Synthesis of tricyclic 3-hydroxyisoindolin-1-ones via triethylamine-catalyzed domino reactions of electron-deficient alkynes with phthalimidomalonate derivatives

Abstract

A triethylamine-catalyzed domino reaction between phthalimidomalonate derivatives and activated alkynes has been developed. This catalytic protocol is applicable to the synthesis of structurally functionalized N-fused tricyclic 3-hydroxyisoindolin-1-ones.

---

The development of new methods that can construct diversely nitrogen-containing heterocycles remains a preeminent goal and a great challenge in the pharmaceutical and agrochemical industries. Towards this target, domino reactions are among the most powerful processes in synthetic organic chemistry, allowing rapid generation of molecular complexity and diversity in a one-pot transformation from easily available starting materials.

Isoindolones are structural motifs that are widely present in natural products and medicinally important agents. In particular, 3-hydroxyisoindolin-1-ones bearing a hydroxyl group at the 3-position are key structural elements that appear in the core structures of a lot of naturally occurring isoindolone alkaloids and pharmaceutically relevant compounds (Fig. 1).¹ For instance, chilenine (1), an isoindole-containing benzazepine, is isolated from Chilean barberries.^1a Chlorphthalidone (2) has been widely used as an antihypertensive agent.^1c The isoindolin-1-one derivative (3) has demonstrated antibacterial activity.^1e Other interesting 3-hydroxyisoindolone derivatives have also shown inhibition of the MDM2-p53 protein–protein interaction,^2a HIV integrase^2b as well as cyclin-dependent kinases (CDKs).^2c In addition, 3-hydroxyisoindolones are also versatile building blocks that can undergo synthetically useful transformations to other heterocycles.³ However, only a few approaches have been reported thus far to construct this novel N-fused structure, which mainly included (1) nucleophilic addition, reduction or photolysis reactions of phthalimide derivatives;⁴ (2) cyclization of ortho functional benzoic acid derivatives.⁵ Despite these presented methods for the synthesis of isoindolin-1-ones, it is in high demand to develop new and efficient protocols for the synthesis of 3-hydroxyisoindolin-1-one derivatives due to the unique status of 3-hydroxyisoindolin-1-ones in pharmaceutical chemistry and synthetic chemistry.

Fig. 1 The related structures of 3-hydroxyisoindolin-1-ones.

Yavari and Esmaili reported a strategy for the synthesis of tetraalkyl 2,3-dihydro-5-oxopyrrolo[2,1-a]isoindole-1,2,3,3-tetracarboxylates using dialkyl acetylenedicarboxylate, dialkyl phthalimidomalonates and a stoichiometric amount of triphenylphosphine (Scheme 1a).⁶ Very recently, we found that phosphine can effectively catalyse the [3 + 2] annulation of electron-deficient alkynes with phthalimide derivatives for the synthesis of 3-hydroxyisoxazolo[3,2-a]isoindol-8(3aH)-ones (Scheme 1b).⁷ Based on these earlier studies, we hypothesized that phthalimidomalonates could react with electron-deficient alkynes in the presence of a phosphine catalyst to yield 3-hydroxyisoindolin-1-ones. To our surprise, no product was found when the reaction was disposed with a phosphine catalyst, but the expected tricyclic 3-hydroxyisoindolin-1-one was formed under the influence of a tertiary amine catalyst (Scheme 1c). To the best of our knowledge, this is the first example of an amine-catalyzed domino reaction of activated alkynes with an amide as electrophile partners, although the reactivity of electron-deficient alkynes with aldehyde, imine or activated ketone electrophiles to construct azacyclic compounds has been studied extensively.⁸ Herein, we wish to report a novel example of amine-catalyzed domino reactions of phthalimidomalonates with electron-deficient alkynes for the construction of highly functionalized 3-hydroxyisoindolin-1-one derivatives.

Scheme 1 Base-mediated synthesis of isoindolones using phthalimide derivatives.

