Progress in Neurobiology Vol. 58, pp. 31–59, 1999
`# 1999 Elsevier Science Ltd. All rights reserved
`Printed in Great Britain
`0301-0082/99/$ - see front matter
`
`PII: S0301-0082(98)00075-6
`
`VALPROATE: A REAPPRAISAL OF ITS
`PHARMACODYNAMIC PROPERTIES AND MECHANISMS
`OF ACTION
`
`WOLFGANG LO¨ SCHER*
`Department of Pharmacology, Toxicology and Pharmacy, School of Veterinary Medicine, Bu¨nteweg 17,
`30559 Hannover, Germany
`
`(Received 13 July 1998)
`
`Abstract—Valproate is currently one of the major antiepileptic drugs with e(cid:129)cacy for the treatment of
`both generalized and partial seizures in adults and children. Furthermore, the drug is increasingly used
`for therapy of bipolar and schizoa€ective disorders, neuropathic pain and for prophylactic treatment of
`migraine. These various therapeutic e€ects are reflected in preclinical models, including a variety of ani-
`mal models of seizures or epilepsy.
`The incidence of toxicity associated with the clinical use of valproate is low, but two rare toxic e€ects,
`idiosyncratic fatal hepatotoxicity and teratogenicity, necessitate precautions in risk patient populations.
`Studies from animal models on structure-relationships indicate that the mechanisms leading to hepato-
`toxicity and teratogenicity are distinct and also di€er from the mechanisms of anticonvulsant action of
`valproate.
`Because of its wide spectrum of anticonvulsant activity against di€erent seizure types, it has repeatedly
`been suggested that valproate acts through a combination of several mechanisms.
`As shown in this review, there is substantial evidence that valproate increases GABA synthesis and
`release and thereby potentiates GABAergic functions in some specific brain regions, such as substantia
`nigra, thought to be involved in the control of seizure generation and propagation. Furthermore, valpro-
`ate seems to reduce the release of the epileptogenic amino acid g-hydroxybutyric acid and to attenuate
`neuronal excitation induced by NMDA-type glutamate receptors.
`In addition to e€ects on amino acidergic neurotransmission, valproate exerts direct e€ects on excitable
`membranes, although the importance of this action is equivocal. Microdialysis data suggest that valpro-
`ate alters dopaminergic and serotonergic functions.
`Valproate is metabolized to several pharmacologically active metabolites, but because of the low
`plasma and brain concentrations of these compounds it is not likely that they contribute significantly to
`the anticonvulsant and toxic e€ects of treatment with the parent drug.
`By the experimental observations summarized in this review, most clinical e€ects of valproate can be
`explained, although much remains to be learned at a number of di€erent levels of valproate’s mechan-
`isms of action. # 1999 Elsevier Science Ltd. All rights reserved.
`
`CONTENTS
`
`1. Introduction
`2. Chemistry and physicochemical properties of valproate
`3. Pharmacokinetics of valproate in di€erent species
`4. Clinical use of valproate
`4.1. Epilepsy
`4.2. Other clinical indications
`5. Preclinical pharmacodynamics of valproate
`5.1. Anticonvulsant e€ects of valproate in animal models
`5.2. Anticonvulsant e€ects of valproate in in vitro models
`5.3. Other pharmacodynamic e€ects of valproate in animal models
`6. Adverse e€ects and toxicity of valproate
`7. Mechanisms of action of valproate
`7.1. Neurochemical e€ects of valproate on the GABA system
`7.2. Neurochemical e€ects of valproate on amino acids other than GABA
`7.3. Neurophysiological e€ects of valproate on amino acidergic neurotransmitter functions
`7.4. Neurophysiological e€ects of valproate on neuronal membranes
`7.5. Neurochemical e€ects of valproate on nonamino acidergic neurotransmitters
`7.6. Neurochemical and neurophysiological e€ects of active metabolites of valproate
`7.7. Putative mechanisms involved in the early and late anticonvulsant e€ects of valproate
`8. Conclusions
`References
`
`* Tel.: +49-511-953-8720; Fax: +49-511-953-8581; E-mail: wloscher@pharma.tiho-hannover.de
`
`31
`
`32
`32
`32
`33
`33
`35
`35
`35
`38
`39
`39
`40
`41
`48
`48
`49
`51
`52
`54
`54
`54
`
`Ranbaxy Ex. 1019
`IPR Petition - USP 9,050,302
`
`

`

`32
`
`W. Lo¨ scher
`
`Cyclic adenosine monophosphate
`cAMP
`Guanosine 3’,5’-monophosphate
`cGMP
`Central nervous system
`CNS
`Cerebrospinal fluid
`CSF
`DOPAC 3,4-Dihydroxyphenylacetic acid
`EEG
`Electroencephalogram
`g-Aminobutyric acid
`GABA
