Role of spinal cord alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors in complete Freund's adjuvant-induced inflammatory pain

Male Sprague-Dawley rats (250–300 g) were housed in cages on a standard 12:12 h light/dark cycle. Water and food were available ad libitum until rats were transported to the laboratory approximately 1 h before experiments. The animals were used in accordance with protocols that were approved by the Animal Care and Use Committee at the Johns Hopkins University and were consistent with the ethical guidelines of the National Institutes of Health and the International Association for the Study of Pain. All efforts were made to minimize animal suffering and to reduce the number of animals used.

Intrathecal catheters were implanted into animals under isoflurane anesthesia. A polyethylene (PE-10) tube was inserted into the subarachnoid space at the rostral level of the spinal cord lumbar enlargement segment through an incision at the atlanto-occipital membrane according to the method described previously [16, 17]. The animals were allowed to recover for 5–10 days before being used experimentally. Rats showing any neurologic deficits postoperatively were discarded from the study. The position of the PE-10 catheter was confirmed in each animal after behavioral testing.

CFA was purchased from Sigma Chemical Co. (St. Louis, MO). 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, a competitive AMPA/kainate receptor antagonist), CFM-2 (a selective non-competitive AMPAR antagonist), and GYKI 52466 (a selective non-competitive AMPAR antagonist) were purchased from Tocris Bioscience (Ellisville, MO). CNQX was dissolved in 0.9% physiologic saline, whereas CFM-2 and GYKI 52466 were dissolved in 10% dimethylsulfoxide (DMSO).

To induce persistent inflammatory pain, the rats were placed under isoflurane anesthesia, and 100 μl of CFA (1 mg/ml Mycobacterium tuberculosis) solution was injected into the plantar side of one hind paw. Our previous studies showed that significant CFA-induced thermal and mechanical pain hypersensitivities appeared at 2 h, reached a peak level between 6 and 24 h, and were maintained for at least 72 h [17]. Therefore, we chose 2 h and 24 h post-CFA injection to represent the development and maintenance phases of CFA-induced persistent inflammatory pain for the pharmacologic and biochemical studies.

To examine the role of spinal cord AMPARs in persistent inflammatory pain, two selective competitive AMPAR antagonists, CFM-2 (5, 25, or 50 μg/10 μl) and GYKI 52466 (50 μg/10 μl) were injected intrathecally followed by 10 μl of saline to flush the catheter (total volume of the catheter: 10 μl) at 2 h or 24 h post-CFA injection. To compare the effects of selective and non-selective AMPAR antagonists on persistent inflammatory pain, a competitive AMPA/kainate receptor antagonist, CNQX (6.1 μg/10 μl), was given according to the same protocol. Saline and 10% DMSO were used as controls. The behavioral tests described below were performed 1 day before CFA injection (baseline) and at 20 min after drug administration. Separate groups of rats were used for the 2-h and 24-h behavioral tests. The antagonist doses used were based on data from previous studies [18‐21] and our pilot work. The experimenters were blinded to the treatment groups.

To measure paw withdrawal response to noxious heat stimuli, each animal was placed in a Plexiglas chamber on a glass plate located above a light box. Radiant heat from a Model 336 Analgesic Meter (IITC, Inc./Life Science Instruments, Woodland Hills, CA) was applied by aiming a beam of light through a hole in the light box through the glass plate to the middle of the plantar surface of each hind paw. When the animal lifted its foot, the light beam was turned off. The length of time between the start of the light beam and the foot lift was defined as the paw withdrawal latency. Each trial was repeated five times at 5-min intervals for each paw. A cut-off time of 20 s was used to avoid paw tissue damage.

To measure paw withdrawal response to repeated mechanical stimuli, each animal was placed in a Plexiglas chamber on an elevated mesh screen. A single trial of mechanical stimuli consisted of eight applications of a calibrated von Frey filament (8.01 mN, Stoelting Co., Wood Dale, IL) within a 2–3-s period. Each trial was repeated 10 times at 3-min intervals on each hind paw. The occurrence of hind paw withdrawal in each of these 10 trials was expressed as a percent response frequency, and this percentage was used as an indication of the amount of hind paw withdrawal.

