Cyclooxygenase-1/2 (COX-1/COX-2) and 5-lipoxygenase (5-LOX) inhibitors of the 6,7-diaryl-2,3-1H-dihydropyrrolizine type

Article abstract of DOI:10.1016/S0223-5234(02)01418-6

A series of 6,7-diaryl-2,3-1H-dihydropyrrolizines was prepared as COX-1/COX-2 and 5-LOX inhibitors. The inhibition of COX-1 was evaluated using intact bovine platelets as the enzyme source, whereas LPS-stimulated human monocytes served as the enzyme source for inducible COX-2. The determination of arachidonic metabolites was performed by HPLC for COX-1 and RIA for COX-2. The balance between COX-1/COX-2 and 5-LOX inhibition can be shifted by modifying the substitution pattern of the phenyl moiety at the 6- and 7-position of the pyrrolizine nucleus. Structure-activity relationships are discussed.

Full text of DOI:10.1016/S0223-5234(02)01418-6

European Journal of Medicinal Chemistry 37 (2002) 953ꢁ959
/
www.elsevier.com/locate/ejmech
Original article
Cyclooxygenase-1/2 (COX-1/COX-2) and 5-lipoxygenase (5-LOX)
inhibitors of the 6,7-diaryl-2,3-1H-dihydropyrrolizine type
Holger Ulbrich a,b,ꢀ, Bernd Fiebich c, Gerd Dannhardt a
a Institute of Pharmacy, Johannes Gutenberg-University of Mainz, Staudingerweg 5, D-55099 Mainz, Germany
b Department of Physiology and Pharmacology, Karolinska Institutet, Von Eulers Vag 4, SE-171 77 Stockholm, Sweden
¨
c Department of Psychiatry and Psychotherapy, Hospital of the Albert Ludwigs, University of Freiburg, Hauptstr. 5, D-79104 Freiburg, Germany
Received 31 May 2002; received in revised form 23 September 2002; accepted 26 September 2002
Abstract
A series of 6,7-diaryl-2,3-1H-dihydropyrrolizines was prepared as COX-1/COX-2 and 5-LOX inhibitors. The inhibition of COX-1
was evaluated using intact bovine platelets as the enzyme source, whereas LPS-stimulated human monocytes served as the enzyme
source for inducible COX-2. The determination of arachidonic metabolites was performed by HPLC for COX-1 and RIA for COX-
2. The balance between COX-1/COX-2 and 5-LOX inhibition can be shifted by modifying the substitution pattern of the phenyl
moiety at the 6- and 7-position of the pyrrolizine nucleus. Structureꢁ
# 2002 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved.
/
activity relationships are discussed.
´
Keywords: Non-steroidal anti-inflammatory drugs; Diarylpyrrolizines; COX-1, COX-2 and 5-LOX inhibition; Structureꢁactivity relationship
/
1. Introduction
COX-2, on the other hand, is induced by many kinds
of inflammatory mediators and plays an important role
in the prostaglandin biosynthesis associated with in-
flammatory responses. Therefore, NSAIDs with selec-
tive COX-2 inhibitory activity should be more useful for
the treatment of inflammatory diseases. However, the
simple concept of COX-2 being an exclusively pro-
inflammatory inducible enzyme can no longer be
upheld. Recently, it was found that COX-2 is expressed
constitutively in the brain, spinal cord and kidney, as
well as in numerous other organs [3]. It was hypothe-
NSAIDs are widely used for the treatment of pain,
inflammation and fever. The main mechanism of action
of these drugs is believed to be the inhibition of the
cyclooxygenase enzymes (COX-1 and COX-2) [1] and
consecutively the conversion of arachidonic acid to
prostaglandins.
The major side effects associated with all the currently
available NSAIDs are gastrointestinal (GI) hemorrhagia
and ulceration [2]. These side effects during anti-
inflammatory therapy are caused by interference with
the physiological properties of prostaglandins. Recently
several approaches have been proposed to reduce the
unwanted side effects of NSAIDs. Among these, one
involves the selective inhibition of COX isoenzymes;
COX-1 is constitutively expressed and probably plays a
role as a ‘housekeeping enzyme’, for example in main-
taining the lining of the stomach and in endothelial cells
contributing to the normal function of the cardiovas-
cular system via the release of prostacyclin (PGI2).
sised that selective inhibition of COX-2, may*
inhibition of endothelial prostacyclin synthesis*
/
through
have
/
severe side effects in the cardiovascular system and
promote thrombogenesis [4,5]. Selective COX-2 inhibi-
tion with celecoxib elevates blood pressure and pro-
motes leukocyte adherence to the endothelium [6].
