Post-processed data for: Dopamine D1 receptor activation in the striatum is sufficient to drive reinforcement of anteceding cortical patterns

This study employs two-photon calcium imaging to investigate neural activity in mice during a neuroprosthetic task. Using two-photon microscopy, we captured GCaMP8s fluorescence at a wavelength of 920nm. Mice, head-fixed on a spherical treadmill, controlled a neural cursor by modulating activity in two motor cortex neuron ensembles. The task provided auditory feedback and measured neuronal activity using dF/F ratios. Contrary to traditional neuroprosthetics, we delivered activation of dopamine receptors with photoswitches instead of water reward. We explored various experimental conditions to assess the effect of receptor activation on task performance and neuronal activity. This data set contains the post-processed data.

Methods
We employed a Bruker Ultima Investigator system for calcium imaging, using a Chameleon Ultra II Ti: Sapphire laser tuned to 920nm to visualize GCaMP8s combined with a Olympus XLUMPLFLN 20XW objective and two GaAsP photomultiplier tubes.

Mice were secured in place on a styrofoam ball, where they could move freely under the two-photon microscope. We captured 512×512 pixel frames, approximately 300µm in size, at a rate of 29.7Hz using Prairie View software. The system enabled real-time communication between Prairie View and the BMI code running on Matlab 2017b, facilitating smooth experimental control and data acquisition.

We made manual adjustments during the recordings to address potential motion drifts. 

Calcium transients were extracted using Suite2p. Initially, an anatomical-only analysis captured the neuropil fluorescence signal, which was averaged and denoised to correct for photobleaching. Frames affected by photostimulation were identified and excluded from the second pass of Suite2p (anatomical + functional) .
In addition to the default Suite2p classifier results (probability threshold >0.1), we applied additional criteria to define a region of interest (ROI) as a neuron. The criteria included skewness of the neuropil-corrected fluorescence trace (0.4 to 10), neuron compactness (<1.4), ROI footprint (0 to 3), and the number of pixels (>80). We also required an SNR greater than two and stability, defined by less than 20% signal amplitude change and a standard deviation of the low-pass filtered signal under 1 during the experiment.

Imaging Setup and Data Acquisition

System: Bruker Ultima Investigator with Chameleon Ultra II Ti: Sapphire laser.

Wavelength: 920nm for GCaMP8s.

Detection: Two GaAsP photomultiplier tubes with an Olympus objective.

Head-Fixation: Mice were head-fixed on a styrofoam ball, allowing free movement.

Frame Details: 512x512 pixels, approximately 300µm, at 29.7Hz.

Software: Prairie View for imaging, Matlab 2017b for experimental control and data acquisition.

Motion Correction: Manual adjustments during recordings.

Neuroprosthetic Task

Neural Cursor: Mapped from motor cortex neuron ensembles E1 and E2.

Success Criteria: Neural cursor crossing a threshold (T1).

Auditory Feedback: Tone mapping corresponding to neural cursor values.

Task Duration: 30 minutes, with a 1-minute stabilization period.

Online Processing

Data Acquisition: PrairieLink with Matlab 2020b.

Activity Measurement: dF/F ratios, with dynamic baseline fluorescence (F0) calculation.

Experimental Conditions

Main Experiments: 'D1R activation', 'CONTROL', 'CONTROL_LIGHT', 'CONTROL_AGO', 'DELAY', 'RANDOM'.

Daily Selection: Randomly selected with a preference for 'D1R activation'.

Technical Issues: Variable mouse availability due to technical difficulties.

Optical Stimulation: Used in 'D1R activation',  'CONTROL_LIGHT', 'DELAY', and 'RANDOM'.

Conditions: Blue light for D1R activation, UV light for reset.

Image Pre-Processing

Software: Suite2p for calcium transient extraction.

Initial Analysis: Anatomical-only for neuropil signal, denoising for photobleaching.

Functional Segmentation: Exclusion of photostimulation-affected frames.

Manual Neuron Identification: Additional criteria for SNR, stability, and other metrics.

Motor Analysis

Setup: Head-fixed mice on a spherical treadmill.

Movement Recording: Positional coordinates smoothed and analyzed using Traja.

Threshold for Movement: Speed > 5mm/sec.

Please see manuscript for further methods.