Brønsted Acid-Catalyzed Synthesis of 4-Functionalized Tetrahydrocarbazol-1-ones from 1,4-Dicarbonylindole Derivatives

A p-toluenesulfonic acid-catalyzed cascade reaction is reported for the synthesis of 4-functionalized tetrahydrocarbazolones via the reaction of 4-(indol-2-yl)-4-oxobutanal derivatives with a variety of nucleophiles in acetonitrile or hexafluoroisopropanol. After the initial intramolecular Friedel–Crafts hydroxyalkylation, the 3-indolylmethanol intermediate is subsequently activated and reacted with the external nucleophile. The reaction conditions are crucial to avoid alternative reaction pathways, allowing direct substitution reaction with thiols, (hetero)arenes, alkenes, or sulfinates. The procedure features high overall yields to access a diverse family of compounds bearing the tetrahydrocarbazole core.


■ INTRODUCTION
Friedel−Crafts (FC) alkylation represents a key tool for the functionalization of (hetero)arenes and the preparation of relevant aromatic compounds in organic synthesis. 1Recent advances in the field are mainly related to the development of suitable mild catalysts and the use of alternative and more environmentally friendly electrophilic partners such as alcohols, aldehydes and ketones. 2 The intermolecular Brønsted or Lewis acid-catalyzed reaction between aldehydes and arenes (Scheme 1a, eq 1) represents a useful tool for the construction of relevant triarylmethanes or 1,1,-diarylalkanes (when NuH = Ar) 3 or benzyl-functionalized substrates (for other NuH). 4 This type of reaction is a tandem process consisting of a FC hydroxyalkylation followed by direct nucleophilic substitution of the resulting alcohol, which can also be considered as a FC alkylation.However, the version in which one of the steps proceeded intramolecularly remains underexplored, even though this strategy provides efficient access to the construction of unsymmetrical cyclized products.In this case, the other step could also take place intramolecularly (Scheme 1a, eq 2a), 5 or with an external nucleophile (Scheme 1a, eq 2b). 6n the other hand, the indole core is ubiquitous in numerous biologically active compounds, and therefore the development of methodologies for direct indole functionalization is an open challenge in organic synthesis, in which its FC reaction through the nucleophilic C3 position is one of the most useful approaches. 7n this field, previous work involving the methodology shown in eq 2b, with indole derivatives bearing an aldehyde group, is limited to a few examples where the aldehyde is aromatic, leading to five-membered rings and requiring the use of metallic catalysts (Scheme 1b, eq 3). 8A related substrate, such as 4-(indol-2-yl)-4-oxobutanal, has been prepared by Moody et al., but it cyclizes to 1-methoxycarbazol on treatment with BF 3 /MeOH, where the external nucleophile attacks the ketone instead of the double reaction with the aldehyde (Scheme 1b, eq 4). 9n this context, and following our interest in the development of Brønsted acid-catalyzed functionalization of indole derivatives via direct nucleophilic substitution reactions, 10 we proposed to switch the reactivity described for 4-(indol-2-yl)-4-oxobutanals, preventing their carbazole formation, and thus allowing the synthesis of functionalized tetrahydrocarbazolones via the strategy established in eq 2b, i.e. a tandem intramolecular FC hydroxyalkylation/intermolecular direct S N (Scheme 1c).
The tetrahydrocarbazole scaffold, 11 including tetrahydrocarbazolone derivatives, is an important structural motif that appears in various molecules with biological activity, 12 and has been shown to be a useful intermediate platform for the synthesis of several carbazole derivatives. 13Thus, the development of new approaches to functionalized tetrahydrocarbazolones, in addition to the classical Fischer indole synthesis with arylhydrazones derived from 1,2-cyclohexanediones, α-aminocylcohexanones, or generated via the Japp−Klingemann, 14 intramolecular FC of indolecarboxylic acids, 15 and the oxidation of tetrahydrocarbazoles, 16 remains an interesting goal in the field. 17In fact, only particular examples of tetrahydrocarbazol-1-ones further functionalized at C-4 have recently been reported in the literature, 18 and to the best of our knowledge, there is no specific synthetic route for their preparation.
