Designing multi-targeted agents: An emerging anticancer drug discovery paradigm

Designing multi-targeted agents: An emerging anticancer drug discovery paradigm

Accepted Manuscript Designing multi-targeted agents: An emerging anticancer drug discovery paradigm Rong-geng Fu, Yuan Sun, Wen-bing Sheng, Duan-fang ...

950KB Sizes 0 Downloads 16 Views

Accepted Manuscript Designing multi-targeted agents: An emerging anticancer drug discovery paradigm Rong-geng Fu, Yuan Sun, Wen-bing Sheng, Duan-fang Liao PII:

S0223-5234(17)30377-X

DOI:

10.1016/j.ejmech.2017.05.016

Reference:

EJMECH 9446

To appear in:

European Journal of Medicinal Chemistry

Received Date: 2 March 2017 Revised Date:

30 April 2017

Accepted Date: 4 May 2017

Please cite this article as: R.-g. Fu, Y. Sun, W.-b. Sheng, D.-f. Liao, Designing multi-targeted agents: An emerging anticancer drug discovery paradigm, European Journal of Medicinal Chemistry (2017), doi: 10.1016/j.ejmech.2017.05.016. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT P2

P1

P1

P2

P2

P1

linker

P1 P2

conjugated-pharmacophore

cleavable linker conjugated-pharmacophore

RI PT

linker

fused-pharmacophore

merged-pharmacophore

Increasing degree of overlap of two pharmacophores, P1 and P2

AC C

EP

TE D

M AN U

SC

Strategies in the designing of multi-targeted agents

ACCEPTED MANUSCRIPT

Designing multi-targeted agents: An emerging anticancer drug discovery paradigm

M AN U

SC

*Corresponding author: Rong-geng Fu Tel: +86-731-88458227; E-mail: [email protected] Duan-fang Liao Tel: +86-731-88458227; E-mail: [email protected]

RI PT

Rong-geng Fu*a,b, Yuan Sunc, Wen-bing Shenga, Duan-fang Liao*a,b a School of Pharmacy, Hunan University of Chinese Medicine, Changsha, Hunan, 410208, China b Sino-Luxemburg TCM Research Center, Hunan University of Chinese Medicine, Changsha, Hunan, 410208, China c Department of Chemistry and Biochemistry, The Ohio State University, 43210, Columbus, Ohio, USA

AC C

EP

TE D

ABSTRACT The dominant paradigm in drug discovery is to design ligands with maximum selectivity to act on individual drug targets. With the target-based approach, many new chemical entities have been discovered, developed, and further approved as drugs. However, there are a large number of complex diseases such as cancer that cannot be effectively treated or cured only with one medicine to modulate the biological function of a single target. As simultaneous intervention of two (or multiple) cancer progression relevant targets has shown improved therapeutic efficacy, the innovation of multi-targeted drugs has become a promising and prevailing research topic and numerous multi-targeted anticancer agents are currently at various developmental stages. However, most multi-pharmacophore scaffolds are usually discovered by serendipity or screening, while rational design by combining existing pharmacophore scaffolds remains an enormous challenge. In this review, four types of multi-pharmacophore modes are discussed, and the examples from literature will be used to introduce attractive lead compounds with the capability of simultaneously interfering with different enzyme or signaling pathway of cancer progression, which will reveal the trends and insights to help the design of the next generation multi-targeted anticancer agents. Keywords: multi-targeted agents, anticancer therapy, multi-pharmacophore modes, pharmacophore combination, drug design 1. Introduction Towards the end of the 20th century, most drugs have emerged from the current ‘one molecule - one target - one disease’ philosophy, and it will undoubtedly remain dominant in medicinal chemistry research for many years to come. The main driver behind this strategy is to discover small molecules that are able to modulate the biological function of a single drug target which is fully responsible for a certain disease whilst reducing the risk of off-target related side effects. However, there is now a growing recognition that drugs designed to act on individual

1

ACCEPTED MANUSCRIPT molecular targets are usually inadequate for multigenic diseases such as cancer. [1] The initiation and progression of cancer depended on more than one receptor or signaling

pathway, suggesting that multi-targeted therapeutics may be advantageous over mono-therapies.

RI PT

For example, a recent study by Stratikopoulos and co-workers [2] illustrates that targeted inhibition of PI3K may not provide a long-lasting therapeutic effect and is often followed by the development of resistance to the drug, in which multiple signaling cascades are impaired; instead, combined inhibition of PI3K and BET shows a beneficial antitumor effect in a model of metastatic breast cancer driven by PI3K and MYC. Multi-target therapeutics can be accomplished by two

AC C

EP

TE D

M AN U

SC

different strategies, a combination of single targeted drugs or administration of a multi-targeted agent. In the last decade, significant efforts have been directed toward the development of combination therapies in the treatment of different types of cancers. [3, 4] Certain cancers have been effectively treated with such combinatorial approach. However, the benefits of administering multiple agents are often limited by dose-limiting toxicities and drug-drug interactions. In addition, the cost of a single agent could be less than two separate agents. (Table. 1) Therefore, an alternative strategy for efficiently eliminating cancer cells is developed with single agent simultaneity modulating the biological function of multiple targets. Some such advantages include the avoidance of different bioavailabilities, pharmacokinetics and metabolism of each component within the combination regimen. The risk of possible drug–drug interactions would be avoided and the dosing regimen would be greatly simplified to enhance compliance and therapeutic efficacy. Increasingly, it is widely being recognized that a multi-targeted agent can provide a superior therapeutic effect and minimal side effect profile compared to the action of a mono-targeted agent. [5, 6] Positive clinical and preclinical data suggest that the next wave of new treatments in cancer will involve multi-targeted agent approaches to therapy.[7, 8] Although the numerous current drugs have been found to possess activity for more than one target, the design of multi-targeted drugs is just becoming a popular trend. Morphy and Rankovic elegantly discussed this approach in recent articles, which was mostly concerned with non-cancer disease. [9, 10] This review on new trends and approaches toward anticancer multi-targeted agents is not meant to be exhaustive. Rather, our intent here is to continue from the excellent perspective by Morphy and Rankovic [11] to report the progress on the discovery progression of multi-targeted agents for cancer therapy and exemplary for the different strategies that could be pursued and those that have already proved to be valuable and fruitful. 2. .Strategies for discovering multi-targeted agents Screening approaches and knowledge-based approaches are two fundamentally different methods for the discovery of multi-targeted agents in literature. To obtain novel multi-targeted agents, a design strategy is usually applied in which distinct pharmacophores of different drugs are integrated into the same structure to yield hybrid molecules. In principle, each pharmacophore of these multi-targeted agents should retain the ability to interact with its specific site(s) on the target and, consequently, to produce specific pharmacological responses that, taken together, should interrupt the cancer progression. The commonness of multi-targeted agents is multi-pharmacophore mode (Fig. 1) which can be divided into the following four types: non-cleavable conjugated-pharmacophore mode, cleavable conjugated-pharmacophore mode, fused-pharmacophore mode and merged-pharmacophore mode. In the conjugated-pharmacophore mode, two distinct pharmacophores are connected

2

ACCEPTED MANUSCRIPT

SC

RI PT

through a linker which does not exist in either of two pharmacophores. In some cases, conjugated-pharmacophore agents contain a metabolically cleavable linker designed to release two agents with distinct in vivo activities. In fused-pharmacophore mode, two distinct pharmacophores are slightly overlapped. More commonly, in order to produce smaller and simpler molecules with favorable physicochemical properties, medicinal chemists aspire to maximize the degree of pharmacophore overlap using merged-pharmacophore mode. Given the broad anticancer activity of multi-targeted agents and their ability to act synergistically with different targets, these four types of multi-pharmacophore modes should be broadly applicable to design small multi-targeted agents for the treatment of cancer. Computational techniques can assist medicinal chemists in the efforts to design novel multi-targeted agents interacting simultaneously with multiple targets and alternative bioactive molecules at an early stage. Some applications such as molecular docking, fragment based design and multi-target quantitative structure–activity relationship (mt-QSAR) analysis would be of assistance to predict biological activity against multiple targets or optimize multi-targeted agents for better lead molecules and drugs. [12, 13, 14, 15, 16]

TE D

M AN U

3. Recent advances of multi-targeted anticancer agents The use of drugs to target different, often unrelated, pathways or multiple biological targets, with the expectation of synergistic effects and low toxicity than combinational therapy, becomes an approach of increasing interest. Therefore, the discovery of multi-targeted agents through rational design has been a subject of growing interest in anticancer drug discovery. Here, we will discuss a number of examples of multi-targeted agents that have just been identified as multi-targeted agents with anticancer activities via knowledge-based approaches.

EP

3.1 Multi-targeted agents targeting tyrosine kinase and inducing DNA-damage Epidermal growth factor receptor (EGFR), also known as erbB1 or HER1, is overexpressed in many solid tumors and play a critical role in the progression of aggressive tumors. Small molecule inhibitors of EGFR, such as gefitinib and erlotinib, have gained widespread acceptance for the treatment of certain cancers with overexpression of EGFR. [17] However, most EGFR inhibitors induce reversible growth inhibitory activity and in many cases inhibiting one single target does not suffice to eradicate the tumor. To circumvent this problem, the combination of

AC C

potent tyrosine kinase inhibitors with cytotoxic agents has become an actively explored strategy. As the forerunners in the rational design of ‘combi-molecules’ with binary epidermal growth factor receptor (EGFR) targeting and significant DNA-damaging properties, Jean-Claude and co-workers reported several combi-triazene molecules in last decade. Firstly, they demonstrated the feasibility of the principle with triazene carrying a methylating agent designed to alkylate guanine at the O-6 and N-7 positions. The principle is outlined in Fig. 2: A combi-molecule termed SMA41 (TZ-I), in which TZ represents the DNA-damaging tail and I the EGFR TK recognition moiety, is designed to penetrate cells by passive diffusion and bind to the ATP site of EGFR on its own (TZ-I-EGFR) or to be hydrolyzed under physiological conditions to generate a DNA damaging agent (TZ) leading to adducted DNA (TZ-DNA) and another EGFR inhibitor SMA52 (I) blockading EGFR (see I-EGFR). Based on this combi-targeting principle, combi-molecule SMA41 was obtained by appending the alkyltriazene moiety to the 6-position of the quinazoline heterocycle where substituents can be altered without affecting inhibitor binding

3

ACCEPTED MANUSCRIPT

RI PT

affinity for the EGFR ATP binding site. [18] SMA41 is 1.8-fold more potent (IC50 = 36 µM) than its metabolite SMA52 alone (IC50 = 59 µM), and 10-fold more potent than TEM (IC50 = 366 µM) in the O6-methylguanine-DNA methyltransferase-proficient skin cancer cell line A431. In contrast to SMA52 alone, the combi-molecule SMA41 and TEM induced significant DNA damage in A431 cells after 30-min or 2-h drug exposures. SMA41 showed 5-fold greater affinity for the ATP binding site of EGFR than independently synthesized SMA52 in an enzyme assay and blocked EGF induced tyrosine phosphorylation and EGFR autophosphorylation in A431 cells in a dose-dependent manner; Despite SMA41 possessing potent antiproliferative activity against alkylating-agent-resistant cells expressing EGFR in vitro, its antitumor activity in vivo remained moderate due to low solubility and lack of potency of the released free inhibitor and the DNA damaging species.[19]

SC

Considering that the TZ group could be replaced, a few alterations have been made on this scaffold, such as replacing monoalkyltriazene with 3,3-disubstituted nitrosourea, [20] using the 1,2,3-triazene as a carrier of the bulky mustard, [21] masking the monoalkyltriazene with

M AN U

carbamates [22], grafting on the Capecitabine (an orally available prodrug of 5-fluorouracil) by di-carbamate linker, [23] which leads to the discovery of several combi-molecules with desired properties such as good potency of EGFR tyrosine kinase inhibitory, high selectivity to target cells expressing EGFR, optimal stability and degradation half-life.