After the introduction of the highly selective COX-2
inhibitors celecoxib (Celebrex†) and rofecoxib (Vioxx†;
Fig. 1) onto the market, the VIOXX gastrointestinal
Outcomes Research (VIGOR) and the Celecoxib long-
term Arthritis Safety Study (CLASS) studies suggest
that these agents causing also GI injury like conven-
tional NSAIDs and increased systemic blood pressure
[7]. A more recent strategy for devising a gastric-sparing
ꢀ Correspondence and reprints:
E-mail address: holger.ulbrich@fyfa.ki.se (H. Ulbrich).
´
0223-5234/02/$ - see front matter # 2002 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved.
PII: S 0 2 2 3 - 5 2 3 4 ( 0 2 ) 0 1 4 1 8 - 6

954
H. Ulbrich et al. / European Journal of Medicinal Chemistry 37 (2002) 953ꢁ959
/
metabolite of tebufelone, which acts as an antioxidant
[17,18].
This paper deals with diarylpyrrolizines as a class of
compounds with multiple inhibitory effects on both
COX enzymes, especially COX-2, and 5-LOX, which
may display therapeutic advantages over NSAIDs in
treating anti-inflammatory diseases as discussed above.
Fig. 1. Structures of selective COX-2 inhibitors.
2. Chemistry
NSAID involves chemically coupling a nitric oxide
(NO)-releasing moiety to the parent NSAID [8]. Along
with prostaglandins NO appears to play an important
cytoprotective role in GI homeostasis [9].
The synthetic route is analogous to that of Dannhardt
[19] and is outlined in Fig. 3. 4-Chloro-butyronitril was
condensed with
a commercially available benzyl-
Grignard and ring closure to 2-benzyl-D1-pyrroline
was accomplished in situ by heating. The D1-pyrrolines
were cyclised with the corresponding bromoacetophe-
nones (hetero-aromatic 1-bromo-1-ethanones) in abs.
CH2Cl2 by stirring for 24 h at room temperature.
Pyrrolinium intermediates were not isolated. Then a
solution (5%) of NaHCO3 was added and stirring
continued for a further 24 h. After addition of water,
the organic layer was separated and dried (Na2SO4 sicc.)
and the solvent evaporated. Variation at the substitution
pattern was accomplished by using different kinds of 2-
benzyl-D1-pyrrolines and bromoacetophenones (hetero-
aromatic 1-bromo-1-ethanones). The data of the com-
pounds are summarised in Tables 3 and 4.
Another interesting class of compounds to minimise
the toxicity of NSAIDs are dual COX/LOX-inhibitors.
Besides interfering with the production of prostaglan-
dins these substances inhibit the biosynthesis of chemo-
tactic leukotrienes, which are another important
mediator in inflammatory processes, i.e. they induce
the invasion of neutrophils into the inflamed area as a
prerequisite for the formation of gastric ulcers [10].
Diarylpyrrolizines have been widely investigated in-
hibitors of COX-1, COX-2 and 5-LOX pathways. A
compound of these series Licofelone (ML-3000; Fig. 2)
showed anti-inflammatory and analgesic activity in
osteoarthritis in clinical trials comparable to conven-
tional NSAIDs and selective COX-2 inhibitors with a
safer GI profile [11ꢁ15]. COX-1/COX-2 inhibitors with
/
3. Biological assays
balanced inhibiting capacities (Gkz) on both enzymes
could also improve the therapeutic benefit in the
treatment of inflammatory diseases [16]. 7-tert-Butyl-
2,3-dihydro-3,3-Dimethyl derivatives and 1,2-isothiazo-
lidine-1,1-dioxide (g-sultam) derivatives (S-2474) (Fig.
2) were recently reported as COX-2/5-LOX-inhibitors.
One of these substances was found to be the active
All the compounds were tested in intact cell assays
described earlier [19ꢁ22].
/
4. Results and discussion
Fig. 3 shows the synthetic pathway of the general
procedure.
First all the compounds were tested for their ability to
inhibit COX-1 and 5-LOX. The most interesting com-
pounds underwent further investigation to ascertain
their COX-2 inhibiting activity. Table 1 summarises
the results of the compounds tested.
4.1. 6,7-Diaryl-2,3-dihydropyrrolizines
The derivatives 1ꢁ
9 are different in R1 and R2.
/
Table 1 summarises the impact of these substituents.
All the compounds except 7 inhibit the COX-1 pathway.
