# Neurotrophins, Trk Receptors and p75

> A developing neuron survives when the neurotrophin its target supplies reaches a Trk receptor, a receptor tyrosine kinase that starts survival signals. A neuron reads a neurotrophin through two kinds of receptor: three Trk kinases, each preferring particular factors, and one shared receptor, p75, from the TNF-receptor family. This lesson matches the four mammalian neurotrophins to their receptors, uses mouse knockouts to show which neurons need which pair, traces how the receptors were found, and explains how the uncut precursor of NGF can kill a cell that the trimmed protein would keep alive. Afterwards you can predict how a neuron is likely to respond from the form of the factor and the receptors it carries.

Course: Developmental Neuroscience (Survival and competition). Level: advanced university and graduate. About 25 minutes.
Lesson page (link this one when citing): https://stemhop.com/lessons/trk-and-p75/

Builds on: The Neurotrophic Hypothesis Tested · Too Many Neurons: Naturally Occurring Cell Death

## What you'll learn

- Match NGF, BDNF, NT-3 and NT-4 to their preferred Trk receptors, explain how a factor dimer activates a Trk kinase to start survival signals, and describe p75 as a shared receptor of the TNF-receptor family.
- Use mouse knockout phenotypes to predict which neuron populations depend on a given neurotrophin or receptor.
- Explain why mature NGF acting through TrkA supports survival while uncleaved proNGF acting through p75 and sortilin can trigger apoptosis, why the cell's balance of receptors changes the outcome, and how the concentration of mature NGF changes which binding sites it fills.
- Describe how p75 and the Trk receptors were identified, including the trk oncogene and the 1991 papers that tied Trk to NGF.

**Many developing neurons live only if their target supplies enough of the right neurotrophin, and each neuron reads that factor with a Trk receptor, a kinase that switches on survival signals.**

The [last lesson](https://stemhop.com/lessons/neurotrophic-hypothesis/) left you with a rule: a target makes a limited amount of a survival factor, and neurons that fail to get enough die. This lesson follows the factor to the receptors that read it.

## The layout

Most of this lesson happens at one cell. Picture a sympathetic neuron in the superior cervical ganglion, a cluster of nerve cell bodies in the neck. Its axon runs out to a target, such as a blood vessel or a gland, that makes only a small amount of nerve growth factor (NGF). Many neurons send axons to that same target.

NGF is one of four mammalian **neurotrophins**: NGF, brain-derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3) and neurotrophin-4 (NT-4). Each is a **dimer**, two identical chains held together, and all four share one fold.

A neuron reads them with two kinds of receptor.

- **Three Trk receptors** (pronounced “track”): TrkA, TrkB and TrkC. Each is a receptor tyrosine kinase, and each prefers particular factors.
- **One shared receptor, p75** (also written p75NTR). It binds all four factors and has no kinase. It belongs to the family of the tumour necrosis factor (TNF) receptor, and it is a relative of Fas, the death receptor from [that lesson](https://stemhop.com/lessons/death-receptors-bcl2-caspases/). Like Fas, it carries a **death domain**, a stretch inside the cell that can bind proteins that lead toward apoptosis, though not the same proteins Fas uses.

| Neurotrophin | The Trk receptor it prefers |
| --- | --- |
| NGF | TrkA |
| BDNF | TrkB |
| NT-4 | TrkB |
| NT-3 | TrkC, and more weakly TrkA and TrkB |

All four also bind p75. NGF is made first as a longer precursor, proNGF, with an extra stretch at its front, the pro-domain, which proteases cut off to leave the mature protein. What p75 does depends on its partners: with TrkA it can work alongside the kinase, and with sortilin, a membrane protein that grips the pro-domain, it can kill.

The map raises two puzzles.

1. **How does a neuron know which factor is its own?**
2. **How can one factor keep one cell alive and kill another?** Add mature NGF to cultured sympathetic neurons and they survive. Add the same protein before its pro-domain has been cut off, and in Barbara Hempstead’s 2001 experiments the neurons died.

> **The one idea to carry through this lesson.** The target supplies the factor; the neuron’s receptors decide what it means. A Trk receptor with its ligand usually reads a neurotrophin as “stay alive”. p75 can read it as “die” in some cells, and with its partner sortilin it reads the uncut precursor that way. Cutting the precursor removes the stretch sortilin grips, so the cut changes which receptors the factor works through best.

