- Email: [email protected]
Anticonvulsant valproate reduces seizure-susceptibility in mutant Drosophila
Brain Research 958 (2002) 36–42 www.elsevier.com / locate / brainres
Research report
Anticonvulsant valproate reduces seizure-susceptibility in mutant Drosophila Daniel Kuebler a , Mark Tanouye a,b , * a
b
Department of Molecular and Cell Biology, Division of Neurobiology, University of California, Berkeley, CA 94720, USA Department of Environmental Science, Policy, and Management, Division of Insect Biology, University of California, Berkeley, CA 94720, USA Accepted 12 August 2002
Abstract Despite the frequency of seizure disorders in the human population, the genetic basis for these defects remains largely unclear. Currently, only a fraction of the epilepsies can be linked conclusively to a genetic determinant. In addition, a significant number of epileptics do not respond to the current anticonvulsant therapies. We have turned to Drosophila as a model to address these problems and have identified genetic mutants that are more sensitive to seizures, bang-sensitive (BS) mutants, such as slamdance (sda), bangsenseless (bss) and easily shocked (eas), as well as mutants that are resistant to seizures, such as paralytic, maleless napts , shaking-B 2 and Shaker. Here, we have developed a new method for evaluating compounds with anticonvulsant activity. The methodology uses Drosophila BS mutants to assay the ability of compounds to suppress the seizure susceptible phenotype normally seen in the BS mutants. To test the effectiveness of this method, two BS mutant strains were administered the anticonvulsant valproate and in both cases the drug was able to suppress seizures. The Drosophila system provides a potentially powerful way of developing and testing new drugs with anticonvulsant properties. 2002 Elsevier Science B.V. All rights reserved. Theme: Neurotransmitters, modulators, transporters and receptors Topic: Behavioral pharmacology Keywords: Seizure; Drosophila; Valproate; Seizure suppression; Genetics of seizure disorders; Epilepsy
1. Introduction Genetic and molecular analyses of Drosophila mutants have provided a model for examining fundamentally important problems in biology, particularly in developmental biology and neurobiology. An important lesson from these studies is that fundamental processes and many of the essential gene products are conserved across species. Thus, findings are generally applicable to other biological systems such as mouse and human biology. One implication from this cross-species conservation is that Drosophila has the potential to be an attractive system for developing and evaluating new drug therapies to treat human *Corresponding author. Department of Environmental Science, Policy, and Management, 201 Wellman Hall, University of California, Berkeley, CA 94720, USA. Tel.: 11-510-642-9404; fax: 11-510-643-6791. E-mail address: [email protected] (M. Tanouye).
pathologies. The same features that serve well for mutant isolation [8], could potentially be applied to advantage in drug testing. Flies are small in size allowing easy manipulation and the rapid testing of large populations of subjects. Suppressor mutations may be used to identify new drug targets (see [13], for example). Feeding methods are welldeveloped for mutagens such as ethylmethanesulfonate and can be adapted for testing candidate drugs. In the case of neurological syndromes, behavioral tests and electrophysiological assays are well-developed allowing additional testing to determine drug mode of action. Drug effects may be examined in a variety of different genotypes by manipulating genetic backgrounds. Finally, there are no major animal welfare concerns associated with testing of drugs in Drosophila. Given the advantages of using Drosophila, it has the potential to be a highly effective tool in developing drug therapies for human conditions such as epilepsy where
0006-8993 / 02 / $ – see front matter 2002 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 02 )03431-5
D. Kuebler, M. Tanouye / Brain Research 958 (2002) 36–42
there is a critical unmet need for new drugs. Seizures are a serious problem in many human neuropathologies: they are a major symptom associated with fever, mechanical trauma, electroconvulsive shock, drugs, alcohol, tumors, and especially epilepsy. The human epilepsies are a group of about 40 seizure disorders affecting 1% of the US population, about 2.5 million people [3,10,24,25,28,34]. Epilepsy is a chronic neurological disorder characterized by repeated, spontaneous seizures. Although drug therapy is usually effective, for many patients there are sideeffects, some patients only respond to a combination of drugs, and a significant portion, approximately 20–30%, have intractable epilepsy, i.e. they still suffer spontaneous seizures despite drug treatment. New, more effective anticonvulsants with reduced side-effects are needed to combat intractable epilepsy. Although the potential is promising and the need is evident, there are two important prerequisites that must be satisfied in order for Drosophila to be an effective tool in developing drug therapies for seizure disorders: (1) a good Drosophila model of the human pathology and (2) some assurance that currently available human treatments are effective in Drosophila. In this paper, we examine the effect of the anticonvulsant valproate on bang-sensitive (BS) paralytic mutants, a Drosophila model for human seizure disorders [12]. The BS mutant class includes several mutants such as bangsenseless (bss), easily shocked (eas), slamdance (sda) and technicalknockout (tko). All BS mutants suffer from cycles of intense behavioral hyperactivity and temporary paralysis caused by a mechanical shock, such as a tap of the culture vial on the bench top or brief vortex mixing (a ‘bang’) [2,6]. The hyperactivity phenotype is characterized by intense, uncoordinated motor activity including wing flapping, leg shaking, and abdominal muscle contractions; the paralytic phenotype, on the other hand, is observed as a cessation of all physical activity [2,26]. The hyperactivity and paralysis can be mimicked on the electrophysiological level by stimulating and recording from the central nervous system (CNS) [12,27]. These analyses show that the BS hyperactivity is due to neurological seizure and that the mutants are five to ten times more sensitive to evoked seizure than wild type flies. Paralysis is due to a failure of chemical synaptic transmission in several neural circuits. Thus, Drosophila BS mutants are a potentially useful model for human seizure-sensitivity, although the model is limited by a lack of some of the complex neural networks that may play an important role in mammalian epileptigenesis [24]. Valproate (sodium valproate, valproic acid) is a shortbranched fatty acid that is the most widely used antiepileptic drug for the treatment of generalized and partial epilepsy [11,19]. Its antiepileptic effectiveness was discovered serendipitously when it was used as a carrier for testing other hydrophobic compounds in experimental animal models of epilepsy. Diverse forms of epilepsy have been found to respond to valproate and this has been taken
37
to indicate that the drug may act to control seizures through multiple targets. One mechanism of seizure control appears to be through a valproate effect on voltagegated sodium channels. Valproate causes a decrease in peak sodium conductance that is voltage-dependent and delays recovery from inactivation [33; see however, Ref. 1]. These effects appear to limit high frequency firing of neurons thereby interfering with the ability to generate and propagate seizures [23,31]. Valproate also appears to interfere with seizures by increasing the amount of synaptic inhibition by g-aminobutyric acid (GABA). Valproate rapidly increases total GABA levels in the rodent brain with significant changes observed within 5 min and overall increases reaching 15–45% [7,9,14,20]. Valproate is believed to alter GABA levels through both the stimulation of GABA biosynthetic enzymes and the inhibition of enzymes involved in GABA degradation. In this study, we show that valproate is able to suppress seizure in BS flies in a dosage-dependent manner and that the drug does not appear to affect the firing threshold of individual neurons. This is the first report that we are aware of that examines the effects of valproate on the Drosophila nervous system. Further investigations may be able to determine the mode of action for valproate effect in Drosophila and assess if it is mechanistically homologous to the human anticonvulsant effectiveness. Also, these investigations may serve as foundation for the use of Drosophila in the screening and evaluation of novel anticonvulsant agents.
2. Materials and methods
2.1. Animals Normal wild type Drosophila were the Canton Special (CS) strain. The bang senseless (bss) and slamdance(sda) mutants are bang-sensitive (BS) paralytics that are 5–10 times more susceptible to seizures than normal flies. The mutant behavioral phenotypes of seizure and paralysis, the corresponding electrophysiological phenotypes of seizure and synaptic failure, and the threshold for seizure susceptibility of bss and sda alleles compared to normal have been described [6,12,13,26,27]. The sda gene encodes a protein homologous to human aminopeptidase N [21,35]. The bss gene product has not been described.
2.2. Electrophysiology Electrophysiology experiments on unanesthetized intact male flies were performed as described previously [12,30]. For experiments involving injected solutions, the fly was mounted to an insect pin and then while affixed, moved to a petri dish containing a piece of packaging foam (1.030.5 in.; 1 in.52.54 cm) glued in the middle. The fly was placed dorsal-side up on the foam and the pin was stabilized with
38
D. Kuebler, M. Tanouye / Brain Research 958 (2002) 36–42
plasticine. A small incision was made through the head cuticle between the interocellar setae [5] using a small dissecting knife (Beaver姠 1.5 mm). Hemolymph solution (JR solution) was injected into the head through the incision using a glass micropipette (tip diameter 0.5 mm). Enough JR solution was injected to cause the proboscis to extend. Particular care was taken to avoid over-injection of JR solution which could cause damage by rupture. Any flies in which the proboscis failed to extend or showed damage by rupture were discarded. The composition of JR solution was (in mM): 100 NaCl, 25 KCl, 6 CaSO 4 , 10 MgSO 4 , 4 NaHCO 3 , 1 NaH 2 PO 4 , 5 trehalose, 75 sucrose, 5 HEPES at pH 7.2 [32]. JR solution containing valproate (Na-valproate, Sigma) was made fresh before each experiment. Valproate concentrations are indicated in the text and refer to the value of Na-valproate. Three minutes after injection of JR solution, bipolar tungsten stimulating electrodes were inserted into the head of the fly as described previously [12]. Single-pulse stimuli (0.2 ms duration) were delivered to the brain to drive the giant fiber (GF) and GF-driven muscle potentials were recorded in the dorsal longitudinal muscles (DLM) using tungsten recording electrodes. GF thresholds were determined as the lowest voltage at which the short latency GF pathway responded. Five minutes following the injection, a single high frequency electrical stimulus (HFS) wavetrain with parameters 0.5 ms pulses at 200 Hz for 400 ms was delivered to the brain at the voltage indicated to elicit seizures. Previously, we have shown that seizures are elicited in an all-or-nothing manner [12]. Seizures consist of high frequency activity in at least seven different muscle groups and over thirty muscle fibers in the thorax. This activity in each muscle fiber corresponds to seizure activity in the motoneuron that innervates it. In the present paper, recordings of DLM muscle potentials were used to denote the occurrence of seizures as described previously [12]. The seizure threshold graph found in Fig. 2 was generated by calculating the percentage of flies that seized at each HF stimulus intensity for a particular genotype. The number of animals examined was n . 20 for each experimental group tested. The midpoint of the curve was taken as the seizure threshold for that strain and the slope of the curve indicates the level of variability within each genotype. For control animals, the values obtained by this method show good agreement to those described previously [12,13].
3. Results
3.1. Seizure susceptibility in BS and wild type Drosophila In intact Drosophila, seizures are elicited by high frequency electrical stimuli (HFS) delivered to the brain (Fig. 1) [12,13,27]. These evoked seizures are character-
Fig. 1. Seizures in intact sda and CS flies. The BS mutant sda fly is more susceptible to seizures than the wild type CS fly and therefore has a much lower seizure threshold. (A) A seizure is elicited in a sda fly by a high frequency stimulus of low strength (8 V) and displayed at a high sweep speed. The HF stimulus (labeled HF) is a short wavetrain (0.5 ms pulses at 200 Hz for 300 ms) of electrical stimuli delivered to the brain. Recording is from a DLM muscle fiber and reflects the activity of the single DLM motoneuron that innervates it. The ‘seizure’ (labeled SEI) is abnormal high frequency firing of the DLM motoneuron. The seizure is widespread as similar activity can be found in recordings from seven different muscle groups in the fly following HF stimulation (Kuebler and Tanouye [12]). (B) A low-voltage HF stimulus of 8 V fails to elicit a seizure in a wild type CS fly because the stimulus is below the seizure threshold. Following the HF stimulus artifact, there is no seizure activity observed in this recording displayed at a high sweep speed. Note also that there is no period of synaptic failure and single-pulse stimulation of the GF (0.5 Hz) continues to evoke DLM potentials. Two such effective single-pulse stimuli are depicted in this trace; each was effective in evoking a DLM potential. (C) A seizure is elicited in a wild type CS fly by a high voltage HF stimulus (30 V) which is above the threshold for seizure. The seizure in this recording begins within the large stimulus artifact and is displayed at a high sweep speed. (D) Same recording as for (A) from a sda fly displayed at a slow sweep speed. In this recording, the HF stimulus and seizure (SEI) are followed by a quiescent period (labeled SYNAPTIC FAILURE) that is characterized by synaptic failure within the GF circuit (Pavlidis and Tanouye [27]). During this period, there are stimulus artifacts (downward-going) from continuous singlepulse stimulation of the GF (0.5 Hz), but no evoked DLM potentials. Spontaneous activity (labeled SPON) or ‘recovery seizure’ appears as additional seizure-like activity occurring just after the synaptic failure period and just prior to recovery (recovery not evident in this trace). Vertical calibration bar is 20, 40, 40 and 10 mV for (A), (B), (C) and (D), respectively. Horizontal calibration bar is 300 ms, 1.2 s, 1.2 s, and 1.5 s for (A), (B) (C) and (D), respectively (figure modified from Kuebler et al. [13], Fig. 1).
D. Kuebler, M. Tanouye / Brain Research 958 (2002) 36–42
ized by aberrant high frequency spiking in a large population of motoneurons and muscles throughout the fly. For normal Canton-Special (CS) males, an HFS of 0.5 ms pulses at 200 Hz for 300 ms causes an all-or-nothing seizure at a threshold voltage of 30.163.8 V. The bangsensitive (BS) paralytic mutants are 5–10 times more sensitive to seizures than normal flies: the mutants bss and sda have seizure thresholds to HFS of only 3.260.6 and 6.260.8 V, respectively. The seizures are followed by a period of quiescence that is due to synaptic failure in many chemical synapses and an additional seizure just prior to recovery (Fig. 1) [12,27]. This quiescent period is apparently responsible for behavioral paralysis in BS mutant flies [27]. Immediately following a bout of seizure–quiescence–recovery seizure, there is a transient increase in seizure threshold lasting several minutes that contributes to a refractory period [12]. We examined the effects of fly surgery and hemolymph solution injection on excitability and seizure initiation. Our general observations were that stable electrical recordings could be maintained for about 10 min. However, longer term recordings were unreliable because of spontaneous excitability changes as determined by threshold changes in evoked GF responses to single pulses and HFS-evoked seizures. This necessitated a change in our general protocol. For each fly undergoing surgery, the GF threshold was determined initially by single pulse stimulation at 3 min after surgery, then a single HFS (0.5 ms pulses at 200 Hz for 400 ms) was delivered at a preset voltage at 5 min after surgery. The preparation was then discarded because of uncertainties due to refractory period and postsurgery changes in threshold. The seizure threshold curves found in Fig. 2 indicate the level of seizure susceptibility in CS, bss and sda strains. The rank order of the genotypes is identical to that seen previously in studies that did not employ surgery and injection: bss was the most susceptible to seizure, CS was the least susceptible, and sda was intermediate, although closer to bss than CS. The apparent seizure thresholds for bss, sda, and CS are about 6, 10 and 40 V, respectively. These values are similar to, although slightly higher than, those found for intact flies.
