Search

A recent study with Ca++-sensitive-dyes in neurons in whole DRGs (Table 5) found that much lower percentages of nociceptors were polymodal-nociceptors (PMNs) (Emery et al., 2016), than the 50–80% values in many electrophysiological fiber studies.This conflict highlighted the lack of knowledge about percentages of nociceptor-subtypes in the DRG. This was analysed from intracellularly-recorded neurons in rat lumbar DRGs stimulated from outside the skin. Polymodal nociceptors (PMNs) were 11% of all neurons and 19% of all nociceptors. Most PMNs had C-fibers (CPMNs). Percentages of C-nociceptors that were CPMNs varied with receptive field (RF) depths, whether superficial (∼80%), dermal (25%), deep (0%) or cutaneous (superficial + dermal) (40%). This explains CPMN percentages 40–90%, being highest, in electrophysiological studies using cutaneous nerves, and lowest in studies that also include deep RFs, including ours, and the recent Ca++-imaging studies in whole DRGs.Despite having been originally described in 1967 (Burgess and Perl), both Aβ-nociceptors and Aβ-moderate pressure receptors (MPRs) remain overlooked. Most A-fiber nociceptors in rodents have Aβ-fibers. Of rat lumbar Aβ-nociceptors with superficial RFs, 50% were MPRs with variable medium-low trkA-expression. Despite having conduction velocities at the two extremes for nociceptors, both CPMNs and MPRs have relatively low thresholds, superficial/epidermal RFs and low trkA-expression. For abbreviations used see Table 5.

The hyperpolarization-activated current (I(h)) has been implicated in nociception/pain, but its expression levels in nociceptors remained unknown. We recorded I(h) magnitude and properties by voltage clamp from dorsal root ganglion (DRG) neurons in vivo, after classifying them as nociceptive or low-threshold-mechanoreceptors (LTMs) and as having C-, Aδ- or Aα/β-conduction velocities (CVs). For both nociceptors andLTMs, I(h) amplitude and I(h) density (at -100 mV) were significantly positively correlated with CV.Median I(h) magnitudes and I(h) density in neuronal subgroupswere respectively:muscle spindle afferents(MSAs):-4.6 nA,-33 pA pF(-1); cutaneous Aα/β LTMs: -2.2 nA, -20 pA pF(-1); Aβ-nociceptors: -2.6 nA, -21 pA pF(-1); both Aδ-LTMs and nociceptors: -1.3 nA, ∼-14 pA pF(-1); C-LTMs: -0.4 nA, -7.6 pA pF(-1); and C-nociceptors: -0.26 nA, -5 pApF(-1). I(h) activation slow time constants (slow τ values) were strongly correlated with fast τ values; both were shortest in MSAs. Most neurons had τ values consistent with HCN1-related I(h); others had τ values closer to HCN1+HCN2 channels, or HCN2 in the presence of cAMP. In contrast, median half-activation voltages (V(0.5)) of -80 to -86 mV for neuronal subgroups suggest contributions of HCN2 to I(h). τ values were unrelated to CV but were inversely correlated with I(h) and I(h) density for all non-MSA LTMs, and for Aδ-nociceptors. From activation curves ∼2-7% of I(h)would be activated at normal membrane potentials. The high I(h) may be important for excitability of A-nociceptors (responsible for sharp/pricking-type pain) and Aα/β-LTMs (tactile sensations and proprioception). Underlying HCN expression in these subgroups therefore needs to be determined. Altered high I(h) may be important for excitability of A-nociceptors (responsible for sharp/pricking-type pain) and Aα/β-LTMs (tactile sensations and proprioception). Underlying HCN expression in these subgroups therefore needs to be determined. Altered Ih expression and/or properties (e.g. in chronic/pathological pain states) may influence both nociceptor and LTM excitability.expression and/or properties (e.g. in chronic/pathological pain states) may influence both nociceptor and LTM excitability.

Summary Objective Osteoarthritis (OA) pain mechanisms are poorly understood. We used the monosodium iodoacetate (MIA) model of knee OA to characterize changes in excitability during the course of OA in different classes of mechanosensitive afferents projecting to joint-associated tissues, and examine whether these afferent responses and pain behavior are correlated. Methods Rats were injected intra-articularly with MIA (1mg in 50μl). Hind-limb weight bearing was studied 3 (MIA3) and 14 (MIA14) days after MIA, followed by deep anesthesia and teased-nerve-fiber recordings. Spontaneous activity (SA) and mechanically evoked responses of A- and C-mechanosensitive fibers (AM and CM respectively, probably nociceptive) innervating tissues associated with the ipsilateral knee joint were examined. Results MIA3 and MIA14 rats exhibited reduced ipsilateral weight bearing. SA (>0.02impulses/s) occurred in ∼50% of CMs from MIA rats vs 0% in normals. SA firing rates in CMs were significantly higher than normal; decreased weight bearing was correlated with increased CM SA rates. Neither percentages of AMs with SA (20%) nor their firing rates (0–0.01impulses/s) significantly increased after MIA. In contrast, in MIA rats AMs, but not CMs, exhibited decreased mechanical thresholds and increased firing rates in response to suprathreshold mechanical stimulation. Conclusions These findings of increased SA firing rate in CMs but not AMs and increased mechanical sensitivity of AMs, but not CMs, have not previously been reported. These are two distinct important physiological mechanisms that may underpin spontaneous pain (CMs) and stimulus-evoked pain (AMs) in OA. Our data contribute to a mechanism-based understanding of OA pain.

Two pore domain potassium (K2P) channels (KCNKx.x) cause K + leak currents and are major contributors to resting membrane potential. Their roles in dorsal root ganglion (DRG) neurons normally, and in pathological pain models, are poorly understood. Therefore, we examined mRNA levels for 10 K2P channels in L4 and L5 rat DRGs normally, and 1 day and 4 days after unilateral cutaneous inflammation, induced by intradermal complete Freund's adjuvant (CFA) injections. Spontaneous foot lifting (SFL) duration (spontaneous pain behaviour) was measured in 1 day and 4 day rats TRESK(KCNK18) > TRAAK(KCNK4) > TREK2(KCNK10) = TWIK2(KCNK6) > TREK1 (KCNK2) = THIK2(KCNK12) > TASK1(KCNK3) > TASK2(KCNK5) > THIK1(KCNK13) = TASK3(KCNK9). During inflammation, the main differences from normal in DRG mRNA levels were bilateral, suggesting systemic regulation, although some channels showed evidence of ipsilateral modulation. By 1 day, bilateral K2P mRNA levels had decreased (THIK1) or increased (TASK1, THIK2) but by 4 days they were consistently decreased (TASK2, TASK3) or tended to decrease (excluding TRAAK). The decreased TASK2 mRNA was mirrored by decreased protein (TASK2-immunoreactivity) at 4 days. Ipsilateral mRNA levels at 4 days compared with 1 day were lower (TRESK, TASK1, TASK3, TASK2 and THIK2) or higher (THIK1). Ipsilateral SFL duration during inflammation was positively correlated with ipsilateral TASK1 and TASK3 mRNAs, and contralateral TASK1, TRESK and TASK2 mRNAs. Thus changes in K2P mRNA levels occurred during inflammation and for 4 K2P channels were associated with spontaneous pain behaviour (SFL). K2P channels and their altered expression are therefore associated with inflammation-induced pain.

Summary After partial nerve injury, changes in uninjured nociceptors and nonnociceptors (spontaneous firing, decreased electrical thresholds, hyperpolarisation) may account for the different aspects of neuropathic and paresthesias., Partial nerve injury leads to peripheral neuropathic pain. This injury results in conducting/uninterrupted (also called uninjured) sensory fibres, conducting through the damaged nerve alongside axotomised/degenerating fibres. In rats seven days after L5 spinal nerve axotomy (SNA) or modified-SNA (added loose-ligation of L4 spinal nerve with neuroinflammation-inducing chromic-gut), we investigated a) neuropathic pain behaviours and b) electrophysiological changes in conducting/uninterrupted L4 dorsal root ganglion (DRG) neurons with receptive fields (called: L4-receptive-field-neurons). Compared to pretreatment, modified-SNA rats showed highly significant increases in spontaneous-foot-lifting duration, mechanical-hypersensitivity/allodynia, and heat-hypersensitivity/hyperalgesia, that were significantly greater than after SNA, especially spontaneous-foot-lifting. We recorded intracellularly in vivo from normal L4/L5 DRG neurons and ipsilateral L4-receptive-field-neurons. After SNA or modified-SNA, L4-receptive-field-neurons showed the following: a) increased percentages of C-, Ad-, and Ab-nociceptors and cutaneous Aa/b-low-threshold mechanoreceptors with ongoing/spontaneous firing; b) spontaneous firing in C-nociceptors that originated peripherally; this was at a faster rate in modified-SNA than SNA; c) decreased electrical thresholds in A-nociceptors after SNA; d) hyperpolarised membrane potentials in A-nociceptors and Aa/b-low-threshold-mechanoreceptors after SNA, but not C-nociceptors; e) decreased somatic action potential rise times in C- and A-nociceptors, not Aa/b-low-threshold-mechanoreceptors. We suggest that these changes in subtypes of conducting/uninterrupted neurons after partial nerve injury contribute to the different aspects of neuropathic pain as follows: spontaneous firing in nociceptors to ongoing/spontaneous pain; spontaneous firing in Aa/b-low-threshold-mechanoreceptors to dysesthesias/paresthesias; and lowered A-nociceptor electrical thresholds to A-nociceptor sensitization, and greater evoked pain.

