Design, synthesis and biological evaluation of 8-(2-amino-1-hydroxyethyl)-6-hydroxy-1,4-benzoxazine-3(4H)-one derivatives as potent β2-adrenoceptor agonists

Article abstract of DOI:10.1016/j.bmc.2019.115178

A series of β₂-adrenoceptor agonists with an 8-(2-amino-1-hydroxyethyl)-6-hydroxy-1,4-benzoxazine-3(4H)-one moiety is presented. The stimulatory effects of the compounds on human β₂-adrenoceptor and β₁-adrenoceptor were characterized by a cell-based assay. Their smooth muscle relaxant activities were tested on isolated guinea pig trachea. Most of the compounds were found to be potent and selective agonists of the β₂-adrenoceptor. One of the compounds, (R)-18c, possessed a strong β₂-adrenoceptor agonistic effect with an EC₅₀ value of 24 pM. It produced a full and potent airway smooth muscle relaxant effect same as olodaterol. Its onset of action was 3.5 min and its duration of action was more than 12 h in an in vitro guinea pig trachea model of bronchodilation. These results suggest that (R)-18c is a potential candidate for long-acting β₂-AR agonists.

Full text of DOI:10.1016/j.bmc.2019.115178

Journal Pre-proofs
Design, synthesis and biological evaluation of 8-(2-amino-1-hydroxyethyl)-6-
hydroxy-1,4-benzoxazine-3(4H)-one derivatives as potent β2-adrenoceptor ag-
onists
Ce Yi, Gang Xing, Siqi Wang, Xiaoran Li, Yichuang Liu, Jinyan Li, Bin Lin,
Anthony Yiu-Ho Woo, Yuyang Zhang, Li Pan, Maosheng Cheng
PII:
S0968-0896(19)31475-0
https://doi.org/10.1016/j.bmc.2019.115178
BMC 115178
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
28 August 2019
18 October 2019
22 October 2019
Please cite this article as: C. Yi, G. Xing, S. Wang, X. Li, Y. Liu, J. Li, B. Lin, A. Yiu-Ho Woo, Y. Zhang, L.
Pan, M. Cheng, Design, synthesis and biological evaluation of 8-(2-amino-1-hydroxyethyl)-6-hydroxy-1,4-
benzoxazine-3(4H)-one derivatives as potent β2-adrenoceptor agonists, Bioorganic & Medicinal Chemistry
(2019), doi: https://doi.org/10.1016/j.bmc.2019.115178
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2019 Published by Elsevier Ltd.

Design, synthesis and biological evaluation of 8-(2-amino-1-hydroxyethyl)-6-
hydroxy-1,4-benzoxazine-3(4H)-one derivatives as potent β2-adrenoceptor
agonists
Ce Yia,b, Gang Xinga,b, Siqi Wangc, Xiaoran Lic, Yichuang Liua,b, Jinyan Lia,b, Bin Lina,b,
Anthony Yiu-Ho Wooc, Yuyang Zhangc, Li Pana,b**, Maosheng Chenga,b*
a Department of Medicinal Chemistry, School of Pharmaceutical Engineering,
Shenyang Pharmaceutical University, Shenyang 110016, China.
b Key Laboratory of Structure-Based Drug Design and Discovery of Ministry of
Education, Shenyang Pharmaceutical University, Shenyang 110016, China.
c Department of Pharmacology, School of Life Sciences and Biopharmaceutics,
Shenyang Pharmaceutical University, Shenyang 110016, China.
*Corresponding author.
** Corresponding author.
Email address: mscheng@syphu.edu.cn (Maosheng Cheng)，panli65@aliyun.com (Li
Pan)
Abstract
A series of β2-adrenoceptor agonists with an 8-(2-amino-1-hydroxyethyl)-6-
hydroxy-1,4-benzoxazine-3(4H)-one moiety is presented. The stimulatory effects of the
compounds on human β2-adrenoceptor and β1-adrenoceptor were characterized by a
cell-based assay. Their smooth muscle relaxant activities were tested on isolated guinea
pig trachea. Most of the compounds were found to be potent and selective agonists of
the β2-adrenoceptor. One of the compounds, (R)-18c, possessed a strong β2-
1

adrenoceptor agonistic effect with an EC50 value of 24 pM. It produced a full and potent
airway smooth muscle relaxant effect same as olodaterol. Its onset of action was 3.5
min and its duration of action was more than 12 h in an in vitro guinea pig trachea
model of bronchodilation. These results suggest that (R)-18c is a potential candidate
for long-acting β2-AR agonists.
Keyword: Asthma, COPD, β2-adrenoceptor agonist, isoprenaline, salmeterol,
olodaterol.
1. Introduction
Chronic obstructive pulmonary disease (COPD), characterized by persistent
expiratory limitation and progressive airway inflammation, is a major cause of
morbidity and mortality in the world.1-3 It is the most common noncommunicable
disease of the respiratory system secondary to asthma. The death rate of COPD ranks
fifth globally nowadays and is expected to rise to the third place in 2030.4-6 Asthma is
a chronic inflammatory disorder of the airway characterized by hyperresponsiveness
and structural change of the airway. In susceptible individuals, the inflammation causes
recurrent episodes of widespread but variable airflow obstruction.7 The airflow
limitation in asthma is reversible either spontaneously or with treatment. β2-
Adrenoceptor (β2-AR) agonists are the most effective bronchodilator drugs to treat
asthma.8 The β2-AR, a member of the G protein-coupled receptor superfamily, is highly
expressed in smooth muscles of the respiratory tract. Stimulation of the β2-AR causes
airway smooth muscle relaxation. Thus, β2-AR agonists such as those presented in Fig.
2

1 are widely used to relieve symptoms of airway obstruction. As highly effective
bronchodilators, inhaled long-acting β2-AR agonists (LABAs) have become standard
of care for the control of airway obstruction.9-11 β2-AR agonists are mainly categorized
into three generations. First-generation drugs, such as salbutamol (1),12 are short acting
β2-AR agonists (SABA) which are characterized by fast action onset and short duration
of action (3-6 h). Salmeterol (2)13 and formoterol (3)13 are second-generation β2-AR
agonists with duration of action of about 12 h. Third-generation agents (also called
ultra-LABAs) including indacaterol (4)14,15 and olodaterol (5)16,17 with duration of
action of about 24 h are used with a once-daily dosing regimen.
3

OH
H
N
OH
H
O
N
HO
HO
HO
OH
Salbutamol 1
Salmeterol 2
OH
H
N
OH
H
H
N
O
N
HO
HN
HO
O
O
Formoterol 3
Indacaterol 4
OH
H
OH
H
HO
N
N
OH
O
O
HO
HN
HN
O
O
Olodaterol 5
9g
OH
H
N
HO
HN
O
(R)-18c
Fig. 1 Structures of β2-adrenoceptor agonists: salbutamol (1), salmeterol (2), formoterol (3), indacaterol (4),
olodaterol (5), 9g (Ref. 18) and (R)-18c (Ref. 18).
Using various medicinal chemistry strategies, we aim to discover new LABAs, but
it is becoming more difficult to construct truly innovative compounds. Our group has
previously synthesized some new β2-AR agonists. Compounds 9g (Ref. 18) and (R)-
18c (Ref. 18) (Fig. 1) with a common 5-(2-amino-1-hydroxyethyl)-8-hydroxyquinolin-
2(1H)-one moiety have shown high β2-AR agonistic activities and β2/β1-selectivities.18
The difference between the two compounds lies on their tail groups. In the present study,
4

different tail groups were combined with 8-(2-amino-1-hydroxyethyl)-6-hydroxy-1,4-
benzoxazine-3(4H)-one 6 (Fig. 2), the head group of olodaterol, to produce a series of
compounds. Olodaterol is an ultra-LABA developed by Boehringer-Ingelheim.
OH
H
HO
N
R
O
HN
O
6
Fig. 2 The structure of 8-(2-amino-1-hydroxyethyl)-6-hydroxy-1,4-benzoxazine-3(4H)-one analogues (6)
2. Results and discussion
2.1 Chemistry
Firstly, we synthesized a series of tail groups: 10a-l and 14a-h. Tail groups 10a-l
were synthesized in three steps from the respective aldehydes 7a-l which went through
the Wittig reaction with serial reduction of the double bond and the cyano group
(Scheme 1). In addition, different ketones 11a-l went through the Grignard reaction,
the Ritter reaction and deacetylation to yield the respective tail groups 14a-h (Scheme
2).
The head group was synthesized and linked to the tail groups. Briefly, acetophenone
15 was prepared as previously described.19 The key steps involved oxidation of 15 to
yield compound 16 and addition of different primary amines (10a-l and 14a-h) to 16
followed by removal of the benzyl protecting group to deliver the target compounds
17a-l and 18a-h. Compound 16 reacted with purchased commercially amines
(phenethylamine, 4-fluorophenethylamine, 4-methoxyphenethylamine, 4-phenylbutan-
5

1-amine) to yield 17-1 – 17-4. To synthesize the chiral target compounds (R)-18a-h,
acetophenone 15 was first brominated to give compound 19. Then, (-)-
diisopinocampheyl chloroborane ((-)-DIP-Cl) was added in the presence of 2 mol/L
NaOH to induce chiral reduction and epoxidation of compound 19 to give the important
chiral intermediate 20. Compounds 14a-h were reacted with compound 20 and
balanced with ethyl acetate-HCl to yield the respective HCl salts, followed by
hydrogenation to yield the target compounds (R)-18a-h (Scheme 3). Biological tests
were performed on the free base of salmeterol and compound 18a, and the HCl salts of
olodaterol, isoprenaline and all the new compounds except 18a.
O
CN
CN
a
b
c
H
NH2
R1
R1
R1
R1
7a-l
8a-l
9a-l
10a-l
10a:R1= H
10e:R1= m-OH 10f:R1= p-OCH3 10g:R1= o-OCH3 10h:R1= m-OCH3
10i:R1= p-CH3 10j:R1= m-CH3 10k:R1= m-CF3 10l:R1= 3,5-F,F
10b:R1= p-F
10c:R1= p-OH
10d:R1= o-OH
Scheme 1. Reagents and conditions: (a) 2-diethoxyphosphorylacetonitrile, t-BuOK, anhydrous THF, ice-bath 30
∙ THF,
min, 90-95%; (b) 5% Pd/C, H2, ethyl acetate, 1 h, 95-98%; (c) BH3
anhydrous THF, 60 °C, 2 h, 90-95%.
R2
O
R2
O
R2
R2
c
a
b
R3
N
R3
R3
OH
12a-h
R3
NH2
H
11a-h
13a-h
14a-h
14a:R2= H , R3= -phenyl
14b:R2= H , R3= p-methoxyphenyl
14d:R2= H , R3= m-methoxyphenyl
14f:R2= H , R3= m-hydroxyphenyl
14c:R2= H , R3= o-methoxyphenyl
14e:R2= H , R3= p-hydroxyphenyl
14g:R2= -phenyl , R3= -phenyl
14h:R2= p-fluorophenyl , R3= p-fluorophenyl
Scheme 2. Reagents and conditions: (a) MeMgBr, anhydrous THF, ice-bath, 40 min, 85-90%; (b) acetonitrile,
concentrated sulfuric acid, ice-bath, 3 h, 70-80%; (c) KOH, ethane-1,2-diol, 200 °C, 15 h, 85-90%;
6

OH
O
H
R1
HO
N
10a-l
•HCl
O
O
O
O
b
HN
BnO
BnO
OH
OH
a
O
17a-l
OH
HN
HN
H
HO
N
R2
R3
O
15
O
14a-h
16
c
O
•HCl
18a-h
OH
HN
O
d
H
HO
N
R
g
O
•HCl
HN
17-1 to 4
O
O
O
OH
O
H
BnO
Br
BnO
HO
N
R2
R3
14a-h
e
O
O
f
HN
HN
•HCl
HN
O
O
O
19
20
(R)-18a-h
Scheme 3. Reagents and conditions: (a) 48% HBr, DMSO, 80°C, 1.5 h, 90-98%; (b) (i) NaBH4, DMSO, MeOH, 1
h, 45-60%; (ii) 5% Pd/C, H2, MeOH, 1 h, 90-98%; (iii) ethyl acetate-HCl, 2 h; (c) (i) NaBH4, DMSO, MeOH, 15 h,
45-60%; (ii) 5% Pd/C, H2, MeOH, 1 h, 90-98%; (iii) ethyl acetate-HCl, 2 h; (d) TBABr3, 1,4-dioxane, MeOH, 2 h,
85-90% (e) (i) (-)-DIP-Cl, anhydrous THF, ice-bath, 1 h, 80-84%; (ii) 2 mol/L NaOH (aq), ice-bath, 2 h, 95-98%;
(f) (i) isopropanol, reflux, 15 h, 65-70%; (ii) 5% Pd/C, H2, MeOH , 1 h, 90-98%; (iii) ethyl acetate-HCl, 2 h; (g) (i)
purchased amines (phenethylamine, 4-fluorophenethylamine, 4-methoxyphenethylamine, 4-phenylbutan-1-amine),
isopropanol, reflux, 15 h, 65-70%; (ii) 5% Pd/C, H2, MeOH , 1 h, 90-98%; (iii) ethyl acetate-HCl, 2 h.
2.2. Biological tests
2.2.1. Activities and selectivities in HEK-293 cells
Respiratory tissues including the lung and the airway express both the β1-subtype
and the β2-subtype of the β-ARs, in which the β2-AR is the dominant subtype. The
reverse is true for the heart with the β1-AR being the dominant subtype. Thus, β2-AR
agonists that cross-activate the β1-AR are more prone to side effects such as tachycardia.
In the present study, isoprenaline was used as a reference for non-selective β1,β2-AR
agonist.20 In response to stimulation of the β2-AR, intracellular cAMP levels of airway
smooth muscle cells increase and a cascade of intracellular events occur ultimately
7

leading to airway smooth muscle relaxation.21 To evaluate the biological activities of
the target compounds, cell-based assays were conducted using human embryonic
kidney 293 (HEK-293) cells overexpressing β1-AR or β2-AR. cAMP accumulation was
measured after exposure of the cells to the test compounds for 1 h. The potencies of the
compounds were expressed in terms of the pEC50 values. The pEC50 value at the β1-AR
was subtracted from the pEC50 value at the β2-AR to obtain the β2/β1-selectivity (Table
1).
We began the study by comparing compounds having different lengths of the
carbon chain in the tail group. Compounds 17-1, 17a and 17-4 with 2, 3 and 4 carbon
atoms between the secondary amine group and the terminal benzene group were found
to have pEC50 values of 8.62, 9.50 and 9.60, respectively. The potency of 17-1 was one
order of magnitude lower than that of 17-4, and this held true even when the para-
position of the phenyl group in 17-1 was substituted with a fluoride group (17-2) or a
methoxy group (17-3). Note that both 17-3 and olodaterol have the same backbone with
olodaterol having an extra dimethyl group on the α-carbon of the N-ethylbenzene group.
The pEC50 values of 17-4 and 17a were similar. Based on our previous study with the
discovery of a highly potent β2-adrenoceptor agonist (R)-18c (Ref. 18) from the
indacaterol derivatives, we deduced that compound 17a with an extra carbon atom in
the carbon chain as compared to olodaterol would be a more promising lead compound
as compared to 17-4 for further modification and testing. We explored whether further
substitution in the phenyl ring of 17a might improve activity. A series of compounds
(17b-17l) were synthesized. As exemplified in compounds 17c-e and 17f-h,
substitution in the ortho- or meta-position of the phenyl group resulted in higher activity
than substitution in the para-position. For compounds 17c-e, substitution with a
hydroxyl group in the meta-position achieved the best result. But for compounds 17f-h
8

(substituted with a methoxy), the ortho-position was the best. Comparing compounds
17b (fluoride) and 17f (methoxy), compound 17b was minimally more active than 17f,
suggesting that the ability of the para-substituent to supply or absorb electrons makes
little difference. The same structure-activity relationship applies to 17i (methyl) and
17k (trifluoromethyl) with similar activities as the methyl group and the trifluoromethyl
group occupy similar space but have opposite electophilicity. Activities of compounds
17d and 17e were at the same level as that of the parent compound 17a. Interestingly,
the compound 17l bearing two fluorine atoms showed the highest activity in this series
with a pEC50 value of 9.89. In summary, most compounds in this series 17a-17l were
more active than 17-1, 17-2, 17-3 (compound 17i was less active than 17-3), albeit less
potent than isoprenaline, salmeterol and olodaterol by about 10 folds.
Comparison of the activities of compound 17-3 and olodaterol revealed that
introduction of a dimethyl group on the α-carbon of the secondary amine group can
greatly increase activity. Therefore, compounds 18a-18h were synthesized with a
dimethyl substitution on the α-carbon of the secondary amine group. The pEC50 value
of 18a was 10.49, a potency similar to those of isoprenaline and salmeterol. However,
18a was less selective to the β2-AR than salmeterol and olodaterol as suggested by its
low β2-β1 value. Therefore, we intended to improve the potency of 18a by introducing
a range of substituent groups on the phenyl ring (18g-18h). When 18a was substituted
with a methoxy group (18b-d), ortho-substitution (18c) was found to be superior,
albeit no observable improvements in activity and selectivity were noted. Among the
compounds with a hydroxyl substituent 18e-f, compound 18f (meta-hydroxyl) with a
pEC50 value of 10.51 was better than compound 18e (para-hydroxyl). Comparing
selectivity, 18f was similar to 18c but inferior to 18a while the selectivity of 18e was
slightly improved. To find out whether steric hindrance had a deleterious influence on
9

potency, compounds 18g-h were synthesized and studied. They were found to be less
potent than 18a.
Next, we focused on the stereoselective synthesis of (R)-18a-h, and the biological
activities of this set were compared with those of 18a-h. We found that activities of all
(R)-enantiomers were enhanced while the effects of the substituent groups on pEC50
values largely remained. Except for (R)-18g-h, all the pEC50 values were above 10.45.
The highest one was 11.04 ((R)-18a) which was basically the same as that of olodaterol.
However, the selectivity of (R)-18a was not as good. Compound (R)-18f with a slightly
lower activity (pEC50 = 10.79) exhibited a selectivity approaching those of olodaterol
and salmeterol.
Table 1
Activities of the test compounds in inducing cAMP accumulation in HEK293 cells
expressing β2- or β1-adrenoceptors. Isoprenaline, salmeterol and olodaterol were used
as references.
OH
H
HO
N
R
O
HN
O
a
b
Compound
17-1
R
β2pEC50
8.62
β1pEC50
β2-β1
17-2
8.31
F
17-3
8.79
OCH3
17-4
9.60
17a
9.50
10

17b
17c
17d
17e
17f
17g
17h
9.13
9.05
9.67
9.76
9.03
9.42
9.37
F
OH
HO
OH
OCH3
H3CO
OCH3
17i
17j
8.65
9.16
CH3
CH3
17k
17l
8.87
9.89
CF3
F
F
18a
18b
18c
10.49±0.08
10.31±0.16
10.54±0.10
7.83±0.05
2.66
OCH3
8.41±0.62
7.91±0.80
2.13
2.40
H3CO
18d
10.31±0.16
OCH3
18e
18f
10.40±0.05
10.51±0.31
7.54±0.04
8.38±0.58
2.86
2.13
OH
OH
11

18g
18h
9.21
9.10
F
F
(R)-18a
(R)-18b
(R)-18c
11.04±0.08
10.45±0.03
10.62±0.04
8.91±0.73
8.09±0.51
8.85±0.45
2.13
2.36
1.77
OCH3
H3CO
(R)-18d
10.56±0.24
8.25±0.81
2.31
OCH3
(R)-18e
(R)-18f
10.59±0.30
10.79±0.07
8.27±0.61
7.44±0.22
2.32
3.35
OH
OH
(R)-18g
(R)-18h
9.85
9.76
F
F
OH
H
N
isoprenaline
10.26
8.81±0.19
1.45
HO
HO
salmeterol
olodaterol
10.35
11.07
6.50±0.07
7.56±0.64
3.85
3.51
a. Data represent mean ± SEM obtained from at least four independent experiments. HCl omitted from structures.
b Selectivity is defined as pEC50 at the β2-adrenoceptor - pEC50 at the β1-adrenoceptor.
2.2.2 Intrinsic activity in non-overexpressed HEK293 cells
The results described so far were obtained from HEK293 cells overexpressing β2-
12

or β1-AR. To study the efficacies of the compounds, it is necessary to conduct
experiments on cells expressing the β2-AR at levels resembling those in airway smooth
muscles. The non-overexpressed HEK293 cells expressing endogenous human β2-AR
was used for this purpose. The intrinsic activities of the compounds were determined
and compared against that of isoprenaline which was defined as 100%. Compounds
with high activities in the cell-based assay, for instance 18a, 18c-f and (R)-18a-f, were
chosen for testing. Salmeterol and olodaterol were used as reference drugs (Table 2).
The intrinsic activities of salmeterol (28.76%) and olodaterol (30.60%) were found to
be lower than that of isoprenaline, so as the other compounds tested. Compounds 18f,
(R)-18e and (R)-18f showed the highest intrinsic activities with values of 26.48%,
21.61% and 28.68%, respectively. Therefore, in non-overexpressed HEK293 cells, all
the test compounds behaved as partial β2-AR agonists.
Table 2
Intrinsic activities of selected compounds 18a, 18c-f and (R)-18a-f, salmeterol,
olodaterol and isoprenaline on HEK293 cells endogenously expressing human β2-
adrenoceptors
Intrinsic activity
Compound
(vs. isoprenaline %)
18a
18c
18d
18e
18f
32.91
28.79
38.90
33.98
26.48
13

(R)-18a
(R) -18b
33.58
27.87
32.82
32.48
21.61
28.68
28.76
30.60
100
(R)-18c
(R)-18d
(R)-18e
(R)-18f
salmeterol
olodaterol
isoprenaline
2.2.3 Tracheal muscle relaxant activities
To predict the bronchodilator effects of the compounds in vivo, we examined the
smooth muscle relaxant effects of the compounds on isolated guinea pig tracheal strips.
Compounds 18a-f and (R)-18a-f with high activities in the cell-based assay were
chosen for the study. Relaxant responses were expressed as percentages of the maximal
contractile effect induced by histamine. Concentration-relaxation curves were
constructed (Fig. 3-5). Emax represents the maximal effect attainable from the
compounds. Potency (pD2) represents the negative logarithm of the concentration of
the compounds at which the relaxant effect equals to 50% of the drug’s own maximal
effect. The pD2 values were determined by non-linear regression analysis using
GraphPad Prism, version 5.0. Comparison between sets of data was performed by
paired Student’s test with statistical software SPSS 20.0 and P<0.05 was considered
significant. The Emax values for 18b, 18c, 18e, (R)-18a and (R)-18e were at the same
level and substantially less than those of isoprenaline, salmeterol and olodaterol (Fig.
14

3). The pD2 values of 18c and 18e were 6.28 and 8.08, respectively, which were
comparable to those of isoprenaline (6.00) and olodaterol (7.50) (Table 3). The Emax
values of 18a, 18f, (R)-18b and (R)-18f approached those of isoprenaline, salmeterol
and olodaterol (Fig. 4). The pD2 values of 18a, 18f and (R)-18f were similar to that of
isoprenaline but less than those of salmeterol and olodaterol. Compound (R)-18b had
a pD2 value of 8.04. The Emax values of 18d, (R)-18c and (R)-18d were similar or higher
than those of isoprenaline, salmeterol and olodaterol (Fig. 5). Compound (R)-18c had
a pD2 value of 7.78 and an Emax value of 219.43%. These results suggest that 18d, (R)-
18c and (R)-18d are full β2-AR agonists in this assay, and 18a-c, 18e, 18f, (R)-18a,
(R)-18b, (R)-18e and (R)-18f are partial agonists.
Fig. 3 Concentration–relaxation curves of olodaterol (Olo), salmeterol (Sal),
isoprenaline (Iso), 18c, 18e, 18b, (R)-18a and (R)-18e on contracted guinea pig trachea.
Data are expressed as relaxation percentages of the maximum contraction induced with
histamine
(5×10-5
M).
Error
bar
represents
S.D.
(n=4).
15

Fig. 4 Concentration–relaxation curves of olodaterol (Olo), salmeterol (Sal),
isoprenaline (Iso), 18f, (R)-18b, 18a and (R)-18f on contracted guinea-pig trachea.
Data are expressed as relaxation percentages of the maximum contraction induced with
histamine (5×10-5 M). Error bar represents S.D. (n=4).
Fig.5 Concentration–relaxation curves of olodaterol (Olo), salmeterol (Sal),
isoprenaline (Iso), (R)-18c, (R)-18d and 18d on contracted guinea pig trachea. Data are
expressed as relaxation percentages of the maximum contraction induced with
histamine (5×10-5 M). Error bar represents S.D. (n=4).
Table 3
Emax and pD2 of the smooth muscle relaxant responses of selected compounds 18a-f and
(R)-18a-f, isoprenaline, salmeterol and olodaterol.
16

Compound
18a
pD2
Emax (%)
6.14±0.32
5.76±0.25
6.28±0.29
6.56±0.34
8.08±0.32*&
6.46±0.29
5.92±0.26
8.04±0.54*&
7.78±0.48*
5.56±0.83
5.61±0.26
6.50±0.35
6.00±0.29
7.23±0.51
7.50±0.61
99.57±5.24
67.48±10.37
72.74±7.05
184.28±9.80
71.47±9.78
106.86±5.71
61.84±6.14
105.07±8.60
219.43±11.74*&#
203.79±10.92#
47.67±7.49
92.60±2.58
178.71±9.37
173.61±2.86
150.92±11.76
18b
18c
18d
18e
18f
(R)-18a
(R)-18b
(R)-18c
(R)-18d
(R)-18e
(R)-18f
isoprenaline
salmeterol
olodaterol
The pharmacodynamic parameters were derived from non-linear regression analysis of data from the experiments. *p <0.05, compared with
that of isoprenaline; &p <0.05, compared with that of salmeterol; #p <0.05, compared with that of olodaterol.
2.2.4 In vitro study of onset time and duration of action of compounds
Compounds producing strong relaxant effects on isolated tracheal strips, for instance
18d, (R)-18c and (R)-18d, were selected for further testing. The onset times and the
duration of action of olodaterol, salmeterol, isoprenaline, 18d, (R)-18c and (R)-18d
17

were determined graphically from the relaxation-time profiles (Fig. 6) and the
recovery-time profiles (Fig. 7), respectively (Table 4). The onset time is explained by
the Ot50 value which is the duration from application of the compound till attainment
of 50% of the Emax of the compound. Duration of action (DoA) is defined as the duration
from withdrawal of the compound producing a relaxant response till the relaxant effect
reduces by half. Compound 18d had an Ot50 value about two folds larger than that of
isoprenaline (3.10 min), the largest among the compounds tested. The Ot50 value of
(R)-18d was 3.73 min, which was smaller than that of salmeterol. Compound (R)-18c
had the smallest Ot50 value (3.50 min) among the compounds tested which was close to
that of isoprenaline. The Ot50 values of the three compounds were larger than that of
olodaterol, and their DoAs were longer than 12 h. Therefore, the three compounds are
expected to have equally fast action onset in vivo and behave as long-acting
bronchodilators.
Fig. 6 Relaxation-time profiles of olodaterol (Olo), salmeterol (Sal), isoprenaline (Iso),
(R)-18c and (R)-18d on guinea pig tracheal muscles pre-contracted with histamine.
Data are expressed as relaxation percentages of Iso (mean ± S.D., n=4).
18

Fig. 7 Recovery-time profiles of olodaterol (Olo), salmeterol (Sal), (R)-18c and (R)-
18d. Guinea pig tracheal muscles were pre-contracted with 5×10-5 M of histamine and
relaxed by addition of the test compounds. Recording of the recovery process began
upon withdrawal of the test compounds. Data are expressed as recovery percentages of
the test compounds (mean ± S.D., n=4).
Table 4
Onset time (Ot50) and duration of action (DoA) of the test compounds.
Compound
18d
Ot50 (min)
5.86±0.21
3.50±0.38
3.73±0.89
3.10±0.42
4.23±0.34
2.31±0.49
DoA (h)
﹥12
﹥12
﹥12
-
(R)-18c
(R)-18d
isoprenaline
salmeterol
olodaterol
﹥12
﹥12
The pharmacodynamic parameters of the tested compounds were determined on guinea pig tracheal muscles pre-contracted with 5×10-5
M
19

of histamine by non-linear regression analysis of the data in Fig. 6 and Fig. 7 using GraphPad Prism version 5.0. Data are expressed as
means ± S.D. (n=4).
2.3 Molecular modeling
In the subsequent molecular modeling study, flexible structures of the compounds
were applied to automated docking simulations. The crystallographic structure of the
22
β2-AR co-crystallized with hydroxybenzyl isoproterenol (HBI) (PDB: 4LDL) was
used as the template for docking. Docking was performed using Glide in Schrödinger
Maestro. Compounds (R)-18a, (R)-18c, (R)-18f with high biological activities were
chosen for the study. The binding pose of HBI in Fig. 8A shows that interactions with
the β2-AR involved hydrogen bonding of the two hydroxyl groups on the catechol head
with residues THR118 and SER207, which are considered necessary for β2-AR
activation.22 Within the common 8-(2-amino-1-hydroxyethyl)-6-hydroxy-1,4-
benzoxazine-3(4H)-one moiety of olodaterol, (R)-18a, (R)-18c and (R)-18f, the
oxygen atom of the amide also interacted with these two amino acid residues (Fig. 8B-
E). Besides, the oxygen atom, the nitrogen of the amide and the hydroxyl group in this
moiety also formed hydrogen bonds with ASP113, SER203 and ASN293, respectively.
These interactions may explain the superior activities of this series of compounds. The
2-amino-1-phenylethanol structure is regarded as the backbone of classical β2-AR
agonists because the interactions between the charged secondary amine group and the
β-hydroxyl group with ASN312 and ASP113 are indispensable for anchorage of the
agonist in the ligand-binding pocket of β2-AR. The docking simulations reveal that the
hydroxyl group and the secondary amine group of HBI, olodaterol, (R)-18a and (R)-
18c interacted with ASN312 and ASP113. But for (R)-18f, the secondary amine group
interacted with ASP113 and TYR316, and the β-hydroxyl group interacted with
TYR308. This maybe an alternative way to activate the receptor. The difference in the
lengths of carbon chains in the tail groups of the compounds could explain why the
20

para-hydroxyphenyl group of HBI interacted with CYS191 in extracellular loop 2
while the respective benzene moiety of (R)-18c and olodaterol interacted with PHE193
through π-σ interactions. Similarly, the meta-hydroxyphenyl group of (R)-18f formed
a hydrogen bond with ASP192. In contrast, no interaction was found between the
terminal phenyl group of (R)-18a and any amino acid residue in this region. The
identical interaction patterns of olodaterol and (R)-18c may explain the high and very
similar activities of the two compounds in various assays. Interestingly, for (R)-18f
compared to 18f, there was an increase in β2-mediated activity but a decrease in β1-
mediated activity. According to the modeling results, (R)-18f has a slightly different
interaction pattern with β2-adrenoceptor as compared to (R)-18a and (R)-18c. The
meta-hydroxyphenyl group of (R)-18f formed a hydrogen bond with ASP192 which is
not present in the interaction patterns of (R)-18a and (R)-18c. This hydrogen bonding
with ASP192 also appeared in the interaction pattern of olodaterol with the β2-
adrenoceptor, and (R)-18f were comparable to olodaterol in terms of activity and
selectivity. Therefore, the interaction with ASP192 may be specific to the β2-
adrenoceptor but not to the β1-adrenoceptor and important for β2-agonistic activity. The
superior activity and selectivity of (R)-18f as compared to 18f also implies that this
interaction plays an important role on β2-agonistic activity and β2/β1-selectivity. (R)-
18a and (R)-18f with slightly different interaction patterns were similarly active as
olodaterol and (R)-18c in the cell-based cAMP assay but somewhat less active in the
tracheal muscle relaxation assay. Together, the molecular modeling results suggest that
while the 8-(2-amino-1-hydroxyethyl)-6-hydroxy-1,4-benzoxazine-3(4H)-one head
group in this series of compounds contributes to a high receptor binding affinity, a high
activity also relies on proper binding interactions of the 2-amino-1-phenylethanol core
and the tail group with the receptor.
21

Fig. 8 Binding modes of HBI (A), olodaterol (B), compound (R)-18a (C), compound
(R)-18c (D) and compound (R)-18f (E).
2.4 General discussion
We have previously synthesized and tested a series of 5-(2-amino-1-
hydroxyethyl)-8-hydroxyquinolin-2(1H)-one derivatives for β2-AR stimulatory effects
and have identified two compounds 9g (Ref. 18) and (R)-18c (Ref. 18) with high
potency and β2/β1-selectivity.18 In the present study, compounds with the 8-(2-amino-
1-hydroxyethyl)-6-hydroxy-1,4-benzoxazine-3(4H)-one head group were studied.
Compound (R)-18c in this series has been found to possess a high activity with a
selectivity towards the β2-AR, a strong smooth muscle relaxant effect, a rapid onset of
action and a long duration of action. Molecular modeling reveals that compound (R)-
18c (present study) has 10 interactions with the β2-AR molecule (9 hydrogen bonds or
electrostatic interactions and 1 π-σ interaction). Compound 9g (Ref. 18) has 9 hydrogen
22

bonds or electrostatic interactions, and compound (R)-18c (Ref. 18) has 9 hydrogen
bonds or electrostatic interactions and 1 π-σ interaction.18 Their activities in stimulating
the β2-AR have been found to be 24 pM, 36 pM and 21 pM, respectively18. Thus,
activities of the compounds are proportional to the number of interactions with the β2-
AR molecule. However, compound (R)-18c possesses higher β1-adrenoceptor activity
and this influences its β2/β1-selectivity. In considering the high β2-AR agonistic
activities of compounds (R)-18a-f, the 8-(2-amino-1-hydroxyethyl)-6-hydroxy-1,4-
benzoxazine-3(4H)-one scaffold was a promising structure for further development.
Another class of β2-AR agonists synthesized in our laboratory include trantinterol23 and
its derivatives24,25, which contain a non-classical 2-amino-2-phenylethanol core
structure. This class of compounds are characterized by high β2/β1-selectivity but low
β2-AR agonistic activity (pEC50 values: 6 - 9) as compared with compounds containing
the 2-amino-1-phenylethanol structure (pEC50 values: 8 - 11).18,24,25
3. Conclusion
In summary, several potent β2-AR agonists have been identified from 8-(2-
amino-1-hydroxyethyl)-6-hydroxy-1,4-benzoxazine-3(4H)-one derivatives. Among
these compounds, compound (R)-18c with an EC50 value of 24 pM possesses the
highest β2-AR agonistic activity. It also has a strong airway smooth muscle relaxant
effect with a rapid onset of action (3.50 min) and a long duration of action (>12 h) in
isolated guinea pig trachea. In pharmacological assays, compound (R)-18c exhibits
similar efficacy and potency as compared with olodaterol though with a lower β2/β1-
23

selectivity. Molecular modeling reveals that (R)-18c assumes an optimal receptor
binding pattern same as olodaterol. These findings provide a theoretical basis for further
optimization of β2-AR agonists.
4. Experimental section
4.1. Chemistry
All reagents and solvents were purchased through commercial sources and were
used without further purification. Anhydrous solvents were prepared by distillation
over a silk of sodium. Reaction products were monitored by TLC (Silica gel60 GF254,
Merck) and visualized under UV light. If necessary, crude products were purified on
silica gel (200-300 mesh) by column chromatography. The purity of all test compounds
was more than 95%. 1H-NMR and 13C-NMR spectra were recorded on a Bruker ARX-
400-MHz or Bruker ARX-600-MHz spectrometer. High resolution mass spectrometry
(HRMS) was performed on a Bruker Micromass Time of Flight mass spectrometer
equipped with an electrospray ionization (ESI) probe. Melting points were measured
by a Buchi B-540 melting point meter.
4.1.1. General procedures for the synthesis of compounds 10a-l.
t-BuOK (5.66 g, 50.0 mmol) was added to a stirred solution of anhydrous THF
containing an appropriate benzaldehyde derivative (7a-l) (5.00 g, 42.0 mmol) and 2-
diethoxyphosphorylacetonitrile (8.10 g, 46.0 mmol) at 0°C for 30 min. Then, the
reaction was maintained at 25°C for a further 30 min. The mixture was extracted with
ethyl acetate and the combined organics were washed with brine, dried over sodium
24

sulfate, and concentrated to give compounds 8a-l. To a solution of 8a-l in ethyl acetate,
5% Pd/C (0.2 equiv.) was added. The suspension was filtered and evaporated. Then,
the mixture was maintained at 30°C under 1 atm of hydrogen for 1 h. Removal of the
solvent by rotary evaporation gave 9a-l. Under nitrogen protection, BH3∙THF (2 mol/L,
20.7 mL, 41.4 mmol) was added slowly to a solution of 9a-l (4.12 g, 27.6 mmol) in
anhydrous THF at room temperature. Then, the reaction was refluxed at 60°C for 2 h.
The mixture was extracted with ethyl acetate and the combined organics were washed
with brine and dried over sodium sulfate, and concentrated to give compounds 10a-l.
4.1.2. General procedures for the synthesis of compounds 14a-h.
Compounds 14a-h were synthesized according to reported methods.18
4.1.3. 6-(benzyloxy)-8-(2,2-dihydroxyacetyl)-2H-benzo[b][1,4]oxazin-3(4H)-one (16)
Intermediate 15 (20.00 g, 67.2 mmol) was suspended in DMSO (120 mL) and 48%
hydrobromic acid (40.78 g, 201.6 mmol) was added. Then the suspension was stirred
for 1.5 h at 80°C. The resulting yellow solution was cooled to 25°C and extracted three
times with ethyl acetate. The combined organics were washed with brine, dried over
sodium sulfate, and concentrated in a rotary evaporator to give a white solid 16.
4.1.4. General procedures for the synthesis of compounds 17-1 - 17-4.
The purchased amines (1.1 equiv.) was added into intermediate 16 (1.0 equiv.)
dissolved in DMSO and the reaction was stirred at 35°C for more than 4 h. Then,
sodium borohydride (1.0 equiv.) was added with a suitable amount of methanol and the
reaction was stirred for a further 1 h at 0°C. After removal of the methanol by rotary
25

