The roles of spatial recruitment and discharge frequency in spinal cord coding of pain: a combined electrophysiological and imaging investigation

Abstract

An investigation was conducted to examine both temporal and spatial factors likely to be involved in spinal cord nociceptive coding by wide cord nociceptive neurons. Three separate methodologies were employed. First, the impulse frequency responses of L 4 spinal cord wide-dynamic-range (WDR) neurons to gentle mechanical stimulation, vigorous but innocuous brushing, warmth (43°C), and nociceptive thermal stimuli (45–49°C) were electrophysiologically characterized in unanesthetized, spinal cord-transected rats. Second, the spatial distribution of evoked activity in response to the same types of mechanical and thermal stimuli was examined utilizing the 14C-2-deoxyglucose (2-DG) metabolic mapping method in the same type of animal preparation. Finally, the contributions of impulse frequency and numbers of neurons activated to encoding the distinction between painful and non-painful sensations were directly evaluated by electrically stimulating axons within the spinal cord anterolateral quadrant (ALQ) of conscious human subjects. Electrophysiological findings revealed that vigorous but innocuous brushing produced intermediate rates of impulse discharge significantly greater than those produced by 35 and 43°C stimuli, yet indistinguishable from those produced by relatively low nociceptive temperatures (45–47°C). Thus, the discharge frequencies of individual dorsal horn WDR neurons alone do not provide sufficient information to encode the distinction between innocuous and low intensity nociceptive stimuli. Mapping of spinal cord activity by the 2-DG method revealed that nociceptive stimuli activated extensive rostro-caudal regions extending from L 1–L 5. In contrast, vigorous but innocuous brushing evoked metabolic activity that was confined to a narrow zone within L 3. Thus, as predicted from previous studies, the distinction between nociceptive and non-nociceptive sensory events may be encoded, in part, by differences in the spatial distribution, and hence, the relative numbers of spinal cord neurons activated by nociceptive and innocuous stimuli. The responses of conscious human subjects to varying frequencies and intensities of electrical ALQ stimulation clarify the significance of the large numbers of spinal cord neurons activated by nociceptive stimuli. With stimulus frequency held constant at 50 Hz, low stimulus currents, sufficient to activate only small numbers of ALQ axons, produced innocuous sensations. Higher stimulus currents, sufficient to activate larger numbers of neurons, consistently produced painful sensations. Increasing ALQ stimulus frequency at currents subthreshold for pain or increasing stimulus currents at frequencies subthreshold for pain resulted in painful sensations, thus indicating that both discharge frequency and numbers of neurons activated are both important factors in the encoding of pain. The combination of observations from single neuron recording, 2-DG mapping, and ALQ stimulation experiments provide strong evidence that the number of spinal cord neurons activated by peripheral stimuli and the frequencies at which they discharge are both crucial factors that combine to encode the distinction between innocuous and nociceptive somatosensory events as well as the intensity of nociceptive stimulation. This combination of spatial and temporal factors provides a mechanism by which WDR neurons that respond to both innocuous and nociceptive stimuli can encode pain.