The capacity of anaerobically grown bacteria to exchange the 2H+ of the cell for the K+ of the medium and to maintain a high K+ distribution between cell and medium

The N,N'-dicyclohexylcarbodiimide sensitive exchange of 2H+ of a cell for K+ of medium stable to pH, K+ activity and temperature changes has been discovered in anaerobically grown gram-negative Escherichia coli, Salmonella typhimurium. S. enteritidis, Proteus mirabilis, P. vulgaris, anaerobic gram-positive bacteria Streptococcus faecalis, Lactobacillus salivarius, L. lactis in the presence of exogenic energy source. This exchange in gram-negative bacteria is operating only at increase of medium osmolarity. The high K+ distribution between cell and medium has been reached during the exchange of 2H+ for one K+ and the corresponding potassium equilibrium potential is much more than the measured delta psi. In aerobically grown E. coli, S. typhimurium, Brevibacterium flavum and aerobic Micrococcus luteus exchange of 2H+ for K+ does not take place, the K+ distribution is lower and in good conformity with the measured delta psi. It is assumed that exchange of 2H+ for K+ in anaerobic bacteria is carried out by the H+-ATPase complex and the Trk (or Trk-like) system of K+ absorption united into the same membrane supercomplex which functions as the H+-K+-pump and supports the high K+ distribution between cell and medium.     

Modulation of the mitochondrial cyclosporin A-sensitive permeability transition pore by the proton electrochemical gradient. Evidence that the pore can be opened by membrane depolarization.

This paper reports an investigation on the relationship between the proton electrochemical gradient (delta mu H+) and the cyclosporin A-sensitive permeability transition pore (PTP) in rat liver mitochondria. Using the SH group cross-linker phenylarsine oxide as the inducer, we show that both matrix pH and the membrane potential can modulate the process of PTP induction independently of Ca2+. We find that membrane depolarization induces the PTP per se when pHi is above 7.0, while at acidic matrix pH values PTP induction is effectively prevented. Since Ca2+ uptake leads to major modifications of the delta mu H+ (i.e. matrix alkalinization and membrane depolarization), we have explored the possibility that the Ca(2+)-induced changes of the delta mu H+ may contribute to PTP induction by Ca2+. Our data in mitochondria treated with Ca2+ plus N-ethylmaleimide and Ca2+ plus phosphate show that membrane depolarization is a powerful inducer of the PTP. Taken together, our observations indicate that the PTP can be controlled directly by the delta mu H+ both in the absence and presence of Ca2+, and suggest that a collapse of the membrane potential may be the cause rather than the consequence of PTP induction under many experimental conditions. Thus, many inducers may converge on dissipation of the membrane potential component of the delta mu H+ by a variety of mechanisms.     

Effect of the membrane potential on the Mg2+,ATP-dependent transport of Ca2+ across smooth muscle sarcolemma

The effect of the membrane potential (K(+)-valinomycin system) on the Mg2+, ATP-dependent transport of Ca2+ in inside-out vesicles of myometrium sarcolemma has been studied. The membrane potential was identified by using a cyanine potential-sensitive probe, diS-C3-(5). In the presence of valinomycin (5.10(-8) M) the inside-out directed K+ gradient (delta psi = -86 mV, with a negative charge inside) stimulated the initial rate of the energy-dependent accumulation of Ca2+ transfer whereas the oppositely directed K+ gradient (delta psi = +72 mV, with a positive charge inside) had no effect on this process. The K+ gradient was formed by isotonic substitution of K+ in intra- or extravesicular space for choline +. At the same time, in the absence of K+ gradient the Mg2+, ATP-dependent accumulation of Ca2+ in membrane vesicles did not depend on the chemical nature of the cations (K+ or choline+) used for isotonicity. The decrease of delta psi from 0 to -86 mV affects the initial rate of Ca2+ accumulation but not the maximal content of the accumulated cation. Preliminary dissipation of the membrane potential (delta psi = -86 mV) in Mg2(+)-free isotonic (with respect of K+ and choline+) media containing ATP and Ca2+ resulted in the inhibition of Mg2+, ATP-dependent Ca2+ transport induced by subsequent addition of Mg2+. These results indicate that the negative (intravesicular) electrical potential activates the Ca-pump of smooth muscle sarcolemma. This activation is based on the increase in the turnover number of the Ca2+ transporting system but not on its affinity for the transfer substrate. The use of the absolute reaction rates theory made it possible to establish that the Ca-pump effectuates the transport of a single positive charge in inside-out vesicles of smooth muscle plasma membranes, i.e., the energy-dependent transport of Ca2+ occurs either as a symport (with an anion (Cl-) or an antiport with a monovalent cation (K+) or a proton. It is assumed that the potential dependence of the Ca-pump in the smooth muscle plasma membrane plays a role in the realization of effects of mediators and physiologically active substances that are manifested as stimulation of the contractile response and depolarization of the sarcolemma. In is quite probable that the delta psi-dependent Ca-pump is also responsible for the maintenance of intracellular homeostasis of monovalent cations (K+, H+, Cl-) in smooth muscle tissues.     

The nature of H+-K+-exchange in anaerobically and aerobically grown S. typhimurium

H+-K+-exchange via the Trk-like system of K+ accumulation takes place in anaerobically grown S. typhimurium LT-2 with stable ratio of DCC-sensitive ionic fluxes, equal to 2H+ of a cell for one K+ of the medium. This exchange is now observed in the mutant S. typhimurium TH-31 with unfunctional H+-ATPase. H+-K+-exchange in aerobically grown S. typhimurium LT-2 has unstable ratio of ionic fluxes. The rate of K+ uptake in anaerobically grown bacteria is higher than that in the aerobically grown ones. Q10 is about 1.8 both for H+ transfer and K+ uptake in anaerobically grown bacteria, but it is 1.7 and 0.9 respectively in the aerobically grown ones. Delta psi is not changed by different temperatures both in anaerobically and aerobically grown bacteria. The distribution of K+ in anaerobically grown bacteria is higher than 10(3) and the potassium equilibrium potential is much higher than the measured delta psi. In aerobically grown bacteria the distribution of K+ is in good conformity with the measured delta psi. H+ and K+ transport in anaerobically grown cells is likely to proceed by the same mechanism, which includes H+-ATPase and the Trk-like system. In aerobically grown bacteria these transport systems work separately, and the Trk-like system as K+-ionophore serving for K+ uptake across the electrical field on the membrane.     

Changes in host cell energetics in response to bacteriophage PRD1 DNA entry.

Double-stranded DNA bacteriophage PRD1 infects a variety of gram-negative bacteria harboring an IncP-type conjugative plasmid. The plasmid codes for the DNA transfer phage receptor complex in the cell envelope. Our goal was, by using a collection of mutant phage particles for which the variables are the DNA content and/or the presence of the receptor-binding protein, to obtain information on the energy requirements for DNA entry as well as on alterations in the cellular energetics taking place during the first stages of infection. We studied the fluxes of tetraphenylphosphonium (TPP+), phenyldicarbaundecaborane (PCB-), and K+ ions as well as ATP through the envelope of Salmonella typhimurium cells. The final level of the membrane voltage (delta psi) indicator TPP+ accumulated by the infected cells exceeds the initial level before the infection. Besides the effects on TPP+ accumulation, PRD1 induces the leakage of ATP and K+ from the cytosol. All these events were induced only by DNA-containing infectious particles and were cellular ATP and delta psi dependent. PRD1-caused changes in delta psi and in PCB- binding differ considerably from those observed in other bacteriophage infections studied. These results are in accordance with the presence of a specific channel engaged in phage PRD1 DNA transport.     

Thrombin-induced membrane depolarization of platelets and its inhibition by cetiedil

When 50 microM cetiedil alone was added to a platelet suspension, increase in Na+ content, decrease in K+ content, and depolarization of platelet membrane were observed without change in the intracellular concentration of free Ca2+ ([Ca2+]1) or in the morphology of platelets. The cetiedil-induced depolarization was attenuated by the reduction of extracellular sodium concentration, while sodium transport inhibitors such as procaine and tetrodotoxin failed to modify the depolarization. On the other hand, thrombin caused such changes in platelets as increases in Na+ content, 22Na space and [Ca2+]1, decrease in K+ content, and membrane depolarization. All these changes caused by thrombin were inhibited by cetiedil. It is suggested that cetiedil brought the increased ion transport and subsequent partial depolarization, which might lead to modification of the reaction of platelet membrane induced by thrombin.     

Influence of membrane potential changes on cytoplasmic Ca2+ concentration in an electrically excitable cell, the insulin-secreting pancreatic B-cell.

Glucose stimulation of insulin release involves metabolism of the sugar and elevation of cytoplasmic calcium (Ca2+i) in pancreatic B-cells. We compared the dynamic changes of metabolism (fluorescence of endogenous reduced pyridine nucleotides, NAD(P)H), membrane potential (intracellular microelectrodes), and Ca2+i (fura-2 technique), in intact mouse islets. Glucose (15 mM) sequentially triggered an increase in NAD(P)H fluorescence, a depolarization with electrical activity, and a rise in Ca2+i. The change in NAD(P)H was monophasic and regular, whereas the changes in membrane potential and Ca2+i were multiphasic, with steady-state regular oscillations of similar average frequencies (about 2.2/min). Digital image analysis revealed that Ca2+i oscillations were synchronous in all regions of the islets. Omission of extracellular Ca2+ abolished the rise in Ca2+i but not the increase in NAD(P)H. Both electrical and Ca2+i oscillations disappeared in low external Ca2+ (1 mM), and became larger but slower in high Ca2+ (10 mM). Sustained depolarization (by tolbutamide, arginine, or high K+) and hyperpolarization (by diazoxide) of B-cells caused sustained increases and decreases of Ca2+i, respectively. In conclusion, the changes in membrane potential induced by various secretagogues trigger synchronous changes in Ca2+i in all B-cells of the islets. The oscillatory pattern of the electrical and Ca2+i responses induced by glucose is not accompanied by and thus probably not due to similar oscillations of metabolism.     

Cation fluxes cause plasma membrane depolarization involved in beta-glucan elicitor-signaling in soybean roots

Inducible and specific ion fluxes on plasma membranes represent very early events during elicitation of plant cells. The hierarchy of such ion fluxes involved is still unknown. The effect of Phytophthora sojae-derived beta-glucan elicitors on the plasma membrane potential as well as on surface K+, Ca2+, and H+ fluxes has been investigated on soybean roots using ion-selective microelectrodes. Beta-Glucans with different degrees of polymerization transiently depolarized the plasma membrane. The elicitor concentration necessary for half-maximal depolarization closely resembled the corresponding binding affinities of soybean root membranes toward the respective beta-glucans. Upon repeated elicitor treatment, the root cells responded partially refractory, suggesting a complex responsiveness of the system. Within the root hair space, characteristic decreasing K(+)- and Ca(2+)-free concentrations were induced by the elicitors, probably causing depolarization through the influx of positive charges. Whereas K+ fluxes were inverted after passing the K+ equilibrium (Nernst-) potential, Ca2+ influx continued. No anion fluxes sufficient to account for charge compensation were observed under the same experimental conditions. K+ and Ca2+ fluxes as well as depolarization were inhibited by 100 microM or less of the Ca2+ antagonist La3+. Contrasting other systems, in soybean the main cause for elicitor-induced plasma membrane depolarization is the activation of cation instead of anion fluxes.     

Conditions for S. typhimurium transfection by isolated DNA from bacteriophage P22 H5]

Transfection of spontaneous S. typhimurium LT2 WTR mutant, sensitive to bacteriophages FO, Ffm, 6SR, and resistant to bacteriophages P22 H5, C-21 and P1 vir, by DNA of P22 H5 bacteriophage in the presence of Ca2+ ions was demonstrated. The following were conditions of infection providing maximal and reproducible results: concentration of the recipient cells of 6--7x 109 cells/ml, DNA concentration of 30 microng/ml, CaC12 concentration of 0,25 M. The efficacy of transfection was 5 x 10-8.     

The role of Ca2+ ions in a phage infection of Corynebacterium glutamicum

It was shown that neither uncouplers of oxidative phosphorylation, nor lack of Ca2+ ions affected the normal MC-2 phage absorption on Corynebacterium glutamicum cells, while the phage development was repressed under these conditions. Simultaneous measurement of Ca2+, K+ and H+ ion flows, as well as measurement of membrane potential showed that the addition of the phage into the experimental medium led to significant depolarization of the membrane from -160 mV to -100 mV due to the penetration of Ca2+ ions into the cells followed by K+ and H+ efflux. The (Ca2+) to (K+ + H+) ratio was shown to be 1 : 1. Phage DNA is supposed to be injected into the host cells as a positively charged (Ca2+-DNA) complex.     

Mutants defective in the 33K outer membrane protein of Salmonella typhimurium.

Salmonella typhimurium LT2 lines, if phenotypically rough, are fully sensitive to bacteriocin 4-59, produced by Salmonella canastel strain SL1712. Bacteriocin-resistant mutants fell into three classes. Those resistant to phage ES18 and to albomycin proved to be mutants of class chr (equivalent to tonB of Escherichia coli); these mutants still adsorb the bacteriocin and so are classified as tolerant. Another class of (incompletely) tolerant mutants was resistant to phage PH51; their envelope fractions lacked the band corresponding to outer membrane protein 34K, known to serve for adsorption of phage PH51. A third class of mutants, which did not adsorb the bacteriocin, was unaltered in sensitivity to phages. Their envelopes lacked the 33K band, indicating absence of the outer membrane protein 33K, considered to correspond to outer membrane protein II* of E. coli, which in that species is determined at locus ompA (formerly tolG or con). Phage P22 HT105/1 cotransduced the 33K S. typhimurium gene (to be called ompA, to accord with E. coli usage) with pyrD+ at about 30% frequency when the donor allele was ompA+ or one ompA, but at only 3 to 11% when the donor allele was another ompA. When the donor carried either of two long deletions of the put (proline utilization) operon, phage P22 HT105/1 cotransduced put (and ompA+) with pyrD+ at low frequency. The cotransduction data indicate that ompA of S. typhimurium is located between pyrD and put, nearer the former. This corresponds to the map position of ompA in E. coli K-12.     

Commitment to differentiation of murine erythroleukemia cells involves a modulated plasma membrane depolarization through Ca2+-activated K+ channels

The role of the plasma membrane potential (delta psi p) in the commitment to differentiation of murine erythroleukemia (MEL) cells has been studied by analyzing the ionic basis and the time course of this potential in the absence or the presence of different types of inducers. delta psi p was determined by measuring the distribution of tetraphenylphosphonium (TPP+) across the plasma membrane and displayed a 22-hour depolarization phase (from -28 to +5 mV) triggered by factors contained in foetal calf serum (FCS) and followed by a nearly symmetrical repolarization phase. After measuring the electrochemical equilibrium potential of Na+, K+, and Cl-, the relative contribution of these ions to delta psi p was evaluated by means of ion substitution experiments and by the addition of ion flux inhibitors (tetrodotoxin [TTX], 4-acetoamide-4'-isothiocyanostilbene-2,2'-disulfonate [SITS]) and ionophores (Valinomycin, A23187). The Na+ contribution to delta psi p appeared negligible, the potential being essentially generated by K+ and Cl- fluxes. When evaluated by a new mathematical approach, the effects of Valinomycin and A23187 at different times of incubation provided evidence that both the depolarization and the repolarization phase were due to variations of the K+ permeability across the plasma membrane (PK) mediated by Ca2+-activated K+ channels. All the inducers tested (dimethylsulfoxide [DMSO], hexamethylen-bis-acetamide [HMBA], diazepam), although they did not modify the ionic basis of delta psi p, strongly attenuated the depolarization rate of this potential. This attenuation was not brought about when the inducers were added to noninducible MEL cell clonal sublines. Cell commitment occurred only during the depolarization phase and increased proportionally to the attenuation of this phase up to a threshold beyond which the further increase of the attenuation was associated with the inhibition of commitment. The major role of the inducers apparently consisted of the stabilization of the Ca2+-activated K+ channels, suggesting that a properly modulated delta psi p depolarization through these channels is primarily involved in the signal generation for MEL cell commitment to differentiation.     

Plasma membrane depolarization induced by abscisic acid in Arabidopsis suspension cells involves reduction of proton pumping in addition to anion channel activation, which are both Ca2+ dependent

In Arabidopsis suspension cells a rapid plasma membrane depolarization is triggered by abscisic acid (ABA). Activation of anion channels was shown to be a component leading to this ABA-induced plasma membrane depolarization. Using experiments employing combined voltage clamping, continuous measurement of extracellular pH, we examined whether plasma membrane H(+)-ATPases could also be involved in the depolarization. We found that ABA causes simultaneously cell depolarization and medium alkalinization, the second effect being abolished when ABA is added in the presence of H+ pump inhibitors. Inhibition of the proton pump by ABA is thus a second component leading to the plasma membrane depolarization. The ABA-induced depolarization is therefore the result of two different processes: activation of anion channels and inhibition of H(+)-ATPases. These two processes are independent because impairing one did not suppress the depolarization. Both processes are however dependent on the [Ca2+]cyt increase induced by ABA since increase in [Ca(2+)](cyt) enhanced anion channels and impaired H(+)-ATPases.     

Investigation of Ca2+ -induced changes of membrane potential of smooth muscle mitochondria using flow cytometric analysis

Using flow cytometric analysis and potential-sensitive fluorescent dye TMRM Ca2+ -induced changes of membrane potential of isolated smooth muscle mitochondria were studied. It was shown, that Ca2+ (100 microM) addition to the incubation medium induced mitochondrial membrane depolarization that probably could be explained by Ca2+/H+ -exchanger activation which functioning lead to membrane potential dissipation. In the case of ruthenium red (10 microM) preliminary presence in incubation medium, Ca2+ (100 microM) addition did not lead to membrane potential dissipation. Hence, membrane potential dissipation was caused by an increase of matrix Ca2+ concentration. In the presence of Mg2+ (3 mM) and ATP (3 mM), Ca2+ addition did not cause depolarization. It was supposed that in this case ATP synthase acted in the opposite direction as H+ -pump and prevented from mitochondrial membrane potential dissipation. Thus, the flow cytometry method allows to register membrane potential of isolated smooth muscle mitochondria and also to test the effectors, capable to modulate this parameter.     

Ca2+/H+ exchange through plasma membrane in the system of transport of calcium and hydrogen ions

The survey is aimed to review the data from literature, concerning possible mechanisms of Ca2+ and H+ transport through the plasma membrane of a cells, and also possibility of existence of Ca2+/H(+)-exchange in the plasma membrane of the muscle cells. It is known that the modification of pHl (delta pH) also can influence the work of the contractile system of muscle cells, and the transition of Ca2+ through the plasma membrane of the cells. Thus, one can suppose a direct relation between Ca2+ and H+ transport, through Ca2+/H+ exchange, and indirect relation through connection with other systems of transport of both Ca2+ (Ca(2+)-ATPase, Na+/Ca2+ exchange), and H+ (Na+/H(+)-exchange, H(+)-ATPase). For example it is shown, that the activator (inhibitor) of the Na+/H(+)-exchange through the plasma membrane of muscle cells, influence the work of the retractive system. And as is known, Ca2+ takes main part in involvement in the system excitation--contraction, and, thus, influencing the work of the Na+/H(+)-exchange, it is possible to regulate transport of Ca2+ through the plasma membrane of a muscle cell. The problem about a possibility of existence of Ca2+/H+ exchange, or functioning of Ca2+/H(+)-exchanger, is still far from the solution. Therefore, in the given review the attempt is made to analyze available information about possible connection between Ca2+ and H+ transport through the plasma cell membrane.     

Interconversion of components of the bacterial proton motive force by electrogenic potassium transport.

The influence of K+ ions on the components of the transmembrane proton motive force (delta mu H+) in intact bacteria was investigated. In K+-depleted cells of the glycolytic bacterium STreptococcus faecalis the addition of K+ ions caused a depolarization of the membrane by about 60 mV. However, since the depolarization was compensated for by an increase in the transmembrane pH gradient (delta pH), the total proton motive force remained almost constant at about 120 mV. Half-maximal changes in the potential were observed at K+ concentrations at which the cells accumulated K+ ions extensively. In EDTA-treated, K+-depleted cells of Escherichia coli K-12, the addition of K+ ions to the medium caused similar, although smaller changes in the components of delta mu H+. Experiments with various E. coli K-12 K+ transport mutants showed that for the observed potential changes the cells required either a functional TrkA or Kdp K+ transport system. These data are interpreted to mean that the inward movement of K+ ions via each of these bacterial transport systems is electrogenic. Consequently, it leads to a depolarization of the membrane, which in its turn allows the cell to pump more protons into the medium.     

Proton transport is necessary for divalent metal cations release from deenergized mitochondria

The release of divalent cations (Ca2+ and Sr2+) from rat liver mitochondria after membrane depolarization with protonophore (carbonyl cyanide m-chlorophenyl hydrazone, CCCP), sodium azide and K(+)-ionophore (valinomycin) was studied. It is stated that membrane depolarization itself is not sufficient for cations release from mitochondrial matrix (provided that mitochondrial permeability transition pore is blocked by cyclosporin A). Complete delivering of divalent cations is observed only after protonophore (CCCP) addition to suspension of deenergized mitochondria. The data show that membrane permeabilisation to hydrogen ions (H+) is necessary for complete cation release from the mitochondrial matrix. The enhancement in K(+)-conductivity of mitochondrial membrane (by valinomycin), on the contrary, is not able to provide complete delivering of cations from mitochondria. It is shown that quantity of divalent metal cation released from mitochondria (depolarized and permeabilized for K+ as well) is proportional to the concentration of protonophore (but not K(+)-ionophore) introduced in the incubation medium. The data obtained lead to the conclusion that H(+)-permeabilization of the mitochondrial membrane is necessary for the complete release of Ca2+ and Sr2+ from mitochondria after membrane depolarization. The possible mechanism of divalent metal cations release from deenergized mitochondria is discussed.     

ATP activates a Ca2+-dependent Cl− current in the rat thyroid cell line,FRTL-5

The effect of external ATP on both the membrane potential and the transmembrane current of the thyroid cell line FRTL-5 has been investigated in the patch-clamp whole-cell recording configuration. In the resting situation the membrane potential is around -70 mV and the membrane acts like a K(+)-sensitive electrode. Application of ATP at concentrations higher than 1 microM elicited an increase in Cl- conductance, responsible for a membrane depolarization which could be blocked by preincubation with the P2-antagonist quinidine. Chelation of intracellular Ca2+ also blocked the ATP induced changes in membrane potential and Cl- current. Intracellular perfusion with inositol trisphosphate (IP3) (50 microM) also stimulated a Cl- current which mimicked the response induced by ATP. ATP is able to initiate a response in the absence of extracellular Ca2+, but also opens a Ca(2+)-influx pathway, as demonstrated by a secondary response upon Ca2+ readmission in the external medium, in the continued presence of ATP. ADP and ATP gamma S were able to mimic the ATP response, whereas AMP and adenosine were unable to elicit a Cl- current. The P2X receptor agonist alpha,beta-methyleneATP was without effect as was the P2Y receptor agonist 2-methylthio ATP. We conclude that ATP is able to elicit a large IP3-mediated Ca(2+)-dependent Cl- current and membrane depolarization via a novel P2-type purinergic receptor.     

Passive Ca2+ transport in a suspension of isolated smooth-muscle cells and the contribution of the basal calcium permeability of the plasma membrane to the activation of a tonic contraction

Ca2+ concentration has been estimated in isolated myometrium cells using Ca2(+)-sensitive quin-2 fluorescent probe. Two components of Ca permeability of the plasmatic membrane have been determined, a potential-independent one (activated by K+ depolarization and nitrendipine-sensitive), and a basal one (not sensitive to nitrendipine). Smooth muscle cells could maintain intracellular Ca2+ concentration at the physiological level. In the presence of nitrendipine, orthovanadate, an inhibitor of sarcolemma Ca pump, induced the increase in the basal tonus depending on the presence of the Ca2+ in the medium. This suggests that in conditions of the blockage of electrically controlled Ca channels and Ca pump of the plasmatic membrane, the noncompensated basal Ca2+ influx activates the tonic contraction of smooth muscles.     

Electrochemical potential difference for H+-ions as a regulator of redox profile of membrane during ATP-dependent ion transport in E. coli

The importance of delta mu H+ for transport of K+ via K(+)-ionophore and H(+)-K(+)-pump was studied. It was shown that the operation of the pump was decelerated by oxidant ferrycyanide, whereas sulfhydryl reagent dithiothreitol (DTT) drastically accelerated ATP driven ion exchange. Introduction of protonophore CCCP into the medium completely blocked the pump operation. However, the addition of DTT after CCCP restored the high level activity of the pump. At the same time DTT was unable to restore K+ accumulation after CCCP in aerobically grown bacteria for which the K+ uptake was performed across the electrical field gradient. Thus it was established that delta mu H+ was necessary for ATP driven ionic systems as a regulator of the membrane redox state.