Herein, we present our results on the development of an efficient strategy for the synthesis of 4-functionalized tetrahydrocarbazol-1-ones from readily available 4-(indol-2yl)-4-oxobutanals.

■ RESULTS AND DISCUSSION
For the preparation of the required 4-(indol-2-yl)-4-oxobutanal 1a we envisaged a two-step synthetic route consisting of the reaction of 2-lithio-1-methylindole with γ-butyrolactone, 19 leading to the ketoalcohol 2a, and further oxidation of the primary hydroxyl group (Scheme 2a).However, the overall yield was moderate due to the competitive double addition of the organolithium to the lactone.An alternative approach involved the use of 2-hydroxycyclobutanone as an electrophilic reagent for 2-lithio-1-methylindole, giving rise to cyclobutane-1,2-diol derivative 3a.Subsequent oxidative cleavage with DMSO under dioxomolybdenum catalysis 20 afforded 1a but, again, with moderate overall yield (Scheme 2b).In search of a better result, we tested the reaction of the lithiated indole with the Weinreb amide, 21 which gave access to silyl-protected 2a, which was further desilylated and oxidized with the Dess-Martin reagent, providing 1a in a significant overall yield of 60% (Scheme 2c).On the other hand, for the preparation of the previously described 1b, we initially followed the reported procedure.However, in our hands, the final oxidation delivered a very poor yield of the desired 1b, which was not improved by using other oxidants, such as Dess-Martin periodinane or Swern reaction, instead of PCC (Scheme 2d).We therefore developed an alternative route involving lithiation of the Nsulfonyl protected indole and trapping with γ-butyrolactone to give 2b.Its oxidation, acetalization, 22 and further basic hydrolysis yielded the dimethylacetal 4b with a useful overall yield (Scheme 2e).We also carried out the acetalization of 1a to give 4a in high yield.
First, 1a was treated with MeOH under the conditions reported by Moody et al. 9a and, as in their results, 1methoxycarbazole 5 was isolated in high yield (Scheme 3).This product formally arises from the annulation of the indole with the aldehyde and the reaction of the ketone with the external nucleophile.At this point, we thought that softer nucleophiles, rather than O-centered ones, might lead to the desired concurrent addition of the indole and the external nucleophile to the same carbonyl group.Gratifyingly, when 4chlorothiophenol was employed under the same reaction conditions, tetrahydrocarbazolone 6a was obtained in good yield, instead of the thio-functionalized carbazole, analog to 5 (Scheme 3).
Considering the potential usefulness of this process for the synthesis of functionalized tetrahydrocarbazolone derivatives, we next further evaluated the reaction of indolyl-functionalized The Journal of Organic Chemistry γ-ketoaldehyde 1a with 4-chlorothiophenol, looking for optimal and softer conditions than those employed in Scheme 3 (Table 1).First, BF 3 could be lowered to catalytic amounts with an even better yield, but with trace amounts of dithioacetal 7a, derived from a competitive reaction of the thiol with the aldehyde (entries 1 and 2).Other Lewis acids, such as Cu(OTf) 2 , also promoted the reaction, but with more competitive dithioacetal formation (entry 3).Gratifyingly, a simple Brønsted acid, such as p-toluenesulfonic acid monohydrate (p-TsOH), provided similar results to BF 3 •OEt 2 (entry 4).However, diphenyl phosphate was less effective in this transformation (entry 5).Due to its easy availability and handling, we selected p-TsOH as the catalyst for the subsequent studies.The use of 1 equiv of p-TsOH resulted in a higher amount of the competitive dithioacetal 7a (entry 6).The same tendency was observed when increasing the amount of the thiol (entry 7).We checked that using 10 mol % of p-TsOH the reaction was completed in 2 h with an even better yield and with only trace amounts of 7a (entry 8).Other solvents such as hexafluoroisopropanol (HFIP), MeNO 2 or toluene provided 6a with similar efficiency (entries 9−11).Interestingly, the reaction is also completed in HFIP without the acid catalyst, although it requires heating at 60 °C for a longer time (entry 12).In the absence of the thiol, only decomposition was observed in both MeCN and HFIP (entries 13 and 14).
After optimizing the reaction conditions (entry 8, Table 1), we evaluated the scope of the reaction for the synthesis of 4thiotetrahydrocarbazol-1-ones 6 (Table 2).Thiophenols bearing electron-withdrawing substituents led to tetrahydrocarbazolones 6a−e in good yields (entries 1 and 3−6).We also checked that the dimethylacetal 4a, derived from 1a, behaves similarly in its reaction with a thiol, although a slightly lower yield was obtained (entry 2).Other arylthiols are useful counterparts (entries 7 and 8), but when a more electron-rich thiophenol such as 3-methoxybenzenethiol was used, the corresponding carbazolone 6h was obtained in only moderate yield due to the competitive formation of the corresponding dithioacetal 7h (entry 9).This effect was even more pronounced when an aliphatic thiol was used, leading in this case to a lower yield of the 4-alkylthio carbazolone 6i (entry 10), indicating that competitive thioacetalization is favored with more nucleophilic thiols.Fortunately, ethyl mercaptoacetate could be engaged in the process leading to the esterfunctionalized carbazolone 6j (entry 11  The Journal of Organic Chemistry moderate yield (entry 12).Interestingly, as expected from the result shown in Table 1 (entry 9), the reactions proceed with similar efficiency in HFIP. 23t this point, we decided to extend the scope of the process by testing other π-nucleophiles.We focused our attention on indoles and started with N-methylindole under the optimal conditions described for thiols (Scheme 4).Surprisingly, an almost equimolar mixture of tetrahydrocarbazolone 8a and carbazole 9a was obtained, which could be isolated independently.After some experimentation trying to control the selectivity of the reaction, 24 and considering that HFIP has relevant properties (high polarity, relatively acidic OH) that make it a suitable solvent for the direct nucleophilic substitution of alcohols, 25 we found that a simple change of solvent from MeCN to HFIP led exclusively to 8a (Scheme 4).In addition, HFIP is known to increase the acidity of p-TsOH through hydrogen bonding interactions.3f,6d Once the reaction conditions were reoptimized, the scope of the 4-indolyltetrahydrocarbazol-1-ones 8 was evaluated using different indoles as nucleophiles (Scheme 5).N-Methylindoles with different substitution at C-2 led to tetrahydrocarbazolones 8a−c in high yields.The structure of 8a was further confirmed by X-ray analysis. 26The use of NH-indoles was equally effective and allowed the synthesis of 8d,e which were also obtained in high yields.5-Substituted indoles with both electron-withdrawing and electron-donating groups could also be employed to give the corresponding indolyl carbazolones 8f−i.Similarly, 6-nitroindole delivered 8j in high yield.Finally, when skatole was reacted with 1a, the attack through C-2 led to the carbazolone derivative 8k also in very high yield (Scheme 5).Surprisingly, when the reaction of the indoles with 1a was carried out in MeCN, the corresponding 1indolylcarbazoles 9 were only produced in trace amounts, with the exception of N-methylindole and NH-indole, which allowed the isolation of 9a and 9d in 30% and 10% yield, respectively. 27In any case, the yields for the synthesis of 8 were consistently higher in HFIP than in MeCN.
With suitable catalytic conditions for the preparation of 4indolyltetrahydrocarbazol-1-ones 8, we turned our attention to evaluating the applicability of this strategy to the synthesis of various 4-(hetero)aryltetrahydrocarbazolones 10 by employing other suitable electron-rich (hetero)aromatics as nucleophiles (Scheme 6).A selection of these functionalized carbazolones 10 were readily prepared by varying the nucleophilic partner with oxoaldehyde 1a and dimethylacetal 4b.Methoxyfunctionalized benzenes, including 1,3,5-trimethoxybenzene, 1,3-dimethoxybenzene, or 3,4-dimethoxyphenol, reacted regioselectively with 1a to give 4-aryltetrahydrocarbazolones Scheme 4. Brønsted Acid Catalyzed Reaction of 1a with N-Methylindole Scheme 5. Synthesis of 4-Indolyltetrahydrocarbazol-1-ones 8 Scheme 6. Brønsted Acid-Catalyzed Reaction of 1a and 4b with (Hetero)aromatics, 1,1-Diphenylethylene and Sodium Benzenesulfinate The Journal of Organic Chemistry 10a−d, which were isolated in high yields.Interestingly, 10a could be prepared on the 2 mmol-scale, allowing the isolation of 534 mg (73% yield) of this substrate.However, other arenes attempted, such as 1,2,3-trimethoxybenzene and 2,6-dimethoxyphenol, only led to decomposition, showing that a delicate balance between the nucleophilicity of the external and the internal nucleophiles is essential for the success of the reaction.Interestingly, electron-rich heteroaromatics, such as selected furans, thiophenes and pyrroles, were able to participate in this process, yielding the 4-heteroaryltetrahydrocarbazolones 10e− k in good yields and complete regioselectivity except in the case of employing 3-methoxythiophene, which afforded 10j as an approximately 5/1 mixture of regioisomers.Other heteroarenes like N-methylpyrrole, benzofuran or benzothiophene were unsuccessful partners. 28In addition, sodium benzenesulfinate could also be used as an external nucleophile, giving rise to 4-sulfonyltetrahydrocarbazolone 10l in moderate yield, as complete conversion could not be achieved.Gratifyingly, 1,1-diphenylethylene efficiently participated in the reaction to give 4-alkenyltetrahydrocarbazolones 10m,n in high yields (Scheme 6).In general, slightly lower yields were obtained for the NH-carbazolones derived from 4b.
Our proposal for the formation of the tetrahydrocarbazolone derivatives 6, 8 and 10 from 1,4-ketoaldehyde 1a is outlined in Scheme 7.Although PSTA in MeCN efficiently catalyzes the process, the reaction is enhanced when employing HFIP as the solvent.In this sense, the complexation of the PTSA with HFIP molecules increases its acidity, facilitating the interaction with the solvated ketoaldehyde 1a.So, initially, the activation of 1a by the acid catalyst ([HA]) could generate the intermediate A. Then, the acid-catalyzed intramolecular attack of the indole to the activated aldehyde would release a solvated 4-hydroxy-2,3,4,9-tetrahydro-1H-carbazol-  29 It is worth pointing out that HFIP is not only a stronger acid (pK a = 9.3) than related alcohol iPrOH and an excellent hydrogen bond donor, but also its lower nucleophilicity and strong ionization power make HFIP an ideal medium for generating cations facilitating a variety of synthetic transformations. 30In this sense, as mentioned in the optimization (Table 1, entry 12), the HFIP molecules could also activate the 1,4-ketoaldehyde 1a to obtain A′, facilitating an intramolecular attack of the indole to generate B′, although with lower efficiency than when PTSA is used.Then, solvated intermediated B′ could be directly transformed into cationic indoleneiminium D′ that, after the attack of the thiol nucleophile, released the functionalized tetrahydrocarbazolone 6a.
In addition, taking advantage of γ-ketoaldehyde 1a, we envisaged that tetrahydrocarbazolones 12 functionalized with an acylmethyl group at C-4 could be accessed in a two-step process (Scheme 8).First, a selective Wittig reaction with selected stabilized ylides provided the corresponding diketones 11.Then, the intramolecular Michael addition of the indole could be efficiently catalyzed by AuCl 3 31 leading to the expected 4-acylmethyltetrahydrocarbazolones 12 in high yields (Scheme 8).
At this point, we tried to increase the synthetic value of our protocol for the synthesis of tetrahydrocarbazolones by their further transformation.For example, treatment of 10a with EtMgBr gave rise to the expected alcohol, which was purified by silica gel chromatography to afford dihydrocarbazole 13 in moderate yield (Scheme 9).Selected carbazolone 10f was αalkylated by base-mediated-enolization and subsequent reaction with methyl iodide, giving carbazolone 14 with low The Journal of Organic Chemistry diastereoselectivity (Scheme 9).Finally, the reduction of indole-functionalized carbazolone 8f led to the expected alcohol 15 as a mixture of diastereoisomers (Scheme 9).

■ CONCLUSIONS
In conclusion, an efficient cascade reaction has been described for the synthesis of 4-functionalized tetahydrocarbazolones involving a tandem intermolecular FC hydroxyalkylation/ intermolecular direct nucleophilic substitution of readily accessible 1,4-dicarbonylindole compounds using p-TsOH as a cheap, readily available, and easy to handle Brønsted acid catalyst.After the initial attack of the indole to the aldehyde, which enables the formation of the tetrahydrocarbazolone core, the key intermediate bearing the structure of 3indolylmethanol could be reactivated by the action of the acid catalyst, allowing the subsequent reaction with a wide variety of external nucleophiles such as thiols, (hetero)arenes, alkenes or sulfinates.By fine-tuning the reaction conditions, the competitive dehydration of this crucial intermediate, which leads to the alternative and previously described carbazole formation process, is prevented.Moreover, the tetrahydrocarbazolones obtained are suitable for further derivatization reactions, providing access to a variety of compounds containing the valuable tetrahydrocarbazole core.
■ EXPERIMENTAL SECTION General Methods.All reactions involving air-sensitive compounds were carried out under an N 2 atmosphere in oven-dried glassware.All common reagents and solvents were purchased from commercial suppliers and used without any further purification.TLC was performed on alumina-backed plates coated with silica gel 60 with F 254 indicator, using UV light or Ce/Mo solution and heat as a visualization agent.Flash silica gel chromatography was performed using Merk silica gel 60, 230−240 mesh.NMR spectra were recorded on a Varian Mercury Plus or Bruker Advanced III HD (300 MHz 1 H; 75.4 MHz 13 C, 282 MHz 19 F) or Bruker Advanced NEO 4500 (500 MHz 1 H, 126 MHz 13 C) instrument at room temperature.Chemical shifts (δ) are reported in ppm, using residual solvent peak as the internal reference (CDCl 3 : δ H = 7.26 and δ C = 77.16;(CD 3 ) 2 CO: δ H = 2.05 and δ C = 29.84 and 206.26;DMSO-d 6 : δ H = 2.50 and δ C = 39.50).Coupling constants (J) are given in hertz (Hz).Data are reported as follows: chemical shift, multiplicity (s: singlet, bs: broad single, bm: broad multiplet, d: doublet, dd: doublet of doublets, ddd: doublet of doublets of doublets, dddd: doublet of doublets of doublets of doublets, dq: doublet of quartets, dt: doublet of triplets, ddt: doublet of doublets of triplets, dtd: doublet of triplets of doublets, td: triplet of doublets, t: triplet, tt: triplet of triplets, q: quartet, m: multiplet), coupling constants and integration.Carbon multiplicities have been assigned by DEPT experiments.Low-resolution electron impact mass spectra (EI-LRMS) were obtained at 70 eV, and only the molecular ion and/or base peaks and significant MS peaks are given.High-resolution mass spectra (HRMS) were recorded on an instrument equipped with a QTOF analyzer using ESI (+) or APCI (+).Melting points were measured on a Gallenkamp apparatus using open capillary tubes and were uncorrected.For simplicity, the ptoluenesulfonic acid monohydrate is represented as p-TsOH.
Synthesis of 4-Indol-2-yl-4-oxobutanal 1a: Procedure I-a.To a stirred solution of N-methylindole (5.24 g, 40 mmol) in anhydrous Et 2 O (40 mL) was added n-BuLi (16 mL, 40 mmol, 2.5 M solution in hexane) at 0 °C, and the resulting mixture was heated at 40 °C for 2 h.Next, γ-butyrolactone (5.16 g, 60 mmol) was added at 0 °C and the resulting mixture was stirred for 2 h at 0 °C.Then, the mixture was quenched with aq.NH 4 Cl (5 mL).THF was removed under reduced pressure, and the aqueous layer was extracted with EtOAc (3 × 15 mL).The combined organic layers were dried over anhydrous Na 2 SO 4 , filtered and concentrated in vacuo.The residue was filtered through a pad of silica to remove the excess of Nmethylindole using a mixture of hexane/EtOAc (2/1) to afford the alcohol 2a (3.47 g), which was not isolated in pure form.To a solution of the obtained alcohol 2a (3.47 g) in DCM (40 mL) was added DMP (22 g, 52 mmol) and the resulting mixture was stirred at rt for 1 h.Then, volatiles were removed under reduced pressure and the residue was purified by flash column chromatography using a 5/1 mixture of hexane/EtOAc as eluent to afford ketoaldehyde 1a as a brown solid (2.67 g, 31% referred to N-methylindole).
