Characterization of the Menaquinone-dependent Disulfide Bond Formation Pathway of Escherichia coli
2004; Elsevier BV; Volume: 279; Issue: 45 Linguagem: Inglês
10.1074/jbc.m407153200
ISSN1083-351X
AutoresYoh-hei Takahashi, Kenji Inaba, Koreaki Ito,
Tópico(s)Bacterial Genetics and Biotechnology
ResumoIn the protein disulfide-introducing system of Escherichia coli, plasma membrane-integrated DsbB oxidizes periplasmic DsbA, the primary disulfide donor. Whereas the DsbA-DsbB system utilizes the oxidizing power of ubiquinone (UQ) under aerobic conditions, menaquinone (MK) is believed to function as an immediate electron acceptor under anaerobic conditions. Here, we characterized MK reactivities with DsbB. In the absence of UQ, DsbB was complexed with MK8 in the cell. In vitro studies showed that, by binding to DsbB in a manner competitive with UQ, MK specifically oxidized Cys41 and Cys44 of DsbB and activated its catalytic function to oxidize reduced DsbA. In contrast, menadione used in earlier studies proved to be a more nonspecific oxidant of DsbB. During catalysis, MK8 underwent a spectroscopic transition to develop a visible violet color (λmax = 550 nm), which required a reduced state of Cys44 as shown previously for UQ color development (λmax = 500 nm) on DsbB. In an in vitro reaction system of MK8-dependent oxidation of DsbA at 30 °C, two reaction components were observed, one completing within minutes and the other taking >1 h. Both of these reaction modes were accompanied by the transition state of MK, for which the slower reaction proceeded through the disulfide-linked DsbA-DsbB(MK) intermediate. The MK-dependent pathway provides opportunities to further dissect the quinone-dependent DsbA-DsbB redox reactions. In the protein disulfide-introducing system of Escherichia coli, plasma membrane-integrated DsbB oxidizes periplasmic DsbA, the primary disulfide donor. Whereas the DsbA-DsbB system utilizes the oxidizing power of ubiquinone (UQ) under aerobic conditions, menaquinone (MK) is believed to function as an immediate electron acceptor under anaerobic conditions. Here, we characterized MK reactivities with DsbB. In the absence of UQ, DsbB was complexed with MK8 in the cell. In vitro studies showed that, by binding to DsbB in a manner competitive with UQ, MK specifically oxidized Cys41 and Cys44 of DsbB and activated its catalytic function to oxidize reduced DsbA. In contrast, menadione used in earlier studies proved to be a more nonspecific oxidant of DsbB. During catalysis, MK8 underwent a spectroscopic transition to develop a visible violet color (λmax = 550 nm), which required a reduced state of Cys44 as shown previously for UQ color development (λmax = 500 nm) on DsbB. In an in vitro reaction system of MK8-dependent oxidation of DsbA at 30 °C, two reaction components were observed, one completing within minutes and the other taking >1 h. Both of these reaction modes were accompanied by the transition state of MK, for which the slower reaction proceeded through the disulfide-linked DsbA-DsbB(MK) intermediate. The MK-dependent pathway provides opportunities to further dissect the quinone-dependent DsbA-DsbB redox reactions. Disulfide bond formation, an important maturation process for envelope and secreted proteins, depends on dedicated cellular factors (Dsb proteins) (1Collet J.-F. Bardwell J.C.A. Mol. Microbiol. 2002; 179: 2465-2471Google Scholar, 2Kadokura H. Katzen F. Beckwith J. Annu. Rev. Biochem. 2004; 72: 111-135Crossref Scopus (444) Google Scholar). DsbA, a periplasmic enzyme of Escherichia coli, introduces disulfide bonds into newly exported proteins using its Cys30–Cys33 active-site disulfide bond. DsbB, a plasma membrane protein, re-oxidizes the DsbA cysteines to enable catalytic functioning of DsbA (3Bardwell J.C. Lee J.O. Jander G. Martin N. Belin D. Beckwith J. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 1038-1042Crossref PubMed Scopus (355) Google Scholar, 4Missiakas D. Georgopolous C. Raina S. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 7084-7088Crossref PubMed Scopus (194) Google Scholar, 5Guilhot C. Jander G. Martin N.L. Beckwith J. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 9895-9899Crossref PubMed Scopus (128) Google Scholar, 6Kishigami S. Akiyama Y. Ito K. FEBS Lett. 1995; 364: 55-58Crossref PubMed Scopus (67) Google Scholar). The ultimate source of an oxidizing equivalent for this system is oxygen under aerobic conditions (7Bader M. Muse W. Ballou D.P. Gassner C. Bardwell J.C. Cell. 1999; 98: 217-227Abstract Full Text Full Text PDF PubMed Scopus (310) Google Scholar, 8Kobayashi T. Ito K. EMBO J. 1999; 18: 1192-1198Crossref PubMed Scopus (91) Google Scholar), in which respiratory chain components mediate electron transport (9Kobayashi T. Kishigami S. Sone M. Inokuchi H. Mogi T. Ito K. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 11857-11862Crossref PubMed Scopus (206) Google Scholar). Most importantly, ubiquinone-8 (UQ8) 1The abbreviations used are: UQ, ubiquinone; MK, menaquinone; AMS, 4-acetamido-4′-maleimidylstilbene-2,2′-disulfonate; DTT, dithiothreitol; DDM, n-dodecyl β-d-maltoside; DMK8, demethylmenaquinone-8; HPLC, high pressure liquid chromatography.1The abbreviations used are: UQ, ubiquinone; MK, menaquinone; AMS, 4-acetamido-4′-maleimidylstilbene-2,2′-disulfonate; DTT, dithiothreitol; DDM, n-dodecyl β-d-maltoside; DMK8, demethylmenaquinone-8; HPLC, high pressure liquid chromatography. interacts directly with DsbB to receive electrons from the DsbA-DsbB oxidative system (7Bader M. Muse W. Ballou D.P. Gassner C. Bardwell J.C. Cell. 1999; 98: 217-227Abstract Full Text Full Text PDF PubMed Scopus (310) Google Scholar, 10Bader M.W. Xie T. Yu C.-A. Bardwell J.C.A. J. Biol. Chem. 2000; 275: 26082-26088Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar). As a more thorough introduction and discussion of the modes of DsbB function in the oxidation of DsbA, including the current controversies, has been presented in our previous study (11Inaba K. Takahashi Y.-H. Fujieda N. Kano K. Miyoshi H. Ito K. J. Biol. Chem. 2004; 279: 6761-6768Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar), we recapitulate some salient features of UQ-DsbB interaction here. Ubiquinone interacts with DsbB with a Km of ∼2.0 μm and activates its DsbA-oxidizing activity in vitro (10Bader M.W. Xie T. Yu C.-A. Bardwell J.C.A. J. Biol. Chem. 2000; 275: 26082-26088Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar). Of the two pairs of essential cysteines in the periplasmic domains of DsbB (Cys41/Cys44 in the N-terminal loop and Cys104/Cys130 in the C-terminal loop) (12Jander G. Martin N.L. Beckwith J. EMBO J. 1994; 13: 5121-5127Crossref PubMed Scopus (104) Google Scholar), the Cys41/Cys44 pair is strongly oxidized by the respiratory components in vivo (8Kobayashi T. Ito K. EMBO J. 1999; 18: 1192-1198Crossref PubMed Scopus (91) Google Scholar). Indeed, UQ1 or decylubiquinone oxidizes this pair (but not the other pair) of cysteines in vitro (13Inaba K. Ito K. EMBO J. 2002; 21: 2646-2654Crossref PubMed Scopus (77) Google Scholar, 14Regeimbal J. Bardwell J.C.A. J. Biol. Chem. 2002; 277: 32706-32713Abstract Full Text Full Text PDF PubMed Scopus (47) Google Scholar). Although the UQ-binding site on DsbB has not been directly determined (15Xie T. Yu L. Bader M.W. Bardwell J.C.A. Yu C.