The Journal of Neurophysiology Vol. 80 No. 3 September 1998, pp. 1082-1093
Copyright ©1998 by the American Physiological Society

Myelinated Mechanically Insensitive Afferents From Monkey Hairy Skin: Heat-Response Properties

Rolf-Detlef Treede^{1, 2}, Richard A. Meyer^{1, 3}, and James N. Campbell^{1, 3}

¹ Department of Neurosurgery, School of Medicine, The Johns Hopkins University, Baltimore 21205; ² Institute of Physiology and Pathophysiology, Johannes Gutenberg University, D-55099 Mainz, Germany; and ³ Applied Physics Laboratory, The Johns Hopkins University, Laurel, Maryland 20723

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Treede, Rolf-Detlef, Richard A. Meyer, and James N. Campbell. Myelinated mechanically insensitive afferents from monkey hairy skin: heat-response properties. J. Neurophysiol. 80: 1082-1093, 1998. To compare the heat responses of mechanically sensitive and mechanically insensitive A-fiber nociceptors, an electrical search technique was used to locate the receptive fields of 156 A-fibers that innervated the hairy skin in the anesthetized monkey (77 A-fibers, 79 A-fibers). Two-thirds of these afferents were either low-threshold mechanoreceptors (n = 91) or low-threshold cold receptors (n = 11). Nine A-fibers and 41 A-fibers were cutaneous nociceptors, and four A-fibers innervated subcutaneous tissue. The majority of cutaneous A-fiber nociceptors were heat sensitive (43/50 = 86%). Heat-insensitive cutaneous A-fiber nociceptors consisted of one cold nociceptor, three silent nociceptors, and three high-threshold mechanoreceptors. Two types of response were observed to an intense heat stimulus (53°C, 30 s). Type I (n = 26) was characterized by a long latency (mean: 5 s) and a late peak discharge (16 s). Type II (n = 17) was characterized by a short latency (0.2 s) and an early peak discharge (0.5 s). Type I fibers exhibited faster conduction velocities (25 vs. 14 m/s) and higher heat thresholds (>53 vs. 47°C, 1-s duration) than type II fibers. The possibility that the type I heat response was a result of sensitization was tested in three fibers by determining the heat threshold to 30-s duration stimuli (42-46°C). For this long stimulus duration heat thresholds were reproducible across multiple runs, and the threshold to the 1-s duration stimulus was not altered by these tests. Thus fibers with a type I heat response were not high-threshold mechanoreceptors that developed a heat response through sensitization. Fibers with a type II heat response had significantly higher mechanical thresholds (median: 15 bar) than fibers with a type I heat response (5 bar). This finding accounts for the observation that type II heat responses were infrequently observed in earlier studies wherein the search technique depended on mechanical responsiveness. Fibers with a type II response exhibited a graded response to heat stimuli, marked fatigue to repeated applications of heat stimuli, and adaptation to sustained heat stimuli similar to that seen in C-fiber nociceptors. First pain sensation to heat is served by type II A-fiber nociceptors that are mechanically insensitive. Type I A-fiber nociceptors likely signal pain to long-duration heat stimuli and may signal first pain sensation to mechanical stimuli.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Mechanically and heat-sensitive A-fiber nociceptors (AMHs) from monkey skin exhibit two distinct heat transduction mechanisms (Treede et al. 1995). Type I AMHs have a delayed onset of response to heat stimuli, and their discharge rate typically increases during a prolonged heat stimulus. Their median heat threshold is high (>53°C), and they are found in glabrous and hairy skin. These fibers encode the sustained pain during a prolonged, intense heat stimulus and the primary hyperalgesia to heat that develops after a burn to the glabrous skin (Meyer and Campbell 1981b). In contrast, type II AMHs have a short-latency adapting response to heat stimuli. Their median heat threshold is low (46°C), and they are found only in hairy but not glabrous skin. The response properties and conduction velocities of type II AMHs are such that this fiber type is an ideal candidate to subserve the sensation of "first pain" to stepped heat stimuli. Although first pain to heat is a prominent sensation in psychophysical studies, relatively few type II AMHs are evident with conventional neurophysiological search techniques. This paradox prompted us to consider whether other search techniques might demonstrate a higher incidence of type II AMHs.

Most investigations of A-fiber nociceptors depended on a technique that uses responses to mechanical stimuli to locate cutaneous receptive fields (e.g., Adriaensen et al. 1983; Burgess and Perl 1967; Fitzgerald and Lynn 1977; Georgopoulos 1976; Treede et al. 1995). This common search technique may introduce a selection bias, especially because recent studies showed that sensitivity of nociceptors to heat and chemical stimuli may be unrelated to their mechanical sensitivity (Davis et al. 1993; Schmidt et al. 1995). Indeed more than one-half of the A-fiber nociceptor population exhibits high mechanical thresholds >6 bar (100 mN) and thus are unlikely to be found with a mechanical search technique (Meyer et al. 1991). The prime motivation for this study was to determine the heat-response characteristics of mechanically insensitive afferents (MIAs).

An electrical search technique (Meyer et al. 1991) was used to determine the receptive field locations of all A-fiber afferents in monkey cutaneous nerves. We demonstrate that A-fiber MIAs, like mechanically sensitive afferents, exhibit two distinct heat transduction mechanisms. However, a large proportion of the MIAs exhibits a type II response, whereas a large proportion of mechanically sensitive afferents exhibits a type I response.

	METHODS

Abstract Introduction Methods Results Discussion References

A standard teased fiber technique (e.g., Campbell and Meyer 1983) was used to record from single primary afferents in cutaneous, peripheral nerves in monkeys (Macaca fascicularis, 5-7 kg). The monkeys were initially sedated by intramuscular injection of ketamine (10 mg/kg), and anesthesia was maintained with a mixture of pentobarbital sodium (3 mg·kg¹·h¹) and morphine sulfate (0.5 mg·kg¹·h¹). At the beginning of each experiment, penicillin G (450,000 U) was administered for prophylaxis against infection. To minimize muscle artifacts as well as to facilitate respiratory control, the animals were paralyzed with pancuronium bromide (0.1 mg/kg) and artificially ventilated (end-tidal pCO₂ maintained at 32-40 Torr). The electrocardiogram was monitored aurally and visually. Changes in heart rate were used to monitor the level of anesthesia. Core temperature was measured via a rectal probe and maintained ~38°C with use of circulating water heating pads. Dextrose (5%) in normal saline was infused intravenously throughout the experiment to maintain hydration.

An incision was made in the skin over the cutaneous nerve of interest, and the edges of the incision were tied to a ring to form a well into which paraffin oil was placed. At the proximal end of this well, small bundles of axons were cut away from the nerve trunk and dissected to fine strands that were placed on the recording electrode. The strands, from which recordings were made, were cut proximally so that only centripetally conducted action potentials were recorded. A bipolar, nerve stimulation electrode (cathode proximal) was placed on the nerve 37.3 ± 6.3 mm (mean ± SD) distal to the recording electrode but within the well. The number of fibers at the recording electrode activated after stimulation at the nerve stimulation electrode was determined. Small strands were dissected for recording to enable each myelinated fiber to be identified on the basis of its electrical threshold, latency, and action-potential shape. A search was then made to locate the receptive field for each of these fibers. For every receptive field that was located, mechanical-electrical interaction consisting of latency shifts in the electrically evoked action potential by intervening responses to natural stimuli (Hallin and Torebjörk 1974; Meyer et al. 1985; Schmelz et al. 1995) or collision techniques (Meyer et al. 1985) were used to verify that a given receptive field was associated with the action-potential shape identified from stimulation at the nerve stimulation electrode.

