Tryptamine-based human β3-adrenergic receptor agonists. Part 3: Improved oral bioavailability via modification of the sulfonamide moiety

Article abstract of DOI:10.1016/j.bmcl.2004.12.033

The continued SAR investigation of tryptamine-based human β₃-adrenergic receptor (AR) agonists is reported. Prior efforts resulted in the identification of 2 as a potent β₃-AR agonist. Further modification of the left side arylsulfonamide portion in 2 provided compounds with good cell permeability, which have potent agonistic activity for β₃-AR. Cinnamylamine analog 16i exhibited an excellent agonistic profile in vitro and good oral bioavailability in rats.

Full text of DOI:10.1016/j.bmcl.2004.12.033

Bioorganic & Medicinal Chemistry Letters 15 (2005) 1061–1064
Tryptamine-based human b3-adrenergic receptor agonists.
Part 3: Improved oral bioavailability via modification
of the sulfonamide moiety
Masaaki Sawa,a,* Kazuhiro Mizuno,a Hiroshi Harada,a Hirotaka Tateishi,a Yukiyo Arai,b
Shinya Suzuki,b Mayumi Oue,c Hiroshi Tsujiuchi,c Yasuji Furutanic and Shiro Katoa
aChemistry Research Loboratories, Dainippon Pharmaceutical Co., Ltd, Enoki 33-94, Suita, 564-0053, Japan
bPharmacokinetics & Physico-Chemical Property Research Laboratories, Dainippon Pharmaceutical Co., Ltd, Enoki 33-94,
Suita, 564-0053, Japan
cPharmacology & Microbiology Research Laboratories, Dainippon Pharmaceutical Co., Ltd, Enoki 33-94, Suita, 564-0053, Japan
Received 26 October 2004; revised 10 December 2004; accepted 10 December 2004
Available online 18 January 2005
Abstract—The continued SAR investigation of tryptamine-based human b3-adrenergic receptor (AR) agonists is reported. Prior
efforts resulted in the identification of 2 as a potent b3-AR agonist. Further modification of the left side arylsulfonamide portion
in 2 provided compounds with good cell permeability, which have potent agonistic activity for b3-AR. Cinnamylamine analog
16i exhibited an excellent agonistic profile in vitro and good oral bioavailability in rats.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
of the indole ring was important for both activity and
selectivity. Although the optimization of b3-agonistic
activity within this series led us to several potent and
selective b3-agonists exemplified by 2, these compounds
had extremely low oral bioavailability due to the lack of
cellular permeability.
In an effort to find novel therapy for the treatment of
diseases such as non-insulin dependent diabetes mellitus
(NIDDM), obesity, and frequent urination, we have re-
cently disclosed a series of selective human b3-adrenergic
receptor (AR) agonists based on a tryptamine scaffold
(Fig. 1).1–3 It has been suggested that stimulation of
b3-AR results in various pharmacological effects,
such as lipolysis and thermogenesis in adipocytes,4 and
relaxation of urinary bladder detrusor urination.5 The
structure–activity relationship (SAR) study of the
tryptamine-based analogs revealed that the 7-position
In the course of our work on the tryptamine-based ana-
logs, it became clear that Caco-2 cell monolayer perme-
ability was greatly decreased due to the acidic property
of incorporating a sulfonamide moiety into the left-hand
side of the molecule. In fact, few compounds in this ser-
ies lacking an acidic proton of the sulfonamide exhibited
improved Caco-2 cell permeability (3a vs 3b and 3c, Ta-
ble 1), although they showed moderate agonistic activ-
ity. These results prompted us to further optimize the
left side of the molecule. We now wish to report further
on the results of our investigation into the SAR of the
tryptamine-based b3-AR agonists in activity and Caco-
2 cell permeability.
OH
OH
H
N
H
N
H
N
Cl
Ar
S
O
O
N
H
N
H
Me
AJ-9677 (1)
2
O
CO2H
O
S
O
O
Figure 1. Structures of tryptamine-based human b3-AR agonists.
2. Chemistry
Keywords: b3-Adrenergic receptor; Agonist; Tryptamine.
Corresponding author. Tel.: +81 6 6337 5904; fax: +81 6 6338
7656; e-mail: masaaki-sawa@dainippon-pharm.co.jp
*
The benzenesulfonamide 3a and N-methyl benzenesulf-
onamide 3b were prepared as shown in Scheme 1. Thus,
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.12.033

1062
M. Sawa et al. / Bioorg. Med. Chem. Lett. 15(2005) 1061–1064
Table 1. Activity and cell permeability of tryptamine-based b3-agonists
OH
H
N
X
N
H
Me
O
S
O
O
Compd
X
EC50, nMa (IA, %)b
Permeabilityc
b3
b1
b2
Papp · 10À6 cm/s
3a
3b
3c
3d
3e
3f
–SO2NH–
–SO2N(Me)–
–SO2O–
0.54 (90)
52 (38)
32 (60)
160 (96)
100 (54)
15 (90)
11 (87)
13 (62)
(3)d
(13)d
120 (96)
(4)d
19 (46)
(11)d
nd (low)
4.7 (high)
7.0 (high)
(4)d
–CONH–
–SO2CH2–
–CH2O–
69 (82)
(7)d
0.9 (moderate)
0.9 (moderate)
6.7 (high)
(21)
(28)
46 (35)
16 (43)
3g
–CH2NH–
9.0 (high)
a Agonistic activity was assessed by measuring cAMP accumulation in CHO cells expressing human b-ARs.
b Values in parentheses represent the intrinsic activity (IA) as a percentage of maximal stimulation with isoproterenol.
c Permeability of compounds was determined in Caco-2 cell monolayer permeability assay. The test compounds were classified for their absorption
potential using high (>1.0), moderate (0.5–1.0) and low (<0.5) descriptions. nd = Not detected.
d EC50 was not determined. Values in parentheses represent percent activity at 1000 nM.
O
O
OTBDMS
Br
O
OH
OTBDMS
Br
H
N
b, c, d
O2N
Br
a
O2N
Br
HO
BnO
Br
BnO
a, b
c, d
S
O
O
4
5
6
8
9
10
OH
OH
H
N
H
H
N
H2N
e, f
BnO
e, f
N
O
6
S
+
O
N
H
N
H
N
OMs
H
OMs
OMs
3f
7
3a
OH
OH
H
g, h, i, j
H
g, h, i
N
N
O
O
N
S
S
O
O
O
N
H
N
OMs
H
3b
OMs
3c
Scheme 1. Reagents: (a) (R)-2-methyl-CBS-oxazaborolidine, BH3,
THF; (b) TBDMSCl, imidazole, DMF; (c) Fe, NH4Cl, EtOH, H2O;
(d) benzenesulfonyl chloride, pyridine, CH2Cl2; (e) i-Pr2NEt, KI,
MeCN; (f) 4 N HCl–AcOEt, EtOH; (g) (Boc)2O, CHCl3; (h) MeI,
K2CO3, DMF; (i) 4 N HCl–AcOEt.
Scheme 2. Reagents: (a) benzyl bromide, K2CO3, DMF; (b) phenyl-
trimethylammonium tribromide, THF; (c) (R)-2-methyl-CBS-oxaza-
borolidine, BH3, THF; (d) TBDMSCl, imidazole, DMF; (e) 7,
i-Pr2NEt, KI, MeCN; (f) 4 N HCl–AcOEt, EtOH; (g) (Boc)2O,
CHCl3; (h) H2, 20% Pd(OH)2, EtOH; (i) benzenesulfonyl chloride,
Et3N, CH2Cl2; (j) 4 N HCl–AcOEt.
commercially available m-nitrophenacyl bromide 4 was
reduced stereoselectively with borane–THF complex in
the presence of (R)-2-methyl-CBS-oxazaborolidine pro-
viding chiral alcohol 5.2 Protection of the chiral alcohol
5 with TBDMS, followed by reduction of the nitro
group with Fe, gave the corresponding aniline, which
was treated with benzenesulfonyl chloride to afford 6.
Coupling of the resulting bromide 6 with (R)-a-
methyl-7-methanesulfonyloxytryptamine 72 in the pres-
ence of diisopropylethylamine and KI, followed by
deprotection of TBDMS afforded the benzenesulfon-
amide 3a, which was then converted to the N-methyl
benzenesulfonamide 3b through a three-step procedure
involving protection of the secondary amine, N-methyl-
ation, and deprotection.
ammonium tribromide gave 9. Stereoselective reduction
of 9 and protection of the resulting alcohol afforded
10. The bromide 10 was then allowed to react with trypt-
amine 7 and deprotected to give the benzyloxy analog
3f. Protection of the secondary amine with Boc and
the benzyl group was removed by hydrogenolysis,
exposing the phenolic alcohol, which was functionalized
through esterification with benzenesulfonyl chloride.
Deprotection of the Boc group afforded the desired benz-
enesulfonate 3c.
The benzenesulfonylmethyl analog 3e was prepared
according to Scheme 3. The key step was the palla-
dium-catalyzed conversion of an aryl bromide to the
corresponding acetophenone 12. The reaction of 3-bro-
mobenzyl bromide 11 with benzenesulfinic acid sodium
salt gave the benzenesulfonylmethylphenyl bromide,6
which was used to palladium-catalyzed a-arylation of
butyl vinyl ether.7 The resulting acetophenone 12 was
brominated, reduced stereoselectively, protected as de-
The benzenesulfonate 3c was synthesized similarly to the
preparation of 3a, except reversing the order of the benz-
enesulfonyl group introduction (Scheme 2). Protection
of the phenol 8 with the benzyl group, followed by bro-
mination of the acetyl group using phenyltrimethyl-

M. Sawa et al. / Bioorg. Med. Chem. Lett. 15(2005) 1061–1064
1063
OH Boc
N
O
OTBDMS
Br
c, d, e
O
a, b
Br
Br
H2N
O2N
a
b, c, d
S
S
5
O
O
O
O
N
H
11
12
13
14
15
OMs
OH
e, f
f, g
H
N
g, f
S
O
O
N
H
OH
OH
3e
OMs
H
N
H
N
H
N
H
N
R
O
Scheme 3. Reagents: (a) benzenesulfinic acid sodium salt, DMF; (b)
butyl vinyl ether, K2CO3, Pd(OAc)2, 1,3-bis(diphenylphosphino)pro-
pane, DMF, H2O; (c) phenyltrimethylammonium tribromide, THF;
(d) (R)-2-methyl-CBS-oxazaborolidine, BH3, THF; (e) TBDMSCl,
imidazole, DMF; (f) 7, i-Pr2NEt, KI, MeCN; (g) 4 N HCl–AcOEt,
EtOH.
N
H
N
H
3d
16a - i
OMs
OMs
Scheme 4. Reagents: (a) 2 N NaOH, THF; (b) 7, DMF; (c) (Boc)2O,
CHCl3; (d) Fe, NH4Cl, EtOH, H2O; (e) PhCOCl, Et3N, CHCl3; (f) 4 N
HCl–AcOEt; (g) R–Br, K2CO3 or Et3N, DMF.
All final compounds synthesized were characterized
using NMR, mass spectrometry, and HPLC.
scribed in Scheme 2, and provide bromide 13. Alkylation
of tryptamine 7 with bromide 13 and the subsequent
deprotection of Boc afforded the benzenesulfonylmethyl
analog 3e.
3. Biological evaluation
The general procedure for the preparation of 3-aniline
derivatives with various R groups, through the alkyl-
ation of the key aniline intermediate 15, is listed in Table
2 and shown in Scheme 4,. Therefore, the chiral alcohol
5 was treated with alkaline to form epoxide 14. The reac-
tion of the resulting epoxide 14 with tryptamine 7 affor-
ded a secondary amine, which was then protected with
Boc, and subsequently reduced the nitro group to give
the aniline 15. The functionalization of the amino group
was accomplished by alkylation with the appropriate al-
lyl or aralkyl bromide. Finally, deprotection of the Boc
group furnished the desired anilines 16a–i. Alternatively,
the reaction of the aniline 15 with benzoyl chloride, fol-
lowed by deprotection of the Boc group afforded the
amide 3d.
All compounds were tested in vitro for their ability to
stimulate cAMP accumulation in CHO cells expressing
the cloned human b1-, b2-, and b3-ARs.8 Compounds
with high potency were evaluated for their ability to per-
meate Caco-2 cell monolayers.9
To identify compounds with improved Caco-2 cell per-
meability, we initially prepared tryptamine analogs with
various linkages instead of the sulfonamide linkage as
shown in Table 1. The replacement of the sulfonyl group
with carbonyl (amide 3d) resulted in a substantially
diminished potency for b3-AR. The geometry of the pla-
nar amide in this position is not well-tolerated by
b-ARs. A dramatic decrease in potency for all b-ARs
was observed when the sulfonamide nitrogen atom was
exchanged for a methylene group (sulfone 3e), indicat-
ing that the heteroatoms play a role in receptor binding.
The importance of this heteroatom was further con-
firmed with the improved potency of the benzyloxy
(3f) and benzylamine (3g) analogs. Interestingly, the
benzyloxy analog 3f exhibited good agonistic activity
for b3-AR with improved Caco-2 cell permeability,
whereas the sulfonate analog 3c was a weak agonist
and was only partially agonistic for b3-AR. The benzyl-
amine analog 3g also exhibited good potency for b3-AR
with excellent subtype selectivity. It was noteworthy that
benzylamine 3g showed considerably improved Caco-2
cell monolayer permeability.
Table 2. Activity of tryptamine-based b3-agonistsa
OH
H
R
N
N
H
Me
O
S
O
O
Compd
R
EC50, nM (IA, %)
b3
b1
b2
16a
16b
16c
3.2 (110)
1.2 (102)
9.0 (101)
(49)d
2.2 (78)
H
N
H
170 (30)
(38)d
10 (64)
11 (49)
N
H
N
H
N
Since benzylamine analog 3g exhibited a good agonistic
profile with high cellular permeability properties, we
used the aniline portion of 3g as a template for a second
SAR study wherein substituents on amino groups were
examined (Table 2). We designed some allyl derivatives
to reduce the molecular weight in anticipation of further
improvement of cellular permeability. The replacement
of the benzene ring in 3g with a simple olefin dramati-
cally improved agonistic activity for b3-AR, but resulted
in an increase in potency against b2-AR (16a–c vs 3g).
Crotyl analog 16b exhibited maximum potency in this
series (EC50 = 1.2 nM, IA = 102%). Geminal substitu-
tion of the double bond in 16a was unfavorable to
16d
36 (79)
160 (51)
36 (59)
(27)d
1.5 (60)
9.6 (67)
18 (61)
H
N
16e
16f
16g
17 (114)
10 (102)
6.5 (90)
H
N
45 (52)
(45)d
H
N
H
N
16h
16i
2.9 (107)
1.0 (97)
44 (58)
(36)d
7.6 (75)
13 (81)
H
N
a See footnotes in Table 1.
d EC50 was not determined. Values in parentheses represent percent
activity at 1000 nM.

1064
M. Sawa et al. / Bioorg. Med. Chem. Lett. 15(2005) 1061–1064
Table 3. Binding affinity and cell permeability of tryptamine-based
b3-agonists
monolayer permeability. The cinnamylamine analog
16i exhibited good permeability and showed improved
oral bioavailability in rats. Further studies are being
conducted to optimize the in vitro profile of this series
and will be reported in due course.
Compd
Binding Ki, nMa
Permeabilityb
b3
3.3
b1
b2
Papp · 10À6 cm/s
3a
44
150
48
23
18
25
nd (low)
7.3 (high)
2.1 (high)
16b
16i
33
15
Acknowledgements
a Binding potency is reported as Ki, the binding inhibition constant,
determined by inhibition of 125I-iodocyanopindolol binding.
b See footnote c in Table 1.
We would like to thank San Francisco Edit (www.sfedit.
net) for their assistance in editing this manuscript.
b3-agonistic activity (16d and 16e). Interestingly, 16e
showed strong activity against b2-AR (EC50 = 1.5 nM),
while the potency against b1-AR was extremely weak.
Acetylene substituents tended to exert little influence
on the potency for b3-AR (16f and 16g vs 3g).
References and notes
1. Harada, H.; Hirokawa, Y.; Suzuki, K.; Hiyama, Y.; Oue,
M.; Kawashima, H.; Yoshida, N.; Furutani, Y.; Kato, S.
Bioorg. Med. Chem. Lett. 2003, 13, 1301–1305.
2. Mizuno, K.; Sawa, M.; Tateishi, H.; Harada, H.; Oue, M.;
Tsujiuchi, H.; Furutani, Y.; Kato, S. Bioorg. Med. Chem.
Lett. 2004, 14, 5959–5962.
3. Sawa, M.; Tateishi, H.; Mizuno, K.; Harada, H.; Oue, M.;
Tsujiuchi, H.; Furutani, Y.; Kato, S. Bioorg. Med. Chem.
Lett. 2004, 14, 5963–5966.
4. de Souza, C. J.; Burkey, B. F. Curr. Pharm. Des. 2001, 7,
1433–1449.
5. Yamaguchi, O. Urology 2002, 59(Suppl 5A), 25–29.
6. Ghera, E.; Plemenitas, A.; Ben-David, Y. Synthesis 1984,
504–506.
A very interesting observation was made upon the
expansion of the benzene ring to the naphthalene ring
(Table 2). The b3-agonistic potency of 16h increased
4-fold compared to that of 3g. As with the olefin com-
pounds, geminal substitution of the double bond was ex-
pected to decrease the potency for b3-AR, we therefore,
designed cinnamylamine analog 16i. As expected, cinn-
amylamine analog 16i showed strong activity for b3-
AR (EC50 = 1.0 nM, IA = 97%). This compound also
showed excellent selectivity against b1-AR and modest
activity against b2-AR.
7. Vallin, K. S. A.; Larhed, M.; Hallberg, A. J. Org. Chem.
2001, 66, 4340–4343.
8. In a previous report,1 we used CHO cells expressing a high
level of b3-AR to measure compound activity, that is the
receptor densities were 150,000 receptors/cell (b3-AR),
12,000 receptors/cell (b1-AR), and 30,000 receptors/cell
(b2-AR). To better evaluate the subtype selectivity, we
used CHO cells expressing low densities of b3-AR (13,000
receptors/cell), and high densities of b1- and b2-ARs in this
report (320,000 and 600,000 receptors/cell, respectively).
The CHO cells expressing either human b1-, b2-, or b3-AR
were prepared as described in Kato, S.; Harada, H.;
Taoka, I.; Kawashima, H. PCT Patent Application, WO
2000044721, 2000; Kato, S.; Harada, H.; Hirokawa, Y.;
Yoshida, N.; Kawashima, H. PCT Patent Application,
WO 9616938, 1996.
9. The apparent permeability coefficients (Papp) were deter-
mined in the apical-to-basolateral (A-to-B) direction
across the cellÕs monolayers cultured on polycarbonate
membrane filters.
10. Compound 16i was dosed orally at 5 mg/kg in 1%
hydroxypropyl cellulose aqueous solution and intrave-
nously at 0.5 mg/kg in a vehicle consisting of 10% DMSO,
10% EtOH, 20% polyethylene glycol and saline. After
dosing, blood samples were collected at 5 and 15 min (iv
only), and 0.5, 1, 2, 4, 6, 8, 24 h (ivand po) to determine
plasma drug concentrations by LC/MS/MS.
To evaluate the possibility of this series for drug candi-
dates, the selected compounds were subjected to recep-
tor binding assays. As shown in Table 3, the binding
potency of crotyl analog 16b decreased by one order
of magnitude compared to that of 3a (Ki = 33 and
3.3 nM, respectively). Furthermore, 16b exhibited rever-
sal selectivity in binding regarding b3 versus b2-AR
compared with the agonistic activity. Cinnamylamine
analog 16i also showed a decreased binding constant
(Ki = 15 nM), but possessed moderate selectivity against
b1 and 2-AR. Since the cinnamylamine analog 16i
showed high Caco-2 cellular permeability (Papp = 2.1
Papp · 10À6 cm/s), the pharmacokinetic property was
determined in rats (5 mg/kg po, 0.5 mg/kg iv).10 As ex-
pected from the Caco-2 experiments, a large improve-
ment was observed with the cinnamylamine analog
16i; its oral bioavailability in the rat was 21%. In addi-
tion, the half-life of 16i was 4.2 h. These data indicate
that the sulfonamide moiety is detrimental to oral
absorption.
In conclusion, we have discovered aniline analogs as po-
tent and selective b3-agonists with excellent Caco-2 cell