Identification of Tricyclic Agonists of Sphingosine-1-phosphate Receptor 1 (S1P1) Employing Ligand-Based Drug Design

Article abstract of DOI:10.1021/acs.jmedchem.6b01099

Fingolimod (1) is the first approved oral therapy for the treatment of relapsing remitting multiple sclerosis. While the phosphorylated metabolite of fingolimod was found to be a nonselective S1P receptor agonist, agonism specifically of S1P₁ is responsible for the peripheral blood lymphopenia believed to be key to its efficacy. Identification of modulators that maintain activity on S1P₁ while sparing activity on other S1P receptors could offer equivalent efficacy with reduced liabilities. We disclose in this paper a ligand-based drug design approach that led to the discovery of a series of potent tricyclic agonists of S1P₁ with selectivity over S1P₃ and were efficacious in a pharmacodynamic model of suppression of circulating lymphocytes. Compound 10 had the desired pharmacokinetic (PK) and pharmacodynamic (PD) profile and demonstrated maximal efficacy when administered orally in a rat adjuvant arthritis model.

Full text of DOI:10.1021/acs.jmedchem.6b01099

Subscriber access provided by MT ROYAL COLLEGE
Article
Identification of Tricyclic Agonists of Sphingosine-1-Phosphate
Receptor 1 (S1P1) Employing Ligand-Based Drug Design
Hai-Yun Xiao, Scott H. Watterson, Charles M. Langevine, Anurag S. Srivastava, Soo S Ko, Yanlei
Zhang, Robert Joseph Cherney, Weiwei Guo, John L Gilmore, James E Sheppeck, Dauh-Rurng
Wu, Peng Li, Duraisamy Ramasamy, Pirama Nayagam Arunachalam, Arvind Mathur, Tracy L.
Taylor, David J. Shuster, Kim W. McIntyre, Ding Ren Shen, Melissa Yarde, Mary Ellen Cvijic,
Anthony M. Marino, Praveen V. Balimane, Zheng Yang, Dana M. Banas, Georgia Cornelius,
Celia J. D´Arienzo, Bethanne M. Warrack, Lois D. Lehman-McKeeman, Luisa M. Salter-
Cid, Jenny H. Xie, Joel C. Barrish, Percy H Carter, Alaric J. Dyckman, and T. G. Murali Dhar
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b01099 • Publication Date (Web): 11 Oct 2016
Downloaded from http://pubs.acs.org on October 11, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Lehman-McKeeman, Lois; Bristol-Myers Squibb Company, Investigative
Toxicology (MS H24-03)
Salter-Cid, Luisa; Bristol-Myers Squibb Pharmaceutical Research and
Development, Discovery Biology
Xie, Jenny; Bristol-Myers Squibb Pharmaceutical Research and
Development, Discovery Biology
Barrish, Joel; Achillion Pharmaceuticals Inc
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Carter, Percy; BMS, R&D
Dyckman, Alaric; Bristol-Myers Squibb, Discovery Chemistry
Dhar, T. G. Murali; Bristol-Myers Squibb, Discovery Chemistry
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Identification of Tricyclic Agonists of Sphingosine-1-
Phosphate Receptor 1 (S1P ) Employing Ligand-
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Based Drug Design
‡
‡
‡
‡
‡
Hai-Yun Xiao , Scott H. Watterson , Charles M. Langevine , Anurag S. Srivastava , Soo S. Ko , Yanlei
‡
‡
‡
‡
‡
‡
Zhang , Robert J. Cherney , Wei-Wei Guo , John L. Gilmore , James E. Sheppeck II , Dauh-Rurng Wu ,
‡
#
#
‡
‡
Peng Li , Duraisamy Ramasamy , Piramanayagam Arunachalam , Arvind Mathur , Tracy L. Taylor ,
‡
‡
‡
‡
‡
David J. Shuster , Kim W. McIntyre , Ding-Ren Shen , Melissa Yarde , Mary Ellen Cvijic , Anthony M.
‡
‡
‡
‡
‡
‡
Marino , Praveen V. Balimane , Zheng Yang , Dana M. Banas , Georgia Cornelius , Celia J. Darienzo ,
‡
‡
‡
‡
‡
Bethanne M. Warrack , Lois Lehman-McKeeman , Luisa M. Salter-Cid , Jenny Xie , Joel C. Barrish ,
‡
‡
‡
Percy H. Carter , Alaric J. Dyckman and T. G. Murali Dhar *
‡
Research and Development, BristolꢀMyers Squibb Company, P.O. Box 4000, Princeton, New
Jersey 08543ꢀ4000, United States
#
BMS Biocon Research Center, Bangalore, India
RECEIVED DATE
ABSTRACT: Fingolimod (1) is the first approved oral therapy for the treatment of relapsing remitting
multiple sclerosis. While the phosphorylated metabolite of fingolimod was found to be a nonꢀselective
S1P receptor agonist, agonism specifically of S1P is responsible for the peripheral blood lymphopenia
1
believed to be key to its efficacy. Identification of modulators that maintain activity on S1P while
1
sparing activity on other S1P receptors could offer equivalent efficacy with reduced liabilities. We
disclose in this paper a ligand based drug design approach that led to the discovery of a series of potent
ACS Paragon Plus Environment

Page 3 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
tricyclic agonists of S1P with selectivity over S1P and were efficacious in a pharmacodynamic model
1 3
of suppression of circulating lymphocytes. Compound 10 had the desired pharmacokinetic (PK) and
pharmacodynamic (PD) profile and demonstrated maximal efficacy when administered orally in a rat
adjuvant arthritis model.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
INTRODUCTION
The Gꢀprotein coupled receptor (GPCR) S1P (sphingosineꢀ1ꢀphosphateꢀsubtype 1) has emerged
1
as an attractive drug target in recent years. Activation of S1P by synthetic agonists, but not by
1
activation with the endogenous ligand S1P, leads to sustained internalization and degradation of cell
1
surface receptors. This functional antagonism of the receptor results in immunosuppression through
disruption of normal lymphocyte trafficking, with lymphocytes sequestered in lymph nodes and
2
secondary lymphoid organs. Extensive validation of this mechanism has been demonstrated across a
3
range of preꢀclinical models of chronic autoimmune disease. The success of the nonꢀselective S1P
receptor agonist fingolimod (FTY720, 2, Figure 1), approved by the FDA in 2010 for the treatment of
relapsing remitting multiple sclerosis (RRMS), has spurred interest in the clinical evaluation of it and
additional S1P agonists for a growing list of disorders (Crohn’s disease, ulcerative colitis, psoriasis,
1
4
polymyositis, progressive MS, lupus). Fingolimod is a proꢀdrug, requiring in vivo phosphorylation to
2
form FTY720ꢀP (2-P, Figure 1) which is a potent agonist of 4 of the 5 S1P receptors (S1P1,3,4,5). In
addition to the proꢀdrug class of S1P agonists, exemplified by fingolimod, directꢀacting agonists that
1
replace the aminoꢀphosphonate group with moieties that mimic its interactions and thus do not require
5ꢀ13
bioactivation have been reported by a number of research teams and are in clinical development.
The
focus of research programs following fingolimod has primarily been aimed at identifying agonists with
improved S1P family selectivity, specifically to avoid interaction with S1P , which was believed to be
3
associated with clinical side effects such as the transient reduction in heart rate (bradycardia). In rodent
studies, fingolimod was unable to elicit heart rate reductions in S1P knockꢀout mice, whereas it did
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
4
induce bradycardia in wild type controls. Furthermore, synthetic S1P ꢀselective agonists that lacked
1
activity on S1P did not induce bradycardia in rodents. Surprisingly, in human trials multiple S1P ꢀ
3
3
sparing S1P agonists have now been shown to offer no substantial advantage over fingolimod in
1
1
5
regards to bradycardia. In humans, as opposed to nonꢀclinical species, regulation of heart rate appears
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
15
to have a significant S1P related component. Nevertheless, as S1P does not contribute to the desired
1
3
efficacy, a drive for more selective S1P agonists has remained a viable approach for research programs,
1
including our own.
Figure 1. Structure of S1P, fingolimod (FTY720), and its phosphate
We recently disclosed the identification of compound 3 as a potent functional antagonist of S1P1
1
6
with selectivity over S1P (Figure 2). The single crystal Xꢀray structure of 3 indicated that the phenyl
3
ring and the 1,2,4ꢀoxadiazole ring are coplanar in orientation (Figure 2). This suggested that ring fusion
of the oxadiazole ring with the 3ꢀposition of the phenyl ring would maintain planarity of the scaffold
thus leading to a novel series of tricyclic S1P receptor modulators. However, in the absence of Xꢀray
1
coꢀcrystal structures of S1P with functional antagonists, it was not clear if substituents at the 3ꢀposition
1
1
7
of the phenyl ring would be tolerated by the receptor. Yan et al. disclosed a series S1P receptor
1
modulators with a methyl substituent at the 3ꢀposition of the phenyl ring (4, Figure 2). These
compounds were potent functional antagonists at S1P with excellent selectivity over S1P , suggesting
1
3
that substituents at the 3ꢀpostion were tolerated by the receptor. These data gave us the impetus to
followꢀup on our ligand based drug design hypothesis for the generation of novel S1P modulators
1
(Figure 2). As discussed below, this novel scaffold (5) allowed us to explore the structureꢀactivity
ACS Paragon Plus Environment

