Design, Synthesis and Mechanistic Study of New Dual Targeting HDAC/Tubulin Inhibitors

Abstract

Aim:

The purpose of this work is to create and synthesize a new class of chemicals: 3-cyano-2-substituted pyridine compounds with expected multitarget inhibition of histone deacetylase (HDAC) and tubulin.

Materials & methods:

The target compounds (3a–c, 4a–c and 5a–c) were synthesized utilizing 6-(4-methoxyphenyl)-2-oxo-4-(3,4,5-trimethoxyphenyl)-3-cyanopyridine, with various linkers and zinc-binding groups (ZBGs).

Results:

Most of the tested compounds showed promising growth inhibition, and hydroxamic acid-containing hybrids possessed higher HDAC inhibition than other ZBGs. Compound 4b possessed the highest potency; however, it showed the most tubulin polymerization inhibition. Docking studies displayed good binding into HDAC1 and six pockets and tubulin polymerization protein.

Conclusion:

Compound 4b could be considered a good antitumor candidate to go further into in vivo and clinical studies.

Graphical abstract

Keywords::

• capping
• cyanopyridine
• docking
• HDAC
• multitargeting
• tubulin

The emergence and development of human cancers are significantly influenced by epigenetic control, which includes DNA methylation and histone acetylation/deacetylation [1]. Malignant tumors are primarily caused by abnormal gene expression introduced by cellular gene mutations [2]. Enzymes that are involved in epigenetic control offer multiple targets for the development of anticancer medications [3]. Two categories of enzymes regulate the acetylation and deacetylation of histones: histone acetyltransferases (HATs) and histone deacetylases (HDACs). Tumor suppressor gene expression can be reduced by histone deacetylation and can be induced by histone hyperacetylation, which leads to apoptosis [4]. HDACs also deacetylate nonhistone proteins including α-tubulin, which are essential for the migration and proliferation of cancer cells [5]. HDAC enzymes are categorized as NAD⁺ dependent (class III) or zinc dependent (class I, II and IV) [6]. Small molecules with zinc-binding groups (ZBGs) can inhibit class I and II HDACs by binding to a Zn²⁺ ion in the active site of the enzyme through a reversible process via mono or bidentate complexation [7]. These ZBGs involve hydroxamates, carboxylates, cyclic peptides and benzamides [7]. There are two additional pharmacophoric features for HDAC inhibitors in addition to ZBGs. These are a surface recognition cap that interacts with the amino acid residues near the active site’s entrance and a linker that joins the ZBG and surface recognition cap [8]. Up until now, five HDAC inhibitors, including vorinostat (SAHA; I), belinostat II, panobinostat III, chidamide IV and romidepsin V, have been approved for the treatment of human cancers (Figure 1A) [9]. However, their clinical application has been constrained by their ineffectiveness against solid tumors [10,11]. Many studies have recently reported using HDAC inhibitors in combination with chemotherapeutic drugs to combat tumors.

Figure 1. Reported agents.

(A) US FDA-approved HDAC inhibitors and chidamide (approved in China). (B) Reported tubulin polymerization inhibitors. (C) Reported dual tubulin polymerization and histone deacetylase inhibitors.

HDAC: Histone deacetylase; ZBG: Zinc binding group.

On the other hand, microtubule dysfunction causes abnormal mitotic spindles to form arrest of the cell cycle in the G2/M phase. Consequently, it is crucial to use drugs in anticancer therapy that control microtubule assembly by impeding either tubulin polymerization or microtubule disassembly [12]. Recently, a number of tubulin inhibitors have been developed that can bind to the colchicine site, destabilize tubulin’s dynamic equilibrium, stop mitosis and initiate apoptosis [13–18]. Among them are synthetic combretastatin A4 (CA-4; VI) and isocomberastatin isoA-4, VII, which bind to the colchicine site causing tubulin polymerization inhibition [19]. In addition, some pyridine-containing derivatives, VIII, have been proven to be inhibitors of β-tubulin polymerization (Figure 1B) [20–22].

According to a variety of studies [23–25], HDAC inhibitor and tubulin inhibitor combination may have synergistic anticancer effects. Therefore, numerous dual HDAC/tubulin inhibitors have been identified [26–28]. Consequently, the first SAHA/colchicine hybrid IX with promising anticancer and HDAC1 inhibitory activities was reported by Zhang et al. [29]; and recently 2,6-diarylpyridine-based hydroxamic acid derivative, X [30], quinoline-2-carbonitrile-based hydroxamic acids, XI [31], as well as 2-methoxyestradiol-based benzamide (2ME2) derivatives have been reported. Compound XII was found to be a potential dual tubulin polymerization and histone deacetylase inhibitor (Figure 1C) [32].

Depending on the aforementioned studies and on the finding that pyridine nucleus displayed an attractive core for the discovery of new HDACs inhibitors [33], confirmed by our previously work [34], we were encouraged to design and synthesize new multitargeting hybridized HDAC inhibitors utilizing 3-cyano-2-substituted pyridine as a capping group. First, we synthesized new series of potential anticancer hybrids (3a–c, 4a–c and 5a–c) with expected dual-tubulin polymerization and HDAC inhibitory activity (Figure 2). This was accomplished by using 6-(4-methoxyphenyl)-2-oxo-4-(3,4,5-trimethoxyphenyl)-3-cyanopyridine, a rigid analogue of CA-4 (I) with tubulin polymerization inhibitory activity being a capping group. In addition, variant linker lengths, either aromatic or aliphatic were designed; the ZBG was suggested to be carboxylic acid, hydroxamic acid or 2-aminoanilide to investigate the impact of linker type, length and ZBG nature on the activity.

Figure 2. Rationale of design of the target tubulin/histone deacetylase inhibitor hybrids.

HDAC: Histone deacetylase; ZBG: Zinc-binding group.

Experimental

Chemistry

All the chemicals bought from Aldrich (MI, USA), Merck (Darmstadt, Germany), Fluka (India), Cambrian Chemicals (Newtown, UK) and El-Nasr Pharmaceutical Chemicals (Egypt) are of commercial quality, and intermediate and target compounds were produced using these chemicals. None of the compounds were further purified before usage. El-Nasr Pharmaceutical Chemicals, Aldrich, Merck, Cambrian Chemicals and Fluka supplied the commercial-grade solvents that were used. Dimethylformamide (DMF) was refined by distillation on a rotary evaporator after being dried with anhydrous sodium sulphate.

Uncorrected melting points were identified using the Stuart electro-thermal melting point equipment. At Assiut University, IR spectra were recorded utilizing the Nicolet iS5 FT-IR spectrometer at Assuit University.

Using TMS as an internal reference, NMR spectra were gathered at the Ain Shams University-Faculty of Pharmacy using a Bruker Advance 400 MHz NMR spectrometer. Chemical shifts (δ) values were collected in parts per million (ppm) in relation to DMSO-d6 (proton: 2.50, carbon: 39.50) and coupling constants (J) in Hertz were provided. The following characters follow the designations for splitting patterns: s for singlet; d for doublet; t for triplet; q for quartet; dd for doublet of doublet; m for multiplet. Mass spectra were acquired at Al-Azhar University’s Regional Center for Mycology and Biotechnology utilizing a Varian MAT 311-A (70 eV). Thin-layer chromatography was employed to track the development of the compound preparation reactions with Merck 9385 precoated aluminum plate silica gel (Kieselgel 60) 5 × 20 cm plates with a 0.2 mm-layer thickness. By exposing the spots to a UV lamp at λ = 254 nm, the spots were found. The Regional Center for Mycology and Biotechnology (Cairo, Egypt) employed the Vario El Elementar CHN Elemental analyzer for elemental analyses. The findings were collected within ± 0.4% of the theoretical values.

Synthesis of 2-oxo-6-(4-methoxyphenyl)-4-(3,4,5-trimethoxyphenyl)-1,2-dihydropyridine-3-carbonitrile 1

A mixture of (1 mmol) of 3,4,5-trimethoxybenzaldehyde, (1 mmol) of 4-methoxyacetophenone, (1 mmol) of ethyl cyanoacetate and (1 mmol) of ammonium acetate were combined and heated to 120–130°C for 5–10 min while vigorously stirring. The crude product crystallized after cooling. result from pure ethanol. After cooling, the crude product was crystallized from absolute ethanol. Yellow crystals (yield: 3.25 g, 90%) mp = 305–307°C, recorded mp <300°C. ¹H NMR (400 MHz, DMSO-d₆) δ (ppm): 3.76 (s, 3H, OCH₃), 3.85 (s, 3H, OCH₃), 3.86 (s, 3H, OCH₃), 3.88 (s, 6H, two-OCH₃), 6.80 (s, 1H, pyridine-C₅-H), 7.00–7.15 (m, 4H, 3,4,5-trimethoxyphenyl-C_2,6-H and 4-methoxyphenyl- C_3,5-H), 8.22 (d, 2H, J = 8.00 Hz, 4-methoxyphenyl-C_2,6-H), 13.09 (br s, 1H, NH, D₂O exchangeable).

