Abstract
Functional 5-hydroxytryptamine type 3 (5-HT3) receptors can be formed by 5-HT3A subunits alone or in combination with the 5-HT3B subunit, but only the 5-HT3A receptor has been previously studied with respect to the modulation by volatile anesthetics and n-alcohols. Using two-electrode voltage-clamp, we show for the first time the modulation of heteromeric human (h)5-HT3AB receptors, expressed in Xenopus oocytes, by a series of n-alcohols and halogenated volatile anesthetics. At twice their anesthetic concentration, compounds having a molecular volume of less than 110 Å3 enhanced submaximal 5-HT-evoked current. Compounds larger than 110 Å3 inhibited submaximal 5-HT-evoked current. In experiments examining 5-HT concentration-response relationships, chloroform and butanol caused a slight decrease in the 5-HT EC50. Sevoflurane and octanol inhibited 5-HT-evoked current at all 5-HT concentrations tested but had no effect upon the 5-HT EC50. Compared with previous data on homomeric h5-HT3A receptors, the presence of the h5-HT3B subunit reduces the enhancement of h5-HT3 receptors by smaller halogenated volatile anesthetics and n-alcohols. In summary, these results suggest that heteromeric h5-HT3AB receptors are modulated by halogenated volatile anesthetics at clinically relevant concentrations, in addition to n-alcohols, suggesting that these receptors may be another physiological target for these compounds. The modulation is dependent upon the molecular volume of the compound, further supporting the concept of an anesthetic binding pocket of limited volume common on other Cys-loop ligand-gated ion channels. Incorporation of the 5-HT3B subunit alters either the anesthetic binding site or the allosteric interactions between anesthetic binding and channel opening.
5-Hydroxytryptamine (serotonin) type 3 (5-HT3) receptors belong to a superfamily of Cys-loop ligand-gated ion channels that includes the nicotinic acetylcholine, γ-aminobutyric acid type A (GABAA), glycine, and zinc-activated channel receptors (Davies et al., 2003). The structure of the 5-HT3 receptor is presumed to be similar to all receptors in the Cys-loop superfamily of ligand-gated ion channels, namely, oligomers consisting of five subunits arranged to form a nonselective cation channel (Reeves and Lummis, 2002). The human genome contains five genes encoding for different 5-HT3 subunits (5-HT3A–E) (Niesler et al., 2003). Presently, only two of these subunits, 5-HT3A and 5-HT3B, have been shown to be involved with the formation of functional receptors. Unlike the 5-HT3A subunit, the 5-HT3B subunit alone is unable to form homomeric receptors, but it can combine with 5-HT3A subunits to form heteromeric 5-HT3AB receptors that differ from homomeric 5-HT3A receptors in pharmacological and biophysical properties (Davies et al., 1999; Dubin et al., 1999). Although the exact subunit composition of receptors in vivo is not fully known, there does appear to be a region specific expression of the different subunits (Morales et al., 2001; Morales and Wang, 2002; Niesler et al., 2003). Expression of the 5-HT3B subunit may even be species-dependent, since Northern blot analysis and reverse transcription-polymerase chain reaction revealed messages for 5-HT3B subunits in human and monkey brain (Davies et al., 1999; Dubin et al., 1999), whereas in situ hybridization and reverse transcription-polymerase chain reaction techniques failed to find 5-HT3B subunits in rat brain (Morales and Wang, 2002). However, interneurons expressing the 5-HT3B subunit have been identified in the rat hippocampus with the use of a polyclonal antibody (Monk et al., 2001).
Within the central and peripheral nervous systems, activation of presynaptic 5-HT3 receptors is associated with a modulation in release of neuropeptides and neurotransmitters such as acetylcholine, GABA, dopamine, glutamate, and vasoactive intestinal peptide (Wang et al., 1998; Koyama et al., 2000; Férézou et al., 2002); hence, any modulation of 5-HT3-mediated release by anesthetics will have potentially significant physiological effects. For example, 5-HT3 receptors are involved with autonomic reflexes, including emesis (Hornby, 2001), blood pressure, and heart rate (Comet et al., 2004). In fact, 5-HT3 receptor antagonists are used to prevent and treat postoperative nausea and vomiting, which has a strong association with the use of halogenated volatile anesthetics (Apfel et al., 2002).
5-HT3 receptors, along with nicotinic acetylcholine, GABA, and glycine receptors, are anesthetic-sensitive. Previous studies have shown that volatile anesthetics and n-alcohols modulate currents mediated by rodent 5-HT3A (Machu and Harris, 1994; Jenkins et al., 1996; Zhou and Lovinger, 1996) and human 5-HT3A receptors (Suzuki et al., 2002; Stevens et al., 2005). The molecular volume of the anesthetic determines the modulation of submaximal 5-HT-evoked currents with currents being enhanced by smaller (<120 Å3) compounds (Stevens et al., 2005). Because the subunit composition of ligand-gated ion channels can influence the modulation by anesthetics, the aim of this study was to examine the modulation of heteromeric h5-HT3AB receptors by anesthetic compounds and determine the influence the 5-HT3B subunit has on their modulation.
This is the first report on the effects of halogenated volatile anesthetics and n-alcohols on heteromeric human 5-HT3AB receptors. Here, we show that a variety of volatile anesthetics and n-alcohols do indeed modulate human 5-HT3AB receptor-mediated currents and that the modulation characteristics differ compared with those of 5-HT3A receptors.
Materials and Methods
Xenopus Oocyte Preparation and Receptor Expression. Oocytes from human chorionic gonadotropin-injected adult female Xenopus laevis frogs (Xenopus One, Ann Arbor, MI) were surgically removed under anesthesia with 0.2% tricaine and hypothermia. cDNA encoding the human 5-HT3A and 5-HT3B subunits were generously supplied by E. Kirkness (TIGR, Rockville, MD) and transcribed into mRNA using the mMESSAGE mMACHINE High Yield Capped RNA Transcription Kit (Ambion, Inc., Austin, TX). Defolliculated stage V and VI oocytes were injected with 25 to 50 nl of cRNA encoding for the h5-HT3A subunit or a cRNA mixture encoding for both the h5-HT3A and h5-HT3B subunits (in the ratio of 1:2:1 by volume for 5-HT3A subunit/5-HT3B subunit/water). Oocytes were incubated at 18°C for 18 h to 8 days in filter-sterilized ND-96 solution: 96 mM NaCl, 2 mM KCl, 10 mM HEPES, 1.8 mM CaCl2, 1.0 mM MgCl2, 5 units/ml penicillin, and 5 μg/ml streptomycin, pH adjusted to 7.5 with NaOH prior to performing electrophysiological experiments.
Electrophysiological Recording. Currents from oocytes expressing 5-HT3 receptors were recorded using the two-electrode voltage-clamp technique. Oocytes were placed in a 40-μl recording chamber, impaled with two capillary glass electrodes (A-M Systems, Carlsborg, WA) filled with 3 M KCl (resistance <5 MΩ), and were voltage-clamped at -50 mV using a GeneClamp 500B amplifier (Axon Instruments, Union City, CA). During electrophysiological recording, oocytes were constantly superfused at a rate of 5 ml/min with buffer solution: 96 mM NaCl, 2 mM KCl, 10 mM HEPES, 1.0 mM CaCl2, and 0.8 mM MgCl2, pH adjusted to 7.5 with NaOH, using a closed-syringe superfusion system when volatile anesthetics were present. Currents were filtered at 1 KHz, sampled at 33.3 Hz, and recorded using Clampex v.9 software (Axon Instruments).
Protocols and Data Analysis. The effects of several volatile and alcohol anesthetics on agonist-mediated currents were surveyed using 2 μM 5-HT, an agonist concentration that evokes approximately 10% of the maximal peak current for 5-HT3AB receptors (EC10 concentration). The EC10 concentration is commonly measured when examining the modulation of Cys-loop ligand-gated ion channels since it allows any enhancements to be measured. Oocytes were preincubated with anesthetic for 30 s prior to coapplication of anesthetic plus 5-HT for 60 s. Each experiment was preceded and followed by a control application of 5-HT, both to normalize data and to ensure the reversibility of any drug-induced modulation of currents.
