A Comparative Study on the Effects of the Lysine Reagent Pyridoxal 5-Phosphate and Some Thiol Reagents in Opening the Tl+-Induced Mitochondrial Permeability Transition Pore

Lysine residues are essential in regulating enzymatic activity and the spatial structure maintenance of mitochondrial proteins and functional complexes. The most important parts of the mitochondrial permeability transition pore are F1F0 ATPase, the adenine nucleotide translocase (ANT), and the inorganic phosphate cotransporter. The ANT conformation play a significant role in the Tl+-induced MPTP opening in the inner membrane of calcium-loaded rat liver mitochondria. The present study tests the effects of a lysine reagent, pyridoxal 5-phosphate (PLP), and thiol reagents (phenylarsine oxide, tert-butylhydroperoxide, eosin-5-maleimide, and mersalyl) to induce the MPTP opening that was accompanied by increased swelling, membrane potential decline, and decreased respiration in 3 and 3UDNP (2,4-dinitrophenol uncoupled) states. This pore opening was more noticeable in increasing the concentration of PLP and thiol reagents. However, more significant concentrations of PLP were required to induce the above effects comparable to those of these thiol reagents. This study suggests that the Tl+-induced MPTP opening can be associated not only with the state of functionally active cysteines of the pore parts, but may be due to a change in the state of the corresponding lysines forming the pore structure.

It was previously shown that the MPTP opening becomes more probable if ANT is in the c-conformation, and, conversely, it visibly decreases when this exchanger is stabilized in the m-conformation [1,3,[10][11][12]. So, our results above [5][6][7][8][9] are because the former reagents (PAO, DIDS, high MSL, high NEM, tBHP, and Diam) stabilize the ANT c-conformation, and the latter ones (ADP, EMA, BKA, low NEM, and low MSL) favor this exchanger in the m-conformation [10]. It is well known that the ANT cysteines play an essential role in The swelling of succinate-energized mitochondria was slightly increased in the presence of 2-6 mM PLP ( Figure 1A). At once, calcium-loaded rat liver mitochondria swelled a little more with 2-6 mM PLP in comparison to the control experiments ( Figure 1A). ADP inhibited this swelling with minimal effect at 6 mM PLP ( Figure 1B). The MPTP inhibitors significantly prevented the swelling of calcium-loaded RLM in the presence of 6 mM PLP ( Figure 1C). This swelling decreased in the series CsA or PLP alone > control > ADP or NEM > CsA + NEM > ADP + CsA or ADP + NEM > "0" (free of Ca2+ and additions) ( Figure 1C). CsA is interested in has not inhibited this swelling even in similar experiments with 2 mM PLP (not shown here). The swelling of succinate-energized mitochondria in experiments with thiol reagents and PLP increased in the series EMA < "0" (free of additions) < PLP or NEM(1) < NEM(2) < high MSL < tBHP < PAO ( Figure 1D). A minor swelling increase ( Figure 1E) was found in similar experiments with calcium-loaded mitochondria in the presence of EMA, PLP, MSL, and tBHP. Herewith, the similar effect of PAO was maximal. experiments with 2 mM PLP (not shown here). The swelling of succinate-energized mitochondria in experiments with thiol reagents and PLP increased in the series EMA < "0" (free of additions) < PLP or NEM(1) < NEM(2) < high MSL < tBHP < PAO ( Figure 1D).

