Secondary Logo

Journal Logo

Isoflurane Selectively Inhibits Distal Mitochondrial Complex I in Caenorhabditis Elegans

Kayser, Ernst-Bernhard, PhD*; Suthammarak, Wichit, PhD*; Morgan, Phil G., MD*,†; Sedensky, Margaret M., MD*,†

doi: 10.1213/ANE.0b013e3182121d37
Anesthetic Pharmacology: Research Reports

BACKGROUND: Complex I of the electron transport chain (ETC) is a possible target of volatile anesthetics (VAs). Complex I enzymatic activities are inhibited by VAs, and dysfunction of complex I can lead to hypersensitivity to VAs in worms and in people. Mutant analysis in Caenorhabditis (C.) elegans suggests that VAs may specifically interfere with complex I function at the binding site for its substrate ubiquinone. We hypothesized that isoflurane inhibits electron transport by competing with ubiquinone for binding to complex I.

METHODS: Wildtype and mutant C. elegans were used to study the effects of isoflurane on isolated mitochondria. Enzymatic activities of the ETC were assayed and dose–response curves determined using established techniques. Two-dimensional native gels of mitochondrial proteins were performed after exposure of mitochondria to isoflurane.

RESULTS: Complex I is the most sensitive component of the ETC to isoflurane inhibition; however, the proximal portion of complex I (the flavoprotein) is relatively insensitive to isoflurane. Isoflurane and quinone do not compete for a common binding site on complex I. The absolute rate of complex I enzymatic activity in vitro does not predict immobilization of the animal by isoflurane. Isoflurane had no measurable effect on stability of mitochondrial supercomplexes. Reduction of ubiquinone by complex I displayed positive cooperative kinetics not disrupted by isoflurane.

CONCLUSIONS: Isoflurane directly inhibits complex I at a site distal to the flavoprotein subcomplex. However, we have excluded our original hypothesis that isoflurane and ubiquinone compete for a common hydrophobic binding site on complex I. In addition, immobilization of the nematode by isoflurane is not due to limiting absolute amounts of complex I electron transport as measured in isolated mitochondria.

Published ahead of print April 5, 2011

From the *Center for Developmental Therapeutics, Seattle Children's Research Institute, Seattle, Washington; and Department of Anesthesiology, University of Washington, Seattle, Washington.

Funding: EBK, WS, PGM, and MMS were supported by National Institutes of Health grant #GM58881 and ARRA Supplement 3R01GM58881-12S1.

The authors declare no conflict of interest.

Reprints will not be available from the authors.

Address correspondence to Phil G. Morgan, MD, Department of Anesthesiology, University of Washington and Center for Developmental Therapeutics, Seattle Children's Research Institute, 1900 Ninth Avenue, Seattle, WA 98101-1304. Address e-mail to

Accepted January 18, 2011

Published ahead of print April 5, 2011

The nematode Caenorhabditis (C.) elegans has been established as a tractable genetic model for studying anesthetic action.1,2 Although reversible immobility produced by volatile anesthetics (VAs) occurs at concentrations higher than minimum alveolar concentration in humans, this end point obeys the Meyer–Overton rule, exhibits stereoselectivity, is not affected by nonimmobilizers, and fits well with the LC50/EC50 ratio seen in mammals for VAs.3 Most important, the molecular mechanisms that determine this end point control the anesthetic behavior of other organisms, including humans.46

Data from both humans and C. elegans indicate that mitochondrial physiology plays a role in controlling anesthetic sensitivity. The electron transport chain (ETC) is a collection of interacting protein complexes and mobile electron carriers that resides within the inner mitochondrial membrane.79 In C. elegans, animals that are defective in respiration through complex I of the ETC are hypersensitive to VAs.4,6,10 The same is true for children who present for a muscle biopsy to diagnose mitochondrial dysfunction.5

Complex I is the point of entry for electrons from mitochondrial nicotinamide adenine dinucleotide (NADH) into the respiratory chain in the inner mitochondrial membrane (for excellent reviews, see references 9,11,12). Complex I transfers electrons from NADH, through a series of iron sulfur clusters, finally to the hydrogen carrier ubiquinone (UQ) (coenzyme Q, CoQ).7,11 This molecule consists of a hydrophobic tail embedded within the inner mitochondrial membrane, and a more polar aromatic headgroup that is the redox active moiety. Driven by the free enthalpy of the electron transfer, complex I pumps protons from the matrix to the intermembrane space. This contributes to the proton motive force that energizes the inner mitochondrial membrane, and is crucial for the aerobic regeneration of adenosine triphosphate (ATP). Complex I is an enormous complex consisting of at least 45 distinct subunits (in eukaryotes). Most of these subunits are encoded by nuclear genes and have to be imported into the mitochondrion. Only the 7 most hydrophobic subunits are encoded by mitochondrial DNA and expressed inside the mitochondria. Insight into subunit composition, structure, and function has been compiled from studies of a wide variety of biological sources for both the eukaryotic and prokaryotic forms of the enzyme.9,11,12 Both forms are roughly L shaped with a matrix arm and a membrane arm. It is important, however, to appreciate that the bacterial complex I is composed of only 14 subunits while fulfilling the same basic function as the eukaryotic form. All of the prokaryotic subunits have clear orthologs in the eukaryotic enzyme. These are referred to as the core subunits. Subunits that do not have counterparts in bacteria are often called supernumerary, reflecting our present ignorance about their possible functions. The recent progress in crystallizing bacterial complex I allowed development of a detailed spatial model for the prokaryotic complex and by analogy for the arrangement of the core subunits of the eukaryotic form.9

