- Email: [email protected]
The formalin test in mice: dissociation between inflammatory and non-inflammatory pain
103
Pain, 30 (1987) 103-114 Elsevier
PAI 01057
The formalin test in mice: dissociation between inflcimm2tnrv 5mt-l ___- non-infl2mmrttnrv __V__““‘-“‘““““-““J “““““‘““‘“‘““J Steinar Hunskaar
twin r--‘-
and Kjell Hole
Department of Physiology, University of Bergen, N-5009 Bergen (Norway)
(Received 9 September 1986, accepted 1 December 1986)
The formahn test in mice is a vahd and reliable model of nociception and is sensitive for various classes of analgesic drugs. The noxious stimulus is an injection of dilute formahn (1% in saline) WI&~ c~wfmv= nf hidnnaw * n.e ~e~nnnre ic the nf rl_.“- __ .._”-amnnnt .___^_. __time _*.._the _._-animsllc 1.1_.. -_ “_..“. the s&n of the . .._ dnrc~l -_.,.“- ..-.*..“_ the . ..- rirrht ..CT-..-*-r..” spend lick.ing the injected paw. Two distinct periods of high licking activity can be identified, an early phase lasting the first 5 mm and a late phase lasting from 20 to 30 mm after the injection of formalin. In order to elucidate the involvement of inflammatory processes in the two phases, we tested different classes of drugs in the two phases ind~endently. Morphine, codeine, nefopam, and orphenadrine, as examples of centrally acting analgesics, were antinociceptive in both phases. In contrast, the non-steroid anti-inflammatory drugs indomethacin and naproxen and the steroids dexamethasone and hydrocortisone inhibited only the late phase, while acetylsalicylic acid (ASA) and paracetamol were antinociceptive in both phases. The results demonstrate that the two phases in the formalin test may have different nociceptive mechanisms. It is suggested that the early phase is due to a direct effect on nociceptors and that prostaglandins do not play an important role during this phase. The late phase seems to be an infla~ato~ response with infl~ato~ pain that can be inhibited by anti-infla~ato~ drugs. ASA and paracetamol seem to have actions independent of their i~bition of prostaglandin synthesis and they _1__nave -L_.._CIICCLS _#C--.- on _- n”n-lnmlmmarory .__- f_~-__.._.r-_. pam. _-I. LtlS”
S-=Y
Key words: Pam test; Mice; Infl~ation
A large number of behavioural methods have been developed in order to study nociception and the action of various analgesic drugs in animals. Most of these tests readily reveal the activity of narcotic analgesics, but are insensitive to mild, non-narcotic antinociceptive drugs [24:46:56]. Attempts to demonstrate the antinociceptive effect of acetyls~icylic acid (ASA) and related drugs by applying the painful stimulus to intact tissue, have generally been unsuccessful, whereas techCorrespondence to: Steinar Hunskaar, Arstadveien 19, N-5009 Bergen, Norway.
03~-39S9/87/$03.50
M.D., Department
of Physiology, University
0 1987 Elsevier Science Publishers B.V. (Biomedical Division)
of Bergen,
104
niques involving inflammation show analgesic activity of these drugs [20,34,38,55]. The inflammatory and non-inflammatory pain are by some authors suggested to represent different physiological entities. The inflammatory pain is thought to be selectively inhibited by the non-steroid anti-inflammatory drugs (NSAIDs) and steroids because they reduce inflammation, and not because of any specific antinociceptive action [9,53,54]. However, an increasing amount of data supports the possibility that NSAIDs have antinociceptive actions independent of their anti-inflammatory effects [5,7,8,24,25,27,34]. Hyperalgesia, an essential feature of inflammatory pain, consists of two different components which involve the action of prostaglandins: the sensitization of local, peripheral receptors, and a less well documented central effect on pathways mediating nociception [13]. There are, however, many other substances that are released in acute inflammation, such as histamine, serotonin and bradykinin. Data from the carrageenan induced inflammation in rats, indicate distinct phases in the inflammatory process [10,40,50]; an initial phase where histamine and serotonin are released, followed by kinins, lasting for some 30-60 min, and then a phase mediated by prostaglandins. Thus prostaglandins act relatively late in the development of inflammation, although prostacyclin (PGI,) induces an early and short-lived hyperalgesia [14]. It is well established that NSAIDs prevent the hyperalgesia of inflammation by blocking the prostaglandin synthesis [15,51]. The formalin test in mice [24] is sensitive to NSAIDs and other mild analgesics. The test employs a chemical nociceptive stimulus that elicits a spontaneous response indicative of pain. The test has two different phases, possibly reflecting different types of pain [11,24,27]. As inflammation occurs at the site of formalin injection it should be possible to elucidate the role of inflammation on the responses in the two phases. We have previously shown that ASA and indomethacin are antinociceptive through partially different modes of action in this test [27]. We have also shown that ASA does not have any delay of onset of its action in the early phase compared to morphine [23]. Both of these studies suggested that ASA has some effects which cannot be attributed to the inhibition of prostaglandin synthesis alone. In the present study, 3 groups of drugs with well-established effects were used to study the two phases of the formalin test: (1) Standard narcotic analgesics (morphine and codeine) [28] and centrally acting non-narcotic analgesics without known mode of action (nefopam and orphenadrine) [21,22]. (2) Prostaglandin synthesis inhibitors (ASA, paracetamol, indomethacin and naproxen) [6,16,17,39]. (3) Anti-inflammatory steroids (hydrocortisone and dexamethasone) [43,46] which inhibit inflammation through a variety of actions including the inhibition of prostaglandin synthesis. Determinations of serum concentrations of ASA, paracetamol and morphine after antinociceptive doses of each drug were included in the study. Methods
Animals Male albino mice (Born: NMRI, Msllegird, Denmark) weighing 25-35 g were used in the experiments. The animals were housed in colony cages (15-16 mice in
105
each) with free access to food and water prior to the experiments. They were maintained in climate- and light-controlled rooms (23 f 0.5 ’ C,12/12 h dark/light cycle with lights on at 07:OO) for at least 2 weeks prior to the experiments. Testing took place during the first hours of the light phase. The mice were brought to the test room the day before testing, and allowed at least 18 h to adapt to the testing environment. Drugs and administration routes
All injections were made intraperitoneally (i.p.) in a volume of 10 ml/kg. In all experiments an equal volume of vehicle was used as control for the injections. The drugs were injected 30 min prior to testing unless otherwise stated. Acetylsalicylic acid (Svaneapoteket, Bergen) and indomethacin (Merck, Sharp and Dohme) were dissolved in Tris buffer 0.1 M, pH 7.4-7.6. Morphine hydrochloride (NAF-Lab A.S.), codeine phosphate (Svaneapoteket, Bergen), nefopam hydrochloride (3M Biker Lab., Inc.), orphenadrine citrate (3M Biker Lab., Inc.), naproxen (Astra), and dexamethasone 21-phosphate injectable (Merck, Sharp and Dohme) were dissolved in 0.9% NaCl. Paracetamol (Svaneapoteket, Bergen), and hydrocortisone (Norsk Medisinaldepot) were dissolved in 12.5% 1,2-propanediol in 0.9% NaCl. Control experiments did not show any altered responses due to the vehicles alone. Formalin test
Separate groups of animals were used for the different experiments. Each mouse was used on one occasion only. Testing or data recording were performed by an observer unaware of the drug treatment in each experiment. The formalin test [11,24] was performed in mice that had been individually exposed to the observation chamber (macrolone cage, 30 cm x 12 cm x 13 cm) for 2 h. Using a minimum of restraint, 20 ~1 of 1% formalin in 0.9% NaCl was injected subcutaneously into the dorsal hind paw of the mouse using a microsyringe with a 26-gauge needle. The mouse was then put back into the chamber and the observation period started. The amount of time the animal spent licking the injected paw or leg was recorded. On the basis of the response pattern described elsewhere [24], two distinct periods of intensive licking activity were identified and scored separately unless otherwise stated. The first period (early phase) was recorded O-5 min after the injection of formalin and the second period (late phase) was recorded 20-30 min after the injection. Thus, when testing the late phase, formalin was injected 10 min after the injection of the drug in order to obtain the same period of time (30 min) between injection of drug and testing. In all experiments attention was paid to the ethical guidelines for investigations of experimental pain in conscious animals [57], and the procedures were approved by the Norwegian Committee for Experiments on Animals. Determination of serum concentrations of analgesic drugs
Mice were given i.p. morphine (5 and 10 mg/kg), acetylsalicylic acid or paracetamol (200 and 400 mg/kg), and blood samples were obtained after 30 min by puncture of the heart during combined pentobarbital and chloral hydrate anaesthesia. The serum concentration of morphine was determined with HPLC and electrochem-
106
ical detection by the method of Bolander et al. [l] with small modifications. The samples were extracted twice with CHCl, and run on a 3 pm C,, column without an internal standard. A phosphate buffer (pH 6.2) containing 10% ethanol was used to increase the retention time. The serum salicylate concentration was determined by calorimetry after complex formation with ferric nitrite (Du Pont ACA, Wilmington, DE) [48]. Paracetamol was measured in serum using enzyme multiplied immunoassay technique (EMIT, Syva Co., Palo Alto, CA) [33]. Statistical analysis
The data were examined by analysis of variance (ANOVA) and Dunnett’s procedure for multiple comparisons with a single control group. When the analysis was restricted to two means, Student’s t test (two-tailed) was used. Level of significance was set to 5% (P < 0.05). Results are given as mean f S.E.M. Results The antinociceptive effects of different classes of drugs in the early and late phases of the formalin test
The centrally acting analgesic drugs morphine, codeine, nefopam, and orphenadrine all showed dose-dependent antinociceptive effects during both the early and
Drug. Dose: (mg/kgl
\ F** F 111’ Morphtne
*0
1
0
5
‘i
10 l
l
*
Codeine
0
25 50
Nefopam
0
10 20
Orphenadnne 0
10 20
08
*
l
l.
