Mechanical and heat sensitization of cutaneous nociceptors in rats with experimental peripheral neuropathy.

This study examined whether or not the properties of cutaneous nociceptive fibers are altered in the neuropathic state by comparing lumbars 5 and 6 spinal nerve ligation (SNL) rats with sham-operated controls. The rats with the unilateral SNL developed mechanical allodynia in the ipsilateral hind limb, whereas the sham group did not. Two to 5 weeks after the neuropathic or sham surgery, rats were subjected to single fiber-recording experiments to examine the properties of afferent fibers in the sural and plantar nerves. A total of 224 afferents in the C- and Adelta-ranges were characterized in the neuropathic and sham groups. Spontaneous activity was observed in 16 of 155 fibers in the neuropathic group and one of 69 fibers in the sham group. The response threshold of both the C- and Adelta-fibers to mechanical stimuli was lower in the neuropathic group than the sham group. The afferent fibers responsive to heat stimuli were all C-fibers, and none were Adelta-fibers. The response threshold of the C-fibers to the heat stimuli was lower in the neuropathic group than the sham group. The magnitude of the responses of both C- and Adelta-fibers to the suprathreshold intensity of the mechanical stimulus was greater in the neuropathic group than the sham group. However, the magnitude of the responses of C-fibers to the suprathreshold intensity of the heat stimulus in the neuropathic group was not different from that in the sham group. These results suggest that after a partial peripheral nerve injury, the nociceptors on the skin supplied by an uninjured nerve become sensitized to both mechanical and heat stimuli. This nociceptor sensitization can contribute to neuropathic pain.

Fiber counts at multiple sites along the rat ventral root after neonatal peripheral neurectomy or dorsal rhizotomy.

We hypothesized that the afferent fibers in the ventral root of the rat are the third branches of dorsal root ganglion cells; these afferent processes in the ventral root are of varying length and end bluntly along the length of the root. In the case of an injury at either the central or the peripheral processes of the dorsal root ganglion cells in the neonatal stage, these fibers sprout at the blunt endings along the length of the ventral root. We cut either the sciatic nerve or the dorsal root on one side in neonatal rats. After the rats were fully grown, the number of both myelinated and unmyelinated fibers was counted in electron photomicrographs at multiple sites along the length of the ventral root. We observed a greatly increased number of unmyelinated fibers in the ventral root after the sciatic nerve had been cut at the neonatal stage. The magnitude of increase was more at the distal than at the proximal portion of the ventral root, suggesting that added fibers originated from the distal side. Neonatal dorsal rhizotomy, however, did not produce the same result. These results are consistent with our hypothesis that peripheral nerve injury at the neonatal stage triggers sprouting of the third branches of the dorsal root ganglion cells which end bluntly along the length of the ventral root in the normal animal.

Peripheral glutamate receptors contribute to mechanical hyperalgesia in a neuropathic pain model of the rat.

We hypothesized that glutamate (Glu) released from the peripheral terminals of primary afferents contributes to the generation of mechanical hyperalgesia following peripheral nerve injury. Nerve injury was performed on rats with a lumbar 5 spinal nerve lesion (L5 SNL), which was preceded by L5 dorsal rhizotomy (L5 DR) to avoid the potential central effects induced by L5 SNL through the L5 dorsal root. Mechanical hyperalgesia, as evidenced by a reduction in paw withdrawal threshold (PWT), was short-lasting (<6 days) after L5 DR, but persistent (>42 days) after L5 SNL preceded by L5 DR. When an intraplantar injection into the affected hind paw was given immediately before L5 SNL, non-competitive N-methyl-D-aspartate (NMDA) receptor antagonist MK-801 (20 nmol), group-I metabotropic Glu (mGlu) receptor antagonist DL-amino-3-phosphonopropionic acid (DL-AP3; 70 nmol), and selective group-II mGlu receptor agonist 4-aminopyrrolidine-2,4-dicarboxylate (APDC; 20 nmol) delayed the onset of PWT reduction for 1-4 days. However, this onset was not affected by alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid/kainate receptor antagonist 2,3-dioxo-6-nitro-1,2,3,4,-tetrahydrobenzo[f]quinoxaline-7-sulfonamide (NBQX; 100 nmol). When the same injection was given after L5 SNL-induced mechanical hyperalgesia had been established, MK-801 reversed the PWT reduction for 30-75 min, whereas NBQX, DL-AP3, or APDC had no effect. These results suggest that the manipulation of the peripheral Glu receptors reduces neuropathic pain, by blocking NMDA and group-I mGlu receptors and by stimulating group-II mGlu receptor during the induction phase of neuropathic pain, but only by blocking the NMDA receptor during its maintenance phase.

