Studies towards topical selective β2-adrenoceptor agonists with a long duration of action

Article abstract of DOI:10.1016/j.bmcl.2009.06.104

β₂-Adrenoceptor agonists with basic and acidic groups attached via an alkyl linker to the phenyl ethanolamine core were prepared and investigated in vitro and in vivo. The compounds exhibited a high potency in a functional cellular assay and a bronchoprotective effect in a guinea pig model which lasted over the complete study period of 5 h.

Full text of DOI:10.1016/j.bmcl.2009.06.104

Bioorganic & Medicinal Chemistry Letters 19 (2009) 5237–5240
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Studies towards topical selective b2-adrenoceptor agonists with a long
duration of action
Thierry Bouyssou b, Klaus Rudolf a, Christoph Hoenke a, Philipp Lustenberger a,
Andreas Schnapp b, Ingo Konetzki a,
*
a Department of Chemical Research, Boehringer Ingelheim Pharma GmbH & Co. KG, 88397 Biberach, Germany
b Department of Pulmonary Diseases Research, Boehringer Ingelheim Pharma GmbH & Co. KG, 88397 Biberach, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
b2-Adrenoceptor agonists with basic and acidic groups attached via an alkyl linker to the phenyl ethanol-
amine core were prepared and investigated in vitro and in vivo. The compounds exhibited a high potency
in a functional cellular assay and a bronchoprotective effect in a guinea pig model which lasted over the
complete study period of 5 h.
Received 26 May 2009
Revised 26 June 2009
Accepted 30 June 2009
Available online 4 July 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Beta2
Adrenoceptor agonists
Duration of action
Benzoxazinone
Asthma
COPD
b2-Adrenoceptor agonists are well established in the treatment
of both asthma and chronic obstructive pulmonary disease (COPD)
due to their bronchodilatory effects. The two most prescribed rep-
resentatives of this class are salmeterol 1 and formoterol 2, which
have been approved for a twice-daily (b.i.d.) regimen and are both
used as monotherapy and in combination with steroids. A consid-
erable interest has been developed to identify long-acting beta2
agonists (LABAs) with a 24 h duration of action (q.d. dosing) and
an improved therapeutic window compared to the marketed
b.i.d. LABAs (Fig. 1).1
To avoid unwanted b-adrenoceptor mediated effects like tachy-
cardia, hypokalemia or muscle tremor it is necessary that the sys-
temic exposure of the b-agonist be as low as possible. Low oral
bioavailability would be beneficial in this respect, because a certain
proportion of the inhaled drug is always swallowed and therefore
available for absorption through the gut wall of the intestine. But
even in such cases the agonist might become systemically available
through pulmonary absorption. It has been found that drugs which
show otherwise poor oral bioavailability are absorbed via the highly
permeable pulmonary epithelium and it seems that only the pul-
monary absorption rate can be influenced by the physicochemical
properties of a compound.2,3 Thus, a small pulmonary absorption
rate in combination with poor oral bioavailability should contribute
to low systemic exposure of a b-agonist. We hypothesized that such
characteristics could be achieved through the introduction of addi-
tional basic or acidic groups to the phenyl ethanolamine structure
which is common to all known b-agonists. The incorporation of such
groups appeared to us most practicable at the amine residue (here,
right-hand side of the molecule) since the ethanolamine backbone
as well as the phenolic aryl moiety (here, left-hand side of the mol-
ecule) are important pharmacophores and highly sensitive towards
minor structural modifications.
OH
H
N
HO
HO
O
1
OH
H
N
H
N
O
HO
O
2
O
O
OH
H
HN
HO
N
O
R
N
H
3
* Corresponding author. Tel./fax: +49 7351 5498716.
E-mail address: ingo.konetzki@boehringer-ingelheim.com (I. Konetzki).
