Synthesis and pharmacological characterization of 4-substituted-2- aminobicyclo[3.1.0]hexane-2,6-dicarboxylates: Identification of new potent and selective metabotropic glutamate 2/3 receptor agonists

Article abstract of DOI:10.1021/jm4000165

As part of our ongoing interest in identifying novel agonists acting at metabotropic glutamate (mGlu) 2/3 receptors, we have explored the effect of structural modifications of 1S,2S,5R,6S-2-aminobicyclo[3.1.0]hexane-2,6- dicarboxylate (LY354740), a potent

Full text of DOI:10.1021/jm4000165

Subscriber access provided by Temple University Libraries
Article
Synthesis and Pharmacological Characterization of 4-Substituted-2-
Aminobicyclo[3.1.0]hexane-2,6-Dicarboxylates: Identification of New
Potent and Selective Metabotropic Glutamate 2/3 Receptor Agonists.
James A Monn, Matthew J Valli, Steven M Massey, Junliang Hao, Matthew R Reinhard, Mark G Bures,
Beverly A Heinz, Xu-Shan Wang, Joan H Carter, Brian G Getman, Gregory A. Stephenson, Marc
Herin, John T Catlow, Steven Swanson, Bryan G Johnson, David L McKinzie, and Steven S. Henry
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm4000165 • Publication Date (Web): 15 May 2013
Downloaded from http://pubs.acs.org on May 17, 2013
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 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Synthesis
and
Pharmacological
Characterization
of
4-Substituted-2-
Aminobicyclo[3.1.0]hexane-2,6-Dicarboxylates: Identification of New Potent and
+
Selective Metabotropic Glutamate 2/3 Receptor Agonists.
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
*,#
#
#
#
James A. Monn, Matthew J. Valli, Steven M. Massey, Junliang Hao, Matthew R.
#
#
#
#
#
Reinhard, Mark G. Bures, Beverly A. Heinz, Xushan Wang, Joan H. Carter, Brian G.
#
†
§
§
§
Getman, Gregory A. Stephenson, Marc Herin, John T. Catlow, Steven Swanson,
‡
‡
#
Bryan G. Johnson, David L. McKinzie and Steven S. Henry
#
†
§
Discovery Chemistry Research and Technologies, Preformulation, Drug Disposition
‡
and Neuroscience Research, Eli Lilly and Company, Indianapolis, IN 46285.
+
th
Dedicated to Professor Gary L. Grunewald, Ph.D. on the occasion of his 75 birthday.
Abstract
As part of our ongoing interest in identifying novel agonists acting at metabotropic
glutamate (mGlu) 2/3 receptors, we have explored the effect of structural modifications
of 1S,2S,5R,6Sꢀ2ꢀaminobicyclo[3.1.0]hexaneꢀ2,6ꢀdicarboxylate (LY354740), a potent
and pharmacologically balanced mGlu2/3 receptor agonist. Incorporation of relatively
small substituents (e.g. F, O) at the C4 position of this molecule resulted in additional
highly potent mGlu2/3 agonists that demonstrate excellent selectivity over the other
mGlu receptor subtypes, while addition of larger C4ꢀsubstituents (e.g. SPh) led to a loss
of agonist potency and/or the appearance of weak mGlu2/3 receptor antagonist activity.
Further characterization of the αꢀfluoroꢀsubstituted analog (LY459477) in vivo revealed
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 55
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
that this molecule possesses good oral bioavailability in rats and effectively suppresses
phencyclidineꢀevoked locomotor activity at doses that do not impair neuromuscular
coordination.
This molecule therefore represents a valuable new addition to 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
arsenal of pharmacological tools competent to investigate mGlu2/3 receptor function
both in vitro and in vivo.
ACS Paragon Plus Environment

Page 3 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Introduction
Metabotropic glutamate (mGlu) receptors constitute an important class of cellꢀsurfaceꢀ
localized, glutamateꢀsensing proteins which have drawn considerable attention as
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
targets for novel therapeutic agents. Of the eight known mGlu receptor clones, mGlu2
and mGlu3 receptors are among the mostꢀwell studied owing in part the availability of
potent and selective small molecule agonist, antagonist and allosteric modulator tools
that have been used in preclinical models to validate the role of these receptors in a
2
3
4
diverse array of psychiatric, neurologic and proliferative disorders.
Identifying
molecules capable of increasing mGlu2/3 receptor signaling (orthosteric agonists and/or
positive allosteric modulators) has been the subject of particular interest owing to
clinical evidence that oral prodrugs LY544344 (1) and LY2140023.H O (2) of mGlu2/3
2
receptor agonists LY354740 (3) and LY404039 (4) (Figure 1) were effective in
alleviating core symptoms associated with generalized anxiety disorder (1) and
5,6
schizophrenia (2).
Figure 1. Metabotropic glutamate 2/3 receptor agonists (3ꢀ6) and clinically assessed
oral prodrugs (1, 2).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 55
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
We have previously disclosed that constrained bicyclic amino acids such as those
represented by compounds 3 and 4 (Figure 1) possessed highly potent and selective
7
agonist effects at mGlu2/3 receptors. In addition, we found that substitution at the C4ꢀ
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
position of 3 with a methyl group led to 5, a balanced mGlu2/3 agonist and 6
(LY541850), a molecule that exhibits unprecedented mGlu2 agonist/mGlu3 antagonist
pharmacology both in cells expressing recombinant human receptors and in native rat
8
brain preparations. In this account, we have extended our investigation of the effect of
C4ꢀsubstituent structure and stereochemistry on mGlu2/3 receptor function. This effort
has led to the identification of multiple new highly potent and selective mGlu2/3 receptor
agonists. Since methyl substitution at the C4ꢀposition demonstrated a dramatic
influence of stereochemistry on mGlu2/3 receptor functional responses, we sought to
further explore small substitutions at C4 on either the αꢀface (side opposite of ring fusion
hydrogens) or βꢀface (same side as ring fusion hydrogens) of the constrained
bicyclohexane framework.
Chemistry: Our synthetic strategy was enabled by versatile intermediate 8, which by
virtue of the ketone functionality present at the C4 position of the bicyclic nucleus
allowed us to incorporate a variety of substituents at this site with a high degree of
stereochemical control. The synthesis started with the known enantiomerically pure
9
enone 7. Carboxyꢀcyclopropanation of 7 provided ketone 8 in high yield and
10,11
diastereoselectivity
(ds
>95:5).
Further
treatment
of
8
with
methyltriphenylphosphonium ylide gave exocyclic olefin 9 (Scheme 1).
ACS Paragon Plus Environment