The potent COX-1 inhibiting substances 1 and 8 bear a
methyl sulphide group at the C-6 phenyl moiety with an
approximate IC50 of 1.1 versus 0.03 mM. The most
potent COX-1 inhibitor 8 bears an additional lipophilic
substituent R2. The thioether compounds may act as
Fig. 2. Structures of a well balanced COX-1/2 and dual inhibitors of
COX-2/5-LOX (for explanations see text).

H. Ulbrich et al. / European Journal of Medicinal Chemistry 37 (2002) 953ꢁ
/
959
955
Fig. 3. Synthetic pathway. (a) CH2Cl2(abs.) room temperature 24 h. (b) NaHCO3 (5%), 24 h.; for R1, R2 and Y see Table 1.
Table 1
In vitro inhibitory potencies of the 6,7-diaryl-2,3-dihydropyrrolizines and the 6-heteroaryl-7-aryl-2,3-dihydropyrrolizines
a
a
a
a
Compound
R1
R2
Y
IC50 (mM) COX-1
IC50 (mM) 5-LOX
IC50 (mM) COX-2
IC50 COX-1/IC50 COX-2
1
SCH3
H
H
H
F
ꢁ
ꢁ/
ꢁ/
ꢁ/
ꢁ/
ꢁ/
ꢁ/
ꢁ/
/
1.1
0%
10
30% (1 mM)
0.03
3.6
n.d.
n.d.
n.d.
n.d.
27
2
SO2CH3
NHSO2CH3
SCH3
25%
10%
28%
4.3
3
3.4
0%
0%
0%
0%
0%
10
4
26% (0.1 mM)
0.16
5
SO2CH3
SO2CH3
F
Cl
6
8.2
0%
0.008
n.t.
0.9
1025
n.d.
0.03
140
n.d.
n.d.
n.d.
n.d.
n.d.
0.43
7
7
NHSO2CH3 Cl
8
SCH3
SO2CH3
H
CH3
0.03
0.7
9
10
CH3
H
ꢁ/
0.005
20%
S
S
S
S
O
33%
3.9
10
11
12
Cl
Br
H
0%
8.4
0%
10
n.t.
n.t.
H
25%
0%
13
14
CH3
H
H
1.8
n.t.
H
20%
0.16
0.0028
0.004
ꢀ10
Licofelone
Diclofenac
Indomethacin
4ƒ-Cl-SC 57666
0.21
n.t.
n.t.
n.t.
0.37
0.0004
0.0005
0.016
8
ꢀ625
a
IC50 values of compounds tested represent the mean value of four determinations (mM or percentage inhibition at a concentration of 10 mM),
n.t., not tested; n.d., not determined.

956
H. Ulbrich et al. / European Journal of Medicinal Chemistry 37 (2002) 953ꢁ959
/
Table 2
Chemical structures of diclofenac, indomethacin and cyclopentene derivative 4ƒ-Cl-SC-57666, IC50 values, COX-1/COX-2 ratios
peroxidase-reducing agents by consuming electrons and
oxygen, hence destroying the peroxidase-derived oxi-
dant, which is thought to oxidise the putative tyrosine
residue to a tyrosyl radical, and thereby initiating the
cyclooxygenase pathway [23]. Sulindac, an anti-inflam-
matory drug with a sulphoxide moiety has to be
metabolised to the corresponding thioether in order to
become a potent inhibitor of cyclooxygenase. It was
shown that the lipophilicity of the thioethers is 1000-fold
higher, thus demonstrating the requirements for lipo-
philic interactions of drug and enzyme.
sulphone it is possible to shift the specificity for 5-LOX
as well as for COX-2. The methanesulphonamide 3
shows moderate dual inhibition of COX-2 and 5-LOX
(IC50: 3.6 vs. 3.4 mM). The sulphone 2 inhibits COX-2
potently (IC50: 0.03 mM) and shows significant inhibi-
tion of 5-LOX (IC50: 10 mM). Compound 5 possesses a
COX-1/COX-2 ratio of 27. The selective COX-2 in-
hibitors 6 and 9 in this range are sulphones, which bear a
lipophilic chloro- or methyl-moiety R2 documenting a
COX-1/COX-2 ratio of 1025ꢁ140.
/
Compound 9 combines the structural requirements
for COX-1/2 and 5-LOX inhibition (IC50: 0.7, 0.005 and
10 mM). Summarising these data, it can be stated that
the combination of a methyl sulphonyl group or
methanesulphonamide residue R1 with a lipophilic
moiety R2 produces triple inhibitors of COX-1/2 and
5-LOX (2, 3 and 9).
