RpoS-dependent expression of the second lysine decarboxylase gene in Escherichia coli

The second lysine decarboxylase gene (ldc) is at 4.7 min on the Escherichia coli chromosome [Kikuchi et al., J. Baceriol. 179, 4486-4492 (1997)]. This report showes that the expression of ldc as well as cadA was induced at stationary phase in the wild type of E. coli. The ldc was not expressed in a rpoS deletion mutant of E. coli at any growing stage. In contrast, cadA was expressed in the rpoS mutant. Thus, we conclude that the expression of ldc but not cadA at stationary phase is regulated by a RpoS-dependent mechanism (s) in E. coli.     

Characterization of alpha-ketoglutarate-dependent taurine dioxygenase from Escherichia coli

The Escherichia coli tauD gene is required for the utilization of taurine (2-aminoethanesulfonic acid) as a sulfur source and is expressed only under conditions of sulfate starvation. The sequence relatedness of the TauD protein to the alpha-ketoglutarate-dependent 2,4-dichlorophenoxyacetate dioxygenase of Alcaligenes eutrophus suggested that TauD is an alpha-ketoglutarate-dependent dioxygenase catalyzing the oxygenolytic release of sulfite from taurine (van der Ploeg, J. R., Weiss, M. A., Saller, E., Nashimoto, H., Saito, N., Kertesz, M. A., and Leisinger, T. (1996) J. Bacteriol. 178, 5438-5446). TauD was overexpressed in E. coli to approximately 70% of the total soluble protein and purified to apparent homogeneity by a simple two-step procedure. The apparent Mr of 81,000 of the native protein and the subunit Mr of 37,400 were consistent with a homodimeric structure. The pure enzyme converted taurine to sulfite and aminoacetaldehyde, which was identified by high pressure liquid chromatography after enzymatic conversion to ethanolamine. The reaction also consumed equimolar amounts of oxygen and alpha-ketoglutarate; ferrous iron was absolutely required for activity; and ascorbate stimulated the reaction. The properties and amino acid sequence of this enzyme thus define it as a new member of the alpha-ketoglutarate-dependent dioxygenase family. The pure enzyme showed maximal activity at pH 6.9 and retained activity on storage at -20 degrees C for several weeks. Taurine (Km = 55 microM) was the preferred substrate, but pentanesulfonic acid, 3-(N-morpholino)propanesulfonic acid, and 1,3-dioxo-2-isoindolineethanesulfonic acid were also desulfonated at significant rates. Among the cosubstrates tested, only alpha-ketoglutarate (Km = 11 microM) supported significant dioxygenase activity.     

Resistance to fluoroquinolones in Escherichia coli isolated from poultry

Quinolone-resistant Escherichia coli strains were isolated from poultry clinical samples in Saudi Arabia. The poultry flocks had been treated with oxolinic acid or flumequine prophylaxis. The measure of the uptake of fluoroquinolones showed that none of the strains had a reduced accumulation of quinolones. The result of complementation with the wild-type E. coli gyrA gene, which restored fluoroquinolone susceptibility, and the isolation of DNA gyrase from six isolates indicated that the resistant strains had an altered DNA gyrase. The minimum effective dose of ciprofloxacin for inhibition of supercoiling catalyzed by the isolated gyrases varied from 0.085 microgram/ml for a susceptible isolate (MIC < 4 micrograms/ml) up to 96 micrograms/ml for the more resistant one (strain 215, MIC > 64 micrograms/ml). For the same two isolates, the minimum effective doses of sparfloxacin varied from 0.17 up to 380 micrograms/ml. The in vitro selection of spontaneous single-step fluoroquinolone-resistant mutants using ciprofloxacin suggested that the more resistant mutants are likely the result of several mutations. These results also show that, as in human medicine, cross-resistance between older quinolones and fluoroquinolones can exist in veterinary isolates and reiterate the need for the prudent use of these drugs.     

