Abstract:
In this thesis, I have investigated the functional outcomes of the recently identified de novo KCNA2 mutations on neuronal firing behavior. To do so, induced pluripotent stem cells were derived from the fibroblasts of the patients carrying four different KCNA2 mutations as well as of healthy individuals. Patient and healthy control iPSCs were then differentiated into cortical excitatory neurons, and their electrophysiological features were assessed and compared on both single cell and network levels, using patch-clamp and microelectrode array techniques, respectively.
In the first part, the aim was to reveal the pathophysiological outcome of the KCNA2-T374A mutation that causes the most severe phenotype among KCNA2-mediated DEEs and was found to have both gain- and loss-of-function effect on the KV1.2 channels. The functional outcomes of the KCNA2-T374A mutation on single neuron physiology have been investigated longitudinally, from week one to four after the neuronal differentiation start. To do so, the patient iPSCs carrying the KCNA2-T374A mutation and the healthy control iPSCs were differentiated into neurons, and active and passive electrical properties of these neurons were assessed and compared over four consecutive weeks. Although the KCNA2-T374A patient-derived neurons’ input resistances decreased over the weeks, confirming changing ion channel composition of maturing neuronal membranes, the patient-derived neurons had higher input resistances during early development, indicating slower maturation of the neurons compared to the controls. At early timepoints (week two and three), the patient-derived neurons were found to have longer action potential time-to-peak durations and half-widths while their evoked firing response to depolarizing current stimulations were altered only at week four compared to the controls. In the second part of this thesis, the patient-derived neuronal populations carrying the KCNA2-T374A mutation was also examined on a network level, and they were found to be more active, with a significantly higher mean firing rate compared to the controls, from early time points (week two) on. They had higher bursting rates and lower burst spike frequencies at different time points during the development, that were observed to be comparable at later time points to the controls. Nevertheless, burst durations were observed to be prolonged from week three on, remaining to be the robust indication of hyperexcitable networks.
The KCNA2-R297Q mutation, that was identified in the patients with severe DEE phenotypes with generalized epilepsy and was found to have prominent gain-of-function effect on the channel function, also caused a strong phenotype in iPSC-derived neuronal cultures. Compared to the controls, the patient iPSC-derived neuronal populations had significantly increased mean firing rates on the network level already after one week of development in vitro. Increased burst rates (with a narrowing gap over weeks) and burst durations (with robustly strong difference over weeks) were observed consistently whereas the burst spike frequency was higher compared to the controls only at later stages of development. On a single cell level, these patient-derived neurons responded to depolarizing current stimulations differently than the controls. The change in the pattern of evoked action potentials from week four to six was also different than the observed pattern in control case, that was opposite to what is expected from developing neurons, revealing a pathophysiological difference in patient-derived networks. Moreover, the patient-derived neurons were found to have increased action potential time-to-peak and half-width durations, especially at later time points.
The KCNA2-L328V mutation, that was identified in a severely affected DEE patient and was found to cause both loss- and gain-of-function of the KV1.2 channels, gave rise to increased firing activities in patient iPSC-derived neuronal cultures compared to the controls. While the burst spike frequency was lower compared to the controls at earlier weeks, burst rates were higher in these patient populations, nevertheless by week six, which is the last examination time point in this study, both burst rates and burst spike frequencies were stabilized at the levels comparable to the controls. Interestingly, burst durations were observed to remain longer than those of the controls even at week six. Evoked firing response of the single neurons was different than that of the controls at both time points, week four and six, when these populations were examined on a single cell level. The change in the evoked firing response of the patient-derived neurons from week four to six, similarly to the neurons carrying the KCNA2-R297Q mutation, was in the opposite manner to what was observed in control cultures, indicating a crucial pathophysiological difference in firing behavior of developing patient-derived neurons.
The KCNA2-P405L mutation, that was identified in patients with milder DEE phenotypes with focal epilepsy and was found to have a pure loss-of-function effect on the channel function, also caused the mildest phenotype among the iPSC-derived neuronal populations carrying KCNA2 mutations. This mutation caused a developmental delay in neurons. Although the patient cultures always had comparable mean firing rates to the controls, appearance of bursting activity and localization of spikes within bursts, which are important steps in neuronal development, were observed with a delay. The mean input resistance of the patient-derived neurons was also found to be higher than that of the controls at an earlier time point, confirming the developmental delay of these patient-derived neurons, whereas at a later time point there was no difference anymore, suggesting a comparable ion channel composition on membranes of both patient and control-derived neurons by week six. Evoked firing patterns of patient-derived neurons was different from controls at week four, when the patient-derived neurons could fire only a smaller number of action potentials, whereas at week six there was no difference anymore. On a network level, although the bursting rate was not different in the patient cultures, burst spike frequencies were lower and the burst durations were consistently longer compared to the control-derived networks.
Overall, the KCNA2 mutations examined in this study have been found to cause prolonged burst durations in neuronal networks. Although the underlying mechanism leading to longer bursts must be different between GOF- and LOF-causing mutations, it has been revealed that the mutations with a gain-of-function effect on the channel function prolonged the action potential durations mainly slowing down the rising phase of action potentials. The patient-derived neurons carrying these mutations were found to have prolonged action potential time-to-peak and/or half-width durations at different time points they were examined. Although these differences were time-specific, all these patient lines had hyperexcitable phases because of widened action potentials. Even though the differences in single action potential parameters were not observed at later time points, the cultures remained to be more active, exhibiting prolonged bursts. These findings raised a question whether structural changes may have happened during hyperexcitable periods in early development, resulting in sustained hyperexcitability of networks at later time points, potentially due to increased synaptic connectivity. To follow up the findings of this thesis, further experiments not only on single cell and network levels but also on a transcriptomic level should performed to shed light on the cause of robustly prolonged burst durations.
Print-on-Demand