Report

Active site rearrangement and structural divergence in prokaryotic respiratory oxidases

See allHide authors and affiliations

Science  04 Oct 2019:
Vol. 366, Issue 6461, pp. 100-104
DOI: 10.1126/science.aay0967

Hemes switch spots in a terminal oxidase

Reduction of molecular oxygen to water is the driving force for respiration in aerobic organisms and is catalyzed by several distinct integral membrane complexes. These include an exclusively prokaryotic enzyme, cytochrome bd–type quinol oxidase, which is a potential antimicrobial target. Safarian et al. determined a high-resolution cryo–electron microscopy structure of this enzyme from the enteric bacterium Escherichia coli. Comparison to a homolog reveals a complete relocation of the site of oxygen binding and reduction caused by a change in the arrangement of heme cofactors and channels in the protein scaffold. This switch illustrates the diversity of structure and function in this family of enzymes and might reflect different biochemical roles of these homologs.

Science, this issue p. 100

Abstract

Cytochrome bd–type quinol oxidases catalyze the reduction of molecular oxygen to water in the respiratory chain of many human-pathogenic bacteria. They are structurally unrelated to mitochondrial cytochrome c oxidases and are therefore a prime target for the development of antimicrobial drugs. We determined the structure of the Escherichia coli cytochrome bd-I oxidase by single-particle cryo–electron microscopy to a resolution of 2.7 angstroms. Our structure contains a previously unknown accessory subunit CydH, the L-subfamily–specific Q-loop domain, a structural ubiquinone-8 cofactor, an active-site density interpreted as dioxygen, distinct water-filled proton channels, and an oxygen-conducting pathway. Comparison with another cytochrome bd oxidase reveals structural divergence in the family, including rearrangement of high-spin hemes and conformational adaption of a transmembrane helix to generate a distinct oxygen-binding site.

Cytochrome bd oxidases are quinol-dependent terminal oxidases found exclusively in prokaryotes. They catalyze the reduction of molecular oxygen to water in the respiratory chain without formation of reactive oxygen species (ROS). This exergonic reaction is coupled to the generation of an electrochemical proton gradient across the periplasmic membrane by vectorial release and uptake of protons (1). Cytochrome bd oxidases are unrelated to the heme-copper oxidases (HCO), including the cytochrome c oxidases of mitochondria (2). Cytochrome bd oxidases have a very high oxygen affinity and are indispensable for pathogenic bacteria during host infection, making them ideal targets for antimicrobial drug development (3). The cytochrome bd oxidases are divided into two subfamilies, featuring either a short (S) or a long (L) quinol-binding domain (Q-loop) (4). Most enterobacterial pathogens that cause acute infectious diseases, such as Salmonella, uropathogenic E. coli (UPEC), and enterohemorrhagic E. coli (EHEC), rely on L-subfamily bd oxidases for their survival (57).

Using cryo–electron microscopy (cryo-EM), we determined the structure of the E. coli L-subfamily bd-I oxidase in lipid nanodiscs with a bound Fab fragment to 2.7 Å resolution (Fig. 1A and figs. S1 and S2). The hetero-oligomeric bd oxidase consists of a pseudo-symmetric CydAB core dimer and two accessory single-transmembrane subunits, CydX and CydH (Fig. 1 and fig. S3) (8, 9). Subunits CydA and CydB share a common architecture of two four-helix bundles and an additional peripheral helix (Fig. 1C). The oxygen reduction reaction is confined to CydA, which contains all three heme cofactors and the periplasmically exposed Q-loop (10). The low-spin heme b558 is the initial electron acceptor and transfers electrons from a quinol to the diheme active site composed of two high-spin hemes (b595 and d), where molecular oxygen is reduced to water. CydB harbors a structural ubiquinone-8 molecule that occupies a hydrophobic pocket in a near-symmetric position relative to the b-type hemes (Fig. 1C and fig. S4). At a distance of 3.5 nm to heme d, this UQ-8 is unlikely to be involved in electron transfer or oxygen reduction, and is more likely to have a role in CydAB dimer assembly.

