Tropomyosin Tpm3.1 is required to maintain the structure and function of the axon initial segment

Background

Neurons are highly polarized cells, composed of two distinct domains that have specialized form and function: the somatodendritic and the axonal compartments. This segregation is crucial for the neuron to ensure the correct electric information input (dendrites) and output (axon) within  the brain circuitry. One of the actors contributing to this polarity process is the axonal initial segment (AIS). Located within the proximal part of the axon, the AIS is the site for action potential initiation (due to its high concentration of voltage-gated channels) and is the physical boundary between the somatodendritic and the axonal domains. It allows the selective transport of axonal proteins and lipids through a membrane diffusion barrier and a cytoplasmic filter, preserving the identity of the neuron.

The AIS has a strong cytoskeleton network, based on compact shafts of microtubules and also actin filaments, which can be forming periodic rings (spaced by 190 nm) or patches. While a clear function for the microtubules in establishing and maintaining the AIS has been determined, the role of actin in these processes is still under study. Different articles have shown that, while intact actin is required for AIS formation, actin-disrupting drugs have no structural effects at the established and mature AIS. Does this mean that actin is not involved in AIS maintenance? The authors of this study have found that a specific tropomyosin isoform, Tpm3.1, localizes to the actin cytoskeleton of the AIS. The inhibition of Tpm3.1 in cultured neurons leads to a collapse of the actin cytoskeleton of the AIS, a disruption of the somatodendritic vesicles actin-based filter, a decrease of AnkG and sodium channel (Nav) levels and a reduced firing frequency.

 

Key findings

The authors found, by using structured illumination microscopy, that Tpm3.1 localizes to the AIS with a periodic pattern that is similar to the one displayed by the actin rings. Using phalloidin to co-stain the actin, they saw that the intensity profiles for both markers were periodic and partially matching, showing that Tpm3.1 localizes to the AIS actin cytoskeleton.

Actin is also present at the AIS in the form of patches, whose function has been associated with the filtering of somatodendritic vesicles. This study takes a step forward in characterizing the actin dynamics of these structures. Cultured hippocampal neurons were transfected with a photoactivable GFP-tagged actin (PAGFP-actin); by briefly illuminating the AIS area with a 405nm laser, the authors detected high and stable fluorescence points corresponding to the actin patches, showing the robustness of their structures. To go further into depolymerization dynamics, the authors focused on a particular AIS patch and compared its dynamics with patches outside the AIS and dendritic segments free of spines and branching points. By taking a picture every 3 seconds, they could see that the actin patches were highly stable as their time fluorescence decay was higher than the other groups. Altogether, the data shows that these patches are formed by very stable filaments with a low depolymerization rate. Importantly, Tpm3.1 was also detected at these patches, showing that this protein is localized to both actin structures of the AIS.

The authors also found that Tpm3.1 is required for maintaining the AIS structure, since the treatment of cultured neurons with two Tpm3.1 inhibitors (TR100 and Anisina) led to the decrease of AnkG fluorescence intensity versus the control (DMSO alone), as well as other AIS markers like EB1, TRIM46 and NF186. Importantly, these results were confirmed in cultured neurons from a Tpm3 knockout mouse model (which includes the Tpm3.1 isoform), which shows a decrease in AnkG. In light of this data, this work states the importance of Tpm3.1 for AIS maintenance in vitro and in vivo.

How does the inhibition of Tpm3.1 affect AIS function? First, the authors assessed if the actin-based filter for somatodendritic cargoes at the AIS was still existing after the inhibition of Tpm3.1. For that, they stained GluA1, a somatodendritic protein, in both the control (DMSO) and the inhibitor (TR100) treated neurons. They saw that GluA1 was invading the axon only in the TR100 neurons, showing that the filter is no longer capable of excluding somatodendritic proteins from the axon. The other important function of the AIS is action potential initiation and tuning, as it displays a high concentration of ion channels. The inhibition of Tpm3.1 led to a decrease in sodium channels on the AIS versus the control. By performing current-clamp experiments, the authors also found that this decrease in Nav channels had consequences at the action potential firing frequency, which was lower for the Tpm3.1 inhibitor-treated neurons. Thus, Tpm3.1 is crucial for the clustering of Nav channels at the AIS and for regular spike firing frequency.

Given the localization of the Tpm3.1 to the actin and its importance in AIS maintenance, the authors wanted to know the specific consequences of Tpm3.1 inhibition on the actin structures of the AIS. They found that the neurons treated with the inhibitors had a lower frequency of actin patches within the AIS, but also a less uniform distribution of actin rings, shown by an effect on their periodicity. Importantly, this last effect was not seen when the neurons were treated with latrunculin B, an actin-disrupting drug. Tpm3.1 is, therefore, necessary for actin cytoskeleton integrity in the mature AIS.

 

Why did I choose this preprint?

While more and more studies are showing the relevance of the AIS for correct neuronal function, the mechanisms of its composition, formation, maintenance and function remain largely unknown. This study shows that a new protein at the AIS, Tpm3.1, modulates the actin cytoskeleton and is key for AIS maintenance. I found it very impressive that the inhibition of Tpm3.1 provoked a partial perturbation on the periodicity of the actin rings in the established AIS, while most of the actin-depolymerizing drugs have no effect on them. At the same time, the absence of Tpm3.1 led to a decrease in the frequency of actin patches at the AIS, showing the role of this protein on the two specific forms of the AIS actin. This study shows, through Tpm3.1, that the actin cytoskeleton is necessary for AIS maintenance, something for which there was no evidence before.

 

Questions for the authors

• The inhibition of Tpm3.1 leads to a decrease of AnkG and also promotes the disruption of the somatodendritic vesicles filter at the AIS. It is known that this last effect is also triggered with the invalidation of AnkG. Is it possible that Tpm3.1 is promoting such an effect by disrupting the actin and decreasing AnkG by direct interaction? Do you plan to run interaction experiments between Tpm3.1 and AnkG / other AIS components, considering also that some of them show periodic patterns within the AIS, like Tpm3.1?

 

• You show that Tpm3.1 is expressed in the mature AIS, do you know if it is also true for the stages where the AIS is developing? In this case, would it be possible to downregulate Tpm3.1 at DIV0 with a shRNA and assess the consequences on AIS formation?

 

References

1. Song, A. et al. A Selective Filter for Cytoplasmic Transport at the Axon Initial Segment. Cell 136, 1148–1160 (2009).
2. Xu, K., Zhong, G. & Zhuang, X. Actin, Spectrin, and Associated Proteins Form a Periodic Cytoskeletal Structure in Axons. Science 339, 452–456 (2013).
3. Abouelezz, A., Micinski, D., Lipponen, A. & Hotulainen, P. Sub-membranous actin rings in the axon initial segment are resistant to the action of latrunculin. Biol. Chem. 400, 1141–1146 (2019).
4. Jones, S. L., Korobova, F. & Svitkina, T. Axon initial segment cytoskeleton comprises a multiprotein submembranous coat containing sparse actin filaments. J Cell Biol 205, 67–81 (2014).
5. Watanabe, K. et al. Networks of polarized actin filaments in the axon initial segment provide a mechanism for sorting axonal and dendritic proteins. Cell Rep 2, 1546–1553 (2012)

 

 

Posted on: 8 November 2019 , updated on: 27 November 2019

doi: https://doi.org/10.1242/prelights.15059

Read preprint (No Ratings Yet)