Page 5 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 1. Preparation of versatile intermediate 8.
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
Reagents and conditions. (a) tꢀbutyl 2ꢀ(2,3,4,5ꢀtetrahydrothiophenꢀ1ꢀyl)acetate,
F CCH OH (64%); (b) Ph PCH Br, NaHMDS, THF (99%).
3
2
3
3
Preparation of analogs possessing substituents oriented on the αꢀface of 3 is depicted
in Scheme 2. Reduction of ketone 8 with Lꢀselectride proceeded with hydride attack
exclusively from the concave face of the bicyclic ring system to give βꢀcarbinol 10 which
was then smoothly converted to tosylate 11 under standard conditions. Nucleophilic
displacement of the tosylate proceeded with complete inversion of stereochemistry to
yield intermediates 12a-c. Azide 12b was reduced to primary amine 12d, which was in
turn acetylated to provide 12e. Finally, alcohol 10 was treated with DAST to provide
C4αꢀfluoro intermediate 12f. Analogs possessing substituents oriented on the βꢀface of
3 were prepared as shown in Scheme 3. Stereoselective conversion of 10 to C4αꢀ
bromide derivative 13 was achieved in high yield under standard conditions.
Displacement of the bromide with either azide or thiophenol nucleophiles proceeded
with inversion of stereochemistry to give intermediates 14a and 14b, respectively.
Azide 14a was reduced to amine derivative 14c which was then acetylated to give 14d.
Unlike the smooth conversion of carbinol 10 to C4αꢀfluoro analog 12f, attempts to
prepare useful quantities of C4βꢀfluoro analog 14e were largely unsuccessful.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 55
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
Scheme 2. Preparation of C4α amino acid precursors.
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
Reagents and conditions. (a) Lꢀselectride, THF (92%); (b) TsCl, NEt , DMAP, CH Cl
2
3
2
(70%); (c) CF CO Na, DMF; then aq. NaHCO (62%); (d) TMSN , TBAF, THF (92%);
3
3
3
3
(e) PhSH, K CO , DMF (89%); (f) H , Pd/C, EtOH (97%); (g) Ac O, NEt , DMAP, CH Cl
2 3 2 2 3 2 2
(
94%); (h) DAST, CH Cl (71%).
2
2
Reaction of alcohol 12a with DAST led to the isolation lactone 15 as the primary product
(88%), whose structure was unambiguously determined by single crystal Xꢀray
12
analysis. This product may have been produced via intramolecular trapping of a
transient C4ꢀcarbocation by the C2ꢀester with concomitant loss of the tert-butyl group,
though this proposed mechanistic path has not been rigorously established. Most of the
remaining mass was comprised of previously prepared αꢀF intermediate 12f (8%), with
only trace quantities (2%) of desired βꢀfluoro analog 14e being isolated. Attempts to
displace bromide 13 with fluoride were unsuccessful, leading exclusively to elimination
product 16.
ACS Paragon Plus Environment

Page 7 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 3. Preparation of C4β amino acid precurosors
Br
N
3
NHR
H
H
H
H
tꢀBuO
2
C
tꢀBuO
2
C
2
tꢀBuO C
a
b
d
10
H
CO
2
tꢀBu
H
CO
2
tꢀBu
H
CO
2
tꢀBu
H
H
NHBoc
NHBoc
NHBoc
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
14a
SPh
1
1
4c (R=H)
4d (R=Ac)
e
H
H
c
tꢀBuO C
2
H
2
CO tꢀBu
NHBoc
14b
F
F
H
H
H
tꢀBuO C
tꢀBuO C
tꢀBuO C
O
f
2
2
2
1
2a
+
+
H
CO
2
tꢀBu
tꢀBu
H
2
CO tꢀBu
H
O
H
H
H
H
NHBoc
NHBoc
NHBoc
12f
14e
15
2
tꢀBuO C
g
13
H
CO
2
H
NHBoc
16
Reagents and conditions. (a) PPh , Br , PhMe (67%); (b) TMSN , TBAF, THF (91%); (c) PhSH,
3
2
3
K CO , DMF (95%); (d) H , Pd/C, EtOH (97%); (e) Ac O, NEt , DMAP, CH Cl (90%); (f) DAST,
2
3
2
2
3
2
2
CH Cl (88% of 15); (g) KHF / TBADHF, dodecane or TBAF, ACN (48%). Inset: Xꢀray crystal
2
2
2
structure of compound 15.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 55
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
Finally, fully protected amino acids were converted to their corresponding zwitterions
through our recently reported microwaveꢀassisted solvolysis with AcOH and water
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
13
(Scheme 4).
In the case of 12d and 14c, hydrolysis under these conditions provided
the corresponding acetate salts.
Scheme 4. Deprotection of NꢀBoc, diꢀtertꢀbutyl ester intermediates.
Results: Compounds 3, 5, 6, 17a,b, 18a-f and 19a-e were assessed for their functional
14
effects in recombinant human mGlu2 and mGlu3ꢀexpressing cells (Table 1). Each
compound was also assessed for functional activity in cells expressing each of the other
six human mGlu receptor subtypes. Except for relatively weak potency responses in
15
hmGlu6 and hmGlu8ꢀreceptor expressing cells as noted in Table 1, no discernable
ACS Paragon Plus Environment

