To investigate whether there are subsets of dopaminergic neurons that are responsible for “reward” or “punishment” in the absence of stimuli, different transgenic Drosophila lines were screened for approach or avoidance of neuronal activation. The different lines expressed the optogenetic “Chrimson” channel in different sets of dopaminergic neurons and were tested in different operant self-stimulating experiments (T-Maze (red light), T-Maze (yellow light), Y-Maze and JoyStick). Except for two, none of the tested lines displayed consistent preference or avoidance in all four different experiments. Both lines (TH-D’ and TH-D1) drive expression in the same DAN clusters including PPL1_FB, PPM2 and PPM3.

Since there are now more refined Gal4 driver lines, enabling expression of the optogenetic channel in only subsets of neurons compared to the original driver lines, T-Maze and JoyStick experiments were repeated with these new lines.

Rescreening in JoyStick and T-Maze showed that out of the three tested lines only one line displayed (tendencies for) light-avoidance as the original driver line. This was the line driving expression in the PPM2 cluster. (JoyStick data from other lines not shown, PPM3 could not be tested in T-Maze due to technical limitations). Interestingly in the JoyStick, this line displayed a shift in preference that was similar to the shift observed in the original TH-D’ driver line.

As mentioned above, JoyStick experiments had shown that after initial avoidance of the light (represented by negative PIs), in the last training periods flies would show PIs close to zero. In the T-Maze experiments flies are usually tested for only 1 minute. To find out whether fly preference would show a same development when choice time was extended, I conducted a set of T-Maze experiments, with a choice time of 10 minutes.

These results suggest that in the flies expressing the Chrimson channel in PPM2 DANs, prolonged choice time leads to a shift in “valence”, as also observed in the broadly expressing driver line TH-D’. This shift was only observed under yellow light.

Comparing control groups in yellow and red light, there seem to be differences in the activation of the Chrimson channel under red and yellow light. Whether this can be explained by different penetration capabilities due to the different wave lengths will be subject to further research.

Final Results

3IY; N=16

Sucrose; N=16

Buridan Test WTB; 5%Sucrose

PPM2 flies showed light-avoidance in T-Maze experiments when tested in yellow light for 1 minute. In the JoyStick setting, where flies are tested for 10 one-minute periods flies displayed a shift from negative (light-avoidance) values in the beginning to no preference in the last training period. For this reason I started a set of T-maze experiments where flies could decide between light and dark tubes for 10 minutes.

In the red light 1-minute test setting flies did not show avoidance of light as strong as in the yellow light setting, but still tendencies for avoidance were observable. Therefore another set of 10-minute experiments will also be conducted with red light.

Update 03.02.25: Added more flies to the experiments and included some control flies.

Update 09.02.25: Added more flies; excluded all experiments where combined number of flies in light, dark and elevator tube was less than 30. Also added first red light results.

Update 14.02.25: Experiments with yellow light are almost finished, red light still needs some work

Update: 22.02.25: Added more flies to red light.

13.0273 (Red)

For all Figures the left side displays all 10 testing and training periods for the JoyStick experiment. The right side compares the PIs of the last training period between the groups. Graphics indicate whether red or yellow light was used.

13.0273 (Yellow)

SIFa (Red)

SIFa (Yellow)

The left-hand side of the figure displays the choice indices (CIs) for the different groups tested for 1 minute in the T-Maze setting. On the right-hand side, the preference indices (PIs) for the final training period in the JoyStick setting are shown. Since flies of the TH_Flp_p10;64H06 line were not blind, they could not be tested in the T-Maze setting. The upper part of the figure refers to experiments conducted with yellow light, while the bottom part to experiments with red light, as indicated by the graphics.

In previous posts I referred to the different dopaminergic neurons (DA neurons) with the names of the driver lines used for the crossings. The table below the figure connects the fly strains to the targeted neurons and gives the reference. Gr28bd+TrpA1 target heat sensing neurons and acted as a control, since flies expressing the chrisom channel in these heat sensing neurons would avoid light activation. Flies were fed with all-trans retinal (ATR) for 2 days before the experiments, to enable light activation of the targeted neurons. For the negative control ethanol was used.

In the T-Maze experiments, flies were tested for 1 minute without prior exposure to light, whereas the JoyStick results reflect preferences after nine 1-minute training periods. Therefore, the T-Maze experiments should be repeated using longer testing periods. Additionally, PIs from the initial training periods in the JoyStick experiments will be included to allow for better comparison.

Hulse et al. eLife 2021: https://doi.org/10.7554/eLife.66039

Xie et al. Cell Reports 2018: https://doi.org/10.1016/j.celrep.2018.03.068

The positive control was tested once without food and once with food to observe its effect on general behavior, particularly on learning behavior. NF = No Food / WF = With Food