Systematic Study of the Glutathione Reactivity of N-Phenylacrylamides: 2. Effects of Acrylamide Substitution

Article abstract of DOI:10.1021/acs.jmedchem.0c00749

A comprehensive understanding of structure-reactivity relationships is critical to the design and optimization of cysteine-targeted covalent inhibitors. Herein, we report glutathione (GSH) reaction rates for N-phenyl acrylamides with varied substitutions at the α- and β-positions of the acrylamide moiety. We find that the GSH reaction rates can generally be understood in terms of the electron donating or withdrawing ability of the substituent. When installed at the β-position, aminomethyl substituents with amine pKa's > 7 accelerate, while those with pKa's < 7 slow the rate of GSH addition at pH 7.4, relative to a hydrogen substituent. Although a computational model was able to only approximately capture experimental reactivity trends, our calculations do not support a frequently invoked mechanism of concerted amine/thiol proton transfer and C-S bond formation and instead suggest that protonated aminomethyl functions as an electron-withdrawing group to reduce the barrier for thiolate addition to the acrylamide.

Full text of DOI:10.1021/acs.jmedchem.0c00749

Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE
Article
Systematic Study of the Glutathione (GSH) Reactivity of N-
Phenylacrylamides: 2. Effects of Acrylamide Substitution
Adam Birkholz, David J Kopecky, Laurie P Volak, Michael D. Bartberger, Yuping Chen,
Chris Tegley, Tara L. Arvedson, John D McCarter, Christopher H. Fotsch, and Victor J. Cee
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c00749 • Publication Date (Web): 23 Sep 2020
Downloaded from pubs.acs.org on September 24, 2020
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 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Systematic Study of the Glutathione (GSH)
Reactivity of N-Phenylacrylamides: 2. Effects of
Acrylamide Substitution
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
Adam Birkholz*, David J. Kopecky, Laurie P. Volak, Michael D. Bartberger, Yuping Chen,
Christopher M. Tegley, Tara Arvedson, John D. McCarter, Christopher Fotsch, and Victor J.
Cee
AMGEN Research, One Amgen Center Drive, Thousand Oaks, California 91320, United States
KEYWORDS Acrylamide, glutathione, intrinsic reactivity, conjugate addition, covalent
inhibitor, inductive effects, steric effects, transition state analysis.
ABSTRACT
A comprehensive understanding of structure-reactivity relationships is critical to the
design and optimization of cysteine-targeted covalent inhibitors. Herein we report glutathione
(GSH) reaction rates for N-phenyl acrylamides with varied substitutions at the - and -
positions of the acrylamide moiety. We find that the GSH reaction rates can generally be
understood in terms of the electron donating or withdrawing ability of the substituent. When
installed at the β-position, aminomethyl substituents with amine pKa’s > 7 accelerate, while
those with pKa’s < 7 slow the rate of GSH addition at pH 7.4, relative to a hydrogen substituent.
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 42
1
2
3
4
5
6
7
8
9
Although a computational model was able to only approximately capture experimental reactivity
trends, our calculations do not support a frequently invoked mechanism of concerted amine/thiol
proton transfer and C-S bond formation, and instead suggest that protonated aminomethyl
functions as an electron-withdrawing group to reduce the barrier for thiolate addition to the
acrylamide.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
INTRODUCTION
In recent years irreversible covalent small molecule inhibitors have received renewed
interest in the drug discovery field.1 Recently-approved EGFR- or EGFR/HER2-targeting
covalent inhibitor drugs for the treatment of cancers include afatinib2 (Boehringer Ingelheim),
neratinib3 (Puma Biotechnology), and osimertinib4 (AstraZeneca). These drugs contain an N-aryl
acrylamide structural motif that is responsible for covalent reaction with the non-catalytic
Cys797 residue of the target protein. (Figure 1). The discovery and optimization of afatinib,
neratinib, osimertinib, and other acrylamide covalent inhibitors5 has contributed to the increased
understanding of the structure-reactivity relationships of acrylamides with sulfur nucleophiles,
but we still possess limited data and understanding with regard to the impact of diverse
acrylamide substituents on acrylamide reactivity.6 This work represents a continuation of our
previously published systematic study of the impact of aromatic ring substitution on the reaction
rate of N-aryl acrylamides with glutathione (GSH)7, and seeks to define the impact of direct
acrylamide substitution on the reaction rate of N-phenyl acrylamides with glutathione (GSH)
(Figure 2).
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
2

Page 3 of 42
Journal of Medicinal Chemistry
1
2
3
4
O
O
5
6
7
8
H
N
O
N
N
O
N
O
O
N
N
N
N
N
N
N
H
N
N
H
N
HN
9
HN
O
HN
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
O
F
Cl
Cl
afatinib
neratinib
osimertinib
Figure 1. Marketed EGFR or EGFR/HER2-targeting covalent inhibitor drugs afatinib, neratinib,
and osimertinib each contain an N-aryl acrylamide as the cysteine-reactive moiety.
NH2
O
O
5 mM GSH
R
HO2C