Page 5 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
relationship (SAR) of the polar head piece, phenyl ring (A), central ring (B), heterocyclic ring (C) and
the lipophilic tail piece as outlined in Fig. 2.
Figure 2. Design of tricyclic modulators of S1P1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CHEMISTRY
Schemes 1ꢀ4 outline the synthesis of representative examples of compounds with variations to
1
8
the phenyl ring, central ring, heterocyclic ring, the lipophilic tail piece and the polar head piece. The
key step in the synthesis of the bisꢀisoxazole 10 (Scheme 1) is the condensation of dilithiated 1ꢀ
1
9
tetralone oxime with methyl 3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazoleꢀ5ꢀcarboxylate. For optimal yields,
LiTMP was the base of choice to affect this condensation ꢀ LDA or nꢀBuLi gave poor yields. In the case
of the tricyclic chroman (24), the tetralone oxime route led to intractable mixtures of products during the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
condensation step. An alternative approach employing oxyamination of a β−diketone with
hydroxylamine via an eneamine led to the desired product, albeit in low yield (Scheme 2, see
2
0
transformation 18 to 19). The key step in the synthesis of the regioisomeric isoxazole tailpiece (36)
was an intramolecular 1,3ꢀdipolar cycloaddition of a nitrile oxide which was generated in situ from the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
21
oxime precursor 32 (Scheme 3). The synthesis of the thiazole (47, Scheme 4) was accomplished in
2
2
good yield via condensation of the acylamino ketone (41) with Lawesson’s reagent.
a
Scheme 1 . Synthesis of 10
a
o
o
Reagents and conditions: (a) (1) Tf
2
O, Py, 0 C to RT, 4 h, 94%; (2) tributyl(vinyl)stannane, LiCl, Pd(PPh
3
)
4
, dioxane, 100 C, 14 h, 80%; (3) NH
2
OH.HCl,
o
o
16
NaOAc, MeOH, 80 C, 1.5 h, 70%; (b) (1) TMSꢀCl, DIPEA, THF, 0 C; (2) LiTMP, THF, methyl 3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazoleꢀ5ꢀcarboxylate , ꢀ
7
o
o
o
2 4 4 2
8 C to 0 C; (3) SOCl , Py, PhMe, RT to 90 C, 25% for 3 steps; (c) OsO , NMO, THF, overnight then NaIO , H O, 30 min, 100%; (d) (1) tertꢀbutyl
azetidineꢀ3ꢀcarboxylate, NaBH(OAc)
3
, Ti(OiPr)
4
, MeOH, DCE, 2 h, 79%; (2) TFA, 90%
a
Scheme 2 . Synthesis of 24
ACS Paragon Plus Environment

Page 7 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
o
o
Reagents and conditions: (a) 3ꢀchloropropanoic acid, CF
6%; (d) tributyl(vinyl)stannane, lithium chloride, Pd(PhP)
, morpholine, toluene, 89%; (h) 3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazoleꢀ5ꢀcarbonyl fluoride , Et
O, 45 C, 17%; (i) SOCl , pyridine, toluene, 62%; (j) TFA, 100%; (k) NaIO , THF, H O; (l) tertꢀButyl azetidineꢀ3ꢀcarboxylate, acetic acid
, 64%; (m) TFA; MeOH slurry crystallization, 99%.
3
ꢀSO
4
3
H, 80 C, 75%; (b) 2M aq. NaOH, 5 CꢀRT; 6M aq. H
2
SO
4
, 81%; (c) (CF
3
ꢀSO
2
)
2
O, pyridine,
o
8
, dioxane, 100 C, 80%; (e) OsO
4
, NMO, THF, 99%; (f) dimethoxypropane, (1R)ꢀ(ꢀ)ꢀ
16
camphorsulfonic acid, 58%; (g) TiCl
HCl, NaOAc, H
salt, AcOH, titantium(IV) isopropoxide, MeOH, CH Cl
2 2
4
N, CH CN: NH OHꢀ
3 3 2
o
2
2
4
2
a
Scheme 3 . Synthesis of 36
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Reagents and conditions: (a) iꢀBuOCOCl, Nꢀmethylmorpholine, THF; NaBH
4
, THFꢀMeOH, 63%; (b) DessꢀMartin reagent, CH
2
Cl
2
, 99%%; (c) CBr
4
, PPh
OHꢀ
, NMO,
2
, 58%; (l) TFA;
3
,
THF, 67%; (d) nꢀBuLi,THF; paraformaldehyde, 63%; (e) PBr
HCl, Et
THF; NaIO
3
, CH Cl
2 2
, 40ꢀ60%; (f) 4ꢀbromoꢀ2ꢀhydroxybenzaldehyde, K
2
CO
3
, DMF, >85%; (g) NH
2
o
3
N, MeOH, 97%; (h) NaOCl, Et
3
N, CH
2
Cl
2
, 73%; (i) tributyl(vinyl)stannane, lithium chloride, (PhP)
4
Pd, dioxane, 100 C, 67%; (j) OsO
Cl
4
4
, THF, H O, 95%; (k) tertꢀButyl azetidineꢀ3ꢀcarboxylate, acetic acid salt, AcOH, titantium(IV) isopropoxide, MeOH, CH
2
2
neutralization; MeOH; slurry crystallization, 65%.
a
Scheme 4 . Synthesis of isomeric tricyclic S1P modulator 47
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
3 2
Reagents and conditions: (a) NBS, pTsOH, 60 °C, 11 min; (b) NaN1, acetone, water, 3h rt, 69% over two steps; (c) Pd/C, 2N HCl, MeOH, 50 psi H , rt 4h,
6
74%; (d) (1) 3ꢀPhenylꢀ4ꢀtrifluoromethyl)isoxazoleꢀ5ꢀcarboxylic acid , oxalyl chloride, DMF, DCM, 0 °C; (2) DMAP, Hunig’s base, DCM, rt overnight,
40%; (e) Lawesson’s reagent, THF, 30 min, 120 °C ꢀꢀwave, 60%; (f) BBr , DCM, 6h, rt, 100%; (g) Triflic anhydride, pyridine, 0 °C, 89%; (h)
, dioxane, 100 C, overnight; (i) OsO , NMO, THF, 3h, rt then NaIO , H O, 90 min, 47% over two steps; (j) (1)
3
o
Tributyl(vinyl)stannane, LiCl, Pd(PPh
3
)
4
4
4
2
3
Azetidineꢀ3ꢀcarboxylic acid, AcOH, MeOH, 1,2ꢀdichloroethane, 1h, 80 °C; (2) NaCNBH , rt, 1h, 25%
RESULTS AND DISCUSSION
All compounds were evaluated in a functional GTPγS assay for both human S1P and S1P
1
3
receptor agonism. Compounds with good S1P potency, and selectivity over S1P , and acceptable
1
3
profiles in liability assays (metabolic stability, CYP inhibition, etc.) were initially evaluated in a rat
pharmacokinetic/ pharmacodynamic (PK/ PD) model where rats were dosed orally with the compound.
Blood was drawn at the 4 and 24 hour time points to measure both circulating lymphocytes relative to
control and plasma concentration of each compound. Compounds which showed reduction in
circulating lymphocytes at low doses were then evaluated in PK studies and then progressed to chronic
ACS Paragon Plus Environment

