Aberystwyth University Confocal Microscopy Article Discussion

Can you please analyze figure 2A from the attached article?

And write me the goal of the article  and what interest you from the aticle and what concerns or questions you have about the article?

Articles
Cite This: ACS Chem. Biol. 2019, 14, 636−643
pubs.acs.org/acschemicalbiology
Reverse Chemical Proteomics Identifies an Unanticipated Human
Target of the Antimalarial Artesunate
Michael P. Gotsbacher,†,⊥,§ Sung Min Cho,‡,⊥ Nam Hee Kim,‡ Fei Liu,† Ho Jeong Kwon,*,‡
and Peter Karuso*,†
†
Department of Molecular Sciences, Macquarie University, Sydney, NSW 2109, Australia
Chemical Genomics Global Research Laboratory, Department of Biotechnology, College of Life Science & Biotechnology, Yonsei
University, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-749, South Korea
Downloaded via DEPAUL UNIV on February 10, 2020 at 02:33:44 (UTC).
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
‡
S Supporting Information
*
ABSTRACT: Artemisinins are the most potent and safe
antimalarials available. Despite their clinical potential, no
human target for the artemisinins is known. The unbiased
interrogation of several human cDNA libraries, displayed on
bacteriophage T7, revealed a single human target of
artesunate; the intrinsically disordered Bcl-2 antagonist of
cell death promoter (BAD). We show that artesunate inhibits
the phosphorylation of BAD, thereby promoting the
formation of the proapoptotic BAD/Bcl-xL complex and the
subsequent intrinsic apoptotic cascade involving cytochrome c
release, PARP cleavage, caspase activation, and ultimately cell
death. This unanticipated role of BAD as a possible drug
target of artesunate points to direct clinical exploitation of artemisinins in the Bcl-xL life/death switch and that artesunate’s
anticancer activity is, at least in part, independent of reactive oxygen species.
T
of artemisinin is activated and cleaved through a hemedependent, radical mechanism that alkylates parasite proteins.18−20 This has recently been supported through chemical
proteomics investigations in which at least 124 parasite
proteins were shown to be alkylated by an artesunate analogue
in infected red blood cells.21 Notwithstanding this, proposed
specific targets include SERCA,22 membrane glutathione Stransferase (PfEXP1),23 or phosphatidylinositol-3-kinase
PfPI3K in P. falciparum.24
Interestingly, artemisinins display polypharmacology with
profound and selective anticancer activity25,26 and, it has been
suggested, promote apoptosis via mitochondrial pathways in
cancer cell lines.27−29 It is known that many genes are
upregulated upon 1 treatment in cancer cell lines. These
include BUB3, cyclins, CDC25A (proliferation), VEGF, MMP9, angiostatin, thrombospondin-1 (angiogenesis), and Bcl-2,
BAX, and NF-κB (apoptosis).27,30 Despite their clinical
potential for treating cancer and the extensive primary
literature on their anticancer activity, 1’s human target remains
elusive.31−34
he study of weak but selective protein−ligand interactions has always been a challenge and is exemplified by
gamma-hydroxybutyric acid, which is a 3 mM agonist of
GABA-B receptors but highly selective.1 Low affinity effectively
hinders the identification of small molecule protein targets
using genome-wide, forward proteomic methods such as
affinity capture MS or affinity chromatography.2 Even high
affinity interactions can be problematic to identify in the
presence of large amounts of nonspecific interactions. An
attractive alternative approach, particularly suited for weak
interactions with low abundance proteins, is reverse chemical
proteomics using phage display where there is a physical link
between the genome and proteome.3−13 The rapid life cycle of
bacteriophages allows iterative purification of rare and/or low
abundance proteins from highly complex mixtures based on
relatively weak affinity to a tagged small molecule. For
example, we were able to rapidly isolate a human protein
target for kahalalide F that showed ∼50 μM KD with ribosomal
protein S25.9 Here, we report the identification of the Bcl-2
agonist of cell death (BAD) promotor as the first human target
of the natural product drug artesunate (1) using T7 phage
display.
Artemisinins are sesquiterpenes from the sweet wormwood
(Artemisia annua).14,15 They are the most potent antimalarials
available with 500 million doses/annum prescribed to
eradicate Plasmodium falciparum infections,16 but their mode
of action is still under debate.17 It is, however, widely accepted
that inside the infected red blood cell, the endoperoxide bridge
© 2019 American Chemical Society
■
RESULTS AND DISCUSSION
We began with the synthesis of a biotinylated analogue (3) of
1 from artemisinin (2) containing a long, hydrophilic linker
Received: November 13, 2018
Accepted: March 6, 2019
Published: March 6, 2019
636
DOI: 10.1021/acschembio.8b01004
ACS Chem. Biol. 2019, 14, 636−643
Articles
ACS Chemical Biology
not converge (Figure 1A; Figure S4). For all four tumor
libraries, the phage titer increased exponentially after round 8
Scheme 1. Synthesis of Probes and NMR Correlations for
Probe 3a
Figure 1. BAD is the common protein binding partner of 1, identified
by biopanning of phage displayed cDNA libraries from various cancer
cells. (A) Agarose gel electrophoresis of phage DNA inserts, amplified
by PCR from phage sublibraries after biopanning against 5
immobilized on neutravidin-coated microtiter plates. See Figures S4
and S5 for more detail. (B) Phage titer after each round of biopanning
showing convergence after rounds 8−10 for the 4 tumor cDNA
libraries and (C) on-phage binding study showing 100-fold higher
titer of the BAD-displaying phage clone for the 1-immobilized support
than the support with immobilized negative control (4). The wildtype phage (no insert) does not differentiate the two supports.
