Phosphodiesterase 4 inhibitors as airways smooth muscle relaxant agents: synthesis and biological activities of triazine derivatives

Article abstract of DOI:10.1016/S0968-0896(98)00266-1

A series of triazine derivatives was synthesized. The compounds were evaluated for tracheal smooth muscle relaxant and type 4 phosphodiesterase inhibitory activities. A highly significant correlation was observed between the two effects. Two compounds exh

Full text of DOI:10.1016/S0968-0896(98)00266-1

Bioorganic & Medicinal Chemistry 7 (1999) 509±516
Phosphodiesterase 4 Inhibitors as Airways Smooth Muscle
Relaxant Agents: Synthesis and Biological Activities of
Triazine Derivatives
F. Leroux,a B. J. van Keulen,b J. Daliers,b N. Pommerya and J. P. Henicharta,*
aInstitut de Chimie Pharmaceutique, 3 rue du Professeur Laguesse, BP 83, F-59006 Lille, France
bUCB Pharma Sector, Chemin du Foriest, B-1420, Braine-l'Alleud, Belgium
Received 15 September 1998; accepted 5 November 1998
AbstractÐA series of triazine derivatives was synthesized. The compounds were evaluated for tracheal smooth muscle relaxant and
type 4 phosphodiesterase inhibitory activities. A highly signi®cant correlation was observed between the two e€ects. Two com-
pounds exhibited potent relaxant activity (EC50: 17 and 24 nM) and might be useful for the treatment of asthma. # 1999 Elsevier
Science Ltd. All rights reserved.
as rolipram or RP 73401 in experimental studies,3,4,18±23
it is generally accepted that inhibitors of this isoenzyme
Introduction
Cyclic nucleotide phosphodiesterases (PDE) control the
rate of breakdown of the intracellular messengers, 30,50-
cyclic adenosine and guanosine monophosphates
(cAMP, cGMP), by hydrolyzing the 30-ribose phosphate
bond of the nucleotides to form 50-nucleoside mono-
phosphates. Currently, more than 30 isoenzymes are
recognized and grouped into at least seven di€erent
families, each with distinct substrate speci®cities and reg-
ulatory characteristics.1±6 The concept of di€erently
expressed and regulated PDE isoenzymes implies that
individual PDEs are likely to be good targets for ther-
apeutic intervention in diseases caused or regulated by
cyclic nucleotide-modulated transduction mechanisms.
may have useful bronchodilator and anti-in¯ammatory
properties for treating both the symptoms and the
underlying causes of the disease.
Asthma is an in¯ammatory condition characterized by
bronchoconstriction, microvascular leakage, mucus
hypersecretion, bronchial hyperresponsiveness and by
invasion of the airways by in¯ammatory cells.7,8 The
PDE 4 family has a marked preference for cAMP as a
substrate and, as a consequence, inhibition of PDE 4
activity results in increased intracellular levels of this
nucleotide, which is generally associated with dampen-
ing e€ects on airway smooth muscle tone and on acti-
vation and mediator release from in¯ammatory cells.
Thus, on the basis of the presence of the type 4 PDE in
airway smooth muscle9±12 and in proin¯ammatory
cells,13±17 and of the e€ects of the PDE 4 inhibitors such
This study focuses on the synthesis and pharmacologi-
cal evaluation of triazine derivatives, designed by struc-
tural analogy with cAMP. We present the guinea pig
tracheal relaxant e€ect of these compounds, which was
found to be highly correlated with their ability to inhibit
PDE 4 catalytic activity. Two compounds exhibit potent
relaxant activity, comparable to the activity displayed
by RP 73401.
Chemistry
The s-triazine derivatives reported in this paper are lis-
ted in Table 1. In general, these derivatives and inter-
mediates were prepared by the methods illustrated in
Scheme 1 and Scheme 2. 2-(Cyclo)alkyl-4,6-dichloro-s-
triazines (II) were synthesized by the reaction of 2,4,6-
trichloro-s-triazine (I) with Grignard reagents according
Key words: Bronchodilators; enzyme inhibitors; substituent e€ects;
antiin¯ammatories.
*Corresponding author. Fax: +33-3-20-96-43-61; e-mail:
henicha@phare.univ-lille2.fr
0968-0896/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(98)00266-1

510
F. Leroux et al. / Bioorg. Med. Chem. 7 (1999) 509±516
Table 1. Comparative e€ects of triazine derivatives, rolipram, and RP73401 on guinea pig tracheal smooth muscle and on polymorphonuclear type
4 PDE activity
Tracheal relaxation
PDE 4 inhibition
IC50(mM)
R1
R2
R3
EC50 (mM)
1 mM-induced e€ecta
1
2
3
0.052b‹0.029
0.49‹0.30
58‹9
0.16‹0.11(3)
67‹15
63‹7
5.68
2.94
0.71‹0.22
4
1.5‹0.2
42‹3
24.5
5
6
0.15‹0.06
0.47‹0.11
83‹8
0.69
5.50
69‹10
7
2.5‹1.4
39‹13
30.1‹12.8(2)
8
9
0.017b‹0.002
1.7‹0.9
77‹7
0.14‹0.10(4)
53.3
50‹23
10
11
12
0.076b‹0.027
0.12b‹0.07
90‹3.3
45‹2
61‹15
20‹3
0.38‹0.39(4)
0.33‹0.19(3)
61.2‹16.0(3)
13
0.021b‹0.023
0.024b‹0.011
0.16‹0.02
59‹4
78‹10
77‹2
0.18‹0.04(2)
0.15‹0.01(6)
1.96‹0.33(3)
14(d)
15
Rolipram
RP73401
0.014b‹0.008
0.010‹0.006
40‹11
76‹7
0.09‹0.03(8)
0.004‹0.001(2)
Relaxation values are the means of three determinations; the number of PDE 4 inhibitory experiments is indicated in brackets; a% of maximal
relaxation to isoprenaline (1 mM); bEC50 values for the ®rst relaxant e€ect below 1 mM (agents giving biphasic curves, see text); (d) dextrogyre.

F. Leroux et al. / Bioorg. Med. Chem. 7 (1999) 509±516
511
to Hirt et al.24 The intermediates III were obtained by
Results and Discussion
treatment of II with an equimolar amount of the
appropriate amine in the presence of K2CO3, according
to the method of Nishigaki et al.25 Compounds 1, 2, 4±7,
10, and 11 were obtained from the reaction of III with
two equivalents of the appropriate amine or one
equivalent of amine in the presence of triethylamine in
dioxane at re¯ux temperatures.26 Substituted cycloalkyl-
s-triazines 13 and 14 were prepared from the corres-
ponding biguanides (IV) and the appropriate sub-
stituted cycloalkyl carboxylic esters or carboxylic acid
chloride27±30 (Scheme 2). Compound 12 was obtained
by this procedure using ethyl succinyl chloride and sub-
sequent aminolysis of the intermediary ester. S,S-diox-
ides 8, 9, 14, and 15 were obtained by oxidation with
oxone31 of the corresponding sul®de derivatives, pre-
pared according to either of the procedures described.
Triazine derivatives relaxed the guinea pig tracheal
smooth muscle with varying potencies and ecacies
(testing concentrations from 1 nM to 100 mM). Two
groups were discerned. In the ®rst group (1, 8, 10, 11,
13, 14; EC50 from 17 to 120 nM, Table 1), compounds
relaxed isolated trachea in a biphasic manner as
demonstrated by the di€erences in curve slope at low,
middle and high concentrations (Fig. 1). They were
more potent than the compounds of the second group
whose concentration±relaxation relationships appeared
to follow sigmoidal curves (2±7, 9, 12, 15; EC50 from
150 nM to 9 mM). The ®rst phase was observed for con-
centrations below 1 mM. Depending on the compound
tested, this ®rst phase ended in a marked plateau or an
in¯exion of the curve, between 46 and 73% of the maximal
Scheme 1.
Scheme 2.

512
F. Leroux et al. / Bioorg. Med. Chem. 7 (1999) 509±516
relaxation achieved with 1 mM isoprenaline. The second
phase occurred with upper concentrations (>1 mM) and
gave nearly the maximal smooth muscle relaxation in
most cases.
curve, rolipram and the triazine derivatives 1, 10, 11, 13,
displayed a smaller absolute e€ect at 1 mM than some
compounds giving a sigmoidal curve (Table 1).
The biphasic nature of some concentration±relaxation
relationships suggested that there might be at least two
mechanisms at the origin of the triazine derivatives
mediated-relaxation of guinea pig trachea. We exam-
ined the inhibitory activity of these compounds on type
4 phosphodiesterase isoenzymes from polymorpho-
nuclear leukocytes and determined IC50 values ranging
from 0.14 to 61.2 mM (Table 1). Moreover, we observed
a highly signi®cant correlation (R2=0.93, n=15, Student:
p>0.1%, Fig. 2) between the ability of triazine deriva-
tives to inhibit PDE 4 isoenzymes and their ability to
relax trachea (using EC50 value for the ®rst relaxant
e€ect occurring below 1 mM in the case of biphasic
curves). Thus, one part of the triazine-relaxant e€ect,
and in particular the ®rst e€ect that occurred with
biphasic curves giving compounds used at concentra-
tions below 1 mM, may be due to PDE 4 isoenzyme
inhibition. For these last compounds (1, 8, 10, 11, 13,
14), complete relaxation may be achieved when another
mechanism is associated. For instance, A. L. Harris and
his collaborators32 observed that pretreatment of the
trachea with a ®xed concentration of CI-930, a selective
inhibitor of PDE 3 isozyme, converted the rolipram
biphasic curve into a sigmoidal curve, near complete
below 1 mM. Thus, in the case of rolipram, the ®rst
relaxant e€ect may be due to the speci®c PDE 4 iso-
enzyme inhibition, as very elevated concentrations of
agent may inhibit PDE 4 and PDE 3 isoenzymes.
Among the triazine derivatives giving biphasic curves,
two di€erent types could be discerned: the rolipram
type and the RP73401 type. In the case of rolipram
(Fig. 1A), the concentration±response curve was mark-
edly biphasic and reached only 37% of the maximal
isoprenaline-induced tracheal relaxation during the ®rst
phase. On the other hand, RP 73401 mean curve was
sigmoidal (Fig. 1B) and reached 76% of the maximal
isoprenaline-induced tracheal relaxation at 1 mM. This
relaxation was not signi®cantly improved with higher
RP73401 concentrations (80% at 100 mM). Despite dif-
ferences between the pro®le of their relaxation curve, the
EC50 values for rolipram and RP73401 relaxant e€ect
below 1 mM were nearly identical: 14 and 10 nM respec-
tively. The relaxant e€ects elicited by these speci®c PDE
4 inhibitors, used as reference products, were consistent
with the results provided in the literature.19,32
The concentration±relaxation curves for compounds 8
and 14 resembled the curve obtained for RP73401
(Fig. 1B) with an additional e€ect above 1 mM. On the
contrary, the concentration±response curve for 10 was
rather similar to rolipram (Fig. 1A), but shifted to
higher concentrations. The derivatives 1, 11, and 13
gave an intermediary response (Fig. 1A). As a con-
sequence of the intermediate plateau in their relaxation
In summary, these results show that compounds 8 and
14 may be the most interesting agents in this series of
triazine derived PDE 4 isoenzyme inhibitors because
they combine a low EC50 (respectively 17 and 24 nM)
with an important e€ect during the ®rst phase of the tra-
cheal relaxation (73 and 64% of maximal relaxation to
isoprenaline). These [1,3,5]triazine derivatives are ideally
substituted with a cyclopropylamino group into the 2
position (substituent R1, Table 1), with a cyclopropyl or
Figure 2. Guinea pig tracheal relaxation as a function of poly-
morphonuclear PDE 4 inhibition. EC50 values (expressed in ordinate)
are either the concentrations of relaxant agent that produced 50% of
the maximal e€ect, or the concentrations that produced 50% of the
®rst relaxant phase, according to the sigmoidal or biphasic aspect of
the curves. Abscissa: inhibition of polymorphonuclear PDE 4.
Figure 1. Biphasic cumulative concentration±relaxation curves of
rolipram and compounds 1, 10, 11, and 13 (A) and RP73401 and
compounds 8 and 14 (B). For clarity, SE have not been reported.

F. Leroux et al. / Bioorg. Med. Chem. 7 (1999) 509±516
513
a 2-methyl-cyclopropyl group into the 4 position (R2) and
with a 1,1-dioxo-1,3-thiazolidine or a 1,1-dioxo-1,4-thio-
morpholine ring in the 6 position (R3).
Cyclopropyl-(4-cyclopropyl-6-morpholin-4-yl-[1,3,5]triazin
-2-yl)-methyl-amine (3). 3.5 g (11 mmol) of compound 12
was dissolved in 35 mL of dimethylformamide and
0.76 g (25 mmol) of 80% NaH in mineral oil were
added. The mixture was heated at 50 ꢀC for 30 min.
After cooling to room temperature 1.66 g (1.1 mL,
40 mmol) of methyl iodide was added. The mixture was
heated at 50 ꢀC for 1 h, cooled and concentrated. The
residue was partitioned between 50 mL of dichloro-
methane and 50 mL of water and the layers were sepa-
rated. The organic phase was dried over sodium sulfate,
®ltered, and concentrated. The residue was puri®ed over
silica gel, eluent dichloromethane:ethanol (49:1, v/v). The
product was recrystallized as the hydrochloride from a
mixture of methanol anꢀd diethylether giving 2.9 g of 3:
Yield 80%; mp 109±110 C; 1H NMR (DMSO-d6) d 0.79
(m, 2H), 0.93 (m, 2H), 1.15 (d, 3H), 2.9 (m, 1H), 3.65
(m, 4H), 3.80 (m, 4H); MS M+ 275; Anal. (C14H21N5O
The replacement of the pre-cited substituents with a pro-
pyl or a cyclobutyl group into the 4 position (R2) or with a
morpholine ring into the 6 position (R3) lead to com-
pounds (1, 10, 11, 13) slightly less potent than 8 and 14
(according to their EC50 values), but less ecacy during
their ®rst relaxant phase (according to the percentage of
maximal relaxation achieved). All the other modi®cations
listed in Table 1 are associated with a great decrease on the
relaxant and PDE 4 inhibition parameters.
Since the triazine derivatives 8 and 14 are more selective
inhibitors of the type 4 PDE (8 and 14 platelet PDE 3
inhibitory activity33: IC50>100 mM, versus 0.14 and
0.15 mM for the PDE 4 inhibitory activity), our results
con®rm that the PDE 4 isoenzyme is functionally
important in modulating the tracheal smooth muscle
tone in the guinea pig. It should be noted that selective
inhibitors of the type 3 PDE relax guinea pig trachea in
vitro.4,32 However, PDE 3 inhibitors have been shown
to induce positive inotropism and vasodilation whereas
inhibition of type 4 PDE is not associated directly with
such cardiovascular e€ects.3 Thus, type 4 PDE inhibi-
tors as 8 and 14 may be more selective and useful for the
treatment of asthma.
.
HCl).
(4-Cyclopropyl-6-morpholin-4-yl-[1,3,5]triazin-2-yl)-dicy-
clopropylmethyl-amine (4). 2.4 g (10 mmol) of 2-chloro-
4-cyclopropyl-6-morpholin-4-yl-[1,3,5]triazine,26 1.1 g (10
mmol) of dicyclopropylmethylamine (prepared from
dicyclopropylketone and hydroxylamine and sub-
sequent reduction of the imine with sodium metal in
ethanol)34 and 1.5 mL of triethylamine in 100 mL of
dioxane were heated at re¯ux (100 ꢀC) for 3 h. The mix-
ture was cooled to room temperature and concentrated
under vacuum. The residue was partitioned between
50 mL of dichloromethane and 50 mL of water. The
layers were separated and the organic phase was dried
over sodium sulfate, ®ltered, and concentrated. The
residue was puri®ed over silica gel, eluent ethylacetate-
n-hexane (1:9, v/v). The product was recrystallized as
the hydrochloride from a mixture of methanol and ethyl
acetate giving 1.3 g of 4: Yield 37%; 1H NMR (DMSO-
d6) d 0.3±0.6 (m, 8H), 1.00±1.3 (m, 6H), 1.95 (m, 1H),
3.21 (m, 1H), 3.6 (m, 4H), 3.77 (m, 4H), 8.41 (d, 1H);
Experimental
Chemistry
1H NMR spectra were recorded on a Bruker AM-250
spectrometer unless otherwise stated. Low-resolution
mass spectra were recorded with a Kratos or a Finnigan
TSQ 700 mass spectrometer. Normal-phase silica gel
(Merck, silica gel 60 F254) was used for chromato-
graphy. Products were puri®ed by preparative HPLC
over silica gel (Merck, silica gel 60, size 0.014±
0.040 mm).
MS M+ 315; Anal. (C17H25N5O HCl).
.
6,N-Dicyclopropyl-N0,N0-bis(2-methoxyethyl)-[1,3,5]tria-
zine-2,4-diamine (6). 6.3 g (0.03 mol) of 2-chloro-4-
cyclopropyl-6-cyclopropylamino-[1,3,5]triazine26 and 8 g
(0.06 mol) of bis(2-methoxyethyl)-amine in 100 mL of
dioxane were heated at 100 ꢀC for 2 h. The brown mix-
ture was cooled to room temperature, concentrated
under vacuum, and taken up in 100 mL of dichlor-
omethane. After washing with water (2Â100 mL) and
drying over sodium sulfate, the solution was con-
centrated leaving 10g of product that was puri®ed over
silica gel using eluent dichloromethane:ethanol (49:1,
v/v). After crystallization from n-hexane there were
obtained 5.5 g of 6: Yield 59.7%; mp 64±66 ꢀC; 1H NMR
(CDCl3) d 0.50 (m, 2H), 0.73 (m, 2H), 0.84 (m, 2H), 1.05
(m, 2H), 1.76 (m, 1H), 2.74 (m, 1H), 3.35 (s, 6H), 3.56
(m, 4H), 3.76 (m, 4H), 5.13 (s, 1H); MS M+ 307; Anal.
(C15H25N5O2).
