In this study, we find a synergistic effect of ketamine on morphine/fentanyl-mediated signaling that appears to be selective to the ERK1/2 pathway. Although ketamine was shown to inhibit the formation of cAMP in a dose-dependent manner in CHO cells expressing recombinant μ-, δ- and κ-opioid receptors, this effect was found to be neither additive nor synergistic when ketamine was used in conjunction with opioid receptor agonists (Hirota et al. 1999
). Interestingly, we find that ketamine is more effective in potentiating the phosphorylation of ERK1/2 by lower doses of opioid agonists than by higher doses. This correlates with studies showing that ketamine administration increases the effectiveness of opioids used in pain management (Abdel-Ghaffar et al. 1998
Although ketamine acts primarily as a non-competitive NMDA-receptor antagonist (Willets et al. 1990
; Yamamura et al. 1990
), several studies have shown that it also exhibits non-NMDA-mediated effects by modulating the activity of nicotinic (Scheller et al. 1996
), muscarinic (Hustveit et al. 1995
), opioid (Smith et al. 1980
; Finck and Ngai 1982
; Hustveit et al. 1995
; Hirota et al. 1999
) as well as AMPA receptors (Li et al. 2010
; Zarate et al. 2010
). However, the molecular mechanisms underlying this are not clearly understood. A recent study found that intraperitoneal administration of ketamine led to activation of the mTOR and ERK pathways in synaptoneurosomes of prefrontal cortex and this was completely blocked by pretreatment with a selective AMPA receptor inhibitor (Li et al. 2010
). In vitro
studies examining intracellular signaling have shown that ketamine interacts stereoselectively with μOR and κOR (Finck and Ngai 1982
; Hustveit et al. 1995
). Studies examining the effects of ketamine on binding of opiate ligands find that high but clinically achievable concentrations of ketamine (> 100 μM) are able to displace a non-selective opioid receptor antagonist ([3
H] diprenorphine) binding to recombinant μOR (Hirota et al. 1999
). In these studies, ketamine was found to significantly increase the affinity of [3
H]diprenorphine binding with no change in total binding. In this study, we find that ketamine significantly increases the affinity of [3
H]morphine or [3
H]fentanyl albeit only at high doses (100 μM) in cells endogenously expressing μ receptors. The fact that we find that ketamine significantly potentiates morphine- or fentanyl-mediated signaling at doses that do not affect binding supports a role for interactions at the level of intracellular signaling by these drugs.
An interesting observation made in the current study is that while ketamine had no significant effect on the potency of morphine (EC50
from 5.8 to 5.9 nM) in inducing ERK1/2 phosphorylation, it substantially increased the potency of fentanyl (EC50
from 68 to 0.5 nM). This could reflect that ketamine differentially modulates the pathways leading to ERK1/2 phosphorylation following μOR activation by morphine or fentanyl. Studies show that Gi/o coupled receptor-mediated ERK1/2 phosphorylation can occur via G-protein-dependent/-independent mechanisms (May and Hill 2008
). In this context, a recent study showed that while morphine could induce ERK1/2 phosphorylation primarily via a G protein-dependent pathway through PKCε, fentanyl could activate both PKCε and β-arrestin-mediated pathways (Zheng et al. 2011
). Given that fentanyl can activate ERK1/2 phosphorylation by multiple mechanisms compared with morphine, this would suggest that the robust effects observed with a combination of ketamine and fentanyl could be due to greater effects of ketamine on G protein-independent pathways of ERK1/2 phosphorylation. This is supported by our observation that ketamine did not potentiate opioid (morphine or fentanyl)-mediated signaling at the level of receptor associated G-proteins. Thus, our results suggest that ketamine stabilizes μOR conformations that promote its association with β-arrestin thereby potentiating G protein-independent signaling. Alternatively, ketamine may be changing the localization of the μOR from lipid to non-lipid rafts given the evidence showing that G-protein-mediated ERK1/2 phosphorylation requires μOR localization to lipid rafts whereas β-arrestin-mediated ERK1/2 phosphorylation requires its localization to non-lipid raft domains (Zheng et al. 2008
). Further studies are needed to evaluate these possibilities.
The ability of ketamine to increase the duration of opioid-induced analgesia suggests that it could also modulate opioid-induced signaling in such a way that these receptors remained in an active state for a longer time. This could be achieved by influencing the time course of desensitization and resensitization of signaling by opioid receptors. Receptor desensitization which occurs upon short-term (sec to min) exposure of cells to agonists is mediated by uncoupling of activated receptors from G-proteins and this effectively terminates the signaling process. Receptor resensitization occurs after withdrawal of agonists from the media leading to a new round of signaling events following a second challenge with agonists. Our results demonstrate that ketamine delays the desensitization and improves the resensitization of ERK1/2 signaling. The overall effect of ketamine would thus appear to be in keeping opioid-induced ERK1/2 signaling active for a longer time period.
Interestingly, several studies have demonstrated the involvement of ERK1/2 in nociception. For example, noxius stimulus leads to an increase in ERK1/2 phosphorylation in DRG neurons (Ji et al. 1999
; Huang et al. 2000
) whereas administration of an inhibitor of ERK1/2 phosphorylation leads to pain alleviation (Ji et al. 1999
; Karim et al. 2001
). However, activation of μOR by opiates leads to ERK1/2 phosphorylation and antinociception (Fukuda et al. 1996
; Li and Chang 1996
; Gutstein et al. 1997
; Trapaidze et al. 2000
) and this can be potentiated by ketamine. This indicates that ERK1/2 plays a central role in pain perception that may be modulated by down-stream signaling events eventually leading to either nociception or antinociception.
In conclusion, our results implicate ERK1/2 signaling in ketamine–opioid interactions. They also suggest that events involved in ERK1/2 signaling may have a significant role in delineating the putative beneficial effects of ketamine in preemptive analgesia. As our results indicate that ketamine delays the desensitization and improves the resensitization of signaling via this pathway the overall effect would be in maintaining this pathway active for a longer time period. This could have profound effects on the physiologic responses that occur as a result of the activation of opioid receptors and could account for the observed effects of ketamine on the duration of opioid-induced analgesia.