European Journal of Pharmacology 748 (2015) 115–122

Contents lists available at ScienceDirect

European Journal of Pharmacology

journal homepage: www.elsevier.com/locate/ejphar

Neuropharmacology and analgesia

Effects of metabolites of the analgesic agent dipyrone (metamizol)

on rostral ventromedial medulla cell activity in mice
Sabatino Maione a, Lilyana Radanova b, Danilo De Gregorio a, Livio Luongo a,
Luciano De Petrocellis c, Vincenzo Di Marzo c,n, Peter Imming b
a
Endocannabinoid Research Group at the Department of Experimental Medicine, Division of Pharmacology “L. Donatelli”, The Second University of Naples,
80138 Naples, Italy
b
Institut fuer Pharmazie, Martin-Luther-Universitaet Halle, Germany
c
Endocannabinoid Research Group at the Institute of Biomolecular Chemistry, Consiglio Nazionale delle Ricerche, Pozzuoli (NA), Italy

art ic l e i nf o a b s t r a c t

Article history: The molecular mechanism of action of dipyrone, a widely used antipyretic and non-opioid analgesic
Received 25 August 2014 drug, is still not fully understood. Actions upon peripheral inflamed tissues as well as the central nervous
Received in revised form system, especially upon the PAG-RVM axis, have been suggested. Dipyrone is a prodrug and its activity is
13 November 2014
due to its immediate conversion to its active metabolites. We tested the effect of two recently discovered
Accepted 15 December 2014
Available online 31 December 2014
metabolites of dipyrone, the arachidonoyl amides of 4-methylaminoantipyrine and 4-aminoantipyrine,
on the neurons of the rostral ventromedial medulla (RVM), which are part of the descending pathway of
Keywords: antinociception. These compounds reduced the activity of ON-cells and increased the activity of OFF-
Dipyrone cells. Both CB1 and TRPV1 blockade reversed these effects, suggesting that the endocannabinoid/
Arachidonic acid amides endovanilloid system takes part in the analgesic effects of dipyrone.
Periaqueductal gray
& 2015 Published by Elsevier B.V.
Rostral ventromedial medulla
Analgesia

1. Introduction stimulation of the PAG-RVM-axis (Vazquez et al., 2007), COX-1/

Dipyrone (1, metamizole) is one of the most commonly used
analgesic and antipyretic drugs worldwide (Nikolova et al., 2012).
In contrast to acidic NSAIDs, it is only weakly anti-inflammatory,
but has spasmolytic effects and causes no gastrointestinal lesions.
After oral administration, dipyrone is rapidly hydrolyzed to 4-
methylaminoantipyrine (2) (Scheme 1). Compound 2 is further
metabolized to 4-aminoantipyrine (3) and 4-formylaminoantipyrine
(4) (Levy et al., 1995). Two novel metabolites were recently identified
by us in mice: the arachidonoyl amides of 2 and 3, viz. 5 and 6
(Rogosch et al., 2012).
The molecular mechanism of action of dipyrone is still not fully
explained. Several mechanisms were proposed, including the invol-
vement of endogenous opioids (Vanegas and Tortorici, 2002),
Abbreviations: CB1, cannabinoid receptor type 1; COX, cyclooxygenase; I-RTX,
iodoresiniferatoxin; NSAIDs, non-steroidal antiinflammatory dugs; PAG, periaque-
ductal gray; RVM, rostral ventromedial medulla; TRPV1, transient receptor poten-
tial vanilloid 1
n
Correspondence to: Endocannabinoid Research Group at the Institute of
Biomolecular Chemistry, Consiglio Nazionale delle Ricerche, Pozzuoli (NA), Italy.
Tel.: þ 39 0818675093.
E-mail address: vdimarzo@icb.cnr.it (V. Di Marzo).
COX-2 inhibition by dipyrone and its metabolites 2 and 3 (Pierre
et al., 2007), COX-2 inhibition (Hinz et al., 2007), and interference
with components of the endocannabinoid system (Rogosch et al.,
2012). According to recent findings, the latter mechanism is con-
nected with a central analgesic effect (Rogosch et al., 2012, Escobar
et al., 2012), although we showed that cannabinoid CB1 receptors
clearly do not mediate the thermal antinociceptive actions of
dipyrone (Schlosburg et al., 2012). A recent study suggested that a
CB1 receptor inverse agonist and an anti-sense oligonucleotide
against CB1 receptors reversed the anti-hyperalgesic effect of 3,
but not that of dipyrone or 2 (Dos Santos et al., 2014). Finally, it was
reported that 2 induced hypothermia and inhibited PGE2-dependent
and –independent fever, while 3 only inhibited PGE2-dependent
fever (Malvar et al., 2014).
The periaqueductal gray-rostral ventromedial medulla (PAG-
RVM) pathway plays an important role in pain processing. PAG
modulates nociception via a descending pathway that relays in the
RVM and terminates in the spinal cord (Vanegas and Tortorici,
2002,Urban and Gebhart, 1992). There are three classes of neurons
in the RVM: ON-cells, OFF-cells and neutral cells that respond
differently to pain stimuli. Activation of ON-cells results in nocicep-
tion, and this makes them an interesting target for pain research
(Urban and Gebhart, 1992). The activity the ON-cells was suggested

http://dx.doi.org/10.1016/j.ejphar.2014.12.022
0014-2999/& 2015 Published by Elsevier B.V.
116 S. Maione et al. / European Journal of Pharmacology 748 (2015) 115–122

Scheme 1. Metabolites of dipyrone (1).

to serve as an electrophysiological marker of nociception

(De Novellis et al., 2012). OFF-cells are inhibited by pain stimuli,
and neutral cells show no response. AP:-4.84 mm
We have reported that the arachidonic acid amide 5 inhibits L:0.5 mm
TRPV1 (Sinning et al., 2008a), and that some fatty acid amides V:3.3 mm
exert their antinociceptive effect partly by acting on the spinal cord
and the PAG (De Novellis et al., 2012), and hence we have hypothe-
sized that the dipyrone metabolites 5 and 6 also act upon the PAG-
RVM axis, contributing to the strong analgesic effect of dipyrone. The
central analgesic effects of this compound were suggested to be
effected through PAG neurons and the descending antinociceptive
pathway already in 1986 (Carlsson et al., 1986). We monitored the
activity of ON- and OFF-cells after injection of 2, 5 and 6 into the PAG
and the influence of coinjection of TRPV1 and CB1 antagonists. The
aim of this study was to evaluate the interaction of the arachidonic
acid conjugates of 5 and 6 with the PAG-RVM axis and thus further
AP:-6.48 mm
evaluate how and how much the analgesic activities of dipyrone are
mediated through this axis. L:0.3-0.5 mm
V:4.5-6 mm

