Nerve growth factor determines neuronal cell fate during development or after injury. A newly identified ‘death factor’, an unprocessed form of this protein, induces cell death by activating two receptors in concert.
The death of nerve cells is a key aspect in the establishment of functional neural circuits during development, but inevitably also features in injury or degenerative conditions. On page 843 of this issue, Nykjaer et al.1 provide some surprising molecular insights into cellular demise. Two unrelated cell-surface proteins, the p75 neurotrophin receptor (p75NTR) and sortilin, collaborate to induce death in responsive cells — including neurons. This joint effort involves direct interactions between p75NTR and sortilin, and is a consequence of the ability of sortilin to bind to an unprocessed form of nerve growth factor (NGF) as well as to the protein neurotensin. p75NTR has been implicated in cell death during both development and injury, but this unexpected partnership has profound consequences for our understanding of cell death in the nervous system, where these receptors cohabit on many of the same cells.
What is surprising about this ‘death system’ is that NGF was discovered as a factor that promotes the survival and growth of neurons, a finding for which Rita Levi-Montalcini and Stanley Cohen were awarded the Nobel prize in 1986 (ref. 2). Although it binds to p75NTR3, attempts to demonstrate that this receptor mediates NGF's effects largely failed. A second NGF receptor, TrkA, was later shown4 to mediate most of the survival and growth effects of NGF (Fig. 1a). So p75NTR, as it physically interacts with TrkA in some circumstances, was largely thought to have some sort of positive accessory role.
A breakthrough came when p75NTR was shown to be essential for the death of several cell types, including developing neurons4,5. But how could NGF promote survival through TrkA and death through p75NTR — sometimes in the same cell type? It soon became clear that the function of p75NTR depends on the cellular context; activation of p75NTR promotes death in numerous cells, including injured neurons, but promotes migration, growth and survival in other cells4.
In the next development, NGF was found to exist in both unprocessed (‘pro’) and mature forms. On some cells the mature NGF preferentially activates TrkA, whereas proNGF only activates p75NTR5. Importantly, proNGF is much more efficient than NGF at inducing the death of responsive cells, leading to speculation that the nature of the NGF itself is a key determinant of the ultimate outcome of p75NTR activation.
It is within this context that the paper of Nykjaer et al.1 now appears. Their results provide a molecular explanation for why p75NTR signals the downfall of some cell types but not others. First, they found that sortilin, a protein that shuttles other proteins within cells and acts as a receptor for neurotensin6, interacted with proNGF. Specifically, sortilin bound the ‘pro’ region of NGF, which is cleaved off during the formation of mature NGF, whereas p75NTR bound regions in mature NGF. Together, a sortilin–p75NTR complex bound proNGF (Fig. 1b). But preventing proNGF binding to sortilin blocked formation of this complex and stopped proNGF binding to p75NTR with high affinity. Such high-affinity interactions occur at the very low concentrations at which proNGF induces cell death.
Nykjaer and colleagues1 then examined whether sortilin was essential for proNGF to induce cell death through p75NTR. Inhibiting sortilin expression in neurons that express p75NTR, and that normally die when exposed to proNGF, prevented cell death. So what happened when sortilin was put into cells that don't normally have it? They died in response to proNGF. Thus, the presence or absence of sortilin determines whether or not p75NTR functions as a death receptor within a given cell type (Fig. 1).
These findings therefore represent a considerable advance in our understanding of cell death in the nervous system, and clarify many of the seemingly contradictory results obtained with p75NTR in different cell types. Sortilin is broadly expressed in the nervous system7, but p75NTR is present during development and when cells of many types are injured3. So this co-receptor system could have a major influence on whether neurons live or die.
Nykjaer and colleagues' study typically raises as many questions as it answers. First, sortilin is required for p75NTR-induced cell death in cultured cells, but does it function similarly in the whole organism? In fact, p75NTR activation does not inevitably cause death, even in cells expressing sortilin4. Is this because unprocessed forms of other NGF family members are not synthesized or secreted in their vicinity and/or because all p75NTR in these cells is bound to other receptors? In this regard, p75NTR binds at least two other receptors, TrkA and the Nogo receptor3,8,9. The latter binds proteins in myelin (which ensheathes neurons) that inhibit the regeneration of neuronal processes9. The interaction between p75NTR and TrkA enhances NGF binding to TrkA (Fig. 1a), and the interaction between p75NTR and the Nogo receptor is required for growth inhibition3,8,9 (Fig. 1c). So, the various and contradictory effects of p75NTR could come down to its choice of co-receptor in a particular cell, or even regions of those cells.
Second, the death signal that emanates from the p75NTR–sortilin complex still eludes us. As sortilin was first found to regulate protein movement within cells6, perhaps it chaperones p75NTR to specific compartments in cells for a rendezvous with death-inducing proteins. Third, little is known about how the expression of proNGF and neurotensin is regulated during development and injury. Nykjaer et al.1 show here that neurotensin prevents proNGF's actions, so the relative ratio of these two proteins can determine whether the p75NTR–sortilin complex induces cell death. Answering these questions will be essential for improving our understanding of the basic biology of the nervous system. It will also be necessary for designing effective treatments for the devastating neurological conditions that cause neuronal death.
Nykjaer, A. et al. Nature 427, 843–848 (2004).
Levi-Montalcini, R. Science 237, 1154–1162 (1987).
Hempstead, B. L. Curr. Opin. Neurobiol. 12, 260–267 (2002).
Kaplan, D. R. & Miller, F. D. Curr. Opin. Neurobiol. 10, 381–391 (2000).
Lee, R. et al. Science 294, 1945–1948 (2001).
Mazella, J. Cell Signal. 13, 1–6 (2001).
Sarret, P. et al. J. Comp. Neurol. 461, 483–505 (2003).
Kaplan, D. R. & Miller, F. D. Nature Neurosci. 6, 435–436 (2003).
McKerracher, L. & Winton, M. J. Neuron 36, 345–348 (2002).
About this article
Frontiers in Neuroscience (2019)
Journal of Internal Medicine (2018)
Expert Review of Ophthalmology (2017)
Effects of age and insulin-like growth factor-1 on rat neurotrophin receptor expression after nerve injury
Muscle & Nerve (2016)
Glioma invasion mediated by the p75 neurotrophin receptor (p75NTR/CD271) requires regulated interaction with PDLIM1