Description Of The Model

Panic Away

Panic Away Ebook

Get Instant Access

Behavioral Manifestations

The clinical significance of any experimental model of epilepsy is limited unless, like humans, a stereotyped set of behavioral changes are observed concomitant with a seizure. Because a wealth of behavioral information is available from the rodent literature, including a detailed and well-accepted set of behavioral seizure stages (Racine, 1972), we began by investigating freely swimming zebrafish larvae exposed to PTZ. One approach to quantify seizure behaviors was to videotape the zebrafish and then review the recordings at a slower speed (and with the benefit of tape replay). For these studies, individual zebrafish were placed in 96-well Falcon plates and the solution was exchanged for normal bathing medium containing 15 mM PTZ. Fish were monitored for 30 to 60 minutes using a digital video camera. From these videotapes, three distinct stages of behavior were noted. In stage I, zebrafish larvae showed a general increase in swim activity. This increase occurred in the first few minutes of PTZ exposure and was not seen on exposure to fresh bathing medium. Subsequently, stage II behaviors characterized by a rapid "whirlpool-like" circling around the outer edge of the well were observed. Finally, in stage III, zebrafish larvae exhibited brief head-to-tail convulsions followed by a loss of posture (e.g., fish floating on its side). Stage III convulsions, associated with very rapid movement across the dish (see dashed lines in Figure 1), lasted between 1 and 3 seconds and persisted as long as the fish were exposed to PTZ. Once established as distinct seizure stages, we compared the latency and prevalence of these episodes at varying concentrations of PTZ. As expected, seizure behaviors were elicited in a concentration-dependent fashion. These findings are not terribly surprising and are entirely consistent with well-established rodent pediatric epilepsy models (Jensen et al., 1991; Priel et al., 1996; Sankar et al., 1999).

Visual observation of freely behaving animals is a reasonable and commonly used method to document seizure behaviors. However, automated observation systems can provide several distinct advantages. In particular, behaviors can be recorded more reliably because computer algorithms do not suffer from observer fatigue or interobserver variability. Further, computer automated detection is open to a wide array of measurements, for example, distance traveled, duration of movement, and velocity. For automated detection of zebrafish seizure behavior, we set up a computer-based locomotion tracking system. Using a stereomicroscope and high-speed charge-coupled device (CCD) camera acquiring images at 30 Hz, we can record the behavior of single zebrafish larvae during exposure to PTZ. Subsequently video output is digitized by a frame grabber and passed directly to a computer running EthoVision software (Noldus Information Technology, Wageningen, The Netherlands). Image-processing algorithms are then applied to analyze each frame and to distinguish the zebrafish against a background image and, on the basis of user-defined gray-scale values, track the position of the zebrafish in one well (Noldus et al., 2002). From these types of recordings, we obtain locomotion plots as shown in Figure 1. Note that each of the seizure stages defined initially by direct visual observation can be reproduced and quantified using this locomotion tracking system. Of many potential uses for an automated seizure detection system, one could be a comparison between epileptic phenotypes in various wild-type and mutant zebrafish, thus facilitating phenotype-genotype strategies currently in vogue (Lewis et al., 2004; Zumkeller et al., 2004).

Electrophysiologic Manifestations

By definition, epilepsy models must exhibit some form of abnormal electrical discharge in a CNS structure, and the

FIGURE 1 Behavioral seizure stages. a: Frame-grabber image of a 7 days post fertilization zebrafish larvae in one well of a 96-well Falcon plate. b: Sample locomotion tracking plots are shown for individual zebrafish in normal Ringer's medium (baseline), during stage I (increased swimming activity), stage II (circling), and stage III (clonus-like convulsions). Solid lines indicate movement; dashed lines indicate rapid convulsive seizure activity (this fish exhibited >25 convulsive episodes). Plots were obtained from recording epochs 2 minutes in duration. (See color insert.)

FIGURE 1 Behavioral seizure stages. a: Frame-grabber image of a 7 days post fertilization zebrafish larvae in one well of a 96-well Falcon plate. b: Sample locomotion tracking plots are shown for individual zebrafish in normal Ringer's medium (baseline), during stage I (increased swimming activity), stage II (circling), and stage III (clonus-like convulsions). Solid lines indicate movement; dashed lines indicate rapid convulsive seizure activity (this fish exhibited >25 convulsive episodes). Plots were obtained from recording epochs 2 minutes in duration. (See color insert.)

