- Email: [email protected]
Acute morphine administration and withdrawal from chronic morphine increase afterdepolarization amplitude in rat supraoptic nucleus neurons in hypothalamic explants
Neuropharmacology 61 (2011) 789e797
Contents lists available at ScienceDirect
Neuropharmacology journal homepage: www.elsevier.com/locate/neuropharm
Acute morphine administration and withdrawal from chronic morphine increase afterdepolarization amplitude in rat supraoptic nucleus neurons in hypothalamic explants Ming Ruan a, John A. Russell b, Colin H. Brown a, * a b
Centre for Neuroendocrinology and Department of Physiology, Otago School of Medical Sciences, University of Otago, P.O. Box 913, Dunedin 9054, New Zealand Centre for Integrative Physiology, University of Edinburgh, Edinburgh EH8 9XD, UK
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 January 2011 Received in revised form 3 May 2011 Accepted 18 May 2011
Supraoptic nucleus (SON) neurons secrete either oxytocin or vasopressin into the bloodstream from their axon terminals in the posterior pituitary gland. SON neurons are powerfully inhibited by the classical m-opioid receptor agonist, morphine. Oxytocin neurons develop morphine dependence when chronically exposed to this opiate, and undergo robust withdrawal excitation when morphine is subsequently acutely antagonized by naloxone. Morphine withdrawal excitation is evident as an increased firing rate and is associated with an increased post-spike excitability that is consistent with the expression of an enhanced post-spike afterdepolarization (ADP) during withdrawal. Here, we used sharp electrode recording from SON neurons in hypothalamic explants from morphine naïve and morphine treated rats to determine the effects of morphine on the ADP, and to test the hypothesis that morphine withdrawal increases ADP amplitude in SON neurons. Acute morphine administration (0.05e5.0 mM) caused a dosedependent hyperpolarization of SON neurons that was reversed by concomitant administration of 10 mM naloxone, or by washout of morphine; counter-intuitively, acute exposure to 5 mM morphine increased ADP amplitude by 78 11% (mean SEM). Naloxone-precipitated morphine withdrawal did not alter baseline membrane potential in SON neurons from morphine treated rats, but increased ADP amplitude by 48 11%; this represents a hyper-activation of ADPs because the basal amplitude of the ADP was similar in SON neurons recorded from explants prepared from morphine naïve and morphine treated rats. Hence, an enhanced ADP might contribute to morphine withdrawal excitation of oxytocin neurons. Ó 2011 Elsevier Ltd. All rights reserved.
Keywords: Afterdepolarization Afterhyperpolarization Intrinsic excitability Oxytocin Vasopressin
1. Introduction Morphine is the classical agonist of m-opioid receptors, which are widely expressed within the central nervous system (Mansour et al., 1995). However, relatively few neuronal phenotypes develop dependence when chronically exposed to morphine; these include hypothalamic supraoptic nucleus (SON) and paraventricular nucleus (PVN) magnocellular oxytocin neurons (Brown and Russell, 2004; Brown et al., 2000) that project to the posterior pituitary gland where they secrete oxytocin into the systemic circulation. Morphine acutely inhibits oxytocin neurons in vivo (Ludwig et al., 1997), but during chronic intracerebroventricular (ICV) administration, oxytocin neurons develop dependence on morphine. Dependence is revealed by a marked and sustained increase in firing
* Corresponding author. Tel.: þ64 3 479 7354; fax: þ64 3 479 7323. E-mail addresses: [email protected] (M. Ruan), [email protected] (J.A. Russell), [email protected] (C.H. Brown). 0028-3908/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuropharm.2011.05.025
rate of oxytocin neurons (and a consequent large increase in oxytocin secretion) upon morphine withdrawal in vivo (Bicknell et al., 1988; Blackburn-Munro et al., 2000; Brown et al., 1996, 1998; Bull et al., 2003) that is accompanied by increased Fos protein and oxytocin heteronuclear RNA expression in oxytocin neurons (Johnstone et al., 2000), as well as increased oxytocin release into the brain from oxytocin neuron somata and dendrites (Brown et al., 1997; Russell et al., 1992). Withdrawal excitation occurs in oxytocin neurons without a marked change in the activation of their major afferent inputs (Murphy et al., 1997). Moreover, oxytocin neurons are excited by direct administration of the opioid receptor antagonist, naloxone, into the SON of morphine treated rats (Johnstone et al., 2000; Ludwig et al., 1997). Thus, oxytocin neurons evidently develop dependence at a cellular level and provide a robust model that is amenable to detailed analysis of the cellular mechanisms of morphine dependence (Brown and Russell, 2004). By contrast to oxytocin neurons, SON vasopressin neurons do not show morphine dependence; after chronic morphine treatment, naloxone administration excites some vasopressin neurons
790
M. Ruan et al. / Neuropharmacology 61 (2011) 789e797
and inhibits others (Bicknell et al., 1988; Brown et al., 2005), so the overall effect is to induce only a modest increase in plasma vasopressin concentrations (Bicknell et al., 1988). Excitability of SON neurons is strongly influenced by nonsynaptic post-spike potentials (Brown, 2004; Brown and Bourque, 2006), including an afterdepolarization (ADP) and a medium afterhyperpolarization (mAHP). mAHP amplitude is reduced in SON neurons during morphine withdrawal (Brown et al., 2005), indicating that a reduced mAHP might contribute to the increased firing rate of oxytocin neurons evident during morphine withdrawal. However, the changes in post-spike excitability that accompany the withdrawal-induced increase in firing rate in vivo indicate that morphine withdrawal might also expose an ADP in oxytocin neurons (Brown et al., 2005), which could drive withdrawal excitation. Here, we used sharp electrode recording from SON neurons in hypothalamic explants from morphine naïve and morphine treated rats to determine the effects of chronic morphine on the ADP and to test the hypothesis that morphine withdrawal increases ADP amplitude in SON neurons. 2. Material and methods 2.1. Ethical approval All experimental procedures were approved by the University of Otago Animal Ethics Committee and were carried out in accordance with the recommendations of the Australian and New Zealand Council for the Care of Animals in Research and Teaching.
2.2. Electrophysiology Female Sprague-Dawley rats (200e350 g) were restrained in a soft plastic cone (5e10 s) and decapitated. The brains were rapidly removed and a block of tissue 8 8 2 mm (dorso-ventral) containing the basal hypothalamus was excised using razor blades and pinned, ventral surface up, to the silicone elastomer base of a superfusion chamber. Within 2e3 min, the excised hypothalamic explant was superfused (0.5e1.0 ml min1, 32e33 C) with carbogenated (95% oxygen and 5% carbon dioxide) artificial CSF (aCSF; see below) delivered via tubing placed over the medial tuberal region. The arachnoid membrane overlying the SON was removed using fine forceps, and a tissue paper wick was placed at the rostral tip of the explant to facilitate aCSF drainage. The aCSF (pH 7.4; 295 3 mOsmol kg1) consisted of (in mM): 120 NaCl, 3 KCl, 1.2 MgCl2, 26 NaHCO2, 2.5 CaCl2, and 10 glucose (Sigma). Intracellular recordings were made using sharp micropipettes prepared from glass capillaries (1.5 mm O.D. 0.86 mm I.D.) pulled on a P-97 FlamingeBrown puller (Sutter Instruments, Novato, CA). Micropipettes were filled with 2 M potassium acetate to yield DC resistances of 70e160 MU to a AgeAgCl wire electrode immersed in aCSF. Voltage recordings were obtained using an Axoclamp 2B amplifier (Molecular Devices, Foster City, CA) in continuous current-clamp (‘bridge’) mode. Digitized signals (10 kHz; DigiData 1320 Interface, Molecular Devices) were stored on a personal computer running pClamp9.2 (Molecular Devices) and analyzed offline. Recordings were made from SON neurons impaled with sharp electrodes in superfused hypothalamic explants. The cells had resting membrane potentials more negative than 50 mV, input resistances of >150 MU, and spike amplitudes of >60 mV when measured from baseline. Each cell displayed frequency-dependent spike broadening and transient outward rectification when depolarized from initial membrane potentials more negative than 75 mV (characteristics specific to SON neurons (Renaud and Bourque, 1991)). To determine the effects of acute morphine, and withdrawal from chronic morphine, on SON neuron membrane potential, recordings were made from an initial baseline membrane potential 5e10 mV below spike threshold (maintained by injection of hyperpolarizing current if necessary, with a constant current injected throughout the recording from each neuron). Depolarizing current injection (80 ms) was applied to elicit a train of 4e5 action potentials (the number of spikes in the trains was kept constant for each cell). Subthreshold ADPs were recorded from cells maintained at a baseline membrane potential approximately 5e10 mV below the spike threshold whereas spontaneous action potential afterdischarge during suprathreshold ADPs was measured from cells held approximately 2 mV below threshold. To determine the effects of the drugs on ADP and mAHP amplitude, as well as afterdischarge, measurements in the presence of drug were made from a baseline membrane potential between 1 mV more positive and 2 mV more negative than that used for pre-drug measurements.
2.3. Induction of morphine dependence Morphine dependence was induced as previously described (Rayner et al., 1988). Briefly, virgin female Sprague-Dawley rats (bodyweight 200e350 g) were anaesthetized with halothane (5% in O2). An Alzet model 2001 mini-osmotic pump (Charles River Ltd.) was placed subcutaneously and connected via polythene tubing to a 21 gauge stainless steel cannula implanted into the right lateral cerebral ventricle (3.0 mm caudal, 2.0 mm lateral to bregma and 4.5 mm below the surface of the skull). The pump and tubing were filled with morphine sulfate (Sigma) in sterile pyrogen-free water to deliver increasing doses (10 mg h1, 20 mg h1 for 40 h each and 50 mg h1 for the remainder at 1 ml h1) over 5 days. The cannula was secured using dental acrylic bonded to stainless steel screws inserted in the skull. Following surgery rats were housed individually with food and water available ad libitum. For electrophysiological recording (the sixth day following minipump implantation for morphine treated rats) in explants from morphine treated rats, the aCSF contained 5 mM morphine. 2.4. Statistics All averaged data are expressed as the mean the standard error of the mean (SEM). All differences within groups were evaluated with SigmaPlot software (SPSS Science, Chicago, IL, USA) using Student’s t-tests, paired t-tests or one-way repeated measures (RM) ANOVA, where appropriate. Where the F-ratio was significant, posthoc comparisons were completed as described in the figure legends.
3. Results 3.1. Effects of acute morphine and naloxone-precipitated withdrawal from chronic morphine on SON neuron membrane potential To determine the effects of acute morphine administration, membrane potential was measured during superfusion of 50 nM, 500 nM and 5 mM morphine followed by 10 mM naloxone (in the continued presence of morphine). Morphine caused a small,
Fig. 1. Acute morphine hyperpolarizes SON neurons in vitro. A, Example of the reversible, dose-dependent effects of morphine on SON neuron membrane potential, typical of all five neurons tested. The segments of recording shown were all made during injection of 25 pA of hyperpolarizing current and show the steady-state effects of each treatment on baseline membrane potential (relative to control: dashed line after >5 min superfusion of drug); continuous gap free recordings were interrupted to permit measurement of post-spike potentials. B, Change in membrane potential (mean SEM) of five SON neurons during superfusion of morphine (MOR) and morphine þ 10 mM naloxone (MOR þ NLX), showing a dose-dependent, naloxonereversible hyperpolarization of SON neurons by morphine (P ¼ 0.012, one-way RM ANOVA). *P < 0.05 versus pre-drug and MOR þ NLX, StudenteNewmaneKeuls posthoc tests.
M. Ruan et al. / Neuropharmacology 61 (2011) 789e797
791
dose-dependent hyperpolarization in all five SON neurons tested (P ¼ 0.01, one-way RM ANOVA; Fig. 1), similar to that reported previously in response to a m-opioid receptor agonist, endomorphin-1, in SON neurons in hypothalamic slices (Doi et al., 2001). The morphine-induced hyperpolarization was reversed by administration of 10 mM naloxone. In three other neurons, naloxone alone (10 mM) did not consistently alter membrane potential (0.7 1.0 mV change; P ¼ 0.54, paired t-test). Membrane potential was measured in nine neurons in explants from morphine treated rats during superfusion of 10 mM naloxone (in the continued presence of 5 mM morphine and with a constant hyperpolarizing current throughout each recording). Similar to the effects of naloxone in SON neurons from morphine naïve rats, naloxone did not consistently alter membrane potential (1.2 mV change, measured w10 min after the onset of naloxone administration; P ¼ 0.17, paired t-test).
