Biochem. J. (1960) 76, 47
Barbiturates and Oxidative Phosphorylation
By W. N. ALDRIDGE AND V. H. PARKER
Unitfor Research in Toxicology, M.R.C. Laboratories, Carshalton, Surrey
(Received 23 October 1959)
It has been claimed that barbiturates uncouple
oxidative phosphorylations (Brody & Bain, 1954;
Brody, 1955). These authors have demonstrated
that during the oxidation of pyruvate by liver and
brain mitochondria, phosphate uptake was lowered
proportionately more than oxygen uptake. Support
for the uncoupling theory was derived from certain
similarities between the barbiturates and
dinitrophenol (Brody & Bain, 1954). 'The slopes of
the inhibition curves of these compounds are
remarkably similar although dinitrophenol is the
more potent agent. Both depress fatty acid oxid-
ation and stimulate oxidative rate in a phosphate-
deficient system. The addition ofexcess magnesium
ion does not reverse the uncoupling action of either
the barbiturate or dinitrophenol' (Brody, 1955).
Further, barbiturate hypnosis is potentiated by
2:4-dinitrophenol (Killam, Brody & Bain, 1958).
None of these arguments is conclusive evidence for
uncoupling, and they ignore the major difference
that, in vitro, barbiturates inhibit (Bain, 1952)
whereas 2:4-dinitrophenol stimulates respiration
(Loomis & Lipmann, 1948; Parker, 1958). This
difference is readily demonstrated in vivo, for after
administration of barbiturates oxygen consump-
tion (Costa & Bonnycastle, 1955) and body temper-
ature (Birnie & Grayson, 1952; Lessin & Parkes,
1957) fall, whereas 2:4-dinitrophenol causes a rise
of both oxygen consumption (Cameron, 1958) and
body temperature (Stoner, 1956).
ences in behaviour in isolated systems have also
been demonstrated (Johnson & Quastel,
Jalling, Low, Ernster & Lindberg, 1957; Messer,
1958). Although the supporting evidence for the
Bain (1954) have nevertheless demonstrated a
lowering of the phosphorylation quotient (P/O
Respiration, with pyruvate as substrate, of the
liver mitochondria used in this paper can be in-
hibited 50% without any decrease of theP/Oratio
(Aldridge, 1957). With such preparations we have
re-examined the effect ofoxy- and thio-barbiturates
upon oxidative phosphorylation.
dissociating the process of oxidation from that of
oxidative phosphorylation by the intervention of
1958), we have studied the effects of
is not conclusive, Brody &
A preliminary report of these results has already
appeared (Aldridge & Parker, 1958).
Manometric experiments. The reasons for the composi-
tion and pH of the medium have been discussed by
Aldridge (1957, 1958). For measurements of respiration
each flask contained 3 ml. of a solution containing adeno-
sine 5-phosphate (1-15 mM), adenosine triphosphate (ATP;
1.06 mm), KC1 (10 mm), MgCl2 (14mm), ethylenediamine-
tetra-acetic acid (EDTA; lO mM), potassium phosphate
(50 mm), sucrose (30 mm) and substrates (10 mM, except
fumarate mm). For measurements ofoxidative phosphoryl-
ation, glucose (60 mM), glycylglycine (16.7 mM) and hexo-
kinase (200-400 units) were added to the above-mentioned
mixture. In each case the medium was adjusted to pH 6.7-
6-8 with KOH. For either type of experiment 0*3 ml. of a
suspension of mitochondria in 0-3M-sucrose, equivalent to
150 mg. wet wt. of liver, was used. For experiments on
oxidative phosphorylation, when a range of concentrations
of barbiturate was examined (method 1), P uptake was
measured between 10 and 22 min. after placing the flasks
in the bath at 370. Uptake of02for the same period was
calculated from the slope of the best straight line through
manometer readings at 10 min., 13 min. 20 sec., 16 min.
40 sec. and 20 min. For more accurate determinations of
P/0 ratio six flasks were used for each determination
(method 2). Manometer readings were taken at 10, 14, 18,
22 and 26 min. and the reaction was stopped by the addi-
tion of 6-5 ml. ofice-cold 5% (w/w) perchloric acid at 7, 11,
15, 19, 23 and 27 min. Thus P uptake was calculated from
the slope ofthe regression line through six values. Owing to
the successive removal of flasks for the determination of
inorganic phosphate, the slope of the regression line for 02
uptake was calculated from four readings at 10 min., four
at 14 min., three at 18 min., two at 22 min. and one at
26 min. (a total of 14 readings). The standard errors ofthese
slopes were calculated (Snedecor, 1946) and thus the phos-
phorylation quotient (P/O ratio) and its standard error
could be determined (Langer, 1951).
ratios have standard errors less than 10% of the ratio, the
difference between the control and experimental values has
been tested for significance (t test) on the basis of their
being derived from a large sample (Snedecor, 1946).