(e.g., Figure 1C). Random plaques were picked after the last
round and fingerprinted by Hinf1 digestion (Figure S5). From
sequencing, in all converged tumor libraries, the dominant
clone (Table S4) was the Bcl-2 antagonist of cell death
promoter (BAD). Comparison of the DNA sequences showed
that for each library the phage selection produced BAD clones
of different lengths, but all were in-frame with the phage coat
protein and covered the full coding sequence of BAD (Table
1). No other gene was represented by in-frame clones that
appeared in more than one tumor library.
Several attempts to clone and overexpress BAD as hexaHis
or GST fusions failed due to the intrinsic disorder of the BAD
protein.36 Attempted purification led to extensive degradation,
so it was not possible to obtain a pure sample for reliable KD
measurements. Consequently, an on-phage binding assay
(Figure 1C) was used and showed that the control phage
(no insert) produced a background of ∼107 phage particles
upon elution with SDS in wells derivatized with 3 or 4. In
contrast, BAD-phage (with BAD insert) produced 100× more
phage particles upon elution from 3-derivatized wells
compared to 4, suggesting a specific interaction between
BAD and 1. As an alternative to quantitative KD measurements,
we constructed a peptide library of BAD peptides (31 × 20mers shifted by 5 aa each time; Table S5) arrayed onto a glass
slide. Staining with 6, with butylamine-NBD as a negative
control (Figure S13), indicated preferential binding of 6 to a
section around S136, just before the BH3 domain, and to the
C-terminal of the protein (Figure S14).
BAD, as a BH3-only, proapoptotic protein, was originally
identified as a partner for Bcl-2 and Bcl-xL in a yeast twohybrid screen and shown to displace BAX to induce
cytochrome c release and caspase-dependent apoptosis.37 To
a
Red = HMBC, green = ROESY, black = COSY, showing key
correlations only. Reagents and conditions: (a) 2, NaBH4 (3.5
equiv)/MeOH, 0−5 °C, 2 h, 81% yield; (b) 5, succinic anhydride
(1.6 equiv), imidazole (0.9 equiv)/DCM, rt 2 h, 91% yield; (c) biotinPEG-NH2, 1 (2 equiv), EDC (4 equiv), HOBt (4 equiv)/DMF, 0 °C
over 30 min, then rt 18 h, then excess water, 39% yield; (d) biotinPEG-NH2, valeric acid (2 equiv), EDC (3 equiv), HOBt (3 equiv)/
DMF, 0 °C over 30 min, then rt 18 h, then excess water, 28% yield;
(e) NBD-Cl (1.2 equiv)/ACN, β-alanine (1.2 equiv), and NaHCO3
(3 equiv)/water, 55 °C, 1 h; then remove ACN, pH to 2 (1 N HCl);
solvent removal, then 5 (1 equiv)/DCM, DMAP (1.2 equiv) under
N2; then EDC (1.2 equiv), rt 18 h, 42% yield over two steps.
(biotin-PEG-NH2; Scheme 1). Derivatization at C12 is known
to not affect the anticancer activity of artemisinins.35 The
structure, purity, stereochemistry, and stability were verified by
HPLC, NMR spectroscopy, and HRMS (Scheme 1; Figures S1
and S2). Likewise, a biotinylated “blank probe” reagent (4)
was synthesized from valeric acid, along with 6, a fluorescent
analogue of 1 for imaging (F-ART, Scheme 1; Figure S3).
Compounds 3 and 4 were immobilized in separate neutravidin
coated microtiter strip wells. Five human cDNA libraries,
expressed in bacteriophage T7, were panned against 3-coated
wells, after preclearing in 4-coated wells to remove nonspecific
binders. After 9 rounds of biopanning, all four tumor libraries
produced dominant clones but the normal colon library did
637
DOI: 10.1021/acschembio.8b01004
ACS Chem. Biol. 2019, 14, 636−643
Articles
ACS Chemical Biology
Table 1. Sequence Alignment of BAD Clones from Colon Tumor (CoT), Liver Tumor (LiT), Lung Tumor (LuT) and Breast
Tumor (BrT) cDNA Phage Libraries and Consensus with Human BADa
a
The BH3 domain of BAD is highlighted in blue and the phosphorylated Ser residues in green. Each library converged on full-length BAD with
many different clones varying in N- and C-terminal sequences, but all were in-frame with the coat protein. The bolded/underlined residues are the
consensus binding domains from the peptide dot blot experiments (see Table S5, Figures S13 and S14). bThe mouse numbering system is widely
used in the literature and is 36 amino acids longer than human BAD. Thus, S99 above is S136 in the mouse sequence, and to avoid confusion, the
mouse numbering is used throughout.
mL−1) and 1 (0 to 400 μM) concentrations, following the
literature recommendations to apply 10× higher ligand
concentrations as the upper limit compared to the effective
concentration.38 For actin (Figure 3D) and BAX (Figure 3C),
hydrolysis proceeded in a 1-independent manner. In contrast,
BAD was more readily proteolyzed in a clear dose-dependent
manner (Figure 3B; p = 0.0063), suggesting that 1 selectively
binds to BAD but not the related BAX protein or the actin
loading control.
The proapoptotic action of 1 was abrogated when BAD
expression was knocked down using a mixture of four siRNAs
against BAD (siBAD) (Figure 4). In the presence of 20 nM
siBAD, BAD expression in HeLa cells was reduced by ∼60%
compared to random siRNA (Figures 4A and 4B). Compound
1’s apoptotic effect increased in a time- and dose-dependent
manner (Figure 4C) that was only slightly affected by the
introduction of a mixture of random siRNA (Figure 4D). In
cells treated with siBAD with 40 μM 1, the suppression of 1’s
apoptotic effect was significant (87% cells surviving with BAD
knock-down vs 65% without; Figure 4E). The loss of 1’s
apoptotic effect with 20 nM siBAD was most prominent after
72 h of growth (71% surviving with BAD knock-down vs 20%
without; Figure 4E). The same effect was observed for all
concentrations of 1 tested again in a dose-dependent manner
(p = 0.055). In contrast, the apoptotic effect of camptothecin
was not responsive to changes of BAD expression (Figure
S11C) because camptothecin induces apoptosis by targeting
topoisomerase I.39
validate the dependence of 1’s apoptotic activity on BAD, cell
proliferation assays were performed first using HeLa (human
cervical cancer) cells (Figure S6), providing LD50 values of 63,
38, and 12 μM (R2 = 0.94−0.98) for 24, 48, and 72 h 1
treatments, respectively, and established upper limits on the
concentrations of 1 that can be used in live cells. The apoptotic
effect of 1 was also confirmed in HeLa cells by observation of a
dose-dependent release of cytochrome c into the cytosol,
PARP cleavage, and caspase activation (Figures S7−S9).
Under the microscope, cells treated with ≥10 μM 1 showed
typical signs of apoptosis (Figure S12).
The binding of ART to BAD was further investigated in
HeLa cells through confocal microscopy. To investigate the
binding of ART to BAD in living cells, F-ART (6) was
synthesized and used in a competitive binding assay. Cells that
were pretreated with unlabeled ART (1) and then stained with
6 (Figure 2A, panel 2), showing a 60% reduction in
fluorescence, demonstrating that 6 and 1 share the same
binding site in the cytosol of HeLa cells (Figure 2A). BAD and
6 were also found to partially colocalize in the cytosol of HeLa
cells (Figure 2B), and there is a positive correlation (r2 > 0.75)
between the localization of 6 and BAD-Ab, suggesting that 6
and 1 bind to BAD in living cells. Similar results were obtained
for HEK293 cells (Figure S12). A DARTS assay with HEK293
cell lysates (Figure 3), in which the binding of a ligand to a
protein is detected by either the enhanced or reduced rate of
proteolysis,38 suggested the binding of 1 to BAD. The
proteolysis of BAD, BAX, and actin in the cell lysate was
monitored in the presence of increasing Pronase (0 to 10 μg
638
DOI: 10.1021/acschembio.8b01004
ACS Chem. Biol. 2019, 14, 636−643
Articles
ACS Chemical Biology
Figure 4. Validation of the interaction between 1 and BAD using
siRNA. (A) BAD can be knocked down in live HeLa cells with 20 nM
mixture of 4 siRNAs for BAD (lane 3) relative to random mixed
siRNA (lane 2). (B) Graphical representation of (A) carried out in
triplicate. (C) Percentage (relative to control) cell survival of HeLa
cells treated with variable 1 for 2472 h. (D) Same as B except HeLa
cells were also treated with a 20 nM mixture of random siRNA. (E)
The same as D except cells were treated with 20 nM siRNA against
BAD. ***Designates p < 0.001 (Student’s t-test). various derivatives) has been reported, but no target was identified.42 More recently, Button et al. argued that necroptosis is 1’s main mode of action on Schwannoma cells.43 Even more recently, Hamacher suggests “ferroptosis” in pancreatic cancer cells and that 1 did not induce apoptosis or necroptosis.44 A recent chemical proteomics study in this journal suggested that due to heme activation, 1 covalently modified many proteins (in HeLa cell lysates treated with 10 μM hemin) to achieve its anticancer activity in parallel to its antimalarial activity.45 However, this is not consistent with other studies that discount oxidative damage as necessary for cytotoxicity of 1 and demonstrate 1-induced apoptosis in a ROS-independent, Bax-mediated manner.46 The use of a cell lysate with a high concentration of added hemin is also not physiologically relevant. The endoperoxide of ART is known to react with free hemin.47 Our identification of BAD as a target of 1 opens new questions in ROS-independent apoptosis and potential avenues for targeted therapies. Given that targeting Bcl-2/Mcl-1 family members is an emerging strategy in development of anticancer therapeutics48 and these life/ death switches have been suggested as the Achilles’ heel of many tumors,49 a BAD-targeted mechanism of action for 1 in a specific apoptotic pathway could lead to synergy with other anticancer therapies.50 BAD can reverse the pro-survival activity of Bcl-xL but not that of Bcl-2. Also, in rat ovaries, mutation of BADS136A results in the reported binding of BAD to Mcl-1, whereas wild-type phosphorylated BAD binds exclusively to 14-3-3.51 Regulation of BAD is achieved by cytokine and growth factor signaling and likely influences numerous aspects of metabolism, autophagy, and apoptosis.48,52 In a breast cancer model where PTEN is mutated, BAD is constitutively phosphorylated and sequestered by 14-3-3, completely inhibiting its proapoptotic activity.53 In AML, it was found that BAD was phosphorylated at S112 and S136 in 41/42 clinical samples tested.54 BAD’s sensitizer activity toward apoptosis is negatively regulated by kinases (JNK, PKA, and AKT) through phosphorylation at S112, S136, and S155.55−57 It has been suggested the phosphorylation within the