Escherichia coli glyoxalase II is a binuclear zinc-dependent metalloenzyme

Cytotoxic methylglyoxal is detoxified by the two-enzyme glyoxalase system. Glyoxalase I (GlxI) catalyzes conversion of non-enzymatically produced methylglyoxal-glutathione hemithioacetal into its corresponding thioester. Glyoxalase II (Glx II) hydrolyzes the thioester into d-lactate and free glutathione. Glyoxalase I and II are metalloenzymes, which possess mononuclear and binuclear active sites, respectively. There are two distinct classes of GlxI; the first class is Zn2+-dependent and is composed of GlxI from mainly eukaryotic organisms and the second class is composed of non-Zn2+-dependent (but Ni2+ or Co2+-dependent) GlxI enzymes (mainly prokaryotic and leishmanial species). GlxII is typically Zn2+-activated, containing Zn2+ and either Fe3+/Fe2+ or Mn2+ at the active site depending upon the biological source. To address whether two classes of GlxII might exist, glyoxalase II from Escherichia coli was cloned and overexpressed and characterized. Unlike E. coli GlxI, which is non-Zn2+-dependent, Zn2+ activates the E. coli GlxII enzyme, with no evidence for Ni2+ metal utilization.     

Pseudomonas aeruginosa contains multiple glyoxalase I-encoding genes from both metal activation classes

The glyoxalase (Glx) system is a critical detoxification enzyme system that is widely distributed in prokaryotic and eukaryotic organisms. Glyoxalase I (GlxI), the first enzyme in the system, is a divalent metal-ion dependent lyase (isomerizing), and its homologs have recently been categorized into two metal activation classes which are either Zn2+-dependent or non-Zn2+ dependent (Ni2+-/Co2+-activated). The latter class encompasses enzymes of predominantly bacterial origin. We have identified two genes in Pseudomonas aeruginosa PAO1 encoding glyoxalase I enzymes in addition to the gloA1 sequence recently reported and characterized. The gloA1 and gloA2 genes encode non-Zn2+ dependent glyoxalase I enzymes and the gloA3 gene remarkably encodes a Zn2+-dependent homolog. To our knowledge this is the first report of a eubacterial species with several GlxI encoding genes, and also of an organism possessing GlxI enzymes from both metal activation classes.     

The role of glyoxalase I in the detoxification of methylglyoxal and in the activation of the KefB K+ efflux system in Escherichia coli

The glyoxalase I gene ( gloA ) of Escherichia coli has been cloned and used to create a null mutant. Cells overexpressing glyoxalase I exhibit enhanced tolerance of methylglyoxal (MG) and exhibit elevated rates of detoxification, although the increase is not stoichiometric with the change in enzyme activity. Potassium efflux via KefB is also enhanced in the overexpressing strain. Analysis of the physiology of the mutant has revealed that growth and viability are quite normal, unless the cell is challenged with MG either added exogenously or synthesized by the cells. The mutant strain has a low rate of detoxification of MG, and cells rapidly lose viability when exposed to this electrophile. Activation of KefB and KefC is diminished in the absence of functional glyoxalase I. These data suggest that the glutathione-dependent glyoxalase I is the dominant detoxification pathway for MG in E . coli and that the product of glyoxalase I activity, S-lactoylglutathione, is the activator of KefB and KefC.     

Identification of Sequences Encoding the Detoxification Metalloisomerase Glyoxalase I in Microbial Genomes from Several Pathogenic Organisms

The ubiquitous glyoxalase system, which is composed of two enzymes, removes cellular cytotoxic methylglyoxal (MG). In an effort to identify critical residues conserved in the evolution of the first enzyme in this system, glyoxalase I (GlxI), as well as the structural implications of sequence alterations in this enzyme, a search of the National Center for Biotechnology Information (NCBI) database of unfinished genomes was undertaken. Eleven putative GlxI sequences from pathogenic organisms were identified and analyses of these sequences in relation to the known and previously identified GlxI enzymes were performed. Several of these sequences show a very high similarity to the Escherichia coli GlxI sequence, most notably the 79% identity of the sequence identified from Yersinia pestis, the causative agent of bubonic plague. In addition to the conservation of residues critical to binding the catalytic metal in all of the proposed GlxI enzymes, four regions in the Homo sapiens GlxI enzyme are absent in all of the bacterial GlxI sequences, with the exception of Pseudomonas putida. Removal of these regions may alter the active-site conformation of the bacterial enzymes in relation to that of the H. sapiens. These differences may be targeted for the development of inhibitors selective to the bacterial enzymes. Received: 13 October 1999 / Accepted: 17 January 2000     

