Influence of Tl+ on The ca2+ and na+ movemenT across raT neonaTal cardIomyocyTes and raT hearT mITochondrIa membranes

thallium is known to produce one of the most complex and serious patterns of toxicity, involving a wide range of human organs and tissues. the toxic impact on biologic organisms is linked especially to the ability of tl+ to disturb calcium homeostasis and to permeate easily the inner mitochondrial membrane (IMM). the aim of this work was to study the effects of Tl+ on intracellular Ca2+ dynamics in rat neonatal cardiomyocytes as well as on sodium penetrability of the IMM and tl+-induced mitochondrial permeability transition pore (MPtP) opening in isolated Ca2+-loaded rat heart mitochondria (RHM). The use of the fluorescent calcium indicator Fura 2 aM showed that tl+ induced calcium influx across the plasmatic membrane, resulting in calcium ([Ca]i) increase in the cytoplasm. this increase was even more pronounced in experiments with accelerating of tl+-transmembrane fluxes by nonactin. It was nevertheless abolished by the removal of extracellular Ca2+ ions, but was not inhibited by a calcium-channel blocker (nifedipine). tl+ did not release calcium from the intracellular stores. tl+ potentiated sodium permeability of the IMM because swelling of nonenergized rhM in medium containing tlNo3 and NaNo3 was enhanced at high tl + concentration. the calcium load of rhM induced MPtP opening which was accompanied by the increase of the swelling as well as the decrease of the inner membrane potential and of state 40 (basal) and state 3UDNP (2,4-dinitrophenol-uncoupled) respiration. These effects of Tl+ were suppressed by MPtP inhibitors (cyclosporine a, aDP and n-ethylmaleimide). the data obtained showed that tl+-stimulated influx of extracellular calcium into cardiomyocytes could cause calcium and sodium rhM overload, which lead to the MPtP opening, thus determining the sensitivity of heart muscle to thallium intoxication.

the liver, kidney, or skin [811]. At the same time, considerab le structural damage and grains were ob served in SR, nuclei, myofibrils, and mitochondria of the rats. Additionally, cell muscle granules were deprived of glycogen. The toxic impact on biologic organisms is linked to the ability of Tl + to perme ate easily the inner mitochondrial membrane (IMM) and to substitute K + in K + dependent biochemical re actions or transport processes [1,12].
It is generally known that a moderate calcium load of mitochondria facilitates the appearance of the permeability transition pore (MPTP) in the inner mi tochondrial membrane (IMM) [13]. In this case, the MPTP is in a low conductance state, and the IMM becomes penetrable to H + , K + , Na + , and Ca 2+ with massive mitochondrial swelling and decreased mem brane potential (ΔΨ mito ). It was previously found that Tl + raised the concentration of Ca 2+ and Na + in iso lated rat hepatocytes [14]. The calcium load of succi nateenergized rat liver mitochondria (RLM) in me dium containing TlNO 3 and nitrates (KNO 3 , NaNO 3 , NH 4 NO 3 ) caused Tl + induced MPTP opening in their inner membrane. This was manifested as an increase in swelling and a decrease in both ΔΨ mito and state 4, state 3, and state 3U DNP (2,4dinitrophenol(DNP) stimulated) respiration [15]. The Tl + induced MPTP with the latter three effects was found in similar ex periments with Ca 2+ loaded RHM, injected into me dium containing TlNO 3 and KNO 3 [16]. The opening of the MPTP in calciumloaded RLM and RHM, as well as the decrease in both ΔΨ mito and respiration, and the increase in swelling were slowed down in the presence of MPTP inhibitors (ADP and cyclo sporin A (CsA)) [15,16]. The effects of Tl + have not been investigated in intracellular calcium in cardiac myocytes. There are no data on the joint effect of Ca 2+ and Na + on the Tl + induced MPTP opening in the inner membrane of RHM. The aim of this work is to study the effects of Tl + on intracellular Ca 2+ dynamics in rat neonatal cardiomyocytes. Another goal of our work was to study the effect of Tl + on both the sodium penetrability of the IMM and the Tl + induced MPTP in Ca 2+ loaded rat heart mito chondria. We, therefore, examined the swelling, ΔΨ mito decrease, as well as state 4 0 , state 3, and state 3U DNP respiration in the RHM in medium containing TlNO 3 and NaNO 3 in the presence of Ca 2+ as well as the MPTP inhibitors (CsA, ADP, nethylmaleimide (NEM)).

