New anti-inflammatory N-pyridinyl(alkyl)phthalimides acting as tumour necrosis factor-α production inhibitors

Article abstract of DOI:10.1016/S0223-5234(01)01254-5

This paper describes the synthesis of N-pyridinyl(alkyl)phthalimides related to N-phenyl-4,5,6,7-tetrafluorophthalimides known to be inhibitors of tumour necrosis factor-α (TNFα) production. Pharmacomodulation at the phthalimidic nitrogen led to the selection of two pharmacophoric fragments (2,4-lutidinyl and β-picolyl), allowing significant inhibition of TNFα production (compounds 12 and 17). Variation of the substituents linked to the homocycle of their phthalimide scaffold indicated that high (TNFα production) inhibitory potency could be achieved, notably by 5-fluoro, 4- or 5-nitro, 5-amino and especially tetrafluoro substitution. The most active compound, N-(pyridin-3-ylmethyl)-4,5,6,7-tetrafluorophthalimide (32) (84% inhibition at 10 μM), also produced an anti-oedematous effect in the PMA-induced mouse-ear swelling test. Although less active than dexamethasone, it exerted a marked reduction in ear thickness after oral administration (63% vs. 85% for dexamethasone at 0.2 mM kg^-1) and remained efficient after topical application (46% vs. 96% for the dexamethasone). It also induced potent inhibition in the rat carrageenan foot oedema test with an ID₅₀ (0.14 μM kg^-1) comparable with that of N-(2,6-diisopropylphenyl)phthalimide (4) (0.15 μM kg^-1).

Full text of DOI:10.1016/S0223-5234(01)01254-5

Eur. J. Med. Chem. 36 (2001) 639–649
´
© 2001 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved
PII: S0223-5234(01)01254-5/FLA
Original article
New anti-inflammatory N-pyridinyl(alkyl)phthalimides acting as tumour
necrosis factor-a production inhibitors
Xavier Collina,*, Jean-Michel Roberta, Gae´tane Wielgosza, Guillaume Le Bauta,
Christine Bobin-Dubigeonb, Nicole Grimaudb, Jean-Yves Petitb
aDepartment of Organic Chemistry and Medicinal Chemistry, Faculty of Pharmacy, 1 rue Gaston Veil,
F-44035 Nantes Cedex, France
bDepartment of Pharmacology, Faculty of Pharmacy, 1 rue Gaston Veil, F-44035 Nantes Cedex, France
Received 12 March 2001; revised 19 June 2001; accepted 19 June 2001
Abstract – This paper describes the synthesis of N-pyridinyl(alkyl)phthalimides related to N-phenyl-4,5,6,7-tetrafluorophthalimides
known to be inhibitors of tumour necrosis factor-a (TNFa) production. Pharmacomodulation at the phthalimidic nitrogen led to
the selection of two pharmacophoric fragments (2,4-lutidinyl and b-picolyl), allowing significant inhibition of TNFa production
(compounds 12 and 17). Variation of the substituents linked to the homocycle of their phthalimide scaffold indicated that high
(TNFa production) inhibitory potency could be achieved, notably by 5-fluoro, 4- or 5-nitro, 5-amino and especially tetrafluoro
substitution. The most active compound, N-(pyridin-3-ylmethyl)-4,5,6,7-tetrafluorophthalimide (32) (84% inhibition at 10 mM), also
produced an anti-oedematous effect in the PMA-induced mouse-ear swelling test. Although less active than dexamethasone, it
exerted a marked reduction in ear thickness after oral administration (63% vs. 85% for dexamethasone at 0.2 mM kg−1) and
remained efficient after topical application (46% vs. 96% for the dexamethasone). It also induced potent inhibition in the rat
carrageenan foot oedema test with an ID50 (0.14 mM kg−1) comparable with that of N-(2,6-diisopropylphenyl)phthalimide (4) (0.15
mM kg−1). © 2001 Editions scientifiques et me´dicales Elsevier SAS
´
N-pyridinylphthalimides / non-acidic NSAIDs / TNFa inhibitors / PMA oedema / carrageenan-induced oedema
1. Introduction
implicated in inflammatory processes and the patho-
genesis of various human disorders such as cachexia
(cancer, infection) [2], septic shock [3], multiple sclero-
sis [3], rheumatoid arthritis [4], brain injury and in-
flammation following stroke [5], psoriatic skin [6] or
inflammatory bowel disease (Crohn’s disease, ulcera-
tive colitis) [7]. Drugs inhibiting the production or
maturation of TNFa may have excellent therapeutic
potential, constituting an interesting alternative to
Tumour necrosis factor-a (TNFa), a protein
secreted by monocytes/macrophages in response to
many inflammatory stimuli, was initially detected be-
cause of its ability to induce hemorrhagic necrosis in
transplanted solid tumours [1]. TNFa is now gener-
ally regarded as a cytokine that displays pleiotropic
biological effects, such as anti-tumour activity, and
that can also stimulate the immune system and gener-
ate pro-inflammatory activity. As a pro-inflammatory
agent, TNFa elicits the production of other cytokines
(interleukins IL-1, IL-6 and IL-8) and upregulates
surface adhesion ligands on neutrophils and mono-
cytes [2]. Excessive production of TNFa has also been
classical
cyclooxygenase
inhibitors
in
these
pathologies.
In this context, thalidomide (1, figure 1), which was
used as a hypnotic/sedative agent but withdrawn from
the market because of its teratogenicity, could repre-
sent a lead compound. In fact, the effect of this drug
on TNFa production is cell-type specific, i.e. thalido-
mide enhances TNFa production in phorbol myris-
tate acetate-stimulated HL-60 cells [8], but inhibits
* Correspondence and reprints
E-mail address: x.collin@sante.univ-nantes.fr (X. Collin).

