INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Infection with Helicobacter pylori, a microaerophilic, gram-negative bacterium, is associated with type B gastritis and peptic ulcer disease and is a risk factor for gastric carcinomas in humans (8, 17, 27). It so prevalently affects the world population that H. pylori has been described as "the most common chronic infection" (http://www.cdc.gov/ncidod/dbmd/diseaseinfo/hpylori_t.html), with some developing countries experiencing nearly 100% infection rates. While H. pylori infection is generally controlled with a cocktail of antibiotics and bismuth (60), the specter of emerging antibiotic resistance necessitates the search for alternative drug strategies and a clearer understanding of bacterial defense systems.

H. pylori colonizes the mucosal layer of the stomach and secretes immunogenic products that recruit macrophages and polymorphonuclear leukocytes to the site of infection (59). Here, the resulting oxidative burst by the phagocytic cells produces reactive oxygen species (ROS), such as superoxide anion (O₂^·), hydrogen peroxide (H₂O₂), and the hydroxyl radical (OH^·), that damage gastric tissues. Increased ROS levels are present in H. pylori patients (19), and it is thought that long-term exposure to ROS contributes to the development of cancerous gastric cells (16).

To resist oxidative damage from chronic inflammation, H. pylori relies on a variety of protective enzymatic systems, including catalase (the katA gene product) (47) and superoxide dismutase (the sodB gene product) (63), which eliminate H₂O₂ and O₂^·, respectively; however, it is unknown how H. pylori coordinates its oxidative stress response because the bacterium lacks the oxidatively activated regulatory genes, soxRS and oxyR, common to other eubacteria (67). We report herein that H. pylori also expresses a thioredoxin (Trx)-dependent alkyl hydroperoxide reductase (AhpC) protein, a member of the peroxiredoxin (Prx) family (62), whose activity can detoxify lipid hydroperoxides analogous to those created in membranes exposed to ROS (32). All Prx enzymes are dimers, decamers, or, in some cases, higher-order aggregates (1, 38, 61) with one essential, N-terminal Cys residue per subunit (Cys46 of Salmonella typhimurium AhpC) (21). Some Prxs contain only this single, conserved Cys (the 1-Cys Prxs), but most include another conserved Cys residue (the 2-Cys Prxs) analogous to Cys165 in S. typhimurium AhpC, which links the two subunits via an intersubunit disulfide bond with the N-terminal Cys in the oxidized protein (21, 53). Reduced S. typhimurium AhpC directly converts hydroperoxides to alcohols with concomitant formation of a sulfenic acid (Cys-SOH) at Cys46; condensation between the two active-site cysteinyl derivatives then regenerates the stable intersubunit disulfide bond of the oxidized protein (20, 21).

Our search of the H. pylori genomic database for an AhpC homologue revealed a gene (ahpC, HP1563) that had previously been reported to encode a species-specific antigen by O'Toole et al. (48). Although not identified as a peroxidase at that time, the H. pylori AhpC sequence containing the two conserved Cys residues was later classified as a 2-Cys Prx (12). To reduce AhpC, most bacteria, including S. typhimurium, express a specialized flavoprotein reductase, AhpF, which is homologous to Escherichia coli Trx reductase (TrxR), except that AhpF contains an additional N-terminal region directly involved in AhpC reduction (54, 55). Alternatively, some eukaryotic Prx systems utilize TrxR and Trx to reduce the peroxidase (sometimes referred to as Trx-dependent Prxs or TPxs), where TrxR catalyzes the reduction of Trx, which, in turn, reduces the Prx disulfide bond. In these cases, electron transfer proceeds along the following path: NADPH  TrxR  Trx  AhpC (TPx)  ROOH.

While eukaryotic examples of Trx-dependent Prx proteins abound (e.g., Entamoeba histolytica AhpC [52], Saccharomyces cerevisiae TPx [10], and multiple human Prx homologues [11]), to our knowledge, only one other bacterial Trx-dependent Prx, besides the H. pylori AhpC described herein, has been experimentally demonstrated [TPx from the cyanobacterium Synechocystis (75)]. A distantly related Prx family member, E. coli bacterioferritin comigratory protein, has also been shown to exhibit low levels of Trx-dependent peroxidase activity (34). Inspection of the H. pylori genome yielded no ahpF, but an E. coli trxA homologue (trx1, HP0824 in the annotation of Tomb et al. [67]) encoding Trx1 was identified along with an E. coli trxB homologue (trxR, HP0825) encoding TrxR. Further examination uncovered HP1164, annotated as a TrxR locus due to its similarity to Plasmodium falciparum trxR (21% identity). HP1164 does not, however, encode a putative catalytic disulfide motif (CXXC or CXXXXC) indicative of such a redox center and was therefore excluded from these studies. Further genomic searches revealed a second Trx locus (trx2, HP1458) encoding Trx2. Therefore, we considered TrxR (HP0825)-Trx1 (HP0824) and TrxR (HP0825)-Trx2 (HP1458) to be good candidates as AhpC-reducing systems.

Herein, we present the first example of a Trx-dependent alkyl hydroperoxide reductase system from a gastric pathogen and describe the cloning, purification, and kinetic characterization of AhpC, Trx1, Trx2, and TrxR from H. pylori. Along with the genetic characterization of H. pylori ahpC, we have also shown that AhpC plays a critical role in the defense against oxygen toxicity that is essential for survival and growth, even in microaerophilic environments.

(Abstracts reporting some of this information have been published earlier [5, 6]).

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Materials. Sodium dodecyl sulfate (SDS), ultrapure glycine, ultrapure urea, EDTA, dithiothreitol, Tris base, and other buffer reagents were purchased from Research Organics (Cleveland, Ohio). Bacteriological medium components were from Difco Laboratories (Detroit, Mich.). Ethanol was obtained from Warner Graham Company (Cockeysville, Md.). Isopropyl--D-thiogalactopyranoside (IPTG) and 5-bromo-4-chloro-3-indolyl--D-galactopyranoside (X-Gal) stocks were from Inalc (Milan, Italy). Vent DNA polymerase, Moloney murine leukemia virus reverse transcriptase, and T4 polynucleotide kinase were purchased from New England Biolabs (Beverly, Mass.). Restriction enzymes, ligase, calf intestinal phosphatase, and restriction buffers were obtained from Promega (Madison, Wis.). Agarose medium EEO (electrophoresis grade) and H₂O₂ were from Fisher (Fairlawn, N.J.). Acrylamide-Bis (40%) solution was purchased from Bio-Rad (Hercules, Calif.). Ampicillin (AMP) powder, cumene hydroperoxide (CHP), glucose-6-phosphate, glucose-6-phosphate dehydrogenase, linoleic acid, lipoxygenase, and insulin were from Sigma (St. Louis, Mo.). Ethyl hydroperoxide (EtOOH) was from Polysciences, Inc. (Warrington, Pa.), and tert-butyl hydroperoxide (t-BOOH) was from Aldrich (Milwaukee, Wis.). NADPH and NADH were from Roche Molecular Biochemicals (Mannheim, Germany).

