Discovery of potent α1L-adrenoceptor agonists: Design and synthesis of bicyclic derivatives

Article abstract of DOI:10.1016/j.bmcl.2015.05.049

Abstract We aimed to create a novel and potent α_1L-adrenoceptor agonist because such agonists are possible drug candidates for stress urinary incontinence. We used ligand-based drug design and evaluated the α_1L-adrenoceptor agonist a

Full text of DOI:10.1016/j.bmcl.2015.05.049

Accepted Manuscript
Discovery of potent α1L-adrenoceptor agonists: Design and synthesis of bicyclic
derivatives
Shinya Suzuki, Tsubasa Okano, Rie Horiuchi, Nana Hareyama, Kazutoshi
Amikura, Naoyoshi Yamamoto, Yoshitaka Yoshizawa, Mai Yagi, Kanako
Serizawa, Ryoji Hayashi
PII:
S0960-894X(15)00511-9
DOI:
http://dx.doi.org/10.1016/j.bmcl.2015.05.049
BMCL 22739
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
15 January 2015
29 April 2015
15 May 2015
Please cite this article as: Suzuki, S., Okano, T., Horiuchi, R., Hareyama, N., Amikura, K., Yamamoto, N.,
Yoshizawa, Y., Yagi, M., Serizawa, K., Hayashi, R., Discovery of potent α1L-adrenoceptor agonists: Design and
synthesis of bicyclic derivatives, Bioorganic & Medicinal Chemistry Letters (2015), doi: http://dx.doi.org/10.1016/
j.bmcl.2015.05.049
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. 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.

Discovery of potent α1L-adrenoceptor agonists: Design and synthesis of bicyclic derivatives
Shinya Suzuki*, Tsubasa Okano, Rie Horiuchi, Nana Hareyama, Kazutoshi Amikura, Naoyoshi
Yamamoto, Yoshitaka Yoshizawa, Mai Yagi, Kanako Serizawa, and Ryoji Hayashi
Pharmaceutical Research Laboratories, Toray Industries, Inc., 6-10-1 Tebiro,Kamakura, Kanagawa
248-8555, Japan
E-mail addresses:
Shinya Suzuki: Shinya_Suzuki@nts.toray.co.jp
Tsubasa Okano: Tsubasa_Okano@nts.toray.co.jp
Rie Horiuchi: Rie_Horiuchi@nts.toray.co.jp
Nana Hareyama: Nana_Hareyama@nts.toray.co.jp
Kazutoshi Amikura: Kazutoshi_Amikura@nts.toray.co.jp
Naoyoshi Yamamoto: Naoyoshi_Yamamoto@nts.toray.co.jp
Yoshitaka Yoshizawa: Yoshitaka_Yoshizawa@nts.toray.co.jp
Mai Yagi: Mai_Yagi@nts.toray.co.jp
Kanako Serizawa: Kanako_Serizawa@nts.toray.co.jp
Ryoji Hayashi: Ryoji_Hayashi2@nts.toray.co.jp
*Corresponding author.
Pharmaceutical Research Laboratories, Toray Industries, Inc., 6-10-1 Tebiro, Kamakura, Kanagawa
248-8555, Japan
Tel.: +81 467 32 9643
Fax: +81 467 32 9931
E-mail: Shinya_Suzuki@nts.toray.co.jp

Abstract
We aimed to create a novel and potent α1L-adrenoceptor agonist because such agonists are possible
drug candidates for stress urinary incontinence. We used ligand-based drug design and evaluated the
α1L-adrenoceptor agonist activity of the designed compounds. Among them, tetrahydroquinoline
derivative 50 showed the most potent activity (ratio of noradrenaline half maximal effective
concentration, 0.0028) and effectively induced contraction of rat bladder neck.
Keywords: Drug discovery, Adrenoceptor agonist, Stress urinary incontinence.
Abbreviations: CHO, Chinese hamster ovary; EC50, half maximal effective concentration; DCE,
1,2-dichloroethane; DMF, dimethylformamide; DMSO, dimethyl sulfoxide; NA, noradrenaline; NBS,
N-bromosuccinimide; PMB, 4-methoxybenzyl ether; SUI, stress urinary incontinence; TosMIC,
toluenesulfonylmethyl isocyanide; TFA, trifluoroacetic acid; THF, tetrahydrofuran; TMS, trimethylsilyl.

α1-Adrenoceptors are members of the 7TM superfamily of G protein-coupled receptors and are
classified as α1H-adrenoceptors and α1L-adrenoceptors. α1H-Adrenoceptors have high affinity for
prazosin, which is a specific α1-adrenoceptor antagonist1, and they are classified into three subtypes
(α1A, α1B, and α1D). In contrast, α1L-adrenoceptors, which result from expression of the
α1a-adrenoceptor gene2, have low affinity for prazosin. Putative α1L-adrenoceptors had been
discussed for several years and recent research results have confirmed the existence of
α1L-adrenoceptor subtypes. Muramatsu and co-workers have reported an
α1L-adrenoceptor-expressing cell line3. The α1L-adrenoceptor mediate contraction of the rabbit
prostate 4, the mouse prostate5, 6, the rat urethra7, the human bladder7 and prostate8, and other
mammalian urogenital structures. Accordingly, we hypothesized that strong α1L-adrenoceptor
agonists may be useful drugs for stress urinary incontinence (SUI). SUI is a common health problem
caused by injury to the tissue of the bladder neck or pelvic floor muscle by childbirth or obesity. SUI
significantly reduces quality of life.
Although the development of α1L-adrenoceptor agonists would be useful for SUI drugs, little
research on these agonists has been conducted. MK017 is one of the few known α1L-adrenoceptor
agonists (Fig. 1) 9. Against this background, we commenced research aimed at finding stronger
α1L-adrenoceptor agonists than existing agonists for SUI drugs.
Figure 1. The structure of MK017. This compound is a known α1L-adrenoceptor agonist.
To begin, we took MK017 as a starting point for ligand-based drug design because there is little
ligand information available. We evaluated substituent effects for MK017 and discovered the
3-isopropyl substituent is essential for α1L-adrenoceptor agonist activity (Table 1). Agonist activity is
indicated as the ratio of noradrenaline (NA) half maximal effective concentration (EC50) in Chinese
hamster ovary (CHO) cells stably expressing human α1L-adrenoceptor3, 10, 11. We subsequently

hypothesized that conformational restricted compounds in which the 2′ carbon was connected to the
1′, 3′, or 3′′ carbon via a C1 or C2 unit, while retaining the isopropyl unit, would achieve high
α1L-adrenoceptor agonist activity. Accordingly, we designed the indane and tetraline frameworks
shown in Fig. 2.
Table 1 In vitro α1L-adrenoceptor agonist activity of MK017 derivatives.
Table 1
In vitro 1L-adorenoceptor agonist activity of MK017 derivatives
Compound
Ratio of NA EC50
Me Me
H
N
Me
MK017
0.043
N
Cl
Me Me
H
N
ND (36%)**
ND (0%) **
0.52
1
Me
N
N
Me
H
N
2
3
Cl
Me
H
N
Me
N
Cl
* NA EC50 = 1019 247 nM
** ND : Not Determined
Emax responce to NA is taken as 100%
Figure 2. Designed compounds. Various indane and tetraline derivatives were designed to serve as
α1L-adrenoceptor agonists.
The key intermediates in our synthetic plan were bromoindanone 13 and bromotetralone 14. These
were chosen because the bromide can be converted to isopropenyl or methylene nitrile by
Suzuki-Miyaura cross-coupling using isopropenyl borane reagent or isoxazole borane reagent
followed by hydroxylation, respectively. Furthermore, the carbonyl oxygen can be converted to a

geminal dimethyl group or nitrile group by treatment with dimethylzinc or toluenesulfonylmethyl
isocyanide (TosMIC), respectively. To prepare the key intermediates, we started from the
corresponding carboxylic acids (9, 10). Bromination and subsequent cyclization resulted in
formation of indanone and tetralone (Scheme 1).
Scheme 1
Compounds 4 and 5 were synthesized according to Scheme 2. The isopropenyl substituent was
installed by Suzuki-Miyaura cross-coupling. The carbonyl oxygen was converted to a nitrile with
TosMIC. Cyclization with ethylenediamine at the nitrile afforded imidazoline products 19 and 20.
Finally, reduction of the olefin could be accomplished under hydrogen (1 atm) using PtO2, leading to
the desired compounds.
Scheme 2
Compounds 6 and 7 were synthesized according to Scheme 3. Dimethylation with dimethylzinc
afforded 21 and 22. Suzuki-Miyaura cross-coupling with isoxazole borane reagent, and subsequent
hydroxylation gave the methylene nitrile derivatives 23 and 2412. Because isoxazole intermediate
remained in this step, yield was low. Finally, cyclization with ethylenediamine afforded the desired
compounds. Additionally, monomethyl- and spirocyclopropyl-substituted derivatives (8 and 30)
were prepared from 14. The monomethyl substituent was installed by Wittig reaction and reduction.

The cyclopropyl substituent was installed by Wittig reaction and Simmons-Smith reaction (Scheme
4).
Scheme 3
Scheme 4
The results of α1L-adrenoceptor agonist activity are shown in Table 2. Whereas indane derivatives
tended to have lower agonist activity, tetraline derivatives tended to have higher agonist activity (4
vs. 5, 6 vs. 7). Furthermore, a substituent at the 3-position had a marked effect on agonist activity (7,
30 vs. 8). At this position, we presume that hydrophobic interaction and steric bulk played important
roles in α1L-adrenoceptor agonist activity.
Table 2 In vitro α1L-adrenoceptor agonist activity of indane and tetraline derivatives.

Compound
Ratio of NA EC50
0.043
Me Me
H
N
Me
MK017
N
Cl
Cl
H
Me
Me
N
Me
N
4
5
0.08
0.024
0.12
H
N
Me
Me
N
N
Cl
Cl
H
N
Me
6
H
N
Me
Me
7
0.028
0.063
N
N
Cl
Cl
H
N
Me
8
H
N
30
0.025
N
Cl
* NA EC50 = 1019 247 nM
Aiming to further improve the α1L-adrenoceptor agonist activity, we introduced a heteroatom at the
2-position of the benzene ring in tetralines 5 and 30, while retaining the same structure
conformation. We expected that the effect of hydrogen bonding and strong π-π stacking interaction
with the α1L-adrenoceptor protein could be altered by changing the electron density of the benzene
ring.
Chromane derivatives were synthesized by the same method as in the benzene derivative syntheses.
The isopropenyl substituent was installed by Suzuki-Miyaura cross-coupling from bromide 31.
Olefin reduction afforded the isopropyl substituent. The carbonyl oxygen was converted to a nitrile
by treatment with trimethylsilyl cyanide (TMS-CN) and olefin reduction. Finally, cyclization with
ethylenediamine at the nitrile afforded imidazoline 36 (Scheme 5).

Scheme 5
Derivatives with a cyclopropyl group were synthesized by the method used for 30. An exo olefin
was installed by Wittig reaction and converted to a cyclopropyl group by Simmons-Smith reaction
using trifluoroacetic acid (TFA), diethylzinc, and diiodomethane. Then, Suzuki-Miyaura
cross-coupling with isoxazole borane reagent at the bromide followed by hydroxylation gave the
methylene nitrile derivative 39. Finally, cyclization with ethylenediamine at nitrile afforded target
product 40 (Scheme 6).
Scheme 6
To prepare tetrahydroquinoline derivatives, first we constructed the tetrahydroquinolinone
framework by a general method from corresponding aniline 4113. After that, the isopropenyl
substituent was installed by Suzuki-Miyaura cross-coupling from the bromide. Olefin reduction
afforded the isopropyl substituent. The carbonyl oxygen was converted to a nitrile with TosMIC
after protecting the aniline nitrogen with a 4-methoxybenzyl ether (PMB) group. Cyclization with
ethylenediamine at the nitrile afforded imidazoline 49 and deprotection with acid afforded target
product 50 (Scheme 7).

Scheme 7
We attempted to synthesize tetrahydroquinoline derivative 58 in the same way (Scheme 8).
Unfortunately, we could not obtain the target product. In the final step, deprotection with TFA
afforded a pentacyclic heterocompound (59). Its structure was determined by X-ray analysis (Fig.
3)14. By mass spectroscopy, we detected 58 but not 59 under acidic conditions, so we speculate that
59 was formed by the synthetic mechanism shown Scheme 9. After the reaction mixture was
neutralized, the tetrahydroquinoline nitrogen attacked the C2 position of the imidazoline. An
unstable intermediate was oxidized in air and then intramolecular attack by the nitrogen anion gave
the pentacyclic framework. Finally, compound was oxidized again, giving 59.
Scheme 8

Scheme 9
Figure 3. ORTEP drawing of compound 59. This product was unexpectedly formed in the attempted
synthesis of 58. Thermal ellipsoids show 50% probabilities (white, carbon; green, chlorine; purple,
nitrogen; blue, hydrogen).
Chromane compound 40 which was cyclized at the 2 and 3 positions in the benzene ring had lower
α1L-adrenoceptor agonist activity compared with 30. In contrast, both heterocyclic compounds
cyclized at the 1 and 2 positions of the benzene ring (36, 50) had higher α1L-adrenoceptor agonist
activity compared with 5. Tetrahydroquinoline 50 showed the most potent α1L-adrenoceptor agonist
activity with a ratio of NA EC50 of 0.0028 (Table 3). We presumed that the nitrogen and oxygen
atoms effectively adjust the electron density of the benzene ring, which effectively interacts with the
α1L-adrenoceptor. The effect of the aniline NH is not yet fully understood, but it might serve as a
proton donor.
Table 3 In vitro α1L-adrenoceptor agonist activity of chromane and tetrahydroquinoline

derivatives.
Because 50 showed potent α1L-adrenoceptor agonist activity in CHO cells stably expressing human
α1L-adrenoceptor, we next evaluated the pharmacological effect of 50 in isolated rat bladder
neck15.Contraction of the bladder neck was induced dose-dependently by 50, which showed stronger
potency than NA (Fig. 4).
100
75
50
NA
25
0
Compound 50
-9
-8
-7
-6
-5
-4
Concentration (Log[mol/L])
Figure 4. Log concentration-response curve in isolated rat bladder neck. Compound 50 induced
contraction of isolated rat bladder neck dose-dependently and showed stronger potency than NA.
We successfully designed and discovered novel potent α1L-adrenoceptor agonists, including

