Targeting the Allosteric Pathway That Interconnects the Core-Functional Scaffold and the Distal Phosphorylation Sites for Specific Dephosphorylation of Bcl-2

Dephosphorylation of Bcl ‑2

Article

Targeting the Allosteric Pathway That Interconnects the Core-

Functional Scaﬀold and the Distal Phosphorylation Sites for Speciﬁc

⊥

,⊥

Ziqian Wang, Ting Song,* Zongwei Guo, Keke Cao, Chao Chen, Yingang Feng, Hang Wang,

Fangkui Yin, Sheng Zhou, Jian Dai, and Zhichao Zhang*

Cite This: https://dx.doi.org/10.1021/acs.jmedchem.0c01290

Read Online

ACCESS

Metrics & More

Article Recommendations

sı Supporting Information

ABSTRACT: Protein phosphorylation is the most signiﬁcant post-translational modiﬁcation for regulating cellular activities, but

site-speciﬁc modulation of phosphorylation is still challenging. Using three-dimensional NMR spectra, molecular dynamics

simulations, and alanine mutations, we identiﬁed that the interaction network between pT69/pS70 and R106/R109 residues

prevents the phosphorylation sites from exposure to phosphatase and subsequent dephosphorylation. A Bcl-2-dephosphorylation

probe, S1-6e, was designed by installing a carboxylic acid group to a Bcl-2 inhibitor. The carboxyl group competitively disrupts the

interaction network between R106/R109 and pT69/pS70 and subsequently facilitates Bcl-2 dephosphorylation in living cells. As a

result, S1-6e manifests a more eﬀective apoptosis induction in pBcl-2-dependent cancer cells than other inhibitors exhibiting a

similar binding aﬃnity for Bcl-2. We believe that targeting the allosteric pathways interconnecting the core-functional domain and

the phosphorylation site can be a general strategy for a rational design of site-speciﬁc dephosphorylating probes, since the allosteric

pathway has been discovered in a variety of proteins.

INRODUCTION

activated proapoptotic proteins (Bax and Bak) form oligomers

on the outer mitochondrial membrane, causing its permeabi-

lization (MOMP) and the release of cytochrome c. The

antiapoptotic members (such as Bcl-2, Bcl-xl, and Mcl-1)

physically interact and sequester the activated proapoptotic

proteins to block apoptosis. The BH3-only proteins are further

classiﬁed as “activators” (Bim, Puma, and Bid) and “sensitizers”

(Bad, Noxa, and Bmf), which allosterically activate proapop-

totic proteins or competitively interact with antiapoptotic

■

Protein phosphorylation is the most signiﬁcant post-transla-

tional modiﬁcation for regulating cellular activities, protein

functions, and protein−protein interactions (PPIs). Phosphor-

ylation on serine, threonine, and tyrosine residues not only

directly induces protein−protein interactions via the phospho-

protein-binding domain (such as SH2, WW, and FHA

domains) but also allosterically regulates the biological

−3

function of the phosphorylated proteins.

For example,

10−16

phosphorylation of conserved sites on a distal surface loop of

many kinases can modulate the enzyme activity of the catalytic

domain by inducing conformational changes, and abnormal

phosphorylation of these kinases is a key pathologic event that

leads to the occurrence and progression of a variety of

proteins.

Similar to its functions in kinases, phosphorylation also plays

a critical role in modulating the antiapoptotic function of Bcl-

10,17−25

one of the most attractive targets for anticancer

−6

diseases.

Received: July 24, 2020

Bcl-2 family proteins are the key regulators of the intrinsic

mitochondrial apoptotic pathway that consists of proapoptotic

(

Bax and Bak), antiapoptotic (Bcl-2, Bcl-xl, Mcl-1, Bcl-w, and

Bﬂ-1), and BH3-only (Bim, Puma, Bid, Bad, Noxa, and Bmf)

subgroups. The protein−protein interaction (PPI) network of

these subgroups determines cell fate.

−9

In this model,

XXXX American Chemical Society

Journal of Medicinal Chemistry

J. Med. Chem. XXXX, XXX, XXX−XXX

Article

Figure 1. Bcl-2 phosphomimetic mutant induced a helical conformational change in the R106-D111 region. (a, b) Chemical shift perturbations

CSPs) of Bcl-2 residues in Bcl-2-EK (a) and EEE-Bcl-2-EK (b) in comparison with that in Bcl-2 isoform 2. The residues with large CSPs (>0.2

ppm) are shown in red, and the residues with newly emerged sharp resonances are shown in green. (c) 3D NOESY-HSQC strip plot of the R106-

(

D111 region in EEE-Bcl-2-EK. The sequential NH/NH (i, i + 1) NOEs C H/NH (i, i + 1) and C H/NH (i, i + 3) NOEs are indicated with

dashed lines. (d) EEE-Bcl-2-EK with a helical conformational change in the R106-D111 region, as indicated by NOESY-HSQC data, is shown in

the blue cartoon. The NMR structure of Bcl-2 isoform 2 (PDB: 1GJH) with an unstructured R106-D111 region is shown in the gray cartoon. The

R106-D111 region is highlighted by yellow rectangles in (a) and (d).

drugs. The antiapoptotic protein Bcl-2 comprises seven helical

and dephosphorylation via an allosteric pathway formed by the

interconnected residues between the nucleotides binding the

core-functional domain and the phosphorylation sites. In our

opinion, this allosteric pathway oﬀers a potential avenue for

site-speciﬁc dephosphorylating probe discovery.

segments (α1−α7), and it sequesters proapoptotic or BH3-

only proteins by seizing their BH3 α-helical domains within its

hydrophobic BH3-binding groove, which is surrounded by α2,

10−16

α3, α4, and α5 helices.

The multiple phosphorylation

sites of the Bcl-2 protein (T69, S70, and S87) are located in

Herein, we established a general strategy for the rational

design of site-speciﬁc dephosphorylating small-molecule

probes that target the allosteric pathways interconnecting the

core-functional domain and the phosphorylation site. These

dephosphorylating probes consist of two parts: a recognizing

group that is located at the core-functional domain on one side

of the allosteric pathway and a group that targets and perturbs

the residue−residue interactions in the allosteric pathway. For

example, we designed and synthesized a Bcl-2-dephosphorylat-

ing probe, S1-6e, by adding a carboxylic acid group to our

previously identiﬁed Bcl-2/pBcl-2 inhibitor S1-6 after fully

dissecting the structural basis of the allosteric pathway

interconnecting the core-functional BH3 groove with the

phosphorylation sites in Bcl-2. As expected, S1-6e allosterically

induced Bcl-2 dephosphorylation by competitively disrupting

the interaction network between the pT69/pS70 and R106/

R109 residues, which veriﬁed the usefulness of the strategy.

the loop region (65 residues) between α1 and α 2, which is

18,19,27−29

spatially distant from the BH3 groove

(Figure S1).

Previous studies illustrated that phosphorylated Bcl-2 (pBcl-2)

confers poor survival to leukemia cells and resistance to Bcl-2

inhibitors, such as Obatoclax, S1, ABT-737, ABT-263, and

ABT-199 (the ﬁrst FDA-approved drug for Bcl-2-dependent

cancer treatment) by impeding their direct binding to the BH3

groove. Moreover, pBcl-2 exhibits diﬀerent binding aﬃnities

for proapoptotic members than npBcl-2. For example, pBcl-2

binds with higher aﬃnity to Bax and Bak than that to npBcl-2

but with lower aﬃnity to Bim than that to npBcl-2, suggesting

that the apoptosis network is reshaped by Bcl-2 phosphor-

1−25,30,31

ylation.

Hence, the modulation of protein phosphorylation is of

great importance to understand the pathological processes in

response to phosphorylation, drug discovery, and disease

therapy. The reversible phosphorylation of proteins is executed

by kinases and phosphatases, and therefore, regulation of

relative kinases and phosphatases is a generally accepted

RESULTS AND DISCUSSION

■

“Arginine−Phosphate” Interactions between the

Phosphates and the R106/R109 Side Chains Mediate

an Allosteric Network Interconnecting the Phosphor-

ylation Sites in the Loop Region and the BH3 Groove of

Bcl-2. To date, the structural information regarding the Bcl-2

protein is based on a structurally modiﬁed Bcl-2/Bcl-xL

chimera named Bcl-2 isoform 2, in which the loop region

bearing the three phosphorylation sites is replaced by a

32−34

strategy for modulating protein phosphorylation.

ever, since every single kinase or phosphatase is able to induce

de)phosphorylation of several proteins, this strategy lacks

How-

(

site/protein speciﬁcity.

Interestingly, researchers have reported that nucleotide

(

ATP, AMP, ADP, etc.) binding can regulate the phosphor-

22−24

ylation of many proteins, especially kinases.

Nucleotides

protect the distal phosphorylated sites from phosphatase access

sequence from Bcl-xL. Most recently, to explore the

Journal of Medicinal Chemistry

J. Med. Chem. XXXX, XXX, XXX−XXX

Article

Figure 2. MD simulations of pBcl-2 (derived from PDB: 2XA0, phosphate groups were added to T69, S70, and S87) identiﬁed an allosteric

network connecting phosphorylation sites to BH3-binding sites. (a) pBcl-2 structure at the end of the simulation. (b−d) Structural parameters

(

hydrogen-bond distance and dihedral angle Ψ values) between diﬀerent residues characterizing the conformational change are shown as a function

of the simulation time. The x-axis presents the amino acid sequence number, and the y-axis presents the hydrogen-bond distance and dihedral angle

Ψ values. The light blue region highlights the narrow time window in which the concerted conformational change occurred.

signiﬁcance of Bcl-2 phosphorylation, our group established

Bcl-2-EK and EEE-Bcl-2-EK proteins. Bcl-2-EK was generated

by the cleavage of residues 2-61 in recombinant Bcl-2 to

enhance its solubility and stability, retaining the three

phosphorylation sites (T69, S70, and S87). EEE-Bcl-2-EK

was generated by replacing the three phosphorylation sites in

Bcl-2-EK with glutamic acids to mimic native multisite

phosphorylated Bcl-2₂_‑₂₀₆(pBcl-2). Reconstruction of the

binding functionality of native pBcl-2 by EEE mutagenesis

ns) suggested that R106 and R107 form backbone hydrogen

bonds with R110 and D111, respectively (Figure 2a,b). In

addition, Y108 and R110 are changed from being in a ﬂexible

state to forming a stable conformation, as shown by the

dihedral angle Ψ values that changed from ﬂuctuating (SD >

30°) to stabilizing at −42 ± 5 and −71 ± 6°, respectively

(Figure 2c). Consistent with 3D NOESY-HSQC, these results

demonstrated the formation of a continuous helical structure

in the region from R106 to D111 (Figure 2a). Simultaneously,

the phosphate group on T69 was observed to form a hydrogen

bond with the side chain of R109, and the phosphate group on

S70 is hydrogen-bound to R106 and R109 (Figure 2a,d). This

binding occurred over the same time period that the helical

structure formed, suggesting that the helical conformation was

stabilized by the newly formed arginine−phosphate interaction

network established between the side chains of pT69/pS70

and R106/R109. Because both Y108 and R110 are involved

with the BH3-binding interface, we proposed an extensive

allosteric network driven by hydrogen bonds between the

phosphate groups, with R106/R109 enabling communication

between the phosphorylation sites and the BH3-binding sites.