**What the rule covers.** The survival and death of developing neurons, and of neurons and glia in culture, in the systems used here: sympathetic and sensory ganglia in mice, cultured sympathetic neurons, oligodendrocytes and the chick retina. It has three limits. “Decides” means the whole set of receptors and how much of each, not one receptor alone. It does not say how much uncut precursor reaches neurons in a normal embryo, which is still unsettled. And it leaves out the neurotrophins’ other jobs, such as axon growth, BDNF acting through TrkB at adult synapses and p75’s roles in axon growth and pruning, as well as reports that TrkA and TrkC without ligand can signal death in some neurons.

## How a Trk receptor turns on

**Film transcript: Four Factors, Three Kinases**

### Film transcript: Four Factors, Three Kinases

A factor lands on the outside of a neuron. How does that switch on a signal inside? Start with the receptor. A Trk receptor sits in the neuron's membrane. Outside, it binds the factor. Inside, it carries a tyrosine kinase domain. Now the factor. Each neurotrophin is a dimer, so it has two binding faces. A dimer has two faces. What can it do to two receptors at once?

It can hold two Trk receptors at once, and bring them together. A crystal structure shows this. One NGF dimer holds two TrkA domains, one on each side. Their kinase domains, now side by side, phosphorylate each other on tyrosines. The phosphorylated tyrosines become docking sites for proteins inside the cell. They start two survival branches you have met before. One runs from Ras to ERK. The other runs from PI3K to Akt. Now back to the first puzzle. How does a neuron know which factor is its own? Each neurotrophin prefers particular Trk receptors. NGF prefers TrkA. BDNF and NT-4 prefer TrkB. NT-3 prefers TrkC, and activates TrkA and TrkB more weakly.

The selective differences lie in a few contact surfaces, such as the factor's amino terminus. The overall plan is the same in another pair, NT-4 with TrkB. So the answer is this. A neuron answers the factor whose Trk receptor it makes. A neuron that makes TrkA is switched on by NGF. One that makes TrkC is switched on by NT-3. The factor does not know which neuron it is for. The receptor sorts it. And p75? It binds NGF at a site different from where TrkA binds. What it does with that binding is the rest of this lesson. The takeaway: the factor pairs two receptors, and the receptor a neuron makes decides which factor it hears.

From the film

- A neurotrophin dimer can hold two Trk receptors together so that their kinase domains phosphorylate each other.
- The phosphorylated tyrosines start two survival branches, Ras to ERK and PI3K to Akt.
- Each neurotrophin prefers particular Trk receptors, so a neuron answers the factor whose Trk receptor it makes.

The factor dimer has two binding faces, so it can hold two Trk receptors at once and bring them together. Their kinase domains, now side by side, phosphorylate each other on tyrosines, as in [the lesson on receptor tyrosine kinases](https://stemhop.com/lessons/receptor-tyrosine-kinases/). The phosphorylated tyrosines become docking sites that start two survival branches you have met: the Ras to extracellular signal-regulated kinase (ERK) cascade ([that lesson](https://stemhop.com/lessons/ras-to-mapk/)) and the phosphoinositide 3-kinase (PI3K) to Akt route ([that lesson](https://stemhop.com/lessons/pi3k-akt-insulin/)).

*Figure (AI-generated illustration).* Many axons share one target that makes only a little NGF. At each axon tip the NGF dimer brings two TrkA receptors together, their kinase domains phosphorylate each other, and the Ras to ERK and PI3K to Akt survival branches start. Image: a sympathetic ganglion sends several axons to one blood vessel that releases a few NGF dots. An inset at the axon tip shows three steps: an NGF dimer and two separate TrkA receptors; the dimer holding both receptors together while their kinase domains phosphorylate each other; and arrows from the phosphorylated pair to Ras to ERK and PI3K to Akt.

That answers the first puzzle in outline: a neuron answers the factor whose Trk receptor it makes. A neuron that makes TrkA is switched on by nerve growth factor (NGF), one that makes TrkC by neurotrophin-3 (NT-3). The factor does not know which neuron it is for; the receptor sorts it.