tetraline, chromane, and tetrahydroquinoline derivatives. Among the compounds synthesized in this
study, 50 exhibited the highest agonist activity (ratio of NA EC50, 0.0028) and induced contraction
of isolated rat bladder neck. Further optimization of the tetrahydroquinoline derivatives is in
progress and will be reported in the future.
The authors thank Dr. Ikunobu Muramatsu (Division of Pharmacology, Department of
Biochemistry and Bioinformative Sciences, School of Medicine) for supporting our research.
1. Bremner, J. B.; Coban, B.; Groenewoud, K. M.; Yates, B. F. Bioorgan. Med. Chem. 2000, 8,
201.
2. Nelson, C. P. Brit. J. Pharmacol. 2008, 155, 1.
3. Nishimune, A.; Suzuki, F.; Yoshiki, H.; Morishita, S.; Muramatsu, I. Journal of
Pharmacological Sciences. 2010, 113, 169.
4. Su, T-H.; Morishima, S.; Suzuki, F.; Yoshiki, H.; Anisuzzaman, A.; Tanaka, T.; Cheng, J-T.;
Muramatsu, I. Brit. J. Pharmacol. 2008, 155, 906.
5. Gray, K. T.; Ventura, S. Eur. J. Pharmacol. 2006, 540, 155.
6. Gray, K. T.; Short, J. L.; Ventura, S. J. Pharmacol. 2008, 155, 103.
7. Yoshiki, H.; Uwada, J.; Umeda, H.; Kobayashi, T.; Takahashi, T.; Yamakawa, T.; Yamaguchi,
A.; Yokoyama, O.; Muramatsu, I. Brit. J. Pharmacol. 2013, 170, 1242.
8. Morishima, S.; Suzuki, F.; Nishimune, A.; Yoshiki, H.; Akino, H.; Yokoyama, O.; Muramatsu,
I. J. Urology 2010, 183, 812.
9. Nippon Chemiphar Co., Ltd. WO2009/128479 A1.
10. CHO cells stably expressing human α1L-adrenoceptor are defined as the cells (α1L-dominant)
expressing human α1A-AR and human CRELD1α.
11. The α1L-adrenoceptor agonist activity was evaluated by measuring the ability to induce a
fluorescence-based calcium mobilization signal in a fluorometric imaging plate reader assay

using the cells (α1L-dominant) expressing human α1A-AR and human CRELD1α. The cells were
plated onto 96-well plates and grown overnight at 37°C in a CO2 incubator. The cells were
incubated with calcium indicator dye (FLIPR® Calcium 5 Assay kit, Molecular Devices) for 60
min at 37°C. Calcium influx was measured using the fluorometric imaging plate reader
(FLIPRTETA®, Molecular Devices).
12. Velcicky, J.; Soicke, A.; Steiner, R.; Schmalz, H-G. J. Am. Chem. Soc. 2011, 133, 6948.
13. Anderson, K. W.; Tepe, J. J. Org. Lett. 2002, 4, 459.
14. X ray experimental details : space group ; Pccn(#56), Z value ; 8, Dcalc ; 1.390 g/cm3, Residuals [
R1 (I>2.00s(I)) ; 0.0616, R (All reflections) ; 0.0663, wR2 (All reflections) ; 0.2043]. CCDC
No. ; 1059398.
15. Male Wistar rats were anesthetized with sodium pentobarbital (50 mg/kg) and killed. The
bladder neck were isolated and cleaned in a modified Krebs-Henseleit solution aerated with 95%
O2 and 5% CO2. Strips of bladder neck were placed in organ baths containing a modified
Krebs-Henseleit solution (37°C). NA and other test agonists were applied cumulatively and the
isometric tension changes were recorded through a transducer. This study was performed
according to the Guidelines for Animal Experiments, University of Fukui.
List of captions
Figure 1. The structure of MK017. This compound is a known α1L-adrenoceptor agonist.
Figure 2. Designed compounds. Various indane and tetraline derivatives were designed to serve as
α1L-adrenoceptor agonists.
Figure 3. ORTEP drawing of compound 59. This product was unexpectedly formed in the attempted
synthesis of 58. Thermal ellipsoids show 50% probabilities (white, carbon; green, chlorine; purple,
nitrogen; blue, hydrogen).
Figure 4. Log concentration-response curve in isolated rat bladder neck. Compound 50 induced
contraction of isolated rat bladder neck dose-dependently and showed stronger potency than NA.

