [14] J.C. Fontecilla-Camps, A. Volbeda, C. Cavazza, Y. Nicolet, Chem. Rev. 107 (2007) 4273–4303. [15] K.A. Vincent, A. Parkin, F.A. Armstrong, Chem. Rev. 107 (2007) 4366–4413. [16] H.B. Gray, J.R. Winkler, Annu. Rev. Biochem. 65 (1996) 537–561. [17] F. Millett, B. Durham, Biochemistry 38 (2002) 11315–11324. [18] H.B. Gray, J.R. Winkler, Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 3534–3539. [19] B.L. Greene, C.A. Joseph, M.J. Maroney, R.B. Dyer, J. Am. Chem. Soc. 134 (2012) 11108–11111. [20] D. Streich, Stepping into Catalysis: Kinetic and Mechanistic Investigations of Photoand Electrocatalytic Hydrogen Production with Natural and Synthetic Molecular Catalysts(Dissertation) Acta Universitatis Upsaliensis, Uppsala, 2013. [21] T. Noji, M. Kondo, T. Jin, T. Yazawa, H. Osuka, Y. Higuchi, M. Nango, S. Itoh, T. Dewa, J. Phys. Chem. Lett. 5 (2014) 2402–2407. [22] F.W. Studier, Protein Expr. Purif. 41 (2005) 207–234. [23] S. Fukuzumi, T. Kobayashi, T. Suenobu, Angew. Chem. 50 (2011) 728–731. [24] B. Shan, T. Baine, X.A.N. Ma, X. Zhao, R.H. Schmehl, Inorg. Chem. 52 (2013) 4853–4859. [25] A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser, A. von Zelewsky, Coord. Chem. Rev. 84 (1988) 85–277. [26] W.V. Sweeney, J.C. Rabinowitz, Annu. Rev. Biochem. 49 (1980) 139–161. [27] M.L. Antonkine, M.S. Koay, B. Epel, C. Breitenstein, O. Gopta, W. Gärtner, E. Bill, W. Lubitz, Biochim. Biophys. Acta Bioenerg. 1787 (2009) 995–1008. [28] M.M.-J. Couture, V.J.J. Martin, W.W. Mohn, L.D. Eltis, Biochim. Biophys. Acta Protein Proteomics 1764 (2006) 1462–1469. [29] S.G. Mayhew, Eur. J. Biochem. 85 (1978) 535–547. [30] M.M. Roessler, R.M. Evans, R.A. Davies, J. Harmer, F.A. Armstrong, J. Am. Chem. Soc. 134 (2012) 15581–15594. [31] J. Fischer, A. Quentmeier, S. Kostka, R. Kraft, C.G. Friedrich, Arch. Microbiol. 165 (1996) 289–296. [32] O. Schröder, B. Bleijlevens, T.E. de Jongh, Z. Chen, T. Li, J. Fischer, J. Forster, C.G. Friedrich, K.A. Bagley, S.P. Albracht, W. Lubitz, J. Biol. Inorg. Chem. 12 (2007) 212–233.
