Novel 9-Benzylaminoacridine Derivatives as Dual Inhibitors of Phosphodiesterase 5 and Topoisomerase II for the Treatment of Colon Cancer

It has been shown that phosphodiesterase 5 (PDE5) inhibitors have anticancer effects in a variety of malignancies in both in vivo and in vitro experiments. The role of cGMP elevation in colorectal carcinoma (CRC) has been extensively studied. Additionally, DNA topoisomerase II (Topo II) inhibition is a well-established mechanism of action that mediates the effects of several approved anticancer drugs such as doxorubicin and mitoxantrone. Herein, we present 9-benzylaminoacridine derivatives as dual inhibitors of the PDE5 and Topo II enzymes. We synthesized 31 derivatives and evaluated them against PDE5, whereby 22 compounds showed micromolar or sub-micromolar inhibition. The anticancer activity of the compounds was evaluated with the NCI 60-cell line testing. Moreover, the effects of the compounds on HCT-116 colorectal carcinoma (CRC) were extensively studied, and potent compounds against HCT-116 cells were studied for their effects on Topo II, cell cycle progression, and apoptosis. In addition to exhibiting significant growth inhibition against HCT116 cells, compounds 11, 12, and 28 also exhibited the most superior Topo II inhibitory activity and low micromolar PDE5 inhibition and affected cell cycle progression. Knowing that compounds that combat cancer through multiple mechanisms are among the best candidates for effective therapy, we believe that the current class of compounds merits further optimization and investigation to unleash their full therapeutic potential.


Introduction
Phosphodiesterase 5 (PDE5) inhibitors are a well-established pharmacological class that act by increasing intracellular levels of the second messenger cGMP [1]. Due to their effect on smooth muscles, PDE5 inhibitors have been approved for the treatment of

Chemistry
The syntheses of the planned compounds were achieved through a four-step sis as illustrated in Scheme 1. The synthesis of substituted N-phenylanthranilic acid A3) was carried out using the Ullmann reaction between 2-chlorobenzoic acid a respective aniline derivative in the presence of copper as a catalyst under high-te ture conditions. Subsequently, the N-phenylanthranilic acid was self-condense conc. sulfuric acid to produce acridones (B1-B3), which were then converted to spective 9-chloroacridines (C1-C3) via reflux in POCl3. The target compounds w tained by reacting benzyl amine analogs with 9-chloroacridine derivatives in etha der reflux in the presence of potassium carbonate as a base. (

Chemistry
The syntheses of the planned compounds were achieved through a four-step synthesis as illustrated in Scheme 1. The synthesis of substituted N-phenylanthranilic acids (A1-A3) was carried out using the Ullmann reaction between 2-chlorobenzoic acid and the respective aniline derivative in the presence of copper as a catalyst under high-temperature conditions. Subsequently, the N-phenylanthranilic acid was self-condensed with conc. sulfuric acid to produce acridones (B1-B3), which were then converted to the respective 9-chloroacridines (C1-C3) via reflux in POCl 3 . The target compounds were obtained by reacting benzyl amine analogs with 9-chloroacridine derivatives in ethanol under reflux in the presence of potassium carbonate as a base.
Molecules 2023, 28, x FOR PEER REVIEW 3 of 21 Figure 1. General structure representing 9-benzylaminoacridines as well as our previously reported thienopyrimidine PDE5 inhibitors.