Compound 1 (R2ꢂ
COX-2 almost equipotently (IC50: :
cant increase of COX-2 inhibition is observed for 4
(R2ꢂ
/
H) and 8 (R2ꢂ
/
CH3) inhibit
/
1 mM). A signifi-
/
F) (26% inhibition at 0.1 mM). Whereas all the
thioethers (1, 4, and 8) show no inhibiting potency
against 5-LOX, by oxidation of the sulphur atom to a
Table 3
Melting points, yields and analytical data of new compounds
a
Compound
Formula (all compound were analysed for C, H, N)
Yield (%)
Melting point (8C)
R1
R2
1
2
C20H19NS
C20H19NO2S
C20H20N2O2S
60
55
47
65
58
60
55
58
61
65
59
58
61
55
99
170
168
120
128
146
190
190
155
115
87
SCH3
SO2CH3
NHSO2CH3
SCH3
SO2CH3
SO2CH3
NHSO2CH3
SCH3
SO2CH3
H
H
H
3
H
4
C20H18FNS
C20H18FNO2
F
5
F
6
C
C
20H18ClNO2S
20H19ClN2O2S
Cl
Cl
CH3
CH3
H
7
8
C21H19NS
C21H21NO2S
C17H15NS
9
10
11
12
13
14
C
C
17H14ClNS
17H14BrNS
Cl
Br
H
105
94
H
C18H15NS
C17H15NO
CH3
H
H
95
H
a
Elemental analysis is within 90.4% for elements indicated unless otherwise.

H. Ulbrich et al. / European Journal of Medicinal Chemistry 37 (2002) 953ꢁ
/
959
957
Table 4
Physical data of compounds synthesised
Compound IR (KBr)
1H-NMR (200 MHz), d (ppm), J (Hz)/MS (EI, 70 eV): m/z
1
2
3
1600 (CꢀC), 1320, 1150
(SO2ꢁN)
2.53 (s, 3H, ꢁSCH3), 2.57ꢁ
1H, vinyl-H), 7.12ꢁ
7.27 (m, 9H, aromatic) 306/305 (32, 3%, Mꢃ+), 290 (11%, C19H16NS), 258 (2%,
19H16NS), 202.07 (2%, C16H10), 56 (12%, C3H6N)
2.55ꢁ2.60 (m, 2H, H-2), 2.95 (t, 2H, H-1, Jꢂ7.30), 3.07 (s, 3H, ꢁSO2CH3), 4.03 (t, 2H, H-3, Jꢂ7.30), 6.85
(s, 1H, vinyl.-H), 7.12ꢁ7.30 (m, 5H, aromatic), 7.45 (d, 2H, AA?BB?, Jꢂ8.3 H-aromatic), 7.75 (d, 2H,
AA?BB?, Jꢂ8.3 H-aromatic) 339/338/337 (6, 22, 100%, Mꢃ+) 258 (11%, C19H16N), 181 (4%, C13H11N)
2.49ꢁ2.60 (m, 2H, H-2), 2.92ꢁ2.99 (m, 2H, H-1), 3.06 (s, 3H, ꢁCH3), 4.04 (t, 2H, H-3, Jꢂ7.0), 6.50 (s, 1H,
ꢁNH), 6.74 (s, 1H, vinyl-H), 7.04ꢁ
7.24 (m, 9H, aromatic) 355/354/353 (2, 6, 24%, Mꢃ+), 272 (100%,
19H16N2)
2.51 (s, 3H, ꢁSCH3), 2.57ꢁ
1H, vinyl-H), 6.99ꢁ
7.24 (m, 8H, aromatic) 323 (100%, Mꢃ+), 304 (14%, Mꢃ+ꢄF), 276 (1%, C19H15NF)
/
2.61 (m, 2H, H-2), 2.99 (t, 2H, H-1, Jꢂ7.0), 4.03 (t, 2H, H-2, Jꢂ7.0), 6.76 (s,
/
C
1600 (CꢀC), 1320, 1160
(SO2ꢁN)
/
/
1600 (CꢀC), 1320, 1160
(SO2ꢁN)
/
/
/
C
4
5
1640 (CꢀC)
/2.65 (m, 2H, H-2), 2.99 (t, 2H, H-1, Jꢂ7.0), 4.05 (t, 2H, H-2, Jꢂ7.0), 6.77 (s,
/
1600 (CꢀC), 1320, 1160
(SO2ꢁN)
2.45ꢁ/2.60 (m, 2H, H-2), 2.95 (t, 2H, H-1, Jꢂ7,3), 3.08 (s, 3H, SO2CH3, Jꢂ7,3), 6.88 (s, 1H, vinyl-H),
6.90ꢁ
/
7.20 (m, 4H, aromatic), 7.38 (d, 2H, AA?BB?, Jꢂ8.3, H-aromatic), 7.78 (d, 2H, AA?BB?, Jꢂ8.3, H-
aromatic) 356/355/354 (49, 100, 20, Mꢃ+), 276 (11%, C19H15NF), 248 (17%), 220 (7%)
6
1600 (CꢀO), 1320, 1160
(SO2ꢁN)
