Roles of peripheral terminals of transient receptor potential vanilloid-1 containing sensory fibers in spinal cord stimulation-induced peripheral vasodilation

Background: Spinal cord stimulation (SCS) is used to relieve ischemic pain and improve peripheral blood flow in selected patients with peripheral arterial diseases. Our previous studies show that antidromic activation of transient receptor potential vanilloid-1 (TRPV1) containing sensory fibers importantly contributes to SCS-induced vasodilation. Objectives: To determine whether peripheral terminals of TRPV1 containing sensory fibers produces vasodilation that depends upon the release of calcitonin gene-related peptide (CGRP) and nitric oxide (NO) during SCS. Methods: A unipolar ball electrode was placed on the left dorsal column at lumbar spinal cord segments 2-3 in sodium pentobarbital anesthetized, paralyzed and ventilated rats. Cutaneous blood flow from left and right hindpaws was recorded with laser Doppler flow perfusion monitors. SCS was applied through a ball electrode at 30%, 60%, 90% and 300% of motor threshold. Resiniferatoxin (RTX; 2 mu g/ml, 100 mu l), an ultra potent analog of capsaicin, was injected locally into the left hindpaw to functionally inactivate TRPV-1 containing sensory terminals. In another set of experiments, CGRP(8-37), an antagonist of the CGRP-1 receptor, was injected at 0.06, 012 or 0.6 mg/100 mu l into the left hindpaw to block CGRP responses; N-omega-nitro-L-arginine methyl ester (L-NAME), a nonselective nitric-oxide synthase (NOS) inhibitor, was injected at 0.02 or 0.2 mg/100 mu l into the left hindpaw to block nitric oxide synthesis; (4S)-N-(4-Amino-5[aminoethyl] aminopentyl)-N'-nitroguanidine, TFA, a neuronal NOS inhibitor, was injected at 0.02 or 0.1 mg/100 mu l into the left hindpaw to block neuronal nitric oxide synthesis. Results: SCS at all intensities produced vasodilation in the left hindpaw, but not in the right. RTX administration attenuated SCS-induced vasodilation at all intensities in the left hindpaw (P < 0.05, n = 7) compared with responses before RTX. CGRP(8-37) administration attenuated SCS-induced vasodilation in the left hindpaw in a dose dependent manner (linear regression, P < 0.05) compared with responses before CGRP(8-37). In addition, L-NAME at a high dose, but not (4S)-N-(4-Amino-5[aminoethyl]aminopentyl)-N'-nitroguanidine, TFA, decreased SCS-induced vasodilation (P < 0.05, n = 5). Conclusion: While TRPV1, CGRP and NO are known to be localized in the same nerve terminals, our data indicate that SCS-induced vasodilation depends on CGRP release, but not NO release. NO, released from endothelial cells, may be associated with vascular smooth muscle relaxation and peripheral blood flow increase in response to SCS. (c) 2007 Elsevier B.V. All rights reserved.