Effective Synthesis and Biological Evaluation of Natural and Designed Bis(indolyl)methanes via Taurine-Catalyzed Green Approach.

An ecofriendly, inexpensive, and efficient route for synthesizing 3,3'-bis(indolyl)methanes (BIMs) and their derivatives was carried out by an electrophilic substitution reaction of indole with structurally divergent aldehydes and ketones using taurine and water as a green catalyst and solvent, respectively, under sonication conditions. Using water as the only solvent, the catalytic process demonstrated outstanding activity, productivity, and broad functional group tolerance, affording the required BIM natural products and derivatives in excellent yields (59-90%). Furthermore, in silico based structure activity analysis of the synthesized BIM derivatives divulges their potential ability to bind antineoplastic drug target and spindle motor protein kinesin Eg5. The precise binding mode of BIM derivatives with the ATPase motor domain of Eg5 is structurally reminiscent with previously reported allosteric inhibitor Arry520, which is under phase III clinical trials. Nevertheless, detailed analysis of the binding poses indicates that BIM derivatives bind the allosteric pocket of the Eg5 motor domain more robustly than Arry520; moreover, unlike Arry520, BIM binding is found to be resistant to drug-resistant mutations of Eg5. Accordingly, a structure-guided mechanism of Eg5 inhibition by synthesized BIM derivatives is proposed.

Introduction

3,3′-Bisindolylmethane (BIM) and its functionalized class of natural products were originally isolated from cruciferous plants and terrestrial and marine organisms and shown to contain a broad array of medicinal properties.[1−3] Key members of this class of natural products include arundine, vibrindole A, streptindole, arsindoline A, tris(1H-indol-3-yl)methane, and trisindoline (Figure ).[4−10] A unique structure of this bioactive intermediate contains two indole units which are responsible for exhibiting a range of essential biological activities such as anti-neurodegenerative, anti-inflammatory, anticancer, antibiotic, insecticidal, antimicrobial, antifungal, and antioxidant properties (Figure ).[7,11−20] Interestingly, some of these BIM derivatives are used as a sedatives in the treatment of AIDS, chronic fatigue, irritable bowel syndrome, and cancer, while bis(indolyl)arylmethane-bearing compounds are used as drugs for Parkinson’s disease, obesity, and bacterial and oncolytic viruses.[18,21−29] In addition, BIMs have also been found to have the potential to normalize abnormal cell growth associated with cervical dysplasia.[30]

Figure 1

Representative analogues of naturally occurring 3,3′-BIMs with relevant bioactivity.

Moreover, BIMs have grown in prominence in recent years, particularly after the discovery of their effective involvement in suppressing prostate cancer cells in humans.[31,32] Because of their impressive pharmaceutical properties coupled with their high demand in medicinal chemistry, drug discovery, and agrochemicals, BIMs have become a frequent target for synthetic organic chemists worldwide, and intense research activities toward their structural modification and biological evaluation have been pursued. As a result of this surge, numerous methods have been devised for the production of naturally occurring BIMs and their functionalized analogues (see Scheme ).[7,11,33,34] Interestingly, most methods utilize 3-indole aromatic alcohols to afford symmetrical and unsymmetrical 3,3′-BIMs using a condensation strategy with appropriate nucleophiles.[19,35] In addition, the most common reaction protocols used in the synthesis include expensive and excess amounts of catalysts such as Cu-catalysts,[36] TsOH, Ga(OTf)3, Zn(OTf)2, etc.[37−39] Several catalytic pathways were also developed to afford BIMs, which includes Bronsted or Lewis acid, methanol, iodine, electrolysis methods, metal salts, visible light, and enzymes.[40−42] Moreover, solid acidic catalysts such as Amberlyst, montmorillonite clay K-10, TiO2, ZrOCl2/SiO2, HClO4–SiO2, and P2O5/SiO2 have also been investigated so far.[43−46] Over the last decades, methods for fabricating symmetrical and unsymmetrical 2,3′- and 3,3′-BIMs have primarily focused on the indolylation of 3-indolylmethyl alcohol, sulfone, malononitrile, 2,4,6-trimethoxybenzene, etc. with indole derivatives.[47−51] Although the divulged methodologies are phenomenal, they have some downsides, such as the need for expensive and higher catalytic loading, harsh reaction conditions, time-consuming reaction procedures, use of hazardous organic solvents, inadequate yields, and a constrained substrate scope. Considering an excellent medicinal profile and necessity, BIMs always demand a scalable method and new functionalized analogues for further drug discovery and development. As a corollary, there is an undeniable potential for an efficient and environmentally friendly approach to synthesize BIMs.

