Synthesis, Molecular Structure, Anticancer Activity, and QSAR Study of N-(aryl/heteroaryl)-4-(1H-pyrrol-1-yl)Benzenesulfonamide Derivatives

A series of N-(aryl/heteroaryl)-4-(1H-pyrrol-1-yl)benzenesulfonamides were synthesized from 4-amino-N-(aryl/heteroaryl)benzenesulfonamides and 2,5-dimethoxytetrahydrofuran. All the synthesized compounds were evaluated for their anticancer activity on HeLa, HCT-116, and MCF-7 human tumor cell lines. Compound 28, bearing 8-quinolinyl moiety, exhibited the most potent anticancer activity against the HCT-116, MCF-7, and HeLa cell lines, with IC50 values of 3, 5, and 7 µM, respectively. The apoptotic potential of the most active compound (28) was analyzed through various assays: phosphatidylserine translocation, cell cycle distribution, and caspase activation. Compound 28 promoted cell cycle arrest in G2/M phase in cancer cells, induced caspase activity, and increased the population of apoptotic cells. Relationships between structure and biological activity were determined by the QSAR (quantitative structure activity relationships) method. Analysis of quantitative structure activity relationships allowed us to generate OPLS (Orthogonal Projections to Latent Structure) models with verified predictive ability that point out key molecular descriptors influencing benzenosulfonamide’s activity.


Introduction
Nowadays, cancer causes one in seven deaths worldwide, and has become more fatal than AIDS (acquired immune deficiency syndrome), tuberculosis, and malaria combined. As was estimated by the International Agency for Research on Cancer, total cancer deaths in 2012 reached 8.2 million and it is expected that, by 2030 cancer deaths will surpass 13 million per year due to the growth and aging of the population [1].
In view of the importance of sulfonamides and nitrogen-containing heterocycles as privileged structures for the design of anticancer agents, we decided to explore the synthesis and anticancer activity of molecular hybrids obtained by the combination of benzenesulfonamide and pyrrole fragments. These compounds were evaluated for their cytotoxicity against three human cancer cell lines: HeLa (cervical cancer), HCT-116 (colon cancer), and MCF-7 (breast cancer). For the most active compound, apoptosis-inducing activity toward cancer cell lines was further investigated. Analysis of the quantitative structure activity relationships (QSAR) allowed us to generate OPLS models (Orthogonal Projections to Latent Structure) with verified predictive ability that defined the key descriptors affecting the N-(aryl/heteroaryl)-4-(1H-pyrrol-1-yl)benzenesulfonamide's activity.
IR, 1 H NMR, and 13 C NMR spectroscopy confirmed the structures of the compounds. Additionally, 25 an X-ray analysis was undertaken. 4-Acetamido-N-(aryl/heteroaryl)benzenesulfonamides presented in the IR spectra typical absorption bands at 1662-1692 cm −1 , associated with C=O bond stretching. In the 1 H NMR spectra it showed singlets at approximately 2 ppm corresponding to an acetyl group and two NH singlets at about 10 ppm (see the Supplementary Material). An acetyl group has been also recognized as a peak at approximately 170 ppm in the 13 C NMR spectra (see the Supplementary Material). Hydrolysis of 4-acetamido-N-(aryl/heteroaryl)benzenesulfonamides caused the appearance of NH 2 stretching, as well as bending bands at 3498-3008 cm −1 and 1658-1620 cm −1 , respectively.
Details on data collection, structure solution, and refinement are given in Table 1. Compound 25 crystallized in the monoclinic system, and the space group P2 1 /n, with one molecule in the asymmetric unit, and four in the unit cell, Z = 4. The atom numbering scheme is presented in Figure 3. The sulfur atom coordination is close to tetrahedral. The amide group arrangement S(1)-N(1)-C(1) and phenyl ring C1-C6 are not in the same plane, suggesting a lack of coupling between the nitrogen lone pair and aromatic electrons. The phenyl ring C1-C6 and its etheric counterpart C7-C12 are inclined at a dihedral angle of 86.96 • (almost perpendicular), while the angle at O1 is rather typical, at 116 • .
Because of the substitution of the sulfonamide group, only one N-H hydrogen-bond donor is available. The intermolecular hydrogen bonding, of the N-H···O type, forms infinite chains, spreading parallel to the b axis ( Figure 4, Table 2). The oxygen atom O(3) involved in the interaction is a little more distant from S(1) than the other oxygen atom O(2) (1.4336 (14) vs. 1.4287(14) Å), which is expected and confirms hydrogen bonding. The chain has no additional symmetry so it is described by the simplest p1 (R1) rod group symmetry [22], and its topology is characterized as a chain with four atoms in the links, graph set symbol C(4) [23]. A weak hydrogen bond of the CH...O type could be also found, which operates within the main chains and may induce the actual conformation of aromatic rings ( Table 2). The ring-ring stacking interaction does not play a significant role, as the shortest distance between the ring's centroids is large (4.47 Å between C7-C12 and C13-C18). The crystal packing diagram suggests the chains are simply packed to maximize the solid density and van der Waals interactions ( Figure 5).