2. Materials and methods

2.1. Animals
Adult male CD1 mice (20–25 g) were purchased from Harlan
Laboratories (Milan, Italy) and were housed under controlled
illumination (12:12 h light/dark cycle; light on 06.00 h) and envir-
onmental conditions (room temperature 20–22 1C, humidity 55–
60%) for at least 1 week before the beginning of experiments. Chow
and tap water were available ad libitum. All experimental proce-
dures were conducted in conformity with protocols approved by
the Animal Ethics Committee of the Second University of Naples.
Animal care was in compliance with the IASP and European
Community (E.C. L358/1 18/12/86) guidelines on the use and
protection of animals in experimental research. All efforts were
made to minimize animal suffering and to reduce the number of
animals used.
Fig. 1. Schematic illustration of the location of ventrolateral periaqueductal gray
(VL PAG) microinjection sites (A) and RVM ON or OFF cell recording sites (B).
Vehicle or drug microinjections were performed in the VL PAG (filled circles). The
open circles indicate microinjections accidentally or intentionally performed out-
side of VL PAG, the effects of which have been considered in the study for location
specificity (A). Cell recordings were performed by lowering a tungsten electrode
into the RVM. ON cells (filled triangles) or OFF cells (open triangles) recording sites
are shown in B.
2.2. Surgical preparation for intra-PAG microinjection
(Fig. 1) For electrophysiological experiments, mice were anesthe-
tized with pentobarbital (50 mg/kg, i.p.) and a 26-gauge, 10 mm long
stainless steel guide cannula was stereotaxically lowered until its tip
was 1 mm above the left VL-PAG by applying coordinates (AP:
 4,84 mm from bregma, L: 0.5 mm from midline, V: 3,3 mm below
 S. Maione et al. / European Journal of Pharmacology 748 (2015) 115–122
the dura) from the Atlas of Paxinos and Watson (1986). The cannulae
were anchored with dental cement to a stainless steel screw in the
skull. We used a David Kopf stereotaxic apparatus (David Kopf
Instruments, Tujunga, CA, USA) with the animal positioned on a
homeothermic temperature control blanket (Harvard Apparatus
Limited, Edenbridge, Kent).
2.3. Intra-VL PAG microinjections
Direct intra-VL PAG administration of drugs, or respective
vehicles, was conducted with a stainless steel cannula connected
by a polyethylene tube to a SGE 1-microlitre syringe, inserted
through the guide cannula and extended 1 mm beyond the tip of
the guide cannula to reach the VL PAG. Vehicle or drug solutions
were administered into the VL-PAG in a final volume of 0.8 mL.
Microinjection was performed over a period of 60 s and the
injection cannula gently removed 2 min later. At the end of the
experiment, a volume of 0.8 μL of neutral red (0.1%) was also
injected into the VL PAG 30–40 min before killing the mouse. Mice
were then perfused intracardially with 20 ml phosphate buffer
solution (PBS) followed by 200 ml 4% paraformaldehyde solution
in PBS. The brains were removed and immersed in a saturated
paraformaldehyde solution for 24 h. The injection sites were
ascertained by using 2 consecutive sections (40 μm), one stained
with cresyl violet to identify nuclei and the other unstained in
order to determine dye spreading.
2.4. RVM extracellular recording
After implantation of the guide cannula into the VL PAG, a glass
insulated tungsten filament electrode (3–5 MΩ) (Frederick Haer & Co.,
ME, USA) was stereotaxically lowered, through a small craniotomy,
into the RVM (6,48 mm caudal to bregma, 0.3–0.5 mm lateral and
4–6 mm depth from the surface of the brain) using the coordinates
from the atlas of Paxinos and Watson (1986). The jugular vein was
cannulated to allow intravenous anesthetic administration (propofol,
8–10 mg/kg/h, i.v.). As described by Fields et al. (1982) RVM cells were
identified by their response to thermal noxious stimulus by applica-
tion of a thermoceptive stimulation of the tail able to evoke a
withdrawal (tail flick response) (Fields et al., 1982). These neurons
were also identified by their response to noxious stimulation at the
hind paw (press/pinch). RVM ON cells were identified by a burst of
activity that begins just after mechanical stimulus applied on the hind
paw while OFF cells were identified by the fact that they cease firing at
that time. Mechanical stimuli were applied using a spring-operated
forceps with a force (4300 g/10 mm2 ando400 g/10 mm2) that
squeezed the tissue (painful pressure). In our experiments nociceptive
mechanical stimuli were elicited every 5 min for at least 15 min prior
to microinjecting drugs, or the respective vehicle, 0.05% dimethyl
sulfoxide (DMSO) in artificial cerebrospinal fluid (ACSF, composition in
mM: KCl 2.5; NaCl 125; MgCl2 1.18; CaCl2 1.26) into the VL-PAG. The
recorded signals were amplified and displayed on an analog and
digital storage oscilloscope to ensure that the unit under study was
unambiguously discriminated throughout the experiment. Signals
were also processed by an interface (CED 1401) (Cambridge Electronic
Design Ltd., UK) connected to a Pentium III PC. Spike2 software (CED,
version 4) was used to create peristimulus rate histograms on-line and
to store and analyze digital records of single-unit activity off-line.
Configuration, shape, and height of the recorded action potentials
were monitored and recorded continuously, using a window discri-
minator and Spike2 software for on-line and off-line analysis. Once an
RVM cell was identified from its background activity, we optimized
spike size before all treatments. This study only included neurons
whose spike configuration remained constant and could clearly be
discriminated from activity in the background throughout the experi-
ment, indicating that the activity from one neuron only and from the
117
same one neuron was measured. The recording site was marked with
a 20 μA DC current for 20 s. After fixation by immersion in a 4%
paraformaldehyde, the recording sites were identified. In each mouse,
only one neuron was recorded before and after vehicle or drug
administration. The neuron responses, before and after intra-VL PAG
vehicle or drug microinjections were measured and expressed as
spikes/s (Hz). At the end of the experiment, each animal was killed
with a lethal dose of urethane, the microinjection site was marked
with 0.2 μl of a Cresyl Violet solution and the recording site marked
with a 20 μA DC current for 20 s. After fixation by immersion in 10%
formalin, the microinjection and recording sites were identified.
These procedures were used and parameters measured for
6 groups of mice (n ¼5 for each treatment):
1. Mice who received intra-PAG injection of vehicle (0.05% dimethyl
sulfoxide).
2. Mice who received intra-PAG injection of 2 (2 nmol).
3. Mice who received intra-PAG injection of 6 (2 nmol).
4. Mice who received intra-PAG injection of 5 (2 nmol).
5. Mice who received intra-PAG injection of 5 (2 nmol) in combi-
nation with AM 251 (0.5 nmol).
6. Mice who received intra-PAG injection of 5 (2 nmol) in combi-
nation with I-RTX (0.5 nmol)
The doses were adjusted for mouse from (Escobar et al., 2013).
2.5. Data analysis and statistical procedures
All data are given as means7SEM. For electrophysiology experi-
ments mixed design two-way ANOVA followed by the Bonferroni
post-hoc test was used to analyze statistical differences between the
different groups of mice. P valueso0.05 were considered statistically
significant. F values and the degree of freedom were calculated with
SPSS statistic software.
2.6. Substances
2 (4-methylaminoantipyrine) was a generous gift from Sanofi-
Aventis Deutschland GmbH, Berlin. AM251 and I-RTX were purchased
from Tocris, UK. Arachidonoyl-4-methylaminopyrine (5Z,8Z,11Z,14Z)-
Icosa-5,8,11,14-tetraenoic acid ((1,5-dimethyl-3-oxo-2-phenyl-2,3-
dihydro-1 H-pyrazol-4-yl)-methyl-amide), 5, was prepared according
to the following protocol:
Arachidonic acid (123.3 mg, 0.405 mmol) and benzotriazol-1-
yloxytripyrrolidinophosphonium hexafluorophosphate (PyBOP;
191mg, 0.368 mmol) were dissolved in dichloromethane (10 ml). The
mixture was cooled to 0 1C and a solution of 4-methylaminoantipyrine
(2, 80 mg, 0.368 mmol in 5 ml) and 1.5 eq of N,N-diisopropylethylamine
were added. The mixture was stirred for 3 h at 0 1C. The solvent was
evaporated to dryness, redissolved in methyl tert-butyl ether and
washed with aqueous 0.1 M HCl and saline. The organic phase was
dried over anhydrous magnesium sulfate and evaporated to dryness.
The residue was purified twice by flash chromatography on normal
phase silica gel (40–63 mm), eluting with trichloromethane/methanol
(98/2). Yield, 118 mg (64%).
Analytical data were as follows: ESI-MS (5 kV), m/z (%): 504.5
(14, M þ ), 526.5 (100, M þ Na), 1030.3 (31, 2 M þ Na). – 1H NMR
(400 MHz, CDCl3), δ: 7.29–7.51 (m, 5H, Ar-H), 5.44–5.25 (m, 8 H, 4x
CH¼ CH), 3.30 (s, peak of rotational isomer of CO-N-CH3), 3.14 (s,
3H, CO-N-CH3), 3.12 (s, 3H, N-CH3), 3.10 (s, peak of rotational
isomer of N-CH3), 2.85-2.71 (m, 6H, 3x CH ¼CH-CH2-CH¼CH),
2.49 (t, 3J¼7.5 Hz, peak of rotational isomer of CH2-CO), 2.32–2.00
(m, 9H, 2x CH2-CH¼ CH, CH2-CO, C ¼C-CH3), 1.77 (tt, peak of
rotational isomer of CH2-CH2-CO), 1.68 (tt, 3J¼ 3J¼ 7.4 Hz, 2H,
CH2-CH2-CO), 1.38-1.21 (m, 6 H, 3xCH2), 0.88 (t, 3J¼ 6.8 Hz, 3H,
CH3). – 13C NMR (100 MHz, CDCl3), δ: 174.3, 161.6, 151.6, 134.5,
118 S. Maione et al. / European Journal of Pharmacology 748 (2015) 115–122

