Axonal thermosensitivity and mechanosensitivity of cutaneous afferent neurons

We hypothesized that cutaneous afferent myelinated fibers (A-fibers) and afferent unmyelinated fibers (C-fibers) respond to the same natural stimuli applied to their axons as to their terminals in the skin. In anesthetized rats, activity was recorded from afferent axons in strands isolated proximally from the sural nerve. Mechanical, cold or heat stimuli were applied to the skin or along a 15-mm length of the distal sural nerve. One-hundred and eighteen A-fibers and 109 C-fibers were characterized by their conduction velocity and/or shape of their action potentials, and by their responses to natural stimulation of the skin. Then, these fibers were tested for their responses to the same stimuli applied to the nerve. In some cases, the nerve was crushed distally after the nerve fibers had been characterized by their responses to physiological stimulation of the skin, and the responses to stimuli applied to the nerve proximal to the lesion were tested again. Almost all non-nociceptive cold-sensitive (type 1) C-fibers (97%) could be activated by cold stimuli applied to the nerve. Of nociceptive cold-sensitive (type 2) C-fibers, 39% were activated by cold stimuli applied to the nerve. Furthermore, 34% of heat-sensitive C-fibers could be activated by heating the nerve. In contrast, only 2-4% of mechanosensitive A-fibers and C-fibers responded to mechanical stimuli applied to the nerve. In conclusion, cold and heat sensitivity of cutaneous afferent neurons is not restricted to their terminals in the skin, but often extends along the axons in the nerve. Mechanosensitivity is restricted to the afferent endings in the skin.     

Antidromic nerve stimulation in monkey does not sensitize unmyelinated nociceptors to heat

Spread of sensitization of nociceptors has been proposed as the mechanism for the secondary hyperalgesia in the region that surrounds a cutaneous injury. One possibility, tested in this study, is that antidromically propagated action potentials in nociceptors result in the release of substances at their cutaneous arborization which sensitize adjacent nociceptors. Using standard teased-fiber techniques, we recorded from single C-fiber nociceptive afferents innervating hairy skin in the monkey. Heat testing was performed before and after electrical stimulation of the parent nerve at strengths sufficient to activate C-fibers. The heat thresholds and total response to the heat sequences were not changed by the electrical stimulation. We conclude that antidromic stimulation of C-fibers in the monkey does not sensitize C-fiber nociceptors to heat stimuli.     

Ectopic activity in cutaneous regenerating afferent nerve fibers following nerve lesion in the rat

Spontaneous activity, and mechanical and thermal sensitivity were investigated in regenerating afferent nerve fibers within 4-21 days post sural nerve lesion (crush or transection and resuturing) in anaesthetized rats. About 33-40% of the myelinated (A) and 22-27% of the unmyelinated (C) fibers excited by electrical nerve stimulation exhibited at least one of these ectopic discharge properties. In total 177 A- and 169 C-fibers with ectopic activity were analysed. Most A-fibers (161/177) were mechanosensitive. Spontaneous activity (median 1 imp/s) was present in 23/177 and thermosensitivity in 14/177 A-fibers (13 of them being activated by heat stimuli). Almost all A-fibers (159/177) exhibited only one type of ectopic discharge property. Most C-fibers (94/169) were thermosensitive responding either to cold (n = 45) or to heat stimuli (n = 33) or to both (n = 16). Eighty-four of 169 C-fibers were spontaneously active (median 0.3 imp/s) and 75/169 C-fibers were mechanosensitive. Both the proportion and the discharge rate of spontaneously active C-fibers were significantly higher after crush than after section and resuturing of the nerve. About 60% of the C-fibers (101/169) had only one ectopic discharge property and 40% two or three. In conclusion, regenerating cutaneous afferent A- and C-fibers may develop mechano- and/or thermosensitivity as well as spontaneous activity. We suggest that spontaneous and evoked ectopic activity in regenerating cutaneous afferents are a function of the intrinsic functional properties of these neurons and of the interaction between the regenerating nerve fibers and non-neural cells during Wallerian degeneration in the nerve distal to the nerve lesion.     

