Effect of Xenon on Excitatory and Inhibitory Transmission in Rat Spinal Ventral Horn Neurons

• The site of action of anesthetics to prevent movement during surgery largely resides in the spinal cord

• Xenon produces anesthesia in part by reducing N -methyl-D-aspartate (NMDA) receptor and α-amino-3-hydroxy-5-methyl-4-isoxazole-4-propionic acid (AMPA) receptor activation, but whether it does so in the ventral horn is not known

• In the lamina IX neurons of rat spinal cord, xenon inhibits excitation of AMPA receptors but not NDMA receptors

• The immobilizing effects of xenon may be through an unexpected mechanism, reduction in excitation of AMPA receptors

XENON, as an inhalational anesthetic, combines profound anesthetic and analgesic properties with a low side-effect profile, and if used with an elaborate low-flow delivery system, it has the potential to become a main-line anesthetic.1,2Significant progress has been made toward elucidating how xenon produces anesthesia, and it has been revealed that xenon suppresses the excitatory synaptic transmission in the central nervous system.3,–,8A study on cultures of rat hippocampal neurons showed that xenon inhibited N -methyl-D-aspartate (NMDA) receptor-mediated synaptic transmission but not α-amino-3-hydroxy-5-methyl-4-isoxazole-4-propionic acid (AMPA) receptor-mediated synaptic transmission.4In later studies, xenon blocked both NMDA receptor-mediated and AMPA receptor-mediated synaptic transmission in cultures of mouse embryonic cortical neurons,5rodent amygdala,6prefrontal cortex,7and spinal cord dorsal horn.7,8By contrast, xenon had no effect on inhibitory synaptic transmission.6,–,8 

Originally, the brain was considered to be the principal effect site of anesthetics and the determiner of the minimum alveolar concentration (MAC) for volatile anesthetics, a common means to express the strength or potency of anesthetics. However, the MAC merely refers to the anesthetic concentration needed to prevent movement in response to noxious stimuli9and is caused by anesthetic actions in the spinal cord, rather than in the brain.10,11 

In the spinal cord, anesthetic actions might result from a combination of a reduction of the sensory transmission of nociceptive signals within the spinal dorsal horn and a suppression of the neuronal activity in the spinal ventral horn.12We reported previously that 50% xenon inhibited both NMDA receptor-medicated and AMPA receptor-mediated glutamatergic excitatory transmission via  a postsynaptic mechanism in the spinal dorsal horn8; however, xenon's actions in the spinal ventral horn have not yet been clarified. Therefore, the current study investigated the effects of xenon on spinal ventral horn neurons.

All experimental procedures involving the use of animals were approved by the Animal Care and Use Committee at Niigata University Graduate School of Medical and Dental Sciences (Niigata, Japan).

Neonatal Wistar rats (200 rats, 6–14 days old) were anesthetized with urethane (1.5 g/kg, intraperitoneal). Dorsal laminectomy was performed, and the lumbosacral segment of the spinal cord was removed.13The rats were killed immediately by exsanguination. Each spinal cord was placed in preoxygenated ice-cold artificial cerebrospinal fluid containing 117 mM NaCl, 3.6 mM KCl, 2.5 mM CaCl², 1.2 mM MgCl², 1.2 mM NaH²PO⁴, 25 mM NaHCO³, and 11.5 mM D-glucose. After all the ventral and dorsal roots were cut, the pia-arachnoid membrane was removed. The spinal cord was mounted on the metal stage of a microslicer (DTK-1500; Dosaka, Kyoto, Japan) and cut into 500-μm transverse slices. Each spinal cord slice was transferred to a recording chamber and placed on the stage of an upright microscope equipped with an infrared-differential interference contrast system (E600FN; Nikon, Tokyo, Japan). The slice was fixed with an anchor and superfused at 4–6 ml/min with artificial cerebrospinal fluid equilibrated with a gas mixture of 95% O²and 5% CO²(pH = 7.4) and maintained at 36°C using a temperature controller (TC-324B; Warner Instruments, Hamden, CT).

