Discovery of Aminopiperidine Indoles That Activate the Guanine Nucleotide Exchange Factor SOS1 and Modulate RAS Signaling

Article abstract of DOI:10.1021/acs.jmedchem.8b00360

Deregulated RAS activity, often the result of mutation, is implicated in approximately 30% of all human cancers. Despite this statistic, no clinically successful treatment for RAS-driven tumors has yet been developed. One approach for modulating RAS activity is to target and affect the activity of proteins that interact with RAS, such as the guanine nucleotide exchange factor (GEF) son of sevenless homologue 1 (SOS1). Here, we report on structure-activity relationships (SAR) in an indole series of compounds. Using structure-based design, we systematically explored substitution patterns on the indole nucleus, the pendant amino acid moiety, and the linker unit that connects these two fragments. Best-in-class compounds activate the nucleotide exchange process at submicromolar concentrations in vitro, increase levels of active RAS-GTP in HeLa cells, and elicit signaling changes in the mitogen-activated protein kinase-extracellular regulated kinase (MAPK-ERK) pathway, resulting in a decrease in pERK1/2^T202/Y204 protein levels at higher compound concentrations.

Full text of DOI:10.1021/acs.jmedchem.8b00360

Subscriber access provided by University of Winnipeg Library
Article
Discovery of Aminopiperidine Indoles That Activate the Guanine
Nucleotide Exchange Factor SOS1 and Modulate RAS Signaling
Jason R. Abbott, Timothy R. Hodges, R Nathan Daniels, Pratiq A. Patel, Jack Phillip Kennedy,
Jennifer E. Howes, Denis T. Akan, Michael C. Burns, Jiqing Sai, Tammy Sobolik, Yugandhar
Beesetty, Taekyu Lee, Olivia W. Rossanese, Jason Phan, Alex G. Waterson, and Stephen W. Fesik
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b00360 • Publication Date (Web): 01 Jun 2018
Downloaded from http://pubs.acs.org on June 1, 2018
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 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 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.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
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 2 of 57
1
2
3
4
5
6
7
8
9
Discovery of Aminopiperidine Indoles That Activate the
Guanine Nucleotide Exchange Factor SOS1 and Modulate
RAS Signaling
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
Jason R. Abbott,◊,# Timothy R. Hodges,◊,# R. Nathan Daniels,◊,† Pratiq A. Patel,◊,‡ J.
Phillip Kennedy,◊,⊥ Jennifer E. Howes,◊ Denis T. Akan,◊ Michael C. Burns,◊,Ω Jiqing
Sai,◊ Tammy Sobolik,◊,¶ Yugandhar Beesetty,◊ Taekyu Lee,◊ Olivia W. Rossanese,◊,£
Jason Phan,◊ Alex G. Waterson,§,¥ and Stephen W. Fesik◊,§,¥,*
◊Department of Biochemistry, Vanderbilt University School of Medicine, Nashville,
Tennessee 37232-0146, USA
§Department of Pharmacology, Vanderbilt University School of Medicine, Nashville,
Tennessee 37232-0146, USA
¥Department of Chemistry, Vanderbilt University, Nashville, Tennessee 37232-0146,
USA
ABSTRACT: Deregulated RAS activity, often the result of mutation, is implicated in
approximately 30% of all human cancers. Despite this statistic, no clinically successful
treatment for RAS-driven tumors has yet been developed. One approach for modulating
RAS activity is to target and affect the activity of proteins that interact with RAS, such as
the guanine nucleotide exchange factor (GEF) son of sevenless homologue 1 (SOS1).
ACS Paragon Plus Environment

Page 3 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Here, we report on structure–activity relationships (SAR) in an indole series of
compounds. Using structure-based design, we systematically explored substitution
patterns on the indole nucleus, the pendant amino acid moiety, and the linker unit that
connects these two fragments. Best-in-class compounds activate the nucleotide exchange
process at sub-micromolar concentrations in vitro, increase levels of active RAS-GTP in
HeLa cells, and elicit signaling changes in the mitogen-activated protein
kinase/extracellular regulated kinase (MAPK/ERK) pathway, resulting in a decrease in
pERK1/2T202/Y204 protein levels at higher compound concentrations.
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
n INTRODUCTION
Proteins in the RAS family control an array of important cellular processes,
including differentiation, proliferation, and survival.1-3 Cycling between an active GTP-
bound state and an inactive GDP-bound state, these proteins relay signals from cell
surface receptors to intracellular signaling pathways.4-8 Aberrations in RAS signaling can
confer upon cells many of the hallmarks of cancer, such as uncontrolled proliferation,
avoidance of apoptosis, induction of angiogenesis, and evasion of immune destruction.9-11
Mutationally activated RAS proteins are implicated in approximately 30% of all human
cancers, with the highest incidences occurring in tumors of the pancreas, colon, and
lung.11-13
Despite the significance of RAS in the field of oncology, the development of a
therapeutic means for targeting oncogenic RAS has met with limited success. While
direct inhibition is the most straightforward approach, accomplishing this via traditional
drug discovery strategies has proven to be a formidable challenge.14-23 An alternative
approach is to target proteins that regulate RAS activity, such as the guanine nucleotide
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 57
1
2
3
4
5
6
7
8
9
exchange factor (GEF) son of sevenless homologue 1 (SOS1).24-26 GEFs promote RAS
activation by catalyzing the exchange of GDP for intracellular GTP. During the
nucleotide exchange process, RAS engages in a protein–protein interaction with SOS1
and forms a complex (RAS:SOS1:RAS) containing one molecule of SOS1 and two
molecules of RAS—one bound to the allosteric site on SOS1 and one bound to the
catalytic site (CDC25 domain) on SOS1.27
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
Recently, our laboratory reported on compounds derived from an aminopiperidine
indole scaffold that bind to the CDC25 domain of SOS1, adjacent to the Switch II region
of RAS in the RAS:SOS1:RAS complex. This binding event stimulates nucleotide
exchange on RAS in vitro and increases the levels of RAS-GTP in cancer cells.24,28
Paradoxically, compound-mediated activation of nucleotide exchange results in the
inhibition of downstream extracellular regulated kinase (ERK1/2) phosphorylation
through negative feedback on SOS1.28
Figure 1. Mechanistic hypothesis for the action of SOS1-binding indole compounds.
ACS Paragon Plus Environment

