Asymmetric Synthesis of the Major Metabolite of a Calcitonin Gene-Related Peptide Receptor Antagonist and Mechanism of Epoxide Hydrogenolysis

Article abstract of DOI:10.1021/acs.joc.7b00052

An asymmetric synthesis of the major metabolite of the calcitonin gene-related peptide recepotor antagonist BMS-846372 is presented. The variously substituted cyclohepta[b]pyridine ring system represents an underexplored ring system and showed some unexpe

Full text of DOI:10.1021/acs.joc.7b00052

Subscriber access provided by HACETTEPE UNIVERSITESI KUTUPHANESI
Article
Asymmetric Synthesis of the Major Metabolite of a CGRP
Receptor Antagonist and Mechanism of Epoxide Hydrogenolysis
Guanglin Luo, Ling Chen, Charles M. Conway, Walter Kostich, Benjamin
M. Johnson, Alicia Ng, John E Macor, and Gene M. Dubowchik
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.7b00052 • Publication Date (Web): 17 Mar 2017
Downloaded from http://pubs.acs.org on March 17, 2017
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.
The Journal of Organic 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 37
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Asymmetric Synthesis of the Major Metabolite of a CGRP Receptor
Antagonist and Mechanism of Epoxide Hydrogenolysis
Guanglin Luo,* Ling Chen, Charles M. Conway, Walter Kostich, Benjamin M. Johnson, Alicia
Ng, John E. Macor, and Gene M. Dubowchik
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
BristolꢀMyers Squibb Research & Development, BristolꢀMyers Squibb Company, 5 Research Parkway,
Wallingford, CT 06443, USA.
guanglin.luo@bms.com
RECEIVED DATE (to be automatically inserted after your manuscript is accepted if required
according to the journal that you are submitting your paper to)
*Corresponding author. Tel: +12036776640. Fax: +12036777702.
ABSTRACT
An asymmetric synthesis of the major metabolite of the CGRP recepotor antagonist BMSꢀ
846372 is presented. The variously substituted cyclohepta[b]pyridine ring system represents an underꢀ
explored ring system, and showed some unexpected chemistry. Reactivities of epoxide and ketone
1
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 37
1
2
3
4
5
6
7
8
9
functional groups on cycloheptane ring were greatly controlled by a remote bulky TIPS group. The rate
difference of the hydrogenolysis between two diastereomeric epoxide intermediates shed some light on
the mechanism of epoxide hydrogenolysis, and further deuterium labeling studies unravelled more
mechanistic details on this well known chemical transformation for the first time.
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
INTRODUCTION
Migraine is a very common and disabling medical illness that affects approximately 12% of the
adult population.1 The current standard of care for migraneurs is the use of the triptan class of drugs,
which are vasoconstrictors and suffer from mechanismꢀbased cardiovascular risks.2 Calcitonin geneꢀ
related peptide (CGRP), a 37ꢀamino acid peptide, is closely associated with the pathophysiology of
migraine.3 Small molecule inhibitors of the CGRP receptor have been clinically shown to be effective
and safe treatment for people who suffer from migraine.4 In contrast to the triptans, CGRP receptor
antagonists should inhibit vasodilation, rather than directly constrict dilated arteries, thereby
circumventing cardiovascular risks.5 Effective treatment of migraine by CGRP receptor antagonists have
been clinically proven by numerous small molecules, but so far none has reached the market.5,6 In one of
our medicinal chemistry programs directed at discovering novel CGRP receptor antagonists for the
treatment of migraine, BMSꢀ846372 (Figure 1) emerged as a potential clinical candidate.7 An
enantioselective synthesis of BMSꢀ846372 has recently been reported.8 During preclinical studies,
compound 1 (Figure 1) was proposed as the major metabolite of BMSꢀ846372 in various animal species
(data not shown). A synthesis of 1 would not only verify the structure of the major metabolite of BMSꢀ
846372, but could also determine if it were an active metabolite, which with the added ꢀOH group,
might have increased aqueous solubility. This was the key issue that complicated further development
of BMSꢀ846372.
2
ACS Paragon Plus Environment

Page 3 of 37
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Figure 1. CGRP antagonist BMSꢀ846372 and its major metabolite 1.
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
RESULTS AND DISCUSSION
A brief retrosynthetic analysis of 1 led to the diol 2, from which selective carbamate formation at
the secondary alcohol should be straightforward (Figure 2). The tertiary alcohol could come from the
key cyclohepta[b]pyridine hydroxyl ketone intermediate 3 by a reaction with an appropriate aryl anion.
By choosing a bulky TIPS protecting group, an aryl lithium anion might attack the ketone group from
the opposite face to achieve the desired relative stereochemistry as shown.8 Furthermore, we envisioned
that the desired ketone 3 could be transposed from the chiral ketone 58 via an epoxide intermediate 4
(Figure 2).
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 37
1
2
3
4
5
6
7
8
9
F
F
Li
HO
O
F
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
1
F
N
N
HO
TIPSO
2
3
O
O
N
N
TIPSO
TIPSO
4
5
Figure 2. Retrosynthetic analysis of 1.
To test the reactions of epoxide formation and opening, and the aryl lithium addition to the ketone in
which the enolizable ketone might be an issue, we pursued a model reaction before starting with the
chiral material. As a simpler version of 3, compound 10 was prepared as shown in Scheme 1. Starting
with the commercially available aldehyde 6, Wittig reaction with methyltriphenylphosphonium bromide
treated with nꢀBuLi afforded vinyl pyridine 7 in 89% yield.9 A Negishi reaction of 7 with commercially
available 4ꢀpentenyl zinc bromide provided the diene 8 in 65% yield. The free base 8 failed RCM, but
with its in situ prepared HCl salt,10 diene 8 cyclized smoothly under GrubbsꢀII catalysis conditions to
afford the cycloheptene 9 in excellent yield. Compound 9 was converted to the desired ketone 10 in a
straightforward 4ꢀstep sequence in 86% overall yield requiring no chromatographic purification. That is,
the double bond was first converted to the bromohydrin intermediate, followed by MeONa treatment to
form the epoxide. The epoxide was regioselectively opened by hydrogenolysis to generate the alcohol
intermediate, which was oxidized by Swern oxidation to 10. Hydrogenolysis was exceptionally easy
4
ACS Paragon Plus Environment

Page 5 of 37
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
(Pd/C with 1 atm. H2, 2 h), and each of the four steps was essentially spotꢀtoꢀspot conversion by TLC
analysis. With 10 available, we tested the aryl addition reaction. Indeed, the enolizable ketone posed a
problem, and the reaction stalled after some conversion. Nonetheless the tertiary alcohol 11 was
obtained in 21% yield with 38% recovered ketone.
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
Scheme 1. Synthesis of model compound 11
O
4-pentenylzinc bromide
Pd(PPh3)4,THF
Ph3PMeBr, n-BuLi, THF
H
0 °C to rt, 72 h
89%
70 °C, 17 h
65%
N
N
Br
N
Br
8
7
6
AcO
Br
i) HCl (5 equiv), ether
ii) GrubbsII (5 mol%), CH2Cl2
N-bromoacetamide, AcOLi, AcOH
rt, 16 h
N
N
40 °C, 5h
92%
9
oxalyl chloride, DMSO,
CH2Cl2, Et3N
-55 °C - -50 °C, 30 min
O
OH
MeONa, THF
Pd/C, H2 (1 atm), MeOH
rt, 2 h
N
86% for 4 steps
N
rt, 2 h
F
F
1-bromo-2,3-difluorobenzene
n-BuLi, THF
HO
O
-78 °C to rt, 1 h,
21% (with 38% 10 recovered)
N
N
11
10
The first synthesis of the transposed ketone compound 3 was accomplished as shown in Scheme
2. The TMS enol ether was obtained by treating the ketone 58 with LiHMDS followed by quenching
with TMSCl. After brief workup and with no purification, treatment of the TMS enol ether intermediate
with NBS afforded the bromoꢀketone 12 as a single diastereomer in 84% yield. It is conceivable that
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 37
1
2
3
4
5
6
7
8
9
NBS reacted from the side opposite to the bulky TIPS protected alcohol. Stereospecific reduction of the
ketone group of 12 by superhydride (LiEt3BH) also occurred from the opposite face of the TIPSO group
to afford stereospecifically the bromohydrin 13, which upon treatment with NaH gave the epoxide 14 in
excellent yield. However, it was very surprising and interesting to find out that this cisꢀepoxide was very
much resistant to typical hydrogenolysis conditions, especially as compared to the hydrogenation
reaction shown in Scheme 1. Only 8.5% desired alcohol 15 was obtained, even with prolonged reaction
time (18 h, with partial pyridine ring reduction and 36% recovered 14). The alcohol 15 was then
converted to the desired ketone 3 under Swern oxidation condition in good yield.
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
Scheme 2. Synthesis of the intermediate ketone 3
Intrigued by the cause of very low reactivity of the chiral epoxide 14 towards hydrogenolysis, we
intended to prepare the diasteromer of 14, as shown in Scheme 3 for comparison. Nonselective
6
ACS Paragon Plus Environment

Page 7 of 37
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
reduction of 5 by NaBH4, followed by dehydration with Burgess’ reagent ((Methoxycarbonylsulfamoyl)ꢀ
triethylammonium hydroxide, inner salt) afforded the chiral cycloheptene derivative 16 in 56% yield.
Stereoselective epoxidation of 16 under Jacobsen’s conditions with (1S, 2S)ꢀ(+)ꢀ[1,2ꢀ
cyclohexanediaminoꢀN,N'ꢀbis(3,5ꢀdiꢀtꢀbutylsalicylidene)] manganese (III) chloride,11 as a match case,
afforded 17 in 58% yield. Gratifyingly hydrogenolysis of 17 went smoothly to 18 (diastereomer of 15
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
1
with distinctive H NMR) in quantitative yield. Similar to 15, alcohol 18 was converted to the desired
ketone 3 by Swern oxidation in good yield. To definitively determine the relative stereochemistry of
17/18 and 14/15, an xꢀray structure of 18 was obtained.12
Scheme 3. Synthesis of the epoxide 17 and its hydrogenolysis
The drastically different rates of hydrogenolysis of the diastereomeric epoxides 14 and 17 helped
to explain some experimental results that had puzzled us. In a previous racemic synthesis of BMSꢀ
846372, 16 served as an important intermediate,7 which, following the 4ꢀstep reaction sequence as
shown in Scheme 1, afforded racemic 3 in only 18% yield (Scheme 4). The reason for the poor
conversion is clear now, as the sequence likely generated more cisꢀepoxide 14 (as a racemate) than
transꢀepoxide 17. The yield was greatly improved by using racemic Jacobsen’s catalysts (equal mixture
of the commercial (1S, 2S)ꢀ and (1R, 2R)ꢀ(±)ꢀ[1,2ꢀcyclohexanediaminoꢀN,N'ꢀbis(3,5ꢀdiꢀtꢀ
7
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 37
1
2
3
4
5
6
butylsalicylidene)]manganese (III) chloride) but not the chiral form).7 It is apparent now that the transꢀ
epoxide 17 (as a racemate) was mostly formed in the latter case (Scheme 4).
7
8
9
Scheme 4. Synthesis and reactivity of diastereomeric/racemic epoxides
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
Reductive opening of epoxide by hydrogen catalyzed by palladium on carbon is an important reaction
in organic chemistry,13 but its mechanism remains underꢀinvestigated. On the other hand, catalytic
hydrogenation of alkenes by Pd/C had been well studied.14 Presumably the oxygen atom of the epoxide
is absorbed onto the catalyst surface with either of its two electron pairs and a nearby Pdꢀbound hydride
attacks the weaker CꢀO bond to open the epoxide ring (Figure 3). For 17, binding of oxygen and attack
of hydride pose no problem as the oxygen is unencumbered on both sides of the epoxy ring similar to
the simpler epoxide in Scheme 1. For epoxide 14, however, binding of the oxygen to the catalyst must
occur opposite to the bulky TIPS group, but this should pose no problem and be even more facile than
17 if the hydride comes from nearby same site. The fact that 14 reacted much more slowly might
indicate that binding and delivery of hydride occur from different sites and the hydride was delivered
8
ACS Paragon Plus Environment

Page 9 of 37
The Journal of Organic Chemistry
1
2
3
4
5
from the side opposite that where PdꢀO binding occurs, and which would be significantly hampered by
the presence of bulky TIPS group.
6
7
8
9
Pd
H
X
O
H
O
Pd-H
N
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
Pd
H
H
H
N
O
O
Si
Si
17
14
Figure 3. Hydrogenolysis of diastereomeric epoxides 14 and 17. The ballꢀstick structures were
generated by ChemBio3D Ultra without energy minimization and the electron long pairs were depicted
as the pink balls.
To further investigate the epoxide hydrogenolysis mechanism, deuterium labeling studies were
carried out. Both epoxides 14 and 17 were treated with deuterium gas over Pd/C as before. For epoxide
17, reaction was complete in 2 h and deuterium was incoported almost exclusively in an SN2 fashion,
clearly indicated by the H NMR coupling patterns of 18 and Dꢀ18 (Figures 4).15 On the other hand,
1
reaction of 14 was again extremely slow (requiring 18 h) and overꢀreduction of the pyridine was a
competing reaction. Nontheless, desired product D-15 was obtained. Deuterium incorporation in D-15
was approximately 60%, but also occurred stereospecifically in an SN2 fashion as indicated by 1H NMR
analysis (Figures 4).15 Careful purification of the hydrogenolysis products of 14 also provided the other
isomeric alcohol product 19 (Figure 5).16 Originally we thought that 19 was formed as the completing
9
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 37
1
2
3
4
hydrogenolysis of the other CꢀO bond. In the deuteration reaction of 14, we purified 19 as well.
5
6
7
1
Surprisingly there was no incorporation of deuterium in 19 as shown by H NMR analysis.12 It is
8
9
reasonable to propose that some Pd itself may serve simply as a reducing reagent to the epoxide 14 as a
slower competing pathway, which after protonolysis by MeOH generates 19 without deuterium
incorporation (Figures 4). This pathway likely also generated the other 40% of 15, which had no
deuterium incorporation. With regard to epoxide hydrogenolysis, these experiments clearly indicated an
SN2 mechanism, as the hydride/deuteride must approach from the opposite face of the broken CꢀO bond.
As in the case of 14, the bulky TIPS group potentially forced catalyst adsorption chelation onto the
opposite side, sterically blocking potential SN2 reaction. But with the higher energy transition states
where 14 was adsorbed onto the more hindered TIPSO face, SN2 hydrogenolysis occurred, to generate
stereospecific D-15. In the case of 17, both sides of the epoxide were open for adsorption but SN2
hydrogenolysis proceeded rapidly and productively. In a previous publication regarding hydrogenolysis
of aziridines, the stereochemical outcome appeared to be solventꢀdependent.17 In our case, runing the
reaction of 14 in MeOH or EtOH did not affect the reaction rates. In ethyl acetate the conversion was
even lower (3.6% yield) for the same amount of time, but cleaner with mainly starting material
remaining. Deuterium labeling study of 14 was only carried out in MeOH. With regard to 17, reaction
rate was also significantly slower in EtOAc. Deuterium reaction of 17 was also carried out in EtOAc.
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
1
For the 2 h reaction time, there was only around 1/3 conversion and 35% of D-18 was obtained. H
NMR indicated that the deuterium labeling outcome was the same as in MeOH.
10
ACS Paragon Plus Environment

Page 11 of 37
The Journal of Organic Chemistry
1
2
3
4
5
H
>95%
OH
H
D
6
7
8
17
N
9
TIPSO
D-18
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
~60%
D
H
OH
OH
H
H
H
H
14
N
N
TIPSO
D-15/15
TIPSO
19
MeOH
MeOH
*Pd
O
O
H
Pd*
H
H
H
N
N
TIPSO
Pd as reducing reagent
TIPSO
Figure 4. Stereospecific incoporation of deuterium and proposed competing mechanism for 14.
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
With ketone 3 in hand, aryl addition was performed as shown in Scheme 5. In the presence of the
bulky TIPS group, enolization of ketone 3 may have been prevented, and a single diastereomerically
pure alcohol 20 was formed in 82% yield. The TIPS group was removed by treatment with TBAF at
elevated temperature overnight to afford the diol 21 in good yield. Following previously reported
conditions,7 selective carbamate formation at the secondary alcohol was achieved in good yield.
However this product did not match the analytical data of the major metabolite of BMSꢀ846372.
Furthermore this product was inactive in a hCGRP binding assay. This compound was believed to be the
ꢀOH diastereomer of 1 as drawn in Scheme 5.
Scheme 5. Synthesis of diastereomeric 1
11
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 37
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
It was surprising that the aryllithium attacked the ketone exclusively from the seemingly more
crowded face occupied by the bulky TIPS group (TIPS is likely too bulky to allow Oꢀmediated reaction)
as compared with the formation of bromide 12 and the arylation of a previously reported isomeric αꢀ
ketone8. Simple modeling of compound 3 could not easily explain why the aryllithium species would
attack exclusively from Si face, but probably the ketone group adopted a conformation with Si face fully
exposed as shown by a simple ballꢀstick structure generated by ChemDraw3D (Figure 5).
12
ACS Paragon Plus Environment