pattern of electrographic activity can be used to classify the model in clinical terms (Engel and Pedley, 1997). To meet this "gold standard," we sought a means to document PTZ-induced electrographic seizure activity in developing zebrafish. The obvious problem with this model criterion is technical: how to record electrical events in a larval zebrafish brain. A standard experimental strategy in rodent pediatric epilepsy models is to implant indwelling elec-troencephalographic (EEG) recording electrodes in "complex" brain structures such as neocortex or hippocampus. At first glance, one might consider zebrafish larvae incapable of generating EEG-like discharge as their brains lack a highly ordered laminar cortical structure (Wulliman et al., 1996). Because larval zebrafish are only 3 to 4mm long, implantation of a chronic EEG monitoring device in a freely swimming fish is not possible. As an alternative we developed a procedure to immobilize zebrafish larvae in a low-melting-point agar block. The principal advantage of this procedure is that invasive surgical techniques and anesthetic drugs (with their undesirable potential for interfering with CNS function) are not necessary. The limitation of this protocol is that we cannot simultaneous monitor electrical activity and behavior. Agar-embedded zebrafish can be placed on the recording chamber of an upright or stereo microscope, which allows for perfusion with various con-vulsant (or anticonvulsant) solutions. Because zebrafish larvae are translucent, we employed a relatively simple procedure to place glass microelectrodes in visualized central brain structures using a three-dimensional micromanipula-

tor. In our laboratory, successful field recordings have been obtained from the optic tectum, telencephalon, and cerebellum in zebrafish as young as 4DPF.

In 6 to 7DPF zebrafish, bath application of 15 mM PTZ induced a progressive increase of tectal electrical activity, resulting in population bursts with typical features of inter-ictal discharges, for example, abrupt in onset, primarily monophasic, and brief duration. Interictal-like events were observed within 10 to 20 minutes of drug exposure. Continuous application of PTZ resulted in interictal activity followed by ictal-like discharges lasting 4 to 6 seconds and composed of large-amplitude population spikes. Ictal-like discharge was usually followed by a period of suppressed electrical activity or postictal depression (Figure 2a). Simultaneous field recordings in optic tectum and telencephalon indicate a seizure pattern that could be classified as "generalized" epilepsy; epileptic discharge appeared nearly simultaneously in all brain regions (Figure 2c). Electrographic seizure events generated in CNS structures (optic tectum or telencephalon) preceded convulsive activity monitored as an electromyogram in the fish-tail muscle (Figure 2c). Electrographic activity was similar in fish exposed to curare (10-250 mM, a neuromuscular blocking agent), further demonstrating that agar-embedding is a sufficient method to immobilize zebrafish larvae for electrophysiol-ogy. Spontaneous discharges were never observed in fish exposed to normal bathing medium. Of course, convulsant agents effective in eliciting epileptiform-like activity in immature animals are not limited to PTZ, and this review

FIGURE 2 Epileptiform-like electrographic activity. a: Representative tectal field recording from a zebrafish larva exposed to 15 mM pentylenetetrazol (PTZ) for 45 minutes. Note the presence of interictal, ictal, and postictal phases of the electrical recording. b: Schematic of the configuration used to obtain electrophysiologic recordings from agar-embed-ded zebrafish larvae. c: Representative dual field recordings from a zebrafish exposed to 15 mM PTZ. In one set of recordings (left), a nearly simultaneous epileptiform burst discharge is observed in the zebrafish forebrain and optic tectum. In a second set of recordings (right), an epileptiform burst discharge is seen to originate in the optic tectum prior to its appearance in a distal section of the zebrafish tail muscle. EMG, electromyelograph.

FIGURE 2 Epileptiform-like electrographic activity. a: Representative tectal field recording from a zebrafish larva exposed to 15 mM pentylenetetrazol (PTZ) for 45 minutes. Note the presence of interictal, ictal, and postictal phases of the electrical recording. b: Schematic of the configuration used to obtain electrophysiologic recordings from agar-embed-ded zebrafish larvae. c: Representative dual field recordings from a zebrafish exposed to 15 mM PTZ. In one set of recordings (left), a nearly simultaneous epileptiform burst discharge is observed in the zebrafish forebrain and optic tectum. In a second set of recordings (right), an epileptiform burst discharge is seen to originate in the optic tectum prior to its appearance in a distal section of the zebrafish tail muscle. EMG, electromyelograph.

highlights one of what are potentially many methods to induce epilepsy in zebrafish. As a brief example of the possibilities, we also found that electrographic seizures were elicited on exposure to high potassium, picrotoxin, or pilo-carpine (e.g., manipulations used to elicit seizure-like activity in rodent models) (Figure 3). In each case, a distinct pattern of abnormal electrical discharge was observed. These findings demonstrate the broad nature of using a zebrafish model and suggest that epilepsy research in this species is not limited to PTZ-induced seizures. A further application could be the introduction of gene mutations previously shown to elicit epileptic phenotypes (e.g., generation of "epileptic" zebrafish). In general, field potentials recorded in larval zebrafish are remarkably similar in waveform to those reported in chronic seizure models in vivo (Bragin et al., 1999a), acute seizure models in vitro (Dzhala et al., 2003), and in the epileptogenic region of patients with temporal lobe epilepsy (Bragin et al., 1999b).

Research into the basic mechanisms of seizure genesis and propagation has provided evidence consistent with a critical role for excitatory glutamate-mediated synaptic transmission. In the context of pediatric epilepsy models, antagonists to the alpha-amino-3-hydroxy-5-methyl-4-isox-azole propionic acid (AMPA) glutamate receptor subtype block hypoxia-induced seizures in the immature rat (Jensen et al., 1995), and 3-(RS)-2-carboxypiperazine-4-yl)-propyl-1-phosphonic acid (CPP) blocks 4-aminopyridine-induced spontaneous episodes of spreading depression in hippocam-

pal slices from immature rats (Psarropoulou and Avoli, 1996). As such, agents that block presynaptic glutamate release or activation of postsynaptic ionotropic glutamate receptors should be effective at terminating chemically induced seizures in larval zebrafish. Because ionotropic glutamate receptors are expressed in the larval zebrafish brain (Edwards and Michel, 2003), we predicted a role for gluta-mate-mediated synaptic transmission in the generation or propagation of electrographic seizure discharge in our model. Pharmacologic studies performed on zebrafish larvae at 7DPF confirmed this hypothesis. First, bath application of tetrodotoxin (TTX) to block Na-dependent action potentials, and therefore synaptic transmission, abolished PTZ-induced electrographic seizures. Second, application of either a nonspecific glutamate receptor antagonist (kynurenate) or a "cocktail" of AMPA/N-methyl-D-aspartate (NMDA) receptor antagonists 6-cyano-7-introquinoxaline-2,3-dione-2amino-5-phosphonovalerate (CNQX-APV) abolished PTZ-induced electrographic seizures. These types of studies further support the general concept that chemically induced seizure activity in zebrafish larvae is similar to that observed in higher vertebrates.

Molecular Alterations

Revolutionary advances in molecular biology have contributed to our nascent understanding of the many changes in gene expression that can occur prior to, during, and after

FIGURE 3 Electrographic seizure activity in zebrafish larvae with various convulsants. a: Representative 1-minute field recording traces are shown following 45 to 60 minutes of bath exposure to convulsant agents: 25 mM KCl or "high-potassium" model; 100 mM picrotoxin, a g-aminobutyric acid (GABA) receptor antagonist; and 120 mM pilocarpine, a muscarinic acetylcholine receptor agonist. b: Isolated burst discharge shown at a faster time resolution as indicated by the asterisk in (a). Note that different type of burst discharge waveforms are elicited with the different mechanisms of seizure initiation.

FIGURE 3 Electrographic seizure activity in zebrafish larvae with various convulsants. a: Representative 1-minute field recording traces are shown following 45 to 60 minutes of bath exposure to convulsant agents: 25 mM KCl or "high-potassium" model; 100 mM picrotoxin, a g-aminobutyric acid (GABA) receptor antagonist; and 120 mM pilocarpine, a muscarinic acetylcholine receptor agonist. b: Isolated burst discharge shown at a faster time resolution as indicated by the asterisk in (a). Note that different type of burst discharge waveforms are elicited with the different mechanisms of seizure initiation.

a seizure episode. These changes in gene expression presumably reflect (1) molecular mechanisms that contribute to an acute hyperexcitable state (e.g., upregulation of immediate early gene expression) (Morgan et al., 1987); (2) compensatory alterations in genes that may limit the spread of epileptic activity (e.g., increased neuropeptide Y expression) (Rizzi et al., 1993); or (3) molecular alterations that underlie a process of epileptogenesis (e.g., changes in GABAa receptor subunit expression) (Brooks-Kayal et al., 1998). Like most research in the epilepsy field, findings were first reported in adult seizure models (or tissue obtained from patients with intractable forms of epilepsy) and have only recently been studied in pediatric epilepsy models. Because zebrafish larvae represent a simple vertebrate system for "reductionist" questions related to how seizures develop in an immature nervous system, we were interested in reproducing at least one of the molecular alterations that would be expected in an established seizure model. Because one of the most robust and widely reproduced examples of a seizure-induced change in gene expression is the dramatic upregulation of the immediate early gene (IEG) c-fos in brain regions participating in seizure genesis, we began with this IEG (Dragunow and Robertson, 1987; Morgan et al., 1987). It is important to emphasize that c-fos experiments are discussed as "proof-of-principle" and are not intended to diminish the importance of other molecular alterations in epileptogenesis.

