Hyperalgesia and sensitization of dorsal horn neurons following activation of NK-1 receptors in the rostral ventromedial medulla

It is known that activation of neurokinin-1 (NK-1) receptors in the rostral ventromedial medulla (RVM), a main output area for descending modulation of pain, produces hyperalgesia. Here we show that activation of NK-1 receptors produces hyperalgesia by sensitizing nociceptive dorsal horn neurons. Targeting this pathway at its origin or in the spinal cord may be an effective approach for pain management.

Abstract

Neurons in the rostral ventromedial medulla (RVM) project to the spinal cord and are involved in descending modulation of pain. Several studies have shown that activation of neurokinin-1 (NK-1) receptors in the RVM produces hyperalgesia, although the underlying mechanisms are not clear. In parallel studies, we compared behavioral measures of hyperalgesia to electrophysiological responses of nociceptive dorsal horn neurons produced by activation of NK-1 receptors in the RVM. Injection of the selective NK-1 receptor agonist Sar9,Met(O₂)¹¹-substance P (SSP) into the RVM produced dose-dependent mechanical and heat hyperalgesia that was blocked by coadministration of the selective NK-1 receptor antagonist L-733,060. In electrophysiological studies, responses evoked by mechanical and heat stimuli were obtained from identified high-threshold (HT) and wide dynamic range (WDR) neurons. Injection of SSP into the RVM enhanced responses of WDR neurons, including identified neurons that project to the parabrachial area, to mechanical and heat stimuli. Since intraplantar injection of capsaicin produces robust hyperalgesia and sensitization of nociceptive spinal neurons, we examined whether this sensitization was dependent on NK-1 receptors in the RVM. Pretreatment with L-733,060 into the RVM blocked the sensitization of dorsal horn neurons produced by capsaicin. c-Fos labeling was used to determine the spatial distribution of dorsal horn neurons that were sensitized by NK-1 receptor activation in the RVM. Consistent with our electrophysiological results, administration of SSP into the RVM increased pinch-evoked c-Fos expression in the dorsal horn. It is suggested that targeting this descending pathway may be effective in reducing persistent pain.

NEW & NOTEWORTHY It is known that activation of neurokinin-1 (NK-1) receptors in the rostral ventromedial medulla (RVM), a main output area for descending modulation of pain, produces hyperalgesia. Here we show that activation of NK-1 receptors produces hyperalgesia by sensitizing nociceptive dorsal horn neurons. Targeting this pathway at its origin or in the spinal cord may be an effective approach for pain management.

pain and hyperalgesia are mediated in part by sensitization of nociceptive dorsal horn neurons, referred to as central sensitization. Descending pathways from the brain stem, including the rostral ventromedial medulla (RVM), modulate nociceptive transmission in the spinal cord and contribute to central sensitization and persistent pain. Although early studies focused on descending inhibition of pain, parallel descending pathways have been identified that facilitate nociceptive transmission and contribute to the development of chronic pain and hyperalgesia (Kovelowski et al. 2000; Ossipov et al. 2014; Porreca et al. 2002; Vanderah et al. 2001).

In the RVM, there are two main functional groups of neurons, ON and OFF cells (Fields et al. 1983) that facilitate or diminish nociceptive transmission in the spinal cord, respectively (Fields 2004; Fields et al. 1998, 1991; Heinricher et al. 2009). Substance P (SP) is present in the RVM, and NK-1 receptors are located on RVM neurons (Budai et al. 2007; Khasabov and Simone 2013; Marson and Loewy 1985; Nakaya et al. 1994; Saffroy et al. 1988). Activation of NK-1 receptors in the RVM by SP increased excitability of ON cells (Budai et al. 2007). Sensitization of ON cells produced by SP injection into the RVM, intraplantar injection of capsaicin, or peripheral inflammation was decreased by a NK-1 receptor antagonist (Brink et al. 2012; Budai et al. 2007; Khasabov et al. 2012), suggesting NK-1 receptors were located on cells. Behavioral studies showed that microinjection of SP into the RVM produced dose-dependent hyperalgesia to heat, and blockade of NK-1 receptors, or selective ablation of RVM neurons that express NK-1 receptors, attenuated hyperalgesia produced by inflammation or intraplantar injection of capsaicin (Brink et al. 2012; Hamity et al. 2010; Khasabov and Simone 2013; Khasabov et al. 2012; Pacharinsak et al. 2008). Although these studies support the notion that activation of NK-1 receptors on cells in the RVM facilitates nociceptive transmission in the spinal cord, the effects of NK-1 receptor activation in the RVM on response properties of dorsal horn neurons are unknown.

It has been proposed that descending facilitation and central sensitization are initiated and maintained by spinal-supraspinal circuitry (Dickenson et al. 2005; Suzuki et al. 2004a, 2004b). One possibility is that ascending projection neurons in the dorsal horn excite ON cells, which in turn sensitizes dorsal horn neurons, including the projection neurons that drive descending facilitation. One pathway involved in driving descending modulation is the spinoparabrachial tract (SPbT) (Lapirot et al. 2009; Vera-Portocarrero et al. 2007). SPbT neurons are located in the superficial dorsal horn (Kitamura et al. 1993; Menétrey and De Pommery 1991; Murphy et al. 2009; Saper 1995; Todd et al. 2000; Jansen and Giesler 2015), encode noxious stimuli (Andrew 2010; Andrew 2009; Bester et al. 2000; Jansen and Giesler 2015), and are sensitized following tissue injury (Andrew 2009; Ikeda et al. 2003; Jansen and Giesler 2015). SPbT neurons project to the parabrachial area (PbA) of the dorsolateral pons, and their activation affects neurons involved in descending modulation (Wall et al. 1988). Inactivation of the lateral PbA decreased evoked responses of RVM neurons directly (Beitz 1982; Hermann et al. 1997; Roeder et al. 2016; Verner et al. 2008) or indirectly (McGaraughty et al. 2004; McGaraughty and Heinricher 2002) through connections between the PbA and the RVM.

Therefore, in correlative behavioral and electrophysiological studies, we compared the effects of NK-1 receptor activation in the RVM on behavioral measures of nociception and on response properties of nociceptive dorsal horn neurons, including identified SPbT neurons.

MATERIALS AND METHODS

Animals

Adult, male, Sprague–Dawley rats (Harlan, Indianapolis, IN) weighing 250–400 g were housed in a climate-controlled room maintained on a 12-h dark-light cycle with food and water available ad libitum. All experiments were performed during the light cycle and were conducted in accordance with the guidelines recommended by the International Association for the Study of Pain. All procedures were approved by the Animal Care and Use Committee at the University of Minnesota.

Behavioral Studies

Chronic cannulation of the RVM.

Rats were anesthetized with acepromazine maleate (2 mg/kg sc) and ketamine HCl (80 mg/kg sc) and placed in a stereotaxic apparatus. Body temperature was maintained near 37°C using a feedback-controlled electrical blanket (Harvard apparatus, Holliston, MA). A small craniotomy was made, and a guide cannula (17.5 mm in length, 26 gauge; Plastics One, Roanoke, VA) was lowered toward the medial portion of the RVM. The interaural stereotaxic coordinates were as follows: anterior-posterior (AP) = 2.3 mm; media-lateral (ML) = 0 mm; and dorsal-ventral (DV) = 0.5 mm according to the rat brain atlas (Paxinos and Watson 1998). The guide cannula was secured using dental resin (Duralay, Dental, Worth, IL) that was cemented to three stainless steel machine screws that were screwed to the skull. A dummy cannula (33 gauge; Plastics One) was inserted into the guide cannula to maintain patency. Animals were allowed to recover for 5–7 days before behavioral testing. To control for injection site specificity, some rats were implanted with a cannula outside the RVM (AP = −2.3 mm; ML = 0–1.4 mm; and DV = 1.4–2 mm.).

Drug administration.