RNA isolation and primer extension analysis. For RNA isolation, 50 ml of H. pylori strain HP 26695 (67) was grown in Brucella broth supplemented with 10% fetal calf serum (Sigma) in a microaerobic environment (7% O₂, 5% CO₂) at 37°C to an optical density at 590 nm of 1.0 and harvested by centrifugation at 4°C for 10 min. Cells were lysed in 50 mM Tris-HCl (pH 8.0)-1 mM EDTA-50 mM NaCl buffer supplemented with 1.5% SDS for 5 min at 95°C. Following phenol extraction and precipitation, the RNA was dissolved in diethylpyrocarbonate-treated water and the total RNA concentration was determined at 260 nm. For the primer extension studies, an oligonucleotide (5'-GCCAGAAGAAAAGAATCGCACCGT-3'; Gibco-BRL Custom Primers) was 5'- end labeled in the presence of [-³²P]ATP (5,000 Ci/mmol; Amersham) and 20 U of T4 polynucleotide kinase. Total H. pylori RNA (100 µg) was incubated with the ³²P-labeled oligonucleotide and annealed under the following conditions: 80°C for 5 min, 65°C for 5 min, 42°C for 10 min, and 37°C for 20 min. Following annealing, the RNA was precipitated with ethanol, dried, and resuspended in 7 µl of diethylpyrocarbonate-treated water and 13 µl of reverse transcriptase buffer (50 mM Tris-HCl at pH 8.3, 30 mM KCl, 10 mM MgCl₂, 1 µg of actinomycin D [Sigma], 1 mM dithiothreitol, 2.5 mM each deoxynucleoside triphosphate). One microliter of Moloney murine leukemia virus reverse transcriptase (50 U/µl) was added to the sample, and reverse transcription was carried out at 42°C for 60 min; 1 µl of 0.5 M EDTA and 1 µl of RNase (10 mg/ml) were added, and the reaction mixture was incubated at 37°C for an additional 30 min. The reaction was stopped by phenol-chloroform extraction, followed by ethanol precipitation. The sample was then dried and resuspended in 10 µl of sequencing loading buffer. In parallel, dideoxy DNA sequencing of the cloned ahpC gene and the upstream region was performed using the same oligonucleotide primer to determine the transcriptional start site. After denaturation at 95°C for 2 min, aliquots were subjected to 6% urea polyacrylamide gel electrophoresis (PAGE) and autoradiographed.

Allelic replacement mutagenesis of ahpC. Genomic DNA isolated from H. pylori strain 26695 (67) was used to amplify ahpC with forward and reverse oligonucleotide primers (Table 1) under typical PCR conditions to generate a product 930 bp long. Isolated ahpC was digested at flanking EcoRI and SacI sites and then inserted into pBluescript-SK (+) plasmid vector (Stratagene, La Jolla, Calif.) digested with the same restriction enzymes. Vector derivatives of pBluescript were stably maintained in E. coli strains but were not replicated in H. pylori hosts. The ahpC-containing plasmid was then digested with XbaI (located approximately 330 bp from the ahpC translational start site), and a chloramphenicol resistance (camR) cassette originating from Campylobacter coli (70) digested with XbaI was inserted to interrupt the coding sequence of ahpC. Four H. pylori strains were transformed with the ahpC::camR construct: 26695 (67), SS1 (40), HP1 (28), and HP1061 (25). The same strains were also transformed with a construct in which the nitroreductase gene (rdxA) was interrupted with camR (26). H. pylori transformation (electroporation) and subsequent evaluation of any colonies obtained after up to 7 days of incubation in selective medium were carried out as described elsewhere (57).

Cloning of ahpC, trx1, trx2, and trxR into expression vectors. The H. pylori genomic database was searched for genes corresponding to the E. coli sequences for ahpC, trxA, trxB, and trxC (www.tigr.org/tdb/mdb/hpdb/hpdb.html). Clones GHPDQ26, GHPAB86, GHPCL04, and GHPAE26, containing trx1, trx2, trxR, and ahpC, respectively, were ordered from The Institute of Genomic Research-ATCC Microbial Genome Special Collection. Plasmid DNA was purified from the E. coli host using the Wizard Miniprep Kit (Promega). The genes of interest were amplified using PCR primers (listed in Table 1) synthesized in the DNA Synthesis Core Laboratory of the Comprehensive Cancer Center of Wake Forest University. PCR mixtures (50 µl) contained 300 µM deoxynucleoside triphosphates, 1 U of vent polymerase, 10 pmol each of the forward and reverse oligonucleotides, 0.5 mM MgCl₂, and 0.5 µg of template DNA. Reactions were carried out as follows in a Mini Cycler (MJ Research, Waltham, Mass.): 95°C for 30 s, 40°C for 45 s, and 72°C for 1.5 min (5 cycles); 98°C for 30 s and 72°C for 1 min (35 cycles); and then 72°C for 15 min. The PCR products were purified using the QIAquick PCR Cleanup kit (Qiagen, Studio City, Calif.). The PCR products, 600 bp for ahpC, 300 bp for trx1 and trx2, and 932 bp for trxR, were each separately ligated into the pCR2.1 TA cloning vector (Invitrogen, Carlsbad, Calif.). Plasmids containing each insert were digested with the restriction enzymes corresponding to the engineered restriction sites (Table 1), and then DNA fragments of interest were excised from agarose gels and purified using the Gene Clean II Kit (Bio 101, Inc., Vista, Calif.). These fragments were inserted, using T4 DNA ligase, into pPROK1 expression vectors (Clontech, Palo Alto, Calif.; expression under control of the tac promoter), appropriately digested, isolated from agarose gels, and pretreated with calf intestinal phosphatase.

Bacterial strains and culture procedures. Ligated DNA was transformed into competent E. coli XL-1 Blue cells (Stratagene), except the AhpC-expressing plasmid, which was transformed into an ahpCF deletion strain of E. coli, TA4315 (65). Single colonies were selected on Luria-Bertani (LB) plates containing AMP at 50 µg/ml, and those containing the recombinant DNA were evaluated for protein expression on SDS-polyacrylamide gels after induction with 0.4 mM IPTG. Isolated plasmid DNA for each construct was sequenced throughout the coding region by automated DNA sequencing at the Comprehensive Cancer Center of Wake Forest University. Bacterial stocks containing each plasmid with the subcloned gene were prepared from a single colony and stored at 80°C in LB broth containing 15% (vol/vol) glycerol. Culture procedures were generally the same as reported earlier (53).

