antinociception induced by intravenous dipyrone (metamizol) upon dorsal horn neurons: involvement of...

7
Research report Antinociception induced by intravenous dipyrone (metamizol) upon dorsal horn neurons: Involvement of endogenous opioids at the periaqueductal gray matter, the nucleus raphe magnus, and the spinal cord in rats Enrique Vazquez, Norma Hernandez, William Escobar, Horacio Vanegas * Instituto Venezolano de Investigaciones Cientificas (IVIC), Centro Biofisica/Bioquimica, 8424 NW 56th Street, Suite CCS 00202, Miami, FL 33166, USA Accepted 28 April 2005 Available online 25 May 2005 Abstract Microinjection of dipyrone (metamizol) into the periaqueductal gray matter (PAG) in rats causes antinociception. This is mediated by endogenous opioidergic circuits located in the PAG itself, in the nucleus raphe magnus and adjacent structures, and in the spinal cord. The clinical relevance of these findings, however, is unclear. Therefore, in the present study, dipyrone was administered intravenously, and the involvement of endogenous opioidergic circuits in the so-induced antinociception was investigated. In rats, responses of dorsal spinal wide- dynamic range neurons to mechanical noxious stimulation of a hindpaw were strongly inhibited by intravenous dipyrone (200 mg/kg). This effect was abolished by microinjection of naloxone (0.5 Ag/0.5 Al) into the ventrolateral and lateral PAG or into the nucleus raphe magnus or by direct application of naloxone (50 Ag/50 Al) onto the spinal cord surface above the recorded neuron. These results show that dipyrone, a non-opioid analgesic with widespread use in Europe and Latin America, when administered in a clinically relevant fashion causes antinociception by activating endogenous opioidergic circuits along the descending pain control system. D 2005 Elsevier B.V. All rights reserved. Theme: Sensory systems Topic: Pain modulation: pharmacology Keywords: Descending pain control; Endogenous opioids; Nucleus raphe magnus; NSAIDs; Periaqueductal gray; Spinal nociceptive neurons 1. Introduction Non-opioid analgesics exert their effects by acting upon peripheral tissues as well as upon central nervous system structures. Central targets of non-opioid analgesics include the periaqueductal gray matter (PAG) [3,27,28,31], the rostral ventromedial medulla (RVM), i.e., the nucleus raphe magnus (NRM) and adjacent structures [16], and the spinal cord (see [30] for review). Dipyrone (metamizol) is an antipyretic and non-opioid analgesic with widespread clinical use in Europe and Latin America [4,14,15]. This pyrazolone derivative readily forms neutral solutions in water and has inhibitory activity upon cyclooxygenases 1, 2, and 3 [2,5,7,14,22]. The biologically active metabolites of dipyrone quickly enter the cerebrospinal fluid and reach a concentration in brain tissue of about 50% plasma concen- tration [9]. We have shown [27] that dipyrone micro- injection into PAG induces changes in the activity of spinally projecting neurons located in the RVM, specifically, the so-called on- and off-cells (see [10] for review). These changes are in the expected direction for the proposed role of on- and off-cells as pain-modulating intermediaries between the PAG and spinal nociceptive circuits [27]. PAG-microinjected dipyrone consequently induces inhib- ition of spinal neuronal responses to peripheral noxious stimulation [31–33] and inhibition of the tail flick reflex [3,27]. The antinociceptive effects of PAG-microinjected dipyrone thus mimic the effects of PAG-microinjected opioids [6,11,35] and are abolished by naloxone admin- 0006-8993/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2005.04.083 * Corresponding author. Fax: +58 212 504 1093. E-mail address: [email protected] (H. Vanegas). Brain Research 1048 (2005) 211 – 217 www.elsevier.com/locate/brainres

Upload: enrique-vazquez

Post on 04-Sep-2016

215 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Antinociception induced by intravenous dipyrone (metamizol) upon dorsal horn neurons: Involvement of endogenous opioids at the periaqueductal gray matter, the nucleus raphe magnus,

www.elsevier.com/locate/brainres

Brain Research 1048

Research report

Antinociception induced by intravenous dipyrone (metamizol) upon dorsal

horn neurons: Involvement of endogenous opioids at the periaqueductal

gray matter, the nucleus raphe magnus, and the spinal cord in rats

Enrique Vazquez, Norma Hernandez, William Escobar, Horacio Vanegas*

Instituto Venezolano de Investigaciones Cientificas (IVIC), Centro Biofisica/Bioquimica, 8424 NW 56th Street, Suite CCS 00202, Miami, FL 33166, USA