evaporation, the concentrate was purified by column chromatography (silica gel,
CH2Cl2-MeOH 50:1) to give a white solid. The white solid dissolved in MeOH and
added ethyl acetate-HCl, stirred at room temperature for 2 h, the suspension was filtered
and yield white solid. The desired fraction was dissolved in methanol and mixed with
5% Pd/C (0.2 equiv.). The reaction was maintained at 30°C under 1 atm of hydrogen
for 1 h. The suspension was filtered and evaporated to afford compounds 17-1 - 17-4.
4.1.4.1. 6-hydroxy-8-(1-hydroxy-2-(phenethylamino)ethyl)-2H-benzo[b][1,4]oxazin-
3(4H)-one hydrochloride (17-1).
White solid; mp: 148.1-150.4 °C; yield 96 %; 1H-NMR (600 MHz, DMSO-d6) δ 10.67
(s, 1H), 9.36 (s, 1H), 9.15 (s, 1H), 8.85 (s, 1H), 7.36 (t, J = 7.4 Hz, 2H), 7.28 (d, J = 7.5
Hz, 3H), 6.56 (s, 1H), 6.38 (s, 1H), 6.13 (d, J = 3.4 Hz, 1H), 5.14 (d, J = 9.8 Hz, 1H),
13
4.55 – 4.47 (m, 2H), 3.23 (m, 3H), 3.04 (m, 3H). C-NMR (151 MHz, DMSO-d6) δ
165.8, 153.0, 137.8, 132.8, 130.4, 130.0, 129.4, 129.2, 129.1, 128.2, 127.2, 107.0, 102.7,
67.4, 63.3, 52.3, 48.6, 31.8. Calcd. for C18H20N2O4 [M+H]+ 329.1423; found 329.1449.
4.1.4.2.
8-(2-((4-fluorophenethyl)amino)-1-hydroxyethyl)-6-hydroxy-2H-
benzo[b][1,4]oxazin-3(4H)-one hydrochloride (17-2).
White solid; mp: 148.1-149.7 °C; yield 97 %; 1H-NMR (600 MHz, DMSO-d6) δ 10.65
(s, 1H), 9.34 (s, 1H), 9.14 (s, 1H), 8.81 (s, 1H), 7.33 – 7.28 (m, 2H), 7.17 (t, J = 8.6 Hz,
2H), 6.54 (d, J = 2.1 Hz, 1H), 6.37 (s, 1H), 6.12 (s, 1H), 5.12 (d, J = 9.6 Hz, 1H), 4.52
– 4.45 (m, 2H), 3.14 (dd, J = 22.0, 6.5 Hz, 3H), 2.96 (ddd, J = 33.6, 17.9, 8.6 Hz, 3H).
13C-NMR (151 MHz, DMSO-d6) δ 165.8, 162.4, 160.8, 153.0, 133.9, 132.7, 131.0,
26

130.4, 128.2, 115.9, 115.7, 107.0, 102.7, 67.4, 63.3, 52.3, 48.5, 30.9. Calcd. for
C18H19FN2O4 [M+H]+ 347.1329; found 347.1413.
4.1.4.3.
6-hydroxy-8-(1-hydroxy-2-((4-methoxyphenethyl)amino)ethyl)-2H-
benzo[b][1,4]oxazin-3(4H)-one hydrochloride (17-3)
White solid; mp: 229.9-230.2 °C; yield 95 %; 1H-NMR (600 MHz, DMSO-d6) δ 10.65
(s, 1H), 9.34 (s, 1H), 9.10 (s, 1H), 8.78 (s, 1H), 7.17 (d, J = 8.6 Hz, 2H), 6.89 (d, J =
8.6 Hz, 2H), 6.53 (d, J = 2.7 Hz, 1H), 6.36 (d, J = 2.7 Hz, 1H), 6.11 (d, J = 4.0 Hz, 1H),
5.14 – 5.09 (m, 1H), 4.49 (q, J = 15.0 Hz, 2H), 3.73 (s, 3H), 3.11 (s, 3H), 2.91 (pd, J =
13.3, 6.2 Hz, 3H). 13C-NMR (151 MHz, DMSO-d6) δ 165.8, 158.5, 153.0, 132.7, 130.4,
130.1, 130.1, 129.5, 128.2, 114.5, 114.5, 107.0, 102.7, 67.4, 63.3, 55.5, 52.3, 48.8, 30.9.
Calcd. for C19H22FN2O5 [M+H]+ 359.1529; found 359.1610.
4.1.4.4.
6-hydroxy-8-(1-hydroxy-2-((3-phenylpropyl)amino)ethyl)-2H-
benzo[b][1,4]oxazin-3(4H)-one hydrochloride (17-4)
White solid; mp: 203.7-205.0 °C; yield 97 %; 1H-NMR (400 MHz, DMSO-d6) δ 10.60
(s, 1H), 9.28 (s, 1H), 8.82 (s, 1H), 8.62 (s, 1H), 7.27 (t, J = 7.5 Hz, 2H), 7.17 (dd, J =
15.3, 7.5 Hz, 3H), 6.50 (d, J = 2.0 Hz, 1H), 6.34 (s, 1H), 6.03 (s, 1H), 5.06 (d, J = 9.8
Hz, 1H), 4.44 (t, J = 8.9 Hz, 2H), 3.02 (s, 1H), 2.95 – 2.79 (m, 3H), 2.57 (t, J = 6.8 Hz,
2H), 1.60 (d, J = 3.5 Hz, 4H). 13C-NMR (151 MHz, DMSO-d6) δ 165.7, 153.0, 142.1,
132.7, 130.4, 128.7, 128.7, 128.2, 126.4, 126.2, 107.1, 102.7, 67.4, 63.3, 52.3, 47.3,
35.0, 28.2, 25.2. Calcd. for C20H24N2O4 [M+H]+ 357.1736; found 357.1812.
4.1.5. General procedures for the synthesis of compounds 17a-l.
27