Fig. 1 Cryo-EM structure of the bd-I oxidase from E. coli.

(A) Surface representation of the bd-I oxidase cryo-EM density map at 2.68 Å resolution. (B) Subunit CydH binds to a hydrophobic cleft in CydA and blocks oxygen accessibility to heme b595 from the hydrophobic lipid bilayer. (C) E. coli bd-I contains 20 membrane-spanning helices. (D) The Q loop is divided into flexible QN and rigid, well-ordered QC segments. (E) The substrate inhibitor AD3-11 reduces deuterium exchange rate only in the flexible QN region. Relative differences of deuterium uptake are mapped on the structure and the CydA amino acid sequence. Abbreviations: A, Ala; C, Cys; D, Asp; E, Glu; F, Phe; G, Gly; H, His; I, Ile; K, Lys; L, Leu; M, Met; N, Asn; P, Pro; Q, Gln; R, Arg; S, Ser; T, Thr; V, Val; W, Trp; Y, Tyr.

The small noncatalytic accessory subunit CydX is positioned in a groove formed by TMHs 1 and 6 of CydA (Fig. 1A and fig. S3) where it promotes the assembly or stability of the oxidase complex (8). The arrangement and architectures of subunits CydA, CydB, and CydX resemble the overall structure of the S-subfamily cytochrome bd oxidase from Geobacillus thermodenitrificans (G. th) (10). The similar location of CydS in the G. th S-subfamily enzyme indicates an analogous structural role for CydX.

A previously unknown single transmembrane subunit CydH binds in the cleft between transmembrane helices (TMHs) 1 and 9 of CydA (Fig. 1B). CydH is encoded by the orphan gene ynhF (fig. S5), which is neither part of the cyd (bd-I) nor the app (bd-II) operon. Homologs of CydH are exclusively found in the proteobacterial clade, where they correlate with the presence of L-subfamily bd oxidases. Two tightly bound lipid molecules close to CydH seal the cleft between CydH and TMHs 1 and 8 of CydA (Fig. 1B). CydH occupies the putative oxygen entry site in the bd oxidase from G. th, where CydH is absent. The bd-I oxidase of E. coli may thus possess an alternative oxygen-conducting channel, because it is unlikely that CydH undergoes rapid rearrangement to open and close an oxygen-conducting channel in a gating-like mechanism without impairing efficient turnover.

We examined the role of the Q-loop in the as-isolated state of the enzyme without substrate, in the presence of a synthetic substrate quinone (UQ-1), and in the presence of a quinolone-type inhibitor (AD3-11) (fig. S6). The Q-loop starts near the periplasmic section of TM6, where we observe a helix break at the strictly conserved Asp239.A near propionate B of heme b558 (Fig. 1D and fig. S7). Map features in the region between Glu240.A and Asp247.A are noisy and difficult to interpret. However, we were able to trace the polypeptide chain, although with less precision, to the short horizontal helix Qh1 that comprises the highly conserved residues Lys252.A and Glu257.A, critical for quinol binding and electron transfer (11). The entire polypeptide stretch between the helix break of TM6 and Lys252.A forms a dynamic loop, covering heme b558 that would otherwise be surface-exposed. This conformational heterogeneity extends to the entire N-terminal part of the Q-loop (QN) domain. The previously uncharacterized helix-turn-helix motif of the C-terminal Q-loop domain (QC) is rigid and covers the entire periplasmic surface of CydA (Fig. 1D and fig. S7).

Hydrogen/deuterium exchange mass spectrometry (HDX-MS) measurements showed that binding of the aurachin-D–type competitive inhibitor AD3-11 exclusively affects the flexible and disordered QN-loop (Fig. 1E). No substrate-induced effect was observed in the QC-loop, implying that this insertion, which defines the L-subfamily, does not participate in substrate binding. Although AD3-11 decreased the flexibility of the QN-loop (Fig. 1E), it did not give rise to a distinct conformation that we could identify by cryo-EM (fig. S3). The flexibility seems to be required for the transient interaction of quinols with the bd oxidase and their rapid release upon oxidation. Apart from their interaction with the flexible QN-loop, UQ-1, or AD3-11 do not affect the bd-I oxidase structure, making it unlikely that additional substrate binding sites exist.