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry journal homepage: www.elsevier.com/locate/jinorgbio
Photoinduced reduction of the medial FeS center in the hydrogenase small subunit HupS from Nostoc punctiforme Patrícia Raleiras, Leif Hammarström, Peter Lindblad, Stenbjörn Styring ⁎, Ann Magnuson ⁎⁎ Department of Chemistry, Ångström Laboratory, Uppsala University, Box 523, SE-75120 Uppsala, Sweden
a r t i c l e
i n f o
Article history: Received 19 December 2014 Received in revised form 22 March 2015 Accepted 24 March 2015 Available online 14 April 2015 Keywords: Hydrogenase Iron–sulfur Electron transfer Flash photolysis Electron paramagnetic resonance
a b s t r a c t The small subunit from the NiFe uptake hydrogenase, HupSL, in the cyanobacterium Nostoc punctiforme ATCC 29133, has been isolated in the absence of the large subunit (P. Raleiras, P. Kellers, P. Lindblad, S. Styring, A. Magnuson, J. Biol. Chem. 288 (2013) 18,345–18,352). Here, we have used flash photolysis to reduce the iron-sulfur clusters in the isolated small subunit, HupS. We used ascorbate as electron donor to the photogenerated excited state of Ru(II)-trisbipyridine (Ru(bpy)3), to generate Ru(I)(bpy)3 as reducing agent. Our results show that the isolated small subunit can be reduced by the Ru(I)(bpy)3 generated through flash photolysis. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Hydrogenases are metalloenzymes that catalyze the reversible oxidation of H2: þ
−
H2 ⇌2H þ 2e :
Hydrogenases belong to either one of three categories: the [NiFe] hydrogenases, containing an inorganic nickel–iron complex in the catalytic center, the [FeFe] hydrogenases, containing a binuclear iron complex, and the [Fe] hydrogenases, containing a mononuclear iron center [1]. The [NiFe] hydrogenases consist of two protein subunits, referred to as the large and small subunit respectively [2,3]. The large subunit contains the active site where H2 oxidation or production is catalyzed by the [NiFe] complex. The typical electron transfer motif in the small subunit is three iron– sulfur (FeS) clusters: a proximal [4Fe–4S] cluster closest to the active site, a medial [3Fe–4S] and a distal [4Fe–4S] cluster. The three FeS clusters are aligned so that they connect the protein surface with the active site in the large subunit, and it is generally believed that electron transfer in the small subunit goes via this array of FeS clusters. Depending on whether the hydrogenase catalyzes proton reduction or hydrogen oxidation, the electrons are transported to, or away from the active site. ⁎ Corresponding author. ⁎⁎ Corresponding author. Fax: +46 18 471 6844. E-mail addresses: [email protected] (S. Styring), [email protected] (A. Magnuson).
http://dx.doi.org/10.1016/j.jinorgbio.2015.03.018 0162-0134/© 2015 Elsevier Inc. All rights reserved.
Cyanobacteria are phototrophic microorganisms that can produce H2 from solar energy and water. They are therefore attractive targets for efforts to improve their productivity via genetic engineering. All cyanobacteria with known gene sequence possess at least one hydrogenase copy. In cases where more than one copy exists within a genus, they usually differ in metabolic purpose and activity [4]. In nitrogen-fixing strains, an uptake hydrogenase is usually present which recycles H2 that is evolved in the nitrogenase reaction. Other, bidirectional, hydrogenases are known to perform both hydrogen evolution and uptake under different metabolic conditions. Only one cyanobacterial hydrogenase, the bidirectional hydrogenase from Synechocystis PCC 6803, has so far been isolated and characterized [5]. The filamentous, heterocystous cyanobacterium Nostoc punctiforme ATCC 29133 has only one hydrogenase, the [NiFe] uptake hydrogenase HupSL. We have previously isolated the small subunit, HupS, in the absence of the large subunit by expressing it in Escherichia coli as a fusion protein, from now on referred to as f–HupS [6]. This enabled us to characterize the FeS centers in the small subunit without spectroscopic or magnetic interference from the active site. Since HupSL is a H2-oxidizing enzyme, the electron transfer in HupS is expected to be directed away from the active site. Due to the linear arrangement of the FeS clusters in HupS, the electron transfer route from the active site is believed to be first to the proximal cluster, then to the medial and distal clusters, and finally to the native redox partner. The relative reduction potentials of the three FeS clusters have been suggested to play an important role in steering the electron transfer directionality in known NiFe hydrogenases [7,8]. The reduction potentials of the proximal and distal [4Fe–4S] clusters have been determined to −290 to −360 mV [7,9–13]. The medial [3Fe–4S] cluster on the other
