Development of (4-Cyanophenyl)glycine Derivatives as Reversible Inhibitors of Lysine Specific Demethylase 1

Article abstract of DOI:10.1021/acs.jmedchem.7b00462

Inhibition of lysine specific demethylase 1 (LSD1) has been shown to induce the differentiation of leukemia stem cells in acute myeloid leukemia (AML). Irreversible inhibitors developed from the nonspecific inhibitor tranylcypromine have entered clinical trials; however, the development of effective reversible inhibitors has proved more challenging. Herein, we describe our efforts to identify reversible inhibitors of LSD1 from a high throughput screen and subsequent in silico modeling approaches. From a single hit (12) validated by biochemical and biophysical assays, we describe our efforts to develop acyclic scaffold-hops from GSK-690 (1). A further scaffold modification to a (4-cyanophenyl)glycinamide (e.g., 29a) led to the development of compound 32, with a K_d value of 32 nM and an EC₅₀ value of 0.67 μM in a surrogate cellular biomarker assay. Moreover, this derivative does not display the same level of hERG liability as observed with 1 and represents a promising lead for further development.

Full text of DOI:10.1021/acs.jmedchem.7b00462

Subscriber access provided by UNIVERSITY OF THE SUNSHINE COAST
Article
Development of (4-Cyanophenyl)glycine Derivatives as
Reversible Inhibitors of Lysine Specific Demethylase 1
Daniel P. Mould, Cristina Alli, Ulf Bremberg, Sharon Cartic, Allan M. Jordan, Matthis Geitmann, Alba Maiques-
Diaz, Alison E McGonagle, Tim C. P. Somervaille, Gary J. Spencer, Fabrice Turlais, and Donald J. Ogilvie
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b00462 • Publication Date (Web): 11 Sep 2017
Downloaded from http://pubs.acs.org on September 12, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
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 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Development of (4ꢀCyanophenyl)glycine Derivatives as Reversible Inhibitors of Lysine Specific
Demethylase 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
†
⊥
§
⊥
†,*
Daniel P. Mould , Cristina Alli , Ulf Bremberg , Sharon Cartic , Allan M. Jordan , Matthis
§
‡
†
‡
‡
Geitmann , Alba MaiquesꢀDiaz , Alison E. McGonagle , Tim C. P. Somervaille , Gary J. Spencer ,
⊥
†
Fabrice Turlais , and Donald Ogilvie
†
Drug Discovery Unit, Cancer Research UK Manchester Institute, University of Manchester,
Wilmslow Road, Manchester, M20 4BX, UK
‡
Leukaemia Biology Laboratory, Cancer Research UK Manchester Institute, University of
Manchester, Wilmslow Road, Manchester, M20 4BX, UK
§
Beactica AB, Uppsala Business Park, Virdings allé 2, 75450, Uppsala, SE
⊥
CRT Discovery Laboratories, Babraham Campus, Babraham, Cambridgeshire, CB22 3AT, UK
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 61
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
Abstract
Inhibition of lysine specific demethylase 1 (LSD1) has been shown to induce the differentiation of
leukemia stem cells in acute myeloid leukaemia (AML). Irreversible inhibitors developed from the
nonꢀspecific inhibitor tranylcypromine have entered clinical trials; however, the development of
effective reversible inhibitors has proved more challenging. Herein, we describe our efforts to identify
reversible inhibitors of LSD1 from a high throughput screen, and subsequent in silico modelling
approaches. From a single hit (12) validated by biochemical and biophysical assays, we describe our
efforts to develop acyclic scaffoldꢀhops from GSKꢀ690 (1). A further scaffold modification to a (4ꢀ
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
d
cyanophenyl)glycinamide (e.g. 29a) led to the development of compound 32, with a K value of 32
nM and an EC50 value of 0.67 µM in a surrogate cellular biomarker assay. Moreover, this derivative
does not display the same level of hERG liability as observed with 1 and represents a promising lead
for further development.
ACS Paragon Plus Environment

Page 3 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Introduction
1
Lysine specific demethylase 1 (LSD1) was the first histone demethylase to be discovered, and has
2
been found to play an important role in normal and malignant cells as a transcriptional repressor.
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
LSD1 belongs to the FADꢀdependent amine oxidase family of demethylases and catalyzes the
3
4
demethylation of lysine residues of histones, specifically H3K4me1/2 and H3K9me1/2. Aside from
this role, it has become clear that LSD1 plays important scaffolding roles as part of the CoREST
5
complex, especially in controlling the interaction with transcription factors such as growth factor
6
ꢀ7
independence (GFI) 1 and 1b. LSD1 and GFI1 in combination appear to balance the regulation of
8
hematopoietic cell proliferation and differentiation in hematopoietic stem and progenitor cells. GFI1
interacts with LSD1 through binding of its Cꢀterminal SNAG domain in the active site of LSD1,
suggesting that LSD1 inhibitors may induce their phenotype through abrogation of this proteinꢀ
protein interaction, and the downstream changes in gene expression this effects, rather than by
9
inhibition of histone demethylation. In addition, it has been established that LSD1 plays a key role in
10
acute myeloid leukaemia (AML) driven by mixed lineage leukaemia (MLL) fusions, whereby
overexpression of LSD1 maintains malignant stem and progenitor cells in a selfꢀrenewing state. This
has helped to establish a novel treatment hypothesis in a hematological malignancy where the
1
1
standard of care has changed little in the past three decades.
There is also evidence that LSD1 inhibition could have a beneficial effect in a number of other disease
areas. It has been shown that small cell lung cancer cell lines have high levels of LSD1
overexpression in almost all cases, and some show sensitivity to LSD1 inhibition in vitro and in
12
tumour xenografts. The authors suggest a correlation between response and DNA hypomethylation.
The consequence of inhibition is similar to AML, in that cells show a differentiation phenotype
towards their original neuroendocrine state. LSD1 inhibition has also shown promise for the treatment
1
3, 14
15
of herpes simplex infection,
memory deficit, and patents have been filed describing the potential
use of LSD1 inhibitors in diseases associated with alterations in protein conformation, such as
1
6, 17
Alzheimer’s, Parkinson’s and Huntingdon’s diseases.
Irreversible inhibitors of LSD1 developed
from the clinical monoamine oxidase (MAO) inhibitor tranylcypromine entered clinical trials in 2014.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 61
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
These inhibitors will provide valuable insights into the validation of targeting malignant stem and
1
progenitor cells in AML patients. Preliminary clinical trials data of relꢀN ꢀ[(1R,2S)ꢀ2ꢀ
phenylcyclopropyl]ꢀ1,4ꢀcyclohexanediamine (ORYꢀ1001, Oryzon Genomics) has demonstrated
promotion of blast cell differentiation in roughly twoꢀthirds of patients in an extension cohort with
specific subtypes of relapsed or refractory AML, although adverse effects limited the maximum
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
8
tolerated dose. Additional clinical trials are underway with tranylcypromine derivatives from
19
GlaxoSmithKline (Clinicaltrials.gov identifier NCT02177812),
and Incyte Corporation
2
0
(
Clinicaltrials.gov identifier NCT02712905), although no results have yet been disclosed. The
clinical efficacy of LSD1 inhibitors may ultimately be enhanced by way of combination therapies. For
example, allꢀtransꢀretinoic acid (ATRA) therapy is already successfully used as a differentiation
therapy in acute promyelocytic leukaemia (APL), and research has suggested addition of an LSD1
21
inhibitor could enable this therapy to be more widely applicable to AML. Emerging data suggests
22
the wider applicability of this strategy to other epigenetic targets such as bromodomains, and histone
23
deacetylases.
The development of reversible LSD1 inhibitors is not trivial, due to the large size and polarity of the
LSD1 substrate binding pocket. As such, progress has not matched that of the covalent inactivators of
2
4
LSD1. Reversible inhibitors such as GSKꢀ690 (1, Figure 1) have achieved good potency in
2
5
biochemical and cellular assays, however this compound strongly inhibits the human etherꢀaꢀgoꢀgoꢀ
26
related gene (hERG) cardiac ion channel, which prevented the progression of this series to the clinic.
Many other published reversible inhibitor series appear to have significant offꢀtarget and nonꢀspecific
2
4
effects. In the past year, it has become apparent that there is commercial interest in developing
reversible inhibitors, as evidenced by a number of patent applications from Quanticel Pharmaceuticals
2
7–32
(
2–6, Figure 1) and Celgene.
Recently, the discovery and optimization of a series of thieno[3,2ꢀ
b]pyrroleꢀ5ꢀcarboxamides, and the crystal structures of quinazoline based reversible inhibitors has
3
3–35
also been described.
ACS Paragon Plus Environment

Page 5 of 61
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 1. Compound 1 and examples from several series disclosed by Quanticel Pharmaceuticals(2–
2
8
6).
Herein, we describe our efforts towards the development of reversible inhibitors of LSD1 identified
by highꢀthroughput screening, and using in silico design to develop subꢀmicromolar inhibitors of
LSD1 that show activity in cellular assays.
Chemistry
Aniline 9 (Scheme 1) was prepared from nitrobenzene 7 via aromatic nucleophilic substitution with
3,5ꢀdimethylpyrazole to afford 8, which was reduced with zinc metal and ammonium formate. Aniline
9 was reacted with the requisite sulfonyl chloride in pyridine to afford compounds 10a–b. As
described in Scheme 2, compound 12 was formed by the reaction of aniline 11 with 4ꢀ
acetatamidobenzenesulfonyl chloride. Compound 11 was also reacted with 4ꢀmethylphenylsulfonyl
chloride in acetonitrile in the presence of pyridine to afford sulfonamide 13. Compound 13 was
alkylated with tertꢀbutyl bromoacetate to afford the ester 14. Deprotection with trifluoroacetic acid
3
6
afforded acid 15 as a key intermediate. This was reacted with a variety of diꢀamines under standard
amide coupling conditions, with a Bocꢀdeprotection step using 4M HCl/dioxane if required, to give
amides 16a–m. Sulfonamides 18–20 were formed under standard conditions, then alkylated with a
Bocꢀprotected chloroacetylpiperidine derivative (see Supporting Information for synthesis) to afford
derivatives 21–23 (Scheme 3). To synthesize the methylene replacement analogues (Scheme 4),
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 61
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
commercially available (4ꢀcyanophenyl)glycine was reacted with a variety of Bocꢀprotected
substituted piperidines under standard amide coupling conditions to afford intermediates 25–28.
Alkylation with a variety of substituted benzyl bromides in DMF in the presence of potassium
carbonate, and subsequent Bocꢀdeprotection, afforded compounds 29a–e, 30a–f, 31 and 32.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Scheme 1: Synthesis of compounds 10a–b
a
Reagents and conditions: (i) 3,5ꢀdimethylpyrazole, K CO , DMF, 80 °C, 16 h, 99%; (ii) Zn,
2
3
ammonium formate, MeOH, 40 °C, 16 h, 78%; (iii) appropriate sulfonyl chloride, pyridine, 100 °C, 1
h, 41–70%.
a
Scheme 2: Synthesis of compounds 12 and 16a–m
a
Reagents and conditions: (i) 4ꢀacetamidobenzenesulfonyl chloride, DMAP, MeCN, µW, 100 °C, 1 h,
3
7%; (ii) 4ꢀmethylphenylsulfonyl, pyridine, MeCN, RT, 12 h, 89%; (iii) tertꢀbutyl bromoacetate,
CO , DMF, 1 h, 96%; (iv) TFA, DCM, 1 h, 98%; (v) amine, DIPEA, COMU, DMF, then 4M
HCl/dioxane if Boc deprotection required, RT, 12–68%.
K
2
3
ACS Paragon Plus Environment