## Where the receptors came from

**Film transcript: The Kinase from a Tumour**

### Film transcript: The Kinase from a Tumour

Why does a growth factor for nerve cells have a receptor named after a tumour? Start with the first neurotrophin receptor anyone cloned. It could not explain survival. In 1986, Moses Chao and colleagues transferred human DNA into mouse cells. They found cells that made the human receptor, and cloned its gene. Eric Shooter's laboratory reported the rat receptor the next year. This protein, p75, bound NGF with low affinity. It had no kinase. Neurons were known to have higher affinity sites as well. On its own, p75 could not account for them. The kinase came from somewhere else entirely: a human colon tumour. In 1986, Mariano Barbacid's laboratory described an oncogene from that tumour.

Part of a tropomyosin gene had fused to the gene for a tyrosine kinase. The fusion kept the kinase on, without any ligand. The name comes from that fusion: tropomyosin receptor kinase. The normal gene encoded a membrane receptor with no known ligand. If DCC from the netrin lesson comes to mind, it should. Here again, a gene first studied in cancer had a job in wiring the nervous system. Now a test. Take a fibroblast that makes no Trk receptor and ignores NGF. Give it the gene for the normal Trk protein. Then add NGF. What should happen?

In spring 1991, two groups showed that NGF binds the Trk protein and causes its tyrosine phosphorylation. Barbacid's group also found that fibroblasts carrying Trk divided in response to NGF, without needing p75. The orphan kinase from a tumour was a receptor for NGF. Earlier, in Barbacid's laboratory, Ruediger Klein went looking for the Trk gene in mouse brain. In his later account, he recovered a related gene instead. It was reported in 1989 as TrkB. A later search, with Fabienne Lamballe, found a third relative, TrkC, reported in 1991. A search for one receptor had turned up a family of three. That is why the original Trk protein is now called TrkA.

One question stayed open. How p75 and TrkA together give high-affinity binding is still argued. The takeaway: the survival receptor was found in cancer, and NGF turned it on.

From the film

- The first neurotrophin receptor cloned, p75, bound NGF with low affinity and had no kinase.
- The trk gene was first found as an oncogene from a human colon tumour, and in 1991 two groups showed that NGF binds the Trk protein and causes its tyrosine phosphorylation.
- The original Trk protein is now called TrkA, because two relatives, TrkB and TrkC, were found soon after.

**The first neurotrophin receptor cloned could not explain survival.** In 1986 Moses Chao and colleagues isolated the gene for the human nerve growth factor (NGF) receptor by transferring human DNA into mouse cells, finding cells that made the human receptor, and cloning the gene. Eric Shooter’s laboratory reported the rat receptor in 1987. This protein, p75, bound NGF with low affinity, and it had no kinase. Neurons were known to have higher-affinity sites as well, and on its own p75 could not account for them. Which protein makes those sites was the puzzle of the 1980s.

**The kinase came from a tumour.** In 1986 Mariano Barbacid’s laboratory described an oncogene from a human colon carcinoma in which part of a tropomyosin gene had fused to the gene for a tyrosine kinase ([that lesson](https://stemhop.com/lessons/oncogenes/)). The fusion kept the kinase on without any ligand. The normal gene, *trk*, encoded a membrane receptor with no known ligand. If deleted in colorectal cancer (DCC) from [the netrin lesson](https://stemhop.com/lessons/netrin-and-dcc/) comes to mind, it should: here again a gene first studied in cancer turned out to have a job in wiring the nervous system.

**Predict.** A fibroblast makes no Trk receptor and does not respond to NGF. Give it the gene for the normal Trk protein, then add NGF. What should NGF do to the Trk protein, and what would that tell you about what Trk is sufficient for?

> **Check your prediction.** In spring 1991 two groups showed that NGF binds the Trk protein and causes its tyrosine phosphorylation: David Kaplan, Barbara Hempstead, Dionisio Martin-Zanca, Moses Chao and Luis Parada, and, separately, Ruediger Klein, Barbacid and colleagues. Barbacid’s group also found that fibroblasts carrying Trk divided in response to NGF without needing p75. The orphan kinase from a tumour was a receptor for NGF, and in those cells it was enough for the factor to be read.

The original Trk protein is now called TrkA, because two relatives, TrkB and TrkC, were found soon after. How p75 and TrkA together give high-affinity binding was disputed and is still not settled; Going deeper follows the dispute.

## Three deletions, three populations

**Film transcript: Delete One, Count the Survivors**

### Film transcript: Delete One, Count the Survivors

If you remove one receptor from a mouse, which neurons disappear, and when? A knockout mouse is one in which a chosen gene has been disrupted. It makes no working protein, or no working form of one protein. Delete each neurotrophin or Trk gene, one at a time. The family sorts by which neurons are lost. Start with TrkA, and the sympathetic ganglion in the neck. Anne Fagan and colleagues counted its cells at several ages. At embryonic day 13.5, the count was normal. If TrkA is needed for survival, not for making the neurons, what should happen next?

The mutant fell steadily behind. At birth it had about half as many neurons as control ganglia. By postnatal day 9, nearly all of its principal neurons were gone. Over those embryonic ages, dividing cells were no fewer than in controls. Dying cells, with shrunken nuclei, increased. So the neurons were made, and then they died. Now the other pairs. Mice lacking NGF or TrkA lose sympathetic neurons. They also lose many small sensory neurons that sense pain and heat. Mice lacking BDNF lose vestibular neurons, which sense balance and head movement. They also lose many nodose-petrosal neurons, which carry signals from the heart, lungs and gut. Mice lacking NT-4 lose nodose-petrosal neurons too.

And mice lacking NT-3 or TrkC lose the sensory neurons that report where a limb is. These mice move and hold their limbs abnormally. In this grid, lose means fewer neurons. Usually part of a ganglion survives, but some groups are lost almost entirely. Each pair has its own pattern of loss. But the map has overlaps. NT-3 prefers TrkC, yet it also activates TrkA and TrkB. Should a mouse lacking NT-3 lose more neurons, or fewer, than one lacking TrkC?

Expect more. In many ganglia at the same age, mice lacking NT-3 lose more neurons than mice lacking TrkC. That fits NT-3 also acting through TrkA and TrkB. The size of the gap depends on the ganglion and age. Some populations also need more than one factor, at different stages. The takeaway: each factor and receptor pair has its own pattern of loss. In the sympathetic ganglion, neurons were made, then lost.

From the film

- In mice lacking TrkA, sympathetic ganglion neurons were made and then died, leaving about half as many neurons as controls at birth.
- Each factor and receptor pair has its own pattern of neuron loss, and some populations need more than one factor at different stages.
- In many ganglia, mice lacking NT-3 lose more neurons than mice lacking TrkC, which fits NT-3 also acting through TrkA and TrkB.

A [knockout mouse](https://stemhop.com/methods/gene-targeting-mice/) is one in which a chosen gene, or part of it, has been disrupted so that it makes no working protein, or no working form of one protein. Deleting each neurotrophin or Trk gene in mice, one at a time, sorts the family by which neurons are lost. Start with the best-followed case.

**Predict.** In mice lacking TrkA, Anne Fagan and colleagues followed the superior cervical ganglion through development. At embryonic day 13.5 it had a normal number of cells. If TrkA is needed for the neurons to survive, rather than to be made, what should the neuron counts, the dividing cells and the dying cells do from then on?

> **Check your prediction.** The mutant ganglion fell steadily behind: about 15% fewer cells at day 15.5, 35% fewer at day 17.5 and about half as many neurons as control ganglia at birth. Over those embryonic ages, dividing cells (mitotic figures) were no fewer than in controls, while dying cells, with shrunken, condensed (pyknotic) nuclei, increased. By postnatal day 9 the ganglion had lost nearly all of its principal neurons. The neurons were made and then died.

The other pairs give a map. “Lost” means fewer neurons: usually part of a ganglion survives, but some defined groups, like the TrkA-null sympathetic neurons above, are lost almost entirely.

| Mice lacking | Neurons lost |
| --- | --- |
| NGF or TrkA | sympathetic neurons; many small sensory neurons that sense pain and heat |
| BDNF | vestibular neurons, which sense balance and head movement; many nodose-petrosal neurons, which carry signals from the heart, lungs and gut |
| NT-4 | nodose-petrosal neurons |
| the kinase-containing form of TrkB | neurons in both the central and peripheral nervous systems |
| NT-3 or TrkC | proprioceptive afferents, the sensory neurons that report where a limb is; the mice move and hold their limbs abnormally |

One result marks the edge of the map. In NGF-null mice the basal forebrain cholinergic neurons still develop.

Each factor-receptor pair has its own pattern of loss, though deleting a factor and its main receptor removes many of the same neurons. The four factors share a fold and a co-receptor, but to a neuron they are separate signals.