materials and methods
animals. Female and male Wistar rats (250 300 g) were used in the research. The animals were kept at 20-23 °C under 12h light/dark cycle with free access to the standard rat diet and water ad libi tum. All treatment procedures of animals were per formed in accordance with the Animal Welfare act and the Institute Guide for Care and Use of Labora tory Animals.
Cultivation of rat neonatal cardiomyocytes. Methods of cultivation and Ca 2+ registration have been described in detail [17]. Briefly, newborn Wistar rats (1dayold) were used for the prepara tion of the myocytes' culture. Rats were subjected to carbon dioxide. Dissected hearts were thorough ly disin tegrated and incubated at 37° С for 40-50 min in Ringer's solution for rodents: 146 mM NaCl, 5 mM KCl, 2 mM CaCl 2 , 1 mM MgCl 2 , 11 mM glu cose, 10 mM HEPES (pH 7.4), containing type IA collagenase (1 mg/ml, Sigma) and trypsin (0.12%, Bio lot, Russia). After filtration and centrifuga tion at 1000 rpm for 10 min, the cells were resus pended in the medium containing DMEM medium with 10% fetal serum, 50 IU/ml penicillin and 50 μg/ml of streptomycin. Nonmuscular cells were de leted with a preliminary incubation on a Petri dish over 40-50 min at 37 °С. Non-attached cells were then incubated on cover slips (12×24 mm) covered with a polyDlysine (0.1 mg/ml, MP Biomedicals, USA). The cells were incubated in a СО 2 incubator (Jouan, France) at 5% of СО 2 , a humidity of 95% and at 37 °С. The medium was replaced every two days. The cells were cultured for 5-10 days before the experiments.
Measurements of intracellular calcium. Cover slips with cells were loaded in Ringer's solution for rodents with Fura 2 AM at a concentration of 5 μM for 1 h at 26 °С. Then, cells were washed with a Ringer's solution for 30 min and the intracellular cal cium concentration ([Ca 2+ ] i ) was measured at room temperature on cover slips placed above an inverted microscope lens. Nonactin of 1 μM we used to en hance the entrance of Tl + into the cardiomyocytes (Fig. 1). During Ca 2+ experiments in Ca 2+ free EGTA solution, all Ca 2+ was replaced with 2 mM EGTA (Fig. 3). [Ca 2+ ] i was measured using a computer analy sis system for intracellular ion content (Intra cellular Imaging & Photometry System, USA), with inverted microscope Nikon TMS (objective ×30) and monochrome digital video camera, RS_170 CCD (Cohu Inc., USA). The cells were excited with light at wavelengths of 340 and 380 nm, and emission was registered at 510 nm. The builtin computer InCytIm2TM software allowed measuring the intra cellular calcium ion concentration as the ratio of the fluorescence emission intensities excited at 340 and 380 nm (F 340 /F 380 ), using a calibration curve based on calibration solutions (Fluka Chemicals) with known Ca 2+ concentrations.
