1. The aim of this study was to utilize new quantitative behavioral methods in rats to investigate the effects of electrical stimulation in midbrain analgesia areas on the magnitude of flexion hindlimb withdrawal and tail flick reflexes evoked by graded noxious heating. Electrophysiological experiments were then done with the use of these animals to correlate behavioral data with the effects of identical midbrain stimulation on sacral dorsal horn neuronal responses to graded heating of the tail. 2. To quantify limb withdrawals, electromyographs (EMGs) were recorded in biceps femoris during withdrawals elicited by noxious heat stimuli (40-52 degrees C, 5 s, 2-min intervals) delivered to the plantar surface of the hind paw, without and during concomitant electrical stimulation (100 ms, 100-Hz trains, 3/s, 10-600 mu A) in midbrain periaqueductal gray (PAG) or laterally adjacent reticular formation (LRF) via previously implanted electrodes. The same animals were tested with the use of a tail flick paradigm modified to allow measurement of the force of tail movements in three orthogonal planes and thereby calculate the overall force vector of tail flicks elicited by graded noxious radiant heat pulses (38-58 degrees C, 5 s, 2-min intervals) delivered to the tail, again with and without concomitant PAG or LRF stimulation. Finally, the same rats were anesthetized with pentobarbital sodium and microelectrode recordings made from single sacral dorsal horn neurons responsive to the same noxious heat stimuli delivered to the tail to assess effects of PAG and LRF stimulation. 3. PAG and LRF stimulation suppressed the magnitude of limb flexor EMGs, and tail flick force vectors, in an intensity-dependent manner. Recruitment of suppression of both limb withdrawal EMGs and tail flicks was generally more effective for LRF compared with PAG stimulation, although mean thresholds for suppression were similar. Tail flick force and limb withdrawal EMGs recorded from the same rat in separate sessions were suppressed about equally in a majority of cases. 4. Limb withdrawal EMG magnitude increased monotonically from threshold (similar to 40 degrees C) to 52 degrees C. The population stimulus-response function was fit equally well by linear regression or a 2 degrees polynomial function (r(2) = 0.79 for bath). PAG stimulation significantly reduced the mean slope of the stimulus-response function (to 73%; n = 15), whereas LRF stimulation shifted it toward the right with a smaller slope reduction (to 85%) and 3 degrees C increase in threshold. 5. Tail flick force vectors typically increased linearly from threshold (40 degrees C) to similar to 48-52 degrees C and then leveled off. PAG stimulation reduced the slope of the stimulus-response function (50%) with little change in threshold. LRF stimulation produced a near-parallel rightward shift with an similar to 3 degrees C increase in threshold. 6. In 14 behaviorally tested rats, 41 sacral units responded to noxious tail heating. Units additionally responded to a range of mechanical brush, tap, and pressure stimuli and had receptive fields ranging in size from a few millimeters squared to the entire tail. They were thus classified as multireceptive or wide dynamic range type. 7. Unit responses to noxious heat were suppressed in an intensity-dependent manner by PAG and LRF stimulation. Neuronal responses were generally less effectively suppressed by midbrain stimulation compared with previously measured tail flicks in the same rats. 8. Neuronal responses increased linearly from threshold (mean 38.6 degrees C) to similar to 52 degrees C, beyond which they often plateaued. The slope of the population neuronal stimulus-response function was reduced (to 53%) by PAG stimulation with little threshold change. LRF stimulation reduced the slope less (to 61%) and increased the threshold by similar to 4 degrees C. 9. These results indicate that midbrain PAG and LRF stimulation may exert parametrically distinct suppressive effects on limb withdrawal and tail flick reflexes, and sacral dorsal horn neurons, in accordance with previously reported effects on lumbar dorsal horn neurons.
