Novel Inhibitors of Human DOPA Decarboxylase Extracted from

Letters pubs.acs.org/acschemicalbiology

Novel Inhibitors of Human DOPA Decarboxylase Extracted from Euonymus glabra Roxb. Jie Ren,‡,⊥ Yuanyuan Zhang,†,∥,⊥ Huizi Jin,‡ Jing Yu,§ Yueyang Zhou,† Fang Wu,*,† and Weidong Zhang*,‡ †

Key Laboratory of Systems Biomedicine (Ministry of Education), Shanghai Center for Systems Biomedicine, ‡School of Pharmacy, and §State Key Laboratory of Microbial Metabolism & School of Life Sciences and Biotechnology, Shanghai Jiao Tong University, Shanghai 200240, China ∥ Key Laboratory of Exploration and Utilization of Aquatic Genetic Resources, Ministry of Education, College of Fisheries and Life Science, Shanghai Ocean University, Shanghai 201306, China S Supporting Information *

ABSTRACT: Dopamine, a biogenic amine with important biological functions, is produced from L-DOPA by DOPA decarboxylase (DDC). DDC is a potential target to modulate the production of dopamine in several pathological states. Known inhibitors of DDC have been used for treatment of Parkinson’s disease but suffered low specificity and diverse side effects. In the present study, we identified and characterized a novel class of natural-product-based selective inhibitors for DDC from the extract of Euonymus glabra Roxb. by a newly developed high-throughput enzyme assay. The structures of these inhibitors are dimeric diarylpropane, a unique chemical structure containing a divalent dopamine motif. The most effective inhibitors 5 and 6 have an IC50 of 11.5 ± 1.6 and 21.6 ± 2.7 μM in an in vitro purified enzyme assay, respectively, but did not inhibit other homologous enzymes. Compound 5 but not 6 dosedependently suppressed the activity of hDDC and dopamine levels at low micromolar concentrations in cells. Furthermore, structure−activity relationship analyses revealed that p-benzoquinone might be a crucial moiety of these inhibitors for inhibiting hDDC. The natural-product-based selective inhibitors of hDDC could serve as a chemical lead for developing improved drugs for dopamine-related disease states.

P

high-throughput mode, a biochemical assay has developed to link the decarboxylase activity of hDDC with the absorbance of NADH, which has been used for measuring of the activity of ornithine decarboxylase.11,12 For screening of inhibitors, we chose the concentration of substrate near its Km value (∼350 μM; ref 13), i.e., 750 μM L-DOPA. An optimal window for measuring the specific activity of DDC was observed when the assay was run at the time of ∼60 min under these assay conditions (Supplementary Figure S1A), which could be inhibited by methyldopa, a known competitive inhibitor of DDC (Figure 1A).14 Furthermore, methyldopa showed a dose−response curve with an IC50 of 1200 ± 3.9 μM for DDC in this in vitro purified enzyme assay but did not affect the activities of coupling phosphoenolpyruvate carboxylase (PEPC) and malate dehydrogenase (MDH) enzymes (Figure 1A and Supplementary Figure S1B). The well-to-well reproducibility of the assay was validated with 1000 μM (∼IC50) methyldopa as inhibitor. The results showed distinct differences between the control and the methyldopa-treated groups (Supplementary Figure S1C). The average Z′ values of the assay were ∼0.55

arkinson’s disease (PD) is a frequent neurodegenerative disorder with more than 6 million patients worldwide.1 Affected patients have lost substantial dopaminergic neurons within the substantia nigra, which results in a lack of the neurotransmitter dopamine.2,3 Raising dopamine levels in the substantia nigra of brain is currently one of the main therapeutic approaches for PD.4 The combination of administrating L-DOPA and a human DOPA decarboxylase (hDDC) inhibitor is the classic treatment for PD patients.2,4,5 hDDC catalyzes L-DOPA into dopamine and is a validated drug target for PD by increasing the half-life of L-DOPA in blood plasma, thereby increasing the dopamine content in the brain.2,6 Known inhibitors of DDC are mainly of the irreversible type and bind covalently to the pyridoxal-5′phosphate (PLP) cofactor of DDC,7 therefore suffering a notorious unspecificity to other PLP-dependent enzymes and resulting in clinical side effects.8,9 To date, no high-throughput assay has been available for monitoring the activity of DDC. Current procedures for measuring the activity of DDC need additional steps for separation and detection of the product dopamine from the reaction mixtures and often are carried out in vials with large reaction volumes;9,10 thus they are not suitable for highthroughput screening. To monitor the activity of hDDC in a © 2014 American Chemical Society