In addition, the crystal structure of Bcl-2 complexed with the

Bcl-2 inhibitor ABT-263 (PDB: 4LXD) or ABT-199 (PDB:

4MAN) suggested that these complexes favor a β-turn

conformation in the R106-D111 region (Figure S5), which

diﬀers from the helical conformation in the R106-D111 region

of EEE-Bcl-2-EK. These diﬀerences reasonably explain the

signiﬁcantly decreased aﬃnity of these two molecules toward

pBcl-2₂_‑₂₀₆, as shown herein (Table 1) and in previous reports;

that is, the helical conformational undergoes a transition in the

R106-D111 region.

30,31,35,36

was conﬁrmed both in vitro and in living cells.

Here, we ﬁrst perform 3D heteronuclear triple resonance

NMR spectroscopy of the C, N-labeled Bcl-2 isoform 2, Bcl-

-EK, and EEE-Bcl-2-EK proteins. Compared to those of the

Bcl-2 isoform 2, Bcl-2-EK residues showed signiﬁcant chemical

shift perturbations (CSPs) located in α5, α6, and α7 helices,

which may be due to these helices being spatially close to and

forming hydrogen bonds with the truncated α1 (CSP > 0.2

ppm, shown in red in Figures 1a and S2a ). However, when

introducing the E mutation into Bcl-2-EK, signiﬁcant CSPs

were additionally formed for D103, R106, and Y108 and were

accompanied by three newly discovered sharp resonance from

residues R109, R110, and D111 (shown in green, Figures 1b

and S2b ). Notably, the signals are located in the R106-R110

region between α2 and α4, which indicates a high degree of

ﬂexibility and shows diﬀerent conformations of Bcl-2/Bcl-xL in

0,31

complex with diﬀerent partners or artiﬁcial ligands.

This

disordered region is essentially ﬁxed following phosphomi-

micking mutations at sites approximately 40 residues away.

Consistently, 3D NOESY-HSQC spectra showed a series of

sequential NH/NH (i, i + 1) nuclear Overhauser eﬀects

(

NOEs) and medium-range C H/NH (i, i + 1) and C H/NH

i, i + 3) NOEs from R106 to D111 in EEE-Bcl-2-EK (Figures

Arginine−Phosphate Interaction Network Modulates

the Binding Proﬁle of the BH3 Groove. To investigate

whether the allosteric network discussed above manifests BH3-

binding aﬃnity, we performed R106A/R109A mutagenesis on

full-length Bcl-2₂_‑₂₀₆to construct Bcl-2_R₁₀₆_A_/_R₁₀₉_A, and the

prephosphorylated Bcl-2 variants pBcl-2₂_‑₂₀₆and pBcl-

2_R₁₀₆_A_/_R₁₀₉_Awere yielded using in vitro kinase phosphorylation.

The binding aﬃnities of Bax BH3 and Bim BH3 peptides for

these proteins were tested using a ﬂuorescence polarization

assay (FPA). Consistent with the ﬁndings of our groups and

c and S3 ), demonstrating the formation of a continuous

helical structure in the intrinsically disordered region upon Bcl-

phosphorylation (Figure 1d).

Based on the 3D NMR results, we performed an all-atom

explicit-solvent MD simulation. The simulation was initiated

from a homology-modeled full-length Bcl-2₂_‑₂₀₆protein (PDB:

XA0) with a loop region in which phosphate groups were

added to T69, S70, and S87 (Figures 2 and S4 ). Six separate

simulations ranging from 100 to 200 ns in length (totaling 600

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Table 1. Binding Aﬃnities of BH3 Peptides or Artiﬁcial

BH3 Ligands with Diﬀerent Bcl-2 Variants As Determined

by the Fluorescence Polarization Assay (FPA) and

Isothermal Titration Calorimetry (ITC) (K , nM)

Bcl-

R106A/R109A

pBcl-

^B^c^l^-²2

‑206

^p^B^c^l^-²2‑206

2.1 ± 0.5

21.4 ± 2.5

1803 ± 200

705 ± 144

R106A/R109A

Bax BH3

Bim BH3

ABT-263

ABT-199

19.6 ± 1.3

2.7 ± 0.6

20.6 ± 3.8

3.1 ± 0.9

15.5 ± 2.3

4.1 ± 0.6

27.8 ± 3.6

4.5 ± 1.2

16.2 ± 1.8

4.5 ± 0.8

30.6 ± 5.4

5.5 ± 1.8

All values represent means ± standard error of the mean (s.e.m.)

from three independent experiments. K determined by FPA. K

determined by ITC.

0,31,35,36

those of others,

pBcl-2₂_‑₂₀₆exhibited enhanced Bax

Figure 3. In vitro and in vivo dephosphorylation assay of Bcl-2

characterized that the arginine−phosphate interactions prevent the

phosphorylation sites from dephosphorylation. (a) In vitro

dephosphorylation assay of prephosphorylated pBcl-2₂_‑₂₀₆or pBcl-

2_R₁₀₆_A_/_R₁₀₉_Ain phosphatase assay buﬀer with or without PP2A-C. (b)

The level of pBcl-2(Ser70) in endogenous Bcl-2-silenced HL-60 after

Bcl-2₂_‑₂₀₆and Bcl-2_R₁₀₆_A_/_R₁₀₉_Atransfection. Protein levels were

determined by immunoblotting using speciﬁc antibodies, and ﬁgures

represent the results from n = 3 independent samples.

BH3 binding and reduced Bim BH3 binding. As shown in

Table 1 and Figure S6a , R106A/R109A mutagenesis in Bcl-

²2

‑206

had no inﬂuence on the binding aﬃnities for the Bax

BH3 and Bim BH3 peptides (15.5 vs 19.6 and 4.1 vs 2.7 nM

for Bcl-2_R₁₀₆_A_/_R₁₀₉_Aand Bcl-2₂_‑₂₀₆, respectively), R106A/R109A

in pBcl-2₂_‑₂₀₆abolished the eﬀect of Bcl-2 phosphorylation on

enhancing Bax BH3 binding and decreasing Bim BH3 binding,

resulting in K values similar to those of Bcl-2₂_‑₂₀₆(16.2 vs 19.6

and 4.5 vs 2.7 nM for pBcl-2_R₁₀₆_A_/_R₁₀₉_Aand Bcl-2₂_‑₂₀₆

respectively). In addition, the isothermal titration calorimetry

experiment (ITC) showed that the R106A/R109A mutation

attenuated the pBcl-2-induced lower binding aﬃnity for ABT-

carboxylic acid group to our previously identiﬁed Bcl-2/pBcl-2

dual-inhibitor compound S1-6, which has been shown to bind

in the BH3 groove of Bcl-2, whether phosphorylated or not,

but does not access the R106-D111 motif (Figure 4a). The

carboxylic acid group was linked to the 3-position of S1-6 via a

variety of linkers with lengths from 4.8 to 11.2 Å. The 3-CN in

S1-6 was demonstrated by NMR-derived molecular docking to

point to the R106-D111 motif from a distance of

approximately 10 Å, yielding S1-6a−S1-6e. Moreover,

compounds S1-6f−S1-6h, in which the carboxylic acid group

was removed (S1-6f) and replaced with a neutral hydrophilic

short PEG group (S1-6g) or with an oppositely charged

tertiary amino group (S1-6h), respectively, were designed as

the controls. Resulting compounds S1-6a−S1-6h were

synthesized according to the general synthetic route used for

S1-series compounds, which has been previously developed by

63 and ABT-199, resulting in K values similar to those of

Bcl-2₂_‑₂₀₆(Table 1 and Figure S6b ). Thus far, the results

identiﬁed that the phosphorylation of Bcl-2 in the loop region

modulates its binding proﬁle through an allosteric network

driven by arginine−phosphate interactions.

Arginine−Phosphate Interactions Prevent the De-

phosphorylation of Phosphorylation Sites. We next

determined whether the disruption of the interaction network

established by R106/R109 translates to the promotion of

phosphatase-dependent Bcl-2 dephosphorylation. An in vitro

dephosphorylation assay of recombinant phosphorylated Bcl-2

by phosphatase PP2A was carried out in a previously reported

cell-free EAT system, and the degree of Bcl-2 phosphor-

ylation was determined by western blotting. As shown in

Figure 3a, R106A/R109A mutagenesis signiﬁcantly accelerated

PP2A-induced Bcl-2 dephosphorylation, but this mutagenesis

did not induce Bcl-2 dephosphorylation without PP2A. To

further test this mechanism in cells, Bcl-2₂_‑₂₀₆and Bcl-

2_R₁₀₆_A_/_R₁₀₉_Awere transfected into pBcl-2-dependent HL-60

cells, in which endogenous Bcl-2 was silenced. As shown in

Figure 3b, a signiﬁcantly lower amount of phosphorylated Bcl-

36,38,39

our group

(Scheme 1).