2.45ꢁ2.65 (m, 2H, H-2), 2.98 (t, 2H, H-1, Jꢂ7,3), 3.10 (s, 3H, SO2CH3), 4.13 (t, 2H, H-3, Jꢂ7.3), 6.85 (s,
/
1H, vinyl-H), 7.10 (d, 2H, AA?BB?, Jꢂ8.3), 7.26 (d, 2H, AA?BB?, Jꢂ8.3), 7.38 (d, 2H, AA?BB?, Jꢂ8.3),
7.78 (d, 2H, AA?BB?, Jꢂ8.3) 373/372/371 (38, 25, 100%, Mꢃ+), 292 (9% C19H15NCl), 257 (7%, C19H15N),
215 ( 2%, C16H9N)
7
8
1590 (CꢀO), 1320, 1160
(SO2ꢁN)
1630 (CꢀC)
2.49ꢁ
6.74 (s, 1H, vinyl-H), 7.00ꢁ
2.34 (s, 3H, ꢁCH3), 2.47 (s, 3H, ꢁSCH3), 2.58 (m, 2H, H-2), 3.01 (t, 2H, H-1, Jꢂ7,3), 4.01 (t, 2H, H-3,
Jꢂ7,3), 6.75 (s,1H, vinyl-H), 7.08ꢁ
7.24 (m, 8H, aromatic) 319/318 (100, 9%, Mꢃ+), 304 (23%, C20H18NS),
228 (2%, C14H14NS), 123 (2%, C7H7S), 105 (3%, C7H7N)
2.36 (s, 3H, ꢁCH3), 2.50ꢁ2.65 (m, 2H, H-2), 3.00 (t, 2H, H-1, Jꢂ7,3), 3.08 (s, 3H, ꢁSO2CH3), 4.04 (t, 2H,
/
2.65 (m, 2H, H-2), 2.92ꢁ
/
3.01 (m, 2H, H-1), 3.09 (s, 3H, ꢁCH3), 4.03 (t, 2H, H-3), 6.38 (s, 1H, ꢁNH),
/7.30 (m, 8H, aromatic)
/
9
10
11
12
13
1590 (CꢀC), 1320, 1150
(SO2ꢁN)
/
H-3, Jꢂ7.3), 6.85 (s, 1H, vinyl-H), 7.10 (s, 4H, H-aromatic), 7.40 (d, 2H, AA?BB?, Jꢂ8.3), 7.75 (d, 2H,
AA?BB?, Jꢂ8.3). 352/351 (23%, 100% Mꢃ+), 272 (11%, C20H18N), 257 (4%, C19H15N), 215 (1%, C16H9N)
1600 (CꢀC)
1600 (CꢀC)
1600 (CꢀC)
1595 (CꢀC)
2.44ꢁ
(t, 1H, H-4), 7.10 (d, 1H, H-5), 7.18ꢁ
C15H13NS), 188 (8%), 140 (3%, C10H6N)
2.45ꢁ2.59 (m, 2H, H-2), 2.96 (t, 2H, H-1, Jꢂ7.3), 4.01 (t, 2H, H-3, Jꢂ7.3), 6.51 (d, 1H, H-3 thiophen),
6.72 (d, 1H, 4-H thiophen), 6,79 (s,1H, vinyl-H), 7.14ꢁ7.36 (m, 5H, aromatic) 299/298/297 (100, 12, 3%,
Mꢃ+), 264 (3%, C17H14NS), 207 (3%, C15H13N), 165 (3%, C12H7N)
2.43ꢁ2.60 (m, 2H, H-2), 2.97 (t, 2H, H-1, Jꢂ7.3), 4.00 (t, 2H, H-3, Jꢂ7.3), 6.51 (d, 1H, H-3 thiophen),
6.79 (s 1H, vinyl-H), 6.84 (d, 1H, H-4 thiophen), 7.14ꢁ7.36 (m, 5H, aromatic) 345/344/343 (11, 65, 20%,
Mꢃ+), 265 (100%, C17H14NS), 236 (16%, C15H10NS), 188 (18%), 146 (16%), 90.9 (17%)
2.44 (s, 3H, ꢁCH3), 2.50ꢁ2.65 (m, 2H, H-2), 3.00 (t, 2H, H-1, Jꢂ7.3), 4.05 (t, 2H, H-3, Jꢂ7.3), 6.48ꢁ
(m, 2H, H-3, H-4, thiophen), 6.81 (s, 1H, vinyl-H), 7.14ꢁ7.5 (m, 5H, aromatic) 2.43ꢁ2.57 (m, 2H, H-2),
2.87ꢁ3.05 (t, 2H, H-1, Jꢂ7.3), 3.90ꢁ4.10 (t, 2H, H-3, Jꢂ7.3), 5.99ꢁ6.05 (d, 1H, H-3 furan), 6.30ꢁ6.37 (t,
1H, H-5 furan), 6.90ꢁ6.95 (s, 1H, vinyl-H), 7.10ꢁ7.50 (m, 6H, aromatꢃfuran)
/
2.60 (m, 2H, H-2), 2.98 (t, 2H, 4,03 (t, 2H, H-3, Jꢂ8.3), 6.75 (d, 1H, H-3), 6.83 (s, 1H, vinyl-H), 6.95
/
7.45 (m, 5H, aromatic) 265/264/263 (100, 21, 5%, Mꢃ+), 239 (13%,
/
/
/
/
/
/
6.70
/
/
/
/
/
/
/
/
14
1605 (CꢀC)
2.43-2.57 (m, 2H, H-2), 2.96 (t, 2H, H-1, Jꢂ7.3), 4.00 (t, 2H, H-3, Jꢂ7.3), 6.02 (d,1H, H-3 furan), 6.34 (t,
1H, H-5 furan), 6.93 (s, 1H, vinyl-H), 7.10-7.50 (m, 6H, aromatꢃfuran) 250/249/248 (23, 100, 16%, Mꢃ+),
