| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
 
|
|
|
||
|
|
Anesth Analg 2003;97:494-505 © 2003 International Anesthesia Research Society
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Indirect evidence indicates that morphine-3-glucuronide (M3G) may contribute significantly to the neuro-excitatory side effects (myoclonus and allodynia) of large-dose systemic morphine. To gain insight into the mechanism underlying M3G’s excitatory behaviors, we used fluo-3 fluorescence digital imaging techniques to assess the acute effects of M3G (5–500 µM) on the cytosolic calcium concentration ([Ca2+]CYT) in cultured embryonic hippocampal neurones. Acute (3 min) exposure of neurones to M3G evoked [Ca2+]CYT transients that were typically either (a) transient oscillatory responses characterized by a rapid increase in [Ca2+]CYT oscillation amplitude that was sustained for at least
30 s or (b) a sustained increase in [Ca2+]CYT that slowly recovered to baseline. Naloxone-pretreatment decreased the proportion of M3G-responsive neurones by 10%–25%, implicating a predominantly non-opioidergic mechanism. Although the naloxone-insensitive M3G-induced increases in [Ca2+]CYT were completely blocked by N-methyl-D-aspartic acid (NMDA) antagonists and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) (alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid/kainate antagonist), CNQX did not block the large increase in [Ca2+]CYT evoked by NMDA (as expected), confirming that M3G indirectly activates the NMDA receptor. Additionally, tetrodotoxin (Na+ channel blocker), baclofen (
-aminobutyric acidB agonist), MVIIC (P/Q-type calcium channel blocker), and nifedipine (L-type calcium channel blocker) all abolished M3G-induced increases in [Ca2+]CYT, suggesting that M3G may produce its neuro-excitatory effects by modulating neurotransmitter release. However, additional characterization is required. IMPLICATIONS: Large systemic doses of morphine administered to some patients for cancer pain management have been reported to produce myoclonus and allodynia. Indirect evidence implicates the major morphine metabolite, morphine-3-glucuronide (M3G), in these neuro-excitatory side effects. Hence, this study was designed to gain insight into the cellular mechanism responsible for M3G’s neuro-excitatory actions.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The success of the World Health Organization’s recommendation of morphine as the drug of choice for the alleviation of moderate to severe cancer pain is best illustrated by the exponential increase in global morphine consumption from 1986–1995 (1). Not only are there more prescriptions for morphine being written, but larger morphine doses are also being prescribed (1) as a means of increasing the likelihood that patients with cancer pain will receive satisfactory pain relief. However, one unintended consequence of the increased prescribing of large systemic doses of morphine and its close structural analog, hydromorphone (HMOR), for cancer pain management is the now relatively frequent occurrence of opioid-related myoclonus and allodynia progressing to frank seizures in some patients (2).
Although the mechanistic basis for the neuro-excitatory consequences (myoclonus, allodynia, and seizures) of large-dose systemic morphine and HMOR is unknown, indirect evidence implicates the 3-glucuronide metabolites (2). After a systemic administration, more than 50% of every dose of morphine is metabolized to morphine-3-glucuronide (M3G), a metabolite that has no intrinsic pain relieving effects (3). Instead, intracerebroventricular (icv) M3G and its structural analog, hydromorphone-3-glucuronide (H3G), the major metabolite of HMOR, evoke a range of dose-dependent neuro-excitatory behaviors (4–6) including myoclonus, allodynia, and seizures in rodents. Although large icv doses of the parent opioid, morphine, produce behavioral excitation in rodents (5,7), morphine is at least 10-fold less potent as a neuro-excitant than M3G/H3G.
Importantly, M3G crosses the blood-brain barrier sufficiently that the mean steady-state cerebrospinal fluid (CSF) M3G concentration exceeds that of morphine by two- to 10-fold after chronic oral and subcutaneous morphine dosing in cancer patients (8–11). In the clinical setting, changing from large-dose systemic morphine to small-dose icv morphine ensures resolution of opioid-related neuro-excitatory behaviors in cancer patients. After initiation of small-dose icv morphine, there is an 
50-fold increase in the trough CSF morphine concentrations relative to the steady-state CSF morphine concentrations achieved during previous oral or subcutaneous morphine dosing (10). The icv-dosing route also facilitates clearance of M3G from patient CSF while maintaining pain relief through delivery of large concentrations of morphine into the lateral ventricle, i.e., adjacent to regions of the brain containing large densities of opioid receptors (10). These findings provide indirect evidence to suggest that the neuro-excitatory behaviors associated with large systemic doses of morphine are produced primarily by M3G and not the parent opioid (morphine) because the icv-dosing route bypasses liver/gut wall metabolism, and the brain has a small propensity for glucuronidating morphine (3).
Another empirically derived strategy for resolution of opioid-related behavioral excitation in cancer patients is opioid rotation, whereby the opioid medication is rotated from morphine/HMOR to a structurally dissimilar opioid, such as methadone or fentanyl (12,13). This facilitates clearance of M3G/H3G from patient CSF with subsequent resolution of the excitatory behaviors and restoration of analgesia with the dissimilar opioid. Additionally, stopping or markedly reducing the dose of systemic morphine/HMOR often results in temporal resolution of behavioral excitation (12) consistent with the clearance of M3G/H3G from patient CSF.