l
160J
I,i
Fig. 1. Antinociceptive effects of intraperitoneally administered morphine, codeine, nefopam, or orphenadrine in the formalin test. Nociceptive behaviour in the early phase (O-S mm after the injection of formalin) and the late phase (20-30 min after the injection of formalin) was scored as the amount of time spent licking the injected hind paw or leg (xc, mean f S.E.M.). Hatched columns indicate a statistically significant dose-response relationship. Significant differences between vehicle and drug treated groups arc indicated by * (P < 0.05) or * * (P < 0.01) (Dunnett’s test subsequent to ANOVA). The statistical calculations for the early (ER) and late (LR) response for dose-response relationships were as follows 43,31, (ANOVA): morphine, P < o,ool (ER), FF 2.21 = 41.71, P < 0.001 (RR), &,a, =L5.55, P < 0.001 (LR): a,,s =17.08, P -C 0.001 (LR); nefopam, F2,2,= 10.38, P < 0.001 codeine, F2,*,= F2,35 = 24.17,P < 0.001 (ER), F2,+ = 6.90, P < 0.005 (RR). Fa.20 = 22.53, P < 0.001(LR); orphenadrine, (LR).
Drug:
A
Dose:
v
ASA
Paracetamol
lndomethacln
n
Naproxen
Fig. 2. Antinociceptive effects of intraperitoneally administered acetylsalicylic acid (ASA), paracetamol, indomethacin, or naproxen in the formalin test. Nociceptive behaviour in the early phase (O-5 mm after the injection of formalin) and the late phase (20-30 mitt after the injection of formalin) was scored as the amount of time spent licking the injected hind paw or leg (XC,mean f S.E.M.). Hatched columns indicate a statistically significant dose-response relationship. Significant differences between vehicle and drug treated groups are indicated by * (P < 0.05) or * * (P < 0.01) (Dunnett’s test subsequent to ANOVA). The statistical calculations for the early (ER) and late (LR) response for dose-response relationships were as follows (ANOVA): ASA, FzT12 = 32.95, P < 0.001 (ER), F2,2, =19.52, P < 0.001 = (LR); paracetamol, F2,21 = 37.10, P -c0.001 (ER), F2,21 = 83.64, P < 0.001 (LR); indomethacin, F,.,, 0.63, P > 0.5 (ER), F2,*,, = 18.97, P-z0.001 (LR); naproxen, F2,,9 = 0.02, P > 0.9 (ER), F,.,, = 20.51, P c 0.001 (LR).
late phases (Fig. 1, statistics shown in the figure legend). There was no apparent difference in the effect of each drug in the early phase versus the late phase. No motor, neurological, or other behavioural deficits were observed. Fig. 2 shows the effects of 4 different NSAIDs. In doses without overt toxic effects, clear dose-dependent antinociception was found during the late phase for all the drugs In addition, ASA and paracetamol both significantly suppressed the licking activity in the early phase, while no effect was found for indomethacin and naproxen (statistics shown in the figure legend). The steroids hydrocortisone and dexamethasone both significantly suppressed the licking activity in the late phase, but were without effect in the early phase (Fig. 3, statistics shown in the figure legend). Possible delayed effects of such drugs were investigated by injecting dexamethasone (10 mg/kg) 3 h before testing. This treatment did not affect the response to formalin in the early phase of the test. Dexamethasone reduced the licking activity in the late phase with 40% compared to controls, but the difference did not reach statistical significance (tl,lS = 1.34, P = 0.20) (Fig. 3). Analysis of the time course of the effect of indomethacin Indomethacin and naproxen (NSAIDs) and the steroids hydrocortisone
and
Hydrocortlsone
Drug b
Dose. (mgikgi
Dexamethasane
( 3 h)
Dexamethasone
* P
O-
l
160-
Fig. 3. Antinociceptive effects of intraperitoneally administered hydrocortisone and dexamethasone in the formalin test. Nociceptive behaviour in the early phase (O-5 mm after the injection of formalin) and the late phase (20-30 min after the injection of formalin) was scored as the amount of time spent licking the injected hind paw or leg (set, mean + S.E.M.). Dexamethasone treated groups (10 mg/kg) were also tested 3 h after drug injection and compared to similar vehicle treated groups. The animals were maintained in the observation chambers during these 3 h. Hatched columns indicate a statistically significant dose-response relationship. Significant differences between vehicle and drug treated groups are indicated by * * (P -c0.01) (Dunnett’s test subsequent to ANOVA). The statistical calculations for the early (ER) and late (LR) response for dose-response relationships were as follows (ANOVA): = 8.51, P -c0.001 (LR); dexamethasone, F,.,, = 0.50, hydrocortisone, F,, ,6= 1.33, P > 0.2 (ER), F2,42 8.78, P < 0.002 (LR); dexamethasone 3 h after injection, Pr.,s = 0.20, P > 0.3 P > 0.6 (ER), F, ,,, = _,_, (ER), F,,,5 = 1.79, P > 0.05(LR). Earlyphase ‘1
I’
Late phase I,
Cl
^
1
Early phase
Late phase
m
/4
;; I
0 Vehicle 500-
c
. lndomethocln
2
oI o-1 2-3
4-5
O-5
10-15
20-25
Minutes
30-35
after inlectlon of formalin
Fig. 4. Time course of antinociceptive effects of intraperitoneally administered indomethacin (30 mg/kg) or vehicle in the formalin test. Nociceptive behaviour in the early (O-S mitt after the injection of formalin) and the late phase (20-30 min after the injection of formalin) was scored as the amount of time spent licking the injected hind paw (set, mean + S.E.M.). A: pain intensity score during the first 5 mm recorded in 1 mm periods. B: pain intensity score during the first 40 mm after the injection of formalin. Drugs were injected 30 mm before injection of formalin. Significant differences between drug and vehicle treated groups calculated by means of Student’s t test for pairs of data from each 5 mitt period, are indicated by * (P < 0.05) or * * (P < 0.01). C: accumulated response during the first 40 mm (5 mm periods) after the injection of formalin. Statistically significant differences between the two groups are * * * (P < 0.001) (Student’s r test calculated on concomitant pairs). indicated by ** (P
109
dexamethasone all showed a similar pattern of effects in the two phases of the formalin test (Figs. 2 and 3), and preliminary studies indicated that the time course of the effects also was similar for these drugs (data not shown). Indomethacin (30 mg/kg) was therefore used in a more detailed experiment of the time course (Fig. 4). Again, no statistically significant difference was found in the total amount of licking during the early phase between the indomethacin and vehicle treated groups (t 1 12< 0.1, P > 0.7). In order to detect possible differences during this period, each 1 min period was compared in the two groups, but no statistically significant difference was found ( F1,i8 < 0.1, P > 0.7 for overall effect, F4,72= 16.38, P < 0.001 for trial effect and F4,72< 0.1, P > 0.9 for interaction, ANOVA) (Fig. 4A). Time course (O-40 min) of paw licking response after treatment with indomethacin or vehicle is shown in Fig. 4B. ANOVA revealed statistically significant differences between the two groups (Fl,12 = 16.82, P < 0.002 for overall effect, F 7,84= 15.92, P < 0.001 for trial effect and F,,+, = 3.18, P < 0.01 for interaction). In contrast to the early phase, the results from the late phase were significantly different as shown by one-way ANOVA calculated on the two 5 min periods included in the late phase (20-25 and 25-30 rnin) ( F,,12 = 11.10, P -c 0.01 for overall effect, Fl, 12= 1.38, P > 0.2 for trial effect, and Fl,12 < 0.1, P > 0.7 for interaction). In the periods O-15 min and 25-40 min no statistically significant differences were found between the drug and vehicle. When the accumulated response to the nociceptive stimulus was analysed by Student’s t test for each pair of data points, the difference between the two groups reached statistical significance in the fourth 5 min period (15-20 min after formalin injection) (Fig. 4C). Serum concentrations of morphine, acetylsalicylic acid and paracetamol
Determinations of the serum concentrations (Table I) were analysed 30 min after injection, the same interval as between injection and behavioural testing. Preliminary studies indicated that the peak concentrations also occurred at that time (data not shown).
TABLE I SERUM CONCENTRATIONS MOL IN MICE
OF MORPHINE,
ACETYLSALICYLIC
ACID AND PARACETA-
Drug
Dose (mg/kg i.p.)
No. of animals
Serum concentration (mean f S.E.M.)
