Start/stop signals emerge in nigrostriatal circuits during sequence learning

Abstract

Learning new action sequences subserves a plethora of different abilities such as escaping a predator, playing the piano, or producing fluent speech. Proper initiation and termination of each action sequence is critical for the organization of behaviour, and is compromised in nigrostriatal disorders like Parkinson's and Huntington's diseases. Using a self-paced operant task in which mice learn to perform a particular sequence of actions to obtain an outcome, we found neural activity in nigrostriatal circuits specifically signalling the initiation or the termination of each action sequence. This start/stop activity emerged during sequence learning, was specific for particular actions, and did not reflect interval timing, movement speed or action value. Furthermore, genetically altering the function of striatal circuits disrupted the development of start/stop activity and selectively impaired sequence learning. These results have important implications for understanding the functional organization of actions and the sequence initiation and termination impairments observed in basal ganglia disorders.

Figure 1. Mice learn to perform a specific sequence of actions

a, b, Example of the microstructure of the behavior of a mouse during a session on (a) the first day of FR8 training and (b) the last day of FR8 training. Each dot indicates a lever press, with red and black color indicating the first and final press of each individual sequence. The black and red solid lines on the X axis represent magazine entries and licks, respectively. The black dashed lines indicated the reward timing and corresponding lever press. c, Average number of lever presses per sequence, d, sequence duration, e, inter-sequence interval, and f, within-sequence press rate, during the first, sixth and twelfth day of FR8 training. g–j, Variability, measured as coefficient of variance (CV), of (g) sequence length, (h) sequence duration, (i) inter-sequence interval, and (j) within-sequence press rate during the first, sixth and twelfth day of FR8 training. Error bars denote s.e.m., same for all figures below.

Figure 2. Lever press-related activity in nigrostriatal circuits

a, b, Examples of MSNs showing (a) positive and (b) negative modulation of firing rate in relation to lever pressing. c, d, Examples of (c) positive and (d) negative modulation of firing rate of SN GABA neurons relative to lever pressing. e, Example of positive modulation of firing rate in a SN DA neuron before lever pressing. For each panel, top: raster plot where each dot indicates a spike; bottom: PETH for the neuron; time zero is the time of lever pressing. f, Percentage of MSNs (left panel), SN GABA (middle panel) and SN DA (right panel) neurons displaying press-related activity throughout learning, respectively.

Figure 3. Neural activity signaling the initiation and termination of action sequences emerges in nigrostriatal circuits during learning

a–c, Example of (a) a striatal MSN, (b) a SN GABA neuron and (c) a SN DA neuron showing significantly higher phasic firing selectively before the first press of a sequence. d, Example of a striatal MSN showing a phasic increase in activity preferentially before the last press of a sequence. e, Example of a SN GABA neuron showing both sequence initiation and termination related activity. For each panel, top: raster plot where each black dot indicates a spike and each orange triangle represents a lever press; bottom: PETH for the neuron. From left to right, each column shows the PETH of the same neuron related to the 1^st, 2^nd, 3^rd and 3^rd to final, 2^nd to final and final press of each sequence. f–h, Proportion of (f) striatal MSNs, (g) SN GABA neurons and (h) SN DA neurons exhibiting sequence start/stop related activity (including just initiation, just termination or both) throughout training. i–k, Proportion of striatal MSNs (i), SN GABA neurons (j), and SN DA neurons (k) displaying activity signaling sequence initiation, termination, or both initiation and termination (day 12).

Figure 4. Sequence start/stop related activity does not reflect differences in expected value and can be action-specific

a, Total number of licks for left and right single-lever FR8 forced choice sessions, where the left lever sequences led to a small reward and the right lever to a larger reward; after six days the lever-reward magnitude contingency was switched. b, Mice prefer the lever leading to larger reward during two-lever choice extinction tests on day 7 and day 13. c, Percentage of striatal MSNs, SN GABA neurons, and SN DA neurons exhibiting action-specific start/stop related activity. d, g, j, Firing rate modulation of (d) striatal MSNs, (g) SN GABA neurons and (j) SN DA neurons in relation to lever pressing during small and large reward sessions, and e, h, k for corresponding population average. f, i, l, The rate modulation for each press within the action sequence for (f) striatal MSNs, (i) SN GABA neurons, and (l) SN DA neurons.

Figure 5. Striatal-specific deletion of NMDA receptors disrupts the development of start/stop activity impairs sequence learning

a, Average lever pressing rate per session during three days of CRF followed by six days of FR8 training for striatal NR1-KO mutants and their littermate controls. b, Proportion of MSNs in striatal NR1-KO mutants and littermate controls displaying press-related activity during early and. late stages of training. c, Proportion of MSNs in striatal NR1-KO mutants and littermate controls exhibiting sequence start/stop related activity during the early and late stages of training. d, e, Example of the behavior microstructure of the same striatal-KO mouse during (d) the first day of FR8 training, and (e) the sixth day of FR8 training. All markers and insets are the same as used in Fig. 1a, b. f–i, Average (f) sequence length, (g) duration, (h) ISI and (i) within-sequence press rate for mutants and littermate controls. j–m, CV of sequence (j) length, (k) duration, (l) ISI and (m) within-sequence press rate during the first and sixth day of FR8 training for striatal NR1-KO mice and littermate controls.