JOURNAL OF BACTERIOLOGY, May 2011, p. 2487–2497 0021-9193/11/$12.00 doi:10.1128/JB.01427-10 Copyright © 2011, American Society for Microbiology. All Rights Reserved.
Vol. 193, No. 10
The Maturation Factors HoxR and HoxT Contribute to Oxygen Tolerance of Membrane-Bound [NiFe] Hydrogenase in Ralstonia eutropha H16䌤† Johannes Fritsch, Oliver Lenz, and Ba¨rbel Friedrich* Institut fu ¨r Biologie/Mikrobiologie, Humboldt-Universita ¨t zu Berlin, Chausseestrasse 117, 10115 Berlin, Germany Received 25 November 2010/Accepted 6 March 2011
The membrane-bound [NiFe] hydrogenase (MBH) of Ralstonia eutropha H16 undergoes a complex maturation process comprising cofactor assembly and incorporation, subunit oligomerization, and finally twinarginine-dependent membrane translocation. Due to its outstanding O2 and CO tolerance, the MBH is of biotechnological interest and serves as a molecular model for a robust hydrogen catalyst. Adaptation of the enzyme to oxygen exposure has to take into account not only the catalytic reaction but also biosynthesis of the intricate redox cofactors. Here, we report on the role of the MBH-specific accessory proteins HoxR and HoxT, which are key components in MBH maturation at ambient O2 levels. MBH-driven growth on H2 is inhibited or retarded at high O2 partial pressure (pO2) in mutants inactivated in the hoxR and hoxT genes. The ratio of mature and nonmature forms of the MBH small subunit is shifted toward the precursor form in extracts derived from the mutant cells grown at high pO2. Lack of hoxR and hoxT can phenotypically be restored by providing O2-limited growth conditions. Analysis of copurified maturation intermediates leads to the conclusion that the HoxR protein is a constituent of a large transient protein complex, whereas the HoxT protein appears to function at a final stage of MBH maturation. UV-visible spectroscopy of heterodimeric MBH purified from hoxR mutant cells points to alterations of the Fe-S cluster composition. Thus, HoxR may play a role in establishing a specific Fe-S cluster profile, whereas the HoxT protein seems to be beneficial for cofactor stability under aerobic conditions. prerequisite for biotechnological applications, such as enzymatic fuel cells (74) or light-driven H2 production by coupling oxygenic photosynthesis to hydrogenase (22, 34, 65). When establishing these systems in living cells, not only do aspects of protein stability or catalysis need to be considered but also posttranslational maturation processes should be well adapted to O2 exposure. Assembly and incorporation of complex metallocenters that are often deeply buried inside the protein represent enormous challenges for a living cell. Such processes are frequently assisted by a number of accessory proteins and chaperones which guarantee correct metal acquisition, cofactor assembly, controlled folding, proteolytic processing, and targeted transport to a subcellular location. [NiFe] hydrogenases undergo a particularly complex maturation process in order to synthesize the heterobimetallic active site with biologically uncommon endogenous CO and CN⫺ ligands (50). The megaplasmid pHG1-encoded gene cluster required for active MBH expression in R. eutropha encompasses 21 genes (64) (Fig. 1), of which only three have coding functions for the structural polypeptides of the enzyme. Cofactor incorporation into the active-site subunit HoxG is accomplished by at least six Hyp proteins (Fig. 1) (12, 30). Moreover, a specific chaperone, HoxL, and a transfer protein, HoxV, which is most likely involved in the shuttle of the Fe(CN)2CO cofactor intermediate, are required for MBH maturation in R. eutropha (42). Interestingly, genes coding for HoxV homologues were found only in the gene clusters of cytochrome b-linked, MBH-like [NiFe] hydrogenases. A dual function was assigned to the accessory proteins HoxO and HoxQ (61). They interact with the small
Oxidation of molecular hydrogen under aerobic conditions is the driving force for autotrophic growth of the betaproteobacterium Ralstonia eutropha H16. This model organism harbors genes encoding at least three oxygentolerant [NiFe] hydrogenases (15): a membrane-bound (MBH), a cytoplasmic NAD⫹-reducing, and a regulatory hydrogenase. The MBH consists of a large subunit HoxG (67.1 kDa) containing the Ni-Fe active site and a small subunit HoxK (34.6 kDa) accommodating three Fe-S clusters. Physiologically active MBH is exposed to the periplasm and connected to a membrane-integral cytochrome b, denoted HoxZ, which has a dual function as electron acceptor and anchor to the membrane. Electrons derived from H2 oxidation are conducted through the Fe-S cluster relay in HoxK via the cytochrome to the quinone pool of the respiratory chain (8, 57). The MBH from R. eutropha has been shown to be remarkably O2 and CO tolerant, i.e., it performs H2 conversion in the presence of these usually highly inhibiting agents (18, 41, 74). This feature contrasts with [NiFe] hydrogenases abundant in anaerobic microbes. These standard types of hydrogenases need reductive activation upon exposure to O2 to slowly recover catalytic activity (20, 75). O2 tolerance is an important
* Corresponding author. Mailing address: Institut fu ¨ r Biologie/ Mikrobiologie, Humboldt-Universita¨t zu Berlin, Chausseestrasse 117, 10115 Berlin, Germany. Phone: 49-30-2093-8101. Fax: 49-302093-8102. E-mail:
[email protected]. † Supplemental material for this article may be found at http://jb .asm.org/. 䌤 Published ahead of print on 25 March 2011. 2487
2488
J. BACTERIOL.
FRITSCH ET AL.
oxidizing bacteria, known as “Knallgas bacteria,” that thrive in aerobic habitats (62, 73). This obvious correlation might indicate a protective role of these additional proteins against detrimental effects of O2. To explore the function of hoxR and hoxT in more detail, we investigated mutants carrying in-frame deletions within the respective genes for MBH-mediated growth on H2, enzymatic activity, and maturation intermediates. Copurification results gave insights into interactions of HoxR and HoxT with the MBH and its maturation factors, leading to the conclusion that HoxR affects MBH biosynthesis at an early stage of maturation, whereas HoxT appears to protect the MBH at the completion of the process. MATERIALS AND METHODS
FIG. 1. MBH gene cluster and model of the MBH maturation pathway. The genes hoxKGZ encode the small and large hydrogenase subunits (blue) and a membrane-integral b-type cytochrome (gray); the hoxMLOQRTV genes (yellow) code for MBH-specific accessory proteins; the hypA1B1F1CDEX genes (green) are responsible for active site assembly (downstream genes involved in the regulation of hydrogenase gene expression are not shown for simplification). The maturation model represents the current stage of our work. Isc and Suf symbolize the general machinery for assembly and insertion of Fe-S clusters (52) and are likely involved in incorporation of the Fe-S centers into HoxK. For further details and references, see the text.
