INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Spinal nociceptive transmission is subject to descending modulatory influences from supraspinal sites (see Fields and Basbaum 1999; Mason 2001; Sandkühler 1996 for recent overviews). Inhibitory modulation is most commonly studied and is generally represented as selective for spinal nociceptive transmission (as opposed to non-nociceptive transmission). For example, inhibitory modulation from the periaqueductal gray (PAG) or rostral medial medulla (RMM) has been reported to be relatively selective for nociceptive responses of spinal dorsal horn neurons, including spinothalamic tract cells (e.g., Bennett and Mayer 1979; Duggan and Griersmith 1979; Fields et al. 1977; Oliveras et al. 1974a), although some have reported inhibitory effects on non-nociceptive spinal input (e.g., Carstens et al. 1981; Du et al. 1984; Gray and Dostrovsky 1983; McCreery et al. 1979).

It is also appreciated that descending modulatory influences can enhance or facilitate spinal nociceptive transmission (e.g., Haber et al. 1980; McCreery et al. 1979; Zhuo and Gebhart 1992, 1997). The RMM has been well documented to contribute both inhibitory and facilitatory influences on spinal nociceptive transmission. Importantly, is has been shown that chemical activation of cell bodies in the RMM, and not only fibers of passage that also would be affected by electrical stimulation, produce inhibitory and/or facilitatory effects on spinal nociceptive transmission (e.g., Drower and Hammond 1988; Helmchen et al. 1995; Thomas et al. 1995; Urban and Gebhart 1997; Urban and Smith 1994; Zhuo and Gebhart 1990, 1992, 1997). Most studies have employed thermal stimulation as the peripheral noxious stimulus. Fewer studies (e.g., Gerhart et al. 1981; Gray and Dostrovsky 1983; Haber et al. 1980; Kajander et al. 1984; McCreery et al. 1979) have studied the effects of stimulation in the RMM on spinal transmission of cutaneous mechanical input. McCreery et al. (1979) and Haber et al. (1980) noted excitatory effects of electrical stimulation in RMM on spinal mechanical transmission but did not investigate biphasic or bidirectional influences from RMM and were uncertain whether effects were produced by excitation of cells in RMM. The present study was thus undertaken to evaluate parametrically the effects of electrical and chemical stimulation in the RMM on noxious and non-noxious spinal mechanical transmission.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals

Adult, male Sprague-Dawley albino rats (Biolab., St. Paul, MN) weighing 270-450 g were used in experiments. Rats were anesthetized with an intraperitoneal injection of pentobarbital sodium (45 mg/kg; Nembutal, Abbott Laboratories, Abbott Park, IL), and catheters were inserted into the trachea for mechanical ventilation, into a femoral vein for intravenous administration of drugs, and into a femoral artery for the measurement of arterial blood pressure and heart rate. All wound margins were covered with a local anesthetic ointment. The Animal Care and Use Committee, The University of Iowa approved the research protocol.

The lumbar spinal cord was exposed by laminectomy between T₁₃ and L₃. Rats were suspended by vertebral clamps rostral and caudal to the laminectomy, and a pool for agar (1.75% in saline) was made to minimize respiratory movements of the spinal cord. The head of the rat was fixed in a stereotaxic apparatus, and a hind foot was placed in a paraffin wax model with the plantar surface upward.

During the recording session, rats were paralyzed with pancuronium bromide (0.4 mg initially and 0.2 mg/h intravenous thereafter) and mechanically ventilated. Anesthesia was maintained by inhalation of a gaseous mixture of nitrous oxide and oxygen (2:1) and a continuous intravenous infusion of pentobarbital sodium (5-10 mg · kg¹ · h¹). Arterial blood pressure and heart rate were recorded continuously throughout the experiments, and body temperature was maintained at 37 ± 0.5°C with a water-circulating heating pad and overhead lamps. Tungsten microelectrodes (Micro Probe, Clarksburg, MD; 0.8-0.95 M) were used for extracellular recording of single neurons in the L₄-L₅ spinal segments.

Peripheral stimulation

Mechanical stimulation (touch, pressure, and pinch) applied to the glabrous skin of the plantar surface of the ipsilateral hind foot was used to search for spinal units. Isolated units, continuously monitored by analog delay, were subsequently tested for responses to non-noxious brush, noxious pressure stimuli, and noxious heat.

BRUSH. A reproducible non-noxious mechanical stimulus was produced using an instrument (B413 Tactile Stimulator, Bioengineering, The University of Iowa) that moved a soft camera lens brush (5 × 10 mm) across the skin at rate of 1 cm/2.2 s. Each stimulus consisted of one continuous forward and reverse excursion of the brush across the glabrous skin of the hind foot within the receptive field of the unit. Baseline responses of spinal units to brush of the skin were defined as the mean of three consecutive measurements at 3-min intervals.

PRESSURE. Von-Frey-like stimulation with nylon monofilaments (Semmes-Weinstein Anesthesiometer; Stoelting, Chicago, IL) was applied within the receptive field of the unit. Filaments of different thickness, requiring different pressures to bow the filament (28.8-125.9 g), were applied for 10 s at 3-min intervals. These intensities of stimulation were considered in the noxious range based on stimulus duration and hindpaw withdrawal in rats. Baseline responses of spinal units to pressure was the mean of three consecutive measurements at 3-min intervals.

HEAT. Radiant heat from a 50-W projector lamp (50°C, 15 s) was focused on the glabrous skin within the receptive field. A copper-constantan thermocouple (ANSI type T, 0.13 mm diam, Omega Engineering, Stamford, CT) placed in the center of the field of heat stimulation allowed for feedback control of temperature at the air-skin interface. Heat stimuli were given at 3-min intervals; this results in stable spinal unit responses over the course of an experiment (e.g., Zhuo and Gebhart 1992, 1997). Baseline responses of spinal units to noxious heating was the mean of three consecutive measurements taken at 3-min intervals.

