Formation of perineuronal nets within a thalamocortical circuit shapes mechanical and thermal pain thresholds in mice with neuropathic pain

Abstract

We moved from the hypothesis that perineuronal nets (PNNs), which are condensed structures of the extracellular matrix surrounding GABAergic interneurons in the forebrain, contribute to mechanisms of maladaptive neuronal plasticity underlying chronic pain. Here, we found that the density of PNNs labelled with the lectin Wisteria Floribunda Agglutinin (WFA) increased in the contralateral somatosensory cortex (SSC), medial prefrontal cortex (mPFC), reticular thalamic nucleus (RTN), and insular cortex of mice developing neuropathic pain in response to unilateral chronic constriction injury of the sciatic nerve. These regions are involved in neuronal circuits underlying perception, sufferance, embodiment, and top-down control of pain. At least in the SSC and mPFC, the increased density of WFA + PNNs was associated with an up-regulation of the proteoglycans, brevican and neurocan, as shown by immunoblot analysis. Enzymatic degradation of PNNs caused by local infusion of chondroitinase ABC in the contralateral SSC or RTN enhanced both mechanical and thermal pain thresholds in chronic constriction injury mice. In contrast, siRNA-induced knock-down of the PNN-degrading enzyme, type-9 matrix metalloproteinase (MMP-9), in the SSC or RTN lowered pain thresholds in sham-operated mice. These data, combined with our previous findings obtained in mice with chronic inflammatory pain, suggest that an enhanced formation/reduced degradation of WFA + PNNs in regions of the pain matrix is associated with different types of chronic pain and may drive mechanisms of nociceptive sensitization leading to reduced mechanical and thermal pain thresholds.

Keywords: Chondroitinase ABC; Neuropathic pain; Perineuronal nets; Prefrontal cortex; Reticular thalamic nuclei; Somatosensory cortex; Type-9 matrix metalloproteinase.

Figure 1.

Chronic constriction injury of the sciatic nerve causes a reduction in the mechanical and thermal pain thresholds after 14 days. Mechanical and thermal pain thresholds are shown in (A and B), respectively. Note that CCI reduced mechanical and thermal pain thresholds to the same extent after 8 and 14 days. Statistical significance was determined by 2-way ANOVA for repeated measures followed by Sidak multiple comparison test. Data are means + S.E.M. of 16 mice per group. *P < 0.05; ***P < 0.001; ****P < 0.0001. SO, sham operated.

Figure 2.

Chronic constriction injury causes an increased density of WFA⁺ PNNs in the contralateral SSC. Representative images of WFA⁺PNNs/PV⁺ cells (A). The density of WFA⁺ and PV⁺ cells enwrapped by WFA⁺ is shown in (B and C). Statistical significance was determined by 2-way ANOVA followed by Fisher LSD test where values are means + S.E.M. of 7 mice per group. ***P < 0.001; ****P < 0.0001. SO, sham operated; ipsi, ipsilateral SSC; contra, contralateral SSC; ROI, region of interest. Representative panels of WFA⁺PNNs/SSt⁺ cells and WFA⁺PNNs/5HT3A⁺ cells are shown in (D and E), respectively. Scale bars are 100 μm and 50 μm in the upper and lower panel in (A), respectively.

Figure 3.

No changes in the density of Cat-315⁺ PNNs and aggrecan⁺ PNNs in the contralateral SSC. Representative images of Cat-315⁺ PNNs/PV⁺ cells (A) and aggrecan⁺ PNNs in (D). The density of Cat-315⁺ PNNs and PV⁺ cells enwrapped by Cat-315 is shown in (B and C), respectively. Two-way ANOVA followed by Fisher LSD test where values are means + S.E.M. of 7 mice per group. SO, sham operated; ipsi, ipsilateral SSC; contra, contralateral SSC. No changes in the density of aggrecan⁺ PNNs (E). Scale bars 100 μm.

Figure 4.

Changes in the proteoglycan constituents of PNN caused by CCI in the SSC. Representative immunoblots of aggrecan, neurocan, low and high molecular size brevican, and tenascin are shown in (A–D). Statistical significance was determined by 2-way ANOVA followed by Fisher LSD test where densitometric values are means + S.E.M. of 5 mice per group. *P < 0.05: **P < 0.01; ****P < 0.0001. Levels of the transcripts of the genes encoding aggrecan (Acan), neurocan (Ncan), and brevican (Bcan) are shown in (E-G), where values are means + S.E.M. of 4 to 5 mice per group. *P < 0.05. SO, sham operated; ipsi, ipsilateral SSC; contra, contralateral SSC.

