Inflammation reduces mechanical thresholds in a population of transient receptor potential channel A1-expressing nociceptors in the rat

Abstract

Inflammatory hypersensitivity is characterized by behavioural reductions in withdrawal thresholds to noxious stimuli. Although cutaneous primary afferent neurones are known to have lowered thermal thresholds in inflammation, whether their mechanical thresholds are altered remains controversial. The transient receptor potential channel A1 (TRPA1) is a receptor localized to putative nociceptive neurones and is implicated in mechanical and thermal nociception. Herein, we examined changes in the properties of single primary afferents in normal and acutely inflamed rats and determined whether specific nociceptive properties, particularly mechanical thresholds, are altered in the subpopulation of afferents that responded to the TRPA1 agonist cinnamaldehyde (TRPA1-positive afferents). TRPA1-positive afferents in normal animals belonged to the mechanonociceptive populations, many of which also responded to heat or capsaicin but only a few of which responded to cold. In acute inflammation, a greater proportion of afferents responded to cinnamaldehyde and an increased proportion of dorsal root ganglion neurones expressed TRPA1 protein. Functionally, in inflammation, TRPA1-positive afferents showed significantly reduced mechanical thresholds and enhanced activity to agonist stimulation. Inflammation altered thermal thresholds in both TRPA1-positive and TRPA1-negative afferents. Our data show that a subset of afferents is sensitized to mechanical stimulation by inflammation and that these afferents are defined by expression of TRPA1.

Fig. 1

Activation thresholds in primary afferents in normal and inflamed animals. (A) Mechanical thresholds of the TRPA1-positive and TRPA1-negative afferents in normal (n= 55 units, 38 animals) and inflamed (n= 17 units, 8 animals) animals. Thresholds were significantly decreased in inflammation only in the TRPA1-positive population (19 normal, 10 inflamed). [Box and whisker plots show the 25th and 75th centiles (box), median (central line) and range (whiskers)] (P = 0.01). NB: Log scale on y-axis. (B) The heat thresholds for both TRPA1-positive (6 normal, 11 inflamed) and TRPA1-negative (6 normal, 4 inflamed) polymodal nociceptors are significantly lowered in inflammation (mean ± SEM; P = 0.005). (C) Cold thresholds were increased in C-fibre polymodal nociceptors in the TRPA1-positive and TRPA1-negative populations (2 to 3 normal, 2 to 3 inflamed; mean ± SD). (D) A representative digitized data trace of evoked activity in a single polymodal unit innervating inflamed skin. The dashes in the lowest trace denote the time of the mechanical stimuli, the intensity of which is given next to the elicited response (action potentials μV, top trace). Repeated stimulation with 1 g von Frey hairs (825–840 s) elicited no responses, unlike 1.4 and 2 g hairs (∼815 and 790 s, respectively). The middle trace shows the interface temperature of the Peltier device, which increases from 30 to 45 °C (beginning at ∼875 s), evoking a response at ∼35 °C. NS, not significant. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 2

Primary afferent characteristics in CFA-induced inflammation. (A) CFA significantly enhanced cinnamaldehyde-evoked activity (number of action potentials) in afferents innervating inflamed skin (n= 42) more than in afferents innervating normal skin (n= 30; P < 0.01) (medians and interquartile range). (B) Functional TRPA1 responses could be evoked in 36% of fibres in normal animals (n= 59) and this was increased to ∼60% (n= 20) after CFA inflammation (P < 0.001). (C) Ankle joint diameters in normal and inflamed rats at 3 days after CFA injection. CFA injection caused significant swelling of the joints at this time (n= 4; P < 0.001). **P < 0.01, ***P < 0.001.

Fig. 3

Pharmacological characterization of agonist and antagonist responses at TRPA1-positive primary afferents. (A) Response to three sequential doses of cinnamaldehyde (80 mm) does not desensitize with an interstimulus interval of 5 min (n= 27 units from naive and inflamed rats). (B) Ruthenium red significantly attenuated intra-arterial 80 mm cinnamaldehyde-evoked activity (mean number of action potentials and SEMs, P < 0.01, n= 5 units). (C) Capsazepine (500 μm) did not affect activity evoked by intra-arterial 80 mm cinnamaldehyde (means and SEMs, n= 6 units). NS, not significant. **P < 0.01.

Fig. 4

The number of DRG neurones showing TRPA1-like immunoreactivity is increased in acute inflammation. (A and B) TRPA1-like immunoreactivity was evident in small but not large neuronal cell bodies (black arrowhead) in sections from rat L4 DRG from control (A) and day 3 CFA (B) animals. Scale bar, 50 μm. (C) Omission of primary antibody resulted in loss of staining in neurones but not in satellite cells (white arrowhead) or DRG nuclei (black arrow), suggesting that this nuclear staining was non-specific. (D) The proportion of TRPA1-immunoreactive neurones increased in rat L4 DRG ganglia 3 days after CFA injection (P < 0.0.01, n= 4–6 per group). (E) Knee diameters were also significantly increased by CFA treatment 3 days after CFA injection (P < 0.001, n= 6 per group). ***P < 0.001.