Subcutaneous formalin and intraplantar carrageenan increase nitric oxide release as measured by in vivo voltammetry in the spinal cord.

The paper describes in vivo voltammetric detection of nitric oxide with carbon fibre microelectrodes at the lumbar spinal dorsal horn level of decerebrated-spinalized rats during peripheral noxious inflammatory processes. At the lumbar (L3-L4) dorsal horn level, a nitric oxide dependent peak of oxidation current (650 mV), remaining stable for up to 4h ((92 +/- 5)% of control) could be detected indicating that significant amounts of nitric oxide are produced continuously. Following subcutaneous injection in the hindpaw of 50 microl of 0.5% formalin the oxidation current rapidly increased ((115 +/- 5)% of control at 25 min) and reached (120 +/- 6)% of control 1h later. Subsequently the voltammograms stabilized for up to 90 min and decreased ((107 +/- 4)% at 124 min). After an injection in the hindpaw of 150 microl of 4% carrageenan, the voltammograms remained at control level for 1h and then the oxidation current increased continuously for up to 4h ((145 +/- 16)% of control at 240 min); such an increase was reversed by ketamine. In these two models of inflammation, the delay in onset and the duration of the increases in NO release within the dorsal horn relate, to some extent, to the time course of the peripheral inflammatory processes, since they are shorter after formalin than after carrageenan. The results provide a direct in vivo demonstration that the intercellular messenger nitric oxide participates in the transmission of noxious afferent messages within the dorsal horn of the spinal cord following peripheral inflammation.

Nitric oxide (NO) release by glutamate and NMDA in the dorsal horn of the spinal cord: an in vivo electrochemical approach in the rat.

Glutamate acts as a neurotransmitter of primary afferent messages in the spinal cord. Through glutamatergic mechanisms nitric oxide (NO) is also a potential intermediary in the transmission of sensory messages, particularly nociceptive, at the spinal level. The aim of the present study was, by using electrochemical monitoring of NO, to determine if the activation of glutamatergic transmission, particularly through NMDA receptors, could increase NO production within the dorsal horn of the lumbar spinal cord in the rat. 30 micrometers diameter treated carbon fiber electrodes coated with nickel-porphyrine and Nafion(R), and associated with differential normal pulse voltammetry, have been used in vivo to monitor NO within the dorsal horn of the lumbar spinal cord of decerebrated-spinalized rats. A NO-dependent peak of oxidation current (650 mV vs. Ag-AgCl), remaining stable for up to 3 h (+/-5%) could be detected under basal conditions, which indicates that significant amounts of NO are produced continuously. The non-competitive N-methyl-d-aspartate (NMDA) channel blockers, Ketamine (100 mg kg-1 i.p.) and MK-801 (10 mg kg-1 i.p.), decreased the voltammograms to 70+/-5% and 69+/-2% of controls at 120 min, respectively. Glutamate (10 mM), when directly superfused upon the spinal cord (20 min at 50 microliters min-1) induced a rapid and significant increase of the 650 mV peak, with a maximum at around 90 min (148+/-6% of control) followed by a slow decay (138+/-4% of control at 150 min). This increase could be totally reversed or blocked by i.p. injection of 100 mg kg-1 of Ketamine. NMDA (30 mg kg-1 i.p.) induced a long-lasting increase in the peaks (149+/-11% at 90 min and 162+/-20% at 120 min), which was also fully reversed by Ketamine or MK-801. These results provide in vivo direct evidence of a glutamate- and/or NMDA-induced release of NO at the spinal level, and is discussed in relation to the glutamatergic transmission of primary afferent messages.

Cholinergic modulation of the picrotoxin-induced electrocorticographical events and behavioral "pain-like" symptoms at somatomotor cortical level in the rat.