Effects of capsaicin applied to a peripheral nerve on the responses of primate spinothalamic tract cells

Capsaicin is a neurotoxin that appears to affect unmyelinated nociceptive sensory fibers selectively. We examined the effects of capsaicin applied topically to the sural nerve on peripheral nerve volleys and on the responses of neurons belonging to the spinothalamic tract (STT) in the monkey. The responses examined included those following electrical stimulation of the sural nerve and also those produced by more natural forms of noxious and innocuous stimuli applied to the skin. Capsaicin (1% solution) applied onto the sural nerve for 15 min resulted in a reduction of the sizes of A delta- and C-fiber afferent volleys. These changes paralleled the reduction of A- and C-fiber responses of the STT cells elicited by electrical stimulation of the sural nerve. During capsaicin application onto the sural nerve, the background activity of STT cells increased for 5-10 min. After capsaicin treatment, the responses of STT cells to innocuous mechanical stimuli applied to the cutaneous receptive field were increased, whereas the responses to noxious mechanical stimuli were decreased. However, topical capsaicin application almost eliminated the responses of STT cells to noxious heat stimuli. The results of the present study suggest that topical capsaicin application onto a peripheral nerve produces a transient nociceptive response followed by a decrease in sensitivity to noxious stimuli, particularly to noxious heat. These changes are due to conduction block of the nerve fibers at the site of capsaicin application.     

Identification of afferent C units in intact human skin nerves

Single unit potentials were recorded with microelectrodes from intact human skin nerves, and the unitary responses to electrical and natural skin stimuli were studied. The unitary discharges were derived from afferent C fibres since the impulses were conducted at C velocity and persisted after preferential blocking of activity in myelinated fibres by nerve compression, whereas they were abolished before the A fibre discharges by lidocaine.The afferent C units adapted slowly to mechanical stimuli and some of them exhibited afterdischarges following withdrawal of the stimuli. The excitability decreased as a consequence of repeated mechanical stimulation. The most vigorous responses were elicited by various intense stimuli such as pinpricks or heat stimulation. Although no definite classification was made, the responses of the human C units to different stimuli reminded of responses in ‘polymodal’ C receptors identified in the cat and the monkey.Painful stimuli elicited the most intense discharges in the units, but considerable activity was also elicited by non-painful stimuli. Some evidence suggested that single or a few repeated impulses in an afferent C unit need not reach consciousness.     

Reactions of cardiac postganglionic sympathetic neurons to movements of normal and inflamed knee joints

The effects of passive movements of normal and inflamed knee joints on unitary activity in filaments of the inferior cardiac nerve (ICN) were studied in cats anesthetized with chloralose and urethane. The effects were compared with those obtained by electrical stimulation of afferent A- and C-fibers in the medial articular nerve, in muscle and in cutaneous hind limb nerves. The vagus nerves were cut and the right carotid artery was tied off. The left carotid sinus was intact. All ICN units used in this study displayed spontaneous activity which was usually related to the cardiac and respiratory rhythms. The ICN units were regularly excited by electrically evoked single or short repetitive A-volleys in articular, cutaneous and muscle nerves. The excitation was followed by a silent period. Inclusion of C-fibers in the afferent volleys gave a second, long-latency burst of impulses which was seen only with short repetitive stimulation. Passive movements in the normal working range of the joint did not influence the activity of ICN units. However, noxious joint movements, particularly of inflamed joints, led to pronounced excitation of ICN units accompanied by rises in blood pressure. Most of these effects could still be seen after all nerves to the hind limbs, except the medial articular nerve, were cut. It is proposed (a) that ICN units form a homogeneous population of sympathetic postganglionic units whose reaction pattern to somatovisceral input is distinctly different from that of other sympathetic subsystems, and (b) that articular receptors make a substantial contribution to the ICN input particularly when many fine afferent units are sensitized to mechanical stimulation by an acute joint inflammation.     

Does sympathetic nerve discharge affect the firing of myelinated cutaneous afferents in humans?

In clinical practise, the notion that some complex regional pain syndromes (CRPS) are associated with sympathetic facilitation of nociceptive transmission is widespread. However, physiological increases in cutaneous sympathetic nerve activity have not been found to influence the firing properties of cutaneous polymodal nociceptive (high-threshold mechano-heat sensitive) fibers in human subjects. Whether the same applies to low-threshold cutaneous mechanoreceptors is not known. Such an effect could be relevant for sympathetic facilitation of nociception, given that tactile afferents are implicated in the allodynia associated with CRPS. This issue was addressed by recording the responses of single cutaneous mechanoreceptors in the glabrous skin of the finger pads to constant mechanical stimuli, at rest and during physiological increases in cutaneous sympathetic activity produced by arousal stimuli. Unitary recordings were made from 17 rapidly adapting (15 FAI and 2 FAII) and 20 slowly adapting (9 SAI and 11 SAII) afferents located in the finger pads via tungsten microelectrodes inserted percutaneously into the median nerve at the wrist in nine subjects. A servomotor applied 1 s constant-displacement ramp-and-hold indentations to the receptor-bearing digit every 3 s. Displacement and compression force were recorded. Blood flow in the finger pad and sweating in the palm were measured contralaterally. Increases in cutaneous sympathetic outflow caused only modest changes in the spontaneous and compression-evoked firing of tactile afferents. These changes were usually (for 26/37 afferents) related to the associated decreases in skin blood flow. The latency from the start of the ramp stimulus to the onset of firing was inversely correlated to flow (i.e. unit response was delayed during vasoconstriction) for 11/31 units (7/15 FAI, 1/2 FAII, 2/9 SAI, 1/5 SAII), whereas no units showed a positive correlation. Compression-evoked firing rates were positively correlated to flow (i.e. vasoconstriction reduced firing rates) for 14/31 units (2/15 FAI, 1/2 FAII, 7/9 SAI, 4/5 SAII), whereas no units showed a negative correlation. 10/11 SAII afferents exhibited spontaneous background firing, which increased for 4 and decreased for 4 in response to arousal stimuli, presumably reflecting their sensitivity to changes in skin stretch associated with sympathetically mediated reductions in blood volume in the finger pad. Two afferents showed no change, but nor was there significant vasoconstriction in these recordings. Thus, arousal stimuli reduced rather than augmented tactile afferent firing. The close relation to blood flow for all types of afferents, and the different responses among SAII afferents, suggest that sympathetically mediated changes in afferent firing properties are indirect, i.e. secondary to changes in the mechanoreceptors' tissue environment rather than to a direct sympathetic effect on the endings.     

TRPV1 antagonist, A-889425, inhibits mechanotransmission in a subclass of rat primary afferent neurons following peripheral inflammation

TRPV1 (transient receptor potential vanilloid family type 1) is a nonselective cation channel that is activated and/or sensitized by noxious heat, protons, and other endogenous molecules released following tissue injury. In addition, a role for TRPV1 in mechanotransmission is emerging. We have recently reported that a selective TRPV1 receptor antagonist, A-889425, reduces mechanical allodynia and spinal neuron responses to mechanical stimulation of complete Freund's adjuvant (CFA)-inflamed rat hind paws. The population of peripheral nerve fibers through which TRPV1 antagonists mediate their effect on mechanotransmission have not yet been described. The objective of this study was to characterize TRPV1-mediated modulation of mechanically evoked activity in sensory axons innervating rat hind paws. We used an in vitro skin-nerve preparation to record neural activity from single axons isolated from rat tibial nerve. Single fibers were classified by conduction velocity, mechanical threshold, and stimulus-response relationships. We used A-889425 to investigate uninjured and inflamed skin afferent neuron populations to evoked mechanical stimulation. Application of A-889425 had no effect on the mechanical responsiveness of Aδ and C-fiber units innervating uninjured skin. In contrast, A-889425 inhibited responses of slowly conducting Aδ fiber units to noxious mechanical stimulation in a population of axons innervating CFA-inflamed hind paws. These data support a role for TRPV1 in mechanotransmission following peripheral inflammation, and highlight the importance of a distinct subclass of primary afferent neurons in mediating this effect.     

