O2-Tolerant H2 Activation by an Isolated Large Subunit of a [NiFe

Subscriber access provided by Kaohsiung Medical University

Article 2

2

O-tolerant H activation by an isolated large subunit of a [NiFe] hydrogenase Sven Hartmann, Stefan Frielingsdorf, Alexandre Ciaccafava, Christian Lorent, Johannes Fritsch, Elisabeth Siebert, Jacqueline Priebe, Michael Haumann, Ingo Zebger, and Oliver Lenz Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.8b00760 • Publication Date (Web): 15 Aug 2018 Downloaded from http://pubs.acs.org on August 17, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

O2-tolerant H2 activation by an isolated large subunit of a [NiFe] hydrogenase Sven Hartmann†,‡, Stefan Frielingsdorf†,‡,*, Alexandre Ciaccafava†, Christian Lorent†, Johannes Fritsch§, Elisabeth Siebert†, Jacqueline Priebe†, Michael Haumann∥, Ingo Zebger†, and Oliver Lenz†,*

†

Department of Chemistry, Sekr. PC14, Technische Universität Berlin, 10623 Berlin, Germany

§

Department of Biology, Humboldt-Universität zu Berlin, 10115 Berlin, Germany

∥

Department of Physics, Freie Universität Berlin, 14195 Berlin, Germany.

‡

These authors contributed equally to this work.

Corresponding Authors * Stefan Frielingsdorf, Oliver Lenz: Department of Chemistry, PC14, Technische Universität Berlin, Straße des 17. Juni, 10623 Berlin, Germany; [email protected], [email protected]

Keywords: metalloenzyme, nickel, iron, hydrogenase, maturation, hydrogen-deuterium exchange, Fourier transform infrared (FTIR), electron paramagnetic resonance (EPR)

1 ACS Paragon Plus Environment

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT The catalytic properties of hydrogenases are nature’s answer to the seemingly simple reaction H2 ⇌ 2H+ + 2e-. Members of the phylogenetically diverse subgroup of [NiFe] hydrogenases generally consist of at least two subunits, where the large subunit harbors the H2-activating [NiFe] site and the small subunit contains iron-sulfur clusters mediating e- transfer. Typically, [NiFe] hydrogenases are susceptible to inhibition by O2. Here, we conducted systems’ minimization by isolating and analyzing the large subunit of one of the rare members of O2-tolerant [NiFe] hydrogenases, namely the preHoxG protein of the membrane-bound hydrogenase from Ralstonia eutropha. Unlike previous assumptions, preHoxG was able to activate H2 as it clearly performed catalytic hydrogen/deuterium exchange. However, it did not execute the entire catalytic cycle described for [NiFe] hydrogenases. Remarkably, H2 activation was performed by preHoxG even in the presence of O2 although the unique [4Fe3S] cluster located in the small subunit and ascribed to be crucial for tolerance toward O2 was absent. These findings challenge the current understanding of O2 tolerance of [NiFe] hydrogenases. The applicability of this minimal hydrogenase in basic and applied research is discussed.

INTRODUCTION Microbes widely use molecular hydrogen (H2) as energy source by H2 oxidation as well as electron sink in the course of H+ reduction yielding H2.1 The corresponding enzymes enabling microbes to utilize H2 are termed hydrogenases. Their reactivity depends on metal-based cofactors found in their active sites and accordingly, they can be divided into three classes, i.e. [Fe] , [FeFe] and [NiFe] hydrogenases.2–4 Sensitivity toward O2 is widespread among hydrogenases, consistent with the fact that most microorganisms catalyze H2 conversion