O
R
HN
O

N
H
S
N
H
67 mM potassium
phosphate buffer in water/
1.0-1.5% DMSO,
R

R

NH
CO2H
pH 7.4, 37 oC
Figure 2. Reaction conditions for the conjugate addition of glutathione (GSH) to substituted N-
phenylacrylamides (1 M) in PBS buffer. Reaction progress was monitored by LC/MS analysis
of reaction samples at predetermined time intervals over a 120 min period.
Assay conditions for the determination of reaction rates of acrylamide substrates with
GSH are identical to those which we have reported previously.7 The experiment involves
incubation of 1 M of acrylamide substrate in 5 mM GSH in 67 mM pH 7.4 phosphate buffer
containing 1.0-1.5% DMSO at 37 °C. Reaction progress is monitored by analysis of peak area
ratios on an AB Sciex TripleTOF 5600 LC/MS system at an initial 2 min time point followed by
periodic analysis during an additional 120 min time period. From this data, the percent remaining
at the tested time points relative to zero can be calculated and then half-lives can be determined
by the assumption of pseudo first-order kinetics for the conjugate addition reaction. As before,
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 42
1
2
3
4
5
6
7
8
9
the maximal half-life for a 120 min incubation period is inferred to be 512 min based on the
assumption that no more than 15% depletion of parent molecule can be accurately measured at
this time point. The determination of an exact GSH t1/2 for the compounds listed as >512 min
would require additional experiments with a longer incubation time, which is beyond the scope
of this manuscript.
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
RESULTS AND DISCUSSION
-substituted acrylamides. Eleven distinct -substituted N-phenyl acrylamides were
synthesized and tested (Table 1). Fluorine- (2) and chlorine- (3) substitution exhibited striking
differences in reactivity relative to hydrogen (1), with greatly reduced and accelerated reaction
rates, respectively. The reduced reactivity of -fluoro acrylamides relative to their non-
fluorinated counterparts have been described previously8 and the mesomeric electron-donating
ability of fluorine is believed to dominate the inductive electron-withdrawing effects. Methyl
substitution (4) reduces reactivity with GSH relative to hydrogen (1, GSH t1/2 = 179 min), and
derivatives containing mildly electron withdrawing groups exhibit similar reactivity (5-8); a
derivative containing the strongly electron withdrawing trifluoromethyl group (9, GSH t1/2 = 15
min) exhibits enhanced reactivity. The -aminomethyl acrylamides (10 GSH t1/2 <2 min; 11
GSH t1/2 = 10 min) exhibited very rapid reactions with GSH, with the trend of the higher amine
pKa being associated with the faster reaction rate (10 calc. pKa=8.0; 11 calc. pKa=5.7). As
reported previously with other substrates, the addition of GSH to -aminomethyl acrylamides
10 and 11 occurs with concomitant loss of the amino group, yielding the -thiomethyl
acrylamide product.5f
Table 1. GSH Reaction Rates for -Substituted-N-Phenylacrylamides
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
4

Page 5 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
O
N
H
R
7
8
9
compound
R
GSH t1/2 (min)
kpseudo1st (min-1 x 10-3)
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
H
F
179
>512
3
3.87
<1.35
231
3
Cl
4
CH3
>512
>512
>512
241
>512
15
<1.35
<1.35
<1.35
2.88
5
Bn
6
CH2OH
CH2OMe
CH2CH2OH
CF3
7
8
<1.35
46.2
9
10
11
CH2NMe2
<2
>347
69.3
10
N
O
12
Ph
>512
<1.35
Mechanistic possibilities for the exceptional rate acceleration of GSH addition associated
with the -DMAM group on compound 10 are shown in Figure 3. It is important to note that
under the aqueous buffered conditions of the GSH addition assay (pH 7.4), both nucleophile and
electrophile are in equilibrium with their respective neutral and charged forms, with the highly
reactive thiolate constituting a minority of the total nucleophile population (glutathione Cys pKa
9.2-9.4)9 and the protonated ammonium species a majority of the total electrophile population
(calculated pKa 8). The reaction of the s-cis acrylamide conformer could occur via three distinct
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 42
1
2
3
4
5
6
7
8
9
arrangements – (a) nucleophilic addition of the thiolate to the ammonium acrylamide species; (b)
concerted deprotonation/nucleophilic addition of thiol by neutral amino acrylamide10; (c)
nucleophilic addition of thiolate to the neutral amino acrylamide, with possible acceleration due
to the amino group engaging in an intramolecular hydrogen bond with the amide NH. For the
reaction of the s-trans acrylamide conformer, three similar arrangements are possible - (d)
nucleophilic addition of the thiolate to the ammonium acrylamide species; (e) concerted
deprotonation/nucleophilic addition of thiol by neutral amino acrylamide10; (f) nucleophilic
addition of thiolate to the neutral amino acrylamide. Computational assessment of different
transition structure arrangements is provided in the Computational Discussion section.
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
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
6

Page 7 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
-SR
O-
SR
O
O
(a)
O
SR
RS-
N
H
N
H
N
+Me2NH
N
H
H
H
H
H
N+
N+
N+
7
8
9
-DMAM TS1

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
O
O
N
(b)
R
O-
SR
RSH
S
N
O
SR
H
H
+Me2NH
H
N
N
N
N
H
H
H
N+
-DMAM TS2

O
-SR
O-
N
SR
RS-
O
SR
O
N
H
N
H
(c)
N
H
+Me2NH
N
H
N
H
N
-DMAM TS3

H
H
O
N+
O- N+
O
H
(d)
O
N+
RS-
N
H
N
H
N
H
+Me2NH
N
H
SR
SR
-SR
-DMAM TS4

H
O
O- N+
O
O
RSH
N
H
N
N
H
(e)
N
H
N
H
+Me2NH
N
H
SR
SR
S
R
-DMAM TS5

O
O-
N
O
O
RS-
H
N
H
N
N
H
N
(f)
N
H
N
+Me2NH
H
SR
SR
S-
R
-DMAM TS6

Figure 3. Mechanistic possibilities for the reaction of -dimethylaminomethyl acrylamide 10
with GSH. (a-c) s-cis conformation (d-f) s-trans conformation. The nucleophilic addition is
assumed to be the rate limiting step.
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 42
1
2
3
4
5
6
-substituted acrylamides. We turned our attention to the reaction of GSH with acrylamides
substituted at the -position of the acrylamide (Table 2). In comparison to the moderately
7
8
9
reactive unsubstituted parent N-phenylacrylamide 1, as expected, a methyl substituent at the -
position (compound 13) resulted in substantially reduced reaction with GSH. This has been
suggested to arise from either an increase in the steric hindrance at the site of nucleophilic attack
or from deactivation of the acrylamide via hyperconjugation of the methyl group’s electron
density into the acrylamide antibonding -orbital. Fluoromethyl substitution was observed to
lead to a wide range of GSH reaction rates, with trifluoromethyl acrylamide 14 and
difluoromethyl acrylamide 15 both highly reactive (t1/2 = 16 min and 44 min respectively), while
monofluoromethyl acrylamide 16 showed behavior like the parent methyl species (13) in this
assay. We suggest that the order of reactivity of CF3>CF2H>CH2F~CH3 reflects increased partial
positive charge at the acrylamide -carbon, that can both outweigh the steric bulk due to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
substitution at the -carbon center, or reduce the extent of hyperconjugative stabilization. Other
simple substitutions at the acrylamide -carbon were non-reactive under the conditions of our
assay (Table 2, entries 17-19), indicating that aryl groups, acetic acid and acetamide groups are
not sufficiently electron-deficient to overcome the increased steric constraints and/or
hyperconjugative stabilization imparted by these groups on the acrylamide system.
Table 2. GSH Half-Lives for Non-Amino -Substituted-N-Phenylacrylamides
O
R
N
H
Compound
R
H
GSH t1/2 (min)
179
kpseudo1st (min-1 x 10-3)
1
3.87
13
CH3
>512
<1.35
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
8

Page 9 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
14
15
16
17
18
19
CF3
16
43.3
15.8
CF2H
CH2F
44
>512
>512
>512
>512
<1.35
<1.35
<1.35
<1.35
Ph
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
CH2CO2H
NH
O
The -dimethylaminomethyl (DMAM) substituent has long been known to accelerate
nucleophilic addition relative to the corresponding -methyl substituent.10 We conducted an
extensive investigation of -position aminomethyl substituents spanning a range of steric and
electronic properties to understand the relationship between structure and intrinsic reactivity for
this structural motif and this data is arranged by decreasing calculated pKa value (Table 3). The
parent N,N-dimethylaminomethyl (DMAM) acrylamide 23 showed a modest level of reactivity
with GSH (t1/2 = 116 min), as compared to methyl acrylamide 13 (Table 2) which showed a GSH
t1/2
=
>512 min.
Consistent with the observations of prior studies, the N,N-
dimethylaminomethyl acrylamide is unique in that it exhibits qualitatively similar reactivity to
cysteine nucleophiles as the parent completely unsubstituted acrylamide (1, GSH t1/2 = 179
min).10 Acrylamides with aminomethyl groups at the -position possessing calculated pKa
values >7 exhibited comparable GSH reactivities to 23 (Table 3, compounds 20, 21, 22, 24).
Acrylamides with aminomethyl groups at the -position possessing pKa values calculated by
ACD Percepta <7 exhibited significantly reduced rates of reaction with GSH (Table 3,
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 42
1
2
3
4
5
6
7
8
9
compounds 25-32). Finally, we note that the homologated derivative 33 shows that increasing
the number of intervening atoms between the amino group and the acrylamide by one carbon can
result in a significant reduction in reactivity (compare with 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
Table 3. GSH Half-Lives for Amino -Substituted-N-Phenylacrylamides
O
R
N
H
Compound
R
GSH t1/2
(min)
kpseudo1st
calc. pKaa
(min-1 x 10-3)
20
21
NEt2
170
162
4.08
4.28
9.2
9.2
N
22
177
3.92
8.4
N
23
24
NMe2
116
181
5.97
3.83
8.2
7.3
N
OMe
25
26
326
2.13
6.7
6.4
N
N
F
>512
<1.35
O
27
28
452
<1.35
<1.35
6.1
5.7
N
F
>512
N
CO2Me
O
S
29
30
>512
>512
<1.35
<1.35
4.6
4.4
N
O
F
F
N
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
10

Page 11 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
31
32
>512
<1.35
<1.35
2.8
0.5
N
N
F
F
>512
F
N
F
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
33
CH2NMe2
443
1.56
9.2
a’Classic’ pKa calculated in ACD Percepta 2017.2.1
The results in Table 3 suggest the existence of a direct correlation between the basicity of
the -aminomethyl moiety and the intrinsic reactivity of the acrylamide. Indeed, when the
calculated pKa values of all examined -aminomethyl acrylamides studied are plotted versus
their corresponding half-lives in 5 mM GSH it is observed that amino groups with calculated
pKa values >6 have a relationship with the half-lives that is well described by a sigmoidal curve
(Figure 4). These results suggest that protonation of the amine plays a key role in the intrinsic
reactivity of a -aminomethyl-substituted acrylamide, and that reactivity can be fine-tuned by
choosing an amino substituent with an appropriate calculated pKa value.
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 42
1
2
3
4
5
6
Figure 4. Plot of calculated pKa (ACD Percepta 2017.2.1, ‘classic’) vs. GSH t1/2 for -
aminomethylacrylamide analogs. GSH half-life of 512 min (dotted line) is the assay maximum;
five points are >512 min.
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
Mechanistic possibilities for the rate acceleration of GSH addition associated with the -
aminomethyl group are shown in Figure 5. We propose that six different arrangements may be
possible. The reaction of the s-cis acrylamide conformer could occur via three distinct
arrangements – (a) nucleophilic addition of the thiolate to the ammonium acrylamide species; (b)
concerted deprotonation/nucleophilic addition of thiol by neutral amino acrylamide as originally
proposed by Tsou, et al.10; (c) nucleophilic addition of thiolate to the neutral amino acrylamide.
For the reaction of the s-trans acrylamide conformer, three arrangements are plausible - (d)
nucleophilic addition of the thiolate to the ammonium acrylamide species; (e) concerted
deprotonation/nucleophilic addition of thiol by neutral amino acrylamide10 and (f) nucleophilic
addition of thiolate to the neutral amino acrylamide. Computational assessment of different
transition structure arrangements is provided in the Computational Discussion section.
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
12