Accepted 28 April 2005

Available online 25 May 2005

Abstract

Microinjection of dipyrone (metamizol) into the periaqueductal gray matter (PAG) in rats causes antinociception. This is mediated by

endogenous opioidergic circuits located in the PAG itself, in the nucleus raphe magnus and adjacent structures, and in the spinal cord. The

clinical relevance of these findings, however, is unclear. Therefore, in the present study, dipyrone was administered intravenously, and the

involvement of endogenous opioidergic circuits in the so-induced antinociception was investigated. In rats, responses of dorsal spinal wide-

dynamic range neurons to mechanical noxious stimulation of a hindpaw were strongly inhibited by intravenous dipyrone (200 mg/kg). This

effect was abolished by microinjection of naloxone (0.5 Ag/0.5 Al) into the ventrolateral and lateral PAG or into the nucleus raphe magnus or

by direct application of naloxone (50 Ag/50 Al) onto the spinal cord surface above the recorded neuron. These results show that dipyrone, a

non-opioid analgesic with widespread use in Europe and Latin America, when administered in a clinically relevant fashion causes

antinociception by activating endogenous opioidergic circuits along the descending pain control system.

D 2005 Elsevier B.V. All rights reserved.

Theme: Sensory systems

Topic: Pain modulation: pharmacology

Keywords: Descending pain control; Endogenous opioids; Nucleus raphe magnus; NSAIDs; Periaqueductal gray; Spinal nociceptive neurons

1. Introduction

Non-opioid analgesics exert their effects by acting upon

peripheral tissues as well as upon central nervous system

structures. Central targets of non-opioid analgesics include

the periaqueductal gray matter (PAG) [3,27,28,31], the

rostral ventromedial medulla (RVM), i.e., the nucleus raphe

magnus (NRM) and adjacent structures [16], and the spinal

cord (see [30] for review). Dipyrone (metamizol) is an

antipyretic and non-opioid analgesic with widespread

clinical use in Europe and Latin America [4,14,15]. This

pyrazolone derivative readily forms neutral solutions in

water and has inhibitory activity upon cyclooxygenases 1, 2,

0006-8993/$ - see front matter D 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.brainres.2005.04.083

* Corresponding author. Fax: +58 212 504 1093.

E-mail address: [email protected] (H. Vanegas).

and 3 [2,5,7,14,22]. The biologically active metabolites of

dipyrone quickly enter the cerebrospinal fluid and reach a

concentration in brain tissue of about 50% plasma concen-

tration [9]. We have shown [27] that dipyrone micro-

injection into PAG induces changes in the activity of

spinally projecting neurons located in the RVM, specifically,

the so-called on- and off-cells (see [10] for review). These

changes are in the expected direction for the proposed role

of on- and off-cells as pain-modulating intermediaries

between the PAG and spinal nociceptive circuits [27].

PAG-microinjected dipyrone consequently induces inhib-

ition of spinal neuronal responses to peripheral noxious

stimulation [31–33] and inhibition of the tail flick reflex

[3,27]. The antinociceptive effects of PAG-microinjected

dipyrone thus mimic the effects of PAG-microinjected

opioids [6,11,35] and are abolished by naloxone admin-

(2005) 211 – 217

Page 2: Antinociception induced by intravenous dipyrone (metamizol) upon dorsal horn neurons: Involvement of endogenous opioids at the periaqueductal gray matter, the nucleus raphe magnus,

E. Vazquez et al. / Brain Research 1048 (2005) 211–217212

istration to the same PAG site [29], to the RVM [32], or to

the spinal cord [13]. Therefore, the antinociceptive effect of

PAG-microinjected dipyrone is mediated by endogenous

opioids at the PAG, the RVM, and the spinal cord, i.e., along

the descending pain control system. The clinical relevance

of these findings remains unknown, however, because

dipyrone is not normally administered by microinjection

into the PAG. In the present study, dipyrone was therefore

administered intravenously, as often done to induce analge-

sia in humans [4,14,15], and naloxone was subsequently

applied at various levels of the descending pain control

system in order to investigate whether the antinociceptive

effect of systemically administered dipyrone is also medi-

ated by endogenous opioids acting at such levels. Some of

the results have been presented in preliminary form [12].

2. Methods

2.1. General

Recommendations of the Society for Neuroscience and

the International Association for the Study of Pain regarding

experiments in animals were followed throughout. Male

Sprague–Dawley rats (260–320 g), bred at the Instituto

Venezolano de Investigaciones Cientificas, were deeply

anesthetized with thiopental (60 mg/kg i.p. initial dose and

3.5–5 mg/kg/h i.v. continuously for maintenance). After

insertion of a tracheal cannula, a carotid catheter, and a

jugular catheter, a lumbar laminectomy was performed.

Carotid pressure remained within normal range, and rectal

temperature was kept around 37 -C. The animals were

neither paralyzed nor artificially ventilated.