Muscle spindle afferent (MSA) neurons can show rapid and sustained firing. Immunostaining for the α3 isoform of the Na+/K+-ATPase (α3) in some large dorsal root ganglion (DRG) neurons and large intrafusal fibres suggested α3 expression in MSAs (Dobretsov et al. 2003), but not whether α3-immunoreactive DRG neuronal somata were exclusively MSAs. We found that neuronal somata with high α3 immunointensity were neurofilament-rich, suggesting they have A-fibres; we therefore focussed on A-fibre neurons to determine the sensory properties of α3-immunoreactive neurons. We examined α3 immunointensity in 78 dye-injected DRG neurons whose conduction velocities and hindlimb sensory receptive fields were determined in vivo. A dense perimeter or ring of staining in a subpopulation of neurons was clearly overlying the soma membrane and not within satellite cells. Neurons with clear α3 rings (n = 23) were all MSAs (types I and II); all MSAs had darkly stained α3 rings, that tended to be darker in MSA1 than MSA2 units. Of 52 non-MSA A-fibre neurons including nociceptive and cutaneous low-threshold mechanoreceptive (LTM) neurons, 50 had no discernable ring, while 2 (Aα/β cutaneous LTMs) had weakly stained rings. Three of three C-nociceptors had no rings. MSAs with strong ring immunostaining also showed the strongest cytoplasmic staining. These findings suggest that α3 ring staining is a selective marker for MSAs. The α3 isoform of the Na+/K+-ATPase has previously been shown to be activated by higher Na+ levels and to have greater affinity for ATP than the α1 isoform (in all DRG neurons). The high α3 levels in MSAs may enable the greater dynamic firing range in MSAs.

Intracellular recordings were made from 1022 somatic lumbar dorsal root ganglion (DRG) neurones in anaesthetized adult rats, classified from dorsal root conduction velocities (CVs) as C, Aδ or Aα/β, and according to their responses to mechanical and thermal stimuli as nociceptive (including high-threshold mechanoreceptive (HTM) units), and non-nociceptive (including low-threshold mechanoreceptive (LTM) and cooling units). Of these, 463 met electrophysiological criteria for analysis of action potentials (APs) evoked by dorsal root stimulation. These included 47 C-, 71 Aδ- and 102 Aα/β-nociceptive, 10 C-, 8 Aδ− and 178 Aα/β-LTM, 18 C- and 19 Aδ- unresponsive, and 4 C-cooling units. Medians of AP and afterhyperpolarization (AHP) durations and AP overshoots were significantly greater for nociceptive than LTM units in all CV groups. AP overshoots and AHP durations were similar in nociceptors of all CV groups whereas AP durations were greater in slowly conducting, especially C-fibre, nociceptors. C-cooling units had faster CVs, smaller AP overshoots and shorter AP durations than C-HTM units. A subgroup of Aα/β-HTM, moderate pressure units, had faster CVs and AP kinetics than other Aα/β-HTM units. Of the Aα/β-LTM units, muscle spindle afferents had the fastest CV and AP kinetics, while rapidly adapting cutaneous units had the slowest AP kinetics. AP variables in unresponsive and nociceptive units were similar in both C- and Aδ-fibre CV groups. The ability of fibres to follow rapid stimulus trains (fibre maximum following frequency) was correlated with CV but not sensory modality. These findings indicate both the usefulness and limitations of using electrophysiological criteria for identifying neurones acutely in vitro as nociceptive.

The TTX-sensitive Na(v)1.7 (PN1) Na(+) channel alpha subunit protein is expressed mainly in small dorsal root ganglion (DRG) neurones. This study examines immunocytochemically whether it is expressed exclusively or preferentially in nociceptive primary afferent DRG neurones, and determines the electrophysiological properties of neurones that express it. Intracellular somatic action potentials (APs) evoked by dorsal root stimulation were recorded in L6/S1 DRG neurones at 30 +/- 2 degrees C in vivo in deeply anaesthetised young guinea-pigs. Each neurone was classified, from its dorsal root conduction velocity (CV) as a C-, Adelta- or Aalpha/beta-fibre unit and from its response to mechanical and thermal stimuli, as a nociceptive, low threshold mechanoreceptive (LTM) or unresponsive unit. Fluorescent dye was injected into the soma and Na(v)1.7-like immunoreactivity (Na(v)1.7-LI) was examined on sections of dye-injected neurones. All C-, 90 % of Adelta- and 40 % of Aalpha/beta-fibre units, including both nociceptive and LTM units, showed Na(v)1.7-LI. Positive units included 1/1 C-LTM, 6/6 C-nociceptive, 4/4 C-unresponsive (possible silent nociceptive) units, 5/6 Adelta-LTM (D hair), 13/14 Adelta-nociceptive, 2/9 Aalpha/beta-nociceptive, 10/18 Aalpha/beta-LTM cutaneous and 0/9 Aalpha/beta-muscle spindle afferent units. Overall, a higher proportion of nociceptive than of LTM neurones was positive, and the median relative staining intensity was greater in nociceptive than LTM units. Na(v)1.7-LI intensity was clearly positively correlated with AP duration and (less strongly) negatively correlated with CV and soma size. Since nociceptive units tend overall to have longer duration APs, slower CVs and smaller somata, these correlations may be related to the generally greater expression of Na(v)1.7 in nociceptive units.

To establish the afferent receptive properties of lumbosacral dorsal root ganglion (DRG) neurones that express calcitonin gene-related peptide (CGRP), intracellular recordings were made with fluorescent dye-filled electrodes in deeply anaesthetised young guinea-pigs. After determination of neuronal functional properties, dye was injected into the soma. CGRP-like immunoreactivity (CGRP-LI) was examined on histological sections of dye-marked neurones. Fourteen of 34 C-fibre neurones showed CGRP-LI. These included 10/21 C-fibre nociceptive neurones. All C-polymodal nociceptors in glabrous (n = 4) but none in hairy skin (n = 4) were positive. Positive C-fibre high threshold mechanoreceptive (HTM) units had receptive fields in dermal or deeper tissue. Four (n = 6) unresponsive or unidentified C-fibre units were positive. Neither C-fibre cooling sensitive (n = 4) nor C-fibre low threshold mechanoreceptive (LTM) units (n = 3) had CGRP-LI. Six of 23 A-fibre nociceptive cells were positive including one Aα/β unit. Three of these positive cells had epidermal and three had dermal/deep receptive fields. Three of 36 A-fibre LTM units exhibited CGRP-LI, all were Aα/β-fibre G hair units. All glabrous skin and muscle spindle units and in hairy skin slowly adapting and field units, and some G-hair units lacked CGRP-LI. CGRP-LI stained fibres were found in tissues containing receptive fields of positive DRG neurones: glabrous skin, near hair follicles and in skeletal muscle. A few substance P-labelled neurones did not exhibit CGRP-LI and vice versa. Thus CGRP expression was detected in under half the nociceptive neurones, was not limited to nociceptive neurones and apart from receptive properties was also related to location/depth in the tissues of a DRG neurone's peripheral terminals.