Page 7 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
Scheme 3: Synthesis of compounds 21a–b, 22 and 23
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
Reagents and conditions: (i) 1ꢀacetylindolineꢀ5ꢀsulfonyl chloride, pyridine, MeCN, RT, 12 h, 84%;
ii) (21a) tertꢀbutyl Nꢀ[(3R)ꢀ1ꢀ(2ꢀchloroacetyl)ꢀ3ꢀpiperidyl]carbamate, K CO , cat. KI, DMF, 80 °C, 3
CO , 80
C, 6 h, then 4M HCl/dioxane, 70%; (iii) 2ꢀmethylꢀ2Hꢀindazolꢀ5ꢀamine, pyridine, MeCN, rt, 12 h,
(
2
3
h; then TFA, 1 h, rt, 42%; (21b) tertꢀbutyl Nꢀ[1ꢀ(2ꢀchloroacetyl)ꢀ4ꢀpiperidyl]carbamate, K
2
3
°
5
4%; (iv) tertꢀbutyl Nꢀ[(3R)ꢀ1ꢀ(2ꢀchloroacetyl)ꢀ3ꢀpiperidyl]carbamate, K CO , cat. KI, DMF, 80 °C, 3
2 3
h; then TFA, 1 h, rt, 25%; (v) 1ꢀacetylindolineꢀ5ꢀsulfonyl chloride, pyridine, MeCN, RT, 12 h, 54%;
vi) tertꢀbutyl Nꢀ[1ꢀ(2ꢀchloroacetyl)ꢀ4ꢀpiperidyl]carbamate, K CO , DMF, 80 °C, 6 h, then 4M
HCl/dioxane, 1 h RT, 52%.
(
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 61
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
Scheme 4: Synthesis of compounds 29a–e, 30a–f, 31 and 32
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Reagents and conditions: i) Bocꢀprotected diamine, DIPEA, COMU, DMF, rt, 1 h, 67 – 99%; ii)
2 3
substituted benzylꢀbromide, K CO , DMF, 80°C, 1 h, then 4 M HCl/dioxane, rt, 1 h, 6–58%
Results and Discussion
A library of ~150,000 compounds was screened against the LSD1 enzyme in a timeꢀresolved
fluorescence resonance energy transfer (TRꢀFRET) assay at 30 µM, with an initial hit rate of 0.3% (as
defined by >70% inhibition). The resulting 472 compounds were tested for interference of the TRꢀ
FRET assay and for purity by LCꢀMS, then taken into 7ꢀ and 12ꢀ point IC50 testing, yielding 72 hits
with IC50 values in the range of 4–102 µM.
After resynthesis of the original hit matter for validation, we chose to develop a series of sulfonamides
that appeared to offer an attractive startꢀpoint with multiple vectors for diversification (Figure 2). This
series was optimized from compound 10a to give 10b with subꢀmicromolar activity in the
biochemical assay. Both sides of the sulfonamide could be extensively modified while retaining
potency; however the aniline portion originating from the HTS could not be improved upon.
3
7
Unfortunately, when tested in our previously described cellular assay, 10b and related compounds
ACS Paragon Plus Environment

Page 9 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
were found to cause no induction of CD86 at 50 µM in human THPꢀ1 AML cells (Figure 2B). This
disconnect was reinforced when the compounds were tested by surface plasmon resonance (SPR),
where the binding response increased linearly with compound concentration (Figure 2C, inset),
suggesting that no specific binding between compound and LSD1 protein was occurring. In addition,
a competition assay with tranylcypromine demonstrated no abrogation of the interactions of these
compounds with LSD1 (data not shown). Several other chemotypes emerging from the HTS were
shown to display a similar lack of activity by SPR, suggesting the biochemical assay may be
particularly susceptible to false positives, and demonstrating the value of an orthogonal biophysical
assay in the screening cascade.
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. Development and characterization of compound 10b. A: Biochemical assay; B: Surrogate
cellular biomarker assay; C: Surface plasmon resonance sensorgram of the interaction between LSD1
and 10b in twoꢀfold dilution series (highest concentration indicated in the graph).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 61
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
Figure 3. Structure of compound 12 with biochemical and SPR sensorgram indicating the highest
tested concentration.
d d
K = 9.5 µM. K value was determined by SPR at 15 °C as the average of two
experimental series (typically 10 concentrations).
SPR analysis did identify a single compound (12) that displayed reversible binding activity (Figure 3),
a sulfonamide that we hypothesized might be acting as a ringꢀopened isostere of the pyridine scaffold
of compound 1. While 1 was disclosed by GSK at the AACR Annual Meeting 2013 (Washington DC,
38
USA), very little work to develop this ligand had been published before 2015. Certainly, the scope
for potential coreꢀhops had been seldom explored at the point at which we began to pursue this
approach. To attempt to validate sulfonamides as a potential new scaffold, we used Cresset Spark and
Torch software (Welwyn Garden City, U.K.) to align and identify potential new cores by searching
3
9
fragment libraries for replacements with similar spatial and electronic features.
ACS Paragon Plus Environment

Page 11 of 61
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 4. Cresset Torch alignment and structures of compound 1 (left) and 16g (right). Similarity
score 0.712. Blue field points (spheres) highlight energy minima for a positively charged probe, red
for a negative probe. Yellow spheres represent an attractive van der Waals minimum for a neutral
probe and brown spheres represent hydrophobic regions. Oxygen atoms are shown in red, nitrogen in
blue. The similarity score is based on the likeness of the field points in terms of their magnitude and
position. A score of greater than 0.7 is considered a ‘good’ score.
From this output we designed initial targets, a series of sulfonamidoacetamides (Figure 4) that looked
to offer ease of synthesis while offering low twoꢀdimensional (2D) similarity in comparison to more
recognizable pyridine replacements such as pyridazines or pyrazines. We incorporated the paraꢀcyano
and paraꢀtolyl aryl groups from 1 for the initial development, focused on optimizing the basic center,
by coupling a series of amines to a key sulfonamidoacetic acid intermediate to give compounds 16a–
m. (Table 1). In contrast to the sulfonamides 10a–b, these compounds showed specific and reversible
binding to LSD1 by SPR, validating them as start points for further assay. Both primary and
secondary amides decorated with basic amines were tolerated and displayed significantly improved
levels of potency over 16a. Initial optimization gave the 4ꢀaminopiperidine derivative 16g.
Methylation αꢀ to the basic center (16k) was also tolerated, alongside methylation on the nitrogen
itself (16l, 16m), however the activity appeared have reached a plateau, with no compounds achieving
a K value below 2 µM.
d
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 61
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
Table 1. Optimization of basic center on the acetamidosulfonamide scaffold
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Compound
1
R
IC50 (µM)
d
SPR K (µM)
ꢀ
0.037 (0.012)
0.009
90
1
6a
6b
>150
1
11.9 (0.2)
3.6
1
6c
14.2 (0.3)
13.1 (0.2)
3.5
2.4
16d
1
6e
13.9 (0.1)
24.3 (0.4)
4.5
1
6f
10.3
N
1
6g
6h
7.6 (1.2)
2.6
6.5
NH2
1
15.6 (1.3)
16i
16j
22.7 (1.6)
16.4 (0.6)
4.7
3.2
16k
5.2 (0.7)
2.5
ACS Paragon Plus Environment

Page 13 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
16l
7.5 (1.2)
2.9
2.9
N
1
6m
12.6 (1.8)
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
a
IC and K values for selected compounds against the LSD1 enzyme in biochemical and biophysical
50
d
assays. Standard deviation is given in parentheses. IC50 determined from 10 point concentration/effect
experiments. Geometric mean of at least two independent experimental determinations given. Kd
values were determined by SPR at 15 °C as the average of two experimental series (typically 10
concentrations).
Shortly after synthesizing these initial compounds, a patent from Quanticel disclosed several series of
reversible LSD1 inhibitors featuring 6ꢀ and 5ꢀmembered monocyclic and 6,5ꢀbicyclic coreꢀhops
2
8
(
Figure 1). Compounds from each series were claimed to display IC50 values against LSD1 of less
than 100 nM in biochemical and cellular assays. While the patent reduced the diversity of chemical
space available to explore, it also provided useful learning, both in terms of the essential features of
the compounds, and which positions are more tolerant to change. For instance, the paraꢀcyanophenyl
and the presence of a basic nitrogen center are conserved throughout the examples in the patent,
whereas the tolyl moiety was more amenable to optimization, and was successfully replaced with a
number of monoꢀ and bicyclic systems (e.g. 3, Figure 1).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 61
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
Table 2. Aryl modifications on the acetamidosulfonamide scaffold
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
2
3
IC50
(µM)
SPR K
d
(µM)
Compound
1
R
R
R
0
.037
ꢀ
ꢀ
ꢀ
0.009
(
0.012)
5.2
21a
21b
2.4
(0.4)
4.4
1.8
9.6
(1.0)
30.0
2
2
2
3
(
2.4)
>30
>100
a
IC50 and K
d
values for selected compounds against the LSD1 enzyme in biochemical and biophysical
assays. Standard deviation is given in parentheses. IC determined from 10ꢀpoint concentration/effect
5
0
experiments. Geometric mean of at least two independent experimental determinations given. K
d
values were determined by SPR at 15 °C as the average of two experimental series (typically 10
concentrations).
With this in mind, we investigated how the modification of the two aryl groups affected the activity
(
Table 2). Derivatives with an acetylindoline group were shown to give slightly better potency than
the parent tolyl (21a–b). On the other hand, compound 22, where the sulfonamide was reversed and
ACS Paragon Plus Environment

Page 15 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the tolyl replaced with the favored 2ꢀmethylindazole from the Quanticel patent, showed a
disappointing dropꢀoff in activity. Adopting the substituted aniline group from the prior HTS series
(23) showed no activity by biochemical assay or SPR, while the parent 20 (biochemical IC50 = 1.6
d
µM; K > 100 µM) displayed the same nonꢀspecific binding effects by SPR. Despite our optimization
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
attempts, we were disappointed to find that potency could not be significantly improved further, and
thus we looked to modify the core itself to establish whether potency could be improved..
Toward this aim, we replaced the SO
2
moiety with a methylene linker to give a 4ꢀ
cyanophenylglycinamide core. This alteration has the benefit of a significant reduction in molecular
weight and total polar surface area, which could afford increased membrane permeability and
bioavailability. However, this modification also introduces an aniline functionality that could
40
potentially be a metabolic liability. Initially, we chose to adopt the 4ꢀaminopiperidine from
compound 16g as the basic center, and focused on modifications of the tolylꢀ group. Compound 29a
demonstrated a 5ꢀfold improvement in potency compared to the matchedꢀpair sulfonamide 16h by
both biochemical assay and SPR. In addition, a range of phenyl groups were tolerated and, in one
case, improved potency further (29c). These beneficial modifications could be extended to bicyclic
systems (15e). We then returned to investigate the basic center and found that homologating the 4ꢀ
aminopiperidine unit to a piperidinꢀ4ꢀylmethanamine gave a useful ~4ꢀfold increase in potency (e.g.
29a vs. 30a). Modifying the pKa of the basic center by introducing of a fluorine atom at the βꢀcarbon
to the amine was deleterious to activity, however decreasing the flexibility of the amine by
incorporation into a spirocycle (32) afforded a compound with <100 nM activity in the biochemical
assay.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 61
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
Table 3. Aryl optimization on the (4ꢀcyanophenyl)glycine scaffold
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
SPR
K
d
Compound
1
R
IC50 (µM)
(
µM)
-
0.037 (0.012)
1.3 (0.02)
0.009
0.43
29a
29b
5.5 (0.07)
1.9
29c
0.81 (0.02)
4.1 (0.1)
0.62 (0)
0.19
1.1
29d
29e
0.14
a
d
IC50 and K values for selected compounds against the LSD1 enzyme in biochemical and biophysical
assays. Standard deviation is given in parentheses. IC50 determined from 10 point concentration/effect
experiments. Geometric mean of at least two independent experimental determinations given. Kd
values were determined by SPR at 15 °C as the average of two experimental series (typically 10
concentrations).
ACS Paragon Plus Environment

Page 17 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
Table 4. Basic center optimization and aryl variation on the (4ꢀcyanophenyl)glycine scaffold
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
2
SPR
K
d
Compound
R
R
IC50 (µM)
0.037 (0.012)
0.29 (0.03)
(
µM)
1
-
-
0.009
0.11
30a
30b
0.40 (0.11)
1.3 (0.15)
0.29
0.46
30c
30d
30e
1.0 (0.14)
0.42
0.49
0.83 (0.08)
30f
0.21 (0.06)
0.99 (0.23)
0.058
0.23
31
32
0.083 (0.003)
0.032
a
IC and K values for selected compounds against the LSD1 enzyme in biochemical and biophysical
5
0
d
assays. Standard deviation is given in parentheses. IC50 determined from 10 point concentration/effect
experiments. Geometric mean of at least two independent experimental determinations given. K
d
values were determined by SPR at 15 °C as the average of two experimental series (typically 10
concentrations).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 61
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
In an attempt to rationalize the observed SAR, we performed docking studies, using the previously
4
1
reported structure of LSD1 bound to tetrahydrofolate (PDB accession code 4KUM). Protein
42
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
preparation and docking were performed in Glide (Schrödinger, New York, USA). The crystal
26
structure of 1 bound to LSD1 has been disclosed, and from these data, the binding mode of
compound 30f was predicted, as displayed in Figure 5. In this predicted pose, the nitrile forms a key
hydrogen bonding interaction with K661. The basic center forms a network of ionic interactions with
Asp555 and Asp556. This ionic interaction has been exploited previously by Vianello and coꢀ
3
3–34
workers.
The 3ꢀfluoroꢀ4ꢀmethoxyphenyl group sits in a channel formed by Trp695, Ile356,
Leu677 and Leu693. The ether oxygen is predicted to make a hydrogen bonding interaction with
Gln358, however, if the phenyl group were to flip then it could also potentially form a similar
interaction with Asn535. It is worthy of note that a wide variety of scaffoldꢀhops can be tolerated, and
this may be explained by a lack of contact between the protein and the scaffold itself, with the key
interactions being located at distal ends of the ligands.
ACS Paragon Plus Environment