Received: June 19, 2013 Accepted: January 28, 2014 Published: January 28, 2014 897

dx.doi.org/10.1021/cb500009r | ACS Chem. Biol. 2014, 9, 897−903

ACS Chemical Biology

Letters

Figure 1. Methyldopa and natural products (1−14) isolated from E. glabra on the activities of human DOPA decarboxylase (DDC) and cystathionine γ-lyase (CSE). (A) The structure of methyldopa and its inhibitory effects (IC50; μM) on hDDC and hCSE. (B) Structures for 1−14 and their IC50’s (μM) on hDDC and hCSE.

using 1000 μM methyldopa as positive controls (Supporting Information). With the validated high-throughput assay for DDC in hand, we started to search novel inhibitors of DDC from unexplored extractions of various plants or microorganisms, which resulted in identifying an active extract from Euonymus glabra Roxb. (E. glabra), a medicinal plant located in Xishuangbanna of southernmost China. Fractionation of the ethanolic extract of this plant was carried out and yielded nine members of the rare class of dimeric diarylpropanes (1−9; Figure 1B), i.e., euonydiarylpropanes A− D (1−4), euonyquinone A (5), combrequinone B (6),15 euonyquinone B (7), (−)-euonyquinone C (8), and (+)-euonyquinone C (9), along with three diarylpropanes

(10−12)16,17 and two flavans (13 and 14).18,19 Their structures were elucidated on the basis of HR-ESIMS, 1D and 2D NMR spectroscopic methods, which discovered eight new dimeric diarylpropanes (1−5 and 7−9), one known dimeric diarylpropane (6, the only known compound for this class), and five other known compounds (10−14). The assignments for the structures of newly discovered compounds 1−5 are discussed below, and those for 7−9 are discussed in Supporting Information. Compound 1 was isolated as light yellow solid. Its molecular formula of C35H40O9 was deduced from a [M + Na]+ ion peak at m/z 627.2559 in the positive HR-ESIMS, indicating 16 degrees of unsaturation. The IR spectrum exhibited the absorption bands of hydroxyl groups (3428 cm−1) and benzene 898