Effects of Tl + and Thiol-Modifying Agents on Respiration and ∆Ψ mito of Succinate-Energized Rat Liver Mitochondria
One can see in Figure 2A that PLP partly inhibited states 3 and 3U DNP respiration; however, the PLP effect on DNP-uncoupled respiration was not so pronounced. As for other thiol reagents, their inhibitory effect on RCR ( Figure 2B) and state 3 respiration ( Figure 2C) [6,9] increased in the series tBHP < EMA < PLP or MSL < PAO. DNP-uncoupled respiration [6,7,9] and RCR DNP ( Figure 2D) decreased similarly in experiments with these reagents. However, the inhibitory effect of PAO on RCR DNP ( Figure 2D) was more visible due to the PAO-induced state 4 increase [6]. The decrease in state 3U DNP respiration and RCR DNP in calcium-loaded RLM (control) was visibly greater in increasing PLP concentration in the medium containing TlNO 3 and 2-6 mM PLP (Figs. 3A and B). The mitochondrial response to DNP added into the medium was preserved in this case. Thiol reagents (PAO and tBHP), similar to PLP and unlike EMA or MSL, contributed to a further decrease in both 3U DNP respiration and RCR DNP in experiments with calcium-loaded mitochondria ( Figure 3A,B) [6,7,9]. The MPTP inhibitors (ADP with CsA or NEM alone) noticeably prevented a Ca 2+ -induced decrease in state 3U DNP respiration or RCR DNP in these experiments with/without PLP ( Figure 3A,C). The Ca 2+ -induced respiration decrease was markedly prevented by the MPTP inhibitors in the medium containing both PLP and tBHP or PAO ( Figure 3B,C) [6,7,9]. On the other hand, the above mitochondrial parameters were affected by EMA or MSL, similar to the MPTP inhibitors used. The additive effect of MSL with EMA was observed in this case. The fluorescent dye, safranin O, was used to evaluate the inner mitochondrial membrane potential (∆Ψ mito ). The addition of succinate into the medium led to the uptake of dye by succinate-energized mitochondria due to the inner membrane potential appearance. As a result, a decrease in the mitochondrial suspension fluorescence was observed. Calcium visibly decreased the membrane potential. An even greater decrease in the potential (Figure 4) was observed after adding calcium to the medium containing thiol reagents (PLP, EMA, MSL, tBHP, and PAO) [6,7,9]. However, this decrease was completely eliminated in the medium containing ADP and CsA.      (5). Besides, 500 μM ADP and 1 μM CsA (where indicated) were correspondingly added into the medium before mitochondria and Ca 2+ . Next, 75 μM Ca 2+ was injected into the medium after mitochondria. * shows significant differences from the control experiments with Ca 2+ and free of the reagents (p < 0.05).

Discussion
PLP was shown to penetrate slowly and passively into the mitochondrial matrix along a concentration gradient regardless of the presence of inhibitors and oxidative phosphorylation uncouplers [21,48]. The manganese transporter (Mtm1) transfers PLP into the matrix with micromolar affinity [49]. PLP and PAO were found to have produced increased swelling, decreased respiration in 3 and 3UDNP states, and ΔΨmito decline in the RLM experiments (Figures 1, 2, and 4) [6]. However, the above PLP effects occurred at a much lower rate since PLP concentrations of three orders of magnitude more than PAO concentrations are required to produce an impact of similar magnitude. Other thiol reagents (tBHP, Diam, DIDS, MSL, and high MSL) similarly affected swelling and respiration in 3 and 3UDNP states in experiments with calcium-free media (Figures 1, 2, and 4) [50]. The similarity of these effects may be because PLP can react not only with the ANT lysine residues but PLP can also interact with the ANT cysteine residues, similar to the reaction of PAO and above reagents with the cystein ones [10,11,16,24,28,29,50]. Low concentrations of PLP and thiol reagents were insufficient to inhibit visible RCR and state 3 respiration (Figure 2A-C) [6][7][8][9]50]. Wherein, state 3, not state 4, was markedly inhibited by these reagents at high concentrations. The state 3 inhibition increased in the series tBHP, Diam < PAO < EMA < PLP, FITC < high MSL, DIDS, high NEM (Figure 2,C) [6,50]. RCR showed similar series, but PAO-induced RCR decrease was maximal due to the state 3 decline and state 4 increase [6].
State 3 respiration is known to depend on the activities of F1F0-ATP synthase, the adenine nucleotide translocase (ANT), and the mitochondrial phosphate cotransporter (PiC), which are necessary for the ATP synthesis carried out by mitochondria. PLP and DIDS can inhibit F1F0-ATP synthase, gastric H + /K + -ATPase, and tonoplast ATPase [6,51-  (4), and 2 µM PAO (5). Besides, 500 µM ADP and 1 µM CsA (where indicated) were correspondingly added into the medium before mitochondria and Ca 2+ . Next, 75 µM Ca 2+ was injected into the medium after mitochondria. * shows significant differences from the control experiments with Ca 2+ and free of the reagents (p < 0.05).

