Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Microbiology. Author manuscript; available in PMC Aug 3, 2009.
Published in final edited form as:
PMCID: PMC2720019

Pathway for H2O2 and O2 detoxification in Clostridium acetobutylicum


An unusual non-haem diiron protein, reverse rubrerythrin (revRbr), is known to be massively upregulated in response to oxidative stress in the strictly anaerobic bacterium Clostridium acetobutylicum. In the present study both in vivo and in vitro results demonstrate an H2O2 and O2 detoxification pathway in C. acetobutylicum involving revRbr, rubredoxin (Rd) and NADH: rubredoxin oxidoreductase (NROR). RevRbr exhibited both NADH peroxidase (NADH: H2O2 oxidoreductase) and NADH oxidase (NADH: O2 oxidoreductase) activities in in vitro assays using NROR as the electron-transfer intermediary from NADH to revRbr. Rd increased the NADH consumption rate by serving as an intermediary electron-transfer shuttle between NROR and revRbr. While H2O2 was found to be the preferred substrate for revRbr, its relative oxidase activity was found to be significantly higher than that reported for other Rbrs. A revRbr-overexpressing strain of C. acetobutylicum showed significantly increased tolerance to H2O2 and O2 exposure. RevRbr thus appears to protect C. acetobutylicum against oxidative stress by functioning as the terminal component of an NADH peroxidase and NADH oxidase.


Reactive oxygen species (ROS) are formed when molecular oxygen diffuses into cells and is adventitiously reduced at the active sites of redox enzymes containing flavins or quinones (Valentine et al., 1998; Imlay, 2003). Aerobic micro-organisms scavenge these toxic ROS via the superoxide dismutase (SOD)/catalase system. In anaerobes this system sometimes seems to be absent or incomplete (Imlay, 2002; Dolla et al., 2006). Thus, in the genome of the strict anaerobe Clostridium acetobutylicum, no catalase was annotated (Nölling et al., 2001), and a function has not yet been demonstrated for the annotated SODs. Regeneration of molecular oxygen by the SOD/catalase system may be a disadvantage for anaerobes. Nevertheless, C. acetobutylicum survives short periods of aeration (O’Brien & Morris, 1971), and therefore an alternative detoxification system must exist. Interestingly, all genes for a superoxide reductase (SOR) and peroxidase-dependent detoxification pathway for ROS proposed for Desulfovibrio vulgaris (Lumppio et al., 2001) are also present in clostridial species. The purified C. acetobutylicum SOR homologue, desulfoferrodoxin (Dfx), was shown to be capable of functioning in superoxide reduction (Riebe et al., 2007). In other bacteria and archaea, rubrerythrin (Rbr) has been assigned a role as a peroxidase (NAD(P)H: H2O2 oxidoreductase) in this alternative pathway (Coulter et al., 1999; Weinberg et al., 2004; Kurtz, 2006). A Porphyromonas gingivalis rbr mutant showed increased sensitivity to ROS (Sztukowska et al., 2002) and reduced virulence in mice (Mydel et al., 2006). A putative SOD activity attributed to Clostridium perfringens Rbr (Lehmann et al., 1996) may be an artefact (Coulter et al., 1999; Fournier et al., 2003), and, in any case, is much lower than those of classical SODs.

Rbrs are characterized by a unique combination of a non-sulfur, oxo-bridged diiron site similar to that of ferritin and a rubredoxin-like [Fe(SCys)4] domain (Kurtz, 2006). The evidence indicates that during Rbr’s peroxidase function, the reduced diiron site reacts directly with hydrogen peroxide, while the [Fe(SCys)4] site transfers electrons from an exogenous donor to the oxidized diiron site. The reactivity of Rbrs with dioxygen is typically low relative to that with hydrogen peroxide.

Interestingly, the C. acetobutylicum genome encodes four proteins homologous to Rbrs. The arrangement of the domains in two of these Rbrs is the same as that of most other known ‘classical’ Rbrs, with an N-terminal diiron domain and a C-terminal [Fe(SCys)4] domain. However, in the other two C. acetobutylicum Rbrs, the positions of these domains are reversed. These latter have, therefore, been named ‘reverse’ Rbrs (revRbrs) (May et al., 2004). The 99 % identical open reading frames of the two C. acetobutylicum revRbrs, cac3597 (rbr3B) and cac3598 (rbr3A), are in adjacent positions in the genome and form an operon. In previous studies, we demonstrated an increased expression of the revRbrs during oxidative or heat stress (May et al., 2004). Kawasaki et al. (2007) showed that recombinant, purified, reduced revRbr could be oxidized by hydrogen peroxide and that cell extracts from microxically grown C. acetobutylicum could catalyse reduction of as-isolated (partially oxidized) revRbr by NAD(P)H. However, they did not identify the NAD(P)H: revRbr oxidoreductase, nor did they report any catalytic activities of revRbr.

We now present new in vitro and in vivo data demonstrating that revRbr functions as the terminal component of an NADH peroxidase (NADH: H2O2 oxidoreductase) and an NADH oxidase (NADH: O2 oxidoreductase) in C. acetobutylicum. We also compare these revRbr activities with those of one of the C. acetobutylicum Rbrs, RubY, which has the ‘classical’ Rbr domain structure, and which is apparently not upregulated upon oxidative stress. These findings demonstrate a key role for revRbr in oxidative stress tolerance of C. acetobutylicum, including the aerotolerance of a recently described ΔperR mutant (Hillmann et al., 2008).


Reagents and proteins

Reagents and buffers were at least analytical grade. Ferredoxin-NADP+ reductase (FNR) from spinach, horse-radish peroxidase (HRP), glucose oxidase and E. coli iron superoxide dismutase (Fe-SOD) were purchased from Sigma. Concentrations were assumed to be those provided by Sigma. 10-Acetyl-3,7-dihydroxyphenoxazine (Amplex Red) was from Synchem OHG. Reagents were used without further purification.

Construction of Rd, NROR, RubY and revRbr expression plasmids and transformations

Clostridium acetobutylicum ATCC 824 genomic DNA was isolated according to Bertram & Dürre (1989). C. acetobutylicum Rd was cloned and purified as described in our previous work (Riebe et al., 2007). The rbr3B (cac3597, encoding revRbr), rubY (cac2575) and nror (cac2448) genes were amplified using the following oligonucleotides: Rbr_Fw_BamHI (5′-GAAAGGATCCATGAAAAAATTTAAATGTGTTG-3′), Rbr_Rev_XmaI (5′-TTATCCCGGGTTTGAAATATCTGTTTAATAAAC-3′), RubY_Fw_ BamHI (5′-GAGGGGATCCATGAAATCACTTAAAGG-3′), RubY_ Rev_XmaI (5′-CTTATCCCGGGATAGTTTTCACTTAATATTTC-3′), NROR_Fw_BamHI (5′-GATGGGATCCATGAAAAGCACAAAAATTTT-3′) and NROR_Rev_XmaI (5′-TAAATTCCCGGGTAAATTATTTAATATTGCATC-3′), introducing a BamHI or a XmaI restriction site. The rbr3B, rubY and nror fragments were cloned into a pThydA vector (Girbal et al., 2005). The resulting plasmids pTrbr3B, pTrubY and pTnror were transformed into Escherichia coli DH5α. These strains were used for overexpression of the respective proteins.

These same plasmids could also be used for overexpression of the proteins in C. acetobutylicum. For transformation in C. acetobutylicum, plasmids were reisolated and subcloned into a methylation strain, E. coli ER2275 (Bermejo et al., 1998)/pANI (Mermelstein & Papoutsakis, 1993). The methylated plasmids were electroporated into C. acetobutylicum ATCC 824 as follows. Cultures (100 ml) were grown in clostridial growth medium (CGM, Roos et al., 1985) at 37 °C to OD600 0.8–1.0 in anaerobic serum flasks and then harvested with SS34 tubes in an anaerobic chamber. Cells were centrifuged at 5000 g for 1 min and then washed once with 10 ml electroporation buffer containing 270 mM sucrose and 200 mM phosphate at pH 7.4. After a second centrifugation step the pellet was suspended in 2 ml prechilled electroporation buffer. One microgram of DNA was used for anaerobic electroporation of 400 μl competent cells in a 4 mm electroporation cuvette. Electroporation conditions were 2 kV, 50 μF, 600 Ω. After electroporation, 600 μl fresh medium was added and cells were regenerated for 3 h at 37 °C. Then 200 μl aliquots were plated on reinforced clostridium agar (RCA) plates supplemented with erythromycin (20 mg l−1).

Protein expression and purification

The thiolase (thl) promoter in the pThydA vector is active in both Gram-positive and Gram-negative organisms. Proteins could, therefore, be expressed in either E. coli DH5α or C. acetobutylicum strains using the same vector constructs.

E. coli DH5α overexpression cultures were grown aerobically in 500 ml Luria–Bertani broth supplemented with 100 μg ampicillin ml−1 at 30 °C under continuous shaking for 18 h. A higher level of iron incorporation into revRbr and Rd was achieved by addition of 40 μg FeSO4.7H2O ml −1 to the cultures (Coulter & Kurtz, 2001). Cultures were harvested at 8000 g for 10 min and pellets were suspended in 5 ml buffer W (100 mM Tris/HCl, 150 mM NaCl, pH 8.0) without EDTA. Disruption of the cells was achieved by ultrasonication under anaerobic conditions in a MAC 5000 anaerobic chamber. The lysates were centrifuged for at least 30 min at 12 000 g at 4 °C in anaerobic SS34 tubes. The resulting crude extract was loaded onto columns containing a Strep-Tactin-Sepharose matrix. Further purification steps were carried out as described in the IBA standard protocol under anaerobic conditions (Schmidt & Skerra, 2007). Protein concentrations were determined using the Bradford assay (Bradford, 1976). Purity of the elution fractions was analysed by SDS-PAGE (Laemmli, 1970).

Analytical gel filtration

Analytical gel filtration was performed using a Superose 12 10/300 GL (GE Healthcare) by loading 100 μl revRbr (10 mg ml−1) and developed using a 0.5 ml min−1 isocratic flow of buffer containing 50 mM MOPS and 250 mM NaCl pH 7.3. Molecular mass was estimated by comparing retention time to that of known standards in the LMW Gel Filtration Calibration kit (GE Healthcare).

Response to oxidative stress

All experiments were carried out at 37 °C. Ten-millilitre cultures of either wild-type or pTrbr3B-transformed (revRbr-overexpressing) C. acetobutylicum were grown anaerobically to OD600 0.8–1.0 in CGM. Medium for the revRbr overexpression strain was supplemented with 20 mg erythromycin l−1. Aliquots (1 ml) of each culture were treated anaerobically with H2O2 (at concentrations of 0.125, 0.25, 0.375, 0.5, 0.625, 0.75, 0.875 or 1.0 mM) for 30 min, then plated. Another set of 1 ml aliquots were flushed with air for 15 min, then returned to the anaerobic chamber and plated after an additional 15 min. A further 1 ml aliquot of each culture was treated anaerobically only with 50 μl 50 mM Tris/HCl pH 7.5. Aliquots (25 μl) of the treated 1 ml cultures were plated on RCA containing 2.5 μg erythromycin ml−1 for the revRbr overexpression strain or no antibiotic for the wild-type strain. The plates were incubated for 2–4 days at 37 °C and c.f.u. were then counted. C.f.u. from plates of the Tris/HCl buffer-treated cultures were taken as the 100 % survival reference.

Redox-dependent spectral changes of revRbr

Reduction of oxidized (as-isolated) revRbr by a mixture of NROR, Rd and NAD(P)H was monitored spectrophotometrically. Conditions are listed in the figure legends.

For oxidation of reduced revRbr, the purified revRbr was reduced by an equimolar amount of sodium dithionite under anaerobic conditions. This pretreated revRbr (0.1 mM) was then incubated in anaerobic 50 mM MOPS and 0.1 mM EDTA pH 7.0 with 0.5 mM H2O2 or diluted in aerobic MOPS buffer.

NAD(P)H peroxidase/oxidase activities

These assays were modified from that described by Coulter & Kurtz (2001) using either C. acetobutylicum NROR or FNR from spinach plus Rd as electron donors to revRbr. In some cases direct reaction of NROR with revRbr was tested without addition of Rd. The activities were measured at room temperature by rates of either NADH (with NROR) or NADPH (with FNR) consumption as the decrease in absorbance at 340 nm (ε340 6220 M−1 cm−1). All reagents and proteins were added from anaerobic stock solutions into a 400 μl cuvette. The reaction mix contained 0.25 mM H2O2, 0.1 mM NAD(P)H, 1 μM FNR or NROR in 50 mM MOPS and 0.1 mM EDTA at pH 7.0. After 30 s C. acetobutylicum Rd (1 μM, Bradford) and after a further 30 s revRbr was added at various concentrations (Bradford) to this premix. NAD(P)H oxidase activities of revRbr were measured analogously, but in aerobic buffer without added H2O2. Further variations are described in the text and figure legends. All reactions were also tested using RubY instead of revRbr.

H2O2 consumption

In the presence of H2O2, HRP oxidizes 10-acetyl-3,7-dihydroxyphenoxazine (Amplex Red) to the coloured resorufin. The assay procedure was adapted from that described by Seaver & Imlay (2001) as follows. One hundred microlitres of a 0.2 mM stock solution of Amplex Red in 50 mM potassium phosphate pH 7.8 was added to 250 μl of the sample mix. In the assay a calibrated flux of H2O2 was produced by the glucose/glucose oxidase reaction (as used for a peroxidase assay described by Smith et al., 1974). The composition of the sample mix was as follows: 200 μl 50 mM Tris/HCl buffer at pH 7.0 containing 2 μg glucose oxidase ml−1, 0.1 mM NADPH and 1 μM FNR. After 30 s, 10 μl of a 0.2 mg ml−1 HRP stock solution in a buffer of 50 mM MOPS and 0.1 mM EDTA at pH 7.0 was added, and after a further 30 s, 1 μM Rd. After addition of HRP a small increase of absorbance occurred. Purified revRbr was added to various concentrations after a constant baseline was obtained. To start the H2O2 generation, 10 μl of a glucose stock solution containing 0.15 M glucose, 0.1 M NH4Cl and 0.1 M CaCl2 was added to the reaction mix. H2O2 consumption by the peroxidase activity of revRbr was measured as a decreased H2O2 generation by the glucose/glucose oxidase reaction. In some cases the reaction was started by addition of HRP. Production of resorufin was followed by the increase in absorbance at 560 nm.

O2 consumption

O2 consumption during the revRbr reactions was measured at room temperature using a Fibox 3 Single Channel Fiber Optic oxygen meter (PreSens Precision Sensing). The premix contained all components listed above for the NAD(P)H peroxidase assay except H2O2. Protein components were added at 2 min intervals with revRbr added last. The reaction buffer was aerobic 50 mM MOPS with 0.1 mM EDTA at pH 7.0. The total volume of the mix was 4 ml. In some cases 0.3 mM H2O2 was also added to the premix prior to addition of revRbr.

O2 consumption rates were measured for whole-cell suspensions of the revRbr overexpession and wild-type strains of C. acetobutylicum as follows. Ten-millilitre cultures were grown as described above, and 450 μl aliquots were suspended with 50 μl 37 % formaldehyde for cell counting. Then 4.5 ml aliquots were flushed with air for 5 min and loaded into the cuvette of an optode. The cuvette was sealed, and O2 consumption was followed at room temperature using the Fiber Optic oxygen meter. Fresh CGM was used as a control. O2 concentrations of 280–320 μM were taken as saturated solutions. This concentration was reached 1–2 min after start of the measurement due to the breadboard construction. Consumption rates were calculated as fmol h−1 per cell.


Isolation and purification of C. acetobutylicum revRbr, RubY, NROR and Rd

Rd, revRbr, RubY and NROR were expressed in E. coli or in C. acetobutylicum and purified via Strep-Tactin column chromatography (Fig. 1); 5 mg revRbr, 1 mg RubY, 3.5 mg Rd and up to 1.5 mg NROR were obtained from 500 ml E. coli culture volumes.

Fig. 1
SDS-PAGE of purified proteins. The proteins were overexpressed in E. coli DH5α and purified via Strep-Tactin-affinity chromatography. The purified proteins were run on a 20 % SDS-PAGE gel, which was then stained with Coomassie brilliant blue. ...

Visible absorption spectra (330–800 nm) of the purified revRbr, Rd and NROR are shown in Fig. 2. Spectra of the completely air-oxidized forms of revRbr and Rd showed similar spectral features, with absorption maxima at 350, 375, 492 and 565 nm (Fig. 2, spectra a and b). The peak at 492 nm and shoulder at 565 nm are characteristic of the [Fe(SCys)4] sites in Rds and in Rbrs (Gupta et al., 1995; Jin et al., 2004a, b). The higher absorption in the range 330–400 nm of revRbr compared to that of Rd is due to the additional oxo-bridged diferric site in the former protein, as can be seen in the difference absorption spectrum (Fig. 2, inset II). The absorption spectra thus indicated that the respective centres of the purified proteins are occupied with iron and should, therefore, be functional. Both recombinant proteins, Rd and revRbr, were completely reducible by addition of equimolar sodium dithionite under anaerobic conditions, as indicated by the disappearance of their visible absorption features (data not shown). Cell lysates of E. coli or C. acetobutylicum overexpressing these proteins as well as purified protein fractions showed a red colour characteristic of the oxidized [Fe(SCys)4] sites. The colour intensity increased after exposure to air and, in cell lysates, decreased again after a few minutes. This latter colour regeneration decreased after repeated mixing of the cell lysates with air.

Fig. 2
Absorption spectra of C. acetobutylicum revRbr, Rd and NROR. (a) Oxidized C. acetobutylicum revRbr; (b) C. acetobutylicum Rd. Insets: (I) spectrum of C. acetobutylicum NROR; (II) revRbr (a) minus Rd (b) difference spectrum. Spectra were normalized to ...

Protein fractions containing NROR had a yellow colour, and showed absorption maxima at 380 and 450 nm and a shoulder at 490 nm (Fig. 2, inset I) characteristic of a flavin. These results are in agreement with previous studies on NROR (Guedon & Petitdemange, 2001).

Analytical gel filtration experiments indicated that the native molecular mass of as-isolated revRbr is ~80 kDa (data not shown). Based on the calculated monomer molecular mass for the revRbr construct, 21 576.5, these results are consistent with a tetrameric oligomer for revRbr.

NAD(P)H peroxidase/oxidase activities of C. acetobutylicum revRbr

According to the activities of other Rbrs (Coulter et al., 1999; Coulter & Kurtz, 2001; Weinberg et al., 2004), C. acetobutylicum revRbr should be able to function as the terminal component of an NAD(P)H peroxidase and/or oxidase, transferring electrons to H2O2 and/or O2 via an NAD(P)H oxidoreductase. We have reconstituted such a pathway using heterologously expressed and purified C. acetobutylicum revRbr, Rd and NROR. NADH consumption during the reaction was monitored via decreased absorption at 340 nm. In the reconstituted pathway NADH consumption was obtained with O2, as well as with H2O2 as a substrate (Fig. 3, time-courses a and c). In both cases NROR alone was able to transfer the electrons to revRbr (time-courses b and c). Addition of 1 μM Rd doubled the reaction rates of both H2O2 and O2 consumptions (Fig. 3, time-courses b and c). Addition of H2O2 to an aerobic reaction slightly enhanced the reaction rate (Fig. 3, time-course c). No consumption of NADH occurred when either NROR or revRbr was omitted from an otherwise complete assay mixture. The specific NADH peroxidase and NADH oxidase activities of revRbr of C. acetobutylicum are given in Table 1.

Fig. 3
NADH oxidase/peroxidase activity of revRbr using Rd and NROR. Activities were monitored at room temperature as the decrease in absorbance at 340 nm due to NADH consumption. In the premix 0.1 mM NADH was supplemented to the anaerobic reaction buffer of ...
Table 1
NADH peroxidase and NADH oxidase activities of C. acetobutylicum and purified proteins

NADH and C. acetobutylicum NROR could be replaced by NADPH and spinach FNR for the revRbr peroxidase/oxidase activities. FNR has been previously used in comparable systems of other bacteria (Coulter et al., 1999; Coulter & Kurtz, 2001; Riebe et al., 2007). Thus, C. acetobutylicum Rd could be reduced by spinach FNR using NADPH as an electron donor. However, in contrast to NROR, Rd was absolutely necessary for the peroxidase/oxidase activity when using FNR. As shown in Fig. 4, NADPH consumption was observed only when FNR, Rd and revRbr were all present in the reaction mix; no consumption of NADPH occurred in either the aerobic oxidase (time-course a) or the anaerobic peroxidase (time-course c) activities if any one of the proteins was omitted from the assay mixture. Generation of H2O2 by glucose/glucose oxidase reaction in aerobic solutions increased the NADPH consumption rate over that of the O2-only reactions (Fig. 4, time-course b). Time-course (c) in Fig. 4 shows that when a small amount (25 μM) of H2O2 was added to the anaerobic reaction mixture, NADPH consumption ceased when the H2O2 was consumed. A second anaerobic addition of the same amount of H2O2 resulted in the same rate and extent of NADPH consumption as did the first addition. Subsequent aeration of the reaction mix by vortexing resulted in further consumption of NADPH, confirming the oxidase and peroxidase activities under these conditions. The NAD(P)H peroxidase activity of revRbr was found to be in the same order of magnitude as reported for other Rbrs (Coulter et al., 1999; Coulter & Kurtz, 2001).

Fig. 4
NADPH oxidase/peroxidase activity of revRbr using Rd and FNR. The activities were monitored at room temperature as the decrease in absorbance at 340 nm due to NADPH consumption. In the premix 0.1 mM NADPH was supplemented to the anaerobic reaction buffer ...

NADH peroxidase and NADH oxidase activities of C. acetobutylicum RubY

The two C. acetobutylicum genes that encode Rbr homologues with the ‘classical’ diiron and [Fe(SCys)4] domain structure (rubY and cac3018) are not induced under oxidative stress conditions (Hillmann et al., 2006). We nevertheless overexpressed and purified one of these classical Rbrs, RubY, and tested its oxidase and peroxidase activities. The purified RubY was found to be a homodimer under non-denaturing conditions (data not shown) and exhibited peroxidase activity comparable to that of revRbr (Table 1). As was the case for revRbr, Rd was not necessary for RubY’s peroxidase activity, but the presence of Rd increased this activity. However, the oxidase activity of RubY was very low compared to that of revRbr. With NROR in the absence or presence of Rd the NADH oxidase activity of RubY was only ~4 % or ~21 %, respectively, of the corresponding revRbr NADH oxidase activities.

H2O2 is the preferred substrate of revRbr

In addition to the consumption of NAD(P)H due to the activity of revRbr, the consumption of O2 and H2O2 was also followed. HRP catalyses oxidation of Amplex Red to the dye resorufin, by an H2O2 flux generated via glucose/glucose oxidase (Seaver & Imlay, 2001; Smith et al., 1974). As shown in Fig. 5, this oxidation was inhibited when revRbr was added to an assay mixture also containing NADPH, FNR and Rd, consistent with consumption of H2O2. This inhibition was proportional to the concentration of revRbr (Fig. 5, time-course b).

Fig. 5
H2O2 generation is inhibited by revRbr. H2O2 generation due to glucose/glucose oxidase reaction was measured by the Amplex Red/HRP system. Generation of resorufin was measured at 560 nm. (a) Control reaction containing only Amplex Red/HRP and glucose/glucose ...

Addition of revRbr (1 μM) to an aerobic premix containing NROR (1 μM), NADH (0.1 mM) and Rd (1 μM) caused a decrease of the dissolved O2 with a maximal activity of 12.5 μM O2 per μM revRbr per min (Fig. 6, time-course b). However, addition of 0.25 mM H2O2 to an aerobic assay mixture followed by addition of revRbr resulted in no measurable O2 consumption (Fig. 6, time-course a). This inhibition of O2 consumption by H2O2 coupled with the increase in NADH consumption when H2O2 is generated in the presence of O2 (Fig. 4, time-courses a vs b) means that H2O2 outcompetes O2 as a substrate for revRbr.

Fig. 6
H2O2 inhibits O2 consumption by revRbr. O2 consumption rates were measured directly using an O2 consumption meter in aerobic reaction mixtures containing 0.1 mM NADH, 1 μM NROR and 0.1 mM EDTA in 50 mM MOPS, pH 7.0. Rd and revRbr (both dithionite ...

Reduction of revRbr by NAD(P)H and reoxidation by H2O2

Fig. 7 represents a reductive titration of revRbr. Oxidized revRbr showed a characteristic absorbance peak at 492 nm (line a). After addition of NROR this peak disappeared, indicating a NROR-catalysed reduction of revRbr by NADH (line c). Addition of an excess of H2O2 rapidly reoxidized the NADH-reduced revRbr under anaerobic assay conditions (line d). Analogous spectral titrations (Supplementary Fig. S1, available with the online version of this paper) showed that FNR and Rd together catalyse reduction of revRbr by NADPH under anaerobic conditions, and the resulting reduced revRbr can then be fully reoxidized by H2O2. These results are fully consistent with the NAD(P)H peroxidase activity of revRbr described above.

Fig. 7
Reductive titrations of revRbr with NADH/NROR and reoxidation by H2O2. UV-vis absorption spectra are shown for (a) 100 μM as-isolated revRbr and (b) 100 μM as-isolated revRbr plus 125 μM NADH. (c) Same sample as in (b) after addition ...

Effect of revRbr overexpression on oxidative stress resistance of C. acetobutylicum

To analyse the effect of revRbr on oxidative stress tolerance in vivo, one of the two (99 % identical) revRbr genes, rbr3B, was cloned into an expression vector, pTrbr3B, and overexpressed in C. acetobutylicum. In survival experiments (Fig. 8), the revRbr overexpression strain exhibited a greatly increased tolerance to added H2O2 compared to the wild-type strain. The revRbr overexpression strain survived 30 min exposure to concentrations up to 0.3 mM added H2O2 at least as well as the non-H2O2-treated control, whereas corresponding survival of the wild-type strain steadily decreased over the same H2O2 concentration range. The wild-type strain did not survive added H2O2 concentrations >0.3 mM, whereas the revRbr overexpression strain exhibited 80 % survival up to 0.8 mM H2O2. Similarly, compared to the wild-type (18 % c.f.u.), the survival rate of the overexpression strain doubled (44 % c.f.u.) in 10 ml cultures that were flushed with air (500 ml min−1) for 15 min. In agreement with this result the overexpression strain consumed 1.71 fmol O2 h−1 per cell compared to 0.93 fmol O2 h−1 per cell for the wild-type. See also Table 1 for the NADH peroxidase and NADH oxidase activities of C. acetobutylicum wild-type and revRbr-overexpressing cells.

Fig. 8
Survival of C. acetobutylicum after exposure to H2O2. Anaerobic C. acetobutylicum cultures in CGM medium at 37 °C were exposed to various concentrations of added H2O2 for 30 min. Then 50 μl aliquots of these exposed cultures were plated ...


Previous results demonstrated increased revRbr expression upon exposure of C. acetobutylicum to air or H2O2 (Kawasaki et al., 2004; May et al., 2004) and clearly implicated revRbr in oxidative stress protection. The combined in vitro and in vivo results reported in this work indicate that this protection most likely arises from revRbr’s reductive scavenging of H2O2 and possibly O2.

The significant new in vivo result is that a C. acetobutylicum strain overexpressing revRbr showed greatly increased tolerance to both air (44 % survival compared to 14 % in case of the wild-type) and H2O2 exposure (see Fig. 8). Complementary in vitro results show that a reconstituted system consisting of NROR, Rd and revRbr, all from C. acetobutylicum, can function as both an NADH peroxidase and an NADH oxidase. The electron-transfer pathway from NADH to O2 or H2O2 via revRbr incorporating these other components is shown diagrammatically in Fig. 9. While this pathway did not absolutely require Rd, its presence increased the NADH peroxidase and oxidase activities. NROR most likely either directly reduces the [Fe(SCys)4] site of revRbr or preferentially reduces the analogous site in Rd, which then transfers its electrons to revRbr. Consistent with this notion, spinach FNR was absolutely dependent on Rd for catalysis of NADPH reduction of revRbr, as was shown earlier for D. vulgaris Rbr (Coulter & Kurtz, 2001). The peroxidase activity of other Rbrs requires the [Fe(SCys)4] site (Jin et al., 2004a), presumably to funnel electrons from external donors to the diiron site. The tetrameric oligomer of revRbr could conceivably support either intra- or intersubunit electron transfer between [Fe(SCys)4] and diiron sites. Both cases have been observed in other Rbrs (Jin et al., 2004b; Iyer et al., 2005).

Fig. 9
Detoxification pathways for ROS and oxygen in C. acetobutylicum. Rd, rubredoxin; revRbr, reverse rubrerythrin; Dfx, desulfoferrodoxin; X, unknown protein; red, reduced; ox, oxidized. The stoichiometry of the reactions has not been observed. (a) and (b), ...

The in vitro results show that revRbr can use both O2 and H2O2 as electron acceptors, but that H2O2 is the preferred substrate. The inhibition of O2 consumption in the presence of H2O2 (Fig. 6, time-course a) is not due to an inactivation of revRbr by H2O2, since NADH consumption in the presence of H2O2 occurred under both aerobic and anaerobic conditions after addition of H2O2 (see Fig. 4, time-courses b and c). The inhibition of O2 consumption by H2O2 is, therefore, probably the result of a higher reactivity of the diiron site of revRbr with H2O2 than with O2, as has been demonstrated for other Rbrs (Coulter & Kurtz, 2001). H2O2 probably undergoes a concerted two-electron reduction by abstracting two electrons from the diferrous site of Rbrs, which would minimize Fenton-type chemistry, i.e. the one-electron reduction of H2O2 to OH. (Kurtz, 2006). In contrast, four electrons are necessary for the complete reduction of O2 to water. The two additional electrons are transferred from external donors (NROR or Rd) probably via the [Fe(SCys)4] domain of revRbr. If the reduction of O2 to water by revRbr occurs via an H2O2 intermediate, it apparently is not released from the active site. We did not detect evolution of H2O2 during the reaction of revRbr with O2 (data not shown). We also could not detect any significant catalase activity of revRbr.

While H2O2 seems to be the preferred substrate, revRbr uses O2 as a substrate in vitro at a significantly higher relative level than reported for other Rbrs (Coulter et al., 1999, Coulter & Kurtz, 2001). This observation is consistent with the increased consumption of O2 by the revRbr-overexpressing C. acetobutylicum strain. The revRbr oxidase activity could, thus, be relevant when C. acetobutylicum is exposed to air, where the intracellular O2 levels may be higher than the H2O2 levels. Exposure of C. acetobutylicum to air results in massive increases in transcription (May et al., 2004) and translation (Kawasaki et al., 2004) of revRbr.

In this work we have also shown that one of the two C. acetobutylicum Rbrs, RubY, with the ‘classical’ N-terminal diiron and C-terminal [Fe(SCys)4] domain structure, has peroxidase (H2O2 reductase) specific activity comparable to that of the revRbr in vitro. It is, therefore, somewhat surprising that the intracellular levels of these classical Rbrs do not seem to be affected by oxidative stress (May et al., 2004; Hillmann et al., 2006), either in C. acetobutylicum or in other clostridia (Geissmann et al., 1999). Our recent microarray data (unpublished) confirmed the very low expression level of the C. acetobutylicum classical Rbr genes and their lack of upregulation in response to air exposure. This is in contrast to classical Rbrs from sulphate-reducing bacteria (Lumppio et al., 1997, 2001; Dolla et al., 2006) and archaea (Weinberg et al., 2004), which are typically expressed at significant constitutive levels even under anaerobic growth conditions.

The present results explain the observed aerotolerance of a C. acetobutylicum deletion mutant of the putative rbr3AB repressor, PerR (Hillmann et al., 2008). Deletion of perR resulted in massive overproduction of revRbr. Crude cell extracts of this ΔperR strain have a red colour due to high amounts of revRbr and show >10-fold higher NADH oxidase and peroxidase activities than the wild-type strain.

The C. acetobutylicum ΔperR strain also shows superior O2 consumption and aerotolerance over the revRbr over-expression strain (16 fmol h−1 per cell for ΔperR vs 1.71 fmol h−1 per cell for the revRbr overexpression strain), consistent with apparent regulation of other O2 detoxification proteins by perR (Hillmann et al., 2008).

The present results clearly demonstrate reductive H2O2 and O2 scavenging by C. acetobutylicum revRbr both in vitro and in vivo. The NADH peroxidase and oxidase activities of revRbr combined with the superoxide reductase activity of Dfx constitute an efficient reductive scavenging system for ROS in C. acetobutylicum. Fig. 9 summarizes the proposed ROS and O2 detoxification pathways involving revRbr, Dfx, and yet to be identified proteins. Candidates for X in Fig. 9 are FprA1 (cac1027) and FprA2 (cac2449), which are upregulated in response to oxidative stress (Kawasaki et al., 2004, 2005) and homologues of which from other organisms show NAD(P)H oxidase activity (Kurtz, 2007). We have so far been unable to construct an rbr3AB deletion strain, which is consistent with a more global stress defence role previously identified for revRbr (Hillmann et al., 2006).

Supplementary Material


O. R. was supported by a fellowship of the Graduiertenförderung of the Federal State of Mecklenburg-Vorpommern. This work was partially supported by NIH grant GM040388 to D. M. K. and in part by the SysMO project COSMIC (http://www.sysmo.net) to R.-J. F. and H. B.


ferredoxin-NADP+ reductase
horseradish peroxidase
NADH:rubredoxin oxido-reductase
reverse rubrerythrin
reactive oxygen species
superoxide dismutase


A supplementary figure is available with the online version of this paper.


  • Bermejo LL, Welker NE, Papoutsakis ET. Expression of Clostridium acetobutylicum ATCC 824 genes in Escherichia coli for acetone production and acetate detoxification. Appl Environ Microbiol. 1998;64:1079–1085. [PMC free article] [PubMed]
  • Bertram J, Dürre P. Conjugal transfer and expression of streptococcal transposons in Clostridium acetobutylicum. Arch Microbiol. 1989;151:551–557.
  • Bradford MM. A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of protein-dye-binding. Anal Biochem. 1976;72:248–254. [PubMed]
  • Coulter ED, Kurtz DM., Jr A role for rubredoxin in oxidative stress protection in Desulfovibrio vulgaris. Catalytic electron transfer to rubrerythrin and two-iron superoxide reductase. Arch Biochem Biophys. 2001;394:76–86. [PubMed]
  • Coulter ED, Shenvi NV, Kurtz DM., Jr NADH peroxidase activity of rubrerythrin. Biochem Biophys Res Commun. 1999;255:317–323. [PubMed]
  • Dolla A, Fournier M, Dermoun Z. Oxygen defense in sulfate-reducing bacteria. J Biotechnol. 2006;126:87–100. [PubMed]
  • Fournier M, Zhang Y, Wildschut JD, Dolla A, Voordouw JK, Schriemer DC, Voordouw G. Function of oxygen resistance proteins in the anaerobic, sulfate-reducing bacterium Desulfovibrio vulgaris Hildenborough. J Bacteriol. 2003;185:71–79. [PMC free article] [PubMed]
  • Geissmann TA, Teuber M, Meile L. Transcriptional analysis of the rubrerythrin and superoxide dismutase genes of Clostridium perfringens. J Bacteriol. 1999;181:7136–7139. [PMC free article] [PubMed]
  • Girbal L, Von Abendroth G, Winkler M, Benton PMC, Meynial-Salles I, Croux C, Peters JW, Happe T, Soucaille P. Homologous and heterologous overexpression in Clostridium acetobutylicum and characterization of purified clostridial and algal Fe-only hydrogenases with high specific activities. Appl Environ Microbiol. 2005;71:2777–2781. [PMC free article] [PubMed]
  • Guedon E, Petitdemange H. Identification of the gene encoding NADH-rubredoxin oxidoreductase in Clostridium acetobutylicum. Biochem Biophys Res Commun. 2001;285:496–502. [PubMed]
  • Gupta N, Bonomi F, Kurtz DM, Jr, Ravi N, Wang DL, Huynh BH. Recombinant Desulfovibrio vulgaris rubrerythrin. Isolation and characterization of the diiron domain. Biochemistry. 1995;34:3310–3318. [PubMed]
  • Hillmann F, Fischer RJ, Bahl H. The rubrerythrin-like protein Hsp21 of Clostridium acetobutylicum is a general stress protein. Arch Microbiol. 2006;185:270–276. [PubMed]
  • Hillmann F, Fischer RJ, Bahl H. PerR acts as a switch for oxygen tolerance in the strict anaerobe Clostridium acetobutylicum. Mol Microbiol. 2008;68:848–860. [PubMed]
  • Imlay JA. How oxygen damages microbes: oxygen tolerance and obligate anaerobiosis. Adv Microb Physiol. 2002;46:111–153. [PubMed]
  • Imlay JA. Pathways of oxidative damage. Annu Rev Microbiol. 2003;57:395–418. [PubMed]
  • Iyer RB, Silaghi-Dumitrescu R, Kurtz DM, Jr, Lanzilotta WN. High-resolution crystal structures of Desulfovibrio vulgaris nigerythrin: facile, redox-dependent iron movement, domain interface variability, and peroxidase activity in the rubrerythrins. J Biol Inorg Chem. 2005;10:407–416. [PubMed]
  • Jin S, Kurtz DM, Jr, Liu ZJ, Rose J, Wang BC. Displacement of iron by zinc at the diiron site of Desulfovibrio vulgaris rubrerythrin. X-ray crystal structure and anomalous scattering analysis. J Inorg Biochem. 2004a;98:786–796. [PubMed]
  • Jin S, Kurtz DM, Jr, Liu ZJ, Rose J, Wang BC. X-ray crystal structure of Desulfovibrio vulgaris rubrerythrin with zinc substituted into the [Fe(SCys)4] site and alternative diiron site structures. Biochemistry. 2004b;43:3204–3213. [PubMed]
  • Kawasaki S, Ishikura J, Watamura Y, Niimura Y. Identification of O2-induced peptides in an obligatory anaerobe, Clostridium acetobutylicum. FEBS Lett. 2004;571:21–25. [PubMed]
  • Kawasaki S, Watamura Y, Ono M, Watanabe T, Takeda K, Niimura Y. Adaptive responses to oxygen stress in obligatory anaerobes Clostridium acetobutylicum and Clostridium aminovalericum. Appl Environ Microbiol. 2005;71:8442–8450. [PMC free article] [PubMed]
  • Kawasaki S, Ono M, Watamura Y, Sakai Y, Satoh T, Arai T, Satoh J, Niimura Y. An O2-inducible rubrerythrin-like protein, rubperoxin, is functional as a H2O2 reductase in an obligatory anaerobe Clostridium acetobutylicum. FEBS Lett. 2007;581:2460–2464. [PubMed]
  • Kurtz DM., Jr Avoiding high-valent iron intermediates: superoxide reductase and rubrerythrin. J Inorg Biochem. 2006;100:679–693. [PubMed]
  • Kurtz DM., Jr Flavo-diiron enzymes: nitric oxide or dioxygen reductases? Dalton Trans. 2007:4115–4121.
  • Laemmli UK. Cleavage of structural proteins during the assembly of the head of the bacteriophage T4. Nature. 1970;227:680–685. [PubMed]
  • Lehmann Y, Meile L, Teuber M. Rubrerythrin from Clostridium perfringens: cloning of the gene, purification of the protein, and characterization of its superoxide dismutase function. J Bacteriol. 1996;178:7152–7158. [PMC free article] [PubMed]
  • Lumppio HL, Shenvi NV, Garg RP, Summers AO, Kurtz DM., Jr A rubrerythrin operon and nigerythrin gene in Desulfovibrio vulgaris (Hildenborough) J Bacteriol. 1997;179:4607–4615. [PMC free article] [PubMed]
  • Lumppio HL, Shenvi NV, Summers AO, Voordouw G, Kurtz DM., Jr Rubrerythrin and rubredoxin oxidoreductase in Desulfovibrio vulgaris: a novel oxidative stress protection system. J Bacteriol. 2001;183:101–108. [PMC free article] [PubMed]
  • May A, Hillmann F, Riebe O, Fischer RJ, Bahl H. A rubrerythrin-like oxidative stress protein of Clostridium acetobutylicum is encoded by a duplicated gene and identical to the heat shock protein Hsp21. FEMS Microbiol Lett. 2004;238:249–254. [PubMed]
  • Mermelstein LD, Papoutsakis ED. In vivo methylation in Escherichia coli by the Bacillus subtilis phage phi 3T I methyltransferase to protect plasmids from restriction upon transformation of Clostridium acetobutylicum ATCC 824. Appl Environ Microbiol. 1993;59:1077–1081. [PMC free article] [PubMed]
  • Mydel P, Takahashi Y, Yumoto H, Sztukowska M, Kubica M, Gibson FC, III, Kurtz DM, Jr, Travis J, Collins LV, et al. Roles of the host oxidative immune response and bacterial antioxidant rubrerythrin during Porphyromonas gingivalis infection. PLoS Pathog. 2006;2:e76. [PMC free article] [PubMed]
  • Nölling J, Breton G, Omelchenko MV, Makarova KS, Zeng Q, Gibson R, Lee HM, Dubois J, Qiu D, et al. Genome sequence and comparative analysis of the solvent-producing bacterium Clostridium acetobutylicum. J Bacteriol. 2001;183:4823–4838. [PMC free article] [PubMed]
  • O’Brien RW, Morris JG. Oxygen and the growth and metabolism of Clostridium acetobutylicum. J Gen Microbiol. 1971;68:307–318. [PubMed]
  • Riebe O, Fischer RJ, Bahl H. Desulfoferrodoxin of Clostridium acetobutylicum functions as a superoxide reductase. FEBS Lett. 2007;581:5605–5610. [PubMed]
  • Roos JW, McLaughlin JK, Papoutsakis ET. The effect of pH on nitrogen supply, cell lysis and solvent production in fermentations of Clostridium acetobutylicum. Biotechnol Bioeng. 1985;27:681–694. [PubMed]
  • Schmidt TGM, Skerra A. The Strep-tag system for onestep purification and high-affinity detection or capturing of proteins. Nat Protoc. 2007;2:1528–1535. [PubMed]
  • Seaver LC, Imlay JA. Alkyl hydroperoxide reductase is the primary scavenger of endogenous hydrogen peroxide in Escherichia coli. J Bacteriol. 2001;183:7173–7181. [PMC free article] [PubMed]
  • Smith AL, Rosenberg I, Averill DR, Moxon ER, Stossel T, Smith DH. Brain polymorphonuclear leukocyte quantitation by peroxidase assay. Infect Immun. 1974;10:356–360. [PMC free article] [PubMed]
  • Sztukowska M, Bugno M, Potempa J, Travis J, Kurtz DM., Jr Role of rubrerythrin in the oxidative stress response of Porphyromonas gingivalis. Mol Microbiol. 2002;44:479–488. [PubMed]
  • Valentine JS, Wertz DL, Lyons TJ, Liou LL, Goto JJ, Gralla EB. The dark side of dioxygen biochemistry. Curr Opin Chem Biol. 1998;2:253–262. [PubMed]
  • Weinberg MV, Jenney FE, Jr, Cui X, Adams MW. Rubrerythrin from the hyperthermophilic archaeon Pyrococcus furiosus is a rubredoxin-dependent, iron-containing peroxidase. J Bacteriol. 2004;186:7888–7895. [PMC free article] [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Compound
    PubChem Compound links
  • MedGen
    Related information in MedGen
  • Pathways + GO
    Pathways + GO
    Pathways, annotations and biological systems (BioSystems) that cite the current article.
  • Protein
    Published protein sequences
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...