640
X. Collin et al. / European Journal of Medicinal Chemistry 36 (2001) 639–649
This paper describes the synthesis of a new series of
N-pyridinyl(alkyl)phthalimides (7–32) and the evalu-
ation of their in vitro activity in the inhibition of
TNFa (tables I and II); a first approach to the struc-
ture–activity relationships will be presented. Nine
phthalimides exhibiting an inhibition percentage of
more than 30% at 10 mM were evaluated as inflamma-
tion inhibitors after oral administration (table III) or
topical application (for the most efficient compound,
table IV). The five most potent compounds were
studied for their systemic anti-inflammatory activity
in the carrageenan-induced rat-paw oedema test (table
V).
TNFa production in lipopolysaccharide-stimulated
human monocytes [9]. On the basis of these findings,
a few groups have investigated the possible structural
modifications of thalidomide in order to create more
potent inhibitors of TNFa production. An increase in
potency of about 100-fold was obtained by the re-
placement of the glutarimide moiety of thalidomide
with the methyl 3-(3,4-dimethoxyphenyl)propionate
synthon or the 2,6-diisopropylphenyl moiety, in con-
junction with fluorine or amino substitution in the
homocycle of thalidomide or other N-aryl-phthal-
imides (2–4) [10, 11]. Fluorine substitution might also
prevent the formation of an arene oxide metabolite of
thalidomide, thought to be responsible for the terato-
genic effect of this drug [10].
We had described earlier the synthesis of struc-
turally related compounds, N-(4,6-dimethylpyridin-2-
yl) heteroarylcarboxamides, and evaluated their
inhibitory activity on the production of TNFa by
zymosan or lipopolysaccharide-activated macro-
phages [12]. Furan-2-carboxamide (5) was found to
inhibit TNFa production at an IC50 (ca. 70 mM) lower
than that of thalidomide (ca. 200 mM). Cytokine-sup-
pressing anti-inflammatory drugs (CSAIDs), typified
by bicyclic imidazole (6) [13], are another class of
compounds with high TNFa inhibitory activity. Like
the most potent compounds in this series, imidazole is
substituted by 4-pyridinyl and 4-fluorophenyl moi-
eties.
2. Chemistry
The preparation of target N-pyridinyl(alkyl)-
phthalimides is shown in figure 2. Classical condensa-
tion of amino(alkyl)pyridines with phthalic anhy-
drides in refluxing acetic acid (method a) afforded
these imides in satisfactory yields that could some-
times be increased by adding acetic anhydride
(method b) or by heating the mixture at 160–200 °C
in the absence of solvent (method c). Alternative
methods consisted of cyclizing the intermediate
phthalamic acids (methods d and e) or the corre-
sponding phthalamic ester (methods f and g). At-
tempted arylation in the presence of K2CO3 –DMF
(method h) gave poor results (18%) that could not be
improved by operating under microwave. The amidic
function in compound 20 was created after the activa-
tion of the corresponding acid by 1,1-carbonyldiimi-
dazole (method j) or phenyl dichlorophosphate
(method k). Physicochemical data for target com-
pounds 7–32 are summarised in tables I and II.
These findings led us to synthesize the closely re-
lated N-azaaryl(alkyl)phthalimides incorporating
amino(alkyl)pyridines and especially 6-amino-2,4-lu-
tidine [14], which are known to enhance anti-inflam-
matory activity appreciably in this series [15].
3. Results and discussion
3.1. In vitro inhibition of TNFh production
As the azaaryl ring (present in compounds 5 and 6)
appears to be important for optimum potency against
LPS-induced TNFa production by murine macrophages
[16], we evaluated the influence of: (i) the variation of
the position of the nitrogen atom; (ii) the number and
position of methyl or bromine substituents; and (iii) the
existence and length variation of an alkyl chain binding
the N-azaaryl and phthalimide rings (table I). In a-
Figure 1. Inhibitors of TNFa production.

X. Collin et al. / European Journal of Medicinal Chemistry 36 (2001) 639–649
641
Figure 2. Synthetic procedure to compounds 7–32. Reagents and conditions: (a) AcOH, reflux; (b) AcOH, Ac2O, reflux; (c)
160–200 °C; (d) THF; (e) Dean Stark, toluene, reflux; (f) SOCl2, reflux; (g) TEA, THF, reflux; (h) (1) K2CO3, DMF, (2) RꢀX,
180 °C; (i) Na2CO3, H2O; (j) 1,1’-carbonyldiimidazole, THF; (k) phenyl dichlorophosphate, TEA, 1,2-dichloroethane; (l) H2, Pd/C,
EtOH.
pyridyl(alkyl) substituents introduced at the phthalimido
nitrogen, we investigated the effect of substituents intro-
duced at the phthalimide moiety of compounds 12 and
15 (table II). In the two sub-series (2,4-lutidinyl and
b-picolyl series), higher activity was observed with 5-
fluoro derivatives (24: 40% inhibition, and 31: 41%
inhibition) than with their 4-fluoro analogues (23 and
30: <10%). However, the introduction of the electron-
withdrawing nitro group into the b-picolyl series exerted
a positive effect when fixed at position 4 (26: 72%
inhibition) or 5 (27: 73% inhibition), yielding activity
equivalent to that of the reference compound (4: 73%
inhibition). Contrary to the results obtained by Muller
et al. [11], the introduction of an electron-donating
group, such as an amino group, at position 4 in the
2,4-lutidinyl series (22) definitely decreased the TNFa
inhibitory activity (less than 10%). Nevertheless, the
corresponding 5-substituted analogue in the b-picolyl
series (29) was more potent, showing moderately lower
activity than the 5-nitro analogue (27: 62% vs. 73%). As
reported earlier [10], TNFa inhibitory activity was in-
creased more markedly by tetrafluorination of the
phthalimide moiety (25: 50% inhibition, and 32: 84%
inhibition). Phthalimide 32 was the most potent com-
pound of the series studied, showing slightly higher
pyridinyl derivatives, unsubstituted N-pyridinylphthal-
imide (7) lacked significant inhibition (less than 10%) of
TNFa production. The presence of a methyl substituent
in position 3, 4, 5 or 6 (8–11) induced no increase in
activity. The most potent compound in this series was
obtained by the introduction at nitrogen of a 4,6-
dimethylpyridin-2-yl moiety corresponding to our lead
pharmacophore (12: 37% inhibition at 10 mM) described
earlier [15]. Introduction of one (13) or two (14)
bromine atoms at positions 3 and 5 of 12 resulted in a
total loss of activity. Displacement of the nitrogen atom
to the b position (16) failed to enhance TNFa inhibi-
tion, whereas activity was clearly increased by the intro-
duction of a methylene group between pyridine and
phthalimide, leading to 17 (34% inhibition at 10 mM).
This phenomenon was not observed when the same
spacer was introduced into the a sub-series (15 was
inactive at the same concentration). Further elongation
of the alkyl chain in this sub-series allowed a moderate
increase in activity (20% inhibition with 19). Although
amide 20 incorporated the 3-(3,4-dimethoxyphenyl)-
propionyl moiety present in leads 2 and 3, it exerted
very poor inhibition (<10%).
Once it was determined that 2,4-lutidinyl and b-pi-
colyl moieties were the most potent of the different

642
X. Collin et al. / European Journal of Medicinal Chemistry 36 (2001) 639–649
Table I. Physical properties and TNFa inhibitory activity of N-pyridinyl-(alkyl)phthalimides 7–19.

X. Collin et al. / European Journal of Medicinal Chemistry 36 (2001) 639–649
643
Table II. Physical properties and TNFa inhibitory activity of N-pyridinyl (alkyl)phthalimides 20 and 22–32.

644
X. Collin et al. / European Journal of Medicinal Chemistry 36 (2001) 639–649
5-aminophthalimide 28 (which could not be tested in the
in vitro assay because of its very low solubility in the
solvent system used) were evaluated in the acute phorbol
myristyl acetate (PMA)-induced mouse-ear oedema
model of topical inflammation [17]. The activity of these
inhibitors was compared with that of phthalimide 4 and
reference drugs (dexamethasone and ibuprofen). Al-
though ibuprofen induced a marked increase of TNFa
production, it was selected, as a classical arylalkanoic
NSAID for comparison with the TNFa inhibitors stud-
ied (table III). Although the correlation between inhibi-
tion percentages after oral administration at 0.2 and 0.4
mM kg−1 was not quite satisfactory, all the compounds
tested exhibited significant ear oedema reduction at a
lower concentration. Determination of the ID50 of the
two most potent compounds (29 and 32) and compari-
son with the reference compound 4 and ibuprofen
brought to the fore that they were essentially equiactive:
0.12 0.03, 0.14 0.02, 0.13 0.01 and 0.18 0.05
mM kg−1). N-b-Picolyltetrafluorophthalimide (32) also
proved to be highly efficient in mouse-ear thickness
reduction after topical application of 2×500 mg (79%
inhibition). Although less potent than dexamethasone, it
remained significantly active at 2×100 mg (46% vs. 96%)
(table IV).
The five most active phthalimides in the mouse-ear
swelling test (25, 26, 28, 29 and 32) were selected,
together with the reference drugs (4, ibuprofen and
dexamethasone), for the determination of their in-
hibitory effect on systemic inflammation. Evaluation of
their ID50 in the carrageenan-induced rat-paw oedema
model [18] after oral administration (table V) confirmed
that compounds 25, 26 and 29 possess an activity level
comparable with that of ibuprofen (ID50=0.20
mM kg−1). Phthalimide 32 was equipotent to reference
compound 4 (0.14 vs. 0.15 mM kg−1).
inhibition of TNFa production than lead compound 4
[11] (84 and 73% inhibition, respectively).
3.2. In vivo pharmacological studies
Psoriatic skin shares many of the pathologic features
of phorbol ester-treated mouse skin, including elevated
levels of arachidonic acid metabolism products, inflam-
matory cells and cell proliferation. The nine most potent
TNFa inhibitors (% inhibition: >30% at 10 mM) and
Table III. Inhibition of phorbol myristate acetate-induced
oedema in mice (orally administered drugs).
Compound
Mouse-ear oedema (% inhibition)
0.2 mM kg−1
0.4 mM kg−1
12
5594
5896
5293
5595
5498
4195
4794
7692
5395
6393
6197
5694
8592
6998
6594
5695
7093
8391
6695
8092
9493
5994
9592
7994
8696
17
24
25
26
27
28
29
31
32
4
Ibuprofen
Dexamethasone
100
Table IV. Inhibition of phorbol myristate acetate-induced
oedema in mice (topically applied drugs).
Compound
Mouse-ear oedema (% inhibition)
2×100 mg
2×500 mg
32
4695
9692
7991
100
Dexamethasone
Finally, the acute toxicity of the new lead compound
32 was evaluated in male rats, showing an LD50 of
1.05 0.27 mM kg−1 after oral administration. This indi-
cated that its therapeutic coefficient, compared to that
of ibuprofen [19], remained moderate (8 vs. 25).
Table V. Inhibition of carrageenan-induced oedema in rats
(orally administered drugs).