Page 5 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In the proposed biological mechanism, depicted in Figure 1, binding of an indole
compound (SOS1 ligand) to SOS1 increases the rate of nucleotide exchange on RAS.
Consequently, the levels of active RAS-GTP increase within the cell, which amplifies
signaling down the RAS/RAF/MEK/ERK pathway, and increases the levels of
phosphorylated ERK1/2 (pERK1/2T202/Y204) protein. Via a negative feedback loop,
pERK1/2T202/Y204 then phosphorylates SOS1 at Ser1178. This phosphorylation event
disrupts the interaction between SOS1 and growth factor receptor bound protein 2
(GRB2), likely delocalizing SOS1 from the vicinity of RAS. As a result, the levels of
RAS-GTP decrease, RAS signaling is dampened, and the levels of pERK1/2T202/Y204
decrease.28
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
At the time that we first proposed this mechanism, only a few compounds from
the indole series were sufficiently potent to induce biphasic signaling changes in cells. To
further explore our mechanistic hypothesis, we required more active compounds. Here,
we describe the design and synthesis of the aminopiperidine indole compounds that were
used in our initial biological studies.24,28 We further describe the synthesis of more
advanced, structurally related analogues that demonstrate markedly improved potency in
vitro relative to the first-generation aminopiperidine indoles and, importantly, are more
active in a cellular context. We used a structure-based design approach to assess the
effects of various substituents on the indole ring system, examine an assortment of amino
acid and non-amino acid subunits, and investigate the linker unit that ties these two
pieces together. Optimized compounds activate nucleotide exchange at sub-micromolar
concentrations in vitro, increase the levels of RAS-GTP in HeLa cells, and elicit biphasic
modulation of ERK1/2 phosphorylation.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 57
1
2
3
4
5
6
n RESULTS AND DISCUSSION
7
8
9
Chemistry. The general strategies developed for compound synthesis are
described in Schemes 1–5. The synthesis of compounds devoid of substitution on the
indole nucleus (4a–k, Table 1) started from the commercially available 1H-indole
carbaldehyde 1 (Scheme 1). Reductive amination with tert-butyl piperidin-4-ylcarbamate,
followed by acid-mediated removal of the N-Boc protecting group gave amine 2.
Coupling of 2 with carboxylic acids 3a–k furnished the aminopiperidine indole analogues
4a–k shown in Table 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
Scheme 1. Synthesis of Compounds 4a–ka
H
N
NH2
R
N
N
O
O
a, b
c, b
N
H
N
H
N
H
1
2
4a−k
O
O
O
O
O
O
O
O
3a :
3g :
3b :
3c :
3d :
3e :
3f :
Me
OH
NH2
HO
HO
HO
HO
HO
HO
HO
HO
N
NH
O
O
O
3j :
3h :
3i :
3k :
HO
HO
HO
BocHN
BocHN
NH
NH
NH
N
aReagents and conditions: (a) tert-Butyl piperidin-4-ylcarbamate, NaBH(OAc)3, DCM.
(b) TFA, DCM. (c) 3a–k, PyBOP, i-Pr2NEt, DMSO or 3a–k, HATU, i-Pr2NEt, DMF.
Analogues substituted on the indole ring were prepared as shown in Scheme 2.
The substituted 1H-indole carbaldehydes 5a–k were advanced through a four step
sequence comprising: (1) reductive amination, (2) N-Boc deprotection, (3) coupling with
(tert-butoxycarbonyl)-L-tryptophan, and (4) N-Boc deprotection. The aminopiperidine
indole analogues 7a–k depicted in Table 2 were thus obtained.
ACS Paragon Plus Environment

Page 7 of 57
Journal of Medicinal Chemistry
1
2
3
4
Scheme 2. Synthesis of Compounds 7a–ka
5
6
7
8
a : R = 7-azaindole
b : R = 2-Me
c : R = 4-Cl
H
NH2
NH2
N
d : R = 5-Cl
NH
N
N
O
4
7
e : R = 6-Cl
O
5
6
9
a, b
c, b
f : R = 7-Cl
R
2
R
R
g : R = 5-OMe
h : R = 5-Me
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
N
H
N
H
N
H
5a−k
6a−k
7a−k
: R = 5-F
: R = 5-Br
: R = 5-CF
i
j
k
3
aReagents and conditions: (a) tert-Butyl piperidin-4-ylcarbamate, NaBH(OAc)3, DCM.
(b) TFA, DCM or 4 M HCl in dioxane, DCM. (c) (tert-Butoxycarbonyl)-L-tryptophan,
PyBOP, i-Pr2NEt, DMSO or (tert-Butoxycarbonyl)-L-tryptophan, HATU, i-Pr2NEt,
DMF.
The synthesis of indole analogues featuring substituted tryptophan moieties (12a–
h, Table 3) was enabled by chemistry developed by Jia and Zhu (Scheme 3).29
Commercially available anilines 8a–h were elaborated to the tryptophan methyl esters
10a–h via palladium-catalyzed annulation with (S)-2-N,N-di-(tert-butoxycarbonyl)-5-
oxopentanoate 9. Hydrolysis of methyl esters 10a–h gave the corresponding carboxylic
acids 11a–h, which were advanced to analogues 12a–h via coupling with amine 6d and
deprotection of the two N-Boc protecting groups from the tryptophan subunit.
Scheme 3. Synthesis of Compounds 12a–ha
a : R' = 5'-OMe
O
b : R' = 6'-OMe
c : R' = 5'-CF3
d : R' = 6'-CF3
e : R' = 5'-Me
f : R' = 6'-Me
g : R' = 5'-Cl
H
OR
NH2
N
NH
5'
N
N(Boc)2
Cl
I
O
5'
6'
5'
6'
a
c, d
R'
R'
6'
R'
N
H
N
H
NH2
: R' = 6'-Cl
h
8a−h
10a−h : R = Me
11a−h : R = H
12a−h
b
O
O
H
OMe
N(Boc)2
9
aReagents and conditions: (a) 9, Pd(OAc)2, DABCO, DMF, 85 °C. (b) KOH,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 57
1
2
3
4
THF/EtOH/H2O (1:1:1 v/v/v), 50 °C. (c) 6d, HATU, i-Pr2NEt, DMF. (d) TFA, DCM.
5
6
7
8
9
For the synthesis of compounds containing various linker units (15a–f, Table 4
and 16–19, Table 5), 5-chloro-1H-indole-3-carbaldehyde 5d was subjected to reductive
amination with mono-N-Boc-diamines 13a–f, and the products treated with TFA to give
subunits 14a–f (Scheme 4). Coupling of 14a–f with (tert-butoxycarbonyl)-L-tryptophan,
followed by N-Boc deprotection gave the unsubstituted tryptophan–indoles 15a–f
displayed in Table 4. Similarly, compounds 16–19, which comprise the final matrix
depicted in Table 5, were prepared via the union of amines 14d and 14f with carboxylic
acids 11f and 11h (Scheme 5).
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 4. Synthesis of Compounds 15a–fa
H
N
H2N
linker
linker
N
N
NH
N
O
O
Cl
Cl
Cl
a, b
c, b
N
H
N
H
N
H
5d
14a−f
15a−f
NHBoc
13a :
13d :
13c :
HN
HN
HN
13b :
13e :
HN
HN
NBoc
NBoc
H
N
NBoc
13f :
NBoc
NBoc
aReagents and conditions: (a) 13a–f, NaBH(OAc)3, DCM. (b) TFA, DCM. (c) (tert-
Butoxycarbonyl)-L-tryptophan, PyBOP, i-Pr2NEt, DMSO.
Scheme 5. Synthesis of Compounds 16–19a
H
N
6'
H2N
R'
16 : R' = 6'-Me
17 : R' = 6'-Cl
N
N
linker
linker
N
N
NH
N
O
Cl
Cl
18 : R' = 6'-Me
19 : R' = 6'-Cl
a, b
N
N
N
H
N
H
14d, 14f
16−19
ACS Paragon Plus Environment