dx.doi.org/10.1021/cb500009r | ACS Chem. Biol. 2014, 9, 897−903

ACS Chemical Biology

Letters

rings (1605, 1514 cm−1). The 1H NMR spectrum of 1 showed the presence of two ABX-substituted benzene rings (rings B and D), two AX-substituted benzene rings (rings A and C), one methine [δH 4.50 (1H, t, J = 7.8 Hz)], five methylene groups, and five methyl groups (Supplementary Table S1), which suggested two diarylpropane units in 1. The assumption was further confirmed by analyses of 2D NMR data. Two spin systems [H-1/H2-2/H2-3 (unit I) and H2-1/H2-2/H2-3 (unit II)] deduced from 1H−1H COSY experiment coupled with HMBC correlations [unit I: H-1/C-1′, 2′, 6′, H2-3/C-1″, 2″, 6″, unit II: H2-1/C-1′, 2′, 6′, H2-3/C-1″, 2″, 6″ and H-1 (unit I)/ C-4′, 5′, 6′ (unit II)] established the planar structure of 1 (Supplementary Figure S2). Compound 1 was given the name euonydiarylpropane A. Compounds 2, 3, and 4 all exhibited similar NMR patterns compared to those of compound 1 (Supplementary Tables S1 and S2). Their structures were determined by 2D NMR studies and named euonydiarylpropanes B, C, and D, respectively. Compound 5 was obtained as a brown amorphous solid. Its molecular formula C34H34O10 was characterized by a [M + H]+ ion peak at m/z 603.2225 in the positive HR-ESIMS spectrum, indicating 18 degrees of unsaturation. The IR spectrum exhibited the presence of hydroxyl groups (3438 cm−1) and carbonyl groups (1645 cm−1). The 1H NMR spectrum of 5 displayed two ABX-substituted benzene rings (rings B and D), an AX-substituted benzene ring (ring C), an olefinic proton [δH 6.00 (1H, s)], five methylene groups, and four methyl groups (Figure 1B and Supplementary Table S3). In accordance with the IR absorptions, two carbonyl signals (δC 182.0, 186.9) were observed, which indicated the presence of a p-benzoquinone residue. Thus, the 1H and 13C NMR signals of rings A−D were both assigned by HMQC correlations. The analysis of the 1 H−1H COSY spectrum revealed the spin system of H2-1/H22/H2-3 (unit II) as shown in Supplementary Figure S3. Additionally, the HMBC correlations of H2-1/C-1′, 2′, 6′ and H2-3/C-1″, 2″, 6″ established unit II as a diarylpropane, and the correlation from H-6′ (unit II) to C-2′ (unit I) indicated ring C to be connected to ring A through C-5′ (unit II) and C-2′ (unit I). Similarly in unit I, the HMBC correlations of H2-1/C-2′, 3′, 4′ and H2-3/C-1″, 2″, 6″ revealed that C-1 and C-3 were connected to ring A and ring B through C-3′ and C-1″, respectively. Further, H2-1 and H2-3 were observed to correlate to δC 203.6, which suggested that the remaining carbonyl group was attached to C-2 (unit I). Therefore, the structure of 5 was assigned and named euonyquinone A. A possible biosynthetic pathway and hypothesis of synthetic pathway for 5 are depicted in Supplementary Figures S4 and S5. Among the isolated natural products from E. glabra (1−14), seven dimeric diarylpropanes (1, 2, and 5−9) showed inhibitory effects to the activity of hDDC in the in vitro purified enzyme assay (Figure 1B). Compounds 5 and 6 were the most effective inhibitors with IC50 of 11.5 ± 1.6 and 21.6 ± 2.7 μM, respectively (Figure 2A), whereas 1, 2, and 7−9 displayed weaker inhibitory effects on the activity of hDDC (Figure 1B). Furthermore, 5 or 6 up to 200 μM did not influence the activity of coupling PEPC and MDH enzymes, indicating that they directly inhibited the hDDC (Figure 2B). All of the active compounds contained two fragments of benzene rings modified by 3,4-dihydroxy or methoxy groups, which is the analogue of dopamine (Figure 1). On the contrary, 10−14 with only one hydroxyl modified fragment were inactive to hDDC (Figure 1). Apart from the feature of being analogues of the dopamine product of hDDC, the p-benzoquinone moiety

Figure 2. Effects of dimeric diarylpropanes on the activity of purified hDDC in vitro. (A) Dose-dependent effects of 5 (●) and 6 (○) on the activity of hDDC in the DDC-MDH-PEPC-Linked assay (for details, see Supporting Information). (B) Effects of 5 and 6 on the activity of malate dehydrogenase (MDH) and phosphoenolpyruvate carboxylase (PEPC) in the MDH-PEPC-Linked assay (Supporting Information). Various concentrations (0−200 μM) of 5 (●) and 6 (○) were measured for their inhibitory effects to the activity of the linked assays in present (DDC-MDH-PEPC-Linked assay, panel A) or absent (MDH-PEPC-Linked assay, panel B) of hDDC. The data are presented as percent of control (DMSO, 100%). Error bars represent the standard deviation (n ≥ 3). Compound 5 at 200 μM and 6 at 100 or 200 μM were found to have substantial absorbance and interfere with the NADH readout at 340 nM (Supplementary Figure S8), and therefore we have made the background corrections for these inhibitors at the concentrations.