Page 9 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
efficacy models such as the rat adjuvant arthritis model. The following tables summarize SAR related to
the lipophilic tail piece (Table 1), central ring B (Table 2), heterocyclic ring C (Table 3), phenyl ring A
(Table 4) as well as the PK/PD properties of these compounds.
Table 1. Lipophilic tail piece SAR.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In Vitro Activity
hS1P
hS1P GTPγS
Rat Blood Lymphocyte Reduction Assay
1
3
S1P
S1P
Sel.
3
/
Reduction
at 4 h
Plasma
Conc.at
4h (nM)
Reduction
at 24 h
Plasma
Conc. at
24h (nM)
Conc.
4h/24h
ratio
a
Cmpd
GTPγS
EC50 nM
1
a
EC50 nM
(1.0 mpk)
(1.0 mpk)
3
0.47±0.25
0.59±0.19
1,600±320
4,400±330
3,400x
7,500x
78%
2,100
91%
700
3
8
6%
120
55
85%
80% (0.3)
27
13
4
4
10
48
8
3% (0.3)
0.48±0.27
0.58±0.16
0.33 (n=2)
1.0±0.86
1,900 (n=2)
4,600±2,900
1,400±870
4,000x
7,900x
4,200x
1,300x
>2,100x
1,000x
17,000x
83%
160
190
150
120
520
190
28
87%
69%
43
32
4
6
4
9
74%
50
82% (0.4)
82% (0.5)
86%
85% (0.4)
69% (0.5)
80%
26
6
b
5
1
1,300
1
120
9
b
b
52
1.5
>3,100
61
b
b
5
5
5
5
5
5
5
6
3
4
5
6
7
8
9
0
1.3
1,300
85%
34%
4
48
6
b
1.5±1.5
4.6±2.6
26,000
57%
21%
5
>63,000±25,000 >14,000x
ꢀ5%
<LLQ
ND
110
68
12%
<LLQ
ND
50
b
b
24
11,000
460x
ND
ND
b
1.3±1.0
0.98±0.29
310±32
2,800
2,200x
2,100x
>200x
8,800x
74%
76%
2
4
2,100 (n=2)
>63,000±30,000
35% (0.5)
ND
32% (0.5)
ND
17
ND
ND
ND
ND
b
1.6±1.4
14,000
ND
ND
a
b
EC50 values are shown as mean values of at least three determinations; EC50 values are shown as a single determination; ND = Not Determined
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
6
As expected, based on our earlier effort lipophilic groups are well tolerated in this region of the
receptor leading to potent functional antagonists of S1P . Within this group however, both isomeric
1
isoxazoles, (10 and 48) and isothiazoles (49 and 50) exhibited picomolar potencies for S1P in the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
GTPγS assay with good selectivity against S1P . It is also interesting to note the SAR trends in the 5ꢀ
3
membered heteroaryl vs. the phenyl analogs – for example, a CF group on the 5ꢀmembered heteroaryl
3
ring, which is adjacent to the carbon atom anchoring the tricyclic group (R, in Table 1), led to very
potent functional antagonism at S1P . In the case of the phenyl analogs, however, ortho substitution of
1
the CF group to the R group led to a relative loss in potency at S1P in the functional assay (compare
3
1
compound 52, 57, and 54 with 56). Potency is regained when the CF group is moved to the meta
3
position in relation to the R group (compare 56 and 54). It is also important to note that lipophilic
moieties at the para position of the phenyl ring are critical in imparting functional potency at S1P – the
1
data for 59 is consistent with this SAR. Although subtle SAR trends described above are difficult to
explain from a functional potency perspective, it is quite possible that optimal occupancy of the
lipophilic pocket and the dihedral angle between the R group and the lipophilic tail piece may be
important factors contributing to the modest potency differences observed in this region of the molecule.
Since a number of compounds in Table 1 showed excellent functional potency for S1P while
1
maintaining selectivity for S1P they were advanced to the rat blood lymphocyte reduction (BLR)
3
,
pharmacokinetic/ pharmacodynamic (PK/PD) model. Compounds with the isoxazole and isothiazole tail
pieces showed sustained PD effects at 24 hours at low plasma concentrations – see for example, data for
5
0, 48 and 10 in Table 1. Since we had a good understanding of the liability profile of the isoxazoles
16
compared to the isothiazoles based on our earlier work, the isoxazoles (48 and 10) were determined to
be the optimized lipophilic tail pieces for exploring SAR of the rest of the tricyclic scaffold.
ACS Paragon Plus Environment

Page 11 of 57
Journal of Medicinal Chemistry
1
2
3
4
Table 2. SAR of the central ring, B.
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In Vitro Activity
hS1P GTPγS
Rat Blood Lymphocyte Reduction Assay
hS1P
1
3
S1P
S1P
Sel.
3
/
Reduction
at 4 h
(1.0 mpk)
Plasma
Conc. at
4h (nM)
Reduction
at 24 h
(1.0 mpk)
Plasma
Conc. at
24h (nM)
4
ratio
h/24h
a
Cmpd
GTPγS
EC50 nM
EC50 nM
1
a
1
0
0.59±0.19
6.8±1.4
4,400±330
7,500x
4,700x
83% (0.3)
14%
55
21
80% (0.3)
3%
13
4
b
61
62
63
32,000
<LLQ
8
1%
6% (0.3)
200
34
85%
79% (0.3)
46
8
4
4
b
1.3±0.37
31±23
4,400
3,400x
390x
6
12,000 (n=2)
ND
ND
330
230
ND
ND
ND
89% (0.5)
93% (0.5)
ND
ND
56
2
4
0.25 (n=2) >31,000±15,000 >120,000x
82% (0.5)
87% (0.5)
ND
6
5
36
0.27±0.18
1.6±0.32
1,200±1,300
4400x
46
6
6
4
5
>63,000±31,000 >39,000x
ND
ND
b
b
33
1,600
48x
ND
ND
a
b
EC50 values are shown as mean values of at least three determinations; EC50 values are shown as a single determination; ND = Not Determined
Table 2 summarizes SAR of the central ring B. Of the carbocyclic analogs examined, the six
membered ring (10, 48) was optimal – the five membered analog (61) was approximately an order of
magnitude less potent in the S1P GTPγS assay. Comparing the six and seven membered analogs (62
1
vs. 63), the six membered analog was significantly more potent. This data was also consistent with the
data for the seven membered oxepane analog 65. The significant loss in potency for the seven
membered analogs compared to the five and six membered compounds supports our initial hypothesis
about the importance of maintaining planarity between the phenyl group (which anchors the polar
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
headpiece) and the central heterocyclic ring for optimal potency and selectivity for the S1P receptor.
1
Of the heterocyclic rings examined, the pyran analogs (24 and 36) were equipotent to the carbocyclic
analogs (10 and 48) in the S1P GTPγS assay. Since both 24 and 36 were about equipotent in the S1P
1
1
GTPγS assay, we tested them in the rat BLR assay for further progression. As outlined in Table 2, both
analogs were efficacious in reducing lymphocyte count at very low doses and these reductions were
similar to what was observed with the carbocyclic analog 10. We decided to use the carbocyclic analog
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
0 for further analoging, since the synthetic feasibility of exploring SAR appeared to be easier with this
compound compared to pyrans 24 and 36.
Table 3. SAR of the heterocyclic ring, C.
In Vitro Activity
Rat Blood Lymphocyte Reduction Assay
hS1P
1
hS1P
EC50 nM
3
GTPγS
S1P
S1P
Sel.
3
/
Reduction
at 4 h
(1.0 mpk)
Plasma
Conc. at
4h (nM)
Reduction
at 24 h
(1.0 mpk)
Plasma
Conc. at 24h
(nM)
4
ratio
h/24h
a
Cmpd
GTPγS
EC50 nM
1
a
1
6
6
6
6
4
0
6
7
8
9
7
0.59±0.19
190±23
380±150
21±16
4,400±330
7,500x
>330x
>170x
>3,000x
170x
83% (0.3)
ND
55
80% (0.3)
ND
13
ND
ND
ND
11
4
>63,000±30,000
>63,000±30,000
>63,000±35,000
ND
ND
ND
110
230
ND
ND
ND
ND
b
5.5±3.6
2.4 (n=2)
920
62%
2%
10
3
b
12,000
5,000x
80%
85%
84
a
b
EC50 values are shown as mean values of at least three determinations; EC50 values are shown as a single determination; ND = Not Determined
ACS Paragon Plus Environment

Page 13 of 57
Journal of Medicinal Chemistry
1
2
3
4
Table 3 summarizes the SAR of the heterocyclic ring C, (Fig. 1). It is clear from Table 3 that of
the various heterocyclic rings examined, the [1,2ꢀc]isoxazole (10) is the most potent in the S1P GTPγS
1
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
assay. It is of interest to note that the isomeric [2,1ꢀd]oxazole (69) is an order of magnitude less potent
vs. S1P in the GTPγS assay compared to compound 10. Additionally, in the rat BLR PK/PD assay
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
compound 69 was significantly less efficacious in reducing lymphocyte counts through 24 hours
compared to the isoxazole (10). On the contrary, the thiazole analog 47, which is similar in potency to
6
9 in the S1P GTPγS assay, was highly efficacious in reducing lymphocyte counts in the rat BLR assay.
1
This is likely due to the higher plasma concentration of 47 compared to 69 at the 24 hour time point.
Although we did not do exhaustive SAR studies on the aryl ring (A, in Fig. 2), the limited SAR
work suggested that the phenyl analog (10) provided the best balance of S1P potency vs. selectivity for
1
S1P in the GTPγS assay (Table 4).
3
Table 4. SAR of Ring A.
In Vitro Activity
hS1P hS1P
Rat Blood Lymphocyte Reduction Assay
1
3
S1P
S1P
Sel.
3
/
Reduction
at 4 h
Plasma
Conc. at
4h (nM)
Reduction
at 24 h
Plasma
Conc. at
24h (nM)
4
ratio
h/24h
Cmpd
GTPγS
GTPγS
1
a
a
EC50 nM
EC50 nM
(1.0 mpk)
(1.0 mpk)
1
7
7
0
0
1
0.59±0.19
4,400±330
7,500x
1,400x
9,400x
83% (0.3)
85%
55
80% (0.3)
34%
13
4
4
48
7
b
b
0.87
1,200
190
320
b
1.6±0.32
15,000
74% (0.5)
79% (0.5)
46
>
(
63,000
n=2)
b
7
2
17
>3,700x
76%
460
32%
47
10
a
b
EC50 values are shown as mean values of at least three determinations; EC50 values are shown as a single determination; ND = Not Determined
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Since we have shown that the azetidine carboxylic acid head piece can be replaced by an
2
3
ethanolamine group, we decided to explore this SAR along with other polar moieties in the tricylic
chemotype (Table 5).
Table 5. SAR of polar head piece.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In Vitro Activity
hS1P hS1P
Rat Blood Lymphocyte Reduction Assay
1
3
S1P
S1P
Sel.
3
/
Reduction
at 4 h
Plasma
Conc. at
4h (nM)
Reduction
at 24 h
Plasma
Conc. at
24h (nM)
4
ratio
h/24h
Cmpd
GTPγS
GTPγS
1
a
a
EC50 nM
EC50 nM
(1.0 mpk)
(1.0 mpk)
1
0
0.59±0.19
4.0±1.9
4,400±330
7,500x
1,900x
83% (0.3)
70%
55
80% (0.3)
72%
13
4
1
b
73
74
7,400
190
160
8
1%
280
93
87%
78% (0.3)
77
32
4
3
0.33±0.20
1,700 (n=2)
5,200x
7
0% (0.3)
13,000
(n=2)
7
7
5
6
0.21±0.045
62,000x
76%
140
76%
49
3
7
b
26±7.8
200±89
1.6±0.58
51±9.0
16,000
620x
90x
28%
ND
140
ND
190
ND
20%
ND
21
ND
26
b
77
18,000
b
7
8
6,200
3,900x
450x
50 (0.5)
43 (0.5)
ND
b
79
23,000
ND
ND
a
b
EC50 values are shown as mean values of at least three determinations; EC50 values are shown as a single determination; ND = Not Determined
As is evident from Table 5, ethanolamines such as 74 and 75 and αꢀhydroxy amides such as 78
showed potencies comparable to compound 10 in the S1P GTPγS assay while showing good selectivity
1
for S1P . It is interesting to note that chirality at the carbon anchoring the hydroxy group is important in
3
determining potency and selectivity for S1P in both the ethanolamine and the αꢀhydroxy amide series
1
(
the absolute configuration of the isomers 73-78 was not established in this series).
It is clear from the discussion above, that a number of compounds show excellent lymphopenia
at low doses in the rat BLR assay. However, compound 10 remained one of the lead compounds in
ACS Paragon Plus Environment