58
P. Raleiras et al. / Journal of Inorganic Biochemistry 148 (2015) 57–61
hand, is more oxidizing with a potential of −70–+100 mV in known cases. It might thus act as an electron trap in the electron transfer chain [14,15]. It remains unclear whether electron transfer proceeds via the medial cluster, or if the electron tunnels the ca. 17 Å distance from the proximal to the distal cluster avoiding the medial cluster entirely. All previous investigations of electron transfer reactions in NiFe hydrogenases have been performed in the intact enzyme including the large subunit. The spectroscopic and redox properties of the small subunit have therefore been affected by the presence of the nearby active site. Our aim with this study is to investigate if the FeS clusters in the isolated f-HupS are accessible for photo-induced reduction from the surface of the protein. Flash photolysis using ruthenium-based photosensitizers, has been used for investigating electron transfer pathways in a number of metalloproteins [16–18] and recently to initiate turnover in hydrogenases [19–21]. We used ascorbate as electron donor to the photogenerated excited state of Ru(II)-trisbipyridine (Ru(bpy)3), to generate Ru(I)(bpy)3 as reducing agent, and monitored photoreduction of f-HupS using electron paramagnetic resonance (EPR) spectroscopy. Our results show that the isolated f-HupS can be reduced by flash photolysis. 2. Materials and methods 2.1. Protein expression and purification The HupS protein was heterologously expressed and isolated as a fusion protein, f-HupS, with NusA, in E. coli BL21(DE3) as described previously [6]. Briefly: E. coli BL21(DE3) (Novagen), carrying the pET431HupS plasmid vector, were grown aerobically for about 20–24 h in autoinduction medium ZYP-5052 [22]. The cells were then collected by centrifugation, washed once in buffer W (100 mM Tris–HCl pH 7.5 containing 150 mM NaCl) and frozen at −20 °C until further use. The cells were broken by sonication after suspension in buffer W containing a glucose/glucose oxidase/catalase mixture and a protease inhibitor. A soluble fraction was obtained after addition of avidin, DNAse I and RNAse A and centrifugation of the resulting crude extract at 184,000 ×g. Protein purification was carried out by loading the soluble fraction on a Strep-Tactin column (IBA) in a glove box (MBraun) under an argon atmosphere. The Strep-tagged fusion protein, f-HupS, was eluted with three bed volumes of buffer W containing 5 mM desthiobiotin. f-HupS was aliquoted either directly into EPR tubes (150 μL each), capped with rubber septa, removed from the glove box and frozen (and kept frozen) in liquid nitrogen; or into 2 mL screw-cap tubes capped with rubber septa, removed from the glove box and frozen at −80 °C. Fractions containing the highest concentration of pure f-HupS were used in EPR experiments. All solutions used in the purification step were deoxygenated by purging for at least 15 min with N2 prior to use. 2.2. Sample preparation and flash photolysis EPR samples were prepared from f-HupS that had either been directly added to EPR tubes immediately after purification in the glove box, or from stock solutions kept anaerobic outside the glove box as detailed above. In the latter case, transfer to the EPR tubes was performed under a flow of argon gas. Ru(bpy)3Cl2 (20 mM) and sodium ascorbate (2 M) stock solutions were prepared in buffer W, purged for at least 15 min with N2 and kept in the dark until further use. Ru(bpy)2+ and 3 ascorbate were added anaerobically (under an argon gas flow) to each EPR tube, to a final concentration of 1.7 mM and 170 mM, respectively. All manipulations were at this stage performed under dim red light, chosen to avoid photoreactions by the sensitizer and electron donor. The final protein concentration in the EPR tubes was 13.6 μM. The samples were subjected to a train of laser flashes at room temperature and then immediately frozen at 77 K. Flashes were provided by a Nd:
YAG laser (Spectra Physics, USA) at 532 nm, 850 mJ/pulse and a frequency of 1 Hz, or in some cases 5 Hz. 2.3. EPR spectroscopy Samples were investigated by continuous wave X-band EPR directly as purified, after addition of Ru(bpy)23 + and sodium ascorbate in the dark, and after flash photolysis. EPR measurements were performed on a Bruker ELEXYS E500 spectrometer using an ER049X SuperX microwave bridge, in a Bruker SHQ0601 cavity equipped with an Oxford Instruments continuous flow cryostat. Measurement temperature was 7 K, using an ITC 503 temperature controller (Oxford Instruments) and liquid helium as coolant. Signal processing and quantification was performed using the Xepr software package (Bruker). 3. Results and discussion To achieve photoreduction of the isolated f-HupS protein, we used Ru(bpy)2+ as photosensitizer and sodium ascorbate as electron donor. 3 Scheme 1 illustrates the reactions involved in this system. When the photosensitizer is excited by a laser flash (Scheme 1, (1)) in the presence of an excess amount of ascorbate (ca. 100 times the concentration of Ru(bpy)23 +), the excited state Ru(bpy)23 + ⁎ is reductively quenched by ascorbate (Scheme 1, (2)), generating Ru(bpy)+ 3 [23,24]. The reduction potential of Ru(bpy)+ 3 in aqueous solution at neutral pH is about −1.3 V vs. the standard hydrogen electrode (SHE) [25], which is theoretically well below the potentials of the FeS clusters in HupS and should therefore be sufficient to reduce all three clusters. The Ru(bpy)+ 3 then transfers an electron to the f-HupS protein (Scheme 1, (3)) where one of the FeS clusters is reduced. There is sufficient electron donor in the reaction mix to allow for this reaction cycle to occur several times on continued flashing. FeS clusters are known to display changes in the UV-visible (UV– VIS) absorption spectrum upon reduction [26–28]. However, the presence of three FeS clusters with overlapping absorption spectra in the same protein makes the observation and assignment of reduction in the different clusters difficult. On the other hand, the spectral signatures of FeS clusters as observed by EPR spectroscopy are distinguishable from each other. In fact, we have previously observed the EPR signatures of all three FeS clusters in the isolated f-HupS [6]. We therefore utilized EPR spectroscopy to monitor the photoreduction of FeS clusters in f-HupS. Fig. 1A shows the EPR spectrum of the medial, [3Fe–4S] cluster before and after flash photolysis of f-HupS at room temperature. When f-HupS is isolated from the cell culture, the medial cluster is in the oxidized form in a variable portion (15–45%) of the proteins, indicated by the presence of the medial cluster EPR signature (Fig. 1A, top spectrum). When Ru(bpy)23 + and ascorbate were added to the protein in the dark, the EPR spectrum was virtually identical to that of the as-purified protein (not shown). This shows that, in the absence of light, the photosensitizer and in particular the electron donor were incapable of reducing the FeS clusters on their own, on the timescale of the experiment including freezing of the sample. We then exposed samples containing f-HupS, Ru(bpy)2+ 3 and ascorbate to a train of 150 laser flashes at 532 nm, which efficiently excites Ru(bpy)23 +. The EPR signal from the [3Fe–4S] cluster then decreased in amplitude (Fig. 1A, middle spectrum). Fig. 1B shows a plot of the decrease in EPR signal intensity in five individual samples given different numbers of flashes. It shows how the EPR signal from the medial cluster decreased successively with higher number of flashes. After ca. 20 flashes, the signal amplitude was reduced by 50%, and the signal was completely absent after 190 flashes. This suggests that the reduction yield per flash is in the 5% range. A difference spectrum, obtained by subtracting the top and middle spectra in Fig. 1A, shows the EPR signal which had disappeared with a total number of 150 flashes (Fig. 1A, bottom spectrum). This g = 2.023 signal is typical of the oxidized [3Fe–4S]
P. Raleiras et al. / Journal of Inorganic Biochemistry 148 (2015) 57–61
59
Scheme 1. Reaction mechanism for light-induced reduction of the fusion protein f-HupS, using a photosensitizer (Ru(bpy)2+ 3 ) and an electron donor (sodium ascorbate). See text for details.