Page 9 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
activity was observed for these molecules up to the highest test concentration (25 ꢀM
for agonist, 12.5 ꢀM for antagonist) in cells expressing the other mGlu receptor
subtypes (mGlu1, mGlu4, mGlu5 and mGlu7; Table S2). Introduction of a carbonyl
group at the C4ꢀposition of 3 led to a dramatic enhancement in agonist potency for
compound 17a in both mGlu2 (EC50 = 0.16 nM) and mGlu3 (EC50 = 0.11 nM)
expressing cells. In contrast, enhanced potency at mGlu2 (EC50 = 1.3 nM) but not
mGlu3 (EC50 = 15.5 nM) was observed for exocyclic methylene analog 17b compared
to 3, and this molecule was found to be considerably more potent than either of the
corresponding singleꢀbond linked methyl analogs 5 or 6.
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
Substitution at C4 with either hydroxyl (18a and 19a) or azide (18b and 19b)
groups did not substantially enhance or diminished agonist potency compared to
unsubstituted 3, though these substitutions did lead to enhanced potency when
compared to the corresponding methylꢀsubstituted variants. Introduction of the more
polar amino substituent in either αꢀ or βꢀ configurations resulted in a modest loss of
agonist potency relative to 3 (αꢀNH , 18c, EC = 47.9 nM and 57.1 nM at mGlu2 and
2
50
mGlu3; βꢀNH , 19c, EC = 21.2 nM and 24.9 nM at mGlu2 and mGlu3). Interestingly,
2
50
acetylated variants of 18c and 19c led to differential functional outcomes. Acetylation of
βꢀamino derivative 19c led to diminished potency, but maintained mGlu2 and mGlu3
receptor agonist activity for 19d (EC50 = 257 nM and 193 nM at mGlu2 and mGlu3,
respectively), whereas acetylation of αꢀamino analog 18c afforded 18d, a relatively low
potency antagonist at both mGlu2 (IC = 4990 nM) and mGlu3 (IC = 3390 nM).
50
50
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 55
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
Table 1. Functional Agonist or Antagonist Responses of C4ꢀSubstituted Analogs in Cells Expressing Recombinant
1
0
Human mGlu Receptor Subtypes.
Functional responses in hmGlu receptorꢀexpressing cells (inhibition of forskolinꢀ
14,15
stimulated cAMP)
a
EC50 (nM) ± SEM; (% max effect) or
b
IC50 (nM)± SEM; (% max effect)
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
No.
R
Face
mGlu2
mGlu3
mGlu6
mGlu8
3
5
6
H
CH
CH
n/a
β
5.6 ± 0.97 (100)
28.6 ± 7.01 (98)
63.6 ± 11.8 (98)
0.16 ± 0.03 (98)
1.3 ± 0.14 (99)
7.6 ± 0.52 (99)
7.5 ± 0.95 (99)
8.2 ± 1.03 (97)
27.1 ± 5.7 (98)
166 ± 66.3 (88)
3420 ± 2100 (100)
0.11 ± 0.03 (99)
15.5 ± 4.25 (97)
14.2 ± 3.34 (99)
6.6 ± 1.42 (98)
12.3 ± 2.94 (96)
3360 ± 706 (96)
9280 ± 3730 (94
7540 ± 1570 (94)
1380 ± 167 (100)
267 ± 11 (100)
>25000
>25000
3
3
α
>25000
1
7a
7b
8a
9a
8b
=O
=CH
OH
n/a
n/a
α
6530 ± 1990 (98)
1590 ± 172 (96)
>25000
1
1
1
1
2
4370 ± 348 (98)
3040 ± 298 (100)
780 ± 38.9 (99)
OH
β
>25000
N
3
3
α
12400 ± 5200 (82)
1
9b
N
β
12.2 ± 2.26 (99)
23.5 ± 2.92 (98)
1890 ± 428 (100) 12000 ± 1560 (91)
18c
19c
18d
NH
2
α
β
α
47.9 ± 6.90 (98)
21.2 ± 2.17 (99)
4990 ± 1070 (92)
57.1 ± 10.5 (99)
24.9 ± 3.73 (98)
3390 ± 543 (100)
11800 ± 588 (96)
9370 ± 1420 (97)
6690 ± 824 (87)
>25000
>25000
>25000
NH
2
NHAc
1
1
9d
8e
NHAc
SPh
β
257 ± 16.2 (93)
193 ± 43.2 (81)
709 ± 69.1 (99)
1270 ± 254 (98)
4300 ± 776 (85)
>25000
c
α
7020± 770 (100)
895± 350 (100)
1
170 ± 239 (49)
1
9e
SPh
F
β
3370± 1210 (100)
3250 ± 535 (98)
1410 ± 129 (99)
>25000
1
5300± 5400 (53)
1
8f
α
0.56 ± 0.13 (98)
0.24 ± 0.04 (99)
11500 ± 4090 (92)
a
b
Maximal response achieved compared to glutamate (for mGlu2, mGlu3, mGlu8) or LꢀAP4 (mGlu6). Maximal %
c
inhibition achieved in reversing a 100% glutamate (mGlu2, mGlu3, mGlu8) or LꢀAP4 (mGlu6) agonist response. In
two out of eight agonist format assays, low potency and submaximal (~30%) efficacy agonist responses were
observed. The other six agonist format assays failed to demonstrate an agonist response above baseline. In
comparison, four out of four antagonist format assays displayed consistent antagonist effects with 100% maximal
inhibition of glutamate.
ACS Paragon Plus Environment

Page 11 of 55
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
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 55
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
Incorporation of the larger and more lipophilic phenylthioether substituent in
either stereoꢀconfiguration at C4 led to a loss of agonist efficacy, with both 18e and 19e
producing relatively low potency antagonist effects in both mGlu2 and mGlu3 receptorꢀ
expressing cells. These latter findings pertaining to aminoacetyl and thiophenyl
substituents suggest that the region of the agonistꢀreceptor complex adjacent to the C4
position, especially in the area occupied by C4α substituents, has limited tolerance for
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
steric bulk. Finally, introduction of fluorine at the C4α position of 3 led to a significant
enhancement of agonist potency at both mGlu2 and mGlu3 receptors, with 18f (mGlu2
EC = 0.56 nM; mGlu3 EC = 0.24 nM) being particularly noteworthy owing to its subꢀ
50
50
nanomolar potencies at these receptors.
Molecular modeling. In an effort to rationalize the observed structureꢀactivity
relationships of C4 substituted variants of 3, we investigated possible mGlu3 receptor
binding modes for selected compounds. A published Xꢀray crystal structure of the
reference mGlu2/3 receptor agonist (1R,2R)ꢀ3ꢀ[(1S)ꢀ1ꢀaminoꢀ2ꢀhydroxyꢀ2ꢀoxoꢀ
ethyl]cyclopropaneꢀ1,2ꢀdicarboxylic acid (DCGꢀIV, 20) bound within the amino terminal
16
domain (ATD) of mGlu3 (PDB ID = 2e4v served as a useful template for generating
proposed binding modes for compounds exhibiting agonist behavior. All modeling was
17
carried out in the Schrodinger Maestro program, version 9.2. Images depicting
18
proposed binding modes were generated using PyMOL, version 1.2r3.
As can be seen in Figure 2, compound 20 is involved in a network of hydrogenꢀbinding
interactions at the mGlu3 glutamate binding site. These hydrogenꢀbonds include
ACS Paragon Plus Environment

