Life Sciences 77 (2005) 85–95

www.elsevier.com/locate/lifescie

Enhancement of fentanyl antinociception by subeffective doses of

nitroparacetamol (NCX-701) in acute nociception and in

carrageenan-induced monoarthritis

Gema Gaitan, F. Javier Ahuir, Juan F. HerreroT

Departamento de Fisiologıa, Facultad de Medicina, Campus Universitario, Universidad de Alcala,

Alcala de Henares, 28871 Madrid, Spain

Received 14 July 2004; accepted 21 December 2004

Abstract

We have reported that subanalgesic doses of new generation non-steroidal anti-inflammatory drugs

(NSAIDs) enhance the antinociceptive activity of the mu-opiate fentanyl, and the duration of its effect, in

acute nociception. Since this therapy is intended for situations of hyperalgesia, we have compared the

antinociceptive activity of fentanyl in the absence and in the presence of subeffective doses of NCX-701

(nitroparacetamol) in normal animals and in animals with carrageenan-induced monoarthritis. Subanalgesic

dose of NCX-701 did not modify any of the nociceptive responses on its own but reduced the ID50 of

fentanyl more than two-fold in both the normal and sensitized states. When administered alone, full recovery

from fentanyl was always observed within 15 to 20 minutes, however, full recovery was not observed in the

presence of NCX-701. Naloxone was unable to reverse the effect, suggesting a possible reduction of other

opiate-mediated secondary effects. We therefore studied the possibility that combining administration of

fentanyl and nitroparacetamol (NCX-701) would reduce the development of acute tolerance to fentanyl in

behavioral experiments. Acute tolerance to fentanyl in behavioral nociceptive reflexes was developed within

72 h after the constant infusion of the drug, whereas in animals treated with small doses of NCX-701

tolerance was not observed. In summary, our results, both in normal animals and in animals with hyperalgesia,

show that fentanyl antinociception can be strongly potentiated with subanalgesic doses of the NSAID NCX-

0024-3205/$ -

doi:10.1016/j.l

T Correspond

E-mail add

see front matter D 2005 Elsevier Inc. All rights reserved.

fs.2004.12.022

ing author. Tel.: +34 91 885 45 16; fax: +34 91 885 45 90.

ress: juanf.herrero@uah.es (J.F. Herrero).

G. Gaitan et al. / Life Sciences 77 (2005) 85–9586

701 and that the development of acute tolerance to fentanyl in normal animals is prevented by this

combination of drugs.

D 2005 Elsevier Inc. All rights reserved.

Keywords: Pain; Analgesia; Spinal cord; Opioids; Non-steroidal anti-inflammatory drugs; Nitric oxide; Wind-up

Introduction

The combined administration of some cyclooxygenase (COX) inhibitors with other analgesic drugs

induces an enhancement of some of their therapeutic actions, that has led to combined prescriptions in

the treatment of severe pain. This therapy has been used either to enhance the analgesic properties or to

decrease the incidence of unwanted side-effects caused by non-steroidal anti-inflammatory drugs

(NSAIDs), especially those related to gastro-intestinal (GI) and renal toxicity. Among the best studied is

the combination of NSAIDs and opiates, frequently used in the treatment of postoperative pain, and in

the treatment of cancer to decrease the required dosage of opiates, and thereby to reduce the incidence of

adverse effects (Burns et al., 1991). This therapy seems to be effective in the treatment of postoperative

pain and in pain due to spinal cord sensitization (Gillies et al., 1987; Laitinen and Nuutinen, 1992; Bosek

and Miguel, 1994), that severely impoverishes the quality of life of many patients. In most animal

experiments, the administration of NSAIDs like ketorolac (Malmberg and Yaksh, 1993; Maves et al.,

1994; Lashbrook et al., 1999) potentiates morphine antinociception in different situations of

sensitization. In all cases, the doses of the opiate and the NSAID chosen were pharmacologically

effective and, therefore, likely to by accompanied by the development of secondary effects induced by

either opiates (i.e. constipation, blood pressure changes, tolerance. . .) or NSAIDs (i.e. gastric and renal

problems, among others). These secondary effects might be, at least in the long term, stronger than those

observed when the drugs were given separately. In any case, tolerance and even dependence to the opiate

treatment did not seem to be modified by the presence of the NSAID, as the effect observed was reversed

by the administration of the opioid receptor antagonist naloxone. We observed that the administration of

subanalgesic doses of new generation NSAIDs like dexketoprofen trometamol (Gaitan and Herrero,

2002) enhances the antinociceptive activity of the mu-opiate fentanyl, and the duration of its effect, in

acute nociception by a naloxone-independent mechanism. Nevertheless, this type of therapy is intended

for situations of sensitization and since our observation was made on normal animals in the acute

nociceptive situation, it was important to assess if the enhancement of fentanyl antinociceptive activity

by new generation NSAIDs also happens in animals with spinal sensitization. We have studied the

antinociceptive activity of fentanyl in normal animals and in animals with carrageenan-induced

monoarthritis in the absence and presence of subeffective doses of NCX-701 (nitroparacetamol). This

might open a new perspective in the treatment of pain, since the low doses of NSAIDs used, the

significant increase in the duration of the effect and the lack of reversal by naloxone suggest reduced side

effects, and, perhaps, even a reduction in the development of acute tolerance. This is supported by the

fact that the effectiveness of fentanyl was not only enhanced in terms of antinociceptive potency, but also

in the duration of the effect. We therefore wondered if the combined administration of fentanyl and

NCX-701 would reduce the development of acute tolerance to the continuous administration of fentanyl.

We tested this hypothesis in behavioral experiments performed in rats with a constant subcutaneous (s.c.)

G. Gaitan et al. / Life Sciences 77 (2005) 85–95 87

infusion of fentanyl over the course of five days. Preliminary results have been published in abstract

form (Gaitan et al., 2003).

Materials and methods

Electrophysiological experiments

Experiments were performed on 250 to 330 g male Wistar rats following the technique described

previously in detail (Herrero and Headley, 1991; Solano and Herrero, 1997; Romero-Sandoval et al.,

2003). Briefly, rats were initially anaesthetized with halothane (5% in oxygen for induction and 2% for

maintenance). Cannulation of the trachea, one carotid artery to register the blood pressure, and two

superficial jugular veins for the administration of the maintenance anesthetic and drugs, respectively,

was performed under this anesthesia. After the surgery, halothane was ceased and the anesthesia

continued with alpha-chloralose (Sigma) at an induction dose of 50 mg/kg and a dose of 30 mg/kg/h, by

perfusion pump, for maintenance at a rate of 1 ml/h. This rate allowed for correct hydration of the

animal. The right hind limb was fixed in inframaximal extension in a Perspex block using plaster.

Throughout the experiment, the core body temperature of the rat was monitored and maintained (36.5–

37.5 8C) by means of a heating blanket connected to a rectal thermal probe via an automatic feedback

control unit.

Nociceptive withdrawal reflexes were recorded as single motor units (SMU) by means of bipolar

tungsten electrodes inserted percutaneously into muscles of the right hind limb. The units were activated

by both noxious natural (pinch) and electrical stimuli (wind-up) applied to the cutaneous receptive fields

in three minute cycles. Noxious mechanical stimulation was applied during 10 s over an area of 14 mm2,

using a computer controlled pincher at a force of 200 mN over threshold intensity. The threshold was

considered as the minimum pressure required to trigger the withdrawal reflex for at least ten seconds of

stimulation. Electrical stimulation was applied through two 0.2 mm needles, with 16 pulses of 2 ms

width, 1 Hz and an intensity of twice the threshold current for C-fiber responses. Only units with a stable

firing rate and summation of responses to constant intensity repetitive electrical stimulation (wind-up)

were selected for the experiments (Herrero et al., 2000). The administration of drugs only commenced if

the responses recorded were stable, with a variation lower than 20%, in at least three cycles of

stimulation.

