Neurotrophin channels excitement

Sodium-ion channels usually open in response to a voltage change across the membrane in which they sit. But, surprisingly, a growth factor secreted by neurons also rapidly triggers the opening of a specific sodium channel.

Ion channels are remarkable proteins that regulate the transfer of ions across the lipid bilayers that make up cellular membranes. These proteins are required for a variety of physiological processes; for instance, a group of cation-selective channels is responsible for conducting action potentials along nerve cells, allowing them to transfer information from one place to another. In vertebrates, the ions that are typically responsible for the propagation of action potentials are sodium ions. These are temporarily allowed to enter neurons through channels in the cell membrane that open as a result of a transmembrane voltage change. Surprisingly, however, Konnerth and colleagues1 now report (page 687 of this issue) that a brain protein secreted by neurons is involved in opening a particular sodium channel — one already known to occur in sensory neurons of the peripheral nervous system.

The secreted protein in question is a growth factor called brain-derived neurotrophic factor (BDNF), and the story began three years ago when Konnerth and colleagues2 found that BDNF causes membrane depolarization within milliseconds. (Membrane depolarization refers to a shift in the potential difference across the membrane that, beyond a certain threshold, can trigger action potentials.) This was a remarkable finding because, until then, only neurotransmitters had been found to have such a rapid effect on the membrane potential of neurons. Typical neurotransmitters are low-molecular-weight compounds such as acetylcholine and the amino acids glutamate and glycine. They act on a millisecond timescale because they bind to receptors that are also ion channels, causing them to open briefly.

By contrast, BDNF is a much larger molecule, consisting of two identical chains of about 100 amino acids each, and neither of the two known receptors to which it binds is an ion channel. In fact, one BDNF receptor is an enzyme, the transmembrane tyrosine kinase TrkB, which, with a certain delay, mediates some of the typical biological activities of BDNF — such as the prevention of neuronal death3 (Fig. 1). Yet Konnerth and colleagues clearly showed that BDNF causes rapid membrane depolarization, and that briefly exposing TrkB-expressing neurons to puffs of BDNF triggers trains of action potentials2. Other neurotrophins that have different receptor specificities failed to do this.

Figure 1: Brain-derived neurotrophic factor (BDNF) is a versatile protein.

There are two receptors for BDNF on the surface of neurons: the neurotrophin receptor p75, which binds all neurotrophins, and TrkB. Each receptor alone binds BDNF with nanomolar affinity, but when they associate, the affinity and selectivity of TrkB for BDNF increase. TrkB can also associate with two different ion channels. TRPC3 is a non-selective cation channel that needs to be phosphorylated by TrkB to open; this is a slow process (acting in the range of minutes). Nav1.9 is selective for Na+ ions and, as Konnerth and colleagues1 show, opens within milliseconds following the binding of BDNF to TrkB. The figure also shows some of the functional outcomes of the activation of the BDNF receptors.

Although surprising, these results helped to explain previous observations indicating that BDNF modulates synaptic transmission4 (neurons typically communicate at synapses, which are specialized junctions). One particularly famous example of such modulation is the long-term potentiation (LTP) of synaptic transmission. This is a widely used cellular model of memory, and numerous studies have shown that in the hippocampus — a brain structure crucial to memory formation in rodents and humans — BDNF plays a role in LTP that is essentially as significant as that of glutamate4. The observed BDNF-induced membrane depolarization triggered an increase in the levels of Ca2+ ions in the protrusions (spines) of postsynaptic neurons — those on the receiving side of the synapse — and immediately induced LTP when paired with a burst of synaptic stimulation5.

In an impressive reconstitution experiment, Konnerth and colleagues1 now show which molecules are needed for BDNF to rapidly depolarize neurons. Briefly, when the authors created fibroblast cells containing a particular subtype of Na+ channel, Nav1.9, together with the BDNF receptor TrkB, they found that BDNF caused an inward flow of Na+ ions, and hence membrane depolarization, within milliseconds. This is simply too rapid to be attributed to the enzymatic activity of TrkB. Moreover, in these experiments, Nav1.9 could not be activated by voltage changes alone, although the related channel Nav1.7 could. So the explanation must be that binding to BDNF produces a conformational change in TrkB that is transferred without delay to Nav1.9, such that it briefly becomes permeable to Na+ ions. Finally, Konnerth and colleagues found that, in hippocampal neurons, the targeted elimination of Nav1.9 blocks BDNF-induced depolarization. This implies that these neurons, at least, contain Nav1.9 and require it for the BDNF-mediated influx of Na+ ions.