Previous behavioral studies in our laboratory indicate that M3G may evoke its neuro-excitatory behaviors via indirect activation of N-methyl-D-aspartic acid (NMDA) receptors (4) in a manner similar to that which occurs secondary to nerve or tissue injury (14). To gain insight into the cellular mechanism mediating the neuro-excitatory pharmacological effects of M3G, we have established primary cultures of embryonic rat hippocampal neurones as a model system, consistent with our behavioral studies in rats showing that the rat hippocampus has a range of subregional sensitivities to M3G (15). As the cytosolic calcium concentration ([Ca2+]CYT) plays a crucial role in neuronal processing, drugs that increase [Ca2+]CYT have the potential to profoundly amplify neurotransmission. Hence, we have used fluo-3 fluorescence digital imaging techniques to qualitatively assess the acute effects of M3G on [Ca2+]CYT in cultured hippocampal neurones as a means of gaining insight into the cellular mechanism(s) through which M3G evokes neuro-excitatory behaviors.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Ethical approval was obtained from The University of Queensland Animal Experimentation Ethics Committee. Pregnant Sprague-Dawley (SD) rats were purchased from the Herston Medical Research Centre, The University of Queensland (Brisbane, Australia) or from the Animal Resources Center (Perth, Western Australia). M3G, NMDA, alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), naloxone, Poly-D-lysine (PDL), papain, bovine serum albumin (BSA), dimethyl sulfoxide (DMSO), rabbit anti-glial fibrillary acidic protein (GFAP) antibody, mouse monoclonal anti-synaptophysin antibody, goat anti-rabbit immunoglobulin (Ig)G TRITC conjugate, goat anti-mouse IgG FITC conjugate, and all buffer reagents were obtained from Sigma (St Louis, MO). Tissue culture media and Hank’s balanced salt solution (HBSS) without calcium chloride were from Invitrogen (Melbourne, Australia). Fluo-3/AM (Fluo-3) was from Molecular Probes (Eugene, OR). MK-801 (dizocilpine) and CNQX (6-cyano-7-nitroquinoxaline-2,3-dione) were from ICN Biomedicals Inc (Aurora, OH). Tetrodotoxin (TTX) was obtained from Alomone Labs Ltd (Jerusalem, Israel). LY274614 (±-decahydro-6-(phosphonomethyl)-3-isoquinolinecarboxylic acid) was a generous gift from Eli Lilly Pty Ltd (Sydney, Australia). AM336 (
-conotoxin CVID) was a generous gift from AMRAD Corporation Ltd (Melbourne, Australia). Baclofen (Lioresal® intrathecal) vials (2 mg/mL) were purchased from the Royal Brisbane Hospital Pharmacy (Brisbane, Australia). -Conotoxin MVIIC was purchased from Peninsula Laboratories, Inc (San Carlos, CA). Nifedipine was obtained from Calbiochem (La Jolla, CA). Flunarizine was from BIOMOL Research Laboratories, Inc (PA). Midazolam was a generous gift from Roche Products Pty Ltd (Sydney, Australia). Primary cultures of rat embryonic hippocampal neurones, obtained from SD rats at 18–19 days of gestation, were prepared as described by Brewer et al. (16) with modifications. Briefly, hippocampi were dissociated by treatment with 2 mg/mL of papain/Hibernate-E for 30 min at 30°C, followed by trituration. Cells were pelleted and resuspended in neurobasal plating medium containing: L-glutamine (0.5 mM), glutamate (2.5 µM), 2% B-27®, and neurobasal medium. Cells were plated onto 25-mm round glass coverslips coated with PDL (0.05 mg/mL) at 6–7 x 105 cells/mL. Neurones were grown in a humidified atmosphere of 5% CO2 at 37°C. The medium was changed on Day 3 of culture by exchanging half of the medium with neurobasal feeding medium containing 2% B-27® and L-glutamine (0.5 mM). Thereafter, culture media were changed weekly.
Hippocampal neurones aged 14–18 days in vitro were used in these studies. Because experiments were performed over a long period and involved imaging at high rates of acquisition, our studies, similar to others, used the dye fluo-3 to assess Ca2+ levels in a semiquantitative manner. The use of ratio-metric dyes such as fura-2 would have led to significant phototoxicity in our model system and would not have allowed characterization of high frequency Ca2+ oscillations in individual neurones.
Neurones were loaded with 2 µM of fluo-3 for 45 min at 37°C in HEPES HHSS containing 3 mg/mL of BSA and incubated for an additional 15 min to promote hydrolysis of the AM ester. HHSS comprised the following (in mM): HEPES 20, NaCl 137, CaCl2 1.3, MgSO4 0.4, MgCl2 0.5, KCl 5.0, KH2PO4 0.4, Na2HPO4 0.6, NaHCO3 3.0, glucose 5.6, and 10 µM of glycine; the pH value was adjusted to 7.45 with NaOH. Dye-loaded cells were transferred to the recording chamber of an inverted microscope, and cells were superfused with HHSS, containing 0.9 mM of [Mg2+]o, in the presence or absence of naloxone (1 µM) at a rate of 2.5–3 mL/min for 15 min before each experiment. Because our preliminary studies showed that the acute effects of M3G to increase oscillations in [Ca2+]CYT involved both opioid and non-opioid mechanisms, with the latter predominating, the majority of the studies described herein were undertaken in the presence of an opioid receptor-saturating concentration of naloxone (1 µM) to facilitate characterization of the non-opioid mechanism.
