Autism spectrum disorders (ASDs) are most often described in terms of their associated cognitive, social, and communication deficits, but ASDs are also characterized by extensive changes to sensory perception, including pain sensation. Major advances have been made in identifying genes associated with ASDs, but none of these genes have been directly demonstrated to affect pain—until now.
Researchers led by Ru-Rong Ji, Duke University School of Medicine, Durham, US, find changes in heat hyperalgesia in a newly developed ASD mouse model that has a deletion of SHANK3, a gene that when mutated plays a causative role in ASDs. Characterization of this model reveals that SHANK3 interacts with TRPV1 ion channels and regulates their function, providing a mechanistic understanding of how pain perception may be modified in ASD individuals that harbor a mutation in this gene.
“It’s a very nice paper in that it merges these two worlds of ASDs and pain, providing an additional perspective on how pain abnormalities and ASDs may be related to one another,” says Michael Caterina, Johns Hopkins University, Baltimore, US, who was not involved in the study.
The new research was published online December 1 in Neuron.
It’s been challenging to study modification of sensory perception, especially pain, in ASD individuals because of difficulties in communicating with and evaluating them. Perhaps indicative of these difficulties, enhanced pain sensitivity has been reported in some studies of ASDs, and decreased sensitivity in others (Allely, 2013). Phelan-McDermid syndrome (PMS), which features ASD symptoms, is less ambiguous when it comes to pain. Individuals with PMS have a complete deletion of SHANK3, with clear resultant effects on pain perception.
“Whereas many genes contribute to autism spectrum disorders, PMS seems to be primarily associated with a deletion of SHANK3 and 77% of these individuals also have decreased pain sensitivity, indicating that SHANK3 may play a critical role in pain,” said Ji.
Interestingly, another ASD mouse model that harbors mutations within the SHANK3 gene was found to have enhanced touch sensitivity (Orefice et al., 2016).
“A lot of PMS individuals have hypersensitivity to innocuous touch, but hyposensitivity to pain. We showed in our paper that SHANK3 mutant mice have increased sensitivity to light touch, but we didn’t look at pain behaviors,” said Lauren Orefice, Harvard Medical School, Boston, US, first author of that paper.
What about pain?
This led Ji to speak with Yong-Hui Jiang, also at Duke University School of Medicine. Jiang, co-corresponding author on the current study, had already generated a PMS-like mouse model where the entire SHANK3 gene is deleted. These knockout mice exhibited ASD-like behaviors, as reported earlier this year (Wang et al., 2016).
“During my initial discussions with Dr. Jiang, I thought we could test pain in these animals,” said Ji.
Co-first authors Qingjian Han, Yong Ho Kim, Xiaoming Wang and colleagues initially saw no effect of SHANK3 knockout on baseline pain sensitivity. Using cold and hot plate assays to test thermal sensitivity and Von Frey hairs to test mechanical sensitivity, they found that baseline pain thresholds of the knockout mice were no different from those of wild-type mice.
The researchers then turned to two chronic pain models, the complete Freund’s adjuvant model of inflammatory pain, and the chronic constriction injury model of neuropathic pain. This time, they found that the heat hyperalgesia that normally develops in each model was not as intense in the SHANK3 knockouts, but saw no differences in cold sensitivity, compared to wild-type mice.
SHANK3 acts via TRPV1
Given that only heat sensitivity was altered in the SHANK3 mice during chronic pain, the researchers suspected that perhaps heat-sensing TRPV1 ion channels were at play. Whereas injection of the TRPV1 agonist capsaicin into the hindpaw of wild-type mice caused robust pain behavior, the response to capsaicin in SHANK3 knockout mice was significantly weaker, suggesting that the channel might be involved in the effects of SHANK3 deletion on pain.
Molecular characterization revealed that SHANK3 was expressed in primary dorsal root ganglia (DRG) sensory neurons, including those responsible for pain transmission, and their spinal cord terminals, further suggesting a role for SHANK3 in pain processing. The investigators next discovered that SHANK3 co-localized with TRPV1 in DRG neurons, with co-immunoprecipitation experiments suggesting an interaction between the two.
Moreover, in SHANK3 knockout mice, subcellular TRPV1 levels were not altered, but the amount of TRPV1 that made its way to the cell surface was reduced. This suggested that SHANK3 plays an active role as a TRPV1 scaffold protein, regulating expression of the channel on the cell surface. Electrophysiological recordings in dissociated DRG neurons from SHANK3 knockout mice revealed diminished currents through TRPV1 channels, compared to those in wild-type mice, further suggesting an active role of SHANK3 in TRPV1 function.
TRPV1 channels in central terminals of pain-sensing neurons are known to modulate excitatory input into interneurons within the spinal cord and contribute to central sensitization in chronic pain. Recording electrical activity of these interneurons in response to capsaicin, using spinal cord slices, the investigators found that TRPV1-induced modulation of excitatory activity was reduced in slices from SHANK3 knockouts, compared to wild type slices, suggesting that SHANK3 may contribute to spinal cord synaptic plasticity and pain.
Next, the researchers generated another mouse model where SHANK3 was deleted only in pain-sensing neurons and some low-threshold A-fibers, instead of being deleted throughout the entire body. This allowed them to test SHANK3’s role specifically in pain modulation. Interestingly, unlike global SHANK3 knockout, sensory neuron-specific deletion not only decreased heat hyperalgesia during chronic pain but also decreased baseline heat sensitivity.
Ji explained that this finding may suggest opposing roles for central vs. peripheral SHANK3, where decreased baseline heat transduction in primary sensory neurons is somehow masked by changes within the spinal cord or brain. When SHANK3 is missing only from sensory neurons, spinal cord and brain SHANK3 still function normally, removing this potential masking effect, which then reveals decreases in baseline pain sensitivity.
“Another possibility is timing,” says Caterina. “In the global knockout, these sensory neurons never have SHANK3, whereas in the sensory-specific knockout, SHANK3 disappears when Nav1.8 [a marker for pain neurons] is first expressed. It’s possible that the differential timing of the knockout could be important for the magnitude of the phenotype,” he explained.
From mouse to human pain neurons
To determine the translational implications of their findings, the investigators obtained DRG neurons from healthy humans and used small interfering RNA (siRNA) to knock down SHANK3. Similar to the electrophysiological data from mouse pain neurons, blocking SHANK3 expression in human DRGs also decreased TRPV1 function. Strikingly, partial knockdown of SHANK3 expression also substantially blocked TRPV1 function, suggesting that haploinsufficiency of SHANK3 is enough to disrupt function of the channel.
Ji has hopes of exploring these findings clinically. “It would be exciting to specifically test heat sensitivity in PMS individuals,” he said. It’s also possible that different members of the SHANK family of proteins may regulate ion channels involved in other forms of pain or even other sensory systems. “This is a question we are going to explore in the future,” says Ji.
While the interaction between SHANK3 and TRPV1 may help to explain changes in pain perception in PMS individuals, there are also broader implications for pain patients. “In people with chronic pain, this is a potential interaction that we could inhibit. This could be an approach to develop an analgesic,” said Ji.
Nathan Fried is a postdoctoral fellow at the University of Pennsylvania, Philadelphia, US.