- Email: [email protected]
Adrenergic receptor modulation of hippocampal CA3 network activity
Epilepsy Research 66 (2005) 117–128
Adrenergic receptor modulation of hippocampal CA3 network activity Chris W.D. Jurgens a , Sarah J. Boese a , Jacob D. King a , Sally J. Pyle b , James E. Porter a,∗ , Van A. Doze a,∗ a
b
Department of Pharmacology, Physiology and Therapeutics, University of North Dakota School of Medicine and Health Sciences, 501 N. Columbia Road, Grand Forks, ND 58203, USA Department of Anatomy and Cell Biology, University of North Dakota School of Medicine and Health Sciences, 501 N. Columbia Road, Grand Forks, ND 58203, USA Received 28 April 2005; received in revised form 14 July 2005; accepted 27 July 2005 Available online 2 September 2005
Abstract Norepinephrine (NE) has demonstrated proconvulsant and antiepileptic properties; however, the specific pharmacology of these actions has not been clearly established. To address this, we studied the effect of NE on hippocampal CA3 epileptiform activity. Frequency changes of burst discharges in response to NE were biphasic; low concentrations increased the number of bursts, while higher concentrations reduced their frequency, suggesting the involvement of multiple adrenergic receptor (AR) types. This hypothesis was confirmed when, in the presence of AR blockade, increasing concentrations of NE caused a monophasic decrease in epileptiform activity. Antagonists selective for ␣1 or ␣2 ARs were then used to determine which ␣AR type was involved. While discriminating concentrations of the ␣1 AR antagonists prazosin and terazosin had no effect, selective amounts of the ␣2 AR antagonists RS79948 and RX821002 significantly reduced the potency of NE in decreasing epileptiform activity. Furthermore, this antiepileptic action of NE persisted when all GABA-mediated inhibition was blocked. This data suggests that, under conditions of impaired GABAergic inhibition, the excitatory and inhibitory effects of NE on hippocampal CA3 epileptiform activity are mediated primarily via  and ␣2 ARs, respectively. Moreover, our results imply that the antiepileptic effect of ␣2 AR activation in CA3 is not dependent on the GABAergic system. © 2005 Elsevier B.V. All rights reserved. Keywords: Norepinephrine; Hippocampal CA3 region; ␣2 Adrenergic receptor; Field potentials; Bicuculline; Epileptiform burst activity
1. Introduction ∗ Corresponding authors. Tel.: +1 701 777 4293; fax: +1 701 777 4490. E-mail addresses: [email protected] (J.E. Porter), [email protected] (V.A. Doze).
0920-1211/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.eplepsyres.2005.07.007
Norepinephrine (NE), an endogenous neurotransmitter, has profound effects on seizure activity (Giorgi et al., 2004). While these actions may have poten-
118
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
tial therapeutic relevance, the specific pharmacology of these effects has not been unequivocally established (Weinshenker and Szot, 2002). This lack of knowledge has hindered the determination of these underlying mechanisms and limited their possible clinical application. Delineating which adrenergic receptors (ARs) mediate the various actions of NE would not only be extremely useful for examining the role of NE in epilepsy, but could be exploited pharmacologically for future antiepileptic drug development. Among the major targets of the NE system in the brain is the hippocampus (Loy et al., 1980), which contains one of the highest densities of NE-containing terminals (Schroeter et al., 2000). As part of the limbic system, the hippocampus is an essential component of the neural circuitry which governs emotions, attention, and certain memory processes (Milner et al., 1998; Aston-Jones et al., 1999; Eichenbaum, 2000). The hippocampus also plays a major role in epilepsy. It has an extremely low seizure threshold and thus is frequently involved in hyperexcitable episodes (Johnston and Amaral, 2004). Furthermore, the disinhibited hippocampus is often used as a model of acute focal epilepsy (Schwartzkroin, 1986), since the excitatory pyramidal neurons in the hippocampus, particularly those in the CA3 region, can develop synchronous depolarizations under conditions of impaired synaptic inhibition. The NE system has been demonstrated to play a critical role in modifying epileptic activity (Chauvel and Trottier, 1986; Weinshenker and Szot, 2002). Stimulation of ARs has profound anticonvulsant properties in many seizure models (Weiss et al., 1990; Ferraro et al., 1994). Conversely, destruction of the NE system using pharmacological or transgenic techniques impairs the ability to prevent seizures (Arnold et al., 1973; Weinshenker and Szot, 2002). Furthermore, an intact NE system appears to be requisite for the antiepileptic effects of the ketogenic diet and vagal nerve stimulation (Krahl et al., 1998; Szot et al., 2001). It should be noted, however, that depending on the epilepsy model and brain region studied, NE exhibits both proconvulsant and antiepileptic actions (Weinshenker and Szot, 2002). Uncovering the multiple functions of NE has been complicated by the diversity of ARs that it binds and activates. Based on both pharmacological and molecular biological evidence, ARs are classified into three major
types (␣1 , ␣2 , and ), each of which consists of three subtypes (Bylund et al., 1994). Although all ARs are coupled to G proteins, each receptor type appears to have its own distinct pharmacological characteristics (Pupo and Minneman, 2001). Activation of these different AR types is often seen to produce different, if not opposing, effects within the same cell or system. The effect of AR activation in the rat hippocampus is still under debate, as most of the past investigations used either non-selective pharmacological agents or concentrations of agents which were not discriminatory. While several studies have suggested that NE’s antiepileptic actions are mediated through ␣1 AR activation (Stanton et al., 1992; Rutecki, 1995), some have indicated a role for ARs (Ferraro et al., 1994), and others, ␣2 AR involvement (Stoop et al., 2000). Conversely, other studies found NE to be proepileptic, further confounding the role of AR stimulation in the hippocampus (Mueller and Dunwiddie, 1983; Leung and Miller, 1988). The specific mechanisms underlying the antiepileptic actions of NE in the hippocampus are also not known. A potential mechanism could be a NEmediated increase in synaptic inhibition. In the hippocampus, the synaptic inhibition is generated by the inhibitory neurotransmitter, ␥-amino-butyric acid (GABA) acting through GABAA and GABAB receptors. Since NE increases the release of GABA from hippocampal interneurons (Doze et al., 1991; Bergles et al., 1996), the NE-enhanced GABA release would be expected to increase GABA receptormediated inhibition, and thus could reduce hyperexcitability. The aim of this study was to determine the AR types mediating the various actions of NE on hippocampal CA3 burst activity using selective concentrations of AR antagonists. Another goal was to clarify the role of the GABAergic system in the anticonvulsant effects of NE. Our results indicate that, under conditions where the GABAergic inhibition is impaired, the excitatory and inhibitory effects of NE on hippocampal CA3 epileptiform activity are primarily due to activation of  and ␣2 ARs, respectively. Furthermore, these results imply that enhanced synaptic inhibition is not the underlying mechanism for the antiepileptic effects of NE-mediated through ␣2 AR activation.
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
2. Materials and methods 2.1. Animals Sprague-Dawley rats, postnatal day 14–32 (weighing 27–95 g) were housed with their mothers in cages (size 16.5 in. × 8.5 in.) kept in rooms maintained at a temperature of ∼22 ◦ C with a relative humidity of ∼55%. Water and dried laboratory food (Teklad Global 18% Protein Rodent Diet) from Harlan Teklad (Madison, WI, USA) were provided ad libitum. Lighting was set to a 12:12 h light-dark cycle (lights on at 7:00 a.m.). Rats were allowed to acclimate for four days after arrival from Harlan (Madison, WI, USA) prior to their use. All protocols described were approved by the Institutional Animal Care and Use Committee of the University of North Dakota in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals. 2.2. Reagents The pharmacological agents and their sources were: (−)-norepinephrine (+)-bitartrate, phentolamine methanesulfonate (Sigma–Aldrich, St. Louis, MO, USA); (−)-bicuculline methobromide, CGP55845 [(2S)-3-[[(1S)-1-(3,4-dichlorophenyl)ethyl]amino-2hydroxypropyl](phenylmethyl) phosphinic acid], pindolol, prazosin hydrochloride, RS79948 [(8aR,12aS, 13aS)-5,8,8a,9,10,11,12,12a,13,13a-dechydro-3-methoxy-12-(ethylsulfonyl)-6H-isoquino[2,1-g][1,6]naphthyridine hydrochloride], RX821002 [2-(2,3-dihydro-2-methoxy-1,4-benzodioxin-2-yl)-4,5-dihydro1H-imidazole hydrochloride], terazosin hydrochloride (Tocris Cookston Inc., Ellisville, MO, USA); and isoflurane (Abbott Laboratories, North Chicago, IL, USA). All reagents used to make the artificial cerebrospinal fluid (ACSF) were from J.T. Baker, Inc. (Phillipsburg, NJ, USA). 2.3. Hippocampal slice preparation Hippocampal brain slices were prepared as follows. Sprague-Dawley rats were deeply anesthetized with isoflurane, sacrificed by decapitation, and their brains rapidly removed. Hippocampi were dissected from each hemisphere and placed into a beaker of ice-cold saline solution containing, in mM: choline
119
chloride 110, KCl 2.5, MgSO4 7, CaCl2 0.5, NaH2 PO4 1.25, NaHCO3 25, glucose 25, sodium ascorbate 11.6, and sodium pyruvate 3.1. The hippocampi were sectioned transversely at 500 m intervals using a conventional tissue sectioning apparatus (Stoelting, Wood Dale, IL). The hippocampal slices were incubated at 34 ± 1 ◦ C in ACSF containing in mM: NaCl 119, KCl 5, MgSO4 1.3, CaCl2 2.5, NaH2 PO4 1, NaHCO3 26.2, and glucose 11, for 30 min before being allowed to recover for at least an additional 30 min at room temperature (22 ± 1 ◦ C) before experimentation. All solutions were continually aerated with 95% O2 :5% CO2 . 2.4. Electrophysiological recordings A single slice was transferred to the recording chamber, where it was submerged and continuously superfused at a rate of ∼4 ml/min with ACSF. Glass microelectrodes were made using a two-stage puller (PP-830, Narashige, Japan). Extracellular field potentials were recorded using microelectrodes filled with 3 M NaCl and placed in the stratum pyramidale of the CA3 region of the hippocampus. Potentials were detected using an Axoclamp 2B (Axon Instruments, Union City, CA), amplified at 10× or 100× using a Brownlee 440 instrumentation amplifier and signal conditioner (Brownlee Precision, San Jose, CA), digitized at 1 kHz with a Digidata 1322A analog-todigital converter (Axon Instruments), and recorded using Axoscope 9.0 software (Axon Instruments). All experiments were performed at room temperature (22 ± 1 ◦ C). 2.5. Generation of epileptiform burst activity To elicit epileptiform burst discharges in the hippocampal CA3 region, slices were continuously perfused with ACSF containing 20 M of the GABAA receptor antagonist, bicuculline, for the entire duration of the experiment. Using this model of epileptiform activity allowed comparison of our results with other pharmacological studies that used this same model system. Furthermore, this model also allowed for the examination of NE’s antiepileptic effects under conditions in which the influence of the inhibitory GABAergic interneurons had been pharmacologically removed.
