Intrathecal rosiglitazone acts at peroxisome proliferator-activated receptor-gamma to rapidly inhibit neuropathic pain in rats

Abstract

In this report, we demonstrate the transcription, expression, and DNA-binding properties of the peroxisome proliferator-activated receptor (PPAR)-gamma subtype of the peroxisome proliferator-activated nuclear receptor family to the spinal cord with real-time PCR, Western blot, and electrophoretic mobility shift assay. To test the hypothesis that activation of spinal PPAR-gamma decreases nerve injury-induced allodynia, we intrathecally administered PPAR-gamma agonists and/or antagonists in rats after transection of the tibial and common peroneal branches of the sciatic nerve. Single injection of either a natural (15-deoxy-prostaglandin J2, 15d-PGJ2) or synthetic (rosiglitazone) PPAR-gamma agonist dose-dependently decreased mechanical and cold hypersensitivity. These effects were maximal at a dose of 100 microg and peaked at approximately 60 minutes after injection, a rapid time course suggestive of transcription-independent mechanisms of action. Concurrent administration of a PPAR-gamma antagonist (bisphenol A diglycidyl ether, BADGE) reversed the effects of 15d-PGJ2 and rosiglitazone, further indicating a receptor-mediated effect. In animals without nerve injury, rosiglitazone did not alter motor coordination, von Frey threshold, or withdrawal response to a cool stimulus. Intraperitoneal and intracerebroventricular administration of PPAR-gamma agonists (100 microg) did not decrease mechanical and cold hypersensitivity, arguing against effects subsequent to diffusion from the intrathecal space. We conclude that ligand-induced activation of spinal PPAR-gamma rapidly reverses nerve injury-induced mechanical allodynia. New or currently available drugs targeted at spinal PPAR-gamma may yield important therapeutic effects for the management of neuropathic pain.

Perspective: PPAR-gamma receptor agonists such as rosiglitazone and pioglitazone are approved as insulin sensitizers by the United States Food and Drug Administration. We demonstrate PPAR-gamma expression in the spinal cord and report that activation of these receptors inhibits allodynia. BBB-permeant PPAR-gamma agonists may yield important therapeutic effects for the management of neuropathic pain.

Figure 1. PPARγ expression in the rat dorsal horn

(A) Expression of PPARγmRNA in spinal cord. Positive controls include liver, brain, and spleen. (B) Representative Western blots of PPARγ immunoreactivity. This gel illustrates the results of two rats from an enriched nuclear fraction of lumbar dorsal spinal cord (lanes 1-2) or liver (lanes 3-4). MW markers (kD) are indicated on the left. A band of expected size (~67 kD) for PPARγ was present. The higher MW band (=~135 kD) was no longer apparent with the addition of β-mercaptoethanol (not shown). These experiments were repeated several times with similar results. (C) Specificity of protein-binding to DNA in spinal cord (lanes 1-4) and liver (lanes 5-8). Lanes 1-5 show the location of the biotinylated probe alone (free probe, “FP”). Lanes 2-3 and 6-7 illustrate examples of shift bands (S), indicating that protein is bound to the probe, thus reducing its mobility. Lanes 4-8 show that addition of a competitor untagged oligonucleotide (COMP), in high molar excess relative to the probe, leads to disappearance of S. This indicates that all transcriptional complexes are bound to unlabeled probe rather than labeled probe, and demonstrate specificity of the nuclear protein complex (likely a PPARγ/RXR heterodimer) to the oligonucleotide probe. (D) Specificity of DNA-binding to PPARγ in spinal cord. To further demonstrate specificity of the consensus PPARresponse element (probe) to PPARγ heterodimer complexes, a supershift assay was performed. Lane 3 repeats our finding that excess unlabeled probe competes with labeled probe, thus eliminating the S band. Lanes 4-6 show that 2 μg of PPARγ-specific antibodies (sc-6284x and sc-7273x) further decrease electrophoretic mobility, leading to the appearance of supershift (SS) bands. The formation of this antibody/protein/DNA complex indicates that a PPARγ-heterodimer binds to the DNA probe.

