- Email: [email protected]
Biochemical properties and catalytic domain structure of the CcmH protein from Escherichia coli
Biochimica et Biophysica Acta 1824 (2012) 1394–1400
Contents lists available at SciVerse ScienceDirect
Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbapap
Biochemical properties and catalytic domain structure of the CcmH protein from Escherichia coli Xue-Ming Zheng a , 1, Jing Hong b, d, 1, Hai-Yin Li a, Dong-Hai Lin b, c, Hong-Yu Hu a,⁎ a
Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai 200031, China Shanghai Institute of Materia Medica, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai 201203, China The key Laboratory for Chemical Biology of Fujian Province, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, China d Department of Food Science and Engineering, College of Biological Science and Technology, Fuzhou University, Fuzhou 350108, China b c
a r t i c l e
i n f o
Article history: Received 20 March 2012 Received in revised form 11 June 2012 Accepted 29 June 2012 Available online 10 July 2012 Keywords: CcmH protein topology solution structure redox potential cytochrome c maturation periplasm
a b s t r a c t In the Gram-negative bacterium of Escherichia coli, eight genes organized as a ccm operon (ccmABCDEFGH) are involved in the maturation of c-type cytochromes. The proteins encoded by the last three genes ccmFGH are believed to form a lyase complex functioning in the reduction of apocytochrome c and haem attachment. Among them, CcmH is a membrane-associated protein; its N-terminus is a catalytic domain with the active CXXC motif and the C-terminus is predicted as a TPR-like domain with unknown function. By using SCAM (scanning cysteine accessibility mutagenesis) and Gaussia luciferase fusion assays, we provide experimental evidence for the entire topological structure of E. coli CcmH. The mature CcmH is a periplasm-resident oxidoreductase anchored to the inner membrane by two transmembrane segments. Both N- and C-terminal domains are located and function in the periplasmic compartment. Moreover, the N-terminal domain forms a monomer in solution, while the C-terminal domain is a compact fold with helical structures. The NMR solution structure of the catalytic domain in reduced form exhibits mainly a three-helix bundle, providing further information for the redox mechanism. The redox potential suggests that CcmH exhibits a strong reductase that may function in the last step of reduction of apocytochrome c for haem attachment. © 2012 Elsevier B.V. All rights reserved.
1. Introduction The maturation of c-type cytochromes is a post-translational modification process, characterized by covalent attachment of haem to the thiols that occurs in the CXXC motif of apocytochrome c [1]. In the Gram-negative bacterium Escherichia coli, eight genes located at the minute 47 as an operon (ccmABCDEFGH) encode the indigenous proteins that are essential for the biogenesis of cytochrome c under anaerobic condition [2–4]. Among their gene products, CcmA is an ATP-binding cassette protein forming a haem release complex together with CcmB, CcmC and CcmD [5–7]. The chaperone protein CcmE can bind haem covalently at a conserved histidine residue and carry it to the lyase CcmF, which can catalyze the formation of covalent bond between haem and reduced apocytochrome c [8–10]. Both CcmH and CcmG are characterized by the CXXC motif in the catalytic domain oriented in periplasm and act as thiol reductases for Abbreviations: AMS, 4-acetamido-4′-maleimidylstilbene-2,2′-disulfonic acid; CcmH, cytochrome c maturation protein H; CD, circular dichroism; CTZ, coelenterazine; DTT, dithiothreitol; Gluc, luciferase from Gaussia princeps; GSH, glutathione; MPB, N-(3-maleimidopropionyl) biocytin; NMR, nuclear magnetic resonance; SCAM, scanning cysteine accessibility mutagenesis; TPR, tetratricopeptide repeat ⁎ Corresponding author. Tel./fax: +86 21 54921121. E-mail address: [email protected] (H.-Y. Hu). 1 These authors contributed equally to this work. 1570-9639/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.bbapap.2012.06.017
apocytochrome c [11,12]. In other species, another protein CcmI is also involved to form a complex together with CcmF and CcmH and functions in catalyzing the final steps of haem attachment to apocytochromes c [13]. CcmH of E. coli is encoded by the last gene in the ccm operon required for the maturation of cytochrome c. The N-terminal region characterized by the CXXC motif is homologous to CycL from B. japonicum, whereas the C-terminus is the paralog of CycH [2]. The CcmH protein was proposed to be essential for the cytochrome c biosynthesis [12]. The structures of the N-terminal domain from P. aeruginosa and E. coli in the oxidized form exhibit a three-helix bundle different from the canonical thioredoxin domain [14,15]. Deletion of the C-terminus of CcmH, a putative TPR-like domain, has little impact on the maturation of cytochrome c [16]. Although the function of CcmG and CcmH has been defined as reductases of apocytochrome c, the exact mechanism in the reduction cascade is still obscure. Apocytochrome c is oxidized by the thiol oxidase DsbA in the oxidized state of periplasm and must be reduced by reductases prior to ligation with haem [17]. CcmG is reduced by DsbD, which also functions in this pathway in mediating the transport of electron from cytoplasm to periplasm [11,18]. Intriguingly, CcmG remains predominantly in the oxidized form in the mutants defective in CcmF and CcmH, indicating that CcmG functions after them in the redox cascade [19]. Since no interaction has been detected between CcmG and CcmH, the reductive steps in the assembly of c-type cytochromes remain uncertain. The
X.-M. Zheng et al. / Biochimica et Biophysica Acta 1824 (2012) 1394–1400
finding that both CcmH and CcmG can bind apocytochrome c directly increases the complexity of this problem [14,17]. As predicted, CcmH is a transmembrane redox protein with some regions in the periplasmic lumen. However, the topological structure and its functional relationship are not defined in detail. Here we report characterization of the topological structure of CcmH by using SCAM (scan cysteine availability mutagenesis) method [20] and a newly developed luciferase fusion assay [21]. In addition, biochemical and structural studies of the redox properties suggest that the CcmH protein is a strong reductase as compared with another reductant CcmG (DsbE) [11], which functions in the last reduction step prior to haem attachment in cytochrome c maturation. 2. Materials and methods 2.1. Strains, plasmids and chemicals All the single-cysteine mutations of CcmH were generated by regular PCR and cloned into plasmid pET-22b + using NdeI and XhoI restriction sites. The E. coli strain BL21 (DE3) was used as the host for all the ccmH-containing plasmids. Luciferase from Gaussia princeps (Gluc) with the deletion of signal sequence was fused to the carboxyl end in frame with different N-terminal fragments of CcmH. The fusion genes were also cloned into the plasmid pET-22b +. All mutants and fusions were verified by DNA sequencing. The luciferase substrate CTZ was purchased for Chisso Company; the reagents MPB and AMS were from Sigma or Invitrogen. Streptavidin-conjugated horseradish peroxidase and other reagents were obtained from Pierce and other comparable suppliers. 2.2. SCAM (scanning cysteine accessibility mutagenesis) assay Each plasmid harboring a single-cysteine mutant of CcmH was transformed into E. coli strain BL21 (DE3), and cultured in LB medium supplemented with ampicillin (100 μg/mL) at 37 ° C. When A600 reaches ~ 0.6, IPTG (finally 0.2 mM) was added for inducing expression of the exogenous CcmH mutants and cultured for another 4 h before harvest. After sedimentation of aliquot culture, the cell pellet was resuspended in buffer A (50 mM Hepes, pH 7.6, 250 mM sucrose, 5 mM EDTA) and labeled with MPB (75 μM) on ice for 3 min. In another parallel experiment, the same amount of resuspended pellet from the same culture was incubated with 0.1% Triton X-100 for 15 min prior to MPB labeling [20]. When specified, the pellet was incubated with 5 mM AMS for 30 min before MPB labeling. All the labeling reactions were on ice and handled carefully avoiding possible impairment of the cell. The labeling reaction was stopped by addition of 2 mM DTT and incubated on ice for 5 min. Then, the cell pellet was collected by centrifugation. In order to get rid of other soluble proteins, the cell pellet was resuspended and lysed with lysozyme. The pellet was collected through centrifugation to exclude other endogenous proteins, which are probably labeled due to the reactive cysteines. The MPB labeling products were analyzed via Western blotting with streptavidin-conjugated horseradish peroxidase. Triplicated independent experiments were performed for each singlecysteine mutant of CcmH. 2.3. Gaussia luciferase fusion assay The luminescence assay of the CcmH fragments fused with luciferase was carried out accordingly [21]. Six DNA sequences coding for different N-terminal fragments of CcmH were fused with Gluc gene using regular PCR, cloned into the pET-22b + plasmid, transformed into the BL21 (DE3) strain, and plated on the LB plate with 100 μM of ampicillin. Cells with respective plasmid were allowed to culture at 37 °C in 5-mL LB medium. When A600 reaches ~ 0.6, IPTG (0.2 mM) was added for inducing expression and the
1395
