Motor skill learning and reward consumption differentially affect VTA activation

Abstract

Dopamine release from the ventral tegmental area (VTA) terminals in the primary motor cortex (M1) enables motor skill acquisition. Here, we test the hypothesis that dopaminergic VTA neurons projecting to M1 are activated when rewards are obtained during motor skill acquisition, but not during task execution at plateau performance, or by rewards obtained without performing skilled movements. Rats were trained to perform a skilled reaching task for 3 days (acquisition) or 7 days (plateau). In combination with retrograde labelling of VTA-to-M1 projection neurons, double immunofluorescence for c-fos and tyrosine hydroxylase (TH) was used to assess activation of dopaminergic and non-dopaminergic VTA neurons. Dopaminergic VTA-to-M1 projection neurons were indeed activated during successful motor skill acquisition, but not when rats failed to learn or had reached plateau performance, nor by food rewards alone. By contrast, dopaminergic VTA neurons that did not project to M1 were activated by both skilled reaching and food rewards. Non-dopaminergic neurons were found to be activated by motor task performance at plateau, but not during skill acquisition. These results indicate that distinct populations of VTA neurons are activated by motor skill acquisition and task performance. Moreover, this activation is not merely related to consumption of food rewards.

Figure 1

Overview of experimental setup and typical histological results. (A) Schematic drawing showing the front of the training apparatus and placement of food pellets (red) during pre-training and single pellet reaching (SPR) training for SPR rats (left part of the cage) and RC rats (right part of the cage). The pellet is placed within tongue-reach for SPR rats during pre-training and on a pedestal, that can only be reached by grabbing during single pellet reaching training. For RC rats, pellets are delivered to the inside of the cage whenever the SPR rat consumes one in both the pre-training and training phase. (B) Schematic of experimental setup for successful acquisition (SA), unsuccessful acquisition (UA) and plateau (PL) experiments. Numbers indicate time after FastBlue injection (FB). hab = habituation, training = single pellet reaching training. (C) Distribution of FastBlue in M1 after cortical injection. Representative photomicrographs showing dye injection and needle track in cortex (left) and extend of dye spread in sections 2.7-2 mm rostral of bregma (right). (D) Representative examples of VTA neurons positive for tyrosine hydroxylase (TH-FITC, green), FastBlue (FB, blue) and/or c-fos (c-fos-Cy3, red). Top row: 10x magnification, bottom rows: 40x magnification. Arrows indicate FB⁺ cell bodies. Asterisks denote position of c-fos⁺ nuclei.

Figure 2

Behaviour of SPR rats during task execution in SA (N = 8), UA (N = 6) and PL (N = 8) experiments. Values are mean ± s.e.m. (A) Learning curves showing percentage of successfully obtained pellets per training session. (B) Distance walked during the final training session. (***p ≤ 0.001 in Tukey’s HSD test) (C) Number of trials executed per minute during the final training session. (***p ≤ 0.001 in Tukey’s HSD test) (D) Number of grabbing attempts per trial in the final training session of SA (N = 8), UA (total and fail: N = 5 of 6, success: N = 2 of 6) and PL (N = 8) rats. 3 rats in the UA experiment did not achieve any successful trials.

Figure 3

c-fos expression in dopaminergic neurons during successful acquisition (SA, N = 8 per group). SPR = single pellet reaching, RC = reward control, CC = cage control. T = trained, U = untrained, L = left, R = right hemisphere. Asterisks indicate significant differences (*p < 0.05, **p < 0.01, ***p ≤ 0.001 in Tukey’s HSD test), values are mean ± s.e.m. (A) Superposition of TH⁺c-fos⁺ (green) and FB⁺ TH⁺c-fos⁺ (pink) neurons in VTA derived from 8 animals (2 sections per position per animal). Numbers indicate position relative to bregma. (B) Number of FB⁺ TH⁺c-fos⁺ neurons in rostral and caudal VTA sections. (C) Number of TH⁺c-fos⁺ neurons in rostral and caudal VTA sections.

Figure 4

c-fos expression in dopaminergic neurons during plateau task execution (PL, N = 8 per group). SPR = single pellet reaching, RC = reward control, CC = cage control. T = trained, U = untrained, L = left, R = right hemisphere. Values are mean ± s.e.m. (A) Superposition of TH⁺c-fos⁺ (green) and FB⁺ TH⁺c-fos⁺⁺ (pink) neurons in VTA derived from 8 animals (2 sections per position per animal). Numbers indicate position relative to bregma. (B) Number of FB⁺TH⁺c-fos⁺ neurons in rostral and caudal VTA sections. (C) Number of TH⁺c-fos⁺ neurons in rostral and caudal VTA sections. (***p ≤ 0.001 in Tukey’s HSD test).

Figure 5

c-fos expression in dopaminergic neurons during unsuccessful acquisition (UA, N = 6 per group). SPR = single pellet reaching, CC = cage control. T = trained, U = untrained, L = left, R = right hemisphere. Values are mean ± s.e.m. (A) Superposition of TH⁺c-fos⁺ (green) and FB⁺TH⁺c-fos⁺ (pink) neurons in VTA derived from 6 animals (2 sections per position per animal). Numbers indicate position relative to bregma. (B) Number of FB⁺ TH⁺c-fos⁺ neurons in rostral and caudal VTA sections. (C) Number of TH⁺c-fos⁺ neurons in rostral and caudal VTA sections.

Figure 6

c-fos expression in non-dopaminergic (TH⁻) VTA neurons. SPR = single pellet reaching, RC = reward control, CC = cage control. T = trained, U = untrained, L = left, R = right hemisphere. Values are mean ± s.e.m. (A) Superposition of TH⁻c-fos⁺ (red) neurons in VTA derived from 8 animals at plateau (2 sections per position per animal). Numbers indicate position caudal of bregma. (B) Numbers of TH⁻c-fos⁺ neurons in rostral and caudal VTA in SA rats (N = 8). (C) Numbers of TH⁻c-fos⁺ neurons in rostral and caudal VTA in UA rats (N = 6). (D) Numbers of TH⁻c-fos⁺ neurons in rostral and caudal VTA in PL rats. (**p < 0.01 in Tukey’s HSD test)