Page 9 of 57
Journal of Medicinal Chemistry
1
2
3
4
aReagents and conditions: (a) 11f or 11h, HATU, i-Pr2NEt, DMF. (b) TFA, DCM.
5
6
7
8
9
Compound Design. Aminopiperidine Indole Extensions. The first set of
analogues was designed to evaluate extensions from the exocyclic amine of the
aminopiperidine unit initially identified as the minimum pharmacophore necessary for
nucleotide exchange activation. Accordingly, a variety of amino acid and non-amino acid
aminopiperidine amides—a selection of which are shown in Table 1—were synthesized.
The ability of these compounds to affect SOS1-mediated nucleotide exchange was
assessed using a high throughput nucleotide exchange assay.24,30 Compound efficacy in
vitro was expressed in terms of maximal percent activation (Max. Act. %), defined as the
maximum exchange rate elicited by an individual compound at 100 µM relative to the
DMSO vehicle control alone. We hypothesized that compounds should elicit a marked
increase in the maximum rate of nucleotide exchange in order to modulate RAS-GTP
signaling and achieve enhanced potency in cancer cells, as previously discussed.30 In
vitro compound potency was defined as the half maximal effective concentration (EC50),
calculated from the concentration response curve (0–100 µM compound) generated from
the nucleotide exchange assay.
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
Table 1. SAR of the Aminopiperidine Extensiona
Max.
Compd
2
Structure
EC50 (µM)b
Act. (%)c
NH2
N
—
222 ± 52.2
N
H
H
N
N
Me
O
4a
4b
—
—
157 ± 38.8
200 ± 62.0
N
H
H
N
N
OH
O
N
H
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 57
1
2
3
4
5
6
7
8
9
H
N
N
NH2
O
4c
4d
4e
4f
—
225 ± 48.1
323 ± 222.0
110 ± 4.7
N
H
H
N
N
O
—
N
H
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
N
N
O
—
N
H
H
N
N
O
—
263 ± 134.4
67 ± 7.2
N
H
H
N
N
N
N
N
N
N
O
4g
4h
4i
—
N
H
H
N
O
—
170 ± 20.6
585 ± 304.3
311 ± 90.8
641 ± 225.3
NH
N
H
H
N
NH
O
—
—
N
H
H
NH2
N
NH
O
4j
N
H
H
NH2
N
NH
O
4k
89.7 ± 12.34
N
H
aValues reported as the mean ± SD of at least two separate experiments.
b “—” denotes an EC50 value of >100 µM, or an EC50 value that was not calculated due to low efficacy in vitro.
cActivation values represent percentage relative to DMSO control at a ligand concentration of 100 µM.
Most of the initial analogues, exemplified by 4a–c in Table 1, only weakly
stimulated SOS1-catalyzed nucleotide exchange at the maximum evaluated ligand
concentration (100 µM), and reliable EC50 values could not be derived from the resulting
concentration response curves. However, many of these derivatives possessed adequate
solubility profiles for use in X-ray crystallographic studies, from which the X-ray
structure of glycine analogue 4c bound to the RAS:SOS1:RAS complex was obtained
(Figure 2).24
ACS Paragon Plus Environment

