Chem. -Biol. Zntemctions, 12 (1976) 71-79 0 Elsevier Scientific Publishing Company, Amsterdam
- Printed in The Netherlands
RESPIRATORY CONTROL DEPRESSION BY TETRAALKYLAMMONIUM BROMIDES IN RAT LIVER MITOCHONDRIA
KENNETH S. ROGERS and EDWIN S. HIGGINS Department of Biochemistry, Medical College of Virginia, Virginia Commonwealth University, Richmond, Va. 23298
(U.S.A.)
(Received April 24th, 1975) (Revision received July 14th, 1975) (Accepted August 7th, 1975)
SUMMARY
Six different lipophilic (hydrophobic) organic cations, tetraethyl-, tetrapropyl, tetrabutyl-, tetrapentyl-, tetrahexyl-, and tetraheptylanimonium bromide, depressed respiratory control in rat liver mitochondria. Evaluation of mitochondrial responses in terms of a quadratic equation in log 1) (an index of lipophilicity) indicated that the NADH dehydrogenase receptor site for inhibitor (diminution of control of glutamate, cu-ketoglutarate, and /3-hydroxybutyrate respiration) was more lipophilic than receptor sites for flavin-linked substrates (reduction of control of succinate, choline and CXglycerophosphate respiration). The succinate dehydrogenase receptor site for inhibition by the tetraalkylammonium bromides was more hydrophilic (less lipophilic) than the choline or a-glycerophosphate dehydrogenase receptor sites. Depression of respiratory control may be a function of charge density and of lipophilicity at specific inner membranal sites and the susceptible site may differ for different respiratory substrates.
INTRODUCTION
Analysis [l] of the interaction of symmetrical tetraalkylammonium bromides with the inner membrane of mitochondria during depression of respiratory control indicated that the NADH dehydrogenase receptor site (diminution of control of glutamate respiration) was more lipophilic (hydrophobic) than the succinate dehydrogenase receptor site (reduction of control of succinate respiration). Depression of respiratory control occurred by two different mechanisms depending upon inhibitor concentrations (a) inhibition
71
of phosphorylation at low concentrations of the alkyi bromide and (b) uncoupling of oxidative phosphorylation at higher concentrations of the alkyl bromide. An alternative explanation for the loss of respiratory control during glutamate oxidation might involve interference with transfer of reducing equivalents from initial substrate to the NADH dehydrogenase site by the symmetrical alkyl bromide rather than lipophilic perturbation of the inner membrane constituents. It is the purpose of this work to resolve the contingency of the two possible explanations for depression of respiratory control during glutamate respiration and provide additional information about the lipophilic character of the mitochondrial inner membrane. The following substrates for different mitochondrial receptor activities were used: L-glutamate, succinate, flhydroxybutyrate, or-ketoglutarate, choline and a-glycerophosphate. MATERIALS
AND METHODS
Hepatic mitochondria were prepared from Sprague-Dawley rats as described previously [l] . Respiratory rates were determined polarographically [l] and the conventions of Chance and Williams [Z] were used in identifying respiratory states and in computing respiratory control ratios (RCR). Mitochondrial protein was estimated in the presence of 1% deoxycholate by a biuret method [3] with crystalline bovine serum albumin as standard. Tetraethyl-, tetrapropyl-, tetrabutyl-, tetrapentyl-, tetrahexyl-, and tetrah8eptylammonium bromide were obtained from Eastman Organic Chemicals. Sucrose, EDTA and sodium deoxycholate were from Fisher Scientific Co., and mannitol was from Pfanstiehl Laboratories. Adenosine diphosphate, substrates and other biochemicals were from Sigma Chemical Company. Estimation of membrane Lipophilicity Logarithms of the tetraalkylammonium bromides’ partition coefficients [l] for the distribution of the chemical between the nonpolar : polar phases 1-octanol and water were used in a quadratic equation to estimate the hypothetical lipophilicity for a portion of the mitochondrial inner membrane. Thus pl,,
= a log P - b(log P)’ + c
where Iso is the concentration of chemical responsible for depression of respiratory control by 5096, p is the negative logarithm, and P is the partition coefficient of the chemical. A hypothetical value of logp may be obtained that indicates membrane lipophilicity [l] by taking the first derivative of the quadratic equation and setting the quantity (d log Z50 /d log P) equal to zero. Success of the quadratic equation in characterizing the data was evaluated statistically using the method of least squares [4] in a multiple linear regression model.
