Optogenetics—using light to control cells in genetically engineered animals expressing light-sensitive proteins—has been a part of the scientific research landscape for over a decade. But only in the last few years have investigators begun to apply this technique to the peripheral nervous system. A slew of recent studies now use optogenetics to activate peripheral nociceptive neurons in freely moving mice, providing researchers another opportunity to revisit some of the pain field’s most fundamental concepts.
In the first study, published July 5 in Cell Reports, Clifford Woolf, Boston Children’s Hospital and Harvard Medical School, US, and colleagues use optogenetics to examine the protective withdrawal reflex in response to noxious stimuli. By expressing the light-sensitive protein Channelrhodopsin-2 (ChR2) in small-diameter C-fiber and A-delta-fiber nociceptors innervating the skin of mice, they show that this reflex entails a coordinated and global behavioral response across the entire body that differs depending upon the state of the animal, including its posture, activity, and whether it is asleep or awake.
This research builds upon work published in Neuron at the end of last year by Stefan Lechner, Heidelberg University, Germany, and colleagues, who express ChR2 in myelinated A-fiber nociceptors and in low-threshold mechanoreceptors, to confirm key predictions of the gate control theory of pain. They demonstrate that activation of low-threshold mechanoreceptors alleviates acute A-fiber-evoked pain, and was necessary to produce a coordinated nociceptive reflex response to harmful stimuli. They also identify a subpopulation of A-fiber nociceptors that specifically mediate the response to sharp objects, in this case pinprick pain.
“Both of these papers advance the field,” says Philippe Séguéla, McGill University, Montreal, Canada, who along with colleagues became the first to apply optogenetics to pain circuits in 2013 (Daou et al., 2013; also see PRF related news story). “I think optogenetics is the best tool we have right now to investigate how peripheral somatosensory circuits are activated, and this research shows how it is possible to characterize those circuits with more and more precision—genetically, physiologically, and at the behavioral level,” he said.
Séguéla’s group makes a further contribution to the field as reported online July 12 in Pain. That investigation focuses on two different nociceptive neuron populations—one peptidergic, the other non-peptidergic—and uses a conditioned place aversion assay to reveal differing behavioral responses upon light stimulation of each set of C-fibers.
Back to the future
The impetus for the new Cell Reports study, according to first author Liam Browne, was to bring a new technique to build upon an age-old idea—the nociceptive withdrawal reflex—first described a century ago by the British neurophysiologist and Nobel laureate Sir Charles Sherrington.
“We wanted to apply new tools developed over the past few years to fundamental concepts first described in Sherrington's work,” Browne said. “We thought optogenetics would be a good candidate to do that because of the genetic specificity and high spatial and temporal precision it provides, allowing us to selectively target nociceptive cell populations of interest to us,” according to Browne, who carried out the research as a postdoc in Woolf’s lab and is now at University College London, UK.
The investigators began by expressing ChR2 in TRPV1+ and Tac1+ neurons, two different dorsal root ganglion (DRG) nociceptive populations of small-diameter C-fibers and A-delta-fibers, respectively. Using high-speed cameras to make their behavioral observations, the group found that just a few milliseconds of blue light applied to a tiny area of the animals’ glabrous (hairless) hind paw skin consistently evoked hind limb withdrawal. Half of the time, it took about 30 milliseconds before the animals first withdrew their paw, and about 140 milliseconds the other half of the time. The authors say this corresponds to the different conduction properties of A-delta-fibers and C-fibers, respectively, that they saw in electrophysiological experiments.
The researchers then confirmed expression of ChR2 in the two nociceptive populations, whose peripheral terminals ended in the skin epidermis and whose central axons reached lamina I-II of the spinal cord dorsal horn. Electrophysiological recordings in vitro from these DRG neurons indicated that one three-millisecond pulse of light produced just a single action potential, a finding confirmed in vivo in anesthetized mice.