Preparation of the mitochondria. Mitochondria were pre-
pared as previously described (Aldridge, 1957, 1958), with
a Potter-Elvehjem-type homogenizer with a smooth glass
tube and Perspex pestle, with a total clearance of 0-02 in.
for rat liver and 0.01 in. for rat brain.
Adenosine-triphosphatase activity. Each beaker contained
3 ml. of a solution containing ATP (3 mm), KCI 5
Since these P/O
W. N. ALDRIDiE D V. H. PARKER
,MgCl2 (14 mM), EDTA (mMJA and sucrose (30 mM). The
beakers were shaken in air at 37°.
equilibration mitochondria were added and the mixture
was incubated for 10 min. The reaction was stopped by the
addition of 0 5 ml. of ice-cold 65% (w/w) perchloric acid
and inorganic phosphate was determined.
Special chemicals and reagents. The following chemicals
were obtained from the sources indicated: adenosine 5-
phosphate, glycylglycine, sodium pyruvate (Roche Pro-
ducts Ltd.); disodium salt of ATP (Sigma Chemical Co.,
St Louis, Mo., U.S.A.); glucose, sodium fumarate, 2:4-
dinitrophenol (DNP; British Drug Houses Ltd.). Phenyl-
arsenious acid was synthesized and used as described by
Aldridge (1958). The following barbiturates have also been
used: sodium 5-ethyl-5-phenyl barbiturate (Phenobarbital:
methylbarbituric acid (Hexobarbital: Mayand Baker Ltd.);
sodium 5-ethyl-5-isoamylbarbiturate (Amytal: Eli Lilly
and Co. Ltd.); sodium 5-ethyl-5-(1-methylbutyl)-2-thio-
thal: Imperial Chemical Industries Ltd.); sodium 5-allyl-5-
isobutylthiobarbiturate (Baytenal: Farbenfabriken Bayer,
A.G., Wuppertal-Elberfeld, Germany).
Hexokinase was prepared from baker's yeast by a modi-
fication by V. H. Parker (unpublished work) of the method
of Berger, Slein, Colowick & Cori (1946). The preparation
was taken to the equivalent of step 3a and, when assayed
by their procedure at 370 (instead of 300), had an activity
of 3500 units/ml.
Potato apyrase was prepared by the
method of Lee & Eiler (1951). This preparation liberated
1100ug.atoms of inorganic phosphate/hr./ml.
Inorganic phosphate was deter-
mined by the method of Fiske & Subbarow (1925). Protein
was measured by the biuret method ofRobinson & Hogden
(1940) as modified by Ald
pressed as mg. of albumin.
Purification of barbiturat
supplied were dissolved in
acids by the addition ofHC
were washed repeatedly a
were recrystallized as fobo
ethanol, m.p. 176° (uncorre
m.p. 146-5-147-5°; Amyta
Thiopental from water-d
1620; Baytenal from benz
was recrystallized from w
forms, one melting at 1240 a
is converted into the latter
Codex, p. 394, 1953). All
stock solutions of the free a
0-03 ml. of various concent
a constant final concentra
formamide always being pr
Units. The oxidative acti
expressed aspl.of O0/m
activity asjug.atomof P li
inhibitory power of a subs'
the negative logarithm of
will produce 50% inhibitio
are expressed as mean±S.E
tions upon separate pre]
A comparison of the in]
produced by the oxyb
barbital and Phenobari
lridge (1957), and has been ex-
tes. The sodium barbiturates as
water, converted into the free
11 and the precipitated free acids
After drying they
ws: Phenobarbital from water-
ycted); Hexobarbital as supplied,
al from benzene, m.p. 156-.5;
rene, m.p. 147-148°. Kemithal
It exists in two
and the other at 140°; the former
on prolonged heating (Brit. Vet.
barbiturates were prepared as
acids in dimethylformamide and
trations was added to the flasks,
Ltion of 1% (v/v) of dimethyl-
ivity ofthe mitochondria (qo2) is
ig. of protein/hr. and ATPase
iberated/mg. of protein/hr. The
;tance is given as itspI50 value,
the molar concentration which
n. Where errors are given these
with the number ofdetermina-
,parations of mitochondria
ation by liver mitochondria.
hibition of 02 and P uptake
arbiturates Amytal, Hexo-
bital is illustrated in Fig. 1.
