Skip Navigation

The itch without the pain

August 6, 2009 By Lizzie ETB Buchen This article courtesy of Nature News.

A special set of neurons may be dedicated to sensing itchiness.

Researchers have discovered nerve cells that are crucial for sensing itchiness — but are not needed to detect pain.

It might seem obvious that an itch and a pain are different sensations — that itch you want to scratch feels quite different from being stabbed — but neuroscientists have long puzzled over how the nervous system teases them apart.

When neuroscientist Zhou-Feng Chen and his colleagues at Washington University in St Louis, Missouri, destroyed a particular set of neurons in the spinal cords of mice, they found that the animals no longer scratched themselves when injected with itch-inducing chemicals. But they still responded normally to painful stimuli1.

"Most people accept that there are specific, highly specialized neurons for sensations like taste," says Chen. "But for pain and itch this is much more controversial. This is the first time itch perception has been shown to be independent from pain."

The idea that separate circuits of nerve cells convey itchiness and pain is attractive in its simplicity but has been difficult to prove. Peripheral sensory neurons, which carry information from the skin to the spinal cord, all seem to respond to multiple sensations, including pain and itchiness. Neurons that respond only to pain had previously been discovered in the spinal cord — the next stage in the nerve relay to the brain — but no equivalents specific for itchiness. Two high-profile papers2,3 that claimed to have discovered itch-specific neurons turned out to be false alarms4.

Ticklish problem

Two years ago, Chen's group discovered that a cell-surface protein called the gastrin-releasing peptide receptor (GRPR) is important for sensing itchiness but not pain in mice5. Although this suggested the existence of itch-specific neurons in the nervous system, it wasn't conclusive — neurons producing GRPR might also be transmitting pain.

This is the first behavioural evidence that there are itch-specific neurons. People have been looking for these for many years.
Zhou-Feng Chen
Washington University

Now Chen and his colleagues have taken a more rigorous approach. They destroyed GRPR-bearing neurons by means of a cell toxin attached to bombesin, a peptide found in frog skin that closely resembles mouse gastrin-releasing peptide, the molecule that normally binds to GRPR. When they injected the complex into the spinal cords of adult mice, neurons bearing GRPR bound the bombesin-attached toxin, took it into the cell and were killed.

Mice that had lost the GRPR-producing neurons reacted to painful stimuli just like normal mice, licking themselves and flinching or jumping in response to heat, highly irritant chemicals and mechanical pressure. But when the researchers injected the animals with chemicals that normally cause scratching, such as histamine, they barely responded. Chen and his team found that the greater the number of GRPR-expressing neurons destroyed, the more subdued was the scratching response.

"This is the first behavioural evidence that there are itch-specific neurons," says Chen. "People have been looking for these for many years."

Scratching the surface?

But some researchers say that behavioural evidence isn't enough. Neuroscientist Earl Carstens at the University of California in Davis thinks the findings "make an important contribution" by identifying neurons that are crucial for itch and not for pain, but says that this doesn't prove specificity.

"[The neurons] might signal both itch and pain based on some property of their firing, rather than signalling only itch," he explains, referring to another leading hypothesis in which pain and itchiness are the results of different firing patterns of neurons that respond to both.

Chen acknowledges that his results do not disprove the firing pattern model, but also notes that "so far there is no evidence to support that theory".

"Even if some GRPR neurons may be responding to painful stimuli — and note that this is a big if — they are dispensable for pain sensation," he says.

Chen's findings may not end the debate, but they highlight a new target for therapy says clinician Gil Yosipovitch of Wake Forest University in North Carolina. "This is very exciting to all of us who are waiting for general-purpose itch drugs," he says. "It opens the field for specific treatments for itch that don't affect pain."


  1. Sun, Y.-G. et al. Science, doi:10.1126/science.1174868 (2009).
  2. Schmelz, M. et al. J. Neurosci. 17, 8003-8008 (1997).
  3. Andrew, D. & Craig, A.D. Nature Neurosci. 4, 72-77 (2001).
  4. Schmelz, M. et al. J Neurophysiol. 89, 2441-2448 (2003).
  5. Sun, Y.-G. & Chen, Z.-F. Nature 448, 700-704 (2007).


Need Assistance?

If you need help or have a question please use the links below to help resolve your problem.