In pursuit of a selective hepatocellular carcinoma therapeutic agent: Novel thalidomide derivatives with antiproliferative, antimigratory and STAT3 inhibitory properties

Article abstract of DOI:10.1016/j.ejmech.2021.113353

Advanced stage liver cancer is predominantly treated with the multi-kinase inhibitor sorafenib; however, this therapeutic agent lacks selectivity in its cytotoxic actions and is associated with poor survival outcomes. Herein we report the design and preparation of several thalidomide derivatives, including a variety of novel thioether-containing forms that are especially rare in the literature. Importantly, two of the derivatives described are potent antiproliferative agents with dose-dependent selectivity for tumorigenic liver progenitor cells (LPC) growth inhibition (up to 36% increase in doubling time at 10 μM) over non-tumorigenic cells (no effect at 10 μM). Furthermore, these putative anti-liver cancer agents were also found to be potent inhibitors of tumorigenic LPC migration. This report also describes these derivatives’ effects on several key signalling pathways in our novel liver cell lines by immunofluorescence and AlphaLISA assays. Aryl thioether derivative 7f significantly reduced STAT3 phosphorylation (23%) and its nuclear localisation (16%) at 10 μM in tumorigenic LPCs, implicating the IL-6/JAK/STAT3 axis is central in the mode of action of our derivatives.

Full text of DOI:10.1016/j.ejmech.2021.113353

European Journal of Medicinal Chemistry 217 (2021) 113353
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
In pursuit of a selective hepatocellular carcinoma therapeutic agent:
Novel thalidomide derivatives with antiproliferative, antimigratory
and STAT3 inhibitory properties
,
,
Michael J. Nutt a b, Yeung Sing Yee a, Amanda Buyan c, Neil Andrewartha a b, Ben Corry c,
George C.T. Yeoh a **, Scott G. Stewart a
,
b,
, *
a School of Molecular Sciences, The University of Western Australia, Crawley, WA, 6009, Australia
b The Centre for Medical Research, The Perkins Institute of Medical Research, Nedlands, WA, 6009, Australia
c Research School of Biology, Australian National University, Acton, ACT, 2601, Australia
a r t i c l e i n f o
a b s t r a c t
Article history:
Advanced stage liver cancer is predominantly treated with the multi-kinase inhibitor sorafenib; however,
this therapeutic agent lacks selectivity in its cytotoxic actions and is associated with poor survival
outcomes. Herein we report the design and preparation of several thalidomide derivatives, including a
variety of novel thioether-containing forms that are especially rare in the literature. Importantly, two of
the derivatives described are potent antiproliferative agents with dose-dependent selectivity for
Received 4 January 2021
Received in revised form
25 February 2021
Accepted 26 February 2021
Available online 10 March 2021
tumorigenic liver progenitor cells (LPC) growth inhibition (up to 36% increase in doubling time at 10
mM)
over non-tumorigenic cells (no effect at 10 M). Furthermore, these putative anti-liver cancer agents
m
were also found to be potent inhibitors of tumorigenic LPC migration. This report also describes these
derivatives’ effects on several key signalling pathways in our novel liver cell lines by immunofluores-
cence and AlphaLISA assays. Aryl thioether derivative 7f significantly reduced STAT3 phosphorylation
(23%) and its nuclear localisation (16%) at 10
mM in tumorigenic LPCs, implicating the IL-6/JAK/STAT3 axis
is central in the mode of action of our derivatives.
© 2021 Elsevier Masson SAS. All rights reserved.
1. Introduction
myelodysplastic syndrome (MDS), erythema nodosum leprosum
(ENL) and psoriatic arthritis [7e9].
In the last 20 years a resurgence of thalidomide use in drug
therapy has uncovered interesting and desirable biological activ-
ities including enhanced T-cell stimulation and proliferation, as
well as anti-tumour and anti-angiogenic effects [1e3]. Further-
more, thalidomide and its derivatives (lenalidomide, pomalido-
mide, avadomide, iberdomide, and CC-885 from Celgene, Fig. 1) can
modulate the immune response in several cell types, as well as
indirectly influence cell signalling pathways involved in growth,
proliferation, and apoptosis [4e6]. Currently, this family of drugs
has been approved use for treating a range of disease states
including multiple myeloma (MM), 5q deletion-associated
For decades the mechanism(s) underlying thalidomide’s pleio-
tropic biological activities remained unclear, and presumed the
involvement of multiple distinct drug targets [10]. In 2010 however,
a landmark paper from the Ito group revealed cereblon (CRBN) as
the primary target of thalidomide, and significantly, it was
responsible for its teratogenic effects in early-term pregnancies.
CRBN forms part of an E3 ubiquitin ligase complex (known
collectively as CRL4CRBN) that regulates proteosomal degradation
via substrate ubiquitination, and upon binding of thalidomide to
CRBN substrate specificity of the ligase activity is altered, inducing
protein degradation of downstream targets [11,12]. While originally
described as immunomodulatory drugs (IMiDs), thalidomide and
its derivatives are now also referred to as CELMoDs (cereblon
modulators), to better reflect its more expanded mechanism of
action.
* Corresponding author. School of Molecular Sciences, The University of Western
Australia, Crawley, WA, 6009, Australia.
** Corresponding author. The Centre for Medical Research, The Perkins Institute of
Medical Research, Nedlands, WA, 6009, Australia.
In the last decade, protein crystallography studies have revealed
the CRBN binding site of thalidomide and other CELMoDs, to be
E-mail address: scott.stewart@uwa.edu.au (S.G. Stewart).
https://doi.org/10.1016/j.ejmech.2021.113353
0223-5234/© 2021 Elsevier Masson SAS. All rights reserved.

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
approach to assess both selectivity and efficacy in our screens by
comparing responses in matched pairs of BMOL-NT cells versus
BMOL-Tcells. Our early structure-activity relationship (SAR) studies
of thalidomide analogues in TNFa expression inhibition and
apoptosis studies highlighted several analogues with higher po-
tency than thalidomide and lenalidomide [34e36]. Suppression of
TNFa expression is a well-documented property of the IMiDs, and
its role in HCC warranted further study of our derivatives as po-
tential therapeutic agents [37e39]. In a preliminary investigation
we recently identified a standout aryl amine-bearing thalidomide
analogue which inhibited the growth of tumorigenic liver cells
without affecting BMOL-NT cell growth [22]. The present work
seeks to discover related compounds with improved anticancer
potency while maintaining the desired selectivity for tumorigenic
liver cells. We also seek to understand its mechanism of action by
identifying signalling pathways involved in this selective growth
inhibition. The potential role of CRBN binding in eliciting these
actions is also examined both experimentally and through molec-
ular dynamics (MD) simulation approaches.
Fig. 1. Structure of thalidomide, sorafenib, and thalidomide derivatives developed by
Celgene.
2. Results and discussion
highly conserved between species from bacterial to mammalian
[13,14]. It is now known that the glutarimide ring shared by these
compounds is essential for CRBN binding, while the phthaloyl
scaffold remains partially exposed where it forms part of the
protein-protein interface between CRBN and the target of ubiqui-
nation [15e17]. Unique neosubstrates of CRBN have been identified
that mediate both the therapeutic effects of these drugs in MM (the
zinc finger transcription factors IKZF1 and IKZF3) and MDS (the
2.1. Compound design
From the structure of our previously identified selective BMOL-T
growth inhibitor we envisioned several points of structural modi-
fication (Fig. 2). Firstly, the amine linker could be exchanged with
other heteroatoms or a methylene group, to probe for any favour-
able binding interactions via hydrogen bonding with a biological
target. Similarly, bioisosteric replacement of the aromatic ring with
heterocycles was desired as, if tolerated, they would provide a
means to further modulate lipophilicity and solubility of our
compounds. Only derivatives with alkyl-substituted pendant aryl
groups had been assayed in our previous study; installation of a
diverse range of functional groups would help establish SAR re-
lationships to better inform drug design efforts and identify any
opportunities to exploit hydrogen bonding networks with any
biological targets.
serine/threonine kinase CK1a), as well as some of thalidomide’s
maligned birth defects (SALL4, another zinc finger) [18]. The ability
to induce degradation of these targets differs between each
member of the CELMoD class, and thus it is now understood that
subtle structural differences in the substitution at the phthaloyl
ring can create various opportunities for further intermolecular
interactions that reflect their unique degradation profiles [19].
While thalidomide and some derivatives have been especially
successful in treating hematological malignancies, reports of their
use in the treatment of HCC is rare in the literature [20e23]. He-
patocellular carcinoma is an especially lethal disease: estimates
place it as the sixth most common cancer, and yet responsible for
the fourth highest number of cancer-related deaths worldwide
[24,25]. Liver cancers are notoriously resistant to chemotherapeutic
agents due to the liver’s inherent ability to rapidly metabolise and
efflux exogenous agents. Partial liver resection or transplant remain
the best curative options for HCC patients, but require sufficiently
early diagnosis. Only seven drugs currently hold FDA approval for
the treatment of HCC, including first-line choice Sorafenib (Fig. 1),
which unfortunately suffers from poor selectivity and provides only
modest improvements to median patient survival i.e. 3 months in
clinical trials [26,27]. There is therefore a pressing need for the
development of improved chemotherapeutic agents for the treat-
ment of HCC.
Our research team’s establishment of a liver progenitor cell
(LPC) line designated BMOL (for Bipotential Mouse Oval Liver)
provides a unique platform for compound evaluation [22,28]. The
ability for BMOL cells to differentiate into both cholangiocytes and
hepatocytes in culture confirms their progenitor cell status by
function. In the context of this communication, their association
with liver pathologies and cancer is significant: reports suggest
some forms of HCC are derived from LPCs, and it is further claimed
some LPCs may function as liver cancer stem cells [29e33].
Extensive passage of some of these cell lines, including BMOL cells,
generates tumorigenic (T) cell lines from what were originally non-
tumorigenic (NT) [22,28]. This methodology provides a novel
2.2. Synthesis
As a preliminary focus for our structure modifications, we pre-
pared thalidomide derivatives functionalised at the C4 position
with a range of atoms including C, N, S and O. As we have
demonstrated previously, a library of thalidomide analogues could
be obtained through cross-coupling techniques from appropriately
halogenated thalidomide precursors (Scheme 1) [34]. Thus, readily
available 2,3-dimethylaniline (1) was first converted efficiently to
3-iodophthalic anhydride (3), which is subsequently condensed
with the trifluoroacetic acid salt of aminoglutarimide to afford 4-
iodothalidomide 4a in 41% overall yield (4 steps) [40]. This pre-
cursor could be conveniently prepared on a multi-gram scale and
served as a key intermediate to several classes of thalidomide an-
alogues. Similarly, application of this condensation to the
Fig. 2. Rationale of compound design, including three sites of structural optimisation
from our previously identified active thalidomide derivative [36].
2

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
The C4-benzyl functionalised analogue 5e was accessed through
a
Suzuki-Miyaura cross-coupling between benzylboronic acid
pinacol ester and 4a in good yield (67%, Scheme 1). This approach
was not amenable for the equivalent C4-ether-bridged isostere 5g
proved unobtainable through both palladium cross-coupling and
copper-catalysed Ullmann-type approaches. Instead, attaching the
desired phenoxy-substituent onto the phthalimide scaffold was
completed firstly through a SNAr reaction between 3-fluorophthalic
anhydride (4c) and phenol to furnish ether 4d (Scheme 1) [41],
which was subsequently reacted with the trifluoroacetate salt of
glutarimide to provide ether 5g (28% yield over two steps, Scheme
1).
The related C4-sulfur linked derivative was prepared in excel-
lent yield (94%) through a C-S cross-coupling reaction of iodotha-
lidomide
4 and thiophenol, using the catalyst system of
Pd2(dba)3ꢀCHCl3 and Xantphos (Scheme 1) [42]. Thioether-
containing derivatives of thalidomide are especially rare in the
literature, where previously lenalidomide was subjected to a S-
transfer reaction [43] and more recently via a photocatalysed
approach [44]. This work is the first report of Pd-catalysed C-S
cross-coupling to afford thalidomide thioethers, and complements
the other recent approaches developed elsewhere. Initially, as the
para-isopropyl amine 5b was considered a starting point for this
study [22], each of the synthetic procedures outlined was also
repeated to afford the para-isopropyl derivatives 5b, 5f, 5h and 5j.
In the case of ispropyl-bearing thioether 5j treatment under strong
oxidative conditions (m-CPBA) also provided the corresponding
desired sulfonyl analogue (5k) in 58% (Scheme 1).
The versatile nature of this outlined amination process allowed
for the coupling of various electronically diverse anilines to func-
tionalised thalidomide analogues in yields ranging from 15% to 98%
(Scheme 2). A further 15 analogues were obtained for biological
study with this approach, featuring both electron-rich (e.g. alkoxy
and dimethylamino substituents) and electron deficient (e.g. tri-
fluoromethyl, acetyl and ethyl ester motifs) para-substitued ex-
amples. The general coupling conditions could also be carried out
with ortho-aminopyridine, ortho-toluidine and 2-aminotetralin to
furnish compounds 7a (90%), 7b (58%) and 7h (72%) respectively.
In the case of the C-S cross-coupling, a diverse set of 14
thioether-linked analogues of thalidomide were prepared in yields
ranging from fair to near-quantitative (Scheme 3). These reactions
produced thioether-bearing derivatives of thalidomide including
electron-donating tert-butyl (7f, 97%) and methoxy (7i, 72%) sub-
stituents, and those bearing an electron-withdrawing substituent
such as CF3 (7g, 70%). It was suspected that the low yield of thio-
anisole 7j (38%) resulted from undesired Pd-catalysed C-S cleavage
of its methyl sulfide group, a process which has also been reported
previously [45]. Interestingly, more diverse thioethers could be
produced under these robust cross coupling conditions such as
those thalidomide analogues bearing N-heterocycles (7a-b and 7h)
and linear (7l) or bulky (8m-o) alkyl variants. Overall, this approach
was deemed an efficient and high-yielding approach to thioether-
linked derivatives of thalidomide, which complements other
existing approaches that have recently been disclosed [43,46]. In
addition to these cross-coupling products, a late-stage functional-
isation of compound 7f was performed, by treatment with methyl
iodide, to give N-methylated analogue 7p in excellent yield
(Scheme 4).
Scheme 1. Reagents and Conditions: a) NaNO2, HCl, ꢁ5 ꢂC then KI, 62%; b) KMnO4, H2O/
pyridine, reflux, 24 h, 87%; c) Ac2O, reflux, 22 h, 79%; d) glutarimide trifluoracetate,
NEt3, THF, reflux, 72 h, 73%; e) aniline (1.5 equiv), Pd2(dba)3ꢀCHCl3 (4 mol%), XPhos
(16 mol%), K2CO3 (2 equiv), 1,4-dioxane, reflux, 18 h; f) aniline (1.5 equiv),
Pd2(dba)3ꢀCHCl3 (2 mol%), XPhos (8 mol%), K2CO3 (2 equiv), 1,4-dioxane, reflux, 18 h; g)
Pd2(dba)3ꢀCHCl3 (8%) PPh3 (1 equiv), Ag2O, THF, 70 ꢂC, 24h; h) phenol, KF, DMF; i)
glutarimide trifluoroacetate, NEt3, THF, reflux, 72 h,; j) thiophenol, iPr2NEt (2 equiv),
Pd2(dba)3ꢀCHCl3 (2.5 mol%), Xantphos (5 mol%), 1,4-dioxane (2 mL), reflux, 14 h; k) m-
CPBA (3 equiv), DCM, rt, 24 h, 58%.
commercially available brominated anhydride 3b afforded 5-
bromothalidomide 4b in 68% yield (Scheme 1). Given the success
of the commercial thalidomide derivatives lenalidomide and
pomalidomide (Fig. 1) and recent results within ours and other
groups, a synthetic pathway for access to C4 nitrogen linked com-
pounds was required. Fortunately, robust in-house conditions
previously optimized for Buchwald-Hartwig aminations of our
halogenated thalidomide precursors could be applied in these ex-
amples [36]. Using the precatalyst Pd2(dba)3.CHCl3 and the
biphenyl ligand XPhos, a coupling between iodothalidomide 4a and
aniline to 5a could be achieved 97% yield (Scheme 1). Coupling of
bromothalidomide 4b in an analogous fashion required only half
the catalytic loading to give 5c in near-quantitative yield (99%).
2.3. Selective targeting of tumorigenic cell growth
To identify agents with selective antiproliferative effects in liver
cancer cells, we screened for growth inhibition in a matched pair of
BMOL-T/BMOL-NT cell lines established from the same lineage. The
Incucyte Zoom (Essen BioScience) live-cell analysis system was
3

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
Scheme 2. General Reagents and Conditions: a) aniline (1.5 equiv), Pd2(dba)3ꢀCHCl3 (4 mol%), XPhos (16 mol%), K2CO3 (2 equiv), 1,4-dioxane, reflux, 18 h, 15e99%.
Scheme 3. General Reagents and Conditions: a) generic thiophenol or thiol, iPr2NEt (2 equiv), Pd2(dba)3ꢀCHCl3 (2.5 mol%), XantPhos (5 mol%), dioxane (2 mL), reflux, 14 h, 38e99%.
used to monitor cell growth in real-time. In contrast to traditional
end-point assessments of proliferation (such as the more common
MTT assay), this platform offers dynamic real-time measurements
of cell growth across the entire course of an experiment. The
doubling time of each cell population during treatment can then be
determined specifically during the log phase of cell growth. We had
previously established selective antiproliferative activity of 5b in
Scheme 4. Reagents and Conditions: a) MeI (1.2 equiv), K2CO3 (2 equiv), DMF, 0 ꢂC e rt,
18 h, 95%.
these cell lines at a 10
mM concentration [22]. This extension of the
previous study seeks to identify novel derivatives of 5b with
4

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
Table 1
Differential effects of several commercial drugs on growth inhibition of BMOL-NTand BMOL-Tcells. Growth is reported as mean doubling times in hours (n ¼ 3 in all cases, with
SEM in parentheses), control represents 0.1% DMSO vehicle in supplemented growth medium. Statistical significance is indicated as follows: * ¼ p < 0.05, ** ¼ p < 0.01,
*** ¼ p < 0.001.
BMOL-NT
BMOL T
Treatment
Growth (SEM)
Change Over Control
p
Growth (SEM)
Change Over Control
p
Control
18 (0.4)
18 (0.3)
18 (0.2)
18 (0.6)
50 (8.2)
e
NS
NS
NS
NS
***
15 (0.3)
15 (0.6)
15 (0.5)
15 (0.5)
71 (7.6)
e
NS
NS
NS
NS
***
Thalidomide
Pomalidomide
Lenalidomide
Sorafenib
e
e
e
e
e
e
181%
382%
enhanced efficacy at this dose, while maintaining selectivity for the
BMOL-T cell line. Concurrently we sought to elucidate SARs which
could be used as tools to predict this selectivity, to better inform
drug design approaches. Structural modifications to the aromatic
phthalimide ring have led to improved multiple myeloma activity
through initial CRBN binding and further interaction with the
IKAROS family of lymphocyte transcription factors (IKZF1/3) [47].
Thus, we hoped to determine if variation in aromatic substitution
could result in an improvement in activity in the BMOL cell models.
To first establish a baseline response in the BMOL cell lines used
role in our compounds’ efficacy. This tolerance of various bridging
atoms was not without limits, however: both ethers (5g-h) and
sulfonyl (5k) derivatives failed to elicit growth inhibition in either
liver cell line. This, taken together with the apparent requirement of
iPr-substitution for selective growth inhibition may indicate a
general preference for compounds with higher lipophilicity.
Notably, moving substituent to the adjacent C5 position and
i
removal of the Pr-moiety both ablated antiproliferative activity
(compound 5a) e a result mirrored in both benzyl and thioether
derivatives. Thus, this early study revealed examples of benzyl,
amine, and sulfide derivatives of thalidomide (5f, 5b, and 5j
respectively), each differing in their H-bonding donor and acceptor
capacities, were found to be selective BMOL-T growth inhibitors.
Furthermore, exchange of these successful linkers with either sul-
fonyl or ether functionalities was not tolerated, and the para-iso-
propyl aryl group was deemed essential for more potent cell
growth inhibition. Each of these factors was next examined care-
fully through a more thorough SAR investigation.
A range of compounds bearing modifications to the amino-
phenyl group were investigated for selective BMOL-T growth in-
hibition (Fig. 3). Having already highlighted the importance of the
isopropyl group contained within our previous lead, we sought to
further optimise and investigate the importance of this unique
substitution. Encouragingly, the only compound in this series to
selectively inhibit growth of the non-tumorigenic cells (BMOL-NT)
in this study, treatments of 10
domide, pomalidomide, or sorafenib (Fig. 1) were used in the
Incucyte proliferation assay (Table 1). At 10 M concentration, only
mM with either thalidomide, lenali-
m
sorafenib produced any statistically significant effect on the growth
of either cell line. While increasing doubling time for BMOL-T cells
(2.1-fold greater than that of the NT line) sorafenib nevertheless
demonstrated severe antiproliferative activity in BMOL-NT cells. In
contrast, at the same dose our previously described derivative 5b
inhibited BMOL-T growth without any concomitant effect on NTcell
growth [22]. Interestingly, thalidomide and its derivatives were
well-tolerated by both cell lines and were completely inactive as
antiproliferative agents at this dose. These findings are in accor-
dance with both evidence from clinical trials [48] and our own prior
work [22] in these cell lines, where no effect on BMOL cell growth
was observed at 10 mM, and at higher (100 mM) doses only indis-
criminate growth inhibition of both T and NT cell lines was
observed. This also held true for pomalidomide which to our
knowledge has never been assessed for inhibitory activity in liver
cancer. Furthermore, this data suggests that the structural modifi-
cations introduced in our thalidomide derivative 5b are required
for anticancer activity in these cell lines, and that the parent drug
does not share its selective activity against BMOL-T proliferation.
In the case of TNF inhibition, C4-amine substituted lenalidomide
is much more potent than its C5-amino analogue e a trend we also
identified in the TNF inhibition profiles of our amine-derived
compounds previously [36,49]. Nevertheless, given the aniline
derived 5b exhibited selective cell growth inhibition, the equivalent
C5-substituted compound was considered in this current study to
confirm the optimal position of substitution (Table 2). Additionally,
to examine any contribution of the bridging aniline nitrogen atom
to the observed efficacy and selectivity of 5b against HCC, we
sought isosteric and analogous compounds. Thus, derivatives
featuring methylene-, oxygen- and sulfur-bridged aryl groups were
preferentially was the 2-pyridyl compound 6a (10 mM). Introduc-
tion of a single methyl substituent alone to the phenyl ring was not
enough to induce the activity observed in 5b, and moreover the
position of substitution (ortho vs. para; 6b and 6c) had negligible
effect on compound potency or selectivity. Of the alkyl derived
compounds related to iPr-substituted compound 5b, five de-
rivatives (6d e 6h) demonstrated comparable or greater growth
inhibition towards tumourigenic cells at 10
mM. Chain length
appeared to be a crucial factor determining selectivity; both the
shorter ethyl chain of 6d and the larger 4-carbon substituents of
nBu (6g) and tetralin (6h) derivatives introduced modest inhibition
(up to 10%) of BMOL-NT proliferation. Negligible differences in ac-
i
tivity and selectivity were observed when comparing the Pr and
nPr analogues 5b and 6e, while the tBu derivative 6f was the most
potent compound of all analogues in this series. Treatment of the
BMOL cells with 6f produced 31% inhibition of BMOL-T growth,
with no concomitant decrease in BMOL-NT growth observed.
To determine whether the importance of alkyl substituent
choice stemmed from its impact on compound lipophilicity or was
instead governed by steric requirements for target binding, we
calculated several physicochemical parameters of all novel com-
pounds including LogP (see supporting information for these data).
The calculated logP (cLogP) values of alkyl-substituted compounds
6c e 6g were in the range 1.56e3.45 (see supporting information
for a summary of physicochemical parameters). High cLogP alone
was not a reliable predictor of efficacy or selectivity however; nBu-
derivative 6g (cLogP ¼ 3.45) was less selective and no more potent
also subjected to our cell growth assays at a 10 mM dose (Table 2).
Remarkably, the para-isopropyl examples of benzyl, thioether and
aniline derivatives (5b, 5f and 5j) all demonstrated near-identical
levels of cancer growth inhibition, and all three remained selec-
tive towards the BMOL-T line. Each of these functionalities offer a
unique set of hydrogen bonding capacities (5b as acceptor and
donor, 5i as acceptor only, and 5e with none at all), and the ho-
mogeneity of their responses in BMOL cells indicates that hydrogen
bonding between this site and a potential cellular target plays no
5

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
Table 2
Differential effects of thalidomide derivatives on growth inhibition of BMOL-NT and BMOL-T cells. Growth is reported as mean doubling times in hours (n ¼ 3 in all cases, with
SEM in parentheses), control represents 0.1% DMSO vehicle in supplemented growth medium. Statistical significance is indicated as follows: * ¼ p < 0.05, ** ¼ p < 0.01,
*** ¼ p < 0.001..
Ring Position
Y
R
BMOL-NT
BMOL-T
Compd.
Growth (SEM)
% Change
p
Growth (SEM)
% Change
p
5a
5b
5c
5d
5e
5f
5g
5h
5i
4
4
5
5
4
4
4
4
4
4
4
N
N
N
N
C
H
iPr
H
iPr
H
iPr
H
iPr
H
iPr
iPr
17.4 (0.4)
17.3 (0.4)
17.4 (0.1)
17.2 (0.4)
17.7 (0.9)
17.5 (0.3)
16.9 (0.6)
17.5 (0.6)
16.8 (0.4)
17.2 (0.8)
17.4 (0.8)
-
-
-
-
2%
-
-
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
14.3 (0.4)
16.3 (0.7)
14.7 (0.7)
14.1 (0.3)
13.8 (0.5)
17.0 (0.6)
14.3 (0.2)
15.4 (0.3)
13.8 (0.5)
16.2 (1.1)
14.9 (0.7)
7%
19%
4%
3%
-
17%
6%
5%
2%
19%
3%
NS
***
NS
NS
NS
***
NS
NS
NS
**
C
O
O
S
S
SO2
-
1%
3%
3%
5j
5k
NS
Fig. 3. Aniline derived thalidomide derivatives and BMOL-T and BMOL-NT growth inhibition. Values reported as percentage change in mean doubling time over control for each cell
line. Error bars represent SEM of three independent experiments. * ¼ p < 0.05, ** ¼ p < 0.01, *** ¼ p < 0.001.
6

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
than the isopropyl equivalent 5b (cLogP ¼ 2.93). These data imply
that modulation of the overall compound lipophilicity is not wholly
responsible for the observed selectivity and activity in anti-
proliferative actions of our alkyl-subsituted compounds against the
BMOL line. Aniline derived thalidomide analogues with para-polar
groups were designed to confirm this theory, and to probe whether
H-bond acceptor/donors in this position could influence potency.
Important to note was compounds bearing OH, OMe, OiPr, and
NMe2 groups were all inhibitors of BMOL-T growth and did not
significantly affect BMOL-NT cells, though they proved less prom-
ising in selectivity and potency that the alkyl derivatives described
above. It would appear from these results that H-bond donors or
acceptors at this position are tolerated but do not improve activity.
Diaryl amine thalidomide derivatives with carbonyl-containing
moieties in this position were not effective in inhibiting cell
growth. Promisingly, from this study seven new analogues (6e-f
and 6h-l) were identified as selective growth inhibitors of tumor-
igenic liver progenitor cells.
demonstrated remarkable selective activity, with 36% growth in-
hibition of the BMOL-T cell lines and no significant inhibition of the
NT cell line. Importantly, 7f has a 9-fold factor of selectivity for
BMOL-T cells (compared to 2.1-fold for sorafenib as described in
Table 1) and was the most potent compound identified in this study
at the 10 mM dose. More polar isosteres of 7f for example those
featuring a p-CF3 group (7g-h) failed to elicit any growth inhibition.
Unfortunately, our set of aliphatic thioethers were also all poor
inhibitors of growth, and no predictive trend was found between
their lipophilicity and BMOL cell response (cLogP values ranged
from 1.05 to 2.78). Interestingly, pyrimidine and quinoline derived
compounds 7a-b produced no measurable effect on BMOL-NT
growth, unlike pyridyl aniline 6a above. This would suggest het-
erocycles may be incorporated into thalidomide thioether de-
rivatives to further modulate drug properties such as solubility,
without compromising overall selectivity in HCC.
t
Having identified Bu-functionalised 6f and 7f as the most se-
lective and potent in the class of amine and sulfide derivatives
respectively, we set out to confirm that their effects were dose-
dependent and identify the lowest active dose for each (Fig. 5).
Following the improved growth inhibition displayed by com-
pound 6f a series of thioether-containing derivatives were also
investigated (Fig. 4). Overall, our thioether compounds showed less
promise than the above aniline derived compounds. Similar pro-
files of activity were observed in those compounds with an aniline
derived derivative bearing identical substituents (7c, 7e, 7f-g, and
7i) apart from anisole 7i which was not consistently tolerated by
BMOL-NT cells at this dose and exhibited some indiscriminate
growth inhibition. The promising selectivity and potency of iPr and
tBu substituted 7e-f however proves that thioether motifs can serve
Thioether 7f was active at doses as low as 6
rivative 6f remained active at the lowest tested concentration of
M; half that of the lead compound 5b in our previous study [22].
mМ, while amine de-
4
m
From this data we projected that at high enough doses, both
compounds would begin to show the undesired inhibition of
BMOL-NT cells and diminished selectivity. As expected, a clear
dose-dependent increase in growth inhibition of BMOL-T cells was
observed for both compounds. Furthermore, inhibition of non-
t
equally as well as our amine candidates (Fig. 3). Bu-thioether 7f
tumorigenic cells occurred at all doses greater than 10 mM for
Fig. 4. Sulfide derived thalidomide derivatives and BMOL-T and BMOL-NT growth inhibition. Values reported as percentage change in mean doubling time over control for each cell
line. Error bars represent SEM of at least three independent experiments. * ¼ p < 0.05, ** ¼ p < 0.01, *** ¼ p < 0.001.
7

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
Fig. 5. Effect of dose of lead compounds 6f (A) and 7f (B) on BMOL-NT and BMOL-T growth. Values reported as mean % change in doubling time over control (n ¼ 3 in all cases), error
bars represent SEM. * ¼ p < 0.05, ** ¼ p < 0.01, *** ¼ p < 0.001.
Table 3
both compounds
dependent for sulfide 7f which remained essentially constant be-
tween doses of 14e26 M. While amine 6f was undoubtedly the
more potent compound in BMOL-T cells, its reduced selectivity
from 23-fold preference for BMOL-T at 10 M, to 3-fold at the
e however, this inhibition was not dose-
Effect of N-methylation on activity and selectivity of compound 7e against BMOL cell
growth. Growth is reported as mean doubling times in hours (n ¼ 3 in all cases, with
SEM in parentheses), control represents 0.1% DMSO vehicle in supplemented growth
medium. Statistical significance is indicated as follows: * ¼ p < 0.05, ** ¼ p < 0.01,
*** ¼ p < 0.001.
m
m
highest tested concentration, renders it less promising than its
sulfur analogue 7f, which retains an excellent 9-fold selectivity at
the higher concentration of 26 mM. It is noteworthy that in some
BMOL-NT
BMOL-T
Compd.
Growth (SEM) % Change
p
Growth (SEM) % Change
p
Control 19.0 (0.5)
-
e
14.3 (0.1)
NS 20.1 (2.4)
** 16.3 (0.2)
-
e
cases, very small changes to the pendant substituent can give rise to
modified biological activities. For example, 6f bearing two addi-
tional methyl groups is more selective for BMOL-T cells growth
inhibition than 6d. This is also observed in the case of 7f when
compared to compound 7i however, this change in selectivity could
be attributed to the introduction of a heteroatom. This would imply
that these substituents may play some critical role in target bind-
ing, however overall as illustrated in Figs. 3 and 4, there seems to be
an overall trend for BMOL-T growth inhibition selectivity.
The modern paradigm surrounding CELMoDs and their de-
rivatives implies that their pleiotropic actions stem from the
recruitment and ubiquitination of unique protein neosubstrates to
the CRL4-CRBN complex; we inferred that our related compounds
might also elicit their HCC anticancer actions through this same
7e
7p
19.3 (0.6)
24 (2.5)
þ2%
þ51%
***
*
þ26%
þ14%
mode of action. It has previously been shown that methylation of a
CELMoD s glutarimide nitrogen abrogates cereblon binding, thus
0
reducing induced protein degradation (Fig. 6) [13,47,50,51].
We therefore investigated the effect of N-methylation of our
most potent selective derivative 7f (Table 3). This comparatively
small structural modification had a dramatic effect on both cell
proliferation inhibition potency and selectivity; decreasing activity
against tumorigenic cells by 73%, while also causing significant
inhibition of the non-tumorigenic line that was not previously
observed. Methylation appeared to invert selectivity, from a strong
(26-fold) preference for inhibition of BMOL-T growth, to a 2-fold
preference for the non-tumorigenic line. This would strongly sug-
gest that the imide proton is an essential motif for selective anti-
cancer activity, presumably serving as a key hydrogen bond donor
during binding. These data provide initial evidence supporting the
hypothesis that our new thalidomide derivatives share CRBN as a
binding target with the rest of the CELMoD class of drugs.
2.4. Molecular dynamics simulations of CRBN binding
When considering the potential role of CBRN as a target for the
resulting pharmacology of our compounds, it should be noted that
despite high residue conservation between species, CELMoD
binding to rodent CRBN does not induce the degradation of IKZF1/3
or CK1a seen in humans [52]. Mutation studies have identified a
single variant residue in the ligand binding domain of CRBN
responsible for this insensitivity; human valine 387 is replaced
with isoleucine in the matching position for rodent CRBN (residue
Fig. 6. Examples from the literature of N-methylation at glutarimide as a method of
inhibiting CRBN binding, and the structure of our N-methyl derivative 7p [47,51].
8