Inhibition of 02 uptake (%)
Fig. 1. Effect of oxybarbiturates on oxidativephosphoryl-
ation by rat-liver mitochondria withpyruvateas substrate.
All values were obtained by method 1. The broken line
shows where points should be if theP/0ratio isunchanged
and the continuous line the calculated beststraightline.
1.0 mM); *, Phenobarbital (0.5-1.0 mM). The mean qO2
was 113-2±4-6 (6) and theP/Oratio 2-67±0-08 (6).
Inhibition of 02 uptake (%6)
Fig. 2. Effect ofthiobarbiturates on oxidative phosphoryl-
ation of rat-liver mitochondria with pyruvate as substrate.
All values were obtained by method 1. The broken line
shows where points should be if the P/0 ratio is unchanged
and the continuous line the calculated best straight line.
*, Thiopental (0.06-1.0 mM); A, Baytenal (0.2-1.0 mM);
*, Kemithal (0.1-0.4 mM). The meanqO2was 107-8±3-4
(5) and the P/0 ratio 2-56±0 11 (5). When 02 uptake is
50% inhibited, P uptake is inhibited by 68-69 %.
BARBITURATES AND OXTDATIVE PHOSPHORYLATION
Table 1. Effect of barbiturate8 on pyruvate oxidation and coupled pho8phorylation of liver mitochondria
For details of technique used see method 2, described under Methods. The control and experimental values
were obtained by parallel experiments upon the same preparation of mitochondria. Results are expressed as
mean±s.E. Meanqo for controls is 90-5±1-2 (12). pC = -log conen. (M).
Oxidative phosphorylation (P/O ratio)
1*90 (P 0 06)
1-55 (P 0.13)
2-04 (P 0*05)
0-87 (P 0*4)
0-81 (P 0-41)
8.18 (P < 0 01)
2-66 (P 0-013)
5.77 (P < 0*01)
10*4 (P <0-01)
941 (P < 0.01)
4.6 (P < 0 01)
2-4 (P 0 02)
Table 2. Influence of barbiturate8 on adenosine-triphosphatase activity, 8welling
and uptake of oxygen of un8timulated mitochondria
Values for ATPase are all corrected for the activity of the mitochondria in the presence of solvent (1%, v/v,
dimethylformamide), which was 073±007(6)pg.atomofP/mg. ofprotein/hr. In the presence of0 03 mm-DNP
their activity was 11x9±0-13 (6). Swelling was measured by measurements of E at 530mpin a Unicam DG.
spectrophotometer, the mitochondria being suspended and incubated at 370 for 10 min. under conditions identical
with those used for the ATPase assay. Uptake of02was measured in the presence of pyruvate and fumarate.
Meanq°2was 22*4+0-28 (6) and, in the presence of0 03 mm-DNP, 82-9±1.6 (6), giving a stimulation of 3-7-fold.
pC = -log conen. (M).
(yg.atoms of P/mg.
The broken line shows where the points will fall
with an unchanged P/O ratio. There is clearly little
indication that the inhibition of P uptake is any
greater than that of 02 uptake. In contrast, the
thiobarbiturates Thiopental, Kemithal and Bay-
tenal impaired P uptake more than 0, uptake, so
giving a lower P/O ratio (Fig. 2). As pointed out
in the Methods section these results were obtained
with a technique when the P/O ratio for each con-
measurements of 02 uptake in one Warburg flask
only. By using more flasks for the assessment of
the rates of 02 and P uptake, the standard error of
each P/O ratio can be obtained. In Table 1 are
shown the results of such experiments with six
flasks for each concentration of barbiturate. For
Hexobarbital and Phenobarbital the difference
between the P/O ratios when
02 uptake is 40-
Uptake of 02
or decrease ( - )
60% inhibited was not significant; the significance
for Amytal wasP approx. 0 1. For thiobarbiturates
there was a significant difference between the P/O
ratios.Even with quite a small lowering of 02
uptake (cf. Kemithal 0 1 mm) there was a marked
lowering of theP/Oratio.
Adeno8ine triphosphatase of liver mitochondria.