In worms, one of the mutants most sensitive to VAs, gas-1(fc21), cannot readily transport electrons from complex I to complex III of the ETC.7 The mutation alters an evolutionarily strictly conserved arginine (R290K) in a core subunit of complex I.13 This subunit of complex 1 termed GAS-1 (orthologous to 49kD, Nuo4, NDUFS2),14 together with another core subunit PSST, holds the terminal iron sulfur cluster N2 and forms part of the binding site for the headgroup of UQ (Fig. 1A).7,9,11 Crystallographic data from the bacterial complex I showed that the cavity believed to bind the UQ headgroup is in part formed by a 4-α helical bundle of the GAS-1 ortholog.9 Amphiphilic peptides with 4 helical bundles have been extensively used to model the binding of halothane to membrane proteins.15,16 Halothane binds to these synthetic peptides with an affinity approximately within physiologic ranges and had dose-dependent effects on the structure of these synthetic molecules.

Figure 1

Figure 1

Several findings suggested that VAs like isoflurane inhibit complex I near the site of UQ reduction: the gas-1 mutation strongly affects anesthetic sensitivity; the GAS-1 protein forms part of a specific binding site of the UQ headgroup; and GAS-1 has a structural motif suspected to bind halothane. Furthermore, knockdown of the UQ binding site subunits PSST and GAS-1 also increase sensitivity to VAs.14 Moreover, the UQ biosynthesis mutant clk-1(qm30), which can only produce demethoxyubiquinone, an UQ analog with an aberrant headgroup, is hypersensitive to anesthetics.6 Because interference with UQ reduction by VAs is a possible site of anesthetic action, we tested the hypothesis that isoflurane competes with UQ for a common hydrophobic pocket in complex I (for a model, see Fig. 1). If so, VAs could inhibit electron flow and energy production by displacing UQ from its normal binding site at complex I.

Complex I exists primarily as a component of a very large, intricately entwined, functional structure, the mitochondrial supercomplex.17,18 In worms, virtually all of complex I is associated with complexes III and IV, in I:III2:IVn supercomplexes.19 Different members of this supercomplex have allosteric effects on the activity of complex I, such that complex I activity is maximized when it is part of the I:III:IV supercomplex.1720 In addition, Xi et al. have shown that complex IV of the ETC binds labeled halothane.21 It is possible that anesthetic binding to complex III or IV could indirectly change the activity of complex I in an allosteric manner, by disturbing the normal tight associations between members of the supercomplex (e.g., Fig. 1, B and C). To interpret the anesthetic inhibition as a direct interaction between complex I and anesthetic, we had to exclude indirect inhibition via the respiratory supercomplex. Therefore we measured activities of each of these enzymatic protein complexes. We also asked whether isoflurane can indirectly inhibit complex I by dissociating mitochondrial supercomplexes or by weakening the associations between members of supercomplexes I:III2:IVn.

Back to Top | Article Outline



C. elegans wildtype N2 (Bristol) was obtained from the Caenorhabditis Genetics Center (Minneapolis, MN). The mutant gas-1(fc21) was isolated in our laboratory.1,13 Worms were maintained according to standard techniques.22 Mitochondrial preparation was as previously described,10 with the minor change that worms were separated from bacteria by settling in ice-cold S-basal. In addition, the protease inhibitor cocktail was omitted.

Back to Top | Article Outline

Mitochondrial Assays

Spectrophotometric ETC assays were performed as previously described.10 Fresh mitochondria preparations were solubilized with cholate, and zero-order rates were determined spectrophotometrically for the following enzyme activities: rotenone-sensitive NADH cytochrome c reductase (complexes I–III), succinate cytochrome c reductase (complexes II–III), antimycin A-sensitive decylubiquinol: cytochrome c reductase (complex III), NADH ferricyanide reductase (NFR), and rotenone-sensitive NADH decylubiquinone reductase (complex I). Assays and calculations were performed as described by Hoppel et al.23

Azide-sensitive TMPD/ascorbate:oxidase (complex IV) was measured with a Hansatech Oxytherm Clark electrode at 30°C. Assay conditions were fresh intact mitochondria (0.3 mg/mL) in 0.5 mL gemisch (0.1 M KCl, 0.1% bovine serum albumine, 2 mM EGTA, 50 mM MOPS, 2 mM potassium phosphate pHed to 7.4), either with or without 4 mM KCN, energized with 25 mM ascorbate, 2.5 mM N,N,N′,N′-tetramethyl-p-phenylene-diamine (TMPD), and 2 mM ADP.