For analysis, the average of these two measurements was used as the control. A 3-min recovery period was allowed after each application of agonist (with or without anesthetic). Experiments were repeated in at least three oocytes.
The molecular volumes of all anesthetic agents were determined using Mac Spartan Pro v1.01 (Wavefunction, Inc., Irvine, CA) on an Apple MacIntosh G4 computer. Geometry optimization was performed using ab initio molecular orbital calculations (Hartree-Fock, 3–21G basis set). The correlation coefficient relating anesthetic and n-alcohol action with molecular volume was calculated using SPSS 9.0 (SPSS, Inc., Chicago, IL).
For experiments examining the effects of halogenated volatile anesthetics and n-alcohols on agonist concentration-response relationships, the oocyte was first exposed for 15 s to 300 μM 5-HT, a concentration that yields maximal current. Following a 5-min recovery period, a control current was measured at a specific 5-HT concentration (15- to 75-s exposure time, depending on concentration). Following another recovery period (3–5 min, depending on 5-HT concentration), the maximal current response was again determined using 300 μM 5-HT. After a further 5-min recovery period, the oocyte was preincubated with anesthetic for 30 s prior to the coapplication of anesthetic and 5-HT. Finally, after another 3- to 5-min recovery period, the maximal current response was again measured using 300 μM 5-HT to ensure reversibility.
For analysis, all currents were normalized to the average maximal current elicited by 300 μM 5-HT immediately preceding and following each measurement. Normalized data were plotted as mean ± standard deviation. The 5-HT concentration-response curves were fitted to the following Hill equation: where I is the peak current at a certain concentration of 5-HT, Imax is the maximum 5-HT-evoked current, EC50 is the concentration of 5-HT that elicits 50% of the maximal response, and n is the Hill coefficient. Serotonin EC50 values were calculated to evaluate shifts in the 5-HT concentration-response relationship curves in the presence of volatile anesthetics and alcohols.
Data were analyzed post hoc using Graphpad Prism v.4 software (Graphpad Software, Inc., San Diego, CA). Statistical analysis was performed by using a Student's t test with statistical significance set at p < 0.05.
Drugs and Chemicals. 5-Hydroxytryptamine (serotonin), aminobenzoic acid ethyl ester (tricaine), collagenase IA, and hexanol were purchased from Sigma-Aldrich (St. Louis, MO). Pentanol and octanol were purchased from Aldrich Chemical Co. (Milwaukee, WI), and butanol was purchased from Fluka Chemika (Buchs, Switzerland). Isoflurane was purchased from Baxter Healthcare Corp. (Deerfield, IL). Sevoflurane was purchased from Abbott Laboratories (Chicago, IL). Saturated solutions of volatile anesthetics were prepared by adding an excess of agent to a sealed bottle of recording solution and stirred overnight. These saturated solutions of known concentration were then diluted using gas-tight syringes to yield the final desired anesthetic concentration for experimentation. The anesthetizing concentrations of halogenated volatile anesthetics were defined as the aqueous concentrations corresponding to one minimal alveolar concentration (MAC) calculated using the aqueous/ gas partition coefficient at 37°C. Except for chloroform, MAC for all volatile anesthetics was in humans (Strum and Eger, 1987; Eger et al., 2003; Wadhwa et al., 2003). MAC for chloroform was taken as 0.5% the average value of two studies in mice (Miller et al., 1973; Deady et al., 1981). The anesthetizing concentrations of n-alcohols were defined as the aqueous concentrations that cause loss of righting reflex in tadpoles (Alifimoff et al., 1989).
Results
Application of 5-HT to oocytes injected with mRNA encoding for h5-HT3A and h5-HT3B subunits caused a concentration-dependent activation of inward current that was desensitized in the continued presence of a high concentration of 5-HT. Examining the 5-HT concentration-response of h5-HT3A and h5-HT3AB receptors expressed in oocytes demonstrated a difference between the two receptors. The measured EC50 values were 1.8 ± 0.05 and 20 ± 3 μM for h5-HT3A and h5-HT3AB receptors, respectively (Fig. 1A). In addition, heteromeric h5-HT3AB receptors had a lower Hill coefficient (1.2 ± 0.2) compared with homomeric h5-HT3A receptors (2.9 ± 0.4).
