Nerve injury-induced epigenetic silencing of opioid receptors controlled by DNMT3a in primary afferent neurons

Abstract

Opioids are the gold standard for pharmacological treatment of neuropathic pain, but their analgesic effects are unsatisfactory in part due to nerve injury-induced downregulation of opioid receptors in dorsal root ganglia (DRG) neurons. How nerve injury drives such downregulation remains elusive. DNA methyltransferase (DNMT)-triggered DNA methylation represses gene expression. We show here that blocking the nerve injury-induced increase in DRG DNMT3a (a de novo DNMT) rescued the expression of Oprm1 and Oprk1 mRNAs and their respective encoding mu-opioid receptor (MOR) and kappa-opioid receptor (KOR) proteins in the injured DRG. Blocking this increase also prevented the nerve injury-induced increase in DNA methylation in the promoter and 5'-untranslated region of the Oprm1 gene in the injured DRG, restored morphine or loperamide (a peripheral acting MOR preferring agonist) analgesic effects, and attenuated the development of their analgesic tolerance under neuropathic pain conditions. Mimicking this increase reduced the expression of Oprm1 and Oprk1 mRNAs and their coding MOR and KOR in DRG and augmented MOR-gated neurotransmitter release from the primary afferents. Mechanistically, DNMT3a regulation of Oprm1 gene expression required the methyl-CpG-binding protein 1, MBD1, as MBD1 knockout resulted in the decreased binding of DNMT3a to the Oprm1 gene promoter and blocked the DNMT3a-triggered repression of Oprm1 gene expression in DRG neurons. These data suggest that DNMT3a is required for nerve injury-induced and MBD1-mediated epigenetic silencing of the MOR and KOR in the injured DRG. DNMT3a inhibition may serve as a promising adjuvant therapy for opioid use in neuropathic pain management.

Figure 1

DNMT3a represses MOR and KOR expression in DRG. ShRNA: AAV5-Dnmt3a shRNA. Scr: AAV5-scambled shRNA. DNMT3a: AAV5-Dnmt3a. GFP: AAV5-Gfp. A. Expression of DNMT3a, MOR and KOR proteins in the ipsilateral L4/5 DRGs of naïve rats 5 weeks after DRG microinjection of PBS, ShRNA or Scr. Left: representative Western blots. Right: statistical summary of the densitometric analysis. n = 6 rats/group. *P < 0.05, **P < 0.01 vs the corresponding PBS-treated group. B and C. Expression of MOR and KOR proteins in the ipsilateral L5 DRG (B) and expression of Oprm1, Oprd1, and Oprk1 mRNAs in the ipsilateral (Ipsi) and contralateral (Con) L5 DRGs (C) from the treatment groups indicated 7 days after sham or SNL surgery. n = 6 rats/group. *P < 0.05, **P < 0.01 vs the corresponding sham plus PBS group. #P < 0.05, ##P < 0.01 vs the SNL plus Scr group. D and E. Expression of Dnmt3a (left in D), Oprm1 (E), Oprd1 (E) and Oprk1 (E) mRNAs in the ipsilateral (Ipsi) and contralateral (Con) L4/5 DRGs and expression of DNMT3a protein in the ipsilateral L4/5 DRGs (right in D) 5 weeks after DRG microinjection of GFP or DNMT3a. n = 5 rats/group. **P < 0.01 vs the corresponding GFP group. F. Expression of MOR and KOR proteins in the ipsilateral L4/5 DRGs 5 weeks after DRG microinjection of PBS, GFP or DNMT3a. n = 5 rats/group. *P < 0.05, **P < 0.01 vs the corresponding PBS-treated group. G. Co-localization of DNMT3a (red) with MOR (green) or KOR (green) in the DRG neurons of naïve rats. Scale bar: 40 μm.

