Synthesis and β-blocking activity of (R,S)-(E)-oximeethers of 2,3-dihydro-1,8-naphthyridine and 2,3-dihydrothiopyrano[2,3-b]pyridine: Potential antihypertensive agents - Part IX

Article abstract of DOI:10.1016/S0223-5234(00)00173-2

The synthesis of oximeethers of 2,3-dihydro-1,8-naphthyridine and 2,3-dihydrothiopyrano[2,3-b]pyridine is described. These compounds exhibit a selective β-blocking activity, with a selectivity towards β₂-receptors. Groups in the N₁ position giving rise to a considerable steric hindrance led to a higher β₂-blocking selectivity, whereas groups creating a moderate hindrance caused a weak but significant decrease in β₂-antagonist potency. Substitution of the N₁-R group with a sulfur atom led to compounds possessing β₁-, β₂- and β₃-blocking properties. Compounds 9c₁ and 10a₁ showed a β₃-antagonist activity slightly lower than that of propranolol. (C) 2000 Editions scientifiques et medicales Elsevier SAS.

Full text of DOI:10.1016/S0223-5234(00)00173-2

Eur. J. Med. Chem. 35 (2000) 815−826
© 2000 Éditions scientifiques et médicales Elsevier SAS. All rights reserved
S0223523400001732/FLA
Original article
Synthesis and ꢀ-blocking activity of (R,S)-(E)-oximeethers of 2,3-dihydro-
1,8-naphthyridine and 2,3-dihydrothiopyrano[2,3-b]pyridine:
potential antihypertensive agents – Part IX
Pier Luigi Ferrarinia*, Claudio Moria, Muwaffag Badawnehb, Vincenzo Calderonec, Rosamiria
Grecoc, Clementina Maneraa, Adriano Martinellia, Paola Nieric, Giuseppe Saccomannia
aDipartimento di Scienze Farmaceutiche, Università di Pisa, via Bonanno 6, 56126 Pisa, Italy
bPhiladelphia University, P.O. Box 1101 Sweileh, Jordan
cDipartimento di Psichiatria, Neurobiologia, Farmacologia e Biotecnologie, Università di Pisa, sez. via Bonanno 6, 56126 Pisa, Italy
Received 8 June 1999; revised and accepted 22 March 2000
Abstract – The synthesis of oximeethers of 2,3-dihydro-1,8-naphthyridine and 2,3-dihydrothiopyrano[2,3-b]pyridine is described. These
compounds exhibit a selective ꢀ-blocking activity, with a selectivity towards ꢀ2-receptors. Groups in the N1 position giving rise to a
considerable steric hindrance led to a higher ꢀ2-blocking selectivity, whereas groups creating a moderate hindrance caused a weak but
significant decrease in ꢀ2-antagonist potency. Substitution of the N1-R group with a sulfur atom led to compounds possessing ꢀ1-, ꢀ2- and
ꢀ3-blocking properties. Compounds 9c1 and 10a1 showed a ꢀ3-antagonist activity slightly lower than that of propranolol. © 2000 Éditions
scientifiques et médicales Elsevier SAS
ꢀ1-, ꢀ2- and ꢀ3-adrenergic blockers / oximeethers / 1,8-naphthyridine derivatives / thiopyrano[2,3-b]pyridine derivatives
1. Introduction
In a previous paper [1], we described the synthesis of
a series of (R,S)-(E)- and (R,S)-(Z)-oximeethers of 1,8-
naphthyridine 1 and 2 (figure 1) and their ꢀ2-stimulating
and ꢀ1-blocking properties as evaluated in vitro on
isolated preparations. Recently, we have synthesized the
(R)- and (S)- enantiomers of oximeethers 1 and 2 in order
Figure 1. Structures of (R, S)-(E)- and (R, S)-(Z)-4(1H)oxi-
to examine their behaviour towards ꢀ-adrenergic recep-
meether-2,3-dihydro-1,8-naphthyridines.
tors. The enantiomeric compounds obtained showed an
interesting ꢀ-blocking activity: compounds with bromine
in positions 6 and/or 7 exhibited pKb values for the
ꢀ1-adrenoceptor similar to that of practolol [2].
On the basis of these results, it appeared to be of
interest to continue our chemical and pharmacological
investigation in this field. We therefore decided to prepare
a new series of oximeethers of 1,8-naphthyridine with an
7-position, to study the influence of these substituents on
the activity versus ꢀ-adrenoceptor subtypes. We chose the
latter substituent, because, in previous papers, the
oximeethers 1a and 2a (R = CH3, R1 = H) were consid-
ered as lead compounds [1, 2].
N-alkyl group or a sulfur atom in the 1-position of the
Moreover, we decided to prepare only the (R,S)-(E)-
1,8-naphthyridine nucleus and with a methyl group in the
oximeethers, because the (R)- and (S)-enantiomers of 1
and 2 exhibited no chiral preference, and the (E)-isomer
1 generally presented a slightly higher affinity for both
* Correspondence and reprints: ferrarini@farm.unipi.it