As previously observed (Davies et al., 1999; Dubin et al., 1999; Stewart et al., 2003), incorporation of h5-HT3B subunits into heteromeric h5-HT3 receptors resulted in 5-HT-evoked currents that were distinct from those mediated by homomeric h-5HT3A receptors (Fig. 1B). Serotonin-evoked currents from heteromeric h5-HT3AB receptors decayed faster relative to that observed with homomeric h5-HT3A receptors. The decay of current was examined by calculating the percentage of the maximal 5-HT-evoked current amplitude remaining after a 15-s application of 5-HT (Fig. 1C). No difference in maximal 5-HT-evoked current amplitude was observed between h5-HT3A receptors (100 μM) and h5-HT3AB receptors (300 μM). However, a significant difference in current decay at the end of a 15-s application of maximal 5-HT was observed between h5-HT3A receptors (43.6 ± 21.6%, n = 14) and h5-HT3AB receptors (93.5 ± 2.6%, n = 11, p < 0.0001).
Initial screening of volatile anesthetic and n-alcohol effects on heteromeric h5-HT3AB receptors was performed using a concentration of 2 μM 5-HT, which elicits approximately 10% of maximal current (EC10). The volatile anesthetics and n-alcohols were all used at equipotent concentrations, specifically, at twice their anesthetizing concentrations (see Materials and Methods). Anesthetic modulation of agonist-elicited currents varied with anesthetic molecular volume; a plot of the modulation by n-alcohols and halogenated volatile anesthetics of 2 μM 5-HT-evoked currents versus agent molecular volume shows a negative correlation, rs = -0.962 (Fig. 2). The physically smaller (molecular volumes <110 Å3) volatile anesthetics chloroform (1.7 mM) and halothane (0.43 mM) along with the n-alcohol butanol (21.6 mM) enhanced 2 μM 5-HT-elicited currents by 121 ± 50.7, 42.6 ± 4.5, and 48.6 ± 23.2%, respectively. However, the larger (molecular volumes >110 Å3) volatile anesthetic sevoflurane (0.66 mM) and n-alcohols hexanol (1.14 mM) and octanol (0.11 mM) inhibited currents by -49.6 ± 9, -37.8 ± 14.7, and -65.7 ± 8.1%, respectively. Amplitudes of 2 μM 5-HT-evoked currents were minimally affected by 5.8 mM pentanol (13.9 ± 21%) and 0.55 mM isoflurane (-10.3 ± 9.5%), an alcohol and anesthetic that are intermediate in molecular volume.
Anesthetic modulation of EC10 5-HT-evoked currents mediated by heteromeric h5-HT3AB receptors was compared with that mediated by homomeric h5-HT3A receptors (Stevens et al., 2005). Chloroform, halothane, butanol, and pentanol were significantly less effective at enhancing EC10 5-HT-evoked current amplitudes meditated by h5-HT3AB receptors compared with h5-HT3A receptors (Fig. 3). Isoflurane caused a small enhancement of h5-HT3A-mediated currents but slightly inhibited currents mediated by h5-HT3AB receptors. Although hexanol caused a significantly greater inhibition of currents recorded from h5-HT3AB compared with h5-HT3A receptors, inhibition by sevoflurane and octanol appeared identical for both receptors.
Agonist concentration-response relationships were constructed in the absence and presence of halogenated volatile anesthetics (chloroform, isoflurane, and sevoflurane) and n-alcohols (butanol and octanol) to resolve anesthetic actions on agonist EC50 and maximal agonist-elicited currents in oocytes expressing h5-HT3AB receptors. Each anesthetic and alcohol was applied at 1 and 2 times their anesthetizing concentrations. None of the halogenated volatile anesthetics tested had any significant effect on 5-HT EC50 values. Chloroform (0.85 and 1.7 mM) had no effect on maximally 5-HT-evoked currents and only slightly decreased 5-HT EC50 (Fig. 4A). At 2 MAC, chloroform (1.7 mM) decreased the EC50 by only 22.5% from 16 ± 0.9 μM to 12.4 ± 1.1 μM with no change in the Hill slope. Two MAC isoflurane (0.55 mM) increased the 5-HT EC50 by only 23.2% in addition to an inhibition of peak current amplitude (Fig. 4B). Sevoflurane caused a concentration-dependent decrease in maximal 5-HT-evoked current; 2 MAC sevoflurane (0.66 mM) decreased peak current by 73% while causing only a small (5.6%) increase in EC50 (Fig. 4C). All compounds tested had no effect on the Hill coefficient.
