Characterization of the Artemisinin Binding Site for Translationally

Nov 16, 2016 - Controlled Tumor Protein (TCTP) by Bioorthogonal Click Chemistry ... binding site of P. falciparum translationally controlled tumor pro...
0 downloads 0 Views 3MB Size
Communication pubs.acs.org/bc

Characterization of the Artemisinin Binding Site for Translationally Controlled Tumor Protein (TCTP) by Bioorthogonal Click Chemistry Weichao Li,†,‡,∥ Yiqing Zhou,†,∥ Guanghui Tang,§ and Youli Xiao*,†,‡ †

CAS Key Laboratory of Synthetic Biology, CAS Center for Excellence in Molecular Plant Sciences, Institute of Plant Physiology and Ecology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai 200032, China ‡ University of Chinese Academy of Sciences, Beijing 100039, China § School of Chemical Biology and Biotechnology, Peking University Shenzhen Graduate School, Shenzhen 518055, China S Supporting Information *

ABSTRACT: Despite the fact that multiple artemisinin-alkylated proteins in Plasmodium falciparum have been identified in recent studies, the alkylation mechanism and accurate binding site of artemisinin−protein interaction have remained elusive. Here, we report the chemical-probe-based enrichment of the artemisinin-binding peptide and characterization of the artemisininbinding site of P. falciparum translationally controlled tumor protein (TCTP). A peptide fragment within the N-terminal region of TCTP was enriched and found to be alkylated by an artemisinin-derived probe. MS2 fragments showed that artemisinin could alkylate multiple amino acids from Phe12 to Tyr22 of TCTP, which was supported by labeling experiments upon site-directed mutagenesis and computational modeling studies. Taken together, the “capture-and-release” strategy affords consolidated advantages previously unavailable in artemisinin−protein binding site studies, and our results deepened the understanding of the mechanism of protein alkylation via heme-activated artemisinin.

A

enable the profiling of proteome-wide interactions of artemisinin in complex biological systems.8,9 More proteins were identified as specific artemisinin-alkylated targets in P. falciparum by using the activity-based probes of artemisinin. This further confirmed the “cluster bomb” hypothesis and the multitarget property of artemisinin inside the parasite.10 Although the protein targets of artemisinin have been extensively studied, the molecular mechanism of the alkylation reactions and the accurate modification site of artemisininprotein adduct remain controversial. In the past decades, a lot of attention has been paid to the binding of artemisinin with model proteins and synthetic peptides, but it has not been understood well in terms of determining the interaction between artemisinin and its native target proteins.11−13 The translationally controlled tumor protein (TCTP) is a highly conserved housekeeping protein that is widely expressed in all

rtemisinin, a sesquiterpene lactone extracted from the Chinese herb Artemisia annua, is highly effective against drug-resistant Plasmodium falciparum strains during multiple stages of the parasite development.1 The endoperoxide bridge has been suggested as the pharmacophore of artemisinin and its semisynthetic derivatives, which is indispensable for their antimalarial activities.2 It is widely accepted that inside the parasite, the endoperoxide bridge can be activated and cleaved by heme-Fe(II), which releases carbon-centered radicals.2,3 These highly reactive radicals can alkylate heme and form covalent adducts with essential parasite macromolecules, which results in the rapid death of the parasite.4 The C4-centered radical can also react with glutathione to alkylate the peptide through a thioether linkage, indicating that a covalent bond between artemisinin and protein may be formed in this way (Scheme 1).5 Early studies have revealed two direct targets of artemisinin in P. falciparum, including the sarco−endoplasmic reticulum Ca2+-ATPase (SERCA, PfATP6) and translationally controlled tumor protein (Pf TCTP).6,7 Recently, chemoproteomics technology has arisen as a powerful strategy to © XXXX American Chemical Society

Received: September 25, 2016 Revised: November 15, 2016 Published: November 16, 2016 A