Drug solutions (0.5 µl) were injected into the RVM through a 33-gauge injection cannula that extended 1 mm below the tip of the guide cannula (AP = 2.3 mm; ML = 0 mm; and DV = −0.5 mm). The selective NK-1 receptor agonist, Sar9,Met(O₂)¹¹-Substance P (SSP) (cat. no. 1178; Tocris Bioscience, Bristol, UK) was administered in doses of 0.1, 0.3, 1, 3, 5, and 10 nmol. The selective NK-1 receptor antagonist L-733,060 (cat. no. 1145; Tocris Bioscience), was given at a dose of 1.5 pmol. All drugs were dissolved in saline. The injection cannula was attached to PE-10 tubing (Plastics One), and injections were made using a 1-µl microsyringe (Hamilton, Reno, NV) and administered slowly over a period of 2 min in a volume of 0.5 µl. The cannula was left in position for 90 s after injection to allow for diffusion and then removed and replaced with the dummy cannula. At the end of each experiment, 0.5 µl of pontamine sky blue were injected through the cannula and placement in the RVM was verified histologically. As we have described previously (Khasabov and Simone 2013), the RVM is considered an area of the medulla oblongata with rostra-caudal dimension from the caudal margins of the trapezoid body to the rostral pole of the inferior olives. In the coronal plane, this was a rectangular area with the ventral border (horizontal) located on the dorsal margins of the pyramids, the dorsal border (horizontal) at the level of most dorsal margins of facial nuclei, and lateral borders (vertical) midway between the lateral edge of the pyramid and the most medial edge of the facial nucleus (Fig. 1E). Only rats with a cannula correctly placed in the RVM were used.

Fig. 1.

Injection of Sar9,Met(O₂)¹¹-substance P (SSP) into the rostral ventromedial medulla (RVM) produces dose-dependent hyperalgesia. A: mean (±SE) 50% paw withdrawal thresholds (g) to mechanical stimuli before, and up to 85 min after injection of vehicle or 0.5, 3, and 10 nmol of SSP into the RVM. Doses of 3 and 10 pmol SSP produced mechanical hyperalgesia as compared with vehicle (*P < 0.05 and ***P < 0.005). BL, baseline. B: dose-dependent mechanical hyperalgesia represented as mean (±SE) areas under the curve (AUC). Doses are displayed on a log scale. The AUC following vehicle injection is indicated by the open triangle. AUCs for 3 doses (3, 5, and 10 pmol SSP) were statistically significantly lower as compared with vehicle (*P < 0.05, ****P < 0.001). The ED₅₀ value was 2.75 nmol. C: mean (±SE) paw withdrawal latencies (in seconds) before and up to 85 min after injections of vehicle or 0.5, 3, and 10 nmol SSP. The dose of 10 pmol has reduced withdrawal latencies to heat compared with vehicle injection (***P < 0.005). D: mean (±SE) AUC for withdrawal latencies to heat. AUC following vehicle is illustrated by the open triangle. ****P < 0.001, a difference from vehicle. E: histological verification of the location of all injection sites in the RVM. Schematics show cross sections of the medulla adapted from the stereotaxic atlas (Paxinos and Watson 2007) with permission. Coordinates below bregma are indicated on the left. Structures in the brain stem are indicated by arrows: py, pyramidal tract; 7n, facial nerve; RMg, raphe magnus nucleus; RPa, raphe pallidus nucleus; GiA, gigantocellular reticular nucleus pars alpha; LPGiA, lateralis paragigantocellular reticular nucleus pars alpha. The gray rectangles represent the area of the RVM according to our previous stereological studies (Khasabov and Simone 2013).

Withdrawal responses to mechanical stimuli.

Rats were placed under a clear plastic cage on an elevated plastic mesh floor and acclimated to the environment for 15 min before testing. Fifty-percent mechanical withdrawal thresholds were determined in previously used setup (Khasabov and Simone 2013). Mechanical stimulation was applied to the plantar surface of each hind paw using calibrated von Frey monofilaments. The 50% withdrawal response threshold (g) was determined for each paw using the up–down method (Chaplan et al. 1994) and averaged. Stimuli were each applied for durations of 1–2 s approximately every 10 s. Mechanical hyperalgesia was defined as a significant decrease in withdrawal response threshold.

Withdrawal responses to heat stimuli.

Radiant heat was applied to the plantar surface of each hind paw, and withdrawal response latencies were determined according to a method similar to that described previously setup (Khasabov and Simone 2013). Before testing, rats were placed under a clear plastic cage (23 × 13 × 13 cm) on a clear, 3-mm thick glass elevated to allow maneuvering of a controlled radiant heat source underneath. Animals were acclimated to this testing environment daily for 15 min before testing. Heat stimuli were produced by a 50-W light bulb placed in a custom case, which allowed focusing the light source (8-mm diameter) on the plantar surface of one hind paw. The intensity of the lamp was adjusted and maintained to produce stable withdrawal latencies ~10–15 s. Withdrawal latencies were measured to the nearest 0.1 s using a photo-cell that terminated the trial and a timer upon a withdrawal response. Each hind paw received four stimuli, alternating between each hind paw, with a minimum of 1 min between trials. Withdrawal latency for each hind paw was defined as the mean of the last three trials, and values for each paw were averaged. A 19-s cutoff was imposed on the stimulus duration to prevent tissue damage. Heat hyperalgesia was defined as a significant decrease in withdrawal response latency.

Electrophysiological Studies

Surgical preparation for spinal cord electrophysiology.

Rats were anesthetized with ketamine (100 mg/kg) and acepromazine (45 mg/kg), and a catheter was inserted into the external jugular vein for supplemental anesthesia with pentobarbital sodium (5–10 mg·kg⁻¹·h⁻¹). The trachea was cannulated to provide unobstructed ventilation or to provide artificial ventilation for experiments using antidromic stimulation, and end-tidal CO₂ was maintained within normal levels. The carotid artery was cannulated, and blood pressure was monitored continuously with a pressure transducer. Experiments were terminated if mean pressure dropped below 60 mmHg. Core body temperature was maintained at 37–38°C by a feedback-controlled heating pad. Areflexia was maintained by monitoring the corneal reflex or withdrawal responses to mild pinching applied to the hind paw at frequent intervals throughout the experiment. The lumbar enlargement was exposed by laminectomy and secured within a spinal frame. The spinal cord was continually bathed in a pool of warm mineral oil.

Electrophysiological recording from spinal neurons.

Extracellular recordings of single dorsal horn neurons with receptive fields (RFs) located on the plantar surface of the hind paw were obtained using stainless steel microelectrodes (Frederick Haer, Brunswick, ME). Recording electrodes were lowered into the spinal cord using an electronic micromanipulator (Kopf) in 5-µm steps. Recordings were made only from single neurons whose amplitude could be easily discriminated. Electrophysiological activity was amplified and displayed on an oscilloscope before being sent to a computer for data collection and offline analyses using Spike II (Cambridge Electronics Design) software that stored raw data, discriminated impulses, stimulus temperature, and time of application of mechanical stimuli.

Antidromic stimulation of identified SpPB neurons.

Cathodal current pulses (500 µA, 200 µs, 10 Hz) delivered through a stimulating electrode placed in the contralateral PbA served as search stimuli. The criteria used to identity a spinal neuron as being activated antidromically were as follows: 1) evoked responses occurred at a constant latency (≤0.2 ms variability); and 2) evoked responses were able to follow high-frequency (>333 Hz) stimulation; collision of the antidromic action potential with an orthodromic action potential (Lipski 1981).

To determine the area of termination for each axon, microantidromic mapping was done as previously described (Dado et al. 1994; Davidson et al. 2008; Zhang et al. 1995). Briefly, the stimulating electrode was moved rostrally throughout the PbA in a series of electrode penetrations across the mediolateral extent of the brain. Antidromic tracking for each axon was begun at a level near the rostral extent of the PbA. After multiple tracks were made across the most rostral level, the stimulating electrode was moved 0.5–1.0 mm caudally, and the procedure was repeated at the new level. Tracks were separated by 300–500 µm mediolaterally. Within each track, the electrode was lowered from the dorsal to the ventral surface and antidromic thresholds were determined at 200-µm intervals. A series of electrode tracks was made until a point was located at which the antidromic threshold was ≤30 µA (low-threshold point). Current pulses ≤30 µA have been shown to activate spinohypothalamic tract axons at a distance of ≤400 µm from the stimulating electrode (Burstein et al. 1991; Dado et al. 1994). To locate the apparent terminal location of the axons, we surrounded penetrations containing the most rostral low threshold point dorsally, ventrally, rostrally, medially, and laterally with penetrations in which the axon could not be activated antidromically with <500 µA. This is interpreted as evidence that axons of passage were not stimulated (Zhang et al. 1995).

