TREK-1 and TREK-2 Knockout Mice Are Not Resistant to Halothane or Isoflurane

Additional methods are found in Supplemental Digital Content 1 (https://links.lww.com/ALN/D129).

All animal experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health (Bethesda, Maryland) and were approved by the Institutional Animal Care and Use Committee of Seattle Children’s Research Institute (Seattle, Washington). The mice were housed at 22°C with a 12-h light–dark cycle and maintained on a standard rodent diet. Food and water were available ad libitum. Male and female mice were used for each experiment, and the total number of mice used are listed in the tables. Female and male mice were used in approximately equal numbers in all experiments.

When referring to the protein product of a gene, all letters are capitalized and not italicized (e.g., TREK-1). When referring to the gene or the mutant strain, the first letter is capitalized, and the term is italicized (e.g., Trek-1). In general, all mutants used are in a C57Bl/6 background. Since there were no sensitivity differences between the homozygote wild type and heterozygotes for either Trek-1, Trek-2, or Trek-1;Trek-2, heterozygotes are included in the wild-type designation in our tables.

The generation²¹  and functional^22,23,28  characteristics of the Ndufs4 strain have been previously described. Trek-1^tm1Lex was purchased from the Mutant Mouse Resource and Research Center at University of California Davis (mmrrc.ucdavis.edu). Trek-1^tm1Lzd is the allele studied by Heurteaux et al.¹³  and was carried in a previously characterized triple mutant, Trek-1^tm1Lzd,Trek-2,Traak.^29,30  It was the kind gift of Dr. Florian Lesage (French Institute of Health and Medical Research, Paris, France) and Dr. Andreas Schwingshackl (University of California Los Angeles, Los Angeles, California). In the previous version of this article, which used this strain, several reviewers requested that we isolate Trek-1^tm1Lzd from a potentially confounding genetic background. Therefore, Trek-1^tm1Lzd was isolated from the triple mutant by outcrossing, and its loss was confirmed by polymerase chain reaction. This isolated allele is the source for all data reported here as Trek-1^tm1Lzd.

Using CRISPR/CAS9 in a C57Bl/6 background, we contracted with Taconic-Cyagen (Leverkusen, Germany) to construct two alleles of Trek-2, each of which unequivocally produced knockout alleles and removed all of exon 2 (Fcc1 and Fcc2). The CRISPR guide sequences, which flanked exon 2, are available on request. The two new alleles were used for all studies reporting results for individual Trek-2 data. Genotypes of Trek-1, Trek-2, and Ndufs4 were confirmed by polymerase chain reaction in each mouse studied. The loss of the TREK-1 and TREK-2 proteins was confirmed by immunohistochemistry (see Results). The primer sequences are also available on request.

Anesthetic endpoints were loss of righting reflex or nonmovement during a noncrushing tail clamp using the methods described by Sonner et al.^31,32  Mice between postdelivery age 23 and 30 days (if containing Ndufs4) or between 23 and 30 or 57 and 70 days (if not containing Ndufs4) were anesthetized with either halothane or isoflurane, while their temperature was maintained by a water-filled heating pad. The anesthetic exposures did not affect survival, and all animals intended for study were successfully included. Temperature of the mice was monitored with a skin monitor and maintained between 37° and 38°C. The mice were exposed to both anesthetics, separated by 48 h between exposures; the order of anesthetic treatment was randomized. The concentrations of halothane and isoflurane were monitored using a calibrated inline AA-8000 anesthetic agent analyzer (BC Biomedical, Canada).