General procedure for the synthesis of compounds 2a–c

In DMF (20 ml) containing anhydrous potassium carbonate (0.690 g, 5 mmol), a mixture compound 1 (1.81 g, 5 mmol) and the appropriate ester (5 mmol), namely methyl 4-bromomethyl benzoate (1.145 g), ethyl 5-bromopentanoate (1.045 g) or ethyl 6-bromohexanoate, was stirred for 24 h at room temperature. Crushed ice was added to dilute the reaction mixture, and the precipitate that formed was filtered, rinsed with water and dried.

The aqueous ethanol solvent was used in the crystallization process of the crude product.

Methyl 4- {[3-cyano-6-(4-methoxyphenyl)-4-(3,4,5-trimethoxyphenyl)pyridin-2-yloxy]methyl} benzoate 2a

White powder (yield 2.45 g, 98%); mp = 170–172°C; ¹H NMR (400 MHz, DMSO-d₆) δ (ppm): 3.76 (s, 3H, OCH₃), 3.85 (s, 3H, OCH₃), 3.86 (s, 3H, OCH₃), 3.88 (s, 6H, two-OCH₃), 5.75 (s, 2H, O-CH₂), 7.00–7.15 (m, 4H, 3,4,5-trimethoxyphenyl-C_2,6-H and 4-methoxyphenyl-C_3,5-H), 7.69 (d, 2H, J = 8.00 Hz, benzoate-C_3,5-H), 7.80 (s, 1H, pyridine-C₅-H), 8.02 (d, 2H, J = 8.00 Hz, benzoate-C_2,6-H), 8.22 (d, 2H, J = 8.00 Hz, 4-methoxyphenyl-C_2,6-H); ¹³C-NMR (100 MHz, DMSO-d₆) δ (ppm): 52.62 (OCH₃), 55.87 (OCH₃), 56.64 (two-OCH₃), 60.60 (OCH₃), 67.96 (OCH₂), 106.86, 114.74 (CN), 128.02, 129.39, 129.56, 129.73, 129.83, 131.54, 136.36, 139.37, 142.69, 153.42, 157.39, 161.96, 163.90, 166.44 (CO); Elemental analysis for C₃₁H₂₈N₂O₇ (540.56)(Calcd./Found); C, 68.88/69.09; H, 5.22/5.26; N, 5.18/5.26.

Ethyl-5-[3-cyano-6-(4-methoxyphenyl)-4-(3,4,5-trimethoxyphenyl)pyridin-2-yloxy]pentanoate 2b

White powder (yield 2.41 g, 92.7%); mp = 96–98°C; ¹H NMR (400 MHz, DMSO-d6) δ (ppm): 1.17 (t, 3H, J = 7.20 Hz, CH₂-CH₃), 1.55–1.98 (m, 4H, CH₂-CH₂-CH₂-CO), 2.41 (t, 2H, J = 7.20 Hz, CH₂-CO), 3.76 (s, 3H, OCH₃), 3.85 (s, 3H, OCH₃), 3.88 (s, 6H, two-OCH₃), 4.05 (q, 2H, J = 7.20 Hz, O-CH₂-CH₃), 4.56 (t, 2H, J = 7.20 Hz, O-CH₂), 7.06–7.10 (m, 4H, 3,4,5-trimethoxyphenyl-C_2,6-H and 4-methoxyphenyl-C_3,5-H), 7.74 (s, 1H, pyridine-C₅-H), 8.21 (d, 2H, J = 8.00 Hz, 4-methoxyphenyl-C_2,6-H); ¹³C-NMR (100 MHz, DMSO-d₆) δ (ppm): 14.54 (CH₃), 21.60 (CH₂), 28.23 (CH₂), 33.62 (CH₂), 55.83 (OCH₂), 56.61 (OCH₃), 60.19 (OCH₃), 60.58 (OCH₃), 62.09 (OCH₃), 66.97 (OCH₂), 91.58, 106.53, 106.80, 113.02, 114.68 (CN), 129.61, 129.69,129.90, 131.65, 132.73, 139.32, 153.40, 157.41, 161.86, 164.43, 173.19 (CO); Elemental analysis for C₂₉H₃₂N₂O₇ (520.57)(Calcd./Found); C, 66.91/67.04; H, 6.20/6.25; N, 5.38/5.45.

Ethyl-6-[3-cyano-6-(4-methoxyphenyl)-4-(3,4,5-trimethoxyphenyl)pyridin-2-yloxy]hexanoate 2c

White powder (yield 2.5 g, 93%); mp = 101–103°C; ¹H NMR (400 MHz, DMSO-d6) δ (ppm): 1.17 (t, 3H, J = 7.20 Hz, CH₂-CH₃), 1.42–1.81 (m, 6H, CH₂-CH₂-CH₂-CH₂-CO), 2.32 (t, 2H, J = 7.20 Hz, CH₂-CO), 3.76 (s, 3H, OCH₃), 3.84 (s, 3H, OCH₃), 3.88 (s, 6H, two-OCH₃), 4.05 (q, 2H, J = 7.20 Hz, O-CH₂-CH₃), 4.53 (t, 2H, J = 7.20 Hz, O-CH₂), 7.06–7.08 (m, 4H, 3,4,5-trimethoxyphenyl-C_2,6-H and 4-methoxyphenyl-C_3,5-H), 7.73 (s, 1H, pyridine-C₅-H), 8.21 (d, 2H, J = 8.00 Hz, 4-methoxyphenyl-C_2,6-H); ¹³C-NMR (100 MHz, DMSO-d₆) δ (ppm):, 14.54 (CH₃), 24.66 (CH₂), 25.42 (CH₂), 28.45 (CH₂), 33.93 (CH₂), 55.82 (OCH₃), 56.59 (OCH₃), 60.12 (OCH₃), 60.57 (OCH₃), 66.90 (OCH₂), 67.16 (OCH₂), 91.56, 106.79, 112.98, 114.67, 114.78, 129.59, 129.90, 131.56, 139.32, 153.39, 156.43, 157.39, 161.85, 164.47, 173.23 (CO); Elemental analysis for C₃₀H₃₄N₂O₇ (534.60)(Calcd./Found); C, 67.40/67.52; H, 6.41/6.49; N, 5.24/5.35.

General procedure for the synthesis of compounds 3a–c

The appropriate ester (1 mmol) was suspended in 20 ml 20% methanolic potassium hydroxide and stirred at room temperature for 16 h to produce the corresponding acid. After filtering the reaction mixture, the filtrate was acidified with hydrochloric acid and chilled to 0°C. The produced solid was separated by filtration, washing with water, drying and further crystallization from the aqueous ethanol.

4-{[3-Cyano-6-(4-methoxyphenyl)-4-(3,4,5-trimethoxyphenyl)pyridin-2-yloxy]methyl}benzoic acid 3a

Yellow powder (yield 0.462 g, 88%); mp = 225–227°C; IR (cm^-1): 3466–3120 (OH), 2218 (CN), 1719 (C=O); ¹H NMR (400 MHz, DMSO-d₆) δ (ppm): 3.76 (s, 3H, OCH₃), 3.86 (s, 3H, OCH₃), 3.88 (s, 6H, two-OCH₃), 5.74 (s, 2H, O-CH₂), 7.03–7.14 (m, 4H, 3,4,5-trimethoxyphenyl-C_2,6-H and 4-methoxyphenyl-C_3,5-H), 7.62 (d, 2H, J = 8.00 Hz, benzoate-C_3,5-H), 7.81 (s, 1H, pyridine-C₅-H), 8.00 (d, 2H, J = 8.00 Hz, benzoate-C_2,6-H), 8.24 (d, 2H, J = 8.00 Hz, 4-methoxyphenyl-C_2,6-H), 12.96 (br s, 1H, OH, D₂O exchangeable); ¹³C-NMR (100 MHz, DMSO-d₆) δ (ppm): 55.86 (OCH₃), 56.62 (OCH₃), 56.64 (OCH₃), 60.61 (OCH₃), 67.96 (CH₂), 106.63, 114.51, 114.76 (CN), 126.50, 127.67, 129.82, 130.45, 133.26, 139.46, 152.68, 153.44, 161.97, 165.82, 172.06 (CO); Elemental analysis for C₃₀H₂₆N₂O₇ (526.54)(Calcd./Found); C, 68.43/68.63; H, 4.98/5.01; N, 5.32/5.39.