cells were cultured for another 2 h. Subsequently the tubes containing cell culture were put on ice for 30 min before activity assay. Cell suspension of 10 μL was took out and diluted to 500 μL with sterile LB medium, then 1 μM of CTZ was added and mixed, and immediately the luminescence intensity was recorded on a GloMaxTM 20/20 Luminometer. 2.4. NMR spectroscopy and structure determination The cDNA encoding for CcmH 19–99was cloned into pET-22b + vector by NdeI and XhoI restriction sites, and all the CcmH 19–99proteins were purified from E. coli BL21 (DE3) cells with Ni-ATA column, followed by Superdex G-75 gel filtration chromatography. The reduced 15N/ 13C-labeled samples were prepared with 1 mM protein dissolved in 90% H2O/10% D2O buffer containing 20 mM sodium phosphate (pH 6.5), 50 mM NaCl, 5 mM DTT, and were argon-flushed. All NMR experiments were performed at 25 °C on Varian Unity INOVA 600 MHz NMR spectrometer equipped with four RF channels and a triple‐resonance pulsed field gradient probe. A set of two-dimensional 15N- and 13C-edited HSQC spectra and three-dimensional HNCA, HN (CO) CA, HNCO, HNCACB, CBCA (CO) NH, (H) CC (CO) NH-TOCSY, (H) CCH-TOCSY and HCCH-TOCSY spectra were collected to obtain the backbone and side‐chain assignments. The three-dimensional 15N-edited and 13C-edited NOESY spectra (mixing times of 100 ms) were used to confirm the chemical shift assignments and to generate distance restraints for structure calculation. All NMR data were processed using the program NMRPipe/NMRDraw [22] and analyzed using the program Sparky [23]. The structure calculations were performed using the program ARIA2.2 [24]. Distance restraints were derived from inter-proton NOEs. Dihedral angles (ϕ and φ) were calculated from backbone chemical shifts using the program TALOS [25]. In the calculation, 200 structures were calculated in the last iteration and the 20 structures with the lowest energy were selected for refinement in water, so as to no distance violations larger than 0.3 Å. The final structures were analyzed using the program packages MOLMOL [26] and PROCHECK [27]. 2.5. Determination of the redox potential Due to no tryptophan in CcmH 19–99, we introduced a Trp residue by mutagenesis at three positions respectively, L41W, I51W and F90W, near the reactive CXXC motif based on the structure, in order to gain a suitable mutant reporting the fluorescence difference between the reduced and oxidized forms. Each mutant was cloned into pET-22b + plasmid, expressed in E. coli BL21 (DE3), and purified with Ni-NTA column (Qiagen). Fluorescence emission spectra were recorded on a Hitachi F-4010 fluorescence spectrophotometer with the excitation at 280 nm. To determine the equilibrium redox constant of CcmH 19–99, 15 μM protein in a reaction buffer (100 mM phosphate, 1 mM EDTA, 0.1 mM GSSG, pH7.0) was incubated with different concentration of GSH at 37 °C for 8 h under nitrogen atmosphere. The relative amount of the reduced protein at equilibrium was measured using the specific fluorescence emission at 340 nm. From the equilibrium constant as well as the standard potential of glutathione (E0′GSH/GSSG = − 0.205 V), the standard redox potential of CcmH 19–99 was calculated [28]. 2.6. Unfolding and refolding equilibrium The reversible GdmCl-induced unfolding/refolding of CcmH 19–99was performed by measuring the circular dichroic (CD) ellipticities at 222 nm [29]. Because the purified protein was in the oxidized form as determined by DTNB, detection of the reduced form was performed in the presence of 2 mM DTT. For unfolding equilibrium, the protein was dissolved in different concentration of GdmCl and
1396
X.-M. Zheng et al. / Biochimica et Biophysica Acta 1824 (2012) 1394–1400
incubated for 3 h at 25 ° C. Data were analyzed according to two-state assumption [30]. 2.7. Homology modeling of the C-terminal CcmH 175–350 The structure modeling of the C-terminal domain of CcmH was performed by homology modeling program Modeller9v3 [31]. To obtain homologous sequences with the structures solved, the sequence was submitted to NCBI for BLASTp analysis. Following BLASTp query, the TPR domain of NrfG from E. coli (PDB ID: 2E2E) [32] was selected as a template for structure modeling. The structure model of CcmH 175–350 was displayed by using MOLMOL and the quality of the overall structure was analyzed by PROCHECK.
fusion proteins CcmH123–Gluc and CcmH138–Gluc have negligible activities, indicating the cytoplasmic location of this region within residues 123 to 138, which is in consistent with the SCAM analysis (Table 1). All of the constructs with Gluc fused to different CcmH fragments at their C-termini, CcmH196, CcmH256, CcmH288 and CcmH345, have high luminescence intensities (Fig. 1A), indicating that the luciferase moieties are translocated into the periplasm. Since the Gluc fusions have a little variation in the protein levels in cells (Fig. 1B), the ratios of the luminescence intensities over protein amounts may objectively reflect the translocation efficiencies of the fusion proteins to the periplasm (Fig. 1C). Taken together, the Gluc fusion experiment further demonstrates that the C-terminal domain
3. Results 3.1. Biochemical evidence for the topological structure of CcmH The CcmH protein was proposed as a cytochrome c reductase located in the inner membrane of E. coli [12], but little experimental evidence has been provided for the entire topology. CcmH contains four hydrophobic regions that are the putative transmembrane segments (Suppl. Fig. S1). We firstly performed SCAM assay to characterize the topological structure of CcmH [20] (Suppl. Fig. S2). Table 1 summarizes the experimental data from SCAM analysis. The N-terminal domain (Thr19–Gly88) including the active-site motif is entirely located in the periplasm, in agreement with the previous studies [12]. Residues Ser123 to Ala145 are localized in the cytoplasm, while two regions (around Gly113, Gly153–Gln173) flanking this cytoplasmic loop are tentatively assigned to transmembrane segments. For the C-terminal segment, two regions (Thr181–Thr213, Ala315– Ser349) are definitely periplasmic, but the residues of the middle part are still undefined. Thus, so far whether this region (Gly224–Ala303) is located in a transmembrane segment or involved in a hydrophobic core of the compact structure is still controversial. We then applied a newly developed method, Gaussia luciferase assay, to determine the topology of transmembrane proteins [21]. Gluc contains 10 Cys residues [33], and only the correctly folded enzyme has a high activity and shed luminescence in the presence of substrate CTZ [21]. When the enzyme is located in the periplasm, where many thiol isomerase exists, it can hold a correctly folded form with high enzymatic activity. Because of the reductive environment in the cytoplasm, few luciferase molecules can form correct disulfides and then exert luciferase activity. As shown in Fig. 1, the Table 1 Summary of the SCAM analysis for residue locations of the single-cysteine CcmH mutants. Cys site
C0 T19C A29C C43 C46 G88C T92C T98C G113C S123C S138C A145C G153C S164C T168C a b
Tritona –
+
Ub S W S S S U U U U U U U U U
U S S S S S U S U S S S U U U
Residue location
Cys site
Triton –
+
– Periplasm Periplasm Periplasm Periplasm Periplasm UD UD Membrane Cytoplasm Cytoplasm Cytoplasm Membrane Membrane Membrane
Q173C T181C A196C T213C G224C S239C T247C S256C G261C S297C A300C A303C A315C T328C S349C
U S S S U U U U U U U U S S S
U S S S U U U S W U U U S S S
Residue location
Membrane Periplasm Periplasm Periplasm UD UD UD UD UD UD UD UD Periplasm Periplasm Periplasm
The data are from MPB labeling with or without Triton X-100 treatment. U, unlabeled; W, weakly labeled; S, strongly labeled; UD, undefined.
Fig. 1. Characterization of the topological structure of CcmH protein by Gluc fusion assay. (A) Luminescence intensities of different CcmH fragments fused with Gluc. The data are presented as Mean ± SD. (B) Western blotting showing the expression of these CcmH fragments fused with Gluc in E. coli. (C) Relative luminescence intensities of different CcmH fragments fused with Gluc. The data are the ratios of luminescence intensities over protein amounts from Western blotting. The protein amounts are from the band intensities analyzed with Scion Image software.
X.-M. Zheng et al. / Biochimica et Biophysica Acta 1824 (2012) 1394–1400
(residues 175–350) of CcmH protein is entirely located in the bacterial periplasm. 3.2. Redox potential of the N-terminal catalytic domain To compare the stabilities of the oxidized and reduced forms of the N-terminal catalytic domain of CcmH (CcmH 19–99), we performed the GdmCl-induced unfolding by monitoring CD ellipticities at 222 nm. Both the oxidized and reduced forms show a two-state transition profile, suggesting that breakdown of the hydrophobic core and the secondary structures undergoes the same process (Fig. 2A). The two forms have the similar stability under the unfolding condition, which is different from the thioredoxin domain of CcmG, in which the reduced form is more stable than the oxidized form [34]. For the sake of fluorescence detection, we generated three mutants of the redox domain, L41W, I51W and F90W, in which each has an introduced Trp residue in different potential site based on the structure solved (see Fig. 4B). Of them, the I51W mutant is suitable for analyzing the redox properties, because it exhibits large difference of the fluorescence intensities between the oxidized and reduced forms (Fig. 2B), while the overall structure of this mutant is not considerably changed as evidenced by circular dichroic analysis
1397
(Suppl. Fig. S3). An over two-fold increase of the fluorescence intensity can be observed when the oxidized form is reduced by addition of DTT. It is likely that the Trp fluorescence is quenched by the disulfide bond in the oxidized form [34]. The profile for the redox transition equilibrium is shown in Fig. 2C. Based on Nernst equation and the standard potential of glutathione (E0′GSH/GSSG = − 0.205 V), the redox potential of the CcmH 19–99 domain (I51W mutant) is calculated to be − 0.23 V, consistent with that of Pa-CcmH* (− 0.215 V) [14]. These data indicate that CcmH 19–99is more reductive than the thioredoxin domain of CcmG (− 0.175 V), a weak reductant [34], implying that the N-terminal domain of CcmH may act in the last reduction step in cytochrome c maturation. 3.3. Solution structure of the N-terminal catalytic domain in reduced form The N-terminal domain of CcmH (CcmH 19–99) is periplasm resident, functioning as a reductase in the redox reaction [12,16]. To get insights into the redox properties of CcmH, we studied the catalytic domain in the reduced form by NMR techniques. Since the domain structure is dimeric in crystal [15], we performed size exclusion chromatography (SEC) to characterize its form. The SEC profile
Fig. 2. Redox properties of the N-terminal domain CcmH19–99. (A) GdmCl-induced unfolding of the oxidized (solid) and reduced (open) forms of CcmH19–99monitored by CD spectroscopy. The native fraction in different concentration of GdmCl was estimated from CD ellipticity at 222 nm for both forms. For unfolding equilibrium, the protein (0.02 mM) was dissolved in different concentration of GdmCl and incubated for 3 h at 25 °C. For the reduced form, 2 mM DTT was included for incubation for at least 5 min. (B) Fluorescence spectra of the oxidized (square) and reduced (circle) forms of the I51W mutant. The fluorescence emission spectra were recorded at 300–400 nm with the excitation at 280 nm. The reduced protein was prepared in the presence of 2 mM DTT for at least 5 min. The reduced form exhibits a higher intensity of fluorescence than the oxidized. (C) Redox equilibrium of CcmH19–99 (I51W mutant) with glutathione. The oxidized CcmH19–99was incubated with variable concentration of GSH in the reaction buffer (100 mM phosphate, 1 mM EDTA, 0.1 mM GSSG, pH7.0) for 8 h at 37° C. The fraction of the reduced protein at equilibrium at different molar ratio of GSH and GSSG was measured by recording the fluorescence intensities at 340 nm.
