Aminocyclohexylsulfonamides: Discovery of metabolically stable α1a/1d-selective adrenergic receptor antagonists for the treatment of benign prostatic hyperplasia/lower urinary tract symptoms (BPH/LUTS)

Article abstract of DOI:10.1016/j.bmcl.2007.09.051

Benign prostatic hyperplasia/lower urinary tract symptoms (BPH/LUTS) can be effectively treated by α₁ adrenergic receptor antagonists, but these drugs also produce side effects that are related to their subtype non-selective nature. To overcome

Full text of DOI:10.1016/j.bmcl.2007.09.051

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 17 (2007) 6123–6128
Aminocyclohexylsulfonamides: Discovery of metabolically stable
a
1a/1d-selective adrenergic receptor antagonists for the treatment
of benign prostatic hyperplasia/lower urinary tract
symptoms (BPH/LUTS)
George Chiu,*, Shengjian Li,à Hong Cai,à Peter J. Connolly,à Sean Peng,à Kathe Stauber,à
Virginia Pulito,§ Jingchun Liu and Steven A. Middleton–
Johnson & Johnson Pharmaceutical Research and Development L.L.C., PO Box 300, 1000 Route 202 South, Raritan, NJ 08869, USA
Received 25 June 2007; revised 9 September 2007; accepted 11 September 2007
Available online 18 September 2007
Abstract—Benign prostatic hyperplasia/lower urinary tract symptoms (BPH/LUTS) can be effectively treated by a1 adrenergic
receptor antagonists, but these drugs also produce side effects that are related to their subtype non-selective nature. To overcome
this limitation, it was hypothesized that an a1a/1d subtype-selective antagonist would be efficacious while keeping side effects to a
minimum. To discover a1a/1d-selective antagonists and improve metabolic stability of our previously reported compounds, we have
designed and synthesized a series of (phenylpiperazinyl)- or (phenylpiperidinyl)-cyclohexylsulfonamides. By incorporating the infor-
mation obtained from metabolism studies, we were able to discover several compounds that are both a1a/1d adrenoceptor subtype
selective and show increased stability toward human liver microsomal metabolism. The selectivity profile of these compounds pro-
vides great improvement over the commercial drug tamsulosin, hence may pave the way to the development of new and efficacious
therapeutic agents with reduced side effects.
Ó 2007 Elsevier Ltd. All rights reserved.
The prostate is a male sex auxiliary gland located just
below the bladder and surrounding the urethra. Exces-
sive growth of the prostate with age results in benign
prostatic hyperplasia (BPH), which causes obstruction
of the bladder outlet and leads to lower urinary tract
symptoms (LUTS). These symptoms include increased
urinary frequency, decreased urine stream, increased ur-
gency and feeling of irritation, and sensation of incom-
plete bladder emptying.1,2a,b Since there are two
components in BPH, namely the increased size and ele-
vated muscle tone of the prostate gland, drug therapy
for BPH/LUTS has been classified into two categories.
Drugs in the first category, 5-a-reductase inhibitors
(e.g., finasteride and dutasteride), work by reducing
the size of the prostate; drugs in the second category,
a1-adrenergic receptor antagonists (e.g., tamsulosin
and terazosin), work by relaxing prostate muscle. The
a1 blockers have an advantage over 5-a-reductase inhib-
itors because they can provide effective relief of symp-
toms in very short period of time. Unfortunately, all
a1 drugs currently on the market also produce cardio-
vascular related side effects such as orthostatic
hypotension.3,4
Keywords: BPH/LUTS; a1a/1d adrenergic receptors; a1 blockers; a1a/1d
adrenoceptor selective antagonists.
*
Corresponding author. Tel.: +1 908 231 4697; fax: +1 908 231
4766; e-mail: George.Chiu@sanofi-aventis.com
Three a1-adrenergic receptor subtypes have been discov-
ered to date. They are termed a1a, a1b, and a1d,5–7 and
current a1 drugs are known to bind to all of them indis-
criminately or with low selectivity.8 It is speculated that
orthostatic hypotension caused by present a1 blockers is
a result of their subtype non-selective nature. In the
early 1990s, the a1a-adrenoceptor subtype was deter-
mined to play a dominant role in controlling prostatic
muscle contraction.8 Although many a1a-adrenoceptor
Present address: Sanofi-Aventis, 1041 Route 202-206, PO Box 6800,
Bridgewater, NJ 08807, USA.
à
§
Present address: Johnson & Johnson Pharmaceutical R&D, 8 Clark
Drive, Cranbury, NJ 08512, USA.
Present address: Wyeth Research, 865 Ridge Road, Monmouth
Junction, NJ 08852, USA.
–
Present address: Hoffmann-La Roche, Inc., 340 Kingsland Street,
Bldg. 76/5E11, Nutley, NJ 07110, USA.
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.09.051

6124
G. Chiu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6123–6128
subtype-selective antagonists have been discovered,9a–c
the failure of these compounds in clinical trials indicates
that other a1-receptor subtypes might be implicated in
the BHP/LUTS.10 Recent studies have provided evi-
dence that the a1d subtype is involved in the mediation
of LUTS.11a–c In addition, experimental data suggest
that the a1b subtype may be associated with CV related
side effects.12 These results, combined with the fact that
a moderately a1a/1d-selective drug, tamsulosin (1, Fig. 1),
is capable of treating both BPH and LUTS, led to the
new hypothesis that an antagonist with balanced a1a/1d
selectivity profile will be efficacious yet produce less se-
vere side effects.13a–e So far, validation of this hypothesis
has been hampered by the fact that no a1 compound
with high selectivity for a1a/1d versus a1b is currently
available.
product C, were identified. Judging by their structure
and relative abundance, the metabolic pathway of 8
could be constructed (Scheme 1). It was discovered that
the principal degradation steps for compound 8 were de-
alkylation of the alkoxy group and hydroxylation on the
aromatic rings, although cleavage of the sulfonamide
group also played a relatively minor role.
The study of metabolic fate of 8 provided valuable infor-
mation for the further development of metabolically sta-
ble analogues. The metabolic pathways of 8 clearly
suggested three possible modifications of the core struc-
ture to increase its resistance to metabolism. These mod-
ifications are: (1) changing the alkoxy group to block
Path A; (2) reducing electron density of the alkoxylated
aromatic ring attached to the piperazine to retard Path
B; and (3) increasing steric bulk of the sulfonamide
group to slow down Path C (Fig. 2). More specifically,
we believed cyclizing the isopropoxy group or introduc-
ing fluorine atoms on the alkoxy group might be effec-
tive ways to block de-alkylation. To retard
hydroxylation of the alkoxy-phenyl ring, we decided to
replace the piperazine ring with a piperidine ring. We
also tried to incorporate alkyl groups on the sulfon-
amide nitrogen atom to increase steric hindrance and re-
duce metabolic cleavage. A set of compounds were
designed and synthesized according to these three
approaches.
We initiated a research program to validate this hypoth-
esis by first identifying a a1a/1d-selective compound, then
studying it in established animal models. Our primary
goal was to design and synthesize potent and a1a/1d sub-
type-selective antagonists with superior selectivity pro-
files compared to the marketed drug tamsulosin. In
our previous papers,14 we described the discovery of ser-
ies of cis-(phenylpiperazinyl)-cyclohexylsulfonamides (2,
Fig. 1) and cis-(phenylpiperidinyl)-cyclohexylsulfona-
mides (3, Fig. 1). These compounds proved to have
equal affinity for both a1a and a1d subtypes, with good
selectivity against the a1b subtype. Encouraged by these
results, our effort was shifted to evaluate the pharmaco-
kinetic properties of these compounds. We first investi-
gated the metabolic stability of piperazine analogues
(4). Six analogues were chosen for evaluation in a hu-
man liver microsomal (HLM) metabolism assay. Half-
lives were determined and are listed in Table 1. It was
disappointing to discover that these compounds have
short in vitro metabolic half-lives, typically less than
5 min, that strongly indicates that these compounds
would not survive first pass metabolism.
The (phenylpiperazinyl)- or (phenylpiperidinyl)-cyclo-
hexylsulfonamides were generally prepared by the fol-
lowing sequence (Scheme 2).14 A suitably substituted
phenylpiperazine or phenylpiperidine was subjected to
t
the reductive alkylation with Boc protected 4-aminocy-
clohexanone to give a cis/trans mixture of diaminocyclo-
Table 1. Half-lives of 5–10 in HLM experiments (T1/2, min)
O
To avoid the low bioavailability predicted by these re-
sults, we needed to establish the metabolic pathway of
these compounds to identify potential weak points on
the molecule and modify the structure accordingly. This
would serve as a good foundation for the further devel-
opment of metabolically stable analogues. To accom-
plish this, compound 8 was chosen as a substrate in an
in vitro metabolism study. After incubation with HLM
for up to 60 min at 37 °C in the presence of NADPH
and HPLC/MS analysis, three major metabolites, which
included de-alkylation/hydroxylation product A, double
hydroxylation product B, and sulfonamide cleavage
O
X
S
O
N
N
N
H
4
Compound
Configuration
X
T1/2
5
6
cis
cis
cis
cis
cis
cis
4-SO2Me
2.5
4.9
3,4-diOMe
2-MeO-5-Me
2-MeO-5-F
2-MeO-5-Cl
2,4-diCl
7
<2.0
<2.0
<2.0
3.5
8
9
10
SO2NH2
O
O
S
N
R2
X
N
O
O
N
H
R3
R1
O
2
3
X = N
X = C
tamsulosin (1)
Figure 1. Structure of tamsulosin, 2 and 3.

G. Chiu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6123–6128
6125
MeO
O
MeO
OH
F
OH
F
O
S
S
O
O
N
N
N
H
N
N
N
H
OH
A
Path A
MeO
O
O
F
S
O
N
N
N
H
Path C
8
Path B
MeO
O
O
O
F
OH
S
O
N
N
N
N
OH
N
H
OH
B
C
Scheme 1. Major metabolic pathways of compound 8.
1. Modify alkoxy group
Y
R2
O
O
S
O
X
N
N
R1
3. Alkylate sulfonamide bond
2. Make phenyl ring less electron rich
Figure 2. Three modifications of structure to enhance metabolic stability.
R2
R2
O
O
O
NHBoc
b
N
X
NHBoc
X
NH
a
R1
R2
R1
R2
O
O
O
S
N
H
R3
c
O
N
X
NH2
N
X
R1
R1
R2
R2
d
O
O
N
O
O
N
R3
+
R3
S
H
S
H
O
O
N
X
X
N
R1
R1
Scheme 2. Reagents and conditions: (a) Na(AcO)3BH, HOAc, CH2Cl2, rt, 8 h, 40–65% yield; (b) CF3CO2H/CH2Cl2, rt, 2 h, 90–100% yield; (c)
arylsulfonyl chloride/CH2Cl2/Na2CO3 (aq), rt, 8 h, 60–90% yield; (d) SiO2 column or prep TLC.

6126
G. Chiu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6123–6128
R2
R2
O
O
O
a
b
X
N
NH2
N
X
N
H
R1
R1
R2
R2
O
O
O
N
c
R3
S
N
X
N
H
O
N
X
R1
R1
R2
d
R2
O
O
N
O
O
R3
+
S
R3
S
O
O
N
X
N
X
N
R1
R1
Scheme 3. Reagents and conditions: (a) CH3COCl, Et3N, CH2Cl2, rt, 8 h, 85% yield; (b) LiAlH4/THF, reflux, 2 h, 90% yield; (c) arylsulfonyl
chloride/CH2Cl2/Na2CO3 (aq), rt, 8 h, 63% yield; (d) SiO2 column or prep TLC.
hexane intermediates. Treatment with TFA produced
the free amine, which was acylated by various sulfonyl
chlorides. Final chromatographic separation gave the
desired isomers.15
A series of (phenylpiperazinyl)-cyclohexylsulfonamides
and (phenylpiperidinyl)-cyclohexylsulfonamides (11)
were synthesized and tested for binding to cloned hu-
man a1a, a1b, and a1d adrenergic receptors and the dopa-
mine D2 receptor. Results for several typical examples
are summarized in Table 2. To assess metabolic stability,
these compounds were tested in HLMs; their half-lives
are also listed in Table 2. Data for the unmodified refer-
ence compound, (phenylpiperazinyl)-cyclohexylsulfona-
mide cis-6, are included for comparison. Based on our
previous experience, 3,4-dimethoxy was chosen as the
optimum substitution on the phenyl ring attached to
For N-alkylated analogues, the following synthetic pro-
cedure was used (Scheme 3). The cyclohexylamine was
first acylated with acetyl chloride followed by LAH
reduction to give the N-ethylamine, which was allowed
to react with a sulfonyl chloride to yield the cis/trans
mixture of sulfonamides. Chromatographic separation
gave the desired isomers.16
Table 2. Binding profile (Ki, nM) and metabolic half-lives (min) of 6–18
O
R1
O
O
O
S
O
N
X
N
R2
11
Compound
Configuration
X
R2
R1
Ki (nM)
a1d
0.68
HLM
T1/2
4.9
a1a
2.0
a1b
106
D2
6
12
13
14
15
16
17
18
cis
N
N
N
N
C
C
C
C
C
C
C
C
N
N
C
C
H
H
Et
Et
H
H
H
H
H
H
H
H
H
H
H
H
(CH3)2CH
(CH3)2CH
(CH3)2CH
(CH3)2CH
(CH3)2CH
(CH3)2CH
Cyclopropyl
Cyclopropyl
CH2FCH2
CH2FCH2
CHF2CH2
CHF2CH2
CF3CH2
136
62
trans
cis
23
15
23
126
326
109
151
148
141
133
354
4403
278
681
86
24
22
113
trans
cis
6.7
1.7
40
2.0
34
8.8
49
8.6
86
2.7
1.8
28
0.91
165
55
12
18
20
21
35
56
trans
cis
144
133
330
284
228
88
trans
cis
11
11
69
12
31
trans
cis
trans
cis
1.2
54
1.8
21
0.35
8.3
192
114
112
83
trans
cis
CF3CH2
CF3CH2
CF3CH2
39
123
368
2.2
trans
23

G. Chiu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6123–6128
6127
the sulfonamide. We found that, as in our previously re-
ported sulfonamide compounds,14 the cis isomers of this
series also have better selectivity profiles than their trans
counterparts. In addition, they also have lower affinities
for the D2 receptor. Compound cis-12, an N-ethylated
analogue that was originally designed to slow down
the cleavage of sulfonamide bond (Path C), shows much
reduced affinity for a1a and a1d. Since this may indicate
the importance of an N–H hydrogen bond interaction
with the receptor, the N-alkylation approach was not
pursued further. As for the activity of N–H compounds,
except for the slight loss of a1a and a1d affinities for fluo-
rinated analogues cis-15 and 16, compounds cis-13, 14,
17, and 18 all show equal affinity for both a1a and a1d
subtypes and excellent selectivity profiles. Their a1a/a1b
and a1d/a1b ratios range from about 60-fold to more
than 200-fold. These selectivity ratios are much im-
proved over commercial drug tamsulosin (1), which
are about 10-fold (Ki values for tamsulosin, 1, in the
a1a, a1b, and a1d binding assays were 0.19, 2.0, and
0.2 nM, respectively).
HLM metabolism. The selectivity profile exhibited by
these compounds provides great improvement over the
commercial drug tamsulosin, and hence paves the way
toward development of new, efficacious therapeutic
agents with fewer side effects.
Acknowledgments
We express our gratitude to Ms. Sally Varga and Ms
Aida Howell for their technical assistance in this
research.
References and notes
1. Rosini, M.; Bolognesi, M.; Giardina, D.; Minarini, A.;
Tumiatti, Vincenzo; Melchiorre, C. Curr. Top. Med.
Chem. 2007, 7, 147.
2. (a) Michelotti, G. A.; Schwinn, D. A. Curr. Urol. Rep.
2004, 5, 258; (b) Cutis, N. J. Urology 2003, 62, 34.
3. Lowe, F. C. Clin. Ther. 2004, 26, 1701.
4. Muramatsu, I.; Suzuki, F.; Tanaka, T.; Yamamoto, H.;
Morishima, S. J. Pharm. Soc. Jpn. 2006, 126, 187.
5. Bruno, J. F.; Whittaker, J.; Song, J.; Berelowitz, M.
Biochem. Biophys. Res. Commun. 1991, 179, 1485.
6. Ramarao, C. S.; Kincade Denker, J. M.; Perez, D. M.;
Galvin, R. J.; Riek, R. P.; Graham, R. M. J. Biol. Chem.
1992, 267, 21936.
7. Hirasawa, A.; Horie, K.; Tanaka, T.; Takagaki, K.;
Murai, M.; Yano, J.; Tsujimoto, G. Biochem. Biophys.
Res. Commun. 1993, 195, 902.
Perhaps the most striking differences among these ana-
logues are their half-lives against HLM metabolism.
Our reference compound cis-6, a 2-i-propoxyphenylpip-
erazine analogue, has
a short (<5 min) half-life.
Replacement of piperazine with piperidine, which re-
duces electron density on the phenyl ring (Fig. 2, mod-
ification 2), results in the still short yet noticeably
improved half-life of 12 min (cis-13 vs cis-6). Cycliza-
tion of i-propyl group into a cyclopropyl group (cis-
14) extends the half-life of the analogue even further
(Fig. 2, modification 1). When fluorine atoms were
introduced to the alkoxy side chain to block de-alkyl-
ation (Path A), we were encouraged by the results.
Monofluoroethoxy or difluoroethoxy substitution was
able to effectively increase half-life several fold (cis-15
and 16 vs cis-6). When the trifluoroethoxy group was
introduced, the half-life of the piperazine analogue
(cis-17) reached a satisfactory level (35 min for cis-17
vs 4.9 min for cis-6). Finally when the 2-trif-
luoroethoxyphenylpiperidine analogue cis-18 was pre-
pared and tested, the half-life neared the 1-h mark.
By systematically altering the structure to block meta-
bolic pathways, we were able to produce compounds
such as cis-17 and 18 that not only had excellent a1a
and a1d affinity and selectivity versus a1b, but also
showed favorable metabolic stability.
8. Forray, C.; Bard, J. A.; Wetzel, J.; Chiu, G.; Shapiro, E.;
Tang, R.; LePor, H.; Hartig, P. R.; Weinshank, R. L.;
Branchek, T. A.; Gluchowski, C. Mol. Pharmacol. 1994,
45, 703.
9. (a) Wetzel, J. M.; Miao, S. W.; Forray, C.; Borden, L. A.;
Branchek, T. A.; Gluchowski, C. J. Med. Chem. 1995, 38,
1579; (b) Wong, W. C.; Chiu, G.; Wetzel, J. M., et al. J.
Med. Chem. 1998, 41, 2643; (c) Lagu, B.; Tian, D.; Chiu,
G.; Nagarathram, D.; Fang, J.; Shen, Q.; Forray, C.;
Ransom, R. W.; Chang, R. S. L.; Vyas, K. P.; Zhang, K.;
Gluchowski, C. Bioorg. Med. Chem. Lett. 2000, 10, 175.
10. Blue, D. R., Jr.; Grino, P. B.; Jung, D. T.; et al. [abstract].
In Proceedings of the Fifth International Consultation on
BPH, June 25, 2000.
11. (a) Ilampel, C.; Dolber, P. C.; Smith, M. P., et al. J. Urol.
2002, 167, 1513; (b) Gu, B.; Reiter, J. P.; Schwinn, D. A.,
et al. J. Urol. 2004, 172, 758; (c) Smith, M. S.; Scharnbra,
U. M.; Wilson, K. H.; Page, S. O.; Schwinn, D. A. Brain
Res. Mol. Brain Res. 1999, 63, 254.
12. Cavalli, A.; Lattion, A. L.; Hummler, E., et al. Proc. Natl.
Acad. Sci. U.S.A. 1997, 94, 11589.
In conclusion, to discover a a1a/1d-selective adrenoceptor
antagonist as a new drug for the treatment of BPH/
LUTS and improve metabolic stability of our previously
discovered compounds, we carried out an in vitro
metabolism study of compound 8. Identification of the
structure of metabolites helped us to establish the major
metabolic pathways. A series of (phenylpiperazinyl)- or
(phenylpiperidinyl)-cyclohexylsulfonamides were de-
signed and synthesized by incorporating the information
obtained from metabolism studies. Through systematic
introduction of structural features that blocked meta-
bolic pathways, we were able to discover compounds
cis-17 and cis-18 that are both selective ligands for the
13. (a) Schwinn, D. A.; Price, D. T.; Narayan, P. Mayo Clin.
Proc. 2004, 79, 1423; (b) Roehrborn, C. G.; Schwinn, D. J.
Urol. 2004, 171, 1029; (c) Sarma, P. K. S.; Tiwari, A.; Pal,
A. Expert Opin. Ther. Patents 2005, 15, 1333; (d) Daniels,
D. V.; Gever, J. R.; Jasper, J. R.; Kava, M. S., et al. Eur.
J. Pharmacol. 1999, 370, 337; (e) Lowe, F. C. Clin. Ther.
2004, 26, 1701.
14. (a) Chiu, G.; Li, S.; Connolly, P. J.; Pulito, V.; Liu, J.;
Middleton, S. A. Bioorg. Med. Chem. Lett. 2007, 17, 3292;
(b) Chiu, G.; Li, S.; Connolly, P. J.; Pulito, V.; Liu, J.;
Middleton, S. A. Bioorg. Med. Chem. Lett. 2007, 17, 3930.
15. The isomers with higher Rf values in TLC are the cis
isomers. The isomers with lower Rf values are the trans
isomers. This assignment is based on the NMR pattern of
a1a and a1d adrenoceptor subtypes and are stable against

6128
G. Chiu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6123–6128
similar (piperazinyl)-cyclohexylsulfonamide compounds
with known cis or trans configurations.
3.40 (m, 1H), 3.72 (m, 1H), 3.94 (s, 3H), 3.96 (s, 3H), 5.20
(d, J = 7.5 Hz, NH, 1H), 6.8–7.6 (m, 7H) MS: 515 (M+1).
trans-14 NMR: d (CDCl3) 0.76 (m, 4H), 1.2–1.5 (m, 4H),
1.58 (m, 2H), 1.70 (m, 2H), 1.8–2.0 (m, 4H), 2.23 (m, 3H),
2.80 (m, 1H), 2.90 (br d, 2H), 3.02 (m, 1H), 3.72 (m, 1H),
3.92 (s, 3H), 3.96 (s, 3H), 4.50 (d, J = 7.6 Hz, NH, 1H),
6.8–7.6 (m, 7H) MS: 515 (M+1). cis-17 NMR: d (CDCl3)
1.4–1.8 (m, 8H), 2.26 (m, 1H), 2.77 (br s, 4H), 3.10 (br s,
4H), 3.42 (m, 1H), 3.93 (s, 3H), 3.95 (s, 3H), 4.41 (q,
J = 14 Hz, 2H), 4.62 (d, J = 7.2 Hz, NH, 1H), 6.8–7.6 (m,
7H) MS: 558 (M+1). trans-17 NMR: d (CDCl3) 1.1–1.4
(m, 4H), 1.93 (m, 4H), 2.22 (m, 1H), 2.68 (m, 4H), 3.05 (br
s, 4H), 3.14 (m, 1H), 3.94 (s, 3H), 3.96 (s, 3H), 4.38 (q,
J = 15 Hz, 2H), 4.85 (d, J = 7.2 Hz, NH, 1H), 6.8–7.6 (m,
7H) MS: 558 (M+1). cis-18 NMR: d (CDCl3) 1.4–1.8 (m,
10H), 1.84 (br d, 2H), 2.25 (m, 3H), 2.90 (m, 3H), 3.42 (m,
1H), 3.92 (s, 3H), 3.94 (s, 3H), 4.35 (q, J = 15 Hz, 2H),
5.20 (d, J = 7.8 Hz, NH, 1H), 6.7–7.6 (m, 7H) MS: 557
(M+1). trans-18 NMR: d (CDCl3) 1.1–1.4 (m, 4H), 1.65
(m, 2H),1.8–2.0 (m, 6H), 2.30 (m, 3H), 2.93 (m, 3H), 3.04
(m, 1H), 3.92 (s, 3H), 3.94 (s, 3H), 4.33 (q, J = 14.8 Hz,
2H), 4.80 (d, J = 8.1 Hz, NH, 1H), 6.7–7.5 (m, 7H) MS:
557 (M+1).
16. NMR and MS data of representative compounds. cis-12
NMR: d (CDCl3) 1.28 (t, J = 8.2 Hz, 3H), 1.3–1.5 (m, 4H),
1.34 (d, J = 8.1 Hz, 6H), 1.81 (m, 2H), 2.06 (br d, 2H), 2.15
(br s, 1H), 2.60 (br s, 4H), 3.10 (br s, 4H), 3.28 (q,
J = 8.2 Hz, 2H), 3.74 (m, 1H), 3.92 (s, 3H), 3.94 (s, 3H),
4.58 (m, 1H), 6.8–7.6 (m, 7H) MS: 546 (M+1). trans-12
NMR: d (CDCl3) 1.24 (t, J = 8.0 Hz, 3H), 1.3–1.6 (m, 4H),
1.34 (d, J = 8.1 Hz, 6H), 1.85 (br d, 2H), 1.98 (br d, 2H),
2.27 (m, 1H), 2.73 (m, 4H), 3.13 (br s, 4H), 3.24 (q,
J = 8.0 Hz, 2H), 3.62 (m, 1H), 3.92 (s, 3H), 3.94 (s, 3H),
4.60 (m, 1H), 6.8–7.6 (m, 7H) MS: 546 (M+1). cis-13
NMR: d (CDCl3) 1.33 (d, J = 6.0 Hz, 6H), 1.4–1.9 (m,
12H), 2.22 (m, 3H), 2.8–3.1 (m, 3H), 3.44 (m, 1H), 3.92 (s,
3H), 3.94 (s, 3H), 4.52 (m, 1H), 5.08 (d, J = 7.5 Hz, NH,
1H), 6.8–7.6 (m, 7H) MS: 517 (M+1). trans-13 NMR: d
(CDCl3) 1.25 (m, 4H), 1.34 (d, J = 6.0 Hz, 6H), 1.65 (m,
2H), 1.8–2.1 (m, 6H), 2.30 (m, 3H), 2.95 (m, 3H), 3.08 (m,
1H), 3.93 (s, 3H), 3.95 (s, 3H), 4.52 (m, 1H), 4.90 (d,
J = 7.3 Hz, NH, 1H), 6.8–7.6 (m, 7H). MS: 517 (M+1).
cis-14 NMR: d (CDCl3) 0.75 (m, 4H), 1.22 (m, 2H), 1.3–
1.8 (m, 10H), 2.32 (m, 3H), 2.85 (m, 1H), 3.05 (br d, 2H),