120
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
2.6. Drug application Hippocampal slices were continually perfused with ACSF containing 20 M bicuculline and the applicable AR ligands. If no burst discharges were detected after 20 min of perfusion, the slices were discarded. Once burst discharges were evident, at least 30 min of baseline data was recorded before exposure to NE. Preliminary experiments were conducted to assure that AR antagonists had no effect of their own, and that each NE concentration produced a maximal effect in the allotted 8 min time period (data not shown). AR antagonists were continuously applied to the bath containing ACSF starting at least 30 min prior to the application of NE. All drugs used in this study were made fresh daily and dissolved in water, except for CGP55845 and pindolol, which were solubilized in DMSO and 0.01% HCl, respectively. All drugs were diluted at 1:1000 directly into the superfusing ACSF and to minimize oxidation, NE was added to the ACSF just before its application to the hippocampal slice. 2.7. Data analysis Burst discharge frequency was analyzed using Mini Analysis 6.0 (Synaptosoft, Decatur, GA). Frequency versus receptor agonist concentration data was then entered into Prism 4.0 (GraphPad Software, San Diego, CA) and concentration–response curves constructed using a non-linear least squares curve fitting method. Concentration–response curves for each receptor agonist were plotted as a percent of its own maximal response and fit with either a standard (slope = unity) or variable slope. The curve that best-fit the data was determined using an F-test with a P < 0.05 value considered as significant. In all instances, the standard slope curve best fit the individual raw data. From these curves a concentration of NE that produced a half maximal response (EC50 ) was calculated and used as a measurement of potency. A modified Schild analysis was used to functionally estimate the equilibrium dissociation constant (pKb ) for AR antagonists (Arunlakshana and Schild, 1959). For each experiment, cumulative NE concentration–response curves were performed in the absence and presence of a single concentration of AR antagonist. The logarithm of the dose-ratio using NE EC50 values calculated from these concentration–response curves was plotted ver-
sus the log of the AR antagonist concentration. A line with a slope of unity was drawn through this plotted point with the x-intercept representing an estimate of the pKb for the AR antagonist used to inhibit NE’s effects on burst discharges. For these experiments, cumulative concentration–response curves were performed in a single hippocampal slice (three curves per slice). Between curves, the slice was washed until the frequency of epileptiform bursts returned to baseline levels. Significance between groups was tested using an unpaired two-tailed Student’s t-test with a P < 0.05 value considered as significant. Measurements are expressed as the mean ± S.E. for n number of experiments, unless otherwise indicated. 3. Results 3.1. Adrenergic effects on epileptiform activity To study the effect of adrenergic activation on epileptiform activity, we first examined the effects of NE alone on spontaneous epileptiform burst discharges recorded extracellularly from the CA3 pyramidal cell layer of the rat hippocampus. Application of increasing concentrations of NE from 1 nM to 300 M produced profound changes in the frequency of these events (Fig. 1). As illustrated in Fig. 1A, the effects of NE were biphasic (n = 12), with an increase in the number of burst discharges at low concentrations (<1 M) and a decrease in their frequency at higher concentrations (≥1 M). Concentrations of NE ≤30 nM had no observable effect. Furthermore, the effects of NE were best illustrated using separate concentration–response curves for the excitatory and inhibitory phases, respectively (Fig. 1B). The biphasic nature of this response has been demonstrated by other investigators who have used a range of single NE concentrations to selectively support their hypotheses (Mueller et al., 1981). However, the use of cumulative NE concentrations in this single study suggests that multiple AR types may be responsible for NEs effects on epileptiform activity recorded from the hippocampal CA3 region. 3.2. Role of ␣ and AR types on epileptiform activity To further establish which AR types are involved, we pharmacologically isolated the action of a single NE
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
Fig. 1. Effects of NE on epileptiform burst activity. (A) Illustrated is a plot of the number of burst discharges vs. time for experiments in which NE was applied at increasing concentrations. Each bin represents the frequency averaged over a 150 s epoch. NE was applied for the periods (∼8 min) indicated by the bars. Burst discharges were elicited by including 20 M of the GABAA receptor blocker, bicuculline, in the perfusing ACSF. Under these conditions, increasing concentrations of NE had a biphasic effect; with lower concentrations causing an increase in the frequency of spontaneous burst discharges, and higher concentrations producing a reduction in epileptiform activity. All of these effects were completely reversible with ACSF wash, in that the frequency of events would return to the baseline level. Measurements are presented as the mean ± S.E. for n = 12 experiments. (B) Concentration–response curves derived from the frequency histogram (i.e., plot of burst discharge frequency vs. NE concentration). The data points are plotted as percent change in epileptiform activity. The effects of NE were best illustrated using separate concentration–response curves for the excitatory and inhibitory phases, respectively.
concentration on hippocampal CA3 burst discharge frequency (Table 1). Application of 10 M NE to the perfused ACSF caused a modest but significant reduction in the frequency of burst discharges when compared to control. The discharge frequency of 0.055 ± 0.01 Hz
121
in the presence of 10 M NE (n = 10) represented a significant reduction in activity frequency when compared to the 0.079 ± 0.01 Hz control recording (n = 10) and is similar to the diminished burst rate observed by other investigations for this single concentration of NE (Rutecki, 1995; Stoop et al., 2000). However, the inhibitory effect of 10 M NE on epileptiform activity was more effective following pretreatment with a blocking concentration of the AR antagonist, pindolol, when compared to slices not pretreated with AR antagonists. The 0.009 ± 0.003 Hz frequency of burst discharges in the presence of 3 M pindolol (n = 7) is significantly different from the 0.055 ± 0.01 Hz recorded for 10 M NE in the absence of AR antagonist (n = 10) and suggests that activation of an ␣AR type is mediating this inhibitory effect of NE on epileptiform activity. In contrast, 10 M NE caused an increase in epileptiform activity in slices pretreated with an effective concentration of the ␣AR antagonist phentolamine, when compared to the effect of 10 M NE alone. Pretreatment with 10 M phentolamine caused a 0.118 ± 0.01 Hz frequency of burst activity in response to 10 M NE (n = 7) that was significantly increased from the NE response in control slices (0.055 ± 0.01 Hz; n = 10), indicating a role for AR activation leading to hippocampal CA3 hyperexcitability. Together, these results suggest that the biphasic effects of NE on hippocampal CA3 epileptiform burst activity shown in Fig. 1 are mediated via  (excitatory) and ␣ (inhibitory) AR types, respectively. 3.3. Potential involvement of inhibitory GABAergic system Although GABAA receptors have been blocked in this model of epileptiform activity, a possible mechanism for ␣AR-mediated reduction in burst discharge frequency could still be enhanced GABAB receptormediated inhibition secondary to a NE-mediated increase in GABA release (Doze et al., 1991; Bergles et al., 1996). To address this possibility, we examined the effect of NE on epileptiform activity in the presence of both GABAB and AR blockade, in addition to the GABAA receptor blockade (Table 1). Pretreatment with 10 M of the selective GABAB receptor antagonist, CGP55845 (Davies et al., 1993) and 3 M pindolol, followed by application of 10 M NE still caused a significant reduction in the frequency
122
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
Table 1 Effects of AR, ␣AR and complete GABAergic blockade on the NE-mediated epileptiform burst discharge frequency recorded from the hippocampal CA3 region AR ligands
Control (Hz)
Treatment (Hz)
Maximal % change
Norepinephrine (10 M) Norepinephrine (10 M) + pindolol (3 M) Norepinephrine (10 M) + phentolamine (10 M) Norepinephrine (10 M) + pindolol (3 M) + CGP55845 (10 M)
0.079 ± 0.01 (n = 10) 0.066 ± 0.008 (n = 7) 0.079 ± 0.008 (n = 7) 0.067 ± 0.007 (n = 5)
0.055 ± 0.01 (n = 10)a 0.009 ± 0.003 (n = 7)a,b 0.118 ± 0.01 (n = 7)a,b 0.012 ± 0.007 (n = 5)a
−35.4 −87.5 55.6 −84.0
± ± ± ±
13.8 4.0 13.5 10.1
Data is presented as the frequency ± S.E. (in Hz). Control frequency (bursts/s) was measured in the CA3 region after pretreatment with receptor antagonists but before NE application. a Represents significance vs. control response. b Represents significance vs. 10 M NE alone response.
of burst discharges (0.012 ± 0.007 Hz; n = 5) when compared to control (0.067 ± 0.007 Hz; n = 5). Moreover, this ␣AR-mediated inhibitory effect of 10 M NE on epileptiform activity was not significantly different from responses documented in the absence of GABAB receptor blockade (0.009 ± 0.003 Hz; n = 7). These results suggest that a NE-mediated increase in GABA release is not the underlying mechanism for the ␣AR inhibitory effect on burst discharges in this model of epileptiform activity. 3.4. ␣AR-mediated decrease of epileptiform activity To specifically examine the inhibitory effect of cumulative NE concentrations on epileptiform activity, we pharmacologically blocked the contribution of AR activation in this preparation by pretreating the hippocampal slice with 3 M pindolol. Increasing concentrations of NE in the presence of pindolol, produced a concentration-dependent reduction in epileptiform burst discharge frequency without affecting the amplitude of these events (Fig. 2). A frequency histogram of the NE-induced decrease in burst discharges (Fig. 3A) was used to construct a plot of maximal burst discharge frequency versus NE concentration (Fig. 3B). For this particular experiment, a concentration–response curve was fit best with a standard slope, non-linear regression model, which was then used to calculate an EC50 value for NE of 1.2 M. Corresponding experiments were performed on hippocampal slices from different animals (n = 8) resulting in a calculated EC50 for NE to inhibit the frequency of epileptiform burst discharges of 1.2 ± 0.4 M. Similar results were observed when effective concentrations of other AR antagonists,
including 1 M propranolol or 1 M timolol, were used to pharmacologically isolate the NE-mediated effects of ␣AR activation (data not shown). These data confirm that NE inhibits epileptiform burst discharge activity through activation of an ␣AR population in the CA3 region of the hippocampus. 3.5. Role of ␣2 AR activation in CA3 epileptiform activity To determine which type of ␣AR mediates the inhibitory effect on CA3 burst discharges, we used discriminating concentrations of selective ␣1 or ␣2 AR competitive antagonists, based on their affinities for ␣AR subtypes, in an attempt to moderately limit the potency of NE. For these experiments, NE concentration–response curves in the presence of 3 M pindolol () were generated, with 1 nM of the selective ␣1 AR antagonist prazosin (), or 1 nM of the selective ␣2 AR antagonist, RS79948 () and 1 nM of prazosin (Fig. 4). The potency of NE to inhibit CA3 burst discharges in the presence of 3 M pindolol (1.2 ± 0.4 M; n = 3) was not significantly different when 1 nM prazosin was added to the perfusing ACSF (1.0 ± 0.3 M; n = 3). However, there was a significant decrease in the potency of NE when 1 nM of RS79948 was added to the AR antagonist cocktail (2.7 ± 1.3 M; n = 3). This concentration of RS79948 is relative to the calculated equilibrium dissociation constant for this selective AR antagonist when used by other investigators to identify rat ␣2 ARs (Clark et al., 1989; Uhl´en et al., 1998). In similar experiments, NE concentration–response curves in the presence of 3 M pindolol () were generated, with 10 nM of the selective ␣1 AR antagonist ter-
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
Fig. 2. Effects of ␣AR activation on epileptiform burst activity. Illustrated are 150 s long chart recordings of spontaneously occurring epileptiform discharges produced by 20 M bicuculline (BIC). In the presence of AR blockade with 3 M pindolol (PIN), application of increasing concentrations of NE reduced the burst frequency from 0.067 Hz bursts in control ringer to 0.047 Hz in 0.3 M NE to 0.027 Hz in 1 M NE to 0.013 Hz in 3 M NE to ≤0.007 Hz in 10 M NE. This effect was completely reversed with wash. Although not shown here, no epileptiform burst discharges were observed for a >3 min period in the presence of 10 M NE.