Compound
Rat-paw oedema ID50 (mM kg−1
)
25
0.1890.06
0.2190.06
0.4790.04
0.2090.05
0.1490.04
0.1590.05
0.2090.01
0.02390.003
4. Conclusion
26
28
29
Overexpression of inflammatory cytokines, particu-
larly TNFa and IL-1, appears to play a key role in
the progression of inflammatory disease. In accor-
dance with its inhibitory activity in vitro on TNFa
production, 32 potently inhibited MPA-induced
32
4
Ibuprofen
Dexamethasone

X. Collin et al. / European Journal of Medicinal Chemistry 36 (2001) 639–649
645
mouse-ear swelling and carrageenan rat-paw oedema.
5.1.1. Synthesis of intermediary amines
This tetrafluorophthalimide is now being evaluated in
our laboratory in a mouse model of multiple PMA-in-
duced chronic inflammation [20] and a rat model of
trinitrobenzenesulfonic acid (TNBS)-induced colitis
[21], which are considered to be relevant for human
psoriasis and Crohn’s disease, respectively. The objec-
tive of the ongoing pharmacomodulation studies in
this series is to identify new low-molecular-weight
TNFa inhibitors with enhanced efficacy and better
tolerability. As the incorporation of amino acid
residues into 2-amino-4,6-dimethylpyridine has pro-
duced potent topical inflammation inhibitors [22],
studies are now in progress to incorporate such moi-
eties into our compounds.
5.1.1.1. 6-Amino-3-bromo-2,4-lutidine
Monobromination of 6-amino-2,4-lutidine, according
to Fox et al. [23] led to the 3-bromo derivative. M.p.
(dec.): 143 °C; lit. value [23]: 145–146 °C; 1H-NMR
(CDCl3): l 2.28 (s, 3H, 4-CH3), 2.50 (s, 3H, 6-CH3),
4.32 (s, 2H, NH2), 6.25 (s, 1H, Pyr H3).
5.1.1.2. 6-Amino-3,5-dibromo-2,4-lutidine
Dibromination of 6-amino-2,4-lutidine was carried
out using the same protocol [23], with the addition of 2
equiv. of bromine (instead of 1). M.p. (dec.): 135 °C; lit.
1
value [24]: 136–137 °C; H-NMR (CDCl3): l 2.49 (s,
3H, 4-CH3), 2.53 (s, 3H, 6-CH3), 4.92 (s, 2H, NH2).
5.1.1.3. 3-Amino-3-(3,4-dimethoxyphenyl)propionic acid
This b-amino-b-aryl acid was prepared by the treat-
ment of 3,4-dimethoxybenzaldehyde with ammonium
acetate and malonic acid in refluxing ethanol [25]. M.p.
5. Experimental protocols
1
5.1. Chemistry
(dec.): 203 °C. H-NMR (D2O): l 2.76–2.84 (m, 2H,
CH2), 3.79 (s, 3H, CH3O), 3.81 (s, 3H, CH3O), 4.55 (m,
1H, CH), 6.90–7.02 (m, 3H, ArꢀH).
Melting points (m.p. (dec.)) determined on a Tottoli–
Bu¨chi apparatus were uncorrected. The structures of the
products described were confirmed by IR, 1H- and
13C-NMR and microanalytical data. IR spectra were
run with KBr pellets on a Beckman IR 4230 IR spec-
trophotometer. 1H- and 13C-NMR spectra were
recorded on a Bruker AC 250 spectrometer (250 and 62
MHz), using CDCl3 or Me2SO-d6 as the solvent. Chem-
ical shifts (l) are reported in parts per million (ppm)
downfield from tetramethylsilane (Me4Si, 0.00 ppm) as
the internal standard, and coupling constants in hertz.
Microanalyses were performed on a Perkin–Elmer
CHN 240 apparatus. Analyses indicated by the symbols
of the elements were within 0.4% of the theoretical
values. Analytical thin-layer chromatography was per-
formed on pre-coated silica-gel aluminium plates (0.2
mm, GF254, E Merck). Spots were located by UV
illumination. Evaporations were performed in vacuo
(rotary evaporator). Sodium sulphate or phosphorus
pentoxide was used as the drying agent. Most crude
products were passed through short silica gel columns
(silica gel 60, 70–230 mesh, E Merck), using an appro-
priate mixture of dichloromethane and ethanol.
5.1.2. Synthesis of phthalimides
The following experimental protocols illustrate the
general procedures used to synthesise all the
compounds.
5.1.2.1. Method A: 2-(4,6-dimethylpyridin-2-yl)-4-
nitro-1H-isoindole-1,3-dione (21)
A solution of 2-amino-4,6-dimethylpyridine (1.83 g,
15 mmol) and 3-nitrophthalic anhydride (2.88 g, 15
mmol) in glacial acetic acid (15 mL) was heated under
reflux for 24 h. After the evaporation of the reaction
mixture to dryness under reduced pressure, the residue
was neutralised by a solution of sodium hydrogenocar-
bonate (4%) until effervescence ceased. The precipitate
obtained was washed with water, dried (P2O5) and re-
crystallised from diisopropyl ether to produce 21 (3.1 g,
70%). M.p. (dec.): 184 °C. IR (KBr, cm−1): ws(CꢁO
imide) 1780, was(CꢁO imide) 1742, was(NO2) 1532,
1
ws(NO2) 1372. H-NMR (CDCl3): l 2.40 (s, 3H, 4-CH3),
2.56 (s, 3H, 6-CH3), 7.03 (s, 1H, Pyr H5), 7.10 (s, 1H,
Commercially available solvents and chemicals were
used for syntheses, with the exception of the following
polyfunctional amines that were prepared using methods
reported in the literature.
