Activation of the pathway in freely moving animals

As it was described in the previous section, selective activation of the glycinergic pathway in anaesthetized animals induced rhythmic firing in tonically-active TC cells, and synchronized TC activity after a desynchronizing painful stimulus. The evoked oscillatory activity in the cortex had increased power in the slow (1-6) frequency range. Considering the fast, strong and reliable effect of the inhibition, and the fact that the altered features of thalamocortical activity are also affected by the aforementioned ascending activating system, it is reasonable to ask if this prominent ascending inhibition had any behavioral manifestation.

To investigate this question, we used the previously described GlyT2::cre transgenic strain with transfected glycinergic neurons (n = 7). We implated screw electrodes over the ipsi- and contralateral frontal and parietal corticies to monitor changes in cortical activity during behavior, and an optic fiber into the ipsilateral IL to deliver laser pulses (Fig. 4.4.1).

44

Figure 4.4.1. Electrode arrangement for the activation of the glycinergic fibers in the IL thalamus in freely behaving animals.

After implantation, animals were left to recover from surgery for one week. Following the recovery period the optical fiber and the recording cable were connected 30 minutes before each recording session. During this 30 minutes, animals were kept in their homecage to get used to the wires and move naturally. After 30 minutes, mice were placed in a 42 x 52 cm open arena to test the effect of the activation of the glycinergic pathway on their behavior.

One recording session usually lasted for 50-60 minutes and 5-10 stimuli were applied.

In the arena at the begining of the sessions, mice spent most of their time near the walls, and as they got used to the environment they started to explore it. We delivered 30-seconds-long stimuli at 30 Hz first systematically separated by 5 or 10 minutes, and then at random times to avoid the effect of expectation caused by the patterned stimulus. No difference between these two paradigms could be found. During each stimulus we observed reduced bradykinetic motion, but the animals did not seem to loose consciousness, fall asleep or show freezing behaviour. To characterize the effect of the stimulus and to quantify the reduction in motion, we calculated the distance travelled by the animals during stimulated versus non-stimulated periods. The distance travelled was normalized to the distance travelled during the whole

45

session. We observed a 30.88% ± 4.68% reduction (n = 9 animals, Wilcoxon signed rank test, p = 0.008) in distance travelled compared to the pre-stimulus period (Fig. 4.4.2 a).

To better understand the nature of the evoked behavioral response and the network behind it, we tested if decreasing the power of the illumination (and thus the amount of glycinergic fibers activated) would evoke a graded response, or if the phenomenon is an all-or-none event. When we used maximal laser power, in most cases the animal showed complete behavioral arrest. Depending on the exact location of the optic fiber relative to the glycinergic terminals, the movement reduction varied among animals. When we reduced the intensity of the illumination, the distance travelled increased compared to the maximal intensity stimulation (Fig. 4.4.2 b), but instead of moving straight ahead, the animal turned contralaterally to the side of stimulation.

Figure 4.4.2. Behavioral effects of the photoactivation of glycinergic fibers.

a) Decreased distance travelled during 30 Hz activation of the glycinergic fibers in virus-injected compared to control animals. b) Gradual decrease of distance travelled with increasing laser intensities.

Together with the behavioral changes, we also monitored the parallel transitions in cortical activity. During exploratory behavior, as expected, we recorded high frequency asynchronous activity with small amplitude. Once the stimulation began, the increased dominance of slow frequency ranges could be detected on the wavelet transform, which was persistent during the whole stimulation. After the stimulus offset, both the altered cortical activity and behavior returned to the pre-stimulus state (Fig. 4.4.3). Note that this was also

46

the case regarding the firing pattern of the thalamic unit and the parallel cortical activity described previously in anaesthetised animals.

Figure 4.4.3. Cortical activity during stimulation in freely behaving animals.

a) Example traces of cortical LFP before (1) and during (2) photostimulation. b) Changes in the dominant frequency range of cortical activity before and during stimulus shown on a wavelet transform. Note the transient increase of slow frequencies in the signal.

To compare the effect of the stimulation on cortical activity across sessions, we standardized the LFP signal and then calculated the power spectra during and before each stimulation. In these power spectra pairs, we compared the difference in each bin in four animals. On figure 4.4.4 the difference in LFP power between control and stimulated conditions is shown in one representative animal. We detected a significant power increase in the bins between 2-6 Hz during stimulation compared to control (Mann-Whitney U-test for stimulation sections and paired control periods, p < 0.01)

47

Figure 4.4.4. Power spectrum of the cortical LFP during stimulus and control.

In freely moving experiments, we have shown that, by activating the glycinergic fibers, a prominent behavioral effect could be evoked during which all ongoing behavioural activity was suspended. By gradually decreasing the intensity of the illumination, the effect was also gradually decreased. Parallel to the behavioral changes, altered cortical activity could be detected. Both the behavioral effect and the cortical activity change were transient, lasted only during the stimulation, and no long term state transition was observed. These results, together with the IL thalamic unit recordings in anesthetized animals, imply that the effect of the inhibition is motor-related and conveyed to the cortex via the IL thalamus.

48