Purification of recombinant AhpC. Purification of AhpC was performed as described earlier (53), with some modifications. All procedures were carried out in a standard buffer (pH 7.0) consisting of 25 mM potassium phosphate with 1.0 mM EDTA. Briefly, 100 ml of E. coli TA4315 harboring the pPROK1/ahpC plasmid was added to 10 liters of LB medium containing 0.5 g of AMP supplemented with 0.2% glucose in a BioFlo 2000 fermentor (New Brunswick Scientific, Edison, N.J.). IPTG (0.4 mM) was added at A₆₀₀ = 0.9, and bacteria were harvested by centrifugation 16 h after induction. Pelleted bacteria were disrupted with a Bead Beater (BioSpec Products, Bartlesville, Okla.), cell extracts were treated with streptomycin sulfate to precipitate nucleic acids, and 20 and 60% ammonium sulfate treatments were carried out to precipitate proteins as described previously (53). The protein resuspended in standard buffer containing 20% ammonium sulfate was applied to a Phenyl Sepharose 6 Fast Flow Column (24 by 2.5 cm; Pharmacia LKB Biotechnology Inc.), washed with 20% ammonium sulfate buffer, and eluted with a linear gradient of 20 to 0% ammonium sulfate in standard buffer. Protein fractions were evaluated for purity of overexpressed AhpC by SDS-PAGE, and the purest fractions were pooled. After dialysis against 10 mM potassium phosphate buffer (pH 7.0), the protein was loaded onto a DEAE-cellulose column (24 by 2.5 cm; Whatman DE52) and eluted with a linear gradient of 10 to 100 mM potassium phosphate (1-liter total volume). Again, fractions were analyzed for AhpC by SDS-PAGE and pure fractions were pooled, concentrated, and aliquotted for storage at 20°C.

Purification of recombinant Trx1. A 10-liter culture of E. coli XL-1 Blue harboring the pPROK1/trx1 plasmid was grown in the fermentor and induced as described above for AhpC expression. The crude cell extract obtained as described above was treated with streptomycin sulfate, and following centrifugation, the supernatant was subjected to heat denaturation at 70°C for 4 min to remove contaminating proteins. Denatured proteins were removed via centrifugation at 23,000 × g for 20 min, and the supernatant was dialyzed overnight in three changes (6 liters each) of 5 mM potassium phosphate buffer (pH 7.0). The dialyzed protein was applied to a DEAE-cellulose column (24 by 2.5 cm; Whatman DE52) pre-equilibrated with 10 mM potassium phosphate (pH 7.0). Trx1 eluted upon application of a linear gradient of 10 to 100 mM potassium phosphate (1-liter total volume) and was identified by SDS-PAGE. Pooled fractions were concentrated to 2 ml, and the concentrate was applied to a BioGel A-0.5 m agarose column (Bio-Rad) equilibrated in the standard buffer. The fractions were collected, assessed for purity using SDS-PAGE, pooled, and aliquotted for storage at 20°C. Further molecular weight and purity analysis required the use of Tris-Tricine gels as described by Ausubel et al. (3).

Purification of recombinant TrxR. A 10-liter culture of XL-1 transformed with the pPROK1/trxR plasmid was grown, induced, and harvested; the crude extract was prepared essentially as described for AhpC. After the 60% ammonium sulfate treatment, the resuspended and dialyzed protein was loaded onto the DEAE-cellulose column and eluted as described for Trx1. Purified fractions, as assessed by the ratio of A₂₈₀ to A₄₅₀, were pooled and loaded onto an Affi-Gel Blue column (24 by 2.5 cm; Bio-Rad). Fractions were eluted with a linear gradient of 25 mM potassium phosphate buffer to 600 mM potassium phosphate and 0.3 M NaCl (pH 7.0; 700-ml total volume). Protein purity was assessed by A₂₈₀/A₄₅₀ ratio determination and SDS-PAGE. Pure fractions were pooled, dialyzed in standard buffer, and aliquotted for storage at 20°C.

Purification of recombinant Trx2. A 10-liter culture of XL-1 transformed with the pPROK1/trx2 plasmid was grown under the conditions described for Trx1. Bacteria containing Trx2 were treated as described for Trx1, except for the following modifications. Instead of heat denaturation, the supernatant from an initial treatment with 50% ammonium sulfate was brought to 80%. The pelleted protein was resuspended in 5 mM potassium phosphate buffer, dialyzed overnight in the same buffer, and applied to the DEAE-cellulose column pre-equilibrated with 5 mM potassium phosphate buffer (pH 7.0). The nonbinding fraction was then applied to a carboxymethyl cellulose column (24 by 2.5 cm; Whatman CM) pre-equilibrated with 5 mM potassium phosphate buffer (pH 7.0); this was followed by elution with a linear gradient of 5 to 60 mM potassium phosphate (700-ml total volume). SDS-PAGE was again used to assess purity, and concentrated protein was aliquotted for storage at 20°C in standard buffer.

Other protein purifications. Purification of E. coli TrxR and Trx1 was done essentially as described by Poole et al. (54). S. typhimurium AhpF and AhpC were purified as reported previously (53).

Sequence analysis. AhpC homologues were identified through BLAST (2) searches with the H. pylori AhpC sequence. Pairwise comparisons of the H. pylori AhpC sequence were generated with AhpC sequences from other organisms using the BLAST Two Sequences program at the National Center for Biotechnology Information website (http://www.ncbi.nlm.nih.gov/).

Spectroscopic experiments. Most spectral assays were carried out using a thermostatted Gilford 220 updated recording spectrophotometer with a Beckman DU monochromator and a Kipp and Zonen (Delft, The Netherlands) chart recorder. Extinction coefficient determination using microbiuret assays for proteins at 280 nm and for bound flavin at 450 nm, thiol content determination using 5,5'-dithiobis(2-nitrobenzoic acid) (DTNB), and disulfide detection using 2-nitro-5-thiosulfobenzoate (NTSB) were all performed as described previously (51, 53). Further quantification of H. pylori proteins relied on the following extinction coefficients experimentally obtained at 280 nm (except where noted): AhpC, 24,800 ± 300 M¹ cm¹; Trx1, 17,200 ± 1,000 M¹ cm¹; Trx2, 23,300 ± 300 M¹ cm¹; TrxR, 11,900 ± 110 M¹ cm¹ (450 nm). To quantitate the concentration of other proteins and reagents by absorbance, the following extinction coefficients were used: E. coli TrxR, 11,300 M¹ cm¹ (454 nm) (73); E. coli Trx1, 13,700 M¹ cm¹ (280 nm) (31); S. typhimurium AhpF, 13,100 M¹ cm¹ (450 nm); AhpC, 24,300 M¹ cm¹ (280 nm) (53); TNB², 14,150 M¹ cm¹ (412 nm) (58); NADPH, 6,200 M¹cm¹ (340 nm); NADH, 6,220 M¹ cm¹ (340 nm).