Page 2 of 43

Page 3 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

exclusively under anaerobic conditions. However, certain H2 oxidizers such as the βproteobacterium Ralstonia eutropha are able to use H2, CO2 and O2 for lithoautotrophic growth and contain O2-tolerant [NiFe] hydrogenases.5 [NiFe] hydrogenases consist of at least two subunits forming the hydrogenase module (Figure 1A), which is generally connected to further protein modules with various functions.1,2 The hydrogenase module comprises a large subunit harboring the active nickel-iron center and a small subunit carrying at least one iron-sulfur cluster 2 connecting the active site with the physiological electron acceptor/donor.3 Four cysteine residues coordinate the active site nickel, and two of these cysteines serve as bridging ligands to the iron. Two CN ligands and one CO are additionally coordinated to the iron.6,7 Substrate binding occurs at a vacant coordination site between the iron and nickel.6,8 At present it is assumed that at least one [FeS] cluster in electron-transfer distance to the [NiFe] site is a prerequisite for H2 activation,9–11 although separate [NiFe] center-carrying subunits have been rarely investigated so far.12–14 However, the investigation of single large subunits is of interest since they represent a minimal setup of a [NiFe] hydrogenase comprising exclusively the [NiFe] center embedded in a protein shell. Such size reduction will facilitate spectroscopic investigations of the catalytic center, which have thus far been challenging because of the presence of the [FeS] clusters of the small subunit. Therefore, we here investigated the isolated large subunit of the well-characterized membrane-bound hydrogenase (MBH) 5,15 of R. eutropha and found clear evidence for H2 activation mediated by this subunit’s [NiFe] center.


Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

EXPERIMENTAL PROCEDURES Strains and plasmids Bacterial strains and plasmids are listed in Table 1. R. eutropha HF649 harbors plasmid pGE636 which encodes the entire MBH operon with a Strep-tag®II-encoding sequence at the 3’ end of hoxK.16 R. eutropha HF687 is a derivative of R. eutropha HF686 (17) and carries a hoxM deletion and a Strep-tag®II-encoding sequence at the 3’ end of hoxG. R. eutropha HF795 was generated by deletion of the small subunit gene hoxK in HF687 17 using plasmid pCH499. 18 Apart from preHoxG, strain HF795 synthesizes the regulatory hydrogenase (RH, HoxBC). The gene cluster encoding the Hyp machinery was deleted by double homologous recombination 19 through conjugation of E. coli S17-1 (pCH547) with R. eutropha HF795. This resulted in R. eutropha HF796. Media and growth conditions All media for R. eutropha cultures were composed of an H16 buffer system (25 mM Na2HPO4, 11 mM KH2PO4, pH 7.0) and contained mineral salts (37.4 mM NH4Cl, 1 µM NiCl2, 18 µM FeCl3, 68 µM CaCl2 and 810 µM MgSO4). Pre-cultures of R. eutropha grew in a medium containing additional 0.4 % fructose (FN medium) for 48 h at 30 °C and 120 rpm. Main cultures for large subunit preparations were grown in mineral salts medium containing 0.2 % fructose and 0.2 % glycerol (FGN medium) in baffled Erlenmeyer flasks (filled to 20 % of their capacity) at 30 °C and 120 rpm until an optical density at 436 nm (OD436) of approximately 11 was reached (after approximately 72 h). For strain HF796, no NiCl2 but 15 µM nitrilotriacetic acid was added to the main culture. Main cultures for isolation of fully matured MBH were grown in medium containing 0.05 % fructose and 0.4 % glycerol (FGNmod medium) in Erlenmeyer flasks (filled to 80 % of their capacity) at 30 °C and 120 rpm until an OD436 of 11 was reached (after


Page 4 of 43

Page 5 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

approximately 7-9 days). Cells were harvested by centrifugation (11500 x g, 4 °C, 12 min). The resulting cell pellet was frozen in liquid nitrogen and stored at -80 °C.

Table 1. Collection of strains used in this study.

Strain/

Relevant characteristics

plasmid

Source or reference

R. eutropha HF649

pGE636 in HF631, hoxK-StrepTagII

16

HF687

Derivative of HF686 (SH-, AH-); ∆hoxM, hoxG-Strep-tag®II

17

HF795

Derivative of HF687; ∆hoxM, ∆hoxK, hoxG-Strep-tag®II

This study

HF796

Derivative of HF795; ∆hoxM, ∆hoxK, ∆hypA1B1F1CDEX, hoxG-Strep-tag®II

This study

E. coli S17-1

Tra+ recA, pro thi, hsdR chr:RP4-2

20

Plasmid pCH499

pCH499, 2 kbp PvuII-SspI fragment harboring a ∆hoxK allele in pLO2

18

pCH547

pCH547, 2.7 kbp SspI fragment of the hyp gene cluster in pLO1 designed to

21

introduce a 6.3 kbp in-frame deletion into the genomic hypA1B1F1CDEX gene cluster

Protein purification Soluble proteins. Cell pellets were resuspended in 3 mL (per 1 g cells, wet weight) resuspension buffer (50 mM K2HPO4/KH2PO4 at pH 7.0) containing protease inhibitor cocktail (Complete EDTA-Free, Roche) and DNase I (Roche). The resuspended cells were disrupted in a French


Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

pressure cell (G. Heinemann Ultraschall- und Labortechnik, Schwäbisch Gmünd, Germany) at 124.11 MPa. Cell debris and membranes were sedimented by ultracentrifugation (100000 x g, 4 °C, 45 min) yielding soluble protein extract in the supernatant. Membrane-bound proteins. Cell pellets were resuspended in 3 mL (per 1 g cells, wet weight) resuspension buffer containing protease inhibitor mixture and DNase I. The resuspended cells were disrupted in a French pressure cell at 124.11 MPa. Unbroken cells and cell debris were sedimented by centrifugation (4000 x g, 4 °C, 20 min) yielding an emulsion composed of membranes and soluble proteins. The emulsion was treated with K3Fe(CN)6 to a final concentration of 50 mM and subsequently ultracentrifuged (100000 x g, 4 °C, 1 h), yielding the cell membranes as pellet. The membrane pellet was washed through homogenization with a Potter-Elvehjem homogenizer in 3 mL membrane buffer (50 mM K2HPO4/KH2PO4, pH 7.0, 150 mM NaCl) containing protease inhibitor cocktail. The suspension was then ultracentrifuged (100000 x g, 4 °C, 35 min) yielding clean membranes as a pellet. The membrane pellet was frozen in liquid nitrogen and stored at -80 °C. The frozen membrane pellet was homogenized in 10 mL protease inhibitor-containing membrane buffer per 1 g membrane pellet and the suspension was supplemented with Triton X-114 (BioChemica, Applichem) as detergent to a final concentration of 2 % (v/v). Ultracentrifugation (100000 x g, 4 °C, 25 min) yielded solubilized membrane proteins in the supernatant. Purification of MBHStrep was carried out as described previously. 22 For purification of preHoxGStrep and apo-preHoxGStrep the corresponding supernatant was loaded on a Strep-Tactin high capacity column (IBA, Göttingen, Germany). For 50 mL of soluble extract, 2 mL of bed volume was used. The column was washed with 10 bed volumes washing buffer (50 mM K2HPO4/KH2PO4, pH 7.0, 150 mM NaCl), and the preHoxGStrep and apo-preHoxGStrep proteins were eluted six times with 0.5 bed volumes of


Page 6 of 43

Page 7 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

elution buffer (50 mM K2HPO4/KH2PO4, pH 7.0, 150 mM NaCl, 10 % glycerol, 3 mM ddesthiobiotin). The eluate was concentrated by centrifugation (4000 x g, 4 °C, 20 min) using a centrifugal ultrafiltration device (Amicon Ultra Ultracel 30K, Millipore). The resulting concentrate was frozen in liquid nitrogen and stored at -80 °C. Protein concentrations were determined with the PierceTM BCA Protein Assay Kit (Thermo Scientific) method, using bovine serum albumin as a standard. The samples were checked for purity by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis, and the gels were stained with Coomassie brilliant blue G-250.23 Western blots were performed following published procedures.24 Proteins were transferred to a nitrocellulose membrane (BioTrace™, Pall Corporation, Dreieich, Germany) using a fast semidry transfer buffer.25 Polyclonal antibodies raised against the MBH large subunit (anti-HoxG) were used for detection of HoxG.18 Polyclonal antibodies raised against the MBH small subunit (anti-HoxK) were used for detection of HoxK.16 As secondary antibody, alkaline phosphatase-labeled goat anti-rabbit IgG (Dianova, Hamburg, Germany) was used.

Size-exclusion chromatography Size-exclusion chromatography was carried out as described previously using buffer composed of 50 mM K2HPO4/KH2PO4, pH 7.0 and 150 mM NaCl.26

Activity assays H/D exchange reaction. The activity of the enzymes was calculated from the D2 and HD production rates upon uptake of H2 (H/D exchange) in D2O-based buffer. The amount of D2 and HD produced was measured in a modified O2 electrode chamber (DW2/2, Hansatech, King’s Lynn, UK) connected by a homemade brass adaptor to a mass spectrometer (OmniStar™ G3D


Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

301 T3 Quadrupole, Pfeiffer Vacuum, Berlin, Germany). Settings for the mass spectrometer have been described previously.27 The measured mass-to-charge ratios (m/z) were 2 for H2, 3 for HD and 4 for D2. As controls and for leakage detection, m/z values of 18 (H2O), 28 (N2), 32 (O2) and 44 (CO2) were recorded. The reaction chamber was separated from the vacuum side of the mass spectrometer by a gas-permeable Teflon membrane (0.0125 mm, Hansatech). The reaction volume was 1.9 mL, and the measurement was carried out at 25 °C under continuous stirring on a magnetic plate with a magnetic stir bar at 1250 rpm. The reaction buffer (50 mM citric acid/100 mM K2HPO4, pH 5.5, in D2O) was bubbled with H2 gas until saturation. The bubbling was stopped and a protein sample was injected using a Hamilton syringe. The D2 and HD production rates were determined as described previously.27 Apo-preHoxG served as negative control and was used for baseline correction. Notably, the marginal slope determined for apopreHoxG was unaffected by the protein amount in the assay, which was tested in a range from 1 µg – 1000 µg. Furthermore, the D2 production rate was only negligibly increased in comparison to background signal fluctuations observed for H2-saturated D2O-based buffer without protein. pH value influence. To test the influence of the pH value on the D2 production rates, pH values of 3, 4, 5, 5.5, 6 and 7 were adjusted by adequately changing the ratio of the buffer components citric acid and K2HPO4. O2 and CO influence. To test the influence of O2 and CO on the D2 production rates, an aliquot of the reaction buffer was saturated with O2 or CO gas. The reaction chamber was filled with 1.52 mL of reaction buffer and saturated with H2 gas. Desired O2 and CO concentrations at constant H2 level were achieved by adding corresponding amounts of O2, CO and N2-saturated buffer before injection of the protein sample.


Page 8 of 43

Page 9 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Hydrogen oxidation. H2-mediated reduction of redox dyes was measured spectrophotometrically as previously described 28 using a Cary®50 UV-Vis spectrophotometer (Varian, Agilent, California, USA). The reaction was carried out in a 3 mL glass cuvette with a layer thickness of 1 cm at 30 °C. The reaction volume was 1.9 mL and a buffer composed of 50 mM K2HPO4/KH2PO4, pH 5.5 saturated with H2 gas served as reaction buffer. The reaction buffer was filled into the cuvette and methylene blue, phenazine methosulfate, or methyl viologen was added to a final concentration of 0.21 mM, 0.05 mM, and 20 mM, respectively. The wavelengths and the molar extinction coefficients were 570 nm and 13.1 mM-1 cm-1 for methylene blue, 388 nm and 22 mM-1 cm-1 for phenazine methosulfate and 605 nm, 14.5 mM-1 cm-1 for methyl viologen. The mixtures were again bubbled with H2 for 2 min before the protein samples were injected. Proton reduction. H2 evolution using chemically reduced methyl viologen was measured using a modified Clark-type electrode.29 The reaction was carried out as described previously 30 at 30 °C under continuous stirring in 1.3 mL N2-saturated reaction buffer (50 mM K2HPO4/KH2PO4, pH 7.0). For measurement, methyl viologen was added to a final concentration of 5 mM. Then, sodium dithionite was added to a final concentration of 8 mM to reduce methyl viologen. H2 evolution was measured upon sample injection.

Spectroscopy IR spectroscopy. IR spectra were recorded with a spectral resolution of 2 cm−1 using a Bruker Tensor 27 FTIR spectrometer, equipped with a liquid nitrogen-cooled MCT detector. The sample compartment was purged with dry air, and the sample was held in a temperature-controlled (10 °C) gas-tight IR transmission cell for liquid samples (10 µL volume, 50 µm optical path


Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

length), equipped with CaF2 windows. The Bruker OPUS software, version 5.5 or higher, was used for data acquisition and evaluation. Aerobically isolated, H2-saturated and CO-saturated preHoxG samples, as well as aerobically isolated apo-preHoxG samples were analyzed at a concentration of 0.7 mM, while native, heterodimeric MBH was analyzed at a concentration of 0.3 mM. EPR spectroscopy. EPR samples of apo-preHoxG, preHoxG, and native MBH were analyzed at a concentration of 0.5 mM, 0.8 mM, and 0.2 mM, respectively. The measurements were carried out as previously described.31

Metal content analysis Total reflection X-ray fluorescence (TXRF) analysis 32 was used for simultaneous quantification of nickel and iron in a Picotax spectrometer at Röntec (Berlin, Germany) using 1 µL of concentrated and dried protein solutions. Nickel and iron contents were determined relative to a gallium standard. All TXRF-generated values were derived from two biological replicates with five technical replicates each (n = 10).


