вход по аккаунту


Anticonvulsant properties of acetone a brain ketone elevated by the ketogenic diet.

код для вставкиСкачать
Anticonvulsant Properties of Acetone, A
Brain Ketone Elevated by the Ketogenic Diet
Sergei S. Likhodii, PhD,1 Irina Serbanescu, BA,2,3 Miguel A. Cortez, MD,2,3 Patricia Murphy, PhD,1
O. Carter Snead III, MD,1–3 and W. McIntyre Burnham, PhD1
The ketogenic diet (KD), a treatment for drug-resistant epilepsy, elevates brain acetone. Acetone has been shown to
suppress experimental seizures. Whether elevation of acetone is the basis of the anticonvulsant effects of the KD and
whether acetone, like the KD, antagonizes many different types of seizures, however, is unknown. This study investigated
the spectrum of the anticonvulsant effects of acetone in animal seizure models. Rats were injected with acetone intraperitoneally. Dose–response effects were measured in four different models: (1) the maximal electroshock test, which
models human tonic-clonic seizures; (2) the subcutaneous pentylenetetrazole test, which models human typical absence
seizures; (3) the amygdala kindling test, which models human complex partial seizures with secondary generalization;
and (4) the AY-9944 test, which models chronic atypical absence seizures, a component of the Lennox–Gastaut syndrome. Acetone suppressed seizures in all of the models, with the following ED50’s (expressed in mmol/kg): maximal
electroshock, 6.6; pentylenetetrazole, 9.7; generalized kindled seizures, 13.1; focal kindled seizures, 26.5; AY-9944, 4.0.
Acetone appears to have a broad spectrum of anticonvulsant effects. These effects parallel the effects of the KD. Elevation
of brain acetone therefore may account for the efficacy of the KD in intractable epilepsy.
Ann Neurol 2003;54:219 –226
The ketogenic diet (KD) is a nondrug therapy for epileptic seizures. It was developed in the 1920s and was
based on the observation that fasting suppresses seizures, perhaps through the induction of ketosis.1,2 The
high-fat, low-carbohydrate KD mimics the ketogenic
effects of fasting and forces the body into a constant
state of ketosis.2– 6
The KD suppresses many different types of seizures,
including those which do not respond to the conventional anticonvulsants. The KD, in fact, is administered only after drug therapy has failed to provide seizure control.6,7 The KD is effective against tonic-clonic
seizures, absence seizures, complex partial seizures, and
the multiple types of intractable seizures associated
with the Lennox–Gastaut syndrome.6 – 8 The KD apparently works through a unique and broad-spectrum
anticonvulsant mechanism.
There have been many hypotheses concerning the
mechanism of action of the KD (for reviews see Nordli9 and Schwartzkroin10). The KD induces a cascade
of metabolic changes, including ketosis, a shift in electrolytes, elevation of plasma lipids, and changes in cerebral bioenergetics.9,10 Identifying a particular anti-
From the 1Department of Pharmacology
Research Program, Faculty of Medicine,
Division of Neurology, The Brain and
Department of Pediatrics, The Hospital
ronto, Ontario, Canada.
and Bloorview Epilepsy
University of Toronto;
Behavior Program; and
for Sick Children, To-
Received Dec 5, 2002, and in revised form Apr 5, 2003. Accepted
for publication Apr 8, 2003.
convulsant mechanism among these many changes has
been difficult.
Acetone is one of three ketone bodies elevated by the
KD.3,11,12 The idea that acetone might be anticonvulsant was formulated in the 1930s by Helmholz and
Keith13 and Keith.14 The idea received little attention,
however, in subsequent discussions of the KD. Interestingly, the anticonvulsant activity of acetone has been
independently rediscovered in studies unrelated to the
KD.15–22 The possible link between acetone and the
actions of the KD, however, remains unexplored.
If the anticonvulsant effects of the KD result from
an elevation of brain acetone, then the actions of acetone should resemble the actions of the KD. We have
recently shown that the KD and acetone are both effective against pentylenetetrazole (PTZ) seizures in
rats.23,24 It is not known, however, whether acetone
can reproduce the broad spectrum of anticonvulsant
actions of the KD. The objective of this study therefore
was to investigate the spectrum of acetone’s anticonvulsant effects.
Dose–response experiments were performed using
four different seizure tests: (1) the maximal electro-
Address correspondence to Dr Likhodii, Department of Pharmacology, Medical Sciences Building, University of Toronto, Toronto,
Canada, M5S 1A8.
© 2003 American Neurological Association
Published by Wiley-Liss, Inc., through Wiley Subscription Services
shock seizure (MES) test, which models human tonicclonic seizures25; (2) the subcutaneous PTZ test, which
models human typical absence seizures25,26; (3) the
amygdala kindling test, which models human complex
partial seizures with secondary generalization25,27; and
(4) the AY-9944 test, which models atypical absence
seizures, a component of the Lennox–Gastaut syndrome.28
Materials and Methods
All experimental procedures were approved by the University
of Toronto Animal Care Committee.