**Predict.** Neurotrophin-3 (NT-3) prefers TrkC, but it can also activate TrkA and TrkB. Should a mouse lacking NT-3 lose more neurons or fewer than a mouse lacking TrkC?

> **Check your prediction.** Expect more. Removing TrkC takes away only NT-3’s action through TrkC, while removing NT-3 also takes away its weaker help through TrkA and TrkB. In many ganglia compared at the same age, NT-3-null mice do lose more neurons than TrkC-null mice, though the size of the gap depends on the ganglion, the age and which form of TrkC was deleted. The map has overlaps like this, and some populations need more than one factor at different stages.

## The receptor that can kill

**Film transcript: The Receptor That Can Kill**

### Film transcript: The Receptor That Can Kill

Every neurotrophin binds p75. So what does p75 do on its own? In the mid-1990s, two groups found that it can kill. Barde's group showed this in the developing chick retina. Chao's group reported that NGF kills cultured rat oligodendrocytes. These are the myelin-making cells of the brain and spinal cord. They lack TrkA. With no Trk receptor to read stay alive, p75 can read die. The oligodendrocyte experiments were done in culture, where NGF is often added at nanomolar levels. The retina result involved NGF the tissue makes itself. Chapter six shows why dose matters. Yet the precursor has something mature NGF lacks: a pro-domain that sortilin grips, so a p75 and sortilin pair can hold it tightly.

Like most secreted proteins, NGF is made as a longer chain, proNGF. The extra stretch at its front is called the pro-domain. The accepted view had been that the pro-domain is trimmed inside the cell. In 2001, Ramee Lee, Pouneh Kermani, Kenneth Teng and Barbara Hempstead added to that picture. Cells can secrete some proNGF uncut. Enzymes outside the cell can then cut it. Plasmin can, a protease made from the blood protein plasminogen. So can particular matrix metalloproteinases. To test the precursor alone, mutate its cut sites so proteases cannot trim it. Take cultured cells with p75 and sortilin, but little TrkA. Give them mature NGF, or the uncleavable precursor, at the same concentration.

Then block p75. Predict each result. In this worked example, at a low dose, mature NGF added no death. The uncleavable precursor killed the cells. An antibody that blocks p75 prevented that killing. So did cells from mice lacking p75. So the precursor kills through p75, without being trimmed. In Lee's experiments, uncut proNGF also triggered apoptosis of cultured sympathetic neurons. It bound p75 tightly, and activated TrkA very little. Mature NGF kept the same kind of neuron alive, through TrkA. In 2004, Anders Nykjær, Claus Munck Petersen and colleagues found the missing partner. p75 needs sortilin to deliver this death signal. Sortilin grips the pro-domain, while p75 grips the mature part.

This killing was shown in culture and in rodent injury models. How much it adds to the normal deaths of development is less clear. Another group found that uncleavable proNGF kept sympathetic neurons alive, through TrkA. It did so less well. The takeaway: p75 can read a neurotrophin as die. With sortilin, it reads the uncut precursor that way.

From the film

- p75 can read a neurotrophin as die, as when NGF killed cultured oligodendrocytes, which lack TrkA.
- In the cells tested, uncleavable proNGF killed through p75 and needed sortilin, which grips the pro-domain.
- How much precursor killing adds to normal developmental death is less clear, and one group found proNGF could keep sympathetic neurons alive through TrkA, though less well.

Every factor binds p75, yet nothing so far says what p75 does on its own. In the mid-1990s two groups found that it can kill. Barde’s group reported that nerve growth factor (NGF) kills cells in the developing chick retina through p75, and Chao’s group reported that NGF kills cultured rat oligodendrocytes, the myelin-making cells of the brain and spinal cord, which lack TrkA. A cell with p75 and no Trk to read the factor as “stay alive” can read it as “die”. In the retina the NGF was the tissue’s own: an antibody to NGF reduced the death. The oligodendrocytes were given NGF in culture, where doses are often nanomolar; the last section shows why that matters.

### The uncut precursor

Like most secreted proteins, NGF starts as a longer precursor, and the accepted view had been that its pro-domain is trimmed inside the cell. In 2001 Ramee Lee, Pouneh Kermani, Kenneth Teng and Barbara Hempstead reported that cells secrete proNGF uncut, and that it can be cut outside the cell by plasmin, a protease made from the blood protein plasminogen, and by particular matrix metalloproteinases.