An appropriate amine (10a-l) (1.1 equiv.) was added into intermediate 16 (1.0
equiv.) dissolved in DMSO and the reaction was stirred at 35°C for more than 4 h.
Then, sodium borohydride (1.0 equiv.) was added with a suitable amount of methanol
and the reaction was stirred for a further 1 h at 0°C. After removal of the methanol by
rotary evaporation, the concentrate was purified by column chromatography (silica gel,
CH2Cl2-MeOH 50:1) to give a white solid. The white solid dissolved in MeOH and
added ethyl acetate-HCl, stirred at room temperature for 2 h, the suspension was filtered
and yield white solid. The desired fraction was dissolved in methanol and mixed with
5% Pd/C (0.2 equiv.). The reaction was maintained at 30°C under 1 atm of hydrogen
for 1 h. The suspension was filtered and evaporated to afford compounds 17a-l.
4.1.5.1.
6-hydroxy-8-(1-hydroxy-2-((3-phenylpropyl)amino)ethyl)-2H-
benzo[b][1,4]oxazin-3(4H)-one hydrochloride (17a).
White solid; mp: 148.1-150.4 °C; yield 96%; 1H-NMR (600 MHz, DMSO-d6) δ 10.64
(s, 1H), 9.33 (s, 1H), 8.99 (s, 1H), 8.76 (s, 1H), 7.31 (t, J = 7.4 Hz, 2H), 7.26 – 7.16 (m,
3H), 6.52 (s, 1H), 6.36 (s, 1H), 6.07 (s, 1H), 5.09 (s, 1H), 4.48 (q, J = 15.0 Hz, 2H),
3.06 (d, J = 12.2 Hz, 1H), 2.94 (t, J = 7.7 Hz, 2H), 2.87 (t, J = 11.2 Hz, 1H), 2.64 (t, J
= 7.6 Hz, 2H), 1.96 (d, J = 7.3 Hz, 2H). 13C-NMR (151 MHz, DMSO-d6) δ 165.8, 153.0,
141.2, 132.7, 130.4, 128.9, 128.9, 128.7, 128.7, 128.2, 126.5, 107.1, 102.6, 67.4, 63.3,
52.3, 47.1, 32.4, 27.3. Calcd. for C19H22N2O4 [M+H]+ 343.1580; found 343.1651.
4.1.5.2.
8-(2-((3-(4-fluorophenyl)propyl)amino)-1-hydroxyethyl)-6-hydroxy-2H-
benzo[b][1,4]oxazin-3(4H)-one hydrochloride (17b).
28

White solid; mp: 148.1-149.7 °C; yield 97%; 1H-NMR (600 MHz, DMSO-d6) δ 10.64
(s, 1H), 9.33 (s, 1H), 9.00 (s, 1H), 8.74 (s, 1H), 7.29 – 7.23 (m, 2H), 7.13 (dd, J = 12.3,
5.4 Hz, 2H), 6.52 (d, J = 2.7 Hz, 1H), 6.36 (d, J = 2.8 Hz, 1H), 6.07 (d, J = 3.9 Hz, 1H),
5.09 (dd, J = 6.8, 3.3 Hz, 1H), 4.48 (q, J = 15.0 Hz, 2H), 3.06 (d, J = 11.9 Hz, 1H), 2.92
(t, J = 7.7 Hz, 2H), 2.89 – 2.85 (m, 1H), 2.63 (t, J = 7.6 Hz, 2H), 1.98 – 1.90 (m, 2H).
13C-NMR (151 MHz, DMSO-d6) δ 165.8, 162.0, 160.4, 153.0, 137.3, 130.5, 130.5,
130.4, 130.4, 128.2, 115.6, 107.1, 102.6, 67.4, 63.3, 52.3, 47.0, 31.5, 27.4. Calcd. for
C19H21FN2O4 [M+H]+ 361.1485; found 361.1567.
4.1.5.3.
6-hydroxy-8-(1-hydroxy-2-((3-(4-hydroxyphenyl)propyl)amino)ethyl)-2H-
benzo[b][1,4]oxazin-3(4H)-one hydrochloride (17c).
White solid; mp: 239.7-240.8 °C; yield 98%; 1H-NMR (600 MHz, DMSO-d6) δ 10.64
(s, 1H), 9.32 (s, 1H), 9.25 (s, 1H), 8.92 (s, 1H), 8.69 (s, 1H), 6.99 (d, J = 8.4 Hz, 2H),
6.69 (d, J = 8.4 Hz, 2H), 6.52 (d, J = 2.6 Hz, 1H), 6.35 (d, J = 2.8 Hz, 1H), 6.06 (d, J =
3.6 Hz, 1H), 5.10 – 5.06 (m, 1H), 4.48 (q, J = 15.0 Hz, 2H), 3.05 (s, 1H), 2.93 – 2.83
(m, 3H), 2.52 (d, J = 9.2 Hz, 2H), 1.89 (dt, J = 12.5, 6.9 Hz, 2H).13C-NMR (151 MHz,
DMSO-d6) δ 165.8, 156.0, 153.0, 132.7, 131.1, 130.4, 130.4, 129.5, 128.2, 115.6, 115.6,
107.0, 102.6, 67.4, 63.3, 52.3, 47.1, 31.6, 27.6. Calcd. for C19H22N2O5 [M+H]+
359.1529; found 359.1602.
4.1.5.4.
6-hydroxy-8-(1-hydroxy-2-((3-(2-hydroxyphenyl)propyl)amino)ethyl)-2H-
benzo[b][1,4]oxazin-3(4H)-one hydrochloride (17d).
White solid; mp: 212.1-213.0 °C; yield 96%; 1H-NMR (600 MHz, DMSO-d6) δ 10.64
(s, 1H), 9.32 (s, 1H), 9.25 (s, 1H), 8.92 (s, 1H), 8.69 (s, 1H), 6.99 (d, J = 8.4 Hz, 2H),
6.69 (d, J = 8.4 Hz, 2H), 6.52 (d, J = 2.6 Hz, 1H), 6.35 (d, J = 2.8 Hz, 1H), 6.06 (d, J =
3.6 Hz, 1H), 5.10 – 5.06 (m, 1H), 4.48 (q, J = 15.0 Hz, 2H), 3.05 (s, 1H), 2.93 – 2.83
29