The following is Part 3 of a three-part series of selected presentations from the 35th Annual Scientific Meeting of the American Pain Society (APS) held May 11-14, 2016, in Austin, Texas, US. Also see Part 1 and Part 2.
Noxious stimuli activate sensory neurons in the peripheral nervous system, ultimately giving rise to the sensation of pain. But pain is not the product of a simple relay from peripheral nerves to the brain; sensory signals undergo extensive processing along the way, beginning at the first way station, the spinal cord. In a symposium held Thursday, May 12, three researchers examined how plasticity in ascending and descending pain pathways creates a “pain biography,” a lifelong experiential history that shapes the experience of and even future susceptibility to pain.
First, Rebecca Seal, University of Pittsburgh, US, presented data that help to elucidate pain micro-circuitry in the dorsal horn (DH) of the spinal cord. The work is a continuation of the group’s recently published report in Neuron (Peirs et al., 2015; see PRF related news), which identified a novel population of excitatory interneurons in a spinal circuit that gives rise to mechanical allodynia. The neurons, found in lamina III of the DH, transiently express the vesicular glutamate transporter 3 (VGLUT3) during development and receive input from low-threshold mechanically sensitive afferents. (In adulthood, these neurons rely on VGLUT2 to package glutamate.) Without VGLUT3 expression specifically in those neurons during development, mice displayed blunted responses to acute mechanical pain and did not develop mechanical allodynia as wild-type mice did in models of inflammatory and neuropathic pain. Using a DREADD (designer receptors activated exclusively by designer drugs) approach, the researchers then activated the neurons in adulthood—when VGLUT3 is no longer found in the cells—which was sufficient to induce mechanical allodynia.
Now, the team has performed a complementary experiment using an inhibitory DREADD to reversibly silence the neurons. “When we shut down the cells,” with the inhibitory designer ligand, Seal said in her talk, “it blocked mechanical allodynia” in two inflammatory pain models. Acute pain, in contrast, was unaffected. “Whereas the VGLUT3 transporter itself was important for acute and persistent mechanical pain,” Seal said, the new results indicate that, in adulthood, “these neurons transmit persistent pain, but they don’t seem to be important for acute mechanical pain.” That finding suggests a developmental role for VGLUT3. “VGLUT3 expressed by the cells is critical for setting up circuits for both acute, high-threshold pain and persistent mechanical allodynia, but the cells themselves appear to be required for only mechanical allodynia” Seal told PRF in an email.
“One of the questions we have is, Is mechanical allodynia in each type of injury encoded the same way in the dorsal horn?” Previous experiments revealed that with activation of the VGLUT3 neurons comes activation of excitatory neurons containing the gamma isoform of protein kinase C (PKCγ). These neurons had previously been shown to be activated in a neuropathic pain model, and inhibiting PKCγ itself blocks the mechanical allodynia resulting from nerve injury. The researchers also found that VGLUT3 neurons activated another population of neurons containing calretinin, which had not previously been associated with allodynia. When the researchers silenced the calretinin neurons by using the inhibitory DREADD system, mechanical allodynia was blocked in inflammatory models of pain but not in a neuropathic pain model. Thermal and acute mechanical pain were unaffected. In the calretinin neurons, Seal said, “We have identified a population of neurons in the circuitry that seems to be important for inflammatory pain but not for neuropathic pain.” Next, they investigated the role of the spinal PKCγ neurons also contacted by VGLUT3 cells. Intrathecal injection of a PKCγ inhibitor prevented mechanical allodynia in the spared nerve injury model of neuropathic pain but not in models of inflammatory pain.
“This work is starting to reveal the circuits for mechanical allodynia, and there appears to be a requirement for different neurons depending on the type of injury, suggesting that different micro-circuits might contribute to allodynia,” Seal said. She postulates that the calretinin neurons are normally not responsive to touch, but that inflammation somehow disinhibits the neurons. PKCγ neurons, she believes, are also normally inhibited and become disinhibited following nerve injury. “That disinhibition allows the sensation of pain to come through, instead of touch.” In future work, Seal aims to identify how the neurons become disinhibited and which primary afferent neurons are driving activity.
In a second talk, Theodore Price, University of Texas at Dallas, US, presented data suggesting that descending neurons from the brain are constantly remodeling spinal pain circuits. In 2015, Price’s group published work demonstrating that dopaminergic neurons projecting from the hypothalamus to the spinal cord were responsible for maintaining a primed state that rendered animals vulnerable to chronic pain (see PRF related coverage). Now, Price and his team have shown that surprisingly dynamic GABAergic synapses influence the development of chronic pain.
The researchers used the hyperalgesic priming model of the transition to chronic pain in which animals receive an injection of an inflammatory substance such as interleukin 6 (IL-6) to the hindpaw, which causes acute pain that resolves quickly but leaves them “primed.” A week later, injection of the inflammatory molecule prostaglandin E2 (PGE2) produces long-lasting mechanical hyperalgesia not seen in unprimed animals. Dopaminergic neurons signaling through D1/D5-type dopamine receptors were required for maintenance of the primed state.
In the current work, presented at APS and published in June in Pain (see Kim et al., 2016), the researchers first wanted to identify which neurons were contacted by the dopaminergic neurons, so they used a D1/D5 receptor agonist to activate the neurons. Subsequent measurement of c-fos, a marker of neuronal activity, indicated that the targets were mainly inhibitory neurons in lamina III and IV of the DH. “From that result, it became clear that maybe priming in the spinal cord had something to do with GABA signaling, so we went after that,” Price told PRF.