Page 19 of 61
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 5. Left: Predicted binding mode of compound 30f in the LSD1 active site (4KUM). Visualized
using Maestro. Right: Proteinꢀligand interaction diagram with key hydrogen bond interactions
highlighted (pink arrows).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 61
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
To assess the effects of these derivatives on cells, compounds 1, 30f and 32 were tested at 7
concentrations in the THPꢀ1 AML cell line to measure their effect on expression levels of CD86, a
3
7
surrogate cellular biomarker of LSD1 inhibition (Figure 6). Although compound 1 displayed the
highest levels of potency in this assay (EC50 = 0.30 µM), we were satisfied to find only a 2ꢀfold drop
off in activity to our most active derivative, 32 (EC50 = 0.67 µM). The less active analogue 30f
displayed more modest activity (EC50 = 4.1 µM), suggesting good correlation between the
biochemical and cellular assay.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. CD86 cell assay results for compounds 1, 30f and 32.
To further assess cellular effects, we performed colony forming assays using two AML cell lines. This
assay format indicates both proliferation of a cell population and the percentage of cells within that
10
population with colony initiating activity. THP1, and MV4ꢀ11 cells were treated with 10 µM of 1 or
32 for 7 days in semiꢀsolid methylcelluloseꢀbased culture. Both compounds substantially reduced both
colony forming frequency and the size of the colonies that did develop (Figure 7) indicating
significant antiꢀproliferative activity, in addition to inducing upꢀregulation of the cellular
differentiation marker CD86. To explore whether these compounds induced apoptosis, we performed
ACS Paragon Plus Environment

Page 21 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Annexin VꢀperCP EF710/7AAD staining on THPꢀ1 cells treated for four days with 10 µM of each of
the compounds. A modest but significant increase in apoptosis was observed (Figure 8). These results,
combined with the CD86 data, suggest that compounds such as 32 act predominantly to promote
differentiation of AML cells, with concomitant reduction in clonogenic potential.
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 22 of 61
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
Figure 7. Top: Colony formation assay for THPꢀ1 and MV4ꢀ11 cells dosed with 10 µM of
compounds 1 and 32 for 7 days. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, statistical
significance calculated using a oneꢀway ANOVA with Dunnett’s tests between vehicle treated and
compound dosed cells. Bottom: representative images of colonies.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 8. Annexin VꢀperCP EF710/7AAD double staining experiments results with THPꢀ1 cell dosed
pos
with vehicle, or 10 µM 1 or 32 for 4 days in liquid culture. Early apoptotic cells = Annexin V 7ꢀ
neg
pos
pos
AAD ; Apoptotic cells = Annexin V 7ꢀAAD . *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001,
statistical significance calculated using oneꢀway ANOVA analysis with Dunnett’s tests.
in vitro ADME properties
As inhibition of the hERG ion channel proved to be an insurmountable issue in furthering the
2
6
development of 1 towards the clinic, we were keen to assess our compounds against this target.
Compounds 29c, 30f, and 32 were tested in a patch clamp assay, alongside being profiled for their
metabolic stability (Table 5). Gratifyingly these compounds displayed improved selectivity against
hERG in comparison to 1 (IC50 = 3.2 µM), with compound 32 displaying a ~30ꢀfold window of
selectivity between LSD1 inhibition in cells and hERG inhibition. In addition, we observed low levels
of clearance in mouse microsomes with all three compounds. Disappointingly, these compounds
displayed poor membrane permeability and high levels of efflux in comparison to compound 1, which
ACS Paragon Plus Environment

Page 23 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
could make achieving oral bioavailability challenging, although this liability does not appear to
preclude the ability of the compounds to inhibit LSD1 in a cellular context.
Table 5. In vitro parameters for 1, 29c, 30f and 32
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
29c
30f
32
Caco2 A–B mean P
(10 cm/s); (efflux
ratio)
app
ꢀ6
5.6 (3.2)
0.44 (48)
0.26 (58)
0.09 (58)
Mo Mic CLint
6
3.43
404
3.6
385
>25
4.5
305
<1
(ꢁL/min/10 cells)
Mo Mic T
½
(min)
>1000
hERG EC (µM);
5
0
3.2
12.4 (3.4)
20.1 (4.6)
(
standard error)
CONCLUSION
In conclusion, we have developed a series of (4ꢀcyanophenyl)glycine derivatives that show potent
activity against LSD1 by both biochemical assay and SPR. Starting from a weakly active HTS hit 12,
we identified acyclic leadꢀlike analogues of compound 1. We successfully developed these initial start
points, leading to the development of compound 32, which potently inhibits LSD1 in biochemical,
biophysical and cellular assays. These compounds display much greater levels of selectivity over the
hERG cardiac ion channel than compound 1, a liability which precluded further development of that
series. Compound 32 is an attractive tool compound for use in in vitro systems and startꢀpoint for
future development. More generally, this work suggests the scaffold of compound 1 is broadly
modifiable, and while the pꢀcyanophenyl group appears to form a key interaction with K661,
modification of the tolylꢀ functionality and the position and orientation of basic center is much more
tolerant to modification.
Experimental Section
Chemistry
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 61
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
Flash column chromatography was performed using preꢀpacked silica gel cartridges (KPꢀSil SNAP,
Biotage, Hengoed UK or RediSep Rf, Isco). Thin layer chromatography was conducted with 5 × 10
cm plates coated with Merck Type 60 F254 silica gel to a thickness of 0.25 mm. All reagents obtained
from commercial sources were used without further purification. Anhydrous solvents were obtained
from the SigmaꢀAldrich Company Ltd. (Dorset, UK) or Fisher Scientific UK Ltd. (Loughborough,
UK), and used without further drying. HPLC grade solvents were obtained from Fisher Chemicals
Ltd.
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
All compounds prepared were >95% purity as determined by examination of both the LCꢀMS and H
NMR spectra unless otherwise indicated. Where Cl or Br were present, expected isotopic distribution
patterns were observed.
1
13
Proton ( H) and carbon ( C) NMR spectra were recorded on a 300 MHz Bruker Avance
spectrometer. Solutions were typically prepared in either deuterochloroform (CDCl ), deuterated
dimethylsulfoxide (DMSOꢀd ) or deuterated methanol (methanolꢀd ) with chemical shifts referenced
3
6
4
1
to tetramethylsilane (TMS) or deuterated solvent as an internal standard. H NMR data are reported
indicating the chemical shift (δ), the integration (e.g. 1H), the multiplicity (s, singlet; d, doublet; t,
triplet; q, quartet; m, multiplet; br, broad; dd, doublet of doublets etc.), and the coupling constant (J)
in Hz. Deuterated solvents were obtained from Goss Scientific (Crewe, UK). In general, basic NꢀH
protons were not observed.
LCꢀMS analyses were performed on a Waters (Elstree, UK) Acquity UPLC system fitted with BEH
C18 1.7 ꢁM columns (2.1 × 50 mm) and with UV diode array detection (210–400 nm) with a Waters
SQD detector. Analyses were performed with either buffered acidic or basic solvents, and using
gradients detailed in the Supporting Information.
Some compounds were purified by preparative (prep.) HPLC on a Waters FractionLynx MS
autopurification system, with a Waters XBridge 5 µm C18, 100 mm × 19 mm i.d. column, running at
a flow rate of 20 mL/min with UV diode array detection (210–400 nm) and massꢀdirected collection
using both positive and negative mass ion detection. Purifications were performed using buffered
ACS Paragon Plus Environment