subunit HoxK, via the Tat signal peptide, and likely protect the Fe-S clusters in HoxK against reactive oxygen species. Furthermore, HoxO and HoxQ were suggested to prevent premature export of the MBH small subunit. Folding and oligomerization of the HoxG subunit of the MBH are triggered by removal of a C-terminal peptide catalyzed by a specific endoprotease, HoxM (10), which was proposed to have a proofreading function (44). The cofactor-containing MBH precursor is a substrate of the Tat translocon (9) that exports the completely folded proteins from the cytoplasm to the periplasm (58). Since only the small subunit contains an N-terminal Tat leader peptide, the large subunit is suggested to be cotranslocated through the Tat pathway by a hitchhiker mechanism (54). Homologous clusters of the accessory genes hoxO, hoxQ, hoxR, and hoxT are found predominantly in aerobically H2-
Media and growth conditions. Basic media and growth conditions for R. eutropha and Escherichia coli strains have been described previously (42). Antibiotics for R. eutropha were used at the following concentrations: kanamycin, 400 g/ml; tetracycline, 10 g/ml. Antibiotics for E. coli were used at the following concentrations: kanamycin, 50 g/ml; tetracycline, 10 g/ml; ampicillin, 100 g/ml. Ralstonia eutropha was cultivated heterotrophically at 30°C in mineral medium containing 0.2% (wt/vol) fructose and 0.2% (wt/vol) glycerol (FGN) in baffled 500-ml Erlenmeyer flasks filled with 150 ml medium shaken at 120 rpm under air, which is defined as standard cultivation. For comparative activity measurements, cells were grown in 1.5 liters of FGN in a 5-liter fermentor (Biostat MD; Braun) at continuous stirring at 400 rpm and aeration with 1.33 liters/min air (21% [vol/vol] O2) or with a gas mixture of 10% (vol/vol) O2 and 90% (vol/vol) N2 until optical density (⫾standard deviation [SD]) of the culture at 436 nm (OD436) reached 10 ⫾ 1. Severe O2 limitation was achieved during cultivation of cells in modified FGN (FGN*) containing 0.05% (wt/vol) fructose and 0.4% (wt/vol) glycerol in baffled 250-ml Erlenmeyer flasks filled with 200 ml culture and shaken under air at 120 rpm for about 150 h at 30°C (OD436 ⫾ SD, 10 ⫾ 2). For purification of MBH variants, cells were grown in baffled 5-liter Erlenmeyer flasks filled with 2 liters FGN* medium for approximately 96 h (OD436 ⫾ SD, 11 ⫾ 1). Cells were harvested by centrifugation at 6,000 ⫻ g for 15 min at 4°C. For comparative Western blot analysis and real-time reverse transcription (RT)-PCR, cells were grown in FGN* medium to an OD436 ⫾ SD of 4 ⫾ 0.5 in 100-ml Erlenmeyer flasks filled with 10 ml medium incubated in gas-tight jars filled with gas mixtures of various O2 partial pressures (pO2) (10 to 100% [vol/vol]) balanced by N2 or under severe O2 limitation as described above. Lithoautotrophic cultures were grown on mineral agar plates at 30°C under an atmosphere of 60% (vol/vol) H2, 10% (vol/vol) CO2, and various O2 concentrations balanced by N2. For expression of genes under the control of the acoX promoter (51), cells were grown in FGN under standard conditions. Gene expression was induced in the late growth phase through the addition of 20 mM acetoin, and cells were harvested after 3 h. Isolation of membranes and Strep-Tactin affinity chromatography. All fractionation and purification steps were performed at 4°C. R. eutropha cells were resuspended in resuspension buffer (3 ml 50 mM potassium phosphate [KPO4] buffer, pH 7.0, per 1 g wet weight) containing Complete EDTA-free protease inhibitor cocktail (Roche Applied Science) and DNase I. The cell suspension was subsequently disrupted in a French pressure cell (SLM Aminco) via two passages at 18,000 lb/in2. The resulting crude extract was treated by sonication (Branson Sonifier), and the cell debris was removed by low-speed centrifugation (4,000 ⫻ g, 30 min). Membrane and soluble fractions were separated by ultracentrifugation (100,000 ⫻ g for 60 min). The membrane pellet was washed by resuspension in an appropriate volume of 50 mM KPO4 buffer (pH 7.0) followed by ultracentrifugation (100,000 ⫻ g for 35 min). In the case of MBH purification, K3[Fe(CN)6] was added to the crude extract at a final concentration of 50 mM immediately after the removal of cell debris. The resulting oxidized membranes were further processed as described above. Membrane proteins were solubilized in 10 ml solubilization buffer (65 mM KPO4, 300 mM NaCl, 2% [wt/vol] Triton X-114, Complete protease inhibitor cocktail [Roche Applied Science, pH 7.02]) per 1 g of membrane pellet by stirring on ice for 2 h. After ultracentrifugation (100,000 ⫻ g, 45 min), the supernatant containing the solubilized membrane extract was loaded onto Strep-Tactin Superflow columns (IBA, Go ¨ttingen, Germany; 1-ml bed volume for up to 25 ml of solubilized membrane extract), which were run by gravity flow. To remove unbound proteins, the columns were washed with 12 bed volumes of washing buffer (65 mM KPO4, 300 mM NaCl, pH 7.02),
VOL. 193, 2011
HoxR AND HoxT AID O2 TOLERANCE OF MBH IN R. EUTROPHA
and proteins were eluted with elution buffer (50 mM KPO4, 240 mM NaCl, 5 mM desthiobiotin, 20% [wt/vol] glycerol, pH 7). MBH-containing fractions were pooled and concentrated, and the buffer was exchanged to 50 mM KPO4, 150 mM NaCl, and 20% (wt/vol) glycerol (pH 5.5) with a centrifugal filter device (Amicon Ultra-15 PL-30; Millipore). Protein concentrations were determined with the bicinchoninic acid (BCA) kit (Pierce) with bovine serum albumin as the standard. Purity of the samples was estimated by visual inspection after separation of the proteins via SDS-polyacrylamide gel electrophoresis (SDS-PAGE) (35) and subsequent staining with Coomassie brilliant blue G-250 (79). Hydrogenase activity assay. MBH activity was measured by a spectrophotometric assay in an H2-flushed cuvette sealed with a rubber septum, with methylene blue as the electron acceptor (59). Activity measurements of isolated membrane fractions were conducted at 30°C in H2-saturated 50 mM KPO4 buffer at pH 7.0, whereas measurements on purified MBH were carried out at pH 5.5. Purification of protein complexes. Cells were grown in FGN under standard aeration and harvested as described above. All purification steps were performed at 4°C. Cells were resuspended (2 ml per 1 g wet weight) in buffer A (100 mM Tris-HCl [pH 8.0], 100 mM NaCl) containing 0.2 U/ml avidin from egg white (Sigma), Complete EDTA-free protease inhibitor cocktail, and DNase I. Cell disruption was conducted in a French pressure cell as described above. The soluble fractions were cleared from cell debris and membranes by ultracentrifugation (100,000 ⫻ g for 50 min), and soluble extracts were loaded onto Strep-Tactin Superflow columns (1-ml bed volume for up to 70 ml of soluble extract), which were run by gravity flow. The columns were washed with 12 bed volumes of buffer A, and Strep-tagged protein was eluted from the column with buffer A containing 5 mM desthiobiotin. Fractions were pooled and concentrated with centrifugal filter devices (Amicon Ultra-15 PL-3 and Microcon YM-3; Millipore). Protein concentrations were determined by the Bradford method (13) with bovine serum albumin as the standard. Purity of the samples was estimated by visually inspecting SDS-PAGE gels stained by Coomassie brilliant blue G-250. Western blot analysis. Proteins separated by SDS-PAGE were transferred onto a nitrocellulose membrane (BA 85; Schleicher & Schu ¨ ll) according to a standard protocol (72). For immunological detection of MBH-related proteins, rabbit-derived antisera were applied in the following dilutions: antiHoxK, 1:5,000; anti-HoxG, 1:10,000; anti-HoxO, 1:5,000; anti-HoxQ, 1:10,000; anti-HoxV, 1:5,000; anti-HypC, 1:500; anti-HypD, 1:1,000. For detection of Strep-tagged and FLAG-tagged proteins, StrepTag antibodies (IBA, Go ¨ ttingen, Germany) and FLAG tag antibodies (Sigma), respectively, conjugated with alkaline phosphatase were applied. Alkaline phosphatase-labeled goat anti-rabbit and goat anti-mouse immunoglobulin G were obtained from Dianova (Hamburg). Preparation of crude extract and the periplasmic fraction. Crude cell extract for SDS-PAGE and Western blot analysis was obtained as follows: an equivalent of 1 ml culture at an OD436 of 2.5 was harvested (2 min, 21,000 ⫻ g), resuspended in 100 l SDS sample buffer (5% [wt/vol] SDS, 20% [wt/vol] glycerol, 250 mM dithiothreitol, 0.03% [wt/vol] bromophenol blue), and boiled for 10 min. Preparation of the periplasmic fraction was performed as described before (61). Sequence analysis. The hoxT gene sequence was analyzed with the transmembrane prediction methods TMHMM (http://www.cbs.dtu.dk/services/TMHMM/) (71) and Split4 (http://split.pmfst.hr/split/4/) (31) and by the signal peptide prediction tool SignalP (http://www.cbs.dtu.dk/services/SignalP/) (5). The molar extinction coefficient of MBHStrepTag at 280 nm was calculated to be 173,150 M⫺1 cm⫺1 with the ExPASy ProtParam tool (78) (http://www.expasy.org/cgi-bin /protparam), assuming that all cysteine residues were reduced. Multiple sequence alignments were done with ClustalW2 (http://www.ebi.ac.uk/Tools/clustalw2/). Comparative analysis of annotated genomes was conducted with the SEED viewer (http: //seed-viewer.theseed.org/). Bacterial strains and plasmids. Bacterial strains and plasmids are listed in Table 1. Primer oligonucleotide sequences are shown in Table S1 in the supplemental material. R. eutropha H16 is the wild-type strain harboring the indigenous megaplasmid pHG1. The MBH-overexpressing strain HF632 is a derivative of the megaplasmid-free strain R. eutropha HF631 carrying plasmid pLO6 that contains the complete MBH gene cluster from R. eutropha H16 and a single-copy ⌽(hoxK⬘-lacZ) translational fusion in the chromosome (37). E. coli JM109 (80) was used as the host in standard cloning procedures. For attachment of a StrepTag II sequence to hoxR and hoxT on pLO6, conditional lethal suicide plasmids were constructed. A StrepTag II sequence was fused to the 3⬘ end of hoxR by applying inverse PCR on plasmid pTO#217 with primers T25 and T27. The PCR product was digested with NheI and religated, resulting in plasmid pJF8, from which a 0.53-kbp NdeI-SacI fragment, containing the hoxR-StrepTag II fusion, was transferred to the pLO2 plasmid digested with the same restriction
