SUMMARY

This thesis describes the research that was performed in the past five years concerning the generation of fast brain waves, and the cellular mechanisms that are involved. These fast brain waves, in the so called beta (15-30 Hz) and gamma (30-100 Hz) frequencies are mostly prominent while people are engaged in a task that requires working memory and attention. Therefore our rationale has been that the better understanding of the generation of these fast brain oscillations (which cortical area, which cortical layer, which cell types are involved, etc), will eventually lead to improved treatment of patients suffering from abnormal “attention”, such as children and adults with AD(H)D and patients suffering from schizophrenia.

We have approached this by inducing fast brain waves in brain slices from rats. This can done in very simple and direct way, namely via the application of the neurotransmitter acetylcholine, or a substitute such as carbachol. This same neurotransmitter is also released in the brain during tasks that require attention. Nicotine acts partly in a similar way. It is probably not by chance that ADHD and schizophrenic patients are severe smokers.

In chapter 2 and 3 we have investigated whether the small amplitude and frequency variations that can be seen in human EEG data (Figure 2.1b) are the result of the “normal” variation in the underlying system or network, or that these variations are in fact the result from multiple active networks. We investigated this by measuring the brain activity using multi-electrode grids of 8 × 8 electrodes with a high spatial resolution: the distance between electrodes was only 150 µ m, and the total area 1.1 mm 2 . Thus we could measure the activity in all layers of the cortex (Figure 2.3a,b). From these experiments it became clear that there are in fact two subnetworks that are active at the same time. One subnetwork gives rise to activity seen in layer 2 and layer 3/5, and the other to activity seen in layer 5 and 6 (Figure 2.3e). These two subnetworks give rise to oscillations that differ in frequency, layer 3/5 always being a bit faster than layer 6. Interestingly, both oscillations were present in layer 5. We wondered therefore if the nerve cells in layer 5 were divided in two subpopulations, one population involved in the slower and the other population involved in the faster oscillations, or that individual cells were contributing to both oscillations. To answer this question we recorded the activity from individual cells (Figure 2.4a,b). These measurements showed that the same cell was active and firing action potentials during the slow and fast oscillations (Figure 2.4d, 3.4, 3.5), due to the fact that these cells received timed synaptic input, both excitatory and inhibitory, during both oscillations (Figure 2.6, 2.7). This was in contrast to cells in layer 6, which only fired action potentials related to the slow oscillations (Figure 2.5). These results implicate that cells from layer 5 and 6 fire synchronously during the slow oscillation episodes, but not during the fast episodes. Layer 5 and layer 6 project to different brain areas: layer 5 primarily to a brain region called the hypothalamus that regulates the level of alertness. Layer 6 contacts the mediodorsal thalamus, from which the sensory input is send to the prefrontal cortex. Therefore, activation of the subnetwork that generates the fast oscillations could change alertness, while activation of the subnetwork that generated the slow oscillations could influence the input to the prefrontal cortex. The simultaneous activation of both subnetworks (Figure 2.3f,h) may reflect parallel information processing.

In chapter 4 we studies the interaction between two brain areas, namely between the prelimbic and the infralimbic cortex. Both brain areas are part of the prefrontal cortex and are important during tasks that require attention and/or working memory, but each in a different way. When prelimbic cortex functioning is disturbed, rats will make more mistakes in a test where a light indicates which of five holes contains a reward. The difficulty of the task being that there is a delay in time between when the light is shown and when the reward can be obtained. However, the disturbed function of the prelimbic cortex does not affect the number of premature responses: when the rat tries out a hole too soon after the presentation of the light, or before a light is even shown. This is reversed when the infralimbic cortex is affected: now the rats also perform worse than normal but due to an increase in impulsive responses. Given the fact that these brain regions are important in different aspects of attention behavior, we wondered whether this would result in different characteristics of fast brain oscillations, and whether a possible interaction between these connected areas could be revealed during these oscillations. We therefore used a multi-electrode grid of 8 × 8 electrodes covering an area of 4.4 mm 2, so that we could measure the electrical activity in both areas simultaneously. These experiments showed that the prelimbic and the infralimbic cortex generate their own specific oscillation in about 50% of the brain slices. The oscillations in the infralimbic cortex are faster (14.7 ± 0.9 Hz) than those in the prelimbic cortex (12.7 ± 0.7 Hz). The amplitude of the oscillations was also different between the two areas (Figure 4.1). To investigate whether these areas can really generate oscillation by themselves, we made so called mini slices: brain slices that contained only the prelimbic or only the infralimbic area. The measurements from these mini slices are surprising: they confirmed that both areas can generate oscillations independent from each other, but interestingly there is no longer a difference in oscillation frequency (in contrast: the difference in amplitude is still existing, Figure 4.4). Apparently it is the interaction between these areas that is causing the subtle difference in oscillation frequency. Interestingly, it is these kinds of interactions between brain areas that are abnormal in schizophrenic patients.