Synaptic Adaptations In CA1 Area Of The Mouse Hippocampus Following Conditioned Place Preference To Morphine

Pete Rigby1, Laura McNair1,2, Chris Bailey1. 1University of Bath, Bath, UK, 2University of Copenhagen, Copenhagen, Denmark

Aberrant alterations in the process underlying reward-driven behavioural adaptation are widely assumed to be a root cause of compulsive drug use. Although several studies have investigated the neuronal effects of non-contingent administration of drugs of abuse, few studies have attempted to observe directly changes in synaptic function after animals have been trained to demonstrate drug-seeking behaviour. In this study, we used conditioned place preference (CPP) to first demonstrate behavioural adaptation. We then used whole-cell patch clamp and field excitatory post-synaptic potential (fEPSP) recordings to study changes in the functioning of the ventral hippocampal CA3-CA1 synapse.

4 week old C57BL/6J mice were trained in an unbiased counterbalanced CPP paradigm (2 intraperitoneal injections of 10mg.kg1 morphine counterbalanced with two injections of saline), resulting in significant preference for the morphine paired side (63±3% time spent in morphine-paired side, P<0.01, n=10, 1-sample t-test). Immediately following CPP testing, mice were killed and ventral hippocampal slices prepared (Nicholls et al., 1996). Separate groups of animals (‘non-contingent morphine’) received identical injections but were confined to home cage.

fEPSPs were recorded from the CA1 dendritic layer (350 μm-thick slices) following stimulation of Schaffer collateral fibres. Long-term potentiation (LTP) was induced with theta-burst stimulation and fEPSPs recorded every 2 minutes for 50 minutes post-theta burst. Analysis by 2-way ANOVA. For patch-clamp recordings, 275 μm slices were taken and cell-attached voltage-clamp recordings made from CA1 cell bodies. Excitatory post-synaptic currents (EPSCs) were recorded by stimulation of Schaffer collateral fibres; glutamatergic currents were isolated by including QX314 (5mM) and Fl- ions (63mM) in the recording pipette solution. AMPA/NMDA ratios were measured (Vh:+40mV), taking current values at 5ms post-stimulation (AMPA-mediated) and 50ms post-stimulation (NMDA-mediated). Paired-pulse ratios were taken with 2 paired stimuli 50ms apart. Analyses by 1-way ANOVA and Dunnett’s post-hoc test.

Stimulus-induced LTP was significantly reduced in mice that received non-contingent morphine ( 134±20% fEPSP size 30 minutes after theta burst vs. 174±23% (control); P<0.001, n=12 & 9) with a similar non-significant trend in mice that underwent morphine CPP training (154±43%; P=0.15 vs control, n=17). Similarly, AMPA/NMDA ratios were significantly increased following morphine CPP training (4.8±0.7 vs. 2.8±0.4 (control); P<0.05; n=11 & 8) with a similar non-significant trend in mice that received non-contingent morphine (4.0±0.3; n=11).

This apparent potentiation in synaptic efficacy was countered by the observation that paired-pulse facilitation was significantly increased in mice that received non-contingent morphine (1.43±0.05 vs. 1.24±0.07 (control); P<0.05; n=13&6), indicating a decrease in glutamate release probability. Surprisingly, in mice that had undergone morphine CPP there was no change from control (1.21±0.04; n=7).

These data suggest that LTP may have already occurred in vivo in response to both non-contingent morphine and morphine CPP (Billa et al., 2010). But, further presynaptic adaptations occur with non-contingent morphine that are either reversed or masked by concurrent CPP training. This as yet unidentified mechanism may comprise adaptations specific to drug-conditioned learning and may be of clinical relevance.

Billa SK et al. (2010) Mol. Pharmacol., 77, 874-83.

Nicholls RE et al. (1996) Proc. Natl. Acad. Sci. U.S.A .,103, 6380-5.

Funded by BBSRC