Male Sprague-Dawley rats were anesthetized with ether (22 vol % in air) and the brain was rapidly removed and placed in ice cold (5°C) and pregassed (95/5 % O2
, carbogen) artificial cerebral spinal fluid (ACSF). The ACSF had the following composition (in mM): Na 151.25; K 2.5; Ca 2.0; Mg 2.0; Cl 131.5; HCO3
1.25; and glucose 10. Whole brain coronal slices (450 μm) were cut using a vibratome (Campden Instruments), following careful removal of the dura and pia membranes. Hemisected brain slices were equilibrated for at least one hour at room temperature in an incubation chamber filled with ACSF and continually bubbled with carbogen. Individual slices were transferred to a recording chamber and equilibrated for an additional 10 minutes prior to electrophysiological recording. Oxygenated ACSF solution was continuously perfused through the chamber at a flow rate of 3.0 ml/min and maintained at 22 ± 1°C. The present studies were carried out at room temperature because synaptic responses recorded from cooler brain slices exhibit considerably better baseline stability and the tissue remains viable for many more hours in vitro
compared to slices maintained at physiological temperatures. Room temperature also facilitates the use of submerged preparations (oxygen solubility and delivery to slices is increased), which allows the use of 60× optics to visualize single neurons for the patch clamp recordings used in some experiments. Previous studies comparing both volatile and intravenous anesthetic effects at physiological and cooler temperatures in brain slices found that there were no apparent differences in effects [43
]. The most important effect of lower temperature is to increase the aqueous solubility of the volatile anesthetics and previous work from our laboratory has described in detail the solubility changes observed at 22 vs
. 35°C and our methods for measuring and compensating for changed aqueous solubility, as well as the remarkably similar physiological responses recorded from brain slices at these two temperatures [43
To measure population spikes, bipolar tungsten microelectrodes were placed on Schaffer-collateral fibers to electrically stimulate inputs to hippocampal CA1 pyramidal neurons. Glass recording electrodes filled with ACSF (2 to 5 KOhm) were placed in stratum pyramidale to record stimulus-evoked population spike field potentials, or in stratum radiatum to record field EPSPs. Single stimulus pulses (0.01 to 0.05 ms duration; 10 to 80 μA @ 1.0 to 5.0 V) were delivered via constant current isolation units (Grass Instruments, SIU 6D) from a Grass S8800 two channel stimulator; at stimulus rates of 0.05 Hz. Field potential signals were amplified (× 1000), filtered (1 Hz to 10 KHz, bandpass), conditioned (DC offset), and digitally stored for later analysis (A/D with 20 μs resolution on a 486, and 50 MHz microcomputer using Data Wave Systems Corp. or Strathclyde Electrophysiological software).
Whole cell patch-clamp recordings were made using thin-walled borosilicate capillaries (1.5 mm O.D.) pulled in two stages on a Narishige PP83 pipette puller. Patch electrodes were filled with the following intracellular solution (in mM): potassium gluconate or CsCl2 – 100, EGTA – 10, MgCl2 – 5, HEPES free acid – 40, ATP disodium salt – 0.3, and GTP sodium salt -0.3. The electrode solution also contained the local anesthetic QX 314 (1.0 mM) in some experiments, to prevent action potential discharge that would contaminate recordings of IPSCs. Electrode solutions were filtered and pH adjusted to 7.2 using KOH or CsOH and had a final osmolarity of 260 to 270 mOSM. Patch electrodes with a DC resistance of 4 to 5 MOhm were used. Recordings were made using an Axoclamp 2A preamplifier (Axon Instruments) in single electrode voltage clamp mode with > 80 % series resistance compensation and > 5 GOhm seals. Patch-clamp current signals were filtered (0.1 Hz to 10 KHz, bandpass), amplified (× 100) and digitized (10 KHz) for storage and analysis. Frequency and amplitudes of IPSCs were analyzed using Data Wave Technologies and Strathclyde Electrophysiological software in a continuous data recording configuration.
The intravenous anesthetics (propofol and pentobarbital) were made fresh for each experiment, solubilized using 0.5% dimethyl sulfoxide (DMSO) and sonicated immediately prior to test administration in stock solutions and serially diluted into ACSF to achieve the final concentrations for testing. Volatile anesthetics (halothane and isoflurane) were applied in the perfusate at equilibrated concentrations, delivered from calibrated vaporizers and bubbled into the perfusate for at least 10 min prior to switching from control ACSF, to ensure steady-state concentrations were achieved. The concentration of volatile anesthetics in the gas phase were continually measured using a Puritan-Bennett anesthetic monitor. Only a single concentration of a given anesthetic was tested on each brain slice.
Data are expressed as the mean ± standard deviation and statistical analysis (ANOVA with post Tukey test) was performed using Instat from GraphPad Software. For drug effects on paired pulse inhibition, the percent change was first calculated as: Drug 1st/Control 1st = (0.5 × 100) - 100 % = 50 % depression; and Drug 2nd/Control 2nd = (X × 100) - 100 % = X % depression. Then the percent increase in paired pulse inhibition was = (X / 50) × 100 %. This approach has the advantage of normalizing paired responses with respect to varying degrees of population spike facilitation observed during control recordings, i.e. differing degrees of EPSP facilitation on a background of differing degrees of inhibition from preparation to preparation.