For all assays, the isoflurane concentration was adjusted by mixing appropriate volumes of reaction buffer and buffer saturated with anesthetic before mitochondria were added. Isoflurane concentrations are given as percentage equivalent in air. We determined by gas chromatography that fully isoflurane-saturated air contains 38% anesthetic at 30°C. Electron transport activities were measured in methacrylate disposable cuvettes after confirming that results were indistinguishable from those achieved in optical glass cuvettes. Loss of isoflurane during the course of kinetic measurements was negligible, judged by the observation that complex I activity did not increase during runs.

Back to Top | Article Outline

Blue Native/High-Resolution Clear Native Gel Electrophoresis (BN/hrCNE)

For the first dimension of blue native gels (BNG), digitonin was used to solubilize frozen −80°C) mitochondrial samples with a detergent/protein mass ratio of 6/1 (19,20,24with minor modifications). For the second dimension, excised lanes from the first-dimension gel were incubated for 1 hour at 20°C in a gastight container in either anesthetic-free BNG or the same buffer half saturated with isoflurane (equivalent 19% isoflurane in air; ∼7.4 mM). Immediately afterward, each gel strip was placed horizontally on top of a 3.5%–11% clear native gradient polyacrylamide gel and run at 125 V, 4°C for 15 hours. The cathode buffer contained 0.05% deoxycholate and 0.02% dodecyl maltoside.

Back to Top | Article Outline


Equine heart cytochrome c was procured from Calbiochem/E.M.D., and isoflurane from Abbot Laboratories (North Chicago, IL). All other reagents were from Sigma-Aldrich (St. Louis, MO).

Back to Top | Article Outline

Statistical Analysis

Experimental data points in all figures are given as arithmetic means of independent mitochondria isolations with error bars representing the SD. SigmaPlot 11.2 (Systat Software, Inc., Chicago, IL) was used for plotting data and curve fitting. This software reports uncertainties for the fit parameters as asymptotic standard errors. Errors for IC50s were calculated using the macro Propagation25 for Microsoft Excel. P values for the comparison of fit parameters were reported with asterisks if P < 0.01.

Back to Top | Article Outline


Direct Inhibition of Complex I

Isoflurane inhibited complex I activity (rotenone-sensitive electron transfer from NADH to decylubiquinone) in a dose-dependent manner (Fig. 2A) with an IC50 of 6.1% (2.4 mM) (Table 1). The rotenone-resistant activity of complex I was an order of magnitude lower (15 to 30 nmol/min/mg) and not affected by isoflurane (data not shown).

Figure 2

Figure 2

Table 1

Table 1

Isoflurane inhibited flow of electrons from complex I to complex III (I–III activity, NADH:cytochrome c reductase) with similar characteristics (Fig. 2B, Table 1) as in the complex I assay: IC50 = 8.4% (3.3 mM). By contrast, complexes II–III activity (succinate:cytochrome c reductase) was very resistant to isoflurane (Fig. 2C, Table 1) such that half-maximal inhibition could not be achieved by 19% (7.4 mM) isoflurane. We next measured the activities of complexes III (Fig. 2D) and IV (Fig. 2E) as antimycin A-sensitive decylubiquinone:cytochrome c reductase and azide-sensitive TMPD/ascorbate:oxidase, respectively. Neither activity was inhibited by up to 19% isoflurane.

Back to Top | Article Outline

Site of Action Within Complex I

NFR (NADH:ferredoxin reductase) measures the very first step of electron transfer through complex I.23 The artificial electron acceptor ferricyanide diverts electrons from the most proximal subcomplex of complex I, the flavoprotein (the NADH dehydrogenase).23 The flavoprotein sits at the tip of the matrix arm of complex I (Fig. 1E). Isoflurane has no significant effect on this activity (Fig. 2F).

