Design, Synthesis, and Antiproliferative Activity of Benzopyran-4-One-Isoxazole Hybrid Compounds

The biological significance of benzopyran-4-ones as cytotoxic agents against multi-drug resistant cancer cell lines and isoxazoles as anti-inflammatory agents in cellular assays prompted us to design and synthesize their hybrid compounds and explore their antiproliferative activity against a panel of six cancer cell lines and two normal cell lines. Compounds 5a–d displayed significant antiproliferative activities against all the cancer cell lines tested, and IC50 values were in the range of 5.2–22.2 μM against MDA-MB-231 cancer cells, while they were minimally cytotoxic to the HEK-293 and LLC-PK1 normal cell lines. The IC50 values of 5a–d against normal HEK-293 cells were in the range of 102.4–293.2 μM. Compound 5a was screened for kinase inhibitory activity, proteolytic human serum stability, and apoptotic activity. The compound was found inactive towards different kinases, while it completely degraded after 2 h of incubation with human serum. At 5 μM concentration, it induced apoptosis in MDA-MB-231 by 50.8%. Overall, these findings suggest that new benzopyran-4-one-isoxazole hybrid compounds, particularly 5a–d, are selective anticancer agents, potentially safe for human cells, and could be synthesized at low cost. Additionally, Compound 5a exhibits potential anticancer activity mediated via inhibition of cancer cell proliferation and induction of apoptosis.