Morphine
5 10 200 400 200 400
3 4 4 4 4 4
115k16 pg/l 578 f 65 pg/I 200 f 37 mg/l 314 f 79 mg/I 102* 8 mg/l 253 f 14 mg/l
Acetylsalicylic acid Paracetamol
110
Discussion The present study demonstrates that the early and late nociceptive phases of the formalin test in mice are dissimilar. The various effects of 3 classes of analgesic drugs, tested separately on the two phases, indicate that the early phase may be due to direct effects on nociceptors; this phase can be inhibited by centrally acting analgesics. In contrast, the late phase seems to be due to an inflammatory response partly mediated by prostaglandins and can be inhibited by NSAIDs and steroids, as well as the centrally acting drugs. The drug doses used are within the range used by others in comparable studies [24,32,37,39]. Serum concentrations of morphine, ASA, and paracetamol in nearly equipotent antinociceptive doses were found to be on the order of high normal to subtoxic levels for humans. Serum levels of paracetamol (400 mg/kg) were relatively high compared to human data for therapeutic doses [17]. Other studies in animals have also shown that such doses are necessary to induce antinociceptive effects in rodents [19,24,26,35,37,53]. For ASA, the serum levels found here are below the toxic range 30 min after administration in humans [17], and also in accordance with another study in mice [31]. The serum level of morphine found after 5 mg/kg is comparable to human values measured after 10 mg administered intravenously [44]. It may be concluded that compared to human data, moderate to high serum levels of analgesics are required to induce antinociception in the formalin test in mice. The difference between effective human and animal serum level appears to be greatest for morphine. Subcutaneous formalin injection initiates a series of events with an essentially uniform course [24]. This model (as well as the Randall-Selitto test and carrageenan induced inflammation test, for example) reproduces various aspects of acute inflammation, but should not be considered as model for chronic inflammatory responses which often involve immune reactions to antigenic substances. The early phase of the formalin test is of particular interest as the differences between the drugs appeared here. Indomethacin, naproxen and the steroids were without effect in this phase. This argues against an important role of prostaglandins or prostacyclin in the nociceptive responses during the first minutes in this test. ASA has been shown to induce antinociception as early as 30 set after the injection of formalin; the same latency as for morphine [23]. A possible dissociation between analgesic and anti-inflammatory effects of ASA has also been shown in other models [l&41]. From the experiments on the mediators of carrageenan oedema, distinct phases have been demonstrated. ASA inhibits oedema formation in all phases, while all the steroids and other NSAIDs tested inhibit oedema formation only after some time [52]. Indomethacin and dexamethasone failed to suppress the oedema in rats deprived of endogenous prostaglandin precursors, while ASA fully maintained its effect even under these conditions [2,3,27]. Okuyama and Aihara [34] studied NSAIDs in theadjuvant arthritic rat model and found that much smaller doses were needed for antinociception than in normal rats. These authors reported EDso ratios between 1.7 and 1.8 for narcotic analgesics while indomethacin and naproxen had
111
values of 25.7 and 37.4 respectively. Interestingly, ASA and paracetamol had EDs0 values much lower than the other NSAIDs, indicating effects on non-inflammatory pain. Paracetamol was included among the NSAIDs in this study although it has been claimed to have only weak anti-inflammatory effects [17], and not to inhibit prostaglandin synthesis in the periphery [16]. Recent studies in animals demonstrated that also the inhibition of brain prostaglandin synthesis is relatively weak [4,47], and that paracetamol inhibits carrageenan induced oedema in rats in doses with antinociceptive effects [19,53]. The anti-inflammatory effects of steroids are due to several mechanisms, including reduced vascular permeability, exudation, vasodilation, leucocyte emigration, phagocytosis and release of lysosomal enzymes. The current view is that the steroids exert these effects by controlling the rate of synthesis of regulatory proteins. This explains the lag time necessary for the effects in some models [32,49]. Moreover, several authors have demonstrated inhibition of prostaglandin synthesis by steroids [29,30]. The effect is not mediated via the cycle-oxygenase reaction. These observations could indicate that steroids may have analgesic activities comparable to NSAIDs. Some clinical data support the view [42,43], but little experimental evidence for this exists [32]. In the present experiments we postulated that an inhibitory effect on behavioural responses would be due to anti-inflammatory effects and subsequently reduced symptoms of inflammation, including nociception. Both dexamethasone and hydrocortisone significantly suppressed the responses in the late phase. Relatively high doses were used, but in order to suppress the licking by 75% it was presumably necessary to almost eliminate the inflammatory reaction. Injection of dexamethasone 3 h before testing did not yield significant effects on licking, but the tendency was clearly in the direction of a sustained effect in the late phase. There was low licking activity in the control group in this experiment. This was likely an effect of adaptation to the observation chamber, as the late phase is more prone to adaptation than the early phase with its intensive licking activity. It may be concluded that the inhibitory effects of steroids on the late phase indicate that this phase is an inflammatory response with subsequent inflammatory pain. The centrally acting analgesics used in the present study suppressed the responses in the two phases in a similar manner. This may suggest a common way of reducing inflammatory and non-inflammatory pain by these drugs. Despite this, recent studies have indicated that the endogenous opioid-peptidergic and serotonergic systems modulate the early and late phases differently [12,45]. The various effects of the NSAIDs may be due to the fact that ASA and paracetamol have central actions while indomethacin and naproxen do not. However, most NSAIDs have been shown to reduce carrageenan evoked hyperalgesia when given intracerebroventricularly while morphine also reduces non-inflammatory pain [13]. Systemically administered ASA or paracetamol have been shown to inhibit various non-inflammatory nociceptive responses [24-26,341. They inhibit pain related behaviour induced by substance P injected intrathecally or substance P released by intrathecal administration of capsaicin, indicating a central antinociceptive effect of these drugs [25]. Our hypothesis is that the antinociceptive effects of
112
ASA and paracetamol in the early phase of the formalin test are due to a central action, which is probably not related to the inhibition of prostaglandin synthesis. Furthermore, the central effects of some NSAIDs do not exclude the possibility that the late phase is inflammatory and prostaglandin-dependent in nature. In conclusion, the two phases of the formalin test in mice seem to be different, involving different mechanisms of nociception [11,12,24,45]. The data support the view of Dubuisson and Dennis [ll] that the first phase is due to a direct effect of formalin on nociceptors and the second due to inflammation. These findings may significantly extend the use of the test as a tool for studying the processes underlying pain and mechanisms of old and new analgesic drugs.
The authors thank Ms. Anita Kloster and Ms. Trine Tydal for excellent technical assistance. Morphine was analysed by the courtesy of Prof. Ole J. Broth and paracetamol and acetylsalicylic acid by Prof. Olav M. Bakke at the Department of Pharmacology and Toxicology, University of Bergen. Drugs were generously supplied by Merck, Sharp and Dohme (indomethacin), Astra-Syntex A.S. (naproxen), 3M Piker Laboratories, Inc. (nefopam and orphenadrine). The financial support from Norske Kvinners Sanitetsforening is gratefully acknowledged.
References 1 Bolander, H., Kourtopoulos, H., Lundberg, S. and Person, S.-A., Morphine concentrations in serum, brain and cerebrospinal fluid in the rat after intravenous administration of a single dose, J. Pharm. Pharmacol., 35 (1983) 656-659. 2 Bonta, IL., Bult, H., v.d. Ven, L.L.M. and Noordhoek, J., Essential fatty acid deficiency; a condition to discriminate prostaglandin and non-prostaglandin mediated components of inflammation, Agents Actions, 6 (1976) 154-158. 3 Bonta, I.L., Bult, H., Vincent, J.E. and Zijlstra, F.J., Acute anti-inflammatory effects of aspirin and dexamethasone in rats deprived of endogenous prostaglandin precursors, J. Pharm. Pharmacol., 29 (1977) l-7. 4 Bruchhausen, F. and Baumann, J., Inhibitory actions of desacetylation products of phenacetin and paracetamol on prostaglandin synthetase in neuronal and glial cell lines and rat renal medulla, Life Sci., 30 (1982) 1783-1791. 5 Brune, K., Bucher, K. and Walz, D., The avian microcrystal arthritis. II. Central versus peripheral effects of sodium salicylate, acetaminophen and colchicine, Agents Actions, 4 (1974) 27-33. 6 Brune, K., Glatt, M. and Graf, P., Mechanisms of action of anti-inflammatory drugs, Gen. Pharmacol., 7 (1976) 27-33. 7 Capetola, R.J., Sbriver, D.A. and Rosenthale, M.E., Suprofen, a new peripheral analgesic, J. Pharmacol. exp. Ther., 214 (1980) 16-23. 8 Capetola, R.J., Rosenthale, M.E., Dubinsky, B. and McGuire, J.L., Peripheral antialgesics: a review, J. clin. Pharmacol., 23 (1983) 545-556. 9 Crepax, P. and Silvestrini, B., Experimental evaluation in laboratory animals of anti-inflammatory analgesic drugs, Arch. ital. Biol., 101 (1963) 444-457.
