# 19.1: Electron-Transfer Reactions in Mitochondria

## An Overview of Mitochondrial Electron Transport

The main oxidizing agent used during aerobic metabolism in the citric acid cycle is NAD+ (although FAD is used in one step). In the process, these oxidizing agents get reduced to form NADH (and FADH2). Unless NAD+ is regenerated, glycolysis and the Kreb's cycle will grind to a halt. This occurs in anaerobic condition when NADH formed in glycolysis is reoxidized back to NAD+ by pyruvate which is converted to lactate.

Under aerobic, we are continually breathing one of the best oxidizing agents around, dioxygen. NADH is oxidized back to NAD+ not directly by dioxygen, but indirectly as electrons flow from NADH through a series of electron carriers to dioxygen, which gets reduced to water. This process is called electron transport. No atoms of oxygen are incorporated into NADH or any intermediary electron carrier. Hence the enzymes involved in the terminal electron transport step, in which electrons pass to dioxygen, is an oxidase. The enzymes of the Kreb's cycle and electron transport are localized in mitochondria, which is shown in Figure $$\PageIndex{1}$$ below.

By analogy to the coupling mechanism under anaerobic conditions, it would be useful from a biological perspective if this electron transport from NADH to dioxygen, a thermodynamically favorable reaction (ΔG of about -55 kcal/mol), were coupled to ATP synthesis. It is! For years scientist tried to find a high energy phosphorylated intermediate similar 1,3-bisphosphoglycerate (formed by glyceraldehyde-3-phosphate dehydrogenase in glycolysis), which could drive ATP synthesis in the mitochondria. None could be found. A startling hypothesis was put forward by Peter Mitchell, which was proven correct and for which he was awarded the Nobel Prize in Chemistry in 1978. The immediate source of energy to drive ATP synthesis was shown to come not from a phosphorylated intermediate, but a proton gradient across the mitochondrial inner membrane. All the enzymes complexes in electron transport are localized in the inner membrane of the mitochondria, as opposed to the cytoplasmic enzymes of glycolysis. A pH gradient is formed across the inner membrane occurs in respiring mitochondria. In electron transport, electrons are passed from mobile electron carriers through membrane complexes back to another mobile carrier.

There are four inner mitochondrial membrane complexes involved in the flow of electrons to dioxygen. The first, Complex I, passes two electrons from NADH, which engages like NAD+ in 2 electron redox steps), to a flavin derivative, FMN, covalently attached to Complex I. The oxidation of NADH occurs, as expected, by hydride transfer. The reduced form of FMN then passes electrons in single electron steps (characteristic of FAD-like molecules, which can undergo 1 or 2 electrons transfers) through the complex to the lipophilic electron carrier, ubiquinone, UQ as shown in Figure $$\PageIndex{2}$$ below.

This then passes electrons through Complex III to another mobile electron carrier, a small protein, cytochrome C. This protein has covalently attached heme containing a central iron ion that can undergo one electron redox reactions (Fe3+ + 1e- ↔ Fe2+). Reduce cytochrome C (Fe2+) passes electrons through Complex IV, cytochrome C oxidase, to dioxygen to form water. At each step electrons are passed to better and better oxidizing agents, as reflected in their increasing positive standard reduction potential. Hence the oxidation at each complex is thermodynamically favored. 4 electrons must be added to dioxgen, along with 4H+ to form water.

$\ce{O_2 + 4e^{-} + 4H^{+} → 2H_2O}.$

What happened to Complex II? Complex II (also called succinate:quinone oxidoreductase) contains the citric cycle enzyme succinate dehydrogenase, which catalyzes the oxidation of succinate to fumarate by FAD. The bound FADH2 does not dissociate from the enzyme, which is functionally dead for further formation of fumarate until it is oxidized back to FAD by additional components of Complex II. In this process, electrons from FADH2 are passed to the same lipophilic mobile electron carrier from Complex I, ubiquinone. Reduced ubiquinone then can transfer electrons to cytochrome C through Complex III.

For Complexes I, III, and IV, the energy released by the oxidative event is used to drive protons from the matrix to the intermembrane space of the mitochondria. The oxidative energy releases is NOT used to form a high energy mixed anhydride as we saw in the glyceraldehyde-3-phosphate dehydrogenase reaction from glycolysis. The resulting proton gradient collapses in a thermodynamically favorable process that energetically drives the endergonic synthesis of ATP by the last inner membrane complex, ATP synthase, which is sometimes called Complex V. This aerobic coupling of electron transfer (from NADH to dioxygen) and ATP synthesis, is shown schematically in Figure $$\PageIndex{3}$$ below. The electron transfer is usually called electron transport, which might suggest to you that it requires energy as in the case of active transport. It does not, however, since it is strongly favored thermodynamically with dioxygen used as the final electron acceptor.

You may wish to question the need for such a complicated series of reactions to move the simple electron from one chemical species to another and a simple H+ from one side of a membrane to another to form a proton gradient which collapses anyway. Not only that, but the movers of these species are aligned in a membrane to a way which optimize the processes and allow the vectorial, nonrandom flow of electrons and protons. This is what evolution has produced. One reason must to maintain optimal control over the efficiency of the coupled reaction. Another powerful explanation is to reduce dangerous side reactions. In redox chemistry, this involves the formation of highly reactive free radicals that can react quickly and uncontrollably with any species nearby. This is especially true when dioxygen is involved, as we saw in Chapter 13.4 - The Chemistry and Biology of Dioxygen. It takes 4 electrons to fully reduce dioxygen to water. If this process is interrupted, highly reactive oxygen species (ROS) like superoxide and peroxide can form, which can damage membrane, proteins, and nucleic acids.

