INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The subnucleus caudalis (Vc) of the trigeminal (V) spinal tract nucleus is generally considered to play an integral role in the brain stem processing of orofacial nociceptive information and to represent the V brain stem homologue of the spinal dorsal horn and has also been termed the medullary dorsal horn. However, recent studies have also documented the involvement in orofacial pain mechanisms of the rostral components of the V spinal tract nucleus, including subnucleus oralis (Vo). Lesioning of Vc or V tractotomy at the obex level does not completely abolish nociceptive responses evoked by noxious intraoral or perioral stimuli, including tooth pulp stimulation, and disruption of rostral components including Vo has been reported to interfere with nociceptive orofacial sensation or behavior (for review, see Bereiter et al. 2000; Sessle 2000). Electrophysiological studies also have demonstrated that nociceptive neurons responding to high-intensity electrical stimuli or formalin applied to the intraoral or perioral region occur in Vo, and some of these neurons are sensitive to diffuse noxious inhibitory controls (Azerad et al. 1982; Dallel et al. 1990; Hu and Sessle 1984; Hu et al. 1992; Parada et al. 1997; Raboisson et al. 1995). We have also recently shown that application of mustard oil (MO), a small-fiber excitant and inflammatory irritant, to the tooth pulp can produce "central sensitization" reflected in neuroplastic changes in nociceptive neurons of Vo as well as in Vc (the medullary dorsal horn) (Chiang et al. 1998; Park et al. 2001).

Earlier studies have documented that Vc neurons project to Vo and other rostral components of the V brain stem sensory nuclear complex via ascending intersubnuclear pathways (Gobel and Purvis 1972; Hu and Sessle 1979; Ikeda et al. 1982, 1984; Jacquin et al. 1990; Nasution and Shigenaga 1987; Panneton and Burton 1982) and that Vc exerts a net facilitatory influence on Vo neurons (Dallel et al. 1998; Greenwood and Sessle 1976; Khayyat et al. 1975; Woda et al. 2001; Young and King 1972). These considerations raise the possibility that the central sensitization recently documented in Vo nociceptive neurons (Park et al. 2001) is mediated by Vc. To address this issue, the present study used microinjection into Vc of CoCl₂, an effective synaptic transmission blocker in the CNS (Allen and Pronych 1997; Hochstenbach and Ciriello 1997; Lee and Malpeli 1985; Malpeli 1983; Mooney et al. 1992; Nuseir et al. 1999), to determine whether the Vo neuroplastic changes induced by MO application to the tooth pulp depend on an ascending influence from Vc. The data have been briefly presented in abstract form (Chiang et al. 2000).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal preparation

Thirty-two Sprague-Dawley male rats (285-410 g) were used. The methods used for animal preparation, stimulation, and neuronal recording and classification were similar to those described previously in detail (Chiang et al. 1998; Park et al. 2001) and so will only be briefly outlined. The animal was anesthetized by a mixture of alpha-chloralose (50 mg/kg ip) and urethan (1 g/kg ip). The trachea and the left external jugular vein were cannulated. To expose the pulp of the right maxillary first molar, a cavity was prepared with a low-speed dental drill and temporarily filled with a cotton pellet soaked with saline. The animal was then placed in a stereotaxic apparatus, a craniotomy was performed on the right side to expose the cerebral cortex for insertion of the recording electrode into Vo, and the caudal medulla oblongata was also exposed. After 1 h, a supplemental dose of urethan (200-300 mg/kg iv) was administered, the animal was immobilized with gallamine triethiodide (initial dose, 35 mg/kg; maintenance dose, 14 mg/h; iv) and ventilated, and then microelectrode recordings were begun (see following text). An adequate level of anesthesia was confirmed periodically by the lack of spontaneous movements and responses to pinching the paw when the gallamine-induced muscle paralysis was allowed to wear off. In addition, pupil size and heart rate were routinely monitored to ensure their stability when noxious pinch stimuli were applied. The expired percentage CO₂ and rectal temperature were also continuously monitored and maintained at physiological levels of 3.5-4.5% and 37-38°C, respectively. All surgeries and procedures were approved by the University of Toronto Animal Care Committee in accordance with the regulations of the Ontario Animal Research Act (Canada).