-A. J. Biol. Chem. 2002; 277: 1649-1652Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar), the functional importance of the region around Cys44 in this interaction has been suggested. Mutations of residues 42 and 43 affect the redox potential and reactivity of DsbB, and those of the short segment between Cys44 and the second transmembrane region impair UQ8-dependent DsbB oxidation in vivo (13Inaba K. Ito K. EMBO J. 2002; 21: 2646-2654Crossref PubMed Scopus (77) Google Scholar, 16Kadokura H. Bader M. Tian H. Bardwell J.C.A. Beckwith J. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 10884-10889Crossref PubMed Scopus (60) Google Scholar, 17Kobayashi T. Takahashi Y. Ito K. Mol. Microbiol. 2001; 39: 158-165Crossref PubMed Scopus (19) Google Scholar). In particular, the importance of the length of this segment (17Kobayashi T. Takahashi Y. Ito K. Mol. Microbiol. 2001; 39: 158-165Crossref PubMed Scopus (19) Google Scholar) as well as of Arg48 (16Kadokura H. Bader M. Tian H. Bardwell J.C.A. Beckwith J. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 10884-10889Crossref PubMed Scopus (60) Google Scholar) for the quinone-DsbB interaction has been proposed. The peculiar states of DsbB-bound UQ have been described recently. Bardwell and co-workers (18Regeimbal J. Gleiter S. Trumpower B.L. Yu C.-A. Diwaker M. Ballou D.P. Bardwell J.C.A. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 13779-13784Crossref PubMed Scopus (61) Google Scholar) suggested that UQ on DsbB constitutively forms a quinhydrone-type charge transfer complex consisting of a hydroquinone and a benzoquinone moiety and having a strong absorption at 505–510 nm. In contrast, we have shown that UQ undergoes a spectroscopic transition to exhibit an absorption peak at ∼500 nm only when it is on DsbB in certain forms (11Inaba K. Takahashi Y.-H. Fujieda N. Kano K. Miyoshi H. Ito K. J. Biol. Chem. 2004; 279: 6761-6768Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar). One of these forms is the DsbA-DsbB complex, in which DsbA and DsbB are held together by an intermolecular disulfide bond between Cys30 of DsbA and Cys104 of DsbB. Mutational analyses indicated that generation of reduced Cys44 in DsbB is primarily responsible for the color development and that the important sequence/structural features around this residue (as discussed above) are also required for the UQ "red shift." Although DsbA itself is not directly required for the UQ anomaly, a disulfide rearrangement in the DsbA-DsbB complex, as demonstrated by Kadokura and Beckwith (19Kadokura H. Beckwith J. EMBO J. 2002; 21: 2354-2363Crossref PubMed Scopus (80) Google Scholar), will generate reduced Cys44 and hence the spectroscopic transition of UQ. More specifically, engagement of Cys104 in the mixed disulfide formation with Cys30 of DsbA renders reduced Cys130, which in turn attacks the Cys41–Cys44 disulfide to form a Cys41–Cys130 disulfide and to reduce Cys44. On the basis of these observations and considerations, we proposed that the spectroscopic transition of UQ represents its activated state that drives electron transfer and resolution of the DsbA-DsbB complex in a manner overcoming the reversed redox potential difference (Refs. 13Inaba K. Ito K. EMBO J. 2002; 21: 2646-2654Crossref PubMed Scopus (77) Google Scholar and 14Regeimbal J. Bardwell J.C.A. J. Biol. Chem. 2002; 277: 32706-32713Abstract Full Text Full Text PDF PubMed Scopus (47) Google Scholar; see also Ref. 20Grauschopf U. Fritz A. Glockshuber R. EMBO J. 2003; 22: 3503-3513Crossref PubMed Scopus (52) Google Scholar for a contrasting view) between DsbA and DsbB active-site cysteine pairs (11Inaba K. Takahashi Y.-H. Fujieda N. Kano K. Miyoshi H. Ito K. J. Biol. Chem. 2004; 279: 6761-6768Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar). DsbA-dependent disulfide bond formation is not impaired in anaerobically growing E. coli cells (7Bader M. Muse W. Ballou D.P. Gassner C. Bardwell J.C. Cell. 1999; 98: 217-227Abstract Full Text Full Text PDF PubMed Scopus (310) Google Scholar, 10Bader M.W. Xie T. Yu C.