Back to Top | Article Outline

Supercomplex Stability

Mitochondrial supercomplexes can be identified on native gels, where they retain enzymatic activity and respiratory capability.9 Mutations in complexes III or IV can weaken the interactions within supercomplexes, altering their appearance on BNGs, and indirectly decreasing the measurable activity of complex I itself.19,20 Analogously, isoflurane may inhibit complex I activity by weakening the integrity of the supercomplex. We investigated this possibility by single-dimension BNG electrophoresis of mitochondrial proteins after exposure of the mitochondria to isoflurane. No changes were seen in comparison with controls (data not shown). In addition, after running a BNG in a single dimension, we exposed the gel to 19% (7.4 mM) isoflurane. Immediately after exposure to isoflurane, a lane of the BNG was run in a second dimension in a stronger detergent, maltoside. A control BNG was treated similarly without the isoflurane exposure. If the anesthetic had weakened or strengthened interactions between components of the supercomplex, then exposure to isoflurane would alter separation, in comparison with the untreated control, of complexes from the supercomplex in the second dimension. Increased dissociation of supercomplexes would be seen as more bands migrating below a diagonal line that represents protein complexes that are unchanged by electrophoresis in a second dimension. We observed that even after treatment of the gel with 19% isoflurane, the supercomplexes did not differ from controls in patterns of dissociation (Fig. 3). The same technique (but without isoflurane treatment) has been successfully used to demonstrate that mutations in complex III weakened interaction between complex IV and the supercomplex.26

Figure 3

Figure 3

Back to Top | Article Outline

Mechanism of Inhibition

Both UQ and isoflurane are lipophilic and interact with the distal subcomplexes of complex I. The protein encoded by gas-1 is part of this segment of complex I. We therefore hypothesized that the substrate UQ and inhibitor isoflurane compete for binding of the free enzyme.

Complex I from bovine heart follows hyperbolic (“Michaelis–Menten”) kinetics for UQ substrates.27,28 This type of kinetics allows differentiation among 3 possible mechanisms of inhibition (competitive, noncompetitive, and uncompetitive) by comparing Vlim and apparent K for UQ measured in the absence and presence of the inhibitor.29 We observed 510 ± 13 and 332 ± 8 nmol/min/mg for Vlim in the absence and presence of 5.7% isoflurane, respectively (Fig. 4A and Table 2). These clearly distinct values excluded competition of UQ and isoflurane for free complex I as the underlying mechanism of inhibition. We also found that nematode complex I systematically deviated from hyperbolic kinetics (Fig. 4, A and B). Instead, the data fit sigmoidal kinetics with Hill coefficients above 2 (Table 2), suggesting positive cooperation between UQ binding sites. The cooperative nature of UQ kinetics was not disrupted by the presence of anesthetic (Fig. 4A).

Figure 4

Figure 4

Table 2

Table 2

Back to Top | Article Outline

Complex I Mutant Gas-1 (fc21)

In the absence of isoflurane, complexes I and I–III were severely depressed by the gas-1 mutation alone: 62 nmol/min/mg and 85 nmol/min/mg, respectively, which are 25% and 8% of the wildtype level (Fig. 2, A and B, Table 1). Isoflurane minimally inhibited these 2 assays in gas-1.

NFR activity in the mutant was about 60% of wildtype (Fig. 2F). As in wildtype, NFR activity in gas-1 was not inhibited by isoflurane. Hence, the presence of a mutant subunit in the neighboring iron–sulfur–protein subcomplex does not induce a new target for isoflurane on the flavoprotein. Complexes II–III activity of gas-1 was similar to that of wildtype (Fig. 2C), as can be expected for an activity that does not involve the mutated peptide. Complexes III and IV were not studied in gas-1 because the mutation does not change these activities from wildtype, and the wildtype activities are not inhibited by isoflurane.

Back to Top | Article Outline


We have shown for the first time that isoflurane inhibition of complex I, the most susceptible member of the ETC to inhibition by VAs, is not due to competition of the agent for the hydrophobic binding site of UQ within complex I. In addition, it appears that isoflurane has no effect on the flow of electrons through the flavoprotein, the proximal tip of the matrix arm of complex I that protrudes into the aqueous matrix of the mitochondrion. Enzymatic rates of both complex I and complexes I–III were very sensitive to inhibition by isoflurane. Also, complex I activity is measured using a water-soluble form of UQ, whereas I–III activity uses the native quinone. Because both I and I–III activity are inhibited by isoflurane, anesthetic inhibition of complex I does not involve the binding sites for the UQ isoprenyl tail.a

Our results are fully consistent with selective inhibition of complex I by isoflurane at a site distal to the flavoprotein subcomplex (Fig. 1E). Considering that both cytochrome c reductase assays (I–III Fig. 2B, II–III Fig. 2C) use a shared electron path from UQ to cytochrome c, the marked difference in sensitivity to isoflurane confirms that complex I is more sensitive to the drug than are complexes II and III. Our results for isoflurane inhibition of complex I in C. elegans are in agreement with previous findings for halothane in rat liver mitochondria30 and for 3 VAs, including isoflurane, in pig heart mitochondria.31 However, these earlier reports did not establish IC50s for isoflurane, nor did they definitively establish that the complex I inhibition was selective to complex I or a portion of complex I. Our study, because it interrogates each individual relevant step of electron flow, represents a comprehensive evaluation of the possible indirect effects of isoflurane on complex I that have not been previously reported.