Pyr H3), 7.98 (dd, 1H, H6, JH H5 =8.1 Hz, J 7 =8.0
6
6
H H
Hz), 8.16 (dd, 1H, H7, JH H6 =8.0 Hz, J 5 =1.1 Hz),
7
7
H H
8.22 (dd, 1H, H5, JH H6 =8.1 Hz, J 7 =1.1 Hz). Anal.
5
5
H H
(C15H11N3O4): C, H, N.

646
X. Collin et al. / European Journal of Medicinal Chemistry 36 (2001) 639–649
5.1.2.2. Method B:
2-(pyridin-4-ylmethyl)-1H-isoindole-1,3-dione (18)
cm−1): ws(CꢁO imide) 1740, was(CꢁO imide) 1730. 1H-
NMR (CDCl3): l 2.48 (s, 3H, 4-CH3), 7.43 (dd, 1H, Pyr
H5, JH H6 =4.8 Hz, J 3 =1.6 Hz), 7.45 (dd, 1H, Pyr
A mixture of 4-aminomethylpyridine (2.16 g, 20
mmol) and phthalic anhydride (2.96 g, 20 mmol) in
glacial acetic acid (20 mL) was refluxed for 3 h. The
solvent was removed under reduced pressure until the
volume reached ca. 6 mL. After addition of 7 mL of
acetic anhydride, the solution was refluxed again for 15
h. The solvent was removed under reduced pressure.
The residue was neutralised by a solution of sodium
hydrogenocarbonate (4%) until effervescence ceased.
The precipitate obtained was washed with water, dried
(P2O5) and recrystallised from ethanol 95% to give 18
(4.3 g, 91%). M.p. (dec.): 166 °C. IR (KBr, cm−1):
ws(CꢁO imide) 1740, was (CꢁO imide) 1720. 1H-NMR
(CDCl3): l 5.02 (s, 1H, NꢀCH2), 7.51 (d, 2H, Pyr H3,
Pyr H5, JH H2 =J 6 =5 Hz), 8.01 (m, 4H, ArꢀH), 8.55
5
5
H H
H3, JH H6 =1.0 Hz, J 5 =1.6 Hz), 8.08 (m, 4H,
3
3
ArꢀH), 8.58 (dd, 1H, PHyrHH6, JH H5 =4.8 Hz, J
=
6
6 3
1.0 Hz). 13C-NMR (DMSO-d6): l 20.75 (CH3), H12H3.91
(Pyr C3), 124.00 (C4,C7), 125.30 (Pyr C5), 131.71
(C3a,C7a), 135.34 (C5,C6), 146.28 (Pyr C4), 149.37 (Pyr
C6), 150.07 (Pyr C2), 166.86 (C1,C3). Anal.
(C14H10N2O2): C, H, N.
5.1.2.5. Methods F and G:
2-(4-methylpyridin-2-yl)-1H-isoindole-1,3-dione (9)
Methyl hydrogenophthalate (1.8 g, 10 mmol) was
dissolved in thionyl chloride (5 mL), and the solution
was refluxed for 1 h. The solvent was removed, and the
acid chloride obtained was dissolved in THF (5 mL) and
added to a mixture of 2-amino-4-methylpyridine (1.08 g,
10 mmol) and triethylamine (1.45 mL, 10 mmol) in THF
(20 mL). The solution was heated under reflux for 15 h.
The precipitate of triethylamine hydrochloride was sepa-
rated by filtration. The solvent was removed under
reduced pressure. The residue was washed with water
and dried (P2O5). Recrystallization from 95% ethanol
afforded 9 (1 g, 42%). M.p. (dec.): 176 °C.
3
5
(d, 2H, Pyr H2, HPyHr H6, JH H3 =J 5 =5 Hz).). 13C-
2
6
H H
NMR (DMSO-d6): l 40.24 (NꢀCH2), 122.38 (C4,C7),
123.66 (Pyr C3,C5), 131.95 (C3a,C7a), 134.95 (C5,C6),
145.75 (Pyr C4), 150.15 (Pyr C2,C6), 168.00 (C1 and C3).
Anal. (C14H10N2O2): C, H, N.
5.1.2.3. Method C:
2-(pyridin-4-ylmethyl)-1H-isoindole-1,3-dione (18)
A mixture of 4-aminomethylpyridine (1.08 g, 10
mmol) and phthalic anhydride (1.48 g, 10 mmol) was
heated to 180 °C for 2 h. The oil was cooled and
recrystallised from ethanol 95%; 1.9 g of pure product
18 was obtained: Yield: 80%; m.p. (dec.): 165 °C.
5.1.2.6. Method H:
2-(4,6-dimethylpyridin-2-yl)-1H-isoindole-1,3-dione (12)
A mixture of phthalimide (1.47 g, 10 mmol) and
K2CO3 (1.37 g, 10 mmol) in DMF (15 mL) was stirred
at r.t. for 6 h. 2-Bromo-4,6-dimethylpyridine (1.86 g, 10
mmol) was then added and the solution was heated
under reflux for 8 h. The solvent was removed under
reduced pressure and the residue obtained was washed
with water, dried (P2O5) and purified by column chro-
matography using dichloromethane–ethanol 95:5 as the
eluent. Recrystallization from 95% ethanol afforded 12
(0.45 g, 18%). M.p. (dec.): 191–193 °C. IR (KBr, cm−1):
ws(CꢁO imide) 1775, was(CꢁO imide) 1725. 1H-NMR
(CDCl3): l 2.39 (s, 3H, 4-CH3), 2.56 (s, 3H, 6-CH3),
7.02 (s, 1H, Pyr H5), 7.05 (s, 1H, Pyr H3), 7.78–7.93 (m,
4H, ArꢀH). 13C-NMR (DMSO-d6): l 20.65 (4-CH3),
23.85 (6-CH3), 121.20 (Pyr C3), 123.95 (C4,C7), 124.72
(Pyr C5), 131.73 (C3a,C7a), 135.31 (C5,C6), 145.55 (Pyr
C4), 150.11 (Pyr C6), 158.25 (Pyr C2), 166.98 (C1,C3).