2489
endonucleases, yielding pCH1372. A StrepTag II sequence was fused to the 3⬘ end of hoxT by applying inverse PCR on plasmid pJF7 with primers T26 and T28. The PCR product was digested with NheI and religated, resulting in plasmid pJF9, from which a 0.68-kbp MscI-SacI fragment, containing the hoxT-StrepTag II fusion, was transferred to the PmeI-SacI-digested plasmid pLO2, yielding pCH1373. The gene region hoxQ (nucleotide [nt] 6362) to hoxV (nt 8012) was amplified with primers J1 and J2, and the 2.03-kbp XbaI-SphI-digested product was transferred to pLO2 digested with the same enzymes, yielding pCH1376. For introduction of a FLAG tag sequence at hoxR and hoxT on pHG1, pCH1376 derivatives were constructed as follows: a FLAG tag sequence was fused to the 5⬘ end of hoxT by applying inverse PCR on plasmid pJF28 with primers J8 and J9. The PCR product was digested with HindIII and religated, resulting in plasmid pJF29. A 1.21-kbp SalI-XbaI fragment from pJF29, containing the FLAG tag-hoxT fusion, was transferred to the plasmid pCH1376 cut with the same enzymes, yielding pCH1377. A FLAG tag sequence was fused to the 3⬘ end of hoxR by applying inverse PCR on plasmid pJF28 with primers J12 and J13. The product was digested with SphI and religated, resulting in pJF31, from which a 275-bp RsrII-NheI fragment, containing the hoxR-FLAG tag fusion, was transferred to pCH1376, digested with the same enzymes, yielding pCH1378. For genetic complementation experiments, expression plasmids containing the hoxR and hoxT genes under the control of PhoxF were constructed as follows: first, hoxR was amplified by PCR with primers T33 and T34 and plasmid pCH1351 as the template. From the resulting PCR product, a 0.234-kbp NcoI-BamHI fragment was transferred to NcoI-BglII-cut pLO12, resulting in pCH1379. The hoxT sequence was amplified by PCR with primers T5 and T6 and plasmid pCH1351 as the template. From the resulting PCR product, a 0.537-kbp NcoI-BglII fragment was transferred to pLO12 cut with the same enzymes, resulting in pCH1380. For expression of a FLAG tag-hoxT fusion under the control of the inducible acoX promoter (PacoX) (51), inverse PCR was applied on pLO12 with primers J23 and J24. A 7.66-kbp NheI fragment from the PCR product was religated, yielding pCH1381. Subsequently, a 0.537-kbp NcoI-BglII fragment from pCH1380, containing hoxT, was cloned into pCH1381 digested with the same enzymes, yielding pCH1382. Conjugative plasmid transfer and gene replacement. Mobilizable plasmids were transferred from E. coli S17-1 to R. eutropha by spot mating (70). Gene replacement in R. eutropha was achieved by the allelic exchange procedure based on the conditionally lethal sacB gene (39). The StrepTag II sequences attached to hoxR and hoxT were introduced on pLO6 in R. eutropha HF632 with the pCH1372 and pCH1373 plasmids, respectively, resulting in the R. eutropha strains HF757 and HF758. FLAG-tagged hoxR and hoxT alleles were introduced into pHG1 in R. eutropha HF388 with the plasmids pCH1378 and pCH1377, respectively, resulting in the R. eutropha derivatives HF841 and HF843. pCH411, pCH424, and pCH499 were employed for the introduction of in-frame deletions into the hoxG, hoxM, and hoxK genes, respectively, on plasmid pGE647 in R. eutropha HF758, resulting in the HF791, HF792, and HF793 strains. pCH426 was used for introduction of an in-frame deletion into the gene hoxR on plasmid pGE636 in R. eutropha HF649, resulting in strain HF851. UV-visible absorption spectroscopy. UV-visible measurements were carried out on a CARY 5000 UV-Vis-NIR spectrophotometer (Varian). The protein solution was filled into a 100-l quartz cuvette with an optical path length of 1 cm and measured against 50 mM KPO4 (pH 5.5), 150 mM NaCl, and 20% (wt/vol) glycerol at room temperature. For reduction of MBH samples, the cuvette was sealed with a rubber septum and flushed with moisturized N2 for 15 min to remove O2 and subsequently for 15 min with moisturized H2. Real-time RT-PCR. Cultures were incubated as described above, and total RNA was isolated with the RiboPure bacteria system (Applied Biosystems, Darmstadt) after quenching the cells with 50% (vol/vol) methanol (⫺80°C) and stabilizing the RNA with RNAprotect (Qiagen, Hilden). The integrity of the RNA was checked with RNA 6000 Nano assay chips on a Bioanalyzer 2100 (Agilent). In total, 2 g of total RNA was reverse transcribed with the highcapacity RNA-to-cDNA kit (Applied Biosystems). Diluted cDNA samples were used as templates in real-time qPCR analysis with specific primer pairs and SYBR green fluorescent dye. Real-time PCR was performed with FastSYBR Green PCR Mastermix on a model 7500 Fast PCR cycler (Applied Biosystems). Uniformity of the product was checked for every PCR by the determination of a dissociation curve. Pairs of primers with lengths of 20 to 23 nucleotides were optimized for use at an annealing temperature of 58 to 60°C. Each primer pair amplified a fragment of 150 to 200 bp. Relative expression ratios were determined by the ⌬⌬CT method, with gyrB as a constitutive control. Primers used for qPCR analysis are listed in Table S1 in the supplemental material.
2490
FRITSCH ET AL.
J. BACTERIOL. TABLE 1. Bacterial strains and plasmids used in this study
Strain or plasmid
Ralstonia eutropha H16 HF210 HF361 HF363 HF388 HF536 HF537 HF538 HF539 HF540 HF541 HF631 HF632 HF649 HF757 HF758 HF791 HF792 HF793 HF841 HF843 HF851 E. coli JM109 S17-1 Plasmids pHG1 pLO1/pLO2 pLO12 pJF7 pJF8 pJF9 pJF28 pCH411 pCH424 pCH426 pCH428 pCH499 pCH1351 pCH1372 pCH1373 pCH1376 pCH1377 pCH1378 pCH1379 pCH1380 pCH1381 pCH1382 pCH#2588 pBluescript II KS⫹ pTO#217 Litmus 28 (⫺KpnI) a
Relevant characteristic(s)
Source or reference
Wild type; MBH⫹ SH⫹ RH⫹ HoxJ⫺ pHG1⫺ (MBH⫺ SH⫺ RH⫺) Derivative of H16, ⌬hoxR Derivative of H16, ⌬hoxT Derivative of H16, SH⫺ (⌬hoxH) Derivative of HF388, ⌬hoxL Derivative of HF388, ⌬hoxO Derivative of HF388, ⌬hoxQ Derivative of HF388, ⌬hoxR Derivative of HF388, ⌬hoxT Derivative of HF388, ⌬hoxV pHG1⫺ ⌬nor(R2A2B2)::⌽关hoxK⬘-lacZ兴 pLO6 (MBH overexpression) in HF631 pGE636 (derivative of pLO6, hoxK-StrepTag II) in HF631 pGE646 (derivative of pLO6, hoxR-StrepTag II) in HF631 pGE647 (derivative pLO6, hoxT-StrepTag II) in HF631 pGE665 (derivative of pGE647, ⌬hoxK) in HF631 pGE666 (derivative of pGE647, ⌬hoxG) in HF631 pGE667 (derivative of pGE647, ⌬hoxM) in HF631 Derivative of HF388, FLAG-Tag-hoxT on pHG1 Derivative of HF388, hoxR-FLAG-Tag on pHG1 pGE674 (derivative of pGE636, ⌬hoxR) in HF631
DSM 428, ATCC 17699 33 10 10 10 10 10 10 10 10 10 37 37 61 This study This study This study This study This study This study This study This study
endA1 recA1 gyrA96 thi-1 hsdR17(rK⫺ mK⫺) relA1 supE44 ⌬(lac-proAB) 关F⬘ traD36 proAB lacIqZ⌬M15兴 Tra⫹ recA pro th hsdR chr:RP4-2
80
Megaplasmid encoding all hydrogenases known for the wild-type strain R. eutropha H16 Kmr sacB, RP4 oriT, ColE1 ori PacoX, MCS with the sequence for an N-terminal StrepTag II 1.44-kbp NheI-HindIII fragment (hoxT nt 7496 to hoxV nt 8362)a on pBluescript II KS⫹ 0.82-kbp MfeI-NheI fragment containing hoxR-StrepTag II on pBluescript II KS⫹ 1.46-kbp NheI-HindIII fragment containing hoxT-StrepTag II on pBluescript II KS⫹ 1.19-kbp SalI-XbaI fragment (hoxT nt 7534 to hoxV nt 8724)a on pBluescript II KS⫹ 2.0-kbp SphI fragment containing ⌬hoxM (⌬180 bp) on pLO1 2.2-kbp SalI-SmaI fragment containing ⌬hoxG (NdeI, BsaBI, ⌬1.524 kbp) on pLO1 1.31-kbp SalI-NheI fragment containing ⌬hoxR (SfuI, RsrII, ⌬0.195 kbp) on pLO2 1.5-kbp AspI-SphI fragment containing ⌬hoxT (SalI, EcoRV, ⌬0.165 kbp) on pLO1 2.04-kbp PvuII-SspI fragment containing ⌬hoxK (NaeI, ⌬0.438 kbp) on pLO1 8.96-kbp PstI-SacI fragment containing hoxKGZMLOQRTV on LITMUS 28 (⫺KpnI) 0.53-kbp NdeI-SacI fragment containing hoxR-StrepTag II on pLO2 0.68-kbp MscI-SacI fragment containing hoxT-StrepTag II on pLO2 2.03-kbp XbaI-SphI fragment (hoxQ nt 6362 to hoxV nt 8012)a on pLO2 1.21-kbp SalI-XbaI fragment containing FLAG tag-hoxT on pCH1376 1.21-kbp SalI/XbaI fragment containing hoxR-FLAG tag on pCH1376 hoxR-StrepTag II under the control of PhoxF on pCH#2588 hoxT-StrepTag II under the control of PhoxF on pCH#2588 Derivative of pLO12, PacoX, MCS with sequence for an N-terminal FLAG tag FLAG tag-hoxT under the control of PacoX on pCH1381 PhoxF, MCS with the sequence for a C-terminal StrepTag II PT7, PT3, Plac lacZ⬘, Ampr, ColE1 ori 0.79-kbp MfeI-NheI fragment (hoxQ nt 6706 to hoxT nt 7496)a on pBluescript II KS⫹ Apr lacZ⬘, ColE1 ori, KpnI site eliminated
70 64 39 O. Lenz, unpublished This study This study This study This study 10 10 10 10 10 61 This study This study This study This study This study This study This study This study This study O. Lenz, unpublished Stratagene Cloning Systems This study 61
Nucleotide 1 corresponds to the first nucleotide of hoxK.
