Molecular mechanisms for activity-dependent control of neuronal excitability in the central auditory pathway

Voltage-gated potassium channels of the Kv3 subfamily mediate fast repolarisation of action potentials, allowing neurons to fire at high frequencies. High-frequency firing is especially important in the auditory system, where fast and precise information transmission is crucial for the flawless perception of sound.

Previous experiments from the laboratory have shown that only two (Kv3.1 and Kv3.3) out of four Kv3 subunits are expressed in the Medial Nuclei of Trapezoid Body (MNTB) in the auditory brainstem. Tetraethylammonium (TEA) is known to block all Kv3 channels, but a key unknown is whether Kv3.1 or Kv3.3 subunits confer any unique properties on the Kv3 channels formed from these subunits. Since there are no subunit-specific antagonists, my project aimed to combine transgenic manipulation with light-activated pharmacology to investigate the roles of these subunits. This technology involved tethering a light-activated TEA moiety to the pore vestibule of Kv3.3 or Kv3.1; where the respective subunit had been mutated to possess a highly reactive cysteine substitution at the appropriate site. The hypothesis was that by blocking one specific subunit in a Kv3 channel, we could selectively suppress those Kv3 channels mediated by either Kv3.1 or Kv3.3 subunits, which should reveal any subunit-specific physiological functions.

This project investigated the action of light-activated pharmacology based on the photochromic tethered ligand MAQ. Electrophysiology was performed ex vivo on brainstem slices from CRISPR/Cas9-edited mice with the single amino acid substitution (for MAQ anchoring) in either Kv3.1 (E380C in mouse kcnc1) or Kv3.3 (N483C in mouse kcnc3) subunit. MAQ, tethered to either Kv3.1^E380C or Kv3.3^N483C, introduced a reversible light activated block of the Kv3 channel pore, studied in the principal MNTB neurons of transgenic lines. However, the portion of light-controlled potassium current was too small to answer scientific questions (8.6% of total potassium current at +40 mV). Therefore much of the experimental work revolved

around testing the materials and assumptions of the original hypothesis.

Several conditions were tested to determine the reason for the partial light-induced block by MAQ. The quality of the MAQ was verified using the NMR analysis of the synthesized compound. Homology modelling of the Kv3.3^N483C channel with docked MAQ ligand confirmed its binding ability when tethered to the channel pore. The tests with the non-specific photochromic ligand AAQ had shown that azobenzene moiety, a key part of both AAQ and MAQ, successfully undergoes a conformational change when switched between 380 nm and 500 nm using the current optical setup and thus, the set light intensity is enough to convert trans-MAQ molecule to its cis form. I also confirmed that the experimental conditions provide maximum achievable block and that the quality of the brain slice preparation did not undermine the effect achieved using the photo-activated pharmacology. However, the immunofluorescent studies showed that large portions of edited Kv3.1 were retained in the cytosol in the Kv3.1^E380C mouse, while edited Kv3.3 and non-edited Kv3.1 in Kv3.3^N483C mouse were almost absent, suggesting that Kv3.3 subunits are essential to achieve trafficking of Kv3 channels to the presynaptic terminal. Together, these alterations in Kv3 channel expression in CRISPR/Cas9-edited strains caused reduced photo-controlled portions of Kv3.1-specific and Kv3.3-specific potassium currents. In addition, MAQ showed strong neuronal toxicity that was not reported before.

Further immunofluorescent work revealed for the first time that the Kv3.3 subunit is present in the axon initial segments and the nodes of Ranvier. I also deployed the expansion microscopy technique to bypass the resolution limit of confocal microscopes combined with the superresolution post-acquisition algorithms to establish the role of the Kv3.1 and Kv3.3 subunit through their location in the subcellular structures of MNTB neurons. I found that Kv3.1 is mostly present in somatic Kv3 channels (post-synaptic MNTB cell membrane), while Kv3.3 is present in both somatic and presynaptic Kv3s. My findings support the hypothesis that Kv3.3 has a presynaptic function in regulating action potential duration at the synapse in addition to its postsynaptic role in regulating somatic action potentials. It also supported the electrophysiological evidence obtained previously in the lab.