Ubiquinone limits oxidative stress in Escherichia coli

Britta Søballe1 and Robert K. Poole1

Krebs Institute for Biomolecular Research, Department of Molecular Biology and Biotechnology, The University of Sheffield, Firth Court, Western Bank, Sheffield S10 2TN, UK1

Author for correspondence: Robert K. Poole. Tel: +44 114 222 4447. Fax: +44 114 272 8697. e-mail: r.poole{at}sheffield.ac.uk


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Ubiquinone is an essential redox component of the aerobic respiratory chains of bacteria and mitochondria. It is well established that mammalian ubiquinone can function in its reduced form (ubiquinol) as a lipid-soluble antioxidant preventing lipid peroxidation. The objective of this study was to test the hypothesis that prokaryotic ubiquinone is involved in the defence against oxidative stress in the cytoplasmic membrane. The rate of superoxide production by rapidly respiring wild-type Escherichia coli membranes was twofold higher than in the slowly respiring membranes from a ubiCA knockout mutant. However, large amounts of superoxide accumulated in the Ubi- membranes compared to wild-type membranes, which possess superoxide-scavenging ubiquinol. Likewise, the rate of H2O2 production was twofold higher in the wild-type, but the overall production of H2O2 was again significantly higher in the Ubi- membranes. Inclusion of a water-soluble ubiquinone homologue (UQ-1) effectively decreased the amount of H2O2 produced in the Ubi- membranes in a concentration-dependent manner. Addition of UQ-2 to the membranes was even more effective in limiting accumulation of H2O2 than was UQ-1, suggesting a role for the side-chain in conferring liposolubility in the antioxidative defence mechanism. Intracellular H2O2 concentration was increased 1·8-fold in the ubiCA mutant, and expression of the katG gene, encoding the catalase hydroperoxidase I, as well as catalase enzyme activity, were increased twofold in this mutant. The ubiCA mutant was hypersensitive to oxidative stress mediated by CuSO4 or H2O2; sensitivity to the latter could be abolished by addition of cysteine. This phenotype was also exhibited by a ubiG mutant, defective in the last step of UQ biosynthesis and therefore expected to accumulate several UQ biosynthetic intermediates. These observations support the participation of reduced ubiquinone as an antioxidant in E. coli. The ubiCA mutant exhibited a pleiotropic phenotype, being resistant to heat, linolenic acid and phleomycin. Resistance to the two latter compounds is probably due to reduced uptake. Like mutants unable to synthesize the quinol oxidase, cytochrome bd, the ubiCA mutant was also sensitive to dithiothreitol, an effect that is attributed to inability of the respiratory chain to maintain an appropriate redox balance in the periplasm.

Keywords: ubiquinone, Escherichia coli, superoxide, peroxide, oxidative stress

Abbreviations: HRP, horseradish peroxidase; SOD, superoxide dismutase; UQ, ubiquinone; UQH2, ubiquinol (fully reduced form)


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Ubiquinone (Coenzyme Q or UQ) is a lipid-soluble component of membrane-bound electron-transport chains, where it is present in large molar excess over other respiratory components (Søballe & Poole, 1999 ). In animal cells, UQ is found not only in the inner mitochondrial membrane, but also in endoplasmic reticulum, Golgi, lysosomes, peroxisomes, and plasma membrane (Kalén et al., 1989 ). Recently, eukaryotic UQ has acquired renewed interest due to the increasing body of evidence suggesting that reduced UQ (i.e. ubiquinol or UQH2) is able to function as a lipid-soluble antioxidant (for a review, see Ernster & Dallner, 1995 ). UQH2 scavenges lipid peroxyl radicals and thereby prevents a chain reaction causing oxidative damage to polyunsaturated fatty acids of biological membranes, a process known as lipid peroxidation (Forsmark-Andrée et al., 1995 ). The amount of UQH2 and other antioxidants, such as vitamin E, present in low-density lipoprotein, is of vital importance for the prevention of atherosclerosis. The UQH2 form is maintained by quinone reductases (e.g. hepatocyte DT-diaphorase and lipoamide dehydrogenase) and thus protects against cytotoxic and carcinogenic effects (Beyer et al., 1996 ; Olsson et al., 1999 ). Inhibition of quinone reductase activity results in an increase in free radical damage (Beyer et al., 1996 ). The antioxidant properties of UQH2 have led to its clinical use in the treatment of various diseases, e.g. heart disease (Ernster & Dallner, 1995 ).

In the facultatively anaerobic bacterium Escherichia coli, UQ with an 8-unit isoprenoid side-chain is an essential component of the aerobic respiratory chain, whereas an alternative quinone, menaquinone or MQ, is more functional in anaerobic respiration (Gennis & Stewart, 1996 ). The most commonly shown scheme for the sequence of carriers in the aerobic respiratory chains consists of a single UQ pool located immediately upstream of the oxidases and downstream of the respiratory dehydrogenases (e.g. Gennis & Stewart, 1996 ). However, this scheme may be oversimplified, since we recently demonstrated that UQ also functions as electron carrier between cytochromes b and the terminal oxidases by using dual-wavelength spectrophotometry to monitor cytochrome reduction levels in a UQ-deficient strain (ubiCA) (Søballe & Poole, 1998 ). The ubiCA operon encodes the enzymes chorismate lyase and 4-hydroxybenzoate transferase for the first two committed steps of UQ biosynthesis (Søballe & Poole, 1999 ).

There remains controversy as to whether respiratory-chain quinones and quinone-like compounds such as menadione or anthracycline antibiotics might actually stimulate superoxide production (see Afanas’ev et al., 1990 ). In E. coli, it was recently demonstrated that NADH dehydrogenase II is a major source of superoxide and hydrogen peroxide production by autoxidation of its reduced FAD cofactor (Messner & Imlay, 1999 ). Membranes from a UQ- and MQ-deficient mutant produced more superoxide and peroxide; this was attributed to electrons ‘backing up’ at NADH dehydrogenase II. Sulfite reductase was found to be a second flavin-containing autoxidizable electron-transport enzyme of E. coli (Messner & Imlay, 1999 ). However, in bacteria, it is still not clear whether, in addition to its respiratory roles, UQH2 acts as an antioxidant or pro-oxidant. E. coli does contain a quinone oxidoreductase (Qor) (Thorn et al., 1995 ), which may be important in maintaining UQ in its fully reduced state by 2-electron reductions, but the role of this enzyme is unknown. In Saccharomyces cerevisiae, a UQ-deficient strain showed enhanced sensitivity to products of autoxidized polyunsaturated fatty acids, indicating a protective role of UQH2 (Do et al., 1996 ).

Some studies directed at determining the roles of quinones in bacterial physiology (e.g. Imlay, 1995 ) have used UQ-deficient mutants isolated after chemical mutagenesis, which are recognized to be unstable or leaky. Therefore, we have used a stable knockout mutant having a deletion and insertion at the junction of the ubiC and ubiA genes (Søballe & Poole, 1998 ) and tested the hypothesis that UQ in E. coli acts as an antioxidant in the cell’s defence against oxygen-derived radicals and oxidative stress in the cytoplasmic membrane.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Bacterial strains and growth conditions.
The E. coli strains used in this work are described in Table 1. P1 transductions were performed using a modification of the protocol of Miller (see Poole et al., 1996 ). All cultures were grown at 37 °C with vigorous shaking (200 r.p.m.) in conical flasks containing one-fifth their volume of medium. Culture optical density was measured with a Pye-Unicam SP6–550 spectrophotometer at 600 nm. However, cultures for ß-galactosidase assays were grown in 250 ml flasks with matched glass tubes of Klett dimensions as a side arm and the culture densities were measured with a Klett–Summerson photoelectric colorimeter (Manostat Corp.) fitted with a red filter.


View this table:
[in this window]
[in a new window]
 
Table 1. E. coli strains used in this study

 
Growth media were prepared as described by Poole et al. (1996) . For membrane preparations, strains MG1655 (wild-type) and RKP4152 (ubiCA) were grown aerobically in LB with 0·5% (w/v) xylose and harvested at OD600 0·55. For catalase assays, the strains were also grown to exponential phase (OD600 0·5) but in LB with 40 mM glucose. For ß-galactosidase assays, strains were grown in 10 ml LB with 40 mM xylose added. Strains were treated with paraquat (50 µM) or H2O2 (50 µM) at 50 Klett units and harvested in the exponential phase of growth at 100 Klett units.

Viable counts were performed after treating mid-exponential cultures of wild-type and ubi cells grown in MOPS/glucose (40 mM) medium with 0·03% (w/v) H2O2 or phleomycin (10 µg ml-1). Heat resistance was analysed by aerobic shock treatment at 52 °C. A portion (0·5 ml) of a mid-exponential culture was transferred to 4·5 ml MOPS/glucose medium, preincubated at 52 °C, and shaken at this temperature. When exposed to CuSO4 (5 mM) or linolenic acid (1 mM) the strains were grown in LB medium with xylose (40 mM). Samples of all the treated cultures were taken at appropriate time intervals and a dilution series (10-1 to 10-6) was performed in 0·9% NaCl. A portion (20 µl) of each dilution was spotted on LB agar plates and incubated at 37 °C overnight. DTT sensitivity studies were performed as described by Goldman et al. (1996a) . Wild-type and ubi mutant cells were streaked on nutrient agar plates containing various concentrations of DTT (0·8–24 mM) and grown at 30 °C for 2 d. To obtain aerobic growth curves, three flasks were inoculated with 1% of an overnight culture of wild-type or ubiCA cells in LB with glucose (40 mM). To one of the flasks was added H2O2 to a final concentration of 2·5 mM, whereas to another was added H2O2 (2·5 mM) plus cysteine to a final concentration of 0·83 mM.

Preparation of membranes.
This was performed as described by Søballe & Poole (1998) . Protein concentrations were determined by the method of Markwell et al. (1978) .

Superoxide anion detection.
The rate of production during reactions in vitro was measured as the superoxide dismutase (SOD)-sensitive rate of cytochrome c reduction (Imlay & Fridovich, 1991 ) in a Beckman DU 650 spectrophotometer. The 1 ml reaction cuvette contained 50 mM potassium phosphate buffer, 20 µM cytochrome c, and wild-type or ubi membranes (200 µg protein ml-1, final concentration). Reduction of cytochrome c was initiated by the addition of 100 µM NADH (final concentration) and monitored spectrophotometrically at 550 nm. Duplicate reactions were performed with the addition of 250 units SOD (Sigma). The extent of cytochrome reduced was calculated using an absorption coefficient, {epsilon}, of 21·0 mM-1 cm-1.

H2O2 production.
The amount of H2O2 produced by respiring membranes was measured using the scopoletin assay in a Hitachi F-2500 fluorescence spectrophotometer in time-scanning mode with excitation at 350 nm and emission at 460 nm (Loshen et al., 1971 ). The reaction mix (2·5 ml) in phosphate-buffered saline (PBS) buffer contained membranes (200 µg protein ml-1, wild-type or ubiCA) and 0·76 µM horseradish peroxidase (HRP). The fluorescent substrate scopoletin was added to a final concentration of 0·2 µM and the reaction was initiated by addition of 25 mM glucose. The amount of H2O2 generated was determined from a standard curve, in which the H2O2 concentration (0–0·15 M) was directly proportional to the quenching of scopoletin fluorescence. When indicated, water-soluble ubiquinone (UQ-1 or UQ-2, purchased from Sigma) was added to a final concentration of 0·2–2·0 µM (from a 10 mM stock in 1:1 ethanol/water) before the reaction was initiated by the addition of 5 mM glycerol.

Determination of intracellular H2O2 production was based on the assumption that free diffusion of H2O2 through the cell membrane allows an equilibrium to occur after about 15 min (Gonzalez-Flecha & Demple, 1994 ). Cells from 2·5 ml of a culture of the wild-type or ubiCA strain, grown in LB to the exponential phase (50 Klett units), were harvested and resuspended in 25 ml phosphate-buffered saline. Samples were taken at 5 min intervals and spun briefly in a microfuge before assaying the H2O2 content of the supernatant using the above fluorometric assay.

Catalase assay.
Cell pellets were washed in 100 mM potassium phosphate buffer (pH 7·0) and disrupted by sonication (five periods of 1 min each). Cell debris and unbroken cells were removed by centrifugation in a microfuge for 30 min at 13000 r.p.m. The protein concentration of the supernatant extract was determined by the method of Markwell et al. (1978) . The catalase activity was measured by monitoring {Delta}A at 240 nm in time-scanning mode in a Beckman DU 650 spectrophotometer (Gonzalez-Flecha & Demple, 1994 ). The 1 ml UV-cuvette contained 510 µg supernatant protein in potassium phosphate buffer and the reaction was initiated by adding H2O2 to a final concentration of 5 mM. The initial rate of decomposition of H2O2 was determined using an {epsilon}240 of 43·6x10-3 mM-1 cm-1. One unit of catalase activity is defined as the change in H2O2 concentration (mM) per min.

ß-Galactosidase assays.
Assays were carried out at room temperature as described before (Søballe & Poole, 1997 ). Each culture was assayed in triplicate; results were confirmed in at least two independent experiments.


   RESULTS AND DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
UQ limits accumulation of superoxide generated in aerobic respiration
A major site of generation in E. coli is the respiratory chain of the cell membrane (Gonzalez-Flecha & Demple, 1995 ; Imlay, 1995 ). To study the role of UQ in this process, was quantified by NADH-induced, SOD-sensitive cytochrome c reduction at 550 nm using membrane preparations of the wild-type (MG1655) and a UQ-deficient strain (ubiCA, RKP4152). Addition of SOD inhibited the initial rate of cytochrome c reduction by 44% in the wild-type and by 36% in the ubiCA mutant. SOD-insensitive reduction of cytochrome c presumably reflects direct interaction with respiratory chain components. Additional quantities of SOD did not further decrease the rate of cytochrome c reduction. The initial rate of production expressed per mg membrane protein in the ubiCA mutant was about half of that in wild-type cells (Fig. 1; Table 2). However, the respiration rate of NADH-treated membrane preparations is decreased in ubiCA mutants by about 80% (Søballe & Poole, 1998 ) and therefore correction of Fig. 1 for the electron transfer rates to O2 would emphasize the antioxidant role of UQ. Fig. 1 also shows that the rapid production of in the wild-type strain ceased after about 2–3 min, whereas in the ubiCA strain, production continued until the cytochrome c present in the assay was fully reduced (Fig. 1). Thus, the total amount of produced expressed per mg membrane protein was 2·3-fold higher in the ubiCA mutant compared to the wild-type strain (Table 2) under these assay conditions. This result clearly indicates the importance of UQ in maintaining low levels.



View larger version (17K):
[in this window]
[in a new window]
 
Fig. 1. Effect of UQ deficiency on superoxide production. The SOD-sensitive rate of cytochrome c reduction was measured in membranes from wild-type ({square}) and ubi ({bullet}) strains. The reaction, containing 20 µM cytochrome c and 0·2 mg membrane protein ml-1, was initiated by NADH (100 µM) and monitored at 550 nm. The results shown are typical of three similar determinations.

 

View this table:
[in this window]
[in a new window]
 
Table 2. Effects of UQ deficiency on superoxide and H2O2 production and metabolism

 
Messner & Imlay (1999) used vesicles produced from a chemically produced ubiA mutant (AN385) and found a high rate of production when expressed per electron transferred to O2. Titration of the substrate for UbiA (4-hydroxybenzoate) to the growing ubiA culture resulted in the production of UQ, increasing respiration rate and causing a decrease in the rate of production.

UQ limits accumulation of H2O2 in vitro
H2O2 production in membranes from wild-type and ubiCA cells was measured using a HRP-dependent assay, in which the quenching of the fluorescent substrate scopoletin is directly proportional to the production of H2O2 (not shown). The initial rate of H2O2 production was 37% lower in the ubiCA mutant compared to the wild-type (Table 2). However, in the ubiCA membranes, all the available scopoletin was quenched within 6 min (Fig. 2b), whereas the substrate was only 50% quenched in the wild-type strain in the same time period (Fig. 2a); overall, there was a twofold increase in the accumulation of H2O2 in the ubiCA mutant under these assay conditions (Fig. 2, Table 2).



