| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the *Department of Pharmacology, Gülhane Academy of Medicine, Etlik, Ankara, Turkey; Department of Pharmacology and Clinical Pharmacology and Headache Research Center, Medical School of Ankara University, Ankara, Turkey; Departments of Pathology, Analytic Toxicology, Gülhane Academy of Medicine, Etlik, Ankara, Turkey; and ||Department of Pharmacology, Arizona Health Sciences Center, University of Arizona, Tucson, Arizona.
Address correspondence and reprint requests to F. Cankat Tulunay, Department of Pharmacology and Clinical Pharmacology, Medical School of Ankara University, Ankara, Turkey. Address e-mail to f.cankat.tulunay{at}dialup.ankara.edu.tr.
Abstract
BACKGROUND: While many preclinical models detect the analgesic activity of nonsteroidal antiinflammatory drugs (NSAIDs), the radiant heat tail-flick response has repeatedly been insensitive to this class of drugs. As the tail-flick test involves nociceptive processing at spinal circuits with supraspinal modulation, it seems reasonable to assume that the NSAIDs should not modify strong nociceptive stimuli, since the primary site of action of NSAIDs is likely to be in the periphery.
METHODS: We injected 3–300 µg of diclofenac, dipyrone, ketorolac, lysine acetyl salicylate, and sodium salicylate intradermally into mice tails and evaluated the tail-flick response to radiant heat. These results were compared with intraperitoneally injected controls. We also evaluated the ability of naloxone to reverse the observed effects.
RESULTS: Intradermal injection of each NSAID produced a dose-dependent increase in tail-flick latency. Intraperitoneal NSAIDs injection produced no antinociceptive effects. Naloxone pretreatment had no effect on the antinociceptive effects of intradermal diclofenac, ketorolac, lysine acetyl salicylate, and sodium salicylate. Naloxone completely blocked the antinociceptive effects of intradermal dipyrone.
CONCLUSIONS: Local, but not systemic, administration of NSAIDs produced antinociception in the tail-flick thermal assay. The endogenous opioid system contributes to the peripheral antinociceptive effects of dipyrone, but not to that of diclofenac, ketorolac, lysine asetyl salicylate, or sodium salicylate, suggesting differences in the mechanisms of action among the NSAIDs.
The antiinflammatory and antinociceptive activities of nonsteroidal antiinflammatory drugs (NSAIDs) are attributed to inhibition of the cyclooxygenase (COX) enzymes, thus blocking the synthesis of prostaglandins that promote inflammatory responses and enhanced sensitivity to pain at the peripheral site of tissue injury (1–6). The radiant tail-flick test is a commonly used preclinical models of nociception. However, it is widely regarded as resistant to the actions of NSAIDs (1–6). The mechanisms and sites of action of NSAIDs are consistent with the failure of these compounds to inhibit strong nociceptive stimuli, such as those associated with radiant heat applied to the tail. Preclinical studies have emphasized that NSAIDs typically do not increase the pain threshold in animal models as measured in threshold escape paradigms, such as tail-flick and hotplate tests, but they normalize the exaggerated pain behavior, which is observed after tissue injury and inflammation mechanism (1–6).
In the radiant tail flick test, a thermal stimulus is focused on the skin of the animal's tail, activating nociceptors in the surface layers of the skin (1). The activation of peripheral nociceptors triggers a complex series of processes at the spinal level, resulting in the tail-flick response (1,7). The tail-flick response is modulated by descending influence from suprapinal sites (1). Because the primary site of action of NSAIDs is likely to be in the periphery, and the tail-flick test involves nociceptive processing at spinal circuits with supraspinal modulation, it seems reasonable that these drugs should not modify strong nociceptive stimuli. Since most investigations of the possible actions of NSAIDs in modulation of responses to strong nociceptive stimuli have used the systemic or intrathecal routes (4,5,8,9), we hypothesized that there may be an inadequate drug concentration in the vicinity of peripheral nociceptive terminals, which may lead to a failure to observe an antinociceptive response. Consistent with this hypothesis, the skin concentration of acetylsalicylic acid after systemic administration was found to be about 80–100 times lower than that seen after topical administration (10,11). In this study, we injected NSAIDs intradermally into mice tails and measured the local nociceptive threshold in the tail-flick test.