`GABA-T GABA aminotransferase
`GAD
`Glutamic acid decarboxylase
`g-Hydroxybutyric acid
`GHB
`
`ABBREVIATIONS
`Homovanillic acid
`HVA
`5-HIAA 5-Hydroxyindoleacetic acid
`KDHC a-Ketoglutarate dehydrogenase complex
`NMDA N-Methyl-D-aspartate
`SN
`Substantia nigra
`SNR
`Substantia nigra pars reticulata
`SRF
`Sustained repetitive firing
`SSA
`Succinic semialdehyde
`SSADH Succinic semialdehyde dehydrogenase
`SSAR
`Succinic semialdehyde reductase
`
`1. INTRODUCTION
`
`Valproate (valproic acid; usually used as its sodium
`salt), also referred to as di-n-propylacetic acid, is a
`simple eight-carbon branched-chain fatty acid with
`unique anticonvulsant properties. Valproic acid was
`first synthesized in 1882 by Burton (1882), but there
`was no known clinical use until its anticonvulsant
`activity was
`fortuitously discovered by Pierre
`Eymard in 1962 in the laboratory of G. Carraz,
`which was published by Meunier et al. (1963). The
`first clinical trials of the sodium salt of valproate
`were reported in 1964 by Carraz et al. (1964). It was
`marketed in France in 1967 and was released sub-
`sequently in >100 other countries (in the USA in
`1978) for the treatment of epilepsy. Since then,
`valproate has established itself worldwide as a major
`antiepileptic drug against several types of epileptic
`seizures. Clinical experience with valproate has con-
`tinued to grow in recent years,
`including use of
`valproate for diseases other
`than epilepsy,
`for
`example, in bipolar disorders and migraine.
`The present review is not meant to be an exhaus-
`tive survey on valproate; detailed summaries on var-
`ious aspects of valproate’s actions are already
`available (Pinder et al., 1977; Meldrum, 1980;
`Turner and Whittle, 1980; Chapman et al., 1982;
`Hammond et al., 1981; Kerwin and Taberner,
`1981; Johnston, 1984; Morre et al., 1984; Lo¨ scher,
`1985; Macdonald and McLean, 1986; Cotariu et al.,
`1990; Lo¨ scher, 1991, 1993a; Davis et al., 1994;
`Fariello et al., 1995; Lo¨ scher, 1998b). This review
`concentrates on preclinical studies with particular
`emphasis on valproate’s actions that appear to be of
`importance for its diverse therapeutic e€ects.
`
`2. CHEMISTRY AND PHYSICOCHEMICAL
`PROPERTIES OF VALPROATE
`
`Valproic acid or valproate is the trivial name for
`2-propylpentanoic acid (also called n-dipropylacetic
`acid). As a simple branched-chain carboxylic acid it
`di€ers markedly in structure from all other antiepi-
`leptic drugs in clinical use. Its structural formula is
`as follows:
`
`Valproic acid (molecular weight, 144.21; melting
`point 120–1218C) is a colourless liquid with a pKa
`value of 4.56 (Lo¨ scher, 1985). The partition coe(cid:129)-
`cients of valproic acid between organic solvents and
`bu€er at pH 7.4 have been reported as 0.013 for
`heptane, 0.064 for benzene and 0.21 for chloroform
`(Lo¨ scher and Frey, 1984). Thus, because of its high
`degree of ionization at pH 7.4, valproic acid is much
`less lipid-soluble than any other standard anticon-
`vulsant drug (Lo¨ scher and Frey, 1984). This
`explains why the volume of distribution of valproic
`acid is so low (see Section 3), because only the non-
`ionized, lipid-soluble part of a drug can distribute
`from blood to tissues by passive di€usion. However,
`the rapid entry of valproate in the brain is not com-
`patible with its physicochemical properties and is
`thought to be mediated by active transport mechan-
`isms (see Section 3).
`Valproic acid is usually used as its sodium salt,
`which has a molecular weight of 166.198. Sodium
`valproate is a hygroscopic white powder which dis-
`solves readily in polar solvents (e.g. water, ethanol,
`methanol) but is poorly soluble in solvents of lower
`polarity. In the body, however, sodium valproate is
`rapidly dissociated to valproic acid with the physico-
`chemical properties described above. In addition to
`sodium valproate, the drug is available in several
`forms,
`including the parent compound,
`its mag-
`nesium salt and a combination of the parent com-
`pound and its sodium salt (divalproex sodium). In
`this review, the term valproate will be used for all of
`these formulations. For clinical use, valproate is
`available in capsule, tablet, enteric-coated tablet,
`sprinkle, liquid, intravenous, suppository and con-
`trolled-release formulations. Pharmacokinetic and
`tolerability di€erences between formulations have
`been reviewed recently (Davis et al., 1994).
