Thiol Reactivity of N-Aryl α-Methylene-γ-lactams: A Reactive Group for Targeted Covalent Inhibitor Design

Article abstract of DOI:10.1021/acs.joc.1c01335

Kinase activity can be modulated reversibly or irreversibly by the reaction of targeted covalent inhibitors with nucleophilic residues in protein active sites. Herein, we present thiol reactivity studies that support α-methylene-γ-lactams as tunable surro

Full text of DOI:10.1021/acs.joc.1c01335

pubs.acs.org/joc
Article
Thiol Reactivity of N‑Aryl α‑Methylene-γ-lactams: A Reactive Group
for Targeted Covalent Inhibitor Design
̆
Tugcȩ G. Erbay, Daniel P. Dempe, Bhaskar Godugu, Peng Liu,* and Kay M. Brummond*
Cite This: https://doi.org/10.1021/acs.joc.1c01335
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Kinase activity can be modulated reversibly or
irreversibly by the reaction of targeted covalent inhibitors with
nucleophilic residues in protein active sites. Herein, we present
thiol reactivity studies that support α-methylene-γ-lactams as
tunable surrogates for the highly reactive α-methylene-γ-lactones.
The reactivity of the α-methylene is modulated via the N
substituent, and the reaction rates toward glutathione were
determined via mass spectrometry. Density functional theory
calculations of transition states of thiol additions to α-methylene-γ-
lactams revealed that the use of the M06-2X functional with the
SMD solvation model and methyl thiolate as a model nucleophile
reliably predicts the relative reactivities of the α-methylene-γ-
lactams, and quasiharmonic approximations improve the agreement between experiment and computation.
scaffold, combined with need for new reactive groups in TCI
INTRODUCTION
■
design, inspired us to ask if α-methylene-γ-lactams can be used
in a more general way as a cysteine-targeting electrophile.12−14
Herein, we present a deeper understanding of how the
structure and electronics impact the electrophilic reactivity of
an α-methylene-γ-lactam as the next step toward the
development of this new covalent reactive group (Figure 1C).
We hypothesized that the electrophilic reactivity of the α-
methylene-γ-lactam can be predictably modulated by extending
the conjugation to substitution on the lactam nitrogen. To test
our hypothesis, we prepared a series of electronically different
α-methylene-γ-lactams and determined their reactivity toward
glutathione (GSH) using mass spectrometry. We performed
density functional theory (DFT) calculations to establish a
computational methodology for accurate prediction of the thiol
reactivity of N-aryl α-methylene-γ-lactams. We systematically
evaluated the predicted reactivity trends from different
computational methods, including different functionals
(M06-2X and B3LYP), solvation models (SMD and PCM),
and using different nucleophiles (methyl thiolate, cysteamine
thiolate, and GSH). Comparison of the computed activation
free energies and the experimental rate constants revealed a
Targeted covalent inhibitors (TCIs) offer a powerful way to
selectively modulate kinase activity by zeroing in on non-
catalytic, nucleophilic amino acids near the ATP binding
site.1−3 For example, osimertinib is a compound with a thiol-
reactive N-aryl acrylamide attached to a scaffold having
specificity toward the binding site of EGFR tyrosine kinase
(Figure 1A).4 However, a developed resistance to osimertinib
stemming from an EGFR mutation where the targeted cysteine
residue (Cys797) is replaced with a less nucleophilic serine
prevents covalent bond formation and reduces drug
efficacy.5−7 To overcome resistance that occurs due to point
mutations and other TCI limitations, new reactive groups in
combination with a deeper understanding of how the structure
and electronics impact electrophilic reactivity are needed to
design more effective covalent ligands.8−10
To date, α,β-unsaturated carbonyls are the most commonly
used covalent reactive groups with N-aryl acrylamides having
the most success, in part due to the readily tuned electrophilic
reactivity toward thiols.3 Recently, one of our groups applied
this tunable design element to the preparation of guaianolide
analogues where the reactive α-methylene-γ-lactone, one of
Nature’s most common warheads, was replaced with an α-
methylene-γ-lactam (Figure 1B).11 Substitution of the lactam
nitrogen with electronically different groups afforded a series of
compounds that reacted with cysteamine via a thio-Michael
addition with half-lives ranging from days to seconds.
Additionally, for one analogue having intermediate thiol
reactivity, ALARM NMR and ALARM MSPS data support
discriminate reactivity with proteins containing cysteinyl
residues. These encouraging results involving the guaianolide
Received: June 7, 2021
https://doi.org/10.1021/acs.joc.1c01335
J. Org. Chem. XXXX, XXX, XXX−XXX
© XXXX American Chemical Society
A

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
affording boc-protected α-methylene-γ-lactam 2 (37% yield
over two steps). This lactam was reacted with trifluoroacetic
acid (TFA) in methylene chloride (CH2Cl2) to afford 3-
methylene-2-pyrrolidinone (3) in 45% yield (Scheme 1).
N-Arylations of lactam 3 with para-, meta-, and ortho-
substituted aryl iodides were achieved using a modified
procedure previously reported by Buchwald,17 and the results
are depicted in Table 1. The reaction of lactam 3 with CuI
Table 1. Coupling of Aryl Iodides with 3-Methylene-2-
pyrrolidinone (3) Affording N-Aryl α-Methylene-γ-lactams
5a−m
a
entry
R
equiv of ArI
time (h)
yield (%)
1
2
3
4
5
6
7
8
H (5a)
3
3
1.6
3
3
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
24
25
18
24
23
24
21.5
24
24
81
79
69
26
65
68
72
57
60
31
60
81
80
p-CF3 (5b)
p-CN (5c)
p-OMe (5d)
p-F (5e)
p-NO2 (5f)
p-NMe2 (5g)
m-OMe (5h)
m-F (5i)
Figure 1. (A) Osimertinib and the interaction with EGFR-T790M/
V948R kinase.15 (B) Lactam-containing guaianolide analogues where
R1 regulates electrophilic reactivity. (C) Summary of experimental
and computational contents presented in this work.
9
10
11
12
13
m-NO2 (5j)
o-OMe (5k)
o-F (5l)
18.5
23
25
o-NO2 (5m)
72
level of theory that gives reliable predictions of relative
reactivities of the N-aryl α-methylene-γ-lactam.
a
Isolated yields.
RESULTS AND DISCUSSION
■
(0.15 equiv), N,N′-dimethyl ethylenediamine (0.3 equiv),
iodobenzene 4a (3 equiv), and tripotassium phosphate (2
equiv) in toluene at 80 °C afforded N-arylation product 5a in
81% yield (Table 1, entry 1). To facilitate the reaction of these
heterogeneous mixtures, an oversized stir bar was used.
Subsequent experiments with different aryl iodides revealed
that using less aryl iodide (1.5 equiv) resulted in an easier
purification of the N-arylated product.
Synthesis of α-Methylene-γ-lactams. Thirteen elec-
tronically different α-methylene-γ-lactams were prepared by a
late-stage N-arylation of lactam 3, which is synthesized in four
steps from commercially available 2-pyrrolidinone (Scheme
1).16 Accessible aryl substituents representing a broad range of
Scheme 1. Synthesis of 3-Methylene-2-pyrrolidinone (3)
Next, the N-arylation reaction of lactam 3 with aryl iodides
4b−m having electron-donating or -withdrawing groups was
investigated. Reaction of aryl iodides substituted at the para
position with groups that are electron-withdrawing gave N-
arylation products with yields ranging from 65% to 79% (Table
1, entries 2, 3, 5, and 6). Reaction of lactam 3 with aryl iodides
having electron-donating groups, 1-iodo-4-methoxybenzene
(4d) and 1-iodo-4-(N,N-dimethylamino)benzene (4g), af-
forded 5d and 5g in 26% and 72% yields, respectively
(Table 1, entries 4 and 7, respectively). The low yield for 5d
was due to purification issues stemming from excess aryl iodide
co-eluting with the product. Reaction of lactam 3 with aryl
iodides substituted at the meta position afforded products 5h−
j in 31−60% yields (entries 8−10). Due to the light sensitivity
of 1-fluoro-3-iodobenzene (4i), the reaction flask was wrapped
with aluminum foil, affording N-arylation product 5i in 60%
yield after 24 h (Table 1, entry 9). Reaction of lactam 3 with
aryl iodides substituted at the ortho position afforded products
5k−m in 60−81% yields (entries 11−13, respectively). The N-
arylation reaction with 1-fluoro-2-iodobenzene (4l) under the
standard conditions gave 5l in 81% yield (Table 1, entry 12).
The reaction of 1-iodo-2-nitrobenzene (4m) was slow due to
Hammett values with minimal steric impact were chosen, along
with those groups providing a direct comparison of structural
and electronic properties with acyclic N-aryl acrylamides.12
The nitrogen of 2-pyrrolidinone was boc-protected using
boc-anhydride and catalytic 4-(N,N-dimethylamino)pyridine
(DMAP) to form 1 in 83% yield. Installation of the α-
methylene group was accomplished via a two-step process.
First, boc-protected 2-pyrrolidinone 1 was deprotonated with
lithium bis(trimethylsilyl)amide (LHMDS) (1.3 equiv) and
alkylated with N,N-dimethylmethylene iminium iodide,
evidenced by crude 1H NMR (see the Supporting Information,
compound S1). The quaternary ammonium salt was obtained
by adding allyl bromide in ethanol, which was subjected to
sodium carbonate (Na2CO3) to effect an elimination reaction
B
https://doi.org/10.1021/acs.joc.1c01335
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
Table 2. Measured Rates of GSH Addition, Derived Half-Lives, and Spectroscopic Data
b
Hammett parameter
t1/2
(min)
kpseudo1st
log kGSH
Hβ1
δ
Hβ2
δ
Cβ δ
(ppm)
CO λ−1
−
entry
compound
σp or σm
(×10−3 min−1
)
(M−1 s−1
)
(ppm)
(ppm)
(cm−1
)
a
1
2
3
4
5
6
7
8
NPA
−
0.00
0.65
1.00
−0.26
−0.03
1.27
−0.12
0.12
0.34
0.71
−0.26
−0.03
1.27
119
1386
525
5.83
0.50
1.32
3.35
0.24
0.37
9.52
−
0.32
0.83
1.54
0.17
0.39
2.08
−2.01
−3.08
−2.66
−2.25
−3.40
−3.21
−1.80
−
−3.28
−2.86
−2.59
−3.55
−3.19
−2.46
−
−
−
−
1673
1685
1691
1681
1681
1692
1668
1676
1679
1682
1685
1685
1693
H (5a)
5.46
5.53
5.54
5.42
5.47
5.57
5.39
5.46
5.50
5.55
5.41
5.49
5.51
5.90
5.96
5.97
5.86
5.90
6.00
5.83
5.89
5.93
5.98
5.80
5.88
5.86
115.8
117.0
117.4
115.3
116.0
117.7
114.7
116.0
116.6
117.1
115.2
116.1
117.1
p-CF3 (5b)
p-CN (5c)
p-OMe (5d)
p-F (5e)
p-NO2 (5f)
p-NMe2 (5g)
m-OMe (5h)
m-F (5i)
207
2888
1873
73
−
2166
835
450
4077
1777
333
9
10
11
12
13
14
m-NO2 (5j)
o-OMe (5k)
o-F (5l)
o-NO2 (5m)
NPA, N-phenylacrylamide. For ortho- and para-substituted compounds, σp is listed; for meta-substituted compounds, σm is listed.20
a
b
−
steric effects; however, N-arylation product 5m was obtained
in 80% yield after 72 h (Table 1, entry 13).
log was plotted against time. The pseudo-first-order rate
constant was obtained as the slope of the line of best fit
following eq 1, and the half-lives were determined with eq 2.
Experimental Determination of Reaction Rates for
the Conjugate Addition of GSH to α-Methylene-γ-
lactams 5a−m.18 Reactions were performed with a large
excess of GSH to ensure pseudo-first-order kinetics and rate
constants, and reaction half-lives were determined in a manner
analogous to that reported by Cee et al. with minor
modifications.12,13 Lactams 5a−m were reacted with 10 mM
GSH in 67 mM potassium phosphate buffer (pH 7.4) with 1−
1.5% DMSO at 37 °C. These conditions were selected to
mimic intracellular GSH concentration and to benchmark
reactivities with those previously reported for N-aryl
acrylamides.12,14,19 Lactam stability was monitored with
control batches. A Dionex Ultimate 3000RS liquid chromato-
graph and a Thermo Scientific Q-Exactive mass spectrometer
were used. Compounds were separated using a Phenomenex
Luna Omega 1.6 μm Polar C18 100 Å 200 mm × 2.1 mm
column, which was maintained at 37 °C. Compounds were
separated with a linear gradient beginning with 5% acetonitrile
and 0.1% formic acid (v/v) in water and ending with 95%
acetonitrile and 0.1% formic acid in water with a run time of 5
min. Aliquots (2 μL) were injected every 5 min over the span
of 9.3 h, alternating between the batch with GSH and the
control batch. Peak area ratios were determined by dividing the
total ion count (TIC) peak area of the parent compound by
that of the internal standard, phenacetin. For each batch, a
calibration curve was obtained by measuring the peak area
ratio of seven standard solutions of compound 5a ranging in
concentration from 0.2 to 2.0 μM. In all instances, the
calibration curves displayed parity between response and
concentration within this range. A linear concentration−
response relationship was assumed for the internal standard
and the lactams. The relative concentrations for each lactam
were calculated by fitting the measured peak area ratio to the
calibration curve.
ln([analyte]t ) = −kpseudo1stt + ln([analyte]o)
(1)
ln 2
kpseudo1st
t1/2
=
(2)
Each lactam was analyzed in duplicate or triplicate on
different days, and the mean half-lives from multiple runs are
reported in Table 2.12 The data showed low variability over
multiple analyses. The control batches gave linear plots with a
slope of 0, indicating lactam stability and a GSH-dependent
response. The mean half-life of N-phenylacrylamide (NPA)
(119 min, entry 1) was within the range of half-lives for the
same compound reported previously.12,14,19,21
The measured pseudo-first-order rate constants for the
reactions of lactams 5a−f and 5h−m with GSH are listed in
Table 2. Relative to the electronically neutral 5a (t1/2 = 1386
min), the electron-donating methoxy substituent on the para,
meta, and ortho positions of the aryl slowed the rate of reaction
(2888, 2166, and 4077 min, respectively). The fluoro
substituent slowed the reaction when positioned para or
ortho (1873 or 1777 min, respectively). With its π-donating
ability diminished in the meta position, the inductive nature of
the fluoro group resulted in a shorter half-life (835 min).
Electron-withdrawing substituents, such as p-, m-, and o-NO2
and p-CF3 and p-CN groups, accelerated the reaction.
Measurement of the half-life for the reaction of p-
dimethylamino-substituted lactam 5g was unsuccessful. We
attribute this to the insolubility of a dimethylammonium salt of
the lactam formed under the acidic chromatography
conditions, evidenced by a substantial increase in the column
pressure. In most cases, a larger difference between measured
k
pseudo1st values was observed between duplicate measurements
of lactams with longer half-lives. As illustrated in Figure 2, for
each compound, the rate of Michael addition reaction with
GSH is closely associated with its Hammett parameter. The
From these relative concentrations, the percent compound
remaining was determined relative to time zero and the natural
C
https://doi.org/10.1021/acs.joc.1c01335
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
Figure 2. Relationship of Hammett parameters σp− or σm with the rate
of GSH addition, for N-aryl α-methylene-γ-lactams with (a) para and
ortho and (b) meta substitutions on the aryl ring. The reactivities of
lactams with para and ortho substitutions on the aryl ring correlate
well with σp , and the lactams with meta substitutions on the aryl ring
correlate well with σm. o-NO2 is excluded from the linear fitting line.
1
Figure 3. Correlation of log kGSH with (a) 13C NMR and (b) H
NMR shifts.
electrophilicity indices, HOMO−LUMO energies, and atomic
charges have been used toward the development of
quantitative structure−activity relationship (QSAR) models
for the reactivity prediction of TCI fragments.27−29 DFT
methods have also been used to calculate thio-Michael
addition transition states and predict the relative reactivity
trends via transition state theory (TST). These calculations
were often performed to evaluate kinetic barriers and
thermodynamic preferences of additions of a model
nucleophile (usually methyl thiolate ion) to Michael acceptors,
the rate-determining step of the thio-Michael addi-
tions.12−14,23−25 Good correlations between DFT-calculated
activation energies (Ea) and experimentally determined rate
constants (k, derived from eq 3) were obtained.30 However,
quantitative predictions of experimental reactivity differences
between different TCI fragments are still challenging, and a
systematic assessment of different DFT methods has not been
performed. This has limited the applications of DFT methods
to quantitatively predict kinetic profiles of new TCIs. In
addition, such DFT models have yet to be applied for the
lactam reactivity predictions.
−
influence of resonance on reactivity is demonstrated by the
−
better correlation with σp than σp for para- and ortho-
substituted N-aryl α-methylene-γ-lactams [R2 = 0.98 (Figure
2a) and R2 = 0.93 (Figure S10)] and with σm for meta-
substituted N-aryl α-methylene-γ-lactams [R2 = 0.81 (Figure
2b)]. The reactivities of the ortho-substituted compounds can
−
be correlated with σp , except o-NO2. We attribute this to the
steric effects on the conformation of the lactam, evidenced by
1
peak broadening in the H NMR spectrum, indicating the
2
hindered rotation about the N−C(sp ) bond (see the
−
Supporting Information). The positive ρ values for σp and
σm (1.01 and 0.86, respectively) agree with the computed
mechanistic pathway of the addition of thiolate to the N-aryl α-
methylene-γ-lactams (vide infra). The ρ value of 1.01 for para-
and ortho-substituted lactams demonstrates that they are
sensitive to the substituent effects, whereas the smaller ρ value
for meta-substituted lactams (0.86) suggests that these
compounds are slightly less sensitive to the substituent effects.
The greater ρ value for para- and ortho-substituted lactams is
consistent with the sensitivity of the Michael addition to the
resonance effects.
−ΔGE⧧xperiment/RT
kBT
h
k =
e
(3)
Cee et al. studied the addition of methyl thiolate to a series
of N-aryl acrylamides at the B3LYP level of theory using the
IEFPCM solvation model and demonstrated a linear
correlation between the calculated ΔG⧧ and the N-aryl
acrylamide reactivity toward glutathione (log kGSH) with an
R2 of 0.75 and a slope of −0.25.12 Similarly, Flanagan et al.
calculated the ΔG⧧ at the B3LYP level of theory with the SMD
solvation model and found a correlation with the reaction half-
lives of various acrylamides with glutathione with an R2 of
0.91.14 These studies demonstrated the utility of the
application of TST via DFT to predict the relative reactivity
of TCI fragments. However, accurate calculation of ΔG⧧ using
DFT methods remains a challenging task. The performance of
the DFT calculations is affected by multiple factors, such as the
density functional,31 solvation model, and the choice of the
model system (i.e., the active nucleophile). In this study, we
aim to evaluate the performance of various DFT methods by
comparing the computed reactivity difference between N-aryl
α-methylene-γ-lactams and experimental rate constants to
identify the most robust computational method for predicting
reactivity trends.
Because the reaction rate is controlled by the substituent’s
effect on the electronic nature of the molecule, it was predicted
that simple spectroscopic characterization methods that probe
the effective electronic environment around key atoms or
bonds could be utilized to establish trends with the rate of
reaction.22 Indeed 1H NMR, 13C NMR, and infrared
spectroscopy (IR) showed varying degrees of parity with the
measured rate constants. The highest degrees of correlation
were observed for the 13C NMR shifts of the Cβ atom [R2 =
1
0.92 (Figure 3a)] and the H NMR shifts of the methylenyl
proton Hβ1 [R2 = 0.85 (Figure 3b)]. A weaker correlation was
observed for the 1H NMR shift of Hβ2 and the 13C NMR of the
carbonyl carbon [R2 = 0.63 and 0.62, respectively (see Figure
S11)]. As predicted, electron-donating substituents increased
the sp3 character of the CO bond, as shown by lower IR
frequencies. However, these data displayed the weakest
correlation with reaction rate [R2 = 0.43 (see Figure S11)].