Recording and stimulation procedures

Single neuronal activity was recorded extracellularly in the right Vo (2.5-2.7 mm lateral, P 1.3-2.6 mm) (Paxinos and Watson 1986) by an epoxy-resin-coated tungsten microelectrode with a rostral inclination of 26°. As the microelectrode was advanced through the Vo region, natural stimuli (see following text) were applied to the orofacial tissues to search for responsive neurons. The Vo was explored 2.5-2.7 mm lateral to the midline and between frontal planes P1.3 and P2.6 mm referred to the interaural line. Neuronal activity was amplified, displayed on oscilloscopes and also led to a window discriminator connected to an A/D converter (1401plus, Cambridge Electronic Designs (CED) and a personal computer. Data were collected with Spike2 for Windows (CED) and analyzed off-line.

A wide range of mechanical (brush, pressure, and pinch), electrical and noxious thermal (radiant heat, 51-53°C) stimuli were applied to the orofacial skin or intraoral mucosa to classify each neuron as a low-threshold mechanoreceptive neuron (LTM) or a nociceptive neuron [wide dynamic range (WDR) or nociceptive-specific (NS)] (Chiang et al. 1998; Hu 1990; Park et al. 2001). The presence of a deep nociceptive input was considered to occur if the application of a blunt probe to muscle, bone, tendon, or temporomandibular joint (TMJ) evoked a neuronal response at a mechanical threshold >5 g but no response could be evoked by the wide range of cutaneous stimuli used (Chiang et al. 1994; Iggo 1960; Park et al. 2001; Schaible and Schmidt 1983; Yu et al. 1993). Electrical stimuli of constant-current single pulses (0.2 ms and <1 mA for A-fiber inputs; 2 ms and <5 mA for C-fiber inputs) were applied within the delineated mechanoreceptive field (RF) to determine the existence of A- or C-fiber afferent inputs, and monopolar stimuli of cathodal current single pulses (0.2 ms, <2 mA) were applied to the exposed maxillary molar tooth pulp to determine the existence of a dental afferent input (Chiang et al. 1998; Park et al. 2001); we did not attempt to determine if the evoked Vo neuronal responses reflected monosynaptic or polysynaptic afferent inputs.

The spontaneous activity, RF size, and responses to mechanical stimuli were assessed at the time intervals specified in the following text. Spontaneous activity was measured as the average frequency (Hz) of spikes occurring for 2 min. The RF of each neuron was determined through the use of a brush, blunt probe, and a pair of nonserrated forceps. As previously described (Park et al. 2001), the orofacial region was empirically divided into 20 small areas (see Fig. 1), and the size of an orofacial RF was quantified by summing the number of these areas included in the RF. To evaluate the reliability of this area-summing method, the extent of the neuron's cutaneous RF was also outlined on a life-size drawing of the rat's head and measured by a computer-aided device (SigmaScan, Jandel, CA) as previously described (Chiang et al. 1997, 1998; Yu et al. 1993). Mechanical threshold of the neuronal RF was not studied because of difficulty to access the intraoral RF of most neurons. Consistent with our other recent study of Vo neurons (Park et al. 2001), we tested for responses to graded mechanical pinch or pressure stimuli (5, 10, and 20 g for WDR neurons; 50, 100, and 200 g for NS neurons) that were applied to the neuronal orofacial RF were tested with a force-monitoring forceps; each stimulus was applied for 3 s at an interval of >30 s. As in our previous study (Park et al. 2001), the lower stimulus intensity range for WDR neurons was chosen because these neurons have a lower mechanical activation threshold and intensity for mid-range response than NS neurons (Chiang et al. 1994, 1998; Hu et al. 1981), and we chose an intensity that would elicit a mid-range suprathreshold response for each WDR neuron and NS neuron and applied it three times at the same site within the neuron's RF. The responses were quantified as the average of the number of spikes produced by this standardized stimulus.

View larger version (26K): [in this window] [in a new window]   Fig. 1. The orofacial region was divided into the areas indicated in the diagram based on anatomical demarcation combined with neuronal functional characteristics. The size of a neuronal orofacial mechanoreceptive field (RF) was quantified by summing the number of areas included in the RF. AHPal, anterior hard palate; Fr, frontal; LB, lower buccal; LF, lateral face; LInc, lower incisor and gum; LLmu, lower lip mucosa; LLsk, lower lip skin; LMol, lower molar and gum; MF, medial face; Ment, mental; N, nasal; Orb, orbital; PHPal, posterior hard palate; Tg, tongue; UB, upper buccal; UInc, upper incisor and gum; ULmu, upper lip mucosa; ULsk, upper lip skin; UMol: upper molar and gum, Vib: vibrissal pad.

