Document LpLemKkwmaQx19OE5vxb1JNE5

A R ^ - 0,Gf THE EFFECT OF PERFLUORINATED ARYLALKYLSULFONAMIDES ON BIOENERGETICS OF RAT LIVER MITOCHONDRIA Kendall B. Wallace and Anatoli Starkov Department o fBiochemistry and Molecular Biology University o f M innesota School o f Medicine Duluth, M N 55812, U.S.A. Key words: Mitochondria, Membrane potential, Respiration, Uncoupling, Detergents. Supported by a grant from The 3M Company. 004150 MATERIALS AND METHODS KBW 2/4/98 T he isolation of m itochondria. M itochondria were isolated from liver of adult male Sprague-Dawley rats (-200 g body weight) by a conventional differential centrifugation procedure. Animals were killed by decapitation. Liver was excised and weighed and cooled in 40 ml o f isolation medium (210 mM m annitol, 10 mM sucrose, 5mM HEPES-KOH (pH 7.4), 1 mM EGTA). Cooled liver was minced with scissors and w ashed twice w ith 20 ml o f isolation medium, then diluted with the same medium and hom ogenized for 1 m in w ith a motor-driven Potter hom ogenizer (Teflon pestle, glass beaker). The tissue to medium ratio was 1:8 (g:m l). The homogenate was filtered through gauze and centrifuged for 10 min x 700 g, f = 4 C, and the m itochondrial pellet was recovered from the supernatant by centrifugation at 10,000 g x 10 min. The pellet was resuspended in 10 ml o f washing medium (210 mM m annitol, 10 mM sucrose, 5mM HEPES-KOH, pH 7.4) supplem ented w ith bovine serum album in (BSA, 1 mg x m l'1). The suspension of m itochondria was diluted to 35 ml w ith the same medium without BSA, and centrifuged at 10,000 g x 10 min. The final m itochondrial pellet was resuspended in washing medium to a protein concentration o f 70-80 m g x m l'1and stored on ice. M easurem ents. Mitochondrial membrane potential (A*P) was estim ated from TPP+ ion distribution m easured w ith a TPP+- selective electrode constructed according to Kamo et al., 1979. M itochondrial membrane potential was calculated as described elsewhere (Rottenberg, 1984 ). The rate o f oxygen consumption by m itochondria was measured with a hand-made Clark -type oxygen electrode. B oth the m itochondrial membrane potential and the respiration rate were recorded sim ultaneously using a m ultichannel incubation chamber equipped w ith a m agnetic stirrer. The volume o f the chamber was 1.8 ml. All experim ents were perform ed at room tem perature (25 C). The TPP+ -sensitive electrode was calibrated by sequential additions o f known amounts o f TPP+C1' before the addition o f m itochondria (see F ig .l). The respiration rates were calculated assum ing the initial oxygen concentration to be equal to 240 pM. The m itochondrial membrane potentials were slightly (-15 %) underestim ated and the oxygen consum ption rates were overestim ated because no correction was made for lower TPP+ binding constants and for low er equilibrium concentration o f dissolved oxygen in the high ionic strength medium used in our experiments. This is fully satisfactory because the values are used to compare the effects o f different compounds under the same conditions rather than for energetic calculations. Protein concentration was determ ined by the Bradford assay. Bovine serum album in was used as a standard. Additions. PF compounds were dissolved in absolute ethanol (for except for PF143 which was dissolved in deionised water). Prelim inary study revealed that all PF compounds are very hydrophobic and tend to precipitate in our incubation medium. Due to this, a set o f dilutions was made for each PF compound to obtain an array o f concentrations starting from 100 pM down to 6.25 pM. Dilutions were made by adding a volume of 2 004151 KBW 2/4/98 absolute ethanol to a volume of PF com pound stock solution. PF compounds were added to mitochondria as 1.8 pi volum e o f a solution o f desirable initial PF concentration. This approach allowed us to obtain satisfactory reproducible results. Reagents. M annitol was from Aldrich, sucrose U ltra Pure from ICN, all other reagents were from Sigma. Bovine serum album in was essentially fatty acid free. 3 004152 RESULTS AND DISCUSSION KBW 2/4/98 1. The effects of PF compounds on oxidative phosphorylation in rat liver mitochondria Because this in vitro study was aim ed to reveal the potential in vivo toxic properties o f a set of compounds, a buffered high ionic strength potassium chloride medium resem bling cellular cytoplasm m ilieu was chosen. By the same reason, glutamate plus m alate were used as m itochondrial respiratory substrates. The oxidation o f glutam ate plus m alate involves the m ost vulnerable obligatory complex for hydrophobic compounds, Complex I of mitochondrial respiratory chain, as well as all other respiratory enzymes. We also developed a special assay procedure. It allows obtaining detailed data on the effects o f a com pound on m itochondrial energetics. The procedure also allows to com pare in a universal way the efficiency o f various compounds affecting the energy production in mitochondria. Fig. 1 explains the design o f experiments. The picture shows typical data obtained by sim ultaneous recording o f m itochondrial respiration and AT. The first addition o f ADP induced a transient increase in respiration rate Fig.l A typical example oi simultaneous recording of the mitochondrial respiration rate and changes in , \ lH Incubation medium contained 125 mMKCI, 10 mMMOPS, 4mM KH2PO4. 