39
Fig. 2. Surgery and hemolymph injection do not greatly affect the seizure threshold. A single HFS stimulus of a fixed voltage was delivered to each fly 5 min after surgery and hemolymph injection at the voltage indicated and its effectiveness in generating a seizure noted. The curves depicted here show the percentage of flies that have a seizure (n . 20 flies for all genotypes). The apparent seizure threshold for CS is about 40 V, the voltage where seizure was initiated in 50% of flies. The apparent seizure thresholds for bss and sda are about 6 and 10 V, respectively.
V indicates the ability to suppress seizures to wild type levels. We found that 5 mM valproate suppressed seizures in most sda flies and 10 mM completely suppressed seizures (Fig. 3). In the case of bss, 10 mM valproate suppressed seizures in most flies and 25 mM valproate completely suppressed seizures. Given what is known about valproate [19,11], it is likely
3.2. Valproate suppresses seizures in bss and sda Valproate is a clinically-used antiepileptic drug that is effective for treating both generalized and partial seizures [11,19]. Valproate has also been shown to be effective in suppressing seizures in a variety of animal models. Here, we found valproate can suppress the seizure susceptible defect normally seen in the bss and sda mutant strains. We tested a range of increasing concentrations of valproate on both strains and assayed the susceptibility to an HFS wavetrain of fixed voltage (40 V, 0.5-ms pulses at 200 Hz for 400 ms). This voltage was chosen because it is about the seizure threshold for CS wild type flies. Thus, the ability of a given concentration to suppress seizures at 40
Fig. 3. Valproate reduces seizure susceptibility in bss and sda flies. A single HFS stimulus of 40 V (0.5 ms pulses at 200 Hz for 400 ms) was delivered to each fly 5 min after surgery and hemolymph plus valproate injection and its effectiveness in generating a seizure noted. The curves depicted here show the percentage of flies that have a seizure for various concentrations of valproate as indicated (n . 20 flies for each genotype).
40
D. Kuebler, M. Tanouye / Brain Research 958 (2002) 36–42
Fig. 4. Giant fiber threshold is not affected by valproate. Single pulse stimuli (0.2 ms duration, 0.8 Hz) were delivered to each fly 3 min after surgery and hemolymph plus valproate injection and the GF threshold determined. Thresholds remained the same even at the highest drug concentration.
that it exerts its anticonvulsant effect in flies by a variety of mechanisms. One possibility is that valproate may alter seizure susceptibility by altering the thresholds of individual neurons. We examined the giant fiber (GF) neuron threshold at the various concentrations of valproate in both sda and bss and found that it was not affected (Fig. 4). Even at the highest valproate concentrations tested bss had a threshold of 2.7860.34 V at 25 mM and sda had a threshold of 2.6460.33 V at 50 mM; both of which are not significantly different from the GF threshold in the absence of valproate that was for bss and sda 2.4260.37 and 2.3560.40 V, respectively. Thus, at the concentrations we examined, valproate was able to raise the seizure threshold while the single neuron excitability, at least as indicated by GF threshold, remained unchanged. Although the GF threshold was unaffected, we cannot rule out the possibility that other unidentified neurons, critical to the generation of seizures, had significant alterations in excitability.
4. Discussion Although epilepsy affects over 1% of the US population, an understanding of the underlying mechanisms is in the vast majority of cases lacking. One reason that reaching an understanding of the etiology of human seizure disorders has been difficult is that there is such a variety of both causes and types of seizures. Dissecting the pathogenesis of seizure disorders is extremely difficult and is compli-
cated by the heterogeneity of defects as well as the vast array of molecular lesions known to be involved in both human and mouse seizure disorders. The complexity and hetereogeneity of epilepsy is also evident by the number and apparent functional diversity of anticonvulsants used to treat the disorder. Although drug therapy is often successful, some patients only respond to a combination of drugs and there are a substantial number of epilepsy cases that are intractable, unresponsive to drug treatments [17,18]. This is despite the emergence of second generation anticonvulsants such as lamotrigine, vigabatrin, gabapentin and topiramate [4]. What makes the cases of intractable epilepsy particularly intriguing is that often people with identical symptoms can have completely different responses to the same drug [29]. The reasons behind this are unclear, but it demonstrates both the extreme heterogeneity of these disorders and the need for the development of new, more effective anticonvulsants to combat them. An especially attractive approach for developing novel anticonvulsants is to take advantage of Drosophila seizuresuppressor mutations. We have shown previously that certain genetic mutations can both elevate the seizure threshold in Drosophila and suppress the seizure-susceptible phenotype seen in BS mutants [13]. All of these mutants, in double-mutant combinations with BS mutants, provided a genetic background that elevated the seizure threshold to values above those seen in the BS mutants alone. In each case in which the BS seizure susceptible phenotype was altered, a single mutation, which becomes a possible target for the development of novel anticonvulsants, is responsible. Further identification of additional seizure-suppressor mutations through genetic screens could provide an extremely rich source that could serve as a basis for a an extensive drug development program. This study is the first description of a method for testing the ability of an anticonvulsant to suppress seizures in Drosophila. By injecting drug into the head cavity and immediately measuring the effect on seizure susceptibility, we are able to measure the effectiveness of valproate, a commonly-used anticonvulsant in humans, to rescue the seizure-sensitive phenotype that is characteristic of Drosophila BS mutants. The ability to suppress seizures in the BS strains was found to vary depending on genotype. Higher concentrations were required to suppress bss mutants, the more susceptible of the two genotypes, than were required for sda mutants. We have previously suggested that seizures in Drosophila are due to an interplay between excitatory or seizure-initiating neuronal circuits and inhibitory or seizure-suppression circuits [12]. The excitatory circuits may contain circuit elements that are linked through reciprocal excitatory synapses that provide positive feedback, as described for mammalian seizure models [24]. Similarly, the inhibitory circuits may contain circuit elements that feed back inhibition through presynaptic or postsynaptic synapses. Seizure control in
D. Kuebler, M. Tanouye / Brain Research 958 (2002) 36–42
Drosophila mutants by valproate could be through a valproate effect on voltage-gated sodium channels [33]. Indeed, we have previously shown that mutations affecting voltage-gated sodium channels can act as seizure-suppressor mutations in double mutant combinations with BS mutations [13]. A less likely possibility is that valproate is increasing the amount of synaptic inhibition by GABA [7,9,14,20]. GABA is a major inhibitory transmitter in Drosophila, however, in the short drug application times used in the present experiments, it is unlikely that biosynthetic or degradation enzymes are causing much change in GABA levels. Suppression of seizures by valproate is one way Drosophila seizures are similar to mammalian seizures. This adds to previous observations [12,13] that have shown several similarities between seizures in mammals and Drosophila. For example, just as in mammals, seizures in Drosophila exhibit the following characteristics: all individuals have a seizure threshold; seizure susceptibility can be modulated by genetic mutations; electroconvulsive therapy in flies raises the threshold for subsequent seizures; seizures in flies can be spatially segregated into particular regions of the central nervous system; and seizures spread through the nervous system along particular pathways that are dependent on functional synaptic connections and recent electrical activity. Although there are, at present, several superficial similarities between Drosophila and mammalian seizures, there has not yet been a direct demonstration of identical mechanistic homology underlying seizure-susceptibility. Additional study is required to determine the extent of this homology and validate further the utility of this Drosophila model. Many anticonvulsant drugs have, unlike valproate, a specific mode of action [15,16,22,23]. Thus, future testing of different standard anticonvulsant drugs with different modes of action on a variety of different BS mutants would be especially valuable for defining the predictivity for special types of seizures. This would be particularly important in drug development to exclude the possibility that only drugs with a special mechanism are selected. In addition, the specificity of certain drugs can help decipher the processes involved in the BS defect. Also, elucidating different anticonvulsant profiles for the BS strains may indicate underlying differences in the pathogenesis of seizures in different BS mutants and could help in understanding the specific mechanisms that lead to seizure susceptibility in each specific mutant. Finally, the ability to screen multiple compounds makes the system amenable for screening for novel anticonvulsants. Because a currently used anticonvulsant, valproate, has been demonstrated to be effective in suppressing the BS phenotype, the identification of clinically-relevant compounds appears plausible by this method. The technique we have described here has many potential uses and should open lines of research that were previously inaccessible.
41
Acknowledgements The authors thank Sally Faulhaber for assistance in the maintenance of Drosophila stocks. We thank Drs. Jeremy Lee and Charlie Oh for discussion throughout this project and Drs. David Bentley and Geoff Owen for their insightful comments on the manuscript. Part of this work was supported by USPHS grant NS31231 and an Epilepsy Foundation grant to MAT.
References [1] H. Albus, R. Williamson, Electrophysiologic analysis of the actions of valproate on pyramidal neurons in the rat hippocampal slice, Epilepsia 39 (1998) 124–139. [2] S. Benzer, From the gene to behavior, J. Am. Med. Assoc. 218 (1971) 1015–1022. [3] Commission on Classification and Terminology of the International League Against Epilepsy, Proposal for revised classification of epilepsies and epileptic syndromes, Epilepsia 30 (1989) 389–399. [4] S. Czuczwar, P. Patsalos, The new generation of GABA enhancers. Potential in the treatment of epilepsy, CNS Drugs 15 (2001) 339– 350. [5] G.F. Ferris, External morphology of the adult, in: M. Demerec (Ed.), Biology of Drosophila, Hafner, New York, 1950, pp. 368–418. [6] B. Ganetzky, C.-F. Wu, Indirect suppression involving behavioral mutants with altered nerve excitability in Drosophila melanogaster, Genetics 100 (1982) 597–614. [7] Y. Godin, L. Heier, J. Mark, P. Mandel, Effects of di-n-propylacetate, an anticonvulsant compound on GABA metabolism, J. Neurochem. 16 (1969) 869–873. [8] R. Greenspan, Fly Pushing: The Theory and Practice of Drosophila Genetics, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1997. [9] C. Hariton, S. Ciesielsky, S. Simler, M. Valli, G. Jadot, S. Gobaille, E. Mesdijan, P. Mandel, Distribution of sodium valproate and GABA metabolism in CNS of the rat, Biopharm. Drug Dispos. 5 (1984) 409–414. [10] W.A. Hauser, D.C. Hesdorffer, Epilepsy: Frequency, Causes, and Consequences, Demos, New York, 1990. [11] C.U. Johannessen, Mechanisms of action of valproate: a commentatory, Neurochem. Intl. 37 (2000) 103–110. [12] D. Kuebler, M.A. Tanouye, Modifications of seizure susceptibility in Drosophila, J. Neurophysiol. 83 (2000) 998–1009. [13] D. Kuebler, H.G. Zhang, X. Ren, M.A. Tanouye, Genetic suppression of seizure susceptibility in Drosophila, J. Neurophysiol. 86 (2001) 1211–1225. [14] K. Kukino, T. Deguchi, Anticonvulsant activity and effects of sodium dipropylacetate on cerebral 5-hydroxytryptamine and gaminobutyric acid in reserpinized mice, Chem. Pharmacol. Bull. 26 (1978) 3551–3555. [15] C.C. Kuo, A common anticonvulsant binding site for phenytoin, carbamazepine, and lamotrigine in neuronal Na 1 channels, Mol. Pharmacol. 54 (1998) 712–721. [16] C.C. Kuo, R. S Chen, L. Lu, R.C. Chen, Carbamazepine inhibition of neuronal Na 1 currents: quantitative distinction from phenytoin and possible therapeutic implications, Mol. Pharmacol. 51 (1997) 1077–1083. [17] I.E. Leppik, Intractable epilepsy: prognostic evaluation, in: W.H. Theodore (Ed.), Surgical Treatment of Epilepsy, Elsevier, Amsterdam, 1992, pp. 7–11.
42
D. Kuebler, M. Tanouye / Brain Research 958 (2002) 36–42
[18] W. Loscher, Animal models of intractable epilepsy, Prog. Neurobiol. 53 (1997) 239–258. [19] W. Loscher, Valproate: a reappraisal of it pharmacodynamic properties and mechanism of action, Prog. Neurobiol. 58 (1999) 31–59. [20] W. Loscher, M. Vetter, Relationship between drug-induced increases of GABA levels in discrete brain areas and different pharmacological effects in rats, Biochem. Pharmacol. 33 (1985) 1907–1914. [21] A. Look, R. Ashmun, L. Shapiro, S. Peiper, Human myeloid plasma membrane glycoprotein CD13 (gp150) is identical to aminopeptidase N, J. Clin. Invest. 83 (1989) 1299–1307. [22] M.J. McLean, R.L. MacDonald, Multiple actions of phenytoin on mouse spinal cord neurons in cell culture, J. Pharmacol. Exp. Ther. 227 (1983) 779–789. [23] M.J. McLean, R.L. MacDonald, Sodium valproate, but not ethosuximide, produces use- and voltage-dependent limitation of high frequency repetitive firing of action potentials of mouse central neurons in cell culture, J. Pharmacol. Exp. Ther. 237 (1986) 1001– 1011. [24] J.O. McNamara, Cellular and molecular basis of epilepsy, J. Neurosci. 14 (1994) 3413–3425. [25] J.O. McNamara, Emerging insights into the genesis of epilepsy, Nature 399 (Suppl.) (1999) A15–A22. [26] P. Pavlidis, M. Ramaswami, M.A. Tanouye, The Drosophila easily shocked gene: a mutation in a phospholipid pathway causes seizure, neuronal failure, and paralysis, Cell 79 (1994) 23–33. [27] P. Pavlidis, M.A. Tanouye, Seizures and failures in the giant fiber
[28]
[29]
[30] [31]
[32]
[33] [34] [35]
pathway of Drosophila bang-sensitive paralytic mutants, J. Neurosci. 15 (1995) 5810–5819. R.S. Puranam, J.O. McNamara, Seizure disorders in mutant mice: relevance to human epilepsies, Curr. Opin. Neurobiol. 9 (1999) 281–287. D. Schmidt, Diagnostic and therapeutic management of intractable epilepsy, in: D. Schmidt, P.L. Morselli (Eds.), Intractable Epilepsy: Experimental and Clinical Aspects, Raven Press, New York, 1986, pp. 237–258. M.A. Tanouye, R.J. Wyman, Motor outputs of the giant nerve fiber in Drosophila, J. Neurophysiol. 44 (1980) 405–421. A.B. Thomas, G.C. Mathews, J.A. Ferrendelli, Mechanisms of action of antiepileptic drugs, in: E. Wyllie (Ed.), The Treatment of Epilepsy: Principles and Practice, 2nd Edition, Williams and Wilkins, Baltimore, MD, 1996, pp. 700–711. J.R. Trimarchi, R.K. Murphey, The shaking-B 2 mutation disrupts electrical synapses in a flight circuit in adult Drosophila, J. Neurosci. 17 (1997) 4700–4710. R.J. Van den Berg, P. Kok, R.A. Voskuyl, Valproate and Na 1 currents in cultured hippocampal cells, Exp. Brain Res. 93 (1993) 279–287. L. Walton, in: Essentials of Neurology, 6th Edition, Churchill Livingstone, London, 1989, pp. 77–86. H.G. Zhang, J. Tan, E. Reynolds, D. Kuebler, S. Faulhaber, M.A. Tanouye, The Drosophila slamdance gene: a mutation in an aminopeptidase can cause seizure, paralysis and neuronal failure, Genetics, in press.
Research report
Anticonvulsant valproate reduces seizure-susceptibility in mutant Drosophila Daniel Kuebler a , Mark Tanouye a,b , * a
b
Department of Molecular and Cell Biology, Division of Neurobiology, University of California, Berkeley, CA 94720, USA Department of Environmental Science, Policy, and Management, Division of Insect Biology, University of California, Berkeley, CA 94720, USA Accepted 12 August 2002
Abstract Despite the frequency of seizure disorders in the human population, the genetic basis for these defects remains largely unclear. Currently, only a fraction of the epilepsies can be linked conclusively to a genetic determinant. In addition, a significant number of epileptics do not respond to the current anticonvulsant therapies. We have turned to Drosophila as a model to address these problems and have identified genetic mutants that are more sensitive to seizures, bang-sensitive (BS) mutants, such as slamdance (sda), bangsenseless (bss) and easily shocked (eas), as well as mutants that are resistant to seizures, such as paralytic, maleless napts , shaking-B 2 and Shaker. Here, we have developed a new method for evaluating compounds with anticonvulsant activity. The methodology uses Drosophila BS mutants to assay the ability of compounds to suppress the seizure susceptible phenotype normally seen in the BS mutants. To test the effectiveness of this method, two BS mutant strains were administered the anticonvulsant valproate and in both cases the drug was able to suppress seizures. The Drosophila system provides a potentially powerful way of developing and testing new drugs with anticonvulsant properties. 2002 Elsevier Science B.V. All rights reserved. Theme: Neurotransmitters, modulators, transporters and receptors Topic: Behavioral pharmacology Keywords: Seizure; Drosophila; Valproate; Seizure suppression; Genetics of seizure disorders; Epilepsy
1. Introduction Genetic and molecular analyses of Drosophila mutants have provided a model for examining fundamentally important problems in biology, particularly in developmental biology and neurobiology. An important lesson from these studies is that fundamental processes and many of the essential gene products are conserved across species. Thus, findings are generally applicable to other biological systems such as mouse and human biology. One implication from this cross-species conservation is that Drosophila has the potential to be an attractive system for developing and evaluating new drug therapies to treat human *Corresponding author. Department of Environmental Science, Policy, and Management, 201 Wellman Hall, University of California, Berkeley, CA 94720, USA. Tel.: 11-510-642-9404; fax: 11-510-643-6791. E-mail address: [email protected] (M. Tanouye).
pathologies. The same features that serve well for mutant isolation [8], could potentially be applied to advantage in drug testing. Flies are small in size allowing easy manipulation and the rapid testing of large populations of subjects. Suppressor mutations may be used to identify new drug targets (see [13], for example). Feeding methods are welldeveloped for mutagens such as ethylmethanesulfonate and can be adapted for testing candidate drugs. In the case of neurological syndromes, behavioral tests and electrophysiological assays are well-developed allowing additional testing to determine drug mode of action. Drug effects may be examined in a variety of different genotypes by manipulating genetic backgrounds. Finally, there are no major animal welfare concerns associated with testing of drugs in Drosophila. Given the advantages of using Drosophila, it has the potential to be a highly effective tool in developing drug therapies for human conditions such as epilepsy where
0006-8993 / 02 / $ – see front matter 2002 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 02 )03431-5
D. Kuebler, M. Tanouye / Brain Research 958 (2002) 36–42
there is a critical unmet need for new drugs. Seizures are a serious problem in many human neuropathologies: they are a major symptom associated with fever, mechanical trauma, electroconvulsive shock, drugs, alcohol, tumors, and especially epilepsy. The human epilepsies are a group of about 40 seizure disorders affecting 1% of the US population, about 2.5 million people [3,10,24,25,28,34]. Epilepsy is a chronic neurological disorder characterized by repeated, spontaneous seizures. Although drug therapy is usually effective, for many patients there are sideeffects, some patients only respond to a combination of drugs, and a significant portion, approximately 20–30%, have intractable epilepsy, i.e. they still suffer spontaneous seizures despite drug treatment. New, more effective anticonvulsants with reduced side-effects are needed to combat intractable epilepsy. Although the potential is promising and the need is evident, there are two important prerequisites that must be satisfied in order for Drosophila to be an effective tool in developing drug therapies for seizure disorders: (1) a good Drosophila model of the human pathology and (2) some assurance that currently available human treatments are effective in Drosophila. In this paper, we examine the effect of the anticonvulsant valproate on bang-sensitive (BS) paralytic mutants, a Drosophila model for human seizure disorders [12]. The BS mutant class includes several mutants such as bangsenseless (bss), easily shocked (eas), slamdance (sda) and technicalknockout (tko). All BS mutants suffer from cycles of intense behavioral hyperactivity and temporary paralysis caused by a mechanical shock, such as a tap of the culture vial on the bench top or brief vortex mixing (a ‘bang’) [2,6]. The hyperactivity phenotype is characterized by intense, uncoordinated motor activity including wing flapping, leg shaking, and abdominal muscle contractions; the paralytic phenotype, on the other hand, is observed as a cessation of all physical activity [2,26]. The hyperactivity and paralysis can be mimicked on the electrophysiological level by stimulating and recording from the central nervous system (CNS) [12,27]. These analyses show that the BS hyperactivity is due to neurological seizure and that the mutants are five to ten times more sensitive to evoked seizure than wild type flies. Paralysis is due to a failure of chemical synaptic transmission in several neural circuits. Thus, Drosophila BS mutants are a potentially useful model for human seizure-sensitivity, although the model is limited by a lack of some of the complex neural networks that may play an important role in mammalian epileptigenesis [24]. Valproate (sodium valproate, valproic acid) is a shortbranched fatty acid that is the most widely used antiepileptic drug for the treatment of generalized and partial epilepsy [11,19]. Its antiepileptic effectiveness was discovered serendipitously when it was used as a carrier for testing other hydrophobic compounds in experimental animal models of epilepsy. Diverse forms of epilepsy have been found to respond to valproate and this has been taken
37
to indicate that the drug may act to control seizures through multiple targets. One mechanism of seizure control appears to be through a valproate effect on voltagegated sodium channels. Valproate causes a decrease in peak sodium conductance that is voltage-dependent and delays recovery from inactivation [33; see however, Ref. 1]. These effects appear to limit high frequency firing of neurons thereby interfering with the ability to generate and propagate seizures [23,31]. Valproate also appears to interfere with seizures by increasing the amount of synaptic inhibition by g-aminobutyric acid (GABA). Valproate rapidly increases total GABA levels in the rodent brain with significant changes observed within 5 min and overall increases reaching 15–45% [7,9,14,20]. Valproate is believed to alter GABA levels through both the stimulation of GABA biosynthetic enzymes and the inhibition of enzymes involved in GABA degradation. In this study, we show that valproate is able to suppress seizure in BS flies in a dosage-dependent manner and that the drug does not appear to affect the firing threshold of individual neurons. This is the first report that we are aware of that examines the effects of valproate on the Drosophila nervous system. Further investigations may be able to determine the mode of action for valproate effect in Drosophila and assess if it is mechanistically homologous to the human anticonvulsant effectiveness. Also, these investigations may serve as foundation for the use of Drosophila in the screening and evaluation of novel anticonvulsant agents.
2. Materials and methods
2.1. Animals Normal wild type Drosophila were the Canton Special (CS) strain. The bang senseless (bss) and slamdance(sda) mutants are bang-sensitive (BS) paralytics that are 5–10 times more susceptible to seizures than normal flies. The mutant behavioral phenotypes of seizure and paralysis, the corresponding electrophysiological phenotypes of seizure and synaptic failure, and the threshold for seizure susceptibility of bss and sda alleles compared to normal have been described [6,12,13,26,27]. The sda gene encodes a protein homologous to human aminopeptidase N [21,35]. The bss gene product has not been described.
2.2. Electrophysiology Electrophysiology experiments on unanesthetized intact male flies were performed as described previously [12,30]. For experiments involving injected solutions, the fly was mounted to an insect pin and then while affixed, moved to a petri dish containing a piece of packaging foam (1.030.5 in.; 1 in.52.54 cm) glued in the middle. The fly was placed dorsal-side up on the foam and the pin was stabilized with
38
D. Kuebler, M. Tanouye / Brain Research 958 (2002) 36–42
plasticine. A small incision was made through the head cuticle between the interocellar setae [5] using a small dissecting knife (Beaver姠 1.5 mm). Hemolymph solution (JR solution) was injected into the head through the incision using a glass micropipette (tip diameter 0.5 mm). Enough JR solution was injected to cause the proboscis to extend. Particular care was taken to avoid over-injection of JR solution which could cause damage by rupture. Any flies in which the proboscis failed to extend or showed damage by rupture were discarded. The composition of JR solution was (in mM): 100 NaCl, 25 KCl, 6 CaSO 4 , 10 MgSO 4 , 4 NaHCO 3 , 1 NaH 2 PO 4 , 5 trehalose, 75 sucrose, 5 HEPES at pH 7.2 [32]. JR solution containing valproate (Na-valproate, Sigma) was made fresh before each experiment. Valproate concentrations are indicated in the text and refer to the value of Na-valproate. Three minutes after injection of JR solution, bipolar tungsten stimulating electrodes were inserted into the head of the fly as described previously [12]. Single-pulse stimuli (0.2 ms duration) were delivered to the brain to drive the giant fiber (GF) and GF-driven muscle potentials were recorded in the dorsal longitudinal muscles (DLM) using tungsten recording electrodes. GF thresholds were determined as the lowest voltage at which the short latency GF pathway responded. Five minutes following the injection, a single high frequency electrical stimulus (HFS) wavetrain with parameters 0.5 ms pulses at 200 Hz for 400 ms was delivered to the brain at the voltage indicated to elicit seizures. Previously, we have shown that seizures are elicited in an all-or-nothing manner [12]. Seizures consist of high frequency activity in at least seven different muscle groups and over thirty muscle fibers in the thorax. This activity in each muscle fiber corresponds to seizure activity in the motoneuron that innervates it. In the present paper, recordings of DLM muscle potentials were used to denote the occurrence of seizures as described previously [12]. The seizure threshold graph found in Fig. 2 was generated by calculating the percentage of flies that seized at each HF stimulus intensity for a particular genotype. The number of animals examined was n . 20 for each experimental group tested. The midpoint of the curve was taken as the seizure threshold for that strain and the slope of the curve indicates the level of variability within each genotype. For control animals, the values obtained by this method show good agreement to those described previously [12,13].
3. Results
3.1. Seizure susceptibility in BS and wild type Drosophila In intact Drosophila, seizures are elicited by high frequency electrical stimuli (HFS) delivered to the brain (Fig. 1) [12,13,27]. These evoked seizures are character-
Fig. 1. Seizures in intact sda and CS flies. The BS mutant sda fly is more susceptible to seizures than the wild type CS fly and therefore has a much lower seizure threshold. (A) A seizure is elicited in a sda fly by a high frequency stimulus of low strength (8 V) and displayed at a high sweep speed. The HF stimulus (labeled HF) is a short wavetrain (0.5 ms pulses at 200 Hz for 300 ms) of electrical stimuli delivered to the brain. Recording is from a DLM muscle fiber and reflects the activity of the single DLM motoneuron that innervates it. The ‘seizure’ (labeled SEI) is abnormal high frequency firing of the DLM motoneuron. The seizure is widespread as similar activity can be found in recordings from seven different muscle groups in the fly following HF stimulation (Kuebler and Tanouye [12]). (B) A low-voltage HF stimulus of 8 V fails to elicit a seizure in a wild type CS fly because the stimulus is below the seizure threshold. Following the HF stimulus artifact, there is no seizure activity observed in this recording displayed at a high sweep speed. Note also that there is no period of synaptic failure and single-pulse stimulation of the GF (0.5 Hz) continues to evoke DLM potentials. Two such effective single-pulse stimuli are depicted in this trace; each was effective in evoking a DLM potential. (C) A seizure is elicited in a wild type CS fly by a high voltage HF stimulus (30 V) which is above the threshold for seizure. The seizure in this recording begins within the large stimulus artifact and is displayed at a high sweep speed. (D) Same recording as for (A) from a sda fly displayed at a slow sweep speed. In this recording, the HF stimulus and seizure (SEI) are followed by a quiescent period (labeled SYNAPTIC FAILURE) that is characterized by synaptic failure within the GF circuit (Pavlidis and Tanouye [27]). During this period, there are stimulus artifacts (downward-going) from continuous singlepulse stimulation of the GF (0.5 Hz), but no evoked DLM potentials. Spontaneous activity (labeled SPON) or ‘recovery seizure’ appears as additional seizure-like activity occurring just after the synaptic failure period and just prior to recovery (recovery not evident in this trace). Vertical calibration bar is 20, 40, 40 and 10 mV for (A), (B), (C) and (D), respectively. Horizontal calibration bar is 300 ms, 1.2 s, 1.2 s, and 1.5 s for (A), (B) (C) and (D), respectively (figure modified from Kuebler et al. [13], Fig. 1).
D. Kuebler, M. Tanouye / Brain Research 958 (2002) 36–42
ized by aberrant high frequency spiking in a large population of motoneurons and muscles throughout the fly. For normal Canton-Special (CS) males, an HFS of 0.5 ms pulses at 200 Hz for 300 ms causes an all-or-nothing seizure at a threshold voltage of 30.163.8 V. The bangsensitive (BS) paralytic mutants are 5–10 times more sensitive to seizures than normal flies: the mutants bss and sda have seizure thresholds to HFS of only 3.260.6 and 6.260.8 V, respectively. The seizures are followed by a period of quiescence that is due to synaptic failure in many chemical synapses and an additional seizure just prior to recovery (Fig. 1) [12,27]. This quiescent period is apparently responsible for behavioral paralysis in BS mutant flies [27]. Immediately following a bout of seizure–quiescence–recovery seizure, there is a transient increase in seizure threshold lasting several minutes that contributes to a refractory period [12]. We examined the effects of fly surgery and hemolymph solution injection on excitability and seizure initiation. Our general observations were that stable electrical recordings could be maintained for about 10 min. However, longer term recordings were unreliable because of spontaneous excitability changes as determined by threshold changes in evoked GF responses to single pulses and HFS-evoked seizures. This necessitated a change in our general protocol. For each fly undergoing surgery, the GF threshold was determined initially by single pulse stimulation at 3 min after surgery, then a single HFS (0.5 ms pulses at 200 Hz for 400 ms) was delivered at a preset voltage at 5 min after surgery. The preparation was then discarded because of uncertainties due to refractory period and postsurgery changes in threshold. The seizure threshold curves found in Fig. 2 indicate the level of seizure susceptibility in CS, bss and sda strains. The rank order of the genotypes is identical to that seen previously in studies that did not employ surgery and injection: bss was the most susceptible to seizure, CS was the least susceptible, and sda was intermediate, although closer to bss than CS. The apparent seizure thresholds for bss, sda, and CS are about 6, 10 and 40 V, respectively. These values are similar to, although slightly higher than, those found for intact flies.
39
Fig. 2. Surgery and hemolymph injection do not greatly affect the seizure threshold. A single HFS stimulus of a fixed voltage was delivered to each fly 5 min after surgery and hemolymph injection at the voltage indicated and its effectiveness in generating a seizure noted. The curves depicted here show the percentage of flies that have a seizure (n . 20 flies for all genotypes). The apparent seizure threshold for CS is about 40 V, the voltage where seizure was initiated in 50% of flies. The apparent seizure thresholds for bss and sda are about 6 and 10 V, respectively.
V indicates the ability to suppress seizures to wild type levels. We found that 5 mM valproate suppressed seizures in most sda flies and 10 mM completely suppressed seizures (Fig. 3). In the case of bss, 10 mM valproate suppressed seizures in most flies and 25 mM valproate completely suppressed seizures. Given what is known about valproate [19,11], it is likely
3.2. Valproate suppresses seizures in bss and sda Valproate is a clinically-used antiepileptic drug that is effective for treating both generalized and partial seizures [11,19]. Valproate has also been shown to be effective in suppressing seizures in a variety of animal models. Here, we found valproate can suppress the seizure susceptible defect normally seen in the bss and sda mutant strains. We tested a range of increasing concentrations of valproate on both strains and assayed the susceptibility to an HFS wavetrain of fixed voltage (40 V, 0.5-ms pulses at 200 Hz for 400 ms). This voltage was chosen because it is about the seizure threshold for CS wild type flies. Thus, the ability of a given concentration to suppress seizures at 40
Fig. 3. Valproate reduces seizure susceptibility in bss and sda flies. A single HFS stimulus of 40 V (0.5 ms pulses at 200 Hz for 400 ms) was delivered to each fly 5 min after surgery and hemolymph plus valproate injection and its effectiveness in generating a seizure noted. The curves depicted here show the percentage of flies that have a seizure for various concentrations of valproate as indicated (n . 20 flies for each genotype).
40
D. Kuebler, M. Tanouye / Brain Research 958 (2002) 36–42
Fig. 4. Giant fiber threshold is not affected by valproate. Single pulse stimuli (0.2 ms duration, 0.8 Hz) were delivered to each fly 3 min after surgery and hemolymph plus valproate injection and the GF threshold determined. Thresholds remained the same even at the highest drug concentration.
that it exerts its anticonvulsant effect in flies by a variety of mechanisms. One possibility is that valproate may alter seizure susceptibility by altering the thresholds of individual neurons. We examined the giant fiber (GF) neuron threshold at the various concentrations of valproate in both sda and bss and found that it was not affected (Fig. 4). Even at the highest valproate concentrations tested bss had a threshold of 2.7860.34 V at 25 mM and sda had a threshold of 2.6460.33 V at 50 mM; both of which are not significantly different from the GF threshold in the absence of valproate that was for bss and sda 2.4260.37 and 2.3560.40 V, respectively. Thus, at the concentrations we examined, valproate was able to raise the seizure threshold while the single neuron excitability, at least as indicated by GF threshold, remained unchanged. Although the GF threshold was unaffected, we cannot rule out the possibility that other unidentified neurons, critical to the generation of seizures, had significant alterations in excitability.
4. Discussion Although epilepsy affects over 1% of the US population, an understanding of the underlying mechanisms is in the vast majority of cases lacking. One reason that reaching an understanding of the etiology of human seizure disorders has been difficult is that there is such a variety of both causes and types of seizures. Dissecting the pathogenesis of seizure disorders is extremely difficult and is compli-
cated by the heterogeneity of defects as well as the vast array of molecular lesions known to be involved in both human and mouse seizure disorders. The complexity and hetereogeneity of epilepsy is also evident by the number and apparent functional diversity of anticonvulsants used to treat the disorder. Although drug therapy is often successful, some patients only respond to a combination of drugs and there are a substantial number of epilepsy cases that are intractable, unresponsive to drug treatments [17,18]. This is despite the emergence of second generation anticonvulsants such as lamotrigine, vigabatrin, gabapentin and topiramate [4]. What makes the cases of intractable epilepsy particularly intriguing is that often people with identical symptoms can have completely different responses to the same drug [29]. The reasons behind this are unclear, but it demonstrates both the extreme heterogeneity of these disorders and the need for the development of new, more effective anticonvulsants to combat them. An especially attractive approach for developing novel anticonvulsants is to take advantage of Drosophila seizuresuppressor mutations. We have shown previously that certain genetic mutations can both elevate the seizure threshold in Drosophila and suppress the seizure-susceptible phenotype seen in BS mutants [13]. All of these mutants, in double-mutant combinations with BS mutants, provided a genetic background that elevated the seizure threshold to values above those seen in the BS mutants alone. In each case in which the BS seizure susceptible phenotype was altered, a single mutation, which becomes a possible target for the development of novel anticonvulsants, is responsible. Further identification of additional seizure-suppressor mutations through genetic screens could provide an extremely rich source that could serve as a basis for a an extensive drug development program. This study is the first description of a method for testing the ability of an anticonvulsant to suppress seizures in Drosophila. By injecting drug into the head cavity and immediately measuring the effect on seizure susceptibility, we are able to measure the effectiveness of valproate, a commonly-used anticonvulsant in humans, to rescue the seizure-sensitive phenotype that is characteristic of Drosophila BS mutants. The ability to suppress seizures in the BS strains was found to vary depending on genotype. Higher concentrations were required to suppress bss mutants, the more susceptible of the two genotypes, than were required for sda mutants. We have previously suggested that seizures in Drosophila are due to an interplay between excitatory or seizure-initiating neuronal circuits and inhibitory or seizure-suppression circuits [12]. The excitatory circuits may contain circuit elements that are linked through reciprocal excitatory synapses that provide positive feedback, as described for mammalian seizure models [24]. Similarly, the inhibitory circuits may contain circuit elements that feed back inhibition through presynaptic or postsynaptic synapses. Seizure control in
D. Kuebler, M. Tanouye / Brain Research 958 (2002) 36–42
Drosophila mutants by valproate could be through a valproate effect on voltage-gated sodium channels [33]. Indeed, we have previously shown that mutations affecting voltage-gated sodium channels can act as seizure-suppressor mutations in double mutant combinations with BS mutations [13]. A less likely possibility is that valproate is increasing the amount of synaptic inhibition by GABA [7,9,14,20]. GABA is a major inhibitory transmitter in Drosophila, however, in the short drug application times used in the present experiments, it is unlikely that biosynthetic or degradation enzymes are causing much change in GABA levels. Suppression of seizures by valproate is one way Drosophila seizures are similar to mammalian seizures. This adds to previous observations [12,13] that have shown several similarities between seizures in mammals and Drosophila. For example, just as in mammals, seizures in Drosophila exhibit the following characteristics: all individuals have a seizure threshold; seizure susceptibility can be modulated by genetic mutations; electroconvulsive therapy in flies raises the threshold for subsequent seizures; seizures in flies can be spatially segregated into particular regions of the central nervous system; and seizures spread through the nervous system along particular pathways that are dependent on functional synaptic connections and recent electrical activity. Although there are, at present, several superficial similarities between Drosophila and mammalian seizures, there has not yet been a direct demonstration of identical mechanistic homology underlying seizure-susceptibility. Additional study is required to determine the extent of this homology and validate further the utility of this Drosophila model. Many anticonvulsant drugs have, unlike valproate, a specific mode of action [15,16,22,23]. Thus, future testing of different standard anticonvulsant drugs with different modes of action on a variety of different BS mutants would be especially valuable for defining the predictivity for special types of seizures. This would be particularly important in drug development to exclude the possibility that only drugs with a special mechanism are selected. In addition, the specificity of certain drugs can help decipher the processes involved in the BS defect. Also, elucidating different anticonvulsant profiles for the BS strains may indicate underlying differences in the pathogenesis of seizures in different BS mutants and could help in understanding the specific mechanisms that lead to seizure susceptibility in each specific mutant. Finally, the ability to screen multiple compounds makes the system amenable for screening for novel anticonvulsants. Because a currently used anticonvulsant, valproate, has been demonstrated to be effective in suppressing the BS phenotype, the identification of clinically-relevant compounds appears plausible by this method. The technique we have described here has many potential uses and should open lines of research that were previously inaccessible.
41
Acknowledgements The authors thank Sally Faulhaber for assistance in the maintenance of Drosophila stocks. We thank Drs. Jeremy Lee and Charlie Oh for discussion throughout this project and Drs. David Bentley and Geoff Owen for their insightful comments on the manuscript. Part of this work was supported by USPHS grant NS31231 and an Epilepsy Foundation grant to MAT.
References [1] H. Albus, R. Williamson, Electrophysiologic analysis of the actions of valproate on pyramidal neurons in the rat hippocampal slice, Epilepsia 39 (1998) 124–139. [2] S. Benzer, From the gene to behavior, J. Am. Med. Assoc. 218 (1971) 1015–1022. [3] Commission on Classification and Terminology of the International League Against Epilepsy, Proposal for revised classification of epilepsies and epileptic syndromes, Epilepsia 30 (1989) 389–399. [4] S. Czuczwar, P. Patsalos, The new generation of GABA enhancers. Potential in the treatment of epilepsy, CNS Drugs 15 (2001) 339– 350. [5] G.F. Ferris, External morphology of the adult, in: M. Demerec (Ed.), Biology of Drosophila, Hafner, New York, 1950, pp. 368–418. [6] B. Ganetzky, C.-F. Wu, Indirect suppression involving behavioral mutants with altered nerve excitability in Drosophila melanogaster, Genetics 100 (1982) 597–614. [7] Y. Godin, L. Heier, J. Mark, P. Mandel, Effects of di-n-propylacetate, an anticonvulsant compound on GABA metabolism, J. Neurochem. 16 (1969) 869–873. [8] R. Greenspan, Fly Pushing: The Theory and Practice of Drosophila Genetics, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1997. [9] C. Hariton, S. Ciesielsky, S. Simler, M. Valli, G. Jadot, S. Gobaille, E. Mesdijan, P. Mandel, Distribution of sodium valproate and GABA metabolism in CNS of the rat, Biopharm. Drug Dispos. 5 (1984) 409–414. [10] W.A. Hauser, D.C. Hesdorffer, Epilepsy: Frequency, Causes, and Consequences, Demos, New York, 1990. [11] C.U. Johannessen, Mechanisms of action of valproate: a commentatory, Neurochem. Intl. 37 (2000) 103–110. [12] D. Kuebler, M.A. Tanouye, Modifications of seizure susceptibility in Drosophila, J. Neurophysiol. 83 (2000) 998–1009. [13] D. Kuebler, H.G. Zhang, X. Ren, M.A. Tanouye, Genetic suppression of seizure susceptibility in Drosophila, J. Neurophysiol. 86 (2001) 1211–1225. [14] K. Kukino, T. Deguchi, Anticonvulsant activity and effects of sodium dipropylacetate on cerebral 5-hydroxytryptamine and gaminobutyric acid in reserpinized mice, Chem. Pharmacol. Bull. 26 (1978) 3551–3555. [15] C.C. Kuo, A common anticonvulsant binding site for phenytoin, carbamazepine, and lamotrigine in neuronal Na 1 channels, Mol. Pharmacol. 54 (1998) 712–721. [16] C.C. Kuo, R. S Chen, L. Lu, R.C. Chen, Carbamazepine inhibition of neuronal Na 1 currents: quantitative distinction from phenytoin and possible therapeutic implications, Mol. Pharmacol. 51 (1997) 1077–1083. [17] I.E. Leppik, Intractable epilepsy: prognostic evaluation, in: W.H. Theodore (Ed.), Surgical Treatment of Epilepsy, Elsevier, Amsterdam, 1992, pp. 7–11.
42
D. Kuebler, M. Tanouye / Brain Research 958 (2002) 36–42
[18] W. Loscher, Animal models of intractable epilepsy, Prog. Neurobiol. 53 (1997) 239–258. [19] W. Loscher, Valproate: a reappraisal of it pharmacodynamic properties and mechanism of action, Prog. Neurobiol. 58 (1999) 31–59. [20] W. Loscher, M. Vetter, Relationship between drug-induced increases of GABA levels in discrete brain areas and different pharmacological effects in rats, Biochem. Pharmacol. 33 (1985) 1907–1914. [21] A. Look, R. Ashmun, L. Shapiro, S. Peiper, Human myeloid plasma membrane glycoprotein CD13 (gp150) is identical to aminopeptidase N, J. Clin. Invest. 83 (1989) 1299–1307. [22] M.J. McLean, R.L. MacDonald, Multiple actions of phenytoin on mouse spinal cord neurons in cell culture, J. Pharmacol. Exp. Ther. 227 (1983) 779–789. [23] M.J. McLean, R.L. MacDonald, Sodium valproate, but not ethosuximide, produces use- and voltage-dependent limitation of high frequency repetitive firing of action potentials of mouse central neurons in cell culture, J. Pharmacol. Exp. Ther. 237 (1986) 1001– 1011. [24] J.O. McNamara, Cellular and molecular basis of epilepsy, J. Neurosci. 14 (1994) 3413–3425. [25] J.O. McNamara, Emerging insights into the genesis of epilepsy, Nature 399 (Suppl.) (1999) A15–A22. [26] P. Pavlidis, M. Ramaswami, M.A. Tanouye, The Drosophila easily shocked gene: a mutation in a phospholipid pathway causes seizure, neuronal failure, and paralysis, Cell 79 (1994) 23–33. [27] P. Pavlidis, M.A. Tanouye, Seizures and failures in the giant fiber
[28]
[29]
[30] [31]
[32]
[33] [34] [35]
pathway of Drosophila bang-sensitive paralytic mutants, J. Neurosci. 15 (1995) 5810–5819. R.S. Puranam, J.O. McNamara, Seizure disorders in mutant mice: relevance to human epilepsies, Curr. Opin. Neurobiol. 9 (1999) 281–287. D. Schmidt, Diagnostic and therapeutic management of intractable epilepsy, in: D. Schmidt, P.L. Morselli (Eds.), Intractable Epilepsy: Experimental and Clinical Aspects, Raven Press, New York, 1986, pp. 237–258. M.A. Tanouye, R.J. Wyman, Motor outputs of the giant nerve fiber in Drosophila, J. Neurophysiol. 44 (1980) 405–421. A.B. Thomas, G.C. Mathews, J.A. Ferrendelli, Mechanisms of action of antiepileptic drugs, in: E. Wyllie (Ed.), The Treatment of Epilepsy: Principles and Practice, 2nd Edition, Williams and Wilkins, Baltimore, MD, 1996, pp. 700–711. J.R. Trimarchi, R.K. Murphey, The shaking-B 2 mutation disrupts electrical synapses in a flight circuit in adult Drosophila, J. Neurosci. 17 (1997) 4700–4710. R.J. Van den Berg, P. Kok, R.A. Voskuyl, Valproate and Na 1 currents in cultured hippocampal cells, Exp. Brain Res. 93 (1993) 279–287. L. Walton, in: Essentials of Neurology, 6th Edition, Churchill Livingstone, London, 1989, pp. 77–86. H.G. Zhang, J. Tan, E. Reynolds, D. Kuebler, S. Faulhaber, M.A. Tanouye, The Drosophila slamdance gene: a mutation in an aminopeptidase can cause seizure, paralysis and neuronal failure, Genetics, in press.