Bidirectional modulation of substantia nigra activity by motivational state

Abstract

A major output nucleus of the basal ganglia is the substantia nigra pars reticulata, which sends GABAergic projections to brainstem and thalamic nuclei. The GABAergic (GABA) neurons are reciprocally connected with nearby dopaminergic neurons, which project mainly to the basal ganglia, a set of subcortical nuclei critical for goal-directed behaviors. Here we examined the impact of motivational states on the activity of GABA neurons in the substantia nigra pars reticulata and the neighboring dopaminergic (DA) neurons in the pars compacta. Both types of neurons show short-latency bursts to a cue predicting a food reward. As mice became sated by repeated consumption of food pellets, one class of neurons reduced cue-elicited firing, whereas another class of neurons progressively increased firing. Extinction or pre-feeding just before the test session dramatically reduced the phasic responses and their motivational modulation. These results suggest that signals related to the current motivational state bidirectionally modulate behavior and the magnitude of phasic response of both DA and GABA neurons in the substantia nigra.

Figure 1. Single unit recording in substantia nigra.

(a) Representative photomicrograph of electrode tracks in the pars compacta (SNC; arrow; left). Schematic representation of the electrode placement (right; SNR, pars reticulata). Coordinates are relative to Bregma. Sample traces of DA (b) and GABA (c) neurons showing the narrower spike waveform of GABA neurons. Insets are inter-spike interval histograms. (d) Average waveforms (±s.e.m.) for all recorded DA and GABA neurons. We classified the neurons based on the waveforms of their action potentials. DA neurons, for example, are characterized by longer spike durations than GABA neurons. n is indicated by the numbers. (e) Full width half max (FWHM) values of classified DA and GABA neurons. (f) Schematic illustration of the FT60 task. One pellet was delivered into the food cup every 60 s (vertical black lines). An auditory cue preceded the pellet delivery by ∼550 ms (red lines). Vertical scale bars represent 68 µV (b) and 23 µV (c), and horizontal scale bars represent 500 µm in a, and 200 µs in b and c (10 ms in insets).

Figure 2. Effects of quinpirole on dopaminergic and GABAergic neurons.

To confirm the classification of cell types, we performed additional pharmacological experiments. Quinpirole (1 mg/kg), a D2 receptor agonist, was injected intraperitoneally during a recording session (red arrows). Example responses of DA (a–c) and GABA (d–f) neurons to quinpirole treatment are shown. Immediately following injection, the rate of dopaminergic firing decreased (b), whereas GABAergic neurons were unaffected (e). (c, f) Inter-spike interval histograms for the neurons in a and d. In a and d, vertical scale bars represent 76 µV and 137 µV, respectively. Horizontal scale bars represent 200 µs. Scale bars in c and f represent 25 ms.

Figure 3. Dopaminergic and GABAergic neurons showed a phasic response to reward cue.

(a) Activity of neurons that increased following reward cue normalized to the maximum firing rate; sorted according to increasing cumulative maximum values. Each row represents one neuron. X axis is time from reward cue onset (s). n is indicated in parentheses. (b) Representative peri-event histograms and rasters for individual DA (top) and GABA (bottom) neurons showing phasic responses following the reward cue (10 ms bins).

Figure 4. Synchrony is enhanced immediately after the reward cue.

Examples of synchrony between pairs of SN neurons following the reward cue. (a) Example of a corrected joint peri-stimulus time histograms (JPSTH) in the left column, corresponding predictor matrix (middle) and histogram of JPSTH diagonal (right; 500 ms width) for a pair of simultaneously recorded DA neurons (a), GABA neurons (b), and DA-GABA pairs (c); DA activity is plotted on the horizontal axis. Axes are time relative to the reward cue. The predictor matrix is subtracted bin-by-bin from the raw JPSTH, and then normalized by the standard deviation of the predictor matrix (100 ms×100 ms bins).

Figure 5. Population JPSTH analysis.

(a) Normalized population JPSTH (pJPSTH) of all DA-DA pairs (n = 185 pairs). Synchrony was enhanced immediately after the onset of the cue. (b) Synchrony among GABA-GABA pairs (n = 475 pairs) was not affected by the reward cue. (c) Synchrony between DA and GABA neurons (n = 394 pairs) was enhanced following the reward cue. pJPSTH bins are 100 ms×100 ms. Values are normalized by the standard deviation of predictor matrices. (d) The diagonal of each corrected pJPSTH is plotted (width  = 500 ms, bins  = 100 ms). DA-DA and DA-GABA pairs showed enhanced synchrony following the cue, whereas cue-elicited synchrony was not observed among GABA-GABA pairs. Only those neurons showing significant excitation following the reward cue were included in the analysis.

Figure 6. Motivational modulation of phasic responses to reward cue.

Motivational modulation of DA and GABA activity in SN. (a) Sample PSTH illustrating the motivational shift in phasic cue response of decreasing (a) and increasing (b) DA neurons separated into four 30 min blocks (bin size  = 30 ms). GABA neurons also showed similar decreasing (c) and increasing (d) motivational modulation.

Figure 7. Modulation is not due to changes in neuron isolation.