Discussion
PLP was shown to penetrate slowly and passively into the mitochondrial matrix along a concentration gradient regardless of the presence of inhibitors and oxidative phosphorylation uncouplers [21,48]. The manganese transporter (Mtm1) transfers PLP into the matrix with micromolar affinity [49]. PLP and PAO were found to have produced increased swelling, decreased respiration in 3 and 3U DNP states, and ∆Ψ mito decline in the RLM experiments (Figures 1, 2 and 4) [6]. However, the above PLP effects occurred at a much lower rate since PLP concentrations of three orders of magnitude more than PAO concentrations are required to produce an impact of similar magnitude. Other thiol reagents (tBHP, Diam, DIDS, MSL, and high MSL) similarly affected swelling and respiration in 3 and 3U DNP states in experiments with calcium-free media (Figures 1, 2 and 4) [50]. The similarity of these effects may be because PLP can react not only with the ANT lysine residues but PLP can also interact with the ANT cysteine residues, similar to the reaction of PAO and above reagents with the cystein ones [10,11,16,24,28,29,50]. Low concentrations of PLP and thiol reagents were insufficient to inhibit visible RCR and state 3 respiration (Figure 2A-C) [6][7][8][9]50]. Wherein, state 3, not state 4, was markedly inhibited by these reagents at high concentrations. The state 3 inhibition increased in the series tBHP, Diam < PAO < EMA < PLP, FITC < high MSL, DIDS, high NEM ( Figure 2C) [6,50]. RCR showed similar series, but PAO-induced RCR decrease was maximal due to the state 3 decline and state 4 increase [6].
State 3 respiration is known to depend on the activities of F 1 F 0 -ATP synthase, the adenine nucleotide translocase (ANT), and the mitochondrial phosphate cotransporter (PiC), which are necessary for the ATP synthesis carried out by mitochondria. PLP and DIDS can inhibit F 1 F 0 -ATP synthase, gastric H + /K + -ATPase, and tonoplast ATPase [6,[51][52][53][54]. The ANT inhibition was observed in the presence of PLP [24,55], EMA [30], and FITC (discussed in [9]) as well as PAO, tBHP, and Diam (discussed in [6]). This inhibition resulted in the reaction of these reagents with the ANT cysteine residues. However, FITC does not interact with ANT cysteines, but inhibits ADP transport in mitochondria and reacts with the PiC cysteines [9,30]. The mitochondrial H + /Pi cotransporter activity was notably decreased in experiments with PLP [16,41], low MSL [7], EMA [56], and FITC [9]. State 3 respiration decrease may be due to inhibiting the enzymes involved in the oxidative phosphorylation processes and blocking the respiratory substrate transport into mitochondria. PLP and some thiol reagents (PAO, DIDS, FITC, and high MSL) inhibit a succinate transport into the matrix [6,7,9]. At the same time, state 3U DNP respiration decrease [6,7,9] was insignificant in experiments with mitochondria, energized substrates of the first respiratory complex (glutamate with malate) in the presence of reagents (DIDS, MSL, EMA, and FITC). Thus, the succinate transport inhibition may explain the marked decrease in state 3U DNP respiration and RCR DNP values (Figure 2A,D) in experiments with the latter reagents. By comparing the effects of other reagents (PLP, tBHP, Diam, and EMA) on states 3 and 3U DNP (Figure 2), it can be concluded that they have a weak impact on succinate transport. Some used reagents (PAO, MSL, DIDS, tBHP, and Diam) showed increased swelling in experiments with Ca 2+ -loaded mitochondria ( Figure 1) [9,50]. A similar effect of PLP was minor (Figure 1), whereas EMA and DIDS partially prevented this increase [6,9]. The