azosin (), or 10 nM of the selective ␣2 AR antagonist RX821002 () and 10 nM of terazosin (Fig. 5). Again, the potency of NE to inhibit burst discharges in the presence of 3 M pindolol (1.3 ± 0.3 M; n = 3) was not significantly different when 10 nM terazosin was added to the perfusing ACSF (1.5 ± 0.5 M; n = 3). However, there was a significant decrease in the potency of NE
123
when 10 nM of RX821002 was added to the AR antagonist cocktail (9.0 ± 2.9 M; n = 4). This concentration of RX821002 is approximately 10-fold greater than the calculated equilibrium dissociation constant for this selective AR antagonist when used by other investigators to identify rat ␣2 ARs (Welbourn et al., 1986; Uhl´en et al., 1998). The ability of selective ␣2 and not ␣1 AR antagonists to significantly lessen the potency of NE implies that activation of ␣2 ARs are important for mediating the inhibition of network activity in the hippocampal CA3 region. In Figs. 4 and 5, parallel rightward shifts for concentration–response curves in the presence of ␣2 AR blockade, with no decrease in the NE maximal response when compared to control, satisfy an important criteria for calculating an equilibrium dissociation constant (pKb ) using the method of Schild (Arunlakshana and Schild, 1959). If a line generated with a slope of unity is produced from a Schild plot using the data described in Figs. 4 and 5, a one-point pKb value can be calculated which estimates the equilibrium dissociation constant of ␣2 AR antagonists for the AR causing inhibition of hippocampal CA3 hyperexcitability. The 2.7 and 6.9-fold NE EC50 dose ratio calculated in the presence of RS79948 and RX821002, relate to estimated Kb values of 0.9 ± 0.3 and 2.0 ± 0.9 nM, respectively for these selective ␣2 AR antagonists. These estimated values for RS79948 and RX821002 are comparable to affinities calculated by others when using these AR antagonists to identify ␣2 ARs in their investigations (Welbourn et al., 1986; Clark et al., 1989; Uhl´en et al., 1998).
4. Discussion In the present study, we show pharmacological evidence for an ␣2 AR-mediated inhibition of hippocampal CA3 network activity. Increasing concentrations of the endogenous AR agonist, NE, produced a biphasic change in the rate of CA3 hippocampal epileptiform burst discharges, suggesting that more than one type of AR was mediating these effects. This hypothesis was confirmed when, in the presence of AR blockade, NE produced only a decrease in the frequency of epileptiform activity, while in the presence of ␣AR blockade, NE caused only an increase in the number of burst discharges. Furthermore, these effects of
124
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
Fig. 3. Determination of NE potency to cause reduced hippocampal CA3 burst activity. Illustrated is the frequency histogram (A) used to derive the concentration–response curve (B). (A) Plot of the number of burst discharges vs. time for an experiment in which NE was applied at increasing concentrations. Each bin represents the frequency averaged over a 150 s epoch. In the presence of 3 M pindolol, NE was bath-applied for the periods (∼8 min) indicated by the bars. (B) Concentration–response curve derived from the frequency histogram (i.e., plot of burst discharge frequency vs. NE concentration). The data points are plotted as percent of maximal inhibitory response and the curve was constructed using non-linear regression analysis. For this particular experiment, the concentration–response curve for NE was fit best by a standard slope model with a calculated EC50 value of 1.2 M.
NE were not dependent on a functioning GABAergic system. Finally, there was no significant change in EC50 values for NE calculated from cumulative concentration–response curves generated in the presence of selective ␣1 AR antagonists when compared to control. Conversely, there was a significant difference in the potency of NE between control experiments and those performed in the presence of selective ␣2 AR antagonists. In addition, estimations of the equilibrium dissociation constant for these selective ␣2 AR antagonists were similar to values calculated by others when used to identify rat ␣2 AR populations. Therefore, our results support the concept that activation of a population of ␣2 ARs directly inhibits the network activity of the hippocampal CA3 region. A potential mechanism of NE-mediated reduction in epileptiform activity could be enhanced GABA release from hippocampal interneurons (Doze et al., 1991; Bergles et al., 1996). Although GABAA receptors were blocked in the epilepsy model used in this study, increased GABA release would still be expected to enhance GABAB receptor-mediated inhibition, which in turn, might reduce hippocampal excitability. However, the fact that the antiepileptic effects of NE persisted in the presence of GABAB receptor blockade, would seem to rule out this possibility. Furthermore, if increased GABA-mediated inhibition had been the underlying mechanism in this model, we would have
expected ␣1 ARs to have been involved, as we have previously shown (Bergles et al., 1996). However, there was no difference in the NE-mediated inhibition of CA3 activity in the presence of ␣1 AR antagonists when compared to control. In fact, we wonder if the reason an ␣1 AR-mediated inhibition of the hippocampal CA3 burst discharges was not observed was because the influence of inhibitory GABAergic interneurons had been pharmacologically removed. To examine this possibility, investigations using epilepsy models in which the GABAergic inhibition has not been impaired are currently being performed. The effects of NE on epileptiform activity not only generated monophasic responses in the presence of AR blockade, but the responses were also considerably larger. For example, 10 M NE alone caused a modest 35% reduction in the frequency of burst discharges very similar in magnitude to that observed previously by other investigators (Rutecki, 1995; Stoop et al., 2000). However, after pretreatment with 3 M of the selective AR antagonist, pindolol, 10 M NE essentially inhibited all epileptiform burst activity. This increased effect of NE was likely due to pharmacological isolation of the ␣AR response (i.e., blocking the AR-mediated excitatory action of NE), thereby leaving the ␣AR-mediated inhibitory effect unimpeded. Furthermore, this increased NE response allowed for the generation of reliable concentration–response
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
125
Fig. 4. NE concentration–response curves in the presence of prazosin and RS79948. Prazosin has an equilibrium dissociation constant for ␣1 ARs of ∼0.3 nM and an ␣1 /␣2 AR selectivity between 20- and 9000-fold depending on the ␣2 AR subtype (Doxey et al., 1984; Ireland et al., 1997). In contrast, RS79948 has an equilibrium dissociation constant for ␣2 ARs of ∼0.7 nM and an ␣2 /␣1 AR selectivity of about 15,000-fold (Clark et al., 1989; Uhl´en et al., 1998). Increasing amounts of NE were used to generate concentration–response curves associated with the inhibition of epileptiform discharge frequency for rat hippocampal slices in the absence of antagonist (), with 1 nM prazosin (), or in the presence of both 1 nM RS79948 and 1 nM prazosin (). The calculated EC50 values from these curves showed no difference for the potency of NE in the presence of prazosin (1.0 ± 0.3 M) when compared to control (1.2 ± 0.4 M). Conversely, there was a significant difference in the EC50 calculated for NE (2.7 ± 1.3 M) in the presence RS79948 and prazosin when compared to control. All experiments were performed in the presence of 20 M bicuculline to generate epileptiform burst activity and 3 M pindolol to block ARs. Measurements are presented as the mean ± S.E. for n = 3 experiments.
Fig. 5. NE concentration–response curves in the presence of terazosin and RX821002. Terazosin has an equilibrium dissociation constant for ␣1 ARs of ∼3 nM and an ␣1 /␣2 AR selectivity between 10and 500-fold depending on the ␣2 AR subtype (Hancock et al., 1995). In contrast, RX821002 has an equilibrium dissociation constant of ∼0.8 nM and an ␣2 /␣1 AR selectivity of about 80-fold (Welbourn et al., 1986; Uhl´en et al., 1998). Increasing amounts of NE were used to generate concentration–response curves associated with the inhibition of epileptiform discharge frequency for rat hippocampal slices in the absence of antagonist (), with 10 nM terazosin (), or in the presence of both 10 nM RX821002 and 10 nM terazosin (). The EC50 calculated from these curves showed no difference for the potency of NE in the presence of terazosin (1.5 ± 0.5 M) when compared to control (1.3 ± 0.3 M). Conversely, there was a significant difference in the EC50 calculated for NE (9.0 ± 2.9 M) in the presence of RX821002 and terazosin when compared to control. All experiments were performed in the presence of 20 M bicuculline to generate epileptiform burst activity and 3 M pindolol to block ARs. Measurements are presented as the mean ± S.E. for n = 3–4 experiments.
curves, which consequently permitted a more detailed and accurate pharmacological analysis. While several studies have evaluated the pharmacological effect of NE in models of epilepsy, the specific type of AR involved has not yet been unequivocally established. According to Giorgi et al. (2004), conflicting results have been reported not only because there are significant differences in AR distribution between species, but also because previous investigations have used non-discriminating high concentrations of selective AR ligands. In this study, we were able to avoid these problems by using concentrations of AR antagonists selective for only one ␣AR type based on the equilibrium dissociation constant calculated for native and recombinant receptors. For example, prazosin and terazosin, when used at con-
centrations specific for blocking ␣1 ARs, did not shift the NE concentration–response curve. However, addition of RS79948 or RX821002, at concentrations that would selectively inhibit ␣2 ARs, significantly shifted the NE concentration–response curve to the right when compared with control. These results suggested that the antiepileptic effects of NE in the hippocampal CA3 region are mediated through ␣2 AR activation. NE potency changes in the presence of selective ␣2 AR antagonists compared to control were used to calculate a one-point Kb value using a modified method of Schild (Arunlakshana and Schild, 1959). The estimated Kb values of 0.9 ± 0.3 and 2.0 ± 0.9 nM for RS79948 and RX821002, respectively, are similar to equilibrium dissociation constants calculated by others for these selective ␣2 AR antagonists (Welbourn et al., 1986; Clark
126
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
et al., 1989; Uhl´en et al., 1998). Our results using discriminating concentrations of selective ␣AR antagonists suggest that in this model of epileptiform activity, the NE-mediated decrease in hippocampal CA3 hyperexcitability is due primarily to activation of ␣2 ARs. In the rat hippocampus, antagonism of GABAA receptors creates a bursting phenomenon in the hippocampal CA3 region commonly referred to as epileptiform activity. This activity is thought to originate from the extensive recurrent collateral connections between CA3 pyramidal cells (Li et al., 1994; Johnston and Amaral, 2004). Under conditions of GABAergic inhibition, these profuse glutamatergic connections allow for rapid synchronization of pyramidal cell firing. It has been hypothesized that a similar process occurs in temporal lobe epilepsy, where the hippocampal CA3 region is thought to be a common focus (Yaari and Beck, 2002). As such, this epileptiform model of neuronal synchronization is often used as a model for temporal lobe epilepsy, as well as for other epilepsies. In general, pharmacological agents which are seen to decrease the frequency of epileptiform bursting are believed to be anticonvulsant, while drugs which increase epileptiform bursting are thought to be proconvulsant. In fact, there is a good correlation between the effectiveness of antiepileptic agents in vivo and the effectiveness of antiepileptic agents in hippocampal CA3 epileptiform models (Schneiderman and Schwartzkroin, 1982; Ault et al., 1986; Smith and Swann, 1987). Based on this, we speculate that ␣2 AR-mediated inhibition of hippocampal CA3 burst activity could account for some of the anticonvulsant properties of NE observed in vivo. NE presents an exciting prospect in the treatment of epilepsy. Evidence suggests that, unlike most current antiepileptic drugs, NE does not inhibit such cognitive functions as learning and memory (Devauges and Sara, 1991; Thomas and Palmiter, 1997). Indeed, NE may even facilitate certain types of memory processes (O’Carroll et al., 1999; Murchison et al., 2004). Furthermore, indirect evidence suggests that NE could be effective against seizures that are refractory to current antiepileptic drugs. As mentioned previously, both the ketogenic diet and vagal nerve stimulation treatment modalities currently used in patients with medically intractable epilepsies cease to provide their anticonvulsant effects in the absence of an intact NE system (Krahl et al., 1998; Weinshenker and Szot, 2002).
In this study, we found that NE both increased and decreased hippocampal CA3 epileptiform activity depending upon the concentration. The enhancement of this activity occurs at lower doses, while the mitigation of epileptiform activity predominates at higher doses. We speculate that these differential effects at different concentrations of NE could account for how NE both enhances certain memory processes and yet is antiepileptic in vivo. In summary, this investigation attempts to further characterize the excitatory and inhibitory effects of NE on hippocampal CA3 network activity. While this study provides only an initial clarification of the AR types and site of action for these NE effects, future studies can now be narrowed in scope to determine the specific mechanisms for ␣2 AR-mediated antiepileptic effects in the hippocampal CA3 region. This knowledge will not only expand our understanding of the role of NE in epilepsy, but may also reveal novel targets for future antiepileptic drug development research.
Acknowledgements We are grateful to Karen L. Cisek and Katie E. Rau for their assistance with the experiments and the manuscript. This investigation was supported in part by North Dakota EPSCoR through NSF grant EPS0132289 (V.A.D.), NIH grant 5P20RR017699 from the COBRE program (V.A.D. and J.E.P.), and the University of North Dakota Faculty Research Seed Money Council (V.A.D. and S.J.P.).
References Arnold, P.S., Racine, R.G., Wise, R.A., 1973. Effects of atropine, reserpine, 6-hydroxydopamine, and handling on seizure development in the rat. Exp. Neurol. 40, 457–470. Arunlakshana, O., Schild, H.O., 1959. Some quantitative uses of drug antagonist. Br. J. Pharmacol. 14, 48–58. Aston-Jones, G., Rajkowski, J., Cohen, J., 1999. Role of locus coeruleus in attention and behavioral flexibility. Biol. Psychiatry 46, 1309–1320. Ault, B., Gruenthal, M., Armstrong, D.R., Nadler, J.V., Wang, C.M., 1986. Baclofen suppresses bursting activity induced in hippocampal slices by differing convulsant treatments. Eur. J. Pharmacol. 126, 289–292. Bergles, D.E., Doze, V.A., Madison, D.V., Smith, S.J., 1996. Excitatory actions of norepinephrine on multiple classes of hippocampal CA1 interneurons. J. Neurosci. 16, 572–585.
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128 Bylund, D.B., Eikenberg, D.C., Hieble, J.P., Langer, S.Z., Lefkowitz, R.J., Minneman, K.P., Molinoff, P.B., Ruffolo, R.R., Trendelenbyurg, A.U., 1994. IV. International Union of Pharmacology, nomenclature of adrenoceptors. Pharmacol. Rev. 46, 121–136. Chauvel, P., Trottier, S., 1986. Role of noradrenergic ascending system in extinction of epileptic phenomena. In: Delgado-Escueta, Ward, Woodbury, Porter (Eds.), Advances in Neurology, vol. 44, Basic Mechanisms of the Epilepsies. Raven Press, New York, pp. 475–487. Clark, R.D., Repke, D.B., Kilpatrick, A.T., Brown, C.M., MacKinnon, A.C., Clague, R.U., Spedding, M., 1989. (8a␣, 12a␣, 13a␣)-5,8,8a,9,10,11,12,12a,13,13a-decahydro-3-methoxy12-(methylsulfonyl)-6H-isoquino[2,1-g][1,6]naphthyridine, a potent and highly selective ␣2 -adrenoceptor antagonist. J. Med. Chem. 32, 2034–2036. Davies, C.H., Pozza, M.F., Collingridge, G.L., 1993. CGP55845A: a potent antagonist of GABAB receptors in the CA1 region of rat hippocampus. Neuropharmacology 32, 1071–1073. Devauges, V., Sara, S.J., 1991. Memory retrieval enhancement by locus coeruleus stimulation: evidence for mediation by betareceptors. Behav. Brain Res. 43, 93–97. Doxey, J.C., Lane, A.C., Roach, A.G., Virdee, N.K., 1984. Comparison of the alpha-adrenoceptor antagonist profiles of idazoxan (RX781094), yohimbine, rauwolscince and corynanthine. Naunyn Schmiedebergs Arch. Pharmacol. 325, 136–144. Doze, V.A., Cohen, G.A., Madison, D.V., 1991. Synaptic localization of adrenergic disinhibition in the rat hippocampus. Neuron 6, 889–900. Eichenbaum, H., 2000. A cortical-hippocampal system for declarative memory. Nat. Rev. Neurosci. 1, 41–50. Ferraro, G., Sardo, P., Sabatino, M., La Grutta, V., 1994. Locus coeruleus noradrenaline system and focal penicillin hippocampal epilepsy: neurophysiology study. Epilepsy Res. 19, 215–220. Giorgi, F.S., Pizzanelli, C., Biagioni, F., Murri, L., Fornai, F., 2004. The role of norepinephrine in epilepsy: from the bench to the bedside. Neurosci. Biobehav. Rev. 28, 507–524. Hancock, A.A., Buckner, S.A., Ireland, L.M., Knepper, S.M., Kerwin Jr., J.F., 1995. Actions of terazosin and its enantiomers at subtypes of ␣1 - and ␣2 -adrenoceptors in vitro. J. Recept. Signal. Transduct. Res. 15, 863–885. Ireland, L.M., Cain, J.C., Rotert, G., Thomas, S., Shoghi, S., Hancock, A.A., Kerwin Jr., J.F., 1997. [3 H]R-terazosin binds selectivity to ␣1 -adrenoceptors over ␣2 -adrenoceptors – comparison with racemic [3 H]terazosin and [3 H]prazosin. Eur. J. Pharmacol. 327, 79–86. Johnston, B., Amaral, D.G., 2004. In: Shepherd (Ed.), Hippocampus, in the Synaptic Organization of the Brain. Oxford University Press, New York, pp. 455–498. Krahl, S.E., Clark, K.B., Smith, D.C., Browning, R.A., 1998. Locus coeruleus lesions suppress the seizure-attenuating effects of vagus nerve stimulation. Epilepsia 39, 709–714. Leung, P.P., Miller, J.J., 1988. Dual role of norepinephrine in the hippocampal CA1 region of the rat: inhibition and disinhibition. Can. J. Physiol. Pharmacol. 66, 814–819. Li, X.G., Somogyi, P., Ylinen, A., Buzsaki, G., 1994. The hippocampal CA3 network: an in vivo intracellular labeling study. J. Comp. Neurol. 339, 181–208.
127
Loy, R., Koziell, D.A., Lindsey, J.D., Moore, R.Y., 1980. Noradrenergic innervation of the adult rat hippocampal formation. J. Comp. Neurol. 189, 699–710. Milner, B., Squire, L.P., Kandel, E.R., 1998. Cognitive neuroscience and the study of memory. Neuron 20, 445–468. Mueller, A.L., Dunwiddie, T.V., 1983. Anticonvulsant and proconvulsant actions of ␣- and -noradrenergic agonists on epileptiform activity in rat hippocampus in vitro. Epilepsia 24, 57– 64. Mueller, A.L., Hoffer, B.J., Dunwiddie, T.V., 1981. Noradrenergic responses in rat hippocampus: evidence for mediation by ␣- and -receptors in the in vitro slice. Brain Res. 214, 113–126. Murchison, C.F., Zhang, X.Y., Zhang, W.P., Ouyang, M., Lee, A., Thomas, S.A., 2004. A distinct role for norepinephrine in memory retrieval. Cell 117, 131–143. O’Carroll, R.E., Drysdale, E., Cahill, L., Shajahan, P., Ebmeier, K.P., 1999. Stimulation of the noradrenergic system enhances and blockade reduces memory for emotional material in man. Psychol. Med. 29, 1083–1088. Pupo, A.S., Minneman, K.P., 2001. Adrenergic pharmacology: focus on the central nervous system. CNS Spectr. 6, 656–662. Rutecki, P.A., 1995. Noradrenergic modulation of epileptiform activity in the hippocampus. Epilepsy Res. 20, 125–136. Schneiderman, J.H., Schwartzkroin, P.A., 1982. Effects of phenytoin on normal activity and on penicillin-induced bursting in the guinea pig hippocampal slice. Neurology 32, 730–738. Schroeter, S., Apparsundaram, S., Wiley, R.G., Miner, L.H., Sesack, S.R., Blakely, R.D., 2000. Immunolocalization of the cocaineand antidepressant-sensitive l-norepinephrine transporter. J. Comp. Neurol. 420, 211–232. Schwartzkroin, P.A., 1986. Hippocampal slices in experimental and human epilepsy. In: Delgado-Escueta, Ward, Woodbury, Porter (Eds.), Advances in Neurology, vol. 44, Basic Mechanisms of the Epilepsies. Raven Press, New York, pp. 991–1010. Smith, K.L., Swann, J.W., 1987. Carbamazepine suppresses synchronized after discharging in disinhibited immature rat hippocampus in vitro. Brain Res. 400, 371–376. Stanton, P.K., Mody, I., Zigmond, D., Sejnowski, T., Heinemann, U., 1992. In: Avanzini, G., Ngel, R., Fariello, R., Heinemann, U. (Eds.), Neurotransmitters in Epilepsy (Epilepsy Res. Suppl. 8), Chapter 40, Noradrenergic Modulation of Excitability in Acute and Chronic Model Epilepsies. Elseiver Science Publishers, New York, pp. 321–334. Stoop, R., Epiney, S., Meier, E., Pralong, E., 2000. Modulation of epileptiform discharges in the rat limbic system in vitro by noradrenergic agents. Neurosci. Lett. 287, 5–8. Szot, P., Weinshenker, D., Rho, J.M., Storey, T.W., Schwartzkroin, P.A., 2001. Norepinephrine is required for the anticonvulsant effect of the ketogenic diet. Brain Res. Dev. Brain Res. 129, 211–214. Thomas, S.A., Palmiter, R.D., 1997. Disruption of the dopamine beta-hydroxylase gene in mice suggests roles for norepinephrine in motor function, learning, and memory. Behav. Neurosci. 11, 579–589. Uhl´en, S., Dambrova, M., Nasman, J., Schioth, H.B., Gu, Y., Wikberg-Matsson, A., Wikberg, J.E.S., 1998. [3 H]RS79948197 binding to human, rat, guinea pig and pig ␣2A -, ␣2B - and
128
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
␣2C -adrenoceptors. Comparison with MK912, RX821002, rauwolscine and yohimbine. Eur. J. Pharmacol. 343, 93–101. Weinshenker, D., Szot, P., 2002. The role of catecholamines in seizure susceptibility: new results using genetically engineered mice. Pharmacol. Ther. 94, 213–233. Weiss, G.K., Lewis, J., Jimenez-Rivera, C., Vigil, A., Corcoran, M.E., 1990. Antikindling effects of locus coeruleus stimulation: mediation by ascending noradrenergic projections. Exp. Neurol. 108, 136–140.
Welbourn, A.P., Chapleo, C.B., Lane, A.C., Myers, P.L., Roach, A.G., Smith, C.F.C., Stillings, M.R., Tulloch, I.F., 1986. ␣Adrenoceptor reagents. 4. Resolution of some potent selective prejunctional ␣2 -adrenoceptor antagonists. J. Med. Chem. 29, 2000–2003. Yaari, Y., Beck, H., 2002. “Epileptic neurons” in temperol lobe epilepsy. Brain Pathol. 12, 234–239.
Adrenergic receptor modulation of hippocampal CA3 network activity Chris W.D. Jurgens a , Sarah J. Boese a , Jacob D. King a , Sally J. Pyle b , James E. Porter a,∗ , Van A. Doze a,∗ a
b
Department of Pharmacology, Physiology and Therapeutics, University of North Dakota School of Medicine and Health Sciences, 501 N. Columbia Road, Grand Forks, ND 58203, USA Department of Anatomy and Cell Biology, University of North Dakota School of Medicine and Health Sciences, 501 N. Columbia Road, Grand Forks, ND 58203, USA Received 28 April 2005; received in revised form 14 July 2005; accepted 27 July 2005 Available online 2 September 2005
Abstract Norepinephrine (NE) has demonstrated proconvulsant and antiepileptic properties; however, the specific pharmacology of these actions has not been clearly established. To address this, we studied the effect of NE on hippocampal CA3 epileptiform activity. Frequency changes of burst discharges in response to NE were biphasic; low concentrations increased the number of bursts, while higher concentrations reduced their frequency, suggesting the involvement of multiple adrenergic receptor (AR) types. This hypothesis was confirmed when, in the presence of AR blockade, increasing concentrations of NE caused a monophasic decrease in epileptiform activity. Antagonists selective for ␣1 or ␣2 ARs were then used to determine which ␣AR type was involved. While discriminating concentrations of the ␣1 AR antagonists prazosin and terazosin had no effect, selective amounts of the ␣2 AR antagonists RS79948 and RX821002 significantly reduced the potency of NE in decreasing epileptiform activity. Furthermore, this antiepileptic action of NE persisted when all GABA-mediated inhibition was blocked. This data suggests that, under conditions of impaired GABAergic inhibition, the excitatory and inhibitory effects of NE on hippocampal CA3 epileptiform activity are mediated primarily via  and ␣2 ARs, respectively. Moreover, our results imply that the antiepileptic effect of ␣2 AR activation in CA3 is not dependent on the GABAergic system. © 2005 Elsevier B.V. All rights reserved. Keywords: Norepinephrine; Hippocampal CA3 region; ␣2 Adrenergic receptor; Field potentials; Bicuculline; Epileptiform burst activity
1. Introduction ∗ Corresponding authors. Tel.: +1 701 777 4293; fax: +1 701 777 4490. E-mail addresses: [email protected] (J.E. Porter), [email protected] (V.A. Doze).
0920-1211/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.eplepsyres.2005.07.007
Norepinephrine (NE), an endogenous neurotransmitter, has profound effects on seizure activity (Giorgi et al., 2004). While these actions may have poten-
118
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
tial therapeutic relevance, the specific pharmacology of these effects has not been unequivocally established (Weinshenker and Szot, 2002). This lack of knowledge has hindered the determination of these underlying mechanisms and limited their possible clinical application. Delineating which adrenergic receptors (ARs) mediate the various actions of NE would not only be extremely useful for examining the role of NE in epilepsy, but could be exploited pharmacologically for future antiepileptic drug development. Among the major targets of the NE system in the brain is the hippocampus (Loy et al., 1980), which contains one of the highest densities of NE-containing terminals (Schroeter et al., 2000). As part of the limbic system, the hippocampus is an essential component of the neural circuitry which governs emotions, attention, and certain memory processes (Milner et al., 1998; Aston-Jones et al., 1999; Eichenbaum, 2000). The hippocampus also plays a major role in epilepsy. It has an extremely low seizure threshold and thus is frequently involved in hyperexcitable episodes (Johnston and Amaral, 2004). Furthermore, the disinhibited hippocampus is often used as a model of acute focal epilepsy (Schwartzkroin, 1986), since the excitatory pyramidal neurons in the hippocampus, particularly those in the CA3 region, can develop synchronous depolarizations under conditions of impaired synaptic inhibition. The NE system has been demonstrated to play a critical role in modifying epileptic activity (Chauvel and Trottier, 1986; Weinshenker and Szot, 2002). Stimulation of ARs has profound anticonvulsant properties in many seizure models (Weiss et al., 1990; Ferraro et al., 1994). Conversely, destruction of the NE system using pharmacological or transgenic techniques impairs the ability to prevent seizures (Arnold et al., 1973; Weinshenker and Szot, 2002). Furthermore, an intact NE system appears to be requisite for the antiepileptic effects of the ketogenic diet and vagal nerve stimulation (Krahl et al., 1998; Szot et al., 2001). It should be noted, however, that depending on the epilepsy model and brain region studied, NE exhibits both proconvulsant and antiepileptic actions (Weinshenker and Szot, 2002). Uncovering the multiple functions of NE has been complicated by the diversity of ARs that it binds and activates. Based on both pharmacological and molecular biological evidence, ARs are classified into three major
types (␣1 , ␣2 , and ), each of which consists of three subtypes (Bylund et al., 1994). Although all ARs are coupled to G proteins, each receptor type appears to have its own distinct pharmacological characteristics (Pupo and Minneman, 2001). Activation of these different AR types is often seen to produce different, if not opposing, effects within the same cell or system. The effect of AR activation in the rat hippocampus is still under debate, as most of the past investigations used either non-selective pharmacological agents or concentrations of agents which were not discriminatory. While several studies have suggested that NE’s antiepileptic actions are mediated through ␣1 AR activation (Stanton et al., 1992; Rutecki, 1995), some have indicated a role for ARs (Ferraro et al., 1994), and others, ␣2 AR involvement (Stoop et al., 2000). Conversely, other studies found NE to be proepileptic, further confounding the role of AR stimulation in the hippocampus (Mueller and Dunwiddie, 1983; Leung and Miller, 1988). The specific mechanisms underlying the antiepileptic actions of NE in the hippocampus are also not known. A potential mechanism could be a NEmediated increase in synaptic inhibition. In the hippocampus, the synaptic inhibition is generated by the inhibitory neurotransmitter, ␥-amino-butyric acid (GABA) acting through GABAA and GABAB receptors. Since NE increases the release of GABA from hippocampal interneurons (Doze et al., 1991; Bergles et al., 1996), the NE-enhanced GABA release would be expected to increase GABA receptormediated inhibition, and thus could reduce hyperexcitability. The aim of this study was to determine the AR types mediating the various actions of NE on hippocampal CA3 burst activity using selective concentrations of AR antagonists. Another goal was to clarify the role of the GABAergic system in the anticonvulsant effects of NE. Our results indicate that, under conditions where the GABAergic inhibition is impaired, the excitatory and inhibitory effects of NE on hippocampal CA3 epileptiform activity are primarily due to activation of  and ␣2 ARs, respectively. Furthermore, these results imply that enhanced synaptic inhibition is not the underlying mechanism for the antiepileptic effects of NE-mediated through ␣2 AR activation.
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
2. Materials and methods 2.1. Animals Sprague-Dawley rats, postnatal day 14–32 (weighing 27–95 g) were housed with their mothers in cages (size 16.5 in. × 8.5 in.) kept in rooms maintained at a temperature of ∼22 ◦ C with a relative humidity of ∼55%. Water and dried laboratory food (Teklad Global 18% Protein Rodent Diet) from Harlan Teklad (Madison, WI, USA) were provided ad libitum. Lighting was set to a 12:12 h light-dark cycle (lights on at 7:00 a.m.). Rats were allowed to acclimate for four days after arrival from Harlan (Madison, WI, USA) prior to their use. All protocols described were approved by the Institutional Animal Care and Use Committee of the University of North Dakota in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals. 2.2. Reagents The pharmacological agents and their sources were: (−)-norepinephrine (+)-bitartrate, phentolamine methanesulfonate (Sigma–Aldrich, St. Louis, MO, USA); (−)-bicuculline methobromide, CGP55845 [(2S)-3-[[(1S)-1-(3,4-dichlorophenyl)ethyl]amino-2hydroxypropyl](phenylmethyl) phosphinic acid], pindolol, prazosin hydrochloride, RS79948 [(8aR,12aS, 13aS)-5,8,8a,9,10,11,12,12a,13,13a-dechydro-3-methoxy-12-(ethylsulfonyl)-6H-isoquino[2,1-g][1,6]naphthyridine hydrochloride], RX821002 [2-(2,3-dihydro-2-methoxy-1,4-benzodioxin-2-yl)-4,5-dihydro1H-imidazole hydrochloride], terazosin hydrochloride (Tocris Cookston Inc., Ellisville, MO, USA); and isoflurane (Abbott Laboratories, North Chicago, IL, USA). All reagents used to make the artificial cerebrospinal fluid (ACSF) were from J.T. Baker, Inc. (Phillipsburg, NJ, USA). 2.3. Hippocampal slice preparation Hippocampal brain slices were prepared as follows. Sprague-Dawley rats were deeply anesthetized with isoflurane, sacrificed by decapitation, and their brains rapidly removed. Hippocampi were dissected from each hemisphere and placed into a beaker of ice-cold saline solution containing, in mM: choline
119
chloride 110, KCl 2.5, MgSO4 7, CaCl2 0.5, NaH2 PO4 1.25, NaHCO3 25, glucose 25, sodium ascorbate 11.6, and sodium pyruvate 3.1. The hippocampi were sectioned transversely at 500 m intervals using a conventional tissue sectioning apparatus (Stoelting, Wood Dale, IL). The hippocampal slices were incubated at 34 ± 1 ◦ C in ACSF containing in mM: NaCl 119, KCl 5, MgSO4 1.3, CaCl2 2.5, NaH2 PO4 1, NaHCO3 26.2, and glucose 11, for 30 min before being allowed to recover for at least an additional 30 min at room temperature (22 ± 1 ◦ C) before experimentation. All solutions were continually aerated with 95% O2 :5% CO2 . 2.4. Electrophysiological recordings A single slice was transferred to the recording chamber, where it was submerged and continuously superfused at a rate of ∼4 ml/min with ACSF. Glass microelectrodes were made using a two-stage puller (PP-830, Narashige, Japan). Extracellular field potentials were recorded using microelectrodes filled with 3 M NaCl and placed in the stratum pyramidale of the CA3 region of the hippocampus. Potentials were detected using an Axoclamp 2B (Axon Instruments, Union City, CA), amplified at 10× or 100× using a Brownlee 440 instrumentation amplifier and signal conditioner (Brownlee Precision, San Jose, CA), digitized at 1 kHz with a Digidata 1322A analog-todigital converter (Axon Instruments), and recorded using Axoscope 9.0 software (Axon Instruments). All experiments were performed at room temperature (22 ± 1 ◦ C). 2.5. Generation of epileptiform burst activity To elicit epileptiform burst discharges in the hippocampal CA3 region, slices were continuously perfused with ACSF containing 20 M of the GABAA receptor antagonist, bicuculline, for the entire duration of the experiment. Using this model of epileptiform activity allowed comparison of our results with other pharmacological studies that used this same model system. Furthermore, this model also allowed for the examination of NE’s antiepileptic effects under conditions in which the influence of the inhibitory GABAergic interneurons had been pharmacologically removed.
120
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
2.6. Drug application Hippocampal slices were continually perfused with ACSF containing 20 M bicuculline and the applicable AR ligands. If no burst discharges were detected after 20 min of perfusion, the slices were discarded. Once burst discharges were evident, at least 30 min of baseline data was recorded before exposure to NE. Preliminary experiments were conducted to assure that AR antagonists had no effect of their own, and that each NE concentration produced a maximal effect in the allotted 8 min time period (data not shown). AR antagonists were continuously applied to the bath containing ACSF starting at least 30 min prior to the application of NE. All drugs used in this study were made fresh daily and dissolved in water, except for CGP55845 and pindolol, which were solubilized in DMSO and 0.01% HCl, respectively. All drugs were diluted at 1:1000 directly into the superfusing ACSF and to minimize oxidation, NE was added to the ACSF just before its application to the hippocampal slice. 2.7. Data analysis Burst discharge frequency was analyzed using Mini Analysis 6.0 (Synaptosoft, Decatur, GA). Frequency versus receptor agonist concentration data was then entered into Prism 4.0 (GraphPad Software, San Diego, CA) and concentration–response curves constructed using a non-linear least squares curve fitting method. Concentration–response curves for each receptor agonist were plotted as a percent of its own maximal response and fit with either a standard (slope = unity) or variable slope. The curve that best-fit the data was determined using an F-test with a P < 0.05 value considered as significant. In all instances, the standard slope curve best fit the individual raw data. From these curves a concentration of NE that produced a half maximal response (EC50 ) was calculated and used as a measurement of potency. A modified Schild analysis was used to functionally estimate the equilibrium dissociation constant (pKb ) for AR antagonists (Arunlakshana and Schild, 1959). For each experiment, cumulative NE concentration–response curves were performed in the absence and presence of a single concentration of AR antagonist. The logarithm of the dose-ratio using NE EC50 values calculated from these concentration–response curves was plotted ver-
sus the log of the AR antagonist concentration. A line with a slope of unity was drawn through this plotted point with the x-intercept representing an estimate of the pKb for the AR antagonist used to inhibit NE’s effects on burst discharges. For these experiments, cumulative concentration–response curves were performed in a single hippocampal slice (three curves per slice). Between curves, the slice was washed until the frequency of epileptiform bursts returned to baseline levels. Significance between groups was tested using an unpaired two-tailed Student’s t-test with a P < 0.05 value considered as significant. Measurements are expressed as the mean ± S.E. for n number of experiments, unless otherwise indicated. 3. Results 3.1. Adrenergic effects on epileptiform activity To study the effect of adrenergic activation on epileptiform activity, we first examined the effects of NE alone on spontaneous epileptiform burst discharges recorded extracellularly from the CA3 pyramidal cell layer of the rat hippocampus. Application of increasing concentrations of NE from 1 nM to 300 M produced profound changes in the frequency of these events (Fig. 1). As illustrated in Fig. 1A, the effects of NE were biphasic (n = 12), with an increase in the number of burst discharges at low concentrations (<1 M) and a decrease in their frequency at higher concentrations (≥1 M). Concentrations of NE ≤30 nM had no observable effect. Furthermore, the effects of NE were best illustrated using separate concentration–response curves for the excitatory and inhibitory phases, respectively (Fig. 1B). The biphasic nature of this response has been demonstrated by other investigators who have used a range of single NE concentrations to selectively support their hypotheses (Mueller et al., 1981). However, the use of cumulative NE concentrations in this single study suggests that multiple AR types may be responsible for NEs effects on epileptiform activity recorded from the hippocampal CA3 region. 3.2. Role of ␣ and AR types on epileptiform activity To further establish which AR types are involved, we pharmacologically isolated the action of a single NE
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
Fig. 1. Effects of NE on epileptiform burst activity. (A) Illustrated is a plot of the number of burst discharges vs. time for experiments in which NE was applied at increasing concentrations. Each bin represents the frequency averaged over a 150 s epoch. NE was applied for the periods (∼8 min) indicated by the bars. Burst discharges were elicited by including 20 M of the GABAA receptor blocker, bicuculline, in the perfusing ACSF. Under these conditions, increasing concentrations of NE had a biphasic effect; with lower concentrations causing an increase in the frequency of spontaneous burst discharges, and higher concentrations producing a reduction in epileptiform activity. All of these effects were completely reversible with ACSF wash, in that the frequency of events would return to the baseline level. Measurements are presented as the mean ± S.E. for n = 12 experiments. (B) Concentration–response curves derived from the frequency histogram (i.e., plot of burst discharge frequency vs. NE concentration). The data points are plotted as percent change in epileptiform activity. The effects of NE were best illustrated using separate concentration–response curves for the excitatory and inhibitory phases, respectively.
concentration on hippocampal CA3 burst discharge frequency (Table 1). Application of 10 M NE to the perfused ACSF caused a modest but significant reduction in the frequency of burst discharges when compared to control. The discharge frequency of 0.055 ± 0.01 Hz
121
in the presence of 10 M NE (n = 10) represented a significant reduction in activity frequency when compared to the 0.079 ± 0.01 Hz control recording (n = 10) and is similar to the diminished burst rate observed by other investigations for this single concentration of NE (Rutecki, 1995; Stoop et al., 2000). However, the inhibitory effect of 10 M NE on epileptiform activity was more effective following pretreatment with a blocking concentration of the AR antagonist, pindolol, when compared to slices not pretreated with AR antagonists. The 0.009 ± 0.003 Hz frequency of burst discharges in the presence of 3 M pindolol (n = 7) is significantly different from the 0.055 ± 0.01 Hz recorded for 10 M NE in the absence of AR antagonist (n = 10) and suggests that activation of an ␣AR type is mediating this inhibitory effect of NE on epileptiform activity. In contrast, 10 M NE caused an increase in epileptiform activity in slices pretreated with an effective concentration of the ␣AR antagonist phentolamine, when compared to the effect of 10 M NE alone. Pretreatment with 10 M phentolamine caused a 0.118 ± 0.01 Hz frequency of burst activity in response to 10 M NE (n = 7) that was significantly increased from the NE response in control slices (0.055 ± 0.01 Hz; n = 10), indicating a role for AR activation leading to hippocampal CA3 hyperexcitability. Together, these results suggest that the biphasic effects of NE on hippocampal CA3 epileptiform burst activity shown in Fig. 1 are mediated via  (excitatory) and ␣ (inhibitory) AR types, respectively. 3.3. Potential involvement of inhibitory GABAergic system Although GABAA receptors have been blocked in this model of epileptiform activity, a possible mechanism for ␣AR-mediated reduction in burst discharge frequency could still be enhanced GABAB receptormediated inhibition secondary to a NE-mediated increase in GABA release (Doze et al., 1991; Bergles et al., 1996). To address this possibility, we examined the effect of NE on epileptiform activity in the presence of both GABAB and AR blockade, in addition to the GABAA receptor blockade (Table 1). Pretreatment with 10 M of the selective GABAB receptor antagonist, CGP55845 (Davies et al., 1993) and 3 M pindolol, followed by application of 10 M NE still caused a significant reduction in the frequency
122
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
Table 1 Effects of AR, ␣AR and complete GABAergic blockade on the NE-mediated epileptiform burst discharge frequency recorded from the hippocampal CA3 region AR ligands
Control (Hz)
Treatment (Hz)
Maximal % change
Norepinephrine (10 M) Norepinephrine (10 M) + pindolol (3 M) Norepinephrine (10 M) + phentolamine (10 M) Norepinephrine (10 M) + pindolol (3 M) + CGP55845 (10 M)
0.079 ± 0.01 (n = 10) 0.066 ± 0.008 (n = 7) 0.079 ± 0.008 (n = 7) 0.067 ± 0.007 (n = 5)
0.055 ± 0.01 (n = 10)a 0.009 ± 0.003 (n = 7)a,b 0.118 ± 0.01 (n = 7)a,b 0.012 ± 0.007 (n = 5)a
−35.4 −87.5 55.6 −84.0
± ± ± ±
13.8 4.0 13.5 10.1
Data is presented as the frequency ± S.E. (in Hz). Control frequency (bursts/s) was measured in the CA3 region after pretreatment with receptor antagonists but before NE application. a Represents significance vs. control response. b Represents significance vs. 10 M NE alone response.
of burst discharges (0.012 ± 0.007 Hz; n = 5) when compared to control (0.067 ± 0.007 Hz; n = 5). Moreover, this ␣AR-mediated inhibitory effect of 10 M NE on epileptiform activity was not significantly different from responses documented in the absence of GABAB receptor blockade (0.009 ± 0.003 Hz; n = 7). These results suggest that a NE-mediated increase in GABA release is not the underlying mechanism for the ␣AR inhibitory effect on burst discharges in this model of epileptiform activity. 3.4. ␣AR-mediated decrease of epileptiform activity To specifically examine the inhibitory effect of cumulative NE concentrations on epileptiform activity, we pharmacologically blocked the contribution of AR activation in this preparation by pretreating the hippocampal slice with 3 M pindolol. Increasing concentrations of NE in the presence of pindolol, produced a concentration-dependent reduction in epileptiform burst discharge frequency without affecting the amplitude of these events (Fig. 2). A frequency histogram of the NE-induced decrease in burst discharges (Fig. 3A) was used to construct a plot of maximal burst discharge frequency versus NE concentration (Fig. 3B). For this particular experiment, a concentration–response curve was fit best with a standard slope, non-linear regression model, which was then used to calculate an EC50 value for NE of 1.2 M. Corresponding experiments were performed on hippocampal slices from different animals (n = 8) resulting in a calculated EC50 for NE to inhibit the frequency of epileptiform burst discharges of 1.2 ± 0.4 M. Similar results were observed when effective concentrations of other AR antagonists,
including 1 M propranolol or 1 M timolol, were used to pharmacologically isolate the NE-mediated effects of ␣AR activation (data not shown). These data confirm that NE inhibits epileptiform burst discharge activity through activation of an ␣AR population in the CA3 region of the hippocampus. 3.5. Role of ␣2 AR activation in CA3 epileptiform activity To determine which type of ␣AR mediates the inhibitory effect on CA3 burst discharges, we used discriminating concentrations of selective ␣1 or ␣2 AR competitive antagonists, based on their affinities for ␣AR subtypes, in an attempt to moderately limit the potency of NE. For these experiments, NE concentration–response curves in the presence of 3 M pindolol () were generated, with 1 nM of the selective ␣1 AR antagonist prazosin (), or 1 nM of the selective ␣2 AR antagonist, RS79948 () and 1 nM of prazosin (Fig. 4). The potency of NE to inhibit CA3 burst discharges in the presence of 3 M pindolol (1.2 ± 0.4 M; n = 3) was not significantly different when 1 nM prazosin was added to the perfusing ACSF (1.0 ± 0.3 M; n = 3). However, there was a significant decrease in the potency of NE when 1 nM of RS79948 was added to the AR antagonist cocktail (2.7 ± 1.3 M; n = 3). This concentration of RS79948 is relative to the calculated equilibrium dissociation constant for this selective AR antagonist when used by other investigators to identify rat ␣2 ARs (Clark et al., 1989; Uhl´en et al., 1998). In similar experiments, NE concentration–response curves in the presence of 3 M pindolol () were generated, with 10 nM of the selective ␣1 AR antagonist ter-
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
Fig. 2. Effects of ␣AR activation on epileptiform burst activity. Illustrated are 150 s long chart recordings of spontaneously occurring epileptiform discharges produced by 20 M bicuculline (BIC). In the presence of AR blockade with 3 M pindolol (PIN), application of increasing concentrations of NE reduced the burst frequency from 0.067 Hz bursts in control ringer to 0.047 Hz in 0.3 M NE to 0.027 Hz in 1 M NE to 0.013 Hz in 3 M NE to ≤0.007 Hz in 10 M NE. This effect was completely reversed with wash. Although not shown here, no epileptiform burst discharges were observed for a >3 min period in the presence of 10 M NE.
azosin (), or 10 nM of the selective ␣2 AR antagonist RX821002 () and 10 nM of terazosin (Fig. 5). Again, the potency of NE to inhibit burst discharges in the presence of 3 M pindolol (1.3 ± 0.3 M; n = 3) was not significantly different when 10 nM terazosin was added to the perfusing ACSF (1.5 ± 0.5 M; n = 3). However, there was a significant decrease in the potency of NE
123
when 10 nM of RX821002 was added to the AR antagonist cocktail (9.0 ± 2.9 M; n = 4). This concentration of RX821002 is approximately 10-fold greater than the calculated equilibrium dissociation constant for this selective AR antagonist when used by other investigators to identify rat ␣2 ARs (Welbourn et al., 1986; Uhl´en et al., 1998). The ability of selective ␣2 and not ␣1 AR antagonists to significantly lessen the potency of NE implies that activation of ␣2 ARs are important for mediating the inhibition of network activity in the hippocampal CA3 region. In Figs. 4 and 5, parallel rightward shifts for concentration–response curves in the presence of ␣2 AR blockade, with no decrease in the NE maximal response when compared to control, satisfy an important criteria for calculating an equilibrium dissociation constant (pKb ) using the method of Schild (Arunlakshana and Schild, 1959). If a line generated with a slope of unity is produced from a Schild plot using the data described in Figs. 4 and 5, a one-point pKb value can be calculated which estimates the equilibrium dissociation constant of ␣2 AR antagonists for the AR causing inhibition of hippocampal CA3 hyperexcitability. The 2.7 and 6.9-fold NE EC50 dose ratio calculated in the presence of RS79948 and RX821002, relate to estimated Kb values of 0.9 ± 0.3 and 2.0 ± 0.9 nM, respectively for these selective ␣2 AR antagonists. These estimated values for RS79948 and RX821002 are comparable to affinities calculated by others when using these AR antagonists to identify ␣2 ARs in their investigations (Welbourn et al., 1986; Clark et al., 1989; Uhl´en et al., 1998).
4. Discussion In the present study, we show pharmacological evidence for an ␣2 AR-mediated inhibition of hippocampal CA3 network activity. Increasing concentrations of the endogenous AR agonist, NE, produced a biphasic change in the rate of CA3 hippocampal epileptiform burst discharges, suggesting that more than one type of AR was mediating these effects. This hypothesis was confirmed when, in the presence of AR blockade, NE produced only a decrease in the frequency of epileptiform activity, while in the presence of ␣AR blockade, NE caused only an increase in the number of burst discharges. Furthermore, these effects of
124
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
Fig. 3. Determination of NE potency to cause reduced hippocampal CA3 burst activity. Illustrated is the frequency histogram (A) used to derive the concentration–response curve (B). (A) Plot of the number of burst discharges vs. time for an experiment in which NE was applied at increasing concentrations. Each bin represents the frequency averaged over a 150 s epoch. In the presence of 3 M pindolol, NE was bath-applied for the periods (∼8 min) indicated by the bars. (B) Concentration–response curve derived from the frequency histogram (i.e., plot of burst discharge frequency vs. NE concentration). The data points are plotted as percent of maximal inhibitory response and the curve was constructed using non-linear regression analysis. For this particular experiment, the concentration–response curve for NE was fit best by a standard slope model with a calculated EC50 value of 1.2 M.
NE were not dependent on a functioning GABAergic system. Finally, there was no significant change in EC50 values for NE calculated from cumulative concentration–response curves generated in the presence of selective ␣1 AR antagonists when compared to control. Conversely, there was a significant difference in the potency of NE between control experiments and those performed in the presence of selective ␣2 AR antagonists. In addition, estimations of the equilibrium dissociation constant for these selective ␣2 AR antagonists were similar to values calculated by others when used to identify rat ␣2 AR populations. Therefore, our results support the concept that activation of a population of ␣2 ARs directly inhibits the network activity of the hippocampal CA3 region. A potential mechanism of NE-mediated reduction in epileptiform activity could be enhanced GABA release from hippocampal interneurons (Doze et al., 1991; Bergles et al., 1996). Although GABAA receptors were blocked in the epilepsy model used in this study, increased GABA release would still be expected to enhance GABAB receptor-mediated inhibition, which in turn, might reduce hippocampal excitability. However, the fact that the antiepileptic effects of NE persisted in the presence of GABAB receptor blockade, would seem to rule out this possibility. Furthermore, if increased GABA-mediated inhibition had been the underlying mechanism in this model, we would have
expected ␣1 ARs to have been involved, as we have previously shown (Bergles et al., 1996). However, there was no difference in the NE-mediated inhibition of CA3 activity in the presence of ␣1 AR antagonists when compared to control. In fact, we wonder if the reason an ␣1 AR-mediated inhibition of the hippocampal CA3 burst discharges was not observed was because the influence of inhibitory GABAergic interneurons had been pharmacologically removed. To examine this possibility, investigations using epilepsy models in which the GABAergic inhibition has not been impaired are currently being performed. The effects of NE on epileptiform activity not only generated monophasic responses in the presence of AR blockade, but the responses were also considerably larger. For example, 10 M NE alone caused a modest 35% reduction in the frequency of burst discharges very similar in magnitude to that observed previously by other investigators (Rutecki, 1995; Stoop et al., 2000). However, after pretreatment with 3 M of the selective AR antagonist, pindolol, 10 M NE essentially inhibited all epileptiform burst activity. This increased effect of NE was likely due to pharmacological isolation of the ␣AR response (i.e., blocking the AR-mediated excitatory action of NE), thereby leaving the ␣AR-mediated inhibitory effect unimpeded. Furthermore, this increased NE response allowed for the generation of reliable concentration–response
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
125
Fig. 4. NE concentration–response curves in the presence of prazosin and RS79948. Prazosin has an equilibrium dissociation constant for ␣1 ARs of ∼0.3 nM and an ␣1 /␣2 AR selectivity between 20- and 9000-fold depending on the ␣2 AR subtype (Doxey et al., 1984; Ireland et al., 1997). In contrast, RS79948 has an equilibrium dissociation constant for ␣2 ARs of ∼0.7 nM and an ␣2 /␣1 AR selectivity of about 15,000-fold (Clark et al., 1989; Uhl´en et al., 1998). Increasing amounts of NE were used to generate concentration–response curves associated with the inhibition of epileptiform discharge frequency for rat hippocampal slices in the absence of antagonist (), with 1 nM prazosin (), or in the presence of both 1 nM RS79948 and 1 nM prazosin (). The calculated EC50 values from these curves showed no difference for the potency of NE in the presence of prazosin (1.0 ± 0.3 M) when compared to control (1.2 ± 0.4 M). Conversely, there was a significant difference in the EC50 calculated for NE (2.7 ± 1.3 M) in the presence RS79948 and prazosin when compared to control. All experiments were performed in the presence of 20 M bicuculline to generate epileptiform burst activity and 3 M pindolol to block ARs. Measurements are presented as the mean ± S.E. for n = 3 experiments.
Fig. 5. NE concentration–response curves in the presence of terazosin and RX821002. Terazosin has an equilibrium dissociation constant for ␣1 ARs of ∼3 nM and an ␣1 /␣2 AR selectivity between 10and 500-fold depending on the ␣2 AR subtype (Hancock et al., 1995). In contrast, RX821002 has an equilibrium dissociation constant of ∼0.8 nM and an ␣2 /␣1 AR selectivity of about 80-fold (Welbourn et al., 1986; Uhl´en et al., 1998). Increasing amounts of NE were used to generate concentration–response curves associated with the inhibition of epileptiform discharge frequency for rat hippocampal slices in the absence of antagonist (), with 10 nM terazosin (), or in the presence of both 10 nM RX821002 and 10 nM terazosin (). The EC50 calculated from these curves showed no difference for the potency of NE in the presence of terazosin (1.5 ± 0.5 M) when compared to control (1.3 ± 0.3 M). Conversely, there was a significant difference in the EC50 calculated for NE (9.0 ± 2.9 M) in the presence of RX821002 and terazosin when compared to control. All experiments were performed in the presence of 20 M bicuculline to generate epileptiform burst activity and 3 M pindolol to block ARs. Measurements are presented as the mean ± S.E. for n = 3–4 experiments.
curves, which consequently permitted a more detailed and accurate pharmacological analysis. While several studies have evaluated the pharmacological effect of NE in models of epilepsy, the specific type of AR involved has not yet been unequivocally established. According to Giorgi et al. (2004), conflicting results have been reported not only because there are significant differences in AR distribution between species, but also because previous investigations have used non-discriminating high concentrations of selective AR ligands. In this study, we were able to avoid these problems by using concentrations of AR antagonists selective for only one ␣AR type based on the equilibrium dissociation constant calculated for native and recombinant receptors. For example, prazosin and terazosin, when used at con-
centrations specific for blocking ␣1 ARs, did not shift the NE concentration–response curve. However, addition of RS79948 or RX821002, at concentrations that would selectively inhibit ␣2 ARs, significantly shifted the NE concentration–response curve to the right when compared with control. These results suggested that the antiepileptic effects of NE in the hippocampal CA3 region are mediated through ␣2 AR activation. NE potency changes in the presence of selective ␣2 AR antagonists compared to control were used to calculate a one-point Kb value using a modified method of Schild (Arunlakshana and Schild, 1959). The estimated Kb values of 0.9 ± 0.3 and 2.0 ± 0.9 nM for RS79948 and RX821002, respectively, are similar to equilibrium dissociation constants calculated by others for these selective ␣2 AR antagonists (Welbourn et al., 1986; Clark
126
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
et al., 1989; Uhl´en et al., 1998). Our results using discriminating concentrations of selective ␣AR antagonists suggest that in this model of epileptiform activity, the NE-mediated decrease in hippocampal CA3 hyperexcitability is due primarily to activation of ␣2 ARs. In the rat hippocampus, antagonism of GABAA receptors creates a bursting phenomenon in the hippocampal CA3 region commonly referred to as epileptiform activity. This activity is thought to originate from the extensive recurrent collateral connections between CA3 pyramidal cells (Li et al., 1994; Johnston and Amaral, 2004). Under conditions of GABAergic inhibition, these profuse glutamatergic connections allow for rapid synchronization of pyramidal cell firing. It has been hypothesized that a similar process occurs in temporal lobe epilepsy, where the hippocampal CA3 region is thought to be a common focus (Yaari and Beck, 2002). As such, this epileptiform model of neuronal synchronization is often used as a model for temporal lobe epilepsy, as well as for other epilepsies. In general, pharmacological agents which are seen to decrease the frequency of epileptiform bursting are believed to be anticonvulsant, while drugs which increase epileptiform bursting are thought to be proconvulsant. In fact, there is a good correlation between the effectiveness of antiepileptic agents in vivo and the effectiveness of antiepileptic agents in hippocampal CA3 epileptiform models (Schneiderman and Schwartzkroin, 1982; Ault et al., 1986; Smith and Swann, 1987). Based on this, we speculate that ␣2 AR-mediated inhibition of hippocampal CA3 burst activity could account for some of the anticonvulsant properties of NE observed in vivo. NE presents an exciting prospect in the treatment of epilepsy. Evidence suggests that, unlike most current antiepileptic drugs, NE does not inhibit such cognitive functions as learning and memory (Devauges and Sara, 1991; Thomas and Palmiter, 1997). Indeed, NE may even facilitate certain types of memory processes (O’Carroll et al., 1999; Murchison et al., 2004). Furthermore, indirect evidence suggests that NE could be effective against seizures that are refractory to current antiepileptic drugs. As mentioned previously, both the ketogenic diet and vagal nerve stimulation treatment modalities currently used in patients with medically intractable epilepsies cease to provide their anticonvulsant effects in the absence of an intact NE system (Krahl et al., 1998; Weinshenker and Szot, 2002).