Page 13 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
interactions with the backbone and with the side chains of nearby polar residues. For
example, the carboxyl group of the αꢀamino acid portion of 20 interacts with the
backbone of Serꢀ151 and Thrꢀ174, while side chains Lysꢀ389 and Argꢀ68 interact with
the C2’ carboxyl group. Of particular note are the bound water molecule which appears
to form Hꢀbonds with both C2’ and C3’ carboxylate functionalities and the backbone NꢀH
of Serꢀ278, which also acts as an Hꢀbond donor to the C3’ꢀcarboxylate.
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
16
Figure 2. Xꢀray crystal structure of 20 bound to mGlu3 (PDB ID = 2e4v) showing the
hydrogenꢀbonding network (panel A) and a special representation depicting the location of a
bound water molecule interacting with the C2’ and C3’ carboxylates present in 20 as well as a
likely Hꢀbond between the C3’ꢀcarboxylate and a backbone NꢀH arising from Serꢀ278 (panel B).
We used the structure of mGlu3 complexed with 20 to construct a binding mode for 3
and C4ꢀsubstituted analogs of 3 that maintained the analogous key hydrogenꢀbonding
interactions with the embedded glutamate backbone in these molecules (Figure 3).
Thus, the glutamate pharmacophore present in 3 is predicted to form Hꢀbond
interactions with most of the mGlu3 residues that directly interact with 20, but is capable
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 55
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
of only a single Hꢀbond contact to the bound water molecule and no Hꢀbond interactions
with Serꢀ278 (Figure 3A). Based on this proposed binding mode, it is plausible that
small polar hydrogen bondꢀacceptor substituents suitably positioned at the C4 carbon of
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
might result in more tightly bound compounds owing to the formation of additional Hꢀ
bond interactions. This is illustrated for analogs 17a and 18f in Figures 3B and 3C,
respectively. Introduction of these substituents places the carbonyl oxygen of 17a and
fluorine of 18f in very close proximity of both the bound water molecule and backbone
NꢀH of Serꢀ278. It is hypothesized that these additional binding interactions might in
turn result in a more highly stabilized closed (Gꢀproteinꢀactivating) form of the ATD
leading to the observed increases in agonist potencies compared to 3. Though
speculative, the lack of a potency increase for analog 18a compared to that seen with
1
8f may be the result of an inability of the hydroxyl functionality of 18a to achieve a
rotomer that directs the lone pair electrons on the oxygen in such a way that would form
a productive Hꢀbond with Serꢀ278 and the crystal water. Similarly, the reduced potency
of other analogs containing hydrogen bond donating substituents at the C4 position,
such as 18c, may be due in part to the lack of a productive interaction with the
backbone NꢀH of Serꢀ278 (Figure 3D).
We also examined the impact of C4 thiophenyl substitution (e.g. 18e and 19e) on
predicted docking to the mGlu3 ATD in either closed or open forms. Based on the
observed antagonist pharmacology for these molecules and the close proximity of
receptor residues (e.g. Serꢀ278) to the C4ꢀposition of bound agonists (Figures 2 and 3),
we anticipated severe steric clashes arising from C4ꢀSPh substitution with the closed
form of the protein, and this was indeed observed for both 18e (Figure 4, panel A) and
ACS Paragon Plus Environment

Page 15 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
9e (not shown). Conversely, each of these molecules could be readily accommodated
when docked to the published antagonist bound form of the mGlu3 ATD (PDB ID =
19
3
sm9) as exemplified by 18e (Figure 4, panel B).
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
Figure 3. Proposed binding mode for compounds 3 (panel A), 17a (panel B), 18f (panel C) and
16
18c (panel D) docked in the ATD of mGlu3 (2e4v).
Arrow in panel A indicates ligandꢀ
associated space into which appropriate polar substituents capable of hydrogenꢀbond acceptor
interactions with Serꢀ278 and the bound water molecule. These possible interactions are
evident in the docked structures of 17a (panel B) and 18f (panel C). Hydrogen bond donating
substituents at C4 make fewer productive interactions (panel D) resulting in a loss of potency
relative to 17a and 18f.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 55
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
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
6
Figure 4. The closed, agonistꢀbound, form of the mGlu3 ATD (2e4v) cannot accommodate
steric bulk at C4 as evident by significant steric clashes between a portion of the protein chain in
the lower lobe and the thiophenyl group in 18e (green), indicated by the arrow in Panel A.
Compound 18e, which exhibits antagonist activity, is more likely to form productive interactions
19
with the open form of the mGlu3 ATD (3sm9) as depicted in Panel B.
Rat Pharmacokinetics. Compounds of this structural class (constrained glutamate
analogs) possess negative cLogP / cLogD values. They typically display very high
20
aqueous solubility, low passive permeability and, with one known exception, are not
metabolized. In spite of their low passive permeability, several members of this class
have been reported to exhibit good to excellent oral absorption in preclinical
4c,21a,b
species
Intestinal absorption in the rat has been suggested to be transporterꢀ
7c
mediated owing to the observation of enantioselective absorption , however, the
identity of this putative transporter remains unknown. Owing to its subꢀnanomolar
ACS Paragon Plus Environment