DOI: 10.1021/acs.bioconjchem.6b00556 Bioconjugate Chem. XXXX, XXX, XXX−XXX

Communication

Bioconjugate Chemistry

investigate the exact binding site of artemisinin within TCTP with conventional mass spectrometry method but failed to produce any artemisinin-modified peptides by mass spectrometry. This might be explained by the insufficient amount of each kind of artemisinin-modified peptides generated by nonselective promiscuous radical alkylation.13 In the present study, to investigate the artemisinin-binding site within TCTP by mass spectrometry, we overexpressed P. falciparum TCTP in Escherichia coli (Figure S1) and incubated the recombinant protein with artemisinin in the presence of hemin and sodium ascorbate (NaVc). Unfortunately, after tryptic digestion we were not able to find any ART-modified peptides (Figure S3A), which was similar to previously reported results by Efferth et al.13 Alternatively, we attempted to investigate the binding site of artemisinin with TCTP by employing an artemisinin-derived clickable probe ART-P1 (Figure 1A) and an acid-cleavable biotin−azide tag DADPS (Figure 1B) through a bioorthogonal “capture-and-release” strategy to enrich the artemisinin-specific peptides by using an artemisinin activity-based probe (ART-P1) prior to mass spectrometry analysis (Figure 1C).18

Scheme 1. Heme-Induced Activation of Endoperoxides and Formation of Reactive Radical Intermediates Capable of Reacting with Parasite Proteins

eukaryotic organisms.14 Although the physiological role of this protein in P. falciparum remains elusive, more evidence suggests that human TCTP plays a significant role in cell cycle regulation, malignant transformation, and immunological functions.14,15 Artemisinin and its derivatives have been found to be able to specifically interact with both P. falciparum and Homo sapiens TCTPs.7,16 Heme-dependent covalent alkylation of P. falciparum TCTP (Pf TCTP) by artemisinin has been validated both in vitro and in vivo.7 Efferth’s group has tried to



RESULTS AND DISCUSSION Gel-Based Labeling of TCTP with Artemisinin-Derived Probe ART-P1. First, to assess the feasibility of ART-P1 in TCTP alkylation, we conducted gel-based labeling experiments with recombinant TCTP in vitro as previously described.17 The detailed procedures of TCTP treatment were provided in the

Figure 1. (A) Structures of artemisinin (ART) and artemisinin activity-based probe ART-P1. (B) Structure of acid-cleavable biotin-azide (DADPS) and its reaction with formic acid (FA). (C) Workflow of “capture-and-release” strategy for the identification of the binding site of artemisinin and Pf TCTP. B

DOI: 10.1021/acs.bioconjchem.6b00556 Bioconjugate Chem. XXXX, XXX, XXX−XXX

Communication

Bioconjugate Chemistry

Figure 2. Gel-based labeling of recombinant TCTP by ART-P1. (A) Hemin-mediated probe labeling of TCTP can be competed by ART but not by D-ART. (B) Labeling of TCTP by ART-P1 with increasing concentrations of hemin. (C) Labeling of TCTP by increasing concentrations of ARTP1 with 10 μM of hemin. (D) Upper panel: The effect of thiol-blocking reagents (IAA and NEM) and reducing agent (NaVc) on hemin-mediated ART-P1 labeling of TCTP. Lower panel: corresponding intensity of ART-P1 labeling of TCTP in the fluorescent gel above. Fluorescence intensity were measured spectrophotometrically. Bars represent the average of at least three replicates measurements, and error bars represent standard error of the mean.

