Voltage-gated potassium channel Kv3.3 is present in the axon initial segments and the nodes of Ranvier of the murine Medial Nucleus of the Trapezoid Body in the auditory brainstem

Session

Poster Session Two

Authors

Dr. Kseniia Bondarenko (2, 3), Dr. Jonatan Nordmark (1), Prof. Ian Forsythe (3)

Affiliations

1. Karolinska Institute (Department of Clinical Neuroscience)
2. University of Edinburgh (Institute of Immunology and Infection Research)
3. University of Leicester (Department of Neuroscience, Psychology and Behaviour)

Keywords

Voltage-gated potassium channels; Kv3.3; Kv3.1; axon initial segments; nodes of Ranvier; confocal; expansion microscopy; clustering; MNTB; auditory brainstem

Abstract text

           The perception of sound in humans and other mammals relies on the ability of neurons to fire at high frequencies to convey temporally precise transmission about the encoded sound in the auditory pathway. Neuronal action potential (AP) firing requires voltage-gated sodium and potassium ion channels; the former mediate the AP upstroke, while the latter cause repolarisation. Fast potassium channels facilitate rapid repolarisation and sub-millisecond inter-spike intervals.

Voltage-gated potassium channels of the Kv3 subfamily (kcnc gene family) mediate fast repolarisation of APs, facilitating high frequency firing. Previously we have shown that only two (Kv3.1 and Kv3.3) out of four Kv3 subunit genes are expressed in the Medial Nucleus of the Trapezoid Body (MNTB) in the auditory brainstem. 

MNTB receives the encoded sound from globular bushy cells, which project thick axons ending in one giant synapse (known as the calyx of Held) on each principal MNTB neuron. The huge excitatory postsynaptic potential created by this synapse triggers a single action potential, maintaining the temporal precision of the propagating sound across this relay.

Previous reports suggested that MNTB neurons express Kv3.1 subunits in all key excitable structures: soma, axon initial segments (AIS), and nodes of Ranvier (NoR) (Berret et al., 2016), while Kv3.3 was observed only in the soma (Chang et al., 2007). However, Kv3.3 has a major role in action potential repolarization in the somata (Richardson et al., 2022), and we postulated that Kv3.3 might contribute and/or form heteromeric channels with Kv3.1 in the nodes. 

            The subcellular location of Kv3.1 and Kv3.3 subunits were examined using immunohistochemistry of 12 µm tissue cryosections from three murine genotypes: CBA WT, Kv3.1KO, Kv3.3KO. Kv3.1 (n=16) and Kv3.3 (n=16) staining at NoR and AIS in the MNTB was measured as the sum intensity across an area defined by Ankyrin G staining (AnkG, intracellular scaffolding protein coordinating AIS and NoR assembly; Leterrier, 2018).

Differences in Kv3.1 vs Kv3.3 subunit expression were assessed with subsequent data clustering using a Gaussian Mixture Model. Co-immunostaining with Kv3.1 and Kv3.3 antibodies was not possible due to cross-reactivity. The AnkG signal defined the width of NoRs that expressed either Kv3.1 (1540 nodes) or Kv3.3 (300 nodes) from two separate stainings: Kv3.1+AnkG and Kv3.3+AnkG. The AnkG signal level served as an internal control for staining quality, being similar between data sets and their means, confirming that neither of the sub-sets was affected by the quality of tissue.

Expansion microscopy (ExM) was used for sub-cellular localisation of the Kv3 subunits within the NoR. ExM is a recent addition to enhanced optical imaging protocols and is based on the physical expansion of the tissue embedded in a polyacrylate gel matrix (Chen et al., 2015). We used protein-retention ExM (proExM), achieving a stable 4.5x expansion of mouse brain tissue from 40-100 µm tissue cryosections.

Images were acquired using Zeiss LSM 980 Airyscan 2 confocal microscope with Plan-Apochromat 63x/1.40 Oil DIC f/ELYRA objectives and processed using ZEN 3.1 and Fiji software. Data visualization was performed using GraphPad Prism 9. For clustering, data associated with each feature were mean-subtracted and unit-variance scaled. Raw width values of NoRs associated with Kv3.1 and Kv3.3 were plotted against each other to determine the number of potential clusters in MATLAB.