Page 13 of 37
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Figure 5. Si face attack on compound 3 by 2,3ꢀdifluorophenyllithium. The ballꢀstick structure of 3 was
generated by ChemBio3D Ultra without energy minimization and Si face of the ketone was fully
exposed.
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
The absolute stereochemistry of the tertiary alcohol was determined by the stereochemistry of the
TIPS protected alcohol, albeit not the isomer/metabolite desired. Thus, if we started with the enantiomer
of 5, we should be able to set the desired correct absolute stereochemistry of the tertiary alcohol for
metabolite 1, and then correct the stereochmistry of the secondary alcohol by a simple Mitsunobu
inversion. Stereospecific synthesis of 1 was finally achieved as shown in Scheme 6. Following the
synthesis described for 17, the (S)ꢀchiral hydroxyl ketone18 was converted to entꢀ17 after the alcohol
group was protected with TIPS, and epoxidized with the matched (1R, 2R)ꢀJacobsen’s catalyst in the
epoxidation step. Following previously described conditions, entꢀ3 was obtained in good yield.
Aryllithium addition proceeded smoothly with the lithium species generated directly from 1,2ꢀ
difluorobenzene.19 After TBAF deprotection, entꢀ21 was obtained in good yield. Mitsunobu reaction of
entꢀ21 with 4ꢀnitrobenzoic acid afforded the desired 4ꢀnitrobenzoate in 18%. The major side product
was the bridged ether (not unexpected from the relative stereochemistry of the diol entꢀ21), and some
dehydration products were obtained as well. After hydrolysis of the 4ꢀnitrobenzoate by LiOH, diol 2 was
obtained in 86% yield. An Xꢀray structure of 2 was obtained,12 which proved proof of the relative
stereochemistry of all previous intermediates and the final products. Finally, carbamate formation went
smoothly to afford the desired product 1. This product matches in every respect the major metabolite of
BMSꢀ846372 and demonstrated hCGRP receptor binding of 1.4 nM, 20ꢀfold less potent than the parent
compound (BMSꢀ846372: hCGRP receptor Ki = 0.070 nM). More interestingly, the aqueous solubility
was greatly improved to 90 µg/ml for 1 (compared with < 2 µg/ml for BMSꢀ846372).7
13
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 37
1
2
3
4
Scheme 6. Synthesis of 1
5
6
7
8
9
O
i) TIPSOTf, CH2Cl2, Et3N, 0 °C, 1 h
ii) NaBH4, MeOH, rt, 1 h
iii) Burgess's reagent, benzene, reflux, 5 h
O
i) Pd/C, H2 (1 atm), MeOH
rt, 2 h, 97%
N
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
ii) Oxalyl chloride, DMSO,
CH2Cl2, Et3N
-55 °C - -50 °C, 30 min, 72%
iv) (1R,2R)-(-)-Jacobsen's catalysis (12 mol%),
NaOCl, Na2HPO4, CH2Cl2
N
(S)
TIPSO
HO
0 °C to rt, 17 h, 50% for 4 steps
ent-17
OH
OH
O
TBAF, THF, 50 °C
18 h, 72%
1,2-Difluorobenzene
F
F
n-BuLi, THF
F
F
N
N
N
-65 °C to -78 °C to rt
2 h, 79%
TIPSO
ent-3
TIPSO
HO
ent-21
ent-20
F
HO
F
i) 4-nitrobenzoic acid
Ph3P, DIAD
Last step of
scheme 5
N
O
OH
THF, 15 h, 18%
O
N
NH
N
N
F
F
O
ii) LiOH, THF
rt, 2 h, 86%
N
HO
1
2
hCGRP Ki = 1.4 nM
Aq. Sol.: 90 g/ml
In conclusion, a stereospecific synthesis of a potential metabolite of BMSꢀ846372 was
accomplished. Some interesting, and unexpected chemistry around the underꢀexplored
cyclohepta[b]pyridine ring system was discovered. Two diastereomeric epoxides were excellent
substrates for studying the epoxide hydrogenolysis mechanism, and deutorium labeling studies clearly
indicated an SN2 mechanism and necessary separation of the adsorption site and reaction site. Even
though compound 1 did show only nanomolar potency (1.4 nM – 20ꢀfold less potent than its’ parent)
against the CGRP receptor, installation of the ꢀOH group improved intrinsic solubility and led to further
modifications of BMSꢀ846372, ultimately leading to the discovery an improved CGRP receptor
antagonist that has progressed into clinical development. 20
14
ACS Paragon Plus Environment

Page 15 of 37
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
Experimental Section
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
1
13
General information. H and C NMR spectra were recorded on a 400 MHz or 500 MHz Bruker
1
13
spectrometer. In H and C NMR spectra, chemical shifts were expressed in δ (ppm) downfield from
CHCl3 (7.26 ppm) and CDCl3 (77.0 ppm), respectively. Mass spectra were recorded by electron impact
or chemical ionization. High resolution mass spectrometry (HRMS) analyses were performed on a
Fourier Transform Orbitrap mass spectrometer (Exactive, Thermo Fisher Scientific, San Jose, CA) in
positive or negative ionization electrospray mode operating at 25,000 resolution (full width at half
height maximum, FWHM). The instrument was daily calibrated according to manufacturer's
specifications resulting in mass accuracy of or better than 5 ppm. The operating software Xcalibur was
used to calculate theoretical massꢀtoꢀcharge values and to process the obtained data. Column
chromatography was performed on silica gel packed in Biotage Horizon columns. 2ꢀ
Bromonicotinaldehyde (6) was commerically available from Aldrich and was directly used. Deuterium
gas was purchased from Aldrich with 99.96% purity.
2-Bromo-3-vinylpyridine (7).9 Butyllithium (22.75 mL, 59.1 mmol) was added to the THF (450 mL)
suspension of methyltriphenylphosphonium bromide (21.13 g, 59.1 mmol) at 0 °C. The solution turned
to orange and the reaction was lift to room temperature for 30 min before cooled it back to 0 °C. 2ꢀ
bromonicotinaldehyde (10 g, 53.8 mmol) in 50 mL THF was added through canula to the reaction
solution. The precipitate was formed and the reaction was lift to room temperature. The color of the
reaction turned to green. After a while, the color of te reaction became orange again. The reaction was
stirred at room temperature over weekend. The solvent was removed mostly via vacuum and the crude
was partitioned between water and diethyl ether. The organic layer was separated and the aqueous layer
was extract twice with diethyl ether. The diethyl ether layer was combined, dried (Na2SO4), filtered and
15
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 37
1
2
3
4
concentrated. The product was obtained by flash column eluted with ethyl acetate in hexane (10%) as
5
6
7
1
yellow oil (8.78 g, 89%): MS(ESI) [M+H+] = 184.04; H NMR δ ppm (400 MHz, CHLOROFORMꢀd)
8
9
8.21 ꢀ 8.29 (m, 1 H), 7.78 (dd, J = 7.68, 1.89 Hz, 1 H), 7.20 ꢀ 7.28 (m, 1 H), 6.96 (dd, J = 17.37, 11.08
Hz, 1 H), 5.72 (d, J = 17.37 Hz, 1 H), 5.46 (d, J = 11.08 Hz, 1 H). The proton NMR data matached that
of the literature report.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
2-(Pent-4-enyl)-3-vinylpyridine (8).
In a 500 mL roundꢀbottomed flask was 2ꢀbromoꢀ3ꢀ
vinylpyridine (4.151 g, 22.56 mmol) (azeotroped with dry benzene) in THF (40 mL) to give a colorless
solution. Pd(Ph3P)4 (0.782 g, 0.677 mmol) was added under nitrogen. While stirring under nitrogen, 4ꢀ
pentenylzinc bromide (46 mL, 23.00 mmol) was added via syringe, and the resulted dark mixture was
stirred at rt for 5 min. It was then heated at reflux (70 °C) overnight. After 17 h, LCMS indicated a
complete conversion to the desired product. THF was stripped off. The reaction was quenched with
NH4Cl solution and diluted with EtOAc. The layers were separated and the organic layer was washed
with brine, dried and concentrated to a yellow oil. FCC up to 30% EtOAc/hexane afforded the desired
1
product as a colorless oil (2.54 g, 65%): MS(ESI) [M+H+] = 174.10; H NMR (400 MHz,
CHLOROFORMꢀd) δ ppm 8.39 (dd, J = 4.78, 1.51 Hz, 1 H), 7.69 (dd, J = 7.81, 1.76 Hz, 1 H), 7.07 (dd,
J = 7.81, 4.78 Hz, 1 H), 6.89 (dd, J = 17.37, 11.08 Hz, 1 H), 5.80 (dddd, J = 17.06, 10.26, 6.67, 6.55 Hz,
1 H), 5.62 (d, J = 17.37 Hz, 1 H), 5.34 (d, J = 11.08 Hz, 1 H), 4.85 ꢀ 5.09 (m, 2 H), 2.76 ꢀ 2.92 (m, 2 H),
2.11 (q, J = 7.13 Hz, 2 H), 1.67 ꢀ 1.83 (m, 2 H); 13C NMR (101 MHz, CHLOROFORMꢀd) δ ppm 159.2,
148.3, 138.4, 133.2, 133.2, 131.7, 121.4, 117.2, 114.9, 35.0, 33.6, 28.6.
(Z)-8,9-Dihydro-7H-cyclohepta[b]pyridine (9). In a 2 L roundꢀbottomed flask was 2ꢀ(pentꢀ4ꢀenyl)ꢀ
3ꢀvinylpyridine (2.1 g, 12.12 mmol) in ether (4 mL) to give a colorless solution. HCl (30 mL, 60.0
mmol) was added, and the mixture was stirred for 5 min. The volatiles were evaporated to give a
colorless oil, which was then azeotroped with dry benzene to a white solid. It was then dissolved in
CH2Cl2 (1 L) (degassed with argon) to give a colorless solution. GrubbsII (0.515 g, 0.606 mmol) was
16
ACS Paragon Plus Environment