Activity-dependent conduction latency changes in A beta fibers of neuropathic rats.

Activity-dependent changes of the conduction latency of single A beta fibers of primary afferent neurons were characterized in both neuropathic (L4 and L6 ligated) and normal rats. Activity-dependent increases in conduction latency of dorsal root fibers in neuropathic rats were significantly stronger than those in normal rats. Different profiles of activity dependence were also observed between injured and adjacent intact dorsal root fibers of neuropathic rats. However, activity-dependent latency changes in sciatic nerves distal to the dorsal root ganglion were not different between neuropathic and normal rats. These results suggest that partial nerve injury induces activity-dependent excitability changes in the dorsal root fibers of neuropathic rat and that these changes may be responsible for the altered sensory processing such as those seen in allodynia.

Activity-dependent conduction velocity changes of A(delta) fibers in a rat model of neuropathy.

Activity-dependent changes of conduction velocity (CV) and conduction block in single A(delta) fibers of primary afferent neurons were characterized in a rat model of neuropathy (NP). Injured dorsal root (DR) fiber in NP rats exhibited profoundly greater decreases of CV following impulse activity than did DR fiber in normal rats. Activity-dependent conduction block was absent up to 100 Hz of activity rate in DR fiber of NP rats, but was present above 25 Hz in normal rats. Profiles of activity dependence in sciatic fibers were similar in both NP and normal rats. These results suggest that nerve injury may alter activity-dependent hypoexcitability of A(delta) DR fibers. Furthermore, this excitability change may be responsible for the elevated pain perception in neuropathy.

Somatic afferent fibers which continuously discharge after being isolated from their receptors.

A previous study in our laboratory has shown that some afferent axons produce prolonged discharges after the axons have been completely isolated. We have attempted to identify the type of afferent fibers displaying such activity. Single unit activity was recorded from a filament dissected from the distal stump of the cut sural or plantar nerve of an anesthetized rat. After thorough identification of the receptor type, the nerve was cut at a site between the recording electrode and the receptive field, completely isolating the fiber being recorded. Unit activity was recorded up to 1 h after sectioning the nerve. Upon sectioning the nerve, most units showed brief injury discharges lasting only a few seconds. However, 21 of 70 units exhibited prolonged discharges lasting at least 30 min after having been isolated from their receptors. These 21 units included 8 slowly adapting type II cutaneous mechanoreceptors, 3 Pacinian corpuscles and 10 muscle spindle afferent units. These results suggest that prolonged injury discharges can be produced in the axons of the slowly adapting type II cutaneous mechanoreceptors, Pacinian corpuscles, and muscle spindle afferents. This phenomenon may have important clinical and experimental consequences.

Electrophysiological evidence for the antinociceptive effect of transcutaneous electrical stimulation on mechanically evoked responsiveness of dorsal horn neurons in neuropathic rats.

Using a rat model of peripheral neuropathy induced by a tight ligation of L5-6 spinal nerves, the effects of transcutaneous electrical stimulation on the mechanical responses of wide dynamic range (WDR) dorsal horn neurons were investigated. The responses of the WDR neurons to both the brush and pinch stimuli were found to be enhanced in the neuropathic rats compared to those in the normal rats. These enhanced responses were depressed by low-frequency and high-intensity transcutaneous electrical stimulation (2 Hz, 4-5 mA) applied to the somatic receptive field. The durations of the depressive effects on the brush responses ranged between 30 and 45 min and those on the pinch responses were 60-90 min. These results imply that the transcutaneous electrical stimulation used here produces an antinociceptive effect via a depressive action on the enhanced mechanical responsiveness of the spinal neurons in this rat model of peripheral neuropathy.

