Fig 1.
mGluR5 receptor decreased glycinergic currents.
(A–B) Bath perfusion of DHPG for 3 min reduced GlyR-IPSCs (A; 70.2 ± 4.2% of baseline at 15–20 min post-DHPG, t[16] = 6.707, p < 0.001, paired Student t test), but not GABAA-receptor–mediated IPSCs (B; 102.9 ± 3.3% of baseline at 15–20 min post-DHPG, t[5] = 0.347, p = 0.743) in spinal slices of mice. The horizontal bar indicated the period of DHPG perfusion. The original traces were taken at the time points indicated by the numbers 1–3. (C) The inhibitory effect of DHPG on GlyR-IPSCs was blocked by MPEP (92.4 ± 4.4% of baseline at 15–20 min post-DHPG, t[5] = 1.573, p = 0.177) but not by CPCCOEt (53.5 ± 3.7% of baseline at 15–20 min post-DHPG, t[5] = 14.343, p < 0.001). (D) Bath perfusion of CHPG suppressed GlyR-IPSCs (72.6 ± 5.3% of baseline at 15–20 min post-CHPG, t[8] = 6.51, p < 0.001). (E) The inhibitory effect of CHPG on GlyR-IPSCs was not blocked by intracellularly loaded GDP-β-S (62.3 ± 6.4% of baseline at 15–20 min post-CHPG, t[6] = 2.782, p = 0.032), chelerythrine (60.9 ± 11.3% of baseline at 15–20 min post-CHPG, t[6] = 2.705, p = 0.035), or Ro-32-0432 (69.9 ± 3.2% of baseline at 15–20 min post-CHPG, t[5] = 8.495, p < 0.001). (F) Postsynaptic loading of U-0126 or PD98059 prevented CHPG from decreasing glycinergic responses (U-0126, 107.6 ± 10.4% of baseline at 15–20 min post-CHPG, t[8] = 0.997, p = 0.348; PD98059, 93.1 ± 5.0% of baseline at 15–20 min post-CHPG, t[5] = 0.883, p = 0.418). (G) Effects of DHPG (10 μM, 3 min) on the whole-cell currents evoked by exogenously applied glycine (1 mM; 5 s) in the absence (control) or presence of intracellular U-0126 loading. *p < 0.001 versus baseline (t[10] = 5.235, paired t test). n = 11 (control) and 8 neurons (U-0126). (H) DHPG did not affect the paired-pulse ratios of GlyR-IPSCs (n = 10). The underlying data for this figure can be found in S1 Data. Error bars indicated SEM. CHPG, α-amino-2-chloro-5-hydroxybenzeneacetic acid; DHPG, (S)-3,5-Dihydroxyphenylglycine; GABAA receptor, γ-Aminobutyric acid type A receptor; GDP-β-S, Guanosine 5ʹ-O-(2-Thiodiphosphate); GlyR, glycine receptor; IPSC, inhibitory postsynaptic current; mGluR5, metabotropic glutamate receptor 5; MPEP, 6-Methyl-2-(phenylethynyl) pyridine; PD98059, 2ʹ-Amino-3ʹ-methoxyflavone; Ro-32-0432, 2-{8-[(Dimethylamino)methyl]-6,7,8,9-tetrahydropyrido[1,2-a]indol-3-yl}-3-(1-methyl-1H-indol-3-yl)maleimide.
Fig 2.
mGluR5 specifically decreased α1ins currents in HEK293T cells through ERK signaling.
(A) Effects of DHPG on glycine (1 mM, 10 ms)-induced whole-cell currents in HEK293T cells co-transfected with mGluR5a and α1ins (68.3 ± 5.9% of baseline at 10–15 min post-DHPG, t[7] = 2.944, p = 0.022, paired Student t test), α1 (94.3 ± 4.8% of baseline at 10–15 min post-DHPG, t[7] = 0.138, p = 0.894), or α3L (101.8 ± 9.2% of baseline at 10–15 min post-DHPG, t[7] = 0.49, p = 0.639). (B) The inhibitory effect of DHPG on α1ins currents was blocked by U-0126 (99.5 ± 7.0% of baseline at 10–15 min post-DHPG, t[7] = 0.032, p = 0.975) and dynasore (102.3 ± 10.4% of baseline at 10–15 min post-DHPG, t[6] = 0.714, p = 0.502) but not by chelerythrine (64.7 ± 5.1% of baseline at 10–15 min post-DHPG, t[5] = 4.487, p = 0.006). The underlying data for this figure can be found in S1 Data. Error bars indicated SEM. DHPG, (S)-3,5-Dihydroxyphenylglycine; ERK, extracellular signal-regulated kinase; HEK, human embryonic kidney; mGluR5, metabotropic glutamate receptor 5.
Fig 3.
Role of α1ins in spinal nociceptive modification.
(A) Immunohistochemical analysis of α1ins (red) and gephyrin (green) expression in the spinal cord of mice. Scale bar: 100 μm. (B) α1ins and gephyrin puncta in the superficial dorsal horn at high magnification. The arrows indicated α1ins colocalization with gephyrin. The graphs showed the percent of α1ins puncta at synapses (synaptic α1ins/total α1ins puncta) and percent of synapses expressing α1ins (synaptic α1ins/total gephyrin puncta). n = 21 slices from 5 mice. Scale bar: 5 μm. (C) GlyR-mIPSC amplitudes were decreased at day 28 after intraspinal injection of AAV encoding shRNA-α1ins. *p = 0.038 versus NC (Mann–Whitney U test). n = 8 neurons/group. (D–F) shRNA-α1ins elicited mechanical allodynia (D, F[4, 72] = 3.268, p = 0.016, repeated measures ANOVA, n = 10 mice/group), heat hyperalgesia (E, F[4, 56] = 3.405, p = 0.015, n = 8 mice/group), and cold hyperalgesia (F, F[4, 56] = 3.733, p = 0.009, n = 8 mice/group). *p < 0.05 versus NC (post hoc Bonferroni test). (G) shRNA-α1ins did not change the maximum tolerated RPM in accelerating rotarod test (F[4, 56] = 0.536, p = 0.71, n = 8 mice/group). (H) c-fos expression at day 28 after NC or shRNA-α1ins injection. *p < 0.001 versus NC (Mann–Whitney U test). n = 12 sections/group. Scale bar: 100 μm. (I) shRNA-α1ins blocked DHPG from reducing GlyR-IPSCs (104.5 ± 3.7% of baseline at 15–20 min post-DHPG, t[12] = 1.8, p = 0.097, paired Student t test), while NC had no effect (69.3 ± 9.2% of baseline at 15–20 min post-DHPG, t[8] = 2.774, p = 0.024). The underlying data for this figure can be found in S1 Data. Error bars indicated SEM. AAV, adeno-associated virus; ANOVA, Analysis of Variance; DHPG, (S)-3,5-Dihydroxyphenylglycine; GFP, green fluorescent protein; GlyR, glycine receptor; IPSC, inhibitory postsynaptic current; mIPSC, miniature IPSC; NC, negative control shRNA; RPM, rounds per minute; shRNA, short hairpin RNA.
Fig 4.
(A–B) Co-IP with ERK1/2 antibody from lysates of HEK293T cells transfected with GlyR-α1ins (A) or GlyR-α1 (B). The immunoprecipitates were probed with anti-α1 antibody. n = 3 experiments/group. (C) Co-IP with ERK1/2 antibody from spinal dorsal horn of mice. n = 3 experiments. (D) GST-α1ins-IL or GST-α1-IL was used to pull down ERK1/2 and pERK1/2 from lysates of DHPG (10 μM; 0–3 min)-treated spinal slices. The TCLs were also immunoblotted. The graph showed the percentage changes of ERK1/2 and pERK1/2 contents precipitated by GST-α1ins-IL. *p = 0.029 versus DHPG-untreated control (Mann–Whitney U test), n = 4 experiments. (E) The purified His-pERK2 was precipitated by GST-α1ins-IL but not by GST-α1-IL, GST-α3L-IL, or GST-β-IL. The total His (middle) and GST proteins (bottom) were also shown. n = 4 experiments. (F) GST-α1ins-IL, GST-α1-IL, GST-α3L-IL, and GST-β-IL did not pull down the purified nonphosphorylated His-ERK2. n = 4 experiments. (G) TAT-pep-α1ins reduced His-ERK2 contents precipitated by GST-α1ins-IL from lysates of HEK293T cells coexpressing MEK1(S218D/S222D). TAT-Scram was used as control. *p = 0.027, #p = 0.013 versus peptide-untreated control (one-way ANOVA with post hoc Bonferroni test), n = 6 experiments. (H) Intrathecal DHPG application (10 nmol, 10 min) increased α1ins/ERK interaction, which was blocked by pretreatment with TAT-pep-α1ins (200 pmol) for 30 min. *p = 0.023 versus control, #p = 0.006 versus DPHG (one-way ANOVA with post hoc Bonferroni test), n = 6. (I) Intracellular loading of TAT-pep-α1ins prevented DHPG from inhibiting glycine-evoked whole-cell currents in HEK293T cells coexpressing mGluR5a and α1ins (93.1 ± 4.5% of baseline at 10–15 min post-DHPG, t[6] = 1.331, p = 0.231, paired Student t test), whereas TAT-Scram had no effect (57.3 ± 6.1% of baseline at 10–15 min post-DHPG, t[5] = 3.829, p = 0.012). (J–K) Postsynaptic