Chemistry
The syntheses of the planned compounds were achieved through a four-step synthesis as illustrated in Scheme 1. The synthesis of substituted N-phenylanthranilic acids (A1-A3) was carried out using the Ullmann reaction between 2-chlorobenzoic acid and the respective aniline derivative in the presence of copper as a catalyst under high-temperature conditions. Subsequently, the N-phenylanthranilic acid was self-condensed with conc. sulfuric acid to produce acridones (B1-B3), which were then converted to the respective 9-chloroacridines (C1-C3) via reflux in POCl3. The target compounds were obtained by reacting benzyl amine analogs with 9-chloroacridine derivatives in ethanol under reflux in the presence of potassium carbonate as a base. ( All the synthesized acridine derivatives were tested for their in vitro ability to inhibit PDE5A at a screening dose of 10 µM. IC 50 values were determined for compounds exhibiting percentage inhibitions above 50%. The results are shown in Table 1.
Structure-activity relationship of the PDE5 inhibitory activity: Compound 1 was initially synthesized with an unsubstituted benzyl giving an IC 50 = 9.33 µM, indicating that our assumption about the possible PDE5 inhibition by this scaffold was correct. Compounds with different substituents on the benzyl ring were synthesized in attempts to enhance their potency, as discussed below.
Monosubstitution of the phenyl ring: Monosubstitution with electron-donating groups was investigated, starting with the lipophilic, electron-donating methyl group in the ortho, meta, and para positions, as seen in compounds 2, 3, and 4, respectively. The methyl group in the para position showed slightly higher activity (compound 4, IC 50 = 4.44 µM) compared with the ortho (2) and meta (3) isomers, which exhibited IC 50 values of 9.87 µM and 6.40 µM, respectively. On the other hand, the use of the more polar electron-donating methoxy group resulted in the deterioration of the inhibitory activity whether in the ortho, meta, or para positions, as exhibited by compounds 5, 6, and 7, respectively.
Mono-halogen substitutions with fluorine and chlorine were made at different positions of the phenyl ring to produce compounds 8-13. Fluoro-and chloro-substituents in the para position (compounds 10 and 13, respectively) reduced the inhibitory activity compared with the unsubstituted benzyl in compound 1. Meanwhile, similar inhibitory activity was exhibited by compounds that had either of these substituents in the meta position (9 and 12, IC 50 = 6-7 µM). However, the fluoro-substituent in the ortho position (compound 8, IC 50 = 7.16 µM) showed activity 1.5 times lower than that of the chloro-substituent in the same position (compound 11, IC 50 = 4.33 µM).
Disubstitution of the phenyl ring: All compounds containing dimethoxy phenyl (16)(17)(18), similar to the mono-substituted methoxy analogs, did not show any improvement in the activity.
Additionally, disubstitution of the phenyl ring with fluorine at variable positions (compounds 20, 21, and 22) led to a reduction in potency. Exceptionally, compound 20 (2,3difluorobenzyl) showed an IC 50 of 7.25 µM, which was comparable to the mono-substituted ortho and meta fluoro-analogs. While the methoxy group failed to boost the activity against PDE5 in mono-and disubstituted methoxy compounds, combining the p-methoxy with m-fluoro in compound 19 (3-fluoro-4-methoxy) seemed to have a synergetic effect that resulted in an improvement in the activity (IC 50 = 3.39 µM).
Replacement of the phenyl ring from the benzylamino group: Modifications in the structure via the replacement of the phenyl ring with furan, thiophene, and 4-(2thiophenephenyl) gave rise to compounds 14, 15, and 24, respectively. This resulted in some improvement in the PDE5 inhibitory activity compared with compound 1, with the exception of compound 24, which exhibited reduced activity.
Increasing the spacer length: Compound 23 demonstrated the effect of increasing the linker length in the benzylamino moiety via its replacement with phenylethyl amino. This modification showed some improvement in the activity (IC 50 = 6.68 µM) compared with the benzyl derivative, compound 1, exhibiting an IC 50 = 9.33 µM. All the synthesized acridine derivatives were tested for their in vitro ability to inhibit PDE5A at a screening dose of 10 µM. IC50 values were determined for compounds exhibiting percentage inhibitions above 50%. The results are shown in Table 1. Structure-activity relationship of the PDE5 inhibitory activity: Compound 1 was initially synthesized with an unsubstituted benzyl giving an IC50 = 9.33 µM, indicating that our assumption about the possible PDE5 inhibition by this scaffold was correct. Compounds with different substituents on the benzyl ring were synthesized in attempts to enhance their potency, as discussed below. Substitution at C2 of the acridine: A few derivatives were synthesized to have methoxy or methyl substituent at C2 of the acridine. These modifications were adopted together with the use of different para-monosubstituted benzylamino groups at C9. The methyl group in the para position of the benzylamino along with a methoxy group at C2 of the acridine exhibited the highest activity against PDE5 in the present series (compound 26, IC 50 = 0.83 µM), indicating that the methoxy group at C2 can be a key modification for boosting the inhibitory potency of the scaffold against PDE5. This observation was further confirmed by comparing compound 25 (containing 2-methoxy acridine and an unsubstituted benzylamino, IC 50 = 3.61 µM) with compound 1 (IC 50 = 9.33 µM), the respective analog with unsubstituted acridine. Such a comparison showed that the methoxy at position two led to about a three-fold improvement in the activity. Ultimately, the scaffold exhibited low micromolar inhibition against PDE5 with several substituents tolerated on the benzyl moiety. Moreover, the use of a methoxy substituent at position two of the acridine boosted the inhibitory activity to sub-micromolar values.