was also found to be present in the two most active inhibitors of hDDC, namely, 5 and 6 (Figure 1), indicating that the pbenzoquinone group was indispensable for strongly suppressing the activity. Furthermore, introduction of an additional carbonyl group, besides the two in the p-benzoquinone unit of 6, resulted in a 2-fold enhanced inhibitory activity for 5 (Figure 1B). Converting the diarylpropane unit of the dimer (5 or 6) to a cyclized flavan unit (7, 8, or 9) largely reduced the inhibition (Figure 1B). To address the selectivity of these dimeric diarylpropane-based inhibitors of hDDC, we tested whether they influence the activity of human cystathionine γlyase (CSE), a PLP-dependent enzyme and protein homologue of hDDC. The results showed that none of these inhibitors inhibited the activity of hCSE up to 200 μM, implying that this newly discovered moiety is selective for hDDC (selectivity >15fold; Figure 1B). To investigate the effects of the new selective inhibitors of dimeric diarylpropanes on the endogenous DDC activity, 5 and 6 as well as methyldopa, were incubated with human HepG2 cells, a human liver carcinoma cell line. Compound 5 dosedependently inhibited the intracellular activity of hDDC (Figure 3A) with an IC50 of ∼20 μM, and a similar effect was also observed for the levels of cellular dopamine (Figure 3B). In contrast, 6 did not affect the cellular activity of hDDC 899

dx.doi.org/10.1021/cb500009r | ACS Chem. Biol. 2014, 9, 897−903

ACS Chemical Biology

Letters

and dopamine content despite the fact that it also dosedependently inhibited hDDC in the in vitro purified enzyme assay (Figure 3A,B and Figure 2A). Additionally, both 5 and 6 did not affect the cell viability up to 100 μM as accessed by cytotoxicity assays (Figure 3C). Methyldopa at 1000 μM inhibited more than 80% of the content of dopamine and 60% cellular activity of hDDC but reduced only ∼20% dopamine or cellular activity of hDDC at 250 or 50 μM (Figure 3A and B). In our in vitro purified enzyme assay, methyldopa was also found to be a weaker inhibitor (IC50 of 1200 ± 3.9 μM; Figure 1A) for hDDC and also affected the activity of hCSE with an IC50 of 3000 ± 13.4 μM (selectivity ≤3-fold). Taken together, these data implicated that methyldopa is a much weaker inhibitor than 5 both in in vitro purified enzyme or cell-based hDDC assays. Thus, these results demonstrated that 5 is a bioactive inhibitor for hDDC with an improved potency and selectivity. In order to elucidate the action mode of 5 and 6, surface plasmon resonance experiments (SPR; Biacore; Supplementary Methods) were employed to study the interaction between these inhibitors and hDDC. SPR assay showed that 5 and 6 bind directly to hDDC with an equilibrium dissociation constant (KD) of 35 and 25 μM, respectively (Figure 4A). The KD values of 5 and 6 are 3- or 1.2-fold higher than the IC50 of 5 (11.5 ± 1.6 μM) and 6 (21.6 ± 2.7 μM), respectively, a discrepancy that may be attributed to the different measurement conditions used in our DDC and Biacore assay. To illustrate this interaction between this kind of inhibitor and hDDC, docking analysis was utilized to identify the potential binding site of 5 in the crystal structure of hDDC liganded with PLP (PDB code: 3RBF; ref 20). The result indicated that 5 occupied the L-DOPA binding region,8,20 and the pbenzoquinone, hydroxyl, or carbonyl groups in the unit I of 5 formed three hydrogen bonds (HB) with Tyr79, Gly102, or Ser149, respectively. Additional HBs were also observed between 5 with PLP cofactor (Figure 4B). In contrast, the inactive compound 12, which does not contain p-benzoquinone motif, cannot form any hydrogen bond with the active site of hDDC (Supplementary Figure S6A). Furthermore, 5 predominantly occupies the cavity of the active site of hDDC more than 12 (Supplementary Figure S6B) and interacts with known substrate binding residues Ile101, Gly102, Phe103, and Phe309.8,20 Among these hydrogen bonds, the p-benzoquinone group of 5 formed a HB with Tyr79, which was reported to be a key residue involved in the switch between apoenzyme (without PLP) or holoenzyme (with PLP), and this implies that the p-benzoquinone group of newly discovered dimeric diarylpropanes is likely the key chemical moiety for maintaining the inhibitory effects on hDDC. We also accessed the mode of action of 5 and 6 in the reversibility assay. In this assay, the sample pretreated with 5 and 6 retained only ∼75% and 60% activity compared to the DMSO-treated sample, respectively, but did not affected by the treatment of inactive compound 12 (Supplementary Figure S7 and Supplementary Methods). It indicated that the inhibition mechanism of these naturalproduct-based inhibitors 5 or 6 display a combination inhibitory mode of partial irreversible and reversible manner for inactivating of hDDC, which necessitates an additionally elaborate investigation. DDC inhibitors for the treatment of PD have to selectively inhibit the conversion of L-DOPA in peripheral tissues and impermeable to blood−brain barrier (BBB; ref 2). Compounds 5 and 6 were predicated to have large polar surfaces