Introduction
The primary factors influencing cancer treatment are the patient's cancer type and stage. There are several methods of treatment, including surgery, chemotherapy, radiation, hormone therapy, immunotherapy, and targeted therapy [1,2]. Eliminating cancer cells, limiting their growth, and spread, and ultimately curing the illness or lengthening the patient's life are the goals of therapy [3]. Despite significant progress in cancer therapy, chemotherapy, which has many inherent drawbacks, remains a primary choice for treatment. To overcome the shortcomings of chemotherapy, including drug resistance, side effects, low effectiveness, lack of personalization, and high drug costs, novel compounds are still needed to enhance efficacy, safety, and personalization while being affordable [4][5][6]. In terms of effectiveness and safety, targeted therapeutic "small molecule" drugs have exceeded traditional chemotherapeutic drugs, and are now often employed to treat cancer [4].
Benzopyran-4-ones (also known as chromones) have been referred to as a "Privileged Scaffold in Drug Discovery" and have been extensively reviewed in the last few years [7][8][9][10][11]. The benzopyran-4-one skeleton has a variety of pharmacological properties,
A total of 14 benzopyran-4-one-isoxazole conjugates were synthesized, out of which 13 are new, and one is known (5c), while out of 12 benzopyran-4-one precursors, 11 are reported in the literature. A few of the proposed hybrid compounds (*6c, 7b, 8b-d, and 9b) were not obtained via the synthetic protocols followed due to the changes in the reactivity of the intermediates (2a-d and 4a-d) on -OMe substitution at different positions on the benzopyran-4-one nucleus. Compounds 5a-d, 6a, 6b, 6d, 7a, 7c, 7d, 8a, 9a, 9c, and 9d were synthesized and purified with the laboratories techniques and were further fully characterized on the basis of their 1 H NMR, 13 C NMR, 13 C-DEPT NMR, IR, and Q-TOF mass spectral data (cf. Supplementary Materials (SM), page no. 2-65). The structures of known compounds were confirmed via a comparison of their spectral data with those reported in the literature.
A total of 14 benzopyran-4-one-isoxazole conjugates were synthesized, out of which 13 are new, and one is known (5c), while out of 12 benzopyran-4-one precursors, 11 are reported in the literature. A few of the proposed hybrid compounds (*6c, 7b, 8b-d, and 9b) were not obtained via the synthetic protocols followed due to the changes in the reactivity of the intermediates (2a-d and 4a-d) on -OMe substitution at different positions on the benzopyran-4-one nucleus. Compounds 5a-d, 6a, 6b, 6d, 7a, 7c, 7d, 8a, 9a, 9c, and 9d were synthesized and purified with the laboratories techniques and were further fully characterized on the basis of their 1 H NMR, 13  All the 14 hybrid compounds synthesized via Scheme 1, along with the intermediates and starting compounds, were tested in vitro for their cytotoxicity against a panel of 6 human cancer cell lines. This panel involved human leukemia carcinoma (CCRF-CEM), human ovarian adenocarcinoma (SKOV-3), human breast tumor (MDA-MB-231), human prostate cancer (PC-3), androgen-independent human prostate cancer (DU-145), and human renal carcinoma (iSLK) cell lines. The compounds were tested at a concentration of 25 µM for incubation periods of 24 and 72 h. Standard anticancer drug doxorubicin at a concentration of 5 µM and DMSO were used as controls in all the cell assays. The antiproliferative profile of benzopyran-4-one-isoxazole conjugates is shown in Figure 2 (cf. Supplementary Materials for the data obtained for intermediates and starting materials Figure S28.1-S28.6, page no. 71-76).
Among all the 14 compounds tested, a series of benzopyran-4-one-isoxazole esters, i.e., 5a, 5b, 5c, and 5d, exhibited selective antiproliferative activity against the panel of all six cancer cell lines tested (Figure 2a-f). Although among them, compound 5a exhibited exceptional selectivity and cytotoxicity in all the cell lines tested. However, these conjugates, corresponding intermediates, and starting materials were almost inactive in the six tested cancer cell lines (cf. Supplementary Materials, Figure S28.1-S28.6, page no. 71-76). Table S2 (cf. Supplementary Materials, page no. 79) highlights the percentage inhibition of the cancer cell lines on treatment with doxorubicin (at a concentration of 5 µM) and conjugates 5a-d (at a concentration of 25 µM) after 72 h of incubation.
The structure-activity relationship of the benzopyran-4-one-isoxazole hybrid compounds was evaluated from the data in Table S2 (Supplementary Materials, page no. 79) and Figure 2, which clearly indicates that the conjugation of benzopyran-4-one with isoxazole via a particular ester linkage, i.e., esterification of 3,5-dimethylisoxazole-4-carboxylic acid with 3-(hydroxymethyl) substituted-4-oxo-4H-1-benzopyrane (5a-d) had significantly higher selectivity toward the cancer cell lines compared to the hybrid compounds obtained via other chemical linkers, such as acetal (6a, 6b, and 6d), amide (7a, 7c, and 7d), reverse ester (8a), and acrylic ester (9a, 9c, and 9d). Furthermore, the structure-activity relationship indicates that substitution of a methoxy group at 5/6/7-position of benzopyran-4-one moiety decreases the activity slightly (5b, 5c, and 5d), while no substitution on the aromatic ring (5a) leads to higher selectivity and cytotoxicity against the cancer cell lines tested. Compounds 5a-d led to high percentage inhibition (81-83%) of CCRF-CEM cancer cells, while a moderately high percentage of antiproliferative activity for the other five cell lines. Comparing the data for 5a with 5b-d, compound 5a had the highest percentage of inhibition (44-81%) among this series of compounds.
Among all the 14 compounds tested, a series of benzopyran-4-one-isoxazole esters, i.e., 5a, 5b, 5c, and 5d, exhibited selective antiproliferative activity against the panel of all six cancer cell lines tested (Figure 2a-f). Although among them, compound 5a exhibited exceptional selectivity and cytotoxicity in all the cell lines tested. However, these conjugates, corresponding intermediates, and starting materials were almost inactive in the six tested cancer cell lines (cf. Supplementary Materials, Figure S28.1-S28.6, page no. 71-76). Table S2 (  The structure-activity relationship of the benzopyran-4-one-isoxazole hybrid compounds was evaluated from the data in Table S2 (Supplementary Materials, page no. 79) and Figure 2, which clearly indicates that the conjugation of benzopyran-4-one with isoxazole via a particular ester linkage, i.e., esterification of 3,5-dimethylisoxazole-4-carboxylic acid with 3-(hydroxymethyl) substituted-4-oxo-4H-1-benzopyrane (5a-d) had significantly higher selectivity toward the cancer cell lines compared to the hybrid compounds