5 mM glutamate, and 5 mM malate, pH 7.4. R e d cuive. mitochondrial respiration; blue curve, the changes in VK. Additions; Mito. ret liver mitochondria, 1 mg/ml, AD P, 100 |iM; PF12M, 6.25 |iM;DNP, 40 uM 2,4-dinitrophenot T P P *. 0.2,0.2.04.0.0, and 0.4 uM (from top to bottom, total concentration 2 mM). o f m itochondria which was paralleled by a drop in AT. It indicates that m itochondria entered socalled m etabolic State 3, that is they phosphorylate ADP for the expense of respiratory substrates. The values of respiration rate during (V5t3) and after phosphorylation (V,t4) were used to calculate respiratory control index (RCI) equal to the ratio o f Vst3 to Vst4. The amount o f oxygen consumed by m itochondria during phosphorylation was used to calculate A D P:0 ratio. A fter the phosphorylation o f ADP was completed, the rate o f respiration decreased and AT in m itochondria spontaneously restored indicating the onset of State 4. The addition o f a compound (PF12M in Fig. 1) to m itochondria was followed by the second addition of ADP, and second RCI (RCI-compound) and A D P:0 (ADP:0-com pound) ratios were calculated. A fter the onset of second State 4, an uncoupler was added at 4 004153 KBW 2/4/98 a concentration that stimulated the respiration o f m itochondria to the maximal level determ ined by the activity o f respiratory chain. This relatively simple approach allows to estim ate several essential properties o f a compound in relation to the energetics of mitochondria. By the analysis o f data obtained in such assay, it is possible to show if a compound (a) inhibits or stim ulates the respiratory chain; (b) inhibits the entry of substrates into m itochondria; (c) specifically inhibits the enzymes o f the oxidative phosphorylation system; (d) uncouples the oxidation and phosphorylation in m itochondria. The approach described above was applied to all PF compounds. Fig.2 shows the typical record for PF10. 1 min_W, Fig.2. The effect ofPFlO on the respiration rate and AW ofrat liver mitochondria. The composition of incubation medium and other conditions, as in Fig.l. Additions: PF10,62.5 pM; ADP, 100 pM; DNP, 40 pM PF10 in relatively high concentration stim ulates the respiration and decreases AW in State 4 whereas has no influence under m etabolic State 3. It affected neither State 3 respiration rate nor ADPiO ratio (Table 1.) 5 004154 KBW 2/4/98 Table 1. The effect ofPFlO on the respiration rates o f rat liver mitochondria under various metabolic states. The experimental conditions and additions were as in Fig.2 Abbreviations used: Vim, the rate of oxygen consumption by mitochondria before any additions were made; Vsq, the respiration rate in the presence of 100 pM ADP; Vst4, the respiration rate after phosphorylation of ADP has been completed; RCI, the respiratory control index, calculated as VsQto Vst4ratio; ADP:0, the efficiency index, calculated as the ratio of ADP (nmol) added to oxygen (nmol O) consumed during phosphorylation of that amount of ADP; RC^uncoupier), the respiratory control index calculated as the ratio of the respiration rate in the presence of the uncoupler (V(lmcouplcr)) to Vim- Index (PF) means that the parameter was measured in the presence of the PF compound. The data of 5 experiments are presented as mean values S.E. PF10 Parameter mean S .R v mi 18.8 + 1.9 Vst3 84.3 3.9 V st4 18.6 + 2.6 R C I 4.778 + 0.4 ADP:0 1.844 0.1 V(PF) 27.3 2.5 Vst3(PF) 83.7 9.6 Vst4(PF) 25.8 1.9 R C I(pf) 3.256 0.3 A D P : 0 ( pf) 1.84 0.1 V(uncoupler) 111.0 10.4 RCI(uncoupler) 5.992 0.6 RCI(PF) R C I , % 71.0 10.4 A D P : 0 ( pf) : ADP:0, % 99.8 3.5 The effect o f PF10H is presented by Fig.3. and Table 2. Being added at low concentrations, this compound stim ulates the State4 respiration and decreases the RCI and A D P:0 indexes, whereas it does not affect the State 3 respiration (see Table 2). 