Page 10 of 43

Page 11 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

RESULTS Purification of the MBH large subunit For isolation of the large [NiFe] hydrogenase subunit, HoxG, from R. eutropha, a suitable strain was constructed. A previous attempt to purify MBH by affinity chromatography via the HoxG subunit carrying an N-terminal Strep-tag II failed.33 We therefore fused by genetic engineering a Strep-tag II sequence to the C terminus of the precursor of HoxG. In the course of HoxG maturation, a C-terminal peptide of preHoxG is cleaved off by the specific endopeptidase, HoxM, once the [NiFe] site is fully assembled.16,18 To prevent cleavage of the C terminus and the concomitant loss of the affinity tag, the hoxM gene was deleted, resulting in strain R. eutropha HF687.17 Immunologic analysis of preHoxG isolated from R. eutropha HF687 showed that minor amounts of the small hydrogenase subunit, HoxK, were co-purified (Figure S1). In order to exclude any contamination of preHoxG with HoxK, the hoxK gene was deleted as well. The resulting strain was named R. eutropha HF795 and synthesized the Strep-tagged HoxG precursor, preHoxG (Figure 1B). Notably, analyses of this particular preHoxG protein allow for conclusions on the influence of the C-terminal extension on the catalytic properties of HoxG. In order to generate a negative control, we aimed at isolation of a preHoxG variant devoid of the [NiFe] cofactor, termed apo-preHoxG (Figure 1C).


Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

A

Page 12 of 43

B

H2

C Ni-Fe

Ni-Fe 2H+

e-

2e-

MBH heterodimer

preHoxG

apo-preHoxG

Figure 1. Schematic representations of A, MBH heterodimer, B, preHoxG, and, C, apo-preHoxG. HoxG is colored in blue and HoxK is depicted in green. The transmembrane region of HoxK is drawn as a green stick. The C-terminal extension of HoxG, which is cleaved off in the course of maturation, is shown as blue stick. The location of the Strep-tag II peptide is indicated in purple. The blow-up shows the composition of the catalytic center with an “X” representing the position of variable bridging ligands.

For that purpose, we deleted from R. eutropha HF795 the genes encoding the Hyp machinery - a set of auxiliary proteins responsible for [NiFe] cofactor generation and its insertion into the active site of preHoxG.34 The proteins, preHoxG, apo-preHoxG, and the native MBH heterodimer were purified via Strep-Tactin affinity chromatography as described in the experimental procedures. According to SDS-PAGE analyses, protein bands corresponding to a molecular mass of approximately 70 kDa were identified, matching well the expected molecular masses of preHoxG and apo-preHoxG (Figures S2 and S3). To examine if the observed bands are related to preHoxG and apo-preHoxG we performed an immunoblot analysis of the different purification steps using an antibody raised against HoxG.18 The elution fractions of both HoxG variants displayed only one characteristic band at approximately 70 kDa (Figures S2 and S3). This showed that we were able to produce and purify the isolated large subunit of MBH. According to visual inspection of the SDS-PAGE gels, the HoxG derivatives were obtained with 12 ACS Paragon Plus Environment

Page 13 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

a purity of approximately 80 %. The two major contaminations at 58 and 72 kDa are typically observed in Strep-Tactin-based purifications from R. eutropha soluble extracts.17,35 From 100 g (wet weight) of cells, we routinely obtained 4-6 mg of preHoxG and apo-preHoxG.