Price took inspiration from a previous report showing that, with persistent inflammation, GABAergic signaling in the spinal cord switches valence from inhibitory to excitatory signaling (Anseloni and Gold, 2008). In the current work, in naïve animals, injection of IL-6 produced a rapid reduction in withdrawal threshold lasting several hours that was attenuated by simultaneous intrathecal injection of the GABAA receptor agonist muscimol, indicating that GABA signaling inhibited the acute inflammatory pain as expected. In primed animals, however, intrathecal injection of muscimol had no effect on the hyperalgesia that followed intraplantar injection of PGE2. Conversely, the GABAA receptor antagonist gabazine did not worsen mechanical hypersensitivity in naïve mice injected with IL-6, but the drug reduced hypersensitivity in primed mice injected with PGE2. Intrathecal injection of gabazine alone into naïve mice produced mechanical hyperalgesia as previously reported. Together, the results suggest that GABA signaling inhibits acute pain as expected in naïve mice, but in primed mice, the signal somehow promotes pain signaling. In an unpublished experiment, Price and colleagues found in primed mice that gabazine also attenuated hyperalgesia resulting from intrathecal injection of a D1/D5 dopamine receptor agonist, indicating that dopaminergic and GABAergic signaling interact.
Price’s team next wanted to find the root of the priming-related GABA plasticity. They found that the synaptic adhesion molecule neuroligin 2 was persistently upregulated in the spinal cord of primed mice. Expression of the potassium-chloride pump KCC2, which has been shown to mediate a switch in GABAergic signaling under different conditions, did not change in primed mice (see PRF related news).
Price was not sure what to make of the findings. “Then we got lucky,” Price told PRF, “when a paper published around that time reported a synthetic peptide engineered to disrupt neuroligin 2” (van der Kooij et al., 2014). When they used the peptide to interfere with neuroligin 2 expression, priming was reversed. “Now we are excited about the idea of a novel descending modulatory circuit that involves GABA-regulated synaptogenesis via neuroligin 2,” Price said in his talk.
Together, the findings begin to sketch out a picture of priming as highly plastic modulation of pain-related synapses in the spinal cord by dopaminergic inputs, and a switch in GABA from an inhibitory, analgesic messenger to an excitatory, pro-nociceptive signal. “The results have really important implications for treatment, such as the idea that chronic pain is truly a disease of the CNS [central nervous system]. This may help us discover new mechanisms of plasticity that offer disease-modifying therapies to reverse plasticity and chronic pain states,” Price said.
Primed for pain
Simon Beggs, who recently moved from the University of Toronto, Canada, to University College London, UK, presented thought-provoking data from animals “primed” for pain in adulthood by a neonatal incision. The findings could have clinical implications for the millions of preterm babies born each year who are subjected to repeated needle pokes and other medical procedures.
Beggs and colleagues previously found that rats that received a single incision during a critical period in the first postnatal week experienced longer and more intense hyperalgesia following a minor injury in adulthood than did naïve rats (Beggs et al., 2012). “The injury in naïve adult rats sensitizes them for a week, but then they’re back to normal,” Beggs told PRF. Following the same minor injury in adulthood, the primed rats, in contrast, “became more sensitized, and the sensitization lasted longer.” The researchers determined that the early injury primed rats for a microglial neuroimmune response in the spinal cord activated by adult injury. When the researchers delivered the microglial inhibitor minocycline intrathecally before the adult injury, the enhanced hyperalgesia and microglial response were blocked.
Next, Beggs, in collaboration with Suellen Walker, University College London, UK, wanted to test whether minocycline given at the time of the neonatal incision would also be protective three days later. Ongoing experiments suggest that minocycline blocks the increased microglial reactivity seen in the spinal cord three days following the neonatal injury—but only in male rats. Pain sensitivity was likewise affected. “When we gave minocycline at the time of the neonatal incision [and then tested pain responses following minor injury in adulthood], in males, both mechanical and thermal reflex responses were reduced back to control levels, but the neonatal dose of minocycline had no effect in females,” Beggs told PRF. “Minocycline given at the time of the initial injury completely abolished priming in males, which might have implications for neonates undergoing surgery—perhaps we can prevent priming by pretreating with minocycline, at least in boys.” However, Beggs noted, although minocycline is approved by the U.S. Food and Drug Administration for some uses, prophylactic pain treatment in neonates is likely a long way off.
Could neonatal priming for pain also affect brain structures? To find out, Beggs collaborated with Jason Lerch, University of Toronto, Canada, who performed ex vivo structural magnetic resonance imaging (MRI) on brains of mice that received neonatal injury, adult injury, or both. “One important thing we found is that the brain changes very rapidly after injury, with the same time course as pain behaviors,” Beggs told PRF. Brain structures did undergo changes following injury at either age, but how they changed was determined by when the injury occurred. “These are not small, tiny changes,” Beggs said. “We saw up to a 28 percent change in volume in some cases. What they mean is another matter,” he added.
“The second thing we realized is that the changes not only appear quickly, but they’re also long-lasting” into adulthood, Beggs told PRF, as indicated by the differences seen between brains of naïve adult mice and those with a neonatal injury. Finally, Beggs said, there seemed to be an interaction between the neonatal and adult injuries. For example, adult injury led to a decrease in volume in the cerebellar vermis, but in primed mice, it caused an expansion of the same region. “Priming seems to change the brain’s response to injury.”
Together, the three lines of research begin to reveal an extremely dynamic pain-processing system that can be reshaped by injuries throughout life to influence future pain experiences.
Stephani Sutherland, PhD, is a neuroscientist, yogi, and freelance writer in Southern California.