Preliminary X-ray crystallographic analysis of biotin carboxylase isolated from Escherichia coli

Acetyl CoA carboxylase catalyzes the first committed step in the biosynthesis of long chain fatty acids. In Escherichia coli, the enzyme consists of three subunits that are isolated separately and display distinct functional properties. Here we report the crystallization and preliminary X-ray analysis of one of these components, namely biotin carboxylase. The crystals are grown by microdialysis against 10 mM potassium phosphate (pH 7.0), 1 mM EDTA, 2 mM DTT and 1 mM NaN3 at 4 degrees C. They belong to the space group P2(1)2(1)2(1) with unit cell dimensions of a = 61.9 A, b = 96.1 A and c = 180.6 A and contain one dimer per asymmetric unit. The crystals diffract to a nominal resolution of 2.2 A. From a mechanistic standpoint, biotin carboxylase is especially interesting in that it is the smallest protein within its class and is one of only two carboxylases that can utilize free biotin as a substrate.     

Characterization of pyruvate decarboxylase genes from rice

The pdc1 gene encoding pyruvate decarboxylase has been isolated and sequenced from an IR54 rice genomic library. In contrast to a previously isolated intron-less rice genomic pdc, pRgpdc3, this gene contains five intervening introns in the coding region and corresponds to a cDNA clone, pRcpdc1, isolated from an IR54-cDNA library constructed from anaerobically-induced mRNAs. Comparison of the deduced amino acid sequence of this gene with that of the rice pdc2 and pdc3 showed 88% and 89% similarity, and 78% and 79% identity, respectively. Southern blots indicated that more than three genes constitute the pdc gene family in rice. pdc1 is highly inducible under anaerobic conditions. Rice pdc2 is also inducible by anoxia but to a much lesser extent than pdc1.     

Identification and characterization of a newly isolated shiga toxin 2-converting phage from shiga toxin-producing Escherichia coli

Shiga toxins 1 (Stx1) and 2 (Stx2) are encoded by toxin-converting bacteriophages of Stx-producing Escherichia coli (STEC), and so far two Stx1- and one Stx2-converting phages have been isolated from two STEC strains (A. D. O'Brien, J. W. Newlands, S. F. Miller, R. K. Holmes, H. W. Smith, and S. B. Formal, Science 226:694-696, 1984). In this study, we isolated two Stx2-converting phages, designated Stx2Phi-I and Stx2Phi-II, from two clinical strains of STEC associated with the outbreaks in Japan in 1996 and found that Stx2Phi-I resembled 933W, the previously reported Stx2-converting phage, in its infective properties for E. coli K-12 strain C600 while Stx2Phi-II was distinct from them. The sizes of the plaques of Stx2Phi-I and Stx2Phi-II in C600 were different; the former was larger than the latter. The restriction maps of Stx2Phi-I and Stx2Phi-II were not identical; rather, Stx2Phi-II DNA was approximately 3 kb larger than Stx2Phi-I DNA. Furthermore, Stx2Phi-I and Stx2Phi-II showed different phage immunity, with Stx2Phi-I and 933W belonging to the same group. Infection of C600 by Stx2Phi-I or 933W was affected by environmental osmolarity differently from that by Stx2Phi-II. When C600 was grown under conditions of high osmolarity, the infectivity of Stx2Phi-I and 933W was greatly decreased compared with that of Stx2Phi-II. Examination of the plating efficiency of the three phages for the defined mutations in C600 revealed that the efficiency of Stx2Phi-I and 933W for the fadL mutant decreased to less than 10(-7) compared with that for C600 whereas the efficiency of Stx2Phi-II decreased to 0.1% of that for C600. In contrast, while the plating efficiency of Stx2Phi-II for the lamB mutant decreased to a low level (0.05% of that for C600), the efficiencies of Stx2Phi-I and 933W were not changed. This was confirmed by the phage neutralization experiments with isolated outer membrane fractions from C600, fadL mutant, or lamB mutant or the purified His6-tagged FadL and LamB proteins. Based on the data, we concluded that FadL acts as the receptor for Stx2Phi-I and Stx2Phi-II whereas LamB acts as the receptor only for Stx2Phi-II.     

Virulence genes of Shiga toxin-producing Escherichia coli isolated from food, animals and humans

The cold shock protein CspB from Bacillus subtilis is only marginally stable, but it folds extremely fast in a simple N reversible U two-state reaction. The corresponding cold shock proteins from the thermophile Bacillus caldolyticus and the hyperthermophile Thermotoga maritima show strongly increased conformational stabilities, but unchanged very fast two-state refolding kinetics. The absence of intermediates in the folding of B. subtilis CspB is thus not a corollary of its low stability. Rather, two-state folding and an unusually native-like activated state of folding seem to be inherent properties of these small all-beta proteins. There is no link between stability and folding rate, and numerous sequence positions exist which can be varied to modulate the stability without affecting the rate and mechanism of folding.     