Induction of monoarthritis was achieved by the administration of 50 Al of carrageenan E (Sigma)

10mg/ml, in distilled water, into the right knee cavity under brief halothane anesthesia (5% in oxygen for

induction and 2% for maintenance), 16h before the experiment.

Two dose-response curves of intravenous fentanyl were studied in the same unit in four groups of

animals, leaving a minimum gap of one hour between the measurements. In two groups of experiments

(normal animals, n = 22 and animals with carrageenan-induced inflammation, n = 7) fentanyl was

injected in log 2 cumulative doses every two cycles of stimulation (6 minutes) starting with a dose of 1

Ag/kg and increasing the dose until the responses were reduced to less than of 25% of control (32 or 64

Ag/kg). The control value was the average of the three responses previous to the administration of the

first dose. In another two groups (normal animals, n = 7, and animals with carrageenan-induced

inflammation, n = 7), the second dose-response curve of fentanyl was preceded by the administration of

three cumulative doses (4.25, 8.5 and 17 mg/kg) of nitroparacetamol (NCX-701). Although these doses

G. Gaitan et al. / Life Sciences 77 (2005) 85–9588

have been shown to be ineffective in reducing nociceptive responses in the same preparation (Romero-

Sandoval et al., 2003), the possible effect of nitroparacetamol was also studied.

Nitroparacetamol (NicOx) was dissolved in dimethyl sulfoxide (Sigma, 50%) and polyethylene glycol

(Panreac, 50%, final concentration of 50 Amol/ml) and diluted in saline. Fentanyl (Sigma) was dissolved

in saline. In addition, the effect of fentanyl in the presence of nitroparacetamol was challenged with the

opioid receptor antagonist naloxone (Sigma, 200 Ag/kg 45 min after fentanyl) and with the a2-

adrenoceptor antagonist atipamezol (Sigma, 6 min after naloxone). The reduction of responses by

fentanyl was compared between the experiments performed in the presence and absence of

nitroparacetamol, and presence and absence of inflammation, as well as between the first and the

second dose-response curve, with the one-way analysis of variance (ANOVA) with post-hoc Dunnett’s

test. Comparison of responses of fentanyl between groups was performed using the non-parametric

Mann-Whitney U-test. The comparisons of ID50 were made by the Student’s t-test. (GraphPad-Prism

and GraphPad-Instat for windows). Data are expressed as mean F S.E.M. The animals were killed with

an overdose of sodium pentobarbital (Euta-Lender, Normon).

Behavioral experiments

Male Wistar rats (250–320 g) were tested twice a day (10:00 a.m. and 5.30 p.m.) on five consecutive

days for behavioral responses to mechanical stimulation. Mechanical sensitivity was assessed through

the measurement of foot withdrawal frequencies to a sequential series of calibrated von Frey filaments

(80, 100 and 200 mN) applied to the plantar surface of the paw. Each test was made on the two paws and

consisted of 10 applications of each filament, with each application lasting no more than 3 seconds. The

occurrence of foot withdrawals recorded in animals treated with NCX-701 was compared with that in

animals treated with vehicle. In all cases the animals were allowed to habituate for 4 days to the testing

environment and for the time necessary to remain calm when handled by the experimenter. The animals

were placed in individual boxes on a 12 h light-dark cycle and were provided with food and water ad

libitum.

The first test was made on the morning of the first day (10:00 a.m.) and was followed by the

implantation of an Alzet 2ML1 osmotic mini-pump (mean pumping rate 10 Al/h) s.c. on the back of

the animal through a small incision, under brief halothane anesthesia (at the same concentrations as

given for previous experiments). The pumps were filled with fentanyl (Sigma) dissolved in saline to

be delivered at a rate of about 0.15 mg/kg per day. In one group of experiments, NCX-701 was

dissolved as for electrophysiological experiments and administered p.o. twice a day, immediately after

each test, at a dose of 100 Amol/kg (28 mg/kg). The rest of the animals were treated with the

equivalent dose of vehicle. Drug doses and time of administration were chosen on the basis of

preliminary experiments and previous studies investigating these compounds alone in animal models

of nociception (Gaitan et al., 2003; Romero-Sandoval et al., 2003). The dose of 100 Amol/kg of NCX-

701 was not antinociceptive either i.v. in single motor unit experiments (Romero-Sandoval et al.,

2003) or p.o. in preliminary behavioral experiments in which doses of at least 250 Amol/kg were

required for significant effectiveness.

The experimenter was unaware of the treatment of the animals. Data were normalized as percentage of

change of the baseline test prior to the implantation, and expressed as mean F S.E.M. Comparison of

data was made with of the non-parametric Mann-Whitney U-test (GraphPad Prism and GraphPad Instat

for Windows). The animals were used for one procedure only and were humanely killed on completion

G. Gaitan et al. / Life Sciences 77 (2005) 85–95 89

of testing by an overdose of pentobarbitone. European Union legislation regulating animal experiments

was followed and all efforts were made to minimize suffering and to reduce the number of animals used.

Results

Electrophysiological experiments

The mean number of spikes recorded in the control responses to noxious mechanical stimulation were

similar in all the experimental groups, both in normal animals: 311 F 24, and in animals with

inflammation 327 F 29. The administration of carrageenan induced a significant increase in the

perimeter of the knee: 11+2% (P b 0.01), and no differences were observed in the amount of

inflammation in the two groups of monoarthritic animals.

The administration of fentanyl induced a dose-dependent inhibition of responses to pinch in all the

experiments performed. No significant differences were observed in the potency of fentanyl alone in

normal animals (ID50 of 21 F 2 Ag/kg) and in animals with inflammation (29 F 2 Ag/kg, Fig. 1). Theminimum effective dose (MED) was 16 Ag/kg in normal animals and 32 Ag/kg in monoarthritic animals.

The administration of a cumulative dose of 17 mg/kg NCX-701 did not modify any of the nociceptive

responses: 116 F 11% in normal animals and 94F4% in monoarthritic rats. The antinociceptive activity

Fig. 1. Pooled data of the antinociceptive effects obtained with the cumulative i.v. administration of the A-opioid agonist

fentanyl previous to (Fent) and after the administration of 17 mg/kg of NCX-701 Fent (N) in normal animals (Norm) and in

animals with monoarthritis (Infl). The diagram represents the percentage of reduction of single motor unit responses to noxious

mechanical stimulation. The potency of fentanyl was enhanced by more than two-fold in the presence of NCX-701 either in the

normal or sensitized state (inset). The administration of the same dose of NCX-701 by itself did not significantly reduce the

responses (**P b 0.01, comparison vs. control response using the one-way ANOVA, with the post-hoc Dunnett test; #P b 0.05;##P b 0.01, comparison between Fent (N) and Fent using the non-parametric Mann-Whitney U-test).

G. Gaitan et al. / Life Sciences 77 (2005) 85–9590

of fentanyl was higher in the presence of NCX-701 in both groups of experiments. In normal animals the

ID50 was 8 F 1.4 Ag/kg, more than two-fold lower and significantly different than that observed in the

absence of the NSAID (P b 0.01, Fig. 1). The MED was of 4 Ag/kg, 8-fold lower than in the absence of

NCX-701. An enhancement of the potency of fentanyl was also observed in the inflammatory state, were

the ID50 was 11.6F1.2 Ag/kg (P b 0.01), more than two-fold lower than in the absence of the NSAID. In

this case, however, the MED of 16 Ag/kg, though lower than that observed in the absence of NCX-701,

was not as low as in animals without inflammation.

The animals recovered from the antinociceptive effect of fentanyl in 15 to 20 min when it was

administered on its own (Fig. 2). In the presence of NCX-701, however, a full recovery was never

observed. In animals without inflammation, the level of nociceptive responses remained below 25% of

the control response for at least 45 minutes (21 F 7% 45 min after the administration, P b 0.01). In

monoarthritic animals a partial, but not full, recovery was observed 45 min after the last dose studied:

50 F 15% of control, and this response was still significantly different from the control response (P b

0.05, Fig. 2). The administration of 200 Ag/kg of the non-selective opioid receptor antagonist naloxone

was unable to reverse the effect observed. The antinociceptive effect was also challenged with 100 Ag/kgof the selective a2-adrenoceptor antagonist atipamezol, since a2-adrenoceptors have been shown to be

involved in opiate analgesia in situations of inflammation (Herrero and Solano, 1999), but no reversion

of the effect was observed.