BH3 domain (S155; Table 1) renders the binding of BAD to the hydrophobic BH3 Figure 2. Validation of the interaction between 1 and BAD using fluorescently labeled 1 (F-ART, 6). (A) Competition between 1 (ART) and 6 (F-ART). Panel 1: Treatment of HeLa cells with Hoechst (blue) and 6 (green). Panel 2: Treatment of HeLa cells with Hoechst (blue) and 6 (green) in the presence of 25 μM 1. The intensity of 6 was measured using Image J and displayed as a bar graph. (B) Colocalization of 6 (green) and BAD-Ab (red). Panels 1 and 2: Treatment of HeLa cells with Hoechst (blue) and 10 μM 6 (green) and BAD-Ab (red). Panel 3: Additional cells. The colocalization of F-ART and BAD was measured using Image J and expressed as a Pearson correlation curve. Figure 3. Validation of the interaction between 1 and BAD using the DARTS assay. (A) Western blotting of BAD, BAX, and actin (loading control) with variable 1 and Pronase. (B−D) Graphical representation of (B) for BAD (p = 0.00152; one-way ANOVA), BAX (p = 0.8968), and actin (p = 0.6592), respectively, run in triplicate. **Designates p < 0.01 (Student’s t-test). The identification of BAD as a target of 1’s apoptotic effect came as a surprise. Many mechanisms for the anticancer activity of artemisinins have been proposed with reactive oxygen species (ROS) generation or mitochondrial induced apoptosis, involving Bcl-2 family genes, particularly prominent in the literature.33,40,41 A cell line-dependent effect of 1 (or its 639 DOI: 10.1021/acschembio.8b01004 ACS Chem. Biol. 2019, 14, 636−643 Articles ACS Chemical Biology (Figure 6A), even though BAD and ABT-737 both putatively bind Bcl-xL to achieve their apoptotic effect. As the apparent cytotoxicity of 1 is abrogated if BAD is knocked down in HeLa cells (Figure 4) and the level of pBAD decreases in a dose-dependent way with siBAD (Figure 5), this suggests that binding to BAD is required for 1’s apoptotic effect and that ROS are not necessarily involved in 1’s anticancer activity. Because phosphorylation is a major regulatory mechanism by which cancer cells inhibit BAD function to suppress apoptosis, an elegant way to restore BAD activity, and thus the sensitivity of cells to apoptotic signaling without affecting the critical roles of kinases through kinasebased drugs, would be to inhibit only the phosphorylation of BAD. From our results, it is possible that this is indeed how 1 achieves its proapoptotic activity in HeLa cells. Theoretically, 1’s binding to BAD can either stabilize the Bcl-xL−BAD interaction, reduce the sequestration of pBAD by 14-3-3 proteins (leaving free pBAD susceptible to phosphatases), or inhibit the phosphorylation of BAD, which favors the formation of the proapoptotic Bcl-xL−BAD complex. While our observations (Figure 1C, Figure S14) support the last mode of action, the first two possibilities remain to be refuted. The combined effects of 1 and ABT-737, a BH3 mimic that binds to Bcl-2, Bcl-xL, and Bcl-w but not Mcl-1,59 surprisingly revealed a highly synergistic action (Figure 6). HeLa cells are resistant to ABT-737 (LD50 > 200 μM) because of high levels
of Mcl-1. Our results suggest that the dephosphorylation (or
inhibition of phosphorylation) of BAD by 1 results in a marked
increase in sensitivity of HeLa cells to ABT-737 (LD50 16 μM)
in the presence of 10 μM 1 and 1.2 μM with 20 μM 1. This
synergy suggests that 1 may do more than just promote the
binding of BAD to Bcl-xL, and we anticipate that this target
identification will enable the future exploration of many of
these interesting possibilities in the clinic. Intrinsically
disordered proteins (such as BAD) have the potential to
bind to multiple partners depending on their conformation and
post-translational modifications. Future clinical investigations
of how 1 may sensitize tumors to other genotoxic agents will
shed more light on the tumor selectivity of 1 and inform
clinical repositioning of this fascinating natural product.
Indeed, it is known that 1 synergistically induces apoptosis
of HeLa cells after IR but not SiHa cells 60 and
dihydroartemisinin (5) synergistically induces apoptosis in
OVCAR-3 and A2780 (but not IOSE144) ovarian cancer cells
treated with carboplatin.61 Similarly, A2780, HO8910, and
HEY ovarian cancer cells responded to 1 in a dose-dependent
manner and are synergistic with carboplatin, but SKOV3 cells
were totally unresponsive to 1.62 1 has also been reported to
sensitize breast cancer cells to the chemotherapeutic agent
epirubicin.63
Follow-up studies will include further characterization of the
interaction between ART and BAD using other biophysical
techniques and phosphoproteomics. The synergy between
ART and AZD-59912 and S63845 (inhibitors of Mcl-1) and
the interaction of ART with other BH3-only proteins will help
to further characterize the cellular mechanism of ART that
would not be obvious without first identifying BAD as a target
of ART.
In summary, phage display is an underutilized but powerful
technique for the genome wide, unbiased, reverse chemical
proteomics identification of potential protein targets of small
molecules. We have identified the Bcl-2 associate death
promoter (BAD) as a possible human target of artesunate
binding domain of Bcl-xL unfavorable, resulting in inhibition of
BAD’s proapoptotic function.58 The level of S136-pBAD was
therefore measured in HeLa cells 24 h after treating with 1 (0−
40 μM) or camptothecin (positive control; 1−10 μM) (Figure
5A). A dose-dependent decrease in phosphorylation on S136
Figure 5. 1 reduces the level of S136 phosphorylation on BAD and
Bcl-xL expression levels in HeLa cells. (A) Dose-dependent reduction
in S136 phosphorylation on BAD in HeLa cells on treatment with 1.
