| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the Department of Anesthesiology and Resuscitology, Okayama University Medical School, Okayama City 700-8558, Japan.
Address correspondence and reprint requests to M. Yokoyama, MD, Department of Anesthesiology and Resuscitology, Okayama University Medical School, 2-5-1, Shikata-cho, Okayama City 700-8558, Japan. Address e-mail to masayoko{at}cc.okayama-u.ac.jp.
Abstract
BACKGROUND: The status of neuropathic pain alters the responsiveness to formalin injection in rats. However, the mechanism by which this alteration occurs is unknown.
METHODS: We used immunocytochemistry to examine the expression of brain-derived neurotrophic factor (BDNF) and calcitonin gene-related peptide (CGRP) in the spinal cord of rats with L5 spinal nerve ligation (SNL)-induced neuropathy, and investigated the expression of c-Fos in the spinal cord after injection of formalin in the hindpaw of rats with SNL.
RESULTS: Four weeks after SNL, the withdrawal threshold was significantly lower in the SNL group than in the sham-operated (sham) group (n = 12 per group, P < 0.05). In the SNL group, expression of BDNF in the L4 (P < 0.05) and L5 (P < 0.01) superficial dorsal horn was significantly decreased compared to that in the sham group. CGRP protein in the L5 but not in the L4, dorsal horn was significantly decreased compared to that in the sham group (P < 0.01). After formalin injection, spontaneous pain responses in the SNL group were significantly decreased compared to those in the sham group (P < 0.05). Immunolabeling for c-Fos was significantly decreased in the L4 and L5 dorsal horn in the SNL group (P < 0.01).
CONCLUSION: Our examination of c-Fos distribution indicates that decreased neuronal activity in the spinal cord in response to inflammatory pain may be important for altering the perception of acute pain. Decreased BDNF expression in response to SNL-induced neuropathy may be involved in this alteration.
Patients with neuropathic pain often suffer acute pain in response to inflammation, trauma, or surgery (1). Although remarkable progress in the study of pain has shed light on the mechanisms of neuropathic pain (2–4), little is known about the interaction of different pain states over time. Peripheral nerve injury causes a reduction in the total number of nerve fibers (5,6) and induces hypersensitivity to non-noxious stimulation (2,4). Therefore, the response of surviving nerve fibers to different stimulation modalities can be altered (7).
Chronic constriction injury (CCI) of the sciatic nerve (2) and L5 SNL (4) in rats are two different models of neuropathic pain. These models show neuropathic pain, including tactile and thermal allodynia and hyperalgesia (2,4). Injection of formalin into the plantar aspect of the hindpaw in rats induces acute pain behaviors such as biting, licking, and limb flinching (8). Two (9,10) reports have shown that the status of neuropathic pain alters the responsiveness to formalin injection in rats. However, LaBuda et al. (9) reported that SNL enhances formalin-induced nociceptive responses, and Vissers et al. (10) reported that CCI of the sciatic nerve reduces these responses. Nonetheless, these opposing results may be due to differences in the animal models used. The mechanism by which neuropathic pain alters the response to formalin remains to be elucidated.
In addition to behavioral studies, molecular studies may be useful in elucidating this mechanism. Sensory inputs, particularly nociceptive inputs, increase expression of the immediate-early gene c-fos and its protein product c-Fos (11,12). In this study, we measured c-Fos immunolabeling in the spinal cord to examine changes in neuronal activity in central pathways after injection of formalin in the hindpaw of rats with SNL-induced neuropathy.
The role of brain-derived neurotrophic factor (BDNF) as a neuromodulator in the spinal dorsal horn, particularly in inflammatory pain, has been investigated (13). BDNF also contributes to hyperalgesia in SNL-induced pain (14). Calcitonin gene-related peptide (CGRP) is a well-characterized neuropeptide that is released from primary nociceptive afferents and is involved in nociception in the peripheral and central nervous systems (15). It has been reported that spinal CGRP is involved in mediating chronic neuropathic pain (16). BDNF and CGRP may be involved in the modulation of formalin-induced nociceptive transmission at the spinal level after SNL. In addition to the examination of c-Fos immunolabeling, we examined the BDNF and CGRP immunolabeling in the spinal cord to investigate possible pathways involved in altered sensory neuronal transmission.