To examine whether the antagonists used in behavioral testing affected the locomotor function, three reflexes (placing, grasping, and righting) were tested as described previously [16, 22]. In brief, the naïve animals received a 10-μl intrathecal injection of vehicle (saline or 10% DMSO), CNQX (6.1 μg/10 μl), CFM-2 (50 μg/10 μl), or GYKI 52466 (50 μg/10 μl). Twenty minutes later, the following tests were performed with the experimenter blind to drug treatment: (1) Placing reflex: The experimenter held the rat with hind limbs slightly lower than the forelimbs and brought the dorsal surfaces of the hind paws into contact with the edge of a table. The experimenter recorded whether the hind paws were placed on the table surface reflexively; (2) Grasping reflex: The experimenter placed the rats on a wire grid and recorded whether the hind paws grasped the wire on contact; (3) Righting reflex: The experimenter placed the rat's back on a flat surface and noted whether it immediately assumed the normal upright position. Scores for placing, grasping, and righting reflexes were based on counts of each normal reflex exhibited in five trials. In addition, the rat's general behaviors, including spontaneous activity (e.g. walking and running), were observed.

Expression of GluR1 and GluR2 proteins in the total soluble fraction, crude membrane fraction, and crude cytosolic fraction was examined. In brief, the animals were sacrificed by decapitation at 2 and 24 h after injection of saline (100 μl) or CFA into the plantar side of a hind paw. Naïve animals were used as controls. The L_4–5 spinal cord ipsilateral and contralateral to CFA or saline injection was removed. The dorsal part of the spinal cord was separated from the ventral part and collected. Total soluble fraction, crude membrane fraction, and crude cytosolic fraction were prepared as described below. The samples were heated for 5 min at 99°C and then loaded onto 4% stacking/7.5% separating SDS-polyacrylamide gels. The proteins were electrophoretically transferred onto nitrocellulose membrane. The blotting membrane was blocked with 3% non-fat dry milk for 1 h and incubated overnight at 4°C with rabbit anti-GluR1 (1:200; Upstate/CHEMICON, Temecula, CA), rabbit anti-GluR2 (1:500; Upstate/CHEMICON), rabbit anti-N-cadherin (1:1,000; BD Biosciences, Palo Alto, CA), mouse anti-PSD-95 (1:1,000; Upstate/CHEMICON), or monoclonal mouse anti-β-actin (1:10,000; Santa Cruz Biotechnology, Inc., Santa Cruz, CA). N-cadherin and PSD-95 were used as loading controls and markers for crude membrane fraction, whereas β-actin was used as a loading control for total soluble fraction and cytosolic fraction. The proteins were detected with anti-rabbit or anti-mouse secondary antibody and visualized using the chemiluminescence reagents provided with the ECL kit (Amersham Pharmacia Biotech, Piscataway, NJ) and exposure to film. The intensity of blots was quantified with densitometry. The blot density from naïve animals was set as 100%.

After tissues were homogenized in lysis buffer (10 mM Tris-HCl, 250 mM sucrose, 5 mM MgCl₂, 2 mM EGTA, 1 mM phenylmethylsulfonyl fluoride, 2 mM sodium orthovanadate, 1 mM NaF, 1 mM leupeptin, 2 mM pepstatin A, 1 mM dithiothreitol), the crude homogenates were centrifuged at 4°C for 15 min at 900 g. The supernatant was collected, and the pellet (nuclei and debris fraction) discarded. After the measurement of protein concentration, 20% of the supernatant was removed and considered to be the total soluble fraction. The remaining supernatant (80%) was centrifuged at 37,000 g for 1 h at 4°C. The supernatant was considered to be the crude cytosolic fraction, and the pellet, the crude plasma membrane fraction. The pellet was dissolved in buffer (10 mM Tris-HCl, pH 7.4, 1.5% SDS, and 0.1% Triton X-100). Protein concentrations of the three fractions were measured, and samples were prepared for Western blotting, as described above.

