Mechanism of quinol oxidation by the bc1-complex
Use the scroll bar to read an explanation, and for access to supplementary
The initial view is of the three subunits from a "functional" monomer.
Coordinates from the Zhang et al. native complex from chicken heart mitochondrial
complex have been supplemented by additional coordinates to provide the
working model (see below). Click here to view
the complete chicken complex using the coordinate data set submitted
to the Brookhaven Protein Data Bank. Click here for a tutorial
on the structure/function interface, based on the three catalytic subunits
of a "functional" monomer.
The structures used in this animation are a composite from the sets
of coordinates kindly made available by Dr. Ed Berry, of the bc1-complex
from chicken heart mitochondria, and by Dr. S. Iwata, of the beef heart
mitochondrial complex in P6522 and P65 crystals.
The coordinates from the solved structures were supplemented with models
for the quinone, quinol, and semiquinone substrates needed to show their
involvement in the catalytic cycle. The composite file includes coordinates
for the different components, in different configurations, involved in
turnover of the complex. Click here for a
summary of the files, and their construction.
The electron transfer reactions
The model is set up to operate under "conditions" in which both cyt c1
and the ISP are initially, reduced, and the Q-pool is about half reduced.
The pH is 8.0. A quinol occupies the Qo-site, and a quinone
occupies the Qi-site. The animation shows the turnover of the
complex, following delivery of an oxidizing equivalent to cyt c1.
This would be appropriate for activation of the Rb. sphaeroides photosynthetic
chain by a ~2 ms saturating flash (Click here
to see a brief summary
of the modified Q-cycle as it operates in Rb. sphaeroides).
Note: The stoichiometry of three electrons delivered to the high potential
chain, shown in the animation, would not be seen experimentally after flash
activation of the Rb. sphaeroides photosynthetic chain. The stoichiometry
of RC:cyt c2:bc1 is 2:1:1, so the number of oxidizing
equivalents provided by the reaction centers is only 2/bc1 complex,
and the ISP would remain reduced after the second turnover of the Qo-site.
Cyt c1 becomes oxidized, and is reduced by the ISP. Since the
pH is above the pK of His-161, the ISP looses a proton (represented by
a white sphere) on oxidation (see Note 1).
The ISP moves to the docking site on cyt b, and forms a reaction complex
with the quinol bound in the domain of the Qo-site distal from
heme bL. The reaction complex involves a H-bond between quinol
and His-161 of ISP, represented by the H as a white sphere (see
Note 2). The movement of the ISP is represented by successive occupation
of the four structures currently available, starting from that in the Iwata
et al. P6522 crystals, through the Zhang et al. native chicken
complex, and the Iwata et al. P65 crystals, to that in the Zhang
et al. chicken stigmatellin complex.
We have suggested that the pH dependence of the rate in the acidic range
reflects a requirement that His-161 must be in the dissociated form for
formation of the reaction complex with quinol, through formation of the
H-bond (see Note 3, and see links below for additional
information). The probability of formation
of the reaction complex is determined by the availability of the substrates,
- QH2 and the dissociated form of the ISPox. The
rate of quinol oxidation in the "substrate-limited" range is determined
by the probability of formation of the reaction complex; it increases as
[QH2] increases (see Note 4), and increases
with increasing pH (over the range below a pK of 7.5 on ISP-His-161), as
the concentration of the dissociated form increases. The binding constants for
formation of the ES-complex give rise to the 50-fold tighter binding of quinol
than quinone, and the shift in apparent pK from
the value of 7.6 measured through titration of the Em value for ISP,
to the pK of ~6.3 measured from
the pH for half-maximal rate at saturating quinol oxidation. In this animation,
the first proton from quinol is transfered to the ISP with the electron, and
is lost from His-161 on oxidation of the ISP, regenerating the dissociated
We have suggested that Glu-272, the second ligand to the occupant indicated
in the stigmatellin structure (2bcc), is also a second ligand to quinol.
In the native (1bcc) and myxothiazol or MOA-stilbene structures, Glu-272
is rotated by 120o out of its liganding position, to bring the
acidic side chain in contact with a water channel into cyt b from the P-side
aqueous phase. It seems likely that a similar movement provides a pathway
for release of the second proton from quinol. This would require that the
neutral semiquinone, generated at the site after the first electron (and
H+-) transfer to ISP, donates a H+ to Glu-272-
to leave the semiquinone anion as the donor to heme bH .
In the step with a high activation energy, the reaction complex dissociates
to products. With an activation barrier of 40-65 kJ/mol, dissociation to
products is improbable. Formation of semiquinone in this process is not
detected, suggesting a very small equilibrium constant (+ve DGo').
We had previously suggested that the positive DGo'
for formation of semiquinone might approach the value for the activation
barrier, so that both might reflect the same process (see
Note 5). If this were the case, DGo'
= DG# ~= DE#,
the equilibrium constant for formation of products would be very small,
and semiquinone would not normally be within the detection range of current
instrumentation. However, studies of mutant strains in which the Em
value for ISP was modified have shown that the change in rate with Em
value was less than expected from a simple interpretation along these lines.
A detailed study of the activation barriers of all aprtial reactions,
and their pH dependence, has suggested severe
constraints on possible mechanisms.
The model involves a single quinone at the Qo-site, rather than
two, as in double-occupancy models (see Note 6). We
have suggested that, during dissociation of the reaction complex, the semiquinone
moves from the distal domain of the Qo pocket to a domain (the
proximal lobe) close to heme bL. This movement could also contribute
to the activation barrier. By positioning the semiquinone close to heme
bL, the movement would allow a more rapid electron transfer
reaction. It would also remove the semiquinone from the ISP docking interface,
and from potential oxidation by either ISPox or O2.
Although the probability of forming semiquinone is low, the increased rate
of superoxide formation observed at the Qo-site in the presence
of antimycin, or under static head conditions (see Note
7), suggests that it may nevertheless be kinetically significant.
On dissociation of the reaction complex to products, the mobile extrinsic
domain (head) of reduced ISP is liberated from the complex, and rotates
to dock with cyt c1, close to the heme. The reaction complex
with cyt c1 involves a H-bond contact between His-161 of ISP
and a heme c1 propionate -C=O (see Note 8).
In this animation, the position of the ISP in the P6522 crystals
from Iwata's structure is used to show this contact. Electrons are transfered
from ISP to heme c1, and (see above) from semiquinone to heme
bL. Because both reactions occur after the rate limiting step,
they appear experimentally to be concerted. In this animation, the two
electron transfer reactions are shown separately, for clarity.
Electron transfer across the membrane through the b-heme chain reduces
quinone at the Qi-site to semiquinone (see
Quinone leaves the Qo-site, and is replaced by quinol, so that
the second turn-over of the Qo-site can occur (see
Steps 2-7 are repeated, and the second electron in the b-heme chain reduces
the semiquinone at the Qi-site to quinol, which vacates the
site, to restore the starting condition.
©Copyright 1996, Antony
Crofts, University of Illinois at Urbana-Champaign,