In this study, we examined the modulation by acetylcholine of electrocorticographical (ECoG) ictal events and spontaneous pain-like behaviors following cortical application of the GABA(A) antagonist picrotoxin in the awake rat. Distilled water as vehicle, the cholinomimetic substance eserine, and the general muscarinic antagonist atropine were microinjected 10 min before the second microinjection of 2 microg picrotoxin into the hind paw region of the somatomotor cortex (SmI). Under these conditions, we observed that eserine (physostigmine, 1 microg, 10 microg, and 20 microg) did not consistently modify the number of the picrotoxin-induced ECoG spikes and bursts, but instead produced a massive enhancement of the number of hind paw licks compared with vehicle at 10 microg and, to a lesser extent, the number of the stereotyped "turn-in" and "neglected" paws following picrotoxin. In contrast, atropine (1 microg, 10 microg, and 20 microg) increased the number of the picrotoxin-induced spikes and bursts at 10 microg and, at all doses, decreased the number of the picrotoxin-induced pain-like symptoms. Statistically significant changes for the number of paw lifts, licks, and "turn-in" paws were observed only with 10 microg. These results tend to show that epilepsy and pain are not strictly related to each other and also emphasize the cortex as a target for interactions between GABA and acetylcholine relative to "central" pain.

Picrotoxin produces a "central" pain-like syndrome when microinjected into the somato-motor cortex of the rat.

In this study, we report the possibility of producing marked electrocorticographic changes and "pain-like" reactions, when the GABAA antagonist picrotoxin is microinjected unilateraly into the rat somato-motor Sml cortex in the region of the hind paw. After the microinjection, we observed continuous seizure isolated spikes, spikes-and-waves, bursts, and pain-like reactions, almost exclusively confined to the hind paw. These reactions considered of lifting off the floor, licking of the paw palm or digits, biting, paw tremors, and a peculiar paw position that we called "turn-in" paw. We also noted other behaviors, such as "limping," "neglected" paw, or rearing. The "pain-like" character of these manifestations was suggested by the fact that similar qualitative and quantitative data occurred consequent to the administration of 2.5% diluted formalin into the palm of the hind paw in different rats. Bringing together the electrocorticographic events and the behavioral reactions produced by Sml picrotoxin indicated that there was no obvious correlation between the phenomena, except that the tremor was always associated with the bursts. Sensory denervation of the hind paw, produced by sciatic and saphenous nerve transections, did not significantly modify either the ictal activity or the behavior. Finally, microinjection of naloxone prior to picrotoxin did not change the cortical events, but greatly diminished the "pain-like" reactions. All these results favor the cortical microinjection of a GABAA receptor antagonist as a good rat model for studying pain of "central" origin. They emphasize the possible role of the Sml cortex in such a phenomenon, and the deficit of cortical GABAergic processing, which can include an opioid link.

The "plantar test" apparatus (Ugo Basile Biological Apparatus), a controlled infrared noxious radiant heat stimulus for precise withdrawal latency measurement in the rat, as a tool for humans?

In the present study, we precisely and automatically measured the withdrawal latency to noxious radiant heat application in unrestrained male rats and in human subjects of both sexes, by means of the "plantar test" apparatus (Ugo Basile Biological Apparatus). The infrared light stimulus of this tool was applied underneath the hindpaws of rats and the middle fingers of human subjects. With one right and one left stimulation every 10 min, we observed a decrease in latency over a 40-min testing period in rats; the latency reached a mean value of 5.08 +/- 0.25 sec after 40 min with a 36-W stimulus, which corresponded to 46.5 degrees C. In pilot experiments, also performed on rats, we showed that the opiate morphine (10 mg/kg, i.p.) produced remarkable increases of the withdrawal latency only in "naive" animals (i.e., ones that had never experienced the plantar test stimulus) and not in animals "habituated" to it. Among humans, we noted gender differences, such as less sensitivity to the infrared noxious radiant heat for women, particularly during the menstrual period. A difference from rats was that there was no significant latency modification along the 40-min testing period for either women or men, with a mean latency of 5.61 +/- 0.18 sec (47.5 degrees C) for the women and 4.39 +/- 0.10 sec (45.5 degrees C) for the men. These data confirm the reliability of the plantar test in rats, and demonstrate the possible use of an infrared source in human subjects as a noxious heat stimulus; the withdrawal reaction to this stimulus is emphasized as a good index of nociception in humans.

