Proteolysis Targeting Chimeras for the Selective Degradation of Mcl-1/Bcl-2 Derived from Nonselective Target Binding Ligands

Article abstract of DOI:10.1021/acs.jmedchem.9b00919

Proteolysis targeting chimera (PROTAC) recruits an E3 ligase to a target protein to induce its ubiquitination and subsequent degradation. We reported success in the development of two PROTACs (C3 and C5) that potently and selectively induce the degradation of Mcl-1 and Bcl-2 (DC₅₀ = 0.7 and 3.0 μM), respectively, by introducing the E3 ligase cereblon-binding ligand pomalidomide to Mcl-1/Bcl-2 dual inhibitors S1-6 and Nap-1 with micromolar-range affinity. C3-induced Mcl-1 ubiquitination translated into much more lethality in Mcl-1-dependent H23 cells than the most potent Mcl-1 occupancy-based inhibitor A-1210477 with nanomolar-range affinity. Moreover, structure-activity relationship analysis and molecular dynamic simulations discovered the structural basis for turning nonselective or promiscuous Bcl-2 family ligands into selective PROTACs. C3 and C5 exhibited reversible depletion in living cells, which provides a new potent toolkit for gain-of-function studies to probe the dynamic roles of Bcl-2 and Mcl-1 in apoptosis networks.

Full text of DOI:10.1021/acs.jmedchem.9b00919

Subscriber access provided by UNIV AUTONOMA DE COAHUILA UADEC
Article
Proteolysis Targeting Chimeras for the selective degradation
of Mcl-1/Bcl-2 derived from nonselective target binding ligands
Ziqian Wang, Nianzhe He, Zongwei Guo, Cuili Niu, Ting Song, Yafei Guo,
Keke Cao, Anhui Wang, Junjie Zhu, Xiaodong Zhang, and Zhichao Zhang
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b00919 • Publication Date (Web): 07 Aug 2019
Downloaded from pubs.acs.org on August 7, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Proteolysis Targeting Chimeras for the selective
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
degradation of Mcl-1/Bcl-2 derived from
nonselective target binding ligands
Ziqian Wang,a Nianzhe He,a Zongwei Guo,b Cuili Niu,a Ting Song,a Yafei Guo,b Keke
Cao,b Anhui Wang,a Junjie Zhu,a Xiaodong Zhang a and Zhichao Zhang a*
aState Key Laboratory of Fine Chemicals, Zhang Dayu School of Chemistry, Dalian
University of Technology, Dalian, Liaoning (China), 116024.
bSchool of Life Science and Technology, Dalian University of Technology, Dalian,
Liaoning (China), 116024.
KEYWORDS: Proteolysis Targeting Chimeras; protein-protein interaction; Bcl-2 family
protein; linker; selectivity
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 43
1
2
3
4
Abstract
5
6
7
8
Proteolysis Targeting Chimera (PROTAC) recruits an E3 ligase to a target protein to induce
its ubiquitination and subsequent degradation. We reported the success in the development
of two PROTACs (C3 and C5) that potently and selectively induce the degradation of Mcl-
1 and Bcl-2 (DC50 = 0.7 and 3.0 μM), respectively, by introducing the E3 ligase cereblon
(CRBN)-binding ligand pomalidomide to Mcl-1/Bcl-2 dual inhibitors S1-6 and Nap-1 with
μM-range affinity. C3-induced Mcl-1 ubiquitination translated into much more lethality in
Mcl-1-dependent H23 cells than the most potent Mcl-1 occupancy-based inhibitor A-
1210477 with nM-range affinity. Moreover, structure-activity relationship (SAR) analysis
and molecular dynamic (MD) simulations discovered the structural basis for turning
nonselective or promiscuous Bcl-2 family ligands into selective PROTACs. C3 and C5
exhibited reversible depletion in living cells, which provides a new potent toolkit for gain-
of-function studies to probe the dynamic roles of Bcl-2 and Mcl-1 in apoptosis networks.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
ACS Paragon Plus Environment

Page 3 of 43
Journal of Medicinal Chemistry
1
2
3
4
Introduction
5
6
7
8
Occupancy-based pharmaceutical strategies need high local inhibitory percentages
(usually 90-95 %) of disease-related proteins, which is necessary to assure therapeutic
efficacy1-2. Therefore, a heavy burden has been imposed on developing high-affinity small
molecule ligands, which is always forlorn for targets mediating protein-protein interactions
(PPIs).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The recent development of Proteolysis Targeting Chimeras (PROTACs), which
comprise a ligand binding an E3 ubiquitin ligase conjugated to a small molecule warhead
binding the target protein, offer a new pharmacological strategy to solve these limitations1-5.
PROTACs trigger target ubiquitination and subsequent proteasomal degradation by
recruiting targets into spatial proximity with E3 ubiquitin ligases1-7. Thus, PROTACs can
act at substoichiometric occupancies for the targets in a catalytic mechanism, do not require
excessive drug concentrations to ensure occupancy-based inhibition and can avoid adverse
side effects8,9.
To date, PROTACs have succeeded in degrading kinases10-15, bromodomains and extra-
terminal (BET) proteins16-20, and hormone receptors21,22 by using warheads that show
nanomolar affinity for these targets. However, PROTACs for α-helix-mediated PPI target
degradation are still rare23-25. One of the major challenges is that binding ligands for these
PPI targets are usually large hydrophobic molecules (Mw > 800 and LogP > 5), as they are
required to make discontinuous, noncovalent contacts over a large and flat binding surface
area (>800 Å2)26-28. Therefore, it is difficult for PROTACs obtained by conjugating these
PPI inhibitors with a hydrophobic E3 ligase ligand (Mw > 250)1-5 via an alkyl or PEG
linker (usually Mw = 100-200)1-5 to exhibit acceptable solubility and membrane
permeability.
Nevertheless, PPIs are involved in most cellular processes, including differentiation,
apoptosis, signal transduction and transcription29. Therefore, the modulation of PPIs is
considered as a promising strategy toward next-generation therapeutics27-29. For example,
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 43
1
2
3
4
5
6
7
8
9
Bcl-2 family proteins play essential roles in the regulation of intrinsic apoptosis via α-helix-
mediated PPIs between pro- and anti-apoptotic members30-32. Anti-apoptotic proteins,
which are represented by Mcl-1 and Bcl-2, are considered as attractive targets for potential
cancer treatments33.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Recently, researchers have demonstrated that the basis of degradation efficiency rely on
protein-protein interactions between the E3 ubiquitin ligase and the target protein, rather
than the binding affinity of ligands34,35. Therefore, the potent and selective PROTACs for
degrading PPI targets can be developed using target binding ligands with relative lower
Mw via modulating the length and conformation of the linkers to compensate for the low-
affinity of ligands, leading to a dramatic enhancement of target protein removal.
In this article, we constructed two series of PROTACs, H1-H5 and C1-C6, by
introducing the E3 ligase cereblon (CRBN)-binding ligand pomalidomide to Bcl-2/Mcl-1
dual inhibitors S1-636-38 and Nap-139 (Fig. 1a), which exhibits a lower Mw (approximately
500) and a lower binding affinity for both targets in the μM-range, via alkyl and PEG
linkers of different lengths. Among them, PROTACs C3 and C5 were observed to potently
and selectively induce the ubiquitination and proteasomal degradation of Mcl-1 and Bcl-2
in cellulo, respectively, by hijacking a CRBN ubiquitin ligase to form a ternary complex
with the target protein. Moreover, structure-activity relationship (SAR) analysis and
molecular dynamic (MD) simulations in a certain extent revealed the structural basis for
turning nonselective or promiscuous Bcl-2 family ligands into more selective degraders by
changing the length and conformation of their linkers, which is valuable for the design of
other PROTACs. Moreover, C3 and C5 exhibited reversible depletion in living cells, which
provides a new potent toolkit for gain-of-function studies to probe the dynamic roles of
Bcl-2 and Mcl-1 in apoptosis networks.
Results and Discussion
4
ACS Paragon Plus Environment