2.2. Preparation for naloxone administration

When naloxone was to be microinjected, a stainless-

steel guide cannula (22-gauge) was stereotaxically [23]

driven through a small craniotomy to reach 2 mm above

the intended target in the PAG, the NRM, or their

vicinities. When naloxone was to be applied to the spinal

cord, the dura mater over the lumbar enlargement was

opened, and a thin-walled plastic ring was sealed with

grease onto the dorsal spinal cord surface over the

intended recording site. This ring was filled with 50 Alnormal saline, and 2% agar was poured around it to cover

the surgical area as far as the stretched skin flaps. When

naloxone was not to be applied to the spinal cord, no ring

was installed, and the whole area was covered with agar

except for a small saline-filled window above the spinal

cord recording site (see below).

2.3. Recording and stimulation

Tungsten microelectrodes were introduced into the

spinal cord through the saline solution in the plastic ring

or in the agar window in order to record action potentials

from dorsal horn neurons with receptive fields in the

ipsilateral hindpaw. The neurons chosen for study had no or

negligible spontaneous activity and were differentially

excited by the dorsoventral application to the hindpaw of

a weak clamp (innocuous when applied to the experimen-

ter’s fifth finger) or a strong clamp (noxious to the

experimenter). Both the innocuous and the noxious clamp

were spring-loaded so that the pressure applied in each case

was maintained during the 10 s stimulation period (see

below). The neurons chosen for study were also excited by

noxious heat and pinch applied to their receptive field skin

and can thus be classed as wide-dynamic range (WDR) or

multireceptive neurons. When two or (seldom) three

neurons were simultaneously recorded, their spikes were

discriminated by means of the BrainWave\ software. The

number of neurons reported may thus be larger than the

number of animals.

2.4. Experimental protocol

The following recording protocol was carried out every

5 min: (a) 1 min on-going activity, (b) 10 s application of

the innocuous clamp, (c) 1 min on-going activity, and (d)

10 s application of the noxious clamp. After three or more

cycles with stable responses were obtained (baseline),

dipyrone (Novalcina\, formerly Hoechst–Marion–Rou-

sell) was injected (200 mg in 0.8 ml saline per kilogram of

body weight) through the jugular catheter in 10 s. Sixteen

minutes after the dipyrone injection, naloxone was

administered. For administration to the PAG or the

NRM, a stainless-steel microinjection cannula (29-gauge),

connected by polyethylene tubing to a 1 Al Hamilton

syringe, was introduced through the guide cannula to reach

the desired target, and 0.5 Ag naloxone in 0.5 Al saline was

microinjected in 10 s. For application to the spinal cord,

the saline in the ring was replaced with 50 Ag naloxone in

50 Al saline. Only one protocol was performed per animal.

We have previously shown that 200 mg/kg and 400 mg/

kg i.v. dipyrone dose-dependently inhibit the tail flick reflex

in rats [27]. According to the equation of Pong et al. [24], an

oral dose of 200 mg/kg of dipyrone in mice would be

equivalent to the recommended analgesic oral dose of 500

mg for adult humans (see [26], p. 538, and [15], p. 13). If the

Pong equation also holds for rats, the i.v. dose of dipyrone

chosen for the present study (200 mg/kg) is equivalent to the

recommended single oral or i.v. dose for humans. On the

other hand, the doses of naloxone used in the present study

have been effective in the RVM (0.5 Ag/0.5 Al) [32] and on

the spinal cord (50 Ag/50 Al) [13] for blocking the anti-

nociceptive action of PAG-microinjected dipyrone.

At the end of the experiment, the microinjection site was

marked by microinjecting 0.5 Al cresyl violet, and the

microelectrode recording site was marked electrolytically

(20 AA, 20 s). The animal was killed with an overdose of

thiopental, and the brain and lumbar spinal cord were fixed

Page 3: Antinociception induced by intravenous dipyrone (metamizol) upon dorsal horn neurons: Involvement of endogenous opioids at the periaqueductal gray matter, the nucleus raphe magnus,

E. Vazquez et al. / Brain Research 1048 (2005) 211–217 213

by immersion in 10% formalin. The marked sites were

identified in wet and unstained 50 Am transverse sections of

the brainstem [23] and of the spinal cord [19].

2.5. Data processing

Neuronal responses were expressed as number of action

potentials within the 10 s stimulation period. Population

values were expressed as mean T SEM. Baseline values of

different rat groups were tested for homogeneity of variances

and compared by means of one-way ANOVA followed by

post-hocBonferroni’s t test. The time-course of the resultswas

evaluated by one-way ANOVA for repeated measures, and

point differenceswere comparedbyDunnett’s t test. Statistical

significance was set at P < 0.05. The SPSSi Statistical

Software Release 10.0 was used for statistical analysis.

