A New Explanation for Itchy Eyes

by Matthew Soleiman on 9/24/2018

When it comes to the eye, itch has its quirks. It’s usually felt in the conjunctiva, the membrane that lines the eyelids and covers the sclera (the white of the eye). Yet it’s rarely felt in the cornea, the outermost layer of the eye that focuses incoming light. Now, Qin Liu, Washington University, Saint Louis, US, and her colleagues have found a new explanation for this division of sensation.

The researchers show that, in mice, the conjunctiva is innervated by itch-evoking neurons that express the Mas-related G protein-coupled receptor member A3 (MrgprA3). In the cornea, however, these cells are absent, innervated instead by nociceptive fibers. Consistently, silencing the MrgprA3 population in the conjunctiva reduced eye scratching in allergen-exposed mice, including in animals engineered to carry the human variant of MrgprA3.

Along with identifying circuitry for itch and pain in the eye, the team also found surprising differences between itch in the eye and itch in the skin. For example, unlike the latter, which depends on at least three populations of neurons, the former seems to be driven exclusively by MrgprA3 cells.

“This is a nicely put together set of experiments that tells a very logical story of specialization of itch in the eye,” says Mark Hoon, National Institute of Dental and Craniofacial Research, Bethesda, US, who was not involved in the work. “In the end, it gives us a little glimpse of what might be possible for treatment of a condition that’s really unpleasant.”

The study was published in the August 2018 issue of Nature Medicine.

Itchy eyes

Liu’s itch to study the eye, so to speak, came from her own unpleasant experience with seasonal allergies. “My eyes feel itchy every spring,” she said. “I use eye drops, but the antihistamine doesn’t work so well. And, it makes my eyes dry.”

To find a therapeutic target better than histamine, she began asking, “What’s going on in the eye? What are the mechanisms of itch there?”

Within the skin, pruriceptive neurons can be divided into those that express MrgprA3, the related receptor MrgprD, or the peptide somatostatin. Yet when co-first authors Cheng-Chiu Huang, Weishan Yang, and their colleagues looked for these cell populations in the eye, they found only two of the three. In the conjunctiva of mice that were genetically modified such that each cell population could be visually labeled, the researchers saw many fibers with MrgprA3, far fewer with MrgprD, but none with somatostatin.

“We didn’t expect that,” said Liu. “Since the somatostatin population doesn’t project to the conjunctiva, it’s probably not involved in ocular itch.”

What’s more, none of these three populations were present in the cornea, a structure that’s easily damaged with excessive scratching. Instead, there were fibers containing vesicular glutamate transporter 3 (VGLUT3). And, relative to the conjunctiva, there were more cells positive for the transient receptor potential cation channel subfamily M, member 8 (TRPM8).

This anatomy pointed to a clear hypothesis—that MrgprA3 neurons contributed the most to ocular itch. To test the idea, the researchers first ablated these cells in mice, and then delivered a set of itch-causing chemicals, including histamine, neuropeptide FF, serotonin, and the antimalarial drug chloroquine, to the animals’ eyes. With each chemical, mice missing MrgprA3 cells scratched themselves less than their unaltered counterparts. In normal mice, however, stimulating MrgprD neurons had no clear effect. In other words, the MrgprA3 population seemed to be the only one that could trigger ocular itch, a key difference from itch in the skin.

Calcium imaging would reveal that MrgprA3 fibers in the conjunctiva were excited by all four pruritic chemicals. But in the skin, such fibers were unresponsive to serotonin.

These mechanisms of ocular itch stand out to Hoon. Before this study, “people had never looked at itch in specialized structures,” he said. “Because we use mice, and because of the assays we use for behavioral testing, most work on somatosensation focuses on hairy skin. This says that not all external tissues are going to be the same.”

As for neurons in the cornea, activating the TRPM8 population caused mice not to scratch but to blink and close their eyes, two pain-related responses. Together with the organization of sensory fibers, these behavioral findings suggested that, within the eye, itch and pain might rely on distinct circuitry. In additional experiments, the team showed that the central circuit for ocular itch, but not that for ocular pain, involves neuromedin B signaling.

Silencing neurons to ease itch

Now knowing the molecular identity of the eye’s pruriceptive neurons, Liu’s group next tried to ease ocular itch in mice exposed to the allergen ovalbumin. The eyes of the animals were pretreated with a low dose of chloroquine, an agonist for MrgprA3, as well as QX-314, a positively charged sodium channel blocker. As MrgprA3 activation by chloroquine leads to the opening of the TRPA1 ion channel, this would allow QX-314 to enter the MrgprA3 neurons via the open channel, ultimately blocking their activity.

With this approach, the researchers could quell allergy-related scratching for at least 24 hours. In contrast, relief from an antihistamine lasted no more than an hour. QX-314 achieved the same outcome in mice engineered to lack MrgprA3 and instead express the receptor’s human variant, MrgprX1. And, in human tissue, MrgprX1 neurons were found to innervate the conjunctiva, but not the cornea.

The itching from ovalbumin didn’t only involve MrgprA3 neurons, however. As Liu’s team went on to discover, MrgprA3 fibers also recruited mast cells, a type of immune cell, likely through calcitonin gene-related peptide signaling. These mast cells release a flurry of their own itch-evoking chemicals.

To Liu, it makes sense that QX-314 would do more than an antihistamine for itchy eyes. “The ineffectiveness of antihistamines is due to the fact that the histamine-dependent pathway is not the only pathway in allergic itch,” she said. MrgprA3 fibers presumably carry receptors for most if not all of the signaling molecules involved in these pathways. “So if you silence the whole population of MrgprA3 neurons, all of the receptors expressed by them will not work anymore.”

The idea to use QX-314 to silence active neurons is not new. A 2007 study showed that the compound, when given at the same time as the TRPV1 agonist capsaicin, could quiet TRPV1-expressing nociceptors in mice (Binshtok et al., 2007). Years later, some of the same researchers used QX-314 to block itch (see related PRF story).

“QX-314 could be remarkably effective for treating allergic itch, because it would only act where there is a sensitization by an allergen. Neurons that weren’t active wouldn’t be affected,” Hoon said. “That’s the nice thing about QX-314. In normal circumstances, it isn’t a blocker.”

Even so, “there’s always a worry that things aren’t going to work exactly the same way as in a model system. As soon as you start using a compound in people or even other animals, you start finding side effects that are perhaps unanticipated,” he added. “QX-314 sounds great in principle, but it needs to be tested in humans.” The compound has not yet been evaluated in a clinical trial.

In the meantime, Liu intends to learn more about how itching originates in the eye. Although it’s clear that mast cells release molecules such as histamine and serotonin, which then rouse MrgprA3 neurons, the ligand for MrgprA3 is still unknown. “I hypothesize it is neuropeptide FF or some other small-molecule peptide.”

Matthew Soleiman is a graduate student in the Department of History and Science Studies Program at the University of California, San Diego, US. Follow him on Twitter @MatthewSoleiman.

Image credit: stockgiu/123RF Stock Photo.