Screening for Anticancer Agents
Compounds 23-40 were evaluated in vitro for their effects on the viability of three human cancer cell lines: HCT-116 (colon cancer), HeLa (cervical cancer), and MCF-7 (breast cancer). Analysis was performed using the MTT assay after 72 h of incubation, and the calculated IC 50 values (a concentration required for 50% inhibition of cell viability) are given in Table 3.
In spite of our observed structure activity relationships, for a more objective correlation between structure and activity the QSAR methodology was applied.

QSAR Study
We decided to apply an OPLS-based approach that was successful in our previous studies [17,[19][20][21]. Three dimensional structures of all compounds were prepared by Gaussian software by density functional theory (DFT), which is a computational quantum mechanical modeling method. Next, in order to obtain easily interpretable QSAR models, we carefully selected molecular descriptors with clear definition directly reflecting the chemical structures.
Before application of the regression technique, outlying IC 50 values were discarded from further analysis. The OPLS calculations, as well as leave-one-out cross validation of the obtained model, were performed by SIMCA. We were able to achieve significant models for all tested cell lines ( Figure 6).
The interpretation of the obtained model was based on the calculated VIP (variable importance for projection) values, which sort molecular descriptors by their relative importance for prediction of the IC 50 . The table of VIP values for all descriptors is included in the supplementary material (Table S1). The supplementary material also contains the values of the selected descriptors (Tables S2-S4).

Cytotoxic Activity
On the basis of the results of the cytotoxic screening, compound 28 was chosen to investigate apoptosis-inducing activity in cancer cells. First, the cytotoxic activity of 28 was determined in a time-dependent manner towards HCT-116, HeLa, and MCF-7 cells with the MTT assay ( Figure 7). Cells were treated with 28 in the concentration range of 1-25 µM. For the HCT-116 cells after a 24 h incubation, the IC 50 value was reached at 4 µM, and a further 24 h incubation decreased this value to 3 µM. It did not decrease further with further treatment. In the case of the HeLa cells, the IC 50 values were 22, 10, and 7 µM after 24, 48, and 72 h, respectively. The cytotoxic activity of 28 also increased in a concentration-dependent manner towards the MCF-7 cells. After 24 h of incubation, the IC 50 value was 9.5 µM, further decreased to 7 µM after 48 h, and after 72 h of incubation the IC 50 value was 5 µM.

Apoptosis Induction
In apoptotic cells, the membrane phospholipid, phosphatidylserine, is exposed on the outer leaflet of the plasma membrane and can be detected by fluorochrome-conjugated phosphatidylserine-binding proteins such as Annexin V. In order to quantitatively determine the percentage of cells within a population that are actively undergoing apoptosis, phosphatidylserine externalization induced by 28 was examined by flow cytometric analysis. Cells were treated with 2.5, 5, and 10 µM of 28 for 48 h (HCT-116) and 72 h (HeLa and MCF-7), and stained with Annexin V-PE and 7-AAD. The results shown in Figure 8 indicate that compound 28 induced an increase in the population of early apoptotic cells (~10%) and also significantly increased the percentage of late apoptotic cells (35-40%) in HTC-116 at a concentration range of 2.5-10 µM. Furthermore, treatment of MCF-7 cells with 28 resulted in an increase of cells in the late stage of apoptosis (~40%). In the case of the HeLa cell line, early apoptotic cells (10%) were visible from a concentration of 2.5 µM, while late apoptotic cells increased visibly in a concentration-dependent manner from 7-35%.

Caspase Activation
Apoptotic cells were also identified by determination of levels of active caspases in the cells. Increased caspase activity is caused in response to pro-apoptotic signals and ensures the execution of apoptosis in the cell. Caspase activity induction was determined with the use of a fluorescently labelled caspase inhibitor-FAM-VAD-FMK (a carboxyfluorescein derivative of valylalanylaspartic acid fluoromethyl ketone)-that, through covalently binding with an active enzyme, allowed us to detect the fluorescent enzyme-inhibitor complex using flow cytometry. Increased caspase activation is shown by the increased fluorescence of the caspase inhibitor in the cell population, as indicated by the marker M1 in Figure 9. The results shown in Figure 9 indicate that 28 induced caspase activity in all tested cells in a dose-dependent manner, confirming apoptosis induction by this compound in HCT-116, HeLa, and MCF-7 cells.