113 10 Di Rosa, M., Giroud, J.P. and Willoughby, D.A., Studies of the mediators of the acute inflammatory response induced in rats in different sites by carrageenan and turpentine, J. Pathol., 104 (1971) 15-29. 11 Dubuisson, D. and Dennis, S.G., The formahn test: a quantitative study of the analgesic effects of morphine, meperidine, and brain stem stimulation in rats and cats, Pam, 4 (1977) 161-174. 12 Fasmer, O.B., Berge, O.G. and Hole, K., Changes in nociception after lesions of descending serotonergic pathways induced with 5,6dihydroxytryptamine. Different effects in the formalin and tail-flick tests, Neuropharmacology, 24 (1985) 729-734. 13 Ferreha, S.H., Lorenzetti, B.B. and Correa, F.M.A., Central and peripheral antialgesic action of aspirin-like drugs, Europ. J. Pharmacol., 53 (1978) 39-48. 14 Ferreira, S.H., Nakamura, M. and Castro, M.S.A., The hyperalgesic effects of prostacyclin and prostaglandin E2, Prostaglandms, 16 (1978) 31-37. 15 Flower, R.J., Drugs which inhibit prostaglandin biosynthesis, Pharmacol. Rev., 26 (1974) 161-174. 16 Flower, R.J. and Vane, J.R., Inhibition of prostaglandin synthetase in brain explains the anti-pyretic activity of paracetamol (Cacetamidophenol), Nature (Lond.), 240 (1972) 410-411. 17 Flower, R.J., Moncada, S. and Vane, J.R., Analgesic-antipyretics and anti-inflammatory agents. Drugs employed in the treatment of gout. In: A. Goodman Giiman, L.S. Goodman and A. Gilman (Eds.), The Pharmacological Basis of Therapeutics, MacMillan, New York, 1980, pp. 682-728. 18 Gilfoil, T.M., Klavins, I. and Grumbach, L., Effects of acetylsalicylic acid on the edema and hyperesthesia of the experimental inflamed rat’s paw, J. Pharmacol. exp. Ther., 142 (1963) 1-5. 19 Glenn, E.M., Bowman, B.J. and Rohloff, N.A., Anti-inflammatory and PG inhibitory effects of phenacetin and acetaminophen, Agents Actions, 7 (1977) 513-516. 20 Guilbaud, G. and Iggo, A., The effect of lysine acetylsalicylate on joint capsule mechanoreceptors in rats with polyarthritis, Exp. Brain Res., 61 (1986) 164-168. 21 Heel, R.C., Brogden, R.N., Pakes, G.E., Speight, T.M. and Avery, G.S., Nefopam, a review of its pharmacological properties and therapeutic efficacy, Drugs, 19 (1980) 249-267. 22 Hunskaar, S., Antinociceptive effects of orphenadrine citrate in mice, Europ. J. Pharmacol., 111 (1985) 221-226. 23 Hunskaar, S., Similar effects of acetylsalicylic acid and morphine on immediate responses to acute noxious stimulation, Pharmacol. Toxicol., 60 (1987) 167-170. 24 Hunskaar, S., Fasmer, O.B. and Hole, K., Formalin test in mice, a useful technique for evaluating mild analgesics, J. neurosci. Methods, 14 (1985) 69-76. 25 Hunskaar, S., Fasmer, O.B. and Hole, K., Acetylsalicylic acid, paracetamol and morphine inhibit behavioural responses to intrathecally administered substance P or capsaicin, Life Sci., 37 (1985) 1835-1841. 26 Hunskaar, S., Berge, O.-G. and Hole, K., A modified hot-plate test sensitive to mild analgesics, Behav. Brain Res., 21 (1986) 101-108. 27 Hunskaar, S., Berge, O.-G. and Hole, K., Dissociation between antinociceptive and anti-inflammatory effects of acetylsalicylic acid and indomethacin in the formalin test, Pain, 25 (1986) 125-132. 28 Jaffe, J.H. and Martin, W.R., Gpioid analgesics and antagonists. In: A. Goodman Gilman, L.S. Goodman and A. Gilman (Eds.), The Pharmacological Basis of Therapeutics, MacMillan, New York, 1980, pp. 494-534. 29 Kantrowitz, F., Robinson, D.R., McGuire, M.B. and Levine, L., Corticosteroids inhibit prostaglandin production by rheumatoid synovia, Nature (Lond.), 258 (1975) 737-739. 30 Lewis, G.P. and Piper, P.J., Inhibition of release of prostaglandins as an explanation of some of the action of antiinflammatory corticosteroids, Nature (Lond.), 254 (1975) 308-311. 31 Ljungberg, S., Otto, U. and Paalzow, L., Analgesic activity of acetylsalicylic and salicylic acid and its relation to blood concentrations in mice, Acta pharm. suet., 5 (1968) 489-500. 32 Nakamura, H., Mizushima, Y., Seto, Y., Motoyoshi, S. and Kadokawa, T., Dexamethasone fails to produce antipyretic and analgesic actions in experimental animals, Agents Actions, 16 (1985) 542-547. 33 Oellerich, M., Enzyme immunoassays in clinical chemistry. Present status and trends, J. clin. Chem. clin. B&hem., 18 (1980) 197-208. 34 Okuyama, S. and Aihara, H., Inhibition of electrically-induced vocalization in adjuvant arthritic rats as a novel method for evaluating analgesic drugs, Jap. J. Pharmacol., 34 (1984) 67-77.
114 35 Okuyama, S. and Aihara, H., The site of action of morphine and indomethacin differs with electrical stimulation of cutaneous or tibia1 nerves in normal and adjuvant arthritic rats, Arch. int. Pharmacodyn., 276 (1985) 133-141. 36 Okuyama, S. and Aihara, H., Hyperalgesic action in rats of intracerebroventricularly administered arachidonic acid, PG E2 and PG F2a: effects of analgesic drugs on hyperalgesia, Arch. int. Pharmacodyn., 278 (1985) 13-22. 37 Pong, SF., Demuth, S.M., Kinney, C.M. and Deegan, P., Prediction of human analgesic dosages of nonsteroidal anti-inflammatory drugs (NSAIDs) from analgesic ED50 values in mice, Arch. int. Pharmacodyn., 273 (1985) 212-220. 38 Randall, L.O. and Selitto, J.J., A method for measurement of analgesic activity on inflamed tissue, Arch. int. Pharmacodyn., 111 (1957) 409-419. 39 Roszkowski, A.P., Rooks II, W.H., Tomolonis, A.J. and Miller, L.M., Anti-inflammatory and analgetic properties of D-2-(6’-methoxy-2’-naphthyl)-propionic acid (naproxen), J. Pharmacol. exp. Ther., 179 (1971) 114-123. 40 Sedgwick, A.D. and Willoughby, D.A., Initiation of the inflammatory response and its prevention. In: I.L. Bonta, M.A. Bray and M.J. Pamham (Eds.), The Pharmacology of Inflammation, Handbook of Inflammation, Vol. 5, Elsevier, Amsterdam, 1985, pp. 27-48. 41 Skjelbred, P., The effects of acetylsalicylic acid on swelling, pain and other events after surgery, Brit. J. clin. Pharmacol., 17 (1984) 379-384. 42 Skjelbred, P. and L&ken, P., Post-operative pain and inflammatory reaction reduced by injection of a corticosteroid. A controlled trial in bilateral oral surgery, Europ. J. clin. Pharmacol., 21 (1982) 391-396. 43 Skjelbred, P. and L&ken, P., Reduction of pain and swelling by a corticosteroid injected 3 hours after surgery, Europ. J. clin. Pharmacol., 23 (1982) 141-146. 44 Spector, S. and Vesell, E.S., Disposition of morphine in man, Science, 174 (1971) 421-422. 45 Sugimoto, M., Kuraishi, Y., Satoh, M. and Takagi, H., Involvement of medullary opioid-peptidergic and spinal noradrenergic systems in the regulation of formalin-induced persistent pain, Neuropharmacology, 25 (1986) 481-485. 46 Taber, RI., Predictive value of analgesic assays in mice and rats. In: M.C. Braude, L.S. Harris, E.L. May, J.P. Smith and J.E. Villarreal (Eds.), Narcotic Antagonists, Advances in Biochemical Psychopharmacology, Vol. 8, Raven Press, New York, 1974, pp. 191-211. 47 Tolman, E.L., Fuller, B.L., Marinan, B.A., Capetola, R.J., Levinson, S.L. and Rosenthale, M.E., Tissue selectivity and variability of effects of acetaminophen on arachidonic acid metabolism, Prostaglandins Leukotrienes Med., 12 (1983) 347-356. 48 Trinder, P., Rapid determination of salicylate in biological fluids, B&hem. J., 57 (1954) 301-303. 49 Tsurufuji, S., Sugio, K., Takemasa, F. and Yoshizawa, S., Blockade by antiglucocorticoids, actinomytin D and cycloheximide of anti-inflammatory action of dexamethasone against bradykinin, J. Pharmacol. exp. Ther., 212 (1980) 225-231. 50 Tyers, M.B. and Haywood, H., Effects of prostaglandins on peripheral nociceptors in acute inflammation, Agents Actions, Suppl. 6 (1979) 65-78. 51 Vane, J.R., Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs, Nature new Biol., 231 (1971) 232-235. 52 Vinegar, R., Schreiber, W. and Hugo, R., Biphasic development of carrageenin edema in rats, J. Pharmacol. exp. Ther., 166 (1969) 96-103. 53 Vinegar, R., Truax, J.F. and Selph, J.L., Quantitative comparison of the analgesic and anti-inflammatory activities of aspirin, phenacetin and acetaminophen in rodents, Europ. J. Pharmacol., 37 (1976) 23-30. 54 Winder, C.V., Wax, J., Burr, V., Been, M. and Rosiere, C.E., A study of pharmacological influences on ultraviolet erythema in guinea pigs, Arch. int. Pharmacodyn., 116 (1958) 261-292. 55 Winter, C.A. and Flataker, L., Reaction thresholds to pressure in edematous hindpaws of rats and responses to analgesic drugs, J. Pharmacol. exp. Ther., 150 (1965) 165-171. 56 Wood, P.L., Animal models in analgesic testing. In: M. Kuhar and G. Pastemak (Eds.), Analgesics; Neurochemical, Behavioral, and Clinical Perspectives, Raven Press, New York, 1984, pp. 175-194. 57 Zimmermann, M., Ethical guidelines for investigations of experimental pain in conscious animals, Pain, 16 (1983) 109-110.