Figure 2. The naturally-occurring PPARγ agonist, 15d-PGJ2, dose-dependently increases mechanical threshold

Panel A illustrates the time course of mechanical threshold following intrathecal injection of saline or 15d-PGJ2 in rats, 7d after spared nerve injury (SNI). Panel B illustrates the change in mechanical threshold averaged from 30 to 90 min after injection (Mean Δ in Threshold). Panel C illustrates the log dose-response curve at 60 min after injection. The 100 μg and 200 μg doses significantly increased mechanical threshold at the 60 and 90 min time points. Values represent mean ± SEM, n=5-7. *P < 0.05 vs saline control.

Figure 3. The PPARγ antagonist, bisphenol A diglycidyl ether (BADGE), dose-dependently reverses the antiallodynic effect of 15d-PGJ2

Panel A illustrates the time course of mechanical threshold following intrathecal co-injection of 15d-PGJ2 (100 μg) and BADGE, 7d after spared nerve injury (SNI). Panel B illustrates the change in mechanical threshold averaged from 30 to 90 min after injection (Mean Δ in Threshold), and indicates that thresholds in the 5, 25, and 100 μg BADGE+15d-PGJ2 groups were different from 15d-PGJ2 controls. Panel C illustrates the log dose-response curve at 60 min after injection. Vehicle for BADGE was DMSO. Values represent mean ± SEM, n=4-8. *P < 0.05.

Figure 4. The PPARγ agonist, rosiglitazone, dose-dependently inhibits behavioral hypersensitivity

Panel A illustrates the time course of mechanical threshold following intrathecal injection of saline or drug in rats, 7d after spared nerve injury (SNI). Panel B illustrates the time course of cold response following intrathecal injection of saline or drug in rats, 7d after spared nerve injury (SNI). The 100 μg dose significantly increased mechanical threshold at the 60 and 90 min time points, and significantly decreased cold response at the 60, 90, and 120 min time points. The 50 μg dose significantly decreased cold response at the 60 and 90 min time points. Vehicle represents 5μL 30% DMSO +10μL saline flush Values represent mean ± SEM, n=6. *P < 0.05 vs saline control.

Figure 5. BADGE dose-dependently reversed the antiallodynic effect of rosiglitazone

Panel A illustrates the time course of mechanical threshold following intrathecal co-injection of drugs, 7d after spared nerve injury (SNI). Panel B illustrates the change in mechanical threshold at 90 min after injection. Panel C illustrates the time course of cold response following intrathecal co-injection of drugs, 7d after spared nerve injury (SNI). Panel D illustrates the change in cold response at 90 min after injection. A 100 μg dose of rosiglitazone (Rosi) was used. The vehicle (DMSO/saline) and Rosi + 25 μg BADGE groups were significantly different from Rosi at the 90 min time point for mechanical allodynia. The vehicle and Rosi + 100 μg BADGE groups were significantly different from Rosi at the 90 min time point for cold response. Vehicle represents 5μL 30% DMSO + 5μL 100% DMSO + 10μL saline flush Values represent mean ± SEM, n=4-8. *P < 0.05.

Figure 6. Rosiglitazone does not affect mechanical threshold, cold response, IR latency, or motor coordination in animals without nerve injury

Panels A, B, and C illustrates the time course of mechanical threshold, cold response, and IR latency, respectively, following intrathecal injection of drug, 7d after sham SNI surgery. The intrathecal administration of rosiglitazone did not change mechanical threshold, cold response, or IR latency in sham animals. Panel D illustrates the effects of saline/DMSO or 100 μg, 300 μg, or 1000 μg rosiglitazone, administered intrathecally, on motor coordination. Motor coordination is measured by duration spent walking on rotarod. The administration of intrathecal rosiglitazone at anti-allodynic doses does not change motor coordination, when compared to baseline measures (P > 0.05). Vehicle represents 5μL 30% DMSO + 10μL saline flush.

Figure 7. Supraspinal or peripheral administration of PPARγ agonists (at the intrathecal doses used above) did not reduce mechanical or cold hypersensitivity

An intracerebroventricular (i.c.v.) or intraperitoneal (i.p.) injection of saline, 15d-PGJ2, rosiglitazone, and/or morphine was administered to rats, 7d after spared nerve injury (SNI). Panels A and C illustrate the time course of mechanical threshold. Panels B and D illustrate the time course of cold response. In contrast to morphine, a positive control that reduced mechanical and cold allodynia (Panels A-B, P<0.05), neither 15d-PGJ2, rosiglitazone, nor vehicle significantly changed mechanical threshold or cold response. Values represent mean ± SEM, n=4.