Spinal-Specific Super Enhancer in Neuropathic Pain

Abstract

Dysfunctional gene expression in nociceptive pathways plays a critical role in the development and maintenance of neuropathic pain. Super enhancers (SEs), composed of a large cluster of transcriptional enhancers, are emerging as new players in the regulation of gene expression. However, whether SEs participate in nociceptive responses remains unknown. Here, we report a spinal-specific SE (SS-SE) that regulates chronic constriction injury (CCI)-induced neuropathic pain by driving Ntmt1 and Prrx2 transcription in dorsal horn neurons. Peripheral nerve injury significantly enhanced the activity of SS-SE and increased the expression of NTMT1 and PRRX2 in the dorsal horn of male mice in a bromodomain-containing protein 4 (BRD4)-dependent manner. Both intrathecal administration of a pharmacological BRD4 inhibitor JQ1 and CRISPR-Cas9-mediated SE deletion abolished the increased NTMT1 and PRRX2 in CCI mice and attenuated their nociceptive hypersensitivities. Furthermore, knocking down Ntmt1 or Prrx2 with siRNA suppressed the injury-induced elevation of phosphorylated extracellular-signal-regulated kinase (p-ERK) and glial fibrillary acidic protein (GFAP) expression in the dorsal horn and alleviated neuropathic pain behaviors. Mimicking the increase in spinal Ntmt1 or Prrx2 in naive mice increased p-ERK and GFAP expression and led to the genesis of neuropathic pain-like behavior. These results redefine our understanding of the regulation of pain-related genes and demonstrate that BRD4-driven increases in SS-SE activity is responsible for the genesis of neuropathic pain through the governance of NTMT1 and PRRX2 expression in dorsal horn neurons. Our findings highlight the therapeutic potential of BRD4 inhibitors for the treatment of neuropathic pain.SIGNIFICANCE STATEMENT SEs drive gene expression by recruiting master transcription factors, cofactors, and RNA polymerase, but their role in the development of neuropathic pain remains unknown. Here, we report that the activity of an SS-SE, located upstream of the genes Ntmt1 and Prrx2, was elevated in the dorsal horn of mice with neuropathic pain. SS-SE contributes to the genesis of neuropathic pain by driving expression of Ntmt1 and Prrx2 Both inhibition of SS-SE with a pharmacological BRD4 inhibitor and genetic deletion of SS-SE attenuated pain hypersensitivities. This study suggests an effective and novel therapeutic strategy for neuropathic pain.

Keywords: BRD4 inhibitor; Ntmt1; Prrx2; neuropathic pain; super enhancer.

Figure 1.

Ntmt1/Prrx2-linked-SE is a spinal-specific super enhancer. a, Super enhancer signal distribution. The signal strength of SE signals from −5000 to about +5000 of all genes (transcription start site denoted as +1) were detected in the dorsal horn of male mice with CCI or sham surgery. Dorsal horns were collected on day 7 after surgery and subjected to anti-H3K27ac ChIP-Seq; n = 2 repeats. Male mice were used in the experiments unless female was mentioned. b, c, SEs were identified in the dorsal horn of sham mice (b) or CCI mice (c) according to the ROSE algorithm. d, e, Gene Ontology (d) and KEGG pathway (e) analysis for activated SE-related genes in the dorsal horn on day 7 after CCI surgery. f, The chromosome location, potential associated genes (Ntmt1, Prrx2, Asb6, and Ptges), and the constituent enhancers (E1–E3) of a super enhancer with significantly increased signal after nerve injury. Input is the noise signal from secondary antibody IgG in ChIP-Seq. g, Identification of spinal cord–specificity of the Ntmt1/Prrx2-SE by comparing enhancer characteristics in the same genomic region (Chr 2, 30645088–30797673) from five nervous system tissues and six peripheral tissues. The constituent enhancer information for the other 11 tissues was downloaded from the SEA database of super enhancers.

Figure 2.

Suppression of Ntmt1/Prrx2-SE activity by a pharmacological BRD4 inhibitor alleviates neuropathic pain sensitivities. a, The expression of BRD4 in dorsal horn on day 7 after CCI surgery was not changed; n = 6 mice/group. b, Binding activity of BRD4 to three SS-SE clusters. ChIP-PCR with anti-BRD4 in the dorsal horn of Sham mice or CCI mice on day 7 after surgery; n = 4 mice/group' *p < 0.05, **p < 0.01 versus Sham. c, Binding activity of RNA polymerase II (RPII) to Ntmt1/Prrx2-SE clusters. ChIP-PCR with anti-RPII in the dorsal horn of Sham mice or CCI mice on day 7 after surgery; n = 3 mice/group; ***p < 0.001 versus Sham+DMSO; ^#p < 0.05, ^##p < 0.01 versus CCI+DMSO. d, Intrathecal injection of JQ1 7 d after CCI reduced the activity of Ntmt1/Prrx2-SE in ipsilateral dorsal horn 24 h after injection. SE activity was identified by H3K27ac-ChIP-qPCR. DMSO was used as the vehicle control; n = 4 mice/group, *p < 0.05, **p < 0.01, ***p < 0.001 versus Sham+DMSO; ^#p < 0.05, ^###p < 0.001, CCI+DMSO. e–j, Time course of the effect of intrathecal injection of JQ1 or vehicle control into CCI or sham male (e–g) and female (h–j) mice on the ipsilateral PWF to 0.07 g and 0.4 g von Frey filaments and on PWLs to heat stimuli; n = 8 mice/group, ***p < 0.001, versus the Sham+DMSO group at the corresponding time points; ^#p < 0.05, ^##p < 0.01, ^###p < 0.001 versus CCI+DMSO at the corresponding time points; two-way ANOVA, post hoc Tukey's tests. Red arrows indicate CCI or sham surgery. Blue arrows indicate JQ1 or vehicle control injection.

Figure 3.

CRISPR/Cas9-mediated genetic deletion of Ntmt1/Prrx2-SE attenuates nerve-injury-induced nociceptive hypersensitivity. a, Agarose gel electrophoresis showing the activity of the Ntmt1/Prrx2-SE cluster in dorsal horn on day 5 after intrathecal injection of gRNA-74 (74) or gRNA-107 (107) and Lenti-CRISPR/Cas9 for SE1 in mice 7 d after CCI or sham surgery; n = 3–4 mice/group, *p < 0.05, **p < 0.01 versus Sham+Scr; ^##p < 0.01 versus CCI+Scr. Top left, Diagram illustrates CRISPR/Cas9-mediated SEs knockout strategy by deleting the two regions flanking the enhancer cluster of Ntmt1/Prrx2-SE. gRNA, guide RNA. The number following gRNA indicates the gRNA location in each SE cluster (the first nucleotide of each cluster is designated as +1). b–d, Time course of the effect of intrathecal injection of gRNA-74 and Lenti-CRISPR/Cas9 in CCI or sham mice on mechanical allodynia (b, c) and heat hyperalgesia (d). Red arrows indicate CCI or sham surgery. Blue arrows indicate gRNAs and CRISPR-Cas9 injection; n = 8 mice/group, ***p < 0.001 versus Sham+Scr; ^#p < 0.05, ^##p < 0.01, ^###p < 0.001 versus CCI+Scr. e, The activity of the Ntmt1/Prrx2-SE2 cluster in the dorsal horn on day 5 after intrathecal injection of sgRNA-134 (134) or gRNA-91 (91) and CRISPR/Cas9 for SE2 in CCI or sham mice; n = 3–4 mice/group, *p < 0.05 versus Sham+Scr, ^#p < 0.05 versus CCI+Scr. f–h, Time course of the effect of intrathecal injection of sgRNA-134 and CRISPR/Cas9 in CCI or sham mice on mechanical sensitivity (f, g) and heat hyperalgesia (h); n = 8 mice/group, **p < 0.01, ***p < 0.001 versus Sham+Scr mice; ^##p < 0.01, ^###p < 0.001 versus CCI+Scr mice. i, Ntmt1/Prrx2-SE3 activity on day 5 after intrathecal injection of gRNA-104 (104) or gRNA-187 (187) and CRISPR/Cas9 for SE3 in the dorsal horn 7 d after CCI or sham surgery; n = 3–4 mice/group, *p < 0.05 versus Sham+Scr, ^#p < 0.05 versus CCI+Scr. j–l, The effect of intrathecal injection of sgRNA-104 and CRISPR/Cas9 in CCI or sham mice on mechanical sensitivity (j, k) and heat hyperalgesia (l) stimuli; n = 8 mice/group, ***p < 0.001 versus Sham+Scr; ^##p < 0.01, ^###p < 0.001 versus CCI+Scr.

Figure 4.