Determination of H₂O₂ content using ferrithiocyanate. To monitor the disappearance of hydrogen peroxide in solution, experiments were conducted essentially as described by Thurman et al. (66), except that peroxidase reactions (0.5 ml of AhpC [20 µM], Trx1 [2.0 µM], and TrxR [0.2 µM]) were initiated with the addition of H₂O₂ (1 mM). Reactions mixtures were incubated at 37°C, and aliquots were removed at various time points up to 30 min. Reactions were terminated by the addition of 0.5 ml of 12.5% trichloroacetic acid, precipitated proteins were removed by centrifugation, and peroxide content was analyzed by ferrithiocyanate complex formation [0.2 ml of 10 mM Fe(NH₄)₂(SO₄)₂ and 0.1 ml of 2.5 mM KSCN were added to a 50-µl aliquot of supernatant]. Red ferrithiocyanate complex was measured at 480 nm, and peroxide concentration was determined from a standard curve with H₂O₂.

AhpC activity assays and reductase system determination. All AhpC activity assays were conducted on an Applied Photophysics DX.17 MV stopped-flow spectrophotometer at 25°C, except where noted otherwise. Aerobic, steady-state AhpC activity assays were monitored by following the decrease in A₃₄₀ due to NADPH oxidation. All peroxidase assays were conducted with 50 mM potassium phosphate buffer (pH 7.0)-0.1 M ammonium sulfate-0.5 mM EDTA. Proteins in one syringe were mixed with substrates in the other syringe, except where indicated otherwise. The rate of change in A₃₄₀ was assessed by linear regression analysis of the first 10% of the linear region of the resulting trace. Preparation of the reaction chamber for anaerobic assays was conducted essentially as described previously (44).

Steady-state kinetic analysis of TrxR with Trx1. K_m and k_cat measurements of TrxR for NADPH and Trx1 were conducted essentially as previously described (41), except that they were carried out on the stopped-flow spectrophotometer. Assays contained 0 to 40 µM NADPH (0.2 µM at the lowest concentration), 0 to 100 µM Trx1, and 34 nM TrxR (final concentrations are indicated), and activity was followed by monitoring the change in NADPH fluorescence over time (excitation at 340 nm and emission recorded at 90° using a 400-nm filter). The decrease in fluorescence over time was converted to absorbance units per minute using a standard curve. Primary rate data were fitted to a rectangular hyperbolic function obtained using the Marquardt-Levenberg curve-fitting algorithm in SigmaPlot (Jandel Scientific, San Rafael, Calif.). The rate data obtained after varying each substrate were transformed and displayed in a primary Hanes-Wolf plot and yielded lines that intersected on the y axis, indicating a substituted mechanism for TrxR. A substituted (ping-pong) mechanism was then assumed for all equations, and the initial rate in the absence of products was represented by the following equation (15): v = Vab/(K_mBa + K_mAb + ab), where a is the concentration of Trx1, b is the concentration of NADPH, V is the maximum velocity, and K_m is the Michaelis constant. Slopes and y intercepts obtained from the linear regression lines for each concentration of Trx1 were replotted in a secondary Hanes-Wolf plot to obtain the true V_max and K_m for each substrate. The y intercept of the secondary plot was equivalent to K_mB/V, and the slope was equal to 1/V. To solve for K_mA, the following relationship was used: V(app)/K_m(app) = V/K_mA, where the V(app)/K_m(app) ratio was obtained from Michaelis-Menten plots of the initial-rate data.

2

Bisubstrate steady-state kinetic analysis of AhpC. Aerobic peroxidase assays monitoring the A₃₄₀ change of NADPH were carried out on a stopped-flow spectrophotometer as described above. In one syringe, peroxide substrates (10 to 40 µM) were incubated in the peroxidase assay buffer. (All concentrations shown are final concentrations achieved upon mixing.) In the other syringe, AhpC (0 to 2 µM), Trx1 (20 µM), and E. coli TrxR (2 µM) were incubated with NADPH (150 µM) for 5 min prior to assay with the peroxide substrate. (Use of E. coli TrxR rather than H. pylori TrxR in this assay did not significantly affect the results but did greatly decrease the problem of AhpC-independent NADPH oxidation.) The kinetic coefficients were obtained as described by Forstrom et al. (24). Data were analyzed using the integrated Dalziel rate equation for a two-substrate enzymatic system in which enzyme-substrate complexes were not experimentally observed: E₀(t/[ROH]_t) = ₁ ln([ROOH]_t/([ROOH]₀[ROH]_t)/[ROH]_t) + ₂/[Trx1] + ₀, where [ROOH]₀ is the initial hydroperoxide concentration, [ROH]_t is the product (alkyl alcohol) concentration at time t, and ₁, ₂, and ₀ are the Dalziel coefficients. Data were chosen and analyzed as described previously for tryparedoxin peroxidase from Crithidia fasciculata (46). Linoleic acid hydroperoxide (LOOH) was prepared by enzymatic oxidation with soybean lipoxygenase as described by Maiorino et al. (43).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Amino acid sequence alignments. In an amino acid sequence comparison of the putative AhpC from H. pylori with the more extensively studied S. typhimurium enzyme, Cys49 in H. pylori AhpC aligns perfectly with the essential, conserved Cys46 from S. typhimurium AhpC (21). Cys49 in H. pylori AhpC is, therefore, most likely the site of interaction with peroxides and the site of sulfenic acid formation. The H. pylori AhpC amino acid sequence was also compared to the deduced amino acid sequences for other 1- and 2-Cys AhpC homologues from a wide range of organisms (Table 2). Of all the known homologues, H. pylori AhpC shares the most sequence identity with a select group of bacterial Prxs including AhpC from Campylobacter jejuni (67%), TPx from Rickettsia prowazekii (54%), and AhpC from Legionella pneumophila (52%). Other than the proteins from these three bacterial sources, however, H. pylori AhpC is more similar to 2-Cys Prx protein sequences from higher organisms than to other bacterial AhpC proteins (Table 2). Of the sequences shown, S. cerevisiae Prx (2-Cys), C. elegans Prx, the human proliferation-associated gene (2-Cys), and Synechocystis sp. strain PCC6803 (a cyanobacterium) TPx have all been shown experimentally to be recycled by a Trx-reducing system (10, 36, 52).