Pain, 30 (1987) 103-114 Elsevier
PAI 01057
The formalin test in mice: dissociation between inflcimm2tnrv 5mt-l ___- non-infl2mmrttnrv __V__““‘-“‘““““-““J “““““‘““‘“‘““J Steinar Hunskaar
twin r--‘-
and Kjell Hole
Department of Physiology, University of Bergen, N-5009 Bergen (Norway)
(Received 9 September 1986, accepted 1 December 1986)
The formahn test in mice is a vahd and reliable model of nociception and is sensitive for various classes of analgesic drugs. The noxious stimulus is an injection of dilute formahn (1% in saline) WI&~ c~wfmv= nf hidnnaw * n.e ~e~nnnre ic the nf rl_.“- __ .._”-amnnnt .___^_. __time _*.._the _._-animsllc 1.1_.. -_ “_..“. the s&n of the . .._ dnrc~l -_.,.“- ..-.*..“_ the . ..- rirrht ..CT-..-*-r..” spend lick.ing the injected paw. Two distinct periods of high licking activity can be identified, an early phase lasting the first 5 mm and a late phase lasting from 20 to 30 mm after the injection of formalin. In order to elucidate the involvement of inflammatory processes in the two phases, we tested different classes of drugs in the two phases ind~endently. Morphine, codeine, nefopam, and orphenadrine, as examples of centrally acting analgesics, were antinociceptive in both phases. In contrast, the non-steroid anti-inflammatory drugs indomethacin and naproxen and the steroids dexamethasone and hydrocortisone inhibited only the late phase, while acetylsalicylic acid (ASA) and paracetamol were antinociceptive in both phases. The results demonstrate that the two phases in the formalin test may have different nociceptive mechanisms. It is suggested that the early phase is due to a direct effect on nociceptors and that prostaglandins do not play an important role during this phase. The late phase seems to be an infla~ato~ response with infl~ato~ pain that can be inhibited by anti-infla~ato~ drugs. ASA and paracetamol seem to have actions independent of their i~bition of prostaglandin synthesis and they _1__nave -L_.._CIICCLS _#C--.- on _- n”n-lnmlmmarory .__- f_~-__.._.r-_. pam. _-I. LtlS”
S-=Y
Key words: Pam test; Mice; Infl~ation
A large number of behavioural methods have been developed in order to study nociception and the action of various analgesic drugs in animals. Most of these tests readily reveal the activity of narcotic analgesics, but are insensitive to mild, non-narcotic antinociceptive drugs [24:46:56]. Attempts to demonstrate the antinociceptive effect of acetyls~icylic acid (ASA) and related drugs by applying the painful stimulus to intact tissue, have generally been unsuccessful, whereas techCorrespondence to: Steinar Hunskaar, Arstadveien 19, N-5009 Bergen, Norway.
03~-39S9/87/$03.50
M.D., Department
of Physiology, University
0 1987 Elsevier Science Publishers B.V. (Biomedical Division)
of Bergen,
104
niques involving inflammation show analgesic activity of these drugs [20,34,38,55]. The inflammatory and non-inflammatory pain are by some authors suggested to represent different physiological entities. The inflammatory pain is thought to be selectively inhibited by the non-steroid anti-inflammatory drugs (NSAIDs) and steroids because they reduce inflammation, and not because of any specific antinociceptive action [9,53,54]. However, an increasing amount of data supports the possibility that NSAIDs have antinociceptive actions independent of their anti-inflammatory effects [5,7,8,24,25,27,34]. Hyperalgesia, an essential feature of inflammatory pain, consists of two different components which involve the action of prostaglandins: the sensitization of local, peripheral receptors, and a less well documented central effect on pathways mediating nociception [13]. There are, however, many other substances that are released in acute inflammation, such as histamine, serotonin and bradykinin. Data from the carrageenan induced inflammation in rats, indicate distinct phases in the inflammatory process [10,40,50]; an initial phase where histamine and serotonin are released, followed by kinins, lasting for some 30-60 min, and then a phase mediated by prostaglandins. Thus prostaglandins act relatively late in the development of inflammation, although prostacyclin (PGI,) induces an early and short-lived hyperalgesia [14]. It is well established that NSAIDs prevent the hyperalgesia of inflammation by blocking the prostaglandin synthesis [15,51]. The formalin test in mice [24] is sensitive to NSAIDs and other mild analgesics. The test employs a chemical nociceptive stimulus that elicits a spontaneous response indicative of pain. The test has two different phases, possibly reflecting different types of pain [11,24,27]. As inflammation occurs at the site of formalin injection it should be possible to elucidate the role of inflammation on the responses in the two phases. We have previously shown that ASA and indomethacin are antinociceptive through partially different modes of action in this test [27]. We have also shown that ASA does not have any delay of onset of its action in the early phase compared to morphine [23]. Both of these studies suggested that ASA has some effects which cannot be attributed to the inhibition of prostaglandin synthesis alone. In the present study, 3 groups of drugs with well-established effects were used to study the two phases of the formalin test: (1) Standard narcotic analgesics (morphine and codeine) [28] and centrally acting non-narcotic analgesics without known mode of action (nefopam and orphenadrine) [21,22]. (2) Prostaglandin synthesis inhibitors (ASA, paracetamol, indomethacin and naproxen) [6,16,17,39]. (3) Anti-inflammatory steroids (hydrocortisone and dexamethasone) [43,46] which inhibit inflammation through a variety of actions including the inhibition of prostaglandin synthesis. Determinations of serum concentrations of ASA, paracetamol and morphine after antinociceptive doses of each drug were included in the study. Methods
Animals Male albino mice (Born: NMRI, Msllegird, Denmark) weighing 25-35 g were used in the experiments. The animals were housed in colony cages (15-16 mice in
105
each) with free access to food and water prior to the experiments. They were maintained in climate- and light-controlled rooms (23 f 0.5 ’ C,12/12 h dark/light cycle with lights on at 07:OO) for at least 2 weeks prior to the experiments. Testing took place during the first hours of the light phase. The mice were brought to the test room the day before testing, and allowed at least 18 h to adapt to the testing environment. Drugs and administration routes
All injections were made intraperitoneally (i.p.) in a volume of 10 ml/kg. In all experiments an equal volume of vehicle was used as control for the injections. The drugs were injected 30 min prior to testing unless otherwise stated. Acetylsalicylic acid (Svaneapoteket, Bergen) and indomethacin (Merck, Sharp and Dohme) were dissolved in Tris buffer 0.1 M, pH 7.4-7.6. Morphine hydrochloride (NAF-Lab A.S.), codeine phosphate (Svaneapoteket, Bergen), nefopam hydrochloride (3M Biker Lab., Inc.), orphenadrine citrate (3M Biker Lab., Inc.), naproxen (Astra), and dexamethasone 21-phosphate injectable (Merck, Sharp and Dohme) were dissolved in 0.9% NaCl. Paracetamol (Svaneapoteket, Bergen), and hydrocortisone (Norsk Medisinaldepot) were dissolved in 12.5% 1,2-propanediol in 0.9% NaCl. Control experiments did not show any altered responses due to the vehicles alone. Formalin test
Separate groups of animals were used for the different experiments. Each mouse was used on one occasion only. Testing or data recording were performed by an observer unaware of the drug treatment in each experiment. The formalin test [11,24] was performed in mice that had been individually exposed to the observation chamber (macrolone cage, 30 cm x 12 cm x 13 cm) for 2 h. Using a minimum of restraint, 20 ~1 of 1% formalin in 0.9% NaCl was injected subcutaneously into the dorsal hind paw of the mouse using a microsyringe with a 26-gauge needle. The mouse was then put back into the chamber and the observation period started. The amount of time the animal spent licking the injected paw or leg was recorded. On the basis of the response pattern described elsewhere [24], two distinct periods of intensive licking activity were identified and scored separately unless otherwise stated. The first period (early phase) was recorded O-5 min after the injection of formalin and the second period (late phase) was recorded 20-30 min after the injection. Thus, when testing the late phase, formalin was injected 10 min after the injection of the drug in order to obtain the same period of time (30 min) between injection of drug and testing. In all experiments attention was paid to the ethical guidelines for investigations of experimental pain in conscious animals [57], and the procedures were approved by the Norwegian Committee for Experiments on Animals. Determination of serum concentrations of analgesic drugs
Mice were given i.p. morphine (5 and 10 mg/kg), acetylsalicylic acid or paracetamol (200 and 400 mg/kg), and blood samples were obtained after 30 min by puncture of the heart during combined pentobarbital and chloral hydrate anaesthesia. The serum concentration of morphine was determined with HPLC and electrochem-
106
ical detection by the method of Bolander et al. [l] with small modifications. The samples were extracted twice with CHCl, and run on a 3 pm C,, column without an internal standard. A phosphate buffer (pH 6.2) containing 10% ethanol was used to increase the retention time. The serum salicylate concentration was determined by calorimetry after complex formation with ferric nitrite (Du Pont ACA, Wilmington, DE) [48]. Paracetamol was measured in serum using enzyme multiplied immunoassay technique (EMIT, Syva Co., Palo Alto, CA) [33]. Statistical analysis
The data were examined by analysis of variance (ANOVA) and Dunnett’s procedure for multiple comparisons with a single control group. When the analysis was restricted to two means, Student’s t test (two-tailed) was used. Level of significance was set to 5% (P < 0.05). Results are given as mean f S.E.M. Results The antinociceptive effects of different classes of drugs in the early and late phases of the formalin test
The centrally acting analgesic drugs morphine, codeine, nefopam, and orphenadrine all showed dose-dependent antinociceptive effects during both the early and
Drug. Dose: (mg/kgl
\ F** F 111’ Morphtne
*0
1
0
5
‘i
10 l
l
*
Codeine
0
25 50
Nefopam
0
10 20
Orphenadnne 0
10 20
08
*
l
l.