Focal electrical brain stimulation

Focal electrical brain stimulation, 5-100 µA, consisted of continuous 100-Hz constant current cathodal pulses of 100-µs duration. Brain stimulation was started 5 s before and continued during peripheral stimulation of the skin of the hind foot. Monopolar stimulating electrodes (34-gauge; 0.15 mm diam), guided stereotaxically in the vertical plane (incisor bar at +3.3 mm) (Paxinos and Watson 1986), were inserted into the brain through a 26-gauge (0.45 mm OD) guide cannula. The electrodes were cut to extend 2 mm beyond the tip of the guide cannula. Typically, an electrode was lowered to a site in the RMM, and the effect of stimulation on spinal sensory transmission was tested as described in the preceding text before advancing the electrode 0.5 mm and repeating the procedure; typically, two sites in one electrode track were tested in each experiment.

Glutamate microinjection

Three-barrel glass micropipette electrodes were constructed with tip diameter of 15-40 µm. Barrels contained 10 or 100 mM L-glutamate in saline, or saline (control), or a tungsten microelectrode (Micro Probe; 0.8-0.95 M). A pneumatic picopump was used to inject small volumes of L-glutamate. One or more short-duration (10-50 ms) pressure (10-25 psi) pulses to a pipette barrel was employed to administer glutamate or saline into the RMM. Injection volumes were measured directly by monitoring the movement of the fluid meniscus in a pipette barrel with aid of a ×120 compound microscope equipped with a fine reticule. The volume of glutamate was determined by the distance the fluid meniscus was moved by pressure and the inside diameter of the pipette barrel (i.e., volume = distance × inside area). Before glutamate microinjection, electrical stimulation at the same site on responses of spinal units to brush, pressure, or noxious heating of the skin was always tested.

Spinal dorsolateral funiculus (DLF) transection

In some experiments, the cervical spinal cord was also exposed. To transect the DLF, a small pledget of Gelfoam soaked in 1% lidocaine was applied briefly to the cervical spinal cord. The ipsilateral and/or contralateral DLF was then cut using a pair of fine scissors. A reversible fall in arterial blood pressure was produced by DLF transection; all measurements were made only when arterial blood pressure recovered to near the pretransection baseline (30-60 min later).

Ventrolateral funiculus (VLF) lidocaine microinjections

In some experiments, two 26-gauge guide cannulas, 2.0 mm apart, were advanced into the cervical spinal cord (C₁-C₃) in the coronal plane to penetrate the pia matter. Microinjection of lidocaine (4%, 0.5 µl) was made into the ipsilateral and/or contralateral VLF(s) through a 33-gauge (0.20 mm, OD) injection cannula inserted through and extending 2 mm beyond the end of the 26-gauge guide cannula. Injection of lidocaine was done by an electrically driven syringe pump at a speed of 0.5 µl/1.5 min. Progress of the microinjection was continuously monitored by following the movement of an air bubble in a length of calibrated tubing between the syringe and the cannula. This procedure produced a reversible functional blockade in the ventral part of spinal cord (Jones and Gebhart 1987; Sandkühler et al. 1987; Zhuo and Gebhart 1997).

Histology

At the end of experiments, rats were killed with an intravenous overdose of pentobarbital sodium. Anodal electrolytic lesions were made in the brain stem and spinal cord to mark the sites of stimulation and spinal cord recording as well as lidocaine microinjection. The brain, lumbar, and cervical regions of the spinal cord were removed and fixed in 10% Formalin, frozen and cut in 40 µm coronal sections, mounted on glass slides and stained with cresyl-violet. The extent of transection of the DLF(s) was reconstructed.

Data and statistics

Spontaneous activity of spinal dorsal horn neurons was counted during the first 5 s of the period of analysis (before brain stimulation or stimulation of the skin was started). Responses to mechanical pressure, brush, or noxious heating of the skin were counted during stimulus application and are presented as total number of impulses (or as a percentage of that number) minus baseline activity. Data are presented as means ± SE. Statistical comparisons were made using either one- or two-way ANOVAs (Newman-Keuls test for post hoc comparison). Student's t-test was applied for comparisons between paired groups. In all cases, P < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Unit sample

A total of 42 units recorded in the lumbar spinal cord of 29 rats were studied. All units responded to noxious and non-noxious mechanical stimulation. Thirty-two units responded only to mechanical stimuli (non-noxious brush and noxious pressure), and 10 units responded to non-noxious brush, noxious mechanical pressure, and noxious heating (50°C) of the plantar surface of the glabrous skin of the ipsilateral hind foot. Accordingly, all units studied were wide dynamic range type, class 2 spinal units that respond to noxious and non-noxious stimuli. Microelectrode penetrations were made only to 1.2 mm below the dorsum of the spinal cord, and all units studied were histologically confined to dorsal laminae I-VI. The effects of electrical stimulation or glutamate microinjection into the RMM on spontaneous unit activity as well as unit responses to hind foot stimulation of the skin were studied.

Spinal nociceptive mechanical transmission

GENERAL. Both descending facilitatory and inhibitory effects of electrical stimulation on responses of spinal units to noxious pressure (28.8-125.9 g) applied to the skin of the hind foot were observed. At 24 of a total 61 sites in the RMM, electrical stimulation produced intensity-dependent inhibition of spinal unit response to noxious pressure of the skin. At 28 of the remaining 37 sites in the RMM, electrical stimulation produced biphasic effects, facilitating responses to noxious pressure at lesser intensities (5-25 µA) and inhibiting responses of the same spinal units at greater intensities (50-100 µA) of stimulation. At the nine remaining sites in the RMM, electrical stimulation at all intensities tested (5-100 µA) only facilitated responses of spinal units to noxious pressure applied to the skin. The effects of stimulation at 57/61 sites are described below; four sites were located outside the areas of principal interest.

