| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
BACKGROUND: Synaptic plasticity is thought to provide a molecular mechanism for learning and memory. N-methyl-d-aspartate receptor-mediated plasticity requires that N-methyl-d-aspartate receptor activation coincides with postsynaptic depolarizing potentials (DPSPA's). Pentobarbital, in high concentrations, enhances DPSPA's, but high concentrations suppress synaptic plasticity, probably by impairing glutamatergic transmission. Here we tested the hypothesis that low concentrations of pentobarbital can enhance DPSPA's and modify the induction of synaptic plasticity. METHODS: Studies were performed in vitro on rat hippocampal slices. With glutamate transmission blocked, intracellular recording from CA1 neurons was used to investigate the influence of 5 µM pentobarbital on DPSPA's and neuron excitability evoked by high frequency (100 Hz) stimulation. With glutamate transmission intact, extracellular recording was used to examine the effect of 5 µM pentobarbital on the induction of long-term depression and long-term potentiation of synaptic transmission by conditioning stimuli applied to the Schaffer collateral pathway.
RESULTS: High frequency stimulation generated typical DPSPA's that were mediated by CONCLUSIONS: Despite increasing the amplitude of DPSPA's, 5 µM pentobarbital did not alter the induction of synaptic plasticity by a range of conventional conditioning stimuli. These results do not support the hypothesis that excitatory effects of pentobarbital may alter synaptic plasticity.
Activity-dependent changes in the efficacy of synaptic transmission, referred to as "synaptic plasticity," are believed to be fundamental to the processes of learning and memory (1). Plasticity, in the form of long-term depression (LTD) and long-term potentiation (LTP) of synaptic transmission in the hippocampus, is a putative cellular mechanism for spatial memory and learning in the rat (1). Afferent activity of sufficient intensity, duration, or frequency impinging on CA1 neurons can induce either LTD or LTP, depending on the level of postsynaptic N-methyl-d-aspartate (NMDA) receptor activation, Ca2+ influx, and/or the activity balance between phosphatases and kinases (1). Because anesthetics target many of the molecular mechanisms involved in synaptic plasticity (2), it is not surprising that high concentrations of anesthetics that induce unresponsiveness to stimulation during clinical anesthesia also block synaptic plasticity (36). However, results from several studies (79) raise the possibility that low concentrations of anesthetics may paradoxically facilitate synaptic plasticity by enhancing a novel excitatory action of - aminobutryic acid (GABA). Although GABA is conventionally viewed as an inhibitory neurotransmitter, the relevance of an excitatory depolarizing action of GABA described in early studies has been realized (9). In mature neurons, this novel action of GABA may play a physiological role in the thalamus, the dorsal horn of the spinal cord, the neocortex, and the hippocampus (9). It has been proposed (8,10) that drugs that increase GABA-mediated excitation [such as barbiturates (11)] may enhance the induction of NMDA receptor-dependent synaptic plasticity in the hippocampus. We (12) have reported that low concentrations of barbiturates are associated with excitatory effects in the hippocampus [activation of hippocampal electroencephalographic activity and increased paired-pulse facilitation (13)], and in spinal reflex pathways [decreased nociceptive reflex threshold (14)]. We (13,15) have presented evidence that some of the excitatory effects of pentobarbital are dependent upon HCO3 ion, a finding consistent with a mechanism involving GABA-mediated excitation (7,8,10). The goal of the present study was to seek evidence that 5 µM pentobarbital can alter synaptic plasticity. We investigated the effects of pentobarbital on GABA-mediated excitation, synaptic transmission, LTP, and LTD in the CA1 region of rat hippocampal slices. Using intracellular recording, we first sought to determine whether pentobarbital enhances GABA-mediated excitation, as measured by the size of GABA-mediated depolarizing potentials (7,8). Then, using extracellular recording to measure synaptic transmission in the Schaffer Collateral-CA1 pathway, we sought to determine if pentobarbital could alter synaptic plasticity as measured by the influence of afferent conditioning stimuli (CS) on persistent changes in synaptic response (16,17).
Experimental procedures were approved by the Animal Care Committee, Faculty of Medicine, University of Calgary. Experiments were performed on transverse slices of hippocampus prepared from 20- to 30-day-old male Sprague Dawley rats (Charles River, Montreal, Canada). Each animal was anesthetized with ether and decapitated soon after apnea. The brain was rapidly removed and immersed in cold (8°C10°C) physiologic solution (artificial cerebrospinal fluid, aCSF) containing (in mM) NaCl, 124; KCl, 5; NaHCO3, 26; CaCl2, 2; MgSO4, 2; glucose, 10; and NaH2PO4, 1.25. Slices 400 µm thick were cut using a tissue chopper (Stoelting, IL) and placed on a nylon mesh screen at the gasliquid interface surface of a recording chamber. Slices were continuously perfused with aCSF (1 mL/min) and a prewarmed, humidified gas mixture (95% oxygen-5% carbon dioxide, 1.5 L/min). Slices were incubated without stimulation for 1 h at room temperature and thereafter maintained at 32 °C ± 1°C for intracellular experiments and 35 °C ± 1 °C for extracellular experiments.
For the intracellular experiments, neurons in the stratum pyramidale of the CA1 region were impaled blindly with sharp electrodes, (resistance 50 to 100 M Based upon the studies of Kaila et al. (7) intracellular recordings of GABA-mediated potentials were made with glutamate neurotransmission blocked by the presence of ionotropic glutamate receptor blockers, 10 µM 6-nitro-7-sulfamoylbenzo(f)-quinoxaline-2,3-dione (NBQX), 40 µMD-2-amino-5-phosphonopentoate (AP5), and 50 µM ketamine. Stimuli were applied to the stratum radiatum close (<0.5 mm) to the impaled CA1 neuron. Trains of high frequency stimuli (HFS), (1, 2, 4, 8, 20, and 40 pulses at 100 Hz, 1015 V) were generated by a Grass S88 stimulator and SIU5 isolation unit (Grass Corp., Quincy, MA) and applied with a bipolar tungsten electrode. The stimuli release glutamate and GABA in the vicinity of the impaled neuron. Because glutamate transmission was blocked, the resulting rapid responses in the impaled neuron were mediated primarily by GABA receptors. For convenience, we have referred to the GABAA receptor-mediated depolarizing postsynaptic potentials as DPSPA's. The DPSPA size was characterized with the amplitude and width at half-maximal amplitude (18) (Fig. 1). Because the DPSPA evoked by a train of eight pulses was intermediate in both amplitude and width at half-maximal amplitude, eight pulse trains were used to evaluate the effect of pentobarbital on the DPSPA.
For extracellular experiments, field excitatory postsynaptic potential and population spike (PS) responses were recorded using glass microelectrodes (14 M
In the first experiment we examined the influence of pentobarbital on synaptic plasticity induced by low frequency stimulation. Briefly, according to the synaptic modification threshold model, described by Bear (19), stimulation of the Schaffer Collateral-CA1 pathway with low frequency (13 Hz) CS induces LTD whereas higher frequencies (510 Hz) induce LTP. The critical level of postsynaptic response at which LTD changes to LTP has been called "the synaptic modification threshold" ( In the first experiment, low frequency stimulation [1, 3, 5, or 10 Hz trains of 900 pulses (17)] was applied to the Schaffer Collateral pathway after exposure of the slice to either aCSF (control) or pentobarbital. Field responses were averaged over 10 min epochs before and 40 min after application of drug (aCSF or pentobarbital). One CS was applied to each slice 20 min after the start of the drug infusion (aCSF or pentobarbital [5 µM]). Pentobarbital was infused for a total of 30 min. Thereafter, the slice was perfused with aCSF. For each slice, field responses were normalized to the average response for a 10-min epoch before the start of the drug infusion. The effect of the CS was evaluated using responses averaged over a 10-min epoch 40 min after the end of the drug infusion. In the second experiment, we examined the effect of 5 µM pentobarbital on high frequency-induced LTP. In this protocol slices were exposed to either aCSF or pentobarbital throughout each trial. A 40 Hz CS lasting 500 ms was used to induce LTP for two reasons: 1) a frequency above 1020 Hz is putatively required to allow maximal afferent GABA release (16); 2) a short burst of 40 Hz yields a weak induction and a nonmaximal magnitude of LTP (20), permitting observation of any enhancing effect by low dose pentobarbital. Responses were evoked by paired pulse stimuli (13), with the interval (3070 ms separation) set to produce enhancement of the second response. The PS amplitudes were averaged over 10 min epochs before and 40 min after application of the CS. The effects of HFS and pentobarbital on both responses (PS1 and PS2) were analyzed by two-factor repeated measures ANOVA, using PS1 and PS2 as the first factor and aCSF or pentobarbital as the second factor. Wave forms were recorded and analyzed with customized Labview® based software. Data from all similarly treated slices were averaged and presented as mean change ± se from the average baseline response (average of the first 10 min of responses) for each group. All drugs were added to the physiologic solution perfusing the slice. Pentobarbital (Sigma Aldrich, Canada) was used in 5 and 100 µM concentrations based upon previous studies from our laboratory (13). In selected experiments 10 µM bicuculline and 10 µM saclofen (both from Sigma Aldrich) were used to block GABAA and GABAB receptors respectively. To examine the role of HCO3 ion we used two strategies (7): inhibition of carbonic anhydrase with 50 µM ethoxyzolamide (2% dimethyl sulfoxide) and substitution of HCO3 in the perfusion solution with 10 mM HEPES buffer (all three drugs from Sigma Aldrich) (pH adjusted to 7.4 with NaOH). In control experiments, the dimethyl sulfoxide carrier for ethoxyzolamide did not alter passive membrane properties or the biphasic response to HFS. Each treatment was applied to at least three slices. All data are presented as mean values ± se of the mean. Statistical significance was inferred when P < 0.05.
With glutamate transmission blocked, HFS evoked a triphasic response consisting of IPSPA, IPSPB, and a DPSPA (Fig. 1). The IPSPA was altered and the DPSPA was blocked with 10 µM bicuculline (P < 0.01, n = 3, Fig. 1A), indicating GABAA receptor dependency. In the presence of bicuculline, addition of a high-concentration of pentobarbital (100 µM) did not alter the response to HFS (n = 3, Fig. 1A). The IPSPB was blocked with 10 µM saclofen (P < 0.01, n = 3, Fig. 1B), indicating GABAB receptor dependency. The DPSPA was HCO3 ion-dependent as shown by inhibition of the response with the carbonic anhydrase inhibitor ethoxyzolamide or by substitution of HCO3 in the physiologic solution perfusing the slice with HEPES buffer (Figs. 1C and E). The size of the DPSPA increased with the stimulus strength (the number of pulses in the stimulus train) (Fig. 1D); this effect was dependent upon HCO3 ion (Fig. 1E). Pentobarbital increased the size of the DPSPA, a high concentration (100 µM) increased both the amplitude and width at half-maximal amplitude of the depolarizing response while the low concentration (5 µM) increased only the amplitude (Figs. 2AC). The pentobarbital-induced increase in DPSPA size was excitatory, as shown by bursting activity (Fig. 2A), by an increase in the probability of action potential generation (spiking probability, Fig. 2D) and by an increase in the number of action potentials generated by an eight-pulse train (Fig. 2E).
We observed that, during a 60 min exposure to 5 µM pentobarbital, 92% of the final DPSPA amplitude increase was achieved after 27 min (Fig. 3), confirming the suitability of the 30 min exposure times used in the protocols.
With glutamate transmission intact, extracellular electrodes recorded the CA1 postsynaptic response to stimulation of the Schaffer Collateral pathway. Two features of the synaptic response were measured: field excitatory postsynaptic potential slope from recordings in the dendritic layer; PS amplitude from recordings in the cell body layer (Fig. 4A). Responses were depressed by CS trains of 1 and 3 Hz but potentiated by 10 Hz. The crossover point from LTD to LTP (synaptic modification threshold,
In another experiment, we tested the ability of 5 µM pentobarbital to enhance LTP induced by a 40 Hz stimulus train. The 40 Hz CS lies within a frequency range known to induce submaximal LTP (20) but also shown to maximally evoke GABA activity (16). We observed that the 40 Hz CS was followed by submaximal LTP of PS responses, a finding that was not altered by pentobarbital (Fig. 5).
The findings of the present study revealed that a low concentration of pentobarbital (5 µM), despite exerting a measurable increase in GABA-mediated excitation at the cellular level, did not influence synaptic plasticity in CA1 neurons. These findings do not support the hypotheses (8,10) that anesthetics can enhance synaptic plasticity by increasing GABA excitation. Clinically relevant concentrations of anesthetics suppress the induction of LTP (3,5,6,21). Most (5,21), but not all (6) studies report that anesthetics have a smaller effect on LTD. At clinical concentrations, volatile anesthetics depress glutamatergic transmission, possibly through GABAA receptor-mediated inhibition, reducing the activation of NMDA receptors that are involved in some forms of LTD and LTP (3,4). The anesthetic concentrations reported to suppress LTP induction have been approximately 10 times those associated with the excitatory effects of anesthetics (14,22) which were the subject of the present study. Low concentrations are encountered during induction of and emergence from anesthesia. In contrast to the suppressant effects of clinical concentrations of anesthetics, the present study sought evidence that a low concentration of pentobarbital, shown previously (1315) to be associated with excitatory effects, could alter synaptic plasticity. This hypothesis was suggested to us by reports (7,8) that GABAA receptor activation can cause neuronal excitation (9), findings that have recently lead to a reappraisal of the role of GABA in synaptic transmission. Briefly, in addition to its well-known hyperpolarizing action and inhibitory effects on synaptic transmission, GABA has well-defined excitatory effects (9) that have been demonstrated in neurons in the neocortex, hippocampus, and dorsal root ganglia. Gulledge and Stuart (23) have shown that GABAA receptor activation can have at least three effects. In the dendrites, Cl-mediated hyperpolarizing IPSP's may inhibit glutamate-mediated excitatory potentials. Strong stimulation in the dendrites evokes excitatory DPSPAs, generated when the duration of GABAA ionophore opening results in a shift of the predominant anion moving through the channel from Cl to HCO3 (8). For induction of NMDA receptor-dependent LTP, NMDA receptors must be activated during a period of postsynaptic depolarization, which can be mediated by GABA, that relieves the NMDA channel block by Mg+2 (8). At the cell soma, appropriately timed GABAA receptor-mediated increases in conductance may inhibit signal transmission from the dendrites (9). The overall effect of GABAA receptor activation can no longer be summarized as being "inhibitory" but results from the interactions between dendritic inhibition, dendritic excitation, and inhibitory increases in somatic membrane conductance (9). The observation that drugs such as barbiturates enhance DPSPAs (11,24) has lead to the hypothesis (8,10,13) that GABA effects may contribute to excitation associated with concentrations of these drugs that are low enough to leave glutamatergic transmission intact. Drugs that increase GABA excitation can enhance learning (10), confirming a link between pharmacological effects and behavior. The results of the present study suggest that although pentobarbital can increase the size of excitatory depolarizing potentials, this effect is not sufficient to modify the induction of synaptic transmission. Constraints of the present study include the selection of one concentration of one anesthetic and the limited number of conditioning stimulation paradigms that were applied. Pentobarbital was selected as a prototypical barbiturate for which the effects on DPSPAs have been described previously. The concentration corresponds to that seen during behavioral excitation with pentobarbital (24,25). We believe that 5 µM pentobarbital is unlikely to impair glutamatergic transmission; during paired-pulse stimulation the second response is impaired with an IC50 of 2530 µM (13,21); the effects on the first response are less. There have been many other stimulation strategies reported (1). It is quite possible that a study of other stimulation paradigms or the effects different anesthetics may result in effects different from those reported here. In summary, we report that a low concentration of pentobarbital (5 µM) increases the size of excitatory GABA-mediated depolarizing potentials. The increase in DPSPA size was sufficient to make neurons more excitable. Despite the latter findings, pentobarbital did not alter the induction of LTP or LTD in the Schaffer Collateral-CA1 pathway.
Accepted for publication December 12, 2006. Supported by Canadian Institutes of Health Research. Presented in part at the annual meeting of the International Anesthesia Research Society, Honolulu, Hawaii, March 2005. Reprints will not be available from the author.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|