Novel bioactive metabolites of dipyrone (metamizol)

Article abstract of DOI:10.1016/j.bmc.2011.11.028

Dipyrone is a common antipyretic drug and the most popular non-opioid analgesic in many countries. In spite of its long and widespread use, molecular details of its fate in the body are not fully known. We administered dipyrone orally to mice. Two unknown

Full text of DOI:10.1016/j.bmc.2011.11.028

Bioorganic & Medicinal Chemistry 20 (2012) 101–107
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Novel bioactive metabolites of dipyrone (metamizol)
Tobias Rogosch a, Christian Sinning b, Agnes Podlewski a, Bernhard Watzer a, Joel Schlosburg c,
Aron H. Lichtman c, Maria G. Cascio d, , Tiziana Bisogno d, Vincenzo Di Marzo d, Rolf Nüsing e,
Peter Imming b,
⇑
a Zentrum für Kinder- und Jugendmedizin, Philipps-Universität, Marburg, Germany
b Institut für Pharmazie, Martin-Luther-Universität, Halle, Germany
c Department of Pharmacology and Toxicology, Virginia Commonwealth University, Richmond, VA, USA
d Endocannabinoid Research Group, Istituto di Chimica Biomolecolare, Consiglio Nazionale delle Ricerche, Pozzuoli (NA), Italy
e Institut für Klinische Pharmakologie, Pharmazentrum Frankfurt, Frankfurt/M., Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Dipyrone is a common antipyretic drug and the most popular non-opioid analgesic in many countries. In
spite of its long and widespread use, molecular details of its fate in the body are not fully known. We
administered dipyrone orally to mice. Two unknown metabolites were found, viz. the arachidonoyl
amides of the known major dipyrone metabolites, 4-methylaminoantipyrine (2) and 4-aminoantipyrine
(3). They were identified by ESI-LC–MS/MS after extraction from the CNS, and comparison with reference
substances prepared synthetically. The arachidonoyl amides were positively tested for cannabis receptor
binding (CB1 and CB2) and cyclooxygenase inhibition (COX-1 and COX-2 in tissues and as isolated
enzymes), suggesting that the endogenous cannabinoid system may play a role in the effects of dipyrone
against pain.
Received 4 August 2011
Revised 14 November 2011
Accepted 15 November 2011
Available online 25 November 2011
Keywords:
Dipyrone
Analgesia
Cyclooxygenase
Arachidonic acid amides
Cannabinoid receptors
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
4-aminoantipyrine (3) and 4-formylaminoantipyrine (4), which is
an end-metabolite.6
Dipyrone (1; metamizole sodium; chemical name, sodium
N -(2,3-dimethyl-5-oxo-1-phenyl-3-pyrazolin-4-yl)-N-methylami-
nomethanesulphonate, usually as monohydrate) was first mar-
keted in Germany in 1922.1,2 The low toxicity of dipyrone and its
efficacy support its use in clinical practice. It has been banned in
some countries on account of its potential to cause blood dyscra-
sias, however, reported estimates of the risk of agranulocytosis
vary by several orders of magnitude and may depend on dose,
duration and concomitant medication.3,4
Dipyrone is detectable in the serum for only about 15 min
following intravenous administration, and is not detectable after
oral intake.5,6 The pharmacokinetics of dipyrone (Fig. 1) are charac-
terized by a rapid hydrolytic retro-Mannich reaction to 4-methyla-
minoantipyrine (2), which has 85% bioavailability after oral
administration, and takes a short time to achieve maximal systemic
concentrations (tmax of 1.2–2.0 h). Compound 2 is further metabo-
lised, with an average elimination half-life (t1/2) of 2.6–3.5 h, to
The molecular mechanism of analgesic and antipyretic action of
dipyrone has been under debate for a long time. Both central7,8 and
peripheral inhibition of prostaglandin biosynthesis is presumed to
be at the basis of dipyrone action.9,10 The onset and duration of the
analgesic effects correlate with saliva concentrations of 2 and 3.11
Importantly, 2 was found to inhibit both cyclooxygenase (COX) iso-
forms, COX-1 and COX-2, at concentrations that were found in
plasma after standard doses of dipyrone,12 probably by sequester-
ing radicals that are necessary to initiate the catalytic activity of
the COX enzymes.13 Although the lack of acidity was proposed to
be responsible for the favorable gastrointestinal tolerability of
dipyrone over acidic antiinflammatory drugs,12 the mechanism of
action is still a matter of debate.14
We hypothesize that either unidentified alternative targets of 2,
3, and/or 4, or unidentified active metabolites contribute to the
activity of dipyrone. Considering the pharmacokinetics, high
dosage, and rather short duration of activity, unidentified active
metabolites should be closely related to 2. We reasoned that cen-
tral inhibition of COX-1 and COX-2 and activation of cannabinoid
receptors of type 1 and 2 (CB1 and CB2), two new targets for the
development of new analgesic drugs,15 might account for much
of the clinical profile of dipyrone. Recently, Vazquez-Rodriguez
et al. provided evidence that the antinociceptive effects of dipyrone
⇑
Corresponding author. Tel.: +49 345 5525175; fax: +49 345 5527027.