Scheme 1

Synthesis of Symmetrical and Unsymmetrical BIMs

Herein, with an aim toward developing potential therapeutic antineoplastic leads, we set out to optimize an ecofriendly one-pot organic synthesis method to obtain naturally occurring potent BIMs along with their novel derivatives targeting crucial cancer drug leads. Therefore, in continuation of our unwavering interest in indole compounds,[52−54] we investigated the chemistry of condensation of indoles with aldehydes and ketones catalyzed by taurine (C2H7NO3S) using only water as a green solvent.[55] As part of our effort to overcome drawbacks from the previous report and formulate an efficient green approach, we describe here a combination technique of sonication and aqueous media at a moderate temperature. The described synthetic approach culminated into a simple, ecofriendly, one-pot methodology that produces 17 desired BIM derivatives including four natural products, arundine, vibrindole A, tris(1H-indol-3-yl)methane, and trisindoline, in high yields with no transitional isolation. Considering the myriad applications of BIM derivatives in several anticancer therapies, we conducted in silico structure activity analysis of the synthesized BIM derivatives against anticancer drug targets. In this context, mitotic kinesin Eg5 is an alluring target for antineoplastic drugs as it plays a pivotal role in the formation and stabilization of bipolar spindle architecture and faithful chromosomal segregation.[56−58] Kinesin Eg5 is a multidomain molecular machine that transforms ATP-driven mechanochemical force to glide unidirectionally on a microtubule track through a series of motor domain conformational changes and associated changes in microtubule affinities.[59−61] Intriguingly, the synthesized BIM derivatives were found to dock at one of the well-established allosteric inhibitor (monastrol) binding sites of the Eg5 motor domain which also houses two other distinct allosteric inhibitor-binding sites.[62−64] The monastrol binding allosteric site offers high target specificity but suffers from the emergence of recently reported drug-resistant mutations (D130A and L214A).[65] Hence, since the discovery of monastrol site targeting resistance, resistant therapeutic leads are in high demand to confront drug-resistant Eg5 mutations.[66] Interestingly, the synthesized BIM derivatives were found to be recalcitrant toward drug-resistant Eg5 mutations. The detailed structure activity analysis of BIM derivative(s)/Eg5 binding and a plausible molecular mechanism of inhibition are presented herein.