The E. coli bd-I oxidase contains two b-type hemes and a chlorin-type heme d cofactor that are organized in a triangular geometry near the periplasmic surface of CydA (Fig. 2A). Their arrangement is similar to the redox center organization of our previously reported crystal structure of the S-subfamily homolog, and this cofactor localization is thus an evolutionarily conserved feature of cytochrome bd–type oxygen reductases.

Fig. 2 Cofactor organization and redox reaction cycle.

(A) Triangular arrangement of the heme cofactors in CydA. Heme edge-to-edge distances are indicated by reversed parentheses. (B) Schematic overview of heme positions in E. coli and G. th. (C) Assignment of heme cofactors and their axial amino acid ligands. Illustrated map densities are filtered to equal contour levels.

Heme b558 is found within the hydrophobic membrane section adjacent to the periplasmic QN-loop domain. The two axial ligands of heme b558 are Met393.A and His186.A (Fig. 2C). The location and coordination of this heme are equivalent to the G. th enzyme (12, 13). Propionate A of heme b558 is coordinated by the conserved residues Lys252.A and Lys183.A. Propionate B interacts with the essential Asp239.A. Heme b558 is separated by 6.7 Å from the high-spin active site (Fig. 2A). This distance is bridged by the totally conserved Trp441.A that may mediate electron transfer.

In the E. coli enzyme, hemes b595 and d are interchanged with respect to the G. th oxidase (Fig. 2B) (4, 10, 14). In the E. coli structure, heme b595 is located near the periplasmic surface. Detailed map features allowed us to position the heme plane and its two elongated propionate groups unequivocally (Fig. 2C and fig. S8). The heme b595 iron is pentacoordinated with Glu445.A as its axial ligand (Fig. 2B). Both propionates and Glu445.A form a salt bridge network with the strictly conserved Arg448.A and Arg9.A. The side chain of Gln152.A is positioned between the two b-type hemes and adopts two distinct conformations (Fig. 2C). The two orientations most likely compensate the different protonation states of Glu445.A in the mixed-valence state of the enzyme. In the ferric state of heme b595, Gln152.A may stabilize the deprotonated Glu445.A. In the ferrous state, Glu445.A may become protonated and thus compensate the charge.

The third cofactor, which is the site of oxygen binding and reduction, is a cis–heme d hydroxychlorin γ-spirolactone (15). Heme d is located in the center of CydA near the interface to CydB (Fig. 1C and fig. S4). Because of its well-defined single propionate residue and characteristic spirolactone group, its assignment to this location is unambiguous (Fig. 2C and fig. S8). At a distance of 2.5 Å to the heme iron, the invariant His19.A acts as the axial ligand of heme d (Fig. 2B).

On the opposite side of the heme along the His19-FeD axis, we observed a strong nonpeptide density that accounts for the expected dioxygen molecule bound in the as-isolated state (Fig. 2C). Its slightly elongated shape is held in a hydrophobic pocket formed by the strictly conserved residues Phe140.A and Ile144.A that confine the dioxygen next to the heme d iron (Fig. 2B). The functionally essential Glu99.A completes the dioxygen binding site. In contrast to most carboxylates in our structure, especially those forming hydrogen bonds or salt bridges, the density of Glu99.A is not well defined. In the as-isolated state of the enzyme, heme d is predominantly in the ferrous state, with the charge-compensating Glu99.A protonated (1, 10). The lack of hydrogen-bonding partners in this hydrophobic environment most likely accounts for the flexibility of this residue.

