q Birkhäuser Verlag, Basel, 1999

Inflamm. res. 48 (1999) 120–127

1023-3830/99/030120-08 $ 1.50+0.20/0 Inflammation Research

Original Research Papers

The potentiation of analgesic activity of paracetamol plus morphine

involves the serotonergic system in rat brain
M. Sandrini1, G. Vitale2, A. Ottani1 and L.-A. Pini2
1
Dept of Biomedical Sciences, Section of Pharmacology, University of Modena, Italy
2
Clinical Pharmacology Unit, Dept. of Internal Medicine, University of Modena, Via del Pozzo, 71, I-41100 Modena, Italy, Fax þ39 59 424069,
e-mail: pinila@unimo.it

Received 7 July 1998; returned for revision 31 August 1998; accepted by K. Brune 2 November 1998

Abstract. Objective and Design: We investigated the prostaglandin synthesis owing to cyclooxygenase inhibition.
antinociceptive effect of subactive doses of paracetamol However, whereas the anti-inflammatory potency of
and morphine, given in combination. NSAIDs can be correlated with their inhibition of cyclo-
Material and Treatment: Male Wistar rats were injected with oxygenase, the relationship between the antinociceptive
paracetamol (50 or 100 mg/kg i.p.) and morphine (2, 3 or 5 activity of these drugs and their capacity for cyclooxygenase
mg/kg s.c.) 10 min later and subjected to algesimetric tests inhibition is controversial [1–3]. It has been observed that
20 min thereafter. phenazone and paracetamol significantly inhibit prostaglan-
Methods: Pain threshold was evaluated in the hot-plate and din synthesis in the central nervous system (CNS), and some
formalin tests. 5-HT2 receptor binding capacity and 5-HT findings suggest that paracetamol has a weak inhibitory
and 5-HIAA levels were measured in cortical and pontine activity on the synthesis of peripheral prostaglandin, but is a
areas of the brain by means of radioligand binding technique potent inhibitor of prostaglandin biosynthesis within the
and by HPLC, respectively. Statistical analysis was done CNS [4, 5]. Recently the description of two structurally
using Student-Neuman-Keul’s test and 2 × 2 factorial different forms of cyclooxygenase enzyme (COX-1 and
analysis. COX-2) indicated a possible explanation for the different
Results: Only when given in combination, paracetamol (100 analgesic and anti-inflammatory activities of many NSAIDs
mg/kg) and morphine (2 and 3 mg/kg) were able to evoke an [6], and inhibition of COX-2 represents for some authors the
antinociceptive effect in both tests associated with an increase most likely mechanism of action for NSAID-mediated
in 5-HT levels and a decrease in 5-HT2 receptors in the cortex. analgesia [7, 8], mainly at spinal level as for indomethacin
These effects were prevented by i.p. pretreatment with [9] or peripherically as for meloxicam [10].
naloxone (1 mg/kg i.p.). Other authors have demonstrated that paracetamol has
Conclusions: Subactive doses of paracetamol and morphine only a weak antinociceptive effect when injected intracer-
exert an analgesic effect when given in combination in the rat ebroventricularly (i.c.v.) in rats [11]. This lack of activity
and indicate an involvement of both serotonergic and may result from its inability to inhibit prostaglandin
opiatergic systems. synthesis in rat CNS tissue [12].
The central monoaminergic and serotonergic pathways
Key words: Paracetamol – Morphine – Antinociception – are involved in pain modulation [13] and a possible
5-Hydroxytryptamine (5-HT) – 5-HT2 receptors – Brain connection between the analgesia induced by some
NSAIDs and the increase in the turnover rate of dopamine,
noradrenaline and serotonin in the rat CNS has been
proposed [14, 15]. It has been proposed that many types of
Introduction analgesic drugs (opiates, NSAIDs, etc) act through an
increase in brain serotonin levels. The serotonergic system
The mechanism of action of non-steroidal anti-inflammatory can regulate nociception in different ways, depending on the
drugs (NSAIDs) is believed to result from suppression of receptor subtypes involved, and serotonin has been claimed
to exert its central antinociceptive effect in defined brain
areas through its receptor subtypes, notably 5-HT1A [16],
Correspondence to: L.-A. Pini 5-HT2 [17] and 5-HT3 at the spinal level [18].

m
Vol. 48, 1999 Potentiation of analgesic activity of paracetamol plus morphine 121

The bulk of the data suggests that stimulation of 5-HT1 followed the same experimental procedure. At the end of the
receptors reduces nociceptive sensitivity, whereas activation experiments the rats were anaesthetized, decapitated and their blood
of 5-HT2 receptors increases nociceptive responsiveness and brains were removed and stored until required for analysis. Two
additional groups of rats were tested for motor activity with the use of
after long term treatment [19]. either PARA 100 mg/kg plus morphine 3 mg/kg or vehicle plus saline.
The serotonergic system can play a role in the
antinociceptive mechanism of some NSAIDs; it has been
also proposed that other neurotransmitter systems, including Test procedures
opiatergic pathways, may be involved in the central
analgesic effect of this class of drugs [13, 20–22]. Hot-plate test. The hot-plate test (HP) consisted of an electrically-
Previous observations by our group suggested that the heated surface (Socrel DS-35, Ugo Basile, Comerio, VA, Italy) kept at a
antinociceptive effect of phenazone, acetylsalicylic acid constant temperature of 54 6 0:8 8C. The latencies for paw licking or
(ASA) or paracetamol (PARA) in the hot-plate test in rats jumping were recorded for each animal. The baseline latency in the hot-
plate test ranged from 5:9 6 0:8 to 6:3 6 0:9 sec (ANOVA, p > 0.5).
was associated with a decrease in the number of serotonin The analgesic efficacy of the drugs was evaluated as the percentage of
receptors in some brain areas [23–25]. the maximum possible effect (% MPE) according to the formula: (TL-
Morphine stimulates 5-HT release via a supraspinal BL)/(45-BL) × 100, where TL¼test latency; BL¼baseline latency;
action [26] and the latter depletion in the CNS decreases the 45¼cutoff time in seconds.
analgesic effect of morphine [27]; thus, we can draw the Immediately after the last pain threshold measurement, the animals
conclusion that morphine exerts its analgesic effect, at least were anaesthetized with ethyl ether and decapitated; the blood was
collected and serum was stored at ¹20 8C. The brains were removed,
in part, through the serotonergic system [28, 29]. Recently, weighed and stored at ¹80 8C until required for assays.
our group [30] demonstrated that PARA binds to [3H]nalox-
one sites, although with less potency than morphine; the Formalin test. Two hours before testing the animals were placed
maximum binding of PARA was 70% of that of morphine. individually in standard cages and, after the adaptation period, 50 ml of
Since morphine and PARA may share common pathways in 5% formalin solution was injected subcutaneously into the dorsal
their mechanism of action, it could be expected that the surface of the left hindpaw by use of microsyringe with a 26-gauge
needle. Pain behaviour was monitored for a period of 40 min, 5-min
combination of the two drugs, at subactive doses, should
intervals starting at the time 0. Two phases of spontaneous flinching
result in a potentiation of analgesic efficacy, as has been well behaviour were observed: phase 1 began immediately after formalin
documented in clinical practice with combinations of injection to 10 min thereafter, phase 2 began at time 11 and terminated
analgesics and mild opiates. 40 min after formalin injection.
The aim of this work was to assess: To avoid possible interference of room temperature on skin
1) the analgesic effect of combinations of PARA and temperature, all experiment were performed at room temperature of
morphine in rats used in subactive doses, in the hot- 22 6 1 8C [31].
plate and formalin tests; Motor activity. The experiments were performed between 9:00 and
2) the changes in serotonin and 5-HIAA levels and 5-HT2 12:00 a.m. in a sound proof room by experienced observers unaware of
receptors in rat brain membranes; the treatments. Motor activity was measured in an activity cage by
3) the influence of pretreatment with naloxone on the means of an ultrasound apparatus (Cibertic, S. A., Barcelona, Spain)
behavioural and biochemical changes induced by drug placed on the lid of the cage. The number of movements were recorded
combinations. continuously 1 h after an adaptation period of 30 min.

Paracetamol assay
Materials and methods
Paracetamol were determined in the sera by fluorescence polarization
Animals immunoassay (FPIA). A TDx analyzer was used for drug evaluation
(Abbot Laboratories, Chicago, IL, USA).
Adult male rats (Harlan-Nossan, SPF, Correzzano, Italy) weighing
180–200 g at the beginning of the experiments, were housed in
Plexiglas cages, four per cage, with free access to food and water, and Serotonin (5-HT) and 5-hydroxyindoleacetic acid (5-HIAA)
maintained on a 12 h dark/light cycle (light on at 7:00 a.m.) under determination
controlled environmental conditions (temperature, 22 6 1 8C; humidity,
60%). The ethical guidelines for investigation of experimental pain in
conscious animals were followed in all tests, and the procedures were The brain areas were assayed for 5-HT and 5-HIAA determinations by
carried out according to the EEC ethical regulation for animal research reverse-phase high-performance liquid chromatography (HPLC),
(EEC Council 86/609; D.L.27/01/1982, No. 116). according to Grossi and coworkers [32] with modifications.
The thawed areas were homogenized with an ultrasonic dismem-
branator, in 0.1 M HClO4 containing 4 mM NaHSO4 (10 ml per mg of
wet weight) and centrifuged at 2000 × g for 15 min at 4 8C. After
Drug treatment centrifugation, the acid supernatant was filtered on 0.22 mm filters
before analysis.
Rats, divided into groups of 8 animals, were injected intraperitoneally 5-HT and 5-HIAA assays were performed with a Beckman System
(i.p.) with paracetamol at the dose of 50 or 100 mg/kg (dissolved in a Gold high-performance liquid chromatograph (Beckman Instruments
vehicle, which consisted of 12.5% of 1,2-propanediol in sterile saline) Inc., San Ramon, CA, USA) equipped with an ESA Coulochem II
or vehicle. Morphine 2, 3 and 5 mg/kg s.c. dissolved in saline, or saline Multi-Electrode high sensitivity electrochemical detector (ESA Inc.,
was injected, 10 min after PARA administration. Rats were subjected to Bedford, MA, USA) with conditioning cell set at ¹0.75 V, detector set
the algesimetric tests 20 min thereafter. Other groups of rats were at þ0.05 V and detector 2 set at þ0.25 V, response 2, gain 10 × 5. A
pretreated with naloxone (1 mg/kg i.p.) in sterile saline or saline 10 min reverse phase C-18 10 cm × 4,6 cm Hypersil column (Labservice
before PARA (100 mg/kg) or morphine (2 or 3 mg/kg) combination and Analytical, Bologna, Italy) packed with 3 mm ODS was used.

m
122 M. Sandrini et al. Inflamm. res.

The mobile phase composed of methanol 15%, acetonitrile 8% and

Nal þ Para þ Morp5

50 mM NaH2PO4, pH 2.8, with 0.2 mM ethylendiaminetetracetic acid
disodium salt and 200 mg/l sodium octyl sulfate, was pumped at a rate
of 1 ml/min
3,4-Dihydroxybenzylamine (DHBA), the internal standard, and

8:6 6 2:3
5-HT and 5-HIAA were used as standards. Standards and samples are
evaluated according to analite/DHBA ratio in a calibration curve.

Binding assay

Nal þ Para þ Morp3

The characteristics of 5-HT1A binding sites were evaluated according to
Gulati and Bhargava [33] using 6 concentrations of [3H]8-OH-DPAT

9:2 6 4:7
(0.18-6nM) (specific activity 142.9 Ci/mmol).
The characteristics of 5-HT2 binding sites were evaluated
according to Leysen and coworkers [34] with minor modifications.
Brain regions were homogenized in 5 ml of ice-cold 0.25 M sucrose (12
strokes of a Teflon pestle at 120 rpm) and centrifuged at 1300 × g for 10
min at 4 8C. This procedure was repeated, the combination of sucrose

Sal þ Para þ Morp5

Table 1. Influence of naloxone (Nal) treatment on the antinociceptive effect of paracetamol (Para) and morphine (Morp) in the hot-plate test.
supernatants was diluted with 10 ml of 50 mM Tris HCl, pH 7.7, and the
suspension was centrifuged at 3500 × g for 10 min. The pellet was

39:7 6 5:7¬;a
resuspended in 20 ml of Tris-HCl buffer and centrifuged once at
50,000 × g for 10 min; the pellet was then homogenized and diluted in

Nal (1 mg/kg i.p.) or saline (Sal, 2 ml/kg) were injected 10 min before Para (100 mg/kg i.p.) plus Morp (3 or 5 mg/kg s.c.) treatment.
Tris-HCl (about 300 mg protein/ml). Aliquots of membranes (800 ml)
were placed in plastic test tubes containing [3H]ketanserin (six
concentrations in 10% ethanol), methysergide (10 mM, dissolved in
Tris-HCl buffer to define nonspecific binding) or Tris buffer at 37 8C for
15 min. The mixture was filtered under reduced pressure through

Sal þ Para þ Morp3

Whatman GF/B glass fiber filters, previously soaked for 5 min in 0.5%
poliethylenimine with use of a Millipore vacuum pomp and rapidly

31:9 6 10:4¬;a
rinsed twice with 5 ml ice-cold Tris buffer. The filters were transferred
to plastic vials containing 6 ml of Packard Optifluor and shaken. The
vials were stored for 20 h at 4 8C in the dark.

Hot-plate started 20 min after the last treatment. % MPE ¼ percentage of the maximum possible effect.
The following concentrations were used: 0.05 to 4 nM [3H]ketan-
serin (specific activity, 87.5 Ci/mmol). The specific binding was 60 to
70% of the total binding for [3H]ketanserin and 80 to 90% for [3H]8-
Nal þ Veh þ Sal

OH-DPAT.
2:1 6 2:2

Statistics

The results obtained from motor activity were analyzed with Student’s
t-test. The results of binding experiments were analyzed according to
Sal þ Veh þ Morp5

the method of Rosenthal [35]. The equilbrium dissociation constant

ANOVA, *p < 0.05 vs. Sal þ Veh þ Sal. ap < 0.05 vs. Sal þ Veh þ Morp.

(kD) and maximum number of binding sites (Bmax) were evaluated

individually for each sample with six concentrations of labeled drug. A
20:9 6 2:5¬

two-way analysis of variance followed by 2 × 2 factorial analysis by

means of orthogonal comparisons was used to analyze the effect of
PARA þ morphine treatment, naloxone treatment and their com-
bination [36].
The data were examined by ANOVA followed by Student-
Newman-Keuls’ test when the effects of naloxone and PARA þ
Sal þ Veh þ Morp3

morphine were being evaluated separately.