The preHoxG active site is partially occupied with [NiFe] cofactor First, we analyzed the metal content of preHoxG and apo-preHoxG in order to inspect whether or not these proteins contain the constituents of the [NiFe] complex. According to TXRF analyses apo-preHoxG contained nearly no nickel (0.01 ± 0.01, n = 10). However, more than 50 % of the proteins (0.56 ± 0.08, n = 10) seem to carry an iron atom (Figure S4). Thus, the active site cysteine residues might be involved in iron binding prior to delivery of the complete [NiFe] cofactor by the Hyp machinery. A previous study even suggested the presence of substoichiometric amounts of a [FeS] cluster in the apo-form of the large subunit of Hyd-2 from Escherichia coli which could be chemically reconstituted to a [4Fe4S] cluster.36 By contrast, the recent crystal structure of the premature [NiFe] subunit preHyhL from Thermococcus kodakarensis revealed an empty [NiFe] site despite significant iron content detected in a metal determination assay.37 In case of the preHoxG protein we found a significantly higher, but still substoichiometric nickel content of 18 % (0.18 ± 0.07, n = 10) and an almost stoichiometric amount of iron (0.97 ± 0.17, n = 10) (Figure S4). For comparison, native heterodimeric MBH contains one nickel and one iron per HoxG subunit.15 Next, we investigated if preHoxG contains the typical active site iron ligands, namely one CO and two cyanides. Purified native MBH possessing the prototypic NiFe(CN)2CO center was used for comparison. Therefore, preHoxG, apo-preHoxG, and MBH were subjected to IR spectroscopy, and the resulting absorption spectra were compared (Figure 2 and Table S1).


Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2. IR absorbance spectra of HoxG derivatives incubated under different conditions. A, aerobically purified apo-preHoxG (as isolated, pH 7.0). B, aerobically purified preHoxG (as isolated, pH 7.0). C, aerobically purified preHoxG subsequently treated with 1 bar of H2 gas for 30 minutes (pH 7.0). D, aerobically purified preHoxG treated with 1 bar CO gas for 30 minutes (pH 7.0). E, aerobically purified MBH (pH 5.5)) mainly residing in Nir-B state. The signal intensities were normalized according to the molar protein concentration (taking molecular weight differences into account). Intensity of spectrum E was divided by 5 for clarity. For A, B and E: n = 2, representative spectra are shown; for C and D: n = 1.

The spectrum (Figure 2E) of native MBH revealed mainly the characteristic CN and CO absorption bands indicative of the Nir-B state (at 2081, 2098 and 1948 cm-1, respectively), which is typically observed upon aerobic protein isolation. Further absorptions (at 1933 and 2105 cm-1)


Page 14 of 43

Page 15 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

with minor band intensities refer to sub-populated “ready” species.38 In case of aerobically isolated preHoxG, CN- and CO-related vibrational bands were clearly visible (Figure 2B). However, the signals exhibited significantly lower intensities. Normalization to the molar protein concentration (taking molecular weight differences into account) revealed an occupancy of ~20 % of the CO/CN ligands when compared to native MBH. For apo-preHoxG we could not detect any characteristic CO/CN stretching vibrations, confirming the absence of the active site CN and CO ligands (Figure 2A). This further indicates an iron coordination in apo-preHoxG that is different from the classical coordination found in fully assembled [NiFe] hydrogenase. Although the IR spectra demonstrate that preHoxG is - at least partially - loaded with a NiFe(CN)2CO complex, the ν(CO) and ν(CN) were found at frequencies different than those of MBH. This indicates an altered structural/electronic environment of the CN and CO ligands in as-isolated preHoxG. In addition, the observed bands are significantly broadened, indicating structural heterogeneities at the active site. To investigate the redox state of nickel in as-isolated preHoxG we performed EPR spectroscopy. In [NiFe] hydrogenases nickel has been found to switch between the diamagnetic 2+ state and the paramagnetic 1+ and 3+ states.3 The aerobically isolated native MBH, for instance, resides predominantly in the Nir-B state (Figure 3C) featuring a Ni3+.38 The EPR spectra of both apopreHoxG and preHoxG revealed no characteristic nickel-derived signals (Figure 3A,B), indicating the exclusive presence of diamagnetic nickel species (i.e. Ni2+/4+). Thus, in case of nickel in preHoxG the EPR-active 3+ state can be excluded. Since Ni4+ species are highly reactive 39,40 we concluded that nickel resides in the 2+ state, in spite of the aerobic conditions during protein purification. The active site iron in fully matured [NiFe] hydrogenases has been found to reside exclusively in the EPR-silent low-spin Fe2+ state throughout all active site states


Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

identified so-far,41,42 and, thus, iron-derived EPR signals were exclusively detected originating from [FeS] clusters (Figure 3F). Measurements of apo-preHoxG and preHoxG at 10 K revealed no characteristic signals indicative of [FeS] clusters in relevant amounts (Figure 3D,E). The minor signals observed at approx. g = 2.0 are likely related to substoichiometric (

O2-Tolerant H2 Activation by an Isolated Large Subunit of a [NiFe

Recommend Documents