To detect whether these compounds formed interactions

with R106 and R109, the panel of compounds was assayed via

FPA to determine the retention upon their binding to the Bcl-

2₂_‑₂₀₆and Bcl-2_R₁₀₆_A_/_R₁₀₉_Aproteins (Table 2 and Figure S7).

Compounds S1-6a−S1-6d all exhibited similar binding

aﬃnities for both Bcl-2

and Bcl-2

comparable

2‑206

R106A/R109A

with the aﬃnity exhibited by the parent molecule S1-6.

Notably, S1-6e, which harbored a p-benzoic acid group, caused

was detected in the Bcl-2_R₁₀₆_A_/_R₁₀₉_A-transfected HL-60 cells

than in the wild-type Bcl-2-transfected HL-60 cells. These

results demonstrated that the arginine cluster (R106/R109)

modulates the phosphorylation degree of Bcl-2 by blocking the

phosphorylation sites from exposure to phosphatase PP2A and

thus preventing their subsequent dephosphorylation.

Small-Molecule Probe Disrupting the Arginine−

Phosphate Interaction Network Allosterically Regulates

the Phosphorylation of Bcl-2. Considering the ﬁndings

described above, we aimed to design and characterize a small-

molecule Bcl-2-dephosphorylating probe located at the BH3

domain of Bcl-2 to regulate phosphorylation allosterically in

the loop region. To ﬂip the “conformational switch” between

the phosphates and the R106/R109 side chains, we added a

a corresponding ﬁvefold increase in binding aﬃnities (K =

0.16 vs 0.85 μM) for Bcl-2₂_‑₂₀₆, and this gain of aﬃnity was

precluded by R106A/R109A mutagenesis in Bcl-2. In contrast,

compounds S1-6f (without the carboxylic acid group) and S1-

6g (with a neutral hydrophilic short PEG group) exhibited

much lower binding aﬃnity (0.16 vs 1.05 and 0.77 μM) for

Bcl-2₂_‑₂₀₆, and their binding aﬃnity for Bcl-2 was not

signiﬁcantly inﬂuenced by R106A/R109A mutagenesis. More-

over, S1-6h, with an oppositely charged tertiary amino group,

showed a much lower binding aﬃnity for Bcl-2₂_‑₂₀₆compared

with the parent compound S1-6 (4.32 vs 0.85 μM), and

R106A/R109A mutagenesis recovered its capacity to bind to

Bcl-2. Collectively, these data suggested that the carboxylic

Journal of Medicinal Chemistry

J. Med. Chem. XXXX, XXX, XXX−XXX

Article

Figure 4. NOE-derived binding mode and dephosphorylation assays determined that S1-6e disrupted the arginine−phosphate interaction network

and allosterically dephosphorylated of Bcl-2. (a) Molecular formulas of S1-6−S1-6f. (b) NOESY-HSQC strip plot of the backbone NH bonds in

Bcl-2 showing NOE signals with the CH bonds in S1-6e. (c) NOE-derived molecular docking result of S1-6e with Bcl-2 using the Bcl-2

conformation in PDB: 4MAN. The residues showing obvious NOE signals with CH in S1-6e (in green) and the key residues R106, R107, and

R109 for the arginine−phosphate interaction network (in white) were shown as sticks. (d) In vitro dephosphorylation assay of prephosphorylated

pBcl-2₂or pBcl-2

upon the addition of S1-6e/S1-6f in phosphatase assay buﬀer with or without PP2A-C. (e) Level of pBcl-2(Ser70)

‑206

R106A/R109A

upon the addition of dimethyl sulfoxide (DMSO)/S1-6e/S1-6f in HL-60. (f) Level of pBcl-xl and pMcl-1 upon the addition of DMSO/S1-6e/S1-

6f in HL-60. All protein levels were determined by immunoblotting using speciﬁc antibodies, and ﬁgures represent the results from n = 3

independent samples.

Scheme 1. Synthesis Route of S1-6−S1-6h

(

a) malononitrile, CH CN, reﬂux; (b) K CO , CH CN, reﬂux; (c) 4-bromobenzenethiol, CH CN, reﬂux; (d) R H, CH CN, rt; (e) NaOH,

3 2 3 3 3 1 3

THF/H O, rt; (f) (o,m,p)-R -methanamine, hexaﬂuorophosphate azabenzotriazole tetramethyl uranium (HATU), N,N-di-isopropylethylamine

(

DIEA), CH Cl , rt; (g) NaOH, THF/H O, rt; (h) H N(CH CH O) H, HATU, DIEA, CH Cl , rt.

2 2 2 2 2 2 2 2 2

Table 2. Binding Aﬃnities of S1-6−S1-6g and S1-11 with Diﬀerent Bcl-2 Variants As Determined by FPA (K , μM)

^B^c^l^-²2

‑206

^B^c^l^-²R106A/R109A

1.67 ± 0.43

1.44 ± 0.17

0.94 ± 0.30

1.75 ± 0.25

1.67 ± 0.33

^B^c^l^-²2‑206

^B^c^l^-²R106A/R109A

1.65 ± 0.44

1.88 ± 0.31

0.77 ± 0.08

0.84 ± 0.11

0.15 ± 0.05

S1-6

0.85 ± 0.16

1.27 ± 0.28

0.78 ± 0.05

1.35± 0.15

1.02 ± 0.27

S1-6e

S1-6f

S1-6g

S1-6h

S1-11

0.16 ± 0.11

1.05 ± 0.28

0.81 ± 0.06

4.32 ± 1.08

0.21 ± 0.06

S1-6a

S1-6b

S1-6c

S1-6d

All values represent means ± s.e.m. from n = 3 independent experiments.

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

these cell lines. Compared to S1-11 , S1-6e exhibited a

Article

acid formed salt bridges and H-bonds with the R106 and R109

residues of Bcl-2, which endows the molecule with the

potential to regulate the phosphorylation of Bcl-2.

Furthermore, the binding mode of S1-6e to Bcl-2 was

clariﬁed by NOE-derived distance restraints (Figure 4b,c). S1-

Bcl-2 family proteins. Notably, S1-6e had no eﬀect on the

phosphorylation levels of Bcl-xl or Mcl-1, further validating the

dephosphorylating speciﬁcity of S1-6e (Figure 4f). Interest-

ingly, S1-11 showed binding aﬃnities for all Bcl-2 family

members similar to the aﬃnity of S1-6e (Table 3 and Figure

S8) and had no eﬀect on the phosphorylation level of Bcl-2

(Figure 4f). Therefore, we deemed that S1-6e and S1-11 may

be a proper pair of chemical tools to investigate the role of Bcl-

2 dephosphorylation in cell apoptosis.

e displayed NOE signals for H1, H2, and H3 in the 1H-

phenalen-1-one moiety to residues L137 and Ala149 in Bcl-2,

suggesting that S1-6e binds in the BH3 groove similar to the

parent molecule S1-6. Importantly, the NOE signal from H8 in

the S1-6e corresponding to residue Y108 and the signal from

H12 in the carboxylic acid group corresponding to residue

R106 illustrated that the carboxylic acid group of S1-6e points

toward and forms a hydrogen bond with R106 and R109.

Considering that the parent compound S1-6 is a pan-Bcl-2

family inhibitor, we further investigated whether S1-6e binds

to and dephosphorylates other Bcl-2 family members, such as

To evaluate the cellular activity, we exposed four cell lines

with a similar background of the Bcl-2 protein level (Figure

24,30,31

5a)

to S1-6e, S1-6f, S1-11, and the FDA-approved Bcl-2

inhibitor ABT-199. Among them, a high level of Bcl-2

phosphorylation occurs in HL-60 and OCI-AML3 cell lines,

while pBcl-2 expresses at a signiﬁcantly lower level in MCF-7

24,30,31

and KG-1 cell lines.

As shown in Figure 5a and Table 4,

Bcl-xl and Mcl-1. Compounds S1-6, S1-6f, and S1-11 (a

derivative of S1-6 exhibiting binding aﬃnity for Bcl-2

comparable with that of S1-6e) were tested in parallel. The

binding aﬃnities of S1-6e for Bcl-xl and Mcl-1 are comparable

to those of S1-6 and S1-6f (Table 3), suggesting that the

Table 4. EC₅₀Values of ABT-199, S1-6e, S1-6f and S1-11 to

Induce Apoptosis in HL-60, OCI-AML3, KG-1, and MCF-7

Cells at 48 h, As Determined by an Annexin V Staining

Assay

HL-60

OCI-AML3

KG-1

MCF-7

Table 3. Binding Aﬃnities of S1-6, S1-6e, S1-6f, and S1-11

ABT-199

S1-6e

S1-6f

8.1 ± 2.2

3.8 ± 1.2

21.7 ± 2.4

16.9 ± 3.3

6.8 ± 1.6

4.0 ± 0.5

22.3 ± 3.5

15.4 ± 2.1

0.85 ± 0.16

4.9 ± 1.6

16.8 ± 1.2

2.7 ± 1.1

1.35 ± 0.15

5.5 ± 1.3

18.5 ± 3.2

3.1 ± 0.8

with Bcl-2, Bcl-xl, and Mcl-1 As Determined by FPA (K ,

μM)

S1-6

S1-6e

S1-6f

S1-11

Bcl-2

Bcl-xl

Mcl-1

0.85 ± 0.16

0.75 ± 0.18

0.49 ± 0.14

0.16 ± 0.11

0.81 ± 0.12

0.32 ± 0.12

1.05 ± 0.28

1.11 ± 0.33

0.56 ± 0.07

0.21 ± 0.06

0.55 ± 0.07

0.35 ± 0.15

All values represent means ± s.e.m. from n = 3 independent

experiments.

All values represent means ± s.e.m. from n = 3 independent

along with increasing level of pBcl-2, HL-60 and OCI-AML3

cells exhibited 5−10-fold resistance to S1-11 and ABT-199

than the MCF-7 and KG-1 cells. In contrast, S1-6e showed

comparable EC₅₀values for apoptosis induction among all of

experiments.

favorable interactions between the carboxylic acid in S1-6e

with the R106 and R109 residues are not conserved in other

Figure 5. S1-6e exhibited a dual-function mechanism that contributed to the anticancer activity in pBcl-2-dependent cancer cells. (a) Level of pBcl-

in HL-60, OCI-AML3, MCF-7, and KG-1 determined by immunoblotting and co-IP using speciﬁc antibodies. (b) PARP cleavage analyzed in

HL-60, OCI-AML3, MCF-7, and KG-1 cells treated with S1-6e, S1-6f, S1-11, and ABT-199 for 24 h. (c) Levels of the Bax/Bcl-2 complex, pBcl-

(Ser70), and total Bcl-2 in HL-60, OCI-AML3, MCF-7, and KG-1 treated with S1-6e, S1-6f, S1-11, and ABT-199 determined by immunoblotting

and co-IP using speciﬁc antibodies. All ﬁgures represent the results from n = 3 independent samples.

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

signiﬁcantly lower EC₅₀in the HL-60 and OCI-AML3 cells

and a slightly higher EC₅₀in the MCF-7 and KG-1 cells,

indicating that S1-6e can overcome the resistance induced by

Bcl-2 phosphorylation. The capabilities of these compounds to

induce apoptosis were further validated by PARP cleavage

experiments (Figure 5b). Furthermore, immunoblotting and

coimmunoprecipitation (co-IP) assays showed that S1-6e

induced a rapid decrease in Bcl-2 phosphorylation levels in

HL-60 and OCI-AML3 accompanied by the release of

proapoptotic Bax from the Bcl-2 complexes. In contrast,

none of S1-6f, S1-11, and ABT-199 induced a decrease in

pBcl-2 levels during the 24 h treatment period, and their

capacity for releasing Bax signiﬁcantly decreased in HL-60 and

OCI-AML3 than in the other two cell lines in which Bcl-2

proteins are much less phosphorylated (Figure 5c). Taken

together, these results suggest that S1-6e takes advantage of a

dual-function mechanism to overcome cancer resistance to Bcl-

molecule cage”, yielding S1-6e. S1-6e exhibits site-speciﬁc

dephosphorylation of pSer70 in Bcl-2 due to its speciﬁc

binding to the BH3 groove and projecting of the carboxylic

acid group speciﬁcally toward R106/R109. Moreover, the

favorable binding energy of the inhibitor itself compensates for

the weak negative charge and the hydrated shell of the carboxyl

group, which are weaker and smaller, respectively, than are

exhibited by the phosphate group, to compete with the

diarginine molecule cage for access to pSer70. As expected,

dephosphorylation of Bcl-2 accompanied higher lethality of

tumor cells with high pBcl-2 levels than was induced by single-

function Bcl-2 inhibitors, as shown by S1-6e.

S1-6e realize target-speciﬁc and site-speciﬁc dephosphor-

ylation of Bcl-2, with no obvious impact on the phosphor-

ylation levels of other PP2A target proteins. To our knowledge,

S1-6e is the ﬁrst site-speciﬁc artiﬁcial dephosphorylation

modulator in the world, in contrast to the widely used

phosphatase modulators that regulate the dephosphorylation

of a mass of proteins indiscriminately. The in vivo anticancer

activity of the compounds should be further evaluated in the

future to translate this mechanistic study into drug discovery.

Notably, arginine involved in the allosteric signal prop-

agation pathway oﬀers a general mechanism for the regulation

of phosphorylation in other proteins, either in the Bcl-2 family

or others. For example, a similar hydrogen-bond network

between pSer62 and the side chains of R100 and R103 is also

present in the Bcl-xL protein to drive allosteric signaling

inhibitors by Bcl-2 phosphorylation.

DISCUSSION

■

Having investigated the allosteric eﬀects of phosphorylation in

the loop region of the BH3-binding groove using ABT

compounds and small-molecule Bcl-2 inhibitors developed by

our group, we studied the structural basis of the phosphor-

ylation-induced allosteric alterations in the distal BH3-binding

groove. Phosphorylation can add a negatively charged center

within the loops, which is well suited for interacting with the

positively charged side chains of arginine, lysine, and histidine

communications. Therefore, it is reasonable to assume that

adding a carboxylic acid group on a Bcl-xL inhibitor can lead to

its suﬃcient dephosphorylation. Additionally, similar hydro-

gen-bond/salt bridge networks between phosphate and

arginine residues are found in many kinase proteins, such as

on the surface of the folded domains. In this manner, the

hydrogen-bond/salt bridge network established between

phosphoamino acids and these alkaline residues has the

potential to create an allosteric pathway through which the

signal propagates between loops and folded domains.

However, in most cases, the pathways are poorly characterized,

and therefore, no such artiﬁcial phosphorylation modulators

have been reported before this report.

In the present study, we revealed that the Bcl-2 protein

structurally couples phosphorylation in the loop region to the

hotspot residues in the active BH3 groove by the formation of

a ﬁnely tuned network of interactions between the multiple

phosphorylation sites and the α2 helix, which is based on a

hydrogen-bond network formed between phosphate groups on

T69 and S70 and the side chains of R106 and R109, driving a

helical conformational change in the R106-D111 region that is

dynamically coupled with repositioning Y108 and R110 in the

BH3-binding groove. In this manner, we illustrated how the

BH3 groove integrates phosphorylation signals to regulate the

biological functions of Bcl-2 and reshape the Bcl-2 antitumor

network. Bcl-2 phosphorylation locks Y108 and R110 in Bax-

favored and Bim-disfavored conformations.

Moreover, in the interaction network, R106/109 and pSer70

serve as the key nodes in the allosteric pathway for allosteric

signal propagation. NMR together with MD revealed that

R106/R109 provides an electrostatic molecular cage to shield

the phosphorylated site and protect pSer70 from phosphatase

attack. An opposite eﬀect was observed when these direct

electrostatic interactions were destroyed upon R106A/R109A

double mutation. Therefore, we decided to uncover the

molecular protecting cage for pSer70 by constructing hydro-

gen-bond/salt bridge networks with R106/R109 competitively

with pSer70. Based on the binding mode of a previously

developed pBcl-2 inhibitor, a carboxylic acid group was added

at an appropriate substitution site pointing to the “diarginine

37,40,42

CDK2, AKT, and p38g,

to prevent the kinases from

dephosphorylation, suggesting the success of our dephosphor-

ylation strategy on these tumor targets of a family very diﬀerent

from that of Bcl-2. Even tracing to the lower kingdom of life,

we found a similar structural coupling between a core-

functional groove and phosphorylation sites for cheY and

43,44

FLiM proteins,

in which the weak hydrogen-bond donor

threonine takes the place of arginine. Taken together, these

ﬁndings illustrate that the allosteric pathways interconnecting

the core-functional domain and the phosphorylation sites are

found in many proteins, suggesting that targeting the allosteric

pathway by modifying inhibitors located in core-functional

domains described herein can serve as a general molecular

design strategy for site-speciﬁc dephosphorylation probes.

CONCLUSIONS

■

In conclusion, our study reveals the structural mechanism of

the allosteric pathway for communication between the

phosphorylation sites in intrinsically disordered loops and

the core-functional BH3 groove in Bcl-2, which was identiﬁed

as a potential avenue for molecular drug design based on site-

speciﬁc dephosphorylation. A Bcl-2-dephosphorylating probe,

S1-6e, was successfully generated by adding a carboxylic acid

group to a Bcl-2 inhibitor, which is located in the BH3 groove.

The carboxyl group in S1-6e forms salt bridges and H-bonds

with the R106/R109 residues, which disrupt the interaction

network between R106/R109 and pT69/pS70 and subse-

quently facilitate Bcl-2 dephosphorylation in living cells. As a

result, S1-6e manifests a more eﬀective apoptosis induction

mechanism in pBcl-2-dependent cancer cells than other

inhibitors exhibiting a similar binding aﬃnity for Bcl-2. We

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

believe that targeting the allosteric pathways interconnecting

the core-functional domain and the phosphorylation site can

be a general strategy for the rational design of site-speciﬁc

dephosphorylating probes, since this type of allosteric pathway

has been discovered in a variety of proteins.

To gain EEE-Bcl-2-EK62‑206 or Bcl-2-EK62‑206 proteins, the

concentrated target protein was ﬁrst bound to the aﬃnity column.

Then, the column was washed with 20 vol Tris-buﬀer (20 mM Tris−

HCl, pH 8.0) and the solution containing enterokinase was added to

the aﬃnity column by an injector. The cleavage reaction incubated at

°C for 36 h, and the cleaved protein was eluted with Tris-buﬀer.

The puriﬁed protein was stored in Tris-buﬀer containing 150 mM

NaCl, 5 mM DTT, and 0.02% sodium azide and concentrated to a

ﬁnal concentration of 0.3−0.5 mM.

EXPERIMENTAL SECTION

■

Reagents. The 36-mer Bax BH3 (residues 49−84) and Bim BH3

Protein concentrations were calculated by the Bradford method

using BSA as the standard. The Bcl-2 isoform 2 expression plasmid

was a gift from Dr. Renxiao Wang, and its expression and puriﬁcation

(

residues 48−83) peptides were synthesized at HD Biosciences

>98% purity, ShangHai, China). The HL-60 cell line was were

purchased from the Cell Bank of the Chinese Academy of Sciences

and used within 3 months. The cells were maintained in a minimum

essential medium supplemented with 10% FCS and 100 U/mL

streptomycin and penicillin. The Annexin V-FITC reagent was from

Invitrogen. Antibodies speciﬁc for Bcl-2, phospho-Ser70-Bcl-2 (Cell

Signaling, Beverly, MA), actin, Bax, Bim, and PARP (Abcam,

Cambridge, MA) were used. ABT-263 and ABT-199 were purchased

from Selleckchem (Houston, TX). The compounds and antibodies

were sealed and maintained at −20 °C and used within 3 months.

Protein Expression and Puriﬁcation. The construction of the

full-length Bcl-2₂_‑₂₀₆expression plasmid was performed by a pairwise

codon-optimized full-length human Bcl-2 gene that was synthesized

by Takara (Dalian, Liaoning, China) and subcloned into pET-28a (+)

Invitrogen, Carlsbad, CA). This cloning generated a construct in

which full-length Bcl-2₂_‑₂₀₆was produced with an N-terminal His -tag.