004155 6 KBW 2/4/98 U L m inu ^ .. Fig.3. The effect o f PF10H on the respiration rate and AW o f rat liver mitochondria. The composition of incubation medium and other conditions, as in Fig. 1. Additions: PF10H, 6.25 pM; ADP, 100 pM; DNP, 40 pM 004156 7 KBW 2/4/98 T able 2. The effect o f PF10H on the respiration rates o f rat liver mitochondria under various metabolic states. The experim ental conditions and additions, as in Fig.3. Abbreviations, see Table 1. The data of 5 experiments are presented as mean values S.E. PF10H Parameter mean +S.K Vim 17.7 + 0.6 Vst3 81.1+3.8 Vst4 16.2 + 0.9 R C I 5.036 + 0.2 ADPO 1.814 + 0.1 V(PF) 30.9+1.9 Vst3(PF) 81.7 + 7.0 Vst4(PF) 23.5 + 3.4 RCI(pf) 3.77 + 0.6 ADP:0(pf) 1.602 + 0.1 V(uncoupler) 101.1 13.0 R C I(u n coupler) 5.694 0.7 RCI(PF) : R C I, % 77.6 17.0 ADP:0(pf) : ADP:0, % 88.4 + 2.5 The compounds PF143 and PF95 exerted sm all stimulatory effect on m itochondrial respiration both in State 3 and 4. PF95 also slightly increased RCI, as it is shown in Table 4. However, PF143 induced small decrease in the A'P o f mitochondria (Fig.4). The "increase" in AT* induced by PF95 (Fig.5) was not related to the membrane potential o f m itochondria because it was observed also in the absence o f mitochondria. Thus, PF95 apparently interferes w ith the TPP+ - selective electrode, w hich renders impossible the measurement of A'P by this method. The other effects o f these compounds on mitochondrial integrity are described later in this report. 8 004157 MiIto X \ ADP PFM 3 KBW 2/4/98 Fig.4. The effect ofPF143 on the respiration rate and A'F o frat liver mitochondria. The composition of incubation medium and other conditions, as in Fig.l. Additions: PF143, 100 pM; ADP, 100 pM; DNP, 40 pM C04158 9 a3 TPP Milo ADP PF95 DNP KBW 2/4/98 [O2] - 0 I\ " \ l I Fig. 5. The effect ofPF95 on the respiration rate and A Y ofrat liver mitochondria. The composition of incubation medium and other conditions, as in Fig.l. Additions: PF95,10 pM; ADP, 100 pM; DNP, 40 pM 004159 10 KBW 2/4/98 T able 3. The effect o f PF143 on the respiration rates o f rat liver mitochondria under various metabolic states. The experimental conditions and additions, as in Fig.4. Abbreviations, see Table 1. The data of 5 experiments are presented as mean values S.E. PF143 Parameter mean + S.R V ini 15.2 2.7 Vst3 80.4 5.3 Vst4 16.3+3.1 R C I 5.618 1.5 ADP:0 1.898 0.2 V(PF) 20.7 3.4 Vgt3(PF) 94.6 6.1 Vst4(PF) 17.4 2.0 RCI(pf) 5.984 0.4 ADP:0(pf) 1.858 0.1 V(uncoupler) 103.0 10.9 R C I ( u n coupler) 7.676 1.3 R C I(pf) :R C I , % 105.1 11.6 ADP:0(pf) :ADP:0, % 98.1 1.3 004160 h KBW 2/4/98 Table 4. The effect o f PF95 on the respiration rates o f rat liver mitochondria under various metabolic states. The experim ental conditions and additions, as in Fig.5. Abbreviations, see Table 1. The data of 7 experiments are presented as mean values S.E. PF95 Parameter mean S.K Vini 18.3+0.6 VstS 79.0 2.3 Vst4 18.0 0.8 R C I 4.434 0.2 A D P:0 1.799 0.1 V(PF) 21.0 + 1.5 Vst3(PF) 98.3 5.1 Vst4(PF) 18.5 1.9 RCI(pf) 5.549 0.4 ADP:0(pf) 1.819 0.1 V(uncoupler) 134.9 9.1 R C I ( u n coupler) 7.38 0.4 RCI(pf) : R C I , % 126.23 9.9 ADP:0(pf) : A D P : 0 , % 101.31 2.3 Tables 5 and 6 shows the effects o f PF12L and PF12M on the respiration rate o f m itochondria (see also F ig .l for PF12L effects). These compounds at low concentrations stim ulated both State 3 and State 4 respiration, decreased the A'F (see Fig.1), and decreased both RCI and ADPiO ratios. Thus, PF12L and PF12M affect the m itochondria in a way, which is typical for a protonophoric uncoupler like 2,4-dinitrophenol, or FCCP. As it clearly seen from Tables 1-6, different PF compounds exert different effects on the energetics o f m itochondria. Low concentrations o f PF10H, PF12L, and PF12M significantly stim ulated the rate of resting respiration of mitochondria. These features indicate that PF10H, PF12L, and PF12M are able to increase the conductivity of inner membrane o f m itochondria to protons or other ions. The suggestion further supported by the measurements o f m itochondrial A'F. All these compounds decreased the A'F o f m itochondria in parallel to stim ulation of respiration. 0041S1 12 KBW 2/4/98 Table 5. The effect o fPF12L on the respiration rates o f rat liver mitochondria under various metabolic states. The experimental conditions and additions, as in Fig.1. Abbreviations, see Table 1. The data of 3 experiments are presented as mean values S.E. PF12L Parameter mean +S.K Vfei 16.1 1.4 V s 80.3 3.0 Vst4 15.5 1.5 R C I 5.23 0.3 ADP:0 1.797 0.1 V(PF) 38.9 1.3 Vst3(PF) 87.0 3.7 Vst4(PF) 31.8 3.1 RCI(PF) 2.78 0.3 ADP:0(pf) 1.403 0.1 V (uncoupler) 95.8 10.7 R C I(u n coupler) 6.03 0.8 RCI(pf) : R C I , % 54.1 7.9 ADP:0(pf) : ADP:0, % 79.1 10.9 PF95 stim ulated the respiration in all metabolic states, the effect resem bling that of ethanol. The eflFect could be suggested to reflect a fluidization o f inner m itochondrial membrane, which increases both the proton leak and the activity o f membrane enzymes. Unfortunately, it was not possible to measure A1? changes induced by PF95 in mitochondria, due to the fact that the compound exerted a strong effect on TPP+ -electrode, both in the presence and in the absence o f mitochondria. Low concentrations of PF143 were w ithout effect on any param eter studied. However, this is the only water-soluble compound and it was possible that its partition coefficient made unfavorable the distribution of PF143 into mitochondrial membranes. Due to this we studied the effects o f higher concentrations o f PF143. Table 1 shows that 100 pM o f PF143 slightly stim ulated both the resting and State 3 respiration o f m itochondria. A t this concentration, PF143 was without effect on A'P o f m itochondria. This also could be explained by a fluidization of inner m itochondrial membrane, as in the case o f PF95. 