Page 17 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
mGlu2/3 receptor potency and excellent selectivity profile, compound 18f was further
10,22,23
assessed in vivo.
The pharmacokinetic attributes of 18f were evaluated in male
SpragueꢀDawley rats (Figure 5) following oral administration at 1, 3 and 10 mg/kg and
intravenously at 3 mg/kg and plasma drug concentrations were monitored over 24 hours
postꢀdosing
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
00000
18fꢀ3 mg/kg IV
1
1
1
8fꢀ1 mg/kg PO
8fꢀ3 mg/kg PO
8fꢀ10 mg/kg PO
10000
1
000
1
00
10
1
0
4
8
12
16
20
24
Time (hr)
Figure 5. Plasma pharmacokinetics for compound 18f following single oral or
intravenous doses in male SpragueꢀDawley rats.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 55
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
Table 2. Mean pharmacokinetic parameters associated with 18f dosed by either oral or
intraveneous route of administration in rats
10,15
Parameter
Route
Units
3 mg/kg
IV
1 mg/kg
PO
3 mg/kg
PO
10 mg/kg
PO
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
AUC 0ꢀ24
nM*Hours
nM
16700±2030
36660±3040
0±0
2790±233
713±73
2.5±0.6
n.c.
10700±1680
3940±1230
1.8±0.5
n.c.
23600±3850
7940±2010
2.0±0
C or C
o
max
Tmax
CL
Hours
mL/min/kg
L/kg
14.8±1.8
1.25±0.25
7.3±2.5
n.c.
Vd ss
T1/2
F
n.c.
n.c.
n.c.
Hours
2.7±0.6
3.5±1.2
5.9±2.3
n.c.
%
n.c.
n.c.
64.4±11.1
AUC 0ꢀ24 Area under the plasma concentration time curve from 0ꢀ24 hours postꢀdosing
Co
Cmax
CL
Plasma concentration after intravenous dosing extrapolated to time 0
Peak plasma concentration after oral dosing
Intravenous clearance
Vd ss
T1/2
F%
Volume of distribution at steady state
Plasma disappearance halfꢀlife
Oral bioavailability
Mean plasma pharmacokinetic parameters are shown in Table 2 and plasma
concentrationꢀtime curves in Figure 5. Oral absorption of 18f was relatively slow with
peak plasma concentrations occurring two hours postꢀdosing.
Plasma AUC and
Cmax values increased with dose in an approximately linear manner over the oral dose
range of 1 to 10 mg/kg. Clearance was low at about 27% of liver blood flow with a
24
volume of distribution of about twice total body water.
Plasma concentrations
declined with a halfꢀlife of 2.7ꢀ7.3 hours. Despite the polar nature of this zwitterionic
compound, 18f was extensively absorbed after oral administration based upon the
calculated oral bioavailability of 64%.
ACS Paragon Plus Environment

Page 19 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Behavioral pharmacology. Given that the mGlu2/3 receptor agonist potencies as well
7c
as plasma exposures for 18f compare favorably to those reported for 4, a molecule
known to produce potent effects in rodent models of psychosis and has demonstrated
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
6a
efficacy in schizophrenic patients, the effect of 18f in the phencyclidine (PCP) model
of psychosis was evaluated. We employed the same dose range in this study as was
assessed in the aforementioned pharmacokinetic studies (Figure 6A). As can be seen,
a doseꢀdependent decrease in PCPꢀelicited ambulations was observed when 18f was
administered orally two hours prior to testing (90 minutes prior to 5 mg/kg PCP), leading
to a calculated ED50 of 2.08 mg/kg in this assay. A oneꢀway betweenꢀgroups ANOVA
was calculated on the ambulations [F(4,35) = 14.75, p < 0.0001] and a Dunnett Multiple
Comparison Test (PCP alone as comparison control; alpha set at p < 0.05) indicated
that only the 1 mg/kg dose group failed to statistically separate from PCP challenge. In
order to rule out potentially confounding effects of motor impairment, 18f was assessed
using the rotorod test (Figure 6B). Oral doses of 3, 10 and 30 mg/kg of 18f were again
given two hours prior to testing. Evidence of motor impairment was observed at 30
mg/kg, this dose reducing the mean amount of time spent on the rotating barrel by
nearly 70% [F(3,28) = 59.68, p < 0.0001; Dunnett Multiple Comparison Test with
Vehicle as comparison control; alpha set at p < 0.05]. However, at doses associated
with efficacy in the PCP model (3 and 10 mg/kg), no disruption in motor coordination
was apparent, indicating the pharmacologic specificity of this antipsychoticꢀlike effect.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 55
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
B
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
2500
6
0
2000
1500
1000
500
0
50
40
30
20
10
0
*
*
*
*
1
8f
0
0
1mg/kg 3mg/kg 10mg/kg
18f
0
3mg/kg 10mg/kg 30mg/kg
PCP
0
--------------5mg/kg-----------------
Figure 6. Effect of oral doses of 18f on PCPꢀevoked ambulations (A) and rotorod performance
B) in rats. Compound 18f was administered as an aqueous solution two hours prior to
(
evaluation in each test. PCPꢀLMA (n=8 / group); Rotorod (n=8 / group). * p < 0.05 versus PCP
(panel A) or Vehicle (panel B) using Dunnett Multiple Comparison Test.
Summary:
Investigation of substituents attached to the C4ꢀposition of 2ꢀ
aminobicyclo[3.1.0]hexaneꢀ2,6ꢀdicarboxylate 3 has led to the identification of multiple
additional highly potent and selective mGlu2/3 receptor agonists, including two (17a,
1
8f) displaying remarkable subꢀnanomolar potency at these targets. Results from this
study have also revealed apparent steric limits adjacent to the C4ꢀposition of this
bicyclic scaffold which, when exceeded (e.g. 18e and 19e), abolish agonist
pharmacology at both mGlu2 and mGlu3 receptors.
prepared thus far in this series, only the previously described C4αꢀmethyl analog 6 was
found to possess mixed mGlu2 agonist / mGlu3 antagonist pharmacology. The
Interestingly, of the compounds
molecular basis for this highly unusual and unexpected functional selectivity remains
ACS Paragon Plus Environment

Page 21 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
elusive. While attempts to explain this functional selectivity using molecular modeling
8a
have been made, it is unlikely that anything short of solved crystal structures of this
molecule with the mGlu2 and mGlu3 proteins will prove helpful. Even then, it is entirely
possible that the reason(s) underlying this functional selectivity will remain enigmatic.
Finally, 18f, one of the more highly potent analogs derived from this work, was found to
be orally bioavailable in rats and to produce potent antipsychoticꢀlike effects in a rodent
model of psychosis at doses that do not impair motor function. Hence, 18f (LY459477)
represents a valuable new pharmacological tool for studying mGlu2/3 receptors both in
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
25
vitro and in vivo.
Experimental Section.
1
13
Synthesis. Hꢀ and CꢀNMR spectra were obtained on a Varian Unity INOVA 400 at
4
00 MHz and 125 MHz respectively, unless noted otherwise. TMS was used as an
internal standard. LCMS data were obtained using an Agilent 1100 Series HPLC on a
GeminiꢀNX C18 110A, 50 X 2.00 mm column. HRMS were obtained using an Agilent
1
100 Series LC and TOF mass spectrometer, on a GeminiꢀNX C18 110A, 50 X 2.00
mm column. Optical rotations were obtained on a Perkin Elmer Polarimeter 341.
Melting points were obtained on an OptiMelt Automated Melting Point System.
Microwave reactions were run in a Biotage Initiator Microwave. Reactions were
monitored by thin layer chromatography using silica gel 60 F254 plates from EMD and
staining with ninhydrin. Unless otherwise noted normal phase purifications were
®
performed using RediSep preꢀpacked columns from Teledyne Isco. On the basis of
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 55
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
combustion analysis or HPLC data all final compounds were >95% pure. Numbering
convention used in NMR assignments is provided in Figure 7. A table is provided with
detailed elemental analysis (C, H, N) as Supporting Information.
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
ACS Paragon Plus Environment