As a test case for the construction of 3-hydroxyisoindolin-1-one derivatives, we subjected the nucleophile diethyl 2-(1,3-dioxoisoindolin-2-yl)malonate (1a) and benzyl propiolate (2a) to our previously reported cyclization reaction conditions: i.e., 20 mol% of triphenylphosphine in DMF at room temperature (Table 1, entry 1). No product was formed. Even when a stronger nucleophilic base, Bu₃P, was used as a catalyst, no product was found (entry 2). To our delight, the 3-hydroxyisoindolin-1-one (3a) with multiple functional groups was obtained with a moderate yield, when triethylamine was applied in the reaction system (entry 3). To improve the chemical yield, various solvents were examined. Acetonitrile proved to be the most efficient reaction medium, providing the 3-hydroxyisoindolin-1-one in 51% isolated yield after 24 h (entry 4). The product 3a was obtained in 13% yield when EtOAc was used as a solvent. Other solvents, such as CH₂Cl₂, toluene, THF, 1,4-dioxane and ether, were infeasible (entries 5–10). Further screening of bases revealed that triethylamine was the optimal base (entries 11–15). DBU could also effectively promote cyclization and give a matched yield. The desired product was given only in 8% yield when DMAP was used as a promoter. No reaction was occurred when the reaction was carried out in the presence of pyridine. Further investigation showed that the temperature was an important parameter for the formation of 3a (entries 16–19). For example, when the reaction is run at 0 °C, 3a was obtained in 55% yield; no reaction was found when the reaction was performed at −35 °C even with a prolonged reaction time. The starting material diethyl 2-(1,3-dioxoisoindolin-2-yl)malonate (1a) was consumed quickly, but the reaction became very complex when the reaction was proceeded at 60 °C. Decreasing the loading of the catalyst Et₃N to 10 mol% did affect the reaction rate and the yield to a certain extent (entry 20). No reaction was found when the reaction was performed without Et₃N.

Table 1 Optimization of reaction conditions^a

Entry	Cat. (mol%)	Solvent	Temp. [°C]	Time [h]	Yield^b [%]
a Unless noted otherwise, reaction of diethyl 2-(1,3-dioxoisoindolin-2-yl)malonate (1a) (0.4 mmol, 122 mg) and benzyl propiolate (2a) (0.8 mmol, 128 mg) was performed in 5 mL of solvent under N2. b Isolated yield based on 1a. c Only the Michael addition product was formed.
1	Ph3P (20)	DMF	25	24	NR
2	Bu3P (20)	DMF	25	24	NR
3	Et3N (20)	DMF	25	24	40
4	Et3N(20)	CH3CN	25	24	51
5	Et3N (20)	CH2Cl2	25	12	5
6	Et3N (20)	Toluene	25	12	7
7	Et3N (20)	EA	25	24	13
8	Et3N (20)	THF	25	48	NR
9	Et3N (20)	1,4-Dioxane	25	48	NR
10	Et3N (20)	Ether	25	48	NR
11	NaOH (20)	CH3CN	25	24	62^c
12	PPh3 (20)	CH3CN	25	24	Trace
13	Pyridine (20)	CH3CN	25	36	NR
14	DBU (20)	CH3CN	25	24	50
15	DMAP (20)	CH3CN	25	24	8
16	Et3N (20)	CH3CN	40	1.5	45
17	Et3N (20)	CH3CN	60	0.5	43
18	Et3N (20)	CH3CN	0	100	55
19	Et3N (20)	CH3CN	−35	200	NR
20	Et3N (10)	CH3CN	0	100	32
21	Et3N (40)	CH3CN	0	100	51
22	Et3N (0)	CH3CN	0	100	0