Novel findings of changes in nociceptive dorsal root ganglion (DRG) neurons during hindlimb inflammation induced by complete Freund's adjuvant (CFA) injections in the hindpaw and hindleg are reported. These include increased maximum fiber following frequency in nociceptive C- and Aδ-fiber units by 2.7 and 3 times, respectively, and increased incidence of ongoing (spontaneous) activity by 3.3 times (to 54%) and 2.4 times (to 27%), respectively. These changes and the CFA-induced changes in somatic action potential (AP) configuration in nociceptive neurons (Djouhri and Lawson, 1999) were incomplete 24 hr after CFA. The nerve growth factor (NGF) dependence of the inflammation-induced changes was examined by injecting a synthetic NGF sequestering protein [tyrosine receptor kinase A Ig2 (trkA Ig2)] with CFA and subsequently into the CFA injection sites. NGF sequestration prevented some CFA-induced changes in nociceptive neurons including: the increased fiber following frequency (C and Aδ), the increased proportions of units with ongoing activity (C and Aδ), the decreased AP duration (C and Aδ), but not the decreased afterhyperpolarization (AHP) durations (C, Aδ, and Aα/β) (Djouhri and Lawson, 1999). AP variables of nociceptive units with spontaneous activity were examined.The time course of electrophysiological changes in nociceptive units is consistent with processes involving altered protein expression and/or retrograde transport of factors. These results (1) implicate NGF in regulating inflammation-induced decreases in AP duration and in increases in firing rate and spontaneous activity but not in decreases in AHP duration and (2) suggest clinical advantages of reducing NGF in some inflammatory pain states.

Decreases in durations of action potentials (C- and Adelta-fibre units) and afterhyperpolarisations (A-fibre units) occur in somata of nociceptive dorsal root ganglion neurons during hindlimb inflammation induced in young guinea-pigs by intradermal injections of Complete Freund's Adjuvant into the ipsilateral leg and foot. Here we present evidence that the single-point conduction velocity (i.e. estimated over a single conduction distance) of these nociceptive neurons is increased during this type of inflammation. The single-point conduction velocities in anaesthetised untreated guinea-pigs (control) were compared with those two and four days after Complete Freund's Adjuvant treatment in two types of experiment. The first involved intracellular voltage recordings from somata of ipsilateral L6 and S1 dorsal root ganglion neurons. Units were classified as C, Adelta or Aalpha/beta on the basis of their dorsal root conduction velocities and characterised as nociceptive, low-threshold mechanoreceptive or unresponsive according to their responses to mechanical and thermal stimuli. Compared with untreated animals, significant increases of 54% for C-fibre nociceptive units and 46% for A-fibre nociceptive units in the medians of dorsal root single-point conduction velocities were found four days after Complete Freund's Adjuvant treatment. These increases were greater at four days than at two days after Complete Freund's Adjuvant. A slight tendency in the same direction (10%) that was not significant was also seen in low-threshold mechanoreceptors four days after treatment, but not after two days. The increased velocities were confirmed with compound action potential recordings from ipsilateral S2 dorsal roots and sural nerves, in treated and control animals. Recordings showed a tendency for increased single-point velocities in C, Adelta and Aalpha/beta waves, with the upper border of the Adelta wave (i.e. the border between Adelta and Aalpha/beta waves) falling at a significantly higher conduction velocity in treated than control animals. This was seen both in S2 dorsal roots and in sural nerves. There was also a significant decrease in the mean electrical threshold for eliciting the C and Adelta components of compound action potentials of both dorsal root and sural nerves during inflammation. No evidence was found for a reduction in utilisation time for any components of the sural nerve compound action potential (C, Adelta or Aalpha/beta). The conduction velocity increases may be due to altered expression or activation/inactivation of certain ion channel types, such as Na(+) channels. The present experiments demonstrate that hindlimb inflammation caused a significant increase in conduction velocity of nociceptive but not of low-threshold mechanoreceptive primary afferent neurons during inflammation, as well as a significant decrease in the mean electrical threshold for eliciting the C and Adelta components of compound action potentials of both dorsal root and sural nerves. These changes, together with the previously described changes in the action potential shape of nociceptive neurons during inflammation, probably reflect alterations in membrane function that contribute to inflammatory hyperalgesia.

Partial sciatic nerve injury, a model of neuropathic pain, elicits a variety of neurochemical, electrophysiological and neuroanatomical changes in primary sensory neurons. We have used the technique of messenger RNA differential display to identify genes with altered expression in these neurons which may contribute to the development of aberrant sensation following such peripheral nerve damage. This approach identified 14 distinct complementary DNA clones, representing transcripts with increased ipsilateral expression in L4/5 dorsal root ganglia, two weeks after unilateral partial ligation of the rat sciatic nerve. Both Zucker diabetic fatty rats and their lean counterparts were used in this study but none of the transcripts identified showed an induction that was confined to one of the two groups. The majority of the clones did not show significant sequence similarity to previously reported genes and therefore may represent novel messenger RNA sequences or, alternatively, unknown regions of partially characterised messenger RNAs. Two of the clones represented transcripts for the known proteins muscle LIM protein and acidic epididymal glycoprotein, neither of which had previously been associated with expression in the nervous system. Reverse transcriptase-polymerase chain reaction analysis and in situ hybridization confirmed that the messenger RNA expression of both muscle LIM protein and acidic epididymal glycoprotein was induced in an ipsilateral-specific manner. Their localisations, examined with in situ hybridization in L5 dorsal root ganglia, were limited in each case to a sub-population of neuronal profiles. Those neuronal profiles that demonstrated muscle LIM protein hybridization were distributed across the profile size range, whereas the distribution of acidic epididymal glycoprotein-positive profiles appeared to be skewed towards smaller profiles. The induction of muscle LIM protein and acidic epididymal glycoprotein in dorsal root ganglia may play an important functional role in the adaptive response of primary sensory neurons following partial sciatic nerve injury.

Intracellular electrophysiological recordings were made in vitro at 36.5°C from lumbar (L4 to L6) dorsal root ganglion neurones of 6–8-week-old female rats. Electrophysiological properties were recorded prior to intracellular injection with fluorescent dye. The following showed substance P-like immunoreactivity (SP-LI): 8/19 C-fibre cells, 6/26 Aδ cells, and 0/52 Aα/β cells. In C-fibre neurones, there were no significant differences in action potential (AP) characteristics between those with SP-LI and those without. In contrast, the Aδ neurones with SP-LI had significantly deeper and longer afterhyperpolarisations (AHPs), but their AP durations did not differ from those without SP-LI. Both SP-LI and calcitonin gene-related peptide-LI (CGRP-LI) were examined on 18 Aδ and 11 C cells. Most (7/8) neurones with SP-LI also showed CGRP-LI, but only 7/21 neurones without SP-LI showed CGRP-LI. One C cell showed SP-LI but no CGRP-LI. Neurones with neither peptide (−/−), with only CGRP (−/CGRP) or with both peptides (SP/CGRP) were compared (n = 28). The numbers in each group were, respectively, 5, 2, and 3 with C-fibres and 9, 5, and 4 with Aδ-fibres. The peptide content and AP shape were related in Aδ neurones. Most −/− Aδ neurones had short APs and short AHPs; most −/CGRP neurones had long APs and long AHPs, and SP/CGRP neurones had short APs with deep, long AHPs. There was a positive correlation between log10 of the area under the AP (AP area) and log10 AHP duration in Aδ neurones. All SP- and most CGRP-containing Aδ neurones had AP shapes similar to those previously described for nociceptive neurones. However, a few without peptide also showed such properties, raising the possibility that some nociceptive neurones did not express these peptides. J. Comp. Neurol. 388:541–549, 1997. © 1997 Wiley-Liss, Inc.

1. The relationship between the afferent properties and substance P-like immunoreactivity (SP-LI) of L6 and S1 dorsal root ganglion (DRG) neuronal somata was examined in anaesthetized guinea-pigs. Glass pipette microelectrodes filled with fluorescent dyes were used to make intracellular recordings and to label DRG somata. The dorsal root conduction velocity (CV) and the afferent receptive properties of each unit were categorized according to criteria established in other species. Categories included a variety of low threshold mechanoreceptive classes, innocuous thermoreceptive and several nociceptive classes. Nociceptive units were further subdivided on the basis of CV and the locus of the receptive field (superficial cutaneous, deep cutaneous or subcutaneous). 2. SP-LI was determined using the avidin-biotin complex method and the relative staining intensity determined by image analysis. The possible significance of labelling intensity is discussed. Clear SP-LI appeared in twenty-nine of 117 dye-labelled neurones. All SP-LI positive units with identified receptive properties were nociceptive but not all categories of nociceptors were positive. The intensity of SP-LI labelling varied, often systematically, in relation to afferent properties. There was a tendency for nociceptive neurones with slower CVs and/or smaller cell bodies to show SP-LI. 3. Nineteen of fifty-one C fibre neurones showed SP-LI. Fewer than half the C polymodal nociceptors (CPMs) were positive. The most intensely labelled units were the deep cutaneous nociceptors and some of the CPMs in glabrous skin. C low threshold mechanoreceptors and cooling-sensitive units did not show SP-LI. 4. Ten of sixty-six A fibre neurones exhibited SP-LI, including eight of sixteen A delta nociceptors and two of fifteen A alpha/beta nociceptors. A fibre neurones exhibiting SP-LI included seven of eight deep cutaneous mechanical nociceptors and some superficial cutaneous mechano-heat nociceptors of hairy skin. In contrast, none of twenty superficial cutaneous A high threshold mechanoreceptor units or the thirty-five A fibre low threshold units (D-hair and other units) showed detectable SP-LI. 5. We conclude that SP-LI labelling in guinea-pig DRG neurones is related to (a) afferent receptive properties, (b) the tissue in which the peripheral receptive terminals are located, (c) the CV and (d) the soma size.

