Green Production of Indolylquinones, Derivatives of Perezone, and Related Molecules, Promising Antineoplastic Compounds

1Departamento de Ciencias Quı́micas, Facultad de Estudios Superiores Cuautitlán, Universidad Nacional Autónoma de México, 54740 Cuautitlán Izcalli, MEX, Mexico 2Facultad de Ciencias Quı́micas, Posgrado en Ciencias en Ingenieŕıa Quı́mica, Universidad Autónoma de San Luis Potośı, 78210 San Luis Potośı, SLP, Mexico 3Escuela Nacional de Ciencias Biológicas, Instituto Politécnico Nacional, Casco de Santo Tomas, 11340 Ciudad de México, CDMX, Mexico 4Laboratorio de Espectrometŕıa de Masas, Centro de Quı́mica, ICUAP, Benemérita Universidad Autónoma de Puebla, 72570 Puebla, PUE, Mexico 5Laboratorio de Biotecnologı́a, Instituto de Ciencias Biomédicas, Universidad Autónoma de Ciudad Juárez, Anillo Envolvente del Pronaf S/N, 32315 Ciudad Juárez, CHIH, Mexico


Introduction
Green Chemistry, with a protocol of a cohesive set of twelve principles [1], is an actual discipline that encourages working at the molecular level, in order to achieve sustainability; since the early 1990s, this paradigm has acquired its up-to-date position as a scientific discipline, preventing pollution via the implementation of reactions without damaging conditions to the environment [2].Under such a protocol, an ideal synthesis would be that by which a target molecule is produced quantitatively in one step from available and inexpensive starting compounds in an ecologically sustainable process; hence, the request of novel sustainable processes is among the major challenges in organic synthesis.In this regard, researchers at the vanguard of innovation must have the knowledge to design chemicals processes with little or no risk to the environment.However, no activity can be completely risk-free or waste-free or have lower harmful emissions.Consequently, the "Twelve Principles of Green Chemistry" should be viewed and applied as a scientific reflection [3].
In recent years, cancer prevention by means of natural products has received considerable attention; consequently, the chemistry of secondary metabolites is a powerful source for novel drug candidates.It is also convenient to underline the notion that the design and implementation of synthetic processes moving toward the Green Chemistry Protocol [2,25] are one of the major challenges in modern organic synthesis.
The indolylquinone moiety is present in many natural products with interesting biological activities [26,27]; consequently, the promising pharmaceutical use of the indolylquinones has attracted attention for the synthesis of new molecules of this class, mainly under a green approach.In this sense, the most common mode for the preparation of 3-indolylquinones is the direct coupling of indole with several quinones, in the presence of a catalyst (HCl, H 2 SO 4 , AcOH, Bi(OTf) 3 , I 2 , or InBr 3 ) in the presence of solvent, using the mantle heating procedure in order to activate the reactions [28].
As part of our ongoing research program, we are interested in the implementation of green synthetic strategies, in order to synergize or modify the pharmacological activities of single known compounds throughout the construction of novel hybrid molecules, mainly using nonconventional activating sources such as microwave (MW) and near-infrared (NIR) irradiation, ultrasound (US), and high speed ball milling (HSBM) in the absence of solvent or using innocuous solvents [29][30][31][32][33][34] and when required employing Tonsil Actisil FF (TAFF), commercial bentonitic clay, as a green catalyst [35].
Taking into account the aforementioned commentaries, the goals of this work are, in a first stage, to offer a green contribution for the production of four pharmacologic candidates, the indolyl derivatives of perezone (1), isoperezone (2), menadione (3), and plumbagin (4), by comparison of typical mantle heating conditions (MH) with uncommon activating modes, MW and NIR irradiation, in addition to the mechanical procedure of HSBM, all them in the absence of solvent and using TAFF as a catalyst.Additionally, we make an evaluation of the cytotoxic effect of the title molecules, in breast cancer cells as complement to theoretical-computation characterization of them.And finally, novel and convenient transformation of perezone into isoperezone, using for the first time HSBM, is also recommended.
1 H and 13 C NMR spectra were performed using a Varian Mercury-300 spectrometer at 300 MHz and 75 MHz for hydrogen and carbon, respectively, employing CDCl 3 as a solvent and TMS as an internal reference.The multiplicities are reported as singlet (s), broad singlet (bs), doublet (d), triplet (t), and multiplet (m).The EIMS and HRMS data were determined using JEOL JMS-700 MStation mass spectrometer.The elemental composition was calculated within a mass range of ±10 ppm from the accurate measured mass.The melting points were determined in a Fisher-Johns apparatus and are uncorrected.The microwave-assisted production of the target compounds was performed using CEM Focused Microwave6 Synthesis System.The near-infrared irradiation was generated using a commercial device "Flavor-Wave5" (1300 W/110 V/120 V-60 Hz|220 V/240 V-60 Hz) [45].The high speed ball milling was generated using Ball Mill PM 100 Retch, with 6 carbon steel balls (weight: 23.85 g, 0.39  diameter).The temperature was determined, for NIR, employing infrared thermometer (Infrared + Type K Thermometer, Extech Instruments, Sigma-Aldrich 2509388-1 EA); herein, the laser pointer was directed to the reaction center.
The human MDA-MB-231 breast cancer and dermal fibroblast normal cells lines were obtained from ATTC; NaH 2 PO 4 and Na 2 HPO 4 for phosphate buffer solution (PBS), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT), and dimethyl sulfoxide (DMSO) were obtained from Sigma-Aldrich without further purification.The cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin G, and 100 g/L streptomycin in humidified atmosphere with 5% CO 2 at 37 ∘ C. The culture medium was replaced every two days and the cells were used at passages between two and five.Statistical analysis of IC 50 was executed by PRISMA statistical program.
The theoretical calculations were performed employing GAUSSIAN09 program [46].

Typical Experiments Using Nonconventional Activating
Sources.A mixture of perezone 1 (120 mg, 0.4839 mmol), isoperezone 2 (120 mg, 0.4839 mmol), menadione 3 (90 mg, 0.5233 mmol) or plumbagin 4 (140 mg, 0.5000 mmol), indole 5 (60 mg, 0.5128 mmol), and 50 mg of TAFF as catalyst was placed in a micro scale glass vessel.The mixtures were treated using and comparing different activation modes: near-infrared irradiation during a period of 10 min at 121 ∘ C; microwave irradiation during a period of 10 min at 100 ∘ C and 250 W power, and high speed ball milling during a period of 60 min with 500 rpm and 27% power; all reactions were carried out in open vessel.The reactions were monitored by TLC using silica gel/n-hexane/ethyl acetate (80 : 20).The corresponding products were purified by column chromatography using the same mobile phase as TLC.(6).Purple solid; mp 65-67 ∘ C.  (7) (8)

The 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl Tetrazolium Bromide Assay (MTT Assay
).This test was performed with MDA-MB231 breast cancer cell line according to Mosmann [47].Cells were incubated at different indolylquinone concentrations.To derive viability curves of 6-9 treated cells, the cells were seeded onto 96-well culture plate at density of 5000 cells/well and grown in DMEM supplemented with 10% FBS.Cells viability was measured using MTT reagent dissolved in PBS (0.5 mg/mL).On the day of measurement, the medium was carefully replaced by fresh DMEM + 10% FBS with diluted MTT (1 : 10, 10% MTT) and incubated for 1 h at 37 ∘ C in a CO 2 incubator to allow the transformation of MTT dye to formazan salt.After removing incubation medium, formazan crystals were dissolved in 100 L solution of DMSO.It is also important to highlight the notion that MTT reduction was quantified by measuring the light absorbance at 570 nm using the Benchmark Plus absorbance microplate reader (Bio-Rad).MTT test was repeated nine times.Percentage of proliferation (% viability) was calculated according to  where  is the absorbance value of the control group and  is the absorbance value of the cells treated with indolylderivatives.

IC 50
Determination.IC 50 was extrapolated from the dose-response graph (Figure 1).The drug concentration that reduced the viability of cells by 50% (IC 50 ) was determined by plotting triplicate data points over a concentration range and calculating values using regression analysis.

Results and Discussion
3.1.Synthesis.The reactions of 1-4 with indole (5), under solventless conditions and with TAFF as a catalyst, were performed using the following three nonconventional modes of activation: microwave and near-infrared irradiation and high speed ball milling, affording the respective indole derivatives 6-9 (Scheme 1); in addition, the target molecules were comparatively obtained using mantle heating conditions; the corresponding results are summarized in Table 1.In general, the pathway must occur by classical Michael addition by the indol to the corresponding quinonic substrates, assisted by an interaction of the oxygen of a carbonylic group with the Lewis acidic sites of TAFF, followed by in situ oxidation, producing 6-9 in moderated yields at low times.It is appropriate to note that 6, 7, and 9 are novel molecules, and according to our knowledge 8 has been previously produced by MH and US in the presence of solvent with the employment of a noninnocuous catalyst [28,55,56].Moreover, literature research shows that this seems to be the first report for the green production of indolylquinones using MW, NIR, and HSBM as the activating modes without a solvent and in the presence of a natural ecofriendly catalyst.
A complementary and interesting result was the transformation of perezone into isoperezone, using for the first time HSBM (Scheme 2), in comparative yields and reaction times, with values previously reported [9,12].For the structural attribution of 2 and 6-9, in a first instance, isoperezone was spectroscopically correlated with literature data [9] as well as its melting point (108-109 ∘ C).Considering the notion that 6, 7, and 9 are novel products and that they are structurally related to 8, appropriate spectroscopic characterizations of the target molecules were performed.Thus, in a first place, the corresponding accurate molecular ions were acquired by HRMS-EI to achieve the elemental compositions; this was obtained in agreement with the expected products.
In regard to the 1 H NMR, it is worth noting the disappearance of the corresponding signals assigned to H-3 of indole and reagents 2-4, in addition to the signal assigned to H-6 of perezone; moreover, the signal due to the proton linked to the nitrogen of the indole,  7.81 ppm, was displaced, for 6-9 to lower fields 8.4-8.9 ppm, unequivocally due to deprotection generated by the high degree of conjugation between the indolic-quinonic moieties of the products.In this sense, the obtained 13 C NMR data complemented the structural characterization of molecules 6-9, and the signals of C-3 and C-2 of indole at 102.22 ppm and 111.10 ppm were, respectively, displaced to lower fields, 107.9-108.9 and 122.3-128.2ppm, respectively.This change in the chemical shift was also attributed to deprotection by the quinonic system.In general, the 1 H and 13 C NMR experimental data correlated appropriately with those values predicted by computational models.

Cytotoxic Activity.
Indolylquinones cytotoxicities were evaluated in human MDA-MB-231 breast cancer cells using the MTT assay, employing as control normal fibroblast cells.The studied compounds were assayed in concentrations of 0 to 100 g/mL for 6, 0 to 150 g/mL for 7, 0 to 110 for 8, and finally 0 to 130 g/mL for 9.According to Figure 1, the decrease of viable cells starts at 24 hours of incubation of the studied compounds; moreover molecule 7 began to decrease at a lower concenttration (15 mg/mL); on the other hand, for the rest of the molecules, the decrease of the viability began at concentrations higher than 20 g/mL.The same behavior was exhibited at 48 h with cancer cell viability in the range from 10 to 20% when the highest concentrations of 100, 150, 110, and 130 g/mL to 6, 7, 8, and 9, respectively, were used.The results after 72 h showed in 7 and 9 similar trends to those at 48 h; however, 6 and 8 increased the cellular viability in the highest concentrations.
The IC 50 values were obtained through extrapolation from the dose-response graphs (Figure 1).The results of the viability in human breast cancer cells against the four molecules tested recognize 7 as the most active molecule with IC 50 of 25 g/mL.In addition, for 6, 8, and 9, the values of IC 50 were 40.6, 41.6, and 32.3 g/mL, respectively.These concentrations were tested against healthy dermal fibroblast.With this result, it is important to point out that only 9 developed a cytotoxic effect with 15% mortality; on the other hand, 6-8, instead of promoting cells death, caused fibroblast proliferation.The effect in healthy cells is very interesting and is currently under study by our research group, as the cytotoxic pathway presented for these compounds.