Testing the Anticancer Activity of the Compounds NCI 60-Cell Line Testing
The newly synthesized compounds were screened for their in vitro anticancer activity at the Developmental Therapeutic Program (DTP) of the National Cancer Institute (NCI). The screening involved a panel of 60 human tumor cell lines consisting of 9 tissue types to screen for potential anti-cancer agents. The primary screening of submitted compounds was performed by testing a single dose (10 −5 M) in the full NCI-60 panel. The values for the mean % growth inhibition of the 60 cell lines by compounds (1-14) and (16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30)(31) are shown in Table 2. The detailed % inhibition by the tested compounds against the 60 cell lines is shown in Tables S1-S3 in the Supporting Information. One dose assay was conducted on 30 compounds accepted by the NCI (1-14 and 16-31). Almost all synthesized compounds exhibited a mean growth inhibition above 50% and exhibited high activity against some of the investigated cell lines. The 2-methoxy acridine along with 4-fluoro-(28) or 4-chloro-(29) on the benzyl ring showed high mean % growth inhibition (92. 16 and 93.27%, respectively). Moreover, the unsubstituted acridine with the extended 4-(2-thienyl) benzylamino (24) showed the highest mean growth inhibition at 96.88%. This suggests that the electron-donating group at C2 of the acridine with the electron-withdrawing group in the para position of the benzylamino (compounds 28 and 29) or the unsubstituted acridine with the extended aryl benzylamino (compound 24) are important for anticancer activity.

Effects of Compounds on Proliferation of Human Colorectal Cancer Cells and Non-Cancer Cells
Since HCT-116 CRC cells are among the most studied cells in the context of the anticancer activity of PDE5 inhibitors [38,39], the compounds were screened against this cancer cell line for more detailed investigations. Additionally, human non-malignant cells (CCD-966SK) were used as a control cell line to study the selectivity of the compounds ( Table 3). The results showed that 19 compounds had a significant inhibitory effect on cell survival, among which compounds 1-4, 10, 8, 12, 20, 21, and 28 exhibited high cytotoxicity (IC 50 < 5 µM). Compounds 5, 9, 14, 11, 24, 25, and 31 were found to have moderate toxicity to HCT-116 cells (IC 50 < 10 µM). Compounds 2, 3, 10, and 28, however, were toxic to human non-malignant dermal fibroblast CCD-966SK cells. Conclusively, the above results indicate that the cytotoxicity of compounds 1, 3, 4, 8, 12, 20, and 21 was more selective for HCT-116 cells. The prominent structural features that enhanced the anticancer activity against HCT-116 cells are certainly worth mentioning. It should be noted that having orthoand/or para-fluorinated benzyl groups at position nine of the acridine is an important feature in many potent growth inhibitors. This could be observed in compounds 8, 10, 21, and 28, such that all these compounds showed an IC 50 < 4 µM. These compounds had 2-fluorobenzyl, 4fluorobenzyl, 2,4-difluorobenzyl, and 4-fluorobenzyl substituents, respectively, at position nine of the acridine nucleus. On the other hand, having a metaor para-methoxy substituent (alone or in combination with other substituents) at the 9-benzyl moiety reduced anticancer activity, as exhibited by compounds 6, 7, 16, 17, 18, 19, and 27 (with ≤50% growth inhibition at 10 µM).