3.2. Effects of acute morphine treatment on the ADP and mAHP in SON neurons To determine the effect of morphine on ADPs, we recorded SON neurons from morphine naïve rats (n ¼ 13) during superfusion of increasing doses of acute morphine (50e5 mM). Trains of four or five spikes were elicited every 20 s by an 80 ms depolarizing current pulse (þ100e300 pA, from 5 mV to 10 mV below spike threshold) to evoke post-train ADPs (and mAHPs). Paradoxically, ADP amplitude was increased by acute morphine superfusion in SON neurons and this effect was reversed by co-administration of 10 mM naloxone (P ¼ 0.008, one-way RM ANOVA; Fig. 2). In four of 13 neurons that did not express a measurable ADP (i.e. <0.5 mV) under basal conditions, an ADP (>0.5 mV) was exposed during superfusion of 500 nM morphine (n ¼ 1) or 5 mM morphine (n ¼ 3). Concomitant with an increase in ADP amplitude, morphine reduced mAHP amplitude, and this effect was partially reversed by co-administration of 10 mM naloxone (n ¼ 23; P < 0.001, one-way RM ANOVA; Fig. 2). In each of five neurons in which spontaneous action potential afterdischarge was superimposed upon evoked suprathreshold ADPs, superfusion of morphine (50 nM, n ¼ 4 or 5 mM, n ¼ 1) increased the number of spikes in the afterdischarge from 1 spike to 4 1 spikes (P ¼ 0.01, paired t-test; Fig. 2).
3.3. Effects of chronic ICV morphine treatment on the ADP and mAHP in SON neurons
Fig. 2. Acute morphine increases ADP amplitude in SON neurons in vitro. A, ADPs and mAHPs (averages of 10 traces) that each follow a 5-spike train (arrowheads, spikes truncated) evoked by a 80 ms depolarizing pulse (þ200 pA) in a SON neuron before (Pre-drug, left) and during superfusion of 50 nM morphine (MOR, middle), and during superfusion of 10 mM naloxone (NLX) in the continued presence of morphine, (right). In A, the ADP amplitude is the distance between the middle pair of dotted lines before morphine (left) and after morphine þ naloxone (right); the ADP amplitude was increased by morphine, as shown by the distance between the middle pair of dotted lines (centre). B, ADP amplitude in SON neurons during superfusion of increasing doses of morphine, showing that acute morphine treatment causes a naloxone-reversible increase in ADP amplitude (P ¼ 0.014, one-way RM ANOVA). C, mAHP amplitude in SON neurons during superfusion of increasing doses of morphine, showing that acute morphine treatment causes a decrease in mAHP amplitude; the effect of morphine was partially reversed by naloxone (P < 0.001, one-way RM ANOVA). *P < 0.05, **P < 0.01 and ***P < 0.001 versus pre-drug, StudenteNewmaneKeuls post-hoc tests. D, Posttrain afterdischarge superimposed on ADPs before and during superfusion of
To determine the effects of chronic morphine treatment on the ADP, rats were treated with ICV morphine for five days and SON neurons recorded from hypothalamic explants superfused with aCSF containing 5 mM morphine. The proportion of SON neurons that expressed a measurable ADP (>0.5 mV) was not significantly different (P ¼ 0.19, Chi-square test) between morphine naïve rats (14 of 27 neurons; 52%) and morphine treated rats (13 of 17 neurons; 76%). Furthermore, the amplitude of the ADP (in SON neurons that expressed a measurable ADP) was not different between morphine naïve (1.7 0.3 mV, n ¼ 14) and morphine treated rats (1.6 0.3 mV, n ¼ 13; P ¼ 0.77; Fig. 3). Similarly, the amplitude of the post-train mAHP was not different between SON neurons from morphine naïve rats (5.3 0.4 mV, n ¼ 27) and morphine treated rats (5.1 0.6 mV, n ¼ 17; P ¼ 0.73; Fig. 3).
morphine. E, Mean number of spikes in the afterdischarge before and during superfusion of morphine (n ¼ 5 cells; *P < 0.05, paired t-test).
792
M. Ruan et al. / Neuropharmacology 61 (2011) 789e797
3.4. Effects of naloxone-precipitated morphine withdrawal on the ADP and mAHP in SON neurons We have previously shown in vivo that naloxone-precipitated morphine withdrawal induces changes in post-spike excitability of SON oxytocin (and vasopressin) neurons that might be explained by an increase in the amplitude of the ADP (Brown et al., 2005). Therefore, we tested the effects of naloxone on the ADP in SON neurons from which the membrane potential was recorded in hypothalamic explants prepared from morphine treated rats and maintained in aCSF containing 5 mM morphine. By contrast to its effects in explants from morphine naïve rats (Fig. 2), superfusion of naloxone (10 mM, in the continued presence of 5 mM morphine to prevent morphine withdrawal after removal and during preparation of the hypothalamic explant from morphine treated rats) caused a consistent increase in ADP amplitude (þ0.5 0.1 mV) in neurons from morphine treated rats (n ¼ 7; P ¼ 0.005, paired t-test; Fig. 4). In two of three further neurons that did not express a measurable ADP (<0.5 mV) before naloxone, naloxone administration exposed an ADP (Fig. 4). This enhancement of the ADP evident during naloxone-precipitated morphine withdrawal is consistent with the increased post-spike excitability observed during morphine withdrawal excitation in vivo (Brown et al., 2005). In one neuron in which spontaneous action potential afterdischarge was superimposed on suprathreshold evoked ADPs,
Fig. 3. Chronic morphine in vivo does not alter ADP amplitude in SON neurons in vitro. A, ADPs and mAHPs (averages of 10 traces) that each follow a 5-spike train (arrowheads, spikes truncated) evoked by a 80 ms depolarizing pulse (þ150 and þ300 pA, respectively) in SON neurons from a morphine naïve rat (left) and a morphine treated rat (right; 5 mM morphine was present throughout recording of the neuron from the morphine treated rat). B, ADP amplitude in SON neurons from morphine naïve rats (left; n ¼ 14) and morphine treated rats (right; n ¼ 13), showing that chronic morphine treatment does not alter ADP amplitude. C, mAHP amplitude in SON neurons from morphine naïve rats (left; n ¼ 27) and morphine treated rats (right; n ¼ 17), showing that, similar to its lack of effect on the ADP, chronic morphine treatment does not alter mAHP amplitude.
superfusion of naloxone increased the number of spikes in the afterdischarge from 1 spike to 2 1 spikes (P < 0.001, Student’s t-test; n ¼ 10 contiguous measurements in each condition). Concomitant with an increase in ADP amplitude, naloxone reduced mAHP amplitude in neurons from morphine treated rats (by 0.9 0.4 mV; n ¼ 10; P ¼ 0.04; Fig. 4), similarly to our previous results from perforated patch experiments in hypothalamic slices (Brown et al., 2005). 3.5. Effects of morphine and morphine withdrawal on the mAHP and sAHP in SON neurons In SON neurons from morphine naïve rats, the sAHP (the average membrane potential measured between 1 and 2 s after the
Fig. 4. Morphine withdrawal increases ADP amplitude in SON neurons in vitro. A and B, ADPs and mAHPs (averages of 10 traces) that each follow a 5-spike train (arrowheads, spikes truncated) evoked by a 80 ms depolarizing pulse (þ200 pA) before (PreNLX, left) and during superfusion of 10 mM naloxone (NLX, right) in a hypothalamic explant prepared from a morphine treated rat (5 mM morphine was present throughout). Note that in A, the ADP was increased by naloxone and that in B, an ADP was exposed by naloxone (dotted lines: control, Pre-NLX values). C, ADP amplitude in individual SON neurons from morphine treated rats (n ¼ 10) before (Pre-NLX; left) and during (NLX; right) superfusion of 10 mM naloxone in the presence of 5 mM morphine, showing that naloxone consistently increased ADP amplitude. D, mAHP amplitude in the same cells from C, showing that naloxone consistently decreased mAHP amplitude concomitant with its effects on the ADP. *P < 0.05 and **P < 0.01, paired t-test.
M. Ruan et al. / Neuropharmacology 61 (2011) 789e797
end of a 25-spike train to avoid potential confounding effects of the faster mAHP; n ¼ 24; Fig. 5A) was not consistently affected by acute morphine or naloxone (maximum change þ85.2 115.5%; P ¼ 0.49; Fig. 5B). The amplitude of the sAHP was not different between SON neurons from morphine naïve rats (n ¼ 27) and from morphine treated rats (n ¼ 14; P ¼ 0.82; Fig. 5D and E). In SON neurons from morphine treated rats, the amplitude of the sAHP was not affected by naloxone (n ¼ 7; P ¼ 0.72; Fig. 5G and H). In the neurons from morphine naïve rats, the combined mAHP and sAHP (the maximum negative membrane potential following the 25-spike train) was not consistently affected by acute morphine or naloxone (Fig. 5A); while one-way ANOVA was significant (P ¼ 0.03), all pairwise post-hoc analyses using StudenteNewmaneKeuls tests
793
did not identify any significant differences between any of the groups (P 0.09; Fig. 5C). The amplitude of the combined mAHP and sAHP was not different between SON neurons from morphine naïve rats and from morphine treated rats (P ¼ 0.44; Fig. 5D and F). In SON neurons from morphine treated rats, the amplitude of the combined mAHP and sAHP was not affected by naloxone (P ¼ 0.79; Fig. 5G and I). 3.6. Effects of morphine and morphine withdrawal on the fAHP in SON neurons In SON neurons from morphine naïve rats, fAHP amplitude (measured from spike threshold of the first spike in the evoked train; n ¼ 23) was decreased by acute morphine (maximum
Fig. 5. Effects of morphine and morphine withdrawal on the mAHP and sAHP in SON neurons in vitro. A, Post-train sAHPs (evoked by a 25-spike train; spikes truncated, arrowhead) before (left) and during (middle) 5 mM morphine and 10 mM naloxone in 5 mM morphine (right), showing no effect of morphine on the mAHP and sAHP in SON cells from morphine naïve rats. B, Mean membrane potential in SON cells from morphine naïve rats measured 1e2 s after the 25-spike train (to avoid possible confounding effects of the mAHP) before (Predrug) and during increasing doses of morphine (MOR) and morphine plus 10 mM naloxone (MOR þ NLX). C, Mean mAHP þ sAHP amplitude immediately after the 25-spike train, before and during increasing doses of morphine and morphine plus 10 mM naloxone in SON cells from morphine naïve rats. D, Post-train mAHP and sAHPs in SON neurons from morphine naïve rats (left) and morphine treated rats (in SON neurons from morphine naïve and morphine treated rats (right)). E, Mean membrane potential (measured 1e2 s after the 25-spike train) in SON neurons from morphine naïve and morphine treated rats. F, Mean mAHP þ sAHP amplitude immediately after the 25-spike train in SON neurons from morphine naïve and morphine treated rats. G, Post-train mAHPs and sAHPs in SON neurons from a morphine treated rat before (left) and during (right) superfusion of 10 mM naloxone, showing no effect of naloxone on the sAHP. H, Mean membrane potential (measured 1e2 s after the 25-spike train) in SON neurons from morphine treated rats before and during naloxone (NLX) superfusion. I, Mean mAHP þ sAHP amplitude immediately after the 25-spike train in SON neurons from morphine treated rats before and during naloxone (NLX) superfusion.
794
M. Ruan et al. / Neuropharmacology 61 (2011) 789e797
change 9.4 2.2%; P ¼ 0.01; Fig. 6A). The amplitude of the fAHP was greater in SON neurons from morphine naïve rats (n ¼ 27) than in those from morphine treated rats (n ¼ 17; P ¼ 0.002; Fig. 6B). In SON neurons from morphine treated rats, the amplitude of the
fAHP was increased by naloxone (by 1.6 0.3 mV, n ¼ 10; P < 0.001; Fig. 5E) and the sAHP was not affected by naloxone (n ¼ 7; P ¼ 0.72; Fig. 6C). 3.7. Effects of morphine and morphine withdrawal on spike properties of SON neurons In SON neurons from morphine naïve rats, the amplitude (measured from baseline) of the first spike in the evoked train (n ¼ 23) was not significantly affected by morphine or naloxone (maximum change 4.3 1.6%; P ¼ 0.24, one-way RM ANOVA; Fig. 7A). The half-width of the first spike in the evoked train (n ¼ 23) was not affected by morphine (maximum change þ1.2 2.9%), but was increased by naloxone co-administration with morphine (þ12.6 3.5% increase; P ¼ 0.01; Fig. 7A). Spike broadening (the percentage increase in spike half-width across a five-spike train; n ¼ 20) was not affected by acute morphine or naloxone (maximum change þ3.2 2.0%; P ¼ 0.40; Fig. 7A). Spike amplitude was not significantly different between SON neurons from morphine naïve rats (n ¼ 27) and morphine treated rats (n ¼ 17; P ¼ 0.16, Student’s t-test; Fig. 7B). By contrast spike half-width was narrower in SON neurons from morphine naïve rats (n ¼ 27) than in those from morphine treated rats (n ¼ 17; P ¼ 0.009; Fig. 7B). Nevertheless, spike broadening was similar in SON neurons from morphine naïve rats (n ¼ 23) and from those in morphine treated rats (n ¼ 17; P ¼ 0.71; Fig. 7B). The amplitude of the first spike in the evoked train was not significantly affected by naloxone in SON neurons from morphine treated rats (n ¼ 10; P ¼ 0.58; Fig. 7C). Similarly, spike half-width was not significantly affected by naloxone in SON neurons from morphine treated rats (n ¼ 10; P ¼ 0.48; Fig. 7C). However, consistent with our previous observations (Brown et al., 2005), the proportionate increase in spike half-width across a five-spike train was increased by naloxone (by 3.7 1.3%; n ¼ 6; P ¼ 0.04; Fig. 7C), which might underpin the increased ADP evident during morphine withdrawal. 4. Discussion
Fig. 6. Effects of morphine and morphine withdrawal on the fAHP in SON neurons in vitro. A, Post-spike (spike truncated, arrowhead) fAHPs (left panel) before (solid line) and during superfusion of 5 mM morphine (dotted line) and 10 mM naloxone in the continued presence of 5 mM morphine (dashed line), showing a decrease in fAHP amplitude in the presence of morphine; for the sake of clarity, fAHPs in the presence of 50 nM and 500 nM morphine have been omitted. Right panel: mean fAHP amplitude before (Pre-drug) and during superfusion of various doses of morphine (MOR) and morphine plus 10 mM naloxone (MOR þ NLX). *P < 0.05 versus pre-drug, StudenteNewmaneKeuls post-hoc tests. B, fAHPs (left panel) from SON neurons from morphine naïve rats (solid line) and morphine treated rats (in the presence of 5 mM morphine; dotted line) showing a decrease in fAHP amplitude after chronic morphine administration. Right panel: mean fAHP amplitude in SON neurons from morphine naïve and morphine treated rats. **P < 0.01, Student’s t-test. C, fAHPs (left panel) from a SON neuron from a morphine treated rat before (solid line) and during (dotted line) superfusion of 10 mM naloxone, showing an increase in fAHP amplitude in naloxone. Right panel: mean fAHP amplitude in SON neurons from morphine treated rats before and during naloxone superfusion. ***P < 0.001, paired t-test.
Using sharp electrode recordings from hypothalamic explants, we have revealed complex effects of morphine and naloxone-induced morphine withdrawal on the electrophysiological properties of SON neurons from morphine naïve rats and morphine-dependent rats. While acute morphine administration increased ADP amplitude, ADP amplitude was similar in SON neurons recorded from explants prepared from morphine naïve rats and rats chronically administered morphine, indicative of the development of tolerance to the effects of morphine on the ADP. Surprisingly, we found that naloxone-precipitated withdrawal from chronic morphine also increased ADP amplitude. Such changes in ADP amplitude likely contribute to the increased excitability of oxytocin neurons evident in vivo after administration of low doses of m-opioid receptor agonists (Wakerley et al., 1983), or during withdrawal from chronic morphine administration (Bicknell et al., 1988). 4.1. Acute and chronic morphine effects on the electrophysiological properties of SON neurons Acute morphine administration consistently inhibits oxytocin neuron firing rate in vivo (Ludwig et al., 1997). Consistent with this observation, we found that morphine caused a small but consistent membrane hyperpolarization in all SON neurons tested. Counterintuitively, we also found that acute morphine administration increased ADP amplitude, which is expected to be excitatory, as seen by the increased spike afterdischarge evident during morphine
M. Ruan et al. / Neuropharmacology 61 (2011) 789e797
795
Fig. 7. Effects of morphine and morphine withdrawal on spike properties of SON neurons in vitro. A, Left: Spikes before (solid line) and during superfusion of 5 mM morphine (dotted line) and 10 mM naloxone in the continued presence of 5 mM morphine (dashed line) in SON cells from morphine naïve rats; for the sake of clarity, spikes in the presence of 50 nM and 500 nM morphine have been omitted. Histograms: left panel: mean spike amplitude of the first spike in a five-spike train before (Pre-drug) and during superfusion of increasing doses of morphine (MOR) and morphine plus 10 mM naloxone (MOR þ NLX) in SON cells from morphine naïve rats. Middle panel: mean spike half-width before and during superfusion of various doses of morphine and morphine plus naloxone. Right panel: mean spike width of the fifth spike in a five-spike train before and during superfusion of various doses of morphine and morphine plus naloxone. **P < 0.01 versus pre-drug, StudenteNewmaneKeuls post-hoc test. B, Left: Spikes in SON neurons from morphine naïve rats (solid line) and morphine treated rats (dotted line). Histograms: left, mean spike amplitude of the first spike in a five-spike train in SON neurons from morphine naïve rats and morphine treated rats. Middle: mean spike width of the first spike in a five-spike train in SON neurons from morphine naïve rats and morphine treated rats. Right: mean spike width of the fifth spike in a five-spike train in SON neurons from morphine naïve rats and morphine treated rats. **P ¼ 0.01, Student’s t-test. C, Left: Spikes from an SON neuron from a morphine treated rat before (solid line) and during (dotted line) superfusion of 10 mM naloxone. Histograms: left, mean spike amplitude of the first spike in a five-spike train from SON neurons from morphine treated rats before and during superfusion of 10 mM naloxone (NLX). Middle: mean spike width of the first spike in a five-spike train from SON neurons from morphine treated rats before and during superfusion of 10 mM naloxone. Right: mean spike width of the fifth spike in a five-spike train from SON neurons from morphine treated rats before and during superfusion of 10 mM naloxone. *P < 0.05, paired t-test.
administration. The morphine-induced enhancement of the ADP was evident at lower doses than those required to induce a hyperpolarization, which might explain the previous observation of morphine excitation of a small proportion of SON neurons in vitro (Wakerley et al., 1983). The morphine-induced enhancement of the ADP was likely mediated by m-opioid receptors because k-opioid agonists decrease ADP amplitude (Brown and Bourque, 2004; Brown et al.,1999, 2006) and SON neurons do not express d-opioid receptors
in morphine naïve rats or morphine-dependent rats (Sumner et al., 1990). Given that the mAHP was reduced by morphine concomitant with an increased ADP following a 4- or 5-spike train but that the combined mAHP and sAHP following a 25-spike train was not affected by morphine, it appears likely that the principal actions of acute morphine were on the ADP, rather than the mAHP. The selective m-opioid agonist D-Ala(2), N-CH(3)-Phe(4), Gly(5)ol-enkephalin (DAGO), decreases the frequency (but not the
796
M. Ruan et al. / Neuropharmacology 61 (2011) 789e797
amplitude or kinetics) of excitatory post-synaptic currents (Liu et al., 1999), indicating that glutamatergic inputs to SON neurons are under presynaptic m-opioid inhibition. Furthermore, morphine inhibits excitatory noradrenaline release within the SON in vivo (Onaka et al., 1995). Here, we found that morphine consistently hyperpolarized SON neurons, similar to the effects of the specific mopioid agonist, endomorphin-1 (Doi et al., 2001). Overall, we conclude that inhibition of excitatory inputs and direct steady-state hyperpolarization likely explain the inhibition by morphine of oxytocin neurons evident in vivo because the excitatory effects of an enhanced ADP will be evident only when a cell is active. 4.2. Morphine withdrawal effects on the electrophysiological properties of SON neurons During morphine withdrawal in vivo, oxytocin neurons increase their firing rate, which underpins a massive and prolonged secretion of oxytocin into the circulation and the brain (Brown and Russell, 2004). While acute morphine administration enhanced ADP amplitude, after five days of morphine treatment, the amplitude of the ADP was indistinguishable from that of morphine naïve rats, indicative of the development of tolerance to the effects of morphine on the ADP. Remarkably, morphine withdrawal did not cause a rebound inhibition of the ADP but rather had an excitatory effect similar to that of acute morphine exposure. However, the mechanisms that underpin each of these excitations might be quite different because (unlike morphine withdrawal) acute morphine did not alter spike broadening. The SON contains both oxytocin and vasopressin neurons. By contrast to oxytocin neurons, vasopressin neurons do not undergo morphine withdrawal excitation; naloxone administration to morphine-dependent rats does not consistently change vasopressin neuron firing rate and only modestly increases vasopressin secretion (Bicknell et al., 1988; Brown et al., 2005). Nevertheless, morphine withdrawal increased the ADP in all SON neurons tested in vitro, similar to our previous observations of a reduced mAHP and transient outward rectification in all SON neurons during morphine withdrawal (Brown et al., 2005). The failure of a consistently enhanced ADP (as seen here in vitro) to precipitate robust morphine withdrawal-induced increases in vasopressin neuron firing in vivo further supports our conclusion that an enhanced ADP likely does not alone drive withdrawal-induced increases in oxytocin neuron firing rate in vivo. However, it should be noted that vasopressin neurons are subject to activity-dependent autocrine inhibition by the k-opioid peptide, dynorphin, via inhibition of the ADP (Brown, 2004; Brown and Bourque, 2006). Because there is no cross-tolerance between mand k-opioid receptors in morphine-dependent rats (Pumford et al., 1993), activity-dependent k-opioid inhibition of vasopressin neuron (but not oxytocin neuron) ADPs will still occur during spontaneous activity, opposing the effects of ADP enhancement induced by morphine withdrawal in vasopressin neurons. In addition to k-opioid inhibition, we have also demonstrated that adenosine (probably generated from breakdown of ATP), also inhibits the activity of vasopressin neurons via activity-dependent activation of A1 receptors (Bull et al., 2006; Ruan and Brown, 2009), which would further oppose ADP enhancement induced by morphine withdrawal in vasopressin neurons. While we focused here on the effects of morphine withdrawal on the ADP, this was not the only property of SON neurons affected by morphine withdrawal. Concomitant with an increased ADP, the mAHP was decreased by morphine withdrawal, while the sAHP was not changed during morphine withdrawal. Because the mAHP and ADP overlap temporally (Greffrath et al., 1998; Ruan and Brown, 2009), and as we did not isolate the ADP, we cannot exclude the