            We found that Kv3.3 is present in AIS and the NoR, consistent with the recent electrophysiological studies showing that Kv3.3 plays a major role in neuronal excitability. We suggest that previous difficulties in detecting Kv3.3 in these structures were due to the different expression pattern compared to Kv3.1. We observed that Kv3.1 was expressed in both large and small diameter NoRs belonging to axons from globular bushy cells (giving rise to the calyx of Held) and other non-principal neuronal projections (making conventional small synapses), respectively. Kv3.3 was present in nodes with a significantly larger mean diameter (two-sample t-test). 

A comparison of the two distributions based on raw width values showed a significant difference between the means of the Kv3.1 and Kv3.3 subsets. The clustering study of the distribution of the NoR diameter suggested a bimodal pattern: nodes could be grouped into either large or small, which alone was sufficient to segregate NoRs into expressing either Kv3.1 or Kv3.3. We also observed an overlap between node widths in both subsets, suggesting that some nodes co-express Kv3.1 and Kv3.3 subunits. Intriguingly, Kv3.3 was absent from AIS in the Kv3.1KO but remained in NoR from these knockouts, consistent with the hypothesis of heteromeric Kv3 channels in the nodes of Ranvier.

 The Kv3.1 subunit was previously reported in the last nodes (heminodes) of axons forming Calyces of Held (Kim and Rutherford, 2016; Berret et al., 2016; Kim et al., 2020). However, the presence or absence of the Kv3.3 subunit in these structures remained unknown. Here, we qualitatively show that both Kv3.1 and Kv3.3 subunits are present in the calyx heminodes, specifically in the heminode walls, thereby contributing to the fast repolarization of action potentials. We did not perform the quantitative comparison of Kv3 expression levels due to the presence of the non-homogeneous tissue behind the heminodes in our tissue sections, especially in the Kv3.3 sub-set.

            Our recent work showed that Kv3.3 has a specific presynaptic function in regulating action potential duration at the calyx of Held synapse, in addition to its postsynaptic role in regulating somatic action potentials (Richardson et al., 2022). The current study shows that both Kv3.1 and Kv3.3 subunits are expressed in the key structures mediating transmission of excitability in the principal neurons of the MNTB: somatic membrane, nodes of Ranvier, and axon initial segments. Thus, we conclude that Kv3.3 is an integral part of the AIS and NoR in the MNTB, in contradiction to observations reported in the literature (Chang et al., 2007). Kv3.3 exhibits a different expression pattern to Kv3.1 in NoRs, where it predominantly resides in larger nodes, likely belonging to the axons from globular bushy cells. Kv3.3 is absent from Kv3.1KO AIS but not the NoR, suggesting that both Kv3.1 and Kv3.3 subunits are required to form functional Kv3 channels in axon initial segments of MNTB neurons.

References

1. Berret, E., Kim, S. E., Lee, S. Y., Kushmerick, C., and Kim, J. H. (2016). Functional and
    structural properties of ion channels at the nerve terminal depends on compact myelin. The Journal of physiology, 594(19):5593–5609
2. Chang, S. Y., Zagha, E., Kwon, E. S., Ozaita, A., Bobik, M., Martone, M. E., Ellisman, M. H.,
    Heintz, N., and Rudy, B. (2007). Distribution of kv3. 3 potassium channel subunits in distinct neuronal populations of mouse brain. Journal of Comparative Neurology, 502(6):953–972
3. Chen, F., Tillberg, P. W., and Boyden, E. S. (2015). Expansion microscopy. Science,
    347(6221):543–548
4. Kim, K. X. and Rutherford, M. A. (2016). Maturation of nav and kv channel topographies in the auditory nerve spike initiator before and after developmental onset of hearing function. Journal of Neuroscience, 36(7):2111–2118.
5. Kim, W. B., Kang, K.-W., Sharma, K., and Yi, E. (2020). Distribution of kv3 subunits
    in cochlear afferent and efferent nerve fibers implies distinct role in auditory processing. Experimental Neurobiology, 29(5):344
6. Leterrier, C. (2018). The axon initial segment: an updated viewpoint. Journal of Neuroscience, 38(9):2135–2145
7. Richardson, A., Ciampani, V., Stancu, M., Bondarenko, K., Newton, S., Steinert, J. R., Pilati N., Graham B. P., Kopp-Scheinpflug C. & Forsythe, I. D. (2022). Kv3. 3 subunits control presynaptic action potential waveform and neurotransmitter release at a central excitatory synapse. Elife, 11, e75219.