DNP, as well as uncoupling oxidative phosphoryl-
ation, inducesATPase activity inlivermitochondtia
(Lardy & Weilman, 1953).
interest to see whether the barbiturates, and parti-
At concentrations which inhibit
uptake about 50%, Amytal, Phenobarbital and
Hexobarbital gave an increase of ATPase activity
of less than
pg.atom of P/mg. of protein/hr.
Under similar conditions the thio-
an activity of more than
It was therefore of
Bioch. 1960, 76
W. N. ALDRIDGE AND V. H. PARKER
3,ug.atomsof P/mg. of protein/hr. This activation
of ATPase was not associated with swelling of the
mitochondria (Table 2).
four times as high as than those given in Table 2 did
not produce measurable changes in the extinction
ofthe mitochondrial suspension. The stimulation of
activity by the thiobarbiturates was
associated with stimulation of 02 uptake (Table 2,
Figs. 3 and 4) with pyruvate as substrate. No
stimulation of 02 uptake was found at any concen-
tration of the oxybarbiturates (Table 2).
Since the ATPase activity elicited by DNP may
be regarded as accounting for its uncoupling action
(see Discussion), a comparison has been made
between the action of the thiobarbiturates and of
DNP in lowering the P/O ratio, by choosing con-
centrations of the different uncoupling agents
that liberated equal activities of ATPase. When 02
uptake is 50% inhibited by thiobarbiturates, P
uptake is 69 % inhibited: an increase of 19% over
that required for the maintenance of a constant
P/O ratio (Fig.
measurements of oxidative phosphorylation was in
the range pC 3 4-3-7 (Tables
ATPase liberated by these concentrations was
3-28 jg.atoms of P/mg. of protein/hr. (Table 2).
The relationship between the concentration of
DNP and the activity of ATPase which it induces
conditions. It was found that 4,uM-DNP induced
the same ATPase activity as did the thiobarbitur-
ates and inhibited P uptake by 25-2 ± 1-8%and 02
uptake by 4-0 + 1-6 0/ (mean + S.E. of three deter-
minations upon the same preparation of mito-
chondria). The difference between the percentage
inhibition ofP and 02 uptakes is 21-2, in agreement
with 19-0 obtained with the thiobarbiturates.
was concluded that the relation between uncoupling
action and ATPase production was the same for the
thiobarbiturates as for DNP.
Adeno8ine-triphosphata8e activity and uptake of
oxygen induced by dinitrophenol. Uptake of 02 in
the presence of DNP reflects the activity of the
non-phosphorylating respiratory chain and ATPase
activity induced by DNP is a measure of the
activity of the energy-transfer mechanism when it
is dissociated from the respiratory chain; hence the
use of DNP enables separate study of these two
parts of the complete mechanism (see Discussion).
A comparison has therefore been made of the
action of the various barbiturates on 02 uptake in
the presence of DNP and the action of DNP in
inducing ATPase activity. The results in Table 3
show that 02 uptake in the presence of DNP was
inhibited by concentrations of oxybarbiturates
similar to those which inhibit 02 uptake in the
presence of hexokinase and glucose. The mito-
2). The concentration of thio-
1, 3). The mean
chondria were consistently rather more sensitive to
barbiturates when in the presence of hexokinase
and glucose than with DNP though in no case was
the difference more than twofold.
ATPase activity induced by DNP is not inhibited
by the similar concentrations of oxybarbiturates
and even concentrations four times as high have no
measurable effect (Table 3).
Experiments similar to the above have been
carried out with the thiobarbiturates, but are more
difficult to interpret. The results obtained with
Kemithal illustrate the difficulties (Figs.
presence of DNP but the precise concentration at
which a 50% inhibition occurs is difficult to assess.
Taking the whole 02 uptake in the presence of
DNP, 50% inhibition
(Fig. 3). If a correction is made for the stimulation
of 02 uptake by thiobarbiturates, 50% inhibition is
at pC 3*55. Kemithal stimulates ATPase (Fig. 4)
and this increases with concentration.
concentrations of Kemithal which produce 50%
inhibition of 02 uptake (pI50 3-60, cf. Table 3) give
only a small lowering of DNP-activated ATPase,
nevertheless, unlike the behaviour of the oxy-
barbiturates, a pronounced lowering occurs with
higher concentrations. The interpretation of this
effect depends on whether or not a correction should
be made for ATPase activated by the barbiturates
themselves (cf. Fig. 4).
is obtained at pC 3-30
and activity of adenosine tripho8phata8e of liver mito-
chondria in the presence of dinitrophenol
Effect of barbiturate8 on uptake of oxygen
Uptake of 02 was measured with pyruvate as substrate.