Drug administration

MO (allyl isothiocyanate, 95%; Aldrich Chemical) was applied locally to the exposed molar pulp to induce central sensitization of Vo nociceptive neurons that typically lasts >40 min (Park et al. 2001). Therefore to test for the possible Vc synaptic mediation of the MO-induced central sensitization of Vo nociceptive neurons, CoCl₂ (Sigma-Aldrich Canada) was freshly dissolved in saline (5 mM) and a 0.3 µl bolus was microinjected by a manual microinjector (Model 5000, David Kopf) over 45-60 s into the right Vc (1.0 mm behind the obex; 1.2 mm lateral to the midline; 1.0 mm below the medullary surface) at ~20 min after MO application. Vehicle (isotonic saline, 0.3 µl) application served as a control. In some experiments (see following text), CoCl₂ was instead microinjected into subnucleus interpolaris (Vi), 1 mm rostral to the obex and 1.5 mm lateral to the midline. To mark the injection site so that the extent of drug diffusion could be assessed, a 0.3 µl bolus of pontamine sky blue solution (2% in saline) was first loaded into the syringe, followed by an air gap (0.1 µl), and then the syringe filled with the CoCl₂ solution.

Experimental paradigm

In each animal, only one neuron was tested with saline or CoCl₂ after the pulp application of MO. Experimental animals were divided into three groups according to the drug or saline application: MO application followed by saline injection into Vc (Saline/Vc); MO application to the pulp followed by CoCl₂ injection into Vc (CoCl₂/Vc); MO application followed by CoCl₂ injection into Vi (CoCl₂/Vi). The latter experiment was designed to test the possibility that any CoCl₂ blocking effect on Vo neurons might be caused by drug diffusion from the injection site (Vc) directly to Vo. A similar experimental paradigm was applied for all three groups: 10 min after a Vo nociceptive neuron was identified, the neuronal spontaneous activity, orofacial RF size and responses to mechanical stimuli were determined and served as baseline values; a complete assessment of these properties took ~6-7 min. Then the saline-soaked cotton pellet in the prepared molar cavity was carefully replaced by a segment of dental paper point soaked with MO (0.2 µl), and the cavity quickly sealed with CAVIT (ESPE, Germany). Another assessment of neuronal properties started at 3 min after MO application, and repeated at 8- to 10-min intervals until 60 min. At ~20 min after MO application, saline or CoCl₂ was locally injected into right Vc (or Vi in the CoCl₂/Vi group).

Histological and statistical analysis

An electrolytic lesion (anodal current, 8 µA, 10 s) was made at the recording site at the end of the experiment. Ten minutes before the beginning of transcardial perfusion, the pontamine sky blue solution (0.3 µl) that was left in the injection syringe was slowly injected at the same site. Verification of the recording and injection sites and the dye's approximate diffusion extent was assessed with conventional histological procedures.

Statistical treatments were performed on the normalized data. For assessing the drug effects, differences between baseline (predrug) values and values at different postdrug time points in each group were treated by repeated-measures ANOVA (RM ANOVA) or RM ANOVA on ranks followed by Dunnett's method. Differences between groups were treated by ANOVA on ranks (Denenberg 1984). Differences between groups at a given time point were treated by a priori Mann-Whitney rank sum test or t-test. All values were expressed as means ± SE except for those measures of orofacial RF size that involved summing the areas (see preceding text); these were expressed as the median and 25th and 75th percentiles. The level of significance was set at P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Twenty-five nociceptive (18 WDR, 7 NS) Vo neurons were recorded and studied with the full range of tests, but 3 LTM neurons recorded did not show central sensitization to pulp application of MO and so were excluded from further analysis. The recording sites of all 25 nociceptive neurons were in the middle and ventral part of Vo, except for one site close to the border of its dorsomedial (DM) part (Fig. 2).

View larger version (28K): [in this window] [in a new window]   Fig. 2. The unit recording sites lesioned by anodal current of 8 µA, 10 s and confirmed histologically by conventional Cresyl violet staining were plotted on diagrams of the medulla. A: coronal sections at 3 different rostrocaudal levels of subnucleus oralis (Vo; numbers on right refer to distance relative to the interaural line). B: a sagittal section of the medulla (number on right refers to distance relative to midline). Filled symbols represent wide-dynamic range (WDR) neurons; opened symbols nociceptive-specific (NS) neurons. The dot, triangle, and square symbols represent the Saline/Vc, CoCl2/Vc, and CoCl2/Vi groups, respectively. 7, facial nucleus; 7n, facial nerve; dm, dorsomedial portion of Vo; Vpr, V main sensory nucleus.