Responses to both the loss of righting reflex and tail clamp assays were measured at each dose of anesthetic after 10 min of equilibration, beginning with a concentration of 0.6% for each anesthetic (or 0.2% for Ndufs4-containing mice) and increasing in steps of 0.2%. Once the endpoint was reached, concentrations were decreased to determine when the animals regained their response. MAC was calculated as described by Sonner et al.³¹ 

The exceptions were the Ndufs4-containing animals, which were anesthetized for both halothane and isoflurane for two behavioral endpoints. In Ndufs4 in pilot studies, we observed that this battery of two long repeated exposures of one individual animal introduced confounding morbidity and mortality, something not observed in our previous paradigm of a single anesthetic exposure for one endpoint in one individual.^22,33  In addition, the rare births of the double mutants (Ndufs4;Trek-1) necessitated modifying the protocol to maximize data obtained from any one individual. Therefore, in this study of Ndufs4 and Ndufs4;Trek-1 animals, only the induction concentrations were determined, not the emergence values. Induction is defined as the midpoint between the last concentration at which the behavior was present and the first at which it is lost. Since this method is not the standard for determining MAC, we termed this endpoint the EC50 for induction. Using this method for Ndufs4, all animals survived, and those animals intended for study were all successfully included.

For a subset of experiments the anesthetic sensitivity was determined using the method described in the report of resistance of Trek-1 to volatile anesthetics.¹³  Briefly, this protocol rapidly increased anesthetic concentrations using clinical overpressure techniques to quickly induce anesthesia in 57- to 70-days-postdelivery mice. Then the anesthetic was slowly decreased to determine the emergence concentration (the midpoint between the last concentration at which the behavior was absent and the first at which it is regained) at which the animal responded to a tail clamp. This concentration was reported as the MAC in the original publication. We report it as the emergence concentration in these studies.

Animal Research: Reporting of In Vivo Experiments (ARRIVE) Guidelines are listed in the Supplemental Digital Content 1 (https://links.lww.com/ALN/D129).

In all cases, cells of the lateral ventral horn were patched. Ventral horn spinal cord cells were visualized using differential interference contrast microscopy. The cells could be identified as cholinergic or noncholinergic in some genotypes (Ndufs4 and Ndufs4;Trek-1) that incorporated a fluorescent tag (Ai14) in cholinergic neurons as described previously.²³  This signal could be seen with fluorescence microscopy and was used in Ndufs4-containing slices to ensure use of noncholinergic neurons. This selection was used since previous studies showed that the increased holding currents at 0.6% isoflurane in Ndufs4 slices were seen only in noncholinergic cells. Increased currents at 1.8% isoflurane in other genotypes were seen in all neuronal types of the spinal cord.

Slices of lumbar spinal cord were isolated and studied as previously described.²³  They were first held in the superfusate for 30 min without isoflurane for baseline, unexposed measurements. Isoflurane was then applied in the superfusate at equilibrated concentrations delivered by passing carbogen (a mixture of 95% O₂ and 5% CO₂) through a calibrated isoflurane vaporizer. The superfusate was sampled during isoflurane exposure, and the isoflurane concentration was determined using gas chromatography. To rule out “run down” of the preparation, recordings were also made with no isoflurane exposure for the same period of time that matched the sum of experimental exposure and wash.

Norfluoxetine (hydrochloride; Cayman Chemical catalog no. 15900) was diluted to a final concentration of 20 µM in recording solutions from a stock solution of 10 mM in dimethyl sulfoxide (DMSO; Sigma-Aldrich, USA), based on the previously published IC₅₀ = 9 μM for TREK blockade.¹⁵  Norfluoxetine was added to the bath following a previously published protocol that compared the effects of isoflurane with and without norfluoxetine.²¹  In short, this protocol recorded at 0% isoflurane and then isoflurane exposure, followed by its washout, and then application of norfluoxetine for 15 min, followed by resumption of isoflurane exposure, now in the presence of norfluoxetine. The exception to this protocol in this study is in the Trek-1;Trek-2 double mutant. These cells were very difficult to maintain in a patched status for a long period of time. Thus, after baseline measurements, Trek-1;Trek-2 slices were exposed to isoflurane, and after 15 min, norfluoxetine was added, and currents were measured in the presence of both agents for 10 min. Trek-2 slices were repeated under this protocol to ensure that the two techniques gave comparable results.