Effects of Compounds on Topo II Activity
It has been previously reported that 9-benzylaminoacridines inhibit Topo II [34]. Consequently, the effects of the synthesized compounds on Topo II activity were evaluated with a cell-free assay using a purified human DNA Topo II-protein-mediated relaxation of the negatively supercoiled pHOT1 plasmid DNA [40]. Compounds with cytotoxicity against HCT-116 (IC 50 < 10 µM) were evaluated for their activity against Topo II. Two Topo II inhibitors (doxorubicin and etoposide) were used as positive controls. The results show that compounds 1-5, 8, 9, 11, 14, 18, 20, 21, 22, 24, 25, 30, and 31 at 1 and 5 µM doses have a larger base pair band than nicked DNA, and these results are similar to those of doxorubicin. It is, therefore, speculated that these compounds may insert themselves into DNA and interfere with DNA replication (Figure 2). Notably, compound 11 was found to be the most potent Topo II inhibitor, such that it could dramatically increase the amount of supercoiled DNA at a very low concentration (0.04 µM). This finding suggests that compound 11 inhibits Topo II activity with a higher potency than doxorubicin and that it can, additionally, intercalate into DNA, interfering with DNA replication at high doses. Furthermore, compounds 12 and 28 are considered the second and third best Topo II inhibitors, showing their effects at 0.2 and 1 µM, respectively. the amount of supercoiled DNA at a very low concentration (0.04 µM). This finding suggests that compound 11 inhibits Topo II activity with a higher potency than doxorubicin and that it can, additionally, intercalate into DNA, interfering with DNA replication at high doses. Furthermore, compounds 12 and 28 are considered the second and third best Topo II inhibitors, showing their effects at 0.2 and 1 µM, respectively.

Chemistry
Solvents and reagents used were obtained from commercial suppliers and were use

Chemistry
Solvents and reagents used were obtained from commercial suppliers and were used without further purification. All organic solvents used were of HPLC grade. All starting materials were obtained from Sigma-Aldrich and Alfa Aesar and used without further purification. Column chromatography was carried out using silica gel 40-60 µM. Reaction progress was monitored with TLC using fluorescent pre-coated silica gel plates, and detection of the components was performed using short UV light (λ = 254 nm). 1 H-NMR spectra were run at 400 MHz, and 13 C-NMR spectra were run at 101 MHz using a Varian Mercury 400 plus in deuterated chloroform (CDCl 3 ) and DMSO-d6 . Chemical shifts (δ) are reported in parts per million (ppm), and all coupling constants (J) are given in Hz. Multiplicities are abbreviated as s: singlet, d: doublet, t: triplet, dd: doublet of doublet, ddd: doublet of doublet of doublet, ddt: doublet of doublet of triplet, and m: multiplet. The mass spectrometric analysis (UHPLC-ESI-MS) was performed using the Waters ACQUITY XevoTQD system, which consisted of an ACQUITY UPLC H-Class system (Waters Corp., Milford, MA, USA) and Xevo TM TQD triple-quadrupole tandem mass spectrometer with an electrospray ionization (ESI) interface (Waters Corp., Milford, MA, USA). An Acquity BEH C18 100 mm × 2.1 mm column (particle size, 1.7 µm) was used to separate analytes (Waters, Ireland). The solvent system consisted of water containing 0.1% formic acid (A) and 0.1% formic acid in acetonitrile (B). HPLC method: The flow rate was 200 µL/min. The percentage of B started at 5% and was maintained for 1 min, and then increased up to 100% during 10 min, kept at 100% for 2 min, flushed back to 5% in 3 min, and then kept at 5% for 1 min. The MS scan was carried out in the following conditions: capillary voltage of 3.5 kV, cone voltage of 20 V, radio frequency (RF) lens voltage of 2.5 V, source temperature of 150 • C, and desolvation gas temperature of 500 • C. Nitrogen was used as the desolvation and cone gas at a flow rate of 1000 and 20 L/h, respectively. System operation and data acquisition were controlled using Mass Lynx 4.1 software (Waters). The purity of the tested compounds was determined using HPLC coupled with a mass spectrometer and was higher than 95% for all compounds. Melting points were determined on a Büchi B-540 Melting Point apparatus (Flawil, Switzerland) and were uncorrected.