A fiber was included in the study if 1) its receptive field was located in the hairy skin, 2) its conduction velocity was clearly in the myelinated fiber range (2.5 m/s), 3) no cutaneous injury occurred within 2 cm of the receptive field before study, and 4) the fiber innervated the skin and not subcutaneous tissue (as demonstrated by gently pulling the skin aside relative to the underlying tissue and reapplying the stimulus). Mechanically sensitive fibers were classified as nociceptors if they responded to pinching of the skin but did not respond to blunt pressure (200-g brass rod, 1-cm diam, 0.26 bar) or light stroking. Low-threshold mechanoreceptors were divided into those with rapidly and slowly adapting responses.

To locate the receptive fields of receptors with low or moderate mechanical thresholds, the skin was pinched firmly in a way so as not to damage the skin. Low-threshold thermoreceptors were identified by their spontaneous activity that could be modified by mild warming of the skin with a heat lamp (only cold fibers were observed in this study, which were silenced by mild heating). For most mechanically sensitive afferents, the receptive field area was mapped under a microscope by marking the skin with dye at spots where the fiber responded to a 0.5-mm-diam nylon monofilament, which exerted a force of 200 mN (10 bar). For fibers with a high mechanical threshold, an 18-bar probe (0.7-mm diam, 720 mN) was used for mapping. After waiting several minutes, the threshold to mechanical stimulation was determined with the use of calibrated nylon monofilaments (von Frey type, Stoelting aesthesiometer set) with pressures ranging from 0.1 to 18 bar. As in a previous study (Meyer et al. 1991), we considered any fiber with a mechanical threshold >6 bar to be an MIA. Based on the calibrated von Frey set used in our laboratory, this corresponds to fibers that did not respond to a 5.1-bar (0.36-mm-diam, 51-mN) probe but did respond to the next higher 7.3-bar (0.51-mm-diam, 98-mN) probe.

An electrical search technique (Meyer et al. 1991) was used to locate an electrical receptive field (eRF) for afferents that were not responsive to the pinch stimulus or the low-intensity thermal stimuli. A saline-soaked cotton swab was used as a monopolar, cathodic search electrode. To locate the eRF, the fiber of interest was first excited transcutaneously at a position just distal to the nerve stimulation electrode, and the time for the action potential to reach the recording electrode was recorded. As the search electrode was moved distally along the course of the nerve, the action-potential latency increased gradually. A region was usually found in which the suprathreshold latency no longer increased as the probe was moved distally. In this region, the latency at threshold usually increased abruptly and the electrical threshold decreased (Meyer et al. 1991). This region of longest conduction latency and minimum threshold was defined as the eRF of the fiber. The reduced threshold and longer latency is thought to be associated with the superficial, branching structure of the cutaneous nociceptor terminals.

The conduction velocity of the fiber under study was estimated at the end of each experiment from electrical stimulation at the eRF. For this purpose, two needles were inserted next to the eRF borders, and stimuli of three times the threshold were used. The conduction distance was determined by the length of a piece of suture placed along the presumed path of the nerve between the receptive field and the recording electrode. Conduction velocities were also obtained from latencies measured after stimulation at the nerve stimulation electrode.

The responsiveness of nociceptors to heat stimuli was tested with a laser thermal stimulator system that delivered stepped increases in skin temperature with rise times of ~100 ms (Meyer et al. 1976). The threshold for heat stimuli was tested with stimuli of 1-s duration presented every 30 s and with stimuli of 30-s duration presented every 60 s. They were given in runs of ascending stimulus intensities in 1°C steps from 39 to 53°C. The base temperature (38 or 35°C) was maintained for 1 min before the first stimulus and between stimuli. The sequence was stopped after the first evoked response occurred. A 2-min, stimulus-free interval between repeated threshold runs was previously found to yield reproducible results in C-fiber nociceptors (Treede et al. 1990) and was also used in this study. Fatigue was examined with use of three 49°C stimuli presented with a repetition period of 30 s.

The response of each nociceptor to an intense (53°C) heat stimulus of 30-s duration was used to classify its heat sensitivity in a manner similar to that done previously in mechanically sensitive afferents (Treede et al. 1995). The response latency was defined as the time between stimulus onset and the initiation of the first action potential. The propagation time along the nerve from the receptive field to the recording site (estimated with electrical stimuli at the receptive field) was subtracted to represent the times of action-potential initiation at the receptive field. We also determined the latency-to-peak firing. For this purpose, peristimulus frequency histograms were constructed in the following manner: the instantaneous action-potential frequency was calculated for each pair of action potentials and then the mean interspike frequencies were averaged over bins of 0.2-s duration (for details see Treede et al. 1995).

For statistical analysis of incidences, we used ²-tests. For group comparisons, nonparametric tests were used (Mann-Whitney U test, Wilcoxon matched-pairs test) because several variables were either not normally distributed or had only rank scale character. Standard deviations are used as indicators of variability despite nonnormal distributions. P-values of 0.05 and below were considered to be significant.

	RESULTS

Abstract Introduction Methods Results Discussion References

A total of 118 strands were investigated in 32 experiments that used the superficial radial (n = 26), medial antebrachial cutaneous (n = 3), and superficial peroneal (n = 3) nerves. Each strand contained 2.3 ± 1.4 (SD) A-fibers that could be activated from the nerve stimulation electrode. The conduction velocity distribution (Fig. 1) revealed two peaks, one at 15 m/s and the other at 43 m/s. The boundary between these two fiber groups was placed at the minimum between these two peaks, which was 30 m/s. Of the 263 fibers identified, 114 were classified as A-fibers and 149 as A-fibers.

View larger version (14K): [in this window] [in a new window]   FIG. 1. Distribution of conduction velocities for afferents included in this study. Conduction velocities determined from the nerve stimulation electrode were plotted in 5-m/s bins (n = 263). Lower cutoff was 2.5 m/s to exclude C-fibers. Distribution of A-fiber conduction velocities revealed 2 distinct peaks. Minimum between the peaks (30 m/s) was used to define the border between A- (n = 149) and A-fibers (n = 114). Filled bars correspond to the subpopulation of afferents for which the properties of the cutaneous terminals were determined.

The receptor type was identified for 77 A- and 79 A-fibers (Table 1). For the majority of A-fibers, a low-threshold mechanoreceptive terminal was identified. Nine of the 50 nociceptors were A-fibers, the fastest nociceptor having a conduction velocity of 70 m/s. Eight of these nine fibers exhibited a type I heat response.

Most cutaneous A-fiber nociceptors (43/50) responded to the range of heat stimuli used in this study. Of the seven heat-insensitive nociceptors, two responded to high-intensity von Frey probes and one to squeezing with small forceps (true high-threshold mechanoreceptors). Three of the seven fibers did not respond to any of the natural stimuli used in this study, and one responded vigorously to an ice stimulus but not to mechanical stimuli.

Four heat-insensitive fibers did not have a cutaneous receptive field and were excluded from further analysis. Their eRF did not move when the skin was moved. They either had a mechanical threshold of 18 bar (n = 2) or did not respond to this highest stimulus intensity used (n = 2). Adequate mechanical stimuli for deep afferents were joint or tendon compression near the metacarpophalangeal joint of the index finger, forced thumb movements, or compression of the web between thumb and index. The latter fiber exhibited a slowly increasing discharge to constant blunt pressure.