`
`3. PHARMACOKINETICS OF VALPROATE IN
`DIFFERENT SPECIES
`
`The main pharmacokinetic data for valproate in
`di€erent
`species are
`summarized in Table 1.
`Valproate is rapidly absorbed after di€erent routes
`of administration, provided that conventional for-
`mulations (e.g. no slow release or retard formu-
`lations)
`are
`used. Bioavailability
`after
`oral
`administration depends on the species. While it is up
`to almost 100% in humans, it is much lower at the
`high doses often given in rodents. While volume of
`distribution is similar in most species (being about
`equal to the extracellular fluid volume), there are
`
`

`

`Valproate
`
`33
`
`Table 1. Pharmacokinetics of valproate in di€erent species
`
`Species
`
`Man
`Rhesus monkey
`Dog
`Cat
`Rat
`Mouse
`
`Apparent volume
`of distribution
`(l kg(cid:255)1)a
`
`Half-life
`[t0.5 (b)] (h)
`
`Bioavailability
`after oral
`application
`
`Plasma protein
`bindingb (%)
`
`Brain/plasma
`ratiob
`
`CSF/plasma
`ratiob
`
`0.13–0.19
`0.17
`0.21–0.77
`0.38
`0.66
`0.33
`
`9–18
`0.66
`1–4
`9
`2–5c
`0.8
`
`70–100
`
`80–90
`
`34–47
`
`80–95
`80
`70–80
`
`63
`12
`
`0.07–0.28
`0.22
`0.28–0.39
`0.2–0.7
`0.18–0.32
`0.15–0.2
`
`0.08–0.25
`0.3
`0.2–0.4
`
`aVd(b) or Vd(ss).bAt ‘‘therapeutic’’ plasma concentrations of 50–80 mg ml(cid:255)1.cNonlinear kinetics. Adapted from Lo¨ scher (1985) and Levy
`and Shen (1995).
`
`dramatic species di€erences in elimination half-life.
`Some species, for example, the rat, exhibit dose-
`dependent nonlinear elimination kinetics. The mark-
`edly lower elimination half-life of most species com-
`pared to humans explains why higher doses of
`valproate are needed in most species to obtain com-
`parable ‘active’ plasma levels as in man. A further
`marked species di€erence is plasma protein binding,
`ranging from extensive binding in humans to almost
`absent binding in mice. Despite this di€erence in
`protein binding, brain/plasma ratios of valproate
`are almost the same in all species investigated in this
`regard. About 20% of the plasma concentration of
`valproate is present in the brain, and similar figures
`are
`present
`in
`cerebrospinal
`fluid
`(CSF).
`Experiments on the kinetics of penetration of com-
`mon antiepileptic drugs into CSF of dogs have
`shown that valproate enters the central nervous sys-
`tem (CNS) rapidly, which is in contrast to its physi-
`cochemical properties (see Section 2) and can best
`be explained by a saturable and probenecid-sensitive
`transport carrier at the blood–brain and blood–CSF
`barrier (Frey and Lo¨ scher, 1978; Lo¨ scher and Frey,
`1984). As shown by experiments with probenecid,
`valproate is also rapidly transported out of the
`brain, which explains the relatively low brain/plasma
`ratios. A further explanation is that valproate does
`apparently not bind to brain proteins. Accordingly,
`acute and subacute studies in rodents showed no
`retention of valproate in the brain (cf. Lo¨ scher,
`1985). However, a gradual accumulation of radioac-
`tivity was observed in the olfactory bulb in mice and
`rats following intraveneous (i.v.) injection as well as
`in monkeys after long-term infusion of radiolabeled
`valproate (Schobben et al., 1980; Hoeppner, 1990).
`Whether this radioactivity was due to covalent
`bound valproate, a valproate metabolite, or other
`degradation products is not clear. Interestingly,
`destruction of the olfactory bulb markedly reduces
`the anticonvulsant e(cid:129)cacy of valproate (Ueki et al.,
`1977). In this regard, it may be important that the
`principal target of olfactory bulb e€erent projections
`is the piriform (primary olfactory) cortex, a region
`that seems to be critical to the amplification and
`generalization of seizures (Lo¨ scher and Ebert, 1996).
`Biotransformation is the major route of elimin-
`ation of
`valproate
`in humans
`and animals.
`Valproate undergoes metabolism by a variety of
`conjugation and oxidative processes, which has been
`reviewed in detail recently (Baillie and She€els,
`1995). Several of the resulting unsaturated and oxy-
`
`genated metabolites of valproate exert anticonvul-
`sant activity (see Section 7.6), although the brain
`concentrations of these metabolites are too low to
`contribute to any significant extent to the anticon-
`vulsant activity of the parent drug. However, it has
`often proposed that metabolites may be involved in
`the toxicity of valproate (see Section 6). Some of the
`major metabolites of valproate are shown in Fig. 1.
