Arylpyrrolizines as inhibitors of microsomal prostaglandin E2 synthase-1 (mPGES-1) or as dual inhibitors of mPGES-1 and 5-lipoxygenase (5-LOX)

Article abstract of DOI:10.1021/jm900481c

We synthesized and evaluated inhibitors for the microsomal prostaglandin E2 synthase-1 (mPGES-1), based on the arylpyrrolizine scaffold. In a cell free mPGES-1 assay several "sulfonimides" exceeded our leadML3000 (3) in potency. The most promising compound, the tolylsulfonimide 11f, revealed an IC50 of 2.1 μM and is equipotent to the literature reference molecule MK886 (1). Selected compounds also potently reduced 5-LOX product formation in intact cells. Inhibition of isolated COX was occasionally remarkably cut down.

Full text of DOI:10.1021/jm900481c

4968 J. Med. Chem. 2009, 52, 4968–4972
DOI: 10.1021/jm900481c
Arylpyrrolizines as Inhibitors of Microsomal Prostaglandin E2 Synthase-1 (mPGES-1) or as Dual
Inhibitors of mPGES-1 and 5-Lipoxygenase (5-LOX)
Andy J. Liedtke,*,† Peter R. W. E. F. Keck,† Frank Lehmann,† Andreas Koeberle,‡ Oliver Werz,‡ and Stefan A. Laufer*,†
†Department of Pharmaceutical and Medicinal Chemistry and ‡Department of Pharmaceutical Analytics, Institute of Pharmacy,
Eberhard Karls University Tuebingen, Auf der Morgenstelle 8, 72076 Tuebingen, Germany
Received May 4, 2009
We synthesized and evaluated inhibitors for the microsomal prostaglandin E2 synthase-1 (mPGES-1),
based on the arylpyrrolizine scaffold. In a cell free mPGES-1 assay several “sulfonimides” exceeded
our lead ML3000 (3) in potency. The most promising compound, the tolylsulfonimide 11f, revealed an
IC50 of 2.1 μM and is equipotent to the literature reference molecule MK886 (1). Selected compounds
also potently reduced 5-LOX product formation in intact cells. Inhibition of isolated COX was
occasionally remarkably cut down.
Introduction
PGE2 from COX-2 derived PGH2. The structure of human
mPGES-1 has very recently been determined in complex with
glutathione (GSH) by electron crystallography from 2D
crystals in the presence of phospholipids.7
Prostaglandin (PGa) E2 is one of the most important and
powerful prostanoids with diverse biological activity.1 As a key
cyclic lipid mediator derived from arachidonic acid (AA), it
is involved in the development and perpetuation of inflam-
mation seen in diseases such as rheumatoid arthritis (RA) and
has been implicated in the development of peripheral and
central sensitization during nociceptive processing (e.g., hy-
peralgesia, allodynia) and in tumorigenesis.2-5 In the eicosa-
noid pathway (AA metabolism; see Supporting Information),
induction of PGE2 biosynthesis during inflammation requires
the enzymatic actions of two cytokine-inducible enzymes:
cyclooxygenase (COX) and prostaglandin E synthase (PGES).
Inhibition of COX results in a reduced synthesis of prosta-
glandins (e.g., PGE2) and thromboxanes (TXA2) and is the
basis for the anti-inflammatory efficacy and probably also for
the analgesic activity of nonsteroidal anti-inflammatory drugs
(NSAIDs). Owing to undesirable adverse effects of COX
inhibitors (NSAIDs) and COX-2-selective inhibitors (COXIBs)
in the gastrointestinal, renal, and cardiovascular systems, the
selective inhibition of PGE2-forming enzymes downstream of
COX, such as the inducible glutathione-dependent microso-
mal PGE2 synthase-1 (mPGES-1), has recently been proposed
as a more promising approach for development of drugs
for anti-inflammatory and pain therapy.3-6 mPGES-1 is a
member of the MAPEG family, including FLAP, and is the
key terminal enzyme in pathology related production of
A few synthetic compounds have been reported to act as
inhibitors of mPGES-1 activity, but none of these inhibitors
showed high activity or specificity of action against mPGES-1
in vivo.6 The potent FLAP indole inhibitor MK-8868 (1)
(Figure 1), an anticancer drug that can more efficiently
bind with at least one other protein (FLAP IC50 =26 nM),
functions as a moderate inhibitor of rat (IC50=3.2 μM) and
human mPGES-1 (IC50=1.6 μM) and has been used as a lead
structure to develop more effective inhibitors. Some of the
indole analogues of 1 have promising higher activity for the
inducible human mPGES-1 membrane protein, with the low-
est IC50 of 3 nM.8 Optimized compounds in this series are
represented by the “biaryl derivatives” exemplified by Merck
Frosst compound 238 (2a) or 308 (2b) and their analogues
(Figure 1). However, the latter were less potent in cellular
systems and lacked inhibitory properties in whole blood
because of strong binding to serum proteins. As of yet, no
effective in vivo inhibitor has been reported in the literature.6
Therefore, the design and discovery of novel mPGES-1
specific inhibitors with different scaffolds are highly desirable.