Classification of heat responses

Response latency and peak discharge latency to an intense heat stimulus (53°C, 30 s), which causes a burn injury, were used to classify the heat responses in a manner similar to that done by us previously to classify mechanically sensitive afferents (Treede et al. 1995). Examples of the two types of heat response that were observed in this study are shown in Fig. 2. In some fibers the latency to the first action potential was long (e.g., Fig. 2A; >6 s), the heat-evoked discharge increased with time in spite of constant skin temperature, and the peak discharge occurred near the end of the stimulus. This type I heat response was found in a few MIAs but was more typical for mechanically sensitive fibers. For other heat-sensitive afferents, the latency to the first action potential was short (e.g., Fig. 2B, 0.136 s), the heat-evoked discharge adapted during a constant heat stimulus, and the peak latency was <1 s. This type II heat response was found predominantly among the MIAs.

View larger version (16K): [in this window] [in a new window]   FIG. 2. Typical examples of the 2 types of heat response observed in the A-fiber nociceptors from monkey hairy skin. A: peristimulus frequency histogram (obtained with 0.2-s binwidth) of the response to the 53°C, 30-s stimulus for an A-fiber nociceptor in the hairy skin (mechanical threshold = 2.8 bar, heat threshold >53°C, 1 s, conduction velocity = 24.6 m/s, J56.5AA). Because of its long response latency (6.8 s) and long peak latency (27.1 s), this fiber was considered to have a type I heat response. B: peristimulus frequency histogram from another A-fiber in hairy skin (mechanical threshold = 7.3 bar, heat threshold = 47°C, 1 s, conduction velocity = 16 m/s, G60.2EF). Because of its short response latency (136 ms) and short peak latency (0.3 s), this fiber was considered to have a type II heat response. C: skin temperature recordings from the radiometer within the laser thermal stimulator.

A scatter plot of peak latency as a function of response latency is shown in Fig. 3A. Because the peak discharge cannot occur before the first action potential, all data points lie above the y = x diagonal. A type II heat response (; ) was exhibited by 13 of the 35 A-fiber nociceptors, for which a response to the 53°C, 30-s stimulus was recorded. The distribution of the type I heat responses (; ) suggests that there may be two subgroups with median response latencies of 0.9 s (Fig. 3A, top left; n = 12) and 9.7 s (Fig. 3A, top right; n = 10). The significant difference in response latency (P < 0.002, U test) accounted for a significant difference in the number of action potentials evoked by the heat stimulus (P < 0.01, U test). Heat threshold, latency-to-peak discharge, conduction velocity, mechanical threshold, and receptive field size, however, were not significantly different between these subgroups.

View larger version (15K): [in this window] [in a new window]   FIG. 3. Response of the A-fiber nociceptor population to the 53°C, 30-s heat stimulus. A: scatter plot of peak discharge latency vs. response latency for mechanically insensitive afferents (MIA; , , n = 16) and mechanically sensitive afferents (MSA; , , n = 19). Mechanical sensitivity was classified according to previously published operational criteria (Meyer et al. 1991). Heat response was classified in a manner similar to that used previously for mechanically sensitive nociceptors only (Treede et al. 1995). Thus receptors that had a long peak discharge latency were considered to have a type I heat response (, ). Receptors that had a short response latency and a peak discharge near the stimulus onset were considered to have a type II heat response (, ). The type II heat response was found more frequently in the MIA group (P  0.05, 2-test). B: average peristimulus frequency histogram (obtained with 0.2-s binwidth) of the response to the 53°C, 30-s stimulus for A-fiber nociceptors that had a type I heat response (n = 21). C: average peristimulus frequency histogram for A-fiber nociceptors that had a type II heat response (n = 12).

The distinguishing characteristic between the types I and II heat responses was the latency to the peak discharge (either <1.2 or >3 s). In addition to the 35 afferents presented in Fig. 3A, 8 other fibers were classified as having a type I or II heat response based on less intense heat stimuli. Of the 43 heat-responsive cutaneous A-fiber nociceptors, 26 (60%) were type I afferents, whereas 17 (40%) were type II afferents. These incidences were not significantly different (² = 1.88, df = 1).

The population responses of A-fiber nociceptors exhibiting either of the two heat response types are shown in Fig. 3, B and C. The slowly increasing discharge of the type I afferents contrasted with the marked adaptation of the type II afferents, many of which stopped discharging before the end of the constant heat stimulus. The adaptation of the type II afferents was fit by a single exponential with a time constant of 2.4 s, which turned into a plateau of <1 Hz after 5 s.

Mechanical thresholds of afferents with a type I or type II heat response

The distributions of mechanical thresholds for the cutaneous A-fiber nociceptors in this study are shown in Fig. 4. Fibers that did not respond to the 18-bar probe were considered to be unresponsive to mechanical stimuli (13/50 = 26%), although a few of these subsequently responded to pinching of the skin with forceps (3/10). Fibers with a mechanical threshold >6 bar (dashed line) were considered to be MIAs because ~95% of the A-fiber nociceptors in hairy skin in previous studies, which used mechanical stimuli to identify the receptive field, had thresholds <6 bar (Meyer et al. 1991). With the electrical search technique used in this study, more than one-half of the cutaneous A-fiber nociceptors were MIAs (29/50 = 58%).

View larger version (17K): [in this window] [in a new window]   FIG. 4. Mechanical threshold distribution for types I and II and heat-insensitive fibers. For the 50 cutaneous A-fiber nociceptors in this study, mechanical thresholds were determined for the most sensitive part of the receptive field with the use of von Frey probes (maximum: 18 bar, 716 mN). Histograms were obtained with 3-bar binwidth separately, according to the heat response of the fiber: A: type I response to heat (n = 26). B: type II response to heat (n = 17). C: nociceptors that were unresponsive to heat (n = 7). The vertical dashed line at 6 bar corresponds to the stimulus level used previously to separate both mechanically sensitive and insensitive afferents (Meyer et al. 1991). Mechanical thresholds for the fibers with a type II heat response were significantly higher than for fibers with a type I heat response (P  0.002, U test).

Notably, however, more than one-half of nociceptors with a type I heat response were mechanically sensitive (15/26 = 58%), whereas the majority of type II responses was observed in MIAs (13/17 = 76%). This difference in incidence was significant (P  0.05, ²-test). Seven of the type II nociceptors were unresponsive to the highest von Frey probe used, and five of them could only be activated by heat; the other two responded to squeezing with small forceps. Consistent with the above analysis, the median mechanical threshold for fibers with a type II heat response was significantly higher than for fibers with a type I response (15 bar vs. 4.9 bar; P  0.002, U test, Table 2).

Conduction velocities of afferents with a type I or type II heat response

All A-fiber nociceptors with a type II heat response had conduction velocities <30 m/s and thus all of them were considered A-fibers (Fig. 5B). In contrast, the conduction velocities of the type I afferents ranged from 8.2 to 70 m/s (Fig. 5A) and thus bracketed the A- and A-fiber classification. The mean conduction velocity of the type II afferents (14.2 ± 5.2 m/s) was significantly slower than the mean conduction velocity of the type I afferents (25.4 ± 15.7 m/s, P  0.05, U test, Table 2).