Figure 3. Characterization of the TCTP binding sites by an integrated HCD spectrum of probe-modified peptides DVFTNDEVCSDSYVQQDPFEVPEFR; a zoomed-in panel displays the diagnostic fragment ion (DFI) peak (m/z: 199.11).

cysteine blocking reagents, namely iodoacetamide (IAA) or Nethylmaleimide (NEM), diminished the labeling, and adding sodium ascorbate (NaVc) recovered the fluorescence signal (Figure 2D). Although the only cysteine residue (Cys19) of TCTP has been proposed as one of possible artemisinin alkylation sites,7 our results suggested this cysteine might not be a specific alkylation site but could serve as electron donor during endoperoxide activation. Enrichment of ART-P1-Alkylated Peptide and Mass Spectrometry Analysis. Encouraged by the results of gelbased labeling experiments, we attempted to enrich and characterize the artemisinin-modified peptide within TCTP.

Biological Experimental Procedures section of the Supporting Information. After incubation with ART-P1 and hemin, protein samples were subjected to Cu (I)-catalyzed click (CuAAC) reaction with tetramethylrhodamine−azide (TAMRA−N3), separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), and detected by in-gel fluorescence. Coomassie blue staining indicated equal amount of recombinant protein loaded in each lane. As predicted, the labeling could be competed by excess artemisinin (ART) but not by the inactive analog deoxyartemisinin (D-ART) (Figure 2A), and the labeling was both probe- and hemin-dependent (Figure 2B,C). In addition, preincubating the proteins with C

DOI: 10.1021/acs.bioconjchem.6b00556 Bioconjugate Chem. XXXX, XXX, XXX−XXX

Communication

Bioconjugate Chemistry Table 1. Detected Peptide Precursors Alkylated by Artemisinin Probe ART-P1a sequence

theoretical [M + H]+

experimental [M + H]3+

charge

ΔM [ppm]

DVFTNDEVCSDSYVQQDPFEVPEFR DVFTNDEVCSDSYVQQDPFEVPEFR+model A1 DVFTNDEVCSDSYVQQDPFEVPEFR+model A1 DVFTNDEVCSDSYVQQDPFEVPEFR+model A2 DVFTNDEVCSDSYVQQDPFEVPEFR+model A2 tagged molecule

2965.2857 3471.5559 3473.5689 3468.5621 3469.5476 molecular formula

989.1001 1157.8568 1158.5278 1156.8589 1157.1874

3 3 3 3 3 theoretical mass

−8.81 −10.92 2.05 −9.24 −8.81

model A1 model A2

C26H42N4O6 C24H38N4O4

506.3104 446.2893

a The HCD spectra of these precursors are shown in Figure S5. The structures of binding models A1 and A2 are shown in Figure S6. : deamination; : carbamidomethyl.

Figure 4. Validation of binding sites of ART and TCTP. (A) Gel-based labeling of wild-type (WT) and mutant TCTPs (F12A, D15A, C19A, and Y22A) by ART-P1 in the absence and presence of NaVc. (B) Docking of the artemisinin into P. falciparum TCTP (PDB ID code: 3P3K) by Molecular Operating Environment software.

Compared to the total ion chromatography (TIC) of artemisinin-treated TCTP, several additional signal peaks were detected in the TIC of probe-treated TCTP subjected to affinity capture and acid release (Figure S3B, arrowed peaks). To search for tandem mass spectra containing the probespecific ions, we extracted the specific diagnostic fragment ions (DFI) at m/z of 199.11, which was corresponding to the characteristic MS2 fragmentation of the acid-cleaved DADPS− probe conjugate (Figures 3 and S4), from the total ion chromatography by using Thermo Xcalibur software. To our surprise, the DFI signals were observed in multiple HCD spectra (>500) with doubly or triply charged precursor ions eluted within 50−52 min, which could be candidates for possible probe-modified peptides and were manually searched for possible tryptic peptide matches. Representative highquality HCD spectra according to the intensities of DFI were