Lamina regions were identified under low magnification (×5 objective lens), and individual neurons were identified using a ×40 objective lens under an infrared-differential interference contrast microscope and monitored by a charge-coupled device camera (C2400–79H; Hamamatsu Photonics, Hamamatsu, Japan) on a video screen. Whole cell patch clamp recordings were made from the large lamina IX neurons (size, 15–25 μm).13In a previous study, these neurons were identified as motoneurons by fluorescence labeling with Evans blue dye injected into the rat hind limb the day before sacrifice.14Whole cell patch pipettes were constructed from borosilicate glass capillaries (1.5 mm OD; World Precision Instruments, Sarasota, FL). The resistance of a typical patch pipette was 4–8 MΩ when filled with internal solution. The patch pipette solutions contained 110 mM Cs²SO⁴, 5 mM tetraethylammonium, 0.5 mM CaCl², 2 mM MgCl², 5 mM EGTA, 5 mM HEPES, and 5 mM ATP-Mg.

Signals were amplified by an Axopatch 200B amplifier (Molecular Devices, Union City, CA), filtered at 2 kHz, and digitized at 5 kHz. All experiments were performed in voltage-clamp mode at a holding potential of −70 mV for recording exogenously applied AMPA-induced currents or AMPA receptor-mediated electrically evoked excitatory postsynaptic current (EPSCs) and miniature EPSCs. Exogenously applied NMDA-induced currents and NMDA receptor-mediated electrically evoked EPSCs were recorded at +40 mV. Inhibitory currents induced by exogenous γ-aminobutyric acid (GABA) or glycine, as well as miniature inhibitory postsynaptic current (IPSCs), were recorded at 0 mV. Electrically evoked EPSCs were elicited by focal stimulation of the deep dorsal horn with a concentric bipolar tungsten electrode (diameter = 10 μm). This stimulation area was identified visually under low magnification (×5 objective lens). Stimulus parameters were as follows: intensity, 1–5 V; duration, 0.05 ms; and frequency, 0.1 Hz. Electrically evoked EPSCs that displayed a constant latency and lack of failures with high frequency stimulation (10 Hz) were classified as monosynaptic. Data were stored and analyzed using a pCLAMP 9.2 data acquisition program (Molecular Devices) and Mini-analysis 6.0.7 (Synaptosoft, Leonia, NJ). Each peak current was measured before and after xenon treatment and expressed as (posttreatment/pretreatment) × 100 (as percentages).

Xenon was acquired from TG Showa (Tokyo, Japan). AMPA, NMDA, GABA, glycine, bicuculline, strychnine, 6-cyano-7-nitroquinoxaline-2,3-dione disodium, and D, L-2-amino-5-phosphonopentanoic acid were acquired from Sigma–Aldrich (St. Louis, MO). Tetrodotoxin was acquired from Wako (Osaka, Japan). Drugs were applied to the whole slice by perfusion via  a three-way stopcock without changing the perfusion rate or temperature. The volume of the recording chamber was approximately 1.2 ml. The drugs reached the recording chamber within 15 s after the stopcock was opened, and the drugs were completely washed out within 30 s after the stopcock was closed. Xenon was mixed in a 1:1 ratio with a prefabricated gas containing 90% O²and 10% CO², so that the composition of the gas mixture was 50% xenon, 45% O², and 5% CO². For the additional experiments, xenon was mixed in a 3:1 ratio with a prefabricated gas containing 80% O²and 20% CO², so that the composition of the gas mixture was 75% xenon, 20% O², and 5% CO². Xenon was delivered through an ultrathin polyethylene tube and applied by bubbling through the perfusing solution saturated with the gas mixture for 30 min.

Numerical data are represented as means ± SD. Statistical significance was determined as P < 0.01 using the Student paired t  test or the Kolmogorov-Smirnov test, as appropriate. When referring to electrophysiological data, “n” indicates the number of neurons studied. The continuous curve for the concentration-response relationship of AMPA, NMDA, GABA, and glycine was drawn according to the Hill equation:

where x  is the agonist concentration, y  is the relative amplitude of agonist-induced current (%), and y^max  is the maximal value of y . The term k  is the half-maximal effective concentration (EC⁵⁰; μM or mM), and n  is the slope of the curve (Hill coefficient).