We considered the similarities between ML3000 (licofelone)9
(3), a well-known dual COX/5-LOX inhibitor (third genera-
tion NSAID, Figure 1), and 1 in terms of structural confor-
mation and mechanism of inhibition of leukotriene synthesis.
Werz and co-workers very recently also addressed the mode of
action of 3 in the suppression of PGE2 formation.1 Their data
indicated that 3 appears to suppress inflammatory PGE2
formation preferentially by inhibiting mPGES-1 at concen-
trations that do not affect COX-2 (mPGES-1 IC50=6 μM),
implying attractive and thus far unique molecular pharmaco-
logical dynamics as an inhibitor of COX-1, 5-LOX, and
mPGES-1. Furthermore, 3 is ∼60-fold more potent in the
cellular assay than in cell-free systems regarding the reduction
of PGE2 formation and thus more than 100-fold more potent
than 1.1
*To whom correspondence should be addressed. For A.J.L.:
telephone, þ49-7071-2975278; fax, þ49-7071-295037; e-mail, andy.
liedtke@uni-tuebingen.de. For S.A.L.: telephone, þ49-7071-2972459;
fax, þ49-7071-295037; e-mail, stefan.laufer@uni-tuebingen.de.
aAbbreviations: COX, cyclooxygenase; CDI, carbonyldiimidazole;
MAPEG, membrane-associated proteins involved in eicosanoid and
glutathione metabolism; FLAP, 5-lipoxygenase activating protein;
mPGES, microsomal prostaglandin E2 synthase; PG, prostaglandin;
NSAID, nonsteroidal anti-inflammatory drug; COXIB, COX-2-selec-
tive inhibitor; prostanoid, a subclass of eicosanoids consisting of
prostaglandins, thromboxanes, and prostacyclins; PMNL, polymor-
phonuclear leukocytes; 12-HHT, 12(S)-hydroxy-5-cis-8,10-trans-hepta-
decatrienoic acid; LTB4, leukotriene B4; LOX, lipoxygenase.
r
pubs.acs.org/jmc
Published on Web 07/14/2009
2009 American Chemical Society

Brief Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 15 4969
Scheme 2. Synthetic Routes to the Target Compoundsa
Figure 1. Structures of anti-inflammatory drugs and literature
mPEGs-1 inhibitors.
Scheme 1. Construction of the Pyrrolizine Scaffolda
a Reagents and conditions: (a) diazoacetic acid ester/Cu0 (add over
1 h), toluene, reflux (110 °C), 15 min, 39%; (b) acrylic acid ester/BF3,
dichloro ethane, room temp, 1 h, 50-70%; (c) KOH, ethanol, reflux,
5 min to 1 h, 50-75%; (d) CH2dNþ(CH3)2Cl-, absolute CH2Cl2;
(e) CH3I, NaCN; (f) NaN3.
Scheme 3. Derivatization of the Acid Functionalities to the
Desired Sulfonylimidesa
a Reagents and conditions: (a) benzylmagnesium chloride (Grignard
species provided in situ from benzylchloride and Mg 1:1), initially
absolute Et2O, 2 h, reflux, then toluene, 3 h, reflux, 23-30%, depending
on the further workup (R3 = H, 89%);11,12 (b) ω-bromoacetophenone,
absolute EtOH, 24 h, room temp, then addition of saturated NaHCO3,
24 h, room temp, 42-57%.
a Reagents and conditions: (a) (1) 10 in absolute THF, CDI, 2 h, room
temp (mixture A); (2) substituted sulfonylamide in absolute THF, argon,
NaH, 1 h, roomtemp (mixture B); (3) combining mixturesA and B under
argon, 40 h, room temp, 39-46%.
In the present study, to improve the activity profile and
at the same time reduce undesirable side effects, we pre-
pared several compounds based on the lead structure of
3 that were modified at the acid group.10,11 After evalua-
tion of existing compounds and their IC50 values against
mPGES-1, both isoforms of the COX enzyme, and 5-LOX,
we devised a concrete derivatization strategy to optimize
the mPGES-1 inhibition potency of these molecules while
at the same time reducing or canceling the COX activity.
Ideally, the 5-LOX inhibitory activity should also be
retained.9
Cyclization and simultaneous incorporation of the first aro-
matic substituent are carried out by Grignard reaction using
benzyl chloride. In a final step both, the obtained 5-ben-
zyl-3,3-dialkyl-3,4-dihydro-2H-pyrrole and the 5-benzyl-3,4-
dihydro-2H-pyrrole (5)10 are ring-closed to the favored
2,3-dihydro-1H-pyrrolizine derivative 7 with (substituted)
ω-bromoacetophenone 6.10 The latter reaction step is carried
out in an aqueous/ethanolic sodium bicarbonate solution at
room temperature (40 h) (Scheme 1). Implementation of the
obligatory acid group can be conducted by two different
methods depending on the preferred alkyl anchor (spacer)
using diazo ethyl acetate and copper in boiling toluol
(affording 8)12 or acryl acid methyl ester under boron
trifluoride etherate catalysis in absolute dichloroethane at
ambient temperature (affording 9).13 Alternatively, at this
stage Friedel-Crafts alkylation may be employed to achieve 9
from8 and 3-bromopropionic acidethylester usingaluminum
chloride as catalyst (Scheme 2).
Chemistry
Arylpyrrolizines are recognized scaffolds for (dual) COX/
LOX inhibitors (e.g., 3, Figure 1), and this compound class
was recently found to comprise putative inhibitors of the
mPGES-1 as well.1 A general synthetic strategy is briefly
summarized in Schemes 1-3.
All bioactive target compounds have a pyrrolizine ring
system in common whose synthesis typically is based on a
differently substituted 4-chlorobutannitrile 411,12 (Scheme 1).

4970 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 15
Liedtke et al.
The free carbonic acids 10a-g10,12-14 (Table 1) arise from
these esters by alkaline hydrolysis with 10% potassium hydro-
xide. Further derivatization of the free alkyl acids to the
desired sulfonylimides 11a-f10,11,15 (Scheme 3) is carried out
by the initial activation of the acid function with CDI in dry
THF (room temp, 2 h) and subsequent reaction of these
intermediates with deprotonated sulfonamides (dry THF,
room temp, 40 h).
Biosteric replacement of a carboxyl group by tetrazole could
also be a useful method to optimize the lead structure (3).
Tetrazole 14 was synthesized by converting 6-(4-chlorophenyl)-
2,2-dimethyl-7-phenyl-2,3-dihydro-1H-pyrrolizine 7 into the
dimethylaminomethyl derivative 1216 using a mixture of for-
maldehyde/dimethylaminohydrochloride. Activation by quar-
ternization (methyl iodide) with sodium cyanide finally
afforded the corresponding nitrile 13, which was then converted
to the tetrazole 14 using sodium azide.
Biological Testing
All compounds were screened in a cell free mPGES-1 assay
at four distinct concentrations between 10 μM and 1 nM,
depending on the estimated intrinsic activity.1 PGE2 synthase
activity was determined in microsomes of A549-cells. Test
results are expressed as a percentage of the PGE2 level
measured in the presence of the vehicle control. Compounds
1 and 2a (Table 2) were used as internal standards. Further-
more, when mPGES-1 IC50 values were e10 μM, the com-
pounds were investigated for their activity on isolated COX-1/
COX-2 and were tested at 30 μM.1 In these cases, data are
presentedasthe percentageof theconcentration of 12-HHT in
the vehicle control. Inhibition of 5-LOX was determined in
PMNL (and 5-LOX) intact cell assay at inhibitor concentra-
tions between 10 μM and 0.1 μM. LTB4 was measured by an
HPLC method.17
Table 1. Substitution Patterns of the Synthesized Compounds
test compd
R1
R2
R3
n
3
COOH
COOH
COOH
COOH
COOH
COOH
COOH
COOH
COOH
4-Cl
4-Cl
Me
H
1
1
2
2
1
1
1
1
2
1
1
1
1
2
2
1
Discussion
10a
10b
10c
10d
10e
10f
10g
10h
11a
11b
11c
11d
11e
11f
14
We recently reported that the arylpyrrolizine carboxylic
acid 3 has good intrinsic activity against mPGES-1 and
5-LOX.1,17 By use of 3 (Figure 1) as a lead structure, struc-
ture-activity relationship (SAR) studies were undertaken
with the aim of identifying more potent and selective inhibi-
tors of mPGES-1 and/or dual inhibitors of mPGES-1 and
5-LOX. The biological data of the tested compounds and of
the standards 1 and 2a are compiled in Table 2.
Substitution of the p-chloro atom of the phenyl ring at the
C-6 position of the central pyrrolizine core by NO2 (10d), CF3
(10e), or tert-butyl (10g) resulted in a significant decrease
in potency against mPGES-1 (IC50>10 μM) compared to 3.
However, the substituent at the C-6 phenyl residue appeared
to be of marked importance for the COX activity. In the
4-Cl
4-Cl
Me
H
4-NO2
4-CF3
benzofuran-6-yla
4-tert-butyl
4-tert-butyl
4-Cl
Me
Me
Me
Me
Me
Me
Et
CO-NH-SO2-CH3
CO-NH-SO2-CH3
CO-NH-SO2-Ph
CO-NH-SO2-Tol
CO-NH-SO2-Ph
CO-NH-SO2-Tol
1H-tetrazol-5-yl
4-Cl
4-Cl
Me
Me
Me
Me
Me
4-Cl
4-Cl
4-Cl
4-Cl
a Instead of the otherwise used phenyl substructure.
Table 2. Results of Biological Assays of mPGES-1 Reference Inhibitors and Arylpyrrolizine Test Compoundsa
COX residual activity at 30 μM [% 12-HHT of
control] ((SEM)
cell free mPGES-1 residual
activity at 10 μM [% PGE2
of control] ((SEM)
cell free mPGES-1b
PMNL 5-LOXc
IC50 [μM]14
compd
IC50 [μM]
COX-1
COX-2
1
2a
22.5 ( 2.3***
11.6 ( 2.9*** at 100 nM
44.9 ( 2.6***
65.3 ( 3.9*
47.4 ( 2.5***
20.8 ( 18.2***
58.8 ( 13.2**
56.1 ( 4.3**
42.1 ( 6.2***
53.8 ( 6.6***
63.1 ( 0.9***
38.5 ( 3.5***
28.7 ( 3.6***
32.2 ( 0.9***
23.2 ( 1.9***
27.9 ( 7.2***
2.11 (lit.: 1.66)
19.5 ( 3.2***
nd
56.7 ( 10.5**
nd
nd
nd
0.18d
1.6d
nd
1.9d
1.5d
0.31d
nd
0.18d
0.23d
nd
0.012 (lit.: 0.0078)
3
6.7
nd
7.2
1.8
nd
nd
6.2
nd
nd
5.9
4.5
4.8
2.1
3.9
12.4 ( 7.6***
nd
53.7 ( 2.4**
nd
10a
10b
10c
10d
10e
10f
10g
11a
11b
11c
11d
11f
14
82.9 ( 13.8
nd
ni
nd
nd
nd
nd
nd
94.0 ( 7.6
ni
nd
75.8 ( 12.8
ni
nd
nd
nd
25.8 ( 10.8***
8.6 ( 7.0***
29.2 ( 5.1***
37.4 ( 3.7***
43.4 ( 7.6**
13.6 ( 2.0***
ni
0.25d
0.26d
nd
46.5 ( 5.1**
0.5d
a n=3-5, (/) p<0.05, (//) p<0.01, or (///) p<0.001 vs vehicle (0.1% DMSO) control, ANOVA þ Tukey HSD post hoc tests. nd=not determined.
ni=no inhibition. b Utilized inhibitor concentrations: 10 μM, 1 μM, 100 nM, 10 nM (1 nM). c Utilized inhibitor concentrations: 10, 1, 3.3, 1 μM.
d Average of at least two determinations.

Brief Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 15 4971
tert-butyl phenyl derivative10g, the inhibitory activity toward
both tested COX isozymes was successfully canceled out at an
inhibitor concentration of 30 μM, whereas the 5-LOX inhibi-
tion remained unaffected (IC50=0.18 μM),14 which is promis-
ing for a possible follow-up investigation toward more
selective mPGES-1 inhibitors.
In contrast, total replacement of the p-chloro phenyl moiety
by benzofuran-6-yl in 10f had only a minimal effect on
potency, showing that such a planar bicyclic moiety is toler-
ated at this position. However, any advances in more selective
mPGES-1 inhibitions were offset by the COX results. Both
COX isoforms were only slightly inhibited by 10f, which still
constitutes a pharmacological benefit compared to 3.
indicating a better selectivity for mPGES-1 than both of the
reference compounds. The data reveal that the propionic acid
derivative 11f also has a better COX inhibition profile (higher
residual COX activity) than its nor-derivative 11d.
To define the minimal requirements for acceptable
mPGES-1 activity, we introduced a small and sterically less
demanding mesylate onto the acetic acid precursor 3 (realized
in 11a). This modification resulted in increased activity of the
isolated mPGES-1 as detected by a 1.4-fold higher PGE2 level
(63.1% at 10 μM compared to control, Table 2). These
findings imply that an optimal pattern may be obtained by a
bulkier terminal aromatic system (R=Ph, Tol). Finally, in
vitro potency could be favorably modulated by variation of
the C-2 substitution position of the arylpyrrolizines. For
example, when the methylsulfonimide analogue 11b is com-
pared to 11a, the observed loss of instrinsic activity could be
successfully compensated by replacement of the two methyl
groups at C-2 by homologous ethyl residues. 11b revealed an
IC50 of 5.9 μM.
As a tetrazole may serve as a biosteric replacement for
carboxylicacids, wealsosynthesized14. Thebiological datain
Table 2 support this concept, with the IC50 of 14 (3.9 μM)
being similar to the those of arylsulfonimides 11c and 11d
(4.5 and 4.8 μM, respectively). Furthermore, the COX inhibi-
tion profile of 14 is comparable to that of 11c, whereas the
COX activity at 30 μM is markedly better suppressed by 11d.
In addition, all tested sulfonimides (11a, 11c, and 11d)
revealed good 5-LOX inhibitory results in the submicromolar
range (IC50 = 0.23-0.26 μM); the tetrazole 14, however,
slightly declines (IC50 = 0.5 mM). Thus, for these molecule
portions the active site of the 5-LOX seem to be relatively
insensitive for sterical factors, as even bulkier residues like
phenyl or tolyl were tolerated. Also, definite fluctuations in
the acidity (a certain acidity is essential as an inevitable
structural property for ligand-protein binding) of the sub-
stituents at the pyrrolizin C-5 position appear to have no
significant influence on the 5-LOX inhibition.
On the basis of the biological data for mPGES-1 of our lead
3 (see above) in a cell-based test matrix,1 we presume that the
potencies of the newly synthesized compounds 10b/f, 11b-f,
and 14 in the whole cell essay should correlate in order
of magnitude with their inhibitory power in the cell free
mPGES-1 test system, because of a presumably moderate or
low degree of protein binding (see Supporting Information,
table of clogP values). Thus, we predict only a small shift in
potency caused by the presence of plasma proteins for these
compounds, even if they revealed an inhibitory activity on
PGE2 in the single-digit micromolar range in the cell free test
assay.
With regard to the future oral bioavailability of these
potential drug candidates, we also intended to maintain an
opportune hydrophilic/lipophilic balance, favoring an ex-
pected absorption or permeation behavior in vivo. The aryl-
pyrrolizine compounds described here, i.e., the sulfimides,
generally appear to better meet with the Lipinski rule-of-five
criteria with respect to the molecular weight, their calculated
logP values, and/or the number of H-bond donor/acceptor
positions (see Supporting Information). These characteristics
are in strong contrast to most of the other published inhibitor
candidates such as the reference 1 and, in particular, its highly
potent but less selective indole analogues.