Determination of the structure of Escherichia coli glyoxalase I suggests a structural basis for differential metal activation

The metalloenzyme glyoxalase I (GlxI) converts the nonenzymatically produced hemimercaptal of cytotoxic methylglyoxal and glutathione to nontoxic S-D-lactoylglutathione. Human GlxI, for which the structure is known, is active in the presence of Zn(2+). Unexpectedly, the Escherichia coli enzyme is inactive in the presence of Zn(2+) and is maximally active with Ni(2+). To understand this difference in metal activation and also to obtain a representative of the bacterial enzymes, the structure of E. coli Ni(2+)-GlxI has been determined. Structures have also been determined for the apo enzyme as well as complexes with Co(2+), Cd(2+), and Zn(2+). It is found that each of the protein-metal complexes that is catalytically active has octahedral geometry. This includes the complexes of the E. coli enzyme with Ni(2+), Co(2+), and Cd(2+), as well as the structures reported for the human Zn(2+) enzyme. Conversely, the complex of the E. coli enzyme with Zn(2+) has trigonal bipyramidal coordination and is inactive. This mode of coordination includes four protein ligands plus a single water molecule. In contrast, the coordination in the active forms of the enzyme includes two water molecules bound to the metal ion, suggesting that this may be a key feature of the catalytic mechanism. A comparison of the human and E. coli enzymes suggests that there are differences between the active sites that might be exploited for therapeutic use.     

Reaction mechanism of the binuclear zinc enzyme glyoxalase II - A theoretical study

The glyoxalase system catalyzes the conversion of toxic methylglyoxal to nontoxic d-lactic acid using glutathione (GSH) as a coenzyme. Glyoxalase II (GlxII) is a binuclear Zn enzyme that catalyzes the second step of this conversion, namely the hydrolysis of S-d-lactoylglutathione, which is the product of the Glyoxalase I (GlxI) reaction. In this paper we use density functional theory method to investigate the reaction mechanism of GlxII. A model of the active site is constructed on the basis of the X-ray crystal structure of the native enzyme. Stationary points along the reaction pathway are optimized and the potential energy surface for the reaction is calculated. The calculations give strong support to the previously proposed mechanism. It is found that the bridging hydroxide is capable of performing nucleophilic attack at the substrate carbonyl to form a tetrahedral intermediate. This step is followed by a proton transfer from the bridging oxygen to Asp58 and finally C-S bond cleavage. The roles of the two zinc ions in the reaction mechanism are analyzed. Zn2 is found to stabilize the charge of tetrahedral intermediate thereby lowering the barrier for the nucleophilic attack, while Zn1 stabilizes the charge of the thiolate product, thereby facilitating the C-S bond cleavage. Finally, the energies involved in the product release and active-site regeneration are estimated and a new possible mechanism is suggested.     

Identification of Glyoxalase I Sequences in Brassica oleracea and Sporobolus stapfianus: Evidence for Gene Duplication Events

Glyoxalase I (GlxI) is the first of two enzymes involved in the cellular detoxification of methylglyoxal. A recent search of the National Center for Biotechnology Information (NCBI) databases with the protein sequence of Salmonella typhimurium GlxI identified two new hypothetical proteins with unassigned function. These two sequences, from Brassica oleracea and Sporobolus stapfianus, have significant sequence similarity to known GlxI sequences, suggesting that these two open reading frames encode for GlxI in these plants. Interestingly, analysis of these two new sequences indicates that they code for a protein composed of two fused monomers, a situation previously found solely in the yeast GlxI enzymes. Received: 10 May 1997 / Accepted: 15 October 1997     

Evidence for an evolutionary antagonism between Mrr and Type III modification systems