130.4, 129.5, 129.2, 128.5, 128.3, 128.1, 127.8, 127.5, 127.0, 124.0,
115.2, 35.9, 35.6, 32.6, 31.4, 29.2, 27.1, 26.7, 25.6, 25.5, 24.9, 22.5,
22.5, 14.0, 10.5. – C32H45N3O2 (503.72): calc. C: 76.30, H: 9.00,
N: 8.34, found C: 75.90H: 9.15, N: 8.50.
Arachidonoyl-4-aminoantipyrine (5Z,8Z,11Z,14Z)-Icosa-5,8,11,14-
tetraenoic acid ((1,5-dimethyl-3-oxo-2-phenyl-2,3-dihydro-1H-pyra-
zol-4-yl)-amide), 6, was prepared according to the following protocol:
Arachidonic acid (201 mg, 0.66 mmol) and dimethylformamide
(48.2 mg, 0.66 mmol) were dissolved in dry tetrahydrofurane
(10 ml). The mixture was cooled to 0 1C, two equivalents of oxalyl
chloride (167.5 mg, 1.32 mmol) were added dropwise, and stirring was
continued for 1 h at 0 1C. A solution of aminoantipyrine (3, 670.7 mg,
3.3 mmol) in 5 ml of dichloromethane was added and stirring
continued for further 30 min at ambient temperature. The mixture
was washed with aqueous 0.1 M HCl, satured aqueous sodium
hydrogen carbonate solution and saline. The organic phase was dried
over anhydrous magnesium sulfate and evaporated to dryness. The
residue was purified by flash chromatography on normal phase silica
gel (40–63 mm), eluting with ethyl acetate/methanol (98/2). Yield,
179.7 mg (54%).
Analytical data were as follows: EI-MS (70 eV), m/z (%) 203 (1 0 0),
204 (82.9), 56 (70), 480 (35.7, Mþ). – 1H NMR (400 MHz, CDCl3),
δ: 8.26 (br s, 1H, CO-NH), 7.45-7.25 (m, 5H, Ar-H), 5.42-5.27 (m, 8H, 4x
CH¼CH), 3.04 (s, 3 H, N-CH3), 2.84-2.76 (m, 6H, 3x CH¼ CH-CH2-
CH¼CH), 2.29 (t, 3J¼7.5 Hz, 2H, CH2-CO), 2.20 (s, 3 H, C¼C-CH3), 2.11-
2.00 (m, 4H, 2x CH2-CH¼CH), 1.70 (tt, 3J¼ 3J¼ 7.5 Hz, 2H, CH2-CH2-
CO), 1.38-1.21 (m, 6H, 3xCH2), 0.88 (t, 3J¼7.1 Hz, 3H, CH3). – 13C NMR
(100 MHz, CDCl3), δ: 172.1, 161.8, 149.6, 134.6, 130.4, 129.3, 129.2,
128.51, 128.48, 128.3, 128.1, 127.9, 127.5, 126.8, 124.2, 109.0, 36.2, 35.6,
31.5, 29.3, 27.2, 26.8, 25.7, 25.6, 22.6, 22.6, 14.1, 12.5. – C31H43N3O2
(489.71): calc. C: 76.03, H: 8.85, N: 8.58, found C: 76.25H: 9.03, N: 8.32.
3. Results
3.1. Effect of intra-PAG administration of metabolites of dipyrone on
the ongoing activity of RVM ON and OFF cell in mice.
Neurons identified in the RVM as ON cells by a burst of activity
just after mechanical responses were spontaneously active in 33.3%
of the cases and inactive in the remaining cases. ON cells with
spontaneous activity were chosen to characterize the postdrug
changes in their spontaneous activity. The population of spontaneous
active ON cells had a mean frequency of spontaneous activity of
7.470.9 spikes/s. Microinjection of vehicle (0.05% dimethyl sulfox-
ide) did not change the spontaneous activity of the ON cells (7.570.8
spikes/s, n¼5) in mice (Fig. 2A). Intra-PAG microinjection of 2
(2 nmol) caused a non-statistically significant decrease in the spon-
taneous firing activity of the ON cells, whereas the administration of
6 (2 nmol) and 5 (2 nmol) induced a significant decrease on the
ongoing activity that lasted until 60 min post injection (3.0170.5
spikes/s, Po0.05, F(3,16)¼ 36.67), n¼5 and 2.7570.28 spikes/s,
Po0.05, F(3,16)¼5.77, n¼ 5, respectively (Fig. 2A).
The population of cells identified as OFF cells were always
spontaneously active and they had a mean frequency of spontaneous
activity of 8.170.8, n¼10 in mice. Microinjection of vehicle did not
change the spontaneous activity of OFF cells (8.271.0 spikes/s, n¼5)
(Fig. 3A). The intra-PAG administration of 2 (2 nmol) did not

