Subscriber access provided by UNIV OF NEWCASTLE
Article
Design, Synthesis, and Evaluation of New Selective NM23-H2 Binders as c-MYC Transcription Inhibitors via Disruption of the NM23-H2/G-Quadruplex Interaction Yu-Qing Wang, Zhou-Li Huang, Shuo-Bin Chen, Chen-Xi Wang, Chan Shan, QiKun Yin, Tian-Miao Ou, Ding Li, Lian-Quan Gu, Jia-Heng Tan, and Zhi-Shu Huang J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b00421 • Publication Date (Web): 17 Jul 2017 Downloaded from http://pubs.acs.org on July 17, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 67
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of Medicinal Chemistry
Design, Synthesis, and Evaluation of New Selective NM23-H2 Binders as c-MYC Transcription Inhibitors via Disruption of the NM23-H2/G-Quadruplex Interaction
Yu-Qing Wang, Zhou-Li Huang, Shuo-Bin Chen, Chen-Xi Wang, Chan Shan, Qi-Kun Yin, Tian-Miao Ou, Ding Li, Lian-Quan Gu, Jia-Heng Tan*, and Zhi-Shu Huang*
School of Pharmaceutical Sciences, Sun Yat-sen University, Guangzhou 510006, People′s Republic of China
ACS Paragon Plus Environment
Journal of Medicinal Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ABSTRACT c-MYC is one of the important human proto-oncogenes, and transcriptional factor NM23-H2 can activate c-MYC transcription by recognizing the G-quadruplex in the promoter of the gene. Small molecules that inhibit c-MYC transcription by disrupting the NM23-H2/G-quadruplex interaction might be a promising strategy for developing selective anticancer agents. In recent studies, we developed a series of isaindigotone derivatives, which can bind to G-quadruplex and NM23-H2, thus down-regulating c-MYC (J. Med. Chem. 2017, 60, 1292-1308). Herein, a series of novel isaindigotone derivatives were designed, synthesized and screened for NM23-H2 selective binding ligands. Among them, compound 37 showed a high specific binding affinity to NM23-H2, effectively disrupting the interaction of NM23-H2 with G-quadruplex, and it strongly down-regulated c-MYC transcription. Furthermore, 37 induced cell cycle arrest and apoptosis, and it exhibited good tumor growth inhibition in a mouse xenograft model. This work provides a new strategy to modulate c-MYC transcription for the development of selective anticancer drugs.
KEYWORDS: isaindigotone derivatives; NM23-H2 binder; c-MYC gene; transcriptional regulation; anticancer
ACS Paragon Plus Environment
Page 2 of 67
Page 3 of 67
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of Medicinal Chemistry
INTRODUCTION
Oncogenes drive tumorigenesis have been proposed as promising therapeutic targets for cancers treating.1 The human c-MYC gene is one of the most well-known oncogenes; it encodes a multifunctional transcription factor that regulates the expression of genes involved in cell growth, proliferation, differentiation and apoptosis.2, 3 The aberrant overexpression of c-MYC is associated with various human cancers, and numerous murine cancer models are c-MYC-dependent.4 Anti-c-MYC therapeutic strategies involve multiple routine approaches, including the prevention of c-MYC binding to its partner proteins or target genes, and c-MYC gene transcription inhibition.5-8 Because c-MYC does not have any cavities which small molecules can easily bind into, the design of direct inhibitors is very complicated.9 The development of effective therapeutic strategies for the inhibition of c-MYC gene transcription has been challenging. Additionally, the inhibition of the bromodomain and extra-terminal (BET) family BRD4 protein can suppress c-MYC transcription and lead to tumor inhibition pre-clinically in vivo.6,10 This finding indicates that down-regulating c-MYC transcription is a feasible strategy to expand cancer therapy. The NM23-H2 protein comes from the human NM23 family, and NM23-H2 has been known to function as both a nucleoside diphosphate kinase (NDPK) and transcriptional factor.11-13 NM23-H2 has been reported to play important roles in many physiological processes, such as gene transcription, DNA repair, oncogenesis, and cellular proliferation.14-17 In fact, aberrant expression of NM23-H2 have been observed in many tumors, such as hepatocellular carcinoma, chronic myeloid leukemia, and breast cancer,16,
18, 19
indictaing NM23-H2 as an ideal target for cancer treatment. Notably,
NM23-H2 has been identified as a transcription factor of the c-MYC oncogene, which activates
ACS Paragon Plus Environment
Journal of Medicinal Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 67
c-MYC transcription by recognizing and unfolding the structural and functional element (G-quadruplex) located in the promoter region of the gene.7,
20-22
Hence, disruption of the
NM23-H2/c-MYC G-quadruplex interaction may be a feasible strategy for down-regulating c-MYC transcription. Several studies have been performed to develop compounds that disrupt the interaction, and most of the compounds are G-quadruplex ligands.20, 23 G-quadruplex structural motifs can form anywhere in the genome in G-quadruplex-forming sequences, whereas single-stranded G-rich DNA is exposed during replication, transcription or recombination. 24, 25 The G-quadruplex ligands may lack targeting specificity and have side effects in oncotherapy. In recent studies, we developed isaindigotone derivative 1 (SYSU-ID-01) and 2 (SYSU-ID-19d) with two amino side chains (Figure 1), which bind to both G-quadruplex DNA and the NM23-H2 protein26 and can also down-regulate c-MYC transcription by disrupting the NM23-H2/G-quadruplex interaction. Therefore, further design and synthesis of highly specific NM23-H2 binding ligands to enhance the target selectivity will supply new and important information for c-MYC transcription inhibition strategies.
Figure 1. Structures of compound 1, 2, and the new isaindigotone derivatives
ACS Paragon Plus Environment
Page 5 of 67
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of Medicinal Chemistry
Table 1. Structures of the new synthesized isaindigotone derivatives (12-56) Compd.
n
12
R1
R2
R3
R4
R5
Compd.
n
3
-H
-H
-F
-H
35
13
3
-H
-H
-CF3
-H
14
3
-H
-H
15
3
-H
-H
16
3
-H
-H
R2
R3
R4
R5
3
-H
-OMe
-OMe
-H
36
3
-H
-OMe
-OMe
-OMe
-H
37
3
-OH
-H
-OCH2Ph
-H
-H
38
3
-H
-OH
-OCH2Ph
-H
-H
39
3
-H
-H
-F
-H
17
3
-H
-H
-H
40
3
-H
-H
-CF3
-H
18
3
-H
-H
-N(CH3)2
-H
41
3
-H
-H
-H
19
3
-H
-H
-OMe
-H
42
3
-H
-H
-H
20
3
-H
-H
-OCH2Ph
-H
43
3
-H
-H
-CH(CH3)2
-H
21
3
-OH
-H
-N(CH3)2
-H
44
3
-H
-H
-N(CH3)2
-H
22
3
-OMe
-H
-H
-H
45
3
-H
-H
-CHO
-H
23
3
H
-OMe
-OMe
H
46
3
-H
-H
-OMe
-H
24
3
H
-OMe
-OMe
-OMe
47
3
-H
-H
-OCH2Ph
-H
25
3
-OH
-H
-OCH2Ph
-H
48
3
H
-OMe
-OMe
-OMe
26
3
-H
-OH
-OCH2Ph
-H
49
3
-OH
-H
-OCH2Ph
-H
27
3
-H
-H
-H
50
2
-H
-H
-CHO
-H
28
3
-H
-H
-H
51
2
-H
-H
- OCH2C≡CH
-H
29
3
-H
-H
-H
52
2
-OH
-H
-N(CH3)2
-H
30
3
-H
-H
-OPh
-H
53
2
-OH
H
-OCH2Ph
H
31
3
-H
-H
-OCH(CH3)2
-H
54
3
-OH
H
-OCH2Ph
H
32
3
-H
-H
-OCH2C≡CH
-H
55
2
-OH
H
-OCH2Ph
H
33
3
-H
-OH
-OH
-H
56
0
-OH
H
-OCH2Ph
H
34
3
-H
-OMe
-OH
-OMe
-CH(CH3)2
R1
In this study, to explore novel NM23-H2-specific ligands, a series of new isaindigotone derivatives were designed based on the structures of our previous isaindigotone derivatives with two
ACS Paragon Plus Environment
Journal of Medicinal Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 67
amino side chains (Figure 1). First, we retained the five-carbon aliphatic ring fused to the pyrrolo[2,1-b]quinazoline moiety because our previous research confirmed that its derivatives exhibit higher binding affinities with the NM23-H2 protein than compounds fusing six-carbon aliphatic ring.26 Second, introducing two amino side chains to the chromophore has been proven to be an effective way to obtain high G-quadruplex DNA binding and stabilizing potency.27, 28 To remove the G-quadruplex ligand properties of the derivatives, only one positive amino side chain at the 6-position was introduced for the requirement of solubility. Meanwhile, we replaced another amino side chain at the 4′-position of the styrene ring with a variety of substituents, such as alkyl, hydroxyl, alkoxy, alkamino, fluoro, trifluoromethyl, and so on. Therefore, a series of new (E)-6-amino-3-benzylidene-7-fluoro-2,3-dihydropyrrolo[2,1-b]quinazoline-9(1H)-one
derivatives
(12-56) were synthesized based on the structural modification described above (Figure 1 and Table 1). Their biological activities and structure–activity relationships were analyzed. After a screening assay, the most outstanding compound was further investigated, including activity of the control of c-MYC gene transcription, cell cycle arrest and apoptosis-inducing activities in SiHa cells, and inhibition of tumor growth in a mouse xenograft model of cervical squamous cancer.
RESULTS AND DISCUSSION
Chemistry. The facilely synthetic pathway for the isaindigotone derivatives is shown in Scheme 1. Intermediate 4 was obtained by the reaction of 2-amino-4,5-difluorobenzoic acid (3) and pyrrolidin-2-one in the presence of POCl3.29, 30 The next step was the nucleophilic substitution of the 6-position fluorine atom with different N-nucleophiles of the primary amines to obtain intermediates
ACS Paragon Plus Environment
Page 7 of 67
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of Medicinal Chemistry
5-11. Treatment of 5-11 with different aldehydes via the Knoevenagel reaction gave the target compounds 12-56.31 The structure of the absolute configuration of compound 37 was finally identified using single crystal X-ray crystallographic analysis via anomalous scattering of Cu Kα radiation (Figure S1).
Scheme 1. Synthesis of the isaindigotone derivatives. Reagents and conditions: (a) pyrrolidin-2-one, POCl3, reflux, 24 h, 84%; (b) R1(CH2)nNH2, 100 °C, 24 h, 52-78%; (c) aldehydes, DMF, TMSCl, 100 °C, 24-48 h, 40-88%.
Binding Affinities of the Derivatives for NM23-H2 Protein and SAR Study. To investigate the binding affinities of the synthesized compounds for the NM23-H2 protein, surface plasmon resonance (SPR) experiments were performed (Figure S2A). The dissociation constants (KD) were achieved from the equilibrium fitting mode. As shown in Table 2, most of the derivatives bound to NM23-H2 protein showed moderate to strong binding affinity, and the abilities of some ligands were superior than 1 and 2. The structure–activity relationship was further explored as described below.
Table 2. The binding affinity of the derivatives to NM23-H2 protein. The equilibrium dissociation constants (KD) were determined with the SPR assay and MST assay. Compd.
KD/µM with SPRa
KD/µM with MSTb
Compd.
KD/µM with SPRa
KD/µM with MSTb
12
‒c
61.6
36
16.8
18.4
13
‒c
59.4
37
3.1
2.0
14
‒c
55.3
38
13.2
16.1
ACS Paragon Plus Environment
Journal of Medicinal Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
a
Page 8 of 67
15
‒c
‒d
39
‒c
44.2
16
‒c
‒d
40
‒c
‒d
17
13.8
20.8
41
‒c
‒d
18
‒c
48.5
42
‒c
‒d
19
15.4
23.5
43
‒c
‒d
20
14.6
20.6
44
‒c
‒d
21
13.9
11.7
45
38.0
‒d
22
‒c
72.8
46
28.4
25.6
23
17.1
25.1
47
18.4
22.3
24
16.3
28.2
48
18.5
23.6
25
4.6
3.4
49
14.3
16.9
26
16.7
25.8
50
‒c
68.0
27
8.5
10.8
51
‒c
27.5
28
10.1
18.9
52
‒c
45.7
29
20.6
21.0
53
8.1
6.4
30
12.4
16.3
54
11.5
10.8
31
9.6
15.4
55
12.2
7.6
32
25.6
8.9
56
‒c
56.0
33
9.2
10.7
1
10.4
24.6
34
13.3
16.5
2
19.8
18.0
35
15.8
19.2
The KD values determined using the SPR assay were achieved from the equilibrium fitting mode of the Analysis module in the
ProteOn XPR36 software. b
The KD values determined using the MST assay were achieved in the Analysis module of the NT Analysis software using the Hill
model. c
No significant binding was found at a concentration up to 40 µM ligand.
d
The KD values of the compound was not obtained by fitting.
First, we investigated the effect of the single substituent of R4 on the styrene portion, which
ACS Paragon Plus Environment
Page 9 of 67
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of Medicinal Chemistry
includes a series of analogs that were designed and synthesized with different R4 substituents. As shown in Table 2, in the series of 6-(3-diethylaminopropyl)amino substituted compounds (12-20) and in the series of 6-(3-morpholinopropyl)amino substituted compounds (39-47), the binding potency of the derivatives is very sensitive to the type of R4 substituent. Compounds (19-20 and 46-47) with an alkoxy group (4′-OCH3 or 4′-OCH2Ph) obviously exhibited better activity than other substituents (alkamino, alkyl, or fluoro groups) except compound 17, and compounds (20 and 47) with the 4′-benzyloxy group exhibited a slightly stronger binding affinity than that of compounds (19 and 46) with the 4′-methoxy group. Similarly, the 6-(3-dimethylaminopropyl)amino substituted compounds (28-32) with different types of alkoxy groups also exhibited considerable activity. All of the above results
indicated
that
introducing
an
alkoxy
group
at
the
4′-position
of
the
3-benzylidene-7-fluoro-2,3-dihydropyrrolo[2,1-b]- quinazolin-9(1H)-one moiety was important for increasing the binding affinity. We also synthesized a series of compounds with two or more alkoxy groups at the 2′-, 3′-, 4′-, or 5′-position of the phenyl ring. As shown in Table 2, upon comparison of the activities of 2′, 4′-disubstituted derivative 25 and 3′,4′-disubstituted derivative 26 with 4′-substituted derivative 20, we found that a hydroxy group introducing at the 2′-position seems to be more beneficial for increasing the activity of the compounds than introduction at the 3′-position. These results were also found for other series of compounds, such as for compounds 37 (2′-OH-4′-OCH2Ph, KD = 3.1 µM) and 38 (3′-OH-4′-OCH2Ph, KD = 13.2 µM) and for compounds 49 (2′-OH-4′-OCH2Ph, KD = 14.3 µM) to 47 (4′-OCH2Ph, KD = 18.4 µM). Amino side chain substituents at the 6-position of 2,3-dihydropyrrolo[2,1-b]quinazolin9(1H)-one also have substantial effects on the binding activity. A series of analogs (25, 37, 49, 53-55) that containing the 4-(benzyloxy)-2-hydroxybenzylidene group at the 3-position of the moiety was
ACS Paragon Plus Environment
Journal of Medicinal Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
analyzed. As shown in Table 2, the introduction of an open-ring basic terminal group, i.e., a dimethylamino group (compound 37) or diethylamino group (compound 25), exerted a remarkable improvement in the binding activity with KD values of 3.1 and 4.6 µM, respectively, compared with the corresponding closed-ring basic terminal group containing compounds 49 (14.3 µM) and 54 (11.5 µM). We also observed that the length of the amino side chain at the 6-position had an effect on the activity. For example, for compounds that had the same terminal side chain, the binding affinity of 37 (n = 3) was better than that of 53 (n = 2), and the binding affinity of 49 (n = 3) was better than that of 56 (n = 0). These results indicated that the length and terminal basic groups of the side chain at the 6-position have important impacts on the compounds regarding their binding activities. In conclusion, the new isaindigotone derivatives synthesized via structural modification were confirmed as NM23-H2 protein binders, and some compounds exhibited a significant binding affinity to NM23-H2 with a KD value below 10 µM. Compound 37 exhibited the best binding affinity (3.1 µM, Figure S2) and was stronger than the previous compounds 1 (10.4 µM) and 2 (19.8 µM). The microscale thermophoresis (MST) assay was also performed to verify and validate the binding activity of the compounds (Table 2 and Figure S3), and the structure–activity relationship was consistent with the SPR assay. To gain more details on the interaction mode of compound 37 with NM23-H2 protein, we further performed molecular docking simulations of compound 37 with NM23-H2 (PDB code: 3BBB) using Schrodinger Maestro 9.3. As shown in Figure 2, compound 37 was well-fitted into the narrow, slightly curved pocket that the dinucleotide possessed as previously reported.7 Compound 37 undergoes hydrogen bonding with residues in the channel of the protein active site (Gly113 and Asp121), hydrophobic interactions with His118 and Lys66, and π-π stacking with Phe60 and Tyr67.
ACS Paragon Plus Environment
Page 10 of 67
Page 11 of 67
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of Medicinal Chemistry
To further validate the binding modes, five single amino acid mutant NM23-H2 proteins (D54A, F60A, G113F, H118F, and D121A) were further investigated.7 The SPR experiment and MST experiment results showed that compound 37 could still bind to the D54A, G113F, and H118F mutants with a reduced binding affinity compared to the wild-type NM23-H2 while losing binding affinity to the F60A and D121A mutant (Figure 2C and Figure S2). These results indicated that compound 37 binds to the NM23-H2 active pocket, and Phe60 as well as Asp121 showed the most important roles.
Figure 2. Predicted binding mode of compound 37 with NM23-H2. (A, B) Binding mode of compound 37 bound to one subunit of NM23-H2 (PDB code: 3BBB). (C) Binding affinity of compound 37 to wild-type and mutant NM23-H2 determined by SPR and MST. “-a” in the table means the KD value was not obtained by fitting.
In addition, we also examined the binding affinities of G-quadruplex DNA to wild-type and mutant NM23-H2 proteins. As shown in Table S1, the KD value of pu22 binding to the wild-type NM23-H2 was 0.45 µM, while the binding affinity reduced for the D54A and H118F mutant. For F60A, G113F and D121A mutants, pu22 lost binding affinity. To some extent, these results were
ACS Paragon Plus Environment
Journal of Medicinal Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
consistent with the behavior of compound 37 binding to the wild-type and mutant NM23-H2 proteins, indicating 37 might have the potential to occupy some binding sites of pu22 towards NM23-H2 and disrupt the NM23-H2/G-quadruplex interaction accordingly.
Stabilizing and Binding Activities of Derivatives on the c-MYC G-Quadruplex. We evaluated the binding affinities of the synthesized isaindigotone derivatives to NM23-H2 protein, but we are also interested in the stabilizing and binding activities of the compounds to G-quadruplex DNA. A fluorescence resonance energy transfer (FRET) assay was performed to evaluate the stabilizing activity of the compounds on c-MYC G-quadruplex DNA (Table S2), and the results (Table S3) showed that most of the compounds exhibited a weak stabilization effect on the G-quadruplex, while compound 37 had no meaningful stabilizing activity. Additionally, a surface plasmon resonance (SPR) assay was performed to investigate the binding activity and selectivity of the compounds to the c-MYC G-quadruplex (Pu22), the telomeric G-quadruplex (Htg21), and duplex DNA (hairpin 18). As shown in Table S4, most of the compounds similarly exhibited a weak activity, and compound 37 had an unmeasurable binding affinity to the c-MYC G-quadruplex. Other tested DNAs at a concentration of 40 µM of compound indicated that this series of isaindigotone derivatives had no or weak stabilizing and binding activities to G-quadruplex DNA.
Disrupting Activity of the Derivatives to the NM23-H2/c-MYC G-Quadruplex Interaction. Since the major objective of this study was to develop disrupting ligands for the NM23-H2/c-MYC G-quadruplex interaction (Figure S4), the disrupting activities of newly synthesized NM23-H2 binders were identified using enzyme-linked immunosorbent assay (ELISA)
ACS Paragon Plus Environment
Page 12 of 67
Page 13 of 67
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of Medicinal Chemistry
methods. The affinity of NM23-H2 bound to G-quadruplex (KD = 0.38 ± 0.047 µM) was similar to a previous report.7 Then, the KD values of NM23-H2/ G-quadruplex interaction were tested in the presence of the compounds (Table S5). We visualized the results as shown in Figure 3, where the KD ratio (KD ratio = KDwith compound/KDwithout compound) is the y-axis of the column graph. A higher ratio represents a stronger disrupting ability of the compounds to the NM23-H2/c-MYC G-quadruplex interaction.
Figure 3. Intervening effects of the isaindigotone derivatives on the interaction of NM23-H2 with the c-MYC G-quadruplex determined by ELISA assays. In total, 100 nM 5′-biotin-labelled c-MYC Pu22 was annealed in 200 µL of Tris-HCl buffer (10 mM, pH 7.4) containing 100 mM KCl. The absorbance reading was performed at 450 nm, and the KD values were fitted and calculated from the original data in Origin 8. Compound 1 was used as the reference compound. The KD of NM23-H2-G-quadruplex binding in the presence of the compounds (KD with ligand) was normalized by the KD of the NM23-H2-DNA interaction without the compound (KD without ligand) and columned. The data were obtained from three individual experiments and are expressed as means ± SD.
Compared with compound 1 (KD ratio = 2.3), some compounds maintained (17, 18, 21, 36, and 47) (KD ratio in the range of 2.1~2.6) or increased (20, 25-27, 31, 37, 38, and 53-55) (KD ratio≥3.0) the disrupting potency. Meanwhile, most of the compounds (20, 25, 27, 31, 37, 38 and 53-55) with a
ACS Paragon Plus Environment
Journal of Medicinal Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
higher KD ratio had a stronger binding with NM23-H2 protein (KD