AC C

EP

TE D

3.2 Multi-targeted agents targeting EGFR and Src kinase While the co-administration of an EGFR inhibitor and a Src kinase inhibitor as two separate agents is a clear choice for providing combined therapy, an alternative approach is to design a dual-pharmacophore single agent to act against both EGFR and Src kinase. With the purpose of developing agents targeted to both EGFR and Src or related kinases, Barchéchath S and co-workers [ 24 ] designed combi-molecule SB163 (Fig. 3), a chimeric molecule termed combi-molecule with activity against EGFR with an IC50 value of 0.32 µM and Src kinase with an IC50 value of 2.9 µM. Based on the binding model of X-ray crystal structures of known EGFR and Src inhibitors, a dual inhibitor SB163 was designed by combining two pharmacophore moieties in the same molecule optimized for the binding at each kinase site and linked by an appropriate linker. The optimal position to attach the linker was 6-position of the quinazoline (directed at EGFR) and the 9-position of the pyrazolopyrimidine (directed at Src), and the length of linker had been verified through molecular modeling. The activity of SB163 was superior to that of an equimolar combination of Iressa and PP2 in the Boyden Chamber assay. Although SB163 was a moderately potent inhibitor of both Src and EGFR, this work, as a preliminary attempt, brought us a new insight into the development of multi-targeted agents based on non-cleavable conjugated-pharmacophore approach. For example, the nature of linker (rigidity, hydrophobicity, length) joining the two pharmacophore moieties may contribute to the overall efficacy of the conjugates. During the design of the conjugates, the availability of structural information related to the ligand-receptor interaction will help though this is not an absolute requirement. 3.3 Multi-targeted agents targeting thymidylate synthase (TS) and dihydrofolate reductase (DHFR) As an attractive target in cancer chemotherapy, folate metabolism has been under research

4

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

investigation for many years. The key enzymes, thymidylate synthase (TS) and dihydrofolate reductase (DHFR), are crucial for the synthesis of thymidylate in dividing cells. Several antifolates inhibiting TS and DHFR have come into clinical practice as anticancer agents including raltitrexed, methotrexate and pemetrexed. [25, 26, 27] Pemetrexed represents an important example of a clinically used classical multi-targeted antifolate that has been reported to possess the inhibitory activity of TS, DHFR and glycinamide ribonucleotide formyltransferase, which generated renewed interest in the design of single agents that function as dual inhibitors against TS and DHFR. [28] Base on the framework of classified antifolates, Gangjee and co-workers [29, 30] reported several series of antifolates as dual inhibitors against TS and DHFR. In order to determine the importance of 2,4-diaminoquinazoline ring in methotrexate framework, they started from the 2,4-diaminoquinazoline scaffold with structural changes, replacing of the quinazoline with furo[2,3-d]pyrimidine. Compound (1) (Fig. 4) was obtained in the rationale of bioisostere strategy. Importantly, they hypothesized that the furo[2,3-d]pyrimidine ring bind in either the 2-amino-4-Oxo mode (“normal” mode) (Fig. 4) to TS, or the 2,4-diamino mode (“flipped” mode) (Fig. 4) to DHFR, which is achieved by rotating the “normal” mode by 180° around its NH2-C2 bond. Subsequently, these two primary binding modes had been determined by the crystal structure. [31] Based on the two types binding mode of 6-5 fused ring system, a variety of classical dual TS and DHFR inhibitors as antitumor agents which including pyrrolo[2,3-d]pyrimidines, pyrrolo[3,2-d]pyrimidines, thieno[2,3-d]pyrimidines structural scaffolds were designed and synthesized by Gangjee’ group in recent years (Fig. 4).[32, 33, 34, 35] The inhibitory activities of some of these dual inhibitors were more potent than pemetrexed against TS and DHFR. Cytotoxicity was evaluated by using a broad panel of tumor cell lines, with some of these dual inhibitors displaying IC50 values in the nanomolar range. Several disadvantages are associated with the clinical use of classical antifolates such as i) dependence on folylpolyglutamate synthetase for their inhibitory activity of the target enzymes, ii) requirement of the reduced folate carrier (RFC) system for active uptake into the cell, which, when impaired, causes tumor resistance. [36, 37] To overcome these potential drawbacks associated with classical antifolates, Gangjee and co-workers reported several lipophilic nonclassical antifolates developed as antitumor agents, based on the structure of nolatrexed (the first antitumor nonclassical TS inhibitor to reach clinical trials). [32, 33, 34, 38] Since these nonclassical antifolates are lipophilic and lack the polar glutamate moiety, they do not require the folate transport system and RFC system but enter cells via diffusion or passive transport. The discovery of classical and nonclassical dual TS and DHFR inhibitors was recognized as using an innovative bioisostere strategy based on the merged-pharmacophore model: 6-5 fused ring scaffold matches the middle hydrophobic center of pemetrexed or nolatrexed and the overall size of the molecule satisfied binding TS and DHFR through the rotation. Recently, Arooj and co-workers reported a novel computational approach by integrating the affinity predictions from structure-based virtual screening with dual ligand-based pharmacophore to discover potential dual inhibitors of human TS and human DHFR, and eight optimized dual hits demonstrated excellent binding features on TS and DHFR. [39] 3.4 Multi-targeted agents targeting aromatase and steroid sulfatase Most cases of breast tumors are found to be hormone-dependent, with estrogens playing a

5

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

vital role in the growth and development of the disease. One strategy to tackle hormone-dependent breast cancer involves the inhibition of the production of oestrogens, thus depriving the tumor of these mitogenic steroids. Aromatase and steroid sulfatase (STS) are two essential enzymes in estrogens biosynthetic pathway. There is now abundant and strong evidence to conlude that the deprivation of estrogen levels in patients treated with aromatase inhibitors (AIs) can be augmented if STS is inhibited at the same time. [40] Adopting merged-pharmacophore approach, Potter and coworkers [41, 42, 43, 44, 45, 46] have previously reported and validated the concept of non-steroidal dual aromatase–sulfatase inhibitors (DASI) whose structures are based on the YM511, letrozole, anastrozole template. (Fig. 5) The first series of DASI was designed by fusing the STS inhibitory pharmacophore, i.e., phenyl sulfamate ester moiety present in irosustat, with the aminotriazole moiety which is the essential pharmacophore of AIs. Compound (2) inhibits sulfatase and aromatase in JEG-3 cells with respective IC50 values of 20 and 2.3 nM (YM511, IC50 = 0.5 nM). Further structure–activity relationship studies performed on 4-{[(4-Cyanophenyl)(4H-1,2,4-triazol-4-yl)amino]methyl}phenyl sulfamate revealed that relocation of the halogen atom and replacement of the triazolyl group with an imidazolyl group gives compound (3) improved inhibitory activities on both the aromatase and STS by one order of magnitude. For symmetrical aromatase inhibitors such as letrozole, which contains a tetrahedral carbon center, replacing one of its two p-cyanophenyl rings with a phenol sulfamate ester moiety will render the resulting DASI compound (4). The bis-sulphamate compound (4) exhibited IC50 values of 3044 nM for aromatase and >10 µM for STS in JEG-3 cells. However, at a single oral dose of 10 mg/kg, compound (4) inhibited aromatase and rat liver STS by 60% and 88%, respectively, 24 h after administration. Adopting a similar strategy, the anastrozole-based DASI compound (5) was designed by retaining the haem-ligating triazole moiety as the key for the reversible inhibition of aromatase, and the removal of one of the two dimethyl-propionitrile side-arms of anastrozole allows the incorporation of a phenol sulfamate ester group via linkers into the hybrid inhibitor. Compound (5) had a high degree of potency against aromatase (IC50 = 3.5 nM), comparable with that of Anastrozole (IC50 = 1.5 nM) whereas, only moderate activity against steroid sulfatase was found. However, when compound (5) was tested in vivo, both the plasma E2 levels and liver STS activity were greatly inhibited (STS: 83% and Aromatase: 70%), confirming that compound (5) is indeed functioning as a DASI. Replacing p-chlorophenyl ring with a phenol sulfamate ester moiety, vorozole-based DASI compound (6) was designed. Compound (6) exhibits good inhibitory activity against aromatase (IC50 = 29 nM), although poor STS inhibition was observed (IC50 >10 µM). Despite the relatively high in vitro potency observed for AIs, only YM511-based DASI showed desirable inhibition activity against STS. 3.5 Multi-targeted agents targeting kinesin spindle protein (KSP) and aurora-A kinase Considering KSP and aurora-A kinase are two key enzymes in the mitotic machinery, and KSP is one of aurora-A kinase substrates, [47] our group proposed the dual targets of KSP and aurora-A kinase may be more effective for blocking multiple key components of the mitotic machinery compared to separate agents and would be beneficial to enhance antitumor activity toward a wide range of cancers. Based on fused-pharmacophore strategy, our group utilized dihydropyrazolo[3,4-b]pyridine moiety as the core structure in which the pyrazole ring was supposed to perform as an aurora-A

6

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

kinase hinge region–interacting group while the dihydropyridine ring maintained in CPUYL064 as a KSP scaffold. [48] (Fig. 6) The structure-activity relationship of serial 1-3 showed the phenyl or thienyl series are generally preferred for the inhibition of KSP whereas furyl series leads to potent inhibition of aurora-A kinase. Some serial 1-3 compounds exhibited the potent inhibitory activity for KSP and moderate activity for Aurora-A kinase; however, the cytotoxicity dramatically decreased. We proposed that the disappointing cytotoxicities of those compounds could be attributed to the relatively poor membrane permeability of them. To optimize further and determine the optimal hinge region–interacting group, we carried out a more extensive exploration of hinge binding group by replacing the pyrazole group with the benzimidazole moiety to afford the target compounds, which employed benzo[4,5]imidazo[1,2-a]pyrimidine moiety as the core structure. Although serial 4 componds were moderately potent KSP and aurora-A kinase inhibitory fused-molecules, most compounds displayed more balanced inhibitory activities against two targets compare to dihydropyrazolo[3,4-b]pyridines and exhibited inhibition on the growth of the HCT116 cells (IC50 = 4.96 - 11.49 µM), much more active than its counterpart series 3 componds (IC50 > 19.76 µM). Correspondingly, mi log P value for benzimidazole series compound (mi logP = 4.0 to 5.0) was higher than for furyl series compound (mi logP = 2.7 to 3.3)indicating better membrane penetration which could at least partially account for the good cytotoxicity obtained despite weaker KSP inhibition and similar Aurora-A kinase inhibition.

AC C

EP

TE D

3.6 Multi-targeted agents targeting tyrosine kinase and microtubule Combination chemotherapy containing both antiangiogenic and cytotoxic agents has shown significant promise, and several studies with such combinations are in progress in the clinic. [49] Adopting merged-pharmacophore approach, Zhang and co-workers reported a novel tumor targeting strategy, transposing the 5-substitution of the receptor tyrosine kinase(RTK) inhibitor compound (7) (Fig. 7) on to the 4-NH moiety of compound (7) and placing a 6-CH3 group on compound (7), producing a hybrid furo[2,3-d]pyrimidine scaffold (Fig. 7) capable of antitubulin activity structurally akin to compound (8) with the ability to access the hydrophobic region compound (7) on the general pharmacophore for RTKs. In the CAM angiogenesis inhibitory assay, compound (9) were the most potent among the series, showing EC50 value of 3.9 µM for the inhibition of angiogenesis, which is about 2-fold less potent than sunitinib and 9-fold more potent than erlotinib. Structure–activity analyses of the other members of this series identified that the size of the N-alkyl group determines the microtubule depolymerization potency. Compound (9) (EC50 = 103 nM) with an N-methyl group is the most potent tubulin depolymerizer in the series. Compound (9) has nanomolar potency (IC50 = 20.8 nM), comparable to cisplatin (IC50 = 10.6 nM), against skin cancer cell line A431 over expressing VEGFR-2 and PDGFR-β. Compound (9) binds at the colchicine site on tubulin, depolymerizes cellular microtubules, inhibits purified tubulin assembly, overcomes both β III-tubulin and P-glycoprotein-mediated drug resistance, and initiates mitotic arrest leading to apoptosis. The corresponding HCl salt form was found to reduce tumor size and vascularity in xenograft and allograft murine models and was superior to docetaxel and sunitinib, without overt toxicity. [ 50 ] Recently, a series of conformationally restricted, 4-substituted 2,6-dimethylfuro[2,3-d]pyrimidines was designed to explore the bioactive conformation required for dual inhibition of microtubule assembly and RTKs. The tetrahydroquinoline analog compound (10) (Fig. 7) showed microtubule depolymerizing activity comparable to or better than the lead compound (9), some of which have nanomolar EC50 values.

7

ACCEPTED MANUSCRIPT Compound (10) also showed potent RTK inhibition with nanomolar IC50 value. Compared with lead compound (9), compound (10) showed a significantly improved potency against most of the NCI 60 cancer cells including several chemo-resistant cell lines. [51]

TE D

M AN U

SC

RI PT

3.7 Multi-targeted agents targeting estrogen receptor α (ERα) and vascular endothelial growth factor receptor-2 (VEGFR-2) In 2014, Tang and co-workers [52] reported a novel tumor targeting strategy, combining the structural features of VEGFR-2 inhibitor with ERα inhibitory groups to overcome drug resistance by synergistic inhibition of two pharmacological targets. The structure of the potent ERα inhibitor acolbifene, which displays only inhibitory activity against ERα, was modified to obtain derivatives capable of simultaneously inhibiting both ERα and VEGFR-2 activities. The design strategy entailed the replacement of the existing 2H-1-benzopyran moiety of acolbifene with groups with specialized side chains, such as 2,3-diaryl isoquinolinone derivatives moiety, which provides the ability to selectively inhibit VEGFR-2. which provides the ability to selective inhibition against VEGFR-2. Compound (11) (Fig. 8) was found to be the most active inhibitor toward ERα with the IC50 value of 1.3 µM which was slightly better than that of tamoxifen (IC50= 1.9 µM). It also displayed moderate inhibition activities toward VEGFR-2 (IC50= 1.4 µM). Moreover, the ability of compound (11) to inhibit MCF-7 cell lines with the IC50 values of 2.73 µM was verified in comparison to that of the lead compound tamoxifen. As their ongoing interest to explore different scaffold structures as multiple ligands of ERα and VEGFR-2, a series of 6-aryl-indenoisoquinolone derivatives with basic side chain as dual ERα and VEGFR-2 inhibitors were designed and synthesized. [53] Compounds (12) containing two hydroxyl groups to mimic estradiol were able to form two hydrogen bonds with ERɑ. Compound (12) (Fig. 8) presented decent ERα binding affinity(IC50 = 7.2 µM) and ERα antagonistic activity, as well as potent VEGFR-2 inhibitory potency (IC50 = 0.099 µM). To evaluate the anticancer activity, compound (12) was screened against human breast cancer cell line MCF-7 (ER+),

AC C

EP

MDA-MB-231 (ER-) and human endometrial cancer cell line Ishikawa, showing IC50 value of 1.2 µM, 0.5 µM and 8.2 µM, respectively. Compound (12) could inhibit the growth of both MCF-7 (ER+) and MDA-MB-231 (ER-) breast cancer cells better than tamoxifen which indicated that the anti-proliferative activity was acquired not only through ERɑ but a multi-target effect. Further investigation of compound (12) showed that it was able to inhibit the activation of VEGFR-2 and the signaling transduction of Raf-1/MAPK/ERK pathway in MCF-7 cells. This work opens a novel way to the design of molecules where inhibitory pharmacophores of two families’ targets can be integrated into a single molecule. These types of bifunctional compounds can be potential double targeting agents for use in the therapy of some breast cancers. 3.8 Multi-targeted agents targeting histone deacetylase (HDAC) and other targets Zinc-containing enzymes, including matrix metalloproteinases (MMPs), carbonic anhydrases (CAs) and histone deacetylases (HDACs), are attractive therapeutic targets for treatment of cancer. In contrast to the “single target” strategy adopted for HDAC inhibitors like SAHA, some designed multi-targeted approaches targeting various key proteins have been reported. Here we give some examples of multi-targeted agents targeting HDACs and other targets. Numerous studies have suggested that dual inhibition of HDACs and other targets induces a synergistic effect on impairing cell proliferation and triggering apoptotic cell death. Moreover,

8

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

HDAC inhibitors are recognized as ‘sensitizing drugs’ that manifest synergistic efficacy with other anticancer drugs, and more interestingly their nature of chemical flexibility, readily embedding other drugs (such as protein kinase inhibitors, 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors, topoisomerases inhibitors, cyclooxygenases inhibiors, estrogen receptor antagonists and androgen receptor antagonists) into the structure of HDAC inhibitors facilitates medicinal chemists to embark the journey toward the dual inhibitor discovery. The residues presented at the outer rim of HDAC enzymes form rugged landscapes designed to flexibly accommodate a diverse class of substrates. Structural studies of the binding between HDAC and its inhibitors have identified a canonical pharmacophore comprised of three distinct motifs, capping group, linker, and zinc binding group (ZBG). Capping group is a surface recognition unit and usually contains a hydrophobic and aromatic moiety to interact with the rim of the binding pocket. Linker domain is designed to connect the cap group to ZBG as a saturated or unsaturated structure. Intriguingly, the multi-targeted agents targeting HDAC and other targets comprise a first pharmacophore that binds zinc ions and inhibits HDACs, and a second pharmacophore, which is combined (or fused) to the zinc-binding moiety, and which inhibits one or more different signaling pathways or biological functions.

EP

TE D

3.8.1 Multi-targeted agents targeting histone deacetylase (HDAC) and kinases To overcome the low response rate and acquired resistance to protein kinase inhibitors (such as EGFR kinase inhibitors), a number of strategies have been tested, including combination therapies and multi-targeted inhibitors to block multiple pathogenic pathways. One particularly promising approach is the modulation of kinase pathways by the inhibition of HDACs. In 2010, Cai and co-workers [54] for the first time introduced multi-targeting HDAC, EGFR, and HER2 Inhibitor. In this study, they covalently combined HDAC inhibitor vorinostat (SAHA) with EGFR inhibitor erlotinib to furnish a novel multi-targeted agent CUDC-101 (Fig. 9). CUDC-101 displays potent in vitro inhibitory activity against HDAC, EGFR, and HER2 with an IC50 of 4.4, 2.4, and 15.7 nM, respectively. In most tumor cell lines tested, CUDC-101 exhibits efficient antiproliferative activity with greater potency than vorinostat, erlotinib, lapatinib, combinations of vorinostat/erlotinib as well as vorinostat/lapatinib. As expected, the study of the molecular mechanism illustrated that CUDC-101 presented inhibition against HDAC, EGFR, and

AC C

HER2 under the cell-free condition and in vitro cell cultures. A phase Ib expansion study investigating the safety, efficacy, and PKs of intravenous CUDC-101 in patients with advanced head and neck, gastric, breast, liver and NSCLC tumors (NCT01171924) and a phase I study of orally administered CUDC-101 to evaluate its bioavailability have been conducted (NCT01702285). In 2012, the same research group reported another multi-targeted agent, CUDC-907 as a follow-up to their previous study. In contrast to the earlier study, they designed PI3Ks -HDAC inhibitor by covalently merging SAHA-like HDAC inhibitor to the GDC-0941 framework. (Fig. 9) [55] The potency of CUDC-907 against class I HDACs was similar to that of panobinostat and greater than vorinostat. CUDC-907 is also a potent inhibitor of class I PI3K kinases with an IC50 value of 19, 54, and 39 nM for PI3Kɑ, PI3Kβ, and PI3Kδ, respectively. This activity is similar to that of the lead PI3K inhibitor, GDC-0941(IC50 value of 19, 54, and 39 nM for PI3Kɑ, PI3Kβ, and PI3Kδ, respectively.). Through its integrated HDAC inhibitory activity, CUDC-907 durably inhibits the PI3K-AKT-mTOR pathway and compensatory signaling molecules such as RAF,