The mean qo0 was 22-4+0-28 (6) and, in the presence of
0-03 mM-DNP, 82-9+ 1-6 (6). The values for po50in the
presence of hexokinase and glucose are the ranges obtained
from the experiments given in Table 1 and Figs. 1 and 2.
ATPase activity in the absence of DNP and barbiturate
but in the presence of solvent (1%, v/v, dimethylform-
amide) was 0-73 +0-07 (6)pg.atomofP/mg. ofprotein/hr.,
and, in the presence of 0-03 mM-DNP, 11-9+0-13 (6). For
the inhibition of ATPase in the presenceof DNP concen-
trations ofoxybarbituratesfour times as high as those in
column 2 were used.
Inhibition of 02 uptake
In the presence
BARBITURATES AND OXIDATIVE PHOSPHORYLATION
It has previously been
demonstrated that succinate oxidation is unaffected
by barbiturates (Quastel & Wheatley, 1932-33) and
more recently that phosphorylation associated with
the oxidation of succinate is likewise not inhibited
by Amytal (Eiler & McEwen, 1949; Low, Ernster &
Lindberg, 1955). In the presence of Amytal one
atom of oxygen is consumed for every molecule of
Uptake of 02 in the presence of DNP
-log concn. (M) of Kemithal (pC).
Fig. 3. Effect ofKemithal on unstimulated02uptake and
on 02uptake stimulated by 2:4-dinitrophenol. Liver mito-
chondria were used with pyruvate as substrate: 0, in the
presence of30pM-DNP;0, in the absence of DNP. The
broken line is the difference between these curves in an
attempt to correct for the stimulation of 0° uptake by
Kemithal. The arrows indicate the concentration where 02
uptake is 50% inhibited: a broken arrow forthe uncorrected
and continuous arrow for the corrected curves.
succinate oxidized (Fig. 5), indicating that other
oxidations ofthe Krebs tricarboxylic acid cycle are
suppressed. We have confirmed that phosphoryl-
ation with succinate as substrate is not affected by
Amytal. Kemithal behaves in the same way with
succinate oxidation, the oxygen consumed being
one atom for every molecule ofsuccinate. However,
in agreement with the
pyruvate as substrate
stimulates rate of 02 uptake in the presence of
In view of the effects of
barbiturates in vivo it would have been desirable to
examine their influence upon the activities ofbrain
mitochondria. However, unlike liver mitochondria,
the respiration of our preparations of brain mito-
chondria declines during the course of the experi-
ment (Aldridge, 1957). This limits the work one can
do on oxidative phosphorylation. In addition, and
phenylarsenious acid which is generally believed to
inhibit a-oxo acid oxidases (Peters, 1955), inhibits
P uptake of brain mitochondria more than 02
uptake. A comparison of the effects of phenyl-
arsenious acid with the barbiturates is shown in
Fig. 6. These results indicate that for a given in-
hibition of 03 uptake (less than 60%) Amytal,
Phenobarbital and Hexobarbital do not inhibit P
uptake more than does phenylarsenious acid. In
results obtained with
3), Kemithal also
ATPase activity in the presence of DNP
-log concn. (M) of Kemithal (pC)
Fig. 4. Effect of Kemithal on ATPase with and without
DNP. The ATPase activity in the absence of DNP and
Kemithal but in the presence of solvent (1%, v/v, di-
methylformamide) was 0-63fg.atomof P/mg. of protein/
hr. This value has been subtracted from all results.
the presence of 0 03 mM-DNP; 0, in the absence of DNP.
The broken line is the difference between these curves.
Fig. 5. Effect of Amytal on 0, uptake with succinate as
Potato apyrase (1OjLg.atoms of P/hr.) was
added to each flask to produce maximal 0, uptake. Mito-
chondria (4-1 mg. of protein) equivalent to 150 mg. wet wt.
of liver were added to each flask. The02uptake during the
10 min. equilibration period was assumed to be at the same
rate as during the following 5 min. period. Each contained
30itmolesof succinate and, on the basis of 1pg.atom of
oxygen for each f&mole of succinate, complete oxidation
would require 336pl. of 02
0, with Amytal (1.0 mM).
0, Control without Amytal;