All 7 NS neurons and 16 of the WDR neurons had at baseline (i.e., before MO application) an ipsilateral RF that involved the maxillary and/or mandibular divisions and that was located in the intraoral and/or perioral regions; only 2 WDR neurons had a cutaneous RF involving the ophthalmic division. In most WDR neurons, the location of the cutaneous or intraoral tactile RF was within the pinch/pressure RF, but in five WDR neurons, the intraoral tactile RF was separate from the intraoral pinch/pressure RF in the same division or in a different division. Most (56%) of the 16 WDR neurons and 50% of the six NS neurons tested received electrically evoked A-fiber inputs from their cutaneous or intraoral mucosal RFs; 33% of these responsive WDR neurons also received C-fiber inputs. Moreover, 56% of the 16 WDR neurons tested, but none of the 6 NS neurons, received electrically evoked A- or C-fiber molar inputs. Spontaneous activity was noted in 28% of the WDR and NS neurons; the spontaneous firing rate was <0.25 Hz for most of these neurons.

Saline injected into Vc does not affect MO-induced neuroplastic changes in Vo nociceptive neurons

In the Saline/Vc group, pulp application of MO induced significant neuroplastic changes in all nine Vo nociceptive (6 WDR, 3 NS) neurons tested. The neuroplastic changes included increases in orofacial RF size and responses to pinch or pressure stimuli (Figs. 3A, 4-6, Table 1), but no consistent changes in spontaneous activity occurred after MO application or saline injection (Table 1). At baseline, three of the nine neurons had both perioral and intraoral RFs, four neurons had only a perioral RF (including the vibrissal pad), and the other two had only an intraoral RF.

View larger version (45K): [in this window] [in a new window]   Fig. 3. Examples of changes in orofacial RF and pinch- or pressure-evoked responses of 3 WDR neurons, each from the Saline/Vc (A), CoCl2/Vc (B), or CoCl2/Vi (C) group. A similar paradigm was used in these 3 different experiments: Soon after a stable nociceptive unit recording in Vo, the baseline values of spontaneous activity, orofacial RF size and response to noxious stimuli were assessed. The mustard oil (MO) was applied to the exposed molar pulp and 3 min later the same assessment was repeated at an interval of 8 min until 60 min. In addition, 20 min after the MO application, either a 0.3 µl of 5 mM CoCl2 solution was slowly microinjected into Vc (B) or Vi (C) or a vehicle (saline) into Vc (A). Note that the pinch RFs are shown in the top panel, the square pulse in the middle panel represents the 3-s stimulation period and the pinch- or pressure-evoked responses are shown in the bottom panel; numbers below each lower panel indicate the time (min) after MO application. Marked neuroplastic changes occurred in all 3 WDR neurons after MO application. While CoCl2 injected into Vc effectively and reversibly blocked these neuroplastic changes (B), saline injected into Vc (A) or CoCl2 injected into Vi (C) did not affect these changes, which lasted beyond 40 min. A smaller-sized unit was active during the neuronal recordings in A; this unit's activity was not included in the data analyses.

OROFACIAL RF SIZE. MO application produced a significant, long-lasting increase in tactile and pinch/pressure RF size in all nine nociceptive neurons tested. As shown in Fig. 4 and Table 1, the pinch/pressure RF size increased significantly throughout the 40-min period following MO application (P < 0.001, RM ANOVA on ranks), with its peak around 26 min (median: 200%; 25th75th percentile: 161-338% of baseline; P < 0.05, Dunnett's method), despite saline injection into Vc at 20 min. MO application produced a novel intraoral or perioral pinch or pressure RF in three WDR and two NS neurons having, respectively, a baseline perioral or intraoral RF. In addition, the cutaneous RF size of six of the nine neurons was quantitatively assessed (see METHODS) before and after MO application. As shown in Fig. 5, the cutaneous pinch RF size showed a significant and prolonged increase following MO application, which peaked at 26 min (mean ± SE: 348 ± 80%; range: 120-592% of baseline; P < 0.001, RM ANOVA). The time course of the cutaneous pinch RF changes of these neurons was similar to the orofacial pinch/pressure RF changes measured by summing the number of areas in the orofacial RF (see preceding text and Fig. 4).