9

ACCEPTED MANUSCRIPT

RI PT

MEK, MAPK, and STAT-3, as well as upstream receptor tyrosine kinases. CUDC-907 shows greater growth inhibition and proapoptotic activity than single-target PI3K or HDAC inhibitors in both cultured and implanted cancer cells. Phase I study to assess the safety, tolerability and pharmacokinetics of CUDC-907 in patients with lymphoma or multiple myeloma are registered at ClinicalTrials.gov, number NCT01742988. [56] In 2016, Yang and co-workers create bispecific single molecules with both JAK2 and HDAC targeted inhibition. [57] In their study, they systematically varied linker length and zinc-binding group to clarify structure-activity relationships. Among a series of hybrid inhibitors they

TE D

M AN U

SC

synthesized, EY3238 (Fig. 9) inhibits JAK2 and HDAC6 in low nanomolar range, <100 nM potent against HDACs 2 and 10, sub-micromolar potent against HDACs 1, 8 and 11, and >50 fold selective for JAK2 in a panel of 97 kinases. EY3238 also demonstrated significant antiproliferative activity towards the several hematological cell lines. This study provides new tool compounds for further exploration of dual JAK-HDAC pathway inhibition achieved with a single molecule. Strategically, these designs can be regarded as merged-pharmacophore approach, involve the incorporation of an HDAC inhibitor pharmacophore to a known potent kinase inhibitor, using the kinase ATP domain hinge-binding fragment as a “capping group” such as 4-anilino-quinazoline which in turn is anticipated to act as an aromatic surface recognition cap group essential for HDAC inhibition while also retaining its EGFR kinase inhibition activity. Considering that the “capping group” could be replaced, other research groups also explored this scaffold in ways such as replacing aniline with 4,5,6,7-tetrabromobenzotriazole or macrocyclic compounds to discover several combi-molecules with desired properties such as good potency of different protein kinase inhibitory. [58, 59] Apparently, the key to this strategy would be to identify a tolerant region (usually located in the kinase solvent exposed region) in the protein kinase inhibitor. These works opened possibilities for the development of novel kinases inhibitors that emerged as potential new cancer therapeutics because of their ability to inhibit the kinases while simultaneously interfering with HDAC.

AC C

EP

3.8.2 Dual inhibitors of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase (HMGR) and HDAC Recently, preclinical studies have employed a variety of breast cancer cell lines and animal models to better comprehend the underlying mechanisms between mevalonate inhibitions to anticancer effects. Much preclinical and epidemiologic evidence were obtained to support the anticancer effects of HMGR inhibitors (lipophilic statins). [60] It has been reported that the combination use of HDAC inhibitor with statins may synergistically prevent mouse lung tumors and induced apoptosis in HeLa cells.[61, 62] In another elegant example of the merged-pharmacophore approach to dual-targeted agents, Chen and co-workers designed a novel strategy that seeks to combine both HMGR and HDAC inhibitory functions within one single molecule, reducing the toxic to normal cells. [63] Based on the HDAC inhibitor pharmacophore (a hydrophobic cap region, an aliphatic linker, and a zinc binding group.), the authors selected lovastatin and SAHA as the starting compounds. HDAC inhibitor SAHA contains six carbon linker and a hydrophobic cap which are found to be essential for activity. In another aspect, hydroxamic acid is considered as a bioisostere of carboxylic acid. The HMGR inhibitor Lovastatin was also found to possess these groups at the same position as the

10

ACCEPTED MANUSCRIPT

RI PT

HDAC inhibitor thereby illustrating a close similarity of the substituents of the inhibitors even though the two proteinases belong to different classes. These two important features were retained and a dual inhibitor compound (13) (Fig. 10) for HMGR and HDAC were designed by introducing specific functional groups, hydroxamic acid group. Compound (13) showed potent inhibitory activities against HDACs and HMGR with IC50 values in the nanomolar range. Compound (13) significantly induced cytotoxicity in the A549 human lung cancer cells (IC50 = 18.2 µM), but was not toxic to the HS68 normal human fibroblast cells at 100 µM. The selectivity index of statins for A549 human lung cancer cells and HS68 normal human fibroblast cells was low (SI > 5.6), while SAHA was toxic to both cancer and normal cells without specificity.

AC C

EP

TE D

M AN U

SC

3.8.3 Multi-targeted agents targeting topoisomerases and HDAC Current chemotherapeutic options for the treatment of cancer are often plagued by debilitating side effects and off-target toxicities. Since both topoisomerases and HDAC are crucial for tumor progression, simultaneous modulation of both independent cellular pathways, by a single agent, could be highly beneficial.[64] Studies aimed at identifying dual inhibitors of topoisomerases and HDAC inhibitory activity as promising pharmacological tools that may be more efficacious for various human cancers than highly selective single-target drugs are ongoing in several academic and industrial pharmaceutical laboratories. Synergistic reports of antitumor effects of HDAC inhibitors with topoisomerase II (Topo II) inhibitors led to the design of anthracycline daunorubicin-SAHA hybrids by the incorporation a hydroxamic acid moiety, the key fragment of SAHA within the Topo II inhibitors. Adopting fused-pharmacophore approach, Guerrant and co-workers [65] designed dual inhibitors of Topo II and HDAC and evaluated them for HDAC inhibition, using a cell-free assay as well as cytotoxicity in three cancer cell lines. Linker length was systematically varied to determine the most suitable molecule. All the hybrids exhibited potent HDAC inhibition activity which suggested that daunorubicin moiety is an appropriate capping group for HDAC inhibitors. Compound (14) (Fig. 11) showed the strongest HDAC inhibitory activity and inhibited Topo II activity at comparable levels to that of daunorubicin at the same drug concentration as well as powerful antiproliferative activity against DU-145 (IC50 = 0.13 µM), SK-MES-1 (IC50 = 0.47 µM) and MCF-7 (IC50 = 0.99 µM). Since both HDAC and Topo I enzymes are localized to the nucleus, the opportunity for dual inhibition from a single agent is a promising possibility. As a follow-up to their work on dual-acting HDAC–Topo II inhibitors, using merged-pharmacophore approach, the same research group reported another class of dual-acting HDAC-Topo I inhibitors derived from the camptothecin ring system. [66] The linker region of SAHA-like HDAC inhibitor is coupled through a triazole moiety to the camptothecin template, which in turn is anticipated to act as an aromatic surface recognition cap group essential for HDAC inhibition while also retaining its Topo I inhibition activity. Compound (15) (Fig. 11) retained inhibitory activities against both target enzymes and inhibited the proliferation of DU-145 prostate carcinoma cells. It was reported that the resistance to Top II inhibitors is often concomitant with a rise in the level of Top I expression and vice versa.[67] In this regard, a single molecule able to inhibit TopI/TopII/HDAC could be helpful to prevent mechanism-based drug resistance and show more potent antitumor activity. 3-Amino-10-hydroxylevodiamine was proven to be a dual Top I and Top II inhibitor by in silico target identification in combination with biological assays.[68, 69] From

11

ACCEPTED MANUSCRIPT

TE D

M AN U

SC

RI PT

previously structure−activity relationship studies on evodiamine derivatives, substitution at the 3-amino group was tolerable. Because of the presence of large hydrophobic patches at the HDAC surface rim, conjugating SAHA with hydrophobic 3-amino-10-hydroxylevodiamine generate potent triple topoisomerase I/II and HDAC inhibitors. [70] 1,2,4-oxadiazoles always used as a flat, aromatic linker to place substituents in the appropriate orientation for ligand binding, was introduced between the evodiamine scaffold and the zinc binding group of SAHA. The results show that all the 1,2,4-oxadiazole derivatives potently inhibited HDAC1 with IC50 values in the nanomolar range. Moreover, it was observed that the HDAC1 inhibitory activity was enhanced with the increasing of the carbon chain length. In particular, Compound (16)(Fig. 11)with a six methylene alkyl linker proving to be optimum for HDAC1 inhibitory activity (IC50 = 24 nM) among the 1,2,4-oxadiazole derivatives, showed comparable activity to SAHA (IC50 = 23 nM). To obtain evidence for the HDAC isoform selectivity, compound (16) was evaluated against recombinant HDAC2, HDAC3, HDAC6 and HDAC8. Compound (16) displayed nanomolar activity against HDAC2 (IC50 = 275 nM), HDAC3 (IC50 = 71 nM) and HDAC6 (IC50 = 13 nM). In contrast, its activity against HDAC8 (IC50 = 2.5 µM) was significantly decreased. As compared with SAHA, compound (16) was more potent against HDAC6 and HDAC8. Compound (16) was found to be active against Top I-mediated DNA relaxation and showed higher Top I inhibitory activity than camptothecin for the HCT116 cell line. Comparable to lead compound 3-amino-10-hydroxylevodiamine, compound (16) was significantly more active than the parental drug SAHA. Three attractive targets, HDAC, topoisomerase I and topoisomerase II, are cellular modulators that can broadly arrest cancer proliferation through a range of downstream effects. In the designed multi-targeted agents, the knowledge of SAR around topoisomerase I/II inhibitor and HDAC inhibitor was critical to identify the key pharmacophoric features and most optimal position for connection of the two frameworks.

AC C

EP

3.8.4 Multi-targeted agents targeting tubulin and HDAC Chemotherapeutic combinations incorporating DNA-damaging and microtubule-targeting agents are often used as first- and second-line treatments for patients with various cancers. HDAC inhibitors can have a synergistic antitumor effect when combined with tubulin inhibitors. [71] Using fused-pharmacophore approach, Zhang and co-workers designed five dual inhibitors of tubulin and HDAC contain a colchicine moiety as a capping group and a hydroxamic acid as a zinc-binding group (ZBG).[72] Among these compounds, compound (17) (Fig. 12) showed medium inhibitory activity against HDAC1 (IC50 = 1.33 µM), HDAC3 (IC50 = 1.36 µM) and HDAC6 (IC50 = 3.43 µM), but showed the strongest anti-proliferative activity against A431 (IC50 = 0.242 µM), A549 (IC50 = 4.672 µM), HCT116 (IC50 = 0.903 µM), MCF-7(IC50 = 0.825 µM) and PC-3 (IC50 = 0.813 µM). Prompted by these efforts to further explore this kind of dual inhibitors, they designed and synthesized a new series of colchicine derivatives with benzamide moiety as HDAC ZBG. Biological evaluation displayed that most of these hybrids with colchicine moiety as HDAC capping group exhibited potent HDAC inhibitory activity. [73] Compared with molecules equipped with unsubstituted benzamide ZBG, hybrids with bis(aryl)-type ZBG showed better anti-HDAC activity. Compound (19) displayed the most efficient anti-HDAC activity which is comparable with mocetinostat and showed comparable tubulin inhibitory activity with the positive control colchicines. Compound (18) with moderate HDAC and tubulin inhibitory activity showed

12

ACCEPTED MANUSCRIPT the best cytotoxicity which is comparable with colchicine (IC50 = 0.106 µM), while compound (19) with powerful HDAC and tubulin inhibitory activity displayed moderate anti-proliferative activity against A549 cell line (IC50 = 2.79 µM). The difference between their cytotoxicity could be due to the different cell membrane penetration. Compound (18) also showed superior cytotoxicity to the positive control compounds (mocetinostat and colchicine) against a panel of human cancer cell lines (IC50 = 2-105 nM).

EP

TE D

M AN U

SC

RI PT

3.8.5 Multi-targeted agents targeting cyclooxygenases(COX) and HDAC COX-2, an inducible enzyme, is associated with inflammatory diseases and carcinogenesis, which is suspected to promote angiogenesis and tissue invasion of tumors. [74] Due to high expression of COX-2 in most tumors, it has been proposed that non-steroidal anti-inflammatory drugs (NSAIDs) could someday find applications in the prevention and/or cure of some cancers, especially for colon and prostate cancer. [75] Several mechanisms of cytotoxicity of NSAIDs towards cancer cells have been reported; most are believed to be irrelevant of COX-2 inhibition. It has been confirmed that celecoxib induces apoptosis in human osteosarcoma cell line MG-63 via down-regulation of PI3K/Akt. [76] The combination of HDAC1/3 and COX-2 inhibition has been reported to significantly impair proliferation of BxPC-3 cells in vitro and entirely stalled the BxPC-3 cells tumor growth onto the chorioallantoic membrane in vivo. [77] Based on the analysis that the sulfonamide (SO2NH2) and trifluoromethyl (CF3) moieties of celecoxib are projected towards different solvent-exposed regions of the COX-2, these two moieties could be suitable points for the attachment of HDAC-inhibiting pharmacophores. Similarly, for indomethacin, the carboxylic acid moiety is projected towards the solvent exposed region of COX-2. Four series of NSAID-HDACi conjugates (celecoxib-SAHA and indomethacin-SAHA linked hybrids ) (Fig. 13) were designed and evaluated for activity against a panel of cancer cell lines: breast (MCF-7), lung (A549), colon (HCT-116), androgen-dependent (LNCaP) and -independent (DU-145) prostate cancer cell lines. [78] These compounds 20-23 potently inhibited the growth of cancer cell lines consistent with their anti-COX and anti-HDAC activities.

AC C

3.8.6 Multi-targeted agents targeting ERα and HDAC Estrogen receptor is a crucial determinant of resistance to endocrine therapy, which may change during the progression of breast cancer. Approximately 70% of breast cancers express the ERα and are treated with the ERα antagonist, tamoxifen. However, resistance to tamoxifen frequently develops in advanced breast cancer, in part due to a down-regulation of ERα corepressors. As an indicator of epigenetic disequilibrium, the loss of ERα corepressors may predispose cancer cells to the cytotoxic effects of HDAC inhibitors, which has been shown to effectively reverse tamoxifen resistance in numerous studies. [79] Based on the merged-pharmacophore approach, the design and synthesis of multi-targeted agents compound 24-26 against ERα and HDAC, a combination of HDAC inhibitor pharmacophore moiety (hydroxamate) to ERα antagonist (tamoxifen, ICI-164,384 and raloxifene), have been reported. (Fig. 14) These molecules satisfy the requirements for the interaction with active site of HDAC and do not disturb the binding mode of the ERα core structure [80, 81, 82 ].The hydroxamate derivative of tamoxifen compound (24) proved to be effective dual inhibitors of ERα (IC50 = 0.248 µM) and HDAC6 (IC50 = 0.567 µM) and afforded micromolar IC50

13

ACCEPTED MANUSCRIPT

RI PT

values against MCF-7, MDA-MB-231and DU-145 cancer cell lines. Compound (25) showed suppression of bioluminescence resonance energy transfer signal induced by 17b-estradiol in a dose-dependent manner, consistent with antiestrogenic activity and similar to control ICI-164,384. Although displaying weaker potency against HDAC3 (IC50 = 0.96 µM) than either SAHA or entinostat, compound (25) had a potency against HDAC6 (IC50 = 1.15 µM) within one order of magnitude of SAHA (IC50 = 0.35 µM). Compound (26) proved to be antiestrogenic in cell culture and to inhibit HDAC1, 2, and 3, with IC50 values of 1–3 µM. The extent of cell killing of ER(-) breast cancer cells by compound (26) was equivalent to combination treatments with 4OHT (the active metabolite of tamoxifen), raloxifene, or desmethylarzoxifene, despite compound (26) having antiestrogenic potency higher by 2–3 orders of magnitude. Moreover, compound (26) caused significant cell death at an earlier time point than combination treatments.

TE D

M AN U

SC

3.9 Multi-target approach for the rational in silico design of anticancer agents In recent years, mt-QSAR models have been reported, representing significant advances with respect to those models generated from classical approaches which use a very limited dataset of structurally related compounds against only one target. In the mt-QSAR models, fragments with positive contributions are considered to be 2D-pharmacophores and the quantitative contributions of different molecular fragments to the activity can be calculated. Focusing on the analysis of the structural information and the quantitative contributions of the suitable fragments,Speck-Planche and co-workers developed the first mt-QSAR discriminant model performing for the virtual screening and rational in silico discovery of anti-cancer agents. [83, 84, 85] The first model was based on the linear discriminant analysis (mt-QSAR-LDA) employing fragment-based descriptors while the second model was obtained using artificial neural networks (mt-QSAR-ANN) with global 2D descriptors. Both models correctly classified more than 90% of active and inactive compounds in training and prediction sets. These mt-QSAR models would be of assistance to automatically extract the desirable fragments for the design of new, potent and multi-target anticancer agents.

AC C

EP

4. Conclusion Multi-target therapeutics has currently been one of the most effective strategies for the treatment and prevention of cancer. Multi-targeted drugs, by their nature, contain privileged substructures and pharmacophores that are relevant to multiple targets. The success of multi-kinase inhibitors for the treatment of cancer promotes the research and development of other novel multi-targeted drugs with the ideal pharmacological profile. The lack of anticancer activity in many cases of this review is most likely due to the poor pharmacodynamic and pharmacokinetic properties of obtained molecules. Therefore, for multi-targeted drugs design, the challenge becomes how to design a drug with the desired balance of two (or more) activities in concert with optimizing the ADME-PK properties and safety profiles. Although many of the examples of current multi-pharmacophores are intriguing, they will undoubtedly come to be seen as simplistic and should be viewed as just the beginning of the long road ahead. Numerous applications such as molecular docking, pharmacophore modeling, multi-target QSAR analysis, scaffold hopping, lead based or fragment based design would be of assistance to optimize multi-targeted agents for better lead molecules and drugs. In conclusion, it is our view that the design of multi-targeted agents will continue to play a crucial role in the discovery of next-generation anticancer drugs. It will be

14

ACCEPTED MANUSCRIPT exciting to observe the systematic discovery of multi-targeted drugs applied to both established and novel cancer targets in the future.

SC

Conflict of interest The authors confirm that this article content has no conflict of interest.

RI PT

ACKNOWLEDGEMENT Acknowledgments are made to the Natural Science Foundation of Hunan Province (Grant No. 2015JJ6085 and 2015JJ6082), Research Fund of the Pharmaceutical Science Key Discipline in the Hunan Province, The special Fund for International Cooperation of Chinese Medicine (Grant No. ZYGZ201502) and Scientific Research Fund of Hunan Provincial Education Department (Grant No. 13B080) for financial supports.

References

AC C

EP

TE D

M AN U

[1] R. Dienstmann, J. Rodon, V. Serra, J. Tabernero, Picking the point of inhibition: a comparative review of PI3K/AKT/mTOR pathway inhibitors, Mol. Cancer. Ther. 13 (2014)1021-1031. [2] E. E. Stratikopoulos, M. Dendy, M. Szabolcs, A. J. Khaykin, C. Lefebvre, M. M. Zhou, R. Parsons, Kinase and BET Inhibitors Together Clamp Inhibition of PI3K Signaling and Overcome Resistance to Therapy, Cancer. Cell. 27 (2015) 837-851.. [3] Z. Guan, B. Xu, M. L. DeSilvio, Z. Shen, W. Arpornwirat, Z. Tong, V. Lorvidhaya, Z. Jiang, J. Yang, A. Makhson, W. L. Leung, M. W. Russo, B. Newstat, L. Wang, G. Chen, C. Oliva, H. Gomez, Randomized trial of lapatinib versus placebo added to paclitaxel in the treatment of human epidermal growth factor receptor 2-overexpressing metastatic breast cancer, Invest. New. Drugs. 31 (2013)734-741. [4] R. B. Mokhtari, T. S. Homayouni, N. Baluch, E. Morgatskaya, S. Kumar, B. Das, H. Yeger,combination therapy in combating cancer, Oncotarget. (2017). doi: 10.18632/oncotarget.16723. [5] P. Mondello, E. Derenzini, Z. Asgari, J. Philip, E. J. Brea, V. Seshan, R. C. Hendrickson, E. de Stanchina, D. A. Scheinberg, A. Younes, Dual inhibition of histone deacetylases and phosphoinositide 3-kinase enhances therapeutic activity against B cell lymphoma, Oncotarget. 8 (2017) 14017-14028. [6] F. H. Andrews, A. R. Singh, S. Joshi, C. A. Smith, G. A. Morales, J. R. Garlich, D. L. Durden, T. G. Kutateladze, Dual-activity PI3K-BRD4 inhibitor for the orthogonal inhibition of MYC to block tumor growth and metastasis, Proc. Natl. Acad. Sci. USA. 114 (2017) E1072-E1080. [7] L. Garuti, M. Roberti, G. Bottegoni, Multi-kinase inhibitors, Curr. Med. Chem. 22 (2015) 695-712. [8] L. L. Covell, A. K. Ganti, Treatment of advanced thyroid cancer: role of molecularly targeted therapies, Target. Oncol. 10 (2015) 311-324. [9] R. Morphy, Z. Rankovic, Designed Multiple Ligands. An Emerging Drug Discovery Paradigm, J. Med. Chem. 48 (2005) 6523-6543. [10] R. Morphy, C. Kay, Z. Rankovic, From magic bullets to designed multiple ligands, Drug. Discov. Today. 9 (2004) 641-651. [11] R. Morphy, Z. Rankovic, Designing Multiple Ligands-Medicinal Chemistry Strategies and Challenges, Current. Pharmaceutical. Design.15 (2009) 587-600. [12] A. Speck-Planche, V.V. Kleandrova, F. Luan, M. N. D. S. Cordeiro, Multi-target inhibitors for proteins associated with Alzheimer: In silico discovery using fragment-based descriptors, Curr. Alzheimer Res. 10 (2013) 117-124. [13] F.J. Prado-Prado, I. Garcia, X. Garcia-Mera, H. Gonzalez-Diaz, Entropy multi-target QSAR 15