---

With the optimized reaction conditions in hand, a variety of 3-hydroxyisoindolin-1-one derivatives were efficiently synthesized in good chemical yields (Table 2). Generally, all terminal alkynoates were proved to be applicable to this reaction under the optimized reaction conditions. The nature of the substituent on the benzene ring of the benzyl propiolate had a great effect on the yields. For example, for a substrate with a strong electron-donating substituent methoxy group attached on the benzene ring, the yield of the corresponding product became very low, while 4-(trifluoromethyl)benzyl propiolate gave the product with a very good yield. The reaction was equally effective at producing the corresponding product in good yield when a 2-naphthyl group was introduced. Gratifyingly, the benzene ring of benzyl propiolate could be replaced with heteroaryl groups, such as furan-2-yl or thiophen-2-yl, and gave the desired products in good yields. The alkynoates bearing little alkyl units, namely ethyl and methyl, could give the products in 53 and 50% yields respectively and the alkynoate with a branched bulky alkyl group, such as t-butyl, gave the product in a very low yield. No product was formed when a β-substituted alkynoate was applied in the reaction. This protocol could be readily extended to various alkynyl ketones. For example, using the same reaction conditions and starting with but-3-yn-2-one, we isolated the corresponding 3-hydroxyisoindolin-1-one 3p in 87% yield. However, the reaction became very complex when 1-phenylprop-2-yn-1-one was used in the reaction (Scheme 2).

Scheme 2 Gram-scale synthesis of 3-hydroxyisoindolin-1-one 3m and further transformations.

Table 2 Reactions of phthalimidomalonates (1) with electron-deficient alkynes (2)^a,b

a The reaction of phthalimidomalonate (1) (0.4 mmol), propiolate (2) (0.8 mmol) and Et₃N (0.08 mmol, 128 mg) in CH₃CN at 0 °C for 100 h under N₂. b Isolated yield based on 1.

---

No reaction was found when N,N-diethylpropiolamide or N,N-dibenzylpropiolamide was used as a reaction partner in place of terminal alkynoates. Furthermore, different phthalimidomalonates, such as dimethyl 2-(1,3-dioxoisoindolin-2-yl)malonate and diisopropyl 2-(1,3-dioxoisoindolin-2-yl)malonate, also worked nicely with ethyl propiolate in the presence of Et₃N (20 mol%), providing 3u and 3v with good yields. However, other CH-active units in place of malonate, such as ethyl 2-(1,3-dioxoisoindolin-2-yl)-3-oxobutanoate or 2-(2,4-dioxopentan-3-yl)isoindoline-1,3-dione, were used as a reaction partner, but no product was formed.

The structures of the products were undeniably confirmed by X-ray crystallographic analysis of compound 3m.⁹ The ORTEP diagram of 3m is shown in Fig. 2.

Fig. 2 X-ray crystal structure of 3m.

To demonstrate the synthetic potential of this catalytic system, the gram-scale preparation of highly functionalized 3-hydroxyisoindolin-1-one 3m was investigated. The reaction of 4 mmol of the starting material (1a) proceeded smoothly, delivering the corresponding product 3m without a significant loss of efficiency (small scale, 50%) (Scheme 3). The 3-hydroxyisoindolin-1-ones can be readily converted into various interesting compounds (Scheme 3). First, treatment of 3m with NaOH in EtOH afforded the functional 1H-pyrrole 4 in 89% yield. Next, the 3-hydroxyisoindolin-1-one 3m was treated with Pd/C under the atmosphere of H₂ in MeOH at room temperature, leading to isoindolin-1-one 5 in 74% yield. Interestingly, triethyl 2-acetoxy-5-oxo-2,5-dihydro-3H-pyrrolo[2,1-a]isoindole-1,3,3-tricarboxylate 6 was obtained by treatment of 3m with AcCl and Et₃N in CH₂Cl₂ at room temperature.

Scheme 3 Mechanism for the formation of 3-hydroxyisoindolin-1-one derivatives 3m.

At present, the reaction mechanism is not clear. However, on the basis of our experimental observations and known literature, we tentatively proposed two mechanisms for the formation of 3-hydroxyisoindolin-1-one derivatives 3 in Scheme 3. For path A, the nucleophile triggers the reaction by Michael type nucleophilic conjugate addition at the carbon–carbon triple bond of 2m to produce vinylammonium carbanion I, which can then deprotonate phthalimidomalonate 1a to generate intermediates II and III. Intermediate III then undergoes Michael addition to II to give IV. Intermediate IV might then undergo intramolecular nucleophilic attack to form intermediate V, followed by proton transfer and elimination of Et₃N to produce the desired product 3m. For path B, Et₃N abstracts a proton from an active methylene of 1a to form the nucleophilic carbanion III. Addition of this anion to the carbon–carbon triple bond of 2m generates intermediate VII. Intramolecular addition in VII produces intermediate VIII, which abstracts a proton from phthalimidomalonate 1a to give intermediate III and product 3m.

Conclusions

In summary, we have developed a triethylamine-catalyzed domino reaction of phthalimidomalonate derivatives with various activated alkynes under mild reaction conditions to give the 3-hydroxyisoindolin-1-one derivatives. This reaction could be performed on the gram scale and is a practical protocol. Subsequent transformations provided a concise access to various heterocycles and pharmaceutically attractive compounds.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (Grant no. 21102179, 21572271 and 81473078), the Qing Lan Project, the project-sponsored by SRF for ROCS, SEM and National Found for Fostering Talents of Basic Science (Grant no. J1030830).

References

1. (a) G. Kim, P. Jung and A. T. Le, Tetrahedron Lett., 2007, 49, 2391; (b) M. H. A. Zarga, S. S. Sabri, S. Firdous and M. Shamma, Phytochemistry, 1987, 26, 1233; (c) J. G. Topliss, L. M. Konzelman, N. Sperber and F. E. Roth, J. Med. Chem., 1964, 7, 453; (d) B. R. Davis, J. A. Cutler, C. D. Furberg, J. T. Wright, M. A. Farber and J. V. Felicetta, Ann. Intern. Med., 2002, 137, 313; (e) A. Mikolasch, S. Hessel, M. G. Salazar, H. Neumann, K. Manda, D. Gördes, E. Schmidt, K. Thurow, E. Hammer, U. Lindequist, M. Beller and F. Schauer, Chem. Pharm. Bull., 2008, 56, 781.
2. (a) I. R. Hardcastle, S. U. Ahmed, H. Atkins, G. Farnie, B. T. Golding, R. J. Griffin, S. Guyenne, C. Hutton, P. Källblad, S. J. Kemp, M. S. Kitching, D. R. Newell, S. Norbedo, J. S. Northen, R. J. Reid, K. Sarvanan, H. M. G. Willems and J. Lunec, J. Med. Chem., 2006, 49, 6209; (b) J. M. Chen, X. Chen, M. Fardis, H. Jin, C. U. Kim and L. N. Schacherer, World Pat, WO2004035577-A3, 2004; Chem. Abstr. 2004 140 375156; (c) T. Honma, N. Yoshizumi, K. Hayashi, N. Kawanishi, K. Fukasawa, T. Takaki, C. Ikeura, M. Ikuta, T. I. Suzuki, T. Hayama, S. Nishimura and H. Morishima, J. Med. Chem., 2001, 44, 4628.
3. (a) G. Blaskó, D. J. Gula and M. Shamma, J. Nat. Prod., 1982, 45, 105; (b) L. Li, M. Wang, X. Zhang, Y. Jiang and D. Ma, Org. Lett., 2009, 11, 1309; (c) M. Lamblin, A. Couture, E. Deniau and P. Grandclaudon, Org. Biomol. Chem., 2007, 5, 1466; (d) S. Couty, C. Meyer and J. Cossy, Tetrahedron Lett., 2006, 47, 767; (e) S. Couty, B. Liegault, C. Meyer and J. Cossy, Tetrahedron, 2006, 62, 3882; (f) P. Pigeon and B. Decroix, Tetrahedron Lett., 1996, 37, 7707; (g) Y. Koseki and T. Nagasaka, Chem. Pharm. Bull., 1995, 43, 1604.