Figure 3. Effects of dimeric diarylpropanes on the activity of endogenous hDDC in cells. (A) Dose-dependent effects of 5 and 6 as well as methyldopa on the activity of intracellular hDDC. Compound 5 or 6 at various concentrations (10, 20, or 50 μM) as well as methyldopa at 50, 250, or 1000 μM were incubated with HepG2 cells overnight, cells were then collected and lysed, and hDDC activity was measured as described in the Supporting Information. The hDDC activity in each sample was normalized by the amount of cellular protein. (B) Dose-dependent effects of 5 and 6 as well as methyldopa on the cellular contents of dopamine. Compound 5 or 6 or methyldopa at indicated concentrations was treated for 8 h with the HepG2 cells, which were preincubated with 100 μM L-DOPA for 6 h (Supporting Information). The cells were then collected, washed, and lysed, and the levels of cellular dopamine were measured by ELISA as described (Supporting Information). The content of dopamine in each sample was normalized by the amount of cellular protein. (C) Effects of 5 and 6 as well as methyldopa on cell viability. Compound 5 or 6 at various concentrations (5, 10, 20, 50, or 100 μM) as well as methyldopa at 250 or 1000 μM were incubated with HepG2 cells overnight, the cell medium was then collected, and the LDH activity was measured (Supporting Information). The data are presented as percent of control (DMSO, 100%). Error bars represent the standard deviation (n ≥ 2). 900

dx.doi.org/10.1021/cb500009r | ACS Chem. Biol. 2014, 9, 897−903

ACS Chemical Biology

Letters

Figure 4. Interaction between dimeric diarylpropanes and hDDC. (A) Surface plasmon resonance assay analysis of the binding of 5 or 6 to hDDC. Solutions with various concentrations of 5 or 6 (0−80 μM) were injected into the chamber with a hDDC-coated sensor chip. The change of response units is shown over time. Dissociation constants for 5 and 6 (KD) were calculated by using Biacore evaluation software (Supporting Information). (B) Stereo view of the binding model of 5 in the active site of hDDC. Residues surrounding the 5 within a distance of 3.5 Å are indicated in gray; PLP, in yellow; 5, in green and by default atom types (O, N); hydrogen bonds, green dotted lines.

(Supplementary Methods), i.e., 118.3 and 183.2 Å2, respectively, indicating that they are impervious to the BBB and have a potential to selectively inhibit the DDC in peripheral tissues.21 Taken into consideration that 5 was the only compound that could suppress the activity of DDC and dopamine levels in purified enzyme and cell -based assays, it indicates that 5 is an interesting lead compound on developing new drug candidates for PD. In summary, we reported the identification, isolation, and structure characterization of natural-product-based inhibitors for human DOPA decarboxylase from the ethanolic extract of E. glabra guided by a newly developed high-throughput assay. The discovery leads to the identification of eight new rare members of dimeric diarylpropanes. The natural product 5, a structural analogue of dopamine, was discovered for the first time from this species and displayed a unique bioactivity to selectively inhibit the human DOPA decarboxylase in in vitro purified enzyme and cell-based assays. Further analyses of structure−activity relationship and molecular modeling revealed that p-benzoquinone might be the crucial chemical moiety for binding and inhibiting human DOPA decarboxylase. The isolated and identified dimeric diarylpropane compounds as well as the extraction itself could provide a basis to develop pharmacologically useful dopamine analogues for treatment of dopamine-related diseases.

