Figure 1.
Scheme of the redox reactions in complex III of mitochondrial respiratory chain.
Explanation is given in the text.
Figure 2.
The prediction of bistability in complex III operation.
The levels of semiquinone (A) and free ubiquinol (B) were chosen as the indicators of state of system. If the system is in highly reduced state initially, it remains in this state (thick gray curves); if initially the levels of reduction are lower, the system evolves to another steady state characterized by lower levels of semiquinone and, respectively, ROS production. The same type of curves in (A) and (B) refers to the same simulation.
Figure 3.
Bi-stability of the reaction system catalyzed by complex III.
(A) shows the time course of transition of complex III from low to high ROS producing steady state. This transition was triggered in the model by a switch to higher succinate supply (increase in succinate concentration) and it is characterized by the increase of content of semiquinone radical bound at outer (Qo) site of the complex (SQ) and free ubiquinol (QH2 free), and decrease ubiquinone content (Q free). Electron transport rate (e_flux) initially increases, but then decreases because the deficit of electron acceptor (Q) impedes the oxidation of cytochrome bH. As a consequence, cytochrome bL remains reduced and cannot accept electrons from bound SQ thus increasing SQ content and related ROS production. (B) shows the computed content of semiquinone radicals bound at Qo site of complex III (SQ) and electron flow rates in steady states reached at various succinate supply from an initial state characterized by low SQ content (SQ_l). When the substrate supply overcomes a certain threshold, the system comes to a steady states characterized by the high levels of SQ and ROS production. If the system initially is in one of such high ROS producing states, it evolves to the different steady states after a decrease of succinate supply. Dashed line shows for comparison the continuum of steady states for SQ reached from an initial state characterized by high SQ content (SQ_h). (C) shows the computed content of semiquinone radicals bound at Qo site of complex III (SQ) and electron flow rates in steady states reached at various succinate supply from an initial state characterized by high SQ content (SQ_h). When the substrate supply decreases below a certain threshold, the system comes to a steady states characterized by the low levels of SQ and ROS production. If the system initially is in a one of such low ROS producing states, it evolves to the different steady states after an increase of succinate supply. Dashed line shows for comparison the continuum of steady states for SQ reached from an initial state characterized by low SQ content (SQ_l).
Figure 4.
Experimental observations of bistability in mitochondrial electron transport and their simulation in the model.
(A) shows the registration of ROS accumulation (traces “ros”) and membrane potential (traces “mp”) and in the suspension of isolated rat brain mitochondria incubated with 5 mM of succinate in the absence (blue traces) and in the presence of 1mM ADP (red traces). In the absence of ADP mitochondria are in high ROS producing state; the presence of ADP switches them to low ROS producing mode. Within the first 3 min initially present ADP is completely converted into ATP; the completion of ATP synthesis could be seen as an increase of membrane potential measured as the quenching of safranin O fluorescence (red trace “mp,ADP”). After the transformation of ADP into ATP and return to the state 4 of respiration, ROS production remained low (red trace “ros,ADP”). Under the conditions where ADP was absent (blue traces), 1 mM of ATP together with 1 uM of oligomycin was added and it did not decrease the rate of ROS production. The reason for addition of oligomycin together with ATP was to avoid effect of small amounts of ADP, which are usually present in preparations of ATP (blue traces). Thus, as the model predicts (Figures 3B and 3C), under the same conditions there exist two different states of ROS production. The complete set of experiments is presented in Text S1. (B) The experiments similar to that shown in (A) performed for various succinate concentrations revealed that two branches of steady state ROS production existed for the same experimental conditions in accordance with the model's prediction (Figures 3B and 3C). Upper branch (asterisks) corresponds to the measured initial rate of ROS production in state 4 (blue curve “ros” in Figure 4A), then 1 mM of ADP was added and, after 3 min of incubation with ADP, oligomycin was added to assure the transition back to state 4 of respiration. The temporal ATP synthesis switched the respiratory chain into low ROS production mode persisted after return to state 4 of respiration. The rate measured in this state (corresponded to the red trace “ros,ADP” in Figure 4A) constituted the lower branch (squares). Lines are the model simulations where the relative succinate concentration is transformed into mM using a scaling factor (the same for all points), chosen so that the threshold for stepwise shift to high ROS production in the upper branch in simulation corresponds to the one obtained experimentally. The succinate dehydrogenase is considered here as a Michaelis' function of not only Q, but also succinate (see eq (17) in “Methods”). Upper steady states were reached in the model when initially the system was in a high ROS producing state, lower steady states were reached when initially the system was in a low ROS producing state.
Figure 5.
Theoretical predictions and experimental observations of the bistability in mitochondrial electron transport triggered by anoxia.
(A) shows that decrease of oxygen concentration down to 0.1% O2 and below induces a switch to high SQ content if initially the system is in a state characterized by low SQ content. The dependence of SQ content at Qo site of complex III was calculated using the oxygen dependence of complex IV activity summarized in eq (18) of Model section. The increase of SQ content at Qo site proceeds in conjunction with decrease of electron flow limited by oxygen availability. Dashed line shows the continuum of steady states for SQ content reached if initially mitochondria are in a high ROS producing state. In this case the increase of oxygen availability does not restore low levels of SQ radicals. (B) shows that mitochondria, switched to a low ROS producing state by a temporal presence of ADP (as in (Fig 4A), could be switched back to high ROS production by a short application of anoxia (red trace), which is in accordance with the model prediction shown in Figure 4A. Anoxia was induced by substitution of atmospheric O2 for N2 and its effectiveness was estimated by the decrease of transmembrane potential (blue trace). After reoxygenation followed by restoration of membrane potential, ROS production is essentially higher than before anoxia.
Figure 6.
Theoretical predictions of ROS production under hypoxic conditions.
The rate of ROS production was calculated using eq. (16) with two different values of KROS indicated in the Figure. Concentrations of SQ necessary for the calculations were the same as in Figure 5A.