1398
X.-M. Zheng et al. / Biochimica et Biophysica Acta 1824 (2012) 1394–1400
suggests that CcmH 19–99forms a monomer in solution (Fig. 3). So we solved the solution structure based on the presumption of a monomer protein under the NMR condition. Actually the line broadness in NMR spectra also supports a monomeric form. The experimental constraints for the structure calculation and analysis are listed in Table 2. As in the oxidized form in crystal structure [15], the solution structure of CcmH19–99comprises mainly a bundle of three α-helices, Helix 1 (residues 28–37), Helix 2 (57–72) and Helix 3 (76–86) (Fig. 4A). There is a short antiparallel β-sheet (41–42, 92–93) near the redox-active site, which appears not existed in the domainswapped dimer [15]. The active-site motif Cys–Pro–Lys–Cys is located in the loop connecting Helix 1 and 2 (Fig. 4B). The distance between the sulfur atoms of Cys43 and Cys46 is 9.1 Å, indicative of a reduced form of CcmH 19–99. This is contrary to the crystal structure of this domain, where it from P. aeruginosa [14] or E. coli [15] is oxidized in the crystal state. Fig. 4C compares the N-terminal domain structures of CcmH from E. coli in the reduced form and from P. aeruginosa in the oxidized form. The largest difference between the two domains lies in Helix 3, where the Pa-CcmH* domain contains an additional short helix, whereas Ec‐CcmH 19–99does not. It appears that the orientation of helix 3 and the loop connecting Helix 1 and 2 are also significantly changed.
Table 2 NMR experimental restraints and structural model of the CcmH19–99domain (residues 19–99).
3.4. The C-terminal domain of CcmH shows a compact fold
4. Discussion
To characterize the structural feature of the C-terminal domain, we prepared the C-terminal fragment CcmH 175–350 and performed CD studies. CcmH 175–350 shows a typical CD spectrum with double‐ negative peaks (Fig. 5), indicative of α-helical secondary structures. In the presence of low-concentration SDS, its helical structure content is significantly reduced rather than increased, suggesting that the C-terminus of CcmH is a non-transmembrane segment (Fig. 5). Since CcmH 175–350 is highly soluble in recombinant purified form, it is proposed to be a soluble domain located in the periplasm. We then tried to perform structural analysis, but unfortunately, we have not got the structure yet, because it is prone to proteolytic degradation in solution. By using homology modeling, CcmH 175–350 shows a relative compact structure of all α-helices (see Fig. 6), reminiscent of the TPR-like domain structure [32].
CcmH is a strategic reductase functioning in the c-type cytochrome maturation [1]. Since it is a transmembrane protein, the precise topology and high-resolution structure are the prerequisite for understanding the function of this protein. In this work, we employed two methods to analyze the topological structure of CcmH. SCAM is a classically biochemical method, but it may sometimes give us ambiguous information (Table 1). However, the luciferase fusion experiment is unambiguous that suggests that the segments around residues Ala196, Ser256 and Thr288 are located in the periplasm (Fig. 1). There is a possibility that these residues are buried in the hydrophobic core of the C-terminal domain, where MPB molecule is difficult to be accessed. Other evidence is from the purified fragment (CcmH 175–350) that is a highly soluble and well structured domain (Fig. 5). Thus, we propose that the C-terminal
Fig. 3. Size exclusion chromatography graph showing the monomeric form of CcmH19–99. The profile of CcmH19–99is shown in black, and thioredoxin as a control is shown in gray. Column: analytical Superdex 75 (GE Healthcare). Comparing the profile of CcmH19–99with thioredoxin (14 kDa), the apparent molecular weight of CcmH19–99is estimated to be 12 kDa. This suggests that CcmH19–99is a monomer in solution.
Fig. 4. Solution structure of the N-terminal redox domain CcmH19–99in reduced form. (A) Ribbon diagram of a representative structure of CcmH19–99. (B) View of the active-site loop region of CcmH19–99highlighting the Ile51 and active Cys43/46 residues. (C) Superposition of the solution structure of reduced Ec-CcmH19–99 (red) and the crystal structure of oxidized Pa-CcmH* (green).
Number of experimental restraints Total unambiguous distance restraints Intra‐residual Sequential (| i − j| = 1) Medium range (2 ≤|i − j| ≤ 4) Long range (|i − j| ≥ 5) Dihdedral angle restraints (ϕ, φ) RMSD from the average atomic coordinates (Å) Secondary structures Backbone atoms All heavy atoms All residues Backbone atoms All heavy atoms Deviations from idealized covalent geometry Bond (Å) Angles (°) Improper (°) Ramachandran analysis (%) Residues in most favored regions Residues in additionally allowed regions Residues in generously allowed regions Residues in disallowed regions
1116 537 238 226 115 50 × 2 0.38 ± 0.08 0.63 ± 0.09 0.88 ± 0.11 1.14 ± 0.12
0.004 ± 0.00006 0.56 ± 0.01 1.50 ± 0.08 86.2 9.81 2.31 1.71
X.-M. Zheng et al. / Biochimica et Biophysica Acta 1824 (2012) 1394–1400
Fig. 5. CD spectra of the C-terminal domain of CcmH. CcmH175–350 shows a spectrum of double negative peaks at 222 and 208 nm (black). Low concentration of SDS (0.4 – 2.0 %) causes a significant decrease of the peak intensity at 222 nm (gray).
domain is entirely located in the periplasm. The forth region predicted by hydrophobicity analysis (Suppl. Fig. 1) is actually not a transmembrane segment. Based on biochemical characterization, and structure and redox potential analysis, we proposed a schematic model for CcmH protein located in the inner membrane of E. coli (Fig. 6). CcmH is expressed in the cytoplasm, and then translocated to the inner membrane with the assistance of the N-terminal signal peptide, which, after translocation, may be removed by the signal peptidase. The middle part of CcmH forms two transmembrane helices anchoring in the inner membrane, while it retains a loop region in the cytoplasm. Both the N-terminal catalytic domain and the C-terminal TPR-like domain orientate toward the periplasm, suggesting that CcmH functions as an oxidoreductase for cytochrome c maturation in the periplasm. Although the function of the C-terminal domain is still unknown, it is speculated that this domain may play a role in protein interaction, as in the case of TPR-containing proteins [32]. Another possibility is that some helical segments of the C-terminal domain are associated with the inner membrane, in which helix–lipids interaction may cause unfolding of some helices (Fig. 5). Thus, understanding how the
1399
C-terminal domain functions in the periplasm awaits elucidation of the membrane-associated structure. Since the N-terminus of E. coli CcmH is homologous to CycL, the C-terminus is like CycH and CcmI [12,35]. This study on the biochemical properties and catalytic domain structure of E. coli CcmH may also provide valuable information to these homologous proteins. The crystal structure of the N-terminal catalytic domain of CcmH showed a helix-swapped dimer and oxidized form [15]. As comparison, our NMR analysis provides an alternative structure in solution with monomeric and reduced form. It seems that the redox change may cause conformational fluctuation at least in the active-site loop, thus I51W mutant is a suitable fluorescence probe for measuring the redox potential. For a thiol oxidoreductase, the redox potential may determine its biological function. Compared with those of CcmG (−0.186 V) [11,34] and DsbA (0.089 V) [28], CcmH (−0.23 V) may act as a strong reductant like DsbD [36]. The reductive activity is just a little weaker than that of the cytoplasmic thioredoxin (−0.269 V) [11]. This redox potential feature confers CcmH a reductase involved in reduction of apocytochrome c, the last step for covalent haem attachment. Our research provides information of the topological structure and redox potential for the E. coli CcmH protein. Sequence alignment and structure modeling also suggest that the C-terminus of CcmH is a TPR-like domain with an all-helical structure fold. The periplasm‐oriented C-terminal domain is suggested for recognition of apocytochrome c, as the TPR domain in NrfG [32]. However, this assumption awaits further exploration in the future.