METHODS
This study was approved by the Animal Care and Use Committee of Okayama University Medical School. Animals were treated in accordance with the Ethical Guidelines for Investigation of Experimental Pain in Conscious Animals issued by the International Association for the Study of Pain (17). Experiments were performed on 7-wk-old male Sprague–Dawley rats housed in pairs before surgery and individually after surgery. Food and water were available ad libitum.
Neuropathic Pain Model
For experiments, rats were placed individually on an elevated plastic mesh floor covered with a clear plastic cage top (21 x 27 x 15 cm3), and baseline withdrawal threshold in response to punctate mechanical stimulation was determined. The mechanical stimulus was applied from underneath the mesh (openings 12 x 12 mm2) to the plantar aspect of the proximal part of the heel by the up/down method with nine von Frey monofilaments (0.4, 0.6, 1, 1.4, 2, 4, 6, 8, and 15 g; Touch-Test® Sensory Evaluator, North Coast Medical, Morgan Hill, CA). Each trial was initiated with a von Frey force of 2 g delivered to the left hindpaw for approximately 1 s. If there was no withdrawal response, the next higher force was delivered. If there was a response, the next lower force was delivered. This procedure was performed six times after the first response. On the basis of the response pattern and the force of the final filament, the 50% response threshold was calculated by the formula described by Chaplan et al. (18).
After the baseline threshold was determined, rats were anesthetized with pentobarbital (50 mg/kg), and nerve injury was produced by tight ligation of the L5 spinal nerve as described previously (3,19). In brief, animals were placed in the prone position to access the left L4–6 spinal nerves. Under magnification, approximately one-third of the L6 transverse process was removed. The left L5 spinal nerve was identified and ligated tightly with 5-0 silk sutures (SNL group, n = 12). The wound was treated with antiseptic solution and closed with wound clips. For sham surgery, the nerve was exposed, but no ligature was applied (sham group, n = 12). To confirm tactile allodynia to the plantar aspect, the withdrawal threshold was measured every week for 4 wk after surgery.
Formalin Test
Four weeks after surgery, six animals in each group were subjected to formalin injection (9). In brief, animals were placed in a 30 x 30 x 30 cm3 Plexiglas chamber and allowed to habituate for at least 20 min. A mirror was placed below the Plexiglas chamber at a 45° angle to allow for easy viewing of behavioral responses. Animals received a 50-µL injection of 5% formalin solution to the plantar aspect of the left hindpaw. Behavioral testing was initiated immediately after formalin injection and lasted for 60 min. The incidences of flinching, biting, and licking were counted per min every 5 min; the first counting was performed after 5 min of formalin injection for 1 min, and the second counting was performed after 10 min of formalin injection for 1 min, and it was done in the same way thereafter.
BDNF and CGRP Immunohistochemistry
Four weeks after surgery, six animals in each group were anesthetized deeply with ether and perfused transvascularly with 50 mL saline, followed by 500 mL 4% formaldehyde in 0.1 M phosphate buffer (pH 7.4). The L4 and L5 spinal cord segments were dissected and postfixed in the same fixative for 30 min, incubated in phosphate-buffered 20% sucrose solution overnight, and 50-µm frozen sections were generated. For simultaneous visualization of BDNF and CGRP in the spinal cord, double immunofluorescence was performed (20). Sections were incubated in a mixture of chicken anti-BDNF IgE (1:200; R&D Systems, Minneapolis, MN) and rabbit anti-CGRP serum (1:1000; Peninsula Labs, Belmont, CA) for 24 h at room temperature (rt). The sections were then treated with a mixture of fluorescein isothiocyanate-conjugated donkey anti-chicken IgE (1:100; Jackson ImmunoResearch Laboratories, West Grove, PA) and lissamine rhodamine B-conjugated donkey anti-rabbit IgG (1:500; Jackson ImmunoResearch Laboratories) for 2 h at rt. Sections were dry-mounted on gelatin-coated glass slides, washed in phosphate-buffered saline (PBS), and coverslipped with Entellan (Merck, Darmstadt, Germany).
For quantification of neurons labeled for BDNF and CGRP, the L4 and L5 segments were subdivided into three rostrocaudal levels (rostral, medium, and caudal) and sections from each level were used. The superficial layer (laminae I–II) and deep layer (laminae III–VI) of the dorsal horn were identified bilaterally by dark- field microscopy, and BDNF-immunoreactive (ir) and CGRP-ir neurons were quantified with LuminaVision software (Mitani, Tokyo, Japan). To quantify the density of labeled neurons in each section, the ratio of labeling on the ipsilateral side to that on the contralateral side was calculated (21). Three to four sections from each level per animal were selected randomly, and the average density of BDNF- or CGRP-ir neurons per section in laminae I–II and laminae III–VI were recorded.
c-Fos Immunohistochemistry
Two hours after formalin injection, six animals from each group were anesthetized deeply with ether and perfused as described earlier. Spinal cords were dissected and post-fixed in the same fixative as above overnight, incubated in phosphate-buffered 20% sucrose solution overnight, and 50-µm frozen sections were generated. Sections were immunolabeled for c-Fos by the peroxidase–anti-peroxidase method of Sugimoto et al. (22) In brief, sections were incubated in 80% ethanol containing 0.3% H2O2 at rt for 30 min, followed by incubation in blocking solution composed of 0.02 M PBS containing 1% normal goat serum at rt for 1 h. Sections were then incubated at 4°C for 48–72 h in primary antibody (1:10,000; rabbit polyclonal antibody to c-Fos; Santa Cruz Biotechnology, Santa Cruz, CA), washed, and incubated at 4°C for 12 h in biotinylated secondary antibody (1:400; goat anti-rabbit IgG; Kirkegaard & Perry Laboratories, Gaithersburg, MD) and peroxidase–anti-peroxidase complex (1:3000; MP Biomedicals, Irvine, CA) for 1 h at rt. All antibodies were diluted in PBS containing 1% normal goat serum and 0.75% Triton X-100. Before incubation with each antibody, sections were rinsed thoroughly in PBS, and pre-incubated with PBS containing 3% normal goat serum for 1 h. Reaction product was visualized with nickel ammonium sulfate-intensified diaminobenzidine as the chromogen. Sections were dry-mounted on gelatin-coated glass slides, dehydrated in an alcohol series, and coverslipped with Entellan (Merck). Locations of c-Fos-ir neurons were plotted with a camera coupled to a bright-field microscope.
For quantification of neurons immunolabeled for c-Fos, the L4 and L5 segments were subdivided into three rostrocaudal levels (rostral, medium, and caudal) and sections from each level were used. The superficial laminae were identified by dark-field microscopy, and the numbers of c-Fos-ir neurons in the superficial laminae (I–II) and deeper laminae (III–VI) were counted. Three to four sections from each level per animal were selected randomly, and the average numbers of c-Fos-ir neurons per section in laminae I–II and laminae III–VI were recorded.
Statistical Analysis
Statistical analysis was performed with two-way ANOVA, followed by Dunn’s test for behavioral experiments and with unpaired t-test for protein expression experiments. Differences were considered statistically significant at P < 0.05.
RESULTS
Tactile allodynia was observed from 1 wk after surgery and before formalin injection in the SNL group, and the withdrawal threshold of the ipsilateral hindpaw was significantly decreased in the SNL group compared to that in the sham group (P < 0.01; Fig. 1). The values of the contralateral side in the SNL group and both sides in the sham group showed no significant differences compared to the values before surgery. The formalin-induced pain response in both groups followed the characteristic biphasic pattern (Fig. 2). There was an initial phase of increased pain behavior (0–5 min), followed by a phase of decreased pain behavior. Analysis of each 5-min test period showed no significant difference in response between the two groups in the first phase. However, the total incidence of flinching, biting, and licking was significantly decreased in the SNL group compared to that in the sham Group 15–35 min (P < 0.01) and 50–55 min (P < 0.05) after formalin injection.