Albino, male Wistar rats (Charles River, St. Constant, Quebec, Canada), weighing 250 to 300gm, were used in the experiments with the MES (n ⫽ 60) and PTZ (n ⫽ 50) models. Wistar rats, matched for age, were used to measure
acetone concentrations in plasma and cerebrospinal fluid
(CSF) after injections of acetone (n ⫽ 30). In the kindling
experiments, albino, male Wistar rats, weighing 500 to
600gm were used (n ⫽ 13). For the AY-9944 experiments,
male Long-Evans hooded rats (Charles River), weighing 250
to 350gm, were used (n ⫽ 12).
Acetone Doses and Injections
Different solutions of acetone in physiological saline were
prepared to provide doses of 2, 4, 8, 16, or 32mmol/kg. The
MES test used an additional dose of 64mmol/kg (this dose
caused anesthesia and was abandoned). Solutions were injected intraperitoneally. The volume of injection was
10ml/kg of body weight. This injection volume allowed the
use of dilute solutions which caused no irritation.
Scoring of Ataxia
Fifteen (for PTZ test) or 30 (for other tests, see below) minutes after acetone injection, rats were rated for ataxia by two
independent observers, using the ataxia scale of Löscher.29
Rats were placed in an open field, observed for 1 minute,
and rated 0 to 5 using the following categories: (0) no ataxia;
(1) a slight ataxia in the hind limbs; (2) a more pronounced
ataxia, with a slight decrease in abdominal muscle tone; (3) a
further increase in ataxia with a more marked decrease in
abdominal tone; (4) marked ataxia with a loss of balance
during forward locomotion, and a loss of abdominal tone;
and (5) very marked ataxia with frequent losses of balance
during forward locomotion, and a loss of abdominal tone.
Procedure for the Maximal Electroshock Seizure Test
The MES test was administered 30 to 32 minutes after acetone injection. The MES stimulus consisted of a 60Hz sinewave current of 150mA with a train duration of 0.2 seconds.
The stimulus was applied via corneal electrodes. Each rat was
tested only once. Seizures were videotaped and scored by two
independent reviewers. Seizures were scored as “present” or
“absent.” Seizure absence was defined as a failure to extend
Annals of Neurology
Vol 54
No 2
August 2003
the hind limbs to an angle greater than 90 degrees during the
tonic period of the convulsion. Rats had been prescreened 6
days before the main study, using the same stimulus parameters. Only rats who exhibited hindlimb extension in prescreening were used.
Procedure for the Pentylenetetrazole Seizure Test
A solution of PTZ (Sigma, St. Louis, MO) in physiological
saline was injected subcutaneously at a dose of 50mg/kg into
a loose fold of skin on the back of the neck. The injection
volume was 7ml/kg of body weight. A dose of 50mg/kg previously had been found to be optimal for detecting small
elevations in seizure threshold.23 The PTZ test was administered 15 to 17 minutes after acetone injection. This interval
was selected because average latency to the first PTZ seizure
is approximately 15 minutes with a 50mg/kg dose.23 The
interval between acetone injection and a PTZ-induced seizure therefore was approximately 30 minutes, an interval
comparable to that in the MES and kindling experiments.
Immediately after PTZ injection, rats were placed in square
boxes (size 1 ⫻ 1m) and videotaped. Two independent reviewers scored the seizures. Seizures were scored as “present”
or “absent.” Seizure protection was defined as the absence of
whole-body clonus within 30 minutes of PTZ injection.
Each rat was tested only once.
Procedures for Kindled Seizures
A protocol of Albright and Burnham (1980) was used.30 Bipolar stimulating/recording electrodes (Plastics ONE,
Roanoke, VA) were implanted into the right basolateral
amygdala. The following coordinates were used: posterior
⫺1.0mm (relative to bregma); right lateral ⫹4.8mm
(bregma); and ventral (from the skull surface) ⫺8.5mm. The
incisor bar was set at ⫹5. Coordinates were derived from the
atlas of Pellegrino (1979).31 Fourteen days after surgery, a
period that allowed for recovery and habituation to the test
box, the kindling stimulations were begun. Each stimulus
consisted of a 1-second train of 60Hz biphasic 1-millisecond
square-wave pulses with a 400␮A peak-to-peak intensity.
During each trial, electrographic activity (electroencephalogram) was recorded. Convulsions, as they developed, were
rated according to the 5-point scale of Racine (1972).32
Stimulations were continued until rats had reached a criterion of 10 stage 5 convulsions.