“We were able to generate a single action potential in genetically targeted nociceptors from one short pulse of light, and this was sufficient to drive robust hind paw withdrawal,” Browne said.
Context matters
But the withdrawal of the hind paw was not an invariant response, but rather one that differed according to the context—that is, it depended upon the animals’ activity, alertness, and posture. For instance, the mice were less likely to withdraw the hind paw if they were grooming at the time the light was administered, or if they were sleeping. Interestingly, sleeping mice woke up in a fraction of a second after light stimulation, much more quickly than when they were woken up by a harmless sound.
Further, if the mice’s forepaws were raised from the floor (“forepaw-up”) at the time of light application, hind paw withdrawal was slower, in comparison to mice with all paws on the floor (“forepaw-down”). This is likely a reflection of the animals’ need to maintain their balance, the authors reasoned. The difference seen between the forepaw-up and forepaw-down conditions appeared to result from a delay in the putative C-fiber response to light, the authors say.
Intriguingly, optogenetic stimulation did not only result in movement of the limb, but also in simultaneous movements across the animals’ entire body. The whiskers, for example, also moved, and did so before hind paw withdrawal most of the time. The presence of this motor response, which is controlled by the brainstem, challenges the view that the protective pain reflex only involves the spinal cord, the researchers say.
Similar globally coordinated behavior was observed during the tail flick response, another spinal withdrawal reflex. Here, mice withdrew their tail upon light stimulation, and at the same time the whiskers, head, body, and limbs of the animals moved. The animals took longer to withdraw the tail than they did to withdraw the hind paw, suggesting that tail withdrawal was mediated primarily by input from the more slowly conducting C-fibers.
Altogether, the behavioral results indicate that the protective pain response does not always look the same. “We show that the nervous system is organized to prioritize different behaviors, reflecting activity of the whole system, that alter posture and rapidly orient and alert the animal to and minimize the risk of danger,” Browne told PRF. “It makes sense that these global behaviors occur to help the animal get away and minimize the contact it has with something that's potentially harmful. It's not just the paw that might be damaged; it may be the entire animal that needs to remove itself.”
Revisiting the gate control theory of pain
It was another core concept in pain science—the gate control theory of pain—that provided the backdrop to the paper from Lechner and colleagues published online last December 15 in Neuron.
“Gate control theory predicted that when you have increased pain sensitivity after an injury or inflammation, tactile input can activate pain-processing neurons in the spinal cord, thereby converting tactile stimuli into pain,” Lechner said. “Another prediction was that, at the same time, in healthy organisms, there must be circuits in the spinal cord where touch receptors inhibit painful input to these circuits.”
To revisit gate control theory using optogenetics, first author Alice Arcourt and colleagues inserted ChR2 into myelinated A-fiber nociceptors expressing the marker neuropeptide Y receptor-2 (NPY2R). Skin-nerve recordings indicated that blue light stimulation of the hind paw in these NPY2R-ChR2 mice activated a subset of A-fiber nociceptors. Other experiments showed that these fibers were activated only in response to noxious mechanical stimuli and not to heat or capsaicin, and had fast axonal conduction velocities.
Consistent with these findings, immunostaining experiments in the DRG revealed that the NPY2R-ChR2 cells contained TRKA, a marker of peptidergic A-fiber nociceptors. And, the peripheral projections of these cells ended in free nerve endings in hind paw skin, while the central projections reached as far as lamina II of the lumbar spinal cord. This was in line with the evidence that ChR2 had been expressed in a subset of peptidergic, myelinated A-fiber nociceptors.
The team next turned to the behavior of the NPY2R-ChR2 animals, reporting two main findings. First, using a diphtheria toxin approach to ablate NPY2R-expressing DRG neurons, they discovered that the animals took much longer to withdraw the paw in response to pinprick touch, pointing to a fundamental role for this population of fibers in detecting this particular type of mechanical stimulation. Second, light applied to the hind paw produced a number of strong nociceptive behaviors in NPY2R-ChR2 animals, including guarding, jumping, and vocalization, but not in littermate controls.