816
P.L. Ferrarini et al. / Eur. J. Med. Chem. 35 (2000) 815–826
ꢀ-adrenoceptor subtypes with a slight selectivity for
ꢀ2-receptors [2].
As compounds binding to ꢀ3-adrenoceptors, whose
propanolamine side chain is attached to the oxygen of an
oxime function, have not been prepared so far [3], we
decided to examine this series of compounds for their
ꢀ3-activity.
2. Chemistry
The ketones 3 were prepared following a new, less
complicated route than that reported by us in a previous
paper [4].
Starting from 2-amino-6-methylpyridine 4, the alky-
lamines 5a [5], 5b [6], 5c [7] and 5d (figure 2) were
obtained by reductive amination with sodium borohy-
dride and the appropriate aldehyde. The subsequent
reaction with ethyl acrylate, in the presence of glacial
acetic acid, led to derivatives 6. The corresponding crude
acids, obtained by alkaline hydrolysis of 6, were then
allowed to react in polyphosphoric acid (PPA) to give the
target compounds 3a–c [4] and 3d (figure 2, tables I and
II). The ketones 3 were converted, by refluxing with
hydroxylamine hydrochloride into a mixture of anhy-
drous ethanol and pyridine, to yield the pure (E)-oximes
7 (figure 3, tables I and II), as demonstrated by 1H-NMR
analysis and in agreement with a previous report [1]. The
sodium salt of the (E)-oximes 7 was allowed to react with
epichlorohydrin in anhydrous toluene to give the (E)-
epoxides 8 (figure 3, tables I and III). Under these
conditions, the corresponding (Z)-epoxides are not ob-
tained, as reported in a previous paper [1]. The ring-
opening addition reaction of epoxides 8 with the suitable
amine carried out at room temperature in acetonitrile and
catalysed by lithium perchlorate, as reported in a previous
paper [2], gave the target compounds 9 in a good yield
(figure 3, table IV and V). It was also of interest to prepare
oximeethers of thiopyrano[2,3-b]pyridine 10, in which
the N-R group, present in the 1-position of the 1,8-
naphthyridine nucleus, was substituted by a sulfur atom
to establish the influence of this part of the structure as
regards the ꢀ-activities. As illustrated in figure 4, the
2-mercapto-6-methylpyridine 11 [8], was allowed to react
with ethyl acrylate, in the presence of glacial acetic acid
to give the derivative 12. The corresponding acid 13 was
then obtained by alkaline hydrolysis. Compound 13 was
also obtained by reaction of the potassium salt of 11 with
sodium chloropropionate. The cyclization of 13 in PPA at
170 °C gave the expected ketone 14 in a moderate yield
(figure 4, tables I and II).
Figure 2. Reagents: A) NaBH4, MeOH, (HCHO, CH3CHO,
PhCHO, PhCH2CHO); B) Ethyl acrylate, AcOH(glac.); C) NaOH
(6N) / EtOH.
respectively. The ring-opening addition reaction of ep-
oxide 16 with the appropriate amine, carried out at 80 °C
in toluene, led to the target compounds 10 (figure 5,
tables IV and V).
The structures assigned were fully confirmed by el-
1
emental analysis, IR and H-NMR spectra.
1
The H-NMR spectrum of ester 6b is quoted as an
example: δ 1.22 (CH3, 6H, m), 2.35 (6-CH3, 3H, s), 2.63
(C–CH2, 2H, t), 3.77 (CH2–CH2, 2H, t), 3.48 (N–CH2,
2H, q); 4.11 (O–CH2, 2H, q), 6.29 (H3 + H5, 2H, m); 7.26
(H4, 1H, m).
1
The esters 6 all exhibit analogous H-NMR spectra.
As reported in previous papers for similar compounds
[9–11], the 1H-NMR spectra of compounds 3d, 7, 14 and
15 (table II) show two doublets ranging from δ 6.37–6.95
and δ 7.80–8.15, due to H6 and H5, respectively, and two
multiplets ranging from δ 2.82–2.99 and δ 3.20–3.40,
due to H3 and H2, respectively. The (E)- conformation
was assigned to oximes 7 and 15, epoxides 8 and 16
(table III) and the final compounds 9 and 10 (table V), on
the basis of their 1H-NMR spectra, which show the
chemical shift of H5, ranging from δ 7.81–8.07, close to
that of ketones 3 and 14. This behaviour was analogous to
that of similar compounds previously reported [1].
The oxime 15 and epoxide 16 (figure 5, tables I–III)
were prepared as reported above for compounds 7 and 8,

P.L. Ferrarini et al. / Eur. J. Med. Chem. 35 (2000) 815–826
Table I. Physical data of ketones, (E)-oximes and (R,S)-(E)-epoxides.
817
Compound
X
Y
Flash chromatography eluent
Yield (%)
M.p. (°C)
3a
3b
3c
3d
7a
7b
7c
7d
8a
NCH3
NC2H5
NCH2Ph
N(CH2)2Ph
NCH3
O
O
O
AcOEt
AcOEt
AcOEt
68
51
61
40
85
87
88
52
67
[4]
[4]
[4]
O
AcOEt/pet. ether (1:4)
oil
NOH
NOH
NOH
NOH
–
–
–
128–130a
132–135a
142–145a
oil
NC2H5
NCH2Ph
N(CH2)2Ph
NCH3
AcOEt/pet. ether (1:4)
AcOEt/pet. ether (1:6)
oil
8b
8c
8d
NC2H5
NCH2Ph
AcOEt/pet. ether (1:6)
AcOEt/pet. ether (1:6)
AcOEt/pet. ether (1:9)
72
60
36
oil
oil
oil
N(CH2)2Ph
14
15
16
S
S
S
O
NOH
AcOEt/pet. ether (2:3)
42
86
48
90–92b
234–236c
oil
–
AcOEt\pet. ether (3:2)
a Petroleum ether 100–140 °C; b cyclohexane; c MeOH.
3. Theoretical calculations
The only difference between the MEPs of 1a3 and 10a3 is
found in the region where the molecules are structurally
different: in compound 1a3, a negative MEP is localized
above and below the ring plane in the region of the N1,
whereas in compound 10a3, an analogous negative MEP
region is localized in the plane of the ring in the region of
the sulphur atom.
The molecular electrostatic potential (MEP) of the
N-unsubstituted 1a3 and 10a3 analogues of 1a1 [1] and
10a1 (figure 6) were calculated, in accordance with the
method used for other adrenergic drugs [12]. Compounds
1a3 and 10a3 were considered in their free base form, and
their conformation was optimized at the semi-empirical
PM3 level [13]. In the case of compound 1a3 it was also
taken into consideration that it could exist in the tauto-
meric form (A) and (B) (figure 7), but a PM3 calculation
including full geometry optimisation indicated that form
(B) was not favoured by more than 10 kcal/mol.
The MEP of 1a3 and 10a3 were calculated from ab
initio SCF STO-3G* wave functions [13] in a three-
dimensional grid around the molecules. Figure 8 shows
the isopotential MEP surfaces corresponding to a value of
–10 kcal/mol [12], indicating the molecular regions pos-
sessing a nucleophilic reactivity. The MEP of 1a3 and
10a3 is very similar in the region of the ethanolaminic
and the oximeether moiety as well as in that of the
pyridinic nitrogens, which should indicate that the two
molecules possess a similar reactivity in these regions.
4. Pharmacological results
The pharmacological results are shown in table VI.
Compound 1a1 [1], showing a ꢀ1-, ꢀ2- and ꢀ3-blocking
activity, was taken as the lead compound.
Many of the compounds tested, 9a1–9c2 and 9d1–10a2,
exhibited a ꢀ-blocking activity, but showed a different
selectivity profile toward the ꢀ1-, ꢀ2- or ꢀ3-subtypes. The
molecules with the 1,8-naphthyridine moiety, 9a1–9c2
and 9d1, 9d2, were more selective as ꢀ2-antagonists than
as ꢀ1-antagonists. Furthermore, it was observed that the
presence of steric hindrances in the N1-position abolished
the ꢀ1-activity. Nevertheless, the N1-substitutions giving
rise to a considerable steric hindrance (benzyl or phenyl-