Involvement of alpha2-adrenoceptors in mediating sympathetic excitation of injured dorsal root ganglion neurons in rats with spinal nerve ligation.

The present study examined the effects of sympathetic stimulation on the activity of primary afferent neurons that had peripheral axons being injured previously by a spinal nerve ligation. About 22% of afferents with injured fibers that showed spontaneous discharge were excited by sympathetic stimulation or systemic injection of adrenaline. Most sympathetically-excited afferent neurons had axons that conducted in the A-fiber range. This sympathetically-evoked afferent excitation was not affected by cutting the spinal nerve at a place close to the dorsal root ganglion (DRG). Yohimbine, alpha2-antagonist, suppressed sympathetically-evoked afferent excitation which was not affected by alpha1-antagonist prazosin. Clonidine, alpha2-agonist, exerted an excitatory effect, whereas alpha1-agonist phenylephrine had no effect on the activity of afferents with injured fibers. No afferent fibers in control preparations responded to sympathetic stimulation. The results suggest that after a spinal nerve ligation, injured DRG neurons with fast-conducting fibers become sensitive to sympathetic activity via activation of alpha2-adrenoceptors.

N-methyl-D-aspartate (NMDA) and non-NMDA glutamate receptor antagonists differentially suppress dorsal horn neuron responses to mechanical stimuli in rats with peripheral nerve injury.

The effects of iontophoretically ejected glutamate receptor antagonists on mechanically evoked responsiveness were examined on wide dynamic range (WDR) dorsal horn neurons in anesthetized rats that received a unilateral ligation of the L5 and L6 spinal nerves 10-15 days previously. Both brush- and pinch-evoked responses of dorsal horn neurons on the nerve-injured side were enhanced. N-Methyl-D-aspartate (NMDA) receptor antagonist, 2-amino-5-phosphonopentanoic acid (AP5), preferentially suppressed the enhanced pinch-evoked response, whereas (RS)-a-amino-3-hydroxy-5-methyl-4-isoxazole-propionate (AMPA) receptor selective antagonist, 6-nitro-7-sulphamoylbenzo(f)quinoxaline-2,3-dione (NBQX), preferentially attenuated the enhanced brush-evoked response. The results indicate that the enhanced responses to noxious and non-noxious peripheral inputs induced in WDR dorsal horn neurons following the nerve injury are mediated by activation of NMDA and AMPA receptors, respectively.

The role of peripheral N-methyl-D-aspartate receptors in Freund's complete adjuvant induced mechanical hyperalgesia in rats.

We investigated the role of excitatory amino acid receptors in mechanical hyperalgesia induced by subcutaneous injection of Freund's complete adjuvant (FCA) into the rat hind paw. In normal rats, an intraplantar (i.pl.) injection of L-glutamate, but not of D-glutamate (3 pmol/0.1 ml each) produced a mechanical hyperalgesia in the hind paw with a lowered paw-withdrawal threshold to pressure. In rats that developed mechanical hyperalgesia associated with inflammation in the hind paw following i.pl. injection of FCA (0.15 ml), the injection of N-methyl-D-aspartate (NMDA) receptor antagonist, MK-801 (1 pmol/0.1 ml) into the inflamed paw increased the paw pressure threshold. On the other hand, the injection of non-NMDA receptor antagonist, 6-cyano-7-nitroqiunoxaline-2,3-dione (CNQX, 10 pmol/0.1 ml) into the inflamed paw had no effect on FCA-induced lowering of the paw pressure threshold. The results suggest that NMDA, but not non-NMDA receptors play a substantial role in mediating the development of mechanical hyperalgesia induced in the inflamed paw following i.pl. FCA injection.