Recombinant TCTP (5.0 mg/mL) was incubated with ART-P1 (10 μM) in the presence of hemin and NaVc for 3 h and subjected to click chemistry with a biotin−azide tag DADPS bearing an acid-cleavable site (Figure 1B).18 After enrichment by pull-down with streptavidin beads, unlabeled proteins were washed out, and captured proteins were released by 5% HCOOH. The eluent was neutralized with NH4HCO3 (100 mM) and concentrated by ultrafiltration, subjected to reductive alkylation by dithiothreitol (DTT) and IAA, and digested by trypsin overnight. The digested peptides were desalted, concentrated, and analyzed by Thermo Q-Exactive mass spectrometry in data-dependent acquisition (DDA) scan mode. We chose the top 20 precursor ions in MS1 spectra for MS2 acquisition using higher-energy collision-induced dissociation (HCD) activation. D

DOI: 10.1021/acs.bioconjchem.6b00556 Bioconjugate Chem. XXXX, XXX, XXX−XXX

Communication

Bioconjugate Chemistry

information on the role of Phe12 in artemisinin binding, we speculated that Phe12 could be essential for the secondary structure of the artemisinin-binding domain of TCTP. Taken together, these results were in agreement with one of the binding mode of artemisinin previously proposed: first, artemisinin noncovalently binds protein within a specific domain; next, the endoperoxide moiety is activated by heme (reduced from hemin) to generate artemisinin-derived reactive radicals; finally, the radicals alkylate the neighborhood amino acid residues around the binding domain.17

shown in Figure S5. By comparing the theoretical mass of the peptide + probe with the experimental data for the deconvoluted precursor mass in each of the spectra in Figure S4, the tryptic peptide DVFTNDEVCSDSYVQQDPFEVPEFR (10−34) was identified as the probe-modified candidate, and the amino acid sequence was confirmed unambiguously according to numerous y-type ions formed after HCD fragmentation (Table 1 and Figure S5). Both modification models before (model A1) and after (model A2) the loss of an acetate molecule followed by protein alkylation (Figure S6) were detected (Table 1). However, according to the information obtained from any single HCD spectrum, we found the existence of multiple alkylation sites within this peptide, which could explain the distinct retention time of multiple precursors corresponding to the same peptide (Figure S5). Therefore, to globally examine the modification sites, we integrated the MS2 data of all HCD spectra within 50−52 min to generate a combined spectrum with all of the fragment ions (Figure 3). As shown in the integrated spectrum, we only found the residues within the ranges of b1−b2 and y1−y12 were never modified, while all other residues among b3−b12 (y13−y22), except for Thr13 and Val18, were able to be modified by the probe (Table S1). Because it has been suggested that the hemeinduced artemisinin radicals are short-lived and can only covalently react with the neighboring amino acid residues,19 we speculated that unlike those site-specific and covalent chemical warheads, such as iodoacetamide, phosphofluoride and sulfonyl fluoride,20−22 the heme-activated artemisinin could randomly modify the amino acid residues nearby with appropriate position and distance. Validation of Artemisinin-Binding Sites by SiteDirected Mutation and Molecular Docking. To validate the results obtained from mass spectrometry, we generated a series of mutants, including F12A, D15A, C19A, and Y22A of Pf TCTP recombinant proteins. These mutants, together with the wild-type TCTP, were reacted equally and treated with ART-P1 in the presence of hemin. After conjugating with TAMRA-N3 the samples were resolved by SDS-PAGE, and the labeling was evaluated by in-gel fluorescence. Interestingly, the results demonstrated that replacing Cys19 with Ala diminished all of the labeling, and replacing Phe12 caused a significant decrease in the fluorescent intensity, whereas the D15A or F22A mutants had only a small effect on the labeling. With the addition of reducing agent NaVc to recover the electron source, all of the mutants were successfully labeled with the fluorescence intensity order of Y22A > D15A > WT > F12A > C19A. Combined with the block-recovery labeling studies mentioned above (Figure 2D), these results highlighted the importance of Cys19 as the electron donor during endoperoxide activation and of both Phe12 and Cys19 as key residues on the interaction of artemisinin with TCTP. In addition, to elucidate the potential mode of interaction, artemisinin was blindly docked into the crystal structure of Pf TCTP (PDB ID code: 3P3K) by Molecular Operating Environment (MOE) software (Figure 4B). The result showed a binding pocket with Lys9, Thr13, Asn14, Asp15, Glu16, Tyr175, His181, and His182 residues that were visualized, in which a binding region with a 3.34 Å was considered and the binding affinity was calculated as −1.0 kcal/mol. A hydrogen bond was observed between Lys9 and the endoperoxide bridge of artemisinin. The endoperoxide bridge was close to the residues from Thr13 to Glu16, which was in accordance with mass spectrometry results. Although the docking model did not provide direct



