Nermin S. Ahmed1 | Jannette Wober2

Abstract
Tamoxifen (TAM) is currently the endocrine treatment of choice for all stages of breast cancer; it has proven success in ER positive and ER negative patients. TAM is activated by endogenous CYP450 enzymes to the more biologically active metabolites 4-hydroxytamoxifen and endoxifen mainly via CYP2D6 and CYP3A4/5. CYP2D6 has been investigated for polymorphism; there is a large interindividual variation in the enzyme activity, this drastically effects clinical outcomes of tamoxifen treatment. Here in we report the design and synthesis of 10 novel compounds bearing a modified tamoxifen skeleton, ring C is substituted with different ester groups to bypass the CYP2D6 enzyme metabolism and employ esterase enzymes for activation. All compounds endorse flexibil- ity on ring A. Compounds (II-X) showed MCF-7% growth inhibition >50% at a screening dose of 10 μM. These results were validated by yeast estrogen screen (YES) and E-Screen assay combined with XTT assay. Compound II (E/Z 4-[1-4-(3-Dimethylamino- propoxy)-phenyl)-3-(4-methoxy-phenyl)-2-methyl-propenyl]-phenol) showed nanomolar antiestrogenic activity (IC50 = 510 nM in YES assay) and was five times more potent in inhibiting the growth of MCF-7 BUS (IC50 = 96 nM) compared to TAM (IC50 = 503 nM). Esterified analogues VI, VII were three times more active than TAM on MCF-7 BUS (IC50 = 167 nM). Novel analogues are prodrugs that can ensure equal clinical outcomes to all breast cancer patients.

KEYWORDS:Hydroxy-Tamoxifen, Carboxylesterases, CYP2D6, E-Screen, MCF-7, polymorphism, tamoxifen, triphenylethylene, XTT assay, YES assay


1 | INTRODUCTION
Tamoxifen (TAM) 1 is a nonsteroidal triphenylethylene derivative; it was considered the first selective estrogen receptor modulator (SERM). SERMs are characterized by its altered pharmacology in various tissues. They can work as antagonist in breast tissues and as agonists in uterine and bone tissues. SERMs are a highly versatile group for the treatment of different conditions associated with postmenopausal women’s health, including hormone responsive cancer, via binding to both estrogen recep- tor subtypes ERα or ERβ. The plasticity of the estrogen receptor-binding pocket allows various ligands to induce different conformations of the internal binding cavity; this can result in differential binding of various cofactors resulting in agonistic or antagonistic conformation (Agnusdei & Iori, 2000; Hillisch et al., 2004; Miller, 2002).
TAM is considered the first-line endocrine therapeutic agent used for treating ER-positive breast cancers. Moreover, it reduces the risk of recur- rence and death when used as adjuvant therapy in early stage or in meta- static cancer (Jordan, 2003). In addition to endocrine therapy, chemotherapy is used for treatment of breast cancer. It is usually rec- ommended postoperative or in case of metastasis. Taxanes, anthracyclines,5-fluorouracil, and cyclophospahmides are common agents used alone or in combination with hormonal therapy. (Bines, Earl, Buzaid, & Saad, 2014) Tamoxifen has mediated its actions via multiple targets other than ERα and ERβ.

Their action extended to include: inhibition of protein Kinase C, metalloproteinases, interaction with multidrug resistance-associated pro- teins.(Bogush et al., 2012) Tamoxifen also inhibits topoisomerases and mitochondrial DNA synthesis and progressively depletes hepatic mitochon- drial DNA in vivo. (Larosche etal., 2007) Targeting topisomerase (Top) as a target for breast cancer has attracted more attention, novel Top1 inhibitors based on the thiohydantoin scaffold proved success in inhibiting the growth of MCF-7 breast cancer cell lines.(Majumdar et al., 2015) One of the most challenging issue with tamoxifen use is the development of resis- tance in an initially responsive breast tumor. Although the molecular mech- anisms underlying resistance to tamoxifen remains unclear, various mechanisms have been proposed; for example, differential metabolic acti- vation of tamoxifen, loss of ER function/expression, alterations in crosstalk between ER and growth factor-mediated signaling pathways, the presence of ER-negative cancer stem cells, and dynamic responses to oxidative stress. Mechanistic understanding of tamoxifen resistance will expand our knowledge on devising new therapy regimens and benefit the breast can- cer patients. Currently both Aromatase inhibitors and selective estrogen receptor degraders (SERD) are evaluated as the extended adjuvant therapy in ER-positive breast cancer with tamoxifen resistance. (Chang, 2012)