RESULTS Accessory genes hoxR and hoxT are required for H2-dependent growth at high O2 concentrations. Of the six MBH-specific accessory genes of R. eutropha, the products of hoxR and
hoxT remained to be explored in more detail. Thus, mutants with in-frame deletions in hoxR and hoxT were examined physiologically for MBH-mediated lithoautotrophic growth on H2 and CO2 as sole energy and carbon sources. The strains were
VOL. 193, 2011
HoxR AND HoxT AID O2 TOLERANCE OF MBH IN R. EUTROPHA
2491
FIG. 2. MBH-driven lithoautotrophic growth of mutant strains at different O2 levels. HF388 (SH⫺) mutant strains with in-frame deletions in the MBH-specific accessory genes hoxO (HF537), hoxQ (HF538), hoxR (HF539), hoxT (HF540), hoxV (HF541), and hoxL (HF536) and the small subunit gene hoxK (HF532) were grown lithoautotrophically on mineral agar plates in the presence of 60% (vol/vol) H2, 10% (vol/vol) CO2, and various concentrations of O2.
cultivated on mineral agar plates under an atmosphere of 60% (vol/vol) H2 and 10% (vol/vol) CO2 and increasing concentrations of O2 (pO2) in N2 (Fig. 2). To prevent interference with the cytoplasmic NAD⫹-reducing soluble hydrogenase (SH) of R. eutropha, all experiments were conducted in a strain background carrying an in-frame deletion in the SH structural gene hoxH (Table 1). For comparison, corresponding mutants defective in hoxK, hoxO, hoxQ, hoxL, and hoxV were included as controls. At elevated O2 concentrations (25% [vol/vol]), only the wild-type strain showed significant growth. Under standard conditions, in the presence of 10% (vol/vol) O2, the hoxT mutant grew like the wild type, whereas the hoxR mutant failed to grow. At low O2 supply (2% [vol/vol] O2), however, the hoxR mutant recovered some growth ability, while the remaining mutant strains did not exhibit lithoautotrophic growth as reported before (10). From these results, we conclude that of the six MBH-specific accessory genes, the products of hoxR and hoxT specifically affect H2/CO2-mediated growth in response to the concentration of O2. To exclude polar effects of gene deletion on the expression of adjacent genes in the MBH cluster, plasmids carrying hoxR and hoxT were used for genetic complementation. The resulting transconjugants fully recovered growth at high O2 concentrations (see Fig. S1 in the supplemental material), supporting the notion that the individual mutations were responsible for the O2-sensitive phenotype. Impact of O2 on MBH activity. To investigate if O2-sensitive growth was the result of decreased MBH activity, the hoxR and hoxT mutants were cultivated heterotrophically under hydrogenase-derepressing conditions (63) in a fermentor aerated continuously with air (21% [vol/vol] O2) or with a gas mixture of 10% (vol/vol) O2 and 90% (vol/vol) N2. Severe O2 limitation was established by incubating the cells in Erlenmeyer flasks filled to 80% (vol/vol) with medium, resulting in a minimal gas exchange of the culture surface with the atmosphere. Membrane fractions were prepared, and hydrogenase activity was determined in an anaerobic assay by H2-dependent methylene blue reduction (Fig. 3; see also Fig. S2 in the supplemental material). Under O2 deprivation, the hoxR and hoxT mutant strains showed almost wild-type-like hydrogenase activity, whereas the supply of 10% (vol/vol) O2 led to a significant
loss of MBH activity in the membrane fraction of the ⌬hoxR mutant. The ⌬hoxT mutant maintained a relatively high activity level. Exposure to high pO2 (21% [vol/vol]) almost completely abolished MBH activity in the membranes of both mutants. These results indicate that the gene products of hoxR and hoxT are indispensable for MBH activity when cells are exposed to high pO2. The different responses to defined O2 concentrations, however, were taken as first evidence that HoxR and HoxT may act at different stages in biosynthesis of the MBH. Maturation state of MBH in the mutants at high pO2. To examine if the hoxR and hoxT deletions have any effect on MBH maturation, the respective mutant proteins were analyzed by immunoblot analysis (Fig. 4). For preparation of crude extracts, cells were grown heterotrophically at different
FIG. 3. Hydrogenase activity in the membrane fraction of hoxR and hoxT mutant strains grown under different O2 partial pressures. Cells (HF361, HF363) were grown heterotrophically in a fermentor continuously aerated with air (21% [vol/vol] O2), with a gas mixture of 10% (vol/vol) O2 and 90% (vol/vol) N2, or under O2-limited conditions in Erlenmeyer flasks. Hydrogenase activity in the membrane fraction was measured by H2-dependent methylene blue reduction in an anaerobic assay. Activity of the wild-type strain H16 under the respective conditions was taken as 100% (O2 limited, 14.3 ⫾ 0.3 U/mg; at 10% O2, 4.5 ⫾ 0.7 U/mg; at 21% O2, 3.1 ⫾ 0.4 U/mg; the absolute values of the mutant strains are depicted above the bars). One unit is defined as 1 mol H2 min⫺1. Values are averages of results from three independent replicates, and error bars represent standard deviations.
2492
FRITSCH ET AL.
FIG. 4. Western blot analysis of cell extracts from mutant strains grown at different pO2. Wild-type (HF388) and mutant strains defective in hoxR (HF539) and hoxT (HF540) were grown heterotrophically under O2-limited conditions and under a 10% (vol/vol) or a 20% (vol/vol) oxygen atmosphere. Proteins from crude extracts (an equivalent of 100 l cell culture at an OD436 of 2.5 in each lane) were separated on 10% or 12% SDS-polyacrylamide gels and identified with antibodies raised against HoxK and HoxG.
O2 concentrations. Under O2 deprivation, the immunological patterns of wild-type and mutant MBH proteins were almost identical. The large MBH subunit HoxG appeared to be fully processed under all conditions. The small subunit HoxK, however, occurred predominantly as preHoxK precursor at high pO2, particularly visible in the mutant extracts. This suggests that HoxR and HoxT are involved in MBH maturation at high O2 concentrations and that the small electrontransferring subunit of the MBH is the major target of HoxR and HoxT. Identification of HoxR and HoxT in cell extracts. To monitor occurrence of HoxR and HoxT in wild-type cell extracts, a sequence encoding a FLAG tag, which can be identified with a highly sensitive antibody, was fused genetically to the 5⬘ end of each gene. The FLAG tags attached to HoxR and HoxT did not affect MBH-mediated lithoautotrophic growth. For immunoblot analysis, cell extracts were prepared from cells grown heterotrophically under hydrogenase-derepressing conditions at various pO2. Under standard aeration, neither HoxR nor HoxT could be detected. Identification of HoxR was hindered due to the presence of nonspecific cross-reacting material of similar size (data not shown). Surprisingly, only in the presence of very high O2 concentrations (⬎50% [vol/vol]) in the headspace of the cultures could HoxT be visualized (Fig. 5). This result raises the possibility that transcription of hoxT could be enhanced under high pO2. However, real-time RT-PCR revealed no increase of transcript levels of either hoxT or hoxR
FIG. 5. Occurrence of HoxT in cells grown under different O2 concentrations. Cells harboring the sequence for an N-terminal FLAG tag fused genetically to the hoxT gene on pHG1 (HF841) were grown heterotrophically under oxygen concentrations of 20 to 100% (vol/vol) balanced by N2. Proteins from crude cell extracts (an equivalent of 100 l cell culture at an OD436 of 2.5 in each lane) were separated on 10% or 12% SDS-polyacrylamide gels, and proteins were identified via Western blot analysis with antibodies raised against HoxK, HoxG, and the FLAG tag.