Differential antinociceptive effect of transcutaneous electrical stimulation on pain behavior sensitive or insensitive to phentolamine in neuropathic rats.

The effects of transcutaneous electrical stimulation and systemic injection of phentolamine, a non-specific alpha-adrenergic antagonist, on the behavioral signs of mechanical allodynia and cold hyperalgesia in rats with nerve injury were investigated. Mechanical allodynia and cold hyperalgesia were evaluated by measuring the paw withdrawal frequency (PWF) resulting from repetitive application of a von Frey hair and the paw lift duration (PLD) at a cold temperature, respectively. After a unilateral nerve injury, both PWF and PLD increased in the injured hind paw. Application of low-frequency, high-intensity transcutaneous electrical stimulation (LFHI-TES) to the injured hind paw depressed the injury-induced increased PWF, whereas it had no effect on the injury-induced increased PLD. Naloxone reversed the LFHI-TES produced depression of PWF. Intraperitoneal administration of phentolamine depressed the injury-induced increased PLD without affecting the injury-induced increased PWF. Our results suggest that LFHI-TES, which activates the endogenous opioid systems, produces an antinociceptive effect that appears to be related to whether or not the pain is mediated by sympathetic activity.

Increased number of unmyelinated fibers in the ventral root after peripheral neurectomy in adult rat.

We examined the possibility that peripheral nerve injury in the adult rat triggers sprouting of unmyelinated ventral root afferent fibers. Three to 5 months after the sciatic nerve was sectioned on one side in the adult rat, myelinated and unmyelinated fibers were counted at 3 sites along the length of the ventral root. A sciatic nerve lesion resulted in about a 3-fold increase in the number of unmyelinated fibers in the L5 ventral root. Our data suggest that a peripheral nerve lesion in the adult rat triggers sprouting of unmyelinated afferent fibers in the ventral root. No evidence was found that dorsal rhizotomy triggers sprouting of afferent fibers.

Conduction block by clonidine is not mediated by alpha2-adrenergic receptors in rat sciatic nerve fibers.

BACKGROUND AND OBJECTIVES: Clonidine, an alpha(2)-adrenergic agonist, has been shown to prolong local anesthesia. It appears that clonidine by itself produces conduction block by acting on peripheral nerves. However, whether clonidine-induced conduction block is mediated through alpha(2)-adrenergic receptors remains unclear. The purpose of this study was to see if clonidine's nerve-blocking action was through alpha(2)-adrenergic receptors by examining clonidine's action in the presence of alpha(2)-adrenergic antagonists. METHODS: The compound action potentials (CAPs) evoked by electrical stimuli were recorded from the isolated rat sciatic nerve in a recording chamber. Conduction block was examined by analyzing CAPs with regard to peak amplitude and time-to-peak in the presence of clonidine alone or clonidine plus alpha(2)-adrenergic antagonist yohimbine or idazoxan. RESULTS: Both clonidine and yohimbine produced concentration-dependent, reversible, conduction block. Based on concentration-response relationships, the 50% of effective concentration (EC(50)) were estimated to be 1.61 +/- 0.51 mmol/L (mean +/- SD) for clonidine and 51.4 +/- 27.2 micromol/L for yohimbine. A mixture of equal volumes of 2.07 mmol/L clonidine and 55.6 micromol/L yohimbine produced conduction block to a level close to the mean value between conduction blocks induced by 2.07 mmol/L clonidine alone and 55.6 micromol/L yohimbine alone. Addition of idazoxan, a more specific alpha(2)-adrenergic antagonist than yohimbine, to clonidine was without effect on clonidine-induced conduction block. CONCLUSIONS: The results indicated that the mixture of clonidine and yohimbine, in which either drug inhibited impulse conduction, produced conduction block in an additive manner, and that clonidine-induced conduction block was not reversed by coapplication with a specific alpha(2)-adrenergic antagonist idazoxan. These data suggest that clonidine's effects likely depend on mechanisms not mediated by alpha(2)-adrenergic receptors.