Synthesis of 4-Indol-2-yl-4-oxobutanal 1a: Procedure I-c.Synthesis of S1: 21 To a solution of N,O-dimethylhydroxylamine hydrochloride (5.4 g, 55 mmol) in anhydrous DCM (100 mL) was added dropwise dimethylaluminum chloride (55 mL, 55 mmol, 1 M in hexane) at 0 °C.The resulting mixture was stirred at this temperature for 1 h.γ-Butyrolactone (4.31 g, 50 mmol) was added slowly and the mixture was stirred for 30 h at rt.Then, the reaction was quenched by slow addition of water (50 mL).The aqueous layer was extracted with DCM (3 × 40 mL) and the combined organic Scheme 9. Derivatization of Selected Functionalized Tetrahydrocarbazolones 8 and 10 The Journal of Organic Chemistry layers were dried over anhydrous Na 2 SO 4 , filtered and concentrated in vacuo yielded 4-hydroxy-N-methoxy-N-methylbutanamide as a yellowish oil which was used in the next step without further purification.Next, to a solution of 4-hydroxy-N-methoxy-Nmethylbutanamide (7.36 g, 50 mmol) and imidazole (10.21 g, 150 mmol) in DMF (50 mL) TBSCl (11.3 g, 75 mmol) was added at 0 °C.The mixture was stirred at rt for 3 h.Then, the reaction was quenched with water and extracted with EtOAc (3 × 30 mL).The combined organic layers were washed with brine (2 × 30 mL), dried over anhydrous Na 2 SO 4 , filtered and concentrated in vacuo.The residue was purified by flash column chromatography using a 5/1 mixture of hexane/EtOAc as eluent affording S1.
General Procedure VI for the Synthesis of Diones 11.To a stirred solution of 4-(1-methyl-1H-indol-2-yl)-4-oxobutanal (1a) (107 mg, 0.5 mmol) in anhydrous DCM (5 mL) was added the corresponding ylide (0.55 mmol), and the resulting solution was stirred at rt for 16 h.The residue was purified by flash column chromatography using silica gel and mixtures of hexane/EtOAc as eluent to afford the corresponding diones 11a−b.

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

Accession Codes
CCDC 2307505 contains the supplementary crystallographic data for this paper.These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing data_request@ccdc.cam.ac.uk, or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.

Table 1 .
Scheme 3. Preliminary Results: Reactions of 1a with MeOH and 4-ClC 6 H 4 SH Optimization of the Reaction Conditions for the Synthesis of Tetrahydrocarbazolone 6a a ClC 6 H 4 SH (0.15 mmol, unless otherwise established), catalyst (10 mol %, unless otherwise established), solvent (1 mL), rt, under N 2 atmosphere.bDeterminedby1H NMR using 1,3,5-trimethoxybenzene as internal standard.c 10 equiv of BF 3 .OEt 2 were used.d 1 equiv of p-TsOH was used.e 1.5 equiv of the thiol were employed.f Carried out at 60 °C.At rt 80% of conversion.g No thiol was added.Only decomposition products were observed.
).Finally, to check if NH substrates could be employed, the acetal 4b was reacted with the model thiol to deliver the NH carbazolone 6k in a Reaction conditions: 1a (0.15 mmol),4-
•Et 2 O (1 mL, 4 mmol, 48% solution in Et 2 O), and the resulting mixture was stirred at rt for 16 h.Then, the reaction was quenched with water (2 mL) and extracted with Et 2 O (3 × 5 mL).