Nerve injury upregulates SE-related Ntmt1 and Prrx2 in the dorsal horn. a–d, Levels of four potential SE-associated genes—Ntmt1 (a), Prrx2 (b), Asb6 (c), and Ptges (d)—in the ipsilateral dorsal horn up to 14 days after CCI of the unilateral sciatic nerve; n = 6 mice/time point/group, *p < 0.05, **p < 0.01, ***p < 0.001 versus the corresponding contralateral side; two-way ANOVA followed by post hoc Tukey's tests. e, mRNA expression levels of Ntmt1, Prrx2, Ptges, and Asb6 mRNA in ipsilateral DRG on day 7 after CCI or Sham surgery of the unilateral sciatic nerve; n = 4 mice/group; CCI group versus Sham group; two-tailed unpaired Student's t test. f, g, Expression levels of NTMT1 (f) and PRRX2 (g) protein in the ipsilateral dorsal horn up to 14 days after CCI surgery of the unilateral sciatic nerve; n = 4 mice/time point/group, *p < 0.05, **p < 0.01, ***p < 0.001 versus the corresponding contralateral side; two-way ANOVA followed by post hoc Tukey's tests. h, Expression levels of NTMT1 (top) and PRRX2 (bottom) in the ipsilateral DRG on day 7 after CCI or Sham surgery; n = 4 mice/group; CCI group versus Sham group; two-tailed unpaired Student's t test. i–l, Combined immunofluorescence staining for NTMT1 (i, j) or PRRX2 (k, l, red) and NeuN (a neuronal marker, green), GFAP (an astrocyte marker, green), and IBA1 (a microglial marker, green) and their overlay analysis in ipsilateral spinal cord of naive mice; n = 3 repeats. Scale bar, 50 μm.

Figure 5.

SS-SE drives Ntmt1 and Prrx2 upregulation after nerve injury. a, Luciferase activity of a reporter inserted by the individual Ntmt1/Prrx2-SE clusters (pGL6-SE1, 2, and 3; SE1, 2, and 3, respectively) at 48 h after transfection of pGL6-SE1, 2, or 3 or the pGL6 control and PBS negative control (Control) in HEK-293T cells; n = 3 repeats/treatment, ***p < 0.001 versus the pGL6 control group; one-way ANOVA followed by post hoc Tukey's tests. b–e, The effect of suppressing Ntmt1/Prrx2-SE activity by intrathecal injection of the BRD1 inhibitor JQ1 or vehicle control (DMSO) on Ntmt1 (b, c) and Prrx2 (d, e) mRNA and protein expression in the dorsal horn 7 d after CCI or Sham surgery. Data were obtained 6 h after the intrathecal injection; n = 5 mice/group, *p < 0.05, ***p < 0.001 versus Sham+DMSO; ^#p < 0.05, ^###p < 0.001 versus CCI+DMSO. f, g, Effect of Ntmt1/Prrx2-SE1 cluster knockout by CRISPR-Cas9 on NTMT1 (f) and PRRX2 (g) protein levels in the dorsal horn. SE1-sgRNA and Lenti-CRISPR/Cas9 were coinjected 7 d after CCI or Sham surgery, and data were obtained 5 d after coinjection; n = 5 mice/group, *p < 0.05, **p < 0.01 versus Sham+Scr; ^#p < 0.05 versus CCI+Scr; two-way ANOVA with repeated measures followed by post hoc Tukey's tests. h, i, NTMT1 (h) and PRRX2 (i) protein levels in the dorsal horn 5 d after coinjection of SE2-sgRNA and Lenti-CRISPR/Cas9 in 7 d post-CCI or after sham surgery mice; n = 4 mice/group, **p < 0.01 versus Sham+Scr; ^#p < 0.05, ^###p < 0.001 versus CCI+Scr; two-way ANOVA with repeated measures followed by post hoc Tukey's tests. j, k, Expression levels of NTMT1 (j) and PRRX2 (k) protein in the dorsal horn after deletion of the Ntmt1/Prrx2-SE2 cluster. Data were obtained 5 d after coinjection of SE3-sgRNA and Lenti-CRISPR/Cas9 in CCI or sham mice; n = 4 mice/group, *p < 0.05 versus Sham+Scr; ^#p < 0.05 versus CCI+Scr; two-way ANOVA with repeated measures followed by post hoc Tukey's tests.

Figure 6.