DFT Studies. DFT calculations have been used to facilitate
the TCI fragment design of various acrylamides, α,β-
unsaturated ketones, aldehydes, and esters.12−14,23−25 Addi-
tionally, DFT calculations provided insights into the structural
and electronic factors that influence the reactivity by the
assessment of the ground state properties of Michael
acceptors.26 DFT-calculated physical descriptors such as
We first compared the optimized transition state geometries
and activation energies of the reaction of α-methylene-γ-lactam
5a (Figure 4). Regardless of the computational method used,
the thio-Michael addition transition state always favors thiolate
D
https://doi.org/10.1021/acs.joc.1c01335
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
Next, we investigated whether using a more realistic model
nucleophile, such as the cysteamine thiolate or the
deprotonated glutathione (hereafter termed glutathione),
would affect the predicted transition state geometries and the
agreement between computed and experimental ΔG⧧ values.
We surmised that these more realistic models would be able to
identify and describe the potential H-bonding interactions
between the NH2 or amide groups on the nucleophile and the
N-aryl α-methylene-γ-lactam and steric effects of the
glutathione. If these factors are prominent in determining the
transition state energies, the use of these thiolates should afford
better agreement with experiment than the use of methyl
thiolate. Similar to the methyl thiolate addition, the syn
additions of the cysteamine thiolate and the glutathione are
favored (TS2s and TS3s in panels b and c, respectively, of
Figure 4). We did not observe strong H-bonding interactions
between the NH2 on the cysteamine thiolate and 5a in TS2s.
On the contrary, H-bonding interactions with one of the amide
groups on glutathione are present in both syn and anti
transition states (TS3s and TS3a, respectively). Therefore, the
potential H-bonding interactions with these more realistic
nucleophiles do not alter the preference for syn transition
states. Distortion−interaction analysis of the syn and anti
transition states with glutathione demonstrated that the syn-TS
(TS3s) has a stronger ΔEint than the anti-TS (TS3a) [−9.3
and −6.1 kcal/mol for TS3s and TS3a, respectively (see Figure
S13)]. These results suggest the dominant factor favoring the
syn-TS is the diminished closed-shell repulsion between an S
lone pair on the nucleophile and the π electrons on the CC
group of the Michael acceptor.24 Compared to the reaction
with methyl thiolate, using cysteamine thiolate as the model
nucleophile only slightly increased the predicted ΔG⧧ values.
On the contrary, the computed activation free energy with
glutathione (ΔG⧧ = 18.2 kcal/mol) is 3.5 kcal/mol higher than
that with methyl thiolate. The ΔG⧧DFT of TS3s (18.2 kcal/mol)
is closer to the ΔGE⧧xperiment for this substrate (22.9 kcal/mol).32
We attribute this to the greater distortion of both the lactam
and the nucleophile in TS3s than in TS1s (see Figure S13),
and the increase in steric repulsion about the forming Cβ···S
bond due to a later transition state with GSH (the Cβ···S
distances in TS3s and TS1s are 2.37 and 2.47 Å, respectively).
Next, we calculated the barriers to the Michael addition of
methyl thiolate to 12 different N-aryl α-methylene-γ-lactams
using different computational methods, including different
density functionals (M06-2X and B3LYP) and solvation
models (SMD and PCM). These functionals were selected
because of the use of B3LYP in previous studies and the broad
applications of the hybrid-meta-GGA functional M06-2X in
recent computational organic chemistry literature. Further-
more, we calculated thermal corrections to Gibbs free energies
using both the harmonic oscillator model and Truhlar’s
quasiharmonic approximation33 to account for the challenges
in computing contributions to entropies from low vibrational
frequencies. Because the thiolate identity did not affect the
preferred Michael addition mechanism (vide supra), the methyl
thiolate was used in these calculations to reduce the
computational cost. The use of M06-2X with the SMD
solvation model and quasiharmonic approximation provided
the best correlation with the experimental reactivities of N-aryl
α-methylene-γ-lactams (Figure 5a). At this level of theory, the
correlation of the calculated and experimental reactivities
provides one of the highest fitting of data (R2 = 0.95) and the
slope of the linear fitting is closest to unity (0.83). A similar
Figure 4. Transition states of the thio-Michael additions to N-phenyl
lactam 5a. Transition states and activation free energies (ΔG⧧) were
calculated at the M06-2X/6-311+G(d,p)(SMD)//M06-2X/6-31+G-
(d)(SMD) level of theory with (a) methyl thiolate, (b) cysteamine
thiolate, and (c) glutathione (−2 charged with both a thiolate and a
carboxylate) as the nucleophile and (d) methyl thiolate at the M06-
2X/6-311+G(d,p)(PCM)//M06-2X/6-31+G(d)(PCM) level of
theory (left) and at the M06-2X/6-311+G(d,p)//M06-2X/6-
31+G(d) level of theory in the gas phase (right). All geometry
optimizations and single-point energy calculations were performed in
water, except for TS1s_gas. All energies are with respect to the
corresponding thiolate and 5a.
approach with a syn orientation placing the methyl group on
the thiolate syn-periplanar with the CC double bond of the
Michael acceptor. For example, the syn TS [TS1s (Figure 4a)]
is preferred over the anti TS (TS1a) by 2.5 kcal/mol at the
M06-2X/6-311+G(d,p)(SMD)//M06-2X/6-31+G(d)(SMD)
level of theory when methyl thiolate was used as the
nucleophile.
This preference is consistent with the previous computa-
tional study by Houk and Krenske of the addition of methyl
thiolate to α,β-unsaturated ketones.24 The solvation model
greatly impacts the transition state geometries and the
activation energy barriers. Calculations in the gas phase not
only significantly underestimate the activation free energy
(ΔG⧧DFT = −1.7 kcal/mol, and ΔGE⧧xperiment = 22.9 kcal/mol for
5a) but also predict a much earlier transition state as evidenced
by the significantly longer Cβ···S distance than those optimized
in solvent (Figure 4d, TS1s_gas). These results indicate that a
suitable solvation model should be used in both geometry
optimization and single-point energy calculations.
E
https://doi.org/10.1021/acs.joc.1c01335
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
Figure 5. Correlation of the DFT-calculated reactivity to the experimentally determined reactivity. Experimental ΔG⧧ values were derived from the
Eyring equation. The free energies of activation of the Michael additions of methyl thiolate to the N-aryl α-methylene-γ-lactams were calculated at
the (a) M06-2X/6-311+G(d,p)(SMD)//M06-2X/6-31+G(d)(SMD), (b) M06-2X/6-311+G(d,p)(PCM)//M06-2X/6-31+G(d)(PCM), (c)
B3LYP/6-311+G(d,p)(SMD)//B3LYP/6-31+G(d)(SMD), and (d) B3LYP/6-311+G(d,p)(PCM)//B3LYP/6-31+G(d)(PCM) levels of theory.
All optimizations and single-point calculations were performed in water. HOA, harmonic oscillator approximation; QHA, quasiharmonic
approximation.
correlation between ΔH⧧DFT and ΔGE⧧xperiment was observed [R2
= 0.96 (see Figure S14)], which indicates entropic effects do
not affect the relative reactivity trends and thus the computed
activation free energies from quasiharmonic approximation and
activation enthalpies may both be used to predict the relative
reactivity trend of different α-methylene-γ-lactam TCIs. On
the contrary, the use of the PCM solvation model in place of
SMD gave a linear fitting line with a much smaller slope (0.43)
and a moderate correlation of data [R2 = 0.84 (Figure 5b)].
We believe this is due to the differences in the TS geometries
optimized with PCM. The PCM-optimized transition states
have 0.16 Å shorter Cβ···S distances on average than those
optimized with the SMD solvation model.
Interestingly, when the PCM solvation model was employed,
the use of quasiharmonic approximation did not improve the
correlation between ΔG⧧DFT and ΔGE⧧xperiment regardless of the
choice of functional (Figure 5b,d). The previously used hybrid-
GGA functional B3LYP with the SMD solvation model
provided a high R2 value of 0.96, but the slope was significantly
Figure 6. Correlation of the DFT-calculated reactivity with the
experimentally determined reactivity. The free energies of activation
of the Michael additions of cysteamine thiolate to the N-aryl α-
methylene-γ-lactams were calculated at the M06-2X/6-311+G(d,p)-
(SMD)//M06-2X/6-31+G(d)(SMD) level of theory. All optimiza-
lower than unity [0.45 (Figure 5c)]. These results indicate that
B3LYP is capable of qualitatively predicting the reactivity trend
but would underestimate the reactivity difference between two
TCIs.
Next, we explored whether the use of deprotonated
tions and single-point calculations were performed in water.
cysteamine in place of methyl thiolate as the nucleophile in
the transition state calculations would affect the predicted
reactivity trend of different Michael acceptors (Figure 6). We
calculated the activation free energies of Michael addition of
the cysteamine thiolate to the 12 different N-aryl α-methylene-
γ-lactams at the M06-2X/6-311+G(d,p)(SMD)//M06-2X/6-
31+G(d)(SMD) level of theory. Experimentally determined
reactivities and DFT-calculated activation barriers showed
good correlations for predicted reactivity for both the
cysteamine thiolate and the methyl thiolate when using the
quasiharmonic approximation (R2 = 0.96 and 0.95, respec-
tively). The slope of the fitted line was slightly smaller with
cysteamine thiolate than with methyl thiolate (0.75 and 0.83,
respectively). Overall, a larger model thiolate having an NH2
group afforded activation barriers in closer alignment with
experiment (∼ 1 kcal/mol higher in energy than methyl
thiolate) but did not improve the accuracy of the predicted
relative reactivities of N-aryl α-methylene-γ-lactams.
The computational results presented above identified the
M06-2X/6-311+G(d,p)(SMD)//M06-2X/6-31+G(d)(SMD)-
QHA level of theory and the use of methyl thiolate in
transition state energy calculations as an effective way to
predict reactivities of N-aryl α-methylene-γ-lactams. We tested
whether this computational approach can be applied to other
types of Michael acceptors, such as N-aryl acrylamides. We
computed the Gibbs free energies of activation of the addition
of methyl thiolate to various N-aryl acrylamides and compared
these predicted values with the activation free energies
reported in a previous experimental study by Cee et al. (see
Figure S15).12 The calculations gave a high fitting of data (R2 =
0.87) and a slope of the linear fitting close to unity (0.89).
F
https://doi.org/10.1021/acs.joc.1c01335
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
removed, and the reaction mixture was allowed to warm to rt. After
2.5 h at rt, TLC showed the complete disappearance of 2-
pyrrolidinone. The reaction mixture was transferred to a 250 mL
separatory funnel and diluted with ethyl acetate (13 mL). HCl (1 M,
25 mL) was added, and the mixture was shaken; then the layers were
separated. The aqueous layer was extracted with ethyl acetate (3 × 13
mL). The combined organic layers were dried over MgSO4, filtered,
and concentrated by rotary evaporation. The crude residue was
purified by silica gel flash column chromatography (2 cm × 20 cm
SiO2 column, 75% ethyl acetate/hexanes) to furnish the title
compound 1 in 83% yield (1.8 g). Characterization data corresponds
to the literature.34 1H NMR (500 MHz, CDCl3): δ 3.74−3.66 (m,
2H), 2.48−2.42 (m, 2H), 2.04−1.90 (m, 2H), 1.48−1.44 (m, 9H).
TLC: Rf = 0.31 (50% ethyl acetate/hexanes) (silica gel, UV, KMnO4).
N-boc-α-Methylene-γ-lactam (2). The synthesis of 2 was
performed in a manner similar to that previously reported by Craven
et al.34 A 100 mL, two-neck, round-bottom flask equipped with a 1 cm
Teflon-coated magnetic stir bar and a rubber septum was charged
with N-boc-lactam 1 (1.8 g, 9.7 mmol, 1 equiv). THF (28 mL) was
added via syringe, and the flask was placed in a dry ice/acetone bath
(−78 °C). Lithium bis(trimethylsilyl)amide (LHMDS) (1.0 M in
THF, 12.8 mL, 12.6 mmol, 1.3 equiv) was added via syringe dropwise
over 20 min, and the solution was maintained at −78 °C. After 1 h,
dimethylmethylideneammonium iodide (2.7 g, 14.6 mmol, 1.5 equiv)
was added and the reaction mixture was maintained at −78 °C for 2.5
h. The reaction mixture was warmed to 0 °C by replacing the dry ice/
acetone bath with an ice bath. After 30 min, saturated aqueous NH4Cl
(75 mL) and H2O (20 mL) were added to the reaction flask and the
mixture was transferred to a separatory funnel and diluted with ethyl
acetate (50 mL). The aqueous layer was separated and extracted with
ethyl acetate (3 × 50 mL). The combined organic layers were dried
over MgSO4, filtered, and concentrated using rotary evaporation to
give the dimethyl(methylamine) intermediate as evidenced by 1H
NMR analysis of the residue (2.18 g).
These results suggest that this level of theory may be applied to
reactivity prediction for a broad range of substrates.
CONCLUSIONS
■
In this work, we present the α-methylene-γ-lactam as a novel
covalent reactive group that can be incorporated as a TCI
fragment to react with cysteine residues. Experimental studies
of our N-aryl α-methylene-γ-lactams showed a 10-fold
reactivity attenuation with GSH compared to the N-aryl
acrylamides. Characterization of the thiol reactivity of this
lactam warhead will be valuable in the rational design of
covalent inhibitors for several enzyme classes. The electro-
philicity of the α-methylene group can be tuned by modifying
the nitrogen substituent as demonstrated by GSH reactivity
studies. The reactivity of the N-aryl α-methylene-γ-lactams
toward GSH shows good correlation with the 13C NMR shifts
of the Cβ resonance of the methylene (R2 = 0.92) and with the
−
Hammett parameters σp for the para and ortho aryl
substitutions (R2 = 0.98) and σm for the meta aryl substitution
(R2 = 0.81). The positive ρ values for σp and σm support the
−
anionic mechanism of addition of thiolate to the N-aryl α-
methylene-γ-lactams. A computational protocol was estab-
lished to predict relative reactivities of the N-aryl α-methylene-
γ-lactams. Systematic studies with different density functionals
(B3LYP and M06-2X), different solvation models (PCM and
SMD), and different model nucleophiles (methyl thiolate,
cysteamine thiolate, and glutathione) demonstrated that the
use of the M06-2X functional with the SMD solvation model
and the methyl thiolate nucleophile serves this purpose
efficiently. The application of quasiharmonic approximations
improves the linear correlation between the experiment and
computation.
The crude residue (2.18 g, 9.0 mmol, 1 equiv) was transferred to a
100 mL, two-neck, round-bottom flask equipped with a 1 cm Teflon-
coated stir bar and two septa. Ethanol (40 mL) and 3-bromo-1-
propene (6.2 mL, 72.0 mmol, 8 equiv) were added sequentially via
syringe. Sodium carbonate (5.72 g, 54.0 mmol, 6 equiv) was added in
a single portion. The mixture was maintained at rt for 18 h, at which
time TLC showed complete disappearance of the starting material.
EXPERIMENTAL PROCEDURES
■
General Information. Unless otherwise stated, all reactions were
performed in flame-dried glassware under an atmosphere of dry
nitrogen. All commercially available starting materials were used as
received, without further purification. Reaction solvents tetrahydro-
furan (THF) and dichloromethane (DCM) were purified via pressure
filtration through an alumina column. Toluene was distilled from
CaH2 before use. Column chromatography was performed using 40−
63 μm, 60 Å pore size silica gel. TLC was performed on Silicycle glass
The mixture was poured into a Buchner funnel (6 cm), and the solid
̈
was removed via vacuum filtration. The filtrate was transferred to a
separatory funnel, and saturated aqueous NH4OH (20 mL), H2O (20
mL), and ethyl acetate (30 mL) were added. The mixture was shaken,
and the aqueous layer was separated and extracted with ethyl acetate
(3 × 20 mL). The combined organic layers were dried over MgSO4,
filtered, and concentrated by rotary evaporation to give a residue
(1.22 g). The residue was purified by silica gel flash column
chromatography (2 cm × 20 cm SiO2 column, 50% ethyl acetate/
hexanes) to furnish the title compound 2 in 37% yield (0.71 g) over
two steps. Characterization data corresponds to the literature.34 1H
NMR (500 MHz, CDCl3): δ 6.09 (t, J = 3.0 Hz, 1H), 5.41 (t, J = 2.5
Hz, 1H), 3.68 (t, J = 7.5 Hz, 2H), 2.77−2.65 (m, 2H), 1.48 (s, 9H).
TLC: Rf = 0.43 (50% ethyl acetate/hexanes) (silica gel, KMnO4).
α-Methylene-γ-lactam (3). The synthesis of 3 was performed in
a manner similar to that previously reported by Craven et al.34 A 50
mL, two-neck, round-bottom flask equipped with a 1 cm Teflon-
coated magnetic stir bar and a rubber septum was charged with 2
(0.65 g, 3.28 mmol, 1 equiv). Dichloromethane (25 mL) and
trifluoroacetic acid (1.26 mL, 16.4 mol, 5 equiv) were added
successively via syringe, and the reaction solution was maintained at rt
for 2.5 h, at which point TLC showed the complete disappearance of
the starting material. Saturated aqueous K2CO3 was added dropwise
until the pH reached 10 (pH test strips), and the solution was diluted
with H2O (10 mL). The mixture was transferred to a separatory
funnel, and the aqueous layer was separated and extracted with
dichloromethane (3 × 20 mL). The combined organic layers were
dried over MgSO4, filtered, and concentrated by rotary evaporation.
The crude residue was purified by silica gel flash column
backed 60 Å plates containing the F254 indicator. H NMR and 13C
1
NMR spectra were recorded using a Bruker Avance 500 MHz
spectrometer. Spectra were referenced to chloroform (1H, 7.26 ppm;
13C, 77.16 ppm) and DMSO (1H, 2.50 ppm; 13C, 39.51 ppm).
Chemical shifts are reported in parts per million, and multiplicities are
indicated by singlet (s), doublet (d), triplet (t), quartet (q), doublet
of doublets (dd), doublet of triplets (dt), triplet of doublets (td), and
multiplet (m). Coupling constants are reported in hertz. All NMR
spectra were recorded at room temperature. Mass spectrometry was
performed with a Thermo Scientific Q-Exactive Oribtrap high-
resolution mass spectrometer. IR spectra were recorded using a
Nicolet Avatar E.S.P. 360 FT-IR instrument. The purities of the
compounds were determined using HPLC and/or qNMR methods,
and compounds (5a−5m) were ≥95% pure. Detailed procedures are
provided in the Supporting Information.
N-boc-Lactam (1). The synthesis of 1 was performed in a manner
similar to that previously reported by Craven et al.34 A 25 mL, single-
neck, round-bottom flask equipped with a 2 cm Teflon-coated
magnetic stir bar and a rubber septum was placed in an ice bath, and
acetonitrile (6 mL) was added via syringe. 2-Pyrrolidinone (1.0 g, 1
equiv, 0.89 mL, 11.8 mmol) was added via syringe, and then di-tert-
butyl dicarbonate (2.69 g, 1.05 equiv, 12.3 mmol) and 4-N,N-
dimethylaminopyridine (0.13 g, 0.09 equiv, 1.06 mmol) were added.