medial cluster [6]. Its disappearance after flash photolsyis shows that the medial cluster was photoreduced. The procedure of flashing and freezing the samples at room temperature introduces idle time when the ascorbate in the solution might reduce the FeS clusters on its own, competing with the photochemical reduction. In a control sample we therefore let the ascorbate react with the protein at room temperature in the dark, in the presence of Ru(bpy)2+ 3 , for a total time of 2.5 min. This corresponds to the time it takes to thaw a sample, flash it at 5 Hz up to 190 flashes, and then freeze it. The EPR spectrum taken in the dark control after 2.5 min showed no discernible change in the EPR signal intensity from the [3Fe–4S] cluster (Fig. 1B, square). In contrast, in a sample that had been given about 190 flashes at 5 Hz, the [3Fe–4S] EPR signal had disappeared completely. It can thus be concluded that the observed reduction of the medial cluster is entirely due to photochemistry, and that the Ru(bpy)2+ or ascorbate 3 in the solution did not reduce f-HupS in the dark. The interesting question now arises whether it is also possible to photoreduce either one or both of the proximal and distal [4Fe–4S] clusters. One can anticipate that this is much more difficult, given the lower redox potentials of these. The proximal and distal clusters are considerably more reducing than the medial cluster, with reduction potentials of − 290 to − 360 mV [7,9–13]. Following the reduction of the medial cluster, we therefore analyzed the EPR spectra for signs of reduction of also the proximal and distal clusters. However, we did not observe any of the EPR signatures typical of the reduced [4Fe–4S] species, which we had observed previously after chemical reduction [6]. In our previous study of the isolated f-HupS protein, we could observe the reduction of all three FeS clusters by addition of 1–2 mM dithionite. At neutral pH the potential of a deoxygenated 1 mM dithionite solution is approximately −0.5 V [29]. It is thus close to the supposed potentials of the [4Fe–4S] clusters, but reducing enough for accomplishing complete reduction of both the proximal and distal cluster. That we obtained different results with the two reduction protocols was unexpected. However, it is conceivable that the method of flashing individual samples, as presented in Fig. 1B, leads to a slow or inefficient reaction, making the protein reduction incomplete. Therefore, in a separate experiment, we did serial flash-photolysis during several freezethaw cycles, where each sample was given several trains of flashes on different occasions. This successive photochemical reduction resulted in the same rate and degree of reduction of the medial [3Fe–4S] cluster EPR signal as when we flashed parallel samples individually (Fig. 1C, filled symbols). Also for this experimental series we kept a control sample in the dark for 15 min at room temperature, to simulate the total amount of
time a sample has spent at room temperature during several cycles of thawing and flashing up to 150 flashes. In the control sample, the EPR signal from the oxidized [3Fe–4S] cluster decreased by about 31% (Fig. 1C, filled square) after 15 min. However, the flash photolysis reduced the cluster to the same level considerably faster, with a comparable reduction yield already after a few flashes (Fig. 1C). Importantly, under these first 150 flashes we were again unable to observe any reduction of the [4Fe–4S] clusters (not shown). We therefore attempted to reduce the protein further, by continuing to flash the samples where all of the [3Fe–4S] clusters already had been photoreduced, since the reaction mixture contains enough electron donor to sustain photoreduction for many more cycles of flash photolysis. In spite of the application of up to 400 flashes, we did not observe any further reactions beyond the reduction of the medial [3Fe–4S] cluster. Thus, even with the serial flashing protocol, and prolonged flashing, we did not observe any EPR signals that could be attributed to the reduced [4Fe–4S] clusters. It is clear that it is possible to completely reduce the medial cluster by flash photolysis using the Ru/ascorbate system at pH 7.5. The lack of photoreduction of the [4Fe–4S] clusters might hypothetically be a result of instability of the protein during the experiment, leading to loss of metal and protein integrity. We have previously observed that the protein is unstable after exposure of the