Mitochondrial isolation. Rat heart mitochon dria were isolated from Wistar adult male rats (200 250 g) according to procedure which explicated more detail in Korotkov et al. [16]. The rat hearts were minced, washed, and homogenized by using an MPW302 polytron homogenizer from Mechanika Precyzyjna (Warsaw, Poland) in the medium I con taining 70 mM sucrose, 220 mM mannitol, 2 mM EGTA, 10 mM TrisHCl (pH 7.3), and 0.1% BSA (Minsk, Belarus). To precipitate nonhomogenized tissue particles, the homogenate was then centri fuged for 10 min at 400 g by a K24 centrifuge (Ger many). Further, the containing mitochondria super natant centrifuged for 10 min at 8500 g and the RHM sediment was kept on ice. To increase the number of mitochondria, the tissue particles were repeatedly homogenized by using a potter homogenizer with Teflon pestle. Next, the tissue homogenate and the supernatant were successively centrifuged at 400 and 8500 g, accordingly. This procedure was per formed twice. At the final stage, the RHM sediment obtained from the three cycles was twice rinsed out in a medium II containing 300 mM sucrose, 10 mM TrisHCl (pH 7.3), 5 µM EGTA, and 0.1% BSA (Sig ma, USA) with followed the RHM sedimentation for 10 min at 6500 g. All the above procedures were per formed on ice. The protein concentration tested by the Bradford method was 2530 mg/ml.
Mitochondrial swelling. Swelling of the RHM was evaluated by a decrease in the appa rent absorban ce of mitochondrial suspension at room temperature on an SF46 spectrophotometer (LOMO, St. Petersburg, Russia) at 540 nm. Mito chondria (1.0 mg/ml of protein) were administrated in 1cm cuvette with 1.5 ml of sucroseadjusted 400 mOsm medium containing 075 mM TlNO 3 (Fig. 4, a) or 25 mM TlNO 3 , 100 mM sucrose, 1 mM TrisPO 4 , and 200 µM CaCl 2 (Fig. 4, b). All media above contained 125 mM NaNO 3 , 5 mM TrisNO 3 (pH 7.3), 2 µM rotenone, and 1 µg/ml of oligomy cin. The following agents were administered into the medium before addition of mitochondria (where indicated): 0.5 mM ADP, 1 µM CsA, and 50 µM NEM. The concentrations of other additions showed in the figure legends. The swelling, respiration rates and ΔΨ mito were tested in the 400 mOsm media in order to check the comparability and consistency be tween the events in different experiments. The media osmo lality was fixed by sucrose addition.
oxygen consumption assay. The oxygen con sumption rates (ng atom O/min per mg of protein) were tested polarographically using Expert001 ana lyzer (EconixExpert Ltd., Moscow, Russia) in a 1. Mitochondrial membrane potential. The ΔΨ mito induced by 5 mM succinate on the inner membrane of the RHM (Fig. 6) was evaluated according to Waldmeier et al. [19]. We measured intensity of safranin fluorescence (arbitrary units) in the mito The use of millimolar thallium(I) concentrations for our study was due to the specifics of isolated mitochondria research that was our discussed earlier [20,21]. Statistical analysis. Data are presented from at least three independent experiments, with 5-25 cells in each experiment. The means ± s.e.m. are shown. N is the number of experiments, n is the number of cells. The statistical differences in results and corre sponding Pvalues were evaluated using two popu lation ttest (Microcal Origin, Version 6.0, Microcal Software). These differences are presented as per cent of the average (P < 0.05) from one of three in dependent experiments (Fig. 16). 2+ ] i in cardiomyocytes. In our experiments with rat neonatal car diomyocytes, we used Tl + concentrations compa rable to those used in other studies [4,17,22,23]. An increase of [Ca 2+ ] i was observed in the myocytes after 300400 sec of Tl + application ( Fig. 1, a). To enhance the entrance of Tl + into the cardiomyocytes, we used nonactin which was found to be markedly increased Tl + penetration into energized rat liver mi tochondria [12,20,24]. It has previously been shown that cardio myocytes were killed after application of nonactin [25]. We, therefore, used essentially a low concentration of nonactin (1 µM) and a short expo sure time of 16 min maximum. Experiments with nonactin showed that the Tl + induced Ca 2+ increase occurred in a shorter time (by 100200 sec) after Tl + nonactin application (Fig. 1, b and c). The effects of 1 µM nonactin (Fig. 1, c) on [Ca 2+ ] i at the beginning of the experiments and before Tl + application were minor and transient, and were seen within a short time span (100200 sec). Only 19±2% of the cells responded to nonactin (n = 37, N = 3). The effects of Tl + on [Ca 2+ ] i dynamics were dependent on the Tl + concentration and duration of Tl + application (Fig. 1,  c). For example, 10±2 and 55±3% of the cells in the presence of 1 µM nonactin responded to 1 mM Tl + and 1.5 mM Tl + (n = 79, N = 3) respectively. It should be noted that the cardiomyocytes did not respond to Tl + simultaneously. The a-c images of cardio myocytes loaded with Fura 2 AM and fluorescing at 380 nm ( Fig. 2) are correspondingly shown at the times indicated by arrows a, b and c in Fig. 1, b. The fluorescent signal of F 380 was not recovered after the end of Tl + exposure (Fig. 2, c). This may be due to the extrusion of Fura 2AM through the plasmalemmal pore(s), which may be formed after Tl + application .