3. Results

Neuronal responses to the innocuous clamp were not

modified by i.v. injection of either dipyrone or saline nor by

subsequent administration of naloxone to PAG, NRM, or

spinal cord (Figs. 1–3). In contrast, i.v. administration of

dipyrone induced an inhibition of spinal neuronal responses

to the noxious clamp (Figs. 1–3), and naloxone adminis-

tration to the PAG, the NRM, or the spinal cord abolished this

inhibition (Figs. 1–3). This is presented in detail forthwith.

3.1. Microinjection of naloxone into PAG or outside PAG

Baseline responses to the innocuous or noxious clamp

were not significantly different among the 3 rat groups

presented in this section (Fig. 1).

Fig. 1. Reversibility of the antinociceptive effect of systemic dipyrone by PAG nal

squares) or saline (SAL, rhomboids) upon responses (mean and SEM) of dorsal sp

the ipsilateral hindpaw for 10 s. After minute 16 post-DIP or post-SAL, naloxone

NAL, rhomboids and filled squares, see Fig. 4 for histology) or outside the ventrol

graph, baseline (BL) values are not statistically different among the rat groups (P =

way ANOVA). *Statistically significant (P < 0.05) difference to the respective B

In a group of 6 rats, the responses of dorsal spinal

neurons to the noxious clamp were inhibited by i.v.

administration of dipyrone (Fig. 1, filled squares) to 20%

of baseline in 16 min (10 neurons, F = 6.494, P < 0.0001

vs. baseline). Naloxone microinjection into the ventrolateral

and lateral PAG (Fig. 4, filled squares) abolished the

dipyrone-induced antinociception so that thereafter the

responses to the noxious clamp were not statistically

different from baseline (Fig. 1, filled squares). The neurons

in this group were approximately located in spinal laminae

II–III and V–VI (Fig. 4C, filled squares). Also in another

group (5 rats), the responses of dorsal spinal neurons to the

noxious clamp were inhibited by i.v. administration of

dipyrone (Fig. 1, empty squares) to 40% of baseline in 16

min (7 neurons, F = 8.526, P < 0.0001 vs. baseline).

Naloxone microinjection outside the ventrolateral PAG

(Fig. 4A, empty squares) did not modify the dipyrone-

induced depression of dorsal spinal neuronal responses to

noxious stimulation (Fig. 1, empty squares). This suggests

that the effect of naloxone microinjection into the PAG was

not due to an anatomically widespread action. The neurons

in this group were approximately located in spinal laminae

II–III and V–VI (Fig. 4D, empty squares). In the third

group (3 rats), i.v. injection of saline alone (0.8 ml/kg) did

not affect the responses of 5 dorsal spinal neurons to

noxious stimulation (Fig. 1, rhomboids) and, also, naloxone

microinjection into the ventrolateral PAG in these cases

(Fig. 4A, rhomboids) did not produce any significant

change in the responses. This shows that neither the i.v.

injection nor the PAG naloxone microinjection had by

themselves any effect on neuronal responses to noxious

stimulation. The neurons in this group were approximately

located in spinal laminae III and V–VI (Fig. 4D,

rhomboids).

oxone. Effect of i.v. injection of dipyrone (DIP, 200 mg/kg, filled and empty

inal neurons to an innocuous (INN) and a noxious (NOX) clamp applied to

was microinjected (0.5 Ag/0.5 Al) into the ventrolateral or lateral PAG (PAG

ateral PAG (off PAG NAL, empty squares, see Fig. 4 for histology). In each

0.747 for innocuous stimulation and P = 0.554 for noxious stimulation, one-

L (one-way ANOVA).

Page 4: Antinociception induced by intravenous dipyrone (metamizol) upon dorsal horn neurons: Involvement of endogenous opioids at the periaqueductal gray matter, the nucleus raphe magnus,

Fig. 2. Reversibility of the antinociceptive effect of systemic dipyrone by NRM naloxone. Effect of i.v. injection of dipyrone (DIP, 200 mg/kg, filled and empty

circles) or saline (SAL, crosses) upon responses (mean and SEM) of dorsal spinal neurons to an innocuous (INN) and a noxious (NOX) clamp applied to the

ipsilateral hindpaw for 10 s. After minute 16 post-DIP or post-SAL, naloxone was microinjected (0.5 Ag/0.5 Al) into the NRM (NRM NAL, filled circles and

crosses, see Fig. 4 for histology) or outside the NRM (off NRM NAL, empty circles, see Fig. 4 for histology). In each graph, baseline (BL) values are not

statistically different among the rat groups (P = 0.961 for innocuous stimulation and P = 0.826 for noxious stimulation, one-way ANOVA). *Statistically

significant (P < 0.05) difference to the respective BL (one-way ANOVA).

