Matteo
Corrieri
,
Lucia De
Crescentini
,
Fabio
Mantellini
,
Giacomo
Mari
,
Stefania
Santeusanio
and
Gianfranco
Favi
*
Department of Biomolecular Sciences, Section of Chemistry and Pharmaceutical Technologies, University of Urbino “Carlo Bo”, Via I Maggetti 24, 61029 Urbino, Italy. E-mail: gianfranco.favi@uniurb.it
First published on 6th September 2022
A practical, robust and chemoselective approach toward the synthesis of pyrrolo[2,3-b]indoles via direct intramolecular C–H bond amination of α-indolylhydrazones has been achieved. This base- and oxidant-free chemoselective transformation relies on a Cu/Fe co-catalyst system that operates at 50 °C in air with water as the only reaction medium. The easy product isolation together with the recyclable catalyst aqueous system (reused at least five times, maintaining over 50% of its catalytic activity) can provide an effective environmentally benign approach to fused N-heterocycles of remarkable interest in pharmaceutical and medicinal chemistry. The ability of the hydrazone residue to act as a chelating/directing group as well as an aminating agent guarantees the success of this C–H functionalization.
Within this context, noble catalysts (e.g., Pd, Rh, Ir and Ru) along with stoichiometric or excess amount of oxidant (e.g., Cu(OAc)2, AgOAc, PhI(OAc)2, CeSO4, and/or F+) and complex/specialized auxiliary ligands and/or additives have been employed predominantly. However, most of the metal catalysts operate at rather high temperature in toxic, polar, aprotic organic solvents (e.g., N,N-dimethylformamide (DMF), N-methylpyrrolidone (NMP), and N,N-dimethylacetamide (DMAc)) which meeting with limited success when used in water. Even with the enormous progress made over the past decade in the most emerging areas of photocatalytic34–36/electrochemical37–43/photoelectrocatalytic44 oxidative cross-coupling, however, there is still a necessity for the development of greener alternative and more efficient applicable methods which do not rely on noble metal catalysts45–48 (e.g., Cu, Fe, Zn, Mn, Co and Ni), avoid the use of toxic and/or hazardous organic solvents, and circumventing the need for external oxidants.49,50 Moreover, despite its conceptual simplicity, the intramolecular C–H bond amination of hydrazones51–56 possessing a “privileged” indole ring to afford value-added N-fused indoles remains elusive. Just recently, we reported a protocol for the synthesis of azacarbolines via PhIO2-promoted six-membered cycloamination-oxidation of α-indolylhydrazones57 (Fig. 1b, previous work). We envisioned that the pendant hydrazone residue in α-indolylhydrazones could be responsible for a five ring-closing C–H amination as result of the potential hydrazone–enamine tautomerization,58,59 thus providing a distinct approach for the C(2)–H functionalization of indoles and expedient synthesis of pyrroloindoles60 (Fig. 1b, this work). To accomplish this, the hydrazonic residue should serve both as a chelating/directing group61–63 and as an intramolecular nitrogen source51–57 under the action of the opportune metal.
In designing a complementary and convenient strategy to produce less saturated version of pyrrolo[2,3-b]indole molecules64–71via C–H bond functionalization, we herein report an unprecedented intramolecular C(sp2)–N bond amination strategy that utilizes a combination of more advantageous Cu72–75/Fe76–78 (the most abundant in the Earth's crust) catalyst, at 50 °C in the presence of air as terminal oxidant in aqueous system/medium (Fig. 1b, this work).
Besides remarkable biological activities exemplified by representative compounds such as pyrroindomycins (PYRs) Sirtuins inhibitor, and inibithor of growth of Bacillus subtilis (Fig. 1c), these 1,8-dihydro pyrrolo[2,3-b]indoles exhibit a broad spectrum of applications in optoelectronic materials and fluorescent probes.