With these initial findings in hand, the group then expressed ChR2 in a different population of cells―low-threshold mechanoreceptors (LTMRs) marked by MafA, a transcription factor expressed only in LTMRs. As with the NPY2R-ChR2 cells, immunostaining and electrophysiology here, too, confirmed the specificity of these optogenetic manipulations. The peripheral nerve endings of these fibers surrounded hair follicles in the skin, and also innervated Meissner corpuscles of the hind paw footpad, structures typically innervated by LTMRs. Meanwhile, the fibers projected centrally mainly to lamina III-IV of the spinal cord, also the area to which LTMRs are known to project.
The stage was now set to return to gate control theory and ask, Could light activation of touch receptors inhibit pain, in particular acute pain? To test this idea, the researchers crossed the NPY2R-ChR2 mice with the MafA-ChR2 mice, creating double transgenic animals in which light could activate both cell populations at the same time. And indeed, it was the case: Activating the LTMRs and the A-fiber nociceptors simultaneously led to a reduction of pain. That is, in comparison to NPY2R-ChR2 mice, the double transgenic mice vocalized less and were less likely to show guarding and jumping behaviors upon light exposure.
Better reflexes, too
Interestingly, the researchers also found that activation of the LTMRs improved the coordination of the nociceptive withdrawal reflex. Namely, while NPY2R-ChR2 mice withdrew both paws in response to light delivered to one hind paw, with some animals even jumping with all four paws, the double transgenic mice exhibited smoother behavior, withdrawing just one paw in response to light. This withdrawal response was similar to that seen with pinprick mechanical stimulation.
“As soon as we mimic natural stimulation by activating nociceptive neurons as well as sensory fibers that normally detect touch—which are inevitably activated by any kind of painful mechanical stimuli—then all of a sudden the reflex coordination is much better,” Lechner told PRF. “The animals can really localize where the stimulus comes from. They keep standing on the other three legs, and just step aside with the stimulated leg. It’s a very well-coordinated paw withdrawal,” he added.
Let there be light … on peptidergic versus non-peptidergic nociceptive neurons
In their recent paper published online July 12 in Pain (also see accompanying commentary by Lechner), Séguéla’s group sought to determine whether optogenetic stimulation would evoke different pain behaviors depending upon the specific nociceptor subpopulation they activated. So co-first authors Hélène Beaudry and Ihab Daou and colleagues expressed ChR2 in primary sensory neurons containing TRPV1, a marker of peptidergic nociceptors. They also placed ChR2 into cells expressing Mas-related G protein-coupled receptor subtype D (MrgD), a marker of non-peptidergic nociceptors.
In each case, they expressed ChR2 only in mature nociceptors, using an adenovirus vector to deliver ChR2 to the TRPV1+ cells, and an inducible Cre recombinase approach to deliver the light-sensitive protein to the MrgD+ neurons. They did so in order to prevent the possibility that developmental mechanisms that regulate TRPV1 and MrgD gene expression could result in non-specific expression of ChR2. Because of this approach, “we are more precise in how we interrogate these classes of somatosensory neurons,” Séguéla told PRF.
Immunofluorescence studies indicated the strategy succeeded. In the case of the TRPV1+ neurons, the group confirmed expression of ChR2 strictly in peptidergic fibers. The central projections of these fibers reached lamina I and outer lamina II in the dorsal horn spinal cord, ending in an area dorsal to non-peptidergic fibers in inner lamina II, while the peripheral projections ended as free nerve endings in glabrous skin; these findings are consistent with the known projection patterns of peptidergic fibers. In addition, the TRPV1-ChR2 DRG neurons expressed the peptidergic marker CGRP. In the case of the MrgD-ChR2 neurons, ChR2 was expressed selectively in non-peptidergic neurons, which sent central projections to inner lamina II of the spinal cord, and reached the periphery as free nerve endings in the superficial area of the skin.