E. Vazquez et al. / Brain Research 1048 (2005) 211–217214

3.2. Microinjection of naloxone into NRM or outside NRM

Baseline responses to the innocuous or noxious clamp

were not significantly different among the 3 rat groups

presented in this section (Fig. 2).

In a group of 6 rats, the responses of dorsal spinal

neurons to the noxious clamp were inhibited by i.v.

administration of dipyrone (Fig. 2, filled circles) to 26%

of baseline in 16 min (7 neurons, F = 2.068, P = 0.015 vs.

baseline). Naloxone microinjection into NRM (Fig. 4B,

filled circles) abolished the dipyrone-induced antinocicep-

tion so that thereafter the responses to the noxious clamp

were not statistically different from baseline (Fig. 2, filled

Fig. 3. Reversibility of the antinociceptive effect of systemic dipyrone by spinal na

or saline (SAL, empty triangles) upon responses (mean and SEM) of dorsal spinal

ipsilateral hindpaw for 10 s. After minute 16 post-DIP or post-SAL, naloxone (50

triangles). In each graph, baseline (BL) values are not statistically different amon

noxious stimulation, one-way ANOVA). *Statistically significant (P < 0.05) diffe

circles). The neurons in this group were approximately

located in spinal laminae II–VI (Fig. 4C, filled circles).

Also in another group (6 rats), the responses of dorsal

spinal neurons to the noxious clamp were inhibited by i.v.

administration of dipyrone (Fig. 2, empty circles) to 42% of

baseline in 16 min (9 neurons, F = 3.129, P = 0.02 vs.

baseline). Naloxone microinjection outside the NRM (Fig.

4B, empty circles) did not modify the dipyrone-induced

depression of dorsal spinal neuronal responses to noxious

stimulation. This suggests that the effect of naloxone

microinjection into NRM was not due to an anatomically

widespread action. The neurons in this group were

approximately located in spinal laminae II–VI (Fig. 4D,

loxone. Effect of i.v. injection of dipyrone (DIP, 200 mg/kg, filled triangles)

neurons to an innocuous (INN) and a noxious (NOX) clamp applied to the

Ag in 50 Al) was pipetted into the spinal ring (spinal NAL, filled and empty

g the rat groups (P = 0.405 for innocuous stimulation and P = 0.830 for

rence to the respective BL (one-way ANOVA).

Page 5: Antinociception induced by intravenous dipyrone (metamizol) upon dorsal horn neurons: Involvement of endogenous opioids at the periaqueductal gray matter, the nucleus raphe magnus,

Fig. 4. Approximate histological location of microinjection sites and recorded neurons. Equal symbols belong to the same experiment type. (A) Microinjection

sites in and outside the ventrolateral and lateral PAG, depicted on a simplified AP 1.6 section [23]. Filled and empty squares: naloxone microinjections.

Rhomboids: saline microinjections. (B) Microinjection sites in the NRM, depicted on a simplified AP � 2.6 section [23]. Filled and empty circles: naloxone

microinjections. Crosses: saline microinjections. (C and D) Neuronal recording sites in the spinal cord, depicted on an L4 section [19]. Triangles represent

neurons in experiments with spinal application of naloxone after i.v. dipyrone (filled triangles) or i.v. saline (empty triangles). All other symbols correspond to

microinjection experiments and match the symbols in panels (A) and (B). CS, colliculus superior. DDT, decussatio dorsalis tegmenti. FLM, fasciculus

longitudinalis medialis. FSV, fasciculus spinalis trigemini. I –VI, dorsal spinal Rexed laminae. LM, lemniscus medialis. NSV, nucleus spinalis trigemini. NGM,

nucleus geniculatus medialis. NIP, nucleus interpeduncularis. NR, nucleus ruber. NRM, nucleus raphe magnus. PAG, periaqueductal gray. SN, substantia nigra.

V, trigeminal nerve. VII, facial nucleus (ventral) and genu (dorsal).

E. Vazquez et al. / Brain Research 1048 (2005) 211–217 215

empty circles). In the third group (3 rats), i.v. injection of

saline alone (0.8 ml/kg) did not affect the responses of 4

dorsal spinal neurons to noxious stimulation (Fig. 2,

crosses) and, also, naloxone microinjection into NRM in

these cases (Fig. 4B, crosses) did not produce any change

in the responses. This shows that neither the i.v. injection

nor the NRM naloxone microinjection had by themselves

any effect on neuronal responses to noxious stimulation.

The neurons in this group were approximately located in

spinal laminae III and V–VI (Fig. 4D, crosses).

3.3. Spinal cord application of naloxone

Baseline responses to the innocuous or noxious clamp

were not significantly different among the 2 rat groups

presented in this section (Fig. 3).