We began our studies by testing the conversion of α-indolylhydrazone 1a to 1-amino pyrrolo[2,3-b]indole 2a using simple copper/iron salts (Table 1). After preliminary screening of a variety of conditions including nature of copper source, (co)catalyst loading, solvent, additive and temperature (see Table S1 of the ESI† for more details), we found that a combination79–87 of catalytic [commercially accessible] Cu(OAc)2·H2O (10 mol%) and FeCl3·6H2O (5 mol%) in an open flask at room temperature using water as the only reaction medium,88–91 was beneficial for this transformation, with the compound 2a being formed in 99% yield (entry 1). Reactions that employed other iron cocatalysts such as Fe2O3, Fe(NO3)3 9H2O, Fe(ClO4)3, Fe2(SO4)3·H2O or Fe(acac)3 led to inferior results in term of reactivity/efficiency (entries 2–6). Reducing the amount of catalyst/co-catalyst did not affect the yield albeit an extended reaction time was required for complete consumption of 1a (entry 7). On the other hand, we found that a mild heating (50 °C) significantly accelerated the reaction, and 1a was produced in almost quantitative yield within 3 h (entry 8).
Entry | Catalyst [mol%] | Co-catalyst [mol%] | T [°C] | t [h] | Yield [%]b |
---|---|---|---|---|---|
a All reaction were performed on 0.2 mmol scale of 1a in 2 mL of solvent (0.1 M) under air atmospher for the indicate time. b Unless noted, yields are referred to the isolated product after column chromatography. c The unreacted starting material was recovered. d Without column chromatography. | |||||
1 | Cu(OAc)2·H2O (10) | FeCl3·6H2O (5) | r.t. | 24 | 99d |
2 | Cu(OAc)2·H2O (10) | Fe2O3 (5) | r.t. | 20 | 98d |
3 | Cu(OAc)2·H2O (10) | Fe(NO3)3 9H2O (5) | r.t. | 10 | 32 |
4 | Cu(OAc)2·H2O (10) | Fe(ClO4)3 (5) | r.t. | 48 | 73 |
5 | Cu(OAc)2·H2O (10) | Fe2(SO4)3·H2O (5) | r.t. | 48 | Tracec |
6 | Cu(OAc)2·H2O (10) | Fe(acac)3 (5) | r.t. | 40 | 99 |
7 | Cu(OAc)2·H2O (5) | FeCl3·6H2O (2.5) | r.t. | 40 | 99d |
8 | Cu(OAc)2·H2O (10) | FeCl3·6H2O (5) | 50 | 3 | 99d |
9 | Cu(OAc)2·H2O (10) | — | r.t. | 48 | Tracec |
10 | — | FeCl3·6H2O (10) | r.t. | 48 | Tracec |
Control experiments revealed that both Cu(OAc)2·H2O and FeCl3·6H2O were essential for this transformation as the omission of one of two catalysts resulted in only a trace amount of 2a (Table 1, entries 9 and 10). To our delight, the product 2a was purified simply by extraction with EtOAc, filtration through a plug of silica gel, concentration, and precipitation without the tedious column chromatography.
With these conditions in hand, the generality and scope of this novel intramolecular C–H amination process was explored (Table 2). First, the effect of different O-alkyl groups on the ester moiety of α-indolylhydrazone 1 was investigated. It was found that, in addition to methyl, ethyl, isopropyl, allyl or benzyl substituted substrates also worked well under the standard conditions, producing the desired products 2a–f in good to excellent yields. Incorporation of a phosphonate residue (R3 = PO(OMe)2) into the product (2g) was tolerated. Notably, the cyclization could also proceed successfully to deliver product 2h when a substrate with an amide N-protective group (R5 = NH2) was used. Compounds 2i,j, which vary in the substitution pattern at the R4, resulted in good yields. An evaluation of the substituents on indolic nitrogen revealed that N–H indole is most sensitive for this transformation. Specifically, for compound 2l, a spontaneous conversion to azacarboline 5a92 was registered. This observation imply that the substituent attached to the nitrogen of the indole nucleus is crucial in precluding ring expansion.
a Reaction conditions: 1 (0.2 mmol), Cu(OAc)2·H2O (10 mol%) and FeCl3·6H2O (5 mol%) in H2O (2.0 mL) at 50 °C, 3–48 h. b 3.15 mmol scale reaction (0.933 g). c Isolated yields after column chromatography. d 50 °C for 36 h then 70 °C for 24 h. e A spontaneous conversion to azacarboline 5a (vide infra) was observed. |
---|
![]() |
Then, we examined the substituent (R2) effect on the indole ring. Delightedly, the reaction well tolerated either EDG or EWG groups at 4-, 5-, 6-, or 7-positions of the indole rings affording the corresponding products 2m–s in excellent yields. Finally, pyrrolo[2,3-b]indole 2t incorporating a ring system between the N and C7 atoms of the indole ring was successfully generated.