View larger version (18K): [in this window] [in a new window]   FIG. 5. Conduction velocity distributions for types I and II and heat-insensitive fibers. For the 50 cutaneous A-fiber nociceptors in this study, histograms of conduction velocity from the nerve stimulation electrode were obtained with 5-m/s binwidth separately, according to the heat response of the fiber: A: type I response to heat (n = 26). B: type II response to heat (n = 17). C: nociceptors that were unresponsive to heat (n = 7). The vertical dashed line at 30 m/s separates A- and A-fibers (see Fig. 1). All nociceptors with a type II heat response were A-fibers, whereas type I heat response was also observed in A-fibers.

Stimulus duration and heat thresholds of types I and II afferents

The median heat threshold for type II afferents for 1-s duration stimuli (48°C) was significantly lower than that of the type I afferents (>53°C, P  0.002, U test). Most (16/17) of the type II afferents responded to 1-s stimuli 49°C, whereas only 4 of the 26 type I afferents responded to such stimuli.

This threshold difference, however, strongly depended on stimulus duration. Three type I afferents and seven type II afferents had threshold determinations with 1-s, then 30-s, then again with 1-s duration stimuli. In contrast to the results with 1-s duration stimuli, the thresholds for the types I and II afferents were similar when the 30-s duration stimuli were used (Fig. 6, A and B). Expressed as temperature increase above the 38°C base, the median threshold decrease for 30-s versus 1-s stimulus duration was 53% for type I afferents versus 18% for type II afferents (P < 0.02, U test). The threshold of four type II afferents studied with a 38°C base temperature was contrasted with the threshold determined with a 35°C base temperature. In contrast to a previous study in C-fiber nociceptors (Tillman et al. 1995), there was no consistent effect of baseline temperature on threshold (Fig. 6C).

View larger version (16K): [in this window] [in a new window]   FIG. 6. Heat thresholds at different stimulus durations and baseline temperatures. Heat thresholds were determined with ascending series of stimuli. A run was terminated when 1 action potential was elicited or 49°C was reached. There was a 2-min, stimulus-free interval between runs. , MIAs; , MSAs. A: heat thresholds at different stimulus durations for A-fiber nociceptors with a type I heat response (n = 3, 1 MIA, 2 MSAs). Although the type I afferents typically had high heat thresholds when tested with 1-s duration stimuli, the thresholds for the 30-s duration stimuli were substantially lower. Expressing threshold as temperature increase above baseline (38°C), the thresholds differed by a factor of 2. The unchanged heat threshold for 1-s duration obtained after the long heat stimuli indicates that this threshold difference was not caused by sensitization. B: heat thresholds at different stimulus durations for A-fiber nociceptors with a type II heat response (n = 7, 4 MIAs, 3 MSAs). The thresholds for the 30-s duration stimuli were only slightly lower than heat threshold obtained with 1-s duration. C: heat thresholds at different baseline temperatures (1-s stimulus duration) for A-fibers with a type II heat response (n = 4, 2 MIAs, 2 MSAs). Heat thresholds at the 35°C base were not significantly different from the thresholds obtained before and after at the 38°C base.

It might be argued that the type I A-fiber nociceptors are actually high-threshold mechanoreceptors (Perl 1968) and that the responses to the long-duration heat stimuli represent sensitization. However, the threshold of the three type I afferents shown in Fig. 6A was stable over multiple runs (Fig. 7A). Moreover, the threshold to the 1-s duration stimulus was not altered by these tests (Fig. 6A). Thus the long-duration threshold sequence did not lead to heat sensitization. The slow response of the type I afferents to the 53°C, 30-s stimulus therefore reflects a slow transduction mechanism and not sensitization. Repeated testing of the type II afferents also revealed that their heat thresholds were stable (except for 1 fiber: Fig. 7B, ···).

View larger version (9K): [in this window] [in a new window]   FIG. 7. Reliability of A-fiber nociceptor heat thresholds. Heat thresholds were determined with ascending series of stimuli from a 38°C base. A run was terminated when 1 action potential was elicited or 49°C was reached. There was a 2-min, stimulus-free interval between runs. , MIA; , MSA. A: heat thresholds for type I afferents were determined with 30-s duration stimuli (n = 3, 1 MIA, 2 MSAs). B: heat thresholds for type II afferents were determined with 1-s duration stimuli (n = 8, 5 MIAs, 3 MSAs). Heat thresholds with this paradigm did not vary substantially with repeated runs for A-fiber nociceptors with either type of heat response except for one type II afferent (···). Thus heat thresholds of A-fiber nociceptors in monkey hairy skin in this study were as reliable as those of C-fiber nociceptors in a previous study (Treede et al. 1990).

Fatigue of type II afferents to repeated heat stimuli

Type II afferents exhibited a marked fatigue to repeated presentations of the same heat stimulus. As shown in Fig. 8A, the response to a second and third presentation of a 49°C stimulus was 40 ± 10% (SE) and 35 ± 13%, respectively, of the first response when the repetition interval was 30 s. For three fibers, the repetition interval was systematically varied (Fig. 8B). The amount of fatigue decreased as the repetition interval increased, but pronounced fatigue was still present at a repetition interval of 10 min (70 ± 7%).

View larger version (10K): [in this window] [in a new window]   FIG. 8. Fatigue of the type II heat response to repeated stimuli. A: average response of A-fiber nociceptors with a type II heat response to repeated presentations of a 49°C stimulus (repetition period 30 s, n = 9). For each fiber, the responses to trials 2 and 3 were normalized by dividing by the response to the 1st trial. There was no significant difference between the 6 MIAs and 3 MSAs (P > 0.2, Mann-Whitney U test). B: average response of type II afferents to pairs of 49°C stimuli with different intertrial intervals ranging from 15 to 600 s (10 min between pairs, n = 3, 1 MIA, 2 MSAs). Data were normalized by dividing by the response to the 1st stimulus in each pair. Fatigue was inversely proportional to the intertrial interval. Even at 10 min, the heat response was not fully recovered. Means ± SE.

Stimulus response functions for type II afferents

The stimulus response functions of four type II afferents were determined with the use of an ascending series of 1-s stimuli (Fig. 9). A monotonically increasing response was observed, indicating that type II afferents can code for stimulus intensity even in the presence of marked fatigue. As another estimate of the encoding capabilities of type II afferents, we evaluated the response to the first 49°C, 1-s stimulus and the response during the first 1 s of the 53°C, 30-s stimulus. A significantly greater response was seen at the higher temperature (P < 0.001, Wilcoxon matched pairs test, n = 12).

View larger version (11K): [in this window] [in a new window]   FIG. 9. Stimulus-response functions for A-fibers with a type II heat response. A: response of 4 A-fiber nociceptors with a type II heat response to an ascending series of heat stimuli (1-s duration, presented at a repetition period of 30 s). In each fiber, the response increased as a function of temperature, in spite of the known fatigue of these afferents.  and , MIAs;  and , MSAs. B: comparison of the population response to 49°C (very 1st heat stimulus applied to the receptive field, 1-s duration) and 53°C (response during the 1st second of the 53°C, 30-s stimulus presented after 10-min break). The 4°C difference in stimulus temperature was coded as a significant increase in number of action potentials (mean ± SE, P < 0.001, Wilcoxon matched pairs test, n = 12), which was not significantly different between the 8 MIAs and 4 MSAs (P > 0.1, Mann-Whitney U test).