More recent work have suggested molecules bearing a chromene scaffold as potent molecules that successfully regulate apoptosis in part via modulating calcium homeostasis with ER calcium transporters. The novel chromene molecules showed nanomolar potency against multiple drug resistance (MDR) cancer cell lines and are of potential clinical application. (Bian et al., 2018) TAM is metabolized to 4-hydroxylated derivatives such as 4-hydroxytamoxifen (4-OHTAM) 2 and endoxifen 3. As compared to the parent drug, those metabolites are almost of 100-fold greater affinity to the estrogen receptor (Lim, Desta, Flockhart, & Skaar, 2005). This metabolism is mainly mediated via cytochrome P450 (CYP) enzymes, specifically the CYP2D6 and CYP3A4 isoforms (Crewe, Notley, Wunsch, Lennard, & Gillam, 2002). TAM is considered as a personalized medicine; this is attributed to the contribution of the polymorphic CYP2D6 enzyme in its metabolism. Patients with inactive CYP2D6 alleles fails to convert the less active TAM to its active metabolites 4-OHTAM and endoxifen. This provides variations in plasma concentrations of active metabolites among TAM-treated women of different populations, and hence the consequent clinical outcome is not equal in all females (Brauch et al., 2013; de Souza & Olopade, 2011; Ramón y Cajal et al., 2010).

In 1990, TAT-59 4, a phosphate ester analogue of TAM was developed;it was a prodrug with better oral availability. Miproxifene, active metabolite of TAT-59, was more potent than TAM in inhibiting the growth of MCF-7, an ER positive breast cancer cell line (Toko et al., 1990). TAT-59 was discontinued in 1999 for two major reasons; they did not optimize the absorptive drug flux of the parent drug and did not provide any biopharmaceutical advan- tage. (Heimbach et al., 2003) The second reason was its failure to provide better pharmacological response compared to TAM. (Shibata et al., 2000).In 2006, Meegan et al. worked on flexible estrogen antagonist bear- ing a pivalolyl esters to provide prodrugs with better bioavailability (M. Meegan et al., 2006). Our research group succeeded in preparing esterified TAM analogues with higher ERα binding affinity and MCF-7 growth inhibition,the new compounds successfully bypassed CYP2D6 metabolism (Ahmed et al., 2016; Elghazawy et al., 2016).Herein, we intend to design, synthesize, and investigate novel TAM analogues, whose metabolites are produced via cleavage of esters. This was implemented to alter the metabolic pathway, so that the active hydroxylated metabolite is produced via carboxylesterases rather than polymorphic CYP2D6 enzymes. Ring C of TAM bears dif- ferent alkyl esters; this alters the novel compounds pharmacokinetic properties.

Structural modifications to TAM also include variations in the alkylaminoalkoxy side chain, which plays a crucial rule in the antiestrogenic action of SERMs. This employs the utilization of dimethylamino propoxy and piperidinyl ethoxy side chains on ring B, substituents differ in terms of length, rigidity, and alteration of basicity of nitrogen atoms. In addition, the impact of varying electron density on ring A was explored by introducing an electron donating methoxy group in the para position. All compounds were designed to endorse flexibility to the rigid triphenylethylene backbone of TAM through introduction of a benzylic methylene spacer between ring A and the ethylene group. Finally, the ethyl vinylic substituent of TAM was changed to a methyl group. The contribution of structural features in inducing an estrogenic/antiestrogenic ER binding conformation was studied. Compounds are depicted in the synthetic scheme.The novel analogues were tested for their relative antiestrogenic activity in a recombinant yeast estrogen screen assay (YES assay). The compounds were further tested for its in vitro antitumor activity against 60 human tumor cell lines by the National Cancer Institute (NCI), compounds showing percent growth inhibition on MCF-7 human breast cancer cell line ≥ 50%, were tested in an E-Screen/XTT assay using most responsive estrogen-sensitive MCF-7 BUS cell sub- line to determine their IC50 values.