Neurokinin A-like immunoreactivity of dorsal root ganglion neurons innervating the kidney were studied with retrograde tracing of FluoroGold dye applied to the cut renal nerves. The proportions and sizes of renal afferent neurons with neurokinin A-like immunoreactivity were quantified in T9-L2 dorsal root ganglia from five rats. Of 240 renal afferent neuronal somata examined, 26 +/- 3% (S.E.M.) showed neurokinin A-like immunoreactivity. Compared with the overall size distribution of renal afferent neurons, those staining for neurokinin A were mostly small-sized neurons with a few medium-sized neurons. All somata with neurokinin A-like immunoreactivity were neurofilament-poor as judged by labelling with an anti-neurofilament antibody, RT97, and it is therefore likely that they had unmyelinated fibres. To examine the sites to which the renal afferent fibres with neurokinin A might project, sections of rat and guinea-pig kidney and upper ureter were examined. Fibres with neurokinin A-like immunoreactivity were found beneath and within the transitional epithelium lining the inner surface of the pelvis, and within the smooth muscle layer beneath the transitional epithelium. Epithelial innervation was found only in regions with underlying smooth muscle and loose connective tissue, and not in sites where the epithelium was closely applied to the renal parenchyma. The network of fibres was most dense towards the pelvo-uretic junction. Fibres with neurokinin A-like immunoreactivity were not seen beneath or within the cuboidal/columnar epithelium covering the papilla within the renal pelvis. Furthermore, only very few fibres with neurokinin A were observed penetrating the transitional epithelium of the upper ureter in both rat and guinea-pig. The distribution of fibres labelled with antibodies to substance P and calcitonin gene-related peptide in the renal pelvis was similar to that for fibres with neurokinin A-like immuno-reactivity, although a few fibres penetrated further into the fornices than fibres with neurokinin-A-like immunoreactivity. Thus, many afferent fibres in the renal pelvis may contain neurokinin A as well as substance P and calcitonin gene-related peptide. These fibres may be the source of the neurokinin A, substance P and calcitonin gene-related peptide which can be released by topical capsaicin treatment. In addition they may be the mechano- and chemo-receptive fibres in the renal pelvis that are known to play important roles in renal haemodynamics. The intra-epithelial position of some of these fibres in the epithelial layer suggests a possible chemosensory or osmosensory role.

Intracellular voltage recordings and fluorescent dye injections were made in vitro in 107 neurons in lumbar dorsal root ganglia (DRGs) of 6- to 8-week-old rats. Calcitonin gene-related, peptide-like immunoreactivity (CGRP-LI) was examined in these neurones, which were divided into C-, A delta-, and A alpha/beta-fibre neurones on the basis of their conduction velocities (CVs). A-fibre neurones with CGRP-LI had significantly longer mean action potential (AP) and afterhyperpolarisation (AHP) durations than those without CGRP-LI. A delta neurones with CGRP-LI had significantly longer AHP durations, slower CVs and slower maximal fibre following frequencies than those without CGRP-LI. They also had longer AP durations (not significant). The largest A delta neurones were CGRP-LI negative, whereas the smaller cells were either positive or negative. A alpha/beta neurones with CGRP-LI also had longer mean APs (not significant) and AHPs (significant) than those without CGRP-LI, and the cell size distributions were similar for positive and negative neurones. Most A-fibre neurones with CGRP-LI had inflections on the falling phase of the somatic AP. Of the A-fibre neurones with such inflections (Ai neurones), those with CGRP-LI had longer AP durations (not significant) and longer AHP durations (significant) than Ai neurones without CGRP-LI, pointing to a functionally distinct subgroup of Ai neurones. There were no significant differences in electrophysiological properties or cell size measurements between C-fibre neurones with and without detectable CGRP-LI. The patterns of electrophysiological properties of A delta neuronal somata with CGRP-LI and of most, but not all, A alpha/beta neuronal somata with CGRP-LI are similar to those reported for cutaneous nociceptors with A fibres in rat (Ritter and Mendell [1992] J. Neurophysiol. 68:2033-2041). Because rat DRG neurones that express CGRP normally also express trkA (Averill et al. [1995] Eur. J. Neurosci. 7:1484-1494), the properties described here of neurones with CGRP-LI are probably the same as those of DRG neurones with trkA.

I(h), which influences neuronal excitability, has recently been measured in vivo in sensory neuron subtypes in dorsal root ganglia (DRGs). However, expression levels of HCN (hyperpolarization-activated cyclic nucleotide-gated) channel proteins that underlie I(h) were unknown. We therefore examined immunostaining of the most abundant isoforms in DRGs, HCN1 and HCN2 in these neuron subtypes. This immunostaining was cytoplasmic and membrane-associated (ring). Ring-staining for both isoforms was in neurofilament-rich A-fiber neurons, but not in small neurofilament-poor C-fiber neurons, although some C-neurons showed cytoplasmic HCN2 staining. We recorded intracellularly from DRG neurons in vivo, determined their sensory properties (nociceptive or low-threshold-mechanoreceptive, LTM) and conduction velocities (CVs). We then injected fluorescent dye enabling subsequent immunostaining. For each dye-injected neuron, ring- and cytoplasmic-immunointensities were determined relative to maximum ring-immunointensity. Both HCN1- and HCN2-ring-immunointensities were positively correlated with CV in both nociceptors and LTMs; they were high in Aβ-nociceptors and Aα/β-LTMs. High HCN1 and HCN2 levels in Aα/β-neurons may, via I(h), influence normal non-painful (e.g. touch and proprioceptive) sensations as well as nociception and pain. HCN2-, not HCN1-, ring-intensities were higher in muscle spindle afferents (MSAs) than in all other neurons. The previously reported very high I(h) in MSAs may relate to their very high HCN2. In normal C-nociceptors, low HCN1 and HCN2 were consistent with their low/undetectable I(h.) In some C-LTMs HCN2-intensities were higher than in C-nociceptors. Together, HCN1 and HCN2 expressions reflect previously reported I(h) magnitudes and properties in neuronal subgroups, suggesting these isoforms underlie I(h) in DRG neurons. Expression of both isoforms was NT3-dependent in cultured DRG neurons. HCN2-immunostaining in small neurons increased 1 day after cutaneous inflammation (CFA-induced) and recovered by 4 days. This could contribute to acute inflammatory pain. HCN2-immunostaining in large neurons decreased 4 days after CFA, when NT3 was decreased in the DRG. Thus HCN2-expression control differs between large and small neurons.

The POU-domain transcription factor Oct-2 is expressed in both B lymphocytes and sensory neurones, where its expression is regulated by nerve growth factor (NGF). In order to define a possible role for Oct-2 in neurotrophin signalling, we examined the expression of an NGF-regulated channel (capsaicin-evoked ion fluxes), neuropeptides (substance P, calcitonin gene-related peptide), structural proteins (neurofilaments and peripherin) and receptors (trks) in dorsal root ganglion neurones derived from perinatal transgenic mice containing a defective Oct-2 structural gene. Northern blots show that central nervous tissue contains a larger than normal (> 10 kb) mRNA transcript corresponding in size to an Oct-2 transcript encoding a defective protein. PCR analysis shows the absence of normal Oct-2 transcripts in dorsal root ganglia. In null mutants, capsaicin sensitivity, and neuropeptide and cytoskeletal protein expression were unaffected by the loss of Oct-2 expression. The number of sensory neurones and the gross morphology of CNS tissues that normally express high levels of Oct-2 were also examined and found to be normal in the null mutant. Heterozygous animals show normal thresholds of sensitivity to noxious heat and normal inflammatory responses. Oct-2 does not therefore play an essential role in the NGF responsiveness of sensory neurones in these animals.