Functional classification of dorsal horn neurons.

The RFs of isolated neurons were mapped with a suprathreshold von Frey monofilament. Each neuron was characterized based on its response to graded intensities of mechanical stimulation applied to the RF. Innocuous stimuli consisted of stroking the skin with a cotton swab. Noxious stimulation included mild pinching using calibrated arterial clips and mild pinch with forceps. Neurons were classed functionally according to responses evoked by mechanical stimuli as 1) low-threshold (LT) if they were excited maximally by innocuous stimulation, 2) wide dynamic range (WDR) if they responded in a graded fashion to increasing intensity of stimulation, and 3) high-threshold (HT) if responses were evoked by noxious stimulation only. Only WDR and HT neurons were studied.

Mechanical and heat stimulation.

Responses evoked by mechanical stimuli were determined using a calibrated von Frey monofilament (26 g) with a bending force of 255 mN. This was applied three times, each for 5 s every 20 s, to the most sensitive area of the RF and the mean number of evoked impulses was determined. Heat stimuli were applied using a feedback-controlled Peltier device (contact area of 1 cm²) that delivered a series of nine heat stimuli from 34 to 50°C (each with a duration of 5 s) in ascending steps of 2°C increments from a base temperature of 32°C. The rise/fall rate for all stimuli was 18°C/s. Stimuli were delivered with an interstimulus interval of 60 s. Mechanical and heat stimuli were applied in the same experiments and heat stimuli always followed mechanical stimulation.

Drug administration.

In electrophysiological experiments, 0.5 µl of SSP (5 nmol) or L-733,060 (1.5 pmol) were injected using 1-µl Hamilton microsyringe connected to a glass micropipette with ~50 µm tip diameter aimed at the following stereotaxic coordinates: AP = 2.3 mm; ML = 0 mm; and DV = −0.5 mm. To ensure that drugs were delivered into the RVM, FluoSpheres Polystyrene Microspheres (cat. no. F13081; Invitrogen, Eugene, OR) were added to solutions and injection sites were visualized histologically after each experiment. Rats that received injection outside of the RVM were excluded. To study the effects of NK-1R antagonist on the development of sensitization, capsaicin (dissolved in 5% Tween-80 and saline) was injected into the plantar hind paw with an insulin syringe and 28-gauge needle at a dose of 10 µg in 10 µl. Responses evoked by capsaicin were determined for 5 min after injection.

Immunohistochemical Studies of c-Fos Expression

Rats were deeply anesthetized with ketamine (67.5 mg/kg) and xylazine (22.5 mg/kg) injected intraperitoneally and perfused through the ascending aorta with 100 ml of phosphate-buffered saline (PBS) at room temperature followed by 800 ml of cold 4% paraformaldehyde in PBS (pH 7.4) for 40 min. Lumbar enlargements were removed, postfixed in the same fixative for 24 h at 4°C and transferred to 30% sucrose solution in PBS (4°C for 48 h). A freezing microtome was used to cut coronal sections (50 μm) of the lumbar segments. Free-floating sections were washed in PBS three times for 10 min, blocked for 1 h in 10% normal donkey serum (Jackson ImmunoResearch Laboratories, West Grove, PA) in PBS with 0.3% Triton X-100 at room temperature, and incubated overnight at 4°C in solutions containing two primary antibodies: 1) rabbit anti-c-Fos (cat. no. PC38-100UL, lot no. 4194-1-2; Oncogene Research Products, La Jolla, CA; AB Registry ID: AB 2106755) at 1:10,000; and 2) mouse anti-NeuN (cat. no. MAB377, lot no. LV1825845; Millipore, Billerica, MA; AB Registry ID: AB 2298772) at 1:500, dissolved in PBS with 1% normal donkey serum and 0.3% Triton X-100. NeuN antisera has been used for clear visualization of the dorsal horn. After application of the primary antibodies, sections were washed in PBS three times for 20 min and incubated for 2 h in two secondary antibodies: 1) biotinylated donkey anti-rabbit IgG (cat. no. 711-065-152, lot no. 100391; Jackson ImmunoResearch Laboratories) 1:500 and 2) Donkey anti-Mouse Alexa Fluor 488 (cat. no. A21202, lot no. 1113537 Invitrogen, Grand Island, NY). Finally, c-Fos protein was visualized by 1-h tissue incubation with Cy3-conjugated Streptavidin (cat. no. 016-160-084, lot no. 100687; Jackson ImmunoResearch Laboratories). After application of secondary antibodies, tissues were washed in PBS three times for 20 min and transferred on gelatinized glass slides. Tissues were dried at ~37°C for 2 h, dehydrated in graded alcohols (50, 70, 80, 90, 96, and 100%, 2 min in each), and cleared in xylene for 2 min. Slides were coverslipped using DPX (VWR, Radnor, PA). To confirm the specificity of the primary antibodies, controls included preincubation with corresponding synthetic peptides or omission of primary antibodies.

Quantification of c-Fos-expressing neurons was made by conventional microscopy using a Nikon E800 epifluorescence microscope equipped with filter sets designed to allow selective visualization of Cy3-conjugated streptavidin and donkey anti-mouse Alexa Fluor 488. For each section, c-Fos-positive neurons in the dorsal horn were quantified under a ×20 objective lens according to the brain atlas (Paxinos and Watson 2007). We compared the mean number of neurons per 50-µm coronal sections of lumbar spinal cord between the following groups of rats: 1) injected only with vehicle into the RVM; 2) injected only with SSP into the RVM; 3) injected with vehicle followed by noxious stimulation of the hind paw (strong pinch with a hemostat for 2 min); and 4) injected with SSP followed by the same noxious stimulation of the hind paw. Rats were anesthetized with isoflurane (2–2.5% with room air) during these experiments. In groups 1 and 2, the numbers of c-Fos protein-expressing neurons were averaged between dorsal horns. In groups 3 and 4, c-Fos-expressing neurons were quantified separately for dorsal horns that were ipsi- and contralateral to noxious stimulation.

Experimental Design

For behavioral studies of the effects of injection of SSP into the RVM on withdrawal responses, rats received an injection of either vehicle or 0.1–10 pmol SSP alone or SSP coadministered with the NK-1 receptor antagonist L-733,060 into or outside the RVM. Withdrawal response thresholds and withdrawal latencies to heat were determined before and at 5 min after injection and every 20 min for 80 min.

For studies of the role of NK-1 receptor activation on responses of dorsal horn neurons, vehicle or SSP (5 nmol) was injected into the RVM. Ongoing spontaneous activity (SA), obtained over a period of 2 min, and responses of dorsal horn neurons evoked by mechanical and heat stimuli were determined before and injection and at 15 min after injection of vehicle or SSP into the RVM.

To determine the role of NK-1 receptors in the RVM on sensitization of dorsal horn neurons following intraplantar injection of capsaicin, vehicle or the selective NK-1 receptor antagonist L-733,060 (1.5 pmol) was injected into the RVM before the injection of capsaicin. SA and responses evoked by mechanical or heat stimuli were determined before and at 15 min after injection of vehicle or L-733,060 into the RVM. Then, capsaicin (10 µg) was injected into the hind paw and ongoing activated was determine for 5 min after injection. Responses evoked by mechanical and heat stimuli were determined at 15 min after the injection of capsaicin.

At the end of each behavior and electrophysiological experiment, animals received an overdose of pentobarbital sodium and were perfused with normal saline followed by 10% formalin containing 1% potassium ferrocyanide. Serial transverse sections (50 µm) of the brainstem and the spinal cord were stained with neutral red. Injection sites in the RVM were identified by fluorescent microspheres added to solutions, injected into the RVM. Recording sites in the spinal cord and stimulation sites in the PbA were identified by Prussian blue marks or by small lesions.

Statistical Analyses

The effects of vehicle and SSP injected into the RVM on mean withdrawal responses to mechanical and heat stimuli were determined using two-way ANOVA with repeated measures. Post hoc comparisons were made using Bonferroni t-tests. The median effective dose (ED₅₀) was determined from the areas under the curve using GraphPad Prism 5 software (La Jolla, CA). Similarly, mean SA and evoked responses of dorsal horn neurons were determined using two-way ANOVA with repeated measures and Bonferroni t-tests. For all analyses, responses of dorsal horn neurons evoked by mechanical stimuli were determined by subtracting the number of spontaneous impulses that occurred for 5 s before each stimulus from the number of impulses evoked during the stimulus (5 s). Similarly, responses to heat stimuli were determined by subtracting the number of impulses that occurred during 20 s preceding each stimulus temperature from the number of impulses evoked for 20 s from the beginning of the stimulus.

