“The future of general anesthetic pharmacology, ideally characterized by low toxicity and favorable pharmacokinetics, depends on identifying key targets that enable precise pharmacologic manipulation.”
SINCE the 1980s, molecular hypotheses attributing general anesthesia to nonspecific membrane biophysical perturbations have largely receded, while researchers have embraced the idea that anesthetic drugs act directly at protein targets, that is, classical receptors.1 However, there is little evidence supporting this hypothesis for volatile anesthetics. Pandit et al.2 in this volume of Anesthesiology add another piece to the challenging puzzle of volatile anesthetic mechanisms.
The new work by Pandit et al.2 relates clinically to a toxic effect of volatile anesthetics: inhibition of the hypoxic ventilatory response. Loss of this important physiologic reflex may contribute to postoperative hypoventilation, hypoxia, and associated morbidity. In the neural circuits underlying the hypoxic ventilatory response, glomus type I cells located in the carotid bodies and aortic bodies are the sensory cells, and their molecular chemosensing elements include TWIK-related acid-sensitive potassium channels (TASK; where TWIK is tandem pore-domain weakly inward-rectifying potassium-conductive), specifically heterodimers of TASK-1 and TASK-3 subunits.3 TASK channels are members of the two pore-domain potassium channel family and are partially active under normal physiologic conditions. They are directly inhibited by acute decreases in pH and indirectly by decreases in Po2, reducing potassium conductance and depolarizing glomus cells, leading to increased intracellular calcium, release of neurotransmitters, and ultimately stimulating both respiratory frequency and tidal volume. Pharmacologic TASK channel inhibitors, such as doxapram, also stimulate pulmonary ventilation.4 On the other hand, volatile anesthetics activate TASK channels, hyperpolarizing glomus cells and antagonizing the hypoxic ventilatory response. Volatile anesthetics also activate some other two pore-domain potassium channels, including TREK-1 (a TWIK-related potassium channel), which contributes to hypnotic and neuroprotective effects.5,6
Pandit et al.2 studied the effects of halothane, isoflurane, and mixtures of these two volatile anesthetics, using calcium-sensitive dye fluorescence to track intracellular [Ca2+] in glomus cells and voltage-clamp electrophysiology to measure potassium currents conducted by native TASK 1/3 channels in glomus cells and TASK-1 homodimeric channels expressed in HEK293 (kidney) cells. Halothane and isoflurane produced concentration-dependent activation of TASK channels and inhibition of hypoxia-induced intracellular [Ca2+] signals. The effects of halothane were consistently much larger than those of isoflurane. When halothane was coapplied with isoflurane in these experiments, the effects were subadditive. Notably, when a fixed halothane concentration was coapplied with increasing isoflurane concentrations, effects were smaller than those of halothane alone, increasingly approaching those of isoflurane alone.
Pandit et al.2 recognized that their results could reflect competitive interactions of two agonists, one high efficacy (halothane) and one low efficacy (isoflurane), both reversibly binding at a shared receptor site. Starting with halothane alone, as the coapplied concentration of isoflurane rises, it competitively replaces halothane at more binding sites, and a higher fraction of receptors display the lower activity characteristic of isoflurane agonism. To check this possibility, the research team applied quantitative analysis, using single drug affinity and efficacy parameters derived from fits to classical receptor models, and found that incorporating competitive binding to a single shared agonist site was consistent with the experimental results for drug combinations. This interpretation is parsimonious (alternative mechanisms would require more elements and more complex interactions between them) and very likely correct.
A growing body of evidence supports receptor-based mechanisms underlying sedation (inhibition of spontaneous activity) and hypnosis (inhibition of responses to stimuli) by intravenous drugs. Familiar examples include dexmedetomidine, a selective orthosteric agonist at α2 adrenergic receptors, and benzodiazepines, which allosterically enhance activation of certain types of γ-aminobutyric acid type A (GABAA) receptors. These two sedative-hypnotics are among the most potent in routine clinical use, acting at submicromolar concentrations. Recent research has also mapped several distinct sets of allosteric GABAA receptor sites that selectively bind intravenous hypnotic drugs, including etomidate, alphaxalone, benzyl alcohols, and certain barbiturates.7,8 These drugs all enhance GABAA receptor activation and in aquatic animals inhibit responses to stimuli at concentrations below 10 μM. Like other classic receptor-based drugs (e.g., opioids), these intravenous hypnotics demonstrate significant stereoselectivity (varying potency or efficacy) among enantiomers, supra-additive effects when combined with hypnotic drugs that act via distinct binding sites, and antagonism by low-efficacy ligands that occupy the same binding sites. These pharmacologic phenomena are demonstrable in both molecular and animal models, while transgenic animal models further link the effects of intravenous anesthetics to target receptors or subunits.
Contrasting with intravenous anesthetics, volatiles are among the oldest and most widely used, yet there is little evidence that specific receptors mediate their therapeutic actions. Volatile anesthetics produce hypnosis in animals at concentrations above 100 μM and display negligible stereoselectivity, and isobolographic analysis of combinations producing hypnotic endpoints consistently indicate additivity.9 While volatile anesthetics affect a wide variety of molecular targets hypothesized to mediate their major effects, no single dominant target has been identified for sedation, hypnosis, or immobility. Thus, while competitive antagonism is demonstrable for several intravenous anesthetics, it is not among the strategies being explored for actively reversing volatile anesthesia.
If volatile anesthetic agonism of TASK channels represents a case of specific receptors contributing to an important effect, previous research suggests a location for the responsible sites. Mutations in a specific region of TASK-1 and TASK-3 subunits ablate activation by volatile anesthetics. Covalent modification of an engineered cysteine in the putative site mimics the effect of volatile anesthetic exposure. Recent structural studies suggest that the putative anesthetic site forms an ion channel gate with a vestibule where antagonists bind.10 However, mutational and chemical modification results do not exclude allosteric effects at a volatile anesthetic binding site located elsewhere. Indeed, Pavel et al.11 reported recently that volatile anesthetic activation of TREK-1 depends on localization of phospholipase D2 to lipid rafts (microdomains), where this enzyme produces phosphatidic acid that agonizes TREK-1. They also showed that genetic inactivation of phospholipase D2 in fruit flies results in volatile anesthetic resistance. Whether such an indirect mechanism might also apply to TASK channels or can explain differential efficacy among volatile agents remains to be seen. Approaches such as photolabeling or cryoelectron microscopy may be useful in locating the relevant volatile anesthetic binding site(s).
There are broader scientific implications of Pandit et al.’s findings for the field of anesthetic mechanisms, in that receptor-mediated pharmacology may underlie other important actions of volatile anesthetics. For example, TASK channel activation may also contribute to organ-protective, arrhythmogenic, and inotropic effects of volatile anesthetics. While volatile anesthetics affect the function of many receptors, those that are associated with important clinical effects represent the most impactful opportunities for research.
The clinical and translational implications of this research deserve further exploration. The current report by Pandit et al.2 does not include animal studies comparing the effects on hypoxic pulmonary ventilation of various volatile anesthetics and mixtures. These crucial experiments are apparently underway, and the outcome may depend on the occupancy status of volatile anesthetic sites affecting TASK receptors, because low occupancy can bias results toward the appearance of additivity.12 TASK inhibitors seem unlikely to become staples in perioperative care, but might improve outcomes for patients at high risk of hypoventilation. Such patients might also be better served with intravenous anesthetics that do not activate TASK channels.
The future of general anesthetic pharmacology, ideally characterized by low toxicity and favorable pharmacokinetics, depends on identifying key targets that enable precise pharmacologic manipulation. A great deal more work is needed to realize that vision.
Supported by the National Institutes of Health (Bethesda, Maryland; grant Nos. R01-GM089745 and R01-GM128989) and the Department of Anesthesia, Critical Care and Pain Medicine, Massachusetts General Hospital (Boston, Massachusetts).
The author is not supported by, nor maintains any financial interest in, any commercial activity that may be associated with the topic of this article.