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

model for prediction of antiviral drug complex networks, Chemometr. Intell. Lab. Syst. 107 (2011) 227-233. [14] A. Speck-Planche, M. N. D. S. Cordeiro, Multi-target QSAR approaches for modeling protein inhibitors. Simultaneous prediction of activities against biomacromolecules present in gram-negative bacteria, Curr. Top. Med. Chem. 15 (2015) 1801-1813. [15] F. Prado-Prado, X. Garcia-Mera, P. Abeijon, N. Alonso, O. Caamano, M. Yanez, T. Garate, M. Mezo, M. Gonzalez-Warleta, L. Muino, F. M. Ubeira, H. Gonzalez-Diaz, Using entropy of drug and protein graphs to predict FDA drug-target network: theoretic-experimental study of MAO inhibitors and hemoglobin peptides from Fasciola hepatica, Eur. J. Med. Chem. 46 (2011) 1074-1094. [16] A. Speck-Planche, V. V. Kleandrova, M. T. Scotti, M. N. D. S. Cordeiro, 3D-QSAR methodologies and molecular modeling in bioinformatics for the search of novel anti-HIV therapies: rational design of entry inhibitors, Curr. Bioinform. 8 (2013) 452-464. [17] F. Cappuzzo, A. Morabito, N. Normanno, P. Bidoli, Conte. A. Del, L. Giannetta, A. Montanino, F. Mazzoni, R. Buosi, M. A. Burgio, G. Cerea, R. Chiari, D. Cortinovis, G. Finocchiaro, L. Foltran, M. R. Migliorino, M. Tiseo, S. Ferrari, F. De Marinis, Efficacy and safety of rechallenge treatment with gefitinib in patients with advanced non-small cell lung cancer, Lung. Cancer. 99 (2016) 31-37. [ 18 ] S. L. Matheson, J. McNamee, B. J. Jean-Claude, Design of a chimeric 3-methyl-1,2,3-triazene with mixed receptor tyrosine kinase and DNA damaging properties: a novel tumor targeting strategy, J. Pharmacol. Exp. Ther. 296 (2001) 832-840. [19] S. L. Matheson, J. P. McNamee, T. Q. Wang, M. A. Alaoui-Jamali, A. M. Tari, and B. J. Jean-Claude, The combi-targeting concept: dissection of the binary mechanism of action of the combi-triazene SMA41 in vitro and antitumor activity in vivo, J. Pharmacol. Exp. Ther. 311 (2004) 1163-1170. [20] J. Domarkas, F. Dudouit, C. Williams, Q. Qiyu, R. Banerjee, F. Brahimi, B. J. Jean-Claude, The Combi-Targeting Concept: Synthesis of Stable Nitrosoureas Designed to Inhibit the Epidermal Growth Factor Receptor (EGFR), J. Med. Chem. 49 (2006) 3544-3552. [21] Z. Rachid, F. Brahimi, Q. Qiu, C. Williams, J. M. Hartley, J. A. Hartley, B. J. Jean-Claude, Novel Nitrogen Mustard-Armed Combi-Molecules for the Selective Targeting of Epidermal Growth Factor Receptor Overexperessing Solid Tumors: Discovery of an Unusual Structure-Activity Relationship, J. Med. Chem. 50 (2007) 2605-2608. [22] Z. Rachid, M. Macphee, C. Williams, M. Todorova, B. J. Jean-Claude, Design and synthesis of new stabilized combi-triazenes for targeting solid tumors expressing the epidermal growth factor receptor (EGFR) or its closest homologue HER2, Bioorg. Med. Chem. Lett. 19 (2009) 5505-5509. [23] M. Ait-Tihyaty, Z. Rachid, A. L. Larroque-Lombard, B. J. Jean-Claude, The first EGFR inhibitor-capecitabine based combi-molecule, requires carboxylesterase-mediated hydrolysis for optimal activity, Invest. New. Drugs. 31 (2013) 1409-1423. [24] S. Barchéchath, C. Williams, K. Saade, S. Lauwagie, B. Jean-Claude, Rational Design of Multitargeted Tyrosine Kinase Inhibitors: A Novel Approach, Chem. Biol. Drug. Des. 73 (2009) 380-387. [25 ] S. Barni, A. Ghidini, A. Coinu, K. Borgonovo, F. Petrelli, A systematic review of raltitrexed-based first-line chemotherapy in advanced colorectal cancer, Anticancer. Drugs. 25 (2014) 1122-1128. [26] H. Akiyama, H. Takase, F. Kubo, T. Miki, M. Yamamoto, M. Tomita, M. Mochizuki, O. Miura, A. Arai, High-dose methotrexate following intravitreal methotrexate administration in preventing central nervous system involvement of primary intraocular lymphoma, Cancer. Sci. 107 (2016) 1458-1464. [27] Z. Song, H. Su, Y. Zhang, Patients with ROS1 rearrangement-positive non-small-cell lung cancer benefit from pemetrexed-based chemotherapy, Cancer. Med. 5 (2016) 2688-2693. [28] E. Esteban, M. Casillas, A. Cassinello, Pemetrexed in first-line treatment of non-small cell lung cancer, Cancer. Treat. Rev. 35 (2009) 364-373. [29] A. Gangjee, R. Devraj, J. J. McGuire, R. L. Kisliuk, S. F. Queener, L. R . Barrows, Classical and nonclassical furo[2,3-d]pyrimidines as novel antifolates: synthesis and biological activities, J. 16

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Med. Chem. 37 (1994) 1169-1176. [30] A. Gangjee, J. Yu, J. J. McGuire, V. Cody, N. Galitsky, R. L. Kisliuk, S. F. Queener, Design, Synthesis, and X-ray Crystal Structure of a Potent Dual Inhibitor of Thymidylate Synthase and Dihydrofolate Reductase as an Antitumor Agent, J. Med. Chem. 43 (2000) 3837-3851. [31] A. Gangjee, J. Yu, R. L. Kisliuk, W. H. Haile, G. Sobrero, J. J. McGuire, Design, Synthesis, and Biological Activities of Classical N-[4-[2-(2-amino-4-ethylpyrrolo[2,3-d]pyrimidin -5-yl)-Ethylpyrrolo[2,3-d]pyrimidin-5-yl]ethyl]benzoyl]-L-glutamic Acid and Its 6-Methyl Derivative as Potential Dual Inhibitors of Thymidylate Synthase and Dihydrofolate Reductase and as Potential Antitumor Agents, J. Med. Chem. 46 (2003) 591-600. [32] A. Gangjee, X. Lin, R. L. Kisliuk, J. J. McGuire, Synthesis of N-{4-[(2,4-diamino-5methyl-4,7-dihydro-3H-pyrrolo[2,3-d]pyrimidin-6-yl)thio]benzoyl}-L-glutamic Acid and N-{4-[(2-amino-4-oxo-5-methyl-4,7-dihydro-3H-pyrrolo[2,3-d]pyrimidin -6-yl)thio]benzoyl}-L-glutamic Acid as Dual Inhibitors of Dihydrofolate Reductase and Thymidylate Synthase, and as Potential Antitumor Agents, J. Med. Chem. 48 (2005) 7215-7222. [33] A. Gangjee, H. D. Jain, J. Phan, X. Lin, X. Song, J. J. McGuire, R. L. Kisliuk, Dual inhibitors of thymidylate synthase and dihydrofolate reductase as antitumor agents: design, synthesis, and biological evaluation of classical and nonclassical pyrrolo[2,3-d]pyrimidine antifolates(1), J. Med. Chem. 49 (2006) 1055-1065. [34] A. Gangjee, W. Li, J. Yang, R. L. Kisliuk, Design, Synthesis, and Biological Evaluation of Classical and Nonclassical 2-Amino-4-oxo-5-substituted-6-methylpyrrolo[3,2-d]pyrimidines as Dual Thymidylate Synthase and Dihydrofolate Reductase Inhibitors, J. Med. Chem. 51 (2008) 68-76. [ 35 ] A. Gangjee, Y. Qiu, W. Li, R. L. Kisliuk, Potent Dual Thymidylate Synthase and Dihydrofolate Reductase Inhibitors: Classical and Nonclassical 2-Amino-4-oxo-5-arylthiosubstituted-6-methylthieno[2,3-d ]pyrimidine Antifolates, J. Med. Chem. 51 (2008) 5789-5797. [36] Y. G. Assaraf, Molecular Basis of Antifolate Resistance, Cancer. Metastasis. Rev. 26 (2007) 153-181. [37] L. H. Matherly, Z. Hou, Structure and Function of the Reduced Folate Carrier a Paradigm of a Major Facilitator Superfamily Mammalian Nutrient Transporter, Vitam. Horm. 79 (2008) 145-184. [38] A. Gangjee, W. Li, R. L. Kisliuk, V. Cody, J. Pace, J. Piraino, J. Makin, Design, synthesis, and X-ray crystal structure of classical and nonclassical 2-amino-4-oxo-5-substituted-6ethylthieno[2,3-d]pyrimidines as dual thymidylate synthase and dihydrofolate reductase inhibitors and as potential antitumor agents, J. Med. Chem. 52 (2009) 4892-4902. [39] M. Arooj, S. Sakkiah, G. p. Cao, K. W. Lee, An innovative strategy for dual inhibitor design and its application in dual inhibition of human thymidylate synthase and dihydrofolate reductase enzymes, PLoS. One. 8 (2013) e60470. [40] J. W. Mueller, L. C. Gilligan, J. Idkowiak, W. Arlt, P. A. Foster, The Regulation of Steroid Action by Sulfation and Desulfation, Endocr. Rev, 36 (2015) 526-563. [41] L. W. Woo, O. B. Sutcliffe, C. Bubert, A. Grasso, S. K. Chander, A. Purohit, M. J. Reed, B. V. Potter, First Dual Aromatase-Steroid Sulfatase Inhibitors, J. Med. Chem. 46 (2003) 3193-3196. [42] P. M. Wood, L. W. Woo, A. Humphreys, S. K. Chander, A. Purohit, M. J. Reed, B. V. Potter , A letrozole-based dual aromatase–sulphatase inhibitor with in vivo activity, J. Steroid. Biochem. Mol. Biol. 94 (2005) 123-130. [43] L. W. Woo, C. Bubert, O. B. Sutcliffe, A. Smith, S. K. Chander, M. F. Mahon, A. Purohit, M. J. Reed, B. V. Potter, Dual Aromatase-Steroid Sulfatase Inhibitors, J. Med. Chem. 50 (2007) 3540-3560. [44] T. Jackson, L. W. Woo, M. N. Trusselle, S. K. Chander, A. Purohit, M. J. Reed, B. V. Potter, Dual aromatase–sulfatase inhibitors based on the anastrozole template: synthesis, in vitro SAR, molecular modelling and in vivo activity, Org. Biomol. Chem. 5 (2007) 2940-2952. [45] P. M. Wood, L. W. Woo, M. P. Thomas, M. F. Mahon, A. Purohit, B. V. Potter, Aromatase and dual aromatase-steroid sulfatase inhibitors from the letrozole and vorozole templates, Chem. Med. Chem. 6 (2011) 1423-1438. [46] L. W. Woo, P. M. Wood, C. Bubert, M. P. Thomas, A. Purohit, B. V. Potter, Synthesis and Structure–Activity Relationship Studies of Derivatives of the Dual Aromatase–Sulfatase Inhibitor 4-{[(4-Cyanophenyl)(4H-1,2,4-triazol-4-yl)amino]methyl}phenyl sulfamate, Chem. Med. Chem. 8 (2013) 779-799. 17