Nucleotides corresponding to both R106 and R109 were

substituted to create an alanine acid (A) with a site-directed

^m^u^t^a^g^eⁿ^e^sⁱ^s^kⁱ^t⁽^C^l^oⁿ^t^e^c^h^,^B^eⁱ^jⁱⁿ^g^,^C^hⁱⁿ^a⁾^,^yⁱ^e^l^dⁱⁿ^g^B^c^l^-²R106A/R109A

Nucleotides corresponding to 70, 87 serine (S), or 69 threonine (T)

residues were substituted to create a triple glutamic acid (E) with a

^sⁱ^t^e^-^dⁱ^r^e^c^t^e^d^m^u^t^a^g^eⁿ^e^sⁱ^s^kⁱ^t^.^T^h^e^g^eⁿ^e^eⁿ^c^o^dⁱⁿ^g^E^E^E^-^B^c^l^-²2

were done as previously described.

NMR Spectroscopy. The 3D NMR spectra were recorded for 0.5

mM N/ C-labeled EEE-Bcl-2-EK or Bcl-2-EK in buﬀer (20 mM

Tris (pH 7.8), 150 mM NaCl, 5 mM DTT-d , 0.02% NaN , and 90%

H 0/10% D O) at 25 °C. The 3D NMR spectra were recorded on a

Bruker 800 MHz spectrometer equipped with a cryoprobe. Backbone

resonance assignments for Bcl-2-EK and EEE-Bcl-2-EK were carried

out using 3D HNCA, 3D HN(CO)CA, 3D HNCO, 3D HN(CA)CO,

and 3D NOESY-HSQC (50−100 ms mixing times). The [ N, H]

HSQC spectra were recorded for 0.35 mM N-labeled EEE-Bcl-2-EK

or Bcl-2-EK on a Bruker Advance DRX600 MHz spectrometer. The

NMR data were processed using NMRPipe (Delaglio et al., 1995) and

analyzed with Sparky (Goddard and Kneller, Sparky 3.114, UCSF,

San Francisco) and the program NMRView. Chemical shift changes

(

2 0.5

were calculated as ((Δ H ppm) + (0.2 × Δ N ppm) ) as a

function of the residue number.

The expected chemical shift (CS) calculations were carried out

using the programs SPARTA+ and SHIFTX2. The average CS

changes were calculated from the simulation using the last 5 ns (500

frames) for pBcl-2 and npBcl-2.

‑206

Unbiased Molecular Dynamics Simulations. All unbiased all-

atom molecular dynamics simulations were performed using the

Gromacs4.6.5 package. First, we modeled the npBcl-2 protein,

including the optimized ﬂexible loop domain (FLD), using

MODELLER. The modeled structure was analyzed by Ramachan-

dran’s map using PROCHECK (Programs to check the stereo-

chemical Quality of Protein Structures), and the environmental proﬁle

was assessed using ERRAT graphs (Structure Evaluation server). Prior

to the construction of pBcl-2, six independent simulations of 200 ns

each were performed on the npBcl-2 model to equilibrate the system.

We then constructed pBcl-2 by adding three phosphates to the T69,

S70, and S87 residues in the loop domain using the PTM2.0 server.

Finally, the six simulations were performed for another 200 ns to

obtain a ﬁnal pBcl-2 structure.

^B^c^l^-²2

‑206

was introduced with an enterokinase-cleaved site between

or Bcl-

^A⁶¹^aⁿ^d^S⁶²ⁱⁿ^aⁿ^a^t^t^e^m^p^t^t^o^p^r^o^d^u^c^e^aⁿ^E^E^E^-^B^c^l^-²^-^E^K6

^-^E^K6

2‑206

protein.

2‑206

^T^h^e^f^u^l^l^-^l^eⁿ^g^t^h^B^c^l^-²2

and various mutants were produced by

‑206

cell-free synthesis with the addition of detergent Brij 58 (1% w/v).

Brieﬂy, Escherichia coli S12 extract-based expression was conducted in

batch mode with plasmid DNA (10 μg/mL) as the template in an

Eppendorf thermomixer at 30 °C for 150 min with agitation at 800

rpm. Protein was puriﬁed on Ni-NTA resin (Qiagen) by following the

manufacturer’s instructions. The protein was further puriﬁed on a

Source Q15 column in Tris-buﬀer (25 mM, pH 8.0) with a NaCl

gradient.

^T^h^e^E^E^E^-^B^c^l^-²^-^E^K6

2‑206

^o^r^B^c^l^-²^-^E^K62‑206 proteins were produced

in E. coli BL21 (DE3) cells. Cells were grown at 37 °C in LB

containing kanamycin (30 μg/mL). When the cell density reached

^O^D⁶00 ∼ 0.8, the protein expression was induced by the addition of

isopropyl-β-D-thiogalactoside (IPTG) to a ﬁnal concentration of 0.6

mM for 5 h at 37 °C. The cells were resuspended and lysed in lysis

buﬀer (50 mM Tris−HCl, pH 7.9, 0.1 mM ethylenediaminetetra-

acetic acid (EDTA), 0.1 mM dithiothreitol (DTT), and 150 mM

NaCl). The lysate was then centrifuged at 9000 rpm for 30 min. The

pellet was washed with a washing buﬀer (lysis buﬀer containing 1%

Triton) once and lysis buﬀer twice. Puriﬁed inclusion bodies were

solubilized in 6 M guanidine HCl, 10 mM Tris−HCl, pH 7.9, 10 mM

DTT, 1 mM EDTA, for 2 h. The denaturation protein solution was

dissolved by pulse sonication for 2 min to dissolve more inclusions.

The guanidine-solubilized protein was centrifuged at 9000 rpm for 20

min, and the supernatant was diluted to 0.1−0.5 mg/mL. Then, it was

stored at 4 °C against dialyzed buﬀer A (2 M guanidine HCl, 0.4 M

arginine, 1 mM dithiothreitol, 10 mM Tris−HCl pH 7.9), dialyzed

buﬀer B (1 M guanidine HCl, 0.4 M arginine, 1 mM dithiothreitol, 10

mM Tris−HCl pH 7.9), and dialyzed buﬀer C (0.4 M arginine, 1 mM

dithiothreitol, 10 mM Tris−HCl pH 7.9) for 24 h. Finally, it was

stored twice against 10 vol borate-buﬀered saline (150 mM NaCl,

The MD simulation was performed using the CHARMM36 force

ﬁeld in the GROMACS format (including CGenFF version 4.0 and

the CHARMM36 m protein force ﬁeld revision) and the

AMBER99SB-ILDN all-atom force ﬁeld. All molecules were

solvated in a dodecahedron box with periodic boundary conditions

using TIP3P water molecules, and the simulation was performed in

the NPT ensemble at room temperature. The analyses were

performed using Gromacs inbuilt tools (VMD1.9.1) and PyMOL.

In Vitro Phosphorylation. To achieve a high degree of

modiﬁcation, 100 μg of puriﬁed Bcl-2₂_‑₂₀₆or Bcl-2_R₁₀₆_A_/_R₁₀₉_Awas

incubated with 5 μg of GST-tagged human recombinant JNK (EMD

Millipore) at 30 °C for 90 min. Aliquots of the reaction mixture were

subjected to sodium dodecyl sulfate-polyacrylamide gel electro-

phoresis (SDS-PAGE), transferred to nitrocellulose, and immuno-

blotted with anti-phospho-Ser70-Bcl-2 to conﬁrm that most of the

Bcl-2 proteins were phosphorylated.

In Vitro Dephosphorylation in Cell-Free EAT System.

Prephosphorylated pBcl-2₂_‑₂₀₆or pBcl-2_R₁₀₆_A_/_R₁₀₉_A, with or without

compounds (10 μM in the ﬁnal volume), was aliquoted in

phosphatase assay buﬀer (500 mM ATP, 50 mM HEPES, 100 mM

NaCl, 10 mM MgCl , 1 mM DTT, pH = 7.5) and incubated at 30°C

.01 M sodium borate, pH 8.5) at 4 °C for 24 h and centrifuged for

for 10 min. PP2A-C (80 ng) was added, and the mixture (ﬁnal volume

= 30 μL) was then incubated at 30°C for 50 min. Phosphatase

reactions were terminated with SDS sample buﬀer, and the

0 min at 5000 rpm. The supernatant was concentrated to 50−70 mL

for ﬁnal aﬃnity chromatography puriﬁcation.

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

phosphorylation level of Bcl-2 Ser70 was analyzed by immunoblot-

ting.

Transfection of cells was performed with lipofectamine (Invi-

trogen) according to the manufacturer’s instructions. Under our

conditions, 20−30% of cells were routinely transfected. For stable

transfection, the transfected cells were selected by the addition of

Geneticin (G418), purchased from Invitrogen (Grand Island, NY), to

the medium at a concentration of 600 μg/mL. After 6 weeks, stably

transfected selected cells were further cultured with G418 at a

concentration of 200 μg/mL.

Immunoblotting and Coimmunoblotting. Cell lysates were

prepared in CHAPS buﬀer (1% CHAPS, w/v, 100 mM NaCl, 5 mM

Na PO , 2.5 mM EDTA, 1 μg/mL leupeptin, 1 μg/mL aprotinin, and

μM PMSF). After 40 min on ice, the lysates were cleared by

centrifugation at 12 000g for 30 min at 4 °C. Total protein (100 μg)

was separated by SDS-PAGE, electrotransferred to poly(vinylidene

diﬂuoride) membranes, and analyzed by following standard

procedures. Signals were detected using Super Signal West Femto

Apoptosis Assays. Phosphatidylserine (PS) exposure was

quantiﬁed by surface Annexin V-FITC staining. Cells were seeded

(

#34096, Thermo Fisher Scientiﬁc Inc., Barrington, IL).

in each well of six-well tissue culture plates (5 × 10 cells/well) and,

For cytochrome c release assay, cells were lysed in digitonin lysis

24 h later, were treated with compounds for 48 h. Control cells were

treated with an equivalent concentration of the solvent DMSO. Cells

were washed twice with phosphate-buﬀered saline (PBS) and

incubated with a 1:40 solution of FITC-conjugated Annexin V for

10 min at room temperature. Stained cells were analyzed by ﬂow

cytometry.

buﬀer (75 mM NaCl, 8 mM Na HPO , 1 mM NaH PO , 1 mM

EDTA, 350 μg/mL digitonin). After centrifugation, the supernatant

S-100, cytosolic fraction) was collected and subjected to

(

immunoblotting.