`With respect
`to pharmacokinetic drug inter-
`actions, valproate may alter the plasma and brain
`levels of other drugs, due to interactions at the level
`of drug metabolism and plasma protein binding (cf.
`Lo¨ scher, 1985; Perucca and Richens, 1985; Davis et
`al., 1994). Similarly, other drugs may a€ect plasma
`and brain levels of valproate, thereby changing its
`pharmacodynamic potencies.
`
`4. CLINICAL USE OF VALPROATE
`
`4.1. Epilepsy
`
`The major use of valproate is in the pharmaco-
`logical therapy of epileptic seizures, although its use
`in other indications, such as psychiatric disorders
`and migraine, is steadily increasing. Epilepsy is one
`of the most common diseases of the brain, a€ecting
`at least 50 million persons worldwide (Scheuer and
`Pedley, 1990). Epilepsy is a chronic and often pro-
`gressive disorder characterized by the periodic and
`unpredictable occurrence of epileptic seizures which
`are caused by an abnormal discharge of cerebral
`neurons. Many di€erent types of seizures can be
`identified on the basis of their clinical phenomena.
`These clinical characteristics, along with their elec-
`troencephalographic (EEG) features, can be used to
`categorize seizures (Commission on Classification
`and Terminology of
`the
`International League
`Against Epilepsy, 1981). Seizures are fundamentally
`divided into two major groups: partial and general-
`ized. Partial (focal, local) seizures are those in which
`clinical or electrographic evidence exists to suggest
`that the attacks have a localized onset in the brain,
`usually in a portion of one hemisphere, while gener-
`alized seizures are those in which evidence for a
`localized onset is lacking. Partial seizures are further
`subdivided into simple partial, complex partial and
`partial seizures evolving to secondarily generalized
`seizures, while generalized seizures are categorized
`into absence (nonconvulsive), myoclonic, clonic,
`tonic, tonic–clonic and atonic seizures. In addition
`to classifying the seizures that occur in patients with
`
`

`

`34
`
`W. Lo¨ scher
`
`Fig. 1. Structure of some valproate metabolites. (A) Structures of metabolites of valproate believed to
`arise through mitochondrial b-oxidation. (B) Metabolites believed to derive from oxidative processes dis-
`tinct from those of mitochondrial b-oxidation (Baillie and Levy, 1991). Most of the metabolites illus-
`trated in this figure have been demonstrated to exert anticonvulsant activity in animal models, although
`all metabolites (except the E-2-en unsaturated one) were less potent than valproate.
`
`epilepsy, patients are classified into appropriate
`types of epilepsy or epileptic syndromes character-
`ized by di€erent seizure types, etiologies, ages of
`onset
`and EEG features
`(Commission
`on
`Classification and Terminology of the International
`League Against Epilepsy, 1989). More than 40 dis-
`tinct epileptic syndromes have been identified, mak-
`ing epilepsy a remarkably diverse collection of
`disorders. The first major division of epilepsy are
`localization-related (focal, local, partial) epilepsies,
`which account for roughly 60% of all epilepsies,
`and generalized epilepsies, which account for ca
`40% of all epilepsies. An epilepsy or epileptic syn-
`
`drome is either idiopathic, which is virtually synon-
`ymous with genetic epilepsy, or symptomatic, that
`is, due to structural
`lesion or major identifiable
`metabolic derangements. Both types of seizure and
`epilepsy determine the choice and prognosis of
`therapy. For instance, the most common and most
`di(cid:129)cult-to-treat type of seizures in adult patients are
`complex partial seizures, while primary generalized
`tonic–clonic (‘grand mal’) seizures respond in most
`patients to treatment with anticonvulsants.
`Various clinical studies and extensive clinical ex-
`perience over the last decades have demonstrated
`that valproate is e€ective in the treatment of various
`
`

`

`Valproate
`
`35
`
`including tonic–clonic, absence and
`seizure types,
`partial seizures, both as add-on and monotherapy
`(cf. Davis et al., 1994). The drug has also demon-
`strated some evidence of e(cid:129)cacy in the treatment of
`infantile spasms (West syndrome), Lennox–Gastaut
`syndrome,
`febrile seizures and status epilepticus
`(Davis et al., 1994). Because of this wide-spectrum
`of anticonvulsant activity and its
`tolerability,
`valproate is a well-established first-line treatment for
`patients with a broad range of seizure types. Indeed,
`the discovery and therapeutic development of
`valproate can be considered a milestone in drug
`therapy of the epilepsies (Lo¨ scher, 1998b).
`
`4.2. Other Clinical Indications
`
`In addition to epilepsy, valproate is increasingly
`used for treatment of other diseases, including bipo-
`lar disorders, migraine
`and neuropathic pain
`(Balfour and Bryson, 1994; Petty, 1995). Valproate
`has been shown to be e€ective in patients with bipo-
`lar and schizoa€ective disorders, including those re-
`sistant to lithium and carbamazepine. The drug is
`particularly e€ective against the manic episodes of
`bipolar disorders, although during long term pro-
`phylaxis both manic and depressive episodes may be
`reduced. Valproate also may have a wider spectrum
`of e(cid:129)cacy than lithium, with accumulating evidence
`of its use in atypical (dysphoric/mixed) mania, rapid
`cycling and secondary manias,
`in which lithium
`appears to be less clinically e€ective. The mood-sta-
`bilizing e€ect of valproate in psychiatric conditions
`is
`shared by some other anticonvulsant drugs,
`namely carbamazepine,
`lamotrigine, and possibly
`also gabapentin. Valproate, carbamazepine and
`gabapentin are also considered e€ective for treat-
`ment of chronic neuropathic pain. In migraine,
`valproate has been shown to be an e€ective prophy-
`lactic treatment in several controlled clinical trials
`(Balfour and Bryson, 1994; Silberstein, 1998).
`
`5. PRECLINICAL PHARMACODYNAMICS OF
`VALPROATE
`
`5.1. Anticonvulsant E€ects of Valproate in Animal
`Models
`
`As noted above, the anticonvulsant properties of
`valproate were serendipitously discovered in France
`in 1962 (Meunier et al., 1963). By using valproate as
`a lipophilic vehicle for dissolving water-insoluble
`khelline derivatives, a significant anticonvulsant
`e€ect against pentylenetetrazol (PTZ)-induced sei-
`zures was observed in the
`vehicle
`controls.
`Subsequent clinical trials substantiated the anticon-
`vulsant activity of valproate in epileptic patients,
`and nowadays valproate is one of the major drugs
`for treatment of di€erent types of epileptic seizures.
`Experimentally, valproate exerts anticonvulsant
`e€ects in almost all animal models of seizure states
`examined in this respect
`(Tables 2–4),
`including
`models of di€erent types of generalized seizures as
`well as focal seizures. The anticonvulsant potency of
`valproate strongly depends on the species, the route
`of administration, the type of seizure induction, and
`the time interval between drug administration and
`seizure induction. Because of the rapid penetration
`into the brain but the short half-life of valproate in
`the most marked
`most species (Lo¨ scher, 1985),
`e€ects are obtained shortly, that is, 2–15 min, after
`parenteral
`[e.g.
`intraperitoneal
`(i.p.)]
`injection.
`Depending on the preparation, onset of action after
`oral administration may be somewhat retarded. In
`most
`laboratory animal species,
`the duration of
`anticonvulsant action of valproate is only short so
`that high doses of valproate are needed to suppress
`long-lasting or repeatedly occurring seizures in ani-
`mal models. In general, the anticonvulsant potency
`of valproate increases in parallel with the size of the
`animal.
`In rodents,
`the highest anticonvulsant
`potencies are obtained in genetically seizure suscep-
`tible species, such as gerbils and rats with spon-
`
`Table 2. Anticonvulsant potency of valproate in di€erent animal models of generalized clonic or tonic seizures with
`chemical seizure induction
`
`Model
`
`E(cid:129)cacy of valproate
`
`Convulsant
`(mg kg(cid:255)1)
`
`Seizure
`type
`
`Species Timea
`(hr)
`
`Route of
`application
`
`ED50
`(mg kg(cid:255)1)
`
`References (examples)
`
`PTZc (85–100 s.c.)
`
`Clonic Mouse
`
`0.25
`
`PTZc (85–100 s.c.)
`
`Clonic
`
`Rat
`
`Picrotoxin (3.2 s.c.)
`Bicuculline (2.7 s.c.)
`3-MPd (66 s.c.)
`Allylglycine (400 i.v.)
`Isoniazide (200 s.c.)
`DMCMf
`Strychnine (1.2 s.c.)
`NMDLAg
`
`Clonic Mouse
`Clonic Mouse
`Clonic Mouse
`Clonic Mouse
`Clonic Mouse
`Clonic/tonic Mouse
`Tonic
`Mouse
`Clonic Mouse
`
`0.5
`
`0.5
`0.5
`0.25
`0.25
`0.5
`0.5
`0e
`0.5
`0.25
`0.5
`
`i.p.
`
`p.o.
`
`i.p.
`p.o.
`i.p.
`i.p.
`i.p.
`i.p.
`p.o.
`i.p.
`i.p.
`i.p.