(B) Graphical representation of (A) carried out in triplicate. (C)
Dose-dependent reduction of Bcl-xL expression levels upon 1
treatment. (D) Graphical representation of (C) carried out in
triplicate. *Designates p < 0.05, **designates p < 0.01, ***designates p < 0.001 (Student’s t-test). was observed (Figure 5B) with 1, consistent with the peptide array data (Figures S13 and S14). There was also a dosedependent decrease in the expression levels of Bcl-xL (Figures 5C and 5D). At 20 μM, BAD phosphorylation was reduced by 60% and Bcl-xL by 20%. The effect of fixed doses of 1 on HeLa cells in the presence of 0, 0.01, 0.1, 1, or 10 μM ABT-737 or camptothecin was subject to isobolographic analysis (Figure 6) to determine if 1 Figure 6. 1 is synergistic with ABT-737 and camptothecin. (A) Isobole analysis (from Figures S15A−C) of the interaction between 1 and ABT-737 shows a strong synergistic effect at 10 μM (square) and 20 μM (diamond) 1. (B) Isobole analysis (from Figures S15d−f) of the interaction between 1 and camptothecin also shows a synergistic effect at 10 μM (square) and 20 μM (diamond). would work synergistically or additively with ABT-737, a clinically useful, known BH3-mimic that binds Bcl-xL.59 HeLa cells were grown for 24 or 48 h after treatment and the LD50 determined in the presence or absence of 1 (Figure S15). As expected, 1 and camptothecin are synergistic (Figure 6B) as they target orthogonal pathways, but surprisingly, 1 and ABT737 camptothecin were also found to be highly synergistic 640 DOI: 10.1021/acschembio.8b01004 ACS Chem. Biol. 2019, 14, 636−643 Articles ACS Chemical Biology (AE-MS) Rather than Affinity Purification Mass Spectrometry (APMS). Mol. Cell. Proteomics 14, 120. (3) McKenzie, K. M., Videlock, E. J., Splittgerber, U., and Austin, D. J. (2004) Simultaneous Identification of Multiple Protein Targets by Using Complementary-DNA Phage Display and a Natural-ProductMimetic Probe. Angew. Chem., Int. Ed. 43, 4052−4055. (4) Boehmerle, W., Splittgerber, U., Lazarus, M. B., McKenzie, K. M., Johnston, D. G., Austin, D. J., and Ehrlich, B. E. (2006) Paclitaxel induces calcium oscillations via an inositol 1,4,5-trisphosphate receptor and neuronal calcium sensor 1-dependent mechanism. Proc. Natl. Acad. Sci. U. S. A. 103, 18356−18361. (5) Izaguirre-Carbonell, J., Kawakubo, H., Murata, H., Tanabe, A., Takeuchi, T., Kusayanagi, T., Tsukuda, S., Hirakawa, T., Iwabata, K., Kanai, Y., Ohta, K., Miura, M., Sakaguchi, K., Matsunaga, S., Sahara, H., Kamisuki, S., and Sugawara, F. (2015) Novel anticancer agent, SQAP, binds to focal adhesion kinase and modulates its activity. Sci. Rep. 5, 15136. (6) Jung, H. J., Shim, J. S., Lee, J., Song, Y. M., Park, K. C., Choi, S. H., Kim, N. D., Yoon, J. H., Mungai, P. T., Schumacker, P. T., and Kwon, H. J. (2010) Terpestacin Inhibits Tumor Angiogenesis by Targeting UQCRB of Mitochondrial Complex III and Suppressing Hypoxia-induced Reactive Oxygen Species Production and Cellular Oxygen Sensing. J. Biol. Chem. 285, 11584−11595. (7) Kim, H., Deng, L., Xiong, X., Hunter, W. D., Long, M. C., and Pirrung, M. C. (2007) Glyceraldehyde 3-Phosphate Dehydrogenase Is a Cellular Target of the Insulin Mimic Demethylasterriquinone B1. J. Med. Chem. 50, 3423−3426. (8) Kim, N. H., Pham, N. B., Quinn, R. J., Shim, J. S., Cho, H., Cho, S. M., Park, S. W., Kim, J. H., Seok, S. H., Oh, J.-W., and Kwon, H. J. (2015) The small molecule R-(−)-β-O-methylsynephrine binds to nucleoporin 153 kDa and inhibits angiogenesis. Int. J. Biol. Sci. 11, 1088−1099. (9) Piggott, A. M., and Karuso, P. (2008) Rapid Identification of a Protein Binding Partner for the Marine Natural Product Kahalalide F by Using Reverse Chemical Proteomics. ChemBioChem 9, 524−530. (10) Piggott, A. M., Kriegel, A. M., Willows, R. D., and Karuso, P. (2009) Rapid isolation of novel FK506 binding proteins from multiple organisms using gDNA and cDNA T7 phage display. Bioorg. Med. Chem. 17, 6841−6850. (11) Shim, J. S., Lee, J., Park, H.-J., Park, S.-J., and Kwon, H. J. (2004) A New Curcumin Derivative, HBC, Interferes with the Cell Cycle Progression of Colon Cancer Cells via Antagonization of the Ca2+/Calmodulin Function. Chem. Biol. 11, 1455−1463. (12) Woolard, J., Vousden, W., Moss, S. J., Krishnakumar, A., Gammons, M. V. R., Nowak, D. G., Dixon, N., Micklefield, J., Spannhoff, A., Bedford, M. T., Gregory, M. A., Martin, C. J., Leadlay, P. F., Zhang, M. Q., Harper, S. J., Bates, D. O., and Wilkinson, B. (2011) Borrelidin modulates the alternative splicing of VEGF in favor of anti-angiogenic isoforms. Chem. Sci. 2, 273−278. (13) Gotsbacher, M. P., Cho, S., Kwon, H. J., and Karuso, P. (2016) Daptomycin, a last-resort antibiotic, binds ribosomal protein S19 in