Likewise, previous investigations of the affects of VAs on ETCs preceded our current knowledge of the importance of supercomplex I:III:IVn for proper complex I function. In vivo complexes I, III, and IV form a supercomplex that is thought to be the functional unit of respiration.1719 Normal associations with complex IV and complex III are necessary for maximal electron transport activity of fully assembled complex I in vitro.1720 The data indicate that supercomplexes remain intact in ETC assays and that complexes III and IV can indirectly affect complex I activity. Specific interactions of VAs with complexes III and IV have been convincingly demonstrated by Xi et al. using radiolabeled halothane.21 It is therefore also conceivable that anesthetic bound to complex III or IV could elicit activity changes in complex I (Fig. 1B). However, such an interaction would be expected to also affect the activity of complex III or IV. Our results indicate that neither complex III nor complex IV is the relevant target for isoflurane. Alternatively, isoflurane might bind to complexes III and IV without affecting their individual activities but indirectly inhibit complex I by destabilizing activating interactions between complex I and its partners (Fig. 1C). Weakened interactions between members of the supercomplexes that are not detected by 1-dimensional BNGs can be unmasked by 2-dimensional (2D) BNGs. We have identified mutations in subunits of complex III that cause an ∼40% decrease in complex I activity and a striking vulnerability of the I:III:IV supercomplex to dissociation by maltoside,26 conditions identical to those we used for our 2D native gels presented here. If isoflurane inhibited complex I by weakening the association of complexes within the supercomplex, we would expect an observable effect on stability after exposure to 19% isoflurane. However, 2D electrophoresis did not reveal any stabilizing or destabilizing effects of 19% isoflurane on the supercomplexes. Thus, isoflurane did not disrupt the binding of complexes I, III, and IV within the supercomplex. These results further validate the conclusion drawn from the ETC assays, that isoflurane interacts directly with complex I.

VAs are a heterogeneous group of chemicals, not identified by a common active chemical group. Their anesthetic potency is largely determined by their gas/oil partitioning coefficient.32 The binding sites for UQ and 3 classes of complex I inhibitors are believed to be located in a hydrophobic fold of the complex.12,33,34 UQ ordinarily has a long hydrophobic polyisoprenyl tail thought to be embedded in the mitochondrial membrane, as well as a relatively hydrophilic head portion that swings between hydrophilic domains of complex I and complex III, passing electrons through the ETC.12,33,34 In the prokaryote Thermus thermophilus, crystallographic data demonstrate that the binding site for quinone in complex I is sufficiently large to chamber VA molecules.9,35 In addition, that part of the protein Nqo4 (also termed 49 kDa), an orthologue of GAS-1, which binds to UQ, is a 4 α-helix bundle, located precisely at the junction of the transmembrane and the hydrophilic domains that compose the quinone binding site. This type of structure, a 4 α-helix bundle, has been used extensively to model the binding of VAs to membrane proteins.15,16 Thus, the hydrophobic binding pocket, containing the UQ binding site, could be a prime candidate for the binding of VAs. Just as some of the inhibitor binding sites of complex I overlap with each other and with UQ binding sites, it is easily conceivable that anesthetics interfere with the binding of UQ, and vice versa. However, such competitive inhibition is unequivocally excluded (Fig. 4, Table 2) because even very high concentrations of decylubiquinone could not out-compete a fixed concentration (5.7%) of isoflurane. We conclude that isoflurane and UQ neither share a common binding site nor bind to overlapping sites (Fig. 1, D and E). Our results furthermore excluded another variant of competitive inhibition: heterotropic allosteric inhibition. Binding of either isoflurane or UQ does not change the conformation of the complex as to negatively affect binding of the other. However, it is likely that isoflurane binds to other sites within complex I, the I:III:IV supercomplex, or even to UQ itself.

Once kinetics indicated that cooperative binding of UQ had to be considered, the simple model to distinguish the remaining 2 mechanisms of inhibition was no longer applicable. Thus, we were unable to determine whether isoflurane binds and blocks exclusively a preformed enzyme-substrate complex (uncompetitive inhibition) or binds to complex I independent of UQ (noncompetitive inhibition). Having established wildtype complex I as a selective target for isoflurane, but not in direct competition with Q for a binding site, we tested whether the hypersensitivity of gas-1 may be caused directly by hypersensitivity of its mutant complex I to isoflurane.