0041S2 13 KBW 2/4/98 Table 6. The effect o fPF12M on the respiration rates o frat liver mitochondria under various metabolic states. The experimental conditions and additions, as in Fig. 1. Abbreviations, see Table 1. The data of 3 experiments are presented as mean values S.E. PF12M Parameter mean S.K V ini 15.6 1.3 Vsta 77.1 2.0 Vst4 15.3 0.4 R C I 5.037 0.2 ADP:0 1.827 0.1 V(PF) 32.6 1.4 Vst3(PF) 81.9 5.1 Vst4(PF) 27.2 1.5 RCI(pf> 3.03 0.2 ADP:0(pf) 1.56 0.1 V(uncoupler) 95.6 6.4 R C I(u n co u p ler) 6.25 0.8 RCI(pf) : R C I , % 60.1 3.6 ADP:0(pf) : ADP:0, % 85.6 4.6 Low concentrations of PF10 affected neither m itochondrial respiration nor AT'. Higher concentrations slightly stim ulated resting respiration, and induced a sm all decrease in A4?. The substances, which are able to increase the conductivity of the inner m itochondrial membrane, could be expected to uncouple oxidative phosphorylation. The uncoupling can be observed as a decrease in RCI and/or in A D P:0 ratio induced by a compound. However, the decrease in intactness of m itochondria occurring due to aging o f isolated organelles, heterogeneity o f m itochondria, and other factors decrease the reproducibility of experim ents. Our assay procedure explained above allows us to elim inate m ost of these factors. Two additions of ADP to mitochondria, first in the absence and the second in the presence o f a compound under study provide an internal standart thus allowing to compare the effects o f different compounds alm ost independently on the variations in mitochondrial preparations. Fig.6 compares the effects o f PF compounds on m itochondrial oxidative 14 004163 KBW 2/4/98 phosphorylation. The effects expressed as percent changes in RCI and A D P:0 induced by different PF compounds. 130 120 V 110 0^ 100 0wmmm 90 (fS CC 80 70 80 SO 40 30 20 10 0 ethanol PF10 HPF10H 0PF12L ADPlO (compound) ------BPI-- Fig.6. The effect o fPF compounds on the efficiency o f oxidative phosphorylation o f rat liver mitochondria. The composition of incubation medium and other conditions, as in F ig .l. For concentrations o f PF compounds, see Tables 1-6. The data shown in Fig.6 are in a good agreem ent w ith the effects o f PF compounds on resting respiration o f m itochondria and on the A'F. These data further clarify the modes o f action of PF compounds on the m itochondrial energetics. Fig.6 shows that low concentrations o f PF12L, PF12M, and PF10H decrease both the RCI and A D P:0 ratios in mitochondria. The action o f these compounds on oxidative phosphorylation resem bles that of classical uncouplers like FCCP or dinitrophenol. PF143 and PF95, as well as ethanol, increased RCI whereas they were practically without effect on ADPiO ratio in mitochondria. This does not contradict our suggestion that these compounds increase the fluidity o f m itochondrial membrane thus activating the enzymes of respiratory chain. The decrease in RCI induced by PF10 in the absence o f the effect on ADPrO ratio suggests that PF10 can increase the proton leak (which is believed to be absent or insignificant in State 3). 004164 15 KBW 2/4/98 2. The mechanisms of the effects of PF compounds on mitochondrial energetic. In the previous section we have shown that PF compounds affect the mitochondrial energetics in at least 3 different ways. Some compounds, such as PF12L, PF12M, and PFIOH were shown to decrease the degree of coupling o f ATP production to oxidation of respiratory substrates. O ther compounds (PF143, PF95) were suggested to affect the fluidity o f m itochondrial membrane, and PF10 apparently exerted an effect on the proton (ion) leak inherent to the inner m itochondrial membrane. We next made an attem pt to investigate the mechanisms involved in the effects o f PF compound on mitochondria. To study the action of PF compounds on the conductivity o f m itochondrial membrane, we took advantage o f sim ultaneous measurements o f respiration and membrane potential changes. Various concentrations o f PF compounds were added to mitochondria and respiration rates and AT changes were recorded. For a comparison, the same experim ents were perform ed w ith a classical uncoupler -protonophore 2,4-dinitrophenol. A typical record of changes in respiration and A T induced by the addition o f 2,4-DNP is shown on Fig.3. Fig.3 The effect o f various 2,4-DNP concentrations on the rate o f oxygen consumption and A Pin rat liver mitochondria. Incubation medium (see Fig. 1) was supplemented with 2 pg/ml oligomycine. All other conditions were as in Fig. 1. The concentrations of 2,4-dinitrophenol (DNP) were as following: 5 pM, 5 pM, 10 pM, 20 pM (40 pM total). 