E-mail address: peter.imming@pharmazie.uni-halle.de (P. Imming).
Present address: Institute of Medical Sciences, University of Aberdeen, Aberdeen,
UK.
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.11.028

102
T. Rogosch et al. / Bioorg. Med. Chem. 20 (2012) 101–107
Figure 1. Metabolites of dipyrone (1).
following microinjection into ventrolateral periaqueductal gray
matter16 were reversed by the CB1 antagonist AM251.17 Both 2
and 3 easily cross the blood–brain barrier, reaching concentrations
The product ions of greatest intensity (m/z 218.2 (5) and m/z 204.2
(6)) resulted from loss of arachidonic acid, as were clearly seen in
tandem mass spectroscopy. Two other different ions were regis-
tered isochronically for 6 ((M)+ m/z 490.4, (MÀH2O)+ m/z 472.4)
and one for 5 ((M)+ m/z 504.5) (data not shown) and were found
to be identical in retention time and relative intensities for both
the standard and the sample. The amount of 5 and 6 detectable
was directly associated with the feeding duration of dipyrone, and
they were not present in negative controls.
of 5–30 l l
g mlÀ1 (2) and 1–7 g mlÀ1 (3) in the cerebrospinal
fluid.18 However, these compounds are very weak inhibitors of
both COX isoenzymes with an IC50 of approximately 10À3 M.19 An-
other study reported comparatively low activity against purified
COX-1 and COX-2, or COX-1 in intact bovine aortic endothelial cells
and in human platelets, but higher activity against COX-2 in mur-
ine macrophages and primary human leukocytes, viz. approxi-
We investigated the involvement of fatty acid amide hydrolase
(FAAH), an enzyme involved in both the hydrolysis and synthesis
of some long chain fatty acid amides23,24 in the formation of 5
and 6. For this purpose, dipyrone was fed to wild-type and FAAH
mately 12 and 21 l . However, the authors stated that
g mlÀ1 20
‘the term metamizol (to) encompasses the entire set of pyrazolone
derivatives’ (dipyrone, 2, 3, 4), without specifying if they added
dipyrone or a mixture of the pyrazolones to the various test sys-
tems. The different pyrazolones have differential effects and fate
when added to systems with active enzymes. Since not even the
relative ratio of the pyrazolones was reported, the results are diffi-
cult to interpret and compare with other findings.
Both 2 and 3 are weakly basic amines. We therefore assumed
that they might be acylated within the CNS, leading to fatty acid
amides. A related mechanism of action was proposed for acetami-
nophen (paracetamol).21,22 The amide that could most likely exhi-
bit COX inhibition is the arachidonoyl amide of 2 (5). As standards,
we prepared the arachidonoyl amides 5 and 6 from 2 and 3 and
arachidonoyl chloride. We then examined whether they are indeed
produced in vivo from dipyrone and tested their activity in a vari-
ety of assays.
À/À
mice25 for 5 days. The arachidonoyl amides 5 and 6 were
detectable in substantial amounts in wild-type mice (FAAH+/+
)
after 2 days of the application of dipyrone, and in slightly higher
amounts after 5 days of application. Compared to wild-type mice,
less than 10% of the amides were formed in FAAHÀ/À mice. In con-
trols (0 days) without addition of dipyrone to the drinking water,
no substantial amount of 5 and 6 was observed (Fig. 3).
In the following set of experiments we characterized 5 and 6
in vitro. Derivatives and mimics of arachidonic acid are central
metabolites in the COX and endocannabinoid (CB) systems,15 so
we studied the inhibitory effect on COX. Both compounds inhibited
the formation of COX products. Following addition of exogenous
arachidonic acid to the brain homogenates, amide 5 inhibited the
formation of PGE2 with an IC50 of 13.2
lM (Fig. 4A) and the forma-
tion of 6-keto-PGF1 , the stable hydrolysis product of PGI2, with an
a
IC50 of 5.2
mation by brain homogenates over a concentration range of
100 nM to 100 M (Fig. 5).