Characterization of purified H. pylori proteins. AhpC, Trx1, Trx2, and TrxR expressed from pPROK1 in the respective E. coli strains were purified to homogeneity (Fig. 1). All proteins were obtained in high yields as soluble proteins after induction with IPTG at 37°C. Migration of the H. pylori proteins was compared to that of pure S. typhimurium or E. coli homologues during SDS-PAGE, which allowed the detection of each protein during purification and assessment of purity. Pure, reduced AhpC corresponded to an apparent molecular mass of 26 kDa, which is slightly higher than its expected molecular mass of 22,235 Da. When prepared in nonreducing sample buffer and analyzed by SDS-PAGE, AhpC migrated as a 50-kDa protein, indicating that the purified protein is oxidized and contains one or more intersubunit disulfide bonds (Fig. 1A, lane 6). TrxR migrated as a 33-kDa protein on SDS-PAGE under both reducing and nonreducing conditions (Fig. 1A, lanes 5 and 7; expected mass, 33,538 Da). Previously, an intersubunit disulfide bond between TrxR subunits was suggested by Windle et al. (74); our results, however, rule out such a linkage between subunits. Spectral analyses (data not shown) of TrxR revealed a strong A₄₅₄, which is typical of enzymes containing a bound flavin cofactor. Trx1 and Trx2 both migrated as monomers in the presence or absence of -mercaptoethanol and gave masses of approximately 12 kDa, close to their expected masses of 11,854 and 11,744 Da, respectively.

View larger version (56K): [in this window] [in a new window]   FIG. 1.   SDS-PAGE analysis of purified AhpC, Trx1, Trx2, and TrxR from H. pylori. (A) Crude lysates and recombinant purified AhpC and TrxR were analyzed on the same 12% polyacrylamide gel in reducing sample buffer (except where noted) as follows: lane 1, molecular mass markers (Broad Range Molecular Weight Standards; Bio-Rad); lane 2, crude extracts of E. coli cells transformed with pPROK1/ahpC and induced with 0.4 mM IPTG for at least 3 h at 37°C; lane 3, pure, recombinant AhpC; lane 4, crude extracts of E. coli cells transformed with pPROK1/trxR and induced as described above; lane 5, pure, recombinant TrxR protein; lane 6, pure AhpC in nonreducing sample buffer; lane 7, pure TrxR in nonreducing sample buffer. (B) Crude lysates and recombinant Trx1 and Trx2 were analyzed on a 10% Tris-Tricine gel as follows: lane 1, molecular mass markers; lane 2, crude extracts of E. coli cells transformed with pPROK1/trx1 after induction with IPTG; lane 3, pure, recombinant Trx1; lane 4, crude extracts of E. coli cells transformed with pPROK1/trx2 after induction with IPTG; lane 5, recombinant purified Trx2 after purification. Equivalent protein masses (10 µg) were loaded in all lanes of both gels. Molecular masses (in kilodaltons) are indicated on the left.

Kinetic characterization of Trx1, Trx2, and TrxR. H. pylori TrxR and Trx1 were shown in a series of assays to act as a general disulfide reductase system analogous to their counterparts from E. coli. When assayed with small-molecule (DTNB) or protein (insulin) disulfide-containing substrates, both Trx1 and Trx2 from H. pylori exhibited TrxR-dependent reductase activities that were hyperbolically dependent on the Trx1 and Trx2 concentrations. Insulin reduction in vitro by TrxR-Trx1 gave a V_max(app) of 19.9 ± 1.4 µM min¹ and a K_m(app) for Trx1 of 13.4 ± 2.7 µM (Fig. 2), while the TrxR-Trx2 system gave a V_max(app) of 10.6 ± 1.5 µM min¹ and a K_m(app) for Trx2 of 11.0 ± 1.9 mM (Fig. 2). While Trx2 exhibited about half of the turnover rate of Trx1 in insulin assays, both Trx1 and Trx2 reduced the small, non-protein disulfide substrate DTNB with about the same catalytic parameters as the TrxR-Trx1-insulin system (Fig. 2). In TrxR-Trx1-insulin experiments in which NADPH was replaced with NADH, no decrease in the NADH A₃₄₀ was observed upon addition of TrxR to the reaction mixture, demonstrating the high specificity of TrxR for NADPH. H. pylori TrxR and Trx1 were also capable of forming an efficient reducing system when mixed with their E. coli counterparts (data not shown). Using the DTNB-linked TrxR assay, reaction mixtures containing one H. pylori reductase component mixed with a corresponding E. coli partner protein gave catalytic efficiencies (k_cat/K_m for Trx1) which were just a fewfold lower for the heterologous systems compared with the natural systems. The fact that the H. pylori TrxR and Trx1 proteins can interact efficiently with the E. coli proteins suggests that the H. pylori proteins share a great deal of structural similarity with their E. coli counterparts and also illustrates the functional homology H. pylori Trx1 and TrxR share with other TrxR systems.

View larger version (32K): [in this window] [in a new window]   FIG. 2.   Comparative kinetic parameters of Trx1 and Trx2 from H. pylori. Trx activity assays using either 200 µM DTNB (black bars) or 80 µM insulin (gray bars) as a substrate for 0 to 50 µM Trx1 or Trx2 were conducted in a standard buffer of 100 mM potassium phosphate and 2 mM EDTA, pH 7.4, with 150 µM NADPH. Assays were started with the addition of 7.0 nM TrxR and were monitored at 340 nm when insulin was used as the substrate or at 412 nm when DTNB was used as the substrate to observe the release of TNB2. Rates were determined from the first 10% of the reaction and then fitted to a hyperbola to determine the Vmax(app) and Km(app) values for each.

View larger version (18K): [in this window] [in a new window]   FIG. 3.   Steady-state kinetic analysis of Trx1 and TrxR from H. pylori. Assays were conducted using a stopped-flow spectrophotometer that monitored the decrease in NADPH fluorescence as described in Materials and Methods. NADPH concentrations (0.8 to 34 µM) were varied at fixed concentrations of Trx1 in the presence of 0.1 µM TrxR. (A) Representative primary Hanes-Wolf plot of one set of initial-rate data for 2 µM (closed circles), 5 µM (open circles), 10 µM (closed triangles), 20 µM (open triangles), and 60 µM (closed squares) Trx1. (B) A replot of the averaged data from three separate experiments was used to calculate the steady-state kinetic parameters for H. pylori TrxR as summarized in Table 3.