To test whether changes in neuronal isolation could account for the modulation observed in single unit activity, waveform characteristics were calculated for all modulated neurons for each half of the recording session. Neither FWHM (a) nor the peak of the inter-spike interval (ISI; b) changed during the two-hour recording sessions (paired t-tests, p>0.05). (c) Sample traces from the first and second half of the session for modulated neurons. The range of ISI values shown is 0–30 ms.

Figure 8. Food-seeking behavior gradually decreases with satiety.

(a) Following reward delivery, mice entered the food cup to collect the pellet. Representative data from one mouse are shown. (b) The rapid increase in food cup entries after the reward cue was eliminated during extinction and following pre-feeding. Traces are from the same mouse. FT60 data is the average from all four blocks in a. (c) Rate of entries decreased as mice became sated during FT60 sessions. It was also reduced during extinction and following pre-feeding (average of all mice on all testing days; 20 min bins). (d) The latency to enter the food cup following reward delivery increased during extinction and following pre-feeding treatment. (e) Latency to enter the food cup for all mice on all testing days was pooled to yield population averages. (f) The proportion of DA and GABA neurons that exhibited significant inhibitions following the reward cue was unaffected by pre-feeding and extinction. (g) The number of excitatory responses varied as a function of motivational state for both DA and GABA neurons. * represents significant difference at alpha of 0.05.

Figure 9. Average response of all excited neurons, including those that do not show motivational modulation during the session.

To compare the basal activity and mean responses between the classes we examined the responses of all neurons that were excited following the cue. The activity (mean ±s.e.m.; 20 ms bins) of all neurons with excitatory responses to the reward cue, regardless of motivational modulation, is plotted for each half of FT60 (a), pre-feeding (b) and extinction (c) sessions. Overall, there was a reduction in the magnitude of the phasic response as mice became sated. During extinction (c) the phasic response is nearly abolished for both DA and GABA neurons. The response to the cue (d) and to reward receipt (e) are dissociable. The population responses are shown for DA (n = 128) and GABA (n = 272) neurons recorded during FT60 sessions. (f–g) Baseline firing rates do not vary between FT60, pre-feed, or extinction sessions for DA or GABA neurons presented in a–c (one-way ANOVAs, F<1.0, p>0.05).

Figure 10. Motivational modulation is absent following pre-feeding and during extinction.

of the response of a dopamine neuron to the cue in FT60 (a), pre-feed (b), and extinction (c) sessions. Each line represents the average response from a 30-trial block. Bins  = 10 ms, Gaussian smooth  = 100 ms. Corresponding waveforms are shown at the right.

Figure 11. Summary of phasic response in dopamine neurons.

To compare the magnitude of the satiety effect to the magnitude of the mean response, we plotted the responses of all dopamine neurons showing a cue-elicited phasic burst. (a) The phasic response to the cue of increasing DA neurons (n = 19) was higher for the second half of FT60 sessions than it was for the first half. Decreasing DA neurons (n = 74) showed the opposite pattern. The phasic response of unmodulated neurons (n = 35) did not change over time. Modulation of the phasic response was reduced by pre-feeding (b; n = 5 increasing, 15 decreasing, 19 unmodulated) and eliminated during extinction (c; n = 2 increasing, 2 decreasing, 18 unmodulated). The phasic response of increasing neurons was nearly abolished following pre-feeding and during extinction, and the phasic response of decreasing neurons was abolished during extinction.

Figure 12. Summary of phasic response in GABA neurons.

GABA neurons show the same pattern of modulation as DA neurons. (a) The phasic response to the reward cue of increasing GABA neurons (n = 68) was higher for the second half of FT60 sessions than it was for the first half. Decreasing GABA neurons (n = 138) showed the opposite pattern. The phasic response of unmodulated neurons (n = 66) did not change. Modulation of the phasic response was reduced by pre-feeding (b; n = 15 increasing, 24 decreasing, 33 unmodulated) and eliminated during extinction (c; n = 7 increasing, 4 decreasing, 73 unmodulated). The phasic response of increasing neurons was nearly abolished following pre-feeding and during extinction, and the phasic response of decreasing neurons was abolished during extinction.

Figure 13. Mice are active throughout the recording session.

To ensure that motivational modulation was not caused by mice falling asleep or losing interest in the pellets, we used video tracking to record the position of mice during recording sessions. Mice spent most of their time near the food cup (a). Speed was calculated (pixels per frame) throughout the session. Mice remained consistently active for the entire two-hour session (b).

Figure 14. Cue-elicited synchrony of neural activity is also modulated by motivational state.

pJPSTH diagonals (mean ±s.e.m.; 100 ms×100 ms bins; 500 ms diagonal width) are shown for DA-DA (n = 185 FT60; n = 81 pre-feed; n = 45 extinction), GABA-GABA (n = 475 FT60; n = 346 pre-feed; n = 389 extinction), and DA-GABA (n = 394 FT60; n = 144 pre-feed; n = 138 extinction) pairs recorded during FT60 (a), pre-feeding (b), and extinction (c). The diagonals of the corrected and normalized pJPSTHs are in the top row, while the diagonals of the predictor matrices are in the bottom row. The predictor was defined as the cross-product of the JPSTH for each pair of neurons. Pre-feeding reduced cue-elicited synchrony for DA-DA and DA-GABA pairs while increasing overall synchrony between GABA-GABA pairs (b). Extinction reduced synchrony between DA-GABA and DA-DA pairs without affecting GABA-GABA pairs (c).