Since heterozygotes for the Trek channels (e.g. Trek-1^+/- and Trek-2^+/-) behave like wild type in isoflurane and halothane, in this report we use the term “wild type” to denote both C57Bl/6 (Trek^+/+ or Trek^+/-) mice. These heterozygotes are included in the wild-type designation in our tables. The term “mutant” refers to Ndufs4(knockout)-, Trek-1(knockout)-, or Trek-2(knockout)-containing mice. All induction, emergence, and MAC values for behavioral testing are expressed as the mean anesthetic concentration with SD in parentheses (mean and SD followed by N in the tables) for the number of animals studied and the corresponding P value. Effect sizes were calculated by dividing the mean of the Trek mutant to the mean of the control using same endpoint and anesthetic and are listed in the tables. In the graphs, error bars refer to 95% CI. Since including all of these values in the prose made the article difficult to read, we included four tables with the values and referenced those tables when appropriate.

We used the percentage of anesthetic for comparison to previous reports; however, we also included temperature- corrected aqueous concentrations of isoflurane for comparison in holding current results. For normally distributed data, we used a paired, one-tailed t test when comparing two groups and a one-way ANOVA with Dunnet’s multiple comparisons test for greater than two groups. P values were determined in Excel or in Prism and calculated for each paired group individually. For non-normally distributed data, we used the Kolmogorov–Smironov test when comparing two groups and the Kruskal–Wallis test with Dunn’s multiple comparison test for greater than two groups.

The P values refer to mutant values compared to CB57Bl/6 values determined in paired experiments (control and mutant done side by side) unless explicitly stated otherwise. In general, Ndufs4;Trek-1 double mutants were limited in number such that having matched side-by-side measurements was not possible. In those cases, comparisons were made between the double mutant and Ndufs4 groups but not as matched pairs. In addition, the EC50 values for the wild-type and Ndufs4 strains listed in table 1 are the cumulative means for those two strains from all experiments. We grouped all values from each of the paired tests, which led to a larger N than for any individual comparison. The significance level was selected as a P value less than 0.05. Post hoc values were subjected to a Bonferroni correction for multiple comparisons to adjust significance for multiple comparisons. For comparisons to C57Bl/6, there were six groups, leading to a corrected P value for significance of 0.01. For comparisons to Ndufs4, there were four groups, leading to a corrected P value for significance of 0.017.

In general, we measured one cell per spinal cord slice and one slice per animal. Thus, our N refers to animals, slices, and cells. For holding current studies, means and 95% CI are given in the Results and shown by box plots in the figures. The P values for comparisons of changes in holding currents are noted in the text and in table 4. Further statistical considerations are listed in the supplemental materials (https://links.lww.com/ALN/D129). For measurements of EC50 values, we defined a change of >10% as an effect size of biologic significance. We used a difference of 10% from the EC50 values for wild-type mice, a SD of 0.1, an α of 0.05, and a desired power of 0.8 to determine adequate sample size (generally 7). In most cases, our sample sizes exceeded these values.

As controls for the effects of TREK-1 loss on Ndufs4, we first measured both induction and emergence in the presence of volatile anesthetics in Trek-1 23- to 30-days-postdelivery mice in a wild-type background for two knockout alleles: tm1Lex³⁴  and tm1Lzd.¹³  In response to a nondamaging tail clamp, we did not observe significant changes in either induction or emergence concentrations of isoflurane or halothane when comparing either allele of Trek-1 mice to wild-type controls (fig. 1, A and B; tables 1 and 2). MACs calculated from averaging the induction and emergence concentrations closely matched those previously published for wild-type mice³¹ : for wild type, MAC (halothane), 1.30% (0.10); MAC (isoflurane), 1.40% (0.11); for Trek-1^tm1Lex, MAC (halothane), 1.27% (0.11; P = 0.387); MAC (isoflurane), 1.38% (0.09; P = 0.268); for Trek-1^tm1Lzd, MAC (halothane); 1.27% (0.11; P = 0.482); MAC (isoflurane), 1.41% (0.12; P = 0.188). The P values compare mutant values to paired wild-type values. Similarly, we also did not detect significant differences between the TREK-1 mutant strains and controls for loss of righting reflex in either anesthetic (tables 1 and 2).