|
|
BDNF staining in the superficial layer of the ipsilateral L4 and L5 dorsal horn was significantly decreased in all (rostral, medium, and caudal) levels in the SNL group compared to that on the contralateral side (P < 0.01; Fig. 3 and Table 1), and these values were significantly decreased compared to those in the sham group (L4: P < 0.05 or 0.01; L5: P < 0.01). No significant difference in BDNF expression was observed in the deeper layer in both groups. CGRP staining in both superficial and deeper layers of ipsilateral L5 dorsal horn was significantly decreased in all levels in the SNL group compared to those on the contralateral side (P < 0.01; Fig. 4 and Table 2), and these values were significantly decreased compared to those in the sham group (P < 0.01). In the superficial and deeper layers of the ipsilateral L4 dorsal horn, a significant decrease in CGRP expression was observed in the caudal level, but not observed in the medium and rostral levels. c-Fos staining in both layers of the ipsilateral L4 and L5 dorsal horn was significantly decreased in all levels in the SNL group compared to that in the contralateral side (P < 0.01; Fig. 5 and Table 3), and these values were also significantly decreased compared to those in the sham group (P < 0.01).
|
|
|
|
|
|
DISCUSSION
Results of this study showed that SNL for 4 wk reduced the incidence of pain response to formalin injection. Examination of c-Fos immunolabeling indicates that decreased neuronal activity in the spinal cord in response to inflammatory pain may be important in altering the perception of acute pain. BDNF may be involved in this alteration.
We showed a reduced pain response, whereas LaBuda et al. (9) reported that SNL enhanced the pain response to formalin. The reason for these opposing results is unclear, but possibilities include the following. First, LaBuda et al. (9) performed the formalin test 2 wk after SNL, whereas we performed it 4 wk after SNL. It has been reported that expression of BDNF in the spinal cord is increased 2 wk after SNL (23). In our study, BDNF immunolabeling was decreased 4 wk after SNL. Different time courses may lead to different results. Although we did not perform the formalin test 2 wk after SNL, the purpose of our study was to observe the effect of chronic neuropathic pain on acute pain. We believe that a 4-wk model is better suited for this objective than a 2-wk model. Second, LaBuda et al. (9) used 1% formalin, and we used 5% formalin. Vissers et al. (10) used different concentrations of formalin (5%, 2.5%, 0.63%, and 0.16%) in rats subjected to CCI, and they observed dose-related alterations in behavior. Thus, differences in formalin concentration may lead to differences in behavioral responses. Third, the method of assessment of pain behavior differed. In the LaBuda et al. study (9), behavior was scored as 2 if the rat licked or bit the injected paw, as 1 if the rat elevated the paw from the floor, and as 0 if any part of the paw other than the tips of the digits was in contact with any surface of the box. Their results showed an increased pain score and increased duration of paw elevation in response to formalin injection in the SNL group. We counted the total incidence of flinching, biting, or licking per minute and showed a decreased total number of pain behaviors in response to formalin injection in the SNL group. Because the paw was not usually in contact with the surface of the box before formalin injection, due to SNL-induced allodynia, we did not use the duration of paw elevation as a formalin-induced pain behavior. Further investigation is necessary to establish the best evaluation method for formalin injection involving interaction of different pain states over time.
Vissers et al. (10) reported that CCI of the sciatic nerve reduced the response to formalin, and their method of assessment of pain behavior was almost similar to that in our study. These results suggest that CCI- and SNL-induced neuropathy cause similar alterations in the perception of acute pain. Vissers et al. (10) measured the levels of stress hormones to examine mechanisms by which CCI reduces the pain response to formalin and showed that increased levels of stress hormones after formalin injection were suppressed in rats subjected to CCI (10). They speculated that in response to nerve damage, neuronal fibers of the affected hindpaw fire spontaneously (24), causing continuous stimulation of inhibitory descending pathways. As a result, transduction of painful stimuli is masked. However, whether the reduced response is caused by decreased pain reactivity at the peripheral, spinal, or supraspinal level, or by changes in stress reactivity or coping strategies remains unknown (10).