Binding to isolectin-B4 (IB4) and expression of tyrosine kinase A (trkA) (the high-affinity NGF receptor) have been used to define two different subgroups of nociceptive small dorsal root ganglion (DRG) neurons. We previously showed that only nociceptors have high trkA levels. However, information about sensory and electrophysiological propertiesin vivoof single identified IB4-binding neurons, and about their trkA expression levels, is lacking. IB4-positive (IB4+) and small dark neurons had similar size distributions. We examined IB4-binding levels in >120 dye-injected DRG neurons with sensory and electrophysiological properties recordedin vivo. Relative immunointensities for trkA and two TTX-resistant sodium channels (Nav1.8 and Nav1.9) were also measured in these neurons. IB4+ neurons were classified as strongly or weakly IB4+.All strongly IB4+ neurons were C-nociceptor type (C-fiber nociceptive or unresponsive). Of 32 C-nociceptor-type neurons examined, ∼50% were strongly IB4+, ∼20% were weakly IB4+ and ∼30% were IB4−. Aδ low-threshold mechanoreceptive (LTM) neurons were weakly IB4+ or IB4−. All 33 A-fiber nociceptors and all 44 Aα/β-LTM neurons examined were IB4−. IB4+ compared with IB4− C-nociceptor-type neurons had longer somatic action potential durations and rise times, slower conduction velocities, more negative membrane potentials, and greater immunointensities for Nav1.9 but not Nav1.8. Immunointensities of IB4 binding in C-neurons were positively correlated with those of Nav1.9 but not Nav1.8. Of 23 C-neurons tested for both trkA and IB4, ∼35% were trkA+/IB4+ but with negatively correlated immunointensities; 26% were IB4+/trkA−, and 35% were IB4−/trkA+. We conclude that strongly IB4+ DRG neurons are exclusively C-nociceptor type and that high Nav1.9 expression may contribute to their distinct membrane properties.

Spontaneous pain, a poorly understood aspect of human neuropathic pain, is indicated in animals by spontaneous foot lifting (SFL). To determine whether SFL is caused by spontaneous firing in nociceptive neurons, we studied the following groups of rats: (1) untreated; (2) spinal nerve axotomy (SNA), L5 SNA 1 week earlier; (3) mSNA (modified SNA), SNA plus loose ligation of the adjacent L4 spinal nerve with inflammation-inducing chromic gut; and (4) CFA (complete Freund's adjuvant), intradermal complete Freund's adjuvant-induced hindlimb inflammation 1 and 4 d earlier. In all groups, recordings of SFL and of spontaneous activity (SA) in ipsilateral dorsal root ganglion (DRG) neurons (intracellularly) were made. Evoked pain behaviors were measured in nerve injury (SNA/mSNA) groups. Percentages of nociceptive-type C-fiber neurons (C-nociceptors) with SA increased in intact L4 but not axotomized L5 DRGs in SNA and mSNA (to 35%), and in L4/L5 DRGs 1-4 d after CFA (to 38-25%). SFL occurred in mSNA but not SNA rats. It was not correlated with mechanical allodynia, extent of L4 fiber damage [ATF3 (activation transcription factor 3) immunostaining], or percentage of L4 C-nociceptors with SA. However, L4 C-nociceptors with SA fired faster after mSNA (1.8 Hz) than SNA (0.02 Hz); estimated L4 total firing rates were approximately 5.0 and approximately 0.6 kHz, respectively. Similarly, after CFA, faster L4 C-nociceptor SA after 1 d was associated with SFL, whereas slower SA after 4 d was not. Thus, inflammation causes L4 C-nociceptor SA and SFL. Overall, SFL was related to SA rate in intact C-nociceptors. Both L5 degeneration and chromic gut cause inflammation. Therefore, both SA and SFL/spontaneous pain after nerve injury (mSNA) may result from cumulative neuroinflammation.

To test the hypothesis that trkA (the high-affinity NGF receptor) is selectively expressed in nociceptive dorsal root ganglion (DRG) neurons, we examined the intensity of trkA immunoreactivity in single dye-injected rat DRG neurons, the sensory receptor properties of which were identified in vivo with mechanical and thermal stimuli. We provide the first evidence in single identified neurons that strong trkA expression in DRGs is restricted to nociceptive neurons, probably accounting for the profound influence of NGF on these neurons. Furthermore, we demonstrate that trkA expression is as high in rapidly conducting (Aalpha/beta) as in more slowly conducting (Adelta and C) nociceptors. All Aalpha/beta low-threshold mechanoreceptors (LTMs) are trkA negative, although weak but detectable trkA is present in some C and Adelta LTMs. NGF can influence electrophysiological properties of DRG neurons, probably by binding to trkA. We found positive correlations for single identified Aalpha/beta (but not C or Adelta) nociceptors between trkA immunocytochemical intensity and electrophysiological properties typical of nociceptors, namely long action potential and afterhyperpolarization durations and large action potential amplitudes. Furthermore, for Aalpha/beta (notCorAdelta) nociceptors, trkA intensity is inversely correlated with conduction velocity. Similar relationships, again only in Aalpha/beta nociceptors, between electrophysiological properties and trkA expression exist for sodium channel Nav1.8 but not Nav1.9 immunoreactivities. These findings suggest that in Aalpha/beta nociceptors, influences of NGF on expression levels of Nav1.8 are related to, and perhaps limited by, expression levels of trkA. This view is supported by a positive correlation between immuno-intensities of trkA and Nav1.8 in A-fiber, but not C-fiber, nociceptors.

The existence of nociceptors with Abeta-fibers has often been overlooked, and many textbooks endorse the view that all nociceptors have either C- or Adelta-fibers. Here we review evidence starting from the earliest descriptions of A-fiber nociceptors, which clearly indicates that a substantial proportion of cutaneous/somatic afferent A-fiber nociceptors conduct in the Abeta conduction velocity (CV) range in all species in which CV was carefully examined, including mouse, rat, guinea pig, cat and monkey. Reported proportions of A-fiber nociceptors with Abeta-fibers vary from 18% to 65% in different species, usually >50% in rodents. In rat, about 20% of all somatic afferent neurons with Aalpha/beta-fibers were nociceptive. Distributions of CVs of A-fiber nociceptors usually appear unimodal, with a median/peak in the upper Adelta or lower Abeta CV range. We find no evidence to suggest discontinuous differences in electrophysiological or cytochemical properties of Adelta and Abeta nociceptors, rather there are gradual changes in relation to CV. However, some functional differences have been reported. In cat, A-fiber nociceptors with lower mechanical thresholds (moderate pressure receptors) tend to have faster CVs [P.R. Burgess, D. Petit, R.M. Warren. Receptor types in cat hairy skin supplied by myelinated fibers. J. Neurophysiol. 31 (1968) 833-848]. In primate (monkey) A-fiber nociceptors that responded to heat were divided into type I A mechano-heat (AMH) units (Adelta and Abeta CVs) with lower mechanical and higher heat thresholds and may include moderate pressure receptors, and type II AMH units (Adelta CVs) with higher mechanical/lower heat thresholds. It is important that the existence of Abeta nociceptors is recognised, because assumptions that fast conducting, large diameter afferents are always low threshold mechanoreceptors might lead/have led to misinterpretations of data.

This is the first examination of sensory receptive properties and associated electrophysiological properties in vivo of dorsal root ganglion (DRG) neurons that express the TTX-resistant sodium channel Na(v)1.9 (NaN). Intracellular recordings in lumbar DRGs in Wistar rats enabled units with dorsal root C-, Adelta-, or Aalpha/beta-fibers to be classified as nociceptive, low-threshold mechanoreceptive (LTM), or unresponsive. Intracellular dye injection enabled subsequent immunocytochemistry for Na(v)1.9-like immunoreactivity (Na(v)1.9-LI). Na(v)1.9-LI was expressed selectively in nociceptive-type (C- and A-fiber nociceptive and C-unresponsive) units. Of the nociceptive units, 64, 54, and 31% of C-, Adelta-, and Aalpha/beta-fiber units, respectively, were positive for Na(v)1.9-LI. C-unresponsive units were included in the nociceptive-type group on the basis of their nociceptor-like membrane properties; 91% were positive. Na(v)1.9-LI was undetectable in Adelta- or Aalpha/beta-fiber LTM units and in one C-LTM unit. Na(v)1.9-LI intensity was correlated negatively with soma size and conduction velocity in nociceptive units and with conduction velocity in C-fiber units. There was a positive correlation with action potential rise time in nociceptive-type units with membrane potentials equal to or more negative than -50 mV. The data provide direct evidence that Na(v)1.9 is expressed selectively in (but not in all) C- and A-fiber nociceptive-type units and suggest that Na(v)1.9 contributes to membrane properties that are typical of nociceptive neurons.