possibility that inhibition of the mAHP might underpin withdrawal-induced enhancement of the ADP. However, there was no correlation between the change in mAHP amplitude and the change in ADP amplitude (following a 4- or 5-spike train) induced by withdrawal (Pearson Correlation coefficient, r ¼ 0.06; P ¼ 0.87) and the amplitude of the combined mAHP and sAHP (following a 25-spike train) was not affected by withdrawal, so it is likely that the withdrawal-induced enhancement of the ADP cannot be fully accounted for by inhibition of the mAHP. The fAHP (the hyperpolarizing phase of spike repolarization) was increased in SON neurons during morphine withdrawal. Intuitively, this would be expected to decrease excitability. However, the fAHP has been shown to increase firing rate in SON neurons via activation of a hyperpolarization-activated inward current (Ghamari-Langroudi and Bourque, 2000). Hence, an increased fAHP might also contribute to withdrawal-induced increases in firing rate. While baseline membrane potential, spike amplitude and spike half-width of SON neurons were not altered during morphine withdrawal, activity-dependent spike broadening was increased during withdrawal. Spike broadening is caused by Ca2þ influx (Bourque and Renaud, 1985) and the ADP is generated by a Ca2þ dependent (Li and Hatton, 1997) non-specific cation channel (Ghamari-Langroudi and Bourque, 2002), indicating that the increased Ca2þ influx at higher firing rates during withdrawal might contribute to the enhancement of the ADP. While spike broadening increased by only 4% and the ADP increased by 48%, the effect of Ca2þ influx on ADP amplitude is amplified by Ca2þ-induced Ca2þ release from intracellular stores (Li and Hatton, 1997). Furthermore, while spike half-width was not increased by withdrawal itself, spikes were broader in neurons from morphine treated rats (before withdrawal) compared to morphine naïve neurons. Hence, the modest increase in spike broadening during withdrawal was superimposed on an already broadened spike. If increased Ca2þ influx does underpin the withdrawal-induced enhancement of ADP amplitude, an increased ADP might be a consequence of withdrawal excitation, rather than a primary cause. 4.3. Consequences of morphine withdrawal excitation of magnocellular oxytocin neurons Although magnocellular oxytocin neurons (that project to the posterior pituitary gland) provide a robust model to study cellular morphine dependence (Brown and Russell, 2004), they might also play a role in the behavioral opiate withdrawal syndrome. Central oxytocin inhibits the development of morphine tolerance and attenuates the behavioral morphine withdrawal syndrome in mice (Kovacs et al., 1985). Centrally released oxytocin can originate from magnocellular neurosecretory neuron dendrites, being secreted by exocytosis (Ludwig and Leng, 2006). Central oxytocin antagonist treatment reduces withdrawal excitation of oxytocin neurons (Brown et al., 1997) and oxytocin release into the cerebrospinal fluid by magnocellular neurons is increased during morphine withdrawal (Coombes et al., 1991). Thus oxytocin release from SON (and PVN) magnocellular neuron dendrites might underpin central actions of oxytocin associated with opiate addiction. 5. Conclusion Here, we show complex effects of morphine on SON neurons. While acute morphine consistently hyperpolarized SON neurons, withdrawal from chronic morphine did not consistently alter SON neuron membrane potential. Hence, a rebound depolarization is unlikely to underpin morphine withdrawal excitation of SON
M. Ruan et al. / Neuropharmacology 61 (2011) 789e797
neurons in vivo. Surprisingly, acute morphine exposure increased ADP amplitude and this might contribute to the morphine-induced excitation observed in a small proportion of SON neurons in vitro (Wakerley et al., 1983). Upon naloxone-induced withdrawal of chronic morphine, ADP amplitude increased, consistent with the observed changes in post-spike excitability seen in SON neurons in vivo (Brown et al., 2005). However, the increase in ADP amplitude was not marked; hence, this is unlikely to be the major contributor to the increased firing rate of oxytocin neurons during morphine withdrawal excitation. While we have previously shown that morphine withdrawal excitation does not increase the activation of afferent inputs to oxytocin neurons (Murphy et al., 1997), taken together with our in vivo observations (of an increased steady-state post-spike excitability during morphine withdrawal (Brown et al., 2005)), our current results (showing no steady-state depolarization during morphine withdrawal in vitro) indicate that increased synaptic input might be the major trigger for morphine withdrawal excitation of firing in oxytocin neurons, but the changes in intrinsic membrane properties found here at the level of SON neurons during morphine withdrawal (i.e. a functional increase in ADP amplitude) would contribute importantly to this withdrawal excitation. Acknowledgments Supported by a University of Otago Research Grant (CHB) and the Wellcome Trust (Grant #070118, CHB and JAR). References Bicknell, R.J., Leng, G., Lincoln, D.W., Russell, J.A., 1988. Naloxone excites oxytocin neurones in the supraoptic nucleus of lactating rats after chronic morphine treatment. J. Physiol. 396, 297e317. Blackburn-Munro, G., Brown, C.H., Neumann, I.D., Landgraf, R., Russell, J.A., 2000. Verapamil prevents withdrawal excitation of oxytocin neurones in morphinedependent rats. Neuropharmacology 39, 1596e1607. Bourque, C.W., Renaud, L.P., 1985. Activity dependence of action potential duration in rat supraoptic neurosecretory neurones recorded in vitro. J. Physiol. 363, 429e439. Brown, C.H., 2004. Rhythmogenesis in vasopressin cells. J. Neuroendocrinol. 16, 727e739. Brown, C.H., Bourque, C.W., 2004. Autocrine feedback inhibition of plateau potentials terminates phasic bursts in magnocellular neurosecretory cells of the rat supraoptic nucleus. J. Physiol. 557, 949e960. Brown, C.H., Bourque, C.W., 2006. Mechanisms of rhythmogenesis: insights from hypothalamic vasopressin neurons. Trends Neurosci. 29, 108e115. Brown, C.H., Ghamari-Langroudi, M., Leng, G., Bourque, C.W., 1999. Kappa-opioid receptor activation inhibits post-spike depolarizing after-potentials in rat supraoptic nucleus neurones in vitro. J. Neuroendocrinol. 11, 825e828. Brown, C.H., Leng, G., Ludwig, M., Bourque, C.W., 2006. Endogenous activation of supraoptic nucleus kappa-opioid receptors terminates spontaneous phasic bursts in rat magnocellular neurosecretory cells. J. Neurophysiol. 95, 3235e3244. Brown, C.H., Munro, G., Johnstone, L.E., Robson, A.C., Landgraf, R., Russell, J.A., 1997. Oxytocin neurone autoexcitation during morphine withdrawal in anaesthetized rats. Neuroreport 8, 951e955. Brown, C.H., Munro, G., Murphy, N.P., Leng, G., Russell, J.A., 1996. Activation of oxytocin neurones by systemic cholecystokinin is unchanged by morphine dependence or withdrawal excitation in the rat. J. Physiol. 496, 787e794. Brown, C.H., Murphy, N.P., Munro, G., Ludwig, M., Bull, P.M., Leng, G., Russell, J.A., 1998. Interruption of central noradrenergic pathways and morphine withdrawal excitation of oxytocin neurones in the rat. J. Physiol. 507, 831e842. Brown, C.H., Russell, J.A., 2004. Cellular mechanisms underlying neuronal excitability during morphine withdrawal in physical dependence: lessons from the magnocellular oxytocin system. Stress 7, 97e107. Brown, C.H., Russell, J.A., Leng, G., 2000. Opioid modulation of magnocellular neurosecretory cell activity. Neurosci. Res. 36, 97e120.
797
Brown, C.H., Stern, J.E., Jackson, K.L., Bull, P.M., Leng, G., Russell, J.A., 2005. Morphine withdrawal increases intrinsic excitability of oxytocin neurons in morphinedependent rats. Eur. J. Neurosci. 21, 501e512. Bull, P.M., Brown, C.H., Russell, J.A., Ludwig, M., 2006. Activity-dependent feedback modulation of spike patterning of supraoptic nucleus neurons by endogenous adenosine. Am. J. Physiol. Regul. Integr. Comp. Physiol. 291, R83eR90. Bull, P.M., Ludwig, M., Blackburn-Munro, G.J., Delgado-Cohen, H., Brown, C.H., Russell, J.A., 2003. The role of nitric oxide in morphine dependence and withdrawal excitation of rat oxytocin neurons. Eur. J. Neurosci. 18, 2545e2551. Coombes, J.E., Robinson, I.C., Antoni, F.A., Russell, J.A., 1991. Release of oxytocin into blood and into cerebrospinal fluid induced by naloxone in anaesthetized morphine-dependent rats: the role of the paraventricular nucleus. J. Neuroendocrinol 3, 551e561. Doi, N., Brown, C.H., Cohen, H.D., Leng, G., Russell, J.A., 2001. Effects of the endogenous opioid peptide, endomorphin 1, on supraoptic nucleus oxytocin and vasopressin neurones in vivo and in vitro. Br. J. Pharmacol. 132, 1136. Ghamari-Langroudi, M., Bourque, C.W., 2000. Excitatory role of the hyperpolarization-activated inward current in phasic and tonic firing of rat supraoptic neurons. J. Neurosci. 20, 4855e4863. Ghamari-Langroudi, M., Bourque, C.W., 2002. Flufenamic acid blocks depolarizing afterpotentials and phasic firing in rat supraoptic neurones. J. Physiol. 545, 537e542. Greffrath, W., Martin, E., Reuss, S., Boehmer, G., 1998. Components of afterhyperpolarization in magnocellular neurones of the rat supraoptic nucleus in vitro. J. Physiol. 513, 493e506. Johnstone, L.E., Brown, C.H., Meeren, H.K., Vuijst, C.L., Brooks, P.J., Leng, G., Russell, J.A., 2000. Local morphine withdrawal increases c-fos gene, Fos protein, and oxytocin gene expression in hypothalamic magnocellular neurosecretory cells. J. Neurosci. 20, 1272e1280. Kovacs, G.L., Horvath, Z., Sarnyai, Z., Faludi, M., Telegdy, G., 1985. Oxytocin and a C-terminal derivative (Z-prolyl-D-leucine) attenuate tolerance to and dependence on morphine and interact with dopaminergic neurotransmission in the mouse brain. Neuropharmacology 24, 413. Li, Z., Hatton, G.I., 1997. Ca2þ release from internal stores: role in generating depolarizing after-potentials in rat supraoptic neurones. J. Physiol. 498, 339e350. Liu, Q.S., Han, S., Jia, Y.S., Ju, G., 1999. Selective modulation of excitatory transmission by mu-opioid receptor activation in rat supraoptic neurons. J. Neurophysiol. 82, 3000e3005. Ludwig, M., Brown, C.H., Russell, J.A., Leng, G., 1997. Local opioid inhibition and morphine dependence of supraoptic nucleus oxytocin neurones in the rat in vivo. J. Physiol. 505, 145e152. Ludwig, M., Leng, G., 2006. Dendritic peptide release and peptide-dependent behaviours. Nat. Rev. Neurosci. 7, 126e136. Mansour, A., Fox, C.A., Akil, H., Watson, S.J., 1995. Opioid-receptor mRNA expression in the rat CNS: anatomical and functional implications. Trends Neurosci. 18, 22e29. Murphy, N.P., Onaka, T., Brown, C.H., Leng, G., 1997. The role of afferent inputs to supraoptic nucleus oxytocin neurons during naloxone-precipitated morphine withdrawal in the rat. Neuroscience 80, 567e577. Onaka, T., Luckman, S.M., Guevara-Guzman, R., Ueta, Y., Kendrick, K., Leng, G., 1995. Presynaptic actions of morphine: blockade of cholecystokinin-induced noradrenaline release in the rat supraoptic nucleus. J. Physiol. 482, 69e79. Pumford, K.M., Russell, J.A., Leng, G., 1993. Effects of the selective kappa-opioid agonist U50,488 upon the electrical activity of supraoptic neurones in morphine-tolerant and morphine-naive rats. Exp. Brain Res. 94, 237e246. Rayner, V.C., Robinson, I.C., Russell, J.A., 1988. Chronic intracerebroventricular morphine and lactation in rats: dependence and tolerance in relation to oxytocin neurones. J. Physiol. 396, 319e347. Renaud, L.P., Bourque, C.W., 1991. Neurophysiology and neuropharmacology of hypothalamic magnocellular neurons secreting vasopressin and oxytocin. Prog. Neurobiol. 36, 131e169. Ruan, M., Brown, C.H., 2009. Feedback inhibition of action potential discharge by endogenous adenosine enhancement of the medium afterhyperpolarization. J. Physiol. 587, 1043e1056. Russell, J.A., Neumann, I., Landgraf, R., 1992. Oxytocin and vasopressin release in discrete brain areas after naloxone in morphine-tolerant and -dependent anesthetized rats: pushepull perfusion study. J. Neurosci. 12, 1024e1032. Sumner, B.E., Coombes, J.E., Pumford, K.M., Russell, J.A., 1990. Opioid receptor subtypes in the supraoptic nucleus and posterior pituitary gland of morphinetolerant rats. Neuroscience 37, 635e645. Wakerley, J.B., Noble, R., Clarke, G., 1983. Effects of morphine and D-Ala, D-Leu enkephalin on the electrical activity of supraoptic neurosecretory cells in vitro. Neuroscience 10, 73.