Anal. (C15H12N2O3): C, H, N.
5.1.2.4. Methods D and E:
2-(4-methylpyridin-2-yl)-1H-isoindole-1,3-dione (9)
2-Amino-4-methylpyridine (1.08 g, 10 mmol) and ph-
thalic anhydride (1.48 g, 10 mmol) were dissolved in 20
mL of dry THF. The reaction mixture was stirred at
room temperature (r.t.) for 4 h and cooled. The suspen-
sion was then filtered, and the precipitate was washed
with THF and dried (P2O5), yielding 2.01 g (82%) of
N-(4-methylpyridin-2-yl) phthalamic acid. A suspension
of 2.01 g (8.2 mmol) of N-(4-methylpyridine-2-yl) phtha-
lamic acid in dry toluene (20 mL) was heated under
reflux for 72 h, using a Dean Stark to separate the
theoretical quantity of water (1.44 mL). The solvent was
removed under reduced pressure. The residue was neu-
tralised by a solution of sodium hydrogenocarbonate
(4%) until effervescence ceased. The collected solid was
washed with water and dried (P2O5). By recrystallization
from 95% ethanol, 1.36 g of pure white product was
obtained. Yield: 70%; m.p. (dec.): 176 °C. IR (KBr,
5.1.2.7. Methods I and J: N-(4,6-dimethylpyridin-2-yl)-
3-phthalimido-3-(3,4-dimethoxyphenyl) propanamide
(20)
A mixture of 3-amino-3-(3,4-dimethoxyphenyl) propi-

X. Collin et al. / European Journal of Medicinal Chemistry 36 (2001) 639–649
647
onic acid (1 g, 4.46 mmol), Na2CO3 (0.46 g, 4.46 mmol)
5.1.2.9. Method L: 2-(4,6-dimethylpyridin-2-yl)-
4-amino-1H-isoindole-1,3-dione (22)
and N-carbethoxyphthalimide (0.96 g, 4.46 mmol) in
water (10 mL) was stirred for 20 min at r.t. The solution
was acidified by 6 M HCl and extracted with diethyl
ether. The organic layer was dried (Na2SO4) and evapo-
rated. Crystallization from diisopropyl ether gave 1.1 g
of 3-phthalimido-3-(3,4-dimethoxyphenyl) propionic
acid. Yield: 69%; m.p. (dec.): 120–121 °C. 1,1-Car-
bonyldiimidazole (0.81 g, 5 mmol) was added portion-
wise to a solution of 3-phthalimido-3-(3,4-dimethoxy-
phenyl) propionic acid (1.77 g, 5 mmol). The mixture
was stirred for 1 h. 2-Amino-4,6-dimethylpyridine (0.79
g, 6.5 mmol) was then added to the solution. Stirring
was continued for 24 h at r.t. The solvent was removed
under reduced pressure. After column chromatography
with dichloromethane–ethanol 95:5 as the solvent sys-
tem, crystallization from diisopropyl ether gave 1.4 g of
pure product 20. Yield: 61%; m.p. (dec.): 157 °C. IR
(KBr, cm−1): w(NH) 3340, ws (CꢁO imide) 1775, was(CꢁO
imide) 1700, w(CꢁO amide) 1650, l(NH) 1540. 1H-NMR
(CDCl3): l 2.24 (s, 3H, 4-CH3), 2.35 (s, 3H, 6-CH3),
3.28 (dd, 1H, CH2, 3J=5.9 Hz, 2J=15.7 Hz), 3.83–3.87
(m, 6H, 3-OCH3, 4-OCH3), 3.91 (m, 1H, CH2), 5.93 (dd,
Palladium on charcoal (5%, 0.24 g) was added to a
solution of 21 (2.38 g, 8 mmol) in ethanol (160 mL). The
mixture was stirred under an H2 atmosphere. When the
consumption of hydrogen ceased, the catalyst was
filtered and the solvent was evaporated under reduced
pressure. The resulting solid was recrystallised from 95%
ethanol to afford 22 (1.54 g, 72%). M.p. (dec.): 200 °C.
IR (KBr, cm−1): w(NH) 3740, ws(CꢁO imide) 1760,
1
was(CꢁO imide) 1715, l(NH) 1645. H-NMR (CDCl3): l
2.39 (s, 3H, 4-CH3), 2.58 (s, 3H, 6-CH3), 5.34 (s, 2H,
NH2), 6.88 (dd, 1H, Ar H5, J 6 =8.2 Hz, J 7 =0.47
5
5
H H
Hz), 7.02 (s, 1H, Pyr H5), 7.05 (s, 1H, Pyr H3H), H7.22 (dd,
1H, Ar H7, JH H6 =7.1 Hz, J 5 =0.47 Hz), 7.46 (dd,
7
7
1H, Ar H6, JH H6 =7.1 Hz, JH H6 =8.2 Hz). 13C-NMR
7
5
H H
(DMSO-d6): l 20.61 (4-CH3), 23.86 (6-CH3), 109.00
(C3a), 111.46 (Pyr C3), 121.17 (Pyr C5), 122.04 (C5),
124.39 (C7), 132.41 (C7a), 135.94 (C6), 145.77 (C4),
147.42 (Pyr C4), 149.92 (Pyr C6), 158.11 (Pyr C2), 167.12
(C3), 168.40 (C1). Anal. (C15H13N3O2): C, H, N.