Page 17 of 37
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
added, and the mixture was heated at a 40 °C oil bath with stirring under nitrogen for 5 h. LCMS
indicated complete conversion. The mixture was concentrated to a tan oil. It was dissolved EtOAc and
washed with saturated NaHCO3 solution, brine, dried with Na2SO4, and concentrated to a tan oil. FCC
up to 40% EtOAc/hexane afforded the desired product as a tan oil (1.79 g, 92%): MS(ESI) [M+H+] =
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
1
146.06; H NMR (400 MHz, CHLOROFORMꢀd) δ ppm 8.20 (d, J = 4.78 Hz, 1 H), 7.32 (d, J = 7.55
Hz, 1 H), 6.99 (dd, J = 7.55, 5.04 Hz, 1 H), 6.21 (dt, J = 12.28, 2.05 Hz, 1 H), 5.90 (dt, J = 12.34, 4.41
Hz, 1 H), 2.93 ꢀ 3.06 (m, 2 H), 2.30 ꢀ 2.49 (m, 2 H), 1.82 ꢀ 2.02 (m, 2 H); 13C NMR (101 MHz,
CHLOROFORMꢀd) δ ppm 160.6, 146.3, 137.8, 134.2, 131.3, 127.2, 121.2, 39.1, 32.5, 24.9.
8,9-Dihydro-5H-cyclohepta[b]pyridin-6(7H)-one (10). This compound was prepared through 4ꢀstep
telescoped reactions. 1. 6ꢀBromoꢀ6,7,8,9ꢀtetrahydroꢀ5Hꢀcyclohepta[b]pyridinꢀ5ꢀyl acetate (unpurified
intermediate): In a 500 mL roundꢀbottomed flask was intermediate 9 (5.16 g, 35.5 mmol), lithium
acetate (9.38 g, 142 mmol) in acetic acid (100 mL) to give a tan suspension under nitrogen. Nꢀ
bromoacetamide (5.00 g, 36.2 mmol) was added. The flask was wrapped with aluminum foil and the
mixture was stirred at rt for 16 h. There was no solids left and LCMS showed complete conversion to
the desired more polar product as a major peak. AcOH was stripped off under high vacuum. The
residue was diluted with water and EtOAc. Na2CO3 was added to neutralize the mixture till no gas
evolved. The layers were separated and the aqueous layer was extracted with EtOAc. The combined
organic layers were washed with brine, dried with Na2SO4, and concentrated to a dense tan oil (10.5 g,
100%): MS(ESI) [M+H+] = 284.17. The crude was used as it was in the next step.
2.
6ꢀBromoꢀ6,7,8,9ꢀtetrahydroꢀ5Hꢀcyclohepta[b]pyridinꢀ5ꢀyl
acetate
epoxide
(unpurified
intermediate): In a 500 mL roundꢀbottomed flask was previous crude intermediate (10.09 g, 35.5 mmol)
(azeotroped with dry benzene) in THF (100 mL) to give a tan solution. Sodium methoxide (9.59 g, 178
mmol) was added, and the mixture was stirred at rt under nitrogen for 2 h. TLC showed complete
conversion to the more polar product spot. THF was stripped off and the residue was partitioned
17
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 37
1
2
3
4
5
6
7
8
9
between water and EtOAc. The aqueous layer was extracted with EtOAc. The combined organic layers
were washed with brine, dried with Na2SO4, and concentrated to a tan oil (5.72 g, 100%): MS(ESI)
[M+H+] = 162.21, which was directly used in the next reaction.
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
3. 6,7,8,9ꢀTetrahydroꢀ5Hꢀcyclohepta[b]pyridinꢀ6ꢀol (unpurified intermediate): In a 500 mL roundꢀ
bottomed flask was the previous epoxide (5.72 g, 35.5 mmol) and Pd/C (1.889 g, 1.775 mmol) in MeOH
(100 mL) to give a black suspension. It was stirred under 1 atm hydrogen (balloon) for 2 h. It was
filtered and concentrated to a tan oil (6 g, 100%). It was used in the next reaction without further
purification and characterizations.
4. In an ovenꢀdried 500 mL roundꢀbottomed flask was oxalyl chloride (3.42 mL, 39.1 mmol) in
CH2Cl2 (100 mL) to give a colorless solution at ꢀ55 °C under nitrogen. DMSO (5.54 mL, 78 mmol) was
added dropwise over 10min. After the solution was stirred for an additional 30 min, previous crude
intermediate (5.79 g, 35.5 mmol) (azeotroped with dry benzene) dissolved in 20 mL CH2Cl2 (plus 20
mL rinse) was added via canuula over 5 min. The reaction mixture was stirred at ꢀ50 ꢀ ꢀ55 °C for an
additional 40 min (the solution became milky). Et3N (24.74 mL, 178 mmol) was added via syringe at ꢀ
50 °C and the reaction mixture was stirred for 30 min. Water (100 mL) was added, and the layers were
separated. The aqueous layer was extracted with CH2Cl2 (2x100 ml). The combined organic layers
were dried with Na2SO4, and concentrated to a tan oil with some solids. Purification by FCC up to 10%
1
MeOH/CH2Cl2 afforded the desired product as an orange oil (4.947 g, 86% for 4 steps): H NMR (500
MHz, CHLOROFORMꢀd) δ ppm 8.33 (dd, J = 4.3, 2.2 Hz, 1H), 7.37 (dd, J = 7.6, 1.7 Hz, 1H), 7.05
(ddd, J = 6.9, 4.9, 1.6 Hz, 1H), 3.65 (d, J = 1.5 Hz, 2H), 3.19 – 3.10 (m, 2H), 2.54 (dd, J = 3.4, 1.7 Hz,
2H), 2.01 (pd, J = 6.6, 1.7 Hz, 2H); 13C NMR (126 MHz, CHLOROFORMꢀd) δ ppm 207.2, 160.2,
148.0, 137.2, 128.8, 122.2, 48.9, 43.9, 36.1, 24.7.
6-(2,3-Difluorophenyl)-6,7,8,9-tetrahydro-5H-cyclohepta[b]pyridin-6-ol (11). In an ovenꢀdried
500 mL roundꢀbottomed flask was BuLi (17.19 mL, 43.0 mmol) in THF (100 mL) to give a colorless
18
ACS Paragon Plus Environment

