Psychopharmacology (2007) 191:843–853 DOI 10.1007/s00213-007-0733-z
ORIGINAL INVESTIGATION
Medial forebrain bundle stimulation evokes endocannabinoid-mediated modulation of ventral tegmental area dopamine neuron firing in vivo Giuliano Pillolla & Miriam Melis & Simona Perra & Anna Lisa Muntoni & Gian Luigi Gessa & Marco Pistis
Received: 18 October 2006 / Accepted: 30 January 2007 / Published online: 16 February 2007 # Springer-Verlag 2007
Abstract Rationale Endocannabinoid-mediated forms of transient synaptic depression have been described in several brain structures, including the dopaminergic ventral tegmental area (VTA). However, their functional and/or behavioural correlates are yet to be determined. Objectives The present study was designed to investigate whether back-propagating action potentials in dopamine (DA) neurons, evoked by the stimulation of the medial forebrain bundle (MFB), could trigger endocannabinoidmediated forms of synaptic modulation. The MFB contains axons ascending from DA neurons to the nucleus accumbens and other forebrain structures, and its stimulation is rewarding because it elicits intra-cranial self-stimulation. Materials and methods Single cell extracellular recordings were carried out from anti-dromically identified VTA DA neurons in chloral hydrate anesthetized rats. Results DA neurons responded to MFB stimulation (1 s, 20–80 Hz) with a frequency-dependent increase in spontaneous firing rate, which was enhanced by the cannabinoid type-1 receptor antagonist SR141716A (1 mg/kg) and depressed by the agonist WIN55212-2 (0.125 mg/kg). G. Pillolla : M. Melis : S. Perra : M. Pistis (*) B.B. Brodie Department of Neuroscience, University of Cagliari, Cittadella Universitaria di Monserrato, 09042 Monserrato , Cagliari, Italy e-mail:
[email protected] G. Pillolla : M. Melis : S. Perra : A. L. Muntoni : M. Pistis Center of Excellence for the Neurobiology of Addiction, University of Cagliari, 09042 Monserrato , Cagliari, Italy A. L. Muntoni : G. L. Gessa C.N.R. Institute of Neuroscience, University of Cagliari, 09042 Monserrato, Cagliari, Italy
Increasing brain levels of the endocannabinoid anandamide by blocking its major hydrolysing enzyme, fatty-acid amide hydrolase, with URB597 (0.1 mg/kg) was ineffective, whereas blockade of the endocannabinoid membrane transporter with UCM707 (1 mg/kg) enhanced poststimulus firing rate. Conclusions Our study indicates that stimulation of the MFB evokes an endocannabinoid-mediated short-term modulation of DA neuron activity. Thus, endocannabinoids might play an important role in the mechanisms underlying the rewarding properties of MFB stimulation. Keywords Endocannabinoids . Dopamine neurons . Electrophysiology . Medial forebrain bundle . Ventral tegmental area . Rat
Introduction Endocannabinoids are emerging as important modulators of synaptic functions in several brain regions (Chevaleyre et al. 2006). Depolarization-induced suppression of excitation and inhibition (DSE and DSI, respectively) are the bestcharacterised phenomena of endocannabinoid-mediated short-term synaptic plasticity (Diana and Marty 2004). Depolarization of postsynaptic neurons evokes mobilisation of endocannabinoids, which, acting pre-synaptically at cannabinoid type-1 (CB1) receptors, depress glutamate or GABA release (Freund et al. 2003). The identification, characterisation and molecular dissection of this mechanism have preceded the discovery of functional and/or behavioural correlates. In this respect, several problems arise when in vitro data are transposed into the in vivo condition. First, the protocols to induce DSE/DSI are often rather extreme. Depolarization of the post-synaptic membrane of
844
tens of mV for tens of seconds are difficult to be considered as physiological, leading to the idea that, DSE in particular, may rather subserve pathophysiological functions, i.e. during ischemia (Melis et al. 2006) or excitotoxicity (Marsicano et al. 2003). Indeed, a demonstration of DSE/ DSI-like phenomena in vivo is still lacking (Hampson et al. 2003). Secondly, in ventral tegmental area (VTA) dopaminergic (Melis et al. 2004a,b; Riegel and Lupica 2004; Lupica and Riegel 2005), cerebellar Purkinje (Kreitzer and Regehr 2001; Diana et al. 2002) and hippocampal pyramidal neurons (Wilson and Nicoll 2001; Ohno-Shosaku et al. 2002), endocannabinoids act retrogradely on CB1 receptors located both on glutamatergic and GABAergic terminals. Thus, the significance of these functional opposing actions is not clear. It was shown earlier that back-propagating action potentials are likely to play an important role in triggering the increases in intra-cellular Ca2+ necessary for DSI (Morishita and Alger 2001). We reasoned that stimulation of the medial forebrain bundle (MFB) could represent an ideal candidate for the demonstration of DSE/DSI-like mechanisms in vivo. The MFB is a large collection of ascending and descending fibres that passes through the lateral region of the hypothalamus. Among those fibres, the MFB contains axons of dopamine (DA) neurons ascending to the nucleus accumbens (NAc), and other forebrain regions, from their origin in the VTA. MFB stimulation induces an enduring potentiation of behaviour directed at obtaining additional stimulation (intra-cranial self-stimulation, ICS). Electrical stimulation at the parameters utilised to elicit ICS is able to produce both orthodromic spikes travelling to axon terminals in the NAc to release DA (Cheer et al. 2005) and antidromic action potentials back propagating to the cell body region of VTA neurons (Kuhr et al. 1987). A considerable body of evidence suggests that DA neurons are involved in the rewarding effects of MFB selfstimulation (Fibiger et al. 1987; Stellar and Corbett 1989), although a role of other neurotransmitters, such as GABA, cannot be excluded (Cheer et al. 2005). At frequencies used for ICS, ranging from 15 to 150 Hz, it is conceivable that most stimuli would induce antidromic spikes travelling to somatodendritic region of DA neurons. The consequences of this large stimulation-induced activity of DA neurons are unknown. Because DA neurons release endocannabinoids after either neuronal depolarization or stimulation of excitatory afferents (Melis et al. 2004a,b), one interesting possibility is that these back-propagating action potentials could induce endocannabinoid-mediated forms of synaptic modulation (Morishita and Alger 2001), which could be demonstrated in vivo. Behavioural evidence indicate that CB1 receptors modulate ICS, given that CB1 receptor antagonists reduce it (Deroche-Gamonet et al.
Psychopharmacology (2007) 191:843–853
2001), whereas the effect of the agonists is still controversial (Kucharski et al. 1983; Gardner et al. 1988; Lepore et al. 1996; Arnold et al. 2001; Antoniou et al. 2005; Vlachou et al. 2005). In search of a functional significance of endocannabinoid-mediated short-term synaptic depression, we tested the hypothesis that stimulation of MFB in vivo is able to evoke endocannabinoid-mediated modulation of DA neuron activity by antidromic action potentials, in a DSE/DSI-like mechanism. As a consequence, the endocannabinoid system might play a role in the mechanisms underlying rewarding properties of MFB stimulation.
Materials and methods Male Sprague–Dawley rats (250–350 g; Harlan, Italy) were housed in groups of three to six in standard conditions of temperature and humidity under 12 h/12 h light/dark cycle (with lights on at 7:00 A.M.) with food and water available ad libitum. Animals were anaesthetised with chloral hydrate (400 mg/kg, i.p.), their femoral vein was cannulated for intravenous administration of pharmacological agents and were placed in the stereotaxic apparatus (Kopf, Tujunga, CA, USA) with their body temperature maintained at 37± 1°C by a heating pad. Thereafter, the scalp was retracted, and one burr hole was drilled above the VTA (AP, 2.0 mm from lambda; L, 0.3–0.6 mm from midline) for the placement of a recording electrode. An additional hole was drilled for the placement of a Formvar-coated stimulating stainless steel bipolar electrode (250 μm tip diameter) above the MFB at the lateral hypothalamus level (AP, +0.1 mm from bregma; L, 2.0 mm from the midline; rostro-caudal angle, 15°; V, 8.0 mm), for antidromic activation of VTA neurons. Structures were localised according to the stereotaxic atlas of Paxinos and Watson (1997). Single unit activity of neurons located in VTA (V, 7.0– 8.0 mm from the cortical surface) was recorded extracellularly with glass micropipettes filled with 2% pontamine sky blue dissolved in 0.5 M sodium acetate (impedance, 2– 5 MΩ). Single unit activity was filtered (bandpass, 500– 5,000 Hz) and individual spikes were isolated by means of a window discriminator (Digitimer, Hertfordshire, UK), displayed on a digital storage oscilloscope (TDS 3012, Tektronics, Marlow, UK) and recorded by a videocassette recorder. Experiments were sampled on line and off line with Spike2 software (Cambridge Electronic Design, Cambridge, UK) by a computer connected to CED 1401 interface (Cambridge Electronic Design, Cambridge, UK). Single units were isolated and identified according to the already published criteria (Guyenet and Aghajanian 1978;
Psychopharmacology (2007) 191:843–853
845
Grace and Bunney 1983, 1984; Ungless et al. 2004). Because only one cell was recorded per rat, VTA DA neurons were selected when all criteria for identification were fulfilled: firing rate, 2.5 ms (Fig. 1a), inhibitory responses to hindpaw pinching. Additionally, VTA DA neurons were identified by antidromic spikes elicited by the stimulation of the MFB (Fig. 1b,c). An antidromic response was defined as the ability of evoked spikes to follow stimulation frequencies of more than 250 Hz, displaying constant latency and collision with spontaneously occurring spikes (Lipski 1981; Fig. 1b,c). Bursts were defined as the occurrence of two spikes at inter-spike interval of less than 80 ms and terminated when the inter-spike interval exceeded 160 ms (Grace and Bunney 1983). At the end of each experiment, apomorphine (40 μg/kg, i.v.) was administered to confirm the DAergic nature of the recorded cells. Hence, only cells whose firing was inhibited by more than 50% were included in this study. Baseline firing rates were obtained for 3 to 5 min, then the MFB was electrically stimulated with trains of monophasic stimuli at 20, 40, or
80 Hz for 1 s (20, 40, 80 pulses; 0.69±0.1 mA; duration of each pulse was 0.3 ms) to analyse the response of DA neurons to different frequencies. Following this set of experiments, stimulation frequency was set to 40 Hz, and drugs were administered intravenously 4 min after the first train. Neuronal activity was then recorded for 4 min before one additional train stimulation identical to the first one was delivered. Only one cell was recorded per rat. Peristimulus time histograms (PSTH) were generated on 2 min recording periods, 10 s bins, by Spike2 software (Cambridge Electronic Design, Cambridge, UK). Spike probability was normalised by dividing the number of action potentials in the bins after the stimulus by the average number of events in the pre-stimulus baseline.
Fig. 1 Stimulation of the MFB induces antidromic action potentials in VTA DA neurons and modulates spontaneous firing. a Superimposed traces, acquired from a digital storage oscilloscope, showing the spontaneous firing of a VTA DA neuron recorded from a chloral hydrate anaesthetized rat. Note the typical, broad, notched waveform. b Representative traces displaying a typical recording from a VTA DA neurons antidromically identified by MFB stimulation. Following the recording of spontaneous activity, which in this particular cell occurs mostly in bursts, the MFB is stimulated at 40 Hz for 1 s (action potentials are illustrated in black, stimulus artefacts in light gray). c At higher magnification, it can be observed that all stimuli are followed
by a single antidromic action potential, except when collisions (arrows) occur with spontaneous spikes (triangles). d Stimulation of MFB at different frequencies induces a short-lasting inhibition of spontaneous firing of VTA DA neurons and a frequency-dependent long-lasting increase. Data are expressed as mean±SEM and normalized to their baseline level (100%). Asterisk indicates p