Page 11 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. X-ray co-crystal structure of compound 4c (violet; PDB ID code 4NYJ) bound
to SOS1 (aqua) in the RAS:SOS1:RAS ternary complex. Yellow circle highlights
hydrophobic subpocket above Phe890, near Ile856.
As illustrated in Figure 2, a hydrogen bond between the indole N–H and the
backbone carbonyl of Met878 anchors compound 4c deep within the hydrophobic
binding pocket of SOS1. The aminopiperidine linker extends outward, orienting the
terminal glycine moiety toward the solvent. In the crystallized binding pose, compound
4c leaves a nearby hydrophobic subpocket above Phe890 and next to Ile856 unoccupied
(Figure 2, yellow circle). We hypothesized that increasing the ligand’s occupation of this
region would potentially enhance the potency of these early stage compounds. Thus,
several aminopiperidine amides featuring aromatic side-chains were synthesized (4d–k,
Table 1). The majority of these new derivatives showed no measureable improvement in
the nucleotide exchange assay. However, compound 4k, which features a tryptophan
subunit, activated the nucleotide exchange process to a desirably high maximal rate as
compared to the other compounds tested in Table 1 (Max. Act. = 641%), and with an
EC50 of 89.7 µM.24 Although similar to 4k in structure, compounds 4i and 4j—the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 57
1
2
3
4
5
6
deamino and enantiomeric analogues of compound 4k, respectively—showed reduced
nucleotide exchange activity.
7
8
9
To guide the next round of optimization, an X-ray structure of compound 4k
bound to SOS1 in the RAS:SOS1:RAS complex was obtained (Figure 3).24 Similar to the
glycine analogue 4c, compound 4k retains a hydrogen bond with the backbone carbonyl
group of Met878. The aminopiperidine linker extends out from the binding pocket and
folds back on itself, positioning the tryptophan arene ring close to the hydrophobic
subpocket formed by Ile856, Phe890, and Ile893. In the crystallized binding pose, the
tryptophan N–H sits in close proximity to Glu902, and engages in a hydrogen bond with
the carboxylate side-chain of this residue.
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. X-ray co-crystal structure of compound 4k (gold; PDB ID code 4NYM) bound
to SOS1 (aqua) in the RAS:SOS1:RAS ternary complex.
SAR of the Indole Ring System. The X-ray co-crystal structure shown in Figure 3,
combined with the data in Table 1, suggested that additional improvements in potency
could be realized by refining three areas of the compound: (1) the indole anchor, (2) the
aromatic ring of the tryptophan subunit, and (3) the aminopiperidine linker connecting
these two pieces. In an attempt to fill the unoccupied space around the indole ring system,
ACS Paragon Plus Environment

Page 13 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
analogues with substituents at the 2-, 4-, 5-, 6-, and 7-positions of the indole nucleus were
prepared (Table 2). Within this iteration of compounds, a preference for substitution at C-
5 of the indole ring system emerged. This is exemplified by the 9-fold improvement in
potency of the 5-chloroindole 7d over the 5H-indole 4k. Conversely, 4-, 6-, and 7-
chloroindoles—7c, 7e, and 7f—demonstrated reduced potency and diminished efficacy
compared to compound 4k.
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
In addition to compound 7d, other 5-substituted indoles demonstrated improved
potency relative to the 5H-indole 4k in the nucleotide exchange assay. These include the
5-methyl analogue 7h, the 5-bromo analogue 7j, and the 5-trifluoromethyl analogue 7k,
which exhibited EC50 values of 29.2, 14.2, and 11.2 µM, respectively. Consistent with the
hydrophobic nature of the binding pocket, the 5-methoxy derivative 7g and the azaindole
7a demonstrated reduced activity. Finally, the 2-methylindole 7b was also only weakly
active, suggesting that substitution at this position was not well tolerated.
Table 2. SAR of the Indole Ring Systema
Max.
Compd
4k
Structure
EC50 (µM)b
Act. (%)c
H
N
NH2
NH
N
4
7
89.7 ± 12.3431
641 ± 225.3
O
5
6
2
N
H
H
NH2
N
NH
NH
NH
N
O
7a
7b
7c
—
—
86 ± 3.6
159 ± 13.9
259 ± 60.6
>1000
N
H
N
H
NH2
N
N
O
Me
N
H
H
NH2
N
Cl
N
O
—
N
H
H
NH2
N
NH
N
Cl
O
7d
9.6 ± 4.27
N
H
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 57
1
2
3
4
5
6
7
8
9
H
NH2
N
NH
N
O
7e
7f
—
—
250 ± 53.0
146 ± 28.6
N
Cl
H
H
NH2
N
NH
N
O
N
H
Cl
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
NH2
N
NH
N
MeO
Me
F
O
7g
7h
7i
—
813 ± 429.4
>1000
N
H
H
NH2
N
NH
N
O
29.2 ± 5.39
15.9 ± 2.30
14.2 ± 3.58
11.2 ± 0.25
N
H
H
NH2
N
NH
N
O
158 ± 15.9
>1000
N
H
H
NH2
N
NH
N
Br
O
7j
N
H
H
NH2
N
NH
N
F3C
O
7k
407 ± 24.1
N
H
aValues reported as the mean ± SD of at least two separate experiments.
b “—” denotes an EC50 value of >100 µM, or an EC50 value that was not calculated due to low efficacy in vitro.
cActivation values represent percentage relative to DMSO control at a ligand concentration of 100 µM.
To explore other areas of the molecule, we chose to fix the indole scaffold upon
which to build further SAR. The 5-chloroindole 7d was chosen for its improved potency
and efficacy relative to the other compounds in Table 2, and for its greater ligand
efficiency (LE) and lipophilic efficiency (LipE) (LE = 0.22; LipE = 1.62) relative to other
5-substitued indoles, such as the 5-bromoindole 7j (LE = 0.21; LipE = 1.37) and the 5-
(trifluoromethyl)indole 7k (LE = 0.20; LipE = 1.28).32-34
Prior to the discovery of compound 7d, activation values in the nucleotide
exchange assay were reported as maximal percent activation (Max. Act. %), relative to
the DMSO vehicle control. Although this format was useful for identifying hit
compounds in the nucleotide exchange assay, a degree of variability in the baseline
activation rate was noted between different experiments. Thus, to allow for more reliable
ACS Paragon Plus Environment

Page 15 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
comparison of activation values between plates, compound 7d was included as an
internal control in every subsequent nucleotide exchange experiment. Compound efficacy
was then expressed in terms of relative nucleotide exchange activation (Rel. Act. %),
which was calculated as the percentage of maximal activation elicited by a given
compound at 100 µM, normalized to the maximal activation elicited by compound 7d at
100 µM. As discussed in a recent publication, compounds that achieved greater than
approximately 50% relative activation in vitro elicited a more robust induction of RAS-
GTP levels in a cellular context.30 This threshold was subsequently used to predict
whether or not a given compound was likely to produce activity in cells.
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
SAR of the Tryptophan Ring System. With the identity of the indole anchor
established and the reporting of activation values standardized, the focus shifted to the
optimization of the tryptophan subunit and the aminopiperidine linker. These two areas of
chemical space were investigated in parallel using the data for compound 7d as a baseline
for comparison. Analysis of the X-ray co-crystal structure of compound 4k shown in
Figure 3 suggested that the tryptophan moiety did not adequately fill the hydrophobic
subpocket above Phe890, near Ile856. This area did, however, appear to be accessible
from the aromatic ring of the tryptophan nucleus. As such, a series of substituted
tryptophan derivatives were prepared and tested in the nucleotide exchange assay. As
shown in Table 3, compounds 12a and 12c–h—substituted at either the 5’- or 6’-position
of the tryptophan ring—all showed improved potency in vitro relative to the parent
compound 7d, while also achieving similar levels of efficacy. Within this series of
analogues, the data for matched pairs of C-5’- and C-6’-substituted tryptophans indicated
a preference for substitution at the 6’-position (12c versus 12d; 12e versus 12f; and 12g
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 57
1
2
3
4
5
6
versus 12h). Most notable was compound 12h, which exhibited an EC50 value of 1.4
µM—7-fold more potent than its predecessor 7d.
7
8
9
Table 3. SAR of the Tryptophan Ring Systema
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
Rel.
Compd
7d
Structure
EC50 (µM)
Act. (%)b
H
NH2
N
NH
5'
N
Cl
7'
6'
O
9.6 ± 4.27
100
4'
N
H
H
N
NH2
NH
N
Cl
O
12a
12b
12c
12d
12e
12f
8.5 ± 0.21
15.1 ± 8.66
7.8 ± 4.76
6.1 ± 2.45
5.9 ± 1.96
3.1 ± 1.04
1.6 ± 0.40
1.4 ± 0.09
105 ± 20.8
87 ± 24.6
72 ± 31.3
105 ± 28.1
104 ± 32.7
120 ± 29.5
87 ± 19.7
107 ± 36.4
N
H
OMe
H
NH2
N
NH
N
Cl
O
OMe
CF3
Me
N
H
H
NH2
N
NH
N
Cl
O
N
H
CF3
H
NH2
N
NH
N
Cl
O
N
H
H
NH2
N
NH
N
Cl
O
N
H
Me
H
NH2
N
NH
N
Cl
O
N
H
H
NH2
N
NH
N
Cl
O
12g
12h
N
H
Cl
H
NH2
N
NH
N
Cl
O
Cl
N
H
aValues reported as the mean ± SD of at least two separate experiments.
bActivation values calculated as the percentage activation for each compound at 100 µM relative to the activation of control compound
7d at 100 µM.
From a structural perspective, the preference for substitution at the 6’-position of
the tryptophan ring system can be rationalized by overlaying the X-ray co-crystal
structures of the regioisomeric matched pairs 12e versus 12f and 12g versus 12h bound to
SOS1 in the RAS:SOS1:RAS ternary complex (Figure 4a,b). As shown in Figure 4a, the
ACS Paragon Plus Environment

Page 17 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
chlorine atom at the 6’-position of the tryptophan nucleus in compound 12h fits into the
hydrophobic pocket near Ile856 (Figure 4a–c, yellow circle). Conversely, the chlorine
atom of its 5’-regioisomer 12g floats above this pocket and is not well ordered in the
crystal structure. Figure 4b shows the same observation to be true for the methyl
analogues 12e and 12f, however the 6’-methyl group on compound 12f does extend as
deeply into the binding pocket as the corresponding 6’-chlorine atom on compound 12h
(Figure 4c). In addition to more adequately filling the hydrophobic space near Ile856, the
6’-chlorotryptophan 12h engages several residues in the binding pocket, including
Met878 via a direct H–bond, and Asp887 via two separate water-mediated interactions
(Figure 4d).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a)
b)
c)
d)
Figure 4. (a) Overlay of compounds 12g (salmon; PDB ID code 6BVJ) and 12h (purple;
PDB ID code 6BVI). (b) Overlay of compounds 12e (cyan; PDB ID code 6BVL) and 12f
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 57
1
2
3
4
5
6
7
8
9
(orange; PDB ID code 6BVK). (c) Overlay of compounds 12f (orange) and 12h (purple).
Yellow circle highlights the hydrophobic pocket near Ile856. (d) X-ray co-crystal
structure of compound 12h (purple) bound to SOS1 (aqua) in the RAS:SOS1:RAS
ternary complex.
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
Aminopiperidine Replacements. In parallel to the compounds in Table 3,
analogues of compound 7d were designed in which the 5-chloroindole and the tryptophan
subunits were connected through a variety of linker moieties. As shown earlier in Figure
3, the aminopiperidine ring of compound 4k adopts a U-shaped configuration upon
binding to SOS1, which is approximately 34 kcal/mol higher in energy than the ground
state conformation of the ligand, calculated using water as the solvent (see Supporting
Information). We hypothesized that reinforcing the binding conformation of the ligand
would potentially reduce the entropic penalty of binding.35-37 Thus, a focused set of
compounds featuring rigidified cyclic and bicyclic amine linkers was synthesized (Table
4), exemplified by compound 15d, for which the difference in energy between the ground
state conformation and a conformation resembling that of the bound ligand is
approximately 29 kcal/mol, calculated using water as the solvent (see Supporting
Information).
The compounds featuring rigidified linker units were then tested in the nucleotide
exchange assay, where analogues 15d–f proved to be more potent than the comparator
compound 7d (Table 4). On the other hand, compounds 15b and 15c, both of which
feature the 2,7-diazaspiro[3.5]nonane linker, were less potent and less efficacious in vitro
than compound 7d. Finally, introducing additional flexibility into the linker unit (e.g.
compound 15a) did not greatly affect compound activity.
ACS Paragon Plus Environment

Page 19 of 57
Journal of Medicinal Chemistry
1
2
3
4
Table 4. SAR of the Diamine Linker Unita
5
6
7
8
Rel.
Compd
7d
Structure
EC50 (µM)b
Act. (%)c
H
NH2
N
NH
N
Cl
O
9.6 ± 4.27
100
9
N
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
H
N
H2N
15a
15b
13.3 ± 5.69
82 ± 40.9
N
N
N
N
N
H
O
Cl
N
H
H
N
H2N
N
—
44 ± 15.3
35 ± 9.2
76 ± 3.1
O
Cl
Cl
Cl
N
H
H
N
H2N
N
15c
15d
37.3 ± 18.36
O
N
H
H
N
H2N
N
6.7 ± 2.61
O
N
H
H
N
NH2
N
N
O
15e
15f
Cl
4.5 ± 1.84
2.9 ± 0.29
150 ± 37.2
97 ± 34.1
N
H
N
Cl
N
N
NH2
H
O
NH
aValues reported as the mean ± SD of at least two separate experiments.
b “—” denotes an EC50 value of >100 µM, or an EC50 value that was not calculated due to low efficacy in vitro.
cActivation values calculated as the percentage activation for each compound at 100 µM relative to the activation of control compound
7d at 100 µM.
As Figure 5a illustrates, the octahydropyrrolo[3,4-c]pyrrole ring system within
compound 15d orients the tryptophan moiety into the back of the binding pocket in a
similar manner as the aminopiperidine linker of the initial lead compound 4k. In the
crystallized binding pose, compound 15d engages in a water-mediated interaction with
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 57
1
2
3
4
5
6
7
8
9
Tyr884 and Asp887, and a water-mediated interaction with the backbone carbonyl of
Phe890 (Figure 5b). Together, these interactions, coupled with the rigidified ligand
conformation, likely explain the improved potency of compound 15d vis-à-vis compound
4k.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a)
b)
Figure 5. (a) Overlay of compounds 15d (marine; PDB ID code 6BVM) and 4k (gold;
PDB ID code 4NYM). (b) X-ray co-crystal structure of compound 15d (marine) bound to
SOS1 (aqua) in the RAS:SOS1:RAS ternary complex.
Having realized potency improvements by adding substituents to the tryptophan
ring system and by changing the aminopiperidine linker separately, we sought to combine
these two structural modifications in an effort to achieve further improvements in
compound activity. Towards this end, the compounds shown in Table 5 were synthesized.
These compounds each feature one of the preferred 6’-substituted tryptophan motifs
combined with one of the preferred linker units. Incorporation of these two design
elements into a single molecule generated the most potent compounds to date in the
indole series. Compounds 17–19 activated SOS1-catalyzed nucleotide exchange above
the relative activation threshold and with EC50 values below 1.0 µM.
Table 5. SAR of Optimized Compoundsa
ACS Paragon Plus Environment

Page 21 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Rel.
EC50 (µM)
Compd
16
Structure
Act. (%)b
H
Me
N
H2N
N
4.4 ± 1.61
0.8 ± 0.47
123 ± 8.8
N
O
Cl
N
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
H
Cl
N
H2N
N
17
18
19
119 ± 16.3
N
O
Cl
N
H
N
Cl
N
N
NH2
H
0.8 ± 0.09
105 ± 32.8
O
NH
Me
N
Cl
N
N
NH2
H
0.8 ± 0.36
78 ± 23.3
O
NH
Cl
aValues reported as the mean ± SD of at least two separate experiments.
bActivation values calculated as the percentage activation for each compound at 100 µM relative to the activation of control compound
7d at 100 µM.
Biological Evaluation. In order to determine whether the improvements in
biochemical potency translated to enhanced cellular potency, the cellular effects on RAS-
GTP and pERK1/2T202/Y204 protein levels that were elicited by four compounds with a
range of EC50 values were measured (Figure 6). To facilitate comparison between these
compounds—12h, 15c, 17, and 19—quantification values for RAS-GTP and
pERK1/2T202/Y204 levels normalized to their respective controls are provided below each
blot. These numbers are also provided in the Supporting Information (Tables S1 and S2),
along with quantification data for compound 7d, which was used in previous mechanistic
studies, and served as the control compound in the nucleotide exchange assay (vide
supra).24,28 Importantly, the trends in RAS-GTP and pERK1/2T202/Y204 modulation were
consistent between independent biological replicates (Tables S1 and S2).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 57
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 6. RAS-GTP and corresponding pERK1/2T202/Y204 levels from HeLa cells that
were treated for 30 min with up to 50 µM of compounds 12h, 15c, 17, or 19. EGF
treatment (50 ng/mL for 5 min) was used as a positive control for pathway activation.
Quantification values for RAS-GTP levels and pERK1/2T202/Y204 levels are displayed
under their respective blots for each compound. The levels of RAS-GTP were normalized
to the total RAS protein levels and to the DMSO control. The levels of pERK1/2T202/Y204
were normalized to the total ERK1/2 protein levels and to the DMSO control. Data are
representative of two independent experiments.
ACS Paragon Plus Environment