Intracellular action potentials evoked by dorsal root stimulation were intracellularly recorded from L6 and S1 dorsal root ganglion neurones in deeply anaesthetised guinea-pigs in vivo. Units were classed as C, Adelta or Aalpha/beta units and as nociceptive, low-threshold mechanoreceptive or unresponsive. Units with membrane potentials of at least -40 mV and action potentials with an amplitude of20 mV were included. Nociceptive neurones had significantly larger somatic action potential overshoots than low-threshold mechanoreceptors in C, Adelta and Aalpha/beta units. A higher proportion of low-threshold mechanoreceptors than of nociceptors had action potentials that failed to overshoot in all conduction velocity groups. 60% of muscle spindle afferents failed to overshoot. The size of the overshoot was correlated positively with log(10) action potential duration, log(10) action potential rise time, log(10) afterhyperpolarisation duration, action potential amplitude and membrane potential and negatively (weakly) with log(10) conduction velocity.We conclude that nociceptive neurones are more likely to have somatic action potential overshoots than low-threshold mechanoreceptors in any conduction velocity group. This effect was not due to electrode properties or conduction failure at site(s) of failure of action potential regeneration. Differences in overshoot may affect the influence of neuronal firing on cellular processes. If an overshooting action potential is used as a selection criterion, a bias towards nociceptive neurones is likely to occur. An overshooting action potential coupled with a long afterhyperpolarisation or broad action potential may help in identifying sensory neurones as nociceptive.

Neuropathic pain is associated with changes in the electrophysiological and neurochemical properties of injured primary afferent neurons. A mRNA differential display study in rat L(4/5) dorsal root ganglia (DRGs) revealed upregulation of the calcium channel alpha2delta-1 subunit 2 weeks after partial sciatic nerve ligation (Seltzer model of neuropathic pain). The upregulated transcript appeared to represent previously unidentified sequence from the 3'-untranslated region of rat alpha2delta-1 mRNA. In situ hybridization using L(5) DRGs from sham operated rats showed that 73, 40 and 19% of small (700 microm(2)), medium (700-1100 microm(2)) and large (1100 microm(2)) neuronal profiles, respectively, expressed alpha2delta-1 mRNA. Two weeks following nerve injury there was a significant ipsilateral increase, both in the percentage of DRG neurons expressing alpha2delta-1 mRNA and in the intensity of the hybridization signal. Comparison of this ipsilateral expression with that in sham animals, revealed that for small, medium and large neurons, respectively, the proportion of neurons labelled increased by 1.2-, 1.8- and 2.7-fold, while the hybridization signal in alpha2delta-1-labelled neurons increased by 2.8-, 2.5- and 3.7-fold. The most intensely labelled neuronal profiles in ipsilateral, sham and contralateral DRGs, were generally those with small cross-sectional areas. The alpha2delta-1 auxiliary subunit is known to modulate calcium channel function in heterologous expression systems via its association with the pore-forming alpha1 calcium channel subunit. Therefore the increased levels of this subunit in the populations of primary afferents described may, via modulation of calcium-dependent processes such as neurotransmitter release and neuronal excitability, influence the processing of sensory information.

Partial injury of the rat sciatic nerve elicits a variety of characteristic chemical, electrophysical and anatomical changes in primary sensory neurons and constitutes a physiologically relevant model of neuropathic pain. To elucidate molecular mechanisms that underlie the physiology of neuropathic pain, we have used messenger RNA differential display to identify genes that exhibit increased ipsilateral expression in L4/5 dorsal root ganglia, following unilateral partial ligation of the rat sciatic nerve. One set of partial complementary DNA clones identified in this screen was found to encode a protein kinase, nerve injury-associated kinase. Cloning of the full-length human nerve injury-associated kinase complementary DNA, together with recombinant expression analysis, reveal nerve injury-associated kinase to be a functional member of a subgroup of sterile 20-like protein kinases characterised by the presence of a putative carboxy terminal autoregulatory domain. Induction of nerve injury-associated kinase expression in dorsal root ganglia in the rat neuropathic pain model was confirmed by quantitative reverse transcription–polymerase chain reaction, and RNA in situ hybridization analysis revealed enhanced levels of nerve injury-associated kinase within neurons. Together, our data implicate nerve injury-associated kinase as a novel upstream component of an intracellular signalling cascade that is up-regulated in dorsal root ganglia neurons in response to sciatic nerve injury.

1. Intracellular voltage recordings were made from the somata of L6 and S1 dorsal root ganglion (DRG) neurones at 28.5-31 C in young guinea-pigs (150-300 g) anaesthetized with sodium pentobarbitone. Action potentials (APs) evoked by dorsal root stimulation were used to classify conduction velocities (CVs) as C, Adelta or Aalpha/beta. Units with overshooting APs and membrane potentials (Vm) more negative than -40 mV were analysed: 40 C-, 45 Adelta- and 94 Aalpha/beta-fibre units. 2. Sensory receptive properties were characterized as: (a) low-threshold mechanoreceptive (LTM) units (5 C-, 10 Adelta- and 57 Aalpha/beta-fibre units); (b) nociceptive units, responding to noxious mechanical stimuli, some also to noxious heat (40 C-, 27 Adelta- and 27 Aalpha/beta-fibre units); (c) unresponsive units that failed to respond to a variety of tests; and (d) C-fibre cooling-sensitive units (n = 4). LTM units made up about 8 % of identified C-fibre units, 36 % of identified Adelta-fibre units and > 73 % of identified Aalpha/beta-fibre units. Compared with LTM units, the nociceptive units had APs that were longer on average by 3 times (C-fibre units), 1.7 times (Adelta-fibre units) and 1.4 times (Aalpha/beta-fibre units). They also had significantly longer rise times (RTs) and fall times (FTs) in all CV ranges. Between Aalpha/beta-nociceptors and Aalpha/beta-LTMs there was a proportionately greater difference in RT than in FT. The duration of the afterhyperpolarization measured to 80 % recovery (AHP80) was also significantly longer in nociceptive than LTM neurones in all CV ranges: by 3 times (C-fibre units), 6.3 times (Adelta-fibre units) and 3.6 times (Aalpha/beta-fibre units). The mean values of these variables in unresponsive units were similar to those of nociceptive units in each CV range; in C- and Adelta-fibre groups their mean AHP duration was even longer than in nociceptive units. 3. A-fibre LTM neurones were divided into Adelta- (D hair units, n = 8), and Aalpha/beta- (G hair/field units, n = 22; T (tylotrich) hair units, n = 6; rapidly adapting (RA) glabrous units, n = 6; slowly adapting (SA) hairy and glabrous units, n = 2; and muscle spindle (MS) units n = 17). MS and SA units had the shortest duration APs, FTs and AHP80s of all these groups. The mean RT in D hair units was significantly longer than in all Aalpha/beta LTM units combined. T hair units had the longest mean FT and AHP of all the A-LTM groups. The mean AHP was about 10 times longer in T hair units than in all other A-LTM units combined (significant), and was similar to that of A-fibre nociceptive neurones. 4. These differences in somatic AP shape may aid in distinguishing between LTM and nociceptive or unresponsive C- and Adelta-fibre units but probably not between nociceptive and unresponsive units. The differences seen may reflect differences in expression or activation of different types of ion channel.

Dorsal root ganglion neurons innervating skin via the saphenous nerve, muscle via the gastrocnemius nerve and viscera via the splanchnic nerve, were identified by retrograde tracing with Fast Blue applied to the cut nerve. Only neuronal profiles with nuclei were counted. At the survival times used no changes in immunohistochemical labelling patterns were detectable in the axotomized neurons. Percentages of Fast Blue-labelled neuronal profiles that were immunolabelled were calculated. The values for markers of carbohydrate groups were for skin, muscle and viscera, respectively: the lectin peanut agglutinin 55%, 24%, and 50%; the lectin soybean agglutinin 72%, 56%, 61%; the antibody 2C5 (against lactoseries groups) 43%, 20%, 6%; the antibodies SSEA-4 (against globoseries groups) 6%, 12%, 0% and SSEA-3 (against globoseries groups) 6%, 5%, 0%. The values for neurofilament rich profiles were for skin, muscle and viscera, respectively: 34%, 43%, 19%, and for carbonic anhydrase were 10%, 33%, 2%. Values for neuropeptides were, for calcitonin gene-related peptide 51%, 70%, 99%, for substance P 21%, 51%, 82%, and for somatostatin 10%, 2% and 0%. The population of skin afferents therefore contained the highest proportion of profiles expressing galactose containing carbohydrate groups labelled by 2C5 and the lectins and the highest proportion of cells with somatostatin. In contrast they had the lowest proportions of cells with calcitonin gene-related peptide and substance P, compared with the other tissues. Muscle afferents had the highest proportions compared with the other tissues of the neurofilament-rich, carbonic anhydrase-positive and SSEA-4-labelled profiles, but the lowest proportions of profiles with lectin binding. The splanchnic visceral afferents had the highest proportions, compared with the other tissues, of neuronal profiles labelled for calcitonin gene-related peptide and substance P, but the lowest proportions of neurofilament rich profiles and of profiles with carbonic anhydrase or 2C5 labelling and they totally lacked any labelling for globoseries carbohydrates and somatostatin. Both the muscle and skin afferent populations had clear small cell and large cell peaks in their size distributions, with the small cell peak being larger for skin than muscle afferents and the large cell peak being more marked for muscle afferents. The visceral afferent profiles had a unimodal size distribution with the peak size being between the small and large cell peaks of the somatic afferent units. This study therefore shows that the patterns of immunohistochemical labelling and cell size of primary afferent neurons differ according to their peripheral target tissue.

Intracellular voltage recordings were made in vitro at 36.5 ± 1°C from 35 rat lumbar dorsal root ganglion (DRG) neurones with a peripheral conduction velocity (CV) in the C-fibre range (0.3–2.2 m/s). The peripheral nerve (PN) was stimulated in one of three different ways, each delivering single stimuli (0.1–1 ms duration, 2–3-times threshold; maximum 50 V) at a low frequency (0.3 Hz). With each of the three stimulation methods used here a similar proportion of cells (approximately 30%) showed changes, either an abrupt latency change or a soma invasion by two action potentials (APs). Both of these changes were consistent with branching of primary afferent C-fibres in the PN.

Immunocytochemical properties of dorsal root ganglion neurons innervating the kidney were studied with retrograde tracing of Fluorogold or Fast Blue dyes applied to the cut renal nerves in the rat. The proportions and sizes of renal afferent neurons labelled with a variety of markers were quantified in T9-L1 dorsal root ganglia from five rats. Compared with the overall size distribution in these ganglia, renal afferent neurons were mainly small with a few medium-sized neurons. The majority (79%) of renal afferent dorsal root ganglion neuronal somata were unlabelled by an anti-neurofilament antibody, RT79, and classified as neurofilament-poor with probable C-fibres. These had an approximately normal distribution of cell sizes. Only 21% were RT79-positive and classified as neurofilament-rich with probable A-fibres, and even these were small to medium sized cells, consistent with them being mostly Aδuˆfibre neurons. Percentages of renal afferent neurons showing labelling were as follows: peripherin-like immunoreactivity: 69%; calcitonin-gene related peptide: 93%; substance P: 37%; the lectins soybean agglutinin: 57% and peanut agglutinin: 68%; Calbindin D 28k -like immunoreactivity: 21% (only weak labelling); carbonic anhydrase like immunoreactivity: 0%. There were differences between probable C-fibre and probable A-fibre neurons, classified according to their labelling with RT97. The percentages of RT97-negative and RT97-positive neurons respectively labelled with the other markers were as follows: peripherin-like immunoreactivity: 82%, 25%; calcitonin gene-related peptide-like immunoreactivity: 99%, 79%; substance P-like immunoreactivity: 43%, 0%; soybean agglutinin: 69%, 24%; peanut agglutinin: 76%, 47%; calbindin-like immunoreactivity: 26%, 0%. Thus, the biggest differences between the probable A-and C-fibre renal afferent neurons were in their peripherin, substance P and calbindin contents. Thus, renal afferent neurons in the dorsal root ganglion are not homogeneous and it is suggested the differences may relate to the known different afferent receptor types within the kidney. It is suggested that the low proportion of neurons with substance P-like immunoreactivity in the renal afferent innervation compared to that of other viscera may relate to the role of the renal vasculature in urine formation.

Neurofilament immunoreactivity was examined in spinal cords of rats and cats with antibodies to all three subunits (68 kD, 155 kD and 200 kD) and to different phosphorylation states of 200 kD. NFHP-, an antibody against non-phosphorylated 200 kD, labelled all rat neuronal perikarya but failed to labet cat neurofilaments. In both species, the perikarya and processes of motoneurones were immunoreactive for all three subunits but most dorsal horn neuronal perikarya were not immunoreactive for 68 kD and 155 kD. Motoneuronal perikarya and proximal processes showed filamentous labelling for 68 kD but not for 155 kD in the rat, while in neither species did these show labelling with RT97, an antibody against a highly phosphorylated form of 200 kD; immunoreactivity for 200 kD was present in both filamentous (probably partially phosphorylated) and non-filamentous (non-phosphorylated) forms, but in dorsal horn neurones only the latter was present. Interpretations consistent with this data are: in rat and possibly also cat, motoneuronal neurofilaments consist of a 68 kD backbone with partially phosphorylated 200 kD sidearms, with both 155 kD and 200 kD (non-phosphorylated) subunits in a non-filamentous form; this neurofilament becomes more highly phosphorylated along the proximal processes. The dorsal horn neurones probably contain 200 kD in a non-filamentous form but may lack the other subunits.

Two populations of cat L4 dorsal root ganglion (DRG) neurones were apparent from their Nissl staining with Toluidine blue. One had neurones of all sizes and the other had predominantly small neurones. The size distribution of neuronal profiles in the two populations overlapped and both were approximately normal. They corresponded to the light (L) and small dark (SD) cell populations previously described in rat DRGs. These neurones were examined with four antibodies to neurofilament: RT97, NFH, 155 and anti-68kD. RT97 is specific for the phosphorylated form of the 200 kDa subunit; NFH recognises both phosphorylated and non-phosphorylated forms of this subunit; 155 and anti-68 kDa recognise the 155 kDa and 68 kDa subunits respectively. The clearest differential labelling was seen with NFH and RT97 and this labelling was compared with cell size. High intensity NFH labelling was in a population of neuronal profiles of all sizes and low intensity labelling in a population of predominantly small neuronal profiles. These populations corresponded respectively to the L and SD populations seen with toluidine blue staining. In the rat, these populations can be demonstrated by both NFH and RT97. In contrast in the cat, high intensity RT97 labelling was seen in only 75% of the L neuronal profiles defined with NFH and was also seen in some SD neuronal profiles defined with NFH. It is thus proposed that L and SD cell types are present in the cat DRG and can be demonstrated using the anti-neurofilament marker, NFH.

This paper reviews and provides new data on the relationship of the peptide content in rat dorsal root ganglion (DRG) neurons to a) the neurofilament content of the soma and b) the conduction velocities of the fibres. The latter involved intracellular recordings made in vitro followed by dye injection and immunocytochemistry. Because neurofilament-poor DRG neurones have C-fibres, and A-fibre neurones are neurofilament rich, the soma neurofilament content of peptide containing neurones allowed predictions to be made about their conduction velocity ranges. Substance P-like immunoreactive (SP-LI) neurones were mostly small, neurofilament poor, but a few (15%) were neurofilament rich. From conduction velocity measurements, about half the C-fibre neurones studied and 10% of A delta-neurones but no A alpha/beta-neurones showed SP-LI. CGRP-LI neurones were also mainly neurofilament poor neurones, but 32% were neurofilament rich, including small, medium, and large neurones. Fibres of CGRP-LI neurones conducted in the C, A delta or A alpha/beta ranges. Neurones with somatostatin-LI (SOM-LI) were all neurofilament poor; preliminary data is consistent with SOM-LI neurones having C-fibres.

1. Intracellular recordings were made in dorsal root ganglia in vitro at 37 degrees C. The L4, L5 and L6 ganglia from 46- to 51-day-old female Wistar rats were used. In each neuron conduction velocity (CV) was measured and fluorescent dye was injected. Later the intensity of the immunoreactivity to RT97 (a monoclonal antibody to the phosphorylated 200 kDa neurofilament subunit) as well as the cell size (cross-sectional area at the nuclear level) were measured in the dye-injected neurons. RT97 was used to distinguish between the L (light, neurofilament-rich) and the SD (small dark, neurofilament-poor) neuronal somata. 2. Neurons were classified as C neurons (CV less than 1.3 m/s), C/A delta neurons (1.3-2 m/s), A delta neurons (2-12 m/s) or A alpha/beta neurons (greater than 12 m/s). 3. All A-fibre somata were RT97 positive (L) and all C-fibre somata were RT97 negative (SD), although in the C/A delta group both positive and negative neurons were seen. Thus, RT97-negative somata had C (unmyelinated) or C/A delta fibres, while RT97-positive somata had A (myelinated) or C/A delta fibres. 4. The size distributions of A neurons and C neurons were consistent with their classification as L- and SD-cell neurons respectively. The size distribution of A delta cells was skewed with a peak of small cells and a tail of medium-sized cells. 5. There was a loose positive correlation between cell size and fibre CV. 6. RT97 intensity was positively correlated with CV if all neurons were considered together, but no correlation was seen within the C, A delta or A alpha/beta CV groups. 7. RT97 intensity was positively correlated with cell size when all neurons were considered together. Although no correlation was seen within the C or the A delta CV groups, a clear positive correlation was seen for A alpha/beta neurons. 8. The relationship of RT97 intensity to cell size was not demonstrably altered by axotomy, time in vitro or the presence of intracellular dye in control experiments. 9. RT97-negative and -positive neurons could be seen in neonatal rat ganglia. Their size distributions resembled, respectively, the SD- and L-neuron populations at this age. RT97 immunoreactivity may therefore be a useful predictor of the cell type and myelinated state which a sensory cell is destined to reach in the adult rat.

Capsaicin (0.01-50 microM) was applied to adult rat vagus nerve in vitro to examine the C-fibre conduction block and to compare its time course with that of the depolarisation caused by this drug. The conduction block was assessed by the reduction in the C-wave of the compound action potential. Capsaicin caused a dose-dependent decrease in the height and area under the C-wave; the threshold dose was between 0.01 and 0.3 microM and maximum C-wave reduction of about 85% occurred with doses of 5 microM or above. The C-wave reduction was divided into reversible and irreversible components which differed in dose dependency. The threshold for the reversible block was below 0.3 microM and for the irreversible block it was about 1 microM. The onset of the block took about 5 min. regardless of dose. Where there was more than 50% recovery after the block the reversible component was measured. This was dose dependent and its duration was 10-90 min. Removal of external calcium did not affect the magnitude of the C-wave block, although it did seem to prevent recovery of the C-wave.

Anterograde transport of horseradish peroxidase (HRP) through somatic and visceral nerves was used to estimate the proportions of somatic and visceral dorsal root ganglion (DRG) cells of the lower thoracic ganglia of the cat. A concentrated solution of HRP was applied for at least 5 hours to the central end of the right greater splanchnic nerve and of the left T9-intercostal nerve of adult cats. Some animals remained under chloralose anaesthesia for the duration of the HRP transport time (up to 53 hours) whereas longer HRP application and transport times (4-5 days) were allowed in animals that recovered from barbiturate anaesthesia. Visceral DRG cells were found in approximately equal numbers in all ganglia examined (T7-T11). Population estimates were obtained for the T8 and T9 ganglia where visceral DRG cells were found to be 6.2% (T8) and 5.2% (T9) of the total cell population. In contrast, somatic DRG cells were found in large numbers in the ganglia examined (T8 and T9) where they amounted to over 90% of the cell population. Measurement of cross-sectional areas and estimates of cell diameters of the DRG cells showed greater proportions of large somatic cells (diameter greater than 40 micron) than of large visceral cells. Similar distributions of cell size were found for both somatic and visceral DRG cells with diameters less than 40 micron. These results show that the proportion of visceral afferent fibres in the dorsal roots that mediate the spinal cord projection of the splanchnic nerve is very small. Since viscerosomatic convergence in the thoracic spinal cord is very extensive, the present results suggest considerable divergence of the visceral afferent input to the central nervous system.

Conduction velocities of rat L4, L5 and L6 dorsal root ganglion neurons were measured in vitro, from several points on the peripheral nerve and dorsal root. Conduction velocities calculated from a single stimulation point (12-26 mm from the ganglion) proved accurate for fibres conducting up to 17 m/s in the peripheral nerve and up to 14 m/s in the dorsal root, but tended to underestimate the value for faster fibres. C-fibre neurons of the L4 and L5 ganglia had a unimodal distribution of conduction velocities below 1.3 m/s. These were discontinuous with A-fibre conduction velocities, which also had a unimodal distribution and had no clear A delta peak. In contrast, conduction velocities of L6 ganglion neurons showed no discontinuity between C- and A-fibres, but had a clear A delta peak. In A-fibre neurons, dorsal root conduction velocities were on average about 14% slower than, and linearly related to, those in the peripheral nerve. However, in individual neurons the dorsal root conduction velocity could be slower or faster than that in the peripheral nerve. In C-fibre neurons dorsal root conduction velocities were almost always slower (average 28%) but not correlated with those of the peripheral nerve. Slowing of conduction velocity along the sciatic nerve was seen in most fibres conducting at less than 2 m/s, but not in faster fibres. The slowing was substantial (up to 60%), sometimes from the A delta to the C-fibre range, and sudden, occurring at a distance of between 15 and 29 mm from the ganglion.

Sensory neurones are commonly classified into conduction velocity (CV) groups, the C-, Aδ - and Aα / β - fibre groups. If the functional significance of these groups is to become clear, the borderlines between the groups need to be clearly defined. However, fibres are sometimes classified as C on the basis of their CV in the periphery, although, as we show below, some of these may conduct in the Aδ range nearer the ganglion. Furthermore, an assumption is sometimes made that small dorsal root ganglion (DRG) neurones have C-fibres and thus that some peptides found predominantly in small neurones are only in C-fibre cells. However it has been shown in the rat that, although C-fibres arise from small neurones, A-fibres can arise from all sizes of neurones including the very small ones (Harper and Lawson 1985). Therefore, we have correlated the fibre CV of neurones directly with their immunoreactivity. The antibodies we have used are anti-substance P, anti-calcitonin gene related peptide, 2C5 and RT97.

Partial nerve injury leads to peripheral neuropathic pain. This injury results in conducting/uninterrupted (also called uninjured) sensory fibres, conducting through the damaged nerve alongside axotomised/degenerating fibres. In rats seven days after L5 spinal nerve axotomy (SNA) or modified-SNA (added loose-ligation of L4 spinal nerve with neuroinflammation-inducing chromic-gut), we investigated (a) neuropathic pain behaviours and (b) electrophysiological changes in conducting/uninterrupted L4 dorsal root ganglion (DRG) neurons with receptive fields (called: L4-receptive-field-neurons). Compared to pretreatment, modified-SNA rats showed highly significant increases in spontaneous-foot-lifting duration, mechanical-hypersensitivity/allodynia, and heat-hypersensitivity/hyperalgesia, that were significantly greater than after SNA, especially spontaneous-foot-lifting. We recorded intracellularly in vivo from normal L4/L5 DRG neurons and ipsilateral L4-receptive-field-neurons. After SNA or modified-SNA, L4-receptive-field-neurons showed the following: (a) increased percentages of C-, Aδ-, and Aβ-nociceptors and cutaneous Aα/β-low-threshold mechanoreceptors with ongoing/spontaneous firing; (b) spontaneous firing in C-nociceptors that originated peripherally; this was at a faster rate in modified-SNA than SNA; (c) decreased electrical thresholds in A-nociceptors after SNA; (d) hyperpolarised membrane potentials in A-nociceptors and Aα/β-low-threshold-mechanoreceptors after SNA, but not C-nociceptors; (e) decreased somatic action potential rise times in C- and A-nociceptors, not Aα/β-low-threshold-mechanoreceptors. We suggest that these changes in subtypes of conducting/uninterrupted neurons after partial nerve injury contribute to the different aspects of neuropathic pain as follows: spontaneous firing in nociceptors to ongoing/spontaneous pain; spontaneous firing in Aα/β-low-threshold-mechanoreceptors to dysesthesias/paresthesias; and lowered A-nociceptor electrical thresholds to A-nociceptor sensitization, and greater evoked pain.