Blocking Ntmt1 and Prrx2 inhibits the genesis of neuropathic pain. a, The Ntmt1 expression level on 48 h after Ntmt1 siRNA-347 and siRNA-548 (transcription start site designated as +1) transfection in HT-22 cell; **p < 0.01 versus Scr. b, c, Expression levels of dorsal horn Ntmt1 mRNA (b) and protein (c) 3 d after intrathecal injection of Ntmt1-siRNA-548 (siR) or scrambled siRNA (Scr) in CCI mice; n = 5 mice/group, *p < 0.05 versus Sham+Scr group; ^##p < 0.01, ^###p < 0.001 versus CCI+Scr group; one-way ANOVA with repeated measures followed by post hoc Tukey's tests. d–f, Effect of knockdown with Ntmt1-siRNA on nerve-injury-induced mechanical (d, e) and thermal (f) hypersensitivities; n = 8 mice/group, *p < 0.05, **p < 0.01, ***p < 0.001 versus Sham+Scr group; ^##p < 0.01, ^###p < 0.001 versus the CCI+Scr group at the corresponding time points; two-way ANOVA with repeated measures followed by post hoc Tukey's tests. g, The Prrx2 expression level on 48 h after Prrx2 siRNA-564 and siRNA-367 transfection in HT-22 cell; **p < 0.01 versus Scr. h, i, Expression levels of dorsal horn Prrx2 mRNA (h) and protein (i) 3 d after intrathecal injection of Prrx2-siRNA-564 (siR) or scrambled siRNA (Scr) in CCI mice; n = 5 mice/group, **p < 0.01, ***p < 0.001 versus Sham+Scr group; ^#p < 0.05, ^##p < 0.01 versus CCI+Scr group; one-way ANOVA with repeated measures followed by post hoc Tukey's tests. j–l, Effect of Prrx2 knockdown with siRNA on CCI-induced mechanical (j, k) and thermal (l) hypersensitivities; n = 6–8 mice/group, ***p < 0.001 versus Sham+Scr group; ^##p < 0.01, ^###p < 0.001 versus CCI+Scr group at the corresponding time points; two-way ANOVA with repeated measures followed by post hoc Tukey's tests.

Figure 7.

Mimicking the upregulation of NTMT1 and PRRX2 induces neuropathic pain-like behaviors. a, b, Ntmt1 mRNA (a) and protein (b) levels on day 5 after intrathecal injection of Lenti-Ntmt1 (Ntmt1) or Lenti-Gfp (Gfp) in naive mice; n = 5 mice/group, *p < 0.05 versus the Gfp group; two-tailed unpaired Student's t test. c–h, Effect of Ntmt1 upregulation by Lenti-Ntmt1 on sensitivity to mechanical and thermal stimuli in male (c–e) and female (f–h) naive mice; n = 6-8 mice/group, *p < 0.05, **p < 0.01, and ***p < 0.001 versus the Gfp group; two-way ANOVA with repeated measures followed by post hoc Tukey's tests. i, j, Prrx2 mRNA (i) and protein (j) levels on day 5 after intrathecal injection of Lenti-Prrx2 (Prrx2) or Lenti-Gfp (Gfp) in naive mice; n = 5 mice/group, **p < 0.01 and ***p < 0.001 versus the Gfp group; two-tailed unpaired Student's t test. k–p, Effect of Prrx2 upregulation on sensitivity to mechanical and thermal stimuli in male (k–m) and female (n–p) naive mice; n = 6–8 mice/group, *p < 0.05, **p < 0.01, and ***p < 0.001 versus the Gfp group; two-way ANOVA with repeated measures followed by post hoc Tukey's tests.

Figure 8.

Ntmt1 and Prrx2 contribute to neuropathic pain by enhancing cellular hyperactivation in the dorsal horn. a, b, Protein level of p-ERK1/2, a marker of spinal cells hyperactivation) and GFAP (a marker for astrocytic hyperactivation) in the ipsilateral dorsal horn on day 2 after intrathecal injection of Ntmt1-siRNA (siR; a), Prrx2-siRNA (siR; b), or scrambled siRNA (Scr) in CCI mice; *p <0.05, **p < 0.01, ***p < 0.001 versus the corresponding groups; two-way ANOVA, post hoc Tukey's test; n = 5 mice/group. c, d, Protein levels of p-ERK1/2, GFAP, and IBA1 (a marker for microglial hyperactivation) in the dorsal horn on day 5 after intrathecal injection of Lenti-Ntmt1 (Ntmt1; c), Lenti-Prrx2 (d), or Lenti-Gfp in naive mice, *p <0.05, **p < 0.01, ***p < 0.001 versus the corresponding groups; two-way ANOVA, post hoc Tukey's test, n = 4 mice/group.

Figure 9.

Activation of the spinal-specific super enhancer contributes to the genesis of neuropathic pain by driving Ntmt1 and Prrx2 expression in dorsal horn neurons. Nerve injury induces the formation of a spinal-specific SE via the binding of H3K27ac and BRD4 to the chromatin region located upstream of the Ntmt1 and Prrx2 genes. This activated SE further recruits transcription factors and RNA polymerase II to promote the transcription of Ntmt1 and Prrx2, resulting in hyperactivation of dorsal horn neurons in the spinal cord. Disruption of this SE by the BRD4 inhibitor JQ1 alleviates nerve-induced pain hypersensitivity by downregulating the NTMT1 and PRRX2 levels in the dorsal horn.