The original reported resistance of Trek-1^tm1Lzd to volatile anesthetics was based on determining an average anesthetic concentration at which emergence occurred in 57- to 70-days-postdelivery mice after a rapid induction.¹³  We determined whether mice anesthetized under that protocol, which measured emergence as anesthetic concentrations were decreased, might display a resistance in Trek-1^tm1Lex or Trek-1^tm1Lzd compared to wild type. Since the differences reported between wild type and Trek-1^tm1Lzd were greatest in halothane, we specifically tested response to tail clamp in halothane, trying to exactly match the published protocol. We were unable to find any difference in sensitivity between the three genotypes using this emergence protocol (fig. 1C; table 3): for wild type, MAC (halothane), 1.23% (0.18); for Trek-1^tm1Lex, MAC (halothane), 1.28% (0.09; P = 0.582); for Trek-1^tm1Lzd, MAC (halothane), 1.25% (0.09; P = 0.943). Of note, the MACs for the Trek-1^tm1Lex and Trek-1^tm1Lzd strains were similar to that earlier reported for Trek-1^tm1Lzd.¹³  However, in that report of halothane resistance for Trek-1, the wild-type MAC was low (0.87%)¹³  compared to values in the literature.^31,32  In our measurements, the wild-type MAC was not shifted from other reports nor from that of either Trek-1 allele.

The reported resistance of Trek-1^tm1Lzd mice also resulted from studies done in mice older than those used for our determinations of MACs.¹³  We therefore also applied the standard MAC determination³¹  to mice of postdelivery age of 57 to 70 days for wild type and Trek-1^tm1Lex. When we repeated the studies in wild-type and Trek-1^tm1Lex mouse strains at the older age, we again found no difference between the mutant and wild type in either isoflurane (fig. 1D; table 3) or halothane (fig. 1E; table 3): for wild type, MAC (halothane), 1.23% (0.07); MAC (isoflurane), 1.22% (0.13); for Trek-1^tm1Lex, MAC (halothane), 1.31% (0.16; P = 0.401); MAC (isoflurane), 1.18% (0.12; P = 0.492). The P values compare mutant values to paired wild-type values.

It remained unclear to what extent the increased holding current seen in spinal neurons from Ndufs4 mice²³  at low doses of isoflurane contributed to the profound hypersensitivity of the animal to isoflurane and halothane. We generated Ndufs4;Trek-1^tm1Lex and Ndufs4;Trek-1^tm1Lzd mice to test whether loss of TREK-1 expression would attenuate the profound anesthetic hypersensitivity of Ndufs4 mice using two behavioral endpoints. As noted in Materials and Methods, for technical reasons, we used only the induction concentrations to determine anesthetic sensitivities of these double mutants; we termed these results EC50 instead of MAC for that reason. There were no differences between Ndufs4 mice and Ndufs4;Trek-1^tm1Lex or Ndufs4;Trek-1^tm1Lzd mice in the concentrations of isoflurane or halothane necessary to induce inhibition of the tail clamp response (fig. 2A; tables 1 and 2): for Ndufs4, EC50 (halothane), 0.65% (0.05); EC50 (isoflurane), 0.63% (0.05); for Ndufs4;Trek-1^tm1Lex EC50 (halothane), 0.58% (0.07; P = 0.004); EC50 (isoflurane), 0.61% (0.06; P = 0.442); for Ndufs4;Trek-1^tm1Lzd, EC50 (halothane), 0.63% (0.09; P = 0.529); EC50 (isoflurane), 0.64% (0.07; P = 0.394). The P values report double mutant values compared to Ndufs4 values. We did measure a small but significant difference in the concentration of halothane required to inhibit the tail clamp response. However, Ndufs4;Trek-1^tm1Lex mice exhibited an increase in sensitivity to halothane compared to Ndufs4 mice (0.58% vs. 0.65%) rather than a resistance as hypothesized (fig. 2A; tables 1 and 2). We did not observe a difference between Ndufs4 mice and Ndufs4;Trek-1^tm1Lex or Ndufs4;Trek-1^tm1Lzd mice in the loss of righting reflex response when exposed to isoflurane or halothane (fig. 2B; tables 1 and 2).