The Mrr protein of Escherichia coli is a laterally acquired Type IV restriction endonuclease with specificity for methylated DNA. While Mrr nuclease activity can be elicited by high-pressure stress in E. coli MG1655, its (over)expression per se does not confer any obvious toxicity. In this study, however, we discovered that Mrr of E. coli MG1655 causes distinct genotoxicity when expressed in Salmonella typhimurium LT2. Genetic screening enabled us to contribute this toxicity entirely to the presence of the endogenous Type III restriction modification system (StyLTI) of S. typhimurium LT2. The StyLTI system consists of the Mod DNA methyltransferase and the Res restriction endonuclease, and we revealed that expression of the LT2 mod gene was sufficient to trigger Mrr activity in E. coli MG1655. Moreover, we could demonstrate that horizontal acquisition of the MG1655 mrr locus can drive the loss of endogenous Mod functionality present in S. typhimurium LT2 and E. coli ED1a, and observed a strong anti-correlation between close homologues of MG1655 mrr and LT2 mod in the genome database. This apparent evolutionary antagonism is further discussed in the light of a possible role for Mrr as defense mechanism against the establishment of epigenetic regulation by foreign DNA methyltransferases.     

The alpha-methylglucoside transport in Escherichia coli K12 cells]

The transport of alpha-methylglucoside (MG) in the wild type cells of Escherichia coli K12 and the isogenic mutant strains, defective in the activity of phosphoenolpyruvate: sugar phosphotransferase system components was studied. It was shown that the enzyme IIB' in the absence of enzyme I and HPr is able to transport MG into the cells by a "facilitated" diffusion mechanism. Compounds which dissipate the energy of membrane protone potential such as NaN3, carbonylcyanide-m-chlorophenylhydrasone, dicyclohexylcarbodiimide, enhance the utilization of MG by the wild-type cells. However, the cells retaining intact enzyme IIB' but deficient in the phospho approximately HPr-generating system, were not sensitive to the action of poisons. The cells possessing the intact phospho HPr-generating system and inactive enzyme IIB' are also unaffected by the poisons. It seems that these results do not confirm the hypothesis of the direct delta mu H+ involvement in the regulation of transmembrane phosphorylation. The hypothesis is postulated that the energy metabolism inhibitors influence the phosphatase activity of factor III of the phosphotransferase system. The present data are well explained by this hypothesis.     

Flux balance analysis of ammonia assimilation network in E. coli predicts preferred regulation point

Nitrogen assimilation is a critical biological process for the synthesis of biomolecules in Escherichia coli. The central ammonium assimilation network in E. coli converts carbon skeleton α-ketoglutarate and ammonium into glutamate and glutamine, which further serve as nitrogen donors for nitrogen metabolism in the cell. This reaction network involves three enzymes: glutamate dehydrogenase (GDH), glutamine synthetase (GS) and glutamate synthase (GOGAT). In minimal media, E. coli tries to maintain an optimal growth rate by regulating the activity of the enzymes to match the availability of the external ammonia. The molecular mechanism and the strategy of the regulation in this network have been the research topics for many investigators. In this paper, we develop a flux balance model for the nitrogen metabolism, taking into account of the cellular composition and biosynthetic requirements for nitrogen. The model agrees well with known experimental results. Specifically, it reproduces all the (15)N isotope labeling experiments in the wild type and the two mutant (ΔGDH and ΔGOGAT) strains of E. coli. Furthermore, the predicted catalytic activities of GDH, GS and GOGAT in different ammonium concentrations and growth rates for the wild type, ΔGDH and ΔGOGAT strains agree well with the enzyme concentrations obtained from western blots. Based on this flux balance model, we show that GS is the preferred regulation point among the three enzymes in the nitrogen assimilation network. Our analysis reveals the pattern of regulation in this central and highly regulated network, thus providing insights into the regulation strategy adopted by the bacteria. Our model and methods may also be useful in future investigations in this and other networks.     

