Pitfalls of Mitochondrial Redox Signaling Research

Abstract

Redox signaling from mitochondria (mt) to the cytosol and plasma membrane (PM) has been scarcely reported, such as in the case of hypoxic cell adaptation or (2-oxo-) 2-keto-isocaproate (KIC) β-like-oxidation stimulating insulin secretion in pancreatic β-cells. Mutual redox state influence between mitochondrial major compartments, the matrix and the intracristal space, and the cytosol is therefore derived theoretically in this article to predict possible conditions, when mt-to-cytosol and mt-to-PM signals may occur, as well as conditions in which the cytosolic redox signaling is not overwhelmed by the mitochondrial antioxidant capacity. Possible peroxiredoxin 3 participation in mt-to-cytosol redox signaling is discussed, as well as another specific case, whereby mitochondrial superoxide release is diminished, whereas the matrix MnSOD is activated. As a result, the enhanced conversion to H₂O₂ allows H₂O₂ diffusion into the cytosol, where it could be a predominant component of the H₂O₂ release. In both of these ways, mt-to-cytosol and mt-to-PM signals may be realized. Finally, the use of redox-sensitive probes is discussed, which disturb redox equilibria, and hence add a surplus redox-buffering to the compartment, where they are localized. Specifically, when attempts to quantify net H₂O₂ fluxes are to be made, this should be taken into account.

Keywords: H2O2 release into intracristal space; MnSOD; cytosolic H2O2 release; matrix H2O2 release; peroxiredoxins; redox buffers; redox signaling from mitochondria; redox-sensitive probes.

Figure 1

Redox signaling upon insulin secretion, stimulated either with glucose [18,19], ketoisocaproate (KIC) [18], or fatty acid [22]. Schemes show that the ATP-sensitive K⁺ channel (K_ATP) is to be closed only when both ATP and H₂O₂ (redox signaling) act in synergy [18], leading to a threshold depolarization (−50 mV) of the plasma membrane and concomitant opening of the voltage-dependent Ca²⁺ channels (Ca_L), allowing the Ca²⁺ entry and insulin granule vesicles’ exocytosis. With glucose, the pentose phosphate shuttle supplies NADPH [24] (besides so-called redox pyruvate transport shuttles, [19]) for the constitutively expressed NADPH-oxidase isoform 4 (NOX4), which provides cytosolic redox signaling [18]. With KIC, its oxidation provides both ATP and H₂O₂, which now originates from the mt-matrix-formed superoxide/H₂O₂ [18]. KIC is oxidized via a series of reactions resembling fatty acid β-oxidation (termed β-like oxidation) [13,14]. Finally, with fatty acid, even at low glucose, fatty acid β-oxidation also provides both ATP and H₂O₂. Again, H₂O₂ is of mt matrix origin and redox signaling from mitochondria to the plasma membrane (PM) has to occur. Simultaneously, H₂O₂ also activates mitochondrial phospholipase iPLA2γ (not shown, for simplicity), which adds a surplus of mt fatty acids for both β-oxidation and the metabotropic GPR40 receptor on PM [22]. GPR40 downstream pathways further stimulate insulin secretion.

Figure 2

Compartments of mitochondrion. Schemes shows examples of the following: (a) Mitochondrial network as imaged using 4Pi microscopy, i.e., 3D high-resolution fluorescence microscopy (images, such as published in Ref. [28]). (b) Section of a mitochondrial network tubule (such as indicated by the red line in (a)), extracted from 3D FIB/SEM images (see Ref. [33]). The outer mitochondrial membrane (OMM) is highlighted by a green color, similarly to the inner boundary membrane (IBM) and the intermembrane space peripheral (IMSp) between them. Yellow color-coding is used for cristae membranes, shown together with the contained but unresolved proteins. (c) Image of a single crista with already-resolved ATP-synthase dimers and respiratory chain (RC) supercomplexes, adopted from Ref. [34]. Diffusion distances for ubiquinol (QH₂) or ubiquinone (Q) are indicated by arrows, as well as the diffusion of protons between the ATP-synthase and RC supercomplexes. Matrix-faced dehydrogenases (DHs) allow QH₂ diffusion on the same leaflet, whereas the QH₂ diffusion between dehydrogenases facing the intracristal space (ICS) requires a flip/flop of QH₂/Q. These are namely glycerol-3-phosphate dehydrogenase (GAPDH) and dihydroorotate dehydrogenase (DHODH).

Figure 3

Peroxiredoxins of 2-Cys type exemplified by cycles of mt PRDX3. The catalytic cycle of the PRDX3 dodecamer emphasizes the deprotonation of the peroxidatic cysteine into S⁻ (red, protein part denoted as “p”), its oxidation into the sulfenyl state, and concomitant formation of S-S bonds with the resolving cysteines in the proximal subunit of the dimeric couple (dark blue, protein part denoted as “r”). After this, dodecamers occur in an unstable conformation, allowing splitting to dimers, which are reduced (regenerated to the original state) by the thioredoxin (TRX) plus TRXR (omitted). Upon intensive H₂O₂ flux, a hyperoxidation cycle starts by oxidation into sulfinyls (SO₂H), which can stack into the HMW filaments. Sulfinyls can be reduced at the expense of the ATP, being phosphorylated during the first step via a sulfinyl reductase reaction (sulfiredoxin, SRX), followed by TRX/TRXR or glutahione (GSH)/GRX. Redox signaling can exist, when a particular target protein with proximal cysteines interacts with just the peroxiredoxin in a state with disulfides. Protein reduces the peroxiredoxin dimer, whereas being oxidized into disulfides typically alters its function.

Figure 4

Considered fluxes and redox buffers for deriving the cytosolic redox signal (see Equation (6)).