[Ca2+]CYT was measured in the somatic region of neurones. All experiments were performed at room temperature (21°C ± 2°C). Neuronal responses were determined using at least three different coverslips from two different cultures, unless otherwise indicated; n values refer to the number of individual neurones assessed.
The fluorescence excitation wavelength (480 nm) was supplied by a Sutter DG-4 wavelength selector (Sutter Instrument Company, Novato, CA) and emission observed through a 550-nm band-pass filter (Omega Optical, Brattleboro, VT). A 40 x oil immersion objective was used. Data points were collected at 0.6-s intervals during rapid characterization or at 10- to 30-s intervals during prolonged incubation or washout periods to minimize bleaching, depending upon the protocol. Images were acquired using MetafluorTM imaging software (West Chester, PA).
Cells were fixed, and the same areas in which [Ca2+]CYT was determined were examined using fluorescence immunocytochemistry immediately after the measurement of fluo-3 fluorescence. Fixed cells were stained using rabbit polyclonal anti-GFAP and mouse monoclonal anti-synaptophysin. Primary antibodies were then treated with goat anti-mouse IgG (FITC) and goat anti-rabbit IgG (TRITC). Neurones were identified by their positive staining for synaptophysin and by the absence of staining for GFAP. Consistent with previous reports using neurobasal media (16), very few of the cells (<5%) were glia in our culture.
Coverslips containing cultured hippocampal neurones were placed in a microscope chamber for Ca2+ measurements as described above. Test compounds and drugs were added to the chamber either through a gravity-fed perfusion system or via direct addition into the chamber.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
By Days 14–18 in culture, the hippocampal neurones had formed extensive processes and synaptic connections. Consistent with these morphological observations,
76% of cultured neurones perfused with HHSS (0.9 mM of [Mg2+]o) exhibited spontaneous [Ca2+]CYT oscillations, whereas the remaining 24% of cultured neurones were quiescent. The acute (3-min exposure) effects of M3G (5–500 µM) on [Ca2+]CYT in cultured embryonic rat hippocampal neurones were assessed using state of the art high-speed fluo-3 fluorescence digital imaging techniques. Increases in [Ca2+]CYT evident as increases in fluo-3 fluorescence intensity were measured in the soma of neurones. Before the experiments, neurones were treated with a single concentration of M3G (50 µM) to assess their M3G-responsiveness. If the M3G-evoked increase in the amplitude of oscillations (as assessed by fluo-3 intensity) was more than two-fold when compared with pretreatment levels, these neurones were classified as M3G-responsive.
Superfusing M3G (50 µM) for 3 min resulted in a rapid increase in [Ca2+]CYT, which was always reversed upon M3G washout (Fig. 2E). In the experiment shown in Figure 1, three nonspontaneously oscillating neurones were sampled to determine the effects of M3G (50 µM) on [Ca2+]CYT, and the fluo-3 fluorescence intensity from three neuronal regions was plotted against time to show the M3G-evoked increases in [Ca2+]CYT in these neurones (Fig. 1E). Pseudo-color images of these same fluo-3 loaded cells before, during, and after M3G-mediated increases in [Ca2+]CYT are shown in Figure 1A–C, respectively, where hot colors, e.g., yellow-red, indicate a large increase in fluorescence intensity, which in turn reflects a large increase in [Ca2+]CYT. These same cells were then characterized using fluorescence immunocytochemistry, such that neurones were labeled for synaptophysin (green) and glia for GFAP (red) (Fig. 1D). As can be seen, very few (<5%) of the cells in our culture were glia, consistent with previous studies where neurones were cultured using neurobasal media (16). Importantly, any glia present in the culture were clearly stained with GFAP using our staining protocols.
|
|
The M3G-mediated [Ca2+]CYT transients were typically one of two types: transient oscillatory responses and slow recoverable responses. The first type of response was characterized by a rapid increase in [Ca2+]CYT oscillation amplitude, which was sustained for at least
30 s in the absence (Fig. 2A) and presence (Fig. 2B) of naloxone (1 µM). By contrast, in nonspontaneously oscillating neurones, M3G often increased [Ca2+]CYT, reaching a plateau concentration that slowly recovered back to baseline in the absence (Fig. 2C) and presence (Fig. 2D) of naloxone (1 µM). The amplitude of the M3G-evoked calcium responses in quiescent neurones (Figs. 2, C and D) seemed to exceed that produced in oscillating neurones (Figs. 2, A and B); however, quantitative comparison of these relative responses using fura-2 remains for future studies. The relative response of the same neurone to each of M3G (50 µM) and glutamate (10 µM) is shown in Figure 4C. The effect of M3G on [Ca2+]CYT responses in this neurone is clearly significant because glutamate is well documented to produce increases in [Ca2+]CYT in the micromolar range in hippocampal neurones (17,18).
|
Increasing the naloxone concentration to 50 µM (Fig. 2E) also failed to block M3G-evoked increases in [Ca2+]CYT. Those neurones that failed to exhibit an increase in [Ca2+]CYT after exposure to 50 µM of M3G in HHSS containing 1 µM of naloxone (positive control) were excluded from further experimentation.