S-D-Lactoylglutathione can be an alternative supply of mitochondrial glutathione

The mitochondrial pool of GSH (glutathione) is considered the major redox system in maintaining matrix redox homeostasis, preserving sulfhydryl groups of mitochondrial proteins in appropriate redox state, in defending mitochondrial DNA integrity and protecting mitochondrial-derived ROS, and in defending mitochondrial membranes against oxidative damage. Despite its importance in maintaining mitochondrial functionality, GSH is synthesized exclusively in the cytoplasm and must be actively transported into mitochondria. In this work we found that SLG (S-D-lactoylglutathione), an intermediate of the glyoxalase system, can enter the mitochondria and there be hydrolyzed from mitochondrial glyoxalase II enzyme to D-lactate and GSH. To demonstrate SLG transport from cytosol to mitochondria we used radiolabeled compounds and the results showed two different kinetic curves for SLG or GSH substrates, indicating different kinetic transport. Also, the incubation of functionally and intact mitochondria with SLG showed increased GSH levels in normal mitochondria and in artificially uncoupled mitochondria, demonstrating transport not linked to ATP presence. As well mitochondrial-swelling assay confirmed SLG entrance into organelles. Moreover we observed oxygen uptake and generation of membrane potential probably linked to D-lactate oxidation which is a product of SLG hydrolysis. The latter data were confirmed by oxidation of D-lactate in mitochondria evaluated by measuring mitochondrial D-lactate dehydrogenize activity. In this work we also showed the presence of mitochondrial glyoxalase II in inter-membrane space and mitochondrial matrix and we investigated the role of SLG in whole cells. In conclusion, this work showed new alternative sources of GSH supply to the mitochondria by SLG, an intermediate of the glyoxalase system.     

sucAB and sucCD are mutually essential genes in Escherichia coli

sucAB and sucCD of Escherichia coli encode enzymes that generate succinyl-CoA from 2-oxoglutarate and succinate, respectively. Their mutual essentiality was studied. sucAB and sucCD could be deleted individually, but not simultaneously. The mutual essentiality of sucAB and sucCD was further confirmed by the conditional expression of sucABCD, sucAB, and sucCD under the control of a P(BAD) in E. coli MG1655, E. coli MG1655 (DeltasucCD), and E. coli MG1655 (DeltasucAB), respectively. These strains grew well in Luria-Bertani medium containing 0.1% arabinose, but not in the absence of arabinose unless the medium was supplemented with succinyl-CoA. Our results indicate that either sucAB or sucCD is enough to produce succinyl-CoA that is essential for cell viability.     

Generation of a non‐transmissive Borna disease virus vector lacking both matrix and glycoprotein genes

Borna disease virus (BoDV), a prototype of mammalian bornavirus, is a non‐segmented, negative strand RNA virus that often causes severe neurological disorders in infected animals, including horses and sheep. Unique among animal RNA viruses, BoDV transcribes and replicates non‐cytopathically in the cell nucleus, leading to establishment of long‐lasting persistent infection. This striking feature of BoDV indicates its potential as an RNA virus vector system. It has previously been demonstrated by our team that recombinant BoDV (rBoDV) lacking an envelope glycoprotein (G ) gene develops persistent infections in transduced cells without loss of the viral genome. In this study, a novel non‐transmissive rBoDV, rBoDV ΔMG, which lacks both matrix (M ) and G genes in the genome, is reported. rBoDV‐ΔMG expressing green fluorescence protein (GFP), rBoDV ΔMG‐GFP, was efficiently generated in Vero/MG cells stably expressing both BoDV M and G proteins. Infection with rBoDV ΔMG‐GFP was persistently maintained in the parent Vero cells without propagation within cell culture. The optimal ratio of M and G for efficient viral particle production by transient transfection of M and G expression plasmids into cells persistently infected with rBoDV ΔMG‐GFP was also demonstrated. These findings indicate that the rBoDV ΔMG‐based BoDV vector may provide an extremely safe virus vector system and could be a novel strategy for investigating the function of M and G proteins and the host range of bornaviruses.

     

Degradation of bacterial DNA by a natural antimicrobial agent with the help of biomimetic membrane system

The antimicrobial efficacy of methylglyoxal (MG) against several gram-negative bacteria including Escherichia coli has been reported. To determine the mechanism of action of MG, molecular interactions between lipid and MG within the liposomal membrane were also investigated. Multilamellar and unilamellar vesicles were prepared from 1, 2-dipalmitoyl-snglycero-3-phosphocholine (DPPC). The effect of MG on DPPC liposomal membrane was studied by fluorescence spectroscopy and differential scanning calorimetry. The results indicate that MG interacts mainly with the DPPC head group that produces a significant increase in the fluidity of liposomal vesicles, which could be the cause of a fusion/aggregation effect in microbial cells. The agarose gel electrophoresis study with the genomic DNA extracted from E. coli ATCC 25922 revealed that addition of MG could completely degrade this DNA within 1 h, pointing out to their distinctly high degree of sensitivity towards MG. Further, the drug was able to cross the cell membranes, penetrating into the interior of the cell and interacting with DNA for demonstrating antibacterial activity of MG.     