The effects of L-733,060 and vehicle on responses evoked by mechanical and heat stimuli applied to each hind paw were determined using one- or two-way ANOVAs with repeated measures followed by Bonferroni t-tests. P < 0.05 was considered significant in all cases.

The mean numbers of c-Fos labeled cells in the dorsal horn after injection of vehicle or SSP into the RVM, with and without the pinch stimulus, were compared using one-way ANOVA. All data are expressed as mean (±SE) unless otherwise stated.

RESULTS

Behavioral Studies

Mechanical and heat hyperalgesia following activation of NK-1 receptors in the RVM.

Forty-six rats were used to determine the effects of NK-1 receptor activation in the RVM on withdrawal responses to mechanical and heat stimuli. Rats received a single injection of vehicle (n = 5) or SSP at doses of 0.1 (n = 6), 0.3 (n = 5), 0.5 (n = 6), 3 (n = 6), 5 (n = 9), and 10 nmol (n = 9) into the RVM. Changes in paw withdrawal threshold (g) and heat withdrawal latency (s) were determined before and 5 min after injection and at 20-min intervals thereafter (25, 45, 65, and 85 min after injection).

Injection of SSP into the RVM produced dose-dependent mechanical and heat hyperalgesia. Figure 1 shows the effects of vehicle and 0.5, 3, and 10 nmol SSP injected into the RVM on withdrawal responses to mechanical and heat stimuli. Only 3 and 10 nmol of SSP decreased withdrawal response thresholds compared with vehicle (Fig. 1A; two-way ANOVA with repeated measures, P < 0.05). Bonferroni t-tests indicated that withdrawal thresholds decreased at 5 min after injections of 3 (P < 0.05) and 10 (P < 0.005) nmol of SSP compared with baseline values before injection. Withdrawal thresholds returned to baseline values at 65 min after the dose of 3 nmol but remained below baseline measures for over 85 min after injection of 10 nmol. Injection of lower doses of SSP or vehicle did not alter withdrawal thresholds. We determined the area under the curve (AUC) during the entire testing time (85 min from the time of injection) to compare dose-response relationships and to determine the median effective dose (ED₅₀). As shown in Fig. 1B, the AUC was inversely related to the increase in pain sensitivity (decrease in withdrawal threshold), compared by one-way ANOVA. The mean AUC decreased after injections of 3 (P < 0.05), 5 (P < 0.001), and 10 (P < 0.001) nmol SSP as compared with vehicle. The ED₅₀ for SSP-evoked mechanical hyperalgesia was 2.75 ± 2.47 nmol.

Injection of 10 nmol SSP, but not 0.5 or 3 nmol, or vehicle significantly decreased withdrawal latencies to heat (two-way ANOVA with repeated measures, P < 0.005; Fig. 1C). A decrease in the mean AUC for withdrawal latency (Fig. 1D) was evident after injections of 3, 5, and 10 nmol SSP as compared with vehicle (one-way ANOVA, P < 0.001). The ED₅₀ for SSP that produced hyperalgesia to heat was 3.4 ± 2.21 nmol. The ED₅₀ doses for evoking mechanical and heat stimuli did not differ (P > 0.05), indicating that SSP-induced activation of NK-1 receptors in the RVM has similar potency for producing mechanical and heat hyperalgesia.

Figure 1E shows histological reconstructions of SSP and vehicle injection sites in the RVM. All injections were located in the RVM and there were no apparent differences in the locations of the injection sites for the various treatments.

Hyperalgesia produced by SSP was due to activation of NK-1 receptors in the RVM.

Coadministration of SSP with the NK-1 receptor antagonist L-733,060 into the RVM was used to determine whether hyperalgesia produced by SSP occurred through NK-1 receptors. The coadministration of 1.5 pmol of L-733,060 and 5 nmol SSP (n = 6) into the RVM blocked the development of mechanical (Fig. 2A) and heat (Fig. 2B) hyperalgesia. Importantly, NK-1 receptor-dependent facilitation was site specific and neither mechanical nor heat hyperalgesia developed after injection of 5 nmol SSP at least 0.5 mm outside the RVM (Fig. 2, C and D, respectively, n = 5). Together, these data indicate that hyperalgesia produced by injection of SSP was due to activation of NK-1 receptors located within the RVM.

Fig. 2.

Hyperalgesia produced by injection of SSP was mediated by neurokinin-1 (NK-1) receptors in the RVM. Hyperalgesia to mechanical (A) and to heat stimuli (B) were completely blocked by coadministration of the specific NK-1 receptor antagonist L-733,060 with SSP. SSP injected alone outside the RVM (open squares) had no effect on mechanical withdrawal threshold (C) or heat withdrawal latency (D), demonstrating site specificity of the hyperalgesia produced by SSP. E: histological verification of injection sites inside (filled circles) and outside (open squares) the RVM. Details of schematic representations (from Paxinos and Watson 2007 with permission) and abbreviations are as described for Fig. 1.

Electrophysiology Studies

General characteristics of spinal dorsal horn neurons.

A total of 35 nociceptive dorsal horn neurons with RFs on the plantar hind paw were studied. Neurons were classed functionally as WDR (n = 24) or HT (n = 11) according to their responses evoked by graded mechanical stimulation of their RF (see materials and methods). The effects of injection of vehicle into the RVM on SA and evoked responses were determined for 17 neurons: 13 WDR (76.5%) and 4 HT (23.5%) cells. Similarly, effects of injection of SSP was studied in 18 neurons: 11 WDR (61.1%) and 7 HT (38.9%) neurons. There were no differences in the proportion of WDR and HT neurons that received vehicle or SSP into the RVM (Fisher exact test, P = 1.00).

Among the 35 neurons, 19 (54%) were responsive to both mechanical and heat stimuli, while the remaining 16 neurons (46%) were responsive only to mechanical stimuli. Importantly, SA, mechanical response thresholds, and responses evoked by suprathreshold mechanical stimuli did not differ between neurons responsive to mechanical and heat stimuli and those responsive to mechanical stimuli only. Also, there were no differences in the proportions of neurons sensitive to mechanical and heat stimuli and those only responsive to mechanical stimuli that were studied in vehicle-treated (53 and 47%, respectively) and SSP-treated (56 and 44%, respectively) groups (Fisher’s exact test, P = 1.00). The localization of injection sites for vehicle and SSP in the RVM and the localization of recording sites in the spinal cord for neurons excited by mechanical and heat stimuli or only by mechanical stimuli did not differ (Fig. 3).

Fig. 3.

Histological reconstruction of injection sites in the RVM (top) and for recording sites in the lumbar dorsal horn (bottom). There were no apparent differences in the location of injection sites in the RVM for vehicle (left) and for SSP (right). Similarly, the distributions of recording sites in the dorsal horn were similar for neurons studied before and after injection of vehicle or SSP into the RVM. HT, high threshold; WDR, wide dynamic range. Schematic representations of brain stems and spinal cords from Paxinos and Watson (2007) with permission.

Activation of NK-1 receptors in the RVM facilitated responses of nociceptive dorsal horn neurons.

mechanical response thresholds.

Injection of SSP, but not vehicle, into the RVM decreased response thresholds of WDR neurons. Mean mechanical thresholds were unchanged after injection of vehicle into the RVM (n = 13; 1.7 ± 0.46 g before and after injection, Student's t-test, P = 1.000). In contrast, injection of SSP into the RVM (n = 11) decreased response threshold from 2.1 ± 0.42 to 0.7 ± 0.23 g (Student's t-test, P < 0.001). Interestingly, mean response thresholds of HT neurons were not altered after injection of vehicle (n = 4; 36.3 ± 13.75 g before and 35.0 ± 14.43 g after, Student's t-test, P = 0.952) or SSP (n = 7; 31.7 ± 11.78 before and 42.0 ± 23.18 g after injection, Student's t-test, P = 0.699) (Fig. 4A).

Fig. 4.