Compounds 1, 5, 8, 10, 11, 13, 14, and 15 have been
described previously.26
4-Cyclopropyl-6-morpholin-4-yl-[1,3,5]triazin-2-yl-amine
(2). 2.69 g (15.7 mmol) of 2-amino-4-chloro-6-cyclopro-
pyl-[1,3,5]triazine (prepared from 2,4-dichloro-6-cyclo-
propyl-[1,3,5]triazine and aqueous ammonia in
acetone25) and 3 g (35 mmol) of morpholine in 50 mL of
dioxane were heated at 100 ꢀC for 2 h. The mixture was
cooled to room temperature, concentrated under
vacuum and partitioned between 50 mL of dichlor-
omethane and 50 mL of water. The layers were sepa-
rated and the organic phase was dried over sodium
sulfate, ®ltered and concentrated. The residue was pur-
i®ed over silica gel, eluent dichloromethane:ethanol
(97.5:2.5, v/v). The product 2 was isolated as the hy-
drochloride from ethanol and diethylether: Yield 3.25 g
2-(4-Cyclopropyl-6-cyclopropylamino-[1,3,5]triazin-2-yl-
amine)-acetamide (7). 6.3 g (0.03 mol) of 2-chloro-4-
cyclopropyl-6-cyclopropylamino-[1,3,5]triazine,26 4.42 g
(0.04 mol) of glycinamide hydrochloride, and 11 g
(80.4%); mp 203±204 ꢀC; H NMR (DMSO-d6) d 1.22
1
(m, 4H), 2.00 (m, 1H), 3.64 (m, 4H), 3.77 (m, 4H), 7.5±
8.7 (broad, 3H); MS M+ 221; Anal. (C10H15N5O HCl).
.

514
F. Leroux et al. / Bioorg. Med. Chem. 7 (1999) 509±516
(0.08 mol) of potassium carbonate in 70 mL of iso-
propanol were heated at re¯ux (80 ꢀC) for 50 h. The
mixture was cooled to room temperature and con-
centrated under vacuum. The residue was partitioned
between 200 mL of dichloromethane and 200 mL of
water. The solids were removed by ®ltration and the
layers were separated. The organic phase was dried over
sodium sulfate, ®ltered, and concentrated. The residue
was puri®ed over silica gel using eluent dichloromethane:
ethanol (9:1, v/v). The product was recrystallized from a
mixture of ethyl acetate and n-hexane giving 1.86g of 7:
Yield 25%; mp 187±188 ꢀC; 1H NMR (DMSO-d6) d 0.46
(m, 2H), 0.62 (m, 2H), 0.84 (m, 2H), 0.93 (m, 2H), 1.68 (m,
1H), 2.72 (m, 1H), 3.76 (m, 2H), 6.88 (broad s, 2H), 7.14
(broad s, 2H); MS M+ 248; Anal. (C11H16N6O).
carboximidamide hydrochloride26 in 1 L of methanol (dis-
tilled from sodium) was added a solution of 15.2 g
(0.66 mol) of sodium in 500 mL of methanol. The mix-
ture was stirred for 30 min at room temperature,
51.3 mL (0.36 mol) of ethyl succinylchloride was added
dropwise with stirring over 30 min while maintaining the
temperature of the mixture between 25±35 ꢀC. The mix-
ture was heated at re¯ux overnight. After cooling and
®ltering, the mixture was concentrated to dryness, leav-
ing a syrup. The product was puri®ed over silica gel
using eluent dichloromethane:methanol (49:1, v/v)
yielding 18.5 g (18.2%) of the ester as an oil that crys-
tallized upon standing. 10.2 g (0.033 mol) of the product
was dissolved in 1 L of methanol, the mixture was
cooled to 15 ꢀC and ammonia gas was bubbled through
the solution for 5 days. The mixture was concentrated and
the residue was puri®ed over silica gel, eluent dichlor-
omethane:methanol (9:1, v/v) and crystallization from a
mixture of methanol and diethylether gave 5.9 g of 12 as a
solid: Yield 61.2%; mp 217±218 ꢀC; 1H NMR (DMSO-d6)
d 0.47 (m, 2H), 0.64 (m, 2H), 2.43 (m, 2H), 2.61 (m, 2H),
2.75 (m, 1H), 3.60 (m, 4H), 3.71 (m, 4H), 6.64 (s, 1H), 7.22
(s, 1H), 7.36 (s, 1H); MS M+ 292; Anal. (C13H20N6O2).
Cyclopropyl-[4-cyclopropyl-6-(2,2-dimethyl-1,1-dioxo-1ꢀ6-
thiazolidin-3-yl)-[1,3,5]triazin-2-yl]-amine (9). 22.4 g (146
mmol) of 2,2-dimethylthiazolidine hydrochloride (pre-
pared from 2-aminoethanethiol and acetone) were dis-
solved in 50 mL of water and 4 N NaOH was added
until the pH of the solution reached between 8 and 9
and the free base was extracted with dichloromethane
(3Â200 mL). The combined organic phases were dried
over magnesium sulfate, ®ltered, and concentrated, giv-
ing an oily residue of 15.5 g (132mmol) of 2,2-dimethyl-
thiazolidine. This residue was dissolved in 200 mL of
dioxane and 13.94 g (66.2 mmol) of 2-chloro-4-cyclo-
propyl-6-cyclopropylamino-[1,3,5]triazine26 were added.