In separate experiments, neurones were assessed for their responsiveness to 5, 50, and 500 µM concentrations of M3G in both the absence and presence of an opioid receptor saturating concentration of naloxone (1 µM). For these studies, all neurones tested in the experiments were included. There was a concentration-dependent increase in the percentage of neurones that responded to M3G (Fig. 2F), which was reduced by 10%–25% in the presence of naloxone (1 µM) (Fig. 2F). However, this decrease only reached statistical significance (P < 0.01; 
2 test) for neurones exposed to M3G (50 µM) (Fig. 2F).
Spontaneous [Ca2+]CYT oscillations were completely abolished by 15 min of pretreatment with LY274614 (50 µM), a competitive NMDA receptor antagonist, MK-801 (30 µM), a noncompetitive NMDA antagonist, and CNQX (10 µM), an AMPA/kainate receptor antagonist. Additionally, the ability of M3G (50 µM) to increase [Ca2+]CYT was blocked by pretreatment with these same concentrations of LY274614 (n = 58; Figs. 3,A and B), MK-801 (n = 17; Fig. 3C), and CNQX (n = 34; Fig. 3D). By contrast, CNQX did not prevent NMDA (125 µM)-evoked increases in [Ca2+]CYT, as expected (Fig. 3E; n = 22 neurones from two coverslips in the same culture). Importantly, subsequent removal of the glutamate receptor antagonists from the perfusion medium resulted in neurones again becoming responsive to M3G (50 µM), confirming that the effects of these antagonists were reversible and that viable neurones were studied throughout the experiments (Fig. 3A–E). As expected, the increase in [Ca2+]CYT in our cultured neurones induced by NMDA (125 µM) was blocked by pretreatment with MK-801 (30 µM; data not shown).
|
Exposure of neurones to extracellular Ca2+-free conditions (i.e., media containing EGTA (100 µM) and no CaCl2) abolished spontaneous calcium oscillations (Fig. 4A; n = 20 neurones from two coverslips in the same culture). In addition, the ability of M3G (50 µM) to evoke increases in [Ca2+]CYT was abolished (Fig. 4A).
The Na+ channel blocker, TTX (1 µM), abolished spontaneous oscillations in calcium (data not shown), consistent with previous studies showing that nerve terminal depolarization is critical to calcium entry to facilitate presynaptic vesicular release of glutamate (19–21). Additionally, TTX (1 µM) completely abolished M3G (50 µM)-evoked increases in [Ca2+]CYT, an effect that was completely reversible upon TTX washout (Fig. 4B; n = 41 neurones from four coverslips in two cultures).
Exposure of cultured hippocampal neurones to the P/Q-type calcium channel blocker, MVIIC (1 µM), for 15 min abolished spontaneous calcium oscillations (Fig. 4C) as well as completely blocking M3G-evoked increases in [Ca2+]CYT (n = 46; Fig. 4C). Consistent with previous observations in the literature (22), the effects of MVIIC were long lasting.
Preincubation with the N-type calcium channel blocker, AM336 (also known as omega-conotoxin CVID) (1 µM), for 2 min completely abolished spontaneous calcium oscillations (Fig. 4D). AM336 did not prevent but possibly attenuated M3G-evoked increases in [Ca2+]CYT (n = 41; Fig. 4D). Clearly, further quantitative investigation using fura-2 is required to determine the extent to which AM336 attenuates M3G’s actions.
After exposure of neurones to nifedipine, a dihydropyridine L-type calcium channel blocker, (10 µM) for 15 min, spontaneous calcium oscillations in cultured hippocampal neurones were completely abolished, as were M3G-evoked increases in [Ca2+]CYT (n = 28; Fig. 4E).
Pretreatment (15 min) of cultured hippocampal neurones with flunarizine (1 µM) had little or no effect on either spontaneous calcium oscillations or on M3G-evoked increases in [Ca2+]CYT (n = 35; Fig. 4F).
Preincubation of neurones with a -aminobutyric acid (GABA)B agonist, baclofen (10 µM), for 15 min before acute M3G (50 µM) exposure completely inhibited spontaneous calcium oscillations as well as M3G-mediated increases in [Ca2+]CYT (n = 29; Fig. 5A). By contrast, the increases in [Ca2+]CYT induced by exposure of neurones to AMPA (10 µM) were not prevented by pretreatment with baclofen (10 µM) (n = 35; Fig. 5B), as expected.
|
Exposure (15 min) of neurones to midazolam, the water-soluble benzodiazepine agonist, in a concentration of 10 µM seemed to attenuate but did not block M3G-induced increases in [Ca2+]CYT in cultured hippocampal neurones (n = 11; Fig. 5C). Clearly, further quantitative investigation of this issue is required.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In this study, we used high-speed fluo-3 fluorescence digital imaging techniques and primary cultures of embryonic rat hippocampal neurones to gain insight into the cellular mechanism(s) mediating the neuro-excitatory behaviors produced by M3G, the major metabolite of morphine in humans and rats. In agreement with previous studies, cultured hippocampal neurones were synaptically active, with
76% of neurones tested exhibiting spontaneous oscillations in [Ca2+]CYT (19,21,23) that were abolished when depolarization of the nerve terminal was prevented by TTX, in accordance with previous reports (19–21). Additionally, synaptic transmission was blocked by removal of extracellular calcium or by increasing extracellular magnesium, consistent with the documented role of NMDA receptors in mediating spontaneous glutamatergic synaptic transmission (24).
After acute (three minutes) exposure to M3G (5–500 µM), the proportion of M3G-responsive neurones increased in a concentration-dependent manner from 30% at 5 µM to 95% at 50 µM (Fig. 2F). Thus, an M3G concentration of 50 µM was chosen for the remainder of our investigations to ensure M3G-responsiveness of the majority of neurones tested. Although this concentration of M3G is 50-fold larger than the mean CSF M3G concentration (1 µM) found in cancer patients dosed chronically with large-dose systemic morphine at the time of ventriculostomy (10), these CSF M3G concentrations would have been significantly underestimated. This is because in many patients, the morphine dose was reduced substantially before ventriculostomy to resolve opioid-related confusion so patients could give informed consent (10). It should also be noted that the typical concentrations of morphine and other opioids (1–30 µM) used to characterize the inhibitory effects of opioids in electrophysiological studies are 25–800 times larger than the mean steady-state CSF morphine concentration (35 nM) determined in patients receiving oral or s.c. morphine for chronic pain (8,9). Almost certainly, the necessity to use larger concentrations of each of M3G and morphine in in vitro studies of the mechanisms underpinning their respective pharmacological effects reflects the fact that integration of simultaneous neuronal responses at multiple levels of the central nervous system occurs in vivo to produce a behavioral response.
The percentage of neurones that responded to M3G in the presence of an opioid receptor saturating concentration of naloxone (1 µM) was reduced by 10%–25% (Fig. 2F), indicating the predominantly non-opioidergic mechanism mediating M3G’s neuro-excitatory actions. Consistent with the low binding affinity of M3G for µ, 
, and 
-opioid receptors (half-maximal inhibitory concentration [IC50] >1 µM) in rat brain homogenate (25,26), a 50-fold increase in the naloxone concentration (from 1 µM to 50 µM), failed to block or attenuate M3G-evoked calcium responses in imaged neuronal cell bodies. These observations confirm our previous in vivo findings in rats that M3G’s neuro-excitatory behavioral effects are mediated by a predominantly non-opioid mechanism (15). Moreover, the naloxone-insensitivity of M3G’s actions has been confirmed in a very recently published electrophysiological study undertaken in rat spinal cord neurones (27). To characterize M3G’s non-opioid mechanism, all of our subsequent experiments were undertaken in the presence of an opioid receptor-saturating concentration of naloxone (1 µM).
The NMDA receptor is a ligand- and voltage-gated channel, which not only requires the binding of glutamate for activation, but also an initial AMPA-receptor mediated depolarization of the cell membrane. Depolarization of the neuronal membrane removes the Mg2+ block from the NMDA receptor ion channel, thereby permitting the influx of Ca2+ and Na+ ions into the cell (28). MK-801 (noncompetitive NMDA antagonist) and LY274614 (competitive NMDA antagonist) completely blocked spontaneous calcium oscillations, consistent with previous reports that these arise from glutamatergic synaptic transmission (19,21,23,24). Furthermore, LY274614 and MK-801 completely blocked the M3G-induced increases in [Ca2+]CYT in both quiescent and spontaneously oscillating cultured hippocampal neurones in accordance with our previous in vivo studies in rats, which showed that LY274614 dose-dependently attenuated the behavioral excitation evoked by icv M3G (4). However, as M3G does not bind to any of the known binding sites on the NMDA receptor complex (25), our current data suggest that M3G increases [Ca2+]CYT by a mechanism that involves indirect rather than direct activation of NMDA receptors. The failure of M3G to increase [Ca2+]CYT in the presence of NMDA antagonists also indicates that M3G does not promote calcium influx by directly opening postsynaptic voltage-gated calcium channels nor does it mobilize calcium from intracellular stores.
As CNQX (AMPA/kainate receptor antagonist) completely blocked M3G-evoked increases in [Ca2+]CYT, but had no effect (as expected) on the increase in [Ca2+]CYT produced by direct exposure of cultured neurones to NMDA (125 µM), this confirms that M3G indirectly activates the NMDA receptor. Additionally, the fact that CNQX blocked M3G-evoked increases in [Ca2+]CYT indicates that M3G does not directly depolarize postsynaptic cells because such depolarization of the postsynaptic cell membrane would have removed the Mg2+ block from the NMDA receptor, thereby allowing Ca2+ influx, which we would have detected as an increase in [Ca2+]CYT. Our findings are consistent with the results of a recent electrophysiological study that showed that M3G did not affect postsynaptic membrane conductance or excitability in slices of rat spinal cord (27).
As the GABAB agonist and P-type calcium channel blocker, baclofen, completely inhibited M3G-evoked increases in [Ca2+]CYT but had no effect on AMPA-induced increases in [Ca2+]CYT, it is clear that M3G does not directly interact with AMPA/kainate receptors, which is consistent with our previous radioligand binding work that showed M3G does not bind to the AMPA/kainate receptor (25). Thus, M3G’s site of action is upstream of the AMPA receptor in the signaling cascade that ultimately leads to indirect activation of AMPA and NMDA receptors. Given that micromolar concentrations of baclofen have been shown to inhibit calcium currents and synaptic transmission in cultured rat hippocampal neurones (29), it is possible that M3G increases [Ca2+]CYT by an augmentation of glutamate release. This proposal is further supported by a previous report showing that activation of presynaptic GABAB receptors reduces neurotransmitter release by inhibition of calcium channels in nerve terminals (30).
Exposure of neurones to TTX, as well as removal of extracellular calcium or increasing the extracellular magnesium (3 mM) concentration (data not shown) abolished M3G-responsiveness, indicating that influx of calcium from the extracellular environment rather than mobilization from intracellular stores is critical for M3G’s pharmacological action. These findings are consistent with the results of a recent study that used whole patch-clamp recordings in substantia gelatinosa neurones in transverse slices of rat spinal cord to show that in the presence of TTX, M3G reduced the miniature inhibitory postsynaptic current (mIPSC) frequency without affecting the amplitude in all cells tested, suggestive of a presynaptic mechanism (27).
We have discounted the possibility that M3G may evoke its neuro-excitatory effects by blocking postsynaptic GABAA or glycine receptors for the following reasons. First, we have previously reported that M3G has insignificant binding affinity for either the benzodiazepine or the GABA binding site on the GABAA-benzodiazepine receptor complex or for inhibitory glycine receptors in neuronal tissue (25). Second, the recently published electrophysiological study mentioned above (27) showed that M3G produced a concentration-dependent decrease in the amplitude of the evoked IPSC (27), such that M3G reduced the frequency but not the amplitude of TTX-insensitive mIPSCs, thereby implicating a presynaptic mechanism of action. Third, because midazolam (10 µM) seemed to attenuate, but did not block, M3G-induced increases in [Ca2+]CYT, it is possible that M3G interacts with one or more subtypes of metabotropic glutamate receptors to modulate GABA or glutamate release from nerve terminals (31); this remains for future examination.
Taken together, the previously mentioned findings collectively suggest that M3G does not have direct effects on the postsynaptic cell; rather, the evidence points to an upstream site of action such as the promotion of neurotransmitter release. This proposal is further supported by our findings that M3G-responsiveness of neurones was abolished when nerve terminal depolarization was blocked by TTX, thereby preventing Ca2+ influx into the nerve terminal, a step critical to neurotransmitter release. We further investigated this issue using selective blockers of each of N-, P/Q-, L-, and T-type calcium channels.
Consistent with previous reports (22), MVIIC (1 µM), the selective P/Q-type calcium channel blocker, produced long-lasting inhibitory effects. Specifically, MVIIC blocked spontaneous oscillations in calcium in accordance with reports that P-type calcium channels have an important role in the modulation of glutamatergic synaptic transmission in hippocampal neurones (32) at all stages of neuronal development. Furthermore, our observation that MVIIC also abolished M3G-evoked increases in [Ca2+]CYT, supports the possibility that M3G elicits its neuro-excitatory effects via indirect augmentation of glutamatergic neurotransmission.
AM336, the selective N-type calcium channel blocker (33), abolished spontaneous [Ca2+]CYT oscillations in a reversible manner, in accordance with the report by Verderio et al. (34) that N-type Ca2+ channels have a pivotal role in modulating presynaptic glutamate release in developing rat hippocampal neurones. However, although AM336 (1 µM) seemed to attenuate M3G’s effects, it did not completely block M3G-evoked increases in [Ca2+]CYT. Clearly, additional research is required to quantify AM336’s inhibitory effects.
Although the release of neurotransmitters is generally reported to be under the influence of both N and P/Q-type calcium channels, there are reports that also implicate L-type calcium channels in this process (35,36). In support of this, blockade of L-type calcium channels with nifedipine (10 µM) in the present study, abolished spontaneous oscillations in calcium as well as preventing M3G-evoked increases in [Ca2+]CYT. Furthermore, pretreatment of cultured hippocampal neurones with the T-type calcium channel blocker, flunarizine (1 µM), seemed to have little or no effect on either spontaneous oscillations in calcium or on M3G-induced increases in [Ca2+]CYT.
Thus, collectively, our findings show that selective blockade of N-, P/Q-, and L-type calcium channels abolishes spontaneous calcium oscillations, consistent with the well-accepted view that calcium influx into the nerve terminal via N-, P/Q-, and L-type calcium channels is critical to facilitate presynaptic neurotransmitter release. By contrast, only P/Q- and L-type calcium channel blockers completely prevented M3G-evoked increases in [Ca2+]CYT, suggesting that M3G may act via pathways that primarily rely on P/Q- and L-type calcium channels to indirectly modulate neurotransmitter release. Our findings described herein are consistent with a recent electrophysiological study undertaken in rat spinal cord slices (published while our manuscript was under review) showing that M3G produced a naloxone-insensitive suppression of inhibitory synaptic transmission via a presynaptic mechanism (27).