AhpC cysteine thiol and disulfide quantification. In previous work, our laboratory has demonstrated that peroxidase activity in S. typhimurium AhpC is reliant on an essential, conserved cysteine residue (Cys46) while a second active-site cysteine (Cys165') contributed by a different AhpC subunit stabilizes the oxidized protein through the formation of an intersubunit disulfide bond (21). Thiol quantification of reduced AhpC revealed the presence of two cysteine thiol groups per monomer (2.10 ± 0.17). Denaturation of reduced H. pylori AhpC did not change the thiol titer, indicating a high degree of accessibility of the cysteine thiol groups in the reduced protein. As isolated from E. coli, overexpressed H. pylori AhpC was in its oxidized form and lacked free thiol groups; NTSB assays revealed that oxidized AhpC contained one disulfide bond per monomer (0.90 ± 0.07). Taken together, along with the presence of intersubunit disulfide bonds in nonreducing SDS-PAGE gels of AhpC (see above), these results support a head-to-tail arrangement of monomers to form two active sites per dimer in H. pylori AhpC, as is the case with S. typhimurium AhpC.

Peroxidase activity of AhpC, Trx1 or Trx2, and TrxR. To test the ability of H. pylori AhpC to reduce hydroperoxides with H. pylori Trx1 and TrxR acting as the reducing system, the proteins were mixed with NADPH and cumene hydroperoxide and the change in A₃₄₀ was monitored. The maximal sustained rate of NADPH oxidation was observed when all three proteins, TrxR, Trx1, and AhpC, were included in the assay mixture (Fig. 4). Alone, TrxR possessed oxidase activity (Fig. 4, long, dashed lines), which was observed as a steady decrease in A₃₄₀ at 0.70 s¹ in the absence of peroxide; E. coli TrxR exhibited a much slower rate of NADPH oxidation (0.01 s¹) under these conditions. In the presence of peroxide, NADPH oxidation by H. pylori TrxR alone increased to 0.80 s¹, which is indicative of weak peroxidatic activity of TrxR. In the absence of AhpC, a small burst in NADPH oxidation was first observed (between 0.3 and 1 s) at a rate of 5.1 s¹ relative to the TrxR concentration, which tapers off to a rate similar to that of TrxR alone (0.9 s¹) once all of the Trx1 had been reduced (reduction of 5 µM Trx1 accounts for a decrease in A₃₄₀ of 0.03). With all three H. pylori proteins present, an initial burst of NADPH oxidation (~6.6 s¹ relative to TrxR), at a rate similar to that of TrxR and Trx1 alone, was observed. The somewhat slower but sustained rate observed between 10 and 20 s (2.5 s¹) was not significantly changed when AhpC concentrations were varied, suggesting that both Trx1 reduction by TrxR and AhpC reduction by Trx1 were partially rate limiting under these conditions.

View larger version (15K): [in this window] [in a new window]   FIG. 4.   AhpC activity assays with H. pylori AhpC, TrxR, and Trx1. NADPH oxidation was monitored at 340 nm on a stopped-flow spectrophotometer for assay mixtures containing TrxR (0.5 µM), Trx1 (5 µM), and AhpC (20 µM; solid line). Other assays were conducted similarly but in the absence of AhpC (short, dashed line), in the absence of Trx1 (dotted line), or in the absence of both AhpC and Trx1 (long, dashed line). NADPH and CHP (150 µM and 1 mM, respectively, after mixing) were incubated in one syringe with peroxidase assay buffer, and assays were initiated when substrates were mixed with proteins incubated in the second syringe with peroxidase buffer.

1

1

View larger version (17K): [in this window] [in a new window]   FIG. 5.   AhpC activity assay with Trx1 or Trx2 as the reductant. The decrease in NADPH absorbance was monitored on a stopped-flow spectrophotometer when AhpC (20 µM) and TrxR (0.5 µM) were assayed with Trx1 (5.0 µM, solid line) or with Trx2 (5.0 µM, dotted line; 10 µM, long dashes) in peroxidase assay buffer as described in the legend to Fig. 4. Assays of TrxR-Trx1 excluding the AhpC protein (dashed-dotted line) and assays of TrxR alone (small dashes) were also conducted. AhpF (0.5 µM) from S. typhimurium was included in place of TrxR-Trx1 with H. pylori AhpC (medium dashes), and in this case, NADH rather than NADPH was used as the reducing substrate under anaerobic conditions.

1

View larger version (16K): [in this window] [in a new window]   FIG. 6.   Peroxide consumption by H. pylori AhpC. The disappearance of peroxide was monitored by ferrithiocyanate complex formation following incubations of a mixture containing 20 µM AhpC (triangles) or 0 µM AhpC (circles) with TrxR (0.2 µM), Trx1 (2.0 µM), and an NADPH-regenerating system consisting of glucose-6-phosphate (10 mM), glucose-6-phosphate dehydrogenase (0.2 U/ml), and NADPH (10 µM) in 500-µl volumes. All reactions were started with the addition of H2O2 (1 mM) as described in Materials and Methods.

Specificity of H. pylori AhpC for TrxR-Trx1 or AhpF-like reductase systems. In S. typhimurium and most other bacterial systems, the AhpC component is reduced by a specialized flavoprotein related to TrxR and known as AhpF, NADH oxidase, Nox-1, or PrxR (55). In a set of experiments designed to test the specificity of H. pylori AhpC for its own reductase system, S. typhimurium AhpF replaced H. pylori Trx1 and TrxR in the stopped-flow peroxidase assay mixture with H. pylori AhpC. No significant rate of NADH oxidation was observed in an anaerobic S. typhimurium AhpF and H. pylori AhpC system (Fig. 5, medium dashes). Higher concentrations of AhpF still showed no activity with H. pylori AhpC (data not shown), indicating a clear specificity of the cysteine-based peroxidase for reduction by Trx1 rather than by AhpF from a different bacterial source. S. typhimurium AhpC (10 µM) also exhibited considerable specificity for its own reductase, AhpF (0.5 µM), compared with reduction by E. coli TrxR (0.5 µM) plus E. coli Trx1 (5 µM); turnover rates were 42 s¹ with AhpF and 1.7 s¹ with E. coli TrxR-Trx1 (data not shown). Nonetheless, the rate of turnover of S. typhimurium AhpC with the E. coli TrxR-Trx1 system was only about twofold lower than that of H. pylori AhpC with its own TrxR-Trx1 system (3.2 s¹ under the same conditions) while H. pylori AhpC and S. typhimurium AhpF interaction was undetectable. Among the proteins under investigation, H. pylori TrxR and Trx1 and E. coli TrxR and Trx1 were the most interchangeable in peroxidase assays. H. pylori AhpC (2 µM) assayed with the E. coli proteins TrxR (2 µM) and Trx1 (25 µM) exhibited a rate of NADPH oxidation that was about the same as that obtained with H. pylori TrxR-Trx1 under the same conditions (1.9 versus 3.2 s¹).

Steady-state kinetics of AhpC. To further investigate the peroxidase activity of H. pylori AhpC, reaction conditions were first established under which initial rates were directly proportional to AhpC at a Trx1 concentration (30 µM) that was at least 10-fold higher than the maximal concentration of AhpC (3 µM). Because the very low intrinsic NADPH oxidase activity of E. coli TrxR allowed the observation of low peroxidase rates above background NADPH turnover, E. coli TrxR replaced H. pylori TrxR using concentrations that were high enough (2 to 3 µM) to support rapid H. pylori Trx1 recycling (i.e., additional TrxR did not further increase observed rates of NADPH oxidation). AhpC-dependent rates of NADPH oxidation measured over a range of Trx1 concentrations (15 to 30 µM) suggested a simple bimolecular interaction between reduced Trx1 and oxidized AhpC at a rate of 1.0 × 10⁵ M¹ s¹ (data not shown).

View larger version (21K): [in this window] [in a new window]   FIG. 7.   Kinetic analysis of AhpC from H. pylori by the Dalziel method. Assays of AhpC were conducted on a stopped-flow spectrophotometer using AhpC (4.0 µM), Trx1 (2.0 µM), and E. coli TrxR (1.0 µM) with NADPH (150 µM) and limiting amounts of peroxide (10 to 40 µM). Proteins were incubated with NADPH for 5 min prior to mixing with peroxide. The resulting reaction traces were used to calculate [ROOH] and [ROH] over time. (A) In the Dalziel primary plot, a least-squares linear regression line was obtained for the calculated data points for five different concentrations of Trx1: 4.0 µM (closed triangles), 2.0 µM (closed circles), 1.6 µM (open triangles), and 1.0 µM (open circles). The resulting lines give slopes of 1 and intercepts of 2/[Trx1]. (B) The apparent maximum velocities (y intercepts from the primary plot) were replotted in a Dalziel secondary plot against the reciprocal of the [Trx1] at which they were obtained. The slope of the resulting linear regression line is 2.

1

1

1

1

1

1

1

1

1

1

1

1

1

1

Essentiality of H. pylori ahpC. Allelic replacement mutagenesis of ahpC was conducted to study the physiological role of AhpC in H. pylori and to determine the effect of AhpC removal on H. pylori's responses to oxidative stress. When ahpC interrupted with the camR cassette was introduced into H. pylori strains HP26695, SS1, HP1061, and HP1, there was no growth of colonies on selective medium after 7 days of incubation, whereas a few thousand colonies are normally obtained after 2 to 3 days for nonessential-gene knockouts carried out in this manner (e.g., with the rdxA knockout control). While single crossover events have been described in H. pylori using this method of transformation (57), no such phenomenon was observed even after several repeated transformation attempts. A similar construct was successfully generated in which the nitroreductase (rdxA) gene was knocked out with the same camR cassette (26) and introduced by transformation and homologous recombination into the same H. pylori strains. Both Cam^r and Mtz^r (metronidazole-resistant) colonies were formed, indicating that rdx is not essential for growth. Therefore, the loss of growth under microaerobic conditions with the loss of ahpC strongly suggests an essential role for AhpC in H. pylori viability.

Genetic characterization of ahpC. The gene encoding AhpC is located between ceuE (HP1562, encoding an iron III ABC transporter) transcribed in the opposite direction and a gene encoding an outer membrane protein (HP1564) in the same orientation as ahpC and containing endogenous promoter sequences (71). The location of the transcriptional start site of ahpC was determined by primer extension analysis. Two prominent reverse transcript bands migrate parallel to G and T residues, 96 and 94 bp upstream, respectively, from the AUG translation initiation codon (Fig. 8A). Sequences resembling the H. pylori ⁷⁰ consensus sequence (25) were located at the expected distance from the transcriptional start site (Fig. 8B, bold). The potential 10 hexamer of the putative ahpC promoter, TATACT, displays a high degree of identity (5 of 6 bp) to the 10 consensus sequence for H. pylori, TAtaaT (25). Other putative regulatory sequence elements were identified in the promoter region centered around 40 and +10 (underlined, Fig. 8B), underscoring the possibility that ahpC transcription is regulated. However, no Fur (ferric uptake regulator) binding site (69) was evident in the H. pylori ahpC promoter region and the H. pylori genome lacks a structural gene for OxyR, the redox-sensitive transcription factor which upregulates AhpC production in response to oxidative stress in other eubacteria (65). Although the deduced H. pylori AhpC amino acid sequence has already been published, the protein was not identified as an alkyl hydroperoxide reductase and the translational start site of ahpC was not correctly positioned by O'Toole et al. (48).

View larger version (30K): [in this window] [in a new window]   FIG. 8.   Primer extension analysis of ahpC. (A) To map the transcriptional start site of H. pylori AhpC mRNA, a [32P]ATP-labeled oligonucleotide complementary to the 5' end of ahpC was hybridized to 100 µg of total RNA and extended using reverse transcriptase. DNA sequencing reactions carried out with the same primer (right) were electrophoresed concomitantly with the primer extension products (left) to the left on a 6% urea polyacrylamide sequencing gel. Two potential 5' ends of the ahpC transcript are 96 (G) and 94 (T) bp upstream from the AUG translation initiation codon. The potential 10 hexamer of the putative ahpC promoter is indicated in bold upstream of the transcriptional start site. (B) Shown is the DNA sequence of the region upstream of ahpC from H. pylori strain 26695. The potential ahpC transcription and translation start signals are shown in bold, as well as the Shine-Dalgarno (SD) ribosome binding site and the putative promoter sites centered at 10 and 35. The inverted repeats are indicated by arrows.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Antioxidant systems critical to the defense of H. pylori against ROS generated by the oxidative burst of macrophages and polymorphonuclear leukocytes are central to the ability of this organism to establish chronic infections in gastric tissues and to combat the high degree of inflammatory responses mounted by the host. To date, Fe-dependent superoxide dismutase (63), catalase (47), and the Trx-dependent peroxidase (AhpC) described herein have been characterized as H. pylori antioxidant enzymes. Deletion of the gene encoding catalase (katA) from the H. pylori chromosome did not affect the viability of the organism (71), whereas we have found that ahpC is essential for H. pylori survival under microaerobic conditions. Previous studies with other organisms in which the ahpC locus was deleted or mutated have not demonstrated AhpC essentiality. For example, in studies of the closely related, but more aerotolerant, enteric pathogen C. jejuni, deletion of sodB decreased its capacity to survive in macrophages (50), while deletion of ahpC increased susceptibility to oxidative stress (4); nonetheless, viability under conditions of low oxygen tension was not affected for either mutant. In support of our findings, two other groups have recently reported attempts to knock out H. pylori ahpC and have reached similar conclusions regarding its essentiality (42, 13). H. pylori sensitivity toward oxidative damage is highlighted by its dependence on microaerophilic growth conditions. Its inability to grow in the absence of ahpC is a clear reflection of the delicate redox balance required to support, yet not inactivate, key metabolic enzymes (37).

Given the apparent requirement of TrxR-Trx1 for AhpC activity, the TrxR-Trx1 reductase system is also likely to be indispensable for H. pylori viability, as well, although this hypothesis has yet to be tested. The alternative, that an as-yet-unidentified reductase is also capable of AhpC reduction, is also a possibility. Interestingly, no homologues for the E. coli glutathione reductase system, such as glutaredoxin (grxA), -L-glutamyl-L-cysteine synthetase (gshA), or glutathione reductase (gorA), can be found in the H. pylori genome. Therefore, H. pylori does not have the main other reductase system that can serve as a compensatory mechanism in most other organisms (56), highlighting the possible fundamental requirement for a functional TrxR-Trx1 system in H. pylori.

Our demonstration that the 26-kDa protein previously described as an abundant "species-specific antigen" by O'Toole et al. (48) is a cysteine-based peroxidase reactivated by reduced Trx agrees nicely with the sequence-based comparisons of H. pylori AhpC with a wide variety of other Prxs. A subset of these proteins, some of which have been classified as TPxs, are not reducible by bacterial AhpF proteins and rely on reducing equivalents from Trx to support turnover with cellular peroxides (10, 33, 36, 52). We have found that this capacity to accept electrons from reduced Trx is a property common to bacterial AhpC proteins as well, which nonetheless show much greater reactivity toward their specialized flavoprotein reductase, AhpF. Some eukaryotic Prxs, on the other hand, are incapable of turnover with reduced Trx proteins, at least under the conditions tested (22, 35). The specificity of such cysteine-based peroxidases for reduced Trx, as implied by the TPx designation, is therefore of questionable validity as a unique functional description distinguishing this group among the diverse Prx proteins.

In cases in which the reactivation of Prx proteins by their electron donor proteins has been investigated in detail, two different kinetic patterns have been observed. For AhpF-AhpC interactions, reduction of AhpC by AhpF is a saturable phenomenon characterized by a K_m for AhpC of around 15 µM and a k_cat/K_m for the flavoprotein of ~10⁷ M¹ s¹ (55). For the Prx system from C. fasciculata, interaction of reduced tryparedoxin with its peroxidase is a bimolecular process with infinite k_cat and K_m values and a second-order rate constant of 1.5 × 10⁶ to 3.5 × 10⁶ M¹ s¹ (46). Here, we have demonstrated that the interaction of Trx1 with H. pylori AhpC is also bimolecular (~10⁵ M¹ s¹) and that therefore no enzyme-substrate complexes between Trx1 and AhpC can be detected kinetically. Presumably, this kinetic pattern will hold for other Trx-dependent Prx proteins as well when such kinetic profiles are investigated.

Using kinetic studies akin to those previously applied to another Prx, tryparedoxin peroxidase from C. fasciculata, the interaction of H. pylori AhpC with peroxides was also shown to be a bimolecular process (like Trx-Prx interactions) lacking detectable enzyme-substrate complexes. This ping-pong mechanism has also been observed for the distantly related glutathione peroxidases (23, 68). The higher rate of enzyme-peroxide interaction for glutathione peroxidase (~10⁸ M¹ s¹) has been attributed to the unique reactivity of the selenocysteine at the active site (64), although at 10⁵ to 10⁶ M¹ s¹, the rate of peroxide reduction is still quite high and indicative of an important role for H. pylori AhpC in cellular peroxide metabolism. Using k₁' to characterize the protein-peroxide interaction, our experiments demonstrated essentially no specificity when AhpC was tested with a wide variety of small, bulky, aromatic, or lipid hydroperoxide substrates, as was true of the C. fasciculata peroxidase (46). Differential reactivities toward particular hydroperoxide substrates have been reported for some other Prx enzymes based on less quantitative analyses (9, 33, 34). In addition, peroxynitrite (OONO) has recently been shown to be a substrate for H. pylori AhpC, with the rate of decomposition occurring at a second-order rate constant of 1.21 × 10⁶ M¹ s¹ (7). These results are all consistent with a minimal binding site on AhpC for hydroperoxides (and peroxynitrite) consisting of little more than the catalytic residue (Cys49) at the active site.

On the basis of results reported here and mechanisms outlined for other 2-Cys AhpC homologues (10, 51), electrons from NADPH proceed along the path outlined in Fig. 9 for the reduction of peroxides to alcohols. This scheme is highly analogous to that for electron transfer through the S. typhimurium AhpC system, except that TrxR and Trxl replace AhpF in the H. pylori system. Nonetheless, one tightly bound flavin and three disulfide redox centers mediate electron transfer from pyridine nucleotide to peroxide in both cases.

View larger version (13K): [in this window] [in a new window]   FIG. 9.   Pathway for transfer of reducing equivalents from NADPH to hydroperoxide in the alkyl hydroperoxide reductase system from H. pylori. Note that the enzyme species shown do not necessarily represent actual catalytic intermediates. FAD, flavin adenine dinucleotide.

Little information exists on potential redox or iron regulation of H. pylori AhpC expression, although in other studies of TrxR and Trx1 from H. pylori, Windle et al. (74) observed that Trxl expression dramatically increased under conditions of oxidative stress; Trx1 was therefore classified as a stress response element in H. pylori. While the proximal location of trxR and trx1 in the chromosome could provide the bacterium with a mechanism for a coordinated response eliciting expression of both proteins, an increase in TrxR expression did not accompany increased Trx1 expression under oxidative stress conditions (74). Interestingly, reductase activity of the TrxR-Trx1 reductase system was also reportedly present in the media of culture