Page 23 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Previously, it was reported that compound 7d (EC50 = 9.6 µM) stimulates
maximum production of RAS-GTP at 50 µM, after a 30 min treatment in HeLa cells.24,28
Relative to compound 7d, compounds 12h, 17, and 19 exhibit lower EC50 values in vitro
(EC50 = 1.4, 0.8, and 0.8 µM, respectively). As illustrated in Figure 6, in HeLa cells,
compounds 12h, 17, and 19 cause dose-dependent increases in RAS-GTP at treatment
concentrations as low as 3.1 µM, and a peak in RAS-GTP levels at 25 µM. In line with
the biochemical assay results, which showed compound 15c to be less potent (EC50 =
37.3 µM) than compounds 12h, 17, and 19, higher treatment concentrations of compound
15c were required to activate RAS-GTP production.
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
In addition to modulating the levels of RAS-GTP, compounds 12h, 17, and 19
also reduced the levels of pERK1/2T202/Y204 below baseline at concentrations of 25–50
µM (Figure 6). On the other hand, neither compound 7d nor compound 15c inhibited
ERK1/2 phosphorylation over the concentration range tested (0–50 µM, Table S2). Thus,
compounds 12h, 17, and 19 achieved a higher potency of pERK1/2T202/Y204 inhibition
than both compounds 7d and 15c.
Taken together, these findings further articulate our recently proposed mechanism
for the modulation of RAS signaling, wherein compound-induced increases in the levels
of RAS-GTP produce biphasic modulation of ERK1/2 phosphorylation via negative
feedback on SOS1 by pERK1/2T202/Y204 28
The results displayed in Figure 6 (and in
.
Tables S1 and S2) also suggest that improvements in biochemical potency generated
through SAR optimization led to compounds with enhanced potency for RAS-GTP
activation and pERK1/2T202/Y204 inhibition in HeLa cells.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 57
1
2
3
4
n CONCLUSION
5
6
7
8
In this study, we have described the structure-based design and synthesis of a series of
SOS1-binding indole compounds. By systematically exploring three different areas of the
initial lead molecule 4k—the arene ring of the indole, the amino acid extension, and the
linker unit—we discovered the most potent compounds in this indole series, 17–19.
These compounds exhibit sub-micromolar activation of SOS1-catalyzed nucleotide
exchange on RAS in vitro. In HeLa cells, compounds 17 and 19 increase the levels of
RAS-GTP and cause biphasic modulation of ERK1/2 phosphorylation, characterized by
an increase in the levels of pERK1/2T202/Y204 at lower compound concentrations and a
decrease in the levels of pERK1/2T202/Y204 below baseline at higher compound
concentrations. Notably, potency improvements realized in vitro translated to enhanced
potency in HeLa cells, relative to previously reported compounds from this series.24
These findings are consistent with our recent mechanistic studies demonstrating that
compound-mediated activation of RAS results in biphasic modulation of ERK1/2
phosphorylation through negative feedback on SOS1; thus, it is likely that these
compounds modulate RAS signaling through the same biological mechanism.28 This
series of indole-derived compounds provides a unique opportunity to explore SOS1-
mediated RAS-dependent signaling and to establish the therapeutic implications of
modulating RAS activity through SOS1 in cancer cells.
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
n EXPERIMENTAL SECTION
General Procedures. All chemical reagents and reaction solvents were purchased
from commercial suppliers and used as received. Proton nuclear magnetic resonance (1H
ACS Paragon Plus Environment

Page 25 of 57
Journal of Medicinal Chemistry
1
2
3
4
NMR) spectra were recorded at either 400 MHz or 600 MHz on a Bruker spectrometer.
5
6
1
For H NMR spectra, chemical shifts are reported in parts per million (ppm) and are
7
8
9
reported relative to residual non-deuterated solvent signals. Coupling constants are
reported in hertz (Hz). The following abbreviations (or a combination, thereof) are used
to describe splitting patterns: s, singlet; d, doublet; t, triplet; q, quartet; quint, quintet; m,
multiplet; comp, overlapping multiplets of non-magnetically equivalent protons; br,
broad. All compounds were of 95% purity or higher, unless otherwise noted, as measured
by analytical reversed-phase HPLC. Analytical HPLC was performed on an Agilent 1200
series system with UV detection at 214 and 254 nm, along with evaporative light
scattering detection (ELSD). Low-resolution mass spectra were obtained on an Agilent
6140 mass spectrometer with electrospray ionization (ESI). LC-MS experiments were
performed with the following parameters: Phenomenex Kinetex 2.6 µm XB-C18 100 Å,
LC column 50 x 2.1 mm; 2 min gradient, 5%–95% MeCN in H2O, and 0.1% TFA or
0.1% formic acid. Analytical thin layer chromatography (TLC) was performed on
Kieselgel 60 F254 glass plates precoated with a 0.25 mm thickness of silica gel. TLC
plates were visualized with UV light and iodine. Silica gel chromatography was
performed using a Teledyne Isco Combiflash® Rf system. Preparative reversed-phase
HPLC was performed on a Gilson instrument equipped with a Phenomenex Kinetex C18
column, using varying concentrations of MeCN in H2O, and 0.1% TFA.
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 synthesis of and characterization data for compounds 2, 4c, 4k, 7d, and 7h
has been reported previously.24
General Procedure 1—Reductive Amination of 1H-Indole Carbaldehydes 1
and 5a–k for the Synthesis of Amines 2, 6a–k, and 14a–f
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 57
1
2
3
4
5
STEP 1: Similar to the previously reported procedure,24 a mono-N-Boc-diamine (1.00
equiv) and a 1H-indole carbaldehyde, such as 1 or 5a–k (1.10 equiv), were dissolved in
6
7
8
9
DCM (0.3 M). The solution was stirred for 0.25–3
h
before sodium
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
triacetoxyborohydride (NaBH(OAc)3) (2.00 equiv) was added. The progress of the
reaction was monitored by LC-MS analysis. When the reaction was complete, the
mixture was diluted with saturated aqueous sodium bicarbonate (NaHCO3) until pH ≈ 7
was achieved. The resulting mixture was diluted with water (H2O) and the organic layer
was collected. The aqueous phase was extracted with three portions of DCM. The organic
phases were combined, concentrated in vacuo, and purified by silica gel chromatography
to give the N-Boc-(1H-indol-3yl)methanamines.
STEP 2: A mixture of an N-Boc-(1H-indol-3yl)methanamine (1.00 equiv) and DCM/TFA
(9:1 v/v) (0.2 M) or DCM/4 M HCl in dioxane (3:1 v/v) (0.1 M) was stirred at room
temperature. The progress of the reaction was monitored by LC-MS analysis. When the
reaction was complete, the mixture was concentrated in vacuo. The residue was either
azeotroped with toluene three times and the product amine 2, 6a–k, or 14a–f used
directly as its corresponding salt, or dissolved in methanol, rotated with MP-carbonate,
filtered, and concentrated to give the amine 2, 6a–k, or 14a–f as its free base.
General Procedure 2—Synthesis of Compounds 4a–b, 4i–j, 7c, 7e–g, 7i–k, and 15a–f
STEP 1: An amine or amine salt, such as 2, 6a–k, or 14a–f (1.00 equiv), and
(benzotriazol-1-yloxy)tripyrrolidinophosphonium hexafluorophosphate (PyBOP) (1.10
equiv) were dissolved in DMSO (0.25 M). The solution was stirred at room temperature.
A carboxylic acid (1.00 equiv) and N,N-diisopropylethylamine (i-Pr2NEt) (2.00 equiv)
ACS Paragon Plus Environment