5. Conclusion In summary, the mature CcmH is a periplasm-resident reductase anchored to the inner membrane by two transmembrane segments. Both N- and C-terminal domains are located and function in the periplasmic compartment. The N-terminal catalytic domain comprises a three-helix bundle and a loop with reactive thiols that functions in the reduction of apocytochrome c, and the C-terminal domain may play a role in apocytochrome c recognition.
6. Coordinates The coordinates and structural information have been deposited in PDB with the accession code 2KW0, and the NMR assignments are deposited in BMRB with No.16802.
Fig. 6. Schematic model of the CcmH protein anchoring in the inner membrane of E. coli. The N-terminal domain is illustrated with the solution structure (PDB: 2KW0), while the C-terminal domain is shown with the structure of a homology model. There are two hydrophobic helices crossing the inner membrane that assist two functional domains toward the periplasm. The C-terminal domain may have some helical segments contacting with the membrane.
1400
X.-M. Zheng et al. / Biochimica et Biophysica Acta 1824 (2012) 1394–1400
Acknowledgements The authors thank Z. R. Zhou, S. Liu and Y. H. Zhang, and Y. Peng for technical assistance and discussion. This work was supported by grants from the National Basic Research Program of China (2011CB911104, 2012CB911003), and the National Natural Science Foundation of China (10979070).
[15]
[16] [17]
[18]
Appendix A. Supplementary data [19]
Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.bbapap.2012.06.017.
[20] [21]
References [22] [1] L. Thony-Meyer, Cytochrome c maturation: a complex pathway for a simple task? Biochem. Soc. Trans. 30 (2002) 633–638. [2] L. Thony-Meyer, F. Fischer, P. Kunzler, D. Ritz, H. Hennecke, Escherichia coli genes required for cytochrome c maturation, J. Bacteriol. 177 (1995) 4321–4326. [3] J. Grove, S. Tanapongpipat, G. Thomas, L. Griffiths, H. Crooke, J. Cole, Escherichia coli K-12 genes essential for the synthesis of c-type cytochromes and a third nitrate reductase located in the periplasm, Mol. Microbiol. 19 (1996) 467–481. [4] C.L. Richard-Fogal, E.R. Frawley, E.R. Bonner, H. Zhu, B. San Francisco, R.G. Kranz, A conserved haem redox and trafficking pathway for cofactor attachment, EMBO J. 28 (2009) 2349–2359. [5] C.L. Richard-Fogal, E.R. Frawley, R.G. Kranz, Topology and function of CcmD in cytochrome c maturation, J. Bacteriol. 190 (2008) 3489–3493. [6] U. Ahuja, L. Thony-Meyer, CcmD is involved in complex formation between CcmC and the heme chaperone CcmE during cytochrome c maturation, J. Biol. Chem. 280 (2005) 236–243. [7] R.E. Feissner, C.L. Richard-Fogal, E.R. Frawley, R.G. Kranz, ABC transportermediated release of a haem chaperone allows cytochrome c biogenesis, Mol. Microbiol. 61 (2006) 219–231. [8] Q. Ren, U. Ahuja, L. Thony-Meyer, A bacterial cytochrome c heme lyase. CcmF forms a complex with the heme chaperone CcmE and CcmH but not with apocytochrome c, J. Biol. Chem. 277 (2002) 7657–7663. [9] H. Schulz, H. Hennecke, L. Thony-Meyer, Prototype of a heme chaperone essential for cytochrome c maturation, Science 281 (1998) 1197–1200. [10] T. Uchida, J.M. Stevens, O. Daltrop, E.M. Harvat, L. Hong, S.J. Ferguson, T. Kitagawa, The interaction of covalently bound heme with the cytochrome c maturation protein CcmE, J. Biol. Chem. 279 (2004) 51981–51988. [11] Q. Li, H.Y. Hu, W.Q. Wang, G.J. Xu, Structural and redox properties of the leaderless DsbE (CcmG) protein: both active-site cysteines of the reduced form are involved in its function in the Escherichia coli periplasm, Biol. Chem. 382 (2001) 1679–1686. [12] R.A. Fabianek, T. Hofer, L. Thony-Meyer, Characterization of the Escherichia coli CcmH protein reveals new insights into the redox pathway required for cytochrome c maturation, Arch. Microbiol. 171 (1999) 92–100. [13] C. Sanders, S. Turkarslan, D.W. Lee, O. Onder, R.G. Kranz, F. Daldal, The cytochrome c maturation components CcmF, CcmH, and CcmI form a membrane-integral multisubunit heme ligation complex, J. Biol. Chem. 283 (2008) 29715–29722. [14] A. Di Matteo, S. Gianni, M.E. Schinina, A. Giorgi, F. Altieri, N. Calosci, M. Brunori, C. Travaglini-Allocatelli, A strategic protein in cytochrome c maturation: three-
[23] [24]
[25]
[26] [27]
[28] [29] [30] [31] [32]
[33]
[34]
[35]
[36]
dimensional structure of CcmH and binding to apocytochrome c, J. Biol. Chem. 282 (2007) 27012–27019. U. Ahuja, A. Rozhkova, R. Glockshuber, L. Thony-Meyer, O. Einsle, Helix swapping leads to dimerization of the N-terminal domain of the c-type cytochrome maturation protein CcmH from Escherichia coli, FEBS Lett. 582 (2008) 2779–2786. J. Grovc, S. Busby, J. Cole, The role of the genes nrf EFG and ccmFH in cytochrome c biosynthesis in Escherichia coli, Mol. Gen. Genet. 252 (1996) 332–341. S. Turkarslan, C. Sanders, S. Ekici, F. Daldal, Compensatory thio-redox interactions between DsbA, CcdA and CcmG unveil the apocytochrome c holdase role of CcmG during cytochrome c maturation, Mol. Microbiol. 70 (2008) 652–666. C.U. Stirnimann, A. Rozhkova, U. Grauschopf, M.G. Grutter, R. Glockshuber, G. Capitani, Structural basis and kinetics of DsbD-dependent cytochrome c maturation, Structure 13 (2005) 985–993. E. Reid, J. Cole, D.J. Eaves, The Escherichia coli CcmG protein fulfils a specific role in cytochrome c assembly, Biochem. J. 355 (2001) 51–58. Q. Zhu, J.R. Casey, Topology of transmembrane proteins by scanning cysteine accessibility mutagenesis methodology, Methods 41 (2007) 439–450. H.Y. Li, X.M. Zheng, M.X. Che, H.Y. Hu, A redox-sensitive luciferase assay for determining the localization and topology of endoplasmic reticulum proteins, PLoS One 7 (2012) e35628. F. Delaglio, S. Grzesiek, G.W. Vuister, G. Zhu, J. Pfeifer, A. Bax, NMRPipe: a multidimensional spectral processing system based on UNIX pipes, J. Biomol. NMR 6 (1995) 277–293. T.D. Goddard, D.G. Kneller, SPARKY 3, University of California, San Francisco, 2008. M. Nilges, M.J. Macias, S.I. O'Donoghue, H. Oschkinat, Automated NOESY interpretation with ambiguous distance restraints: the refined NMR solution structure of the pleckstrin homology domain from beta-spectrin, J. Mol. Biol. 269 (1997) 408–422. G. Cornilescu, F. Delaglio, A. Bax, Protein backbone angle restraints from searching a database for chemical shift and sequence homology, J. Biomol. NMR 13 (1999) 289–302. R. Koradi, M. Billeter, K. Wuthrich, MOLMOL: a program for display and analysis of macromolecular structures, J. Mol. Graph. 14 (1996) 51–55 (29–32). R.A. Laskowski, J.A. Rullmannn, M.W. MacArthur, R. Kaptein, J.M. Thornton, AQUA and PROCHECK-NMR: programs for checking the quality of protein structures solved by NMR, J. Biomol. NMR 8 (1996) 477–486. M. Wunderlich, R. Glockshuber, Redox properties of protein disulfide isomerase (DsbA) from Escherichia coli, Protein Sci. 2 (1993) 717–726. H.Y. Hu, H.N. Du, Alpha-to-beta structural transformation of ovalbumin: heat and pH effects, J. Protein Chem. 19 (2000) 177–183. D. Shortle, Staphylococcal nuclease: a showcase of m-value effects, Adv. Protein Chem. 46 (1995) 217–247. A. Sali, T.L. Blundell, Comparative protein modelling by satisfaction of spatial restraints, J. Mol. Biol. 234 (1993) 779–815. D. Han, K. Kim, J. Oh, J. Park, Y. Kim, TPR domain of NrfG mediates complex formation between heme lyase and formate-dependent nitrite reductase in Escherichia coli O157:H7, Proteins 70 (2008) 900–914. S. Inouye, Y. Sahara, Identification of two catalytic domains in a luciferase secreted by the copepod Gaussia princeps, Biochem. Biophys. Res. Commun. 365 (2008) 96–101. Q. Li, H. Hu, G. Xu, Biochemical characterization of the thioredoxin domain of Escherichia coli DsbE protein reveals a weak reductant, Biochem. Biophys. Res. Commun. 283 (2001) 849–853. C. Sanders, C. Boulay, F. Daldal, Membrane-spanning and periplasmic segments of CcmI have distinct functions during cytochrome c biogenesis in Rhodobacter capsulatus, J. Bacteriol. 189 (2007) 789–800. J.F. Collet, J. Riemer, M.W. Bader, J.C. Bardwell, Reconstitution of a disulfide isomerization system, J. Biol. Chem. 277 (2002) 26886–26892.