Figure 2

DRG DNMT3a knockdown enhanced morphine analgesia and attenuated morphine tolerance in naïve and SNL rats. All of the following experiments were carried out 5 weeks after microinjection of AAV5-Dnmt3a shRNA (ShRNA), AAV5-scambled shRNA (Scr), or PBS into unilateral L4/5 DRGs (A–D) or L5 DRG (E and F). A. Morphine maximal possible analgesic effects (MPAE) measured 20 min after subcutaneous injection of 0.5 mg/kg morphine on the ipsilateral (Ipsi) and contralateral (Con) sides of naïve rats. n = 5 rats/group. **P < 0.01 vs the corresponding PBS-treated group. B and C. Morphine analgesic tolerance developed by subcutaneous injection of 10 mg/kg morphine twice daily for 6 days on the ipsilateral (B) and contralateral (C) sides of naïve rats. MPAEs were measured 20 min after subcutaneous injection of 5 mg/kg morphine on the mornings of days 1, 3, 5, and 7. n = 5 rats/group. **P < 0.01 vs the corresponding MPAE on day 1. ## P < 0.01 vs the PBS-treated group at the corresponding time points. D. Dose-response curve on the ipsilateral side of naïve rats from the PBS-, ShRNA- and Scr-treated groups on day 7 after subcutaneous injection of 10 mg/kg morphine twice daily for 6 days. E. Morphine analgesic tolerance developed by subcutaneous injection of 5 mg/kg morphine twice daily for 4 days starting at day 7 post-SNL on the ipsilateral side. MPAEs were measured 20 min after subcutaneous injection of 1.5 mg/kg morphine on the mornings of days 7, 9, and 11 post-SNL. n = 5 rats/group. *P < 0.05, **P < 0.01 vs the corresponding MPAE on day 7 post-SNL. #P < 0.05, ##P < 0.01 vs the PBS plus SNL group at the corresponding time points. F. Dose-response curve on the ipsilateral side on day 11 post-SNL after subcutaneous injection of 5 mg/kg morphine twice daily for 4 days starting at day 7 post-SNL from the PBS-, ShRNA- and Scr-injected rats. n = 5 rats/group.

Figure 3

DRG DNMT3a knockdown enhanced loperamide analgesia and alleviated loperamide tolerance in naïve and SNL rats. All of the following experiments were carried out 5 weeks after microinjection of AAV5-Dnmt3a shRNA (ShRNA), AAV5-scambled shRNA (Scr), or PBS into unilateral L5 DRG. A and B. Loperamide maximal possible analgesic effects (MPAE) measured 20 min after subcutaneous co-injection of 3 mg/kg loperamide (Lop) plus vehicle (Veh) or 5 mg/kg naltrexone (Naltr) on the ipsilateral (A) and contralateral (B) sides. n = 5 rats/group. **P < 0.01 vs the corresponding PBS group. C. Loperamide analgesic tolerance developed by subcutaneous injection of 10 mg/kg loperamide twice daily for 5 days on the ipsilateral side of naïve rats. MPAEs were measured 20 min after loperamide injection on the mornings of days 1, 3, and 5 post-injection. n = 5 rats/group. **P < 0.01 vs the corresponding MPAE on day 1 post-injection. #P < 0.05, ##P < 0.01 vs the PBS-treated group at the corresponding time points. D. Loperamide analgesic tolerance developed by subcutaneous injection of 0.75 mg/kg loperamide twice daily for 3 days starting on day 7 post-SNL on the ipsilateral side. n = 5 rats/group. *P < 0.05, **P < 0.01 vs the corresponding MPAE on day 7 post-SNL. #P < 0.05, ##P < 0.01 vs the PBS plus SNL group at the corresponding time points.

Figure 4

Overexpression of DNMT3a in DRG promotes neurotransmitter release from primary afferents. All of the following experiments were carried out 5 weeks after microinjection of AAV5-Dnmt3a (DNMT3a), or AAV5-GFP (Control) into unilateral L4/5 DRGs. A. Example traces of miniature EPSC (mEPSC) recorded in lamina II dorsal horn neurons before and after DAMGO application and its wash out. B and C. Effects of DAMGO (1 μM) on the frequency (B) and amplitude (C) of mEPSC in lamina II neurons from the AAV5-Dnmt3a-treated group (n = 19 neurons, 9 rats) and control group (n = 26 neurons, 12 rats). *P < 0.05, **P < 0.01 vs the control group before DAMGO application. D. Example traces of C-fiber input-evoked EPSC in lamina II neurons before and after DAMGO application and its wash out. E and F. Effects of DAMGO (1 μM) on the pair pulse ratio (E) and first peak amplitude of EPSC (F) in lamina II neurons from the AAV5-Dnmt3a-treated group (n = 16 neurons, 19 rats) and control group (n = 23 neurons, 20 rats). *P < 0.05, **P < 0.01 vs the control group before DAMGO application.