Page 27 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
were added. The progress of the reaction was monitored by LC-MS analysis. When the
reaction was complete, the mixture was directly purified by reversed-phase preparative
HPLC. The target compound was then obtained as its free base by passage through an
Isolute® SCX-2 cartridge or by treatment with MP-carbonate.
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
STEP 2: If necessary, a mixture of the coupled product obtained in STEP 1 (1.00 equiv)
and DCM/TFA (9:1 v/v) (0.2 M) was stirred at room temperature. The progress of the
reaction was monitored by LC-MS analysis. When the reaction was complete, the
mixture was concentrated in vacuo. The residue was purified by reversed-phase
preparative HPLC. The target compound was then obtained as its free base by passage
through an Isolute® SCX-2 cartridge or by treatment with MP-carbonate.
General Procedure 3—Synthesis of Compounds 4d–h and 7a–b
STEP 1: A carboxylic acid (1.00 equiv) was dissolved in DMF (0.3 M). Next, 1-
[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium
3-oxid
hexafluorophosphate (HATU) (1.10 equiv) and i-Pr2NEt (3.00 equiv) were added. The
resulting mixture was stirred for 15 min at room temperature before a solution of an
amine or amine salt, such as 2, 6a–k, or 14a–f (1.0 M in DMF) (1.10 equiv), was added.
After 18 h, the reaction mixture was diluted with ethyl acetate (EtOAc) and washed
successively with 1 M citric acid, saturated aqueous NaHCO3, and brine. The organic
phase was dried over sodium sulfate (Na2SO4) and concentrated in vacuo. The residue
was purified by reversed-phase preparative HPLC. The target compound was then
obtained as its free base by passage through an Isolute® SCX-2 cartridge.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 57
1
2
3
4
5
6
7
8
9
STEP 2: If necessary, a mixture of the coupled product obtained in STEP 1 and DCM/4
M HCl in dioxane (3:1 v/v) (0.1 M) was stirred at room temperature. The progress of the
reaction was monitored by LC-MS analysis. When the reaction was complete, the
mixture was concentrated in vacuo. The residue was purified by reversed-phase
preparative HPLC. The target compound was then obtained as its free base by passage
through an Isolute® SCX-2 cartridge.
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
General Procedure 4—Synthesis of Compounds 12a–h and 16–19
STEP 1: According to the procedure reported by Jia and Zhu,29 (S)-2-N,N-di-tert-
butoxycarbonyl-5-oxopentanoate 9 (1.00 equiv), an aniline, such as 8a–h (1.10 equiv),
palladium(II) acetate (Pd(OAc)2) (0.10 equiv), and DABCO (3.00 equiv) were added to a
reaction vial. The vial was sealed, evacuated, and refilled with nitrogen (N2). The
evacuation/refill cycle was repeated two additional times. Degassed DMF (0.3 M) was
then added and the reaction mixture was heated to 85 °C. After 18 h, the reaction mixture
was cooled to room temperature and then diluted with H2O and EtOAc. The resulting
emulsion was treated with sodium chloride (NaCl) and the layers were separated. The
aqueous phase was extracted with two portions of EtOAc. The organic phases were
combined, washed with brine, dried over Na2SO4, and concentrated in vacuo. The residue
was purified by silica gel chromatography to provide the tryptophan methyl esters 10a–h.
STEP 2: A tryptophan methyl ester, such as 10a–h (1.00 equiv), was dissolved in a
mixture of THF/ethanol (EtOH)/H2O (1:1:1 v/v/v) (0.05 M). Powdered potassium
hydroxide (KOH) (8.00 equiv) was added and the mixture was heated to 50 °C. After 6 h,
the mixture was cooled to room temperature and concentrated in vacuo. The residue was
ACS Paragon Plus Environment

Page 29 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
partitioned between H2O and diethyl ether (Et2O). The organic layer was discarded and
the aqueous phase was acidified to pH ≈ 4–5 with 2.5 M citric acid before being extracted
with three portions of EtOAc. The organic phases were combined, washed with brine,
dried over Na2SO4, and concentrated in vacuo. The residue was purified by silica gel
chromatography to give the tryptophan carboxylic acids 11a–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
STEP 3: A tryptophan carboxylic acid, such as 11a–h (1.00 equiv), was dissolved in
DMF (0.4 M). Next, HATU (1.10 equiv) and i-Pr2NEt (3.00 equiv) were added. The
resulting mixture was stirred for 15 min at room temperature before a solution of an
amine or amine salt, such as 2, 6a–k, or 14a–f (0.2 M in DMF) (1.10 equiv) was added.
After 18 h, the reaction mixture was diluted with EtOAc and washed successively with 1
M citric acid, saturated aqueous NaHCO3, and brine. The organic phase was dried over
Na2SO4 and concentrated in vacuo. The residue was used without further purification.
STEP 4: The coupled product obtained in STEP 3 (1.00 equiv) was dissolved in DCM
(0.05 M) and stirred at room temperature. Next, TFA (0.2 M) was added and the mixture
was stirred for 2 h before being concentrated in vacuo. The residue was purified by
reversed-phase preparative HPLC. The target compound was then obtained as its free
base by passage through an Isolute® SCX-2 cartridge.
N-(1-((1H-Indol-3-yl)methyl)piperidin-4-yl)propionamide (4a). Prepared from 1H-
indole-3-carbaldehyde, tert-butyl piperidin-4-ylcarbamate, and propionic acid according
to general procedure 1 followed by general procedure 2 (70% yield over three steps).
1
Glassy solid. H NMR (400 MHz, DMSO-d6, δ): 10.90 (br s, 1H), 7.62 (d, J = 7.8 Hz,
2H), 7.34 (d, J = 7.9 Hz, 1H), 7.20 (d, J = 1.7 Hz, 1H), 7.06 (dd, J = 7.2, 0.9 Hz, 1H),
ACS Paragon Plus Environment