SSP injected into the RVM increased responses of WDR neurons to mechanical stimuli. A: injection of SSP, but not vehicle, into the RVM decreased mean (±SE) response threshold for WDR neurons (**P < 0.001) but had no effect on response thresholds of HT neurons. B: representative examples of responses of two WDR neurons evoked by application of a von Frey monofilament (26 g for 5 s) to the receptive field (RF) before and 15 min after injection of vehicle (top) or SSP (bottom) into the RVM. Times of stimulation are indicated by the horizontal bar; numbers of evoked action potentials are indicated in parenthesis. C: mean (±SE) number of impulses evoked by the von Frey monofilament (26 g) before and at 15 min after injection of vehicle or SSP into the RVM for all examined WDR and HT neurons. Responses were increased in WDR but not HT neurons (*P < 0.05).

responses to suprathreshold mechanical stimuli.

Figure 4B shows representative examples of mechanically evoked responses of WDR neurons before and after injection of vehicle or SSP into the RVM. Whereas injection of vehicle into the RVM did not alter the number of impulses evoked by application of the 26-g monofilament (Fig. 4B, top), evoked responses increased after injection of SSP (Fig. 4B, bottom). Figure 4C shows that the mean number of impulses evoked by 26-g von Frey monofilament for WDR neurons (n = 13) was not altered following injection of vehicle into the RVM (39.2 ± 9.68 and 38.3 ± 8.77 impulses, respectively, Student's t-test, P = 0.945) but increased after SSP (n = 11) by 92.3 ± 16.70% (from 37.8 ± 6.57 impulses before injection to 67.7 ± 11.31 impulses after; Student's t-test, P < 0.05).

Unlike the WDR neurons, evoked responses of HT neurons were not altered following injection of either vehicle (42.2 ± 15.19 impulses before and 40.5 ± 14.66 impulses after, Student's paired t-test, P = 0.939, n = 4) or SSP injected into the RVM (28.9 ± 7.33 impulses before and 27.7 ± 15.72 impulses after, Student's paired t-test, P = 0.949, n = 7). These data demonstrate that descending NK-1 receptor-expressing projections from the RVM increase mechanically-evoked responses of WDR, but not HT, neurons.

responses evoked by heat stimuli.

Heat stimuli were delivered in an ascending series of stimulus intensities so that changes in the stimulus-response function (threshold and suprathreshold evoked responses) could be evaluated. Responses to heat for WDR (n = 13) and HT (n = 6) neurons before any injection did not differ. Figure 5A shows representative examples of responses of two WDR neurons to heat stimuli before and after RVM injections. Injection of vehicle into the RVM did not alter responses whereas SSP increased responses to a wide range of thermal stimuli (from 40°C to 50°C). As is shown in Fig. 5C, the mean heat response thresholds of WDR neurons did not change after injection of vehicle (n = 9) and were 45.8 ± 1.02 and 45.3 ± 1.05°C, respectively (P = 0.65). One HT neuron tested to heat stimuli had a threshold of 44°C before and after vehicle injection into the RVM. However, as was the case for responses evoked by mechanical stimulation of the RF, injection of SSP into the RVM decreased heat response threshold of WDR neurons (from 44.7 ± 0.99 to 40.0 ± 1.46°C, P < 0.05, n = 6) but not HT cells (45.5 ± 2.63°C before and 44.5 ± 0.99°C after injection, P = 0.311, n = 4). Two-way ANOVA with repeated measures was used for comparison between groups injected with vehicle or SSP. Figure 5D shows that vehicle did not alter responses of WDR neurons (n = 9) to suprathreshold stimuli (two-way ANOVA with repeated measures). A similar result was obtained for the HT neuron. In contrast, injection of SSP into the RVM caused a leftward shift in the stimulus-response function for heat in WDR (n = 6) but not HT (n = 4) cells (two-way ANOVA with repeated measures, P < 0.05 and P = 0.16, respectively). The increase in responses to thermal stimuli also did not depend on the location of neurons in the dorsal horn. Importantly, the finding that no changes in responses to heat stimuli occurred after injection of vehicle demonstrates that the sequence of heat stimuli used did not cause sensitization to subsequent heat stimuli.

Fig. 5.

Injection of SSP into the RVM increased responses of spinal WDR neurons to heat stimuli. Representative examples of responses of single WDR neurons to heat stimuli of 40, 46, and 50°C (5 s) applied on the RF before and 20 min after injection of vehicle (A) or SSP (B) into the RVM. C: injection of vehicle into the RVM did not alter mean (±SE) heat response threshold of WDR cells (n = 9) and 1 HT neuron, whereas mean threshold was lower in WDR neurons (n = 6) after injection of SSP (P < 0.05) but not HT neurons (n = 4). D: mean numbers of impulses evoked by a series of heat stimuli were not altered in WDR or HT cells following vehicle. After injection of SSP into the RVM, responses were enhanced in WDR (n = 6) but not HT (n = 4) neurons. *P < 0.05, **P < 0.01, significant differences from responses obtained before injection.

spontaneous activity.

Spontaneous activity was determined over a 2-min period just before injection into the RVM. Injection of vehicle or SSP into the RVM did not alter SA of either WDR or HT neurons. Thus the mean rate of SA for WDR neurons was 1.6 ± 0.42 impulses/s before injection of vehicle into the RVM and 1.9 ± 0.42 impulses/s after (two-way ANOVA with repeated measures, P = 1.40, n = 13).The rates of SA before and after injection of SSP were 2.4 ± 0.85 and 4.0 ± 0.85 impulses/s, respectively (two-way ANOVA with repeated measures, P = 0.53, n = 11). Similar results were found for HT neurons. Mean rates of SA for HT neurons before and after injection of vehicle were 2.3 ± 0.76 and 3.1 ± 0.75 impulses/s, respectively (two-way ANOVA with repeated measures, P = 0.171, n = 4) and were 2.1 ± 1.01 impulses/s before and 2.8 ± 1.47 impulses/s after injection of SSP (two-way ANOVA with repeated measures, P = 0.525, n = 7).

Inhibition of NK-1 receptors in the RVM prevented the development of central sensitization.

We investigated whether NK-1 receptors in the RVM contribute to central sensitization produced by intraplantar injection of capsaicin, a model of central sensitization and hyperalgesia (Johanek and Simone 2005; Khasabov et al. 2002; Simone et al. 1989a, 1991), because our earlier studies showed that capsaicin-evoked hyperalgesia was attenuated following injection of the selective NK-1 receptor antagonist L-733,060 into the RVM (Pacharinsak et al. 2008). Only WDR neurons sensitive to both mechanical and heat stimuli were studied since our results above indicate that activation of NK-1 receptors in the RVM facilitated responses of WDR neurons only and earlier studies showed that capsaicin produced a more robust sensitization in WDR as compared with HT neurons (Simone et al. 1991). Responses evoked by mechanical and heat stimuli were determined before and at 15 min after pretreatment with injection of vehicle (n = 8) or L-733,060 (1.5 pmol in 0.5 µl; n = 8) into the RVM and then at 15 min after intraplantar injection of 0.1% capsaicin.

responses evoked by capsaicin.

Figure 6, A and B, shows examples of responses in two WDR neurons. Following pretreatment with vehicle into the RVM, intraplantar injection of capsaicin evoked a vigorous discharge that lasted for more than 2 min and this was greatly reduced in rats pretreated with L-733,060. Whereas the mean number of action potentials per 15-s bin was increased for the entire 2-min period after capsaicin in rats pretreated with vehicle into the RVM, responses evoked by capsaicin occurred only during the first 15 s after injection in rats pretreated with L-733,060 (two-way ANOVA with repeated measures and post hoc Bonferroni t-tests, P < 0.05; Fig. 6C). The cumulative mean number of impulses evoked during the first 2 min after capsaicin injection was 2598.9 ± 627 in the vehicle-pretreated group and 794.1 ± 202.7 impulses in the group pretreated with L-733,060, a 3.3-fold decrease in the response to capsaicin.

Fig. 6.