5-[3-Cyano-6-(4-methoxyphenyl)-4-(3,4,5-trimethoxyphenyl)pyridin-2-yloxy]pentanoic acid 3b

Yellowish white powder (yield 0.437 g, 89%); mp = 107–109°C; ¹H NMR (400 MHz, DMSO-d₆) δ (ppm): 1.65–1.89 (m, 4H, CH₂-CH₂-CH₂-CO), 2.33 (t, 2H, J = 7.20 Hz, CH₂-CO), 3.76 (s, 3H, OCH₃), 3.85 (s, 3H, OCH₃), 3.88 (s, 6H, two OCH₃), 4.57 (t, 2H, J = 7.20 Hz, O-CH₂), 7.00–7.12 (m, 4H, 3,4,5-trimethoxyphenyl-C_2,6-H and 4-methoxyphenyl-C_3,5-H), 7.75 (s, 1H, pyridine-C₅-H), 8.23 (d, 2H, J = 8.00 Hz, 4-methoxyphenyl-C_2,6-H), 11.60 (br s, 1H, OH, D₂O exchangeable); ¹³C-NMR (100 MHz, DMSO-d₆) δ (ppm): 21.66 (CH₂), 28.34 (CH₂), 33.89 (CH₂), 55.86 (OCH₃), 56.63 (OCH₃), 60.60 (OCH₃), 61.98 (OCH₃), 67.07 (OCH₂), 91.62, 106.82, 114.74 (CN), 129.64, 131.68, 133.65, 134.63, 139.30, 153.40, 156.50, 157.14, 157.42, 161.87, 164.46, 173.65 (CO); EI-MS (70 eV) m/z (%): 492 (25) (M⁺), 491 (21), 43 (100); Elemental analysis for C₂₇H₂₈N₂O₇ (492.52)(Calcd./Found): C, 65.84/65.99; H, 5.73/5.77; N, 5.69/5.81.

6-[3-Cyano-6-(4-methoxyphenyl)-4-(3,4,5-trimethoxyphenyl)pyridin-2-yloxy]hexanoic acid 3c

Yellowish white powder (yield 0.465 g, 93%); mp = 142–144°C; ¹H NMR (400 MHz, DMSO-d6) δ (ppm): 1.42–1.53 (m, 2H, CH₂-CH₂-CH₂-CO), 1.54–1.87 (m, 4H, CH₂-CH₂-CH₂-CH₂-CO), 2.25 (t, 2H, J = 7.20 Hz, CH₂-CO), 3.76 (s, 3H, OCH₃), 3.85 (s, 3H, OCH₃), 3.88 (s, 6H, two OCH₃), 4.55 (t, 2H, J = 7.20 Hz, O-CH₂), 7.03–7.11 (m, 4H, 3,4,5-trimethoxyphenyl-C_2,6-H and 4-methoxyphenyl-C_3,5-H), 7.74 (s, 1H, pyridine-C₅-H), 8.21 (d, 2H, J = 8.00 Hz, 4-methoxyphenyl-C_2,6-H),12.04 (br s, 1H, OH, D₂O exchangeable); ¹³C-NMR (100 MHz, DMSO-d₆) δ (ppm): 24.73 (CH₂), 25.52 (CH₂), 28.52 (CH₂), 34.15 (CH₂), 55.85 (OCH₃), 56.62 (OCH₃), 60.59 (OCH₃), 65.44 (OCH₃), 67.23 (OCH₂), 91.60, 106.82, 113.02, 114.74, 129.61, 131.68, 139.00, 139.30, 140.88, 153.39, 156.49, 157.41, 161.87, 164.49, 174.89 (CO); Elemental analysis for C₂₈H₃₀N₂O₇ (506.55)(Calcd./Found): C, 66.39/66.56; H, 5.97/6.03; N, 5.53/5.63.

General procedure for the synthesis of 4a–c

Dry tetrahydrofuran (THF, 10 ml) was mixed with acids 3a–c (0.01 mol) and then N,N′-carbonyldiimidazole (CDI; 0.04 mol, 6.48 g) was added. The mixture was stirred occasionally for 4 h at 25–30°C. After adding hydroxylamine hydrochloride (0.04 mol, 2.78 g), the mixture was stirred for a further 12 h. The solvent was removed by distillation, followed by the addition of 10 ml of ethyl acetate and (2 × 10 ml) washes of water. The organic layer was then collected, dried over anhydrous sodium sulfate, filtered and evaporated under vacuum. Aqueous ethanol was utilized as a crystallization solvent.

4-{[3-Cyano-6-(4-methoxyphenyl)-4-(3,4,5-trimethoxyphenyl)pyridin-2-yloxy]methyl}-N-hydroxybenzamide 4a

Yellow powder (yield 4.76 g, 88%); mp = 209–212°C; IR (cm^-1): 3307–3100 (OH), 3451 (NH), 2218 (CN), 1719 (C=O); 3.76 (s, 3H, OCH₃), 3.85 (s, 3H, OCH₃), 3.88 (s, 6H, two OCH₃), 5.73 (s, 2H, O-CH₂), 7.02–7.16 (m, 4H, 3,4,5-trimethoxyphenyl-C_2,6-H and 4-methoxyphenyl-C_3,5-H), 7.62 (d, 2H, J = 8.00 Hz, benzoate-C_3,5-H), 7.80 (s, 1H, pyridine-C₅-H), 8.00 (d, 2H, J = 8.00 Hz, benzoate-C_2,6-H), 8.23 (d, 2H, J = 8.00 Hz, 4-methoxyphenyl-C_2,6-H), 9.78 (s, 1H, NH, D₂O exchangeable), 11.13 (brs, 1H, OH, D₂O exchangeable); ¹³C-NMR (100 MHz, DMSO-d₆) δ (ppm): 52.26 (OCH₃), 55.51 (OCH₃), 56.28 (OCH₃), 60.24 (OCH₃), 67.58 (OCH₂), 91.43, 106.50, 114.38 (CN), 127.66, 129.03, 129.20, 129.37, 129.47, 131.18, 139.01, 142.33, 153.06, 157.03, 161.60, 163.54, 166.08, 169.52 (CO); Elemental analysis for C₃₀H₂₇N₃O₇ (541.55)(Calcd./Found); C, 66.53/66.72; H, 5.03/5.06; N, 7.76/7.81.

5-[3-Cyano-6-(4-methoxyphenyl)-4-(3,4,5-trimethoxyphenyl)pyridin-2-yloxy]-N-hydroxypentanamide 4b

Whitish yellow powder (yield 4.61 g, 91%); mp = 96–97°C; IR (cm^-1): 3277 (OH), 3549 (NH), 2217 (CN), 1721 (C=O); ¹H NMR (400 MHz, DMSO-d₆) δ (ppm): 1.67–1.92 (m, 4H, CH₂-CH₂-CH₂-CO), 2.34 (t, 2H, J = 7.20 Hz, CH₂-CO), 3.76 (s, 3H, OCH₃), 3.86 (s, 3H, OCH₃), 3.89 (s, 6H, two OCH₃), 4.58 (t, 2H, J = 7.20 Hz, O-CH₂), 7.06–7.12 (m, 4H, 3,4,5-trimethoxyphenyl-C_2,6-H and 4-methoxyphenyl-C_3,5-H), 7.77 (s, 1H, pyridine-C₅-H), 8.24 (d, 2H, J = 8.00 Hz, 4-methoxyphenyl-C_2,6-H), 9.77 (s, 1H, NH, D₂O exchangeable), 10.41 (brs, 1H, OH, D₂O exchangeable); ¹³C-NMR (100 MHz, DMSO-d₆) δ (ppm): 21.93 (CH₂), 28.56 (CH₂), 33.95 (CH₂), 56.16 (OCH₃), 56.94 (OCH₃), 60.52 (OCH₃), 60.91 (OCH₃), 67.30 (CH₂), 91.91, 106.88, 107.13, 113.35, 115.01 (CN), 129.94, 130.02, 130.23, 131.98, 133.06, 139.65, 153.73, 156.50, 157.74, 162.19, 164.76, 171.80 (CO); EI-MS (70 eV) m/z (%): 507 (7) (M⁺), 508 (11) (M⁺¹), 42 (100); Elemental analysis for C₂₁H₂₀N₂O₄ (507.54)(Calcd./Found): C, 63.89/64.04; H, 5.76/5.79; N, 8.28/8.42.

6-[3-Cyano-6-(4-methoxyphenyl)-4-(3,4,5-trimethoxyphenyl)pyridin-2-yloxy]-N-hydroxyhexanamide 4c

Yellowish white powder (yield 4.63 g, 89%); mp = 132–134°C; 1.42–1.56 (m, 2H, CH₂-CH₂-CH₂-CO), 1.54–1.87 (m, 4H, CH₂-CH₂-CH₂-CH₂-CO), 2.24 (t, 2H, J = 7.20 Hz, CH₂-CO), 3.75 (s, 3H, OCH₃), 3.84 (s, 3H, OCH₃), 3.87 (s, 6H, two OCH₃), 4.55 (t, 2H, J = 7.20 Hz, O-CH₂), 7.03–7.12 (m, 4H, 3,4,5-trimethoxyphenyl-C_2,6-H and 4-methoxyphenyl-C_3,5-H), 7.75 (s, 1H, pyridine-C₅-H), 8.22 (d, 2H, J = 8.00 Hz, 4-methoxyphenyl-C_2,6-H), 9.70 (s, 1H, NH, D₂O exchangeable), 10.40 (brs, 1H, OH, D₂O exchangeable); ¹³C-NMR (100 MHz, DMSO-d₆) δ (ppm): 24.86 (CH₂), 25.62 (CH₂), 28.65 (CH₂), 34.13 (CH₂), 56.02 (OCH₃), 56.79 (OCH₃), 60.32 (OCH₃), 60.77 (OCH₃), 67.36 (OCH₂), 91.76, 106.99, 113.18, 114.87, 129.79, 130.10, 131.85, 139.52, 153.59, 156.63, 157.59, 161.11, 162.05, 164.67, 171.83 (CO); Elemental analysis for C₂₈H₃₁N₃O₇ (521.56)(Calcd./Found): C, 64.48/64.65; H, 5.99/6.03; N, 8.06/8.06.