View larger version (11K):
[in this window]
[in a new window]
 
Fig. 2. Effect of UQ deficiency on H2O2 production. Quenching of the fluorescent substrate scopoletin by glucose-respiring membranes from wild-type (a) and ubi (b) cells was followed in a fluorescence spectrophotometer in time-scanning mode, with excitation at 350 nm and emission at 460 nm. The assay contained 0·2 mg membrane protein ml-1 and 0·76 µM HRP. Addition of 0·2 µM scopoletin (S) and 25 mM glucose (G) initiated the reaction. The results shown are typical of three similar determinations.

 
The effects on H2O2 production of adding a water-soluble ubiquinone homologue, UQ-1, to the ubi membranes are shown in Fig. 3(a). Over the assay period, the accumulation of H2O2 decreased substantially with increasing additions of UQ-1; accumulation of H2O2 decreased by 80% in the presence of 2 µM UQ-1. On addition of 0·4 µM UQ-1 [corresponding to 2 nmol (mg protein)-1], H2O2 accumulation decreased by 50% to a level equivalent to H2O2 accumulation in wild-type membranes. This is in good agreement with the UQ concentration in membranes from a wild-type strain (AN387) grown aerobically, i.e. about 2·26 nmol (mg protein)-1 (Wallace & Young, 1977 ). This UQ concentration is anticipated to offer protection against oxidative stress.



View larger version (20K):
[in this window]
[in a new window]
 
Fig. 3. H2O2 production in ubi membranes with added UQ homologues, UQ-1 or UQ-2. The scopoletin (S) assay (as described in Fig. 2) was initiated by addition of 5 mM glycerol (G). In (a), various concentrations of UQ-1 (0·2–2·0 µM) were included in the assay. In (b), UQ-1 or UQ-2 was added to a final concentration of 0·2 µM.

 
In Fig. 3(b) it is shown that ubiquinone with a 2-unit isoprenoid side-chain, UQ-2, decreased H2O2 accumulation to an even greater extent than UQ-1; addition of 0·2 µM UQ-2 decreased the H2O2 accumulation in the ubi membranes by 57%, whereas 0·2 µM UQ-1 decreased it by only 20%. This observation suggests that the length of the side-chain is important in the effectiveness of UQ as an antioxidant. Addition of UQ with longer side-chains, i.e. UQ-6 and UQ-10, was not possible due to their insolubility in water-based assays. In Sacch. cerevisiae, a broad spectrum of UQs (UQ-5 to UQ-10) has been shown to be biologically functional in a UQ-deficient strain, but the original UQ-6 species showed the highest activity (Okada et al., 1998 ).

Effects of UQ deficiency on oxidative stress in vivo: peroxide levels, catalase activity and expression of katG and sodA
The above results demonstrate that isolated membranes from a UQ-deficient strain accumulate substantially more superoxide and peroxide than do membranes containing a normal complement of UQ. To determine if this has physiological consequences for growth, or whether the additional flux of reactive oxygen species can be accommodated by cytoplasmic SOD and catalase activities, we determined the effects of the ubiCA knockout in intact cells.

The steady-state intracellular H2O2 concentration, which was determined after allowing diffusion of H2O2 into PBS buffer, was increased 1·8-fold in the ubiCA mutant (Table 2). These results mimic the and H2O2 measurements in membranes and support a protective role of UQ in oxygen radical scavenging.

We also assayed levels of total catalase activity in unfractionated cell extracts. Rates of H2O2 conversion were twofold higher in the ubiCA cells compared to wild-type levels (Table 2).

E. coli possesses a bifunctional catalase-peroxidase (HPI, KatG) and a second monofunctional catalase, HPII. HPI, together with an alkyl hydroperoxide reductase, many other proteins with roles in resisting peroxide stress, and a small untranslated RNA are regulated by the OxyR protein in response to peroxide (Demple, 1991 ; Rosner & Storz, 1997 ). The intracellular concentration of H2O2 in E. coli is normally maintained around 0·1–0·2 µM during aerobic growth and catalase activity is regulated to compensate for changes in H2O2 production rates (Gonzalez-Flecha & Demple, 1995 ). Thus, katG transcription is a useful measure of intracellular peroxide levels (Gonzalez-Flecha & Demple, 1995 ). Table 3 shows that expression of {Phi}(katG–lacZ) increased 2·2-fold in a wild-type strain when induced with H2O2 under our experimental conditions. To seek confirmation that the presence of UQ affects H2O2 levels in vivo, the effect of introducing the ubiCA allele into the {Phi}(katG–lacZ) strain was studied. The basal level of katG expression was 2·1-fold higher in the ubi background, but induction with H2O2 still increased the expression a further 1·3-fold (Table 3).


View this table:
[in this window]
[in a new window]
 
Table 3. Effect of UQ deficiency on expression of katG and sodA

 
The manganese-containing superoxide dismutase (MnSOD) enzyme is encoded by the sodA gene, which can be regarded as a monitor of intracellular oxidative stress as it is induced via soxRS regulators by , by increased oxygen pressure and by redox-active compounds (Touati, 1988 ; Demple, 1991 ; Compan & Touati, 1993 ). Paraquat is widely used as a redox-cycling agent; it donates a single electron to oxygen to give superoxide anion and is readily rereduced intracellularly, enabling further rounds of superoxide production. The expression of {Phi}(sodA–lacZ) increased 4·2-fold when induced with paraquat, as expected (Table 3). Surprisingly, the introduction of the ubiCA mutation did not increase the basal level of the aerobic expression of sodA (Table 3). A possible explanation is that, in vivo, the slow rate of superoxide generation (seen in membranes; Fig. 1) might be accommodated by spontaneous dismutation of superoxide to peroxide. Thus catalase levels are raised but SOD levels are not. Paraquat increased {Phi}(sodA–lacZ) expression 2·6-fold in the ubi background (Table 3). The fact that paraquat induction was not as marked in the ubi background could be due to the growth defect exhibited by this strain, especially in the presence of severe oxidative stress mediators such as paraquat (not shown). Alternatively, UQ function may be needed for paraquat redox-cycling and generation.

A ubiCA mutant is hypersensitive to mediators of oxidative stress
In view of the elevated levels of superoxide and peroxide measured in ubiCA cells, the ability of cells to tolerate additional exogenous oxidative stress mediated by H2O2 or CuSO4 was examined (Fig. 4). The ubiCA mutant appeared to be hypersensitive to treatment with 0·03% H2O2 and its viability was reduced about 16-fold relative to the wild-type strain after 1 h (Fig. 4a).



View larger version (14K):
[in this window]
[in a new window]
 
Fig. 4. Sensitivity of the ubi mutant to oxidative stress. Exponential cultures of wild-type ({square}) and ubi ({bullet}) cells were exposed to 0·03% H2O2 (a) and 5 mM CuSO4 (b) and samples for viable counts were taken at appropriate time intervals. The results shown are typical of three similar determinations.

 
We considered the possibility that intermediates in the UQ biosynthetic pathway downstream of the UbiCA-catalysed steps might act as antioxidants. It has recently been demonstrated that UbiG catalyses both O-methyltransferase steps in UQ biosynthesis, one of which is the last step (Hsu et al., 1996 ; Poon et al., 1999 ). However, strains harbouring leaky point-mutant alleles of ubiG (e.g. AN86, AN151; Stroobant et al., 1972 ) were shown to accumulate demethyl-UQ, the last intermediate in UQ biosynthesis. Unlike RKP4152 (ubiCA knockout strain), HW271 (ubiG) retains considerable respiratory activity with several oxidizable substrates (Wu et al., 1992 ), suggesting that this strain also harbours a leaky allele. We found that strain HW271 showed increased sensitivity to H2O2. After 1 h of H2O2 treatment, the viability of the ubiG strain was reduced 28-fold compared to the corresponding wild-type (not shown).

Copper ions (Cu2+/Cu1+) participate in a redox-cycle, resulting in the generation of superoxide anion. They have also been reported to catalyse the conversion of H2O2 to ·OH in vitro (Kimura & Nishioka, 1997 ). When treated with CuSO4, the viability of the ubiCA mutant was reduced 10-fold relative to the wild-type strain after 90 min (Fig. 4b). Likewise, the viability of the ubiG mutant, HW271, was reduced 4·4-fold after 30 min compared to its isogenic wild-type (not shown). These results indicate the importance of the presence of UQ for protection against oxidative stresses generated by H2O2 or CuSO4. Sensitivity to oxidative stress in a UQ-deficient mutant of the fission yeast Schizosaccharomyces pombe has also been reported recently (Suzuki et al., 1997 ).

Cysteine is an amino acid with antioxidant properties due to the presence of the thiol group (Suzuki et al., 1997 ). Sensitivity of the ubiCA mutant to H2O2 (2·5 mM) could be abolished by the addition of this compound to the growing cells (Fig. 5b). In contrast, the presence of H2O2 or cysteine did not affect growth of the wild-type cells (Fig. 5a). Likewise, cysteine or glutathione restores growth of a UQ-deficient (dps) mutant of fission yeast in minimal medium (Suzuki et al., 1997 ).



View larger version (13K):
[in this window]
[in a new window]
 
Fig. 5. Rescue of H2O2 sensitivity of the ubi mutant by cysteine. Growth of (a) the wild-type and (b) the ubiCA strain in LB-glucose ({bullet}), LB-glucose plus H2O2 (2·5 mM) ({blacksquare}), and LB-glucose plus both H2O2 (2·5 mM) and cysteine (0·83 mM) ({triangleup}). The results shown are typical of three similar determinations.

 
Pleiotropic phenotype of the ubiCA mutant
Linolenic acid is a polyunsaturated fatty acid prone to autoxidation and breakdown into toxic products (Do et al., 1996 ); UQ-deficient Sacch. cerevisiae (coq3) and Schiz. pombe have been reported to be hypersensitive to this compound (Do et al., 1996 ). The hypersensitivity could be abolished by addition of the COQ3 gene on a single-copy plasmid, butylated hydroxytoluene, {alpha}-tocopherol or trolox, a vitamin E analogue (Do et al., 1996 ). Surprisingly, when we analysed the sensitivity of E. coli wild-type and ubiCA strains to linolenic acid (1 mM) we found the ubiCA mutant to be highly resistant to this compound, whereas the viability of the wild-type was reduced 500-fold after 3 h (Fig. 6a).



View larger version (11K):
[in this window]
[in a new window]
 
Fig. 6. Pleiotropic phenotype of the ubi mutant. Exponential cultures of wild-type ({square}) and ubi ({bullet}) strains were exposed to 1 mM linolenic acid (a), 10 µg phleomycin ml-1 (b) and heat shock (c) at 52 °C. Samples for viable counts were taken at appropriate time intervals. The results shown are typical of three similar determinations.

 
A similar result was obtained when the cells were treated with phleomycin, which is an antibiotic and antitumour agent produced by Streptomyces verticillus. The drug induces DNA breaks and cell death in prokaryotes and eukaryotes (Collis & Grigg, 1989 ). The wild-type and ubiCA strains were treated with phleomycin at a final concentration of 10 µg ml-1 (Fig. 6b). The ubiCA strain appeared resistant to this concentration, whereas the viability of the wild-type was decreased 10-fold after 1 h of treatment (Fig. 6b). It is likely that the resistance to both linolenic acid and phleomycin is due to reduced uptake of these compounds in the ubiCA strain.

The effects of heat shock at 52 °C on the viability of wild-type and ubiCA strains are shown in Fig. 6(c). The ubiCA mutant proved to be extremely resistant to the lethal effects of heat shock. After 48 min at 52 °C the surviving fraction of the ubiCA mutant was reduced 4·5-fold, whereas that of the wild-type was reduced about 170-fold (Fig. 6c).

The ubiCA mutation causes sensitivity to DTT
Like ubi mutants, mutants (cydAB or cydDC) unable to synthesize the quinol oxidase, cytochrome bd, have a pleiotropic phenotype, including sensitivity to H2O2 (Wall et al., 1992 ; Goldman et al., 1996b ). In addition, certain reducing compounds can suppress the temperature-sensitive phenotype of both cydAB and cydDC mutants. DTT was found to inhibit Cyd- mutants at 3 and 10 mM, concentrations at which the isogenic wild-type strains were resistant (Goldman et al., 1996a , b ). A firm explanation for these effects is not available, but it has been suggested (Goldman et al., 1996a ) that the thiol periplasmic environment of a cydC mutant is oxidized in comparison to the wild-type strain. Using the same assay as Goldman et al. (1996a) , we found that the ubiCA mutant was sensitive to 8 mM DTT, whereas the wild-type was resistant to DTT at this concentration (not shown). Thiol hypersensitivity has also been reported for ubi mutants isolated after chemical mutagenesis (Zeng et al., 1998 ). Thus the DTT sensitivity is not due to the ability to assemble cytochrome bd per se, but to respiratory chain function. We hypothesize that ubi mutants are sensitive to DTT because of the requirement for the respiratory chain to oxidize the essential redox-active CXXC motif of DsbB. This membrane protein in turn oxidizes the active-site cysteines in DsbA, the disulfide bond formation factor in the periplasm (Kobayashi & Ito, 1999 ). Indeed, ubiA menA mutants and hemA mutants do accumulate a reduced form of DsbA (Kobayashi et al., 1997 ) and, intriguingly, dsbA/dsbB mutants are sensitive to DTT (Missiakas & Raina, 1997 ). These findings and the work of Bader et al. (1999) clearly link the function of the respiratory chain with maintenance of an appropriate redox environment in the periplasm.

Conclusions
The importance of UQ as a component of E. coli respiratory chains terminated by oxygen and nitrate is well established. The present studies using a defined knockout allele of ubiCA clearly demonstrate additional roles for UQ in limiting the accumulation of superoxide and peroxide. This contradicts the view that quinones might constitute an important source of superoxide by virtue of the spontaneous autoxidation of the radical form (for references, see Afanas’ev et al., 1990 ). Skulachev (1997) has suggested that respiration that does not involve the Q-cycle – a mechanism that can produce long-lived semiquinone (QH·) – might serve as a defence against reactive oxygen species. Further protection from superoxide production in functioning respiratory chains might be afforded by high rates of electron transfer to the terminal oxidase, thereby avoiding ‘excessive electronegativity of respiratory carriers’ (Papa et al., 1997 ) and the potential for undesirable single-electron donation to oxygen. Both these conditions appear to be met in E. coli, in which a Q cycle need not be invoked (Poole & Ingledew, 1987 ) and in which operation of cytochrome bd, which is not a proton pump, allows very rapid rates of respiration with a phenomenally high apparent affinity for oxygen (Km about 5 nM; D’mello et al., 1996 ). Both cyanide (Imlay, 1995 ) and quinone deficiency (Messner & Imlay, 1999 ) increase superoxide production in membrane vesicles, possibly by electron ‘leakage’ from an upstream component such as NADH dehydrogenase II.

The complex phenotype of ubi mutants is not surprising and arises from at least two important aspects of UQ function. First, UQ is able to limit accumulation of superoxide and peroxide due to its ability to rapidly abstract electrons from upstream dehydrogenases and transfer them to the oxidases. Second, reduced UQ is able to react with superoxide in vitro (Nakayama et al., 1997 ) and functions as an antioxidant in scavenging oxygen radicals such as perferryls or lipid peroxyl radicals (Ernster & Dallner, 1995 ). These studies do not really distinguish between these two mechanisms. Consequent damage in ubi mutants by lipid peroxidation may therefore explain the oxidative-stress-sensitive phenotype and poor growth, as well as the changes in membrane permeability and heat tolerance. Finally, other phenotypes such as sensitivity to DTT are more likely a consequence of impaired respiratory chain function.


   ACKNOWLEDGEMENTS
 
This work was supported by the Biotechnology and Biological Sciences Research Council (UK) through grant P07744 to R.K.P. We are grateful to Danièle Touati and Bruce Demple for providing strains.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Afanas’ev, I. B., Korkina, L. G., Suslova, T. B. & Soodaeva, S. K. (1990). Are quinones producers or scavengers of superoxide ion in cells? Arch Biochem Biophys 281, 245-250.[Medline]

Bader, M., Muse, W., Ballou, D. P., Gassner, C. & Bardwell, C. A. (1999). Oxidative protein folding is driven by the electron transport system. Cell 98, 217-227.[Medline]

Beyer, R. E., Segura-Aguilar, J., Di Bernard, S. & 7 other authors (1996). The role of DT-diaphorase in the maintenance of the reduced antioxidant form of coenzyme Q in membrane systems. Proc Natl Acad Sci USA 93, 2528–2532.[Abstract/Free Full Text]

Collis, C. M. & Grigg, G. W. (1989). An Escherichia coli mutant resistant to phleomycin, bleomycin and heat inactivation is defective in ubiquinone synthesis. J Bacteriol 171, 4792-4798.[Medline]

Compan, I. & Touati, D. (1993). Interaction of six global regulators in expression of manganese superoxide dismutase in Escherichia coli K-12. J Bacteriol 175, 1687-1696.[Abstract]

Demple, B. (1991). Regulation of bacterial oxidative stress genes. Annu Rev Genet 25, 315-337.[Medline]

D’mello, R., Hill, S. & Poole, R. K. (1996). The cytochrome bd quinol oxidase in Escherichia coli has an extremely high oxygen affinity and two oxygen-binding haems: implications for regulation of activity in vivo by oxygen inhibition. Microbiology 142, 755-763.[Abstract]

Do, T. Q., Schultz, J. R. & Clarke, C. (1996). Enhanced sensitivity of ubiquinone-deficient mutants of Saccharomyces cerevisiae to products of autoxidized polyunsaturated fatty acids. Proc Natl Acad Sci USA 93, 7534-7539.[Abstract/Free Full Text]

Ernster, L. & Dallner, G. (1995). Biochemical, physiological and medical aspects of ubiquinone function. Biochim Biophys Acta 1271, 195-204.[Medline]

Forsmark-Andrée, P., Dallner, G. & Ernster, L. (1995). Endogenous ubiquinol prevents protein modification accompanying lipid peroxidation in beef heart submitochondrial particles. Free Radic Biol Med 19, 749-757.[Medline]

Gennis, R. B. & Stewart, V. (1996). Respiration. In Escherichia coli and Salmonella: Cellular and Molecular Biology, 2nd edn, pp. 217–261. Edited by F. C. Neidhardt and others. Washington, DC: American Society for Microbiology.

Goldman, B. S., Gabbert, K. K. & Kranz, R. G. (1996a). Use of heme reporters for studies of cytochrome biosynthesis and heme transport. J Bacteriol 178, 6338-6347.[Abstract]

Goldman, B. S., Gabbert, K. K. & Kranz, R. G. (1996b). The temperature-sensitive growth and survival phenotypes of Escherichia coli cydDC and cydAB strains are due to deficiencies in cytochrome bd and are corrected by exogenous catalase and reducing agents. J Bacteriol 178, 6348-6351.[Abstract]

Gonzalez-Flecha, B. & Demple, B. (1994). Intracellular generation of superoxide as a by-product of Vibrio harveyi luciferase expressed in Escherichia coli. J Bacteriol 176, 2293-2299.[Abstract]

Gonzalez-Flecha, B. & Demple, B. (1995). Metabolic sources of hydrogen peroxide in aerobically growing Escherichia coli. J Biol Chem 270, 13681-13687.[Abstract/Free Full Text]

Hsu, A., Poon, W. W., Shepherd, J. A., Myles, D. C. & Clarke, C. (1996). Complementation of coq3 mutant yeast by mitochondrial targeting of the Escherichia coli UbiG polypeptide: evidence that UbiG catalyzes both O-methylation steps in ubiquinone biosynthesis. Biochemistry 35, 9797-9806.[Medline]

Imlay, J. A. (1995). A metabolic enzyme that rapidly produces superoxide, fumarate reductase of Escherichia coli. J Biol Chem 270, 19767-19777.[Abstract/Free Full Text]

Imlay, J. A. & Fridovich, I. (1991). Assay of metabolic superoxide production in Escherichia coli. J Biol Chem 266, 6957-6965.[Abstract/Free Full Text]

Kalén, A., Norling, B., Appelkvist, E.-L. & Dallner, G. (1989). Ubiquinone biosynthesis by the microsomal fraction from rat liver. Biochim Biophys Acta 926, 70-78.

Kimura, T. & Nishioka, H. (1997). Intracellular generation of superoxide by copper sulphate in Escherichia coli. Mutation Res 389, 237-242.[Medline]

Kobayashi, T. & Ito, K. (1999). Respiratory chain strongly oxidizes the CXXC motif of DsbB in the Escherichia coli disulfide bond formation pathway. EMBO J 18, 1192-1198.[Abstract/Free Full Text]

Kobayashi, T., Kishigama, S., Sone, M., Inokuchi, H., Mogi, T. & Ito, K. (1997). Respiratory chain is required to maintain oxidised states of the DsbA-DsbB disulfide bond formation system in aerobically growing Escherichia coli. Proc Natl Acad Sci USA 94, 11857-11862.[Abstract/Free Full Text]

Loshen, G., Flohe, L. & Chance, B. (1971). Respiratory chain linked H2O2 production in pigeon heart mitochondria. FEBS Lett 18, 261-264.[Medline]

Markwell, M. A. K., Haas, S. M., Bieber, L. L. & Tolbert, N. E. (1978). A modification of the Lowry procedure to simplify protein determination in membrane and lipoprotein samples. Anal Biochem 87, 206-210.[Medline]

Messner, K. R. & Imlay, J. A. (1999). The identification of primary sites of superoxide and hydrogen peroxide formation in the aerobic respiratory chain and sulfite reductase complex of Escherichia coli. J Biol Chem 274, 10119-10128.[Abstract/Free Full Text]

Missiakas, D. & Raina, S. (1997). Protein folding in the bacterial periplasm. J Bacteriol 179, 2465-2471.[Free Full Text]

Nakayama, T., Hashimoto, M. & Hashimoto, K. (1997). Superoxide dismutase inhibition of oxidation of ubiquinol and concomitant formation of hydrogen peroxide. Biosci Biotechnol Biochem 61, 2034-2038.

Okada, K., Kainou, T., Matsuda, H. & Kawamukai, M. (1998). Biological significance of the side chain length in Saccharomyces cerevisiae. FEBS Lett 431, 241-244.[Medline]

Olsson, J. M., Xia, L., Eriksson, L. C. & Bjornstedt, M. (1999). Ubiquinone is reduced by lipoamide dehydrogenase and this reaction is potently stimulated by zinc. FEBS Lett 448, 190-192.[Medline]

Papa, S., Guerrieri, F. & Capitanio, N. (1997). A possible role of slips in cytochrome c oxidase in the antioxygen defense system of the cell. Biosci Rep 17, 23-31.[Medline]

Poole, R. K. & Ingledew, W. J. (1987). Pathways of electrons to oxygen. In Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology, pp. 170–200. Edited by F. C. Neidhardt and others. Washington, DC: American Society for Microbiology.

Poole, R. K., Anjum, M. F., Membrillo-Hernández, J., Kim, S. O., Hughes, M. N. & Stewart, V. (1996). Nitric oxide, nitrite, and Fnr regulation of hmp (flavohemoglobin) gene expression in Escherichia coli K-12. J Bacteriol 178, 5487-5492.[Abstract]

Poon, W. W., Barkovich, R. J., Hsu, A. Y., Frankel, A., Lee, P. T., Shepherd, J. N., Myles, D. C. & Clarke, C. F. (1999). Yeast and rat Coq3 and Escherichia coli UbiG polypeptides catalyze both O-methyltransferase steps in coenzyme Q biosynthesis. J Biol Chem 274, 21665-21672.[Abstract/Free Full Text]

Rosner, J. L. & Storz, G. (1997). Regulation of bacterial responses to oxidative stress. Curr Top Cell Regul 35, 163-175.[Medline]

Skulachev, V. P. (1997). Membrane-linked systems preventing superoxide formation. Biosci Rep 17, 347-366.[Medline]

Søballe, B. & Poole, R. K. (1997). Aerobic and anaerobic regulation of the ubiCA operon, encoding the first two committed steps of ubiquinone biosynthesis in Escherichia coli. FEBS Lett 414, 373-376.[Medline]

Søballe, B. & Poole, R. K. (1998). Requirement for ubiquinone downstream of cytochrome(s) b in the oxygen-terminated respiratory chains of Escherichia coli K-12 revealed using a null mutant allele of ubiCA. Microbiology 144, 361-373.[Abstract]

Søballe, B. & Poole, R. K. (1999). Microbial ubiquinones: multiple roles in respiration, gene regulation and oxidative stress management. Microbiology 145, 1817-1830.[Free Full Text]

Stroobant, P., Young, I. G. & Gibson, F. (1972). Mutants of Escherichia coli K-12 blocked in the final reaction of ubiquinone biosynthesis: characterization and genetic analysis. J Bacteriol 109, 134-139.[Medline]

Suzuki, K., Okada, K., Kamiya, Y., Zhu, X. F., Nakagawa, T., Kawamukai, M. & Matsuda, H. (1997). Analysis of the decaprenyl diphosphate synthase (dps) gene in fission yeast suggests a role of ubiquinone as an antioxidant. J Biochem 121, 496-505.[Abstract]

Thorn, J. M., Barton, J. D., Dixon, N. E., Ollis, D. L. & Edwards, K. J. (1995). Crystal structure of Escherichia coli QOR quinone oxidoreductase complexed with NADPH. J Mol Biol 249, 785-799.[Medline]

Touati, D. (1988). Transcriptional and posttranscriptional regulation of manganese superoxide dismutase biosynthesis in Escherichia coli, studied with operon and protein fusions. J Bacteriol 170, 2511-2520.[Medline]

Wall, D., Delaney, J. M., Fayet, O., Lipinska, B., Yamamoto, T. & Georgopoulos, C. (1992). arc-dependent thermal regulation and extragenic suppression of the Escherichia coli cytochrome d operon. J Bacteriol 174, 6554-6562.[Abstract]

Wallace, B. J. & Young, I. G. (1977). Role of quinones in electron transport to oxygen and nitrate in Escherichia coli: studies with a double quinone mutant. Biochim Biophys Acta 461, 84-100.[Medline]

Wu, G., Williams, H. D., Zamanian, M., Gibson, F. & Poole, R. K. (1992). Isolation and characterization of Escherichia coli mutants affected in aerobic respiration: the cloning and nucleotide sequence of ubiG. Identification of an S-adenosylmethionine-binding motif in proteins, RNA, and small-molecule methyltransferases. J Gen Microbiol 138, 2101-2112.[Medline]

Zeng, H., Snavely, P., Zamorano, P. & Javor, G. T. (1998). Low ubiquinone content in Escherichia coli causes thiol hypersensitivity. J Bacteriol 180, 3681-3685.[Abstract/Free Full Text]

Received 11 August 1999; revised 13 December 1999; accepted 15 December 1999.