protein to oxygen and we developed our experimental protocol to minimize this problem [6]. During all steps of protein isolation and sample handling in the present study, we have therefore avoided air exposure, e.g. preparing samples in a glovebox. It is thus unlikely that oxygen can cause protein instability under our experimental conditions. Another possibility is that prolonged flash photolysis with Ru(bpy)2+ 3 could be detrimental to the protein. To investigate this possibility we made two control experiments. In the first one, we tested the reversibility of the photoinduced reduction. f-HupS samples that had been subjected to flash photolysis were first desalted by gel fitration to remove excess Na-ascorbate. Then a stoichiometric amount of K3Fe(CN)6 (ferricyanide) was added to oxidize all [3Fe–4S] centers. The EPR spectrum obtained after this showed that at least 80% of the medial clusters could be reoxidized by ferricyanide, similar to earlier results (not shown but cf. [6]). In our second control experiment, we investigated if [4Fe–4S] clusters, that have been pre-reduced, would undergo degradation by flash photolysis. Therefore we first reduced the f-HupS protein using sodium dithionite as in our previous study [6]. This chemically reduced sample displayed an EPR spectrum with typical features of the reduced FeS clusters, which we had observed previously (not shown, but cf. ref. [6]). We then added Ru(bpy)2+ and gave the sample 200 laser flashes 3
60
P. Raleiras et al. / Journal of Inorganic Biochemistry 148 (2015) 57–61
at room temperature. In the EPR spectrum which then resulted (Fig. S1), the amplitude of the signal from the reduced [4Fe–4S] clusters was slightly diminished, but otherwise the spectrum looked essentially the same. We then thawed the sample, and let it stand in the dark at room temperature for 30 min, and then froze it again. The EPR spectrum which was recorded after this was nearly identical to the spectrum recorded directly after the first reduction with dithionite (Fig. S1), clearly demonstrating the integrity of the low potential FeS clusters. Thus, we conclude that the protein survives flash photolysis in the presence of a ruthenium photosensitizer, and that the FeS clusters are intact after such a treatment. We can therefore exclude protein damage as explanation for the lack of EPR signals from reduced [4Fe–4S] clusters after flash photolysis. In the f-HupS construct which we use for making the HupS protein soluble in the absence of the large subunit (HupL), the NusA solubilization protein is fused to HupS close to the proximal [4Fe–4S] cluster. It is possible that the NusA protein thus covers the “entry point” for electrons to the proximal cluster, preventing Ru(bpy)+ 3 from donating electrons there. If this is the case, it is possible that the flash photolysis protocol could lead to a situation where both the medial and the distal FeS clusters are reduced, while the proximal cluster remains oxidized. In that case it is clear that the EPR signal from the oxidized medial cluster should disappear, which is what we observe (Fig. 1). In contrast, it is not clear what the EPR signature from the distal cluster would look like. We have earlier studied the fully reduced protein, i.e. where all three FeS clusters have been reduced by dithionite [6]. Then we could observe both EPR signatures from the proximal and distal [4Fe–4S] clusters. As we do not observe any similar [4Fe–4S] EPR signals after flash photolysis, it is possible that the distal [4Fe–4S] cluster in a partially reduced system is EPR silent [30], or is displaying EPR features that we have yet to discover. No cyanobacterial uptake hydrogenase has so far been structurally characterized. As previously discussed [6] there are important differences in the amino acid sequences which distinguish the cyanobacterial uptake hydrogenases from the more well-studied classical NiFe hydrogenases in e.g. the Desulfovibrio family [6,31,32]. It is thus possible that the cluster composition might be different from previously known hydrogenases. The clusters might differ in structure or be differently located. If for example the [3Fe–4S] cluster, which is reduced by flash photolysis, would be located closer to the protein surface than expected, the other two clusters (the [4Fe–4S] clusters) might then be relatively more difficult to reduce with our flash protocol. In summary, it is clearly possible to completely reduce the medial FeS cluster by