Figure 1. Structures of the bid LABAs salmeterol 1 and formoterol 2 and the 5-
hydroxy-4H-benzo[1,4]oxazin-3-ones 3 described within this study.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.06.104

5238
T. Bouyssou et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5237–5240
O
Phenyl ethanolamines like 3 with a 5-hydroxy-4H-benzo[1,4]oxa-
a, b
O
OH
OH
H
N
zin-3-one moiety at the left-hand side and a phenethyl residue bound
to the amine have been already suggested by our company in the past
for use as bronchodilators.4 A patent evaluation gave rise to the
assumption that it should be possible to introduce substituents at
the para-position of the right phenyl group under retention of a suffi-
cient potency towards the b2-adrenoceptor.5,6 The structure 3 was
therefore selected as template for our studies and it was envisaged
that an alkyl linker connected to the phenyl group at the right-hand
side of the molecule should serve as anchor for additional basic or
acidic groups.
HN
HO
8
8
O
H
N
NH
NH
N
H
10
O
c, d
O
H
N
HN
HO
O
O
R
N
H
The synthesis of the 5-hydroxy-4H-benzo[1,4]oxazin-3-ones
described in this study is outlined in Schemes 1 and 2. Acylation
of the phenethylamine 4 with Boc protected 4-amino-butyric acid
and subsequent coupling with the literature known hemi-acetal 6
yielded the ethanolamine 7.4 Boc removal afforded intermediate 8
which was further converted into compound 9 through hydrogena-
tion of the benzyl group.
The introduction of basic and acidic groups to the alkyl amine
side chain of intermediate 8 delivered the target compounds 10–
18. Hence, the primary amine of 8 was converted into a guanidine
through treatment with 1-amidinopyrazole. A TBTU mediated
acylation and removal of the benzyl and Cbz groups through
hydrogenolysis afforded the compounds 11–14. Reaction of the
chloroacetamide obtained from 8 with different amines and depro-
tection yielded the 2-aminoacetamides 15 and 16 as well as the tri-
methyl ammonium compound 17. The sulfonic acid 18 was
obtained from 8 through treatment with 1,3-propane sultone and
subsequent debenzylation.7,8
The in vitro characteristics of the synthesized compounds were
assessed in CHO-K1 cell lines stably expressing the human b1- and
b2-adrenoceptors. The agonist-induced rise of intracellular cAMP
was measured as a functional read out in these assays. The potencies
(EC50) of the compounds were calculated and the intrinsic activity as
a marker for efficacy was reported as percentage of the maximal ef-
fect observed for isoprenaline. Marketed formoterol fumarate salt
was measured in these assays for comparison (Table 1).
The compounds prepared in this first series are potent, full ago-
nists of the human b2-adrenoceptor. The most potent compound,
example 15, exhibited a similar potency and intrinsic activity as
formoterol, while EC50 values around 1 nM were measured for
the majority of the other examples. The least potent b2-agonist
from this set of compounds, example 14, still has an acceptable
H
N
N
NH2
NH2
13 R =
11 R =
O
O
O
NH2
H
N
H
NH2
N
NH2
12 R =
14 R =
O
O
O
OH
e, f, g
H
N
HN
HO
8
O
H
N
N
H
R
H
N
N
N
15 R =
16 R =
O
O
H
N
N
O
O
OH
e, h, i
H
N
HN
HO
8
8
O
O
H
N
N+
N
H
O
17
O
O
OH
j, k
H
N
HN
HO
O
S
O
H
N
OH
N
H
H2N
H2N
a
O
H
N
O
18
NH2
N
H
O
Scheme 2. Reagents and conditions: (a) 1-amidinopyrazole hydrochloride, DMF,
67%; (b) Pd/C, H2, MeOH, 80%; (c) carboxylic acid (benzyloxycarbonylamino-acetic
acid for 11, 3-benzyloxycarbonylamino-propionic acid for 12, 2-benzyloxycarbon-
ylamino-succinic acid 4-benzyl ester for 13, 2,5-bis-benzyloxycarbonylamino-
pentanoic acid for 14), TBTU, DIPEA, THF; (d) Pd/C, H2, MeOH; (e) chloroacetyl
chloride, DIPEA, CH2Cl2, 48%; (f) amine (2 M dimethylamine in THF for 15, N-
methyl-piperazine for 16), DIPEA, THF; (g) Pd/C, H2, MeOH; (h) 4 M trimethylamine
in EtOH, THF, 60 °C, 52%; (i) Pd/C, H2, MeOH, 72%; (j) 1,3-propane sultone, ACN,
EtOH, reflux, 18%; (k) Pd/C, H2, MeOH, 32%.