loading of TAT-pep-α1ins prevented DHPG from suppressing GlyR-IPSCs (J, 92.2 ± 7.1% of baseline at 15–20 min post-DHPG, t[9] = 1.333, p = 0.215, paired Student t test), while TAT-Scram had no effect (K; 67.8 ± 8.2% of baseline at 15–20 min post-DHPG, t[7] = 3.761, p = 0.007). The underlying data for this figure can be found in S1 Data. Error bars indicated SEM. ANOVA, Analysis of Variance; ERK, extracellular signal-regulated kinase; DHPG, (S)-3,5-Dihydroxyphenylglycine; GlyR, glycine receptor; GST, Glutathione S-Transferase; HEK, human embryonic kidney; IgG, immunoglobulin G; IL, intracellular large loop; IP, immunoprecipitation; IPSC, inhibitory postsynaptic current; MEK, mitogen-activated protein kinase kinase; mGluR5, metabotropic glutamate receptor 5; pERK, phosphorylated ERK; TAT, human immunodeficiency virus-type 1 TAT sequence; TAT-pep-α1ins, TAT-fused α1ins-derived peptide; TAT-Scram, TAT-fused scrambled peptide; TCL, total cell lysate.
Fig 5.
ERK phosphorylated α1ins at Ser380.
(A) DHPG reduced glycine (1 mM, 10 ms)-evoked currents in HEK293T cells expressing mGluR5a along with α1ins(S326A) (69.2 ± 3.5% of baseline at 10–15 min post-DHPG, t[8] = 3.978, p = 0.004, paired Student t test) but not with α1ins(S380A) (86.3 ± 5.1% of baseline at 10–15 min post-DHPG, t[8] = 1.974, p = 0.084). (B) LC MS/MS analysis identified Ser380 as a phosphorylation site on GST-α1ins-IL. (C) Myc-α1ins or Myc-α1ins(S380A) was transfected along with mGluR5a in HEK293T cells, immunoprecipitated by Myc antibody, and immunoblotted with pS380-Ab. Representative western blots showed pS380-Ab signals with or without DHPG (10 μM) stimulation. *p = 0.002 (Mann–Whitney U test), n = 6. (D) Ser380 phosphorylation was catalyzed by ERK. GST-α1ins-IL or its mutant GST-α1ins-IL(S380A) was incubated in vitro with purified His-pERK2 for 30 min before immunoblotting with pS380-Ab (up). The total His (middle) and GST proteins (bottom) were also probed. n = 6. (E) TAT-pep-α1ins dose-dependently reduced GST-α1ins-IL phosphorylation catalyzed by His-pERK2 in vitro. TAT-Scram was used as control. *p = 0.026, #p = 0.009 versus peptide-untreated group (one-way ANOVA with post hoc Bonferroni test), n = 6. (F) Intrathecal DHPG application (10 min) increased pS380-Ab signals, which was attenuated by pretreatment for 30 min with TAT-pep-α1ins (left). Preabsorbing with excessive antigen (middle) or treatment with alkaline phosphatase (right) eliminated pS380-Ab signals. *p < 0.001 versus control, #p < 0.001 versus DHPG (one-way ANOVA with post hoc Bonferroni test), n = 8. (G) Effects of intrathecal TAT-pep-α1ins (pep) and TAT-Scram on DHPG (10 nmol)-induced spontaneous pain behaviors. F(25, 175) = 1.644, p = 0.033 (repeated measures ANOVA). n = 8 mice/group. (H) Intraspinal injection of AAV encoding shRNA-α1ins, but not NC, induced spontaneous pain behaviors (F[3, 30] = 15.459, p < 0.001, repeated measures ANOVA, n = 6 mice/group). Note that the painful behaviors in shRNA-α1ins mice partially occluded those caused by subsequent DHPG application. The arrow indicated the time point when intrathecal DHPG was given at day 21 post-viral injection. The underlying data for this figure can be found in S1 Data. Error bars indicated SEM. AAV, adeno-associated virus; ANOVA, Analysis of Variance; DHPG, (S)-3,5-Dihydroxyphenylglycine; ERK, extracellular signal-regulated kinase; GlyR, glycine receptor; GST, Glutathione S-Transferase; HEK, human embryonic kidney; IL, intracellular large loop; LC MS/MS, liquid chromatograph/mass spectrometer; mGluR5, metabotropic glutamate receptor 5; NC, negative control shRNA; pERK, phosphorylated ERK; pS380-Ab, phosphorylation-state–specific antibody against Ser380 on α1ins; shRNA, short hairpin RNA; TAT-pep-α1ins, TAT-fused α1ins-derived peptide; TAT-Scram, TAT-fused scrambled peptide.
Fig 6.
Ser380 phosphorylation promoted α1ins ubiquitination and endocytosis.
(A) Intrathecal application of DHPG (10 nmol, 10 min) enhanced α1ins ubiquitination. Effects of spinal U-0126 (2 nmol) and TAT-pep-α1ins (200 pmol) were also examined. *p < 0.001 versus control, #p < 0.001 versus DHPG (one-way ANOVA with post hoc Bonferroni test), n = 6. (B) Myc-α1ins(S380A) inhibited, while Myc-α1ins(S380D) occluded, the increase of Myc-α1ins ubiquitination caused by DHPG in cultured neurons. *p = 0.004 versus control (one-way ANOVA with post hoc Bonferroni test), n = 6. (C) LC MS/MS analysis identified Lys379 on Myc-α1ins as a ubiquitination site. (D) DHPG-induced Myc-α1ins ubiquitination was attenuated by Lys379 mutation to Arginine. *p < 0.001 versus DHPG-untreated control (one-way ANOVA with post hoc Bonferroni test), n = 6. (E) Eps15 contents precipitated by Myc antibody from neurons transfected with Myc, Myc-α1ins, Myc-α1ins(S380A), Myc-α1ins(S380D), or Myc-α1ins(K379R/S380D). *p = 0.001 and #p = 0.001 versus Myc-α1ins, ##p < 0.001 versus Myc-α1ins(S380D) (one-way ANOVA with post hoc Bonferroni test), n = 6. (F–H) Internalized (red) and surface remaining Myc-α1ins (green, F) or Myc-α1ins(S380A) (green, G) in cultured neurons untreated or treated with DHPG (10 μM). The ratios of internalized to total fluorescence intensities were averaged (H). *p < 0.001 versus DHPG-untreated control (Mann–Whitney U test), n = 30 cells/group. Scale bar, 5 μm. The underlying data for this figure can be found in S1 Data. Error bars indicated SEM. ANOVA, Analysis of Variance; DHPG, (S)-3,5-Dihydroxyphenylglycine; Eps15, epidermal growth factor receptor substrate 15; IP, immunoprecipitation; LC MS/MS, liquid chromatograph/mass spectrometer; MAP2, Microtubule-Associated Protein 2; TAT-pep-α1ins, TAT-fused α1ins-derived peptide; Ubi, ubiquitin.
Fig 7.
TAT-pep-α1ins attenuated inflammatory pain.
(A) Formalin injection into left hindpaws of mice enhanced pS380 in the ipsilateral dorsal horn of spinal cord. Effects of intrathecal MPEP (50 nmol), U-0126 (2 nmol), TAT-pep-α1ins, or TAT-Scram (200 pmol) on pS380 were also examined. *p = 0.017, #p = 0.007 versus contralateral sides (one-way ANOVA with post hoc Bonferroni test), n = 10. (B) Effects of formalin on α1ins ubiquitination. *p = 0.002, #p = 0.002 versus contralateral sides (one-way ANOVA with post hoc Bonferroni test), n = 6. (C) Effects of TAT-pep-α1ins and TAT-Scram on formalin-induced spontaneous pain. F(44, 396) = 1.652, p = 0.007 (repeated measures ANOVA). n = 10 mice/group. (D–E) The first-phase (0–10 min, D) and second-phase (15–60 min, E) behaviors in formalin tests were summarized. *p = 0.004 versus TAT-Scram–treated mice (one-way ANOVA with post hoc Bonferroni test). (F) GlyR-IPSCs in slices from formalin-injected mice were potentiated by TAT-pep-α1ins (144.9 ± 7.6% of baseline at 15–20 min postpeptide, t[15] = 6.495, p < 0.001, paired Student t test) but not by TAT-Scram (88.0 ± 6.5% of baseline at 15–20 min postpeptide, t[10] = 1.856, p = 0.093). The underlying data for this figure can be found in S1 Data. Error bars indicated SEM. ANOVA, Analysis of Variance; GlyR, glycine receptor; IP, immunoprecipitation; IPSC, inhibitory postsynaptic current; MPEP, 6-Methyl-2-(phenylethynyl) pyridine; pS380, phosphorylation at Ser380; TAT-pep-α1ins, TAT-fused α1ins-derived peptide; TAT-Scram, TAT-fused scrambled peptide; Ubi, ubiquitin.