flash photolysis using Ru(bpy)2+ and sodium ascorbate 3 at pH 7.5. This is a useful result, which opens the possibility to study electron transfer and partial reduction of the FeS clusters in the purified f-HupS protein. These studies are helpful in advancing the knowledge of the HupS subunit from the biotechnologically interesting organism N. punctiforme, where the uptake hydrogenase HupSL is present in the anaerobic environment of the heterocysts. Fig. 1. A, X-band EPR spectra of f-HupS. Top spectrum: The EPR signal from the oxidized medial [3Fe–4S] cluster in f-HupS as purified; middle spectrum: after giving the sample 150 laser flashes at room temperature; bottom spectrum: difference spectrum between the top and middle spectra. The remaining EPR spectrum after 150 flashes is different from the EPR signal from the medial cluster and does not reflect the FeS centers. B, Flashing of individual samples: Circles show the normalized EPR signal intensity from the oxidized medial cluster as a function of the number of flashes given to five individual samples. Square: the signal intensity in a control sample which has been standing in the dark at room temperature, for a period of 2.5 min (equal to the time the sample flashed 190 times had been at room temperature during flashing). C, Serial flashing: Filled symbols show the signal intensity of the oxidized medial cluster from three separate samples, each having been flashed multiple times during successive freeze–thaw cycles. The filled square shows the signal intensity in a control sample that was let standing in the dark at room temperature during 15 min, which is the approximate time a sample has spent at room temperature in total, after about 150 flashes including several freeze–thaw-flash cycles. The open symbols show the same data as shown in Fig. 1B for comparison. All experiments were performed in the presence of Ru(bpy)2+ and sodium ascorbate. EPR 3 conditions in all experiments: 2 mW applied microwave power, T = 7 K.
4. Conclusions We have shown that it is possible to inject electrons into the isolated small subunit, f-HupS using flash photolysis. The protein remains intact in spite of intense flashing in the presence of a photosensitizer. This opens up the possibility to continue flash photolysis studies in this protein and eventually in its mutants. Presently we have achieved controlled reduction of the medial, high potential FeS cluster, while we could not photoreduce the proximal and distal [4Fe–4S] clusters, judging from the absence of EPR signatures from these clusters. However, one possibility remains that the distal cluster might be invisible by EPR under these circumstances. We have nevertheless a straightforward method to semi-reduce f-HupS to a redox state that is otherwise hard to achieve chemically, opening possibilities for future studies of electron
P. Raleiras et al. / Journal of Inorganic Biochemistry 148 (2015) 57–61
transfer and magnetic interaction between clusters in this particular redox state. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jinorgbio.2015.03.018. Abbreviations Bpy f-HupS
bipyridine recombinant NusA-HupS fusion protein
Acknowledgments The Swedish Energy Agency (11674-5) and the Knut and Alice Wallenberg Foundation (2011.0067) are gratefully acknowledged for their financial support. References [1] W. Lubitz, H. Ogata, O. Rüdiger, E. Reijerse, Chem. Rev. 114 (2014) 4081–4148. [2] H. Ogata, Y. Mizoguchi, N. Mizuno, K. Miki, S.-I. Adachi, N. Yasuoka, T. Yagi, O. Yamauchi, S. Hirota, Y. Higuchi, J. Am. Chem. Soc. 124 (2002) 11628–11635. [3] J. Fritsch, P. Scheerer, S. Frielingsdorf, S. Kroschinsky, B. Friedrich, O. Lenz, C.M.T. Spahn, Nature 479 (2011) 249–252. [4] P. Tamagnini, E. Leitão, P. Oliveira, D. Ferreira, F. Pinto, D.J. Harris, T. Heidorn, P. Lindblad, FEMS Microbiol. Rev. 31 (2007) 692–720. [5] F. Germer, I. Zebger, M. Saggu, F. Lendzian, R.D. Schulz, J. Appel, J. Biol. Chem. 284 (2009) 36462–36472. [6] P. Raleiras, P. Kellers, P. Lindblad, S. Styring, A. Magnuson, J. Biol. Chem. 288 (2013) 18345–18352. [7] M. Rousset, Y. Montet, B. Guigliarelli, N. Forget, M. Asso, P. Bertrand, J.C. FontecillaCamps, E.C. Hatchikian, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 11625–11630. [8] C.C. Page, C.C. Moser, X.C. Chen, P.L. Dutton, Nature 402 (1999) 47–52. [9] R. Cammack, D. Patil, R. Aguirre, E.C. Hatchikian, FEBS Lett. 142 (1982) 289–292. [10] R. Cammack, K.K. Rao, J. Serra, M.J. Llama, Biochimie 68 (1986) 93–96.