Page 23 of 55
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
Figure 7. Numbering convention used in NMR peak assignments.
tert-Butyl-2-(thiophene)-acetate sulfonium bromide. To a solution of
tetrahydrothiophene (216 mL, 216 g, 1.63 mol) in 325 mL of acetone was added tꢀbutyl
bromoacetate (160 mL, 197 g, 1.08 mol). Precipitate formed within 1 h. After stirring
overnight, the solid was filtered, washed with acetone, and vacuum dried to obtain the
1
sulfonium bromide salt (298 g, 1.05 mol, 97% yield). H NMR (500 MHz, DMSOꢀd , δ):
6
4
.40 (s, 1H), 3.51 (m, 2H), 3.48 (m, 2H), 2.23 (m, 2H), 2.13 (m, 2H), 1.42 (s, 9H).
Sulfonium ylide. To a 0 °C solution of sulfonium bromide salt (757 g, 2.67 mol) in 2.2
L of CH Cl was added 1.7 L of a saturated aqueous K CO solution keeping the
2
2
2
3
temperature below 10 °C. After stirring the biphasic mixture 1.5 h, 223 mL of 50% aq.
NaOH was added portionꢀwise keeping the temperature below 5 °C. The mixture was
stirred for 3 h and filtered. The salts were rinsed with CH Cl . The layers were
2
2
separated and the aqueous layer was extracted with 600 mL of CH Cl . The combined
2
2
organic layers were dried over solid K CO and concentrated under vacuum to give the
2
3
sulfonium ylide (533. g, 2.64 mol, 98% yield) as a clear, yellow oil. Upon storage in the
1
freezer, the oil crystallized giving an offꢀwhite solid, mp 48ꢀ50 °C. H NMR (500 MHz,
CDCl , δ): 3.12 (m, 2H), 3.00 (m, 2H), 2.85 (s, 1H), 2.41 (m, 2H), 1.84 (m, 2H), 1.38 (s,
3
9
H).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 55
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
(
1S,2S,5R,6R)-Di-tert-butyl 2-[(tert-butoxycarbonyl)amino]-4-
oxobicyclo[3.1.0]hexane-2,6-dicarboxylate, (8). To a 0 °C solution of enone 7 (194
g, 653 mmol) in 650 mL of CH Cl was added trifluoroethanol (474 mL, 6.5 mol, 10
2
2
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
equiv). A solution of the sulfonium ylide, prepared above, (396 g, 1.96 mol) in 325 mL
CH Cl was added dropꢀwise over 40 min keeping the temperature below 10 °C. After 1
2
2
h, the ice bath was removed. After 2.5 h, HPLC analysis showed the reaction to be
complete. (The ratio of desired exo to endo diastereomer was > 95:5 by HPLC).
Deionized water (680 mL) was added to the reaction mixture and the layers were
separated. The aqueous layer was extracted with 400 mL CH Cl . The combined
2
2
organic layers were washed with 500 mL of brine, dried with Na SO , and concentrated
2
4
under vacuum to obtain 587 g of an amber solid. The solid was dissolved in 400 mL of
CH Cl and eluted through a 1.6 kg silica gel plug using 5:1:1 hexanes:MTBE: CH Cl
2
2
2
2
as the eluent. A total of 13.2 L of eluent was collected and concentrated to give 398.7 g
of white solid. The solid was dissolved in 3 L of refluxing 70:30 hexanes:MTBE. The
solution was allowed to cool to rt overnight (precipitate formed at 50 °C) and then
cooled in an ice bath for 1 h. The solid was filtered, rinsed with cold solvent
(approximately 700 mL), and vacuum dried at 35 °C to obtain ketone 8 (173 g, 420
20
1
mmol, 64%) as a white solid. mp 144ꢀ46 °C. [α]D +30.5 (c = 1, CHCl ). H NMR (500
3
MHz, CDCl , δ): 5.36 (bs, 1H), 2.88 (m, 1H), 2.64 (dd, J = 5.2, 3.2 Hz, 1H), 2.37 (d, J =
3
13
2
2
2
.7 Hz, 1H), 2.23 (bs, 1H), 1.45 (s, 9H), 1.43 (s, 18H). C NMR (125 MHz, CDCl , δ):
3
06.2, 171.2, 168.5, 155.3, 83.4, 82.7, 80.7, 61.2, 43.2, 36.0, 34.3 28.4, 28.2, 28.0,
5.3. IR (CHCl ): 2982, 1744, 1719, 1485, 1394, 1309 cmꢀ1. MS (ES+) m/e (%
3
ACS Paragon Plus Environment