Page 15 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
terms of showing a sustained PD response at very low doses (Figure 3) with an ED of 0.05 mg/kg
50
(
EC = 2.5 ± 0.8 nM) in addition to displaying an excellent liability profile (Table 6). Moreover, the
50
PK profile of the compound across species (Table 7) was very promising (low clearance, moderate to
high bioavailability and volumes of distribution) and the compound was advanced to the rat adjuvant
arthritis (AA) model, a preclinical model of rheumatoid arthritis. The rationale for testing compound 10
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
in a rat AA model is as follows. While the current market approval for Gilenya, an S1P agonist, is for
1
RRMS, S1P agonists have been or are currently being clinically evaluated for a wide range of
1
autoimmune disorders (psoriasis, ulcerative colitis, Crohn’s disease, organ transplant, lupus). At the
time of these investigations, arthritis was of primary interest for this program.
Figure 3. Rat lymphocyte reduction of compound 10
Compound 10
Rat Lymphocyte Reduction
6
5
4
0%
0%
0%
30%
4
h
2
1
0%
0%
24 h
0
%
0
.05 0.3 1.0
mpk mpk mpk
Dose
Table 6. Partial in vitro profiling data for compound 10
Parameter
Result
Human S1P
Human S1P
Human S1P
Human S1P
Human S1P
Human S1P
1
1
1
3
4
5
binding (IC50
GTPγS EC50 (Ymax)
cAMP EC50 (Ymax)
)
0.38 nM
0.59 nM (95%)
0.032 nM (93%)
7,500
2.4x
2.4x
0.31 nM
6100x
99.6% human
/S1P
/S1P
/S1P
1
1
1
selectivity
selectivity
selectivity
Rat S1P
Rat S1P
1
GTPγS (EC50
)
3
/S1P selectivity
1
Protein Binding (bound)
9
9
9.3% mouse
9.6% rat
99.5% dog
9.5% monkey
9
hERG (Patch Clamp)
Na (Patch Clamp)
IC50 = 2 ꢀM
12% @ 10 ꢀM (1 and 4
+
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
+
Ca (Patch Clamp)
Hz)
PXR TransAct, EC50
11% @ 10 ꢀM
>
50 ꢀM
a
CYP inhibition (IC50
)
>40 ꢀM 1A2, 2B6, 2C9,
2C19, 2D6, 3A4ꢀBZR
=
14 ꢀM 2C8
280 nm/s @ pH 7.7
60 nm/s @ pH 5.5
PAMPA permeability
2
Met Stab, T1/2
CYP = cytochrome P450
>120 min (h, m, r, d, mk)
a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Rat adjuvant arthritis study of compound 10
Compound 10 in Rat Adjuvant Arthritis
2
2
1
.5
.0
.5
Vehicle
Cmpd. 10: 0.03 mg/kg
Cmpd. 10: 0.1 mg/kg
Cmpd. 10: 1.0 mg/kg
Dex-0.1 mg/kg
1.0
0
.5
.0
*
*
*
*
0
7
9
11
13
15
17
19
21
Mean +/- SEM
n=8/group
Study Day
*
p<.05 vs. Vehicle
ANOVA w/ Dunnett's
Table 7. Pharmacokinetic parameters for compound 10
a
a
a,b
a
Parameter
Mouse
5
Rat
1
Cyno
Dog
0.1
po dose (mg/kg)
iv dose (mg/kg)
0.5
0.5
0.41
4
2
1
0.1
C
max (ꢀM)
3.6
3
0.14
4
0.09
8
T
max (h)
AUC (ꢀM*h) PO
30.3
15
1.5
7.9
8.5
4.8
43
3.4
18
2.3
23.1
1.4
2.5
95
T
1/2 (h)
CL (mL/min/kg)
ss (L/kg)
po (%)
1.7
2.2
80
2.4
3.6
80
V
F
a
b
Average of three animals; cynomologus monkey.
Arthritis was induced in male Lewis rats by a subcutaneous injection of complete Freund’s
adjuvant at the base of the tail. Compound 10 was dosed orally (PO), once a day (QD), for 21 days at
0
.03, 0.1 and 1.0 mg/kg. Dexamethasone served as the positive control in the study and was dosed at
.1 mg/kg PO/QD. Paw volumes (day 7 to day 21) were measured using water displacement
0
ACS Paragon Plus Environment

Page 17 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
plethysmometry. As is evident from Figure 4, a dose of 0.03 mg/kg had little impact on disease
development. However, a dose of 0.1 mg/kg provided >50% suppression of paw swelling, and the
higher dose of 1mg/kg resulted in robust inhibition equivalent to the positive control dexamethasone.
On the basis of the suppression of paw swelling observed in this study, the ED for compound 10 was
5
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
determined to be 0.09 mg/kg (EC = 24 nM) which is comparable to other S1P modulators disclosed in
5
0
1
1
6,23
the literature.
CONCLUSION
In summary, a series of tricyclic S1P receptor agonists with selectivity over S1P have been
1
3
identified employing a ligandꢀbased drug design approach. SAR in this series suggested that (a) a
number of lipophilic tail pieces are tolerated (b) the azetidine carboxylic acid (10) and the hydroxy
nipecotic acid (74 and 75) are optimal head pieces (c) six membered carbocylic and pyran rings are
optimal central rings and (d) [1,2ꢀc]isoxazole is the preferred heteroaromatic ring. Based on the
extensive SAR carried out, compound 10 was identified as having the optimal properties in terms of
potency for S1P , selectivity vs. S1P with an excellent liability profile. This compound showed
1
3
sustained pharmacodynamic effects in a rat blood lymphocyte reduction assay and efficacy in a rat
adjuvant arthritis model, with an ED comparable to reported S1P modulators. Based upon this
5
0
1
favorable profile, compound 10 was further evaluated for suitability as a clinical development candidate.
Experimental
All commercially available chemicals and solvents were used without further purification. Reactions
1
were performed under an atmosphere of nitrogen. All new compounds gave satisfactory H NMR,
1
LC/MS and/or HRMS, and mass spectrometry results. H NMR spectra were obtained on a Bruker 400
MHz or a JEOL 500 MHz NMR spectrometer using the residual TMS signal of deuterated NMR solvent
as internal reference. Electrospray ionization (ESI) mass spectra were obtained on a Waters ZQ single
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
quadrupole mass spectrometer. All MS measurements were made on open access systems. The purity of
tested compounds determined by analytical HPLC was >95%. HPLC Methods are listed below:
Method A: A linear gradient using 5% acetonitrile, 95% water, and 0.05% TFA (Solvent A) and 95%
acetonitrile, 5% water, and 0.05% TFA (Solvent B); t = 0 min., 10% B, t = 15 min., 100% B (20 min.)
was employed on a SunFire C18 3.5u 3.0 x 150 mm column. Flow rate was 1.0 ml/min and UV
detection was set to 220/254 nm. The LC column was maintained at ambient temperature.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Method B: A linear gradient using 5% acetonitrile, 95% water, and 0.05% TFA (Solvent A) and 95%
acetonitrile, 5% water, and 0.05% TFA (Solvent B); t = 0 min., 10% B, t = 15 min., 100% B (20 min.)
was employed on a XBridge Ph 3.5u 3.0 x 150 mm column. Flow rate was 1.0 ml/min and UV
detection was set to 220/254 nm. The LC column was maintained at ambient temperature.
Method C. A linear gradient using 10% methanol, 90% water, and 0.2% H PO (Solvent A) and 90%
3
4
methanol, 10% water, and 0.2% H PO (Solvent B); t = 0 min., 0% B, t = 4 min., 100% B (5 min.) was
3
4
employed on a YMC S5 CombiScreen 4.6 x 50 mm column. Flow rate was 4.0 ml/min and UV
detection was set to 220 or 254 nm. The LC column was maintained at ambient temperature.
Method D. A linear gradient using 10% methanol, 90% water, and 0.2% H PO (Solvent A) and 90%
3
4
methanol, 10% water, and 0.2% H PO (Solvent B); t = 0 min., 0% B, t = 4 min., 100% B (5 min.) was
3
4
employed on a Chromolith SpeedROD 4.6 x 50 mm column. Flow rate was 4.0 ml/min and UV
detection was set to 220 nm. The LC column was maintained at ambient temperature.
6
-Vinyl-3,4-dihydronaphthalen-1(2H)-one oxime (7). To a solution of 6ꢀhydroxyꢀ3,4ꢀ
dihydronaphthalenꢀ1(2H)ꢀone (2 g, 12.33 mmol) in anhydrous pyridine (10 mL) was added
trifluoromethanesulfonic anhydride (2.5 mL, 14.80 mmol) at 0 °C over a period of 5 min. The reaction
mixture was allowed to come to room temperature and stirred at room temperature for 4 h. The reaction
mixture was concentrated under reduced pressure. The resulting brownish red residue was partitioned
between ether (60 mL) and water (30 mL). The ether layer was sequentially washed with 1N
ACS Paragon Plus Environment

Page 19 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
hydrochloric acid (20 mL), sat. aq. NaHCO (20 mL), brine (20 mL), dried over sodium sulfate,
3
concentrated, and purified by silica gel column chromatography using hexane/ethyl acetate to yield 5ꢀ
1
oxoꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl trifluoromethanesulfonate (3.4 g, 11.55 mmol, 94 % yield). H
NMR (400 MHz, CDCl ) δ ppm 8.14 (1 H, d, J=8.4 Hz), 7.17ꢀ7.24 (2 H, m), 3.02 (2 H, t, J=6.1 Hz),
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
.69 (2 H, t, J=6.5 Hz), 2.15ꢀ2.22 (2 H, m).
To a solution of 5ꢀoxoꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀylꢀtrifluoromethanesulfonate (3.4 g, 11.55
mmol) in anhydrous dioxane was sequentially added tributyl(vinyl)stannane (3.73 mL, 12.71 mmol),
lithium chloride (1.470 g, 34.7 mmol), and tetrakis(triphenylphosphine)palladium(0) (1.335 g, 1.155
mmol). The reaction mixture was purged with nitrogen gas for 5 min. and heated at 100 °C for 14 h.
The reaction mixture was cooled to room temperature and filtered. The yellow residue was washed with
ethyl acetate (3 x 20 ml). The filtrate was concentrated under reduced pressure. To the red oil was
added ether (50 mL) and the contents were stirred at room temperature for 15 min. The contents were
filtered and the resulting brick red residue was washed with ether (3 x 20 mL). The filtrate was
concentrated under reduced pressure and the resulting red oil was purified by silica gel column
chromatography using hexane/ethyl acetate to yield 6ꢀvinylꢀ3,4ꢀdihydronaphthalenꢀ1(2H)ꢀone (1.6 g,
1
9
1
.29 mmol, 80 % yield). H NMR (400 MHz, CDCl ) δ ppm 8.00 (d, J=8.14 Hz, 1 H), 7.36 (dd, J=8.14,
3
.10 Hz, 1 H), 7.26 (s, 1 H), 6.72 (dd, J=17.61, 11.00 Hz, 1 H), 5.87 (dd, J=17.61, 0.66 Hz, 1 H), 5.37ꢀ
5.41 (m, 1 H), 2.96 (t, J=6.05 Hz, 2 H), 2.63ꢀ2.67 (m, 2 H), 2.10ꢀ2.17 (m, 2 H).
To 6ꢀvinylꢀ3,4ꢀdihydronaphthalenꢀ1(2H)ꢀone (1.6 g, 9.29 mmol) in methanol (10 mL) was
sequentially added hydroxylamine hydrochloride (0.775 g, 11.15 mmol) and sodium acetate (0.915 g,
11.15 mmol). The reaction mixture was heated at 80 °C (oil bath temp.) for 1.5 h. The reaction mixture
was concentrated under reduced pressure and to the residue was added water (30 mL). The contents
were triturated and filtered. The solid material was washed with water (2 x 20 mL) and dried overnight
1
to yield 6ꢀvinylꢀ3,4ꢀdihydronaphthalenꢀ1(2H)ꢀone oxime (1.2 g, 6.41 mmol, 69.0 % yield). H NMR
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(400 MHz, CDCl ) δ ppm 8.23 (br. s., 1 H), 7.85 (d, J=8.14 Hz, 1 H), 7.28 (s, 1 H), 7.18 (s, 1 H), 6.69
3
(dd, J=17.61, 10.78 Hz, 1 H), 5.78 (dd, J=17.61, 0.88 Hz, 1 H), 5.23ꢀ5.32 (m, 1 H), 2.71ꢀ2.87 (m, 4 H),
1
.81ꢀ1.94 (m, 2 H).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3-(3-phenyl-4-(trifluoromethyl)isoxazol-5-yl)-7-vinyl-4,5-dihydronaphtho[1,2-c]isoxazole
8). Preparation of LiTMP: To a stirred solution of 2,2,6,6ꢀtetramethylpiperidine (10.88 mL, 64.5 mmol)
(
in anhydrous THF (25 mL) was added nꢀBuLi (25.8 mL, 64.5 mmol) (2.5M in hexanes) dropwise at 0
°C under nitrogen. The pale yellow solution was then stirred at the same temperature for 20 min.
To a stirred solution of 6ꢀvinylꢀ3,4ꢀdihydronaphthalenꢀ1(2H)ꢀone oxime (7, 4.03 g, 21.50 mmol)
and diisopropylethylamine (6.83 mL, 39.1 mmol) in anhydrous THF (50 mL) was added TMSꢀCl (2.75
mL, 21.50 mmol) dropwise at 0 °C under nitrogen. The mixture was stirred at 0 °C for 30 min. The
previously prepared LiTMP solution was added dropwise at ꢀ78 °C under nitrogen. After the mixture
was stirred at the same temperature for 30 min, a solution of methyl 3ꢀphenylꢀ4ꢀ
1
6
(trifluoromethyl)isoxazoleꢀ5ꢀcarboxylate (5.3 g, 19.54 mmol) in anhydrous THF (10 mL) was added
dropwise at ꢀ78 °C. The solution was stirred at ꢀ78 °C for 30 min and then warmed to 0 °C over 30 min.
The reaction was quenched with saturated aqueous ammonium chloride solution (25 mL) and water (20
mL). EtOAc (50 mL) was added, contents stirred at room temperature for 30 min and filtered through a
pad of celite. The filtrate was separated. The aqueous layer was extracted with EtOAc (2 x 25 mL). The
combined organic solutions were washed with brine (50 mL), dried over sodium sulfate and
concentrated to give a liquid. Flash chromatography purification (330g silica gel column, 15ꢀ30% ethyl
acetate in hexanes) afforded 3ꢀ(3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ7ꢀvinylꢀ3,3a,4,5ꢀ
tetrahydronaphtho[1,2ꢀc]isoxazolꢀ3ꢀol (5 g, 11.73 mmol, 60.0 % yield). The compound had an HPLC
+
retention time = 3.74 min (condition C); LCMS (ESI) m/z Calcd for C23H18F3N2O3 [M + H] 427.13.
Found: 426.9.
ACS Paragon Plus Environment

Page 21 of 57
Journal of Medicinal Chemistry
1
2
3
4
The 3ꢀ(3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ7ꢀvinylꢀ3,3a,4,5ꢀtetrahydronaphtho[1,2ꢀ
c]isoxazolꢀ3ꢀol (5 g, 11.73 mmol) was mixed with thionyl chloride (1.712 mL, 23.45 mmol) and
anhydrous toluene (80 mL). Pyridine (0.190 mL, 2.345 mmol) was then added dropwise. The mixture
was stirred under nitrogen at room temperature for 30 min and 90 °C for 15 min. The mixture was
concentrated under reduced pressure. The residue was mixed with saturated aqueous sodium bicarbonate
solution (50 mL) and extracted with dichloromethane (100 mL, 2 x 50 mL). The combined
dichloromethane extracts were dried over sodium sulfate, filtered through a pad of silica gel and
concentrated under reduced pressure. Flash chromatography purification (220g silica gel column, 25ꢀ
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
>
60% dichloromethane in hexanes) and trituration with methanol gave 3ꢀ(3ꢀphenylꢀ4ꢀ
(
trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ7ꢀvinylꢀ4,5ꢀdihydronaphtho[1,2ꢀc]isoxazole (1.96 g, 4.80 mmol, 40.9 %
yield). The compound had an HPLC retention time = 4.28 min (condition C); LCMS (ESI) m/z Calcd for
+
C H F N O [M + H] 409.12. Found: 408.9.
23
16
3
2
2
3
-(3-phenyl-4-(trifluoromethyl)isoxazol-5-yl)-4,5-dihydronaphtho [1,2-c]isoxazole-7-
carbaldehyde (9). To a stirred mixture of 3ꢀ(3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ7ꢀvinylꢀ4,5ꢀ
dihydronaphtho[1,2ꢀc]isoxazole (1.96 g, 4.80 mmol) and THF (20 mL) were added NMO (50% in
water, 1.493 mL, 7.20 mmol) and osmium tetroxide (4% in water, 1.173 mL, 0.192 mmol) at room
temperature. The mixture was vigorously stirred at room temperature overnight. A solution of sodium
periodate (1.540 g, 7.20 mmol) in water (10 mL) was added. The mixture was stirred at room
temperature under nitrogen for 30 min. Water (20 mL) was then added. The solid was filtered, washed
with water (2 x 5 mL) and ethanol (2 mL) and dried to give 3ꢀ(3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazolꢀ5ꢀ
yl)ꢀ4,5ꢀdihydronaphtho[1,2ꢀc]isoxazoleꢀ7ꢀcarbaldehyde (1.92 g, 4.68 mmol, 98%). The compound had
+
an HPLC retention time = 3.95 min (condition C); LCMS (ESI) m/z Calcd for C H F N O [M + H]
2
2
14
3
2
3
4
11.10. Found: 410.9.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1-((3-(3-Phenyl-4-(trifluoromethyl)isoxazol-5-yl)-4,5-dihydronaphtho[1,2-c]isoxazol-7-
yl)methyl)azetidine-3-carboxylic acid (10). To a stirred solution of 3ꢀ(3ꢀphenylꢀ4ꢀ
(
trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ4,5ꢀdihydronaphtho[1,2ꢀc]isoxazoleꢀ7ꢀcarbaldehyde (2.1 g, 5.12 mmol),
tertꢀbutyl azetidineꢀ3ꢀcarboxylate, AcOH (1.668 g, 7.68 mmol), and acetic acid (0.586 mL, 10.24 mmol)
in anhydrous MeOH (10 mL) and anhydrous 1,2ꢀdichloroethane (30 mL) was added titanium(IV)
isopropoxide (3.05 mL, 10.24 mmol) dropwise at room temperature under nitrogen. The solution was
stirred at room temperature for 1 hr before sodium triacetoxyborohydride (4.34 g, 20.47 mmol) was
added in 10 portions over 3.5 hr. The mixture was stirred at room temperature for 2 hr. Saturated
sodium bicarbonate (70 mL) was added slowly to make the reaction mixture basic. Celite was added,
and the contents were stirred at room temperature for 30 min and then filtered through a pad of celite.
The aqueous filtrate was separated and extracted with dichloromethane (3 x 20 mL). The combined
organic solutions were dried (sodium sulfate) and concentrated. Flash chromatography purification
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
120g silica gel column, 25ꢀ>70% ethyl acetate in hexanes) afforded tertꢀbutyl 1ꢀ((3ꢀ(3ꢀphenylꢀ4ꢀ
trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ4,5ꢀdihydronaphtho[1,2ꢀc]isoxazolꢀ7ꢀyl)methyl) azetidineꢀ3ꢀcarboxylate
(
(2.24 g, 4.06 mmol, 79 % yield). The compound had an HPLC retention time = 3.40 min (condition C);
+
LCMS (ESI) m/z Calcd for C H F N O [M + H] 552.21. Found: 552.1.
3
0
29
3
3
4
To a stirred solution of tertꢀbutyl 1ꢀ((3ꢀ(3ꢀphenylꢀ4ꢀ(trifluoromethyl) isoxazolꢀ5ꢀyl)ꢀ4,5ꢀ
dihydronaphtho[1,2ꢀc]isoxazolꢀ7ꢀyl)methyl)azetidineꢀ3ꢀcarboxylate (2.24 g, 4.06 mmol) in
dichloromethane (4 mL) was added TFA (4 mL) slowly at room temperature. The reaction mixture was
stirred at room temperature for 2 hr. The mixture was concentrated to remove dichloromethane and
TFA (6 mL) was added. The mixture was stirred at room temperature till the starting material
disappeared. Dichloroethane (6 mL) was added and the reaction mixture was concentrated under
reduced pressure. The residue was mixed with water (10 mL), made basic (pH ~8) with 1 N aqueous
NaOH and acidified with 1N aqueous HCl to pH ~5. The mixture was sonicated for 30 min. The solid
ACS Paragon Plus Environment

Page 23 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
was filtered and washed with water (4 x 10 mL). The residue was mixed with methanol (40 mL),
sonicated for 30 min, filtered, washed with methanol (3 x 10 mL), and dried to give 1ꢀ((3ꢀ(3ꢀphenylꢀ4ꢀ
(trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ4,5ꢀdihydronaphtho[1,2ꢀc]isoxazolꢀ7ꢀyl)methyl) azetidineꢀ3ꢀcarboxylic
+
acid. HPLC t = 7.62 min. (Method A); LCMS (ESI) m/z Calcd for C H F N O [M + H] 496.15.
r
26 21
3
3
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
Found: 496.0; H NMR (500 MHz, DMSO) δ ppm 7.85 (1 H, d, J=7.90 Hz), 7.64 ꢀ 7.68 (2 H, m), 7.57 ꢀ
7.65 (3 H, m), 7.37 (1 H, s), 7.33 (1 H, d), 3.60 (2 H, s), 3.42 (2 H, br. s.), 3.23 (3 H, br.s.), 3.10 (2 H, d,
J=6.41 Hz), 3.05 (2 H, d, J=6.41 Hz).
3
-chloro-1-(2,4-dihydroxyphenyl)propan-1-one (12). To a stirred mixture of resorcinol (10 g,
9
3
1 mmol) and 3ꢀchloropropanoic acid (9.95 g, 92 mmol) was added trifluoromethanesulfonic acid (50 g,
33 mmol) in one portion. The solution was heated at 80 °C for 30 min, cooled to room temperature
over 15 min, and poured into dichloromethane (200 mL). The resulting solution was slowly poured into
water (200 mL), and the layers were separated. The aqueous layer was extracted with dichloromethane
(2 x 100 mL). The combined organic solutions were dried over anhydrous sodium sulfate, and filtered.
Concentration under reduced pressure afforded 3ꢀchloroꢀ1ꢀ(2,4ꢀdihydroxyphenyl)propanꢀ1ꢀone (13.62 g,
67.9 mmol, 74.8 % yield) as an orange semiꢀsolid. The compound was used without any further
1
purification. HꢀNMR (400 MHz, CDCl ) δ ppm 12.39 (1H, s), 7.56 (1H, d, J=8.58 Hz), 7.19 (1H, s),
3
6.17 ꢀ 6.49 (1H, m), 5.23 (1H, s), 3.78 ꢀ 4.01 (2H, m), and 3.34 (2H, t, J=6.71 Hz).
7
-hydroxychroman-4-one (13). To a stirred 2M solution of sodium hydroxide (500 mL, 1000
o
mmol) at 5 C was added 3ꢀchloroꢀ1ꢀ(2,4ꢀdihydroxyphenyl)propanꢀ1ꢀone (13.6 g, 67.9 mmol) in one
o
portion. The solution was warmed to room temperature over 2h, reꢀcooled to 5 C, and the pH was
adjusted to ~2 with 6M aqueous sulfuric acid (~50 mL). The orange precipitate formed was collected by
vacuum filtration and dried under reduced pressure. The filtrate was extracted with ethyl acetate (3 x
100 mL), washed with brine, and dried over anhydrous sodium sulfate. Concentration under reduced
pressure gave an orange solid. The two recovered solids were combined to give 7ꢀhydroxychromanꢀ4ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
+
one (9.0 g, 54.8 mmol, 81% yield). LCMS (ESI) m/z Calcd for C H O [M + H] 165.1. Found: 165.0.
9
8
3
1
HꢀNMR (400 MHz, CDCl ) δ ppm 7.77 (1H, d, J=8.58 Hz), 6.45 (1H, dd, J=8.69, 2.31 Hz), 6.33 (1H,
3
d, J=2.42 Hz), 5.62 (1H, br. s.), 4.33 ꢀ 4.66 (2H, m), and 2.58 ꢀ 2.90 (2H, m).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4-oxochroman-7-yl trifluoromethanesulfonate (14). To a solution of 7ꢀhydroxychromanꢀ4ꢀone
(
2.024 g, 12.3 mmol) in pyridine (10 mL) at 0 °C was added trifluoromethanesulfonic anhydride (2.500
mL, 14.8 mmol) over 5 min. The iceꢀbath was removed, and the reaction mixture was stirred at room
temperature for 2.5 h. Pyridine was removed under reduced pressure, and the residue was diluted with
ether (70 mL) and washed with water (30 mL). The organic layer was collected, and the aqueous layer
was backꢀextracted with ether (25 mL). The combined organic layers were washed with 1N aqueous
hydrochloric acid, 1N aqueous sodium hydroxide, brine, and dried over anhydrous sodium sulfate.
Ether was removed under reduced pressure, and the crude product was purified by silica gel flash
chromatography using a mixture of ethyl acetate in hexane (5% ꢀ 15% ꢀ 20%) to afford 4ꢀoxochromanꢀ
7
ꢀyl trifluoromethanesulfonate (3.13 g, 10.6 mmol, 86% yield) as a yellow oil. HPLC t = 2.40 min.
r
+
(
Method D). LCMS (ESI) m/z Calcd for C H F O S [M + H] 297.0. Found: 297.2.
1
0
7
3
5
7-vinylchroman-4-one (15). To a solution of 4ꢀoxochromanꢀ7ꢀyl trifluoromethanesulfonate
2.00 g, 6.75 mmol) in dioxane (10 mL) in a sealed tube was added sequentially tributyl(vinyl)stannane
2.18 mL, 7.43 mmol) and lithium chloride (0.859 g, 20.3 mmol). The mixture was degassed under
(
(
reduced pressure and charged with nitrogen (2x). To the mixture was added
tetrakis(triphenylphosphine)palladium(0) (0.780 g, 0.675 mmol), and the contents stirred under a strong
stream of nitrogen for 5 min. The reaction mixture was sealed, immersed in an oil bath at 100 °C, and
stirred overnight. The reaction mixture was cooled to room temperature and filtered under reduced
pressure. The solid was washed with ethyl acetate (4 x 50 mL), and the filtrate was concentrated. The
crude product mixture was diluted with ether (~100 mL), sonicated for several minutes, and filtered (the
cake was rinsed with ether). The filtrate was concentrated under reduced pressure, and the residue was
ACS Paragon Plus Environment

Page 25 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
purified by silica gel flash chromatography using a mixture of ethyl acetate in hexane (5% ꢀ 12% ꢀ 15%)
to give 7ꢀvinylchromanꢀ4ꢀone (0.938 g, 5.38 mmol, 80 % yield) as a yellow oil. HPLC t = 1.93 min.
r
+
1
(
Method D). LCMS (ESI) m/z Calcd for C H O [M + H] 175.1. Found: 174.9. HꢀNMR (400 MHz,
11 10 2
CDCl ) δ ppm 7.87 (d, J=8.14 Hz, 1H), 7.09 (dd, J=8.25, 1.43 Hz, 1H), 6.98 (d, J=1.54 Hz, 1H), 6.69
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(dd, J=17.61, 11.00 Hz, 1H), 5.88 (d, J=17.61 Hz, 1H), 5.42 (d, J=10.78 Hz, 1H), 4.52 ꢀ 4.58 (m, 2H),
and 2.78 ꢀ 2.85 (m, 2H).
7-(1,2-dihydroxyethyl)chroman-4-one (16). To a mixture of 7ꢀvinylchromanꢀ4ꢀone (0.801 g,
.60 mmol) and a 50% solution of NMO in water (0.953 mL, 4.60 mmol) in tetrahydrofuran (11 mL) at
4
room temperature was added a 4% aqueous solution of osmium tetroxide in water (1.44 mL, 0.184
mmol). The reaction mixture was stirred at room temperature overnight. The reaction mixture was
diluted with dichloromethane (200 mL) and washed with water (40 mL). The organic layer was
collected and was washed with brine (40 mL). The combined aqueous layers were extracted with
dichloromethane (200 mL), and the combined organic layers were dried over anhydrous sodium sulfate.
Concentration under reduced pressure afforded 7ꢀ(1,2ꢀdihydroxyethyl)chromanꢀ4ꢀone (0.950 g, 4.56
mmol, 99% yield) as a dark yellow oil. HPLC t = 0.587 min. (Method D); LCMS (ESI) m/z Calcd for
r
+
C H O [M + H] 209.1. Found: 209.0.
11
12
4
7
-(2,2-dimethyl-1,3-dioxolan-4-yl)chroman-4-one (17). To a mixture of 7ꢀ(1,2ꢀ
dihydroxyethyl)chromanꢀ4ꢀone (0.950 g, 4.56 mmol) and 2,2ꢀdimethoxypropane (1.68 mL, 13.7 mmol)
in acetone (20 mL) at room temperature was added (1R)ꢀ(ꢀ)ꢀcamphorsulfonic acid (0.212 g, 0.913
mmol). The reaction mixture was stirred at room temperature overnight. The reaction mixture was
quenched with a saturated aqueous solution of sodium bicarbonate and concentrated under reduced
pressure. The aqueous residue was diluted with ethyl acetate, washed with a saturated aqueous solution
of sodium bicarbonate, and brine. The organic layer was collected, and the aqueous layer was extracted
with ethyl acetate. The combined organic layers were dried over anhydrous sodium sulfate and
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
concentrated to give the crude product as a dark oil, which was purified by flash silica gel
chromatography using a 20% mixture of ethyl acetate and hexane to afford 7ꢀ(2,2ꢀdimethylꢀ1,3ꢀ
dioxolanꢀ4ꢀyl)chromanꢀ4ꢀone (0.657 g, 2.65 mmol, 58% yield) as a clear, colorless oil. HPLC t = 2.14
r
+
min. (Method D); LCMS (ESI) m/z Calcd for C H O [M + H] 249.1. Found: 248.9.
1
4
16
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
-(7-(2,2-dimethyl-1,3-dioxolan-4-yl)-2H-chromen-4-yl)morpholine (18). To a mixture of 7ꢀ
(2,2ꢀdimethylꢀ1,3ꢀdioxolanꢀ4ꢀyl)chromanꢀ4ꢀone (0.657 g, 2.65 mmol) and morpholine (1.15 mL, 13.2
mmol) in toluene (12 mL) at 0 °C was added a 1.0 M solution of titanium(IV) chloride in toluene (1.46
mL, 1.46 mmol) dropwise. The iceꢀbath was removed, and the orange, heterogeneous reaction mixture
was stirred at room temperature overnight. The heterogeneous, orange reaction mixture was filtered
through a pad of Celite and rinsed with toluene (3 x 20 mL). The pale yellow filtrate was concentrated
under reduced pressure to give 4ꢀ(7ꢀ(2,2ꢀdimethylꢀ1,3ꢀdioxolanꢀ4ꢀyl)ꢀ2Hꢀchromenꢀ4ꢀyl)morpholine
(0.751 g, 2.37 mmol, 89% yield) as a yellow oil.
7-(2,2-dimethyl-1,3-dioxolan-4-yl)-3-(3-phenyl-4-(trifluoromethyl)isoxazol-5-yl)-3a,4-
dihydro-3H-chromeno[4,3-c]isoxazol-3-ol (19). To a homogeneous solution of 4ꢀ(7ꢀ(2,2ꢀdimethylꢀ1,3ꢀ
dioxolanꢀ4ꢀyl)ꢀ2Hꢀchromenꢀ4ꢀyl)morpholine (0.266 g, 0.838 mmol) and triethylamine (0.232 mL, 1.68
mmol) in anhydrous acetonitrile (3.0 mL) at 0°C was added a solution of 3ꢀphenylꢀ4ꢀ
16
(
trifluoromethyl)isoxazoleꢀ5ꢀcarbonyl fluoride (0.217 g, 0.838 mmol) in acetonitrile (1.0 mL). The
reaction mixture was stirred at 0 °C for 30 min. and then at room temperature for 60 min. A
homogeneous solution of hydroxylamine hydrochloride (0.233 g, 3.35 mmol) and sodium acetate (0.275
g, 3.35 mmol) in water (0.544 mL, 30.2 mmol) was added, and the reaction mixture was heated at 45 °C
for 60 min. The reaction mixture was diluted with dichloromethane, washed with water, and dried over
anhydrous sodium sulfate. Concentration under reduced pressure followed by purification by flash silica
gel chromatography using a mixture of ethyl acetate in hexane (5% ꢀ 12% ꢀ 20%) afforded 240 mg of
the product mixture. The product was further purified by reverse phase preparative HPLC (2 x 120 mg)
ACS Paragon Plus Environment

Page 27 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
to give, after neutralization, 7ꢀ(2,2ꢀdimethylꢀ1,3ꢀdioxolanꢀ4ꢀyl)ꢀ3ꢀ(3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazolꢀ
5
ꢀyl)ꢀ3a,4ꢀdihydroꢀ3Hꢀchromeno[4,3ꢀc]isoxazolꢀ3ꢀol (0.070 g, 0.139 mmol, 17% yield). HPLC t = 3.08
r
+
min. (Method D); LCMS (ESI) m/z Calcd for C H F N O [M + H] 503.1. Found: 502.9.
2
5
21
3
2
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
7-(2,2-dimethyl-1,3-dioxolan-4-yl)-3-(3-phenyl-4-(trifluoromethyl)isoxazol-5-yl)-4H-
chromeno[4,3-c]isoxazole (20). (This reaction was run twice.) To a stirred suspension of 7ꢀ(2,2ꢀ
dimethylꢀ1,3ꢀdioxolanꢀ4ꢀyl)ꢀ3ꢀ(3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ3a,4ꢀdihydroꢀ3Hꢀ
chromeno[4,3ꢀc]isoxazolꢀ3ꢀol (0.035 g, 0.070 mmol) in anhydrous toluene (1.0 mL) at room
temperature was added pyridine (0.012 mL, 0.153 mmol) followed by thionyl chloride (8.64 µl, 0.118
mmol). The reaction mixture was stirred at room temperature for 15 min. and then at 80 °C for 15 min.
The reaction mixture was cooled to room temperature and concentrated. The solid residue was diluted
with dichloromethane, washed with a saturated aqueous solution of sodium bicarbonate, and dried over
anhydrous sodium sulfate. Concentration under reduced pressure afforded a yellow film. The combined
crude products were triturated with methanol with sonication. The solid was collected to provide 7ꢀ(2,2ꢀ
dimethylꢀ1,3ꢀdioxolanꢀ4ꢀyl)ꢀ3ꢀ(3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ4Hꢀchromeno[4,3ꢀ
c]isoxazole (0.042 g, 0.087 mmol, 62% yield). HPLC t = 3.63 min. (Method D); LCMS (ESI) m/z
r
+
Calcd for C H F N O [M + H] 485.1. Found: 484.9.
2
5
19
3
2
5
1
-(3-(3-phenyl-4-(trifluoromethyl)isoxazol-5-yl)-4H-chromeno[4,3-c]isoxazol-7-yl)ethane-
1
,2-diol (21). A solution of 7ꢀ(2,2ꢀdimethylꢀ1,3ꢀdioxolanꢀ4ꢀyl)ꢀ3ꢀ(3ꢀphenylꢀ4ꢀ(trifluoromethyl)ꢀ
isoxazolꢀ5ꢀyl)ꢀ4Hꢀchromeno[4,3ꢀc]isoxazole (0.042 g, 0.087 mmol) in trifluoroacetic acid (1.00 ml, 13.0
mmol) was left standing at room temperature for 45 min. The trifluoroacetic acid was removed under
reduced pressure, and the residue was diluted with dichloromethane, washed with a saturated aqueous
solution of sodium bicarbonate, and dried over anhydrous sodium sulfate. Concentration under reduced
pressure afforded 1ꢀ(3ꢀ(3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ4Hꢀchromeno[4,3ꢀc]isoxazolꢀ7ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
yl)ethaneꢀ1,2ꢀdiol (0.039 g, 0.088 mmol, 100% yield). HPLC t = 3.00 min. (Method D); LCMS (ESI)
r
+
m/z Calcd for C H F N O [M + H] 445.1. Found: 444.9.
2
2
15
3
2
5
3
-(3-phenyl-4-(trifluoromethyl)isoxazol-5-yl)-4H-chromeno[4,3-c]isoxazole-7-carbaldehyde
(22). To a homogeneous solution of 1ꢀ(3ꢀ(3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ4Hꢀchromeno[4,3ꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
c]isoxazolꢀ7ꢀyl)ethaneꢀ1,2ꢀdiol (0.039 g, 0.088 mmol) in a mixture of tetrahydrofuran (1.0 mL) and
water (0.065 mL) at room temperature was added sodium periodate (0.028 g, 0.132 mmol). The
reaction mixture was stirred for 30 min. The solvent was removed under reduced pressure, and the
residue was diluted with dichloromethane, washed with water, and dried over anhydrous sodium sulfate.
Concentration under reduced pressure afforded 36 mg of a ~1:1 mixture of 3ꢀ(3ꢀphenylꢀ4ꢀ
(trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ4Hꢀchromeno[4,3ꢀc]isoxazoleꢀ7ꢀcarbaldehyde and 7ꢀ(2,2ꢀdimethylꢀ1,3ꢀ
dioxolanꢀ4ꢀyl)ꢀ3ꢀ(3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ4Hꢀchromeno[4,3ꢀc]isoxazole as a yellow
oil. The product was used in the next step without any further purification. HPLC t = 3.35 min.
r
+
(
Method D); LCMS (ESI) m/z Calcd for C H F N O [M + H] 413.1. Found: 412.9.
2
1
11
3
2
4
tert-butyl 1-((3-(3-phenyl-4-(trifluoromethyl)isoxazol-5-yl)-4H-chromeno[4,3-c]isoxazol-7-
yl)methyl)azetidine-3-carboxylate (23). To the mixture of 3ꢀ(3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazolꢀ5ꢀ
yl)ꢀ4Hꢀchromeno[4,3ꢀc]isoxazoleꢀ7ꢀcarbaldehyde (0.036 g, 0.087 mmol) and 7ꢀ(2,2ꢀdimethylꢀ1,3ꢀ
dioxolanꢀ4ꢀyl)ꢀ3ꢀ(3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ4Hꢀchromeno[4,3ꢀc]isoxazole from the
previous reaction in methanol (0.333 mL) and dichloromethane (1.0 mL) at room temperature was added
the tert-butyl azetidineꢀ3ꢀcarboxylate, acetic acid salt (0.028 g, 0.131 mmol), acetic acid (10.00 µl,
0
.175 mmol), and titanium(IV) isopropoxide (0.051 mL, 0.175 mmol) dropwise. The resulting
homogeneous reaction mixture was stirred for 60 min. To the reaction mixture was added sodium
triacetoxyborohydride (0.056 g, 0.262 mmol) in one portion, and the reaction mixture was stirred at
room temperature for 30 min. The reaction was quenched with a saturated aqueous solution of sodium
bicarbonate until the pH was slightly basic (~5 mL). The resulting emulsion was extracted with
ACS Paragon Plus Environment

Page 29 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
dichloromethane (3x). and the organic layers were combined, dried over anhydrous sodium sulfate, and
concentrated to give a pale yellow solid. The crude product was purified by flash silica gel
chromatography using a mixture of ethyl acetate in hexane (20% ꢀ 35% ꢀ 50%) to afford tert-butyl 1ꢀ((3ꢀ
(3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ4Hꢀchromeno[4,3ꢀc]isoxazolꢀ7ꢀyl)methyl)azetidineꢀ3ꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
carboxylate (0.031g, 0.056 mmol, 64% yield). HPLC t = 3.01 min. (Method D); LCMS (ESI) m/z
r
+
Calcd for C H F N O [M + H] 554.1. Found: 553.9.
29 26
3
3
5
1-((3-(3-phenyl-4-(trifluoromethyl)isoxazol-5-yl)-4H-chromeno[4,3-c]isoxazol-7-
yl)methyl)azetidine-3-carboxylic acid (24). A mixture of tertꢀbutyl 1ꢀ((3ꢀ(3ꢀphenylꢀ4ꢀ
(
trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ4Hꢀchromeno[4,3ꢀc]isoxazolꢀ7ꢀyl)methyl)azetidineꢀ3ꢀcarboxylate (0.031
g, 0.056 mmol) and trifluoroacetic acid (2.58 ml, 33.5 mmol) was left standing at room temperature for
5 min. The trifluoroacetic acid was removed under reduced pressure, and the residue was suspended in
4
water, the pH was adjusted to 5 with sonication, and the resulting suspension was stirred for 1 h. The
pH remained at 5, so the solid was collected by vacuum filtration, washed with water, and dried under
reduced pressure to give a quantitative yield of the product. The compound was suspended in methanol
and sonicated for 15 min. The product was collected by vacuum filtration, washed with methanol, and
dried under reduced pressure to provide 1ꢀ((3ꢀ(3ꢀphenylꢀ4ꢀ(trifluoromethyl)isoxazolꢀ5ꢀyl)ꢀ4Hꢀ
chromeno[4,3ꢀc]isoxazolꢀ7ꢀyl)methyl)azetidineꢀ3ꢀcarboxylic acid (0.029 g, 0.058 mmol, 100% yield).
+
HPLC t = 7.51 min. (Method A); LCMS (ESI) m/z Calcd for C H F N O [M + H] 498.1. Found:
r
25 18
3
3
5
1
4
98.0. HꢀNMR (500 MHz, DMSOꢀd ) δ ppm 7.79 (d, J=7.77 Hz, 1H), 7.59 ꢀ 7.69 (m, 5H), 7.09 (d,
6
J=7.77 Hz, 1H), 7.04 (s, 1H), 5.61 (s, 2H), 3.58 (s, 2H), and 3.20 ꢀ 3.46 (m, 5H).
(5-Phenyl-4-(trifluoromethyl)isoxazol-3-yl)methanol (26). To a solution of 5ꢀphenylꢀ4ꢀ
(trifluoromethyl)isoxazoleꢀ3ꢀcarboxylic acid (6.0 g, 23.33 mmol) in tetrahydrofuran (40 mL) at 0°C was
added Nꢀmethylmorpholine (3.85 mL, 35.0 mmol) followed by isobutyl chloroformate (4.60 mL, 35.0
mmol) via syringe over 5 min. The resulting suspension was stirred at 0°C for 15 min. The suspension
ACS Paragon Plus Environment