humans. Proteome Sci. 15, 16. (14) Tu, Y. (2016) ArtemisininA Gift from Traditional Chinese Medicine to the World (Nobel Lecture). Angew. Chem., Int. Ed. 55, 10210. (15) Klayman, D. L. (1985) Qinghaosu (artemisinin): an antimalarial drug from China. Science 228, 1049−1055. (16) White, N. J. (2008) Qinghaosu (Artemisinin): The Price of Success. Science 320, 330−334. (17) O’Neill, P. M., Barton, V. E., and Ward, S. A. (2010) The molecular mechanism of action of artemisinin - the debate continues. Molecules 15, 1705−1721. (18) Stocks, P. A., Bray, P. G., Barton, V. E., Al-Helal, M., Jones, M., Araujo, N. C., Gibbons, P., Ward, S. A., Hughes, R. H., Biagini, G. A., Davies, J., Amewu, R., Mercer, A. E., Ellis, G., and O’Neill, P. M. (2007) Evidence for a Common Non-Heme Chelatable-IronDependent Activation Mechanism for Semisynthetic and Synthetic Endoperoxide Antimalarial Drugs. Angew. Chem., Int. Ed. 46, 6278− 6283. (1) in several human cancer proteomes displayed on bacteriophage T7 and shown that a possible mode of action is to interfere with BAD phosphorylation at S136. This BADtargeting activity of 1 is highly synergistic with the BH3 mimetic ABT-737 that binds Bcl-xL. Biophysical characterization of the interaction between ART and BAD was made difficult because BAD is an intrinsically disordered protein. Identifying and quantifying such weak and dynamic interactions is problematic, and biophysical evidence is usually more ambiguous than for globular proteins. However, these atypical and often weak interactions, refractory to in vivo characterization, can be very meaningful in higher-order signaling assemblies and regulation that is only now being appreciated.64 However, overall, our data suggest a targeted mechanism of action for 1 for immediate clinical exploitation.50 ■ ASSOCIATED CONTENT S Supporting Information * The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acschembio.8b01004. ■ Methods section, including detailed synthetic procedures and probe stability, detailed target identification and validation procedures, and synergistic activity data (PDF) AUTHOR INFORMATION Corresponding Authors *E-mail: peter.karuso@mq.edu.au. *E-mail: kwonhj@yonsei.ac.kr. ORCID Michael P. Gotsbacher: 0000-0002-7153-1250 Sung Min Cho: 0000-0003-1889-6370 Ho Jeong Kwon: 0000-0002-6919-833X Peter Karuso: 0000-0002-0217-6021 Present Address § M.P.G.: School of Medical Sciences (Pharmacology), The University of Sydney, Sydney, NSW 2006, Australia. Author Contributions ⊥ M.P.G. and S.M.C. contributed equally. Funding This work was supported by ARC grant DP130103281 to P.K. and H.J.K. and NRF grants 2015K1A1A2028365, 2015M3A9C4076321, and BK21plus to H.J.K. Notes The authors declare no competing financial interest. ■ ABBREVIATIONS BAD, Bcl-2 antagonist of cell death; pBAD, S136 phosphorylated BAD; siBAD, a mixture of four specific small interfering RNAs against BAD; ART, artesunate; F-ART, fluorescent ART ■ REFERENCES (1) Mathivet, P., Bernasconi, R., Barry, J. D., Marescaux, C., and Bittiger, H. (1997) Binding characteristics of γ-hydroxybutyric acid as a weak but selective GABAB receptor agonist. Eur. J. Pharmacol. 321, 67−75. (2) Keilhauer, E. C., Hein, M. Y., and Mann, M. (2015) Accurate Protein Complex Retrieval by Affinity Enrichment Mass Spectrometry 641 DOI: 10.1021/acschembio.8b01004 ACS Chem. Biol. 2019, 14, 636−643 Articles ACS Chemical Biology conformational change upon binding to prosurvival Bcl-2 targets. Cell Death Differ. 14, 128−136. (37) Yang, E., Zha, J., Jockel, J., Boise, L. H., Thompson, C. B., and Korsmeyer, S. J. (1995) Bad, a heterodimeric partner for Bcl-xL and Bcl-2, displaces bax and promotes cell death. Cell 80, 285−291. (38) Lomenick, B., Hao, R., Jonai, N., Chin, R. M., Aghajan, M., Warburton, S., Wang, J., Wu, R. P., Gomez, F., Loo, J. A., Wohlschlegel, J. A., Vondriska, T. M., Pelletier, J., Herschman, H. R., Clardy, J., Clarke, C. F., and Huang, J. (2009) Target identification using drug affinity responsive target stability (DARTS). Proc. Natl. Acad. Sci. U. S. A. 106, 21984−21989. (39) Pommier, Y. (2006) Topoisomerase I inhibitors: camptothecins and beyond. Nat. Rev. Cancer 6, 789−802. (40) Efferth, T. (2014) Activation of Mitochondria-Driven Pathways by Artemisinin and Its Derivatives. In Mitochondria: The Anti-cancer Target for the Third Millennium (Neuzil, J., Pervaiz, S., and Fulda, S., Eds.), pp 135−150, Springer, Netherlands. (41) Tran, K. Q., Tin, A. S., and Firestone, G. L. (2014) Artemisinin triggers a G1 cell cycle arrest of human Ishikawa endometrial cancer cells and inhibits cyclin-dependent kinase-4 promoter activity and expression by disrupting nuclear factor-κB transcriptional signaling. Anti-Cancer Drugs 25, 270−281. (42) Efferth, T., Sauerbrey, A., Olbrich, A., Gebhart, E., Rauch, P., Weber, H. O., Hengstler, J. G., Halatsch, M.