Page 19 of 37
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
solution at ꢀ78 ºC under nitrogen. 1ꢀBromoꢀ2,3ꢀdifluorobenzene (4.81 mL, 43.0 mmol) was added
dropwise via syringe. The mixture was stirred at ꢀ78 ºC for 20 min, and intermediate 10 (4.947 g, 30.7
mmol) (azeotroped with dry benzene and dried under high vac) dissolved in 10 mL THF was added
dropwise via canuula (plus 10 mL THF rinse). The mixture was warmed up to rt in 1 h. TLC showed
some conversion to a slightly more polar spot. After quenched with saturated NH4Cl solution, THF was
stripped off. The remaining mixture was partitioned between water and EtOAc. The layers were
separated and the organic layer was washed with brine, dried with Na2SO4, and concentrated to a dark
oil. The residue was purified by FCC up to 10% MeOH/CH2Cl2. The impure fractions were pooled and
purified by FCC with EtOAc/CH2Cl2 up to pure EtOAc. Some starting material was recovered (1.88 g,
38%). The product fractions were pooled and concentrated to a tan solid, which were washed repeatedly
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
1
with Et2O to a tan solid (1.79 g, 21%): H NMR (400 MHz, CHLOROFORMꢀd) δ ppm 8.27 (dd, J =
4.9, 1.8 Hz, 1H), 7.46 – 7.38 (m, 1H), 7.34 (dd, J = 7.5, 1.7 Hz, 1H), 7.12 – 6.99 (m, 3H), 3.93 (dd, J =
14.4, 2.7 Hz, 1H), 3.18 – 3.04 (m, 2H), 2.86 (dd, J = 14.4, 2.0 Hz, 1H), 2.68 – 2.52 (m, 1H), 2.50 – 2.39
(m, 1H), 1.94 – 1.84 (m, 1H), 1.83 – 1.68 (m, 1H); HRMS (ESI), m/z calcd for C16H16NOF2 [M+H]+
276.1197, found 276.1185.
(6R,9R)-6-Bromo-9-(triisopropylsilyloxy)-6,7,8,9-tetrahydro-5H-cyclohepta[b]pyridin-5-one
(12). In a 100 mL roundꢀbottomed flask was compound 5 (2.01g, 6.03 mmol) in THF (20 mL) to give a
colorless solution. After cooling to ꢀ78 °C, LiHMDS (6.63 mL, 6.63 mmol) (1.0M in THF) was added
dropwise via syringe. After stirring for 60min at ꢀ78 °C, TMSꢀCl (1.078 mL, 8.44 mmol) was added.
The stirring was continued for 1h while bath temperature gradually warmed up to ꢀ20 °C. TLC (4/1
hexane/EtOAc) showed complete conversion to a less polar spot. The reaction was quenched with
saturated NaHCO3 solution and diluted with EtOAc. The layers were separated. The organic layer was
washed with brine, dried with Na2SO4, and concentrated to a light yellow oil (2.447 g), which was
directly used in the next step.
19
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 37
1
2
3
4
5
6
7
8
9
In a 250 mL roundꢀbottomed flask was previous crude intermediate (2.447 g, 6.03 mmol) in THF (30
mL) to give a slightly yellow solution. After cooling to 0 °C, NBS (1.084 g, 6.09 mmol) was added in
one portion. The mixture was stirred at 0 °C for 30 min. TLC (spot between the TMS ether and starting
ketone) (Rf = 0.47 with 20% EtOAc/hexane) and LCMS showed complete conversion. Cooling bath
was removed and the mixture was stirred at rt for another 10 min. The reaction was quenched with
saturated NaHCO3 solution. The volatiles were stripped off in vacuo and the residue was partitioned
between water and EtOAc. The layers were separated and the aqueous layer was extracted with EtOAc.
The combined organic layers were washed with brine, dried with Na2SO4, and concentrated to a slightly
yellow oil. Purification by FCC up to 40% EtOAc/hexane afforded the desired product 12 (2.087 g, 84%
for 2 steps) as a colorless dense oil: MS(ESI) [M+H+] = 414.30; 1H NMR (400 MHz, CHLOROFORMꢀ
d) δ ppm 8.56 (dd, J = 5.04, 1.76 Hz, 1 H), 7.83 (dd, J = 7.55, 1.76 Hz, 1 H), 7.25 ꢀ 7.32 (m, 1 H), 5.15
(dd, J = 4.78, 2.27 Hz, 1 H), 4.99 (dd, J = 8.06, 3.02 Hz, 1 H), 2.80 ꢀ 2.99 (m, 1 H), 2.30 ꢀ 2.39 (m, 1 H),
2.19 ꢀ 2.30 (m, 1 H), 2.02 ꢀ 2.14 (m, 1 H), 0.94 ꢀ 1.08 (m, 3 H), 0.88 ꢀ 0.95 (m, 9 H), 0.84 (d, J = 7.30
Hz, 9 H); 13C NMR (101 MHz, CHLOROFORMꢀd) δ ppm 199.0, 160.3, 150.2,137.9, 132.4, 122.9,
75.9, 54.7, 30.4, 30.1, 17.8, 17.7, 12.0.
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
(5R,6R,9R)-6-Bromo-9-(triisopropylsilyloxy)-6,7,8,9-tetrahydro-5H-cyclohepta[b]pyridin-5-ol
(13). In a 500 mL roundꢀbottomed flask was previous intermediate 12 (2.087 g, 5.06 mmol) (azeotroped
with dry benzene) in THF (16 mL) to give a colorless solution. After cooling to ꢀ78 °C under nitrogen,
Superhydride (5.06 mL, 5.06 mmol) (1.0M in THF) was added dropwise via syringe. The resulted
yellow mixture was gradually warmed up to ꢀ60 °C in 2 h. LCMS showed good conversion (might have
a minor cisꢀbromohydrin and partially cyclized to epoxide). After 2.5 h, the reaction was quenched with
water. The solvent was removed and the residue was partitioned between EtOAc and water. The layers
were separated and the aqueous layer was extracted with EtOAc. The combined organic layers were
washed with brine, dried with Na2SO4, and concentrated to a slighly yellow oil. The residue was further
20
ACS Paragon Plus Environment

Page 21 of 37
The Journal of Organic Chemistry
1
2
3
4
5
dried overnight (2.097 g): MS(ESI) [M+H+] = 416.30. It was directly carried onto next reaction without
further purification and characterizations.
6
7
8
9
(1aS,4R,8bR)-4-((Triisopropylsilyl)oxy)-1a,3,4,8b-tetrahydro-2H-
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
oxireno[2',3':3,4]cyclohepta[1,2-b]pyridine (14). In a 250 mL roundꢀbottomed flask was intermediate
13 (2.097 g, 5.06 mmol) (azeotroped with dry benzene) in THF (20 mL) to give a colorless solution.
After cooling to 0 °C, NaH (0.486 g, 20.24 mmol) (excess) was added (Caution: gas evolves). The
cooling bath was removed and the mixture was stirred at rt for 1 h. TLC showed complete conversion to
a major more polar spot (Rf = 0.32 with 20% EtOAc/hexane). After another 1 h, the reaction was slowly
quenched with water (gas evolves!) and THF was removed. The residue was partitioned between EtOAc
and brine. The layers were separated. The organic layer was dried with Na2SO4, concentrated to a yellow
oil (1.8 g, LCMS showed 83% purity after two steps), which was directly carried onto next reaction
without further purification and characterizations. In a separate experiment, 14 was purified and
characterized: 1H NMR (400 MHz, Chloroformꢀd) δ 8.54 (dd, J = 4.8, 1.8 Hz, 1H), 7.89 (dd, J = 7.7, 1.7
Hz, 1H), 7.30 (dd, J = 7.6, 4.9 Hz, 1H), 5.19 (t, J = 3.4 Hz, 1H), 3.18 (ddd, J = 18.3, 13.0, 2.4 Hz, 1H),
2.60 (ddd, J = 18.3, 6.2, 2.4 Hz, 1H), 2.16 (ddt, J = 7.1, 5.3, 3.8 Hz, 2H), 2.02 (dddt, J = 10.7, 8.7, 6.8,
3.4 Hz, 1H), 1.64 (ddtd, J = 17.2, 13.1, 8.3, 2.3 Hz, 1H), 1.03 – 0.93 (m, 3H), 0.91 (d, J = 7.0 Hz, 9H),
0.84 (d, J = 7.1 Hz, 9H); 13C NMR (126 MHz, Chloroformꢀd) δ 161.7, 150.2, 136.8, 134.0, 123.0, 76.9,
40.73, 32.9, 19.9, 17.8, 17.7, 12.1, 11.9; HRMS (ESI), m/z calcd for C19H32NO2Si [M+H]+ 334.2197,
found 334.2182; αD20 +163.9 (c 3.00 CHCl3).
(6S,9R)-9-(Triisopropylsilyloxy)-6,7,8,9-tetrahydro-5H-cyclohepta[b]pyridin-6-ol (15). In a 250
mL roundꢀbottomed flask was epoxide 14 (1.688 g, 5.06 mmol) in MeOH (30 mL) (crude: 83% purity)
to give a yellow solution. Pd/C (0.269 g, 0.253 mmol) was added. The mixture was stirred under
hydrogen (balloon) at rt for 2 h. LCMS showed only 10% conversion. Another 269 mg of Pd/C was
added and the reaction continued overnight for 18 h. It was filtered and washed with MeOH. The
21
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 37
1
2
3
4
5
6
7
8
9
combined organic solution was concentrated to a colorless oil. It was purified by FCC first up to 50%
EtOAC/hexane to obtain the starting material (0.6 g, 36%, not pure, contained impurity later shown to
be intermediate 19) and then up to pure EtOAc to elute the desired product (0.145 g, 8.5% for 3 steps):
1H NMR (400 MHz, CHLOROFORMꢀd) δ ppm 8.28 (dd, J = 4.91, 1.38 Hz, 1 H), 7.37 ꢀ 7.52 (m, 1 H),
7.08 (dd, J = 7.43, 4.91 Hz, 1 H), 5.09 (d, J = 6.55 Hz, 1 H), 3.50 ꢀ 3.78 (m, 2 H), 2.71 (d, J = 12.84 Hz,
1 H), 2.51 ꢀ 2.65 (m, 1 H), 2.44 (d, J = 13.09 Hz, 1 H), 2.09 ꢀ 2.26 (m, 1 H), 1.90 ꢀ 2.05 (m, 1 H), 1.73
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
13
(d, J = 3.02 Hz, 1 H), 1.02 ꢀ 1.18 (m, 3 H), 0.94 ꢀ 1.00 (m, 9 H), 0.90 (d, J = 7.05 Hz, 9 H); C NMR
(101 MHz, CHLOROFORMꢀd) δ ppm 162.6, 145.8, 138.8, 132.3, 122.7, 77.0, 70.8, 43.0, 33.8, 30.9,
18.0, 17.8, 12.1; HRMS (ESI), m/z calcd for C19H34NO2Si [M+H]+ 336.2353, found 336.2339; αD20
+34.2 (c 1.61 CHCl3).
(R)-9-(Triisopropylsilyloxy)-8,9-dihydro-5H-cyclohepta[b]pyridin-6(7H)-one (3). In an ovenꢀ
dried 100 mL roundꢀbottomed flask was oxalyl chloride (0.951 mL, 0.951 mmol) in CH2Cl2 (4 mL) to
give a colorless solution at ꢀ55 °C under nitrogen. DMSO (0.135 mL, 1.901 mmol) was added dropwise
slowly over 2 min. After the solution was stirred for an additional 30 min, intermediate 15 (145 mg,
0.432 mmol) (azeotroped with dry benzene) dissolved in 2 mL CH2Cl2 (plus 2 mL rinse) was added via
canuula over 5 min. The reaction mixture was stirred at ꢀ50 ꢀ ꢀ55 °C for an additional 40 min. Et3N
(0.301 mL, 2.161 mmol) was added via syringe at ꢀ50 °C and the reaction mixture was gradually
warmed up to ꢀ20 °C for 30 min. LCMS showed good conversion. Water and EtOAc were added, and
the layers were separated. The aqueous layer was extracted with EtOAc. The combined organic layers
were dried with Na2SO4, and concentrated to a tan oil. Purification by FCC up to 50% EtOAc/hexane
1
afforded the desired product as an colorless oil (122 mg, 85%): MS(ESI) [M+H+] = 334.34; H NMR
(400 MHz, CHLOROFORMꢀd) δ ppm 8.39 (d, J = 5.04 Hz, 1 H), 7.49 (d, J = 7.55 Hz, 1 H), 7.19 (dd, J
= 7.55, 4.78 Hz, 1 H), 5.26 (dd, J = 4.78, 2.27 Hz, 1 H), 4.69 (d, J = 14.35 Hz, 1 H), 3.29 (d, J = 14.35
Hz, 1 H), 3.02 (ddd, J = 12.15, 9.00, 6.04 Hz, 1 H), 2.45 ꢀ 2.59 (m, 1 H), 2.31 ꢀ 2.45 (m, 1 H), 2.06 ꢀ
22
ACS Paragon Plus Environment