Responses to a second burn

The intense heat stimulus (53°C, 30 s) was applied for a second time to the receptive fields of 15 A-fiber nociceptors. At least 25 min passed between these stimuli. In the nine type I afferents, the number of action potentials elicited increased from 104 ± 64 (SE) to 148 ± 48 (NS), response latency decreased from 7.0 ± 2.2 s to 4.7 ± 1.3 s (NS), and peak latency decreased significantly from 15.2 ± 2.9 s to 7.7 ± 1.7 s (P < 0.01, Wilcoxon matched-pairs tests). In other words, type I afferents continued to show a type I heat response in spite of some signs of sensitization. In the six type II afferents, the response to the second burn was identical to that to the first burn in the three mechanically sensitive fibers (thresholds 4.2-5.1 bar), but no response to the second burn was observed in the other three type II afferents that were MIAs (threshold >18 bar). In two nociceptors it was possible to apply a second burn to a part of the receptive field that was not tested with heat before. One fiber showed a type I heat response for both areas and the other a type II response for both areas. These data indicate that the type of heat response is a property of the afferent fiber under study.

Responses of low-threshold mechanoreceptors to a burn

The receptive fields of 13 low-threshold mechanoreceptors (6 A and 7 A) were also exposed to the 53°C, 30-s heat stimulus. Five of these fibers showed an OFF response to mild heat stimuli, which turned into no response (n = 2), an OFF response (n = 1), or a paradoxical ON response to the burn (n = 2). One fiber was a highly sensitive pacinian-like fiber (17 m/s conduction velocity), which responded with one or two action potentials to the onset of all heat stimuli including the 38°C base. None of these characteristics resembled those of the A-fiber nociceptor population described above. Seven low-threshold mechanoreceptors did not respond to any of the heat stimuli.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

A-fiber nociceptors are of two distinct types with regard to the response to heat stimuli. Type I afferents have a wind-up response to heat stimuli as manifest by the delay in peak latency of several seconds after stimulus onset. Type II afferents have an adapting response to heat stimuli similar to that seen in C-fiber nociceptors. Whereas we previously noted this difference in mechanically sensitive fibers (Treede et al. 1995), the present study establishes that this same dichotomy of heat response types applies to MIAs.

Notably, however, the mechanical sensitivity of types I and II A-fiber nociceptors was markedly different. Type II afferents had significantly higher mechanical thresholds than did type I afferents. A large proportion (76%) of type II afferents met the criterion for being mechanically insensitive (Meyer et al. 1991). Moreover, a substantial number of them (41%) had no response to the highest von Frey force used in this study.

This finding resolves one of the paradoxes regarding peripheral neural mechanisms of pain sensation. The conduction velocity and heat response characteristics of type II A-fiber nociceptors suggest strongly that these afferents signal the first pain sensation evident with stepped heat stimuli. In earlier studies, however, there were only anecdotal reports about type II afferents (Adriaensen et al. 1983; Bromm and Treede 1984; Dubner et al. 1977; Perl 1968). Even our systematic study of mechanically sensitive afferents (Treede et al. 1995) identified only 28% of the hairy skin A-fiber nociceptors as type II, significantly less than type I (² = 8.70, P < 0.01). Given that first pain sensation is an easily demonstrated sensation in hairy skin (1st pain sensation is not evoked by stimulation of the glabrous skin of the hand and type II afferents are not evident in this skin) (Campbell and LaMotte 1983), the dilemma was how this sensation could be evoked so easily if the substrate for this sensation, the type II afferents, was so few in number.

It appears from the current study that type II afferents were infrequently observed previously because of their relative mechanical insensitivity. In this study we used an electrocutaneous search technique to obtain an unbiased sample of the total A-fiber population and to identify the receptive fields independent of the adequate stimulus. With this technique the incidence of types I and II afferents approached equality (60 vs. 40%, NS). As discussed in the following section, the true incidence of type II afferents may be even higher.

Search technique

The search technique used in this study differs from many previous studies in two aspects: 1) the receptor characteristics were determined for all A-fibers identified after electrical nerve stimulation, leading to an unbiased sampling of all fibers (see Fitzgerald and Lynn 1977; Leem et al. 1993 for the rat); and 2) an electrocutaneous search technique allowed the receptive field to be identified independent of the adequate stimulus (see Schmidt et al. 1995 for human C-fiber nociceptors).

Adequate sampling of all sizes of myelinated afferents is indicated by the fact that we found about the same percentage of A-fibers (149/263 = 57%) as was reported in histological studies of the human radial nerve (56 ± 9%, SD) (O'Sullivan and Swallow 1968). We were able to identify and study the receptive fields of 77 of 114 A-fibers (68%) and 79 of 149 A-fibers (53%). Several factors contribute to the relatively high proportion of receptive fields not identified. 1) Several afferents may have been disconnected from their terminal because of the extensive surgical preparation that was necessary for placing stimulation and recording electrodes. In fact, several receptive fields were immediately adjacent to the well, and other fibers could not be followed electrically outside the well. 2) Some fibers had receptive fields that were inaccessible for the laser stimulator. 3) Fibers that innervated previously damaged skin were not studied beyond detecting their eRF. For the low-threshold mechanoreceptor population the standard mechanical search procedure is adequate. In our study, rapidly adapting receptors outnumbered slowly adapting ones (62 vs. 38%) by a similar ratio as in previous studies in monkey hairy skin (69 vs. 31%) (Merzenich and Harrington 1969).

The value of the electrocutaneous search technique is emphasized by comparing the current results with some earlier studies. With the use of mechanical stimuli as the means to search for the receptive field of nociceptors in monkey hairy skin, only 8 of 124 A-fibers had thresholds 100 mN (equivalent to 6 bar, the point chosen as the boundary between MIAs and mechanically sensitive afferents) (Meyer et al. 1991). In contrast, 58% of the cutaneous nociceptors were in the MIA range in the present study. Other studies in rat and rabbit (Fitzgerald and Lynn 1977; Lynn and Carpenter 1982) explored a threshold range up to 50-100 mN. Thresholds in the only extensive study in humans (Adriaensen et al. 1983) covered a range of 20-220 mN, but again 88% of the values were <100 mN. In spite of anecdotal reports of mechanically unexcitable A-fiber nociceptors with a type II heat response (Meyer et al. 1991; Perl 1968), previous studies did not find the significant difference in mechanical thresholds between types I and II afferents (Dubner et al. 1977; Treede et al. 1995). Therefore adequate sampling of the population of A-fiber nociceptors with a type II heat response requires the electrocutaneous search technique.

The electrocutaneous search also enabled us to identify the receptive fields of deep units in the superficial radial nerve. All four deep units were unresponsive to heat stimuli. One unit exhibited a slowly increasing discharge to constant blunt pressure (see Garell et al. 1996) and may account for the slowly increasing pain in humans during skin fold stimulation at forces 4 N (Adriaensen et al. 1984).

Populations of A-fiber nociceptors in monkey hairy skin

Although the numbers are small, the data from this study also provide information regarding the relative incidence of nociceptors in the cutaneous A-fiber population as a whole (at least for the hairy skin of the distal extremity). Approximately 33% of the A-fibers were nociceptive and the majority of these was heat sensitive. Approximately 7% of the A-fibers were cold fibers, and the remaining 60% were low-threshold mechanoreceptors, of which the majority were quickly adapting.