61
[11] M. Teixeira, I. Moura, A.V. Xavier, J.J.G. Moura, J. Legall, D.V. Dervartanian, H.D. Peck, B.H. Huynh, J. Biol. Chem. 264 (1989) 16435–16450. [12] P.J. Silva, E.C.D. van den Ban, H. Wassink, H. Haaker, B. de Castro, F.T. Robb, W.R. Hagen, Eur. J. Biochem. 267 (2000) 6541–6551. [13] C.M. Cordas, I. Moura, J.J.G. Moura, Bioelectrochemistry 74 (2008) 83–89. [14] J.C. Fontecilla-Camps, A. Volbeda, C. Cavazza, Y. Nicolet, Chem. Rev. 107 (2007) 4273–4303. [15] K.A. Vincent, A. Parkin, F.A. Armstrong, Chem. Rev. 107 (2007) 4366–4413. [16] H.B. Gray, J.R. Winkler, Annu. Rev. Biochem. 65 (1996) 537–561. [17] F. Millett, B. Durham, Biochemistry 38 (2002) 11315–11324. [18] H.B. Gray, J.R. Winkler, Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 3534–3539. [19] B.L. Greene, C.A. Joseph, M.J. Maroney, R.B. Dyer, J. Am. Chem. Soc. 134 (2012) 11108–11111. [20] D. Streich, Stepping into Catalysis: Kinetic and Mechanistic Investigations of Photoand Electrocatalytic Hydrogen Production with Natural and Synthetic Molecular Catalysts(Dissertation) Acta Universitatis Upsaliensis, Uppsala, 2013. [21] T. Noji, M. Kondo, T. Jin, T. Yazawa, H. Osuka, Y. Higuchi, M. Nango, S. Itoh, T. Dewa, J. Phys. Chem. Lett. 5 (2014) 2402–2407. [22] F.W. Studier, Protein Expr. Purif. 41 (2005) 207–234. [23] S. Fukuzumi, T. Kobayashi, T. Suenobu, Angew. Chem. 50 (2011) 728–731. [24] B. Shan, T. Baine, X.A.N. Ma, X. Zhao, R.H. Schmehl, Inorg. Chem. 52 (2013) 4853–4859. [25] A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser, A. von Zelewsky, Coord. Chem. Rev. 84 (1988) 85–277. [26] W.V. Sweeney, J.C. Rabinowitz, Annu. Rev. Biochem. 49 (1980) 139–161. [27] M.L. Antonkine, M.S. Koay, B. Epel, C. Breitenstein, O. Gopta, W. Gärtner, E. Bill, W. Lubitz, Biochim. Biophys. Acta Bioenerg. 1787 (2009) 995–1008. [28] M.M.-J. Couture, V.J.J. Martin, W.W. Mohn, L.D. Eltis, Biochim. Biophys. Acta Protein Proteomics 1764 (2006) 1462–1469. [29] S.G. Mayhew, Eur. J. Biochem. 85 (1978) 535–547. [30] M.M. Roessler, R.M. Evans, R.A. Davies, J. Harmer, F.A. Armstrong, J. Am. Chem. Soc. 134 (2012) 15581–15594. [31] J. Fischer, A. Quentmeier, S. Kostka, R. Kraft, C.G. Friedrich, Arch. Microbiol. 165 (1996) 289–296. [32] O. Schröder, B. Bleijlevens, T.E. de Jongh, Z. Chen, T. Li, J. Fischer, J. Forster, C.G. Friedrich, K.A. Bagley, S.P. Albracht, W. Lubitz, J. Biol. Inorg. Chem. 12 (2007) 212–233.