View larger version (22K): [in this window] [in a new window]   Fig. 4. Time courses of the MO-induced changes in orofacial pinch/pressure RF size of Vo nociceptive neurons in the 3 groups of animals, and the effects on these changes of CoCl2 or saline injected into subnucleus caudalis [Vc; or subnucleus interpolaris (Vi)]. Note that significant differences (*P < 0.05) between baseline (0 min) values and values at 18- to 40-min time points in Saline/Vc group (A, n = 9), values at 18- to 40-min time points in CoCl2/Vi group (C, n = 6), and values at 8-, 18-, and 40-min but not at 26- and 32-min time points in CoCl2/Vc group (B, n = 10). Differences in values at the 26-min time point between the CoCl2/Vc group and other 2 groups are also significant (#P < 0.05). Arrow 1 indicates the time when MO was applied to the pulp. Arrow 2 indicates of the time of CoCl2 or saline injection. Median: transverse line within box; 75th percentile: top half box; 25th percentile: bottom half box; 95th percentile: top bar; 5th percentile: bottom bar.

View larger version (21K): [in this window] [in a new window]   Fig. 5. Time courses of the MO-induced changes in facial cutaneous pinch RF size of Vo nociceptive neurons in the 2 groups of animals and the effects on these changes of CoCl2 or saline injected into Vc. Note significant differences (*P < 0.05) between baseline (0 min) values and post-CoCl2 (or saline) values at the different time point in CoCl2/Vc (filled circle, n = 5) and Saline/Vc (empty circle, n = 6). Differences in values at the 26- and 32-min time points between the CoCl2/Vc and Saline/Vc groups are also significant (#P < 0.05). The arrow at 0 min indicates the MO application time. The arrow at 20 min indicates the CoCl2 or saline injection time.

RESPONSES TO SUPRATHRESHOLD MECHANICAL STIMULI. After MO application, neuronal responses to standardized, mid-range suprathreshold pinch or pressure stimuli were increased in all eight nociceptive neurons tested. The response to noxious mechanical stimuli gradually increased to a peak at 18 min that was significantly different from baseline (333 ± 62% of baseline; range: 76-671%; P < 0.02, RM ANOVA; Fig. 6, Table 1), then declined slowly toward baseline level by 60 min. Saline injection into Vc at 20 min did not affect these changes.

View larger version (24K): [in this window] [in a new window]   Fig. 6. The time courses of the MO-induced changes in pinch- or pressure-evoked response of Vo nociceptive neurons in the 3 groups of animals and the effects on these changes of CoCl2 or saline injected into Vc (or Vi). Note significant differences (*P <0.05) between baseline (0 min) values and post-CoCl2 (or saline) values at the different time point in Saline/Vc (, n = 8), CoCl2/Vc (, n = 8) and CoCl2/Vi groups (, n = 6). Differences in values at the 26-min time point between the CoCl2/Vc group and other 2 groups are also significant (#P <0.05).  (left to right), MO and CoCl2 or saline injection times, respectively.

CoCl₂ injected into Vc reversibly blocks MO-induced neuroplastic changes in Vo nociceptive neurons

In the CoCl₂/Vc group, as in the Saline/Vc group and the CoCl₂/Vi group (see following text), pulp application of MO induced significant neuroplastic changes in all Vo nociceptive (7 WDR, 3 NS) neurons tested. These changes involved increases in RF size and responses to pinch or pressure stimuli; no consistent changes in spontaneous activity occurred after MO application or CoCl₂ injection (Table 1). An example is shown in Fig. 3B. At baseline, 3 of the 10 neurons tested had both intraoral and perioral (including facial) RFs, 3 neurons had only a facial and/or perioral RF, and the remaining 4 neurons had only an intraoral RF. An example of the CoCl₂ injection site in Vc is illustrated in Fig. 7A in which the diffusion extent, as marked by the subsequent injection of dye, was confined within a region with a diameter of 1.2 mm around the end of the needle track.

View larger version (107K): [in this window] [in a new window]   Fig. 7. Photomicrographs of parasagittal sections taken from the lower brain stem that displayed the CoCl2 injection site (black arrow) in Vc (A) and the recording site (white arrow) in Vo in 1 experiment (0023) of the CoCl2/Vc group, and the CoCl2 injection site (black arrow) in Vi and the recording site (white arrow) in Vo (D) in another experiment (0035) of the CoCl2/Vi group. Note that the diffusion extent in both Vo and Vc, as marked by the subsequent injection of dye, was confined within a region with a diameter of 1.2 mm around the end of the needle track. The sections were cut by a vibratome in 30 µm and stained with cresyl violet. VII, facial nucleus. Scale: 1 mm.