■

yldopa), bovine serum albumin (BSA), and NADH disodium salt trihydrate were purchased from Sangon. Pyridoxal 5′-phosphate hydrate (PLP) and 3,4-dihydroxy-L-phenylalanine (L-DOPA) were from Sigma, phosphoenolpyruvic acid monopotassium salt (PPPA) was from Alfa Aesar, and β-mercaptoethanol and porcine heart malate dehydrogenase (MDH) were from Amersco. Isolation and Characterization of Natural Products from Ethanolic Extract of E. glabra. The natural products were extracted by ethanol from E. glabra and purified by various methods of chromatography, and the chemical structures were then characterized by HR-ESIMS and 1D and 2D NMR spectroscopic methods. The detailed experimental methods for isolation and structural elucidation of these natural products as well as their spectroscopic data are listed in Supporting Information. Cloning, Overexpression, and Purification of hDDC, E. coli Phosphoenolpyruvate Carboxylase, and Human CSE. hDDC and E. coli phosphoenolpyruvate carboxylase (PEPC) were subcloned into pET28b to generate N- or N- and C-terminal 6× His-tagged protein; the fusion proteins were purified by a Ni2+-agarose column (refs 22−24 and Supporting Information). Similarly, hCSE were subcloned into glutathione S-transferase (GST)-fusion expression vector pGEX-KG, overexpressed, and affinity-purified by a GSTagarose column (refs 25−27; for details, see Supporting Information). DDC Enzyme Assay. To detect the decarboxylase activity of hDDC, we adapted a highly sensitive and efficient method that had been used for measuring the decarboxylase activity of ornithine decarboxylase.11,12 The decarboxylase activity of hDDC was then linked to the consumption of NADH by using PEPC and MDH as coupling enzymes. For details, see Supporting Information. CSE Enzyme Assay. The enzyme reaction of hCSE was carried out according to the conditions as described (Supporting Information, ref 28).

METHODS

Chemicals and Regents. 5,5′-Dithiobis(2-nitrobenzoic acid; DTNB), L-cysteine, 3-(3,4-dihydroxphenyl)-2-methyl-L-alanine (meth901

dx.doi.org/10.1021/cb500009r | ACS Chem. Biol. 2014, 9, 897−903

ACS Chemical Biology

Letters

Cell Culture. Human HepG2 cells were maintained in MEM (Life Technologies) supplemented with 1× nonessential amino acids (NEAA, Life Technologies), 10% fetal bovine serum (FBS; Life Technologies), and 1% (w/v) penicillin and streptomycin (Life Technologies) in a humidified 5% CO2 atmosphere at 37 °C. Cellular Activity of hDDC. To measure the cellular hDDC activity, HepG2 cells were cultured in a poly-D-lysine coated six-well plate with at a density of 5 × 105 per well for 1 day and then treated in the absence or presence of compounds of interest for 16 h. Cellular activities of hDDC were then measured according to the procedure described in Supporting Information. Molecular Modeling. Compound 5 or 12 was docked by molecular modeling into the active site of the crystal structure of the complex of hDDC and PLP cofactor (PDB code: 3RBF; ref 20). The 3D structures of 5 and 12 were drawn and optimized by energy minimization (Discovery Studio version 3.5; Accelrys). The optimized structures were docked into the active site of hDDC by using the Discovery Studio/CDOCKER module, a CHARMM-based molecular dynamics simulated-annealing program.29

