SorCS1-mediated Sorting of Neurexin in Dendrites Maintains Presynaptic Function

Luis Filipe Ribeiro, Ben Verpoort, Julie Nys, Kristel M Vennekens, Keimpe D Wierda, Joris de Wit

Preprint posted on February 17, 2019

What mechanisms instruct axon-dendritic polarity of neurons? The dendritic sorting receptor SorCS1 maintains cell compartment-specific protein composition.

Selected by Carmen Adriaens

Categories: cell biology, neuroscience


Neurons are polarized cells with two distinct compartments: dendrites and axons. These contain the post-and pre-synaptic specialization, respectively, which vary greatly in protein composition. Proper neuronal development, neuronal function and neuronal plasticity depend on the precise targeting of proteins to the right compartment. The bulk of protein synthesis in neurons occurs at the cell body. Thus, a central conundrum in Neuroscience remains: how are axonal and dendritic proteins differentially located to distinct compartments within the same cell?

Neurons have developed different mechanisms to ensure proper axonal protein targeting1. For instance, when protein subcellular distribution initially happens in a non-discriminatory fashion, protein cargo can be removed exclusively from the dendritic surface through a mechanism called selective endocytosis/retention2. As a result, protein cargo becomes enriched at the axonal surface. Alternatively, axonal cargo can be first inserted into the somatodendritic plasma membrane, and only then internalized from the cell surface and rerouted via endosomally derived carriers to the axon through a mechanism called transcytosis2.

In this preprint, Ribeiro et al. examined the role of the sorting receptor SorCS1 in the transcytosis-mediated axonal targeting of Neurexin (Nrxn), a presynaptic cell adhesion molecule essential for synaptogenesis and neuronal transmission3. SorCS1 belongs to the family of VPS10P-domain sorting receptors4, which are prominently expressed in the brain. They have recently emerged as key regulators of intracellular trafficking of synaptic receptors5–7 , but the mechanism through which they do this is yet uncharacterized. Here, Ribeiro and colleagues elucidate part of the cellular and molecular mechanisms that SorCS proteins use to regulate the trafficking of synaptic receptors.


Key Findings

First, the authors show that the axonal surface of mature cortical primary neurons displays twice as much endogenous Nrxn1a as the dendritic surface, indicating polarization towards the axon (see figure below). However, when the cells are permeabilized, Nrxn1a is mostly detected in dendrites. Permeabilization makes it possible to detect also intracellular proteins, indicating that Nrxn1a is trafficked to both compartments. To understand the mechanism responsible for axonal polarization of Nrxn1α the authors performed live-cell imaging experiments to monitor the transport of Nrxn1a throughout the secretory pathway. They showed that Nrxn1a is first trafficked to dendrites, appearing in the axon only after a long delay. Thus, they hypothesized that newly synthesized Nrxn1a is first trafficked to the somatodendritic compartment, where it is inserted into the plasma membrane. Indeed, when they monitored endocytosis, Nrxn1a was internalized from this membrane and further transcytosed, through the cell, from the dendritic surface to the axonal compartment.

Figure. (a) Surface endogenous Nrxn1a is axonally polarized in mouse cortical neurons. Red arrowheads indicate the axon and blue asterisks mark the cell body. (b) ratio of the axonal/dendritic HA intensity. DIV, days in vitro. Figure adapted from the preprint figure 3 by Ribeiro et al. made available under a CC-BY-NC-ND 4.0 International license.


One factor that could be involved in this process is SorCS1, an endosomal sorting receptor that cycles between the plasma membrane and the endosomal pathway4. Indeed, as previous work had shown that SorCS1 directly binds Nrxn in cis7, the authors tested whether this protein was required for Nrxn1α transcytosis. Loss of SorCS1 resulted in an accumulation of Nrxn1a on the dendritic surface and its depletion from the axonal surface, shifting Nrxn1α surface polarization from axonal to dendritic. This polarization defect was due to an impaired transition of internalized Nrxn1a from early endosomes (EEs) to recycling endosomes (REs) in the somatodendritic compartment. Normally, upon internalization from the dendritic plasma membrane, cargo proteins are incorporated into small carrier vesicles that fuse with existing EEs. From EEs, cargo is sorted to REs. When this transition is affected (like in the case of SorCS1 loss), cargo accumulates in EE, and is either degraded via lysosomes or recycled back to the dendritic membrane.

To gain more mechanistic insight into how SorCS1 functions, the authors set out to find factors that interact with this protein. They identified the endosomal associated protein, Rip11, as a new SorCS1 binding partner. Rip11 has been previously involved in sorting of cargo from early to recycling endosomes in non-neuronal cells, and here they showed that it is also required to maintain axon-dendritic polarity of Nrxn1a.