Injection of the NK-1 antagonist L-733,060 into the RVM decreased responses of WDR neurons evoked by intraplantar injection of capsaicin. A: representative examples of responses of WDR neurons evoked by capsaicin for 2.5 min after injection of vehicle (top) or SSP (bottom) into the RVM. B: mean (±SE) number of impulses per 15-s bin before and up to 300 s following injection of capsaicin. Whereas capsaicin caused a robust response for 2 min following pretreatment with vehicle (n = 8) into the RVM, capsaicin-evoked responses were decreased following pretreatment with L-733,060 (n = 8). The arrow indicates the injection of capsaicin. *P < 0.05, significant difference from vehicle. #P < 0.05, ##P < 0.01, ###P < 0.005, significant difference from number of action potential per bin before capsaicin injection in vehicle pretreated animals. $P < 0.05, significant difference from number of action potential per bin before capsaicin injection in L-733,060-pretreated animals.

sensitization to mechanical stimuli.

Pretreatment with vehicle or L-733,060 into the RVM did not alter responses of WDR neurons evoked by the 26-g von Frey monofilament (applied to the RF for 5 s). Intraplantar injection of capsaicin increased mechanically-evoked responses which were blocked by pretreatment with L-733,060, but not vehicle, into the RVM (Fig. 7, A and B). Mean mechanical response thresholds were 1.7 ± 0.34 g before and 0.6 ± 0.11 g after intraplantar injection of capsaicin (one-way ANOVA, P < 0.05) in rats pretreated with vehicle into the RVM. In contrast, mechanical response thresholds did not change after capsaicin in rats pretreated with L-733,060 into the RVM (1.4 ± 0.16 g before and 1.5 ± 0.26 g after injection of capsaicin (one-way ANOVA, P = 0.812; Fig. 7C). Similarly, Fig. 7D shows that the increase in responses evoked by suprathreshold 26-g monofilament following capsaicin was also blocked by pretreatment with injection of L-733,060 into the RVM. The mean number of impulses evoked by the 26-g von Frey monofilament in vehicle-treated group was 42.5 ± 11.79 before and 103.5 ± 19.29 after intraplantar injection of capsaicin (one-way ANOVA, P < 0.05). The increased response following capsaicin was blocked by pretreatment with L-733,060 into the RVM (41.2 ± 12.4 before and 59.8 ± 28.0 impulses after capsaicin; one-way ANOVA, P = 0.779).

Fig. 7.

Injection of L-733,060 into the RVM blocked the sensitization of WDR neurons to mechanical stimuli produced by capsaicin. Representative examples of responses of 2 WDR neurons evoked by a von Frey monofilament (26 g for 5 s) applied to the RF before any injection, 15 min after injection of vehicle (A) or L-733,060 (B) into the RVM, and 15 min after intraplantar injection of capsaicin. Numbers of evoked action potentials are indicated in parenthesis. C: intraplantar injection of capsaicin decreased mean (±SE) mechanical response thresholds of WDR neurons following pretreatment with vehicle (n = 8) into the RVM (one-way ANOVA, P < 0.05) and this was blocked by pretreatment with L-733,060 (n = 8). D: capsaicin increased the mean (±SE) number of impulses evoked by the 26 g von Frey monofilament in rats pre-treated with vehicle in the RVM (one-way ANOVA, P < 0.05) but not following L-733,060 into the RVM. Importantly, injection of vehicle or L-733,060 into the RVM alone did not alter mechanical response thresholds or responses evoked by mechanical stimuli. *P < 0.05, significant difference from responses before vehicle. #P < 0.05, significant difference from responses after vehicle.

sensitization to heat stimuli.

Pretreatment with L-733,060 into the RVM also blocked sensitization to heat produced by intraplantar injection of capsaicin (representative examples are shown in Fig. 8, A and B). Injection of vehicle or L-733,060 into the RVM did not alter heat response thresholds. Whereas mean heat response thresholds decreased from 42.0 ± 0.9°C before capsaicin to 35.5 ± 0.7°C following pretreatment with vehicle into the RVM (one-way ANOVA with post hoc Bonferroni t-tests, P < 0.001; Fig. 8C), thresholds were unchanged after capsaicin following pretreatment with L-733,060 into the RVM (40.8 ± 1.5°C before and 40.4 ± 2.3 after capsaicin; one-way ANOVA with post hoc Bonferroni t-tests, P = 0.42; Fig. 8D). Following pretreatment with vehicle into the RVM, intraplantar injection of capsaicin increased responses to heat over entire range of stimulus intensities, causing a leftward shift in the stimulus-evoked response function (two-way ANOVA with repeated measures and post hoc Bonferroni t-tests, P < 0.05 and P < 0.005; Fig. 8E). Importantly, this increase in capsaicin-evoked responses to heat was blocked and responses to heat were significantly decreased following pretreatment with L-733,060 injected into the RVM (two-way ANOVA with repeated measures and post hoc Bonferroni t-tests, P < 0.05, Fig. 8F). These results suggest that blockade of descending facilitation via NK-1 receptors in the RVM reveals descending inhibition.

Fig. 8.

Injection of L-733,060 into the RVM blocked the sensitization of WSR neurons to heat stimuli produced by capsaicin. Representative examples of responses of WDR neurons evoked by heat stimuli of 40, 46, and 50°C (5 s) applied to the RF before any injection, at 20 min after injection of vehicle (A) or L-733,060 (B) into the RVM, and at 20 min after intraplantar injection of capsaicin. Responses to heat stimuli increased after intraplantar injection of capsaicin following pretreatment with vehicle (n = 8) into the RVM but not following pretreatment with L-733,060 (n = 8) into the RVM. C: mean (±SE) heat response threshold of dorsal horn neurons decreased after intraplantar injection following pretreatment in the RVM with vehicle but not L-733,060 (one-way ANOVA). **P < 0.01, significant difference from the response threshold before vehicle. ##P < 0.01, significant difference from the responses threshold after vehicle. The mean (±SE) numbers of impulses evoked by stimuli of 34–50°C were not altered after injection of either vehicle (E) or L-733,060 (F) alone into the RVM. Following injection of vehicle into the RVM, capsaicin caused a leftward shift in the stimulus-response function for heat (two-way ANOVA with repeated measures, P < 0.05 compared with responses before RVM injection and P < 0.005, compared with responses after vehicle injection). Responses to heat after capsaicin deceased following pretreatment with L-733,060 into the RVM (two-way ANOVA with repeated measures, P < 0.05, compared with responses before and after L-733,060). *P < 0.05, **P < 0.005, significant difference from responses following injection into the RVM.

Sensitization of identified SPbT neurons following injection of SSP into the RVM.

Because neurons in the spinal cord that project to the PBA may drive and maintain descending facilitation of nociceptive transmission, we determined whether these neurons became sensitized following injection of SSP into the RVM. Recordings were made from four neurons that were identified by antidromic activation from the lateral parabrachial nucleus. Identification criteria and representative examples of responses before and after RVM injection of vehicle followed by SSP are shown in Fig. 9. Evoked responses of SPbT neurons to mechanical and heat stimuli were determined before and after injection of vehicle (n = 1) and/or 5 nmol SSP (n = 4) into the RVM. Responses to mechanical and heat stimuli were increased following administration of SSP into the RVM (Fig. 9H). Figure 10 shows that SPbT neurons became sensitized after intraplantar injection of capsaicin. Importantly, as we demonstrated in unidentified spinal neurons, sensitization of SPBT neurons following intradermal injection of capsaicin was blocked and responses were decreased following 1.5 pmol L-733,060 injected into the RVM before the capsaicin injection (n = 2). These data demonstrate that activation of NK-1 receptors in the RVM facilitates responses of SPbT neurons, thereby enhancing ascending nociceptive transmission.

Fig. 9.