l
160J
I,i
Fig. 1. Antinociceptive effects of intraperitoneally administered morphine, codeine, nefopam, or orphenadrine in the formalin test. Nociceptive behaviour in the early phase (O-S mm after the injection of formalin) and the late phase (20-30 min after the injection of formalin) was scored as the amount of time spent licking the injected hind paw or leg (xc, mean f S.E.M.). Hatched columns indicate a statistically significant dose-response relationship. Significant differences between vehicle and drug treated groups arc indicated by * (P < 0.05) or * * (P < 0.01) (Dunnett’s test subsequent to ANOVA). The statistical calculations for the early (ER) and late (LR) response for dose-response relationships were as follows 43,31, (ANOVA): morphine, P < o,ool (ER), FF 2.21 = 41.71, P < 0.001 (RR), &,a, =L5.55, P < 0.001 (LR): a,,s =17.08, P -C 0.001 (LR); nefopam, F2,2,= 10.38, P < 0.001 codeine, F2,*,= F2,35 = 24.17,P < 0.001 (ER), F2,+ = 6.90, P < 0.005 (RR). Fa.20 = 22.53, P < 0.001(LR); orphenadrine, (LR).
Drug:
A
Dose:
v
ASA
Paracetamol
lndomethacln
n
Naproxen
Fig. 2. Antinociceptive effects of intraperitoneally administered acetylsalicylic acid (ASA), paracetamol, indomethacin, or naproxen in the formalin test. Nociceptive behaviour in the early phase (O-5 mm after the injection of formalin) and the late phase (20-30 mitt after the injection of formalin) was scored as the amount of time spent licking the injected hind paw or leg (XC,mean f S.E.M.). Hatched columns indicate a statistically significant dose-response relationship. Significant differences between vehicle and drug treated groups are indicated by * (P < 0.05) or * * (P < 0.01) (Dunnett’s test subsequent to ANOVA). The statistical calculations for the early (ER) and late (LR) response for dose-response relationships were as follows (ANOVA): ASA, FzT12 = 32.95, P < 0.001 (ER), F2,2, =19.52, P < 0.001 = (LR); paracetamol, F2,21 = 37.10, P -c0.001 (ER), F2,21 = 83.64, P < 0.001 (LR); indomethacin, F,.,, 0.63, P > 0.5 (ER), F2,*,, = 18.97, P-z0.001 (LR); naproxen, F2,,9 = 0.02, P > 0.9 (ER), F,.,, = 20.51, P c 0.001 (LR).
late phases (Fig. 1, statistics shown in the figure legend). There was no apparent difference in the effect of each drug in the early phase versus the late phase. No motor, neurological, or other behavioural deficits were observed. Fig. 2 shows the effects of 4 different NSAIDs. In doses without overt toxic effects, clear dose-dependent antinociception was found during the late phase for all the drugs In addition, ASA and paracetamol both significantly suppressed the licking activity in the early phase, while no effect was found for indomethacin and naproxen (statistics shown in the figure legend). The steroids hydrocortisone and dexamethasone both significantly suppressed the licking activity in the late phase, but were without effect in the early phase (Fig. 3, statistics shown in the figure legend). Possible delayed effects of such drugs were investigated by injecting dexamethasone (10 mg/kg) 3 h before testing. This treatment did not affect the response to formalin in the early phase of the test. Dexamethasone reduced the licking activity in the late phase with 40% compared to controls, but the difference did not reach statistical significance (tl,lS = 1.34, P = 0.20) (Fig. 3). Analysis of the time course of the effect of indomethacin Indomethacin and naproxen (NSAIDs) and the steroids hydrocortisone
and
Hydrocortlsone
Drug b
Dose. (mgikgi
Dexamethasane
( 3 h)
Dexamethasone
* P
O-
l
160-
Fig. 3. Antinociceptive effects of intraperitoneally administered hydrocortisone and dexamethasone in the formalin test. Nociceptive behaviour in the early phase (O-5 mm after the injection of formalin) and the late phase (20-30 min after the injection of formalin) was scored as the amount of time spent licking the injected hind paw or leg (set, mean + S.E.M.). Dexamethasone treated groups (10 mg/kg) were also tested 3 h after drug injection and compared to similar vehicle treated groups. The animals were maintained in the observation chambers during these 3 h. Hatched columns indicate a statistically significant dose-response relationship. Significant differences between vehicle and drug treated groups are indicated by * * (P -c0.01) (Dunnett’s test subsequent to ANOVA). The statistical calculations for the early (ER) and late (LR) response for dose-response relationships were as follows (ANOVA): = 8.51, P -c0.001 (LR); dexamethasone, F,.,, = 0.50, hydrocortisone, F,, ,6= 1.33, P > 0.2 (ER), F2,42 8.78, P < 0.002 (LR); dexamethasone 3 h after injection, Pr.,s = 0.20, P > 0.3 P > 0.6 (ER), F, ,,, = _,_, (ER), F,,,5 = 1.79, P > 0.05(LR). Earlyphase ‘1
I’
Late phase I,
Cl
^
1
Early phase
Late phase
m
/4
;; I
0 Vehicle 500-
c
. lndomethocln
2
oI o-1 2-3
4-5
O-5
10-15
20-25
Minutes
30-35
after inlectlon of formalin
Fig. 4. Time course of antinociceptive effects of intraperitoneally administered indomethacin (30 mg/kg) or vehicle in the formalin test. Nociceptive behaviour in the early (O-S mitt after the injection of formalin) and the late phase (20-30 min after the injection of formalin) was scored as the amount of time spent licking the injected hind paw (set, mean + S.E.M.). A: pain intensity score during the first 5 mm recorded in 1 mm periods. B: pain intensity score during the first 40 mm after the injection of formalin. Drugs were injected 30 mm before injection of formalin. Significant differences between drug and vehicle treated groups calculated by means of Student’s t test for pairs of data from each 5 mitt period, are indicated by * (P < 0.05) or * * (P < 0.01). C: accumulated response during the first 40 mm (5 mm periods) after the injection of formalin. Statistically significant differences between the two groups are * * * (P < 0.001) (Student’s r test calculated on concomitant pairs). indicated by ** (P
109
dexamethasone all showed a similar pattern of effects in the two phases of the formalin test (Figs. 2 and 3), and preliminary studies indicated that the time course of the effects also was similar for these drugs (data not shown). Indomethacin (30 mg/kg) was therefore used in a more detailed experiment of the time course (Fig. 4). Again, no statistically significant difference was found in the total amount of licking during the early phase between the indomethacin and vehicle treated groups (t 1 12< 0.1, P > 0.7). In order to detect possible differences during this period, each 1 min period was compared in the two groups, but no statistically significant difference was found ( F1,i8 < 0.1, P > 0.7 for overall effect, F4,72= 16.38, P < 0.001 for trial effect and F4,72< 0.1, P > 0.9 for interaction, ANOVA) (Fig. 4A). Time course (O-40 min) of paw licking response after treatment with indomethacin or vehicle is shown in Fig. 4B. ANOVA revealed statistically significant differences between the two groups (Fl,12 = 16.82, P < 0.002 for overall effect, F 7,84= 15.92, P < 0.001 for trial effect and F,,+, = 3.18, P < 0.01 for interaction). In contrast to the early phase, the results from the late phase were significantly different as shown by one-way ANOVA calculated on the two 5 min periods included in the late phase (20-25 and 25-30 rnin) ( F,,12 = 11.10, P -c 0.01 for overall effect, Fl, 12= 1.38, P > 0.2 for trial effect, and Fl,12 < 0.1, P > 0.7 for interaction). In the periods O-15 min and 25-40 min no statistically significant differences were found between the drug and vehicle. When the accumulated response to the nociceptive stimulus was analysed by Student’s t test for each pair of data points, the difference between the two groups reached statistical significance in the fourth 5 min period (15-20 min after formalin injection) (Fig. 4C). Serum concentrations of morphine, acetylsalicylic acid and paracetamol
Determinations of the serum concentrations (Table I) were analysed 30 min after injection, the same interval as between injection and behavioural testing. Preliminary studies indicated that the peak concentrations also occurred at that time (data not shown).
TABLE I SERUM CONCENTRATIONS MOL IN MICE
OF MORPHINE,
ACETYLSALICYLIC
ACID AND PARACETA-
Drug
Dose (mg/kg i.p.)
No. of animals
Serum concentration (mean f S.E.M.)