221 (29%), 208 (9%, C14H9NO), 165 (9%, C12H7N)
4.2. 6-Heteroaryl-7-aryl-2,3-dihydropyrrolizines
In this series, 13 is the most potent COX-2 inhibitor
(IC50: 1.8 mM).
The potency of anti-inflammatory drugs often de-
pends especially on their lipophilicity. Replacement of a
phenyl residue by a thiophene moiety significantly
increases COX-1 inhibition [24].
5. Summary and conclusions
Contrary to this hypothesis our 6-thiophene and 6-
furane-7-aryl-2,3-dihydropyrrolizines possess no or
poor COX-1/2 potency. The C-6 thiophene derivative
10 shows weak inhibition of COX-1/2 and 5-LOX at
concentration of 10 mM (33, 50, 20%). The highest
COX-1 inhibitory activity is found for the 31-chlor-
othiophenederivative 11 (IC50: 3.9 mM). Compound 12
inhibits the 5-LOX pathway noticeably (IC50: 8.4 mM).
These results indicate that the inhibitory effects of the
compounds tested to inhibit COX-1/2 and 5-LOX can
be shifted by varying R1 and R2:
a) Replacement of the phenyl-residue at C-6 by a
thiophene or a furan moiety produces poor COX-1/
2 or 5-LOX inhibitors.
b) The most potent COX-1 inhibitor (8) bears a methyl
sulphide group at C-6. A lipophilic group at the C-7

958
H. Ulbrich et al. / European Journal of Medicinal Chemistry 37 (2002) 953ꢁ959
/
residue increases COX-1 and especially COX-2
activity: 4 (26% inhibition at 0.1 mM) is ten times
more potent than 1 (30% inhibition at 1 mM).
6.2. Enzyme assays
For cell preparations and apparatus see references
[19ꢁ22].
c) By oxidation of the sulphur atom, it is possible to
shift the specificity for 5-LOX as well as for COX-2.
d) The combination of a methylsulphonyl group or
methanesulphonamide R1 with a lipophilic R2
moiety produces triple inhibitors of COX-1/2 and
5-LOX (2, 3 and 9). Compound 9 is a potent
inhibitor of COX-1/2 and 5-LOX with a high
selectivity to COX-2 in comparison to benzofuran
derivatives and tebufelone with the same selectivity
profile published. This combination of COX and 5-
LOX inhibition with preference for COX-2 could
offer opportunities to treat inflammatory diseases
with better tolerated drugs, thus enhancing patient
compliance. Investigations to confirm this hypoth-
esis are ongoing.
/
All the compounds were tested in an intact cell assay
described earlier [20]. Shortly, the inhibition of COX-1
was determined using bovine thrombocytes and that of
5-LOX using bovine polymorphonuclear leucocytes
(PMNLs). The cells were incubated with the compounds
and stimulated with calcium ionophore A 23187. The
amounts of LTB4 and 5-HETE were quantified to
measure 5-LOX activity. For the evaluation of the
COX-1 12-HHT and PGE2 were determined by HPLC.
