Plasticity in intact A delta- and C-fibers contributes to cold hypersensitivity in neuropathic rats

Abstract

Cold hypersensitivity is a common sensory abnormality accompanying peripheral neuropathies and is difficult to treat. Progress has been made in understanding peripheral mechanisms underlying neuropathic pain but little is known concerning peripheral mechanisms of cold hypersensitivity. The aim of this study was to analyze the contribution of uninjured primary afferents to the cold hypersensitivity that develops in neuropathic rats. Rats with a lumbar 5 (L5) and L6 spinal nerve ligation (SNL, Chung model) but not sham, developed mechanical allodynia, evidenced by decreased paw withdrawal thresholds and increased magnitude of response to von Frey stimulation. Cold hypersensitivity also developed in SNL but not sham rats, evidenced by enhanced nociceptive behaviors induced by placement on a cold plate (6 degrees C) or application of icilin (a transient receptor potential M8 (TRPM8)/transient receptor potential A1 (TRPA1) receptor agonist) to nerve-injured hind paws. Single fiber recordings demonstrated that the mean conduction velocities of intact L4 cutaneous A delta- and C-fibers were not different between naive and SNL rats; however, mechanical thresholds of the A delta- but not the C-fibers were significantly decreased in SNL compared with naive. There was a higher prevalence of C-mechanoheat-cold (CMHC) fibers in SNL compared with naive, but the overall percentage of cold-sensitive C-fibers was not significantly increased compared with naive. This was in contrast to the numerous changes in A delta-fibers: the percentage of L4 cold sensitive A delta-, but not C-fibers, was significantly increased, the percentage of L4 icilin-sensitive A delta-, but not C-fibers, was significantly increased, the icilin-induced activity of L4 A delta-, but not C-fibers, was significantly increased. Icilin-induced activity was blocked by the TRPA1 antagonist Ruthenium Red. The results indicate plasticity in both A delta- and C-uninjured fibers, but A delta fibers appear to provide a major contribution to cold hypersensitivity in neuropathic rats.

Fig. 1

Traces showing standard heat, mechanical and cold stimulation. (A) Action potentials (upper trace) evoked from an Aδ-fiber by the standard 10 s heat ramp starting from 34 and rising to 51 °C on the epidermal surface (lower trace, 51 °C on the epidermal side=47 °C on the corium side). (B) Action potentials (upper trace) evoked from an Aδ-fiber by the standard 20 s mechanical ramp (from 0 to 250 mN) stimulation (lower trace). (C) Action potentials (upper trace) evoked from an Aδ-fiber by the 30 s cold (from 34 to ∼0 °C) stimulation (lower trace).

Fig. 2

Mechanical allodynia in SNL rats. (A) Compared with pre-surgery, the post-surgery mean mechanical threshold of the nerve-injured paw was significantly decreased but there was no change in threshold in sham rats. (B) The mean % maximum response frequency to von Frey filament stimulation (12.9 mN, 22.8 mN and 43.2 mN) was significantly increased in SNL rats comparing pre- to post-surgery (* P<0.05 compared with pre-surgery, Wilcoxon signed rank test).

Fig. 3

Cold plate-induced pain behaviors in SNL rats. (A) Compared with pre-surgery, post-surgery flinching was significantly increased for the nerve-injured paw of SNL rats when placed on a 6 °C cold plate. In contrast, sham rats showed no change in flinching behavior post-surgery and were no different than naive rats. (B) Placement on a 24 °C plate resulted in little flinching behavior in any group. (* P<0.05 compared to sham and naive, Kruskal Wallis test).

Fig. 4

Icilin-induced pain behaviors in SNL rats. Application of 1 mM icilin to the plantar surface of the SNL hind paw induced significant amounts of flinching behavior compared with naïve and sham rats and compared with vehicle in SNL rats (* P<0.05, indicates significant difference from all other groups, one-way ANOVA). (Rats were tested at 7 days post-surgery.)

Fig. 5

Conduction velocities of Aδ- and C-fibers in naive and SNL rats. There appeared to be no difference in the distribution of conduction velocities for Aδ- and C-fibers comparing naive to SNL rats. Furthermore, the C-fiber conduction velocities in SNL rats (median=0.75, 25th percentile=0.64, 75th percentile=0.99) were similar to naive (median=0.70; 25th percentile=0.58, 75th percentile=0.91) and the Aδ-fiber conduction velocities in SNL rats (median=2.77 m/s, 25th percentile=2.12 m/s, 75th percentile=3.78 m/s) were similar to naive (median=2.71 m/s, 25th percentile=2.12 m/s, 75th percentile=3.71 m/s).

Fig. 6

Mechanical threshold of Aδ- and C-fibers in naive and SNL rats. The mechanical threshold of Aδ-fibers was significantly lower in SNL compared with naive rats, but there was no difference in C-fiber mechanical sensitivity (** P<0.01, Wilcoxon signed rank test).

Fig. 7

Icilin-induced dose response for Aδ- and C-fibers. (A) Application of an ascending series of icilin concentrations (0.01, 0.1, 1.0, 10, 100 μM) to the receptive field of C-fibers (A) and Aδ-fibers (B) produced a dose-dependent increase in activity in both fiber populations (⁺ P<0.05, indicates significant difference from vehicle for that group, Friedman's ANOVA). Icilin-induced activity in C-fibers was similar in naive and SNL rats. In contrast, icilin-induced activity in Aδ-fibers was significantly increased at 10 and 100 μM in SNL compared with naive rats (* P<0.05, indicates significant difference from naïve at same concentration, Mann Whitney U test).

Fig. 8

Icilin-induced activity in Aδ-fibers. Examples of nociceptor activity evoked in SNL and naive hind paws following exposure of the receptive field to icilin (10 μM) for 2 min.

Fig. 9

Icilin-induced cytosolic Ca²⁺ responses in L4 DRG cells. Responses to icilin are reflected as changes in FURA-2 F340/F380 nm ratio. During the 2 min application of icilin, the cytosolic Ca²⁺ change was significantly larger in SNL compared with naive (inset, *** P<0.001, Mann Whitney U test).