METHODS
After institutional review and approval in accordance with the Guide for the Care and Use of Laboratory Animals as adopted by the National Institutes of Health we studied adult male Balb-C mice (25–30 g). The animals were placed in a quiet and room with a 12/12 h light-dark cycle (08 am-08 pm light). All efforts were made to minimize animal suffering and to reduce the number of animals used. Each animal was used only once and received only one dose of the drugs being tested. Eight animals were used for each treatment group.
Drugs
The drugs used and their suppliers are as follows: diclofenac sodium (Cayman Chem, Ann Arbor, MI), dipyrone (Sigma, St. Louis, MO), ketorolac tromethamin (Syntex Laboratories, Palo Alto, CA), dl-lysine monoacetylsalicylate (Aspisol®, Bayer Leverkusen, Germany), sodium salicylate (Sigma), and naloxone HCl (Abbott, Abbott Park, IL). All drugs were freshly dissolved in 0.9% saline for intradermal or intraperitoneal (i.p.) injection.
Radiant Heat Tail-Flick Test
Antinociception was assessed using the radiant heat tail-flick test (Columbus, OH; Type 812). The mouse tail was marked with a pen about 3 cm from the tip and the light beam was focused on this marked site. Baseline tail-flick latency for each mouse was determined and designated as the baseline latency. The intensity of light was adjusted so that baseline latencies were 2–3 s, with a cutoff time of 6 s to prevent tissue damage. Animals were restrained and the drugs or 0.9% saline were injected intradermally using a 30 gauge needle attached to a Hamilton syringe with a volume of 10 µL into the skin of the tail on the marked site. Test latencies were measured after drug or 0.9% saline injections. To determine whether NSAIDs acted locally, tail-flick latencies elicited by stimulation of the tail at a site 1–2 cm more proximal from the marked site was concomitantly measured. Tail-flick latencies from the uninjected portion of the tail were similar to baseline latencies. Naloxone (5 mg/kg) was given subcutaneously (s.c.) into the skin on the animal's back in a volume of 0.1 mL/10 g 20 min before injection of peripheral NSAIDs to identify whether the endogenous opioid system was involved in the antinociceptive effects of NSAIDs. In a further set of experiments, a single injection of naloxone (5 mg/kg, s.c. in the back) was compared with the second repeated dose of s.c. naloxone 30 min after the injection of the NSAIDs. In another set of experiments, the NSAIDs were given i.p. and then tail-flick latencies were measured to assess the effects of systemic NSAID administration.
Tail-flick latency was assessed at baseline, then at 10, 20, 30, 45, 60, 120, 180 min after intradermal injection. Another set of animals received maximal NSAID doses, and additional measures of tail-flick latency were made 240, 300, 360, and 480 min after intradermal injection.
Histopathology of Tail Skin After Intradermal NSAIDs Injections
To identify the injection site and to demonstrate that injection of a high dose of NSAID did not affect tissue histology, we injected blue-colored marker (tissue marking dye, blue, Thermo Electron CO, Cat# 3120125, Pittsburgh, PA) along with the highest doses of NSAIDs or 0.9% saline intradermally using a 30 gauge needle attached to a Hamilton syringe. The injectate volume as 10 µL. Thirty min after intradermal injections, mice were killed with ether and the marked tail segment was removed and immersed in 10% phosphate buffered formalin (pH 7.0) at room temperature overnight. The specimens were then rinsed in buffer and decalcified in 10% formic acid for 6 h. They were dehydrated in a graded series of ethanol and embedded in paraffin as transverse tissue sections to tails. Three-five-micrometer thick sections on slides were prepared with a rotary microtome and stained with hematoxylin-eosin. Stained slides were examined under Zeiss photomicroscope. Microphotographs were taken by a computer-assisted system with a Sony video camera. Each tail section was evaluated in a blinded manner, without the investigator's knowing the identity of the treatment group, and scored for histopathological changes. Tissue slides of specimens were systematically evaluated from epidermis to bone structures of tails. Drug injection sides were histomorphologically compared with 0.9% saline or noninjected groups. Epidermis, dermal collagenous matrix, lymphatics, arterioles and nerves, subcutaneous fat and nerves, and striated muscles and nerves next to bone in these tissue sections were examined under a microscope.