Before we explore each complex in more detail, let's get lower resolution view of the entire mitochondrial electron transport and ATP synthesis (often called oxidative phosphorylation or ox-phos) which is shown in cartoon form in Figure $$\PageIndex{4}$$ below.

Especially note:

• the subunit complexity of each Complex;
• the location of the subunits (many integral membrane proteins, others more hydrophilic and located in the polar matrix);
• the mobile electron carriers NADH/NAD+, lipophilic UQH2/UQ used for both complex I and II, cytochrome C (a protein), and O2;
• the number of H+s moved from the matrix to the inner membrane space through Complexes I, III and IV and back thourgh ATP synthase.

Also note that the prokaryotic structures are displayed, even though they are placed in the mitochondrial membrane! Bacteria do not have organelles like mitochondrial but they have the same complexes inserted their plasma membranes. For bacterial, protons are ejected into the periplasm space between the plasma membrane and cell wall.

Diagrams like the one shown in Figure $$\PageIndex{4}$$ above are really useful in simplifying complex biological systems. At the same time, they constrain our ideas about how, in this case, the individual complexes are arranged in the membrane. Much evidence shows that Complex I, III and IV (those that pump protons) form a very large supercomplex called the respirasomes, which has some 80 subunits in mammals. Clustering the subunits in a supercomplex increases the likelihood that shared mobile electron carriers wont' diffuse away, but rather continue passing electrons along the electron transport chain. As mentioned earlier, some reactions generate reactive oxygen species, which need to be kept in the complex to avoid their toxic effects. The stability of the individual complexes is also most likely increased in the larger aggregate.

Using cryoelectron microscopy (cryoEM), the structures of two large super- and mega-complexes have been solved showing the arrangement of all of the respiratory complex. Two different structures showing the SuperComplex-I1III2IV1 (SCI1III2IV1) and MegaComplex-I2III2IV2 (MCI2III2IV2) are shown below in Figure $$\PageIndex{5}$$.

Panel (A) shows human SCI1III2IV1 (PDB ID: 5XTH) in side view (along the inner membrane) and top-down view. Note that CI is adjacent to CIII (located in the center) which is adjacent to CIV. The is the expected orientation given the flow of electrons from NADH to dioxygen. In panel (B) substrates and products are include into the side view, along with the flow of H+ (from all of the complexes) into the matrix. Panel C shows the added cofactors (FeS clusters, FMN, hemes, etc). Panel (D) shows a very beautiful topdown view ofthe human megacomplex, which has two copies of each of subunits. Again note the centrality of Complex III.

Figure $$\PageIndex{6}$$ below shows another model of the megacomplex with Complex II (CII, which does not translocated protons) added. It also shows a tetramer of Complex V (ATP synthase)

With this low resolution background view of the entire mitochondrial oxidative phosphorylation pathway done, now let's explore each pathway in more detail.

## Complex I - NADH-quinone oxidoreductase

Complex I is located in the inner mitochondrial membrane in eukaryotes and in the plasma membrane of bacteria. Bacteria have only 14 subunits while the mammalian Complex I has 45, including the core 14 found in bacteria.  Another 31 are term supernumerary (in excess of the "normal number") subunits that support the function of the core. The significance of the extra mammalian subunits is still unclear. A cartoon model with the actual crystal structure of the Thermus thermophilus (bacterial) complex is shown in Figure $$\PageIndex{7}$$ below. The hydrophilic or peripheral domain catalyzes electrons transfer while the membrane domain (encoded by mitochondrial DNA) is involved in active transport of protons. It is one of the biggest enzymes around, so its exciting to try to understand it.

Figure $$\PageIndex{8}$$ below shows an interactive iCn3D model of the mammalian respiratory complex I (5LDW)

Figure $$\PageIndex{8}$$: Mammalian respiratory complex I (5LDW). (Copyright; author via source).
Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/i...xVjRPLLgaLCi17 (both load slowly given the size of the structures). For the external link, use the mouse scroll to zoom the structure to see the FeS clusters and FMN, all of which are labeled.

The red and blue parallel "dummy" atoms represent the boundary of the mitochondrial inner membrane.

### Electron Transfer in Complex I

Electron flow occurs from NADH to UQ through a series of one electron carriers in the hydrophilic or peripheral domain of complex 1. Initial handoff of electrons occurs to a flavin cofactor, FMN, and then through a series of Fe/S clusters, to ubiquinone. These electron acceptors include tetranuclear (Fe4S4) and binuclear (Fe2S2) iron-sulfur clusters, a FMN flavin mononucleotide, and a Mn (II) ion. These are shown in Figure $$\PageIndex{9}$$ below.

The tetranuclear Fe4S4 cluster is based on the cubane structure with Fe and S occupying alternating corners of a square in a tetrahedral geometry. Each Fe is also coordinated to thiolate (RS-) from coordinating cysteines and also sulfide (S2)-anions. The actual structure is a distorted cube as shown in Figure $$\PageIndex{10}$$ below, along with that of the binuclear Fe2S2 cluster, whose bond angle also deviate from those in a perfect tetrahedron.

Many possible micro-redox states with different standard reduction potentials are possible for tetranuclear Fe4S4 clusters, much as polyprotic acids have multiple pKa values. The two relevant for the tetranuclear clusters in Complex I are shown below:

a. Fe(II)Fe(III)3/S2(CysS)21- + e- ↔ Fe(II)2Fe(III)2/S2(CysS)22- (less positive standard reduction potential)

b. Fe(II)2Fe(III)2/S2(CysS)22- + e- ↔ Fe(II)3Fe(III)/S2(CysS)23- (more positive standard reduction potential)

##### pKas and E0 values

You have studied Bronsted acids in many courses and know the factor that affect the pKa of different acids that are structural similar. These include the initial charge on a potential proton donor. The relative strength of a series of structural similar acids can be predicted by examining the factors that stabilize the negative charge on the resulting conjugate base. The more stable the conjugate base, the more acidic the parent acid. Factors that stabllize a negative conjugate base are electronegativity of the atom holding the negative charge, resonance which might delocalize the charge, inductive/electron release effect of substitutents and the hybridization of the atom holding the charge (sp aniond are more stable due to more s character with electrons pulled more closely to the nucleus).