818
P.L. Ferrarini et al. / Eur. J. Med. Chem. 35 (2000) 815–826
Table II. 1H-NMR chemical shifts (δ) of ketones and (E)-oximes.
Compound
X
Y
O
H2 (m)
3.40
H3 (m)
2.93
H5 (d)
7.90
H6 (d)
6.45
7-CH3 (s)
2.42
Other
3d
N(CH2)2Ph
1-CH2: 3.89 (t);
CH2: 2.52 (t) Ph:
7.22 (m)
7a
7b
NCH3
NC2H5
NOH
NOH
3.20
3.27
2.99
2.83
7.81
7.86
6.41
6.37
2.40
2.37
1-CH3: 3.10 (s)
1-CH2: 3.69 (q);
CH3: 1.15 (t)
1-CH2: 4.88 (s);
Ph: 7.25 (m)
1-CH2: 3.78 (t),
CH2: 2.79 (t) Ph:
7.23 (m)
7c
NCH2Ph
NOH
NOH
3.22
3.24
2.82
2.86
7.82
7.80
6.47
6.41
2.37
2.41
7d
N(CH2)2Ph
14
15
S
S
O
NOH
3.28
3.26
2.91
2.99
8.15
8.00
6.93
6.95
2.53
2.38
–
N–OH: 11.40 (s)
ethyl group 9c1, 9c2, 9d1 and 9d2) led to higher ꢀ2-
blocking properties (figure 9). On the other hand, the
presence of a moderate steric hindrance in position N1 of
the 1,8-naphthyridine heterocyclic system (methyl or
ethyl substituents, 9a1–9b2), caused a weak, but signifi-
cant, decrease in the ꢀ2-antagonist potency. Interestingly,
the thiopyranopyridine derivatives showed both ꢀ1- and
ꢀ2-blocking properties. The i-Pr substituted compound
10a2 blocked the two receptor subtypes with almost
similar potencies. The t-Bu substituted compound 10a1
showed a ꢀ1-blocking activity (figure 10), together with a
weak, albeit significant, ꢀ1-selectivity. Furthermore its
potency was similar to that exhibited by the known
antagonist practolol.
Compounds 1a1, 9a1, 9b1, 9c1, 9c3, 9c4 and 10a1 were
assayed for agonistic and antagonistic activity on ꢀ3-
adrenoceptors of rat white adipose tissue; isoprenaline
and propranolol were taken as reference drugs.
All the compounds were tested in a concentration
range of 10 nM–0.1 mM, except 9c1, for which an
interference with the spectrophotometric determination
was observed at the concentration of 0.1 mM. Com-
pounds 1a1, 9a1, 9b1, 9c1, 9c3, 9c4 and 10a1, showed
ꢀ3-blocking properties but with significantly different
potencies. Compounds 9c1 and 10a1 showed the highest
potency, since they blocked isoprenaline-stimulated li-
polysis at lower concentrations in comparison with the
remaining compounds tested. Furthermore, their potency
values were slightly lower than that of propranolol, the
non-selective antagonist taken as the reference drug [14].
All the remaining compounds exhibited lower potencies
Figure 3. Reagents: A) H2NOH HCl, EtOH, Py; B) EtONa/
EtOH; C) Epichlorohydrin, toluene; D) (t-butylamine,
i-propylamine, a-methylbenzylamine or a-methylphenethyl-
amine), LiClO4 CH3CN.