Therefore we determined the inhibitory potency of 5 on COX-1
and on COX-2 isolated enzyme activity, also using isolated enzyme
preparations. As shown in Figure 6, both enzymes were blocked
lM (Fig. 4B). In contrast, 6 exerted no effect on PGE2 for-
2. Results
l
Wild-type mice fed dipyrone via drinking water possessed
detectable levels of the arachidonoyl amides 5 and 6 in brain and
spinal cord (SC) two days after the application of dipyrone, and
slightly higher amounts after five days (Fig. 2). The following data
and considerations prove 5 and 6 to be the arachidonoyl amides of
2 and 3. In HPLC experiments with different eluents, retention times
of the standards and material from animal tissue were identical. The
LC–MS peak pattern of standards and samples was also identical.
with similar potency. For COX-1, we found an IC50 of 42
(Fig. 6A), and for COX-2 inhibition with an IC50 of 69
l
l
M
M
(Fig. 6B). The precursor of 5, the amine 2, displayed weaker
inhibition at an IC50 of 115 M for COX-1 and 307 M for COX-2
l
l

T. Rogosch et al. / Bioorg. Med. Chem. 20 (2012) 101–107
103
Figure 2. Concentrations of 5 (A) and 6 (B) in the brain (denoted CNS) and the spinal cord (SC) of C57BL/6 wild-type mice (n = 3 mice/group, mean SEM).
Figure 3. Concentrations of 5 (A) and 6 (B) in the brain of wild-type mice (FAAH+/+) and of FAAH knockout mice (FAAHÀ/À) after administration of dipyrone or water for five
days (n = 3 mice/group, mean SEM).
Figure 4. Inhibition of PGE2 formation in brain homogenates in the presence j and absence h of the FAAH inhibitor, 4-benzyloxyphenyl-n-butylcarbamate (URB532) (A) and
inhibition of 6-keto-PGF1 formation in brain homogenates (B) by 5 (n = 4, mean SEM).
a
(Fig. 7). Furthermore, inhibition of PGE2 formation in brain homog-
enates of LPS stimulated COX-1À/À and COX-2À/À mice by 5 was as-
sessed. We did not observe any formation of PGE2 in brain tissue
4-benzyloxyphenyl-n-butylcarbamate (URB532).26 The FAAH
inhibitor shifted the dose–response curve for inhibition of PGE2
formation to the left to yield a calculated IC50 of 58 nM (Fig. 4 A),
thus suggesting that 5 might also be a FAAH substrate. This has
implications for the analgesic activity of 1—see Section 3 below.
Since the amide of arachidonic acid N-arachidonoylethanola-
mide (anandamide) binds to cannabinoid CB1 and CB2 receptors,15
we tested the two compounds 5 and 6 on human recombinant CB1
and CB2 in specific binding assays. Compound 5 exhibited affinity
from COX-1À/À mice with 0.8
l
M 5 (IC50, 355 nM), while formation
of PGE2 in brain tissue from COX-2À/À mice was reduced to 30%
with 0.4 M 5 (IC50 107 nM) (Fig. 8). This can be interpreted by
l
assuming suppression of residual PGE2 formation by COX-1 in
COX-2À/À and of COX-2 in COXÀ/À mice by 5.
The inhibitory potency of 5 was further analyzed in mouse brain
homogenates in the presence or absence of the FAAH inhibitor
for both CB1 (Ki = 7.8 0.8 lM) and CB2 (Ki = 3.0 0.4 lM)

104
T. Rogosch et al. / Bioorg. Med. Chem. 20 (2012) 101–107
Figure 5. Inhibition of PGE2 formation in brain homogenate by
mean SEM).
6 (n = 4,
Figure 7. Inhibition of COX-1 j and COX-2 h (isolated enzyme) by 2 (n = 4,
mean SEM).
receptors. Similarly, compound 6 exhibited affinity for both CB1
(Ki = 2.9 0.3
SEM, N = 3).
lM) and CB2 (Ki = 5.4 0.6 lM) receptors (means
3. Discussion
Our findings suggest that the analgesic effect of dipyrone may
be partly mediated by a dual mechanism of action: The inhibition
of COX enzyme activity and the stimulation of CB receptors. Prere-
quisite for this combinatory action is the acylation of its primary
metabolite 2 with arachidonic acid. Recently a similar analgesic
mechanism has been shown for paracetamol (acetaminophen),
which partly acts after metabolic conversion to N-arachidonoylam-
inophenol (AM 404) through a combination of COX inhibition and
stimulation of CB1 receptors and transient receptor potential vanil-
loid type-1 (TRPV1) channels.21,27 Antinociception was observed
after intracerebroventricular injection of 10 nmol of AM404. Pres-
ently both agonistic and antagonistic effects at TRPV1 are supposed
to lead to antinociception.31 Possibly similar to the multiple mech-
anism of action of paracetamol, we herein report evidence that
dipyrone is similarly activated metabolically by acylation of its pri-
mary metabolite 2 with arachidonic acid to give 5. With an IC50 at
TRPV1 receptors of 3000 nM, as reported by us,32 5 may well con-
tribute to the analgesic activity of dipyrone. For the TRPV1 agonist
AM404, we found an EC50 of 26 nM.33 The N-demethylated ara-
chidonate, 6, had no activity at TRPV1.