tl + induces an increase in [Ca
experiments in Ca 2+ -free eGta solution. In our experiments, the Ca 2+ in the extracellular solution was replaced by 2 mM EGTA before (Fig. 3, a) and after (Fig. 3, b) the addition of Tl + . The extracellular  Fig. 1, b. Panel afluorescence at the start of the experiment (arrow a). Panels b and c -fluorescence at the times indicated by the arrows b and c in Fig. 1, b, respectively a b c Ca 2+ was subsequently restored to 2 mM (Fig. 3, a,  b). As shown in Fig. 3 (a), Tl + did not induce [Ca 2+ ] i release from intracellular stores. When Ca 2+ influx was inhibited in Ca 2+ free EGTA solution, the sus tained increase in Ca 2+ stimulated by application of Tl + was absent (Fig. 3, b). After restoration of ex tracellular Ca 2+ , an increase in [Ca 2+ ] i was observed (Fig. 3, b). To find out which Ca 2+ channels are re sponsible for the entrance of Tl + , we used plasmatic LCa 2+ channels blocker, nifedipine at a concentra tion of 50 µM, which had no effect on Ca 2+ influx stimulated by Tl + application (Fig. 3, c). effects of tl + and Ca 2+ on swelling of rat heart mitochondria in nitrate media. Brierley et al. [26] earlier showed that the IMM passive sodium permeabi lity can be estimated by the swelling of nonener gized mitochondria in medium containing NaNO 3 because the IMM is free penetrable to NO 3 anion. It is known that mitochondrial respiratory enzymes (NADH, succinate, and malate dehydro genase) poorly inhibited by Tl + millimolar concen trations since Tl + showed a weak interaction with molecular thiol groups [27], and it is this fact that was the main reason for the use of millimolar Tl + concentrations [9,28] in experiments with isolated mitochondria what we explained previously in more detail [20]. Earlier, we found that Tl + increased the permeability in experiments with RLM, injected into 400 mOsm medium containing 0-75 mM TlNO 3 and 125 mM NaNO 3 [29]. The swelling of nonener gized RHM increased steadily when the TlNO 3 concentration was increased in the medium (Fig. 4,  a). However, the swelling was almost not observed when 125 mM NaNO 3 in the medium was replaced by 250 mM sucrose (bold dash traces). Injection of succinate caused these organelles to contract, but the contraction abated when 75 mM TlNO 3 were added (Fig. 4, a). The calcium load of mitochon dria is accompanied by massive swelling of these organelles, due to opening of the mitochondrial permeability transition pore (MPTP) in the inner membrane [13]. Swelling of succinateenergized rat liver mitochondria in medium with TlNO 3 , 125 mM nitrates (KNO 3 , NH 4 NO 3 , NaNO 3 ) and 100 μM Ca 2+ was caused by the opening of the MPTP in the in ner membrane [29]. The 3minute swelling of non energized rat heart mitochondria (RHM) in medium containin 25 mM TlNO 3 and 125 mM NaNO 3 (Fig. 4, b) was not affected by 300 µM Ca 2+ (con trol trace with Ca 2+ alone) nor by the MPTP inhibi tors (ADP, CsA, NEM). The addition of succinate resulted in massive mitochondrial swelling (control trace) which was followed by mitochondrial contrac tion, increased in series CsA, NEM < ADP < ADP + CsA (Fig. 4, b).
Influence of Tl + and Ca 2+ on oxygen consumption rates of rat heart mitochondria and the inner membrane potential in nitrate media. State 4 0 of suc cinateenergized RHM was diminished in medium containing 50-75 mM TlNO 3 and 125 mM NaNO 3    (Fig. 5, b), correspondingly. When the TlNO 3 con centration was raised from 25 to 75 mM, state 3 and state 3U DNP respiration of RHM dimini shed in medium containing NaNO 3 (Fig. 5, a) or NH 4 NO 3 (Fig. 5, b). Compared with experiments free of Ca 2+ (Fig. 5 and 6 (free of Ca 2+ trace)), a decrease in state 3U DNP respiration (control trace with Ca 2+ alone) and in RCR DNP was found after admini stration of 100 µM Ca 2+ in the 400 mOsm medium with 25 mM TlNO 3 and 100 mM sucrose, as well as with 125 mM NaNO 3 (Fig. 5, c) or 125 mM NH 4 NO 3 (Fig. 5, d).

Fig. 6. Effects of T l+ and Ca 2+ on the inner membrane potential (ΔΨ mito ) of rat heart mitochondria. Mitochondria (0.5 mg/ml of protein) were added to medium containing 15 mM tlNo 3 , 120 mM sucrose, 5 mM tris-No 3 (ph 7.3), 1 mM tris-Po 4 , 3 μM safranin, 2 μM rotenone, and 1 μg/ml of oligomycin as well as 125 mM NaNO 3 (panel a) or 125 mM Nh 4 No 3 (panel b). Down-directed arrows show additions of mitochondria (rhM) and 5 mM succinate (Succ). Injections of 30 μM Ca
The decrease in the respiration and in RCR DNP was counteracted notably in the presence of MPTP in hibitors (ADP and CsA) but the decrease remained constant in experiments with NEM (Fig. 5). When 25 mM TlNO 3 was replaced by 50 mM sucrose in the medium with 125 mM NaNO 3 (Fig. 5, c, free of Tl + traces), state 3U DNP respiration and RCR DNP after Ca 2+ injection and regardless of ADP with CsA addi tions (bold traces) were similar to ones found in like experiments with this medium without Ca 2+ (Fig. 5,  a). The decrease of ΔΨ mito on the inner membrane of RHM (Fig. 6) was not so strong (free of Ca 2+ trace) after administration of succinate into the 400 mOsm medium containing 15 mM TlNO 3 and 110 mM su crose together with 125 mM NaNO 3 (Fig. 6, a) or 125 mM NH 4 NO 3 (Fig. 6, b). The ΔΨ mito decrease af ter injection of Ca 2+ into the medium (control trace with Ca 2+ alone) revealed a notable acceleration, which was clearly eliminated in experiments with the MPTP inhibitors (ADP + CsA). Ltype Ca 2+ channels are believed to be the main transporter for the transsarcolemmal Ca 2+ influx in adult cardiomyocytes. The impact of Tl + on isolated hepatocytes was accompanied by an in crease in cytosolic concentrations of Ca 2+ [14]. We, therefore, anticipated that Tl + could also stimulate this pathway. Talbot [23] found that Tl + entered the nerve terminal primarily via Ca channels. While nifedipine (> 2 μM) blocked the Ca 2+ transients in cardiomyocytes [30], the influx of Ca 2+ into rat neo natal cardiomyocytes in our experiments was not blocked in the presence of 50 μM nifedipine and 1.5 mM Tl + (Fig. 3, c). Thus, the mechanism of the Tl + induced rise in [Ca 2+ ] i (Fig. 1, a) is likely not as sociated with Tl + activation of the Ltype Ca 2+ chan nels. It can be assumed that the fluxes of extracel lular calcium are most likely due to the activation of the Ca 2+ permeability pathway or calcium channels (other than Ltype) in the plasma cell membrane, which causes the direct influx of Ca 2+ into the cy toplasm.