Cell Cycle Distribution
The influence of 28 on changes in cell cycle phase distribution was analyzed with flow cytometry. HCT-116 cells were treated for 48 h with 2.5, 5, and 10 µM (28), whereas the HeLa and MCF-7 cells were treated with corresponding concentrations of 28 for 72 h. The results presented in the histograms in Figure 10 show

General Information
Melting points were measured using a Thermogalen (Leica, Vienna, Austia) apparatus and were uncorrected. IR spectra were obtained using a Nicolet iS5 FTIR spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) and spectra were measured in KBr pellets; the absorption range was 400-4000 cm −1 . 1 H NMR and 13 C NMR spectra were obtained using a Varian Gemini 200 apparatus (Varian, Palo Alto, CA, USA) or Varian Unity Plus 500 apparatus. Chemical shifts are expressed as δ values, and resonance multiplicity is described as s (singlet), br. s (broad singlet), d (doublet), dd (doublet of doublets), t (triplet), and m (multiplet). Elemental analyses were obtained using a PerkinElmer 2400 Series II CHN Elemental Analyzer (PerkinElmer, Shelton, CT, USA), and the results indicated by the symbols of the elements that were within ±0.4% of the theoretical values. Thin-layer chromatography (TLC), used for monitoring reactions and for the qualitative analysis of reaction products, was performed on Merck Kieselgel 60 F254 plates (Merck, Darmstadt, Germany) and visualized with UV. Gravity liquid chromatography was conducted using silica gel with a pore size of 60 Å, 220-440 mesh particle size, and 35-75 µm particle size.
Method B. To a solution of aminocomponent (3 mmol) in acetone (3 mL) and pyridine (0.9 mL), 4-acetamidobenzene-1-sulfonyl chloride (3 mmol, 701 mg) was added, and reaction mixture was stirred for 18-48 h at room temperature. Afterwards, reaction solvents were evaporated under reduced pressure, and the residue was suspended in ice slush (40 mL). Obtained mixture was acidified with dilute hydrochloric acid (1 M) to pH~2 and stirred for 1 h at room temperature. The precipitated solid was filtered, washed with water, and dried. Pure compounds were obtained after crystallization from ethanol.  119.1, 119.2, 124.1, 124.2, 126.1, 127.1, 128
Method B: The 4-acetamido-N-(aryl/heteroaryl)benzenesulfonamide (2.5 mmol) was heated at 100 • C with hydrochloric acid (4 mL, 36%) in ethanol (10 mL) for 1 h, than the solution was cooled, treated with cold water (20 mL), and basified with ammonia to pH~8-9. The solid was filtered off, washed with water (3 × 10 mL), and purified as indicated below.  The structure was solved using direct methods with SHELXS-13 program and refined by SHELXL-2013 [38] program run under control of WinGx [39]. Positions of the C-H hydrogen atoms were calculated geometrically and taken into account with isotropic temperature factors. The NH hydrogen atom was found in the Fourier residual electron density map and was refined without constraints.
Crystallographic data for structure of 25 has been deposited with the Cambridge Crystallographic Data Centre, No. CCDC 1825991. Copies of this information can be obtained free of charge on application to CCDC, 12 Union Road, Cambridge, CB21EZ, UK (Fax: +44-1223-336033; E-mail: deposit@ccdc.cam.ac.uk or Available online: http://www.ccdc.cam.ac.uk).

Cell Culture and Cell Viability Assay
All chemicals were purchased from Sigma-Aldrich (St. Louis, MO, USA). The MCF-7 and HeLa cell lines were obtained from Cell Lines Services (Eppelheim, Germany), and the HCT-116 cell line was ordered from ATCC (ATCC-No: CCL-247). Cells were cultured in Dulbecco's modified Eagle's medium (DMEM). Medium was supplemented with 10% fetal bovine serum, 100 units/mL penicillin, 2 mM glutamine, and 100 µg/mL streptomycin. Cultures were held in an incubator (HeraCell, Heraeus, Langenselbold, Germany) in a humidified atmosphere with 5% CO 2 at 37 • C.
Screening for cytotoxic activity: The MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide) assay was used to determine cell viability. Stock solutions of different compounds were obtained by dissolving them in 100% DMSO. To prepare working solutions samples, stock solutions were diluted with DMEM medium. The contents of DMSO in the treated samples did not exceed 0.5%. 96-Well plates were inoculated with cells at a density of 5 × 10 3 cells/well; then, medium containing the test compound at different concentrations (1, 10, 25, 50, and 100 µM) was added to each well and incubated for 72 h. After treatment, cells were incubated for 2 h with MTT (0.5 mg/mL) at 37 • C. Subsequently, cells were lysed with DMSO, and absorbance of the formazan solution was measured at 550 nm (1420 multilabel counter, Victor, Jügesheim, Germany). Data were expressed as the mean ± SD of at least three independent experiments carried out in triplicate.
Cells were seeded in 96-well plates at a density of 5 × 10 3 cells/well and treated for 24, 48, and 72 h with the examined compound in the concentration range 1-25 µM. The following steps of the experiment were carried out as indicated above in Screening for cytotoxic activity section.
Detection of apoptosis by Annexin V-PE and 7-AAD staining: Apoptosis induction was detected with an Annexin V-PE Apoptosis Detection Kit I (BD Biosciences, Erembodegem, Belgium) according to the manufacturer's instructions. Cells were treated with 28 (2.5, 5, and 10 µM) for 48 h (HCT-116) and 72 h (HeLa and MCF-7). Following treatment cells were collected, washed with Annexin-binding buffer, and stained with Annexin V-phycoerythrin (PE) and 7-amino-actinomycin (7-AAD) for 15 min at RT in the dark. Acquisition was performed on a FACSCalibur cytometer (BD) and data were analyzed with Flowing software (version 2.5) (Turku, Finland).