5.2. Pharmacology
3
3
1H, CH, J=5.9 Hz, J=5.8 Hz), 6.68 (s, 1H, Ph H2),
6.80 (d, 1H, Ph H5, J=7.5 Hz), 7.13 (s, 1H, Pyr H5),
7.15 (d, 1H, Ph H6, J=7.1 Hz), 7.66–7.78 (m, 4H,
ArꢀH), 7.81 (s, 1H, Pyr H3), 8.06 (s, 1H, NH). 13C-
NMR (DMSO-d6): l 21.05 (4-CH3), 23.67 (6-CH3),
38.31 (CH2 CO), 50.62 (NꢀCH), 55.87 (3-OCH3 and
4-OCH3), 111.22 (Pyr C3), 111.68 (Ph C2), 112.07 (Ph
C5), 119.78 (Ph C6), 119.89 (Pyr C5), 123.50 (C4,C7),
131.56 (Ph C1), 131.87 (C3a,C7a), 134.94 (C5,C6), 148.71
(Ph C3), 148.94 (Ph C4), 149.34 (Pyr C4), 151.57 (Pyr
C6), 156.34 (Pyr C2), 168.05 (C1,C3), 169.44 (CONH),
Anal. (C26H25N3O5): C, H, N.
5.2.1. TNFh inhibitory activity
The effect of the different compounds on the produc-
tion of the major pro-inflammatory cytokine TNFa by
in vitro activated peritoneal macrophages (MF) was
quantified as described earlier in Ref. [16]. Adult male
Swiss CF mice (18–25 g) were used. Briefly, thioglycol-
late-elicited mouse MF were isolated by peritoneal
washing with PBS without Ca–Mg, after cervical dislo-
cation. The cell suspension (1–2×106 cell per well) was
incubated in a 24 well culture plate in RPMI 10% FCS
for 2 h at 37 °C, 5% CO2. After three washes with PBS,
mice MF were pre-incubated with the different com-
pounds solubilised in ethanol (maximal final concentra-
tion of ethanol: 0.5%) at 10 mM. MF were then
stimulated by bacterial lipopolysaccharide or LPS (0.1
mg mL−1 — from Salmonella typhimurium, Sigma) for 4
h. Supernatants were collected and assessed for TNFa
content using the Wehi164 clone 13 cytotoxic assay [16].
5.1.2.8. Method K: N-(4,6-dimethylpyridin-2-yl)-3-
phthalimido-3-(3,4-dimethoxyphenyl) propanamide (20)
A solution of 3-phthalimido-3-(3,4-dimethoxyphenyl)
propionic acid (2.02 g, 5.7 mmol), 2-amino-4,6-
dimethylpyridine (0.69 g, 5.7 mmol) and triethylamine
(2.4 mL, 17.1 mmol) in dry dichloroethane (70 mL) was
cooled in an ice bath. Phenyl dichlorophosphate (0.85
mL, 5.7 mmol) was added dropwise. The mixture was
stirred at r.t. for 24 h. The solvent was removed under
reduced pressure. The residue was purified by column
chromatography using dichloromethane–ethanol 95:5 as
the eluent. By recrystallization from diisopropyl ether,
1.9 g of the pure product was obtained. Yield: 73%; m.p.
(dec.): 157 °C.
5.2.2. PMA-induced mouse-ear oedema (orally
administered drugs)
Induction of mouse-ear oedema was based on the
method of Carlson et al. [17] with some modifications.
Groups of five male Swiss mice weighing 20–25 g were
used. All animals were fasted 24 h before the experi-
ments and maintained in suitable environmental condi-

648
X. Collin et al. / European Journal of Medicinal Chemistry 36 (2001) 639–649
tions throughout the experiments. PMA (phorbol-12-
myristate-13-acetate) was dissolved in 80% aqueous eth-
anol at a concentration of 250 mg mL−1; 10 mL−1 was
applied topically to the anterior and posterior surfaces
of the right ear of each mouse. The left ear (control)
received the vehicle (10 mL of 80% aqueous ethanol).
The products studied were orally administered 1 h be-
fore the PMA application. Ear thickness was measured
with a model micrometer gauge (Oditest Kroeplin) 3 h
and 30 min after treatment. Ear oedema, calculated by
subtracting the thickness of the left ear (vehicle) from
the thickness of the right ear (PMA), was expressed as
an increase in ear thickness. The percentage of inhibi-
tion of the inflammatory reaction was determined for
each animal by the comparison of ear oedema in treated
and non-treated animals.
saline solution. Rats were fasted 24 h before the experi-
ment, and water (1.5 mL/100 g body weight) was orally
administered twice (20 and 4 h before injections). The
volume of both the hind paws of control and treated
animals were measured with a plethysmograph 3 h after
injection. Rats were kept in the same experimental con-
ditions. The inhibition percentage of the inflammatory
reaction was determined for each animal by comparison
with controls, and calculated by the formula I (%)=
100×(1−dt/dc), where dt is the difference in paw vol-
ume in the drug-treated group and dc the difference in
the control group.
5.2.5. Acute toxicity (LD50)
Tests were carried out on rats fasted for 24 h. Groups
of eight animals were treated orally with the compounds
studied at various dose levels. Animals were watched for
mortality until 48 h, and for symptoms until day 8. The
value of the lethal dose (LD50) was determined accord-
ing to the method of Miller and Tainter [26].