Page 5 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. (a) Chemical structures, molecular weights and NMR-derived binding mode of
the parent inhibitors S1-6 (top, left), Nap-1 (bottom, left) and intermediates 1 (top, right),
2 (bottom, right) in complex with Mcl-1 (PDB ID: 2NLA). The pink and yellow surface
highlight p2 and p3 binding pockets, and red arrows highlight exit vectors for linking. (b)
Chemical structures and molecular weights of H1-H5, C1-C6, C3-Me, C5-Me and S1-5.
Herein, we selected two Bcl-2 family inhibitors, S1-6 and Nap-1 (Fig. 1a), whose
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 43
1
2
3
4
5
6
7
8
9
opposite binding orientations to the BH3 groove of Bcl-2 proteins have been identified by
2D and 3D NMR in our previous studies36-39. For S1-6, the solvent exposed 3-cyano group
located in the P3 pocket of the BH3 groove provides a suitable point for linker attachment
without tethering the binding properties (Fig. 1b, top). We then replaced the cyano group
with the 3-aminopropanoic acid methyl ester group, which yielded compound 1 (Fig. 1a).
For Nap-1 (Fig. 1b, bottom), the 4-tert-butylbenzoic acid group bond to the P2 pocket was
replaced by hydrophobic alkyl chains for further linker connection, yielding compound 2
(Fig. 1a).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To evaluate the effects of distance and spatial location between the ligands and the E3
ligase, five different linkers of varying composition and length were attached to 1 and 2
via amide bond formation. Finally, the CRBN binding ligand pomalidomide was attached,
yielding PROTACs H1-H5 and C1-C6 (Fig. 1c).
Scheme 1. Synthesis routes of 1, 2, H1-H5, C1-C6, C3-Me and C5-Me. (a) Reaction
conditions: i) methyl 3-aminopropanoate hydrochloride, NaOH, CH3CN/H2O, rt; ii) NaOH,
THF/H2O, rt; iii) 3, DIEA, HATU, DCM, rt. (b) Reaction conditions: i) 4-Bromothiopheno,
NEt3, DMF, reflux; ii) ethylenediamine, NEt3, EtOH, reflux; iii) 6-Methoxy-6-
oxohexanoic acid, DIEA, HATU, DCM, rt; iv) NaOH, THF/H2O, rt; v) 3/3-Me, DIEA,
HATU, DCM, rt.
6
ACS Paragon Plus Environment

Page 7 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
To achieve the desired ligands, a generally applicable three-step synthetic strategy was
devised (Scheme 1). First, intermediate compounds 1 and 2 were synthesized according to
earlier reported routes36-39. Second, the linkers bearing a free amine group at one end and
a Boc-protected amine group at the other end were attached to the pomalidomide via SNAr
reactions40. In the third step, deprotection of the Boc-protected amine and subsequent
HATU-mediated amide bond formation with the carboxylic acid in intermediate molecules
1 and 2 afforded the desired PROTAC compounds, respectively.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Binding affinities of S1-6, Nap-1, intermediates 1, 2, C1-C5 and H1-H5 with
Bcl-2 and Mcl-1 determined by ELISA, data are expressed as the mean ± s.e.m.
Compounds
IC50 (μM)
Mcl-1
Bcl-2
S1-6
Nap-1
1
2.03 ± 1.02
4.45 ± 1.53
3.46 ± 0.94
5.95 ± 1.49
7.89 ± 2.47
1.51 ± 0.39
0.78 ± 0.12
15.12 ± 3.85
11.81 ± 2.48
5.40 ± 1.25
10.64 ± 4.19
2.31 ± 0.87
3.18 ± 1.39
4.52 ± 1.28
4.78 ± 2.06
15.21 ± 4.33
10.02 ± 2.90
0.54 ± 0.27
13.47 ± 4.57
4.94 ± 1.36
7.66 ± 2.28
10.48 ± 3.66
2
C1
C2
C3
C4
C5
C6
H1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 43
1
2
3
4
5
6
7
8
9
H2
H3
H4
H5
9.95 ± 2.72
7.65 ± 2.56
16.01 ± 4.88
2.29 ± 1.33
9.71 ± 2.13
1.94 ± 0.58
11.65 ± 4.44
2.24 ± 0.89
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Enzyme-linked immunosorbent assay (ELISA) experiments were performed to measure
the binding abilities of the PROTAC molecules toward Mcl-1 and Bcl-2 (Table 1 and Fig.
S1). Compared with the parent inhibitors S1-6 and Nap-1, intermediate molecules 1, 2 and
PROTACs C1-C6 and H1-H5 showed similar overall binding affinities in the μM range
with Mcl-1 and Bcl-2, denoting that the binding modes were still conserved.
To assess the extent of PROTAC-mediated protein removal, we analyzed the protein
levels of Mcl-1 and Bcl-2 after a 12 h treatment with C1-C6 or H1-H5 by immunoblotting
in Hela cancer cells. Hela cells were chosen because they are less susceptible to the
cytotoxic effects of Bcl-2 family protein knockdown or inhibition. The parent inhibitors
S1-6 and Nap-1 were included as controls (Fig. 2). No significant depletion in Mcl-1 or
Bcl-2 was observed in Hela cells upon treatment with 10 μM PROTACs H1-H5 (< 20 %),
and Mcl-1 levels increased beyond the vehicle control levels in some cases, which
suggested that H1-H5 function more as protein inhibitors than as degraders due to the Mcl-
1 upregulation under inhibition stress. The bands at high mass in addition of H4 and H5
were not ubiquitin adducts of Mcl-1 (Fig. S2). In contrast, PROTACs C3 and C5 induced
a marked concentration-dependent depletion of Mcl-1 and Bcl-2, respectively. Compound
C3 selectively induced a 70 % decrease in Mcl-1 levels at 1 μM and progressively
downregulated it to less than 10 % at 10 μM. In contrast, Bcl-2 protein was not depleted,
even upon treatment with 10 μM C3. In comparison, C5 served as a selective Bcl-2 reducer
that reduced protein levels progressively from 10 % to 90 % as the dose increased from 1
to 10 μM. As a control, the parent inhibitors Nap-1 and S1-6 showed no depletion of either
8
ACS Paragon Plus Environment

Page 9 of 43
Journal of Medicinal Chemistry
1
2
3
4
Mcl-1 or Bcl-2 at 10 μM.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. PROTACs downregulated the protein levels of their respective targets. Mcl-1
and Bcl-2 protein levels were analyzed by Western blotting in Hela cells treated with (a)
H1-H5 and (b) C1-C6 for 12 h. All western blot figures represent the results from nꢀ=ꢀ3
independent samples.
Consequently, time- and dose-dependent depletion of Mcl-1/Bcl-2 by C3 and C5 were
tested (Fig. 3 and Fig. S3,S4). Our previously reported Mcl-1 degrader, S1-538 (Fig. 1a),
was added as a control. S1-5 could form hydrogen bonds with His224 from Mcl-1 to drive
a helical QRN conformation and finally lead to the degradation of Mcl-1 by benefiting
from its binding ability to the E3 ligase Mule. As shown in Fig. 3a and Fig. S3, Mcl-1/Bcl-2
levels in Hela cells were monitored over a 24-hour time course upon treatment with 10 μM
of compounds C3, C5 and S1-5. Progressive removal of Bcl-2 family proteins over time
was observed for the three compounds, but with different selectivity and efficiency. Ten
micromolar C3 began to markedly deplete Mcl-1 at 8 h and reached the maximal reduction
at 12 h, when approximately 80 % of the protein had been removed. After that, Mcl-1
recovery occurred probably owing to the possibility of compound instability, for the
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 43
1
2
3
4
5
6
7
8
9
mRNA level over 0-24 h (Fig. S5). In contrast, Bcl-2 remained constant during the time
course. For C5, remarkable Bcl-2 but not Mcl-1 depletion was observed at 2 h, and almost
all the Bcl-2 proteins were depleted after 12 h of treatment. Bcl-2 abundance also recovered
after 16 h. For S1-5, selective depletion of Mcl-1 was observed at 12 h and 80 % of Mcl-1
was depleted at 24 h. Reassuringly, all these compounds exhibited dose-dependent and
selective depletion activities (Fig. 3b and Fig. S4). Compound C3 exhibited Mcl-1 removal
activity with a DC50 value of 0.7 μM, while the DC50 value of C5 against Bcl-2 was 3.0
μM at 12 h. The DC50 value of S1-5 for Mcl-1 was much higher than that of C3, which was
3.6 μM. The much more efficient Mcl-1 depletion induced by C3 than S1-5 was further
verified in K562 cells (Fig. S6).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Interestingly, compared with ELISA data, we found an inconsistency between binding
affinity and degradation potency, and, more importantly, the degradation selectivity. For
example, even though all 10 PROTACs (C1-C6 and H1-H5) exhibited a similar μM
binding affinity to Mcl-1 and Bcl-2, only C3 and C5 degraded Mcl-1 and Bcl-2,
respectively, which demonstrated that potency and selectivity of the PROTACs can be
optimized via modulating the length and conformation of the linkers.
To evaluate the specificity of C3/C5-induced degradation, we tested the protein level of
another of Bcl-2 family member, Bcl-xl, and the potential off-targets of some Bcl-2 family
inhibitors (MDM2 and tubulin)39,41, which has been reported by our group. None of these
protein levels decreased upon addition of 10 μM C3 or C5, further validated the specificity
of C3/C5-induced Mcl-1/Bcl-2 degradation (Fig. S7).
Despite proving selective degradation within the Bcl-2 family members and some
potential off-targets of Bcl-2 family inhibitors, we cannot exclude the degradation of
unrelated targets in the global proteome. Therefore, a global proteomic experiment would
help assess the potential efficacy and safety risks of drug discovery program in the future.
10
ACS Paragon Plus Environment