Trek-1 animals are not resistant to anesthetics. However, TREK-2 is of interest for its potential role in anesthetic behavior in wild-type animals as a cause for the isoflurane-induced, norfluoxetine-inhibitable increase in holding current. We measured both induction and emergence in the presence of volatile anesthetics in Trek-2 mice for two knockout alleles: Fcc1 and Fcc2. In response to tail clamp, we did not observe significant changes in the EC50 values of isoflurane or halothane when comparing either allele of Trek-2 mice to wild-type controls (fig. 1, A and B; tables 1 and 2). MACs closely matched those previously published for wild-type mice and our values reported above³¹ : for Trek-2^(Fcc1), MAC (halothane), 1.29% (0.14; P = 0.073); MAC (isoflurane), 1.41% (0.09); P = 0.412); for Trek-2^(Fcc2), MAC (halothane), 1.36% (0.12; P = 0.330); MAC (isoflurane), 1.36% (0.09; P = 0.086). The P values refer to mutant values compared to paired wild-type values. In the two cases in which there was a trend toward significance, the Trek mutants tended to increased sensitivity, not resistance. Similarly, we did not detect any significant differences between the Trek-2 mutant strains and controls for loss of righting reflex (tables 1 and 2).

In addition, and in response to the editor’s request, although our study was not specifically designed to detect sex-specific differences in anesthetic sensitivity, post hoc evaluation of our data does not suggest differences between male and female animals in terms of EC50 values to either halothane or isoflurane. While we detected statistically significant differences in mean concentrations for Trek-1 male and female responses to isoflurane for loss of righting reflex (MAC [isoflurane], for males, 0.73 [0.04]; for females, 0.82 [0.10]; P = 0.004), these small differences pose little functional significance and did not warrant further investigation (Supplemental Digital Content 2, table S1, https://links.lww.com/ALN/D130).

We finally considered whether the loss of both TREK-1 and TREK-2 channels might affect anesthetic sensitivity of the wild-type animal. We constructed the double mutant Trek-1^tm1Lex;Trek-2^Fcc1 and measured the MACs for halothane and isoflurane. In response to the tail clamp, we did not observe significant changes in the EC50 values of isoflurane or halothane when comparing Trek-1;Trek-2 mice to controls (fig. 1, A and B; tables 1 and 2). MACs calculated from averaging the induction and emergence concentrations closely matched those previously published for wild-type mice and our values reported above³¹ : for Trek-1^tm1Lex;Trek-2^Fcc1, MAC (halothane), 1.26% (0.14; P = 0.661); MAC (isoflurane), 1.46% (0.06; P = 0.373). The P values refer to double mutant values compared to paired control values. Similarly, we did not detect any significant differences between the double mutant strains and controls for loss of righting reflex (tables 1 and 2).

Regarding holding currents, our interest in the TREK-1 channel was initially the result of a norfluoxetine-inhibitable increase in leak current in noncholinergic neurons of the ventral spinal cord of Ndufs4²³  upon exposure to low concentrations of isoflurane, coupled with a report that a Trek-1 mouse was resistant to anesthetics.¹³  Since norfluoxetine blocks both TREK-1 and TREK-2 channels,^15,16  we considered whether TREK-1 or TREK-2 might be up-regulated in Ndufs4 mice. After verifying that our alleles of Trek-1 and Trek-2 did not produce a protein product (fig. 3, A and B), we found that there was no increase in TREK-1 or TREK-2 staining in Ndufs4 spinal cords (fig. 3, C and D) compared to those from wild-type animals.

