SINCE its introduction in the 1980s, propofol has become the most widely used intravenous anesthetic in clinical anesthesia. Its success is attributable to its pharmacokinetic advantages and its versatile applications, including the induction and maintenance of general anesthesia and procedural and intensive care sedation. In addition to its well-known sedative and hypnotic actions, propofol is a potent amnestic agent, a desirable action in many of the unpleasant and stressful applications for its use. Modulation of receptors for the inhibitory neurotransmitter γ-aminobutyric acid is an important molecular mechanism for some of the pharmacologic effects of propofol. Thus, knock-in mice harboring γ-aminobutyric acid receptor type A β3 subunits rendered insensitive to propofol or etomidate by a single point mutation are resistant to the immobilizing effects of propofol and etomidate in vivo .1The molecular mechanism for the amnestic actions of propofol has not been identified directly, but a promising candidate is modulation of the α5 γ-aminobutyric acid receptor type A subunit. This subunit is expressed in extrasynaptic γ-aminobutyric acid type A receptors and is responsible for mediating tonic inhibitory currents in many neurons. Knockout of this receptor subunit in mice eliminates the memory-blocking action of etomidate without affecting sedation or hypnosis.2Another potential target that has been implicated in the amnestic actions of propofol is the endocannabinoid system, a widespread paracrine cell-signaling mechanism integral to several forms of short- and long-term synaptic plasticity.3Now comes a fascinating study published this month in Anesthesiology4showing that the endocannabinoid system indeed contributes to the effects of propofol on memory but counterintuitively by enhancing retrograde memory consolidation rather than suppressing anterograde memory, the better known effect of propofol on memory. This novel twist in the pharmacology of propofol suggests that a single drug blends the waters from Lethe, the river of forgetfulness, with the waters from Mnemosyne, the river of memory, to block consolidation of anterograde memory and facilitate consolidation of retrograde memory.
The endocannabinoid system consists of cannabinoid receptors, of which there are two well-characterized subtypes (CB1 and CB2), endogenous cannabinoids (endocannabinoids) such as anandamide, and the metabolic pathways responsible for endocannabinoid synthesis and degradation. CB1 receptors are highly expressed in brain and mediate most of the psychotropic effects of Δ9-tetrahydrocannabinol (Δ9THC), the principal psychoactive component of cannabis. Propofol has been shown5to inhibit the activity of fatty acid amide hydrolase, a key enzyme in the degradation of anandamide (fig. 1). Patel et al. reported that sedative doses of propofol, but not of thiopental, increased brain concentrations of anandamide in mice in vivo .5Loss of righting reflex, an index of hypnosis in mice, after administration of propofol, but not thiopental, was antagonized by pretreatment with the CB1 receptor antagonist rimonabant, whereas pretreatment with WIN 55212-2, a CB1 receptor agonist, potentiated loss of righting reflex caused by propofol. In addition, propofol, but not thiopental, etomidate, or midazolam, inhibited fatty acid amide hydrolase activity in vitro . These findings implicated the endocannabinoid system in the sedative actions of propofol, but not of the other intravenous general anesthetics tested, through inhibition of the degradation of endogenous CB1 receptor agonists such as anandamide. Interaction with the endocannabinoid system by propofol is an attractive candidate for some of the agent-specific actions of propofol, compared with other general anesthetics, that are shared with cannabinoids including its antiemetic, mood altering, and postoperative dreaming effects.
Fig. 1. Schematic representation of the propofol effect on endocannabinoid signaling at the synapse. Endocannabinoids are synthesized at the postsynaptic membrane and activate presynaptic CB1 cannabinoid receptors to reduce presynaptic activity by inhibiting Ca2+channels and activating K+channels. Anandamide is removed from the synapse by uptake into the postsynaptic neuron, where it is hydrolyzed and inactivated by fatty acid amide hydrolase (FAAH). Propofol indirectly activates endocannabinoid signaling by inhibiting FAAH, which increases intracellular anandamide concentration, reducing anandamide uptake and increasing synaptic anandamide concentrations. This increased concentration of synaptic anandamide activates presynaptic CB1 receptors, and reduces transmitter release. Propofol also inhibits postsynaptic neuronal activity through potentiation of γ-aminobutyric acidA(GABAA) receptors.
Fig. 1. Schematic representation of the propofol effect on endocannabinoid signaling at the synapse. Endocannabinoids are synthesized at the postsynaptic membrane and activate presynaptic CB1 cannabinoid receptors to reduce presynaptic activity by inhibiting Ca2+channels and activating K+channels. Anandamide is removed from the synapse by uptake into the postsynaptic neuron, where it is hydrolyzed and inactivated by fatty acid amide hydrolase (FAAH). Propofol indirectly activates endocannabinoid signaling by inhibiting FAAH, which increases intracellular anandamide concentration, reducing anandamide uptake and increasing synaptic anandamide concentrations. This increased concentration of synaptic anandamide activates presynaptic CB1 receptors, and reduces transmitter release. Propofol also inhibits postsynaptic neuronal activity through potentiation of γ-aminobutyric acidA(GABAA) receptors.
Impaired learning and memory are well-known effects of cannabinoids. These actions can be observed in animal models of memory that depend on the hippocampus, a brain region critical to various forms of memory and known to express high concentrations of CB1 receptors.6A role for the endocannabinoid system in anterograde memory is supported by observations that cannabinoid agonists impair, whereas antagonists or CB1 receptor knock-down enhance, performance on hippocampal-dependent memory tasks.7However, recent studies demonstrate that cannabinoids can facilitate memory consolidation in specific situations, particularly in memory tasks that involve an emotional component such as fear conditioning.8With this in mind, Hauer et al. 4tested the hypothesis that propofol effects on endocannabinoid signaling are involved in enhanced consolidation of emotionally charged memories. Such an effect could be clinically relevant to posttraumatic stress disorder after intraoperative awareness or stressful intensive care unit events.
In studies of human volunteers, treatment with low nonsedative doses of propofol to achieve a state of conscious amnesia during learning produces anterograde amnesia by impairing long-term memory consolidation.9,10Thus, propofol disrupts memory formation when administered during encoding not by disrupting encoding but by preventing consolidation of short-term memory traces to long-term memory.11Similar effects have been demonstrated when subanesthetic doses of propofol are administered during training in an aversive memory paradigm.12Hauer et al. used a well-characterized animal model of aversive training, known as inhibitory avoidance, in which rats are trained to avoid a dark compartment associated with a foot shock. Memory retention was assessed 48 h later by measuring the time until the rat reentered the dark compartment (longer latency reflects stronger memory). Drug treatments at various times after the training period were used to determine their effects on memory consolidation, a paradigm that avoids drug effects on memory acquisition or encoding during training while detecting effects on posttraining consolidation. Relatively high anesthetic doses of propofol administered immediately or 30 min after training increased brain anandamide concentrations and enhanced retention of the aversive memory, whereas lower doses or administration of drug delayed 90 or 180 min after training had no effect. In contrast, anesthetic doses of midazolam or pentobarbital did not enhance retention of inhibitory avoidance or increase brain anandamide concentrations. Involvement of the cannabinoid system was evident in the ability of a CB1 receptor antagonist (rimonabant) coadministered immediately after training to reduce the enhanced retention latency produced by propofol. Thus, propofol enhanced consolidation of aversive memory in a time-sensitive manner involving CB1 receptors and associated with increased brain anandamide. Activation of the endocannabinoid system in the amygdala, a brain region that is important in aversive memory, is involved in enhanced retention of inhibitory avoidance behavior.6It will be interesting to determine whether the amygdala is also involved in propofol actions and whether propofol also affects consolidation in other nonaversive memory consolidation paradigms.
There are a number of important limitations to this study. The doses of propofol used are high for rat13and so might not be relevant to typical doses encountered in clinical anesthesia or intensive care sedation. Rimonabant and other CB1 receptor antagonists have cognitive effects on their own, so it will be important to show that such parallel mechanisms are not involved in antagonizing the propofol memory deficit. Rimonabant has other off-target effects independent of CB1 receptors that should be ruled out as possible mechanisms. Other substrates for fatty acid amide hydrolase (such as oleoylethanolamide) have procognitive effects independent of CB1 receptor agonism14that could contribute to the propofol effect. Resolution of these questions should further focus the precise mechanism or reveal additional mechanisms of propofol action. Finally, the story remains incomplete because other drugs known to inhibit fatty acid amide hydrolase, including some nonsteroidal antiinflammatory drugs, have minimal or no effects on memory. The reasons for this are unclear but might involve effects of cyclo-oxygenase inhibition on endocannabinoid metabolism, for example. This question further reveals the complexities of endocannabinoid signaling and memory mechanisms and highlights the need for additional investigations in this area.
The demonstration that propofol, but not a representative benzodiazepine or barbiturate, indirectly activates the endocannabinoid system to enhance aversive memory consolidation provides in vivo evidence for a unique γ-aminobutyric acid-independent mechanism of propofol action. That propofol has apparently agent-specific effects on the consolidation of aversive memories has potential clinical implications. It is plausible that this effect contributes to an increased incidence of aversive memories after intraoperative awareness associated with propofol-based intravenous anesthesia.15,16In addition, the important role of the endocannabinoid system in aversive memory consolidation provides a potential pharmacologic approach to blocking posttraumatic memory consolidation using CB1 receptor antagonists. Although these hypotheses require further experimental and clinical validation, it appears that propofol can no longer be considered the “milk of amnesia” we remember.
The authors thank Karl Herold, M.D., Ph.D. (Research Associate, Department of Anesthesiology, Weill Cornell Medical College, New York, New York), who assisted in the preparation of the figure.
*Department of Anesthesiology, Weill Cornell Medical College, New York, New York. hchemmi@med.cornell.edu. †Gill Center, Indiana University, Bloomington, Indiana, and Department of Anesthesiology, Indiana University School of Medicine, Indianapolis, Indiana.