In a group of 4 rats, the responses of dorsal spinal

neurons to the noxious clamp were inhibited by i.v.

administration of dipyrone (Fig. 3, filled triangles) to 32%

of baseline in 16 min (7 neurons, F = 2.166, P = 0.02 vs.

baseline). When the saline solution in the spinal plastic ring

was replaced with the naloxone solution, the dipyrone-

induced antinociception was abolished so that the responses

to the noxious clamp in the next three stimulation cycles

were not statistically different from baseline (Fig. 3,

triangles). At 36 min post-dipyrone, the naloxone effect

had ceased, and the antinociception became again evident

(F = 4.838, P = 0.04). The neurons in this group were

approximately located in laminae II–V of the dorsal horn

(Fig. 4C, filled triangles). In another group (2 rats), i.v.

injection of saline alone (0.8 ml/kg) did not affect the

responses of 4 dorsal spinal neurons to noxious stimulation

(Fig. 3, empty triangles); naloxone application to the spinal

cord in these cases did not produce any change in the

responses. This shows that neither the i.v. injection nor the

naloxone application had by themselves any effect on

neuronal responses to noxious stimulation. The neurons in

this group were approximately located in laminae II and IV

of the dorsal horn (Fig. 4D, empty triangles).

4. Discussion

The present results show that i.v. injection of dipyrone

inhibits mechanical nociception in spinal dorsal horn WDR

neurons. This extends previous studies in which systemic

dipyrone induced appropriate changes in the activity of

RVM pain-modulating neurons and inhibited the tail flick

reflex [27], depressed on-going activity and mechanical

nociception in spinal neurons [21], and inhibited activity in

spinal ascending axons evoked by peripheral C-fiber

stimulation [3]. These findings reveal at least some of the

mechanisms whereby dipyrone alleviates clinical pain in

humans [4,14,15]. More interestingly, the present results

show that the antinociceptive effect of systemically admin-

istered dipyrone can be attenuated by naloxone application

to various levels of the descending pain control system,

namely, the ventrolateral and lateral PAG, the NRM, and the

spinal cord. Similar results were obtained in a non-clinical

model, i.e., direct administration of dipyrone to the PAG

followed by administration of naloxone to the same PAG

site [29], the RVM [32], or the spinal cord [13]. Taken

together, these studies show that dipyrone, although

considered to be a non-opioid analgesic, engages endoge-

nous opioids in the descending pain control system for its

analgesic action.

The mechanisms whereby dipyrone may interact with

endogenous opioids are slowly being elucidated. In the first

Page 6: Antinociception induced by intravenous dipyrone (metamizol) upon dorsal horn neurons: Involvement of endogenous opioids at the periaqueductal gray matter, the nucleus raphe magnus,

E. Vazquez et al. / Brain Research 1048 (2005) 211–217216

place, the following observations suggest how opioids act

at least in the PAG. Since GABA antagonists in the PAG

have a similar effect as morphine [20], it has been proposed

that opioids act in the PAG by inhibiting GABAergic

neurons and thus disinhibiting their target neurons,

supposedly the ‘‘output’’ neurons of the PAG. These

neurons would in turn cause antinociception by acting

upon downstream structures like the NRM. There is

consistent evidence [8,18,34] that opioids in the PAG

indeed reduce synaptic GABA release by stimulating the

formation of arachidonic acid metabolites along the 12-

lipoxygenase pathway. It has thus been postulated [34] that

non-opioid analgesics in the PAG synergize with endoge-

nous opioids by blocking, as usual, the cyclooxygenases,

and thereby making more arachidonic acid available to the

12-lipoxygenase pathway and thus further decreasing

GABA release. Since dipyrone has been shown to inhibit

cyclooxygenases 1, 2, and 3 [2,5], an interaction with

endogenous opioids at the PAG might explain the present

and previous results. Once the PAG has been disinhibited

by opioids, by GABA antagonists, or by dipyrone,

opioidergic systems are triggered downstream; indeed,

opioid antagonists block the PAG-induced descending

inhibition of nociception when they are applied at the

RVM [17,25,32] or at the spinal cord ([1,13] and present

study). The analgesic effect of systemically administered

dipyrone thus probably derives to a large extent from its

action upon the PAG and the resulting opioidergic

descending inhibition of nociception. In addition, dipyrone

may of course interact with local opioids at the RVM and

the spinal cord.

One intriguing aspect of this study with systemic admin-

istration as well as of previous studies with PAG micro-

injection [13,31,32] is the lack of a dipyrone effect upon

responses to innocuous mechanical stimulation. If the

inhibition triggered by dipyrone were exerted upon the spinal

WDR neurons studied herein, an inhibition of their responses

to excitation of all afferent fiber types should be expected, but

only the responses to excitation of nociceptive afferents were

actually inhibited. This would be the case if systemic

dipyrone were exerting its main effect at the peripheral

terminals of only nociceptive afferents, but the reversal of the

dipyrone effect by naloxone applied to central nervous

system structures speaks in favor of a mainly central action.