P.L. Ferrarini et al. / Eur. J. Med. Chem. 35 (2000) 815–826
Table III. 1H-NMR chemical shifts (δ) of (R,S)-(E)-epoxides.
819
Compound
X
H2 (m)
H3 (m)
H5 (d)
H6 (d)
7-CH3 (s) a-CH2 (m) ꢀ-CH (m) γ-CH2 (m)
Other
8a
8b
NCH3
NC2H5
3.21
3.22
2.75
2.79
7.82
7.86
6.42
6.37
2.38
2.37
4.18
4.19
3.17
3.26
2.95, 2.55 1-CH3: 3.07 (s)
2.96, 2.61 1-CH2: 3.67 (q)
CH3: 1.14 (t)
8c
NCH2Ph
3.18
3.20
2.80
2.77
7.90
7.85
6.42
6.40
2.38
2.40
4.18
4.19
3.21
3.24
2.90, 2.55 1-CH2: 4.87 (s)
Ph: 7.25 (m)
2.96, 2.58 1-CH2: 3.82 (t)
CH2: 2.30 (t) Ph:
8d
N(CH2)2Ph
7.22 (m)
16
S
3.12
2.75
8.07
6.83
2.49
4.25
3.16
3.40, 3.40
–
in blocking isoprenaline lipolysis, in comparison with the
previous drugs. In particular, the pIC50 value of the
reference compound 1a1 was found to be 5.14 ± 0.17.
The remaining compounds 9a1 and 9b1 showed the
lowest degree of activity (pIC50 ≤ 4.5), and a great
variability in their action, as indicated by the high value
of SEM. Compounds 9c3 and 9c4 did not show any
blocking activity on ꢀ3-adrenoceptors.
The loss of ꢀ1-activity for compounds 9 is explainable
by the steric hindrance due to the substituents present in
the N1-position, whereas the same steric hindrance does
not seem to have a great influence on the ꢀ2- or
ꢀ3-activity.
Moreover, the different biopharmacological profile ver-
sus ꢀ1-adrenergic receptors of 1a1 and 10a1 could be
correlated to the different MEP trend in the region of the
aromatic system corresponding to the 1-position. The
presence of the proton donor N1H group in 1a1 makes an
interaction possible with an electrophilic region of the
receptor above and/or below the ring plane, whereas the
presence of the sulphur in 10a1 moves this kind of
interaction to the ring plane.
None of the compounds tested showed any lipolytic
activity in our suspension test system.
5. Conclusions
The pharmacological functional study of (R,S)-4(1H)-
oximeethers 9 and 10 showed a ꢀ1-, ꢀ2- and ꢀ3-blocking
activity for many compounds tested. An examination of
the pKb values reported in table VI indicates that the
substitution of H in the N1 position by an alkyl group
determined the total absence of any ꢀ1-activity (com-
pounds 9), with a generally lower affinity for ꢀ2- and
ꢀ3-receptors compared with compound 1a1, and a mod-
erate selectivity towards ꢀ2-receptors. The presence of a
moderate steric hindrance in the N1-position (compounds
9a1–9b2) caused a significant decrease in the ꢀ2-antagonist
potency (maximum 14-fold) and the ꢀ3-antagonist po-
tency (maximum 10-fold). The presence of a greater
steric hindrance (compounds 9c1–9c2) led to higher ꢀ2-
and ꢀ3-blocking properties similar to those of 1a1. The
substitution of the N1H group by a sulfur atom (com-
pounds 10a1 and 10a2) leads to a good affinity versus ꢀ1-,
ꢀ2- and ꢀ3-receptors, analogous to that of compound 1a1,
but with a slight selectivity toward the ꢀ1-receptor for
compound 10a1. A further increase in the hindrance of the
side chain leads to a complete abolition of ꢀ3-activity.
6. Experimental protocols
6.1. Chemistry
All compounds were routinely checked for their struc-
1
ture by IR and H-NMR spectroscopy. Melting points
were determined on a Kofler hot-stage apparatus and are
uncorrected. The IR spectra were measured with a Gen-
esis Series FTIR ATI Mattson spectrometer. The 1H-
NMR spectra were determined in DMSO-d6 or CDCl3
with TMS as the internal standard, on a Varian CFT-20
NMR spectrometer. Analytical TLC was carried out on
Merck 0.2 mm pre-coated silica-gel glass plates (60
F-254) and location of spots was detected by illumination
with a UV lamp. Elemental analysis of all synthesized
compounds for C, H and N were within ± 0.4% of
theoretical values and were performed by our analytical
laboratory.

820
P.L. Ferrarini et al. / Eur. J. Med. Chem. 35 (2000) 815–826
Table IV. Physical data of (R,S)-(E)-oximeether derivatives.
Compound
X
R
Flash chromatography eluent
Yield (%)a
61
9a1
NCH3
t-Bu
AcOEt:DEA
(20:1)
9a2
9b1
9b2
9c1
NCH3
NC2H5
NC2H5
NCH2Ph
NCH2Ph
NCH2Ph
NCH2Ph
N(CH2)2Ph
N(CH2)2Ph
S
i-Pr
t-Bu
AcOEt:DEA
(20:1)
AcOEt:DEA
(20:1)
AcOEt:DEA
(20:1)
64
65
62
54
52
29
30
30
27
62
67
i-Pr
i-Pr
AcOEt:pet. eth:DEA
(3:1:0.1)
AcOEt:pet. eth:DEA
(3:1:0.1)
AcOEt:MeOH:DEA
(20:1:0.5)
AcOEt:MeOH:DEA
(20:1:0.5)
9c2
t-Bu
9c3
Ph(CH3)CH
PhCH2(CH3)CH
t-Bu
9c4
9d1
9d2
10a1
10a2
AcOEt:DEA
(3:0.1)
AcOEt:DEA
(3:0.1)
i-Pr
t-Bu
AcOEt:pet. eth:DEA
(3:3:0.5)
AcOEt:pet. eth:DEA
(3:3:0.5)
S
i-Pr
a Oil.
6.1.1. General procedure
72 h. After cooling, the solution was treated with 10%
sodium hydroxide and extracted with chloroform. The
combined extracts were dried (magnesium sulfate) and
evaporated to dryness in vacuo, and the crude oil was
purified by flash chromatography to obtain the target
compounds 6.
for the preparation of 2-alkylamino-6-methylpyridine 5
Sodium borohydride (100 mmol) was added in small
amounts to a solution of 50 mmol of 2-amino-6-
methylpyridine 4 and 100 mmol of appropriate aldehyde
in 10 mL of anhydrous methanol. After stirring for 48 h at
room temperature, water was added, methanol was elimi-
nated by evaporation in vacuo and the mixture was
extracted with chloroform. The combined extracts were
dried (magnesium sulfate) and evaporated to dryness in
vacuo, and the crude oil was purified by flash chroma-
tography to obtain compounds 5.
6.1.3. General procedure
for the preparation of substituted
7-methyl-2,3-dihydro-1,8-naphthyridine-4(1H)-one 3
A mixture of 6 (10 mmol) and 20 mL of 6 N aqueous
sodium hydroxide solution was refluxed for 60 min. After
cooling, the solution was acidified with 10% hydrochloric
acid (pH 4–5) and the mixture was extracted with
chloroform. The combined extracts were dried (magne-
sium sulfate) and the solvent was evaporated to dryness
in vacuo. The crude propionic acid derivative obtained
and 15 g of polyphosphoric acid were heated at 120 °C
for 40 min. After cooling, the mixture was poured into
6.1.2. General procedure
for the preparation of 3-alkylamino-3-
(6-methylpyrid-2-yl)aminopropionic ethyl esters 6
A mixture of 5.0 mmol of the appropriate 6-methyl-2-
alkylaminopyridine 5, 9.5 mL of ethyl acrylate (8.6 mmol)
and 2.0 mL of glacial acetic acid was heated at 100 °C for

P.L. Ferrarini et al. / Eur. J. Med. Chem. 35 (2000) 815–826
Table V. 1H-NMR chemical shifts (δ) of (R,S)-(E)-oximeether derivatives.
821
Compound
X
R
H2
(m)
H3
(m)
H5
(d)
H6
(d)
7-CH3 OCH2 HOCH RNCH2
Other
(s)
(d)
(m)
(m)
9a1
9a2
NCH3 t-Bu
NCH3 i-Pr
3.20
3.20
2.84
2.84
7.83
7.83
6.39
6.39
2.38
2.38
4.12
4.12
3.90
3.90
2.65 (CH3)3C: 1.11 (s) 1-CH3: 3.07 (s)
2.65 CH: 2.62 (m); (CH3)2C: 1.08 (d) 1-CH3:
3.07 (s)
9b1
9b2
9c1
9c2
9c3
9c4
NC2H5 t-Bu
NC2H5 i-Pr
NCH2Ph t-Bu
NCH2Ph i-Pr
3.27
3.27
3.20
3.20
3.18
3.20
3.22
3.22
2.82
2.82
2.78
2.78
2.80
2.70
2.73
2.73
7.81
7.81
7.86
7.86
7.85
7.85
7.83
7.83
6.36
6.36
6.42
6.42
6.45
6.45
6.40
6.40
2.36
2.36
2.38
2.38
2.38
2.38
2.40
2.40
4.12
4.12
4.11
4.11
4.06
4.06
4.11
4.11
3.92
3.92
3.93
3.93
3.80
3.80
3.97
3.97
2.72 (CH3)3C: 1.11 (d) 1-CH2: 3.67 (q); CH3:
1.22 (t)
2.72 CH: 2.57 (m); (CH3)2C: 1.08 (d) 1-CH2:
3.67 (q); CH3: 1.22 (t)
2.68 (CH3)3C: 1.11 (s) 1-CH2: 4.87 (s); Ph:
7.25 (m)
2.68 CH: 2.78 (m); (CH3)2C: 1.08 (d) 1-CH2:
4.87 (s); Ph: 7.25 (m)
2.65 CH3: 1.32 (d); CH: 4.06 (m) 1-CH2: 4.87
(s); Ph: 7.25 (m)
2.69 CH:2.70 (m); CH3: 1.34 (d); CH2: 2.43
(d) 1-CH2: 4.86 (s); Ph: 7.25 (m)
2.80 (CH3)3C: 1.10 (s) 1-CH2: 3.81 (t); CH2:
2.73 (m) Ph: 7.22 (m)
2.77 CH: 2.73 (m); (CH3)2C: 1.07 (d) 1-CH2:
3.81 (t); CH2: 2.73 (m) Ph: 7.23 (m)
2.70 (CH3)3C: 1.11 (s)
NCH2Ph
NCH2Ph
A
B
9d1
9d2
N(CH2)2Ph t-Bu
N(CH2)2Ph i-Pr
10a1
10a2
S
S
t-Bu
i-Pr
3.04
3.04
3.04
3.04
8.03
8.03
6.83
6.83
2.48
2.48
4.15
4.15
3.95
3.95
2.70 CH: 2.75 (m); (CH3)2C: 1.06 (d)
A: CH(CH3)Ph; B: CH(CH3)CH2Ph.
crushed ice and made basic (pH = 8) with cold concen-
trated ammonium hydroxide. The solution was extracted
with chloroform and the combined extracts were dried
(magnesium sulfate) and evaporated in vacuo to obtain
compounds 3, as crude oils, which were purified by flash
chromatography (tables I and II).
A mixture of 12 (10 mmol) and 3 mL of 10% potas-
sium hydroxide was stirred at room temperature for
5 days. The solution was then acidified with hydrochloric
acid (pH = 4–5) and extracted with chloroform. The
combined extracts were dried (magnesium sulfate) and
evaporated in vacuo. The crude residue was purified by
crystallization from petroleum ether 100–140 °C to give
13 in a 14% yield, m.p. 58–59 °C.
6.1.4. 3-(6-Methylpyrid-2-yl)-
mercaptopropionic ethyl ester 12
B) From 11:
A mixture of 1.25 g (10 mmol) of 6-methyl-2-
mercaptopyridine 11 [8], 10 mL of ethyl acrylate
(9.0 mmol) and 0.5 mL of glacial acetic acid was heated
at 170 °C for 5 h. After cooling, the solution was treated
with 10% sodium hydroxide (pH = 7–8) and extracted
with chloroform. The combined extracts were dried
(magnesium sulfate) and evaporated to dryness to obtain
an oily residue which was purified by distillation at
120 °C and 0.3 mm Hg to give 12 in an 86% yield.
A
solution of 2-mercapto-6-methylpyridine 11
(4.0 mmol) in 3 mL of 10% potassium hydroxide was
added to a solution of ꢀ-chloropropionic acid (5.0 mmol)
and NaHCO3 (0.5 mmol) in 3 mL of water and the
mixture was stirred at room temperature for 4 days.
The solution was treated with hydrochloric acid until
the pH was 7 and extracted with chloroform. The com-
bined extracts were dried (magnesium sulfate) and evapo-
rated in vacuo: the starting 2-mercapto-6-methylpyridine
(9% yield) was recovered. The aqueous solution was then
acidified with hydrochloric acid (pH = 4) and extracted
with chloroform. The combined extracts were dried
6.1.5. 3-(6-Methylpyrid-2-yl)mercaptopropionic acid 13
A) From 12:

822
P.L. Ferrarini et al. / Eur. J. Med. Chem. 35 (2000) 815–826
Figure 6. Compounds used for the theoretical calculations.
6.1.6. 7-Methyl-2,3-
dihydrothiopyrano[2,3-b]pyridine-4(4H)-one 14
A mixture of 13 (5.0 mmol) and 10 g of polyphospho-
ric acid was heated at 170 °C for 3 h. After cooling, the
mixture was poured into crushed ice and made basic
(pH = 8) with cold concentrated ammonium hydroxide.
The solution was extracted with chloroform. The com-
bined extracts were dried (magnesium sulfate) and evapo-
rated in vacuo, and the residue was purified by flash
chromatography to give 14 (tables I and II).
Figure 4. Reagents: A) Ethyl acrylate, AcOH(glac.); B) 10%
KOHaq, EtOH; C) ClCH2CH2COONa, H2O.
6.1.7. General procedure for the preparation
of substituted (E)-4(1H)-hydroxyimino-7-methyl-
2,3-dihydro-1,8-naphthyridine 7 and (E)-
4(4H)-hydroxyimino-7-methyl-
(magnesium sulfate) and evaporated to dryness in vacuo,
and the residue was crystallized from petroleum ether
100–140 °C to give 13 in a 63% yield.
2,3-dihydrothiopyrano[2,3-b]pyridine 15
A suspension of 1.0 mmol of the appropriate ketone 3
or 14 and 2.5 mmol of hydroxylamine hydrochloride in
15 mL of anhydrous ethanol and 1 mL of anhydrous
pyridine was refluxed for 90 min. The mixture was
evaporated to dryness in vacuo and the crude residue was
treated with a saturated NaHCO3 solution. The (E)-
oximes 7 or 15 were collected, washed with water and
purified by crystallization or flash chromatography (tables
I and II).
Figure 5. Reagents: A) H2NOH HCl, EtOH, Py; B) EtONa/
EtOH; C) Epichlorohydrin, toluene; D) (t-Butylamine or
i-propylamine), toluene.
Figure 7. Tautomeric forms of 1a3.

P.L. Ferrarini et al. / Eur. J. Med. Chem. 35 (2000) 815–826
823
6.1.9. General procedure for the preparation of (R,S)-
(E)-4(1H)-[1-(3-alkylamino-2-hydroxypropyl)oxyimino]-
7-methyl-2,3-dihydro-1,8–naphthyridine derivatives 9
Two millimoles of the appropriate alkylamine was
added to a mixture of 1.0 mmol of epoxypropyloximino
derivative 8 and 2.0 mmol of lithium perchlorate in 5 mL
of acetonitrile and the mixture was allowed to react at
room temperature for 24 h. The organic solution was
evaporated to dryness in vacuo and the crude residue was
treated with water and extracted with chloroform. The
combined extracts were dried (magnesium sulfate) and
evaporated in vacuo, and the residue was purified by flash
chromatography, to obtain compounds 9 (tables IV and
V).
6.1.10. (R,S)-(E)-4(4H)-[1-(3-alkylamino-
2-hydroxypropyl)oxyimino]-7-methyl-2,3-
dihydrothiopyrano[2,3-b]pyridine derivatives 10
A solution of 1.0 mmol of 16 and 3.0 mmol of suitable
amine in 10 mL of anhydrous toluene was heated at 80 °C
for 24 h in a sealed tube. After cooling, the solution was
evaporated to dryness in vacuo to give an oily residue,
which was purified by flash chromatography to obtain
compounds 10 (tables IV and V).
6.2. Pharmacological methods
All the procedures performed in the pharmacological
screening follow the guidelines of European Community
Council directive 86-609. For the pharmacological study,
male Dunkin-Hurtley guinea-pigs (300–350 g) and male
Wistar rats (200–230 g) were killed by cervical disloca-
tion under light ether anaesthesia. The ꢀ-agonist
L-isoprenaline hydrochloride (IPNA) was dissolved
(1 mM) in distilled water, while the compounds tested
were dissolved (1 mM) in dimethylsulfoxide. All the
further dilutions were performed in distilled water. The
solutions were prepared immediately before the experi-
ments.
Figure 8. Compounds 1a3 and 10a3 in their preferred confor-
mations. MEP contours corresponding to a value of – 10
kcal/mol are shown.
6.1.8. General procedure for the preparation
of (R,S)-(E)-4(1H)-[(2,3-epoxypropyl)oxyimino]-7-
methyl-2,3-dihydro-1,8-naphthyridine derivatives 8
and (R,S)-(E)-4(4H)-[(2,3-epoxypropyl)oxyimino]-
7-methyl-2,3-dihydro thiopyrano[2,3-b]pyridine 16
Compounds 7 or 15 (1.0 mmol) were added to a
solution of 1.0 mmol of freshly prepared sodium meth-
oxide in 15 mL of anhydrous methanol and the mixture
was refluxed for 1 h. The evaporation of the solvent in
vacuo produced the sodium salt, to which 30 mL of
anhydrous toluene and 10 mmol of epichlorohydrin were
added; the resulting suspension was allowed to react at
100 °C for 24 h. The solvent was evaporated in vacuo and
the crude products were purified by flash chromatography
to obtain compounds 8 and 16 (tables I and III).
6.2.1. Isolated guinea-pig atria
A possible ꢀ1-blocking activity was evaluated on
spontaneously beating isolated atria of guinea-pigs, as
previously described [2].
The hearts were rapidly explanted and the atria were
separated from the ventricular tissue and from the major
blood vessels. The left atrium was sutured to a wire-
mounting rod, fixed to a 10 mL chamber of the isolated
organ bath. The right atrium was connected by inexten-
sive thread to an isometric force transducer (Basile mod.
7005) under a pre-load of 1 g. The atrial inotropic tension
developed was recorded on a microdynamometer (Basile
mod. 7050). The bathing fluid (Tyrode saline solution,
composition in mM: NaCl 136.8, KCl 2.95, CaCl2 1.80,

824
P.L. Ferrarini et al. / Eur. J. Med. Chem. 35 (2000) 815–826
Table VI. Antagonistic activity on ꢀ-adrenoceptor.
Compound
R
X
pKb vs ꢀ1
pKb vs ꢀ2
pIC50 vs ꢀ3
1a1
9a1
9a2
9b1
9b2
9c1
9c2
9c3
9c4
9d1
9d2
10a1
10a2
Propranolol
Practolol
t-Bu
t-Bu
i-Pr
t-Bu
i-Pr
NH
NCH3
NCH3
6.48 ± 0.15
≤ 5
6.69 ± 0.20
5.50 ± 0.17
5.75 ± 0.38
5.32 ± 0.33
5.73 ± 0.27
6.46 ± 0.23
6.11 ± 0.08
≤ 5
5.14 ± 0.17
4.51 ± 0.34
n.t.
4.14 ± 0.49
n.t.
≤ 5
≤ 5
≤ 5
≤ 5
≤ 5
≤ 5
≤ 5
≤ 5
NC2H5
NC2H5
NCH2Ph
NCH2Ph
NCH2Ph
NCH2Ph
N(CH2)2Ph
N(CH2)2Ph
S
t-Bu
i-Pr
5.80 ± 0.22
n.t.
Ph(CH3)CH
PhCH2(CH3)CH
t-Bu
≤ 4
≤ 4
n.t.
n.t.
≤ 5
6.38 ± 0.24
6.38 ± 0.12
6.70 ± 0.11
6.13 ± 0.11
8.25 ± 0.12
n.t.
i-Pr
t-Bu
i-Pr
≤ 5
7.65 ± 0.41
5.76 ± 0.46
8.71 ± 0.15
7.51 ± 0.19
5.68 ± 0.21
n.t.
6.15 ± 0.09
n.t.
S
n.t.: not tested.
MgSO4·7H2O 1.05, NaH2PO4 0.41, NaHCO3 11.9, glu-
cose 5.5) was thermostated at 32 °C and continuously
gassed with O2. The preparation was left to equilibrate for
1 h before starting the experimental protocol. A first
Figure 10. Inhibition curve for compound 10a1 versus the
ꢀ1-mediated positive inotropic effect evoked by the reference
concentration of IPNA, in guinea-pig spontaneously beating
atria. The points represent the mean of four experiments; values
of SEM, not drawn for clarity, are < 10%. The responses are
expressed as % of the maximal effect evoked by the reference
concentration of IPNA.
Figure 9. Concentration–response curves for IPNA, in guinea-
pig tracheal preparations (ꢀ2-adrenoceptor stimulation), in con-
trol conditions (™) and in the presence of 9c2 1 µM (♦ ). The
points represent the mean of four experiments; values of SEM,
not drawn for clarity, are < 10%. The responses are expressed as
% of the maximal effect evoked by IPNA.

P.L. Ferrarini et al. / Eur. J. Med. Chem. 35 (2000) 815–826
825
concentration–response curve for IPNA was obtained by
the method of single-concentration administration (start-
ing from 1 nM, with 3-fold increasing steps).
ꢀ1-antagonism was evaluated as the inhibition curve
obtained by the progressive reduction of the positive
inotropic response to a reference concentration (Ar) of
IPNA, induced by increasing concentrations (starting
from 30 nM, with 3-fold increasing steps) of the com-
pound tested. The chosen Ar of IPNA was 100 nM.
were prepared by the method originally described by
Rodbell [15] and subsequently modified by Cushman
[16]. After removal of the major blood vessels, the finely
cut tissue pieces were incubated for 30 min at 37 °C with
collagenase at a concentration of 1 mg/mL in a shaking
water bath gassed with 5% CO2 and 95% O2. Digestion
was terminated by filtration followed by three washes,
each with 10 mL of buffer. Tyrode buffer supplemented
with 30 mM Hepes containing 10 mg/mL bovine serum
albumin at pH 7.4 (THA) was used during collagenase
digestion and incubation. Buffer without bovine serum
albumin was used for the wash. Washed cells were
counted in a Burker chamber and diluted to obtain 106
cells/mL.
Adipocyte lipolysis was measured by incubating
2 × 105 cells in a shaking waterbath at 37 °C for 90 min
in a final volume of 400 µL containing THA buffer and
the agents to be tested. In the tests in which the
antagonistic action was assayed, the cells were pre-
incubated for 15 min with the antagonist or with the
vehicle, before the addition of the agonist. When IPNA
was used, the same concentration of L-ascorbic acid was
added to the incubation tube as an anti-oxidant.
The agents to be tested were dissolved in Tyrode buffer
or in dimethylsulphoxide in a final concentration ranging
from 0.00025–0.025%, and in this case the experiments
were performed with the appropriate control containing
dimethylsulphoxide. At the end of the incubation period,
the reaction was stopped in ice and 200 µL of incubation
medium was taken for enzymatic determination of glyc-
erol [17] released into the incubation medium, using a
Perkin-Elmer Lambda 15 UV/VIS spectrophotometer at
366 nm.
Collagenase (Type II), (–)isoprenaline hydrocloride,
bovine serum albumin (fraction V), ATP (magnesium
salt), ꢀ-NAD, hydrazine hydrate, glycine, and glycerol
were purchased from Sigma (St. Louis, USA). Glycero-
phosphate dehydrogenase (GDH) and glycerokinase (GK)
were obtained from Boehringer (Mannheim, FRG).
6.2.2. Isolated guinea-pig trachea
The compounds were tested on isolated guinea-pig
tracheal smooth muscle in order to evaluate a possible
ꢀ2-antagonist activity, as previously described [2]. The
trachea, explanted and freed from extraneous tissues, was
cut length-wise through the anteromedial cartilage. Fi-
nally, a zig-zag-shaped preparation was obtained by
alternated partial cuts, perpendicular to the length of the
organ. One extremity of the preparation was sutured to a
wire-mounting rod, fixed to a 10 mL chamber of the
isolated organ bath. The other was connected by inexten-
sive thread to an isotonic force transducer (Basile mod.
7006) under a pre-load of 0.5 g. The isotonic changes of
tension were recorded by means of a microdynamometer
(Basile mod. 7050). The bathing fluid was Krebs saline
solution (composition in mM: NaCl 118, KCl 4.75, CaCl2
2.5, MgSO4·7H2O 1.19, NaHCO3 25, glucose 11.5),
thermostated at 37 °C and continuously gassed with a
mixture of O2 (95%) and CO2 (5%).
The preparations were left to equilibrate for 1 h before
starting the experimental protocol. In each preparation,
two concentration–response curves for IPNA were ob-
tained. The equilibration time between the two curves
was 1 h. The second curve was obtained in the presence
of a reference concentration of the test antagonist. Previ-
ous experiments showed that the administration of the
vehicle did not cause any shift of the second concentra-
tion–response curve.
The concentration–relaxing-response curves for IPNA
were obtained as follows: the tracheal smooth muscle was
pre-contracted by the administration of a single concen-
tration (1 mM) of the muscarinic agonist carbamylcho-
line. When the contraction reached a steady plateau, the
trachea was relaxed by the cumulative administration of
increasing concentrations of IPNA (starting from 1 nM,
with 3-fold increasing steps).
6.2.4. Data analysis
6.2.4.1. ꢀ1- and ꢀ2-antagonism
The antagonist potency was expressed as pKb, repre-
senting the value of –log of the dissociation constant, and
it was expressed as the mean ± standard error, for at least
four experiments. The value of the dissociation constant
for the ꢀ1-adrenoceptor/antagonist complex was calcu-
lated by means of the inhibition curve, as previously
described [18]. The value of the dissociation constant for
the ꢀ2-adrenoceptor/antagonist complex was calculated
by means of Gaddum’s method [19]. Data were statisti-
6.2.3. Rat adipocytes
The compounds were tested on adipose cells isolated
from male Wistar rats fed ad libitum, in order to evaluate
a possible ꢀ3-antagonist activity. Epididymal fat pads
were immediately removed and isolated adipose cells

826
P.L. Ferrarini et al. / Eur. J. Med. Chem. 35 (2000) 815–826
cally analysed by ANOVA or by Student’s two-tail t-test;
a value of P < 0.05 was considered as representative of
significant differences. Raw data interpolations and sta-
tistical analyses were performed by a computerized
method (program Graph-Pad Prism TM 2.0).
References
[1] Ferrarini P.L., Mori C., Primofiore G., Da Settimo A., Breschi M.C.,
Martinotti E., Nieri P., Ciucci M.A., Eur. J. Med. Chem. 25 (1990)
489–496.
[2] Ferrarini P.L., Mori C., Badawneh M., Manera C., Saccomanni G.,
Calderone V., Scatizzi R., Barili P.L., Eur. J. Med. Chem. 32 (1997)
955–963.
6.2.4.2. ꢀ3-antagonism
[3] Strosberg A.D., Annu. Rev. Pharmacol. Toxicol. 37 (1997) 421–450.
Different concentrations of IPNA and of the compounds
tested were added to adipocyte preparations to obtain
concentration–response curves. The concentration–
response curve to IPNA was taken as the reference for
agonistic activity and the concentration that induced
about 80% (100 nM) of the maximal response was
employed to assay the activity of antagonists. Antagonis-
tic potency was expressed as pIC50 ± SEM of n experi-
ments, that is the negative logarithm of the molar
concentration of the antagonist that inhibited the stimu-
lant action of IPNA by 50%.
[4] Carboni S., Da Settimo A., Bertini D., Ferrarini P.L., Livi O.,
Tonetti I., Il Farmaco Ed. Sci. 31 (1976) 322–333.
[5] Lindstroem S., Ahmad T., Grivas S., Heterocycles 38 (1994)
529–540.
[6] Katritzky A.R., Rachwal S., Rachwal B., J. Chem. Soc. Perkin
Trans. I (1987) 805–810.
[7] Huttrer C.P., Djerassi C., Beears W.L., Mayer R.L., Scholz C.R.,
J. Am. Chem. Soc. 68 (1946) 1999–2000.
[8] Delarge J., Il Farmaco Ed. Sci. 22 (1967) 1069–1076.
[9] Carboni S., Da Settimo A., Bertini D., Ferrarini P.L., Livi O.,
Tonetti I., Il Farmaco Ed. Sci. 30 (1975) 237–247.
[10] Da Settimo A., Biagi G., Primofiore G., Ferrarini P.L., Livi O., Il
Farmaco Ed. Sci. 33 (1978) 770–780.
The concentration–response curves were calculated by
non-linear regression analysis of data by means of a
computer-aided program (GraphPad Prism 2.0). Statisti-
cal analysis was carried out using Student’s t-test, with a
probability value (P) of less than 0.05 regarded as
significant.
[11] Da Settimo A., Biagi G., Primofiore G., Ferrarini P.L., Livi O.,
Marini A.M., J. Heterocyclic Chem. 17 (1980) 1225–1229.
[12] Balsamo A., Breschi M.C., Chiellini G., Lapucci A., Lazzeri N.,
Macchia M., et al., Bioorg. Med. Chem. 6 (1998) 2151–2160.
[13] Frisch M.J., Trucks G.W., Schlegel H.B., Gill P.M.W., Johnson B.G.,
Robb M.A., et al., Gaussian 94 w (Version 4.0), Gaussian, Inc.,
Pittsburgh PA, 1995.
[14] Murphy G.J., Kirkham D.M., Cawthorne M.A., Young P., Biochem.
Pharmacol. 46 (1993) 575–581.
[15] Rodbell M., J. Biol. Chem. 239 (1964) 375–380.
[16] Cushman S., J. Cell Biol. 46 (1970) 326–341.
[17] Wieland O., Biochem. Z. 239 (1957) 313–319.
[18] Calderone V., J. Pharmacol. Toxicol. Meths. 39 (1998) 129–135.
[19] Gaddum J.H., Pharmacol. Rev. 9 (1957) 211–218.
Acknowledgements
This work was supported by a grant from the Ministero
dell’Università e della Ricerca Scientifica (40%).