Regulatory Effect of General Anesthetics on Activity of Potassium Channels

Since the early 19th century, general anesthetics have been used to induce a state of unconsciousness for surgery. Modern anesthesiology defines a complete general anesthetic effect as including unconsciousness, amnesia, analgesia, and muscle relaxation that is indispensable for modern surgery. Hitherto, myriad molecular targets of general anesthetics have been identified, such as gamma-aminobutyric acid receptor, the N-methyl-D-aspartate receptor families, and ion channels [1‐6]. Among the many kinds of ion channels, potassium (K⁺) channels are the most diverse and ubiquitous, playing important roles in controlling neuronal excitability and neurotransmitter release in the central nervous system by determining the membrane potential of neurons [7‐12]. Thus, many studies have been conducted on the regulatory effects of general anesthetics on K⁺ channel activity. Also, data on these effects have accumulated, and this review summarizes recent research achievements in this area, especially emphasizing studies using K⁺ channel knockout (KO) mice. K⁺ channels that have been studied as general anesthetic targets have been assigned to 4 categories according to a standard nomenclature by The International Union of Basic and Clinical Pharmacology (IUPHAR) and the Human Genome Organisation Gene Nomenclature Committee (HGNC) [7]. They are voltage-gated (Kv) channels, the background/leak or tandem 2-pore (K2P) families, inwardly-rectifying (Kir) channels, and Ca²⁺-activated (K_Ca) channels. Research advances related to the effects of general anesthetics on these channels are summarized based on the above classification.

The opening of Kv channels is regulated by a membrane potential change and triggered by moving of the voltage sensor domain located on the S4 segment. The voltage-gated Kv channel family includes 40 genes encoding pore-forming subunits that are divided into 12 subfamilies (Kv1–Kv12) based on sequence homology (Fig. 1A–C). The HGNC and IUPHAR nomenclature and names based on the homologous channels in Drosophila are shown in Fig. 1A [13].

The two-pore-domain K⁺ (K2P) channel family is composed of fifteen KCNK genes, each of which comprises four transmembrane segments and two pore domains in tandem. Two pore-forming subunits form a functional channel as dimers. K2P channels conduct time- and voltage-independent background, or ‘leak’ currents to generate a negative membrane potential in excitable and non-excitable cells. K2P channels can be classified into six subgroups (TREK, TASK, TWIK, TALK, THIK, and TRESK) based on coding genes and biophysical properties (Fig.  2A). The nomenclature provided by HGNC and IUPHAR, and a popular naming scheme are all assigned sequentially in Fig. 2 [34]. Among them, the TASK and TREK-1 channels were the first to be identified and they are potentiated by general anesthetics. Chloroform, diethyl ether, halothane, and isoflurane potentiate the TREK-1 channel, whereas halothane and isoflurane activate the TASK channel (Table 1) [35]. The C-terminus of the TREK-1 and TASK channels is important for activation by general anesthetics [35]. Subsequent studies have revealed many members of this family to be sensitive to volatile anesthetics. Here, we give an introduction according to the subfamilies.

The mammalian Kir channel family includes 16 genes that encode an α subunit which has two transmembrane segments (M1 and M2) with a pore loop in between (Fig. 3A). Kir channels conduct K⁺ currents more readily in the inward direction than in the outward direction, and they are activated by ligand-stimulated G protein-coupled receptors. With activation of these receptors, G-protein βγ subunits are released from heterotrimeric receptors to interact with GIRK (G protein-coupled inwardly-rectifying K⁺) channels and open them. Among these genes, Kir3.x genes (GIRK1–4), such as GIRK1, GIRK2, and GIRK3 are predominantly expressed in neurons while GIRK4 is mainly expressed in the heart [10]. In neurons, GIRK channels either form GIRK2 homotetramers or form heterotetramers of GIRK1, GIRK2, and/or GIRK3 channels [62]. Tetramer channels with different subunits demonstrate different sensitivity to volatile anesthetics. Halothane enhances background currents through hetero-oligomeric GIRK1/GIRK4 channels but not through homo-oligomeric GIRK2 channels. This activation of basal current also requires the presence of G-protein βγ subunits. In contrast to basal GIRK currents, the agonist-induced GIRK currents (via co-expressed M2 muscarinic receptors) are inhibited by halothane (Table 1). In GIRK1/GIRK4 channels, this inhibition is most pronounced at low anesthetic concentrations (0.1 mmol/L–0.3 mmol/L) and occurs also when channels are activated by guanosine-5’-O-(3-thio) triphosphate. However, high concentrations of halothane (0.9 mmol/L) enhance the agonist-induced currents [63] (Table 1). This increase in agonist-induced currents is never seen with GIRK2 homo-oligomeric channels. Subsequent studies using GIRK1 channel (Del 363) and GIRK2 channel (Del 356) mutants that lack the C-terminal ends confirmed that these different modulatory effects produced by halothane are determined by the C terminus [64]. GIRK channels are resistant to intravenous anesthetics and nitrous oxide gas anesthetic (Table 3) [32]. On the other hand, efforts to investigate the roles of GIRK channels in general anesthesia in vivo using GIRK2-KO mice indicated that the MACs for 50% immobility in the tail clamp test for adult wild-type and GIRK2-KO mice are not significantly different [51] (Table 2). This result suggests that there is insufficient evidence to support the hypothesis that GIRK channels are involved in altering the arousal state in general anesthesia.

Among the Kir channels, Kir6.2 has also been named a KATP channel because it is inhibited by intracellular ATP and is sensitive to intracellular pH. It is composed of the pore-forming inward rectifier Kir6.2 and the regulatory sulfonylurea receptor SUR2A or SUR1. Although isoflurane does not directly interact with the Kir6.2 channel, it interacts with the SUR2A receptor to activate the Kir6.2 channel at low pH (pH 6.8). However, it has no activation effect at pH 7.4. Also, isoflurane cannot activate Kir6.2 channels co-expressed with SUR1 receptors (Table 1). Site-directed mutagenesis in the Walker motif of the SUR2A receptor abolishes the activation of Kir6.2 channels by isoflurane [65, 66]. Based on the recently dissected Kir6.2 channel structure, the nucleotide-binding domain on the SUR receptor is shown in Fig. 3B [67]. This study indicated that isoflurane preferentially enhances opening of the cardiac KATP channel (Kir6.2/SUR2A), suggesting a role of the KATP channel in cardiac protection in the anesthetic state induced by halogenated anesthetics.

The K_Ca family can be divided into three categories: BK (big conductance), IK (intermediate conductance), and SK (small conductance) channels [68]. BK channels are all tetrameric; they are formed by α subunits that have 6/7 transmembrane segments encoded by genes SLO1-3 [69]. But the Slo1 and Slo2 channels are expressed in brain, while the Slo3 channel is specifically expressed in testis and spermatozoa [70]. The Slo2 channel subfamily includes Slo2.1 and Slo2.2, which are activated by Na⁺ rather than Ca²⁺ [71]. The Slo1 channel is activated by both voltage and intracellular Ca²⁺ while the IK and SK channels are gated by intracellular Ca²⁺ alone. BK channels are diffusely expressed in almost the whole brain, whereas IK channels are mainly expressed in T lymphocytes, red blood cells, and parotid glands [72].

The regulatory effects of general anesthetics on BK, IK, and SK channels have been investigated. The volatile anesthetic halothane, at a clinically relevant dose of 0.5 mmol/L, reduces the open probability of BK channels without altering the single-channel conductance. However, this effect is blocked by increasing the concentration of cytoplasmic free Ca²⁺ from 1 µmol/L to 100 µmol/L, suggesting that halothane serves as a closed channel blocker [73]. In addition to halothane, isoflurane and enflurane also inhibit BK channels at a clinically relevant dose (Table 1) [74, 75]. Also, the intravenous anesthetic ketamine (at clinically relevant concentrations: 2–500 µmol/L) selectively blocks the BK channel in a dose-dependent manner. Ketamine shifts the open probability vs voltage curve to a higher potential without changing the slope of the voltage-dependence [76]. Interestingly, another intravenous anesthetic (propofol) activates the BK channel to relax coronary arteries [77]. A study using animal models showed that Slo-1-mutated Caenorhabditis elegans are resistant to volatile anesthetics [78]. But the effect of general anesthetics on Slo1-KO mice has not been reported.

IK channel currents are rapidly and reversibly inhibited by many volatile anesthetics such as halothane, isoflurane, enflurane, and sevoflurane with an EC₅₀ from 0.4 to 1 mmol/L, whereas the SK channel is not inhibited by volatile anesthetics [79, 80]. However, the SK channel is blocked by the intravenous anesthetics ketamine, barbital, and methohexital [80]. Experiments using patch clamp in brain slices have further confirmed that propofol enhances the inhibition of spike firing in retrotrapezoid nucleus neurons by blocking SK channels (Table 3) [81].

Although the effects of general anesthetics on Ca²⁺-activated K⁺ channels have been investigated, the underlying mechanism and the overall roles played by these channels in general anesthesia remain to be explored.

In summary, K⁺ channel activity is widely regulated by general anesthetics. Among them, the mechanisms by which Kv1.2, Kir6.2, and TASK-1 channels are modulated by general anesthetics have been studied in detail. Local structures and amino-acids critical for anesthetic binding to Kv1.2, Kir6.2, and TASK-1 channels revealed by biophysical studies have further clarified the characteristics of the molecular interactions of general anesthetics with these K⁺ channels. In addition, recent research has confirmed that activity changes in Kv1.x, TASK-1, TASK-3, and TREK-1 channels are essential for altering the arousal state when volatile anesthetics play their roles (Tables 2 and 4). In addition, the role of the THIK-1 channel in the retrotrapezoid nucleus is to maintain respiratory motor activity under general anesthesia. However, the biophysical mechanisms of the regulatory effects of general anesthetics on other K⁺ channels remain to be further explored. In the meantime, the roles of many K⁺ channels in general anesthesia remain elusive, especially the nuclei and neuronal circuits in which they play their roles. In vitro biophysical studies combined with optogenetic electrophysiological studies in vivo using K⁺ channel KO mice will greatly facilitate further addressing the role of K⁺ channels in general anesthesia.