Figure 5.

Chronic constriction injury causes an increased density of WFA⁺ PNNs in the contralateral mPFC. Representative images of WFA⁺PNNs/PV⁺ cells in the 3 regions of the contralateral mPFC of SO and CCI mice are shown in (A and D); AC, anterior cingulate cortex; PL, prelimbic cortex; IL, infralimbic cortex. The density of WFA⁺PNNs and PV⁺ cells enwrapped by WFA⁺ is shown in (B and C). Statistical significance was determined by 2-way ANOVA followed by Fisher LSD where values are means + S.E.M. of 7 mice per group. ***P < 0.001; ****P < 0.0001. The density of WFA+PNNs in AC and PL and IL is shown in (E and F). Normal distribution was assessed by Shapiro–Wilk test, unpaired 2-tailed Student t test where values are means + S.E.M. of 7 mice per group. ****P < 0.0001. Scale bars are 200 μm in (A) and 50 μm in (D).

Figure 6.

No changes in the density of Cat-315⁺ PNNs and aggrecan⁺ PNNs in the contralateral mPFC. Representative images of Cat-315⁺ PNNs/PV⁺ cells and aggrecan⁺ PNNs are shown in (A and C), respectively. Density values are shown in (B and D) (n = 7 mice per group). Representative immunoblots of aggrecan, neurocan, low and high molecular size brevican, and tenascin are shown in (E–H). Statistical significance was determined by 2-way ANOVA followed by Fisher LSD test. Densitometric values are means + S.E.M. of 5 mice per group. *P < 0.05; **P < 0.01. SO, sham operated; ipsi, ipsilateral mPFC; contra, contralateral mPFC. Scale bars 100 μm.

Figure 7.

Chronic constriction injury causes an increased density of WFA⁺ PNNs in the contralateral INS and RTN. Representative images of WFA⁺PNNs/PV⁺ cells in the INS and RTN are shown in (A and D), respectively. The density of WFA⁺ PNNs and PV⁺ cells enwrapped by WFA⁺ PNNs in the INS and RTN is shown in (B and C) and (E and F), respectively. Statistical significance was determined by 2-way ANOVA followed by Fisher LSD test where values are means + S.E.M. of 7 mice per group. *P < 0.05; **P < 0.01; ***P < 0.001. SO, sham operated; ipsi, ipsilateral INS or RTN; contra, contralateral INS or RTN. Scale bars 100 μm.

Figure 8.

Enzymatic degradation of PNNs in the contralateral SSC or RTN causes analgesia in CCI mice. Representative images of WFA⁺PNNs/PV⁺ cells after infusion of chondroitinase ABC (ChABC) or vehicle in the contralateral SSC and RTN of CCI mice are shown in (A and B), respectively. Mechanical and thermal pain thresholds after infusions in the SSC are shown in (C and D), respectively. Statistical significance was determined by 2-way ANOVA for repeated measures followed by Sidak multiple comparison test, where values are means + S.E.M. of 8 mice per group. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Mechanical and thermal pain thresholds after infusions in the RTN are shown in (E and F), respectively. Two-way ANOVA for repeated measures followed by Sidak multiple comparison test, where values are means + S.E.M. of 6/8 mice per group. *P < 0.05; **P < 0.01; ***P < 0.001. Scale bars = 100 μm. Pre = preinfusion time (8 days after CCI); Post = postinfusion time (4 days after infusion, corresponding to 14 days after CCI).

Figure 9.

Injection of MMP-9 siRNA in the contralateral SSC or RTN reduces pain thresholds in sham operated mice. Representative images of WFA⁺PNNs/PV⁺ cells after infusion of MMP-9 siRNA in SSC and RTN are shown in (A and B), respectively. The density of WFA⁺ PNNs is shown in (C and D). Statistical significance was determined by 2-way ANOVA followed by Fisher LSD test, where values are means + S.E.M. of 7 mice per group. *P < 0.05; **P < 0.01; ***P < 0.001. Mechanical and thermal pain thresholds are shown in (E) and (F) and (G) and (H), respectively. Two-way ANOVA for repeated measures followed by Sidak multiple comparison test, where values are means + S.E.M. of 8 to 16 mice per group. Post = postinfusion time (4 days after infusion, corresponding to 14 days after CCI). **P < 0.01; ***P < 0.001; ****P < 0.0001; Scale bar 100 μm.