For the majority of the reactions, purification of products by column chromatography can be avoided. Also, the recovery and reutilization of the aqueous solution containing the Cu/Fe co-catalyst system was tested using the formation of compound 2a as model reaction (see (Fig. 2) and Table S2 of the ESI†). Thus, after the isolation of the latter compound, the recovered aqueous solution was used again to accomplish the same transformation up to five times. As shown in (Fig. 2), it is possible to use the water solution two times with no variation in both the yields and times of conversion of 1a to 2a. However, for the second recycling run, a clear decrease in the yield was observed. Comparable yield with a remarkable increase of the time of conversion resulted for the third run. While modest variations are observed for the fourth run, a more sensible change of both the yield and process time was registered for the subsequent run with the aqueous catalytic system that retains over 50% of its activity after five cycling runs. Interestingly, recycling involves not only the catalytic system but also the aqueous reaction medium itself, differently from most of the methods reported in the literature in which the metal species has to be prior separated from the reaction medium and often reactivated before its reuse.
In addition, this intramolecular C(sp2)–H amination was found to be scalable delivering product 2a in 94% yield (0.933 g) on 3.15 mmol scale.
To further explore the synthetic potential of the developed method, transformations of the thus formed pyrrolo[2,3-b]indoles were conducted (Scheme 1). Pyrroloindoles 3 and 4 can be easily obtained by N–N bond reductive cleavage (Magnus's protocol93) and removal of N-Boc protecting group, respectively. Also, azacarboline 5b can be prepared from compound 4 following our previously reported oxidation procedure.57 Alternatively, a Paal–Knorr pyrrole synthesis from 4 generated the N–N indole-pyrrole scaffold 6.94 The acetylenic dienophile DMAD also reacted in Diels–Alder reaction with 2 to give 9H-carbazole 7, after N-aminonitrene extrusion.95
On the basis of literature information and our findings, a tentative mechanism for the transformation of 1a to 2a is illustrated in (Scheme 2). Initial N,O-bidentate coordination of the hydrazone moiety in 1a to the Cu(OAc)2 species occurs to form the metallacycle complex I with the release of AcOH. Oxidation of the Cu(II) adduct I to intermediate II by Fe(III) would then facilitate the subsequent electrophilic aromatic substitution. Proton abstraction at the 2-position of indole through a six-membered transition state III (or via acetate ligand-assisted concerted C–H activation pathway) may be followed to give endo-metallacycle96 intermediate IV. Finally, reductive elimination promoted by CH/NH tautomerization generates product 2a along with the Cu(I) specie. The catalytic cycle is completed by the regeneration of the active Cu(II) catalyst by air oxidation.
Some control experiments to support this mechanistic scenario were carried out. Under the reaction conditions, α-indolylhydrazone 1u bearing a phenyl substitution failed to produce 2u (Scheme 2a), which implies that the electron-withdrawing group (EWG) at the α-position of the hydrazone substrate results indispensable for the reaction to occur. Furthermore, no cycloamination was detected in the case of N-phenyl 1v because the copper complex via a bidentate coordination could not formed during the catalytic cycle, indicating the crucial role of N-carboxyalkyl moiety in directing/assisting the cyclization at 2-position of the indole ring. A pathway in which a hydrogen abstraction promotes the reductive elimination of metallacycle intermediate IV instead of a preliminary CH/NH tautomerization on 1a57 (not shown) is also supported by an experiment in which the addition of K2CO3 as a base resulted in the dramatically decrease of the product yield of 2a (29% yield).
Our new method is clearly distinguished from reported copper-catalyzed aerobic annulation of (hetero)aromatic hydrazones. With respect to Xiao and Xu protocol,56 this method furnishes 1,8-dihydro pyrrolo[2,3-b]indoles (instead of cinnolines) with excellent control of the chemoselectivity and the obvious benefit of providing more sustainable solution with water under mild reaction conditions.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d2gc02340g |
This journal is © The Royal Society of Chemistry 2022 |