Distinct subsets of unmyelinated primary sensory fibers mediate behavioral responses to noxious thermal and mechanical stimuli

Abstract

Behavioral responses to painful stimuli require peripheral sensory neurons called nociceptors. Electrophysiological studies show that most C-fiber nociceptors are polymodal (i.e., respond to multiple noxious stimulus modalities, such as mechanical and thermal); nevertheless, these stimuli are perceived as distinct. Therefore, it is believed that discrimination among these modalities only occurs at spinal or supraspinal levels of processing. Here, we provide evidence to the contrary. Genetic ablation in adulthood of unmyelinated sensory neurons expressing the G protein-coupled receptor Mrgprd reduces behavioral sensitivity to noxious mechanical stimuli but not to heat or cold stimuli. Conversely, pharmacological ablation of the central branches of TRPV1(+) nociceptors, which constitute a nonoverlapping population, selectively abolishes noxious heat pain sensitivity. Combined elimination of both populations yielded an additive phenotype with no additional behavioral deficits, ruling out a redundant contribution of these populations to heat and mechanical pain sensitivity. This double-dissociation suggests that the brain can distinguish different noxious stimulus modalities from the earliest stages of sensory processing.

Fig. 1.

Specific ablation of Mrgprd⁺ neurons in mice. (A) Mrgprd^DTR targeting construct. (B–D) DTR expression in the absence of DTX does not impair the survival of Mrgprd⁺ neurons, determined using an independent farnesylated enhanced green fluorescent protein (EGFPf)-expressing allele in Mrgprd^EGFPf/DTR transheterozygous mice. Sections of the DRG (E–H), spinal cord (I–L), and glabrous skin (M and N) from DTX-treated Mrgprd^EGFPf/+ mice (E, G, I, K, and M) and Mrgprd^EGFPf/DTR mice (F, H, J, L, and N) were stained for the indicated markers. (F, H, J, L, and N). Note selective loss of Mrgprd⁺ cells (green) and fibers. See Table S1 for quantification. (Scale bar in D, F, H, J, L: 50 μm; scale bar in N: 25 μm.)

Fig. 2.

Mice lacking Mrgprd⁺ neurons exhibit selective deficits in mechanosensitivity. (A) Mechanical thresholds determined with the von Frey test before and after DTX injection (Student's t test: *, P < 0.05, **, P < 0.01). (B) von Frey test before (BL) and after CFA injection (two-way ANOVA with Bonferroni posttests: **, P < 0.01, ***, P < 0.001). (C) Normalized mechanical threshold at post-CFA day 1 relative to pre-CFA baseline (Mann-Whitney U test: *, P < 0.05). (D) Tail immersion test. (E) Hot plate test. The 55 °C data used a separate cohort of mice. (F) Radiant heat test before (BL) and after CFA injection. Data represent means ± SEM; n = 6–10 for all tests.

Fig. 3.

Intrathecal (i.t.) capsaicin treatment selectively ablates the central terminals of TRPV1⁺ nociceptors. (A–D) Immunostaining in lumbar dorsal horn 16 days after i.t. injection of vehicle (Left) or capsaicin (Right). Staining for TRPV1 (A) and CGRP (B) was significantly reduced by capsaicin. Staining for IB4 (C) and TRPV2 (D) was unchanged. (E) Density of staining: mean ± SEM (*, P < 0.0005, Student's t test). For TRPV1 staining, n = 8 for capsaicin-treated mice, n = 10 for vehicle-treated mice, and n = 4 per group for all others. (F) Double-labeling for TRPV1 (red) and Mrgprd-GFP (green) in vehicle- (Left) and capsaicin-treated (Right) mice. (Scale bar: 200 μm.)

Fig. 4.

Capsaicin (cap)-treated mice exhibit a complete and selective loss of heat pain sensitivity. Latency to lick or jump on 55 °C hot plate 1 day after intrathecal (i.t.) injection of cap or vehicle (veh) (P̂ < 0.0001, Student's t test) (A) and at weekly intervals after injection of cap (n = 4) (B). (C) CFA-induced heat hypersensitivity (**, P < 0.01, P̂ < 0.0001, two-way repeated measures ANOVA with Bonferroni posttest). (D) Temperature preference assay (**, P < 0.01, Student's t test). (E) Fos induced by hind paw immersion in 55 °C water bath. (F) Fos⁺ neurons in laminae I/II and V per 40-μm section (**, P < 0.01, Student's t test; n = 3 per group). (G) Mechanical threshold (von Frey test). (H) CFA-induced mechanical hypersensitivity (P̂ < 0.0001, two-way repeated measures ANOVA with Bonferroni posttest). Data represent mean ± SEM. Unless otherwise noted, n = 8–10.

Fig. 5.

TRPV1⁺ neurons do not contribute to residual mechanical pain sensitivity in mice lacking Mrgprd⁺ neurons. Mechanical (von Frey test) (A) and heat (radiant heat) (B) responsiveness in DTX-treated WT or Mrgprd^DTR mice before and after intrathecal injection of capsaicin (n = 3 for WT/DTX and n = 6 for Mrgprd^DTR/DTX; **, P < 0.001, ***, P < 0.0001; n.s., not significant; Student's t test). (C) Schematic illustrating modality-specific contributions of Mrgprd⁺ and TRPV1⁺ fibers to behavior and their termination zones in the epidermis and spinal cord (14). The second-order targets of Mrgprd⁺ afferents are shown projecting to deeper laminae before engaging ascending pathways (1). SG, stratum granulosum; SS, stratum spinosum; I, lamina I; IIo, lamina II outer; IIi, lamina II inner.