The Classification and Properties of Nociceptive Afferent Units from the Skin of the Anaesthetized Pig

The afferent properties of nerve fibres innervating the hairy skin of the pig hind limb were investigated by recording from 142 single units from the saphenous nerve. Identified single units were isolated using maximal electrical stimulation of the nerve trunk. Afferent units were classified on the basis of their responses to a range of stimuli, both thermal (heating to 60°C and cooling to 10°C) and mechanical (air jet, von Frey type filaments with forces of 0.1–250 mN, and strong pressure with a blunt needle). A-fibre units (conduction velocity 6.3–64 m/s, n= 60) fell into categories that have been described in hairy skin in other mammalian species. Most were mechanoreceptors, although seven typical A-fibre mechanical nociceptors with large, multipoint fields were also isolated. No cutaneous receptive field could be found for 15% of A-fibre units. Out of 62 C-fibre units (conduction velocity 0.49–2 m/s) 40% had no cutaneous field for pressure, heat or cold. Of the C-fibre units with cutaneous fields, 42% were polymodal nociceptors, 38% were mechanoreceptors with a variety of properties, including some excited by noxious heat, and 19% were heat-only nociceptors. C-polymodal nociceptors had large receptive fields up to 12.5 mm across and did not sensitize following strong heating. Twenty units conducted at 2–6.3 m/s, between the main C- and A-fibre bands, and were varied in their responses. Some had properties identical to C-fibre mechanoreceptors whilst four were sensitive cold thermoreceptors and one was a polymodal nociceptor. Two units were mechanical nociceptors with small receptive fields. The innervation of pig skin thus has some features like that of primates, such as the presence of C-heat nociceptors and the large receptive fields of C-polymodal nociceptors. However, other features were like non-primates (e.g. the minimal heat-sensitivity of A-mechanical nociceptors) or were unique (the heat-sensitive mechanoreceptors and the lack of heat sensitization in C-polymodal nociceptors from hairy skin).     

Muscle receptors with group IV afferent fibres responding to application of bradykinin.

In 60 anaesthetized cats the discharge characteristics of single group IV afferent fibres from skeletal muscle at rest, during local mechanical stimulation of the muscle, and in response to injections of bradykinin have been studied. 250 group IV fibres from the gastrocnemius-soleus muscle have been identified by their conduction velocity. Of these fibres 121 have been recorded from the dorsal root and 129 from the muscle nerve. A resting discharge was present in 52% of the dorsal root fibres and in 57% of the fibres recorded peripherally. Out of 203 units subjected to local mechanical stimulation of the muscle 9 responded to light, innocuous pressure; 129 fibres required strong, presumably noxious stimulation to be excited. The rest of 65 fibres could not be activated by the mechanical stimuli used. Intraarterial injection of bradykinin excited 88 out of 184 group IV fibres. Intramuscular injection of bradykinin was also found to be an effective stimulus. External application of bradykinin to the intact and to the dissected muscle nerve regularly had no excitatory effect on fibres being sensitive to intraarterially administered bradykinin. Injections of vasoactive drugs (nylidrine, noradrenaline) failed to affect the activity of group IV fibres responding to bradykinin. It is concluded that bradykinin is a potent stimulant for the intramuscular endings of afferent group IV fibres; the effective concentration lying within the range observed in pathologically altered tissues. This sensitivity to bradykinin together with a high mechanical threshold suggests that muscular group IV afferent units participate in the reception of muscular pain.     

DEJERINE-SOTTAS DISEASE WITH A NOVEL DE NOVO DOMINANT MUTATION, SER 149 ARG, OF THE PERIPHERAL MYELIN PROTEIN 22

In this study, differences of unmyelinated nerve fiber density in sural nerve biopsy material from patients suffering from neuropathies of unknown origin with (n = 14) or without pain (n = 13) were analyzed. Immunocytochemistry was applied to differentiate afferent sensory and efferent sympathetic nerve fibers. All patients were evaluated for deficits of small fiber function with thermotesting, quantitative sudomotor-axon reflex-testing and testing of painfulness of mechanical stimuli before performing the biopsy. No difference was found between patients with and without pain concerning clinical deficits or results in any of the neurophysiological examinations. There were also no histopathological differences concerning the density of afferent C-fibers. However, absolute and relative density of efferent sympathetic nerve fibers was significantly higher in patients with painful neuropathy (P < 0.001), although none of the patients demonstrated clinical sympathetic abnormalities. We conclude that an imbalance between afferent and sympathetic nerve fiber density in the periphery may contribute to neuropathic pain even in those patients without obvious clinical autonomic disturbances.     

Properties of functionally identified nociceptive and nonnociceptive facial primary afferents and presynaptic excitability changes induced in their brain stem endings by raphe and orofacial stimuli in cats

The activity of 221 single primary afferent units was recorded extracellularly in the trigeminal (V) ganglion of chloralose-anaesthetized cats to examine the receptive field properties of nonnociceptive and nociceptive cutaneous afferents and the effect of conditioning stimulation of the raphe system and orofacial afferents on the antidromic excitability of their brain stem endings in V subnucleus caudalis. In addition to slowly adapting and rapidly adapting low-threshold mechanosensitive afferents, we functionally identified three classes of cutaneous nociceptive afferents: these included A-delta high-threshold mechanoreceptive afferents (A-delta HTMs), C-fiber high-threshold mechanoreceptive afferents (C-HTMs), and C-polymodal nociceptive afferents (CPNs). Most of the CPNs could be activated by light tactile stimuli as well as by heavy pressure and pinch and noxious radiant heat applied to their mechanoreceptive field which usually involved a localized spot (approximately 1 mm in diameter) of skin. In contrast, the C-HTMs and A-delta HTMs could not be activated by radiant heat stimuli although some did show sensitization which was also a feature of the CPNs; they did respond to noxious mechanical stimulation of a localized area of skin. We noted that orofacial conditioning stimulation could produce an increase in antidromic excitability which was considered a reflection of primary afferent depolarization (PAD) in both nociceptive and nonnociceptive afferents innervating the cat's facial skin; nonnoxious mechanical stimuli and electrical stimuli were particularly effective in the low-threshold mechanosensitive afferents and noxious mechanical and high-intensity electrical stimuli were especially effective in the cutaneous nociceptive afferents. Raphe conditioning stimulation also was very effective in inducing PAD in these nociceptive afferents; however, the raphe conditioning effects were not limited to these nociceptive afferents since PAD was also frequently demonstrated in the low-threshold mechanosensitive afferents.     

Stimulation of PAR-2 excites and sensitizes rat cutaneous C-nociceptors to heat

Proteinase-activated receptor 2 (PAR-2) is expressed on many nociceptive neurons. Application of PAR-2 agonists has been shown to induce behavioral signs of hyperalgesia. We investigated effects of the rat PAR-2 agonist SLIGRL-NH2 in the isolated rat skin-saphenous nerve preparation. SLIGRL-NH2 (100 microM) excited 20% of all C-fiber nociceptors tested. In addition, C-fiber nociceptors were sensitized to heat after SLIGRL-NH2 application resulting in an increase in response magnitude and a decrease of heat threshold. The PAR-2-inactive control peptide LRGILS-NH2 had no effect. The mechanical sensitivity of C-fibers was not affected by SLIGRL-NH2. PAR-2-mediated excitation and sensitization of primary nociceptors may contribute to PAR-2-mediated hyperalgesia.     

Effects of sympathetic stimulation on mechanoreceptive and nociceptive afferent units from the rabbit pinna

Single afferent units from the rabbit great auricular nerve were examined using cutaneous stimuli before, during and after tetanic stimulation of the cervical sympathetic trunk for 1-4 min. G-hair units showed a depression of responses during and for 1-2 min after sympathetic stimulation; D-hair units showed a similar pattern but the average depression was less. These changes are not likely to be due to a reduction in blood flow since no comparable effect was produced by arterial occlusion. Sensitive C-mechanoreceptor units were excited transiently by sympathetic stimulation and by arterial occlusion. Responses of nociceptors, both A-fibre high threshold mechanoreceptors and C-fibre polymodal nociceptors were unaffected by sympathetic stimulation.     

Cardiovascular reactivity in mild hypertension

Blood pressure and heart rate were monitored in 18 mild hypertensives and 18 normotensive controls matched for age and sex during rest, isometric exercise, carbon dioxide rebreathing, cold pressor test, head-up tilt, radiant heat, head-back tilt and mental arithmetic. Blood pressure and heart rate were higher in the hypertensive than in the normotensive group at rest. During the instructions prior to the mental arithmetic and head-back tilt tests, heart rate increased more in the hypertensive than in the normotensive group while during mental arithmetic performance, blood pressure levels increased more in hypertensives than in normotensive controls. Diastolic blood pressure levels increased more in hypertensives during head-up tilt. In all other conditions, cardiovascular responses were similar in both groups. The results support the view that the sympathetic nervous system contributes to hypertension since the cardiovasculature of mild hypertensives was hyperreactive during tests where sympathetic activity was likely to predominate.     

Does neurogenic inflammation alter the sensitivity of unmyelinated nociceptors in the rat?

This study on neurogenic inflammation aimed at comparing the effects of antidromic nerve stimulation and of cutaneous application of mustard oil on the mechanical and thermal sensitivity of cutaneous C-fibres in the rat saphenous nerve. They were tested with von Frey hairs and series of radiant heat stimuli in 5-min intervals before and after one of the above treatments. Antidromic electrical stimulation was effective in evoking a plasma extravasation in the saphenous region as revealed by prior intravenous injection of Evans blue. However, it did not evoke spontaneous activity in the polymodal C-fibres tested nor did it markedly affect their mechanical and thermal excitability in periods of up to 1 h after stimulation. Mustard oil, topically applied to the receptive fields, also evoked localized Evans blue extravasation and strongly affected the C-units. They responded to the application of the irritant with sustained firing; most units showed an increased thermal sensitivity after the treatment. This involved a lowering of the threshold and an increase of the suprathreshold discharge. Even low threshold mechanosensitive C-fibres transiently developed heat sensitivity. In some polymodal units a transient sensitization was followed by a persistent desensitization which abolished their thermal and mechanical sensitivity. Mustard oil application and antidromic nerve stimulation seem to be similar in evoking cutaneous vasodilatation and plasma extravasation but different in inducing nociceptor sensitization.     

Differential impairment of the sudomotor and nociceptor axon-reflex in diabetic peripheral neuropathy

It is not known whether C-fiber functional subclasses are differentially affected by diabetes mellitus or whether the patterns of C-fiber dysfunction are different between type 1 and type 2 diabetes. We therefore examined efferent sympathetic sudomotor and primary afferent nociceptor C-fiber function in diabetic patients. Acetylcholine (10%) was used to evoke C-fiber (axon-reflex)-mediated responses. The nociceptor (flare) response was measured using a laser Doppler device. The sudomotor response was quantified with silastic imprints. The nociceptor C-fiber-mediated flare response was reduced in type 2 diabetic patients (P < 0.008) but was similar to controls in type 1 diabetic patients. The sympathetic C-fiber-mediated responses, including sweat volume (P < 0.05) and the number of activated sweat glands (P = 0.003), were increased in patients with type 1 diabetes. There also was a trend toward a larger axon-reflex sweat area in patients with type 1 diabetes (P = 0.09). No differences in these sweat responses were found in patients with type 2 diabetes compared to controls. These findings suggest that the functional abnormalities in diabetic peripheral neuropathy are not homogeneous and that C-fiber subclasses are differentially affected in type 1 and 2 diabetes mellitus.     

C nociceptor activity in human nerve during painful and non painful skin stimulation.

Percutaneous recordings from more than one hundred single C fibres have been performed in the radial nerve of conscious human subjects. All these fibres belong to the poly-modal C nociceptor group, being excited by mechanical and thermal and also by chemical stimulation. Conduction velocities showed a monophasic distribution with a mean value of 0.86 m/s (SD: 0.17). The mechanical threshold, measured with von Frey hairs, varied between 2.3 and 13.1 g. The receptive field was circular or elliptical; for 33 units the mean axes were 6 mm and 7 mm. Mechanically evoked C fibre discharge even up to more than 10 spikes/s was not necessarily accompanied by pain sensation. Nettle sting evoked an irregular C fibre discharge (maximum 10 spikes/s) accompanied by a pricking and burning sensation; the sensation of itch which was sometimes reported, was not correlated with the discharge frequency. C fibre activation by a chemical irritant (paint remover) also evoked an irregular discharge (maximum 3 to 6 spikes/s), accompanied by pricking and burning pain sensation. The C threshold for radiant heat usually lay below the subject's pain threshold. Increasing skin temperature produced increasing neural firing rate. The mean spike frequency rarely exceeded two spikes/s even with stimuli evoking strong heat pain. The occurrence of subjective heat pain response could be as well predicted from th C fibre spike frequency as from the skin temperature. It is concluded that nociceptive C input provoked by thermal or chemical stimuli correlates well with pain sensation. However, similar C input provided by mechanical stimulation which activates also A beta mechanoreceptors, did not necessarily produce pain sensation.     

Functional interactions between tumor and peripheral nerve: changes in excitability and morphology of primary afferent fibers in a murine model of cancer pain

We used a murine model to investigate functional interactions between tumors and peripheral nerves that may contribute to pain associated with cancer. Implantation of fibrosarcoma cells in and around the calcaneus bone produced mechanical hyperalgesia of the ipsilateral paw. Electrophysiological recordings from primary afferent fibers in control and hyperalgesic mice with tumor revealed the development of spontaneous activity (0.2-3.4 Hz) in 34% of cutaneous C-fibers adjacent to the tumor (9-17 d after implantation). C-fibers in tumor-bearing mice exhibited a mean decrease in heat threshold of 3.5 +/− 0.10 °C. We also examined innervation of the skin overlying the tumor. Epidermal nerve fibers (ENFs) were immunostained for protein gene product 9.5, imaged using confocal microscopy, and analyzed in terms of number of fibers per millimeter of epidermal length and branching (number of nodes per fiber). Divergent morphological changes were linked to tumor progression. Although branching of ENFs increased significantly relative to control values, in later stages (16-24 d after implantation) of tumor growth a sharp decrease in the number of ENFs was observed. This decay of epidermal innervation of skin over the tumor coincided temporally with gradual loss of electrophysiological activity in tumor-bearing mice. The development of spontaneous activity and sensitization to heat in C-fibers and increased innervation of cutaneous structures within the first 2 weeks of tumor growth suggest activation and sensitization of a proportion of C-fibers. The decrease in the number of ENFs observed in later stages of tumor development implicates neuropathic involvement in this model of cancer pain.