004165 16 KBW 2/4/98 Fig. 4 shows the changes in respiration rate o f m itochondria plotted against changes in A'F. Fig.4 The relationships between mitochondrial membrane potential and the rate o f oxygen consumption in the presence o fvarious concentrations ofP F compounds. Incubation medium (see Fig.l) was supplemented with 2 pg/ml oligomycine. All other conditions were as in Fig.l. The respiration and membrane potential of mitochondria were sequentially titrated by additions of increasing concentrations of the PF compounds. The concentrations were as following: PF10, 0, 62.5, 125, 187.5,250 jjM; PF10H, 0, 6.25, 12.5,25, 50 pM; PF12L, 0, 2.08, 4.16, 8.32, 16.64 pM; PF12L, 0, 2.08, 4.16, 8.32, 16.64 pM; PF143, 0, 100, 200, 300,400 pM; DNP, 0,2.22,4.44, 8.88,17.76, 35.52 pM. These experim ents revealed that for low concentrations o f PF12M, PF12L, and PF10H the conductivity o f m itochondrial membrane (represented by the rate o f respiration) relates linearly to the membrane potential in a range o f concentrations o f the compounds. The same relationships were shown for a classical protonophore 2,4DNP. This indicates that the mechanism o f increase in the inner membrane conductivity involves proton shuttling as it is firm ly established for a representative uncoupler 2,4-dinitrophenol. However, another kind o f experim ents, in particular the direct measurements o f conductivity w ith artificial bilayer membranes, are necessary to prove this hypothesis. C041S6 17 KBW 2/4/98 PF10 (Fig.5) slightly stim ulates the respiration with a correspondent decrease in AT at relatively high concentration, however further increases in concentration o f this compound produce no further effect. It correlates w ell with our proposal that PF10 increases the proton leak in m itochondrial membranes. The leak is membrane potential dependent in such a way that slight decreases in A T can significantly suppress the leak (Nicholls DG, 1974). Mito / Mito > J PF10 vl k .VI' increase PF;io p f i o il l X PF10 Ni PFIO PFIO k1 .1PF10 V PF10 \i \ DNP l [Oz]-D Vt DNP t 1 o . 1 min , Fig.5. The effect o fPF10 on the respiration and A W. All conditions were as in Fig. 3. Each addition of PF10 was 62.5 nM 004167 18 KBW 2/4/98 PF143 (Fig.6) in the concentration range 100-400 (J.M was practically without effect on the respiration rate o f mitochondria. However, as it clearly seen in Fig.6, the first addition o f 100 pM PF143 iduced a slow decrease in A1? of m itochondria which continued to decline, being apparently not affected by further additions of PF143. .In the previous section we propose that this compound affects the fluidity o f m itochondrial membrane. Due to this it can be expected that PF143 m ight act as a detergent at higher concentrations. W hen higher concentrations (up to 1 mM) o f the compound were added to mitochondria, we indeed observed a transient stim ulation of respiration followed by inhibition of oxygen consumption, which was insensitive to 2,4-DNP (data not shown). PF143 Mito .II, PF143 I PFI43 PF143 E0d-0 Mito / 1 min \ \ v Fig.6. The effect ofPF143 on the respiration and A T/ . All conditions were as in Fig. 3. Each addition of PF143 was 100 jiM, the addition of DNP was 40 jjM. 0041S8 19 KBW 2/4/98 PF10H (Fig.7) at concentrations higher then 25 pM strongly inhibited the respiration o f m itochondria. These experim ents were repeated with a low ionic strength medium and with the use of another respiratory substrate, succinate. Under these conditions, 25-50 pM PF10H also inhibited the respiration. This indicates that the site o f inhibition is located in the region o f ubiquinone:cytochrom e c reductase, which is typical for artificial uncouplers, or it inhibits cytochrome c oxidase. However, it was noted that the addition of 5 pM cytochrome c to m itochondria inhibited by PF10H partially restored the respiration (Fig.8). It is well known that in high ionic strength medium, swelling of mitochondria induces the release o f cytochrome c and the inhibition o f respiration. Mito \ P F 10H P F 10H I\ \\ | P F 10H wj P F 10H A`l' increase Mito Fig.7. The effect o fPF1 OHon the respiration and A tF. All conditions were as in Fig. 3. The additions ofPF10H were 6.25,6.25,11.5 pM (25 pM total). 