-E., Volm, M., Tew, K. D., Ross, D. D., and Funk, J. O. (2003) Molecular modes of action of artesunate in tumor cell lines. Mol. Pharmacol. 64, 382−394. (43) Button, R. W., Lin, F., Ercolano, E., Vincent, J. H., Hu, B., Hanemann, C. O., and Luo, S. (2014) Artesunate induces necrotic cell death in schwannoma cells. Cell Death Dis. 5, No. e1466. (44) Ooko, E., Saeed, M. E. M., Kadioglu, O., Sarvi, S., Colak, M., Elmasaoudi, K., Janah, R., Greten, H. J., and Efferth, T. (2015) Artemisinin derivatives induce iron-dependent cell death (ferroptosis) in tumor cells. Phytomedicine 22, 1045−1054. (45) Zhou, Y., Li, W., and Xiao, Y. (2016) Profiling of Multiple Targets of Artemisinin Activated by Hemin in Cancer Cell Proteome. ACS Chem. Biol. 11 (4), 882−888. (46) Qin, G., Wu, L., Liu, H., Pang, Y., Zhao, C., Wu, S., Wang, X., and Chen, T. (2015) Artesunate induces apoptosis via a ROSindependent and Bax-mediated intrinsic pathway in HepG2 cells. Exp. Cell Res. 336, 308−317. (47) Zhang, F., Gosser, D. K., and Meshnick, S. R. (1992) Hemincatalyzed decomposition of artemisinin (qinghaosu). Biochem. Pharmacol. 43, 1805−1809. (48) Czabotar, P. E., Lessene, G., Strasser, A., and Adams, J. M. (2014) Control of apoptosis by the BCL-2 protein family: implications for physiology and therapy. Nat. Rev. Mol. Cell Biol. 15, 49−63. (49) Adams, J. M., and Cory, S. (2007) The Bcl-2 apoptotic switch in cancer development and therapy. Oncogene 26, 1324−1337. (50) Crespo-Ortiz, M. P., and Wei, M. Q. (2012) Antitumor activity of artemisinin and its derivatives: from a well-known antimalarial agent to a potential anticancer drug. J. Biomed. Biotechnol. 2012, 247597. (51) Leo, C. P., Hsu, S. Y., Chun, S. Y., Bae, H. W., and Hsueh, A. J. (1999) Characterization of the antiapoptotic Bcl-2 family member myeloid cell leukemia-1 (Mcl-1) and the stimulation of its message by gonadotropins in the rat ovary. Endocrinology 140, 5469−5477. (52) Danial, N. N., and Korsmeyer, S. J. (2004) Cell death: critical control points. Cell 116, 205−219. (53) She, Q. B., Solit, D. B., Ye, Q., O’Reilly, K. E., Lobo, J., and Rosen, N. (2005) The BAD protein integrates survival signaling by EGFR/MAPK and PI3K/Akt kinase pathways in PTEN-deficient tumor cells. Cancer Cell 8, 287−297. (54) Andreeff, M., Jiang, S., Zhang, X., Konopleva, M., Estrov, Z., Snell, V. E., Xie, Z., Okcu, M. F., Sanchez-Williams, G., Dong, J., Estey, E. H., Champlin, R. C., Kornblau, S. M., Reed, J. C., and Zhao, S. (1999) Expression of Bcl-2-related genes in normal and AML progenitors: changes induced by chemotherapy and retinoic acid. Leukemia 13, 1881−1892. (1 9) I s ma il , H. M ., B art o n, V . E., P an chan a, M. , Charoensutthivarakul, S., Biagini, G. A., Ward, S. A., and O’Neill, P. M. (2016) A Click Chemistry-Based Proteomic Approach Reveals that 1,2,4-Trioxolane and Artemisinin Antimalarials Share a Common Protein Alkylation Profile. Angew. Chem., Int. Ed. 55, 6401−6405. (20) Robert, A., Cazelles, J., and Meunier, B. (2001) Characterization of the Alkylation Product of Heme by the Antimalarial Drug Artemisinin. Angew. Chem., Int. Ed. 40, 1954−1957. (21) Wang, J., Zhang, C.-J., Chia, W. N., Loh, C. C. Y., Li, Z., Lee, Y. M., He, Y., Yuan, L.-X., Lim, T. K., Liu, M., Liew, C. X., Lee, Y. Q., Zhang, J., Lu, N., Lim, C. T., Hua, Z.-C., Liu, B., Shen, H.-M., Tan, K. S. W., and Lin, Q. (2015) Haem-activated promiscuous targeting of artemisinin in Plasmodium falciparum. Nat. Commun. 6, 10111. (22) Eckstein-Ludwig, U., Webb, R. J., van Goethem, I. D. A., East, J. M., Lee, A. G., Kimura, M., O’Neill, P. M., Bray, P. G., Ward, S. A., and Krishna, S. (2003) Artemisinins target the SERCA of Plasmodium falciparum. Nature 424, 957−961. (23) Lisewski, A. M., Quiros, J. P., Ng, C. L., Adikesavan, A. K., Miura, K., Putluri, N., Eastman, R. T., Scanfeld, D., Regenbogen, S. J., Altenhofen, L., Llinas, M., Sreekumar, A., Long, C., Fidock, D. A., and Lichtarge, O. (2014) Supergenomic network compression and the discovery of EXP1 as a glutathione transferase inhibited by artesunate. Cell 158, 916−928. (24) Mbengue, A., Bhattacharjee, S., Pandharkar, T., Liu, H., Estiu, G., Stahelin, R. V., Rizk, S. S., Njimoh, D. L., Ryan, Y., Chotivanich, K., Nguon, C., Ghorbal, M., Lopez-Rubio, J. J., Pfrender, M., Emrich, S., Mohandas, N., Dondorp, A. M., Wiest, O., and Haldar, K. (2015) A molecular mechanism of artemisinin resistance in Plasmodium falciparum malaria. Nature 520, 683−687. (25) Efferth, T., Dunstan, H., Sauerbrey, A., Miyachi, H., and Chitambar, C. R. (2001) The anti-malarial artesunate is also active against cancer. Int. J. Oncol. 