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

[47] R. Giet, R. Uzbekov, F. Cubizolles, Guellec. K. Le, C. Prigent, The Xenopus laevis aurora-related protein kinase pEg2 associates with and phosphorylates the kinesin-related protein XlEg5, J. Biol. Chem. 274 (1999) 15005-15013. [48] R. G. Fu, Q. D. You, L. Yang, W. T. Wu, C. Jiang, X. L. Xu, Design, synthesis and bioevaluation of dihydropyrazolo[3,4-b]pyridine and benzo[4,5]imidazo[1,2-a]pyrimidine compounds as dual KSP and Aurora-A kinase inhibitors for anti-cancer agents, Bioorg. Med. Chem. 18 (2010) 8035-8043. [49]J. B, Vermorken, S. Rottey, E. Ehrnrooth, K. Pelling, A. Lahogue, S. Wind, J. P. Machiels, A phase Ib, open-label study to assess the safety of continuous oral treatment with afatinib in combination with two chemotherapy regimens: cisplatin plus paclitaxel and cisplatin plus 5-fluorouracil, in patients with advanced solid tumors, Ann. Oncol. 24 (2013) 1392-1400. [50] X. Zhang, S. Raghavan, M. Ihnat, J. E. Thorpe, B. C. Disch, A. Bastian, L. C. Bailey-Downs, N. F. Dybdal-Hargreaves, C. C. Rohena, E. Hamel, S. L. Mooberry, A. Gangjee, The Design and Discovery of Water Soluble 4-Substituted-2,6-dimethylfuro[2,3-d]pyrimidines as Multitargeted Receptor Tyrosine Kinase Inhibitors and Microtubule Targeting Antitumor Agents, Bioorg. Med. Chem. 22 (2014) 3753-3772. [51] X. Zhang, S. Raghavan, M. Ihnat, E. Hamel, C. Zammiello, A. Bastian, S. L. Mooberry, A. Gangjee, The design, synthesis and biological evaluation of conformationally restricted 4-substituted-2,6-dimethylfuro[2,3-d]pyrimidines as multi-targeted receptor tyrosine kinase and microtubule inhibitors as potential antitumor agents, Bioorg. Med. Chem. 23 (2015) 2408-2423. [52] Z. Tang, S. Niu, F. Liu, K. Lao, J. Miao, J. Ji, X. Wang, M. Yan, L. Zhang, Q. You, H. Xiao, H. Xiang, Synthesis and biological evaluation of 2,3-diaryl isoquinolinone derivatives as anti-breast cancer agents targeting ERα and VEGFR-2, Bioorg. Med. Chem. Lett. 24 (2014) 2129-2133. [53] Z. Tang, C. Wu, T. Wang, K. Lao, Y. Wang, L. Liu, M. Muyaba, P. Xu, C. He, G. Luo, Z. Qian, S. Niu, L. Wang, Y. Wang, H. Xiao, Q. You, H. Xiang, Design, synthesis and evaluation of 6-aryl-indenoisoquinolone derivatives dual targeting ERα and VEGFR-2 as anti-breast cancer agents, Eur. J. Med. Chem. 118 (2016) 328-339. [54] X. Cai, H. X. Zhai, J. Wang, J. Forrester, H. Qu, L. Yin, C. J. Lai, R. Bao, C. Qian, Discovery of 7-(4-(3-ethynylphenylamino)-7-methoxyquinazolin-6-yloxy)-N-hydroxyheptanamide (CUDC-101) as a potent multi-acting HDAC, EGFR, and HER2 inhibitor for the treatment of cancer, J. Med. Chem. 53 (2010) 2000-2009. [55] C. Qian, C. J. Lai, R. Bao, D. G. Wang, J. Wang, G. X. Xu, R. Atoyan, H. Qu, L. Yin, M. Samson, B. Zifcak, A. W. Ma, S. DellaRocca, M. Borek, H. X. Zhai, X. Cai, M. Voi, Cancer network disruption by a single molecule inhibitor targeting both histone deacetylase activity and phosphatidylinositol 3-kinase signaling, Clin. Cancer. Res. 18 (2012) 4104-4113. [56] A. Younes, J. G. Berdeja, M. R. Patel, I. Flinn, J. F. Gerecitano, S. S. Neelapu, K. R. Kelly, A. R. Copeland, A. Akins, M. S. Clancy, L. Gong, J. Wang, A. Ma, J. L. Viner, Y. Oki, Safety, tolerability, and preliminary activity of CUDC-907, a first-in-class, oral, dual inhibitor of HDAC and PI3K, in patients with relapsed or refractory lymphoma or multiple myeloma: an open-label, dose-escalation, phase 1 trial, Lancet. Oncol. 17 (2016) 622-631. [57] E. G. Yang, N. Mustafa, E. C. Tan, A. Poulsen, P. M Ramanujulu, W. J. Chng, J. J. Yen, B. W. Dymock, Design and Synthesis of Janus Kinase 2 (JAK2) and Histone Deacetlyase (HDAC) Bispecific Inhibitors Based on Pacritinib and Evidence of Dual Pathway Inhibition in Hematological Cell Lines, J. Med. Chem. 59 (2016) 8233-8262. [58] M. Purwin, J. Hernández-Toribio, C. Coderch, R. Panchuk, N. Skorokhyd, K. Filipiak, B. de Pascual-Teresa,A. Ramos. Design and synthesis of novel dual-target agents for HDAC1 and CK2 inhibition, RSC. Adv. 6 (2016) 66595-6608. [59] C. Q. Ning, C. Lu, L. Hu, Y. J. Bi, L. Yao, Y. J. He, L. F. Liu, X. Y. Liu, N. F. Yu, Macrocyclic compounds as anti-cancer agents: design and synthesis of multi-acting inhibitors against HDAC, FLT3 and JAK2, Eur. J. Med. Chem. 95 (2015) 104-115. [60] T.P. Ahern, T.L. Lash, P. Damkier, P. M. Christiansen, D.P. Cronin-Fenton, Statins and breast cancer prognosis: evidence and opportunities, Lancet. Oncol.15 (2014) e461-e468. [61] Y. Gan, J. Wang, J. Coselli, X. L. Wang, Synergistic induction of apoptosis by HMG-CoA reductase inhibitor and histone deacetylases inhibitor in HeLa cells, Biochem. Biophys. Res. Commun. 365 (2008) 386-392. [62] M.A. Pereira, B.M. Warner, T.J. Knobloch, C.M. Weghorst, R.A .Lubet, V.E. Steele, B.C. 18

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Casto, Chemoprevention of mouse lung and colon tumors by suberoylanilide hydroxamic acid and atorvastatin, Int. J. Cancer. 131 (2012) 1277-1286. [63] J.B. Chen, T.R. Chern, T.T. Wei, C.C. Chen, J.H. Lin, J.M. Fang, Design and synthesis of dual-action inhibitors targeting histone deacetylases and 3-hydroxy-3-methylglutaryl coenzyme A reductase for cancer treatment, J. Med. Chem. 56 (2013) 3645-3655. [64] B. Sarcar, S. Kahali, P. Chinnaiyan. Vorinostat enhances the cytotoxic effects of the topoisomerase I inhibitor SN38 in glioblastoma cell lines, J. Neurooncol. 99 (2010) 201-207. [65] W. Guerrant, V. Patil, J.C. Canzoneri, A.K. Oyelere, Dual targeting of histone deacetylase and topoisomerase II with novel bifunctional inhibitors, J. Med. Chem. 55 (2012) 1465-1477. [66] W. Guerrant, V. Patil, J.C. Canzoneri, L.P. Yao, R. Hood, A.K. Oyelere, Dual-acting histone deacetylase-topoisomerase I inhibitors, Bioorg. Med. Chem. Lett. 23 (2013) 3283-3287. [67] S. Salerno, Settimo. F. Da, S. Taliani, F. Simorini, Motta. C. La, G. Fornaciari, A.M. Marini, Recent advances in the development of dual topoisomerase I and II inhibitors as anticancer drugs, Curr. Med. Chem. 17 (2010) 4270-4290. [68] G. Dong, C. Sheng, S. Wang, Z. Miao, J. Yao, W. Zhang, Selection of evodiamine as a novel topoisomerase I inhibitor by structure-based virtual screening and hit optimization of evodiamine derivatives as antitumor agents, J. Med. Chem. 53 (2010) 7521-7531. [69] G. Dong, S. Wang, Z. Miao, J. Yao, Y. Zhang, Z. Guo, W. Zhang, C. Sheng. New tricks for an old natural product: discovery of highly potent evodiamine derivatives as novel antitumor agents by systemic structure-activity relationship analysis and biological evaluations, J. Med. Chem. 55 (2012) 7593-7613. [70] S. He, G. Dong, Z. Wang, W. Chen, Y. Huang, Z. Li, Y. Jiang, N. Liu, J. Yao, Z. Miao, W. Zhang, C. Sheng, Discovery of Novel Multiacting Topoisomerase I/II and Histone Deacetylase Inhibitors, ACS Med. Chem. Lett. 6 (2015) 239-243. [71] V. Zuco, M. De Cesare, R.Cincinelli, R. Nannei, C. Pisano, N. Zaffaroni, F. Zunino, Synergistic antitumor effects of novel HDAC inhibitors and paclitaxel in vitro and in vivo, PLoS. One. 6 (2011) e29085. [72] X. Zhang, J. Zhang, L. Tong, Y. Luo, M. Su, Y. Zang, J. Li, W. Lu, Y. Chen, The discovery of colchicine-SAHA hybrids as a new class of antitumor agents, Bioorg. Med. Chem. 21 (2013) 3240-3244. [73] X. Zhang, Y. Kong, J. Zhang, M. Su, Y. Zhou, Y. Zang, J. Li, Y. Chen, Y. Fang, X. Zhang, W. Lu. Design, synthesis and biological evaluation of colchicine derivatives as novel tubulin and histone deacetylase dual inhibitors, Eur. J. Med. Chem. 95 (2015) 127-135. [74] W. Li, W. Yue, H. Wang, B. Lai, X. Yang, C. Zhang, Y1. Wang, M. Gu, Cyclooxygenase-2 is associated with malignant phenotypes in human lung cancer, Oncol. Lett. 12 (2016) 3836-3844. [75] S.M. Fischer, E.T. Hawk, R.A Lubet, Coxibs and other nonsteroidal anti-inflammatory drugs in animal models of cancer chemoprevention, Cancer. Prev. Res (Phila). 4 (2011) 1728-1735. [76] B. Liu, S. Yan, L. Qu, J. Zhu, Celecoxib enhances anticancer effect of cisplatin and induces anoikis in osteosarcoma via PI3K/Akt pathway, Cancer. Cell. Int. 17 (2017) 1. [77] O. Peulen, A. Gonzalez, P. Peixoto, A. Turtoi, D. Mottet, P. Delvenne, V. Castronovo, The anti-tumor effect of HDAC inhibition in a human pancreas cancer model is significantly improved by the simultaneous inhibition of cyclooxygenase 2, PloS. One. 8 (2013) e75102. [78] I. Raji, F. Yadudu, E. Janeira, S. Fathi, L. Szymczak, J.R. Kornacki, K. Komatsu, J.D. Li, M. Mrksich, A.K. Oyelere, Bifunctional conjugates with potent inhibitory activity towards cyclooxygenase and histone deacetylase, Bioorg. Med. Chem. 25 (2017) 1202-1218. [79] S. Légaré, M. Basik, Minireview: The Link Between ERα Corepressors and Histone Deacetylases in Tamoxifen Resistance in Breast Cancer, Mol. Endocrinol. 30 (2016) 965-976. [80] B. E. Gryder, M. K. Rood, K. A. Johnson, V. Patil, E. D. Raftery, L. P. Yao, M. Rice, B. Azizi, D. F. Doyle, A. K. Oyelere, Histone deacetylase inhibitors equipped with estrogen receptor modulation activity, J. Med. Chem. 56 (2013) 5782-5796. [81] H. K. Patel, M. I. Siklos, H. Abdelkarim, E.L. Mendonca, A. Vaidya, P. A. Petukhov, G. R. Thatcher, A Chimeric SERM–Histone Deacetylase Inhibitor Approach to Breast Cancer Therapy, Chem. Med. Chem. 9 (2014) 602-613. [82] R. Mendoza-Sanchez, D. Cotnoir-White, J. Kulpa, I. Jutras, J. Pottel, N. Moitessier, S. Mader, J. L. Gleason, Design, synthesis and evaluation of antiestrogen and histone deacetylase inhibitor molecular hybrids, Bioorg. Med. Chem. 23 (2015) 7597-7606. 19