General Synthetic Procedures Procedure A: General Procedure for Preparation of 2-Anilinobenzoic Acid Derivatives (A1-A3)
A mixture of 2-chlorobenzoic acid (1.9 g, 12 mmol), aniline (12 mmol), potassium carbonate (3.45 g, 25 mmol), Cu powder (0.2 g), and KI (0.1 g) in 50 mL of DMF was heated under reflux overnight. After the complete reaction, DMF was evaporated under vacuum followed by extraction using methylene chloride and water. The aqueous layer was collected, filtered under vacuum and subsequently acidified using HCl: H 2 O (1:1). A white precipitate was obtained and left to dry to obtain the 2-anilinobenzoic acid derivatives in different yields. Compounds (A1-A3) were used for the next step without further purification.
2-(4-Methylanilino)benzoic acid (A3). The compound was synthesized according to procedure A, using 2-chlorobenzoic acid and 4-methyl aniline to produce a white powder (51% yield); mp 195-197 • C [41]. The appropriate compound (A1-A3) (14 mmol) was added to 10 mL of conc. H 2 SO 4 and heated at a temperature of 100-110 • C in an oil bath for 4 h. Afterwards, the mixture was poured slowly into an ice/water mixture to produce a yellow precipitate, which was then obtained via vacuum filtration and left to dry to obtain the acridone derivatives in different yields. No further purification was required.
Procedure C: General Procedure for Synthesis of 9-Chloroacridine Derivatives (C1-C3) The appropriate compound (B1-B3) (10 mmol) was added to 10 mL of POCl 3 , and the mixture was heated in an oil bath at a temperature of 100-110 • C for 3 h. After completion of the reaction (confirmed using TLC), the mixture was added dropwise into an excess ice/ammonia solution mixture. Subsequently, extraction using methylene chloride was conducted, whereby the organic phase was collected and dried over anhydrous magnesium sulfate. The methylene chloride layer was then evaporated under reduced pressure to produce a light green powder. Compounds (C1-C3) were used in the next step without further purification. 9-Chloroacridine (C1). The compound was synthesized according to procedure C, using compound B1 to produce a light green powder (92% yield); mp 117-119 • C [44].
9-Chloro-2-methylacridine (C3). The compound was synthesized according to procedure C, using compound B3 to produce a light green powder (83% yield); mp 145-147 • C [46]. The respective benzylamine (2 mmol) was added to 15 mL of absolute alcohol and potassium carbonate (0.276 g, 2 mmol), and the mixture was stirred at room temperature for 45 min. The appropriate compound (C1-C3) (1 mmol) and KI (0.415 g, 0.25 mmol) were then added. The mixture was heated under reflux overnight at 80 • C. After the completion of the reaction, excess ethanol was evaporated under vacuum, followed by extraction performed using ethyl acetate and brine. The organic phase was collected after drying over anhydrous MgSO 4 and concentrated under vacuum. The product was purified using silica gel column chromatography.

Cell Proliferation Activity Assay
In the MTT assay, 96-well culture plates were used. A total of 7 × 10 4 cells per well were seeded, and test materials were added to the cells at various concentrations [47,48]. Human cancer and non-cancer cell lines were purchased commercially from the American Type Culture Collection (ATCC). After 72 h, the cytotoxicity of the compounds and paclitaxel (as a positive control group) was detected using the MTT method (thiazolyl blue tetrazolium bromide, Sigma-M2128). Absorbance values at 570 and 620 nm (OD = OD570-OD620) were measured using an ELISA reader (Anthos Labtec Instruments, Salzburg, Austria). IC 50 values were determined for compounds that showed more than 50% growth inhibition at 10 µM, using concentrations of 1, 2.5, 5, and 10 µM.