Local tissue toxicity was graded for epidermal injury, acute inflammation, edema, and nerve fiber injury. The extent of epidermal injury was graded as 0 = no, 1 = intercellular edema/inflammatory infiltration, 2 = epidermal cell sloughing, and 3 = erosion/ulcer. Edema and inflammatory infiltration were scored as no = 0, 1 = mild, 2 = moderate, and 3 = marked. Nerve fiber injury was graded as 0 = no, 1 = nerve fiber distortion, 2 = shrinkage, 3 = necrosis.
Statistical significances of more than two groups were evaluated by Kruskal-Wallis test (P < 0.05), followed by Dunnett's multiple test for individual comparisons (P < 0.05). To compare two groups, Mann-Whitney U-test was used (P < 0.05). Data were expressed as mean ± sem. To generate a dose-response curve at 30 min after drug administration, the data were converted to % antinociception by the formula: % Antinociception = 100 x (test latency – baseline latency)/(6 – baseline latency). The A50 dose (i.e., dose producing 50% maximum possible effect) and 95% confidence intervals were calculated from the dose-percent inhibition relations by computer log-linear regression analysis (12).
RESULTS
The mean baseline tail-flick latencies when the radiation beam was focused on the marked portion of the tail was 2.93 ± 0.11 s. This baseline latency was not different from stimulation at the more proximal, unmarked segment of the tail (2.82 ± 0.23 s). Local intradermal injection of 0.9% saline did not produce any significant prolongation of tail-flick latencies at any time point during the 120-min observation period. However, the local intradermal administration of diclofenac (1, 3, 10, 30, and 50 µg) (Fig. 1A), dipyrone (1, 10, 30, 100, and 300 µg) (Fig. 2A), ketorolac (1, 10, 30, 100, and 300 µg) (Fig. 3A), lysine acetylsalicylate (1, 10, 30, 100, and 300 µg) (Fig. 4A), and sodium salicylate (1, 10, 30, 100, and 300 µg) (Fig. 5A) produced a significant dose-dependent prolongation of tail-flick latency. The maximal effect of diclofenac (50 µg), dipyrone, ketorolac, lysine acetylsalicylate (300 µg) after their intradermal administration was 5.67 ± 0.12, 4.65 ± 0.18, 4.47 ± 0.07, and 4.98 ± 0.21 s, respectively, at 30 min after drug injection. The maximal effect plateau extended 120 min postinjection (Figs. 1A, 2A, 3A, 4A, and 5A). There was no change in tail-flick latencies with these NSAID-treated mice when the radiant heat was focused on the unmarked proximal part of the drug injection sites during the 120-min period indicating local response (Figs. 1B, 2B, 3B, 4B, and 5B). From the dose-response curve generated at the time of peak effect (at +30 min), the ED50 and 95% confidence intervals of diclofenac, dipyrone, ketorolac, lysine acetyl salicylate, and sodium salicylate were found to be 10.3 (7.6–14.5), 143.8 (62.8–354.7), 228.2 (83.3–612.5), 131.9 (122.3–182.1), and 872.7 (552.3–1342.8), respectively (Fig. 6).
