Shank2 identifies a subset of glycinergic neurons involved in altered nociception in an autism model.

Autism Spectrum Disorders (ASD) patients experience disturbed nociception in the form of either hyposensitivity to pain or allodynia. A substantial amount of processing of somatosensory and nociceptive stimulus takes place in the dorsal spinal cord. However, many of these circuits are not very well understood in the context of nociceptive processing in ASD.We have used a Shank2-/- mouse model, which displays a set of phenotypes reminiscent of ASD, and performed behavioural and microscopic analysis to investigate the role of dorsal horn circuitry in nociceptive processing of ASD.We determined that Shank2-/- mice display increased sensitivity to formalin pain and thermal preference, but a sensory specific mechanical allodynia. We demonstrate that high levels of Shank2 expression identifies a subpopulation of neurons in murine and human dorsal spinal cord, composed mainly by glycinergic interneurons and that loss of Shank2 causes the decrease in NMDAR in excitatory synapses on these inhibitory interneurons. In fact, in the subacute phase of the formalin test, glycinergic interneurons are strongly activated in wild type (WT) mice but not in Shank2-/- mice. Consequently, nociception projection neurons in laminae I are activated in larger numbers in Shank2-/- mice.Our investigation is limited to male mice, in agreement with the higher representation of ASD in males; therefore, caution should be applied to extrapolate the findings to females. Furthermore, ASD is characterized by extensive genetic diversity and therefore the findings related to Shank2 mutant mice may not necessarily apply to patients with different gene mutations. Since nociceptive phenotypes in ASD range between hyper- and hypo-sensitivity, diverse mutations may affect the circuit in opposite ways.Our findings prove that Shank2 expression identifies a new subset of inhibitory interneurons involved in reducing the transmission of nociceptive stimuli and whose unchecked activation is associated with pain hypersensitivity. We provide evidence that ...