To ask what the precursor does on its own, mutate its cut sites so that proteases cannot trim it, and compare it with mature NGF at the same concentration.

**Predict.** Fill in the table before you open the answer. It is a simplified worked example of the comparisons Lee and colleagues (2001) and Nykjær and colleagues (2004) reported. The first test cells express p75 and sortilin but little TrkA, and both forms are given at a low dose at which mature NGF adds no death. (Much higher doses of mature NGF can act through p75, as the last section shows.)

| Test cells | Mature NGF (same low dose) | Uncleavable proNGF | Uncleavable proNGF, p75 blocked or absent |
| --- | --- | --- | --- |
| p75 and sortilin, little TrkA | ? | ? | ? |
| Cultured sympathetic neurons (TrkA) | ? | ? | — |

> **Check your prediction.** | Test cells | Mature NGF (same low dose) | Uncleavable proNGF | Uncleavable proNGF, p75 blocked or absent |
> | --- | --- | --- | --- |
> | p75 and sortilin, little TrkA | no added death | more cells die | no added death (antibody to p75, or cells from p75-null mice) |
> | Cultured sympathetic neurons (TrkA) | survive, through TrkA | died in Lee and colleagues’ experiments; another group found weak survival | — |
>
> In Lee and colleagues’ experiments uncut proNGF bound p75 tightly, activated TrkA very little, and triggered apoptosis of cultured sympathetic neurons through p75. A precursor that cannot be cut still kills, so it does not need to be trimmed first.

In 2004 Anders Nykjær, Claus Munck Petersen and colleagues showed that, in the cells they tested, p75 needs sortilin to deliver this death signal. Sortilin grips the pro-domain while p75 grips the mature part of the factor, so the pair can hold the precursor tightly. p75 alone binds mature NGF at least as well as it binds the precursor.

*Figure (AI-generated illustration).* Mature NGF works best through TrkA and keeps a TrkA-carrying neuron alive. The uncut precursor keeps its pro-domain, which sortilin grips while p75 grips the mature part, and in cells with little TrkA this pair can trigger death. Image: two membrane patches side by side. Left, mature NGF held between two TrkA receptors, with an arrow to 'Stay alive'. Right, uncut proNGF whose pro-domain tails are gripped by sortilin while p75, with a death domain inside the cell, grips the mature part, with an arrow to a dashed box 'Can trigger death'.

Keep the scope the papers give. Precursor killing was shown in cultured neurons, in cells of the oligodendrocyte lineage and in rodent injury models. How much it contributes to the normal deaths of development is less clear. Fahnestock and colleagues found that uncleavable proNGF can keep sympathetic neurons alive through TrkA, though about fivefold less well than mature NGF, so the outcome depends on the cell’s balance of TrkA, p75 and sortilin. What happens after the death signal starts is the subject of [that lesson](https://stemhop.com/lessons/death-machinery-in-neurons/).

## Reading one factor two ways

**Film transcript: Reading One Factor Two Ways**

### Film transcript: Reading One Factor Two Ways

Mature NGF keeps neurons alive. So how could it kill cells that lack TrkA? And why did the precursor kill neurons in one lab, but keep them weakly alive in another? Simple binding arithmetic helps with the first question. The precursor needs its pro-domain. Binding studies on chick sensory neurons found two classes of NGF site. Their dissociation constants were about 20 pM and 2 nM. Treat these as orders of magnitude, about ten picomolar and one nanomolar. The two kinds of site are a hundredfold apart. A dissociation constant, K d, is the free concentration that fills half the sites. Below it, fewer than half fill. Pick a low level for illustration, ten picomolar of mature NGF.

Which sites fill? About half of the Trk sites fill. Only about one percent of the p75 sites do. So at low levels, a much larger fraction of the high-affinity sites fills than of the p75 sites. Survival depends on signals this count leaves out. Now turn the dial to about 2 nM. Culture experiments often add NGF in this range. The Trk sites are nearly full. And roughly half to two thirds of the p75 sites fill too. So in a dish, mature NGF fills many p75 sites. That fits it killing TrkA-lacking cells there. But filling a site is not the same as signaling. The model has no threshold for survival or death. Nor can it explain the precursor. At a low dose where mature NGF added no death, the precursor killed.

What the precursor adds is the pro-domain. Sortilin grips it, and in the cells tested, the precursor's death signal needed sortilin. This lesson puts no binding numbers on proNGF. Its effects are compared by experiment instead. So the cut shifts the balance. It does not flip a switch. Why did the precursor kill in one lab, and support weak survival in another? That is still open. The dose, the receptor balance, and how much precursor was cut during the experiment could each matter. Three variables now go a long way. First, whether the factor has been cut. Second, which receptors the cell makes. Third, how much of each. Change any one, and the same factor can mean something else.

Remember the rule. The target supplies the factor. The receptors decide what it means. Next, see how competing neurons use p75 against each other. The takeaway: dose changes which sites fill. The cut changes which receptors the factor works through best.

From the film

- At low NGF, a much larger fraction of the high-affinity Trk-linked sites fills than of the p75 sites, but at about 2 nM many p75 sites fill too.
- Binding arithmetic cannot explain the precursor, which adds a pro-domain that sortilin grips.
- Whether the ligand has been cut, which receptors the cell makes and how much of each go a long way toward deciding what one neurotrophin does.

The second puzzle now has two halves that seem to clash: mature nerve growth factor (NGF) keeps sympathetic neurons alive yet killed TrkA-lacking cells through p75, and proNGF killed sympathetic neurons in one lab but kept them weakly alive in another. Binding arithmetic helps with the first clash. The second turns on the pro-domain, and why the labs disagreed is still open (Going deeper).

Binding studies on chick embryo sensory neurons found two classes of NGF site, with dissociation constants (Kd) of about 20 picomolar and 2 nanomolar. Treat these as orders of magnitude: about 10 pM (10⁻¹¹ M) for the high-affinity sites and about 1 nM (10⁻⁹ M) for p75 alone, a hundredfold apart. The model treats the high-affinity sites as Trk-linked, setting aside whether p75 helps make them. Kd is the free concentration at which half the sites are occupied. For simple binding:

> fraction of sites occupied = [NGF] / ([NGF] + Kd)

**Predict.** Give mature NGF to a cell at 10 pM, a low concentration chosen for illustration. About what fraction of the Trk-linked sites fill, and what fraction of the p75 sites? Now raise NGF to about 2 nM. Which sites gain the most?

> **Check your prediction.** | Free mature NGF | High-affinity (Trk-linked) sites | p75-alone sites |
> | --- | --- | --- |
> | 10 pM | 50% | 1% |
> | 100 pM | 91% | 9% |
> | about 2 nM | about 100% | about 50–67% |
>
> Raising NGF from 10 pM to about 2 nM fills the p75 sites the most, from almost empty to roughly half to two-thirds full: 67% with the rounded 1 nM Kd, 50% with the measured 2 nM.

Targets supply small amounts of factor, so the high-affinity sites are likely the ones that matter most for survival. Culture experiments often add NGF at around 50 nanograms per millilitre, which for a dimer of about 26.5 kilodaltons is about 2 nM. At that level many p75-alone sites are occupied, which fits mature NGF acting through p75 in TrkA-lacking cells in a dish.

Keep the model’s limits in view. The percentages are fractions of each class of site, not counts of receptors or strengths of signal, and the model has no threshold for survival or death. Nor can it explain the precursor: in the worked example, at a dose where mature NGF added no death, the uncleavable precursor killed. What the precursor adds is the pro-domain, which sortilin grips. This lesson puts no binding numbers on proNGF; its effects are compared by experiment instead.

Put the pieces together and three variables go a long way toward deciding what one neurotrophin does at one cell.

1. **Whether the ligand has been cut.** Uncut proNGF can kill through p75 with sortilin; mature NGF works best through Trk. Neither form is confined to one receptor, as both clashes showed. Enzymes inside and outside the cell decide which form arrives.
2. **Which receptors the cell makes.** Its Trk receptor picks the factor. p75 binds every factor, and what it does with them depends on partners such as sortilin and TrkA.
3. **How much of each it makes.** The ratio of p75 and sortilin to TrkA changes what the same ligand does. With TrkA, p75 may also raise NGF binding in some cells, a point still argued.

### Design problem

Your purified “NGF” kills cultured cells that express p75 but little TrkA, while a second batch keeps sympathetic neurons alive. Design the experiments that find out why.

> **One way to do it.** Start with the protein itself. Run both batches on a gel and probe a western blot with an antibody to NGF, and with one to the pro-domain. If the killing batch carries a larger band that the pro-domain antibody recognises, it contains uncut proNGF.
>
> Next, separate the two forms. Make a proNGF with its cleavage sites mutated so that proteases cannot trim it, and test it beside pure mature NGF at the same concentration on both kinds of cell. Prediction: the uncleavable precursor kills the cells with p75 but little TrkA, while mature NGF does not. On sympathetic neurons mature NGF should support survival; the precursor may kill them or support them weakly, since groups have found both.
>
> Then test the receptors. Block p75 with an antibody, or use cells from p75-null mice, and block sortilin, for example with an inhibitor that blocks its binding to the pro-domain or by knocking it down. If killing by the precursor disappears when either partner is gone, the precursor acts through p75 with sortilin.
>
> Last, ask how the first batch came to contain precursor. Add plasmin to it; if the killing activity falls as the precursor band is converted to mature NGF, the difference between the batches was processing, not two different genes.

## Putting it together

Both puzzles have one answer: the target supplies the factor, and the receptors give it its meaning. Knockouts show each Trk pair keeping its own populations alive, and p75, with sortilin for the uncut precursor, lets the same factor kill.

One step remains. How competing neurons use p75 against each other is the subject of [competition for targets](https://stemhop.com/lessons/competition-for-targets/).

## Going deeper

### The dispute over high-affinity NGF binding

The 1991 papers did not end the argument. Hempstead and colleagues reported that high-affinity nerve growth factor (NGF) binding required TrkA and p75 expressed together. Barbacid’s group, besides the fibroblast result above, reported high-affinity NGF binding to TrkA in nonneural cells that lacked p75. So the groups disagreed about binding itself, and the debate ran for years. A 2022 review concludes that p75 and TrkA form a complex that gives high-affinity NGF binding. A 2007 structural study found no direct contact between the two receptors on cells and proposed that they cooperate through shared signaling partners instead.

### How BDNF, NT-3, NT-4, TrkB and TrkC were found

**The other factors came from brain.** In 1982 Yves-Alain Barde, Deborah Edgar and Hans Thoenen purified brain-derived neurotrophic factor (BDNF) from pig brain as a protein that kept cultured chick sensory neurons alive. Its cloning in 1989 showed that it resembles NGF, and searches for genes resembling both found neurotrophin-3 (NT-3) in 1990 and neurotrophin-4 (NT-4) in 1991. Klein, in Barbacid’s laboratory, had already cloned a relative of *trk*, *trkB*, in 1989. In 1991 three groups showed that TrkB is a receptor for BDNF and NT-3, and Fabienne Lamballe, Klein and Barbacid reported a third relative, TrkC, that prefers NT-3. NT-4 was shown to act through TrkB soon after.

### Names and counts

NT-4 is called NT-4/5 in older papers. Some fish have extra neurotrophins (NT-6 and NT-7), which is why the count of four is given for mammals.

### p75 and Fas: a family resemblance

p75’s death domain is a family resemblance to Fas, not the same machinery. Fas kills through the adaptor FADD and caspase-8. p75’s death signalling, including that of the p75-sortilin complex bound to a precursor, runs through other adaptors and activation of c-Jun N-terminal kinase (JNK).

### Gene dosage matters

Many mice with one copy of the BDNF gene eat more and become obese, and mice with one copy of the NGF gene show shrinkage of basal forebrain cholinergic neurons and poorer memory. Half the factor is not always enough. The obesity is an adult phenotype of the hypothalamus, outside the developmental rule of this lesson.

### Unsettled details

Mice lacking TrkB. Most mice lacking the kinase-containing form of TrkB die within a day of birth. A study of these mice also reported fewer motor neurons, but mice lacking both TrkB ligands, BDNF and NT-4, kept their facial motor neurons, so whether motor neurons need TrkB during development is unsettled.

Why groups disagreed about proNGF. Possible reasons include the dose, the cells’ balance of receptors, how the precursor was prepared and how much of it was cut during the experiment. A 2008 study by Boutilier and colleagues found that TrkA activation by proNGF needed the precursor to be processed.

---

From STEMhop (https://stemhop.com), free lessons by Christopher Deppmann. Lesson text © the authors. Practice questions: https://stemhop.com/lessons/trk-and-p75/