General procedure for the synthesis of compounds 5a–c

CDI (0.39 g, 2.4 mmol) was added to a suspension of 3a–c (2 mmol) in dry THF (10 ml), and the mixture underwent stirring for 4 h at room temperature. Following the addition of 0.28 g (2.60 mmol) of o-phenylenediamine and 0.20 g (0.14 ml, 1.76 mmol) of TFA to the reaction mixture while stirring for 16 h at room temperature, the precipitated solid was filtered, rinsed with water, dried and crystallized from ethanol.

N-(2-Aminophenyl)-4-{[3-cyano-6-(4-methoxyphenyl)-4-(3,4,5-trimethoxyphenyl)pyridin-2-yloxy]methyl}benzamide 5a

Pale brown powder (yield 1.084 g, 86%); mp = 116–118°C; ¹H NMR (400 MHz, DMSO-d₆) δ (ppm): 3.75 (s, 3H, OCH₃), 3.85 (s, 3H, OCH₃), 3.87 (s, 6H, two-OCH₃), 4.90 (s, 2H, NH₂, D₂O exchangeable), 5.75 (s, 2H, O-CH₂), 6.54–6.94 (m, 2H, 2-aminophenyl-C_3,5-H), 7.04–7.20 (m, 6H, 3,4,5-trimethoxyphenyl-C_2,6-H, 4-methoxyphenyl-C_3,5-H and 2-aminophenyl-C_4,6-H), 7.67 (d, 2H, J = 8.00 Hz, benzoate-C_3,5-H), 7.81 (s, 1H, pyridine-C₅-H), 8.01 (d, 2H, J = 8.00 Hz, benzoate-C_2,6-H), 8.24 (d, 2H, J = 8.00 Hz, 4-methoxyphenyl-C_2,6-H), 9.61 (s, 1H, NH, D₂O exchangeable); ¹³C-NMR (100 MHz, DMSO-d₆) δ (ppm): 55.42 (OCH₃), 56.17 (OCH₃), 56.25 (OCH₃), 60.14 (OCH₃), 67.56 (OCH₂), 94.29, 106.39, 113.08, 114.32, 114.54 (CN), 116.22, 127.47, 128.00, 129.29, 129.52, 131.10, 138.88, 147.27, 151.72, 152.96, 154.47, 156.24, 163.48, 163.55, 165.17, 168.03 (CO); Elemental analysis for C₃₆H₃₂N₄O₆ (616.66)(Calcd./Found); C, 70.12/70.30; H, 5.23/5.25; N, 9.09/9.16

N-(2-Aminophenyl)-5-[3-cyano-6-(4-methoxyphenyl)-4-(3,4,5-trimethoxyphenyl)pyridin-2-yloxy]pentanamide 5b

Brown powder (1.024 g, 88% yield); mp = 92–93°C; ¹H NMR (400 MHz, DMSO-d₆) δ (ppm): 1.67–1.89 (m, 4H, CH₂-CH₂-CH₂-CH₂-CO), 2.32 (t, 2H, J = 7.20 Hz, CH₂-CO), 3.75 (s, 3H, OCH₃), 3.84 (s, 3H, OCH₃), 3.87 (s, 6H, two OCH₃), 4.57 (t, 2H, J = 7.20 Hz, O-CH₂), 4.82 (s, 2H, NH₂, D₂O exchangeable), 6.48–7.48 (m, 8H, 3,4,5-trimethoxyphenyl-C_2,6-H, 4-methoxyphenyl-C_3,5-H and 2-aminophenyl-C_3,4,5,6-H), 7.75 (s, 1H, pyridine-C₅-H), 8.22 (d, 2H, J = 8.00 Hz, 4-methoxyphenyl-C_2,6-H), 9.14 (s, 1H, NH, D₂O exchangeable); ¹³C-NMR (100 MHz, DMSO-d₆) δ (ppm): 21.18 (CH₂), 27.87 (CH₂), 33.41 (CH₂), 55.40 (OCH₃), 56.15 (2 OCH₃), 60.13 (OCH₃), 66.61 (OCH₂), 91.17, 106.34, 112.66, 114.29 (CN), 115.73, 125.29, 125.71, 129.14, 129.19, 131.23, 138.81, 141.89, 152.94, 154.82, 156.06, 156.96, 161.42, 164.01, 170.93 (CO); Elemental analysis for C₃₃H₃₄N₄O₆ (582.65)(Calcd./Found): C, 68.03/68.20; H, 5.88/5.92; N, 9.62/9.77.

N-(2-Aminophenyl)-6-[3-cyano-6-(4-methoxyphenyl)-4-(3,4,5-trimethoxyphenyl)pyridin-2-yloxy]hexanamide 5c

Grey powder (yield 1.037 g, 87%); mp = 108–110°C; IR (cm^-1): 3427 (NH₂), 3266 (NH), 2217 (CN), 1707 (C=O); ¹H NMR (400 MHz, DMSO-d6) δ (ppm): 1.42–1.53 (m, 2H, CH₂-CH₂-CH₂-CO), 1.54–1.87 (m, 4H, CH₂-CH₂-CH₂-CH₂-CO), 2.25 (t, 2H, J = 7.20 Hz, CH₂-CO), 3.76 (s, 3H, OCH₃), 3.84 (s, 3H, OCH₃), 3.87 (s, 6H, two OCH₃), 4.54 (t, 2H, J = 7.20 Hz, O-CH₂), 4.83 (s, 2H, NH₂, D₂O exchangeable), 6.49- 6.88 (m, 1H, 2-aminophenyl-C₃-H), 7.03–7.31 (m, 6H, 3,4,5-trimethoxyphenyl-C_2,6-H, 4-methoxyphenyl-C_3,5-H and 2-aminophenyl-C_4,5-H), 7.42–7.63 (m, 1H, 2-aminophenyl-C₆-H), 7.73 (s, 1H, pyridine-C₅-H), 8.21 (d, 2H, J = 8.00 Hz, 4-methoxyphenyl-C_2,6-H), 9.25 (s, 1H, NH, D₂O exchangeable); ¹³C-NMR (100 MHz, DMSO-d₆) δ (ppm): 24.71 (CH₂), 25.51 (CH₂), 28.51 (CH₂), 34.11 (CH₂), 55.84 (OCH₃), 56.61 (OCH₃), 56.68 (OCH₃), 60.59 (OCH₃), 67.22 (OCH₂), 91.58, 106.80, 112.98, 114.72, 125.10, 127.85, 128.62, 129.59, 130.90, 131.67, 139.29, 142.35, 153.39, 157.41, 159.68, 161.86, 164.49, 172.42 (CO); Elemental analysis for C₃₄H₃₆N₄O₆ (596.67)(Calcd./Found): C, 68.44/68.60; H, 6.08/6.14; N, 9.39/9.56.

Biology

In vitro anticancer assay (National Cancer Institute)

This assay evaluated the in vitro anticancer activity of nine selected compounds, namely 3a–c, 4a–c and 5a–c, at a single dose concentration. The cancer cell lines used in the evaluation were derived from nine distinct cancer types, namely lung, leukemia, colon, ovarian, melanoma, renal, prostate, CNS and breast cancers. The examined compounds were incorporated into the culture at a single concentration of 10–5 M and left for 48 h of incubation. Sulforhodamine B, a protein-binding dye, was utilized and applying spectrophotometry, the percentage growth was compared to the control group, which was not exposed to the test substances. After drug addition, cells were incubated at 37°C, 5% CO₂, 95% air and 100% relative humidity for 48 h then stained by sulforhodamine B and the absorbance was evaluated spectrophotometrically via using an automated plate. The percent growth was calculated at different drug concentration levels and at different times. These assay techniques were carried out in accordance with the protocol supplied by the National Cancer Institute (NCI; MA, USA) Drug Evaluation Branch. The methodology details of the assay are available on the NCI website (https://dtp.cancer.gov/discovery_development/nci-60/technique.htm; also see Supplementary File) [35].

In vitro enzymatic activity assay

First HDAC screening on HeLa cell nuclear extract

All of the recently synthesized target hybrids (3a–c, 4a–c and 5a–c) had their in vitro HDAC inhibitory activity measured by assessing against HeLa cell nuclear extract (which primarily contains HDAC1 and HDAC2) using a fluorometric HDAC inhibitor drug-screening kit (BioVision, CA, USA; catalog no. K340-100), in accordance with the manufacturer’s recommended protocol [36]. Briefly, the compounds under testing were dissolved in DMSO to prepare fourfold serial dilutions, beginning at 10 μM, and thoroughly combined with the reagents included in the kit. After adding 50 μl of the reaction mix to each well and mixing thoroughly, the plate was incubated for a minimum of 30 min at 37°C. Finally, the reaction was stopped by adding 10 μl of lysine developer and mixing well. The fluorescence of the plate is measured using a fluorescence plate reader with Ex = 350–380 nm and Em = 440–460 nm after 30 min of incubation at 37°C. Each compound’s IC₅₀ was recorded using the GraphPad Prism program (Supplementary File).

In vitro HDAC inhibitory assay against HDAC1 & HDAC6 isoforms

In the first HDAC screening on HeLa cell nuclear extract, the inhibitory activity and selectivity of HDAC isoforms on class I (HDAC1) and class IIb (HDAC6) isoforms for the most potent hybrids were measured using the manufacturer’s protocols for the Kinetic HDAC1 Assay Kit and the Fluorogenic HDAC6 Assay Kit (BPS Bioscience, CA, USA; catalog nos 53001 and 50076-1, respectively) [37,38]. In DMSO, the substances under examination were dissolved. By utilizing GraphPad Prism, the IC₅₀ for each compound was determined.

Assay protocol

Put every ingredient on ice to defrost, except for the HDAC enzyme. Every reaction needs to be carried out on ice.

Preparation of the master HDAC substrate solution

A total of 237 μl of HDAC assay buffer, 800 μl of BSA (1 mg/ml), and 32 μl of HDAC substrate is mixed to create the master HDAC substrate solution. This will generate 100 responses. The components can be proportionally decreased if there are fewer than 100 reactions. Make the amount required for the assay and keep the extra BSA and 5 mM solution in stock in aliquots at -80°C. The master HDAC substrate solution is then added in 20-μl increments to each well.

Preparation of the inhibitor solutions

Dilute the test inhibitor ten times more than the final concentration you wish to test in 10% DMSO (the final DMSO concentration is 1% in all the reactions). Then, as a control, dilute the identical buffer (inhibitor buffer) without the inhibitor in 10% DMSO. Specifically, 10 μl of the test inhibitor solution and 10 μl of SAHA solution should be given to each well-marked ‘test inhibitor’ and ‘SAHA inhibitor control’ respectively. To the ‘positive control’ and ‘blank’ wells, add 10 μl of the inhibitor buffer, respectively.

Preparation of the developer solution

Developer dilution buffer should be used to dilute the 10× developer solution (1:10). A total of 50 μl of the developer solution should then be inserted into every well.

Preparation of the HDAC1/6 enzyme solution

Dilute HDAC1/6 to 4 ng/μl (80 ng/reaction) in HDAC assay buffer. Any remaining enzyme should be aliquoted and kept undiluted at -80°C. Add 20 μl of HDAC assay buffer to each thoughtfully created ‘blank’ after that. Each carefully created ‘test inhibitor’, ‘positive control’ and ‘SAHA inhibitor control’ should receive 20 μl of the HDAC1 enzyme solution. The enzyme solution is always added last.

A microtiter plate-reading fluorimeter

A microtiter plate-reading fluorimeter measured the plate at 5-min intervals for up to 1 h, with the ability to excite at 350–380 nm and detect light emission at 440–460 nm.

Tubulin polymerization inhibition assay

According to the manufacturer’s recommended protocol, using a Tecan spark reader and a fluorescence kit (Cytoskeleton, cat. no. BK011P), the investigated drugs’ capacity to block tubulin polymerization was evaluated [39]. In triplicate, 5 μl of the test compound solution in 10% DMSO was poured into each well of a 96-well plate. The reaction mixture was then warmed for 1 min at 37°C before the addition of the tubulin. A total of 45 μl of 2 mg/ml (>99% pure) tubulin, 80.0 mM piperazine-N,N′-bis(2-ethanesulfonic acid) sequisodium salt, 0.5 mM EGTA, 2 mM MgCl2, 1 mM GTP and 10.2% glycerol were included, along with 10 μM of DAPI (4′,6-diamidino-2-phenylindole), a fluorescent reporter. Tubulin polymerization was detected by tracking the fluorescence emission at 450 nm (excitation = 360 nm) for 60 min at intervals of 1 min. Using GraphPad Prism software, the IC₅₀ for each compound was determined from the obtained data.

In vitro cytotoxicity assay on normal cell line WI-38 & other cancer cell lines

Dulbecco’s modified Eagle’s medium (Invitrogen [MA,USA]/Life Technologies [CA, USA]) was used to cultivate the cells, and an MTT (3-[4,5-dimethylthiazole-2-yl]-2,5-diphenyltetrazolium bromide) assay was used to measure the cytotoxicity of the hybrids that were examined [40]. The American Type Culture Collection supplied the cells (All of the other chemicals and reagents were from Sigma or Invitrogen) . The cells were plated in a 96-well plate for 24 h at a density of 1.2–1.8 × 10,000 cells/well in a volume of 100 μl of the test substance plus 100 μl of full growth media per well prior to the MTT assay.

Cell culture protocol

The culture medium was removed to a centrifuge tube, and then the cell layer was quickly rinsed with a solution containing 0.53 mM EDTA and 0.25% (w/v) trypsin to completely remove any remaining trypsin inhibitor-containing serum. Next, 2.0–3.0 ml of trypsin EDTA solution was added to the flask, and an inverted microscope was used to observe the cells until the cell layer separated (typically within 5–15 min). The cells were kept at 37°C, added to 6.0–8.0 ml of full growth media and then carefully pipetted out to help with cell dispersal. The cell suspension, medium and cells were placed into a centrifuge tube and centrifuged for 5–10 min at a rate of approximately 125×. It was necessary to resuspend the cell pellet in fresh growing media as well as fill new culture vessels with appropriate aliquots of the cell suspension. The cultures were grown at 37°C for a whole day. Before conducting the MTT assay, the plates were viewed under an inverted microscope following a 48-h incubation period at 37°C at the serial concentrations of the test material in the cells [40].

MTT-cytotoxicity assay protocol

The MTT assay was conducted using multiwell plates, with the final cell density not higher than 10⁶ cells/cm². There was a blank medium with no cells in each test. The cultures were removed from the incubator and placed in a sterile workspace (such as a laminar flow hood). Then each vial of MTT [M-5655] was resuspended in 3 ml of medium or balanced salt solution free of phenol red and serum. A total of 10% of the volume of the culture medium was added to the reconstituted MTT. After that, the cultures were incubated again for 2 h. After incubation, the cultures were removed from the incubator and the same volume of MTT Solubilization Solution (M-8910) was used to dissolve the formazan crystals that resulted. At a wavelength of 450 nm, absorbance was measured spectrophotometrically. The background absorbance of the multiwell plates at 690 nm was subtracted from the 450-nm measurement. Using GraphPad Prism software, the IC₅₀ from the data was calculated (Supplementary File) [40].

Cell cycle analysis

After seeding 1 × 10⁵ cells/well on a six-well plate, T47D cells were incubated for a full day. For a duration of 24 h, 5.50 μM hybrid 4b or 0.1% DMSO were applied to T47D cells. After that, cells were gathered and preserved for 12 h at 4°C in ice-cold 70% ethanol. Furthermore, the cells were placed in 0.5 ml of cooled phosphate-buffered saline (PBS) and incubated for 30 min at 37°C. Propidium iodide (PI) was applied to the cells and left in the dark for 30 min. The DNA content was then detected using a flow cytometer [41].

Annexin (Fluoresceinisothiocyanate) V-FITC assay

1 × 10⁵ concentration of cells/well of T47D cells were put in a six-well plate, and the plate was incubated for 24 h. Hybrid 9e at a concentration of 5.50 μM or 0.1% DMSO was applied to the T47D cells for 24 h. After harvesting, the cells underwent PBS rinsing, staining with annexin V-FITC and PI in binding buffer (10 μM HEPES, 140 μM NaCl and 2.5 μM CaCl2 at pH 7.4), and underwent flow cytometry analysis [39,42].

Docking methodology

Software called Discovery Studio 2.5 was employed for the docking analysis (Accelrys Inc., CA, USA). This is a docking tool that is fully automated and runs on an Intel^® core™ i32370 CPU running at 2.4 GHz with 2 GB of RAM and a Windows 7.0 operating system. The crystal structures of tubulin (Protein Data Bank [PDB] entry: 1SA0), HDAC6 (PDB entry: 5EF8) and HDAC1 (PDB entry: 5ICN) were downloaded from the Protein Data Bank. Drawing standard for chemical structure: Cambridge Soft Corporation (MA, USA; 2010). The docked compounds were created in Chem. 3D ultra 12.0 and then copied to Discovery Studio 2.5.

An automatic protein synthesis module was employed with the MMFF94 force field. The binding site sphere was automatically defined by the software. The receptor that was previously generated was now accepted as input for the ‘input receptor molecule’ parameter in the CDOCKER protocol parameter explorer. Force fields were applied to the compounds in order to reduce their energy structure as much as feasible. After examining the derived postures, the positions with the best protein–ligand interaction energies (as determined by CDOCKER energy calculations) were selected. The receptor–ligand interactions of the complexes were examined in both two and three dimensions (Supplementary File).

Results & discussion

Chemistry

The final target compounds 3a–c, 4a–c and 5a–c were synthesized according to that outlined in Figure 3. By stirring an equimolar amount of the reaction’s component chemicals – 4-methoxyacetophenone, 3,4,5-trimethoxybenzaldehyde, ethyl cyanoacetate and ammonium acetate – at 110°C for 10–15 min, the key product 1 was produced in excellent yield in a one-pot reaction without the use of a solvent [43]. Both the methyl ester 2a and the ethyl esters 2b and 2c were prepared through alkylation of 1 with the respective bromoester in dry DMF and using anhydrous K₂CO₃ as a base [44]. Consequently, the alkaline hydrolysis of the corresponding esters 2a–c with potassium hydroxide (20%) followed by neutralization with HCl afforded the respective acid intermediates 3a–c in a moderate yield [44]. The target hydroxamic acid derivatives 4a–c were obtained by activating the acid intermediates 3a–c with CDI in dry THF followed by adding hydroxylamine hydrochloride [34,45,46]. The structural formulae of the intermediates 2a–c, 3a–c and the final compounds 4a–c were characterized using IR, ¹H NMR, ¹³C NMR as well as elemental analyses. IR data of 4a–c showed a significant stretching vibration band at 3549–3100 cm^-1 assigned as NH and OH. In ¹H NMR spectra, a broad singlet signals in a range of δ 9.70–9.78 ppm assigned as -NH while that at δ 10.40–11.13 ppm corresponds to OH of hydroxamic acids. The remaining protons appeared at their expected δ values. ¹³C NMR, mass spectrometry and elemental analyses data of compounds 4a–c agreed with the suggested structures. The synthesis of the target final benzamide derivatives 5a–c was achieved through reaction of the respective acids 4a–c with CDI in dry THF followed by acylation with benzene-1,2-diamines in trifluoroacetic acid [45,46]. The IR spectra of compounds 5a–c showed significant stretching bands corresponding to NH and NH₂ at 3427–3266 cm^-1. The exchangeable D₂O signals were visible in the spectra of protons 5a–c, ranging from 9.14 to 9.61 δ ppm for NH protons, and 4.82 to 4.90 δ ppm for NH₂ protons. Elemental analyses results were found to be consistent with that calculated for the proposed formulae.

Figure 3. Synthesis of the target compounds 3a–c, 4a–c and 5a–c.

Reagents and conditions: (i) ethyl cyanoacetate, ammonium acetate, fusion, 15 min; (ii) appropriate bromoesters, K₂CO₃, dimethylformamide, room temperature, 24 h; (iii) MeOH, methanolic. Potassium hydroxide, room temperature, 16 h; (iv)N,N′-carbonyldiimidazole, tetrahydrofuran, then NH₂OH.HCl, room temperature, 12 h; or o-phenylenediamine, room temperature, 16 h.

Biology

In vitro anticancer assay

The National Cancer Institute (NCI) chose nine freshly generated target compounds 3a–c, 4a–c and 5a–c for in vitro anticancer screening against all 60 disease-focused human cell lines. The cell lines under examination were collected from nine tumor subpanels, including leukemia, lung, colon, CNS, melanoma, ovarian, renal, prostate and breast cancer, in accordance with the protocol outlined by the Drug Evaluation Branch, NCI (MA, USA).

In vitro single-dose assay on full NCI-60 cell lines

At a single concentration of 10 μM, the nine chosen cyanopyridine hybrids were evaluated.

The growth rate of the cells treated with the tested substance in comparison to the untreated control cells was used to record the screening findings. For each cell line, mean graph midpoints (differential activity patterns on a bar scale) were created to facilitate the process of visually scanning the data for possible NCI patterns of selectivity. The bars depicted each tumor cell line’s deviation from the total mean value of all growth percentages noted for every cell line tested. When the growth is more than average, resistance is shown by bars pointing to the left (positive values), but when the growth is less than normal, sensitivity is indicated by bars pointing to the right (negative values). The term ‘delta’ refers to the logarithm of the mean graph midpoint difference.

The findings of the screening were shown again as a percentage of growth inhibition (GI%). The obtained results showed that substitution with the methoxy moieties on the phenyl groups resulted in a decrease in anticancer activity. Hybrids 3a–c, with COOH as ZBG, and aliphatic linkers or aromatics showed moderate-to-weak anticancer activity against most of the tested cancer cell lines.

Activity and broadness was enhanced (with GI% greater than 10 against varying numbers of 15, 59 and 24 cancer cell lines, respectively) when the ZBG switched from COOH to hydroxamic acid, as in hybrids 4a–c. According to the results of these hydroxamate derivatives, compound 4b is the most active hybrid, with only slight anticancer activity against other cancer cell lines but substantial anticancer activity (growth inhibition 60.00% and up to 74.87) against 13 cancer cell lines, as shown in Supplementary Table 36.

In hybrids 5a–c, benzamide as ZGB replaces the hydroxamic acid functionality, increases the broadness except in 5b but with a moderate-to-weak antiproliferative inhibitory activity (with GI% more than 10 against a number of 23, 33 and 34 cancer cell lines, respectively). Moreover, the screening results showed that leukemia (K-562, RPMI-8226 and SR), non-small-cell lung cancer (NCI-H522), colon cancer (HCT-15), CNS cancer (SNB-75), renal cancer (A498, CAKI-1 and UO-31) and breast cancer (MDA-MB-231/ATCC, HS 578T and T-47D) are sensitive to all the tubulin-HDAC hybrid inhibitors with GI% ranges of 12.04–67.36, 15.74–64.11, 19.47–86.67, 10.50–17.92, 29.87–74.46, 14.44–55.6, 14.27–53.74, 27.14–62.21, 16.51–44.42, 12.31–55.00 and 22.02–49.35%, respectively, except for compound 3a on renal cancer (A498) and breast cancer (MDA-MB-231/ATCC and HS 578T), with a GI% of less than 10%. Regardless of the ZBG, the hybrids with aliphatic linkers were more potent than those with aromatic linkers (Supplementary Tables 27–35).

Antiproliferative assay against UO-31 & T47D cancer cells

The most promising hybrids – 4b, 4c, 5b and 5c – were selected to evaluate their antiproliferative activities. The selection of compounds was based on their first anticancer screening at NCI and on the most sensitive cancer cell lines to them – renal cancer UO-31 and breast cancer T47D cell lines – and used SAHA as reference drug. The results, shown in Figure 4, indicate that hybrid 4b possessed strong antiproliferative properties on UO-31 (IC₅₀ = 3.52 ± 0.30 μM), more than SAHA (IC₅₀ = 3.62 ± 0.30 μM), and had effective antiproliferative measures on T47D cancer cells (IC₅₀ = 5.5 ± 0.40 μM) compared with SAHA (IC₅₀ = 3.53 ± 0.20 μM).

Figure 4. Antiproliferative activity assays of 4b, 4c, 5b and 5c on UO-31 and T47D cancer cells compared with vorinostat (SAHA) ± standard deviation of at least three independent experiments.

According to an analysis of the collected data, both cell lines’ activities are in the following order: 4b > 4c > 5b > 5c. This suggests that the hybrids with hydroxamic acid ZBG are more potent than their corresponding analogues with benzamide ZBG and hybrids with a four-carbon spacer linker are more potent than those with linker of five carbon spacers.

In vitro enzymatic inhibitory activity assay

In vitro HDAC inhibition assay

The 3a–c, 4a–c and 5a–c hybrids were evaluated for their in vitro HDAC inhibitory activity against a HeLa cell nucleus extract (which mostly contains HDAC1 and HDAC2) using SAHA as a positive control in order to determine the mechanism of action of the recently synthesized derivatives. Using a fluorometric HDAC inhibitor drug screening kit (BioVision), the target compounds’ HDAC inhibitory activity was assessed in accordance with the manufacturer’s instructions [47].

According to an analysis of the data, hybrids with hydroxamic acid as the zinc binding group were more potent at inhibiting HDAC than its analogues with other ZBGs (Figure 5A; IC₅₀ ranged from 43.42 ± 2.00 to 146.50 ± 8.80 nM), irrespective of the type of linker. When the 2-aminoanilide group, or COOH group, replaces the hydroxamic acid ZBG group, or an aromatic linker in place of the aliphatic linker, with the notable exception of hybrid 5b and 5c with 2-aminoanilide as ZBG group, which still exhibit strong histone deacetylase inhibitory activity (IC₅₀ = 72.45 ± 4.35 and 68.66 ± 4.13 nM, respectively), the in vitro HDAC assay results clearly show that adjusting the ZBG modifies the HDAC inhibitory activity of the compounds under investigation in the following ways: where the aliphatic linker is more active than the aromatic one, it is hydroxamic acid > 2-aminoanilide > carboxylic acid.

Figure 5. Biological investigation.

(A) In vitro HDAC inhibitory activity (IC₅₀) of hybrids 3a–c, 4a–c and 5a–c against HeLa cell nuclear extract compared with vorinostat (SAHA) ± standard deviation (SD) of at least three independent experiments. (B)In vitro HDAC1 and HDAC6 inhibitory activity (IC₅₀) of hybrids 4b, 4c and 5c showing HDAC isoform profiling compared with SAHA ± SD of at least three independent experiments. (C) Tubulin polymerization IC₅₀ for compounds 1, 4b, 4c, 5b, 5c and vinblastine. (D)In vitro cytotoxicity (IC₅₀) against human normal cell line WI-38 for hybrids 4b, 4c and SAHA ± SD of at least three independent experiments.

HDAC: Histone deacetylase.

A five-carbon spacer between the capping group and ZBG is ideal for this activity. Additional analysis of the results of the most effective compounds, which include hydroxamic acid as ZBG and an aliphatic linker, that is, 4b and 4c, showed that the HDAC inhibitory activity increased with the length of the alkyl chain.

In summary, it is clear that the 4,6-diaryl 3-cynopyridine scaffold is an appropriate capping group for the inhibitory action of HDAC. Furthermore, the results of the in vitro HDAC assay made it more clear how the type of ZBG group and the size and structure of the linker affected activity.

In vitro HDAC inhibitory assay against HDAC1 & HDAC6 isoforms

After demonstrating strong antiproliferative activity and strong in vitro HDAC inhibitory activity, the most effective hybrids, 4b, 4c and 5c (HDAC; IC₅₀ of 54.26, 43.42 and 68.66 nM, respectively), were tested for their HDAC isoforms’ inhibitory activity and selectivity on class I (HDAC1) and class IIb (HDAC6) isoforms using SAHA as a positive control. The HDAC1 activity of the hybrids under analysis followed the same trend as the extract from the HeLa cell nucleus. Among the compounds, 4c (HDAC1; IC₅₀ = 58.57 nM) was the most successful. All of the investigated compounds performed better than SAHA but still exhibited encouraging results. Furthermore, Figure 5B shows that the order of tested hybrid action on HDAC6 matched that of the HeLa cell nucleus extract, with 4c (HDAC6; IC₅₀ = 57.32 nM) being the most active compound and SAHA (HDAC6; IC₅₀ = 21.38 nM) being the least potent of all the compounds. Additional information on iso-selectivity was revealed by the results, which indicated that all compounds tested were nonselective with the exception of hybrid 5c, which was more selective for HDAC1 (IC₅₀ = 78.60 nM) than HDAC6 (IC₅₀ = 145.60 nM).

Tubulin polymerization inhibition assay

Vinblastine was used as a positive control and parent 1, which had been reported to be a tubulin polymerization inhibitor and used as a capping group in the design, was used to validate the design and investigate the tubulin polymerization inhibitory activity for 4b, 4c, 5b and 5c. These target hybrids were the most potent and showed promising antiproliferative activity as well as potent HDAC inhibitory.

The obtained results (Figure 5C) showed that hybrid 4b (IC₅₀ = 3.73 ± 0.20) is the most potent derivative with potency comparable to that of the reference vinblastine (IC₅₀ = 2.96 ± 0.16 nM) and the parent 1 (IC₅₀ = 3.53 ± 0.19 nM). Furthermore, hybrid 5b showed substantial inhibition of tubulin polymerization (IC₅₀ = 5.18 ± 0.28 nM) but was less potent than vinblastine and parent 1. On the other hand, hybrids 4c and 5c showed moderate potency (IC₅₀ = 8.69 ± 0.47 and 9.03 ± 0.49 nM, respectively) but were threefold less potent than vinblastine.

In conclusion, it is evident that the target hybrids that have been rationalized maintain their inhibitory effect on tubulin polymerization. When ZBG was changed from hydroxamic acid to benzamide, the tubulin polymerization inhibitory efficacy somewhat decreased. Finally, a decrease in tubulin polymerization inhibitory action was seen when lengthening the linker to a five-carbon spacer.

Structure–activity relationships

Regarding to the data listed above, the structure–activity correlation of the synthesized antiproliferative target hybrids revealed the following observations:

1. Aliphatic linkers are preferred over aromatic linkers;

2. In this rationalized design, the length of the aliphatic linker is critical and may be arranged as follows: the four-carbon linker is the most active, which may be due to its impact on the activity that inhibits tubulin polymerization;

3. The best for activity is hydroxamic acid as ZBG > benzamide > acrylamide > carboxylic acid.

In vitro HDAC inhibitory activity shows that antiproliferative activity and HDAC inhibitory activity do not directly correlate. This indicates that the prepared derivatives may exert their effects by inhibiting both tubulin polymerization and HDAC. Notably, these results clarify the remarkable range and strength of hybrid 4b.

In vitro cytotoxicity on normal cell line WI-38

Fibroblasts from the lung tissue of a 3-month-old aborted female fetus comprised the human cell line WI-38, which was tested against hybrids 4b and 4c to determine the cytotoxic effects on the normal cell line. When compared with the control medication SAHA (IC₅₀ = 11.0 μM), hybrids 4b and 4c showed reduced cellular cytotoxicity (IC₅₀ = 32.4 and 30.4 μM, respectively), indicating their safety on normal cell lines (Figure 5D).

Cell cycle analysis

For the most potent hybrid in cytotoxicity tests, 4b was selected using flow cytometry analysis to examine the mechanism by which the generated hybrids demonstrated their antiproliferative activity, cell cycle progression and apoptosis induction.

The dual potent activity of 4b with regard to tubulin polymerization and HDAC inhibitory activity (IC₅₀ = 5.50 μM on T47D) led to its selection. When compared with data from untreated T47D cells, the analysis’ conclusions for hybrid 4b, which are shown in Figure 6A & B, indicated a significant disruption in the cell cycle profile that resulted in apoptosis. Hybrid 4b increased the percentage of cells accumulating in the G2/M phase by 3.19 times (from 14.47 to 46.15), while reducing the proportion of cells accumulating during S phase and G1 phase. This suggests that hybrid 4b induces a cell cycle arrest in the G2/M phase. Furthermore, the pre-G1 phase cell count increased 24.85 times (from 1.71 to 42.51%), indicating that hybrid 4b induced apoptosis (Supplementary Table 37).

Figure 6. Investigation of cell apoptosis.

(A) DNA content in Cell cycle of T47D cancer cells after treatment with 4b (red color) and DMSO control (blue color). (B) Cell cycle analysis of T47D treated with DEMSO control (i) and treated with 4b (ii). (C) The effect of DMSO (i) and 4b (ii) on the percentage of annexin C-FITC-positive staining in T47D cells. (D) The ratio of early and late apoptosis in T47D cells upon the effect of 4b (blue color) compared with that of DMSO (red color).

Annexin V-FITC apoptosis determination

To explore the potential of the target dual-acting agents in inducing apoptosis, the capability of the most powerful hybrid, 4b, was evaluated on T47D. Annexin V/PI staining was used to study its effects in various stages with flow cytometry. Phosphatidylserine (PS) was transferred from the inside of the plasma membrane to the outside of the cell during the initial phases of apoptosis [48]. In this experiment, PS interacts with labeled Annexin V (conjugated to FITC), which fluoresces green to detect PS and shows the early phases of apoptosis [42]. PI, a fluorescent counterstain used in this assay, intercalates into the nucleic acids of dead cells only (it cannot enter the nuclei of living cells), indicating that apoptosis is in its final stages. Due to the use of PI/Annexin V, this assay also distinguishes between apoptosis and necrosis. In summary, the hybrid 4b results demonstrated that it raised the ratios of early and late apoptosis from 0.28 to 23.26%, respectively, in comparison to T47D cells treated with DEMSO. This revealed that hybrid 4b raised necrosis from 1.36 to 13.07%, an increase of up to 9.61 and 84.11 times (in comparison to the control) for both early and late apoptosis, suggesting that instead of using the cellular programmed death mechanism, hybrid 4b enhanced the apoptotic necrotic route (Supplementary Table 38 & Figure 6C & D).

Compound 4b induced cell cycle arrest via the HDAC/ tubulin inhibition pathway

Intracellular signaling was investigated using western blotting analysis to gain an understanding of the molecular mechanisms by which 4b induces G2/M phase arrest (Figure 7). In the cellular response to DNA-damaging chemicals, HDAC and tubulin have been found to be essential for controlling cell cycle and cell death (Supplementary Table 37). G2/M phase cell cycle arrest is brought on by a decrease in HDAC expression relative to control. Based on the available data, 4b exhibited a dose-dependent decrease in tubulin protein expression (Figure 7B). These outcomes align with the previous results.

Figure 7. Western blotting analysis of 4b.

Levels of (A) HDAC is decreased in comparison to control as well as (B) tubulin down-expression relative to β-actin as a control.

HDAC: Histone deacetylase; OD: Optical density.

Molecular modeling studies

The most powerful hybrid 4b binding mode with the target enzymes HDAC1 (PDB entry: 5ICN) and HDAC6 (PDB entry: 5EF8) has been predicted via docking study. The most potent hybrid tubulin/HDAC inhibitor, 4b, was investigated using the Discovery Studio software package to determine its binding mode with the tubulin active site (PDB entry: 1SA0). The co-crystallized ligands were redocked into the enzymes active sites and displayed a root mean square deviation of less than 2 (ranging from 0.698 to 1.247 Å), indicating the reliability of the produced docking results.

HDAC1 enzyme crystal structures were docked with the well-known HDAC inhibitor SAHA (PDB entry: 5ICN). The findings are displayed in Supplementary Table 39 & Figure 8A and showed that, as previously reported, SAHA formed four hydrogen bonds with amino acid residues His18, Gly27, Lys31 and Lys331, one pi–cation interaction with Lys331 and one pi–pi T-shaped interaction with Tyr336. Its CDOCKER energy was -33.563 kcal/mol and its CDOCKER interaction energy was -39.455 kcal/mol [46,49,50].

Figure 8. Docking and binding modes.

(A) SAHA into the HDAC1 active site (PDB entry: 5ICN); (i) 3D structure of SAHA (red) and (ii) 2D structure of SAHA (red). (B) Hybrid 4b into the HDAC1 active site (PDB entry: 5ICN). (i) 3D structure of hybrid 4b (violet) and (ii) 2D structure of hybrid 4b (violet). (C) Hybrid 4b into the HDAC6 active site (PDB entry: 5EF8). (i) 3D structure of hybrid 4b (violet) and (ii) 2D structure of hybrid 4b (violet). (D) Combrestatin (blue). (i) 3D structure (ii) 2D structure. (E) Compound 6 (orange); (i) 3D structure, (ii) 2D structure and 4b (violet), (iii) 3D structure and (iv) 2D structure into the active site of tubulin (PDB entry: 1SA0).

PDB: Protein Data Bank.

Hybrid 4b docking into HDAC1 (PDB entry: 5ICN) revealed that it was firmly attached to the enzyme’s binding active site, forming 5-H bonds and having a CDOCKER energy of -27.097 kcal/mol and a CDOCKER interaction energy of -46.273 kcal/mol. As shown in Figure 8B & Supplementary Table 39, the hydroxamic acid group’s carbonyl oxygen (C=O) formed a hydrogen bond with Tyr336, and the oxygen of the group formed a second hydrogen bond with Lys331. Moreover, the proton of the hydroxamic acid group made a third hydrogen link with Glu335 while the nitrogen of the CN group made two hydrogen bonding with Lys31 and Lys331. In addition, hybrid 4b showed many hydrophobic interactions with the His18, Lys31, Arg270, Lys305, Tyr336 and Gln338 amino acid residues.

In addition, hybrid 4b was docked into the HDAC6 active site (PDB entry: 5EF8), where it formed 3-H bonds and were nicely bound to the binding active site of HDAC6 with a CDOCKER energy of -18.350 kcal/mol and a CDOCKER interaction energy of -59.654 kcal/mol. As shown in Figure 8C & Supplementary Table 39, the hydroxamic acid group’s carbonyl oxygen (C=O) established a hydrogen link with His614 and the oxygen of the group’s hydroxamic acid produced a second hydrogen bond with His573, while the group’s proton formed a third hydrogen bond with Gly582. In addition, hydroxamic acid’s carbonyl oxygen (C=O) participated in the metal acceptor link with ZN:2001. There were also a lot of hydrophobic contacts between hybrid 4b and amino acid residues Pro711, Leu712 and Phe583.

To investigate the hybrid 4b capacity to bind to the tubulin active site (PDB entry: 1SA0), CA-4 and the template compound 1 were docked first. The docking results of CA-4 (Figure 8D & Supplementary Table 39) into the active site of tubulin (PDB entry: 1SA0) showed a CDOCKER energy of -11.863 kcal/mol and a CDOCKER interaction energy of -41.168 kcal/mol, and demonstrated that CA-4 interacted hydrophobically with amino acid residues Lys352, Val238, Val328 and Ala316.

The docking results of compound 1 (Figure 8E & Supplementary Table 39) in the active site of tubulin (PDB entry: 1SA0) [51] showed a CDOCKER energy of -14.190 kcal/mol and a CDOCKER interaction energy of -46.841 kcal/mol, and also showed that it interacted in the formation of 3-H bonds with Thr179, Val181 and Cys241 amino acid residues. In addition, compound 1 established several hydrophobic contacts with the Gln11, Ala180, Ala250, Ala316, Leu255, Leu248, Cys248, Lys254, Ala317 and Lys352 amino acid residues. Furthermore, the docking results of hybrid 4b into the tubulin active site (PDB entry: 1SA0) demonstrated that it was tightly bound to the tubulin binding active site, forming 3-H bonds and having a CDOCKER energy of -21.398 kcal/mol and a CDOCKER interaction energy of -61.754 kcal/mol. Hydroxamic acid’s oxygen forms a hydrogen link with Tyr202, its proton forms a hydrogen bond with Val238 and, last, the nitrogen of the CN group forms a hydrogen bond with Ala250, as illustrated in Figure 8 & Supplementary Table 39). In addition, hybrid 4b displayed a high number of hydrophobic interactions with amino acid residues Asn258, Val315, Asn350, Lus254, Gln247, Glu183, Leu248, Met259 and Lys352.

Collectively, once the docking data were obtained in agreement with the findings of the biological study, the authors were able to draw the conclusion that hybrid 4b has the potential to be a highly desirable future lead targeted anticancer candidate with dual HDAC/tubulin inhibitory action.

Conclusion

Herein, a cyano-2-substituted pyridine scaffold was employed as a cap group in the creation of novel HDAC/tubulin inhibitors that target multiple targets. The target compounds were tested using 60 cancer cell lines. In vitro one-dose testing of the target derivatives revealed that hydroxamic acids 4a–c were superior in activity to their corresponding –COOH and benzamide ZBG; the hydoxamate derivative 4b had the highest activity against 13 cancer cell lines. The optimal ZBG for the activity seen in hybrid 4b was hydroxamic acid with an aliphatic four-carbon linker spacer, according to independent cytotoxicity screening of hybrids 4b, 4c, 5b and 5c. Hybrid 4b showed higher activity than SAHA against renal cancer cell line UO-31 and showed promising activity against the breast cancer cell line T47D compared with SAHA (IC₅₀ = 3.52 ± 0.30, 5.5 ± 0.40 and 3.62 ± 0.30 μM, respectively). Screening of HDAC inhibitory activity of the target compounds indicated that those with aliphatic linkers and hydroxamic acid ZBG experienced the maximum activity, with an IC₅₀ of 72.45 ± 4.35 and 68.66 ± 4.13 nM for 5b and 5c, respectively, compared with SAHA (57.63 ± 2.40 nM). Meanwhile, a selectivity study against HDAC1 and HDAC6 isoforms showed that the evaluated compounds were nonselective aside from hybrid 5c, which is more selective against HDAC1 (IC₅₀ = 78.60 nM) over HDAC6 (145 nM). Moreover, 4b showed the most potent tubulin polymerization inhibition compared with vinblastine. Accordingly, cell cycle analysis of 4b possessed pre-G1 apoptosis and arrest. Furthermore, a study on molecular docking was conducted to predict binding into the pocket of HDAC1 and -6 and tubulin polymerization protein. These findings suggest that compound 4b could be viewed as a fundamental discovery in medicinal chemistry that merits additional investigation and advancement as a potential multitarget anticancer agent that works by inhibiting HDAC and dual tubulin polymerization.

Summary points

• This work aimed to synthesize a new series containing a cyano-2-substituted pyridine scaffold (3a–c, 4a–c and 5a–c) as a cap group with various linkers and zinc-binding groups that were identified via different spectroscopic techniques.

• The design of the synthesized compounds generated new multitargeting histone deacetylase (HDAC)/tubulin inhibitors.

• The target compounds were examined against 60 cancer cell lines according to the NCI-60 one-dose assay, which revealed promising growth inhibition for most of the target compounds.

• Hydroxamic acid-containing hybrids possessed higher potency in HDAC inhibition than other zinc binding group hybrids.

• Screening of the HDAC inhibitory activity of the target compounds resulted in potential activity of compound 4b.

• Selectivity studies of the HDAC1 and HDAC6 isoforms were evaluated for the 4b, 4c and 5c hybrids against vorinostat (SAHA).

• Tubulin polymerization inhibition compared with vinblastine indicates comparable potency of 4b to the reference.

• Cell cycle analysis for 4b was assayed to investigate pre-G1 apoptosis, and arrest that revealed 4b exhibited pre-G1 apoptosis and cell cycle arrest at the G2/M phase.

• To demonstrate and validate the strong binding into the pockets of HDAC1 and -6 and tubulin polymerization protein, molecular docking research was conducted.

• Compound 4b has been recognized as a novel candidate for anticancer treatment.

Acknowledgments

The authors extend their appreciation to the co-author Dr Hanan A Al-Ghulikah who provided them with the publication fees through Princess Nourah bint Abdulrahman University Researchers Supporting Project number (PNURSP2024R95), Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia.

Supplementary data

To view the supplementary data that accompany this paper please visit the journal website at: www.tandfonline.com/doi/suppl/10.4155/fmc-2023-0336

Financial disclosure

The work was supported by Princess Nourah bint Abdulrahman University Researchers Supporting Project no. PNURSP2024R95. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Competing interests disclosure

The authors have no competing interests or relevant affiliations with any organization or entity with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Writing disclosure

No writing assistance was utilized in the production of this manuscript.

Additional information

Funding

The authors extend their appreciation to the co-author Dr Hanan A Al-Ghulikah who provided them with the publication fees through Princess Nourah bint Abdulrahman University Researchers Supporting Project number (PNURSP2024R95), Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia.