Abstract
Recent studies have demonstrated that activation of the β-adrenergic receptor (AR) using the selective β-AR agonist isoproterenol (ISO) facilitates pyramidal cell long-term potentiation in the cornu ammonis 1 (CA1) region of the rat hippocampus. We have previously analyzed β-AR genomic expression patterns of 17 CA1 pyramidal cells using single cell reverse transcription-polymerase chain reaction, demonstrating that all samples expressed the β2-AR transcript, with four of the 17 cells additionally expressing mRNA for the β1-AR subtype. However, it has not been determined which β-AR subtypes are functionally expressed in CA1 for these same pyramidal neurons. Using cell-attached recordings, we tested the ability of ISO to increase pyramidal cell action potential (AP) frequency in the presence of subtype-selective β-AR antagonists. ICI-118,551 [(±)-1-[2,3-(dihydro-7-methyl-1H-inden-4-yl)oxy]-3-[(1-methylethyl)amino]-2-butanol] and butoxamine [α-[1-(t-butylamino)ethyl]-2,5-dimethoxybenzyl alcohol) hydrochloride], agents that selectively block the β2-AR, produced significant parallel rightward shifts in the concentration-response curves for ISO. From these curves, apparent equilibrium dissociation constant (Kb) values of 0.3 nM for ICI-118,551 and 355 nM for butoxamine were calculated using Schild regression analysis. Conversely, effective concentrations of the selective β1-AR antagonists CGP 20712A [(±)-2-hydroxy-5-[2-([2-hydroxy-3-(4-[1-methyl-4-(trifluoromethyl)-1H-imidazol-2-yl]phenoxy)propyl]amino)ethoxy]-benzamide methanesulfonate] and atenolol [4-[2′-hydroxy-3′-(isopropyl-amino)propoxy]phenylacetamide] did not significantly affect the pyramidal cell response to ISO. However, at higher concentrations, atenolol significantly decreased the potency for ISO-mediated AP frequencies. From these curves, an apparent atenolol Kb value of 3162 nM was calculated. This pharmacological profile for subtype-selective β-AR antagonists indicates that β2-AR activation is mediating the increased AP frequency. Knowledge of functional AR expression in CA1 pyramidal neurons will aid future long-term potentiation studies by allowing selective manipulation of specific β-AR subtypes.
The CA1 region of the hippocampus has been implicated in the processes of learning and memory for a number of years (Zola-Morgan et al., 1986). Long-term potentiation (LTP), currently the most accepted mechanistic model of learning, has been shown to occur in CA1 pyramidal cells, the principal neurons of this hippocampal region (Bliss and Collingridge, 1993). Recent evidence has demonstrated that β-adrenergic receptor (AR) activation using the selective β-AR agonist isoproterenol (ISO) facilitates LTP in these cells (Thomas et al., 1996; Moody et al., 1998; Lin et al., 2003). In vivo studies support this observation: β-AR activation has been shown to be advantageous in the learning process (Sternberg et al., 1985; Roullet and Sara, 1998), whereas blockade of this AR is detrimental to memory retrieval (Nielson and Jensen, 1994; Cahill et al., 2000). It is not known, however, which β-AR subtype mediates this modulatory effect because investigations have continually used nonsubtype-selective pharmacological agents.
The β-AR subtypes (β1-, β2-, and β3-) account for one-third of the adrenergic G protein-coupled receptor family. Although these ARs are traditionally known to signal through Gs, producing increased concentrations of cAMP and subsequent activation of protein kinase A, recent investigations have suggested that the β-ARs are also capable of interacting with Gi, β-arrestins, and G protein-coupled receptor kinases, thereby diversifying their signaling capabilities (Hall, 2004). Activation of β-ARs in the brain results in increased overall neuronal excitability (Mueller and Dunwiddie, 1983; Stoop et al., 2000), making this receptor class an interesting candidate for not only LTP modulation but also other instances where neuronal excitability is altered, such as epilepsy (Rutecki, 1995). Increased neuronal excitability after β-AR activation has been attributed to a number of mechanisms: facilitation of L-type Ca2+ channels (Hoogland and Saggau, 2004), modulation of K+ channels (Yuan et al., 2002), and phosphorylation of type 1 glutamate receptors (Vanhoose and Winder, 2003). It is likely a combination of these and perhaps other events not yet elucidated that enable adrenergic modulation of LTP in CA1.
β-AR expression in the hippocampus is well documented; however, the specific receptor subtypes present and their localization patterns remain unclear. Autoradiographic studies have reported both β1- and β2-AR expression in rat hippocampus (Booze et al., 1993), whereas molecular studies demonstrate mRNA for either the β2- (Nicholas et al., 1996) or the β3-AR (Summers et al., 1995). Attempts to discriminate β-AR subtypes using immunohistochemistry techniques have been compounded by the cross-reactivity of available β-AR antibodies (Milner et al., 2000). Furthermore, marked differences in neuronal AR distribution exist between species, making it difficult to draw comparisons between animal studies (Reznikoff et al., 1986; Booze et al., 1993).
Despite the number of localization studies examining β-AR distribution in the hippocampus, comprehensive functional studies examining the expression of β-AR subtypes are lacking. Past studies have predominately used nonselective β-AR agents or subtype-selective drugs used at nonselective concentrations. Delineating the specific β-AR subtype expressed on hippocampal pyramidal neurons would not only further our understanding of the AR influence in this region but also would enable greater specificity for future LTP studies examining β-AR modulation of learning and memory. The aim of this investigation was therefore to generate a functional β-AR expression profile for rat hippocampal CA1 pyramidal cells.
Materials and Methods
Materials. APV, atenolol [4-[2′-hydroxy-3′-(isopropyl-amino)propoxy]phenylacetamide], atropine, CGP 20712A, DNQX, and isoproterenol were obtained from Sigma-Aldrich (St. Louis, MO). ICI-118,551 and butoxamine [α-[1-(t-butylamino)ethyl]-2,5-dimethoxy-benzyl alcohol) hydrochloride] were obtained from Tocris Cookson Inc. (Ellisville, MO). All other chemical reagents were of biological grade and ordered through Fisher Scientific Co. (Fairlawn, NJ).
Slice Preparation. Male Sprague-Dawley rats weighing 30 to 60 g were deeply anesthetized with isoflurane and decapitated. Brains were rapidly removed and placed in 4°C Ringer solution containing 110 mM choline chloride, 25 mM NaHCO3, 25 mM d-glucose, 11.6 mM sodium ascorbate, 7 mM MgSO4, 3.1 mM sodium pyruvate, 2.5 mM KCl, 1.25 mM NaPO4, and 0.5 mM CaCl2. Using a tissue chopper, medial to dorsal segments of the hippocampus were sliced into 400-μm sections, which were then transferred to artificial cerebral spinal fluid (aCSF) consisting of 119 mM NaCl, 5 mM KCl, 1.3 mM MgSO4, 2.5 mM CaCl2, 1 mM NaH2PO4, 26.2 mM NaHCO3, and 11 mM d-glucose. The slices were incubated in the aCSF for 30 min at 37°C with and without subtype-selective β-AR antagonists or selective inhibitors of neurotransmission and then transferred to room temperature. All solutions were continually aerated with 95% O2, 5% CO2.
Cell-Attached Recording. Micropipettes were prepared from borosilicate glass using a vertical puller (Narishige, Tokyo, Japan). Pipettes were tip loaded with 50 U of RNase inhibitor III (Eppendorf, Westbury, NY) and backfilled with 135 mM KCH3SO4, 8 mM NaCl, 10 mM HEPES, 2 mM MgATP, 0.3 mM NaGTP, and 0.1 mM BAPTAK4. Using a BX51WI upright microscope (Olympus, Melville, NY), hippocampal slices were visualized under infrared-differential interference contrast optics, and a candidate CA1 pyramidal cell was centered in the field (Fig. 1A). Slices were constantly perfused with either aerated aCSF or aerated aCSF containing the pharmacological agent(s) of study. The micropipette was driven into the tissue while applying slight positive pressure to keep the tip clear of debris and placed flush with the membrane of the candidate pyramidal cell (Fig. 1A). A gigaohm seal was then formed using minimal current pulses, creating an isolated recording without disrupting the integrity of the cell membrane. Baseline action potential (AP) frequencies were recorded for 30 min before increasing log -fold concentrations (1–10,000 nM) of the selective β-AR agonist ISO were added to the perfusion line in 8-min increments. Increases in AP frequency were visualized in real time (Fig. 1B) while being recorded for subsequent off-line analysis. The 30-min baseline recording and ISO challenge were then repeated using adjacent slices that had been pretreated for 30 min with a set concentration of the subtype-selective β-AR antagonist of interest. APs were detected using an Axoclamp 2B (Axon Instruments Inc., Union City, CA), amplified by a Brownlee 440 (Brownlee Precision, San Jose, CA), digitized with a Digidata 1322A analog-to-digital converter (Axon Instruments Inc.), and recorded using Axoscope 9.0 software (Axon Instruments Inc.). Postexperimental analysis was completed using Mini Analysis 5.0 (Synaptosoft, Decatur, GA) and Prism 4.0 (GraphPad Software Inc., San Diego, CA).
Data Analysis. AP frequency was recorded during the course of each functional experiment in 2-min intervals (Fig. 1C). The last interval correlating to each ISO concentration was noted, baseline frequency was subtracted, and that value was used to plot a concentration-response curve expressed as percentage of maximal response (Fig. 1D). For each individual experiment, a fitted iterative nonlinear regression curve was used to determine effective ISO concentrations that caused 50% of the maximal response (EC50). The fitted iterative nonlinear regression curve that best represented the data were determined using a partial F-test, F = [(SS1–SS2)/SS2]/[(DF1–DF2)/DF2], where SS is sum of the squares and DF is degrees of freedom (p < 0.05). Significance between groups was tested using an unpaired two-tailed Student's t test (p < 0.05). All values are reported as the mean ± S.E. for n experiments. Equilibrium dissociation constant values (pKb) for subtype-selective β-AR antagonists were estimated using the method of analysis originally described by Arunlakshana and Schild (1959). Dose ratios were calculated for each individual experiment by dividing the ISO EC50 value that was calculated in the presence of a fixed concentration of AR antagonist by the control EC50 value for ISO. Schild regressions were constructed by plotting the log of the dose ratio –1 versus the log of the AR antagonist concentration. Linear regression analysis of these plotted points was used to determine the slope of the Schild regressions. Schild regression slopes are expressed as the mean ± 95% confidence interval and were only considered different from unity if the 95% confidence interval did not include the value of 1 (Kenakin, 1997). pKb values of subtype-selective β-AR antagonists causing inhibition of ISO-mediated increased AP frequencies were calculated from Schild regression x-intercepts. Differences in pKb values and Schild regression slopes were determined by analysis of covariance with a p < 0.05 level of probability accepted as significant.
Research Animals. Sprague-Dawley rats were obtained from Harlan (Madison, WI) and housed with their mothers before weaning. All protocols described above have been approved by the Institutional Animal Care and Use Committee at the University of North Dakota. Institutional Animal Care and Use Committee approval is in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.
Results
Functional CA1 Pyramidal Cell Responses Support Molecular Expression Patterns. Previous single cell RT-PCR investigations demonstrated genomic expression patterns for both the β1- and β2-AR in CA1 pyramidal neurons (Hillman et al., 2005). Therefore, we decided to examine whether either β-AR subtype is functionally expressed on these same cells. Using a cell-attached configuration, AP frequencies were recorded from a single CA1 pyramidal neuron. As increasing amounts of ISO were added to the perfusion line, a concentration-dependent escalation in AP frequency was observed (Fig. 2). Recordings were then repeated in adjacent slices that had been pretreated for 30 min with subtype-selective β-AR antagonists. Pretreatment with the selective β1-AR antagonist CGP 20712A at 10 nM, a concentration expected to block ∼90% of β1-ARs based upon binding data (Dooley et al., 1986), did not considerably affect the pyramidal cell response to ISO compared with control (EC50 = 84 ± 21 versus 51 ± 8.5 nM for control). However, pre-treatment with an effective concentration (10 nM) of the selective β2-AR antagonist ICI-118,551 resulted in a considerable decreased potency for ISO (EC50 = 1936 ± 320 versus 51 ± 8.5 nM for control), suggesting that the β2-AR is predominantly the functional β-AR subtype expressed by CA1 pyramidal neurons. This initial result corroborates our previous molecular studies that illustrate a transcriptional predominance of the β2-AR from these cells (Hillman et al., 2005).
Subtype-Selective β-AR Antagonists Demonstrate Affinity Values for the β2-AR. Given that single concentrations of subtype-selective β-AR antagonists suggested that the β2-AR was functionally expressed on CA1 pyramidal cells, we wanted to examine this subtype more thoroughly. Cell-attached recordings used to generate ISO concentration response were therefore repeated using fixed amounts of ICI-118,551 to determine the affinity of this AR antagonist for the β-AR subtype mediating increased AP frequencies. Pyramidal cells pretreated with 1, 5, and 10 nM ICI-118,551 produced 5-, 37-, and 100-fold parallel rightward shifts of the fitted ISO concentration-response curve (Fig. 3A). Dose ratios calculated for each individual run were plotted against each ICI-118,551 concentration to generate a straight line using linear regression analysis (Fig. 3B). The Schild regression slope included the value of unity (1.4 ± 0.4) and the x-intercept of the regression line represents the ICI-118,551 equilibrium dissociation constant (pKb) for the β-AR mediating the increased AP frequency. The calculated Kb of 0.3 ± 0.1 nM correlates to previously published values for ICI-118,551 in experiments where the antagonist was used to functionally identify the β2-AR (Table 1).
ISO concentration-response curves were repeated in the absence and presence of butoxamine, an agent that competitively blocks β2-ARs with an approximate 10-fold selectivity over the β1-AR (Tsuchihashi et al., 1990). Pretreatment of hippocampal slices with 700, 1000, and 5000 nM of this selective β2-AR antagonist again produced successive parallel rightward shifts in the ISO concentration-response curve (Fig. 4A). Schild regression analysis generated a line with a slope that included the value of unity (1.3 ± 0.3) and a calculated Kb for butoxamine of 355 ± 107 nM (Fig. 4B). This apparent affinity is similar to previously published values for butoxamine at the β2-AR (Table 1). The calculated high-affinity values of selective β2-AR antagonists imply that blockade of the β2-AR subtype affects the ISO-mediated pyramidal cell response in these studies.
To test the contribution of β1-ARs in the ISO-mediated pyramidal cell response, we used atenolol, an antagonist that demonstrates an approximate 30-fold selectivity for the β1-AR over the β2-AR subtype (Tsuchihashi et al., 1990). In these experiments, pretreatment of slices with a concentration of atenolol (0.3 μM) that selectively blocks β1- versus any β2-ARs present did not produce a significant shift in the ISO concentration-response curve (data not shown). However, when the atenolol concentration was increased to 7, 15, and 40 μM, respectively, there were successive parallel rightward shifts of the fitted concentration-response curves for ISO (Fig. 5A). Schild regression analysis generated a line whose slope value included unity (1.4 ± 0.5) and an x-intercept corresponding to an atenolol Kb value of 3162 ± 278 nM (Fig. 5B). This apparent Kb correlates to a previously published atenolol affinity value for a population of β2-ARs in rat brain (Tsuchihashi et al., 1990). This finding corroborates the previous data using selective β2-AR antagonists and together strongly suggests that the β2-AR subtype is the predominant β-AR mediating an increased AP frequency in response to ISO.
Increased AP Frequency Results from Direct Activation of CA1 Pyramidal Cell β2-ARs. CA1 pyramidal cells receive considerable excitatory input from other regions of the hippocampus, primarily from CA3 via the Schaffer collateral pathway. In situ studies have shown that β-AR transcripts are prevalent in CA3 (Nicholas et al., 1993), and studies from our own laboratory suggest the β1-AR is functionally expressed on CA3 pyramidal cells (Jurgens et al., 2005). Consequently, it is necessary to demonstrate that the ISO-mediated increased AP frequency we have observed in our studies is due to direct β2-AR activation on the CA1 pyramidal neuron itself and not resulting from glutamatergic or cholinergic input from CA3. To pharmacologically isolate the pyramidal cell being analyzed from any primary excitatory drive, a cocktail containing 10 μM DNQX, 50 μM APV, and 1 μM atropine was included in the pretreatment perfusion. These compounds were selected for their ability to block α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid/kainate receptors, N-methyl-d-aspartate receptors, and muscarinic receptors, respectively. In this isolated state, ISO still elicited the increased AP frequency seen previously, indicating that this excitability was due to direct activation of β2-ARs on the CA1 pyramidal cell (Fig. 6). Furthermore, when 10 nM of the selective β2-AR antagonist ICI-118,551 was included in this pretreatment cocktail, there was a significant rightward shift in the in the potency of ISO (1503 ± 372 nM) compared with control (35 ± 18 nM), which is similar to results illustrated in Fig. 3 (2972 ± 377 and 25 ± 4 nM, respectively). These functional data using subtype-selective β-AR antagonists on a pharmacologically isolated preparation, in conjunction with the previously described predominance of β2-AR transcripts from these same cells, strongly suggest that the increased AP frequency initiated by ISO is due to a direct activation of β2-ARs expressed on hippocampal CA1 pyramidal neurons.
Discussion
Adrenergic afferents extending from the locus ceruleus to the hippocampus has been accepted for decades, but to date few studies have carried out thorough characterizations of the AR subtypes expressed in this latter region. With growing evidence that β-AR activation promotes LTP in CA1, knowledge of AR subtype expression patterns is becoming increasingly pertinent. Here, we demonstrate that CA1 pyramidal cells, the principal neuronal cell type involved in LTP, predominantly express the β2-AR subtype. This conclusion is based upon previous work that documents consistent transcriptional expression of the β2-AR subtype using single-cell RT-PCR, and the work presented in this article that demonstrates antagonism of the β2-AR negatively affects the ISO-mediated increase in AP frequency as recorded directly from a single CA1 pyramidal neuron.
Preliminary screening of mRNA transcripts present in CA1 pyramidal cells was used to focus the functional characterization of AR subtypes to be examined in this study (Hillman et al., 2005). Our method of single-cell RT-PCR is capable of detecting as little as 1 pg of RNA, providing successful cDNA amplification ∼60% of the time. Of the nine AR subtypes examined in each sample, only mRNA for the β1- and β2-AR was found in calmodulin-dependent protein kinase II-positive cells (i.e., pyramidal neurons), with β2-AR transcripts most heavily expressed. This predominance of β2-(17 of 17 samples) over β1-AR transcripts (4 of 17) seen in our results is not unique. β-AR mRNA profiles of astrocytes in the hippocampus (Zhu and Kimelberg, 2004) and cortex (Shao and Sutin, 1992) demonstrate similar discrepancies between AR subtype expression. Interestingly, the three positive samples for β1-AR mRNA were taken from cells located in stratum pyramidale closer to the CA1/CA2 region of the slice, possibly indicating stronger coexpression of the two receptors and/or a switch to the β1-AR as one progresses toward CA3 (Jurgens et al., 2005).
Based on this molecular evidence for two β-AR subtypes, we elected to functionally characterize the receptor subtypes for CA1 pyramidal neurons in the hippocampus. Arguments for both pyramidal cell β1- (Vanhoose and Winder, 2003) and β2-AR (Hoogland and Saggau, 2004) expression exist; however, the majority of past studies examining β-AR function in CA1 have used nonsubtype-selective agents (i.e., isoproterenol and propranolol) or concentrations of subtype-selective agents used well outside their window of selectivity. This has rendered it difficult to draw definitive conclusions concerning specific β-AR subtypes expressed by CA1 pyramidal neurons. In addition, discrepancies of AR expression between species necessitate analysis of subtype expression specifically pertaining to the rat model. The aim of this study was therefore to use currently available subtype-selective β-AR antagonists at concentrations within their range of selectivity to decipher the β-AR subtypes functionally expressed on CA1 pyramidal cells.
A summary of our functional analyses for each subtypeselective β-AR antagonist is illustrated in Table 1, with comparisons with previously published equilibrium dissociation constants (Ki) at the β1- or β2-AR calculated from radioligand binding analysis. Presented in this manner, it becomes clear that our apparent Kb values calculated from Schild regression analysis agree with previously published affinities for these same compounds at the β2-AR subtype. Although the β-AR antagonists used in this study had varied levels of receptor subtype selectivity, the apparent Kb values strongly suggest that the heightened excitability seen in response to ISO is mediated by the β2-AR expressed on these CA1 pyramidal neurons with no detectable contribution to the AP frequency by the β1-AR subtype.
The β3-AR was not specifically examined in our functional studies, given that we found no evidence for genomic expression of this AR subtype using single-cell RT-PCR analysis. Aside from a report showing a generalized distribution pattern of β3-ARs throughout the brain (Summers et al., 1995), this AR subtype is minimally present in comparison with the β1- and β2-AR. If the β3-AR were indeed contributing to the ISO response of CA1 pyramidal cells, then Kb values calculated from Schild regression analysis would have likely reflected this. For example, Hoffmann has reported Ki values of 65,100 nM for atenolol and 611 nM for ICI-118,551 at the β3-AR, which are well beyond affinities calculated for these same β-AR antagonists in our study (Hoffmann et al., 2004). Furthermore, ISO has a lower affinity for β3-AR systems; reported Ki values for this AR agonist in transfected cell lines are upwards of 1500 nM for the β3-AR subtype, compared with the 200 to 500 nM range for ISO at the β1- and β2-ARs (Hoffmann et al., 2004). The high-potency response of ISO on CA1 pyramidal neurons observed in our studies would support the idea for functional expression of either the β1- or β2-ARs. Thus, it seems the β3-AR subtype is relatively absent from CA1 pyramidal neurons given the fact that no transcriptional evidence or functional effect was characterized.
It should be noted that β-ARs are not limited to the pyramidal cell layer, nor to just CA1. Much work in recent years has been devoted toward examining the AR expression profile of hippocampal astrocytes. β-ARs, specifically the β1- and β2-AR have been localized to CA1 astrocytes both in dissociated cell and slice preparations (Zhu and Kimelberg, 2004). Activation of these ARs may result in release of modulatory neurotransmitters that can affect neighboring neurons (Winder et al., 1996; Haydon, 2001); thus, the influences of glia on pyramidal cell activity cannot be ignored. Similarly, the presence of functional β1-ARs on CA3 pyramidal cells (Jurgens et al., 2005) must be taken into account given these cells are capable of providing feedforward excitation to CA1. Nonetheless we are confident that the increased ISO-mediated excitability of CA1 pyramidal neurons in our study is due to direct activation of a β2-AR on these cells and not a result of glial-pyramidal communication. This deduction is first based upon previous mRNA studies showing β1- and β2-AR transcription from only calmodulin-dependent protein kinase II-positive and not glial fibrillary acidic protein-positive cells. Second, there was no change in the ISO effect on pyramidal cell recordings observed in slices subjected to pharmacological isolation (i.e., pretreatment with DNQX, APV, and atropine). If ISO was indeed activating glial or CA3 β-ARs, prompting a release of excitatory neurotransmitters to CA1 pyramidal neurons, then results of the isolation experiment shown in Fig. 6 would have been significantly different from control. Although glial- and CA3-AR modulation of pyramidal cell activity cannot be exclusively ruled out, we are confident that in our system the foremost effect of ISO is due to direct activation of the β2-AR expressed on CA1 hippocampal pyramidal neurons.
In summary, this investigation characterizes the functional expression pattern of β-AR subtypes on individual CA1 pyramidal neurons. Previous mRNA evidence is supported in this study by Schild analysis using subtype-selective β-AR antagonists to differentiate the functional β-AR subtype causing increased excitability in this specific region of the hippocampus. Knowledge of β-AR subtype expression patterns not only expands our understanding of the AR influence in CA1 but also may reveal novel targets for future studies that examine AR modulation of LTP.
Acknowledgments
We thank Karen L. Cisek for assistance in editing the manuscript.
Footnotes
-
This study was supported in part by North Dakota Experimental Program to Stimulate Competitive Research (EPSCoR) through National Science Foundation Grant EPS-0132289 (to V.A.D.), National Science Foundation CAREER Grant 0347259 (to V.A.D.), National Institutes of Health Grant 2P20RR016471 from the Biomedical Research Infrastructure Networks program and National Institutes of Health Grant 5P20RR017699 from the Centers of Biomedical Research Excellence program (to V.A.D. and J.E.P.). A preliminary report of these findings was presented at the 2004 annual meeting of the American Society for Pharmacology and Experimental Therapeutics, Neuropharmacology Session, Apr 2–6, San Diego, CA.
-
doi:10.1124/jpet.105.084947.
-
ABBREVIATIONS: CA1, cornu ammonis 1; LTP, long-term potentiation; AR, adrenergic receptor; ISO, isoproterenol; APV, d-(–)-2-amino-5-phosphonopentanoic acid; CGP 20712A, (±)-2-hydroxy-5-[2-([2-hydroxy-3-(4-[1-methyl-4-(trifluoromethyl)-1H-imidazol-2-yl]phenoxy)propyl]amino)ethoxy]-benzamide methanesulfonate; DNQX, 6,7-dinitroquinoxaline-2,3-dione; ICI-118,551 hydrochloride, (±)-1-[2,3-(dihydro-7-methyl-1H-inden-4-yl)oxy]-3-[(1-methylethyl)amino]-2-butanol; aCSF, artificial cerebral spinal fluid; BAPTAK4, 1,2-bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid, tetrapotassium salt; AP, action potential; RT-PCR, reverse transcription-polymerase chain reaction.
-
↵1 These authors contributed equally to this work.
- Received February 22, 2005.
- Accepted May 18, 2005.
- The American Society for Pharmacology and Experimental Therapeutics