This central, descending, and opioid-related action of

dipyrone might instead be exerted upon the spinal terminals

of only nociceptive primary afferents. It is also possible that

the opioid-related effect of dipyrone is exerted upon WDR

spinal neurons but differentially affects EPSPs caused by A-

delta and C (nociceptive) fibers vs. EPSPs caused by A-beta

(non-nociceptive) fibers. These and other possible mecha-

nisms must be explored in the future.

In summary, the present results show that dipyrone, a

non-opioid analgesic with widespread use in Europe and

Latin America, when administered in a clinically relevant

fashion causes antinociception by activating endogenous

opioidergic circuits along the descending pain control

system.

Acknowledgments

Partly supported by grant S1-97000106 of the Venezue-

lan FONACIT. We thank Jenny Nava and Karla Ramirez for

technical help and Dr. V. Tortorici for comments on the

manuscript.

References

[1] D. Budai, H.L. Fields, Endogenous opioid peptides acting at A-opioidreceptors in the dorsal horn contribute to midbrain modulation of

spinal nociceptive neurons, J. Neurophysiol. 79 (1998) 677–687.

[2] C. Campos, R. de Gregorio, R. Garcia-Nieto, F. Gago, P. Ortiz, S.

Alemany, Regulation of cyclooxygenase activity by metamizol, Eur. J.

Pharmacol. 378 (1999) 339–347.

[3] K.-H. Carlsson, J. Helmreich, I. Jurna, Activation of inhibition from

the periaqueductal grey matter mediates central analgesic effect of

metamizol (dipyrone), Pain 27 (1986) 373–390.

[4] O.L. Ceraso, Los Analgesicos antitermicos, Lopez Libreros Editores,

Buenos Aires, 1994.

[5] N.V. Chandrasekharan, H. Dai, K.L.T. Roos, N.K. Evanson, J.

Tomsik, T.S. Elton, D.L. Simmons, COX-3, a cyclooxygenase-1

variant inhibited by acetaminophen and other analgesic/antipyretic

drugs: cloning, structure, and expression, Proc. Natl. Acad. Sci.

U. S. A. 99 (2002) 13926–13931.

[6] Z.-F. Cheng, H.L. Fields, M.M. Heinricher, Morphine microinjected

into the periaqueductal gray has differential effects on 3 classes of

medullary neurons, Brain Res. 375 (1986) 57–65.

[7] O. Christ, H.-M. Kellner, G. Ross, W. Rupp, A. Schwarz, Biopharma-

zeutische und pharmakokinetische Untersuchungen nach Gabe von

Metamizol-14C (NovalginR14C) an Ratte, Hund und Mensch, Arneim.-

Forsch., Drug Res. 23 (1973) 1760–1767.

[8] M.J. Christie, C.W. Vaughan, S.L. Ingram, Opioids, NSAIDs and 5-

lipoxygenase inhibitors act synergistically in brain via arachidonic

acid metabolism, Inflamm. Res. 48 (1999) 1–4.

[9] O. Cohen, E. Zylber-Katz, Y. Caraco, L. Granit, M. Levy,

Cerebrospinal fluid and plasma concentrations of dipyrone metabo-

lites after a single oral dose of dipyrone, Eur. J. Clin. Pharmacol. 54

(1998) 549–553.

[10] H.L. Fields, M.M. Heinricher, P. Mason, Neurotransmitters in

nociceptive modulatory circuits, Annu. Rev. Neurosci. 14 (1991)

219–245.

[11] G.F. Gebhart, S.L. Jones, Effects of morphine given in the brain stem

on the activity of dorsal horn nociceptive neurons, Prog. Brain Res. 77

(1988) 229–243.

[12] N. Hernandez, H. Vanegas, Involvement of supraspinal and spinal

endogenous opioids in the antinociceptive effect of intravenously

injected dipyrone (DIP), Abstr.-Soc. Neurosci. 26 (2000) 661.

[13] N. Hernandez, H. Vanegas, Antinociception induced by PAG-micro-

injected dipyrone (metamizol) in rats: involvement of spinal endog-

enous opioids, Brain Res. 896 (2001) 175–178.

[14] Hoechst, Novalgin (monograph), 1996.

[15] IASP, Management of Acute Pain: A Practical Guide, IASP

Publications, Seattle, 1992.

[16] S.L. Jones, Dipyrone into the nucleus raphe magnus inhibits the rat

nociceptive tail-flick reflex, Eur. J. Pharmacol. 318 (1996) 37–40.

[17] J.M. Kiefel, G.C. Rossi, R.J. Bodnar, Medullary A and y opioid

receptors modulate mesencephalic morphine analgesia in rats, Brain

Res. 624 (1993) 151–161.