YES assay is a gene reporter assay where DNA sequence of human ERα is integrated into the yeast genome completed with an expression plasmids carrying estrogen response elements (ERE) in the promoter controlling the expression of the reporter gene Lac-Z (encoding the enzyme β-galactosidase). In the presence of estro- genic compounds, enzyme is synthesized and secreted into the medium, where it converts the chromogenic substrate chlorophenol red-β-D-galactopyranoside (CRPG) from a yellow to a red product, whose absorbance is measured. Agonistic activity is measured directly whereas antagonistic activity is measured in terms of reduc- tion in color formation in presence of 0.5 nM E2 (Routledge & Sumpter, 1996). YES assay is sensitive, time saving and easily han- dled yet there are some limitations that should be considered when tackling results of the assay. The polygenetic differences among yeast cells may alter the metabolism of the tested compounds. Addi- tionally, the cell wall and chemical transport system can decrease intracellular levels of tested compounds (Coldham et al., 1997). Sev- eral studies investigated the ability of several yeast strains to hydrolyse esters. Results showed that the esterase activity was localized mainly to the cytosol and considerable differences in activ- ity were observed between various yeast strains. The phase of growth also contributed to the variation in esterase activity of the yeast. This diversity is taken in consideration when evaluating the novel compounds (Degrassi, Uotila, Klima, & Venturi, 1999; Kwolek- Mirek, Bednarska, Zadrąg-Tecza, & Bartosz, 2011).
The E-Screen, developed to determine estrogenic properties of a substance, is based on the ability of the MCF-7 BUS cells to prolifer- ate in the presence of estrogens. Compounds, which have ER- antagonistic properties, are able to inhibit this proliferation in the presence of estradiol (E2) (Roehm, Rodgers, Hatfield, & Glasebrook, 1991; Soto et al., 1995).

The resulting cell amount can be determined by a colorimetric assay for cell viability and proliferation.The cleavage of the tetrazolium salt sodium 30 -[1-[(phenylamino)-carbony]-3,- 4-tetrazolium]-bis(4-methoxy-6-nitro)benzene-sulfonic acid hydrate (XTT) by dehydrogenase enzymes of metabolically active cells yields a highly colored formazan product which is water soluble. Bioreduction of XTT is potentiated by addition of electron coupling agent phena- zine methosulfate (PMS) (Roehm et al., 1991; Scudiero et al., 1988).However, this combined E-Screen/XTT assay displays the cell amount after treatment. To understand, which processes cause the obtained results, appropriate molecular endpoints have to be studied, because a decrease of the proliferation or an increase of the apoptosis pathways could affect the inhibition of the cell growth.Results of in silico docking studies aided in investigating the pos- sible interactions of the potent compounds II, IV with specific residues within the human ERα ligand-binding domain to support the biological data provided (Scheme 1).

2 | MATERIALS AND METHODS
Solvents and reagents were obtained from commercial suppliers and were used without further purification. All chemicals were obtained from Sigma-Aldrich and were used without further purification. Col- umn chromatography was carried out using silica-gel 40-60 μM mesh. Reaction progress was monitored by TLC using fluorescent precoated silica gel plates and detection of the components was made by short UV light (λ = 254 nm). 1 H NMR spectra were run at 400 MHz and

SCHEME 1 Synthetic scheme
13C spectra were run at 101 MHz in deuterated chloroform (CDCl3). Chemical shifts (δ) were reported in parts per million (ppm) downfield from TMS; and all coupling constants (J) are given in Hz. Multiplicities are abbreviated as: s: singlet; t: triplet; q: quartet; m: multiplet. The purities of the tested compounds were determined by HPLC coupled with mass spectrometer and were higher than 90% for all compounds. Mass spectra were made on (HPLC-ESI-MS) instrument equipped with an ESI source and a triple quadrupole mass detector (Thermo Fin- nigan). The MS detection was carried out at a spray voltage of 4.2 KV, a nitrogen sheath gas pressure of 4.0105 Pa, an auxiliary gas pressure of 1.0105 Pa, a capillary temperature of 400 οC, capillary voltage of 35 V, and source CID of 10 V. All samples were injected by auto- sampler (Surveyor, Thermo Finnigan) with an injection volume of 10 μl. A RP C18 NUCLEODUR 100−3 (125 mm × 3 mm) column (Macherey-Nagel) was used as stationary phase. The solvent system consisted of water containing 0.1% TFA (A) and 0.1% TFA in acetoni- trile (B). HPLC-method: flow rate 400 μl/min. The percentage of B started at an initial of 5%, was increased up to 100% during 16 min, kept at 100% for 2 min, and flushed back to the 5% in 2 min. All masses were reported as those of the protonated parent ions (M + H)+. All reactions were carried out under argon when inert atmo- sphere was needed.

2.1 | General procedures for the preparation of Compound (I)
Zinc powder (10.11 g,154 mmol) was suspended in dry THF (100 ml), and the mixture was cooled to 0 。C. To the same double necked rounded bottom flask TiCl4(7.5 ml,70 mmol)was added dropwise under nitrogen.When the addition was complete,the mixture was warmed to room temperature and heated to reflux for 2 hr at 70 。C. After cooling down, a solution of 4,40 -dihydroxybenzophenone(2.63 g,12.3 mmol)and 4-Methoxyphenylacetone (6.3 g, 38.4 mmol) in dry THF (100 ml) was added at 0 。C to the same flask and the mixture was heated to reflux at 70 。C in the dark for 2 hr. After being cooled to room tem- perature, the zinc dust was filtered off and THF was evaporated. The residue was dissolved in saturated ammonium chloride aqueous solution (150 ml) and extracted with ethyl acetate (120 ml × 6). The organic layers were combined and dried over Na2SO4, concentrated in vacuo, and further purified using silica gel column chromatogra- phy(99:1 dichloromethane-methanol), the reaction yield was approximately 60%.4,4-[2-(4-methoxybenzyl)prop-1-ene-1,1-diyl] diphenol (I)Yellowish oil, Yield 60%; 1 H-NMR (400 MHz, CDCl3) δ 7.22-7.10 (m, 1H), 7.10-6.91 (m, 5H), 6.91-6.80 (m, 3H), 6.77-6.65 (m, 3H), 5.36-5.22 (m, 2H), 3.79 (q, J = 1.8 Hz, 3H), 3.50-3.33 (m, 2H), 1.72-1.61 (m, 3H).13C-NMR (101 MHz, CDCl3) δ 158.15, 153.43, 137.67, 134.84, 132.71, 130.82, 129.64, 114.99, 113.57, 55.37, 40.08, 20.24. TLC Rf = 0.6 (CH2Cl2/CH3OH 99:1). MS (ESI) [M + H]+ = 347.18.

2.2 | General procedures for the preparation of Compounds (II-III)
A solution of I (47 mmol) in DMF (100 ml) was treated with K2CO3 (19.5 g, 141 mmol) and heated in an oil bath at 80。C. The resulting suspension was treated with the appropriate commercially available base hydrochloride salt namely 3-Dimethylamino-1-propyl chloride hydrochloride and 1-(2-Chloroethyl)piperidine hydrochloride (51 mmol) portion wise over a 2 hr period and stirred for 16 hr. The reaction mixture was cooled to room temperature, quenched with saturated ammonium chloride (300 ml), and extracted with ethyl acetate (4 × 150 ml) (Lv, Liu, Lu, Flockhart, & Cushman, 2013). The combined organic phase was washed with brine (2 × 150 ml), dried, and concentrated. The final product was further purified using silica gel column chromatography (92:8 dichloromethane-methanol) to give (II-III).E/Z 4-[1-4-(3-Dimethylamino-propoxy)-phenyl]- 3-(4-methoxy-phenyl)-[2-methyl-propenyl]-phenol (II)Brown oil, Yield 45%; 1 H-NMR (400 MHz, CDCl3) δ 7.16-6.94 (m, 12H), 6.93-6.67 (m, 12H), 3.94 (dd, J = 11.5, 5.7 Hz,4H), 3.80 (s, 6H), 3.44 (d, J = 8.9 Hz, 4H), 2.91 -2.75 (m, 4H), 2.51 (d, J = 2.4 Hz, 12H), 2.09 (s, 4H), 1.69 (s, 2H), 1.67 (s, 2H). 13C-NMR (101 MHz, CDCl3) δ 157.77, 156.94, 156.82, 155.08, 154.96, 137.98, 136.34, 136.31, 135.02, 132.82, 132.46, 130.93, 130.84, 130.65, 130.59, 129.79, 129.77, 129.52,129.49,115.57,115.20,115.02, 114.38, 113.92,113.76,65.27,56.16,55.27, 44.17, 40.59, 25.93, 19.85, 19.82. TLC Rf = 0.4 (CH2Cl2/CH3OH 92:8). MS (ESI) [M + H]+ = 432.25.

2.3 | General procedures for the preparation of Compounds (IV-X)
Compounds II-III (0.38 mmol) and triethylamine (106 μl, 0.76 mmol) were dissolved in THF (5 ml).The appropriate acid chloride (0.57 mmol) was added and stirring was continued for 30 min at room temperature (M. J.Meegan,Hughes, Lloyd, Williams, & Zisterer, 2001). The reaction mixture was diluted with ethyl acetate (40 ml), quenched with 10% HCl (10 ml), washed with brine, and dried over Na2SO4. The solvent was removed under reduced pressure and the final product was further purified using silica gel column chromatography.E/ZAcetic acid 4-[1-[4-(3-dimethylamino-propoxy)- phenyl]-3-(4-methoxy-phenyl)-2-methyl-propenyl]- phenyl ester (IV)Yellowish brown oil; Yield 65%; 1H-NMR (400 MHz, CDCl3) δ 7.26-7.13 (m, 4H), 7.11 (dd, J = 14.0, 8.5 Hz, 8H), 7.06-6.99 (m, 4H), 6.89-6.73 (m, 8H), 4.04 (q, J = 5.7 Hz, 4H), 3.81 (s, 6H), 3.45 (s, 4H), 2.89-2.80 (m, 4H), 2.55 (d, J = 1.8 Hz, 12H), 2.29 (d, J = 3.9 Hz, 6H), 2.17 (dd, J = 10.5, 4.6 Hz, 4H), 1.69 (d, J = 6.8 Hz, 6H). 13C-NMR (101 MHz, CDCl3) δ 169.48, 169.43, 157.86, 157.16, 157.05, 149.07, 148.93, 140.81, 140.75, 137.47, 135.66, 135.60, 133.76, 133.65, 132.49, 132.45, 130.94, 130.69, 130.67, 130.43, 129.98, 129.68, 129.48, 121.44, 121.07, 120.92, 114.48, 114.04, 113.87, 113.80, 77.35, 77.23, 77.03, 76.71, 65.36, 56.11, 55.26, 44.17, 40.55, 29.70, 26.01, 21.18, 19.87, 19.83. TLC Rf: 0.42 (CH2Cl2/CH3OH 92:8). MS (ESI) 474.6 [M + H]+.E/Z Butyric acid 4-[1-[4-(3-dimethylamino-propoxy)- phenyl]-3-(4-methoxy-phenyl)-2-methyl-propenyl]- phenyl ester (VI)

Brown oil; Yield 54%; 1H-NMR (400 MHz, CDCl3) δ 7.20-7.02 (m,13H), 6.97 (dd, J = 8.5, 2.5 Hz,3H), 6.88-6.74 (m, 8H), 4.03 (dd, J = 11.1, 5.6 Hz, 4H), 3.78 (d, J = 9.1 Hz, 6H), 3.39 (d, J = 6.7 Hz, 4H), 3.29-3.13 (m, 6H), 2.79 (d, J = 2.8 Hz, 12H), 2.56-2.45 (m, 4H), 2.29 (d, J = 29.2 Hz, 6H), 1.80-1.68 (m, 4H), 1.69-1.53 (m, 8H).13C-NMR (101 MHz, CDCl3) δ 172, 157.82, 137.32, 130.92, 130.68, 130.56, 130.34, 129.43, 121.10, 120.94, 113.98, 113.81, 113.76, 77.38, 77.06, 76.74, 64.63, 55.80, 55.23, 45.81, 43.12, 36.21, 24.69, 19.76,
18.42, 13.59, 8.62. TLC Rf: 0.44 (CH2Cl2/CH3OH 92:8). MS (ESI)502.29 [M + H]+.E/Z Hexanoic acid 4-[1-[4-(3-dimethylamino-propoxy)- phenyl]-3-(4-methoxy-phenyl)-2-methyl-propenyl]- phenyl ester (VII)Yellow oil; Yield 58%; 1 H-NMR (400 MHz, CDCl3) 1 H NMR (400 MHz, CDCl3) δ 7.44 -7.36 (m, 2H), 7.12 -7.04 (m, 5H), 7.02 (d, J = 4.2 Hz, 2H), 7.01 -6.97 (m, 4H), 6.92 (dd, J = 8.6, 2.8 Hz, 4H), 6.76 (d, J = 7.9 Hz, 3H), 6.75 -6.69 (m, 4H), 3.98 (dd, J = 11.4, 5.9 Hz, 4H), 3.72 (s, 6H), 3.41 (t, J = 18.9 Hz, 6H), 3.18 -3.08 (m, 4H), 2.75 (t, J = 5.0 Hz, 12H), 2.46 (td, J = 7.5, 3.9 Hz, 4H), 2.27 (dd, J = 14.2, 6.6 Hz, 6H), 1.73 – 1.64 (m, 4H), 1.65 – 1.53 (m, 8H), 1.35 – 1.25 (m, 10H). 13C-NMR (101 MHz, CDCl3) δ 172.38, 157.88, 156.74, 149.05, 140.63, 140.57, 137.37, 136.02, 133.86, 132.39, 131.01, 130.77, 130.63, 130.40, 129.95, 129.47, 128.30, 127.87, 125.82, 121.14, 120.98, 113.99, 113.81, 64.62, 55.94, 55.26, 43.12, 40.53, 34.38, 31.25, 24.68, 24.63, 22.31, 19.85, 19.81, 13.91. TLC Rf: 0.31 (CH2Cl2/CH3OH 92:8). MS (ESI) 530.32 [M + H]+ .


2.4 | In vitro assays
2.4.1 | The recombinant yeast estrogen screen (YES assay)
The yeast estrogen receptor assay was supplied by Dr. J.P. Sumpter (Brunel University, Uxbridge, UK), and was used to determine the rela- tive transactivation activity of the human ERα as formerly described (Routledge & Sumpter, 1996). Briefly, Saccharomyces cerevisiae stably transfected with a human ERα and an estrogen responsive element fused to the reporter gene lacZ encoding for β-galactosidase were treated with the test substances for about 48 hr. The β-galactosidase enzymatic activity was measured in a colorimetric assay using a micro- plate photometer by hydrolysis of the substrate chlorophenol red β-D-galactopyranoside(Roche Diagnostics, Mannheim Germany), which leads to the formation of chlorophenol red. This can be mea- sured as an increased absorption at 540 nm. All compounds were diluted in DMSO. 17β-estradiol (E2) (Sigma, Deissenhofen, Germany) 10 nM was used as positive control and DMSO was used as vehicle control.All compounds, also tamoxifen(TAM) (Biotrend, Cologne, Germany), and 4-hydroxytamoxifen (4-OHTAM) (Sigma,Deissenhofen, Germany), were screened first for antiestrogenic activ- ity in concentration of 1 μM; antiestrogenic assays were performed in combination with 0.5 nM E2. IC50 values were determined by testing a range of seven concentrations (50-10 μM). All compounds were tested in technical quadruplicates and biological triplicates. Statistical analysis was performed by analysis of variance (ANOVA) and Tukey’s post hoc test with the significance level of p < .05.

2.4.2| Determination of cell growth using E-Screen in combination with XTT test
Prior to use in the proliferation assays MCF-7 BUS cells, kindly pro- vided by A. M. Soto and C. Sonnenschein (Tufts University, Boston) were cultured for at least one passage in DMEM high glucose (Sigma, Deissenhofen, Germany), supplemented with 1 mM sodium pyruvate (Gibco, Carlsbad, California) and 5% charcoal stripped fetal calf serum (CSF) (BioWest, Nuaillé, France). Exact experimental details are avail- able in literature (Soto et al., 1995; Villalobos et al., 1995). Briefly, sub- cultured cells of MCF-7 BUS were seeded in 96-well cell culture test plates at a density of 3×103 cells/well and incubated for 24 hr to adhere. Then cell culture medium was removed, 100 μl of fresh medium containing treatment, controls, and DMSO were added, and the cells were incubated again for 6 days. Viability of the MCF-7 BUS cells was measured using the XTT test (Scudiero et al., 1988). This test is based on the mitochondrial activity of the cells and measures the PMS (Sigma, Deissenhofen, Germany) mediated reduction of XTT (Sigma, Deissenhofen, Germany) to its colored formazan derivate, which can be monitored using a microplate reader (infinite F200,TECAN Trading AG, Switzerland) at 492 nm. DMSO (vehicle control), E2 (10 nM, positive control), TAM and 4-OHTAM (1 μM) were used as controls, respectively.Studies were performed as three independent experiments with all blanks, controls, and samples at least Mongolian folk medicine in triplicates. Cell viability was calculated in comparison to the 10 pM E2 treatment (set to 100%) of each plate. Statistical analysis was performed by ANOVA and Tukey’s post hoc test with the significance level of p < .05.

3 | RESULTS AND DISCUSSION
Compounds I was synthesized by standard McMurry coupling reac- tion of 4,40 – dihydroxybenzophenone with 4-methoxyphenylacetone using titanium tetrachloride/zinc as catalysts to give the diphenol I with yield 75% as outlined in the synthetic scheme.(Gauthier, Mailhot, & Labrie, 1996)Compound I was confirmed from its spectral data where its 1H- NMR revealed a signal at 3.79 ppm corresponding to (O CH3) of ring A. Twelve aromatic protons appeared from 6.77-7.22 additionally a signal appeared at 3.33-3.50 ppm corresponding to the benzylic CH2. 13C-NMR revealed a signal at 55.37 ppm corresponding to OCH3 of ring A, a signal at 40.08 ppm which corresponds to the benzylic CH2 and a signal at 20.24 ppm corresponding to the terminal CH3 carbon. The diphenol I was then treated with the appropriate base hydrochlo- ride salts in the presence of potassium carbonate as a part of ether formation by Williamson ether synthesis to yield Compounds II-III. Both a monoalkylated and dialkylated product were obtained in an approximate ratio of 60 to 40%, respectively. Separation using silica gel column chromatography provided the monoalkylated derivatives as E and Z isomers. Attempts to isolate the E and Z isomers using col- umn chromatography were not successful. The E/Z isomeric ratio for the products was assigned using LC/MS as well as relative peak heights in the 1 H-NMR spectrum. Integration of 1H-NMR signals showed double the number of protons present in a single isomer. The monoalkylated compounds were esterified using commercially avail- able acid chlorides in trimethylamine (TEA) to yield Compounds IV-X. 1H-NMR confirmed the success of the reaction by an increase in the integration of aliphatic protons corresponding to the alkyl ester side chain. 13C-NMR showed either one or two signals around 170 ppm corresponding to carbonyl group of the ester.

All compounds were tested for their ability to induce anti- estrogenic activity using YES assay, assays were carried out in the presence of 0.5 nM E2. Table 1 All compounds were screened for their in vitro antitumor activity against 60 human tumor cell lines by the National Cancer Institute (NCI), compounds showing≥50% growth inhibition at 10 μM on MCF-7 cell line were selected for IC50 determination using combination of E-Screen and XTT assay on MCF- 7 BUS breast cancer cell line in presence of 10 pM E2. Table 2.Compound I is a bisphenol derivative with no significant anti- estrogenic activity, the lack of a bulky aminoalkoxy side chain rather induces an estrogenic confirmation. Compound I lacked growth inhibi- tory activity at 10 μM on MCF-7 cell line.Compounds II, III bears an OH group at ring C, a dimethylamino- propoxy and piperidinyl-ethoxy at the para position of ring B, respec- tively. Compound II, III showed a relative antiestrogenic activity of 0.18 and 0.42, respectively. Despite the incorporation of the basic nitrogen in a piperidine ring is expected to increase basicity and improve its anti-estrogenic activity, Compound III was approximately three times less potent than Compound II, this may suggest that the extra methylene spacer in Compound II enhanced the anti-estrogenic activity. This encourages more research toward optimization of the alkyl spacer of ring B.

Compound II was more active than TAM and 10 times less active than its congener 4-OHTAM. The introduction of an electron donating methoxy group at the para position of ring A is supposed to increase the π -π interaction inside the ER binding pocket, whereas an extra benzylic carbon on the ethylene backbone imparts flexibility to the compounds and may allow a better fit. For compound II both flexibility and decreasing lipophilicity of ring A reduce the anti-estrogenic activity of the compound compared to 4-OHTAM.Compounds II, III were esterified at ring C to produce prodrug moities VII-X that can bypass CYP2D6 metabolism and rather recruit carboxyesterases (CE) for their bioactivation. Novel compounds bear methyl, ethyl, propyl, pentyl, and decyl esters. These compounds are designed to be hydrolyzed at different rates. Rate of hydrolysis of ester prodrugs depend mainly on hydrophobic and Immunodeficiency B cell development steric factors of the ester group. (Durrer, Wernly-Chung, Boss, & Testa, 1992)All the esterified compounds showed a relative antiestrogenic activity lower than their hydroxylated congeners. This may be attrib- uted to variation between esterase amounts in the yeast systems

FIGURE 1 ERα LBD Interactions of Compound II with the assay and to the difference in the ability of compounds to pass yeast cell wall. If all ester-bearing compounds were fully metabolized to their hydroxylated congeners, compounds IV-X would have rela- tive antiestrogenic activity equal to Compounds II, III. Therefore, we postulate that the antiestrogenic activity measured is the sum of activities of both intact ester analogues and their metabolites.Compounds III-VII bearing dimethylamino-propoxy side chain on ring B, showed higher relative anti-estrogenic activity compared to Compounds VIII-X bearing piperidinyl-ethoxy side chain.Compound VI bearing a butyrate ester on ring C showed three times higher relative anti-estrogenic activity compared to Compound V bearing an acetate ester, this shows that the size of the ester used has a profound effect on activity and rate of metabolic hydrolysis.All compounds were tested for their growth inhibition on a panel of 60 human cancer cell lines via The National Cancer Institute: cancer drug discovery and development program (Grever, Schepartz,& Chabner, 1992). Compounds II-X showed more than 50% growth inhi- bition on MCF-7 cell line at 10 μM, those compounds were tested for their percent antiproliferation activity on MCF-7 BUS at 1 μM in pres- ence of 10 pM estradiol. Compounds showing antiproliferation activity higher than 50% on MCF-7 BUS were further screened to determine their IC50 values. Compounds II, IV, VI, and VII were more active than TAM (IC50 = 503 nM). Compound II (IC50 = 96 nM) is about five times more potent than TAM whereas Compounds VI, VII showed approxi- mately three times higher antiestrogenic potency (IC50 = 168 nM).There is a clear correlation between anti-estrogenic activity of com- pounds and their growth inhibition potency where Compound II is the most antiestrogenic compound and the most potent on MCF-7 BUS cell line. This correlation may indicate that the growth inhibition effect of the novel compounds occurs mainly via an ER mechanism.

FIGURE 2 Overlay of Compound II (green) on 4-OHTAM (brown)
There has been reports indicating the availability of intracellular CE in MCF-7(Katz,Finlay,Banerjee, & Levitz, 1987). Therefore, esterified analogues can be metabolized intracellularly to produce the active hydroxylated congeners with potent antiproliferation activity. Factors like size, polarity, lipophilicity, and electronic char- acters affect the ability of the compound to cross cell membrane and therefore affects its bioactivation. The difference in amounts of available esterases must be considered when interpreting the cellu- lar assay results.

3.1 | In silico results
The mode of binding of Compounds II (highest ERα antiestrogenic activity) and its acetate ester congener IV was investigated through a

FIGURE 3 ERα LBD Interactions of Compound IV with brief computational docking study using MOE.2009. The crystal struc- ture used in the docking studies was obtained from the cocrystallization of ERα with 4-OHTAM as found in the protein data bank (PDB: 3ERT) (Shiau et al., 1998).The docked geometry for Compound II showed a partial overlay on 4-OHTAM, Compound II successfully formed H-Bond between OH in ring C and Glu353, yet it failed to pick up interaction with Arg 394. Oxygen of the dimethylamino-propoxy substituent in ring B formed a H2O mediated interaction with Thr 347.The incorporation of the additional methylene spacer in case of dimethylamino-propoxy substituent in ring B compared to dimethylamino-ethoxy substituent of 4-OHTAM shifted the distance between Oδ351 and protonated Nlig. The distances of 4.8 Å are measured in 4-OHTAM complex in 3ERT, whereas compound II showed a distance of 6.36 Å. The success of the protonated nitrogen to neutralize the Asp351 is a key element in SERM antiestrogenic activity (Maximov et al., 2014). This might explain the 10-fold decrease in anti-estrogenic activity of Compound II compared to 4-OHTAM. Figures 1 and 2.Introducing flexibility to the rigid skeleton of 4-OHTAM may have contributed to the lower antiestrogenic activity of Compound II. The area accommodating ring A was examined, the extra methy- lene carbon has pushed the phenyl ring into a small lipophilic cav- ity lined with Ile424, Met343, Met421, Leu346, and Met 388, this Overlay of Compound IV (green) on 4-OHTAM lipophilic pocket has been previously defined by Meegan et al. (M. Meegan et al., 2006). The introduction of a methoxy group on ring A was intended to increase electron density on the ring to provide better π -π stacking and pick up H-bonds yet the docking model shows a possible site wall clash labeled with red contour. Figures 1 and 2.Converting the hydroxyl group of Compound II into an ester group did not seem to deteriorate the antiestrogenic activity in YES assay. The ability of yeast cells to fully or partially metabolize the novel drugs during the YES assay is still a question to be answered; this elicits a question if the esterified analogues induce their antiestrogenic effect as intact esters. Our docking model was used to check the possible interaction of the esterified Compound IV inside ERα LBD. Docking results show Com- pound IV formed a H-Bond between the carbonyl of the ester group and Arg394 and it retained the essential cationic interaction with Asp351. A site wall clash is also noted in the small lipophilic cavity-accommodating ring A (Figures 3 and 4).

4 | CONCLUSIONS
We designed and synthesized 10 novel compounds, which bear a flex- ible triphenylethylene backbone. Five of the tested compounds showed higher antiproliferation activity on MCF-7 BUS compared to TAM. Compounds are designed as prodrugs that can be metabolically activated via CE rather than polymorphic CYP2D6, this approach can ensure equal clinical outcomes to breast cancer patients with varia- tions in the CYP2D6 gene resulting in reduced or absent enzyme function, those patients has lower levels of active tamoxifen metabo- lites and reduced treatment efficacy.Identifying cellular targets of our novel compounds is an essential task that help future optimization of our molecules. Recent researches adopted various strategies Bisindolylmaleimide I mw such as affinity chromatography, activity based protein profiling, label-free techniques, expression cloning tech- niques and in silico approaches to achieve this aim.(Anantpadma et al., 2016; Sato, Murata, Shirakawa, & Uesugi, 2010) One recent example is the use of target deconvolution strategy to prove that HDAC2 and Prohibitin 2 as the potential cellular binding targets in MCF-7 for a group of compounds bearing a Spiro[pyrrolidine-3, 30 - oxindole] nucleus. (Hati et al., 2016; Kumar etal., 2017)Our results encourages further research on novel triphenylethylene (TPE) prodrugs with improved pharmacokinetic pro- file, our future work willfocus on the effect of endorsing flexibility to the rigid TPE skeleton and the effect of substitution on ring A and on finding more on the targets of our compounds in addition to ERα .