Receptors on chicken erythrocytes for F42 fimbriae of Escherichia coli isolated from pigs

Haemagglutination of purified F42 fimbriae was found to be inhibited by N-acetyl-galactosamine. Purified F42 fimbrial adhesin reacted with distinct membrane components from chicken erythrocytes (35, 37 and 40 kDa) in immunoblot analysis, suggesting that the binding occurred to proteins or glycoproteins.     

Cardiolipin synthase from Escherichia coli

Escherichia coli cardiolipin synthase catalyzes reversible phosphatidyl group transfer from one phosphatidylglycerol molecule to another to form cardiolipin (CL) and glycerol. The enzyme is specified by the cls gene, located at min 28.02 of the E. coli genetic map. Cells with mutations in cls have longer doubling times, tend to lose viability in the stationary phase, are more resistant to 3,4-dihydroxybutyl-1-phosphonate, and have an altered sensitivity to novobiocin. Although cls null mutants appear to lack CL synthase activity, they are still able to form trace quantities of CL. The enzyme appears to be regulated at both the genetic and enzymatic levels. CL synthase's molecular mass is 45-46 kDa, or about 8 kDa less than the polypeptide predicted by the gene sequence, suggesting that posttranslational processing occurs. CL synthase can use various polyols such as mannitol and arabitol to convert CL to the corresponding phosphatidylglycerol analog. When the amino acid sequences of four bacterial CL synthases are compared, three highly conserved regions are apparent. One of these regions contains a conserved pentapeptide sequence, RN(Q)HRK, and another has a conserved HXK sequence. These two sequences may be part of the active site. E. coli CL synthase has been studied by using a mixed micelle assay. The enzyme is inhibited by CL, the product of the reaction, and by phosphatidate. Phosphatidylethanolamine partially offsets inhibition caused by CL but not by phosphatidate. CDP-diacylglycerol does not appear to affect the activity of the purified enzyme but does stimulate the activity associated with crude membrane preparations.     

Characterization of the gntT gene encoding a high-affinity gluconate permease in Escherichia coli

During the cooking of meats, several highly mutagenic heterocyclic amines (HCAs) are produced. Three HCAs, IQ, MeIQx, and PhIP have been under study for carcinogenicity in cynomolgus monkeys, and to date, IQ has been shown to be a potent hepatocarcinogen. Concomitantly, the metabolic processing of these HCAs has been examined. Metabolism studies show that the potent hepatocarcinogenicity of IQ is associated with the in vivo metabolic activation of IQ via N-hydroxylation and the formation of DNA adducts. In monkeys undergoing carcinogen bioassay with IQ, N-hydroxylation was confirmed by the presence of the N-hydroxy-N-glucuronide conjugate of IQ in urine. The N-hydroxylation of IQ appears to be carried out largely by hepatic CYP3A4 and/or CYP2C9/10, and not by CYP1A2, an isoform not expressed in liver of this species. Notably MeIQx is poorly activated in cynomolgus monkeys and lacks the potency of IQ to induce hepatocellular carcinoma after a 5-year dosing period. The poor activation of MeIQx appears to be due to the lack of constitutive expression of CYP1A2 and an inability of other cytochromes P450, such as CYP3A4 and CYP2C9/10, to N-hydroxylate the quinoxalines. MeIQx is detoxified in monkeys largely by conjugation with glucuronide at the N-1 position. Although the carcinogenicity of PhIP is not yet known, the metabolic data suggest that PhIP will be carcinogenic in this species. PhIP is metabolically activated in vivo in monkeys by N-hydroxylation, as discerned by the presence of the N-hydroxy-N-glucuronide conjugate in urine, bile, and plasma. PhIP also produces DNA adducts that are widely distributed in tissues. The results from these studies support the importance of N-hydroxylation in the carcinogenicity of HCAs in nonhuman primates and by analogy, the importance of this metabolic activation step in the possible carcinogenicity of dietary HCAs in humans.     