Contents lists available at SciVerse ScienceDirect
Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbapap
Biochemical properties and catalytic domain structure of the CcmH protein from Escherichia coli Xue-Ming Zheng a , 1, Jing Hong b, d, 1, Hai-Yin Li a, Dong-Hai Lin b, c, Hong-Yu Hu a,⁎ a
Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai 200031, China Shanghai Institute of Materia Medica, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai 201203, China The key Laboratory for Chemical Biology of Fujian Province, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, China d Department of Food Science and Engineering, College of Biological Science and Technology, Fuzhou University, Fuzhou 350108, China b c
a r t i c l e
i n f o
Article history: Received 20 March 2012 Received in revised form 11 June 2012 Accepted 29 June 2012 Available online 10 July 2012 Keywords: CcmH protein topology solution structure redox potential cytochrome c maturation periplasm
a b s t r a c t In the Gram-negative bacterium of Escherichia coli, eight genes organized as a ccm operon (ccmABCDEFGH) are involved in the maturation of c-type cytochromes. The proteins encoded by the last three genes ccmFGH are believed to form a lyase complex functioning in the reduction of apocytochrome c and haem attachment. Among them, CcmH is a membrane-associated protein; its N-terminus is a catalytic domain with the active CXXC motif and the C-terminus is predicted as a TPR-like domain with unknown function. By using SCAM (scanning cysteine accessibility mutagenesis) and Gaussia luciferase fusion assays, we provide experimental evidence for the entire topological structure of E. coli CcmH. The mature CcmH is a periplasm-resident oxidoreductase anchored to the inner membrane by two transmembrane segments. Both N- and C-terminal domains are located and function in the periplasmic compartment. Moreover, the N-terminal domain forms a monomer in solution, while the C-terminal domain is a compact fold with helical structures. The NMR solution structure of the catalytic domain in reduced form exhibits mainly a three-helix bundle, providing further information for the redox mechanism. The redox potential suggests that CcmH exhibits a strong reductase that may function in the last step of reduction of apocytochrome c for haem attachment. © 2012 Elsevier B.V. All rights reserved.
1. Introduction The maturation of c-type cytochromes is a post-translational modification process, characterized by covalent attachment of haem to the thiols that occurs in the CXXC motif of apocytochrome c [1]. In the Gram-negative bacterium Escherichia coli, eight genes located at the minute 47 as an operon (ccmABCDEFGH) encode the indigenous proteins that are essential for the biogenesis of cytochrome c under anaerobic condition [2–4]. Among their gene products, CcmA is an ATP-binding cassette protein forming a haem release complex together with CcmB, CcmC and CcmD [5–7]. The chaperone protein CcmE can bind haem covalently at a conserved histidine residue and carry it to the lyase CcmF, which can catalyze the formation of covalent bond between haem and reduced apocytochrome c [8–10]. Both CcmH and CcmG are characterized by the CXXC motif in the catalytic domain oriented in periplasm and act as thiol reductases for Abbreviations: AMS, 4-acetamido-4′-maleimidylstilbene-2,2′-disulfonic acid; CcmH, cytochrome c maturation protein H; CD, circular dichroism; CTZ, coelenterazine; DTT, dithiothreitol; Gluc, luciferase from Gaussia princeps; GSH, glutathione; MPB, N-(3-maleimidopropionyl) biocytin; NMR, nuclear magnetic resonance; SCAM, scanning cysteine accessibility mutagenesis; TPR, tetratricopeptide repeat ⁎ Corresponding author. Tel./fax: +86 21 54921121. E-mail address: [email protected] (H.-Y. Hu). 1 These authors contributed equally to this work. 1570-9639/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.bbapap.2012.06.017
apocytochrome c [11,12]. In other species, another protein CcmI is also involved to form a complex together with CcmF and CcmH and functions in catalyzing the final steps of haem attachment to apocytochromes c [13]. CcmH of E. coli is encoded by the last gene in the ccm operon required for the maturation of cytochrome c. The N-terminal region characterized by the CXXC motif is homologous to CycL from B. japonicum, whereas the C-terminus is the paralog of CycH [2]. The CcmH protein was proposed to be essential for the cytochrome c biosynthesis [12]. The structures of the N-terminal domain from P. aeruginosa and E. coli in the oxidized form exhibit a three-helix bundle different from the canonical thioredoxin domain [14,15]. Deletion of the C-terminus of CcmH, a putative TPR-like domain, has little impact on the maturation of cytochrome c [16]. Although the function of CcmG and CcmH has been defined as reductases of apocytochrome c, the exact mechanism in the reduction cascade is still obscure. Apocytochrome c is oxidized by the thiol oxidase DsbA in the oxidized state of periplasm and must be reduced by reductases prior to ligation with haem [17]. CcmG is reduced by DsbD, which also functions in this pathway in mediating the transport of electron from cytoplasm to periplasm [11,18]. Intriguingly, CcmG remains predominantly in the oxidized form in the mutants defective in CcmF and CcmH, indicating that CcmG functions after them in the redox cascade [19]. Since no interaction has been detected between CcmG and CcmH, the reductive steps in the assembly of c-type cytochromes remain uncertain. The
X.-M. Zheng et al. / Biochimica et Biophysica Acta 1824 (2012) 1394–1400
finding that both CcmH and CcmG can bind apocytochrome c directly increases the complexity of this problem [14,17]. As predicted, CcmH is a transmembrane redox protein with some regions in the periplasmic lumen. However, the topological structure and its functional relationship are not defined in detail. Here we report characterization of the topological structure of CcmH by using SCAM (scan cysteine availability mutagenesis) method [20] and a newly developed luciferase fusion assay [21]. In addition, biochemical and structural studies of the redox properties suggest that the CcmH protein is a strong reductase as compared with another reductant CcmG (DsbE) [11], which functions in the last reduction step prior to haem attachment in cytochrome c maturation. 2. Materials and methods 2.1. Strains, plasmids and chemicals All the single-cysteine mutations of CcmH were generated by regular PCR and cloned into plasmid pET-22b + using NdeI and XhoI restriction sites. The E. coli strain BL21 (DE3) was used as the host for all the ccmH-containing plasmids. Luciferase from Gaussia princeps (Gluc) with the deletion of signal sequence was fused to the carboxyl end in frame with different N-terminal fragments of CcmH. The fusion genes were also cloned into the plasmid pET-22b +. All mutants and fusions were verified by DNA sequencing. The luciferase substrate CTZ was purchased for Chisso Company; the reagents MPB and AMS were from Sigma or Invitrogen. Streptavidin-conjugated horseradish peroxidase and other reagents were obtained from Pierce and other comparable suppliers. 2.2. SCAM (scanning cysteine accessibility mutagenesis) assay Each plasmid harboring a single-cysteine mutant of CcmH was transformed into E. coli strain BL21 (DE3), and cultured in LB medium supplemented with ampicillin (100 μg/mL) at 37 ° C. When A600 reaches ~ 0.6, IPTG (finally 0.2 mM) was added for inducing expression of the exogenous CcmH mutants and cultured for another 4 h before harvest. After sedimentation of aliquot culture, the cell pellet was resuspended in buffer A (50 mM Hepes, pH 7.6, 250 mM sucrose, 5 mM EDTA) and labeled with MPB (75 μM) on ice for 3 min. In another parallel experiment, the same amount of resuspended pellet from the same culture was incubated with 0.1% Triton X-100 for 15 min prior to MPB labeling [20]. When specified, the pellet was incubated with 5 mM AMS for 30 min before MPB labeling. All the labeling reactions were on ice and handled carefully avoiding possible impairment of the cell. The labeling reaction was stopped by addition of 2 mM DTT and incubated on ice for 5 min. Then, the cell pellet was collected by centrifugation. In order to get rid of other soluble proteins, the cell pellet was resuspended and lysed with lysozyme. The pellet was collected through centrifugation to exclude other endogenous proteins, which are probably labeled due to the reactive cysteines. The MPB labeling products were analyzed via Western blotting with streptavidin-conjugated horseradish peroxidase. Triplicated independent experiments were performed for each singlecysteine mutant of CcmH. 2.3. Gaussia luciferase fusion assay The luminescence assay of the CcmH fragments fused with luciferase was carried out accordingly [21]. Six DNA sequences coding for different N-terminal fragments of CcmH were fused with Gluc gene using regular PCR, cloned into the pET-22b + plasmid, transformed into the BL21 (DE3) strain, and plated on the LB plate with 100 μM of ampicillin. Cells with respective plasmid were allowed to culture at 37 °C in 5-mL LB medium. When A600 reaches ~ 0.6, IPTG (0.2 mM) was added for inducing expression and the
1395