Page 13 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
(a)
R
-S
O
O-
RS-
SR
O
N+
H
N+
H
N+
N
H
N
N
H
H
H
8
9
-DMAM TS1

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
R
(b)
S
H
O-
O
SR
RSH
O
N+
H
N
N
N
N
H
N
H
H
-DMAM TS2

(c)
R
-S
O
RS-
O-
O
SR
N
N
N
N
H
N
H
N
H
-DMAM TS3

O
O
(d)
O-
O-
O-
RS-
SR
SR
SR
N
H
N
H
N+
H
N
H
S-
H
N+
H
N+
R
-DMAM TS4

O
O
(e)
RSH
N
H
N
H
N+
H
N
H
S
R
N
H
N
-DMAM TS5

O
(f)
O
RS-
N
H
N
H
N
N
H
S-
N
N
R
-DMAM TS6

Amgen Proprietary - Confidential
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 42
1
2
3
4
5
Figure 5. Mechanistic possibilities for the reaction of -dimethylaminomethyl acrylamide 23
with GSH. (a-c) s-cis conformation (d-f) s-trans conformation.
6
7
COMPUTATIONAL DISCUSSION
8
9
Background/Methods
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 our previous work7, we found that barrier heights with geometries and free energies
computed using IEFPCM-B3LYP/6-311+G(d,p) showed good correlation with experimental
GSH reaction rates across a large number of N-phenyl acrylamides. In the present work, with the
modifications made directly on the acrylamide reactive site, acrylamide transition state
optimization was found to be unstable with B3LYP, frequently failing to find a transition state or
converging to a transition structure other than the one of interest. Transition state optimization
with the ωB97X-D functional was found to be significantly more reliable, and recently published
work has shown that ωB97X-D is better suited to describing acrylamide reaction energetics
relative to a CCSD(T) reference11. For these reasons, IEFPCM-ωB97X-D/6-311+G(d,p) was
used for all calculations unless otherwise noted.
To compare the experimental half-lives with the results from DFT, we compute
experimental half-life barriers assuming 1st order reaction kinetics12. This translates to a range for
the experimental results of approximately 21.4 to 24.7 kcal/mol. Before attempting to answer
which of the possible mechanisms outlined in figures 3 and 5 is central to the reaction of
acrylamides with methane thiol, we performed a preliminary benchmark on some of the simple
acrylamides from tables 1 and 2, where the rigidity of the acrylamide substrate allowed for an
exhaustive study of all possible conformers. We also looked the effect of including explicit water
molecules in our calculations, which has been shown to dramatically improve the prediction of
equilibrium properties like pKa13. The results of this benchmark are summarized below, but for
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
14

Page 15 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
the full details of this benchmark, please see the computational section of the supplementary
information.
7
8
9
1) When comparing the barriers between the lowest energy reactant and transition state
structures, correlation between calculation and experiment was positive, but
extremely poor (R2 of 0.14 across 5 structures, acrylamides 1, 3, 9, 14 and 15).
Acrylamide 9 was a significant outlier.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2) Fair correlation (R2 of 0.75 across the same 4 structures, excluding acrylamide 9) can
be achieved by using a fully implicit solvation model, but considering only
conformers which do not have potential hydrogen bonding sites occupied by
intramolecular interactions, even if those conformers are higher in energy.
3) An addition of a single water molecule hydrogen-bound to the acrylamide’s amine
improved correlation dramatically, achieving an R2 of 0.88 across all 5 structures.
This improvement is mostly due to a ~5-10 kcal/mol difference in the calculated
acrylamide 9 barrier, where the bulky tri-fluoromethyl group occupies the N-H
hydrogen bonding site in the fully implicit solvation model.
4) Other potential hydrogen bonding sites (e.g. the partial positive charge on the -
carbon at the transition state, or the thiolate ion) were not stable between reactants
and transition states, so inclusion of additional water molecules in a systematic way
did not appear to be feasible.
These results suggest that explicit solute/solvent interactions are necessary to accurately
describe the differences in methane thiolate reactivity between - and - substituted acrylamides,
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 42
1
2
3
4
5
6
7
8
9
however an approximate correlation can be achieved with a fully implicit solvation model by
looking at conformers which do not have potential solute-solvent interaction sites occupied by
intramolecular interactions.
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
Addition of Methanethiol to Neutral Aminomethyl Acrylamide
We began our computational exploration of -aminomethyl and -aminomethyl
substituted acrylamides with an investigation of a frequently invoked mechanism involving the
transfer of the neutral thiol hydrogen to the neutral -amine concomitant with C-S bond
formation (-DMAM TS2 or TS5, Fig. 3b/3e; -DMAM TS2 or TS5, Fig. 5b/5e). For the -
aminomethyl-substituted acrylamide, we attempted to locate a transition state using the
ConnectivityTS method14 which is particularly well suited to locating transition states involving
the concerted breaking/forming of multiple bonds. Although we were able to use this method to
readily determine prohibitively high energy transition states corresponding to the concomitant
transfer of the thiol proton to the carbon (directly resulting in a keto, e.g. R1R2CC=O product),
and the concomitant transfer of a proton to the carbonyl oxygen (directly resulting in an enol,
e.g. R1R2C=COH product), we could find no evidence for a transition structure that would result
in the zwitterionic, enolate intermediate (-DMAM TS2 or TS5).
To further probe this proposed mechanism, we located a geometry of the zwitterionic
enolate intermediate with the ammonium N-H oriented towards the sulfur atom (Figure 6-v).
From this geometry, a 2-dimensional relaxed potential energy scan was performed along the C-S
bond and amine N-H bond coordinates using the IEFPCM-ωB97X-D/6-31+G(d) level of theory.
The resulting surface clearly demonstrates a preference for a 2-step mechanism involving the
transfer of a proton from the thiol to the amine, followed by addition of methane thiolate to the
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
16