In a recent paper on the different pharmacological actions of
paracetamol and dipyrone, the possibility of formation of 5 was ru-
led out because 4-methylaminoantipyrine was stated to be a ter-
tiary amine.14 In fact, it is a secondary amine, and we were able
to acylate it efficiently with arachidonic acid. Our data show that
the resulting amide 5 and the related amide 6 are indeed formed
in vivo after administration of dipyrone and are thus metabolites
of this drug substance. The arachidonoyl amides 5 and 6 bound
to both human recombinant cannabinoid CB1 and CB2 receptors
Figure 8. Inhibition of PGE2 formation by 5 in brain homogenates of LPS stimulated
COX1À/À and COX2À/À mice (n = 2, mean SEM).
at micromolar concentrations, and 5 concentration-dependently
inhibited both isolated COX isoforms as well as PGE2 and PGI2
and LPS-induced PGE2 formation in mouse brain homogenates at
nano- or micromolar concentrations that are attainable at normal
dosages of dipyrone, which are rather high (500–1000 mg single
dose).
Because FAAH, the enzyme responsible for hydrolysis of several
fatty acid amides, may also act in the reverse direction and catalyze
the condensation between long chain fatty acids and amines, as
shown for AM404,21 we examined whether this enzyme is involved
in the formation and degradation of 5 and 6. Indeed, FAAH seems to
be responsible for the formation of these metabolites because less
than 2% formation of 5 and about 6% formation of 6 was found in
brain tissue from FAAHÀ/À mice as compared to wild type mice. Fur-
thermore, we found that the FAAH inhibitor URB532 shifted the
dose–response curve of 5 for inhibition of PGE2 formation in mouse
Figure 6. Inhibition of COX-1 (isolated enzyme) (A) and COX-2 (isolated enzyme) (B) by 5 (n = 4, mean SEM).

T. Rogosch et al. / Bioorg. Med. Chem. 20 (2012) 101–107
105
brain homogenates to the left resulting in a more than 200-fold low-
er IC50 value. This finding supported our hypothesis that 5 was grad-
ually hydrolyzed in the brain homogenates in which FAAH is very
abundant. Consequently, the lower IC50 value calculated in the pres-
ence of the FAAH inhibitor reflects the fact that 5 can not only be
produced but also hydrolyzed by FAAH. Since FAAH activity is high-
est in the CNS and liver, we hypothesize that an additive part of the
analgesic and antipyretic activity of dipyrone is due to a conversion
to its respective arachidonoyl amides in the CNS. This would also ex-
plain its very weak peripheral antiinflammatory activity.
115.4, 36.0, 35.7, 32.8, 31.6, 29.4, 27.3, 26.8, 25.7, 25.0, 22.6,
14.1, 10.6. C32H45N3O2 (503.77): calcd, C, 76.29; H, 9.00; N, 8.34,
found: C, 76.41; H, 8.89; N, 8.46.
Arachidonoyl-4-aminoantipyrin (6; (5Z,8Z,11Z,14Z)-Icosa-5,8,11,
14-tetraenoic acid (1,5-dimethyl-3-oxo-2-phenyl-2,3-dihydro-1H-
pyrazol-4-yl)-amide). Arachidonic acid (0.200 g, 0.66 mmol) and
dimethylformamide p.a. (48 mg, 0.66 mmol) were dissolved in ben-
zene p.a. (5 ml). The mixture wascooled to 0 °C, two equivalents of oxa-
lyl chloride (0.168 g, 2.0 mmol) were added dropwise, and stirring was
continued for 1 h at 0 °C. After dilution with tetrahydrofuran p.a. (THF,
5 ml), aminoantipyrine (3) (0.671 g, 3.3 mmol) was added in 5 ml THF
solution. The mixture was stirred for another 15 min, diluted with
dichloromethane, and washed with aqueous 10% HCl, 1 M NaOH and
water. The organic phase was dried over anhydrous magnesium sulfate
and evaporated to dryness. The residue was purified by flash chroma-
4. Conclusions
The CB receptor agonist,
D
9-tetrahydrocannabinol, is known to
have spasmolytic activity.28 With this in mind, it is intriguing to
speculate that the spasmolytic effect of dipyrone perhaps is caused
by formation of 5 and 6 in other tissues with resulting stimulation
of CB receptors. Further studies will shed light on this aspect and
on the details of the tissue distribution of the active metabolites
5 and 6 that we identified in the course of this study. We believe
to have identified that dipyrone acts as prodrug for two substances
that potentially elicit analgesic effects through the endocannabi-
noid system. With both paracetamol and dipyrone being activated
through conversion to arachidonoyl amides, it seems likely that
other drug substances might be converted to fatty acid derivatives
that are responsible for at least part of their pharmacological activ-
ity. Moreover, knowledge of this dual mechanism of analgesia by
arachidonoyl amides would allow deliberate development of pain
killers with higher activity than dipyrone and paracetamol, allow-
ing for lower dosages.