Fig. 2. Pooled data of the antinociceptive effects of fentanyl in response to noxious mechanical stimulation after the

administration of the highest cumulative doses (32 and 64 Ag/kg) and recovery from the effect observed 15 (R15), 30 (R30) and

45 min later (R45). Recovery from fentanyl was rapid when injected alone, but full recovery was not observed within 45 min of

recording when administered in the presence of NCX-701, (Fent N) both in normal animals (Norm) or in animals with

monoarthritis (Infl). A slight smaller level of responses was observed at 45 min of recovery when compared to that at 30 min,

although this difference was not significant. The effect was not reversed by the administration of 200 Ag/kg of naloxone

(200Nal) or 100 mg/kg of atipamezol (100Atip, inset). Statistical significance and layout as for Fig. 1.

G. Gaitan et al. / Life Sciences 77 (2005) 85–95 91

Repetitive high intensity electrical stimulation produced a progressive increase in the number of

spikes (wind-up) in all the units studied. Baseline responses for electrical stimulation were similar as

the comparison of the control wind-up curves (previous to the administration of any drug) between

the different experimental groups did not show any significant difference. Control curves were,

therefore, pooled together for a more systematic analysis of drug effects. The administration of

cumulative doses of fentanyl dose-dependently reduced the level of wind-up, with complete inhibition

observed at a dose of 32 Ag/kg in all experiments (Fig. 3). Full recovery from the antinociceptive

effect was observed 15 min after the administration of the highest dose of fentanyl when studied alone

or in the presence of NCX-701 in monoarthritic animals (data not shown). The only difference

observed was a delay in the recovery from the effect of fentanyl in the presence of NCX-701 in

animals without inflammation. In this case, only a partial recovery of 53 F 22% of control response

(P b 0.01) was observed 15 min later, whereas full recovery was seen 30 min after the injection of

fentanyl (Fig. 3).

Behavioral experiments

The electrophysiological experiments showed that the combined administration of subanalgesic

doses of NCX-701 enhanced the analgesic effectiveness of fentanyl and the duration of its effect. The

Fig. 3. Effect of fentanyl on single motor unit wind-up when injected alone or in the presence of NCX-701 (Fent N) in normal

animals. Complete inhibition of wind-up was observed with the dose of 32 Ag/kg of fentanyl in normal animals when given

alone or in the presence of 17 mg/kg of NCX-701. Full recovery of the responses was observed 15 min (R15) after the

administration of fentanyl alone. The recovery of the effect when fentanyl was injected in the presence of NCX-701 was only

partial at that time, and full recovery was only observed 30 min after the administration. Statistical comparison between

responses observed with each dose and the control response was made using the one-way analysis of variance, ANOVA, with

the post-hoc Dunnett test (**P b 0.01).

G. Gaitan et al. / Life Sciences 77 (2005) 85–9592

enhancement of the effect was not reversed by naloxone, indicating that the action is not a direct

consequence of the activation of opioid receptors and, therefore, other opiate actions, such as acute

tolerance might be reduced or even abolished. We therefore hypothesized that the presence of NCX-

701 might reduce the opiate-derived acute tolerance. However, although full recovery from the effects

of fentanyl was never observed in the presence of NCX-701 in animals with hyperalgesia,

nociceptive responses remained at a higher level than that observed in normal animals (see also

Gaitan et al., 2003). We therefore studied our hypothesis in normal animals using a dose of NCX-

701 that showed no analgesic effects in preliminary experiments. The dose of fentanyl was also

submaximal, 0.15 mg/kg per day, corresponding to 6.25 Ag/kg/hour, close to the ID50 observed in

electrophysiological experiments, in responses to mechanical stimulation in normal animals, after i.v.

acute administration. This dose would induce an effect close to 50% presumably avoiding sedation

and motor impairment.

In these experiments, fentanyl was effective in reducing responses to mechanical stimulation both in

animals treated with vehicle or animals treated with NCX-701 for the first two days of the study (Fig. 4).

As hypothesized, the dose of fentanyl used initially induced a reduction of responses to mechanical

stimulation of about 50%. In control animals a reduction of 60 F 15% (P b 0.01) and 30 F 7% (P b

0.05) of responses was observed during days one and two, respectively. The level of the responses,

however, was not significant the third day after the implantation of the pumps and full recovery was

observed during days four and five, indicating that acute tolerance to fentanyl had been developed. In

animals treated with NCX-701, significant reduction of responses was seen during the whole period of

observation. A reduction of responses of 68 F 15% and 76 F 14% was recorded on the fourth and fifth

days, respectively, being in both cases significantly different (P b 0.05) from those responses recorded in

animals treated with vehicle.

Fig. 4. Effect of chronic s.c. infusion of 6.25 Ag/kg/hour fentanyl in behavioral nociceptive reflexes. Significant reduction of

nociceptive responses was observed during the first two days of fentanyl administration in animals treated with vehicle. Acute

tolerance was progressively developed in these animals and no antinociceptive effect was observed thereafter. In animals treated

with 100 Amol/kg (28 mg/kg) NCX-701 p.o., tolerance was not observed and fentanyl remained an effective antinociceptive

agent during the five days of treatment (*P b 0.05, comparison vs. animals treated with vehicle using the non-parametric Mann-

Whitney U-test).

G. Gaitan et al. / Life Sciences 77 (2005) 85–95 93

Discussion

The results observed in the present experiments show that the analgesic efficacy of fentanyl is

significantly increased when given in the presence of subeffective doses of the NSAID NCX-701, both

in models of acute nociception and arthritis. Enhancement of the antinociceptive efficacy of opiates has

been shown previously, especially with morphine, administered in conjunction of effective doses of

various NSAIDs. These experiments did not clarify if the unwanted side-effects induced by both the

opiates and the NSAIDs are also enhanced. In any case, long term use of this combined therapy is likely

to exacerbate the side-effects as well. It was therefore important to assess if the combination of very low

doses of an NSAID would increase the antinociceptive effectiveness of an opiate. The doses used did not

affect nociceptive responses and, presumably, have very little chance of inducing secondary effects, even

when administered in long term therapy. It is hard to imagine that classical NSAIDs would interact with

opioids and enhance their activity this way. We have previously observed, however, that some of the

recently developed NSAIDs, which cross the blood brain barrier efficiently and have a strong effect on

the central nervous system, are very efficacious antinociceptive agents. In terms of potency and efficacy,

they can be many-fold stronger than classic NSAIDs while having reduced unwanted side-effects. Some

of these new compounds can enhance the effectiveness of opiates at very low doses, so low that no effect

is observed in nociceptive responses to noxious mechanical and electrical stimulation. In fact, we have

previously observed that the potency of fentanyl in acute nociception is significantly enhanced by very

low doses of dexketoprofen trometamol (Gaitan and Herrero, 2002). These observations, however, were

made in normal animals, in situations of acute nociception, whereas combined drug treatment of humans

is more likely to be used in situations of spinal sensitization. In the present study we have compared the

enhancement of the antinociceptive potency of fentanyl by NCX-701 in the normal situation and in

sensitization due to carrageenan-induced inflammation. The enhancement observed in monoarthritic

animals was similar to that observed in normal animals, an increase in potency of more than two-fold,

and supports a possible use of this combination of drugs in the treatment of pain in situations of

sensitization. As in previous studies (Gaitan and Herrero, 2002; Gaitan et al., 2003), the recovery from

the antinociceptive effect of fentanyl was only partial in both experimental situations. Whereas full

recovery from the effect was always observed between 15 to 20 minutes after the administration of

fentanyl alone, the presence of low doses of NCX-701 delayed this recovery significantly and only a

very slight recovery was observed 45 minutes after the administration of the opiate.

From the present experiments it is difficult to determine what mechanism of action is involved in the

enhancement of the potency and duration of fentanyl. However, it is possible to infer that the action does

not take place as a result of a direct action on the opioid receptor, since no reversal was observed with the

opioid-receptor antagonist naloxone at a dose capable of reverse antinociceptive actions of mu and kappa

opioid receptors in similar experiments (Herrero and Headley, 1991). This is supported by previous

studies in which a potentiation of morphine antinociception by the NSAID metamizol was not reversed

by naloxone (Taylor et al., 1998). In addition, a2-adrenoceptors, which have been shown to be involved

in opiate analgesia in situations of inflammation (Herrero and Solano, 1999 and references therein) do

not seem to be involved in this phenomenon, since no reversal was observed with the a2-adrenoceptor

antagonist atipamezol. Furthermore, no enhancement of the effect of fentanyl by NCX-701 in the wind-

up phenomenon was observed in animals with inflammation, and only a delay in the return to normal

responses was observed in normal animals. Wind-up is a nociceptive phenomenon mediated by NMDA

and NK1 receptors in spinal cord neurons that can also be depressed by opiates by an unspecific decrease

G. Gaitan et al. / Life Sciences 77 (2005) 85–9594

of the afferent inputs into the spinal cord (see Herrero et al., 2000 for review). The lack of effect on this

phenomenon implies that the enhancement of the effect of fentanyl is not related to an action on spinal

cord nociceptive afferents, and that the presence of a small amount of NCX-701 does not improve the

ability of fentanyl to reduce nociceptive transmission to the spinal cord.

A development of tolerance to constant perfusion of s.c. fentanyl was clear in behavioral experiments.

In animals treated with vehicle, a clear antinociceptive effect was observed within 48 h after the initiation

of the treatment with fentanyl but a progressive decrease in effect was recorded over the following days.

In animals treated with NCX-701, however, the antinociception was seen during the whole study and

was even increased in the last days of treatment. This indicates that the presence of NCX-701 not only

increases its antinociceptive effectiveness, but also inhibits the development of acute tolerance to

fentanyl. . In the case of fentanyl, the dose applied was several times smaller than that used in previous

studies (Celerier et al., 1999; Carter et al., 2000; Thornton et al., 2000). For NCX-701, the dose was less

than half of that observed to be effective in preliminary experiments, and even lower than the minimum

effective dose observed after i.v. administration in electrophysiological experiments (Romero-Sandoval

et al., 2002). This suggests that the interaction occurred within the subanalgesic dose range, supporting

the results observed in electrophysiological experiments.

In conclusion, our results show that subanalgesic doses of the NSAID NCX-701 strongly potentiates

fentanyl antinociception both in normal animals and animals with hyperalgesia. Also, the development

of acute tolerance to fentanyl is prevented by the combined administration with NCX-701. Further

experiments are required to elucidate the mechanisms underlying these effects, although it seems clear

that they do not involve a direct opioid receptor activation, nor an action over a2-adrenoceptors, nor it

seems mediated by a direct modulation of NMDA and NK1 receptors.

Acknowledgements

This work has been supported by NicOx S.A. and by a grant from the Comunidad de Madrid (grant

GR/SAL/0815/2004). Gema Gaitan is a fellow of the Spanish Ministry of Science and Technology. We

are grateful to Dr. Erik Wade for a critical review of the manuscript and for his help with the English

revision.

References

Bosek, V., Miguel, R., 1994. Comparison of morphine and ketorolac for intravenous patient-controlled analgesia in

postoperative cancer patients. Clin. J. Pain 10, 314–318.

Burns, J.W., Aitken, H.A., Bullingham, R.E., McArdle, C.S., Kenny, G.N., 1991. Double-blind comparison of the morphine

sparing effect of continuous and intermittent i.m. administration of ketorolac. Br. J. Anaesth. 67, 235–238.

Carter, B.L., Tiffany, S.T., Conklin, C.A., 2000. Associative and non-associative fentanyl tolerance in the rat: evaluation of

cross tolerance with mu-and kappa-specific opioids. Psychopharmacology 148, 384–392.

Celerier, E., Laulin, J., Larcher, A., Le Moal, M., Simonnet, G., 1999. Evidence for opiate-activated NMDA processes masking

opiate analgesia in rats. Brain Res. 847 (1), 18–25.

Gaitan, G., Herrero, J.F., 2002. Subeffective doses of dexketoprofen trometamol enhance the potency and duration of fentanyl

antinociception. Br. J. Pharmacol. 135, 393–398.

G. Gaitan et al. / Life Sciences 77 (2005) 85–95 95

Gaitan, G., Romero-Sandoval, E.A., Herrero, J.F., 2003. The antinociceptive activity of Nitroparacetamol (NCX-701) is located

in the spinal cord and increases the potency and duration of the mu-opiate fentanyl antinociception. 33rd Annual Meeting,

Society for Neuroscience, New Orleans.

Gillies, G.W., Kenny, G.N., Bullingham, R.E., McArdle, C.S., 1987. The morphine sparing effect of ketorolac tromethamine. A

study of a new, parenteral non-steroidal anti-inflammatory agent after abdominal surgery. Anesthesia 42, 727–731.

Herrero, J.F., Headley, P.M., 1991. The effects of sham and full spinalization on the systemic potency of A- and k-opioids on

spinal nociceptive reflexes in rats. Br. J. Pharmacol. 104, 166–170.

Herrero, J.F., Solano, R.E., 1999. The antinociceptive effect of the A-opioid fentanyl is reduced in the presence of the a(2)-

adrenergic antagonist idazoxan in inflammation. Brain Res. 840, 106–114.

Herrero, J.F., Laird, J.M.A., Lopez-Garcia, J.A., 2000. Wind-up of spinal cord neurones and pain sensation: much ado about

something? Prog. Neurobiol. 61, 169–203.

Laitinen, J., Nuutinen, L., 1992. Intravenous diclofenac coupled with PCA fentanyl for pain relief after total hip replacement.

Anesthesiology 76, 194–198.

Lashbrook, J.M., Ossipov, M.H., Hunter, J.C., Raffa, R.B., Tallarida, R.J., Porreca, F., 1999. Synergistic antiallodynic effects of

spinal morphine with ketorolac and selective COX1- and COX2-inhibitors in nerve-injured rats. Pain 82, 65–72.

Malmberg, B., Yaksh, T.L., 1993. Pharmacology of the spinal action of ketorolac morphine ST-91 U50488H and L-PIA on the

formalin test and an isobolographic analysis of the NSAID interaction. Anesthesiology 79, 70–281.

Maves, T.J., Pechman, P.S., Meller, S.T., Gebhart, G.F., 1994. Ketorolac potentiates morphine antinociception during visceral

nociception in the rat. Anesthesiology 80, 1094–1101.

Romero-Sandoval, E.A., Mazario, J., Howat, D., Herrero, J.F., 2002. NCX-701 (nitroparacetamol) is an effective

antinociceptive agent in rat withdrawal reflexes and wind-up. Br J Pharmacol 135, 1556–1562.

Romero-Sandoval, E.A., Del Soldato, P., Herrero, J.F., 2003. The effects of sham and full spinalization on the antinociceptive

effects of NCX-701 (nitroparacetamol) in monoarthritic rats. Neuropharmacology 45, 412–419.

Solano, R.E., Herrero, J.F., 1997. Cutaneous responsiveness of rat single motor units activated by natural stimulation.

J. Neurosci. Methods 73, 135–140.

Taylor, J., Mellstrom, B., Fernaud, I., Naranjo, J.R., 1998. Metamizol potentiates morphine effects on visceral pain and evoked

c-Fos immunoreactivity in spinal cord. Eur J Pharmacol 351 (1), 39–47.

Thornton, S.R., Lohmann, A.B., Nicholson, R.A., Smith, F.L., 2000. Fentanyl self-administration in juvenile rats that were

tolerant and dependent to fentanyl as infants. Pharmacol Biochem Behav 65 (3), 563–570.