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
Fig. 8. Effect of compound treatment on BMOL-T migration and wound healing assay.
(n ¼ 3 in all cases), error bars represent SEM. DMSO vehicle (0.1% in supplemented
growth medium) served as a control. For associated statistical analyses of these data,
see supporting information.
strong response in murine liver cells elicited by our derivatives is
therefore especially interesting, as it leaves no role for any known
CELMoD neosubstrates that are not ubiquinated by murine CRBN
(such as IKZF1/3 or CK1a) in their mechanism of action. Moreover,
it suggests that any neosubstrate(s) recruited as a result of our
derivatives binding to CRBN may be novel, requiring different in-
teractions between protein, ligand and substrate.
2.5. Cell migration assays
The anti-angiogenic effect of CELMoDs has previously been
attributed to halting of cell motility via inhibiting phosphorylation
of targets including STAT3 and Akt. As these pathways have also
been highlighted in the migration and invasion of HCC, we pro-
posed that our compounds might possess anti-migratory effects in
our unique tumorigenic liver progenitor cells. We therefore con-
ducted a series of wound healing assays in the BMOL-T line with the
Incucyte Zoom live cell imaging system to investigate any influence
on cancer cell migration (Fig. 8). Remarkably, compounds 6f and 7f
produced a pronounced and sustained inhibition of BMOL-T cell
motility over a 24-h period. In both cases, cell migration appeared
to halt completely by 16 h, displaying mean reductions in wound
closure of 94% (6f) and 80% (7f) over control respectively. In com-
parison, treatment with lenalidomide inhibited migration by just
Fig. 7. Binding poses of Thalidomide, 6f and 7f to murine CBRN from MD simulations.
(A) Representative snapshots of the most likely binding pose of each compound is
shown as found from replicate 100ns simulations. Residues are coloured by amino acid
type while ligands are coloured by atom name. Three commonly occupied poses of 6f
are shown. The protruding sec-butyl sidechain of isoleucine 390 is highlighted in
green. (B) Average interaction energy between each compound and each protein
residue highlighting the most significant residues for ligand binding. Shaded regions
show the standard deviation across the three replicate simulations.
390) [53]. To better understand the role of this residue, we per-
formed three replicate MD simulations with each of thalidomide, 6f
and 7f in the binding site. Thalidomide sits centrally in the binding
pocket formed by multiple tryptophan residues as depicted in
Fig. 7A, with two hydrogen bonds between the glutarimide ring and
the protein backbone (in accordance with those seen in the original
crystal structure). The long side chains of 6f and 7f, however, allow
for a range of additional interactions with the protein shown in
Fig. 7A. To better quantify these, we determine the average inter-
action energy between the ligand each protein residue across the
duration of 3 replicate 100ns simulations as shown in Fig. 7B. This
shows that all compounds have the strongest association with
W382, W388, W402 as well as additional interactions with H380
and H399. Interestingly, 6f is seen to be relatively mobile, with the
side chain interchanging between three primary conformations
pointing toward either H380, W382 or H399. Despite having bulky
sidechains, 6f and 7f have only slightly greater interactions with the
protein than the parent compound, and would therefore both likely
have comparable binding affinities to CRBN.
20% at the equivalent dose (10 mM), and statistical analyses of the
wound closure rate of each treatment failed to find a significant
difference over control in the lenalidomide treatment. The dis-
covery of anti-migratory properties in our derivatives is especially
important when considering that the liver is the most common site
of metastatic burden e fibrosis and myeloid cell accumulation often
facilitate seeding and development of invasive tumour cells, which
in HCC may promote intrahepatic metastasis [54,55]. Moreover, the
development of extrahepatic metastatic disease as a complication
of HCC would contraindicate surgical intervention, and so drug
candidates demonstrating potentially anti-metastatic effects such
as ours are especially promising.
2.6. Cell signalling: quantifying levels and cellular localisation of
phosphoproteins
Having identified novel thalidomide derivatives with anti-
proliferative and anti-migratory effects in tumorigenic liver cells,
we sought to identify affected signalling pathways that might be
involved in the respective cellular activities. Pathways that are
known to be important in HCC development, growth, and/or in-
vasion were prioritised, especially where these pathways had
formally been implicated as a target of CELMoD therapy in other
malignant cell lines. Based on these criteria the PI3k/Akt/mTOR, IL-
While I390 does not have significant interactions with any of the
simulated compounds, it does sit directly beside them and is
particularly close to the sidechain of 7f. It has been proposed that
the added bulk of the introduced isobutyl sidechain in rodent
versus human CBRN prevents substrate interactions with either the
CELMoD or other residues peripheral to the binding pocket. The
9

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
6/JAK/STAT3, Ras/Raf/Mek/Erk and canonical NF
k
B (p65) signalling
potency: 7f > 6f > lenalidomide. Compound 7f also reduced mean
nuclear localisation of p65 by 23%, suggesting a concomitant
reduction of downstream transcription events in the canonical
axes were selected for further study [56e59]. As a rapid pre-
liminary screen, both BMOL lines were subjected to immunofluo-
rescence assays of key downstream markers of these pathways
using the CellInsight CX7 high-content screening platform (Ther-
moFisher). Individual fluorescence values per cell were measured,
offering mean values representative of very large technical repli-
cates (n ¼ 1400e4500, depending on cell line) within each exper-
iment. Furthermore, the ratio of signal originating from the nucleus
vs. the mean whole-cell total was calculated based upon cross-
staining of the nucleus, serving as a measure of treatment influ-
ence on nuclear localisation of a target protein. Additionally, the
percentage of a cell population positive for the target protein is also
quantified using this technique.
NFkB pathway. A similar profile of activity was observed in the
immunofluorescence assay of p-STAT3, where lenalidomide and 6f
produced mild and inconsistent inhibition. Again, 7f achieved sig-
nificant selective suppression of both phosphorylated STAT3 levels,
and their nuclear localisation. Many of STAT3’s target genes have
pro-oncogenic roles are implicated in tumour cell proliferation,
survival and invasiveness [59]. Furthermore, STAT3 also inhibits the
related STAT1 transcription factor which possesses tumour sup-
pressive, pro-apoptotic effects [60]. Inhibition of STAT3 therefore is
a prime mechanistic candidate that accounts for the actions of both
thalidomide derivatives 6f and 7f in liver progenitor cells.
BMOL cells treated with either DMSO vehicle, lenalidomide, or
target compounds 6f and 7f were subjected to immunofluorescent
assays of the phosphorylated forms of canonical NFkB (p65), Akt,
2.6.1. Cell signalling: quantifying expression and phosphorylation of
STAT3 by AlphaLISA assay
STAT3 and Erk (Table 4). Unsurprisingly, in our murine model
lenalidomide was mostly ineffective in influencing protein levels or
their nuclear localisation. No compound tested consistently
affected the total cellular levels of p-Akt, but nuclear p-Akt was
supressed by lenalidomide and 7f. In this pathway thioether 7f was
slightly more potent than lenalidomide but displayed no selectivity
between cell lines. All three test compounds were able to selec-
A more rigorous approach to confirm the involvement of the
STAT3 pathway is to determine levels of both total STAT3 and pSTAT
(canonical phosphorylation at Y705) in BMOL-T cells (Fig. 9). Along
with the three previous test compounds 6f, 7f and lenalidomide, N-
methyl derivative 7p was also included to ascertain whether loss of
the glutarimide H-bond donor in 7f would have the same ablative
effect on STAT3 inhibition as it did on BMOL-T growth inhibition
previously. Lenalidomide failed to significantly affect mean total
STAT3, or STAT3 phosphorylation in this assay. While all three novel
derivatives 6f, 7f and 7p demonstrated lower mean values in STAT3
levels across triplicate experiments, inter-experimental variation of
cell responses was particularly high, and no compound produced
statistically significant changes to total STAT3 levels in BMOL-T
cells, implying no influence on STAT3 expression in BMOL-T cells.
Thioether 6e alone significantly inhibited STAT3 phosphorylation
(23% reduction over the control). Since N-methyl analogue 7p did
not share this activity it can be inferred the glutarimide proton is
essential for both growth inhibition and downregulating STAT3
activation. Interestingly, despite the close structural resemblance
and similar profiles of growth inhibition, we were not able to detect
any significant effect on STAT3 signalling through the alphaLISA
method. These results suggest that the nitrogen and sulfur-
containing analogues presented in this study may operate
through distinct mechanisms of action, despite their structural
similarities. In the case of 7f however, these data provide strong
evidence implicating a role for the IL-6/JAK/STAT3 signalling axis in
the biological activity of this compound in liver cancer.
tively reduce total levels of the phosphorylated form of NFkB (p65)
preferentially in tumorigenic cells, in the following order of
Table 4
Summary of immunofluorescent assays of BMOL cells screening for changes to
phosphorylated NFkB (p65), Akt, and STAT3. Values represent the mean of two in-
dividual experiments alongside standard deviation in parentheses. DMSO vehicle
(0.1% in supplemented growth medium) served as a control. Results deemed
indicative of selective activity in tumourigenic cells are highlighted in bold.
Mean Intensity
Mean Ratio (nuclear:total)
pNFkB (p65)
% Change (SD)
% Change (SD)
BMOL-NT
Control
Lenalidomide
6f
e
e
þ13% (14.8)
ꢁ1% (8.5)
0% (5.7)
ꢁ9% (2.1)
ꢁ11% (1.4)
þ11 (15.5)
7f
BMOL-T
Control
Lenalidomide
6f
e
e
ꢁ5% (3.5)
ꢁ7% (9.9)
¡13% (0.7)
þ7% (14.3)
ꢁ9% (8.0)
ꢁ23% (3.2)
7f
pAkt
Mean Intensity
Mean Ratio (nuclear:total)
% Change (SD)
% Change (SD)
3. Conclusion
BMOL-NT
Control
Lenalidomide
6f
e
e
Growth studies in our unique cell lines revealed several novel
thalidomide analogues that demonstrate selective antiproliferative
activity in only tumourigenic cells. Most analogues assessed were
also well tolerated by non-tumourigenic cells, and several demon-
strated comparable or improved selective antiproliferative activity
over our previously reported derivative 5b. Benzyl and aryl sulfide-
linked analogues of 5b (5f and 5j) retained selectivity and potency,
while the equivalent ether (5h) was inactive. Generally, short
branched alkyl substituents in the para position of an aryl ring per-
formed best in terms of both selectivity and potency, while de-
rivatives bearing alkyl thioether motifs or N-heterocyclic
substituents all performed poorly in growth assays. Two tert-butyl
substituted derivates (aryl amine 6f and aryl sulfide 7f) were the
most promising compounds assessed, demonstrating selective in-
þ10% (4.9)
þ4% (0.7)
0% (7.0)
ꢁ6% (1.9)
ꢁ10% (4.1)
ꢁ25% (4.7)
7f
BMOL-T
Control
Lenalidomide
6f
e
e
ꢁ9% (10.6)
þ4% (16.3)
ꢁ3% (25.5)
ꢁ16% (2.5)
þ4% (27.4)
ꢁ19% (1.2)
7f
pSTAT3
Mean Intensity (% Change)
Mean Ratio (nuclear:total)
% Change (SD)
BMOL-NT
Control
Lenalidomide
6f
e
e
þ2% (3.5)
ꢁ4% (0.7)
ꢁ5% (2.1)
ꢁ3% (11.2)
ꢁ13% (4.5)
ꢁ8% (11.8)
7f
BMOL-T
Control
Lenalidomide
6f
hibition of only BMOL-T cells at 10 mM (31% and 36% respectively).
e
e
ꢁ4% (3.5)
ꢁ7% (12.0)
¡22% (8.5)
þ3% (22.6)
ꢁ10% (7.5)
¡16% (15.6)
Both compounds also exhibited remarkable anti-migratory activity,
eliciting 94% (6f) and 80% (7f) inhibition of BMOL-T cell motility at
this dose during wound healing assays. Furthermore, preliminary
7f
10

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
(Greiner Bio-One) in 100
left to attach in the Incucyte instrument overnight. The following
day an additional 100 L of medium was added (either vehicle or
mL supplemented medium per well, and
m
containing a compound treatment as required; see below) and each
plate was imaged every 4 h for three days. Compound treatments
were performed in six replicate wells per cell type, for a total of 12
wells per treatment. The Incucyte instrument’s cell detection al-
gorithm was trained on a set of 20 representative images from both
cell lines at a range of confluence levels. The efficiency of cell
detection was checked manually by visual examination to ensure
only cells were detected. The log of cell confluency (measured as %
well coverage) was plotted against time to generate a growth curve.
The doubling time was determined from the linear portion of each
curve corresponding to the log (exponential) phase of cell growth.
4.3. Treatment of cells with thalidomide derivatives
Solutions of all tested compounds were prepared at 2ꢃ (20
concentrations, with 0.2% DMSO in supplemented medium. Cells
were treated with each compound by addition of 100 L of these
solutions, thereby achieving a final concentration of 1ꢃ and 0.1%
DMSO in a final well volume of 200 L.
mM)
m
m
4.4. Quantification and localisation of cell signalling proteins
Fig. 9. A) AlphaLISA assay measuring total cellular STAT3 in BMOL-T cells upon
treatment with various agents at 10 mM. Values represent mean normalised fluores-
Briefly, BMOL cells were cultured in Williams’ essential medium
(supplemented as described above) in 96-well plates and treated
with indicated compounds from Table 4 or 0.1% DMSO vehicle in
supplemented medium for 6 h. Cells were washed with PBS lx, and
affixed to plate using 4% PFA in phosphate buffered saline for
10 min; washed IX and permeabilized using 0.1% Triton-X-100 in
PBS for 8 min; washed Ix and blocked with protein blocker for
30 min. Cells were then incubated with the following primary an-
tibodies: anti-phos-p65, anti-phos-Akt, anti-phos-STAT3, (diluted
in DAKO diluent as per manufacturer’s recommendations) for 18 h.
Cells were washed 3ꢃ prior to adding secondary antibodies and
incubated for l hr, wash again, incubated with Hoescht reagent for
8 min. Cells were washed 2ꢃ and imaged/analyzed using the CX7
instrument with the HCS studio software package.
cence intensity (arbitrary units), error bars represent SEM (n ꢄ 3 in all cases). B)
AlphaLISA assay measuring ratio of total cellular STAT3 to p-STAT3 in BMOL-T cells
upon treatment with various agents at 10
mM. Values represent mean normalised
fluorescence intensity, error bars represent SEM (n ꢄ 3 in all cases). * ¼ p < 0.05,
** ¼ p < 0.01, *** ¼ p < 0.001. DMSO vehicle (0.1% in supplemented growth medium)
served as a control.
analyses of key signalling pathways revealed 7f inhibits STAT3
phosphorylation and selectively downregulates nuclear localisation
of pSTAT3 in BMOL cells. In summary, this work has highlighted that
modifications to the thalidomide core structure can generate novel
derivatives with promise as selective anticancer agents against HCC,
and that further study in animal models and ultimatelyclinical trials
is warranted to establish their potential as therapeutics.
4. Experimental
4.5. Cell migration assays
4.1. Cell maintenance
BMOL-T cells (20,000 cells per well) were plated in 96-well
plates (Greiner Bio-One) in 100
well, and incubated for 24 h. Wounds were made in wells using the
Woundmaker™ device, then medium was aspirated and replaced
with fresh supplemented medium (100
treated, DMSO vehicle, or containing compound treats as required).
The plate was placed in an IncucyteZOOM system and imaged every
2h for 24 h. Images were processed with the Incucyte software to
obtain measurements of average wound width in micrometres,
which were further processed using the Graphpad Prism 5 software
package.
mL supplemented medium per
BMOL-NT and BMOL-T cells were maintained as previously
described by Woo et al. in 25 cm2 flasks (Corning), with 3 mL of
Williams’ E Medium (WEM) supplemented with; 675
Streptomycin (Life Technologies), 1% v/v Fungizone (Life Technol-
ogies), 48.4 g/mL Penicillin (Calbiochem), 20 ng/mL EGF (Corning),
mg/mL
mL per well; either un-
m
30 ng/mL IGF-II (GroPrep Bioreagents), 0.25 U/mL Humulin R (Eli
Lilly and Company), 2 mM Glutamine (Sigma-Aldrich) and 5% Fetal
Bovine Serum (FBS, Life Technologies). Medium was replaced every
2 days (3 days over the weekend with an additional 2 mL of me-
dium). Once cells reached approximately 70% confluence they were
passaged into new flasks. Cells were detached by rinsing with 1 mL
of Hank’s Balanced Salt Solution (HBSS, Sigma-Aldrich), then 1 mL
of 0.05% Trypsin (Life Technologies) in HBSS, followed by a 5 min
incubation at 37 ꢂC. After cells had rounded and detached, they
were resuspended in 1 mL of medium and passaged as required.
4.6. Statistical analysis
Where appropriate, statistical significance was determined via
one-way analysis of variance (ANOVA) with Dunnett’s test post-hoc
analysis, using the Graphpad Prism 5 software package. p values
below 0.05 were considered statistically significant.
4.2. Growth assay protocol
4.7. Chemistry. General experimental
Proliferation assays were performed as follows, modified from a
procedure described by Woo et al. BMOL cells (BMOL-T: 1000 per
well, BMOL-NT: 1500 per well) were plated into 96-well plates
Starting materials and reagents were available from Merck,
Sigma-Aldrich, Alfa Aesar, TCI, Precious Metals Online, Fluorochem,
11

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
Oakwood Chemical, Fluka, Boron Molecular, and used without
further purification unless otherwise stated. The Pd2(dba)3ꢀCHCl3
used in this work was synthesized according to literature proced-
ures [61]. Halogenated thalidomide derivatives 4a-b and their
precursors (compounds 2 and 3a) were synthesized as previously
described [35]. Compounds 5a-d, 6a-f, 6i-j, and 6n-o were syn-
thesized as previously described [36]. The abbreviated ligand name
XPhos refers to 2-dicyclohexylphosphino-20,40,60-triisopropylbi-
167.5, 143.8, 138.9, 137.3, 136.6, 132.9, 129.7, 122.9, 119.4, 113.4, 111.8,
49.1, 34.7, 33.6, 31.4, 22.6, 22.2, 14.2; IR (neat) 3324 (N-H) 3205 (N-
H), 1757 (C¼O), 1689 (C¼O), 1619 (C¼O), 1401, 1320 cmꢁ1; HRMS
calculated for C23H24N3O4: 406.1767, found: 406.1770 [MþH]þ.
4.8.2. 2-(2,6-Dioxopiperidin-3-yl)-4-(5,6,7,8-
tetrahydronaphthalen-2-ylamino)isoindole-1,3-dione (6h)
Following the general procedure for Buchwald-Hartwig ami-
nation reactions, using the coupling partner 5,6,7,8-tetrahydro-2-
naphthylamine (60 mg, 0.41 mmol), the title compound was ob-
tained after purification by flash column chromatography (20% /
50% ethyl acetate/hexanes) as a yellow solid (76 mg, 72%). Rf ¼ 0.53
(40% ethyl acetate/hexanes); mp ¼ 208e210 ꢂC; 1H NMR (400 MHz,
phenyl,
and
XantPhos
likewise
refers
to
4,5-
bis(diphenylphosphino)-9,9-dimethylxanthene. Anhydrous sol-
vents were distilled over the appropriate drying agents according
to Amarego and Chai [62], acquired from a Pure Solv 5-Mid Solvent
Purification System (Innovative Technology Inc.) and degassed
(argon) prior to use. Hexanes and ethyl acetate for chromatography
were distilled prior to use. 1H, 13C, and 19F Nuclear Magnetic
Resonance (NMR) spectra were acquired on a Varian 300, Varian
400, Bruker AV500 or Bruker V600 spectrometer at 25 ꢂC. All sig-
DMSO‑d6)
d
11.12 (s, 1H), 8.26 (s, 1H), 7.58 (dd, J ¼ 8.6, 7.1 Hz, 1H),
7.34 (d, J ¼ 8.5 Hz, 1H), 7.20 (dd, J ¼ 7.1, 0.6 Hz, 1H), 7.10e6.98 (m,
2H), 5.11 (dd, J ¼ 12.9, 5.4 Hz, 1H), 2.90 (ddd, J ¼ 17.2, 14.0, 5.4 Hz,
1H), 2.71 (d, J ¼ 4.0 Hz, 4H), 2.66e2.52 (m, 2H), 2.08e2.02 (m, 1H),
nals
d
are reported in parts per million (ppm). Chemical shifts in 1H
1.74 (quint., J ¼ 3.0 Hz, 4H); 13C NMR (101 MHz, DMSO‑d6)
d 172.8,
and 13C spectra were referenced to the residual (partially) non-
deuterated solvent according to Fulmer et al.63 19F chemical shifts
were referenced to the external standard hexafluorobeznene
(ꢁ164.9 ppm). Wherever compound solubility allowed, CDCl3 was
used to acquire spectra preferentially over DMSO‑d6. Mass spectra
were recorded on a Waters LCT Premier XE instrument (time-of-
flight, electrospray ionisation in positive mode). Infrared spectra
were acquired on a PerkinElmer Spectrum One spectrometer at
2 cmꢁ1 resolution equipped with an ATR diamond. Melting points
were recorded on a Reichart heated-stage microscope. The reported
retention factors (Rf) were acquired via thin layer chromatography
(TLC) performed on Merck silica gel 60 F254 pre-coated aluminium
sheets. Column chromatography was performed using Davisil
170.0, 168.4, 167.09, 143.5, 137.8, 136.4, 136.2, 132.9, 132.4, 129.8,
122.8, 120.0, 119.1, 112.9, 111.2, 48.7, 31.0, 28.8, 28.3, 22.8, 22.6, 22.1;
IR (neat) 3355 (N-H), 3239 (N-H), 3103, 1761 (C¼O), 1690 (C¼O),
1620 (C¼O), 1604 (C¼O), 1476, 1403, 1321 cmꢁ1; HRMS calculated
for C23H22N3O4: 404.1610, found: 404.1615 [MþH]þ.
4.8.3. 2-(2,6-Dioxopiperidin-3-yl)-4-[(4-isopropoxyphenyl)amino]
isoindole-1,3-dione (6k)
Following the general procedure for Buchwald-Hartwig ami-
nation reactions, using the coupling partner 4-isopropoxyaniline
(52 mg, 0.24 mmol), the title compound was obtained after puri-
fication by flash column chromatography (40% ethyl acetate/hex-
ane) as an orange solid (35 mg, 54%). Rf ¼ 0.36 (40% ethyl acetate/
chromatographic silica gel (40e63
mm) supplied by Adelab. Pre-
hexane); mp ¼ 212e214 ꢂC; 1H NMR (400 MHz, DMSO‑d6)
d 11.12
parative thin layer chromatography (PTLC) was performed with in-
house prepared plates (20 ꢃ 20 ꢃ 0.1 cm) using silica gel 60 PF254
containing gypsum supplied by Merck. The visualisation of the
developed plates was achieved using a 254-nm or 365-nm UV
lamp.
(s, 1H), 8.22 (s, 1H), 7.55 (dd, J ¼ 8.5, 7.2 Hz, 1H), 7.30e7.12 (m, 4H),
6.96 (d, J ¼ 8.9 Hz, 2H), 5.11 (dd, J ¼ 12.9 & 5.4 Hz, 1H), 4.60 (hept,
J ¼ 6.4 Hz, 1H), 2.90 (ddd, J ¼ 18.0, 13.8 & 5.5 Hz, 1H), 2.66e2.53 (m,
2H), 1.99e2.07 (m, 1H), 1.27 (d, J ¼ 6.0 Hz, 6H); 13C NMR (101 MHz,
DMSO‑d6)
d 173.3, 170.5, 168.9, 167.6, 155.2, 144.7, 136.6, 132.8,
132.0, 125.8, 119.0, 116.9, 112.9, 111.1, 69.9, 49.1, 31.4, 22.6, 22.3; IR
(neat) 3320 (N-H), 3229 (N-H), 1759 (C¼O), 1621 (C¼O), 1510,
743 cmꢁ1; HRMS calculated for C22H22N3O5: 408.1559, found:
408.1554 [MþH]þ.
4.8. General procedure for Buchwald-Hartwig amination reactions
To a flame-dried Schlenk flask was added the following in
sequence, under an inert atmosphere of argon: 4-iodothalidomide
(200 mg, 0.52 mmol), XPhos (40 mg, 0.083 mmol) Pd2(dba)3∙CHCl3
(22 mg, 0.021 mmol) K2CO3 (144 mg, 1.04 mmol) 1,4-dioxane
(2 mL) and The desired aniline (0.89 mmol). The reaction mixture
was stirred at 90 ꢂC for 18 h and the resulting mixture then cooled
to room temperature, diluted with ethyl acetate (10 mL) and
filtered through a pad of Celite. The filter cake was washed with
ethyl acetate (ca. 40 mL) and the resulting filtrate adsorbed to silica.
Purification by flash column chromatography (ethyl acetate/hex-
anes) gave the desired aminated thalidomide analogue.
4.8.4. 2-(2,6-Dioxopiperidin-3-yl)-4-{[4-(dimethylamino)phenyl]
amino}isoindole-1,3-dione (6l)
Following the general procedure for Buchwald-Hartwig ami-
nation
reactions,
using
N,N-dimethyl-p-phenylenediamine
(120 mg, 0.89 mmol), the title compound was obtained after pu-
rification by flash column chromatography (50% ethyl acetate/
hexanes) as a brown solid (35 mg, 15%). Rf ¼ 0.21 (50% ethyl ace-
tate/hexanes); 1H NMR (400 MHz, DMSO‑d6)
d 11.11 (s, 1H), 8.10 (s,
1H), 7.62e7.39 (m, 1H), 7.22e6.96 (m, 3H), 6.78 (d, J ¼ 9.0 Hz, 1H),
5.10 (dd, J ¼ 12.9 & 5.4 Hz, 1H), 2.91 (s, 6H), 2.64e2.53 (m, 2H),
4.8.1. 2-(2,6-Dioxopiperidin-3-yl)-4-[(4-butylphenyl)amino]
isoindole-1,3-dione (6g)
Following the general procedure for Buchwald-Hartwig ami-
nation reactions, using 4-butylaniline (123 mg, 0.79 mmol) the title
compound was obtained after purification by flash column chro-
matography (30% / 50% ethyl acetate/hexanes) as a yellow solid
2.10e2.02 (m, 1H); 13C NMR (101 MHz, DMSO‑d6)
d 173.3, 170.5,
169.0, 167.6, 148.9, 145.5, 136.5, 132.8, 128.1, 125.9, 118.8, 113.6,
112.4, 110.5, 49.1, 31.4, 22.6; IR (neat) 3362 (N-H), 3264 (N-H), 1760
(C¼O), 1732 (C¼O), 1683 (C¼O), 1622 (C¼O), 1522 cmꢁ1; HRMS
calculated for C21H21N4O4: 393.1563, found: 393.1574 [MþH]þ
.
(207 mg, 98%). Rf
¼
0.33 (40% ethyl acetate/hexane);
4.8.5. 2-(2,6-Dioxopiperidin-3-yl)-4-{[4-(trifluoromethyl)phenyl]
amino}isoindole-1,3-dione (6m)
mp ¼ 205e207 ꢂC; 1H NMR (500 MHz, DMSO‑d6)
d 11.13 (s, 1H),
8.33 (s, 1H), 7.58 (dd, J ¼ 8.6, 7.1 Hz, 1H), 7.34 (d, J ¼ 8.5 Hz, 1H), 7.21
(dd, J ¼ 8.0, 5.3 Hz, 5H), 5.12 (dd, J ¼ 12.9 & 5.4 Hz, 1H), 2.90 (ddd,
J ¼ 17.3, 14.0 & 5.4 Hz, 1H), 2.66e2.52 (m, 4H), 2.14e1.98 (m, 1H),
1.56 (quint., J ¼ 7.4 Hz, 2H), 1.32 (sext., J ¼ 7.4 Hz, 2H), 0.90 (t,
Following the general procedure for Buchwald-Hartwig ami-
nation reactions, 4-trifluoromethylaniline (64 mg, 0.39 mmol) was
coupled with 4-iodothalidomide (99 mg, 0.26 mmol). Purification
by flash column chromatography (40% ethyl acetate/hexanes)
afforded the desired compound (60 mg, 67%) as a yellow solid.
J ¼ 7.4 Hz, 3H); 13C NMR (126 MHz, DMSO)
d 173.2, 170.5, 168.8,
12

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
Rf ¼ 0.36 (40% ethyl acetate/hexanes); mp ¼ 259e261 ꢂC; 1H NMR
crude 3-phenoxyphthalic anhydride (145 mg, 0.60 mmol), 3-
aminopiperidine-2,6-dione trifluoroacetate (153 mg, 0.63 mmol),
and sodium acetate (59 mg, 0.72 mmol). Glacial acetic acid was
added (1.8 mL, 3 M) and the mixture was heated to reflux with
stirring for 16 h. After cooling the reaction to room temperature,
solvent was removed under reduced pressure and the residue was
redissolved in ethyl acetate and cooled in an ice bath. The resulting
precipitate was collected by vacuum filtration, washed three times
with water and ethyl acetate, then dried under high vacuum to give
the title compound as a white solid (202 mg, 28% over 2 steps).
Rf ¼ 0.37 (40% ethyl acetate/hexanes); mp ¼ 244e246 ꢂC; 1H NMR
(400 MHz, DMSO‑d6) d 11.13 (s, 1H), 8.88 (s, 1H), 7.76e7.61 (m, 4H),
7.43 (dd, J ¼ 24.6, 7.4 Hz, 3H), 5.13 (dd, J ¼ 12.9, 5.4 Hz, 1H’),
2.97e2.82 (m, 1H), 2.70e2.55 (m, 1H), 2.18e2.02 (m, 1H); 13C NMR
(101 MHz, DMSO‑d6)
132.7, 126.5 (q, J ¼ 3.8 Hz), 125.9, 123.2, 122.5, 122.2, 122.0, 119.3,
115.3,114.7, 48.8, 31.0, 22.1; 19F NMR (376 MHz, DMSO‑d6)
d 172.8, 170.0, 167.6, 166.9, 144.3, 140.7, 136.2,
d
ꢁ57.89;
IR (neat) 3316 (N-H), 3210 (N-H), 1765 (C¼O), 1695 (C¼O), 1319,
1112, 741 cmꢁ1; HRMS calculated for C20H15N3O4F: 418.1015, found:
418.1017 [MþH]þ
.
4.8.6. 4-Benzyl-2-(2,6-dioxopiperidin-3-yl)isoindoline-1,3-dione
(5e)
(500 MHz, DMSO‑d6) d 11.12 (s, 1H), 7.84e7.76 (m, 1H), 7.64 (d,
J ¼ 7.2 Hz,1H), 7.48 (t, J ¼ 7.9 Hz, 2H), 7.28 (t, J ¼ 7.4 Hz,1H), 7.17 (dd,
J ¼ 12.8, 8.1 Hz, 3H), 5.13 (dd, J ¼ 12.8, 5.4 Hz,1H), 2.89 (ddd, J ¼ 17.0,
14.0, 5.4 Hz, 1H), 2.65e2.53 (m, 2H), 2.10e2.01 (m, 1H); 13C NMR
To a flame-dried reaction vial, 4-iodothalidomide (100 mg,
0.26 mol), Pd2(dba)3∙CHCl3 (22 mg, 0.021 mmol), triphenylphos-
phine (68 mg, 0.26 mmol), Ag2O (121 mg, 0.52 mmol) were added.
The vial was evacuated and backfilled with argon three times, then
fitted with a PTFE septum and THF (1.5 mL) was added, followed by
(126 MHz, DMSO)
d 172.8, 169.9, 166.6, 164.9, 154.8, 153.9, 137.2,
133.6,130.5,125.1,123.6,119.8,118.5,117.9, 31.0, 22.0; IR (neat) 3100
(N-H), 1775 (C¼O), 1705 (C¼O), 1610, 1588, 1475, 1394, 1256, 1201,
748 cmꢁ1; HRMS calculated for C19H15N2O5: 351.0981, found:
351.0989 [MþH]þ.
benzyl boronic acid pinacol ester (87 mL, 0.39 mmol). The reaction
vial was sealed under argon and heated to reflux for 18 h with
stirring. After cooling to room temperature solvent was removed in
vacuo and the residue was passed through a silica plug (20 / 50%
ethyl acetate/hexanes), then purified by PTLC (50% ethyl acetate/
hexanes) to afford the title compound as a white solid (61 mg, 67%).
Rf ¼ 0.48 (50% ethyl acetate/hexanes); mp ¼ 195e197 ꢂC; 1H NMR
4.8.9. 2-(2,6-Dioxopiperidin-3-yl)-4-(4-isopropylphenoxy)
isoindoline-1,3-dione (5h)
To a round-bottom flask fitted with a reflux condenser was
added 3-fluorophthalic anhydride (332 mg, 2.0 mmol), potassium
fluoride (232 mg, 4.0 mmol), 4-isopropylphenol (409 mg,
3.0 mmol), DMF (2.0 mL), and the mixture was heated to reflux and
stirred for 2 h. After cooling to room temperature, the reaction
mixture was diluted with ethyl acetate (20 mL) then washed with
saturated LiCl solution (3 ꢃ 20 mL), Na2CO3 solution (20 mL), water
(20 mL), and brine (20 mL). The organic layer was dried (MgSO4),
filtered, and removal of solvent under reduced pressure gave an
orange oil (539 mg) that was carried forward without further pu-
rification. To a flame-dried flask under an inert atmosphere of
argon was added the crude 4-(4-isopropylphenoxy)isobenzofuran-
1,3-dione (187 mg, 0.66 mmol), 3-aminopiperidine-2,6-dione tri-
fluoroacetate (160 mg, 0.66 mmol) and sodium acetate (65 mg,
0.66 mg,1.0 mmol) and glacial acetic acid (2.5 mL). The mixture was
heated to reflux with stirring for 16 h. Concentration under reduced
pressure afforded a crude oil which was subjected to column
chromatography (25 / 40% ethyl acetate/hexanes) to give the title
compound as a white solid (68 mg, 25% over 2 steps). Rf ¼ 0.47 (40%
ethyl acetate/hexanes); mp ¼ 232e233 ꢂC; 1H NMR (600 MHz,
(500 MHz, CDCl3)
d
8.04 (s, 1H), 7.74 (d, J ¼ 7.3 Hz, 1H), 7.62 (t,
J ¼ 7.6 Hz, 1H), 7.47 (d, J ¼ 7.8 Hz, 1H), 7.33e7.28 (m, 2H), 7.28e7.21
(m, 4H), 4.99 (dd, J ¼ 12.4, 5.4 Hz, 1H), 4.49 (s, 2H), 2.94e2.70 (m,
3H), 2.20e2.12 (m, 1H); 13C NMR (126 MHz, CDCl3)
d 170.9, 168.1,
168.0, 167.3, 147.4, 142.1, 136.4, 136.4, 134.4, 132.3, 129.2, 128.1,
126.9, 122.0, 49.3, 36.3, 33.9, 31.5, 29.9, 25.0, 24.1, 22.8; IR (neat):
3216 (N-H), 3084, 1770 (C¼O), 1700 (C¼O), 1601, 1388, 1322, 1256,
1194, 1119, 737, 700 cmꢁ1; HRMS calculated for C20H17N2O4:
349.1188, found: 349.1189 [MþH]þ.
4.8.7. 2-(2,6-Dioxopiperidin-3-yl)-4-(4-isopropylbenzyl)
isoindoline-1,3-dione (5f)
Prepared as described for 5e above, using 4-iodothalidomide 4a
(50 mg, 0.18 mmol) and 4-isopropylbenzyl boronic acid pinacol
ester (71 mg, 0.27 mmol). Purification of the crude residue by PTLC
(25% ethyl acetate/hexanes, then redeveloped in 35% ethyl acetate/
hexanes) afforded the desired product as a white solid (22 mg 41%).
Rf ¼ 0.42 (40% ethyl acetate/hexanes); mp 137e140 ꢂC; 1H NMR
(500 MHz, CDCl3)
d
8.03 (s, 1H), 7.73 (dd, J ¼ 7.3, 1.0 Hz, 1H), 7.62 (t,
CDCl3)
d
8.23 (s, 1H), 7.57 (dd, J ¼ 8.3, 7.4 Hz, 1H), 7.53 (dd, J ¼ 7.2,
J ¼ 7.6 Hz, 1H), 7.49 (dd, J ¼ 7.8, 1.0 Hz, 1H), 7.20e7.15 (m, 4H), 4.99
0.6 Hz, 1H), 7.27 (s, 1H), 7.07e7.01 (m, 3H), 5.00 (dd, J ¼ 12.5, 5.4 Hz,
(dd, J ¼ 12.4, 5.4 Hz, 1H), 4.45 (s, 2H), 2.94e2.71 (m, 4H), 2.19e2.13
1H), 2.97e2.72 (m, 4H), 2.19e2.11 (m, 1H), 1.26 (d, J ¼ 6.9 Hz, 6H);
(m, 1H), 1.23 (d, J ¼ 6.9 Hz, 6H); 13C NMR (126 MHz, CDCl3)
d
170.9,
13C NMR (151 MHz, CDCl3)
d 171.1, 168.2, 167.0, 165.4, 156.0, 152.4,
168.1, 168.0,167.3, 147.4, 142.1, 136.4,136.4,134.4,132.3, 129.2,128.1,
126.9, 122.0, 49.3, 36.3, 33.9, 31.5, 29.9, 25.0, 24.1, 22.8; IR (neat)
3214 (N-H), 2959, 1771 (C¼O), 1702 (C¼O), 1616, 1512, 1389, 1258,
1193, 1117, 737 cmꢁ1; HRMS calculated for C23H23N2O4: 391.1658,
found: 391.1657 [MþH]þ.
146.4, 136.3, 133.9, 128.2, 122.6, 120.6, 118.3, 117.5; IR (neat) 3320
(N-H), 3229 (N-H) 1769 (C¼O), 1621 (C¼O), 1510, 743 cmꢁ1; HRMS
calculated for C22H21N2O5: 393.1450, found: 393.1454 [MþH]þ.
4.9. General procedure for the synthesis of thioether derivatives
4.8.8. 2-(2,6-Dioxopiperidin-3-yl)-4-phenoxyisoindoline-1,3-dione
(5g)
To a flame-dried Schlenk flask was added the following in
sequence under an inert atmosphere of argon: 4-iodothalidomide
4a (200 mg, 0.52 mmol), Pd2(dba)3∙CHCl3 (16 mg, 0.016 mmol)
and XantPhos (16 mg, 0.032 mmol), 1,4-dioxane (2 mL), diisopro-
pylethylamine (0.50 mL, 1.04 mmol) and the required thiol
(0.57 mmol). The reaction mixture was heated to 90 ꢂC with stirring
for 16 h, cooled to room temperature and diluted with ethyl acetate
(10 mL). The mixture was filtered through a pad of Celite, washing
with ethyl acetate and DCM (ca. 20 mL each). The filtrate was
adsorbed to silica and purified via flash column chromatography on
silica to furnish the desired sulfide-functionalised thalidomide
derivative.
To a round-bottom flask fitted with a reflux condenser was
added 3-fluorophthalic anhydride (355 mg, 2.2 mmol), phenol
(190 mg, 2.42 mmol), and potassium fluoride (192 mg, 3.3 mmol)
followed by DMF (2.0 mL), and the ensuing solution heated to
reflux for 2 h. After cooling to room temperature, the mixture was
poured over ice cold water (50 mL) and the resulting precipitate
was collected by vacuum filtration. After drying under high vac-
uum, a white solid (155 mg) was obtained that was carried through
to the subsequent step without further purification. To a flame-
dried flask under an inert atmosphere of argon was added the
13

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
4.9.1. 2-(2,6-Dioxopiperidin-3-yl)-4-(phenylsulfanyl)isoindole-1,3-
dione (5i)
4.9.5. 2-(2,6-Dioxopiperidin-3-yl)-4-(quinolin-2-ylthio)
isoindoline-1,3-dione (8b)
Following the above general procedure using thiophenol
(32 mg, 0.29 mmol), the title compound was obtained after puri-
fication via flash column chromatography (40% ethyl acetate/hex-
anes) as a pale yellow solid (89 mg, 94%). mp ¼ 243e245 ꢂC; 1H
Following the general procedure for the synthesis of thioether
derivatives, using the coupling partner 2-quinolinethiol (93 mg),
the title compound was obtained after purification via flash column
chromatography (30 / 60% ethyl acetate/hexanes) as an orange
solid (188 mg, 87%); Rf ¼ 0.23 (40% ethyl acetate/hexanes); 1H NMR
NMR (400 MHz, DMSO‑d6)
d 11.16 (s, 1H), 7.94e7.27 (m, 7H), 6.94
(dd, J ¼ 1 Hz, 1H), 5.17 (dd, J ¼ 12.8, 5.4 Hz, 1H), 2.91 (ddd, J ¼ 16.8,
(500 MHz, CDCl3)
d
8.12 (d, J ¼ 8.5 Hz, 1H), 7.98 (s, 1H), 7.95e7.89
13.7 & 5.4 Hz, 1H), 2.69e2.53 (m, 2H), 2.16e2.03 (m, 1H); 13C NMR
(m, 2H), 7.87e7.80 (m, 2H), 7.75e7.67 (m, 2H), 7.55 (ddd, J ¼ 8.0, 6.9,
1.2 Hz, 1H), 7.46 (d, J ¼ 8.5 Hz, 1H), 4.99 (dd, J ¼ 12.4, 5.4 Hz, 1H),
2.95e2.70 (m, 3H), 2.21e2.12 (m, 1H); 13C NMR (151 MHz, CDCl3)
(126 MHz, DMSO‑d6)
d 172.8, 169.9, 166.7, 166.5, 139.0, 135.5, 135.2,
132.4, 130.6, 130.5, 130.3, 128.1, 124.9, 119.8, 49.0, 31.0, 21.9; IR
(neat) 3190, 1764 (C¼O), 1701 (C¼O), 1253, 1185, 743 cmꢁ1; HRMS
calculated for C19H15N2O4S: 367.0753, found: 367.0762 [MþH]þ.
d
170.8, 167.8, 166.8, 166.2, 155.8, 148.5, 137.8, 137.3, 134.4, 133.7,
133.1, 130.4, 129.7, 128.9, 127.8, 127.0, 126.9, 122.7, 122.7, 49.5, 31.5,
22.7; IR (neat): 3088, 1770 (C¼O), 1703 (C¼O), 1387, 1194,
739 cmꢁ1; HRMS for C22H16N3O4S calculated: 418.0862, found:
418.0865 [MþH]þ.
4.9.2. 2-(2,6-Dioxopiperidin-3-yl)-4-[(4-isopropylphenyl)sulfanyl]
isoindole-1,3-dione (5j)
Following the general procedure for the synthesis of thioether
derivatives, using the coupling partner 4-isopropylbenzenethiol
(81 mg, 0.53 mmol) the title compound was obtained after purifica-
tion via flash column chromatography (35% ethyl acetate/hexanes) as
a pale yellow solid (194 mg, 91%). Rf ¼ 0.50 (40% ethyl acetate/hex-
4.9.6. 2-(2,6-Dioxopiperidin-3-yl)-4-(o-tolylthio)isoindoline-1,3-
dione (7c)
Following the above general procedure for the synthesis of
thioether derivatives, using the coupling partner 2-
methylbenzenethiol (71 mg, 0.57 mmol), the title compound was
obtained after purification via flash column chromatography
(20 / 40% ethyl acetate/hexanes) as an off-white solid (167 mg,
84%). Rf ¼ 0.42 (40% ethyl acetate/hexanes); mp ¼ 249e251 ꢂC; 1H
anes); mp ¼ 238e240 ꢂC; 1H NMR (400 MHz, CDCl3)
d 8.08 (s, 1H),
7.60e7.47 (m, 3H), 7.43 (dd, J ¼ 8.2, 7.3 Hz,1H), 7.35 (d, J ¼ 8.1 Hz, 2H),
6.95 (dd, J ¼ 8.1, 0.9 Hz, 1H), 5.01 (dd, J ¼ 12.2, 5.4 Hz, 1H), 3.06e2.66
(m, 4H), 2.26e2.08(m,1H),1.30 (d, J ¼ 6.9 Hz, 6H); 13C NMR(101 MHz,
CDCl3)
d
170.7,167.9,167.0,166.8,151.3,141.8,136.0,134.2,132.5,131.0,
NMR (500 MHz, CDCl3)
d
8.28 (s, 1H), 7.59 (dd, J ¼ 7.5, 1.4 Hz, 1H),
128.3,125.2,125.0,119.5, 49.2, 34.0, 31.4, 23.8, 22.6; IR (neat) 3213 (N-
H), 1770 (C¼O), 1702 (C¼O), 1602, 1390, 745 cmꢁ1; HRMS calculated
for C22H21N2O4S: 409.1222, found: 409.1233 [MþH]þ.
7.55 (dd, J ¼ 7.3, 0.9 Hz, 1H), 7.45e7.39 (m, 3H), 7.30 (td, J ¼ 7.4,
1.8 Hz, 1H), 6.75 (dd, J ¼ 8.2, 0.8 Hz, 1H), 5.05e4.99 (m, 1H),
2.94e2.74 (m, 3H), 2.40 (s, 3H), 2.21e2.15 (m, 1H); 13C NMR
(126 MHz, CDCl3) d 171.0,168.1,167.1,167.0,143.7,140.8,137.4,134.5,
4.9.3. 2-(2,6-Dioxopiperidin-3-yl)-4-((4-isopropylphenyl)sulfonyl)
isoindoline-1,3-dione (5k)
132.9, 131.5, 130.9, 130.5, 127.8, 127.7, 125.4, 119.7, 49.4, 31.5, 22.8,
20.9; IR (neat): 3185 (N-H), 2918, 1768 (C¼O), 1731 (C¼O), 1693
(C¼O), 1603, 1459, 1388, 1207, 1013, 747 cmꢁ1; HRMS calculated for
A solution of 77% meta-chloroperbenzoic acid (134 g, 0.6 mmol)
in dichloromethane (2 mL) was added dropwise to a stirred solu-
tion of sulfide derivative 5j (85 mg, 0.21 mmol) in dichloromethane
(2 mL). The resulting pale green solution was left to stir for 24 h,
then treated with sodium thiosulfate (2 mL) followed by
dichloromethane (6 mL). The mixture was extracted with sodium
bicarbonate (3 ꢃ 10 mL), washed with brine (2 ꢃ 10 mL), dried
(MgSO4), filtered and solvent removed under reduced pressure.
Purification by PTLC (DCM) afforded the desired sulfone 5k as a
white solid (53 mg, 58%). Rf ¼ 0.63 (10% MeOH/DCM); 1H NMR
C
20H17N2O4S: 381.0909, found: 381.0906 [MþH]þ.
4.9.7. 2-(2,6-Dioxopiperidin-3-yl)-4-(m-tolylthio)isoindoline-1,3-
dione (7d)
Following the general procedure for the synthesis of thioether
derivatives, using the coupling partner 3-methylbenzenethiol
(71 mg, 0.57 mmol), the title compound was obtained after puri-
fication via flash column chromatography (25 / 50% ethyl acetate/
hexanes) as a pale yellow solid (180 mg, 91%). Rf ¼ 0.36 (40% ethyl
acetate/hexanes); mp ¼ 272e274C; 1H NMR (500 MHz, DMSO‑d6)
(400 MHz, CDCl3)
d
8.58 (dd, J ¼ 7.9, 1.1 Hz, 1H, Ar-H), 8.13e8.02 (m,
4H), 7.96 (t, J ¼ 7.7 Hz, 1H), 7.37 (d, J ¼ 8.4 Hz, 2H), 4.96 (dd,
J ¼ 12.4,5.3 Hz, 1H), 3.02e2.61 (m, 4H), 2.19e2.09 (m, 1H), 1.24 (d,
J ¼ 6.9 Hz); IR (neat) 3255 (N-H), 3090 (N-H), 1784 (C¼O), 1706
(C¼O), 1383, 1320 (S¼O), 1160, 1122 (S¼O) cmꢁ1; HRMS calculated
for C22H21N2O6S: 441.1120, found: 441.1118 [MþH]þ.
d 11.15 (s, 1H), 7.67e7.63 (m, 2H), 7.53e7.37 (m, 4H), 6.94 (d,
J ¼ 7.0 Hz,1H), 5.16 (dd, J ¼ 12.7, 5.4 Hz,1H), 2.91 (ddd, J ¼ 15.8,12.4,
5.8 Hz, 1H), 2.67e2.53 (m, 2H), 2.37 (s, 3H), 2.14e2.04 (m, 1H); 13C
NMR (126 MHz, DMSO)
d 172.8, 169.9, 166.7, 166.5, 140.1, 139.2,
135.9, 135.2, 132.6, 132.4, 131.0, 130.6, 130.3, 127.8, 124.8, 119.7, 79.2,
49.0, 31.0, 21.9, 20.8; IR (neat) 3179 (N-H), 2920, 1769 (C¼O), 1731
(C¼O), 1694 (C¼O), 1390, 1207, 747 cmꢁ1; HRMS calculated for
4.9.4. 2-(2,6-Dioxopiperidin-3-yl)-4-(pyrimidin-2-ylthio)
isoindoline-1,3-dione (7a)
C
20H17N2O4S: 381.0909, found: 381.0911 [MþH]þ.
Following the general procedure for the synthesis of thioether
derivatives, using the coupling partner 2-mercaptopyrimidine
(64 mg), the title compound was obtained after purification via
flash column chromatography (40 / 70% ethyl acetate/hexanes) as
a pale yellow solid (158 mg, 82%). Rf ¼ 0.38 (70% ethyl acetate/
4.9.8. 2-(2,6-Dioxopiperidin-3-yl)-4-(p-tolylthio)isoindoline-1,3-
dione (7e)
Following the general procedure for the synthesis of thioether
derivatives, using the coupling partner 4-methylbenzenethiol
(71 mg, 0.57 mmol), the title compound was obtained after puri-
fication via flash column chromatography (35% ethyl acetate/hex-
anes) as a pale yellow solid (102 mg, 52%). Rf ¼ 0.40 (40% ethyl
acetate/hexanes); mp ¼ 256e258 ꢂC; 1H NMR (500 MHz, DMSO)
hexanes); mp ¼ 244e246 ꢂC; 1H NMR (500 MHz, CDCl3)
d 8.52 (d,
J ¼ 4.9 Hz, 2H), 8.05e7.99 (m, 2H), 7.95 (dd, J ¼ 7.5 and 1.0 Hz, 1H),
7.79 (t, J ¼ 7.7 Hz, 1H), 7.06 (t, J ¼ 4.8 Hz, 1H), 4.95 (dd, J ¼ 12.4 and
5.3 Hz, 1H), 2.89 (dt, J ¼ 16.6 and 2.3 Hz, 1H), 2.84e2.69 (m, 2H),
2.16e2.10 (m, 1H);13C NMR (126 MHz, CDCl3)
d
170.8, 170.6, 167.7,
d
11.11 (s, 1H), 7.68e7.58 (m, 3H), 7.52 (t, J ¼ 6.0 Hz, 2H), 7.39 (dt,
166.6, 165.8, 157.8, 140.7, 134.6, 133.5, 132.1, 130.1, 124.2, 118.0, 49.4,
31.5, 22.7; IR (neat) 3277 (N-H),1767 (CþO),1704 (C¼O),1552,1381,
1178, 890, 744 cmꢁ1; HRMS calculated for C17H13N4O4S: 369.0658,
found: 369.0658 [MþH]þ.
J ¼ 8.4, 4.4 Hz, 1H), 6.73 (p, J ¼ 3.7 Hz, 1H), 5.16 (dd, J ¼ 12.8, 5.4 Hz,
1H), 2.91 (ddd, J ¼ 16.8, 13.8, 5.4 Hz, 1H), 2.66e2.53 (m, 2H), 2.34 (s,
3H), 2.13e2.06 (m, 1H); 13C NMR (126 MHz, DMSO)
d 172.8, 169.9,
166.7, 166.6, 142.8, 138.4, 136.8, 135.3, 132.6, 131.5, 131.01, 129.8,
14

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
127.9, 127.1, 124.9, 119.6, 49.0, 31.0, 21.9, 20.2; IR (neat) 3199 (N-H),
3076, 2915, 1769 (C¼O), 1730 (C¼O), 1694 (C¼O), 1458, 1389, 1257,
893, 745 cmꢁ1; HRMS calculated for C20H17N2O4S: 381.0909, found:
381.0906 [MþH]þ.
5.17 (dd, J ¼ 12.8, 5.4 Hz, 1H), 3.85 (s, 3H), 2.91 (ddd, J ¼ 16.7, 13.7,
5.4 Hz, 1H), 2.65e2.52 (m, 2H), 2.09 (ddd, J ¼ 9.7, 5.7, 2.6 Hz, 1H);
13C NMR (126 MHz, DMSO‑d6)
d 172.8, 169.8, 166.7, 166.5, 161.0,
140.4, 137.6, 135.1, 132.3, 130.1, 124.5, 119.4, 117.9, 116.1, 55.5, 49.0,
31.0, 22.0; IR (neat) 3196 (N-H), 3095, 2942, 2688 (OCH3 C-H), 1766
4.9.9. 4-((4-(tert-Butyl)phenyl)thio)-2-(2,6-dioxopiperidin-3-yl)
isoindoline-1,3-dione (7f)
Following the general procedure for the synthesis of thioether
derivatives, using the coupling partner 4-(tert-butyl)-benzenethiol
(C¼O), 1732 (C¼O), 1695 (C¼O), 1593, 1392, 1252, 1198, 741 cmꢁ1
;
HRMS calculated for C20H17N2O5S: 397.0858, found 397.0862
[MþH]þ.
(118
m
L), the title compound xx was obtained after purification via
4.9.13. 2-(2,6-Dioxopiperidin-3-yl)-4-((4-(methylsulfanyl)phenyl)
thio)isoindoline-1,3-dione (7j)
flash column chromatography (20 / 40% ethyl acetate/hexanes) as
a pale yellow solid (212 mg, 97%). Rf ¼ 0.32 (40% ethyl acetate/
Following the general procedure for the synthesis of thioether
derivatives, using the coupling partner 4-(methylsulfanyl)thio-
phenol (45 mg), the title compound was obtained after purification
via flash column chromatography (30 / 50% ethyl acetate/hex-
anes) as a yellow solid (42 mg, 38%). Rf ¼ 0.44 (40% ethyl acetate/
hexanes); mp ¼ 268e270 ꢂC; 1H NMR (500 MHz, CDCl3)
d 8.36 (s,
1H), 7.61e7.46 (m, 5H), 7.42 (t, J ¼ 7.7 Hz, 1H), 6.95 (d, J ¼ 8.1 Hz,
1H), 5.02 (dd, J ¼ 12.1, 5.1 Hz, 1H), 2.95e2.73 (m, 3H), 2.16 (dd,
J ¼ 7.4, 5.2 Hz, 1H), 1.36 (s, 9H); 13C NMR (126 MHz, CDCl3)
d 171.2,
168.2, 167.1, 167.0, 153.7, 141.8, 135.8, 134.4, 132.7, 131.1, 127.4, 125.2,
125.1, 119.7, 49.4, 35.0, 31.5, 31.3, 22.8; IR (neat, cmꢁ1) 3234 (N-H),
3096, 2959, 1769 (C¼O), 1701 (C¼O), 1489, 1388, 1257, 1115, 891,
738; HRMS calculated for C23H23N2O4S: 423.1379, found: 423.1380
[MþH]þ.
hexanes); mp ¼ 242e244 ꢂC; 1H NMR (400 MHz, CDCl3)
8.20 ¼ (s,
d
1H, COCN-H), 7.59e7.40 (m, 4H, Ar-H), 7.36e7.29 (m, 2H, Ar-H),
6.98e6.92 (m, 1H, Ar-H), 5.01 (dd, J ¼ 12.1, 5.4 Hz, 1H), 2.99e2.68
(m, 3H), 2.53 (s, 3H), 2.22e2.10 (m, 1H); 13C NMR (75 MHz, CDCl3)
d
171.1, 168.1, 167.0, 142.3, 141.6, 136.4, 134.5, 132.7, 131.0, 127.20,
124.1, 119.9, 49.4, 31.5, 22.8, 15.2; IR (neat) 3213 (N-H), 3091, 1768
4.9.10. 2-(2,6-Dioxopiperidin-3-yl)-4-((4-(trifluoromethyl)phenyl)
thio)isoindoline-1,3-dione (7g)
(C¼O), 1699 (C¼O), 1385, 1256 1128 749 cmꢁ1; HRMS calculated for
C
20H17N2O4S: 413.0630, found: 413.0627 [MþH]þ.
Following the general procedure for the synthesis of thioether
derivatives, using 4-trifluoromethylbenzenethiol (93 mg), the title
compound was obtained after purification via flash column chro-
matography (40% ethyl acetate/hexanes) as a yellow solid (157 mg,
70%). Rf ¼ 0.33 (40% ethyl acetate/hexanes); mp ¼ 223e226 ꢂC; 1H
4.9.14. 4-(Benzylthio)-2-(2,6-dioxopiperidin-3-yl)isoindoline-1,3-
dione (7k)
Following the general procedure for the synthesis of thioether
derivatives, using the coupling partner benzyl thiol (67 mL), the title
compound was obtained after purification via flash column chro-
matography (20 / 60% ethyl acetate/hexanes) as a yellow solid
NMR (500 MHz, CDCl3)
d
8.01 (s, 1H), 7.75e7.69 (m, 4H), 7.65 (d,
J ¼ 7.3 Hz, 1H), 7.51 (dd, J ¼ 8.1, 7.3 Hz, 1H), 7.05 (d, J ¼ 8.2 Hz, 1H),
5.01 (dd, J ¼ 12.4, 5.3 Hz,1H), 3.02e2.67 (m, 3H), 2.22e2.13 (m, 1H);
(127 mg, 64%). Rf
¼
0.35 (40% ethyl acetate/hexanes);
13C NMR (126 MHz, CDCl3)
d
170.8, 167.9, 166.8, 166.7, 135.4, 134.8,
mp ¼ 223e225 ꢂC; 1H NMR (500 MHz, DMSO‑d6)
d 11.12 (s, 1H),
132.1, 127.0 (q, J ¼ 3.8 Hz), 120.9, 55.9, 30.5, 21.9; 19F NMR (CDCl3,
7.82 (d, J ¼ 8.1 Hz, 1H), 7.76 (t, J ¼ 7.7 Hz, 1H), 7.63 (d, J ¼ 7.1 Hz, 1H),
7.48 (d, J ¼ 7.2 Hz, 2H), 7.35 (t, J ¼ 7.5 Hz, 2H), 7.28 (t, J ¼ 7.3 Hz, 1H),
5.11 (dd, J ¼ 12.9, 5.4 Hz, 1H), 4.45 (s, 2H), 2.88 (ddd, J ¼ 17.1, 14.0,
5.4 Hz, 1H), 2.67e2.52 (m, 2H), 2.10e2.00 (m, 1H); 13C NMR
MHz):
d
ꢁ59.36; IR (neat): 3196 (N-H), 3103, 2900, 1707 (C¼O),
1598, 1397, 1318, 1138, 842, 736 cmꢁ1; HRMS calculated for
C
20H14N2O4F3S: 435.0626, found: 435.0619 [MþH]þ.
(126 MHz, DMSO‑d6)
d 172.8, 169.9, 166.7, 166.6, 138.4, 136.1, 134.9,
4.9.11. 2-(2,6-Dioxopiperidin-3-yl)-4-((5-(trifluoromethyl)pyridin-
2-yl)thio)isoindoline-1,3-dione (7h)
132.5, 130.7, 129.0, 128.6, 127.4, 125.3, 119.1, 79.2, 48.90, 34.0, 30.9,
21.9; IR (neat) 3196 (N-H), 3101, 3021, 1761, 1702 (C¼O), 1598, 1458,
1395, 1326, 1259, 1197, 1019, 736 cmꢁ1; HRMS calculated for
Following the general procedure for the synthesis of thioether
derivatives, using the coupling partner 5-(trifluoromethyl)pyri-
dine-2-thiol (102 mg), the title compound was obtained after pu-
rification via flash column chromatography (20 / 40% ethyl
acetate/hexanes) as a yellow solid (70 mg, 31%). Rf ¼ 0.43 (40% ethyl
acetate/hexanes); mp ¼ 262e265 ꢂC; 1H NMR (500 MHz, CDCl3)
C
20H17N2O4S: 418.0862, found: 418.0865 [MþH]þ.
4.9.15. 4-(Butylthio)-2-(2,6-dioxopiperidin-3-yl)isoindoline-1,3-
dione (7l)
Following the general procedure for the synthesis of thioether
derivatives, using the coupling partner 1-buntanethiol (61 mL), the
d
8.63 (d, J ¼ 1.2 Hz, 1H), 8.00 (s, 1H), 7.92e7.85 (m, 2H), 7.82 (dd,
J ¼ 8.4, 2.3 Hz, 1H), 7.74 (t, J ¼ 7.7 Hz, 1H), 7.45 (d, J ¼ 8.4 Hz, 1H),
4.95 (dd, J ¼ 12.5, 5.4 Hz,1H), 2.93e2.85 (m,1H), 2.84e2.66 (m, 2H),
title compound was obtained after purification via flash column
chromatography (ethyl acetate/hexanes 20:80 / 50:50) as a yel-
low solid (146 mg, 81%). Rf ¼ 0.57 (50% ethyl acetate/hexanes);
2.15e2.10 (m, 1H); 13C NMR (126 MHz, CDCl3)
d 170.6, 167.6, 166.3,
165.7, 161.3, 146.8 (q, J ¼ 4.0 Hz), 139.3, 134.6, 134.5, 133.9 (q,
J ¼ 3.4 Hz), 133.3, 130.9, 130.6, 123.7, 123.1, 49.4, 31.3, 22.5.; IR
(neat): 3196 (N-H), 2901, 1707 (C¼O), 1598, 1454, 1318, 1138, 842,
736 cmꢁ1; HRMS calculated for C19H13N3O4F3S: 436.0579, found:
436.0576 [MþH]þ.
mp ¼ 192e194 ꢂC; 1H NMR (500 MHz, CDCl3)
d 8.29 (s, 1H), 7.59
(dd, J ¼ 18.7, 7.2 Hz, 2H), 7.50 (d, J ¼ 7.6 Hz, 1H), 4.97 (dd, J ¼ 11.4,
4.7 Hz, 1H), 3.04 (t, J ¼ 7.0 Hz, 2H), 2.91e2.71 (m, 3H), 2.12 (d,
J ¼ 5.5 Hz, 1H), 1.74 (quint, 2H), 1.52 (app. dd, J ¼ 14.2, 7.0 Hz, 2H),
0.96 (t, J ¼ 7.1 Hz, 3H); 13C NMR (126 MHz, CDCl3)
d 171.2, 168.2,
167.0, 166.9, 140.5, 134.4, 133.1, 130.0, 126.2, 119.2, 49.3, 31.5, 30.6,
30.4, 22.8, 22.2, 13.7; IR (neat) 3198 (N-H), 3095, 2905, 1769 (C¼O),
1697 (C¼O), 1459, 1389, 1322, 1260, 1199, 1183, 1116, 1014, 893,
742 cmꢁ1; HRMS calculated for C17H19N2O4S: 347.1066, found:
347.1068 [MþH]þ.
4.9.12. 2-(2,6-Dioxopiperidin-3-yl)-4-((4-methoxyphenyl)thio)
isoindoline-1,3-dione (7i)
Following the general procedure for the synthesis of thioether
derivatives, using the coupling partner 4-mercaptoanisole (56 mL),
the title compound was obtained after purification via flash column
chromatography (30 / 50% ethyl acetate/hexanes) as a yellow
solid (148 mg, 72%). Rf ¼ 0.40 (40% ethyl acetate/hexanes);
4.9.16. 4-(tert-Butylthio)-2-(2,6-dioxopiperidin-3-yl)isoindoline-
1,3-dione (7m)
Following the general procedure for the synthesis of thioether
derivatives, using the coupling partner 2-methyl-2-propanethiol
mp ¼ 196e198 ꢂC; 1H NMR (500 MHz, DMSO‑d6)
d
11.15 (s, 1H),
7.68e7.54 (m, 4H), 7.18e7.10 (m, 2H), 6.88 (dd, J ¼ 7.7, 1.2 Hz, 1H),
15

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
(65
m
L), the title compound was obtained after purification via flash
7.42 (dd, J ¼ 8.1, 7.3 Hz, 1H), 6.97e6.93 (m, 1H), 5.03e4.96 (m, 1H),
3.22 (s, 1H), 3.03e2.96 (m, 1H), 2.87e2.74 (m, 1H), 2.17e2.09 (m,
column chromatography (20:80 / 50:50 ethyl acetate/hexanes) as
a yellow solid (178 mg, 99%). Rf ¼ 0.43 (40% ethyl acetate/hexanes);
1H), 1.36 (s, 2H); 13C NMR (126 MHz, CDCl3)
d 171.3, 168.8, 167.3,
mp ¼ 189e191 ꢂC; 1H NMR (500 MHz, CDCl3)
d
8.01 (s, 1H), 7.85
167.1, 153.7, 141.8, 135.8, 134.3, 132.7, 131.0, 127.4, 125.3, 125.1, 119.6,
50.1, 35.0, 32.0, 31.3, 27.4, 22.1; IR (neat) 2961, 1768 (C¼O), 1709
(C¼O), 1679 (C¼O), 1602, 1456, 1389, 1115, 891, 739 cmꢁ1; HRMS
calculated for C24H25N2O4S: 437.1535, found: 437.1530 [MþH]þ.
(dd, J ¼ 7.9, 0.9 Hz, 1H), 7.82 (dd, J ¼ 7.4, 0.9 Hz, 1H), 7.67 (t,
J ¼ 7.6 Hz, 1H), 4.98 (dd, J ¼ 12.4, 5.4 Hz, 1H), 2.94e2.87 (m, 2H),
2.83 (td, J ¼ 12.7, 12.2, 3.9 Hz, 1H), 2.79e2.70 (m, 1H), 2.18e2.11 (m,
2H), 1.43 (s, 8H); 13C NMR (126 MHz, CDCl3)
d 170.9, 167.9, 166.6,
166.4, 142.1, 135.2, 133.7, 133.4, 132.4, 122.9, 49.5, 48.6, 31.5, 31.3,
22.7; IR (neat): 3185 (N-H), 1770 (C¼O), 1697 (C¼O), 1602, 1460,
1391, 1200, 1118, 743 cmꢁ1; HRMS calculated for C17H19N2O4S:
347.1066, found: 347.1068 [MþH]þ.
4.10. Ligand parameterisation
Bonded and van der Waals parameters for Thalidomide, 6f and
7f were parameterised with the CHARMM General Force Field using
the paramchem webserver. To determine partial charges for atoms
in these molecules the geometry of each was optimized at MP2/6-
31þG* level of theory. This geometry was employed in the calcu-
lation of the restrained electrostatic potential (RESP) charges at the
HartreeFock/6-31G* level of theory to be consistent with the rest of
the force field. Quantum mechanical calculations were conducted
with the Gaussian09 package [69].
4.9.17. 4-(Cyclohexylthio)-2-(2,6-dioxopiperidin-3-yl)isoindoline-
1,3-dione (7n)
Following the general procedure for the synthesis of thioether
derivatives, using the coupling partner cyclohexanethiol (90 mL),
the title compound was obtained after purification via flash column
chromatography (30% / 50% ethyl acetate/hexanes) as a pale
yellow solid (170 mg, 88%). Rf ¼ 0.27 (30% ethyl acetate/hexanes);
mp ¼ 205e207 ꢂC. 1H NMR (600 MHz, CDCl3)
d
7.95 (s, 1H),
4.11. Simulation setup
7.65e7.58 (m, 2H), 7.55 (dd, J ¼ 7.9, 0.9 Hz, 1H), 4.96 (dd, J ¼ 12.5,
5.4 Hz, 1H), 3.45 (dd, J ¼ 6.9, 3.7 Hz, 1H), 2.93e2.88 (m, 1H), 2.83
(dd, J ¼ 11.1, 2.4 Hz, 1H), 2.77e2.70 (m, 1H), 2.18e2.07 (m, 3H),
1.89e1.81 (m, 2H), 1.72e1.66 (m, 1H), 1.58e1.50 (m, 2H), 1.48e1.39
The crystal structure of the CRBN domain of the E3 ubiquitin
ligase (PDB ID 4CI1) [70], with thalidomide co-crystallised, was
used as a basis for the simulations. Residues 321e426 of the CRBN
domain were included in all simulations. Our parameterised
thalidomide, 6f and 7f were aligned to the thalidomide present in
the crystal structure (PDB ID 4CI1) [63]. The protein and the drug
(including the Zn2þ ion present in the crystal structure) were then
combined together, and the system was solvated with ~7800 TIP3P
water particles. The system was then neutralised and ionised with
0.15 mM KCl. Simulations used the CHARMM36 m[64] force field
for the protein and ion parameters from Juong and Cheatham [65].
(m, 2H), 1.38e1.29 (m, 1H). 13C NMR (126 MHz, CDCl3)
d 170.8,
167.93, 167.0, 166.9, 139.8, 134.3, 133.3, 131.2, 126.6, 119.5, 49.3, 43.3,
32.96, 32.93, 31.5, 26.1, 25.8, 22.8; IR (neat) 3190 (N-H), 3077, 2922,
1769 (C¼O), 1697 (C¼O), 1389, 1257 893, 741 cmꢁ1; HRMS calcu-
lated for C19H21N2O4S: 373.1222, found: 373.1217 [MþH]þ.
4.9.18. 4-(Adamantylthio)-2-(2,6-dioxopiperidin-3-yl)isoindoline-
1,3-dione (7o)
Following the general procedure for the synthesis of thioether
derivatives, using the coupling partner 1-adamantanethiol (98 mg),
the title compound was obtained after purification via flash column
chromatography (ethyl acetate/hexanes 20:80 / 50:50) as a pale
yellow solid (204 mg, 92%). Rf ¼ 0.41 (40% ethyl acetate/hexanes);
mp could not be determined (turned black at ca. 140 ꢂC, presum-
4.12. Simulation parameters for equilibration and production runs
Simulations were run in NAMD2.13 [66]. For the equilibration
phase, the system was initially minimised for 10,000 steps, and
simulated for 1ns with the protein, drug and Zn2þ ion were
completely restrained, to let the water equilibrate in the system.
Restraints were then put on the Ca carbons of the backbone, as well
as the drug and the Zn2þ ion, and reduced from 5kCal/mol to 0.1kCal/
mol over 2ns increments (5->2->1->0.5->0.2->0.1kCal/mol). For
the production runs, constraints were put on residues 333e340, to
mimic CRBN’s interaction with the DDB1 domain of the E3 ubiquitin
ligase. Additional constraints were also put on the Zn2þ ion, as it was
liable to leave its pocket if these were not included. Production runs
were 100ns in length, with a temperature of 310K and a time step of
2fs. Langevin dynamics for the thermostat and barostat was used,
with isotropic pressure coupling. Particle Mesh Ewald (PME) [67]
was used to calculate electrostatics.
ably decomposed); 1H NMR (600 MHz, CDCl3)
d 8.26 (s, 1H), 7.82 (t,
J ¼ 8.5 Hz, 2H), 7.64 (t, J ¼ 7.6 Hz, 1H), 4.98 (dd, J ¼ 12.3, 5.3 Hz, 1H),
2.94e2.69 (m, 3H), 2.14 (dd, J ¼ 8.8, 3.9 Hz, 1H), 2.04 (s, 3H), 1.93 (s,
6H),1.64 (q, J ¼ 12.7 Hz, 6H).13C NMR (151 MHz, CDCl3)
d 171.1, 168.1,
166.6, 166.4, 143.9, 133.4, 133.4, 133.3, 132.6, 123.4, 51.4, 49.4, 43.8,
36.1, 31.5, 30.2, 22.7; IR (neat) 3256 (N-H), 3103, 2903, 2849, 1770,
1703 (C¼O),1457,1382,1256,1192,1117,1035, 891, 740 cmꢁ1; HRMS
calculated for C23H25N2O4S: 425.1535, found: 425.1525 [MþH]þ.
4.9.19. 4-((4-(tert-Butyl)phenyl)thio)-2-(1-methyl-2,6-
dioxopiperidin-3-yl)isoindoline-1,3-dione (7p)
To a flame-dried Schlenk flask, thiothalidomide 8e (100 mg,
mmol, 1.0 equiv) and K2CO3 (66 mg, 0.47 mmol, 2.0 equiv) were
suspended in DMF (1 mL) and cooled to 0 ꢂC with continuous
stirring. Methyl iodide (40 mg, 0.28 mmol, 1.2 equiv) was added
slowly dropwise, the mixture was allowed to warm to room tem-
perature and then stirred for a further 16 h. The reaction mixture
was diluted with ethyl acetate (15 mL), washed with LiCl (1M,
15 mL) three times, and these combined aqueous fractions were
extracted with additional ethyl acetate (15 mL). The combined
organic layers were washed with brine (30 mL), dried over MgSO4,
filtered and concentrated under reduced pressure. The crude res-
idue was subjected to purification by flash column chromatography
(40% ethyl acetate/hexanes) to give the title compound as a pale
4.13. Visualisation and analysis
Visualisation was done with Visual Molecular Dynamics (VMD)
[68]. The namd_energy plugin was used in NAMD to calculate the
interaction energies. Percentage of interactions between HIS380,
HIS399 and the respective drug targets was calculated using in-house
Tcl and Python scripts, using a cutoff of 7 Å to denote an interaction.
Declaration of competing interest
The authors declare that they have no known competing
financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
yellow solid. Rf
¼
0.34 (40% ethyl acetate/hexanes).
mp ¼ 184e185 ꢂC; 1H NMR (500 MHz, CDCl3)
d 7.55e7.48 (m, 1H),
16

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
803e809, https://doi.org/10.1038/nsmb.2874.
Acknowledgements
€
[16] G. Petzold, E.S. Fischer, N.H. Thoma, Structural basis of lenalidomide-induced
CK1 degradation by the CRL4 CRBN ubiquitin ligase, Nature 532 (7597)
a
The authors acknowledge the CMCA Research Facility at UWA.
The ARC is gratefully acknowledged for support through
DP200100860. Further NMR assistance from Dr. Gareth Nealon is
gratefully acknowledged. Michael J. Nutt is a grateful recipient of
financial support from an Australian RTP scholarship and the Ernest
and Evelyn Havill-Shacklock scholarship.
(2016) 127e130, https://doi.org/10.1038/nature16979.
[17] Q.L. Sievers, G. Petzold, R.D. Bunker, A. Renneville, M. Słabicki, B.J. Liddicoat,
€
W. Abdulrahman, T. Mikkelsen, B.L. Ebert, N.H. Thoma, Defining the human
C2H2 zinc finger degrome targeted by thalidomide analogs through CRBN,
Science 362 (6414) (2018), https://doi.org/10.1126/science.aat0572 eaat0572.
[18] M.E. Matyskiela, S. Couto, X. Zheng, G. Lu, J. Hui, K. Stamp, C. Drew, Y. Ren,
M. Wang, A. Carpenter, C.W. Lee, T. Clayton, W. Fang, C.C. Lu, M. Riley,
P. Abdubek, K. Blease, J. Hartke, G. Kumar, R. Vessey, M. Rolfe, L.G. Hamann,
P.P. Chamberlain, SALL4 mediates teratogenicity as a thalidomide-dependent
cereblon substrate, Nat. Chem. Biol. 14 (10) (2018) 981e987, https://doi.org/
10.1038/s41589-018-0129-x.
Appendix A. Supplementary data
[19] A.S. Sperling, M. Burgess, H. Keshishian, J.A. Gasser, S. Bhatt, M. Jan,
M. Słabicki, R.S. Sellar, E.C. Fink, P.G. Miller, B.J. Liddicoat, Q.L. Sievers,
R. Sharma, D.N. Adams, E.A. Olesinski, M. Fulciniti, N.D. Udeshi, E. Kuhn,
A. Letai, N.C. Munshi, S.A. Carr, B.L. Ebert, Patterns of substrate affinity,
competition, and degradation kinetics underlie biological activity of thalido-
mide analogs, Blood 134 (2) (2019) 160e170, https://doi.org/10.1182/
blood.2019000789.
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.ejmech.2021.113353.
References
[20] Z. Qu, C. Jiang, J. Wu, Y. Ding, Lenalidomide induces apoptosis and inhibits
angiogenesis via caspase-3 and VEGF in hepatocellular carcinoma cells, Mol.
Med. Rep. 14 (5) (2016) 4781e4786, https://doi.org/10.3892/mmr.2016.5797.
[21] Y.-Y. Chen, Thalidomide-based multidisciplinary treatment for patients with
advanced hepatocellular carcinoma: a retrospective analysis, World J. Gas-
troenterol. 18 (5) (2012) 466, https://doi.org/10.3748/wjg.v18.i5.466.
[22] K. Woo, S.G. Stewart, G.S. Kong, M.L. Finch-Edmondson, B.J. Dwyer, S.Y. Yeung,
L.J. Abraham, S.S. Kampmann, L.A. Diepeveen, A.M.M. Passman, C.L. Elsegood,
J.E.E. Tirnitz-Parker, B.A. Callus, J.K. Olynyk, G.C.T. Yeoh, Identification of a
thalidomide derivative that selectively targets tumorigenic liver progenitor
cells and comparing its effects with lenalidomide and sorafenib, Eur. J. Med.
Chem. 120 (2016) 275e283, https://doi.org/10.1016/j.ejmech.2016.03.015.
[23] M.A. El-Zahabi, H. Sakr, K. El-Adl, M. Zayed, A.S. Abdelraheem, S.I. Eissa,
H. Elkady, I.H. Eissa, Design, synthesis, and biological evaluation of new
challenging thalidomide analogs as potential anticancer immunomodulatory
agents, Bioorg. Chem. 104 (2020), 104218, https://doi.org/10.1016/
j.bioorg.2020.104218.
[1] T.R.C. Vergara, S. Samer, J.R. Santos-Oliveira, L.B. Giron, M.S. Arif, M.L. Silva-
Freitas, L.A. Cherman, M.S. Treitsman, A. Chebabo, M.C.A. Sucupira, A.M. Da-
Cruz, R.S. Diaz, Thalidomide is associated with increased T cell activation and
inflammation in antiretroviral-naive HIV-infected individuals in a randomised
clinical trial of efficacy and safety, EBioMedicine 23 (2017) 59e67, https://
doi.org/10.1016/j.ebiom.2017.08.007.
[2] S. Kumar, T.E. Witzig, S.V. Rajkumar, Thalidomide: current role in the treat-
ment of non-plasma cell malignancies,
https://doi.org/10.1200/JCO.2004.10.127.
J Clin Oncol (2004) 2477e2488,
[3] R.J. D’Amato, M.S. Loughnan, E. Flynn, J. Folkman, Thalidomide is an inhibitor
of angiogenesis, Proc. Natl. Acad. Sci. U.S.A. 91 (9) (1994) 4082e4085, https://
doi.org/10.1073/pnas.91.9.4082.
[4] Y.X. Zhu, K.M. Kortuem, A.K. Stewart, Molecular mechanism of action of
immune-modulatory drugs thalidomide, lenalidomide and pomalidomide in
multiple myeloma. Leukemia and lymphoma, Taylor & Francis April 28 (2013)
683e687, https://doi.org/10.3109/10428194.2012.728597.
[5] J.G. Gribben, N. Fowler, F. Morschhauser, Mechanisms of action of lenalido-
mide in B-cell non-hodgkin lymphoma, J. Clin. Oncol. 33 (25) (2015)
2803e2811, https://doi.org/10.1200/JCO.2014.59.5363.
[6] N. Mitsiades, C.S. Mitsiades, V. Poulaki, D. Chauhan, P.G. Richardson,
T. Hideshima, N.C. Munshi, S.P. Treon, K.C. Anderson, Apoptotic signaling
induced by immunomodulatory thalidomide analogs in human multiple
myeloma cells: therapeutic implications, Blood 99 (12) (2002) 4525e4530,
https://doi.org/10.1182/blood.V99.12.4525.
[7] G.H. Jackson, F.E. Davies, C. Pawlyn, D.A. Cairns, A. Striha, C. Collett,
A. Hockaday, J.R. Jones, B. Kishore, M. Garg, C.D. Williams, K. Karunanithi,
J. Lindsay, M.W. Jenner, G. Cook, N.H. Russell, M.F. Kaiser, M.T. Drayson,
R.G. Owen, W.M. Gregory, G.J. Morgan, Lenalidomide maintenance versus
observation for patients with newly diagnosed multiple myeloma (myeloma
XI): a multicentre, open-label, randomised, phase 3 trial, Lancet Oncol. 20 (1)
(2019) 57e73, https://doi.org/10.1016/S1470-2045(18)30687-9.
[24] C. Fitzmaurice, C. Allen, R.M. Barber, L. Barregard, Z.A. Bhutta, H. Brenner,
D.J. Dicker, O. Chimed-Orchir, R. Dandona, L. Dandona, T. Fleming,
M.H. Forouzanfar, J. Hancock, R.J. Hay, R. Hunter-Merrill, C. Huynh,
H.D. Hosgood, C.O. Johnson, J.B. Jonas, J. Khubchandani, G.A. Kumar, M. Kutz,
Q. Lan, H.J. Larson, X. Liang, S.S. Lim, A.D. Lopez, M.F. MacIntyre, L. Marczak,
N. Marquez, A.H. Mokdad, C. Pinho, F. Pourmalek, J.A. Salomon, J.R. Sanabria,
L. Sandar, B. Sartorius, S.M. Schwartz, K.A. Shackelford, K. Shibuya, J. Stanaway,
C. Steiner, J. Sun, K. Takahashi, S.E. Vollset, T. Vos, J.A. Wagner, H. Wang,
R. Westerman, H. Zeeb, L. Zoeckler, F. Abd-Allah, M.B. Ahmed, S. Alabed,
N.K. Alam, S.F. Aldhahri, G. Alem, M.A. Alemayohu, R. Ali, R. Al-Raddadi,
A. Amare, Y. Amoako, A. Artaman, H. Asayesh, N. Atnafu, A. Awasthi,
ꢀ
H.B. Saleem, A. Barac, N. Bedi, I. Bensenor, A. Berhane, E. Bernabe, B. Betsu,
~
ꢀ
A. Binagwaho, D. Boneya, I. Campos-Nonato, C. Castaneda-Orjuela, F. Catala-
ꢀ
Lopez, P. Chiang, C. Chibueze, A. Chitheer, J.Y. Choi, B. Cowie, S. Damtew, J. Das
Neves, S. Dey, S. Dharmaratne, P. Dhillon, E. Ding, T. Driscoll, D. Ekwueme,
A.Y. Endries, M. Farvid, F. Farzadfar, J. Fernandes, F. Fischer, T.T. Ghiwot,
A. Gebru, S. Gopalani, A. Hailu, M. Horino, N. Horita, A. Husseini, I. Huybrechts,
M. Inoue, F. Islami, M. Jakovljevic, S. James, M. Javanbakht, S.H. Jee,
A. Kasaeian, M.S. Kedir, Y.S. Khader, Y.H. Khang, D. Kim, J. Leigh, S. Linn,
R. Lunevicius, H.M.A. El Razek, R. Malekzadeh, D.C. Malta, W. Marcenes,
D. Markos, Y.A. Melaku, K.G. Meles, W. Mendoza, D.T. Mengiste, T.J. Meretoja,
T.R. Miller, K.A. Mohammad, A. Mohammadi, S. Mohammed, M. Moradi-
Lakeh, G. Nagel, D. Nand, Q. Le Nguyen, S. Nolte, F.A. Ogbo, K.E. Oladimeji,
E. Oren, M. Pa, E.K. Park, D.M. Pereira, D. Plass, M. Qorbani, A. Radfar, A. Rafay,
M. Rahman, S.M. Rana, K. Søreide, M. Satpathy, M. Sawhney, S.G. Sepanlou,
M.A. Shaikh, J. She, I. Shiue, H.R. Shore, M.G. Shrime, S. So, S. Soneji,
[8] C. Talati, D. Sallman, A. List, Lenalidomide: Myelodysplastic Syndromes with
Del(5q) and beyond. Seminars in Hematology, vol. 1, W.B. Saunders July, 2017,
pp. 159e166, https://doi.org/10.1053/j.seminhematol.2017.06.003.
[9] M.C. Okafor, Thalidomide for erythema nodosum leprosum and other appli-
cations, Pharmacotherapy 23 (4) (2003) 481e493, https://doi.org/10.1592/
phco.23.4.481.32115.
[10] N. Vargesson, Thalidomide-induced limb defects: resolving a 50-year-old
puzzle, Bioessays 31 (12) (2009) 1327e1336, https://doi.org/10.1002/
bies.200900103.
[11] S. Angers, T. Li, X. Yi, M.J. MacCoss, R.T. Moon, N. Zheng, Molecular architec-
ture and assembly of the DDB1-CUL4A ubiquitin ligase machinery, Nature 443
(7111) (2006) 590e593, https://doi.org/10.1038/nature05175.
[12] T. Ito, H. Ando, T. Suzuki, T. Ogura, K. Hotta, Y. Imamura, Y. Yamaguchi,
H. Handa, Identification of a primary target of thalidomide teratogenicity,
ꢀ
V. Stathopoulou, K. Stroumpoulis, M.B. Sufiyan, B.L. Sykes, R. Tabares-Seis-
dedos, F. Tadese, B.A. Tedla, G.A. Tessema, J.S. Thakur, B.X. Tran, K.N. Ukwaja,
B.S. Chudi Uzochukwu, V.V. Vlassov, E. Weiderpass, M. Wubshet Terefe,
H.G. Yebyo, H.H. Yimam, N. Yonemoto, M.Z. Younis, C. Yu, Z. Zaidi, M.E.S. Zaki,
Z.M. Zenebe, C.J.L. Murray, M. Naghavi, Global, regional, and national cancer
incidence, mortality, years of Life lost, years lived with disability, and
disability-adjusted life-years for 32 cancer groups, 1990 to 2015: a systematic
analysis for the global burden of disease study global burden of disease cancer
collaboration, JAMA Oncol. Am. Med. Assoc. April 1 (2017) 524e548, https://
doi.org/10.1001/jamaoncol.2016.5688.
Science
327
(5971)
(2010)
1345e1350,
https://doi.org/10.1126/
science.1177319.
[13] A.A. Akuffo, A.Y. Alontaga, R. Metcalf, M.S. Beatty, A. Becker, J.M. McDaniel,
R.S. Hesterberg, W.E. Goodheart, S. Gunawan, M. Ayaz, Y. Yang, M. Rezaul
Karim, M.E. Orobello, K. Daniel, W. Guida, J.A. Yoder, A.M. Rajadhyaksha,
€
E. Schonbrunn, H.R. Lawrence, N.J. Lawrence, P.K. Epling-Burnette, Ligand-
mediated protein degradation reveals functional conservation among
sequence variants of the CUL4-type E3 ligase substrate receptor cereblon,
J. Biol. Chem. 293 (16) (2018) 6187e6200, https://doi.org/10.1074/
jbc.M117.816868.
[25] World Health Organization (WHO). Newsroom, fact sheets, Cancer https://
www.who.int/news-room/fact-sheets/detail/cancer (accessed Apr 21, 2020).
[26] Q. Dai, C. Zhang, Z. Yuan, Q. Sun, Y. Jiang, Current discovery strategies for
hepatocellular carcinoma therapeutics, Expet Opin. Drug Discov. 15 (2) (2020)
243e258, https://doi.org/10.1080/17460441.2020.1696769.
[14] A.N. Lupas, H. Zhu, M. Korycinski, The thalidomide-binding domain of cere-
blon defines the CULT domain family and is a new member of the
PLoS Comput. Biol. 11 (1) (2015), https://doi.org/10.1371/
journal.pcbi.1004023.
b-tent fold,
[27] J.M. Llovet, S. Ricci, V. Mazzaferro, P. Hilgard, E. Gane, J.-F. Blanc, A.C. de Oli-
veira, A. Santoro, J.-L. Raoul, A. Forner, M. Schwartz, C. Porta, S. Zeuzem,
[15] P.P. Chamberlain, A. Lopez-Girona, K. Miller, G. Carmel, B. Pagarigan, B. Chie-
Leon, E. Rychak, L.G. Corral, Y.J. Ren, M. Wang, M. Riley, S.L. Delker, T. Ito,
H. Ando, T. Mori, Y. Hirano, H. Handa, T. Hakoshima, T.O. Daniel, B.E. Cathers,
Structure of the human cerebloneDDB1elenalidomide complex reveals basis
for responsiveness to thalidomide analogs, Nat. Struct. Mol. Biol. 21 (9) (2014)
€
L. Bolondi, T.F. Greten, P.R. Galle, J.-F. Seitz, I. Borbath, D. Haussinger,
T. Giannaris, M. Shan, M. Moscovici, D. Voliotis, J. Bruix, Sorafenib in advanced
hepatocellular carcinoma, N. Engl. J. Med. 359 (4) (2008) 378e390, https://
doi.org/10.1056/NEJMoa0708857.
17

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
MEG.0b013e3283036740.
[49] G.W. Muller, R. Chen, S.Y. Huang, L.G. Corral, L.M. Wong, R.T. Patterson,
Y. Chen, G. Kaplan, D.I. Stirling, Amino-substituted thalidomide analogs:
[28] S.G. Stewart, L.A. Ho, M.E. Polomska, A.T. Percival, G.C.T. Yeoh, Rapid evalua-
tion of antrodia camphorata natural products and derivatives in tumourigenic
liver progenitor cells with a novel cell proliferation assay, ChemMedChem 4
(10) (2009) 1657e1667, https://doi.org/10.1002/cmdc.200900238.
potent inhibitors of TNF-a production, Bioorg. Med. Chem. Lett 9 (11) (1999)
[29] K.N. Lowes, B.A. Brennan, G.C. Yeoh, J.K. Olynyk, Oval cell numbers in human
chronic liver diseases are directly related to disease severity, Am. J. Pathol. 154
(2) (1999) 537e541, https://doi.org/10.1016/S0002-9440(10)65299-6.
[30] B. Sun, M. Karin, Inflammation and liver tumorigenesis, Front. Med. 7 (2013)
242e254, https://doi.org/10.1007/s11684-013-0256-4.
1625e1630, https://doi.org/10.1016/S0960-894X(99)00250-4.
[50] B. Zhou, J. Hu, F. Xu, Z. Chen, L. Bai, E. Fernandez-Salas, M. Lin, L. Liu, C.-
Y. Yang, Y. Zhao, D. Mceachern, S. Przybranowski, B. Wen, D. Sun, S. Wang,
Discovery of a Small-Molecule Degrader of Bromodomain and Extra-terminal
(BET) Proteins with Picomolar Cellular Potencies and Capable of Achieving
Tumor Regression, 2017, https://doi.org/10.1021/acs.jmedchem.6b01816.
[31] B.C. Knight George Yeoh, TNF/LT
a double knockout mice display abnormal
€
inflammatory and regenerative responses to acute and chronic liver injury,
Cell Tissue Res. 319 (2005) 61e70, https://doi.org/10.1007/s00441-004-1003-
6.
[51] C. Steinebach, S. Lindner, N.D. Udeshi, D.C. Mani, H. Kehm, S. Kopff, S.A. Carr,
€
M. Gütschow, J. Kronke, Homo-PROTACs for the chemical knockdown of
cereblon, ACS Chem. Biol. 13 (9) (2018) 2771e2782, https://doi.org/10.1021/
acschembio.8b00693.
[32] B.B. Knight, G.C.T. Yeoh, K.L. Husk, T. Ly, L.J. Abraham, C. Yu, J.A. Rhim,
N. Fausto, Impaired preneoplastic changes and liver tumor formation in tumor
necrosis factor receptor type 1 knockout mice, J. Exp. Med. 192 (12) (2000)
1809e1818.
[33] G. He, D. Dhar, H. Nakagawa, J. Font-Burgada, H. Ogata, Y. Jiang, S. Shalapour,
E. Seki, S.E. Yost, K. Jepsen, K.A. Frazer, O. Harismendy, M. Hatziapostolou,
D. Iliopoulos, A. Suetsugu, R.M. Hoffman, R. Tateishi, K. Koike, M. Karin,
Identification of liver cancer progenitors whose malignant progression de-
pends on autocrine IL-6 signaling, Cell 155 (2) (2013) 384, https://doi.org/
10.1016/j.cell.2013.09.031.
[52] M. Chesi, G.M. Matthews, V.M. Garbitt, S.E. Palmer, J. Shortt, M. Lefebure,
A.K. Stewart, R.W. Johnstone, P. Leif Bergsagel, Drug response in a genetically
engineered Mouse model of multiple myeloma is predictive of clinical effi-
cacy, Blood 120 (2) (2012) 376e385, https://doi.org/10.1182/blood-2012-02-
412783.
[53] E.C. Fink, M. McConkey, D.N. Adams, S.D. Haldar, J.A. Kennedy, A.A. Guirguis,
N.D. Udeshi, D.R. Mani, M. Chen, B. Liddicoat, T. Svinkina, A.T. Nguyen,
S.A. Carr, B.L. Ebert, CrbnI391V is sufficient to confer in vivo sensitivity to
thalidomide and its derivatives in mice, Blood 132 (14) (2018) 1535e1544,
https://doi.org/10.1182/blood-2018-05-852798.
[34] S.G. Stewart, D. Spagnolo, M.E. Polomska, M. Sin, M. Karimi, L.J. Abraham,
Synthesis and TNF expression inhibitory properties of new thalidomide ana-
logues derived via heck cross coupling, Bioorg. Med. Chem. Lett 17 (21) (2007)
5819e5824, https://doi.org/10.1016/j.bmcl.2007.08.042.
[54] P. Brodt, Role of the microenvironment in liver metastasis: from pre-to
prometastatic niches, Clin. Canc. Res. 22 (24) (2016), https://doi.org/
10.1158/1078-0432.CCR-16-0460.
[35] S.G. Stewart, C.J. Braun, S.-L. Ng, M.E. Polomska, M. Karimi, L.J. Abraham, New
thalidomide analogues derived through sonogashira or Suzuki reactions and
their TNF expression inhibition profiles, Bioorg. Med. Chem. 18 (2) (2010)
650e662, https://doi.org/10.1016/j.bmc.2009.12.001.
[36] S.Y. Yeung, S. Kampmann, K.A. Stubbs, B.W. Skelton, B.J. Kaskow, L.J. Abraham,
S.G. Stewart, M.-C. Viaud-Massuard, Y. Hashimoto, Novel thalidomide ana-
logues with potent NFkB and TNF expression inhibition, Med. Chem. Com-
mun. 2 (2011) 1073e1078, https://doi.org/10.1039/c1md00184a.
[37] J.A. Keifer, D.C. Guttridge, B.P. Ashburner, A.S. Baldwin, Inhibition of NF-ΚB
[55] J.W. Lee, M.L. Stone, P.M. Porrett, S.K. Thomas, C.A. Komar, J.H. Li, D. Delman,
K. Graham, W.L. Gladney, X. Hua, T.A. Black, A.L. Chien, K.S. Majmundar,
J.C. Thompson, S.S. Yee, M.H. O’Hara, C. Aggarwal, D. Xin, A. Shaked, M. Gao,
D. Liu, M.J. Borad, R.K. Ramanathan, E.L. Carpenter, A. Ji, M.C. de Beer, F.C. de
Beer, N.R. Webb, G.L. Beatty, Hepatocytes Direct the Formation of a Pro-
metastatic Niche in the Liver. Nature, vol. 14, Nature Publishing Group
March, 2019, pp. 249e252, https://doi.org/10.1038/s41586-019-1004-y.
[56] S. Yang, G. Liu, Targeting the RAS/RAF/MEK/ERK Pathway in Hepatocellular
Carcinoma. Oncology Letters, vol. 1, Spandidos Publications March, 2017,
pp. 1041e1047, https://doi.org/10.3892/ol.2017.5557.
[57] T. Luedde, R.F. Schwabe, NF-ΚB in the liver-linking injury, fibrosis and he-
patocellular carcinoma, Nat. Rev. Gastroenterol. Hepatol. (2011) 108e118,
https://doi.org/10.1038/nrgastro.2010.213. NIH Public Access February.
[58] J. Li, G.K.K. Lau, L. Chen, S. sui Dong, H.Y. Lan, X.R. Huang, Y. Li, J.M. Luk,
Y.F. Yuan, X. yuan Guan, Interleukin 17a promotes hepatocellular carcinoma
metastasis via NF-ΚB induced matrix metalloproteinases 2 and 9 expression,
PloS One 6 (7) (2011), https://doi.org/10.1371/journal.pone.0021816.
[59] J. Svinka, W. Mikulits, R. Eferl, STAT3 in hepatocellular carcinoma: new per-
activity by thalidomide through suppression of I
Chem. 276 (25) (2001) 22382e22387,
jbc.M100938200.
[38] W. Tan, X. Luo, W. Li, J. Zhong, J. Cao, S. Zhu, X. Chen, R. Zhou, C. Shang,
Y. Chen, TNF- is a potential therapeutic target to overcome sorafenib resis-
kB kinase activity, J. Biol.
https://doi.org/10.1074/
a
tance in hepatocellular carcinoma, EBioMedicine 40 (2019) 446e456, https://
doi.org/10.1016/j.ebiom.2018.12.047.
[39] B. Knight, G.C. Yeoh, TNF/LT
a double knockout mice display abnormal in-
flammatory and regenerative responses to acute and chronic liver injury, Cell
Tissue Res. 319 (1) (2005) 61e70, https://doi.org/10.1007/s00441-004-1003-
6.
spectives, Hepatic Oncol
hep.13.7.
1 (1) (2014) 107e120, https://doi.org/10.2217/
[40] S.M. Capitosti, T.P. Hansen, M.L. Brown, Facile synthesis of an azido-labeled
thalidomide analogue, Org. Lett. 5 (16) (2003) 2865e2867, https://doi.org/
10.1021/ol034906w.
[41] W.T. Schwartz, Process for the Preparation of Phenoxy Phthalic Anhydrides,
US4827000, 1987.
[60] G. Regis, S. Pensa, D. Boselli, F. Novelli, V. Poli, Ups and Downs: the STAT1:
STAT3 Seesaw of Interferon and Gp130 Receptor Signalling. Seminars in Cell
and Developmental Biology, Elsevier Ltd August 1, 2008, pp. 351e359, https://
doi.org/10.1016/j.semcdb.2008.06.004.
[61] S.S. Zalesskiy, V.P. Ananikov, Pd 2(dba) 3 as a precursor of soluble metal
complexes and nanoparticles: determination of palladium active species for
catalysis and synthesis, Organometallics 31 (6) (2012) 2302e2309, https://
doi.org/10.1021/om201217r.
[42] T. Itoh, T. Mase, A general palladium-catalyzed coupling of aryl bromides/
triflates and thiols, Org. Lett.
6 (24) (2004) 4587e4590, https://doi.org/
10.1021/ol047996t.
[43] D. Xiao, Y. jie Wang, X. bei Hu, W. juan Kan, Q. Zhang, X. Jiang, Y. bo Zhou, J. Li,
W. Lu, Design, synthesis and biological evaluation of the thioether-containing
lenalidomide analogs with anti-proliferative activities, Eur. J. Med. Chem. 176
(2019) 419e430, https://doi.org/10.1016/j.ejmech.2019.05.035.
[44] Y. Wang, T. Mi, Y. Li, W. Kan, G. Xu, J. Li, Y. Zhou, J. Li, X. Jiang, Design, synthesis
and biological evaluation of thioether-containing lenalidomide and pomali-
domide derivatives with anti-multiple myeloma activity, Eur. J. Med. Chem.
(2020), 112912, https://doi.org/10.1016/j.ejmech.2020.112912.
[62] W.L.F. Armarego, C. Chai, Purification of Laboratory Chemicals, sixth ed.,
Butterworth-Heinemann, 2009.
[63] G.R. Fulmer, A.J.
M Miller, N.H. Sherden, H.E. Gottlieb, A. Nudelman,
B.M. Stoltz, J.E. Bercaw, K.I. Goldberg, M. Beckman, NMR chemical shifts of
trace impurities: common laboratory solvents, organics, and gases in
deuterated solvents relevant to the organometallic chemist, Organometallics
29 (9) (2010) 2176e2179, https://doi.org/10.1021/om100106e.
[64] J. Huang, S. Rauscher, G. Nawrocki, T. Ran, M. Feig, B.L. de Groot,
H. Grubmuller, A.D. MacKerell Jr., CHARMM36m: an improved force field for
folded and intrinsically disordered proteins, Nat. Methods 14 (2017) 71e73,
https://doi.org/10.1038/nmeth.4067.
[45] L. Wang, W. He, Z. Yu, Transition-metal mediated carbonesulfur bond acti-
vation and transformations, Chem. Soc. Rev. 42 (2) (2013) 599e621, https://
doi.org/10.1039/C2CS35323G.
[46] Y. Wang, T. Mi, Y. Li, W. Kan, G. Xu, J. Li, Y. Zhou, J. Li, X. Jiang, Design, synthesis
and biological evaluation of thioether-containing lenalidomide and pomali-
domide derivatives with anti-multiple myeloma activity, Eur. J. Med. Chem.
(2020), 112912, https://doi.org/10.1016/j.ejmech.2020.112912.
[47] J.D. Hansen, M. Correa, M.A. Nagy, M. Alexander, V. Plantevin, V. Grant,
B. Whitefield, D. Huang, T. Kercher, R. Harris, R.K. Narla, J. Leisten, Y. Tang,
M. Moghaddam, K. Ebinger, J. Piccotti, C.G. Havens, B. Cathers, J. Carmichael,
T. Daniel, R. Vessey, L.G. Hamann, K. Leftheris, D. Mendy, F. Baculi, L.A. LeBrun,
G. Khambatta, A. Lopez-Girona, Discovery of CRBN E3 ligase modulator CC-
92480 for the treatment of relapsed and refractory multiple myeloma,
J. Med. Chem. 63 (13) (2020) 6648e6676, https://doi.org/10.1021/
acs.jmedchem.9b01928.
[65] I.S. Joung, T.E. Cheatham, Determination of alkali and halide monovalent ion
parameters for use in explicitly solvated biomolecular simulations, J. Phys.
Chem. B 112 (2008) 9020e9041, https://doi.org/10.1021/jp8001614.
[66] J.C. Phillips, R. Braun, W. Wang, J. Gumbart, E. Tajkhorshid, E. Villa, C. Chipot,
ꢀ
R.D. Skeel, L. Kale, K. Schulten, Scalable molecular dynamics with NAMD,
J. Comput. Chem. 26 (2005) 1781e1802, https://doi.org/10.1002/jcc.20289.
[67] U. Essmann, L. Perera, M.L. Berkowitz, A smooth particle mesh Ewald method,
J. Chem. Phys. 103 (1995) 8577e8593, https://doi.org/10.1063/1.470117.
[68] W. Humphrey, A. Dalke, K. Schulten, VMD: visual molecular dynamics, J. Mol.
Graph. 14 (1996) 33e38, https://doi.org/10.1016/0263-7855(96)00018-5.
[69] Gaussian 09, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Petersson, H. Nakatsuji, X. Li, M.
Caricato, A. Marenich, J. Bloino, B. G. Janesko, R. Gomperts, B. Mennucci, H. P.
Hratchian, J. V. Ortiz, A. F. Izmaylov, J. L. Sonnenberg, D. Williams-Young, F.
Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson, D.
Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M.
[48] M. Pinter, M. Wichlas, K. Schmid, C. Plank, C. Müller, F. Wrba, M. Peck-
Radosavljevic, Thalidomide in advanced hepatocellular carcinoma as anti-
angiogenic treatment approach: a phase I/II trial, Eur. J. Gastroenterol. Hep-
atol.
20
(10)
(2008)
1012e1019,
https://doi.org/10.1097/
18

M.J. Nutt, Y.S. Yee, A. Buyan et al.
European Journal of Medicinal Chemistry 217 (2021) 113353
Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, K. Throssell, J. A. Montgomery, Jr, J. E. Peralta, F.
Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T.
Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S.
Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene, C. Adamo, R. Cammi, J. W.
Ochterski, R. L. Martin, K. Morokuma, O. Farkas, J. B. Foresman, and D. J. Fox,
Gaussian, Inc., Wallingford CT, 2016.
[70] E.S. Fischer, K. Bohm, J.R. Lydeard, H. Yang, M.B. Stadler, S. Cavadini, et al.,
Structure of the DDB1-CRBN E3 ubiquitin ligase in complex with thalidomide,
Nature 512 (7512) (2014) 49e53, https://doi.org/10.1038/nature13527.
19