Page 23 of 37
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
2.25 (m, J = 8.53, 8.53, 5.98, 2.39 Hz, 1 H), 1.06 ꢀ 1.19 (m, 3 H), 1.01 (d, J = 7.30 Hz, 9 H), 0.89 ꢀ 0.97
(m, 9 H); 13C NMR (101 MHz, CHLOROFORMꢀd) δ ppm 206.9, 160.5, 146.7, 138.4, 128.7, 123.3,
48.4, 39.9, 33.6, 17.9, 17.8, 12.1; HRMS (ESI), m/z calcd for C19H32O2NSi [M+H]+ 334.2197, found
334.2186.
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
(5S,9R)-9-((Triisopropylsilyl)oxy)-6,7,8,9-tetrahydro-5H-cyclohepta[b]pyridin-5-ol (19). After
the hydrogenolysis reaction of 14, in addition to 15, intermediate 19 was also purified and characterized.
The crude product (on 1.357 g, 4.07 mmol reaction scale) was loaded onto a flash column and eluted
with 0 – 100% CH2Cl2/hexane to obtain the recovered starting material 14 (670 mg, 49% recovered),
and then the title product 19 (131 mg, 9.6%). The column was further eluted with 0 – 50%
EtOAc/hexane to obtain the intermediate 15 (117 mg, 8.5%). 19: 1H NMR (400MHz, CHLOROFORMꢀ
d) δ 8.38 (dd, J = 4.9, 1.6 Hz, 1 H), 7.59 (dd, J = 7.5, 1.3 Hz, 1 H), 7.17 (dd, J = 7.5, 4.8 Hz, 1 H), 5.33
(d, J = 11.5 Hz, 1 H), 5.19 (d, J = 6.5 Hz, 1 H), 4.64 (dd, J = 11.5, 6.0 Hz, 1 H), 2.70 ꢀ 2.54 (m, 1 H),
2.32 ꢀ 2.22 (m, 2 H), 1.82 ꢀ 1.62 (m, 3 H), 1.22 ꢀ 1.12 (m, 3 H), 1.06 ꢀ 1.01 (m, 9 H), 1.01 ꢀ 0.95 (m, 9
H); 13C NMR (101MHz, CHLOROFORMꢀd) δ 161.1, 146.5, 138.8, 137.3, 122.7, 78.9, 74.2, 34.2, 32.6,
17.5, 17.3, 17.1, 11.8; HRMS (ESI), m/z calcd for C19H34NO2Si [M+H]+ 336.2353, found 336.2338;
αD20 +33.5 (c 3.65 CHCl3).
(R,Z)-9-(Triisopropylsilyloxy)-8,9-dihydro-7H-cyclohepta[b]pyridine (16). In a 250 mL roundꢀ
bottomed flask was compound 5 (1.97 g, 5.91 mmol) in MeOH (20 mL) to give a colorless solution.
NaBH4 (0.223 g, 5.91 mmol) was added, and the mixture was stirred at rt for 1 h. LCMS indicated
complete conversion. MeOH was stripped off and the residue was partitioned between water and
EtOAc. The layers were separated. The organic layer was washed with brine, dried with Na2SO4, and
concentrated to a colorless oil (2 g, 100%), which was carried onto next reaction without further
purification and characterizations. In a 250 mL roundꢀbottomed flask was the previous crude alcohol
23
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 37
1
2
3
4
5
6
7
8
9
(2.93 g, 8.74 mmol) and (methoxycarbonylsulfamoyl)triethylammonium hydroxide, inner salt (3.12 g,
13.11 mmol) in benzene (80 mL) to give a suspension. It was heated at reflux (preheated oil bath of 85
°C) with stirring under nitrogen for 5 h. LCMS showed good conversion. Volatiles were stripped off and
the residue was partitioned between water and EtOAc. The layers were separated. The organic layer was
washed with brine, dried and concentrated. Purification by FCC up to 30% EtOAc/hexane afforded the
desired product as a colorless oil (1.64 g, 56.4% for 2 steps): MS(ESI) [M+H+] = 318.38; 1H NMR (400
MHz, CHLOROFORMꢀd) δ ppm 8.25 (d, J = 4.53 Hz, 1 H), 7.41 (d, J = 7.81 Hz, 1 H), 6.96 ꢀ 7.15 (m,
1 H), 6.20 (d, J = 12.59 Hz, 1 H), 5.85 ꢀ 6.02 (m, 1 H), 5.25 (d, J = 7.30 Hz, 1 H), 2.73 ꢀ 2.93 (m, 1 H),
2.35 (dd, J = 19.39, 4.03 Hz, 1 H), 2.17 ꢀ 2.29 (m, 1 H), 1.87 (td, J = 13.53, 4.41 Hz, 1 H), 1.01 ꢀ 1.11
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
13
(m, 3 H), 0.93 ꢀ 1.00 (m, 9 H), 0.79 ꢀ 0.89 (m, 9 H); C NMR (101 MHz, CHLOROFORMꢀd) δ ppm
161.1, 145.2, 138.8, 135.0, 131.1, 125.8, 122.4, 76.1, 29.8, 26.3, 18.0, 17.8, 12.4.
(1aR,4R,8bS)-4-((Triisopropylsilyl)oxy)-1a,3,4,8b-tetrahydro-2H-
oxireno[2',3':3,4]cyclohepta[1,2-b]pyridine (17). In a 1 L roundꢀbottomed flask was sodium
hypochlride (191 mL, 166 mmol). Sodium phophate, dibasic (0.880 g, 6.20 mmol) was added. After
cooling to 0°C, intermediate 16 (1.64 g, 5.16 mmol) and (1S,2S)ꢀ(+)ꢀ[1,2ꢀcyclohexanediaminoꢀN,N'ꢀ
bis(3,5ꢀdiꢀtꢀbutylsalicylidene)]manganese (III) chloride (0.394 g, 0.620 mmol) dissolved in CH2Cl2 (40
mL) was added dropwise over 1 h. The dark reaction mixture was allowed to slowly warm to rt and
stirred overnight for 17 h. LCMS showed good conversion to the desired product. The mixture was
diluted with water and Et2O. The layers were separated and the aqueous layer was extracted with Et2O
twice. The combined organic layers were dried with celite and Na2SO4, filtered, and concentrated to a
tan oil. Purification by FCC up to 50% EtOAc/hexane afforded the desired product as a light yellow oil
1
(1.00 g, 58%): H NMR (400 MHz, CHLOROFORMꢀd) δ ppm 8.25 ꢀ 8.44 (m, 1 H), 7.81 (d, J = 8.31
Hz, 1 H), 7.13 (td, J = 7.05, 3.53 Hz, 1 H), 4.94 ꢀ 5.16 (m, 1 H), 3.88 ꢀ 4.04 (m, 1 H), 3.25 ꢀ 3.48 (m, 1
13
H), 2.18 ꢀ 2.38 (m, 1 H), 1.89 ꢀ 2.11 (m, 2 H), 1.11 ꢀ 1.29 (m, 1 H), 0.62 ꢀ 1.10 (m, 21 H); C NMR
24
ACS Paragon Plus Environment