In this study, we found that NE both increased and decreased hippocampal CA3 epileptiform activity depending upon the concentration. The enhancement of this activity occurs at lower doses, while the mitigation of epileptiform activity predominates at higher doses. We speculate that these differential effects at different concentrations of NE could account for how NE both enhances certain memory processes and yet is antiepileptic in vivo. In summary, this investigation attempts to further characterize the excitatory and inhibitory effects of NE on hippocampal CA3 network activity. While this study provides only an initial clarification of the AR types and site of action for these NE effects, future studies can now be narrowed in scope to determine the specific mechanisms for ␣2 AR-mediated antiepileptic effects in the hippocampal CA3 region. This knowledge will not only expand our understanding of the role of NE in epilepsy, but may also reveal novel targets for future antiepileptic drug development research.
Acknowledgements We are grateful to Karen L. Cisek and Katie E. Rau for their assistance with the experiments and the manuscript. This investigation was supported in part by North Dakota EPSCoR through NSF grant EPS0132289 (V.A.D.), NIH grant 5P20RR017699 from the COBRE program (V.A.D. and J.E.P.), and the University of North Dakota Faculty Research Seed Money Council (V.A.D. and S.J.P.).
References Arnold, P.S., Racine, R.G., Wise, R.A., 1973. Effects of atropine, reserpine, 6-hydroxydopamine, and handling on seizure development in the rat. Exp. Neurol. 40, 457–470. Arunlakshana, O., Schild, H.O., 1959. Some quantitative uses of drug antagonist. Br. J. Pharmacol. 14, 48–58. Aston-Jones, G., Rajkowski, J., Cohen, J., 1999. Role of locus coeruleus in attention and behavioral flexibility. Biol. Psychiatry 46, 1309–1320. Ault, B., Gruenthal, M., Armstrong, D.R., Nadler, J.V., Wang, C.M., 1986. Baclofen suppresses bursting activity induced in hippocampal slices by differing convulsant treatments. Eur. J. Pharmacol. 126, 289–292. Bergles, D.E., Doze, V.A., Madison, D.V., Smith, S.J., 1996. Excitatory actions of norepinephrine on multiple classes of hippocampal CA1 interneurons. J. Neurosci. 16, 572–585.
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128 Bylund, D.B., Eikenberg, D.C., Hieble, J.P., Langer, S.Z., Lefkowitz, R.J., Minneman, K.P., Molinoff, P.B., Ruffolo, R.R., Trendelenbyurg, A.U., 1994. IV. International Union of Pharmacology, nomenclature of adrenoceptors. Pharmacol. Rev. 46, 121–136. Chauvel, P., Trottier, S., 1986. Role of noradrenergic ascending system in extinction of epileptic phenomena. In: Delgado-Escueta, Ward, Woodbury, Porter (Eds.), Advances in Neurology, vol. 44, Basic Mechanisms of the Epilepsies. Raven Press, New York, pp. 475–487. Clark, R.D., Repke, D.B., Kilpatrick, A.T., Brown, C.M., MacKinnon, A.C., Clague, R.U., Spedding, M., 1989. (8a␣, 12a␣, 13a␣)-5,8,8a,9,10,11,12,12a,13,13a-decahydro-3-methoxy12-(methylsulfonyl)-6H-isoquino[2,1-g][1,6]naphthyridine, a potent and highly selective ␣2 -adrenoceptor antagonist. J. Med. Chem. 32, 2034–2036. Davies, C.H., Pozza, M.F., Collingridge, G.L., 1993. CGP55845A: a potent antagonist of GABAB receptors in the CA1 region of rat hippocampus. Neuropharmacology 32, 1071–1073. Devauges, V., Sara, S.J., 1991. Memory retrieval enhancement by locus coeruleus stimulation: evidence for mediation by betareceptors. Behav. Brain Res. 43, 93–97. Doxey, J.C., Lane, A.C., Roach, A.G., Virdee, N.K., 1984. Comparison of the alpha-adrenoceptor antagonist profiles of idazoxan (RX781094), yohimbine, rauwolscince and corynanthine. Naunyn Schmiedebergs Arch. Pharmacol. 325, 136–144. Doze, V.A., Cohen, G.A., Madison, D.V., 1991. Synaptic localization of adrenergic disinhibition in the rat hippocampus. Neuron 6, 889–900. Eichenbaum, H., 2000. A cortical-hippocampal system for declarative memory. Nat. Rev. Neurosci. 1, 41–50. Ferraro, G., Sardo, P., Sabatino, M., La Grutta, V., 1994. Locus coeruleus noradrenaline system and focal penicillin hippocampal epilepsy: neurophysiology study. Epilepsy Res. 19, 215–220. Giorgi, F.S., Pizzanelli, C., Biagioni, F., Murri, L., Fornai, F., 2004. The role of norepinephrine in epilepsy: from the bench to the bedside. Neurosci. Biobehav. Rev. 28, 507–524. Hancock, A.A., Buckner, S.A., Ireland, L.M., Knepper, S.M., Kerwin Jr., J.F., 1995. Actions of terazosin and its enantiomers at subtypes of ␣1 - and ␣2 -adrenoceptors in vitro. J. Recept. Signal. Transduct. Res. 15, 863–885. Ireland, L.M., Cain, J.C., Rotert, G., Thomas, S., Shoghi, S., Hancock, A.A., Kerwin Jr., J.F., 1997. [3 H]R-terazosin binds selectivity to ␣1 -adrenoceptors over ␣2 -adrenoceptors – comparison with racemic [3 H]terazosin and [3 H]prazosin. Eur. J. Pharmacol. 327, 79–86. Johnston, B., Amaral, D.G., 2004. In: Shepherd (Ed.), Hippocampus, in the Synaptic Organization of the Brain. Oxford University Press, New York, pp. 455–498. Krahl, S.E., Clark, K.B., Smith, D.C., Browning, R.A., 1998. Locus coeruleus lesions suppress the seizure-attenuating effects of vagus nerve stimulation. Epilepsia 39, 709–714. Leung, P.P., Miller, J.J., 1988. Dual role of norepinephrine in the hippocampal CA1 region of the rat: inhibition and disinhibition. Can. J. Physiol. Pharmacol. 66, 814–819. Li, X.G., Somogyi, P., Ylinen, A., Buzsaki, G., 1994. The hippocampal CA3 network: an in vivo intracellular labeling study. J. Comp. Neurol. 339, 181–208.
127
Loy, R., Koziell, D.A., Lindsey, J.D., Moore, R.Y., 1980. Noradrenergic innervation of the adult rat hippocampal formation. J. Comp. Neurol. 189, 699–710. Milner, B., Squire, L.P., Kandel, E.R., 1998. Cognitive neuroscience and the study of memory. Neuron 20, 445–468. Mueller, A.L., Dunwiddie, T.V., 1983. Anticonvulsant and proconvulsant actions of ␣- and -noradrenergic agonists on epileptiform activity in rat hippocampus in vitro. Epilepsia 24, 57– 64. Mueller, A.L., Hoffer, B.J., Dunwiddie, T.V., 1981. Noradrenergic responses in rat hippocampus: evidence for mediation by ␣- and -receptors in the in vitro slice. Brain Res. 214, 113–126. Murchison, C.F., Zhang, X.Y., Zhang, W.P., Ouyang, M., Lee, A., Thomas, S.A., 2004. A distinct role for norepinephrine in memory retrieval. Cell 117, 131–143. O’Carroll, R.E., Drysdale, E., Cahill, L., Shajahan, P., Ebmeier, K.P., 1999. Stimulation of the noradrenergic system enhances and blockade reduces memory for emotional material in man. Psychol. Med. 29, 1083–1088. Pupo, A.S., Minneman, K.P., 2001. Adrenergic pharmacology: focus on the central nervous system. CNS Spectr. 6, 656–662. Rutecki, P.A., 1995. Noradrenergic modulation of epileptiform activity in the hippocampus. Epilepsy Res. 20, 125–136. Schneiderman, J.H., Schwartzkroin, P.A., 1982. Effects of phenytoin on normal activity and on penicillin-induced bursting in the guinea pig hippocampal slice. Neurology 32, 730–738. Schroeter, S., Apparsundaram, S., Wiley, R.G., Miner, L.H., Sesack, S.R., Blakely, R.D., 2000. Immunolocalization of the cocaineand antidepressant-sensitive l-norepinephrine transporter. J. Comp. Neurol. 420, 211–232. Schwartzkroin, P.A., 1986. Hippocampal slices in experimental and human epilepsy. In: Delgado-Escueta, Ward, Woodbury, Porter (Eds.), Advances in Neurology, vol. 44, Basic Mechanisms of the Epilepsies. Raven Press, New York, pp. 991–1010. Smith, K.L., Swann, J.W., 1987. Carbamazepine suppresses synchronized after discharging in disinhibited immature rat hippocampus in vitro. Brain Res. 400, 371–376. Stanton, P.K., Mody, I., Zigmond, D., Sejnowski, T., Heinemann, U., 1992. In: Avanzini, G., Ngel, R., Fariello, R., Heinemann, U. (Eds.), Neurotransmitters in Epilepsy (Epilepsy Res. Suppl. 8), Chapter 40, Noradrenergic Modulation of Excitability in Acute and Chronic Model Epilepsies. Elseiver Science Publishers, New York, pp. 321–334. Stoop, R., Epiney, S., Meier, E., Pralong, E., 2000. Modulation of epileptiform discharges in the rat limbic system in vitro by noradrenergic agents. Neurosci. Lett. 287, 5–8. Szot, P., Weinshenker, D., Rho, J.M., Storey, T.W., Schwartzkroin, P.A., 2001. Norepinephrine is required for the anticonvulsant effect of the ketogenic diet. Brain Res. Dev. Brain Res. 129, 211–214. Thomas, S.A., Palmiter, R.D., 1997. Disruption of the dopamine beta-hydroxylase gene in mice suggests roles for norepinephrine in motor function, learning, and memory. Behav. Neurosci. 11, 579–589. Uhl´en, S., Dambrova, M., Nasman, J., Schioth, H.B., Gu, Y., Wikberg-Matsson, A., Wikberg, J.E.S., 1998. [3 H]RS79948197 binding to human, rat, guinea pig and pig ␣2A -, ␣2B - and
128
C.W.D. Jurgens et al. / Epilepsy Research 66 (2005) 117–128
␣2C -adrenoceptors. Comparison with MK912, RX821002, rauwolscine and yohimbine. Eur. J. Pharmacol. 343, 93–101. Weinshenker, D., Szot, P., 2002. The role of catecholamines in seizure susceptibility: new results using genetically engineered mice. Pharmacol. Ther. 94, 213–233. Weiss, G.K., Lewis, J., Jimenez-Rivera, C., Vigil, A., Corcoran, M.E., 1990. Antikindling effects of locus coeruleus stimulation: mediation by ascending noradrenergic projections. Exp. Neurol. 108, 136–140.
Welbourn, A.P., Chapleo, C.B., Lane, A.C., Myers, P.L., Roach, A.G., Smith, C.F.C., Stillings, M.R., Tulloch, I.F., 1986. ␣Adrenoceptor reagents. 4. Resolution of some potent selective prejunctional ␣2 -adrenoceptor antagonists. J. Med. Chem. 29, 2000–2003. Yaari, Y., Beck, H., 2002. “Epileptic neurons” in temperol lobe epilepsy. Brain Pathol. 12, 234–239.