Result and Discussion

Being inspired by the green approach of BIMs synthesis methodology, and our recent report on the taurine-catalyzed multicomponent reaction for the synthesis of densely substituted dihydropyrano[2,3-c]pyrazoles,[66] we became interested in assessing the efficacy of taurine as a catalyst to boost the multicomponent reaction of probe indole and functionalized aldehyde. In our quest to find a best catalytic condition to access BIMs and their derivatives, the reaction was performed with p-methoxybenzaldehyde 1a and indole 2a as a model substrate under a different set of catalysts and conditions. Thus, a mixture of aldehyde 1a (1 mmol) and indole 2a (2 mmol) was treated under various solvent and catalytic conditions to afford 3,3′-((4-methoxyphenyl)methylene)bis(1H-indole) 3a, and the results are summarized in Table . Initially, to understand the role and effect of Lewis acid as a catalyst, reactions of aldehyde 1a and indole 2a were examined under different solvents and temperatures to afford 3a. As a result, the reaction was found to be very slow with a catalytic amount of Cu(OTf)2 in DCM at room temperature to 50 °C and afforded only 50% and 44% yield, respectively (see entries 1 and 2, Table ). Similar findings were observed with BF3·OEt2, where the reaction time was greatly decreased but the yield was reduced to 34%, (see entry 3, Table ). To our delight, the reaction with the newly implemented catalyst taurine proceeded smoothly in CH2Cl2 at room temperature to afford BIMs 3a selectively in 20 h with 52% yield (see entry 4, Table ). However, opposite results were observed when the reaction was performed in water medium leading to only a trace amount of compound (entry 5, Table ). Notably, an increase of the reaction temperature from 23 to 50 °C or increase of catalyst loading from 5 to 10 mol % did not furnish the required product in an isolable yield (see entries 6–8, Table ). Interestingly, the taurine-catalyzed combination technique of sonication and aqueous media delivered the required product in 60% yield (see entry 9, Table ). Remarkably, the highest yield of 69% was obtained when the reaction was performed at 50 °C using the taurine-catalyzed sonication reaction in water medium (see entry 10, Table ). In order to establish the conditions, the reaction was attempted with an excess catalyst but did the yield did not increase (see entry 11, Table ). It is worth mentioning that an attempt was made to validate the sonication conditions by changing the reaction temperature, catalyst loading, and time but keeping water as a green solvent did not improve the yield, favoring sonication as a necessary condition (see entries 5–8, Table ). Moreover, to verify the need of a catalyst, a trial was conducted without any catalyst, in which all of the components were sonicated in H2O at 50 °C for 24 h, which led to no product formation; hence, the requirement of the catalyst was confirmed for the reaction (see entry 12, Table ). Most of the catalytic pathways employed to reach BIMs derivative 3a yielded positive results, as shown in Table , but the best results were obtained with taurine-catalyzed water-mediated reaction at 50 °C that gave 69% yield (see entry 12, Table ). Further, to demonstrate the generality of developed method, the protocol was expanded to synthesize functionalized BIMs by treating the indole with a range of aldehydes, and the findings are reported in Scheme . Interestingly, the reactions of indole (2.0 equiv) with varying aldehydes (1.0 equiv) under the optimized conditions proceeded smoothly and resulted in the newly designed products 3b–3o in 59–87% yield (see Scheme ). The developed strategy was well-tolerated for both the aromatic aldehydes (1a–1i) and aliphatic aldehyde (1l–1n) to construct 17 designed bis(indolyl)methanes including four naturally occurring BIMs called arundine, vibrindole A, tris(1H-indol-3-yl)methane, and trisindole in excellent yields. Moreover, aromatic aldehydes 1a possessing an electron-donating group (−OMe) and (1c–1i) possessing electron-withdrawing groups (F, Br, CN, NO2) afforded the corresponding product in excellent yield (67–87%; Scheme ). Furthermore, the approach was tested for heterocyclic aldehydes 1j,k, and to our delight, both reactions operated well without producing any side products. Next, we aimed to establish a new BIM derivative 4a utilizing a substituted indole skeleton to further investigate the extent of the proposed catalytic system. As a result, biochemical precursor 2-methylindole 2b processed with furfural 1j at taurine produced the optimal sonication conditions.

Table 1

Selected Optimization Studies to Access BIMs and Analogues: Effect of Different Supportsa

entry	catalystb	solventc	temp (°C)	time (h)	yieldd (%)
1	Cu(OTf)2	CH2Cl2	23	16	50
2	Cu(OTf)2	CH2Cl2	50	10	44
3	BF3.OEt2	CH2Cl2	23	4	34
4	taurine	CH2Cl2	23	20	52
5	taurine	H2O	23	24	trace
6	taurine	H2O	50	24	trace
7e	taurine	H2O	23	24	trace
8e	taurine	H2O	50	24	trace
9f	taurine	H2O	23	2	60
10f	taurine	H2O	50	1	69
11e,f	taurine	H2O	50	1	54
12f	no catalyst	H2O	50	24	trace

Reaction conditions: aldehyde (1.0 mmol) and indole (2.0 mmol).

Catalyst: 5 mol %.

Solvent: 3–5 mL.

Isolated yields based on starting material recovered.

Catalyst: 10 mol %,

Sonication.