Page 25 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
relative intensity) 412.2 (M+ + 1, 79), 356.2 (50), 300.1 (97), 276.1 (68), 244.1 (100).
Anal. Calcd. for C H NO : C, H, N..
21 33
7
(1S,2S,5R,6S)-Di-tert-butyl 2-[(tert-butoxycarbonyl)amino]-4-
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
methylenebicyclo[3.1.0]hexane-2,6-dicarboxylate, (9). Methyltriphenylphosphonium
bromide (11.5 g, 31.59 mmol, 1.3 equiv.) was dissolved in anhydrous THF (243 mL) in a
1
L roundꢀbottom flask under nitrogen and cooled to 0 °C. Sodium
bis(trimethylsilyl)amide (1 M in THF, 31.6 mL, 31.59 mmol, 1.3 equiv.) was added dropꢀ
wise over 13 min to the stirred suspension. The resultant brightꢀyellow mixture was
stirred at 0 °C for approximately 20 min at which time a solution of ketone 8 (10.0 g,
24.30 mmol, 1.0 equiv.) in THF (16 mL) and added to the ylide via cannula. The
reaction was slowly warmed to rt and allowed to stir overnight. The reaction was diluted
with ethyl acetate (500 mL) and washed with water (200 mL) and brine (150 mL). The
organic was dried over Na SO , filtered, and concentrated in vacuo to give a crude oil
2
4
that was purified on a 400 g silica column eluting with 85/15 hexanes/ethyl acetate.
Fractions containing product were combined and concentrated in vacuo to give
20
exocyclic olefin 9 (9.95 g, 24.3 mmol, 99%) as a white solid. mp 75ꢀ77 °C. [α]D
ꢀ
+
1
47.45 (c = 1.054, EtOH). MS (ES+) 432.2 [M+Na] . H NMR (400 MHz, CDCl , δ):
3
5
.19 (bs, 1H), 5.02 (d, J = 2.5 Hz, 1H), 4.85 (s, 1H), 3.06 (bm, 1H), 2.42 (m, 2H), 1.96
13
(
m, 1H), 1.85 (t, J = 3.0 Hz, 1H), 1.44 (s, 9H), 1.43 (s, 18H). C NMR (125 MHz,
CDCl , δ): 171.5, 170.5, 155.1, 147.2, 129.5, 120.4, 115.3, 107.2, 81.8, 81.2, 65.0,
3
+
3
3.9, 28.2, 28.1, 27.8, 24.9. HRMS ESI (m/z): calcd. for C H NO Na [M+Na] :
22
35
6
432.2368. Found: 432.2378.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 55
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
(
1S,2S,4S,5R,6R)-di-tert-butyl 2-[(tert-butoxycarbonyl)amino]-4-
hydroxybicyclo[3.1.0]hexane-2,6-dicarboxylate, (10). To a solution of the ketone 8
20.0 g, 48.60 mmol, 1.0 equiv.) in THF (240 mL) at ꢀ5 to ꢀ10 °C was added a 1M
(
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
solution of lithium tri(secꢀbutyl)borohydride in THF (60.0 mL, 60.0 mmol, 1.23 equiv.) in
a dropꢀwise manner maintaining the internal temperature between ꢀ7 and ꢀ3 °C. After
3
.5 h at ꢀ5 to 0 °C, no starting material remained as determined by TLC (70/30
hexanes/ethyl acetate). 2M Aqueous sodium carbonate (88.0 mL) was added dropꢀ
wise followed by 35% aqueous hydrogen peroxide (15.0 mL, 182.13 mmol) in water
(70.0 mL, 3.89 mol) which was also added dropꢀwise over about 10 min. MTBE (200
mL) was added, mixed for 15 min and the layers were separated. The organic layer
was washed with 150 mL each of 40% NaHSO , water, and brine. The organic layer
3
was concentrated in vacuo to a white solid which was dissolved in 400 mL hot
heptane:THF (65:35). The solution was concentrated in vacuo until solid began to
precipitate. The mixture was removed from the rotary evaporator and allowed to stand
at rt for 1 h. The white solid was filtered, washed with heptane and dried in vacuo to
give a white solid. The filtrate was concentrated then dissolved in 200 mL hot
heptane:THF (65:35). The solution was concentrated in vacuo until solid began to
precipitate. The mixture was removed from the rotary evaporator and allowed to stand
at rt for 1 h. The white solid was filtered, washed with heptane and dried in vacuo to
give a white solid. Both lots were combined to give 10 (18.5 g, 44.7 mmol, 92%) as a
20
white crystalline solid. mp 187ꢀ188 °C. [α]D ꢀ28.6 (c = 1, MeOH). MS (ES+) 414.2
+
1
[
M+H] . H NMR (500 MHz, CDCl , δ): 5.31 (bs, 1H), 4.38 (d, J = 10.5 Hz, 1H), 4.30
3
(dd, J = 11.0, 6.0 Hz, 1H), 2.68 (d, J = 15.3 Hz, 1H), 2.17 (m, 1H), 2.07 (m, 1H), 1.58
ACS Paragon Plus Environment