Computational Study.
The molecules obtained were also studied through computational chemistry methods.In the optimization of the target molecules, a conformer distribution and the selection of the most stable conformer were achieved.Additionally, the selected geometries were appropriately optimized by DFT employing B3LYP functional and the 6-311++G(d,p) basis set.The structures obtained in phase gas are depicted in Figure 2.
Molecular Geometries.The calculated geometrical parameters of 6-9 are displayed in Tables 2 and 3. Since the target molecules have no experimental X-rays partners and considering the important motif in the biological activities, the experimental data of the starter substrates and similar indol derivative present in the literature [40][41][42] was compared with the predicted value for the target molecules.
Table 2 shows the bond length of 6 and 7 by similar structure in both molecules.In the case of 6 compared with perezone quinonic ring, the predicted values are very similar to 1.However, the C1-C6 bond in 6 is longer in comparison with the perezone.Additionally, the hydrogen bond present between the O3⋅ ⋅ ⋅ H hydroxyl group and the O4 carbonyl group decreases the distance, signaling a more stronger bond (intramolecular hydrogen bond); the O3⋅ ⋅ ⋅ O4 and H3⋅ ⋅ ⋅ O4 distances are 2.583 and 1.965 Å beside O3-H3⋅ ⋅ ⋅ O4 angle of 118.839 ∘ ; these data agree with those of Soriano-García et al. [40].Considering the indolyl moiety, the values were compared with the experimental values obtained for the Xray diffraction of 2-ethyl-5-(3-indolyl)oxazole [41].
The contrast of the indolyl fragment of 6 with the reference shows very similar values in both molecules; however, the predicted bonds are longer in 6.
The predicted values of 7 compared with the isoperezone show a similar trend in the quinonic ring; however, the bonds C3-C4 and C1-C6 are longer than in the experimental case.In the same sense, the intramolecular hydrogen bond is shorter, indicating the increase of the strength in this bond.
Regarding the indolyl moiety in 7, the estimated values are in good agreement with the experimental values considering the structural difference between 7 and the 2-ethyl-5-(3-indolyl)oxazole.
Table 3 summarizes some theoretical values of bond length for 8 and 9 contrasted with the X-rays values of 3 and 4 [43,44] and the indolyl structure with 5-(3-indolyl) oxazole [41].The calculated values for the bond C3-C4 of 8 and 9 are larger and the hydrogen bond of 9 is shorter following the same trends as the previous compounds ( 3 and 4).The rest of  the theoretical values are coherent and near the experimental values, appropriate with the theory level used.The enlargement of the bonds C1-C6 (for 6 and 7) and C3-C4 (for 7, 8, and 9) in the quinonic rings and shortening in the intramolecular hydrogen bond (6, 7, and 9) can be explained by the resonance effect promoted by the addition of the indolic structure.The electrons of indole moiety can produce greater localization of electrons in the carbonyl group oxygen (C4=O4 in 6 and C1=O1 in 7, 8, and 9) in contrast to 1-4 in which the double bonds of the quinone ring can be delocalized to both carbonyl groups.The consequence of this effect in 6-9 is the major negative charge in the oxygen atom in the previous carbonyl group signaled and therefore greater attractive force to hydrogen and minor electronic delocalization to the bonds C1-C6 and C3-C4, provoking enlargement of these bonds.
NMR Parameters.The experimental values corresponding to the chemical shifts of 1 H and 13 C were correlated with the calculated values by DFT at the B3LYP and using the 6-311++G(d,p) basis set and the GIAO method.
The calculated GIAO 1 H NMR chemical shifts were plotted versus the experimental data obtained for the indolylquinones (Figure 3(a)).
Linear regression analysis of the data set of 1 H NMR shifts provided the following results and presented a regression coefficient of 0.9806 and standard deviation of 0.3930 ppm; the equation to describe the fit is   = 0.9875 ex + 0.1661 ppm, where   is a chemical shift predicted in the base of the experimental values  ex , with the slope and intercept having a standard deviation of 0.0196 and 0.1261, respectively.
The same analysis was made for the theoretical and experimental chemical shifts values obtained for 13 C NMR; the corresponding plot is shown in Figure 3(b); in this regard, the linear regression analysis for 13 C chemical shifts yielded a regression coefficient of 0.9955 and standard deviation of 3.35 ppm; the equation to describe the fit is   = 1.0297 ex + 3.138 ppm, where the slope and intercept showed a standard deviation of 0.0076 and 0.937, respectively.
In other words, the values of the regression coefficients in both analyses reflected a good description of the experimental chemical shifts of 1 H and 13 C NMR by the selected method, theory level, and basis set.

Conclusions
The methodology employed proceeded with a good green approach [57]; it is in agreement with the following Green Chemistry Principles: prevention (principle 1), atom economy (principle 2), less hazardous chemical synthesis (principle 3), safer solvents and auxiliaries (principle 5), design for energy efficiency (principle 6), catalysis (principle 9), and inherently safer chemistry for accident prevention (principle 12).In addition, it is convenient to note that the obtained molecules have a cytotoxic effect in breast cancer cells without apparently toxic effect in healthy cells, and the most cytotoxic molecule (7) corresponds to the minor gap energy (theoretical calculations); and it is also important to mention that the experimental activity will be explained on the base of theoretical studies under the course by our research group.
Finally, taking into account the fact that three of the molecules are new, appropriate spectroscopic characterizations were performed by means of 1 H NMR, 13 C NMR, EIMS, and HRMS, and it is important to highlight that the experimental NMR spectroscopic data of the target compounds showed good agreement with theoretical characterization.

Figure 3 :
Figure 3: Linear regression between experimental and calculated B3LYP/6-311++G(d,p): (a) 1 H NMR chemical shifts and (b) 13 C NMR chemical shifts for the indolylquinones obtained in this work.

Table 1 :
Production of indolylquinones derivatives, using different activating modes.