Page 25 of 37
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
(101 MHz, CHLOROFORMꢀd) δ ppm 159.9, 147.7, 139.1, 131.1, 122.7, 76.4, 55.4, 54.3, 30.3, 25.5,
17.9, 17.7, 12.0; HRMS (ESI), m/z calcd for C19H32NO2Si [M+H]+ 334.2197, found 334.2183; αD20
+16.2 (c 3.00 CHCl3).
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
(6R,9R)-9-(Triisopropylsilyloxy)-6,7,8,9-tetrahydro-5H-cyclohepta[b]pyridin-6-ol (18). In a 250
mL roundꢀbottomed flask was intermediate 17 (1.00 g, 3.00 mmol) in MeOH (20 mL) to give a yellow
solution. Pd/C (0.160 g, 0.150 mmol) was added. The mixture was stirred under hydrogen (balloon) at rt
for 2 h. LCMS showed complete conversion. It was filtered and washed with MeOH. The combined
1
13
organic solution was concentrated to a light yellow oil (1.01g, 100%): H and C NMR were obtained
1
(very different from 15, cisꢀalcohol): H NMR (400 MHz, CHLOROFORMꢀd) δ ppm 8.19 ꢀ 8.27 (m, 1
H), 7.44 (d, J = 7.30 Hz, 1 H), 7.06 (dd, J = 7.43, 4.91 Hz, 1 H), 5.06 ꢀ 5.14 (m, 1 H), 4.21 (br. s., 1 H),
3.68 (d, J = 14.10 Hz, 1 H), 3.37 (s, 1 H), 2.76 (dd, J = 14.10, 6.55 Hz, 1 H), 2.39 ꢀ 2.54 (m, 1 H), 1.88 ꢀ
2.01 (m, 2 H), 1.73 ꢀ 1.89 (m, 1 H), 1.00 ꢀ 1.17 (m, 3 H), 0.92 ꢀ 0.99 (m, 9 H), 0.81 ꢀ 0.91 (m, 9 H); 13C
NMR (101 MHz, CHLOROFORMꢀd) δ ppm 162.2, 145.2, 140.6, 132.6, 122.5, 77.2, 66.9, 39.7, 31.3,
26.7, 18.0, 17.8, 12.1; HRMS (ESI), m/z calcd for C19H34NO2Si [M+H]+ 336.2353, found 336.2339;
αD20 +58.9 (c 3.10 CHCl3).
(R)-9-(Triisopropylsilyloxy)-8,9-dihydro-5H-cyclohepta[b]pyridin-6(7H)-one (3). In an ovenꢀ
dried 250 mL roundꢀbottomed flask was oxalyl chloride (3.30 mL, 6.60 mmol) in CH2Cl2 (14 mL) to
give a colorless solution at ꢀ55 °C under nitrogen. DMSO (0.937 mL, 13.20 mmol) was added dropwise
slowly over 2 min. After the solution was stirred for an additional 30 min, crude intermediate 18 (1.01 g,
3.00 mmol) (azeotroped with dry benzene) dissolved in 4 mL CH2Cl2 (plus 6 mL rinse) was added via
canuula over 5 min. The reaction mixture was stirred at ꢀ50 ꢀ ꢀ55 °C for an additional 40 min. Et3N
(2.091 mL, 15.00 mmol) was added via syringe at ꢀ50 °C and the reaction mixture was gradually
warmed up to ꢀ20 °C for 30 min. TLC showed complete conversion. Water and EtOAc were added, and
the layers were separated. The aqueous layer was extracted with EtOAc. The combined organic layers
25
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 37
1
2
3
4
were dried with Na2SO4, and concentrated to a tan oil. Purification by FCC up to 50% EtOAc/hexane
5
6
7
1
afforded the desired product as an colorless oil (814 mg, 81% for 2 steps). H NMR was taken and was
8
9
the same as that made from intermediate 15.
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
Preparation of D-15 and D-18. The deuterium labeling experiments were carried out following the
same procedures described for preparation of 15 and 18 by using deuterium gas in a small balloon. The
compounds were purified and 1H NMR were obtained. The spectra difference were shown in Figure 4.
D-15: HRMS (ESI), m/z calcd for C19H33DNO2Si [M+H]+ 337.2416, found 337.2397. D-18: HRMS
(ESI), m/z calcd for C19H33DNO2Si [M+H]+ 337.2416, found 337.2404.
(6R,9R)-6-(2,3-Difluorophenyl)-9-(triisopropylsilyloxy)-6,7,8,9-tetrahydro-5H-
cyclohepta[b]pyridin-6-ol (20). In an ovenꢀdried 250 mL roundꢀbottomed flask was 1ꢀbromoꢀ2,3ꢀ
difluorobenzene (1.036 g, 5.37 mmol) in THF (10 mL) under nitrogen. After cooling to ꢀ78 °C, nꢀBuLi
(1.757 mL, 4.39 mmol) was added dropwise via syringe (color turned to faint yellow). After 10 min, a
solution of intermediate 3 (814 mg, 2.440 mmol) in THF (4 mL plus 4 mL rinse) was added via syringe
and the reaction was stirred at ꢀ78 °C for 1 h. LCMS indicated very good conversion. The reaction flask
was raised from the cooling bath and stirred at rt for 1.5 h. The reaction was quenched by saturated
NH4Cl solution. THF was stripped off and the residue was partitioned between water and EtOAc. The
layers were separated. TLC (1/1 EtOAc/hexane) showed a much less polar minor spot as starting
material and a more polar major spot (Rf = 0.28). The organic layer was washed with brine, dried with
Na2SO4, and concentrated to a tan oil. Purification by FCC up to 80% EtOAc/hexane afforded the
recovered starting material (148 mg, 18%) as a yellow oil, as well as the desired product (900 mg, 82%)
as a colorless oil: MS(ESI) [M+H+] = 448.36; 1H NMR (400 MHz, CHLOROFORMꢀd) δ ppm 8.26 (d,
J = 3.53 Hz, 1 H), 7.42 (t, J = 6.42 Hz, 1 H), 7.35 (d, J = 7.30 Hz, 1 H), 6.99 ꢀ 7.13 (m, 3 H), 5.16 (br.
s., 1 H), 4.63 (d, J = 13.85 Hz, 1 H), 3.02 ꢀ 3.28 (m, 1 H), 2.71 (d, J = 14.10 Hz, 1 H), 2.34 (d, J = 4.03
26
ACS Paragon Plus Environment