Antibiotic resistance pattern of enterotoxigenic Escherichia coli isolated from infants and young adults in Israel

BACKGROUND: The usual methods of closure of major chest and abdominal wall defects have significant disadvantages. Skin grafts provide no structural support and result in incisional hernias. Synthetic mesh requires skin cover and is prone to infection and wound breakdown. The tensor fasciae latae (TFL) myocutaneous flap offers skin cover and a semi-rigid fascial layer. We document our unit's experience in pedicled and free TFL flaps. METHODS: The TFL flap closure of trunk defects was undertaken in 10 patients between August 1989 and April 1997. All cases were not amenable to primary closure and repair with synthetic mesh or skin grafts. RESULTS: The defect was satisfactorily repaired in all cases without subsequent herniation. The closure techniques using a pedicled TFL flap and a TFL flap for a free-tissue transfer are described. CONCLUSIONS: We conclude that the TFL flap is the method of choice for repairs of major truncal defects.     

Characterization and functional role of the QH site of bo-type ubiquinol oxidase from Escherichia coli

Cytochrome bo is a four-subunit terminal ubiquinol oxidase in the aerobic respiratory chain of Escherichia coli that vectorially translocates protons not only via directed protolytic reactions but also via proton pumping. Previously, we postulated that a bound quinone in the high-affinity quinone binding site (QH) mediates electron transfer from the low-affinity quinol oxidation site (QL) in subunit II to low-spin heme b in subunit I as an electron gate and a transient electron reservoir [Sato-Watanabe, M., Mogi, T., Ogura, T., Kitagawa, T., Miyoshi, H., Iwamura, H., and Anraku, Y. (1994b) J. Biol. Chem. 269, 28908-28912]. In the present study, we carried out screening of ubiquinone analogues using a bound ubiquinone-free enzyme (DeltaUbiA1) that has been isolated from a ubiquinone biosynthesis mutant, and identified PC24 (2-chloro-4, 6-dinitrophenol), PC32 (2,6-dibromo-4-cyanophenol), and PC52 (2-isopropyl-5-methyl-4,6-dinitrophenol) as potent QH site inhibitors. PC15 (2,6-dichloro-4-nitrophenol) and PC16 (2, 6-dichloro-4-dicyanovinylphenol), potent QL site inhibitors, did not exhibit such a selective inhibition of the QH site. Binding studies using the air-oxidized DeltaUbiA enzyme showed that PC32 and PC52 have 4- to 7-fold higher affinity than ubiquinone-1. Reconstitution of the QH site with PC32 and PC52 resulted in a decrease of the apparent Vmax value to 1/7 and 1/3, respectively, of the control activity. These findings suggest that structural features of the QL and QH sites are different, and provide further support for the involvement of the QH site in intramolecular electron transfer and facile oxidation of quinols at the QL site.     

Characterization and sequence of the Escherichia coli panBCD gene cluster

A 4589 bp DNA segment containing the Escherichia coli panBCD gene cluster was sequenced, and found to contain 6 complete open reading frames. panB, panC, and panD were identified by subcloning and insertional mutagenesis. The orientation of panD was also confirmed by orientation-specific expression of asparate-1-decarboxylase. panB and panC lie adjacent to one another, but are separated from panD by orf3, which is oriented in the opposite direction. Interruptions in the remaining open reading frames did not affect growth on glucose-minimal medium. No significant similarity to sequences in databases was found for orf1 and orf2. Orf3 contained extensive similarity to reading frames defined by E. coli yjiP, yjiQ, yhgA, and yafD. The function of these amino acid sequences is as yet undefined.     

Characterization of mutations that allow p-aminobenzoyl-glutamate utilization by Escherichia coli

An Escherichia coli strain deficient in p-aminobenzoate synthesis was mutagenized, and derivatives were selected for growth on folic acid. Supplementation was shown to be due to p-aminobenzoyl-glutamate present as a breakdown product in commercial folic acid preparations. Two classes of mutations characterized by the minimum concentration of p-aminobenzoyl-glutamate that could support growth were obtained. Both classes of mutations were genetically and physically mapped to about 30 min on the E. coli chromosome. A cloned wild-type gene from this region, abgT (formerly ydaH) could confer a similar p-aminobenzoyl-glutamate utilization phenotype on the parental strain. Interruption of abgT on the plasmid or on the chromosome of the mutant strain resulted in a loss of the phenotype. abgT was the third gene in an apparent operon containing abgA, abgB, abgT, and possibly ogt and might be regulated by a divergently transcribed LysR-type regulator encoded by abgR. Two different single-base-pair mutations that gave rise to the p-aminobenzoyl-glutamate utilization phenotype lay in the abgR-abgA intercistronic region and appeared to allow the expression of abgT. The second class of mutation was due to a tandem duplication of abgB and abgT fused to fnr. The abgA and abgB gene products were homologous to one another and to a family of aminoacyl aminohydrolases. p-Aminobenzoyl-glutamate hydrolysis could be detected in extracts from several of the mutant strains, but intact abgA and abgB were not essential for p-aminobenzoyl-glutamate utilization when abgT was supplied in trans.     

Metabolites influence control of lysine transfer ribonucleic acid synthetase formation in Escherichia coli K-12

A mutant of E. coli K-12 has been isolated which has only 1-3% of the wild-type lysyl-tRNA synthetase activity [L-lysine:tRNA ligase (AMP forming), EC 6.1.1.6]. Additions of 20 mM L-alanine or 6 mM leucine dipeptides to the culture medium can restore the activity of lysyl-tRNA synthetase in the mutant strain to the wild-type level. Experiments on the in vivo charging of lysine tRNA in the mutant show that in the absence of the metabolites lysine tRNA is charged 15-23%. Upon the addition of 3 mM L-leucyl-L-alanine to the medium the lysyl tRNA synthetase activity increases 25-fold and the in vivo charging of lysine tRNA returns to the wild-type level. Experiments with antibody against lysyl-tRNA synthetase show that the stimulation of lysyl-tRNA synthetase activity by the metabolites is the result of new protein synthesis.     

Interaction of ADP-ribosylation factor with Escherichia coli enterotoxin that contains an inactivating lysine 112 substitution

Cholera toxin and Escherichia coli heat-labile enterotoxin (LT) exert their effects on cells through ADP-ribosylation of guanine nucleotide-binding proteins. Both toxins consist of one A subunit, which is an ADP-ribosyltransferase, and five B (or binding) subunits. Their enzymatic activities are latent; activation requires reduction and proteolysis, resulting in a catalytically active A1 protein and a much smaller A2 protein. These ADP-ribosyltransferases are activated by GTP-dependent 20-kDa ADP-ribosylation factors or ARFs. To determine if proteolysis plus reduction is required for appearance of the ARF allosteric site as well as for catalytic activity, an inactive mutant of LT, LT(E112K), with replacement of glutamate by lysine at position 112 of its A subunit, was utilized as a competitor in cholera toxin ADP-ribosyltransferase assays containing limiting amounts of ARF. LT(E112K) required trypsinization and reduction to become a potent, concentration-dependent inhibitor. Inhibition was reversed by increasing concentrations of ARF. Reduction or trypsinization alone did not generate an inhibitory form of LT(E112K). These studies are consistent with the conclusion that the ARF site is not expressed in the latent toxin. Both trypsinization and reduction are required for expression of a functional ARF binding site as well as for catalytic activity.     

Characterization of the coenzyme-B12-dependent glutamate mutase from Clostridium cochlearium produced in Escherichia coli

The glutamate mutase dependent on adenosylcobalamin (coenzyme B12) catalyzes the carbon skeleton rearrangement of (S)-glutamate to (2S,3S)-3-methylaspartate, the first step of the glutamate fermentation pathway of the anaerobic bacterium Clostridium cochlearium. The enzyme consists of two protein components, E, a dimer epsilon 2 (epsilon, 53.5 kDa) and S, a monomer (sigma, 14.8 kDa). The corresponding genes (glmE and glmS) were cloned, sequenced and over-expressed in Escherichia coli. The genes glmS and glmE are separated by glmL encoding a protein of unknown function. The deduced amino acid sequence of GlmL contains an ATP-binding motif which is common to chaperones of the HSP70-type, actin and procaryotic cell-cycle proteins. Both components of glutamate mutase were purified with excellent yields from cell-free extracts of E. coli carrying the corresponding genes. In contrast to component E, component S was shown to bind coenzyme B12. This observation strongly supports the idea that significant similarities of the amino acid sequences of component S and several other cobamide-dependent enzymes represent a common binding motif. Incubation of pure components E and S with coenzyme B12 resulted in the formation of a fully active glutamate mutase heterotetramer (epsilon 2 sigma 2) containing one molecule of coenzyme B12. EPR spectra of recombinant glutamate mutase, now available in sufficiency large amounts, were recorded after incubation of the enzyme with coenzyme B12 and (S)-glutamate. The EPR signals (gx,y approximately 2.1, gz = 1.985) were of much better resolution than observed earlier with the clostridial enzyme. Their typical hyperfine splitting is clearly derived from Co(II), which is involved in the formation of the paramagnetic species but is different from cob(II)alamin (gx,y = 2.25). The spin concentration was 34-50% of the concentration of the enzyme (epsilon 2 sigma 2) coenzyme complex. The competitive inhibitors (2S, 4S)-4-fluoroglutamate and 2-methyleneglutarate induced similar but not identical signals with spin concentrations of 134-148% of the enzyme concentration. Even (S)-[2,3,3,4,4-2H5]glutamate induced a signal significantly different to that of (S)-glutamate with an intensity of only 7%. These data suggest an involvement of the Co(II)-containing paramagnetic species in catalysis, the concentration of which reflects a steady state between its formation and decomposition. The large difference in the spin concentrations observed with (S)-glutamate as compared to the predeuterated glutamate is probably due to a kinetic isotope effect and indicates a cleavage of a C-H bond during formation of the paramagnetic species.(ABSTRACT TRUNCATED AT 400 WORDS)     