Traditionally, A-fiber nociceptors were considered to be mainly high-threshold mechanoreceptors (Lynn and Carpenter 1982; Perl 1968). The current data argue against this classification because only three fibers were found that fit this category. The majority (86%) of A-fiber nociceptors in this unbiased sample from monkey hairy skin responded to heat stimuli. This percentage was even higher than the proportion of nociceptors that responded to mechanical stimuli (74%). Most of the nociceptors with a type I heat response would have been classified as high-threshold mechanoreceptors if the long-duration heat stimulus was not used. The current data show that the adequate heat test stimulus may be of an intensity that does not cause primary afferent sensitization. In addition to their mechanical and heat sensitivity, A-fiber nociceptors were shown to respond more vigorously to chemical stimuli than C-fiber nociceptors (Davis et al. 1993). Thus A-fiber nociceptors predominantly are polymodal nociceptors.

If 30 m/s (see RESULTS for rationale) is considered the boundary between A- and A-fibers, 82% of myelinated nociceptive afferents were in the A-range. Here again, however, there was a significant difference between the types I and II afferents; the mean conduction velocity of the type I afferents approached the boundary between A and A, and 31% of the type I afferents were A-fibers. In contrast, in this sample all type II afferents were A-fibers. Because A-fibers tend to have high electrical thresholds and the mechanical thresholds of type II afferents were also high, type II afferents may constitute a high percentage of those A-fibers for which a receptive field was not found.

Reliability and validity of the types I and II classification

We previously reported that cutaneous injury can sensitize the type I heat response and desensitize the type II response (Treede et al. 1995). One may thus hypothesize that 1) the type I response is a fatigued type II response or 2) the type II response is a sensitized type I response. A fatigued type II heat response loses its initial peak and might thus look similar to a type I heat response (see Fig. 2 in Treede 1995). However, most of the type I afferents did not respond to heat stimuli before the 53°C, 30-s stimulus used to classify their response, and thus fatigue was not possible. Conversely, one may imagine that a sensitized type I response could look like a type II response; this was never observed in this study. In two fibers we were able to classify the type of heat response independently for two different parts of the receptive field, resulting in the same classification for both parts. All these findings taken together indicate that the type of heat response is a property of the primary afferent nerve fiber.

Characteristics of the type I heat response

An important finding in this study is that the heat threshold for type I afferents is not as high as originally reported. Whereas most type I afferents do not respond to a 53°C, 1-s heat stimulus, the three afferents that were tested with an ascending threshold sequence that used long-duration (30-s) stimuli had much lower thresholds. In addition their threshold response was reproducible over multiple runs. Thus type I afferents appear to have a slow response not only to the intense heat stimulus used to classify them but also to less intense stimuli. These results also indicate that a sensitizing stimulus is not needed for these afferents to respond to heat.

Possible explanations of the slow heat response in type I afferents include the following: 1) temperature conduction within the skin could lead to a long response latency if the fiber terminals are situated deep in the skin; 2) the apparent heat response is actually a response to chemical mediators released from nonneuronal cells by the intense, tissue-injuring heat stimulus; and 3) type I afferents have a slow heat transduction mechanism.

DEPTH OF RECEPTOR. The 53% difference in heat threshold between stimuli of 1- and 30-s duration would be compatible with a depth of ~700 µm if these afferents had the same transduction mechanism as C-fiber mechanoheat nociceptors (CMHs), i.e., depended only on local temperature at the nerve terminal (see Tillman et al. 1995). Intracutaneous temperature measurements, however, showed that the skin reaches thermal equilibrium ~5-10 s after stimulus onset (Treede et al. 1995), whereas most (15/22) of the nociceptors with a type I heat response exhibited their peak discharge much later than 10 s, and three fibers did not even start to generate action potentials until after 10 s. Thus receptor depth would not account for the late peak in discharge frequency. Although the radial nerve, which was used for most of the experiments in this study, innervates some subcutaneous tissue (4/79 = 5% fibers in this study), the mechanical receptive field structure of fibers with a type I heat response was not compatible with such a deep location. The majority (15/26) of type I afferents responded to von Frey probes of 1.3-5.1 bar (2.4-51 mN). In contrast, the four deep receptors with receptive fields that did not move with the skin had very high mechanical thresholds to von Frey probes and did not respond to heat stimuli at all.

SENSITIZATION RESPONSE. The long latency of the heat response may indicate involvement of a chemical mediator that is released from damaged tissue. Whereas the 53°C, 30-s stimulus does cause tissue damage, this is not the case for the 42.5-45.5°C stimuli to which the three fibers shown in Fig. 7A exhibited a reproducible response. Moreover, the threshold of these fibers to subsequent 1-s stimuli was not lowered (see Fig. 6A), and thus these long-duration stimuli of threshold intensity did not lead to sensitization. Also, psychophysically, the induction of primary hyperalgesia to heat in human hairy skin requires stimuli of 50°C for 60 s or 48°C for 120 s, whereas 50°C for 20 s is ineffective (Davis et al. 1995; LaMotte et al. 1983). Therefore to explain the type I heat response by a chemical mediator this chemical must be released by very mild heat stimuli that do not produce sensitization or primary hyperalgesia.

SLOW TRANSDUCTION MECHANISM. The long response latency most likely is caused by a slow transduction mechanism within the nerve terminals. An element of this slow transduction mechanism is a wind-up of the action-potential response. If the opening of a cation channel leads to the initial response to heat, the initial influx of cations may perpetuate a further opening of the same or other cation channels. Alternatively, heat transduction in these terminals may involve a slow intracellular signal cascade.

Characteristics of the type II heat response

The heat response of type II A-fiber nociceptors is similar to the heat response of C-fiber nociceptors in many ways. Type II afferents exhibit a rapid adaptation of their response to the 53°C, 30-s stimulus (time constant of 2.4 s) and a sustained response in the adapted state of <10% of the peak. Polymodal C-fiber nociceptors exhibit a similar rate of adaptation ( = 2.5 s) (Treede 1995) and amount of sustained response (Meyer and Campbell 1981a). In addition, both C-fiber and A-fiber nociceptors with a type II heat response exhibit fatigue to repeated heat stimuli; for both, the response is suppressed by about a factor of three at an interstimulus interval of 30 s. Whereas C-fiber nociceptors in glabrous skin fully recover between 4 and 10 min (LaMotte and Campbell 1978) and those in hairy skin are 90% recovered after 10 min (Tillman 1992), A-fiber nociceptors with a type II response were only 70% recovered after 10 min (this study). Adaptation and fatigue may occur in the process of transduction of heat into the generator potential and/or in the process of transformation into trains of action potentials. Recently, dissociated dorsal root ganglion neurons, which are the somata of the primary sensory neurons that innervate the skin, were used to study heat transduction mechanisms with whole cell voltage-clamp techniques (Baumann and Martenson 1994; Cesare and McNaughton 1996; Kirschstein et al. 1997). There is preliminary evidence that the heat-evoked inward currents show adaptation when a heat pulse of 49°C is applied for 3 s (P. McNaughton, personal communication). On the other hand, nociceptive A-fibers have a limited capacity to transmit high action-potential rates on electrical stimulation when compared with cold fibers of the same conduction velocity (Raymond et al. 1990). Slowing of conduction velocity and conduction failure may contribute to the overall adaptation.