|
|
|
|
|
|
Naloxone (5 mg/kg, s.c.) when given alone did not affect tail-flick latencies in control animals (Fig. 7). The local antinociceptive effects of the highest dose of diclofenac (50 µg), ketorolac (300 µg), lysine acetylsalicylate (300 µg), and sodium salicylate (300 µg) were not changed by systemic pretreatment with naloxone (5 mg/kg, i.p.) at any time point (Fig. 7). However, the local antinociceptive effect of dipyrone (300 µg) was significantly blocked by a single pretreatment dose of naloxone (5 mg/kg, s.c.) (Fig. 8). The effect of 5 mg/kg naloxone was significant at 10, 20, 30, and 60 min with tail-flick latencies similar to pretreatment control values, However, antinociceptive responses remained the same at 60 and 120 min, most likely due to the limited duration of naloxone's effect. The second dose of naloxone (5 mg/kg) significantly inhibited the local antinociceptive effects of dipyrone (300 µg) at the 90 and 120 min time points (Fig. 8).
|
|
Because there was no evidence of recovery from the local antinociceptive effects of NSAIDs over the initial 120 min observation period, an additional study was conducted with the highest dose tested of each of the NSAIDs in an attempt to determine if the observed effects were reversible. The peripheral antinociception produced by 300 µg of dipyrone, ketorolac, lysine acetylsalicylate or sodium salicylate was diminished at 180 and 240 min, and completely absent 300 min, since the tail-flick latencies were not significantly different from those of the saline-treated group. The peripheral antinociception induced by sodium salicylate (300 µg) was significantly diminished at 180 min and completely absent at 240 min. In contrast, the antinociceptive effect of diclofenac (50 µg) was longer in duration, demonstrating a time-dependent reduction in tail-flick latencies 240, 300, and 360 min and a return to baseline responses at 480 min. Local intradermal injection of 0.9% saline did not produce any significant change of tail-flick latencies at any time point during the entire 480 min observation period (Fig. 9).
|
In other groups of mice, the i.p. injection of diclofenac (50 mg/kg), dipyrone (300 mg/kg), ketorolac (300 mg/kg), lysine acetylsalicylate (300 mg/kg), and sodium salicylate (300 mg/kg) did not change tail-flick latencies when compared to 0.9% saline-treated control animals (Fig. 10).
|
Tail skin specimens after intradermal injection of blue dye administered in the same manner as the saline or NSAID injections showed that most of the subcutaneous layers were stained by the blue dye (Fig. 11B). The 0.9% saline-injected sham control specimens (Fig. 11C) were indistinguishable from normal specimens (Fig. 11A), with no evidence of epidermal and vascular injury, inflammatory infiltration, and nerve fiber injury (Table 1). However, there was edematous enlargement of subcutaneous tissues after saline injection when compared to tissue from the sham control (Table 1 and Fig. 11C). The injection of diclofenac (50 µg), ketorolac (300 µg), dipyrone (300 µg), lysine acetylsalicylate (300 µg), and sodium salicylate (300 µg) did not show significant histopathological abnormalities at the injection sites when compared to tissue from animals injected with 0.9% saline (Table 1). However, there was evidence of edematous enlargement of the subcutaneous tissues from all NSAID-injected groups that was comparable to that observed with the 0.9% saline-injected group (Table 1) (Figs. 11D–H). As seen in the skin from control animals (Fig. 11A) or those injected with 0.9% saline (Fig. 11C), there was no apparent abnormality of the nerve bundles and no changes were seen in fiber density of the epidermal, subepidermal and subcutaneous regions of the skin of animals that received the NSAIDs (Figs. 11D–H).
|
|
DISCUSSION
The key finding of the present study is that NSAIDs administered locally produce antinociception in the tail-flick thermal assay. The same dose given systemically shows no effect. We suggest that the efficacy of locally administered NSAIDS is due to the larger concentrations achieved at the site of application of the thermal stimulus.