Does barbiturate anesthesia modify the neuronal properties of the somatosensory thalamus? A single-unit study related to nociception in the awake-pentobarbital-treated rat.

By means of extracellular recordings, we studied thalamic ventrobasal complex neurons of rats tested first awake, and then anesthetized with pentobarbital. In both conditions, we found two groups of units in both states. The first group, displaying a spontaneous bursting activity, was not obviously responding to peripheral stimuli. Another group, displaying a single-spike activity, was almost exclusively activated by innocuous and/or noxious and innocuous mechanical stimuli. Still in this group, units specifically driven by noxious stimuli were only found under pentobarbital. These data, different from classical findings, emphasize the interest of the awake preparation in order to study nociceptive cellular mechanisms at the thalamic level.

Single-unit recordings at dorsal raphe nucleus in the awake-anesthetized rat: spontaneous activity and responses to cutaneous innocuous and noxious stimulations.

In this study, we recorded the single-unit activity of the dorsal raphe nucleus (DRN) in rats tested first awake and, a few days later, anesthetized with sodium pentobarbital and recorded again. This was achieved by means of a small chronically implanted device supporting a 25 micron platinum-iridium wire as the recording electrode. In both the awake and anesthetized conditions, and in agreement with most of the studies performed at the DRN level, we found that a vast majority of the units, displaying small amplitude and long-duration action potentials, possessed a low level of spontaneous activity (0.2-4 Hz). Among these units, found in greater number under pentobarbital, it was possible to establish that this activity was regular or irregular, in accordance with the literature reports. However, as opposed to these studies, we determined that the 'regularity' was relative, only noticeable in more or less prolonged phases of activity. In particular, we never recorded the so-called 'clock-like' activity, largely reported as an unambiguous criterion for selecting the serotoninergic neurons. In both the awake and anesthetized conditions, the responses of the DRN neurons to peripheral mechanical innocuous and noxious stimulations were observed in only one-half of the units recorded and were weak in comparison to other results that we obtained at the nucleus raphe magnus level in previous studies. When present, these responses were excitation or inhibition, occurring during or after the stimulus application. These results question the direct involvement of the DRN in acute nociception.

Are ventromedial medulla neuronal properties modified by chronic peripheral inflammation? A single-unit study in the awake, freely moving polyarthritic rat.

In the present work, we recorded the neuronal properties of the ventromedial medulla, a brainstem structure involved in the descending spinal control systems related to nociception, in awake, freely moving healthy and polyarthritic rats. These animals were rendered polyarthritic with a subcutaneous administration of the Freund's adjuvant into the tail, and studied at 20 and 30 days post-inoculation. At the ventromedial medulla level, the single-unit activities were recorded by means of a chronically implanted device supporting a 50 microns platinum-iridium wire as the recording electrode. With a total of 308 recorded neurons, we determined that in both healthy rats, i.e. animals having received mineral oil only and arthritic rats, there were ventromedial medulla units with common physiological properties, but also changes. In agreement with the results from anesthetized arthritic rats at spinal and thalamic levels, the systematic analysis of the responses to light touch and mechanical shock revealed that the 'multimodal, multireceptive' units, excited by innocuous and noxious stimuli, were much more responsive to both modalities in arthritic rats. Approximately 7% of these neurons displayed a 'paroxysmal' spontaneous activity, also reported in the literature for other structures. In addition, we recorded a significant number of neurons inhibited or excited-inhibited by innocuous and noxious cutaneous stimulations, and a few with a regular spontaneous activity, also responding, which has never been the case in healthy rats. We conclude that a peripheral chronic inflammation, such as arthritis, can produce changes of the ventromedial medulla neuronal properties, as compared to healthy animals. Consequently, in addition to its classical role in the spinal control of nociception, the ventromedial medulla is able to develop some form of plasticity in the case of persistent pain of peripheral origin.