In summary, our studies in cultured hippocampal neurones show that M3G indirectly activates the NMDA receptor resulting in increased cytosolic calcium concentrations via a predominantly non-opioidergic mechanism. Our findings add further evidence to that published by Moran and Smith (27), who suggest that M3G produces its neuro-excitatory effects via indirect modulation of neurotransmitter release.
|
Acknowledgments |
|---|
Supported, in part, by the National Health and Medical Research Council of Australia and the Queensland Cancer Fund. KH is supported by an International Postgraduate Student Scholarship (IPRS) and a University of Queensland IPRS.
|
Footnotes |
|---|
Presented, in part, presented at the 2001 annual scientific meetings of the Australian Pain Society and the Australian Society for Clinical and Experimental Pharmacologists and Toxicologists.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
- University of Wisconsin Pain & Policy Studies Group. Global consumption of morphine. WHO Collaborating Center. Available from: http://www.medsch.wisc.edu/painpolicy/publicat/monograp/globm.htm. Accessed November 3, 2001.
- Smith MT. Neuroexcitatory effects of morphine and hydromorphone: evidence implicating the 3-glucuronide metabolites. Clin Exp Pharmacol Physiol 2000; 27: 524–8.[Web of Science][Medline]
- Milne RW, Nation RL, Somogyi AA. The disposition of morphine and its 3- and 6-glucuronide metabolites in humans and animals, and the importance of the metabolites to the pharmacological effects of morphine. Drug Metab Rev 1996; 28: 345–472.[Web of Science][Medline]
- Bartlett SE, Cramond T, Smith MT. The excitatory effects of morphine-3-glucuronide are attenuated by LY274614, a competitive NMDA receptor antagonist, and by midazolam, an agonist at the benzodiazepine site on the GABAA receptor complex. Life Sci 1994; 54: 687–94.[Web of Science][Medline]
- Labella FS, Pinsky C, Havlicek V. Morphine derivatives with diminished opiate receptor potency show enhanced central excitatory activity. Brain Res 1979; 174: 263–71.[Web of Science][Medline]
- Wright AW, Mather LE, Smith MT. Hydromorphone-3-glucuronide: a more potent neuro-excitant than its structural analogue, morphine-3-glucuronide. Life Sci 2001; 69: 409–20.[Web of Science][Medline]
- Urca G, Frenk H, Liebeskind J, Taylor A. Morphine and enkephalin: analgesic and epileptic properties. Science 1977; 197: 83–6.
[Abstract/Free Full Text] - Wolff T, Samuelsson H, Hedner T. Morphine and morphine metabolite concentrations in cerebrospinal fluid and plasma in cancer pain patients after slow-release oral morphine administration. Pain 1995; 62: 147–54.[Web of Science][Medline]
- Wolff T, Samuelsson H, Hedner T. Concentrations of morphine and morphine metabolites in CSF and plasma during continuous subcutaneous morphine administration in cancer pain patients. Pain 1996; 68: 209–16.[Web of Science][Medline]
- Smith MT, Wright AWE, Williams BE, et al. The CSF and plasma concentrations of morphine, M3G and M6G in patients before ventriculostomy and after initiation of intracerebroventricular morphine for cancer pain management. Anesth Analg 1999; 88: 109–16.
[Abstract/Free Full Text] - Goucke CR, Hackett LP, Ilett KF. Concentrations of morphine, morphine-6-glucuronide and morphine-3-glucuronide in serum and cerebrospinal fluid following morphine administration to patients with morphine-resistant pain. Pain 1994; 56: 145–9.[Web of Science][Medline]
- Sjogren P, Jensen NH, Jensen TS. Disappearance of morphine-induced hyperalgesia after discontinuing or substituting morphine with other opioid agonists. Pain 1994; 59: 313–6.[Web of Science][Medline]
- Vigano A, Fan D, Bruera E. Individualized use of methadone and opioid rotation in the comprehensive management of cancer pain associated with poor prognostic indicators. Pain 1996; 67: 115–9.[Web of Science][Medline]
- Mayer DJ, Mao J, Holt J, Price DD. Cellular mechanisms of neuropathic pain, morphine tolerance, and their interactions. Proc Natl Acad Sci USA 1999; 96: 7731–6.
[Abstract/Free Full Text] - Halliday AJ, Bartlett SE, Colditz P, Smith MT. Brain region-specific studies of the excitatory behavioral effects of morphine-3-glucuronide. Life Sci 1999; 65: 225–36.[Web of Science][Medline]
- Brewer GJ, Torricelli JR, Evege EK, Price PJ. Optimized survival of hippocampal neurons in B27-supplemented Neurobasal, a new serum-free medium combination. J Neurosci Res 1993; 35: 567–76.[Web of Science][Medline]
- Adamec E, Didier M, Nixon RA. Developmental regulation of the recovery process following glutamate-induced calcium rise in rodent primary neuronal cultures. Brain Res Dev Brain Res 1998; 108: 101–10.[Medline]
- Limbrick DD, Churn SB, Sombati S, Delorenzo RJ. Inability to restore resting intracellular calcium levels as an early indicator of delayed neuronal cell-death. Brain Res 1995; 690: 145–56.[Web of Science][Medline]
- Bacci A, Verderio C, Pravettoni E, Matteoli M. Synaptic and intrinsic mechanisms shape synchronous oscillations in hippocampal neurons in culture. Eur J Neurosci 1999; 11: 389–97.[Web of Science][Medline]
- Shen M, Piser TM, Seybold VS, Thayer SA. Cannabinoid receptor agonists inhibit glutamatergic synaptic transmission in rat hippocampal cultures. J Neurosci 1996; 16: 4322–34.