BIPHASIC EFFECTS. Most sites of stimulation were located in n. gigantocellularis (NGC; n = 19), five sites of stimulation were located in n. gigantocellularis pars  (NGC), and three were located in n. raphe magnus (NRM). An example of stimulation-produced biphasic modulation is given in Fig. 1A. In this example, the response to 28.8 g pressure during 25 µA electrical stimulation in the NGC is 162.4% of the control response (440 total imp/20 s) while stimulation at 50 µA attenuated the response to 78.1% of control (239 total imp/20 s). Figure 2A summarizes effects of stimulation at 27 sites in RMM; electrical stimulation at 10 µA produced a significant mean 130.5 ± 4.6% facilitation of responses to noxious pressure, whereas 50 µA stimulation significantly reduced responses of the same units to a mean 70.8 ± 7.9% of control. The mean recruitment index for inhibition between 10 and 50 µA stimulation (% inhibition/20 µA increase in stimulation intensity) was 31.3 ± 3.7. 

View larger version (31K): [in this window] [in a new window]   Fig. 1. Examples (2 different spinal neurons) of facilitation (A) and inhibition (B) of spinal nociceptive mechanical transmission produced by stimulation in the rostral medial medulla (RMM). A and B: peristimulus time histograms (1-s binwidth) and corresponding oscillographic records illustrating a control response to noxious pressure (28.8 g) of the skin of the hind foot and the effect on responses of the same units during stimulation in the RMM (intensities given). The duration of pressure application (10 s) is indicated by the horizontal bar and the period of stimulation (25 s) is indicated by  and . C: graphic representation of the data in A and B; the point above 0 represents the response (total number of impulses in 10 s) in the absence of RMM stimulation. D: stimulation sites, illustrated on a representative coronal brain section (Paxinos and Watson 1986) and spinal recording sites corresponding to the examples in A and B. Pyr, pyramidal tract; NGC, n. reticularis gigantocellularis; NGC, NGC pars ; NPGCI, n. reticularis paragigantocellularis lateralis; NRM, n. raphe magnus; VII, facial nucleus; and Sp5, spinal trigeminal tract.

View larger version (33K): [in this window] [in a new window]   Fig. 2. Summary of descending modulation of spinal nociceptive mechanical transmission from the RMM. A: responses to noxious pressure of the skin are represented as a percentage of the control response (total number of impulses in 10 s) against the intensity of stimulation for biphasic (n = 27), only inhibitory (n = 21), or only facilitatory (n = 9) effects. B: stimulation sites drawn on representative coronal brain sections; abbreviations as in Fig. 1. Some symbols are obscured.

FACILITATORY EFFECTS. At nine sites in the RMM (5 in NGC, 1 in NGC, and 3 in NRM), stimulation at all intensities tested (10-100 µA) only enhanced responses to noxious pressure to a mean 131.9 ± 11.6-186.6 ± 34.8% of control, respectively. Data are summarized in Fig. 2.

INHIBITORY EFFECTS. An example of stimulation-produced inhibition is given in Fig. 1B; data from 21 experiments (11 in NGC, 6 in NGC, 4 in NRM) are summarized in Fig. 2. Electrical stimulation (10-100 µA) only inhibited responses of spinal units to noxious pressure applied to the skin. As in the preceding text, we examined whether there were differences in effects produced at sites in NGC, NGC, and NRM and noted none either in the magnitude of inhibition (to a mean 33.8 ± 6.9, 49.1 ± 10.4, and 52.8 ± 9.2% of control, respectively; F(2,36) = 0.03) or intensity of stimulation that produced significant inhibition (6 3.6 ± 9.1, 43.8 ± 4.2, and 40.0 ± 10.0 µA, respectively). We also compared inhibition produced at these 21 sites of stimulation with inhibitory effects produced at biphasic sites of stimulation. The estimated (extrapolated) mean threshold for stimulation-produced inhibition was 8.5 ± 2.5 µA, which is significantly less than the estimated threshold intensity of stimulation that produced inhibition from biphasic sites in the RMM (33.1 ± 3.1 µA). The mean recruitment index for inhibition (%inhibition/20 µA) from these 21 inhibitory sites was 26.9 ± 4.0, which did not significantly differ from the recruitment of inhibition from the biphasic sites of stimulation (31.3 ± 3.7% inhibition/20 µA). The inhibition produced by electrical stimulation at 50 µA, however, produced a significantly greater inhibition of responses (to a mean 39.2 ± 5.1%) than produced from biphasic sites of stimulation at the same intensity (mean: 70.8 ± 7.9%).

SPONTANEOUS ACTIVITY. At sites in the RMM where electrical stimulation biphasically modulated responses to noxious mechanical transmission, spontaneous activity of units was not significantly affected by electrical stimulation at the intensities tested (10-100 µA). Similarly, at sites in the RMM where electrical stimulation only inhibited or only facilitated responses to noxious pressure, spontaneous activity of units was not significantly affected.

INTENSITY CODING. Responses of spinal units to noxious mechanical pressure applied to the skin were positively accelerating functions in the intensity range tested (see Fig. 3, A and C). Stimulus-response functions (SRF) for seven spinal units and their modulation by stimulation at inhibitory intensities of stimulation are presented in Fig. 3, A and B, respectively. Electrical stimulation at a mean intensity of 47.9 ± 11.9 µA significantly inhibited responses of these seven units to 75.9 g pressure applied to the skin to a mean 63.8 ± 7.4% of control and significantly decreased the slope of the SRF without affecting the extrapolated threshold for response (Fig. 3B).

View larger version (29K): [in this window] [in a new window]   Fig. 3. Effects of stimulation in the RMM on stimulus-response functions (SRF) to graded pressure applied to the skin. A and C: SRFs of individual spinal neurons in the absence of stimulation in the RMM. Data are plotted as the response to pressure applied to the skin (total number of impulses in 10 s) against intensities of pressure. B and D: summary of effects of stimulation on SRFs of the units illustrated in A and C, respectively. Data are presented as mean response in the absence () and presence () of stimulation. In B, an inhibitory intensity of stimulation at sites illustrated in E ( and ) significantly reduced the slope of the SRF of the 7 units studied. In D, a facilitatory intensity of stimulation at sites illustrated in E ( and ) shifted the SRF leftward.

LATENCY TO EFFECT. The apparent latency to stimulation-produced facilitation and inhibition of spinal mechanical nociceptive transmission was determined by employing a cumulative sum technique (Ellaway 1978) and bin-by-bin analysis of unit responses (Gerhart et al. 1983). Electrical stimulation was given during a relatively stable rate of unit response (10-s duration) to noxious pressure of the skin. Unit activity for 500 ms before the onset of electrical stimulation was averaged to generate a reference baseline and the cumulative sum of unit activity 500 ms before and 1,500 ms after the onset of stimulation was plotted. The apparent latency to facilitatory or inhibitory effects was defined as the time from the onset of stimulation to the point when the cumulative sum of the histogram significantly deviated from the baseline (Fig. 4A). In eight experiments, the mean latency to inhibition of unit response to noxious pressure applied to the skin was determined to be 112.2 ± 39.7 ms (range, 19.7-339.5 ms; mean intensity of stimulation, 56. 3 ± 3.8 µA). In another five experiments, the apparent latency for facilitation of unit response to noxious pressure applied to the skin was determined to be 290.5 ± 76.9 ms (range, 125.0-511.0 ms; mean intensity of stimulation, 19.0 ± 7.8 µA). The latency to facilitation was significantly longer than for inhibition.

View larger version (45K): [in this window] [in a new window]   Fig. 4. Apparent latency of stimulation-produced effects on spinal mechanical transmission. A and C: the derived cumulative sums of individual experiments on noxious pressure and non-noxious brush, respectively.  and · · · , the onset of stimulation (stim; 100 µs; 100 Hz), which was continued throughout the remaining 1.5 s illustrated. In A, the apparent mean latency for stimulation-produced inhibition of noxious mechanical transmission () was determined to be 112.2 ± 39.7 ms; for stimulation-produced facilitation (), the apparent mean latency was 290.5 ± 76.9 ms. In C, the apparent latency for inhibition of non-noxious brush transmission was determined to be 114.3 ± 26.2 ms; for stimulation-produced facilitation, the mean latency was 169.7 ± 38.4 ms. B and D: stimulation sites drawn on representative coronal brain sections (,  inhibitory;  , facilitatory; and  indicate sites where both facilitatory and inhibitory effects were produced at low and high intensities of stimulation, respectively); see Fig. 1 for abbreviations.

SITE SPECIFICITY. To determine whether descending facilitatory and inhibitory influences from different sites in RMM similarly affect the same spinal unit, the effects of electrical stimulation at the same intensity (10 µA), but different sites in the RMM, were studied on the same spinal units. An electrode was lowered into RMM and the effect of stimulation at 10 µA was characterized. The electrode was subsequently lowered 0.5 or 1.0 mm and the same intensity of stimulation tested on response of the same spinal unit to noxious pressure. Summary data are shown in Fig. 5, A and B. The scatter-diagram clearly shows that for each unit studied, the same intensity of stimulation at two different sites (connected by a line in 5B; mean distance between stimulation sites = 0.9 ± 0.2 mm) produced significantly different effects. At 10 stimulation sites (Fig. 5B, ), electrical stimulation at 10 µA produced an inhibitory effect or no effect on responses of spinal units to noxious pressure applied to the skin; the mean effect was inhibition to 71.8 ± 9.2% of control (Fig. 5A, inset). At the second 10 sites of stimulation in the same experiments, electrical stimulation at the same intensity facilitated responses of the same spinal units to the same intensity of noxious pressure to a mean 126.8 ± 4.8% of control. The effects produced significantly differ.

View larger version (42K): [in this window] [in a new window]   Fig. 5. Effects of electrical stimulation at different sites in the RMM on spinal unit responses to non-noxious brush and noxious pressure of the skin of the hind foot. A: scatter-diagram illustrating responses of the same unit to noxious pressure applied to the skin (total numbers of impulses in 10 s) during stimulation at inhibitory sites () and at a facilitatory sites (). Inset: summary of responses of the same 10 units during inhibitory (inhib, to 71.8 ± 9.2% control) and facilitatory (facil, to 126.8 ± 4.8% control) stimulation. C: scatter-diagram illustrating responses of the same unit to non-noxious brush of the skin (total number of impulses) during stimulation at inhibitory sites () and at facilitatory sites (). Inset: summary of responses of the same 6 units during inhibitory (inhib, to 77.6 ± 7.5% control) and facilitatory (facil, to 116.0 ± 3.7% control) stimulation. - - -, where data points would align if the effects of stimulation were the same at each pair of sites. B and D: sites of brain stem stimulation (,  inhibitory; ,  facilitatory) illustrated on representative coronal brain sections; each pair of stimulation sites are connected by a line.

GLUTAMATE MICROINJECTION. At sites where glutamate was microinjected, the effect of electrical stimulation on response of spinal units to noxious pressure applied to the skin was characterized first. At five sites in the RMM (4 in NGC, 1 in NGC), electrical stimulation at 17.5 ± 4.3 µA increased responses of spinal units to noxious pressure to a mean 123.3 ± 6.3% of control and inhibited responses of the same units to 40.4 ± 17.0% of control at greater intensities of stimulation (75.0 ± 14.4 µA). Microinjection of glutamate (10 mM, 25 nl) produced a rapid onset (mean: 2.6 ± 1.0 min), short-lasting (5-10 min) facilitation of responses of spinal units to noxious pressure of the skin (mean 127.0 ± 6.3% of control; Fig. 6A). This glutamate-produced facilitation was reproducible. Responses to noxious pressure of the skin were increased to a mean 125.8 ± 5.8% of control by a second microinjection of glutamate into the same sites 15 min after the first microinjection of glutamate, a time by which responses of units returned to baseline. There was no significant difference between the magnitude of facilitation produced by first and second microinjection of glutamate, indicating an effect by reversible activation of cell bodies. Microinjection of glutamate at a greater dose (100 mM, 25 nl) into the same five sites after responses to noxious pressure returned to baseline produced significant inhibitory effects on responses of the same units to noxious pressure of the skin (to a mean 63.6 ± 8.8% of control, n = 5; Fig. 6A).