20 004169 KBW 2/4/98 This points to a possibility that the apparent inhibition o f respiration by PF10H is due to this compound inducing the Ca2+ -dependent perm eability transition o f m itochondrial inner membrane, or so-called pore opening. This would results in high am plitude swelling o f m itochondria and in loss of cytochrome c. To clarify the effect, we investigated the action o f PF10H on m itochondria using different respiratory substrates and the low ionic PF10H sj DNP C yt.c Fig.8. The effect ofPFlOH on the respiration o frat liver mitochondria. All conditions as in Fig. 3. Additions: PF10H, 50 (iM; DNP, 40 |iM; Cyt c, 5 ^M. Vx. strength incubation medium which is known to decrease the probability o f pore opening. In these experiments, we used the incubation medium containing 210 mM m annitol, 10 mM sucrose, 5mM HEPES-Tris, pH 7.4, and 2 mM MgCfy w ith or without 2.4 mM EGTA (indicated in Figure Legends) to chelate the residual Ca2+. Glutamate and m alate were used as respiratory substrates in the experim ent shown on Fig.8, which im plies that the inhibition o f respiration by PF10H is not due to it inhibiting the succinate dehydrogenase. Indeed, there was no inhibition of respiration even when the concentration o f PF10H was 100 (oM (Fig.9). However, 50 (iM PF10H caused a strong progressive inhibition o f the succinate-supported respiration o f m itochondria and com pletely discharged the AT if rotenone was excluded from the incubation medium (Fig. 10). This is typical for a protonophorous uncoupler, the inhibition explained by the accum ulation of oxaloacetic acid (a strong inhibitor o f succinate dehydrogenase) in the m itochondial matrix. 21 004170 KBW 2/4/98 Fig.9. The effect ofPFlOH on the respiration and AW o f rat liver mitochondria respiring on succinate. The incubation medium contained 210 mM m annitol, 10 mM sucrose, 5mM HEPES-Tris, pH 7.4, 2 mM MgCl2, 2.4 mM EGTA, 5 mM succinate, 2 pg/ml oligomycine, and 2 pM rotenone. The concentration of mitochondria was 1 mg/ml. Additions: PF10H, 50 pM. 004171 22 Mito \ PFtOH x jl KBW 2/4/98 <?* Oc o 1 min m A4' increase DNP Mito \ \\ V Fig. 10. The inhibition o f respiration and the decrease in AW induced by PF10H in the absence o frotenone. The incubation medium and other conditions were as in Fig.9 for except that rotenone was omitted. Additions: PF1OH, 50 pM, DNP, 40 pM. ' These experim ents allow us to rule out the possibility that PF10H inhibits Complex III o f m itochondrial respiratory chain. However, the cytochrome c - reversible inhibition o f respiration may also occure due to the displacem ent o f cytochrome from the m itochondrial membrane (acetyl-ammonium -lik e effect, the negative charge screening by amphiphilic positively charged compounds) or due to competitive inhibition o f cytochrom e binding to term inal oxidase (a polylysine -lik e effect). To further clarify this, we took advantage o f a classical reducing non-enzymatic system ascorbate + TMPD to reduce m itochondrial cytochrome c. This system is very sensitive to any changes in cytochrome c binding and/or interaction with cytochrome c oxidase. F ig.l 1 shows that 50 mM PF10H decreases the A'F and stim ulates the respiration o f m itochondria oxidising ascorbate. There is no spontaneous inhibition of respiration (as it would be observed if PF10H interferes w ith cytochrom e c binding). These data further prove that PF10H does not inhibit the respiratory chain per se. 23 004172 KBW 2/4/98 Fig. 11. The effect o f PF10H on the respiration and A 'F o f rat liver mitochondria respiring on ascorbate + TMPD. The incubation medium contained 210 mM mannitol, 10 mM sucrose, 5mM HEPES-Tris, pH 7.4, 2 mM MgCl2, 2.4 mM EGTA, 5 mM ascorbate, 100 pM TMPD, 2 pg/m l oligomycin, and 2 pM rotenone. The concentration of mitochondria was 1 mg/ml. Additions: PF10H, 50 pM. The effects o f PF compounds were further studied in experim ents aim ed to reveal the action o f these compounds on m itochondrial integrity. For this, we investigated the effects o f high concentrations o f PF12L, PF12M, PF10, PF95, PF143, and PF10H on the swelling o f m itochondria under various conditions. The following data show that the compounds induce the swelling and disruption of m itochondria by different mechanisms. The swelling o f mitochondria was m easured in high and low ionic strength medium w ith or without respiratory substrates and both in the presence or in the absence o f a Ca2^ chelator EGTA. Fig. 12 shows, that the patterns o f PF10H -induced swelling were different under different conditions. In the presence o f EGTA (Fig. 12, upper curve) or succinate (data not shown) PF10H was less efficient than in the absence of these substances. The efficiency o f PF10H was independent on the ionic strength o f the incubation medium (data not shown). It is clearly seen also, that the presence o f EGTA in the incubation medium significantly suppressed the high-am plitude 24 004173 KBW 2/4/98 sw elling induced by 