The GABAA receptor antagonist picrotoxin induces a 'pain-like' behavior when administered into the thalamic reticular nucleus of the behaving rat: a possible model for 'central' pain?

In this study, we reported that the microinjection of the GABAA antagonist picrotoxin into the rat thalamic reticular nucleus produced a 'pain-like' behavior. This behavior was primarily characterized by repetitive lifting off the hindpaw from the floor contralateral to the injection site, sometimes accompanied by extension of the leg and maximal fingers separation. Surprisingly, these manifestations were not occurring when picrotoxin was applied to the ventrobasal complex itself, alternatively producing 'wet-dog' shakes. These data show that the local administration of picrotoxin is a relevant approach for studying pain of 'central' origin and complex GABAergic modulatory mechanisms within the thalamic sensory complex.

Involvement of ventromedial medulla "multimodal, multireceptive" neurons in opiate spinal descending control system: a single-unit study of the effect of morphine in the awake, freely moving rat.

In the present work, we have studied the effects of systemic morphine on the electrophysiological properties of ventromedial medulla (VMM) neurons in the awake, freely moving rat. By means of a chronically implanted single-unit recording device, a drug delivery catheter, and the use of controlled innocuous and noxious cutaneous stimuli, we were able to study precisely the spontaneous and evoked VMM neuronal activities. We have particularly focused our attention upon the VMM "multimodal, multireceptive" units, excited by non-noxious and noxious stimuli (VMM MULT ON), which we have already determined as the neuronal class potentially involved in nociceptive processes at VMM level. We found that morphine (3 mg/kg, i.v.) does not affect the spontaneous activity of these neurons whereas their responses to noxious heat are strongly attenuated (70%), over a prolonged period (about 2 hr) associated with an increase in the response latency. This action of morphine appears to be pharmacologically specific since it is dose dependent to some extent, and is reversed by 0.3 mg/kg of naloxone. In parallel with this pharmacological specificity, we have also demonstrated a preferential physiological effect since the response of the VMM MULT ON neurons to light touch application is not affected by morphine. This specificity is emphasized by the fact that morphine does not modify the activity of the other VMM neuronal groups such as the units unresponsive to any kind of peripheral stimuli, and does not reveal "new" neuronal classes such as those we have found in previous studies after barbiturate administration. The differential effect upon the noxious versus innocuous inputs of these units produced by the opiate reinforces their participation in nociceptive processing since similar effects have been reported in well-known nociceptive somatosensory structures such as the dorsal horn of the spinal cord. Furthermore, although the precise mechanisms of action have not yet been determined, the spinal projection of the VMM MULT ON neurons, previously demonstrated by our group, suggests their involvement in an opiate descending spinal control system of nociception. Although speculative, one can imagine either a direct facilitatory MULT ON spinal effect being attenuated by morphine (disfacilitation), or a morphine-induced disinhibition of inhibitory GABAergic neurons acting upon the MULT ON neurons.

Drastic changes of ventromedial medulla neuronal properties induced by barbiturate anesthesia. I. Comparison of the single-unit types in the same awake and pentobarbital-treated rats.