4. (a) T. Bootwicha, D. Panichakul, C. Kuhakarn, S. Prabpai, P. Kongsaeree, P. Tuchinda, V. Reutrakul and M. Pohmakotr, J. Org. Chem., 2009, 74, 3798; (b) T. Bousquet, J. F. Fleury, A. Daïch and P. Netchitaïlo, Tetrahedron, 2006, 62, 706; (c) A. G. Griesbeck, K. D. Warzecha, J. M. Neudörfl and H. Görner, Synlett, 2004, 2347; (d) M. Othman, P. Pigeon and B. Decroix, Tetrahedron, 1997, 53, 2495; (e) C. H. Yeh, Y. C. Lin, S. Mannathan, K. Hung and C. H. Cheng, Adv. Synth. Catal., 2014, 356, 831; (f) H. N. Nguyen, V. J. Cee, H. L. Deak, B. Du, K. P. Faber, H. Gunaydin, B. L. Hodous, S. L. Hollis, P. H. Krolikowski, P. R. Olivieri, V. F. Patel, K. Romero, L. B. Schenkel and S. D. Geuns-Meyer, J. Org. Chem., 2012, 77, 3887.
5. (a) G. Kim, P. Jung and L. A. Tuan, Tetrahedron Lett., 2008, 49, 2391; (b) Y. Zhou, Y. Zhai, J. Li, D. Ye, H. Jiang and H. Liu, Green Chem., 2010, 12, 1397; (c) S. Sharma, E. Park, J. Park and I. S. Kim, Org. Lett., 2012, 14, 906; (d) X. X. Zheng, C. Du, X. M. Zhao, X. Zhu, J. F. Suo, X. Q. Hao, J. L. Niu and M. P. Song, J. Org. Chem., 2016, 81, 4002; (e) L. Yang, L. Han, B. Xu, L. Zhao, J. Zhou and H. Zhang, Asian J. Org. Chem., 2016, 5, 62.
6. I. Yavari and A. A. Esmaili, J. Chem. Res., 1998, 714.
7. (a) Q. F. Zhou, X. P. Chu, F. F. Ge, Y. Wang and T. Lu, Adv. Synth. Catal., 2013, 355, 2787; (b) Q. Chen, K. Li, T. Lu and Q. Zhou, RSC Adv., 2016, 6, 24792; (c) X. Chen, F. F. Ge, T. Lu and Q. F. Zhou, J. Org. Chem., 2015, 80, 3295.
8. (a) D. Tejedor, S. López-Tosco, G. Méndez-Abt, L. Cotos and F. García-Tellado, Acc. Chem. Res., 2016, 49, 703; (b) G. L. Zhao and M. Shi, J. Org. Chem., 2005, 70, 9975; (c) C. H. Chen, G. S. Yellol, P. T. Lin and C. M. Sun, Org. Lett., 2011, 13, 5120; (d) Y. G. Wang, S. L. Cui and X. F. Lin, Org. Lett., 2006, 8, 1241; (e) P. D. Armas, F. García-Tellado, J. J. Marrero-Tellado, D. Tejedor, M. A. Maestro and J. Gonzalez-Platas, Org. Lett., 2001, 3, 1905; (f) D. Tejedor, S. López-Tosco, J. González-Platas and F. García-Tellado, J. Org. Chem., 2007, 72, 5454; (g) D. T. Aragón, G. V. López, F. García-Tellado, J. J. Marrero-Tellado, P. D. Armas and D. Terrero, J. Org. Chem., 2003, 68, 3363; (h) S. L. Cui, X. F. Lin and Y. G. Wang, Eur. J. Org. Chem., 2006, 5174; (i) D. Tejedor, F. García-Tellado, J. J. Marrero-Tellado and P. D. Armas, Chem. – Eur. J., 2003, 9, 3122; (j) D. Tejedor, D. González-Cruz, A. Santos-Expósito, J. J. Marrero-Tellado, P. D. Armas and F. García-Tellado, Chem. – Eur. J., 2005, 11, 3502; (k) D. Tejedor, L. Cotos and F. García-Tellado, Org. Lett., 2011, 13, 4422; (l) D. Tejedor, G. Méndez-Abt, J. González-Platas, M. A. Ramírez and F. García-Tellado, Chem. Commun., 2009, 17, 2368.
9. CIF files have also been deposited with the Cambridge Crystallographic Data Centre as “supplementary publication no. CCDC 1493448 for compound 3m”.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 1493448. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6qo00544f

---