View larger version (55K): [in this window] [in a new window]   Fig. 6. Summary of effects of electrical stimulation and glutamate microinjection into the RMM on noxious pressure (A) and non-noxious brush (B) mechanical transmission. A: electrical stimulation at low intensities (mean: 17.5 ± 4.3 uA) facilitated responses of spinal units to noxious pressure applied to the skin of the hind foot and inhibited response at greater intensities (mean: 75.0 ± 14.4 uA) of stimulation at the same sites (n = 5). Glutamate microinjection into the same 5 sites at a low dosage (0.25 nmol) produced reproducible facilitation at the 1st and a 2nd injection 15 min later. Glutamate microinjection at a greater dosage (2. 5 nmol) inhibited responses of the same spinal units. B: electrical stimulation at low intensity (mean: 10.0 ± 0.0 uA) facilitated responses of spinal units to brush of the skin and inhibited responses at greater intensities (66.3 ± 11.8 µA). Glutamate microinjection at a low dosage (0.25 nmol) produced similar facilitation of responses at the 1st and a 2nd injection 15 min later (n = 6). Glutamate microinjection at a greater dosage (2.5 nmol) inhibited responses of the same spinal units. C: sites for stimulation and glutamate microinjection are indicated (, pressure, , brush) on representative coronal sections. See Fig. 1 for abbreviations.

SPINAL DLF TRANSECTION. To investigate the spinal pathway(s) mediating descending facilitation and/or inhibition from the RMM, ipsilateral first (n = 6), contralateral first (n = 3), and ultimately bilateral (n = 9) transections of the DLF were performed at the cervical level of the spinal cord. The spontaneous activity of units (n = 9) was not significantly affected by either a unilateral (ipsi- or contralateral) transection of the DLF or ultimately bilateral transection of the DLFs. Similarly, ipsilateral DLF transection (n = 6) did not significantly affect baseline responses of spinal units to noxious pressure of the skin; responses to noxious pressure were slightly increased from baseline (568 ± 92 total imp/20 s) to 716 ± 169 total imp/20 s (P > 0.05). Similar results were found after contralateral transection of the DLF (from 989 ± 149 total imp/20 s to 1,193 ± 117 total imp/20 s; P > 0.05) and ultimately bilateral transection of the DLFs (from 715 ± 117 to 869 ± 121 imp/20 s; P > 0.05).

View larger version (43K): [in this window] [in a new window]   Fig. 7. Summary of effects of unilateral and bilateral transections of the dorsolateral funiculus (DLF) on responses to noxious mechanical stimulation of the skin. A and B: the effects of ipsilateral first (ipsi: n = 6), contralateral first (contra; n = 3) and ultimately bilateral (n = 9) transections of the DLFs. A and B: effects of DLF transections on stimulation-produced inhibition (A) and facilitation (B). Responses to noxious mechanical pressure applied to the skin are presented as percentage of control (total number of impulses in 10 s). C: histological reconstruction of DLF transections. D: stimulation sites ( and , inhibition;  and , facilitation) are indicated on representative coronal brain section; see Fig. 1 for abbreviations.

VENTRAL SPINAL CORD LIDOCAINE MICROINJECTIONS. In nine experiments, ipsilateral first (n = 6), contralateral first (n = 3) and ultimately bilateral, microinjections of lidocaine were made into the VLF in the cervical spinal cord. The spontaneous activity of spinal units was not significantly affected by either unilateral or bilateral microinjections of lidocaine. Baseline responses of spinal units to noxious pressure were increased (n = 2), decreased (n = 2), or not changed (n = 2) by ipsilateral lidocaine microinjection. Subsequent bilateral blockage of the VLF by lidocaine also did not significantly affect responses of spinal units to noxious pressure (283 ± 65 total imp/20 s vs. 453 ± 167 total imp/20 s; P > 0.0 5).

View larger version (35K): [in this window] [in a new window]   Fig. 8. Summary of effects of lidocaine blockage of the ventral spinal cord on stimulation-produced facilitation of responses to noxious mechanical stimulation of the skin. A: the effects of ipsilateral (ipsi; n = 6) and contralateral (contra; n = 3) injections of lidocaine (4%, 0.5 µl) on stimulation-produced facilitation. B and C: sites of lidocaine injection and RMM stimulation, respectively. See Fig. 1 for abbreviations.

Spinal non-nociceptive mechanical transmission

GENERAL. Electrical stimulation in RMM produced both facilitatory and inhibitory effects on responses of spinal units to non-noxious brush of the skin of the hind foot. At 14 of 31 sites in the brain stem, electrical stimulation produced intensity-dependent inhibition of responses. At 9 of the remaining 17 sites in the brain stem, electrical stimulation produced biphasic effects, facilitating responses to brush at lesser intensities (<50 µA) and inhibiting responses of the same units at greater intensities (50-200 µA) of stimulation. At the remaining eight sites, electrical stimulation at all intensities tested (10-100 µA) only facilitated responses of spinal units to brush of the skin.

BIPHASIC EFFECTS. As summarized in Fig. 9A, 10 µA stimulation in NGC (n = 7) or NCG (n = 2) produced a significant mean 136.9 ± 11.5% facilitation of responses to brush of the skin; 100 µA stimulation significantly reduced responses of the same units to a 62.6 ± 11.2% of control. The mean recruitment index for inhibition at these nine biphasic sites of stimulation (% inhibition/20 µA increase in stimulation intensity) was 29.7 ± 7.5. 

View larger version (30K): [in this window] [in a new window]   Fig. 9. Summary of descending modulation of spinal non-noxious mechanical transmission from the RMM. A: responses to non-noxious brush of the skin are represented as a percentage of the control response (total number of impulses) against the intensity of stimulation for biphasic (n = 9), only inhibitory (n = 14) or only facilitatory (n = 8) effects. B: stimulation sites drawn on representative coronal brain sections; see Fig. 1 for abbreviations.

