Cortical astrocytes rewire somatosensory cortical circuits for peripheral neuropathic pain
Sun Kwang Kim, … , Schuichi Koizumi, Junichi Nabekura
Sun Kwang Kim, … , Schuichi Koizumi, Junichi Nabekura
Published May 2, 2016; First published April 11, 2016
Citation Information: J Clin Invest. 2016;126(5):1983-1997. https://doi.org/10.1172/JCI82859.
View: Text | PDF
Category: Research Article

Cortical astrocytes rewire somatosensory cortical circuits for peripheral neuropathic pain

Abstract

Long-term treatments to ameliorate peripheral neuropathic pain that includes mechanical allodynia are limited. While glial activation and altered nociceptive transmission within the spinal cord are associated with the pathogenesis of mechanical allodynia, changes in cortical circuits also accompany peripheral nerve injury and may represent additional therapeutic targets. Dendritic spine plasticity in the S1 cortex appears within days following nerve injury; however, the underlying cellular mechanisms of this plasticity and whether it has a causal relationship to allodynia remain unsolved. Furthermore, it is not known whether glial activation occurs within the S1 cortex following injury or whether it contributes to this S1 synaptic plasticity. Using in vivo 2-photon imaging with genetic and pharmacological manipulations of murine models, we have shown that sciatic nerve ligation induces a re-emergence of immature metabotropic glutamate receptor 5 (mGluR5) signaling in S1 astroglia, which elicits spontaneous somatic Ca2+ transients, synaptogenic thrombospondin 1 (TSP-1) release, and synapse formation. This S1 astrocyte reactivation was evident only during the first week after injury and correlated with the temporal changes in S1 extracellular glutamate levels and dendritic spine turnover. Blocking the astrocytic mGluR5-signaling pathway suppressed mechanical allodynia, while activating this pathway in the absence of any peripheral injury induced long-lasting (>1 month) allodynia. We conclude that reawakened astrocytes are a key trigger for S1 circuit rewiring and that this contributes to neuropathic mechanical allodynia.

Authors

Sun Kwang Kim, Hideaki Hayashi, Tatsuya Ishikawa, Keisuke Shibata, Eiji Shigetomi, Youichi Shinozaki, Hiroyuki Inada, Seung Eon Roh, Sang Jeong Kim, Gihyun Lee, Hyunsu Bae, Andrew J. Moorhouse, Katsuhiko Mikoshiba, Yugo Fukazawa, Schuichi Koizumi, Junichi Nabekura

×

Figure 1

Increased numbers of astrocytic Ca2+ transients in S1 cortex following PSL injury and their contribution to mechanical allodynia.

Options: View larger image (or click on image) Download as PowerPoint
Increased numbers of astrocytic Ca2+ transients in S1 cortex following P...
(A) In vivo 2-photon Ca2+ imaging of astrocytes (yellow staining) in S1 cortex layer I, loaded with OGB-1–AM (Ca2+ indicator, green) and SR101 (astrocyte marker, red). Scale bar: 30 μm. (B) Representative images of S1 astrocytes (dashed circles) and corresponding Ca2+ transients in control (blue rectangle shown in A), PSL–early (3–6 days after PSL injury) and PSL–late (12–15 days after PSL injury) groups. Arrows indicate individual transients (ΔF/F0 >15%). (C) Frequency of astrocytic Ca2+ transients (red), spine turnover rate (n = 550 spines/15 dendrites/6 mice, blue), and mechanical threshold (n = 6 mice, black). Spine turnover and mechanical threshold data are from our previous study (10) conducted under the same conditions, with additional data from 1 mouse included. Ca2+ transients were quantified in active astrocytes. Con, sham-operated mice (n = 41 cells/3 mice); Early (n = 130 cells/3 mice); Late (n = 71 cells/4 mice). ***P < 0.001 versus control, by Kruskal-Wallis test. (D) Left: Proportion of active astrocytes that showed 1 or more Ca2+ events during a 10-minute recording period in sham control, PSL–early, and PSL–late mice. n = 10–13 imaged planes/group. **P < 0.01, by Kruskal-Wallis test. Right: Corresponding mean amplitudes of Ca2+ transients. NS, P > 0.05, by Kruskal-Wallis test. (E) Top: Representative confocal images of synapses (arrowheads) indicated by colocalization of presynaptic (VGlut1, green) and postsynaptic (PSD-95, red) markers in the sham and PSL injury groups. Scale bar: 5 μm. Bottom: Quantitative analysis demonstrates that PSL injury significantly increased the number of colocalized synaptic puncta. n = 12 image stacks/4 mice per group. ***P < 0.001 versus sham control, by unpaired t test. (F) Mean mechanical thresholds 0–27 days following PSL injury in control (n = 10), IP3R2-KO (7), BAPTA-AM (5), and fluoroacetate-treated (4) mice. *P < 0.05, **P < 0.01, and ***P < 0.001 versus control, by 1-way ANOVA. Error bars represent the mean ± SEM.