Page 25 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
acidic or basic solvent systems as appropriate. Compound retention times on the system were
routinely assessed using a 30–50 ꢀL test injection and a standard gradient, then purified using an
appropriately chosen focused gradient as detailed in the Supporting Information, based upon observed
retention time.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Preparation of compounds 10a–b
1ꢀ(2ꢀMethoxyꢀ4ꢀnitroꢀphenyl)ꢀ3,5ꢀdimethylpyrazole (8).
A suspension of 3ꢀmethoxyꢀ4ꢀfluoronitrobenzene (25 g, 146 mmol), 3,5ꢀdimethylpyrazole (15.5 g,
161 mmol) and potassium carbonate (60.5 g, 438 mmol) in DMF (200 mL) was stirred overnight at 80
°C. The reaction mixture was added to water (500 mL) and the resulting yellow precipitate collected
1
by filtration and dried to the title compound (35.8 g, 99%) as a yellow powder. H NMR (300 MHz,
Chloroformꢀd) δ 7.95 (dd, J = 8.5, 2.4 Hz, 1H), 7.90 (d, J = 2.4 Hz, 1H), 7.54 (d, J = 8.5 Hz, 1H),
R
6.03 (s, 1H), 3.94 (s, 3H), 2.31 (s, 3H), 2.14 (s, 3H). LCꢀMS: (High pH) t 1.09 min, m/z 248.1
+
[M+H] , 100% purity.
4ꢀ(3,5ꢀDimethylpyrazolꢀ1ꢀyl)ꢀ3ꢀmethoxyaniline (9).
A stirred suspension of 1ꢀ(2ꢀmethoxyꢀ4ꢀnitroꢀphenyl)ꢀ3,5ꢀdimethylpyrazole (36 g, 0.145 mol), zinc
powder (28.6 g, 0.44 mol) and ammonium formate (91.2 g, 1.46 mol) in methanol (250 mL) was
heated overnight to 40 °C. The reaction mixture was passed through a hydrophobic frit and
concentrated to dryness. The resulting residue was suspended between brine (250 mL) and DCM (500
mL), the organic layer passed through a hydrophobic frit and concentrated to dryness to afford the
crude product. Recrystallization from hexanes afforded the title compound (24.7 g, 78%) as offꢀwhite
1
needles. H NMR (300 MHz, Chloroformꢀ
d
) δ 7.13–7.03 (m, 1H), 6.36 (m, 2H), 5.94 (s, 1H), 3.74 (s,
+
3H), 2.30 (s, 3H), 2.08 (s, 3H). LCꢀMS: (High pH) t
R
0.88 min, m/z 218.2 [M+H] , 98% purity.
Nꢀ[4ꢀ(3,5ꢀDimethylpyrazolꢀ1ꢀyl)ꢀ3ꢀmethoxyphenyl]benzenesulfonamide (10a).
A solution of benzenesulfonyl chloride (18 ꢁL, 0.14 mmol) and 4ꢀ(3,5ꢀdimethylpyrazolꢀ1ꢀyl)ꢀ3ꢀ
methoxyꢀaniline (30 mg, 0.14 mmol) in pyridine (0.45 mL) was stirred at 100 °C for 1 hour. The
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 61
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
reaction was concentrated to dryness then purified by preparative HPLC (low pH gradient 3) to afford
1
the title compound (20 mg, 41%) as a white powder. H NMR (300 MHz, DMSOꢀd ) δ 8.39 (s, 1H),
6
7.90–7.75 (m, 2H), 7.65–7.42 (m, 3H), 7.02 (d, J = 8.4 Hz, 1H), 6.83 (d, J = 2.2 Hz, 1H), 6.67 (dd, J
= 8.4, 2.2 Hz, 1H), 5.94–5.81 (m, 1H), 3.62 (s, 3H), 2.09 (s, 3H), 1.91 (s, 3H).
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
Nꢀ[4ꢀ(3,5ꢀDimethylpyrazolꢀ1ꢀyl)ꢀ3ꢀmethoxyphenyl]ꢀ1,3ꢀdimethylꢀ2ꢀoxoꢀquinolineꢀ6ꢀsulfonamide
(10b).
A solution of 4ꢀ(3,5ꢀdimethylpyrazolꢀ1ꢀyl)ꢀ3ꢀmethoxyaniline (28 mg, 0.13 mmol) and 1,3ꢀdimethylꢀ2ꢀ
oxoꢀquinolineꢀ6ꢀsulfonyl chloride (34 mg, 0.12 mmol) in pyridine (300 µL) was stirred at room
temperature overnight. The reaction mixture was diluted with 2:1:1 (v:v:v) DMSO:water:MeCN (700
ꢀ
L), then purified by preparative HPLC (low pH) to afford the title compound (39 mg, 70%) as a
1
white powder. H NMR (300 MHz, DMSOꢀd
6
) δ 10.62 (s, 1H), 8.15 (d, J = 2.2 Hz, 1H), 7.97–7.86
(m, 2H), 7.67 (d, J = 9.0 Hz, 1H), 7.08 (d, J = 8.4 Hz, 1H), 6.93 (d, J = 2.2 Hz, 1H), 6.75 (dd, J = 8.4,
13
2
.2 Hz, 1H), 5.90 (s, 1H), 3.66 (s, 3H), 3.64 (s, 3H), 2.12 (s, 3H), 2.07 (s, 3H), 1.91 (s, 3H). C NMR
(
75 MHz, Methanolꢀd
4
) δ 144.1, 140.5, 137.6, 135.1, 133.2, 130.5, 128.7, 124.8, 122.1, 114.7, 114.0,
ꢀ
1
10.8, 106.8, 57.4, 30.9, 22.4, 15.7, 15.2. HRMS (ESI) m/z [MꢀH] calcd for C23
H
23
N
4
O
4
S: 451.1445.
Found: 451.1443.
Preparation of compound 12
Nꢀ[4ꢀ[(4ꢀCyanophenyl)sulfamoyl]phenyl]acetamide (12).
A solution of DMAP (50 mg, 0.42mmol), pꢀcyanoaniline (50 mg, 0.42 mmol) and 4ꢀ
acetamidobenzenesulfonyl chloride (99 mg, 0.42 mmol) in acetonitrile (5 mL) was heated by
microwave irradiation to 100 °C for 1 h. The reaction was concentrated to dryness then purified by
preparative HPLC (low pH gradient 2) to afford the title compound (49 mg, 37%) as colourless
1
crystals. H NMR (300 MHz, DMSOꢀd ) δ 10.95 (s, 1H), 10.34 (s, 1H), 7.83–7.62 (m, 6H), 7.28–7.15
6
(
m, 2H), 2.06 (s, 3H).
Preparation of compounds 16a–16m
ACS Paragon Plus Environment