4
5
O
HN
O
O
O
OH
O
OH
O
BnO
H
N
HN
R1O
6
O
H
N
R2
b
N
H
EC50 value of 5.4 nM. In contrast to formoterol, the compounds
do not show any reasonable selectivity over the b1-adrenoceptor.
The in vivo efficacy, potency, and duration of action in compar-
ison to formoterol were determined in guinea pigs according to the
method of Konzett–Roessler.9 In this model, the test compound is
applied intra-tracheally to anaesthetized guinea pigs and broncho-
constriction is induced by intravenous application of acetylcholine
at various time points after administration of the compound.
7 R1 = Bn, R2 = Boc
8 R1 = Bn, R2 = H
9 R1 = H, R2 = H
c
d
Scheme 1. Reagents and conditions: (a) isobutyl chloroformate, N-BOC-aminobu-
tyric acid, N-methyl-morpholine, À25 °C, then 4, 64%; (b) EtOH, 50–80 °C, then
NaBH4, rt, 84%; (c) TFA, CH2Cl2, 42%; (d) Pd/C, H2, MeOH, 68%.

T. Bouyssou et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5237–5240
5239
Table 1
Potency and intrinsic activity at human b1- and b2-adrenoceptors for compounds 2, 9–12 and 14–21
Compounds
hb1 EC50 (nM)
hb1 IAa (%)
hb2 EC50 (nM)
hb2 IAa (%)
b1/b2
2
9
86 23
3.4 1.1
1.8 0.4
4.5 0.2
4.6 1.5
24.0 11.3
5.3 2.7
5.7
122
7
0.4 0.2
1.1 0.1
1.0 0.1
1.2 0.3
0.9 0.6
5.4 0.7
0.5 0.1
1.2
1.0 0.1
1.8 0.4
9.0 4.2
11.6 7.6
4.1 3.4
133 48
128 28
215
3.1
1.8
3.8
5.1
4.4
10.6
4.8
3.5
5.5
1.2
2.4
4.4
93 33
98 24
125 74
135 50
117 36
76 18
76
72
95 60
14
12
13
10
11
12
14
15
16
17
18
19
20
21
115
2
142 30
161 16
130 16
125 47
126
3.5 0.3
9.9 1.6
9
162 92
121
4
ꢀ11
ꢀ28
ꢀ18
4
8
9
1
1
2
78 15
37
54
0
6
a
Cloned human b1- and b2-adrenoceptors are expressed in CHO-K1 cell lines and the intracellular accumulation of cAMP after addition of various test compounds is
measured. The intrinsic activity (IA) is reported as percentage of the maximal effect of isoprenaline (=100%). Values are means of two independent experiments (exceptions:
compound 2 N = 3, compound 16 N = 1). The standard deviation is given.
0.1 µg/kg
0.3 µg/kg
1 µg/kg
3 µg/kg
A
In addition to measuring bronchoprotection, the heart rate of
the guinea pigs is also recorded in this experiment. As an increase
in heart rate indicates a systemic exposure of the test compound,
the pharmacokinetic properties of the agonists are assessed indi-
rectly in this model via the systemic pharmacodynamic effects.
Four examples from the first series were selected based on the
structurally diversity of their side chains and tested in vivo
(10 = guanidine; 11 = mono-amino; 14 = di-amino; 18 = mono-
amino and sulfonic acid).10 All examples showed a dose dependent
bronchoprotection in this model. The lowest dose causing a com-
plete reversal (=100% bronchoprotection) of the induced broncho-
120
100
80
60
40
20
0
spasm was 10 lg/kg for the three compounds (examples 10, 11,
14) with purely basic side chains. Further, a duration of action over
the complete study period of 5 h was determined for the three
examples at doses of 3 and 10
physiologically significant (>10%) increase in heart rate was al-
ready evident at 10 /kg. A heart rate increase was also observed
for the sulfonic acid bearing example 18 at a sub-maximal dose
of 3 g/kg. All four examples showed a similar time course of the
lg/kg (Fig. 2B). Unfortunately, a
l
l
heart rate inasmuch as the maximum was reached between 10
and 20 min after administration of the test compound indicating
a similar pulmonary absorption rate. Furthermore, a sustained ele-
vation of the heart rate was determined at the highest tested dose
for all four examples although the increase was only marginal after
90 min compared to the heart rate before drug administration.11
The comparator formoterol also exhibited a dose dependent
bronchoprotection in this model with a complete reversal of the in-
time [minutes]
1
µ
g/kg
3
µ
g/kg
10µg/kg
B
120
100
80
60
40
20
0
duced bronchospasm at a dose of 1
and 1 g/kg showed a decline in bronchoprotection after 30 and
150 min, respectively. A significant increase in heart rate was de-
tected for formoterol only at a dose of 10 g/kg, which is 10-fold
lg/kg. The lower doses of 0.3
l
l
higher then the first dose reaching 100% bronchoprotection
(Fig. 2A).