The observed diversity in the high-spin heme configuration is reflected in the electrochemical properties of the cofactors. In the G. th enzyme, the oxygen-binding heme d has the lowest midpoint potential of +15 mV at pH 7 (fig. S9), which would not allow for efficient electron transfer and might explain the very low turnover rates (16). Its structural and electrochemical properties suggest that it primarily acts as an oxygen scavenger, unlike the high-turnover respiratory E. coli enzyme (17). The heme groups in the cytochrome bd oxidases from E. coli have an uphill midpoint potential distribution (b558 < b595 < heme d), consistent with properties of other L-subfamily members (18). E. coli heme d has the highest midpoint potential; the potentials of the two heme b groups are lower by at least 100 mV (fig. S9).

The critical step in the catalytic cycle of direct oxygen reduction is the transition from the fully reduced to the oxoferryl state, where ROS formation needs to be prevented (1923) (fig. S10). In a previous proposal, the dioxygen bond is broken in a simultaneous four-electron transfer step, which avoids the formation of a peroxide intermediate. However, the short edge-to-edge distance (6.7 Å) between the two b-type hemes and the interchanged positions of hemes b595 and d are consistent with a sequential electron transfer from heme b558 via heme b595 to heme d (Fig. 2). This proposal is further supported by the midpoint potential difference between heme b595 and d as well as electron backflow experiments (fig. S9) (18, 24). Earlier studies on the E. coli bd-I oxidase provide evidence for such a sequential electron transfer generating a short-lived peroxide intermediate, similar to Cpd0 in the catalytic cycle of peroxidases (21, 25).

We identified two solvent-accessible areas (26) and assigned all densities found consistently in the three independent cryo-EM maps to water molecules (Fig. 3). The waters define a hydrophilic H-channel that starts at the cytoplasmic CydAB interface and runs perpendicular to the membrane plane between the intersubunit four-helix bundle formed by TMH2/3.A and TMH2/3.B until the central Asp58.B (Fig. 3). The conserved hydrophilic residues Ser108.A, Glu107.A, and Ser140.A continue the direct hydrophilic pathway lined by water molecules to heme d and facilitate proton transfer for either charge compensation or dioxygen reduction (4) (Fig. 3C). Close to Asp58.B, the H-channel branches and runs along the conserved helix segment WDGNQ of TMH2.B in the center of CydB until Trp63.B, where its diameter is constricted to ~2 Å (Fig. 3B) (27). It continues beyond this residue and merges with the oxygen-conducting O-channel, similar to the K-pathway in HCOs (2, 28).

Fig. 3 Proton and oxygen pathways.

(A) Interior H- and O-channel predicted by MOLE2. (B) The hydrophilic, water-filled H-channel connects the heme d oxygen reduction site directly to the cytoplasm by the CydA pathway. A second CydB pathway is outlined by the conserved WDGNQ motif (indicated in yellow). The hydrophobic O-channel runs parallel to the membrane plane and connects the lipid interface with heme d. Water densities are shown with transparent white surfaces. (C) Heme d sits in an amphipathic pocket where the hydrophilic proton channel and hydrophobic oxygen pathway converge at the dioxygen binding site.

The O-channel starts near Trp63.B at the membrane interface between TMH1 and TMH9 of CydB and forms a direct hydrophobic pathway to the dioxygen binding site of heme d (Fig. 4B). This channel appears to be occupied by fewer water molecules, and they are spaced farther apart. Close to the dioxygen binding site of heme d, the O-channel becomes hydrophilic and we observe densities that could either be water or dioxygen molecules. Conceivably, the O-channel might act as a pathway for direct oxygen diffusion from the membrane interior to the heme d reaction site. At the same time, it may conduct protons through its connection with the H-channel (29).

Fig. 4 Structural and evolutionary diversity.

(A) Superimposed dioxygen binding sites from bd oxidase of E. coli (blue and orange) and G. th (transparent beige). Note that positions of heme d and b595 are interchanged. (B) The heme d site from E. coli and the superimposed position of b595 from G. th. (C) Sequence alignment and conservation analysis of TMH3.A. L- and S-subfamily members are indicated as superscripts. The red arrowhead denotes the position of the Leu101 insertion.