The actions of the n-alcohols butanol (11 and 22 mM) and octanol (0.11 and 0.22 mM) on 5-HT concentration-response relationships resembled those described above for the volatile anesthetics chloroform and sevoflurane, respectively. Butanol caused a reduction in the 5-HT EC50, reducing the control EC50 from 15 ± 0.7 to 13.1 ± 3.7 and 9.5 ± 0.8 μM (p < 0.05) for 11 and 22 mM butanol, respectively. Butanol (11 mM) showed negligible inhibition (2.5%) at high 5-HT concentrations, but at 22 mM, butanol caused an inhibition (16%) of maximal 5-HT-evoked current amplitude (Fig. 5A). The n-alcohol of larger molecular volume, octanol, inhibited currents evoked by most 5-HT concentrations but did not affect the 5-HT EC50. At 0.22 mM, octanol inhibited the maximal 5-HT-evoked current by 65%, yet EC50 remained unchanged, 13.4 ± 1.7 and 13.3 ± 0.4 μM for the 5-HT EC50 values in the absence and presence of octanol (Fig. 5B).
Figure 6 summarizes the data on EC50 values and Imax from our 5-HT concentration-response relationship experiments and compares them with what we have previously reported for the 5-HT3A receptor (Stevens et al., 2005). Both chloroform and butanol caused a smaller leftward shift in 5-HT EC50 with less inhibition of maximal evoked currents mediated by h5-HT3AB receptors compared with h5-HT3A receptors. The ability of octanol and sevoflurane to right-shift the 5-HT EC50 was greatly reduced by the incorporation of the 5-HT3B subunit. Sevoflurane was slightly better at blocking maximal-evoked current in h5-HT3AB receptors. Although isoflurane had no effects upon the 5-HT concentration-response relationship of h5-HT3A receptors, it inhibited 5-HT3AB-mediated current evoked by high concentrations of 5-HT. No volatile anesthetic or n-alcohol tested affected the Hill coefficients of any 5-HT concentration-response relationship.
Discussion
This study demonstrates that heteromeric h5-HT3AB receptors are modulated by halogenated volatile anesthetics and n-alcohols. Similar to what we have previously reported for homomeric h5-HT3A receptors (Stevens et al., 2005), molecular volume of n-alcohols and halogenated volatile anesthetics determines their ability to modulate h5-HT3AB-mediated current amplitude at submaximal concentrations of 5-HT. There exists a negative correlation between the molecular volume of compound and its ability to modulate submaximal current amplitude, such that all agents smaller than 110 Å3 (chloroform, halothane, butanol, and pentanol) enhance submaximal 5-HT-evoked current amplitude, and compounds larger than 110 Å3 (isoflurane, sevoflurane, hexanol, and octanol) inhibit submaximal 5-HT-evoked current amplitude.
We propose that allosteric interactions modulate anesthetic action on 5-HT3 receptors and suggest that volatile anesthetics and n-alcohols enhance submaximal 5-HT responses by binding to a small binding site that physically limits the binding of volatile anesthetics or n-alcohol having molecular volumes >110 Å3. The modulation by both halogenated volatile anesthetics and n-alcohols of similar molecular volumes suggests common site(s) of action. An additional larger binding site is also proposed. This site allows for the binding of compounds having molecular volumes in excess of 110 Å3 and mediates the inhibitory actions of the compounds. In contrast to the enhancing properties of the volatile anesthetics and n-alcohols, the incorporation of the 5-HT3B subunit does not appear to affect the inhibitory properties of the compounds. Similar n-alcohol enhancing and inhibitory sites have been shown for the nicotinic receptor. The enhancing site is similar in volume to the one described here for the 5-HT3 receptor. Competition studies between ethanol and octanol have shown the two sites to be separate (Wood et al., 1991).