Ca 2+ -induced decrease in state 3U DNP respiration was more visible in experiments with reagents (PAO, tBHP, Diam, FITC, and PLP) in comparison to DIDS, MSL, and EMA that slowed down these effects ( Figure 3) [9,50]. Similar results are associated with the involvement of ANT in the MPTP opening under the action of thiol reagents (PAO, MSL, and DIDS) and stress inducers (tBHP and Diam) on mitochondria due to the stabilization of this exchanger c-conformation because of their reaction with the ANT cysteines [6,50]. Opposite, DIDS and low MSL decreased the MPTP opening due to ANT conformational changes, altering the thiol group reactivity [6,7]. In addition, MSL or DIDS presence discovered that the state 3U DNP respiration decrease was not so pronounced in calcium-loaded mitochondria injected in the medium with TlNO 3 ( Figure 3B) [50]. These results suggest that the interaction of these reagents with the PiC, not ANT thiol groups, plays a primary role. EMA, compared to NEM, showed a more significant affinity to ANT cysteine thiol groups, but EMA weak penetrates across the inner mitochondrial membrane [9][10][11]14,16]. EMA inhibited beef heart mitochondria swelling due to this reagent interacting with the cytoplasm-faced PiC essential thiol groups [9,56]. If the reaction has a place with Cys 47 in experiments with low NEM and low EMA, then the m-conformation stabilization is observed [10,13,14]. The EMA block of the state 3U DNP respiration decrease ( Figure 3B) may be due to the EMA reaction with the ANT Cys 47 that followed this exchanger mconformation stabilization [10,14,15]. The PLP effects on calcium-loaded mitochondria are most likely associated with the interaction of PLP with the ANT cysteine residues [24].
The MPTP inhibitors (ADP, EMA, and low NEM) are known to fix ANT in the mconformation and to prevent the increased swelling and ∆Ψ mito decline in experiments with Ca 2+ -loaded mitochondria in the presence of thiol reagents (PAO, tBHP, Diam, EMA, arsenite, or menadione) that react with the ANT Cys 159 and Cys 256 [10,11,13,14]. The ADP inhibition of Tl + -induced MPTP opening was less pronounced in experiments with PLP, PAO, and DIDS than with MSL, EMA, and FITC (Figures 1, 3 and 4) [6,7,9,51]. Such a difference in effects may result from the fact that the former actively interacts with ANT cysteines, while the latter did not reveal such a critical effect on the ANT structure [6,7,9,11,24,56]. Thus, this circumstance ultimately allowed ADP to inhibit the MPTP opening. Low NEM (50 µM) prevented the increase in swelling and the decrease in RCR DNP and state 3U DNP respiration in experiments with the used thiol reagents (Figures 1C and 3C) [6][7][8][9]50]. The Tl + -induced MPTP inhibition was due to the NEM interaction with the above ANT cysteines in experiments with PAO, Diam, EMA, tBHP, and PLP [10,11,13,14]. However, FITCand MSL-induced MPTP opening in TlNO 3 media (Figures 1C and 3C) was inhibited by low NEM due to the reaction of NEM with cytoplasm-faced PiC thiols [7,9,50]. CsA visibly inhibited the Tl + -induced MPTP opening in experiments with tBHP, EMA, FITC, and MSL, but not PLP (Figures 1C and 3C) [6,7,9,50]. This result suggests that CyP-D is not involved in the MPTP opening in experiments with PLP. In this case, the increased swelling decreased respiration, and ∆Ψmito decline in experiments with used thiol and lysine (PLP) reagents were noticeably leveled or completely disappeared in simultaneous presencing NEM with ADP or NEM with CsA in the experiments with the TlNO 3 medium. The inhibitory effect of ADP with CsA was maximal ( Figures 1C, 3C and 4) [6,7,9,50].

