Diroximel Fumarate Acts Through Nrf2 to Attenuate Methylglyoxal-Induced Nociception in Mice and Decrease ISR Activation in DRG Neurons

Abstract

Diabetic neuropathic pain is associated with elevated plasma levels of methylglyoxal (MGO). MGO is a metabolite of glycolysis that causes pain hypersensitivity in mice by stimulating the phosphorylation of eukaryotic initiation factor 2α (p-eIF2α) and subsequently activating the integrated stress response (ISR). We first established that Zucker diabetic fatty rats have enhanced MGO signaling, engage ISR, and develop pain hypersensitivity. Since nuclear factor erythroid 2-related factor 2 (Nrf2) regulates the expression of antioxidant proteins that neutralize MGO, we hypothesized that fumarates, like diroximel fumarate (DRF), will stimulate Nrf2 signaling, and prevent MGO-induced ISR and pain hypersensitivity. DRF (100 mg/kg) treated animals were protected from developing MGO (20 ng) induced mechanical and cold hypersensitivity. Mechanistically, DRF treatment protected against MGO-induced increase in p-eIF2α levels in the sciatic nerve and reduced loss of intraepidermal nerve fiber density. Using Nrf2 knockout mice, we demonstrate that Nrf2 is necessary for the antinociceptive effects of DRF. Cotreatment of MGO (1 µmol/L) with monomethyl fumarate (10, 20, and 50 µmol/L), the active metabolite of DRF, prevented ISR in both mouse and human dorsal root ganglia neurons. Our data show that targeting Nrf2 with DRF is a strategy to potentially alleviate pain associated with elevated MGO levels.

Graphical abstract

Figure 1

ZDF rats develop hyperglycemia and pain hypersensitivity that correlate with increased CEL and enhanced ISR in DRG. A: Male ZDF rats carry a mutation in the leptin receptor (fa/fa) and develop hyperglycemia, obesity, and neuropathy when fed a high-fat diet. We tested ZDF rats for mechanical, thermal, and cold hypersensitivity at 12, 14, and 16 weeks of age. B and C: ZDF rats have elevated blood glucose levels and increased weight when compared with lean (fa/+) controls. D–F: Similarly, ZDF rats develop thermal, cold, and mechanical hypersensitivity by 16 weeks of age. G: Since MGO has been linked to neuropathic pain in diabetes, we quantified the levels of CEL, an MGO-specific AGE, in DRG of ZDF and lean rats. We found that CEL levels were significantly increased in DRG of 16-week-old ZDF rats. H–K: At the same time point, phosphorylation of eIF2α was also increased in ZDF rat DRG. These experiments link hyperglycemia, MGO signaling, and ISR in an animal model. For B–F, significance was calculated using repeated measures two-way ANOVA followed by Sidak multiple comparison test. For I–K, statistical significance was determined using unpaired t test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 2

DRF pretreatment protects against MGO-induced pain hypersensitivity. A: MGO stimulates ISR and promotes the phosphorylation of eIF2α. DRF is a prodrug and is metabolized to MMF, which activates the Nrf2 antioxidant response neutralizing MGO. B: Male (n = 8 per condition) and female (n = 8 per condition) mice were treated with DRF (60 or 100 mg/kg) for 5 days. MGO (20 ng) was administered in an intraplantar injection on the third day, 1 h after oral DRF administration for the day. Mechanical and cold sensitivity was assessed using von Frey and acetone tests, respectively. Acetone test was performed an hour after von Frey testing was complete. We observed that male (C) and female (D) mice had reduced von Frey thresholds in response to MGO injection on day 3. Daily treatment with DRF prevented MGO-induced mechanical hypersensitivity following MGO injection on day 3. E and F: Using area under the curve of von Frey thresholds, we determined that DRF at a dose of 100 mg/kg was effective at preventing MGO-induced tactile hypersensitivity in males, while both 60 mg/kg and 100 mg/kg doses were effective in females. MGO produced cold hypersensitivity in male (G) and female (H) mice, which was abrogated in DRF-treated mice at both doses. For C, D, G, and H, significance was calculated with repeated measures two-way ANOVA with Tukey multiple comparison test. *, &, and #: P < 0.05; ** and ##: P < 0.01; *** and &&&: P < 0.001; ****P < 0.0001. *indicates comparison with veh-only, & indicates comparison with MGO + DRF-60, and # indicates comparison with MGO + DRF-100. For E and F, significance was calculated with a one-way ANOVA with Brown-Forsythe correction. *P < 0.05, **P < 0.01.

Figure 3

DRF prevents MGO-induced ISR in the sciatic nerve and the loss of IENF. A: Male (n = 3) and female (n = 3) mice were treated with DRF for 2 days prior to a single MGO (20 ng) injection. The following day, 24 h later, sciatic nerves and hind paw biopsies were obtained. B and C: MGO injection in the hind paw increased phosphorylation of eIF2α in the sciatic nerve, which was prevented in animals treated with DRF (100 mg/kg). D and E: Moreover, DRF treatment increased protein levels of Gclm, a direct transcriptional target of Nrf2 and a critical node in glutathione synthesis. F–I: Hind paw skin biopsies were stained for PGP9.5, and IENF density was quantified as number of dermal-epidermal crossings per millimeter of tissue. We found that DRF treatment protected against MGO-induced reduction in IENF density. Statistical significance was determined using one-way ANOVA. *P < 0.05, **P < 0.01.

Figure 4

MGO induces mechanical pain hypersensitivity in Nrf2KO mice. Wild-type (A) and Nrf2KO (B) animals were injected with MGO (20 ng) or vehicle (saline) in the left paw. Mechanical sensitivity was assessed using the von Frey test for up to 20 days after injection. MGO (20 ng) injection produced mechanical hypersensitivity in both wild-type and Nrf2KO mice. C: Area under the curve of von Frey thresholds was used to directly compare the response of wild-type and Nrf2KO animals to MGO injection. Each condition included four male and four female mice for a total of eight animals. For A and B, statistical analysis was performed with repeated measures two-way ANOVA followed by Sidak multiple comparison test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. In C, two-way ANOVA with Tukey test was used. **P < 0.01, ***P < 0.001.

Figure 5

DRF treatment fails to prevent mechanical hypersensitivity in Nrf2KO mice. Wild-type and Nrf2KO animals were treated with DRF at 100 mg/kg based on the paradigm outlined in Fig. 2A. MGO (20 ng) was injected into the paw on day 3 since the start of oral DRF treatment in wild-type (n = 8, four males and four females) and Nrf2KO (n = 8, four males and four females) littermates. Nrf2KO animals developed mechanical hypersensitivity compared with the wild-type animals, suggesting that DRF treatment was unable to prevent MGO-induced hypersensitivity in Nrf2KO animals. A repeated measures two-way ANOVA was used followed by Sidak post hoc test. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 6

MMF prevents MGO-evoked increases in p-eIF2α levels in mouse DRG neurons. A–C: Representative images of immunohistochemistry of mouse DRG neurons cultured for 24 h and subsequently treated with vehicle only, MGO (1 µmol/L), or MGO (1 µmol/L) plus MMF (10, 20, or 50 µmol/L). D: We observed a significant increase in the levels of p-eIF2α when cells were treated with MGO as compared with vehicle-treated cells. MMF treatment with MGO prevented elevation of p-eIF2a, particularly at 20 µmol/L and 50 µmol/L concentrations. Neurons were identified by their expression of β3 tubulin. Vehicle-only (n = 18), MGO (n = 37), MGO + 10 µmol/L MMF (n = 20), MGO + 20 µmol/L MMF (n = 16), and MGO + 50 µmol/L MMF (n = 32). One-way ANOVA was used to calculate significance. *P < 0.05, **P < 0.01.

Figure 7

MMF prevents MGO-evoked increases in p-eIF2α levels in human DRG neurons. A–C: Representative images of human DRG neurons obtained from organ donors treated with vehicle, MGO (1 µmol/L), and MGO + MMF. D: MGO (1 µmol/L) increases p-eIF2α in human DRG neurons, which is prevented with 20 µmol/L and 50 µmol/L MMF cotreatment. Vehicle only (n = 170), MGO (n = 152), MGO + 10 µmol/L MMF (n = 125), MGO + 20 µmol/L MMF (n = 176), and MGO + 50 µmol/L MMF (n = 111). One-way ANOVA was used to calculate statistical significance. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.