Page 11 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. PROTACs induced time- and concentration-dependent selective depletion of
Mcl-1 or Bcl-2 protein in Hela cells. Protein levels were visualized by immunoblotting and
quantified relative to the DMSO control. (a) Protein levels after treatment with 10 μM C3,
C5 or S1-5 over a 0-24 h time gradient. (b) Protein levels after treatment with a 0-10 μM
concentration gradient of C3, C5 or S1-5 for 12 h. All western blot figures represent the
results from nꢀ=ꢀ3 independent samples, the replicates are shown as Fig. S3, S4 in SI. Error
bars show mean ± s.d. (n=3).
Next, we determined the mechanism of the PROTAC-mediated Mcl-1/Bcl-2 degradation.
First, replacing the pomalidomide in C3 and C5 with N-methylated glutarimide, which
yields C3-Me and C5-Me (Fig. 1b, for synthesis route see Scheme 1b), abolished their
capability of inducing Mcl-1/Bcl-2 degradation (Fig. 4a,b). Simultaneously, pretreatment
with the proteasome inhibitor MG-132, the parent Bcl-2 inhibitor Nap-1 and the CRBN
inhibitor pomalidomide effectively blocked both C3-induced Mcl-1 depletion (Fig. 4a) and
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 43
1
2
3
4
5
6
7
8
9
C5-induced Bcl-2 depletion (Fig. 4b). Second, a CRBN pull-down assay was performed to
trap potential ternary complex members in the presence of PROTAC by co-
immunoprecipitation. When a whole-cell lysate from Hela cells was incubated with
different concentrations of compounds C3 or C5, selective interactions between Mcl-1 or
Bcl-2 with CRBN and dose-dependent CRBN recruitment were detected (Fig. 4c),
consistent with their degrading selectivity. In contrast, no interactions between either Mcl-1
or Bcl-2 with CRBN were detected upon addition of the invalid PROTAC H3 (Fig. S8).
All together, these data provided clear evidence that C3 and C5 mediated target protein
degradation in a bona fide PROTAC-based mechanism, in which the degrading potency
and selectivity was dependent on a ternary complex comprised of the target protein,
PROTAC, and CRBN.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. The stable ternary complexes between Mcl-1/Bcl-2, CRBN and the PROTACs
are required for the superstoichiometric ubiquitination and degradation of target proteins.
(a,b) Immunoblot for Mcl-1 (a) or Bcl-2 (b) after pretreatment with 50 μM MG132, 50 μM
pomalidomide or 50 μM Nap-1, followed by PROTAC treatment (10 μM C3 or C5) in
12
ACS Paragon Plus Environment

Page 13 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Hela cells. (c) Stable interactions between Mcl-1/Bcl-2 and CRBN were induced by the
addition of the PROTACs in Hela cells. Immobilized CRBN was used as a bait to trap Mcl-
1 and Bcl-2 in the presence of a concentration gradient of C3 or C5. Protein levels of the
trapped proteins were visualized by immunobloting in the pull-down reactions. All western
blot figures represent the results from nꢀ=ꢀ3 independent samples.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Favourable protein-protein interactions stabilize the C3/C5-induced complexes
between Mcl-1/Bcl-2 and CRBN. (a) CRBN, C3 and Mcl-1 and (b) CRBN, C5 and Bcl-2
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 43
1
2
3
4
were docked and a 50-ns molecular dynamics simulation relaxed the structure (Mcl-1 PDB
5
6
ID: 2PQK; Bcl-2 PDB ID: 2XA0; CRBN PDB ID:4TZ4)42-44
.
7
8
9
Next, we sought a molecular understanding of how Mcl-1 and Bcl-2 formed high-
stability C3- and C5-induced ternary complexes with CRBN, respectively. To accomplish
this, we docked previously published structures of Mcl-1/Bcl-2 and CRBN (PDB:
2PQK/2XA0 and 4TZ4)42-44 with the C3 and C5 PROTACs, respectively, and applied 50-
ns molecular dynamics simulations to relax each structure into a low energy conformation
(Fig. 5 and Fig. S9).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
According to the model shown in Fig. 5a, when the C3 warhead engages Mcl-1, the C3
alkyl linker region exhibits hydrophobic interactions with Histidine 252 in Mcl-1, and thus
adopts a kinked conformation to accommodate the Mcl-1 α3 helix near CRBN. At the PPI
interface between Mcl-1 and CRBN, Threonine 226 from Mcl-1 provides close,
hydrophilic contacts to Histidine 378 in CRBN, and Histidine 224 from Mcl-1 creates an
additional π-π stacking interaction with Phenylalanine 150 from CRBN. On the other hand,
Bcl-2 forms a stable C5-induced ternary complex with CRBN (Fig. 4b). The C5 PEG linker
region adopts a linear conformation and points straight to the solvent, accommodating the
Bcl-2 α4 helix near CRBN, indicating that the PPI interface is located on the Bcl-2 α4
helix. H-bonds are formed between Aspartic acid 111 from Bcl-2 and Serine 126 from
CRBN; Glutamic acid 136 and Arginine 139 from Bcl-2 and Serine 420 from CRBN;
Arginine 129 from Bcl-2 and Histidine 378 from CRBN; Threonine 122 from Bcl-2 and
Glutamic acid 377 from CRBN; Threonine 132 from Bcl-2 and Tryptophan 400 from
CRBN; and Glutamic acid 135 from Bcl-2 and Glutamine 390 from CRBN. In contrast,
Mcl-1-C5-CRBN and Bcl-2-C3-CRBN showed no favourable energy conformation during
50-ns molecular dynamics simulations (Fig. S9).
Our MD illustrated the structure basis by which C3 and C5 induce ternary complexes
between different target proteins and the E3 ligase, which offered a mechanism for the
14
ACS Paragon Plus Environment

Page 15 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
above-mentioned inconsistency between binding affinity and degradation potency and
selectivity. For example, the PROTAC-mediated protein-protein interfaces in the
CRBN/Mcl-1 complex is not identical to those in CRBN/Bcl-2, illustrating the selectivity
that C3 and C5 show for Mcl-1 and Bcl-2 degradation, respectively. Similarly, the lack of
C1- and C2-induced degradation of Mcl-1 is likely due to steric clashes or hydrophobic
collapse of their shorter hydrophobic linkers compared to C3. The longer linker in C4 is
also unfavorable for ternary complex formation. Meanwhile, PROTACs H1-H5, which
connect to CRBN in an opposite orientation compared with C1-C5, showed almost no
degradation activities, suggesting that conjugating the CRBN ligand near the Bcl-2 family
protein P2 pocket can recruit the target proteins and E3 ligase into productive proximity.
As Bcl-2 family protein degradation can be potently observed with the PROTACs, we
next sought to determine their cellular effects compared with the predominantly
occupancy-based inhibitors. In a cell viability assay (Fig. 6a), we determined that the
selective Mcl-1 degrader C3 (DC50 = 0.76 μM) is much more active against the Mcl-1-
driven H23 cell line than the parent inhibitor Nap-1, the N-methylated control compound
C3-Me and the most powerful selective Mcl-1 inhibitor A-121047745 (IC50 = 0.03 μM,
determined by cellular thermal shift assay, CETSA). The combination with C3 and C5
didn’t show obvious synergistic effect (Fig. 6a), consistent with the fact that H23 is a Mcl-
1-dependent cancer cell line. This results were further validated by PARP cleavage (Fig.
6b). Next, to evaluate whether the observed cytotoxic effects were truly driven by the
efficiency of the PROTACs, we plotted the cytotoxicity % values from the cell viability
assays relative to the Mcl-1 reduction % values (Fig. 6b) induced by C3. Strong correlation
was observed between these two parameters (r2 = 0.97, Fig. 6c), which is consistent with
H23 cell proliferation being Mcl-1-driven. Overall, these results verified that PROTACs
achieving in cellulo protein knockdown exhibited a better capacity to regulate cell
phenotypic changes than the predominantly occupancy-based inhibitors.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 43
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. PROTACs exhibited better capacity to regulate cell phenotypic changes than the
predominantly occupancy-based inhibitors. (a) Cell cytotoxicity assays in H23 cells upon
addition of C3, A-1210477, Nap-1, C3-Me and a combination of C3 with C5 for 24 h.
Percentages of cell viability were calculated by normalization of culture background
without cells against untreated cultures as control. Data are expressed as the mean ± s.e.m.
(n=3). (b) The levels of Mcl-1 and PARP cleavage were measured by western blot in H23
upon addition of a concentration gradient of C3 for 12 h. All western blot figures represent
the results from nꢀ=ꢀ3 independent samples. (c) Plotting the cytotoxicity % values in the
H23 cells vs the Mcl-1 reduction % values induced by C3. Data are expressed as the mean
± s.e.m. (n=3).
Conclusion
In conclusion, we reported the first two PROTACs for PPI target degradation, which
highlighted that PROTACs could expand the “target space” to PPI targets for which only
weak binding ligands are available. These two PROTACs, C3/C5, produce potent,
selective and reversible cellular Mcl-1/Bcl-2 knockdown and exhibit cytotoxicity driven
by PROTAC efficiency, providing a new potent toolkit for selective chemical intervention
of Bcl-2 family proteins in chemical biology research and drug discovery. Moreover,
structure-activity relationship (SAR) analysis and MD simulations in a certain extent
revealed the structural basis for turning nonselective or promiscuous target binding ligands
into more selective degraders by changing the length and conformation of their linkers,
16
ACS Paragon Plus Environment

Page 17 of 43
Journal of Medicinal Chemistry
1
2
3
4
which is valuable for the design of other PROTACs.
5
6
7
8
Experimental Section.
9
Chemistry.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
All commercial reagents were purchased and used without further purification or
1
distillation unless otherwise stated. H-NMR were obtained with a Bruker AV-500 or
Bruker AV-400 spectrometer with chemical shifts reported as ppm (in DMSO, TMS as an
internal standard). The following abbreviations are used for multiplicity of NMR signals:
s=singlet, d=doublet, t=triplet. q=quartet, m=multiplet. High-resolution mass spectra
(HRMS) were obtained on a MALDI-Tof MS (Micro) spectrometer. Column
chromatography was performed on silica gel 200-300 mesh. The purity of all final products
was determined by analytical HPLC-MS to be ≥ 95%.
Chemical synthesis of intermediate 1. (Scheme 1a)
To a suspension of 6-((4-bromophenyl)thio)-1-oxo-1H-phenalene-2,3-dicarbonitrile
(417 mg, 1.0 mmol, 1.0 equiv) in a mixture of acetonitrile/H2O (9:1, 20 mL) were added
methyl 3-aminopropanoate hydrochloride (1.03 g, 10.0 mmol, 10.0 equiv) and NaOH (0.80
g, 20.0 mmol, 20.0 equiv). After stirring at rt for 1 h, the solvent was removed under
reduced pressure and the crude was purified on silica gel (CH2Cl2/CH3OH 30:1).
Methyl
3-((6-((4-bromophenyl)thio)-2-cyano-1-oxo-1H-phenalen-3-
yl)amino)propanoate (1)
1
The expected product compound 1 was isolated as a yellow powder (133 mg, 27 %). H
NMR (400 MHz, CDCl3): δ 8.59 (d, J=8.2 Hz, 2H), 8.46(d, J=8.3 Hz, 1H), 7.89 (t, J= 8.2
Hz, 1H), 7.64 (d, J=8.5 Hz, 2H), 7.57 (d, J=8.2 Hz,1H), 7.35 (d, J=8.5 Hz, 2H), 5.75 (s,
1H), 4.14 (q, J=6.9 Hz, 2H), 3.62 (s, 3H), 2.88 (t,J=6.9 Hz, 2H). TOF MS (ESI+):
C24H17N2O3SBr, [M+Na]+ calculated 517.01, 519.01, found 517.00, 519.03.
Chemical synthesis of intermediate 2. (Scheme 1b)
A mixture of 6-bromo-1H,3H-benzo[de]isochromene-1,3-dione(2.77 g, 10.0 mmol), 4-
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 43
1
2
3
4
5
6
7
8
9
bromobenzenethiol(2.08 g, 11.0 mmol) and triethylamine (1.5 mL) were stirred in DMF at
85 ºC for 6 h. The reaction mixture was first cooled to rt and then the pH was adjusted to
4 using 1 M HCl. The mixture was then filtered and the resulting solid was washed with
water and ether and precipitated in petroleum ether/ethyl acetate (4:1), giving the desired
product 5 (299 mg, 78.0 %) as a yellow solid. TOF MS (ESI+): C18H9BrO3S, [M+H]+
calculated 384.94, 386.94, found 384.97, 386.95.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In a round-bottom flask, a mixture of 5 (691 mg, 1.8 mmol), ethylenediamine (360 μL)
and triethylamine (1.2 mL) were stirred in ethanol at 80 ºC for 6 h. The resulting mixture
was concentrated under reduced pressure and the crude was purified on silica gel
(CH2Cl2/MeOH 30:1). The expected product 6 was isolated as bring yellow flakes (500
mg, 68 %). TOF MS (EI+): C20H15BrO2N2S, [M+H]+ calculated 427.00, 429.00, found
427.01, 429.01.
In a round-bottom flask, a mixture of 6 (300 mg, 0.702 mmol, 1.0 equiv), 6-Methoxy-6-
oxohexanoic acid (124 mg, 0.772 mmol, 1.1 equiv), HATU (320 mg, 0.842 mmol, 1.2
equiv) were stirred in DCM (15 mL) and DIPEA (272 mg, 2.106 mmol, 3.0 equiv) was
added at 0 ºC. The resulting mixture was stirred at rt overnight. The mixture was diluted
with water (30 mL) and extracted with ethyl acetate (30 mL x 3). The combined organic
layers were washed with brine (30 mL x 3), dried over anhydrous sodium sulfate, and
evaporated under reduced pressure. The residue was purified on silica gel (CH2Cl2/MeOH
30:1).
Methyl-6-((2-(6-((4-bromophenyl)thio)-1,3-dioxo-1H-benzo[de]isoquinolin-2(3H)-
yl)ethyl)amino)-6-oxohexanoate (2)
The expected product 2 was isolated as yellow flakes (200 mg, 50 %). 1H NMR (500 MHz,
CDCl3): δ 8.65 (dd, J = 7.2, 6.2 Hz, 2H), 8.39 (d, J = 7.9 Hz, 1H), 7.82 – 7.78 (m, 1H), 7.59
– 7.55 (m, 2H), 7.39 – 7.35 (m, 2H), 7.34 (d, J = 7.9 Hz, 1H), 6.17 (s, 1H), 4.41 – 4.36 (m,
2H), 3.70 – 3.63 (m, 2H), 3.60 (s, 3H), 2.22 (dd, J = 8.7, 5.2 Hz, 2H), 2.11 (t, J = 7.0 Hz,
2H), 1.55 (dt, J = 7.2, 3.7 Hz, 4H). TOF MS (ESI+) [C27H26BrN2O5S]+ calculated 569.06,
18
ACS Paragon Plus Environment

Page 19 of 43
Journal of Medicinal Chemistry
1
2
3
4
571.06, found 569.10, 571.09.
5
6
7
8
Chemical synthesis of CRBN binding intermediate 3 and 3-Me.
Compounds 3 and 3-Me were facilely synthesized in high yield by the procedure as
published in the literature.40
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chemical synthesis of intermediate C1-C6, C3-Me and C5-Me. (Scheme 1b)
General procedure.
Treatment of 2 (99 mg, 1 equiv, 1.0 mmol) with sodium hydroxide (80 mg, 2.00 mmol)
gave compound 7 (88 mg, 92%) as orange flakes. 1H NMR (400 MHz, CDCl3 ): δ 12.19
(s, 1H), 8.58 (d, J=8.0 Hz, 2H), 8.46 (t, J=8.0 Hz, 2H), 7.88 (t, J= 8.0 Hz, 1H), 7.63 (d,
J=8.5 Hz, 2H), 7.57 (d, J=8.1 Hz, 1H), 7.38 (d, J=8.5 Hz, 2H), 5.43 (s, 1H), 4.11 (q, J=6.9
Hz, 2H), 2.82 (t, J=6.9 Hz, 2H). TOF MS (ESI+): C23H15N2O3SBr, [M+Na]+ calculated
577.08, found 577.10.
In a round-bottom flask, a mixture of 3/3-Me (0.089 mmol 1.1 equiv), 7 (45 mg 0.081
mmol 1.0 equiv), HATU (37 mg 0.084 mmol 1.2 equiv) were stirred in DCM (15 mL) and
DIPEA ( 31 mg 0.24 mmol 3.0 equiv) was added at 0 ºC. The resulting mixture was stirred
at rt overnight. The mixture was diluted with water (30mL) and extracted with ethyl acetate
(30 mL × 3). The combined organic layers were washed with brine (30 mL × 3), dried over
anhydrous sodium sulfate, and evaporated under reduced pressure. The residue was
purified on silica gel (CH2Cl2/MeOH 30:1). The expected product C1-C6/C3-Me/C5-Me
was isolated as yellow flakes.
N1-(2-(6-((4-bromophenyl)thio)-1,3-dioxo-1H-benzo[de]isoquinolin-2(3H)-yl)ethyl)-
N6-(2-((2-(2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-yl)amino)ethyl)adipamide
(C1).
41 mg, yield = 60 %. 1H NMR (500 MHz, CDCl3) δ 8.65 – 8.61 (m, 2H), 8.41 (s, 1H),
8.37 (d, J = 7.9 Hz, 1H), 7.81 – 7.77 (m, 1H), 7.60 – 7.55 (m, 2H), 7.49 – 7.44 (m, 1H),
7.39 – 7.36 (m, 2H), 7.31 (d, J = 7.9 Hz, 1H), 7.06 (d, J = 7.1 Hz, 1H), 6.87 (d, J = 8.5 Hz,
1H), 6.32 (s, 1H), 6.19 (t, J = 5.6 Hz, 1H), 6.12 (t, J = 5.6 Hz, 1H), 4.91 (dd, J = 12.2, 5.4
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 43
1
2
3
4
5
6
7
8
Hz, 1H), 4.36 (t, J = 5.6 Hz, 2H), 3.71 (m, J = 13.8, 6.8 Hz, 2H), 3.40 – 3.11 (m, 4H), 2.20
– 2.04 (m, 8H), 1.58 – 1.37 (m, 4H). [TOF MS (ESI+)]: C41H37BrN6O8S [M+Na+],
calculated 875.146917, found 875.14677
9
N1-(2-(6-((4-bromophenyl)thio)-1,3-dioxo-1H-benzo[de]isoquinolin-2(3H)-yl)ethyl)-
N6-(4-((2-(2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-yl)amino)butyl)adipamide
(C2).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
49 mg, yield = 62 %. H NMR (500 MHz, CDCl3) δ 8.80 (s, 1H), 8.61 (t, J = 7.8 Hz,
2H), 8.35 (d, J = 7.9 Hz, 1H), 7.83 – 7.72 (m, 1H), 7.57 (d, J = 8.5 Hz, 2H), 7.50 – 7.42
(m, 1H), 7.37 (d, J = 8.4 Hz, 2H), 7.27 (d, J = 7.5 Hz, 1H), 7.07 (d, J = 7.1 Hz, 1H), 6.92
(d, J = 8.5 Hz, 1H), 6.64 (s, 1H), 6.48 – 6.36 (m, 2H), 4.91 (dd, J = 12.2, 5.4 Hz, 1H), 4.32
(dd, J = 12.5, 7.4 Hz, 2H), 3.61 (tt, J = 14.3, 7.2 Hz, 2H), 2.80 – 2.69 (m, 4H), 2.17 – 2.01
(m, 8H), 1.53 – 1.36 (m, 8H). [TOF MS (ESI+)]: C43H41BrN6O8S [M+Na+], calculated
903.178217, found 903.17781
N1-(2-(6-((4-bromophenyl)thio)-1,3-dioxo-1H-benzo[de]isoquinolin-2(3H)-yl)ethyl)-
N6-(7-((2-(2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-yl)amino)heptyl)adipamide
(C3).
64mg, yield = 56 %. 1H NMR (400 MHz, CDCl3) δ 8.64 (dd, J = 7.9, 3.8 Hz, 2H), 8.38
(d, J = 7.9 Hz, 1H), 8.26 (s, 1H), 7.85 – 7.76 (m, 1H), 7.60 – 7.56 (m, 2H), 7.50 – 7.45 (m,
1H), 7.40 – 7.36 (m, 2H), 7.32 (d, J = 7.9 Hz, 1H), 7.08 (d, J = 7.1 Hz, 1H), 6.86 (d, J =
8.5 Hz, 1H), 6.22 (d, J = 6.2 Hz, 2H), 5.83 (s, 1H), 4.91 (dd, J = 12.1, 5.4 Hz, 1H), 4.41 –
4.34 (m, 2H), 3.66 (m, 2H), 3.25 (dd, J = 12.6, 6.6 Hz, 2H), 3.16 (dd, J = 12.9, 6.9 Hz, 2H),
2.15 – 2.03 (m, 8H), 1.65 (m, 6H), 1.51 – 1.45 (m, 4H), 1.44 (d, J = 6.8 Hz, 2H). [TOF MS
(ESI+)]: C45H45BrN6O8S [M+Na+], calculated 931.209517, found 931.21017.
N1-(2-(6-((4-bromophenyl)thio)-1,3-dioxo-1H-benzo[de]isoquinolin-2(3H)-yl)ethyl)-
N6-(6-((2-(1-methyl-2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-
yl)amino)hexyl)adipamide (C3-Me).
45 mg, yield = 60 %. 1H NMR (400 MHz, CDCl3) δ 8.66 (dd, J = 7.8, 3.8 Hz, 2H), 8.38
20
ACS Paragon Plus Environment

Page 21 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(d, J = 7.8 Hz, 1H), 8.26 (s, 1H), 7.85 – 7.76 (m, 1H), 7.61 – 7.56 (m, 2H), 7.50 – 7.45 (m,
1H), 7.40 – 7.36 (m, 2H), 7.32 (d, J = 7.9 Hz, 1H), 7.11 (d, J = 7.0 Hz, 1H), 6.89 (d, J =
8.5 Hz, 1H), 6.23 (d, J = 6.2 Hz, 2H), 5.85 (s, 1H), 4.91 (dd, J = 12.1, 5.4 Hz, 1H), 4.41 –
4.34 (m, 2H), 3.66 (m, 2H), 3.25 (dd, J = 12.6, 6.6 Hz, 2H), 3.21 (s, 3H), 3.16 (dd, J = 12.9,
6.9 Hz, 2H), 2.15 – 2.03 (m, 8H), 1.65 (m, 6H), 1.51 – 1.45 (m, 4H), 1.44 (d, J = 6.8 Hz,
2H). [TOF MS (ESI+)]: C46H47BrN6O8S [M+Na+], calculated 945.2252, found 945.2258.
N1-(2-(6-((4-bromophenyl)thio)-1,3-dioxo-1H-benzo[de]isoquinolin-2(3H)-yl)ethyl)-
N6-(6-((6-((2-(2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-yl)amino)hexyl)amino)-
6-oxohexyl)adipamide (C4).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
69 mg, yield = 50 %. 1H NMR (500 MHz, CDCl3) δ 8.64 (dd, J = 7.9, 3.0 Hz, 2H), 8.49
(s, 1H), 8.37 (d, J = 7.9 Hz, 1H), 7.80 (dd, J = 8.4, 7.4 Hz, 1H), 7.59 – 7.57 (m, 2H), 7.48
(dd, J = 8.4, 7.2 Hz, 1H), 7.40 – 7.36 (m, 2H), 7.31 (d, J = 7.9 Hz, 1H), 7.07 (d, J = 7.0 Hz,
1H), 6.87 (d, J = 8.6 Hz, 1H), 6.33 (s, 1H), 6.21 (s, 1H), 5.92 (s, 1H), 5.88 (s, 1H), 4.92
(dd, J = 12.3, 5.4 Hz, 1H), 4.40 – 4.34 (m, 2H), 3.70 – 3.60 (m, 2H), 3.23 (m, 4H), 3.14
(m, 2H), 2.82 – 2.68 (m, 2H), 2.16 – 2.08 (m, 8H), 2.06 (d, J = 6.6 Hz, 2H), 1.55 – 1.47
(m, 8H), 1.43 (m, 8H). [TOF MS (ESI+)]: C51H56BrN7O9S [M+Na+], calculated
1044.293581, found 1044.29318
N1-(2-(6-((4-bromophenyl)thio)-1,3-dioxo-1H-benzo[de]isoquinolin-2(3H)-yl)ethyl)-
N6-(2-(2-(2-((2-(2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-
yl)amino)ethoxy)ethoxy)ethyl)adipamide (C5).
35 mg, yield = 52 %. 1H NMR (400 MHz, CDCl3) δ 9.24 (s, 1H), 8.62 (dd, J = 10.3, 7.9
Hz, 2H), 8.37 (d, J = 7.9 Hz, 1H), 7.79 (dd, J = 8.3, 7.5 Hz, 1H), 7.61 – 7.54 (m, 2H), 7.51
– 7.45 (m, 1H), 7.40 – 7.35 (m, 2H), 7.31 (d, J = 7.9 Hz, 1H), 7.09 (d, J = 7.1 Hz, 1H), 6.88
(d, J = 8.5 Hz, 1H), 6.53 (t, J = 5.3 Hz, 1H), 6.41 (t, J = 5.1 Hz, 1H), 6.29 (t, J = 5.2 Hz,
1H), 4.96 – 4.87 (m, 1H), 4.39 – 4.30 (m, 2H), 3.72 (t, J = 5.1 Hz, 2H), 3.66 – 3.61 (m,
6H), 3.55 (t, J = 5.0 Hz, 2H), 3.45 (dd, J = 10.4, 5.2 Hz, 2H), 3.39 (dd, J = 10.5, 5.3 Hz,
2H), 2.07 (dt, J = 6.7, 3.4 Hz, 8H), 1.54 – 1.49 (m, 4H). [TOF MS (ESI+)]: C45H45BrN6O10S
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 43
1
2
3
4
[M+Na+], calculated 963.199346, found 963.19903.
5
6
7
8
N1-(2-(6-((4-bromophenyl)thio)-1,3-dioxo-1H-benzo[de]isoquinolin-2(3H)-yl)ethyl)-
N6-(2-(2-(2-((2-(1-methyl-2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-
yl)amino)ethoxy)ethoxy)ethyl)adipamide (C5-Me).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
43 mg, yield = 55 %. 1H NMR (400 MHz, CDCl3) δ 9.26 (s, 1H), 8.60 (dd, J = 10.3, 7.9
Hz, 2H), 8.39 (d, J = 7.9 Hz, 1H), 7.81 (dd, J = 8.3, 7.5 Hz, 1H), 7.60 – 7.54 (m, 2H), 7.52
– 7.44 (m, 1H), 7.40 – 7.35 (m, 2H), 7.31 (d, J = 7.9 Hz, 1H), 7.09 (d, J = 7.1 Hz, 1H), 6.88
(d, J = 8.5 Hz, 1H), 6.53 (t, J = 5.3 Hz, 1H), 6.41 (t, J = 5.1 Hz, 1H), 6.29 (t, J = 5.2 Hz,
1H), 4.97 – 4.89 (m, 1H), 4.39 – 4.31 (m, 2H), 3.73 (t, J = 5.1 Hz, 2H), 3.66 – 3.61 (m,
6H), 3.57 (t, J = 5.0 Hz, 2H), 3.46 (dd, J = 10.4, 5.2 Hz, 2H), 3.41 (dd, J = 10.5, 5.3 Hz,
2H), 3.21 (s, 3H), 2.08 (dt, J = 6.7, 3.4 Hz, 8H), 1.54 – 1.48 (m, 4H). [TOF MS (ESI+)]:
C46H47BrN6O10S [M+Na+], calculated 977.2150, found 977.2154.
N1-(2-(6-((4-bromophenyl)thio)-1,3-dioxo-1H-benzo[de]isoquinolin-2(3H)-yl)ethyl)-
N6-(8-((2-(2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-yl)amino)octyl)adipamide
(C6).
44mg, yield = 59 %. 1H NMR (400 MHz, CDCl3) δ 8.64 (dd, J = 7.9, 3.8 Hz, 2H), 8.38
(d, J = 7.9 Hz, 1H), 8.26 (s, 1H), 7.85 – 7.76 (m, 1H), 7.60 – 7.56 (m, 2H), 7.50 – 7.45 (m,
1H), 7.40 – 7.36 (m, 2H), 7.33 (d, J = 7.9 Hz, 1H), 7.10 (d, J = 7.1 Hz, 1H), 6.86 (d, J =
8.5 Hz, 1H), 6.22 (d, J = 6.2 Hz, 2H), 5.83 (s, 1H), 4.91 (dd, J = 12.1, 5.4 Hz, 1H), 4.41 –
4.34 (m, 2H), 3.66 (m, 2H), 3.25 (dd, J = 12.6, 6.6 Hz, 2H), 3.16 (dd, J = 12.9, 6.9 Hz, 2H),
2.15 – 2.03 (m, 8H), 1.63 (m, 6H), 1.54 – 1.45 (m, 8H), 1.44 (d, J = 6.8 Hz, 2H). [TOF MS
(ESI+)]: C47H49BrN6O8S [M+H+], calculated 937.2589, found 937.2581.
Chemical synthesis of intermediate H1-H5. (Scheme 1a)
General procedure.
Treatment of 1 (99 mg, 1 equiv, 1.0 mmol) with sodium hydroxide (80 mg, 2.00 mmol)
gave compound S1-5 (88 mg, 92%) as orange flakes. 1H NMR (400 MHz, CDCl3 ): δ 12.19
(s, 1H), 8.58 (d, J=8.0 Hz, 2H), 8.46 (t, J=8.0 Hz, 2H), 7.88 (t, J= 8.0 Hz, 1H), 7.63 (d,
22
ACS Paragon Plus Environment