Caspase Activity Determination:
The FLICA Apoptosis Detection Kit (Immunochemistry Technologies, Bloomington, IN, USA) was used to determined caspase activity. Cultured cells were treated with 28 (2.5, 5, and 10 µM) for 24 h (HCT-116) and 48 h (HeLa and MCF-7). After collection of cells, the caspase inhibitor-a carboxyfluorescein-labeled fluoromethyl ketone peptide-was added. Cells were incubated at 37 • C in a humidified 5% CO 2 incubator for 1 h and washed with washing buffer. Flow cytometry (BD FACSCalibur, BD Bioscience, San Jose, CA, USA) was used to determine the amount of fluorescence emitted from inhibitors bound with the caspases corresponding to their activity.

Cell Cycle Distribution Analysis:
The effects of 28 on cell cycle distribution in HCT-116, HeLa, and MCF-7 cells were determined with flow cytometry analysis. Cells were treated with 28 (2.5, 5, and 10 µM) for 48 h (HCT-116) and 72 h (HeLa and MCF-7), after which they were fixed in cold 70% ethanol for 24 h. Fixed cells were treated with 100 µg/mL RNAse (Invitrogen, Darmstadt, Germany) and stained with 10 µg/mL PI (Invitrogen) for 30 min at RT. Acquisition was performed on a FACSCalibur cytometer (BD), and data were analyzed with Flowing software (version 2.5).

Molecular Modeling and QSAR Study
In order to find the lowest energy geometry of the studied compounds, before starting molecular modeling algorithms, the two-dimensional structure of molecules were created using of Gaussian software (Gaussian G09, v D.01, Gaussian Inc., Wallingford, CT, USA). The obtained structures were optimized by means of density functional theory (DFT) algorithm using B3LYP/6-31G(d) basis set.
The developed three-dimensional structures were submitted to descriptor calculations using Dragon software (Dragon v. 7.0.6, Kode srl, Pisa, Italy). Only a selected block of descriptors were calculated: constitutional descriptors, ring descriptors, functional group counts, atom-centered fragments, CASTS2D descriptors, 2D atom pairs, charge descriptors, and molecular properties, in order to enable easier interpretation of the obtained models based on chemical structure of the proposed benzenesulfonamides.
Activity dataset was evaluated before QSAR analysis in terms of the presence of possible outliers. OPLS calculations, as well as a validation study, were performed using SIMCA (SIMCA v. 13.0.3.0, Umeå, Sweden). Descriptors were used as independent variables, and IC 50 was used as a dependent variable.
All compounds were tested for their in vitro cytotoxic activity against HCT-116, HeLa, and MCF-7 cell lines. It has been found that compounds bearing a quinolin-8y-l moiety showed the best anticancer potential. The most prominent compound 28 exhibited significant cytotoxic activity with IC 50 values of 3, 5, and 7 µM for HCT-116, MCF-7, and HeLa, respectively. Due to high cytotoxic activity, compound 28 was investigated for apoptosis-inducing activity in cancer cells. Appearance of an increased population of early and late apoptotic cells, induction of caspase activity, and cell cycle arrest at G2/M phase confirmed apoptosis of HCT-116, HeLa, and MCF-7 cells in the presence of compound 28.
The applied QSAR approach based on OPLS and corresponding VIP values allowed one to recognize chemical sub-structures related to increased or decreased biological activity. Especially, more rigid Ar/Het substituents are beneficial for activity. We also observed decreased activity related with the presence of oxygen-containing substituents. Those indications can be used for a rational plan of further synthesis.