Fig. 2. “A”, “B” and “C” show the spontaneous firing, evoked frequency and the onset of burst of RVM-ON cells of different groups of mice (n ¼5) which received intra-PAG
injection of vehicle (0.05% dimethyl sulfoxide) or metabolites of dipyrone. The vehicle or drugs were administered at time 0. Data are given as means 7 standard error of the
means (S.E.M). n indicates statistically significant difference versus vehicle-treated mice. P o 0.05 was considered as value of significance, calculated using two-way ANOVA
followed by the Bonferroni post-hoc test.
 S. Maione et al. / European Journal of Pharmacology 748 (2015) 115–122 119

Fig. 3. “A”, “B” and “C” show the spontaneous firing, pause duration and the onset of pause of RVM-OFF cells of different groups of mice (n¼ 5) which received intra-PAG
injection of vehicle (0.05% dimethyl sulfoxide) or metabolites of dipyrone. The vehicle or drugs were administered at time 0. Data are given as means 7 standard error of the
means (S.E.M). n indicates statistically significant difference versus vehicle-treated mice. Po 0.05 was considered as value of significance and was calculated using two-way
ANOVA followed by the Bonferroni post-hoc test.

significantly change the spontaneous activity of OFF cells in mice. Intra-
PAG microinjections of 6 (2 nmol) and 5 (2 nmol) increased the
spontaneous firing activity of OFF cells, an effect that was already
significant after 20 min and lasted until 60 min (12.370.8 spikes/s,
Po0.05, F(3,16)¼299.2, n¼5 and 14.17 1.1 spikes/s Po0.05, F(3,16)¼
25.56, n¼ 5, respectively) (Fig. 3A).
3.2. Effect of intra-PAG administration of dipyrone metabolites on
mechanical stimulation-related ON and OFF cell activity in mice
The population of ON cells had a mechanical stimulation-
induced burst of firing of 13.771.8 spikes/s and the onset of the
burst was 2650740 milliseconds (ms). The population of OFF cells
had a pause of 7.071.5 s and an onset of pause of 2500745 ms.
Microinjections of vehicle did not change ON cell burst
(13.572.8 spikes/s, n¼5), the onset of burst (2590750 ms, n¼ 5)
nor OFF cell pause (771.2 s, n¼5) and the onset of the pause
(4100790 ms, n¼5 Figs. 1B and C, 2B and C). Intra-PAG micro-
injections of 2 (2 nmol) caused a significant decrease in ON cell
burst (5.971.5 spikes/s, Po 0.05, F(3,16)¼58.39, n¼5) and OFF cell
pause (2.9071.8 s, Po0.05, F(3,16)¼3.36; n¼5) and an increase in
both the onset of ON cell burst (41507110 ms, Po0.05, F(3,16)¼
160.4, n¼5) and the onset of OFF cell pause (16,3007230 ms,
Po0.05, F(3,16)¼176.3, n¼ 5) (Figs. 2B and C, 3B, and 3C). Intra-PAG
microinjections of the two amides had a greater effect with respect
to 2. In fact the administration of 6 (2 nmol) and 5 (2 nmol) induced
a decrease in ON cell burst (4.472.3 spikes/s, Po0.05, F(3,16)¼7.3,
n¼ 5 and 3.671.4 spikes/s, Po0.05, F(3,16)¼411.2, n¼ 5), and in
OFF cell pause (2.672.2 s, Po0.05, F(3,16)¼7.96, n¼5; 1.271.4 s,
Po 0.05, F(3,16)¼ 47.72, n¼ 5, respectively) (Figs. 2B and 3B), an
increase in both the onset of ON cell burst (63007150 ms, Po0.05,
F(3,16)¼ 125.2, n¼5; 52707290 ms, F(3,16)¼81.92, respectively)
and OFF cell pause (22,3007260 ms, Po0.05, F(3,16)¼50.16, n¼ 5;
21,2357235 ms, Po0.05, F(3,16)¼46.29, n¼5, respectively)
(Figs. 2C and 3C).
3.3. Effect of intra-PAG administration of 5 in combination with AM
251 or I-RTX on extracellular recording
To investigate the possible involvement of the endocannabi-
noid/endovanilloid system in the pharmacological effects of the
compounds, we carried out further experiments with 5 (2 nmol),
which exerted the strongest effects in the experiments described
above, in combination with AM251 (0.5 nmol) or I-RTX (0.5 nmol).
The intra-PAG administration of vehicle did not change ON cell
burst, ongoing activity and the onset of burst evoked by mechan-
ical stimulation, nor the pause, the onset of pause and ongoing
activity of OFF cells in mice. The effect induced by intra-PAG
injection of 5 (2 nmol), on both ON and OFF cell activity was
prevented by the intra-PAG pre-injection of AM251 (0.5 nmol) or I-
RTX (0.5 nmol). The compound was administered 10 min after
injection of AM251 or I-RTX (Figs. 4 and 5A–C). All the antagonists
were inactive per se at the dose used (data not shown).
4. Discussion
In this study, we investigated whether metabolites of dipyrone
(2, 5 and 6) modified the ongoing activity of ON- and OFF-cells of the
RVM when injected into the PAG. Indeed, we observed that 20 min
120 S. Maione et al. / European Journal of Pharmacology 748 (2015) 115–122