The mixture was re¯uxed for 35h, cooled to room
temperature, and concentrated under vacuum. The resi-
due was partitioned between 200 mL of dichloro-
methane and 200 mL of water. The layers were
separated and the organic phase was dried over magne-
sium sulfate, ®ltered, and concentrated. The yellow
residue (24 g) was chromatographed on silica gel, eluent
dichloromethane: methanol (99:1, v/v). The product was
recrystallized from a mixture of dichloromethane and
n-hexane giving 7.2 g of the thiazolidine precursor of 9
with a yield of 37.4%. This material was used as such in
the next step.
Pharmacology
Materials. Guinea pigs and rats were obtained from
Charles River (Saint Aubin les Elbeuf, France) and I€a
Credo (Brussels, Belgium), respectively. Isoprenaline
and the nucleotidase from snake venom (Crotalus Atrox
V-7 000) were obtained from Sigma (Saint-Quentin-
Fallavier, France). Hank's balanced salt solution,
[2,8-3H]cAMP, AG1-X2 anion exchange resin and Ulti-
magold-RX were purchased, respectively, from Gibco-
Life Technologies (Merelbeke, Belgium), NEN Dupont
(Brussels, Belgium), BioRad (Nazareth Eke, Belgium),
and Canberra-Packard (Zellik, Belgium). Rolipram and
RP73401 were synthesized by UCB Pharma Sector
(Braine-l'Alleud, Belgium). All other reagents were of
the highest available grade and purchased from stan-
dard sources.
6.6 g (22.7 mmol) of the above obtained material, dis-
solved in 300 mL of dichloromethane:methanol (1:2, v/v),
was oxidized by dropwise addition during 15 min of
Tracheal relaxant activity. Tracheal relaxant activity
was measured as previously described with slight mod-
i®cations.35 Male albino guinea pigs weighing 350±450 g
were anesthetized by an intraperitoneal injection of
pentobarbital (50 mg kg 1). The trachea was removed,
trimmed of excess tissue, and opened longitudinally by
cutting through the cartilage. Six transverse segments,
each containing 3 to 4 cartilaginous rings, were pre-
pared. Each strip was suspended in a 15 mL tissue bath
and attached via a bristle thread to a force displacement
transducer connected to a polygraph for isometric
recording of tension. The smooth muscle was bathed in
modi®ed Krebs±Henseleit solution of the following com-
position: NaCl, 120 mM; KCl, 4.75mM; CaCl2, 2.5 mM;
.
.
27.9 g (45.4 mmol) of oxone (2KHSO5 KHSO4 K2SO4) in
100 mL of water at 15ꢀC. The mixture was stirred for
24h at room temperature. Water (1L) was added and the
mixture was extracted with dichloromethane (3Â1 L).
The combined washings were dried over magnesium sul-
fate, ®ltered, and concentrated. The residue (7 g) was
chromatographed on silica gel using eluent dichloro-
methane:methanol (99:1, v/v). The product was recrys-
tallized from a mixture of ethylacetate and n-hexane
giving 4.65g of 9: Yield 63.5%; mp 164±165 ꢀC; 1H NMR
(DMSO-d6) d 0.53 (m, 2H), 0.76 (m, 2H), 0.91 (m, 2H),
1.08 (m, 2H), 1.74±1.85 (m+s, 7H), 2.72 (m, 1H), 3.20 (t,
2H, J=7 Hz), 4.04 (t, 2H, J=7Hz), 5.31 (s, 1H); MS M+
323; Anal. (C14H21N5O2S).
.
MgSO4 7H2O, 1.2 mM; KH2PO4, 1.15mM; NaHCO3,
25 mM; glucose, 10 mM. Tissue baths were maintained at
37 ꢀC and constantly bubbled with air: 5% CO2 mixture.
The tissues were allowed to equilibrate for 90min with
frequent washings under a resting tension of 1.5 g, which
was found optimal for determining changes in tension.
Concentration-response curves were constructed by
3-(4-Cyclopropylamino-6-morpholin-4-yl-[1,3,5]triazin-2-
yl-propionamide (12). To a solution of 82 g (0.33 mol)
of N-[imino(cyclopropylamino)methyl]-4-morpholine-

F. Leroux et al. / Bioorg. Med. Chem. 7 (1999) 509±516
515
successive increases in the bath concentration of agent.
At the end of each experiment, a supramaximal con-
centration of isoprenaline (1 mM) was administered and
relaxant responses were expressed as a percentage of the
maximal isoprenaline relaxation to minimize variability
between tissues. Only one cumulative response curve
was obtained on a single tissue to avoid problems of
previous treatment. EC50 values were calculated by lin-
ear regression of concentration-response curves gener-
ated by concentrations that caused signi®cant e€ect
between 10±90%. Rolipram, RP73401 and triazine
derivatives were dissolved in dimethyl sulfoxide, and the
®nal concentration of dimethyl sulfoxide used in each
experiment was below 0.2% at which the solvent did not
in¯uence assays. Isoprenaline was dissolved in deionized
water.