Na + /K + ATPase catalyses transport of Tl + into myocytes [31,32]. To stimulate the influx of Tl + into the cells, we used nonactin (Fig. 1, b and c). Earlier, it has been shown that cardiomyocytes were killed by the application of a high concentration of non actin (10 µM) during a prolonged exposure time (60 min) [25]. Under these circumstances, 33% of the cardiomyocytes were killed [25]. In our experiments (Fig. 1, b and c) we used essentially a low concen tration of nonactin (1 µM) and short exposure time, 16 min maximum. Nonactin was able to increase the cytoplasmic calcium concentration at 20100 µM in nonexcitable cells (the murine tumor cell lines) [33]. These authors suggested that the rise in [Ca 2+ ] i was most likely due to the release of calcium from intra cellular stores, but they did not perform experiments with a Ca 2+ free solution. Therefore, we conducted some experiments with nonactin alone to discover its ability to increase [Ca 2+ ] i . We showed that the Ca increase induced by nonactin before Tl + application was rather low and transient (Fig. 1, c). Additionally, most of the cells did not respond to a 1 µM concen tration of nonactin. It cannot be ruled out that nonac tin, which stimulates both Tl + entrance into the cells, and the subsequent Tl + induced Ca 2+ increase, is able to facilitate Tl + induced channel opening in the inner mitochondrial membrane.
Electrophysiological research has detected Tl + permeability in the SR channel [34]. Tl + ions had a selective impact on the muscle and its mitochondria, which are the main site of Tl + distribution [11]. Thal lium interacts with the excitability of periphe ral and myocardial muscle cells [1]. Some possible toxic mechanisms of Tl + can be the disturbance of calcium homeostasis, inhibition of cellular respiration, as well as ligand formation of Tl + with protein sulfhy dryl groups [1]. Our in vitro studies with RLM have revealed that Tl + was able to form complexes with mitochondrial proteins [35]. In our experiments, we observed that the inhibition of Ca 2+ influx with Ca 2+ free EGTA solution prevented the sustained/ transient increase in [Ca 2+ ] i induced by the addition of Tl + (Fig. 3, a). Thus, Tl + did not induce [Ca 2+ ] i re lease from intracellular stores. Addition of Ca 2+ free EGTA solution after a Tl + induced calcium rise also led to a reduction in [Ca 2+ ] i almost to the resting level (Fig. 3, b). The reduction of Ca 2+ in cardiomyocytes in Ca 2+ free EGTA solution implies that Ca 2+ se questration and extrusion mechanisms are still able to function after Tl + application. These mechanisms can be extrusion of [Ca 2+ ] i by both plasmalemmal so dium calcium exchanger and plasma membrane Ca 2+ ATPase, and possibly mitochondrial Ca 2+ uptake through the Ca 2+ uniporter. Nevertheless, we believe that Tl + at 1.5 mM is toxic for cells, because we ob served a dramatic loss of intracellular dye Fura 2AM (namely F 380 ) at the end of all experiments. The fluo rescent signal of F 380 was not recovered after the end of Tl + exposure (Fig. 2, c). This may be due to the extrusion of Fura 2AM through the plasmalemmal pore(s), which may be formed after Tl + application.
It is assumed that Tl + ions induced calcium overloading in the cardiomyocytes followed by channel opening in the inner mitochondrial mem brane, and this may contribute to the toxic effect of thallium on the myocardium. The BTC-AM fluores cence assay showed that Tl + can be transported via voltagegated potassium channels, as well as small conductance and largeconductance calciumactiva ted potassium channels in mammalian HEK293 cells and cultured rat muscle cells [36,37]. Experi ments with isolated hepatocytes and PC12 cells in dicated that Tl + induced apoptosis, decreased intra cellular glutathione, and enhanced the production of reactive oxygen species, H 2 O 2 , and lipid peroxidation [38,39]. These deleterious effects of 100 µM Tl + on hepatocytes were reduced considerably by the MPTP inhibitors (cyclosporine A and carnitine) [39]. The mechanistic basis of thallium's toxicity is the abili ty of Tl + to replace K + in K + dependent biochemi cal processes. It penetrates easily the inner mitochon drial membrane, releases Ca 2+ from intracellular compartments, and changes cell cycle regulation [1,12,14]. In our preliminary experiments, we found that 1 mM Tl + exerted a negative inotropic effect on the spontaneous contraction of frog atrium (Sobol and Korotkov, unpublished data). In this study, we showed for the first time that Tl + stimulated the un controlled influx of extracellular calcium ions into the cardiomyocytes.
The Tl + fluorescence assay showed that Tl + is transported into rat heart mitochondria through ATPsensitive (mitoK ATP ) or BKtype Ca 2+ activated (mitoK Ca ) potassium channels [40,41]. Tl + inhibited the Na + dependent Ca 2+ efflux from RLM [42]. The swelling of deenergized rat liver mitochondria (RLM) in nitrate media indicated that the inner mi tochondrial membrane (IMM) displays low perme ability to K + but high permeability to Tl + [12,26,43]. Similar experiments with deenergized RLM in sucroseadjusted 400 mOsm medium containing a mixture of 075 mM TlNO 3 with 125 mM nitrates (KNO 3 , NH 4 NO 3 , NaNO 3 ) correspondingly revealed a Tl + induced increase in IMM permeability to K + , H + , and Na + [29]. This increase resulted in additional mitochondrial swelling and decreased state 3 and state 3U DNP respiration. On the contrary, these effects were not found in like experiments with the medium containing 075 mM TlNO 3 together with 250 mM sucrose instead of 125 mM nitrates. So, these ef fects of Tl + in the experiments with RLM were en hanced in the series with TlNO 3 : sucrose < KNO 3 < NaNO 3 ≤ NH 4 NO 3 [29]. The present results with the medium containing NaNO 3 or sucrose (Figs. 4, a and 5) and our earlier results [16] with medium containing TlNO 3 and nitrates (KNO 3 , NH 4 NO 3 ) revealed a similar series in the experiments with deenergized rat heart mitochondria (RHM). Our experiments with the different mitochondrial prepa rations (RLM and RHM) therefore showed that the Tl + induced increase in the passive ion permeabili ty of the IMM is manifested by visible mitochon drial swelling and a simultaneous decrease in state 3 and state 3U DNP respiration of these organelles. We have shown previously that quinine inhibited the succinateinduced contraction of RLM, preswol len in medium containing TlNO 3 with nitrates [29]. This finding supports the hypothesis that the mito chondrial K + /H + exchanger can catalyze, extruding both Tl + [12] and Tl + induced excess of the univa lent cations (K + , Na + , NH 4 + ) [43] from the mitochon drial matrix. The succinateinduced contraction of the RHM (Fig. 4, a) [16] and RLM [29] can thus be explained also by the participation of the K + /H + ex changer.
The cytotoxic effects of Tl + in experiments with isolated rat hepatocytes were significantly reduced in the presence of CsA and carnitine, which are MPTP inhibitors [39]. The main components of MPTP are believed to be the mitochondrial phosphate carrier (PiC) and cyclophilin D (CyPD), whereas the ade nine nucleotide translocase (ANT) is referred to as its regulatory part [44]. It was earlier found that Tl + in sucrose Ca 2+ free media has not effected on state 3 and state 3U DNP mitochondrial respiration that is in agreement with the finding that Tl + does not inhibit respiratory enzymes [9,28,43]. In our experiments with Ca 2+ loaded RLM, the Tl + induced MPTP opening in medium containing TlNO 3 and nitrates (KNO 3 , NaNO 3 , NH 4 NO 3 ) was facilitated by inor ganic phosphate [43] but impeded in the presence of both the MPTP inhibitors (ADP, CsA, Mg 2+ ) and the PiC inhibitor mersalyl [15,45]. This allowed us to reach a conclusion regarding the participation of the ANT and the PiC in the Tl + induced MPTP opening in the inner membrane [21,45]. We earlier found that a closure of mitochondrial potassium chan nels (mitoK ATP and mitoK Ca ) increased the MPTP opening [46]. However, blocking of the mitochon drial Ca 2+ uniporter by ruthenium red, Y 3+ , La 3+ and Ni 2+ deceased the MPTP opening [15,47].
On the other hand, experiments with the Ca 2+ loaded RHM showed a decrease in state 3U DNP res piration, the RCR DNP , and ΔΨ mito together with the increased swelling in medium containing TlNO 3 as well as KNO 3 and NH 4 NO 3 [16] or NaNO 3 (Figs. 4, b, 5, 6). As in the case of the Ca 2+ loaded RLM [15], these effects were attenuated considerably in the presence of the MPTP inhibitors (ADP, CsA) (Figs. 4, b, 5, 6). The detected contraction of the Ca 2+ loaded RHM, preswollen in the nitrate media, (Fig. 4, b) [16] and also of Ca 2+ loaded RLM [15] in presence of ADP and CsA is in agreement with the proposed possible cooperative interaction [48] be tween these inhibitors and the Ca 2+ binding sites of ANT in mammalian mitochondria. The increase in mitochondrial swelling (Fig. 4, b) and the decrease in both state 3U DNP respiration and the RCR DNP (Fig. 5, c) [16] were visibly eliminated in like experiments with RHM injected into the containing 125 mM NaNO 3 medium with 25 mM TlNO 3 replaced by 50 mM sucrose. On the other hand, the even greater decline in state 3U DNP and the RCR DNP in the presen ce of NEM (Fig. 5, c and d) is most likely associa ted with the inhibition of succinate dehydrogenase that we observed earlier in similar experiments with RLM [21]. Above results of these experiments with Ca 2+ loaded RHM (Figs. 4, b, 5, 6) [16] and Ca 2+ loaded RLM [15] suggest that the Ca 2+ load of mitochondria in medium containing a mixture of TlNO 3 with nitrates (KNO 3 , NaNO 3 , NH 4 NO 3 ) may be reasoned by the opening of CsAinhibited and ADPdependent Tl + induced MPTP pores in the inner membrane of these organelles. The decrease of state 4 0 and state 3U DNP respiration is linked to the increased swelling of succinateenergized Ca 2+ loaded RHM (Figs. 4, b and 5, c) [16] or Ca 2+ loaded RLM [15]. These results may be associated with a decreased activity of respiratory enzymes due to the deformed spatial structure of the inner membrane, caused by the mitochondrial swelling due to the Tl + induced Na + entry (Fig. 4, a) into the matrix of the Ca 2+ loaded RHM [15].
In this study, we showed for the first time that Tl + stimulated the uncontrolled influx of extracellu lar calcium ions into cardiomyocytes. We observed that the inhibition of Ca 2+ influx with Ca 2+ free EGTA solution prevented the sustained increase in [Ca 2+ ] i induced by Tl + application. Based on earlier studies and our research, we conclude that more intensive stimulation of the IMM ionic permeabili ty and the Tl + induced MPTP in experiments with RHM, versus RLM, can be explained by the greater sensitivity of muscle tissue, especially heart mus cle, but not liver, to intoxication by thallium salts. Furthermore, the more potent toxicity of thallium to RHM can thus help us to find a suitable antidote against thallium poisoning in humans as well. Howe ver, Tl + -stimulated influx of extracellular calcium into the cardiomyocytes can cause calcium and so dium overload of RHM, which will then lead to the MPTP opening. The latter hypothesis may shed light on the greater sensitivity of heart and striated mus cle, than of liver, to thallium intoxication in vivo.
acknowledgments. Authors are grateful to Ms Terttu Kaustia for correcting the Eng lish. The research was carried out within the state assignment of FASO of Russia (theme No. АААА-А18-118012290142-9). Safranin fluo rescence was measured using of Research Resource Center equipment for the physiological, biochemical and molecularbiological studies (Sechenov Institute of Evolutionary Physiology and Biochemistry, the Russian Academy of Sciences).