OROFACIAL RF SIZE. After MO application, the neuronal pinch/pressure RF size increased from 8 min and peaked ~18 min (median:171%, 25th75th percentile: 126-200% of baseline; P < 0.001, RM ANOVA on ranks). As in the Saline/Vc group, MO application produced a novel intraoral or perioral RF in four WDR and two NS neurons having, respectively, a baseline perioral or intraoral RF. CoCl₂ solution injected into Vc ~20 min after MO application produced a significant blockade of the increased RF size within 6 min in eight neurons and at 12-15 min in the remaining two neurons; the median value at 12 min after CoCl₂ injection was 105% (25th75th percentile: 100-150% of baseline, P > 0.05, Dunnett's method; Fig. 4). As shown in the example in Fig. 3B, the blockade was typically limited to the expanded portion of the RF, not the baseline RF, and lasted for 15-20 min. Thereafter, the MO-induced neuroplastic changes returned (median:129%, 25th75th percentile: 114-167%; P < 0.05, Dunnett's method) and then gradually diminished (at 60 min: median: 124%, 25th75th percentile: 114-160% of baseline; P > 0.05, Dunnett's method; Fig. 4). The CoCl₂-induced blockade and its time course were also reflected in the changes in orofacial RF size determined quantitatively in five of the neurons (Fig. 5).

RESPONSES TO SUPRATHRESHOLD MECHANICAL STIMULI. After MO application, the neuronal responses to standardized suprathreshold stimuli applied to the neuronal RF significantly increased (P = 0.002, RM ANOVA) and peaked ~18 min (281 ± 53%, range: 95-505% of baseline; P < 0.05, Dunnett's method). CoCl₂ injected into Vc invariably produced an abrupt and significant blockade of the increased responses to suprathreshold stimuli (103 ± 28%, range: 18-233% of baseline; P > 0.05, Dunnett's method; Fig. 6) that occurred simultaneously with the reversal of the orofacial pinch/pressure RF increase (see preceding text). Thereafter, the increased responses returned at ~20 min after the CoCl₂ injection (243 ± 54%, range: 143-573% of baseline; P < 0.05, Dunnett's method; Fig. 6) and slowly returned to baseline levels. The time course of response changes was similar to that of RF size changes (Figs. 4 and 6). An example is shown in Fig. 3B.

CoCl₂ injected into Vi does not affect MO-induced neuroplastic changes in Vo nociceptive neurons

In the CoCl₂/Vi group, pulpal application of MO induced significant increases in orofacial RF size and responses to pinch or pressure stimuli in all six (5 WDR, 1 NS) neurons tested (Figs. 3C, 4, and 6, Table 1), but no consistent changes in spontaneous activity occurred after MO application or CoCl₂ injection (Table 1). At baseline, all these neurons had an intraoral RF, and one WDR neuron also had a perioral RF and one NS neuron also had a RF in the TMJ region. An example of the CoCl₂ injection site in Vi is illustrated in Fig. 7C.

OROFACIAL RF SIZE. After MO application, the intraoral RF of the neurons expanded into the perioral region which was sensitive only to pinch stimuli (e.g., Fig. 3C) and was maintained throughout the 60-min observation period. Significant increases in the pinch/pressure RF size occurred from 18 to 60 min (P < 0.001, RM ANOVA on ranks) and peaked at 32 min after MO application (median: 171%, 25th75th percentile: 147-225% of baseline; P < 0.05, Dunnett's method) despite CoCl₂ solution injected into Vi at 20 min after the MO application (Figs. 4, Table 1). As in the Saline/Vc and CoCl₂/Vc groups, MO application produced a novel perioral RF in four WDR and one NS neurons having a baseline intraoral RF only; two of these WDR neurons also acquired a deep RF involving jaw muscle and TMJ.