Adrenergic sensitivity of uninjured C-fiber nociceptors in neuropathic rats.

We investigated the adrenergic sensitivity of afferent fibers in the L4 dorsal roots of rats with a unilateral ligation of the L5-L6 spinal nerves. About 12% of nociceptive fibers on the affected side were excited by sympathetic stimulation or by intra-arterial injection of norepinephrine which did not affect A beta-fiber activity. Sympathetic excitation of nociceptive fibers was suppressed by alpha 1-antagonist prazosin, while it was unaffected by alpha 2-antagonist yohimbine. Most of these fibers were excited by intra-arterial injection of alpha 1-agonist phenylephrine, without being affected by an injection of alpha 2-agonist clonidine. Sympathetic excitation was blocked by lidocaine applied near the receptive fields of recorded fibers. The results suggested that some nociceptors remaining intact after partial nerve injury become sensitive to sympathetic activity by the mediation of alpha 1-adrenoceptors in the peripheral endings.

Effects of iontophoretically applied substance P, calcitonin gene-related peptide on excitability of dorsal horn neurones in rats.

Spontaneous pain, allodynia and hyperalgesia are well known phenomena following peripheral nerve or tissue injury, and it is speculated that secondary hyperalgesia and allodynia, are generally thought to depend on a hyperexcitability (sensitization) of neurons in the dorsal horn. It is supposed that the sensitization may be due to various actions of neurotransmitters (SP, CGRP, excitatory amino acids) released from the primary afferent fibers. In this study, we examined effects of the iontophoretically applied SP and CGRP on the response to EAA receptor agonists (NMDA and non-NMDA) in the WDR dorsal horn neurones and see if the effects of SP or CGRP mimic the characteristic response pattern known in various pain models. The main results are summarized as follows: 1) SP specifically potentiated NMDA response. 2) CGRP non-specifically potentiated both NMDA and AMPA responses. Potentiation of NMDA response, however, was significantly greater than that of AMPA response. 3) 50% of SP applied cells and 15.8% of CGRP applied cells showed reciprocal changes(potentiation of NMDA response and suppression of AMPA response). These results are generally consistent with the sensitization characteristics in diverse pain models and suggests that the modulatory effects of SP and CGRP on NMDA and non-NMDA (AMPA) response are, at least in part, contribute to the development of sensitization in various pain models.

Peripheral contributions to the mechanical hyperalgesia following a lumbar 5 spinal nerve lesion in rats.

Using lumbar 5 (L5) dorsal root rhizotomy-bearing rats, we examined the extent to which L5 spinal nerve lesion (SNL)-induced mechanical hyperalgesia was governed by two peripheral components, that is Wallerian degeneration (WD) and peripherally-propagating injury discharge (PID). The contribution of WD to SNL-induced hyperalgesia was studied by excluding PID with lidocaine treatment that blocked nerve conduction temporarily, but completely at the time of injury, whereas PID was examined separately by using brief tetanic electrical stimulation of the spinal nerve mimicking PID. Following the disappearance of L5 rhizotomy-induced transient hyperalgesia, L5 SNL resulted in long-lasting mechanical hyperalgesia as early as one day post-SNL despite a PID block, highlighting the role of WD. In a comparative experiment, a delayed onset of hyperalgesia (7 days) was measured in L3 rhizotomy-bearing rats following L3 SNL with a PID block, in which injured fiber (L3) was separated from intact fibers (L4 and L5) anatomically until they meet at the peripheral terminals, supporting the importance of interactions between degenerating and adjacent intact fibers for WD-induced hyperalgesia. Tetanic electrical stimulation of decentralized L5 spinal nerve resulted in mechanical hyperalgesia developing within 1 day and persisting for 7 days. This hyperalgesia was prevented by lidocaine blockade of the L5 nerve, and was unaffected by lidocaine blockades of the central inputs from L3 and L4 fibers during L5 nerve stimulation, suggesting the mediation of PID-induced hyperalgesia by sensitization, not activation, of peripheral terminals of adjacent intact afferents. The similar hyperalgesia was also observed following electrical stimulation of decentralized L3 spinal nerve. Prior elimination of L4 C-fibers by local capsaicin prevented hyperalgesia induced either by L5 SNL with a PID block or by L5 nerve stimulation. These results suggest that neighboring C-afferents remaining intact after partial nerve injury play a critical role in the development of mechanical hyperalgesia through interaction with degenerating afferents, and also via peripheral sensitization by PID.