Page 23 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
J=8.5 Hz, 2H), 7.57 (d, J=8.1 Hz, 1H), 7.38 (d, J=8.5 Hz, 2H), 5.43 (s, 1H), 4.11 (q, J=6.9
Hz, 2H), 2.82 (t, J=6.9 Hz, 2H). TOF MS (ESI+): C23H15N2O3SBr, [M+Na]+ calculated
502.9864, found 502.9870.
9
In a round-bottom flask, a mixture of 3 (0.092 mmol, 1.1 equiv), S1-5 (40 mg, 0.084
mmol, 1.0 equiv), HATU (38 mg, 0.1 mmol, 1.2 equiv) were stirred in DCM (15 mL) and
DIPEA (32 mg, 0.25 mmol, 3.0 equiv) was added at 0 ºC. The resulting mixture was stirred
at rt overnight. The mixture was diluted with water (30 mL) and extracted with ethyl acetate
(30 mL x 3). The combined organic layers were washed with brine (30 mL x 3), dried over
anhydrous sodium sulfate, and evaporated under reduced pressure. The residue was
purified on silica gel (CH2Cl2/MeOH 30:1). The expected product H1-H5 was isolated as
orange flakes.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3-((6-((4-bromophenyl)thio)-2-cyano-1-oxo-1H-phenalen-3-yl)amino)-N-(2-((2-(2,6-
dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-yl)amino)ethyl)propanamide (H1).
39 mg, yield = 61 %. 1H NMR (500 MHz, DMSO) δ 11.08 (s, 1H), 8.71 (s, 1H), 8.57 (d, J
= 8.4 Hz, 1H), 8.45 (d, J = 7.2 Hz, 1H), 8.40 (d, J = 8.1 Hz, 1H), 8.29 (s, 1H), 7.90 – 7.84
(m, 1H), 7.63 (d, J = 8.5 Hz, 2H), 7.50 (t, J = 7.3 Hz, 2H), 7.38 (d, J = 8.4 Hz, 2H), 7.07
(d, J = 8.6 Hz, 1H), 6.97 (d, J = 7.1 Hz, 1H), 6.69 (s, 1H), 5.01 (dd, J = 12.8, 5.5 Hz, 1H),
4.13 (d, J = 6.1 Hz, 2H), 3.25 (m, 4H), 2.64 (t, J = 6.5 Hz, 2H), 2.02 – 1.93 (m, 4H). [TOF
MS (ESI+)]: C38H29BrN6O6S [M+Na+] calculated 799.094487, found 799.09364
3-((6-((4-bromophenyl)thio)-2-cyano-1-oxo-1H-phenalen-3-yl)amino)-N-(4-((2-(2,6-
dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-yl)amino)butyl)propanamide (H2)
35 mg, yield = 52%. 1H NMR (400 MHz, CDCl3) δ 8.77 (s, 1H), 8.51 – 8.38 (m, 3H), 7.91
(d, J = 8.2 Hz, 1H), 7.66 (t, J = 7.8 Hz, 1H), 7.56 (d, J = 8.4 Hz, 2H), 7.36 (d, J = 8.4 Hz,
2H), 7.33 – 7.28 (m, 1H), 7.14 (d, J = 8.1 Hz, 1H), 7.06 (s, 1H), 6.92 (d, J = 7.1 Hz, 1H),
6.70 (d, J = 8.5 Hz, 1H), 6.12 (d, J = 5.3 Hz, 1H), 4.89 (dd, J = 12.1, 5.5 Hz, 1H), 4.28 (s,
2H), 3.33 (d, J = 5.3 Hz, 2H), 3.19 (d, J = 5.4 Hz, 2H), 2.88 – 2.66 (m, 4H), 1.51 – 1.47
(m, 4H), 1.43 (d, J = 6.0 Hz, 2H). [TOF MS (ESI+)]: C40H33BrN6O6S [M+Na+], calculated
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 43
1
2
3
4
827.125788, found 827.12489
5
6
7
8
3-((6-((4-bromophenyl)thio)-2-cyano-1-oxo-1H-phenalen-3-yl)amino)-N-(6-((2-(2,6-
dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-yl)amino)hexyl)propanamide (H3)
52 mg, yield = 56%. 1H NMR (500 MHz, CDCl3) δ 8.58 (d, J = 7.2 Hz, 1H), 8.52 (d, J =
8.4 Hz, 1H), 8.41 (s, 1H), 8.33 (s, 1H), 7.91 (d, J = 8.3 Hz, 1H), 7.72 (t, J = 7.8 Hz, 1H),
7.56 (d, J = 8.3 Hz, 2H), 7.40 (t, J = 7.9 Hz, 1H), 7.36 (d, J = 8.3 Hz, 2H), 7.20 (d, J = 8.1
Hz, 1H), 6.99 (d, J = 7.1 Hz, 1H), 6.79 (d, J = 8.6 Hz, 1H), 6.43 (s, 1H), 6.16 (s, 1H), 4.90
(dd, J = 11.9, 5.4 Hz, 1H), 4.30 (d, J = 4.6 Hz, 2H), 3.29 (d, J = 6.1 Hz, 2H), 3.21 (dd, J =
12.3, 6.4 Hz, 2H), 2.74 (dd, J = 15.3, 4.2 Hz, 2H), 2.01 – 1.91 (m, 4H), 1.65 – 1.54 (m,
2H), 1.36 – 1.17 (m, 6H). [TOF MS (ESI+)]: C42H37BrN6O6S, found 855.1579, 857.1554
6-(3-((6-((4-bromophenyl)thio)-2-cyano-1-oxo-1H-phenalen-3-
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
yl)amino)propanamido)-N-(6-((2-(2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-
yl)amino)hexyl)hexanamide (H4)
42 mg, yield = 45%. 1H NMR (400 MHz, CDCl3) δ 8.71 (s, 1H), 8.57 (t, J = 6.9 Hz, 2H),
8.27 (s, 1H), 8.00 (d, J = 8.3 Hz, 1H), 7.78 – 7.73 (m, 1H), 7.59 – 7.55 (m, 2H), 7.49 – 7.41
(m, 1H), 7.38 – 7.33 (m, 2H), 7.22 (d, J = 8.2 Hz, 1H), 7.04 (d, J = 6.9 Hz, 1H), 6.84 (d, J
= 8.5 Hz, 1H), 6.75 (s, 1H), 6.19 (s, 1H), 5.79 (s, 1H), 4.93 – 4.88 (m, 1H), 4.28 (m, 2H),
3.29 – 3.14 (m, 8H), 2.74 – 2.68 (m, 2H), 2.25 – 2.11 (m, 4H), 1.53 – 1.39 (m, 14H). [TOF
MS (ESI+)]: C48H48BrN7O7S [M+Na+], calculated 968.241152, found 968.24052
3-((6-((4-bromophenyl)thio)-2-cyano-1-oxo-1H-phenalen-3-yl)amino)-N-(2-(2-(2-((2-
(2,6-dioxopiperidin-3-yl)-1,3-dioxoisoindolin-4-
yl)amino)ethoxy)ethoxy)ethyl)propanamide (H5)
38mg, yield = 53%. 1H NMR (400 MHz, CDCl3) δ 9.13 (s, 1H), 8.56 (d, J = 7.3 Hz, 1H),
8.49 (d, J = 8.4 Hz, 1H), 8.23 (s, 1H), 7.91 (d, J = 8.2 Hz, 1H), 7.70 (t, J = 7.8 Hz, 1H),
7.56 (d, J = 8.4 Hz, 2H), 7.37 (d, J = 8.4 Hz, 2H), 7.32 – 7.27 (m, 1H), 7.16 (dd, J = 8.1,
3.8 Hz, 1H), 7.10 (s, 1H), 6.88 (d, J = 7.0 Hz, 1H), 6.78 (d, J = 8.5 Hz, 1H), 6.50 (s, 1H),
4.90 (dd, J = 11.9, 5.7 Hz, 1H), 4.27 (m, 2H), 3.67 (m, 4H), 3.64 – 3.59 (m, 4H), 3.50 –
24
ACS Paragon Plus Environment

Page 25 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
3.41 (m, 6H), 2.18 – 1.93 (m, 4H). [TOF MS (ESI+)]: C42H37BrN6O8S [M+Na+], calculated
887.146917, found 887.1732
7
8
Protein Expression and Purification.
9
The plasmid expressing Mcl-1 and Bcl-2 wconstructed as described in our previous
study.36 The two proteins with an N-terminal 8 × His tag was produced in Escherichia coli
BL21 (DE3) cells. Cells were grown at 37 °C in LB containing 30 μg/mL Kana. Protein
expression was induced by 0.4 mM IPTG at 37 °C for 4h. Cells were lysed in PBS
containing 1 mM PMSF. The protein was purified from the soluble fraction using Ni-NTA
resin (Qiagen, Hilden, Germany), following the manufacturer’s instructions. The protein
was further purified on a Source Q15 column in 25 mM Tris pH 8.0 buffer, with a NaCl
gradient.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Enzyme-linked immunosorbent assay (ELISA)
R-(-)-Gossypol was used as a positive control. For this assay, biotinylated Bim peptide
(residues 81–106, biotin-(β)A-(β)A-D-M-R-P-E-I-W-I-A-Q-E-L-R-R-I-G-D-E-F-N-A-Y-
Y-A-R-R-amide, hereafter called biotin-Bim) was diluted to 0.09 mg/mL in SuperBlock
blocking buffer in PBS (Pierce Biotechnology, Inc, Rockford, IL, catalog #37515) and
incubated for 1.5 h in 96-well microtiter plates already coated with streptavidin (Qiagen,
catalog #15500) to allow the formation of the complex between Biotin-Bim and
streptavidin. All incubations were performed at 4 ℃ unless otherwise noted. Each inhibitor
was first dissolved in pure DMSO to obtain a 10 mM stock solution. Then the stock solution
was diluted successively to get the solution with different concentration gradients (100 μM,
50 μM, 10 μM, 5 μM, 3 μM, 1 μM, 0.5 μM, 0.1 μM). For each tested inhibitor, different
concentrations of the inhibitor were incubated with 200 nM His-tagged Mcl-1 or Bcl-2
protein in PBS for 1 h with a final DMSO concentration of 4%. The plates were washed
three times with PBS containing 0.05% Tween-20. The inhibitor and protein mixture (100
mL) were transferred to the plate containing the biotin-Bim/streptavidin complex and
incubated for 2 h. The plate was then washed as before and mouse anti-His antibody that
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 43
1
2
3
4
5
6
7
8
9
was conjugated with horseradish peroxidase (Qiagen, catalog #34460) was added into the
wells and incubated for 1 h. The plate was then washed with PBS containing 0.05% Tween-
20. Finally, TMB (100 mL, Beyotime, catalog #P0209) was added to each well; the
enzymatic reaction was stopped after 30 min by addition of H2SO4 (100 mL, 2 M).
Absorbances were measured with a TECAN GENios (Swiss, TECAN) microplate reader
using a wavelength of 450 nm. Three independent experiments were performed with each
inhibitor to calculate average IC50 value and standard deviation (SD). (-)-Gossypol was
used as comparison in our assay.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Cell Culture
The human cervical carcinoma cells (Hela) was obtained from the Typical Culture
Preservation Commission Cell Bank (Chinese Acadamy of Sciences, Shanghai, China) and
maintained in DMEM/HIGH GLUCOSE (Hyclone, Beijing, China) supplemented with 10%
fetal bovine serum (Gibco Company, USA). The human chronic myelogenous leukemia
cells (K562) and human lung cells (NCI-H23) were obtained from the Typical Culture
Preservation Commission Cell Bank and maintained in RPMI1640 (Hyclone, Beijing,
China) media supplemented with 10% fetal bovine serum. All cells were cultured at 37°C
in 5% CO2.
Immunoblotting.
Cells were lysed in NP-40 lysis buffer (0.2% NP-40, 142.5 mM KCl, 5 mM MgCl2, 1
mM EGTA, and 20 mM HEPES at pH 7.5) containing 1 mM PMSF, 1 μg/mL aprotinin, 1
μg/mL leupeptin, and 1 μg/mL pepstatin. Protein concentrations were determined by BCA
assay (Pierce Chemical, IL, USA). Equal protein concentrations of each sample were
subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and
electrophoretically transferred to polyvinylidene difluoride membranes. Membranes were
blocked with TBS containing 5% skim milk and 0.1% Tween 20 and then probed with
specific antibodies. Blots were developed using enhanced chemiluminescence kit
(Amersham Biosciences, San Diego, CA, USA) according to the manufacturer’s
26
ACS Paragon Plus Environment

Page 27 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
instructions. All immunonloting experiments were repeated with nꢀ=ꢀ3 independent
samples.
7
8
The ternary complexes detecting assay.
9
Hela (25×106) were lysed in 1% Triton X-100 IP buffer (Beyotime, Shanghai, China).
Whole-cell lysates were obtained, 1 mg total proteins were incubated overnight with 1 μg
of the specific CRBN-antibody. The ternary complexes were captured with protein A+G
Agarose (Santa Cruz Biotechnology, CA, USA), washed three times with Triton X-100
buffer and boiled in loading buffer (Invitrogen, Carlsbad, CA, USA). The constitutes of
the ternary complexes was performed by immunoblotting using specific Mcl-1, Bcl-2 and
CRBN antibodies as described above.8,9 This assay was repeated with nꢀ=ꢀ3 independent
samples.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Molecular docking
The 3D structure of Mcl-1 (PDB ID: 2PQK) was obtained from the protein bank in the
RCSB. The mol2 structures of the inhibitors were generated using Chembio3D Ultra 11.0
followed by energy minimization. AutoDock 4.2 program equipped with ADT was used to
perform the automated molecular docking. Grid maps covering residues that were
perturbed more than the threshold value of 0.1 ppm in the binding groove of the proteins
were defined for all inhibitors in the AutoDock calculations using a grid spacing about of
0.375 Å. For each docking job, 100 hybrid GALS runs were carried-out. A total of 100
possible binding conformations were generated and grouped into clusters based on a 1.0 Å
cluster tolerance. The docking models were analyzed and represented using ADT.
Molecular Dynamics (MD) Simulations
The starting structures for CRBN came from the crystal structure downloaded from
Protein Data Bank (PDB) entry 4TZ4. In order to replace the ligand in this structure with
pomalidomide, the PDB entry 4TZU was used by aligning its protein backbone to that of
4TZ4, transferring pomalidomide to the 4TZ4 structure and replacing the original ligand.
The starting structures for Mcl-1 or Bcl-2 (PDB: 2PQK or 2XA0) and 2 complexes was
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 43
1
2
3
4
5
6
7
8
9
the best pose obtained from the docking study. The starting structures for the Mcl-1 (or
Bcl-2): PROTAC: Cereblon trimer were prepared in a way that the hydrophobic patch of
CRBN-ligand surface opposed different hydrophobic patches and grooves of the Mcl-1-
ligand surface, thus producing different starting modes in terms of the relative dispositions
between CRBN and Mcl-1. For each starting mode, a linker was built to connect
pomalidomide and Mcl-1 ligand and form the full PROTAC. And an energy minimization
was performed for each starting point of trimer. All the molecular-dynamics (MD)
simulations were performed in Gromacs 4.6.7 with GROMOS 96 43A1 force-field for
proteins and ligands were prepared using PRODRG 2.5 server. The protocol of the MD
simulation was as follows: The complex structure was solvated in a truncated
dodecahedron SPC water box 10 Å from the edge and the positive charge of the system
was neutralized by adding Cl- counter ions. The system was energy minimized with the
steepest descent algorithm for 1000 steps. Thereafter, 1 ns isothermal-isochoric (NVT, with
T = 300 K) and 1 ns isothermal-isobaric (NPT, with T = 300 K and P =1 atm) MD
simulations were performed to equilibrate the system. The harmonic restraint with a force
constant of 1000 kJ mol-1 nm-2 was used during the simulations. The equilibrated system
was then subjected to a simulation of 50 ns without any restraint. Two independent MD
runs were performed for each complex for ensemble convergence. During the equilibration
and production runs, the time step was 2 fs, and the LINCS algorithm was applied to
constrain the bonds involving hydrogens. The van der Waals and short-range electrostatic
interactions cutoff was set to 14 Å, and the particle-mesh Ewald (PME) method was used
to treat long-range electrostatics.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Cell Viability Assays
H23 cells were incubated with compounds at the desired concentration for 24 h on a
clear-bottom 96-well plate. Cells were kept in RPMI medium supplemented with 10 %
FBS, L-glutamine, penicillin, and streptomycin. Initial cell density was 4 × 104 per mL.
Cells were treated with various concentrations of compound or vehicle DMSO (0.05 %).
28
ACS Paragon Plus Environment

Page 29 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
After treatment, cell viability was measured with cell counting kit 8 (CCK8) cell viability
assay kit according to the manufacturer instructions. Percentages of cell viability were
calculated by normalization of culture background without cells against untreated cultures
as control.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Supporting Information:
Binding affinity curves to Mcl-1/Bcl-2; Time/dose-dependent western blots and
quantification of protein levels in Hela and K562 cells; mRNA levels during PROTACs
treatment in Hela cells; Protein levels of Bcl-xl, MDM2 and tubulin during PROTACs
treatment in Hela cells; Co-IP of Mcl-1/Bcl-2 and CRBN upon H3 treatment; NMR and
LC-MS spectra of key compounds (PDF)
Molecular formula strings (CSV)
The MD-simulated structure of ternary complexes Mcl1-C3-CRBN and Bcl2-C5-CRBN
(PDB)
Corresponding Author Information:
*Phone: +86 411 84986032. Fax: +86 411 84986032. E-mail: zczhang@dlut.edu.cn.
Acknowledgment: This research was supported by the National Natural Science
Foundation of China (81570129, 21502015, and 81430083) and the China Postdoctoral
Science Foundation (2018M641694).
Abbreviations
SAR, structure-activity relationship; PPI, protein-protein interaction; PROTACs,
Proteolysis Targeting Chimeras; BET, bromodomains and extra-terminal; MD, molecular
dynamics; CRBN, cereblon; DMSO, dimethyl sulfoxide; DMF, N,N-dimethylformamide;
DCM, dimethyl chloride; THF, tetrahydrofuran; DIEA, N,N-di-isopropylethylamine;
HATU, hexafluorophosphate azabenzotriazole tetramethyl uranium.
Reference:
29
ACS Paragon Plus Environment