Because we observed robust electrographic seizure discharge and stage III seizure behavior in zebrafish larvae exposed to 15mM PTZ, our molecular studies were performed at this concentration. Freely swimming zebrafish were exposed to PTZ and monitored for seizures. At 0, 15, 30, and 60 minutes following continuous PTZ exposure, zebrafish were removed and total RNA was isolated. From larval RNA samples, cDNA was synthesized and serial dilutions were amplified by reverse transcriptase (RT)-poly-merase chain reaction (PCR). As expected, PTZ-induced seizures resulted in a significant upregulation of c-fos mRNA expression in all zebrafish. Levels of c-fos mRNA expression in untreated zebrafish larvae were fairly low. Further confirmation of a seizure-induced upregulation of c-fos mRNA expression in the zebrafish CNS was obtained using whole-mount in situ hybridization techniques; an increase in expression was seen in the optic tectum, fore-brain, and cerebellum. At present, antibodies raised against zebrafish c-fos are not available; thus we have not yet confirmed these results using immunohistochemistry. However, as additional zebrafish antibodies are developed, studies to localize the expression of IEGs in neurons participating in seizure genesis in the larval zebrafish brain will be mapped in greater detail. Analysis of postsynaptic GABA receptor subunits, neuropeptides, protein kinases, and the many genes whose expression levels are regulated by seizure activity remains to be performed. It is also likely that this simple model of induced seizures can be combined with gene microarray analysis (or other molecular methodologies) to study an even wider range of known genes, expressed sequence tagging (EST), etc.

Antiepileptic Drugs

It is a well-accepted maxim that testing of new therapeutic agents in humans follows on the preliminary screening and evaluation of these compounds in an experimental animal model. Discovery and screening of antiepileptic drugs have long followed this paradigm, and development of "ideal" animal models closely approximating human epilepsy is critical to this process. What, however, are the important criteria in the development of new animal models? Dixon Woodbury, a pioneer in antiepileptic drug (AED) screening, stated that an experimental model for evaluation of anticonvulsant drugs must meet two criteria: (1) electrographic evidence of epileptic-like activity and (2) clinical seizure-like behaviors manifest as tonic-clonic motor movements (Woodbury 1969). PTZ-induced seizure activity in rodents (commonly referred to as the Metrazol test) is a well-established experimental model fitting these criteria and has been successfully used to test anticonvulsant compounds. As we show here, PTZ-induced seizure activity in larval zebrafish also satisfies both of the Woodbury criteria.

Because our zebrafish work is based on rodent PTZ data, and because a substantial literature exists on AEDs that suppress (or do not suppress) PTZ-induced seizures (Ferrendelli et al., 1989; Krall et al., 1978), we proposed an additional criterion: demonstration of an AED pharmacologic profile identical to that observed in rodents. For these studies we chose suppression of tectal electrographic epileptiform discharge as a sensitive outcome measure of AED efficacy in zebrafish. In all electrophysiologic studies, agar-embedded zebrafish (7DPF) were exposed to 15mM PTZ until a stable level of baseline bursting was established (40-65 minutes; interictal + ictal activity). Next AEDs previously demonstrated to suppress (or abolish) PTZ-induced seizure in rodents were tested e.g., benzodiazepines (clonazepam and diazepam) or valproic acid (Figure 4). Both these agents suppressed PTZ-induced epileptiform discharge in a concentration-dependent manner. Reduction of ictal burst discharge amplitude and frequency could be observed within 30 to 45 minutes of drug exposure. Unlike in vitro slice studies, it was not possible to remove or "wash" the compounds efficiently from our in vivo preparation; thus recovery experiments were not performed. In separate control studies, we also tested AEDs previously shown to have little (or no) effect on PTZ-induced seizures in rodents (e.g., carbamazepine, phenobarbital, ethosuximide, phenytoin). As expected, these AEDs did not suppress electrographic discharge in zebrafish, and we present these findings as a critical validation of our model. Indeed, because rodent PTZ model successfully predicts drugs effective against generalized seizures of the absence (or petit mal) type, we anticipate that our chemically-induced zebrafish seizure model could also identify compounds with therapeutic potential in humans.

Was this article helpful?

0 0
Eliminating Stress and Anxiety From Your Life

Eliminating Stress and Anxiety From Your Life

It seems like you hear it all the time from nearly every one you know I'm SO stressed out!? Pressures abound in this world today. Those pressures cause stress and anxiety, and often we are ill-equipped to deal with those stressors that trigger anxiety and other feelings that can make us sick. Literally, sick.

Get My Free Ebook


Post a comment