Three-Component cine,ipso-Disubstitution of Nitrocoumarins

Abstract

The development of a three-component cine,ipso-disubstitution of nitrocoumarins is reported. The reaction leverages the electrophilicity of nitrocoumarins, the nucleophilicity of nitronates, and the leaving group ability of nitrite (NO₂⁻) to yield complex polyfunctionalized biaryls that often display stable axial chirality.

Graphical Abstract

The three-component synthesis of complex (hetero)biaryls is controllably and rapidly critical considering the prominence of this pattern in pharmaceuticals. Examples related to this include the Catellani reaction,¹ iterative cross-coupling transformations,^2-4 and aryne chemistry.⁵ We surmised that nitrocoumarins may serve as valuable starting materials for controlled, 3-component (hetero)aromatic difunctionalization. Specifically, the nitro group could be harnessed 3-fold (Scheme 1A): (1) as an activator for dearomative nucleophilic addition,^6,7 (2) as a carbanion for diastereoselective alkylation,^8-10 (3) and as a leaving group (e.g., pK_a of nitrous acid (HNO₂) = 3.4).¹¹ These three fundamental steps are understood, but their utility in sequence is surprisingly understudied (Scheme 1A).^12,13 For example, the most similar examples to what we aimed to realize and draw attention to are summarized in Scheme 1B,C. Deredas and co-workers¹² found that 3-diethoxyphosphorylcoumarins can react with indole under base mediated conditions. The phosphonate can then react to yield exocyclic methylene by Horner–Wadworth–Emmons (HWE) with formalin (Scheme 1B). Notably, the phosphonate serves critical roles in each individual step. Tang and co-workers¹³ found that Mg(OTf)₂ could catalyze the addition of indoles to 3-nitrocoumarin, the addition of catalytic triphenylphosphine and methyl vinyl ketone yielded the disubstituted dihydrocoumarin (Scheme 1C). Related three-component aromatic substitution reactions by Zhou¹⁴ and Maes^15,16 also inspires and precedes our work. Additional chemistry of note includes vicarious^17,18 and cine^19-21 nucleophilic aromatic substitution (including a photocatalytic variant²²) and nitroarene cycloaddition^23-25 and cyclization²⁶ reactions. Herein we describe a strategy that harnesses the reactivity described in Scheme 1A and often results in axially chiral biaryls yielding products of high functional, structural, and often chiral complexity.^27,28 Included in this study is an initial method, point-to-axial chirality transfer studies,²⁹ and forays beyond the initial method.

Scheme 1.

(A) cine,ipso-Disubstitution of Nitrocoumarins (B and C) Select Related Three-Component Reactions

To demonstrate the value of the cine,ipso-disubstitution of nitroarenes, we explored the reaction of nitrocoumarins 1 with nucleophilic indoles 2 and allylic electrophiles 3 promoted by a base additive. The isolated products 6, prepared via intermediates 4 and 5, are indole–coumarin biaryls with an alkene functional handle; valuable products considering the ubiquity of indoles³⁰ and coumarins³¹ in drug discovery. The Supporting Information contains optimization studies related to this three-component coupling. Scheme 2 describes the scope studies. First studied was indole substitution on the benzene ring (6a–6h). Electron withdrawing groups at the 5-position of indole were well tolerated (6b–6d) including often sensitive aldehydes (6c) and boronic acids (6d). Product 6e bearing a 4-bromo group yielded the desired product but in modest yield. This was a steric problem associated with the conjugate addition step between nitrocoumarin 1a and indole 2e. Electron donating groups at the 5-position were also well tolerated (6f, 6g). Notably, a free phenol was incorporated into the biaryl scaffold (6g). We were also able to prepare product 6h bearing a basic nitrogen from 7-azaindole. N-Methylindole also was a competent coupling partner, yielding product 6i. Indoles bearing various 2-functional groups were competent coupling partners, generally speaking. For example, methyl (6j), phenyl (6k), ester (6l), amides (morpoline (6m), glycinate (6n), and allyl (6o) containing) and aldehyde (6p) functional groups were tolerated at the 2-position, though the 2-ester indole (2l) was a noticeably poorer nucleophile. Also, all the 2-EWG-substituted indoles required the addition of Et₃N to promote the conjugate addition (e.g., 2l–2p → 5l–5p). With respect to the allyl acetate starting material, we were able to make products 6q–6s derived from cinnamyl acetate, and the acetates of 2-methylenepropane-1,3-diol and cis-butene-1,4-diol, respectively. The last examples in Scheme 2 show that substitution on the nitrocoumarin is also tolerated at many positions. We prepared products containing a bromo-functional handle at the 5–8-positions of coumarin (6t–6x) and examined electron-rich coumarins (6y). Notably, the 5-substituted coumarins suffer from inefficient H─NO₂ elimination chemistry (Scheme 2A,B). For this elimination to occur, the respective intermediates 5t and 5u were first isolated (88% and 74% yield, respectively), and then the elimination step was reoptimized, resulting in 1,1,3,3-tetramethylguanidine (TMG) being identified as the superior base. Considering that the N-Me substrate 5u reacted more efficiently, we suspected that the free indole N─H was hindering the elimination step. Indeed, it was found that 5t, upon in situ protection and TMG-promoted elimination, yielded the N-Boc-protected analogue 7 in improved yield compared to 6t.

Scheme 2.

(A) Scope of Three-Component Biaryl Synthesis; (B) Re-examination of the Elimination Step

The products 6 from the three-component coupling clearly display axial chirality (enantiotopic allylic signals in ¹H NMR). On this line, no signal coalescence was observed for 6a via VT-NMR up to 130 °C in DMSO-d₆ and an enantioenriched sample of 6j was found to be conformationally stable. As such, we wished to examine the possibility of a point-to-axial chirality transfer in this type of system. Point-to-axial chirality transfer would be possible if the potential indole rotamers (5j-rotamers A and B, Scheme 3A) displayed significant energy differences or if the rates at which the rotamers undergo HNO₂ elimination are significantly different (or both). Interestingly, it was found that 5j (er = >99:1; enantiomers separated via chiral supercritical fluid chromatography (SFC)) undergoes an elimination with 62–71% enantiospecificity (up to 85:15 er) (Scheme 3B). While imperfect, this result is a proof of concept for accessing unique axially chiral biaryls via cine,ipso-disubstitution. Understanding more deeply the impact of rotamer-biasing and the elimination step will be key.

Scheme 3.

Point-to-Axial Chirality Transfer Depends on the Energetics of Rotamerization and Elimination

The chemistry developed herein yields complex and functionalized indole–coumarin biaryls. Notably, indoles and coumarins are both common to drugs and drug discovery efforts, though biological/pharmaceutical studies on their biaryl conjugates are minimal.^32,33 It is possible that the bioactivity of indole–coumarins has remained understudied due to the lack of methods capable of constructing diverse and complex (structural, functional, and chiral complexity) analogues. As such, this work represents a direct and simple entry into novel and complex indole–coumarin biaryls. On this line, the indole–coumarin 8a bearing four distinct functional groups (indole N─H, aldehyde, alkene, and Csp2─Br functional groups) was prepared on the gram-scale (Scheme 4A). In unoptimized studies, using standard conditions from the literature, 8a was successfully derivatized via imine formation (8b), Sonogashira cross coupling (8c), ozonolysis (8d) (with NaBH₄ workup), cross metathesis (8e), and Suzuki cross coupling (8f). Thus, from commercial sources, products 8a–8f are prepared in two to three steps, showcasing that diverse, axially chiral indole–coumarins are readily accessible.

Scheme 4.

(A) Functional Group Reactions; (B) Other Coupling Partners for cine,ipso-Disubstitution of Nitrocoumarins

In terms of the methodology, indole,allyl-cine,ipso-disubstitution of nitrocoumarins is a proof-of-concept study: In principle, coupling partners can be exchanged, accessing an even greater diversity of chemical space. In this regard, we examined the coupling of a nucleophilic benzothiophene (Scheme 4B, eq 1) and an arylBF₃K salt (Scheme 4B, eq 2) as well as electrophilic SelectFluor (Scheme 4B, eq 3) and ethyl acrylate (Scheme 4B, eq 4) with nitrocoumarin. Somewhat surprising was that benzothiophene did not react similarly to indole. Neutral conditions (benzothiophene + nitrocoumarin) did not promote the dearomative conjugate addition reaction. However, 1 equiv of AlCl₃ promoted this step as a 2:1 inseparable mixture of benzothiophene isomers 9a and iso-9a. Utilizing tandem Rh(I) and Pd(0) catalysis, it was found that the aryl,allylcoumarins could be prepared. The Rh(I) catalyst first promotes the addition of dearomative conjugate addition. Then Pd(0) and TMG promotes the allylation and eliminative aromatization to 10 in modest, unoptimized 18% yield. By swapping allyl acetate/Pd(0) for SelectFluor, we were able to prepare indole,fluoro-disubstituted coumarin 11 in 35% yield.³⁴ We were also able to perform an indole-alkylation of nitrocoumarin using PPh₃ to promote both the Michael addition³⁵ and the eliminative rearomatization step. In this case, indole–coumarin biaryls 12 now contain a carboxylate functional group in place of the alkene from the standard protocol (Scheme 2). These results suggest that there are many ways to consider difunctionalizing nitrocoumarins via a cine,ipso-disubstitution approach.

One final methodological deviation that yielded interesting results was the cine,ipso-disubstitution of nitrocoumarins with phenolic nucleophiles (Scheme 5). It was observed that treating nitrocoumarins and phenols with AlCl₃ in CH₂Cl₂ followed by allyl acetate, Pd(PPh₃)₄, and Et₃N yielded unexpected isomeric products as the major products (iso-14) and the “originally anticipated” products 14 in minor amounts (Scheme 5A). The p-methoxylphenol nucleophile resulted in good yields of iso-14a (40% yield) and iso-14b (55% yield). The minor products, 14a and 14b, could not be isolated cleanly but were observed during analysis of crude materials. When p-chlorophenol was used as the nucleophile, a 27% yield of iso-14c was achieved along with isolable and characterizable minor product 14c. After careful experimentation and NMR analysis (including ¹H, ¹³C, gHMBC, and NOE analyses), it was determined that the products 14/iso-14 are arising via two distinct trans-lactonizations (Scheme 5B). The dearomative Michael addition yielded 13c and iso-13c. Data supports that 13c with trans-stereochemistry is stable, but the cis-diastereomer undergoes goes trans-lactonization to iso-13c via a tetrahedral intermediate [Int-A]. Pd-catalyzed allylation of 13c/iso-13c yields separable regioisomers allyl-13c and iso-allyl-13c with a cis-relationship between the allyl and the phenolic group, likely due to a phenolic directing effect. When subjected to pyridine-d₅, the aromative elimination reaction took place to yield iso-14c and 14c, respectively, via a trans-lactonization to establish the necessary stereochemistry for E2 elimination. Finally, products 14 clearly display stable axial chirality as determined by ¹H NMR and chiral HPLC analyses.

Scheme 5.

(A) cine,ipso-Disubstitution with Concomitant trans-Lactonization; (B) Proposed Mechanism

We have developed a strategy for the three-component cine,ipso-disubstitution of nitrocoumarins with various nucleophiles and electrophiles. The primary focus in this work was on three-component coupling with indoles and allylic electrophiles, though other nucleophiles and electrophiles were also examined. Further, point-to-axial chirality transfer can be accomplished, though additional design of substrates will need to be considered to improve the stereospecificity. This work will open broad possibilities: (1) What other nitroarenes can participate in cine,ipso-disubstitution? (2) What other nitrocoumarin-disubstitution reactions can be developed (3) Can point-to-axial chirality be improved and understood for enantioselective synthesis? We plan to study these questions and anticipate that we and others may find this strategy useful for streamlining complex biaryl synthesis.

Data Availability Statement

The data underlying this study are available in the published article and its Supporting Information.