丁醇合成途径关键酶基因在大肠杆菌中的克隆和表达

【目的】克隆丙酮丁醇梭状芽胞杆菌(Clostridium acetobutylicum)ATCC824丁醇合成途径关键酶基因,构建产丁醇的工程大肠杆菌。【方法】以C.acetobutylicum ATCC824基因组为模板,分别扩增丁醇合成途径关键酶基因thil,adhE2和BCS operon(crt-bcd-etfB-etfA-hbd)基因序列,构建BCS operon-adhE2-thil/pTrc99a/MG1655(pBAT)。重组菌E.coli pBAT采用0.1 mmol异丙基-β-硫代半乳糖苷(IPTG)诱导5 h,测定乙酰基转移酶(THL)、3-羟基丁酰辅酶A脱氢酶(HBD)、3-羟基丁酰辅酶A脱水酶(CRT)、丁酰辅酶A脱氢酶(BCD)、醛醇脱氢酶(BYDH/BDH)的酶活。并以该基因工程菌作为发酵菌种,采用好氧、厌氧和微好氧三种培养方式,检测丁醇产量。【结果】酶活测定结果显示:THL酶活达到0.160 U/mg protein,酶活力提高了近30倍;HBD酶活力提高了近5倍;CRT酶活达到1.53 U/mg protein,野生菌株无此酶活;BCD酶活力提高了32倍;BYDH/BDH酶活力无显著提高。3种发酵培养结果显示在微好氧和厌氧条件下,均有丁醇产生,且丁醇的最大产量约为84 mg/L。【结论】本实验通过构建产丁醇基因工程大肠杆菌,实现了丁醇关键酶基因在大肠杆菌中的活性表达以及发酵产丁醇,为发酵法生产丁醇开辟了一条新的途径。     

An XAS investigation of product and inhibitor complexes of Ni-containing GlxI from Escherichia coli: mechanistic implications

Escherichia coli glyoxalase I (GlxI) is a metalloisomerase that is maximally activated by Ni(2+), unlike other known GlxI enzymes which are active with Zn(2+). The metal is coordinated by two aqua ligands, two histidines (5 and 74), and two glutamates (56 and 122). The mechanism of E. coli Ni-GlxI was investigated by analyzing Ni K-edge X-ray absorption spectroscopic (XAS) data obtained from the enzyme and complexes formed with the product, S-D-lactoylglutathione, and various inhibitors. The analysis of X-ray absorption near edge structure (XANES) was used to determine the coordination number and geometry of the Ni site in the various Ni-GlxI complexes. Metric details of the Ni site structure were obtained from the analysis of extended X-ray absorption fine structure (EXAFS). Interaction of S-D-lactoylglutathione (product) or octylglutathione with the enzyme did not change the structure of the Ni site. However, analysis of XAS data obtained from a complex formed with a peptide hydroxamate bound to Ni-GlxI is consistent with this inhibitor binding to the Ni center by displacement of both water molecules. XANES analysis of this complex is best fit with a five-coordinate metal and, given the fact that both histidine ligands are retained, suggests the loss of a glutamate ligand. The loss of a glutamate ligand would preserve the neutral charge on the Ni complex and is consistent with the lack of a significant shift in the Ni K-edge energy in this complex. These data are compared with data obtained from the E. coli Ni-GlxI selenomethionine-substituted enzyme. The replacement of three methionine residues in the native enzyme with selenomethionine does not affect the structure of the Ni site. However, addition of the peptide hydroxamate inhibitor leads to the formation of a complex whose structure as determined by XAS analysis is consistent with inhibitor binding via displacement of both water molecules but retention of both histidine and glutamate ligands. This leads to an anionic complex, which is consistent with an observed 1.7 eV decrease in the Ni K-edge energy. Plausible reaction mechanisms for Ni-GlxI are discussed in light of the structural information available.     