Contents lists available at ScienceDirect
Neuropharmacology journal homepage: www.elsevier.com/locate/neuropharm
Acute morphine administration and withdrawal from chronic morphine increase afterdepolarization amplitude in rat supraoptic nucleus neurons in hypothalamic explants Ming Ruan a, John A. Russell b, Colin H. Brown a, * a b
Centre for Neuroendocrinology and Department of Physiology, Otago School of Medical Sciences, University of Otago, P.O. Box 913, Dunedin 9054, New Zealand Centre for Integrative Physiology, University of Edinburgh, Edinburgh EH8 9XD, UK
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 January 2011 Received in revised form 3 May 2011 Accepted 18 May 2011
Supraoptic nucleus (SON) neurons secrete either oxytocin or vasopressin into the bloodstream from their axon terminals in the posterior pituitary gland. SON neurons are powerfully inhibited by the classical m-opioid receptor agonist, morphine. Oxytocin neurons develop morphine dependence when chronically exposed to this opiate, and undergo robust withdrawal excitation when morphine is subsequently acutely antagonized by naloxone. Morphine withdrawal excitation is evident as an increased firing rate and is associated with an increased post-spike excitability that is consistent with the expression of an enhanced post-spike afterdepolarization (ADP) during withdrawal. Here, we used sharp electrode recording from SON neurons in hypothalamic explants from morphine naïve and morphine treated rats to determine the effects of morphine on the ADP, and to test the hypothesis that morphine withdrawal increases ADP amplitude in SON neurons. Acute morphine administration (0.05e5.0 mM) caused a dosedependent hyperpolarization of SON neurons that was reversed by concomitant administration of 10 mM naloxone, or by washout of morphine; counter-intuitively, acute exposure to 5 mM morphine increased ADP amplitude by 78 11% (mean SEM). Naloxone-precipitated morphine withdrawal did not alter baseline membrane potential in SON neurons from morphine treated rats, but increased ADP amplitude by 48 11%; this represents a hyper-activation of ADPs because the basal amplitude of the ADP was similar in SON neurons recorded from explants prepared from morphine naïve and morphine treated rats. Hence, an enhanced ADP might contribute to morphine withdrawal excitation of oxytocin neurons. Ó 2011 Elsevier Ltd. All rights reserved.
Keywords: Afterdepolarization Afterhyperpolarization Intrinsic excitability Oxytocin Vasopressin
1. Introduction Morphine is the classical agonist of m-opioid receptors, which are widely expressed within the central nervous system (Mansour et al., 1995). However, relatively few neuronal phenotypes develop dependence when chronically exposed to morphine; these include hypothalamic supraoptic nucleus (SON) and paraventricular nucleus (PVN) magnocellular oxytocin neurons (Brown and Russell, 2004; Brown et al., 2000) that project to the posterior pituitary gland where they secrete oxytocin into the systemic circulation. Morphine acutely inhibits oxytocin neurons in vivo (Ludwig et al., 1997), but during chronic intracerebroventricular (ICV) administration, oxytocin neurons develop dependence on morphine. Dependence is revealed by a marked and sustained increase in firing
* Corresponding author. Tel.: þ64 3 479 7354; fax: þ64 3 479 7323. E-mail addresses: [email protected] (M. Ruan), [email protected] (J.A. Russell), [email protected] (C.H. Brown). 0028-3908/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuropharm.2011.05.025
rate of oxytocin neurons (and a consequent large increase in oxytocin secretion) upon morphine withdrawal in vivo (Bicknell et al., 1988; Blackburn-Munro et al., 2000; Brown et al., 1996, 1998; Bull et al., 2003) that is accompanied by increased Fos protein and oxytocin heteronuclear RNA expression in oxytocin neurons (Johnstone et al., 2000), as well as increased oxytocin release into the brain from oxytocin neuron somata and dendrites (Brown et al., 1997; Russell et al., 1992). Withdrawal excitation occurs in oxytocin neurons without a marked change in the activation of their major afferent inputs (Murphy et al., 1997). Moreover, oxytocin neurons are excited by direct administration of the opioid receptor antagonist, naloxone, into the SON of morphine treated rats (Johnstone et al., 2000; Ludwig et al., 1997). Thus, oxytocin neurons evidently develop dependence at a cellular level and provide a robust model that is amenable to detailed analysis of the cellular mechanisms of morphine dependence (Brown and Russell, 2004). By contrast to oxytocin neurons, SON vasopressin neurons do not show morphine dependence; after chronic morphine treatment, naloxone administration excites some vasopressin neurons
790
M. Ruan et al. / Neuropharmacology 61 (2011) 789e797
and inhibits others (Bicknell et al., 1988; Brown et al., 2005), so the overall effect is to induce only a modest increase in plasma vasopressin concentrations (Bicknell et al., 1988). Excitability of SON neurons is strongly influenced by nonsynaptic post-spike potentials (Brown, 2004; Brown and Bourque, 2006), including an afterdepolarization (ADP) and a medium afterhyperpolarization (mAHP). mAHP amplitude is reduced in SON neurons during morphine withdrawal (Brown et al., 2005), indicating that a reduced mAHP might contribute to the increased firing rate of oxytocin neurons evident during morphine withdrawal. However, the changes in post-spike excitability that accompany the withdrawal-induced increase in firing rate in vivo indicate that morphine withdrawal might also expose an ADP in oxytocin neurons (Brown et al., 2005), which could drive withdrawal excitation. Here, we used sharp electrode recording from SON neurons in hypothalamic explants from morphine naïve and morphine treated rats to determine the effects of chronic morphine on the ADP and to test the hypothesis that morphine withdrawal increases ADP amplitude in SON neurons. 2. Material and methods 2.1. Ethical approval All experimental procedures were approved by the University of Otago Animal Ethics Committee and were carried out in accordance with the recommendations of the Australian and New Zealand Council for the Care of Animals in Research and Teaching.
2.2. Electrophysiology Female Sprague-Dawley rats (200e350 g) were restrained in a soft plastic cone (5e10 s) and decapitated. The brains were rapidly removed and a block of tissue 8 8 2 mm (dorso-ventral) containing the basal hypothalamus was excised using razor blades and pinned, ventral surface up, to the silicone elastomer base of a superfusion chamber. Within 2e3 min, the excised hypothalamic explant was superfused (0.5e1.0 ml min1, 32e33 C) with carbogenated (95% oxygen and 5% carbon dioxide) artificial CSF (aCSF; see below) delivered via tubing placed over the medial tuberal region. The arachnoid membrane overlying the SON was removed using fine forceps, and a tissue paper wick was placed at the rostral tip of the explant to facilitate aCSF drainage. The aCSF (pH 7.4; 295 3 mOsmol kg1) consisted of (in mM): 120 NaCl, 3 KCl, 1.2 MgCl2, 26 NaHCO2, 2.5 CaCl2, and 10 glucose (Sigma). Intracellular recordings were made using sharp micropipettes prepared from glass capillaries (1.5 mm O.D. 0.86 mm I.D.) pulled on a P-97 FlamingeBrown puller (Sutter Instruments, Novato, CA). Micropipettes were filled with 2 M potassium acetate to yield DC resistances of 70e160 MU to a AgeAgCl wire electrode immersed in aCSF. Voltage recordings were obtained using an Axoclamp 2B amplifier (Molecular Devices, Foster City, CA) in continuous current-clamp (‘bridge’) mode. Digitized signals (10 kHz; DigiData 1320 Interface, Molecular Devices) were stored on a personal computer running pClamp9.2 (Molecular Devices) and analyzed offline. Recordings were made from SON neurons impaled with sharp electrodes in superfused hypothalamic explants. The cells had resting membrane potentials more negative than 50 mV, input resistances of >150 MU, and spike amplitudes of >60 mV when measured from baseline. Each cell displayed frequency-dependent spike broadening and transient outward rectification when depolarized from initial membrane potentials more negative than 75 mV (characteristics specific to SON neurons (Renaud and Bourque, 1991)). To determine the effects of acute morphine, and withdrawal from chronic morphine, on SON neuron membrane potential, recordings were made from an initial baseline membrane potential 5e10 mV below spike threshold (maintained by injection of hyperpolarizing current if necessary, with a constant current injected throughout the recording from each neuron). Depolarizing current injection (80 ms) was applied to elicit a train of 4e5 action potentials (the number of spikes in the trains was kept constant for each cell). Subthreshold ADPs were recorded from cells maintained at a baseline membrane potential approximately 5e10 mV below the spike threshold whereas spontaneous action potential afterdischarge during suprathreshold ADPs was measured from cells held approximately 2 mV below threshold. To determine the effects of the drugs on ADP and mAHP amplitude, as well as afterdischarge, measurements in the presence of drug were made from a baseline membrane potential between 1 mV more positive and 2 mV more negative than that used for pre-drug measurements.
2.3. Induction of morphine dependence Morphine dependence was induced as previously described (Rayner et al., 1988). Briefly, virgin female Sprague-Dawley rats (bodyweight 200e350 g) were anaesthetized with halothane (5% in O2). An Alzet model 2001 mini-osmotic pump (Charles River Ltd.) was placed subcutaneously and connected via polythene tubing to a 21 gauge stainless steel cannula implanted into the right lateral cerebral ventricle (3.0 mm caudal, 2.0 mm lateral to bregma and 4.5 mm below the surface of the skull). The pump and tubing were filled with morphine sulfate (Sigma) in sterile pyrogen-free water to deliver increasing doses (10 mg h1, 20 mg h1 for 40 h each and 50 mg h1 for the remainder at 1 ml h1) over 5 days. The cannula was secured using dental acrylic bonded to stainless steel screws inserted in the skull. Following surgery rats were housed individually with food and water available ad libitum. For electrophysiological recording (the sixth day following minipump implantation for morphine treated rats) in explants from morphine treated rats, the aCSF contained 5 mM morphine. 2.4. Statistics All averaged data are expressed as the mean the standard error of the mean (SEM). All differences within groups were evaluated with SigmaPlot software (SPSS Science, Chicago, IL, USA) using Student’s t-tests, paired t-tests or one-way repeated measures (RM) ANOVA, where appropriate. Where the F-ratio was significant, posthoc comparisons were completed as described in the figure legends.
3. Results 3.1. Effects of acute morphine and naloxone-precipitated withdrawal from chronic morphine on SON neuron membrane potential To determine the effects of acute morphine administration, membrane potential was measured during superfusion of 50 nM, 500 nM and 5 mM morphine followed by 10 mM naloxone (in the continued presence of morphine). Morphine caused a small,
Fig. 1. Acute morphine hyperpolarizes SON neurons in vitro. A, Example of the reversible, dose-dependent effects of morphine on SON neuron membrane potential, typical of all five neurons tested. The segments of recording shown were all made during injection of 25 pA of hyperpolarizing current and show the steady-state effects of each treatment on baseline membrane potential (relative to control: dashed line after >5 min superfusion of drug); continuous gap free recordings were interrupted to permit measurement of post-spike potentials. B, Change in membrane potential (mean SEM) of five SON neurons during superfusion of morphine (MOR) and morphine þ 10 mM naloxone (MOR þ NLX), showing a dose-dependent, naloxonereversible hyperpolarization of SON neurons by morphine (P ¼ 0.012, one-way RM ANOVA). *P < 0.05 versus pre-drug and MOR þ NLX, StudenteNewmaneKeuls posthoc tests.
M. Ruan et al. / Neuropharmacology 61 (2011) 789e797
791
dose-dependent hyperpolarization in all five SON neurons tested (P ¼ 0.01, one-way RM ANOVA; Fig. 1), similar to that reported previously in response to a m-opioid receptor agonist, endomorphin-1, in SON neurons in hypothalamic slices (Doi et al., 2001). The morphine-induced hyperpolarization was reversed by administration of 10 mM naloxone. In three other neurons, naloxone alone (10 mM) did not consistently alter membrane potential (0.7 1.0 mV change; P ¼ 0.54, paired t-test). Membrane potential was measured in nine neurons in explants from morphine treated rats during superfusion of 10 mM naloxone (in the continued presence of 5 mM morphine and with a constant hyperpolarizing current throughout each recording). Similar to the effects of naloxone in SON neurons from morphine naïve rats, naloxone did not consistently alter membrane potential (1.2 mV change, measured w10 min after the onset of naloxone administration; P ¼ 0.17, paired t-test).