cells were cultured for another 2 h. Subsequently the tubes containing cell culture were put on ice for 30 min before activity assay. Cell suspension of 10 μL was took out and diluted to 500 μL with sterile LB medium, then 1 μM of CTZ was added and mixed, and immediately the luminescence intensity was recorded on a GloMaxTM 20/20 Luminometer. 2.4. NMR spectroscopy and structure determination The cDNA encoding for CcmH 19–99was cloned into pET-22b + vector by NdeI and XhoI restriction sites, and all the CcmH 19–99proteins were purified from E. coli BL21 (DE3) cells with Ni-ATA column, followed by Superdex G-75 gel filtration chromatography. The reduced 15N/ 13C-labeled samples were prepared with 1 mM protein dissolved in 90% H2O/10% D2O buffer containing 20 mM sodium phosphate (pH 6.5), 50 mM NaCl, 5 mM DTT, and were argon-flushed. All NMR experiments were performed at 25 °C on Varian Unity INOVA 600 MHz NMR spectrometer equipped with four RF channels and a triple‐resonance pulsed field gradient probe. A set of two-dimensional 15N- and 13C-edited HSQC spectra and three-dimensional HNCA, HN (CO) CA, HNCO, HNCACB, CBCA (CO) NH, (H) CC (CO) NH-TOCSY, (H) CCH-TOCSY and HCCH-TOCSY spectra were collected to obtain the backbone and side‐chain assignments. The three-dimensional 15N-edited and 13C-edited NOESY spectra (mixing times of 100 ms) were used to confirm the chemical shift assignments and to generate distance restraints for structure calculation. All NMR data were processed using the program NMRPipe/NMRDraw [22] and analyzed using the program Sparky [23]. The structure calculations were performed using the program ARIA2.2 [24]. Distance restraints were derived from inter-proton NOEs. Dihedral angles (ϕ and φ) were calculated from backbone chemical shifts using the program TALOS [25]. In the calculation, 200 structures were calculated in the last iteration and the 20 structures with the lowest energy were selected for refinement in water, so as to no distance violations larger than 0.3 Å. The final structures were analyzed using the program packages MOLMOL [26] and PROCHECK [27]. 2.5. Determination of the redox potential Due to no tryptophan in CcmH 19–99, we introduced a Trp residue by mutagenesis at three positions respectively, L41W, I51W and F90W, near the reactive CXXC motif based on the structure, in order to gain a suitable mutant reporting the fluorescence difference between the reduced and oxidized forms. Each mutant was cloned into pET-22b + plasmid, expressed in E. coli BL21 (DE3), and purified with Ni-NTA column (Qiagen). Fluorescence emission spectra were recorded on a Hitachi F-4010 fluorescence spectrophotometer with the excitation at 280 nm. To determine the equilibrium redox constant of CcmH 19–99, 15 μM protein in a reaction buffer (100 mM phosphate, 1 mM EDTA, 0.1 mM GSSG, pH7.0) was incubated with different concentration of GSH at 37 °C for 8 h under nitrogen atmosphere. The relative amount of the reduced protein at equilibrium was measured using the specific fluorescence emission at 340 nm. From the equilibrium constant as well as the standard potential of glutathione (E0′GSH/GSSG = − 0.205 V), the standard redox potential of CcmH 19–99 was calculated [28]. 2.6. Unfolding and refolding equilibrium The reversible GdmCl-induced unfolding/refolding of CcmH 19–99was performed by measuring the circular dichroic (CD) ellipticities at 222 nm [29]. Because the purified protein was in the oxidized form as determined by DTNB, detection of the reduced form was performed in the presence of 2 mM DTT. For unfolding equilibrium, the protein was dissolved in different concentration of GdmCl and
1396
X.-M. Zheng et al. / Biochimica et Biophysica Acta 1824 (2012) 1394–1400
incubated for 3 h at 25 ° C. Data were analyzed according to two-state assumption [30]. 2.7. Homology modeling of the C-terminal CcmH 175–350 The structure modeling of the C-terminal domain of CcmH was performed by homology modeling program Modeller9v3 [31]. To obtain homologous sequences with the structures solved, the sequence was submitted to NCBI for BLASTp analysis. Following BLASTp query, the TPR domain of NrfG from E. coli (PDB ID: 2E2E) [32] was selected as a template for structure modeling. The structure model of CcmH 175–350 was displayed by using MOLMOL and the quality of the overall structure was analyzed by PROCHECK.
fusion proteins CcmH123–Gluc and CcmH138–Gluc have negligible activities, indicating the cytoplasmic location of this region within residues 123 to 138, which is in consistent with the SCAM analysis (Table 1). All of the constructs with Gluc fused to different CcmH fragments at their C-termini, CcmH196, CcmH256, CcmH288 and CcmH345, have high luminescence intensities (Fig. 1A), indicating that the luciferase moieties are translocated into the periplasm. Since the Gluc fusions have a little variation in the protein levels in cells (Fig. 1B), the ratios of the luminescence intensities over protein amounts may objectively reflect the translocation efficiencies of the fusion proteins to the periplasm (Fig. 1C). Taken together, the Gluc fusion experiment further demonstrates that the C-terminal domain
3. Results 3.1. Biochemical evidence for the topological structure of CcmH The CcmH protein was proposed as a cytochrome c reductase located in the inner membrane of E. coli [12], but little experimental evidence has been provided for the entire topology. CcmH contains four hydrophobic regions that are the putative transmembrane segments (Suppl. Fig. S1). We firstly performed SCAM assay to characterize the topological structure of CcmH [20] (Suppl. Fig. S2). Table 1 summarizes the experimental data from SCAM analysis. The N-terminal domain (Thr19–Gly88) including the active-site motif is entirely located in the periplasm, in agreement with the previous studies [12]. Residues Ser123 to Ala145 are localized in the cytoplasm, while two regions (around Gly113, Gly153–Gln173) flanking this cytoplasmic loop are tentatively assigned to transmembrane segments. For the C-terminal segment, two regions (Thr181–Thr213, Ala315– Ser349) are definitely periplasmic, but the residues of the middle part are still undefined. Thus, so far whether this region (Gly224–Ala303) is located in a transmembrane segment or involved in a hydrophobic core of the compact structure is still controversial. We then applied a newly developed method, Gaussia luciferase assay, to determine the topology of transmembrane proteins [21]. Gluc contains 10 Cys residues [33], and only the correctly folded enzyme has a high activity and shed luminescence in the presence of substrate CTZ [21]. When the enzyme is located in the periplasm, where many thiol isomerase exists, it can hold a correctly folded form with high enzymatic activity. Because of the reductive environment in the cytoplasm, few luciferase molecules can form correct disulfides and then exert luciferase activity. As shown in Fig. 1, the Table 1 Summary of the SCAM analysis for residue locations of the single-cysteine CcmH mutants. Cys site
C0 T19C A29C C43 C46 G88C T92C T98C G113C S123C S138C A145C G153C S164C T168C a b
Tritona –
+
Ub S W S S S U U U U U U U U U
U S S S S S U S U S S S U U U
Residue location
Cys site
Triton –
+
– Periplasm Periplasm Periplasm Periplasm Periplasm UD UD Membrane Cytoplasm Cytoplasm Cytoplasm Membrane Membrane Membrane
Q173C T181C A196C T213C G224C S239C T247C S256C G261C S297C A300C A303C A315C T328C S349C
U S S S U U U U U U U U S S S
U S S S U U U S W U U U S S S
Residue location
Membrane Periplasm Periplasm Periplasm UD UD UD UD UD UD UD UD Periplasm Periplasm Periplasm
The data are from MPB labeling with or without Triton X-100 treatment. U, unlabeled; W, weakly labeled; S, strongly labeled; UD, undefined.
Fig. 1. Characterization of the topological structure of CcmH protein by Gluc fusion assay. (A) Luminescence intensities of different CcmH fragments fused with Gluc. The data are presented as Mean ± SD. (B) Western blotting showing the expression of these CcmH fragments fused with Gluc in E. coli. (C) Relative luminescence intensities of different CcmH fragments fused with Gluc. The data are the ratios of luminescence intensities over protein amounts from Western blotting. The protein amounts are from the band intensities analyzed with Scion Image software.
X.-M. Zheng et al. / Biochimica et Biophysica Acta 1824 (2012) 1394–1400
(residues 175–350) of CcmH protein is entirely located in the bacterial periplasm. 3.2. Redox potential of the N-terminal catalytic domain To compare the stabilities of the oxidized and reduced forms of the N-terminal catalytic domain of CcmH (CcmH 19–99), we performed the GdmCl-induced unfolding by monitoring CD ellipticities at 222 nm. Both the oxidized and reduced forms show a two-state transition profile, suggesting that breakdown of the hydrophobic core and the secondary structures undergoes the same process (Fig. 2A). The two forms have the similar stability under the unfolding condition, which is different from the thioredoxin domain of CcmG, in which the reduced form is more stable than the oxidized form [34]. For the sake of fluorescence detection, we generated three mutants of the redox domain, L41W, I51W and F90W, in which each has an introduced Trp residue in different potential site based on the structure solved (see Fig. 4B). Of them, the I51W mutant is suitable for analyzing the redox properties, because it exhibits large difference of the fluorescence intensities between the oxidized and reduced forms (Fig. 2B), while the overall structure of this mutant is not considerably changed as evidenced by circular dichroic analysis
1397
(Suppl. Fig. S3). An over two-fold increase of the fluorescence intensity can be observed when the oxidized form is reduced by addition of DTT. It is likely that the Trp fluorescence is quenched by the disulfide bond in the oxidized form [34]. The profile for the redox transition equilibrium is shown in Fig. 2C. Based on Nernst equation and the standard potential of glutathione (E0′GSH/GSSG = − 0.205 V), the redox potential of the CcmH 19–99 domain (I51W mutant) is calculated to be − 0.23 V, consistent with that of Pa-CcmH* (− 0.215 V) [14]. These data indicate that CcmH 19–99is more reductive than the thioredoxin domain of CcmG (− 0.175 V), a weak reductant [34], implying that the N-terminal domain of CcmH may act in the last reduction step in cytochrome c maturation. 3.3. Solution structure of the N-terminal catalytic domain in reduced form The N-terminal domain of CcmH (CcmH 19–99) is periplasm resident, functioning as a reductase in the redox reaction [12,16]. To get insights into the redox properties of CcmH, we studied the catalytic domain in the reduced form by NMR techniques. Since the domain structure is dimeric in crystal [15], we performed size exclusion chromatography (SEC) to characterize its form. The SEC profile
Fig. 2. Redox properties of the N-terminal domain CcmH19–99. (A) GdmCl-induced unfolding of the oxidized (solid) and reduced (open) forms of CcmH19–99monitored by CD spectroscopy. The native fraction in different concentration of GdmCl was estimated from CD ellipticity at 222 nm for both forms. For unfolding equilibrium, the protein (0.02 mM) was dissolved in different concentration of GdmCl and incubated for 3 h at 25 °C. For the reduced form, 2 mM DTT was included for incubation for at least 5 min. (B) Fluorescence spectra of the oxidized (square) and reduced (circle) forms of the I51W mutant. The fluorescence emission spectra were recorded at 300–400 nm with the excitation at 280 nm. The reduced protein was prepared in the presence of 2 mM DTT for at least 5 min. The reduced form exhibits a higher intensity of fluorescence than the oxidized. (C) Redox equilibrium of CcmH19–99 (I51W mutant) with glutathione. The oxidized CcmH19–99was incubated with variable concentration of GSH in the reaction buffer (100 mM phosphate, 1 mM EDTA, 0.1 mM GSSG, pH7.0) for 8 h at 37° C. The fraction of the reduced protein at equilibrium at different molar ratio of GSH and GSSG was measured by recording the fluorescence intensities at 340 nm.