-A. Bardwell J.C.A. J. Biol. Chem. 2000; 275: 26082-26088Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar, 16Kadokura H. Bader M. Tian H. Bardwell J.C.A. Beckwith J. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 10884-10889Crossref PubMed Scopus (60) Google Scholar), in which menaquinone (MK) is the prevailing quinone species (21Wallace B.J. Young I.G. Biochim. Biophys. Acta. 1977; 461: 84-100Crossref PubMed Scopus (209) Google Scholar). MK is indeed believed to play an electron-accepting role like UQ in the DsbA-DsbB system (7Bader M. Muse W. Ballou D.P. Gassner C. Bardwell J.C. Cell. 1999; 98: 217-227Abstract Full Text Full Text PDF PubMed Scopus (310) Google Scholar, 16Kadokura H. Bader M. Tian H. Bardwell J.C.A. Beckwith J. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 10884-10889Crossref PubMed Scopus (60) Google Scholar). However, MK reactivity with DsbB has not been studied systematically. The first indication for the in vivo importance of MK was provided by our observation that mutational impairment of both the UQ and MK biosynthetic pathways results in dysfunction of the DsbA-DsbB system (9Kobayashi T. Kishigami S. Sone M. Inokuchi H. Mogi T. Ito K. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 11857-11862Crossref PubMed Scopus (206) Google Scholar). Kadokura et al. (16Kadokura H. Bader M. Tian H. Bardwell J.C.A. Beckwith J. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 10884-10889Crossref PubMed Scopus (60) Google Scholar) showed that some Arg48 substitutions in DsbB result in a low activity enzyme that can no longer utilize the MK analog menadione as an in vitro electron-accepting substrate. The aim of this study was to characterize the MK-dependent pathway of oxidative protein folding. Here, we identified the MK molecular species (MK8) that associates with DsbB in the absence of UQ and showed that its interaction with DsbB is similar to that of UQ, including a striking spectroscopic transition elicited by the special states of DsbB. The in vitro reaction of DsbA oxidation with MK8 proved slower than the UQ-dependent reaction, and we characterized this reaction in some detail to gain further insights into the molecular pathways of quinone-coupled and DsbB-dependent oxidation of DsbA. Determination of in Vivo Redox States of DsbA—E. coli strains CU141 (MC4100/F′, lacIq lacPL8 lacZ+Y+A+pro+) (22Akiyama Y. Ogura T. Ito K. J. Biol. Chem. 1994; 269: 5218-5224Abstract Full Text PDF PubMed Google Scholar), SS141 (CU141, dsbB::kan5) (3Bardwell J.C. Lee J.O. Jander G. Martin N. Belin D. Beckwith J. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 1038-1042Crossref PubMed Scopus (355) Google Scholar, 23Kishigami S. Kanaya E. Kikuchi M. Ito K. J. Biol. Chem. 1995; 270: 17072-17074Abstract Full Text Full Text PDF PubMed Scopus (98) Google Scholar), TA36 (KS272, ubiA::Cm) (9Kobayashi T. Kishigami S. Sone M. Inokuchi H. Mogi T. Ito K. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 11857-11862Crossref PubMed Scopus (206) Google Scholar), and AN384 (ubiA420 menA401) (21Wallace B.J. Young I.G. Biochim. Biophys. Acta. 1977; 461: 84-100Crossref PubMed Scopus (209) Google Scholar) were grown at 37 °C in L-broth supplemented with 90 mm sodium phosphate (pH 7.5), 0.1 mm UQ1 (Sigma), and appropriate antibiotics. Exponentially growing cells were harvested by centrifugation, washed with the same medium without UQ1, and then re-inoculated into the UQ1-free medium. After overnight shaking, cells were diluted into the same medium for an additional 3 h of growth at 37 °C. Cultures were directly treated with trichloroacetic acid (5%) to "freeze" the redox states of the proteins, which were dissolved in SDS solution containing AMS and subjected to SDS-PAGE and anti-DsbA immunoblotting (9Kobayashi T. Kishigami S. Sone M. Inokuchi H. Mogi T. Ito K. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 11857-11862Crossref PubMed Scopus (206) Google Scholar). In this assay, reduced DsbA is mobility-shifted due to the AMS modification. Preparation of Various Forms of DsbB and DsbA—All of the DsbB-derived proteins used in this work contained a His6 tag attached to the C terminus, and their nonessential cysteine residues (Cys8 and Cys49) were mutated to Ala and Val, respectively (7Bader M. Muse W. Ballou D.P. Gassner C. Bardwell J.C. Cell. 1999; 98: 217-227Abstract Full Text Full Text PDF PubMed Scopus (310) Google Scholar). The DsbB variant is referred to simply as DsbB in this study. Site-directed mutagenesis was carried out using the QuickChange mutagenesis kit (Stratagene) with appropriate sets of primers. DsbB and its mutant forms were overexpressed and purified as described by Inaba et al. (11Inaba K. Takahashi Y.-H. Fujieda N. Kano K. Miyoshi H. Ito K. J. Biol. Chem. 2004; 279: 6761-6768Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar). DsbA was overexpressed and purified as described by Inaba and Ito (13Inaba K. Ito K. EMBO J. 2002; 21: 2646-2654Crossref PubMed Scopus (77) Google Scholar). The plasmid encoding DsbA(C33S) was described previously (11Inaba K. Takahashi Y.-H. Fujieda N. Kano K. Miyoshi H. Ito K. J. Biol. Chem. 2004; 279: 6761-6768Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar). To prepare DsbB without endogenously bound quinones, it was overproduced in the ubiA menA double mutant strain AN384 (21Wallace B.J. Young I.G. Biochim. Biophys. Acta. 1977; 461: 84-100Crossref PubMed Scopus (209) Google Scholar) as described by Inaba et al. (11Inaba K. Takahashi Y.-H. Fujieda N. Kano K. Miyoshi H. Ito K. J. Biol. Chem. 2004; 279: 6761-6768Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar). This quinone-free DsbB protein is referred to as DsbB(ΔQ) and contained <0.05 molar eq of quinones as determined by UV spectrum measurements before and after reduction with NaBH4 (10Bader M.W. Xie T. Yu C.-A. Bardwell J.C.A. J. Biol. Chem. 2000; 275: 26082-26088Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar). In contrast, DsbB prepared from wild-type cells contains endogenous UQ8 and is referred to as DsbB(UQ). To prepare DsbB with endogenously bound MK, it was overproduced in the ubiA- mutant strain TA36. Plasmid-bearing cells were grown first in the presence of UQ1 and then in its absence as described above, except that the final cell growth was started at a turbidity of 60 with a Klett colorimeter (filter no. 54) in the presence of 10 μm isopropyl 1-thio-β-d-galactopyranoside and continued for 4 h. This DsbB protein was then purified by standard procedures (11Inaba K. Takahashi Y.-H. Fujieda N. Kano K. Miyoshi H. Ito K. J. Biol. Chem. 2004; 279: 6761-6768Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar) and is referred to as DsbB(MK). Purified DsbA was reduced by incubation with 20 mm dithiothreitol (DTT) for 30 min, followed by gel filtration on a Sephadex PD-10 column (Amersham Biosciences) to remove DTT. DsbB was similarly reduced with DTT, but more care was required with the gel filtration steps because DsbB is prone to rapid air oxidation in the absence of DTT. Thus, we equilibrated Bio-Spin resin with degassed buffer (50 mm sodium phosphate (pH 8.0), 0.1 m NaCl, 0.1% n-dodecyl β-d-maltoside (DDM)) containing 20 mm DTT for 30 min, followed by a thorough wash with degassed DTT-free buffer before loading the DTT-reduced DsbB preparation. Identification of DsbB-bound Menaquinone—MK species were extracted from the DsbB(MK) preparation (see above) essentially as described by Bader et al. (10Bader M.W. Xie T. Yu C.