Scheme 2

Taurine-Catalyzed One-Pot Synthesis of BIM Derivatives

Intriguingly, the reaction proceeded very well and produced the required core in 90% yield in about 40 min (see Scheme ). We were indeed inspired by the outcomes and next turned our attention to the development of innovative spiro-analogues 3,3-di(indol-3-yl)indolin-2-ones 6a. Thus, we opted to replace the aldehyde moiety with the bioinspired group isatin 5a in view of obtaining a novel spiro-analogue of the BIM framework. The bioinspired natural product isatin belonging to plants and humans is well-known for its presence in many bioactive compounds and their spiro-congeners and is responsible for a broad array of biological activities such as antiviral, anti-HIV, antitumor, and antitubercular properties.[67−69] Interestingly, taurine-catalyzed water-mediated sonication reaction of indole 2a with isatin 5a under the optimized conditions furnished a well-designed fused spirooxindole product 3,3-di(indol-3-yl)indolin-2-one 6a with 80% yield (Scheme ). In comparison to other functionalized aldehydes, the ketone component of isatin shows a sluggish reactivity. The structure of all of the synthesized BIMs analogues 3a-o, 4a, and 6a were validated unambiguously from their spectroscopic investigation (1H NMR, 13C NMR, IR, HRMS and melting point) (see the Supporting Information for details). As a result of the wide functional group tolerance and extremely effective catalytic support, numerous targeted varied BIMs were synthesized and further screened for their biological evaluation.

Scheme 3

Synthesis of Indole-Functionalized BIM Derivative

Scheme 4

Synthesis of Trisindoline: Spiro-analogue 3,3-Di(indol-3-yl)indolin-2-ones

Plausible Mechanism

A plausible reaction mechanism for the water mediated taurine-catalyzed green multicomponent reaction of indole 2a and p-methoxy benzaldehyde 1a is depicted in Scheme . The catalyst used here is taurine, a naturally occurring β-amino sulfonic acid (pKa = 1.5) that appears to be in zwitterion form when dissolved in water. Thus, taurine behaves as a bifunctional donor–acceptor reagent that facilitates the electrophilic substitution reaction by activating the carbonyl electrophilic site, which is then functionalized by the desired nucleophile.[70,71]

Scheme 5

Plausible Reaction Mechanism for Synthesis of 3,3′-((4-Methoxyphenyl)methylene)bis(1H-indole) 3a

As a result, the taurine-activated aldehyde 1a moiety was apparently invite the activated indole 2a through an addition reaction leading to adduct A. In the presence of taurine, intermediate A experienced protonation followed by dehydration to form another intermediate benzylidene-3H-indole C via intermediate B; subsequently, the next activated indole 2a facilitated the formation of bis indole moiety D, which underwent an imine–enamine tautomerization to deliver the targeted BIM derivative 3a.

Biological Evaluation

Initial ligand-based target identification studies indicate that synthesized BIM derivatives efficiently bind three potential anticancer drug target proteins. Analysis of the free energy of binding (ΔG) for each of these drug target proteins clearly suggests high binding affinities for the synthesized BIM derivatives (Supplementary Table S1).

