Single unit electrophysiology of optogenetically identified dopamine neurons in classical conditioning with probabilistic outcome of reward and punishment

We collected single-unit data (.mat file) from optogenetically identified dopamine neurons in the lateral VTA in mice. These dopamine neurons were recorded while animals performed 4 different variation of classical conditioning tasks with different ratio of probabilistic outcomes of water reward and aversive air puff. In one task, one cue was associated with both water and puff in a probabilistic manner.

Methods
We recorded extracellularly from multiple neurons simultaneously using a custom-built 200-μm-fiberoptic-coupled screw-driven microdrive with six or eight implanted tetrodes (four wires wound together). Tetrodes were glued to the fiber optic (Thorlabs) with epoxy (Devcon). The ends of the tetrodes were 350–500 μm from the end of the fiber optic. Neural signals and time stamps for behavior were recorded using a DigiLynx recording system (Neuralynx). Broadband signals from each wire filtered between 0.1 and 9000 Hz were recorded continuously at 32 kHz. To extract the timing of spikes, signals were band-pass-filtered between 300 and 6000 Hz. Spikes were sorted offline using MClust-3.5 software (David Redish). At the end of each session, the fiber and tetrodes were lowered by 20–80 μm to record new neurons. Sessions of recordings were continued until the tetrodes reached the bottom of the brain where no units were recorded and large fluctuations of voltage traces were recorded from tetrodes. After the completion of the recording sessions, tetrodes were moved up to the depth where units were recorded or the depth where light-identified dopamine neurons were recorded to ensure that the following electrolytic lesions were in the brain.

To verify that our recordings targeted dopamine neurons, we used ChR2 to observe stimulation-locked spikes (Cohen et al., 2012). The optical fiber was coupled with a diode-pumped solid-state laser with analogue amplitude modulation (Laserglow Technologies). Before and after each behavioral session, we delivered trains of 10 light pulses, each 5-ms long, at 1, 2, 5, 10, 20 and 50 Hz at 473 nm at 5–20 mW mm−2. Spike shape was measured using a broadband signal (0.1–9000 Hz) sampled at 30 kHz. This ensured that particular features of the spike waveform were not missed.