`
`120–150 Swinyard (1964); Shuto and Nishigaki (1970);
`Krall et al. (1978)
`220–420 Swinyard (1964); Shuto and Nishigaki (1970);
`Frey and Lo¨ scher (1976)
`Swinyard (1964); Kupferberg (1980)
`74–260
`Kupferberg (1980)
`180
`Kupferberg (1980)
`390
`Kupferberg (1980)
`360
`Lo¨ scher (1980b)
`290
`200 Worms and Lloyd (1981)
`280
`Lo¨ scher and Frey (1977b)
`60
`Petersen (1983)
`290
`Kupferberg (1980)
`340
`Czuczwar et al. (1985)
`
`aTime from application of valproate to injection of convulsant.bDose which protects 50% of animals from seizures.cPentylenetetrazol.d3-
`Mercaptopropionic acid.eValproate and isoniazide were injected simultaneously.fMethyl-6,7-dimethoxy-4-ethyl-b-carbolin-3-carboxylate
`(an inverse agonist at central benzodiazepine receptors).gN-Methyl-D,L-aspartate (an agonist at the NMDA subtype of glutamate recep-
`tors).
`
`

`

`36
`
`W. Lo¨ scher
`
`Table 3. Anticonvulsant potency of valproate in animal models of generalized tonic–clonic or focal seizures with electri-
`cal seizure induction
`
`Model
`
`E(cid:129)cacy of valproate
`
`Name
`
`Stimulus
`
`Seizure type
`
`Species Timea
`(hr)
`
`Route of
`administration
`
`b
`ED50
`(mg kg(cid:255)1)
`
`References (examples)
`
`MESc
`
`50 mA
`
`Tonic
`
`Mouse
`
`0.25
`
`i.p.
`
`235–270
`
`MES
`
`150 mA
`
`Tonic
`
`Rat
`
`MES
`MES
`Amygdala-
`kindling
`
`Tonic
`Tonic
`500 mA Gen. clonusd
`
`Rabbit
`Cat
`Rat
`
`0.5
`0.5
`1.0
`0.5
`0.5
`0.25
`
`Complex-focal
`(clinical)
`Focal (EEG)
`
`Rat
`
`0.25
`
`Rat
`
`0.25
`
`p.o.
`i.p.
`p.o.
`i.p.
`i.p.
`i.p.
`
`i.p.
`
`i.p.
`
`Swinyard (1964);
`Shuto and Nishigaki (1970);
`Krall et al. (1978)
`Shuto and Nishigaki (1970)
`Swinyard (1964); Kupferberg (1980)
`Swinyard (1964); Kupferberg (1980)
`Swinyard (1964)
`Swinyard (1964)
`Lo¨ scher et al. (1986)
`
`315
`140–170
`320–490
`235
`67
`190
`
`220
`
`300
`
`Lo¨ scher et al. (1986)
`
`Lo¨ scher et al. (1986)
`
`aTime between administration of valproate and electrical stimulation.bDose which protects 50% of animals from seizures.cMaximal elec-
`troshock seizure.dSecondarily generalized clonus following focal onset of seizures.
`
`taneously occurring spike-wave discharges (Table 3),
`and against DMCM-induced seizures
`in mice
`(Table 2).
`Three animal models that are commonly used in
`characterization of anticonvulsant drugs are the
`maximal electroshock seizure (MES) test, the subcu-
`taneous (s.c.) PTZ seizure test and kindling. The
`MES test,
`in which tonic hindlimb seizure are
`induced by bilateral corneal or transauricular electri-
`cal stimulation, is thought to be predictive of antic-
`onvulsant drug e(cid:129)cacy against generalized tonic–
`clonic seizures, while the PTZ test, in which general-
`ized myoclonic and clonic seizures are induced by
`systemic (usually s.c.) administration of convulsant
`doses of PTZ, is thought to represent a valid model
`for generalized absence and/or myoclonic seizures in
`humans (Lo¨ scher and Schmidt, 1988). The kindling
`model with electrical stimulation via chronically
`implanted electrodes in amygdala or hippocampus is
`probably the best suited model for focal seizures,
`particularly complex focal seizures as occurring in
`temporal lobe epilepsy (Lo¨ scher and Schmidt, 1988),
`so that by use of these three models the major types
`of epileptic seizures are covered. As shown in
`
`Tables 2 and 3, valproate is e€ective in these three
`models, which reflects the wide spectrum of anticon-
`vulsant activity against di€erent types of seizures
`and epilepsy. In Table 5, the activity of valproate in
`these models is compared with respective activities
`of other ‘old’ or ‘first generation’ drugs, that is,
`drugs developed and introduced before 1970, and
`‘new’ drugs or ‘second generation’ drugs, that is,
`drugs developed
`and introduced
`after
`1970.