Page 7: Antinociception induced by intravenous dipyrone (metamizol) upon dorsal horn neurons: Involvement of endogenous opioids at the periaqueductal gray matter, the nucleus raphe magnus,

E. Vazquez et al. / Brain Research 1048 (2005) 211–217 217

[18] K. Kishimoto, S. Koyama, N. Akaike, Synergistic A-opioid and 5-

HT1A presynaptic inhibition of GABA release in rat periaqueductal

gray neurons, Neuropharmacology 41 (2001) 529–538.

[19] C. Molander, Q. Xu, G. Grant, The cytoarchitectonic organization of

the spinal cord in the rat. I. The lower thoracic and lumbosacral cord,

J. Comp. Neur. 230 (1984) 133–141.

[20] J.-L. Moreau, H.L. Fields, Evidence for GABA involvement in

midbrain control of medullary neurons that modulate nociceptive

transmission, Brain Res. 397 (1986) 37–46.

[21] V. Neugebauer, H.-G. Schaible, X. He, T. Lucke, P. Gundling, R.F.

Schmidt, Electrophysiological evidence for a spinal antinociceptive

action of dipyrone, Agents Actions 41 (1994) 62–70.

[22] H.R. Ochs, D.J. Greenblatt, D.R. Abernethy, R.M. Arendt, J. Gerloff,

W. Eichelkraut, N. Hahn, Cerebrospinal fluid uptake and peripheral

distribution of centrally acting drugs: relation to lipid solubility,

J. Pharm. Pharmacol. 37 (1985) 428–431.

[23] L.J. Pellegrino, A.S. Pellegrino, A.J. Cushman, A Stereotaxic Atlas of

the Rat Brain, Plenum, New York, 1979.

[24] S.F. Pong, S.M. Demuth, C.M. Kinney, P. Deegan, Prediction of

human analgesic dosages of nonsteroidal anti-inflammatory drugs

(NSAIDs) from analgesic ED50 values in mice, Arch. Int. Pharmaco-

dyn. 273 (1985) 212–220.

[25] S.M. Roychowdhury, H.L. Fields, Endogenous opioids acting at a

medullary A-opioid receptor contribute to the behavioral antinocicep-

tion produced by GABA antagonism in the midbrain periaqueductal

gray, Neuroscience 74 (1996) 863–872.

[26] A. Spilva de Lehr, Y. Muktans Spilva, Guia Spilva de las

Especialidades Farmaceuticas, Global Ediciones, C.A., Caracas,

2000/2001.

[27] V. Tortorici, H. Vanegas, Putative role of medullary off- and on-cells in

the antinociception produced by dipyrone (metamizol) administered

systemically or microinjected into PAG, Pain 57 (1994) 197–205.

[28] V. Tortorici, H. Vanegas, Antinociception induced by systemic or

PAG-microinjected lysine-acetylsalicylate in rats. Effects on tail-flick

related activity of medullary off- and on-cells, Eur. J. Neurosci. 7

(1995) 1857–1865.

[29] V. Tortorici, E. Vasquez, H. Vanegas, Naloxone partial reversal of the

antinociception produced by dipyrone microinjected in the periaque-

ductal gray of rats. Possible involvement of medullary off- and on-

cells, Brain Res. 725 (1996) 106–110.

[30] H. Vanegas, H.-G. Schaible, Prostaglandins and cyclooxygenases in

the spinal cord, Prog. Neurobiol. 64 (2001) 327–363.

[31] H. Vanegas, V. Tortorici, A. Eblen-Zajjur, E. Vasquez, PAG-micro-

injected dipyrone (metamizol) inhibits responses of spinal dorsal horn

neurons to natural noxious stimulation in rats, Brain Res. 759 (1997)

171–174.

[32] E. Vasquez, H. Vanegas, The antinociceptive effect of PAG-micro-

injected dipyrone in rats is mediated by endogenous opioids of the

rostral ventromedial medulla, Brain Res. 854 (2000) 249–252.

[33] E. Vasquez, D. Hernandez-Matheus, V. Tortorici, H. Vanegas, PAG-

microinjected dipyrone prevents the late response of spinal nocicep-

tive neurons to subcutaneous formalin in rats, Analgesia 4 (1999)

405–407.

[34] C.W. Vaughan, S.L. Ingram, M.A. Connor, M.J. Christie, How opioids

inhibit GABA-mediated neurotransmission, Nature (Lond.) 390

(1997) 611–614.

[35] T.L. Yaksh, J.C. Yeung, T.A. Rudy, Systematic examination in the rat

of brain sites sensitive to the direct application of morphine:

observation of differential effects within the periaqueductal gray,

Brain Res. 114 (1976) 83–103.