Cutaneous sensory receptors in the rat foot.

1. A total of 574 cutaneous afferent units in the sural and plantar nerves supplying the skin of the rat foot was examined: 399 A beta-units, 55 A delta-units, and 120 C-units. Their receptive-field (RF) properties were similar to those described in other mammals. However, the receptor type composition of units was different between the two nerves. 2. The sural A beta-fiber sample (n = 160) consisted of G-hair (41%), field (11%), rapidly adapting (RA; 6%), slowly adapting type I (SA-I; 7%), and type II (SA-II; 35%) mechanoreceptors. The plantar A beta-fiber sample (n = 239) was composed of G-hair (3%), RA (35%), SA-I (30%), SA-II (24%), and Pacinian corpuscle (PC; 8%) mechanoreceptors. 3. The RFs of SA-II units were located on both hairy and glabrous skin overlying the foot joints. Many of the SA-II units responded to movement of the foot joints. The RFs of both SA-I and RA units were small in size and located in high density on the toe tips and footpads. PC units were very sensitive to vibration and had extremely large RFs as in other species, although they were rare and found only in the plantar nerve. Field units were similar to SA-II units in response properties and RF distribution. 4. The sural A delta-fiber sample (n = 44) included nociceptors (68%), D-hair (27%), and cold (5%) receptors. All sampled plantar A delta-fibers (n = 11) were nociceptors. Of A delta-nociceptor units, A delta-mechanical nociceptors (73%) were dominant. 5. The sural C-fiber sample (n = 85) included nociceptors (44%), C-mechanoreceptors (33%), and cold receptors (21%). The plantar C-fiber sample (n = 35) included nociceptors (77%) and cold receptors (23%). No warm units were found among either the sural or plantar nerve fibers. Of C-nociceptors, C-mechanoheat nociceptors (80%) were dominant. 6. The results indicate that all well-known types of cutaneous receptors, except warm receptors, exist in the foot skin of the rat. On the basis of the fact that RFs of RA and SA-I units are in high density on the toe tips and footpads, it is suggested that those regions may have a spatial discriminating capacity. It is also suggested that SA-II receptors may play a role in proprioception, because they have RFs on the skin over foot joints and respond to joint movement.(ABSTRACT TRUNCATED AT 400 WORDS)

Abnormalities of mechanoreceptors in a rat model of neuropathic pain: possible involvement in mediating mechanical allodynia.

1. Peripheral nerve injury sometimes leads to the development of neuropathic pain. One of the symptoms of such neuropathic pain is mechanical allodynia, pain in response to normally innocuous mechanical stimuli. We hypothesized that sympathetically driven dysfunction of cutaneous mechanoreceptors is responsible for signaling mechanical allodynia. The present study was undertaken to identify the types of sensory receptors that potentially mediate mechanical allodynia, with the use of a rat neuropathic pain model we have developed. 2. One week to 10 days after tight ligations of the L5 and L6 spinal nerves on one side, the rats fully developed behavioral signs of mechanical allodynia on the affected hindlimb. Various cutaneous mechanoreceptors originating from the neuropathic foot were examined by single-fiber recordings from the L4 dorsal root. 3. Although no particular abnormalities were found in other types of cutaneous mechanoreceptors, an unusual type of mechanoreceptor was found to be innervating the neuropathic foot. The response characteristics of this type of receptor resemble those of rapidly adapting mechanoreceptors (RAs), but with low and irregular static discharges during a maintained mechanical stimulus. We termed this unusual type as a "modified rapidly adapting" mechanoreceptor (MRA). 4. The response characteristics of MRAs change to those of typical RAs after a systemic injection of phentolamine, an alpha-adrenergic receptor blocker. 5. We conclude that many RAs become abnormal under the influence of sympathetic efferents in neuropathic pain, so that their response patterns change to those of MRAs. We propose that this abnormality is responsible for signaling the mechanical allodynia that can be seen in neuropathic pain states such as causalgia.