[Abstract/Free Full Text] - Tanaka T, Saito H, Matsuki N. Intracellular calcium oscillation in cultured rat hippocampal neurons: a model for glutamatergic neurotransmission. Jpn J Pharmacol 1996; 70: 89–93.[Medline]
- Gandia L, Lara B, Imperial JS, et al. Analogies and differences between omega-conotoxins MVIIC and MVIID: binding sites and functions in bovine chromaffin cells. Pflugers Arch 1997; 435: 55–64.[Web of Science][Medline]
- Monteith GR, Blaustein MP. Different effects of low and high dose cardiotonic steroids on cytosolic calcium in spontaneously active hippocampal neurons and in co-cultured glia. Brain Res 1998; 795: 325–40.[Web of Science][Medline]
- Ault B, Evans RH, Francis AA, et al. Selective depression of excitatory amino acid induced depolarizations by magnesium ions in isolated spinal cord preparations. J Physiol 1980; 307: 413–28.
[Abstract/Free Full Text] - Bartlett SE, Dodd PR, Smith MT. Pharmacology of morphine and morphine-3-glucuronide at opioid, excitatory amino acid, GABA and glycine binding sites. Pharmacol Toxicol 1994; 75: 73–81.[Web of Science][Medline]
- Loser SV, Meyer J, Freudenthaler S, et al. Morphine-6-O-beta-D-glucuronide but not morphine-3-O-beta-D-glucuronide binds to mu-, delta- and kappa-specific opioid binding sites in cerebral membranes. Naunyn Schmiedebergs Arch Pharmacol 1996; 354: 192–7.[Web of Science][Medline]
- Moran TD, Smith PA. Morphine-3beta-D-glucuronide suppresses inhibitory synaptic transmission in rat substantia gelatinosa. J Pharmacol Exp Ther 2002; 302: 568–76.
[Abstract/Free Full Text] - Ghosh A, Greenberg ME. Calcium signaling in neurons: molecular mechanisms and cellular consequences. Science 1995; 268: 239–47.
[Abstract/Free Full Text] - Scholz KP, Miller RJ. Gaba-B receptor-mediated inhibition of Ca2+ currents and synaptic transmission in cultured rat hippocampal-neurons. J Physiol 1991; 444: 669–86.
[Abstract/Free Full Text] - Travagli RA, Ulivi M, Wojcik WJ. Gamma-aminobutyric acid-B receptors inhibit glutamate release from cerebellar granule cells: consequences of inhibiting cyclic AMP formation and calcium influx. J Pharmacol Exp Ther 1991; 258: 903–9.
[Abstract/Free Full Text] - Dalezios Y, Lujan R, Shigemoto R, et al. Enrichment of mGluR7a in the presynaptic active zones of GABAergic and non-GABAergic terminals on interneurons in the rat somatosensory cortex. Cereb Cortex 2002; 12: 961–74.
[Abstract/Free Full Text] - Takahashi T, Momiyama A. Different types of calcium channels mediate central synaptic transmission. Nature 1993; 366: 156–8.[Medline]
- Lewis RJ, Nielsen KJ, Craik DJ, et al. Novel omega-conotoxins from Conus catus discriminate among neuronal calcium channel subtypes. J Biol Chem 2000; 275: 35335–44.
[Abstract/Free Full Text] - Verderio C, Coco S, Fumagalli G, Matteoli M. Calcium-dependent glutamate release during neuronal development and synaptogenesis: different involvement of omega-agatoxin IVA- and omega-conotoxin GVIA-sensitive channels. Proc Natl Acad Sci U S A 1995; 92: 6449–53.
[Abstract/Free Full Text] - Miller RJ. Presynaptic receptors. Annu Rev Pharmacol Toxicol 1998; 38: 201–27.[Web of Science][Medline]
- Igelmund P, Zhao YQ, Heinemann U. Effects of T-type, L-type, N-type, P-type, and Q-type calcium channel blockers on stimulus-induced pre- and postsynaptic calcium fluxes in rat hippocampal slices. Exp Brain Res 1996; 109: 22–32.[Web of Science][Medline]
This article has been cited by other articles:
|
|
 |
|
  M. P. Davis and M. Angst In Reply J. Clin. Oncol., March 20, 2008; 26(9): 1565 - 1565. [Full Text] [PDF]
|
|
|
|
 |
|
 E. Saboory, M. Derchansky, M. Ismaili, S. S. Jahromi, R. Brull, P. L. Carlen, and H. El Beheiry Mechanisms of Morphine Enhancement of Spontaneous Seizure Activity Anesth. Analg., December 1, 2007; 105(6): 1729 - 1735. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. W. Vaughan and M. Connor In Search of a Role for the Morphine Metabolite Morphine-3-Glucuronide Anesth. Analg., August 1, 2003; 97(2): 311 - 312. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