The solution was maintained at 0 °C for 30 min; the ice bath was
G
https://doi.org/10.1021/acs.joc.1c01335
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
chromatography (2 cm × 20 cm SiO2 column, acetone) to furnish the
title compound 3 in 45% yield (0.319 g). Characterization data
corresponds to the literature.34 1H NMR (500 MHz, CDCl3): δ 6.53
(br s, 1H), 6.00 (t, J = 2.5 Hz, 1H), 5.39−5.36 (m, 1H), 3.44 (t, J =
6.5 Hz, 2H), 2.90−2.84 (m, 2H). TLC: Rf = 0.18 (ethyl acetate)
(silica gel, KMnO4).
7.97 (m, 2H), 7.90−7.86 (m, 2H), 6.00−5.95 (m, 1H), 5.56−5.53
(m, 1H), 3.91 (t, J = 7.2 Hz, 2H), 2.91−2.86 (m, 2H). 13C{1H} NMR
(100 MHz, DMSO-d6): δ 167.0. 143.5, 140.5, 133.1, 119.2, 118.9,
117.4, 106.0, 44.5, 23.0. HRMS (TOF MS ES+) m/z: [M + H]+ calcd
for C12H11N2O 199.0866, found 199.0871. IR (ATR): 2899, 2228,
1691, 1658, 832 cm−1. TLC: Rf = 0.63 (ethyl acetate) (silica gel, UV,
KMnO4). Mp: 126−126.5 °C.
General Procedure: N-Arylation of Lactams 5a−m. The
synthesis of lactams 5a−m was performed in a manner similar to that
previously reported.17 A 5 mL, single-neck, round-bottom flask
equipped with a 1 cm Teflon-coated magnetic stir bar and a rubber
septum with a nitrogen inlet needle was charged with copper(I)
iodide (0.15 equiv) and tripotassium phosphate (2 equiv). The aryl
iodide (1.5−3 equiv) and N,N-dimethylethylenediamine (0.3 equiv)
were added to the flask via syringe. In the case in which the aryl iodide
is solid, it was added along with the copper(I) iodide and potassium
phosphate. In a separate scintillation vial, 3-methylene-2-pyrrolidi-
none (1 equiv) was dissolved in toluene (0.1 M, degassed by bubbling
with argon for 20 min), and this solution was transferred all at once to
the reaction flask via syringe. The flask was placed in a preheated oil
bath (80 °C). After the complete disappearance of 3-methylene-2-
pyrrolidinone was observed by TLC, the resulting suspension was
allowed to cool to rt and passed through a 1 cm × 1 cm pad of silica
gel, eluting with ethyl acetate. The filtrate was concentrated by rotary
evaporation, and the residue was purified by silica gel flash column
chromatography.
3-Methylene-1-[(4-methoxy)phenyl]-2-pyrrolidinone (5d).
The synthesis of 5d was performed according to the General
Procedure, using copper(I) iodide (9 mg, 0.05 mmol, 0.15 equiv),
tripotassium phosphate (0.13 g, 0.62 mmol, 2 equiv), 1-iodo-4-
methoxybenzene (4d) (0.22 g, 0.93 mmol, 3 equiv), N,N-
dimethylethylenediamine (10 μL, 0.09 mmol, 0.3 equiv), 3-
methylene-2-pyrrolidinone (3) (30 mg, 0.31 mmol, 1 equiv), and
toluene (3 mL) at 80 °C for 24 h. Purified by silica gel flash column
chromatography (1 cm × 20 cm, 15% ethyl acetate/hexanes). The
title compound 5d was obtained as a white solid (17 mg, 26% yield).
1H NMR (500 MHz, DMSO-d6): δ 7.70−7.65 (m, 2H), 6.99−6.95
(m, 2H), 5.88−5.84 (m, 1H), 5.44−5.41 (m, 1H), 3.85−3.80 (m,
1H), 3.75 (s, 3H), 2.88−2.82 (m, 2H). 13C{1H} NMR (125 MHz,
DMSO-d6): δ 165.8, 156.1, 141.2, 132.8, 121.1, 115.3 (d, J = 5 Hz),
113.9, 55.2 (d, J = 4.6 Hz), 44.9, 23.2. HRMS (TOF MS ES+) m/z:
[M + H]+ calcd for C12H14NO2 204.1019, found 204.1026. IR
(ATR): 2914, 1681, 1650, 1249, 833 cm−1. TLC: Rf = 0.68 (ethyl
acetate) (silica gel, UV, p-anisaldehyde). Mp: 106−107 °C.
3-Methylene-1-phenyl-2-pyrrolidinone (5a). The synthesis of
5a was performed according to the General Procedure, using
copper(I) iodide (9 mg, 0.05 mmol, 0.15 equiv), tripotassium
phosphate (0.13 g, 0.62 mmol, 2 equiv), iodobenzene (4a) (0.1 mL,
0.93 mmol, 3 equiv), N,N-dimethylethylenediamine (10 μL, 0.09
mmol, 0.3 equiv), 3-methylene-2-pyrrolidinone (3) (30 mg, 0.31
mmol, 1 equiv), and toluene (3 mL) at 80 °C for 24 h. Purified by
silica gel flash column chromatography (1 cm × 20 cm, 15% ethyl
acetate/hexanes). The title compound 5a was obtained as a white
3-Methylene-1-[(4-fluoro)phenyl]-2-pyrrolidinone (5e). The
synthesis of 5e was performed according to the General Procedure,
using copper(I) iodide (9 mg, 0.05 mmol, 0.15 equiv), tripotassium
phosphate (0.13 g, 0.62 mmol, 2 equiv), 1-fluoro-iodobenzene (4e)
(107 μL, 0.93 mmol, 3 equiv), N,N-dimethylethylenediamine (10 μL,
0.09 mmol, 0.3 equiv), 3-methylene-2-pyrrolidinone (3) (30 mg, 0.31
mmol, 1 equiv), and toluene (3 mL) at 80 °C for 23 h. Purified by
silica gel flash column chromatography (1 cm × 20 cm, 15% ethyl
acetate/hexanes). The title compound 5e was obtained as a white
crystalline solid (38 mg, 65% yield). 1H NMR (500 MHz, DMSO-d6):
δ 7.82−7.77 (m, 2H), 7.28−7.22 (m, 2H), 5.90 (td, J = 2.5, 1.0 Hz,
1H), 5.47 (td, J = 2.5, 1.0 Hz, 1H), 3.86 (t, J = 6.5 Hz, 2H), 2.89−
2.84 (m, 2H). 13C{1H} NMR (125 MHz, DMSO-d6): δ 166.1, 158.7
(J = 240 Hz), 140.9, 136.0 (J = 2.0 Hz), 121.4 (J = 7.8 Hz), 116.0,
115.4 (J = 22 Hz), 44.8, 23.1. HRMS (TOF MS ES+) m/z: [M + H]+
calcd for C11H11NOF 192.0819, found 192.0824. IR (ATR): 2915,
1681, 1651, 1392, 831, 737 cm−1. TLC: Rf = 0.74 (ethyl acetate)
(silica gel, UV, KMnO4). Mp: 86−87 °C.
1
solid (43 mg, 81% yield). H NMR (400 MHz, DMSO-d6): δ 7.78−
7.76 (m, 2H), 7.43−7.38 (m, 2H), 7.19−7.15 (m, 1H), 5.91−5.89
(m, 1H), 5.47−5.45 (m, 1H), 3.87 (t, J = 6.8 Hz, 2H), 2.89−2.84 (m,
2H). 13C{1H} NMR (100 MHz, DMSO-d6): δ 166.2, 141.1, 139.6,
128.7, 124.3, 119.4, 115.8, 44.6, 23.2. HRMS (TOF MS ES+) m/z:
[M + H]+ calcd for C11H12NO 174.0913, found 174.0916. IR (ATR):
2907, 1673, 1650, 753 cm−1. TLC: Rf = 0.67 (ethyl acetate) (silica gel,
UV, KMnO4). Mp: 67−71 °C.
3-Methylene-1-[(4-trifluoromethyl)phenyl]-2-pyrrolidinone
(5b). The synthesis of 5b was performed according to the General
Procedure, using copper(I) iodide (9 mg, 0.05 mmol, 0.15 equiv),
tripotassium phosphate (0.13 g, 0.62 mmol, 2 equiv), 1-iodo-4-
(trifluoromethyl)benzene (4b) (250 mg, 0.93 mmol, 3 equiv), N,N-
dimethylethylenediamine (10 μL, 0.09 mmol, 0.3 equiv), 3-
methylene-2-pyrrolidinone (3) (30 mg, 0.31 mmol, 1 equiv), and
toluene (3 mL) at 80 °C for 25 h. Purified by silica gel flash column
chromatography (1 cm × 20 cm, 15% ethyl acetate/hexanes). The
title compound 5b was obtained as a white solid (59 mg, 79% yield).
1H NMR (500 MHz, DMSO-d6): δ 8.00 (d, J = 8.5 Hz, 2H), 7.77 (d,
J = 8.5 Hz, 2H), 5.98−5.94 (m, 1H), 5.55−5.51 (m, 1H), 3.92 (t, J =
6.5 Hz, 2H), 2.91−2.88 (m, 2H). 13C{1H} NMR (125 MHz, DMSO-
d6): δ 166.8, 143.0, 140.6, 125.9 (q, J = 3.5 Hz), 124.3 (q, J = 270
Hz), 124.1 (q, J = 32 Hz), 119.1, 117.0, 44.5, 23.0. HRMS (TOF MS
ES+) m/z: [M + H]+ calcd for C12H11NOF3 242.0787, found
242.0800. IR (ATR): 2907, 1685, 1655, 850 cm−1. TLC: Rf = 0.75
(ethyl acetate) (silica gel, UV, KMnO4). Mp: 135−136 °C.
3-Methylene-1-[(4-nitro)phenyl]-2-pyrrolidinone (5f). The
synthesis of 5f was performed according to the General Procedure,
using copper(I) iodide (9 mg, 0.05 mmol, 0.15 equiv), tripotassium
phosphate (0.13 g, 0.62 mmol, 2 equiv), 1-iodo-4-nitrobenzene (4f)
(115 mg, 0.46 mmol, 1.5 equiv), N,N-dimethylethylenediamine (10
μL, 0.09 mmol, 0.3 equiv), 3-methylene-2-pyrrolidinone (3) (30 mg,
0.31 mmol, 1 equiv), and toluene (3 mL) at 80 °C for 24 h. The
nitrogen inlet was replaced with an Ar-filled balloon before the flask
was placed in the oil bath. Purified by silica gel flash column
chromatography (1 cm × 20 cm, 20% ethyl acetate/hexanes, 30%
ethyl acetate/hexanes, and 50% ethyl acetate/hexanes). The title
1
compound 5f was obtained as a yellow solid (46 mg, 68% yield). H
NMR (500 MHz, DMSO-d6): δ 8.33−8.28 (m, 2H), 8.08−8.04 (m,
2H), 6.05−5.95 (m, 1H), 5.59−5.56 (m, 1H), 3.96 (t, J = 6.5 Hz,
2H), 2.93−2.88 (m, 2H). 13C{1H} NMR (125 MHz, DMSO-d6): δ
167.1, 145.2, 142.7, 140.3, 124.6, 118.8, 117.7, 44.7, 22.9. HRMS
(TOF MS ES+) m/z: [M + H]+ calcd for C11H11N2O3 219.0764,
found 219.0774. IR (ATR): 2909, 1692, 1652, 1505, 817 cm−1. TLC:
Rf = 0.41 (50% ethyl acetate/hexanes) (silica gel, UV, KMnO4). Mp:
171−173 °C.
3-Methylene-1-[(4-(N,N-dimethyl)phenyl]-2-pyrrolidinone
(5g). The synthesis of 5g was performed according to the General
Procedure, using copper(I) iodide (9 mg, 0.05 mmol, 0.15 equiv),
tripotassium phosphate (0.13 g, 0.62 mmol, 2 equiv), 1-iodo-4-N,N-
dimethylaniline (4g) (114 mg, 0.46 mmol, 1.5 equiv), N,N-
dimethylethylenediamine (10 μL, 0.09 mmol, 0.3 equiv), 3-
methylene-2-pyrrolidinone (3) (30 mg, 0.31 mmol, 1 equiv), and
3-Methylene-1-[(4-cyano)phenyl]-2-pyrrolidinone (5c). The
synthesis of 5c was performed according to the General Procedure,
using copper(I) iodide (9 mg, 0.05 mmol, 0.15 equiv), tripotassium
phosphate (0.13 g, 0.62 mmol, 2 equiv), 4-iodobenzonitrile (4c) (113
mg, 0.49 mmol, 1.6 equiv), N,N-dimethylethylenediamine (10 μL,
0.09 mmol, 0.3 equiv), 3-methylene-2-pyrrolidinone (3) (30 mg, 0.31
mmol, 1 equiv), and toluene (3 mL) at 80 °C for 18 h. Purified by
silica gel flash column chromatography (1 cm × 20 cm, 30% ethyl
acetate/hexanes). The title compound 5c was obtained as a white
1
solid (42 mg, 69% yield). H NMR (400 MHz, DMSO-d6): δ 8.02−
H
https://doi.org/10.1021/acs.joc.1c01335
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
toluene (3 mL) at 80 °C for 21.5 h. Purified by silica gel flash column
chromatography (1 cm × 20 cm, 30% ethyl acetate/hexanes). The
title compound 5g was obtained as a pale yellow solid (48 mg, 72%
yield). 1H NMR (500 MHz, DMSO-d6): δ 7.60−7.54 (m, 2H), 6.80−
6.72 (m, 2H), 5.84−5.80 (m, 1H), 5.40−5.37 (m, 1H), 3.80 (t, J =
6.5 Hz, 2H), 2.89 (s, 6H), 2.85−2.80 (m, 2H). 13C{1H} NMR (125
MHz, DMSO-d6): δ 165.5, 147.8, 141.4, 129.3, 120.9, 114.7, 112.4,
44.9, 40.3, 23.3. HRMS (TOF MS ES+) m/z: [M + H]+ calcd for
C13H16N2O 217.1341, found 217.1350. IR (ATR): 2894, 1668, 1649,
1176, 808 cm−1. TLC: Rf = 0.68 (ethyl acetate) (silica gel, UV,
KMnO4). Mp: 108.5−110 °C.
3-Methylene-1-[(2-methoxy)phenyl]-2-pyrrolidinone (5k).
The synthesis of 5k was performed according to the General
Procedure, using copper(I) iodide (9 mg, 0.05 mmol, 0.15 equiv),
tripotassium phosphate (0.13 g, 0.62 mmol, 2 equiv), 3-methylene-2-
pyrrolidinone (3) (30 mg, 0.31 mmol, 1.0 equiv), N,N-dimethyle-
thylenediamine (10 μL, 0.09 mmol, 0.3 equiv), 2-iodoanisole (4k)
(60 μL, 0.46 mmol, 1.5 equiv), and toluene (3 mL) at 80 °C for 23 h.
Purified by silica gel flash column chromatography (1 cm × 15 cm,
30% ethyl acetate/hexanes). The title compound 5k was obtained as a
light yellow/transparent oil that solidified on standing (38 mg, 60%
yield). 1H NMR (400 MHz, DMSO-d6): δ 7.37−7.07 (m, 3H), 7.03−
6.94 (m, 1H), 5.84−5.77 (m, 1H), 5.44−5.40 (m, 1H), 3.78 (s, 3H),
3.68 (t, J = 6.8 Hz, 2H), 2.93−2.84 (m, 2H). 13C{1H} NMR (100
MHz, DMSO-d6): δ 166.5, 154.6, 140.4, 128.8, 128.4, 127.3, 120.5,
115.2, 112.4, 55.6, 46.0, 24.2. HRMS (TOF MS ES+) m/z: [M + H]+
calcd for C12H14NO2 204.1019, found 204.1027. IR (ATR): 2907,
1685, 1658, 1276, 762 cm−1. TLC: Rf = 0.23 (50% ethyl acetate/
hexanes) (silica gel, UV, KMnO4). Mp: 81.5−83 °C.
3-Methylene-1-[(3-methoxy)phenyl]-2-pyrrolidinone (5h).
The synthesis of 4h was performed according to the General
Procedure, using copper(I) iodide (9 mg, 0.05 mmol, 0.15 equiv),
tripotassium phosphate (0.13 g, 0.62 mmol, 2 equiv), 3-methylene-2-
pyrrolidinone (3) (30 mg, 0.31 mmol, 1 equiv), N,N-dimethylethy-
lenediamine (10 μL, 0.09 mmol, 0.3 equiv), 3-iodoanisole (4h) (53
μL, 0.46 mmol, 1.5 equiv), and toluene (3 mL) at 80 °C for 24 h.
Purified by silica gel flash column chromatography (1 cm × 20 cm,
30% ethyl acetate/hexanes). The title compound 5h was obtained as a
3-Methylene-1-[(2-fluoro)phenyl]-2-pyrrolidinone (5l). The
synthesis of 5l was performed according to the General Procedure,
using copper(I) iodide (9 mg, 0.05 mmol, 0.15 equiv), tripotassium
phosphate (0.13 g, 0.62 mmol, 2 equiv), 3-methylene-2-pyrrolidinone
(3) (30 mg, 0.31 mmol, 1 equiv), N,N-dimethylethylenediamine (10
μL, 0.09 mmol, 1 equiv), 2-fluoroiodobenzene 4l (54 μL, 0.46 mmol,
1.5 equiv), and toluene (3 mL) at 80 °C for 25 h. Purified by silica gel
flash column chromatography (1 cm × 15 cm, 30% ethyl acetate/
hexanes). The title compound 5l was obtained as a white crystalline
1
white solid (36 mg, 57% yield). H NMR (400 MHz, DMSO-d6): δ
7.49−7.45 (m, 1H), 7.35−7.25 (m, 2H), 6.79−6.74 (m, 1H), 5.92−
5.88 (m, 1H), 5.48−5.44 (m, 1H), 3.86 (t, J = 6.4 Hz, 2H), 3.76 (s,
3H), 2.89−2.82 (m, 2H). 13C{1H} NMR (100 MHz, DMSO-d6): δ
166.4, 159.5, 141.2, 140.8, 129.6, 116.0, 111.6, 109.8, 105.6, 55.1,
44.8, 23.1. HRMS (TOF MS ES+) m/z: [M + H]+ calcd for
C12H14NO2 204.1019, found 204.1026. IR (ATR): 2914, 1676, 1650,
1278, 873, 668 cm−1. TLC: Rf = 0.79 (ethyl acetate) (silica gel, UV,
KMnO4). Mp: 69−70.5 °C.
1
solid (48 mg, 81% yield). H NMR (400 MHz, DMSO-d6): δ 7.51
(app ddd, J = 15.6, 7.6, 1.6 Hz, 1H), 7.41−7.22 (m, 3H), 5.90−5.86
(m, 1H), 5.50−5.47 (m, 1H), 3.79 (t, J = 6.8 Hz, 2H), 2.96−2.88 (m,
2H). 13C{1H} NMR (100 MHz, DMSO-d6): δ 166.2, 156.6 (d, J =
249.2 Hz), 139.8, 128.8 (d, J = 8.0 Hz), 127.7 (d, J = 1.8 Hz), 126.5
(d, J = 12.1 Hz), 124.8 (d, J = 3.6 Hz), 116.5 (d, J = 20.0 Hz), 116.1,
46.1, 24.3. HRMS (TOF MS ES+) m/z: [M + H]+ calcd for
C11H11NOF 192.0819, found 192.0824. IR (ATR): 2906, 1685, 1657,
1261, 756, 668 cm−1. TLC: Rf = 0.80 (ethyl acetate) (silica gel, UV,
KMnO4). Mp: 80−82 °C.
3-Methylene-1-[(2-nitro)phenyl]-2-pyrrolidinone (5m). The
synthesis of 5m was performed according to the General Procedure,
using copper(I) iodide (9 mg, 0.05 mmol, 0.15 equiv), tripotassium
phosphate (0.13 g, 0.62 mmol, 2 equiv), 1-iodo-3-nitrobenzene (4m)
(115 mg, 0.46 mmol, 1.5 equiv), N,N-dimethylethylenediamine (10
μL, 0.09 mmol, 0.3 equiv), 3-methylene-2-pyrrolidinone (3) (30 mg,
0.31 mmol, 1 equiv), and toluene (3 mL) at 80 °C for 72 h. Purified
by silica gel flash column chromatography (1 cm × 20 cm, 30% ethyl
acetate/hexanes). The title compound 5m was obtained as a yellow
oil (54 mg, 80% yield). 1H NMR (400 MHz, DMSO-d6): δ 7.99 (dd,
J = 8.0, 1.6 Hz, 1H), 7.83−7.78 (m, 1H), 7.68 (dd, J = 8.4. 1.6 Hz,
1H), 7.57−7.51 (m, 1H), 5.87−5.84 (m, 1H), 5.53−5.50 (m, 1H),
3.97 (t, J = 6.8 Hz, 2H), 2.97−2.91 (m, 2H). 13C{1H} NMR (100
MHz, DMSO-d6): δ 166.3, 145.0, 139.4, 134.1, 131.5, 127.5, 126.2,
125.1, 117.1, 45.9, 24.2. HRMS (TOF MS ES+) m/z: [M + H]+ calcd
for C11H11N2O3 219.0764, found 219.0762. IR (ATR): 2895, 1693,
1658, 1525, 755 cm−1. TLC: Rf = 0.25 (50% ethyl acetate/hexanes)
(silica gel, UV, KMnO4). Mp: 111−114.5 °C.
3-Methylene-1-[(3-fluoro)phenyl]-2-pyrrolidinone (5i). The
synthesis of 5i was performed according to the General Procedure,
using copper(I) iodide (9 mg, 0.05 mmol, 0.15 equiv), tripotassium
phosphate (0.13 g, 0.62 mmol, 2 equiv), 3-methylene-2-pyrrolidinone
(3) (30 mg, 0.31 mmol, 1 equiv), N,N-dimethylethylenediamine (10
μL, 0.09 mmol, 0.3 equiv), 1-fluoro-3-iodobenzene (4i) (55 μL, 0.46
mmol, 1.5 equiv), and toluene (3 mL) at 80 °C for 24 h. The reaction
flask was wrapped with aluminum foil before being placed in the oil
bath. Purified by silica gel flash column chromatography (1 cm × 20
cm, 30% ethyl acetate/hexanes). The title compound 5i was obtained
as a white solid (35 mg, 60% yield). 1H NMR (400 MHz, DMSO-d6):
δ 7.78 (app dt, J = 12.0 and 2.0 Hz, 1H), 7.57−7.52 (m, 1H), 7.48−
7.41 (m, 1H), 7.06−6.97 (m, 1H), 5.94−5.90 (m, 1H), 5.52−5.48
(m, 1H), 3.87 (t, J = 6.8 Hz, 2H), 2.90−2.84 (m, 2H). 13C{1H} NMR
(100 MHz, DMSO-d6): δ 166.6, 162.1 (d, J = 240 Hz), 141.2 (d, J =
10.9 Hz), 140.8, 130.4 (d, J = 9.4 Hz), 116.6, 114.8 (d, J = 2.8 Hz),
110.8 (d, J = 21.0 Hz), 106.2 (d, J = 26.3 Hz), 44.7, 23.0. HRMS
(TOF MS ES+) m/z: [M + H]+ calcd for C11H11NOF 192.0819,
found 192.0824. IR (ATR): 2905, 1679, 1653, 1172, 868, 681 cm−1.
TLC: Rf = 0.68 (ethyl acetate) (silica gel, UV, KMnO4). Mp: 82−83
°C.
3-Methylene-1-[(3-nitro)phenyl]-2-pyrrolidinone (5j). The
synthesis of 5j was performed according to the General Procedure,
using copper(I) iodide (9 mg, 0.05 mmol, 0.15 equiv), tripotassium
phosphate (0.13 g, 0.62 mmol, 2 equiv), 1-iodo-3-nitrobenzene (4j)
(115 mg, 0.46 mmol, 1.5 equiv), N,N-dimethylethylenediamine (10
μL, 0.09 mmol, 0.3 equiv), 3-methylene-2-pyrrolidinone (3) (30 mg,
0.31 mmol, 1 equiv), and toluene (3 mL) at 80 °C for 18.5 h. The
nitrogen inlet was then replaced with an Ar-filled balloon before the
flask was placed in the oil bath. Purified by silica gel flash column
chromatography (1 cm × 20 cm, 30% ethyl acetate/hexanes). The
title compound 5j was obtained as a fine yellow powder (20 mg, 31%
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.joc.1c01335.
1
yield). H NMR (500 MHz, DMSO-d6): δ 8.84 (app t, J = 2.5 Hz,
1
Results of GSH reactivity studies, H and 13C NMR
1H), 8.07 (dd, J = 8.0, 2.0 Hz, 1H), 8.02 (dd, J = 8.0, 2.0 Hz, 1H),
7.71 (t, J = 8.0 Hz, 1H), 6.00−5.95 (m, 1H), 5.57−5.52 (m, 1H),
3.95 (t, J = 7.0 Hz, 2H), 2.95−2.86, (m, 2H). 13C{1H} NMR (125
MHz, DMSO-d6): δ 166.9, 147.9, 140.5, 140.4, 130.2, 124.9, 118.6,
117.1, 113.4, 44.6, 23.0. HRMS (TOF MS ES+) m/z: [M + H]+ calcd
for C11H11N2O3 219.0764, found 219.0775. IR (ATR): 3127, 2980,
1682, 1656, 1526, 894 cm−1. TLC: Rf = 0.46 (ethyl acetate) (silica gel,
UV, KMnO4). Mp: 139.3−141.7 °C.
spectra of compounds, Cartesian coordinates, details of
computational studies, and energies of optimized
structures (PDF)
FAIR data, including the primary NMR FID files, for
compounds 1−3 and 5a−m (ZIP)
I
https://doi.org/10.1021/acs.joc.1c01335
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
(8) Gehringer, M.; Laufer, S. A. Emerging and Re-Emerging
Warheads for Targeted Covalent Inhibitors: Applications in Medicinal
Chemistry and Chemical Biology. J. Med. Chem. 2019, 62, 5673−
5724.
AUTHOR INFORMATION
■
Corresponding Authors
Peng Liu − Department of Chemistry, University of Pittsburgh,
Pittsburgh, Pennsylvania 15260, United States; Department
of Chemical and Petroleum Engineering, University of
Pittsburgh, Pittsburgh, Pennsylvania 15261, United States;
orcid.org/0000-0002-8188-632X; Email: pengliu@
pitt.edu
Kay M. Brummond − Department of Chemistry, University of
Pittsburgh, Pittsburgh, Pennsylvania 15260, United States;
orcid.org/0000-0003-3595-6806; Email: kbrummon@
pitt.edu
(9) Dalton, S. E.; Campos, S. Covalent Small Molecules as Enabling
Platforms for Drug Discovery. ChemBioChem 2020, 21, 1080−1100.
(10) Lonsdale, R.; Ward, R. A. Structure-based Design of Targeted
Covalent Inhibitors. Chem. Soc. Rev. 2018, 47, 3816−3830.
(11) Jackson, P. A.; Schares, H. A. M.; Jones, K. F. M.; Widen, J. C.;
Dempe, D. P.; Grillet, F.; Cuellar, M. E.; Walters, M. A.; Harki, D. A.;
Brummond, K. M. Synthesis of Guaianolide Analogues with a
Tunable α-Methylene−γ-lactam Electrophile and Correlating Bio-
activity with Thiol Reactivity. J. Med. Chem. 2020, 63, 14951−14978.
(12) Cee, V. J.; Volak, L. P.; Chen, Y.; Bartberger, M. D.; Tegley, C.;
Arvedson, T.; McCarter, J.; Tasker, A. S.; Fotsch, C. Systematic Study
of the Glutathione (GSH) Reactivity of N-arylacrylamides: 1. Effects
of Aryl Substitution. J. Med. Chem. 2015, 58, 9171−9178.
(13) Birkholz, A.; Kopecky, D. J.; Volak, L. P.; Bartberger, M. D.;
Chen, Y.; Tegley, C. M.; Arvedson, T.; McCarter, J. D.; Fotsch, C.;
Cee, V. J. Systematic Study of the Glutathione Reactivity of N-
Phenylacrylamides: 2. Effects of Acrylamide Substitution. J. Med.
Chem. 2020, 63, 11602−11614.
Authors
̆
Tugcȩ G. Erbay − Department of Chemistry, University of
Pittsburgh, Pittsburgh, Pennsylvania 15260, United States;
orcid.org/0000-0002-9603-324X
Daniel P. Dempe − Department of Chemistry, University of
Pittsburgh, Pittsburgh, Pennsylvania 15260, United States
Bhaskar Godugu − Department of Chemistry, University of
Pittsburgh, Pittsburgh, Pennsylvania 15260, United States
(14) Flanagan, M. E.; Abramite, J. A.; Anderson, D. P.; Aulabaugh,
A.; Dahal, U. P.; Gilbert, A. M.; Li, C.; Montgomery, J.;
Oppenheimer, S. R.; Ryder, T.; Schuff, B. P.; Uccello, D. P.;
Walker, G. S.; Wu, Y.; Brown, M. F.; Chen, J. M.; Hayward, M. M.;
Noe, M. C.; Obach, R. S.; Philippe, L.; Shanmugasundaram, V.;
Shapiro, M. J.; Starr, J.; Stroh, J.; Che, Y. Chemical and
Computational Methods for the Characterization of Covalent
Reactive Groups for the Prospective Design of Irreversible Inhibitors.
J. Med. Chem. 2014, 57, 10072−10079.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.joc.1c01335
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(15) Niggenaber, J.; Heyden, L.; Grabe, T.; Muller, M. P.; Lategahn,
̈
J.; Rauh, D. Complex Crystal Structures of EGFR with Third-
generation Kinase Inhibitors and Simultaneously Bound Allosteric
Ligands. ACS Med. Chem. Lett. 2020, 11, 2484−2490.
This work was financially supported by the University of
Pittsburgh and the National Institutes of Health
(R35GM128779). Calculations were performed at the Center
for Research Computing at the University of Pittsburgh, the
Extreme Science and Engineering Discovery Environment
(XSEDE), and the TACC Frontera supercomputers.
(16) Craven, P.; Aimon, A.; Dow, M.; Fleury-Bregeot, N.; Guilleux,
R.; Morgentin, R.; Roche, D.; Kalliokoski, T.; Foster, R.; Marsden, S.
P.; Nelson, A. Design, Synthesis and Decoration of Molecular
Scaffolds for Exploitation in the Production of Alkaloid-like Libraries.
Bioorg. Med. Chem. 2015, 23, 2629−2635.
REFERENCES
(17) Klapars, A.; Huang, X.; Buchwald, S. L. A General and Efficient
Copper Catalyst for the Amidation of Aryl Halides. J. Am. Chem. Soc.
2002, 124, 7421−7428.
■
(1) Ferguson, F. M.; Gray, N. S. Kinase Inhibitors: The Road Ahead.
Nat. Rev. Drug Discovery 2018, 17, 353.
́
̋
(18) Petri, L.; Abrányi-Balogh, P.; Varga, P. R.; Imre, T.; Keseru, G.
M. Comparative Reactivity Analysis of Small-molecule Thiol
Surrogates. Bioorg. Med. Chem. 2020, 28, 115357.
(2) Barf, T.; Kaptein, A. Irreversible Protein Kinase Inhibitors:
Balancing the Benefits and Risks. J. Med. Chem. 2012, 55, 6243−6262.
(3) Abdeldayem, A.; Raouf, Y. S.; Constantinescu, S. N.; Moriggl, R.;
Gunning, P. T. Advances in Covalent Kinase Inhibitors. Chem. Soc.
Rev. 2020, 49, 2617−2687.
(4) Greig, S. L. Osimertinib: First Global Approval. Drugs 2016, 76,
263−273.
(5) Callegari, D.; Ranaghan, K. E.; Woods, C. J.; Minari, R.; Tiseo,
M.; Mor, M.; Mulholland, A. J.; Lodola, A. L718Q Mutant EGFR
Escapes Covalent Inhibition by Stabilizing a Non-reactive Con-
formation of the Lung Cancer Drug Osimertinib. Chem. Sci. 2018, 9,
2740−2749.
(6) Walter, A. O.; Sjin, R. T. T.; Haringsma, H. J.; Ohashi, K.; Sun,
J.; Lee, K.; Dubrovskiy, A.; Labenski, M.; Zhu, Z.; Wang, Z.; Sheets,
M.; St Martin, T.; Karp, R.; van Kalken, D.; Chaturvedi, P.; Niu, D.;
Nacht, M.; Petter, R. C.; Westlin, W.; Lin, K.; Jaw-Tsai, S.; Raponi,
M.; Van Dyke, T.; Etter, J.; Weaver, Z.; Pao, W.; Singh, J.; Simmons,
A. D.; Harding, T. C.; Allen, A. Discovery of a Mutant-Selective
Covalent Inhibitor of EGFR that Overcomes T790M-Mediated
Resistance in NSCLC. Cancer Discovery 2013, 3, 1404−1415.
(7) Thress, K. S.; Paweletz, C. P.; Felip, E.; Cho, B. C.; Stetson, D.;
Dougherty, B.; Lai, Z.; Markovets, A.; Vivancos, A.; Kuang, Y.; Ercan,
D.; Matthews, S. E.; Cantarini, M.; Barrett, J. C.; Jänne, P. A.; Oxnard,
G. R. Acquired EGFR C797S Mutation Mediates Resistance to
AZD9291 in Non−small Cell Lung Cancer Harboring EGFR T790M.
Nat. Med. 2015, 21, 560−562.
(19) Ward, R. A.; Anderton, M. J.; Ashton, S.; Bethel, P. A.; Box, M.;
Butterworth, S.; Colclough, N.; Chorley, C. G.; Chuaqui, C.; Cross,
́
D. A. E.; Dakin, L. A.; Debreczeni, J. E.; Eberlein, C.; Finlay, M. R. V.;
Hill, G. B.; Grist, M.; Klinowska, T. C. M.; Lane, C.; Martin, S.;
Orme, J. P.; Smith, P.; Wang, F.; Waring, M. J. Structure- and
Reactivity-Based Development of Covalent Inhibitors of the
Activating and Gatekeeper Mutant Forms of the Epidermal Growth
Factor Receptor (EGFR). J. Med. Chem. 2013, 56, 7025−7048.
(20) Hansch, C.; Leo, A.; Taft, R. W. A Survey of Hammett
Substituent Constants and Resonance and Field Parameters. Chem.
Rev. 1991, 91, 165−195.
(21) We attribute the variability in the experimentally measured half-
lives of N-phenylacrylamide to differences in experimental conditions.
Cee reported 179 min for the reaction under the following conditions:
1 μM NPA, 37 °C, 5 mM GSH, 67 mM PPB, pH 7.4 with 1−1.5%
DMSO (ref 12). Flanagan reported 55 min under the following
conditions: 10 mM NPA, 37 °C, 10 mM GSH, 100 mM PPB, pH 7.4
with 10% acetonitrile (ref 14). Ward reported 299 min under the
following conditions: 50 μM NPA, 37 °C, 4.6 mM GSH, phosphate
buffer, pH 7.4 with 10% acetonitrile (ref 10).
(22) Cusack, K. P.; Arnold, L. D.; Barberis, C. E.; Chen, H.;
Ericsson, A. M.; Gaza-Bulseco, G. S.; Gordon, T. D.; Grinnell, C. M.;
Harsch, A.; Pellegrini, M.; Tarcsa, E. A. 13C NMR Approach to
J
https://doi.org/10.1021/acs.joc.1c01335
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
Categorizing Potential Limitations of α,β-Unsaturated Carbonyl
Systems in Drug-like Molecules. Bioorg. Med. Chem. Lett. 2004, 14,
5503−5507.
(23) Krishnan, S.; Miller, R. M.; Tian, B.; Mullins, R. D.; Jacobson,
M. P.; Taunton, J. Design of Reversible, Cysteine-targeted Michael
Acceptors Guided by Kinetic and Computational Analysis. J. Am.
Chem. Soc. 2014, 136, 12624−12630.
(24) Krenske, E. H.; Petter, R. C.; Zhu, Z.; Houk, K. N. Transition
States and Energetics of Nucleophilic Additions of Thiols to
Substituted α,β-Unsaturated Ketones: Substituent Effects Involve
Enone Stabilization, Product Branching, and Solvation. J. Org. Chem.
2011, 76, 5074−5081.
(25) Krenske, E. H.; Petter, R. C.; Houk, K. N. Kinetics and
Thermodynamics of Reversible Thiol Additions to Mono- and
Diactivated Michael Acceptors: Implications for the Design of
Drugs That Bind Covalently to Cysteines. J. Org. Chem. 2016, 81,
11726−11733.
(26) Allgäuer, D. S.; Jangra, H.; Asahara, H.; Li, Z.; Chen, Q.; Zipse,
H.; Ofial, A. R.; Mayr, H. Quantification and Theoretical Analysis of
the Electrophilicities of Michael Acceptors. J. Am. Chem. Soc. 2017,
139, 13318−13329.
(27) Schwöbel, J. A. H.; Wondrousch, D.; Koleva, Y. K.; Madden, J.
C.; Cronin, M. T. D.; Schuurmann, G. Prediction of Michael-type
̈ ̈
Acceptor Reactivity Toward Glutathione. Chem. Res. Toxicol. 2010,
23, 1576−1585.
(28) Mulliner, D.; Wondrousch, D.; Schuurmann, G. Predicting
Michael-acceptor Reactivity and Toxicity Through Quantum
Chemical Transition-state Calculations. Org. Biomol. Chem. 2011, 9,
8400−8412.
(29) Wondrousch, D.; Böhme, A.; Thaens, D.; Ost, N.; Schuurmann,
̈ ̈
G. Local Electrophilicity Predicts the Toxicity-relevant Reactivity of
Michael Acceptors. J. Phys. Chem. Lett. 2010, 1, 1605−1610.
(30) Berteotti, A.; Vacondio, F.; Lodola, A.; Bassi, M.; Silva, C.;
Mor, M.; Cavalli, A. Predicting the Reactivity of Nitrile-Carrying
Compounds with Cysteine: A Combined Computational and
Experimental Study. ACS Med. Chem. Lett. 2014, 5, 501−505.
(31) Cohen, A. J.; Mori-Sánchez, P.; Yang, W. Challenges for
Density Functional Theory. Chem. Rev. 2012, 112, 289−320.
(32) The computed ΔG⧧DFT is with respect to the deprotonated
glutathione and thus is expected to be lower than the experimental
ΔG⧧Experiment because the glutathione exists in the protonated form
under the experimental conditions. On the basis of the reaction pH
and the pKa of glutathione, the deprotonation is expected to be uphill
by 3.1 kcal/mol.
(33) Ribeiro, R. F.; Marenich, A. V.; Cramer, C. J.; Truhlar, D. G.
Use of Solution-phase Vibrational Frequencies in Continuum Models
for the Free Energy of Solvation. J. Phys. Chem. B 2011, 115, 14556−
14562.
(34) Craven, P.; Aimon, A.; Dow, M.; Fleury-Bregeot, N.; Guilleux,
R.; Morgentin, R.; Roche, D.; Kalliokoski, T.; Foster, R.; Marsden, S.
P.; Nelson, A. Design, Synthesis and Decoration of Molecular
Scaffolds for Exploitation in the Production of Alkaloid-like Libraries.
Bioorg. Med. Chem. 2015, 23, 2629−2635.
K
https://doi.org/10.1021/acs.joc.1c01335
J. Org. Chem. XXXX, XXX, XXX−XXX