Next, electrophysiological experiments using dissociated DRG neurons from each of the two cell populations showed that blue light caused inward currents and evoked action potentials. In the TRPV1-ChR2 mice, DRG neurons also showed inward currents in response to the TRPV1 agonist capsaicin, while DRG neurons from the MrgD-ChR2 mice also showed an inward current in response to a selective agonist of P2X3, the latter a purinergic receptor expressed by most non-peptidergic nociceptors. In short, these findings were consistent with the immunofluorescence studies, and indicated specific expression of functional ChR2 in the two cell populations of interest.
Finally, the researchers turned to behavioral observations, identifying distinct responses in freely moving animals depending upon which cell populations were activated. In the TRPV1-ChR2 mice, light stimulation of hind paw glabrous skin resulted mainly in withdrawal and licking of the paw, with less paw lifting. In contrast, while MrgD-ChR2 animals showed paw withdrawal and lifting, licking was less frequent. And, interestingly, in a conditioned place avoidance test, light caused aversion only in the TRPV1-ChR2 mice, who spent more time in a compartment (non-painful) that had been paired with yellow light, compared to the time they spent in a compartment (painful) that had been paired with blue light.
In short, the findings show distinct behavioral responses to light in peptidergic versus non-peptidergic nociceptors, harking back to another core concept in pain, namely that of labeled lines present early on in pain pathways.
Strength in numbers (of approaches)
As these three recent studies show, optogenetics is finally bearing fruit for the study of the peripheral nervous system contribution to pain signaling. But despite the enthusiasm in this field, there is recognition that optogenetics is just one of several techniques to learn more about pain—and carries its own limitations.
For one, light activation of neurons is quite an artificial technique. “It’s important to compare optogenetic stimuli with the comparable natural stimuli,” Lechner stressed. “In vivo, in a natural environment, a single population of sensory neurons alone would never be activated—there is always a symphony of sensory information that arrives simultaneously in the spinal cord. So activating just a single population might cause a very artificial sensory percept.”
Séguéla agrees. “How realistic is optogenetic stimulation? It’s not clear that this is the best way to stimulate a neuron. For example, optical stimulation of TRPV1+ nociceptors does not trigger neurogenic inflammation. So we need to combine optogenetic, naturalistic, and pharmacological approaches,” he said.
Considering the genetic and functional heterogeneity of DRG neurons, another future avenue of research is to target ChR2 to all the different types of DRG neurons involved in pain signaling.
Also up next, Lechner says, is to better understand how the behaviors induced by light evoke a specific kind of sensory experience, which could be tested using training paradigms. “One has to combine optogenetics with more sophisticated behavioral experiments that tell us more about what a mouse really feels when we activate specific receptors. Understanding the complex crosstalk, signal integration, and processing of all of these different types of sensory neurons and how they then, together, form a certain sensory percept is going to be the future.”
He added that future research will also benefit from the possibility of inserting different channelrhodopsins into different populations of sensory neurons, each activated by a different color of light. “Then you could independently control activity in two different types of sensory neurons and see how these lines actually work together, or how important it is to have simultaneous activity in two different populations.”
Optogenetics may be moving toward the periphery, but additional research in the central nervous system, including study of spinal cord circuits, and in chronic pain models as well, is also an important future direction for pain researchers. “Would you see different behaviors when targeting other subpopulations of cells? How is this sensory information processed by different circuits? And what do the responses look like in chronic pain conditions?” Browne said.
Considering all of these areas left to explore—along with the development of implantable, wireless optoelectronic devices that will make the technique less cumbersome to use in animals (see PRF related news story here; RELIEF related interview here; and Samineni et al., 2017)—optogenetics seems set for a bright and colorful future in the pain field.
Neil Andrews is a science journalist and executive editor of PRF.
Image credit: hoperan/123RF Stock Photo.