Table 1. In vitro α1L-adrenoceptor agonist activity of MK017 derivatives.
Table 2. In vitro α1L-adrenoceptor agonist activity of indane and tetraline derivatives.
Table 3. In vitro α1L-adrenoceptor agonist activity of chromane and tetrahydroquinoline derivatives.
Scheme 1. Reagents and conditions: (a) NBS, AuCl3, DCE, 80°C, 96–99%; (b) 1) SOCl2, CH2Cl2,
90°C, 2) AlCl3, CH2Cl2, reflux, 6–17% over 2 steps.
Scheme 2. Reagents and conditions: (a) Pd(OAc)2, Cs2CO3, PPh3, THF, H2O, 70°C, 79–89%; (b)
TosMIC, KOtBu, THF, -78°C, then MeOH, 60°C, 31-56%; (c) P2S5, ethylenediamine, 100°C,
28–76%; (d) 1) PtO2, H2, MeOH, rt, 2) HCl-MeOH, MeOH, rt, 50-65%.
Scheme 3. Reagents and conditions: (a) Dimethylzinc, TiCl4, CH3Cl2, -78–0°C, 83–88%; (b)
PdCl2(dppf), KF, DMSO, H2O, 3–11%; (c) 1) P2S5, ethylenediamine, 100°C, 2) HCl-MeOH, MeOH,
rt, 87%-quant.
Scheme 4. Reagents and conditions: (a) MePPh3Br, tBuOK, THF, rt, 97%; (b) PtO2, H2, MeOH, rt,
88%; (c) PdCl2(dppf), KF, DMSO, H2O, 130°C, 3%; (d) 1) P2S5, ethylenediamine, 100°C, 2)
HCl-MeOH, MeOH, rt, quant.; (e) CH2ICl, Et2Zn, CH2Cl2, rt, 80%; (f) PdCl2(dppf), KF, DMSO,
H2O, 130°C, 26%; (g) 1) P2S5, ethylenediamine, 100°C, 2) HCl-MeOH, rt, 96%.
Scheme 5. Reagents and conditions: (a) Pd(OAc)2, PPh3, Cs2CO3, THF, H2O, 70°C, 91%; (b)
Wilkinson’s reagent, H2, CH2Cl2, rt, 96%; (c) 1) TMS-CN, ZnI2, CH2Cl2, reflux, 2) Amberlyst 15,
toluene, reflux, 39%; (d) NaBH4, EtOH, rt, 97%; (e) 1) P2S5, ethylenediamine, 100°C, 2)
HCl-MeOH, MeOH, rt, 94%
Scheme 6. Reagents and conditions: (a) MePPh3Br, KOtBu, THF, rt, 15%; (b) TFA, Et2Zn, CH2I2,
DCE, 64%; (c) PdCl2(dppf), KF, DMSO, H2O, 16%; (d) 1) P2S5, ethylenediamine, 100°C, 2)
HCl-MeOH, MeOH, rt, 49%
Scheme 7. Reagents and conditions: (a) 3-Bromopropanoyl chloride, K2CO3, CH2Cl2, rt; (b)
NaOtBu, DMF, rt, 2 steps 72%; (c) TfOH, DCE, rt, 34%; (d) Pd(OAc)2, PPh3, Cs2CO3, 70°C, 85%;
(e) Wilkinson’s reagent, H2, CH2Cl2, rt, 99%; (f) PMB-Cl, K2CO3, NaI, DMF, 50°C, 47%; (g)
TosMIC, KOtBu, THF, -78°C, then MeOH, 60°C, 60%; (h) P2S5, ethylenediamine, 100°C, 91%; (i)

1) TFA, CH2Cl2, rt, 2) HCl-MeOH, MeOH, rt, quant.
Scheme 8. Reagents and conditions: (a) PMB-Cl, NaI, K2COe, DMF, 50°C, 56%; (b) MePPh3Br,
KOtBu, THF, rt, 80%; (c) Et2Zn, CH2I2, TFA, DCE, 0°C, 71%; (d) nBuLi, THF, -78°C, then DMF,
0°C; (e) NaBH4, MeOH, THF, rt, 2 steps 76%; (f) 1) SOCl2, CH2Cl2, rt, 16h, 2) NaCN, DMSO, rt,
49%; (g) P2S5, ethylenediamine, 100°C, 76%; (h) TFA, CH2CL2, rt.
Scheme 9. Putative synthetic mechanism of by-product 59.