Fig. 4. “A”, “B” and “C” show the spontaneous firing, evoked frequency and the onset of burst of RVM-ON cells of different groups of mice (n ¼5) which received intra-PAG
injection of vehicle (0.05% dimethyl sulfoxide) or 5 (2 nmol) alone or in combination with AM251 (0.5 nmol) or I-RTX (0.5 nmol). In figure “A” the black arrow indicates the
administration of vehicle, 5 (alone), AM251 and I-RTX. The gray arrow indicates the administration of 5 10 min after the injection of AM251 or I-RTX. Data are presented as
means 7 standard error of the means (S.E.M). n indicates statistically significant difference versus mice who received vehicle. 1 and # indicate statistically significant
difference versus mice who received 5. P o 0.05 was considered as value of significance calculated using two-way ANOVA followed by the Bonferroni post-hoc test.

after intra-PAG injection, the arachidonic acid amides of methyla-
minoantipyrin and aminoantipyrin (5 and 6) caused a significant
reduction of ON cell activity and a significant increase in OFF cell
activity. These arachidonoyl amides had a stronger effect than
methylaminoantipyrine (2). We hypothesize that this is part of the
molecular mechanism of action of dipyrone, as it is known that
reducing RVM ON cells activity leads to antinociception. The same
effect was published by us for the endogenous fatty acid amide
palmitoylethanolamide (PEA), a peroxisome proliferator-activated
receptor-α (PPAR-α) ligand. PEA induced antinociceptive effects in
tail-flick-tests presumably through modifying the activity of RVM
neurons (De Novellis et al., 2012).
The effect of 5 on ON and OFF cell activity was antagonized by
prior injection of AM251 (a CB1 receptor inverse agonist), suggest-
ing that it was mediated in part through the activation of CB1
receptors, in agreement with our previous report of a weak but
measurable affinity of 5 for these receptors (Ki ¼7.8 70.8 mM;
(Rogosch et al., 2012)), and with the well known stimulatory
actions of CB1 agonists over the descending antinociceptive path-
way (Maione et al., 2006). Also in agreement with our present
findings is the recent report of dipyrone acting via CB1 receptors in
the PAG-RVM axis during inflammation in rats ((Escobar et al.,
2012). Intra-VL-PAG microinjection of dipyrone induced antinoci-
ception, which was reduced by AM251. In the light of our present
data, we believe that this effect might have been caused not by
dipyrone itself, because of its rapid non-enzymatic hydrolysis
to 2, nor by this dipyrone metabolite, but instead by the amides
5 and 6.
We also found that the effect of 5 was antagonized by I-RTX, a
selective TRPV1 antagonist. At first glance, this finding is surprising
in view of the fact that 5, unlike 6, behaves as a weak antagonist
(IC50 3 mM), and not as an agonist, at the human recombinant TRPV1
channel (Sinning et al., 2008b), and in view of the previous
observation that activation, rather than antagonism, of TRPV1
enhances the activity of the descending antinociceptive pathway
(Starowicz et al., 2007). Several explanations for this effect are
possible, warranting a separate study to distinguish among them.
First, TRPV1 desensitization (and hence inactivation) in the PAG has
been shown also to activate this pathway (cf. (McGaraughty
et al., 2003)). However, the effect of a weak TRPV1 antagonist such
as 5 should not be antagonized by a more potent TRPV1 antagonist
such as I-RTX. Species-dependent differences are possible with
TRPV1 ligands, and, although unlikely, the possibility cannot be
excluded that 5 might behave as an agonist at murine TRPV1.
Perhaps more importantly, it must be remembered that 5 is also a
weak inhibitor of COX-2 (IC50 ¼69 mM) (Rogosch et al., 2012), which
– like FAAH – is a key enzyme of endocannabinoid metabolism (see
Hermanson et al. (2014) for a review). Thus, COX-2 inhibition by 5
might increase the levels in tissues of the endocannabinoid/endo-
vanilloid anandamide, and indirectly activate CB1 and TRPV1, thus
stimulating the descending antinociceptive pathway also via TRPV1
(Maione et al., 2006; Starowicz et al., 2007). Indeed, we injected a
2.5 nmol/μl concentration of 5, which, even keeping into account
the rapid diffusion of this compound in the tissue, is likely to be
sufficiently high not only to activate CB1 directly, but also to inhibit
COX-2, hence activating CB1 and TRPV1 indirectly, and explaining
 S. Maione et al. / European Journal of Pharmacology 748 (2015) 115–122 121

Fig. 5. “A”, “B” and “C” show the spontaneous firing, pause duration and the onset of pause of RVM-OFF cells in different groups of mice (n¼ 5) which received intra-PAG
injection of vehicle (0.05% dimethyl sulfoxide) or 5 (2 nmol) alone or in combination with AM251 (0.5 nmol) or I-RTX (0.5 nmol). In figure “A” the black arrow indicates the
administration of vehicle, 5 (alone), AM251 and I-RTX. The gray arrow indicates the administration of 5 10 min after the injection of AM251 or I-RTX. Data are given as
means 7 standard error of the means (S.E.M).n indicates statistically significant difference versus mice who received vehicle. 1 and # indicate statistically significant difference
versus mice who received 5. Po 0.05 was considered as value of significance, calculated using two-way ANOVA followed by the Bonferroni post-hoc test.