7:2 6 0:9

Drugs

Paracetamol, was kindly provided by ACRAF SpA. Ancona, Italy, while

morphine hydrochloride, 5-HT, 5-HIAA, DHBA and naloxone were
purchased from Sigma Chemicals Co., St. Louis, MO, USA.
Sal þ Veh þ Sal

[3H]8-hydroxy-2-(D-n-proply-amino)tetralin ([3H]8-OH-DPAT)
and [3H]ketanserin were from Du Pont NEN, Co. Ltd, Milan, Italy.
Treatment

0:8 6 1:7

Formalin was obtained through Bracco Chemical Co., Milan, Italy.

Results
% MPE

Neither PARA alone (50 or 100 mg/kg) nor the single dose of
2 or 3 mg/kg or morphine was able to modify the reaction

m
Vol. 48, 1999 Potentiation of analgesic activity of paracetamol plus morphine 123

Fig. 1. Antinociceptive effect of paracetamol (PARA) and morphine

(MORP) combination in the hot-plate test. % MPE represents the
percentage of the maximum possible effect. MORP (0–5 mg/kg s.c.) Fig. 3. Influence of naloxone (NAL) pretreatment on the antinocicep-
was administered 10 min after PARA (50–100 mg/kg i.p.) and the rats tive effect of a combination of paracetamol (PARA) and morphine
were tested 20 min after morphine. Values are means 6 SEM of 8 rats (MORP) in the formalin test. NAL (1 mg/kg i.p.) or saline (SAL, 2 ml/
for each group. *p < 0.05 vs. control group (VEH þ MORP 0 mg/kg); kg) were injected 10 min before PARA (100 mg/kg i.p.) plus MORP (3
P
p < 0.05 vs VEH þ MORP (ANOVA followed by Student-Newman- mg/kg s.c.) treatment. Each histogram represents the total number of
Keuls’ test). flinches (mean 6 SEM) in phase 1 and in phase 2. *p < 0.05 vs. control
values (ANOVA followed by Student-Newman-Keuls’ test).

time in the HP test. Nor was PARA at the dose of 50 mg plus

2 and 3 mg/kg of morphine able to modify the % MPE in this 5 mg/kg, provoked a significant increase in % MPE that was
test. The combinations of 100 mg/kg of PARA with 2 or 3 potentiated by the combination with PARA (100 mg/kg)
mg/kg of morphine, on the other hand, significantly increased [F(1-32)¼6.2; p < 0.05].
% MPE and exerted an antinociceptive effect (Fig. 1). Naloxone (1 mg/kg) significantly prevented the anti-
The last combination of the two drugs showed a nociceptive effect of PARA plus morphine (3 or 5 mg/kg) in
potentiation effect, demonstrated by the factorial analysis the % MPE (Table 1); the factorial analysis showed a
[F(1-32)¼10.4; p < 0.01], [F(1-32)¼7.6; p < 0.01], for 2 and 3 negative interaction [F(1-28)¼5.4; p < 0.05]; [F(1-28)¼4.4;
mg/kg of morphine, respectively. Morphine, at the dose of p < 0.05].
PARA 100 mg/kg and morphine 2 or 3 mg/kg were
chosen to test the antinociceptive activity in the formalin
test, as these doses did not exert a significant antinociceptive
effect when given alone; a combination of the two drugs
seemed to be the most suitable way of assessing eventual
synergy. The combinations of PARA 100 mg/kg and MORP
2 or 3 mg/kg had a significant effect in both phases of the
formalin test (Fig. 2). On the other hand, the interaction test
showed significant potentiation in the first phase
[F(1-20¼11.5; p < 0.01]; [F(1-20)¼4.8; p < 0.05] for the two
doses of morphine, whereas, in the second phase, only the
combination of PARA plus the dose of 3 mg/kg morphine
caused the potentiation effect [F(1-20)¼6.34; p < 0.05];
[F(1-20)¼1.1; p > 0.05] for the combination of 2 mg/kg.
Naloxone prevented the effect of the combination of
PARA 100 mg/kg and morphine 3 mg/kg in both phases of
the test [F(1-20)¼5.07; p < 0.05], first phase, and [F(1-20)¼8.5;
p <0.01], second phase. Naloxone alone slightly increased
the number of flinches (Fig. 3). None of these combinations
affected motor activity, the number of movements being
1274 6 60 for PARA 100 mg/kg plus morphine 3 mg/kg and
Fig. 2. Antinociceptive effect of a combination of paracetamol (PARA) 1243 6 48 for the control group (p > 0.05).
and morphine (MORP) in the formalin test. PARA (100 mg/kg i.p.) and In order to exclude any pharmacokinetic interaction
MORP (2 or 3 mg/kg s.c.) were administered 10 min after each other
and the rats were tested 20 min after morphine. Each histogram
between naloxone, PARA and morphine we evaluated the
represents the total number of flinches (mean 6 SEM) in phase 1 and in levels of paracetamol in the sera; the concentrations of
phase 2. *p < 0.05 vs. control values (ANOVA followed by Student- PARA were 12:8 6 2:3, 15:2 6 1:6 and 13:1 6 2:1 mg/ml for
Newman-Keuls’ test). saline þ PARA (100 mg/kg)-treated, saline þ PARA þ

m
124 M. Sandrini et al. Inflamm. res.

Fig. 4. The effect of the combination of paracetamol (PARA) and

morphine (MORP) on serotonin content in cortical and pontine
membranes. PARA (100 mg/kg i.p.) and MORP (3 mg/kg s.c.) were
administered 10 min after each other. The rats were killed at the end of
the hot-plate test and brain areas were weighed and frozen at ¹80 8C
until assayed. Values are expressed as mean 6 SEM for 6 rats for each
group. *p < 0.05 vs. control values (ANOVA followed by Student-
Newman-Keuls’ test).

morphine (3 mg/kg)-treated and naloxone (1 mg/kg) þ

PARA þ morphine-treated rats, respectively (p > 0.05).
The levels of serotonin in the cortex and in the pons were Fig. 5. Influence of naloxone (NAL) pretreatment on the effect of
significantly increased by the antinociceptive combination of paracetamol (PARA) and morphine (MORP) on serotonin (5-HT) and
PARA 100 mg/kg and morphine 3 mg/kg, while they did not 5-hydroxyindolacetic acid (5-HIAA) levels in the rat brain. NAL (1 mg/
change the values of controls when given alone (Fig. 4); the kg i.p.) or saline (SAL, 2 ml/kg) was administered 10 min before PARA
interaction test showed a potentiation in the two areas (100 mg/kg) and 20 min before MORP (3 mg/kg); the rats were killed at
the end of the hot-plate test and brain areas were weighed and frozen at
[F(1,20)¼15.2; p < 0.01], [F(1,20)¼28.2; p < 0.01] for pons and ¹80 8C until assayed. Each value represents the mean 6 SEM of 6
cortex respectively. Pretreatment with naloxone prevented separate experiments. *p < 0.05 vs. control values (ANOVA followed
the increase of 5-HT in both areas (Fig. 5); [F(1,20)¼10.7; by Student-Newman-Keuls’ test).
p < 0.01], [F(1,20)¼12.2; p < 0.01] for pons and cortex
respectively.
The levels of 5-HIAA in the cerebral cortex and in the stimulus, species and strain of animals, the type of recording
pons increased but not significantly; this lack of significance device involved in the experimental model.
could be due to the wide variability of the data obtained with Nevertheless, in previous studies the antinociceptive
respect to size of the sample (Fig. 5). effect of phenazone [23] and paracetamol [30] has been
The combination of the same doses of PARA and tested by the hot-plate and formalin tests, which are
morphine provoked a decrease in the number of 5-HT2 generally considered a reliable method of measuring
receptors in the cortex (Table 2), [F(1-20)¼7.0; p < 0.01], but NSAID antinociceptive activity. Indeed, the hot-plate test
not in the pons, while the two drugs did not change the has been widely employed to study nociception, and, in a
maximum number of binding sites in the two areas when review concerning this test, Hunskaar and coworkers [37]
given alone. Naloxone partially prevented the decrease in the demonstrated similar potencies for ASA and morphine in the
Bmax but the interaction test showed no significance rats.
[F(1-20)¼1.76; p >0.05]. The affinity constant remained The behavioural response to the injection of formalin is
unchanged. The characteristics of 5-HT1A receptors never biphasic, with an acute phase followed by a quiescent period
changed, either in the cortex or in the pons (Table 3). and then a prolonged response with a maximum between 10
and 40 min [3]. It has been suggested that the early phase is
caused by a direct effect of formalin on nociceptors, whereas
Discussion the second phase is a tonic response in which inflammatory
processes are involved and neurons of the dorsal horn of the
The evaluation of antinociceptive drug activity in animal spinal cord are activated [38]; no obvious inflammatory state
models depends on several factors, such as the type of has been found 10 min after formalin injection. Thus, the

m
Vol. 48, 1999 Potentiation of analgesic activity of paracetamol plus morphine 125

Table 2. Influence of naloxone (Nal) treat- Cortex Pons

ment on the effect of a combination of
paracetamol (Para) and morphine (Morp) on Bmax kD Bmax kD
the characteristics of 5-HT2 receptors in the Treatment (fmol/mg prot) (nM) (fmol/kg prot) (nM)
rat brain.
Sal þ Veh þ Sal 221:8 6 6:3 1:9 6 0:2 33:4 6 3:9 1:3 6 0:10
Nal þ Veh þ Sal 259:4 6 10:3 1:6 6 0:10 35:4 6 3:9 1:3 6 0:1
Sal þ Para þ Sal 222:3 6 13:7 1:6 6 0:30 n.d. n.d.
Sal þ Morp þ Sal 238:5 6 7:3 1:5 6 0:3 38:4 6 5:4 1:51 6 0:10
Sal þ Para þ Morp3 155:0 6 7:1¬ 20 6 0:2 31:4 6 3:0 1:2 6 0:10
Nal þ Para þ Morp3 202:7 6 9:4 1:9 6 0:1 n.d. n.d.

Rat were injected with saline (Sal, 2 ml/kg), vehicle (Veh, 2 ml/kg), Nal (1 mg/kg i.p.), Para (100
mg/kg i.p.) and Morp (3 mg/kg s.c.). They were killed at the end of the experiment and the brain
were stored at ¹80 8C until assayed. Each value represents the mean 6 SEM of 6 separate
experiments and were derived from Rosenthal plot. Bmax, maximum binding capacity; kD,
equilibrium dissociation constant. ANOVA, p < 0.05 vs. Sal þ Veh þ Sal.

antinociceptive non-antiinflammatory properties of drugs It has been previously shown that m opiate receptor
can be evaluated immediately after formalin injection and agonists suppress the activity of formalin in all the phases of
for 10 minutes thereafter [39]. In this paper we demonstrate the test [40], so we may hypothesize that the block of
that PARA and morphine, when given in combination, even morphine activity in the second phase of the formalin test
in subactive doses, exert a significant analgesic effect in both induced by naloxone can explain also naloxone antagonism
the hot-plate and the formalin test. on the effect of PARA and morphine in combination. On the
The combination of doses of morphine and PARA that other hand, all the doses used were subactive, and when the
are ineffective, when given alone, may be of interest from the activity of morphine was blocked by naloxone the anti-
therapeutic point of view, for the combination showed actual nociceptive activity of PARA did not make itself felt because
additional antinociceptive effects. This combination is largely of the low dose.
used in the treatment of mild to moderate pain in humans. In a recent paper it was suggested that the mechanism of
Our data support the hypothesis that the analgesic action by which some NSAIDs exert their analgesic effect
activity of PARA and morphine in association might occur involves a number of neurotransmitter systems, particularly
as the result of opiatergic activation, it having been adrenergic, serotonergic and opiatergic. An increase in 5-HT
demonstrated that naloxone blocks the potentiation effect. levels in brain areas has been observed by many authors after
We have previously shown that naloxone significantly NSAIDs or PARA injection. Morphine has also been shown
prevents the action of PARA in the hot-plate test and in to act on serotonergic pathways in a way similar to that of
the first phase of the formalin test but does not affect PARA PARA [41], increasing the concentration of 5-HT in the
activity in the second phase, while the effect of morphine is cerebral cortex and decreasing the number of 5-HT2
prevented in both phases of the test [30], which suggests that receptors in the same area. In our experimental conditions
the mechanism of action may be different in the second it is unlikely that 5-HT1A receptors may be involved, at least
phase of the formalin test where inflammatory processes are in the cerebral cortex, because of the different distribution of
involved. In the present paper it has been shown that this type of receptors in CNS. Indeed some authors showed
naloxone prevents this effect of PARA and morphine that 5-HT1A and 5-HT2 receptors are involved in antinoci-
combinations in all the phases of the formalin test. ception at spinal level [42, 43].
PARA and morphine in combination may activate opiate
receptors, which in turn increase 5-HT levels, at least in the
Table 3. Effect of a treatment with paracetamol (Para) and morphine cerebral cortex and pons. Naloxone could therefore block the
(Morp) on the characteristics of 5-HT1A receptors in the cortex of the rat
brain.
opiatergic pathways that in turn activate the serotonergic
system [30].
Treatment Bmax kD Opioid inhibition of GABAergic transmission is poten-
tiated by inhibitors of the enzymes cyclooxygenase and
(fmol/mg prot) (nM) 5-lipoxygenase [44] and the sinergistic analgesic action of
opioids and NSAIDs in the CNS may be mediated by COX1
Veh þ Sal 264:3 6 21:3 1:8 6 0:3
rather than COX2 [45].
Para þ Sal 253:1 6 10:7 1:7 6 0:3
Veh þ Morp 212:7 6 18:4 1:8 6 0:3 Although the mechanism by which PARA interact with
Para þ Morp 266:8 6 22:8 2:1 6 0:3 opiatergic and serotonergic systems has still not been
completely elucidated, we suggest that PARA and morphine
Rat were injected with saline (Sal, 2 mg/kg), vehicle (Veh, 2 ml/kg i.p.), might act through common pathways in the CNS (i.e.
paracetamol (Para, 100 mg/kg i.p.) and morphine (Morp, 3 mg/kg s.c.). serotonergic and opiatergic) that are involved in pain
They were killed at the end of the experiment and the brains were stored modulation, with selective effects in discrete brain areas
at ¹80 8C until assayed. Each value represents the mean 6 SEM of 6
separate experiments and were derived from Rosenthal plot. Bmax, [46]. In a recent paper we observed that PARA was also able
maximum binding capacity (fmol/mg protein); kD, equilibrium to bind to [3H]-naloxone binding sites, although with minor
dissociation constant (nM). potency than morphine, where it might exert a morphine-like

m
126 M. Sandrini et al. Inflamm. res.

action, the final effect being a potentiation of the analgesic [18] Pelissier T, Alloui A, Paeile C, Eschalier A. Evidence of a central
activity when the two drugs are given in combination [30]. antinociceptive effect of paracetamol involving spinal 5-HT3
Many findings indicate the complexity of the central pain receptors. Neuroreport 1995;6:1546–8.
[19] Eide PK, Hole K. The role of 5-hydroxytryptamine (5-HT)
modulating system, where a network of neurotransmitter receptor subtypes and plasticity in the 5-HT systems in the
pathways (i.e. monaminergic and peptidergic) may interact regulation of nociceptive sensitivity. Cephalalgia 1993;13:75–85.
to modulate and perform many behavioural changes [20] Murphy AZ, Murphy RM, Zelman FP. Role of spinal serotonin 1
including nociception [22]. receptor subtypes in thermally and mechanically elicited
The results obtained using this experimental procedure nociceptive reflexes. Psychopharmacology 1992;108:123–30.
could thus probably be transferred to human beings; in fact, [21] Basbaum A, Fields HL. Endogenous pain control mechanisms. In:
Melzack HL, Wall EJ, editors. Textbook of pain. New York:
in human therapeutics as well there are products indicated Churchill Livingstone 1985;142–52.
for mild pain containing PARA and opiate derivatives (like [22] Björkman R. Central antinociceptive effects of non-steroidal anti-
codeine) at ineffective doses when used alone [47]. inflammatory drugs and paracetamol. Acta Anaesthesiol Scan
1995;39:7–43.
[23] Sandrini M, Vitale G, Sternieri E, Bertolini A, Pini LA. Effect of
chronic treatment of phenazone on the hot-plate test and
[3H]serotonin binding sites in pons and cortex membranes in the
References rat. Pharmacology 1993;47:84–90.
[24] Pini LA, Sandrini M, Vitale G. Involvement of brain serotonergic
[1] Brune K, Beck WS, Geisslinger G, Menzel-Soglowek S, Peskar BM, system in the antinociceptive action of acetylsalicylic acid in the
Peskar BA. Aspirin-like drugs may block pain independently of rat. Inflamm Res 1995;44:30–5.
prostaglandin synthesis inhibition. Experientia 1991;47:257–61. [25] Pini LA, Sandrini M, Vitale G. The antinociceptive action of
[2] Brune K, Menzel-Soglowek S, Zeilhofer HU. Differential analge- paracetamol is associated with changes in the serotonergic system
sic effect of aspirin-like drugs. Drugs 1992;44:52–60. in the rat brain. Eur J Pharmacol 1996;308:31–40.
[3] Malmberg AB, Yaksh TL. Antinociceptive actions of spinal anti- [26] Bineau-Thourottes M, Godefroy F, Weil-Fugazza J, Besson JM.
inflammatory agents on the formalin test in the rat. J Pharmacol The effect of morphine on the potassium evoked release of
Exp Ther 1992;263:126–46. tritiated 5-HT from spinal cord slices in the rat. Brain Res
[4] Clissold SP. Paracetamol and Phenacetin. Drugs 1986;32:46–59. 1984;291:293–9.
[5] McCormack K. The Spinal actions of nonsteroidal anti-inflam- [27] Bodnar RJ, Kordower JH, Wallace MM, Tamir H. Stress and
matory-drugs and the dissociation between their anti-inflamma- morphine analgesia: alterations following p-chlorophenylalanine.
tory and analgesic effects. Drugs 1994;47:28–45. Pharmacol Biochem Behav 1980;14:654–51.
[6] Dray A, Bevan S. Inflammation and hyperalgesia: highlighting [28] Yang SW, Zhang ZH, Wang R, Xie YF, Qiao JT, Dafny N.
the team effort. Trends Pharmacol Sci 1993;14:287–90. Norepinephrine and serotonin induced antinociception are blocked
[7] Cashman JN. The mechanisms of action of NSAIDs in analgesia. by naloxone with different dosages. Brain Res Bull
Drugs 1996;52:13–23. 1994;35:113–7.
[8] Lane NE. Pain management in osteoarthritis: the role of COX-2 [29] Drissen B, Reiman W. Interaction of the central analgesic
inhibitors. J Rheumatol 1997;24:20–4. tramadol with the uptake and release of 5-hydroxytryptamine in
[9] Yamamoto T, Nozaki-Taguchi N. Analysis of the effects of vitro. Br J Pharmacol 1992;105:147–51.
cyclooxygenase COX-1 and COX-2 in spinal nociceptive [30] Pini LA, Vitale G, Ottani A, Sandrini M. Naloxone-reversible
transmission using indomethacin, a non-selective COX inhibitor, antinociception by paracetamol in the rat. J Pharmacol Expl Ther
and NS-398, a COX-2 selective inhibitor. Brain Res 1997;280:934–40.
1996;739:104–10. [31] Hole K, Tjølsen A. The tail flick and formalin tests in rodents.
[10] Laird JMA, Herrero JF, Garcia de la Rubia P, Cervero F. Changes in room temperature as confounding factor. Pain
Analgesic activity of the novel COX-2 preferring NSAID, 1993;53:247–51.
meloxicam in mono-arthritic rats. Central and peripheral compo- [32] Grossi G, Bargossi A, Sprovieri G, Benagozzi V, Pasquale R.
nents. Inflamm Res 1997;46:203–10. Full automation of serotonin determination by column switching
[11] Okuyama S, Aihara H. The mode of action of analgesic drugs in and HPLC. Chromatographia 1990;30:61–8.
adjuvant arthritic rats as an experimental model of chronic [33] Gulati GA, Bhargava HN. Down regulation of hypotalamic
inflammatory pain: possible central analgesic action of acid 5-HT1A receptors in morphine-astinent rats. Eur J Pharmacol
non-steroidal anti-inflammatory drugs. Jpn J Pharmacol 1990;82:253–9.
1984;35:95–103. [34] Leysen EJ, Niemegeers CJE, van Neuten JM, Laduron PM.
[12] Abdel-Halim MS, Sjøquist B, Anggard E. Inhibition of prosta- [3H]Ketanserin (R41 468), a selective [3H]-ligand for serotonin2
glandin synthesis in rat brain. Acta Pharmacol Toxicol receptor binding sites. Mol Pharmacol 1981;21:301–14.
1978;43:266–72. [35] Rosenthal H. A graphic method for the determination and
[13] Warner R, Hudson-Howard L, Johnston C, Skolnik M. Serotonin presentation of binding parameters in a complex system. Anal
involvement in analgesia induced by transcranial electrostimula- Biochem 1967;20:520–32.
tion. Life Sci 1990;48:1131–8. [36] Snedecor GW, Cochran WJ. Statistical Methods. Ames: Iowa
[14] Bensemana D, Gascon AL. Relationship between analgesia and State University Press, 1980;298–307.
turnover of brain biogenic amines. Can J Physiol Pharmacol [37] Hunskaar S, Berge OG, Hole K. A modified hot-plate test
1978;56:721–30. sensitive to mild analgesics. Behav Brain Res 1986;21:101–7.
[15] Groppetti A, Braga PC, Biella G, Parenti M, Rusconi L, Mante- [38] Tjølsen A, Berge OG, Hunskaar S, Rosland JH, Hole K. The
gazza P. Effect of aspirin on serotonin and met-enkephalin in formalin test: an evaluation of the method. Pain 1992;51:5–17.
brain: correlation with the antinociceptive activity of the drug. [39] Hunskaar S, Hole K. The formalin test in mice: dissociation between
Neuropharmacology 1988;27:499–505. inflammatory and non-inflammatory pain. Pain 1987;30:103–14.
[16] Eide PK, Hole K. Different role of 5-HT1A and 5-HT2 receptors in [40] Malmberg AB, Yaksh TL. Pharmacology of the spinal action of
spinal cord in the control of nociceptive responsiveness. ketorolac, morphine, ST-91, U50488H, and L-PIA on the formalin
Neuropharmacology 1991;30:727–31. test and an isobolographic analysis of the NSAID interaction.
[17] Alheider AA. Antinociceptive effect of Ketaserin in mice: Anesthesiology 1993;79:270–81.
involvement of supraspinal 5-HT2 receptors in nociceptive [41] Richardson D. Serotonin and nociception. Ann NY Acad Sci
transmission. Brain Res 1991;543:335–40. 1990;600:511–9.

m
Vol. 48, 1999 Potentiation of analgesic activity of paracetamol plus morphine 127

[42] Xu W, Cui X, Han JS. Spinal serotonin 1A and 1C/2 receptors [45] Vaughan CW. Enhancement of opioid inhibition of GABAergic
mediate supraspinal m-opioid-induced analgesia. Neuroreport synaptic transmission by cyclo-oxygenase inhibitors in the
1994;5:2665–8. rat periaqueductal grey neurones. Br J Pharmacol 1998;123:1479–81.
[43] Dazebrink RM, Gebhart GF. Evidence that spinal 5-HT1, 5-HT2 [46] Tao R, Auerbach SB. Involvement of the dorsal raphe but not
and 5-HT3 receptors subtypes mudulate responses to noxious median raphe nucleous in morphine-induced increases in seroto-
colorectal distension in the rat. Brain Res 1991;538:64–75. nin release in the rat forebrain. Neuroscience 1994;68:553–61.
[44] Vaughan CW, Ingram SL, Connor MA, Christie MJ. How [47] Bentley HA, Head TV. The additive analgesic efficacy of
opioids inhibit GABA-mediated neurotransmission. Nature acetaminophen, 1000 mg and codeine, 60 mg, in dental pain.
1997;390:611–4. Clin Pharmacol Ther 1987;42:634–40.