Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

This thesis aims at gaining insight into the performance of statistical methods that can be used to answer a research question which is common in neuroscientific experiments. This research question concerns the determination of the sustainability of long-term potentiation (LTP) which is an enhancement of the synaptic efficacy of neural cells that is experimentally introduced via electrical stimulation. Statistically, the sustainability of LTP can be determined by finding the time point at which the neural response returns to the level before stimulation, i.e. the baseline level. An experiment with mice in the research area mentioned was conducted by the electrophysiology group of the Laboratory of Biological Psychology at the KU Leuven (Laeremans et al. 2014). To answer the research question, Laeremans et al. (2014) applied a procedure that is based on the repeated application of the Wilcoxon matched-pairs signed-rank test. The performance of this procedure as well as the performance of the repeated application of Fisher's randomization test, the Friedman test and the runs test for serial randomness is investigated in a simulation study. The simulation model is based on the characteristics from data originating from a follow-up experiment of the one mentioned above. The performance measured for each statistical method is the percentage of correctly detected 'return to baseline' time points in 500 simulated data sets. Four different data generation models are used, which each imply different assumptions about the individual response profiles of a mouse. Furthermore, several parameters with respect to the data generation are varied within each of the models. The repeated application of the Wilcoxon matched-pairs signed-rank test applied in Laeremans et al. (2014) shows the best performance of the four statistical methods conducted. The performance of the procedure based on Fisher's randomization test is practically the same, but needs much more computational power and time such that it becomes infeasible with a larger sample size than 10 mice. Under conditions that occur in practice, i.e. with a sample size of 10 mice, the percentage of correctly determined 'return to baseline' time points for the Wilcoxon matched-pairs signed-rank test (as well as the randomization test) is low. To receive an acceptable performance with respect to finding the return to baseline, it is recommended to conduct experiments with a sample size of 20 mice. A further increase in sample size from 20 to 30 mice was found to not improve the performance of the method. The procedure based on the Wilcoxon matched-pairs signed-rank test should be conducted with a larger significance level than the standard level of 0.05. A significance level of 0.2 should be used in case only a small number of mice is available (e.g. 10). A smaller significance level can be used if the number of mice is larger (0.15 for 20 mice and 0.1 for 30 mice).

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Alzheimer's disease (AD) is a highly common neurodegenerative disorder, for which no disease-modifying treatment exists today. Pathological synaptic changes, most likely caused by soluble forms of the pathological proteins Aβ and tau and their interactions, are central and early events in AD. Synaptic dysfunction, altered neural network activity and a dramatic degeneration of neurons in important brain regions ultimately lead to dementia and other symptoms.In this thesis, we investigated synaptic function across multiple brain regions and hippocampal subregions in relevant mouse models for AD, using in vitro electrophysiology in brain slices. We started with control mice and found large regional differences, with much higher potentiation levels in the DG of ventral than dorsal slices. This was reflected in a specific ventral DG-deficit in AD mice, while they showed a specific dorsal deficit in the CA1 region. We also assessed regional effects on adult hippocampal neurogenesis and found that generally more new cells were generated in the ventral hippocampus, but the process appeared not to be affected in our AD models.Taken together, our results confirm that amyloid and tau pathology can independently exert synaptotoxic effects. This does not fully support the prevailing amyloid cascade hypothesis, in which amyloid drives and aggravates tau pathology. We also corroborate the sensitivity of LTP to detect early pathological changes caused by amyloid and/or tau in mouse models, long before the onset of plaques, tangles and severe behavioral symptoms.Finally, our results provide additional evidence that the subfields and dorsoventral segments of the hippocampus have different synaptic properties and vulnerabilities to disease, and that they should be clearly distinguished. Overall, the most plastic subregions appeared the most vulnerable to AD pathology. Provided that our findings in the murine hippocampus can be successfully translated to humans with AD, further defining which regions are the earliest affected and most vulnerable will be crucial for the development of preventive and therapeutic interventions, especially those based on the targeted neuromodulation of affected circuits.

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Cross-modal brain plasticity is a typical response of the mammalian brain to maximally compensate for the loss of one sensory function by recruiting and fine-tuning a remaining sense. It improves for instance tactile information processing in Braille-reading patients, and fine-tunes hearing capabilities in echo-locating blind individuals. This life-quality-improving structural and functional phenomenon occurs fast after sensory loss, often well before a bionic device is implanted. By consequence, the mammalian cross-modal sensory cortex fails to interpret the unimodal sensory information transmitted by the electronic device and precludes efficient restoration of the lost sensory function. Especially because our society is strongly vision-directed, neurobiological research questions with a focus on the molecular, anatomical and physiological underpinnings of cross-modal plasticity, are of particular interest. The adult monocular enucleation (ME) mouse model perfectly allows studying unimodal open-eye potentiation in the binocular visual cortex in parallel with whisker-triggered cross-modal plasticity in the medial monocular visual cortex, a process that is completed in seven weeks post injury. The first research question posed in this dissertation aimed at explaining the observation that social isolation abolishes the cross-modal reactivation and thus the reprogramming of the deprived visual cortex. In social species like the mouse, chronic isolation-stress can induce alterations in basic physiology, neuromodulator signaling and behavior. In the mouse model, end-point measurements of plasma corticosterone levels and typical behavioral tests for depression and anxiety in rodents, did not reveal overt differences between adult socially-isolated mice and group-housed mice. In contrast, the single housing had a major impact on the total neuromodulator concentrations in the stress-sensitive medial prefrontal cortex (mPFC), as well as in the somatosensory barrel field (S1BF) and the visual cortex (VC), two cortical areas that are crucially involved in the post-ME cross-modal plasticity phenomenon. We also revealed ME-specific effects on the neuromodulator levels in S1BF and VC, and a combined effect mainly on the mPFC and the VC. Based on these results, we conclude that social isolation exerts its impact by causing changes in serotonin, dopamine and/or noradrenaline modulatory function in the stress-sensitive mPFC and in the sensory cortices directly implicated in post-ME cross-modal plasticity, the VC and S1BF. These results point out that social interaction constitutes an important facet of cross-modal plasticity. We suggest that the implementation of 'social interaction-therapy' during the recovery of sensory loss could improve the communication between a unimodal bionic implant and the cross-modal mammalian brain by increasing multisensory information processing and -interpretation. The second research question posed in this dissertation focused on the role of neuromodulators in the post-ME reactivation of the VC. By performing pharmacological experiments, we discovered that 5-HTR1A exerts a role in the early unimodal open-eye potentiation of binocular cortex, while 5-HTR2A and 5-HTR3A are involved in the late cross-modal whisker-takeover of the monocular cortex. Our finding, that chronic administration of the 5-HTR2A antagonist ketanserin and 5-HTR3A antagonist ondansetron can suppress the cross-modal reorganization after partial vision loss in a cortical brain region- and layer-specific manner, sets the stage for the development of a 5-HT-assisted strategy that allows spatiotemporal control over maladaptive cross-modal cortical adaptations. These results could speed up the current evolution towards neurotransmitter-based bionic implants and may lead to higher success of the existing state-of-the-art retinal implants in functionally restoring vision. The third research question of this dissertation aimed at identifying the proteins and molecular pathways specifically underlying cross-modal plasticity in adult mice, in order to allow steering these cortical processes towards the desired outcome. By applying the ASBA-TMT differential proteomics approach, we revealed and successfully compared the cell surface protein expression patterns of the monocular cortex of single- and group-housed enucleated mice. Neuromodulator (G-protein coupled) receptors, ion channels and transporters displayed at the plasma membrane formed the focus of our study since these molecules mediate important brain functions and brain plasticity in particular. We could conclude that social isolation alters the expression levels of proteins functional in the synaptic vesicle cycle and that these disturbances potentially caused the incomplete cross-modal reactivation of the visual cortex in single-housed mice. Against our expectations, we did not detect plasma membrane proteins directly related to serotonergic neurotransmission, but this was explained to possibly result from either the low abundance of 5-HTRs in the mammalian neocortex, or from the difficulty to biotinylate and enzymatically digest GPCRs and ion-channels, which seems to be imposed by their structural composition and multi-transmembrane character. To conclude, this dissertation answers to a multitude of fundamental neuroscientific questions concerning the unimodal and cross-modal cortical responses to partial vision loss. Our data thereby significantly add to the current knowledge on sensory cortex plasticity. The proposed mechanism underlying ME-induced adult cross-modal plasticity infers a dynamic interplay between social interactions, serotonin and the synaptic vesicle cycle. The gathered insight opens up new avenues of inquiry, and may in future lead to novel therapeutic strategies that can fully restore any lost sensory function.