Differential activation and classification of cutaneous afferents in the rat.

1. A total of 312 cutaneous afferent units identified in the rat foot as belonging to one of nine major types of sensory receptors were included in the present study. A natural stimulus set was defined to differentiate optimally among those receptor types according to the distinguishing response patterns that it produced. It included air puffs, 30- and 300-Hz sinusoids, 200-mN force indentation of the skin, 1.2- and 6-N compressions of a skin fold, cooling the skin by 5 and 20 degrees C, warming by 5 degrees C, and heating by 15 degrees C. 2. The responses to predefined stimuli of 188 units were subjected to multivariate statistical analyses. The responses of an individual unit were measured as the number of impulses evoked by 10 stimuli, each lasting 10 s. Additionally, the number of impulses occurring for 5 s after withdrawal of a 200-mN indentation (1 of the 10 stimuli) was counted. 3. In discriminant analysis, the 11 stimulus variables predicted fairly correctly the grouping of afferent units into nine predetermined receptor categories (175 of 188, 93.1%), indicating a powerful ability to discriminate among different receptor types. Using hierarchical cluster analysis, afferent unit data described by 11 variables were divided into clusters that well represented prior receptor categories (170 of 188, 90.4%), suggesting the reliable application of this procedure to the classification of newly recorded cutaneous sensory receptors. 4. Eleven variables were then reduced to 7 on the basis of the results of factor analysis (95% of variance accounted for). The seven variables corresponded to 1.2-N compression, heating the skin by 15 degrees C, cooling the skin by 20 degrees C, 30- and 300-Hz sinusoids, withdrawal of a 200-mN indentation, and air puffs. 5. The seven selected variables correctly assigned afferent units into five modality-based categories in the discriminant solution (177 of 188, 94.1%). In the cluster solution, afferent units described by the seven selected variables were divided into clusters, most of whose members were modality specific (176 of 188, 93.6%). 6. The results indicate that cutaneous receptors can be divided into modality-specific groups according to similarities in their responses to seven stimulus variables. It is proposed that the stimulus set developed here and multivariate statistical methods can be used as powerful tools for the functional classification of central somatosensory neurons.

Grouping of somatosensory neurons in the spinal cord and the gracile nucleus of the rat by cluster analysis.

1. A set of 11 cutaneous stimuli defined previously to differentiate among different types of cutaneous sensory receptors in the rat hindpaw was also effective in differentially activating second-order sensory neurons in the dorsal horn and the gracile nucleus of rats. 2. All sampled units were responsive to more than 1 of the 11 stimuli. However, none responded to innocuous warming or cooling stimuli. Therefore further analysis was restricted to responses to nine of the selected stimuli. 3. Cluster analysis of the responses to nine selected innocuous and noxious mechanical stimuli and noxious thermal stimuli yielded seven classes that seemed functionally distinct from each other: a class of high-threshold neurons, three classes of convergent (wide dynamic range) neurons, a class of a mixture of poorly responsive neurons and neurons receiving Pacinian inputs, and two classes of low-threshold neurons. 4. High-threshold neurons responded predominantly to noxious mechanical and thermal stimuli and presumably received an input from both mechanically and thermally sensitive nociceptors. These cells were located in the dorsal horn, and some were spinothalamic tract cells. Wide dynamic range neurons were excited by innocuous and noxious stimuli, but better by noxious stimuli. These classes of cells were either in the dorsal horn (some were spinothalamic tract cells) or in the nucleus gracilis.(ABSTRACT TRUNCATED AT 250 WORDS)