This method does not give any information about the
inhibition of COX-2, so the method of Bauer and
Fiebich [21] was used to determine the inhibition of
LPS-induced (10 ng mlꢄ1) COX-2 production of PGE2
in human monocytes. The amounts of PGE2 are
determined by RIA. Another method to confirm the
results, which is described earlier was used [22].
6. Experimental
It allows the determination of IC50-values for inhibi-
tors of COX-1/2 on cells of the same origin: Unstimu-
lated bovine aortic coronary endothelial cells (BAECs)
were used as a source of COX-1 and BAECs pretreated
with 100 mM acetylsalicylic acid (ASA) and activated
with phorbol myristate acetate (PMA) were used as a
source of COX-2. IC50 values were calculated with the
program GRAFIT, Erithacus Software Ltd., UK. The
6.1. Chemistry
Melting points (m.p.) were determined on a Buchi
¨
SMP-20 apparatus and are uncorrected. All the com-
pounds were analysed for C, H and N. Proton and 13C-
NMR were run on a Bruker AC-200 and AC-400
spectrometer using TMS as the internal standard.
Mass spectra (EI) were obtained at 70 eV with a Varian
MAT 7 spectrometer. IR spectra were obtained on a
standard deviations of the obtained values (nꢂ4) were
/
less than 12% of the IC50 values of COX-1. Diclofenac,
indomethacin and 4ƒ-Cl-SC57666 were used as reference
substances (Table 2).
PerkinꢁElmer 1310 spectrometer with KBr disks. Mi-
/
croanalyses were determined on a Haereus CHN rapid
or a Carlo Erba Strumentazione1106. TLC plates of
silica gel (Merck G254) were used to monitor reaction
development. Column liquid chromatography (silica gel
Acknowledgements
200ꢁ
from reaction mixtures; solvent A: toluol, solvent B:
ethyl acetateꢁdiisopropyl ether (8:2), solvent C: ethyl
acetateꢁdiisopropyl ether (6:4), solvent D: etherꢁdi-
chloromethane (6:4), solvent E: ether, solvent F: PEꢁ
/
400 mesh, Merck) was used for product isolation
The financial support of the Fonds der Chemischen
Industrie is gratefully acknowledged.
/
/
/
/
dichloromethane (1:1). All the reagents were of analy-
tical grade and obtained as follows: salt for buffer
solutions, solvents: Merck, Darmstadt (Germany); acet-
References
ophenone derivatives 1ꢁ3, 7, calcium ionophore A
/
[1] O. Laneuville, D.K. Breuer, D.L. Dewitt, T. Hla, C.D. Funk,
23187, diclofenac, indomethacin: Sigma, Munchen
¨
W.L. Smith, J. Pharmcol. Exp. Ther. 271 (1994) 927ꢁ934.
/
[2] H.-C. Huang, J.J. Li, D.J. Garland, T.S. Chamberlain, E.J.
Reinhard, R.E. Manning, K. Seibert, C.M. Koboldt, S.A.
Gregory, G.D. Anderson, A.W. Veenhuizen, Y. Zhang, W.E.
Perkins, E.G. Burton, J.N. Cogburn, P.C. Isakson, D.B. Reitz, J.
(Germany), HPLC reference substances 12-HHT, inter-
nal standards PGB2 and 15-keto-PGE2: Paesel, Frank-
furt/Main (Germany). Bovine blood was obtained from
the local abattoir. Synthesis of 4ƒ-Cl-SC 57666 was
carried out according to reference [25].
Med. Chem. 39 (1996) 253ꢁ
/266.
[3] C. Hoffmann, Curr. Med. Chem. 7 (2000) 1113ꢁ
/1120.
[4] F. Catella-Lawson, B. McAdam, B.W. Morrison, S. Kapoor, D.
Kujubu, L. Antes, K.C. Lasseter, H. Quan, B.J. Gertz, G.A.
6.1.1. General procedure for the preparation of 2,3
FitzGerald, J. Pharmacol. Exp. Ther. 289 (1999) 735ꢁ741.
/
dihydro-1H-pyrrolizines (1ꢁ14)
/
[5] B.F. Mc Adam, F. Catella-Lawson, I.A.: Mardini, S. Kapoor,
J.A. Lawson, G.A. Fitzgerald, Natl. Acad. Sci. USA 96 (1999)
2,3-Dihydro-1H-pyrrolizines were prepared accord-
ing to reference [18].
272ꢁ277.
/

H. Ulbrich et al. / European Journal of Medicinal Chemistry 37 (2002) 953ꢁ
/
959
959
[6] M.N. Muscara, N. Vergnolle, F. Lovren, C.R. Triggle, S.N.