Apoptosis Assay
To assess the mechanism of HCT-116 cell death induced by the compounds, an apoptosis assay was conducted with a flow cytometer using a FITC Annexin-V apoptosis detection kit (BD, Biosciences). The number of apoptotic/necrotic cells was analyzed using flow cytometry following the manufacturer's protocol. Markers were assessed using a FACS-Caliburflow cytometer (Beckman Coulter, Taipei, Taiwan) and analyzed using WinMDI 2.8 software.

Cell Cycle Arrest Analysis
To analyze cell cycle changes, cells permeated at 1 × 10 6 × cells/mL were trypsinized and diluted and then fixed in 70% ethanol at 4 • C. Cell cycle arrest in HCT-116 cells following compound exposure was accomplished using propidium iodide (Sigma-Aldrich, St. Louis, MI, USA). The markers were then assessed using a FACS-Caliburflow cytometer (Beckman Coulter, Taipei, Taiwan) and analyzed using WinMDI software. Markers were assessed using a FACS-Caliburflow cytometer (Beckman Coulter, Taipei, Taiwan) and analyzed using WinMDI 2.8 software [49,50].

Determination of Topoisomerase II Inhibition
The assay was performed following the manufacturer's protocol, using a standard relaxation reaction mix (20 µL) containing 50 mM of Tris-HCl (pH 8.0), 10 mM of MgCl 2 , 200 mM of potassium glutamate, 10 mM of dithiothreitol, 50 µg/mL of bovine serum albumin, 1 mM of ATP, 0.3 µg of pHOT1 plasmid DNA, 8 units of human topo II (Topogen, Columbus, OH, USA), and various compounds (of 1 and 10 µM) at 37 • C for 30 min. The reaction was terminated by adding 2 µL of 10% SDS, followed by 2.5 µL of proteinase K (50 µg/mL) to digest the bound protein. The system was incubated at 37 • C for 15 min. The DNA product was finally analyzed using electrophoresis on a vertical 1.5% agarose gel at 2 volts/cm in 1 × TAE buffer and photographed using the Eagle Eye II system (Stratagene, La Jolla, CA, USA).

Phosphodiesterase 5 Assay
The enzyme activities of phosphodiesterase 5 (PDE5) were measured using an HTRF cGMP assay kit (Cisbio, Codolet, France). PDE5A (0.5 ng/µL; BPS Bioscience, San Diego, CA, USA) were incubated with DMSO (0.1%, as a control), test compounds, or tadalafil at 37 • C for 10 min, and then cGMP (500 nM) was added and incubated for another 45 min. Afterward, the HTRF reagent (cGMP-d2 and anti-cGMP-cryptate) was added for 1 h at 25 • C, and PDE activity was assayed on an HTRF reader (Tecan Infinite F200 Pro). Tadalafil was used as a positive control, and it showed an IC 50 of 10 nM in the assay conditions.

Conclusions
PDE5 inhibitors have been reported in several studies in the context of dually acting compounds [49,[51][52][53]. Herein, we provided first-in-class compounds with dual inhibition of PDE5 and Topo II. The scaffold was found to be active against PDE5 at low micromolar concentrations, with several substituents tolerated on the benzyl moiety. The 2-methoxy substituent of the acridine proved to be an important feature for boosting the activity on a sub-micromolar level, encouraging its presence during future synthesis. Most of the synthesized 9-benzylaminoacridines possessed significant anticancer activity, as indicated by the results obtained via NCI testing. Compounds 11, 12, and 28, which inhibited PDE5 at low micromolar IC 50 values, exhibited the most potent inhibition of Topo II. Finally, active growth inhibitors had variable effects on cell cycle progression and apoptosis in HCT-116 cells, which requires further investigation of the detailed mode of action of the present series.