With these findings in mind, Ribeiro et al. proposed that SorCS1 and Rip11 form a protein complex that localizes to dendritic endosomes and sorts internalized Nrxn1α from early to recycling endosomes. This could explain the increase in surface dendritic levels of Nrxn1α and the concomitant decrease in surface axonal levels observed in Sorcs1 KO cells.

Finally, the authors tested the impact of genetic ablation of SorCS1, which decreases the axonal surface targeting of Nrxn1a, on Nrxn-dependent neuronal transmission and synaptogenesis. Indeed, depletion of SorCS1 compromised Nrxn-dependent synapse assembly and neuronal transmission. In summary, these findings expand the list of neuronal cargoes that undergo transcytosis, which so far is very limited2, and support the notion that SorCS1-mediated sorting in dendrites is required to maintain axon-dendritic polarity and normal neuronal function.


What do I think about this preprint / Open Questions?

I like this preprint a lot! It is a detailed study, and carefully performed. One experiment I particularly like, and that I think should become more common in cell biology research, is the endogenous tagging of proteins using CRISPR/Cas9. For instance, with this technique, the authors showed that Nrxn1α does not only localize to axons, but could also be found (albeit potentially temporarily) in the dendrites. As such, the authors could overcome the lack of good antibodies, and used these observations as the basis for their study. In addition, the live-cell imaging experiments in this work are very impressive. I thought the videos of cargo trafficking from one cellular compartment to the other in real life were awesome!

However, these observations also raise some questions that, in the future, may need further clarification. For instance, what is the biological function of the dendritic pool of Nrxn? Is dendritic Nrxn merely a reservoir of readily-synthesized protein to be delivered to the axonal membrane, or does dendritically localized Nrxn have a function in the dendritic compartment? As dendritic/postsynaptically expressed Nrxn has been shown to inhibit synapse formation8, it would be interesting to further study other possible functions of dendritic Nrxn, particularly of intracellular Nrxn, which is abundantly present in dendrites.

One important finding is that the axonal-dendritic balance of Nrxn is regulated by neuronal activity. Does this mean that SorCS1-mediated sorting can dynamically regulate Nrxn polarity? How do neuronal activity and SorCS1-mediated sorting of Nrxn crosstalk in order to ensure proper protein distribution and neuronal functioning? And further, how much can these Nrxn-specific observations be generalized to explain neuronal activity-dependent compartment diversification?

In summary, I think this study provides important insights into the mechanisms controlling the intracellular trafficking of a compartmentally polarized protein, which helps to better understand subcellular protein distribution in polarized cells. With the toolbox the authors developed for this study, I am convinced many more questions can be answered, and thus I look forward to read the follow up stories, too!



  1. Bentley, M. & Banker, G. Nat. Rev. Neurosci. 17, 611–622 (2016).
  2. Winckler, B. & Mellman, I. Cold Spring Harb. Perspect. Biol. 2, a001826–a001826 (2010).
  3. Südhof, T.C. Cell 171, 745–769 (2017).
  4. Willnow, T.E., Petersen, C.M. & Nykjaer, A. Nat. Rev. Neurosci. 9, 899–909 (2008).
  5. Ma, Q. et al. JCI Insight 2, (2017).
  6. Glerup, S. et al. Mol. Psychiatry 21, 1740–1751 (2016).
  7. Savas, J.N. et al. Neuron 87, 764–780 (2015).
  8. Taniguchi, H. et al. J. Neurosci. 27, 2815–2824 (2007).



Tags: axonal targeting, cell biology, cell compartment specialization, intracellular trafficking, neurexins, neuroscience, sorcs1, transcytosis

Posted on: 27th February 2019

Read preprint (No Ratings Yet)

  • Author's response

    Luis Ribeiro shared

    Thank you for highlighting our work!


    Q1: Is dendritic Nrxn merely a reservoir of readily-synthesized protein to be delivered to the axonal membrane, or does dendritically localized Nrxn have a function in the dendritic compartment?

    Perhaps one of the most puzzling issues that remains to be addressed is the biological function of this circuitous transcytotic trafficking route. One hypothesis that has been put forward is that transcytosis in neurons provides an efficient way of delivering receptors from a reservoir of readily-synthesized proteins (Winckler and Mellman, 2010). Like it has been suggested for other transcytotic axonal cargoes, either constitutively (Ascano et al., 2009; Leterrier, 2006) or in response to ligand-receptor interactions and signalling (Ascano et al., 2009; Yamashita et al., 2017), a dendritic pool of Nrxn may sustain the demand for receptors along the axonal surface. In agreement, neurotrophin signalling (initiated in axon terminals) is necessary for the transport of new TrkA receptors originated from the somatodendritic plasma membrane via transcytosis, increasing receptor availability along the axonal surface (Ascano et al., 2009; Yamashita et al., 2017).


    As you point out in the prelight, dendritically/postsynaptically localized Nrxn might also have a function in this compartment. So far, few hints in the literature have provided insights into a possible postsynaptic function of Nrxns at mammalian CNS synapses. One group suggested that postsynaptic Nrxn1 might interact with Nlgn1 in cis, thereby preventing trans-synaptic interactions of Nlgn1 with presynaptic Nrxn1 (“cis-inhibition hypothesis”). In agreement, postsynaptic expression (cell-autonomous) of Nrxn1 decreased the Nlgn1-dependent synaptogenic activity, reducing synapse density in vitro (Taniguchi et al., 2007). Rather than influencing synaptogenesis, others uncovered a role for postsynaptic Nrxns in NMDA receptor function. Although the exact mechanisms were not fully elucidated, the observed reduction in NMDA receptor-dependent postsynaptic currents in alpha Nrxns KO slices was shown to be cell-autonomous and likely reflects a change in the postsynaptic localization of NMDA receptors (Kattenstroth et al., 2004). Additionally, at the C. elegans neuromuscular junction, the ectodomain of postsynaptic Nrxn is proteolytically cleaved and binds to the presynaptic α2δ calcium channel subunit to inhibit presynaptic release (Tong et al., 2017). In summary, we cannot currently provide a definitive answer, but as suggested by the literature, it may very well be that Nrxns are first trafficked to dendrites, where they can provide a reservoir of Nrxn for the axonal membrane and/or modulate synaptogenesis and postsynaptic receptor function.


    Q2: How do neuronal activity and SorCS1-mediated of Nrxn crosstalk in order to ensure proper protein distribution and neuronal functioning? How much can these Nrxn-specific observations be generalized to explain neuronal activity-dependent compartment diversification?


    Although these are very interesting questions, we cannot currently answer them. Experimentally, we could show that a chronic increase in neuronal firing activity shifts the axonal/dendritic balance of Nrxn from axonal to dendritic. On the other hand, tetradotoxin (TTX) treatment, which decreases neuronal firing activity, augmented the axonal surface polarization of Nrxn. These observations nicely support a model in which neurons can dynamically change the axonal/dendritic surface balance of Nrxn in response to bidirectional changes in activity. However, if these bidirectional changes in neuronal activity can control neuronal compartmentalized distribution of other polarized proteins, to my knowledge, has never been tested. Interestingly, in Sorcs1 KO cells TTX treatment failed to increase the axonal surface polarization of Nrxn, which suggests that SorCS1-mediated sorting is required for TTX-dependent increase in axonal surface levels of Nrxn. Future work is required to understand how neuronal activity regulates SorCS1-dependent transcytosis of Nrxn.



    Ascano, M., Richmond, A., Borden, P., and Kuruvilla, R. (2009). Axonal Targeting of Trk Receptors via Transcytosis Regulates Sensitivity to Neurotrophin Responses. J. Neurosci. 29, 11674–11685.

    Kattenstroth, G., Tantalaki, E., Sudhof, T.C., Gottmann, K., and Missler, M. (2004). Postsynaptic N-methyl-D-aspartate receptor function requires -neurexins. Proc. Natl. Acad. Sci. 101, 2607–2612.

    Leterrier, C. (2006). Constitutive Activation Drives Compartment-Selective Endocytosis and Axonal Targeting of Type 1 Cannabinoid Receptors. J. Neurosci. 26, 3141–3153.

    Taniguchi, H., Gollan, L., Scholl, F.G., Mahadomrongkul, V., Dobler, E., Limthong, N., Peck, M., Aoki, C., and Scheiffele, P. (2007). Silencing of Neuroligin Function by Postsynaptic Neurexins. J. Neurosci. 27, 2815–2824.

    Tong, X.-J., López-Soto, E.J., Li, L., Liu, H., Nedelcu, D., Lipscombe, D., Hu, Z., and Kaplan, J.M. (2017). Retrograde Synaptic Inhibition Is Mediated by α-Neurexin Binding to the α2δ Subunits of N-Type Calcium Channels. Neuron 95, 326-340.e5.

    Winckler, B., and Mellman, I. (2010). Trafficking Guidance Receptors. Cold Spring Harb. Perspect. Biol. 2, a001826–a001826.

    Yamashita, N., Joshi, R., Zhang, S., Zhang, Z.-Y., and Kuruvilla, R. (2017). Phospho-Regulation of Soma-to-Axon Transcytosis of Neurotrophin Receptors. Dev. Cell 42, 626-639.e5.


    Have your say

    Your email address will not be published. Required fields are marked *

    This site uses Akismet to reduce spam. Learn how your comment data is processed.

    Sign up to customise the site to your preferences and to receive alerts

    Register here