Elliott, S. Asfaha, J.L. Wallace, Br. J. Pharmacol. 129 (2000)
Jansen, I.J. Magrisso, K.E. Wehmeyer, D.E. Kuhlenbeck, T.H.
Eichhold, R.L.M. Dobson, S.P. Sirko, R.W. Farmer, J. Med.
1414ꢁ
/
1423.
[7] Notification of ‘Arzneitelegramm’ from 16.02.2001.
Chem. 41 (1998) 1112ꢁ1123.
/
[18] M. Inagaki, T. Tsuri, H. Jyoyama, T. Ono, K. Yamada, M.
Kobayashi, Y. Hori, A. Arimura, K. Yasui, K. Ohno, S. Kakudo,
K. Koizumi, R. Suzuki, M. Kato, S. Kawai, S. Matsumoto, J.
[8] U.K. Bandarage, L. Chen, X. Fang, D.S. Garvey, A. Glavin,
D.R. Janero, L.G. Letts, G.J. Mercer, J.K. Saha, J.D. Schroeder,
M.J. Shumway, S.W. Tam, J. Med. Chem. 43 (2000) 4005ꢁ
[9] S.N. Elliott, W. McKnight, G. Cirino, J.L. Wallace, Gastroenter-
ology 109 (1995) 524ꢁ530.
[10] H. Asako, P. Kubes, J. Wallace, T. Gaginella, R.E. Wolf, D.N.
Granger, Am. J. Physiol. 262 (1992) G903ꢁG908.
[11] Drugs Future 25 (2000) 1093.
[12] S. Laufer, J. Augustin, G. Dannhardt, W. Kiefer, J. Med. Chem.
37 (1994) 1894ꢁ1897.
[13] S. Laufer, S. Tries, J. Augustin, G. Dannhardt, Arzneim.-Forsch./
Drug Res. 44 (1994) 629ꢁ636.
/
4016.
Med. Chem. 43 (2000) 2040ꢁ
[19] G. Dannhardt, R. Obergrusberger, Arch. Pharm. 312 (1979) 896ꢁ
907.
/
2048.
/
/
[20] G. Dannhardt, M. Lehr, J. Pharm. Pharmacol. 44 (1992) 419ꢁ
/
/
424.
[21] K. Lieb, B.L. Fiebich, M. Busse-Grawitz, M. Hull, M. Berger, J.
¨
Bauer, J. Neuroimmun. 67 (1996) 77ꢁ
[22] G. Dannhardt, H. Ulbrich, Inflamm. Res. 50 (2001) 262ꢁ
[23] R. Dietz, W. Nastainczyk, H. Ruf, Eur. J. Biochem. 171 (1988)
321ꢁ328.
/
81.
/
/269.
/
/
[14] J.Y. Reginster, Eular-Congress Satellite Symposium 15.06.2001,
Prag.
[24] J.-Y. Gauthier, Y. Leblanc, W.C. Black, C.-C. Chan, W.A.
Cromlish, R. Gordon, B.P. Kennedey, C.K. Lau, S. Le´ger, Z.
Wang, D. Ethier, J. Guay, J. Mancini, D. Riendeau, P. Tagari, P.
Vickers, E. Wong, L. Xu, P. Prasit, Bioorg. Med. Chem. Lett. 6
[15] S. Laufer, H.-G. Striegel, K. Neher, P. Zechmeister, C. Donat, K.
Stolingwa, S. Baur, S. Tries, T. Kammermeier, G. Dannhardt, W.
Kiefer, Arch. Pharm. 330 (9ꢁ
/
10) (1997) 307ꢁ
¨
/
312.
[16] G. Dannhardt, W. Kiefer, G. Kramer, S. Maehrlein, U. Nowe, B.
(1996) 87ꢁ92.
/
[25] J.J. Li, G.D. Anderson, E.G. Burton, J.N. Cogburn, J.T. Collins,
D.J. Garland, S.A. Gregory, H.-C. Huang, P.C. Isakson, C.M.
Koboldt, E.W. Logusch, M.B. Norton, W.E. Perkins, E.J.
Reinhard, K. Seibert, A.W. Veenhuizen, Y. Zhang, D.B. Reitz,
Fiebich, Eur. J. Med. Chem. 35 (2000) 499ꢁ510.
/
[17] J.M. Janusz, P.A. Young, J.M. Ridgeway, M.W. Scherz, K.
Enzweiler, L.I. Wu, L. Gan, R. Darolia, R.S. Matthews, D.
Hennes, D.E. Kellstein, S.A. Green, J.L. Tulich, J. Rosario-
J. Med. Chem. 38 (1995) 4570ꢁ4578.
/