The dose–response study then was initiated. On the first
test day, each rat received one of six doses of acetone. On the
following test day, rats were reused, each rat receiving a new
(different) dose of acetone. Doses were presented in random
order. By the end of the experiment, each rat had received all
six doses. To allow full recovery, we conducted tests at 48hour (minimum) intervals. Separate dose–response curves
were constructed for the generalized seizures and for spiking
at the amygdaloid focus. Seizures were scored as “present” or
“absent.” Generalized seizures were considered to be absent if
Racine’s stage 4 and 5 convulsions (rearing and rearing/falling) were absent.30 Focal afterdischarge was considered to be
absent if stimulation was followed by a normal electroencephalogram.
Procedures for the AY-9944 Seizure Test
AY-9944 is a trans-1,4-bis(2-chlorobenzylaminomethyl)cyclohexane dihydrochloride. It inhibits the biosynthesis of
cholesterol.33 AY-9944 was administered subcutaneously to
Long-Evans rat pups, at a dose of 7.5mg/kg, every 6 days
from postnatal day (P) 2 to P20. AY-9944 injections result
in atypical absence seizures, which recur throughout the rat’s
life.33 Surgical implantation of two frontal and two parietal
monopolar epidural electrodes was performed at P50, as
previously described. 28 Electrocorticography and dose–
response studies were performed after P55. The AY-9944 –
induced seizures were quantified by measurement of the
onset-to-offset durations of the slow spike-and-wave discharge (SWD) combined over the 20-minute intervals. An
independent electroencephalographer scored the seizures.
Rats were reused. Each rat received all doses of acetone 1
week apart. Doses were presented in random order. On a
test day, before injection of acetone, SWD was first quantified over 20-minute intervals for a total of 1 hour of continuous recordings. The average baseline SWD then was
calculated for each rat. Each rat then was injected with
physiological saline (“zero” dose) or a dose of acetone. Ten
minutes later, a postacetone SWD was quantified over a
20-minute interval.
Acetone Concentrations in Plasma and
Cerebrospinal Fluid
Rats were injected intraperitoneally with different doses of
acetone. Thirty minutes after the injections, rats were anaesthetized with 75mg/kg ketamine and 10mg/kg xylazine, and
CSF was extracted from the cisterna magna, and blood was
drawn by cardiac puncture.
Collected samples were derivatized with 2,4dinitrophenylhydrazine (DNPH; Sigma), as follows. Fifty
microliters of plasma or CSF were mixed with 100␮l of acetonitrile and centrifuged at 105g for 10 minutes to remove
proteins. Fifty microliters of clear supernatant were combined with 50␮l of the DNPH reagent, prepared as follows.
Two and a half milligrams of DNPH were dissolved in a
2.5ml mixture of concentrated HCL and distilled water (4:6
vol/vol) by warming in a water bath at 60°C for 20 minutes.
Derivatized samples were analyzed using Agilent 1100 Series HPLC system equipped with an autosampler (Agilent,
Palo Alto, CA). Samples were injected into a C18 column
Symmetry (5␮m particle size, 25cm ⫻ 4.2mm I.D.; Waters,
Milford, MA). A UV absorbance detector was set to 365nm.
The mobile phase was 63:37 (vol/vol) acetonitrile to water.
The flow rate was 1.0ml/min.
Data Analysis
Dose–response curves were constructed in the MES, PTZ,
and kindling experiments using “%protection” as the response parameter. We defined %protection in each dose
group as the percentage of rats free of generalized seizures or
focal afterdischarge. In the AY-9944 model, average %SWD
decrease was the response parameter. The %SWD decrease
was defined for each rat as 100% ⫻ [SWD(baseline) ⫺
SWD(acetone)]/SWD(baseline), where SWD(baseline) and
SWD(acetone) are the averaged baseline and the postinjec-
tion SWD durations, respectively, quantified over the 20minute intervals. The dose–response data were fitted using a
nonlinear sigmoidal model with a variable slope: Y ⫽ Bottom ⫹ (Top ⫺ Bottom)/(1 ⫹10{[Log(ED50) ⫺ X] ⫻ HillSlope}),
where Bottom, Top, HillSlope, and ED50 were the fitting
parameters, X was the logarithm of acetone concentration,
and Y was the response parameter.
Ataxia scores in each experiment were averaged for each
dose group. The averaged data were fitted using a nonlinear
sigmoidal model similar to the above, with the exception
that the ED50 was replaced with the TD50 fitting parameter,
and Y was the average ataxia score.
Graphing of the dose–response curves and data analysis
were performed using GraphPad Prizm software package version 3.02 for Windows (GraphPad Software, San Diego,
CA). The ED50’s from different seizure tests were compared
using a two-tailed t test built into the Prizm software. A
probability level of p less than 0.05 was considered statistically significant.
Anticonvulsant Potency of Acetone
The Figure, A–D, presents dose–response curves for
seizure suppression in the four different seizure models.
As indicated, acetone completely suppressed generalized seizures in all four models. Complete suppression
of focal seizures (kindling model) was observed in 60%
of kindled mice at 32mmol/kg, the highest dose tested
(see Fig, C).