tography on normal phase silica gel (40–63 lm), eluting with CHCl3/
EtOH (9 + 1). Yield, 0.211 g (65%) of a slightly yellow oil. EI-MS
(30 eV), m/z (%) = 204 (100), 472 (48.7), 490 (M+) (45.2). 1H NMR
(400 MHz, CDCl3), d: 8.26 (br s, 1H, CO–NH), 7.45–7.25 (m, 5H,
5 Â Ar-H), 5.42–5.27 (m, 8H, 4 Â CH@CH), 3.04 (s, 3H, N–CH3),
2.84–2.76 (m, 6H, 3 Â CH@CH–CH2–CH@CH), 2.29 (t, 3J = 7.5 Hz, 2H,
CH2–CH2–CO–NH), 2.20 (s, 3H, C@C–CH3), 2.11–2.00 (m, 4H, 2 Â
CH@CH–CH2), 1.70 (tt, 3J = 3 J = 7.5 Hz, 2H, CH2–CH2–CO–NH), 1.38–
1.22 (m, 6H, CH2–CH2–CH2–CH3), 0.87 (t, 3J = 7.1 Hz, 3H, . ..–CH3);
13C NMR (100 MHz, CDCl3), d: 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, 14.1, 12.5.
C31H43N3O2 (489.76): calcd, C, 76.02; H, 8.84; N, 8.58, found: C,
76.05; H, 9.03; N, 8.43.
5.2. Assays
5. Experimental section
5.1. Syntheses
5.2.1. Feeding experiments
C57BL/6 mice were purchased from Charles River, Sulzfeld,
Germany, housed under controlled conditions, and fed a standard
chow diet. FAAH knock out mice were kindly provided by Beat Lutz
(University of Mainz, Germany). COX-1 and COX-2 knockout mice
were breeded as described by us.29 The study was performed with
permission of the regional animal welfare committee in Giessen
(RP Giessen; Germany). Mice were housed under controlled condi-
tions with standardized air conditioning at 20–22 °C, 50–57% rela-
tive humidity, and a 12-h artificial day/night rhythm. They were
given standard diet and water ad libitum.
Arachidonoyl-4-methylaminoantipyrin (5; (5Z,8Z,11Z,14Z)-Ico-
sa-5,8,11,14-tetraenoic acid (1,5-dimethyl-3-oxo-2-phenyl-2,3-
dihydro-1H-pyrazol-4-yl)-methyl-amide).
Arachidonic
acid
(0.200 g, 0.66 mmol) and dimethylformamide p.a. (48 mg,
0.66 mmol) were dissolved in benzene p.a. (5 ml). The mixture
was cooled to 0 °C, 2 equiv of oxalyl chloride (0.168 g, 2.0 mmol)
were added dropwise, and stirring was continued for 1 h at 0 °C.
After dilution with tetrahydrofurane p.a. (THF, 5 ml), methylami-
noantipyrine (2) (0.717 g, 3.3 mmol) was added in 5 ml THF solu-
tion. The mixture was stirred for another 15 min, diluted with
dichloromethane, and washed with aqueous 10% HCl, 1 M NaOH
and water. The organic phase was dried over anhydrous magne-
sium sulfate and evaporated to dryness. The residue was purified
Dipyrone (from ampule, containing dipyrone-sodium monohy-
drate) was given to the drinking water. Drinking volume was mon-
itored daily and addition of 1 to the drinking water was adjusted
accordingly to result in an approximate daily dose of 600 mg kg
À1
bodyweight. The animals were sacrificed on day 2 or day 5
resulting in time of drug administration of 48 h and 120 h, respec-
tively. Mice were anesthetized with isoflurane, brains quickly re-
moved and chilled in ice-cold H2O. Vertebral column was
dissected and a median laminectomy under a dissecting micro-
scope exposed the spinal cord. The cord was transected at the med-
ullary-spinal junction (C1), gently removed in toto from the
vertebral column, and chilled in ice-cold H2O. Tissue samples were
immediately homogenized and extracted with diisopropyl ether/
ethyl acetate (1:3, v/v). Following centrifugation, the solvents were
evaporated and acetonitrile/water (9:1, v/v) was added. An aliquot
of this solution was analyzed by LC–MS/MS. Different amounts of
synthetic 5 and 6 were used as external calibration samples.
by flash chromatography on normal phase silica gel (40–63 lm),
eluting with CHCl3/EtOH (9+1). Yield, 0.223 g (67%) of a slightly
greenish oil. EI-MS (30 eV), m/z (%) = 218 (100), 504 (M+) (9.5).