By means of single-unit recordings, as we have already performed in other studies, we have found that in the awake, drug-free, freely moving rat, there is only one neuronal class potentially involved in nociception and its control at the ventromedial medulla level (VMM, a structure involved in the spinal descending control systems of nociception): the 'multireceptive multimodal' units. These neurons are always activated by very light mechanical (air puff, light touch) and mechanical (pinch, pin-prick) or thermal noxious stimuli, in addition to an auditory stimulus. During identical VMM penetrations, performed in the same animals tested first awake and then anesthetized a few days later with 30 mg/kg of i.p. pentobarbital, we once again found the 'multireceptive multimodal' units, but this time with physiological properties that were strongly modified: in particular, we noted a disappearance of the nociceptive responses consecutive to a strong noxious heat pulse application (36-51 degrees C), associated sometimes with a reduction of the responses due to innocuous stimulation. This is in agreement with the classical effects of barbiturates. In light of previous observations reported in the literature devoted to the VMM physiology in the anesthetized rat, the most important observation in our study was that, with pentobarbital anesthesia, we recorded 'new' neuronal classes as compared to the awake condition. In these classes, which appeared to be qualitatively similar to those already reported under anesthesia, we found the units exclusively driven by innocuous stimulation (excited for the majority), the units specifically driven by noxious stimulation (half excited, half inhibited) and a 'multireceptive multimodal' group inhibited or excited-inhibited by non-noxious and noxious stimuli (half of the multireceptive group). All these data demonstrate that barbiturate anesthesia strongly modifies the VMM physiology in relation to nociception. Furthermore, since our results, that were obtained in anesthetized rats, were qualitatively identical to those described in the literature under similar experimental conditions, they raise the question of the appropriateness of using a barbiturate anesthetic in order to study the cellular mechanisms related to nociception at this level. In addition, these findings indicate that the obtention of only one neuronal class in the awake, drug-free, freely moving rat (the excited 'multireceptive' neurons) is not due to an experimental bias, which strongly emphasizes the reliability of using awake animals. However, it remains to be determined by which mechanisms pentobarbital 'distorts' the VMM physiology as compared to the normal, standard physiological conditions of the awake animal.

Drastic changes of ventromedial medulla neuronal properties induced by barbiturate anesthesia. II. Modifications of the single-unit activity produced by Brevital, a short-acting barbiturate in the awake, freely moving rat.

In the preceding study, we have found that pentobarbital, a powerful barbiturate substance, strongly modified the ventromedial medulla (VMM) physiology in relation to nociception: indeed, in the same rats, during time-separated similar VMM penetrations, we have recorded, under pentobarbital, 'new' neuronal groups as compared to the awake state, such as the units exclusively driven (excited or inhibited) by cutaneous innocuous or noxious stimulations and the multimodal multireceptive neurons inhibited by non-noxious and noxious stimuli. Still under pentobarbital, we have also recorded the same units found as the rats were awake, i.e., the multimodal multireceptive neurons exclusively excited by various innocuous and noxious stimuli. However, the spontaneous and nociceptive activities of these units were strongly modified as compared to awake animals. Using Brevital (a short-acting barbiturate substance) administration, we have, in the present study, tried to understand the mechanisms underlying these drastic modifications. In particular, one of the questions was whether or not the 'new' neuronal classes recorded under anesthesia resulted from a modification of the physiological properties of the unique VMM neuronal group potentially involved in nociception in awake animals: the multimodal multireceptive units. By following the VMM neuronal activities either before and after or after Brevital administration until recovery from anesthesia, we have determined that the units exclusively driven by innocuous stimulation might result from a modification of the multimodal multireceptive neurons. Alternatively, the multireceptive units inhibited by peripheral stimulations are possibly totally different neurons, silent when the animals are awake.

Back-firing of ventromedial medulla neurons from the spinal cord in awake, freely moving rats.

In the awake, freely moving rat we showed, by means of single-unit recordings and antidromic spinal cord activation, that at the ventromedial medulla level, in these particular experimental conditions, the 'multimodal, multireceptive' units excited by auditory, cutaneous non-noxious and noxious stimuli are possibly involved in a spinal descending control system. These neurons were back-fired from the medial part of the lateral funiculus, hence they probably projected to the dorsal and intermediate horn of the cord, and not to the ventral horn, which emphasizes a role in the control of nociception. Due to their convergent properties, these units are probably involved in nonspecific aspects of nociception such as alertness or stress.