It was recently suggested that the heat thresholds of A-fiber nociceptors depend on the rate of temperature rise (Yeomans and Proudfit 1996). In contrast, the first action-potential threshold of C-fiber nociceptors depends only on local temperature at the nerve terminal, not on its rate of change (Tillman et al. 1995; Yarnitsky et al. 1992). For C-fiber nociceptors in monkey hairy skin, differences in the heat thresholds obtained with different stimulus parameters (base temperature, rate of temperature change, and duration) could be explained by an average depth of their terminals of 150 µm and by the different spatial temperature gradients created by these stimuli (Tillman et al. 1995). The 18% difference in heat threshold between stimuli of 1- and 30-s duration would be compatible with a depth of ~150 µm for the terminals of type II A-fiber nociceptors. From a lower base temperature (35°C instead of 38°C), we would expect a heat threshold that is ~1°C higher because a higher skin surface temperature is necessary to reach the same temperature at 150 µm. Such a difference in surface temperature at threshold between the two base temperatures, however, was not observed for type II A-fiber nociceptors (this study) in contrast to C-fiber nociceptors (Tillman et al. 1995). Thus from the lower base temperature a lower intracutaneous temperature was sufficient to reach threshold. Because the size of the temperature step was larger from the lower base temperature, a rate dependence of A-fiber nociceptor heat threshold could explain this behavior.

The heat response of type II A-fiber nociceptors increased with stimulus intensity, and thus these afferents may play a role in the perceived intensity of noxious heat stimuli. In addition, the threshold distribution of these afferents in monkey matches those of pricking pain and (A-fiber mediated) laser-evoked potentials in humans (Treede et al. 1994). Thus these fibers likely signal pricking pain to heat.

Heat transduction or chemical transduction?

The polymodality of nociceptors suggests that tissue damage and release of inflammatory mediators could be the common pathway for the transduction mechanism of nociception. In this view, nociceptors are considered to be chemoreceptors. However, the current data provide evidence against this concept. Although the slow response of type I afferents to the intense heat stimulus (53°C, 30 s) would be compatible with the concept of chemical mediators, a similar response could be obtained with stimulus intensities that do not damage the skin. The type II heat response is too fast for this mechanism and is likely related to the heat-evoked inward currents that were identified in some acutely dissociated dorsal root ganglion neurons (Cesare and McNaughton 1996; Kirschstein et al. 1997). The range of thresholds of A-fiber nociceptors is very broad, both for mechanical and for heat stimuli. But, even for those A-fibers that had very high thresholds for one stimulus modality, actual tissue damage does not seem to be the adequate stimulus because 1) about one-half of the fibers that did not respond to our strongest mechanical stimulus (7/15) had a type II heat response and 2) most of the fibers with a type I heat response had moderate mechanical thresholds <6 bar (15/26).

In conclusion, A-fiber nociceptors are polymodal and have two distinct types of response to heat. Type I afferents have a slow, wind-up response to heat. These fibers sensitize to injurious heat stimuli, have rapid conduction velocities (at times in the A-range), and generally have mechanical thresholds <6 bar (100 mN). Type I afferents may be primarily responsible for signaling sharpness (Garell et al. 1996) and/or pricking pain to mechanical stimuli and were shown previously to account for the heat hyperalgesia to a burn injury (Meyer and Campbell 1981b). The type II afferents have responses to heat that resemble those of C-fiber nociceptors: early peak frequency and a slowly adapting response. These fibers have slower conduction velocities, always in the A-range, and are characteristically insensitive to mechanical stimuli (thresholds >6 bar). The short response latency, low heat threshold, and graded heat response of type II afferents suggest that their primary function may be to signal heat-induced pain. With little question these fibers signal first pain sensation. Both types I and II afferents respond to chemical stimuli (Davis et al. 1993; Ringkamp et al. 1997) and both likely participate in chemogenic pain. Because A-fiber nociceptors appear to exhibit a higher degree of stimulus specificity than do C-fiber nociceptors, they may provide the most important contribution to discrimination of the quality of a noxious stimulus. A population code for stimulus quality was studied most extensively in the gustatory system, where each primary afferent can be activated by prototypical stimuli for each of the four taste qualities (sour, salty, sweet, and bitter), but each fiber exhibits its lowest threshold to one of the qualities. The differential sensitivity spectra of nociceptive afferents suggest a similar population code for pain quality.

	ACKNOWLEDGEMENTS

  We greatly appreciate helpful discussions with Drs. Karen D. Davis and Srinivasa N. Raja in the course of the experiments and the technical assistance of T. Hartke and S. Horasek.

  This research was supported by National Institute of Neurological Disorders and Stroke Grant NS-14447, by Deutsche Forschungsgemeinschaft Grant Tr 236/1-1, and by North Atlantic Treaty Organization Grant CRG 95032540495.

	FOOTNOTES

  Address for reprint requests: R. A. Meyer, Applied Physics Laboratory, The Johns Hopkins University, Johns Hopkins Rd., Laurel, MD 20723.