These solutions were diluted with bu€er containing
Tris-HCl pH 8, 50 mM, MgCl2, 1 mM and dithio-
threitol, 5 mM, so that dimethyl sulfoxide ®nal con-
centration was in any case equal to 0.1% v/v.
References and Notes
1. Hall, I. P. Br. J. Clin. Pharmacol. 1993, 35, 1.
2. Beavo, J. A.; Conti, M.; Heaslip, R. J. Mol. Pharmacol.
1994, 46, 399.
3. Nicholson, C. D.; Shahid, M. Pulm. Pharmacol. 1994, 7, 1.
4. Souness, J. E.; Giembycz, M. A. In Airways smooth muscle:
biochemical control of contraction and relaxation; Raeburn, D.;
Giembycz, M. A., Eds.; Birkhauser Verlag Basel: Switzerland,
1994; pp 271±308.
5. Loughney, F.; Martins, T. J.; Harris, E. A.; Sadhu, K.;
Hicks, J. B.; Sonnenburg, W. K.; Beavo, J. A.; Ferguson, K. J.
Biol. Chem. 1996, 271, 796.
Inhibition of PDE 4 activity. Polymorphonuclear leuko-
cytes were isolated from pleural cavity of rats locally
injected with autolog serum, as described by J. P. Gir-
oud and his collaborators with slight modi®cations.36
Brie¯y the cells were collected three hours after serum
injection, washed with Hank's balanced salt solution,
made free from erythrocytes by hypotonic shock and
®nally suspended in the following bu€er: Tris-HCl pH 7.4,
6.7 mM; EGTA, 1.7 mM; MgCl2, 0.7 mM; CaCl2, 0.6
mM; KCl, 1.8 mM; NaCl, 91mM; KH2PO4, 0.98mM;
NaH2PO4, 5.4 mM; glucose, 3.7 mM; gelatine, 0.67g/L;
phenylmethylsulfonyl ¯uoride, 0.33 mM; dithiothreitol,
0.67 mM. Cells were disrupted by sonication and
centrifuged (15 000 g, 30 min, 4 ꢀC). Supernatant was
used to measure phosphodiesterase activity. Aliquots
were frozen in liquid nitrogen and preserved at 70 ꢀC.
PDE 4 activity was measured according to Kono37 and
Ong and Rennie38 with slight modi®cations. The incuba-
tions were made 20 min at 37 ꢀC in the following bu€er:
dextrose, 0.3 mM; CaCl2, 0.05 mM; KCl, 0.14 mM;
KH2PO4, 0.08 mM; Na2HPO4, 0.43 mM; NaCl, 7.3 mM;
Tris, 46.5 mM; MgCl2, 9.55 mM; EGTA, 0.13 mM;
dithiothreitol, 4.65 mM; sodium acetate, 4.9 mM. cAMP
was at 1 mM (with 200 000 dpm [3H]cAMP) and di-
methyl sulfoxide 0.1% v/v. The reaction was stopped by
addition of HCl. After enzyme denaturation by heat
and neutralization to pH 7.5 with NaOH, the entire
[3H]5'cAMP produced was converted into [3H]adeno-
sine by incubation with 50nucleotidase from snake
venom and [3H]adenosine was separated from unreacted
[3H]cAMP by column chromatography on AG1-X2
anion exchange resin (0.8 mL) eluted with water.
Radioactivity was quanti®ed in a liquid scintillation
counter after addition of Ultimagold-XR. Percentage of
hydrolyzed cAMP must be included between 6±25%.
Inhibition by 0.1 mM rolipram tested simultaneously to
the assay must be above 33%. The radioactivity mea-
sured without enzyme must be below 8000 dpm. IC50
were calculated using the program NLIN SAS by non
linear ®tting to the equation dpm=a/[1+(x/IC50)nH]+c
de®ning a classical sigmoid curve where a and c repre-
sent respectively the maximum and the minimum dpm
plateau values and x and IC50 represent respectively the
concentrations (mol/L) and the IC50 of the studied
molecules; nH is the Hill number. Starting solutions of
drugs to be tested were made in dimethyl sulfoxide.
6. Houslay, M. D.; Sullivan, M.; Bolger, G. B. Adv. Pharma-
col. 1998, 44, 225.
7. Holgate, S. T. Eur. Respir. J. 1993, 6, 1507.
8. Pare, P. D. Chest 1995, 107, 163S.
9. De Boer, J.; Philpott, A. J.; Van Amsterdam, R. G. M.;
Shahid, M.; Zaagsma, J.; Nicholson, C. D. Br. J. Pharmacol.
1992, 106, 1028.
10. Rabe, K. F.; Tenor, H.; Dent, G.; Schudt, C.; Liebig, S.;
Magnussen, H. Am. J. Physiol. 1993, 264, L458.
11. Torphy, T. J.; Undem, B. J.; Cieslinski, L. B.; Luttmann,
M. A.; Reeves, M. L.; Hay, D. W. P. J. Pharmacol. Exp. Ther.
1993, 265, 1213.
12. Miyamoto, K. I.; Kurita, M.; Sakai, R.; Sanae, F.;
Wakusawa, S.; Takagi, K. Biochem. Pharmacol. 1994, 48,
1219.
13. White, J. R.; Torphy, T. J.; Christensen, S. B.; Lee, J. A.;
Mong, S. FASEB J. 1990, 4, A1987.
14. Wright, C. D.; Kuipers, P. J.; Kobylarz-Singer, D.;
Devall, L. J.; Klinkefus, B. A.; Weishaar, R. E. Biochem.
Pharmacol. 1990, 40, 699.
15. Dent, G.; Giembycz, M. A.; Rabe, K. F.; Barnes, P. J. Br.
J. Pharmacol. 1991, 103, 1339.
16. Souness, J. E.; Carter, C. M.; Diocee, B. K.; Hassall, G.
A.; Wood, L. J.; Turner, N. C. Biochem. Pharmacol. 1991, 42,
937.
17. Kelly, J. J.; Barnes, P. J.; Giembycz, M. A. Br. J. Phar-
macol. 1998, 124, 129.
18. Ashton, M. J.; Cook, D. C.; Fenton, G.; Karlsson, J.-A.;
Palfreyman, M. N.; Raeburn, D.; Ratcli€e, A. J.; Souness, J.
E.; Thurairatnam, S.; Vicker, N. J. Med. Chem. 1994, 37, 1696.
19. Raeburn, D.; Underwood, S. L.; Lewis, S. A.; Woodman,
V. R.; Battram, C. H.; Tomkinson, A.; Sharma, S.; Jordan,
R.; Souness, J. E.; Webber, S. E.; Karlsson, J.-A. Br. J. Phar-
macol. 1994, 113, 1423.
20. Barnes, P. J. Eur. Respir. J. 1995, 8, 457.
21. Souness, J. E.; Maslen, C.; Webber, S.; Foster, M.; Raeburn,
D.; Palfreyman, M. N.; Ashton, M. J.; Karlsson, J.-A. Br. J.
Pharmacol. 1995, 115, 39.
22. Santamaria, L. F.; Palacios, J. M.; Beleta, J. Br. J. Phar-
macol. 1997, 121, 1150.
23. Blease, K.; Burke-Ga€ney, A.; Hellewell, P. G. Br. J.
Pharmacol. 1998, 124, 229.
24. Hirt, R.; Nidecker, H.; Berchtold, R. Helv. Chim. Acta
1950, 33, 1365.
25. Nishigaki, S.; Yoneda, F.; Matsumoto, H.; Morinaga, K.
J. Med. Chem. 1969, 12, 39.
26. Van Keulen, B. J.; Goldstein, S.; Cossement, E.; Gobert,
J.; Wulfert, E. European Patent 0527117, 1992.
27. Shapiro, S. L.; Parrino, V. A.; Rogow, B.; Freedman, L. J.
J. Amer. Chem. Soc. 1959, 81, 3725.

516
F. Leroux et al. / Bioorg. Med. Chem. 7 (1999) 509±516
28. Shapiro, S. L.; Parrino, V. A.; Freedman, L. J. J. Amer.
Chem. Soc. 1959, 81, 3728.
29. Shapiro, S. L.; Parrino, V. A.; Freedman, L. J. J. Amer.
Chem. Soc. 1959, 81, 3996.
30. Sciortino, T.; Maurich, V.; Gratton, S. Boll. Chim. Farm.
1977, 116, 637.
31. Trost, B. M.; Curran, D. P. Tetrahedron Lett. 1981, 22, 1287.
32. Harris, A. L.; Connell, M. J.; Ferguson, E. W.; Wallace,
A. M.; Gordon, R. J.; Pagani, E. D.; Silver, P. J. J. Pharmacol.
Exp. Ther. 1989, 251, 199. See also: Moore, J. B.; Combs, D.
W.; Tobia, A. J. Biochem. Pharmacol. 1991, 42, 679.
33. Data not shown. The PDE 3 activity from human platelets
was measured in a similar way to the method described for the
PDE 4.
34. Arumugam, H.; Shenbagamurthi, P. Tetrahedron Lett.
1972, 2251.
35. Leroux, F.; Goossens, J.-F.; Pommery, N.; Henichart,
J.-P. Regul. Pept. 1994, 52, 119.
36. Giroud, J. P.; Roch-Arveiller, M.; Muntaner, O. Nouv.
Â
Rev. Fr. Hematol. 1978, 20, 535.
37. Kono, T. Methods Enzymol. 1988, 159, 745.
38. Ong, K. K.; Rennie, P. I. C. Anal. Biochem. 1976, 76, 53.