In Vitro Cytotoxicity against the Non-Cancerous Cell Lines
The compounds were further screened for their in vitro cytotoxicity against noncancerous normal cell lines, viz., human embryonic kidney (HEK-293) and normal pig kidney (LLCPK) cell lines. These cell lines were selected to determine whether the compounds had selectivity against cancerous cells in contrast to non-cancerous cells. The data obtained is shown in Figure 3a The compounds were further screened for their in vitro cytotoxicity against non-cancerous normal cell lines, viz., human embryonic kidney (HEK-293) and normal pig kidney (LLCPK) cell lines. These cell lines were selected to determine whether the compounds had selectivity against cancerous cells in contrast to non-cancerous cells. The data obtained is shown in Figure 3a-b (cf. Supplementary Materials for the data obtained for intermediates and starting materials Figure S28.7-S28.8, page no. 77-78).  Compounds 5a-d exhibited potent growth inhibitory activity against a panel of six cancer cell lines and weak to no inhibitory activity against two normal cell lines. These compounds were further analyzed for their IC50 values against five selected cell lines. The data is displayed in Table 1. The compounds displayed weak activity on embryonic kidney (HEK-293) cell lines (>100-200 µM). However, compounds 5a-d displayed potent growth inhibitory activity against all other four cell lines (3-51 µM), again confirming their selectivity towards cancerous cells in contrast to non-cancerous/normal cells. For example, compound 5a showed potent antiproliferative activity against all the cancer cell lines with IC50 in the range of 5.6-17.84 µM and IC50 of 293.2 µM against normal cell lines (HEK-293). Similarly, compound 5c was also potent against all the cancer cell lines with IC50 in the The inhibition percentages of the non-cancerous/normal cell lines on treatment with Dox (at a concentration of 5 µM) and conjugates 5a-d (at a concentration of 25 µM) are represented in Table S3 (cf. Supplementary Materials, page no. 80). The data obtained clearly indicates the selectivity of this series of conjugates (5a-d) towards cancer cell lines (Table S2,

Half-Maximal Inhibitory Concentration (IC 50 )
Compounds 5a-d exhibited potent growth inhibitory activity against a panel of six cancer cell lines and weak to no inhibitory activity against two normal cell lines. These compounds were further analyzed for their IC 50 values against five selected cell lines. The data is displayed in Table 1. The compounds displayed weak activity on embryonic kidney (HEK-293) cell lines (>100-200 µM). However, compounds 5a-d displayed potent growth inhibitory activity against all other four cell lines (3-51 µM), again confirming their selectivity towards cancerous cells in contrast to non-cancerous/normal cells. For example, compound 5a showed potent antiproliferative activity against all the cancer cell lines with IC 50 in the range of 5.6-17.84 µM and IC 50 of 293.2 µM against normal cell lines (HEK-293). Similarly, compound 5c was also potent against all the cancer cell lines with IC 50 in the range of 3.3-12.92 µM and IC 50 of 222.1 µM against normal cell lines (HEK-293). However, compounds 5b and 5d were a few folds less active against respective cancer cell lines (IC 50 = 14.77-51.15 µM for 5b and IC 50 5.2-16.1 µM for 5d) and had low antiproliferative activity against normal HEK-293 cell lines (IC 50 were in the range of 102.4 and 191.5 µM, respectively). Furthermore, compound 5c showed 3-12-fold potency for CCRF-CEM cell lines compared to 5a, 5b, and 5d, whereas compound 5a showed 2-4-fold potency for MDA-MB-231 cell lines compared to 5c and 5d. Compounds 5a, 5c, and 5d were similar in potency against PC3 and DU-145 compared to 5b, which showed threefold weaker activity against DU-145 cell lines. Overall, the safety index of compounds was more than 10-fold based on IC 50 against cancer and normal cell lines.

NCI-60 Cell Lines Test
Among all the compounds tested, the best antiproliferative results were obtained from compound 5a. To further investigate the potency of this compound against different cancer cell lines, it was screened in a single dose (10 µM) experiment on 60 different cancer cell lines by NCI, USA ( Figure 4). The data on screening of compound 5a against 60 cancer cell lines for leukemia, non-small lung, colon, CNS, melanoma, ovarian, renal, prostate, and breast cancer cell lines showed the potency of compound 5a at a dose of 10 µM against leukemia and breast cancer cell lines. The results reveal that compound 5a could serve as a feasible target ( Figure 4) for further investigation in the next round of studies.

Kinase Inhibition Assay
Substituted benzopyran analogs have demonstrated protein kinase inhibitory activity [36]. In our previous studies [37], 4-oxo-4H-1-benzopyran derivatives showed Src kinase inhibitory activity with the IC50 of 52-57 µM. LY294002 has been found to be a selective inhibitor for phosphatidylinositol 3-kinase (PI3K) inhibitor with a cellular IC50 of about 1.4 µM. Therefore, selected compound 5a was screened against several protein ki-

Kinase Inhibition Assay
Substituted benzopyran analogs have demonstrated protein kinase inhibitory activity [36]. In our previous studies [37], 4-oxo-4H-1-benzopyran derivatives showed Src kinase inhibitory activity with the IC 50 of 52-57 µM. LY294002 has been found to be a selective inhibitor for phosphatidylinositol 3-kinase (PI3K) inhibitor with a cellular IC 50 of about 1.4 µM. Therefore, selected compound 5a was screened against several protein kinases to explore the potential of these hybrid compounds as kinase inhibitors. The screening of compound 5a was performed on a panel of 9 kinases, viz., ABL1, c-Kit, c-Src, CDK2/cyclin A1, CSK, EGFR, mTOR/FRAP1, p38a/MAPK14, PKCa, and PI3K, at a dose of 20 µM. The respective positive controls were used for each kinase, and the kinase reaction was performed at 200 µM ATP concentration for all the kinases except for PI3K, which was performed at 10 µM ATP ( Table 2). The results indicated that compound 5a was found to be inactive towards all the kinases screened. It appeared that compound 5a containing a 3-position substitution with an isoxazole ring and a 4-position with a carbonyl group has a detrimental effect on kinase inhibition. Previously 3-substituted benzopyran compound with a short alkyl chain of urea (NH-CO-NH-Et/Me/iPr) was reported with Abl kinase inhibitory activity in the range of 4-27 µM [36]. While compound 5a, containing an ester linker and 2,4-methyl isoxazole moiety at the 3-substitution on the benzopyran ring, showed a detrimental effect due to a lack of hydrogen bond interaction with the key residues in the kinase enzyme binding pocket. Most active benzopyran analogs with kinase inhibitory activity have 2-position substitution, such as LY294002 [38]. Due to prominent antiproliferative activity, these compounds may have some other targets for anticancer activity.

Serum Stability
These compounds were further explored towards degradation by proteolytic enzymes towards degradation. Compound 5a was investigated for proteolytic stability in the presence of human serum. It was incubated with human serum at 37 • C, and aliquots of the reaction mixture were analyzed via HPLC at different time points up to 2 h. The concentration of Compound 5a was measured by integrating the area under the curve (AUC) and correspondingly calculated from the standard curve of 5a at different concentrations (cf. Supplementary Materials, page no. 81-82).
The HPLC chromatogram of serum incubated with compound 5a for 80 min was illustrated in Figure 5. Compound 5a was completely degraded after 2 h of incubation with human serum. Aliquots from the reaction solution were taken at intervals of 0, 2, 5, 10, 20, 40, 80, and 120 min after incubation with human serum. . RP-HPLC chromatograms of conjugate 5a after incubation with human serum. The Conjugate was incubated with human serum for different time intervals of 0, 2, 5, 10, 20, 40, and 80 min before HPLC analysis. Conjugate 5a appears at 38 min, which degrades with time. Degradation products appear at around 34 min. Figure 6 shows the plots of human serum stability of 5a vs. % remaining of compound 5a, and the logarithm of the concentration of 5a vs. time in minutes (cf. Supplementary Materials, page no. 85) for the calculation of the half-life of 5a in human serum. The results indicate that Compound 5a has a half-life of nearly 13.6 min (cf. Supplementary Materials, page no. 84). The short half-life is an indication that this series of compounds stay for a short period in physiological conditions, which means that there is a further need for modification for the next generation of benzopyran-4-one-isoxazole hybrid compounds to increase their stability in human serum.   Figure 6 shows the plots of human serum stability of 5a vs. % remaining of compound 5a, and the logarithm of the concentration of 5a vs. time in minutes (cf. Supplementary Materials, page no. 85) for the calculation of the half-life of 5a in human serum. The results indicate that Compound 5a has a half-life of nearly 13.6 min (cf. Supplementary Materials, page no. 84). The short half-life is an indication that this series of compounds stay for a short period in physiological conditions, which means that there is a further need for modification for the next generation of benzopyran-4-one-isoxazole hybrid compounds to increase their stability in human serum.