18, 767−773. (26) Singh, N. P., and Lai, H. C. (2004) Artemisinin induces apoptosis in human cancer cells. Anticancer Res. 24, 2277−2280. (27) Efferth, T. (2006) Molecular pharmacology and pharmacogenomics of artemisinin and its derivatives in cancer cells. Curr. Drug Targets 7, 407−421. (28) Ho, W. E., Peh, H. Y., Chan, T. K., and Wong, W. S. F. (2014) Artemisinins: Pharmacological actions beyond anti-malarial. Pharmacol. Ther. 142, 126−139. (29) Lai, H. C., Singh, N. P., and Sasaki, T. (2013) Development of artemisinin compounds for cancer treatment. Invest. New Drugs 31, 230−246. (30) Kwok, J. C., and Richardson, D. R. (2002) The iron metabolism of neoplastic cells: alterations that facilitate proliferation? Crit. Rev. Oncol. Hematol. 42, 65−78. (31) Efferth, T. (2007) Willmar Schwabe Award 2006: antiplasmodial and antitumor activity of artemisinin - from bench to bedside. Planta Med. 73, 299−309. (32) Huang, C., Ba, Q., Yue, Q., Li, J., Li, J., Chu, R., and Wang, H. (2013) Artemisinin rewires the protein interaction network in cancer cells: network analysis, pathway identification, and target prediction. Mol. BioSyst. 9, 3091−3100. (33) Odaka, Y., Xu, B., Luo, Y., Shen, T., Shang, C., Wu, Y., Zhou, H., and Huang, S. (2014) Dihydroartemisinin inhibits the mammalian target of rapamycin-mediated signaling pathways in tumor cells. Carcinogenesis 35, 192−200. (34) Thanaketpaisarn, O., Waiwut, P., Sakurai, H., and Saiki, I. (2011) Artesunate enhances TRAIL-induced apoptosis in human cervical carcinoma cells through inhibition of the NF-κB and PI3K/ Akt signaling pathways. Int. J. Oncol. 39, 279−285. (35) Cho, S., Oh, S., Um, Y., Jung, J.-H., Ham, J., Shin, W.-S., and Lee, S. (2009) Synthesis of 10-substituted triazolyl artemisinins possessing anticancer activity via Huisgen 1,3-dipolar cylcoaddition. Bioorg. Med. Chem. Lett. 19, 382−385. (36) Hinds, M. G., Smits, C., Fredericks-Short, R., Risk, J. M., Bailey, M., Huang, D. C., and Day, C. L. (2007) Bim, Bad and Bmf: intrinsically unstructured BH3-only proteins that undergo a localized 642 DOI: 10.1021/acschembio.8b01004 ACS Chem. Biol. 2019, 14, 636−643 Articles ACS Chemical Biology (55) Datta, S. R., Dudek, H., Tao, X., Masters, S., Fu, H., Gotoh, Y., and Greenberg, M. E. (1997) Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell 91, 231− 241. (56) del Peso, L., Gonzalez-Garcia, M., Page, C., Herrera, R., and Nunez, G. (1997) Interleukin-3-induced phosphorylation of BAD through the protein kinase Akt. Science 278, 687−689. (57) Donovan, N., Becker, E. B., Konishi, Y., and Bonni, A. (2002) JNK phosphorylation and activation of BAD couples the stressactivated signaling pathway to the cell death machinery. J. Biol. Chem. 277, 40944−40949. (58) Zha, J., Harada, H., Yang, E., Jockel, J., and Korsmeyer, S. J. (1996) Serine phosphorylation of death agonist BAD in response to survival factor results in binding to 14-3-3 not Bcl-x(L). Cell 87, 619− 628. (59) Oltersdorf, T., Elmore, S. W., Shoemaker, A. R., Armstrong, R. C., Augeri, D. J., Belli, B. A., Bruncko, M., Deckwerth, T. L., Dinges, J., Hajduk, P. J., Joseph, M. K., Kitada, S., Korsmeyer, S. J., Kunzer, A. R., Letai, A., Li, C., Mitten, M. J., Nettesheim, D. G., Ng, S. C., Nimmer, P. M., O’Connor, J. M., Oleksijew, A., Petros, A. M., Reed, J. C., Shen, W., Tahir, S. K., Thompson, C. B., Tomaselli, K. J., Wang, B., Wendt, M. D., Zhang, H., Fesik, S. W., and Rosenberg, S. H. (2005) An inhibitor of Bcl-2 family proteins induces regression of solid tumours. Nature 435, 677−681. (60) Luo, J., Zhu, W., Tang, Y., Cao, H., Zhou, Y., Ji, R., Zhou, X., Lu, Z., Yang, H., Zhang, S., and Cao, J. (2014) Artemisinin derivative artesunate induces radiosensitivity in cervical cancer cells in vitro and in vivo. Radiat. Oncol. 9, 84. (61) Chen, T., Li, M., Zhang, R., and Wang, H. (2009) Dihydroartemisinin induces apoptosis and sensitizes human ovarian cancer cells to carboplatin therapy. J. Cell. Mol. Med. 13, 1358−1370. (62) Wang, B., Hou, D., Liu, Q., Wu, T., Guo, H., Zhang, X., Zou, Y., Liu, Z., Liu, J., Wei, J., Gong, Y., and Shao, C. (2015) Artesunate sensitizes ovarian cancer cells to cisplatin by downregulating RAD51. Cancer Biol. Ther. 16, 1548−1556. (63) Chen, K., Shou, L.-M., Lin, F., Duan, W.-M., Wu, M.-Y., Xie, X., Xie, Y.-F., Li, W., and Tao, M. (2014) Artesunate induces G2/M cell cycle arrest through autophagy induction in breast cancer cells. AntiCancer Drugs 25, 652−662. (64) Berlow, R. B., Dyson, H. J., and Wright, P. E. (2018) Expanding the Paradigm: Intrinsically Disordered Proteins and Allosteric Regulation. J. Mol. Biol. 430, 2309−2320. 643 DOI: 10.1021/acschembio.8b01004 ACS Chem. Biol. 2019, 14, 636−643