72
Fig. 1 summarizes the interference of our model orgwic cation, the symmetrical tet~butyl~onium bromide, with phosphate acceptor control of the mitochondrial oxidation of a-ketoglutarate, &hydroxybutyrate, choline, and cr-glycerophosphate. The effectiveness of the alkyl bromide as an inhibitor of respiratory control depended upon whether reducing equivalents from substrate were linked via a py~dinopro~~ to ~~~dehy~ogen~e (Fig. 1, graph A) or directly to a different mitochondrial flavoprotein (Fig. 1, graph B). A concentration of about 25 @l tetrabutylammonium bromide depressed respiratory control 20-30% during rw-ketoglutarate or P-
Fig. 1. Influences of tstrabutylammonium bromide upon mitochnndrial respiratory control during a-ketoglutarate, fl-hydroxybutyrate, choline and &glycerophosphate oxidations. Oxygen consumption (ng atoms of oxygen per min per mg of mitochondrial protein) was monitored at 30* by a Clark fixed voltage electrode. The reaction mixture of 3 ml contained 0.33 M mannitol; 3.5 mM potassium phosphate (pH 7.4f; 3.5 mM potassium chloride; 0.33 mM disodium ethylenediamine tetraacetic acid; 4 mg of dialyzed bovine serum albumin; and approx. 2.5 mg of mitochondrial protein as intact rat liver mitochondria. Substrates were added in a volume of 20 ~1 to give a final concentration of 1.4 m&I (~-ke~glutarate and choline) or 2.8 mM (DL~-hydroxybutyrate and DL-Q~ycerophosphate) and ADP was added in a volume of 30 ~1 to give a final concentration of 0.13 mM. Respiratory cDntro1 was measured as the ratio of velocity of oxygen consumption during oxidation of substrate in the presence of phosphate acceptor (state 3) to that obtained after exhaust.ion of ADP (state 4). The mean from at least four different mitochondrial preparations is presented with standard errors of the mean in brackets as a single data point for a given conc:ent~tion of tetrabutylammonium bromide. 92~,130,140 and Iso values refer to the respective concentrations of quaternary ammonium salt that diminished respiratory control by 20, 30, 40 and 50%, respectively. Graph A represents ) and @-hydroxybutyrate ( ). data from the substrates CYketoglutarate (W and &glyceropbosphate (o-f. Graph B represents the substrates choline (ci_)
73
hydroxybutyrate oxidations whereas a higher inhibitor concentration uf about 0.1 mM was required to produce the same extent of respiratory control depression during choline or a-glycerophosphate oxidation. Similar findings have been documented for the mitochondrial oxidations of glutamate and succinate in the presence of tetraalkylammonium bromides [l].The two concentration-dependent mechanisms of respiratory control depression (see INTRODUCTION) were not altered here (data not presented) by the type of substrate used for mitochondrial oxidation, i.e., /3-hydroxybutyrate, cr-ketoglutarate, glutamate [ 1 ] , choline, ar-glycerophosphate, or succinate 111. Estimatbn of NADH dehydrogenase, choline dehydrogenase, and a-glycerophosphate dehydrogenase lipophilicity The effects of tetraethyl-, tetrapropyl-, tetmbutyl-, tetrapentyl-, tetrahexyl-, and teiraheptylammonium bromides on the respiratory properties of rat liver mitochondria are presented together with partition coefficient logarithms of the halides in Tables I-IV. At each level of inhibition (20,30,4-O and 50%), the biological data were satisfactorily described by the equations of lipophilicity given in Tables II and IV. Thus, smaller amounts of tetraheptylammonium bromide were required to produce 20% depression of respiratory control than were needed with tetraethylammonium bromide; e.g. in ar-ketoglutarate oxidation (Table I) compare p12,, of 5.77 (1.7 PM) with pIXo of 1.15 (0.07 M). The data were satisfactorily correlated by the quadratic Equations 1 through 16 given in Tables II and IV. Calculated values of pIso from Equations 4 (ar-ketoglutarate respiration), 8 (P-hydroxybutyrate respiration), 12 (choline respiration) and 16 (a!-glycerophosphate respiration) (Tables II and IV) may be compared with the corresponding experimental values in Tables I and III. Predictability was good for inhibition during w ketoglutarate and choline respirations and it was better for inhibition during P-hydroxybutyrate and a -glycerophosphate respirations. Log PO values were calculated from the first derivative of Equations 1 through 16 in Tables II and IV. Regardless of the extent of inhibition produced by the homologous series of tetraalkylammonium bromides, inhibition of respiratory control during a-ketoglutarate respiration provided a log PO value of 2.10 * 0.04 (mean ?1 standard error of mean) and depression of respiratory control during /3-hydroxybutyrate respiration provided a log p value of 2.14 c 0.12. These values may be compared with the other NADH dehydrogenase-linked substrate, glutamate [l] , which has a log PO value of 2.02 + 0.06. Since log P values correlate with membrane lipophilicity [l] at a localized point for inner membrane receptor-tetraalkylammonium bromide intemction, then the observation that the different NADH dehydrogenase linked substrates gave the same calculated log PO value (the three values were not different at the 0.4 level of probability) showed that electron transfer via the individual initial dehydrogenases to the NADH dehydrogenase site was not affected by the quaternary alkylammonium compounds. Thus, the ability to relaite by equations the depression of respiratory control
74
TABLE I DEPRESSION OF MITOCHONDRIAL RESPIRATORY CONTROL DURING ar-KETOGLUTARATE AND &HYDROXYBUTYRATE OXIDATIONS BY SIX LB’OPHILIC TETRAALKYLAMMONIUM BROMIDES The effects of six symmetrical quaternary amines on a general mitochondrial property are summarized. Negative logarithmic values of Z21h Z~O,Z~O~~~Z~O(~Z~O~PZ~O,PZ~O~~~ pZ50, respectively) were obtained from data plof s similar to those given in Fig. 1. Log P values, an index of lipophilicity, were obtained fron calculations given in Table I of ref. 1. Calculated values of pZ,reo were obtained from c*quations 4 and 8 in Table II. The ratio (~2150 calc.IpZ50 obs.) expressed the correlation between calculated and observed values. Compound
log P
PZ20
PZ30
PZ40
PZ50
PZ50
ca1c.l pZ50 obs.
Q-Ketoglutarate
Tetraethylammonium_ bromide Tetrapropylammonium bromide Tetrabutylammonium bromide Tetrapentylammonium bromide Tetrahexylammonium bromide Tetraheptylammonium bromide
~150
respiration
-2.82
1.15
1.00
0.85
0.77
0.76
9.987
-1.78
2.89
2.74
2.51
2.34
2.46
1.051
-0.62
4.65
4.49
4.37
4.12
3.88
3.942
0.54
5.10
4.92
4.82
4.72
4.78
1.013
1.70
5.40
3.26
5.12
5.02
5.18
1.032
2.86
5.77
5.60
5.44
5.15
5.06
0.983
@-Hydroxybutyrate
Tetraethylammonium bromide Tetrapropylammonium bromide Tetrabutylammonium bromide Tetrapentylammonium bromide Tetrahexylammonium bromide Tetraheptylammonium bromide
Ratio:
talc.
respiration
-2.82
1.07
0.92
0.82
0.74
0.69
0.932
-1.78
2.78
2.60
2.40
2.18
2.26
1.037
-0.62
4.52
4.35
4.00
3.56
3.61
1.014
0.54
5.10
4.92
4.80
4.70
4.54
0.966
1.70
5.35
5.19
5.10
4.96
5.04
1.016
2.86
5.52
5.42
5.28
5.12
5.11
0.998
to the increase lipophilicity of inhibitory organic cations was not affected by the substrate used for providing the reducing equivalents, i.e. glutamate [ 11, a-ketoglutarate or P-hydroxybutyrate. Examination of the log PO values for choline and ar-glycerophosphate, 2.87 +- 0.18 and 3.27 + 0.24, respectively, indicated that the receptor sites for the symmetrical tetraalkylammonium bromides were different in character from the receptor site of NADH dehydrogenase. The difference in the log
75
TABLE II DIFFERENTIAL INHIBITION OF MITOCHONDRIAL TETRAALKYLAMMONIUM BROMIDES
RESPIRATORY
CONTROL
BY
pl=alogP-b(logP)“+c Inhibition
aa
b
%
a-Ketoglutamte
20 30 40 50
0.78 0.78 0.78 0.76
C
0.76 0.77 0.78 0.79
0.18 0.18 0.19 0.19
0.20 0.19 0.18 0.16
n
r
F
S.E.
Equation No.
2.17 2.17 2.05 2.00
6 6 6 6
0.993 0.993 0.993 0.996
101 104 104 209
0.28 0.27 0.18 0.19
(1) (2) (3) (4)
6 6 6 6
0.996 0.996 0.999 0.999
183 166 so7 524
0.21 0.22 0.12 0.12
(5) (6) (7) (8)
respiration
/.%Hydroxybutyrate 20 30 40 50
logP
4.87 4.71 4.58 4.42
respiration 4.82 4.64 4.42 4.16
1.90 2.03 2.17 2.47
a a, b and c are regression coefficients of the multiple least squares regression equation. The number of compounds is represented by n; r is the multiple correlation coefficient; the F value describes the significance of the equation; and S.E. is the standard error of the pf estimated by the equation. Log PO is calculated by taking the first derivative of the equation and setting (d pZ/d log P) equal to zero.
P for choline and for cr-glycerophosphate was not unreasonable since the reducing equivalents from these substrates were: transferred to two different mitochondrial flavoproteins. A log PO value of 4.08 f 0.04 has been recorded [l] for the interaction of the symm.etrical tetraalkylammonium bromides with another flavoprotein, succinate dehydrogenase. Because a low value for log P” reflected interaction of inhibitor with a high lipophilic (hydrophobic) site* and a high value for log P* reflected interaction of inhibitor with a low lipophiiic (hydrophilic) site [ l] then an order of lipophilicity has been observed for the interactions of the tetraalkylammonium bromides with the inner membrane of mitochondria. Thus, the NADH dehydrogenase site was more lipophilic than those of choline dehydrogenase and of cu-glycerophosphate dehydrogenase, and the latter two flavoproteins were more lipophilic than succinate dehydrogenase. From these results, we conclude that there are discrete differences in the hydrophobic character of the inner membrane of mitochondria as detected with use of the symmetrical quatemary alkylammonium bromides. l Other possible interpretations. such as steric hindrance, conformational distortion of active site, etc. for th\dparabolic-response of log P with biological activity have been given by C. Hansch and J.M. Clayton, Lipophilic character and biological activity of drugs, II. The parabolic case, J. Pharm. Sci., 62 (1973) 1-21.
76
TABLE III DEPRESSION OF MITOCHONDRIAL RESPIRATORY CONTROL DURING CHOLINE AND a-GLYCEROPHOSPHATE OXIDATIONS BY SIX LIPOPHILIC TETRAALKYLAMMONIUM BROMIDES The effects of six symmetrical quaternary amines on a general mitochondrial property are summarized. Negative logarithmic value8 of I 2o,Z30, ZMI and 150@Z20. ~130, ~140 and PZm) were obtained from data plots similar to Fig. 1. Calculated values of ~150 were obtained from equations 12 and 16 in Table IV. The ratio (p1~ calc./pZs obs.) expressed the correlation between calculated Compound
and observed values. logP
PZ20
PZ30
PZ40
PZ50
talc.