Apoptosis Detection by Flow Cytometry Analysis
In accordance with the MTS assay, compound 5a had promising antiproliferative action. Therefore, we attempted to study the mechanism triggering cell death. It is established that inducing apoptosis is one of the most important methods to lead to cell death via chemotherapy; hence, we performed phosphatidylserine cytosine analysis to check the induction of apoptosis by compound 5a on MDA-MB-231 cells. MDA-MB-231 cells were treated with 5 µM of compound 5a and Dox (5 µM) as a positive control and incubated for 72 h, then double-stained carefully by membrane-linked protein Annexin-V fluorescein isothiocyanate (FITC) and propidium iodide (PI) for flow cytometry analysis. Normally, in the early stages of apoptosis, phosphatidylserine flips from the inner cell membrane to the surface of the cell membrane. The exposed phosphatidylserine could be recognized by an extracellular calcium-dependent phospholipid-binding protein Annexin-V; hence, FITC-labeled Annexin-V could bind specifically to phosphatidylserine and be used to detect apoptosis cells by flow cytometry. PI is a nucleic acid dye that can only cross the membrane of advanced apoptosis and necrotic cells and redden the nucleus. Consequently, when the treated cells are double stained by the two fluorescent dyes, early and late apoptosis and necrotic cells can be clearly distinguished. As shown in Figure 7, untreated control MDA-MB-231 cells had high observed viability of 95.41% ( Figure 7A). After the treatment with compound 5a (5 µM), the percentage of apoptosis cells increased from 3.27% (in untreated control MDA-MB-231 cells) to 47.52% for early apoptosis and from 0.03% (in untreated control MDA-MB-231 cells) to 3.28% for late apoptosis ( Figure 7B); in comparison, doxorubicin (5 µM) induced 61.15% for early apoptosis and 28.29% for late apoptosis to MDA-MB-231 cells ( Figure 7C). These findings illustrated that compound 5a mainly leads to the early apoptosis of MDA-MB-231 cells. The comparison of MDA-MB-231 cell distribution after 5a and doxorubicin treatment 72 h was shown in Figure 8. In accordance with the MTS assay, compound 5a had promising antiproliferative action. Therefore, we attempted to study the mechanism triggering cell death. It is established that inducing apoptosis is one of the most important methods to lead to cell death via chemotherapy; hence, we performed phosphatidylserine cytosine analysis to check the induction of apoptosis by compound 5a on MDA-MB-231 cells. MDA-MB-231 cells were treated with 5 µM of compound 5a and Dox (5 µM) as a positive control and incubated for 72 h, then double-stained carefully by membrane-linked protein Annexin-V fluorescein isothiocyanate (FITC) and propidium iodide (PI) for flow cytometry analysis. Normally, in the early stages of apoptosis, phosphatidylserine flips from the inner cell membrane to the surface of the cell membrane. The exposed phosphatidylserine could be recognized by an extracellular calcium-dependent phospholipid-binding protein Annexin-V; hence, FITC-labeled Annexin-V could bind specifically to phosphatidylserine and be used to detect apoptosis cells by flow cytometry. PI is a nucleic acid dye that can only cross the membrane of advanced apoptosis and necrotic cells and redden the nucleus. Consequently, when the treated cells are double stained by the two fluorescent dyes, early and late apoptosis and necrotic cells can be clearly distinguished. As shown in Figure 7

Instrumentation
Nuclear magnetic resonance spectroscopy: The 1 H NMR and 13 C NMR spectra were recorded on Bruker Acend 400 spectrometer (Billerica, MA, USA), operating at 400. 15 MHz for 1 H and 100.62 MHz for 13 C. The spectra were calibrated using the solvent peaks. The chemical shift values were expressed in δ (ppm) scale relative to tetramethylsilane