RESPONSES TO SUPRATHRESHOLD MECHANICAL STIMULI. Significant increases in responses to suprathreshold mechanical stimuli of the six neurons occurred after MO application and were maintained despite the injection of CoCl₂ into Vi at 20 min after MO application (P < 0.004, RM ANOVA). The increases peaked at 26 min after MO application and 6 min after CoCl₂ injection into Vi (387 ± 97%, range: 167-667% of baseline; P < 0.05, Dunnett's method) and returned toward baseline level by 60 min (Fig. 6; Table 1). The differences in responses between the CoCl₂/Vi and CoCl₂/Vc groups were significant (P < 0.001, ANOVA on ranks, Table 1), including those at the 26-min time point after MO application (P < 0.02, Mann-Whitney test).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Consistent with our previous findings (Park et al. 2001), we have documented that application of MO to the rat molar pulp produces central sensitization in Vo that is reflected in significant neuroplastic changes in Vo nociceptive neurons, including increases in orofacial RF size and responses to noxious mechanical stimuli. We have provided new data that microinjection into Vc of CoCl₂, an effective synaptic blocker, significantly and reversibly reduces the MO-induced neuroplastic changes of Vo nociceptive neurons. These findings, together with our other data that the same quantity of CoCl₂ injected into Vi (which is located between Vc and Vo) did not affect the MO-induced neuroplastic changes of Vo nociceptive neurons, indicate that the central sensitization induced in Vo nociceptive neurons by stimulation of the pulp may be dependent on the functional integrity of the rostral part of Vc.

Cobalt is thought to prevent the release of synaptic transmitters by blocking calcium channels in the presynaptic membrane (Hagiwara and Byerly 1981). CoCl₂ has often been used to block central synaptic transmission in various sensory and autonomic pathways (Allen and Pronych 1997; Hochstenbach and Ciriello 1997; Lee and Malpeli 1985; Malpeli 1983; Mooney et al. 1992; Nuseir et al. 1999). It has several advantages over intracerebral injection of local anesthetics, such as lidocaine, and neurotransmitter modulatory agents, such as the GABA agonist muscimol, which have been used in some other studies. These include a selective block of synaptic transmission rather than fibers of passage (relative to lidocaine), an aqueous solution that is less diffusible to neighboring tissues, a faster onset and shorter duration of blocking action, less toxicity on primary afferent neurons and block of all synaptic transmission (whereas muscimol may not block firing in all neurons i.e., if they do not have GABA receptors) (Gold et al. 1998; Lee and Malpeli 1985; Malpeli 1983; Mooney et al. 1992). Furthermore, a 120 nl 4 mM CoCl₂ solution can safely be used repeatedly without any tissue damage (Lee and Malpeli 1985), and 25 nl 10 mM CoCl₂ can inactivate a volume of tissue with a diameter under 250 µm for 200 s (Mooney et al. 1992). In addition, 100-300 nl 5-10 mM CoCl₂ microinjected into the medulla can block pressor or depressor responses for 40 min (Allen and Pronych 1997; Hochstenbach and Ciriello 1997). However, if too great a volume or concentration of CoCl₂ is injected, fibers of passage or local tissues may be damaged (Lee and Malpeli 1985; Malpeli 1983).

With these findings in mind, we made only one 300 nl injection of a 5 mM CoCl₂ solution into Vc or Vi. We also avoided several smaller injections at different sites or depths because this would take several minutes to complete and be impractical in our experimental paradigm. The injection site selected was that region that contains Vc nociceptive neurons with an orofacial RF (Chiang et al. 1998) and corresponds to the central termination area in the rostral part of the caudal medulla of orofacial primary afferents (Kobayashi and Matsumura 1996; Takemura et al. 1991). Given that the needle tip was placed at 1.0 mm below the medullary surface and we estimated that the extent of CoCl₂ diffusion was 1.2 mm in diameter, it is likely that the CoCl₂ injection blocked most of the intersubnuclear bundles projecting from Vc to Vo (Gobel and Purvis 1972; Ikeda et al. 1982, 1984; Jacquin et al. 1990; Nasution and Shigenaga 1987; Panneton and Burton 1982; Voisin et al. 2002). Furthermore, our findings that CoCl₂ injected into Vi did not affect the neuroplastic changes induced in Vo indicate that the effects on Vo neurons of the CoCl₂ injection into Vc cannot be accounted for by CoCl₂ diffusion directly from Vc to Vo. Although Vi as well as Vc contains intersubnuclear bundles projecting to Vo, the ineffectiveness of CoCl₂ blockade of Vi might be explained by findings that, compared with Vc, Vi lacks laminae I-II (which in Vc contributes to the deep bundles) and has a very limited density of putative neurotransmitters or receptors associated with nociceptive transmission (Mansour et al. 1994; Petralia et al. 1994; Sessle 2000; Tallaksen-Greene et al. 1992). If this explanation is correct, it points to the critical role played by the superficial laminae I-II of Vc in generating central sensitization that might then be mediated to Vo via V intersubnuclear bundles (see following text).