■

H., and Halliday, G. (2010) Missing pieces in the Parkinson’ s disease puzzle. Nat. Med. 16, 653−661. (5) Abdel-Salam, O. M. (2008) Drugs used to treat Parkinson’s disease, present status and future directions. CNS Neurol. Disord.: Drug Targets 7, 321−342. (6) Amadasi, A., Bertoldi, M., Contestabile, R., Bettati, S., Cellini, B., di Salvo, M. L., Borri-Voltattorni, C., Bossa, F., and Mozzarelli, A. (2007) Pyridoxal 5′-phosphate enzymes as targets for therapeutic agents. Curr. Med. Chem. 14, 1291−1324. (7) Wu, F., Christen, P., and Gehring, H. (2011) A novel approach to inhibit intracellular vitamin B6-dependent enzymes: proof of principle with human and plasmodium ornithine decarboxylase and human histidine decarboxylase. FASEB J. 25, 2109−2122. (8) Bertoldi, M., Gonsalvi, M., and Voltattorni, C. B. (2001) Green tea polyphenols: novel irreversible inhibitors of dopa decarboxylase. Biochem. Biophys. Res. Commun. 284, 90−93. (9) Daidone, F., Montioli, R., Paiardini, A., Cellini, B., Macchiarulo, A., Giardina, G., Bossa, F., and Borri Voltattorni, C. (2012) Identification by virtual screening and in vitro testing of human DOPA decarboxylase inhibitors. PLoS One 7, e31610. (10) Komori, H., Nitta, Y., Ueno, H., and Higuchi, Y. (2012) Structural study reveals that Ser-354 determines substrate specificity on human histidine decarboxylase. J. Biol. Chem. 287, 29175−29183. (11) Smithson, D. C., Lee, J., Shelat, A. A., Phillips, M. A., and Guy, R. K. (2010) Discovery of potent and selective inhibitors of Trypanosoma brucei ornithine decarboxylase. J. Biol. Chem. 285, 16771−16781. (12) Smithson, D. C., Shelat, A. A., Baldwin, J., Phillips, M. A., and Guy, R. K. (2010) Optimization of a non-radioactive high-throughput assay for decarboxylase enzymes. Assay Drug Dev. Technol. 8, 175−185. (13) Allen, G. F., Neergheen, V., Oppenheim, M., Fitzgerald, J. C., Footitt, E., Hyland, K., Clayton, P. T., Land, J. M., and Heales, S. J. (2010) Pyridoxal 5'-phosphate deficiency causes a loss of aromatic Lamino acid decarboxylase in patients and human neuroblastoma cells, implications for aromatic L-amino acid decarboxylase and vitamin B(6) deficiency states. Neurochem. 114, 87−96. (14) Bender, D. A., and Coulson, W. F. (1977) Aromatic amino acid decarboxylase: pH-dependence of substrates and inhibitors. Biochem. Soc. Trans. 5, 1353−1356. (15) Wu, M. M., Wang, L. Q., Hua, Y., Chen, Y. G., Wang, Y. Y., Li, X. Y., Li, Y., Li, T., Yang, X. Y., and Tang, Z. R. (2011) New chalcone and dimeric chalcones with 1,4-p-benzoquinone residue from Combretum yunnanense. Planta Med. 77, 481−484. (16) Moosophon, P., Kanokmedhakul, S., and Kanokmedhakul, K. (2011) Diarylpropanes and an arylpropyl quinone from Combretum griffithii. J. Nat. Prod. 74, 2216−2218. (17) Filho, R. B., Dias D, P. P., and Gottlieb, O. R. (1980) Tetronic acid and diarylpropanes from Iryanthera elliptica. Phytochemistry 19, 455−459. (18) Ramadan, M. A., Kamel, M. S., Ohtani, K., Kasai, R., and Yamasaki, K. (2000) Minor phenolics from Crinum bulbispermum bulbs. Phytochemistry 54, 891−896. (19) Garo, E., Maillard, M., Antus, S., Mavi, S., and Hostettmann, K. (1996) Five flavans from Mariscus psilostachys. Phytochemistry 43, 1265−1269. (20) Giardina, G., Montioli, R., Gianni, S., Cellini, B., Paiardini, A., Voltattorni, C. B., and Cutruzzola, F. (2011) Open conformation of human DOPA decarboxylase reveals the mechanism of PLP addition to Group II decarboxylases. Proc. Natl. Acad. Sci. U S A 108, 20514− 20519. (21) Kelder, J., Grootenhuis, P. D., Bayada, D. M., Delbressine, L. P., and Ploemen, J. P. (1999) Polar molecular surface as a dominating determinant for oral absorption and brain penetration of drugs. Pharm. Res. 16, 1514−1519. (22) Bertoldi, M., Moore, P. S., Maras, B., Dominici, P., and Voltattorni, C. B. (1996) Mechanism-based inactivation of dopa decarboxylase by serotonin. J. Biol. Chem. 271, 23954−23959. (23) Montioli, R., Cellini, B., and Borri Voltattorni, C. (2011) Molecular insights into the pathogenicity of variants associated with