CONCLUSIONS In conclusion, we have characterized an irreversible alkylation mechanism of TCTP by a heme-activated artemisinin activitybased probe. Our investigation has led to the identification of an artemisinin-alkylated peptide fragment within TCTP by utilizing a bioorthogonal click chemistry based “capture-andrelease” approach. Our data represent the first mass spectrometry evidence of a specific binding site of artemisinin and its native protein target; the “capture-and-release” strategy affords consolidated advantages not previously available in small molecule−protein binding site studies. Admittedly, the native binding mode and mechanism of artemisinin−TCTP could be impaired or changed by the alkyne modification to the structure of artemisinin, which is a potential limitation of our approach.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.bioconjchem.6b00556. Procedures of the biological experiments and mass spectrometry data. (PDF)



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: 86-21-54924226. ORCID

Weichao Li: 0000-0001-6875-7315 Yiqing Zhou: 0000-0002-6391-3259 Youli Xiao: 0000-0002-4803-3333 Author Contributions ∥

W.L. and Y.Z. contributed equally to this work.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We are grateful for the generous funding support from the National Natural Science Foundation of China (grant nos. 21502206 and 21572243), the “Thousand Talents Program” Young Investigator Award, the Key Projects of Shanghai Committee of Science and Technology (grant no. 15JC1400402), and the National Key Laboratory of Bioorganic and Natural Product Chemistry. We thank Dr. Yuanhong Shan in the Core Facility Centre of the Institute of Plant Physiology and Ecology for mass spectrometry assistance.



REFERENCES

(1) Tu, Y. (2011) The discovery of artemisinin (qinghaosu) and gifts from Chinese medicine. Nat. Med. 17, 1217−1220.