-A. Bardwell J.C.A. J. Biol. Chem. 2000; 275: 26082-26088Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar). Four milliliters of a DsbB(MK) solution (6 mg/ml) was mixed with 6 ml of methanol (-20 °C) and then extracted four times with 10 ml of hexane. The combined hexane phases were evaporated to dryness under vacuum, and residual materials were dissolved in 1 ml of ethanol. A Shimadzu LCMS-2010A instrument was used to separate a 5-μl portion on a Shim-pack VP-ODS reverse-phase chromatography column by elution with methanol/isopropyl alcohol (3:1) at a flow rate of 0.2 ml/min. Peak fractions were then subjected to electrospray ionization mass spectrometry with scanning from m/z 900 to 100 for 1 s. The absorption spectra of the MK species extracted from DsbB(MK) were recorded at wavelengths of 200–400 nm using a Hitachi U-3310 spectrophotometer (1-cm light path). To quantify MK8, the extracted MK sample was dissolved in ethanol containing 0.01 volume of 1 m ammonium acetate buffer (pH 5.0), and a portion was treated with a few pieces of solid NaBH4, which should have reduced any quinone molecules within 2 min at room temperature (24Dunphy P.J. Brodie A.F. Methods Enzymol. 1971; 18: 407-461Crossref Scopus (83) Google Scholar). The extinction coefficient difference between oxidized and reduced forms of MK was assumed to be Δϵ245 = 25.84 mm-1 cm-1 (24Dunphy P.J. Brodie A.F. Methods Enzymol. 1971; 18: 407-461Crossref Scopus (83) Google Scholar) to quantify MK8. In Vitro Assays for DsbB-catalyzed DsbA Oxidation—DsbB activities were followed by several different methods. For kinetic studies, oxidation of DsbA was determined by the decrease in DsbA fluorescence (25Zapun A. Bardwell J.C. Creighton T.E. Biochemistry. 1993; 32: 5083-5092Crossref PubMed Scopus (229) Google Scholar) essentially as described by Bader et al. (26Bader M. Muse W. Zander T. Bardwell J. J. Biol. Chem. 1998; 273: 10302-10307Abstract Full Text Full Text PDF PubMed Scopus (74) Google Scholar). The initial velocity of the reaction was determined from the initial slope of the fluorescence change at 330 nm using a Hitachi F-4500 fluorescence spectrophotometer. Reaction mixtures contained 50 mm sodium phosphate (pH 8.0), 0.1 m NaCl, 0.1% DDM, 20 μm reduced DsbA, 5 nm DsbB, and the indicated concentrations of quinones when necessary. Assays were carried out at 30 °C by manual addition of DsbB. Reactions between DsbB and reduced DsbA in the presence of a quinone species were also determined by gel electrophoresis as described by Inaba et al. (11Inaba K. Takahashi Y.-H. Fujieda N. Kano K. Miyoshi H. Ito K. J. Biol. Chem. 2004; 279: 6761-6768Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar). In brief, reactions were terminated by mixing with an equal volume of 10% trichloroacetic acid for subsequent AMS modification and SDS-PAGE of the proteins involved. Protein species were stained with Coomassie Brilliant Blue. Band intensities for reduced DsbA, oxidized DsbA, partially reduced DsbB, oxidized DsbB, and the DsbA-DsbB disulfide complex were quantified using a Fuji LAS-1000 CCD image analyzer. Finally, the DsbB reactions were followed by spectroscopic changes of a quinone species. To follow absorption changes at 550 nm during the reaction between reduced DsbA and DsbB(MK), absorption spectra from 400 to 700 nm were scanned at various time points after a 1:1 mixing of DsbB(MK) (80 μm) and reduced DsbA (80 μm). The buffer solution was 50 mm sodium phosphate (pH 8.0), 0.1 m NaCl, and 0.02% DDM. MK-dependent in Vivo Oxidation of DsbA—MK is considered to be a major anaerobic quinone species (21Wallace B.J. Young I.G. Biochim. Biophys. Acta. 1977; 461: 84-100Crossref PubMed Scopus (209) Google Scholar). It is also believed to function as an immediate electron acceptor of the DsbA-DsbB redox system under anaerobic conditions (7Bader M. Muse W. Ballou D.P. Gassner C. Bardwell J.C. Cell. 1999; 98: 217-227Abstract Full Text Full Text PDF PubMed Scopus (310) Google Scholar, 10Bader M.W. Xie T. Yu C.