GAS-1, a core subunit of complex I, is 83.4% similar, at the amino acid level, to its human counterpart.13,14 The fc21 missense mutation (R290K) affects a residue in GAS-1 that is conserved even in simple bacterial NADH:UQ reductases. It renders C. elegans exquisitely sensitive to all Vas, but especially so to isoflurane (wildtype EC50 = 7.0 ± 0.1%, fc21 EC50 = 1.3 ± 0.2%).1,10 At 7.0% isoflurane, wildtype complex I is severely inhibited to about 40% of its potential activity, suggesting that the inhibition may cause immobility of the animal. However, this argument loses traction when results from the mutant are considered. The baseline activity of mutant complex I was very low, yet gas-1 is quite mobile in the absence of isoflurane. Its complex I activity measured in the absence of anesthetic is already lower than that of the wildtype mitochondria in the presence of anesthetizing concentrations of isoflurane. A similar pattern as described here for isoflurane has previously been found for halothane.36 Thus, there is no common threshold of complex I activity (measured in vitro) for wildtype and mutant, below which the worms become immobile/anesthetized.

It would appear that although complex I baseline activity (before inhibition) is a fair predictor of the anesthetic sensitivity of an organism,4 the absolute level of complex I activity (resulting from the combination of genetic background and inhibition by anesthetic) is a poor predictor of the anesthetic state of the animal. However, several points indicate that complex I is important in determining anesthetic sensitivity. First, when complex I activity is inhibited by knockdown of multiple subunits of complex I, anesthetic sensitivity is almost uniformly increased.14 Second, children with complex I defects have been noted to have an increased sensitivity to VAs.5 Third, as shown here and by others, of all complexes in the ETC, complex I is by far the most sensitive to VAs.30,31,37 Finally, as shown in this study, complex I activity in gas-1 was resistant to further depression by the anesthetic (the isoflurane IC50 for gas-1 > that for N2). Thus, although isoflurane does not compete directly with quinone for binding to complex I, a mutation in a protein that is part of the UQ binding site does cause resistance to inhibition. The simplest interpretation of these data is that the mutation alters an isoflurane target.

How can one reconcile that complex I dysfunction leads to anesthetic hypersensitivity, yet we cannot correlate levels of complex I activity with the anesthetic state? First, defects in complex I, which changes its sensitivity to isoflurane, may lead to more important secondary changes in pertinent downstream anesthetic targets, resulting in an increase in anesthetic sensitivity. Second, ETC assays measure maximal enzymatic activity, a condition probably not attained in vivo. N2 and gas-1 animals could have similar basal rates of complex I function in vivo. However, this rate may approach gas-1's maximal complex I function and render the mutant complex I more vulnerable than that of wildtype to even small decreases in capacity. This difference could result, in the presence of isoflurane, in a more rapid decrease by gas-1 than N2 to a common in vivo complex I threshold for immobility. Third, chronic loss of complex I activity (as in gas-1) may lead to compensatory changes that allow the animal to function at low complex I rates not tolerated by the wildtype animal during acute anesthetic exposure. For example, the capacity for complex II–dependent respiration is up in gas-110, and we have shown here that complex II activity is not inhibited by isoflurane. Fourth, the anesthetic state may be the result of inhibition of multiple types of targets, of which mitochondria are only one. Studies to differentiate between these possibilities are underway.

Finally, we have found that UQ itself has a cooperative interaction with complex I not previously appreciated. To our knowledge we are the first to report non-Michaelis–Menten behavior of complex I. For complex I the existence of multiple UQ binding sites is supported by experimental evidence and demanded by several mechanistic models;38,39 however, their exact number is still a matter of debate.40 Tocilescu et al. identified residues on the 49 kDa subunit (the GAS-1 orthologue) and the adjacent subunit (PSST) that altered the interaction of complex I with UQ.12 However, this work did not establish whether more than 1 UQ molecule may occupy this redox active site, nor has it been excluded that additional UQ binding sites may exist elsewhere on complex I, which may be merely allosteric. Our measured Hill coefficients of 2.1 ± 0.2 and 2.5 ± 0.4 are consistent with either 2 strongly cooperative binding sites for UQ or more than 2, but weaker, interacting sites. It is noteworthy that isoflurane does not interrupt the cooperative interaction between UQ molecules because the Hill coefficient remains unchanged by the presence of isoflurane. It is crucial to acknowledge, however, that UQ kinetics are not trivial and are affected by multiple complicated interactions.41 The apparent cooperation we observe may be specific to C. elegans or even an artifact of our system. The water-soluble substrate decylubiquinone, which is used for this assay, may have unique properties of binding that are different from the native quinone. Such studies should be repeated in a well-defined system such as bovine mitochondrial preparations to clarify these possibilities.

In summary, we have shown that, in nematodes, isoflurane directly inhibits complex I function, at a site distal to the flavoprotein within the complex. We have excluded our initial hypothesis that the VAs compete with quinones for a hydrophobic binding site on complex I. The demonstrated stability of supercomplexes in the presence of isoflurane supported that complex I inhibition is not caused by isoflurane disrupting necessary interaction with neighboring complexes III and IV. Although complex I activity does predict VA sensitivity, absolute rates of complex I activity do not predict a threshold that correlates with immobility of the animal. Furthermore, we describe non-Michaelis–Menten kinetics, suggesting interacting binding sites, for UQ within complex I.