For the coimmunoprecipitation assay, cells were lysed in 1%

CHAPS buﬀer. Total proteins (500 μg) were precleared with protein

A-Sepharose and incubated overnight with 5 μg of speciﬁc antibody.

Immunocomplexes were captured with either protein A-Sepharose or

protein G-Agarose (Sigma Chemical Co., St. Louis, MO). The

presence of immunocomplexes was determined by immunoblotting.

Chemistry. All commercial reagents were purchased and used

without further puriﬁcation. H NMR and C NMR spectra were

obtained with a Bruker AV-500 spectrometer using TMS as an

internal standard. Chemical shifts were shown in ppm. High-

resolution mass spectra (HRMS) were obtained on an HPLC−Q-

ToF-MS (Micro) spectrometer. Column chromatography was

performed on 200−300 mesh silica gel. The purity of all ﬁnal

products was determined by normal-phase HPLC (XBridge C18, 4.6

mm × 150 mm, 5 μm) to be ≥95%.

Aﬃnity Measurement by FP and ITC. To assay the K values of

BH3 peptides toward diﬀerent Bcl-2 variants, Bcl-xl and Mcl-1, the

BH3 peptide was labeled at the N-terminus with 6-carboxyﬂuorescein

succinimidyl ester (FAM) as the ﬂuoresce tag. FAM-BH3 peptide was

dissolved to 10 nM in the assay buﬀer (100 mM potassium

phosphate, pH = 7.5; 100 μg/mL bovine γ globulin; 0.02% sodium

azide), and 5 μL of gradient concentrations of the Bcl-2 protein

solution was added. After 30 min incubation, the polarization values

were measured using a Spectra Max M5 Detection System in a black

The synthetic procedures and NMR/HRMS data of S1-6, S1-6a,

27−29

and S1-6b compounds were previously reported by our group.

General Procedure for the Synthesis of S1-6c−S1-6h. To a

suspension of S1-6b (0.5 mmol, 1 equiv) in CH Cl (10 mL) were

added corresponding R NH (1.1 equiv, 0.55 mmol) and HATU (1.2

6-well plate at an excitation wavelength of 485 nm and an emission

equiv, 0.6 mmol) successively. After stirring at rt for 6 h, the solvent

was removed under reduced pressure, and the crude was puriﬁed on

silica gel (CH Cl /CH OH = 50) to yield intermediates 3−5 and

wavelength of 530 nm. The IC₅₀value was determined from the

saturation binding curve to give the K_Dvalue. For the competitive

binding assay for S1-6-series compounds, FAM-Bim peptide (10 nM)

and diﬀerent Bcl-2 variants protein (60 nM)/Bcl-xl (60 nM)/Mcl-1

ﬁnal compounds S1-6f, S1-6e, and S1-6h (yields = 43−80%). Then, a

mixture of intermediates 3−5 (0.3 mmol, 1 equiv) and NaOH (6

(

55 nM) were preincubated in the FP assay buﬀer. Then, compounds

mmol, 20 equiv) in the THF/H O (10:1, 11 mL) solvent was stirred

were added in serial dilutions, and the polarization values were

measured by the same means. The IC₅₀value was determined from

for 12 h, poured into ice water, and acidiﬁed with 10% hydrochloric

acid. The precipitate thus formed was collected by ﬁltration, washed

with water, and dried under vacuum to give ﬁnal compounds S1-6c,

S1-6d, and S1-6e (yields = 90−95%).

the competitive binding curve, and the K values were calculated using

the equation

-((3-((6-((4-Bromophenyl)thio)-2-cyano-1-oxo-1H-phenalen-3-

K = IC /([L] /K + [P] /K + 1)

yl)amino)propanamido)methyl)benzoic acid (S1-6c). Yellow solid.

H NMR (500 MHz, DMSO-d ): δ 11.25 (s, 1H), 8.61 (d, J = 8.4 Hz,

IC , the concentration of the free inhibitor at 50% inhibition; [L] ,

the concentration of the free labeled ligand at 50% inhibition; and

1H), 8.58−8.55 (m, 2H), 7.97 (d, J = 7.8 Hz, 1H), 7.91 (t, J = 8.4 Hz,

1H), 7.68−7.66 (m, 3H), 7.53−7.49 (m, 2H), 7.43 (t, J = 7.8 Hz,

2H), 7.38 (t, J = 7.8 Hz, 1H), 4.36 (s, 2H), 4.22 (t, J = 6.4 Hz, 2H),

[

P] , the concentration of the free protein at 0% inhibition.

To assay the K values of ABT-199 or ABT-263 toward diﬀerent

2.80 (t, J = 6.4 Hz, 2H); C NMR (100 MHz, DMSO-d ): δ 178.2,

Bcl-2 variants, isothermal titration calorimetry (ITC) experiments

were performed using a high-precision iTC₂₀₀(Microcal LLC,

Northampton, MA). In this experiment, the heat of the reaction

was obtained by making 16 injections (2.4 μL each) of the ligand

175.3, 173.6, 169.8, 139.1, 138.4, 134.9, 134.5, 133.6, 132.3, 131.8,

131.5, 130.4, 130.0, 129.7, 128.8, 128.1, 126.6, 125.9, 125.3, 121.7,

116.0, 78.5, 40.6, 39.8, 31.9; [ToF-MS (ESI )]: C H BrN O S [M +

H ], calcd 612.0587, found 612.0582.

(

ﬁnal concentration in the syringe of 200 μM in 20 mM Tris, 150 mM

3-((3-((6-((4-Bromophenyl)thio)-2-cyano-1-oxo-1H-phenalen-3-

NaCl, and 5% DMSO, pH 7.8) into the reaction cell that contained

Bcl-2 in the same buﬀer (ﬁnal protein concentration of 20 μM). The

experiments were carried out at 25 °C and a stirring speed of 500

rpm. Then, the resultant titration data were ﬁtted in Origin 7.5

yl)amino)propanamido)methyl)benzoic acid (S1-6d). Yellow solid.

H NMR (500 MHz, DMSO-d ): δ 10.77 (s, 1H), 8.63 (d, J = 8.4 Hz,

1H), 8.58 (d, J = 8.4 Hz, 1H), 8.56 (d, J = 8.4 Hz, 1H), 8.31−8.28

(dd, J = 8.0 Hz, 2H), 7.88 (t, J = 8.4 Hz, 1H), 7.81 (d, J = 8.0 Hz,

1H), 7.68 (t, J = 7.8 Hz, 2H), 7.59 (t, J = 8.0 Hz, 1H), 7.49 (d, J = 8.4

Hz, 1H), 7.42 (t, J = 7.8 Hz, 2H), 4.31 (s, 2H), 4.20 (t, J = 6.4 Hz,

(

Microcal Software Inc., Northampton, MA). All sample data

obtained after data corrections were analyzed to ﬁt to a one-site

model. For the control experiment, the sample cell was ﬁlled with the

assay buﬀer and the compound solution was added.

Cell Line Transfection. Human Bcl-2 cDNA was cloned in

pUC19 plasmid. For cellular expression of Bcl-2₂_‑₂₀₆, PCR-ampliﬁed

full-length Bcl-2 open reading frames were cloned in the pCIneo

vector to construct the WT-Bcl-2-expressing vector. For cellular

expression of Bcl-2_R₁₀₆_A_/_R₁₀₉_A, nucleotides corresponding to 106 and

2H), 2.78 (t, J = 6.4 Hz, 2H); C NMR (100 MHz, DMSO-d ): δ

178.5, 175.4, 173.8, 169.7, 141.5, 139.4, 134.7,132.4, 132.1 131.8,

130.7, 130.1, 128.7, 128.4, 126.2, 125.3, 121.9, 115.7, 40.3, 39.9, 31.8;

[ToF-MS (ESI )]: C H BrN O S [M + H ], calcd 612.0587, found

612.0594.

4-((3-((6-((4-Bromophenyl)thio)-2-cyano-1-oxo-1H-phenalen-3-

yl)amino)propanamido)methyl)benzoic acid (S1-6e). Yellow solid.

09 arginine (R) residues were substituted to create an alteration to

H NMR (500 MHz, DMSO-d ): δ 10.95 (s, 1H), 8.60 (d, J = 8.4 Hz,

alanine (A) with a site-directed mutagenesis kit and then cloned into

pCIneo to construct the Bcl-2 (R106A/R109A)-expressing vector. All

constructs were veriﬁed by sequencing.

1H), 8.58 (d, J = 8.4 Hz, 1H), 8.55 (d, J = 8.4 Hz, 1H), 7.89 (t, J = 8.4

Hz, 1H), 7.72 (t, J = 8.0 Hz, 2H), 7.65 (t, J = 7.8 Hz, 2H), 7.48 (d, J =

8.4 Hz, 1H), 7.41 (t, J = 7.8 Hz, 2H), 7.27 (t, J = 8.0 Hz, 2H), 4.33 (s,

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Technology, Dalian, Liaoning 116024, China; orcid.org/

Article

(

H), 4.19 (t, J = 6.8 Hz, 2H), 2.76 (t, J = 6.8 Hz, 2H); ¹³C NMR

100 MHz, DMSO-d ): δ 178.0, 175.2, 173.7, 169.3, 143.2, 138.8,

0000-0002-2774-2384 ; Email: zczhang@dlut.edu.cn

34.8, 132.2, 131.9, 131.4, 130.6, 130.0, 128.7, 128.4, 128.1, 126.7,

25.8, 125.2, 121.4, 115.8, 78.2, 40.1, 39.9, 31.8; [ToF-MS (ESI )]:

Authors

C H BrN O S [M + H ], calcd 612.0587, found 612.0591.

Ziqian Wang − State Key Laboratory of Fine Chemicals,

Zhang Dayu School of Chemistry, Dalian University of

Technology, Dalian, Liaoning 116024, China

N-Benzyl-3-((6-((4-bromophenyl)thio)-2-cyano-1-oxo-1H-phena-

len-3-yl)amino)propanamide (S1-6f). Yellow solid. H NMR (500

MHz, DMSO-d ): δ 8.61 (d, J = 8.4 Hz, 1H), 8.58 (d, J = 8.4 Hz,

Zongwei Guo − School of Life Science and Technology, Dalian

University of Technology, Dalian, Liaoning 116024, China

Keke Cao − School of Life Science and Technology, Dalian

University of Technology, Dalian, Liaoning 116024, China

Chao Chen − Qingdao Institute of Bioenergy and Bioprocess

Technology, Chinese Academy of Sciences, Qingdao,

Shandong 266101, China

Yingang Feng − Qingdao Institute of Bioenergy and Bioprocess

Technology, Chinese Academy of Sciences, Qingdao,

Shandong 266101, China; orcid.org/0000-0002-0879-

1316

Hang Wang − School of Innovation and Entrepreneurship,

Dalian University of Technology, Dalian, Liaoning 116024,

China

Fangkui Yin − State Key Laboratory of Fine Chemicals, Zhang

Dayu School of Chemistry, Dalian University of Technology,

Dalian, Liaoning 116024, China

H), 8.56 (d, J = 8.4 Hz, 1H), 7.88 (t, J = 8.4 Hz, 1H), 7.72 (t, J = 8.0

Hz, 2H), 7.48 (d, J = 8.4 Hz, 1H), 7.37 (m, J = 7.8 Hz, 2H), 7.33 (t, J

7.8 Hz, 1H), 7.28 (t, J = 8.0 Hz, 2H), 7.25 (t, J = 7.8 Hz, 2H), 4.32

(

s, 2H), 4.20 (t, J = 6.8 Hz, 2H), 2.78 (t, J = 6.8 Hz, 2H); C NMR

(

100 MHz, DMSO-d ): δ 178.1, 175.4, 173.9, 138.7, 137.8, 134.9,

34.6, 132.3, 132.1, 131.9, 131.5, 131.0, 130.6, 128.8, 128.6, 128.1,

27.0, 126.6, 125.8, 125.0, 121.6, 115.8, 78.5, 40.4, 39.7, 31.6; [ToF-

MS (ESI )]: C H BrN O S [M + H ], calcd 568.0689, found

68.0694.

-((6-((4-Bromophenyl)thio)-2-cyano-1-oxo-1H-phenalen-3-yl)-

amino)-N-(2-(2-hydroxyethoxy)ethyl)propanamide (S1-6g). Yellow

solid. H NMR (500 MHz, DMSO-d ): δ 8.61 (d, J = 8.4 Hz, 1H),

.57 (d, J = 8.4 Hz, 1H), 8.54 (d, J = 8.4 Hz, 1H), 7.89 (t, J = 8.4 Hz,

H), 7.72 (t, J = 8.4 Hz, 2H), 7.47 (d, J = 8.4 Hz, 1H), 7.30 (t, J = 8.4

Hz, 2H), 5.43 (s, 1H), 4.20 (t, J = 6.8 Hz, 2H), 3.71−3.65 (m, 4H),

.57 (t, J = 6.4 Hz, 2H), 3.26 (t, J = 6.4 Hz, 2H), 2.78 (t, J = 6.8 Hz,

H); ¹³C NMR (100 MHz, DMSO-d ): δ 178.0, 175.6, 175.1, 138.6,

34.7, 132.4, 132.1, 131.8, 130.6, 128.8, 128.2, 126.0, 125.1, 121.6,

15.8, 78.2, 70.1, 61.4, 41.3, 39.8, 37.1; [ToF-MS (ESI )]:

Sheng Zhou − State Key Laboratory of Fine Chemicals, Zhang

Dayu School of Chemistry, Dalian University of Technology,

Dalian, Liaoning 116024, China

Jian Dai − State Key Laboratory of Fine Chemicals, Zhang

Dayu School of Chemistry, Dalian University of Technology,

Dalian, Liaoning 116024, China

C H BrN O S [M + H ], calcd 556.0744, found 556.0748.

-((6-((4-Bromophenyl)thio)-2-cyano-1-oxo-1H-phenalen-3-yl)-

amino)-N-(4-(dimethylamino)benzyl)propanamide (S1-6h). Yellow

solid. H NMR (500 MHz, DMSO-d ): δ 8.60 (d, J = 8.4 Hz, 1H),

.57 (d, J = 8.4 Hz, 1H), 8.54 (d, J = 8.4 Hz, 1H), 7.89 (t, J = 8.4 Hz,

H), 7.72 (t, J = 8.0 Hz, 2H), 7.48 (d, J = 8.4 Hz, 1H), 7.28 (t, J = 8.0

Hz, 2H), 7.12 (t, J = 7.8 Hz, 2H), 6.85 (t, J = 7.8 Hz, 2H), 4.18 (s,

H), 4.20 (t, J = 6.8 Hz, 2H), 2.78 (t, J = 6.8 Hz, 2H), 2.89 (s, 6H);

Complete contact information is available at:

https://pubs.acs.org/10.1021/acs.jmedchem.0c01290

C NMR (100 MHz, DMSO-d ): δ 178.0, 175.7, 175.2, 173.8, 149.3,

38.6, 134.8, 132.2, 132.0, 131.9, 131.6, 130.7, 128.7, 128.2, 127.9,

27.4, 126.1, 125.1, 121.4, 115.7, 112.7, 78.4, 43.6, 41.5, 39.6, 37.0;

Author Contributions

Z.W. and T.S. contributed equally to this article.

⊥

[

ToF-MS (ESI )]: C H BrN O S [M + H ], calcd 611.1111, found

11.1108.

32 27 4 2

Notes

The authors declare no competing ﬁnancial interest.

ASSOCIATED CONTENT

sı Supporting Information

The Supporting Information is available free of charge at

https://pubs.acs.org/doi/10.1021/acs.jmedchem.0c01290.

■

ACKNOWLEDGMENTS

■

We thank Prof. Donghai Lin who provided valuable

suggestions to this work (Department of Chemical Biology,

Xiamen University, Xiamen, China). Our NMR work was

performed at the National Center for Protein Science Shanghai

List of (φ, ψ) backbone angle pairs for residues R106-

D111, diagram of the phosphorylation sites of Bcl-2, 3D

NMR titration spectra of diﬀerent Bcl-2 variants, crystal

structure of Bcl-2 isoform 2 in complex with ABT-263 or

ABT-199 analogue, plot of the RMSD values of the

backbone atoms in MD, binding aﬃnity curves of BH3

peptides or compounds with diﬀerent Bcl-2 variants,

binding aﬃnity curves of compounds with Bcl-xl and

Mcl-1, and LC−MS spectra of the compounds (PDF)

MD-simulated structure of pBcl-2 (PDB)

(

Shanghai, China). This research was supported by the

National Natural Science Foundation of China (81430083,

1903462, and 82073703), the China Postdoctoral Science

Foundation (2018M641694), and the Fundamental Research

Funds for the Central University (DUT20LK28 and

DUT20YG133). The authors will release the atomic

coordinates upon article publication.

Docking-stimulated structure of S1-6e/Bcl-2 (PDB )

Molecular formula strings (CSV )

ABBREVIATIONS USED

■

CSPs, chemical shift perturbations; NOEs, nuclear Overhauser

eﬀects; FPA, ﬂuorescent polarization assay; ITC, isothermal

titration calorimetry; MD, molecular dynamics; DIEA, N,N-di-

isopropylethylamine; HATU, hexaﬂuorophosphate azabenzo-

triazole tetramethyl uranium

AUTHOR INFORMATION

Corresponding Authors

Ting Song − State Key Laboratory of Fine Chemicals, Zhang

Dayu School of Chemistry, Dalian University of Technology,

Dalian, Liaoning 116024, China; Email: songting@

dlut.edu.cn

■

REFERENCES

■

1) Humphrey, S. J.; James, D. E.; Mann, M. Protein

phosphorylation: a major switch mechanism for metabolic regulation.

Trends Endocrinol. Metab. 2015, 26, 676−687.

Zhichao Zhang − State Key Laboratory of Fine Chemicals,

Zhang Dayu School of Chemistry, Dalian University of

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

3) Nishi, H.; Hashimoto, K.; Panchenko, A. R. Phosphorylation in

Article

(

2) Cohen, P. The origins of protein phosphorylation. Nat. Cell Biol.

002, 4, E127−E130.

protein-protein binding: effect on stability and function. Structure

011, 19, 1807−1815.

4) Liu, S. Q.; Cai, X.; Wu, J. X.; Cong, Q.; Chen, X.; Li, T.; Du, F.

Z.; McCubrey, J.; May, W. S.; Reed, J. C.; Andreeff, M. Mechanisms

of apoptosis sensitivity and resistance to the BH3 mimetic ABT-737

in acute myeloid leukemia. Cancer Cell 2006, 10, 375−388.

(25) Perez-Galan, P.; Roue, P. G.; Lopez-Guerra, M.; Nguyen, M.;

́ ́ ́ ́

(

Villamor, N.; Montserrat, E.; Shore, G. C.; Campo, E.; Colomer, D.

BCL-2 phosphorylation modulates sensitivity to the BH3 mimetic

GX15-070 (Obatoclax) and reduces its synergistic interaction with

bortezomib in chronic lymphocytic leukemia cells. Leukemia 2008, 22,

H.; Ren, J. Y.; Wu, Y. T.; Grishin, N. V.; Chen, Z. J. J.

Phosphorylation of innate immune adaptor proteins MAVS,

STING, and TRIF induces IRF3 activation. Science 2015, 347,

No. aaa2630.

712−1720.

(26) Petros, A. M.; Medek, A.; Nettesheim, D. G.; Kim, D. H.; Yoon,

H. S.; Swift, K.; Matayoshi, E. D.; Oltersdorf, T.; Fesik, S. W. Solution

structure of the antiapoptotic protein Bcl-2. Proc. Natl. Acad. Sci.