-A. Bardwell J.C.A. J. Biol. Chem. 2000; 275: 26082-26088Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar). To mimic the anaerobic cellular quinone composition, we used an E. coli mutant defective in UQ biosynthesis. Whereas the ubiA menA double mutant defective in both the UQ and MK biosynthetic enzymes accumulated reduced DsbA (Fig. 1, lanes 4), the ubiA single mutant contained normally oxidized DsbA (lanes 3) even when cells were deprived of UQ1. The DsbA redox states did not differ significantly between aerobically (Fig. 1A) and semi-anaerobically (Fig. 1B) grown cells. Thus, irrespective of oxygen availability, MK can support the DsbA oxidation function of DsbB in vivo. DsbB Interacts with MK8 (but Not with Demethylmenaquinone-8 (DMK8)) in the Absence of UQ—E. coli contains three major quinone compounds, UQ8, MK8, and DMK8 (24Dunphy P.J. Brodie A.F. Methods Enzymol. 1971; 18: 407-461Crossref Scopus (83) Google Scholar). DsbB is purified from aerobically grown E. coli cells as a complex with UQ8 (10Bader M.W. Xie T. Yu C.-A. Bardwell J.C.A. J. Biol. Chem. 2000; 275: 26082-26088Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar). To examine whether DsbB is still complexed with a quinone species in the absence of UQ, we expressed it in cells of the ubiA mutant that had been deprived of UQ1. DsbB was purified, and the preparation was subjected to extraction with organic solvents (see "Experimental Procedures"). Upon reverse-phase column chromatography, the extracted materials gave two peaks (minor (peak I) and major (peak II)) in addition to some materials with early retention times (Fig. 2A). Electrospray ionization mass spectrometry analysis revealed that the peak I and II compounds had molecular masses of 725.45 and 739.50 Da, respectively, which coincided with those of a single Na+ adduct of DMK8 and of MK8 (Fig. 2B). The DsbB preparation contained only a trace of peak I material, suggesting that DsbB in the UQ-deprived cells was associated with MK8. This notion was supported further by measurement of the NMR spectrum of the DsbB extract, which proved to be almost identical to that of a commercially available MK analog, MK4 (data not shown). The fact that DsbB was purified as a complex with MK8 (but not with DMK8 of ∼3-fold greater cellular abundance (21Wallace B.J. Young I.G. Biochim. Biophys. Acta. 1977; 461: 84-100Crossref PubMed Scopus (209) Google Scholar)) from UQ-depleted E. coli cells indicates that DsbB specifically recognizes MK8. MK8 in ethanol shows several broad absorption peaks at 230–280 nm, the intensity and sharpness of which increase upon reduction (24Dunphy P.J. Brodie A.F. Methods Enzymol. 1971; 18: 407-461Crossref Scopus (83) Google Scholar). Fig. 2C shows the UV-visible spectra of the DsbB extract before and after reduction with NaBH4. The spectroscopic patterns observed were indeed characteristic of the oxidized and reduced forms of MK. Assuming that the extinction coefficient difference between the oxidized and reduced MK forms at 245 nm is 25.84 mm-1 cm-1 (24Dunphy P.J. Brodie A.F. Methods Enzymol. 1971; 18: 407-461Crossref Scopus (83) Google Scholar), we calculated that our DsbB preparation from the UQ-free cells contained equimolar amounts of the DsbB polypeptide and MK8. In conclusion, DsbB specifically interacts with MK8 to form a 1:1 complex. Specific Oxidation of the Cys41/Cys44 Pair in DsbB by MK—To characterize the reactivity between MK and DsbB, we first reduced DsbB(ΔQ), which had been purified as a quinone-free form from the ubiA menA double mutant cells (11Inaba K. Takahashi Y.