Page 27 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
13
(
m, 1H), 1.45 (s, 9H), 1.44 (s, 9H), 1.43 (s, 9H). C NMR (125 MHz, CDCl , δ): 175.1,
3
170.4, 155.5, 83.15, 81.31, 80.09, 73.68, 66.80, 34.33, 28.28, 28.04, 22.02. HRMS ESI
+
(
m/z): calcd. for C H NO Na [M+Na] : 436.2306. Found: 436.2307.
21
35
7
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
(
1S,2S,4S,5R,6R)-Di-tert-butyl 2-[(tert-butoxycarbonyl)amino]-4-(p-
tolylsulfonyloxy)-bicyclo[3.1.0]hexane-2,6-dicarboxylate, (11). A solution of the
alcohol 10 (5.54 g, 13.4 mmol, 1.00 equiv.) in CH Cl (54 mL), was treated with pꢀ
2
2
toluenesulfonyl chloride (2.84 g, 14.8 mmol, 1.10 equiv.) followed by triethylamine (1.90
mL, 13.6 mmol, 1.00 equiv.). The mixture was cooled in an iceꢀwater bath and 4ꢀDMAP
(2.79 g, 22.8 mmol, 1.70 equiv.) was added. After 10 min the ice bath was removed
and stirring was continued at rt overnight. The reaction was washed with 30% NaHSO4,
water and brine. The organic was dried over Na SO , filtered, and concentrated in
2
4
vacuo. The crude orange oil was purified on a 1.5 kg silica column eluting with 80/20
hexanes/acetone. Product fractions were combined and concentrated to give 11 (5.33
20
g, 9.39 mmol, 70%) as a white crystalline solid. mp 144ꢀ146 °C. [α]D ꢀ39.00 (c = 1.00,
+
1
DMSO). MS (ES+) 590.2 [M+Na] . H NMR (400 MHz, CDCl , δ): 4.97 (d, J = 5.9 Hz,
3
1
3
=
5
H), 4.95 (bs, 1H), 2.64 (dd, J = 5.8, 3.3 Hz, 1H), 2.58 Hz (d, J = 15.6 Hz, 1H), 2.43 (s,
H), 2.15 (m, 1H), 1.56 (dd, J = 16.4, 5.9 Hz, 1H), 1.47 (s, 9H), 1.40 (s, 18H), 1.38 (t, J
+
3.3 Hz, 1H). HRMS ESI (m/z): calcd. for C H NO SNa [M+Na] : 590.2394. Found:
28
41
9
90.2402.
(1S,2S,4R,5R,6R)-Di-tert-butyl 2-[(tert-butoxycarbonyl)amino]-4-
hydroxybicyclo[3.1.0]hexane-2,6-dicarboxylate, (12a). A 5 mL microwave vessel
was charged with 11 (703 mg, 1.24 mmol, 1.00 equiv.) and sodium trifluoroacetate (680
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 55
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
mg, 4.94 mmol, 3.99 equiv.). Dimethylformamide (5.0 mL) was added and the reaction
heated to 100 C for 10 h at approximately 30 watts in a microwave. No tosylate was
visible by TLC (80/20 hexanes/acetone; ninhydrin stain). Saturated aqueous NaHCO3
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 mL) was added and the reaction was stirred at rt for 2 h. The mixture was diluted
with ethyl acetate and washed with 20% saturated aqueous NaCl (3X). The organic
was dried over Na SO , filtered, and concentrated in vacuo. The crude material was
2
4
purified on an 80 g silica column eluting with a gradient from 80/20 (hold for 10 column
volumes) to 70/30 hexanes/acetone. Pure fractions were combined and concentrated
20
to give 12a as white solid (315 mg, 62%). [α]D 2.70 (c = 1.00, EtOH). MS (ES+) 436.2
+
1
[M+Na] . H NMR (400 MHz, CDCl3, δ): 5.26 (bs, 1H, NH), 4.76 (m, 1H, H4), 2.83 (bm,
1
H, H3β), 2.15 (m, 1H, H5), 2.11 (dd, J = 6.5, 2.8 Hz, 1H, H1), 2.03 (bs, 1H, OH), 1.97
(t, J = 3.2 Hz, 1H, H6), 1.45 (s, 9H, tꢀBu), 1.43 (s, 9H, tꢀBu), 1.42 (s, 9H, tꢀBu), 1.12 (dd,
13
J = 14, 8.2 Hz, 1H, H3α). C NMR (125 MHz, CDCl , δ): 171.9 (CO ester), 171.2 (CO
3
ester), 155.0 (CO Boc), 81.86 [C(CH ) ], 81.22 [C(CH ) ], 72.18 (C4), 65.01 (C2), 28.26
3 3
3 3
(
C3), 28.07 (CH ), 27.83 (C1, C5), 19.72 (C6). HRMS ESI (m/z): calcd. for
3
+
C H NO Na [M+Na] : 436.2306. Found: 436.2295.
21
35
7
(1S,2S,4R,5R,6S)-Di-tert-butyl 4-azido-2-[(tert-
butoxycarbonyl)amino]bicyclo[3.1.0]hexane-2,6-dicarboxylate, (12b). To a flask
containing 11 (8.04 g, 14.2 mmol, 1.00 equiv.) and azidotrimethylsilane (3.77 mL, 28.3
mmol, 2.0 equiv.) was added a 1M solution of tetrabutylammonium fluoride in THF (28
mL, 28 mmol, 2.0 equiv.) at rt. The mixture was heated to 75 °C. After 17 h the reaction
was cooled to rt and concentrated in vacuo to an oil. The oil was purified on silica
eluting with a gradient from 95/5 (hold for 10 column volumes) to 80/20 hexanes/ethyl
ACS Paragon Plus Environment

Page 29 of 55
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
20
acetate over 10 column volumes to give 12b (5.7 g, 92%) as a viscous oil. [α]D 70.74
+
1
(c = 1.00, CHCl ). MS (ES+) 461.4 [M+Na] . H NMR (400 MHz, CDCl3, δ): 5.23 (bs,
3
1
H, NH), 4.44 (m, 1H, H4), 2.92 (bm, 1H, H3β), 2.15 (m, 1H, H5), 2.10 (dd, J = 6.3, 3.1
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
Hz, 1H, H1), 1.92 (t, J = 3.1 Hz, 1H, H6), 1.46 (s, 9H, tꢀBu), 1.44 (s, 9H, tꢀBu), 1.43 (s,
13
9
H, tꢀBu), 1.17 (dd, J = 14, 9.2 Hz, 1H, H3α). C NMR (125 MHz, DMSO, δ): 172.1
(CO ester), 170.9 (CO ester), 155.5 (CO Boc), 81.16 [C(CH ) ], 80.75 [C(CH ) ], 78.75
3
3
3 3
[
C(CH ) ], 64.84 (C2), 60.93 (C4) 36.89 (C3), 33.12 (C1), 29.76 (C5), 28.58 [ꢀ(CH ) ],
3 3 3 3
2
8.15 [ꢀ(CH ) ], 27.90 [ꢀ(CH ) ], 20.39 (C6). HRMS ESI (m/z): calcd. for C H N O Na
3 3 3 3 21 34 4 6
+
[M+Na] : 461.2371. Found: 461.2372. Anal. calcd. for C H N O : C, H, N.
21
34
4
6
(
1R,2S,4R,5R,6R)-Di-tert-butyl 2-[(tert-butoxycarbonyl)amino]-4-
phenylthio)bicyclo[3.1.0]hexane-2,6-dicarboxylate, (12c). A solution of 11 (400 mg,
.704 mmol, 1.00 equiv.) in dimethylformamide (3 mL) was treated with potassium
(
0
carbonate (200 mg, 1.4 mmol, 2.0 equiv.) and benzenethiol (0.14 mL, 1.4 mmol, 2.0
equiv.). The reaction was heated to 70 °C in a microwave for 40 min. The reaction was
diluted with ethyl acetate and washed with water (2X) and brine. The organic was dried
over Na SO , filtered, and concentrated. The crude material was purified on a 40 g
2
4
silica column eluting with a gradient from 90/10 hexanes/ethyl acetate (hold for 5
column volumes) to 80/20 hexanes/ethyl acetate over 10 column volumes to give 12c
20
(
317 mg, 89%) as a colorless oil. [α]D 81.60 (c = 1.00, CHCl ). MS (ES+) 528.2
3
+
1
[
M+Na] . H NMR (400 MHz, CDCl , δ): 7.42 (m, 2H, ArCH), 7.25 (m, 3H, ArCH), 5.25
3
(bs, 1H, NH), 4.09 (m, 1H, H4), 3.01 (bm, 1H, H3β), 2.15 (dd, J = 5.8, 3.0 Hz, 1H, H1),
2
.07 (m, 1H, H5), 1.84 (m, 1H, H6), 1.45 (s, 9H, tꢀBu), 1.44 (s, 9H, tꢀBu), 1.42 (s, 9H, tꢀ
13
Bu), 1.21 (dd, J = 14, 10 Hz, 1H, H3α). C NMR (125 MHz, CDCl , δ): 171.6 (CO
3
ACS Paragon Plus Environment