Central sensitization is thought to be reflected in neuroplastic changes that can be triggered by nociceptive afferent inputs and that are manifested in spinal and medullary dorsal horn nociceptive neurons as an increase in spontaneous activity, an expansion of the RF, and an increase in the responsiveness to stimuli (Chiang et al. 1998; Coderre and Katz 1997; Dubner 1991; Hu et al. 1992; Li and Woolf 2001; Ma and Woolf 1996; Morisset and Nagy 2000; Sotgiu and Biella 2000; Suzuki et al. 2000; Willis 1993; Woolf 1992; Yu et al. 1993). These neuroplastic changes have also been previously documented in Vo nociceptive neurons following MO application to the tooth pulp (Park et al. 2001). Of particular note are our present findings that the neuroplastic changes in RF size and response properties could be transiently and reversibly reduced by CoCl₂ injection into Vc. The findings are consistent with the earlier observations, although we did not observe a consistent long-lasting increase in spontaneous activity as previously reported (Park et al. 2001). The latter results could be explained by our present sample of nociceptive neurons tested with MO application; it comprised almost 50% of neurons having an RF involving only the intraoral region, and the Park et al. (2001) study showed no change in MO-induced spontaneous activity in such neurons.

Our findings that Vc blockade abolishes pulp-induced neuroplastic changes in Vo nociceptive neurons indicate that the blockade is likely interfering with the relay of MO-evoked pulp afferent inputs via Vc to Vo because Vc is the termination site of many pulp afferents (Clements et al. 1991; Nishikawa et al. 1997; Takemura et al. 1991) that are very effective in inducing central sensitization in Vc neurons (Chiang et al. 1998). The effect of Vc blockade could also be explained by its disruption of a tonic influence that Vc is exerting on Vo central sensitization induced by pulp afferent inputs directly to Vo, although it has been argued that Vo central sensitization most likely depends on the neuronal substrates in Vc (see Chiang et al. 1998; Dallel et al. 1998; Parada et al. 1997; Park et al. 2001; Voisin et al. 2002; Woda et al. 2001; for review, see Sessle 2000). As noted in the preceding text, our data are nonetheless consistent with the considerable anatomical interconnections that exist between Vc and the rostral components of V spinal tract nucleus, including Vo. It has been well documented that the main intranuclear projection of neurons in Vc is via ascending deep intersubnuclear pathways (Gobel and Purvis 1972; Ikeda et al. 1982, 1984; Jacquin et al. 1990; Nasution and Shigenaga 1987; Panneton and Burton 1982). Furthermore, our present data are also consistent with earlier documentation that Vc exerts a net facilitatory influence on Vo (Greenwood and Sessle 1976; Hu and Sessle 1979; Khayyat et al. 1975;Young and King 1972); this is also supported by recent findings that C-fiber-evoked responses of Vo nociceptive neurons can be depressed by morphine microinjection into the superficial laminae of Vc, but not into Vo, and that N-methyl-D-aspartate antagonist MK-801 microinjection into the lateral part of Vc can markedly depress "wind-up" in some Vo nociceptive neurons (Dallel et al. 1998; Woda et al. 2001). Nonetheless, local modulatory mechanisms in Vo itself cannot be neglected as possible factors influencing nociceptive transmission in the rat Vo because morphine microinjected into Vo can reduce the formalin-induced nociceptive behavior (Luccarini et al. 1995) and MK-801 microinjection into Vo can significantly reduce MO-evoked neuroplastic changes in Vo (Park et al. 2001).

Although CoCl₂ microinjected into Vc reversibly blocked the MO-induced neuroplastic changes in Vo nociceptive neurons, both orofacial RF size and responses to noxious stimuli of Vo nociceptive neurons were retained close to their baseline levels. This suggests that Vo orofacial nociceptive processing per se is not influenced by CoCl₂-induced block of Vc, whereas the baseline RF properties of LTM neurons in Vo can be reversibly influenced by cold block of Vc (Greenwood and Sessle 1976). Therefore it seems likely that the mechanisms underlying orofacial nociceptive processing may involve neural substrates in both Vo and Vc; this is consistent with earlier findings (Broton and Rosenfeld 1986; Graham et al. 1988; Luccarini et al. 1998; Pickoff-Matuk et al. 1986), while those underlying MO-induced central sensitization may mainly involve Vc. Thus the functional integrity of Vc appears to be a prerequisite condition for maintenance of the central sensitization in Vo and perhaps in other components of the V brain stem complex or higher brain centers that still have to be explored.