Morphine
5 10 200 400 200 400
3 4 4 4 4 4
115k16 pg/l 578 f 65 pg/I 200 f 37 mg/l 314 f 79 mg/I 102* 8 mg/l 253 f 14 mg/l
Acetylsalicylic acid Paracetamol
110
Discussion The present study demonstrates that the early and late nociceptive phases of the formalin test in mice are dissimilar. The various effects of 3 classes of analgesic drugs, tested separately on the two phases, indicate that the early phase may be due to direct effects on nociceptors; this phase can be inhibited by centrally acting analgesics. In contrast, the late phase seems to be due to an inflammatory response partly mediated by prostaglandins and can be inhibited by NSAIDs and steroids, as well as the centrally acting drugs. The drug doses used are within the range used by others in comparable studies [24,32,37,39]. Serum concentrations of morphine, ASA, and paracetamol in nearly equipotent antinociceptive doses were found to be on the order of high normal to subtoxic levels for humans. Serum levels of paracetamol (400 mg/kg) were relatively high compared to human data for therapeutic doses [17]. Other studies in animals have also shown that such doses are necessary to induce antinociceptive effects in rodents [19,24,26,35,37,53]. For ASA, the serum levels found here are below the toxic range 30 min after administration in humans [17], and also in accordance with another study in mice [31]. The serum level of morphine found after 5 mg/kg is comparable to human values measured after 10 mg administered intravenously [44]. It may be concluded that compared to human data, moderate to high serum levels of analgesics are required to induce antinociception in the formalin test in mice. The difference between effective human and animal serum level appears to be greatest for morphine. Subcutaneous formalin injection initiates a series of events with an essentially uniform course [24]. This model (as well as the Randall-Selitto test and carrageenan induced inflammation test, for example) reproduces various aspects of acute inflammation, but should not be considered as model for chronic inflammatory responses which often involve immune reactions to antigenic substances. The early phase of the formalin test is of particular interest as the differences between the drugs appeared here. Indomethacin, naproxen and the steroids were without effect in this phase. This argues against an important role of prostaglandins or prostacyclin in the nociceptive responses during the first minutes in this test. ASA has been shown to induce antinociception as early as 30 set after the injection of formalin; the same latency as for morphine [23]. A possible dissociation between analgesic and anti-inflammatory effects of ASA has also been shown in other models [l&41]. From the experiments on the mediators of carrageenan oedema, distinct phases have been demonstrated. ASA inhibits oedema formation in all phases, while all the steroids and other NSAIDs tested inhibit oedema formation only after some time [52]. Indomethacin and dexamethasone failed to suppress the oedema in rats deprived of endogenous prostaglandin precursors, while ASA fully maintained its effect even under these conditions [2,3,27]. Okuyama and Aihara [34] studied NSAIDs in theadjuvant arthritic rat model and found that much smaller doses were needed for antinociception than in normal rats. These authors reported EDso ratios between 1.7 and 1.8 for narcotic analgesics while indomethacin and naproxen had
111
values of 25.7 and 37.4 respectively. Interestingly, ASA and paracetamol had EDs0 values much lower than the other NSAIDs, indicating effects on non-inflammatory pain. Paracetamol was included among the NSAIDs in this study although it has been claimed to have only weak anti-inflammatory effects [17], and not to inhibit prostaglandin synthesis in the periphery [16]. Recent studies in animals demonstrated that also the inhibition of brain prostaglandin synthesis is relatively weak [4,47], and that paracetamol inhibits carrageenan induced oedema in rats in doses with antinociceptive effects [19,53]. The anti-inflammatory effects of steroids are due to several mechanisms, including reduced vascular permeability, exudation, vasodilation, leucocyte emigration, phagocytosis and release of lysosomal enzymes. The current view is that the steroids exert these effects by controlling the rate of synthesis of regulatory proteins. This explains the lag time necessary for the effects in some models [32,49]. Moreover, several authors have demonstrated inhibition of prostaglandin synthesis by steroids [29,30]. The effect is not mediated via the cycle-oxygenase reaction. These observations could indicate that steroids may have analgesic activities comparable to NSAIDs. Some clinical data support the view [42,43], but little experimental evidence for this exists [32]. In the present experiments we postulated that an inhibitory effect on behavioural responses would be due to anti-inflammatory effects and subsequently reduced symptoms of inflammation, including nociception. Both dexamethasone and hydrocortisone significantly suppressed the responses in the late phase. Relatively high doses were used, but in order to suppress the licking by 75% it was presumably necessary to almost eliminate the inflammatory reaction. Injection of dexamethasone 3 h before testing did not yield significant effects on licking, but the tendency was clearly in the direction of a sustained effect in the late phase. There was low licking activity in the control group in this experiment. This was likely an effect of adaptation to the observation chamber, as the late phase is more prone to adaptation than the early phase with its intensive licking activity. It may be concluded that the inhibitory effects of steroids on the late phase indicate that this phase is an inflammatory response with subsequent inflammatory pain. The centrally acting analgesics used in the present study suppressed the responses in the two phases in a similar manner. This may suggest a common way of reducing inflammatory and non-inflammatory pain by these drugs. Despite this, recent studies have indicated that the endogenous opioid-peptidergic and serotonergic systems modulate the early and late phases differently [12,45]. The various effects of the NSAIDs may be due to the fact that ASA and paracetamol have central actions while indomethacin and naproxen do not. However, most NSAIDs have been shown to reduce carrageenan evoked hyperalgesia when given intracerebroventricularly while morphine also reduces non-inflammatory pain [13]. Systemically administered ASA or paracetamol have been shown to inhibit various non-inflammatory nociceptive responses [24-26,341. They inhibit pain related behaviour induced by substance P injected intrathecally or substance P released by intrathecal administration of capsaicin, indicating a central antinociceptive effect of these drugs [25]. Our hypothesis is that the antinociceptive effects of
112
ASA and paracetamol in the early phase of the formalin test are due to a central action, which is probably not related to the inhibition of prostaglandin synthesis. Furthermore, the central effects of some NSAIDs do not exclude the possibility that the late phase is inflammatory and prostaglandin-dependent in nature. In conclusion, the two phases of the formalin test in mice seem to be different, involving different mechanisms of nociception [11,12,24,45]. The data support the view of Dubuisson and Dennis [ll] that the first phase is due to a direct effect of formalin on nociceptors and the second due to inflammation. These findings may significantly extend the use of the test as a tool for studying the processes underlying pain and mechanisms of old and new analgesic drugs.
The authors thank Ms. Anita Kloster and Ms. Trine Tydal for excellent technical assistance. Morphine was analysed by the courtesy of Prof. Ole J. Broth and paracetamol and acetylsalicylic acid by Prof. Olav M. Bakke at the Department of Pharmacology and Toxicology, University of Bergen. Drugs were generously supplied by Merck, Sharp and Dohme (indomethacin), Astra-Syntex A.S. (naproxen), 3M Piker Laboratories, Inc. (nefopam and orphenadrine). The financial support from Norske Kvinners Sanitetsforening is gratefully acknowledged.
References 1 Bolander, H., Kourtopoulos, H., Lundberg, S. and Person, S.-A., Morphine concentrations in serum, brain and cerebrospinal fluid in the rat after intravenous administration of a single dose, J. Pharm. Pharmacol., 35 (1983) 656-659. 2 Bonta, IL., Bult, H., v.d. Ven, L.L.M. and Noordhoek, J., Essential fatty acid deficiency; a condition to discriminate prostaglandin and non-prostaglandin mediated components of inflammation, Agents Actions, 6 (1976) 154-158. 3 Bonta, I.L., Bult, H., Vincent, J.E. and Zijlstra, F.J., Acute anti-inflammatory effects of aspirin and dexamethasone in rats deprived of endogenous prostaglandin precursors, J. Pharm. Pharmacol., 29 (1977) l-7. 4 Bruchhausen, F. and Baumann, J., Inhibitory actions of desacetylation products of phenacetin and paracetamol on prostaglandin synthetase in neuronal and glial cell lines and rat renal medulla, Life Sci., 30 (1982) 1783-1791. 5 Brune, K., Bucher, K. and Walz, D., The avian microcrystal arthritis. II. Central versus peripheral effects of sodium salicylate, acetaminophen and colchicine, Agents Actions, 4 (1974) 27-33. 6 Brune, K., Glatt, M. and Graf, P., Mechanisms of action of anti-inflammatory drugs, Gen. Pharmacol., 7 (1976) 27-33. 7 Capetola, R.J., Sbriver, D.A. and Rosenthale, M.E., Suprofen, a new peripheral analgesic, J. Pharmacol. exp. Ther., 214 (1980) 16-23. 8 Capetola, R.J., Rosenthale, M.E., Dubinsky, B. and McGuire, J.L., Peripheral antialgesics: a review, J. clin. Pharmacol., 23 (1983) 545-556. 9 Crepax, P. and Silvestrini, B., Experimental evaluation in laboratory animals of anti-inflammatory analgesic drugs, Arch. ital. Biol., 101 (1963) 444-457.
113 10 Di Rosa, M., Giroud, J.P. and Willoughby, D.A., Studies of the mediators of the acute inflammatory response induced in rats in different sites by carrageenan and turpentine, J. Pathol., 104 (1971) 15-29. 11 Dubuisson, D. and Dennis, S.G., The formahn test: a quantitative study of the analgesic effects of morphine, meperidine, and brain stem stimulation in rats and cats, Pam, 4 (1977) 161-174. 12 Fasmer, O.B., Berge, O.G. and Hole, K., Changes in nociception after lesions of descending serotonergic pathways induced with 5,6dihydroxytryptamine. Different effects in the formalin and tail-flick tests, Neuropharmacology, 24 (1985) 729-734. 13 Ferreha, S.H., Lorenzetti, B.B. and Correa, F.M.A., Central and peripheral antialgesic action of aspirin-like drugs, Europ. J. Pharmacol., 53 (1978) 39-48. 14 Ferreira, S.H., Nakamura, M. and Castro, M.S.A., The hyperalgesic effects of prostacyclin and prostaglandin E2, Prostaglandms, 16 (1978) 31-37. 15 Flower, R.J., Drugs which inhibit prostaglandin biosynthesis, Pharmacol. Rev., 26 (1974) 161-174. 16 Flower, R.J. and Vane, J.R., Inhibition of prostaglandin synthetase in brain explains the anti-pyretic activity of paracetamol (Cacetamidophenol), Nature (Lond.), 240 (1972) 410-411. 17 Flower, R.J., Moncada, S. and Vane, J.R., Analgesic-antipyretics and anti-inflammatory agents. Drugs employed in the treatment of gout. In: A. Goodman Giiman, L.S. Goodman and A. Gilman (Eds.), The Pharmacological Basis of Therapeutics, MacMillan, New York, 1980, pp. 682-728. 18 Gilfoil, T.M., Klavins, I. and Grumbach, L., Effects of acetylsalicylic acid on the edema and hyperesthesia of the experimental inflamed rat’s paw, J. Pharmacol. exp. Ther., 142 (1963) 1-5. 19 Glenn, E.M., Bowman, B.J. and Rohloff, N.A., Anti-inflammatory and PG inhibitory effects of phenacetin and acetaminophen, Agents Actions, 7 (1977) 513-516. 20 Guilbaud, G. and Iggo, A., The effect of lysine acetylsalicylate on joint capsule mechanoreceptors in rats with polyarthritis, Exp. Brain Res., 61 (1986) 164-168. 21 Heel, R.C., Brogden, R.N., Pakes, G.E., Speight, T.M. and Avery, G.S., Nefopam, a review of its pharmacological properties and therapeutic efficacy, Drugs, 19 (1980) 249-267. 22 Hunskaar, S., Antinociceptive effects of orphenadrine citrate in mice, Europ. J. Pharmacol., 111 (1985) 221-226. 23 Hunskaar, S., Similar effects of acetylsalicylic acid and morphine on immediate responses to acute noxious stimulation, Pharmacol. Toxicol., 60 (1987) 167-170. 24 Hunskaar, S., Fasmer, O.B. and Hole, K., Formalin test in mice, a useful technique for evaluating mild analgesics, J. neurosci. Methods, 14 (1985) 69-76. 25 Hunskaar, S., Fasmer, O.B. and Hole, K., Acetylsalicylic acid, paracetamol and morphine inhibit behavioural responses to intrathecally administered substance P or capsaicin, Life Sci., 37 (1985) 1835-1841. 26 Hunskaar, S., Berge, O.-G. and Hole, K., A modified hot-plate test sensitive to mild analgesics, Behav. Brain Res., 21 (1986) 101-108. 27 Hunskaar, S., Berge, O.-G. and Hole, K., Dissociation between antinociceptive and anti-inflammatory effects of acetylsalicylic acid and indomethacin in the formalin test, Pain, 25 (1986) 125-132. 28 Jaffe, J.H. and Martin, W.R., Gpioid analgesics and antagonists. In: A. Goodman Gilman, L.S. Goodman and A. Gilman (Eds.), The Pharmacological Basis of Therapeutics, MacMillan, New York, 1980, pp. 494-534. 29 Kantrowitz, F., Robinson, D.R., McGuire, M.B. and Levine, L., Corticosteroids inhibit prostaglandin production by rheumatoid synovia, Nature (Lond.), 258 (1975) 737-739. 30 Lewis, G.P. and Piper, P.J., Inhibition of release of prostaglandins as an explanation of some of the action of antiinflammatory corticosteroids, Nature (Lond.), 254 (1975) 308-311. 31 Ljungberg, S., Otto, U. and Paalzow, L., Analgesic activity of acetylsalicylic and salicylic acid and its relation to blood concentrations in mice, Acta pharm. suet., 5 (1968) 489-500. 32 Nakamura, H., Mizushima, Y., Seto, Y., Motoyoshi, S. and Kadokawa, T., Dexamethasone fails to produce antipyretic and analgesic actions in experimental animals, Agents Actions, 16 (1985) 542-547. 33 Oellerich, M., Enzyme immunoassays in clinical chemistry. Present status and trends, J. clin. Chem. clin. B&hem., 18 (1980) 197-208. 34 Okuyama, S. and Aihara, H., Inhibition of electrically-induced vocalization in adjuvant arthritic rats as a novel method for evaluating analgesic drugs, Jap. J. Pharmacol., 34 (1984) 67-77.
114 35 Okuyama, S. and Aihara, H., The site of action of morphine and indomethacin differs with electrical stimulation of cutaneous or tibia1 nerves in normal and adjuvant arthritic rats, Arch. int. Pharmacodyn., 276 (1985) 133-141. 36 Okuyama, S. and Aihara, H., Hyperalgesic action in rats of intracerebroventricularly administered arachidonic acid, PG E2 and PG F2a: effects of analgesic drugs on hyperalgesia, Arch. int. Pharmacodyn., 278 (1985) 13-22. 37 Pong, SF., Demuth, S.M., Kinney, C.M. and Deegan, P., Prediction of human analgesic dosages of nonsteroidal anti-inflammatory drugs (NSAIDs) from analgesic ED50 values in mice, Arch. int. Pharmacodyn., 273 (1985) 212-220. 38 Randall, L.O. and Selitto, J.J., A method for measurement of analgesic activity on inflamed tissue, Arch. int. Pharmacodyn., 111 (1957) 409-419. 39 Roszkowski, A.P., Rooks II, W.H., Tomolonis, A.J. and Miller, L.M., Anti-inflammatory and analgetic properties of D-2-(6’-methoxy-2’-naphthyl)-propionic acid (naproxen), J. Pharmacol. exp. Ther., 179 (1971) 114-123. 40 Sedgwick, A.D. and Willoughby, D.A., Initiation of the inflammatory response and its prevention. In: I.L. Bonta, M.A. Bray and M.J. Pamham (Eds.), The Pharmacology of Inflammation, Handbook of Inflammation, Vol. 5, Elsevier, Amsterdam, 1985, pp. 27-48. 41 Skjelbred, P., The effects of acetylsalicylic acid on swelling, pain and other events after surgery, Brit. J. clin. Pharmacol., 17 (1984) 379-384. 42 Skjelbred, P. and L&ken, P., Post-operative pain and inflammatory reaction reduced by injection of a corticosteroid. A controlled trial in bilateral oral surgery, Europ. J. clin. Pharmacol., 21 (1982) 391-396. 43 Skjelbred, P. and L&ken, P., Reduction of pain and swelling by a corticosteroid injected 3 hours after surgery, Europ. J. clin. Pharmacol., 23 (1982) 141-146. 44 Spector, S. and Vesell, E.S., Disposition of morphine in man, Science, 174 (1971) 421-422. 45 Sugimoto, M., Kuraishi, Y., Satoh, M. and Takagi, H., Involvement of medullary opioid-peptidergic and spinal noradrenergic systems in the regulation of formalin-induced persistent pain, Neuropharmacology, 25 (1986) 481-485. 46 Taber, RI., Predictive value of analgesic assays in mice and rats. In: M.C. Braude, L.S. Harris, E.L. May, J.P. Smith and J.E. Villarreal (Eds.), Narcotic Antagonists, Advances in Biochemical Psychopharmacology, Vol. 8, Raven Press, New York, 1974, pp. 191-211. 47 Tolman, E.L., Fuller, B.L., Marinan, B.A., Capetola, R.J., Levinson, S.L. and Rosenthale, M.E., Tissue selectivity and variability of effects of acetaminophen on arachidonic acid metabolism, Prostaglandins Leukotrienes Med., 12 (1983) 347-356. 48 Trinder, P., Rapid determination of salicylate in biological fluids, B&hem. J., 57 (1954) 301-303. 49 Tsurufuji, S., Sugio, K., Takemasa, F. and Yoshizawa, S., Blockade by antiglucocorticoids, actinomytin D and cycloheximide of anti-inflammatory action of dexamethasone against bradykinin, J. Pharmacol. exp. Ther., 212 (1980) 225-231. 50 Tyers, M.B. and Haywood, H., Effects of prostaglandins on peripheral nociceptors in acute inflammation, Agents Actions, Suppl. 6 (1979) 65-78. 51 Vane, J.R., Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs, Nature new Biol., 231 (1971) 232-235. 52 Vinegar, R., Schreiber, W. and Hugo, R., Biphasic development of carrageenin edema in rats, J. Pharmacol. exp. Ther., 166 (1969) 96-103. 53 Vinegar, R., Truax, J.F. and Selph, J.L., Quantitative comparison of the analgesic and anti-inflammatory activities of aspirin, phenacetin and acetaminophen in rodents, Europ. J. Pharmacol., 37 (1976) 23-30. 54 Winder, C.V., Wax, J., Burr, V., Been, M. and Rosiere, C.E., A study of pharmacological influences on ultraviolet erythema in guinea pigs, Arch. int. Pharmacodyn., 116 (1958) 261-292. 55 Winter, C.A. and Flataker, L., Reaction thresholds to pressure in edematous hindpaws of rats and responses to analgesic drugs, J. Pharmacol. exp. Ther., 150 (1965) 165-171. 56 Wood, P.L., Animal models in analgesic testing. In: M. Kuhar and G. Pastemak (Eds.), Analgesics; Neurochemical, Behavioral, and Clinical Perspectives, Raven Press, New York, 1984, pp. 175-194. 57 Zimmermann, M., Ethical guidelines for investigations of experimental pain in conscious animals, Pain, 16 (1983) 109-110.