  Received 27 February 1998; accepted in final form 11 May 1998.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• ADRIAENSEN, H., GYBELS, J., HANDWERKER, H. O., VAN HEES, J. Response properties of thin myelinated (A-) fibers in human skin nerves. J. Neurophysiol. 49: 111-122, 1983.[Free Full Text]
• ADRIAENSEN, H., GYBELS, J., HANDWERKER, H. O., VAN HEES, J. Nociceptor discharges and sensations due to prolonged noxious mechanical stimulationa paradox. Human Neurobiol. 3: 53-58, 1984.[Medline]
• BAUMANN, T. K., MARTENSON, M. E. Thermosensitivity of cultured trigeminal neurons. Soc. Neurosci. Abstr. 20: 1379, 1994.
• BROMM, B., TREEDE, R.-D. Nerve fibre discharges, cerebral potentials and sensations induced by CO₂ laser stimulation. Human Neurobiol. 3: 33-40, 1984.[Medline]
• BURGESS, P. R., PERL, E. R. Myelinated afferent fibres responding specifically to noxious stimulation of the skin. J. Physiol. (Lond.) 190: 541-562, 1967.[Abstract/Free Full Text]
• CAMPBELL, J. N., LAMOTTE, R. H. Latency to detection of first pain. Brain Res. 266: 203-208, 1983.[Medline]
• CAMPBELL, J. N., MEYER, R. A. Sensitization of unmyelinated nociceptive afferents in monkey varies with skin type. J. Neurophysiol. 49: 98-110, 1983.[Free Full Text]
• CESARE, P., MCNAUGHTON, P. A novel heat-activated current in nociceptive neurons and its sensitization by bradykinin. Proc. Natl. Acad. Sci. USA 93: 15435-15439, 1996.[Abstract/Free Full Text]
• DAVIS, K. D., MEYER, R. A., CAMPBELL, J. N. Chemosensitivity and sensitization of nociceptive afferents that innervate the hairy skin of monkey. J. Neurophysiol. 69: 1071-1081, 1993.[Abstract/Free Full Text]
• DAVIS, K. D., MEYER, R. A., TURNQUIST, J. L., FILLOON, T. G., PAPPAGALLO, M., CAMPBELL, J. N. Cutaneous pretreatment with the capsaicin analog NE-21610 prevents the pain to a burn and subsequent hyperalgesia. Pain 62: 373-378, 1995.[Medline]
• DUBNER, R., PRICE, D. D., BEITEL, R. E., HU, J. W. Peripheral neural correlates of behavior in monkey and human related to sensory-discriminative aspects of pain. In: Pain in the Trigeminal Region, edited by D. J. Anderson, and B. Matthews Amsterdam: Elsevier/North-Holland, 1977, p. 57-66.
• FITZGERALD, M., LYNN, B. The sensitization of high threshold mechanoreceptors with myelinated axons by repeated heating. J. Physiol. (Lond.) 365: 549-563, 1977.
• GARELL, P. C., MCGILLIS, S.L.B., GREENSPAN, J. D. Mechanical response properties of nociceptors innervating feline hairy skin. J. Neurophysiol. 75: 1177-1189, 1996.[Abstract/Free Full Text]
• GEORGOPOULOS, A. P. Functional properties of primary afferent units probably related to pain mechanisms in primate glabrous skin. J. Neurophysiol. 39: 71-83, 1976.[Abstract/Free Full Text]
• HALLIN, R. G., TOREBJÖRK, H. E. Activity in unmyelinated nerve fibers in man. Adv. Neurol. 4: 19-27, 1974.
• KIRSCHSTEIN, T., BÜSSELBERG, D., TREEDE, R.-D. Coexpression of heat-evoked and capsaicin-evoked inward currents in acutely dissociated rat dorsal root ganglion neurons. Neurosci. Lett. 231: 33-36, 1997.[Medline]
• LAMOTTE, R. H., CAMPBELL, J. N. Comparison of responses of warm and nociceptive C-fiber afferents in monkey with human judgments of thermal pain. J. Neurophysiol. 41: 509-528, 1978.[Abstract/Free Full Text]
• LAMOTTE, R. H., THALHAMMER, J. G., ROBINSON, C. J. Peripheral neural correlates of magnitude of cutaneous pain and hyperalgesia: a comparison of neural events in monkey with sensory judgments in human. J. Neurophysiol. 50: 1-26, 1983.[Free Full Text]
• LEEM, J. W., WILLIS, W. D., WELLER, S. C., CHUNG, J. M. Differential activation and classification of cutaneous afferents in the rat. J. Neurophysiol. 70: 2411-2424, 1993.[Abstract/Free Full Text]
• LYNN, B., CARPENTER, S. E. Primary afferent units from the hairy skin of the rat hind limb. Brain Res. 238: 29-43, 1982.[Medline]
• MERZENICH, M. M., HARRINGTON, T. The sense of flutter-vibration evoked by stimulation of the hairy skin of primates: comparison of human sensory capacity with the responses of mechanoreceptive afferents innervating the hairy skin of monkey. Exp. Brain Res. 9: 236-260, 1969.[Medline]
• MEYER, R. A., CAMPBELL, J. N. Evidence for two distinct classes of unmyelinated nociceptive afferents in monkey. Brain Res. 224: 149-152, 1981a.[Medline]
• MEYER, R. A., CAMPBELL, J. N. Myelinated nociceptive afferents account for the hyperalgesia that follows a burn to the hand. Science 213: 1527-1529, 1981b.[Abstract/Free Full Text]
• MEYER, R. A., DAVIS, K. D., COHEN, R. H., TREEDE, R.-D., CAMPBELL, J. N. Mechanically insensitive afferents (MIAs) in cutaneous nerves of monkey. Brain Res. 561: 252-261, 1991.[Medline]
• MEYER, R. A., RAJA, S. N., CAMPBELL, J. N. Coupling of action potential activity between unmyelinated fibers in the peripheral nerve of monkey. Science 227: 184-187, 1985.[Abstract/Free Full Text]
• MEYER, R. A., WALKER, R. E., MOUNTCASTLE, V. B. A laser stimulator for the study of cutaneous thermal and pain sensations. IEEE Trans. Biomed. Eng. 23: 54-60, 1976.[Medline]
• O'SULLIVAN, D. J., SWALLOW, M. The fibre size and content of the radial and sural nerves. J. Neurol. Neurosurg. Psychiatry 31: 464-470, 1968.[Free Full Text]
• PERL, E. R. Myelinated afferent fibres innervating the primate skin and their response to noxious stimuli. J. Physiol. (Lond.) 197: 593-615, 1968.[Abstract/Free Full Text]
• RAYMOND, S. A., THALHAMMER, J. G., POPITZ-BERGEZ, F., STRICHARTZ, G. R. Changes in axonal impulse conduction correlate with sensory modality in primary afferent fibers in the rat. Brain Res. 526: 318-321, 1990.[Medline]
• RINGKAMP, M., PENG, Y. B., CAMPBELL, J. N., MEYER, R. A. Intradermal capsaicin produces a vigorous discharge in mechanically-insensitive A-fiber nociceptors of the monkey. Soc. Neurosci. Abstr. 23: 1258, 1997.
• SCHMELZ, M., FORSTER, C., SCHMIDT, R., RINGKAMP, M., HANDWERKER, H. O., TOREBJÖRK, H. E. Delayed responses to electrical stimuli reflect C-fiber responsiveness in human microneurography. Exp. Brain Res. 104: 331-336, 1995.[Medline]
• SCHMIDT, R., SCHMELZ, M., FORSTER, C., RINGKAMP, M., TOREBJÖRK, E., HANDWERKER, H. Novel classes of responsive and unresponsive C nociceptors in human skin. J. Neurosci. 15: 333-341, 1995.[Abstract]
• TILLMAN, D.-B. Heat Response Properties of Unmyelinated Nociceptors (PhD dissertation). Baltimore, MD: Johns Hopkins University, 1992.
• TILLMAN, D.-B., TREEDE, R.-D., MEYER, R. A., CAMPBELL, J. N. Response of C fibre nociceptors in the anaesthetized monkey to heat stimuli: estimates of receptor depth and threshold. J. Physiol. (Lond.) 485: 753-765, 1995.[Abstract/Free Full Text]
• TREEDE, R.-D. Peripheral acute pain mechanisms. Ann. Med. 27: 213-216, 1995.[Medline]
• TREEDE, R.-D., MEYER, R. A., CAMPBELL, J. N. Comparison of heat and mechanical receptive fields of cutaneous C-fiber nociceptors in monkey. J. Neurophysiol. 64: 1502-1513, 1990.[Abstract/Free Full Text]
• TREEDE, R.-D., MEYER, R. A., LESSER, R. P. Similarity of threshold temperatures for first pain sensation, laser-evoked potentials and nociceptor activation. In: Proceedings of the 7th World Congress on Pain. Progress in Pain Research and Management, edited by G. F. Gebhart, D. L. Hammond, and T. S. Jensen Seattle, WA: IASP, 1994, vol. 2, p. 857-865.
• TREEDE, R.-D., MEYER, R. A., RAJA, S. N., CAMPBELL, J. N. Evidence for two different heat transduction mechanisms in nociceptive primary afferents innervating monkey skin. J. Physiol. (Lond.) 483: 747-758, 1995.[Abstract/Free Full Text]
• YARNITSKY, D., SIMONE, D. A., DOTSON, R. M., CLINE, M. A., OCHOA, J. L. Single C nociceptor responses and psychophysical parameters of evoked pain: effect of rate of rise of heat stimuli in humans. J. Physiol. (Lond.) 450: 581-592, 1992.[Abstract/Free Full Text]
• YEOMANS, D. C., PROUDFIT, H. K. Nociceptive responses to high and low rates of noxious cutaneous heating are mediated by different nociceptors in the rat: electrophysiological evidence. Pain 68: 141-150, 1996.[Medline]

0022-3077/98 $5.00 Copyright ©1998 The American Physiological Society