Evaluation of multidrug efflux pump inhibitors by a new method using microfluidic channels

Fluorescein-di-β-D-galactopyranoside (FDG), a fluorogenic compound, is hydrolyzed by β-galactosidase in the cytoplasm of Escherichia coli to produce a fluorescent dye, fluorescein. We found that both FDG and fluorescein were substrates of efflux pumps, and have developed a new method to evaluate efflux-inhibitory activities in E. coli using FDG and a microfluidic channel device. We used E. coli MG1655 wild-type, ΔacrB (ΔB), ΔtolC (ΔC) and ΔacrBΔtolC (ΔBC) harboring plasmids carrying the mexAB-oprM (pABM) or mexXY-oprM (pXYM) genes of Pseudomonas aeruginosa. Two inhibitors, MexB-specific pyridopyrimidine (D13-9001) and non-specific Phe-Arg-β-naphthylamide (PAβN) were evaluated. The effects of inhibitors on pumps were observed using the microfluidic channel device under a fluorescence microscope. AcrAB-TolC and analogous pumps effectively prevented FDG influx in wild-type cells, resulting in no fluorescence. In contrast, ΔB or ΔC easily imported and hydrolyzed FDG to fluorescein, which was exported by residual pumps in ΔB. Consequently, fluorescent medium in ΔB and fluorescent cells of ΔC and ΔBC were observed in the microfluidic channels. D13-9001 substantially increased fluorescent cell number in ΔBC/pABM but not in ΔBC/pXYM. PAβN increased medium fluorescence in all strains, especially in the pump deletion mutants, and caused fluorescein accumulation to disappear in ΔC. The checkerboard method revealed that D13-9001 acts synergistically with aztreonam, ciprofloxacin, and erythromycin only against the MexAB-OprM producer (ΔBC/pABM), and PAβN acts synergistically, especially with erythromycin, in all strains including the pump deletion mutants. The results obtained from PAβN were similar to the results from membrane permeabilizer, polymyxin B or polymyxin B nonapeptide by concentration. The new method clarified that D13-9001 specifically inhibited MexAB-OprM in contrast to PAβN, which appeared to be a substrate of the pumps and permeabilized the membranes in E. coli.     

Specific immobilization of in vivo biotinylated bacterial luciferase and FMN:NAD(P)H oxidoreductase.

Bacterial bioluminescence, catalyzed by FMN:NAD(P)H oxidoreductase and luciferase, has been used as an analytical tool for quantitating the substrates of NAD(P)H-dependent enzymes. The development of inexpensive and sensitive biosensors based on bacterial bioluminescence would benefit from a method to immobilize the oxidoreductase and luciferase with high specific activity. Toward this end, oxidoreductase and luciferase were fused with a segment of biotin carboxy carrier protein and produced in Escherichia coli. The in vivo biotinylated luciferase and oxidoreductase were immobilized on avidin-conjugated agarose beads with little loss of activity. Coimmobilized enzymes had eight times higher bioluminescence activity than the free enzymes at low enzyme concentration and high NADH concentration. In addition, the immobilized enzymes were more stable than the free enzymes. This immobilization method is also useful to control enzyme orientation, which could increase the efficiency of sequentially operating enzymes like the oxidoreductase-luciferase system.     

Induction of oxidative stress by high hydrostatic pressure in Escherichia coli

Using leaderless alkaline phosphatase as a probe, it was demonstrated that pressure treatment induces endogenous intracellular oxidative stress in Escherichia coli MG1655. In stationary-phase cells, this oxidative stress increased with the applied pressure at least up to 400 MPa, which is well beyond the pressure at which the cells started to become inactivated (200 MPa). In exponential-phase cells, in contrast, oxidative stress increased with pressure treatment up to 150 MPa and then decreased again, together with the cell counts. Anaerobic incubation after pressure treatment significantly supported the recovery of MG1655, while mutants with increased intrinsic sensitivity toward oxidative stress (katE, katF, oxyR, sodAB, and soxS) were found to be more pressure sensitive than wild-type MG1655. Furthermore, mild pressure treatment strongly sensitized E. coli toward t-butylhydroperoxide and the superoxide generator plumbagin. Finally, previously described pressure-resistant mutants of E. coli MG1655 displayed enhanced resistance toward plumbagin. In one of these mutants, the induction of endogenous oxidative stress upon high hydrostatic pressure treatment was also investigated and found to be much lower than in MG1655. These results suggest that, at least under some conditions, the inactivation of E. coli by high hydrostatic pressure treatment is the consequence of a suicide mechanism involving the induction of an endogenous oxidative burst.