ASSOCIATED CONTENT

S Supporting Information *

Supplemental tables, figures, methods, isolation and spectroscopic data, structural elucidations for compounds 7−9. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Author Contributions ⊥

These authors contributed equally to this work.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by program NCET Foundation, National Basic Research Program of China (2012CB822103), NSFC (81230090, 31270853, 81102377 and 81102778), Global Research Network for Medicinal Plants (GRNMP) and King Saud University, Shanghai Leading Academic Discipline Project (B906), FP7-PEOPLE-IRSES-2008 (TCMCANCER Project 230232), Key laboratory of drug research for special environments, PLA, Shanghai Engineering Research Center for the Preparation of Bioactive Natural Products (10DZ2251300), the Scientific Foundation of Shanghai China (10DZ1971700, 12401900501), National Major Project of China (2011ZX09307-002-03 and 2011ZX09102-006-02), National Key Technology R&D Program of China (2012BAI29B06), and Shanghai Pujiang Program (12PJ1405000).

■

REFERENCES

(1) Morris, M. E., Martin, C. L., and Schenkman, M. L. (2009) Striding out with Parkinson disease: evidence-based physical therapy for gait disorders. Phys. Ther. 90, 280−288. (2) Hauser, R. A. (2009) Levodopa: past, present, and future. Eur. Neurol. 62, 1−8. (3) Shulman, J. M., De Jager, P. L., and Feany, M. B. (2011) Parkinson’s disease: genetics and pathogenesis. Annu. Rev. Pathol. 6, 193−222. (4) Obeso, J. A., Rodriguez-Oroz, M. C., Goetz, C. G., Marin, C., Kordower, J. H., Rodriguez, M., Hirsch, E. C., Farrer, M., Schapira, A. 902

dx.doi.org/10.1021/cb500009r | ACS Chem. Biol. 2014, 9, 897−903

ACS Chemical Biology

Letters

the aromatic amino acid decarboxylase deficiency. J. Inherit. Metab. Dis. 34, 1213−1224. (24) Chen, L. M., Omiya, T., Hata, S., and Izui, K. (2002) Molecular characterization of a phosphoenolpyruvate carboxylase from a thermophilic cyanobacterium, Synechococcus vulcanus with unusual allosteric properties. Plant Cell Physiol. 43, 159−169. (25) Frank, N., Kent, J. O., Meier, M., and Kraus, J. P. (2008) Purification and characterization of the wild type and truncated human cystathionine beta-synthase enzymes expressed in E. coli. Arch. Biochem. Biophys. 470, 64−72. (26) Oliveriusova, J., Kery, V., Maclean, K. N., and Kraus, J. P. (2002) Deletion mutagenesis of human cystathionine beta-synthase. Impact on activity, oligomeric status, and S-adenosylmethionine regulation. J. Biol. Chem. 277, 48386−48394. (27) Janosik, M., Meier, M., Kery, V., Oliveriusova, J., Burkhard, P., and Kraus, J. P. (2001) Crystallization and preliminary X-ray diffraction analysis of the active core of human recombinant cystathionine beta-synthase: an enzyme involved in vascular disease. Acta Crystallogr. D Biol. Crystallogr. 57, 289−291. (28) Zhou, Y., Yu, J., Lei, X., Wu, J., Niu, Q., Zhang, Y., Liu, H., Christen, P., Gehring, H., and Wu, F. (2013) High-throughput tandem-microwell assay identifies inhibitors of the hydrogen sulfide signaling pathway. Chem. Commun. (Camb) 49, 11782−11784. (29) Wu, G., Robertson, D. H., Brooks, C. L., 3rd, and Vieth, M. (2003) Detailed analysis of grid-based molecular docking: A case study of CDOCKER-A CHARMm-based MD docking algorithm. J. Comput. Chem. 24, 1549−1562.

903

dx.doi.org/10.1021/cb500009r | ACS Chem. Biol. 2014, 9, 897−903