Identification of the coding region for a second poly(A) polymerase in Escherichia coli

We had earlier identified the pcnB locus as the gene for the major Escherichia coli poly(A) polymerase (PAP I). In this report, we describe the disruption and identification of a candidate gene for a second poly(A) polymerase (PAP II) by an experimental strategy which was based on the assumption that the viability of E. coli depends on the presence of either PAP I or PAP II. The coding region thus identified is the open reading frame f310, located at about 87 min on the E. coli chromosome. The following lines of evidence support f310 as the gene for PAP II: (i) the deduced peptide encoded by f310 has a molecular weight of 36,300, similar to the molecular weight of 35,000 estimated by gel filtration of PAP II; (ii) the deduced f310 product is a relatively hydrophobic polypeptide with a pI of 9.4, consistent with the properties of partially purified PAP II; (iii) overexpression of f310 leads to the formation of inclusion bodies whose solubilization and renaturation yields poly(A) polymerase activity that corresponds to a 35-kDa protein as shown by enzyme blotting; and (iv) expression of a f310 fusion construct with hexahistidine at the N-terminus of the coding region allowed purification of a poly(A) polymerase fraction whose major component is a 36-kDa protein. E. coli PAP II has no significant sequence homology either to PAP I or to the viral and eukaryotic poly(A) polymerases, suggesting that the bacterial poly(A) polymerases have evolved independently. An interesting feature of the PAP II sequence is the presence of sets of two paired cysteine and histidine residues that resemble the RNA binding motifs seen in some other proteins.     

Characterization of Escherichia coli NrdH. A glutaredoxin-like protein with a thioredoxin-like activity profile

Ribonucleotides are converted to deoxyribonucleotides by ribonucleotide reductases. Either thioredoxin or glutaredoxin is a required electron donor for class I and II enzymes. Glutaredoxins are reduced by glutathione, thioredoxins by thioredoxin reductase. Recently, a glutaredoxin-like protein, NrdH, was isolated as the functional electron donor for a NrdEF ribonucleotide reductase, a class Ib enzyme, from Lactococcus lactis. The absence of glutathione in this bacterium raised the question of the identity of the intracellular reductant for NrdH. Homologues of NrdH are present in the genomes of Escherichia coli and Salmonella typhimurium, upstream of the genes for the poorly transcribed nrdEF, separated from it by an open reading frame (nrdI) coding for a protein of unknown function. Overexpression of E. coli NrdH protein shows that it is a functional hydrogen donor with higher specificity for the class Ib (NrdEF) than for the class Ia (NrdAB) ribonucleotide reductase. Furthermore, this glutaredoxin-like enzyme is reduced by thioredoxin reductase and not by glutathione. We suggest that several uncharacterized glutaredoxin-like proteins present in the genomes of organisms lacking GSH, including archae, will also react with thioredoxin reductase and be related to the ancestors from which the GSH-dependent glutaredoxins have evolved by the acquisition of a GSH-binding site. We also show that NrdI, encoded by all nrdEF operons, has a stimulatory effect on ribonucleotide reduction.