Page 27 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
N
ꢀ(4ꢀCyanophenyl)ꢀ4ꢀmethylbenzenesulfonamide (13).
To a stirred solution of 4ꢀcyanoaniline (5 g, 42.3 mmol) in pyridine (3.6 mL) and acetonitrile (9 mL)
was added portionwise pꢀtoluenesulfonyl chloride (8.88 g, 46.6 mmol). The reaction mixture was
stirred at room temperature overnight, then water (50 mL) was added and the resulting light orange
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
precipitate collected by filtration, washed with water (3 × 20 mL) and dried under vacuum at 50 °C to
1
afford the title compound (8.92 g, 77%) as a light orange powder. H NMR (300 MHz, DMSOꢀd
6
) δ
10.99 (s, 1H), 7.78–7.61 (m, 4H), 7.43–7.33 (m, 2H), 7.32–7.15 (m, 2H), 2.34 (s, 3H). LCꢀMS: (High
ꢀ
pH) t 0.82 min, m/z 271.1 [M−H] , 100% purity.
R
tertꢀButyl 2ꢀ[4ꢀcyanoꢀNꢀ(pꢀtolylsulfonyl)anilino]acetate (14).
To a stirred suspension of Nꢀ(4ꢀcyanophenyl)ꢀ4ꢀmethylbenzenesulfonamide (2 g, 7.34 mmol) and
potassium carbonate (2.02 g, 14.7 mmol) in DMF (20 mL) was added tertꢀbutyl bromoacetate (1.30
mL, 8.81 mmol) and the reaction mixture heated to 80°C for 1 h. The reaction mixture was diluted
with water (50 mL) then extracted with DCM (2 × 20 mL). The combined organic layers were passed
through a hydrophobic frit and concentrated to dryness to afford the title compound (2.71g, 97%) as
1
an orange oil. H NMR (300 MHz, Chloroformꢀd) δ 7.66–7.51 (m, 4H), 7.41–7.23 (m, 4H), 4.35 (s,
+
2H), 2.43 (s, 3H), 1.40 (s, 9H). LCꢀMS: (High pH) t
R
1.32 min, m/z 387.1 [M+H] , 94% purity.
2ꢀ[4ꢀCyanoꢀNꢀ(pꢀtolylsulfonyl)anilino]acetic acid (15).
A stirred solution of tertꢀbutyl 2ꢀ[4ꢀcyanoꢀNꢀ(pꢀtolylsulfonyl)anilino]acetate (2.9 g, 7.5 mmol) in
trifluoroacetic acid (2.79 mL, 37.5 mmol) and DCM (30 mL) was heated to 30 °C for 1 h. The
reaction mixture was concentrated to dryness to afford an orange oil. Purification by column
chromatography (0 to 60% EtOAc in isohexane, then 10% MeOH in DCM) afforded the title
1
compound as a colourless oil that solidified upon standing to give colourless crystals. H NMR (300
MHz, DMSOꢀd
6
) δ 13.06 (s, 1H), 7.87–7.75 (m, 2H), 7.65–7.53 (m, 2H), 7.39 (m, 4H), 4.55 (s, 2H),
ꢀ
2.38 (s, 3H). LCꢀMS: (High pH) t
R
0.63 min, m/z 319.1 [M−H] , 96% purity.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 61
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
General method A: To a stirred solution of 2ꢀ[4ꢀcyanoꢀNꢀ(pꢀtolylsulfonyl)anilino]acetic acid (30.
mg, 0.09 mmol), the appropriate bisꢀamine (1.2 eq.) and N,NꢀDiisopropylethylamine (45 µL, 0.27
mmol) in DMF (300 µL) was added COMU (58 mg, 0.14 mmol) and the reaction mixture stirred at
room temperature for 6 h. The reaction mixture was concentrated to dryness and purified by mass
directed preparative HPLC (high pH), to afford the desired product.
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
General method B: To a stirred solution of 2ꢀ[4ꢀcyanoꢀNꢀ(pꢀtolylsulfonyl)anilino]acetic acid (30 mg,
0.09 mmol), the appropriate Bocꢀprotected bisꢀamine (1.2 eq.) and N,Nꢀdiisopropylethylamine (45
µL, 0.27 mmol) in DMF (300 µL) was added COMU (58 mg, 0.14mmol) and the reaction mixture
stirred at room temperature for 6 h. The reaction mixture was concentrated to dryness and 4 M HCl in
dioxane (1 mL) was added. This was stirred for 1 h, then concentrated to dryness and purified by mass
directed preparative HPLC (high pH), to afford the desired product.
2ꢀ[4ꢀCyanoꢀNꢀ(pꢀtolylsulfonyl)anilino]ꢀNꢀ(3ꢀpyridyl)acetamide (16a).
Prepared by general method A using 3ꢀaminopyridine to afford the title compound (24 mg, 65%) as a
1
light brown powder. H NMR (300 MHz, Methanolꢀd
4
) δ 8.68 (dd, J = 2.6, 0.8 Hz, 1H), 8.27 (dd, J =
4.8, 1.5 Hz, 1H), 8.07 (ddd, J = 8.4, 2.6, 1.5 Hz, 1H), 7.72–7.65 (m, 2H), 7.63–7.57 (m, 2H), 7.54–
7.43 (m, 2H), 7.44–7.28 (m, 3H), 4.61 (s, 2H), 2.42 (s, 3H).
2ꢀ[4ꢀCyanoꢀNꢀ(pꢀtolylsulfonyl)anilino]ꢀNꢀ[(3R)ꢀquinuclidinꢀ3ꢀyl]acetamide (16b).
Prepared by general method A using (R)ꢀ3ꢀaminoquinuclidine dihydrochloride to afford the title
1
product as an orange oil (14 mg, 35%). H NMR (300 MHz, Methanolꢀd ) δ 8.49 (s, 1H), 7.75–7.63
4
(
(
m, 2H), 7.60–7.50 (m, 2H), 7.42 (m, 2H), 7.40–7.34 (m, 2H), 4.42 (s, 2H), 4.17 (m, 1H), 3.74–3.60
m, 1H), 3.34–3.19 (m, 4H), 3.02 (dd, J = 13.8, 4.7 Hz, 1H), 2.44 (s, 3H), 2.12 (q, J = 3.1 Hz, 1H),
1.99 (s, 3H), 1.86 (s, 1H).
2ꢀ[4ꢀCyanoꢀNꢀ(pꢀtolylsulfonyl)anilino]ꢀNꢀ[(3S)ꢀquinuclidinꢀ3ꢀyl]acetamide (16c).
Prepared by general method A using (S)ꢀ3ꢀaminoquinuclidine dihydrochloride to afford the title
1
product as a yellow powder (11 mg, 28%). H NMR (300 MHz, Methanolꢀd ) δ 8.46 (s, 1H), 7.70 (dd,
4
ACS Paragon Plus Environment