Page 27 of 37
The Journal of Organic Chemistry
1
2
3
Hz, 1 H), 2.02 ꢀ 2.15 (m, 2 H), 1.75 (d, J = 13.85 Hz, 1 H), 1.06 ꢀ 1.19 (m, 3 H), 0.97 ꢀ 1.07 (m, 9 H),
4
5
6
7
0.84 ꢀ 0.97 (m, 9 H). The compound appeared to be a single diastereomer.
8
9
(6R,9R)-6-(2,3-Difluorophenyl)-6,7,8,9-tetrahydro-5H-cyclohepta[b]pyridine-6,9-diol (21). In a
100 mL roundꢀbottomed flask was intermediate 20 (900 mg, 2.011 mmol) in THF (10 mL) to give a
colorless solution. TBAF (2.413 mL, 2.413 mmol) was added, and the mixture was stirred at 50 °C
overnight for 18 h. LCMS indicated complete conversion. THF was stripped off and the residue was
partitioned between water and EtOAc. The layers were separated and the aqueous layer was extracted
with EtOAc. The combined organic layers were washed with brine, dried with Na2SO4, and concentrated
to a tan oil. Purification by FCC afforded the desired product as a white crystalline solid (497 mg, 85%):
MS(ESI) [M+H+] = 292.26; 1H NMR (400 MHz, CHLOROFORMꢀd) δ ppm 8.05 (d, J = 3.78 Hz, 1 H),
7.37 (d, J = 7.55 Hz, 1 H), 7.06 ꢀ 7.21 (m, 1 H), 6.81 ꢀ 7.05 (m, 3 H), 5.63 (br. s., 1 H), 4.82 ꢀ 4.97 (m, 1
H), 3.68 ꢀ 4.11 (m, 2 H), 3.00 (d, J = 14.35 Hz, 1 H), 2.77 (t, J = 10.58 Hz, 1 H), 2.13 (t, J = 11.21 Hz, 1
H), 1.87 ꢀ 2.03 (m, 1 H), 1.69 ꢀ 1.86 (m, 1 H); 19F NMR (376 MHz, CHLOROFORMꢀd) δ ppm ꢀ137.37
(br. s.), ꢀ137.99 (d, J = 15.52 Hz); 13C NMR (101 MHz, CHLOROFORMꢀd) δ ppm 160.5, 152.4 ꢀ 149.9
(dd, J = 13.87 and 248 Hz), 149.9 (dd, J = 12.33 and 249 Hz), 145.3, 139.3, 136.0, 130.7, 123.8, 123.7,
122.6, 121.7, 116.3 (d, J = 16.95 Hz), 73.1, 72.5, 53.5, 44.4, 44.4, 37.8; HRMS (ESI), m/z calcd for
C16H16O2NF2 [M+H]+ 292.1144, found 292.1136.
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
(6R,9R)-6-(2,3-Difluorophenyl)-6-hydroxy-6,7,8,9-tetrahydro-5H-cyclohepta[b]pyridin-9-yl 4-(2-
oxo-2,3-dihydro-1H-imidazo[4,5-b]pyridin-1-yl)piperidine-1-carboxylate (Diastereomeric 1). In an
ovenꢀdried 100 mL roundꢀbottomed flask was intermediate 21 (47.2 mg, 0.162 mmol) (azeotroped with
dry benzene) and 4ꢀnitrophenyl 4ꢀ(2ꢀoxoꢀ2,3ꢀdihydroꢀ1Hꢀimidazo[4,5ꢀb]pyridinꢀ1ꢀyl)piperidineꢀ1ꢀ
carboxylate (93 mg, 0.243 mmol) in DMF (2 mL) to give a light yellow suspension under nitrogen.
NaHMDS (0.648 mL, 0.648 mmol) was added dropwise. The resulted yellow suspension was stirred
under nitrogen at rt overnight for 15 h. LCMS showed the desired product (M + H = 536.27). The
27
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 37
1
2
3
4
5
6
7
8
9
reaction was quenched with saturated NaHCO3 and diluted with EtOAc. The layers were separated and
the aqueous layer was extracted with EtOAc. The combined organic layers were washed with water
(yellow), brine, dried with Na2SO4, and concentrated to a yellow oil. Purification by FCC up to 10%
MeOH/CH2Cl2 afforded one major peak (55 mg) as white foam/solid. It was then further purified by
prepꢀHPLC (30x150mm AXIA LUNA C18 5um Phenominex column, 220nM, 30ml/min, 20ꢀ90%B
over 17min, then 100%B for 3 min, A: 95% H2O/5% MeCN;10mM Ammonium Acetate; B: 95%
MeCN/5% H2O;10 mM Ammonium Acetate) to afford the desired product (41.3 mg, 48%) as a dense
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
1
colorless oil: H NMR (400 MHz, CHLOROFORMꢀd) δ ppm 11.80 ꢀ 10.50 (br. s., 1 H), 8.44 (br. s., 1
H), 8.03 (br. s., 1 H), 7.47 (d, J = 6.80 Hz, 2 H), 6.83 ꢀ 7.31 (m, 5 H), 6.09 (br. s., 1 H), 4.29 ꢀ 4.70 (m, 4
H), 2.84 (d, J = 14.60 Hz, 6 H), 2.11 ꢀ 2.52 (m, 4 H), 1.82 ꢀ 2.01 (m, 2 H); 19F NMR (376 MHz,
13
CHLOROFORMꢀd) δ ppm ꢀ138.31 (d, J = 22.56Hz), ꢀ139.80 (dd, J = 18.80 and 131.60 Hz), C NMR
(101 MHz, Chloroformꢀd) δ 158.2, 153.7, 153.1, 151.9 ꢀ 149.3 (dd, J = 18.23, 308.09 Hz), 148.5 – 145.9
(dd, J = 17.61, 306.83 Hz), 146.6, 142.9, 140.1, 139.8, 138.3 ꢀ 138.1 (d, J = 13.83 Hz), 131.4, 123.8,
122.9, 122.8, 120.3, 116.4, 115.8 ꢀ 115.6 (d, J = 21.38 Hz), 115.0, 79.1, 72.5, 49.9, 43.3, 43.2, 35.4,
29.0, 24.9; HRMS (ESI), m/z calcd for C28H28O4N5F2 [M+H]+ 536.2104, found 536.2088; αD20 +45.99
(c 3.20 CHCl3).
(1aS,4S,8bR)-4-((triisopropylsilyl)oxy)-1a,3,4,8b-tetrahydro-2H-oxireno[2',3':3,4]cyclohepta[1,2-
b]pyridine (ent-17). 1. In an ovenꢀdried 250 mL roundꢀbottomed flask was 5(S)ꢀ9ꢀhydroxyꢀ6,7,8,9ꢀ
tetrahydroꢀ5Hꢀcyclohepta[b]pyridinꢀ5ꢀone (3.16 g, 17.83 mmol) in CH2Cl2 (50 mL) to give a tan
solution. After cooling to 0 °C, TIPSꢀOTf (4.84 mL, 17.83 mmol) and Et3N (4.97 mL, 35.7 mmol) were
added via syringe, and the mixture was stirred at 0 °C for 1h. LCMS indicated complete conversion.
Volatiles were stripped off and the residue was partitioned between NaHCO3 solution and EtOAc. The
layers were separated and the organic layer was washed with brine, dried and concentrated to a tan oil.
The crude was directly used in next reaction. MS(ESI) [M+H+] = 334.28.
28
ACS Paragon Plus Environment

Page 29 of 37
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
2. In a 250 mL roundꢀbottomed flask was previous crude material (5.95 g, 17.83 mmol) in MeOH (50
mL) to give a tan solution. NaBH4 (0.675 g, 17.83 mmol) was added, and the mixture was stirred at rt
for 1 h. LCMS indicated complete conversion. MeOH was stripped off and the residue was partitioned
between water and EtOAc. The layers were separated. The organic layer was washed with brine, dried
with Na2SO4, and concentrated to a tan oil, which was carried onto next reaction without further
purification and characterizations. MS(ESI) [M+H+] = 336.28 (LCMS showed two diastereomers).
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
3. In a 250 mL roundꢀbottomed flask was previous crude material (5.98 g, 17.83 mmol) and
(methoxycarbonylsulfamoyl)triethylammonium hydroxide, inner salt (6.37 g, 26.7 mmol) in benzene
(100 mL) to give a suspension. It was heated at reflux (preheated oil bath of 85 °C) with stirring under
nitrogen for 5 h. LCMS showed complete conversion. Volatiles were stripped off and the residue was
partitioned between water and EtOAc. The layers were separated. The organic layer was washed with
brine, dried and concentrated to a tan oil (5.66 g), which was directly used in the next step without
further purification and characterizations. MS(ESI) [M+H+] = 318.32.
4. In a 2 L roundꢀbottomed flask was sodium hypochlorite (658 mL, 574 mmol). Sodium phosphate,
dibasic (3.04 g, 21.40 mmol) was added. After cooling to 0 °C, previous crude material (5.66 g, 17.83
mmol)
and
manganese(III)
6,6'ꢀ(1E,1'E)ꢀ(1R,2R)ꢀcyclohexaneꢀ1,2ꢀdiylbis(azanꢀ1ꢀylꢀ1ꢀylidene)
bis(methanꢀ1ꢀylꢀ1ꢀylidene)bis(2,4ꢀdiꢀtertꢀbutylphenolate) chloride (1.359 g, 2.140 mmol) dissolved in
CH2Cl2 (140 mL) was added dropwise over 1 h. The dark reaction mixture was allowed to slowly warm
to rt and stirred overnight for 20 h. LCMS showed good conversion to the desired product. The mixture
was diluted with water and Et2O. The layers were separated and the aqueous layer was extracted with
Et2O twice. The combined organic layers were washed with water, brine, dried with celite, filtered, and
concentrated to a dark oil. Purification by FCC up to 50% EtOAc/hexane afforded the desired product as
1
a light yellow oil (2.98 g, 50% for 4 steps): MS(ESI) [M+H+] = 334.35; H NMR (400 MHz,
CHLOROFORMꢀd) δ ppm 8.25 ꢀ 8.44 (m, 1 H), 7.81 (d, J = 8.31 Hz, 1 H), 7.13 (td, J = 7.05, 3.53 Hz,
29
ACS Paragon Plus Environment