5.2.3. PMA-induced mouse-ear oedema (topically
applied drugs)
Groups of five male Swiss mice weighing 20–25 g
were used. Mice were briefly anaesthetised with ether for
ear application. PMA (phorbol-12-myristate-13-acetate)
was dissolved in 80% aqueous ethanol at a concentra-
tion of 250 mg mL−1; 10 mL was applied topically to the
anterior and posterior surfaces of the right ear of each
mouse. The left ear (control) received the vehicle (10 mL
of 80% aqueous ethanol). A solution of 100 mg or 500 mg
of drug in 10 mL of ethanol was applied to the inner
surface of the right ear of treated animals, and 10 mL of
vehicle (ethanol) to the contralateral ear as the control.
These applications were made 30 min before PMA
application, and then again 5 min later. Ear thickness
was measured with a model micrometer gauge (Oditest
Kroeplin) 3 h and 30 min after treatment. Ear oedema,
calculated by subtracting the thickness of the left ear
(vehicle) from the thickness of the right ear (PMA), was
expressed as an increase in ear thickness. The percentage
of inhibition of the inflammatory reaction was deter-
mined for each animal by the comparison of ear oedema
in treated and non-treated animals.
References
[1] Carswell E.A., Old L.J., Kassel R.L., Green S., Foire N.,
Williamson B., Proc. Natl. Acad. Sci. USA 72 (1975) 3666–
3670.
[2] Tracey K.J., Cerami A., Annu. Rev. Cell Biol. 9 (1993) 317–
343.
[3] Beutler B. (Ed.), Tumor Necrosis Factor: The Molecules and
their Emerging Role in Medicine, Raven Press, New York,
1992.
[4] Trembly B., Slikker J., Ann. N.Y. Acad. Sci. 765 (1995) 62–71.
[5] Gater P.R., Determination of the role of TNF in inflammatory
responses using a receptor fusion protein, Communication pre-
sented at the IBC’s Fourth Annual Cytokine Conference,
Philadelphia, PA, 24–25 April 1997.
[6] Schwarz T., Lugert E., in: Maktar H. (Ed.), Pharmacology of
the Skin, CRC Press, Boca Raton, FL, 1992, pp. 283–313.
[7] Schreiber S., Nikolaus S., Hampe J., Homling J., Koop I.,
Grossner B., Lochs H., Raedler A., Lancet 353 (1999) 459–461.
[8] Azuma A., Miyachi H., Shibata Y., Hashimoto Y., Iwasaki S.,
Biol. Pharm. Bull. 19 (1996) 1001–1003.
[9] Sampaio E., Sarno E.N., Gallily R., Cohn Z.A., Kaplan G., J.
5.2.4. Carrageenan-induced rat-paw oedema
Exp. Med. 173 (1991) 699–703.
Anti-inflammatory activity against carrageenan-in-
duced rat-paw oedema was assayed in adult male Wistar
CF rats weighing 180–220 g, according to the method
of Winter et al. [18] with slight modifications. The drugs
were orally administered 1 h before the injection of 0.05
mL of a 1% suspension of carrageenan saline into the
subcutaneous tissue of one hind paw. The other hind
paw was injected in the same way with 0.05 mL of a
[10] Niwayama S., Turk B.E., Liu J.O., J. Med. Chem. 39 (1996)
3044–3045.
[11] Muller G.W., Corral L.G., Shire M.G., Wang H., Moreira A.,
Kaplan G., Stirling D.I., J. Med. Chem. 39 (1996) 3238–3240.
[12] Lang F., Robert J.M., Boucrot P., Welin L., Petit J.Y., J.
Pharmacol. Exp. Ther. 275 (1995) 171–176.
[13] Black R.A., Bird T.A., Mohler K.M., in: Bristol J.A. (Ed.),
Annual Reports in Medicinal Chemistry, vol. 32, Academic
Press, London, 1997, pp. 241–249.

X. Collin et al. / European Journal of Medicinal Chemistry 36 (2001) 639–649
649
Colome J., Sacristan A., Varez G., Pisaturo G., Arzneim.
[14] Collin X., Robert J.M., Robert-Piessard S., Le Baut G., Bobin-
Dubigeon C., Vernhet L., Lang F., Petit J.Y., Pharm. Pharma-
col. Commun. 4 (1998) 27–31.
-Forsch./Drug Res. 27 (1977) 1006–1012.
[20] Alfort J.G., Stanzey P.L., Todderud G., Agents Actions 37
(1992) 260–267.
[21] Morris G.P., Beck P.L., Herridge M.S., Depew W.T., Szewczuk
M.R., Wallace J.L., Gastroenterology 96 (1989) 795–803.
[15] Robert J.M., Robert-Piessard S., Courant J., Le Baut G.,
Robert B., Lang F., Petit J.Y., Grimaud N., Welin L., Eur. J.
Med. Chem. 3 (1995) 915–924.
[22] Duflos M., Courant J., Grimaud N., Renard P., Manechez D.,
Caignard D.H., Eur. J. Med. Chem. 33 (1998) 635–645.
[16] Espevik T., Nissen-Meyer J., J. Immunol. Methods 95 (1986)
99–105.
[23] Fox B.A., Threlfall T.L., Org. Synth. Coll. 44 (1964) 34–39.
[17] Carlson R.P., O’Neill. Davis D., Chang J., Lewis A.J., Agents
Actions 17 (1985) 198–205.
[24] Mariella R.P., Belcher E.P., J. Am. Chem. Soc. 74 (1952)
1916–1921.
[18] Winter C.A., Risley E.A., Nuss G.W., Proc. Soc. Exp. Biol.
[25] Kalvin D., Woodard R., J. Org. Chem. 50 (1985) 2259–2263.
Med. 111 (1962) 544–547.
[26] Miller L.C., Tainter M.L., Proc. Soc. Exp. Biol. Med. 57 (1944)
[19] Orzalesi G., Selleri R., Caldini O., Volpato I., Innocenti F.,
261–265.