Ratio: pZs0 ca1c.l ~15~ obs.
PZ50
Choline respiration Tetraethylammonium bromide Tetrapropylammonium bromide Tetrabutylammonium bromide Tetrapentylammonium bromide Tetrahexylammonium bromide Tetraheptylammonium
-2.82
1.19
0.94
0.79
0.65
0.84
1.292
-1.78
3.15
2.89
2.66
2.55
2.22
0.871
-0.62
4.40
4.10
3.78
3.45
3.46
1.003
0.54
4.85
4.66
4.38
4.22
4.40
1.043
1.70
5.36
5.26
5.12
4.99
5.01
1.004
2.86
5.85
5.64
5.49
5.38
5.31
0.987
&Glycerophosphole Tetraethylammonium bromide Tetrapropylammonium bromide Tetrabutylammonium bromide Tetrapentylammonium bromide Tetrahexylammonium bromide Tetraheptylammonium bromide
respiration
-2.82
0.97
0.83
0.73
0.60
0.71
1.183
-1.78
2.66
2.50
2.39
2.29
2.06
0.900
-0.62
4.12
3.86
3.38
3.20
3.29
1.028
0.54
4.73
4.48
4.25
4.12
4.23
1.027
1.70
5.60
5.38
5.15
4.95
4.88
0.986
2.86
5.72
5.52
5.44
5.24
5.24
1.000
DISCUSSION
The tetraalkylammonium salts interact with the mitochondrial inner membrane per se and their effectiveness varies with the specific respiratory chain involved. Log PO was the same for all substrates requiring the NADH oxidase respiratory chain but log P’ was significantly different, indicating less lipophilic character, for substrates linked through various other flavo-
77
TABLE IV DIFFERENTIAL INHIBITION OF MITOCHONDRIAL TETRAALKYLAMMONIUM’ BROMIDES
RESPIRATORY
CONTROL
BY
pl=alogP-b(logP)2+c Inhibition
aa
b
%
Choline respiration
20 30 40 50
0.76 0.78 0.79 0.79
20 30 40 50
o-Glycerophosphate 0.83 0.15 0.82 0.14 0.82 0.11 0.80 0.11
0.15 0.15 0.13 0.12
C
IogpO
n
r
F
S.E.
Equation No.
4.71 4.49 4.22 4.00
2.53 2.60 3.04 3.29
6 6 6 6
0.990 0.994 0.995 0.994
76 116 136 126
0.31 0.26 0.24 0.24
(9) (10) (11) (12)
respiration 2.77 4.55 2.93 4.30 3.73 3.98 3.64 3.83
6 6 6 6
0.998 0.997 0.997 0.997
309
0.17 0.17 0.16 0.17
(13)
291 295 255
a a, b and c are regression coefficients of the multiple least squares regression Descriptions of log PO, n, r, F and S.E. are given in Table II.
I::; (16) equation.
proteins to a respiratory chain. Therefore, it is proposed that the tetraalkylammonium salts possess relative site selectivity in regard to respiratory control depression. It is of interest that phosphorylation sites I and III are more sensitive to dinitralphenol, for example, than is site II [5] . Hence, one might reasonably expect more pronounced depression of respiratory control with substrates whose electrons have an obligatory route across site I (glutamate, cw-ketoglutarate, P-hydroxybutyrate) compared to those whose entry is subsequent to site I (succinate, a-glycerophosphate, choline). Also, certain drugs have been shown in Pseudomonas saccharophika extracts to inhibit NADH oxidative phosphorylation and ATP-linked reverse electron transport (succinate to NAD’), while succinate oxidative phosphorylation was not depressed by these agents [6] . Bakker et al. [7] recently reported wide variances in the fraction of uncoupler added which is bound by mitochondria (e.g. pentachlorophenol84%, carbonylcyanide p-trifluoromethoxyphenylhydrazone 40%, and 2,4-dinitrophenol