1398
X.-M. Zheng et al. / Biochimica et Biophysica Acta 1824 (2012) 1394–1400
suggests that CcmH 19–99forms a monomer in solution (Fig. 3). So we solved the solution structure based on the presumption of a monomer protein under the NMR condition. Actually the line broadness in NMR spectra also supports a monomeric form. The experimental constraints for the structure calculation and analysis are listed in Table 2. As in the oxidized form in crystal structure [15], the solution structure of CcmH19–99comprises mainly a bundle of three α-helices, Helix 1 (residues 28–37), Helix 2 (57–72) and Helix 3 (76–86) (Fig. 4A). There is a short antiparallel β-sheet (41–42, 92–93) near the redox-active site, which appears not existed in the domainswapped dimer [15]. The active-site motif Cys–Pro–Lys–Cys is located in the loop connecting Helix 1 and 2 (Fig. 4B). The distance between the sulfur atoms of Cys43 and Cys46 is 9.1 Å, indicative of a reduced form of CcmH 19–99. This is contrary to the crystal structure of this domain, where it from P. aeruginosa [14] or E. coli [15] is oxidized in the crystal state. Fig. 4C compares the N-terminal domain structures of CcmH from E. coli in the reduced form and from P. aeruginosa in the oxidized form. The largest difference between the two domains lies in Helix 3, where the Pa-CcmH* domain contains an additional short helix, whereas Ec‐CcmH 19–99does not. It appears that the orientation of helix 3 and the loop connecting Helix 1 and 2 are also significantly changed.
Table 2 NMR experimental restraints and structural model of the CcmH19–99domain (residues 19–99).
3.4. The C-terminal domain of CcmH shows a compact fold
4. Discussion
To characterize the structural feature of the C-terminal domain, we prepared the C-terminal fragment CcmH 175–350 and performed CD studies. CcmH 175–350 shows a typical CD spectrum with double‐ negative peaks (Fig. 5), indicative of α-helical secondary structures. In the presence of low-concentration SDS, its helical structure content is significantly reduced rather than increased, suggesting that the C-terminus of CcmH is a non-transmembrane segment (Fig. 5). Since CcmH 175–350 is highly soluble in recombinant purified form, it is proposed to be a soluble domain located in the periplasm. We then tried to perform structural analysis, but unfortunately, we have not got the structure yet, because it is prone to proteolytic degradation in solution. By using homology modeling, CcmH 175–350 shows a relative compact structure of all α-helices (see Fig. 6), reminiscent of the TPR-like domain structure [32].
CcmH is a strategic reductase functioning in the c-type cytochrome maturation [1]. Since it is a transmembrane protein, the precise topology and high-resolution structure are the prerequisite for understanding the function of this protein. In this work, we employed two methods to analyze the topological structure of CcmH. SCAM is a classically biochemical method, but it may sometimes give us ambiguous information (Table 1). However, the luciferase fusion experiment is unambiguous that suggests that the segments around residues Ala196, Ser256 and Thr288 are located in the periplasm (Fig. 1). There is a possibility that these residues are buried in the hydrophobic core of the C-terminal domain, where MPB molecule is difficult to be accessed. Other evidence is from the purified fragment (CcmH 175–350) that is a highly soluble and well structured domain (Fig. 5). Thus, we propose that the C-terminal
Fig. 3. Size exclusion chromatography graph showing the monomeric form of CcmH19–99. The profile of CcmH19–99is shown in black, and thioredoxin as a control is shown in gray. Column: analytical Superdex 75 (GE Healthcare). Comparing the profile of CcmH19–99with thioredoxin (14 kDa), the apparent molecular weight of CcmH19–99is estimated to be 12 kDa. This suggests that CcmH19–99is a monomer in solution.
Fig. 4. Solution structure of the N-terminal redox domain CcmH19–99in reduced form. (A) Ribbon diagram of a representative structure of CcmH19–99. (B) View of the active-site loop region of CcmH19–99highlighting the Ile51 and active Cys43/46 residues. (C) Superposition of the solution structure of reduced Ec-CcmH19–99 (red) and the crystal structure of oxidized Pa-CcmH* (green).
Number of experimental restraints Total unambiguous distance restraints Intra‐residual Sequential (| i − j| = 1) Medium range (2 ≤|i − j| ≤ 4) Long range (|i − j| ≥ 5) Dihdedral angle restraints (ϕ, φ) RMSD from the average atomic coordinates (Å) Secondary structures Backbone atoms All heavy atoms All residues Backbone atoms All heavy atoms Deviations from idealized covalent geometry Bond (Å) Angles (°) Improper (°) Ramachandran analysis (%) Residues in most favored regions Residues in additionally allowed regions Residues in generously allowed regions Residues in disallowed regions
1116 537 238 226 115 50 × 2 0.38 ± 0.08 0.63 ± 0.09 0.88 ± 0.11 1.14 ± 0.12
0.004 ± 0.00006 0.56 ± 0.01 1.50 ± 0.08 86.2 9.81 2.31 1.71
X.-M. Zheng et al. / Biochimica et Biophysica Acta 1824 (2012) 1394–1400
Fig. 5. CD spectra of the C-terminal domain of CcmH. CcmH175–350 shows a spectrum of double negative peaks at 222 and 208 nm (black). Low concentration of SDS (0.4 – 2.0 %) causes a significant decrease of the peak intensity at 222 nm (gray).
domain is entirely located in the periplasm. The forth region predicted by hydrophobicity analysis (Suppl. Fig. 1) is actually not a transmembrane segment. Based on biochemical characterization, and structure and redox potential analysis, we proposed a schematic model for CcmH protein located in the inner membrane of E. coli (Fig. 6). CcmH is expressed in the cytoplasm, and then translocated to the inner membrane with the assistance of the N-terminal signal peptide, which, after translocation, may be removed by the signal peptidase. The middle part of CcmH forms two transmembrane helices anchoring in the inner membrane, while it retains a loop region in the cytoplasm. Both the N-terminal catalytic domain and the C-terminal TPR-like domain orientate toward the periplasm, suggesting that CcmH functions as an oxidoreductase for cytochrome c maturation in the periplasm. Although the function of the C-terminal domain is still unknown, it is speculated that this domain may play a role in protein interaction, as in the case of TPR-containing proteins [32]. Another possibility is that some helical segments of the C-terminal domain are associated with the inner membrane, in which helix–lipids interaction may cause unfolding of some helices (Fig. 5). Thus, understanding how the
1399
C-terminal domain functions in the periplasm awaits elucidation of the membrane-associated structure. Since the N-terminus of E. coli CcmH is homologous to CycL, the C-terminus is like CycH and CcmI [12,35]. This study on the biochemical properties and catalytic domain structure of E. coli CcmH may also provide valuable information to these homologous proteins. The crystal structure of the N-terminal catalytic domain of CcmH showed a helix-swapped dimer and oxidized form [15]. As comparison, our NMR analysis provides an alternative structure in solution with monomeric and reduced form. It seems that the redox change may cause conformational fluctuation at least in the active-site loop, thus I51W mutant is a suitable fluorescence probe for measuring the redox potential. For a thiol oxidoreductase, the redox potential may determine its biological function. Compared with those of CcmG (−0.186 V) [11,34] and DsbA (0.089 V) [28], CcmH (−0.23 V) may act as a strong reductant like DsbD [36]. The reductive activity is just a little weaker than that of the cytoplasmic thioredoxin (−0.269 V) [11]. This redox potential feature confers CcmH a reductase involved in reduction of apocytochrome c, the last step for covalent haem attachment. Our research provides information of the topological structure and redox potential for the E. coli CcmH protein. Sequence alignment and structure modeling also suggest that the C-terminus of CcmH is a TPR-like domain with an all-helical structure fold. The periplasm‐oriented C-terminal domain is suggested for recognition of apocytochrome c, as the TPR domain in NrfG [32]. However, this assumption awaits further exploration in the future.