Intriguingly, among these three potential anticancer drug target proteins, the synthesized BIM derivatives were found to bind at the allosteric drug binding pocket of mitotic Kinesisn Eg5 (PDB id: 2PG2). On the contrary, the BIM derivative binding hotspots for the other two anticancer drug target proteins [Human p38 α MAP Kinase (PDB Id: 3HVC) and Human 3 alpha HSD type 3 (PDB id: 1J96)] were found to be located at the active site clefts (data not shown) of the respective enzymes, indicating a probabilistic substrate competitive mode of inhibition. Compared to substrate competitive inhibitors, allosteric inhibitors are always in high demand as they can offer more target specificity and thereby less associated side effects. Moreover, mitotic kinesin Eg5 is one of the most well-established anticancer drug targets explored so far. Inhibition of Eg5 leads to mitotic arrest and apoptosis in neoplasms.[62] Among the three novel allosteric drug binding sites of Eg5 motor domain, the monastrol binding site is the most well characterized allosteric inhibitor binding site. A few monastrol binding site targeting Eg5 inhibitors are already in advanced stages of clinical trials,[63] although the fate of those inhibitors is ambiguous as the monastrol binding site of Eg5 kinesin suffers from moderate to low amino acid conservation compared to the nucleotide binding site (Figure C, all panels). In particular, owing to the emergence of specific drug-resistant mutations (D130A and L214A), the therapeutic future of monstrol binding site targeting allosteric inhibitors is uncertain.[64] Interestingly, the synthesized BIM derivatives 3k and 3b also dock at the same monastrol binding site, which is at the interface of α3 helix, α2 helix, and Loop 5 (L5) of the Eg5 motor domain (Figure A, all panels, and Figure B, leftmost panel). The common 3k and 3b binding sites are approximately 12 and 10 Å away from the nucleotide binding site and catalytic Mg2+ binding sites of the Eg5 motor domain, respectively. The binding mode of docked BIM derivative 3k is structurally reminiscent of the crystal structure of allosteric inhibitor Arry520 (commercial name Filanesib; targets Eg5 monastrol binding site) (PDB id: 6hky). Arry520 is currently under an advanced stage of clinical trials for treatment of multiple myeloma.[72] In order to compare the binding efficiency of synthesized BIM derivatives 3k and 3b with Arry-520, we docked the Arry520 molecule using the same docking platform and parameters used to doc k BIM derivatives. The docked conformation of the Arry520 molecule was found to superimpose well with the published crystal structure (Supplementary Figure 2, all panels). Intriguingly, the comparative analysis of free energy of binding (ΔG) clearly indicates the synthesized BIM derivatives bind more efficiently at the same allosteric site of the Eg5 motor domain compared to Arry-520 (ΔGArry-520 = −8.3 kcal/mol versus ΔG = −11.2 kcal/mol; ΔG = −11.2 kcal/mol). Moreover, 3k binding was found to be unaltered in the presence of the recently reported drug resistant variants (D130 and L214A) of Eg5 (Table ). The highly efficient binding of BIM derivatives 3k and 3b at the allosteric drug binding pocket of Eg5 could be exemplified by the extensive array of hydrophobic interactions of the bound ligands with amino acid side chains of α2,α3 helices and allosteric site guarding loop5 (L5) (Figure A all panels; only 3k has been shown for pictorial clarity).

Figure 2

Plausible role of synthesized BIM derivatives as the monastrol site targeting allosteric inhibitor of mitotic kinesin Eg5. (A) Molecular docking of BIM derivative 3k with the motor domain of Eg5 kinesin. The left panel shows the position of the docked ligand at the monastrol binding allosteric site of Eg5. The middle panel emphasizes the three-dimensional interaction pattern of bound ligand 3k with the monastrol binding allosteric site amino acid residues of Eg5. The right panel shows the detailed two-dimensional interaction profile of the docked ligand 3k with the monastrol binding allosteric site amino acid residues of Eg5. Green dotted line represents the formation of the hydrogen bond. Spiked arches represent hydrophobic interactions. (B) Plausible 3k-mediated inhibition mechanism of Eg5 kinesin in comparison with the well-established allosteric inhibitor Arry520. The left panel shows that the cocrystallized allosteric inhibitor Arry520 and the synthesized BIM derivative 3k targets the same inhibitory site of Eg5. The middle panel and the right panel show the structural superimposition of 3k docked and ADP bound state of Eg5 (colored forest green) with its AMPPNP (ATP analogue) bound state (colored golden yellow). In the right panel, the zoomed view of the superimposed allosteric site clearly indicates that 3k binding will hinder the structural transition from ADP to ATP bound conformational states of Eg5. (C) Bonastrol targeting allosteric site of Eg5 kinesin is not highly conserved. Left panel shows the site-specific evolutionary rates of Eg5 kinesin motor domain mapped on its structure surface and color coded from purple (highly conserved sites) to deep cyan (least conserved sites). The middle and the right panels show the positions of recently reported drug-resistant point mutations of Eg5 with respect to the 3k binding site and the nucleotide binding site.

Table 2

In Silico Molecular Docking Based Binding Energy Estimation of Synthesized Bis(indolyl)methane Derivatives with the Motor Domain of Mitotic Kinesin Eg5 and Its Recently Reported Drug-Resistant Point Mutant Variants

	binding energy (ΔG)
		drug-resistant Eg5 mutants
bis(indolyl)methane derivatives	WT Eg5 kinesin (kcal/mol)	D130A (kcal/mol)	L214A (kcal/mol)
3k	–11.2	–11.1	–10.5
3b	–11.2	–11.0	–9.7

Earlier research works suggest the ADP bound conformation of Eg5 Kinesin is structurally more dynamic compared to that of the ATP bound conformational state. In particular, the flexibility of the Switch-I (amino acids: 229–234) region is responsible for the release of ADP[27] (the end product of previous reaction cycle) and subsequent binding of incoming ATP. The structural stability of the Switch-I region in the ATP bound conformation of the Eg5 motor domain stems from the proper positioning of switch-I amino acid residues (Ser233, Arg234) to stabilize the scissile γ-phosphate group of incoming ATP as well as to activate the water nucleophile needed for ATP hydrolysis.[28] Other than the direct catalytic role of the switch-I region, it also plays crucial role in structuring the switch-II region (amino acids: 265–277), which in turn properly positions the microtubule binding α4 relay helix (amino acids: 278–302) of Eg5 kinesin to dock with α/β tubulin heterodimer. Therefore, the nucleotide-driven differential structural dispositions of the Switch-I and Switch-II regions of Eg5 kinesin play a pivotal role in its microtubule gliding activity. Importantly, just like the other monastrol binding site targeting allosteric inhibitors, the synthesized BIM derivatives 3k and 3b freeze the ADP bound conformation of the Eg5 motor domain by inhibiting the structural transition of the α3 helix needed to release ADP as well as to accept the γ-phosphate group of incoming ATP (Figure B all panels; only 3k has been shown for pictorial clarity). In addition, MD simulation analysis divulges 3k binding to Eg5 motor domain in the presence of bound ADP decreases flexibilities of Switch-I and Switch-II regions compared to only the ADP bound conformation (Figure S2). In the Eg5/ADP/3k complex, the reduced flexibilities of switch-I, switch-II and tubulin binding α4 relay helix mimics the ATP bound Eg5 conformation,[73,74] which would ultimately lock the Eg5 in ADP bound state. Altogether, on the basis of these observations, we propose that the synthesized BIM derivatives 3k and 3b would exert Eg5 inhibitory activity by uncoupling the nucleotide driven conformational changes needed for Eg5 microtubule gliding.

Conclusions

In summary, we have reported an unprecedented approach for the first time to access functionalized derivatives of BIMs in an excellent yield via taurine-catalyzed condensation of structurally diverse aldehydes with indole in water medium. This catalytic application has been proven active under milder reaction conditions, providing broad substrate scope toward accessing varied targeted BIMs for their biological evaluation. Furthermore, a detailed in silico based structure activity evaluation of the synthesized BIM derivatives demonstrates their potential to bind the allosteric inhibitory site of kinesin Eg5, a well-established anticancer drug target protein. The molecular mechanism of synthesized BIM derivative mediated allosteric inhibition of Eg5 kinesin clearly demonstrates their ability to lock Eg5 gliding on a microtubule track. The inhibition mechanism is similar to the previously reported Eg5 inhibitors under clinical trials. However, compared to the previously reported allosteric inhibitors, the superior Eg5 binding efficiency of synthesized BIM derivatives and their unique ability to cope with drug-resistant Eg5 mutations confer immense promise for their use as resistance-resistant antineoplastic therapeutic leads. Overall, the procedure’s simplicity and straightforward operational methodology make it an outstanding strategy for these intriguing and appealing reaction products, which continue to be in high demand. Further, in the search for potent drug candidates, an effort toward the construction of bioactive BIMs and obtaining its advanced biological data is ongoing and will be disclosed in due course.