Due to a therapeutic window inferior to the marketed formoter-
ol, the compounds of the first series where not profiled any further.
The small therapeutic window was attributed to the modest selec-
tivity of the compounds over the b1-adrenoceptor and in the fur-
ther course of this study it was attempted to prepare agonists
without this liability.
The phenolic hydroxyl group is important for the potency and
selectivity of a b-adrenoceptor agonist. Precedents from the litera-
ture showed that a shift of this hydroxyl group from the para- to
the meta-position in respect to the ethanolamine substituent in-
creases the b2-selectivity of the agonist.12,13 These findings in-
spired us to prepare the examples 19–21 which differ only in the
position of the phenolic hydroxyl group from the previously inves-
tigated 5-hydroxy-4H-benzo[1,4]oxazin-3-ones and which were
time [minutes]
Figure 2. Bronchoprotective effect of formoterol (A) and example 14 (B) in the
Konzett–Roessler model. Bronchospams were induced in guinea pigs by iv
injections of acetylcholine every 10 min. Bronchoprotection is expressed as the
percentage of inhibition of the increase in pulmonary resistance induced by
acetylcholine (N = 2 animals per dose).

5240
T. Bouyssou et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5237–5240
obtained in an analogous fashion (Scheme 3). The potency of these
compounds towards the b2-adrenoceptor was lower compared to
their 5-hydroxy analogs but still in an acceptable range. Advanta-
geously, the examples 19–21 exhibited only a negligible intrinsic
activity at the b1-adrenoceptor (Table 1).
study period. Further results from our continued efforts towards
the identification of long-acting b2-adrenoceptor agonists with an
improved therapeutic window will be reported in due course.
Supplementary data
These three examples were then tested in the Konzett–Roessler
model at doses of 3 and 10
lg/kg. Again, an increase in heart rate
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2009.06.104.
was observed at 10 g/kg which was the first dose reaching a full
l
reversal of the induced bronchospasm. Consequently, the further
profiling of these compounds was stopped.
References and notes
Consecutively, 6-hydroxy-4H-benzo[1,4]oxazin-3-ones were
prepared with acidic (carboxylic and sulfonic acids) instead of ba-
sic groups but these compounds displayed an insufficient potency
in the cellular cAMP assay and were not submitted to any further
experiments (data not shown).
A potential explanation for the failure of this approach could be
that b2-agonists are already protonated at the basic center of their
ethanolamine moiety under physiological conditions (pH 7.4)
implying that an additional charged group will have not much
influence on the pulmonary absorption rate. Furthermore, to con-
trol the absorption rate seems to be insufficient since a rapid clear-
ance of the absorbed fraction of the agonist is another important
contributor to a low systemic exposure and a beneficial therapeu-
tic window of the drug.
1. Cazzola, M.; Matera, M. G.; Lötvall, J. Exp. Opin. Invest. Drugs 2005, 14, 775.
2. Tronde, A.; Nordén, B.; Marchner, H.; Wendel, A.-K.; Lennernäs, H.; Hultkvist
Bengtsson, U. J. Pharm. Sci. 2003, 92, 1216.
3. Tronde, A.; Nordén, B.; Jeppsson, A.-B.; Brunmark, P.; Nilsson, E.; Lennernäs, H.;
Hultkvist Bengtsson, U. J. Drug Targeting 2001, 11, 61.