EI-MS (60 eV), m/z (%) = 56 (100), 97 (48.7), 159 (19.1), 218
(11.8), 187 (9.0). 1H NMR (400 MHz, CDCl3), d: 7.48–7.28 (m, 5H,
5 Â Ar-H), 5.41–5.27 (m, 8H, 4 Â CH@CH), 3.29 (s, N–CH3, peak of
rotational isomer of the product), 3.13 (s, 3H, N–CH3), 3.11 (s,
3H, CO–N–CH3), 3.07 (s, CO–N–CH3, peak of rotational isomer of
the product), 2.82–2.73 (m, 6H, 3 Â CH@CH–CH2–CH@CH), 2.27
(dt, 2J = 15.0 Hz, 3J = 7.5 Hz, 1H, CH2–CH2–CO–N–CH3), 2.18 (s,
3H, C@C–CH3), 2.12 (dt, 2J = 15.0 Hz, 3J = 7.5 Hz, 1H, CH2–CH2–
CO–N–CH3), 2.10 (s, C@C–CH3, peak of rotational isomer of the
product), 2.07–2.00 (m, 4H, 2 Â CH@CH–CH2), 1.67 (tt,
3J = 3J = 7.5 Hz, 2H, CH2–CH2–CO–N–CH3), 1.36–1.23 (m, 6H,
CH2–CH2–CH2–CH3), 0.87 (t, 3J = 6.8 Hz, 3H, . . .–CH3); 13C NMR
(100 MHz, CDCl3), d: 174.3, 161.7, 151.7, 134.6, 130.5, 129.6,
129.3, 129.1, 128.6, 128.3, 128.1, 127.9, 127.6, 127.1, 124.1,
5.2.2. Determination of 5 and 6 by ESI-LC–MS/MS
Ethyl acetate was purchased from Promochem (Wesel,
Germany), acetic acid was obtained from Aldrich (Taufkirchen,
Germany), diisopropyl ether and acetonitrile from Merck (Darms-
tadt, Germany).

106
T. Rogosch et al. / Bioorg. Med. Chem. 20 (2012) 101–107
Figure 9. Representative chromatograms of 100 pg/sample extracted 5 and 6 standards (A), and mouse brain cell homogenates (B). Upper trace, 2.4 pg/sample 5 (m/z 504.6/
218.2); middle trace, 6 (m/z 490.4/472.4) and lower trace, 6 (m/z 490.4/204.2) with 39 pg/sample.
The HPLC system consisted of a Perkin–Elmer series 200 pump,
a CTC HTS PAL autosampler (CTC Analytics, Zingen, Switzerland)
and an Applied Biosystems API 3000 triple-quad tandem mass
spectrometer (Applied Biosystems, Thornhill, Canada) equipped
with a turbo ion interface.
The sample was sonicated and then centrifuged. A 10 ll aliquot
of this solution was injected in the LC–MS/MS. Semi-quantification
was achieved with external calibration samples (10, 100 and
1000 pg) (Fig. 9).
The isocratic mobile phase consisted of 60% acetonitrile in water
5.2.3. Inhibition of COX activity in tissue extract
with 0.2% acetic acid and was pumped at a flow-rate of 0.2 ml minÀ1
.
Brain tissue homogenates from C57BL/6 mice were prepared on
ice in tissue buffer (50 mM Tris pH 7.4 buffer containing 1 mM
phenol and 1x protease inhibitors (Calbiochem)). For the determi-
Chromatography was performed on a Polaris (Varian Deutschland
GmbH, Darmstadt, Germany) C18-A 3
l
m; 50 Â 2 mm narrow bore
analytical column. The total runtime was 25 min with 5 eluting at
nation of COX activity, 20
tissue buffer containing different amounts of the inhibitors or vehi-
cle. The reaction was started by the addition of 150 M arachidonic
lg of brain homogenate was diluted in
16.5 min and 6 at 13.2 min.