Because there was no database available regarding the proper concentration of the agonists for the experiments in the spinal lamina IX neurons, the concentration-response curve was obtained for the agonists AMPA, NMDA, GABA, and glycine on the spinal lamina IX neurons (fig. 1). Exogenous application of AMPA elicited an inward current in neurons at −70 mV. Exogenous application of NMDA at +40 mV or of GABA or glycine at 0 mV elicited an outward current. Each agonist was applied by perfusion for 15 s and induced a current in a concentration-dependent manner. The EC⁵⁰values for AMPA, NMDA, GABA, and glycine on the spinal lamina IX neurons were as follows: AMPA, 7.0 μM; NMDA, 62.8 μM; GABA, 0.1 mM; and glycine, 0.2 mM. The Hill coefficient values were as follows: AMPA, 1.6; NMDA, 1.7; GABA, 1.1; glycine, 1.3. We set the concentration of the agonists as follows: AMPA, 10 μM; NMDA, 50 μM; GABA, 0.5 mM; glycine, 0.5 mM.

To confirm that these exogenous agonist-induced currents were postsynaptic phenomena, we examined these currents in the presence of tetrodotoxin (1 μM) to remove any possible influence of presynaptic neurons. Tetrodotoxin did not affect the amplitudes or the integrated area: AMPA (amplitude = 105 ± 12% of control values, P = 0.36; area = 106 ± 13% of control values, P = 0.31, n = 6), NMDA (amplitude = 95 ± 13% of control values, P = 0.29; area = 102 ± 23% of control values, P = 0.84, n = 9), GABA (amplitude = 108 ± 13% of control values, P = 0.21; area = 113 ± 12% of control values, P = 0.05, n = 5), glycine (amplitude = 110 ± 14% of control values, P = 0.10; area = 105 ± 18% of control values, P = 0.50, n = 6). Therefore, the following experiments to examine the effects of xenon on exogenous agonist-induced currents were performed in the absence of tetrodotoxin.

Exogenous application of AMPA (10 μM, 15 s) elicited an inward current in the spinal lamina IX neurons at −70 mV (fig. 2A and D). Preapplication of 50% xenon for 5 min reversibly reduced the peak amplitudes of the AMPA-induced currents to 72 ± 9% of the control values (n = 7, P < 0.01; fig. 2B) and the integrated area to 73 ± 13% of the control values (n = 7, P < 0.01; fig. 2C). Exogenous application of NMDA (50 μM, 15 s) induced an outward current at +40 mV (fig. 3A); 50% xenon did not change the peak amplitudes (99 ± 13% of control values, n = 6, P = 0.91) or the integrated area (99 ± 22% of control values, n = 6, P = 0.91) of NMDA-induced currents.

Seventy-five percent xenon reversibly reduced the peak amplitudes of the AMPA-induced currents to 67 ± 12% of the control values (n = 10, P < 0.01; fig. 2E) and the integrated area to 62 ± 12% of the control values (n = 10, P < 0.01; fig. 2F). However, 75% xenon did not change the peak amplitudes (103 ± 7% of control values, n = 10, P = 0.30) or the integrated area (100 ± 14% of control values, n = 10, P = 0.97) of NMDA-induced currents.

To examine the effects of xenon on synaptic transmission, electrically evoked EPSCs were elicited by focal stimulation of the deep dorsal horn. Evoked EPSCs can be divided into different groups. EPSCs that displayed a constant latency and lack of failure with high frequency stimulation (10 Hz) were classified as monosynaptic, whereas EPSCs that displayed variable latencies and failures were considered polysynaptic (fig. 4, A–C). We investigated the effects of xenon on evoked monosynaptic EPSCs.

α-Amino-3-hydroxy-5-methyl-4-isoxazole-4-propionicacid receptor-mediated electrically evoked EPSCs were isolated pharmacologically with D,L-2-amino-5-phosphonopentanoic acid (50 μM), bicuculline (10 μM), and strychnine (2 μM) at −70 mV. Figure 5A shows the representative average traces of AMPA receptor-mediated evoked EPSCs before, during, and after xenon application. Fifty percent xenon reversibly reduced the peak amplitudes of AMPA receptor-mediated evoked EPSCs to 69 ± 13% of the control values (n = 9, P < 0.01). These evoked EPSCs were almost completely abolished by the non-NMDA receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione disodium (10 μM, 10 ± 3% of the control values, n = 9, P < 0.01; fig. 5, A–C), suggesting that the evoked EPSCs were mediated by AMPA receptor.

