3.2 The acidification of the lumen by pmf alters the redox potentials of regulatory thiols, altering the mechanism of redox balancing
At physiological pH (less than about 8.0), the two-electron reduction of protein disulfide occurs with the uptake of 2 H+, as in
-S-S- + 2e- + 2H+ ⇌ (-SH)2
​​
The involvement of one proton per electron in the redox reaction implies that the redox midpoint potentials for the regulatory thiols will shift by -0.06 eV / pH unit, i.e., the lower the pH, the less reducing (more oxidizing) the regulatory thiols will become. The redox potentials of the stromal and lumenal thiol couples will change in opposite directions. Interestingly, reducing the stromal NADP+/NADPH couple, involves only one proton per two electrons, so its redox potential will only shift by about -30 mV/pH unit.
Figure 4 illustrates what would happen if the lumen and stromal pH values started at about 7.0 in the dark and changed to 6.0 and 8.0 in the light. The midpoint potential of thiol couples in the stroma should become more negative relative to dark (more difficult to reduce) and in some cases (e.g., NADPH-MDH) will become more negative than the NADP+/NADPH couple. In this case, reversal of the electron transfer (oxidizing the thiols and reducing NADP+) should result in net oxidation of the regulatory thiols, as discussed above, which can account for at least some of the redox modulation (Kramer et al. 1990).
By contrast, the acidification of the lumen should shift the lumenal thiol couples to less reducing (more oxidizing) redox potentials, making the transfer of electrons from stromal carriers (like NADPH or Trx) strongly favorable, and inhibiting oxidation by reversal processes. Extrapolating from published values, we estimate that VDE and HCF164 will have redox midpoint potentials of -202 and -164 mV at pH = 6.0, reducing in a strongly favorable transfer of electrons from stromal NADPH or Trx, effectively making the transfer of electrons irreversible and preventing stromal carriers from acting as oxidants for lumenal thiol couples.
In the stroma, light-induced electron transport typically results in a net reduction of thiol modulation enzymes that activate assimilation (e.g., FBPase) and inactivate catabolism (e.g., glucose-6-phosphate dehydrogenase). In contrast to stromal thiol-modulated enzymes, which are activated by becoming more reduced, lumenal thiol-modulated enzymes are often activated by becoming more oxidized (compare Figure 3A to 3C, see section 4). For example, VDE is inactivated by the artificial treatment of dithiothreitol, a strong reducing agent (Bilger and Björkman 1990). This occurs despite the fact that acidification of the lumen should make thiol reduction more favorable, strongly implying that the kinetics of oxidative processes, rather than thermodynamic constraints, control the overall redox poise.
The question then is, how do electrons get out of the lumen in a chloroplast system? The above arguments imply that alternative, lumen-associated components are absolutely required for redox balancing and that their redox potentials must be more oxidizing than the least negative thiol component, i.e. midpoint potential at pH = 6 higher than about -160 mV. Molecular oxygen was proposed as an oxidant in the lumen (Buchanan and Luan 2005; Gopalan et al. 2004) (Figure 3B). However, this is unlikely, as the oxygen concentration does not change significantly in the lumen and O2 is a relatively stable form. Obvious candidates include the plastoquinone (PQ) pool and some form of activated oxygen species, e.g. ROS species like H2O2, as seen in 2CP-mediated oxidation in the stroma.
In the Gram-positive bacterial system, thiol oxidation performed by DsbA (Fig. 2A) and a homologous proteins, 2CP, was proposed to oxidant in the stroma (Yoshida, Hara, Sugiura, Fukaya & Hisabori 2018; Yoshida, Yokochi & Hisabori 2019; Vaseghi et al. 2018; Ojeda, Pérez-Ruiz & Cejudo 2018; Yokochi, Fukushi, Wakabayashi, Yoshida & Hisabori 2021). However, 2CP is unlikely to have direct access to lumenal thiols.
LTO1 was proposed as a lumen-localized thiol oxidase by Karamoko et al (2011). LTO1 is an oxidoreductase that belongs to a distinct class of disulfide bond-forming enzymes in bacteria. It has two domains: a lumenal thioredoxin-like domain, which is functionally similar to DsbA, and functions to oxidize (forms disulfide bonds) proteins and a transmembrane domain, that has homology to the mammalian vitamin K epoxide reductase (VKOR) catalytic domain, which is also similar to DsbB (Karamokoet al. 2011; Onda 2013). LTO1 has been shown to oxidize lumenal proteins such as PsbO, STN7, and VDE (Kieselbach 2013; Lu et al. 2013, 2015; Yu et al. 2014; Wu et al. 2021) (see Section 4 for details). However, the final acceptor of LTO1 is still unknown (see Section 5).
Molecular oxygen was suggested as an oxidant in the lumen (Buchanan and Luan 2005; Gopalan et al. 2004) (Figure 3B). However, the oxygen concentration does not change significantly in the lumen and O2 is a relatively stable form.