FACILITATORY EFFECTS. An example of stimulation-produced facilitation of responses of a spinal unit to brush of the skin is given in Fig. 10. In this example, the response to brush of the skin during 10 or 50 µA stimulation was facilitated to 120 and 130% of control (125 total impulses), respectively. Stimulation at eight sites in the RMM (3 in NGC, 2 in NGC, and 3 in NRM) produced only facilitation of responses to brush of the skin at all intensities of stimulation tested (10-100 µA). The data are summarized in Fig. 9A.

View larger version (31K): [in this window] [in a new window]   Fig. 10. Example of facilitation of responses to non-noxious brush stimulation of the skin produced by stimulation in RMM. A: peristimulus time histograms (1-s binwidth) and corresponding oscillographic records illustrating a control response to brush of the skin of the hind foot and the effect on responses of the same unit during stimulation in the RMM (intensities given). The brush stimulus is indicated by the horizontal arrows and the period of RMM stimulation (25 s) is indicated by upward and downward arrows. B: graphic representation of the data in A; the point above 0 represents the response (total number of impulses) in the absence of RMM stimulation. C: Stimulation site, illustrated on a representative coronal brain section (abbreviations as in Fig. 1), and receptive field with orientation of brush stimulus indicated.

INHIBITORY EFFECTS. At 14 sites in RMM (9 in NGC, 3 in NGC, and 2 in NRM), responses of spinal units to brush of the skin were only inhibited by electrical stimulation at all intensities tested (10-100 µA). Data are summarized in Fig. 9A. The estimated mean threshold of stimulation for inhibition was 4.5 ± 2.5 µA, which was significantly less than the estimated threshold for inhibition of response from biphasic sites of stimulation (31.9 ± 5.9 µA). The inhibition of responses to brush produced by 50 µA electrical stimulation at inhibitory sites was significantly greater (to 67 ± 4.6% of control) than the inhibition produced from biphasic sites (mean: 96.3 ± 14.7%) at the same intensity of stimulation. Meaningful comparisons between effects produced by stimulation in NGC, NGC, and NRM are compromised by the limited number of sites (5) studied in NGC and NRM. Stimulation in NGC (n = 9) at a mean intensity of 66.7 ± 8.3 µA inhibited responses to brush to 58.2 ± 8.7% of control.

SPONTANEOUS ACTIVITY. There were no effects on spontaneous activity produced by stimulation at biphasic, inhibitory, or facilitatory sites in the RMM.

LATENCY TO EFFECT. The apparent latencies to stimulation-produced facilitation and inhibition were determined as described in the preceding text. RMM stimulation was given during a relatively stable rate of unit response to non-noxious brush of the skin. The first 500-ms period of recording was used to generate a reference base, and the cumulative sum of unit activity 500 ms before and 1,500 ms during stimulation was plotted. The apparent mean latency for descending facilitation by electrical stimulation was determined to be 169.7 ± 38.4 ms (range, 61.0-310.0 ms; n = 7); the mean latency for descending inhibition was determined to be 114.3 ± 26.2 ms (range, 22.0-306.7 ms; n = 11). There was no significant difference between the latencies to facilitation and inhibition of unit responses to non-noxious brush of the skin (Fig. 4, C and D).

SITE SPECIFICITY. To examine whether a spinal unit received both descending facilitatory and inhibitory influences from different sites in RMM, the effects of electrical stimulation at the same intensity, but different sites in the RMM, were studied on the same spinal units. The scatter diagram in Fig. 5C shows that for each unit studied, the same intensity of stimulation at different sites (connected by a line) in the RMM (mean distance between sites; 0.6 ± 0.1 mm) produced significantly different effects. At six stimulation sites (Fig. 5D, ), electrical stimulation at a mean 16.7 ± 6.7 µA produced inhibition or no effect on responses of spinal units to brush (to a mean 77.6 ± 7.5% of control). However, at the second, more ventral of the six sites, electrical stimulation at the same intensities facilitated responses of the same spinal units to brush to a mean 116.0 ± 3.7% of control (P < 0.001).

GLUTAMATE MICROINJECTION. The example in Fig. 11 shows that L -glutamate microinjection into a site in the RMM facilitated the response of a spinal unit to brush to 194% of control 1 min after glutamate administration. Data from five experiments are summarized in Fig. 6B. Electrical stimulation (10 µA) at five sites facilitated responses to brush to a mean 114.4 ± 4.8% of control and inhibited responses of the same units to 72.4 ± 6.8% of control at greater intensities of stimulation (66.3 ± 11.8 µA). Microinjection of glutamate (10 mM, 25 nl) into these biphasic sites significantly increased responses to brush of the skin to a mean 122.9 ± 7.6% of control. Glutamate-produced facilitatory effects were rapid in onset (mean 3.8 ± 1.0 min), short-lasting (10 min) and reproducible. Responses of spinal units to brush of the skin returned to 95.4 ± 3.4% of control when tested 10 min after administration of glutamate. The facilitatory effect of glutamate was tested again by a second glutamate microinjection into the same five sites. The facilitation produced by the second administration of glutamate was to 130. 3 ± 10.7% of control, which was not significantly different from the facilitation produced by the first microinjection of glutamate. A third, greater dose of glutamate (100 mM; 25 nl) microinjected after responses to brush returned to preinjection baseline produced a significant attenuation of responses of the same units to brush of the skin (to a mean 71.4 ± 11.4% of control, n = 5; Fig. 5B).

View larger version (25K): [in this window] [in a new window]   Fig. 11. Example of the facilitatory effect of glutamate on spinal non-noxious mechanical transmission. A: peristimulus time histograms (1-s binwidth) illustrating a control response to brush of the skin of the hind foot and 1, 3, and 9 min after intra-RMM glutamate microinjection. B: graphic representation of data in A. The point above "c" represents the baseline response to brush of the skin before intra-RMM administration of 0.25 nmol of glutamate. C: glutamate injection site, illustrated on a representative coronal brain section (abbreviations as in Fig. 1).

RMM MODULATION OF NOCICEPTIVE AND NON-NOCICEPTIVE TRANSMISSION. We also compared descending influences on responses of the same nine spinal units to noxious pressure and non-noxious brush. Electrical stimulation in the NGC (n = 7), NGC (n = 1), or NRM (n = 1) produced biphasic (n = 7) or only inhibitory (n = 2) effects on responses to noxious pressure applied to the skin. Responses of the same spinal units to non-noxious brush of the skin were biphasically modulated (n = 4), only facilitated (n = 2), only inhibited (n = 2), or not affected (n = 1) by the same intensities of stimulation (10-100 µA) in the same sites. At three of these nine sites, electrical stimulation biphasically modulated responses of the same three spinal units to both noxious pressure and non-noxious brush of the skin. However, at the other six sites, stimulation modulated responses of the same spinal units to noxious pressure or non-noxious brush differently.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present study demonstrates that electrical stimulation and glutamate microinjection in the RMM produce intensity-dependent biphasic (facilitatory and inhibitory) modulation of spinal mechanical transmission, both noxious and non-noxious. Stimulation-produced inhibition but not stimulation-produced facilitation was also intensity dependent for both noxious and non-noxious mechanical transmission. Both stimulation- and glutamate-produced effects were rapid in onset, short-lasting and reproducible. Descending modulatory effects were specific for the site of stimulation in the brain stem, not for the unit recorded in the spinal cord, because modulation of the same spinal units was shown to be different from different sites in the brain stem. The estimated latencies to effect significantly differed for facilitatory and inhibitory modulation of noxious mechanical spinal transmission and are conveyed in different spinal pathways. Descending inhibitory effects are contained in the DLFs while descending facilitatory effects are contained in the ventral part of the spinal cord.

Descending inhibition of spinal nociceptive mechanical transmission

Spinal nociceptive transmission is known to be subject to descending inhibitory influences from supraspinal structures, including the PAG, NRM, and NGC. Electrical stimulation in these supraspinal structures attenuates or completely inhibits animal behavioral responses to noxious mechanical stimulation of the skin (e.g., pinch, pressure, or squeeze) (Mayer et al. 1971; Oliveras et al. 1974a,b). Complementary electrophysiological experiments documented that electrical stimulation in the brain stem attenuated responses of spinal dorsal horn neurons [including spinothalamic tract (STT) neurons] to noxious mechanical stimulation of the skin (e.g., Fields et al. 1977; Gray and Dostrovsky 1983; Haber et al. 1980; McCreery et al. 1979; Oliveras et al. 1974a; Willis et al. 1977; Yezierski 1990) or stimulation of afferent A and C fibers (Gerhart et al. 1981, 1983; Haber et al. 1980; Willis et al. 1977). In these earlier investigations, only electrical stimulation, which nonselectively activates cell bodies and fibers of passage, was used. We found in the present study that glutamate microinjection replicated the effects of electrical stimulation, producing rapid onset, short-lasting inhibition of responses of spinal units to noxious pressure of the skin, revealing that activation of cell bodies in the RMM is sufficient to inhibit spinal mechanical nociceptive transmission. Stimulation-produced inhibition in the present study was selective for stimulus-evoked responses because spontaneous activity of units was not significantly affected by stimulation at the same intensities.

Descending facilitation of spinal nociceptive transmission

Descending facilitatory influences from supraspinal structures on spinal nociceptive mechanical transmission also have been noted. McCreery et al. (1979) reported that single pulse electrical stimulation in the NGC or NRM of the cat increased the excitability of STT neurons activated by sustained mechanical pressure applied to the skin (followed by prolonged suppression). Haber et al. (1980) reported that electrical stimulation in or near the NGC in the monkey produced excitatory effects on 4 of 41 (9.7%) wide dynamic range STT neurons. When all spinal neuron types are incorporated in the analysis, including five low-threshold STT cells, 9 of 57 (19%) were excited by NGC stimulation. The excitatory effect included increases in spontaneous activity and responses of spinal units to noxious mechanical stimulation of the skin. In another report, electrical stimulation in PAG or NRM (1 site each) facilitated responses of only 2 of 138 spinal dorsal horn neurons to noxious pinch of the skin in cats, but facilitation may have been obscured by the brain stem stimulation artifact (Gray and Dostrovsky 1983). In a study of spinomesencephalic tract cells in the cat, Yezierski (1990) reported that electrical stimulation at 25 of 32 sites (78%) in the brain stem, including the NGC, NRM, and n. reticularis magnocellularis, produced excitation followed by inhibition (n = 16) or only excitation (n = 9) of responses, including to noxious mechanical stimulation of the skin.

In the present study, electrical stimulation at 36 of 57 sites in the RMM (63%) produced biphasic (facilitatory and inhibitory, n = 27) or only facilitatory (n = 9) modulation of spinal nociceptive mechanical transmission. The low percentage of stimulus sites that produced facilitatory effects in earlier work can be explained by the predominant focus in those studies on inhibition and stimulating electrically at threshold or suprathreshold intensities to produce inhibition. Earlier studies either did not parametrically vary stimulation intensity or, if they did, did not test low intensities of stimulation. We have repeatedly noted that facilitatory effects on spinal nociceptive transmission are produced at lesser intensities of electrical stimulation (Zhuo and Gebhart 1992, 1997; Zhuo et al. 2002).

Similarly, low concentrations of glutamate have been demonstrated by us to reliably and reproducibly facilitate spinal nociceptive transmission (Zhuo and Gebhart 1992, 1997; Zhuo et al. 2002). In the present report, glutamate was shown to reproducibly facilitate spinal mechanical nociceptive transmission. The effect of glutamate, like that of stimulation, appeared to be selective for stimulus-evoked responses because spontaneous activity of units was not affected by intra-RMM glutamate injection.

