“…diverse mechanisms under lie ‘anesthesia’ as it pertains to different systems (brain vs. spine vs. autonomic system)…”
The concept of the so-called “minimum”—but actually median—alveolar concentration (MAC) required to prevent movement to a surgical stimulus was the original yardstick that advanced mechanistic anesthesia research by allowing comparisons between agents and species.1 Understanding the processes underlying MAC was the original “holy grail” of research into anesthetic drug actions, but it appeared to be uncrackable. Funding for laboratories to systematically look into mechanisms of immobility dried up as interest moved toward mechanisms of memory suppression and hypnosis, commonly referred to with the trendy but poorly defined term “unconsciousness.” This pitiful paucity of basic science investigation into mechanisms of MAC is unfortunate because of its biologic complexity and obvious clinical relevance. All clinicians will at some time have experienced the frustration of having a patient who is apparently overanesthetized, requires vasoconstrictors for blood pressure support, and displays burst suppression on the electroencephalogram—but then unexpectedly moves or coughs. This is inevitably accompanied by a (misleading) operating room Greek chorus: “The patient is waking up!” The anesthesiologist may decide to solve the “waking up” problem by increasing anesthesia or neuromuscular blockade, but—if s/he is honest—would be almost completely unable to explain to the Greek chorus why the patient moved except that it has nothing to do with “waking up.”
In this edition of the journal is a complex paper by Woods et al.,2 who have made substantial progress by discovering an unexpected molecular context of isoflurane-induced immobility. They found that decreased intrinsic neuronal excitability is the main mechanism of the low MAC seen in mice with a mitochondrial mutation that renders them hypersensitive to isoflurane. This (Morgan/Sedensky) laboratory has long pursued the role of mitochondria in determining responsiveness to anesthetics. Using the worm Caenorhabditis elegans, they identified Complex I of the mitochondrial electron transport chain as a critical determinant of behavioral sensitivity.3 Homozygosity for mutations in the core of Complex I is rare, but it is the most common cause of lethal leukodystrophies referred to as Leigh and Leigh-like syndromes. Patients with Leigh syndrome are extremely sensitive to volatile agents.4 Mice and flies with mutations in Complex I (e.g., NDUFS8 and NDUFS4) replicate cardinal features of Leigh syndrome, including behavioral hypersensitivity to volatile anesthetics.5 Mice with the NDUFS4 mutation show impaired excitatory synaptic function in the stimulated hippocampus, secondary to limitations in neuronal energy supply and presynaptic neurotransmitter kinetics.6 Woods et al., therefore, hypothesized that a similar energetic-synaptic process in the spinal cord lay behind isoflurane impairment of movement responses. They tested this hypothesis by targeting the NDUFS mutation to the spinal cord. They were wrong: instead, it seems that exaggerated isoflurane activation of the TREK potassium channel in some neuronal populations was what drove the lower MAC of isoflurane. The TREK potassium channels are part of the family of two-pore domain potassium channels. These channels are abundant in the central nervous system, where they help regulate intrinsic neuronal excitability and play a role in protection against ischemia. Norfluoxetine, which, in addition to its effects on serotoninergic systems, blocks TREK channels, and restored the MAC to roughly control values.
These results provide the framework to generate a vigorous flock of interesting new questions, many of which are raised by the authors. Figure 1 is a diagram depicting some of the causal interactions between mitochondria, intrinsic neuronal excitability, synaptic efficiency, and resultant behavioral movement in response to noxious stimuli (MAC). The activity of the TREK channel is under a complicated control system involving lipid rafts.7 The first problem is to work out the mechanisms by which intracellular mitochondrial function could alter TREK channel opening in the cell membrane, that is, what are the mechanisms underlying pathway 1 in figure 1? The potentiation of isoflurane’s immobilizing effect is, apparently, not dependent on maximal energy production capability of the mitochondrion. This mechanism (depicted as pathway 2 in fig. 1) does not seem to be important in unstressed circuits. Does this hold true when the neuronal metabolism is stressed by high levels of synaptic activity in the circuit? And why is this effect only seen in the ventral horn of the spinal cord and not in the other areas tested, the vestibular nucleus and central medial thalamus? There is also the problem that other studies involving TREK-1–knockout animals seem to show minimal changes in MAC for volatile agents,8 although mutations in other members of the same large family of tandem-pore potassium channels substantially sensitize mice to volatile anesthetics.9
There are also several potential clinical consequences of this work. First, it emphasizes the fact that diverse mechanisms underlie “anesthesia” as it pertains to different systems (brain vs. spine vs. autonomic system), as has been elegantly demonstrated using differential perfusion of the brain versus the spinal cord in a goat model.10 Thus, there can be no universal “depth-of-anesthesia” measure. Second, we should be aware that drugs that block TREK (like fluoxetine) have the potential to lower the threshold for intraoperative movement,11 while other drugs or physiologic conditions that open TREK might decrease movement, without affecting the threshold for other components of the anesthesia syndrome, like hypnosis. Most importantly, anesthetics are “dirty” drugs. Mitochondria are the crucible of a cell, and any interaction with mitochondria has the potential to cause collateral effects. Depending on the genetic and epigenetic background of the subject, these collateral effects may be irrelevant (probably the majority), may be damaging (neurotoxicity), or may be beneficial (conditioning). We have a good handle on the behavioral effects and the conventional side effects of anesthetics, but to widen the “crack,” we need to place our drugs into a personalized, genetic context.
Supported by grant No. RO1GM134107 from the National Institute of General Medical Sciences (Bethesda, Maryland; to Dr. Perouansky).
The authors declare no competing interests.