E

DOI: 10.1021/acs.bioconjchem.6b00556 Bioconjugate Chem. XXXX, XXX, XXX−XXX

Communication

Bioconjugate Chemistry (2) 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. (3) Robert, A., and Meunier, B. (1997) Characterization of the first covalent adduct between artemisinin and a heme model. J. Am. Chem. Soc. 119, 5968−5969. (4) Meunier, B., and Robert, A. (2010) Heme as trigger and target for trioxane-containing antimalarial drugs. Acc. Chem. Res. 43, 1444− 1451. (5) Wang, D. Y., and Wu, Y. L. (2000) A possible antimalarial action mode of qinghaosu (artemisinin) series compounds. Alkylation of reduced glutathione by C-centered primary radicals produced from antimalarial compound qinghaosu and 12-(2, 4-dimethoxyphenyl)-12deoxoqinghaosu. Chem. Commun. 22, 2193−2194. (6) Eckstein-Ludwig, U., Webb, R. J., Van Goethem, I. D. A., East, J. M., Lee, A. G., Kimura, M., O’Neill, P. M., Ward, S. A., Krishna, S., and Bray, P. G. (2003) Artemisinins target the SERCA of Plasmodium falciparum. Nature 424, 957−961. (7) Bhisutthibhan, J., Pan, X. Q., Hossler, P. A., Walker, D. J., Yowell, C. A., Carlton, J., Dame, J. B., and Meshnick, S. R. (1998) The Plasmodium falciparum translationally controlled tumor protein homolog and its reaction with the antimalarial drug artemisinin. J. Biol. Chem. 273, 16192−16198. (8) Ismail, H. M., Barton, V., Phanchana, M., Charoensutthivarakul, S., Wong, M. H., Hemingway, J., Biagini, G. A., O’Neill, P. M., and Ward, S. A. (2016) Artemisinin activity-based probes identify multiple molecular targets within the asexual stage of the malaria parasites Plasmodium falciparum 3D7. Proc. Natl. Acad. Sci. U. S. A. 113, 2080− 2085. (9) Wang, J., Zhang, C., Chia, W., Loh, C., Li, Z., Lee, Y., He, Y., Yuan, L., Lim, T., Liu, M., Liew, C., Lee, Y., Zhang, J., Lu, N., Lim, C., Hua, Z., Liu, B., Shen, H., Tan, S., and Lin, Q. (2015) Haem-activated promiscuous targeting of artemisinin in Plasmodium falciparum. Nat. Commun. 6, 10111. (10) Ismail, H. M., Barton, V. E., Panchana, 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. 128, 6511−6515. (11) 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. (12) Pandey, A. V., Tekwani, B. L., Singh, R. L., and Chauhan, V. S. (1999) Artemisinin, an endoperoxide antimalarial, disrupts the hemoglobin catabolism and heme detoxification systems in malarial parasite. J. Biol. Chem. 274, 19383−19388. (13) Eichhorn, T., Winter, D., Büchele, B., Dirdjaja, N., Frank, M., Lehmann, W. D., Mertens, R., Krauth-Siegel, R. L., Simmet, T., Granzin, J., and Efferth, T. (2013) Molecular interaction of artemisinin with translationally controlled tumor protein (TCTP) of Plasmodium falciparum. Biochem. Pharmacol. 85, 38−45. (14) Bommer, U. A., and Thiele, B. J. (2004) The translationally controlled tumour protein (TCTP). Int. J. Biochem. Cell Biol. 36, 379− 385. (15) Johnson, T. M., Antrobus, R., and Johnson, L. N. (2008) Plk1 activation by Ste20-like kinase (Slk) phosphorylation and polo-box phosphopeptide binding assayed with the substrate translationally controlled tumor protein (TCTP). Biochemistry 47, 3688−3696. (16) Fujita, T., Felix, K., Pinkaew, D., Hutadilok-Towatana, N., Liu, Z., and Fujise, K. (2008) Human fortilin is a molecular target of dihydroartemisinin. FEBS Lett. 582, 1055−1060. (17) 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, 882−888. (18) Szychowski, J., Mahdavi, A., Hodas, J. J., Bagert, J. D., Ngo, J. T., Landgraf, P., Dieterich, D. C., Schuman, E. M., and Tirrell, D. A.

(2010) Cleavable biotin probes for labeling of biomolecules via azide− alkyne cycloaddition. J. Am. Chem. Soc. 132, 18351−18360. (19) Rhee, H. W., Zou, P., Udeshi, N. D., Martell, J. D., Mootha, V. K., Carr, S. A., and Ting, A. Y. (2013) Proteomic mapping of mitochondria in living cells via spatially restricted enzymatic tagging. Science 339, 1328−1331. (20) Liu, Y., Patricelli, M. P., and Cravatt, B. F. (1999) Activity-based protein profiling: the serine hydrolases. Proc. Natl. Acad. Sci. U. S. A. 96, 14694−14699. (21) Zhou, Y., Guo, T., Tang, G., Wu, H., Wong, N. K., and Pan, Z. (2014) Site-selective protein immobilization by covalent modification of GST fusion proteins. Bioconjugate Chem. 25, 1911−1915. (22) Macpherson, L. J., Dubin, A. E., Evans, M. J., Marr, F., Schultz, P. G., Cravatt, B. F., and Patapoutian, A. (2007) Noxious compounds activate TRPA1 ion channels through covalent modification of cysteines. Nature 445, 541−545.

F

DOI: 10.1021/acs.bioconjchem.6b00556 Bioconjugate Chem. XXXX, XXX, XXX−XXX