Animals
Male Wistar rats (250-300 g) were kept at 20-23 • C under a 12 h light/dark cycle with free access to water ad libitum and the standard rat diet. All treatment procedures of animals were performed according to the Animal Welfare act and the Institute Guide for Care and Use of Laboratory Animals (Protocol # 6/5/2022).

Mitochondrial Isolation
Rat liver mitochondria were isolated according to [9] in a buffer containing 250 mM sucrose, 3 mM Tris-HCl (pH 7.3), and 0.5 mM EGTA; subsequent mitochondrial sediment was washed out twice by resuspension-centrifugation in a medium containing 250 mM sucrose and 3 mM Tris-HCl (pH 7.3) and finally suspended in 1 mL of the last buffer. According to Bradford, the mitochondrial protein content assayed was within 50-60 mg/mL.

Swelling of Mitochondria
Mitochondrial swelling (Figure 1) was tested as a decrease in A 540 at 20 • C using an SF-46 spectrophotometer (LOMO, St. Petersburg, Russia). Mitochondria (1.5 mg of protein/mL) were injected into a 1-cm cuvette with 1.5 mL of 400 mOsm medium containing 75 mM TlNO 3 , 125 mM KNO 3 , 5 mM Tris-NO 3 (pH 7.3), 2 µM rotenone, and 1 µg/mL of oligomycin. PLP, PAO, tBHP, NEM, Ca 2+ , succinate, NEM, ADP, and CsA were added into the medium before or after mitochondria (see Figure 1 legend). The swelling, oxygen consumption rates, and ∆Ψ mito dissipation were investigated in 400 mOsm media to verify the comparability and consistency between data in different experiments.

Mitochondrial Membrane Potential
The inner membrane potential (∆Ψ mito ) induced by injection of 5 mM succinate into a medium was tested according to Waldmeier et al. [57]. The safranin O fluorescence intensity (arbitrary units) in the mitochondrial suspension was tested at 20 • C (Figure 4) using the microplate reader (CLARIOstar ® Plus, BMG LABTECH, Ortenberg, Germany) at 485/590 nm wavelength (excitation/emission). The mitochondria (0.5 mg of protein/mL) were added into cells containing 300 µL of the medium with 20 mM TlNO 3 , 125 mM KNO 3 , 110 mM sucrose, 5 mM Tris-NO 3 (pH 7.3), 1 mM Tris-P i , 2 µM rotenone, 3 µM safranin O, and 1 µg/mL of oligomycin. ADP, CsA, PLP, EMA, PAO, tBHP, and MSL were injected into the medium before mitochondria (see Figure 4 legend). 5 mM succinate, 75 µM Ca 2+ , and 30 µM DNP was administrated into the medium after the mitochondria. The safranin O fluorescence change after the succinate injection was taken as 100% in control experiments free of the above additions (PLP, thiol reagents, ADP, CsA, and Ca 2+ ). The other cases' fluorescence values were calculated relative to this control. A parallel fourfold measurement for each 300 µl aliquot was made from three independent preparations.

Statistical Analysis
The statistical differences in results and corresponding p-values were evaluated using two population t-tests (Microcal Origin, Version 6.0, Microcal Software). These differences are presented as a percent of the average (p < 0.05) from one of three independent experiments (Figures 1-4). A more detailed statistical analysis is in supplementary materialssupplementary materials, Figures S1-S4 and Table S1.

Conclusions
Mitochondrial proteins contain up to 20% acetylated lysine residues, including F 1 F 0 -ATP synthase structural parts [19]. The covalent modification of lysine residues is believed to affect the MPTP open probability [13,20]. The MPTP opening resulted in a fluorescamine reaction with the ANT lysine residues that induced efflux of accumulated Ca 2+ , ∆Ψ mito decline, and mitochondrial swelling [58]. Experiments with BKA-and CAT-treated mitochondria showed that one or more lysine residues could be involved in CAT and BKA binding with the ANT [24,55]. Lys 401 of bovine heart mitochondrial F1F0-ATP synthase was modified explicitly with 4-chloro-7-nitrobenzofurazan [59]. Pyridoxal 5'-diphospho-5'adenosine was shown to bind to the isolated alpha-subunit from E. coli F 1 -ATP synthase [60]. FITC reacts with cysteine and the PiC lysine residues. It quickly penetrates the mitochondrial matrix, binds to α and γ subunits of the F1-ATP synthase, and localizes in the inner membrane lipid part [9,30]. FITC binds the PiC Lys residue more efficiently than PLP [30]. It can be seen that lysine residues are essential for the functioning of the MPTPs most important parts, namely F1F0-ATP synthase, ANT, and Pi cotransporter. An analysis of the study results allows us to conclude that the Tl + -induced MPTP opening is associated not only with functionally active cysteines of the latter three parts, but may also be related to activities of the corresponding lysines that form the pore structure parts.

Data Availability Statement:
The data that support the findings of this study are available from the corresponding authors, [S.M.K.], upon reasonable request.

Acknowledgments:
The authors are grateful to Irina V. Brailovskaya for help in isolating mitochondria and the oxygen consumption rates in rat liver mitochondrial suspensions.

Conflicts of Interest:
The authors declare no conflict of interest.