Page 29 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
J = 1.9, 8.7 Hz, 2H), 7.60–7.50 (m, 2H), 7.47–7.40 (m, 2H), 7.37 (d, J = 7.9 Hz, 2H), 4.43 (s, 2H),
4.19 (m, 1H), 3.72–3.62 (m, 1H), 3.29–3.22 (m, 3H), 3.04 (dd, J = 13.8, 4.7 Hz, 1H), 2.44 (s, 3H),
2.13 (m, 1H), 2.06–1.97 (m, 4H), 1.87 (s, 1H).
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
Nꢀ(4ꢀCyanophenyl)ꢀ
Nꢀ[2ꢀ[(1S,4S)ꢀ2,5ꢀdiazabicyclo[2.2.1]heptanꢀ2ꢀyl]ꢀ2ꢀoxoꢀethyl]ꢀ4ꢀmethylꢀ
benzenesulfonamide (16d).
Prepared by general method B using (1S,4S)ꢀ2ꢀBocꢀ2,5ꢀdiazabicyclo[2.2.1]heptane to afford the title
1
compound (28 mg, 75%) as a white powder. H NMR (300 MHz, Methanolꢀd
4
) δ 8.50 (s, 1H), 7.78–
7.60 (m, 2H), 7.64–7.47 (m, 2H), 7.50–7.25 (m, 4H), 4.43 (s, 2H), 4.18 (ddd, J = 8.3, 5.3, 3.3 Hz,
1H), 3.74–3.59 (m, 1H), 3.34–3.15 (m, 4H), 3.02 (ddd, J = 13.5, 5.3, 2.2 Hz, 1H), 2.43 (s, 3H), 2.18–
1.90 (m, 2H), 1.88– 1.76 (m, 1H)
Nꢀ[2ꢀ[(3
R)ꢀ3ꢀAminoꢀ1ꢀpiperidyl]ꢀ2ꢀoxoꢀethyl]ꢀNꢀ(4ꢀcyanophenyl)ꢀ4ꢀmethylꢀbenzenesulfonamide
(
16e).
Prepared by general method B using (R)ꢀ3ꢀBocꢀaminopiperidine to afford the title compound (14 mg,
1
3
7%) as a white powder. H NMR (300 MHz, DMSOꢀd
6
) δ 6.85–6.78 (m, 2H), 6.78–6.68 (m, 2H),
6.60–6.53 (m, 2H), 6.51 (d, J = 8.1 Hz, 2H), 3.89 (d, J = 16.5 Hz, 1H), 3.81 (d, J = 16.5 Hz, 1H), 3.38
(
d, J = 12.5 Hz, 1H), 3.29–3.08 (m, 1H), 2.96 (d, J = 13.7 Hz, 1H), 2.61–2.06 (m, 3H), 1.58 (s, 3H),
.08–1.00 (m, 1H), 0.92–0.78 (m, 2H).
1
Nꢀ[2ꢀ[(3
S)ꢀ3ꢀAminoꢀ1ꢀpiperidyl]ꢀ2ꢀoxoꢀethyl]ꢀNꢀ(4ꢀcyanophenyl)ꢀ4ꢀmethylꢀbenzenesulfonamide
(
16f).
Prepared by general method B using (S)ꢀ3ꢀBocꢀaminopiperidine to afford the title compound (18 mg,
1
4
8%) as a white powder. H NMR (300 MHz, DMSOꢀd
6
) δ 7.81–7.74 (m, 2H), 7.72–7.60 (m, 2H),
7.43–7.27 (m, 4H), 4.85 (d, J = 17.3 Hz, 1H), 4.70 (d, J = 17.3 Hz, 1H), 4.16–4.04 (m, 1H), 3.71 (d, J
=
12.5 Hz, 2H), 3.10–2.86 (m, 1H), 2.86 (s, 1H), 2.73–2.51 (m, 1H), 2.38 (s, 3H), 1.95–1.12 (m, 4H).
Nꢀ[2ꢀ(4ꢀAminoꢀ1ꢀpiperidyl)ꢀ2ꢀoxoꢀethyl]ꢀNꢀ(4ꢀcyanophenyl)ꢀ4ꢀmethylꢀbenzenesulfonamide
(
16g).
ACS Paragon Plus Environment