Figure 5

DNMT3a is required for the nerve injury-induced increase in Oprm1 DNA methylation in the injured DRG. A. Rabbit anti-DNMT3a immunoprecipitated region 1 (R1, −450/−288), region 5 (R5, +142/+312) and region 6 (R6, +238/+439), but not the remaining 4 regions (R2, −310/−143; R3, −164/−6; R4, −7/+166; R7: +391/+558), within the Oprm1 gene and the rabbit anti-MBD1 immunoprecipitated regions 1–3, but not regions 4–7, within the Oprm1 gene in rat DRGs. Input, total purified fragments. M, ladder marker. n = 3 repeats. B. The binding activity of DNMT3a to regions 1 (R1), 5 (R5) and 6 (R6) within the Oprm1 gene in the injured DRGs of rats on day 7 post-SNL or sham surgery. Left: representative binding between DNMT3a and R1. Right: statistical summary of the quantitative RT-PCR analysis in the binding activity. n = 9 rats/group. *P < 0.05, **P < 0.01 vs the corresponding sham group. C and D. DNA methylation levels at CpG sites of the Oprm1 gene in the injured DRG on day 7 post-SNL or sham surgery detected by bisulfite clone-sequencing assay (C) and bisulfite pyro-sequencing assay (D). n = 6 rats/group. *P < 0.05, **P < 0.01 vs the corresponding sham group. E. The percentage of methylation at the −365 CpG site of the Oprm1 gene in the injured DRG on day 7 post-SNL or sham surgery from the AAV5-GFP (GFP) plus sham group, the AAV5-GFP plus SNL group, AAV5-scrambled shRNA (Scr) plus SNL group, AAV5-Dnmt3a shRNA (ShRNA) plus SNL group, and AAV5-Dnmt3a ShRNA plus sham group. n = 6 rats/group. *P < 0.05 vs the AAV5-GFP plus sham group. #P < 0.05 vs the GFP plus SNL group.

Figure 6

MBD1 is required for DNMT3a repression of MOR expression. A. The binding activity of MDB1 to region 1 (R1) within the Oprm1 gene in the injured DRGs of rats on day 7 post-SNL or sham surgery. Left: representative binding between MBD1 and R1. Right: statistical summary of the quantitative RT-PCR analysis in the binding activity. Input, total purified fragments. M, ladder marker. n = 9 rats/group. *P < 0.05 vs the corresponding sham group. B. The binding activity of DNMT3a to region 1 (R1) within the Oprm1 gene in the DRGs from wild type (WT) and Mbd1 knockout (KO) mice. Left: representative bindings between DNMT3a and R1. Right: statistical summary of the quantitative RT-PCR analysis in the binding activity. Input, total purified fragments. M, ladder marker. n = 12 mice/group. *P < 0.05 vs the corresponding WT mice. C. Expression of Dnmt3a (left) and Oprm1 (right) mRNAs in DRG cultured neurons from WT and Mbd1 KO mice transfected with HSV-GFP (GFP) or HSV-Dnmt3a (DNMT3a). *P < 0.05, **P < 0.01 vs the corresponding HSV-GFP-transfected neurons from WT mice. ##P < 0.01 vs the corresponding HSV-Dnmt3a-transfected neurons from WT mice. D–F. Expression of Mbd1 (D), Dnmt3a (E), and Oprm1 (F) mRNAs in WT DRG cultured neurons co-transfected with HSV-GFP plus AAV5-GFP (control), AAV5-Mbd1 (MBD1) plus AAV5-GFP, HSV-Dnmt3a (DNMT3a) plus AAV5-GFP, or AAV5-Mbd1 plus HSV-Dnmt3a. *P < 0.05, **P < 0.01 vs the control group.