5. Conclusion In summary, the mature CcmH is a periplasm-resident reductase anchored to the inner membrane by two transmembrane segments. Both N- and C-terminal domains are located and function in the periplasmic compartment. The N-terminal catalytic domain comprises a three-helix bundle and a loop with reactive thiols that functions in the reduction of apocytochrome c, and the C-terminal domain may play a role in apocytochrome c recognition.
6. Coordinates The coordinates and structural information have been deposited in PDB with the accession code 2KW0, and the NMR assignments are deposited in BMRB with No.16802.
Fig. 6. Schematic model of the CcmH protein anchoring in the inner membrane of E. coli. The N-terminal domain is illustrated with the solution structure (PDB: 2KW0), while the C-terminal domain is shown with the structure of a homology model. There are two hydrophobic helices crossing the inner membrane that assist two functional domains toward the periplasm. The C-terminal domain may have some helical segments contacting with the membrane.
1400
X.-M. Zheng et al. / Biochimica et Biophysica Acta 1824 (2012) 1394–1400
Acknowledgements The authors thank Z. R. Zhou, S. Liu and Y. H. Zhang, and Y. Peng for technical assistance and discussion. This work was supported by grants from the National Basic Research Program of China (2011CB911104, 2012CB911003), and the National Natural Science Foundation of China (10979070).
[15]
[16] [17]
[18]
Appendix A. Supplementary data [19]
Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.bbapap.2012.06.017.
[20] [21]
References [22] [1] L. Thony-Meyer, Cytochrome c maturation: a complex pathway for a simple task? Biochem. Soc. Trans. 30 (2002) 633–638. [2] L. Thony-Meyer, F. Fischer, P. Kunzler, D. Ritz, H. Hennecke, Escherichia coli genes required for cytochrome c maturation, J. Bacteriol. 177 (1995) 4321–4326. [3] J. Grove, S. Tanapongpipat, G. Thomas, L. Griffiths, H. Crooke, J. Cole, Escherichia coli K-12 genes essential for the synthesis of c-type cytochromes and a third nitrate reductase located in the periplasm, Mol. Microbiol. 19 (1996) 467–481. [4] C.L. Richard-Fogal, E.R. Frawley, E.R. Bonner, H. Zhu, B. San Francisco, R.G. Kranz, A conserved haem redox and trafficking pathway for cofactor attachment, EMBO J. 28 (2009) 2349–2359. [5] C.L. Richard-Fogal, E.R. Frawley, R.G. Kranz, Topology and function of CcmD in cytochrome c maturation, J. Bacteriol. 190 (2008) 3489–3493. [6] U. Ahuja, L. Thony-Meyer, CcmD is involved in complex formation between CcmC and the heme chaperone CcmE during cytochrome c maturation, J. Biol. Chem. 280 (2005) 236–243. [7] R.E. Feissner, C.L. Richard-Fogal, E.R. Frawley, R.G. Kranz, ABC transportermediated release of a haem chaperone allows cytochrome c biogenesis, Mol. Microbiol. 61 (2006) 219–231. [8] Q. Ren, U. Ahuja, L. Thony-Meyer, A bacterial cytochrome c heme lyase. CcmF forms a complex with the heme chaperone CcmE and CcmH but not with apocytochrome c, J. Biol. Chem. 277 (2002) 7657–7663. [9] H. Schulz, H. Hennecke, L. Thony-Meyer, Prototype of a heme chaperone essential for cytochrome c maturation, Science 281 (1998) 1197–1200. [10] T. Uchida, J.M. Stevens, O. Daltrop, E.M. Harvat, L. Hong, S.J. Ferguson, T. Kitagawa, The interaction of covalently bound heme with the cytochrome c maturation protein CcmE, J. Biol. Chem. 279 (2004) 51981–51988. [11] Q. Li, H.Y. Hu, W.Q. Wang, G.J. Xu, Structural and redox properties of the leaderless DsbE (CcmG) protein: both active-site cysteines of the reduced form are involved in its function in the Escherichia coli periplasm, Biol. Chem. 382 (2001) 1679–1686. [12] R.A. Fabianek, T. Hofer, L. Thony-Meyer, Characterization of the Escherichia coli CcmH protein reveals new insights into the redox pathway required for cytochrome c maturation, Arch. Microbiol. 171 (1999) 92–100. [13] C. Sanders, S. Turkarslan, D.W. Lee, O. Onder, R.G. Kranz, F. Daldal, The cytochrome c maturation components CcmF, CcmH, and CcmI form a membrane-integral multisubunit heme ligation complex, J. Biol. Chem. 283 (2008) 29715–29722. [14] A. Di Matteo, S. Gianni, M.E. Schinina, A. Giorgi, F. Altieri, N. Calosci, M. Brunori, C. Travaglini-Allocatelli, A strategic protein in cytochrome c maturation: three-
[23] [24]
[25]
[26] [27]
[28] [29] [30] [31] [32]
[33]
[34]
[35]
[36]
dimensional structure of CcmH and binding to apocytochrome c, J. Biol. Chem. 282 (2007) 27012–27019. U. Ahuja, A. Rozhkova, R. Glockshuber, L. Thony-Meyer, O. Einsle, Helix swapping leads to dimerization of the N-terminal domain of the c-type cytochrome maturation protein CcmH from Escherichia coli, FEBS Lett. 582 (2008) 2779–2786. J. Grovc, S. Busby, J. Cole, The role of the genes nrf EFG and ccmFH in cytochrome c biosynthesis in Escherichia coli, Mol. Gen. Genet. 252 (1996) 332–341. S. Turkarslan, C. Sanders, S. Ekici, F. Daldal, Compensatory thio-redox interactions between DsbA, CcdA and CcmG unveil the apocytochrome c holdase role of CcmG during cytochrome c maturation, Mol. Microbiol. 70 (2008) 652–666. C.U. Stirnimann, A. Rozhkova, U. Grauschopf, M.G. Grutter, R. Glockshuber, G. Capitani, Structural basis and kinetics of DsbD-dependent cytochrome c maturation, Structure 13 (2005) 985–993. E. Reid, J. Cole, D.J. Eaves, The Escherichia coli CcmG protein fulfils a specific role in cytochrome c assembly, Biochem. J. 355 (2001) 51–58. Q. Zhu, J.R. Casey, Topology of transmembrane proteins by scanning cysteine accessibility mutagenesis methodology, Methods 41 (2007) 439–450. H.Y. Li, X.M. Zheng, M.X. Che, H.Y. Hu, A redox-sensitive luciferase assay for determining the localization and topology of endoplasmic reticulum proteins, PLoS One 7 (2012) e35628. F. Delaglio, S. Grzesiek, G.W. Vuister, G. Zhu, J. Pfeifer, A. Bax, NMRPipe: a multidimensional spectral processing system based on UNIX pipes, J. Biomol. NMR 6 (1995) 277–293. T.D. Goddard, D.G. Kneller, SPARKY 3, University of California, San Francisco, 2008. M. Nilges, M.J. Macias, S.I. O'Donoghue, H. Oschkinat, Automated NOESY interpretation with ambiguous distance restraints: the refined NMR solution structure of the pleckstrin homology domain from beta-spectrin, J. Mol. Biol. 269 (1997) 408–422. G. Cornilescu, F. Delaglio, A. Bax, Protein backbone angle restraints from searching a database for chemical shift and sequence homology, J. Biomol. NMR 13 (1999) 289–302. R. Koradi, M. Billeter, K. Wuthrich, MOLMOL: a program for display and analysis of macromolecular structures, J. Mol. Graph. 14 (1996) 51–55 (29–32). R.A. Laskowski, J.A. Rullmannn, M.W. MacArthur, R. Kaptein, J.M. Thornton, AQUA and PROCHECK-NMR: programs for checking the quality of protein structures solved by NMR, J. Biomol. NMR 8 (1996) 477–486. M. Wunderlich, R. Glockshuber, Redox properties of protein disulfide isomerase (DsbA) from Escherichia coli, Protein Sci. 2 (1993) 717–726. H.Y. Hu, H.N. Du, Alpha-to-beta structural transformation of ovalbumin: heat and pH effects, J. Protein Chem. 19 (2000) 177–183. D. Shortle, Staphylococcal nuclease: a showcase of m-value effects, Adv. Protein Chem. 46 (1995) 217–247. A. Sali, T.L. Blundell, Comparative protein modelling by satisfaction of spatial restraints, J. Mol. Biol. 234 (1993) 779–815. D. Han, K. Kim, J. Oh, J. Park, Y. Kim, TPR domain of NrfG mediates complex formation between heme lyase and formate-dependent nitrite reductase in Escherichia coli O157:H7, Proteins 70 (2008) 900–914. S. Inouye, Y. Sahara, Identification of two catalytic domains in a luciferase secreted by the copepod Gaussia princeps, Biochem. Biophys. Res. Commun. 365 (2008) 96–101. Q. Li, H. Hu, G. Xu, Biochemical characterization of the thioredoxin domain of Escherichia coli DsbE protein reveals a weak reductant, Biochem. Biophys. Res. Commun. 283 (2001) 849–853. C. Sanders, C. Boulay, F. Daldal, Membrane-spanning and periplasmic segments of CcmI have distinct functions during cytochrome c biogenesis in Rhodobacter capsulatus, J. Bacteriol. 189 (2007) 789–800. J.F. Collet, J. Riemer, M.W. Bader, J.C. Bardwell, Reconstitution of a disulfide isomerization system, J. Biol. Chem. 277 (2002) 26886–26892.