The reduced enhancement of submaximal 5-HT-evoked currents and the reduced leftward shift in 5-HT EC50 values observed with the introduction of the h5-HT3B subunit can result from a number of possibilities. 1) The number of enhancing sites is reduced in the heteromeric h5-HT3AB receptor. 2) The number of enhancing sites remains the same, but the incorporation of the h5-HT3B subunit renders the receptor unable to undergo the same degree of conformational change induced by anesthetics as that of a homomeric receptor. 3) The h5-HT3AB receptor undergoes an increased desensitization rate in the presence of anesthetic, and hence, the enhancement is underestimated. As seen in Fig. 4A, current amplitudes induced by high concentrations of 5-HT are identical in the absence and presence of chloroform. If chloroform were increasing the rate of desensitization, the maximal current amplitude at high agonist concentrations would be expected to decrease because the rate of desensitization would exceed the solution exchange time. This would produce a left shift in EC50 and a shallow Hill coefficient, neither of which was observed. In addition, we have obtained analogous results using the whole-cell patch-clamp technique in combination with a rapid perfusion system that exchanges solutions in 1 to 3 ms. For h5-HT3AB receptors, the rate of desensitization is much slower than the rate of activation, and the two can clearly be distinguished using this method (data not shown).
Recent work on GABAA and glycine receptors has mapped out a potential binding cavity for n-alcohols and volatile anesthetics. Residues within transmembrane domains TM1, TM2, and TM3 have been shown to be critical for volatile anesthetic and n-alcohol enhancement of these receptors (Mihic et al., 1997; Jenkins et al., 2001). These residues are thought to form a hydrophobic pocket into which volatile anesthetics bind and is separate from the agonist binding site. Although the size of this anesthetic binding site is calculated to be larger than the pocket in 5-HT3 and nicotinic receptors, the location within the receptor is thought to be similar with all the members of the Cys-loop ligand-gated ion channel family.
Incorporation of the h5-HT3B subunit may decrease the number of anesthetic binding sites in the heteromeric receptor. As mentioned above, it is commonly thought that anesthetics and alcohols occupy hydrophobic clefts in ligand-gated ion channel protein structure, where they alter channel function by changing the channels' conformational flexibility. If such sites exist between adjacent h5-HT3A subunits, then incorporation of the h5-HT3B subunit would reduce the number of such sites. Alternatively, the binding site may be present within the h5-HT3A subunit itself and may be absent within the h5-HT3B subunit. A recent study described dramatic changes in the modulation of 5-HT-mediated currents by volatile anesthetics by mutating leucine (L)270, the 15′ residue within TM2 of the murine 5-HT3A receptor (Lopreato et al., 2003). The L270 residue is homologous to the glycine receptor α1 serine (S)267 and GABAA α1 S270 residues that are known to be important for the enhancement by volatile anesthetics and alcohols on those receptors (Mihic et al., 1997). Lopreato et al. (2003) proposed that L270 of the murine 5-HT3A subunits is important for the effects of anesthetics and alcohols and may even form the hydrophobic pocket that is the binding site. Interestingly, the 15′ residue in the TM2 of h5-HT3A is also a leucine, whereas the h5-HT3B subunits have an arginine. Further studies are needed to examine whether this change in residues is sufficient to change the pattern of channel modulation observed between homomeric and heteromeric receptors.