The unexpected conclusion that a secreted growth factor can modulate the opening (gating) of an ion channel on the same timescale as a voltage change raises several questions. The Nav1.9 channel is best known for its presence in small sensory neurons of the dorsal-root and trigeminal ganglia6, where, together with Nav1.8, it may play a key part in the physiology of neurons involved in pain perception6,7. The expression of Nav1.9 is regulated by another growth factor, glial-cell-derived neurotrophic factor (GDNF). These sensory neurons do not express TrkB, so it will be interesting to see whether one component of the GDNF receptor, the tyrosine kinase Ret, is also associated with Nav1.9, and whether GDNF causes rapid changes in membrane potential. If so, this could have implications for our understanding of pain sensing.

In a similar vein, most Nav1.8-expressing and many Nav1.9-expressing neurons also express TrkA7 — the tyrosine kinase receptor for nerve growth factor (NGF). NGF has been associated with pain sensing in many ways8, and one wonders whether it may trigger the rapid opening of Nav1.8 or Nav1.9 via TrkA.

Another question is whether the BDNF-induced, TrkB-mediated depolarization seen in all neurons of the central nervous system (CNS) tested so far can be attributed to Nav1.9. Although Konnerth and colleagues' latest report1 indicates that Nav1.9 is expressed by hippocampal CNS neurons, the levels of expression are probably much lower than those in sensory neurons6. Finally, it has not yet been demonstrated biochemically that TrkB associates directly with Nav1.9, although TrkB does associate with other transmembrane proteins3,9, such as the neurotrophin receptor p75 and the nonspecific cation channel TRPC3 (Fig. 1). Intriguingly, the binding of BDNF to TrkB also increased the permeability of the TRPC3 channel9, although this required the enzymatic activity of TrkB and was considerably slower (minutes instead of milliseconds).

Thus, one of the last remaining distinctions between the physiological properties of neurotrophins and neurotransmitters seems to have disappeared. Work on neurotransmitters may become a source of inspiration for researchers interested in the neurophysiology of neurotrophic factors, and one of the next questions may be how the effects of BDNF on ion channels are rapidly inactivated. Mechanisms involving receptor internalization might be too slow to eliminate this poorly diffusible protein, which is also fairly resistant to protein-degrading enzymes. But BDNF-induced desensitization, as occurs with neurotransmitter receptors, may be an attractive possibility to explore.


  1. 1

    Blum, R., Kafitz, K. W. & Konnerth, A. Nature 419, 687–693 (2002).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Kafitz, K. W., Rose, C. R., Thoenen, H. & Konnerth, A. Nature 401, 918–921 (1999).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Bibel, M. & Barde, Y.-A Genes Dev. 14, 2929–2937 (2000).

    Article  Google Scholar 

  4. 4

    Poo, M. M. Nature Rev. Neurosci. 2, 24–32 (2001).

    CAS  Article  Google Scholar 

  5. 5

    Kovalchuk, Y., Hanse, E., Kafitz, K. W. & Konnerth, A. Science 295, 1729–1734 (2002).

    ADS  CAS  Article  Google Scholar 

  6. 6

    Dib-Hajj, S., Black, J. A., Cummins, T. R. & Waxman, S. G. Trends Neurosci. 25, 253–259 (2002).

    CAS  Article  Google Scholar 

  7. 7

    Benn, S. C., Costigan, M., Tate, S., Fitzgerald, M. & Woolf, C. J. Neurosci. 21, 6077–6085 (2001).

    CAS  Article  Google Scholar 

  8. 8

    Apfel, S. C. Clin. J. Pain 16, 7–11 (2000).

    Article  Google Scholar 

  9. 9

    Li, H.-S, Xu, S.-Z. S. & Montell, C. Neuron 24, 261–273 (1999).

    CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Yves-Alain Barde.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Barde, Y. Neurotrophin channels excitement. Nature 419, 683–684 (2002).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Sign up for the Nature Briefing newsletter for a daily update on COVID-19 science.
Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing