Structure of the bc1- cbb3 respiratory supercomplex from Pseudomonas aeruginosa

Abstract

Energy conversion by electron transport chains occurs through the sequential transfer of electrons between protein complexes and intermediate electron carriers, creating the proton motive force that enables ATP synthesis and membrane transport. These protein complexes can also form higher order assemblies known as respiratory supercomplexes (SCs). The electron transport chain of the opportunistic pathogen Pseudomonas aeruginosa is closely linked with its ability to invade host tissue, tolerate harsh conditions, and resist antibiotics but is poorly characterized. Here, we determine the structure of a P. aeruginosa SC that forms between the quinol:cytochrome c oxidoreductase (cytochrome bc₁) and one of the organism's terminal oxidases, cytochrome cbb₃, which is found only in some bacteria. Remarkably, the SC structure also includes two intermediate electron carriers: a diheme cytochrome c₄ and a single heme cytochrome c₅. Together, these proteins allow electron transfer from ubiquinol in cytochrome bc₁ to oxygen in cytochrome cbb₃. We also present evidence that different isoforms of cytochrome cbb₃ can participate in formation of this SC without changing the overall SC architecture. Incorporating these different subunit isoforms into the SC would allow the bacterium to adapt to different environmental conditions. Bioinformatic analysis focusing on structural motifs in the SC suggests that cytochrome bc₁-cbb₃ SCs also exist in other bacterial pathogens.

Keywords: Pseudomonas aeruginosa; cryoEM; electron transport chain; respiratory supercomplexes; structure.

Fig. 1.

Bacterial cyt. bc₁ and cbb₃. (A) Rhodobacter capsulatus cyt. bc₁ [PDB: 5KLI (28)] and (B) Pseudomonas strutzeri cyt. cbb₃ [PDB:3MK7 (29)]. Subunits are labeled with colored text, electron carriers and substrate-binding sites with black text, substrates and products with purple text, electron transfers are represented with yellow arrows, and the conformational change of the Rieske head domain is represented with a blue arrow. This conformational change brings the FeS cluster from a position adjacent to Q_P site to a position next to heme c₁.

Fig. 2.

Architecture of the cyt. bc₁c₄c₅–cbb₃ SC. (A) CryoEM map and (B) atomic model for cyt. bc₁c₄c₅–cbb₃ SC with ubiquinone (UQ) and glyco-diosgenin (GDN) in gray. (C) Atomic model for cyt. c₅, N-terminal tail was built into density while linker and heme-binding domain are from AlphaFold prediction. (D) Interaction between cyt. bc₁c₄c₅ and cyt. cbb₃ within the membrane regions of the proteins. The sequence between L250 and R253 of cyt. c₁ forms contacts with the region near T94 of CcoN. (E) Close up view of cyt. c₁ ß-hairpin extension, the region between T187 and F178 forms close contacts with CcoO and CcoP of cyt. cbb₃.

Fig. 3.

Dynamics and activity of the cyt. bc₁c₄c₅–cbb₃ SC. (A) Electronic connection between the Q_P site of the cyt. bc₁c₄c₅ monomer and the binuclear center of cyt. cbb₃. Hemes, substrate-binding sites, and ubiquinone (UQ) are labeled with black text; substrates and products with purple text; electrons are indicated with yellow; and inhibitors are labeled with green text and arrows. Electrons from UQ at the Q_P site bifurcate, with one traveling to the FeS cluster (i) and the other through the b hemes to the Q_N site. The FeS undergoes a conformational change (ii) allowing for electron transfer to heme c₁ and then through both hemes of cyt. c₄. (iii) Cyt. c₅ mediates electron transfer from the second heme of cyt. c₄ to the c hemes of cyt. cbb₃ (see B). Electrons can then travel through the c hemes of cyt. cbb₃ to heme b and then the binuclear center where reduction of oxygen occurs (iv). (B) SC subpopulations showing cyt. c₅ density next to cyt. c₄ (i), where electron transfer from cyt. c₄ to the cyt. c₅ heme can occur (yellow arrow) and next to cyt. cbb₃ (ii) where electron transfer from the cyt. c₅ heme to cyt. cbb₃ can occur. (C) Oxygen consumption measurements with NADH, NDH-2, coenzyme Q₁, and SC all present (blue bar). Control experiments, with all reagents except coenzyme Q₁ (orange bar) or NADH (green bar) present. Oxygen consumption with all reagents plus CN^- (red bar) and antimycin A (purple bar). All experiments were performed in triplicate, with two replicates from preparations of enzyme from one growth and the other replicate from a preparation of enzyme from another growth.

Fig. 4.

Modularity of the cyt. bc₁c₄c₅–cbb₃ SC. (A) Cyt. cbb₃ model, sidechains that are variable between isoforms are shown in blue. (B) Isoforms detected by mass spectrometry. (C) Map and model for Y89-W102 and Y89-W93 region of the CcoP₁ (purple structure) and CcoP₂ (pink structure) following computational separation of isoforms. (D) Model in map fits for different CcoN isoforms; CcoN₁, CcoN₂, and CcoN₄ atomic models in green, purple, and red, respectively, suggesting that CcoN₄ is the predominant isoform in the sample.