Although the modulation of 5-HT3 receptors by volatile anesthetics is probably not the main mechanism by which these compounds produce clinical anesthesia, the role played by 5-HT3 receptors in anesthesia and its side effects may not be completely insubstantial. There are many complicated components to the phenomenon of the anesthetized state (hypnosis, amnesia, immobility, and analgesia) occurring at supraspinal and spinal regions (Campagna et al., 2003). It is well documented that anesthetics enhance GABAA receptor mediated synaptic currents, thus increasing the inhibitory drive in neuronal networks. However, it has to be remembered that 5-HT3 receptors are located on some inhibitory GABAergic interneurons in the amygdala (Koyama et al., 2000), cortex (Zhou and Hablitz, 1999; Puig et al., 2004), hippocampus (McMahon and Kauer, 1997), and spinal cord (Alhaider et al., 1991; Tanimoto et al., 2004) and can control the release of GABA into the synapse, presumably through Ca2+-permeable homomeric 5-HT3A receptors (Koyama et al., 2000). Hence, anesthetic modulation of 5-HT3 receptors in these areas will affect GABA release and change the inhibitory drive. For example, anesthetics of low molecular volume can enhance presynaptic Ca2+-permeable homomeric 5-HT3A receptors resulting in a greater release of GABA into the synapse. Anesthetic stimulation of GABA release and potentiation of postsynaptic GABAA receptors provide complementary actions to enhance inhibitory drive.
Recently, a study showed that 5-HT3 receptor antagonists reduce the halothane-mediated inhibition of spinal dorsal horn sensory neuronal responses to noxious peripheral stimulation, indicating that 5-HT3 receptors are anesthetic targets for the reduction in nociception (Koshizaki et al., 2003). Small diameter (<25 μm) dorsal root ganglia neurons innervating the dorsal horn of the spinal cord mainly expressing 5-HT3A subunits are thought to be involved with the processing of nociceptive information, whereas coexpression of both 5-HT3A and 5-HT3B subunits are found in medium (26–40 μm) to large (>40 μm) dorsal root ganglia neurons and are believed to mediate proprioceptive as well as nociceptive information (Morales et al., 2001).
The major side effects of general anesthetic use include postoperative nausea and vomiting and cardiopulmonary depression. 5-HT3 receptors are known to regulate autonomic reflexes within the nucleus tractus solitarius (NTS). Such autonomic reflexes include emesis (Hornby, 2001), blood pressure, and heart rate (Comet et al., 2004). 5-HT3 receptors within the NTS are mainly presynaptic and therefore would influence release of neurotransmitter (Huang et al., 2004). In addition to central innervations, the NTS receives input from the peripheral nodose ganglia (Nosjean et al., 1990). Some nodose ganglia neurons innervating the NTS express 5-HT3A alone, whereas others express both 5-HT3A and 5-HT3B subunits (Morales and Wang, 2002). It remains to be seen whether anesthetic modulation of 5-HT3 receptors in the NTS causes any unwanted side effects of anesthetic administration. The expression of various 5-HT3 subunit combinations in different neurons innervating the spinal cord and NTS could result in a heterogenous response to anesthetics.
This is the first study on the modulation of human heteromeric 5-HT3AB receptors by halogenated volatile anesthetics and n-alcohols. Given the 5-HT3 receptors' role in neurotransmitter release, emesis, cardiovascular reflexes, nociception, and addiction, they constitute an important class of target proteins for general anesthetics and alcohols. A clear dependence upon molecular volume of whether a volatile anesthetic is a potentiator or inhibitor of EC10 5-HT-evoked current amplitudes is observed in both h5-HT3A and h5-HT3AB receptors. Current amplitude is only one part of a multifaceted waveform. In the presence of both agonist and anesthetic, the apparent rate of desensitization appears to increase with anesthetics of small, intermediate, and large molecular volumes. This suggests that the anesthetics may be having multiple effects on the kinetic gating process. Because of the slow solution exchange in the oocyte recording chamber (∼500 ms), we are unable to accurately measure activation, desensitization, or deactivation rates and therefore the effects of anesthetics upon them. Future experiments will need to be performed using rapid solution exchange to examine these important components.
Footnotes
-
Financial support was provided the Department of Anesthesia and Critical Care, Massachusetts General Hospital (to P.A.D.) and from National Institutes of General Medical Sciences Grant GM61927 (to D.E.R.).
-
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
-
doi:10.1124/jpet.105.085076.
-
ABBREVIATIONS: 5-HT, 5-hydroxytryptamine; h, human; MAC, minimal alveolar concentration; TM, transmembrane; NTS, nucleus tractus solitarius.
- Received February 16, 2005.
- Accepted April 6, 2005.
- The American Society for Pharmacology and Experimental Therapeutics