Page 17 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
positively charged acrylamide with no evidence of an alternative, direct pathway (Figure 6). A
similar result was observed for -DMAM: the unphysical transition states resulting in the enol
and keto products were readily located, while no evidence of the transition structure (Fig. 5b/5e)
to produce the zwitterionic intermediate could be found by the Connectivity TS method nor 2-
dimensional PES scan (See Figure S4 in the SI). Therefore, our results do not support the
frequently invoked mechanism of a concerted reaction between neutral thiol and neutral amine
(Fig 3b/e and Fig. 5b/e). We note that even though Figure 6 shows a clear two step mechanism,
the rate limiting step is the final addition of methane thiolate to the acrylamide, and so the
deprotonation of the thiol and the protonation of the acrylamide are more likely to occur
independently of one another at the reaction conditions of pH 7.4.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. 2-dimensional IEFPCM-ωB97X-D/6-31+G(d) potential energy surface for the
reaction of -dimethylamino methyl acrylamide with methanethiol. Each contour corresponds to
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 42
1
2
3
4
5
6
7
8
9
a difference of 2 kcal/mol. The path i  ii  iii  iv  v (green arrow) corresponds to the two-
step formation of the product, while no saddle point for the direct route i  v (red arrow) is
observed.
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
Addition of Methane Thiolate to Neutral or Protonated Aminoalkyl Acrylamides
To gain insight into the transition structures that do contribute to thiolate addition to
aminoalkyl acrylamides, we selected a series of relatively simple -substituted acrylamide
substrates (20, 21, 23 and 33) with high predicted pKAs to investigate whether barrier heights for
methane thiolate addition to either the neutral or charged acrylamide are more consistent with the
values computed in our earlier validation. Additionally, barriers were computed for acrylamide
32, which is structurally similar to 21, but with a significantly lower predicted pKA. Based on the
results of our simple-substituent benchmark (see the SI), we decided to limit our study to the s-
cis-anti barriers as the anti addition showed slightly better agreement with experiment, and the s-
cis conformer was found to be lower in free energy for -DMAM and is expected to be the most
stable conformer for all of the -substituted acrylamides studied.
Barrier heights were computed by first performing an exhaustive conformational search
using Macromodel15 with the acrylamide constrained to be in the s-cis conformer in both the
neutral and positive protonation states, resulting in ~10-100 unique conformers per acrylamide.
These conformers were further refined using unconstrained DFT. To generate guess transition
states, methane thiolate was built onto the DFT-optimized minimum energy structures in the anti
orientation using a z-matrix molecule builder program written in Python. This was repeated for
each minimum with the dihedral angle for the C-S-C-C bond rotated by 180 degrees to add the
methane thiolate to the opposite face of the  carbon. These guess conformers were then refined
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
18

Page 19 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
by constrained minimization with all bond lengths frozen in order to relax any clashes between
methyl thiolate and the substituent on the  carbon. These refined conformers were then fully
optimized without constraints, and Hessian calculations were performed to confirm that the
imaginary frequency vibrational mode corresponded to methane thiolate addition. Free energies
at 310.15K were estimated using scaled (scale factor 0.979) harmonic frequencies, and barriers
were calculated as the free energy difference between the lowest free energy transition state
conformer and the lowest free energy reactant conformer. Barriers for these molecules, plus the 4
acrylamides from the simple substituent benchmark, are shown in Table 4, along with the 4 non-
alkylamino structures with half-lives that fell within the experimental range from the earlier
benchmark study.
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 4. Calculated barrier heights for the s-cis-anti transition state for selected acrylamides.
Separate barriers are shown for the charged alkylammonium acrylamides depending on whether
intramolecular interactions with amine N-H are allowed (NH engaged) or prohibited (NH free).
EXPERIMENT
CALCULATED
Charged,
NH free
21.5
G1
Substituent
1 H
t1/2
(min)
Neutral
Charged,
NH engaged
21.5
179
24.1
21.6
22.6
21.5
17.6
17.9
16.5
24.1
24.0
24.0
23.6
22.3
22.2
3
9
3
16
44
17.6
17.9
16.5
12.7
12.8
13.9
15.7
6.4
17.6
17.9
18.6
17.8
18.2
18.5
19.1
12.1
-Cl
-CF3
-CF2H2
15
20
21
23
33
10
32
23.3
170
162
116
443
<2
24.1
24.1
23.9
24.7
< 21.4
> 24.7
-CH2NEt2
-CH2Pyrrolidine
-CH2NMe2
-(CH2)2NMe2
-CH2NMe2
-CH2Pyrrolidine-F4
>512
10.9
15.4
1 Experimental G is calculated assuming pseudo-first order kinetics. 2 For acrylamide 15, the difference between
engaged and free is distinguished by the orientation of the polar CH on the substituent, with the engaged barrier
used in the Neutral column.
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 42
1
2
3
4
5
6
7
8
9
Acrylamide 32 is predicted to be the most reactive of the 5 -alkylamino acrylamides by
either the charged or the neutral mechanism, however the neutral barrier is approximately 11
kcal/mol higher than the lowest energy charged barrier. This is in good agreement with the
observed pKa dependence on the reactivity of acrylamides at pH 7.4 given the difference in
estimated pKa between acrylamides 32 and 21.
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
For the charged acrylamides, the lowest energy transition states universally had the amine
N-H interacting with the incoming thiolate (Figure 7, charged, NH engaged). These structures
show positive correlation to experimental values (Figure 8ii), but have barriers that are unusually
low relative to the non-amine acrylamides. We also observed that the energy barriers for the
neutral alkyl-amino structures are more consistent with the barriers computed on the non-amino
acrylamides (Figure 8i). This appears to be from a favorable cancelation of errors, since the
correlation between calculation and experiment for just the neutral alkylamino structures is
negative. We also considered barriers involving the lowest energy transition state in which the
ammonium N-H was not making any intramolecular interactions with either the thiolate or the
acrylamide (Figure 7, charged, NH free). These barriers have a similar positive correlation to
experiment (Figure 9iii) as the charged, NH engaged barriers, and when combined with the data
from the non-amine data, show a weak positive correlation to experiment across all 8
acrylamides in Table 4. Both sets of charged transition structures underestimate the barrier
heights relative to the simple substituents, indicating that there is still a component of reactivity
not accounted for by the current model. However, since both charged models show positive
correlation to experiment, it appears the observed acceleration of the -alkylamines can be
explained by the indirect, inductive effect of the charged amine group without needing to
consider alternative mechanisms of reaction.
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
20

Page 21 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Implicit solvent DFT is able to approximate the relative reactivity of acrylamides with
similar substituents at the - and - positions, but large differences can introduce significant
error, possibly due to changes in the free energy of solvation which are not captured by implicit
solvent methods. Some of this error can be mitigated by choosing transition structures which
limit intramolecular interactions; however, it is clear that more sophisticated strategies for
calculating reactivity in solvent, such as explicit solvent QM molecular dynamics, might be
necessary in order to achieve a high level of predictive accuracy across a wide variety of
covalent inhibitor structures.
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
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 42
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
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
22

Page 23 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 7. Reactant and transition structures for a selection of positively charged - and -
alkylamine acrylamides. Separate transition structures are shown depending on whether
intramolecular interactions involving the ammonium are allowed (NH engaged) or prohibited
(NH free).
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 9. Correlation plots comparing the i) Neutral, ii) Charged, NH engaged and iii) Charged,
NH free barriers to the simple substituent benchmark data. Blue dashed trendlines are over all 8
points, while grey dotted trendlines are for the separated simple or alkyl-amino data points.
Equations and R2 are for the blue, dashed trendline. The Neutral barriers have the best
correlation to experiment across the entire set; however, the correlation among just the beta-
alkylamines is strongly negative.
CONCLUSIONS
This systematic study has established benchmark values for rates of GSH addition to a
wide variety of substituted acrylamides. We find that the impact of a substituent on GSH reaction
rates can generally be understood in terms of the electron donating or withdrawing ability of the
substituent. Aminomethyl groups directly attached to the acrylamide were extensively studied.
These substituents exhibit a substantial accelerating effect at both the - and - acrylamide
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 42
1
2
3
4
5
6
7
8
9
positions in a manner that is positively correlated with amine pKa. Density functional quantum
mechanical characterization of transition structure arrangements suggests that rapid nucleophilic
addition likely occurs between the positively charged ammonium acrylamide and negatively
charged thiolate species, and that the observed acceleration is likely due to inductive electron
withdrawal of the ammonium on the acrylamide system, rather than by a direct structural or
mechanistic change in the transition structure relative to the unsubstituted parent. This
comprehensive data should find utility in the design of cysteine-targeting covalent inhibitor drug
molecules containing an N-arylacrylamide as the cysteine-reactive group.
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
AUTHOR INFORMATION
Corresponding Authors
*A.B.: e-mail abirkhol@amgen.com; phone, 805-447-0617
Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval
to the final version of the manuscript.
Funding Sources
The authors declare no competing financial interest.
ABBREVIATIONS
DMSO, dimethylsulfoxide; GSH, glutathione; PBS, phosphate buffer; LCMS liquid
chromatography-mass
spectrometry;
ESI,
electrospray
ionization;
DMAM,
dimethylaminomethyl; B3LYP, Becke three-parameter Lee-Yang-Parr; IEFPCM, Integral
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
24

Page 25 of 42
Journal of Medicinal Chemistry
1
2
3
4
equation formalism polarizable continuum model; DCM, dichloromethane; EtOAc, ethyl acetate;
5
6
TEA, triethylamine; DIPEA, N,N-diisopropylethylamine, DMF, N,N-dimethylformamide; DMA,
7
8
9
N,N-dimethylacetamide;
b]pyridinium 3-oxid
HATU,
(1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-
TBTU, O-(benzotriazol-1-yl)-N,N,N′,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
hexafluorophosphate);
tetramethyluronium tetrafluoroborate; EDC, N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide
hydrochloride; DABCO, 1,4-diazabicyclo[2.2.2]octane; TBDPS, tert-butyldiphenylsilyl; HPLC,
high performance liquid chromatography.
EXPERIMENTAL SECTION
All materials were acquired from commercial suppliers and used without additional purification.
The following acrylamide was commercially available: N-phenylacrylamide (1) and (E)-N,3-
diphenylprop-2-enamide (17). All reactions were run under an inert atmosphere (nitrogen or
argon) using anhydrous solvents were acquired from Sigma-Aldrich. Purifications by flash
column chromatography on silica gel were perfomed on a CombiFlash Companion machine
(Teledyne Isco) using pre-packed RediSep normal-phase 35-60 m silica gel columns with UV
detection at 220 nM and 254 nM. Reverse phase HPLC purifications were performed on an
Agilent 1260 Infinity preparative purification system using a Phenomenex Gemini 30 x 150 mm
5 m C18 column. Gradient elution with a mixture of water containing 0.1% TFA and
acetonitrile containing 0.1% TFA was performed. Preparative supercritical fluid (SFC)
purifications were performed on a Waters preparative SFC system using a Chiralpak AD-H 30 x
150 mm 5 m column. Elution was perfomed using 25% MeOH/CO2 (containing 20 mM NH3)
at a flow rate of 120 mL/min. All final compounds were determined to be >95% purity by
analysis on an Agilent 1100 series HPLC machine using a Phenomenex Synergi column (2 x 50
mm, 3 min run, flow rate = 1.0 mL/min, gradient elution of 0-95% acetonitrile containing 0.1%
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 42
1
2
3
4
5
6
TFA/water containing 0.1% TFA) with UV detection at 254 nM wavelength. Low-resolution
mass spectra were obtained on an Agilent 1100 series LCMS machine with electrospray
7
8
9
1
ionization (ESI) mode and UV detection at 254 nM. H NMR spectra were obtained using a
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
Bruker AV-400 spectrometer (400 MHz) at room temperature. Chemical shifts are reported in
ppm relative to the chemical shift of a deuterated solvent (DMSO-d6=2.50 ppm; CDCl3=7.26
ppm ; CD3OD=3.31 ppm). 1H NMR peak data is reported as follows: chemical shift, multiplicity
(s = singlet, br s = broad singlet, d = doublet, t = triplet, q = quartet, m = multiplet), coupling
constant(s), number of protons. The values of J reported throughout are approximate (+/- 0.5
Hz).
General procedure for the synthesis of α-substituted N-phenylacrylamides: HATU (1 equiv)
was added to an ice-cooled solution of the acrylic acid (1.1 equiv) and DIPEA (1.5 equiv) in
THF (0.15 M concentration). After 15 min, aniline (1 equiv) was added dropwise and the
resulting mixture was warmed to room temperature and stirred overnight. The reaction was
diluted with EtOAc and washed with water. The organic layer was dried over anhydrous sodium
sulfate and concentrated. The residue was purified by flash column chromatography on silica gel
to provide the acrylamide product.
2-Fluoro-N-phenylacrylamide (2). 1H NMR (400 MHz, DMSO-d6) δ 10.26 (br. s, 1H), 7.71 (d,
J=7.6 Hz, 2H), 7.34 (t, J=7.9 Hz, 2H), 7.13 (t, J=7.2 Hz, 1H), 5.71 (dd, J=47.7, 3.5 Hz, 1H), 5.42
(dd, J=15.7, 3.6 Hz, 1H). LCMS-ESI (POS.) M/Z: 166.0 (M+H)+.
1
2-Chloro-N-phenylacrylamide (3). H NMR (400 MHz, DMSO-d6) δ 10.19 (s, 1H), 7.65 (dd,
J=8.0, 2.4 Hz, 2H), 7.30-7.39 (m, 2H), 7.12 (td, J=7.4, 2.1 Hz, 1H), 6.42 (t, J=2.5 Hz, 1H), 6.08
(t, J=2.5 Hz, 1H). GCMS M/Z: 181.1 (M)+.
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
26

Page 27 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
N-Phenylmethacrylamide (4). 1H NMR (400 MHz, DMSO-d6) δ 9.78 (s, 1H), 7.58 (d, J=8 Hz,
2H), 7.29 (t, J = 7.6 Hz, 2H), 7.06 (t, J = 7.2 Hz, 1H). 5.74 (s, 1H), 5.48 (s, 1H), 1.91 (s, 3H).
LCMS-ESI (POS.) M/Z: 162.2 (M+H)+.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
2-Benzyl-N-phenylacrylamide (5). H NMR (400 MHz, CDCl3) δ 9.90 (s, 1H), 7.60-7.69 (m,
2H), 7.14-7.41 (m, 7H), 7.03 (t, J=.7.4 Hz, 1H), 5.85 (s, 1H), 5.46 (s, 1H), 3.67 (s, 2H). LCMS-
ESI (POS.) M/Z: 238.2 (M+H)+.
2-(Hydroxymethyl)-N-phenylacrylamide (6). To a pressure release vial containing N-
phenylacrylamide (1.0 g, 6.8 mmol), DABCO (762 mg, 6.8 mmol), phenol (160 mg, 1.7 mmol)
and tert-butanol (2 mL) was added 37% aqueous formaldehyde solution (556 L, 7.5 mmol).
The vial was sealed and heated at 55 °C overnight. The reaction was allowed to cool to room
temperature and was partitioned between saturated aqueous ammonium chloride and EtOAc. The
organic extract was washed with brine (2X), dried over anhydrous sodium sulfate and
concentrated. The residue was purified by flash column chromatography on silica gel to afford 6
1
(267 mg, 22% yield) as a white solid. H NMR (400 MHz, DMSO-d6) δ 9.84 (s, 1H), 7.66 (d,
J=7.8 Hz, 2H), 7.30 (t, J=7.6 Hz, 2H), 7.06 (t, J=7.1 Hz, 1H), 5.86-5.90 (m, 1H), 5.65 (q, J=1.8
Hz, 1H), 5.12 (t, J=5.6 Hz, 1H), 4.22 (d, J=5.1 Hz, 2H). LCMS-ESI (POS.) M/Z: 178.1 (M+H)+.
1
2-(Methoxymethyl)-N-phenylacrylamide (7). H NMR (400 MHz, DMSO-d6) δ 9.90 (s, 1H),
7.67 (d, J = 8.2 Hz, 2H), 7.34 (m, 1H), 7.31 (t, J=7.2 Hz, 2H), 5.95 (s, 1H), 5.68 (s, 1H), 4.15 (d,
J=1.5 Hz, 2H), 3.30 (s, 3H). LCMS-ESI (POS.) M/Z: 192.2 (M+H)+.
4-Hydroxy-2-methylene-N-phenylbutanamide (8). Step 1. To an ice-cooled solution of 4-
((tert-butyldiphenylsilyl)oxy)-2-methylenebutanoic acid (1.0 g, 2.8 mmol) in DCM (12 mL) was
added oxalyl chloride (2.0 M solution in THF, 2.1 mL, 4.2 mmol) via syringe followed by a
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 42
1
2
3
4
5
6
7
8
9
catalytic amount of DMF via syringe. The resulting mixture was warmed to room temperature
and stirred for 2 h, then was concentrated to afford the crude acid chloride. The crude acid
chloride was dissolved in DCM (5 mL), cooled to 0 °C, and treated with aniline (122 L, 1.3
mmol) dropwise via syringe. The reaction mixture was warmed to room temperature and a
catlytic amount of DMAP was added. After stirring at room temperature overnight the reaction
was concentrated and the residue was purified by flash column chromatography on silica gel to
afford 4-((tert-Butyldiphenylsilyl)oxy)-2-methylene-N-phenylbutanamide (310 mg, 54% yield)
as a yellow oil. LCMS-ESI (POS.) M/Z: 430.2 (M+H)+. Step 2. To a solution of 4-((tert-
Butyldiphenylsilyl)oxy)-2-methylene-N-phenylbutanamide (310 mg, 0.72 mmol) in THF (5 mL)
was added TBAF (1.0 M solution in THF, 900 L, 0.90 mmol) via syringe. The reaction mixture
was stirred at room temperature for 4 h, then was concentrated and the residue was purified by
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
flash column chromatography on silica gel to afford 8 (101 mg, 73% yield) as a yellow oil. H
NMR (400 MHz, DMSO-d6) δ 9.89 (s, 1H), 7.67 (d, J=8.2 Hz, 2H), 7.30 (t, J=7.9 Hz, 2H), 7.00-
7.12 (m, 1H), 5.82 (s, 1H), 5.51 (s, 1H), 4.67 (t, J=5.2 Hz, 1H), 3.50-3.59 (m, 2H), 2.45-2.50 (m,
2H). LCMS-ESI (POS.) M/Z: 192.0 (M+H)+.
N-Phenyl-2-(trifluoromethyl)acrylamide (9). To a solution of 2-(trifluoromethyl)acrylic acid
(301 mg, 2.15 mmol) in DCM (2 mL) was added oxalyl chloride (376 L, 4.3 mmol) via syringe
followed by a catalytic amount of DMF. The resulting mixture was stirred at room temperature
for 1 h and then was concentrated. The crude acid chloride was dissolved in DMF (2 mL) and
aniline (98 L, 1.07 mmol) was added via syringe. After stirring for 1 h at room temperature the
reaction was purified directly by reverse phase preparatory HPLC. The purified product was
dissolved in 1:1 DCM/MeOH and desalted by elution through an Agilent Stratospheres SPE
cartridge to afford 9 (3.7 mg, 2% yield) as a white solid. 1H NMR (400 MHz, CD3OD) δ 7.65 (d,
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
28

Page 29 of 42
Journal of Medicinal Chemistry
1
2
3
4
J=8.0 Hz, 2H), 7.35 (t, J=7.5 Hz, 2H), 7.10-7.19 (m, 1H), 6.80-6.95 (m, 2H). LCMS-ESI (POS.)
5
6
M/Z: 216.2 (M+H)+.
7
8
9
2-((Dimethylamino)methyl)-N-phenylacrylamide (10). Step 1. To an ice-cooled solution of 6
(50 mg, 0.28 mmol) in DMF (110 L) and ether (500 L) was added phosphorus tribromide (38
mg, 0.14 mmol). The resulting slurry was stored in a 0 °C freezer overnight, then was partitioned
between water and EtOAc (4X). The combined organic extracts were washed with water (1X)
and brine (1X), dried over anhydrous sodium sulfate and concentrated. The residue was purified
by flash column chromatography on silica gel to afford 2-(Bromomethyl)-N-phenylacrylamide
(19 mg, 28% yield) as a white solid. 1H NMR (400 MHz, DMSO-d6) δ 10.09 (s, 1H), 7.69 (d,
J=7.8 Hz, 2H), 7.33 (t, J=7.7 Hz, 2H), 7.09 (t, J=7.1 Hz, 1H), 5.97 (s, 2H), 4.42 (s, 2H). LCMS-
ESI (POS.) M/Z: 240.1 (M+H)+. Step 2. To a solution of 2-(Bromomethyl)-N-phenylacrylamide
(16 mg, 0.07 mmol) in THF (800 L) was added dimethylamine (2.0 M solution in THF, 67 L,
0.13 mmol) via syringe. The resulting mixture was stirred at room temperature for 4 h, then was
partitioned between saturated aqueous sodium bicarbonate and EtOAc (4X). The combined
organic extracts were dried over anhydrous sodium sulfate and concentrated to afford 10 (13 mg,
96% yield) as a colorless oil. 1H NMR (400 MHz, CDCl3) δ 11.49 (br. s., 1H), 7.60 (d, J=7.9 Hz,
2H), 7.33 (t, J=7.5 Hz, 2H), 7.09 (t, J=7.2 Hz, 1H), 6.36 (d, J=2.0 Hz, 1H), 5.51 (s, 1H), 3.26 (s,
2H), 2.36 (s, 6H). LCMS-ESI (POS.) M/Z: 205.2 (M+H)+.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2-(Morpholinomethyl)-N-phenylacrylamide (11). Compound 11 was made according to the
procedure for 10 using morpholine in place of dimethylamine. 1H NMR (400 MHz, DMSO-d6) δ
10.98 (s, 1H), 7.64 (d, J=7.9 Hz, 2H), 7.34 (t, J=7.3 Hz, 2H), 7.08 (tt, J=7.41, 1.2 Hz, 1H), 6.06
(d, J=2.0 Hz, 1H), 5.58-5.62 (m, 1H), 3.65 (t, J=4.7 Hz, 4H), 3.31-3.32 (s, 2H), 2.44-2.49 (m,
4H). LCMS-ESI (POS.) M/Z: 247.3 (M+H)+.
Amgen Proprietary - Confidential
ACS Paragon Plus Environment
29