Back to Top | Article Outline


Name: Ernst-Bernhard Kayser, PhD.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Ernst-Bernhard Kayser has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Wichit Suthammarak, PhD.

Contribution: This author helped conduct the study and analyze the data.

Attestation: Wichit Suthammarak has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Phil G. Morgan, MD.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Phil G. Morgan has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Name: Margaret M. Sedensky, MD.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Margaret M. Sedensky has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

“Modified Hill Equation” to describe sigmoidal kinetics:
Cited Here...

Cited Here...

Where velocity v denotes the initial rate of product formation (here: NAD+) when the substrate (here: decylubiquinone) is present at a concentration of [S]. Parameters Vlim, K, h are constants: Vlim is the limiting velocity if enzyme could be fully saturated with substrate; K is the substrate concentration at which halfmaximal velocity is achieved, and h is the so-called Hill coefficient. Note that for the special case of h=1, the equation simplifies to describe hyperbolic (Michaelis–Menten) kinetics.
Cited Here...

Back to Top | Article Outline


1. Morgan PG, Sedensky MM. Mutations conferring new patterns of sensitivity to volatile anesthetics in Caenorhabditis elegans. Anesthesiology 1994;81:888–98
2. Crowder CM, Shebester LD, Schedl T. Behavioral effects of volatile anesthetics in Caenorhabditis elegans. Anesthesiology 1996;85:901–12
3. Kayser B, Rajaram S, Thomas S, Morgan PG, Sedensky MM. Control of anesthetic response in C. elegans. Toxicol Lett 1998;100–101:339–46
4. Humphrey JA, Hamming KS, Thacker CM, Scott RL, Sedensky MM, Snutch TP, Morgan PG, Nash HA. An unusual cation channel and its novel regulator show cross-species conservation of effects on general anesthesia. Current Biol 2007;17:624–9
5. Morgan PG, Hoppel CL, Sedensky MM. Mitochondrial defects and anesthetic sensitivity. Anesthesiology 2002;96: 1268–70
6. Kayser EB, Sedensky MM, Morgan PG. The effects of complex I function and oxidative damage on lifespan and anesthetic sensitivity in Caenorhabditis elegans. Mech Aging Dev 2004;125:455–64
7. Kashani-Poor N, Zwicker K, Kerscher S, Brandt U. A central functional role for the 49-kDa subunit within the catalytic core of mitochondrial complex I. J Biol Chem 2001;276: 24,082–7
8. Lenaz G, Genova ML. Mobility and function of coenzyme Q (ubiquinone) in the mitochondrial respiratory chain. Biochim Biophys Acta 2009;1787:563–73
9. Efremov RG, Baradaran R, Sazanov LA. The architecture of respiratory complex I. Nature 2010;465:441–5
10. Kayser EB, Morgan PG, Hoppel CL, Sedensky MM. Mitochondrial expression and function of GAS-1 in Caenorhabditis elegans. J Biol Chem 2001;276:20,551–8
11. Walker JE. The NADH:ubiquinone oxidoreductase (complex I) of respiratory chains. Q Rev Biophys 1992;25:253–324
12. Tocilescu MA, Fendel U, Zwicker K, Kerscher S, Brandt U. Exploring the ubiquinone binding cavity of respiratory complex I. J Biol Chem 2007;282:29,514–20
13. Kayser EB, Morgan PG, Sedensky MM. GAS-1: a mitochondrial protein controls sensitivity to volatile anesthetics in the nematode Caenorhabditis elegans. Anesthesiology 1999;90:545–54
14. Falk MJ, Rosenjack JR, Polyak E, Suthammarak W, Chen Z, Morgan PG, Sedensky MM. Subcomplex 1λ specifically controls integrated mitochondrial functions in Caenorhabditis elegans. PLoS One 2009;8:e6607
15. Liu J, Strzalka J, Tronin A, Johansson JS, Blasie JK. Mechanism of interaction between the general anesthetic halothane and a model ion channel protein. II: fluorescence and vibrational spectroscopy using a cyanophenylalanine probe. Biophys J 2009;96:4176–87
16. Cui T, Bondarenko V, Ma D, Canlas C, Brandon NR, Johansson JS, Xu Y, Tang P. Four-alpha-helix bundle with designed anesthetic binding pockets. Part II: halothane effects on structure and dynamics. Biophys J 2008;94:4464–72
17. Schagger H, Pfeiffer K. Supercomplexes in the respiratory chains of yeast and mammalian mitochondria. EMBO J 2000;19:1777–83
18. Schafer E, Seelert H, Reifschneider NH, Krause F, Dencher NA, Vonck J. Architecture of active mammalian respiratory chain supercomplexes. J Biol Chem 2006;281:15,370–5
19. Suthammarak W, Yang YY, Morgan PG, Sedensky MM. Complex I function is defective in complex IV-deficient Caenorhabditis elegans. J Biol Chem 2009;284:6425–35
20. Suthammarak W, Morgan PG, Sedensky MM. Mutations in mitochondrial complex III uniquely affect complex I in Caenorhabditis elegans. J Biol Chem 2010;285:40,724–31
21. Xi J, Liu R, Asbury GR, Eckenhoff MF, Eckenhoff RG. Inhalational anesthetic-binding proteins in rat neuronal membranes. J Biol Chem 2004;279:19,628–33
22. Stiernagle T. Maintenance of C. elegans. In: WormBook. The C. Elegans Research Community, ed. , 2006, doi/10.1895/wormbook.1.101.1,
23. Hoppel CL, Kerner J, Turkaly P, Turkaly J, Tandler B. The malonyl-CoA-sensitive form of carnitine palmitoyltransferase is not localized exclusively in the outer membrane of rat liver mitochondria. J Biol Chem 1998;273:23,495–503
24. Wittig I, Karas M, Schägger H. High resolution clear native electrophoresis for in-gel functional assays and fluorescence studies of membrane protein complexes. Mol Cell Proteomics 2007;6:1215–25
25. De Levie R. Advanced Excel for Scientific Data Analysis. New York: Oxford University Press, 2008
26. Suthammarak W, Morgan PG, Sedensky MM. Mutations in mitochondrial complex III uniquely affect complex I in Caenorhabditis elegans. J Biol Chem 2010;285:40724–31
27. Fato R, Estornell E, Di Bernardo S, Pallotti F, Parenti Castelli G, Lenaz G. Steady-state kinetics of the reduction of coenzyme Q analogs by complex I (NADH:ubiquinone oxidoreductase) in bovine heart mitochondria and submitochondrial particles. Biochem 1996;35:2705–16
28. Nakashima Y, Shinzawa-Itoh K, Watanabe K, Naoki K, Hano N, Yoshikawa S. Steady-state kinetics of NADH:coenzyme Q oxidoreductase isolated from bovine heart mitochondria. J Bioenerg Biomembr 2002;34:11–9
29. Copeland RA, Enzymes A. Practical Introduction to Structure, Mechanism, and Data Analysis, 2nd ed. New York: Wiley-VCH, 2008
30. Harris RA, Munroe J, Farmer B, Kim KC, Jenkins P. Action of halothane upon mitochondrial respiration. Arch Biochem Biophys 1971;142:435–44
31. Hanley P, Ray J, Brandt U, Daut J. Halothane, isoflurane and sevoflurane inhibit NADH:ubiquinone oxidoreductase (complex I) of cardiac mitochondria. J Physiol 2002;544:687–93
32. Tanifuji Y, Eger EI II, Terrell R. Some characteristics of an exceptionally potent inhaled anesthetic: thiomethoxyflurane. Anesth Analg 1977;56:387–90
33. Fendel U, Tocilescu M, Kerscher S, Brandt U. Exploring the inhibitor binding pocket of respiratory complex I. Biochim Biophys Acta 2008;1777:660–5
34. Degli-Esposti M. Inhibitors of NADH-ubiquinone reductase: an overview. Biochim Biophys Acta 1998;1364:222–35
35. Sazanov LA, Hinchliffe P. Structure of the hydrophilic domain of respiratory complex I from Thermus thermophilus. Science 2006;311:50,311–7
36. Kayser EB, Morgan PG, Sedensky MM. Mitochondrial complex I function affects halothane sensitivity in Caenorhabditis elegans. Anesthesiology 2004;101:365–72
37. Triepels RH, Van Den Heuvel LP, Trijbels JM, Smeitink JA. Respiratory chain complex I deficiency. Am J Med Genet 2001;106:37–45
38. Tormo JR, Estornell E. New evidence for the multiplicity of ubiquinone- and inhibitor-binding sites in the mitochondrial complex I. Arch Biochem Biophys 2000;381:241–6
39. Dutton PL, Moser CC, Sled VD, Daldal F, Ohnishi T. A reductant-induced oxidation mechanism for complex I. Biochim Biophys Acta 1998;1364:245–57
40. Shinzawa-Itoh K, Seiyama J, Terada H, Nakatsubo R, Naoki K, Nakashima Y, Yoshikawa S. Bovine heart NADH-ubiquinone oxidoreductase contains one molecule of ubiquinone with ten isoprene units as one of the cofactors. Biochem 2010;49:487–92
41. Vinogradov AD. Catalytic properties of the mitochondrial NADH-ubiquinone oxidoreductase (complex I) and the pseudo-reversible active/inactive enzyme transition. Biochim Biophys Acta 1998;1364:169–85
© 2011 International Anesthesia Research Society