the reversal of the effect of this compound not only by AM251, but
also by I-RTX. This effect may have been enhanced by a possible
initial agonistic effect of I-RTX, which was observed with higher
concentration of this compound (1–30 mM; Shimizu et al., 2005).
Evidence accumulates that TRPV1 antagonism is not an all-or-none
affair (Blumberg et al., 2011), and our results with 5 and I-RTX
probably are another instance of the intricacies of intervention with
this highly sensitive and multiply cross-linked ion channel (Garami
et al., 2010).
According to our previous findings, the fatty acid amide hydro-
lase (FAAH) plays an important role for both formation and hydro-
lysis of 5 in mouse brain, and inhibition of this enzyme, and
subsequent endocannabinoid level elevation, also stimulates the
descending antinociceptive pathway (Maione et al., 2006). We
showed that 5 does not inhibit FAAH (Di Marzo, unpublished
results). So FAAH inhibition with concomitant increase of ananda-
mide levels is probably not part of the molecular mechanism of
action of dipyrone and the metabolites investigated here. However,
5 might still occupy FAAH and thus compete with anandamide,
slowing down its metabolic degradation and thus increasing endo-
cannabinoid/endovanilloid tone.
Putting together all the pieces of the dipyrone molecular effects
puzzle (Rogosch et al., 2012), the following concurrence of dipyr-
one metabolite actions has some plausibility. The metabolite 2,
formed non-enzymatically, was reported to inhibit PGE2 biosynth-
esis. PGE2 exerts a pro-nociceptive effect when directly injected
into the PAG (Palazzo et al., 2011). Thus, the metabolism of 2
would reduce the levels of a proalgesic compound in the PAG-RVM
axis. At the same time, 5, which is formed from 2 in a few minutes,
would inhibit the descending antinociceptive pathway through
one or more of the above mentioned mechanisms.
In conclusion, the effects at ON- and OFF-cells, reported here for 5
and 6 for the first time, both point in the same direction: reduced
nociception. We suggest therefore that these dipyrone metabolites
cause antinociception at least in part through the PAG-RVM axis,
acting on the constituents of the endocannabinoid and endovanilloid
system, most likely CB1 and TRPV1 receptors. Whether or not the
formation of these metabolites and the mechanism of action reported
here are at all responsible for the analgesic effects of dipyrone in vivo
will require further investigation.
References
Blumberg, P.M., Pearce, L.V., Lee, J., 2011. TRPV1 activation is not an all-or-none
event: TRPV1 partial agonism/antagonism and its regulatory modulation. Curr.
Top. Med. Chem. 11, 2151–2158.
Carlsson, K.H., Helmreich., J., Jurna, I., 1986. Activation of inhibition from the
periaqueductal gray matter mediates central analgesic effect of metamizol
(dipyrone). Pain 27, 373–390.
De Novellis, V., Luongo, L., Guida, F., Cristino, L., Palazzo, E., Russo, R., Marabese, I.,
D’Agostino, G., Calignano, A., Rossi, F., Di Marzo, V., Maione, S., 2012. Effects of intra-
ventrolateral periaqueductal gray palmitoylethanolamide on thermoceptive thresh-
old and rostral ventromedial medulla cell activity. Eur. J. Pharmacol. 676, 41–50.
Dos Santos, G.G., Dias, E.V., Teixeira, J.M., Athie, M.C., Bonet, I.J., Tambeli, C.H.,
Parada, C.A., 2014. The analgesic effect of dipyrone in peripheral tissue involves
122 S. Maione et al. / European Journal of Pharmacology 748 (2015) 115–122
two different mechanisms: neuronal KATP channel opening and CB1 receptor
activation. Eur. J. Pharmacol. 741C, 124–131.
Escobar, W., Ramirez, K., Avila, C., Limongi, R., Vanegas, H., Vazquez, E., 2012.
Metamizol, a non-opioid analgesic, acts via endocannabinoids in the PAG-RVM
axis during inflammation in rats. Eur. J. Pain 16, 676–689.
Fields, H.L., Bry, J., Hentall, I., Zorman, G., 1982. The activity of neurons in the rostral
medulla of the rat during withdrawal from noxious heart. J. Neurosci. 3,
2545–2552.
Garami, A., Shimansky, Y.P., Pakai, E., Oliveir, D.L., Gavva, N.R., Romanovsky, A.A.,
2010. Contributions of different Modes of TRPV1 activation to TRPV1
antagonist-induced hyperthermia. J. Neurosci. 30, 1435–1440.
Hermanson, D.J., Gamble-George, J.C., Marnett, L.J., Patel, S., 2014. Substrate-