Molecular studies may be useful in elucidating this mechanism. In the neuropathic pain status, the response of surviving and intact nerve fibers to different stimulation modalities can be altered (7). Terminals of primary afferents from the paw are located mostly in the L4 and L5 dorsal horns (25). Therefore, we investigated the expression of pain-related substances in the L4 as well as the L5 spinal cord. We also subdivided the L4 and L5 segments into three rostrocaudal levels (rostral, medium, and caudal), because the caudal level of L4 spinal cord receives the input from the adjacent L5 fibers, but it is likely that the medium and rostral levels of L4 spinal cord receive less input from L5 fibers (26). Therefore, the study of expression of pain-related substances in each level may help to clarify the mechanism by which SNL changes the response in the spinal cord level. Immunocytochemical localization of c-Fos in the spinal cord has been used widely to identify populations of neurons that are activated by exogenous peripheral stimuli (11,12). Our results showed that c-Fos immunolabeling was significantly decreased in the SNL group compared to that in the sham group. Furthermore, c-Fos immunolabeling was decreased in all (rostral, medium, and caudal) levels in the L4 as well as the L5 dorsal horn. It has been reported that the input of peripheral nociceptive stimuli to the L5 dorsal horn is altered by L5 SNL (27). It was surprising that the formalin-based peripheral nociceptive stimulation was significantly suppressed in each level of the L4 region in response to L5 SNL compared to that in the sham group. These results indicate that the peripheral stimuli are conducted to the spinal cord, and that decreased neuronal activity in response to inflammatory pain occurs at the level of the spinal cord. As Vissers et al. (10) speculated, continuous stimulation of descending inhibitory pathways may play an important role in the suppression of c-Fos expression.
To further investigate mechanisms by which SNL reduces pain behavior and c-Fos expression, we quantified BDNF immunolabeling. BDNF immunolabeling was also decreased in the superficial layer in all levels of the L4 and L5 dorsal horn that contain neurons activated by nociceptive stimuli. BDNF exerts a modulatory effect at hippocampal synapses, which are involved in learning and memory, and at the first pain synapse between primary sensory neurons and dorsal horn neurons (28). In the dorsal horn, activation of trkB receptors by endogenous BDNF appears to contribute to hyperalgesia associated with peripheral inflammation (29), and it has been suggested that BDNF contributes to sensitization of dorsal horn neurons (30). BDNF is released in the dorsal horn, along with substance P and glutamate, following patterned high-frequency activation of high-threshold nociceptive fibers (31). Behavioral experiments have provided evidence for a pronociceptive effect of endogenous BDNF (30,32,33), and suppression of BDNF by antibody leads to antinociceptive effects (14). It is possible that decreased BDNF in the spinal cord alleviates formalin-induced pain.
We also quantified CGRP immunolabeling in the spinal cord to investigate possible mechanisms by which SNL alters the response to formalin injection. CGRP, an excitatory neuropeptide localized in primary afferents, plays a key role in nociceptive signaling (34) and is also important in the generation and maintenance of neuropathic pain; CGRP modulates inhibitory descending pathways (15). Our results showed that the CGRP immunolabeling was decreased in all levels of the L5 dorsal horn and the caudal level of the L4 dorsal horn, but not in the medium and rostral levels of the L4 dorsal horn. Therefore, we found no evidence that CGRP is directly involved in the mechanism of altered response to formalin by SNL.
Our results indicate that the status of neuropathic pain can alter the perception of inflammatory pain and that the mechanism occurs at the spinal level. Further studies are needed to fully elucidate the mechanisms involved and should investigate the response to other types of pain, such as that caused by incision, chemicals, or nerve injury, because patients with neuropathic pain often suffer from several types of pain. It is also important to investigate the status of neuropathic pain itself during and after interaction of different pain states over time.
ACKNOWLEDGMENTS
The authors thank Dr. Hiroyuki Ichikawa, Dr. Tomoichiro Yamaai, and Dr. Tomosada Sugimoto, Department of Second Oral Anatomy, Okayama University Dental School, for their instructing immunohistochemical methods.
Footnotes
Accepted for publication December 20, 2006.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