To optimize chemical stability of the ML3000-like com-
pounds (3) bearing an acetic acid group in position C-5 (see
Chemistry), we attempted to extend the alkyl spacer by one
CH2 unit to obtain the corresponding propionic acid (10b and
10c). This modification had little effect on potency, as the IC50
values were comparable to those of the corresponding shorter
acetic acid derivatives. However, as proposed, the propionic
analogues were chemically more stable and thus better suited
for isolation and, ultimately, for further substitution.12
With respect to the C-2 unsubstituted derivatives 10a and
10c and their mPGES-1 activity in the cell free assay at 10 μM
(Table 2), the inhibition values are inverse compared to the
C-2 dimethyl substituted compounds (3 and 10b). Here, the
free propionic acid 10c (C-2 unsubstituted) tends to be the
more potent derivative. In the case of the C-2 substituted
compounds, the propionic acid 10b is almost comparable to
the acetic acid 3 (as described above). In contrast, the dialkyl
substituent at C-2 appears to be of markedly relevance for the
inhibitory potency against 5-LOX, as can be derived from the
data of 10a (R3=H), which has only a moderate 5-LOX IC50
of 1.6 μM compared to its dihomo analogue (3). These results
once more accent a strong positive relationship between the
lipophilicity of the inhibitor molecule and the 5-LOX activity,
also correlating with the earlier finding that one of the key
structural features of the active site of the 5-lipoxygenase is a
hydrophobic domain.14 As a consequence, all other com-
pounds out of Table 2 having a comparably lipophilic core
to 3 and bearing a C-2 dialkyl residue constantly proved to
have submicromolar 5-LOX IC50 values except for the more
hydrophilic 10d (R2=4-NO2).
Substitution of sulfonimides at the free acid functionality
had a great impact on potency toward mPGES-1. Despite
incorporation of rather lipophilic aryl moieties, the acetic
character of the side chain is conserved, which appears to be
a critical determinant for ligand-enzyme binding.7 The tolyl
sulfonimide 11d is already 1.4-fold more potent than 3.
Comparable data were achieved with the substitution of a
simple phenyl group for the terminal tolyl group in 11c.
Moreover, for both compounds residual activities of COX-1
and COX-2 are relatively low. In each case, the COX-1
isozyme is more affected by the inhibitor than is COX-2,
which is consistent with the findings for the lead 3. Having
recognized the importance of the arylsulfonimide substituent,
we transferred this portion of the molecule to the above-
mentioned more stable propionic acid analogues, here exem-
plified by 11f, and achieved a further 2.3-fold increase in
potency, as measured by IC50 values, compared to 11d.
Accordingly, this compound (11f) is 3.2-fold superior to 3 in
the mPGES-1 assay and at least equipotent to the literature
reference molecule 1. Fortunately, in comparison to 1 and 3,
the COX activity at 30 μM is clearly less inhibited by 11f,
To conclude, the concise synthetic method described in
Schemes 1-3 allowed us to prepare a selection of diverse
substituted small arylpyrrolizine derivatives from similar

4972 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 15
Liedtke et al.
starting materials (Table 1). In a cell free mPGES-1 assay,
many of our test compounds exceeded 3 in potency
(Table 2). The incorporation of arylsulfonimides led to a
3.2-fold increase in in vitro potency against mPGES-1
compared to 3. Moreover, the good submicromolar 5-
LOX inhibitory potency remains unaffected for those sul-
fonimides, whereas the COX inhibition could be casually
reduced compared to our lead 3.
These primary results reveal that the structure of 3 can be
successfully modified to yield potent in vitro inhibitors of
mPGES-1. Additional investigations will provide in-depth
information regarding the effect of further structural modifi-
cations of our lead compound and of selective mPGES-1
inhibition in vivo. In this context, mPGES-1 appears to be a
promising future target for the development of anti-inflam-
matory drugs.
Mass Spectrometry (Chemisches Zentralinstitut, Univer-
sity of Tuebingen), and Dr. Dominik Hauser for helpful
discussions. This research on mPGES-1 inhibitors was
financially supported by the Merckle/Ratiopharm Corpo-
rate Group, Blaubeuren, Germany, and the Fonds der
Chemischen Industrie, Germany.
Supporting Information Available: Eicosanoid pathway, com-
parison of the Lipinski rule criteria, synthetic procedures,
routine spectroscopic data, HRMS and HPLC data, biological
testing methods. This material is available free of charge via the
Internet at http://pubs.acs.org.
References
(1) Koeberle, A.; Siemoneit, U.; Buhring, U.; Northoff, H.; Laufer, S.;
Albrecht, W.; Werz, O. Licofelone suppresses prostaglandin
E2 formation by interference with the inducible microsomal pros-
taglandin E2 synthase-1. J. Pharmacol. Exp. Ther. 2008, 326, 975–
982.
(2) Black, A. T.; Gray, J. P.; Shakarjian, M. P.; Mishin, V.; Laskin, D.
L.; Heck, D. E.; Laskin, J. D. UVB light upregulates prostaglandin
synthases and prostaglandin receptors in mouse keratinocytes.
Toxicol. Appl. Pharmacol. 2008, 232, 14–24.
(3) Scholich, K.; Geisslinger, G. Is mPGES-1 a promising target for
pain therapy?. Trends Pharmacol. Sci. 2006, 27, 399–401.
(4) Wang, M.; Song, W. L.; Cheng, Y.; FitzGerald, G. A. Microsomal
prostaglandin E synthase-1 inhibition in cardiovascular inflamma-
tory disease. J. Intern. Med. 2008, 263, 500–505.
(5) Sampey, A. V.; Monrad, S.; Crofford, L. J. Microsomal prosta-
glandin E synthase-1: the inducible synthase for prostaglandin E2.
Arthritis Res. Ther. 2005, 7, 114–117.
Experimental Section
General. All reagents and solvents were of commercial quality
and used without further purification. HPLC analyses (see
Supporting Information for details) were employed for estab-
lishing the grade of purity of each test compound. The purity of
all tested compounds is g95%, if not denoted otherwise.
Synthesis of the Dihydropyrrolizinyl Acetic and Propionic Acid
Derivatives. The three-step general synthetic procedure for the
derivatization of the free acid position with varied sulfonimides
is specified in the Supporting Information.15
Preparation and Analytical Characterization of Exemplified
Test Compounds. 2-(6-(4-Chlorophenyl)-2,2-dimethyl-7-phenyl-
2,3-dihydro-1H-pyrrolizin-5-yl)-N-tosylacetamide (11d). According
to general procedure, 11d was obtained from [6-(4-chlorophenyl)-
2,2-dimethyl-7-phenyl-2,3-dihydro-1H-pyrrolizin-5-yl]acetic acid
(3) (0.5 g, 1.3 mmol) and toluene-4-sulfonamide (0.26 g, 1.5 mmol)
after 40 h at room temperature. The crude product was recrystal-
(6) Friesen, R. W.; Mancini, J. A. Microsomal prostaglandin E2
synthase-1 (mPGES-1): a novel anti-inflammatory therapeutic
target. J. Med. Chem. 2008, 51, 4059–4067.
(7) Jegerschoeld, C.; Pawelzik, S.-C.; Purhonen, P.; Bhakat, P.;
Gheorghe, K. R.; Gyobu, N.; Mitsuoka, K.; Morgenstern, R.;
Jakobsson, P.-J.; Hebert, H. Structural basis for induced formation
of the inflammatory mediator prostaglandin E2. Proc. Natl. Acad.
Sci. U.S.A. 2009, 105, 11110–11115.
(8) Riendeau, D.; Aspiotis, R.; Ethier, D.; Gareau, Y.; Grimm, E. L.;
Guay, J.; Guiral, S.; Juteau, H.; Mancini, J. A.; Methot, N.; Rubin,
J.; Friesen, R. W. Inhibitors of the inducible microsomal prosta-
glandin E2 synthase (mPGES-1) derived from MK-886. Bioorg.
Med. Chem. Lett. 2005, 15, 3352–3355.
(9) Fischer, L.; Hornig, M.; Pergola, C.; Meindl, N.; Franke, L.;
Tanrikulu, Y.; Dodt, G.; Schneider, G.; Steinhilber, D.; Werz, O.
The molecular mechanism of the inhibition by licofelone of the
biosynthesis of 5-lipoxygenase products. Br. J. Pharmacol. 2007,
152, 471–480.
lized from isopropanol to yield 0.28
g (40%) of 11d:
1
C30H29ClN2O3S (Mr = 533.08); H NMR (DMSO-d6) δ (ppm)
1.21 (s, 6H, (-CH3)2), 2.40 (s, 3H, tosyl-CH3), 2.70 (s, 2H,
C1-H2), 3.40 (s, 2H, CH2-CON<), 3.47 (s, 2H, C3-H2), 6.84-
3
7.28 (m, 9H, aryl-H), 7.43-7.47 (d, 2H, J=8.1 Hz, tosyl-C3-/
C5-H), 7.79-7.83 (d, 2H, 3J=8.3 Hz, tosyl-C2-/C6-H); IR (ATR)
3300, 2960, 1725, 1599, 1531, 1489, 1416, 1394, 1354, 1316,
1213, 1189, 1175, 1119, 1087, 1043, 1014, 978, 878, 837, 821, 812,
767, 735, 722, 701, 686, 668, 656 cm-1; HRMS (FT-ICR-MS)
m/z [M þ H]þ calcd for C30H30ClN2O3S 533.166 02, found
533.166 15.
(10) Laufer, S.; Striegel, H. G.; Dannhardt, G. Preparation of N-
Sulfonylpyrrolizineacetamides and Analogs as Cyclooxygenase
and Lipoxygenase Inhibitors. Patent 94-4419247[4419247], 22.
DE. 1-6-1994, 1995 (Merckle GmbH, Germany).
2-(6-(4-Chlorophenyl)-2,2-dimethyl-7-phenyl-2,3-dihydro-1H-
pyrrolizin-5-yl)-N-tosylpropanamide (11f). According to general
procedure, 11f was obtained from 3-(6-(4-chlorphenyl)-2,2-
dimethyl-7-phenyl-2,3-dihydro-1H-pyrrolizin-5-yl)propionic
acid (10b) (0.5 g, 1.3 mmol) and toluene-4-sulfonamide (0.26 g,
1.5 mmol) after 40 h at room temperature. The crude product
was recrystallized from diisopropyl ether to yield 0.3 g (39%) of
11f: C31H31ClN2O3S (Mr = 547.11); 1H NMR (DMSO-d6)
δ (ppm) 1.18 (s, 6H, (-CH3)2), 2.36 (m, 5H, -CH2-CON<,
tosyl-CH3), 2.71 (m, 4H, C5-CH2/C1-H2), 3.61 (s, 2H, C3-H2),
6.89-7.33 (m, 9H, aryl-H), 7.36-7.40 (d, 2H, 3J=8.2 Hz, tosyl-
C3-/C5-H), 7.79-7.83 (d, 2H, 3J=8.2 Hz, tosyl-C2-/C6-H); IR
(ATR) 3234, 3149, 2956, 2843, 1723, 1597, 1530, 1486, 1428,
1409, 1330, 1186, 1171, 1121, 1121, 1085, 1013, 956, 861, 828,
810, 758, 694, 660, 631, 579, 556, 505 cm-1; MS (FAB) m/z (%)
546.1 (Mþ, 45); HPLC (HP1090) 11.3 min, 89.8%; HRMS (FT-
ICR-MS) m/z [M þ H]þ calcd for C31H32ClN2O3S 547.181 67,
found 547.181 78.
(11) Keck, P. R. W. E. F. Arylpyrrolizin-Derivate als Inhibitoren der
mPGES1. Diploma Thesis, University of Tuebingen, 2008.
(12) Dannhardt, G.; Lehr, M. Antiinflammatory 2,3-dihydro-1H-pyr-
rolizines. XIII. Isomeric (diaryldihydropyrrolizinyl)acetic acids and
2-(diaryldihydropyrrolizinyl)ethanols. Arch. Pharm. (Weinheim, Ger.)
1988, 321, 159–162.
(13) Dannhardt, G.; Lehr, M. Antiphlogistic 2,3-dihydro-1H-pyrroli-
zines. 14. Isomer arrangement of diaryldihydropyrrolizinyl-formic
and -propionic acids. Arch. Pharm. (Weinheim, Ger.) 1988, 321,
545–549.
(14) Laufer, S. A.; Augustin, J.; Dannhardt, G.; Kiefer, W. (6,7-
Diaryldihydropyrrolizin-5-yl)acetic acids, a novel class of potent
dual inhibitors of both cyclooxygenase and 5-lipoxygenase. J. Med.
Chem. 1994, 37, 1894–1897.
(15) Laufer, S.; Striegel, H. G.; Dannhardt, G. [a]-Annelated Pyrrole
Derivatives and Pharmaceutical Use Thereof. WO 9532972 (A1),
1995.
(16) Dannhardt, G.; Steindl, L. Antiinflammatory 2,3-dihydro-1H-
pyrrolizines. III: aminomethylation and arylthiolation of 6,7-dia-
ryl-2,3-dihydro-1H-pyrrolizines. Arch. Pharm. (Weinheim, Ger.)
1986, 319, 65–69.
(17) Dannhardt, G.; Lehr, M. In-vitro evaluation of 5-lipoxygenase and
cyclooxygenase inhibitors using bovine neutrophils and platelets
and HPLC. J. Pharm. Pharmacol. 1992, 44, 419–424.
Acknowledgment. We are grateful to Gertrud Kleefeld
for assistance with biological testing, the Department of