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

[83] A. Speck-Planche, V. V. Kleandrova, F. Luan, M. N. D. S. Cordeiro, Unified multi-target approach for the rational in silico design of anti-bladder cancer agents, Anticancer Agents Med. Chem. 13 (2013) 791-800. [84] A. Speck-Planche, V. V. Kleandrova, F. Luan, M. N. D. S. Cordeiro, Chemoinformatics in anti-cancer chemotherapy: multi-target QSAR model for the in silico discovery of anti-breast cancer agents, Eur. J. Pharm. Sci. 47 (2012) 273-279. [85]A. Speck-Planche, V. V. Kleandrova, F. Luan, M. N. D. S. Cordeiro, Rational drug design for anti-cancer chemotherapy: multi-target QSAR models for the in silico discovery of anti-colorectal cancer agents, Bioorg. Med. Chem. 20 (2012) 4848-4855.

20

ACCEPTED MANUSCRIPT

Figures

AC C

EP

TE D

M AN U

SC

RI PT

Fig. 1. The multi-pharmacophore model can be divided into four distinct types: conjugated-pharmacophore; cleavable linker conjugated-pharmacophore; fused-pharmacophore; merged-pharmacophore. Abbreviation: P, pharmacophore. Fig. 2. The principle of combi-molecules combi-targeting Fig. 3. Design of multi-targeted agents targeting EGFR and Src kinase. Fig. 4. The classical and nonclassical dual TS and DHFR inhibitors and binding mode of classical TS and DHFR inhibitors. Fig. 5. Design of multi-targeted agents targeting aromatase and steroid sulfatase. Fig. 6. Design of multi-targeted agents targeting KSP and aurora-A kinase. Fig. 7. Design of multi-targeted agents targeting antitubulin and receptor tyrosine kinase. Fig. 8. Design of multi-targeted agents targeting anti-estrogens and VEGFR-2. Fig. 9. Design of multi-targeted agents targeting HDAC and protein kinases. Fig. 10. Design of multi-targeted agents targeting HMGR and HDAC. Fig. 11. Design of multi-targeted agents targeting topoisomerase and HDAC. Fig. 12. Design of multi-targeted agents targeting tubulin and HDAC. Fig. 13. Design of multi-targeted agents targeting COX-2 and HDAC. Fig. 14. Design of multi-targeted agents targeting Estrogen receptor α and HDAC.

21

ACCEPTED MANUSCRIPT

P2

P1

P1

P2

P2

P1

linker

P1 P2

conjugated-pharmacophore

cleavable linker conjugated-pharmacophore

RI PT

linker

fused-pharmacophore

merged-pharmacophore

Increasing degree of overlap of two pharmacophores, P1 and P2

AC C

EP

TE D

M AN U

SC

Fig. 1

22

AC C

EP

TE D

M AN U

Fig. 2

SC

RI PT

ACCEPTED MANUSCRIPT

23

AC C

EP

TE D

M AN U

SC

Fig. 3

RI PT

ACCEPTED MANUSCRIPT

24

ACCEPTED MANUSCRIPT O

O

COOH

O

N

N N

methotrexate

O

N

H2N

COOH

HN

N

NH2 N

NH2

H2N

O

N

the binding mode of classical dual TS and DHFR inhibitors

N S CH3

HN H2N

SC

O

6-5 fused ring scaffold , classical pharmacophore of dual TS and DHFR inhibitors

N

S

Ar

O

HN H2N

S

HN N

N H

CH3

, H2N

Ar CH3

N

S

O HN , H2N

M AN U

Nolatrexed nonclassical TS inhibitor

O

S

Ar

H N Ar

NH2

HN

N

N H

COOH

HN

2,4-diamino mode DHFR binding mode

COOH

1

N H

pemetrexed dual TS and DHFR inhibitor

rotatution 180 °

O

N

COOH

O

N H2N

COOH

2-amino-4-Oxo mode TS binding mode

CH3 N

NH2

H2N

methotrexate DHFR inhibitor

raltitrexed TS inhibitor

O

HN

N CH3

COOH H2N

O

COOH N H

RI PT

H2N

N H

S N CH3

HN

NH2

CH3 , H2N

N

nonclassical dual inhibitors of TS and DHFR

AC C

EP

TE D

Fig. 4

25

O

O

, H2N

Ar

N

HN

N

CH3

AC C

EP

TE D

M AN U

Fig. 5

SC

RI PT

ACCEPTED MANUSCRIPT

26

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Fig. 6

27

AC C

EP

TE D

M AN U

Fig. 7

SC

RI PT

ACCEPTED MANUSCRIPT

28

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Fig. 8

29

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Fig. 9

30

ACCEPTED MANUSCRIPT O

H N

N H

O

OH H3C

SAHA HDAC inhbitor

O

H3C merged-pharmacophore H3C

OH OH O

H

H3C

N H

O

H3C

13

OH OH O

Hydrophobic region

OH CH3

Spacer

Zn2+ binder

Dual HMGR and HDAC inhibitor HMGR IC50 = 16.8 nM HDAC1 IC50 = 64.8 nM

Lovastatin (active form) HMGR inhbitor

AC C

EP

TE D

M AN U

Fig. 10

SC

H

RI PT

CH3

H3C

OH

31

RI PT

M AN U

SC

5

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Fig. 11

32

AC C

EP

TE D

M AN U

SC

Fig. 12

RI PT

ACCEPTED MANUSCRIPT

33

ACCEPTED MANUSCRIPT

O

H N

N H

O

OH

Vorinostat HDAC inhibitor

O O H2N S

Cl O

CF3

O

HO N H

5

O O N S H

O

N H

N N

O S CH 3

H HO N 6

O

N N N

Cl

O

O

HO CF3

6

H N

N

N N

N

O H3C

CH3

20 Dual-acting HDAC-COX-2 inhibitor HDAC1 IC50 = 1.03 M HDAC3 IC50 = 0.42 M HDAC6 IC50 = 0.066 M COX-2 IC50 = 0.3 M

O

H3C O Indomethacin COX inhibitor

Celecoxib COX-2 inhibitor

CH3 O CH3

SC

H3C

CH3 OH

RI PT

N

N N

H3C O

H3C O

22 Dual-acting HDAC-COX-2 inhibitor HDAC1 IC50 = 0.25 M HDAC3 IC50 = 0.18 M HDAC6 IC50 = 0.005 M COX-2 IC50 = 4.58 M

M AN U

21 Dual-acting HDAC-COX-2 inhibitor HDAC1 IC50 = 0.26 M HDAC3 IC50 = 0.33 M HDAC6 IC50 = 0.05 M COX-2 IC50 = 4.44 M

O

AC C

EP

TE D

Fig. 13

34

CH3 O

N H

O

N OH H 23 Dual-acting HDAC-COX-2 inhibitor M HDAC1 IC50 = 0.98 HDAC3 IC50 = 0.35 M HDAC6 IC50 = 0.049 M COX-2 IC50 = 0.33 M 5

AC C

EP

TE D

M AN U

Fig. 14

SC

RI PT

ACCEPTED MANUSCRIPT

35

ACCEPTED MANUSCRIPT Table

AC C

EP

TE D

M AN U

SC

RI PT

Table. (1). The Advantage and Disadvantage of Combinations of targeted agents and multi-targeted agents

36

ACCEPTED MANUSCRIPT Advantage

Disadvantage

1) improved efficacy and reduced side effects 2) individual targets can be validated 3) increased capability to achieve maximal target inhibition

1) high costs 2) dose-limiting toxicities and unpredictable drug-drug interactions 3)different pharmacokinetics and pharmacologydynamics to combine

multi-targeted agents

1) a lower risk of drug-drug interactions 2) broadly effective in some clinical settings 3) more convenient and less complex for the patient

1) difficulty in optimizing the ratio of activities at the different targets 2) difficult to attribute activity to any one molecular target.

AC C

EP

TE D

SC

M AN U

Table. 1

RI PT

Combinations of single targeted agents

37

ACCEPTED MANUSCRIPT Highlights

AC C

EP

TE D

M AN U

SC

RI PT

Multi-targeted agents contain privileged substructures and pharmacophores that are relevant to multiple targets. More and more of all marketed medicines can be classified as multi-targeted agents The commonness of multi-targeted agents is multi-pharmacophore mode