N -methyl-D-aspartate receptor-mediated electrically evoked EPSCs were isolated pharmacologically with 6-cyano-7-nitroquinoxaline-2,3-dione disodium (10 μM), bicuculline (10 μM), and strychnine (2 μM) at +40 mV. Figure 5D shows the representative average traces of NMDA receptor-mediated evoked EPSCs before and during xenon application. The peak amplitudes of NMDA receptor-mediated evoked EPSCs were not affected by 50% xenon (98 ± 3% of the control values, n = 6, P = 0.09). These evoked EPSCs were almost completely abolished by the NMDA receptor antagonist D,L-2-amino-5-phosphonopentanoic acid (50 μM, 7 ± 3% of the control values, n = 6, P < 0.01; fig. 5, D–F), suggesting that the evoked EPSCs were mediated by NMDA receptor.

To determine the site of the inhibitory action of xenon, we examined the effects of xenon on miniature EPSCs in the presence of tetrodotoxin (1 μM) at −70 mV. Fifty percent xenon decreased the mean amplitudes of the miniature EPSCs to 82 ± 14% of the control values (n = 9, P < 0.01), whereas the mean frequencies of miniature EPSCs remained unchanged (102 ± 40% of the control values, n = 9, P = 0.91, fig. 6A). Figure 6B shows the effects of xenon on cumulative distributions of miniature EPSC amplitudes and interevent intervals. Although xenon increased the proportion of miniature EPSCs with a significantly smaller amplitude compared with that of the control (P < 0.01), it had no effect on the cumulative distribution of the interevent intervals of miniature EPSCs (P = 0.39). The decay time of miniature EPSCs was reversibly prolonged by xenon (111 ± 25% of the control values, n = 9, P = 0.22); however, this change was not statistically significant.

Exogenous application of GABA (0.5 mM, 15 s) and glycine (0.5 mM, 15 s) elicited an outward current at 0 mV (fig. 3, B and C). Fifty percent xenon did not change the peak amplitudes (99 ± 9% of the control values, n = 10, P = 0.76) or the integrated area (101 ± 19% of the control values, n = 10, P = 0.87) of GABA-induced currents. Similarly, xenon did not change the peak amplitudes (105 ± 13% of the control values, n = 9, P = 0.29) or the integrated area (108 ± 12% of the control values, n = 9, P = 0.07) of glycine-induced currents.

We also examined the effects of xenon on miniature IPSCs in the presence of tetrodotoxin (1 μM) at 0 mV. Fifty percent xenon did not change the mean amplitudes (100 ± 32% of the control values, n = 9, P = 0.97; fig. 7A) or the mean frequencies (100 ± 20% of the control values, n = 9, P = 0.95) of miniature IPSCs. Xenon had no effect on the cumulative distributions of miniature IPSC amplitudes (P = 0.15; fig. 7B) or interevent intervals (P = 0.61).

Several lines of evidence suggest that xenon suppresses excitatory synaptic transmission in central nervous system areas such as the brain and spinal cord dorsal horn.3,–,8However, its actions in the spinal ventral horn have not been clarified. Thus, we investigated the effects of xenon on excitatory and inhibitory synaptic transmission in the spinal ventral horn neurons.

In the current study, 50 and 75% xenon inhibited both the peak amplitudes and the integrated area of exogenous AMPA-induced currents, as well as the peak amplitudes of AMPA receptor-mediated evoked EPSCs. In addition, 50% xenon decreased the mean amplitudes of miniature EPSCs, but the mean frequency of miniature EPSCs remained unchanged. However, 50 and 75% xenon had no substantial effect on exogenous NMDA-induced currents or NMDA receptor-mediated evoked EPSCs. In addition, 50% xenon had no effect on currents induced by GABA or glycine or on miniature IPSCs. Taken together, these findings suggest that xenon acts on both synaptic and extrasynaptic AMPA receptors and inhibits AMPA receptor-mediated glutamatergic excitatory transmission via  a postsynaptic mechanism in the spinal lamina IX neurons.

However, several studies have demonstrated that xenon depressed both NMDA and AMPA receptor-mediated glutamatergic excitatory synaptic transmission postsynaptically in the amygdala,6prefrontal cortex,7and substantia gelatinosa,7,8whereas xenon did not affect inhibitory synaptic transmission. This discrepancy between our results and those of previous studies could be attributed to a different composition of NMDA receptors expressed in these neurons.

The NMDA receptor has been proposed as a major target for xenon.4,15,16This receptor is composed of subunits from at least two families: NR1 and NR2. The NR1 subunit is essential for the function of NMDA receptors and is expressed ubiquitously in the entire spinal cord, including the spinal ventral horn motoneurons and the spinal dorsal horn neurons.17,–,20The functional properties of NMDA receptors are determined by the NR2 subunit composition (NR2A–2D).20,–,22The NR3 subunits (NR3A and 3B) occasionally are expressed in addition to NR1 and NR2 subunits and functionally suppress NMDA receptors.23,24The NR2C and NR2D subunits are expressed weakly, but the NR2A and NR2B subunits are not identified in substantia gelatinosa neurons of the adult rat lumbar spinal cord.18We reported previously that 50% xenon depressed NMDA receptor-mediated glutamatergic excitatory transmission via  a postsynaptic mechanism in adult substantia gelatinosa neurons.8By contrast, the NR2A subunit is expressed predominantly in neonatal motoneurons.19However, around postnatal day 10–14, the NR2A subunit decreases, and the major regulatory subunit of the NMDA receptors switches to the NR3B subunit; the NR3B subunit is maintained until the adult stage.19The expression of the NR3B subunit is restricted to somatic motoneurons in the adult brainstem and spinal cord.19Considering the neonatal age of rats in the current study, these reports suggest that we may have examined the effects of xenon mainly on the NR2A subunit; the lack of NMDA receptor response to xenon may be caused by the lack of response of the NR2A subunit to xenon. In short, the subunits of the NMDA receptors expressed in neonatal motoneurons and in adult substantia gelatinosa neurons are different. This different composition of NMDA receptors may underlie the variability among tissues in the effects of xenon on NMDA receptor-mediated synaptic transmission.

In our current study, we were confronted with the issue of xenon concentration. At the beginning, a single concentration of 50% xenon was used in our experiments for technical and financial reasons. The MAC of xenon is estimated to be 86–161% in rats25,26; thus, 50% xenon is calculated to be 0.31–0.58 rat MAC. Xenon is known to have analgesic properties even at subanesthetic (0.3 MAC) concentrations in rats27and humans.28,29Although the concentration of dissolved xenon could not be measured, we calculated the concentration of dissolved 50% xenon at approximately 1.9 mM according to previous studies.4,–,6,30,31A previous study showed that 1.9 mM dissolved xenon depressed the NR2A-containing NMDA receptors in the amygdala.32In addition, we previously demonstrated that 50% xenon depressed NMDA receptor-mediated glutamatergic excitatory transmission in adult substantia gelatinosa neurons.8Therefore, we concluded that this 50% xenon concentration was sufficient to modulate synaptic transmission in the spinal lamina IX neurons. Nevertheless, 50% xenon did not affect NMDA receptor-mediated synaptic transmission in our current investigation. However, xenon may suppress NMDA receptor-mediated synaptic transmission or may produce some other effects at higher concentration. These effects would be more critical to immobilization than the effects on AMPA receptor-mediated synaptic transmission observed in this study. Thus, we also examined the effects of 75% xenon on AMPA- and NMDA-induced currents. Seventy-five percent xenon reversibly reduced the peak amplitudes and the integrated area of the AMPA-induced currents. However, 75% xenon did not change the peak amplitudes or the integrated area of NMDA-induced currents. Considering the maintenance of physiologic concentration of oxygen and carbon dioxide, 75% xenon is the maximum concentration in our current investigation. Nevertheless, NMDA-induced currents were not affected by 75% xenon.

Non-NMDA glutamate receptors include both AMPA receptors and kainate receptors. Neonatal rat motoneurons also express kainate receptors.33However, most of the glutamatergic non-NMDA receptor-mediated synaptic transmissions in neonatal rodent motoneurons are mediated via  AMPA receptors.34,–,36Therefore, we interpreted glutamatergic non-NMDA receptor-mediated synaptic transmission as AMPA receptor-mediated synaptic transmission in this study.

The effects of 50% xenon on AMPA receptor-mediated synaptic transmission in the spinal lamina IX neurons were less than those in the spinal dorsal horn neurons8; however, several previous reports have indicated that the dorsal horn plays a lesser role in immobility and have suggested that the ventral horn is the site of anesthetic actions.37,–,39These results might also suggest that xenon acts differently than other anesthetics by acting predominantly on the spinal dorsal horn neurons, rather than on the spinal ventral horn motoneurons.

In a previous study, Cheng and Kendig35investigated the effects of the inhalational anesthetic enflurane on synaptic transmission in spinal ventral horn neurons. Electrically evoked EPSCs were elicited by electrical stimuli applied to the dorsal root, and the researchers concluded that enflurane depressed glutamatergic excitatory synaptic transmission in motoneurons.35However, the inhalational anesthetic isoflurane40and nitrous oxide41affected synaptic transmission in substantia gelatinosa neurons of the adult rat spinal cord. It was also reported that isoflurane affected electrically evoked EPSCs in substantia gelatinosa neurons of the immature rat spinal cord.42In addition, xenon inhibited AMPA and NMDA receptor-mediated excitatory synaptic transmission in substantia gelatinosa neurons of the adult rat spinal cord.7,8These results suggest that the electrically evoked postsynaptic currents may be affected in the substantia gelatinosa before reaching the spinal lamina IX neurons when the dorsal root or root entry zone is stimulated. Thus, in the current study, we elicited electrically evoked EPSCs by focal stimulation of the deep dorsal horn. Despite using this method, we could not completely exclude the possibility that deep dorsal horn stimulation might evoke monosynaptic responses such as Ia inputs that may not be relevant to the MAC determination.

The ventral interneuronal networks are also important for generating motor responses to noxious stimuli and determining the MAC.37,39The importance of ventral interneurons is supported by the evidence that spinal sensory neurons and motoneurons are relatively resistant to anesthetics, despite showing some individually depressive effects.43 

To identify large spinal lamina IX neurons under an infrared-differential interference contrast microscope, we excluded older rats, whose spinal lamina IX neurons were impossible to identify because of highly developed fibrous tissue. However, the major regulatory subunit of the NMDA receptors switches from NR2A to NR3B in motoneurons around postnatal day 10–14,19and this switch of NMDA receptor subunits indicates a possibility that xenon may depress NMDA receptor-mediated synaptic transmission in adult motoneurons. In contrast, another study showed that the sensitivities of NMDA receptors to Mg²⁺, isoflurane, ketamine, nitrous oxide, and ethanol were not altered by the NR3B subunit, although the NR3B subunit prominently reduced the amplitude of NMDA currents.44These results indicate that NMDA receptor-mediated synaptic transmission may not be affected by xenon in adult motoneurons. In any case, additional studies are required using identified spinal motoneurons in the adult rat.

In conclusion, we investigated the effects of xenon on the spinal lamina IX neurons as one of the components of the spinal reflex pathway and clarified that xenon inhibited AMPA receptor-mediated glutamatergic excitatory synaptic transmission in the spinal lamina IX neurons via  a postsynaptic mechanism. These results indicate that the suppression of neuronal activity in the spinal ventral horn and substantia gelatinosa neurons might contribute to immobility and could account for the mechanism of xenon as an anesthetic for the prevention of movement in response to noxious stimuli and to determine the MAC. However, the modest effects of xenon on AMPA receptors and the lack of effect on the NMDA receptors might mean that the spinal ventral horn neurons are not a likely site for the immobilizing action of xenon; additional investigation of ventral interneurons existing in the lamina VII and VIII would help resolve these questions.