25 pM PF10H, and changed its kinetics. This indicates that high concentrations of PF10H, w hich were shown to inhibit the respiration o f m itochondria can induce Ca2+ -dependent pore opening in the inner m itochondrial membrane. It is well known that the opening of such a pore results in uncoupling, sw elling and eventual disruption of mitochondria. Time, min Fig. 12. The effect o fPF10H on the swelling o f mitochondria Incubation medium was as in F ig .l for except that glutam ate and m alate were excluded. Upper curve, the incubation medium was supplem ented w ith 1 mM EGTA. M itochondrial protein was 1 mg x ml-1. Each addition o f PF10H was 25 pM. In contrast to the action o f PF10H, the swelling o f m itochondria induced by high concentrations o f PF143 was not affected by the presence o f EGTA or respiratory substrates. However, the effect o f PF143 was strongly dependent on the ionic strength o f the incubation medium. Fig. 13 shows, that in a low potassium medium, concentrations of PF143 as high as about 4 mM were w ithout effect on the swelling o f mitochondria. For a comparison, the effect o f the strong non-ionic detergent Triton X-100 on rat liver m itochondria is shown (Fig.13). In a potassium chloride medium, PF143 appeared to be a strong detergent w ith the efficiency o f about 1/3 relative to that o f Triton. All other PF compounds (PF10, PF12L, PF12M, and PF95) were without effect on the sw elling of m itochondria. 004174 25 Absorbance KBW 2/4/98 [Compound], j.ig/mg protein Fig. 13. The effect o f PF143 on the swelling o fmitochondria. High ionic strength medium was as in Fig.4, low er curve. The mannitol -containing medium was 210 mM mannitol, 10 mM sucrose, 4 mM KH2P 0 4>10 mM MOPS, pH 7.4. M itochondrial were added at 1 m g x m l'1. 26 004175 KBW 4/24/98 Fig.S-1. Effect o f "PF95M" on the rate o f respiration and membrane potential o f rat liver mitochondria. Medium composition and other conditions were as in Fig.l. Red curve, the respiration of mitochondria; blue curve, the changes in membrane potential. Additions: TPP+or 0.2, 0.2, 0.4, 0.8, and 0.2 pM TPP+C1" (2 pM, total); Mito, 1 mg/ml rat liver mitochondria; ADP, 200 pM ADP; PF95M, 0.5 pM "PF95M"; DNP, 40 pM 2,4dinitrophenol. 004178 2 3. Conclusions KBW 2/4/98 This study reveals that PF compounds affect various aspects o f m itochondrial energetics. Some of compounds, like PF12L, PF12M, PF10H, at relatively low concentrations decrease the degree o f coupling of respiratory substrate oxidation to ATP production. The m olecular mechanism o f the action o f PF12L and PF12M compounds apparently involves the shuttling o f protons across inner m itochondrial membrane, although further experim ents are needed. The mechanism of action o f PF10H apparently involves the non-specific changes in the perm eability o f the inner m itochondrial membrane. M ost probably, this compound can change the m icrodom ain structure o f m itochondrial lipid membranes (by inducing lipid clustering or by changing lipid phase state from a bilayer to hexagonal). The effects o f PF143 and PF95 are small but significant. These compounds slightly increase the activity o f enzymes o f oxidative phosphorylation and respiratory chain, the most probable m echanism being the fluidisation o f the inner m itochondrial membrane, ft m ight be expected that the accum ulation o f such compounds in a tissue could results in tissue damage. The data obtained w ith PF10 are the most interesting from a bioenergetic point o f view. If our interpretaion is correct and this compound increases the activity o f the pathway(s) o f proton leak in m itochondrial membrane, it may be a very useful tool to study flow /force relationships o f various energy-dependent processes in mitochondria. 4.Literature cited. Kamo N, M uratsugu M, Hongoh R, Kobatake Y "M embrane potential o f m itochondria m easured with an electrode sensitive to tetraphenyl phosphonium and relationship between proton electrochem ical potential and phosphorylation potential in steady state." J M embr Biol 1979 Aug;49(2): 105-121 N icholls DG "The non-Ohmic proton leak--25 years on." Biosci Rep 1997 Jun;17(3):251-257 Rottenberg H "M embrane potential and surface potential in mitochondria: uptake and binding of lipophilic cations." J M embr Biol 1984;81(2):127-138 27 004176 KBW 4/24/98 Supplement 1. Effects of "PF95M" and "Sal" on mitochondrial energetics. All the conditions, procedures, and the logics of experiments were as described in "Materials and Methods". Being added at relatively low concentrations, "PF95M" stimulates the rate of respiration and decreases membrane potential in rat liver mitochondria (Fig.S-1). Other compound, "Sal", exerts the same effects but at much higher concentration (Fig.S-2). Both these compounds decrease RCI and ADP:0 ratio of mitochondria (Fig.S-3). For these experiments, the unequal concentrations of "PF95M" and "Sal" were chosen which approximately double (that means the increase by 100 %) the State 4 respiration rate of mitochondria (see Fig.S-1 and Fig.S-2). An average RCI in these experiments was about 4.16 (416 % increase in respiration rate under addition of ADP) so 100 % increase induced by "PF95M" or "Sal" could not mask the State 4 - State 3 - State 4 transition and RCI and ADP:0 estimation. At these concentrations, both "PF95M" and "Sal" decreased RCI (by ~ 40% and ~ 30%, respectively, comparing to abs. ethanol control incubation) and ADP:0 ratio (by ~ 12 % and ~ 10 %, respectively, comparing to ethanol control incubation). Thus, both these compounds are uncouplers of oxidative phosphorylation, although of different efficiency, "PF95M" being about four hundred times more strong, then "Sal". These experiments were repeated with the use of low ionic strenght medium (225 mM mannitol, 5 mM Hepes (pH 7.4), 4 mM KH2PO4, 5 mM glutamate, and 5 mM malate). With the use of this medium, qualitatively the same results were obtained, however the concentrations doubling the respiration were 1 pM for "PF95M", and 400 pM for "Sal" (data not shown). In order to reveal the putative mechanism of uncoupling, we compared the effects of different concentrations of `TF95M" and "Sal" on mitochondrial respiration and membrane potential with these of a "classical" uncoupler 2,4-dinitrophenol. The changes in respiration rate and in membrane potential were recorded simultaneously. A typical record example is shown by Fig.S-4. 1 004177 KBW 4/24/98 Fig.S-2. Effect of "Sal" on the rate of respiration and membrane potential of rat liver mitochondria. Medium composition and other conditions were as in Fig. 1. Red curve, the respiration of mitochondria; blue curve, the changes in membrane potential. Additions: TPP+or 0.2, 0.2, 0.4, 0.8, and 0.2 pM TPP+C1' (2 pM , total); M ito, 1 mg/ml rat liver m itochondria; ADP, 200 pM ADP; Sal, 200 pM "Sal"; DNP, 40 pM 2,4-dinitrophenol. 004179 3 KBW 4/24/98 Fig.S-3. Effects of low concentrations of "PF95M" and "Sal" on the degree of coupling and efficiency of oxidative phosphorylation in rat liver mitochondria. Respiratory control indexes (RCI) and ADP:0 ratios before and after addition of "PF95M" or "Sal" to mitochondrial suspension were measured as described in "Materials and Methods" (see Fig.S-1, Fig.S-2). Each column represents averaged data from 4 experiments and error bars show S.E. See text for further explanations. 0041S0 4 KBW 4/24/98 Fig.S-4. Typical record of changes in respiration rate and membrane potential of mitochondria induced by sequential additions of "PF95M". Medium composition and other conditions were as in Fig.l. Red curve, the respiration of mitochondria; blue curve, the changes in membrane potential. Additions: TPP+or 0.2, 0.2, 0.4, 0.8, and 0.2 jjM TPP+C1' (2 pM, total); Mito, 1 mg/ml rat liver mitochondria; PF95M, 0.5, 1, 1, 1,2, and 2 pM (7.5 pM total) "PF95M". 5 C041S1 KBW 4/24/98 The values of membrane potential and the rates of respiration were calculated as described in "Material and Methods". Fig.S-5 shows the effect of different concentrations of all three compounds ("PF95M", "Sal", and dinitrophenol) on mitochondrial respiration and Fig.S-6 show the changes in membrane potential plotted against changes in Fig.S-5. The increase in respiration rate of rat liver mitochondria induced by "PF95M", "Sal", and 2,4-dinitrophenol (DNP). All the conditions were as in Fig.S-4. Note that the concentration of "Sal" is expressed as pM x 100 (O"4M). respiration. Fig.S-5 shows the different uncoupling efficiency of the compounds, "PF95M" and "Sal" being the most and the less potent then "classical" dinitrophenol, respectively. Fig.S-6 shows that all these compounds at applied concentrations do not inhibit the respiratory chain of mitochondria. 004132 6 KBW 4/24/98 Fig.S-6. Changes in membrane potential of mitochondria plotted against changes in respiration rate induced by different concentrations of "PF95M", "Sal", and 2-4dinitrophenol (DNP). Conclusion. Our experiments revealed that the compound named "PF95M" is a very potent uncoupler of oxidative phosporylation in liver mitochondria. The uncoupling efficiency of this compound is comparable with that of CCCP, one of potent "classical" uncouplers. Another compound, "Sal", also appeared to be an uncoupler, although of relatively low efficiency. The relationship between changes in membrane potential and increase in the rate of respiration of mitochondria allows us to propose the increase in proton permeability of inner mitochondrial membrane as the mechanism of uncoupling action of these compounds. Further experiments are needed to elucidate the mechanism o f the uncoupling at the molecular level. 0041S3 7