Detection by tandem mass spectrometry was based on precur-
sor ion transitions to the strongest intensity product ions. Instru-
mental conditions were optimized to afford best sensitivity
(dwell time 400 ms, DP 50 V, FP 160 V, EP 9 V). For positive ioniza-
tion mode, the first quadrupole, Q1, was set to monitor the proton-
ated molecules (M+H)+ at m/z 504.6 for 5, and m/z 490.4 for 6 with
collision-induced fragmentation at Q2 (collision gas nitrogen,
collision energy 30 eV), and monitoring the product ions via Q3
at m/z 218.2 for 5 and m/z 472.4 for the first measured ion and
m/z 204.2 for the second product ion of 6 (data not shown).
l
acid at 37 °C. After 5 min, the reaction was stopped by acidification
to pH 2.5 with formic acid (1%) and immediately transferred onto
ice. Prostanoids were extracted with ethyl acetate/hexane (7:1, v/
v). PGE2 and 6-keto-PGF1 were determined by GC–MS/MS.30 In or-
a
der to analyze inhibition of LPS-induced COX activity, we injected
LPS ip (Sigma) at 50 l
g kgÀ1 body weight. After 6 h, brain tissue
homogenates were prepared, and COX activity was determined in
the presence or absence of inhibitors.
Cell homogenates (200
External calibration samples were generated with 0, 10, 100 and
1000 pg of 5 and 6. The analytes were extracted with 1000 l diiso-
propyl ether/ethyl acetate (1:3, v/v, recovery about 90%). The sam-
ple was centrifuged at 1000g for 10 min. The solvent was
l
l) were diluted with 500
l
l water.
5.2.4. Inhibition of COX activity in isolated enzymes
Isolated COX-1 and COX-2 enzymes (Cayman) were diluted in
l
tissue buffer at a concentration of 0.53 U l À1
l
for COX-1 and
0.51 U l À1
The reaction was started by the addition of 50
l
for COX-2 in the presence of the inhibitors or vehicle.
l
M arachidonic acid
evaporated and 100
ll acetonitrile/water (9:1, v/v) were added.
at 37 °C. After 5 min, the reaction was stopped by acidification to

T. Rogosch et al. / Bioorg. Med. Chem. 20 (2012) 101–107
107
8. Tortorici, V.; Vanegas, H. Eur. J. NeuroSci. 1995, 7, 1857.
pH 2.5 with formic acid (1%). The samples were kept at 4 °C. Pros-
9. Abbate, R.; Gori, A. M.; Pinto, S.; Attanasio, M.; Paniccia, R.; Coppo, M.;
Castellani, S.; Giusti, B.; Boddi, M.; Neri Serneri, G. G. Prostaglandins Leukot.
Essent. Fatty Acids 1990, 41, 89.
tanoids were extracted with ethyl acetate/hexane (7:1, v/v). PGE2
and 6-keto-PGF1 were determined by GC–MS/MS.30
a
10. Shimada, S. G.; Otterness, I. G.; Stitt, J. T. Agents Actions 1994, 41, 188.
11. Rohdewald, P.; Drehsen, G.; Milsmann, E.; Derendorf, H. Arzneimittelforschung
1983, 33, 985.
12. Hinz, B.; Cheremina, O.; Bachmakov, J.; Renner, B.; Zolk, O.; Fromm, M. F.;
Brune, K. FASEB J. 2007, 21, 2343.
13. Pierre, S. C.; Schmidt, R.; Brenneis, C.; Michaelis, M.; Geisslinger, G.; Scholich, K.
Br. J. Pharmacol. 2007, 151, 494.
14. Rezende, R. M.; Franca, D. S.; Menezes, G. B.; dos Reis, W. G.; Bakhle, Y. S.;
Francischi, J. N. Br. J. Pharmacol. 2008, 153, 760.
15. Di Marzo, V.; Bifulco, M.; De Petrocellis, L. Nat. Rev. Drug Disc. 2004, 3, 771.
16. Vazquez-Rodriguez, E. E. W.; Ramirez, K.; Avila, C.; Vanegas, H. In 12th World
Congress on Pain, Glasgow, 2008.
17. Gatley, S. J.; Gifford, A. N.; Volkow, N. D.; Lan, R.; Makriyannis, A. Eur. J.
Pharmacol. 1996, 307, 331.
18. Cohen, O.; Zylber-Katz, E.; Caraco, Y.; Granit, L.; Levy, M. Eur. J. Clin. Pharmacol.
1998, 54, 549.
19. Weithmann, K. U.; Alpermann, H. G. Arzneimittelforschung 1985, 35, 947.
20. Campos, C.; de Gregorio, R.; Garcia-Nieto, R.; Gago, F.; Ortiz, P.; Alemany, S. Eur.
J. Pharmacol. 1999, 378, 339.
5.2.5. Cannabinoid receptor binding assays
For CB1 and CB2 receptor binding assays, the new compounds
were tested by using P2 membranes from HEK cells transfected
with either the human recombinant CB1 or CB2 receptor and
(3H)-(À)-cis-3-(2-hydroxy-4-(1,1-dimethylheptyl)-phenyl)-trans-
4-(3-hydroxy-propyl)-cyclohexanol ((3H)CP-55,940) (Kd = 0.27
nM) as the high affinity ligand, as described by the manufacturer
(Perkin Elmer, Italy). Displacement curves were generated by incu-
bating drugs with 0.27 nM of (3H)CP-55,940. In all cases, Ki values
were calculated by applying the Cheng–Prusoff equation to the IC50
values (obtained by GraphPad) for the displacement of the bound
radioligand by increasing concentrations of the test compounds.