J. BACTERIOL.
FIG. 6. Localization of HoxT in different cell fractions. (A) Crude extract (CE), soluble extract (SE), membrane fraction (MF), and periplasmic fraction (PE) were prepared from the wild-type strain R. eutropha H16 producing FLAG-tagged HoxT from pCH1382. Membrane fraction was washed with 50 mM KPO4 buffer (pH 7) or alternatively with 100 mM Na2CO3 buffer (pH 11). (B) Crude extract, soluble extract, and membrane fraction (washed with 50 mM KPO4 buffer, pH 7) were prepared from the hydrogenase-free strain R. eutropha HF210 carrying pCH1382. Proteins were separated on a 15% SDS-polyacrylamide gel (25 g protein per lane) and examined by Western blot analysis with FLAG tag antibody.
mRNA at high pO2 supply (see Fig. S3 in the supplemental material), suggesting an effect of O2 on the posttranscriptional level. Figure 5 shows a delayed proteolytic processing of both the small and the large MBH subunits at high pO2, indicating that O2 has a general inhibitory effect on maturation. To get further insight into the cellular location of the HoxT protein under more physiological conditions, an inducible expression plasmid was constructed encoding an N-terminally FLAG-tagged HoxT protein. The recombinant plasmid was transferred to the wild-type strain R. eutropha H16, and the transconjugant cells were grown heterotrophically under standard aeration. Transcription of the hoxT-FLAG tag was induced by the addition of acetoin in the late growth phase as described in Materials and Methods. Crude extracts as well as the soluble, membrane, and periplasmic fractions were subjected to Western blot analysis. The HoxT protein was detected in the crude extract and in the soluble and in the membrane fraction. It was absent in the periplasm (Fig. 6A). Even after washing the membrane with alkaline carbonate buffer, which is commonly used to dissociate loosely membrane-attached proteins (25), HoxT remained in the membrane fraction (Fig. 6A). A faint, faster-moving band was below the detection level in crude cell extracts (Fig. 6A) and may be a result of proteolytic degradation during cell fractionation. Similar results were obtained when the hoxT gene was expressed in the megaplasmid-free strain HF210 lacking the MBH gene cluster (Fig. 6B). The latter observation shows that the membrane attachment of HoxT is independent of the presence of the MBH protein complex. Furthermore, the amino acid sequence deduced from the hoxT gene sequence was analyzed in silico by TMHMM, Split4, and SignalP (see Materials and Methods). No indication was obtained for a transmembrane helix or the presence of an N-terminal signal peptide. HoxR and HoxT copurify with the maturation complex. The question arose of whether HoxR and HoxT are components of the maturation complex. To identify potential interaction partners, the hoxR and hoxT genes, as constituents of the MBH overexpression plasmid pMBH (37), were fused separately with a sequence encoding a C-terminal StrepTag II (pGE646, HoxRStrep; pGE647, HoxTStrep). In both cases, the fusion did not affect the MBH activity (not shown). Transconjugant cells were grown by standard cultivation. Despite the use of an
VOL. 193, 2011
HoxR AND HoxT AID O2 TOLERANCE OF MBH IN R. EUTROPHA
FIG. 7. Copurification experiments with StrepTag II-tagged versions of HoxR and HoxT as baits. HoxRStrep and HoxTStrep were enriched from soluble extracts of MBH-overproducing strains (HF757 and HF758) by Strep-Tactin affinity chromatography. In a control experiment, the same purification procedure was done with extracts from R. eutropha HF632 that encodes wild-type HoxR and HoxT. Eluates (2.5 g of protein) were separated on 10% (for the detection of HoxG) or 15% (for the detection of HoxT, HoxR, HoxK, HoxO, HoxQ, HoxV, HypC, and HypD) SDS-polyacrylamide gels. MBH-related proteins were identified by Western blot analysis with antibodies raised against StrepTag II, HoxK, and HoxG (A) and HoxO, HoxQ, HoxV, HypC, and HypD (B).
overexpression system, HoxR and HoxT proteins could not be detected in cell extracts via immunoblot analysis. However, trace amounts of HoxRStrep and HoxTStrep were enriched from soluble extracts upon Strep-Tactin affinity chromatography. This enabled us to conduct copurification experiments in order to detect proteins that stably interact with HoxR and/or HoxT. Indeed, with HoxRStrep as the bait, preHoxK, HoxG, HoxO, HoxQ, HoxV, HypC, and HypD were copurified (Fig. 7A and B), suggesting that HoxR is an integral component of the maturation complex. For HoxTStrep, the situation was different, since we predominantly found the mature forms of HoxK and HoxG as interaction partners (Fig. 7A). Only weak signals for preHoxK and HoxQ were observed upon immunoblot analysis (Fig. 7B). As a control, the same purification procedure was conducted with cell extracts from R. eutropha HF632 containing wild-type HoxR and HoxT. Immunoblot analysis did not reveal specific signals for any MBH-related protein, underlining that the observations shown in Fig. 7 result from copurification. To obtain further insights into the interaction of HoxT with the MBH protein, we tried to purify the HoxTStrep protein from genetic backgrounds devoid of hoxK (pGE665), hoxG (pGE666), and the endoprotease gene hoxM (pGE667). Remarkably, HoxT could not be enriched from any of these strains (see Fig. S4 in the supplemental material). This suggests that HoxT becomes unstable when the final maturation product, the HoxKG heterodimer, is absent. Analysis of the MBH purified from a ⌬hoxR background. The HoxR protein is homologous to rubredoxin-type Fe-S proteins, suggesting a redox function in the Fe-S cluster synthesis or repair at elevated pO2. Multiple sequence alignments of HoxR orthologs identified the four highly conserved cysteine residues typical for class I rubredoxins (36). Interestingly, HoxR-related proteins from H2-oxidizing bacteria contain a conserved extended N terminus (15 to 20 amino acids), and the two classical CXXCG motifs found in rubredoxins from strict anaerobes are replaced by CXXCW and CXXCD/E motifs (see Fig. S5 in the supplemental material). This feature may
2493
indicate an altered structure/function of the redox-active center in HoxR-like proteins. To collect further information on the function of HoxR, we focused our attention on the characterization of mature MBH from the ⌬hoxR mutant (MBH⌬hoxR; HF851). Strep-Tactin affinity chromatography was used to purify the MBH from the membrane fraction, which previously had been completely oxidized by ferricyanide. Table 2 shows that the lack of hoxR led to a decrease of both yield and specific activity of the mutant MBH, which is consistent with the observations described above. To investigate the redox behavior of MBH⌬hoxR in comparison to that of MBHWT, the proteins were subjected to UVvisible absorption spectroscopy. The oxidized forms of both proteins showed a broad shoulder at around 415 nm (Fig. 8A), which is typical for [3Fe4S]1⫹ and [4Fe4S]2⫹ cluster-containing proteins (40, 47). The lower absorption of MBH⌬hoxR (see also Fig. 8B, thin line) may indicate either slightly diminished iron content and/or a different redox state compared to those of the wild type. Reduction of MBH by incubation of the samples under a 100% H2 atmosphere at room temperature for 15 min yielded spectra with clearly decreased absorption between 300 nm and 800 nm and a weak shoulder shifted to 370 nm (Fig. 8A, dotted lines). The difference spectrum of the H2-reduced samples (Fig. 8B, dotted line) revealed a less pronounced shoulder at 370 nm for MBH⌬hoxR. An extended incubation under H2 did not lead to a further decrease in absorption, indicating that reduction was completed after 15 min. Changes in absorbance upon reduction became more obvious in the oxidized-minus-reduced difference spectra (Fig. 8A, inset). Two major maxima (bleachings) at 350 nm and 430 nm and a minor broad peak at 750 nm were visible in both wild-type and mutant proteins. A clearly diminished bleaching at 350 nm represents a prevailing characteristic of MBH⌬hoxR (see also Fig. 8B, bold line). The difference spectrum in Fig. 8B (bold line) revealed three additional maxima (A422 ⫾ 4, A499 ⫾ 2, and A661 ⫾ 10), clearly indicating a different bleaching behavior of MBH⌬hoxR upon reduction. All spectral features were reproducible in three independent experiments. In conclusion, the UV-visible experiments unambiguously show a distinct redox behavior of MBH⌬hoxR, pointing to an alteration of the Fe-S cluster composition.
TABLE 2. MBH yield and hydrogenase activities of purified MBH protein obtained from the wild-type strain (HF649) and the hoxR deletion strain (HF851)a Strain
MBH yield ⫾ SD (mg/g of membrane)
HF649 (MBHWT) HF851 (MBH⌬hoxR)
1.3 ⫾ 0.3 0.4 ⫾ 0.1
MBH activity in the Activity of purified membrane ⫾ SD MBH ⫾ SD (units/mg (units/mg of of protein) protein) 4.7 ⫾ 0.6 0.8 ⫾ 0.3
70 ⫾ 13 22 ⫾ 4
a Cells were grown heterotrophically under standard conditions. MBH was purified from membranes obtained from MBH-overproducing strains via StrepTactin affinity chromatography. H2-dependent methylene blue reduction activity was measured spectrophotometrically in an anaerobic assay. One unit is defined as 1 mol H2 min⫺1. Values were calculated from results from three independent experiments.
2494
FRITSCH ET AL.
FIG. 8. UV-visible absorption spectroscopy of MBHWT and MBH⌬hoxR. MBH was purified from K3[Fe(CN)6]-oxidized membranes obtained from the MBH-overproducing strains HF649 (MBHWT) and HF851 (MBH⌬hoxR) via affinity chromatography (see Materials and Methods). MBH samples (43 M, A280 ⫽ 7.5, pH 5.5) were measured in the as-isolated, oxidized state (ox) and after reduction (15 min) with 100% H2 (red). (A) Absorption spectra of as-isolated and H2-reduced MBH proteins. The inset shows difference spectra calculated from oxidized-minus-reduced (ox ⫺ red) spectra of MBHWT and MBH⌬hoxR. (B) In order to visualize the differences between MBHWT and MBH⌬hoxR, double difference spectra were prepared. Wavelengths of relevant absorption maxima or bleachings are marked with a triangle.
DISCUSSION Maturation of the MBH from R. eutropha is a complex process that depends on the products of at least six pleiotropic hyp genes (12, 30) and seven MBH-specific genes arranged in the accessory gene cluster hoxMLOQRTV (10, 42, 61). In this study, we have presented new insights into the functions of hoxR and hoxT. Mutant studies revealed that HoxR and HoxT are crucial for H2-driven growth and synthesis of active MBH in the presence of elevated pO2. Copurification experiments in
J. BACTERIOL.
combination with immunologic analyses showed that HoxR and HoxT fulfil individual functions at sequential stages of MBH maturation (Fig. 1). Hydrogenase accessory gene clusters similar to that of the MBH are present in N2-fixing bacteria, including Rhizobium leguminosarum (73). Using an artificial expression system enabling hydrogenase synthesis under microoxic conditions, the accessory genes hupGHIJ (homologous to hoxOQRT) were shown to have an impact on hydrogenase activity. The analysis of mutants revealed that hydrogenase activity was severely reduced under microaerobic free-living conditions and less affected under nitrogen fixation conditions in bacteroids. From these results, it was proposed that the hupGHIJ gene products play a role in the maturation of the small hydrogenase subunit (46). Proposed role of HoxR in Fe-S cluster biogenesis. The amino acid composition of HoxR indicates that the protein belongs to the family of rubredoxins, which are small (⬃6 kDa), redox-active iron-sulfur proteins found mainly in microaerophilic and anaerobic prokaryotes (11, 23, 77). The redox-active site consists of a central iron coordinated by four cysteines in a tetrahedral arrangement and switches between oxidation states of ⫹2 and ⫹3 at a midpoint potential around 0 mV (16). The exact role of most rubredoxins remains elusive, but some of them were shown to be involved in electron transfer reactions (27, 53, 66) and oxidative stress response in anaerobes (23, 55, 77). In contrast to common rubredoxins that contain two CXXCG sites, we identified a conserved CXXCW and CXXCD/E motif in the HoxR-related proteins (see Fig. S5 in the supplemental material). Exchanges of the conserved glycine residues by site-directed mutagenesis significantly altered the redox potential (2) but caused only minor shifts in the overall protein structure (45) of a rubredoxin from Clostridium pasteurianum. Therefore, it is likely that HoxR-related proteins possess a redox potential that is different from that of canonical rubredoxins. This property might correlate with the oxidizing conditions at which the HoxR proteins are active. In the cyanobacterium Synechococcus sp. PCC 7002, deletion of rubA, a gene encoding a rubredoxin with a single Cterminal transmembrane domain, resulted in specific loss of the Fe-S cluster Fx of photosystem I (68, 69). However, the exact role of RubA in Fx cluster assembly is not clear. It has been proposed to act as an electron shunt to prevent overreduction of the transient cluster during biogenesis. Provided that the HoxR protein has a redox function, it is possibly involved in synthesis and/or repair of the Fe-S cluster of MBH at high pO2. In fact, the analysis of the MBH isolated from the ⌬hoxR background presents the first experimental evidence that HoxR is necessary for the establishment of a specific Fe-S cluster profile in HoxK at high pO2. On the basis of protein sequence comparison and EPR spectroscopy, it has been proposed that HoxK contains a [4Fe4S] cluster distal to the Ni-Fe active site, a [3Fe4S] cluster in the medial position, and an unusually high potential Fe-S cluster proximal to the active site (32, 38, 56, 57, 60). Interestingly, the proximal cluster is surrounded by six instead of four coordinating cysteines (57). The UV-visible absorption spectra presented here are in agreement with the presence of [3Fe4S] and [4Fe4S] clusters in the MBH. Upon reduction, the broad major
VOL. 193, 2011
HoxR AND HoxT AID O2 TOLERANCE OF MBH IN R. EUTROPHA
bleaching around 430 nm (inset in Fig. 8A) could result from an overlap of [4Fe4S]2⫹/1⫹ and [3Fe4S]1⫹/0 redox transitions (see Table S2 in the supplemental material) (26, 40, 47). The significant decrease of A350, A750 (inset in Fig. 8A), and the shoulder at 370 nm observed in reduced MBHWT samples (Fig. 8A) reflect to our knowledge novel features that may be attributed to the unusual proximal Fe-S cluster. Based on this assumption, the less pronounced shoulder at 370 nm (Fig. 8B, dotted line) and the diminished bleaching upon reduction at 350 nm (inset in Fig. 8A, bold line) observed for MBH⌬hoxR suggest a specific role of HoxR in biogenesis of the proximal cluster under aerobic conditions. The copurification experiments imply that HoxR acts on the level of the HoxKOQ complex (61) (Fig. 1). This notion gains further support by a translational fusion (hyaF2) of hoxQ and hoxR that has been annotated in the hydrogenase-1 (Hyd-1) gene cluster of Salmonella species (see Fig. S5 in the supplemental material). The fact that HoxR was found only in substoichiometric amounts in MBH protein complexes supports a catalytic role rather than a protective role in MBH maturation. It is interesting to note that the occurrence of HoxR correlates with membrane-bound hydrogenases whose proximal Fe-S cluster is surrounded by six instead of the conventional four cysteines. Only a few organisms which carry MBH-like hydrogenases with a modified Fe-S cluster, such as Hyd-1 from E. coli (43) and MBH-1 from Aquifex aeolicus (14), are devoid of hoxR. These bacteria, however, produce hydrogenase only under microaerobic or anaerobic conditions (19, 24), under which HoxR seems to be dispensable. Proposed role of HoxT in the export of MBH. Of the three hydrogenases that are well characterized in E. coli, Hyd-1 shares the highest similarity with the MBH. However, in contrast to the complex MBH gene cluster in R. eutropha, the Hyd-1 structural genes are associated only with three accessory genes, hyaD, hyaE, and hyaF (21). These are homologous to hoxM, hoxO, and hoxQ, respectively, and hoxR and hoxT are missing in this context. HybE, a HoxT homologue with relatively low sequence identity (27%), is encoded in the operon of the O2-sensitive hydrogenase-2 (Hyd-2) of E. coli. In a bacterial two-hybrid study, HybE was shown to interact with the premature forms of Hyd-2 subunits (21). The ⌬hybE mutant strain contained very low Hyd-2 activity, and premature as well as mature large subunits of Hyd-2 accumulated in the cytoplasm. Interestingly, the processed small subunit was found in the membrane fraction, but the large subunit was missing (29). The authors suggested a function of the HybE protein in quality control and direct interaction with the Tat leader peptide to prevent wasteful export of immature inactive hydrogenase. This hypothesis gained support by the discovery of a conserved protein fold suitable for binding of the Tat signal peptide in the NMR structure of HybE (67). On the basis of the observations with Hyd-2 from E. coli, copurification of the HoxT protein from R. eutropha with leader peptide-free soluble small MBH subunit was therefore unexpected since the mature form of MBH is usually found exclusively in the membrane fraction (61). This result points to a role of HoxT in the translocation of the MBH heterodimer through the Tat secretion system (Fig. 1). Furthermore, we have evidence that the HoxT protein is destabilized in cells lacking the MBH structural genes or the large-subunit-specific
2495
endopeptidase gene. This indicates that interaction with export-competent MBH may stabilize the HoxT protein. A role of HoxT in Tat-mediated secretion is further deduced from its membrane association and the presence of a conserved C-terminal twin-arginine motif (see Fig. S6 in the supplemental material). Due to the lack of a lysine residue at position ⫹5⬘ relative to the first invariant arginine, which occurs in canonical Tat signal sequences (6), the motif found in HoxT proteins shares similarities with the N-terminal motif in signal peptides of Rieske proteins (3). We consider that the twin-arginine motif in HoxT might interact with the Tat translocon (1, 7) to guide or accelerate export of mature MBH. This enhancement might protect an oxygen-sensitive state of maturation prior to attachment of the MBH with the cytochrome HoxZ (8). Previously, it was shown by electron paramagnetic resonance (EPR) and Fourier transform infrared spectroscopy (FTIR) (57) that membrane-associated HoxZcoupled MBH is well protected against O2, whereas solubilized heterodimeric MBH is damaged under oxidative conditions. Activity staining of MBH from extracts of hoxT mutant cells revealed a similar activity pattern as that derived from hoxZ mutant cells (10). This result suggests that HoxT affects attachment of the MBH heterodimer to the cytochrome HoxZ by preventing oxidative damage of the distal [4Fe4S] cluster in HoxK. This Fe-S cluster is supposed to be located close to the protein surface (48, 49, 76) and is therefore likely prone to oxidative damage (28). Notably, the hupJ gene in the hydrogenase gene clusters of several aerobic hydrogen oxidizers, including Rhodobacter capsulatus, Azorhizobium caulinodans, and Paracoccus denitrificans, encodes an HoxR-HoxT protein fusion (4, 17), pointing to a concerted action of the proteins HoxR and HoxT. Concluding remarks. The results of this study strongly support the model that HoxR and HoxT function at sequential steps in the maturation of the MBH from R. eutropha, thereby contributing to O2 tolerance of the process. HoxR presumably acts on the proximal Fe-S cluster in a large maturation complex, comprising the MBH subunit precursors, a number of MBH-specific maturases, and metal center-assembling Hyp proteins. HoxT might come into play upon resolution of the maturation complex, protecting the distal Fe-S cluster and thereby facilitating association of the MBH with the membrane. Detailed spectroscopic analyses of the mutant proteins will be a key issue of future studies, providing further insights into an O2-tolerant hydrogenase system. ACKNOWLEDGMENTS This work was supported by the EU/Energy Network project FP7 SOLAR-H2 Program (contract no. 212508) and the DFG Cluster of Excellence “Unifying Concepts in Catalysis.” We thank Torsten Schubert for preliminary experiments with hoxR and hoxT mutant strains, Josta Hamann, Janna Schoknecht, and Angelika Strack for excellent technical assistance, and John Golbeck for valuable discussions. Anne Pohlmann is acknowledged for analyzing the real-time RT-PCR data. REFERENCES 1. Alami, M., et al. 2003. Differential interactions between a twin-arginine signal peptide and its translocase in Escherichia coli. Mol. Cell 12:937–946. 2. Ayhan, M., et al. 1996. The rubredoxin from Clostridium pasteurianum: mutation of the conserved glycine residues 10 and 43 to alanine and valine. Inorg. Chem. 35:5902–5911.
2496
FRITSCH ET AL.
3. Bachmann, J., B. Bauer, K. Zwicker, B. Ludwig, and O. Anderka. 2006. The Rieske protein from Paracoccus denitrificans is inserted into the cytoplasmic membrane by the twin-arginine translocase. FEBS J. 273:4817–4830. 4. Baginsky, C., J. M. Palacios, J. Imperial, T. Ruiz-Argueso, and B. Brito. 2004. Molecular and functional characterization of the Azorhizobium caulinodans ORS571 hydrogenase gene cluster. FEMS Microbiol. Lett. 237: 399–405. 5. Bendtsen, J. D., H. Nielsen, G. von Heijne, and S. Brunak. 2004. Improved prediction of signal peptides: SignalP 3.0. J. Mol. Biol. 340:783–795. 6. Berks, B. C. 1996. A common export pathway for proteins binding complex redox cofactors? Mol. Microbiol. 22:393–404. 7. Berks, B. C., T. Palmer, and F. Sargent. 2005. Protein targeting by the bacterial twin-arginine translocation (Tat) pathway. Curr. Opin. Microbiol. 8:174–181. 8. Bernhard, M., B. Benelli, A. Hochkoeppler, D. Zannoni, and B. Friedrich. 1997. Functional and structural role of the cytochrome b subunit of the membrane-bound hydrogenase complex of Alcaligenes eutrophus H16. Eur. J. Biochem. 248:179–186. 9. Bernhard, M., B. Friedrich, and R. A. Siddiqui. 2000. Ralstonia eutropha TF93 is blocked in Tat-mediated protein export. J. Bacteriol. 182:581–588. 10. Bernhard, M., E. Schwartz, J. Rietdorf, and B. Friedrich. 1996. The Alcaligenes eutrophus membrane-bound hydrogenase gene locus encodes functions involved in maturation and electron transport coupling. J. Bacteriol. 178:4522–4529. 11. Blake, P. R., et al. 1991. Determinants of protein hyperthermostability: purification and amino acid sequence of rubredoxin from the hyperthermophilic archaebacterium Pyrococcus furiosus and secondary structure of the zinc adduct by NMR. Biochemistry 30:10885–10895. 12. Bo ¨ck, A., P. W. King, M. Blokesch, and M. C. Posewitz. 2006. Maturation of hydrogenases. Adv. Microb. Physiol. 51:1–71. 13. Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248–254. 14. Brugna-Guiral, M., et al. 2003. [NiFe] hydrogenases from the hyperthermophilic bacterium Aquifex aeolicus: properties, function, and phylogenetics. Extremophiles 7:145–157. 15. Burgdorf, T., et al. 2005. [NiFe]-hydrogenases of Ralstonia eutropha H16: modular enzymes for oxygen-tolerant biological hydrogen oxidation. J. Mol. Microbiol. Biotechnol. 10:181–196. 16. Capozzi, F., S. Ciurli, and C. Luchinat. 1998. Coordination sphere versus protein environment as determinants of electronic and functional properties of iron-sulfur proteins. Metal Sites Proteins Models 90:127–160. 17. Colbeau, A., et al. 1993. Organization of the genes necessary for hydrogenase expression in Rhodobacter capsulatus. Sequence analysis and identification of two hyp regulatory mutants. Mol. Microbiol. 8:15–29. 18. Cracknell, J. A., A. F. Wait, O. Lenz, B. Friedrich, and F. A. Armstrong. 2009. A kinetic and thermodynamic understanding of O2 tolerance in [NiFe]hydrogenases. Proc. Natl. Acad. Sci. U. S. A. 106:20681–20686. 19. Deckert, G., et al. 1998. The complete genome of the hyperthermophilic bacterium Aquifex aeolicus. Nature 392:353–358. 20. De Lacey, A. L., V. M. Fernandez, M. Rousset, and R. Cammack. 2007. Activation and inactivation of hydrogenase function and the catalytic cycle: spectroelectrochemical studies. Chem. Rev. 107:4304–4330. 21. Dubini, A., and F. Sargent. 2003. Assembly of Tat-dependent [NiFe] hydrogenases: identification of precursor-binding accessory proteins. FEBS Lett. 549:141–146. 22. Esper, B., A. Badura, and M. Ro ¨gner. 2006. Photosynthesis as a power supply for (bio-)hydrogen production. Trends Plant Sci. 11:543–549. 23. Fareleira, P., et al. 2003. Response of a strict anaerobe to oxygen: survival strategies in Desulfovibrio gigas. Microbiology 149:1513–1522. 24. Forzi, L., and R. G. Sawers. 2007. Maturation of [NiFe]-hydrogenases in Escherichia coli. Biometals 20:565–578. 25. Fujiki, Y., A. L. Hubbard, S. Fowler, and P. B. Lazarow. 1982. Isolation of intracellular membranes by means of sodium-carbonate treatment—application to endoplasmic reticulum. J. Cell Biol. 93:97–102. 26. Green, J., et al. 1996. Reconstitution of the [4Fe-4S] cluster in FNR and demonstration of the aerobic-anaerobic transcription switch in vitro. Biochem. J. 316(Pt. 3):887–892. 27. Hagelueken, G., et al. 2007. Crystal structure of the electron transfer complex rubredoxin rubredoxin reductase of Pseudomonas aeruginosa. Proc. Natl. Acad. Sci. U. S. A. 104:12276–12281. 28. Imlay, J. A. 2006. Iron-sulphur clusters and the problem with oxygen. Mol. Microbiol. 59:1073–1082. 29. Jack, R. L., et al. 2004. Coordinating assembly and export of complex bacterial proteins. EMBO J. 23:3962–3972. 30. Jones, A. K., O. Lenz, A. Strack, T. Buhrke, and B. Friedrich. 2004. NiFe hydrogenase active site biosynthesis: identification of Hyp protein complexes in Ralstonia eutropha. Biochemistry 43:13467–13477. 31. Juretic, D., L. Zoranic, and D. Zucic. 2002. Basic charge clusters and predictions of membrane protein topology. J. Chem. Infect. Comput. Sci. 42: 620–632. 32. Knu ¨ttel, K., et al. 1994. Redox properties of the metal centers in the mem-
J. BACTERIOL.
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.
brane-bound hydrogenase from Alcaligenes eutrophus CH34. Bull. Polish Acad. Sci. 42:495–511. Kortlu ¨ke, C., and B. Friedrich. 1992. Maturation of membrane-bound hydrogenase of Alcaligenes eutrophus H16. J. Bacteriol. 174:6290–6293. Krassen, H., et al. 2009. Photosynthetic hydrogen production by a hybrid complex of photosystem I and [NiFe]-hydrogenase. ACS Nano. 3:4055–4061. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685. LeGall, J., et al. 1998. Characterisation of a new rubredoxin isolated from Desulfovibrio desulfuricans 27774: definition of a new family of rubredoxins. FEBS Lett. 429:295–298. Lenz, O., A. Gleiche, A. Strack, and B. Friedrich. 2005. Requirements for heterologous production of a complex metalloenzyme: the membrane-bound [NiFe] hydrogenase. J. Bacteriol. 187:6590–6595. Lenz, O., et al. 2010. H2 conversion in the presence of O2 as performed by the membrane-bound [NiFe]-hydrogenase of Ralstonia eutropha. Chemphyschem 11:1107–1119. Lenz, O., E. Schwartz, J. Dernedde, M. Eitinger, and B. Friedrich. 1994. The Alcaligenes eutrophus H16 hoxX gene participates in hydrogenase regulation. J. Bacteriol. 176:4385–4393. Lippard, S. J., and J. M. Berg. 1995. Bioanorganische Chemie, p. 124–135. Spektrum Akademischer Verlag, Heidelberg, Germany. Ludwig, M., J. A. Cracknell, K. A. Vincent, F. A. Armstrong, and O. Lenz. 2009. Oxygen-tolerant H2 oxidation by membrane-bound [NiFe] hydrogenases of Ralstonia species. Coping with low level H2 in air. J. Biol. Chem. 284:465–477. Ludwig, M., et al. 2009. Concerted action of two novel auxiliary proteins in assembly of the active site in a membrane-bound [NiFe] hydrogenase. J. Biol. Chem. 284:2159–2168. Lukey, M. J., et al. 2010. How Escherichia coli is equipped to oxidize hydrogen under different redox conditions. J. Biol. Chem. 285:3928–3938. Magalon, A., M. Blokesch, E. Zehelein, and A. Bo ¨ck. 2001. Fidelity of metal insertion into hydrogenases. FEBS Lett. 499:73–76. Maher, M. J., Z. Xiao, M. C. Wilce, J. M. Guss, and A. G. Wedd. 1999. Rubredoxin from Clostridium pasteurianum. Structures of G10A, G43A and G10VG43A mutant proteins. Mutation of conserved glycine 10 to valine causes the 9-10 peptide link to invert. Acta Crystallogr. D Biol. Crystallogr. 55:962–968. Manyani, H., L. Rey, J. M. Palacios, J. Imperial, and T. Ruiz-Argu ¨eso. 2005. Gene products of the hupGHIJ operon are involved in maturation of the iron-sulfur subunit of the [NiFe] hydrogenase from Rhizobium leguminosarum bv. viciae. J. Bacteriol. 187:7018–7026. Nakamaru-Ogiso, E., T. Yano, T. Yagi, and T. Ohnishi. 2005. Characterization of the iron-sulfur cluster N7 (N1c) in the subunit NuoG of the protontranslocating NADH-quinone oxidoreductase from Escherichia coli. J. Biol. Chem. 280:301–307. Ogata, H., et al. 2005. Activation process of [NiFe] hydrogenase elucidated by high-resolution X-ray analyses: conversion of the ready to the unready state. Structure 13:1635–1642. Ogata, H., P. Kellers, and W. Lubitz. 2010. The crystal structure of the [NiFe] hydrogenase from the photosynthetic bacterium Allochromatium vinosum: characterization of the oxidized enzyme (Ni-A state). J. Mol. Biol. 402:428–444. Pierik, A. J., W. Roseboom, R. P. Happe, K. A. Bagley, and S. P. Albracht. 1999. Carbon monoxide and cyanide as intrinsic ligands to iron in the active site of [NiFe]-hydrogenases. NiFe(CN)2CO, biology’s way to activate H2. J. Biol. Chem. 274:3331–3337. Priefert, H., et al. 1991. Identification and molecular characterization of the Alcaligenes eutrophus H16 aco operon genes involved in acetoin catabolism. J. Bacteriol. 173:4056–4071. Py, B., and F. Barras. 2010. Building Fe-S proteins: bacterial strategies. Nat. Rev. Microbiol. 8:436–446. Ragsdale, S. W., L. G. Ljungdahl, and D. V. DerVartanian. 1983. Isolation of carbon monoxide dehydrogenase from Acetobacterium woodii and comparison of its properties with those of the Clostridium thermoaceticum enzyme. J. Bacteriol. 155:1224–1237. Rodrigue, A., A. Chanal, K. Beck, M. Mu ¨ller, and L. F. Wu. 1999. Cotranslocation of a periplasmic enzyme complex by a hitchhiker mechanism through the bacterial Tat pathway. J. Biol. Chem. 274:13223–13228. Rodrigues, J. V., I. A. Abreu, L. M. Saraiva, and M. Teixeira. 2005. Rubredoxin acts as an electron donor for neelaredoxin in Archaeoglobus fulgidus. Biochem. Biophys. Res. Commun. 329:1300–1305. Saggu, M., et al. 2010. Comparison of the membrane-bound [NiFe] hydrogenases from R. eutropha H16 and D. vulgaris Miyazaki F in the oxidized ready state by pulsed EPR. Phys. Chem. Chem. Phys. 12:2139–2148. Saggu, M., et al. 2009. Spectroscopic insights into the oxygen-tolerant membrane-associated [NiFe] hydrogenase of Ralstonia eutropha H16. J. Biol. Chem. 284:16264–16276. Sargent, F. 2007. The twin-arginine transport system: moving folded proteins across membranes. Biochem. Soc. Trans. 35:835–847. Schink, B., and H. G. Schlegel. 1979. The membrane-bound hydrogenase of
VOL. 193, 2011
60.
61.
62.
63. 64.
65.
66.
67. 68.
69.
HoxR AND HoxT AID O2 TOLERANCE OF MBH IN R. EUTROPHA
Alcaligenes eutrophus. I. Solubilization, purification, and biochemical properties. Biochim. Biophys. Acta 567:315–324. Schneider, K., D. S. Patil, and R. Cammack. 1983. Electron-spin-resonance properties of membrane-bound hydrogenases from aerobic hydrogen bacteria. Biochim. Biophys. Acta 748:353–361. Schubert, T., O. Lenz, E. Krause, R. Volkmer, and B. Friedrich. 2007. Chaperones specific for the membrane-bound [NiFe]-hydrogenase interact with the Tat signal peptide of the small subunit precursor in Ralstonia eutropha H16. Mol. Microbiol. 66:453–467. Schwartz, E., and B. Friedrich. 2006. The H2-metabolizing prokaryotes, p. 496–563. In M. Dworkin, S. Falkow, E. Rosenberg, K.-H. Schleifer, and E. Stackebrandt (ed.), The prokaryotes. Springer, New York, NY. Schwartz, E., U. Gerischer, and B. Friedrich. 1998. Transcriptional regulation of Alcaligenes eutrophus hydrogenase genes. J. Bacteriol. 180:3197–3204. Schwartz, E., et al. 2003. Complete nucleotide sequence of pHG1: a Ralstonia eutropha H16 megaplasmid encoding key enzymes of H2-based lithoautotrophy and anaerobiosis. J. Mol. Biol. 332:369–383. Schwarze, A., M. J. Kopczak, M. Ro ¨gner, and O. Lenz. 2010. Requirements for construction of a functional hybrid complex of photosystem I and [NiFe]hydrogenase. Appl. Environ. Microbiol. 76:2641–2651. Seki, S., A. Ikeda, and M. Ishimoto. 1988. Rubredoxin as an intermediary electron carrier for nitrate reduction by NAD(P)H in Clostridium perfringens. J. Biochem. 103:583–584. Shao, X., J. Lu, Y. Hu, B. Xia, and C. Jin. 2009. Solution structure of the Escherichia coli HybE reveals a novel fold. Proteins 75:1051–1056. Shen, G., et al. 2002. Assembly of photosystem I. II. Rubredoxin is required for the in vivo assembly of F(X) in Synechococcus sp. PCC 7002 as shown by optical and EPR spectroscopy. J. Biol. Chem. 277:20355–20366. Shen, G., et al. 2002. Assembly of photosystem I. I. Inactivation of the rubA gene encoding a membrane-associated rubredoxin in the cyanobacterium Synechococcus sp. PCC 7002 causes a loss of photosystem I activity. J. Biol. Chem. 277:20343–20354.
2497
70. Simon, R., U. Priefer, and A. Pu ¨hler. 1983. A broad host range mobilization system for in vivo genetic engineering—transposon mutagenesis in Gramnegative bacteria. Nat. Biotechnol. 1:784–791. 71. Sonnhammer, E. L., G. von Heijne, and A. Krogh. 1998. A hidden Markov model for predicting transmembrane helices in protein sequences. Proc. Int. Conf. Intell. Syst. Mol. Biol. 6:175–182. 72. Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. U. S. A. 76:4350–4354. 73. Vignais, P. M., B. Billoud, and J. Meyer. 2001. Classification and phylogeny of hydrogenases. FEMS Microbiol. Rev. 25:455–501. 74. Vincent, K. A., et al. 2005. Electrocatalytic hydrogen oxidation by an enzyme at high carbon monoxide or oxygen levels. Proc. Natl. Acad. Sci. U. S. A. 102:16951–16954. 75. Vincent, K. A., A. Parkin, and F. A. Armstrong. 2007. Investigating and exploiting the electrocatalytic properties of hydrogenases. Chem. Rev. 107: 4366–4413. 76. Volbeda, A., et al. 1995. Crystal structure of the nickel-iron hydrogenase from Desulfovibrio gigas. Nature 373:580–587. 77. Wildschut, J. D., R. M. Lang, J. K. Voordouw, and G. Voordouw. 2006. Rubredoxin:oxygen oxidoreductase enhances survival of Desulfovibrio vulgaris Hildenborough under microaerophilic conditions. J. Bacteriol. 188: 6253–6260. 78. Wilkins, M. R., et al. 1999. Protein identification and analysis tools in the ExPASy server. Methods Mol. Biol. 112:531–552. 79. Wilson, C. M. 1983. Staining of proteins on gels: comparisons of dyes and procedures. Methods Enzymol. 91:236–247. 80. Yanisch-Perron, C., J. Vieira, and J. Messing. 1985. Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33:103–119.