What about a single acid? The pKa of given acid is a constant for a given set of conditions (temperature, solvent polarity, etc) but it can be altered by changing conditions. The diagram below in Figure $$\PageIndex{11}$$shows way to "tip" the pKa of a given acid (in this case a simple carboxylic acid) to a lower value, making the acid stronger in a the new set of conditions.

The same ideas apply to the strength of a potential oxidizing agent (electron acceptor) as described by the E0 value (higher for stronger oxidizing agents/electron acceptors). Figure $$\PageIndex{12}$$ below shows the factors that could alter the E0 value for a binuclear FeS cluster. In this case, changes in coordinating ligands, analogous to changes in substituents for an acid) are included since most students have probably spent more time thinking about pKa values than E0 values.

In either case, it really about Le Chatelier's principle. The equilibrium can be shifted to the right (stronger acids with lower pKa or stronger oxidizing agent/electron acceptor with a higher E0) by preferentially stabilizing products compared to the initial reactants.

Electrons are passed singly to oxidized UQ in one electron steps to form UQH2.

Fe-S clusters are synthesized predominately in the mitochondria where they serve as redox cofactors in electron transport as described above. They are ubiquitous in all life forms and serve roles in addition to redox cofactors per se as they serve structural roles in proteins and are used in redox signaling within the cell as they change oxidation states. Many proteins that interact with DNA (repair enzymes, polymerases and helicases) contain an Fe-S cluster.

Evidence suggests that they placed critical roles in the abiotic evolution of life in the absence of oxygen as a a terminal electron acceptor in exergonic oxidation reactions. When oxidation became available, they became potentially became toxic to the cell as the Fe2+ could participate in reactions (such as the Fenton reaction) leading to the generation of deleterious reactive oxygen species (such as superoxide). To prevent toxicity, when delivered to the cytoplasm and nucleus they are carried and delivered by cytoplasmic iron-sulfur assembly (CIA) proteins.

Figure $$\PageIndex{13}$$ below shows a clear models of the complex and path for electron flow.

Electrons "tunnel" from one cofactor to another in Complex I through quantum mechanical "hops". Since all of the redox cofactors after the favored hand off of electrons from NADH to FMN are FeS clusters, which would have similar E0 values (modulated only by local environments), there is no significant thermodynamic barrier to electron flow. The final N2 FeS cluster has a E0 value about 100 meV higher than the started ones so it most readily accepts an electron.

The separation of the hydrophilic mitochondrial matrix domains, where electron transfer occurs, from the transmembrane domains, where proton transfer occurs, is an elegant design that keeps the key and reactive particles, e- and H+, from reacting with each other. The electrons stay on one side of the membrane, as the protons move to the other side (until the return in the last step when Complex V produces ATP).

### Proton Transfer in Complex I

Proton transport occurs in the membrane domain. Available evidence suggests that 4 protons move from the cytoplasm to the periplasmic space against a concentration gradient during a catalytic cycle of Complex I in bacteria and other organisms.

Figure $$\PageIndex{14}$$ below shows multiple views of both key electron transfer step and H+ ejection from the matrix into the inner membrane space. Key to proton transports are the discrete membrane domains shown in the bottom left part of panel (a). There are 4 such domains involved from left to right, each one associated with the transport of 1H+. The individual membrane domains have different names and abbreviations depending on the organism. They are from left to right:

• Nqo12, NuoL or ND5
• Nqo13, NuoM, or ND4
• Nquo14, NuoN or ND2
• Nqo8, NuoH, or ND1.

The Nqo8, NuoH, or ND1 domain is proximal and closely linked to the site for reduction of UQ to UQH2.

The NuoN/Nqo14/ND2, NuoM/Nqo13/ND4 and NuoL/Nqo12/ND5 domains act as and are structural similar to antiporters. Note also a very long helix that crosses the left three-most antiporter domains. This helix runs parallel to the membrane and probably acts a lever allowing coupled conformational changes of the three H+ transporters. We looked in the previous chapter at transporters and showed both in cartoon models and structural models that they alternated between an inward and outward open conformations.

Figure $$\PageIndex{15}$$ below shows more details of the antiporter domains and the unique structural features allowing proton transport.  Each antiporter domain has two 5-helix bundles that display pseudo-symmetry with transmembrane helices number TM4-8 and TM9-13.  The next to last helix in each bundle (7 and 9) are interpreted, with unbroken parts of the helices number TM7a/b and TM12a/b.  Running along the middle of each of the antiporter-like domains of  the membrane domain are a series of conserved charged amino acids (aspartate, glutamate, lysine and histidine) that participate in hydrogen bonding network. Molecular dynamics simulation show water molecules (panel b), represented by red dots (for the oxygen atoms), transiently occupying a H-bonded network of waters that serve as a conduit for H+ flow. An unbound "naked" proton (H+) would not persist long in any aqueous biological system.  Panel C shows the conformational flexibility in broken interrupted helices that presumably gates the proton pore open and closed.

Figure $$\PageIndex{15}$$: Structure of the proton-pumping membrane domain in complex I.