3.2. Effects of acute morphine treatment on the ADP and mAHP in SON neurons To determine the effect of morphine on ADPs, we recorded SON neurons from morphine naïve rats (n ¼ 13) during superfusion of increasing doses of acute morphine (50e5 mM). Trains of four or five spikes were elicited every 20 s by an 80 ms depolarizing current pulse (þ100e300 pA, from 5 mV to 10 mV below spike threshold) to evoke post-train ADPs (and mAHPs). Paradoxically, ADP amplitude was increased by acute morphine superfusion in SON neurons and this effect was reversed by co-administration of 10 mM naloxone (P ¼ 0.008, one-way RM ANOVA; Fig. 2). In four of 13 neurons that did not express a measurable ADP (i.e. <0.5 mV) under basal conditions, an ADP (>0.5 mV) was exposed during superfusion of 500 nM morphine (n ¼ 1) or 5 mM morphine (n ¼ 3). Concomitant with an increase in ADP amplitude, morphine reduced mAHP amplitude, and this effect was partially reversed by co-administration of 10 mM naloxone (n ¼ 23; P < 0.001, one-way RM ANOVA; Fig. 2). In each of five neurons in which spontaneous action potential afterdischarge was superimposed upon evoked suprathreshold ADPs, superfusion of morphine (50 nM, n ¼ 4 or 5 mM, n ¼ 1) increased the number of spikes in the afterdischarge from 1 spike to 4 1 spikes (P ¼ 0.01, paired t-test; Fig. 2).
3.3. Effects of chronic ICV morphine treatment on the ADP and mAHP in SON neurons
Fig. 2. Acute morphine increases ADP amplitude in SON neurons in vitro. A, ADPs and mAHPs (averages of 10 traces) that each follow a 5-spike train (arrowheads, spikes truncated) evoked by a 80 ms depolarizing pulse (þ200 pA) in a SON neuron before (Pre-drug, left) and during superfusion of 50 nM morphine (MOR, middle), and during superfusion of 10 mM naloxone (NLX) in the continued presence of morphine, (right). In A, the ADP amplitude is the distance between the middle pair of dotted lines before morphine (left) and after morphine þ naloxone (right); the ADP amplitude was increased by morphine, as shown by the distance between the middle pair of dotted lines (centre). B, ADP amplitude in SON neurons during superfusion of increasing doses of morphine, showing that acute morphine treatment causes a naloxone-reversible increase in ADP amplitude (P ¼ 0.014, one-way RM ANOVA). C, mAHP amplitude in SON neurons during superfusion of increasing doses of morphine, showing that acute morphine treatment causes a decrease in mAHP amplitude; the effect of morphine was partially reversed by naloxone (P < 0.001, one-way RM ANOVA). *P < 0.05, **P < 0.01 and ***P < 0.001 versus pre-drug, StudenteNewmaneKeuls post-hoc tests. D, Posttrain afterdischarge superimposed on ADPs before and during superfusion of
To determine the effects of chronic morphine treatment on the ADP, rats were treated with ICV morphine for five days and SON neurons recorded from hypothalamic explants superfused with aCSF containing 5 mM morphine. The proportion of SON neurons that expressed a measurable ADP (>0.5 mV) was not significantly different (P ¼ 0.19, Chi-square test) between morphine naïve rats (14 of 27 neurons; 52%) and morphine treated rats (13 of 17 neurons; 76%). Furthermore, the amplitude of the ADP (in SON neurons that expressed a measurable ADP) was not different between morphine naïve (1.7 0.3 mV, n ¼ 14) and morphine treated rats (1.6 0.3 mV, n ¼ 13; P ¼ 0.77; Fig. 3). Similarly, the amplitude of the post-train mAHP was not different between SON neurons from morphine naïve rats (5.3 0.4 mV, n ¼ 27) and morphine treated rats (5.1 0.6 mV, n ¼ 17; P ¼ 0.73; Fig. 3).
morphine. E, Mean number of spikes in the afterdischarge before and during superfusion of morphine (n ¼ 5 cells; *P < 0.05, paired t-test).
792
M. Ruan et al. / Neuropharmacology 61 (2011) 789e797
3.4. Effects of naloxone-precipitated morphine withdrawal on the ADP and mAHP in SON neurons We have previously shown in vivo that naloxone-precipitated morphine withdrawal induces changes in post-spike excitability of SON oxytocin (and vasopressin) neurons that might be explained by an increase in the amplitude of the ADP (Brown et al., 2005). Therefore, we tested the effects of naloxone on the ADP in SON neurons from which the membrane potential was recorded in hypothalamic explants prepared from morphine treated rats and maintained in aCSF containing 5 mM morphine. By contrast to its effects in explants from morphine naïve rats (Fig. 2), superfusion of naloxone (10 mM, in the continued presence of 5 mM morphine to prevent morphine withdrawal after removal and during preparation of the hypothalamic explant from morphine treated rats) caused a consistent increase in ADP amplitude (þ0.5 0.1 mV) in neurons from morphine treated rats (n ¼ 7; P ¼ 0.005, paired t-test; Fig. 4). In two of three further neurons that did not express a measurable ADP (<0.5 mV) before naloxone, naloxone administration exposed an ADP (Fig. 4). This enhancement of the ADP evident during naloxone-precipitated morphine withdrawal is consistent with the increased post-spike excitability observed during morphine withdrawal excitation in vivo (Brown et al., 2005). In one neuron in which spontaneous action potential afterdischarge was superimposed on suprathreshold evoked ADPs,
Fig. 3. Chronic morphine in vivo does not alter ADP amplitude in SON neurons in vitro. A, ADPs and mAHPs (averages of 10 traces) that each follow a 5-spike train (arrowheads, spikes truncated) evoked by a 80 ms depolarizing pulse (þ150 and þ300 pA, respectively) in SON neurons from a morphine naïve rat (left) and a morphine treated rat (right; 5 mM morphine was present throughout recording of the neuron from the morphine treated rat). B, ADP amplitude in SON neurons from morphine naïve rats (left; n ¼ 14) and morphine treated rats (right; n ¼ 13), showing that chronic morphine treatment does not alter ADP amplitude. C, mAHP amplitude in SON neurons from morphine naïve rats (left; n ¼ 27) and morphine treated rats (right; n ¼ 17), showing that, similar to its lack of effect on the ADP, chronic morphine treatment does not alter mAHP amplitude.
superfusion of naloxone increased the number of spikes in the afterdischarge from 1 spike to 2 1 spikes (P < 0.001, Student’s t-test; n ¼ 10 contiguous measurements in each condition). Concomitant with an increase in ADP amplitude, naloxone reduced mAHP amplitude in neurons from morphine treated rats (by 0.9 0.4 mV; n ¼ 10; P ¼ 0.04; Fig. 4), similarly to our previous results from perforated patch experiments in hypothalamic slices (Brown et al., 2005). 3.5. Effects of morphine and morphine withdrawal on the mAHP and sAHP in SON neurons In SON neurons from morphine naïve rats, the sAHP (the average membrane potential measured between 1 and 2 s after the
Fig. 4. Morphine withdrawal increases ADP amplitude in SON neurons in vitro. A and B, ADPs and mAHPs (averages of 10 traces) that each follow a 5-spike train (arrowheads, spikes truncated) evoked by a 80 ms depolarizing pulse (þ200 pA) before (PreNLX, left) and during superfusion of 10 mM naloxone (NLX, right) in a hypothalamic explant prepared from a morphine treated rat (5 mM morphine was present throughout). Note that in A, the ADP was increased by naloxone and that in B, an ADP was exposed by naloxone (dotted lines: control, Pre-NLX values). C, ADP amplitude in individual SON neurons from morphine treated rats (n ¼ 10) before (Pre-NLX; left) and during (NLX; right) superfusion of 10 mM naloxone in the presence of 5 mM morphine, showing that naloxone consistently increased ADP amplitude. D, mAHP amplitude in the same cells from C, showing that naloxone consistently decreased mAHP amplitude concomitant with its effects on the ADP. *P < 0.05 and **P < 0.01, paired t-test.
M. Ruan et al. / Neuropharmacology 61 (2011) 789e797
end of a 25-spike train to avoid potential confounding effects of the faster mAHP; n ¼ 24; Fig. 5A) was not consistently affected by acute morphine or naloxone (maximum change þ85.2 115.5%; P ¼ 0.49; Fig. 5B). The amplitude of the sAHP was not different between SON neurons from morphine naïve rats (n ¼ 27) and from morphine treated rats (n ¼ 14; P ¼ 0.82; Fig. 5D and E). In SON neurons from morphine treated rats, the amplitude of the sAHP was not affected by naloxone (n ¼ 7; P ¼ 0.72; Fig. 5G and H). In the neurons from morphine naïve rats, the combined mAHP and sAHP (the maximum negative membrane potential following the 25-spike train) was not consistently affected by acute morphine or naloxone (Fig. 5A); while one-way ANOVA was significant (P ¼ 0.03), all pairwise post-hoc analyses using StudenteNewmaneKeuls tests
793
did not identify any significant differences between any of the groups (P 0.09; Fig. 5C). The amplitude of the combined mAHP and sAHP was not different between SON neurons from morphine naïve rats and from morphine treated rats (P ¼ 0.44; Fig. 5D and F). In SON neurons from morphine treated rats, the amplitude of the combined mAHP and sAHP was not affected by naloxone (P ¼ 0.79; Fig. 5G and I). 3.6. Effects of morphine and morphine withdrawal on the fAHP in SON neurons In SON neurons from morphine naïve rats, fAHP amplitude (measured from spike threshold of the first spike in the evoked train; n ¼ 23) was decreased by acute morphine (maximum
Fig. 5. Effects of morphine and morphine withdrawal on the mAHP and sAHP in SON neurons in vitro. A, Post-train sAHPs (evoked by a 25-spike train; spikes truncated, arrowhead) before (left) and during (middle) 5 mM morphine and 10 mM naloxone in 5 mM morphine (right), showing no effect of morphine on the mAHP and sAHP in SON cells from morphine naïve rats. B, Mean membrane potential in SON cells from morphine naïve rats measured 1e2 s after the 25-spike train (to avoid possible confounding effects of the mAHP) before (Predrug) and during increasing doses of morphine (MOR) and morphine plus 10 mM naloxone (MOR þ NLX). C, Mean mAHP þ sAHP amplitude immediately after the 25-spike train, before and during increasing doses of morphine and morphine plus 10 mM naloxone in SON cells from morphine naïve rats. D, Post-train mAHP and sAHPs in SON neurons from morphine naïve rats (left) and morphine treated rats (in SON neurons from morphine naïve and morphine treated rats (right)). E, Mean membrane potential (measured 1e2 s after the 25-spike train) in SON neurons from morphine naïve and morphine treated rats. F, Mean mAHP þ sAHP amplitude immediately after the 25-spike train in SON neurons from morphine naïve and morphine treated rats. G, Post-train mAHPs and sAHPs in SON neurons from a morphine treated rat before (left) and during (right) superfusion of 10 mM naloxone, showing no effect of naloxone on the sAHP. H, Mean membrane potential (measured 1e2 s after the 25-spike train) in SON neurons from morphine treated rats before and during naloxone (NLX) superfusion. I, Mean mAHP þ sAHP amplitude immediately after the 25-spike train in SON neurons from morphine treated rats before and during naloxone (NLX) superfusion.
794
M. Ruan et al. / Neuropharmacology 61 (2011) 789e797
change 9.4 2.2%; P ¼ 0.01; Fig. 6A). The amplitude of the fAHP was greater in SON neurons from morphine naïve rats (n ¼ 27) than in those from morphine treated rats (n ¼ 17; P ¼ 0.002; Fig. 6B). In SON neurons from morphine treated rats, the amplitude of the
fAHP was increased by naloxone (by 1.6 0.3 mV, n ¼ 10; P < 0.001; Fig. 5E) and the sAHP was not affected by naloxone (n ¼ 7; P ¼ 0.72; Fig. 6C). 3.7. Effects of morphine and morphine withdrawal on spike properties of SON neurons In SON neurons from morphine naïve rats, the amplitude (measured from baseline) of the first spike in the evoked train (n ¼ 23) was not significantly affected by morphine or naloxone (maximum change 4.3 1.6%; P ¼ 0.24, one-way RM ANOVA; Fig. 7A). The half-width of the first spike in the evoked train (n ¼ 23) was not affected by morphine (maximum change þ1.2 2.9%), but was increased by naloxone co-administration with morphine (þ12.6 3.5% increase; P ¼ 0.01; Fig. 7A). Spike broadening (the percentage increase in spike half-width across a five-spike train; n ¼ 20) was not affected by acute morphine or naloxone (maximum change þ3.2 2.0%; P ¼ 0.40; Fig. 7A). Spike amplitude was not significantly different between SON neurons from morphine naïve rats (n ¼ 27) and morphine treated rats (n ¼ 17; P ¼ 0.16, Student’s t-test; Fig. 7B). By contrast spike half-width was narrower in SON neurons from morphine naïve rats (n ¼ 27) than in those from morphine treated rats (n ¼ 17; P ¼ 0.009; Fig. 7B). Nevertheless, spike broadening was similar in SON neurons from morphine naïve rats (n ¼ 23) and from those in morphine treated rats (n ¼ 17; P ¼ 0.71; Fig. 7B). The amplitude of the first spike in the evoked train was not significantly affected by naloxone in SON neurons from morphine treated rats (n ¼ 10; P ¼ 0.58; Fig. 7C). Similarly, spike half-width was not significantly affected by naloxone in SON neurons from morphine treated rats (n ¼ 10; P ¼ 0.48; Fig. 7C). However, consistent with our previous observations (Brown et al., 2005), the proportionate increase in spike half-width across a five-spike train was increased by naloxone (by 3.7 1.3%; n ¼ 6; P ¼ 0.04; Fig. 7C), which might underpin the increased ADP evident during morphine withdrawal. 4. Discussion
Fig. 6. Effects of morphine and morphine withdrawal on the fAHP in SON neurons in vitro. A, Post-spike (spike truncated, arrowhead) fAHPs (left panel) before (solid line) and during superfusion of 5 mM morphine (dotted line) and 10 mM naloxone in the continued presence of 5 mM morphine (dashed line), showing a decrease in fAHP amplitude in the presence of morphine; for the sake of clarity, fAHPs in the presence of 50 nM and 500 nM morphine have been omitted. Right panel: mean fAHP amplitude before (Pre-drug) and during superfusion of various doses of morphine (MOR) and morphine plus 10 mM naloxone (MOR þ NLX). *P < 0.05 versus pre-drug, StudenteNewmaneKeuls post-hoc tests. B, fAHPs (left panel) from SON neurons from morphine naïve rats (solid line) and morphine treated rats (in the presence of 5 mM morphine; dotted line) showing a decrease in fAHP amplitude after chronic morphine administration. Right panel: mean fAHP amplitude in SON neurons from morphine naïve and morphine treated rats. **P < 0.01, Student’s t-test. C, fAHPs (left panel) from a SON neuron from a morphine treated rat before (solid line) and during (dotted line) superfusion of 10 mM naloxone, showing an increase in fAHP amplitude in naloxone. Right panel: mean fAHP amplitude in SON neurons from morphine treated rats before and during naloxone superfusion. ***P < 0.001, paired t-test.
Using sharp electrode recordings from hypothalamic explants, we have revealed complex effects of morphine and naloxone-induced morphine withdrawal on the electrophysiological properties of SON neurons from morphine naïve rats and morphine-dependent rats. While acute morphine administration increased ADP amplitude, ADP amplitude was similar in SON neurons recorded from explants prepared from morphine naïve rats and rats chronically administered morphine, indicative of the development of tolerance to the effects of morphine on the ADP. Surprisingly, we found that naloxone-precipitated withdrawal from chronic morphine also increased ADP amplitude. Such changes in ADP amplitude likely contribute to the increased excitability of oxytocin neurons evident in vivo after administration of low doses of m-opioid receptor agonists (Wakerley et al., 1983), or during withdrawal from chronic morphine administration (Bicknell et al., 1988). 4.1. Acute and chronic morphine effects on the electrophysiological properties of SON neurons Acute morphine administration consistently inhibits oxytocin neuron firing rate in vivo (Ludwig et al., 1997). Consistent with this observation, we found that morphine caused a small but consistent membrane hyperpolarization in all SON neurons tested. Counterintuitively, we also found that acute morphine administration increased ADP amplitude, which is expected to be excitatory, as seen by the increased spike afterdischarge evident during morphine
M. Ruan et al. / Neuropharmacology 61 (2011) 789e797
795
Fig. 7. Effects of morphine and morphine withdrawal on spike properties of SON neurons in vitro. A, Left: Spikes before (solid line) and during superfusion of 5 mM morphine (dotted line) and 10 mM naloxone in the continued presence of 5 mM morphine (dashed line) in SON cells from morphine naïve rats; for the sake of clarity, spikes in the presence of 50 nM and 500 nM morphine have been omitted. Histograms: left panel: mean spike amplitude of the first spike in a five-spike train before (Pre-drug) and during superfusion of increasing doses of morphine (MOR) and morphine plus 10 mM naloxone (MOR þ NLX) in SON cells from morphine naïve rats. Middle panel: mean spike half-width before and during superfusion of various doses of morphine and morphine plus naloxone. Right panel: mean spike width of the fifth spike in a five-spike train before and during superfusion of various doses of morphine and morphine plus naloxone. **P < 0.01 versus pre-drug, StudenteNewmaneKeuls post-hoc test. B, Left: Spikes in SON neurons from morphine naïve rats (solid line) and morphine treated rats (dotted line). Histograms: left, mean spike amplitude of the first spike in a five-spike train in SON neurons from morphine naïve rats and morphine treated rats. Middle: mean spike width of the first spike in a five-spike train in SON neurons from morphine naïve rats and morphine treated rats. Right: mean spike width of the fifth spike in a five-spike train in SON neurons from morphine naïve rats and morphine treated rats. **P ¼ 0.01, Student’s t-test. C, Left: Spikes from an SON neuron from a morphine treated rat before (solid line) and during (dotted line) superfusion of 10 mM naloxone. Histograms: left, mean spike amplitude of the first spike in a five-spike train from SON neurons from morphine treated rats before and during superfusion of 10 mM naloxone (NLX). Middle: mean spike width of the first spike in a five-spike train from SON neurons from morphine treated rats before and during superfusion of 10 mM naloxone. Right: mean spike width of the fifth spike in a five-spike train from SON neurons from morphine treated rats before and during superfusion of 10 mM naloxone. *P < 0.05, paired t-test.
administration. The morphine-induced enhancement of the ADP was evident at lower doses than those required to induce a hyperpolarization, which might explain the previous observation of morphine excitation of a small proportion of SON neurons in vitro (Wakerley et al., 1983). The morphine-induced enhancement of the ADP was likely mediated by m-opioid receptors because k-opioid agonists decrease ADP amplitude (Brown and Bourque, 2004; Brown et al.,1999, 2006) and SON neurons do not express d-opioid receptors
in morphine naïve rats or morphine-dependent rats (Sumner et al., 1990). Given that the mAHP was reduced by morphine concomitant with an increased ADP following a 4- or 5-spike train but that the combined mAHP and sAHP following a 25-spike train was not affected by morphine, it appears likely that the principal actions of acute morphine were on the ADP, rather than the mAHP. The selective m-opioid agonist D-Ala(2), N-CH(3)-Phe(4), Gly(5)ol-enkephalin (DAGO), decreases the frequency (but not the
796
M. Ruan et al. / Neuropharmacology 61 (2011) 789e797
amplitude or kinetics) of excitatory post-synaptic currents (Liu et al., 1999), indicating that glutamatergic inputs to SON neurons are under presynaptic m-opioid inhibition. Furthermore, morphine inhibits excitatory noradrenaline release within the SON in vivo (Onaka et al., 1995). Here, we found that morphine consistently hyperpolarized SON neurons, similar to the effects of the specific mopioid agonist, endomorphin-1 (Doi et al., 2001). Overall, we conclude that inhibition of excitatory inputs and direct steady-state hyperpolarization likely explain the inhibition by morphine of oxytocin neurons evident in vivo because the excitatory effects of an enhanced ADP will be evident only when a cell is active. 4.2. Morphine withdrawal effects on the electrophysiological properties of SON neurons During morphine withdrawal in vivo, oxytocin neurons increase their firing rate, which underpins a massive and prolonged secretion of oxytocin into the circulation and the brain (Brown and Russell, 2004). While acute morphine administration enhanced ADP amplitude, after five days of morphine treatment, the amplitude of the ADP was indistinguishable from that of morphine naïve rats, indicative of the development of tolerance to the effects of morphine on the ADP. Remarkably, morphine withdrawal did not cause a rebound inhibition of the ADP but rather had an excitatory effect similar to that of acute morphine exposure. However, the mechanisms that underpin each of these excitations might be quite different because (unlike morphine withdrawal) acute morphine did not alter spike broadening. The SON contains both oxytocin and vasopressin neurons. By contrast to oxytocin neurons, vasopressin neurons do not undergo morphine withdrawal excitation; naloxone administration to morphine-dependent rats does not consistently change vasopressin neuron firing rate and only modestly increases vasopressin secretion (Bicknell et al., 1988; Brown et al., 2005). Nevertheless, morphine withdrawal increased the ADP in all SON neurons tested in vitro, similar to our previous observations of a reduced mAHP and transient outward rectification in all SON neurons during morphine withdrawal (Brown et al., 2005). The failure of a consistently enhanced ADP (as seen here in vitro) to precipitate robust morphine withdrawal-induced increases in vasopressin neuron firing in vivo further supports our conclusion that an enhanced ADP likely does not alone drive withdrawal-induced increases in oxytocin neuron firing rate in vivo. However, it should be noted that vasopressin neurons are subject to activity-dependent autocrine inhibition by the k-opioid peptide, dynorphin, via inhibition of the ADP (Brown, 2004; Brown and Bourque, 2006). Because there is no cross-tolerance between mand k-opioid receptors in morphine-dependent rats (Pumford et al., 1993), activity-dependent k-opioid inhibition of vasopressin neuron (but not oxytocin neuron) ADPs will still occur during spontaneous activity, opposing the effects of ADP enhancement induced by morphine withdrawal in vasopressin neurons. In addition to k-opioid inhibition, we have also demonstrated that adenosine (probably generated from breakdown of ATP), also inhibits the activity of vasopressin neurons via activity-dependent activation of A1 receptors (Bull et al., 2006; Ruan and Brown, 2009), which would further oppose ADP enhancement induced by morphine withdrawal in vasopressin neurons. While we focused here on the effects of morphine withdrawal on the ADP, this was not the only property of SON neurons affected by morphine withdrawal. Concomitant with an increased ADP, the mAHP was decreased by morphine withdrawal, while the sAHP was not changed during morphine withdrawal. Because the mAHP and ADP overlap temporally (Greffrath et al., 1998; Ruan and Brown, 2009), and as we did not isolate the ADP, we cannot exclude the
possibility that inhibition of the mAHP might underpin withdrawal-induced enhancement of the ADP. However, there was no correlation between the change in mAHP amplitude and the change in ADP amplitude (following a 4- or 5-spike train) induced by withdrawal (Pearson Correlation coefficient, r ¼ 0.06; P ¼ 0.87) and the amplitude of the combined mAHP and sAHP (following a 25-spike train) was not affected by withdrawal, so it is likely that the withdrawal-induced enhancement of the ADP cannot be fully accounted for by inhibition of the mAHP. The fAHP (the hyperpolarizing phase of spike repolarization) was increased in SON neurons during morphine withdrawal. Intuitively, this would be expected to decrease excitability. However, the fAHP has been shown to increase firing rate in SON neurons via activation of a hyperpolarization-activated inward current (Ghamari-Langroudi and Bourque, 2000). Hence, an increased fAHP might also contribute to withdrawal-induced increases in firing rate. While baseline membrane potential, spike amplitude and spike half-width of SON neurons were not altered during morphine withdrawal, activity-dependent spike broadening was increased during withdrawal. Spike broadening is caused by Ca2þ influx (Bourque and Renaud, 1985) and the ADP is generated by a Ca2þ dependent (Li and Hatton, 1997) non-specific cation channel (Ghamari-Langroudi and Bourque, 2002), indicating that the increased Ca2þ influx at higher firing rates during withdrawal might contribute to the enhancement of the ADP. While spike broadening increased by only 4% and the ADP increased by 48%, the effect of Ca2þ influx on ADP amplitude is amplified by Ca2þ-induced Ca2þ release from intracellular stores (Li and Hatton, 1997). Furthermore, while spike half-width was not increased by withdrawal itself, spikes were broader in neurons from morphine treated rats (before withdrawal) compared to morphine naïve neurons. Hence, the modest increase in spike broadening during withdrawal was superimposed on an already broadened spike. If increased Ca2þ influx does underpin the withdrawal-induced enhancement of ADP amplitude, an increased ADP might be a consequence of withdrawal excitation, rather than a primary cause. 4.3. Consequences of morphine withdrawal excitation of magnocellular oxytocin neurons Although magnocellular oxytocin neurons (that project to the posterior pituitary gland) provide a robust model to study cellular morphine dependence (Brown and Russell, 2004), they might also play a role in the behavioral opiate withdrawal syndrome. Central oxytocin inhibits the development of morphine tolerance and attenuates the behavioral morphine withdrawal syndrome in mice (Kovacs et al., 1985). Centrally released oxytocin can originate from magnocellular neurosecretory neuron dendrites, being secreted by exocytosis (Ludwig and Leng, 2006). Central oxytocin antagonist treatment reduces withdrawal excitation of oxytocin neurons (Brown et al., 1997) and oxytocin release into the cerebrospinal fluid by magnocellular neurons is increased during morphine withdrawal (Coombes et al., 1991). Thus oxytocin release from SON (and PVN) magnocellular neuron dendrites might underpin central actions of oxytocin associated with opiate addiction. 5. Conclusion Here, we show complex effects of morphine on SON neurons. While acute morphine consistently hyperpolarized SON neurons, withdrawal from chronic morphine did not consistently alter SON neuron membrane potential. Hence, a rebound depolarization is unlikely to underpin morphine withdrawal excitation of SON
M. Ruan et al. / Neuropharmacology 61 (2011) 789e797
neurons in vivo. Surprisingly, acute morphine exposure increased ADP amplitude and this might contribute to the morphine-induced excitation observed in a small proportion of SON neurons in vitro (Wakerley et al., 1983). Upon naloxone-induced withdrawal of chronic morphine, ADP amplitude increased, consistent with the observed changes in post-spike excitability seen in SON neurons in vivo (Brown et al., 2005). However, the increase in ADP amplitude was not marked; hence, this is unlikely to be the major contributor to the increased firing rate of oxytocin neurons during morphine withdrawal excitation. While we have previously shown that morphine withdrawal excitation does not increase the activation of afferent inputs to oxytocin neurons (Murphy et al., 1997), taken together with our in vivo observations (of an increased steady-state post-spike excitability during morphine withdrawal (Brown et al., 2005)), our current results (showing no steady-state depolarization during morphine withdrawal in vitro) indicate that increased synaptic input might be the major trigger for morphine withdrawal excitation of firing in oxytocin neurons, but the changes in intrinsic membrane properties found here at the level of SON neurons during morphine withdrawal (i.e. a functional increase in ADP amplitude) would contribute importantly to this withdrawal excitation. Acknowledgments Supported by a University of Otago Research Grant (CHB) and the Wellcome Trust (Grant #070118, CHB and JAR). References Bicknell, R.J., Leng, G., Lincoln, D.W., Russell, J.A., 1988. Naloxone excites oxytocin neurones in the supraoptic nucleus of lactating rats after chronic morphine treatment. J. Physiol. 396, 297e317. Blackburn-Munro, G., Brown, C.H., Neumann, I.D., Landgraf, R., Russell, J.A., 2000. Verapamil prevents withdrawal excitation of oxytocin neurones in morphinedependent rats. Neuropharmacology 39, 1596e1607. Bourque, C.W., Renaud, L.P., 1985. Activity dependence of action potential duration in rat supraoptic neurosecretory neurones recorded in vitro. J. Physiol. 363, 429e439. Brown, C.H., 2004. Rhythmogenesis in vasopressin cells. J. Neuroendocrinol. 16, 727e739. Brown, C.H., Bourque, C.W., 2004. Autocrine feedback inhibition of plateau potentials terminates phasic bursts in magnocellular neurosecretory cells of the rat supraoptic nucleus. J. Physiol. 557, 949e960. Brown, C.H., Bourque, C.W., 2006. Mechanisms of rhythmogenesis: insights from hypothalamic vasopressin neurons. Trends Neurosci. 29, 108e115. Brown, C.H., Ghamari-Langroudi, M., Leng, G., Bourque, C.W., 1999. Kappa-opioid receptor activation inhibits post-spike depolarizing after-potentials in rat supraoptic nucleus neurones in vitro. J. Neuroendocrinol. 11, 825e828. Brown, C.H., Leng, G., Ludwig, M., Bourque, C.W., 2006. Endogenous activation of supraoptic nucleus kappa-opioid receptors terminates spontaneous phasic bursts in rat magnocellular neurosecretory cells. J. Neurophysiol. 95, 3235e3244. Brown, C.H., Munro, G., Johnstone, L.E., Robson, A.C., Landgraf, R., Russell, J.A., 1997. Oxytocin neurone autoexcitation during morphine withdrawal in anaesthetized rats. Neuroreport 8, 951e955. Brown, C.H., Munro, G., Murphy, N.P., Leng, G., Russell, J.A., 1996. Activation of oxytocin neurones by systemic cholecystokinin is unchanged by morphine dependence or withdrawal excitation in the rat. J. Physiol. 496, 787e794. Brown, C.H., Murphy, N.P., Munro, G., Ludwig, M., Bull, P.M., Leng, G., Russell, J.A., 1998. Interruption of central noradrenergic pathways and morphine withdrawal excitation of oxytocin neurones in the rat. J. Physiol. 507, 831e842. Brown, C.H., Russell, J.A., 2004. Cellular mechanisms underlying neuronal excitability during morphine withdrawal in physical dependence: lessons from the magnocellular oxytocin system. Stress 7, 97e107. Brown, C.H., Russell, J.A., Leng, G., 2000. Opioid modulation of magnocellular neurosecretory cell activity. Neurosci. Res. 36, 97e120.
797
Brown, C.H., Stern, J.E., Jackson, K.L., Bull, P.M., Leng, G., Russell, J.A., 2005. Morphine withdrawal increases intrinsic excitability of oxytocin neurons in morphinedependent rats. Eur. J. Neurosci. 21, 501e512. Bull, P.M., Brown, C.H., Russell, J.A., Ludwig, M., 2006. Activity-dependent feedback modulation of spike patterning of supraoptic nucleus neurons by endogenous adenosine. Am. J. Physiol. Regul. Integr. Comp. Physiol. 291, R83eR90. Bull, P.M., Ludwig, M., Blackburn-Munro, G.J., Delgado-Cohen, H., Brown, C.H., Russell, J.A., 2003. The role of nitric oxide in morphine dependence and withdrawal excitation of rat oxytocin neurons. Eur. J. Neurosci. 18, 2545e2551. Coombes, J.E., Robinson, I.C., Antoni, F.A., Russell, J.A., 1991. Release of oxytocin into blood and into cerebrospinal fluid induced by naloxone in anaesthetized morphine-dependent rats: the role of the paraventricular nucleus. J. Neuroendocrinol 3, 551e561. Doi, N., Brown, C.H., Cohen, H.D., Leng, G., Russell, J.A., 2001. Effects of the endogenous opioid peptide, endomorphin 1, on supraoptic nucleus oxytocin and vasopressin neurones in vivo and in vitro. Br. J. Pharmacol. 132, 1136. Ghamari-Langroudi, M., Bourque, C.W., 2000. Excitatory role of the hyperpolarization-activated inward current in phasic and tonic firing of rat supraoptic neurons. J. Neurosci. 20, 4855e4863. Ghamari-Langroudi, M., Bourque, C.W., 2002. Flufenamic acid blocks depolarizing afterpotentials and phasic firing in rat supraoptic neurones. J. Physiol. 545, 537e542. Greffrath, W., Martin, E., Reuss, S., Boehmer, G., 1998. Components of afterhyperpolarization in magnocellular neurones of the rat supraoptic nucleus in vitro. J. Physiol. 513, 493e506. Johnstone, L.E., Brown, C.H., Meeren, H.K., Vuijst, C.L., Brooks, P.J., Leng, G., Russell, J.A., 2000. Local morphine withdrawal increases c-fos gene, Fos protein, and oxytocin gene expression in hypothalamic magnocellular neurosecretory cells. J. Neurosci. 20, 1272e1280. Kovacs, G.L., Horvath, Z., Sarnyai, Z., Faludi, M., Telegdy, G., 1985. Oxytocin and a C-terminal derivative (Z-prolyl-D-leucine) attenuate tolerance to and dependence on morphine and interact with dopaminergic neurotransmission in the mouse brain. Neuropharmacology 24, 413. Li, Z., Hatton, G.I., 1997. Ca2þ release from internal stores: role in generating depolarizing after-potentials in rat supraoptic neurones. J. Physiol. 498, 339e350. Liu, Q.S., Han, S., Jia, Y.S., Ju, G., 1999. Selective modulation of excitatory transmission by mu-opioid receptor activation in rat supraoptic neurons. J. Neurophysiol. 82, 3000e3005. Ludwig, M., Brown, C.H., Russell, J.A., Leng, G., 1997. Local opioid inhibition and morphine dependence of supraoptic nucleus oxytocin neurones in the rat in vivo. J. Physiol. 505, 145e152. Ludwig, M., Leng, G., 2006. Dendritic peptide release and peptide-dependent behaviours. Nat. Rev. Neurosci. 7, 126e136. Mansour, A., Fox, C.A., Akil, H., Watson, S.J., 1995. Opioid-receptor mRNA expression in the rat CNS: anatomical and functional implications. Trends Neurosci. 18, 22e29. Murphy, N.P., Onaka, T., Brown, C.H., Leng, G., 1997. The role of afferent inputs to supraoptic nucleus oxytocin neurons during naloxone-precipitated morphine withdrawal in the rat. Neuroscience 80, 567e577. Onaka, T., Luckman, S.M., Guevara-Guzman, R., Ueta, Y., Kendrick, K., Leng, G., 1995. Presynaptic actions of morphine: blockade of cholecystokinin-induced noradrenaline release in the rat supraoptic nucleus. J. Physiol. 482, 69e79. Pumford, K.M., Russell, J.A., Leng, G., 1993. Effects of the selective kappa-opioid agonist U50,488 upon the electrical activity of supraoptic neurones in morphine-tolerant and morphine-naive rats. Exp. Brain Res. 94, 237e246. Rayner, V.C., Robinson, I.C., Russell, J.A., 1988. Chronic intracerebroventricular morphine and lactation in rats: dependence and tolerance in relation to oxytocin neurones. J. Physiol. 396, 319e347. Renaud, L.P., Bourque, C.W., 1991. Neurophysiology and neuropharmacology of hypothalamic magnocellular neurons secreting vasopressin and oxytocin. Prog. Neurobiol. 36, 131e169. Ruan, M., Brown, C.H., 2009. Feedback inhibition of action potential discharge by endogenous adenosine enhancement of the medium afterhyperpolarization. J. Physiol. 587, 1043e1056. Russell, J.A., Neumann, I., Landgraf, R., 1992. Oxytocin and vasopressin release in discrete brain areas after naloxone in morphine-tolerant and -dependent anesthetized rats: pushepull perfusion study. J. Neurosci. 12, 1024e1032. Sumner, B.E., Coombes, J.E., Pumford, K.M., Russell, J.A., 1990. Opioid receptor subtypes in the supraoptic nucleus and posterior pituitary gland of morphinetolerant rats. Neuroscience 37, 635e645. Wakerley, J.B., Noble, R., Clarke, G., 1983. Effects of morphine and D-Ala, D-Leu enkephalin on the electrical activity of supraoptic neurosecretory cells in vitro. Neuroscience 10, 73.