Historically, the tail-flick test has reliably predicted the antinociceptive activity of opioids. The insensitivity of this test to NSAIDs is widely accepted (1–3,6). Many studies have failed to observe antinociceptive activity for NSAIDs in this assay after systemic administration (2,3,6,13,14). Consistent with these previous studies, our study also failed to observe antinociceptive activity for systemically administered NSAIDs in the tail-flick test. However, the results of the present study strongly suggest that NSAIDs demonstrate a prominent localized antinociceptive effect at peripheral sites. Localized activity of NSAIDs is underscored by the observations that the local NSAIDs did not block nociception when the heat stimulus was applied to the surrounding region (i.e., a portion of the tail more proximal to the site stimulated), and relatively large i.p. doses of the NSAIDs were also without antinociceptive effect. Thus, considering the fact that local injection of NSAID into the skin could achieve a higher drug concentration at the injection sites when compared to systemic administration, one possible interpretation is that the amount of NSAIDs absorbed from the systemic circulation into the skin of the tail is insufficient to produce antinociception. Despite numerous studies reporting the local tissue disposition of NSAIDs after topical administration, the disposition of NSAIDs in local tissue after skin injection has rarely been investigated (15,16). However, it has been reported that the epidermal concentration of some NSAIDs after topical administration was about 200 times higher than that after systemic administration (15). Thus, it can be concluded that the antinociceptive effect of NSAIDs in the tail-flick test can only be observed after local administration, as the skin concentration of NSAIDs would be much higher than when administered systemically. It has been shown that most of the NSAIDs were retained in the skin after topical administration (15). Hence, the retaining of NSAIDs after our injection into the tail skin may explain our observation that intradermal NSAIDs-induced antinociception lasted beyond 300 min.
There is a possibility that high doses of NSAIDs could produce toxic effects at the site of administration, leading to false positive responses. We observed edema in the subcutaneous layer of skin at the site of injection of the NSAIDs and none in non-injected tail skin. However, the injection of the vehicle, 0.9% saline, also produced edema that was comparable to that produced by the NSAIDs, indicating that localized edema is a consequence of the volume of injection. No histopathological abnormalities were observed in the skin of animals injected with NSAIDs or saline vehicle. Thus, it is unlikely that a localized toxic effect of NSAIDs on the skin is responsible for the antinociceptive effects.
The mechanism of action of local antinociceptive effects of NSAISs in the tail-flick test remains to be clarified. Nociceptors activated by noxious heat belong to the vanilloid receptor-system (VR1). (17,18) The skin receives major projections from primary sensory neurons, and previous studies indicated that VR1 expression in skin is localized on the terminals of the intracutaneous epidermal nerve fibers that convey pain sensation evoked by noxious heat (18,19). Activation of VR1 receptors induces eicosanoid formation and releases other proinflammatory mediators (19). It is speculated that the production and release of proinflammatory mediators by secondary activation of the VR1 could act on terminals of sensory neurons in the skin and modulate nociceptive signaling (20). Indeed, prostaglandin-E2 is major prostaglandin in the skin and it enhances VR1-activated release of substance P from sensory neurons (21). Diclofenac, dipyrone, ketorolac, and lysine acetylsalicylate are nonselective inhibitors of COX-1 and COX-2, and inhibition of prostaglandin synthesis in the skin is likely to contribute to the antinociceptive effects of these drugs (22,23). In this study, it is possible that blockade of COX activity with locally administered diclofenac, dipyrone, ketorolac, and lysine acetylsalicylate significantly reduced the availability of prostaglandin in the tail skin, thus revealing a local role of prostaglandins by EP receptors to modulate the responsivity of VR1 through intracellular cascades in acute thermal nociception. In our study, diclofenac was the most potent and sodium salicylate was the least potent in the production of local antinociception. The reason for the difference in the antinociceptive potency may be related to the rank of potency of COX inhibition by diclofenac, dipyrone, lysine acetylsalicylate, ketorolac, and sodium salicylate (22). However, in our study, we were not able to measure COX expression at the site of injection before and after NSAIDs administration. Additionally, COX-1 is constitutively expressed, whereas COX-2 is not normally expressed in the skin (23). In an interesting study, dexamethasone, in contrast to ketorolac, lacked analgesic efficacy in spite of causing a reduction in peripheral prostaglandin-E2 levels through COX-1 inhibition, suggesting that mechanisms other than inhibition of prostaglandin synthesis may be involved in the antinociceptive effects of peripheral NSAIDs (24). In addition, NSAIDs inhibit COX activity at much lower doses than those required to produce peripheral antinociception in our study, with the exception of diclofenac. Thus, it remains to be clarified if other mechanisms besides COX inhibition contribute to the peripheral antinociceptive effects of NSAIDs in the tail-flick test.