(a) The membrane domain of complex I showing conserved buried charged/hydrophilic residues. NuoL (red); NuoM (blue); NuoN (yellow); NuoH (green); Ubiquone - Q (blue van der Waals representation). Inset: The conformation of the Lys/Glu ion pair (here Lys-204M/Glu-123M) can modulate the pKa of the middle Lys (Lys-235M).

(b) The NuoM  antiporter-like subunit with water molecules (red dots).  The arrow shows the flow of protons through a H-bond linked network of waters that "visit" the structure during μs molecular dynamics simulation.

(c) Snapshot of structures obtained from molecular dynamic simulations of open (in blue) and closed (in red) proton channels from the N-side in Nqo13 (NuoM/ND4), showing conformational changes in the broken helix element.

(d ) The structural symmetry of the antiporter-like subunits with an N-side input channel near broken helix TM7b and output channel near broken helix TM12b.

Kaila VRI. 2018 Long-range proton-coupled electron transfer in biological energy conversion: towards mechanistic understanding of respiratory complex I. J. R. Soc. Interface 15: 20170916. http://dx.doi.org/10.1098/rsif.2017.0916. Creative Commons Attribution License http://creativecommons.org/licenses/by/4.0/,

How are the reduction of ubiquinone and proton transport coupled?  The mechanism likely involves conformational changes in flexible regions in the interface of the hydrophilic and membrane domains of the complex.

### Inhibition of Complex I

Complex I is inhibited by more than 60 different families of compounds. They include the classic Complex I inhibitor rotenone and many other synthetic insecticides/acaricides. The classes include: Class I/A (the prototype of which is Piericidin A), Class II/B (the prototype of which is Rotenone) and Class C (the prototype of which is Capsaicin). They appear to bind at the same site. Figure $$\PageIndex{16}$$ below shows their structures.  From the structure of the 3 prototypes, what are the characteristics of the pharmacophore, the “ideal binding ligand”? Where do they likely bind? How “promiscuous” is the binding site?

Many devastating neurological diseases are associated with defects in Complex I. In addition to major problems with oxidative ATP production, reactive oxygen species (ROS) increase. The major sites for generation of ROS are Complex 1 and Complex III. Given the locations of the electron carriers at the periphery and internal within the protein complex, which electron carriers might most readily leak electrons to dioxygen? What ROS is likely to form in the process?

Inhibitors might block access of UQ or conformational changes necessary for final reduction of the ubiqinone free radical. Class A inhibitors dramatically increase ROS production. The actual site of ROS production in Complex I is a bit controversial. One possible electron donor to dioxygen is FMN since both can engage in 1 e- transfers. Mutants that lack N2 iron-sulfur cluster showed ROS production.

In submitochondrial preparations, normal Complex I activity occurs (which leads to formation of a sustained proton gradient). Also reverse electron transport, powered by an artificial proton gradient can occur, which leads to the reduction of NAD+, as shown in Figure $$\PageIndex{17}$$ below.

Figure $$\PageIndex{17}$$:  Overview of forward and reverse electron transport in Complex I.

A summary of finding on superoxide production by Complex I show that:

• superoxide production is inhibited by flavin site inhibitors but not Q site inhibitors.
• Reverse electron transport superoxide production is inhibited by both flavin and Q site inhibitors

Figure $$\PageIndex{18}$$ below shows an expanded view of electron transport and site where ROS production is enhanced.  Refer to the legend for details.

## Complex III

Complex III is a complicated, multisubunit protein that is at the heart of the respirasome. The subunits involved in electron transfer are cytochrome b, cytochrome c1and the Rieske iron sulfur protein (ISP). Cytochrome b has two hemes. One is cyto b562 which is also called the low potential heme or cyto bL. The other is cyto b566 which is called the high potential heme or cyto bH. The cytochrome c1 subunit has one heme. The Reiske iron sulfur protein contains a Fe2S2 cluster.  Figure $$\PageIndex{19}$$ below shows the relative position of the bound mobile electron carrier, cytochrome C, and the internal ones, the Rieske Fe/S cluster and cytochrome bL and bH. Note also the molecule stigmatellin A, which binds to the site where UQ becomes reduced (called the Qo site) and inhibits the complex. This shows that UQ/UQH2 are in position to react readily with the Rieske canter and cytochrome bL heme.

Figure $$\PageIndex{20}$$ below shows an interactive iCn3D model of a eukaryotic respiratory complex III (3CX5)

Figure $$\PageIndex{20}$$: eukaryotic respiratory complex III (3CX5). (Copyright; author via source).
Click the image for a popup or use this external link:https://structure.ncbi.nlm.nih.gov/i...5riT4GSSGhCHfA.  (loads slowly given the size of the structures).

The Rieske iron sulfur protein has a Fe2S2 iron sulfur cluster which differs from other such clusters in that each Fe is also coordinated to two His side changes, as shown in Figure $$\PageIndex{21}$$ below. Alterations in H bonds to the histidines and to the sulfurs in the complex can dramatically affect the standard reduction potential of the cluster.

Figure $$\PageIndex{21}$$:  The Rieske Fe2S2 cluster

As with complex I and IV, proton and electron transfer are coupled processes. However, in contrast to Complex I, in which protons pass through protein domains that have homology to K+/H+ antiporters, and Complex IV, in which they pass through a combination of a water channel and the H-bond network, the protons in Complex III are carried across the inner membrane by ubiquinone itself. Two reduced ubiquinones (UQH2) from complex I pass their four matrix-derived protons into the inner membrane space. In the process four electrons are removed in a multiple step process called the Q cycle.

The two electrons from each UQH2 take different paths. One electron moves to a Fe/S Rieske cluster and the other to cytochrome bL. The electrons moved to the Rieske center then moves to cytochrome c1s and then to the mobile electron carrier cytochrome C which is bound to the complex in the intermolecular space. The electrons moved to cyto bLs are transferred to cytochrome bH in the complex. Though this latter path, two electrons (from two UQH2) are then moved to oxidized UQ, and two matrix protons are added to reform one UQH2. Hence, only one UQH2 participates in the net reaction shown as below.

QH2 + 2 cyto c3+ + 2H+matrix → Q + 2 cyto c 2+ + 4H+IMS

This net overall reaction, the Q cycle, is illustrated below. This net overall reaction, the Q cycle, is illustrated in Figure $$\PageIndex{22}$$ below.

Figure $$\PageIndex{22}$$: Q cycle of Complex III

Once again, there are no “proton” channels or H bonded networks in the protein for proton transfer across the inner membrane.

Another way to think about the electron transfer process from UQH2 to cytochrome C is that the 2 electrons from UQH2 take two different paths, one a high potential path to the Rieske center and on to cytochrome C, and another low potential path to the bL heme and on to the bH heme and then to UQ to reform UQH2 (see figure above).

Complex III, along with Complex I, can also produce unwanted reactive oxygen species (ROS). Only three of the protein subunits, cytochrome b (with the bL and bH hemes), cytochrome c1, and the Rieske iron sulfur protein (ISP) are involved in electron transfer, so one of those is mostly likely involved in ROS production. Experiments and mathematical models support a mechanism that involves a reduction of UQ by addition of one electron from cytochrome bL to form UQ. which then passes its electron on to dioxygen to form superoxide (O2-.).

As two ubiquinones must bind to the complex, there must be two proximal sites. One is the Qi site where oxidized UQ binds and receive an electron. The other is the Qo site where UQH2 binds.

From a kinetic perspective, the first UQH2 binds and transfers two electrons, one to the Rieske cluster (and on to cytochrome c1 and then to cytochrome C) and one to cytochrome bL (and on to heme bH) and then to an oxidized UQ bound at the Qi site. The UQ. radical is stabilized by the adjacent bH heme which has a lower affinity for electrons. Now a second UQH2 binds to the Qo site, and transfers two electrons, again one via the Rieske cluster and the second through cytochrome bL and bH to the UQ. radical present at the Qi site to form UQH2 after two protons are transferred to it from the matrix.

Now a second UQH2 binds to the Qo site, and transfers two electrons, again one via the Rieske cluster and the second through cytochrome bL and bH to the UQ. radical present at the Qi site to form UQH2 after two protons are transferred to it from the matrix.

Antimycin A, an extremely toxic drug, binds to the UQ Qi site and hence blocks electron transfer from cytochrome bL to bH at the Qi site. Heme bL can then pass its electron to dioxygen to produce superoxide.

## Complex IV - Cytochrome C oxidase (CCOx)

This is the final complex in the electron transport chains and is the one that passes 4 electrons  and 4 "substrate" protons to dioxygen to produce water. In addition it moves as many as 4 H+s from the matrix to the inner membrane space, for a net change of 8 H+s.  Here is a modified version of the Equation 1 that represents the transported Hs.

$\ce{O_2 + 4e^{-} + 4H^{+} + 4H^{+}_(in) → 2H_2O + 4H^{+}_(out)}.$

The structure of complex IV is shown in the Figure $$\PageIndex{23}$$ below.  KEGG pathways (with permission).

The 4 H+s that are moved across the membrane are not transported by protein domains that are structural equivalents of antiporters, as in the case for Complex I.  Rather elaborate H-bonded water channels with contribution from a space of polar backbone and side chain H-bond donors and acceptors are involved in proton transport.  Hence a key question to understand is how electron transfer is couple to H+ transfer through this hydrogen-bonded network.  Since its quite complicated we will present multiple different but seemingly reduntant figures of low to high resolution to help explain this coupled process.

Figure $$\PageIndex{24}$$ below shows an interactive iCn3D model of the 14-subunit human cytochrome c oxidase (5z62) with bound cofactors required for electron transfer.  Note that it does not show cytochrome C.

Figure $$\PageIndex{24}$$: 14-subunit human cytochrome c oxidase (5z62). (Copyright; author via source).
Click the image for a popup or use this external link:https://structure.ncbi.nlm.nih.gov/i...XoHRyCkpeRN917.  (loads slowly given the size of the structure).

Reduced cytochrome C, a mobile electron carrier protein, binds to Complex IV in the inner membrane spaces and singly passes 4 electrons to a series on electron carriers (oxidzing agents) in the protein, eventually to dioxogen as the terminal electron acceptor.  Figure $$\PageIndex{25}$$ below shows the arrangement of the electron acceptor cofactors in Complex IV.  Electron transfer presumably occurs through quantum mechanical tunneling.

Figure $$\PageIndex{25}$$: Electron acceptor cofactors in Complex IV.

Cytochrome C, the initial “substrate” of this complex, delivers electrons from its heme cofactor to a dinuclear copper cluster, Cu A, where the copper ions are collective coordinated to two histidines and two cysteine and a methionine.  Figure $$\PageIndex{26}$$ below shows an interactive iCn3D model  of the coordinating ligands for Cu A cluster in Complex IV (5z62)

Figure $$\PageIndex{26}$$: coordinating ligands for Cu A cluster in Complex IV (5z62). (Copyright; author via source).
Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/i...7ARphDU4oyDS4A

From there electrons flow to an adjacent heme a (low spin), which has two coordinating histidine ligands.  This heme hence does not bind dioxygen.  From there electrons move to another heme, heme a3 (high spin) and then finally to dioxygen which is coordinated to the Fe in heme a3 and to an adjacent Cu B.  Heme a3 has only coordinating 1 histidine ligand, which allow dioxygen binding to the unligated site.  The heme a3 Fe:Cu dinuclear cluster is unique among all hemes. Figure $$\PageIndex{27}$$ below shows an interactive iCn3D model  of the coordinating ligands for the hemes and Cu B in Complex IV (5z62)

Figure $$\PageIndex{27}$$: coordinating ligands for CuB, heme a and heme a3 in Complex IV (5z62). (Copyright; author via source).
Click the image for a popup or use this external link:  https://structure.ncbi.nlm.nih.gov/i...4EUooXkwh13Fm8

First let’s consider the transfer of electrons from heme a to heme a3 to dioxygen (we will consider entry of electrons into the complex later). If dioxygen, a substrate for the reaction, dissociated from the heme a3 Fe before it were completely reduced, toxic ROS would result.  This suggest a reason for evolution of this key enzyme to have produced the unique heme a3 Fe:Cu dinuclear cluster

Heme a and a3 vary from the heme in hemoglobin as they both have a formyl group replacing a methyl and a hydroxyethylfarnesyl group added to a vinyl substituent. Its structure is shown in Figure $$\PageIndex{28}$$ below. What is its overall charge of the heme in its reduced state? In its oxidized state?

Figure $$\PageIndex{28}$$:  Heme-Formyl group of Cytochrome C Oxidase

As the heme are prosthetic group, they must bind to the apoprotein. The hydroxyethylfarnesyl group, with its nonpolar tail, facilitates binding of the heme to both the Fe+2 and Fe+3 forms by about 6.3 kcal/mol.  The formyl group in heme a increases the E0 value (i.e. make the heme a better oxidizing agent) by about 179 mV (4.1 kcal/mol).  The effect appears to be mediate by the binding of the heme group to the protein again, only in this case there is a differential affect. The formyl group preferentially stabilizes the binding of the Fe+2 heme compare to the Fe+3 heme, which in a thermodynamic cycle, would promote binding of an electron to Fe+3-heme (increasing its E0 value) to form the more stably bound Fe+2 heme.

The key challenge has been to understand the redox coupling to H+ transport. How is this done? The electrostatic environment of the hemes must be considered.  In heme a, the two axial ligands are uncharged imidazole side chains of histidines. The overall charge on Fe2+-heme is zero as the two proprionates cancel the charge on the Fe2+.  However there is a net +1 charge in the heme in the Fe3+-heme.  This charge could be delocalized within the conjugated pi electrons of the planar heme, which could contribute to deprotonation of a side chain near heme a in process which would couple redox and H+ movement.  The formyl group of heme a is coplanar with the heme ring in both Fe3+-heme and Fe2+-heme, which allows it to participate in the pi conjugated electron system of the ring.  As the formyl group form an H bond with Arg 38, change in oxidation might affect Arg 38 as well.

Many other amino acids are involved as well.  One is Asp-51 of subunit I which contains heme a.  This undergoes a conformational change which moves it to the surface of the inner membrane on reduction.  It is near the matrix in the oxidized state. Hence it is likely involved in a proton transport pathway.   In fact, crystal structures of oxidized and reduced Complex IV show water channels and small “cavities” which calculations show can hold 1-3 water molecules. Hence groups around the heme, including R38 are assessable to water. Water would be involved in H+ transport thought "handshaking" transfer of a H+ through a hydrogen-bonded network of waters and selective side chains.  Some of the amino acid residues associated with the water channels are shown in Figure $$\PageIndex{29}$$: and include R38 (the one hydrogen bonded to the formyl group of the heme, S34, T 424, S461, S382, H413.

Figure $$\PageIndex{29}$$: Possible amino acids involved in  H+ transport through a hydrogen-bonded network in Complex IV

Another consequence of electron transfer to heme a involves the interaction of S382 and the farnesyl OH group, which are close in space and proximal to a water cavity. On reduction of the heme, a conformation change occurs which increases the S382-farnesyl OH group. What effect would this have on the interaction of the two and the S382-L381-Val380 localized conformation? A new water cavity appears to emerge on reduction in this region. How might this impact proton transfer from the matrix?

Now let’s consider the entry site of electrons into the complex and how they might influence proton transport. As mentioned above, Asp 51 (D51) appears to play a key role.  It is shown in Figure $$\PageIndex{x}$$ above and also in Figure $$\PageIndex{x}$$ below.  In the oxidized state, D51 interacts with two OH side chains (S205 and S441) and amide NH backbone groups but is not exposed to water. On reduction, D51 lies on the surface in an aqueous environment. Near D51 is Y440-S441. The backbone carbonyl group between 440 and 441 forms an “indirect” interaction with R38 (right panel in Figure $$\PageIndex{30}$$ below, fwhich we showed earlier is affected by the redox state of heme a. They are too far apart to form H bonds. Add two waters molecules and envision a bridging interaction between the carbonyl O and the side chain R38 via Y371. Likewise in your mind add a water to allow a bridged hydrogen bond interaction between Y371 also forms a H bond and the heme a proprionate.

Figure $$\PageIndex{30}$$:   Local environment of D51, R38 in hemes in Complex IV

A more nuanced understanding of the mechanism and linkage between H+ and e- movement derives from high resolution structures determined by Yano et al (2016). In their model (shown in the figures below based on the oxidized form of the protein, pdb 5b1a), protons from the negative (matrix) N side of the complex enter through a water channel and proceed to the positive (intermembrane side) through a H bond network (as described above and depicted below). These comprise the H-Pathway for H+ transfer across the membrane.  Two other H+ transfer pathways, the D- and K-pathways are use as source of substrate H+ for the 4 H+s added to dixogen to form 2H2Os.

Directional movement is mediated by proton:proton repulsion aided by an increase in + charge on heme a when it transfers an electron to heme a3. Of course, proton:proton repulsion would move protons in both directions. Reverse flow back through the water channel is prevented by a conformational change on oxygen binding that closes the channel.

Ultimately 4 electrons are transferred from cytochrome Cs (in single electron steps) to the dicopper cluster, CuA, and then sequentially to heme a to heme a3 (near the copper B ion) to dioxygen to form water. The motion of electrons and protons are coupled electrostatically.

Figure $$\PageIndex{31}$$ below gives an overview of these movements. The small red dots are the oxygen atoms of internal water molecules (the rest have rest have been removed using Pymol). It should be apparent, given the number and location of the internal water molecules, that many would be involved in the proton translocation pathways.

Figure $$\PageIndex{31}$$: Cartoon showing movement of H+s and e-s in Complex II

What's so interesting about this model is the detailed description of two types of protons, the ones that add to dioxygen and end up in water (substrate protons), and those that are vectorially (directionally) transported to the IMS. In their model, the H+s that end up being transported move through the water and H bond network through a connecting H bond link region to a Mg2+/water cluster. Since binding of oxygen leads to structural changes that closes off the water channel, all protons to be transported to the IMS must be bound in the cluster before dioxygen binding.

Figure $$\PageIndex{32}$$ below show that initially, 4 H+s move through the H system to the Mg2+/H2O cluster. Oxygen binding the closes the water channels. This buildup of positive charges would certainly lead to a enhanced electrostatic attractions for the next phase of the reaction, the movement of electrons into the heme cofactors. Additionally, the 4 H+s in the cluster are probably prevented from leaking to the P side through water that are proximal (see above figure) by proline cluster, which presumably restricts the dynamical motion of the protein in that region necessary for proton movement. The figure does not show charge changes in the electron carriers.

The figure below breaks downs the mechanism to show the addition of the first electron to the CuA (dicopper cluster), delivered from cytochrome C, and the subsequent transport of one proton from the fully proton-loaded Mg2+/water cluster after dioxygen binding. This figure does show the stepwise redox changes in the electron carriers.

After CuA receives an electron from cytochrome C, it donates it to heme a and not to heme a3, even though both are close. The extra negative on heme a facilitates proton pumping thought the H pathways shown.

Figure $$\PageIndex{33}$$ belows shows a cartoon description of the movement of protons and electrons through Complex IV.

Figure $$\PageIndex{33}$$:A schematic representation of the role of the Mg/H2O cluster in the proton-pumping mechanism of bovine heart CcO. The Mg/H2O cluster and the H-pathway are connected with the short hydrogen bond network, as labeled in A (a). The location of heme a is shown by Fea3+ or Fea2+ attached to the hydrogen bond network of the H-pathway. The CuA site is on the water-accessible surface of the Mg/H2O cluster. The conformation of the CuA-Mg2+ complex is shown by the shape of a line connecting the two metal ions. The proton-accepting sites are shown by the four hollows on the water-accessible surface of the Mg/H2O cluster. The O2 reduction site is shown by CuB1+ and Fea32+. The pumping and chemical (water-forming) protons are labeled in green and blue, respectively. Electron transfers are shown by red curves. A, overall catalytic cycle of CcO. B, a typical single electron transfer from CuA to the O2 reduction site, coupled with the uptake and release of protons. For the sake of simplicity, uptake of one chemical proton equivalent (the average number) is given in B. Various oxidation and ligand binding states in both CuB and Fea3 in the intermediate states shown by A (d) and B (d) are not included for the sake of simplicity. Yano N et al. J Biol Chem. 2016 Nov 11;291(46):23882-23894. doi: 10.1074/jbc.M115.711770. Epub 2016 Sep 7. PMID: 27605664; PMCID: PMC5104913. Creative Commons Attribution (CC BY 4.0)

Figure $$\PageIndex{34}$$ below converts the schematic representation into a "structural cartoon" representation showing one cycle of coupled electron and proton transfer into Complex IV.

The H- (for transport of 4H+s across the membrane) and the D- and K-pathways (sources of substrate 4H+s for water synthesis from O2 are shown in Figure $$\PageIndex{35}$$ below.

Figure $$\PageIndex{35}$$: The H-, D- and K-pathways in bovine heart Complex IV (5B1A). Atsuhiro Shimada et al. J. Biol. Chem.(2021) 297(3) 100967.  Creative Commons.  https://creativecommons.org/licenses/by/4.0/

Two black arrows show the D- and K- for substrate H+s that form water, while the blue arrow shows the H-pathway for the hydrogen bonding network of waters in the water channel.   The Mg/H2O cluster (blue area) is attached to the hydrogen-bond network of the H-pathway by a short hydrogen-bond network (gray area with water shown as blue sphere). The formyl group and one of the propionate groups of heme a are hydrogen-bonded with Arg38 and afixed watermolecule in the hydrogen-bond network of the H-pathway. Compare this diagram to the ones presented in the figures above it.  Again we show multiple representations to give readers many opportunities to conceptualize this complicated complex.

## Complex II - succinate:ubiquinone oxidoreductase (SQR) - succinate dehydrogenase

You've seen complex II before with another name, succinate dehydrogenase from step 6 of the citric acid cycle, which catalyses the following reaction.

The ΔGo for this reaction is about 0 kcal/mol so it is readily reversible .

However, something vital is left out of this description.  The above reaction appears to suggest that the FAD/FADH2 are readily diffusable and bind and unbind after catalysis. That's not true.  Instead, FAD/FADH2 are covalently attached to the enzyme so after one cycle of enzyme catalysis, the enzyme is dead.  Another substrate/product pair must interact with enzyme to reconvert the covalently attached FADH2 back to FAD so catalysis can continue.  That other pair is ubiquinone (UQ/UQH2).  A better description of the reaction equation is shown in Figure $$\PageIndex{36}$$ below.  Complex II clearly links the carboxylic acid cycle to electron transport!

Figure $$\PageIndex{36}$$:  Complex II - Succinate Dehydrogenase Reaction

As with the other mitochondrial electron transport complexes, Complex II is a multisubunit integral membrane protein. The hydrophilic section protruding into the matrix contains two proteins, a flavin binding protein (Fp) and an iron/sulfur protein (Ip).  The transmembrane domain has two "cytochrome binding proteins", a lighter one (CybL) and a heavier one (CybS).  The complex has binding sites for dicarboxlic acids (succinate/fumarate) and two binding sites for ubiquinone, a proximal higher affinity site (Qp) on the matrix side closer to the succinate binding site, and a distal lower affinity site (Qd) near .  The ubiquinone sites bind 2-thenoyltrifluoroacetone (TTFA) which acts as an inhibitor of ubiquinone reduction.  3-nitropropionate - (NO2)(CH2)2CO2- is structural similar to succinate - (CO2-)(CH2)2CO2- and is an inhibitor of succinate oxidation.

Figure $$\PageIndex{37}$$ below shows an interactive iCn3D model of the Mitochondrial Respiratory Complex II from porcine heart (1ZOY).  The chains are:

• hydrophilic FAD binding protein (lavender)
• hydrophilic FeS protein (blue)
• integral membrane large cytochrome binding protein (brown)
• integral membrane small cytochrome binding protein (green)

Figure $$\PageIndex{37}$$: Mitochondrial Respiratory Complex II from porcine heart (1ZOY) . (Copyright; author via source).
Click the image for a popup or use this external link:  https://structure.ncbi.nlm.nih.gov/icn3d/share.html?KKPWG9dW1ARq3wQf9.  (loads slowly given the size of the structure).

Given the role of the enzyme in both the mitochondrial citric acid and electron transfer pathways, you would expect high electron transfer rates between the dicarboxylic acid and the ubiquinone binding sites.   Any disruptions in electron transfer (by inhibitors or mutations) could lead to the production of toxic reactive oxygen species (ROS).  Binding an inhibitor to the proximal Qp site would lead to a buildup of FADH2 which could transfer electrons in single electron step to dioxygen to produce ROS.  This would occur if the dicarboxylic site were empty.

As there is no H+ transfer from the matrix to the inner membrane space for Complex II, we'll only focus on electron transfer from succinate to FAD and on to ubiquinone.

The transfer of electrons from succinate to ubiquinone hence takes place in two steps.

Step 1:  Succinate + FAD ↔ Fumarate + FADH2.  Figure $$\PageIndex{38}$$ below shows a plausible mechanism using a histidine in the active site as a general base.  This is a simple hydride (2e-) transfer reaction to FAD.  Note that no FeS cofactors are required.

Figure $$\PageIndex{8}$$: A mechanism for Succinate oxidation by FAD to  Fumarate and FADH2 (after Schaller, LibreText)

Step 2:  FADH2 + UQ ↔ FAD + UQH2.  This does not proceed by transfer of 2 e- from FADH2 directly to UQ.  Rather it occurs through a series of the FeS cofactors (Fe2S2, Fe4S4, and Fe3S4 clusters) and heme. First let's consider the path of electron transfer which will occur in single electron steps through the intermediary FeS clusters to ubiquinone as shown in Figure $$\PageIndex{39}$$ below. The standard reduction potential E0 is shown for each step.  Note that FAD is shown instead of the actual electron donor, FADH2, to correspond to use of the E0 values.

Figure $$\PageIndex{39}$$:  Pathway for electron transfer in Complex II

Electrons flow from FADH2 (lowest, most negative E0 value, -79 mV) to UQ (highest, most positive value E0 value, +113 mV) or from weakest to strongest oxidizing agent.  Electron transfer depends on distance as well.  It appears that transfer of electrons to heme b is not likely, given the presence of the closer UQ with the most positive E0 value.  Hence the function of heme b is unclear.  However if electrons are transferred to it, it would quickly and favorably send them on to UQ.  Note that the FeS cluster have progressively more positive E0 values except for Fe4S4 whose E0 = -260 mV.  Its E0 is depressed by the presence of charged/polar groups which would make the transfer of a negative electron less likely.  The net transfer standard potential is clearly favored thermodynamically.  As the E0 and corresponding G0 values are state functions, the final ΔG0 doesn't depend on path.  The high value of E0 for Fe4S4 presents an activation energy barrier for transfer in both directions and probably prevents reverse electron transport in this complex.

The membrane domains have two polar (Asn, Ser) and 5 polar-charged (Asp, Glu and Lys) side chains.

Now let's look at the terminal step in the transfer of electrons from Fe3S4 to ubiquinone.  Figure $$\PageIndex{40}$$ below shows a plausible mechanism.  Ubiquinone is shown in black.  TTFA, which binds to the ubiquinone site and inhibits electron transfer, is shown in brackets.  Tyrosine 91, Serine 41 and Histidine 216 are likely candidates for involvement in catalysis as shown below. Arginine 46 and aspartate 90 might be involved in protonation steps of UQH2.

Figure $$\PageIndex{40}$$:

Figure $$\PageIndex{41}$$ below shows an interactive iCn3D model of the iCn3D TTFA1 binding site (where ubiquinone binds).

Figure $$\PageIndex{41}$$:Mitochondrial Respiratory Complex II bound with 2-thenoyltrifluoroacetone (1zp0). (Copyright; author via source).
Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/icn3d/share.html?dY3niSfLVa9WiNNe7