The interchanged positions of hemes b595 and d entail an adaptation of electron, proton, and dioxygen pathways (10). In G. th, where the oxygen-binding heme d is located close to the membrane interface, a direct oxygen entry site was postulated close to heme d between TMH1 and 8. In the E. coli enzyme, this site is blocked by the single-transmembrane subunit CydH (Fig. 4A). At position 101, next to the heme d site, the E. coli bd-I oxidase has a Leu inserted, which causes one helix turn to bulge, displacing Glu99.A (Glu101 in G. th) from the central heme iron to allow for dioxygen binding at this position (Fig. 4B). At the same time, Phe104.A is flipped and forms, together with Ile144.A, a hydrophobic binding site for dioxygen. The Leu101.A insertion coincides with a displacement of TMH9.B that forms the O-channel (Fig. 4A). The occurrence of this Leu is not unique for the E. coli bd-I oxidase or even members of the L-subfamily (Fig. 4C), but is a strong indication for the position of the oxygen-binding heme in structurally uncharacterized bd oxidases.

The observed flexibility of the Q-loop poses a challenge for structure-driven design of substrate-type inhibitors that mimic quinones and abolish election transfer. In case of the E. coli bd-I oxidase, it may be possible to design inhibitors that compete with UQ-8 for the CydB binding site and interfere with the assembly of a functional CydAB core dimer. An alternative target is the membrane-embedded entry site of the oxygen-conducting O-channel that would be targeted by hydrophobic compounds to block the channel and abolish oxygen uptake. Further structural studies of bd oxidases are needed to explore whether the observed structural differences between E. coli and G. th are specific for their respective bacterial clades or result from a distinct adaptation to their ecological niche and physiological roles.

Supplementary Materials

science.sciencemag.org/content/366/6461/100/suppl/DC1

Materials and Methods

Figs. S1 to S10

Table S1

Movie S1

References (3053)

References and Notes

Acknowledgments: We thank R. Zimmermann for outstanding technical assistance, S. Fritz for support in generating hybridoma cell lines, and J. Vonck, B. Murphy, and J. Vinnemann for carefully reading the manuscript and valuable discussions. Funding: Supported by the Max Planck Society, the Deutsche Forschungsgemeinschaft (Cluster of Excellence Macromolecular Complexes Frankfurt, SPP 2002), a Grant-in-Aid for Scientific Research (C) (16K07299 to J.S.) from the Japan Society for the Promotion of Science, the University of Strasbourg, CNRS (France), and a Nobel Laureate Fellowship of the Max Planck Society. Author contributions: S.S. designed experiments, purified bd oxidase and mAb20A11, prepared grids, collected cryo-EM data, built the model, drafted the manuscript, and prepared figures. A.H. prepared grids, collected cryo-EM data, refined the structure, drafted the manuscript, and prepared figures. D.J.M. aligned the microscope and collected cryo-EM data. M.R. performed activity measurements and processed HDX-MS data. M.L.E. collected, processed and evaluated HDX-MS data. J.M.-C. sequenced CydH via MS/MS. A.N. and F.M. performed potentiometric titrations and evaluated data. H.M. synthesized AD3-11 and provided the compound. R.B.G. initiated the project and supplied strains and plasmids. J.D.L. evaluated MS data. J.S. provided strains and samples for potentiometric titrations. P.H. evaluated electrochemical data. W.K. and H.M. supervised the project. Competing interests: The authors declare no conflict of interest. Data and materials availability: Cryo-EM maps are deposited at the Electron Microscopy Data Bank under accession numbers EMD-4908, EMD-4911, and EMD-4916. The model of the cytochrome bd-I oxidase structure was submitted to the Protein Data Bank with accession number 6RKO. All other data are presented in the main text or the supplementary materials.
View Abstract

Stay Connected to Science

Navigate This Article