[3 + 2] Annulations between indoles and α,β-unsaturated ketones: access to pyrrolo[1,2-a]indoles and model reactions toward the originally assigned structure of yuremamine

Abstract

A direct [3 + 2] annulation reaction between indoles and α,β-unsaturated ketones is reported, which allows for the efficient assembly of densely substituted, highly functionalized pyrrolo[1,2-a]indoles. Model reactions toward the originally assigned structure of yuremamine are also described, leading to the successful construction of the core with required functionality.

---

The efficient transformations of simple starting materials into complex structures that resemble the core of natural products and medicinally important molecules are of great significance in organic synthesis. Pyrrolo[1,2-a]indole tricyclic ring system¹ is a basic nucleus of a number of complex natural products and biologically important molecules. For example, mitomycins and mitosenes containing the pyrrolo[1,2-a]indole core have received considerable attention from synthetic chemists and medicinal chemists because of their intricate structures and potent biological activities.² Yuremamine, a new member of the indole alkaloids, was isolated from the stem bark of Mimosa hostiles in 2005 by Callaway and co-workers and has been shown to have hallucinogenic and psychoactive effects.³ The originally assigned structure of yuremamine (1) features a densely decorated pyrrolo[1,2-a]indole core, three contiguous stereogenic centers and the polyphenolic rings. Since its isolation, several synthetic strategies toward the core structure have been developed by Kerr, Shi, Dethe, Chen, France and You.⁴ While these approaches are significant and have led to the successful construction of the pyrroloindole core, none of them could allow for the complete incorporation of all substituents/functionality in a single step that are positioned for the eventual chemical synthesis of 1. Very recently, Sperry and co-workers reported a concise biomimetic total synthesis of yuremamine, leading to its structural revision from the pyrroloindole (1) to the flavonoidal indole (2), which was initially hypothesized as a biosynthetic intermediate (Scheme 1).⁵

Scheme 1 Synthetic strategy toward the originally assigned structure of yuremamine.

Our synthetic strategy was inspired by the intricate molecular architecture of 1 without realizing that the structure of the natural product was mis-assigned when we started our study. Our approach to 1 is depicted in Scheme 1. We conceived that this pyrroloindole alkaloid (1) could be derived from pyrrolo[1,2-a]indole A involving functional group transformations and structural isomerization. The densely decorated pyrrolo[1,2-a]indole A could in turn be produced by the direct [3 + 2] annulation between a tryptamine derivative and the corresponding multi-substituted α,β-unsaturated ketone. One appealing feature of the synthetic design is that the pyrrolo[1,2-a]indole core (A), which bears useful substituents/functionality as required by the structure of 1, could be efficiently assembled from readily available starting materials in a single step operation.

While the reactions between indoles and α,β-unsaturated carbonyls to furnish the alkylated products at N1 or C2 or C3 via conjugate addition have been widely investigated,⁶ the annulation process to form a new ring is rare.⁷ Particularly, the related reactions of tryptamine derivatives with α,β-unsaturated ketones, as depicted in our synthetic design (Scheme 1), have been well studied for the preparation of fused indolines in the presence of organocatalysts and organometallic reagents.⁸ We envisioned that under suitable dehydrative conditions at elevated temperature, the thermodynamically more stable pyrrolo[1,2-a]indoles (A) could be formed, and thus the frame work of 1 could be assembled in a highly efficiently manner.

A model reaction between 3a and 4a was carried out first to identify the optimized reaction conditions (Table 1). Gratifyingly, a variety of acids, both Bronsted acids and Lewis acids, could be utilized to catalyze the transformation, producing 5a with moderate to good yields. The best result (87% yield) was obtained using 20 mol% p-TsOH·H₂O as the catalyst and toluene as the reaction media. As expected, high reaction temperature was essential for the formation of 5a, indicating that the pyrrolo[1,2-a]indole should be the thermodynamically more stable product.

Table 1 Optimization of reaction conditions^a

Entry	Acid	Yield^b of 3a (%)
a Reaction conditions: to a solution of 3a (62.8 mg, 0.2 mmol) and 4a (49.0 mg, 0.24 mmol) in toluene was added the specified acid. The resulting mixture was heated to 110 °C for 2 h.b Isolated yield after silica gel chromatography.
1	None	0
2	Benzoic acid	47
3	AcOH	38
4	TFA	50
5	p-TsOH·H2O	87
6	FeCl3	57
7	ZnCl2	50

---

The substrate scope of the [3 + 2] annulations was next investigated (Scheme 2). Under optimized reaction conditions, a variety of 3-substituted indoles and α,β-unsaturated ketones proved to be suitable substrates, delivering the pyrrolo[1,2-a]indoles with good to excellent yields. Indoles with different tethered nucleophiles at 3-position as well as an alkyl bromide side chain were well tolerated, furnishing the desired products with different functionalized side chains (5a–h). A series of α,β-unsaturated ketones with different substitution patterns turned out to be efficient substrates, generating the pyrrolo[1,2-a]indoles bearing useful functionality. The electron withdrawing group R₃ could be varied (5j, 5k), but is not required for the reaction to occur (5n). When R₂ is aliphatic, the annulation reaction also proceeded with high efficiency (5i). It should be noted that a variety of functional groups (protected amine, alcohol, alkyl bromide, ketone, ester, and amide) were compatible with the reaction conditions. Finally, two complex pyrrolo[1,2-a]indoles (5o, 5p) that contain the substituents/functionality for the chemical synthesis of 1 were successfully constructed.

Scheme 2 The substrate scope of the [3 + 2] annulation reactions.

Next, we turned our attention to the construction of the core of 1. Our first model reaction involved the Curtius rearrangement of the related carboxylic acid 6 with the hypothesis that the in situ generated isocyanate B could be hydrolyzed and isomerized to deliver ketone 7 (Scheme 3). The saponification of 5l underwent well using potassium hydroxide, affording 6 with good yield. The Curtius rearrangement initiated cascade sequence presumably involving B and C as the intermediates proceeded smoothly, producing ketone 7 in 89% yield. Unfortunately, when the similar synthetic sequence was applied to a more advanced substrate 8, the formation of ketone 9 was not observed. Under certain forcing reaction conditions, the generated isocyanate intermediate D was trapped by the nearby phenyl ring via Friedel–Crafts acylation reaction. Considering that the 1 contains more nucleophilic polyphenolic rings, which would attack the isocyanate intermediate more readily, we turned to an alternative approach as depicted in Scheme 4.

Scheme 3 Model reactions via Curtius rearrangement.

Scheme 4 Model reactions via Baeyer–Villiger oxidation.

The second model reaction for the synthesis of the core of 1 centered on the Baeyer–Villiger oxidation of aldehyde 10 and related structural isomerization (Scheme 4). Compound 10 was produced in good yield by a two step synthetic manipulation from 5g involving DIBAl-H reduction of the ester followed by Dess–Martin oxidation of the resulted alcohol. The Baeyer–Villiger oxidation of 10 proceeded well in the presence of oxone, delivering 11 in 69% yield. Under basic conditions, the deprotection and structural isomerization of 11 did not occur to afford the desired ketone, but instead 2,3-disubstituted indole 12 was isolated as the major product, presumably because that the electron rich system was further oxidized by molecular oxygen. A solution to this problem was realized by carrying out the reaction under N₂ and adding sodium boron hydride to the reaction to reduce the in situ generated ketone. By doing this, alcohol 13 was generated in good yields as a mixture of three diastereomeric isomers. Although the diastereoselectivity of the reduction need to be further improved, the structure of 13 contains the side chain and correct substitution pattern, which resembles the key structural features of the originally assigned structure of yuremamine (1).

Conclusions

In conclusion, we have developed a direct annulation reaction between indoles and α,β-unsaturated ketones, which allows for the rapid assembly of densely substituted, highly functionalized pyrrolo[1,2-a]indoles in a single step. Model reactions toward the originally assigned structure of yuremamine were carried out, leading to the successful construction of the core with required functionality and the discovery of several interesting synthetic transformations.