4. Schromm, K.; Mentrup, A.; Renth, E. O.; Fügner, A. U.S. Patent 4460,581, 1984.
5. Moran, E. J.; Griffin, J. H.; Choi, S.-K. Patent WO200142193.
6. Moran, E. J.; Jacobsen, J. R.; Leadbetter, M. R.; Nodwell, M. B.; Trapp, S. G.;
Aggen, J.; Church, T. J. Patent WO2003042164.
7. The purity and identity of the synthesized compounds detailed in this
publication were determined by 1H NMR and mass spectroscopy.
8. Synthesis protocol of
(0.86 mmol) benzyloxycarbonylamino-acetic acid, 290 mg (0.90 mmol) TBTU
and 160 (0.92 mmol) ethyl-diisopropyl-amine in 30 mL anhydrous THF
a representative example (compound 12): 190 mg
l
L
were stirred for 30 min at ambient temperature. 500 mg (0.92 mmol) of
compound 8 were added and stirring was continued overnight. The solvent was
removed in vacuo and the residue was dissolved in dichloromethane and
washed with aqueous potassium carbonate solution. The organic phase was
dried over sodium sulfate, concentrated and purified by chromatography on
silica gel eluting with dichloromethane/methanol/ammonia (95:4.5:0.5?
60:35:5 by volume). The product containing fractions were combined and
the solvents evaporated. The Cbz-protected compound 12 was obtained as pale
solid after triturating with diisopropyl-ether. Yield: 240 mg (38%). 210 mg
(0.28 mmol) Cbz-protected compound 12 were hydrogenated in the presence
of 80 mg palladium on carbon (10 wt %) in 20 mL methanol at ambient
temperatures and a hydrogen pressure of 50 psi. The catalyst was filtered off
and the solvent was removed in vacuo. Trituration of the residue with
diisopropyl-ether yielded compound 12 as white solid. Yield: 95 mg (65%).
Analytical data of compound 12: mass spectroscopy: [M+H]+ = 528. 1H NMR
(400 MHz, methanol-d4) d = 7.45 (d, 2H, J = 8.4 Hz), 7.10 (d, 2H, J = 8.4 Hz), 6.94
(d, 1H, J = 8.5 Hz), 6.51 (d, 1H, J = 8.5 Hz), 4.98 (m, 1H), 4.52 (d, 1H, J = 15.3 Hz),
4.48 (d, 1H, J = 15.3 Hz), 3.28 (t, 2H, J = 6.7 Hz), 2.96 (t, 2H, J = 6.8 Hz), 2.84 (m,
2H), 2.70 (m, 2H), 2.40 (m, 4H), 1.89 (m, 2H), 1.09 (s, 3H), 1.08 (s, 3H). 13C NMR
(100 MHz, methanol-d4) d = 174.1, 173.8, 166.9, 147.0, 143.0, 138.3, 134.5,
131.9 (2C), 122.4, 122.1, 121.0 (2C), 116.3, 110.5, 68.1, 68.0, 55.1, 46.9, 39.8,
38.6, 38.0, 35.2, 26.6, 26.4, 26.1.
In conclusion, a number of highly potent b2-adrenoceptor ago-
nists have been identified. Selected examples from this series were
tested in a guinea pig in vivo model and exhibited a dose depen-
dent bronchoprotective effect which lasted over the complete
O
O
OH
H
N
HN
O
R
N
H
OH
NH2
19 R =
20 R =
H
N
N+
9. Walland, A.; Palluk, R.; Burkard, S.; Hammer, R. Eur. J. Pharmacol. 1997, 330,
213.
10. The results from the anaesthetized guinea pig experiments are available in the
Supplementary data for all compounds from this publication tested in the
model.
O
H
N
NH2
NH
21 R =
11. Similar time courses of the heart rate were observed for the examples 19–21.
See also Supplementary data.
12. Tamura, Y.; Yoshizaki, S. J. Med. Chem. 1981, 24, 634.
13. Schwender, C. F.; Sunday, B. R.; Shavel, J. J. Med. Chem. 1974, 17, 1112.
Scheme 3. 6-Hydroxy analogs of examples 8, 10 and 17.