Sensitization of an identified spinoparabrachial tract (SPbT) neuron following injection of SSP into the RVM. A and B: recording site in the superficial dorsal horn (A) and location of its RF (shaded area) (B). C: the axon terminated in the right lateral parabrachial nucleus (LPbN). (dashed lines). Vertical lines show antidromic mapping tracks in dorso-ventral directions with locations of electrical stimuli up to 500 µA (200-µs pulses) indicated by the horizontal lines. Big black dots indicate sites of antidromic activation for this cell from 101 to 500 µA and small black dots are from 31 to 100 µA . Low threshold points (<30 µA) for activation indicate sites of axon termination and are shown by small red dots. Activation thresholds (µA) are indicated on the left. Prussian blue mark shows the location of the lowest antidromic threshold point. D: antidromic activation of this cell occurred rostrally only using high amplitude pulses, but at the same latency as that seen at the lowest threshold point, indicating that the axon did not project rostral from the LPbN. Distances from the bregma are indicated on the lower left corners of tissues. Distances from the midline, above tissues (C and D); distances from the surface of the cortex, on the right from the vertical axes (D). Criteria for identification of SPbT neurons: E: antidromic responses evoked by stimulation at the lowest-threshold point had a stable latency (5.2 ms) shown in 3 overlapping traces. The circle indicates electrical stimulus artifacts, arrowheads indicate action potentials. F: 3 antidromic action potentials followed high-frequency (333 imps/s) stimuli (1) and collided with an orthodromic action potential that occurred just before electrical stimulation (2). *Time at which the antidromic spike would have occurred. Other symbols are as in E. G: location of vehicle and SSP injections indicated by fluorescent microspheres (arrow). The RVM is indicated by dashed rectangular. Py, pyramidal tract. All coordinates are according to Paxinos and Watson (2007). H and I: responses of this neuron evoked by heat (H) and mechanical (I) stimuli applied to the RF before and after injection of vehicle and SSP into the RVM. Responses were not altered after injection of vehicle but increased at 15 min after injection of SSP and returned to near baseline values at 60 min. Responses are illustrated as histograms (1 s/bin). Heat stimuli (5 s each) are indicated above the histograms (H) and mechanical stimuli (10 s each) are indicated below the histograms (I).

Fig. 10.

Administration of L-733,060 into the RVM blocked capsaicin-evoked sensitization in identified SPbT neurons. Representative examples of responses are shown for 2 neurons. Responses evoked by heat (A) and mechanical (B) stimuli before any injection (left), 15 min after injection of vehicle into the RVM (middle), and 15 min after intraplantar injection of capsaicin (right). Vehicle into the RVM did not alter responses to heat (44–50°C; 5 s each) or mechanical (4 applications of 26 g, 5 s each) applied to the RF, but responses to heat and mechanical stimuli increased 15 min after capsaicin. Values above the histograms indicate the mean (±SE) number of impulses evoked by the 26 g stimulus. C and D: are the same format as in A and B except that L-733,060, rather than vehicle, was injected into the RVM before capsaicin. For this cell, pretreatment with NK-1 receptor antagonist L-733,060 into the RVM did not alter responses to heat or mechanical stimuli, but capsaicin injections reduced responses of the neurons compared with those before any injections. Data are illustrated as histograms with 1-s bins. *P < 0.05, significant difference from responses before any injection.

Activation of NK-1 receptors in the RVM increased stimulus-evoked c-Fos expression in the spinal cord.

The expression of c-Fos protein is an indirect marker of neuronal activation (VanElzakker et al. 2008) including that produced by noxious stimulation (Bullitt 1990; Hunt et al. 1987) and was used to gain insight into the spatial distribution of dorsal horn neurons whose activity is facilitated following the activation of NK-1 receptors in the RVM. The number of c-Fos-positive neurons in the dorsal horn was counted and compared between rats injected with vehicle into the RVM, SSP into the RVM, and those given vehicle or SSP into the RVM followed 15 min later by pinching the hind paw for 2 min. The mean number of c-Fos-positive neurons in each condition is provided in Fig. 11A. c-Fos expression was increased only on the side ipsilateral to noxious stimulation and its expression was greater following pretreatment with SSP injected into the RVM then after pretreatment with vehicle. Thus the mean of number of c-Fos-labeled neurons was not different at 105 min following RVM injection of vehicle or SSP without noxious stimulation (1.5 ± 0.27 and 1.6 ± 0.35 neurons/50 µm, respectively). Pinching the paw for 2 min beginning 15 min after injection of vehicle into the RVM increased c-Fos expression, and this was greater in rats pretreated with SSP. Following vehicle or SSP pretreatment into the RVM, pinch resulted in c-Fos expression in 9.3 ± 1.33 neurons/50 µm (P < 0.05) and 20.2 ± 3.77 neurons/50 µm (P < 0.001), respectively. The mean numbers of c-Fos-positive neurons in contralateral dorsal horns were significantly lower (P < 0.001) and did not differ between the groups (one-way ANOVA followed by Bonferroni t-tests, four rats per treatment group). Therefore, activation of NK-1 receptors in the RVM enhances rather than induces c-Fos expression in the dorsal horn (Fig. 11B), indicating the importance of descending NK-1 receptor-dependent facilitation for the maintenance of hyperalgesia.

Fig. 11.

SSP injected into the RVM increased c-Fos expression in dorsal horn neuron evoked by noxious stimulation. A: mean (± SE) numbers of c-Fos-ir neurons in the entire dorsal horn per 50 µm coronal sections. Comparisons were performed by one-way ANOVA. *P < 0.05, **P < 0.001, significant difference between groups of rats injected into the RVM with and without noxious pinch stimulation. $P < 0.01, significant difference in pinch-evoked c-Fos expression in the ipsilateral dorsal horn in rats pretreated with vehicle or SSP into the RVM. #P < 0.05, ##P < 0.001, significant difference between ipsi- and contralateral dorsal horns in groups pretreated with vehicle or SSP (n = 4/group). Numbers indicate representative examples of c-Fos (red) and NeuN (green) expression in coronal sections of related dorsal hors. B: examples of expression of c-Fos (red) and NeuN (green) in confocal images of coronal sections (8 µm) of dorsal horns in treated groups indicated by numbers in A. Images were taken with a 10 × objective. Insets a and b marked by dashed rectangles are single optical confocal sections (0.5 µm) of dorsal horn taken with a ×40 oil emersion objective. Open arrowheads indicate c-Fos-positive neurons with visualized NeuN-ir nuclei.

DISCUSSION

We examined the role of NK-1 receptors in the RVM on the development of hyperalgesia and on nociceptive transmission in the spinal cord. It was found that 1) activation of NK-1 receptors in the RVM produced robust mechanical and heat hyperalgesia and sensitization of nociceptive dorsal horn neurons, including those that project to the PBA; 2) blockade of NK-1 receptors in the RVM prevented sensitization of dorsal horn neurons produced by intraplantar injection of capsaicin; and 3) activation of NK-1 receptors in the RVM enhances the expression of c-Fos protein in the spinal cord evoked by noxious stimulation. Our results are consistent with previous studies indicating a role for NK-1 receptors in the RVM in the descending facilitation of pain (Brink et al. 2012; Hahm et al. 2011; Hamity et al. 2010; Khasabov and Simone 2013; Lagraize et al. 2010; Maduka et al. 2016; Pacharinsak et al. 2008; Zhang and Hammond 2009) and provide new information regarding characteristics and mechanisms underlying descending facilitation of pain from activation of neurons in the RVM that possess NK-1 receptors.

Role of NK-1 Receptor-Expressing Neurons in the RVM to Hyperalgesia

Activation of NK-1 receptor-expressing neurons in the brainstem dose dependently produced heat hyperalgesia in the hind paw, consistent with earlier reports (Hamity et al. 2010; Lagraize et al. 2010). We also found that injection of SSP into the RVM produced robust hyperalgesia to mechanical stimuli. The similarity in ED₅₀ values for evoking mechanical and heat hyperalgesia suggests that NK-1 receptor-dependent descending facilitation affects both pain modalities similarly. Consistent with previous studies (Hamity et al. 2010; Pacharinsak et al. 2008), we showed that hyperalgesia produced by injection of SSP into the RVM occurred via activation of NK-1 receptors in the RVM since hyperalgesia was prevented by pretreatment with a NK-1 receptor antagonist delivered into, but not adjacent to, the RVM (Pacharinsak et al. 2008). We suggest that NK-1 receptor-expressing neurons in the RVM, which are presumably located on ON cells (Budai et al. 2007), are part of descending pathways incorporated into a proposed pronociceptive spino-bulbo-spinal loop that promotes facilitation of nociceptive transmission in the spinal cord (Khasabov et al. 2002, 2005; Rahman et al. 2007; Rygh et al. 2006; Suzuki et al. 2002).

Contribution NK-1 Receptors in the RVM to Sensitization of Dorsal Horn Neurons

Central sensitization is a fundamental process that underlies the development of hyperalgesia (Cervero 2009; Woolf and Salter 2000). Our electrophysiological studies showed that activation of NK-1 receptors in the RVM produced sensitization of WDR, but not HT, neurons. Importantly, there was no effect on responses evoked by mechanical or heat stimuli after injection of vehicle into the RVM, indicating that repeated application of the stimuli used did not alter subsequent evoked responses. The physiological significance of selective sensitization of WDR neurons following activation of NK-1 receptors in the RVM is unclear since specific roles of WDR and HT neurons in nociceptive processing have been debated (Craig 2003; Price 2000). It was suggested that HT cells play a greater role in the localization of nociceptive stimuli, while WDR neurons are responsible for precise encoding of stimulus intensity (Coghill et al. 1993a; Hoffman et al. 1981; Price et al. 1978). Moreover, increases in the activity of WDR neurons following sensitization correlated with higher psychophysical ratings for pain (Coghill et al. 1993b; Simone et al. 1991; Suzuki and Dickenson 2006; Urch et al. 2003). Thus we propose that activation of NK-1 receptor-expressing neurons in the RVM produces mechanical and heat hyperalgesia by facilitating evoked responses of WDR neurons only. Earlier studies suggest that ON cells are the origin of descending NK-1 receptor-dependent facilitation and are themselves sensitized after intraplantar injection of capsaicin, inflammation or nerve injury (Khasabov and Simone 2013; Brink et al. 2012; Khasabov et al. 2012; Edelmayer et al. 2009; Carlson et al. 2007).

SPbT neurons are the main nociceptive ascending output from the lumbosacral dorsal horn in rats (Al-Khater and Todd 2009; Polgár et al. 2010; Williams and Ivanusic 2008), and we showed that activation of NK-1 receptors in the RVM sensitized identified SPbT neurons. Neurons in the lateral PBA project directly to the RVM (Roeder et al. 2016), activating ON and inhibiting OFF cells (Chen et al. 2017). Although projections from PBA to the RVM could contribute to activation a spino-bulbo-spinal circuit that maintains central sensitization (Dickenson et al. 2005; Suzuki et al. 2004a, 2004b) it is unknown whether this pathway is involved in NK-1 receptor-dependent modulation of RVM, since SP expression has not been demonstrated in this pathway. Known sources of SPergic projections to the RVM are the periaqueductal gray, the dorsal raphe nucleus, and the nucleus cuneiformis (Beitz 1982; Chen et al. 2013), which are directly driven by collaterals of ascending lumbar SPbT neurons (Al-Khater et al. 2008; Polgar et al. 2010). In addition, the periaqueductal gray receives direct projections from the PBA (Bianchi et al. 1998; Hayward et al. 2004). We suggest that these indirect connections between SPbT neurons and the RVM drive NK-1 receptor-dependent modulation.

Interestingly, recent studies showed that SPbT neurons also encode pruriceptive stimuli (Jansen and Giesler 2015), suggesting that NK-1 receptor-expressing neurons in the RVM may be involved in the modulation of itch.

Importantly, administration of the NK-1 receptor antagonist L-733,060 alone into the RVM and in the absence of injury did not alter spontaneous or evoked activity of dorsal horn neurons but rather prevented their sensitization. These data are consistent with earlier behavior studies indicating that descending NK-1 receptor-expressing pathways do not exert a tonic influence on nociception (Hamity et al. 2010; Khasabov and Simone 2013; Lagraize et al. 2010; Pacharinsak et al. 2008). Earlier studies showed that blockade of NK-1 receptors in the RVM did not alter response properties of ON cells under normal conditions (Brink et al. 2012; Budai et al. 2007; Khasabov et al. 2012). Together these findings indicate that SP in the RVM is not tonically active but requires noxious stimulation.

Surprisingly, blockade of NK-1 receptors in the RVM not only prevented sensitization of dorsal horn neurons produced by intraplantar injection of capsaicin, including SPbT neurons, but also reduced their responses to mechanical and noxious heat stimuli (Figs. 8 and 10) following injection of capsaicin. This inhibition might reflect unmasking of descending antinociception from the RVM, which is enhanced by noxious stimulation (Kauppila et al. 1998; Kauppila 1997; Ren and Dubner 1996) and originates from neuronal groups distinctive from NK-1 receptor-expressing cells, which appear to be pronociceptive. We showed (Khasabov and Simone 2013) that deletion of NK-1-expressing neurons in the RVM did not alter morphine analgesia which is believed to occur by activation of mu opioid receptors located on OFF cells (Fields 2004) . It was shown that during the development of hyperalgesia descending facilitation prevails over descending antinociception (Guan et al. 2002; Terayama et al. 2000; Zhuo et al. 2002), which may have been unmasked in our experiments by blockade of NK-1 receptors.

Spatial Distribution of Dorsal Horn Neurons That Are Sensitized Following Activation of NK-1 Receptors in the RVM

Central sensitization results from neuronal plasticity due to enhanced intracellular signaling (Brenner et al. 2004; Ji et al. 2003; Ji and Woolf 2001; Kawasaki et al. 2004) with downstream phosphorylation of cytoplasmic mitogen-activated protein kinases (Ji et al. 2003, 2009; Ji and Woolf 2001) and is reflected by the neuronal expression of c-Fos protein (Giorgi et al. 1997; Kerr et al. 1999; Leah et al. 1996; Lima and Avelino 1994). Using immunohistochemistry, we found that activation of NK-1 receptors in the RVM did not evoke c-Fos expression in dorsal horn neurons but increased its expression throughout the ipsilateral dorsal horn after noxious stimulation. This pattern of expression was consistent with our electrophysiological experiments showing that neurons located throughout the dorsal horn became sensitized after injection of SSP into the RVM. This is also consistent with our findings that activation of NK-1 receptors in the RVM did not increase spontaneous activity of dorsal horn neurons, indicating that activation of this descending system does not produce ongoing pain.

Conclusions

Our results demonstrate that the main function of descending NK-1 receptor-dependent facilitation is to increase the excitability of dorsal horn neurons and support the suggestion that the RVM plays a modulating, rather than a permissive, role in the development of central sensitization (Foo and Mason 2003; Jinks et al. 2004; Khasabov and Simone 2013). Although we focused on NK-1 receptors, other mediators in the RVM also contribute to pain facilitation, including cholecystokinin (Marshall et al. 2012; Wang et al. 2013; Zhang et al. 2009), brain-derived neurotrophic factor (Guo et al. 2006; Yin et al. 2014), tumor necrosis factor-α (Wei et al. 2008), and serotonin (Gu et al. 2011; Rahman et al. 2004; Suzuki et al. 2002, 2004a, 2005). A better understanding of interactions between NK-1 receptor-expressing neurons with other neurons in the RVM may lead to novel approaches to disrupt circuits that are involved in central sensitization and hyperalgesia. Also, the NK-1 receptor-dependent descending system might be a useful model for determining spinal mechanisms by which descending pathways facilitate nociceptive transmission in the dorsal horn and thereby provide novel treatment strategies.

Although it was surprising that NK-1 receptor antagonists were found not be beneficial clinically for chronic pain treatment (Boyce and Hill 2004; Hill 2000; Hill and Oliver 2007; Steinhoff et al. 2014), it was recently shown that during chronic pain NK-1 receptors are redistributed from the cell membrane into endosomes and located intracellularly where they activate pronociceptive signaling pathways (Jensen et al. 2017). Thus it is likely that the lipophobic nature of conventional NK-1 receptor antagonists prevents targeting of endosomal NK-1 receptors located intracellularly, which might explain the poor analgesia in clinical trials.

GRANTS

This work was supported by National Institutes of Health Grants DA-011471 (to D. A. Simone) and NS-089647 (to G. J. Giesler, Jr.).

DISCLOSURES

The authors have no conflicts of interest, financial or otherwise, to disclose.

AUTHOR CONTRIBUTIONS

S.G.K., G.J.G., and D.A.S. conceived and designed research; S.G.K., P.M., J.N., J.T., and D.A.S. performed experiments; S.G.K., P.M., J.N., J.T., and D.A.S. analyzed data; S.G.K., G.J.G., and D.A.S. interpreted results of experiments; S.G.K. and G.J.G. prepared figures; S.G.K. drafted manuscript; S.G.K., G.J.G.J., and D.A.S. edited and revised manuscript; S.G.K., G.J.G., and D.A.S. approved final version of manuscript.