Instrumentation
Nuclear magnetic resonance spectroscopy: The 1 H NMR and 13 C NMR spectra were recorded on Bruker Acend 400 spectrometer (Billerica, MA, USA), operating at 400.15 MHz for 1 H and 100.62 MHz for 13 C. The spectra were calibrated using the solvent peaks. The chemical shift values were expressed in δ (ppm) scale relative to tetramethylsilane (TMS) as an internal reference, and the coupling constant values (J) were given in Hz. Assignments were also made from DEPT (distortionless enhancement by polarization transfer) (underlined values).
Infrared spectroscopy: Infrared spectra were recorded on Bruker Alpha FT-IR spectrophotometer (Billerica, MA, USA); the frequency values were recorded in cm −1 scale.
Mass spectroscopy: The mass spectra were recorded on Bruker Q-TOF LC/MS mass spectrophotometer (Billerica, MA, USA); the data were reported as m/z (% of relative intensity of the most important fragments).
Analytical HPLC: The purity of the final products (>97% purity) was verified by highperformance liquid chromatography (HPLC) equipped with a UV detector. Chromatograms were obtained in an HPLC/DAD system, Jasco instrument (pumps model 880-PU and solvent mixing model 880-30, Tokyo, Japan), equipped with a commercially prepacked Nucleosil RP-18 analytical column (250 mm Å~4.6 mm, 5 µm, Macherey-Nagel, Duren, Germany) and UV detection (Jasco model 875-UV) at the maximum wavelength of 254 nm. The mobile phase consisted of acetonitrile/water (gradient mode, room temperature) at a flow rate of 1 mL/min. The chromatographic data were processed with a Compaq computer fitted with CSW 1.7 software (DataApex, Podohradska, Czech Republic).

In Vitro Screening for Compound 5a
Compound 5a was screened in a single dose (10 µM) against NCI-60 cell panel by Developmental Therapeutics Program, National Cancer Institute, Division of Cancer Treatment and Diagnosis, Bethesda, MD, USA, as reported. The cell lines in the NCI-60 panel include leukemia, non-small lung, colon, CNS, melanoma, ovarian, renal, prostate, and breast cancer cell lines. The data was reported for one dose as the mean of the percent growth of treated cells related to the no-drug control.

Serum Stability
HPLC analysis was carried out to evaluate the proteolytic stability of selected conjugate 5a in the presence of human serum. First, different concentrations of Compound 5a were analyzed using the analytical HPLC, and a standard curve was plotted for the different concentrations of Compound 5a versus the area under the curve. Second, the stability experiment was performed by the following procedure. Biological conditions were mimicked by suspending human serum (250 µL) in RPMI media (700 µL) in Eppendorf tube (1.5 mL). The mixture was equilibrated to 37 ± 1 • C for 15 min, and then 50 µL of conjugate stock solution (1 mM in DMSO) was added to it. As Control 1, one vial was treated with 50 µL of DMSO without Compound 5a, and as Control 2, another vial was treated with Compound 5a but without any human serum (instead used 250 µL of deionized water). The initial time was recorded, and 100 µL aliquots of the reaction mixture were removed and added to methanol (200 µL) for precipitation of serum proteins in human serum. The solution turned cloudy, which was cooled down to 4 • C for 15 min. It was then spun at 500× g for 15 min to pellet the serum proteins. A 50 µL of the supernatant was injected into an RP-HPLC Vydac C18 column using an autoinjector. A linear gradient from 10% to 100% acetonitrile/water in 65 min with a flow rate of 1.0 mL/min was used, and the absorbance of the eluting peaks was detected at 254 nm. An aliquot of reaction solution (100 µL) was removed after regular time intervals of 2, 5, 10, 20, 40, and 80 min, and the procedure was repeated. The area under the curve was used to calculate the percentage of remaining Compound 5a at a given time. The relative percentage of Compound 5a was then plotted on a graph to obtain the stability in the human serum at the particular time intervals of incubation. Then, using the standard curve and relative percentage curve, the half-life of Compound 5a was calculated.