While norfluoxetine affects other targets,^35–37  the isoflurane-induced currents were blocked by norfluoxetine. This indicates that the increases in holding current upon anesthetic exposure, indicative of hyperpolarization of the cell, are most likely carried by TREK channels. Blocking TREK channels therefore eliminates the rise in holding current upon isoflurane exposure. We determined whether the isoflurane-induced holding current was still present in Ndufs4;Trek-1^tm1Lex and Trek-1^tm1Lex mice and whether an isoflurane-induced increase in these animals would be sensitive to norfluoxetine. Baseline spinal cord holding currents did not differ significantly between wild-type and mutant strains (table 4).

In agreement with our previous results, holding currents increased in Ndufs4 slices at 0.6% isoflurane but not in wild-type slices (Supplemental Digital Content 3, fig. S1, https://links.lww.com/ALN/D131).²³  Holding currents also increased in Ndufs4;Trek-1^tm1Lex exposed to 0.6% isoflurane (~0.25 mM; isofluranev 0.6%; change in holding current, 110.9 [61.45 to 169.18]; P = 0.015 compared to no isoflurane; fig. 4A). Norfluoxetine approached significance in blocking the rise in holding current in Ndufs4;Trek-1^tm1Lex (change in holding current, 42.3 [23.1 to 61.7]; P = 0.077 norfluoxetine plus isoflurane compared to isoflurane; table 4). Representative curves are shown for all comparisons in figure 4 (A to D).

In slices of Trek-1^tm1Lex, Trek-2^Fcc1, and Trek-1^tm1Lex;Trek-2^Fcc1 slices, isoflurane (0.74 mM) increased holding currents, just as it did in wild-type slices (fig, 4, B to D). However, norfluoxetine did not block the increase seen in any of the mutants compared to the increases at 1.8% isoflurane in the absence of norfluoxetine (fig, 4, B to D; table 4). These data implicate recruitment of norfluoxetine-insensitive channels to increase holding currents caused by isoflurane in spinal cord neurons.

In mice, loss of TREK-1 or TREK-2 channels individually or as a double mutant has no significant impact on the tail clamp response to isoflurane or halothane. Neither induction nor emergence values separately were significantly different in Trek(KO) mice compared to wild-type controls, and there were no significant differences in MAC values. The same was true for the loss of righting reflex; loss of TREK channels did not affect that endpoint. Furthermore, we failed to replicate the resistance seen during emergence from volatile anesthetics observed in Trek-1 mice following a nonstandard induction protocol.¹³  However, the resistance to halothane reported previously for Trek-1^tm1Lzd appears to be based on a low MAC value in control animals compared to that customarily found in the literature.³¹  Since those data have been frequently cited as evidence for the K2P channels as anesthetic targets, it is important to note that we were unable to repeat these findings of anesthetic resistance.

Interest in the TREK channels and behavior in anesthetic stems from initial work in the marine mollusk, Aplysia californica. It was discovered there that volatile anesthetics activate an outwardly rectifying potassium channel, the S channel.³⁸  Mammalian TREK-1 and TREK-2 are proposed orthologs of the Aplysia S channel; they share many biophysical and pharmacologic properties.^39–41  Their activation hyperpolarizes neurons causing a predicted decrease in activity. In addition, TREK-1 and TREK-2 are activated by several volatile anesthetics including isoflurane,^25,26  and both channels are widely expressed in the mammalian spinal cord.⁴²  Recent data have also shown the important sites in TREK-1 for transducing the effect of isoflurane on the channel to increase conductance.²⁷ 

It is well established that volatile anesthetics can hyperpolarize neurons,^43,44  and our data corroborate those findings in the spinal cords of adult mice.²³  These data thus fit well with the previous report of resistance to volatile anesthetics of Trek-1^tm1Lzd mice.¹³  Cells in the spinal cord expressing TREKs would be predicted to become hyperpolarized, and therefore less active, with exposure to volatile anesthetics. We therefore expected that this hyperpolarization, induced by volatile anesthetic concentrations that correlated to whole animal phenotypes, would contribute to the anesthetic response in both wild type and the mitochondrial mutant Ndufs4. A role for TREK-1 channels in the hypersensitivity of Ndufs4 would have indicated a potential unifying model for anesthetic mechanisms involving those channels.

However, loss of either or both TREK channels did not change the behavior of wild type or Ndufs4 animals in isoflurane or halothane. In addition, an isoflurane-inducible rise in holding current remained present, with TREK loss in both wild-type and Ndufs4 animals. However, removal of TREK channels in wild-type animals caused the isoflurane-inducible increase in holding currents to become norfluoxetine insensitive. Other (norfluoxetine-insensitive) potassium channels, such as the TASK channels, have also been shown to be induced by volatile anesthetics and may also affect volatile anesthetic sensitivity.^45,46  We have not ruled out compensatory changes in TASK channels, or other possible effectors, when TREK channels are removed; in fact, our data indicate that they likely exist. They may be recruited to conduct the current seen when either TREK is removed. However, the blockade of the holding current in the wild type by norfluoxetine would appear to rule out the TASK family of K2P channels for the potassium current seen in “normal,” wild-type slices.⁴⁷ 

The loss of TREK-1 also failed to inhibit the increase in holding current of Ndufs4 neurons to 0.6% isoflurane. However, the interpretation of the effect of norfluoxetine on the rise is less clear in Ndufs4;Trek-1^tm1Lex (fig. 4B). It may be that there is a partial block, but clarification awaits further studies. At this time, we hypothesize that if our data represent a partial block, then the hyperpolarizing current is a mix of both norfluoxetine-sensitive and -insensitive channels. Construction of a Trek-1;Trek-2;Ndufs4 triple knockout is underway to clarify our data. It is noteworthy that since TASK channels affect current in cholinergic neurons, we did not see any rise in holding currents when we measured signal from cholinergic cells of the spinal cord in Ndufs4.²³  This makes a role for TASK channels unlikely in mediating an isoflurane-induced hyperpolarizing current in Ndufs4.

Most importantly, the whole animal behavior of two different knockout alleles of both Trek-1 and Trek-2, as well as in the Trek-1;Trek-2 double mutant, did not reveal any resistance to either of two different volatile anesthetics for two different endpoints. The original report of resistance to volatile anesthetics in a Trek-1^tm1Lzd animal was performed nearly two decades ago.¹³  It is possible that the background genetics of that animal were different than the current C57Bl/6 or that genetic drift has occurred over time, causing a novel compensatory change and a return to wild-type sensitivity. Regardless, loss of these particular K2P channels does not change anesthetic resistance in our hands. The interpretation of the increase in holding current seen with the application of isoflurane to spinal cord slices also needs re-evaluation. The electrophysiologic phenomenon of isoflurane-induced increased potassium currents is maintained even in the absence of TREK channels and may contribute to the action of volatile anesthetics.

The authors are indebted to Florian Lesage, Ph.D. (Université Côte d’Azur, Nice, France), and Andreas Schwingshackl, Ph.D. (UCLA, Los Angeles, California), for sharing the Trek-1^tm1Lzd,Trek-2(KO),Traak(KO) mice. The authors thank Beatrice Predoi, M.D., Ph.D., and Miranda Howe for their excellent technical assistance and Julia Stokes and Angelina Zimenko for technical support. They also thank Sangwook Jung, Ph.D., for his insightful discussions and reading of the manuscript.

Supported in part by National Institutes of Health (Bethesda, Maryland) grant No. R01GM105696 and by continued support from the Northwest Mitochondrial Research Guild (to Drs. Sedensky, Ramirez, Johnson, Spencer, and Woods) and by National Institutes of Health grant Nos. T32 GM086270 (to Dr. Spencer) and R35GM139566 (to Drs. Morgan and Woods).

The authors declare no competing interests.

Supplemental Digital Content 1. Additional Methods and ARRIVE details, https://links.lww.com/ALN/D129

Supplemental Digital Content 2. Table S1. Effect of Gender on MAC, https://links.lww.com/ALN/D130

Supplemental Digital Content 3. Figure S1. Effect of Isoflurane on holding currents in Ndufs4 mice, https://links.lww.com/ALN/D131