Acknowledgments
21. Mallet, C.; Barrière, D. A.; Ermund, A.; Jönsson, B. A.; Eschalier, A.; Zygmunt, P.
M.; Högestätt, E. D. PLoS One 2010, 5, e12748.
22. Högestätt, E. D.; Jonsson, B. A.; Ermund, A.; Andersson, D. A.; Bjork, H.;
Alexander, J. P.; Cravatt, B. F.; Basbaum, A. I.; Zygmunt, P. M. J. Biol. Chem. 2005,
280, 31405.
23. Cravatt, B. F.; Giang, D. K.; Mayfield, S. P.; Boger, D. L.; Lerner, R. A.; Gilula, N. B.
Nature 1996, 384, 83.
The authors wish to thank Prof. Dr. Beat Lutz, Department of
Physiological Chemistry, Johannes-Gutenberg-Universität Mainz,
Germany, and Prof. Dr. Benjamin F. Cravatt, Department of Cell Biol-
ogy, The Scripps Research Institute, La Jolla, California, for providing
the FAAH knockout mice. The LC/MS studies were supported by DFG
grant SE 263/17-1. AHL acknowledges support by National Institutes
of Health Grants P01DA009789, P01DA017259, F31DA026279 and
T32DA007027. We thank Sanofi-Aventis Deutschland GmbH, Berlin,
Germany, for a gift of 2 and 3.
24. Kurahashi, Y.; Ueda, N.; Suzuki, H.; Suzuki, M.; Yamamoto, S. Biochem. Biophys.
Res. Commun. 1997, 237, 512.
25. Cravatt, B. F.; Demarest, K.; Patricelli, M. P.; Bracey, M. H.; Giang, D. K.; Martin,
B. R.; Lichtman, A. H. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 9371.
26. Kathuria, S.; Gaetani, S.; Fegley, D.; Valino, F.; Duranti, A.; Tontini, A.; Mor, M.;
Tarzia, G.; La Rana, G.; Calignano, A.; Giustino, A.; Tattoli, M.; Palmery, M.;
Cuomo, V.; Piomelli, D. Nat. Med. 2003, 9, 76.
27. Ottani, A.; Leone, S.; Sandrini, M.; Ferrari, A.; Bertolini, A. Eur. J. Pharmacol.
2006, 531, 280.
28. Carlini, E. A. Toxicon 2004, 44, 461.
References
29. Popp, L.; Haussler, A.; Olliges, A.; Nusing, R.; Narumiya, S.; Geisslinger, G.;
Tegeder, I. Eur. J. Pain 2009, 13, 691.
30. Schweer, H.; Watzer, B.; Seyberth, H. W. J. Chromatogr. 1994, 652, 221.
31. Maione, S.; De Petrocellis, L.; de Novellis, V.; Moriello, A. S.; Petrosino, S.;
Palazzo, E.; Rossi, F. S.; Woodward, D. F.; Di Marzo, V. Br. J. Pharmacol. 2007,
150, 766.
32. Sinning, C.; Watzer, B.; De Petrocellis, L.; Di Marzo, V.; Imming, P.
ChemMedChem 2008, 3, 1956.
33. De Petrocellis, L.; Bisogno, T.; Davis, J. B.; Pertwee, R. G.; Di Marzo, V. FEBS Lett.
2000, 13, 52–56.
1. Fendrich, Z. Cas. Lek. Cesk. 2000, 139, 440.
2. Edwards, J. E.; Meseguer, F.; Faura, C. C.; Moore, R. A.; McQuay, H. J. Cochrane
Database Syst. Rev. 2001. CD003227.
3. Maj, S.; Centkowski, P. Med. Sci. Monit. 2004, 10, PI93.
4. Ibanez, L.; Vidal, X.; Ballarin, E.; Laporte, J. R. Eur. J. Clin. Pharmacol. 2005, 60,
821.
5. Vlahov, V.; Badian, M.; Verho, M.; Bacracheva, N. Eur. J. Clin. Pharmacol. 1990,
38, 61.
6. Levy, M.; Zylber-Katz, E.; Rosenkranz, B. Clin. Pharmacokinet. 1995, 28, 216.
7. Carlsson, K. H.; Helmreich, J.; Jurna, I. Pain 1986, 27, 373.