It has been suggested that NSAIDs produce their analgesic effect, in part, through other mechanisms independent of suppression of peripheral COX inhibition (5,8). Consistent with this assumption, we observed that naloxone significantly inhibited the local antinociceptive effects of dipyrone. The pretreatment with a single systemic dose of naloxone (5 mg/kg) blocked the peripheral antinociceptive effects of dipyrone for nearly 60 min. The addition of a second dose of naloxone 30 min after dipyrone injection fully blocked the peripheral antinociceptive effects of dipyrone for 120 min. Temporary blockade of dipyrone-induced antinociception with a single dose of naloxone pretreatment is most likely due to the limited duration of naloxone's effect. The data suggest that the local analgesic pathways involving the opioid system are activated after dipyrone injection. It has been suggested that opioid receptors located on nociceptor terminals contribute to peripheral endogenous antinociception mediated by ß-endorphin secreted from keratinocytes in response to triggering events (25). It is possible that locally administered dipyrone might trigger keratinocytes to release ß-endorphin, thus inhibiting nociception. Unlike dipyrone, the lack of antagonistic effects of naloxone against diclofenac, ketorolac, lysine acetylsalicylate, and sodium salicylate suggest that the opioid system does not contribute to the antinociceptive mechanisms of these NSAIDs. These results suggest that there may be differences in the mechanisms of peripheral antinociceptive action among the NSAIDs. Moreover, it has been reported that some NSAIDs induce peripheral antinociceptive effects through the activating of l-arginine-NO-cGMP pathways (26). Thus, it is possible that l-arginine-NO-cGMP pathways are contributing to the peripheral antinociceptive effects of diclofenac, ketorolac, lysine acetylsalicylate, and sodium salicylate.
Although there is a possibility that NSAIDs might exert a moderate local anesthetic, electrophysiological studies comparing the local effect of NSAIDs with diltiazem on evoked responses of corneal sensory fibers of cats found that, whereas NSAIDs have a mild local anesthetic effect, attenuation of sensitization of sensory fibers caused by prostaglandin release accounted for a significant component of the antinociceptive effect of topically applied NSAIDs (27). High concentrations of local anesthetics have been shown to be neurotoxic to isolated dorsal root ganglion neurons (28), but a neurotoxic effect is unlikely in the present investigation, since the effect of the locally injected NSAIDs were reversible within 4 h and nociceptive responses returned to baseline levels. Additionally, neurotoxic or cytotoxic effects of NSAIDs are unlikely to underlie the antinociceptive effects observed in the present investigation, since these drugs are used in ophthalmic preparations for topical application for treatment of ocular pain after surgery or injury, and do not produce signs of epithelial damage in studies performed with volunteers (29–34).
In summary, these results show that NSAIDs, when given locally, are effective antinociceptive drugs in the radiant tail-flick test. The data support further research into the antinociceptive actions of locally applied NSAIDs, a route of delivery that might preclude the toxicity associated with systemic NSAID administration. Activation of the endogenous opioid system may be involved in the peripheral antinociceptive effects of dipyrone, but not diclofenac, ketorolac, and lysin asetyl salicylate, suggesting differences in the mechanisms of action among the NSAIDs.
Footnotes
Accepted for publication December 20, 2006.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