ED50’s are presented in Table 1, along with 95%
confidence intervals and HillSlopes. As indicated,
ED50’s ranged from 6.6 to 13.1mmol/kg in the MES,
PTZ, and generalized kindling tests. (Expressed in milligrams per kilogram, they ranged from 383.3 to
760.9.) The ED50 was higher for the focal component
of amygdala-kindled seizures. (Expressed in milligrams
per kilogram, it was 1539.1.) Acetone was most potent
at suppressing seizures in the AY-9944 model, the
ED50 in this model was 4.0mmol/kg, which corresponds to 232.3mg/kg. The AY-induced seizures returned to the original baseline 4 to 5 hours after the
acetone treatment.
Toxicity and Therapeutic Index
Table 1 gives TD50’s for the occurrence of toxicity in
each experiment and presents calculated therapeutic indices (TI ⫽ TD50/ED50). As indicated, TD50’s were
not much different in the different experiments. The
TD50’s for the ataxia tests performed 15 and 30 minutes after acetone administration were also similar.
This suggests that acetone is rapidly absorbed and distributed in the body. Differences in the TI values in
different experiments were largely caused by the differences in the ED50’s.
Likhodii et al: Anticonvulsant Properties of Acetone
Acetone Concentrations in Plasma and
Cerebrospinal Fluid
Table 2 shows concentrations of acetone in plasma and
CSF 30 minutes after the injection with various doses
of acetone. Acetone concentrations in plasma and CSF
were directly proportional to the injected dose. Acetone
concentrations in CSF were similar to concentrations
in plasma.
Acetone: Anticonvulsant Spectrum and Potency
The major finding of these experiments was that acetone, like the KD in humans, displayed a broad spectrum of anticonvulsant effects, including effects in
models of tonic-clonic seizures, typical absence seizures,
atypical absence seizures, and complex partial seizures
with secondary generalization.
Our data are consistent with past reports of acetone’s
anticonvulsant effects in a variety of animal models. In
1931, Keith provided the first evidence of acetone’s anticonvulsant activity in a model utilizing the chemical
convulsant thujone.14 In later reports, acetone was
shown to increase threshold for electrically induced
convulsions34 and to block repeated maximal seizures
produced by semicarbazide, a powerful convulsant.15,16,19 Interestingly, diphenylhydantoin and pentobarbital failed to antagonize semicarbazide-induced
seizures.15 Acetone was also effective against clonictonic convulsions induced by isonicotinic acid and
electroshock.17 Additional evidence came from a series
of studies related to neurotoxicity of the volatile solvents. It was shown that short-term inhalation of acetone suppressed both audiogenic20 and electrically induced seizures.21,22 A recent report has documented
anticonvulsant effects of acetone in a mouse model of
audiogenic seizures.35 Finally, these results are consistent with our own reports of the anticonvulsant effects
of acute and chronic acetone in a threshold PTZ seizure test.24
These results suggest that acetone might be effective
in the treatment of seizures which resist anticonvulsant
drugs. In particular, acetone blocked seizures in models
of chronic atypical absence and complex partial seizures. These types of seizures are often resistant to the
traditional anticonvulsants28,30 but do appear to respond to the KD.6,8
Annals of Neurology
Vol 54
No 2
August 2003
Fig. Dose–response curves for seizure suppression in maximal
electroshock seizure (MES, panel A), pentylenetetrazole (PTZ,
panel B), amygdala kindling, (panel C) and AY-9944 (panel
D) seizure models. Experimental points in each panel were
fitted using a nonlinear sigmoidal dose–response model, which
allowed calculation of ED50 shown in the inserts. The parameters of the sigmoidal curves are shown in Table 1. SWD ⫽
spike-and-wave discharge.
Table 1. Parameters of the Dose–Response Curves
Type of Test
* , mmol/kg
(95% CI)
Hillslope ⫾ SE
TD50, mmol/kg
(95% CI)
6.6a (4.2–10.2)
9.7a,b (7.6–12.3)
2.2 ⫾ 0.6
3.9 ⫾ 1.0
28.4 (26.8–30.1)
27.5 (25.6–29.7)
13.1b (7.9–21.8)
26.5 (21.3–32.8)
4.0 (3.7–4.4)
2.8 ⫾ 1.2
3.8 ⫾ 1.3
5.5 ⫾ 1.6
30.5 (28.6–32.5)
24.0 (21.5–26.9)
*The differences between ED50’s in different tests were statistically significant ( p ⬍ 0.05) except for those sharing the same superscript letters.
CI ⫽ confidence interval; SE ⫽ standard error; MES ⫽ maximal electroshock; PTZ ⫽ pentylenetetrazole.
Table 2. Acetone Concentrations in Plasma and CSF
Dose, mmol/kg
Plasma, mmol/l
CSF, mmol/l
33.7 ⫾ 2.0
15.1 ⫾ 3.4
5.3 ⫾ 1.0
2.6 ⫾ 0.2
1.7 ⫾ 0.2
31.8 ⫾ 3.2
6.6 ⫾ 2.1
1.8 ⫾ 0.3
Data are given as mean ⫾ standard deviation of three to five measurements
amygdala focus (26.5mmol/kg) was higher than the
ED50’s in the other, generalized seizure tests. This is in
agreement with past findings in standard anticonvulsants.30 In most cases, the focal ED50 in standard anticonvulsants was four to six times higher.30 It is interesting that the focal ED50 for acetone is only twice as
high. Thus, acetone suppresses activity in the amygdala
focus at doses that are relatively close to the doses
which suppress generalized convulsions.
CSF ⫽ cerebrospinal fluid; ND ⫽ not determined.
It appears that acetone is most effective against models of chronic atypical absence, and less potent against
models of complex partial seizures. The anticonvulsant
potencies of acetone in the different seizure models
were AY-9944 ⬎ MES ⱖ PTZ ⱖ (kindling generalized) ⬎ (kindling focal). This ranking may reflect the
differences in intensity of the seizure stimuli, but it
may also relate to the mode of action by which acetone
suppresses seizures.
Interestingly, the difference in ED50’s of acetone in
the MES and PTZ tests was only approximately 1.5fold (see Table 1). Such a small difference is unusual.
Most standard anticonvulsants are much more effective
in one test than in the other. This is true in both
mice36 and in rats.37,38 Valproate, a classic “broad
spectrum” anticonvulsant, however, is also, like acetone, slightly more potent in the MES than in the
PTZ test, with the same ratio of potencies of approximately 1.5.37,38
The absolute potency of acetone is less than the
potency of most of the anticonvulsant drugs. It is
broadly comparable, however, to the potency of valproic acid. Acetone’s ED50’s of 383.3 and
563.4mg/kg in the MES and PTZ tests correspond to
valproate’s ED50’s of 110 to 178mg/kg and 157 to
242mg/kg.37 The ED50’s for orally administered valproic acid in rats were as high as 332 to 441mg/kg
and 469 to 985mg/kg for the MES and PTZ tests,
Acetone’s ED50 for suppression of spiking in the
Acetone: Plasma and Cerebrospinal
Fluid Concentrations
The concentrations of acetone in plasma and CSF
were similar and directly proportional to the injected
doses. The protein contents in the CSF and plasma,
however, are quite different; proteins are low in the
CSF and high in plasma. The similarity of acetone
concentrations in the CSF and plasma therefore suggests that acetone does not significantly bind to
plasma proteins. This is in contrast with many of the
standard anticonvulsants; the fraction of proteinbound valproate, phenytoin, and carbamazepine is as
high as 80 to 90%.39
Acetone: Toxicity and Therapeutic Index
The TIs for acetone ranged in generalized seizure
models from 6.0 in the AY-9944 test to 2.3 for kindled seizures (see Table 1). The TI against the kindled amygdala focus, however, was only slightly above
These scores are comparable to those for the standard anticonvulsants. The TIs for many anticonvulsants fall in a range of 11.0 (phenytoin) to 1.9 (valproate).37 Note that the open-field test (used in this
study) is more sensitive to motor impairment than
other, more commonly used tests, including rotorod.40
Our estimates of the TI therefore may be conservative.
Indeed, Kohli and colleagues (1967) found that the TI
for acetone in the MES tests in rats was as high as 11.1
(the authors did not specify the toxicity test used).17
Likhodii et al: Anticonvulsant Properties of Acetone
Their estimate of safety margin for acetone17 therefore
equals that for phenytoin.37
The toxicological profile of acetone has been examined in many studies unrelated to epilepsy. These have
involved both animals and human subjects.41
These studies generally have found no evidence of
serious neurotoxic, neurobehavioral, or other pathological effects of acetone.41 Although Mitran and colleagues (1997) seem to suggest possible neuropathy
from long-term acetone exposure in factory workers,42
the validity of their data and conclusions have been
Note that acetone is a natural, endogenous metabolite with potentially important functions.44 Significant
elevation of acetone occurs in fasting individuals and in
children on the KD.3,11 It is not known to be associated with any serious side effects.
Acetone: Links to Anticonvulsant Effects of the
Ketogenic Diet
According to Livingston (1972), the benefit of fasting
for controlling epileptic seizures has been known since
biblical times.5 In the early 1900s, several researchers
reported improved seizure control in patients undergoing fasting.1,45,46 The obvious limitations of fasting led
Wilder (1921) to suggest the use of a high-fat, lowcarbohydrate diet.2 The diet, which received the name
“ketogenic,” was designed to induce metabolic changes
similar to those occurring during fasting and, in particular, to induce ketosis.2 Talbot and colleagues
(1927) confirmed that the diet proposed by Wilder,
like fasting, elevated blood ketones and reduced blood
glucose.3 Of particular importance, fasting and the KD
both elevated acetone in the blood.3
Wilder (1921) postulated that the anticonvulsant effects of fasting and of the KD were caused by the “sedative” effects of acetoacetate.2 Keith (1931, 1932,
1933) reported that acetone and acetoacetate, but not
␤-hydroxybutyrate, protected rabbits from experimentally produced seizures.14,47,48 According to Keith, sodium acetoacetate had the most marked anticonvulsant
Other researchers, however, proposed alternative
mechanisms to explain the KD’s actions. Thus,
Bridge and Iob (1931) suggested involvement of compensatory changes in response to acidic properties of
ketone bodies.49 Millichap and colleagues (1964) believed in participation of disturbed balances of sodium and potassium ions.50 Dekaban (1966) linked
the beneficial effects of the KD to an increase in
plasma lipids.51
Huttenlocher (1976) thought that anticonvulsant effect was mostly related to elevated plasma acetoacetate
and ␤-hydroxybutyrate (acetone was not measured).52
The more recent study of Likhodii and colleagues
Annals of Neurology
Vol 54
No 2
August 2003
(2000), however, did not find any correlation between
␤-hydroxybutyrate and seizure protection.23 In an in
vitro model, Thio and colleagues (2000) did not find
any effects of acetoacetate or ␤-hydroxybutyrate on excitatory or inhibitory hippocampal synaptic transmission.53
Although Seymour and colleagues (1999) reported
an elevation in brain acetone in patients with epilepsy
treated with the KD,11 the authors did not link their
findings to those in Talbot and colleagues’ (1927) report of elevated blood acetone3 or Keith’s (1931) report of the anticonvulsant effects of acetone.14
Over the years, acetone has received little attention
in discussions of the possible mechanisms of action of
the KD. Until recently, Keith’s experiment of 193114
and that of Driver (1947)34 remained the only ones
that tested acetone for anticonvulsant properties and
suggested its possible involvement into the effects of
the KD.
Our own interest in acetone started with the idea of
using breath acetone to study a correlation between ketosis and seizure control.12,54 Our studies on the potential direct involvement of acetone into the anticonvulsant effects of the KD24,55 were prompted by a
considerable number of reports on the anticonvulsant
effects of acetone mostly from workers not involved in
KD research.14 –22,34
The fact that acetone has anticonvulsant properties
is important. It does not, however, prove that it is the
exclusive anticonvulsant mechanism of the KD. We
recently have outlined possible experiments that could
provide further evidence to link acetone to the effects
of the KD.24 These experiments were designed to
show (1) that the KD elevates acetone to anticonvulsant levels in the brain; (2) that the spectra of anticonvulsant actions of acetone and of the KD are similar; (3) that elevation of acetone in the brain after
initiation of the KD correlates with elevation of the
seizure threshold; and (4) that the KD loses its anticonvulsant effects if acetone does not increase in
blood and brain.
This investigation was one of these. It indicates that
acetone does have a broad spectrum of anticonvulsant
effects. Given that the KD may elevate plasma acetone
to concentrations of 1 to 4mmol/l in humans,24 these
results therefore support the hypothesis that acetone
may be the mechanism by which the KD suppresses
This study was designed to determine whether acetone
has a broad spectrum of anticonvulsant activity, a feature that is thought to be characteristic of the KD. Acetone suppressed acute and chronic experimental seizures of different types, indicating that it does have a
broad spectrum of anticonvulsant action. These suggest
that acetone would be capable clinically of suppressing
tonic-clonic, typical absence, complex partial, and
chronic atypical absence seizures. Perhaps most important, acetone was effective against kindled focal seizures, which, like complex partial seizures, are notoriously drug resistant.
These results are consistent with the hypothesis that
acetone is a causal factor of the KD’s anticonvulsant
effects. Considerably more data are needed, however,
before the “acetone hypothesis” will be proved. The related studies are currently under way.
This work was supported by grants from the Bloorview Children’s
Hospital Foundation (00024256, W.M.B.), Savoy Foundation (fellowship to S.S.L.), and NSERC (CHRP238016-00, W.M.B.).
Wyeth-Ayerst (Philadelphia, PA) is thanked for the gift of the AY9944. The investigation of the AY-9944 model is supported by The
Canadian Institutes of Health Research (14329MOP, O.C.S.) and
the Hospital for Sick Children Pediatric Consultants (77710-62300,
We are grateful to M. McLaughlin for assistance in preparation of
the manuscript.
The study was presented at the Pediatric Epilepsy Research Center
Ketogenic Diet Workshop, Seattle, WA, 2/8/01, and at the Annual
Meeting of the Eastern Association of Electroencephalographers,
New York, 3/24/01.
1. Geyelin HR. Fasting as a method for treating epilepsy. Med
Rec 1921;99:1037–1039.