The animals were deeply anesthetized with isoflurane and perfused with 4% paraformaldehyde in phosphate buffer (0.1 M, pH 7.4) at 2 or 24 h after saline or CFA injection. Naïve rats were used as controls. The L_4–5 spinal cord segments were harvested, post-fixed in the same fixative solution for 2–4 h, cryoprotected by immersion in 30% sucrose overnight at 4°C, and frozen-sectioned at 25 μm. Every fourth section was collected (at least 30 sections/rat). The sections were blocked for 1 h at 37°C in 0.01 M phosphate-buffered saline (PBS) containing 10% normal goat serum plus 0.3% Triton X-100. Half of the sections (approximately 15 sections/rat) were incubated in primary rabbit antibody for GluR1 (1:500; Upstate/CHEMICON) and the remaining sections in primary mouse antibody for GluR2 (1:1,000; Upstate/CHEMICON) for 48 h at 4°C. The sections were finally incubated in biotinylated goat anti-rabbit or anti-mouse IgG (1:200; Vector Laboratories, Burlingame, CA) for 1 h at 37°C followed by avidin-biotin-peroxidase complex (1:100; Vector) for 1 h at 37°C. The immune reaction product was visualized by catalysis of 3,3-diaminobenzidine by horseradish peroxidase in the presence of 0.01% H₂O₂.

Immunostained spinal cord sections were quantified with an Olympus microscope (Olympus, Tokyo, Japan) linked to a Toshiba 3CCD camera (Toshiba, Japan) and I-Cube computer image analysis system (I-Cube, Cambridge, MA). The entire superficial dorsal horn was delimited to quantify the optical density of GluR1 and GluR2 immunoreactivity. Five spinal cord sections randomly selected from a total of 15 sections from each animal were analyzed, and the relative optical densities of five sections were averaged to provide a mean for each animal.

The effect of intrathecal administration of CFM-2 on mechanical and thermal hypersensitivities in animals at 2 h or 24 h after CFA injection was first examined. Consistent with previous studies [17], CFA injection produced a significant increase in paw withdrawal frequency in response to mechanical stimulation (indicating mechanical hypersensitivity; Fig. 1A) and a remarkable decrease in paw withdrawal latency in response to thermal stimulation (indicating thermal hypersensitivity; Fig. 1C) on the ipsilateral side at both 2 h (n = 14, 7/test) and 24 h (n = 14, 7/test) post-CFA injection in the vehicle group. Intrathecal administration of CFM-2 dose-dependently attenuated CFA-induced mechanical hypersensitivity at 2 h and 24 h post-CFA injection (Fig. 1A). Compared with the corresponding vehicle-treated groups, the 25-μg dose of CFM-2 reduced paw withdrawal frequencies by 26% (p < 0.05) and 35% (p < 0.01) at 2 h and 24 h (n = 6/time point) post-CFA injection, respectively, and the 50-μg dose reduced paw withdrawal frequencies by 37% (p < 0.05) and 49% (p < 0.01) at these two time points (n = 6/time point). Paw withdrawal frequencies after administration of 5 μg of CFM-2 at 2 h (n = 6) and 24 h (n = 6) post-CFA injection were not significantly different from those of the vehicle group at those time points (p > 0.05). Interestingly, only the highest CFM-2 dose (50 μg) significantly reduced the CFA-induced thermal hypersensitivity (Fig. 1C). Compared with vehicle-treated rats, paw withdrawal latencies were increased by 31% at 2 h post-CFA injection (n = 6; p < 0.05) and by 70% at 24 h post-CFA injection (n = 6; p < 0.05). Neither the vehicle nor any of the three CFM-2 doses significantly altered basal paw withdrawal responses to mechanical and thermal stimuli applied to the contralateral hind paw at 2 h or 24 h post-CFA injection (Fig. 1B,D). Likewise, none of the doses of CFM-2 significantly affected basal paw withdrawal responses to mechanical and thermal stimuli in naïve rats (data not shown).

To further confirm the effect of selective non-competitive AMPAR antagonists on persistent inflammatory pain, we intrathecally injected another selective non-competitive AMPAR antagonist, GYKI 52466 [15], at 2 h and 24 h post-CFA injection. A 50-μg injection of GYKI 52466 (n = 6/test/time point) attenuated both CFA-induced mechanical and thermal hypersensitivities on the ipsilateral hind paw (Fig. 2A,E). Paw withdrawal frequencies were 31% and 38% lower than those of the vehicle-treated group at 2 h (p < 0.05) and 24 h (p < 0.05) post-CFA injection, respectively (Fig. 2A). Paw withdrawal latencies were 38% and 92.5% higher than those of the vehicle-treated group at 2 h (p < 0.05) and 24 h (p < 0.05) post-CFA injection, respectively (Fig. 2E). Similar to CFM-2, 50 μg GYKI 52466 had no effect on paw withdrawal responses to thermal or mechanical stimuli applied to the contralateral hind paw at either time point (Fig. 2B,F) or in naïve rats (data not shown).

CNQX is a non-selective competitive AMPAR antagonist. Compared with the responses of the vehicle-treated group (n = 20; 5/test/time point), intrathecal CNQX (6.1 μg) significantly reduced CFA-induced mechanical and thermal hypersensitivities at 2 h (p < 0.01) and 24 h (p < 0.01) after CFA injection (Fig. 2C,G). However, the same dose of CNQX also significantly reduced basal paw withdrawal responses to mechanical stimulation on the contralateral hind paw at both time points (both p < 0.05; Fig. 2D), although it had no effect on basal paw withdrawal responses to thermal stimulation on the contralateral hind paw after CFA injection (Fig. 2H).

To exclude the possibility that the behavioral effects described above were due to impaired motor functions caused by these antagonists, we then examined the effects of CFM-2, GYKI 52466, and CNQX on locomotor functions in naive animals. As indicated in Table 1, at the doses used, none of the antagonists significantly altered placing, grasping, or righting reflexes. In addition, no significant differences were observed in the rats' general behaviors (such as walking and running) between the vehicle-treated and the antagonist-treated groups.

To further define the involvement of spinal cord AMPARs in persistent inflammatory pain, we examined whether CFA-induced inflammation produced the changes in expression of total GluR1 and GluR2 proteins in total soluble fraction derived from dorsal horn. In the ipsilateral and contralateral L_4–5 dorsal horn, neither total GluR1 nor total GluR2 protein level was significantly different from that of naïve animals (n = 4) at 2 and 24 h post-saline (n = 4/time point) and post-CFA (n = 4/time point) injection (p > 0.05; Fig. 3A, B).

We next examined whether peripheral inflammatory insult affected the distribution of total GluR1 and GluR2 immunoreactivities in the spinal dorsal horn at 2 and 24 h after CFA injection. Consistent with previous studies [12, 23, 24], GluR1 immunoreactivity was distributed mainly in laminae I and II, whereas GluR2 immunoreactivity was concentrated in inner lamina II and lamina III in naïve rats (n = 3). Under high magnification, some GluR1- and GluR2-positive cell bodies were observed in both superficial and deep dorsal horn (Fig. 3C). Neither intraplantar saline nor CFA produced significant changes in the distribution or optical density of GluR1 and GluR2 immunoreactivities in L_4–5 spinal dorsal horn on the ipsilateral or contralateral side at 2 h (p > 0.05) or 24 h (p > 0.05) post-saline (n = 6, 3/time point) or post-CFA (n = 6, 3/time point) injection (Fig. 3D–F).

Finally, we examined whether CFA-induced peripheral inflammation affected the subcellular distribution of GluR1 and GluR2 proteins in dorsal horn neurons. Quantification of the protein levels revealed that, compared with the naïve animals (n = 4), the amount of GluR2 in the CFA-treated group was 90% higher in the cytosolic fraction (Fig. 4A; p < 0.01) and correspondingly 26% lower in the membrane fraction (Fig. 4B; p < 0.05) at 24 h post-CFA injection (n = 4). Conversely, the amount of GluR1 was decreased by 25% in the cytosolic fraction (Fig. 4C; p < 0.05) and correspondingly increased by 23% in the membrane fraction (Fig. 4D; p < 0.05) at 24 h post-CFA injection (n = 8) compared with the values of the naïve group (n = 8). Neither the GluR1 nor the GluR2 levels were significantly changed in either the cytosolic (Fig. 4A, C; p > 0.05) or membrane (Fig. 4B, D; p > 0.05) fraction at 2 h post-CFA injection (n = 4). As expected, no significant differences were observed between the saline-treated and naïve groups in either the cytosolic or membrane fractions at 2 or 24 h post-saline injection (Fig. 4; n = 4–8/time point).

To test the specificity of the fractionation procedure, the expression levels of a plasma membrane-specific protein (N-cadherin), a postsynaptic density protein (PSD-95), and an intracellular protein (β-actin) were assessed in both crude plasma membrane and cytosolic fractions from naïve rats (n = 3). Consistent with previous studies [9, 22], N-cadherin and PSD-95 were detected at high levels in the crude plasma membrane fraction but were barely detectable in the crude cytosolic fraction (Fig. 4E), whereas β-actin was expressed predominately in the crude cytosolic fraction (Fig. 4E), indicating that the fractionation procedure effectively separated cytosolic proteins from plasma membrane proteins.