The discharge of single afferent units recorded from filaments of the
mixed ciliary nerve of anaesthetised cats was used to study the effect
s of CO2, capsaicin and the capsaicin antagonist, capsazepine, on sens
ory nerve terminals in the cornea. Units were selected on the basis of
their sensitivity to CO2 (98.5% applied to the cornea for 30 s in a m
oist gas stream). Of these units, about 50% (18/38) also responded to
capsaicin (0.1 mu M applied as droplet and washed off after 20 a), wit
h a discharge of similar magnitude to that produced by CO2. The other
CO2-sensitive units (20/38) did not respond to capsaicin. Capsaicin-se
nsitive units responded more rapidly to CO2 (mean latency to first spi
ke 0.7 +/- 0.2 s) than capsaicin-insensitive units (mean latency to fi
rst spike 5.1 +/- 1.2 s). On the basis of their conduction velocity, b
oth the capsaicin-sensitive and the capsaicin-insensitive groups inclu
ded both A- and C-fibres. Application of capsazepine (10 mu M for 5-20
min) to the cornea reversibly blocked the response of the units to ca
psaicin without affecting responses to CO2. Units that were acutely de
sensitised by exposure to capsaicin (0.1 mu M) for 30 s, during which
time the discharge evoked by capsaicin declined to zero, still respond
ed to CO2, though the response was reduced by 44% compared with contro
ls. It is concluded that activation of sensory afferents in the cat co
rnea by capsaicin and by CO2 appear to involve distinct mechanisms, si
nce: (a) many CO2-sensitive units are not excited by capsaicin; (b) ca
psazepine selectively blocks excitation by capsaicin without affecting
responses to CO2; and (c) desensitisation to capsaicin impairs only p
artially the responsiveness to CO2. (C) 1997 International Association
for the Study of Pain.