Spinal pathways

Descending influences traveling in the DLFs are generally considered responsible for inhibitory modulation of spinal nociceptive transmission from supraspinal structures, including the PAG and RMM (see Basbaum and Fields 1984; Gebhart and Randich 1990 for reviews). In the present study, descending inhibition of spinal nociceptive mechanical transmission was blocked by transection of the DLFs, consistent with previous results (Zhuo and Gebhart 1992, 1997). Further, removal of this descending inhibitory pathway uncovered descending facilitatory effects. Electrical stimulation at intensities that inhibited responses to noxious mechanical stimulation before bilateral DLF transections produced a modest enhancement of responses to the same stimulus after DLF transections. These results are in good agreement with our previous studies of modulation of noxious thermal stimulation (Zhuo and Gebhart 1990, 1992, 1997) and with the work of others (e.g., Jones and Gebhart 1987; McCreery et al. 1979; Mokha et al. 1986; Sandkühler et al. 1987; Willis et al. 1977). One interpretation of these outcomes is that descending inhibitory and facilitatory influences are simultaneously active/engaged in the RMM. We have previously suggested that this outcome likely reflects the presence of prepotent, tonic descending inhibition, which when removed by transection of the DLFs, permits expression of normally present, but overridden facilitatory influences.

Spinal pathways for descending facilitatory or excitatory modulation have been less well studied. We found here that stimulation-produced facilitation but not inhibition of spinal mechanical nociceptive transmission was blocked by reversible lidocaine-produced local anesthesia of the ventral part of spinal cord, suggesting that descending facilitation is primarily conveyed in the ventral spinal cord. This is consistent with previous results (Zhuo and Gebhart 1990, 1992, 1997) and related work in which we found that descending facilitatory influences on spinal neurons, whether produced directly in RMM (Urban and Gebhart 1997) or indirectly by activation of vagal afferent fibers (see Randich and Gebhart 1992), were confined to the ventral part of the spinal cord in the rat.

Modulation of spinal nociceptive and non-nociceptive transmission

INHIBITORY MODULATION. Previous studies have addressed the selectivity of descending inhibitory influences on spinal nociceptive and non-nociceptive transmission. In studies of spinal wide dynamic range or class 2 dorsal horn neurons (including ascending tract neurons), it has been reported that responses of spinal units to noxious stimulation (e.g., pinch, squeeze, heating of the skin or stimulation of afferent A and C fibers) are more effectively inhibited by electrical stimulation in the RMM or PAG in terms of percentage of units inhibited and/or the magnitude of inhibition produced (Beall et al. 1976; Carstens et al. 1981; Gebhart et al. 1981, 1983; Haber et al. 1980; Lovick and Wolstencroft 1979; Willis et al. 1977; Zhang et al. 1991). Preferential effects of supraspinal modulation for spinal nociceptive transmission are not always noted (e.g., Duggan and Griersmith 1979; Oliveras et al. 1974a), and animals are reported sometimes to be hyperreactive during stimulation-produced analgesia to non-noxious stimuli (e.g., Mayer et al. 1971; Oliveras et al. 1974a).

FACILITATORY MODULATION. Facilitation of responses of spinal dorsal horn neurons (including STT neurons) to non-noxious stimuli by electrical stimulation in supraspinal structures has been previously reported. Haber et al. (1980) reported that electrical stimulation in the NGC produced excitatory or mixed (excitatory and inhibitory) effects on three of five STT class 1 neurons. Stimulation in the NRM also facilitates responses of class 1 neurons to non-noxious stimuli (Fields et al. 1977). Recording intracellularly, Light et al. (1986) demonstrated that stimulation in the NRM (n = 22) or PAG (n = 4) produced an excitatory postsynaptic potential (EPSP) following an inhibitory postsynaptic potential (IPSP) in 31 of 46 (67%) dorsal horn neurons (class 1) recorded in spinal laminae I and II. Dubuission and Wall (1979) reported that medullary raphe stimulation produced descending excitatory effects on responses of spinal dorsal horn laminae 1 and 2 neurons in the cat, including those only responding to non-noxious stimuli (brush and touch). It was demonstrated in the present study that spinal class 2 neurons are subject to descending facilitatory effects from the RMM. In 17 of 31 sites in RMM (55%), electrical stimulation produced biphasic modulation (n = 9) or only facilitatory (n = 8) effects on responses of spinal units to non-noxious brush of the skin. Such facilitatory effects were reproduced by glutamate microinjection in the RMM, suggesting that facilitatory influences on spinal nociceptive and non-nociceptive transmission can arise from cells located in RMM.

Significance

Accumulating evidence suggests an important role for the RMM in the development and maintenance of exaggerated responses to peripheral stimuli after tissue injury (i.e., hyperalgesia and allodynia). Supraspinal contributions to hyperalgesia have been established in inflammatory, neurogenic, neuropathic, and illness-induced models of hyperalgesia (see Urban and Gebhart 1999 for review). In experiments where hyperalgesia has been established, spinal cord transection, intra-RMM lidocaine or ibotenic acid, and electrolytic lesions of the RMM all have been reported to reverse or block the hyperalgesia. In review of these studies, we (Urban and Gebhart 1999) concluded that the RMM plays a prominent role in mediating the development and maintenance of secondary hyperalgesia and hypothesized that facilitatory influences from the RMM were central to these findings. More recently, Porreca and colleagues have investigated the role of the RMM in the allodynia that characterizes a model of neuropathic pain produced by ligation of the L₅ and L₆ spinal nerves. Sun et al. (2001) established that the tactile allodynia, which develops after nerve ligation, is dependent on inputs to supraspinal sites. Porreca et al. (2001) subsequently documented that selective ablation of RMM cells that express the µ-opioid receptor can prevent the development or reverse established allodynia in this model of neuropathic pain. These data contribute to the growing appreciation that descending facilitatory influences likely underlie some chronic pain states. The present report documents that both nociceptive and non-nociceptive spinal mechanical transmission is subject to tonic facilitatory modulation from cells located in RMM, reinforcing their potential role in the exaggerated responses to peripheral stimuli that characterize some chronic pain states.