U.S.A. 2001, 98, 3012−3017.

5) Zimmermann, S.; Moelling, K. Phosphorylation and regulation

of Raf by Akt (Protein Kinase B). Science 1999, 286, 1741−1744.

6) Sarbassov, D. D.; Guertin, D. A.; Ali, S. M.; Sabatin, D. M.

Phosphorylation and regulation of Akt/PKB by the rictor-mTOR

complex. Science 2005 , 307 , 1098 −1101.

(

27) Chang, B. S.; Minn, A. J.; Muchmore, S. W.; Fesik, S. W.;

Thompson, C. B. Identification of a novel regulatory domain in Bcl-x _L

7) Reed, J. C. Bcl-2 family proteins. Oncogene 1998, 17, 3225−

and Bcl-2. EMBO J. 1997, 16, 968 −977.

9) Sharpe, J. C.; Arnoult, D.; Yould, R. J. Control of mitochondrial

236.

8) Cory, S.; Huang, D. C.; Adams, J. M. The Bcl-2 family: roles in

cell survival and oncogenesis. Oncogene 2003 , 22 , 8590 −8607.

28) Blagosklonny, M. V. Unwinding the loop of Bcl-2

phosphorylation. Leukemia 2001, 15, 869−874.

29) Yamamoto, K.; Ichijo, H.; Korsmeyer, S. J. BCL-2 is

phosphorylated and inactivated by an ASK1/Jun N-terminal protein

kinase pathway normally activated at G2/M. Mol. Cell. Biol. 1999, 19,

permeability by Bcl-2 family members. Biochim. Biophys. Acta, Mol.

Cell Res. 2004, 1644, 107−113.

30) Song, T.; Chai, G.; Liu, Y.; Yu, X.; Wang, Z.; Zhang, Z. Bcl-2

469−8478.

10) Petros, A. M.; Olejniczak, E. T.; Fesik, S. W. Structural biology

of the Bcl-2 family proteins. Biochim. Biophys. Acta, Mol. Cell Res.

004, 1644, 83−94.

11) Hinds, M. G.; Day, C. L. Regulation of apoptosis: Uncovering

the binding determinants. Curr. Opin. Struct. Biol. 2005 , 15 , 690 −699.

12) Van Delft, M. F.; Huang, D. C. How the Bcl-2 family of

proteins interact to regulate apoptosis. Cell. Res. 2006 , 16 , 203 −213.

13) Martinou, J.-C.; Youle, R. J. Mitochondria in apoptosis: Bcl-2

family members and mitochondrial dynamics. Dev. Cell 2011, 21, 92−

01.

14) Ku, B.; Liang, C.; Jung, J. U.; Oh, B. Evidence that inhibition of

phosphorylation confers resistance on chronic lymphocytic leukaemia

cells to the BH3 mimetics ABT-737, ABT-263 and ABT-199 by

impeding direct binding. Br. J. Pharmacol. 2016, 173, 471−483.

(31) Song, T.; Wang, P.; Yu, X.; Wang, A.; Chai, G.; Fan, Y.; Zhang,

Z. Systems analysis of phosphorylation-regulated Bcl-2 interactions

establishes a model to reconcile the controversy over the significance

of Bcl-2 phosphorylation. Br. J. Pharmacol. 2019 , 176 , 491 −504.

(32) Tonks, N. K. Protein tyrosine phosphatases-from housekeeping

33) Shi, Y. Serine/threonine phosphatases: mechanism through

enzymes to master regulators of signal transduction. FEBS J. 2013,

80, 346−378.

structure. Cell 2009 , 139 , 468 −484.

34) Roskoski, R. Cyclin-dependent protein serine/threonine kinase

inhibitors as anticancer drugs. Pharmacol. Res. 2019, 139, 471−488.

35) Song, T.; Cao, K.; Fan, Y.; Zhang, Z.; Guo, Z.; Zhang, M.; Liu,

BAX activation by BCL-2 involves its tight and preferential interaction

with the BH3 domain of BAX. Cell Res. 2011, 21, 627−641.

15) Moldoveanu, T.; Follis, A. V.; Kriwacki, R. W.; Green, D. R.

16) Kale, J.; Osterlund, E. J.; Andrews, D. W. BCL-2 family

Many players in BCL-2 family affairs. Trends Biochem. Sci. 2014, 39,

01−111.

P. Expression and solution NMR study of multi-site phosphomimetic

proteins: changing partners in the dance towards death. Cell Death

Differ. 2018, 25, 65−80.

mutant BCL-2 protein. Protein Pept. Lett. 2019, 26, 449−457.

(36) Song, T.; Yu, X.; Liu, Y.; Li, X.; Chai, G.; Zhang, Z. Discovery

17) Deng, X.; Ruvolo, P.; Carr, B.; May, W. S. Survival function of

of a small-molecule pBcl-2 inhibitor that overcomes pBcl-2-mediated

ERK1y2 as IL-3-activated, staurosporine-resistant Bcl2 kinases. Proc.

Natl. Acad. Sci. U.S.A. 2000, 97, 1578−1583.

resistance to apoptosis. ChemBioChem 2015, 16, 757−765.

(37) Lu, S.; Deng, R.; Jiang, H.; Song, H.; Li, S.; Shen, Q.; Huang,

W.; Nussinov, R.; Yu, J.; Zhang, J. The mechanism of ATP-dependent

18) Ruvolo, P. P.; Deng, X.; May, W. S. Phosphorylation of Bcl-2

allosteric protection of Akt kinase phosphorylation. Structure 2015,

and regulation of apoptosis. Leukemia 2001, 15 , 515 −522.

19) Deng, X.; Gao, F.; Flagg, T.; May, W. S. Mono- and multisite

(

3, 1725−1734.

phosphorylation enhances Bcl2 ’s antiapoptotic function and inhib-

ition of cell cycle entry functions. Proc. Natl. Acad. Sci. U.S.A. 2004,

38) Song, T.; Wang, Z.; Ji, F.; Feng, Y.; Fan, Y.; Chai, G.; Li, X.; Li,

Z.; Zhang, Z. Deactivation of Mcl-1 by dual-function small-molecule

inhibitors targeting the Bcl-2 homology 3 domain and facilitating Mcl-

1 ubiquitination. Angew. Chem., Int. Ed. 2016, 55, 14248−14254.

(

01, 153−158.

20) Bassik, M. C.; Scorrano, L.; Oakes, S. A.; Pozzan, T.;

Korsmeyer, S. J. Phosphorylation of BCL-2 regulates ER Ca

39) Zhang, Z.; Wu, G.; Xie, F.; Song, T.; Chang, X. 3-

phomeostasis and apoptosis. EMBO J. 2004, 23, 1207−1216.

Thiomorpholin-8-oxo-8H-acenaphtho[1,2-b]pyrrole-9-carbonitrile

21) Wang, L.; Chen, L.; Benincosa, J.; Fortney, J.; Gibson, L. F.

S1) based molecules as potent, dual inhibitors of B-cell lymphoma 2

Bcl-2) and myeloid cell leukemia sequence 1 (Mcl-1): structure-

VEGF-induced phosphorylation of Bcl-2 influences B lineage

leukemic cell response to apoptotic stimuli. Leukemia 2005, 19,

based design and structure-activity relationship studies. J. Med. Chem.

2011, 54, 1101−1105.

(

44−353.

22) Kurinna, S.; Konopleva, M.; Palla, S. L.; Chen, W.; Kornblau,

(40) Chan, T. O.; Zhang, J.; Rodeck, U.; Pascal, J. M.; Armen, R. S.;

S.; Contractor, R.; Deng, X.; May, W. S.; Andreeff, M.; Ruvolo, P. P.

Bcl2 phosphorylation and active PKC a are associated with poor

survival in AML. Leukemia 2006, 20, 1316−1319.

Spring, M.; Dumitru, C. D.; Myers, V.; Lia, X.; Cheung, J. Y.;

Feldman, M. A. Resistance of Akt kinases to dephosphorylation

through ATP-dependent conformational plasticity. Proc. Natl. Acad.

Sci. U.S.A. 2011, 108, E1120−E1127.

23) Blagosklonny, M. V.; Alvarez, M.; Fojo, A.; Neckers, L. M. Bcl-

protein downregulation is not required for differentiation of

multidrug resistant HL60 leukemia cells. Leuk. Res. 1996, 20, 101−

07.

24) Konopleva, M.; Contractor, R.; Tsao, T.; Samudio, I.; Ruvolo,

P. P.; Kitada, S.; Deng, X.; Zhai, D.; Shi, Y.; Sneed, T.; Verhaegen, M.;

Soengas, M.; Ruvolo, V. R.; McQueen, T.; Schober, W. D.; Watt, J.

C.; Jiffar, T.; Ling, X.; Marini, F. C.; Harris, D.; Dietrich, M.; Estrov,

(41) Follis, A. V.; Llambi, F.; Kalkavan, H.; Yao, Y.; Phillips, A.;

Park, C.; Marassi, F. M.; Green, D. R.; Kriwacki, R. W. Regulation of

apoptosis by an intrinsically disordered region of Bcl-xL. Nat. Chem.

Biol. 2018, 14, 458−465.

(42) Chandrashekarappa, D. G.; McCartney, R. R.; Schmidt, M. C.

Ligand binding to the AMP-activated protein kinase active site

(

J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

mediates protection of the activation loop from dephosphorylation. J.

Article

Biol. Chem. 2013, 288, 89−98.

43) Formaneck, M. S.; Ma, L.; Cui, Q. Reconciling the “old ” and

“

new ” views of protein allostery: a molecular simulation study of

chemotaxis Y protein (CheY). Proteins 2006, 63, 846−867.

44) Petros, A. M.; Medek, A.; Nettesheim, D. G.; Kim, D. H.; Yoon,

(

H. S.; Swift, K.; Matayoshi, E. D.; Oltersdorf, T.; Fesik, S. W. Solution

structure of the antiapoptotic protein Bcl-2. Proc. Natl. Acad. Sci.

U.S.A. 2001, 98, 3012−3017.