Notes and references

1. For selected 9H-pyrrolo[1,2-a]indoles synthesis: (a) A. Padwa and J. R. Gasdaska, J. Am. Chem. Soc., 1986, 108, 1104; (b) S. J. Hwang, S. H. Cho and S. Chang, J. Am. Chem. Soc., 2008, 130, 16158; (c) S. J. Mahoney and E. Fillion, Chem.–Eur. J., 2012, 18, 68; (d) L. Hao, Y. Pan, T. Wang, M. Lin, L. Chen and Z. Zhan, Adv. Synth. Catal., 2010, 352, 3215; (e) H. Liu, M. El-Salfiti and M. Lautens, Angew. Chem., Int. Ed., 2012, 51, 9846; (f) H. Ren and P. Knochel, Angew. Chem., Int. Ed., 2006, 45, 3462; (g) K. S. Feldman, M. R. Iyer and D. K. Hester II, Org. Lett., 2006, 8, 3113; (h) X. Huang, S. Zhu and R. Shen, Adv. Synth. Catal., 2009, 351, 3118.
2. For selected references about mitomycins and mitosenes: (a) H. Namiki, S. Chamberland, D. A. Gubler and R. M. Williams, Org. Lett., 2007, 9, 5341; (b) A. L. Williams, J. M. Srinivasan and J. M. Johnston, Org. Lett., 2006, 8, 6047; (c) R. S. Coleman, F.-X. Felpin and W. Chen, J. Org. Chem., 2004, 69, 7309; (d) S. J. Danishefsky and J. M. Schkeryantz, Synlett, 1995, 475; (e) T. Fukuyama and L. Yank, J. Am. Chem. Soc., 1989, 111, 8303; (f) T. Fukuyama, F. Nakatsubo, A. J. Cocuzza and Y. Kishi, Tetrahedron Lett., 1977, 4295; (g) Z. Zheng, M. Touve, J. Barnes, N. Reich and L. Zhang, Angew. Chem., Int. Ed., 2014, 53, 9302; (h) G. R. Allen Jr and M. J. Weiss, J. Org. Chem., 1965, 30, 2904; (i) J. C. Hodges, W. A. Remers and W. T. Bradner, J. Med. Chem., 1981, 24, 1184; (j) R. M. Coates and P. A. MacManus, J. Org. Chem., 1982, 47, 4822; (k) J. R. Luly and H. J. Rapoport, Org. Chem., 1984, 49, 1671.
3. J. J. Vepsalainen, S. Auriola, M. Tukiainen, N. Ropponen and J. Callaway, Planta Med., 2005, 71, 1053.
4. (a) M. B. Johansen and M. A. Kerr, Org. Lett., 2008, 10, 3497; (b) D. V. Patil, M. A. Cavitt and S. France, Org. Lett., 2011, 13, 5820; (c) K. Chen, Z. Zhang, Y. Wei and M. Shi, Chem. Commun., 2012, 48, 7696; (d) H.-L. Cui, X. Feng, J. Peng, J. Lei, K. iang and Y.-C. Chen, Angew. Chem., Int. Ed., 2009, 48, 5737; (e) W. B. Liu, X. Zhang, L. X. Dai and S. L. You, Angew. Chem., Int. Ed., 2012, 51, 5183; (f) D. H. Dethe, R. Boda and S. Das, Chem. Commun., 2013, 49, 3260.
5. M. B. Calvert and J. Sperry, Chem. Commun., 2015, 51, 6202.
6. (a) J. A. Joule and K. Mills, Heterocyclic Chemistry, Blackwell, Oxford, 5th edn, 2009; (b) T. Eicher and S. Hauptmann, The Chemistry of Heterocycles, Wiley-VCH, Weinheim, 2003.
7. For indoline synthesis by direct annulation: Q. Cai and S. L. You, Org. Lett., 2012, 14, 3040 . For pyrroloindole synthesis by direct annulation employing rather specialized substrates: K. Wood, D. S. Black and N. Kumar, Tetrahedron Lett., 2009, 50, 574.
8. (a) J. F. Austin, S. G. Kim, C. J. Sinz, W. J. Xiao and D. W. C. MacMillan, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 5482; (b) Z. Zhang and J. C. Antilla, Angew. Chem., Int. Ed., 2012, 51, 11778; (c) Q. Cai, C. Liu, X. W. Liang and S. L. You, Org. Lett., 2012, 14, 4588; (d) C. Liu, W. Zhang, L. X. Dai and S. L. You, Org. Biomol. Chem., 2012, 10, 7177.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra11904a

---