selective COX-2 inhibition as a novel strategy for therapeutic endocannabinoid
augmentation. Trends Pharmacol. Sci. 35, 358–367.
Hinz, B., Cheremina, O., Bachmakov, J., Renner, B., Zolk, O., Fromm, M.F., Brune, K., 2007.
Dipyrone elicits substantial inhibition of peripheral cyclooxygenases in humans:
new insights into the pharmacology of an old analgesic. FASEB J. 21, 2343–2351.
Levy, M., Zylber-Katz, E., Rosenkranz, B., 1995. Clinical pharmacokinetics of
dipyrone and its metabolites. Clin. Pharmacokinet. 28, 216–234.
Maione, S., Bisogno, T., de Novellis, V., Palazzo, E., Cristino, L., Valenti, M., Petrosino, S.,
Guglielmotti, V., Rossi, F., Di Marzo, V., 2006. Elevation of endocannabinoid levels
in the ventrolateral periaqueductal gray through inhibition of fatty acid amide
hydrolase affects descending nociceptive pathways via both cannabinoid recep-
tor type 1 and transient receptor potential vanilloid type-1 receptors.
J. Pharmacol. Exp. Ther. 316, 969–982.
Malvar, D.D., Aguiar, F.A., Vaz, A.D., Assis, D.C., de Melo, M.C., Jabor, V.A.,
Kalapothakis, E., Ferreira, S.H., Clososki, G.C., de Souza, G.E., 2014. The dipyrone
metabolite 4-MAA induces hypothermia and inhibits PGE2 -dependent and-
independent fever while 4-AA only blocks PGE2 -dependent fever. Br.
J. Pharmacol. 171, 3666–3679.
McGaraughty, S., Chu, K.L., Bitner, R.S., Martino, B., El Kouhen, R., Han, P., Nikkel, A.L.,
Burgard, E.C., Faltynek, C.R., Jarvis, M.F., 2003. Capsaicin infused into the PAG
affects rat tail flick responses to noxious heat and alters neuronal firing in the
RVM. J. Neurophysiol. 90, 2702–2710.
Nikolova, I., Tencheva, J., Voinikov, J., Petkova, V., Benbasat, N., Danchev, N., 2012.
Metamizole: a review profile of a well-known “forgotten” drug. Biotechnol.
Biotechnol. Eq 26, 3329–3337.
Palazzo, E., Guida, F., Gatta, L., Luongo, L., Boccella, S., Bellini, G., Marabese, I., de
Novellis, V., Rossi, F., Maione, S., 2011. EP1 receptor within the ventrolateral
periaqueductal gray controls thermonociception and rostral ventromedial
medulla cell activity in healthy and neuropathic rat. Mol. Pain 7, 82.
Paxinos, G., Watson, C., 1986. The Rat Brain in Stereotaxic Coordinates, 2nd ed.
Academic Press, San Diego.
Pierre, S.C., Schmidt, R., Brenneis, C., Michaelis, M., Geisslinger, G., Scholich, K., 2007.
Inhibition of cyclooxygenases by dipyrone. Brit. J. Pharmacol. 151, 494–503.
Rogosch, T., Sinning, C., Podlewski, A., Watzer, B., Schlosburg, J., Lichtman, A.H.,
Cascio, M.G., Bisogno, T., Di Marzo, V., Nüsing, R., Imming, P., 2012. Novel
bioactive metabolites of dipyrone (metamizol). Bioorg. Med. Chem. 20,
101–107.
Schlosburg, J.E., Radanova, L., Di Marzo, V., Imming, P., Lichtman, A.H., 2012.
Evaluation of the endogenous cannabinoid system in mediating the behavioral
effects of dipyrone (metamizol) in mice. Behav. Pharmacol. 23, 722–726.
Shimizu, I., Iida, T., Horiuchi, N., Caterina, M.J., 2005. 5-Iodoresiniferatoxin evokes
hypothermia in mice and is a partial transient receptor potential vanilloid
1 agonist in vitro. J. Pharmacol. Exp. Ther. 314, 1378–1385.
Sinning, C., Watzer, B., De Petrocellis, L., Di Marzo, V., Imming, P., 2008a.
Dopamides, vanillylamides, ethanolamides, and arachidonic acid amides of
anti-inflammatory and analesig drug substances as TRPV1 ligands. ChemMed-
Chem 3, 1956–1964.
Sinning, C., Watzer, B., Coste, O., Nüsing, R.M., Ott, I., Ligresti, A., Di Marzo, V.,
Imming, P., 2008b. New analgesics synthetically derived from the paracetamol
metabolite N-(4-hydroxyphenyl)-(5Z,8Z,11Z,14Z)-icosatetra-5,8,11,14-enamide.
J. Med. Chem. 51, 7800–7805.
Starowicz, K., Maione, S., Cristino, L., Palazzo, E., Marabese, I., Rossi, F., de Novellis,
V., Di Marzo, V., 2007. Tonic endovanilloid facilitation of glutamate release in
brainstem descending antinociceptive pathways. J. Neurosci. 27, 13739–13749.
Urban, M.O., Gebhart, G.F., 1992. Supraspinal contributions to hyperalgesia. Proc.
Natl. Acad. Sci. USA 96, 7687–7692.
Vanegas, H., Tortorici, V., 2002. Opioidergic effects of nonopioid analgesics on the
central nervous system. Cell. Mol. Neurobiol. 22, 655–661.
Vazquez, E., Escobar, W., Ramirez, K., Vanegas, H., 2007. A nonopioid analgesic acts
upon the PAG-RVM axis to reverse inflammatory hyperalgesia. Eur. J. Neurosci.
25, 471–479.