-H. Fujieda N. Kano K. Miyoshi H. Ito K. J. Biol. Chem. 2004; 279: 6761-6768Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar), and incubated it with MK4 (a commercial product). In addition to wild-type DsbB (DsbB(CCCC)), we used its derivatives lacking the C-terminal (DsbB(CCSS)) or N-terminal (DsbB(SSCC)) pair of essential cysteines. The redox states of DsbB were determined by the modifiability with AMS, which alkylates a thiol and thereby causes mobility retardation upon SDS-PAGE. Reduced DsbB(CCCC) was converted to the fully oxidized form when it was incubated with MK4 at 30 °C for 60 min or longer (Fig. 3A, lanes 2–5). DsbB(CCSS), retaining the Cys41/Cys44 pair, was MK4-dependent and completely oxidized within 120 min (Fig. 3A, lanes 7–10), but DsbB(SSCC), retaining the Cys104/Cys130 pair, was only insignificantly oxidized (lanes 12–15). These results indicate that, like UQ (8Kobayashi T. Ito K. EMBO J. 1999; 18: 1192-1198Crossref PubMed Scopus (91) Google Scholar, 13Inaba K. Ito K. EMBO J. 2002; 21: 2646-2654Crossref PubMed Scopus (77) Google Scholar, 14Regeimbal J. Bardwell J.C.A. J. Biol. Chem. 2002; 277: 32706-32713Abstract Full Text Full Text PDF PubMed Scopus (47) Google Scholar), MK4 selectively oxidizes the Cys41/Cys44 pair of essential cysteine residues in DsbB. It was noted that MK-dependent DsbB oxidation was slower than the UQ-dependent reaction. Also, wild-type DsbB was more rapidly oxidized than DsbB(CCSS). When reduced DsbA was included in the above reactions, it was oxidized only by the wild-type DsbB protein in an MK4-dependent manner (Fig. 3B, lanes 1–4). In the experimental settings used, oxidation of DsbB appeared to precede the oxidation of DsbA, indicating that DsbB must first be oxidized by MK4 to initiate DsbA oxidation. Neither DsbB(CCSS) nor DsbB(SSCC) was active in DsbA oxidation in the presence of MK4, consistent with the essentiality of all of the cysteine residues at positions 41, 44, 104, and 130 in DsbB (12Jander G. Martin N.L. Beckwith J. EMBO J. 1994; 13: 5121-5127Crossref PubMed Scopus (104) Google Scholar, 27Kishigami S. Ito K. Genes Cells. 1996; 1: 201-208Crossref PubMed Scopus (47) Google Scholar). Thus, like UQ, MK4 activates the DsbA oxidation pathway(s), in which all of these DsbB cysteines play crucial roles. In previous studies, Bader et al. (7Bader M. Muse W. Ballou D.P. Gassner C. Bardwell J.C. Cell. 1999; 98: 217-227Abstract Full Text Full Text PDF PubMed Scopus (310) Google Scholar) and Kadokura et al. (16Kadokura H. Bader M. Tian H. Bardwell J.C.A. Beckwith J. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 10884-10889Crossref PubMed Scopus (60) Google Scholar) used menadione, an MK analog lacking the hydrophobic carbon chain, as an MK equivalent for analysis of the DsbB functions in vitro. We found, however, that the menadione reactivity was nonspecific in that it oxidized not only the Cys41/Cys44 pair, but also the Cys104/Cys130 pair in DsbB (Fig. 3C). Similar reactivity was also noted for UQ0, a UQ analog without a hydrophobic side chain. 2Y.-h. Takahashi, unpublished data. These results suggest that a hydrophobic isoprenyl chain of quinone molecules is important for their specific interaction with DsbB to create the correct intramolecular disulfide bond. MK Acts as a Competitive Inhibitor in the UQ-dependent DsbA Oxidation Reaction Mediated by DsbB—The DsbB-mediated DsbA oxidation reaction in vitro is a few orders of magnitude slower in the presence of MK4 than UQ. 3Y.-h. Takahashi and K. Inaba, unpublished data. If MK4 and UQ share a specific binding site on DsbB, MK4 may interfere with the UQ-dependent rapid DsbA oxidation reaction. Indeed, we observed such inhibit
Referência(s)