`Furthermore, the clinical spectrum of anticonvulsant
`activities is shown in Table 5. It can be seen that
`only few drugs compete with valproate in terms of
`its wide spectrum of anticonvulsant activity both
`preclinically and clinically,
`thus
`illustrating the
`unique profile of this drug.
`In addition to animal models of generalized or
`focal seizures, valproate also has been evaluated in
`models of status epilepticus. As shown by Ho¨ nack
`and Lo¨ scher (1992) in a mouse model of generalized
`convulsive (grand mal) status epilepticus, i.v. injec-
`tion of valproate was as rapid as benzodiazepines to
`suppress generalized tonic–clonic seizures, which
`was related to the instantaneous entry of valproate
`into the brain after this route of administration. We
`
`Table 4. Anticonvulsant potency of valproate in genetic animal models of epilepsy
`
`Model
`
`E(cid:129)cacy of valproate
`
`Species
`
`Seizure type
`
`Induction
`
`Route of
`administration
`
`b (mg kg(cid:255)1)
`ED50
`
`References (examples)
`
`Epileptic rats
`
`Epileptic gerbils
`
`Myoclonic
`Clonic/tonic
`Clonic/tonic
`Petit mal
`(spike/waves)
`Clonic/tonic
`Epileptic rats
`Clonic
`DBA/2 Mice
`Photosensitive baboons Myoclonic
`Epileptic dogs
`Tonic/clonic
`
`Air blast
`Air blast
`Air blast
`Spontaneous
`seizures
`Audiogenic
`Audiogenic
`Photic
`Spontaneous
`seizures
`
`p.o.
`p.o.
`i.p.
`i.p.
`
`210
`280
`73
`81
`
`Frey et al. (1983)
`Frey et al. (1983)
`Lo¨ scher et al. (1984)
`Lo¨ scher et al. (1984)
`
`115–150
`i.p.
`55–300
`i.p.
`200
`i.v.
`Ine€ective because of too short
`action
`
`Dailey and Jobe (1985)
`Lo¨ scher and Meldrum (1984)
`Lo¨ scher and Meldrum (1984)
`Lo¨ scher et al. (1985)
`
`aDose which protects 50% of animals from seizures.
`
`

`

`Valproate
`
`37
`
`Table 5. Anticonvulsant e€ect of old (first generation) and new (second generation) antiepileptic drugs against di€erent
`types of seizures in animal models and in human epilepsy
`
`Drug
`
`Anticonvulsant activity in experimental
`models
`
`Clinical e(cid:129)cacy
`
`MES test
`(mice or rats,
`tonic seizures)
`
`s.c. PTZ test
`(mice or rats,
`clonic seizures)
`
`Partial
`seizures
`
`Amygdala-
`kindling
`(rats, focal
`seizures)
`
`Generalized seizures
`
`Tonic–clonic Absence Myoclonic
`
`First generation drugs
`Valproate
`Carbamazepine
`Phenytoin
`Phenobarbital
`Primidone
`Benzodiazepines
`Ethosuximide
`
`Second generation drugs
`Lamotrigine
`Topiramate
`Oxcarbazepine
`Felbamate
`Vigabatrin
`Tiagabine
`Gabapentin
`NMDA antagonists
`
`+
`+
`+
`+
`+
`+
`NE
`
`+
`+
`+
`+
`NE
`NE
`2
`+
`
`+
`NE
`NE
`+
`+
`+
`+
`
`NE
`NE
`2
`+
`+
`+
`2
`2
`
`+
`+
`+
`+
`+
`+
`NE
`
`+
`+
`?
`+
`+
`+
`+
`NE
`
`+
`+
`+
`+
`+
`+
`NE
`
`+
`+
`+
`+
`+
`+
`+
`NE
`
`+
`+
`+
`+
`+
`+
`NE
`
`+
`+
`?
`+
`?
`?
`?
`?
`
`+
`NE
`NE
`NE
`NE
`+
`+
`
`+
`2
`NE
`2
`NE
`?
`NE
`?
`
`+
`NE
`NE
`+
`+
`+
`2
`
`+
`+
`NE
`+
`NE
`NE
`NE
`?
`
`E€ect is indicated by: +, e€ective; 2, inconsistent data; NE, not e€ective; ?, no data available (or found). MES, Maximal electroshock
`seizure; PTZ, pentylenetetrazole. Adapted from Lo¨ scher (1998a,b).
`
`formulation of
`therefore proposed that an i.v.
`valproate might be a useful alternative to other anti-
`epileptic drugs such as phenytoin as a nonsedative
`anticonvulsant
`for diazepam-resistant grand-mal
`status (Ho¨ nack and Lo¨ scher, 1992). In view of the
`di€erent mechanisms presumably involved in antic-
`onvulsant activity of valproate against di€erent sei-
`zure types (see below), the situation may be di€erent
`for other types of status epilepticus, because not all
`cellular e€ects of valproate occur rapidly after ad-
`ministration. This is substantiated by accumulating
`clinical experience with parenteral formulations of
`valproate in treatment of di€erent types (e.g. con-
`vulsive vs nonconvulsive) of status epilepticus. In a
`monkey model for status epilepticus, in which a sta-
`tus of focal and secondarily generalized tonic–clonic
`seizures was induced by focal brain injection of alu-
`minia gel combined with systemic administration of
`4-deoxypyridoxine, i.v. administration of valproate
`delayed seizures but did not prevent their occurrence
`(Lockard et al., 1983). In rats with cortical cobalt
`lesions injected with homocysteine thiolactone to
`induce a status of secondarily generalized tonic–clo-
`nic seizures, i.p. injection of valproate blocked sei-
`zures, although only at
`relatively high doses
`(Walton and Treiman, 1992).
`In addition to the acute short-lasting anticonvul-
`sant e€ects of valproate in diverse animal models
`after single dose administration (Tables 2–5), several
`studies have examined the anticonvulsant e(cid:129)cacy of
`valproate during chronic administration. During the
`first days of treatment of amygdala-kindled rats, a
`marked increase in anticonvulsant activity was
`observed which was not related to alterations in
`brain or plasma drug or metabolite levels (Lo¨ scher
`
`et al., 1988a, 1989). Similarly, when anticonvulsant
`activity was measured by means of timed i.v. infu-
`sion of PTZ, prolonged treatment of mice with
`valproate resulted in marked increases in anticonvul-
`sant activity on the second day of treatment and
`thereafter compared to the acute e€ect of valproate,
`although plasma levels measured at each seizure
`threshold determination did not di€er significantly
`(Lo¨ scher and Ho¨ nack, 1995). This ‘late e€ect’ of
`valproate developed irrespective of
`the adminis-
`tration protocol (once per day, three times per day,
`continuous infusion) used for treatment with valpro-
`ate in the animals. Such an increase in anticonvul-
`sant activity during chronic treatment was also
`observed in epileptic patients and should be con-
`sidered when acute anticonvulsant doses or concen-
`trations of valproate in animal models are compared
`to e€ective doses or concentrations in epileptic
`patients during chronic treatment (Table 6). In other
`words, doses or plasma levels being ine€ective after
`acute administration can become e€ective during
`chronic administration. The possible mechanisms
`involved in ‘early’ (i.e. occurring immediately after
`first administration of an e€ective dose) and ‘late’
`(i.e. developing during chronic administration)
`anticonvulsant e€ects of valproate will be discussed
`later in this review. In this respect, it is important to
`note that early and late e€ects of valproate have
`been also observed in in vitro preparations (Altrup
`et al., 1992).
`All e€ects of valproate in animal models of sei-
`zures or epilepsy summarized so far were acute or
`chronic anticonvulsant e€ects. In fact, all currently
`available drugs are anticonvulsant
`(antiseizure)
`rather than antiepileptic. Accurately, the latter term
`
`

`

`38
`
`W. Lo¨ scher
`
`Table 6. Concentrations of valproate in plasma and brain after administration ofanticonvulsant doses
`
`Mouse
`TICa
`50
`
`Humans
`during chronic oral treatment
`(average doses 15–20 mg kg(cid:255)1)
`
`Plasma
`
`Brain
`
`Plasma
`
`Brain
`
`mg ml(cid:255)1
`
`mMol
`
`mg g(cid:255)1
`
`mMol
`
`mg ml(cid:255)1
`
`mMol
`
`120–150
`
`830–1040
`
`25–40
`
`170–280
`
`40–100
`
`280–690
`
`mg g(cid:255)1
`
`6–27b
`
`mMol
`
`42–190
`
`aConcentration which increases the threshold for clonic PTZ seizures by 50%; determined 5 min after doses (TID50) of 80–100 mg kg(cid:255)1
`i.p. (Lo¨ scher and Fiedler, unpublished data).bVajda et al. (1981).
`
`should only be used for drugs which prevent or treat
`epilepsy and not solely its symptoms. Traditionally,
`pharmacological strategies for treatment of epilepsy
`have aimed at seizure initiation and propagation
`rather than the processes leading to epilepsy. As a
`result, none of the currently available antiepileptic
`drugs clinically evaluated in this respect (e.g. pheny-
`toin and carbamazepine) is capable of preventing
`epilepsy, for example, after brain injury (Hernandez,
`1997). Furthermore,
`there is increasing evidence
`that—despite early onset of treatment and suppres-
`sion of seizures—antiepileptic drugs do not a€ect
`the progression or underlying natural history of epi-
`lepsy (Shinnar and Berg, 1996). Thus, one important
`goal for the future will be to develop antiepilepto-
`genic drugs, that is, drugs which prevent or tr