2. Wilder RM. The effects of ketonemia on the course of epilepsy.
Mayo Clin Proc 1921;2:307–308.
3. Talbot FB, Metcalf KM, Moriarty ME. Epilepsy: chemical investigations of rational treatment by production of ketosis.
Am J Dis Child 1927;33:218 –225.
4. McQuarrie I, Keith HM. Relationship of variations in the degree of ketonuria to occurrence of convulsions in epileptic
children on ketogenic diets. Am J Dis Child 1927;34:
5. Livingston S. Comprehensive management of epilepsy in infancy, childhood and adolescence. Springfield: Thomas, 1972.
6. Swink TD, Vining EP, Freeman JM. The ketogenic diet: 1997.
Adv Pediatr 1997;44:297–329.
7. Hemingway C, Freeman JM, Pillas DJ, Pyzik PL. The ketogenic diet: a 3- to 6-year follow-up of 150 children enrolled
prospectively. Pediatrics 2001;108:898 –905.
8. Crumrine PK. Lennox-Gastaut syndrome. J Child Neurol
2002;17(suppl 1):S70 –S75.
9. Nordli D. The ketogenic diet: uses and abuses. Neurology
2002;58(12 suppl 7):S21–S24.
10. Schwartzkroin PA. Mechanisms underlying the anti-epileptic efficacy of the ketogenic diet. Epilepsy Res 1999;37:171–180.
11. Seymour KJ, Bluml S, Sutherling J, et al. Identification of cerebral acetone by 1H-MRS in patients with epilepsy controlled
by ketogenic diet. MAGMA 1999;8:33– 42.
12. Likhodii SS, Musa K, Cunnane SC. Breath acetone as a measure of systemic ketosis assessed in a rat model of the ketogenic
diet. Clin Chem 2002;48:115–120.
13. Helmholz HF, Keith HM. Eight years’ experience with the ketogenic diet in the treatment of epilepsy. JAMA 1930;95:
14. Keith HM. The effect of various factors on experimentally produced convulsions. Am J Dis Child 1931;41:532–543.
15. Jenney EH, Lee LD. The convulsant effect of semicarbazide.
J Pharmacol Exp Ther 1951;103:349.
16. Jenney EH, Pfeiffer CC. The convulsant effect of hydrazides
and the antidotal effect of anticonvulsants and metabolites.
J Pharmacol Exp Ther 1958;122:110 –123.
17. Kohli RP, Kishor K, Dua PR, Saxena RC. Anticonvulsant activity of some carbonyl containing compounds. Indian J Med
Res 1967;55:1221–1225.
18. Postolache V, Safta L, Cuparencu B, Steiner L. [Effect of acetone upon pentamethylenetetrazole convulsions] Arch Toxikol
19. Yamashita J. Convulsive seizure induced by intracerebral injection of semicarbazide (an anti-vitamin B6) in the mouse. J Nutr
Sci Vitaminol (Tokyo) 1976;22:1– 6.
20. Frantik E, Horvath M, Vodickova L. Effects of solvent mixtures
on behaviour and seizure characteristics at the utmost additive.
Act Nerv Super (Praha) 1988;30:260 –264.
21. Vodickova L, Frantik E, Vodickova A. Neutrotropic effects and
blood levels of solvents at combined exposures: binary mixtures
of toluene, o-xylene and acetone in rats and mice. Cent Eur J
Public Health 1995;3:57– 64.
22. Frantik E, Vodickova L, Hornychova M, Nosek M. Pattern of
inhalation exposure: blood levels and acute subnarcotic effects
of toluene and acetone in rats. Cent Eur J Public Health 1996;
4:226 –232.
23. Likhodii SS, Musa K, Mendonca A, et al. Dietary fat, ketosis,
and seizure resistance in rats on the ketogenic diet. Epilepsia
2000;41:1400 –1410.
24. Likhodii S, Burnham M. Ketogenic diet: does acetone stop seizures? Med Sci Monit 2002;8:HY19 –24.
25. Fisher RS. Animal models of the epilepsies. Brain Res Brain Res
Rev 1989;14:245–278.
26. Snead OC 3rd. Pharmacological models of generalized absence
seizures in rodents. J Neural Transm Suppl 1992;35:7–19.
27. Sato M, Racine RJ, McIntyre DC. Kindling: basic mechanisms
and clinical validity. Electroencephalogr Clin Neurophysiol
1990;76:459 – 472.
28. Cortez MA, McKerlie C, Snead OC 3rd. A model of atypical
absence seizures: EEG, pharmacology, and developmental characterization. Neurology 2001;56:341–349.
29. Löscher W, Honack D, Hashem A. Anticonvulsant efficacy of
clonazepam and the beta-carboline ZK 93423 during chronic
treatment in amygdala-kindled rats. Eur J Pharmacol 1987;143:
403– 414.
30. Albright PS, Burnham WM. Development of a new pharmacological seizure model: effects of anticonvulsants on cortical- and
amygdala-kindled seizures in the rat. Epilepsia 1980;21:
681– 689.
31. Pellegrino A. A stereotaxic atlas of the rat brain. 2nd ed. New
York: Plenum Press, 1979.
32. Racine RJ. Modification of seizure activity by electrical stimulation. II. Motor seizure. Electroencephalogr Clin Neurophysiol
33. Cortez MA, Cunnane SC, Snead OC 3rd. Brain sterols in the
AY-9944 rat model of atypical absence seizures. Epilepsia 2002;
43:3– 8.
34. Driver RL. Isopropyl alcohol, other ketogens, and miscellaneous
agents on thresholds for electrical convulsions and diphenylhydantoin. Proc Soc Exp Biol Med 1947;64:248 –251.
Likhodii et al: Anticonvulsant Properties of Acetone
35. Rho JM, Anderson GD, Donevan SD, White HS. Acetoacetate,
acetone, and dibenzylamine (a contaminant in l-(⫹)-betahydroxybutyrate) exhibit direct anticonvulsant actions in vivo.
Epilepsia 2002;43:358 –361.
36. Krall RL, Penry JK, White BG, et al. Antiepileptic drug development. II. Anticonvulsant drug screening. Epilepsia 1978;19:
409 – 428.
37. Löscher W, Nolting B. The role of technical, biological and
pharmacological factors in the laboratory evaluation of anticonvulsant drugs. IV. Protective indices. Epilepsy Res 1991;
38. White HS, Woodhead JH, Franklin MR, et al. General
principles: experimental selection, quantification, and evaluation of antiepileptic drugs. In: Levy RH, Mattson RH, Meldrum BS, eds. Antiepileptic drugs. New York: Raven Press,
1995:99 –109.
39. Shorvon SD. Handbook of epilepsy treatment. Malden, MA:
Blackwell Science, 2000.
40. Wláz P, Löscher W. Evaluation of associated behavioral and
cognitive deficits in anticonvulsant drug testing. In: Peterson
SL, Albertson TE, eds. Neuropharmacology methods in epilepsy research. Boca Raton: CRC Press, 1998:171–192.
41. World Health Organization. Environmental Health Criteria
207. Acetone. Geneva: World Health Organization, 1998.
42. Mitran E, Callender T, Orha B, et al. Neurotoxicity associated
with occupational exposure to acetone, methyl ethyl ketone,
and cyclohexanone. Environ Res 1997;73:181–188.
43. Graham DG. Critical analysis of Mitran et al. (1997). Neurotoxicity associated with occupational exposure to acetone,
methyl ethyl ketone, and cyclohexanone. Environ Res 2000;82:
Annals of Neurology
Vol 54
No 2
August 2003
44. Kalapos MP. Possible physiological roles of acetone metabolism
in humans. Med Hypotheses 1999;53:236 –242.
45. Guelpa G, Marie A. La lutte contre l’épilepsie par la désintoxication et par la réeducation alimentaire. Rev Ther Med Chir
[Paris] 1911;78:8 –13.
46. Conklin HW. The cause and treatment of epilepsy. Am J Osteo
Assoc 1922;22:11–14.
47. Keith HM. Further studies of the control of experimentally
produced convulsions. J Pharmacol Exp Ther 1932;44:
449 – 455.
48. Keith HM. Factors influencing experimentally produced convulsions. Arch Neurol Psychol 1933;29:148 –154.
49. Bridge EM, Iob LV. The mechanism of the ketogenic diet in
epilepsy. Bull Johns Hopkins Hosp 1931;48:373–389.
50. Millichap JG, Jones JD, Rudis BP. Mechanism of anticonvulsant action of ketogenic diet. Am J Dis Child 1964;107:
593– 604.
51. Dekaban AS. Plasma lipids in epileptic children treated with
the high fat diet. Arch Neurol 1966;15:177–184.
52. Huttenlocher PR. Ketonemia and seizures: metabolic and anticonvulsant effects of two ketogenic diets in childhood epilepsy.
Pediatr Res 1976;10:536 –540.
53. Thio LL, Wong M, Yamada KA. Ketone bodies do not directly
alter excitatory or inhibitory hippocampal synaptic transmission. Neurology 2000;54:325–331.
54. Musa-Veloso K, Likhodii SS, Cunnane SC. Breath acetone is a
reliable indicator of ketosis in adults consuming ketogenic
meals. Am J Clin Nutr 2002;76:65–70.
55. Likhodii SS, Snead OC, Burnham WM. The ketogenic diet.
Does acetone stop seizures? 55th Annual Meeting of the Eastern Association of Electroencephalographers, New York, March
24, 2001. Clin Neurophysiol 2002;113:1365.
Без категории
Размер файла
103 Кб
properties, diet, anticonvulsant, ketone, elevated, acetone, brain, ketogenic
Пожаловаться на содержимое документа